- Email: [email protected]
Thiolated polymers as mucoadhesive drug delivery systems
Accepted Manuscript Thiolated polymers as mucoadhesive drug delivery systems
Sarah Duggan, Wayne Cummins, Orla O' Donovan, Helen Hughes, Eleanor Owens PII: DOI: Reference:
S0928-0987(17)30020-9 doi: 10.1016/j.ejps.2017.01.008 PHASCI 3868
To appear in:
European Journal of Pharmaceutical Sciences
Received date: Revised date: Accepted date:
28 November 2016 4 January 2017 9 January 2017
Please cite this article as: Sarah Duggan, Wayne Cummins, Orla O' Donovan, Helen Hughes, Eleanor Owens , Thiolated polymers as mucoadhesive drug delivery systems. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Phasci(2017), doi: 10.1016/j.ejps.2017.01.008
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Thiolated polymers as mucoadhesive drug delivery systems
Sarah Duggan *a, Wayne Cummins a, Orla O’ Donovan a, Helen Hughes a, Eleanor Owens a Pharmaceutical and Molecular Biotechnology Research Centre, Department of Chemical
PT
a
and Life Sciences, Waterford Institute of Technology, Waterford, Ireland.
SC
AC
CE
PT E
D
MA
NU
Email address: [email protected] (Sarah Duggan)
RI
*Corresponding author. Tel: +353 51 834126
1
ACCEPTED MANUSCRIPT Abstract Mucoadhesion is the process of binding a material to the mucosal layer of the body. Utilising both natural and synthetic polymers, mucoadhesive drug delivery is a method of controlled drug release which allows for intimate contact between the polymer and a target tissue. It has the potential to increase bioavailability, decrease potential side effects and offer protection to
PT
more sensitive drugs such as proteins and peptide based drugs. The thiolation of polymers
RI
has, in the last number of years, come to the fore of mucoadhesive drug delivery, markedly
SC
improving mucoadhesion due to the introduction of free thiol groups onto the polymer backbone while also offering a more cohesive polymeric matrix for the slower and more
NU
controlled release of drug. This review explores the concept of mucoadhesion and the recent
PT E
D
mucoadhesive drug delivery devices.
MA
advances in both the polymers and the methods of thiolation used in the synthesis of
AC
polymers
CE
Keywords: mucoadhesion, controlled drug delivery, thiolation, natural and synthetic
2
ACCEPTED MANUSCRIPT 1 Introduction The area of mucosal surfaces of the body far exceeds that of the area of the skin, with the area of the small intestine alone being 100 x greater than the surface of the skin (Khanenko et al., 2009). With this in mind and together with the need for a protective delivery system for drugs such as peptides and proteins (Khutoryanskiy, 2011), the development of mucoadhesive drug
PT
delivery systems has been of great interest over the past number of years (Bernkop-Schnürch
RI
et al., 1999; Bernkop-Schnürch et al., 2004a; Hornof et al., 2003; Iqbal et al., 2012b).
SC
The transmucosal route of drug administration can have up to 4 x the rate of absorption than skin, depending on the site of administration, potentially making it a more resourceful route
NU
for drug delivery (Madhav et al., 2009). There are many advantages to the development of bio- and mucoadhesive drug delivery systems, including: Localised/site specific drug delivery
Reduced likelihood of systemic side effects
Better residence time/more intimate contact to the target surface
Avoidance of peak concentrations due to controlled/slow release of drug (zero order
PT E
D
MA
kinetics)
CE
Improved bioavailability of the drug may be observed due to the intimate and increased
AC
contact time between the drug-loaded polymer and the target area, thus reducing the time of action of the drug at the site. A lower concentration of drug can also be used as the drug could potentially be absorbed directly onto the intended tissue, not systemically, which reduces the potential for toxic side effects. Both synthetic and naturally derived polymers have demonstrated potential as mucoadhesives. Many synthetic polymers are hydrophobic and most natural polymers are hydrophilic (Yasukawa et al., 2004). Synthetic polymers are known to exhibit a longer duration in the body and, therefore, sustain the release of the drug over a prolonged time 3
ACCEPTED MANUSCRIPT period in comparison to the short duration of natural polymers. However, the conditions in which synthetic polymers are formulated, particularly in the formulation of nanoparticulate systems, can be harsher in comparison to natural polymers, using organic solvents in the process (Panyam and Labhasetwar, 2003). Synthetic polymers are also less likely to be biodegradable and may cause an immune response. The majority of polymers used in
PT
mucoadhesion are hydrophilic. They are soluble in water and have a high density of
RI
functional groups which are able to interact through hydrogen bonding with mucins, leading
SC
to stronger adhesion (Andrews et al., 2009).
Andreas Bernkop-Schürch pioneered research in the area of thiolated polymers using a range
NU
of different approaches (Bernkop-Schnürch and Steininger, 2000; Hornof et al., 2003; Andreas Bernkop-Schnürch et al., 2003; Bernkop-Schnürch et al., 2004a; Bernkop-Schnürch
MA
et al., 2004b; Grabovac et al., 2005; Palmberger et al., 2007; Schmitz et al., 2008; Vigl et al., 2009; Iqbal et al., 2011; Wang et al., 2012; Hauptstein et al., 2013; Hauptstein et al., 2014;
D
Hauptstein et al., 2015). By thiolating well established mucoadhesive polymers,
PT E
mucoadhesive properties were greatly improved, with up to a 140-fold increase in adhesive properties (Bernkop-Schnürch, 2005). This is due to the formation of inter- and intra-
CE
disulphide bridges with the cysteine rich areas of the mucosal layer. These covalent bonds
AC
formed with the mucosal layer are much stronger than the non-covalent bonds formed by non-thiolated polymers, and mirror the natural bonds formed within the mucosal layer itself.
2 Mucoadhesion 2.1 Mucosal layer The mucosal layer acts as a lubricating and protective barrier and lines the gastrointestinal (GI) tract, eye, mouth, nasal cavity, reproductive tract and the respiratory tract. It consists of approximately 95% water, with cholesterol, lipids, proteins and other elements making up the 4
ACCEPTED MANUSCRIPT remainder (Bansil and Turner, 2006). The viscoelastic properties of the mucosal layer are controlled generally by the glycoproteins, mucins, but water and ion content also contribute (Ensign et al., 2012). The surface will allow small molecules and viruses to filter through freely (Cone, 2009) but it may become a potential barrier for larger molecules and, therefore, drug delivery. Mucus is constantly secreted, but the amount and the thickness of the layer
PT
varies depending on the organ, with highest levels of secretion in the GI tract (Sigurdsson et
RI
al., 2013). Mucus secretion alters in some disease states, an example of which is cystic
SC
fibrosis, a disease where there is an increase in mucus production. The rate of secretion is important to the development of site specific drug delivery systems, particularly when
which may affect the absorption of the drug.
NU
targeting the GI or reproductive tracts, both of which have faster mucosal layer turnover rates
MA
The mucoadhesive properties of mucus are due to the presence of glycoproteins called mucins. There are two types of mucins: soluble and membrane bound. Mucins are secreted by
D
goblets cells in the epithelium. Mucins are highly glycosylated, large molecules which are
PT E
negatively charged at physiological pH, ~7.4. They have domains which are cysteine rich which are targeted in mucoadhesion using thiolated polymers due to the formation of
CE
intermolecular disulphide bonds, with secondary bonding, such as hydrogen bonding, Van
AC
der Waals forces, also coming into play. 2.1.1 Areas of administration The oral mucus is easily accessible for drug delivery and devices, such as implants or inserts, can be easily removed, if necessary. The surface area is relatively small, but it is highly vascularised leading to rapid drug absorption. Both the buccal (the cheek tissue) and sublingual (under the tongue) routes bypass the GI tract degradation and the hepatic first pass metabolism, both of which often lead to decreased drug plasma levels and are important considerations in drug delivery. The sublingual area is constantly being washed by saliva, 5
ACCEPTED MANUSCRIPT which may impede adhesion or wash the polymer tablet away. It has been observed that involuntary swallowing and increased saliva production can decrease the quantity of drug absorbed. Similar to the oral mucosa, ocular mucosa has the advantage of being easily accessible. However, the eye is a notoriously difficult target for drug delivery as structural barriers and
PT
functional barriers, such as drug clearance by the conjunctival tissues, affect the delivery of
RI
drug to the target tissue (Anderson et al., 2010). Topical preparations, eye drops etc. are the
SC
most convenient for patient use, however, they have a very short contact time on the eye and approximately 90% of the administered dosage is cleared within two minutes from blinking
NU
and lacrimation. Topical preparations require frequent administration and there is often a phase of over-dosage followed by under-dosage. The use of mucoadhesive systems for drug
MA
delivery could be highly valuable, particularly for drug delivery to the posterior segment of the eye, due to the increased residence time of the device on the surface of the eye. Hornof et
D
al. (2003) developed a thiolated PAA-cysteine ocular insert containing diclofenac samples
PT E
and evaluated the inserts in human volunteers. Thiolated PAA inserts were compared to unmodified PAA insert controls. Inserts were positioned in the cul-de-sac of the eye and
CE
various different physical attributes of the insert were analysed including sensation, blurring
AC
and irritation. The controls dissolved within 15 min of administration whereas the thiolated inserts were retained for up to 8 h. The drug release and drug removal patterns of fluorescein, a fluorophotometric agent, from the thiolated PAA insert, an unmodified insert and aqueous eye drops were all compared. The eye drop administration resulted in a high concentration of drug immediately after administration which decreased substantially after 2 h. The unmodified insert had a similar profile to that of eye drops, with an uncontrolled release of drug to the surface of the eye and fast removal. This was due to the lack of cohesion and quick disintegration of the insert. It was in stark contrast to the thiolated polymer which had a 6
ACCEPTED MANUSCRIPT more prolonged and controlled release of the fluorescein over 7 h. Commercially, Pilogel Ophthalmic Gel®, a pilocarpine HCl based gel, has the addition of Carbopol 940, a high MW polyacrylic acid in its formation, added to increases the residence time of the gel on the surface of the eye due to its inherent mucoadhesive properties. Similarly, Novartis’ Nyogel®, another eye gel, uses carbomer 934 in its formulation. A thiolated chitosan based eye drop,
PT
Lacrimera® (Chroma-Pharm), has also been developed and released onto the European
RI
market as a daily treatment to dry eye syndrome, lubricating for approximately 12 h.
SC
In the nasal cavity, mucus turnover is as fast as every 15 - 20 min and as much as two litres of nasal mucus is produced each day (Ugwoke et al., 2005), which could cause complications
NU
with regards to drug delivery through this route. The microvilli in the nasal cavity increase the surface area and increase the area of absorption. The epithelium in the cavity is thin,
MA
porous and highly vascularised (Ugwoke et al., 2005). Many factors, including environmental factors, such as temperature or inhaling cigarette smoke, and pathological factors, (such as
D
allergies, having a cold or asthma), all affect the rate of elimination and of mucus production,
PT E
i.e. lower temperatures decrease both the viscosity of mucus and its production. These could all have a great effect on the use of mucoadhesive devices for nasal drug delivery. Drugs
CE
administered via this route are capable of passing through the olfactory region of the brain
AC
and, consequently, pass into the cerebrospinal fluid (CSF) for clearance which is renewed approximately 4 - 5 times a day. Absorbed substances are quickly circulated with little restriction and in certain cases, substances are capable of travelling directly into the central nervous system (CNS), bypassing the blood-brain barrier (Ugwoke et al., 2005). Enzymatic activity is lower via the nasal route in comparison to the GI tract and because of this, there is a higher bioavailability. In terms of mucoadhesion, initially the turn-over rate of nasal mucosa slows down with mucoadhesion. However, over time, this is reversed and the production and elimination rates return to normal (Ugwoke et al., 2005). A mucoadhesive 7
ACCEPTED MANUSCRIPT tablet system for nasal administration will affect the mucociliary transport system (the elimination system). Mucoadhesion opposes this system and prolonged contact, which is key to the design of such a system, may cause irritation and even toxicity. This does not happen with the administration of an aerosol, as the contact time is minimal. The GI tract is possibly the most important area of administration, as oral administration is
PT
the preferable route of drug administration for both patients and doctors. Oral delivery is the
RI
easiest route for administration i.e. does not need professional administration and is most
SC
likely to have higher patient compliance (Patel et al., 2011). The GI tract includes the mouth, stomach and small and large intestines. It has a large surface area, increased by the presence
NU
of microvilli within the intestine which also increases the total rate of absorption. Mucosal turnover rate is fast and mucosal density varies throughout the GI tract (Ensign et al., 2012).
MA
In the stomach, mucus is secreted to protect the cell lining from the acidic nature of the stomach. Within the mucosal layer of the stomach, bicarbonate ions are released and trapped
D
which creates a pH gradient between the lumen (pH 1 – 2) and the mucosal layer (pH 6 – 7).
PT E
Prostaglandins are also released locally which promote the excretion of both more mucus and bicarbonate (H.P. Rang et al., 2003). Depending on the mobility in the GI tract, a specific
CE
area of the colon can be targeted for drug release either by coating the polymer with an
AC
enteric coating to prevent drug release in the upper GI due to low pH levels, or by increasing the crosslinking on the polymer matrix and thus slowing drug release from the matrix.
2.2 Adhesion theories In general, there are six theories of adhesion to the mucosal surface (Roy et al., 2009; Andrews et al., 2009; Khutoryanskiy, 2011): electronic theory, wetting theory, adsorption theory, diffusion theory, fracture theory, mechanical theory. All of the above adhesion theories are valid and it is most likely a combination of all six that result in adhesion. There are two stages of adhesion: the wetting/contact stage and the consolidation stage (Smart, 8
ACCEPTED MANUSCRIPT 2005). The wetting stage is the intimate contact between the polymer and the mucosal surface, which is required for mucoadhesion (Ludwig, 2005). Interpenetration of the adhesive polymer occurs within the mucus followed by non-covalent bonding, which is due to the presence of hydrophilic groups on the polymer. The extent of the swelling of the polymer and the penetration of the polymer chains into the mucosal surface is often governed by the initial
PT
contact time. If there is an increase in the initial contact time, a stronger adhesive bond is
RI
created (Asane et al., 2008). In the case of easily accessible areas, this intimate contact can
SC
occur by holding the polymer onto the surface until adhesion is apparent. For less accessible areas, e.g. GI tract, the movement of the muscle itself will allow for adhesion (Smart, 2005).
NU
The consolidation stage is important when a strong adhesion is required, particularly when the target area is under constant stress, for example targeting the blinking eye. Mucoadhesive
MA
strength is strongest when the polymer itself is completely dehydrated and is placed onto the mucosal surface. The polymer will begin to absorb the moisture from the mucosal layer
D
which, once dehydrated, switches from having lubricating properties to having adhesive
PT E
properties. The moisture will allow the polymeric matrix to become pliable and to adapt to the shape of the surface it is on, thus increasing the contact surface area. It will also allow for
CE
the formation of bonds between the mucus and the polymer (Smart, 2005).
AC
2.3 Methods of measuring mucoadhesion Once modified, the thiolated polymers must be tested for their mucoadhesive properties. A number of methods have been outlined; mostly commonly used in vitro methods include the use of a rotating cylinder method, rheology, tensile strength tests and flow-through tests. The rotating cylinder test was first described by Bernkop-Schnürch et al. (1999) where a cylinder was submerged in a phosphate buffer at 37 °C. To this cylinder, segments of porcine gastrointestinal tract were glued and to these, the thiolated PAA discs were allowed to adhere.
9
ACCEPTED MANUSCRIPT The cylinder was set to rotate and the time taken for the thiolated PAA discs to dissolved or dislodge was measured. This gave an indication of mucoadhesive properties in terms of time when compared to an unmodified or control sample. Rheological mucoadhesive testing gives an indication of the strength of the bond formed
PT
between the polymer and the mucin solution when compared with the mucin solution and the polymer solution on their own, with observed increased viscosity (Iqbal et al., 2012a) and
RI
increased storage (G') and loss modulus (G'') (Duggan et al., 2016b; Duggan et al., 2016a)
SC
occurring in a thiolated polymer-mucin mix. Rheology can also give an indication of the structural bonds occurring within the polymer-mucin mix, while allowing for a direct
NU
comparison of bond strength between different polymers (Duggan et al., 2016b). Adhesion
MA
forces between polymers and mucin have also been measured using atomic force microscopy (AFM), where fully hydrated pectin samples were evaluated for their mucoadhesive
D
properties and compared to PAA as a positive control (Joergensen et al., 2011).
PT E
Tensile testing, measuring the total work of adhesion and maximum force of detachment, are often conducted as a measurement of mucoadhesive forces. This can be conducted using a
CE
texture analyser (Ivarsson and Wahlgren, 2012), or a similar setup (Palmberger et al., 2007) where a segment of mucosal tissue is attached to the base of a beaker submerged in a buffer
AC
at physiological pH and temperature; the polymer is attached to a steel disc, hanging on a string off a retort stand. The polymer is allowed to adhere to the mucosal tissue, before slowing lowering the mucosal tissue, thus measuring the total work of adhesion and maximum force of detachment (Palmberger et al., 2007). Flow through or wash away tests give an understanding of mucoadhesion under stress, due to the flow of fluid over the polymeric tablet. This test is particularly useful when targeting areas, such as the eye where constant lacrimation and blinking could easily dislodge the
10
ACCEPTED MANUSCRIPT mucoadhesive tablet (Khutoryanskiy, 2011). This style of testing can also evaluate drug release in areas where flow of biological fluid is constant (Horvát et al., 2015).
2.4 Factors affecting mucoadhesion There are a number of factors that must be considered when designing a mucoadhesive polymer. Environmental factors, including pH, can affect the binding properties of the
PT
polymer to the mucosal surface. Properties of the polymer itself, such as molecular weight
RI
and structural modification of the polymer, can have an effect on the adhesive properties, the
SC
degree of and rate at which the polymer swells, and drug release profiles. The biodegradable properties of polymers, specifically synthetic polymers, are important for the removal of the
NU
carrier from the body and the degradation rate of the polymer may have an effect on the drug
MA
release profile.
2.4.1 Charge on the polymer Commonly used polymers have either an anionic or cationic charge, but some non-ionic
D
polymers are also utilised. Thiol groups are introduced to anionic or cationic polymers
PT E
through different reaction processes. Amide bonds are easily formed from the carboxylic acid groups on anionic polymers and thiolation can be achieved by the introduction of cysteine,
CE
mediated by carbodiimide coupling reagent (Bernkop-Schnürch, 2005). A complication with thiolation is the formation of disulphide bridges during the synthesis. This is due to the
AC
unintended oxidation of the thiol groups and may be avoided by performing the reaction under inert conditions or with the addition of EDTA in the buffer solution to chelate any metals that may catalyse the oxidation process. Cationic polymers are a desirable polymer type to use as they encourage ionic interactions between the positive charges of the polymer and the negative charges, such as the sialic acid residues, of the mucosal layer, (Makhlof et al., 2008), and, in the case of nanoparticles, increases the rate of internalisation into the cell (Kumari et al., 2010). A commonly used cationic polymer is chitosan. Thiolation can occur
11
ACCEPTED MANUSCRIPT through the formation of an amide bond, similar to the anionic polymers through the coupling of cysteine in the presence of a carbodiimide, or through the amidine bond with the reaction of Traut’s reagent (2-iminothiolane) (Bernkop-Schnürch, 2005). 2.4.2 pH Studies on the effects of pH changes have been conducted on various aspects of
PT
mucoadhesion: On its effect during the production stage of the polymer, and also its effect on
RI
the mucoadhesive and thiolation properties of the polymer. Grabovac et al. (2005) conducted a study of 19 different mucoadhesive polymers, including polyacrylic acid (PAA), chitosan,
SC
polycarbophil, thiolated PAA, thiolated chitosan and thiolated polycarbophil; thiol content of
NU
the thiolated samples were 498.8 µmol/g, 243.4 µmol/g and 95.7 µmol/g, respectively. The polymers were dissolved in D.I. water and the pH of the polymer solutions was adjusted to
MA
either pH 3 or pH 7, after which the polymers were dried using various methods. The polymers were compressed into discs and the mucoadhesive properties were evaluated in a
D
phosphate buffer at pH 6.8. It was observed that overall the thiolated polymers had a greater
PT E
level of adhesion in comparison to the unmodified samples due to the formation of intermolecular disulphide bonds between the thiolated polymers and mucin. The thiolated
CE
samples which had been adjusted to pH 3 displayed greater mucoadhesive properties than those adjusted to pH 7. It was surmised that the lower pH inhibited the formation of
AC
intramolecular disulphide bonds. Once the polymer solutions had been adjusted to pH 7, the tendency towards disulphide bond formation increased, crosslinking the polymers. Therefore, there were fewer free thiol groups to interact with mucins during mucoadhesive testing within the samples adjusted to pH 7. The pH 3 samples were less reactive and, once tested in the phosphate buffer (pH 6.8) for mucoadhesive properties, had a greater ratio of free thiol groups available to bond with mucin, thus increasing mucoadhesion.
12
ACCEPTED MANUSCRIPT Guggi et al. (2004) examined the influence pH had on pre-thiolated PAA samples; the thiol content of the samples prior to the pH change was not stated. The PAA-cysteine conjugates were dissolved in water, aliquots of which were then adjusted to different pHs between 3 and 8 and the samples were lyophilised. By reducing the samples with NaBH4, the disulphide bond formation was also measured. After the pH adjustment of the samples, the levels of
PT
thiolation were measured and cohesive and mucoadhesive properties were also examined. It
RI
was shown that the degree of thiolation decreased when the samples were adjusted to a higher
SC
pH. Therefore, the sample adjusted to pH 3 displayed the highest levels of thiolation while the lowest levels were observed when adjusted to pH 8. It was also noted that the disulphide
NU
bond content of the samples decreased with increasing pH; therefore, the samples adjusted to pH 8 displayed lower levels of disulphide bond content than that of the samples adjusted to
MA
pH 3. It was thought the formation of sodium carboxylate substructures within the samples which were adjusted to higher pH values increased the molecular weight of those samples,
D
resulting in the lower thiol content observed. The swelling abilities of the higher pH samples
PT E
were greater than that of the lower pH samples, and this also correlated to an increase in disintegration times with the higher pH samples displaying higher levels of cohesion.
CE
Mucoadhesion was greater in lower pH samples. It was concluded that the increase in molecular weight and faster swelling ability of the less acidic PAA-cysteine samples affected
AC
the mucoadhesive and cohesive nature of those samples. It was also thought that the lower pH levels of samples below pH 5 prevented the premature oxidation of the thiol groups within those samples, which allowed for higher levels of mucoadhesion and cohesion. It was, therefore, concluded that higher mucoadhesion could be obtained when the pH of the samples was adjusted to pH 3, but greater levels of cohesion and water uptake would require a higher pH. 2.4.3 Molecular weight and chain length 13
ACCEPTED MANUSCRIPT Molecular weight (MW) can influence the mucoadhesion and cohesive properties of a polymer due to the increase in chain length. MWs of between 100 kDa and 4000 kDa were indicated as being optimal for mucoadhesion by Smart (2005). The MW of modified polymers is equally important; thiolated polymers, both anionic and cationic were analysed and it was observed that thiolated polymers with medium MW, between 400 kDa and 600
PT
kDa, had the strongest mucoadhesion (Bernkop-Schnürch, 2005). The MW of the polymer
RI
may also effect the diffusion of the polymer into the mucus. It must also be noted, however,
SC
that it is difficult to define an optimum MW as each polymer system reacts differently and has a unique structure (Andrews et al., 2009); what is optimal for one polymer, may not be
NU
for another.
Chain length of the polymer and flexibility of the chain are considered to play an important
MA
role in mucoadhesion. Adhesive properties increase with a longer and more flexible chain (Roy et al., 2009). This is due to the interpenetration of the chains into the mucosal layer,
D
allowing for greater ease of interaction between functional groups of the polymer and mucins.
PT E
The diffusion theory of mucoadhesion suggests the diffusion of polymer chains into the mucosal layer and also the diffusion of the glycoprotein chains of mucin into the polymer
CE
matrix both occur. However, diffusion occurs more frequently with shorter polymers chains
AC
and chain lengths which are excessively long may impede mucoadhesion, as diffusion into the mucosal layer is lost (Huang et al., 2000). Crosslinking of the polymer reduces the flexibility of the chain and high degrees of crosslinking can reduce the levels of mucoadhesion due to fewer interactions with and less diffusion into the mucosal layer (Andrews et al., 2009). 2.4.4 Swelling Once the polymer has adhered to the mucosal layer, it will begin to extract the moisture from the surface and start to swell. This is advantageous as the surface area between the polymer 14
ACCEPTED MANUSCRIPT and the mucosal layer will increase (Andrews et al., 2009). By modifying the polymer with crosslinking moieties, such as the introduction of thiol groups, the degree of swelling can be controlled (Huang et al., 2000). This is an important consideration when creating a hydrogel type polymeric structure, as the release of drug is due to the uptake of water and swelling of the matrix. Polymers that are activated by moisture, such as hydrogels, risk becoming overly
PT
hydrated and may lose their adhesion to the surface. Non-specific bioadhesive polymers are
RI
moisture activated and can bind to both the mucosal layer and to cell surfaces (Roy et al.,
SC
2009; Smart, 2005).
NU
3 Mucoadhesive Polymers
A number of polymers, both natural and synthetic, have mucoadhesive properties without
MA
modification. These include chitosan, polyacrylic acid (PAA) and pectin.
3.1 Chitosan
D
Derived from the deacetylation of chitin (Makhlof et al., 2008), chitosan, Figure 1 (A), is the
PT E
most commonly used material for mucoadhesive polymers. It is a naturally occurring polysaccharide which is cationic and hydrophilic in nature. Cationic polymers are thought to
CE
be far superior mucoadhesives due to the interactions between the positive charge of chitosan
AC
and the negatively charged residues of the mucosal layer, such as sialic acid (Ludwig, 2005; Nagarwal et al., 2009). Chitosan has many advantages for use in drug delivery, including its biodegradability, biocompatibility and non-toxic properties (Ludwig, 2005). It is also seen to have antibacterial properties which is advantageous, particularly in ocular drug delivery where infections are common (Ludwig, 2005). As a polymeric delivery system, it is well tolerated in the body, highly stable, and has the capability to enter epithelial cells making it an excellent carrier for drugs. In oral delivery, chitosan is also seen to improve the delivery of hydrophilic 15
ACCEPTED MANUSCRIPT drugs, such as peptides and proteins, as it has the ability to open up tight junctions in the (Nagarwal et al., 2009)mucosal layer by affecting para- and intracellular pathways (actin pathways) (Dodane et al., 1999). It is available in a range of MW, with weights between 10 450 kDa being normal for drug administration (Makhlof et al., 2008). Chitosan is easily modified under mild conditions and is capable of incorporating macromolecules into the
PT
polymeric framework and is, therefore, useful for drug, protein and peptide delivery
RI
(Salamanca et al., 2006). It is insoluble at neutral and high pHs and soluble at low pHs. Once
SC
it is thiolated, it has cohesive properties and also displays gel-like properties in solution (Bernkop-Schnürch et al., 2004b).
NU
3.2 Polyacrylic acid (PAA)
PAA, Figure 1 (B), is an anionic polymer. The main method of adhesion of anionic polymers
MA
is due to hydrogen bonding between the polymer and mucins (Ludwig, 2005). Interactions of PAA with mucins are strongest at an acidic pH, indicating that the polymer in its protonated
D
state is best for mucoadhesion (Ludwig, 2005). This was documented further when the
PT E
addition of urea disrupted the hydrogen bonds between the polymer and mucus resulting in a reduction in mucoadhesion (Mortazavi, 1995). PAA is soluble across a broad range of pHs
CE
(Makhlof et al., 2008). It is not absorbed in the GI tract but it has been shown to have antienzymatic properties, particularly to the endopeptidase, trypsin (Luessen et al., 1995). This
AC
property is ideal for the delivery of protein and peptide drugs through oral administration as the polymer will protect the peptide drug from degradation. PAA is a commonly used polymer, and one of the first polymers used, in the production of thiolated polymers, or thiomers as they are also termed. The mucoadhesive properties of a thiolated polymer depend on the degree of swelling and of crosslinking of the polymer. The mobility and the flexibility of the polymer chain are also of great importance for mucoadhesion, as the chain will
16
ACCEPTED MANUSCRIPT influence the interactions between the polymer and the mucus. Crosslinking will reduce the flexibility of the chain and may have an effect on the mucoadhesive properties.
3.3 Pectin Similar to chitosan, pectin, as shown in Figure 1 (D), is a polysaccharide isolated from citrus peel or apples. This natural polymer has non-toxic properties and is often used within the
PT
food and pharmaceutical industries. Pectin is an anionic, complex heteropolymer which
RI
contains residues of galacturonate and rhamnose and has high levels of carboxyl groups along
SC
its backbone. Due to the levels of hydroxyl groups, pectin is also highly hydrophilic making it ideal for swelling in drug delivery (Pratt and Cornely, 2011).
NU
Modification of the pectin backbone often occurs through methoxylated or amidated carboxyl
MA
groups, and the degree of modification can allow for altered levels of mucoadhesion (Joergensen et al., 2011). Modification can occur under mild conditions due to pectin’s water
D
solubility. The MW of pectin has been proven to be important in mucoadhesion; using
PT E
pectins of varying MW ranging from 80 – 200 kDa and also with varying levels of modification, Thirawong et al. (2007) investigated the mucoadhesive properties of pectin using a texture analyser. Samples were classified as either high methoxy pectin or low
CE
methoxy pectin and one low methoxy pectin sample also had a degree of amidation. In
AC
general, the high methoxy pectin samples showed improved mucoadhesion in comparison to the low methoxy samples. The addition of amide groups to the low methoxy sample did also improve mucoadhesive properties. The results were in contrast to previous studies where low methoxy samples had higher mucoadhesive properties and it was suggested that the MW of the samples used by Thirawong et al. had a marked influence on the mucoadhesive properties of the samples; previous studies examined mucoadhesive properties of varying methoxylation but comparable MW (Liu et al., 2005). The highest level of methoxylation and the highest MW demonstrated the greatest mucoadhesive properties, thus showing the importance of 17
ACCEPTED MANUSCRIPT polymer properties in mucoadhesive drug delivery. Interestingly, the area of GI tissue that the samples were tested on was also of importance. Tissue was taken from the whole length of the GI tract, from the mouth to the large intestine, and pectin samples were tested on each section for mucoadhesion. Samples of chitosan and carbomer946P, known mucoadhesive polymers, were also tested as positive controls in the mucoadhesive testing. Each tissue
PT
sample was submersed in media relevant for the area, i.e. lower pH buffer for stomach tissue
RI
in comparison to intestinal tissue etc. Mucoadhesion was noted to be highest in the large
SC
intestine and there was little variability between the buccal, stomach and small intestine tested. It was thought that the variability in mucosal levels between the different areas of the
NU
GI tract played a marked role in the mucoadhesive levels of the samples; the large intestine has a high ratio of goblet cells resulting in high levels of mucus.
MA
3.4 Thiolated Polymers (Thiomers)
The introduction of compounds which contain free thiol groups to the polymer backbone
D
further increases mucoadhesion, allowing for the formation of strong disulphide bonds
PT E
between the modified polymer and the cysteine-rich mucin domains of the mucosal surface, prolonging adhesion of the polymer to the mucosal layer. Ideally for drug delivery systems
CE
and mucoadhesion, a thiol content value of at least 400 µmol/g of polymer is required, although mucoadhesion is improved with a higher level of thiolation. The degree of intra- and
AC
intercrosslinking depends on the amount of free thiol moieties present on the polymer backbone; the more thiol groups attached, the more cohesive and adhesive the polymer. However, unbound thiol will disrupt mucoadhesion, so it is imperative that molecular residues which possess free thiol groups be removed from the polymer prior to use. Leitner et al. (2003) investigated the mucoadhesive properties of thiolated PAA in comparison to a control PAA sample. All samples were dissolved in water with and without the addition of 1% cysteine (m/m), therefore, the affect unbound cysteine may have on mucoadhesive 18
ACCEPTED MANUSCRIPT properties of thiolated polymers was also examined. The samples were frozen and lyophilised. Mucoadhesive testing was conducted on the tabletted samples with and without the addition of cysteine and, as shown in Figure 2, the presence of unbound cysteine had a significant influence on the mucoadhesive properties of the thiolated sample, decreasing adhesion by up to 50% due to the interruption of the polymer-mucin disulphide bond. The
PT
addition of cysteine to the control sample had no effect on mucoadhesion, thus highlighting
RI
the importance of the removal of unbound thiols from the thiolated polymer matrix.
SC
Adhesion of the polymer to the surface is futile if there is no cohesion in the polymer itself (without the polymer itself binding together, it is unlikely to interact and bind with mucus).
NU
Thiomers show a high level of cohesion, due to the intramolecular disulphide bond formation. Bernkop-Schnürch and Steininger (2000) compared the cohesive properties of
MA
thiolated polycarbophil and carboxymethylcellulose and unmodified samples. It was shown that the disulphide bonds formed within the thiolated polymers created a much more effective
PT E
D
and beneficial delivery system than the unmodified versions. The hydrophobic and biodegradable polymer polylactic acid (PLA) is often modified to form a co-polymer with polyglycolic acid (PGA), poly(lactic-co-glycolic acid) (PLGA), Figure 1
CE
(C). Controlled release from PLA is problematic, and, similar to many biodegradable
AC
polymer matrices, appears to have a triphasic pattern of release, with an initial burst of drug once the polymer implant is administered into the body followed by a slow and sustained release. PLGA, itself, is not mucoadhesive; however, PLGA nanoparticles were synthesised and coated with known mucoadhesive polymers, either chitosan or PAA, to form mucoadhesive drug delivery systems (Kawashima et al., 2000). The nanoparticles were loaded with protein-based drugs and the mucoadhesive properties of the particles were tested on the intestinal tissue of a rat. The particles coated in chitosan showed improved mucoadhesion in comparison to those coated with PAA and it was surmised that the increase 19
ACCEPTED MANUSCRIPT in mucoadhesion was due to the electrostatic interactions between chitosan and the mucosal surface. Similarly, Grabovac and Bernkop-Schnürch (2007) coated PLGA nanoparticles with thiolated chitosan to increase its mucoadhesive properties further. The PLGA nanoparticles were initially coated with chitosan by reaction with EDC and the chitosan-coated particles were then thiolated using Traut’s reagent. Using a wash-away method on porcine intestinal
PT
tissue, the mucoadhesive properties of the nanoparticles were investigated, using both PLGA
RI
nanoparticles and the thiolated chitosan-coated PLGA nanoparticles. Over a three hour
SC
period, the percentage of PLGA nanoparticles adhering to the intestinal tissue decreased, with approximately 50 % detaching between hours one and two. The thiolated chitosan-coated
NU
PLGA particles did not show a significant decrease in detachment over the experimental period.
MA
The crosslinking action of thiolation changes the degree of swelling of the polymer, theoretically allowing for a more sustained and controlled release of drug. Thiolation will
D
prevent over hydration and change the degradation pattern of the polymer, particularly
PT E
important in hydrogels (Tabata and Ikada, 1998). However, high degrees of crosslinking may affect the mucoadhesive properties of the polymer, as the chain length and flexibility are
CE
altered. A commonly used molecule in the creation of a thiomer is L-cysteine (Figure 3 (A)).
AC
3.4.1 Thiolation with cysteine Leitner et al. (2003) investigated the properties of cohesion, mucoadhesion, swelling and disintegration on linear PAA-cysteine conjugates of different MW and, therefore, different chain lengths, ranging from 2 kDa PAA to 450 kDa PAA, and a polycarbophil-cysteine conjugate of MW 750 – 3000 kDa. All polymers had comparable levels of immobilised thiols to allow for a more direct comparison, and results were also compared to unmodified PAA controls. Mucoadhesive studies were analysed by two different methods: Tensile studies and a rotating cylinder method using porcine intestinal mucosa. Both methods concluded that 20
ACCEPTED MANUSCRIPT PAA with MW of 250 kDa and 450 kDa exhibited the highest mucoadhesive levels. In addition to these tests, the amount of residual unbound cysteine and its influence on mucoadhesion was measured using 450 kDa PAA (both thiolated and unmodified PAA). Free cysteine had little to no effect on the mucoadhesion of the unmodified PAA controls, however, the addition of 2.5% free cysteine showed a significant decrease in mucoadhesion
PT
time in comparison to 0% free cysteine on the thiolated PAA sample, again highlighting the
RI
need to completely remove any unbound cysteine during the creation of a mucoadhesive drug
SC
delivery system. Thiolated PAA 450 kDa sample showed the highest levels of mucoadhesion and excellent cohesion. Although the low MW PAA had strong penetration into the mucosa
NU
due to the flexibility of the chain, cohesive properties were poor and mucoadhesion was affected. Similarly, higher weighted polymers had better cohesion but poor penetration due to
MA
the lack of chain flexibility which, again, affected mucoadhesion. In order to evaluate the effect of varying amounts of L-cysteine bound to PAA, five different
D
PAA-cysteine conjugates were synthesised, each with a different amount of immobilised thiol
PT E
groups (Palmberger et al., 2007). Levels of thiolation (measured in µmol/gram of polymer) bound to the 450 kDa PAA ranged from 53 µmol/g up to 767 µmol/g. Various studies,
CE
including tensile and mucoadhesive studies, were performed. The degree of thiolation was
AC
measured using two methods: 1. Ellman’s reagent to measure the amount of free thiol groups on the PAA backbone and 2. TNBS (trinitrobenzene sulfonic acid) to measure the amount of remaining unbound cysteine in the reaction solution by measuring the amine groups on cysteine. The total amount of thiol groups on the polymer was also determined by reducing the disulphide bonds with sodium borohydride and then analysing with Ellman’s reagent. It was concluded that both mucoadhesion and disintegration of the conjugates strongly depended on the amount of thiol groups, with higher levels of thiolation resulting in longer times of mucoadhesion and disintegration. Mucoadhesion studies were all carried out at pH 21
ACCEPTED MANUSCRIPT 6.8 as intermolecular disulphide bonds between the polymer and the mucus can easily be formed at this pH. It was also discovered that without the addition of the coupling reagent, 1Ethyl-3-(3-dimethylaminoprpyl) carbodiimide hydrochloride (EDC), no amide bonds were formed between the polymer and the cysteine. Kafedjiiski et al. (2007) showed the significant difference between crosslinked polymers and
PT
non-crosslinked polymers in relation to rates of degradation and adhesion times. The polymer
RI
used was hyaluronic acid (HA) and was crosslinked using L-cysteine ethyl ester
SC
hydrochloride. Degradation of HA is due to hydrolysis by hyaluronidase. Both the thiolated and unmodified HA rates of degradation were slow but the thiolated polymer rate was
MA
process inhibited enzymatic hydrolysis.
NU
significantly slower, indicating that the crosslinked disulphide bonds formed in the thiolation
Grabovac et al. (2005) showed thiomers, in general, had higher mucoadhesive properties than
D
their equivalent unmodified polymers and that the mucoadhesion was pH dependent. At a pH
PT E
over 5.0, the thiol groups in the polymer were more reactive and began to form intramolecular disulphide bonds and in turn did not form them with the cysteine rich mucin domains. At pH 3.0, the thiols groups were much less reactive and formed bonds with the
CE
mucosal layer. Bernkop-Schnürch (2005) explained that the pH of the medium surrounding
AC
the surface of the polymer controlled the reactivity of the polymer as a whole, whereas the pH of the polymeric framework controlled the reactivity of the thiol groups. With regards to drug delivery systems, the pH of the epithelium would always be around pH 7 and the thiol groups would be reactive enough to penetrate mucus. Thiomers form inter- and intramolecular disulphide bonds at physiological pH due to the oxidation of the thiol groups, classified as in situ gelling properties. Polymers that demonstrate a phase transition, sol-to-gel transition, due to the change in the environment or physiochemical changes (pH, temperature) have been designed. Srividya et al. (2001) synthesised PAA based eye drops that were pH 22
ACCEPTED MANUSCRIPT triggered and changed phase (liquid to gel) once they had come in contact with the eye. The carrier drug used was ofloxacin, an ocular anti-bacterial agent usually given topically as an eye drop. PAA was used in this experiment due to its capacity to form a stiff gel at elevated pH. The trigger in this case was the change from the pH at formulation, pH 6, to the pH of the eye, pH 7.4. The irritation levels of the PAA formulation was measured using the Draize
PT
technique on rabbits. No ocular irritation was observed.
RI
Krauland and Bernlop-Schürch (2004) showed a 3-fold improvement in residence time of
SC
thiolated PAA, as shown in Figure 3 (B), versus unmodified PAA in the small intestinal mucosal layer. This study used insulin as the model peptide drug for delivery and combined a
NU
copolymer of Eudragit RS® (a copolymer consisting of ethyl acrylate and methyl methacrylate) into the thiolated PAA matrix. Previous studies using insulin and thiolated
MA
PAA without Eudragit RS® had observed that the insulin had burst-released within a number of minutes, before the polymer had adhered to the surface. A more controlled and prolonged
D
release of insulin was observed when Eudragit RS® was incorporated into the microparticles.
PT E
The delivery of insulin is a constant problem, with oral delivery of the peptide not being
injection.
CE
viable due to the denaturation of the protein. Current delivery is only by subcutaneous
AC
Grabovac et al. (2008) designed a three layered mucoadhesive patch for delivery of insulin by oral administration and mucoadhesion to the small intestine. The mucoadhesive patch consisted of a polycarbophil-cysteine matrix within which was a mixture of insulin, mannitol and gluthatione, all dissolved in water. Due to the small surface area of the patch and low levels of water uptake and, therefore, limited swelling of the polymer at the site of action, a single dosage of insulin was created by sticking two separate polymeric matrices together which were then surrounded by ethylcellulose and then Eudragit, which acts as an enteric coating and has a pH threshold of 5.5. Plasma insulin levels and blood glucose levels over 23
ACCEPTED MANUSCRIPT time were examined in three separate cohorts of rats; one cohort receiving the polymeric patch, another had subcutaneous delivery of insulin and finally, oral administration of an insulin solution used as the control group. The polymeric matrix showed a more sustained insulin level in the blood plasma, which is to be expected as the insulin was released in a more controlled manner in comparison to the subcutaneous injection. After oral
PT
administration of insulin solution to the control group, there was no change in the insulin
RI
levels in the blood. This shows the efficiency of the mucoadhesive polymeric system.
SC
Chondroitin sulphate (CS) is a glycosaminoglycan found in bone and cartilage tissues. CS is commonly used in the treatment for joint pain and osteoarthritis, given as an oral dosage or
NU
injection to the site of pain. It was thought that prolonging the residence time of CS at the site of action would be beneficial, improving the anti-inflammatory and chondroprotective
MA
properties of this nutraceutical. Therefore, CS was thiolated using L-cysteine and EDC (Suchaoin et al., 2016a). The resulting thiolated CS sample yielded a thiol content of 421.17
D
± 35.14 µmol/g; the sample was then reduced using Tris (2-carboxyethyl) phosphine
PT E
hydrochloride (TCEP HCl) to break any disulphide bond formed during the reaction process and this sample was then dialysed and freeze dried, thus yielding a reduced thiol content of
CE
675.09 ± 39.67 µmol/g. The reduced thiolated CS sample was used throughout the remaining
AC
experiments. The rotating cylinder method on porcine tissue and rheological studies were utilised to investigate the mucoadhesive properties of the thiolated CS sample. Using the rotating cylinder method, thiolated CS showed a 2.48-fold adhesion improvement in comparison to an unmodified CS control, and it was noted that the unmodified CS detached from the intestinal tissue in one piece, whereas the thiolated CS sample gradually dislodged overtime. A CS control, mucin control and thiolated CS mixed with mucin were investigated for their viscosity levels, and a 4.57-fold increase was observed in the thiolated CS/mucin mix in comparison to the mucin solution alone, indicating the mucoadhesive interactions 24
ACCEPTED MANUSCRIPT occurring between the thiolated CS and mucin. It was thought that, although mucoadhesion was improved due to thiolation, higher levels of mucoadhesion would be observed with cationic polymers due to the charge interactions occurring between the negative residues of the mucosal layer and of the CS backbone.
PT
3.4.2 Thiolation with Traut’s reagent (2-iminothiolane) Traut’s reagent, Figure 4 (A), reacts with primary amines groups on the polymer backbone in
RI
a ring opening reaction, yielding free thiol groups. Bernkop-Schnürch et al. (2003) used Traut’s reagent to thiolate chitosan. Chitosan had previously been thiolated using thioglycolic
SC
acid, creating a thiomer with an uncharged amide bond linkage (Kast and Bernkop-Schnürch,
NU
2001); by thiolating with Traut’s reagent, a cationic linkage bond was formed which had the potential to form ionic interactions within the mucosal membrane with residues such as sialic
MA
acid or sulfonic acid, as well as the introduction of thiol groups to the polymer backbone. The reaction of chitosan with Traut’s reagent was conducted at three different pH values: pH 5,
D
pH 7 and pH 10. Chitosan was insoluble at pH 10, but a certain degree of derivatisation was
PT E
observed and a thiol content of 177.2 µmol/g was measured at this pH with the addition of 0.2 g of Traut’s reagent to 1 g of chitosan. When the same ratio of Traut’s reagent was added
CE
at pH 5, although chitosan was fully dissolved at this pH, a lower level of thiol content was observed, measuring 59.8 µmol/g; this was due to lower reactivity of Traut’s reagent at this
AC
pH. Conducting the reaction at pH 7 yielded the highest level of thiol content of the chitosan backbone, with a degree of thiolation of 225.7 µmol/g measured. When the ratio of Traut’s reagent was increased to 0.4 g/g of chitosan, a thiol content of 408.9 µmol/g was measured at pH 7. Mucoadhesive testing was conducted on porcine mucosal tissue using the rotating cylinder method and compared the mucoadhesive properties of the chitosan-Traut’s reagent conjugate samples to that of unmodified chitosan and chitosan-thioglycolic acid conjugate samples. Mucoadhesion time was increased by up to 140-fold after thiolation with Traut’s
25
ACCEPTED MANUSCRIPT reagent, which was thought to be due to the improved cationic character of the polymer, as well as the presence of thiol groups on the chitosan backbone. 3.4.2.1 Polyallylamine Polyallylamine is a synthetic, cationic polymer, shown in Figure 1 (F) as (i) its free base form, and (ii) its salt form. It is a linear polymer with primary amines along its backbone.
PT
Polyallylamine hydrochloride (PAH) is commonly used in the treatment of
RI
hyperphosphatemia in patients with end-stage renal failure. Crosslinked with epichlorohydrin
SC
and known in this form as sevelamer hydrochloride, it acts as a phosphate binder, removing excessive phosphate ions as it is excreted (Hudson et al., 2012). In this form, it is non-
NU
biodegradable but is not absorbed into the blood stream. It is also excreted unchanged and therefore, is not metabolised to toxic by-products or metabolites. In the form of a polymeric
MA
micelle, PAH was used as a delivery system to the GI tract for siRNA (small interfering RNA) (Guo et al., 2013). Polymeric micelles are emerging as a platform in the oral delivery
D
of gene based drugs as the drug is protected in the hydrophobic centre of the micelle. PAH
PT E
was modified with cholesteryl and palmitoyl moieties and also a hydrophilic ammonium moiety, creating a so called quaternised PAH product. This product contained a hydrophobic
CE
interior and hydrophilic exterior, which was thought to be more stable in an acidic environment such as the stomach; amine groups on an unmodified PAH sample would be
AC
affected greatly by this acidic environment. siRNA bound to the PAH by electrostatic interactions and its stability in simulated GI fluids was measured. Stability of siRNA in the quaternised PAH was improved in both simulated gastric fluid and simulated intestinal fluid in comparison to an unmodified PAH system, implying the protective nature of this modified PAH polymeric micelle. Although not analysed in this paper, the quaternised PAH was also seen to interact with the mucosal layer in the GI tract, again through electrostatic interactions between the polymer and mucins. This resulted in increased residence and contact time.
26
ACCEPTED MANUSCRIPT Cationic polymers are often used as vectors for gene transfer and polyallylamine is one such polymer that is utilised. Polyallylamine, however, can be toxic to cells but modification of the backbone has been observed to decrease the toxic effects (Oskuee et al., 2015) (Boussif et al., 1999). Bacalocostantis et al. (2012) thiolated polyallylamine (PAAm) with Traut’s reagent, the proposed structure of which is given in Figure 4 (B). Different ratios of Traut’s reagent
PT
were added to the reactions to result in an overall 5%, 13%, or 20% conversion of amines to
RI
thiol groups. The resulting thiolated polymer was used as a delivery vector for DNA in which
SC
the cationic domains of the polymer were electrostatically bound to anionic charged phosphate groups on DNA. Each thiolated polymer and unmodified polymer was tested for
NU
DNA binding and release rates, and protection of DNA. The 13% conversion rate showed the greatest potential, with higher DNA binding ability and better release of plasmid in
MA
comparison to both unmodified polyallylamine and to the 5% and 20% counterparts. Ibie et al. (2015) thiolated polyallylamine with Traut’s reagent and compared the thiol
PT E
D
content and mucoadhesive properties to a polyallylamine sample which was thiolated with Nacetyl cysteine/EDC and used PAH in both thiolation reactions. The pH of the PAH used was, however, adjusted with NaOH and dialysed to yield the free base form of
CE
polyallylamine, PAAm, which was then thiolated with both Traut’s reagent and N-acetyl
AC
cysteine. Similar to results observed by Vigl et al. (2009), the thiol content of the resulting samples thiolated with N-acetyl cysteine were low, measuring 60 µmol/g which was in vast contrast to the greatly improved thiolation levels observed with Traut’s reagent, which resulted in a thiol content of 490 µmol/g. Disulphide bond formation of the samples was also examined. The samples displayed high levels of disulphide bond formation, with samples thiolated with N-acetyl cysteine resulting in disulphide bond levels of 280 µmol/g and reaction with Traut’s reagent measuring 590 µmol/g formed as disulphide bonds. Mucoadhesion testing was conducted using a porcine mucin solution; solutions of both the 27
ACCEPTED MANUSCRIPT thiolated polymers and unmodified polyallylamine were mixed with the mucin solution and tested by UV/Vis spectrometry to determine the levels of adsorption of mucin to the polymer. Thiolation of the polymer displayed improved mucin interactions in comparison to unmodified samples. The percentage of mucin adsorption increased from approximately 40%
PT
in the unmodified PAAm sample to 65 – 70% in the thiolated PAAm samples. PAAm (15 kDa MW) was thiolated with varying ratios of Traut’s reagent, creating three
RI
thiolated PAAm products (Duggan et al., 2016a). Thiol content of the three products were
SC
134 µmol/g, 329.5 µmol/g and 487.4 µmol/g which equated to 1%, 2.5% and 5% total thiolation of the PAAm backbone, respectively. The three thiolated products were tested for
NU
their mucoadhesive capabilities by investigating their swelling ability, mucoadhesion on
MA
porcine tissue and rheological properties with and without the addition of a mucin solution. The swelling abilities of the three thiolated PAAm samples were fast in the initial minutes of the test, with high levels of fluid uptake, up to 1500% in the first five minutes of the swelling
PT E
D
studies. Surprisingly, the swelling abilities of the thiolated PAAm samples increased with increasing thiol content, and therefore, the 5% thiolated PAAm sample displayed the greatest degree of swelling. It was thought that a higher level of disulphide bond formation was
CE
occurring in the less thiolated sample (1% thiolated PAAm) due to steric hindrance within the
AC
polymer. Mucoadhesive testing on porcine tissue using the rotating cylinder method was investigated and, interestingly, the mucoadhesive properties, regardless of the thiol content, were similar and all three thiolated samples adhered for a minimum of 22 h. The thiolated PAAm samples were also tested for their rheological properties, both alone and mixed with an 8 % mucin solution. Although the mucoadhesive properties using the rotating cylinder method were similar, the rheological studies showed that the increase in thiol content increased the relative G', by approximately 3-fold, indicating that higher levels of interactions
28
ACCEPTED MANUSCRIPT were occurring within the 5% thiolated PAAm sample with the mucin solution in comparison to the other two thiolated PAAm samples. A further study into the properties of thiolated PAAm (15 kDa MW) was conducted by comparing the swelling abilities and mucoadhesive properties of a thiolated PAAm sample
PT
(thiol content of 487.4 ± 18.3 µmol/g) to that of thiolated PAA with comparable thiol content (400.6 ± 61.3 µmol/g) (Duggan et al., 2016b). The PAAm sample was thiolated using Traut’s
RI
reagent while the PAA sample was thiolated using L-cysteine and EDC. The thiolation of
SC
PAAm improved the swelling properties of the polymer in comparison to a control sample, although the swelling abilities of the thiolated PAAm were far greater than that of thiolated
NU
PAA, particularly within the initial minute of testing. The release of the model drug,
MA
chlorpheniramine maleate, was influenced by this increased swelling in the thiolated PAAm sample and drug release was quicker in comparison to the thiolated PAA sample. However, despite greater swelling rates, the mucoadhesive properties of the two thiolated polymers
PT E
D
were comparable, as measured by the rotating cylinder method using porcine intestinal tissue, with the thiolated PAAm samples adhering for 35 ± 18 h and the thiolated PAA adhering for 23 ± 1 h. Comparable mucoadhesive properties were also observed using rheological
AC
3.4.2.2 Gelatin
CE
methodologies when the polymers were mixed with a mucin solution.
Gelatin is a natural polymer derived from Type 1 collagen (which contains no cysteine) of animals and fish. Gelatin is hydrophilic, is degraded enzymatically and, as it is a natural polymer, it does not induce an immune response, therefore, making gelatin an excellent component for potential drug delivery applications. Depending on the pre-treatment process from the collagen, two different forms of gelatin can be produced: acidic (Type A) or basic (Type B). Each process creates a gelatin form with a different isoelectric point (IEP), the point at which a molecule carries no net charge. This varying IEP permits for a greater 29
ACCEPTED MANUSCRIPT flexibility when it comes to drug delivery systems, allowing the gelatin carrier to be of opposite charge to the drug and for the potential formation of a polyion complex, a strong bond formed due to electrostatic interactions between oppositely charged entities (Young et al., 2005; Tabata and Ikada, 1998). The amino acid composition of native gelatin is 16% carboxylate and 13% primary amine. It
PT
has a general structure similar to collagen containing repeating sequences of glycine–X-Y
RI
triplets, where X is proline and Y is often hydroxyproline, as shown in Figure 1 (E). Due to
SC
the presence of these glycine–X-Y triplets, gelatin can exist in a triple helical structure as its most stable conformation (Pratt and Cornely, 2011). As there are both free carboxylate and
NU
amine groups throughout the gelatin backbone, it offers many routes of modification to improve and create its mucoadhesive properties. Gelatin dissolves in aqueous solutions at
MA
approximately body temperatures (37 °C), forming single, flexible coils and becoming transparent upon cooling (Vlierberghe et al., 2011). Because of this, gelatin must be
D
chemically crosslinked or modified in order to become insoluble at body temperature
PT E
(Dinarvand et al., 2005). Gelatin is easily modified under mild conditions, which is ideal for drug delivery carriers. Mild conditions can also be used to load the drug into the polymer
CE
matrix (Young et al., 2005). However, when a protein based drug is present during the
AC
production of gelatin hydrogels, the proteins are seen to lose pharmacological activity due to the chemical crosslinking required in the creation of the hydrogel (Tabata and Ikada, 1998). This problem can be avoided by allowing the protein to absorb into the hydrogel postproduction. A gelatin based mucoadhesive delivery system was devised for the delivery of amoxicillin to the gastric mucosa, targeting the bacteria, Helicobacter pylori, the main cause of ulcers and gastritis (Wang et al., 2000). A constant level of antibiotic in the gastric mucosa is required to treat the condition. This study compared the mucoadhesive and drug release properties of 30
ACCEPTED MANUSCRIPT gelatin microspheres versus modified gelatin microspheres which were firstly aminated with ethylene diamine and secondly crosslinked using varying levels of gluteraldehyde (0.06%, 0.12%, 0.18%). Different crosslinking times were also measured. The unmodified gelatin microspheres had an initial burst, releasing up to 80% of its drug load within 30 min. The varying levels of crosslinked microspheres showed decreased bursts within the same time
PT
period. The highest level of crosslinkage, 0.18% gluteraldehyde, had a slower and more
RI
controlled release of amoxicillin but the increased levels of crosslinkage affected the
SC
mucoadhesion, with adhesion levels being similar to that of the unmodified gelatin. Highest levels of mucoadhesion were observed with 0.06% gluteraldehyde. This indicates that a
NU
compromise between levels of crosslinkage was required to establish an efficient mucoadhesive system which will allow for slow, controlled release.
MA
In terms of thiolation, Kommareddy and Amiji (2005) thiolated gelatin nanoparticles using Traut’s reagent for use as a non-viral delivery vector for DNA. Thiolation, in this case, was
D
used as a means of stabilising the nanoparticle; the disulphide bonds formed due to oxidation
PT E
in the body strengthened the gelatin’s structure. Upon entering a reducing environment, these disulphide bonds would break and the drug would be delivered. Because of this theory, the
CE
delivery of the model drug, fluorescein isothiocyanate dextran, was tested in the presence of
AC
the reducing agent glutathione. Additionally to this, plasmid DNA was inserted into the nanoparticles. The encapsulation efficiency and stability were measured as well as transfection into murine fibroblast cells. In cytotoxicity testing, the gelatin nanoparticles were shown to be non-toxic with up to 92% cell viability seen at high concentrations; it was observed that the higher levels of thiolation resulted in higher levels of cytotoxicity. Varying levels of thiolation were analysed for both drug release and plasmid DNA encapsulation and they were compared to an unmodified gelatin nanoparticle. Additionally to this, nanoparticles were either crosslinked or non-crosslinked. In the case of loading efficiency of fluorescein 31
ACCEPTED MANUSCRIPT isothiocyanate dextran, the thiolated nanoparticles were seen to have a higher loading capacity than the unmodified counterparts. Release was measured over a 5 h period and rates varied depending on the amount of glutathione added, with higher amounts of glutathione resulting in enhanced release when compared to release in PBS. Thiolation levels affected the percentage release, depending on the concentration of glutathione added, with unmodified
PT
particles releasing the drug much quicker. Similarly, crosslinking of the particles resulted in
RI
slowing the release of the drug in comparison to non-crosslinked particles, regardless of the
SC
amount of glutathione added. Thiolation levels did, however, affect the loading of plasmid DNA into the nanoparticles; unmodified gelatin particles had a loading capacity of 99% but
NU
this figure fell to 90% with thiolated nanoparticles of thiol content 16.41 mM/g gelatin. The plasmid DNA did remain stable within the thiolated particles and were comparable to the
MA
unmodified particles and also to naked plasmid DNA.
Vlierberghe et al. (2011) compared two different thiolating agents, N-acetylhomocysteine
D
thiolactone and Traut’s reagent, in their reaction with gelatin. The main body of analysis in
PT E
this paper was conducted using N-acetylhomocysteine thiolactone, however, thiolation levels using both compounds, N-acetylhomocysteine thiolactone and Traut’s reagent, were
CE
compared. Both thiolated gelatin products were analysed using UV/Vis spectrometry and size exclusion chromatography. Thiolation levels were analysed using both Ellman’s reagent and
AC
using the ortho-phthalic dialdehyde (OPA) method; Ellman’s reagent analysed the thiol content whereas the OPA method measured the free amine content, which when compared to unmodified gelatin, gave the degree of substitution. The OPA method indicted higher levels of substitution being present than the Ellman’s method, implying the presence of disulphide bonds in the product. When one equivalent of Traut’s reagent was added to the reaction, 70% degree of modification was achieved. This is in comparison to only 30% with Nacetylhomocysteine thiolactone. In a subsequent part of the experiment, the product thiolated 32
ACCEPTED MANUSCRIPT with N-acetylhomocysteine thiolactone was examined. In the form of a hydrogel, and thiolated at different levels of thiolation, various factors including swelling tests, rheology and texturometry were investigated. Rheological results showed that the storage modulus (G') was higher in the thiolated samples in comparison to unmodified gelatin and that as thiol content increased, so did G'; this was due to the ability of the thiolated samples to crosslink
PT
and form a gel. When comparing G' at different temperatures, room temperature (21 °C) and
RI
50 °C which is the temperature above the sol-gel transition of gelatin, it was shown that
SC
disulphide bond formation contributed alone to G' at 50 °C whereas at 21 °C, the physical properties of gelatin itself also contributed. Swelling properties decreased as the thiol content
NU
increased showing that crosslinking of the polymer influences its swelling ability. Gelatin was thiolated for use as a mucoadhesive drug delivery device using a two-step
MA
reaction process (Duggan et al., 2015). Native gelatin has a high carboxylate content and, in the initial reaction step, unmodified gelatin was reacted with ethylene diamine mediated by
D
an EDC coupling reagent to convert the carboxylic acids along its backbone to amide bonds
PT E
which contained primary amines as side chains, thus increasing the total amine content of the gelatin. In the second step of the reaction process, this aminated gelatin was then thiolated
CE
using Traut’s reagent, yielding a highly thiolated gelatin product. Due to the inherent amine
AC
content of gelatin, it was also directly thiolated with Traut’s reagent which, along with unmodified native gelatin, acted as controls. The swelling abilities and cohesive of the thiolated samples was vastly improved in comparison to control samples and the mucoadhesive properties, as measured by the rotating cylinder method on porcine intestinal tissue, were also markedly increased with up to a 10-fold increase in mucoadhesion upon thiolation using this two-step reaction (Figure 5). A range of gelatin samples with varying MW’s were thiolated using this two-step method and it was noted that MW had an important effect on both the cohesive and mucoadhesive properties of the thiolated gelatin samples 33
ACCEPTED MANUSCRIPT with, interestingly, a gelatin of MW of 20 – 25 kDa displaying the highest levels of cohesion and mucoadhesive properties.
3.5 Other methods of thiolation Polyvinyl alcohol (PVA) is commonly used in pharmaceutical formulations but does lend itself to short residence time after administration. Upon thiolation, this residence time could
PT
be improved. PVA was thiolated using two different methods of thiolation: by reaction with
RI
thiourea, and further neutralisation and acidification to yield the thiolated product, and also
SC
by reaction with 3-mercaptopropionic acid (Suchaoin et al., 2016b). Thiol content of the sample thiolated with thiourea (described as PVA1) measured 130.44 ± 14.99 µmol/g and it
NU
was observed that the reduction using NaBH4 was required as without reduction, the sample was not soluble due to the high levels of crosslinking forming in the sample. The sample
MA
thiolated with 3-mercaptopropionic acid (described as PVA2) yielded a thiol content of 958.35 ± 155.27 µmol/g. In mucoadhesive testing, both thiolated PVA samples displayed
D
higher levels of mucoadhesion in comparison to an unmodified PVA sample; PVA2 showed
PT E
longer adhesion times on porcine tissue, with an increase of 15.8-fold in comparison to the unmodified sample. This was higher than PVA1, which displayed a 3.2-fold increase and
CE
may be due to the increased free thiol content of the PVA2 sample. The swelling abilities of
AC
the two thiolated PVA samples were improved in comparison to the unmodified PVA sample, where both thiolated samples displayed a prolonged disintegration time and slower buffer uptake. The unmodified and thiolated samples were also tested for their drug release capabilities using the model drug, irbesartan, a drug to treat hypertension. Interestingly, there was no marked difference in the drug release profiles of the PVA1 sample versus the unmodified PVA sample. The PVA2 sample, with its higher free thiol content, displayed a slower release of drug, with a 0.6-fold decrease in drug release in comparison to the unmodified sample over a period of three hours. Additionally, the thiolated PVA exhibited no 34
ACCEPTED MANUSCRIPT cytotoxic effects of Caco-2 cells; after 24 h, unmodified PVA was the most cytotoxic, with PVA1 the least cytotoxic, displaying 72.8 % cell viability over the same time period. It was thought the addition of mercaptopropionic acid, and the higher thiol content, to PVA2 may have affected the mitochondria of the cells, resulting in higher cytotoxicity in comparison to PVA1. Thiolated PVA overall displayed lower levels of cytotoxicity, improved
PT
mucoadhesive properties and a slower release of drug over a period of time in comparison to
RI
an unmodified PVA sample.
SC
Carrageenan was also thiolated with thiourea (Suchaoin et al., 2015). The natural polysaccharide, which is derived from seaweeds, is known to have antiviral and antitumor
NU
growing properties. The thiolation of carrageenan could improve its residence time which, for
MA
deliveries of, for example, vaginal or nasal of antiviral medication, may be advantageous. A two-step reaction process was performed to introduce free thiol groups to the polysaccharide: bromination, followed by thiolation with thiourea. Carrageenan can be defined into three
PT E
D
groups, depending on gelling properties and position of the suphate group: λ, κ and ι; both κ and ι carrageenans were evaluated in this study. The thiol content of the κ and ι carrageenans measured 176.57 ± 20.11 and 109.51 ± 18.26 µmol/g, respectively. With the increase in thiol
CE
content, mucoadhesive properties of both the thiolated carrageenan samples were improved in
AC
comparison to unmodified samples; the κ carrageenan displayed a 6.4-fold increase in adhesion time on porcine intestinal tissue using the rotating cylinder method, while the ι carrageenan displayed a 1.8-fold increase using the same test. These mucoadhesive properties were further evaluated by rheology, in which the thiolated ι carrageenan, when mixed with a mucin solution, had an increase of 2-fold in its dynamic viscosity whereas the κ carrageenan displayed a 3.9-fold increase when compared to an unmodified carrageenan-mucin mixture. Thiogylcolic acid (TGA), with the addition of EDC, has also been used in the thiolation of chitosan (Kast and Bernkop-Schnürch, 2001). The reaction was conducted at a number of 35
ACCEPTED MANUSCRIPT different pH values ranging from pH 3 – 6 and also with different weight ratios of chitosan:TGA to investigate possible influences on the thiolation reaction. pH was shown to have a strong impact on the reaction, as below pH 5 the EDC was not as efficient and above pH 5, disulphide bond formation was higher. The different chitosan:TGA weights ratios at pH 5 also showed some interesting results: the highest degree of thiolation was observed at a
PT
1:1 ratio of chitosan:TGA. When the ratio was increased to 2.5:1 chitosan:TGA, the thiol
RI
content decreased marginally, whereas when the ratio was decreased to 1:2 chitosan:TGA,
SC
there was an approximate 50% decrease in thiol content. Swelling studies were conducted on unmodified chitosan samples, on control samples which had been treated with TGA but
NU
without the addition of EDC, and on the thiolated chitosan-TGA conjugates. The swelling study results showed no significant difference in swelling abilities between the thiolated
MA
chitosan sample and the control sample; there was, however, a difference between the control sample and the unmodified chitosan sample, and it was suggested that the change in pH
D
during the dialysis process (conducted in 5 mM HCl and final addition of 1% NaCl) on the
PT E
control sample altered the conformation of chitosan in comparison to the unmodified and undialysed chitosan sample, thus changing its swelling profile. In mucoadhesive testing, both
CE
the control and the unmodified chitosan samples demonstrated similar mucoadhesive properties. Thiolated chitosan samples showed higher levels of mucoadhesion in comparison
AC
to the control and unmodified samples and an increase in degree of thiolation increased the levels of mucoadhesion. This was displayed by both the rotating cylinder method and by tensile strength tests. Similar to PAA and chitosan, pectin is often thiolated to improve its mucoadhesive properties. Sharma and Ahuja (2011) thiolated pectin using thioglycolic acid and examined the thiolated pectin product for its mucoadhesive properties. Beads of both thiolated and unmodified pectin were synthesised and were loaded with the drug, metformin. It was 36
ACCEPTED MANUSCRIPT observed that thiolation improved the mucoadhesive nature of pectin by up to 2-fold. Pectin, due to the presence of hydroxyl groups, is naturally mucoadhesive with electrostatic interactions occurring with the mucosal layer and, in a wash-off bioadhesive test on goat intestinal tissue, the unmodified pectin beads remained adherent for up to 2.5 h. This is in comparison to the thiolated beads which remained adherent for up to 5 h using the same test
PT
procedure. Similarly, in the swelling tests, the thiolated beads had a lower rate of swelling in
RI
comparison to the unmodified samples and it was also observed that the unmodified samples
SC
eroded at a faster rate than the thiolated samples. Therefore, the lower levels of adhesion observed by the unmodified pectin beads were deemed to be due to increased swelling ability
NU
and faster erosion in comparison to the thiolated samples. Interestingly, in drug release studies of metformin, the release profiles of both the unmodified and thiolated beads were
MA
comparable. Therefore, the improved swelling ability of the thiolated sample did not result in a more controlled release of drug over time.
D
3.6 Preactivated and S-protected polymers
PT E
More recent advances in thiomer synthesis include the development of so called preactivated and S-protected thiomers. The concept of preactivated thiomers using PAA (Iqbal et al.,
CE
2012a) came to light due to the oxidation of thiols at pHs above pH 5, and since then, the mucoadhesive properties of many other preactivated and S-protected polymers have been
AC
investigated including chitosan, pectin, crosslinked PAA, alginate and gelatin (Dünnhaupt et al., 2012; Hintzen et al., 2013; Bonengel et al., 2014; Hauptstein et al., 2015; Rohrer et al., 2015). The preactivation of thiomers uses theory based on covalent chromatography where proteins were successfully linked to thiol-bearing resins once they were preactivated with pyridyl structures (Iqbal et al., 2012a; Brandt et al., 1977). The preactivation of the thiol group means it was essentially protected by a non-toxic aromatic leaving group attached by a disulphide bond, protecting the thiol from oxidation. This bond was broken by a disulphide 37
ACCEPTED MANUSCRIPT exchange between the thiol groups of the mucosal layer and the polymer thus releasing a free thiol which binds to the mucosal layer, improving mucoadhesion. The oxidation effect of thiols was seen to lessen the mucoadhesive properties of certain thiomers, therefore, protecting the thiol groups of the polymers allows for greater stability over a broader range of pH. The structure of preactivated PAA is shown in Figure 6.
PT
Iqbal et al. (2012a) investigated the preactivation of PAA, using three different MW PAA:
RI
100, 250 and 450 kDa and compared properties of hardness, swelling and mucoadhesion to
SC
thiolated PAA (thiolated with L-cysteine and EDC) and unmodified PAA over the same MW ranges. Tablet hardness was conducted at a force of 10 kN; the thiolated PAA samples
NU
showed an improvement in hardness of up to 5.4-fold in comparison to the unmodified PAA, but the preactived samples improved by up to 6.8-fold. In swelling studies, all samples were
MA
tested in both 0.1 M HCl and in 0.1 M phosphate buffer. All samples were observed to increase to a higher degree in the phosphate buffer in comparison to the HCl solution, and a
D
higher degree of swelling was also observed using the 450 kDa PAA sample, for the
PT E
unmodified, thiolated and preactivated samples. Overall, the preactivated PAA samples showed the highest degree of swelling over the longest period of time. In mucoadhesive
CE
testing using the rotating cylinder method on porcine intestinal tissue, the preactivated
AC
samples displayed the highest levels of mucoadhesion with a 960-fold increase in mucoadhesive time using the 450 kDa PAA in comparison to the unmodified PAA sample; a 452-fold improvement observed by the thiolated PAA sample. Higher levels of interactions with mucin glycoproteins were also observed with the preactivated thiomers, which was due to the presence of hydroxyl groups on the preactivated thiomers. This was shown through rheological testing where higher levels of apparent viscosity in preactivated polymer and mucin mixtures were measured in comparison to both the unmodified, and thiolated (non-
38
ACCEPTED MANUSCRIPT preactivated) polymer blends. Overall, enhanced stability of preactivated PAA blends was observed in comparison to unmodified and thiolated PAA. A similar study for vaginal delivery using preactivated chitosan thiolated with thioglycolic acid was conducted (Friedl et al., 2013). Thiolated chitosan and preactivated chitosan samples were compared to commercially available products in terms of their mucoadhesion
PT
properties over a 24 hour period. An in-vitro test model of the vagina was created using
RI
bovine vaginal tissue and the mucoadhesive properties were tested using a wash-off test using
SC
an artificial vaginal fluid, pH 4.5. The thiolated and preactivated chitosan formulations showed increased mucoadhesion in comparison to the commercial available products, with
NU
the preactivated samples displaying a 1.5-fold increase in mucoadhesion after 24 h in comparison to the most adhesive commercial product. Tensile strength tests were also
MA
performed on the commercially available and thiolated products. Again, the preactivated chitosan formulation displayed the highest total work of adhesion, indicating high degrees of
D
mucoadhesion. The use of thiolated polymers for drug delivery to more acidic regions of the
PT E
body, gastric or vaginal, has also been impeded due to the reactivity of the thiol group in acidic pH, lessening mucoadhesion.
CE
S-protected pectin was examined and compared to unmodified pectin and thiolated pectin for
AC
its mucoadhesive properties (Hintzen et al., 2013). To create the S-protected pectin compound, a 2-mercaptonicotinic acid dimer was reacted with L-cysteine through a disulphide exchange. The product was purified through filtration and an ion exchange chromatography to remove any unbound L-cysteine or L-cystine and was reacted, in the presence of EDC, with pectin to form the S-protected pectin product. Thiolated pectin was also thiolated with EDC and L-cysteine. The thiolated and S-protected samples were tested for their disintegration rates in a pH 1 HCl media and a pH 6.8 buffer, and compared to that of unmodified pectin. The unmodified pectin sample quickly dissolved in the pH 6.8 buffer 39
ACCEPTED MANUSCRIPT whereas the thiolated sample, due to the formation of disulphide bonds, remained stable at this pH. In both pH solutions, the S-protected pectin displayed decreased rates of disintegration. Swelling studies of the three samples also showed the greater ability of the Sprotected pectin sample, with a higher capacity to swell and greater cohesive properties. The mucoadhesive properties, as measured by the rotating cylinder method, were also increased in
PT
the S-protected pectin sample in comparison to both the thiolated and unmodified sample,
RI
although it was suggested that this increase was due to the more cohesive nature of the S-
SC
protected, and indeed the thiolated pectin, due to slower rates of disintegration. Interestingly, the S-protected sample was tested for its long term stability after 6 months, comparing the
NU
results to that of 1 day. There was no formation of intra- or inter-molecular disulphide as measured by the viscosity levels, which would increase with their formation. A non-
MA
significant decrease was observed in 2-mercaptonicotinic acid dimer and disulphide bond content, but free thiol groups, as measured by Ellman’s reagent, were not observed.
D
Gastric drug delivery with preactivated thiomers was examined in greater detail using pectin
PT E
(Hauptstein et al., 2013). Pectin was initially thiolated with cysteine and was then activated with dimers of 2-mercaptonicotinic acid to create the preactivated thiomer product. Swelling,
CE
disintegration and mucoadhesive properties were all improved in comparison to the
AC
unmodified polymer. Mucoadhesion was tested in both acidic and basic conditions (pH 6.8 buffer), simulating gastric and intestinal adhesion. Mucoadhesion was vastly improved in the acidic conditions using the preactivated polymer over the thiolated counterpart, due to the increased stability of the preactivated thiomer at lower pH. In basic conditions, the preactivated samples increased in mass to a greater extent than in comparison to the acidic conditions, with an increase in mass of 16 – fold in acidic conditions, and a 22 – fold increase in basic buffer, thus highlighting the higher cohesive properties in the acidic buffer, but also displaying the ability of the preactivated pectin sample to swell, an important feature for drug 40
ACCEPTED MANUSCRIPT release in a mucoadhesive tablet. In drug release studies of the model drug, rosuvastatin calcium, the acidic conditions displayed a slower and more controlled release from the preactivated sample in comparison to release in pH 6.8 buffer, with 50% release occurring after 10 h in acidic conditions but after 4 h in the pH 6.8 buffer. The modification of the pectin backbone, by thiolated or by preactivation, showed no significant changes to the low
PT
cytotoxic properties of unmodified pectin, as tested by resazurin assay.
RI
Gelatin was also modified to form a preactivated gelatin product by initially thiolating a
SC
native gelatin sample (50 kDa) with Traut’s reagent and reacting it further with a mercaptonicotinic acid (MNA) dimer (Rohrer et al., 2015). Gelatin thiolated with Traut’s
NU
reagent yield a thiol content of 243 ± 19 µmol/g, and once it was reacted with MNA dimer, the thiol content decreased to 53 ± 6 µmol/g, indicating a successful attachment to the free
MA
thiols of the gelatin-Traut’s reagent complex. Both tensile strength tests and rotating cylinder tests were used to identify the mucoadhesive properties of the preactivated gelatin sample and
D
results were compared to an unmodified gelatin sample and the thiolated gelatin-Traut’s
PT E
reagent sample. The tensile strength tests were conducted on porcine intestinal tissue; the preactivated gelatin sample showed a marked increase of approximately 5 – 6-fold in the total
CE
work of adhesion and maximum detachment force when compared to the thiolated gelatin
AC
sample. This was observed at both pH 5 and pH 6.8, again showing the importance of the concept of polymer preactivation. The rotating cylinder method, also on porcine intestinal tissue, reiterated the tensile strength tests, and the adhesion times of the preactivated gelatin sample were far superior to that of the thiolated gelatin and unmodified gelatin samples. The rotating cylinder method was again tested at both pH 5 and pH 6.8 and interestingly, there was a greater improvement in mucoadhesion observed in the pH 5 buffer, with adhesion of up to 5 days measured, approximately 5-fold higher than adhesion in pH 6.8 buffer. The improved mucoadhesion at pH 5 was also observed in the thiolated gelatin sample too. It was 41
ACCEPTED MANUSCRIPT thought that this was due to the isoelectric point of both the preactivated gelatin and thiolated gelatin samples, measuring 5.7 and 4.7 respectively, both of which were closer to pH 5 than to pH 6.8. Thus, less erosion of the polymer was occurring which in turn increased the time of adhesion of the samples.
3.7 Permeation enhancing effects of thiolated polymers
PT
Apart from forming strong bonds between the polymer and the mucosal surface and
RI
prolonging residence times at the target site, it has also been observed that thiolation of a
SC
polymer has other benefits, including a permeation enhancing effect. Previously, to enhance the uptake of a drug in the intestine, a permeation enhancer agent would also be administered;
NU
however, these agents were generally of low MW and were often absorbed more quickly that the target drug itself, resulting in possible toxic side effects (A. Bernkop-Schnürch et al.,
MA
2003). It was found that polymeric compounds, such as chitosan and PAA displayed permeation enhancing effects and had the benefits of not being absorbed by the intestine due
D
to their higher MW. As both chitosan and PAA have inherent mucoadhesive properties, it
PT E
was postulated that thiolated versions of the polymers would display further enhancement. Clausen and Bernkop-Schnurch (2000) demonstrated the permeation enhancing effects of
CE
thiolated polymers on tight junctions in tissues. In this study, the permeation of thiolated polycarbophil with peptide based drugs on mucosal tissues was investigated and compared to
AC
that of unmodified polycarbophil. Initial studies using sodium fluorescein as the model drug, showed a higher level of transport using the thiolated polycarbophil in comparison to its unmodified counterpart, and it was also showed that the degree of thiolation too had an effect on transport, with higher levels of thiolation resulting in more transport of sodium fluorescein over a three hour period. The study also investigated the transport of both insulin and of bacitracin, which had both been labelled with fluorescein isothiocyanate, using unmodified and thiolated polycarbophil. Here, there was no significant improvement of transport of the 42
ACCEPTED MANUSCRIPT peptide drugs due to the thiolated polycarbophil. It was thought that the observed permeation enhancing effects were due to the bonding of the thiolated polycarbophil with extracellular Ca2+, therefore allowing the tight junctions to open. They also observed a decrease in the transepithelial electrical resistance (TEER), which is the measurement of tightness between mature epithelial cell junctions (Sigurdsson et al., 2013). They concluded this decrease in
PT
TEER as the loosening of the membrane’s tight junctions, which contributed to the passive
RI
diffusion of the peptide drugs across the membrane.
SC
Vigl et al. (2009) investigated the effect thiolated PAH had on efflux pump inhibitory properties. Initially, PAH was crosslinked by reacting the polymer with triethylamine
NU
followed by the addition of dimethylsuccinate. This crosslinked polymer of 70 kDa MW,
MA
along with two uncrosslinked samples of 15 kDa and 70 kDa MW, were thiolated using Nacetlycysteine in the presence of the carbodiimide coupling reagent, EDC. Thiol content was measured as 77.6 µmol/g in the 15 kDa sample, 83.1 µmol/g in the 70 kDa uncrosslinked
PT E
D
sample and 162.5 µmol/g in the crosslinked 70 kDa sample. Inhibition of P-glycoprotein, an efflux transport molecule found within the intestinal epithelium, was tested using thiolated PAH and unmodified PAH. Rhodamine-123 was used as a model substrate for P-
CE
glycoprotein. A portion of rat ileum tissue was mounted into an Ussing type diffusion
AC
chamber and was incubated at 37 °C in artificial intestinal media. This media was removed and was replaced with a solution of rhodamine – 123 with or without the addition of thiolated or unmodified PAH. Samples were then taken either side of the intestinal tissue to investigate the permeation effect of the polymers. Additionally, the paracellular mechanism was also examined using FD-4, which is only transported through a paracellular route. Rhodamine transport was positively increased by the thiolated polymers, to a greater extent than the unmodified sample and the transport of FD-4 was not influenced by either the unmodified or thiolated PAH, therefore, it was surmised that the increase in rhodamine permeation was due 43
ACCEPTED MANUSCRIPT to the inhibition of P-glycoprotein. Cell cytotoxicity was also conducted using Caco-2 cells on unmodified and thiolated PAH samples. Although cell cytotoxicity was low and both cell permeation and efflux pump inhibition were positively affected by the thiolated PAH polymers, it was acknowledged that a higher thiol content would have vastly improved the results, as seen by Clausen and Bernkop-Schnürch (2000).
PT
The permeation enhancing effects were also examined in preactivated PAA samples (Wang
RI
et al., 2012). PAA was preactivated with 2-mercaptonicotinic acid and the influence that both
SC
MW and preactivation levels had on enhancing permeation were investigated. PAA of MW 100 kDa, 250 kDa or 450 kDa were used which all had similar preactivation levels of over
NU
700 µmol/g (high levels), investigating MW influences. PAA 450 kDa was then also preactivated to 300 - 400 µmol/g (medium levels) and also to less than 200 µmol/g (low
MA
levels) to examine the effects of the degree of preactivation. Sodium fluorescein was used as the model drug for the analysis of permeation across rat intestinal tissue. When investigating
D
the influence of MW, it was shown that higher MW samples, therefore PAA 450 kDa,
PT E
displayed greater permeation effects in comparison to both the 100 kDa and 250 kDa samples. Reduced glutathione was added as a permeation mediator and again PAA 450 kDa
CE
displayed higher levels of permeation in comparison to both other samples, although
AC
permeation using PAA 250 kDa was marginally improved with the addition of reduced glutathione. When comparing the preactivated PAA 450 kDa sample to an unmodified PAA 450 kDa and a thiolated PAA 450 kDa, the preactivated sample displayed greater degrees of permeation. It was surmised that this was due to the increased mucoadhesive nature of the preactivated samples, as well as the potential interaction with tight junction proteins which could lead to conformation change in the protein, enhancing drug uptake. Interestingly, when comparing the results of the three preactivated 450 kDa PAA samples of varying degrees of preactivation, low, medium and high, the medium sample displayed the highest levels of 44
ACCEPTED MANUSCRIPT permeation. This was in contrast to previous studies investigating thiolated PAA samples, in which higher levels of thiolation improved the permeation effects (Kast and BernkopSchnurch, 2002). It was thought that perhaps this was due to the decrease in the hydrophilic nature of the polymer upon the addition of the 2-mercaptonicotinic acid, which decreased further with the highest level of preactivation. Having said that, the medium preactivated
PT
sample displayed the highest permeation levels and it was thought this was due to the
RI
appropriate levels of thiol groups within the 300 - 400 µmol/g preactivated sample.
SC
As discussed above, κ and ι carrageenans were thiolated with thiourea and the efflux pump inhibitory actions of both thiolated carrageenan were also investigated using Caco-2 cells,
NU
focusing on the multidrug resistance protein (MRP-2) receptor (Suchaoin et al., 2015). MRP2 can be found on the surface of cancerous cells and may have a negative influence on the
MA
uptake of anticancer drugs. Using the fluorescent marker of sulforhodamine 101, it was shown that thiolated carrageenan increased the amount of intracellular sulforhodamine 101
D
once the cells were pretreated with the thiolated samples when compared to their unmodified
PT E
counterparts. Additionally to this, the thiolated carrageenans were also shown to improve the uptake of sulforhodamine 101 across a Caco-2 monolayer, thus indicating the blocking effect
CE
of MRP-2. It was thought this inhibitory effect was due to the binding of the thiolated
AC
carrageenan to the thiol moieties found on MRP-2.
4 Conclusion
The thiolation of a number of different natural and synthetic polymers has led to increased mucoadhesive properties, slower and more controlled release of drug, and enhanced permeation effects. Further developments in the thiolation of polymers, including Sprotecting and preactivation of the thiol groups, have opened up a new generation of thiolated polymers which allows for mucoadhesion to occur over a wider range of pH levels. This
45
ACCEPTED MANUSCRIPT could equate to a mucoadhesive delivery device which has more scope for the administration, improved adhesion and subsequent controlled drug delivery throughout a larger area of the body, including the full length of the gastrointestinal tract and vagina; further work may be required in order to create and develop a commercially viable drug delivery device for use in these less accessible areas, which resist becoming over hydrated, thus losing its adhesive
PT
properties while retaining its controlled drug release profile. The further gelling properties of
RI
thiolated polymers could prolong the residence time and inhibit the premature removal of
SC
dosage forms, semi-solid for examples, in highly turbulent areas, such as the vagina or the eye. Targeting the eye, for example to treat posterior segment diseases, using a mucoadhesive
NU
dosage form could have many benefits, including fewer systemic side effects, however, poor patient compliance due to the insert of a drug releasing tablet for prolonged periods into the
MA
eye could affect its popularity. Having said that, the release of gel-like dosage forms, such as Lacrimera ®, to the market for treatment to the dry eye syndrome and the founding of
D
Thiomatrix®, a drug delivery company specialising in thiomer technology, is testament to the
AC
CE
PT E
future of mucoadhesive polymeric devices.
46
ACCEPTED MANUSCRIPT 5 References Anderson, O. A., Bainbridge, J. W. B. and Shima, D. T. (2010) 'Delivery of anti-angiogenic molecular therapies for retinal disease', Drug Discovery Today, 15(7-8), pp. 272-282. Andrews, G. P., Laverty, T. P. and Jones, D. S. (2009) 'Mucoadhesive polymeric platforms for controlled drug delivery', European Journal of Pharmaceutics and Biopharmaceutics, 71(3), pp. 505-518.
RI
PT
Asane, G. S., Nirmal, S. A., Rasal, K. B., Naik, A. A. and Mahadik, M. S. (2008) 'Polymers for Mucoadhesive Drug Delivery Systems: A Current Status', Drug Development and Industrial Pharmacy, 34, pp. 1246-1266.
NU
SC
Bacalocostantis, I., Mane, V. P., Kang, M. S., Goodley, A. S., Muro, S. and Kofinas, P. (2012) 'Effect of Thiol Pendant Conjugates on Plasmid DNA Binding, Release, and Stability of Polymeric Delivery Vectors', Biomacromolecules, 13(5), pp. 1331-1339.
MA
Bansil, R. and Turner, B. S. (2006) 'Mucin structure, aggregation, physiological functions and biomedical applications', Current Opinion in Colloid & Interface Science, 11(2-3), pp. 164-170.
D
Bernkop-Schnürch, A. (2005) 'Thiomers: A new generation of mucoadhesive polymers', Advanced Drug Delivery Reviews, 57(11), pp. 1569-1582.
PT E
Bernkop-Schnürch, A., Guggi, D. and Pinter, Y. (2004a) 'Thiolated chitosans: development and in vitro evaluation of a mucoadhesive, permeation enhancing oral drug delivery system', Journal of Controlled Release, 94(1), pp. 177-186.
AC
CE
Bernkop-Schnürch, A., Hornof, M. and Guggi, D. (2004b) 'Thiolated chitosans', European Journal of Pharmaceutics and Biopharmaceutics, 57(1), pp. 9-17. Bernkop-Schnürch, A., Hornof, M. and Zoidl, T. (2003) 'Thiolated polymers—thiomers: synthesis and in vitro evaluation of chitosan–2-iminothiolane conjugates', International Journal of Pharmaceutics, 260(2), pp. 229-237. Bernkop-Schnürch, A., Kast, C. E. and Guggi, D. (2003) 'Permeation enhancing polymers in oral delivery of hydrophilic macromolecules: thiomer/GSH systems', Journal of Controlled Release, 93(2), pp. 95-103. Bernkop-Schnürch, A., Schwarz, V. and Steininger, S. (1999) 'Polymers with thiol groups: a new generation of mucoadhesive polymers?', Pharmaceutical Research, 16(6), pp. 876-881.
47
ACCEPTED MANUSCRIPT Bernkop-Schnürch, A. and Steininger, S. (2000) 'Synthesis and characterisation of mucoadhesive thiolated polymers', International Journal of Pharmaceutics, 194(2), pp. 239247. Bonengel, S., Haupstein, S., Perera, G. and Bernkop-Schnürch, A. (2014) 'Thiolated and Sprotected hydrophobically modified cross-linked poly(acrylic acid) – A new generation of multifunctional polymers', European Journal of Pharmaceutics and Biopharmaceutics, 88(2), pp. 390-396.
RI
PT
Boussif, O., Delair, T., Brua, C., Veron, L., Pavirani, A. and Kolbe, H. V. J. (1999) 'Synthesis of Polyallylamine Derivatives and Their Use as Gene Transfer Vectors in Vitro', Bioconjugate Chemistry, 10(5), pp. 877-883.
NU
SC
Brandt, J., Svenson, A., Carlsson, J. and Drevin, H. (1977) 'Covalent coupling of unsaturated compounds to thiol agrarose using γ-Radiation', Journal of Solid-Phase Biochemistry, 2(2), pp. 105-109.
MA
Clausen, A. E. and Bernkop-Schnürch, A. (2000) 'In vitro evaluation of the permeationenhancing effect of thiolated polycarbophil', Journal of Pharmaceutical Sciences, 89(10), pp. 1253-1261.
D
Cone, R. A. (2009) 'Barrier properties of mucus', Advanced Drug Delivery Reviews, 61(2), pp. 75-85.
PT E
Dinarvand, R., Mahmoodi, S., Farboud, E., Salehi, M. and Atyabi, F. (2005) 'Preparation of gelatin microspheres containing lactic acid – Effect of cross-linking on drug release', Acta Pharmaceutica, 55, pp. 57-67.
AC
CE
Dodane, V., Amin Khan, M. and Merwin, J. R. (1999) 'Effect of chitosan on epithelial permeability and structure', International Journal of Pharmaceutics, 182(1), pp. 21-32. Duggan, S., Hughes, H., Owens, E., Duggan, E., Cummins, W. and O’ Donovan, O. (2016a) 'Synthesis and characterisation of mucoadhesive thiolated polyallylamine', International Journal of Pharmaceutics, 499(1–2), pp. 368-375. Duggan, S., O’Donovan, O., Owens, E., Cummins, W. and Hughes, H. (2015) 'Synthesis of mucoadhesive thiolated gelatin using a two-step reaction process', European Journal of Pharmaceutics and Biopharmaceutics, 91, pp. 75-81. Duggan, S., O’Donovan, O., Owens, E., Duggan, E., Hughes, H. and Cummins, W. (2016b) 'Comparison of the mucoadhesive properties of thiolated polyacrylic acid to thiolated polyallylamine', International Journal of Pharmaceutics, 498(1–2), pp. 245-253.
48
ACCEPTED MANUSCRIPT
Dünnhaupt, S., Barthelmes, J., Thurner, C. C., Waldner, C., Sakloetsakun, D. and BernkopSchnürch, A. (2012) 'S-protected thiolated chitosan: Synthesis and in vitro characterization', Carbohydrate Polymers, 90(2), pp. 765-772. Ensign, L. M., Cone, R. and Hanes, J. (2012) 'Oral drug delivery with polymeric nanoparticles: The gastrointestinal mucus barriers', Advanced Drug Delivery Reviews, 64(6), pp. 557-570.
RI
PT
Friedl, H. E., Dünnhaupt, S., Waldner, C. and Bernkop-Schnürch, A. (2013) 'Preactivated thiomers for vaginal drug delivery vehicles', Biomaterials, 34(32), pp. 7811-7818.
SC
Grabovac, V. and Bernkop-Schnürch, A. (2007) 'Development and In Vitro Evaluation of Surface Modified Poly(lactide-co-glycolide) Nanoparticles with Chitosan-4Thiobutylamidine', Drug Development and Industrial Pharmacy, 33(7), pp. 767-774.
MA
NU
Grabovac, V., Föger, F. and Bernkop-Schnürch, A. (2008) 'Design and in vivo evaluation of a patch delivery system for insulin based on thiolated polymers', International Journal of Pharmaceutics, 348(1-2), pp. 169-174.
D
Grabovac, V., Guggi, D. and Bernkop-Schnürch, A. (2005) 'Comparison of the mucoadhesive properties of various polymers', Advanced Drug Delivery Reviews, 57(11), pp. 1713-1723.
PT E
Guggi, D., Marschütz, M. K. and Bernkop-Schnürch, A. (2004) 'Matrix tablets based on thiolated poly(acrylic acid): pH-dependent variation in disintegration and mucoadhesion', International Journal of Pharmaceutics, 274(1-2), pp. 97-105.
AC
CE
Guo, J., O’Mahony, A. M., Cheng, W. P. and O’Driscoll, C. M. (2013) 'Amphiphilic polyallylamine based polymeric micelles for siRNA delivery to the gastrointestinal tract: In vitro investigations', International Journal of Pharmaceutics, 447(1–2), pp. 150-157. H.P. Rang, M.M. Dale, J.M. Ritter and Moore, P. K. (2003) Pharmacology. 5th ed., Elsevier. Hauptstein, S., Bonengel, S., Rohrer, J. and Bernkop-Schnürch, A. (2014) 'Preactivated thiolated poly(methacrylic acid-co-ethyl acrylate): Synthesis and evaluation of mucoadhesive potential', European Journal of Pharmaceutical Sciences, 63(0), pp. 132-139. Hauptstein, S., Dezorzi, S., Prüfert, F., Matuszczak, B. and Bernkop-Schnürch, A. (2015) 'Synthesis and in vitro characterization of a novel S-protected thiolated alginate', Carbohydrate Polymers, 124(0), pp. 1-7.
49
ACCEPTED MANUSCRIPT Hauptstein, S., Muller, C., Dunnhaupt, S., Laffleur, F. and Bernkop-Schnurch, A. (2013) 'Preactivated thiomers: evaluation of gastroretentive minitablets', Int J Pharm, 456(2), pp. 473-479. Hintzen, F., Hauptstein, S., Perera, G. and Bernkop-Schnürch, A. (2013) 'Synthesis and in vitro characterization of entirely S-protected thiolated pectin for drug delivery', European Journal of Pharmaceutics and Biopharmaceutics, 85(3, Part B), pp. 1266-1273.
RI
PT
Hornof, M., Weyenberg, W., Ludwig, A. and Bernkop-Schnürch, A. (2003) 'Mucoadhesive ocular insert based on thiolated poly(acrylic acid): development and in vivo evaluation in humans', Journal of Controlled Release, 89(3), pp. 419-428.
NU
SC
Horvát, G., Gyarmati, B., Berkó, S., Szabó-Révész, P., Szilágyi, B. Á., Szilágyi, A., Soós, J., Sandri, G., Bonferoni, M. C., Rossi, S., Ferrari, F., Caramella, C., Csányi, E. and BudaiSzűcs, M. (2015) 'Thiolated poly(aspartic acid) as potential in situ gelling, ocular mucoadhesive drug delivery system', European Journal of Pharmaceutical Sciences, 67, pp. 1-11.
MA
Huang, Y., Leobandung, W., Foss, A. and Peppas, N. A. (2000) 'Molecular aspects of mucoand bioadhesion:: Tethered structures and site-specific surfaces', Journal of Controlled Release, 65(1-2), pp. 63-71.
PT E
D
Hudson, S. P., Owens, E., Hughes, H. and McLoughlin, P. (2012) 'Enhancement and restriction of chain motion in polymer networks', International Journal of Pharmaceutics, 430(1–2), pp. 34-41.
CE
Ibie, C. O., Thompson, C. J. and Knott, R. (2015) 'Synthesis, characterisation and in vitro evaluation of novel thiolated derivatives of polyallylamine and quaternised polyallylamine', Colloid and Polymer Science, pp. 1-12.
AC
Iqbal, J., Sakloetsakun, D. and Bernkop-Schnürch, A. (2011) 'Thiomers: Inhibition of cytochrome P450 activity', European Journal of Pharmaceutics and Biopharmaceutics, 78(3), pp. 361-365. Iqbal, J., Shahnaz, G., Dünnhaupt, S., Müller, C., Hintzen, F. and Bernkop-Schnürch, A. (2012a) 'Preactivated thiomers as mucoadhesive polymers for drug delivery', Biomaterials, 33(5), pp. 1528-1535. Iqbal, J., Shahnaz, G., Perera, G., Hintzen, F., Sarti, F. and Bernkop-Schnürch, A. (2012b) 'Thiolated chitosan: Development and in vivo evaluation of an oral delivery system for leuprolide', European Journal of Pharmaceutics and Biopharmaceutics, 80(1), pp. 95-102.
50
ACCEPTED MANUSCRIPT Ivarsson, D. and Wahlgren, M. (2012) 'Comparison of in vitro methods of measuring mucoadhesion: Ellipsometry, tensile strength and rheological measurements', Colloids and Surfaces B: Biointerfaces, 92(0), pp. 353-359. Joergensen, L., Klösgen, B., Simonsen, A. C., Borch, J. and Hagesaether, E. (2011) 'New insights into the mucoadhesion of pectins by AFM roughness parameters in combination with SPR', International Journal of Pharmaceutics, 411(1–2), pp. 162-168.
RI
PT
Kafedjiiski, K., Jetti, R. K. R., Föger, F., Hoyer, H., Werle, M., Hoffer, M. and BernkopSchnürch, A. (2007) 'Synthesis and in vitro evaluation of thiolated hyaluronic acid for mucoadhesive drug delivery', International Journal of Pharmaceutics, 343(1-2), pp. 48-58.
SC
Kast, C. and Bernkop-Schnurch, A. (2002) 'Influence of the molecular mass on the permeation enhancing effect of different poly(acrylates)', Techniques Pharmaceutiques, 12(6), pp. 351-356.
MA
NU
Kast, C. E. and Bernkop-Schnürch, A. (2001) 'Thiolated polymers — thiomers: development and in vitro evaluation of chitosan–thioglycolic acid conjugates', Biomaterials, 22(17), pp. 2345-2352.
D
Kawashima, Y., Yamamoto, H., Takeuchi, H. and Kuno, Y. (2000) 'Mucoadhesive DLLactide/Glycolide Copolymer Nanospheres Coated with Chitosan to Improve Oral Delivery of Elcatonin', Pharmaceutical Development and Technology, 5(1), pp. 77-85.
PT E
Khanenko, E. A., Larionova, N. I. and Demina, N. B. (2009) 'Mucoadhesive Drug Delivery Systems (Review)', Pharmaceutical Chemistry Journal, 43(4), pp. 21 - 29.
CE
Khutoryanskiy, V. V. (2011) 'Advances in mucoadhesion and mucoadhesive polymers', Macromolecular Bioscience, 11, pp. 748-764.
AC
Kommareddy, S. and Amiji, M. (2005) 'Preparation and evaluation of thiol-modified gelatin nanoparticles for intracellular DNA delivery in response to glutathione', Bioconjugate Chemistry, 16, pp. 1423-1432. Krauland, A. H. and Bernkop-Schnürch, A. (2004) 'Thiomers: development and in vitro evaluation of a peroral microparticulate peptide delivery system', European Journal of Pharmaceutics and Biopharmaceutics, 57(2), pp. 181-187. Kumari, A., Yadav, S. K. and Yadav, S. C. (2010) 'Biodegradable polymeric nanoparticles based drug delivery systems', Colloids and Surfaces B: Biointerfaces, 75(1), pp. 1-18.
51
ACCEPTED MANUSCRIPT Leitner, V. M., Marschütz, M. K. and Bernkop-Schnürch, A. (2003) 'Mucoadhesive and cohesive properties of poly(acrylic acid)-cysteine conjugates with regard to their molecular mass', European Journal of Pharmaceutical Sciences, 18(1), pp. 89-96. Leitner, V. M., Walker, G. F. and Bernkop-Schnürch, A. (2003) 'Thiolated polymers: evidence for the formation of disulphide bonds with mucus glycoproteins', European Journal of Pharmaceutics and Biopharmaceutics, 56(2), pp. 207-214.
PT
Liu, L., Fishman, M. L., Hicks, K. B. and Kende, M. (2005) 'Interaction of various pectin formulations with porcine colonic tissues', Biomaterials, 26(29), pp. 5907-5916.
SC
RI
Ludwig, A. (2005) 'The use of mucoadhesive polymers in ocular drug delivery', Advanced Drug Delivery Reviews, 57(11), pp. 1595-1639.
MA
NU
Luessen, H. L., Verhoef, J. C., Borchard, G., Lehr, C.-M., Boer, A. G. d. and Junginger, H. E. (1995) 'Mucoadhesive polymers in peroral peptide drug delivery. II. Carbomer and polycarbophil are potent inhitiors of the intestinal proteolytic enzyme trypsin', Pharmaceutical Research, 12, pp. 1293-1298. Madhav, N. V. S., Shakya, A. K., Shakya, P. and Singh, K. (2009) 'Orotransmucosal drug delivery systems: A review', Journal of Controlled Release, 140(1), pp. 2-11.
PT E
D
Makhlof, A., Werle, H. and Takeuchi, H. (2008) 'Mucoadhesive drug carriers and polymers for effective drug delivery', Journal of Drug Delivery Science and Technology, 18, pp. 375386.
CE
Mortazavi, S. A. (1995) 'An in vitro assessment of mucus/mucoadhesive interactions', International Journal of Pharmaceutics, 124(2), pp. 173-182.
AC
Nagarwal, R. C., Kant, S., Singh, P. N., Maiti, P. and Pandit, J. K. (2009) 'Polymeric nanoparticulate system: A potential approach for ocular drug delivery', Journal of Controlled Release, 136(1), pp. 2-13. Oskuee, R. K., Dosti, F., Gholami, L. and Malaekeh-Nikouei, B. (2015) 'A simple approach for producing highly efficient DNA carriers with reduced toxicity based on modified polyallylamine', Materials Science and Engineering: C, 49(0), pp. 290-296. Palmberger, T. F., Albrecht, K., Loretz, B. and Bernkop-Schnürch, A. (2007) 'Thiolated polymers: Evaluation of the influence of the amount of covalently attached l-cysteine to poly(acrylic acid)', European Journal of Pharmaceutics and Biopharmaceutics, 66(3), pp. 405-412.
52
ACCEPTED MANUSCRIPT Panyam, J. and Labhasetwar, V. (2003) 'Biodegradable nanoparticles for drug and gene delivery to cells and tissue', Advanced Drug Delivery Reviews, 55(3), pp. 329-347. Patel, V. F., Liu, F. and Brown, M. B. (2011) 'Advances in oral transmucosal drug delivery', Journal of Controlled Release, 153(2), pp. 106-116. Pratt, C. W. and Cornely, K. (2011) Essential Biochemistry. Second ed., Wiley.
RI
PT
Rohrer, J., Zupančič, O., Suchaoin, W., Netsomboon, K., Laffleur, F., Oh, S. and BernkopSchnürch, A. (2015) 'Synthesis and in vitro characterisation of preactivated thiolated gelatin', European Polymer Journal, 73, pp. 268-277.
SC
Roy, S., Pal, K., Anis, A., Pramanik, K. and B.Prabhakar (2009) 'Polymers in Mucoadhesive Drug Delivery System: A Brief Note', Designed Monomers and Polymers, 12, pp. 483-495.
MA
NU
Salamanca, A. E. d., Diebold, Y., Calonge, M., Garcıa-Vazquez, C., Callejo, S., Vila, A. and Alonso, M. J. (2006) 'Chitosan Nanoparticles as a Potential Drug Delivery System for the Ocular Surface: Toxicity, Uptake Mechanism and In Vivo Tolerance', Investigative Ophthalmology and Visual Science, 47(4), pp. 1416-1425.
PT E
D
Schmitz, T., Grabovac, V., Palmberger, T. F., Hoffer, M. H. and Bernkop-Schnürch, A. (2008) 'Synthesis and characterization of a chitosan-N-acetyl cysteine conjugate', International Journal of Pharmaceutics, 347(1–2), pp. 79-85. Sharma, R. and Ahuja, M. (2011) 'Thiolated pectin: Synthesis, characterization and evaluation as a mucoadhesive polymer', Carbohydrate Polymers, 85(3), pp. 658-663.
AC
CE
Sigurdsson, H. H., Kirch, J. and Lehr, C.-M. (2013) 'Mucus as a barrier to lipophilic drugs', International Journal of Pharmaceutics, 453(1), pp. 56-64. Smart, J. D. (2005) 'The basics and underlying mechanisms of mucoadhesion', Advanced Drug Delivery Reviews, 57(11), pp. 1556-1568. Srividya, B., Cardoza, R. M. and Amin, P. D. (2001) 'Sustained ophthalmic delivery of ofloxacin from a pH triggered in situ gelling system', Journal of Controlled Release, 73(2-3), pp. 205-211. Suchaoin, W., Bonengel, S., Grießinger, J. A., Pereira de Sousa, I., Hussain, S., Huck, C. W. and Bernkop-Schnürch, A. (2016a) 'Novel bioadhesive polymers as intra-articular agents: Chondroitin sulfate-cysteine conjugates', European Journal of Pharmaceutics and Biopharmaceutics, 101, pp. 25-32.
53
ACCEPTED MANUSCRIPT
Suchaoin, W., Bonengel, S., Hussain, S., Huck, C. W., Ma, B. N. and Bernkop-Schnürch, A. (2015) 'Synthesis and In Vitro Evaluation of Thiolated Carrageenan', Journal of Pharmaceutical Sciences, 104(8), pp. 2523-2530.
PT
Suchaoin, W., Pereira de Sousa, I., Netsomboon, K., Rohrer, J., Hoffmann Abad, P., Laffleur, F., Matuszczak, B. and Bernkop-Schnürch, A. (2016b) 'Mucoadhesive polymers: Synthesis and in vitro characterization of thiolated poly(vinyl alcohol)', International Journal of Pharmaceutics, 503(1–2), pp. 141-149.
RI
Tabata, Y. and Ikada, Y. (1998) 'Protein release from gelatin matrices', Advanced Drug Delivery Reviews, 31(3), pp. 287-301.
NU
SC
Thirawong, N., Nunthanid, J., Puttipipatkhachorn, S. and Sriamornsak, P. (2007) 'Mucoadhesive properties of various pectins on gastrointestinal mucosa: An in vitro evaluation using texture analyzer', European Journal of Pharmaceutics and Biopharmaceutics, 67(1), pp. 132-140.
MA
Ugwoke, M. I., Agu, R. U., Verbeke, N. and Kinget, R. (2005) 'Nasal mucoadhesive drug delivery: Background, applications, trends and future perspectives', Advanced Drug Delivery Reviews, 57(11), pp. 1640-1665.
PT E
D
Vigl, C., Leithner, K., Albrecht, K. and Bernkop-Schnurch, A. (2009) 'The efflux pump inhibitory properties of (thiolated) polyallylamines', Journal of Drug Delivery Science and Technology, 19(6), pp. 405-411.
CE
Vlierberghe, S. V., Schacht, E. and Dubruel, P. (2011) 'Reversible gelatin-based hydrogels: Finetuning of material properties', European Polymer Journal, 47(5), pp. 1039-1047.
AC
Wang, J., Tauchi, Y., Deguchi, Y., Morimoto, K., Tabata, Y. and Ikada, Y. (2000) 'Positively Charged Gelatin Microspheres as Gastric Mucoadhesive Drug Delivery System for Eradication of H. pylori', Drug Delivery, (7), pp. 237-243. Wang, X., Iqbal, J., Rahmat, D. and Bernkop-Schnürch, A. (2012) 'Preactivated thiomers: Permeation enhancing properties', International Journal of Pharmaceutics, 438(1–2), pp. 217-224. Yasukawa, T., Ogura, Y., Tabata, Y., Kimura, H., Wiedemann, P. and Honda, Y. (2004) 'Drug delivery systems for vitreoretinal diseases', Progress in Retinal and Eye Research, 23(3), pp. 253-281.
54
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
PT
Young, S., Wong, M., Tabata, Y. and Mikos, A. G. (2005) 'Gelatin as a delivery vehicle for the controlled release of bioactive molecules', Journal of Controlled Release, 109(1-3), pp. 256-274.
55
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
Figure 1 Structures of mucoadhesive polymers (A) chitosan, (B) polyacrylic acid (PAA), (C) polylactic-coglycolide (PLGA), (D) pectin, (E) gelatin (Elzoghby, 2013) and (F) polyallylamine in its (i) free base and (ii) salt forms (Vigl et al., 2009).
56
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
Figure 2 Mucoadhesive testing on thiolated PAA (black) and control PAA (white) samples, with and without the addition of 1% (m/m) cysteine (Verena M. Leitner et al., 2003)
57
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
Figure 3 (A) Structure of L-cysteine and (B) the proposed structure of PAA-cysteine conjugate (BernkopSchnürch and Steininger, 2000)
58
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
Figure 4 Structure of (A) Traut’s reagent (2-iminothiolane) and (B) the proposed structure of PAAm thiolated with Traut’s reagent
59
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
Figure 5 Mucoadhesive testing of a thiolated gelatin control (a native gelatin sample which had been directly thiolated with Traut’s reagent) and a thiolated gelatin sample following the two-step reaction (Duggan et al., 2015)
60
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
Figure 6 Structure of preactivated PAA (Iqbal et al., 2012a)
61
AC
CE
Graphical abstract
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
62
Sarah Duggan, Wayne Cummins, Orla O' Donovan, Helen Hughes, Eleanor Owens PII: DOI: Reference:
S0928-0987(17)30020-9 doi: 10.1016/j.ejps.2017.01.008 PHASCI 3868
To appear in:
European Journal of Pharmaceutical Sciences
Received date: Revised date: Accepted date:
28 November 2016 4 January 2017 9 January 2017
Please cite this article as: Sarah Duggan, Wayne Cummins, Orla O' Donovan, Helen Hughes, Eleanor Owens , Thiolated polymers as mucoadhesive drug delivery systems. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Phasci(2017), doi: 10.1016/j.ejps.2017.01.008
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Thiolated polymers as mucoadhesive drug delivery systems
Sarah Duggan *a, Wayne Cummins a, Orla O’ Donovan a, Helen Hughes a, Eleanor Owens a Pharmaceutical and Molecular Biotechnology Research Centre, Department of Chemical
PT
a
and Life Sciences, Waterford Institute of Technology, Waterford, Ireland.
SC
AC
CE
PT E
D
MA
NU
Email address: [email protected] (Sarah Duggan)
RI
*Corresponding author. Tel: +353 51 834126
1
ACCEPTED MANUSCRIPT Abstract Mucoadhesion is the process of binding a material to the mucosal layer of the body. Utilising both natural and synthetic polymers, mucoadhesive drug delivery is a method of controlled drug release which allows for intimate contact between the polymer and a target tissue. It has the potential to increase bioavailability, decrease potential side effects and offer protection to
PT
more sensitive drugs such as proteins and peptide based drugs. The thiolation of polymers
RI
has, in the last number of years, come to the fore of mucoadhesive drug delivery, markedly
SC
improving mucoadhesion due to the introduction of free thiol groups onto the polymer backbone while also offering a more cohesive polymeric matrix for the slower and more
NU
controlled release of drug. This review explores the concept of mucoadhesion and the recent
PT E
D
mucoadhesive drug delivery devices.
MA
advances in both the polymers and the methods of thiolation used in the synthesis of
AC
polymers
CE
Keywords: mucoadhesion, controlled drug delivery, thiolation, natural and synthetic
2
ACCEPTED MANUSCRIPT 1 Introduction The area of mucosal surfaces of the body far exceeds that of the area of the skin, with the area of the small intestine alone being 100 x greater than the surface of the skin (Khanenko et al., 2009). With this in mind and together with the need for a protective delivery system for drugs such as peptides and proteins (Khutoryanskiy, 2011), the development of mucoadhesive drug
PT
delivery systems has been of great interest over the past number of years (Bernkop-Schnürch
RI
et al., 1999; Bernkop-Schnürch et al., 2004a; Hornof et al., 2003; Iqbal et al., 2012b).
SC
The transmucosal route of drug administration can have up to 4 x the rate of absorption than skin, depending on the site of administration, potentially making it a more resourceful route
NU
for drug delivery (Madhav et al., 2009). There are many advantages to the development of bio- and mucoadhesive drug delivery systems, including: Localised/site specific drug delivery
Reduced likelihood of systemic side effects
Better residence time/more intimate contact to the target surface
Avoidance of peak concentrations due to controlled/slow release of drug (zero order
PT E
D
MA
kinetics)
CE
Improved bioavailability of the drug may be observed due to the intimate and increased
AC
contact time between the drug-loaded polymer and the target area, thus reducing the time of action of the drug at the site. A lower concentration of drug can also be used as the drug could potentially be absorbed directly onto the intended tissue, not systemically, which reduces the potential for toxic side effects. Both synthetic and naturally derived polymers have demonstrated potential as mucoadhesives. Many synthetic polymers are hydrophobic and most natural polymers are hydrophilic (Yasukawa et al., 2004). Synthetic polymers are known to exhibit a longer duration in the body and, therefore, sustain the release of the drug over a prolonged time 3
ACCEPTED MANUSCRIPT period in comparison to the short duration of natural polymers. However, the conditions in which synthetic polymers are formulated, particularly in the formulation of nanoparticulate systems, can be harsher in comparison to natural polymers, using organic solvents in the process (Panyam and Labhasetwar, 2003). Synthetic polymers are also less likely to be biodegradable and may cause an immune response. The majority of polymers used in
PT
mucoadhesion are hydrophilic. They are soluble in water and have a high density of
RI
functional groups which are able to interact through hydrogen bonding with mucins, leading
SC
to stronger adhesion (Andrews et al., 2009).
Andreas Bernkop-Schürch pioneered research in the area of thiolated polymers using a range
NU
of different approaches (Bernkop-Schnürch and Steininger, 2000; Hornof et al., 2003; Andreas Bernkop-Schnürch et al., 2003; Bernkop-Schnürch et al., 2004a; Bernkop-Schnürch
MA
et al., 2004b; Grabovac et al., 2005; Palmberger et al., 2007; Schmitz et al., 2008; Vigl et al., 2009; Iqbal et al., 2011; Wang et al., 2012; Hauptstein et al., 2013; Hauptstein et al., 2014;
D
Hauptstein et al., 2015). By thiolating well established mucoadhesive polymers,
PT E
mucoadhesive properties were greatly improved, with up to a 140-fold increase in adhesive properties (Bernkop-Schnürch, 2005). This is due to the formation of inter- and intra-
CE
disulphide bridges with the cysteine rich areas of the mucosal layer. These covalent bonds
AC
formed with the mucosal layer are much stronger than the non-covalent bonds formed by non-thiolated polymers, and mirror the natural bonds formed within the mucosal layer itself.
2 Mucoadhesion 2.1 Mucosal layer The mucosal layer acts as a lubricating and protective barrier and lines the gastrointestinal (GI) tract, eye, mouth, nasal cavity, reproductive tract and the respiratory tract. It consists of approximately 95% water, with cholesterol, lipids, proteins and other elements making up the 4
ACCEPTED MANUSCRIPT remainder (Bansil and Turner, 2006). The viscoelastic properties of the mucosal layer are controlled generally by the glycoproteins, mucins, but water and ion content also contribute (Ensign et al., 2012). The surface will allow small molecules and viruses to filter through freely (Cone, 2009) but it may become a potential barrier for larger molecules and, therefore, drug delivery. Mucus is constantly secreted, but the amount and the thickness of the layer
PT
varies depending on the organ, with highest levels of secretion in the GI tract (Sigurdsson et
RI
al., 2013). Mucus secretion alters in some disease states, an example of which is cystic
SC
fibrosis, a disease where there is an increase in mucus production. The rate of secretion is important to the development of site specific drug delivery systems, particularly when
which may affect the absorption of the drug.
NU
targeting the GI or reproductive tracts, both of which have faster mucosal layer turnover rates
MA
The mucoadhesive properties of mucus are due to the presence of glycoproteins called mucins. There are two types of mucins: soluble and membrane bound. Mucins are secreted by
D
goblets cells in the epithelium. Mucins are highly glycosylated, large molecules which are
PT E
negatively charged at physiological pH, ~7.4. They have domains which are cysteine rich which are targeted in mucoadhesion using thiolated polymers due to the formation of
CE
intermolecular disulphide bonds, with secondary bonding, such as hydrogen bonding, Van
AC
der Waals forces, also coming into play. 2.1.1 Areas of administration The oral mucus is easily accessible for drug delivery and devices, such as implants or inserts, can be easily removed, if necessary. The surface area is relatively small, but it is highly vascularised leading to rapid drug absorption. Both the buccal (the cheek tissue) and sublingual (under the tongue) routes bypass the GI tract degradation and the hepatic first pass metabolism, both of which often lead to decreased drug plasma levels and are important considerations in drug delivery. The sublingual area is constantly being washed by saliva, 5
ACCEPTED MANUSCRIPT which may impede adhesion or wash the polymer tablet away. It has been observed that involuntary swallowing and increased saliva production can decrease the quantity of drug absorbed. Similar to the oral mucosa, ocular mucosa has the advantage of being easily accessible. However, the eye is a notoriously difficult target for drug delivery as structural barriers and
PT
functional barriers, such as drug clearance by the conjunctival tissues, affect the delivery of
RI
drug to the target tissue (Anderson et al., 2010). Topical preparations, eye drops etc. are the
SC
most convenient for patient use, however, they have a very short contact time on the eye and approximately 90% of the administered dosage is cleared within two minutes from blinking
NU
and lacrimation. Topical preparations require frequent administration and there is often a phase of over-dosage followed by under-dosage. The use of mucoadhesive systems for drug
MA
delivery could be highly valuable, particularly for drug delivery to the posterior segment of the eye, due to the increased residence time of the device on the surface of the eye. Hornof et
D
al. (2003) developed a thiolated PAA-cysteine ocular insert containing diclofenac samples
PT E
and evaluated the inserts in human volunteers. Thiolated PAA inserts were compared to unmodified PAA insert controls. Inserts were positioned in the cul-de-sac of the eye and
CE
various different physical attributes of the insert were analysed including sensation, blurring
AC
and irritation. The controls dissolved within 15 min of administration whereas the thiolated inserts were retained for up to 8 h. The drug release and drug removal patterns of fluorescein, a fluorophotometric agent, from the thiolated PAA insert, an unmodified insert and aqueous eye drops were all compared. The eye drop administration resulted in a high concentration of drug immediately after administration which decreased substantially after 2 h. The unmodified insert had a similar profile to that of eye drops, with an uncontrolled release of drug to the surface of the eye and fast removal. This was due to the lack of cohesion and quick disintegration of the insert. It was in stark contrast to the thiolated polymer which had a 6
ACCEPTED MANUSCRIPT more prolonged and controlled release of the fluorescein over 7 h. Commercially, Pilogel Ophthalmic Gel®, a pilocarpine HCl based gel, has the addition of Carbopol 940, a high MW polyacrylic acid in its formation, added to increases the residence time of the gel on the surface of the eye due to its inherent mucoadhesive properties. Similarly, Novartis’ Nyogel®, another eye gel, uses carbomer 934 in its formulation. A thiolated chitosan based eye drop,
PT
Lacrimera® (Chroma-Pharm), has also been developed and released onto the European
RI
market as a daily treatment to dry eye syndrome, lubricating for approximately 12 h.
SC
In the nasal cavity, mucus turnover is as fast as every 15 - 20 min and as much as two litres of nasal mucus is produced each day (Ugwoke et al., 2005), which could cause complications
NU
with regards to drug delivery through this route. The microvilli in the nasal cavity increase the surface area and increase the area of absorption. The epithelium in the cavity is thin,
MA
porous and highly vascularised (Ugwoke et al., 2005). Many factors, including environmental factors, such as temperature or inhaling cigarette smoke, and pathological factors, (such as
D
allergies, having a cold or asthma), all affect the rate of elimination and of mucus production,
PT E
i.e. lower temperatures decrease both the viscosity of mucus and its production. These could all have a great effect on the use of mucoadhesive devices for nasal drug delivery. Drugs
CE
administered via this route are capable of passing through the olfactory region of the brain
AC
and, consequently, pass into the cerebrospinal fluid (CSF) for clearance which is renewed approximately 4 - 5 times a day. Absorbed substances are quickly circulated with little restriction and in certain cases, substances are capable of travelling directly into the central nervous system (CNS), bypassing the blood-brain barrier (Ugwoke et al., 2005). Enzymatic activity is lower via the nasal route in comparison to the GI tract and because of this, there is a higher bioavailability. In terms of mucoadhesion, initially the turn-over rate of nasal mucosa slows down with mucoadhesion. However, over time, this is reversed and the production and elimination rates return to normal (Ugwoke et al., 2005). A mucoadhesive 7
ACCEPTED MANUSCRIPT tablet system for nasal administration will affect the mucociliary transport system (the elimination system). Mucoadhesion opposes this system and prolonged contact, which is key to the design of such a system, may cause irritation and even toxicity. This does not happen with the administration of an aerosol, as the contact time is minimal. The GI tract is possibly the most important area of administration, as oral administration is
PT
the preferable route of drug administration for both patients and doctors. Oral delivery is the
RI
easiest route for administration i.e. does not need professional administration and is most
SC
likely to have higher patient compliance (Patel et al., 2011). The GI tract includes the mouth, stomach and small and large intestines. It has a large surface area, increased by the presence
NU
of microvilli within the intestine which also increases the total rate of absorption. Mucosal turnover rate is fast and mucosal density varies throughout the GI tract (Ensign et al., 2012).
MA
In the stomach, mucus is secreted to protect the cell lining from the acidic nature of the stomach. Within the mucosal layer of the stomach, bicarbonate ions are released and trapped
D
which creates a pH gradient between the lumen (pH 1 – 2) and the mucosal layer (pH 6 – 7).
PT E
Prostaglandins are also released locally which promote the excretion of both more mucus and bicarbonate (H.P. Rang et al., 2003). Depending on the mobility in the GI tract, a specific
CE
area of the colon can be targeted for drug release either by coating the polymer with an
AC
enteric coating to prevent drug release in the upper GI due to low pH levels, or by increasing the crosslinking on the polymer matrix and thus slowing drug release from the matrix.
2.2 Adhesion theories In general, there are six theories of adhesion to the mucosal surface (Roy et al., 2009; Andrews et al., 2009; Khutoryanskiy, 2011): electronic theory, wetting theory, adsorption theory, diffusion theory, fracture theory, mechanical theory. All of the above adhesion theories are valid and it is most likely a combination of all six that result in adhesion. There are two stages of adhesion: the wetting/contact stage and the consolidation stage (Smart, 8
ACCEPTED MANUSCRIPT 2005). The wetting stage is the intimate contact between the polymer and the mucosal surface, which is required for mucoadhesion (Ludwig, 2005). Interpenetration of the adhesive polymer occurs within the mucus followed by non-covalent bonding, which is due to the presence of hydrophilic groups on the polymer. The extent of the swelling of the polymer and the penetration of the polymer chains into the mucosal surface is often governed by the initial
PT
contact time. If there is an increase in the initial contact time, a stronger adhesive bond is
RI
created (Asane et al., 2008). In the case of easily accessible areas, this intimate contact can
SC
occur by holding the polymer onto the surface until adhesion is apparent. For less accessible areas, e.g. GI tract, the movement of the muscle itself will allow for adhesion (Smart, 2005).
NU
The consolidation stage is important when a strong adhesion is required, particularly when the target area is under constant stress, for example targeting the blinking eye. Mucoadhesive
MA
strength is strongest when the polymer itself is completely dehydrated and is placed onto the mucosal surface. The polymer will begin to absorb the moisture from the mucosal layer
D
which, once dehydrated, switches from having lubricating properties to having adhesive
PT E
properties. The moisture will allow the polymeric matrix to become pliable and to adapt to the shape of the surface it is on, thus increasing the contact surface area. It will also allow for
CE
the formation of bonds between the mucus and the polymer (Smart, 2005).
AC
2.3 Methods of measuring mucoadhesion Once modified, the thiolated polymers must be tested for their mucoadhesive properties. A number of methods have been outlined; mostly commonly used in vitro methods include the use of a rotating cylinder method, rheology, tensile strength tests and flow-through tests. The rotating cylinder test was first described by Bernkop-Schnürch et al. (1999) where a cylinder was submerged in a phosphate buffer at 37 °C. To this cylinder, segments of porcine gastrointestinal tract were glued and to these, the thiolated PAA discs were allowed to adhere.
9
ACCEPTED MANUSCRIPT The cylinder was set to rotate and the time taken for the thiolated PAA discs to dissolved or dislodge was measured. This gave an indication of mucoadhesive properties in terms of time when compared to an unmodified or control sample. Rheological mucoadhesive testing gives an indication of the strength of the bond formed
PT
between the polymer and the mucin solution when compared with the mucin solution and the polymer solution on their own, with observed increased viscosity (Iqbal et al., 2012a) and
RI
increased storage (G') and loss modulus (G'') (Duggan et al., 2016b; Duggan et al., 2016a)
SC
occurring in a thiolated polymer-mucin mix. Rheology can also give an indication of the structural bonds occurring within the polymer-mucin mix, while allowing for a direct
NU
comparison of bond strength between different polymers (Duggan et al., 2016b). Adhesion
MA
forces between polymers and mucin have also been measured using atomic force microscopy (AFM), where fully hydrated pectin samples were evaluated for their mucoadhesive
D
properties and compared to PAA as a positive control (Joergensen et al., 2011).
PT E
Tensile testing, measuring the total work of adhesion and maximum force of detachment, are often conducted as a measurement of mucoadhesive forces. This can be conducted using a
CE
texture analyser (Ivarsson and Wahlgren, 2012), or a similar setup (Palmberger et al., 2007) where a segment of mucosal tissue is attached to the base of a beaker submerged in a buffer
AC
at physiological pH and temperature; the polymer is attached to a steel disc, hanging on a string off a retort stand. The polymer is allowed to adhere to the mucosal tissue, before slowing lowering the mucosal tissue, thus measuring the total work of adhesion and maximum force of detachment (Palmberger et al., 2007). Flow through or wash away tests give an understanding of mucoadhesion under stress, due to the flow of fluid over the polymeric tablet. This test is particularly useful when targeting areas, such as the eye where constant lacrimation and blinking could easily dislodge the
10
ACCEPTED MANUSCRIPT mucoadhesive tablet (Khutoryanskiy, 2011). This style of testing can also evaluate drug release in areas where flow of biological fluid is constant (Horvát et al., 2015).
2.4 Factors affecting mucoadhesion There are a number of factors that must be considered when designing a mucoadhesive polymer. Environmental factors, including pH, can affect the binding properties of the
PT
polymer to the mucosal surface. Properties of the polymer itself, such as molecular weight
RI
and structural modification of the polymer, can have an effect on the adhesive properties, the
SC
degree of and rate at which the polymer swells, and drug release profiles. The biodegradable properties of polymers, specifically synthetic polymers, are important for the removal of the
NU
carrier from the body and the degradation rate of the polymer may have an effect on the drug
MA
release profile.
2.4.1 Charge on the polymer Commonly used polymers have either an anionic or cationic charge, but some non-ionic
D
polymers are also utilised. Thiol groups are introduced to anionic or cationic polymers
PT E
through different reaction processes. Amide bonds are easily formed from the carboxylic acid groups on anionic polymers and thiolation can be achieved by the introduction of cysteine,
CE
mediated by carbodiimide coupling reagent (Bernkop-Schnürch, 2005). A complication with thiolation is the formation of disulphide bridges during the synthesis. This is due to the
AC
unintended oxidation of the thiol groups and may be avoided by performing the reaction under inert conditions or with the addition of EDTA in the buffer solution to chelate any metals that may catalyse the oxidation process. Cationic polymers are a desirable polymer type to use as they encourage ionic interactions between the positive charges of the polymer and the negative charges, such as the sialic acid residues, of the mucosal layer, (Makhlof et al., 2008), and, in the case of nanoparticles, increases the rate of internalisation into the cell (Kumari et al., 2010). A commonly used cationic polymer is chitosan. Thiolation can occur
11
ACCEPTED MANUSCRIPT through the formation of an amide bond, similar to the anionic polymers through the coupling of cysteine in the presence of a carbodiimide, or through the amidine bond with the reaction of Traut’s reagent (2-iminothiolane) (Bernkop-Schnürch, 2005). 2.4.2 pH Studies on the effects of pH changes have been conducted on various aspects of
PT
mucoadhesion: On its effect during the production stage of the polymer, and also its effect on
RI
the mucoadhesive and thiolation properties of the polymer. Grabovac et al. (2005) conducted a study of 19 different mucoadhesive polymers, including polyacrylic acid (PAA), chitosan,
SC
polycarbophil, thiolated PAA, thiolated chitosan and thiolated polycarbophil; thiol content of
NU
the thiolated samples were 498.8 µmol/g, 243.4 µmol/g and 95.7 µmol/g, respectively. The polymers were dissolved in D.I. water and the pH of the polymer solutions was adjusted to
MA
either pH 3 or pH 7, after which the polymers were dried using various methods. The polymers were compressed into discs and the mucoadhesive properties were evaluated in a
D
phosphate buffer at pH 6.8. It was observed that overall the thiolated polymers had a greater
PT E
level of adhesion in comparison to the unmodified samples due to the formation of intermolecular disulphide bonds between the thiolated polymers and mucin. The thiolated
CE
samples which had been adjusted to pH 3 displayed greater mucoadhesive properties than those adjusted to pH 7. It was surmised that the lower pH inhibited the formation of
AC
intramolecular disulphide bonds. Once the polymer solutions had been adjusted to pH 7, the tendency towards disulphide bond formation increased, crosslinking the polymers. Therefore, there were fewer free thiol groups to interact with mucins during mucoadhesive testing within the samples adjusted to pH 7. The pH 3 samples were less reactive and, once tested in the phosphate buffer (pH 6.8) for mucoadhesive properties, had a greater ratio of free thiol groups available to bond with mucin, thus increasing mucoadhesion.
12
ACCEPTED MANUSCRIPT Guggi et al. (2004) examined the influence pH had on pre-thiolated PAA samples; the thiol content of the samples prior to the pH change was not stated. The PAA-cysteine conjugates were dissolved in water, aliquots of which were then adjusted to different pHs between 3 and 8 and the samples were lyophilised. By reducing the samples with NaBH4, the disulphide bond formation was also measured. After the pH adjustment of the samples, the levels of
PT
thiolation were measured and cohesive and mucoadhesive properties were also examined. It
RI
was shown that the degree of thiolation decreased when the samples were adjusted to a higher
SC
pH. Therefore, the sample adjusted to pH 3 displayed the highest levels of thiolation while the lowest levels were observed when adjusted to pH 8. It was also noted that the disulphide
NU
bond content of the samples decreased with increasing pH; therefore, the samples adjusted to pH 8 displayed lower levels of disulphide bond content than that of the samples adjusted to
MA
pH 3. It was thought the formation of sodium carboxylate substructures within the samples which were adjusted to higher pH values increased the molecular weight of those samples,
D
resulting in the lower thiol content observed. The swelling abilities of the higher pH samples
PT E
were greater than that of the lower pH samples, and this also correlated to an increase in disintegration times with the higher pH samples displaying higher levels of cohesion.
CE
Mucoadhesion was greater in lower pH samples. It was concluded that the increase in molecular weight and faster swelling ability of the less acidic PAA-cysteine samples affected
AC
the mucoadhesive and cohesive nature of those samples. It was also thought that the lower pH levels of samples below pH 5 prevented the premature oxidation of the thiol groups within those samples, which allowed for higher levels of mucoadhesion and cohesion. It was, therefore, concluded that higher mucoadhesion could be obtained when the pH of the samples was adjusted to pH 3, but greater levels of cohesion and water uptake would require a higher pH. 2.4.3 Molecular weight and chain length 13
ACCEPTED MANUSCRIPT Molecular weight (MW) can influence the mucoadhesion and cohesive properties of a polymer due to the increase in chain length. MWs of between 100 kDa and 4000 kDa were indicated as being optimal for mucoadhesion by Smart (2005). The MW of modified polymers is equally important; thiolated polymers, both anionic and cationic were analysed and it was observed that thiolated polymers with medium MW, between 400 kDa and 600
PT
kDa, had the strongest mucoadhesion (Bernkop-Schnürch, 2005). The MW of the polymer
RI
may also effect the diffusion of the polymer into the mucus. It must also be noted, however,
SC
that it is difficult to define an optimum MW as each polymer system reacts differently and has a unique structure (Andrews et al., 2009); what is optimal for one polymer, may not be
NU
for another.
Chain length of the polymer and flexibility of the chain are considered to play an important
MA
role in mucoadhesion. Adhesive properties increase with a longer and more flexible chain (Roy et al., 2009). This is due to the interpenetration of the chains into the mucosal layer,
D
allowing for greater ease of interaction between functional groups of the polymer and mucins.
PT E
The diffusion theory of mucoadhesion suggests the diffusion of polymer chains into the mucosal layer and also the diffusion of the glycoprotein chains of mucin into the polymer
CE
matrix both occur. However, diffusion occurs more frequently with shorter polymers chains
AC
and chain lengths which are excessively long may impede mucoadhesion, as diffusion into the mucosal layer is lost (Huang et al., 2000). Crosslinking of the polymer reduces the flexibility of the chain and high degrees of crosslinking can reduce the levels of mucoadhesion due to fewer interactions with and less diffusion into the mucosal layer (Andrews et al., 2009). 2.4.4 Swelling Once the polymer has adhered to the mucosal layer, it will begin to extract the moisture from the surface and start to swell. This is advantageous as the surface area between the polymer 14
ACCEPTED MANUSCRIPT and the mucosal layer will increase (Andrews et al., 2009). By modifying the polymer with crosslinking moieties, such as the introduction of thiol groups, the degree of swelling can be controlled (Huang et al., 2000). This is an important consideration when creating a hydrogel type polymeric structure, as the release of drug is due to the uptake of water and swelling of the matrix. Polymers that are activated by moisture, such as hydrogels, risk becoming overly
PT
hydrated and may lose their adhesion to the surface. Non-specific bioadhesive polymers are
RI
moisture activated and can bind to both the mucosal layer and to cell surfaces (Roy et al.,
SC
2009; Smart, 2005).
NU
3 Mucoadhesive Polymers
A number of polymers, both natural and synthetic, have mucoadhesive properties without
MA
modification. These include chitosan, polyacrylic acid (PAA) and pectin.
3.1 Chitosan
D
Derived from the deacetylation of chitin (Makhlof et al., 2008), chitosan, Figure 1 (A), is the
PT E
most commonly used material for mucoadhesive polymers. It is a naturally occurring polysaccharide which is cationic and hydrophilic in nature. Cationic polymers are thought to
CE
be far superior mucoadhesives due to the interactions between the positive charge of chitosan
AC
and the negatively charged residues of the mucosal layer, such as sialic acid (Ludwig, 2005; Nagarwal et al., 2009). Chitosan has many advantages for use in drug delivery, including its biodegradability, biocompatibility and non-toxic properties (Ludwig, 2005). It is also seen to have antibacterial properties which is advantageous, particularly in ocular drug delivery where infections are common (Ludwig, 2005). As a polymeric delivery system, it is well tolerated in the body, highly stable, and has the capability to enter epithelial cells making it an excellent carrier for drugs. In oral delivery, chitosan is also seen to improve the delivery of hydrophilic 15
ACCEPTED MANUSCRIPT drugs, such as peptides and proteins, as it has the ability to open up tight junctions in the (Nagarwal et al., 2009)mucosal layer by affecting para- and intracellular pathways (actin pathways) (Dodane et al., 1999). It is available in a range of MW, with weights between 10 450 kDa being normal for drug administration (Makhlof et al., 2008). Chitosan is easily modified under mild conditions and is capable of incorporating macromolecules into the
PT
polymeric framework and is, therefore, useful for drug, protein and peptide delivery
RI
(Salamanca et al., 2006). It is insoluble at neutral and high pHs and soluble at low pHs. Once
SC
it is thiolated, it has cohesive properties and also displays gel-like properties in solution (Bernkop-Schnürch et al., 2004b).
NU
3.2 Polyacrylic acid (PAA)
PAA, Figure 1 (B), is an anionic polymer. The main method of adhesion of anionic polymers
MA
is due to hydrogen bonding between the polymer and mucins (Ludwig, 2005). Interactions of PAA with mucins are strongest at an acidic pH, indicating that the polymer in its protonated
D
state is best for mucoadhesion (Ludwig, 2005). This was documented further when the
PT E
addition of urea disrupted the hydrogen bonds between the polymer and mucus resulting in a reduction in mucoadhesion (Mortazavi, 1995). PAA is soluble across a broad range of pHs
CE
(Makhlof et al., 2008). It is not absorbed in the GI tract but it has been shown to have antienzymatic properties, particularly to the endopeptidase, trypsin (Luessen et al., 1995). This
AC
property is ideal for the delivery of protein and peptide drugs through oral administration as the polymer will protect the peptide drug from degradation. PAA is a commonly used polymer, and one of the first polymers used, in the production of thiolated polymers, or thiomers as they are also termed. The mucoadhesive properties of a thiolated polymer depend on the degree of swelling and of crosslinking of the polymer. The mobility and the flexibility of the polymer chain are also of great importance for mucoadhesion, as the chain will
16
ACCEPTED MANUSCRIPT influence the interactions between the polymer and the mucus. Crosslinking will reduce the flexibility of the chain and may have an effect on the mucoadhesive properties.
3.3 Pectin Similar to chitosan, pectin, as shown in Figure 1 (D), is a polysaccharide isolated from citrus peel or apples. This natural polymer has non-toxic properties and is often used within the
PT
food and pharmaceutical industries. Pectin is an anionic, complex heteropolymer which
RI
contains residues of galacturonate and rhamnose and has high levels of carboxyl groups along
SC
its backbone. Due to the levels of hydroxyl groups, pectin is also highly hydrophilic making it ideal for swelling in drug delivery (Pratt and Cornely, 2011).
NU
Modification of the pectin backbone often occurs through methoxylated or amidated carboxyl
MA
groups, and the degree of modification can allow for altered levels of mucoadhesion (Joergensen et al., 2011). Modification can occur under mild conditions due to pectin’s water
D
solubility. The MW of pectin has been proven to be important in mucoadhesion; using
PT E
pectins of varying MW ranging from 80 – 200 kDa and also with varying levels of modification, Thirawong et al. (2007) investigated the mucoadhesive properties of pectin using a texture analyser. Samples were classified as either high methoxy pectin or low
CE
methoxy pectin and one low methoxy pectin sample also had a degree of amidation. In
AC
general, the high methoxy pectin samples showed improved mucoadhesion in comparison to the low methoxy samples. The addition of amide groups to the low methoxy sample did also improve mucoadhesive properties. The results were in contrast to previous studies where low methoxy samples had higher mucoadhesive properties and it was suggested that the MW of the samples used by Thirawong et al. had a marked influence on the mucoadhesive properties of the samples; previous studies examined mucoadhesive properties of varying methoxylation but comparable MW (Liu et al., 2005). The highest level of methoxylation and the highest MW demonstrated the greatest mucoadhesive properties, thus showing the importance of 17
ACCEPTED MANUSCRIPT polymer properties in mucoadhesive drug delivery. Interestingly, the area of GI tissue that the samples were tested on was also of importance. Tissue was taken from the whole length of the GI tract, from the mouth to the large intestine, and pectin samples were tested on each section for mucoadhesion. Samples of chitosan and carbomer946P, known mucoadhesive polymers, were also tested as positive controls in the mucoadhesive testing. Each tissue
PT
sample was submersed in media relevant for the area, i.e. lower pH buffer for stomach tissue
RI
in comparison to intestinal tissue etc. Mucoadhesion was noted to be highest in the large
SC
intestine and there was little variability between the buccal, stomach and small intestine tested. It was thought that the variability in mucosal levels between the different areas of the
NU
GI tract played a marked role in the mucoadhesive levels of the samples; the large intestine has a high ratio of goblet cells resulting in high levels of mucus.
MA
3.4 Thiolated Polymers (Thiomers)
The introduction of compounds which contain free thiol groups to the polymer backbone
D
further increases mucoadhesion, allowing for the formation of strong disulphide bonds
PT E
between the modified polymer and the cysteine-rich mucin domains of the mucosal surface, prolonging adhesion of the polymer to the mucosal layer. Ideally for drug delivery systems
CE
and mucoadhesion, a thiol content value of at least 400 µmol/g of polymer is required, although mucoadhesion is improved with a higher level of thiolation. The degree of intra- and
AC
intercrosslinking depends on the amount of free thiol moieties present on the polymer backbone; the more thiol groups attached, the more cohesive and adhesive the polymer. However, unbound thiol will disrupt mucoadhesion, so it is imperative that molecular residues which possess free thiol groups be removed from the polymer prior to use. Leitner et al. (2003) investigated the mucoadhesive properties of thiolated PAA in comparison to a control PAA sample. All samples were dissolved in water with and without the addition of 1% cysteine (m/m), therefore, the affect unbound cysteine may have on mucoadhesive 18
ACCEPTED MANUSCRIPT properties of thiolated polymers was also examined. The samples were frozen and lyophilised. Mucoadhesive testing was conducted on the tabletted samples with and without the addition of cysteine and, as shown in Figure 2, the presence of unbound cysteine had a significant influence on the mucoadhesive properties of the thiolated sample, decreasing adhesion by up to 50% due to the interruption of the polymer-mucin disulphide bond. The
PT
addition of cysteine to the control sample had no effect on mucoadhesion, thus highlighting
RI
the importance of the removal of unbound thiols from the thiolated polymer matrix.
SC
Adhesion of the polymer to the surface is futile if there is no cohesion in the polymer itself (without the polymer itself binding together, it is unlikely to interact and bind with mucus).
NU
Thiomers show a high level of cohesion, due to the intramolecular disulphide bond formation. Bernkop-Schnürch and Steininger (2000) compared the cohesive properties of
MA
thiolated polycarbophil and carboxymethylcellulose and unmodified samples. It was shown that the disulphide bonds formed within the thiolated polymers created a much more effective
PT E
D
and beneficial delivery system than the unmodified versions. The hydrophobic and biodegradable polymer polylactic acid (PLA) is often modified to form a co-polymer with polyglycolic acid (PGA), poly(lactic-co-glycolic acid) (PLGA), Figure 1
CE
(C). Controlled release from PLA is problematic, and, similar to many biodegradable
AC
polymer matrices, appears to have a triphasic pattern of release, with an initial burst of drug once the polymer implant is administered into the body followed by a slow and sustained release. PLGA, itself, is not mucoadhesive; however, PLGA nanoparticles were synthesised and coated with known mucoadhesive polymers, either chitosan or PAA, to form mucoadhesive drug delivery systems (Kawashima et al., 2000). The nanoparticles were loaded with protein-based drugs and the mucoadhesive properties of the particles were tested on the intestinal tissue of a rat. The particles coated in chitosan showed improved mucoadhesion in comparison to those coated with PAA and it was surmised that the increase 19
ACCEPTED MANUSCRIPT in mucoadhesion was due to the electrostatic interactions between chitosan and the mucosal surface. Similarly, Grabovac and Bernkop-Schnürch (2007) coated PLGA nanoparticles with thiolated chitosan to increase its mucoadhesive properties further. The PLGA nanoparticles were initially coated with chitosan by reaction with EDC and the chitosan-coated particles were then thiolated using Traut’s reagent. Using a wash-away method on porcine intestinal
PT
tissue, the mucoadhesive properties of the nanoparticles were investigated, using both PLGA
RI
nanoparticles and the thiolated chitosan-coated PLGA nanoparticles. Over a three hour
SC
period, the percentage of PLGA nanoparticles adhering to the intestinal tissue decreased, with approximately 50 % detaching between hours one and two. The thiolated chitosan-coated
NU
PLGA particles did not show a significant decrease in detachment over the experimental period.
MA
The crosslinking action of thiolation changes the degree of swelling of the polymer, theoretically allowing for a more sustained and controlled release of drug. Thiolation will
D
prevent over hydration and change the degradation pattern of the polymer, particularly
PT E
important in hydrogels (Tabata and Ikada, 1998). However, high degrees of crosslinking may affect the mucoadhesive properties of the polymer, as the chain length and flexibility are
CE
altered. A commonly used molecule in the creation of a thiomer is L-cysteine (Figure 3 (A)).
AC
3.4.1 Thiolation with cysteine Leitner et al. (2003) investigated the properties of cohesion, mucoadhesion, swelling and disintegration on linear PAA-cysteine conjugates of different MW and, therefore, different chain lengths, ranging from 2 kDa PAA to 450 kDa PAA, and a polycarbophil-cysteine conjugate of MW 750 – 3000 kDa. All polymers had comparable levels of immobilised thiols to allow for a more direct comparison, and results were also compared to unmodified PAA controls. Mucoadhesive studies were analysed by two different methods: Tensile studies and a rotating cylinder method using porcine intestinal mucosa. Both methods concluded that 20
ACCEPTED MANUSCRIPT PAA with MW of 250 kDa and 450 kDa exhibited the highest mucoadhesive levels. In addition to these tests, the amount of residual unbound cysteine and its influence on mucoadhesion was measured using 450 kDa PAA (both thiolated and unmodified PAA). Free cysteine had little to no effect on the mucoadhesion of the unmodified PAA controls, however, the addition of 2.5% free cysteine showed a significant decrease in mucoadhesion
PT
time in comparison to 0% free cysteine on the thiolated PAA sample, again highlighting the
RI
need to completely remove any unbound cysteine during the creation of a mucoadhesive drug
SC
delivery system. Thiolated PAA 450 kDa sample showed the highest levels of mucoadhesion and excellent cohesion. Although the low MW PAA had strong penetration into the mucosa
NU
due to the flexibility of the chain, cohesive properties were poor and mucoadhesion was affected. Similarly, higher weighted polymers had better cohesion but poor penetration due to
MA
the lack of chain flexibility which, again, affected mucoadhesion. In order to evaluate the effect of varying amounts of L-cysteine bound to PAA, five different
D
PAA-cysteine conjugates were synthesised, each with a different amount of immobilised thiol
PT E
groups (Palmberger et al., 2007). Levels of thiolation (measured in µmol/gram of polymer) bound to the 450 kDa PAA ranged from 53 µmol/g up to 767 µmol/g. Various studies,
CE
including tensile and mucoadhesive studies, were performed. The degree of thiolation was
AC
measured using two methods: 1. Ellman’s reagent to measure the amount of free thiol groups on the PAA backbone and 2. TNBS (trinitrobenzene sulfonic acid) to measure the amount of remaining unbound cysteine in the reaction solution by measuring the amine groups on cysteine. The total amount of thiol groups on the polymer was also determined by reducing the disulphide bonds with sodium borohydride and then analysing with Ellman’s reagent. It was concluded that both mucoadhesion and disintegration of the conjugates strongly depended on the amount of thiol groups, with higher levels of thiolation resulting in longer times of mucoadhesion and disintegration. Mucoadhesion studies were all carried out at pH 21
ACCEPTED MANUSCRIPT 6.8 as intermolecular disulphide bonds between the polymer and the mucus can easily be formed at this pH. It was also discovered that without the addition of the coupling reagent, 1Ethyl-3-(3-dimethylaminoprpyl) carbodiimide hydrochloride (EDC), no amide bonds were formed between the polymer and the cysteine. Kafedjiiski et al. (2007) showed the significant difference between crosslinked polymers and
PT
non-crosslinked polymers in relation to rates of degradation and adhesion times. The polymer
RI
used was hyaluronic acid (HA) and was crosslinked using L-cysteine ethyl ester
SC
hydrochloride. Degradation of HA is due to hydrolysis by hyaluronidase. Both the thiolated and unmodified HA rates of degradation were slow but the thiolated polymer rate was
MA
process inhibited enzymatic hydrolysis.
NU
significantly slower, indicating that the crosslinked disulphide bonds formed in the thiolation
Grabovac et al. (2005) showed thiomers, in general, had higher mucoadhesive properties than
D
their equivalent unmodified polymers and that the mucoadhesion was pH dependent. At a pH
PT E
over 5.0, the thiol groups in the polymer were more reactive and began to form intramolecular disulphide bonds and in turn did not form them with the cysteine rich mucin domains. At pH 3.0, the thiols groups were much less reactive and formed bonds with the
CE
mucosal layer. Bernkop-Schnürch (2005) explained that the pH of the medium surrounding
AC
the surface of the polymer controlled the reactivity of the polymer as a whole, whereas the pH of the polymeric framework controlled the reactivity of the thiol groups. With regards to drug delivery systems, the pH of the epithelium would always be around pH 7 and the thiol groups would be reactive enough to penetrate mucus. Thiomers form inter- and intramolecular disulphide bonds at physiological pH due to the oxidation of the thiol groups, classified as in situ gelling properties. Polymers that demonstrate a phase transition, sol-to-gel transition, due to the change in the environment or physiochemical changes (pH, temperature) have been designed. Srividya et al. (2001) synthesised PAA based eye drops that were pH 22
ACCEPTED MANUSCRIPT triggered and changed phase (liquid to gel) once they had come in contact with the eye. The carrier drug used was ofloxacin, an ocular anti-bacterial agent usually given topically as an eye drop. PAA was used in this experiment due to its capacity to form a stiff gel at elevated pH. The trigger in this case was the change from the pH at formulation, pH 6, to the pH of the eye, pH 7.4. The irritation levels of the PAA formulation was measured using the Draize
PT
technique on rabbits. No ocular irritation was observed.
RI
Krauland and Bernlop-Schürch (2004) showed a 3-fold improvement in residence time of
SC
thiolated PAA, as shown in Figure 3 (B), versus unmodified PAA in the small intestinal mucosal layer. This study used insulin as the model peptide drug for delivery and combined a
NU
copolymer of Eudragit RS® (a copolymer consisting of ethyl acrylate and methyl methacrylate) into the thiolated PAA matrix. Previous studies using insulin and thiolated
MA
PAA without Eudragit RS® had observed that the insulin had burst-released within a number of minutes, before the polymer had adhered to the surface. A more controlled and prolonged
D
release of insulin was observed when Eudragit RS® was incorporated into the microparticles.
PT E
The delivery of insulin is a constant problem, with oral delivery of the peptide not being
injection.
CE
viable due to the denaturation of the protein. Current delivery is only by subcutaneous
AC
Grabovac et al. (2008) designed a three layered mucoadhesive patch for delivery of insulin by oral administration and mucoadhesion to the small intestine. The mucoadhesive patch consisted of a polycarbophil-cysteine matrix within which was a mixture of insulin, mannitol and gluthatione, all dissolved in water. Due to the small surface area of the patch and low levels of water uptake and, therefore, limited swelling of the polymer at the site of action, a single dosage of insulin was created by sticking two separate polymeric matrices together which were then surrounded by ethylcellulose and then Eudragit, which acts as an enteric coating and has a pH threshold of 5.5. Plasma insulin levels and blood glucose levels over 23
ACCEPTED MANUSCRIPT time were examined in three separate cohorts of rats; one cohort receiving the polymeric patch, another had subcutaneous delivery of insulin and finally, oral administration of an insulin solution used as the control group. The polymeric matrix showed a more sustained insulin level in the blood plasma, which is to be expected as the insulin was released in a more controlled manner in comparison to the subcutaneous injection. After oral
PT
administration of insulin solution to the control group, there was no change in the insulin
RI
levels in the blood. This shows the efficiency of the mucoadhesive polymeric system.
SC
Chondroitin sulphate (CS) is a glycosaminoglycan found in bone and cartilage tissues. CS is commonly used in the treatment for joint pain and osteoarthritis, given as an oral dosage or
NU
injection to the site of pain. It was thought that prolonging the residence time of CS at the site of action would be beneficial, improving the anti-inflammatory and chondroprotective
MA
properties of this nutraceutical. Therefore, CS was thiolated using L-cysteine and EDC (Suchaoin et al., 2016a). The resulting thiolated CS sample yielded a thiol content of 421.17
D
± 35.14 µmol/g; the sample was then reduced using Tris (2-carboxyethyl) phosphine
PT E
hydrochloride (TCEP HCl) to break any disulphide bond formed during the reaction process and this sample was then dialysed and freeze dried, thus yielding a reduced thiol content of
CE
675.09 ± 39.67 µmol/g. The reduced thiolated CS sample was used throughout the remaining
AC
experiments. The rotating cylinder method on porcine tissue and rheological studies were utilised to investigate the mucoadhesive properties of the thiolated CS sample. Using the rotating cylinder method, thiolated CS showed a 2.48-fold adhesion improvement in comparison to an unmodified CS control, and it was noted that the unmodified CS detached from the intestinal tissue in one piece, whereas the thiolated CS sample gradually dislodged overtime. A CS control, mucin control and thiolated CS mixed with mucin were investigated for their viscosity levels, and a 4.57-fold increase was observed in the thiolated CS/mucin mix in comparison to the mucin solution alone, indicating the mucoadhesive interactions 24
ACCEPTED MANUSCRIPT occurring between the thiolated CS and mucin. It was thought that, although mucoadhesion was improved due to thiolation, higher levels of mucoadhesion would be observed with cationic polymers due to the charge interactions occurring between the negative residues of the mucosal layer and of the CS backbone.
PT
3.4.2 Thiolation with Traut’s reagent (2-iminothiolane) Traut’s reagent, Figure 4 (A), reacts with primary amines groups on the polymer backbone in
RI
a ring opening reaction, yielding free thiol groups. Bernkop-Schnürch et al. (2003) used Traut’s reagent to thiolate chitosan. Chitosan had previously been thiolated using thioglycolic
SC
acid, creating a thiomer with an uncharged amide bond linkage (Kast and Bernkop-Schnürch,
NU
2001); by thiolating with Traut’s reagent, a cationic linkage bond was formed which had the potential to form ionic interactions within the mucosal membrane with residues such as sialic
MA
acid or sulfonic acid, as well as the introduction of thiol groups to the polymer backbone. The reaction of chitosan with Traut’s reagent was conducted at three different pH values: pH 5,
D
pH 7 and pH 10. Chitosan was insoluble at pH 10, but a certain degree of derivatisation was
PT E
observed and a thiol content of 177.2 µmol/g was measured at this pH with the addition of 0.2 g of Traut’s reagent to 1 g of chitosan. When the same ratio of Traut’s reagent was added
CE
at pH 5, although chitosan was fully dissolved at this pH, a lower level of thiol content was observed, measuring 59.8 µmol/g; this was due to lower reactivity of Traut’s reagent at this
AC
pH. Conducting the reaction at pH 7 yielded the highest level of thiol content of the chitosan backbone, with a degree of thiolation of 225.7 µmol/g measured. When the ratio of Traut’s reagent was increased to 0.4 g/g of chitosan, a thiol content of 408.9 µmol/g was measured at pH 7. Mucoadhesive testing was conducted on porcine mucosal tissue using the rotating cylinder method and compared the mucoadhesive properties of the chitosan-Traut’s reagent conjugate samples to that of unmodified chitosan and chitosan-thioglycolic acid conjugate samples. Mucoadhesion time was increased by up to 140-fold after thiolation with Traut’s
25
ACCEPTED MANUSCRIPT reagent, which was thought to be due to the improved cationic character of the polymer, as well as the presence of thiol groups on the chitosan backbone. 3.4.2.1 Polyallylamine Polyallylamine is a synthetic, cationic polymer, shown in Figure 1 (F) as (i) its free base form, and (ii) its salt form. It is a linear polymer with primary amines along its backbone.
PT
Polyallylamine hydrochloride (PAH) is commonly used in the treatment of
RI
hyperphosphatemia in patients with end-stage renal failure. Crosslinked with epichlorohydrin
SC
and known in this form as sevelamer hydrochloride, it acts as a phosphate binder, removing excessive phosphate ions as it is excreted (Hudson et al., 2012). In this form, it is non-
NU
biodegradable but is not absorbed into the blood stream. It is also excreted unchanged and therefore, is not metabolised to toxic by-products or metabolites. In the form of a polymeric
MA
micelle, PAH was used as a delivery system to the GI tract for siRNA (small interfering RNA) (Guo et al., 2013). Polymeric micelles are emerging as a platform in the oral delivery
D
of gene based drugs as the drug is protected in the hydrophobic centre of the micelle. PAH
PT E
was modified with cholesteryl and palmitoyl moieties and also a hydrophilic ammonium moiety, creating a so called quaternised PAH product. This product contained a hydrophobic
CE
interior and hydrophilic exterior, which was thought to be more stable in an acidic environment such as the stomach; amine groups on an unmodified PAH sample would be
AC
affected greatly by this acidic environment. siRNA bound to the PAH by electrostatic interactions and its stability in simulated GI fluids was measured. Stability of siRNA in the quaternised PAH was improved in both simulated gastric fluid and simulated intestinal fluid in comparison to an unmodified PAH system, implying the protective nature of this modified PAH polymeric micelle. Although not analysed in this paper, the quaternised PAH was also seen to interact with the mucosal layer in the GI tract, again through electrostatic interactions between the polymer and mucins. This resulted in increased residence and contact time.
26
ACCEPTED MANUSCRIPT Cationic polymers are often used as vectors for gene transfer and polyallylamine is one such polymer that is utilised. Polyallylamine, however, can be toxic to cells but modification of the backbone has been observed to decrease the toxic effects (Oskuee et al., 2015) (Boussif et al., 1999). Bacalocostantis et al. (2012) thiolated polyallylamine (PAAm) with Traut’s reagent, the proposed structure of which is given in Figure 4 (B). Different ratios of Traut’s reagent
PT
were added to the reactions to result in an overall 5%, 13%, or 20% conversion of amines to
RI
thiol groups. The resulting thiolated polymer was used as a delivery vector for DNA in which
SC
the cationic domains of the polymer were electrostatically bound to anionic charged phosphate groups on DNA. Each thiolated polymer and unmodified polymer was tested for
NU
DNA binding and release rates, and protection of DNA. The 13% conversion rate showed the greatest potential, with higher DNA binding ability and better release of plasmid in
MA
comparison to both unmodified polyallylamine and to the 5% and 20% counterparts. Ibie et al. (2015) thiolated polyallylamine with Traut’s reagent and compared the thiol
PT E
D
content and mucoadhesive properties to a polyallylamine sample which was thiolated with Nacetyl cysteine/EDC and used PAH in both thiolation reactions. The pH of the PAH used was, however, adjusted with NaOH and dialysed to yield the free base form of
CE
polyallylamine, PAAm, which was then thiolated with both Traut’s reagent and N-acetyl
AC
cysteine. Similar to results observed by Vigl et al. (2009), the thiol content of the resulting samples thiolated with N-acetyl cysteine were low, measuring 60 µmol/g which was in vast contrast to the greatly improved thiolation levels observed with Traut’s reagent, which resulted in a thiol content of 490 µmol/g. Disulphide bond formation of the samples was also examined. The samples displayed high levels of disulphide bond formation, with samples thiolated with N-acetyl cysteine resulting in disulphide bond levels of 280 µmol/g and reaction with Traut’s reagent measuring 590 µmol/g formed as disulphide bonds. Mucoadhesion testing was conducted using a porcine mucin solution; solutions of both the 27
ACCEPTED MANUSCRIPT thiolated polymers and unmodified polyallylamine were mixed with the mucin solution and tested by UV/Vis spectrometry to determine the levels of adsorption of mucin to the polymer. Thiolation of the polymer displayed improved mucin interactions in comparison to unmodified samples. The percentage of mucin adsorption increased from approximately 40%
PT
in the unmodified PAAm sample to 65 – 70% in the thiolated PAAm samples. PAAm (15 kDa MW) was thiolated with varying ratios of Traut’s reagent, creating three
RI
thiolated PAAm products (Duggan et al., 2016a). Thiol content of the three products were
SC
134 µmol/g, 329.5 µmol/g and 487.4 µmol/g which equated to 1%, 2.5% and 5% total thiolation of the PAAm backbone, respectively. The three thiolated products were tested for
NU
their mucoadhesive capabilities by investigating their swelling ability, mucoadhesion on
MA
porcine tissue and rheological properties with and without the addition of a mucin solution. The swelling abilities of the three thiolated PAAm samples were fast in the initial minutes of the test, with high levels of fluid uptake, up to 1500% in the first five minutes of the swelling
PT E
D
studies. Surprisingly, the swelling abilities of the thiolated PAAm samples increased with increasing thiol content, and therefore, the 5% thiolated PAAm sample displayed the greatest degree of swelling. It was thought that a higher level of disulphide bond formation was
CE
occurring in the less thiolated sample (1% thiolated PAAm) due to steric hindrance within the
AC
polymer. Mucoadhesive testing on porcine tissue using the rotating cylinder method was investigated and, interestingly, the mucoadhesive properties, regardless of the thiol content, were similar and all three thiolated samples adhered for a minimum of 22 h. The thiolated PAAm samples were also tested for their rheological properties, both alone and mixed with an 8 % mucin solution. Although the mucoadhesive properties using the rotating cylinder method were similar, the rheological studies showed that the increase in thiol content increased the relative G', by approximately 3-fold, indicating that higher levels of interactions
28
ACCEPTED MANUSCRIPT were occurring within the 5% thiolated PAAm sample with the mucin solution in comparison to the other two thiolated PAAm samples. A further study into the properties of thiolated PAAm (15 kDa MW) was conducted by comparing the swelling abilities and mucoadhesive properties of a thiolated PAAm sample
PT
(thiol content of 487.4 ± 18.3 µmol/g) to that of thiolated PAA with comparable thiol content (400.6 ± 61.3 µmol/g) (Duggan et al., 2016b). The PAAm sample was thiolated using Traut’s
RI
reagent while the PAA sample was thiolated using L-cysteine and EDC. The thiolation of
SC
PAAm improved the swelling properties of the polymer in comparison to a control sample, although the swelling abilities of the thiolated PAAm were far greater than that of thiolated
NU
PAA, particularly within the initial minute of testing. The release of the model drug,
MA
chlorpheniramine maleate, was influenced by this increased swelling in the thiolated PAAm sample and drug release was quicker in comparison to the thiolated PAA sample. However, despite greater swelling rates, the mucoadhesive properties of the two thiolated polymers
PT E
D
were comparable, as measured by the rotating cylinder method using porcine intestinal tissue, with the thiolated PAAm samples adhering for 35 ± 18 h and the thiolated PAA adhering for 23 ± 1 h. Comparable mucoadhesive properties were also observed using rheological
AC
3.4.2.2 Gelatin
CE
methodologies when the polymers were mixed with a mucin solution.
Gelatin is a natural polymer derived from Type 1 collagen (which contains no cysteine) of animals and fish. Gelatin is hydrophilic, is degraded enzymatically and, as it is a natural polymer, it does not induce an immune response, therefore, making gelatin an excellent component for potential drug delivery applications. Depending on the pre-treatment process from the collagen, two different forms of gelatin can be produced: acidic (Type A) or basic (Type B). Each process creates a gelatin form with a different isoelectric point (IEP), the point at which a molecule carries no net charge. This varying IEP permits for a greater 29
ACCEPTED MANUSCRIPT flexibility when it comes to drug delivery systems, allowing the gelatin carrier to be of opposite charge to the drug and for the potential formation of a polyion complex, a strong bond formed due to electrostatic interactions between oppositely charged entities (Young et al., 2005; Tabata and Ikada, 1998). The amino acid composition of native gelatin is 16% carboxylate and 13% primary amine. It
PT
has a general structure similar to collagen containing repeating sequences of glycine–X-Y
RI
triplets, where X is proline and Y is often hydroxyproline, as shown in Figure 1 (E). Due to
SC
the presence of these glycine–X-Y triplets, gelatin can exist in a triple helical structure as its most stable conformation (Pratt and Cornely, 2011). As there are both free carboxylate and
NU
amine groups throughout the gelatin backbone, it offers many routes of modification to improve and create its mucoadhesive properties. Gelatin dissolves in aqueous solutions at
MA
approximately body temperatures (37 °C), forming single, flexible coils and becoming transparent upon cooling (Vlierberghe et al., 2011). Because of this, gelatin must be
D
chemically crosslinked or modified in order to become insoluble at body temperature
PT E
(Dinarvand et al., 2005). Gelatin is easily modified under mild conditions, which is ideal for drug delivery carriers. Mild conditions can also be used to load the drug into the polymer
CE
matrix (Young et al., 2005). However, when a protein based drug is present during the
AC
production of gelatin hydrogels, the proteins are seen to lose pharmacological activity due to the chemical crosslinking required in the creation of the hydrogel (Tabata and Ikada, 1998). This problem can be avoided by allowing the protein to absorb into the hydrogel postproduction. A gelatin based mucoadhesive delivery system was devised for the delivery of amoxicillin to the gastric mucosa, targeting the bacteria, Helicobacter pylori, the main cause of ulcers and gastritis (Wang et al., 2000). A constant level of antibiotic in the gastric mucosa is required to treat the condition. This study compared the mucoadhesive and drug release properties of 30
ACCEPTED MANUSCRIPT gelatin microspheres versus modified gelatin microspheres which were firstly aminated with ethylene diamine and secondly crosslinked using varying levels of gluteraldehyde (0.06%, 0.12%, 0.18%). Different crosslinking times were also measured. The unmodified gelatin microspheres had an initial burst, releasing up to 80% of its drug load within 30 min. The varying levels of crosslinked microspheres showed decreased bursts within the same time
PT
period. The highest level of crosslinkage, 0.18% gluteraldehyde, had a slower and more
RI
controlled release of amoxicillin but the increased levels of crosslinkage affected the
SC
mucoadhesion, with adhesion levels being similar to that of the unmodified gelatin. Highest levels of mucoadhesion were observed with 0.06% gluteraldehyde. This indicates that a
NU
compromise between levels of crosslinkage was required to establish an efficient mucoadhesive system which will allow for slow, controlled release.
MA
In terms of thiolation, Kommareddy and Amiji (2005) thiolated gelatin nanoparticles using Traut’s reagent for use as a non-viral delivery vector for DNA. Thiolation, in this case, was
D
used as a means of stabilising the nanoparticle; the disulphide bonds formed due to oxidation
PT E
in the body strengthened the gelatin’s structure. Upon entering a reducing environment, these disulphide bonds would break and the drug would be delivered. Because of this theory, the
CE
delivery of the model drug, fluorescein isothiocyanate dextran, was tested in the presence of
AC
the reducing agent glutathione. Additionally to this, plasmid DNA was inserted into the nanoparticles. The encapsulation efficiency and stability were measured as well as transfection into murine fibroblast cells. In cytotoxicity testing, the gelatin nanoparticles were shown to be non-toxic with up to 92% cell viability seen at high concentrations; it was observed that the higher levels of thiolation resulted in higher levels of cytotoxicity. Varying levels of thiolation were analysed for both drug release and plasmid DNA encapsulation and they were compared to an unmodified gelatin nanoparticle. Additionally to this, nanoparticles were either crosslinked or non-crosslinked. In the case of loading efficiency of fluorescein 31
ACCEPTED MANUSCRIPT isothiocyanate dextran, the thiolated nanoparticles were seen to have a higher loading capacity than the unmodified counterparts. Release was measured over a 5 h period and rates varied depending on the amount of glutathione added, with higher amounts of glutathione resulting in enhanced release when compared to release in PBS. Thiolation levels affected the percentage release, depending on the concentration of glutathione added, with unmodified
PT
particles releasing the drug much quicker. Similarly, crosslinking of the particles resulted in
RI
slowing the release of the drug in comparison to non-crosslinked particles, regardless of the
SC
amount of glutathione added. Thiolation levels did, however, affect the loading of plasmid DNA into the nanoparticles; unmodified gelatin particles had a loading capacity of 99% but
NU
this figure fell to 90% with thiolated nanoparticles of thiol content 16.41 mM/g gelatin. The plasmid DNA did remain stable within the thiolated particles and were comparable to the
MA
unmodified particles and also to naked plasmid DNA.
Vlierberghe et al. (2011) compared two different thiolating agents, N-acetylhomocysteine
D
thiolactone and Traut’s reagent, in their reaction with gelatin. The main body of analysis in
PT E
this paper was conducted using N-acetylhomocysteine thiolactone, however, thiolation levels using both compounds, N-acetylhomocysteine thiolactone and Traut’s reagent, were
CE
compared. Both thiolated gelatin products were analysed using UV/Vis spectrometry and size exclusion chromatography. Thiolation levels were analysed using both Ellman’s reagent and
AC
using the ortho-phthalic dialdehyde (OPA) method; Ellman’s reagent analysed the thiol content whereas the OPA method measured the free amine content, which when compared to unmodified gelatin, gave the degree of substitution. The OPA method indicted higher levels of substitution being present than the Ellman’s method, implying the presence of disulphide bonds in the product. When one equivalent of Traut’s reagent was added to the reaction, 70% degree of modification was achieved. This is in comparison to only 30% with Nacetylhomocysteine thiolactone. In a subsequent part of the experiment, the product thiolated 32
ACCEPTED MANUSCRIPT with N-acetylhomocysteine thiolactone was examined. In the form of a hydrogel, and thiolated at different levels of thiolation, various factors including swelling tests, rheology and texturometry were investigated. Rheological results showed that the storage modulus (G') was higher in the thiolated samples in comparison to unmodified gelatin and that as thiol content increased, so did G'; this was due to the ability of the thiolated samples to crosslink
PT
and form a gel. When comparing G' at different temperatures, room temperature (21 °C) and
RI
50 °C which is the temperature above the sol-gel transition of gelatin, it was shown that
SC
disulphide bond formation contributed alone to G' at 50 °C whereas at 21 °C, the physical properties of gelatin itself also contributed. Swelling properties decreased as the thiol content
NU
increased showing that crosslinking of the polymer influences its swelling ability. Gelatin was thiolated for use as a mucoadhesive drug delivery device using a two-step
MA
reaction process (Duggan et al., 2015). Native gelatin has a high carboxylate content and, in the initial reaction step, unmodified gelatin was reacted with ethylene diamine mediated by
D
an EDC coupling reagent to convert the carboxylic acids along its backbone to amide bonds
PT E
which contained primary amines as side chains, thus increasing the total amine content of the gelatin. In the second step of the reaction process, this aminated gelatin was then thiolated
CE
using Traut’s reagent, yielding a highly thiolated gelatin product. Due to the inherent amine
AC
content of gelatin, it was also directly thiolated with Traut’s reagent which, along with unmodified native gelatin, acted as controls. The swelling abilities and cohesive of the thiolated samples was vastly improved in comparison to control samples and the mucoadhesive properties, as measured by the rotating cylinder method on porcine intestinal tissue, were also markedly increased with up to a 10-fold increase in mucoadhesion upon thiolation using this two-step reaction (Figure 5). A range of gelatin samples with varying MW’s were thiolated using this two-step method and it was noted that MW had an important effect on both the cohesive and mucoadhesive properties of the thiolated gelatin samples 33
ACCEPTED MANUSCRIPT with, interestingly, a gelatin of MW of 20 – 25 kDa displaying the highest levels of cohesion and mucoadhesive properties.
3.5 Other methods of thiolation Polyvinyl alcohol (PVA) is commonly used in pharmaceutical formulations but does lend itself to short residence time after administration. Upon thiolation, this residence time could
PT
be improved. PVA was thiolated using two different methods of thiolation: by reaction with
RI
thiourea, and further neutralisation and acidification to yield the thiolated product, and also
SC
by reaction with 3-mercaptopropionic acid (Suchaoin et al., 2016b). Thiol content of the sample thiolated with thiourea (described as PVA1) measured 130.44 ± 14.99 µmol/g and it
NU
was observed that the reduction using NaBH4 was required as without reduction, the sample was not soluble due to the high levels of crosslinking forming in the sample. The sample
MA
thiolated with 3-mercaptopropionic acid (described as PVA2) yielded a thiol content of 958.35 ± 155.27 µmol/g. In mucoadhesive testing, both thiolated PVA samples displayed
D
higher levels of mucoadhesion in comparison to an unmodified PVA sample; PVA2 showed
PT E
longer adhesion times on porcine tissue, with an increase of 15.8-fold in comparison to the unmodified sample. This was higher than PVA1, which displayed a 3.2-fold increase and
CE
may be due to the increased free thiol content of the PVA2 sample. The swelling abilities of
AC
the two thiolated PVA samples were improved in comparison to the unmodified PVA sample, where both thiolated samples displayed a prolonged disintegration time and slower buffer uptake. The unmodified and thiolated samples were also tested for their drug release capabilities using the model drug, irbesartan, a drug to treat hypertension. Interestingly, there was no marked difference in the drug release profiles of the PVA1 sample versus the unmodified PVA sample. The PVA2 sample, with its higher free thiol content, displayed a slower release of drug, with a 0.6-fold decrease in drug release in comparison to the unmodified sample over a period of three hours. Additionally, the thiolated PVA exhibited no 34
ACCEPTED MANUSCRIPT cytotoxic effects of Caco-2 cells; after 24 h, unmodified PVA was the most cytotoxic, with PVA1 the least cytotoxic, displaying 72.8 % cell viability over the same time period. It was thought the addition of mercaptopropionic acid, and the higher thiol content, to PVA2 may have affected the mitochondria of the cells, resulting in higher cytotoxicity in comparison to PVA1. Thiolated PVA overall displayed lower levels of cytotoxicity, improved
PT
mucoadhesive properties and a slower release of drug over a period of time in comparison to
RI
an unmodified PVA sample.
SC
Carrageenan was also thiolated with thiourea (Suchaoin et al., 2015). The natural polysaccharide, which is derived from seaweeds, is known to have antiviral and antitumor
NU
growing properties. The thiolation of carrageenan could improve its residence time which, for
MA
deliveries of, for example, vaginal or nasal of antiviral medication, may be advantageous. A two-step reaction process was performed to introduce free thiol groups to the polysaccharide: bromination, followed by thiolation with thiourea. Carrageenan can be defined into three
PT E
D
groups, depending on gelling properties and position of the suphate group: λ, κ and ι; both κ and ι carrageenans were evaluated in this study. The thiol content of the κ and ι carrageenans measured 176.57 ± 20.11 and 109.51 ± 18.26 µmol/g, respectively. With the increase in thiol
CE
content, mucoadhesive properties of both the thiolated carrageenan samples were improved in
AC
comparison to unmodified samples; the κ carrageenan displayed a 6.4-fold increase in adhesion time on porcine intestinal tissue using the rotating cylinder method, while the ι carrageenan displayed a 1.8-fold increase using the same test. These mucoadhesive properties were further evaluated by rheology, in which the thiolated ι carrageenan, when mixed with a mucin solution, had an increase of 2-fold in its dynamic viscosity whereas the κ carrageenan displayed a 3.9-fold increase when compared to an unmodified carrageenan-mucin mixture. Thiogylcolic acid (TGA), with the addition of EDC, has also been used in the thiolation of chitosan (Kast and Bernkop-Schnürch, 2001). The reaction was conducted at a number of 35
ACCEPTED MANUSCRIPT different pH values ranging from pH 3 – 6 and also with different weight ratios of chitosan:TGA to investigate possible influences on the thiolation reaction. pH was shown to have a strong impact on the reaction, as below pH 5 the EDC was not as efficient and above pH 5, disulphide bond formation was higher. The different chitosan:TGA weights ratios at pH 5 also showed some interesting results: the highest degree of thiolation was observed at a
PT
1:1 ratio of chitosan:TGA. When the ratio was increased to 2.5:1 chitosan:TGA, the thiol
RI
content decreased marginally, whereas when the ratio was decreased to 1:2 chitosan:TGA,
SC
there was an approximate 50% decrease in thiol content. Swelling studies were conducted on unmodified chitosan samples, on control samples which had been treated with TGA but
NU
without the addition of EDC, and on the thiolated chitosan-TGA conjugates. The swelling study results showed no significant difference in swelling abilities between the thiolated
MA
chitosan sample and the control sample; there was, however, a difference between the control sample and the unmodified chitosan sample, and it was suggested that the change in pH
D
during the dialysis process (conducted in 5 mM HCl and final addition of 1% NaCl) on the
PT E
control sample altered the conformation of chitosan in comparison to the unmodified and undialysed chitosan sample, thus changing its swelling profile. In mucoadhesive testing, both
CE
the control and the unmodified chitosan samples demonstrated similar mucoadhesive properties. Thiolated chitosan samples showed higher levels of mucoadhesion in comparison
AC
to the control and unmodified samples and an increase in degree of thiolation increased the levels of mucoadhesion. This was displayed by both the rotating cylinder method and by tensile strength tests. Similar to PAA and chitosan, pectin is often thiolated to improve its mucoadhesive properties. Sharma and Ahuja (2011) thiolated pectin using thioglycolic acid and examined the thiolated pectin product for its mucoadhesive properties. Beads of both thiolated and unmodified pectin were synthesised and were loaded with the drug, metformin. It was 36
ACCEPTED MANUSCRIPT observed that thiolation improved the mucoadhesive nature of pectin by up to 2-fold. Pectin, due to the presence of hydroxyl groups, is naturally mucoadhesive with electrostatic interactions occurring with the mucosal layer and, in a wash-off bioadhesive test on goat intestinal tissue, the unmodified pectin beads remained adherent for up to 2.5 h. This is in comparison to the thiolated beads which remained adherent for up to 5 h using the same test
PT
procedure. Similarly, in the swelling tests, the thiolated beads had a lower rate of swelling in
RI
comparison to the unmodified samples and it was also observed that the unmodified samples
SC
eroded at a faster rate than the thiolated samples. Therefore, the lower levels of adhesion observed by the unmodified pectin beads were deemed to be due to increased swelling ability
NU
and faster erosion in comparison to the thiolated samples. Interestingly, in drug release studies of metformin, the release profiles of both the unmodified and thiolated beads were
MA
comparable. Therefore, the improved swelling ability of the thiolated sample did not result in a more controlled release of drug over time.
D
3.6 Preactivated and S-protected polymers
PT E
More recent advances in thiomer synthesis include the development of so called preactivated and S-protected thiomers. The concept of preactivated thiomers using PAA (Iqbal et al.,
CE
2012a) came to light due to the oxidation of thiols at pHs above pH 5, and since then, the mucoadhesive properties of many other preactivated and S-protected polymers have been
AC
investigated including chitosan, pectin, crosslinked PAA, alginate and gelatin (Dünnhaupt et al., 2012; Hintzen et al., 2013; Bonengel et al., 2014; Hauptstein et al., 2015; Rohrer et al., 2015). The preactivation of thiomers uses theory based on covalent chromatography where proteins were successfully linked to thiol-bearing resins once they were preactivated with pyridyl structures (Iqbal et al., 2012a; Brandt et al., 1977). The preactivation of the thiol group means it was essentially protected by a non-toxic aromatic leaving group attached by a disulphide bond, protecting the thiol from oxidation. This bond was broken by a disulphide 37
ACCEPTED MANUSCRIPT exchange between the thiol groups of the mucosal layer and the polymer thus releasing a free thiol which binds to the mucosal layer, improving mucoadhesion. The oxidation effect of thiols was seen to lessen the mucoadhesive properties of certain thiomers, therefore, protecting the thiol groups of the polymers allows for greater stability over a broader range of pH. The structure of preactivated PAA is shown in Figure 6.
PT
Iqbal et al. (2012a) investigated the preactivation of PAA, using three different MW PAA:
RI
100, 250 and 450 kDa and compared properties of hardness, swelling and mucoadhesion to
SC
thiolated PAA (thiolated with L-cysteine and EDC) and unmodified PAA over the same MW ranges. Tablet hardness was conducted at a force of 10 kN; the thiolated PAA samples
NU
showed an improvement in hardness of up to 5.4-fold in comparison to the unmodified PAA, but the preactived samples improved by up to 6.8-fold. In swelling studies, all samples were
MA
tested in both 0.1 M HCl and in 0.1 M phosphate buffer. All samples were observed to increase to a higher degree in the phosphate buffer in comparison to the HCl solution, and a
D
higher degree of swelling was also observed using the 450 kDa PAA sample, for the
PT E
unmodified, thiolated and preactivated samples. Overall, the preactivated PAA samples showed the highest degree of swelling over the longest period of time. In mucoadhesive
CE
testing using the rotating cylinder method on porcine intestinal tissue, the preactivated
AC
samples displayed the highest levels of mucoadhesion with a 960-fold increase in mucoadhesive time using the 450 kDa PAA in comparison to the unmodified PAA sample; a 452-fold improvement observed by the thiolated PAA sample. Higher levels of interactions with mucin glycoproteins were also observed with the preactivated thiomers, which was due to the presence of hydroxyl groups on the preactivated thiomers. This was shown through rheological testing where higher levels of apparent viscosity in preactivated polymer and mucin mixtures were measured in comparison to both the unmodified, and thiolated (non-
38
ACCEPTED MANUSCRIPT preactivated) polymer blends. Overall, enhanced stability of preactivated PAA blends was observed in comparison to unmodified and thiolated PAA. A similar study for vaginal delivery using preactivated chitosan thiolated with thioglycolic acid was conducted (Friedl et al., 2013). Thiolated chitosan and preactivated chitosan samples were compared to commercially available products in terms of their mucoadhesion
PT
properties over a 24 hour period. An in-vitro test model of the vagina was created using
RI
bovine vaginal tissue and the mucoadhesive properties were tested using a wash-off test using
SC
an artificial vaginal fluid, pH 4.5. The thiolated and preactivated chitosan formulations showed increased mucoadhesion in comparison to the commercial available products, with
NU
the preactivated samples displaying a 1.5-fold increase in mucoadhesion after 24 h in comparison to the most adhesive commercial product. Tensile strength tests were also
MA
performed on the commercially available and thiolated products. Again, the preactivated chitosan formulation displayed the highest total work of adhesion, indicating high degrees of
D
mucoadhesion. The use of thiolated polymers for drug delivery to more acidic regions of the
PT E
body, gastric or vaginal, has also been impeded due to the reactivity of the thiol group in acidic pH, lessening mucoadhesion.
CE
S-protected pectin was examined and compared to unmodified pectin and thiolated pectin for
AC
its mucoadhesive properties (Hintzen et al., 2013). To create the S-protected pectin compound, a 2-mercaptonicotinic acid dimer was reacted with L-cysteine through a disulphide exchange. The product was purified through filtration and an ion exchange chromatography to remove any unbound L-cysteine or L-cystine and was reacted, in the presence of EDC, with pectin to form the S-protected pectin product. Thiolated pectin was also thiolated with EDC and L-cysteine. The thiolated and S-protected samples were tested for their disintegration rates in a pH 1 HCl media and a pH 6.8 buffer, and compared to that of unmodified pectin. The unmodified pectin sample quickly dissolved in the pH 6.8 buffer 39
ACCEPTED MANUSCRIPT whereas the thiolated sample, due to the formation of disulphide bonds, remained stable at this pH. In both pH solutions, the S-protected pectin displayed decreased rates of disintegration. Swelling studies of the three samples also showed the greater ability of the Sprotected pectin sample, with a higher capacity to swell and greater cohesive properties. The mucoadhesive properties, as measured by the rotating cylinder method, were also increased in
PT
the S-protected pectin sample in comparison to both the thiolated and unmodified sample,
RI
although it was suggested that this increase was due to the more cohesive nature of the S-
SC
protected, and indeed the thiolated pectin, due to slower rates of disintegration. Interestingly, the S-protected sample was tested for its long term stability after 6 months, comparing the
NU
results to that of 1 day. There was no formation of intra- or inter-molecular disulphide as measured by the viscosity levels, which would increase with their formation. A non-
MA
significant decrease was observed in 2-mercaptonicotinic acid dimer and disulphide bond content, but free thiol groups, as measured by Ellman’s reagent, were not observed.
D
Gastric drug delivery with preactivated thiomers was examined in greater detail using pectin
PT E
(Hauptstein et al., 2013). Pectin was initially thiolated with cysteine and was then activated with dimers of 2-mercaptonicotinic acid to create the preactivated thiomer product. Swelling,
CE
disintegration and mucoadhesive properties were all improved in comparison to the
AC
unmodified polymer. Mucoadhesion was tested in both acidic and basic conditions (pH 6.8 buffer), simulating gastric and intestinal adhesion. Mucoadhesion was vastly improved in the acidic conditions using the preactivated polymer over the thiolated counterpart, due to the increased stability of the preactivated thiomer at lower pH. In basic conditions, the preactivated samples increased in mass to a greater extent than in comparison to the acidic conditions, with an increase in mass of 16 – fold in acidic conditions, and a 22 – fold increase in basic buffer, thus highlighting the higher cohesive properties in the acidic buffer, but also displaying the ability of the preactivated pectin sample to swell, an important feature for drug 40
ACCEPTED MANUSCRIPT release in a mucoadhesive tablet. In drug release studies of the model drug, rosuvastatin calcium, the acidic conditions displayed a slower and more controlled release from the preactivated sample in comparison to release in pH 6.8 buffer, with 50% release occurring after 10 h in acidic conditions but after 4 h in the pH 6.8 buffer. The modification of the pectin backbone, by thiolated or by preactivation, showed no significant changes to the low
PT
cytotoxic properties of unmodified pectin, as tested by resazurin assay.
RI
Gelatin was also modified to form a preactivated gelatin product by initially thiolating a
SC
native gelatin sample (50 kDa) with Traut’s reagent and reacting it further with a mercaptonicotinic acid (MNA) dimer (Rohrer et al., 2015). Gelatin thiolated with Traut’s
NU
reagent yield a thiol content of 243 ± 19 µmol/g, and once it was reacted with MNA dimer, the thiol content decreased to 53 ± 6 µmol/g, indicating a successful attachment to the free
MA
thiols of the gelatin-Traut’s reagent complex. Both tensile strength tests and rotating cylinder tests were used to identify the mucoadhesive properties of the preactivated gelatin sample and
D
results were compared to an unmodified gelatin sample and the thiolated gelatin-Traut’s
PT E
reagent sample. The tensile strength tests were conducted on porcine intestinal tissue; the preactivated gelatin sample showed a marked increase of approximately 5 – 6-fold in the total
CE
work of adhesion and maximum detachment force when compared to the thiolated gelatin
AC
sample. This was observed at both pH 5 and pH 6.8, again showing the importance of the concept of polymer preactivation. The rotating cylinder method, also on porcine intestinal tissue, reiterated the tensile strength tests, and the adhesion times of the preactivated gelatin sample were far superior to that of the thiolated gelatin and unmodified gelatin samples. The rotating cylinder method was again tested at both pH 5 and pH 6.8 and interestingly, there was a greater improvement in mucoadhesion observed in the pH 5 buffer, with adhesion of up to 5 days measured, approximately 5-fold higher than adhesion in pH 6.8 buffer. The improved mucoadhesion at pH 5 was also observed in the thiolated gelatin sample too. It was 41
ACCEPTED MANUSCRIPT thought that this was due to the isoelectric point of both the preactivated gelatin and thiolated gelatin samples, measuring 5.7 and 4.7 respectively, both of which were closer to pH 5 than to pH 6.8. Thus, less erosion of the polymer was occurring which in turn increased the time of adhesion of the samples.
3.7 Permeation enhancing effects of thiolated polymers
PT
Apart from forming strong bonds between the polymer and the mucosal surface and
RI
prolonging residence times at the target site, it has also been observed that thiolation of a
SC
polymer has other benefits, including a permeation enhancing effect. Previously, to enhance the uptake of a drug in the intestine, a permeation enhancer agent would also be administered;
NU
however, these agents were generally of low MW and were often absorbed more quickly that the target drug itself, resulting in possible toxic side effects (A. Bernkop-Schnürch et al.,
MA
2003). It was found that polymeric compounds, such as chitosan and PAA displayed permeation enhancing effects and had the benefits of not being absorbed by the intestine due
D
to their higher MW. As both chitosan and PAA have inherent mucoadhesive properties, it
PT E
was postulated that thiolated versions of the polymers would display further enhancement. Clausen and Bernkop-Schnurch (2000) demonstrated the permeation enhancing effects of
CE
thiolated polymers on tight junctions in tissues. In this study, the permeation of thiolated polycarbophil with peptide based drugs on mucosal tissues was investigated and compared to
AC
that of unmodified polycarbophil. Initial studies using sodium fluorescein as the model drug, showed a higher level of transport using the thiolated polycarbophil in comparison to its unmodified counterpart, and it was also showed that the degree of thiolation too had an effect on transport, with higher levels of thiolation resulting in more transport of sodium fluorescein over a three hour period. The study also investigated the transport of both insulin and of bacitracin, which had both been labelled with fluorescein isothiocyanate, using unmodified and thiolated polycarbophil. Here, there was no significant improvement of transport of the 42
ACCEPTED MANUSCRIPT peptide drugs due to the thiolated polycarbophil. It was thought that the observed permeation enhancing effects were due to the bonding of the thiolated polycarbophil with extracellular Ca2+, therefore allowing the tight junctions to open. They also observed a decrease in the transepithelial electrical resistance (TEER), which is the measurement of tightness between mature epithelial cell junctions (Sigurdsson et al., 2013). They concluded this decrease in
PT
TEER as the loosening of the membrane’s tight junctions, which contributed to the passive
RI
diffusion of the peptide drugs across the membrane.
SC
Vigl et al. (2009) investigated the effect thiolated PAH had on efflux pump inhibitory properties. Initially, PAH was crosslinked by reacting the polymer with triethylamine
NU
followed by the addition of dimethylsuccinate. This crosslinked polymer of 70 kDa MW,
MA
along with two uncrosslinked samples of 15 kDa and 70 kDa MW, were thiolated using Nacetlycysteine in the presence of the carbodiimide coupling reagent, EDC. Thiol content was measured as 77.6 µmol/g in the 15 kDa sample, 83.1 µmol/g in the 70 kDa uncrosslinked
PT E
D
sample and 162.5 µmol/g in the crosslinked 70 kDa sample. Inhibition of P-glycoprotein, an efflux transport molecule found within the intestinal epithelium, was tested using thiolated PAH and unmodified PAH. Rhodamine-123 was used as a model substrate for P-
CE
glycoprotein. A portion of rat ileum tissue was mounted into an Ussing type diffusion
AC
chamber and was incubated at 37 °C in artificial intestinal media. This media was removed and was replaced with a solution of rhodamine – 123 with or without the addition of thiolated or unmodified PAH. Samples were then taken either side of the intestinal tissue to investigate the permeation effect of the polymers. Additionally, the paracellular mechanism was also examined using FD-4, which is only transported through a paracellular route. Rhodamine transport was positively increased by the thiolated polymers, to a greater extent than the unmodified sample and the transport of FD-4 was not influenced by either the unmodified or thiolated PAH, therefore, it was surmised that the increase in rhodamine permeation was due 43
ACCEPTED MANUSCRIPT to the inhibition of P-glycoprotein. Cell cytotoxicity was also conducted using Caco-2 cells on unmodified and thiolated PAH samples. Although cell cytotoxicity was low and both cell permeation and efflux pump inhibition were positively affected by the thiolated PAH polymers, it was acknowledged that a higher thiol content would have vastly improved the results, as seen by Clausen and Bernkop-Schnürch (2000).
PT
The permeation enhancing effects were also examined in preactivated PAA samples (Wang
RI
et al., 2012). PAA was preactivated with 2-mercaptonicotinic acid and the influence that both
SC
MW and preactivation levels had on enhancing permeation were investigated. PAA of MW 100 kDa, 250 kDa or 450 kDa were used which all had similar preactivation levels of over
NU
700 µmol/g (high levels), investigating MW influences. PAA 450 kDa was then also preactivated to 300 - 400 µmol/g (medium levels) and also to less than 200 µmol/g (low
MA
levels) to examine the effects of the degree of preactivation. Sodium fluorescein was used as the model drug for the analysis of permeation across rat intestinal tissue. When investigating
D
the influence of MW, it was shown that higher MW samples, therefore PAA 450 kDa,
PT E
displayed greater permeation effects in comparison to both the 100 kDa and 250 kDa samples. Reduced glutathione was added as a permeation mediator and again PAA 450 kDa
CE
displayed higher levels of permeation in comparison to both other samples, although
AC
permeation using PAA 250 kDa was marginally improved with the addition of reduced glutathione. When comparing the preactivated PAA 450 kDa sample to an unmodified PAA 450 kDa and a thiolated PAA 450 kDa, the preactivated sample displayed greater degrees of permeation. It was surmised that this was due to the increased mucoadhesive nature of the preactivated samples, as well as the potential interaction with tight junction proteins which could lead to conformation change in the protein, enhancing drug uptake. Interestingly, when comparing the results of the three preactivated 450 kDa PAA samples of varying degrees of preactivation, low, medium and high, the medium sample displayed the highest levels of 44
ACCEPTED MANUSCRIPT permeation. This was in contrast to previous studies investigating thiolated PAA samples, in which higher levels of thiolation improved the permeation effects (Kast and BernkopSchnurch, 2002). It was thought that perhaps this was due to the decrease in the hydrophilic nature of the polymer upon the addition of the 2-mercaptonicotinic acid, which decreased further with the highest level of preactivation. Having said that, the medium preactivated
PT
sample displayed the highest permeation levels and it was thought this was due to the
RI
appropriate levels of thiol groups within the 300 - 400 µmol/g preactivated sample.
SC
As discussed above, κ and ι carrageenans were thiolated with thiourea and the efflux pump inhibitory actions of both thiolated carrageenan were also investigated using Caco-2 cells,
NU
focusing on the multidrug resistance protein (MRP-2) receptor (Suchaoin et al., 2015). MRP2 can be found on the surface of cancerous cells and may have a negative influence on the
MA
uptake of anticancer drugs. Using the fluorescent marker of sulforhodamine 101, it was shown that thiolated carrageenan increased the amount of intracellular sulforhodamine 101
D
once the cells were pretreated with the thiolated samples when compared to their unmodified
PT E
counterparts. Additionally to this, the thiolated carrageenans were also shown to improve the uptake of sulforhodamine 101 across a Caco-2 monolayer, thus indicating the blocking effect
CE
of MRP-2. It was thought this inhibitory effect was due to the binding of the thiolated
AC
carrageenan to the thiol moieties found on MRP-2.
4 Conclusion
The thiolation of a number of different natural and synthetic polymers has led to increased mucoadhesive properties, slower and more controlled release of drug, and enhanced permeation effects. Further developments in the thiolation of polymers, including Sprotecting and preactivation of the thiol groups, have opened up a new generation of thiolated polymers which allows for mucoadhesion to occur over a wider range of pH levels. This
45
ACCEPTED MANUSCRIPT could equate to a mucoadhesive delivery device which has more scope for the administration, improved adhesion and subsequent controlled drug delivery throughout a larger area of the body, including the full length of the gastrointestinal tract and vagina; further work may be required in order to create and develop a commercially viable drug delivery device for use in these less accessible areas, which resist becoming over hydrated, thus losing its adhesive
PT
properties while retaining its controlled drug release profile. The further gelling properties of
RI
thiolated polymers could prolong the residence time and inhibit the premature removal of
SC
dosage forms, semi-solid for examples, in highly turbulent areas, such as the vagina or the eye. Targeting the eye, for example to treat posterior segment diseases, using a mucoadhesive
NU
dosage form could have many benefits, including fewer systemic side effects, however, poor patient compliance due to the insert of a drug releasing tablet for prolonged periods into the
MA
eye could affect its popularity. Having said that, the release of gel-like dosage forms, such as Lacrimera ®, to the market for treatment to the dry eye syndrome and the founding of
D
Thiomatrix®, a drug delivery company specialising in thiomer technology, is testament to the
AC
CE
PT E
future of mucoadhesive polymeric devices.
46
ACCEPTED MANUSCRIPT 5 References Anderson, O. A., Bainbridge, J. W. B. and Shima, D. T. (2010) 'Delivery of anti-angiogenic molecular therapies for retinal disease', Drug Discovery Today, 15(7-8), pp. 272-282. Andrews, G. P., Laverty, T. P. and Jones, D. S. (2009) 'Mucoadhesive polymeric platforms for controlled drug delivery', European Journal of Pharmaceutics and Biopharmaceutics, 71(3), pp. 505-518.
RI
PT
Asane, G. S., Nirmal, S. A., Rasal, K. B., Naik, A. A. and Mahadik, M. S. (2008) 'Polymers for Mucoadhesive Drug Delivery Systems: A Current Status', Drug Development and Industrial Pharmacy, 34, pp. 1246-1266.
NU
SC
Bacalocostantis, I., Mane, V. P., Kang, M. S., Goodley, A. S., Muro, S. and Kofinas, P. (2012) 'Effect of Thiol Pendant Conjugates on Plasmid DNA Binding, Release, and Stability of Polymeric Delivery Vectors', Biomacromolecules, 13(5), pp. 1331-1339.
MA
Bansil, R. and Turner, B. S. (2006) 'Mucin structure, aggregation, physiological functions and biomedical applications', Current Opinion in Colloid & Interface Science, 11(2-3), pp. 164-170.
D
Bernkop-Schnürch, A. (2005) 'Thiomers: A new generation of mucoadhesive polymers', Advanced Drug Delivery Reviews, 57(11), pp. 1569-1582.
PT E
Bernkop-Schnürch, A., Guggi, D. and Pinter, Y. (2004a) 'Thiolated chitosans: development and in vitro evaluation of a mucoadhesive, permeation enhancing oral drug delivery system', Journal of Controlled Release, 94(1), pp. 177-186.
AC
CE
Bernkop-Schnürch, A., Hornof, M. and Guggi, D. (2004b) 'Thiolated chitosans', European Journal of Pharmaceutics and Biopharmaceutics, 57(1), pp. 9-17. Bernkop-Schnürch, A., Hornof, M. and Zoidl, T. (2003) 'Thiolated polymers—thiomers: synthesis and in vitro evaluation of chitosan–2-iminothiolane conjugates', International Journal of Pharmaceutics, 260(2), pp. 229-237. Bernkop-Schnürch, A., Kast, C. E. and Guggi, D. (2003) 'Permeation enhancing polymers in oral delivery of hydrophilic macromolecules: thiomer/GSH systems', Journal of Controlled Release, 93(2), pp. 95-103. Bernkop-Schnürch, A., Schwarz, V. and Steininger, S. (1999) 'Polymers with thiol groups: a new generation of mucoadhesive polymers?', Pharmaceutical Research, 16(6), pp. 876-881.
47
ACCEPTED MANUSCRIPT Bernkop-Schnürch, A. and Steininger, S. (2000) 'Synthesis and characterisation of mucoadhesive thiolated polymers', International Journal of Pharmaceutics, 194(2), pp. 239247. Bonengel, S., Haupstein, S., Perera, G. and Bernkop-Schnürch, A. (2014) 'Thiolated and Sprotected hydrophobically modified cross-linked poly(acrylic acid) – A new generation of multifunctional polymers', European Journal of Pharmaceutics and Biopharmaceutics, 88(2), pp. 390-396.
RI
PT
Boussif, O., Delair, T., Brua, C., Veron, L., Pavirani, A. and Kolbe, H. V. J. (1999) 'Synthesis of Polyallylamine Derivatives and Their Use as Gene Transfer Vectors in Vitro', Bioconjugate Chemistry, 10(5), pp. 877-883.
NU
SC
Brandt, J., Svenson, A., Carlsson, J. and Drevin, H. (1977) 'Covalent coupling of unsaturated compounds to thiol agrarose using γ-Radiation', Journal of Solid-Phase Biochemistry, 2(2), pp. 105-109.
MA
Clausen, A. E. and Bernkop-Schnürch, A. (2000) 'In vitro evaluation of the permeationenhancing effect of thiolated polycarbophil', Journal of Pharmaceutical Sciences, 89(10), pp. 1253-1261.
D
Cone, R. A. (2009) 'Barrier properties of mucus', Advanced Drug Delivery Reviews, 61(2), pp. 75-85.
PT E
Dinarvand, R., Mahmoodi, S., Farboud, E., Salehi, M. and Atyabi, F. (2005) 'Preparation of gelatin microspheres containing lactic acid – Effect of cross-linking on drug release', Acta Pharmaceutica, 55, pp. 57-67.
AC
CE
Dodane, V., Amin Khan, M. and Merwin, J. R. (1999) 'Effect of chitosan on epithelial permeability and structure', International Journal of Pharmaceutics, 182(1), pp. 21-32. Duggan, S., Hughes, H., Owens, E., Duggan, E., Cummins, W. and O’ Donovan, O. (2016a) 'Synthesis and characterisation of mucoadhesive thiolated polyallylamine', International Journal of Pharmaceutics, 499(1–2), pp. 368-375. Duggan, S., O’Donovan, O., Owens, E., Cummins, W. and Hughes, H. (2015) 'Synthesis of mucoadhesive thiolated gelatin using a two-step reaction process', European Journal of Pharmaceutics and Biopharmaceutics, 91, pp. 75-81. Duggan, S., O’Donovan, O., Owens, E., Duggan, E., Hughes, H. and Cummins, W. (2016b) 'Comparison of the mucoadhesive properties of thiolated polyacrylic acid to thiolated polyallylamine', International Journal of Pharmaceutics, 498(1–2), pp. 245-253.
48
ACCEPTED MANUSCRIPT
Dünnhaupt, S., Barthelmes, J., Thurner, C. C., Waldner, C., Sakloetsakun, D. and BernkopSchnürch, A. (2012) 'S-protected thiolated chitosan: Synthesis and in vitro characterization', Carbohydrate Polymers, 90(2), pp. 765-772. Ensign, L. M., Cone, R. and Hanes, J. (2012) 'Oral drug delivery with polymeric nanoparticles: The gastrointestinal mucus barriers', Advanced Drug Delivery Reviews, 64(6), pp. 557-570.
RI
PT
Friedl, H. E., Dünnhaupt, S., Waldner, C. and Bernkop-Schnürch, A. (2013) 'Preactivated thiomers for vaginal drug delivery vehicles', Biomaterials, 34(32), pp. 7811-7818.
SC
Grabovac, V. and Bernkop-Schnürch, A. (2007) 'Development and In Vitro Evaluation of Surface Modified Poly(lactide-co-glycolide) Nanoparticles with Chitosan-4Thiobutylamidine', Drug Development and Industrial Pharmacy, 33(7), pp. 767-774.
MA
NU
Grabovac, V., Föger, F. and Bernkop-Schnürch, A. (2008) 'Design and in vivo evaluation of a patch delivery system for insulin based on thiolated polymers', International Journal of Pharmaceutics, 348(1-2), pp. 169-174.
D
Grabovac, V., Guggi, D. and Bernkop-Schnürch, A. (2005) 'Comparison of the mucoadhesive properties of various polymers', Advanced Drug Delivery Reviews, 57(11), pp. 1713-1723.
PT E
Guggi, D., Marschütz, M. K. and Bernkop-Schnürch, A. (2004) 'Matrix tablets based on thiolated poly(acrylic acid): pH-dependent variation in disintegration and mucoadhesion', International Journal of Pharmaceutics, 274(1-2), pp. 97-105.
AC
CE
Guo, J., O’Mahony, A. M., Cheng, W. P. and O’Driscoll, C. M. (2013) 'Amphiphilic polyallylamine based polymeric micelles for siRNA delivery to the gastrointestinal tract: In vitro investigations', International Journal of Pharmaceutics, 447(1–2), pp. 150-157. H.P. Rang, M.M. Dale, J.M. Ritter and Moore, P. K. (2003) Pharmacology. 5th ed., Elsevier. Hauptstein, S., Bonengel, S., Rohrer, J. and Bernkop-Schnürch, A. (2014) 'Preactivated thiolated poly(methacrylic acid-co-ethyl acrylate): Synthesis and evaluation of mucoadhesive potential', European Journal of Pharmaceutical Sciences, 63(0), pp. 132-139. Hauptstein, S., Dezorzi, S., Prüfert, F., Matuszczak, B. and Bernkop-Schnürch, A. (2015) 'Synthesis and in vitro characterization of a novel S-protected thiolated alginate', Carbohydrate Polymers, 124(0), pp. 1-7.
49
ACCEPTED MANUSCRIPT Hauptstein, S., Muller, C., Dunnhaupt, S., Laffleur, F. and Bernkop-Schnurch, A. (2013) 'Preactivated thiomers: evaluation of gastroretentive minitablets', Int J Pharm, 456(2), pp. 473-479. Hintzen, F., Hauptstein, S., Perera, G. and Bernkop-Schnürch, A. (2013) 'Synthesis and in vitro characterization of entirely S-protected thiolated pectin for drug delivery', European Journal of Pharmaceutics and Biopharmaceutics, 85(3, Part B), pp. 1266-1273.
RI
PT
Hornof, M., Weyenberg, W., Ludwig, A. and Bernkop-Schnürch, A. (2003) 'Mucoadhesive ocular insert based on thiolated poly(acrylic acid): development and in vivo evaluation in humans', Journal of Controlled Release, 89(3), pp. 419-428.
NU
SC
Horvát, G., Gyarmati, B., Berkó, S., Szabó-Révész, P., Szilágyi, B. Á., Szilágyi, A., Soós, J., Sandri, G., Bonferoni, M. C., Rossi, S., Ferrari, F., Caramella, C., Csányi, E. and BudaiSzűcs, M. (2015) 'Thiolated poly(aspartic acid) as potential in situ gelling, ocular mucoadhesive drug delivery system', European Journal of Pharmaceutical Sciences, 67, pp. 1-11.
MA
Huang, Y., Leobandung, W., Foss, A. and Peppas, N. A. (2000) 'Molecular aspects of mucoand bioadhesion:: Tethered structures and site-specific surfaces', Journal of Controlled Release, 65(1-2), pp. 63-71.
PT E
D
Hudson, S. P., Owens, E., Hughes, H. and McLoughlin, P. (2012) 'Enhancement and restriction of chain motion in polymer networks', International Journal of Pharmaceutics, 430(1–2), pp. 34-41.
CE
Ibie, C. O., Thompson, C. J. and Knott, R. (2015) 'Synthesis, characterisation and in vitro evaluation of novel thiolated derivatives of polyallylamine and quaternised polyallylamine', Colloid and Polymer Science, pp. 1-12.
AC
Iqbal, J., Sakloetsakun, D. and Bernkop-Schnürch, A. (2011) 'Thiomers: Inhibition of cytochrome P450 activity', European Journal of Pharmaceutics and Biopharmaceutics, 78(3), pp. 361-365. Iqbal, J., Shahnaz, G., Dünnhaupt, S., Müller, C., Hintzen, F. and Bernkop-Schnürch, A. (2012a) 'Preactivated thiomers as mucoadhesive polymers for drug delivery', Biomaterials, 33(5), pp. 1528-1535. Iqbal, J., Shahnaz, G., Perera, G., Hintzen, F., Sarti, F. and Bernkop-Schnürch, A. (2012b) 'Thiolated chitosan: Development and in vivo evaluation of an oral delivery system for leuprolide', European Journal of Pharmaceutics and Biopharmaceutics, 80(1), pp. 95-102.
50
ACCEPTED MANUSCRIPT Ivarsson, D. and Wahlgren, M. (2012) 'Comparison of in vitro methods of measuring mucoadhesion: Ellipsometry, tensile strength and rheological measurements', Colloids and Surfaces B: Biointerfaces, 92(0), pp. 353-359. Joergensen, L., Klösgen, B., Simonsen, A. C., Borch, J. and Hagesaether, E. (2011) 'New insights into the mucoadhesion of pectins by AFM roughness parameters in combination with SPR', International Journal of Pharmaceutics, 411(1–2), pp. 162-168.
RI
PT
Kafedjiiski, K., Jetti, R. K. R., Föger, F., Hoyer, H., Werle, M., Hoffer, M. and BernkopSchnürch, A. (2007) 'Synthesis and in vitro evaluation of thiolated hyaluronic acid for mucoadhesive drug delivery', International Journal of Pharmaceutics, 343(1-2), pp. 48-58.
SC
Kast, C. and Bernkop-Schnurch, A. (2002) 'Influence of the molecular mass on the permeation enhancing effect of different poly(acrylates)', Techniques Pharmaceutiques, 12(6), pp. 351-356.
MA
NU
Kast, C. E. and Bernkop-Schnürch, A. (2001) 'Thiolated polymers — thiomers: development and in vitro evaluation of chitosan–thioglycolic acid conjugates', Biomaterials, 22(17), pp. 2345-2352.
D
Kawashima, Y., Yamamoto, H., Takeuchi, H. and Kuno, Y. (2000) 'Mucoadhesive DLLactide/Glycolide Copolymer Nanospheres Coated with Chitosan to Improve Oral Delivery of Elcatonin', Pharmaceutical Development and Technology, 5(1), pp. 77-85.
PT E
Khanenko, E. A., Larionova, N. I. and Demina, N. B. (2009) 'Mucoadhesive Drug Delivery Systems (Review)', Pharmaceutical Chemistry Journal, 43(4), pp. 21 - 29.
CE
Khutoryanskiy, V. V. (2011) 'Advances in mucoadhesion and mucoadhesive polymers', Macromolecular Bioscience, 11, pp. 748-764.
AC
Kommareddy, S. and Amiji, M. (2005) 'Preparation and evaluation of thiol-modified gelatin nanoparticles for intracellular DNA delivery in response to glutathione', Bioconjugate Chemistry, 16, pp. 1423-1432. Krauland, A. H. and Bernkop-Schnürch, A. (2004) 'Thiomers: development and in vitro evaluation of a peroral microparticulate peptide delivery system', European Journal of Pharmaceutics and Biopharmaceutics, 57(2), pp. 181-187. Kumari, A., Yadav, S. K. and Yadav, S. C. (2010) 'Biodegradable polymeric nanoparticles based drug delivery systems', Colloids and Surfaces B: Biointerfaces, 75(1), pp. 1-18.
51
ACCEPTED MANUSCRIPT Leitner, V. M., Marschütz, M. K. and Bernkop-Schnürch, A. (2003) 'Mucoadhesive and cohesive properties of poly(acrylic acid)-cysteine conjugates with regard to their molecular mass', European Journal of Pharmaceutical Sciences, 18(1), pp. 89-96. Leitner, V. M., Walker, G. F. and Bernkop-Schnürch, A. (2003) 'Thiolated polymers: evidence for the formation of disulphide bonds with mucus glycoproteins', European Journal of Pharmaceutics and Biopharmaceutics, 56(2), pp. 207-214.
PT
Liu, L., Fishman, M. L., Hicks, K. B. and Kende, M. (2005) 'Interaction of various pectin formulations with porcine colonic tissues', Biomaterials, 26(29), pp. 5907-5916.
SC
RI
Ludwig, A. (2005) 'The use of mucoadhesive polymers in ocular drug delivery', Advanced Drug Delivery Reviews, 57(11), pp. 1595-1639.
MA
NU
Luessen, H. L., Verhoef, J. C., Borchard, G., Lehr, C.-M., Boer, A. G. d. and Junginger, H. E. (1995) 'Mucoadhesive polymers in peroral peptide drug delivery. II. Carbomer and polycarbophil are potent inhitiors of the intestinal proteolytic enzyme trypsin', Pharmaceutical Research, 12, pp. 1293-1298. Madhav, N. V. S., Shakya, A. K., Shakya, P. and Singh, K. (2009) 'Orotransmucosal drug delivery systems: A review', Journal of Controlled Release, 140(1), pp. 2-11.
PT E
D
Makhlof, A., Werle, H. and Takeuchi, H. (2008) 'Mucoadhesive drug carriers and polymers for effective drug delivery', Journal of Drug Delivery Science and Technology, 18, pp. 375386.
CE
Mortazavi, S. A. (1995) 'An in vitro assessment of mucus/mucoadhesive interactions', International Journal of Pharmaceutics, 124(2), pp. 173-182.
AC
Nagarwal, R. C., Kant, S., Singh, P. N., Maiti, P. and Pandit, J. K. (2009) 'Polymeric nanoparticulate system: A potential approach for ocular drug delivery', Journal of Controlled Release, 136(1), pp. 2-13. Oskuee, R. K., Dosti, F., Gholami, L. and Malaekeh-Nikouei, B. (2015) 'A simple approach for producing highly efficient DNA carriers with reduced toxicity based on modified polyallylamine', Materials Science and Engineering: C, 49(0), pp. 290-296. Palmberger, T. F., Albrecht, K., Loretz, B. and Bernkop-Schnürch, A. (2007) 'Thiolated polymers: Evaluation of the influence of the amount of covalently attached l-cysteine to poly(acrylic acid)', European Journal of Pharmaceutics and Biopharmaceutics, 66(3), pp. 405-412.
52
ACCEPTED MANUSCRIPT Panyam, J. and Labhasetwar, V. (2003) 'Biodegradable nanoparticles for drug and gene delivery to cells and tissue', Advanced Drug Delivery Reviews, 55(3), pp. 329-347. Patel, V. F., Liu, F. and Brown, M. B. (2011) 'Advances in oral transmucosal drug delivery', Journal of Controlled Release, 153(2), pp. 106-116. Pratt, C. W. and Cornely, K. (2011) Essential Biochemistry. Second ed., Wiley.
RI
PT
Rohrer, J., Zupančič, O., Suchaoin, W., Netsomboon, K., Laffleur, F., Oh, S. and BernkopSchnürch, A. (2015) 'Synthesis and in vitro characterisation of preactivated thiolated gelatin', European Polymer Journal, 73, pp. 268-277.
SC
Roy, S., Pal, K., Anis, A., Pramanik, K. and B.Prabhakar (2009) 'Polymers in Mucoadhesive Drug Delivery System: A Brief Note', Designed Monomers and Polymers, 12, pp. 483-495.
MA
NU
Salamanca, A. E. d., Diebold, Y., Calonge, M., Garcıa-Vazquez, C., Callejo, S., Vila, A. and Alonso, M. J. (2006) 'Chitosan Nanoparticles as a Potential Drug Delivery System for the Ocular Surface: Toxicity, Uptake Mechanism and In Vivo Tolerance', Investigative Ophthalmology and Visual Science, 47(4), pp. 1416-1425.
PT E
D
Schmitz, T., Grabovac, V., Palmberger, T. F., Hoffer, M. H. and Bernkop-Schnürch, A. (2008) 'Synthesis and characterization of a chitosan-N-acetyl cysteine conjugate', International Journal of Pharmaceutics, 347(1–2), pp. 79-85. Sharma, R. and Ahuja, M. (2011) 'Thiolated pectin: Synthesis, characterization and evaluation as a mucoadhesive polymer', Carbohydrate Polymers, 85(3), pp. 658-663.
AC
CE
Sigurdsson, H. H., Kirch, J. and Lehr, C.-M. (2013) 'Mucus as a barrier to lipophilic drugs', International Journal of Pharmaceutics, 453(1), pp. 56-64. Smart, J. D. (2005) 'The basics and underlying mechanisms of mucoadhesion', Advanced Drug Delivery Reviews, 57(11), pp. 1556-1568. Srividya, B., Cardoza, R. M. and Amin, P. D. (2001) 'Sustained ophthalmic delivery of ofloxacin from a pH triggered in situ gelling system', Journal of Controlled Release, 73(2-3), pp. 205-211. Suchaoin, W., Bonengel, S., Grießinger, J. A., Pereira de Sousa, I., Hussain, S., Huck, C. W. and Bernkop-Schnürch, A. (2016a) 'Novel bioadhesive polymers as intra-articular agents: Chondroitin sulfate-cysteine conjugates', European Journal of Pharmaceutics and Biopharmaceutics, 101, pp. 25-32.
53
ACCEPTED MANUSCRIPT
Suchaoin, W., Bonengel, S., Hussain, S., Huck, C. W., Ma, B. N. and Bernkop-Schnürch, A. (2015) 'Synthesis and In Vitro Evaluation of Thiolated Carrageenan', Journal of Pharmaceutical Sciences, 104(8), pp. 2523-2530.
PT
Suchaoin, W., Pereira de Sousa, I., Netsomboon, K., Rohrer, J., Hoffmann Abad, P., Laffleur, F., Matuszczak, B. and Bernkop-Schnürch, A. (2016b) 'Mucoadhesive polymers: Synthesis and in vitro characterization of thiolated poly(vinyl alcohol)', International Journal of Pharmaceutics, 503(1–2), pp. 141-149.
RI
Tabata, Y. and Ikada, Y. (1998) 'Protein release from gelatin matrices', Advanced Drug Delivery Reviews, 31(3), pp. 287-301.
NU
SC
Thirawong, N., Nunthanid, J., Puttipipatkhachorn, S. and Sriamornsak, P. (2007) 'Mucoadhesive properties of various pectins on gastrointestinal mucosa: An in vitro evaluation using texture analyzer', European Journal of Pharmaceutics and Biopharmaceutics, 67(1), pp. 132-140.
MA
Ugwoke, M. I., Agu, R. U., Verbeke, N. and Kinget, R. (2005) 'Nasal mucoadhesive drug delivery: Background, applications, trends and future perspectives', Advanced Drug Delivery Reviews, 57(11), pp. 1640-1665.
PT E
D
Vigl, C., Leithner, K., Albrecht, K. and Bernkop-Schnurch, A. (2009) 'The efflux pump inhibitory properties of (thiolated) polyallylamines', Journal of Drug Delivery Science and Technology, 19(6), pp. 405-411.
CE
Vlierberghe, S. V., Schacht, E. and Dubruel, P. (2011) 'Reversible gelatin-based hydrogels: Finetuning of material properties', European Polymer Journal, 47(5), pp. 1039-1047.
AC
Wang, J., Tauchi, Y., Deguchi, Y., Morimoto, K., Tabata, Y. and Ikada, Y. (2000) 'Positively Charged Gelatin Microspheres as Gastric Mucoadhesive Drug Delivery System for Eradication of H. pylori', Drug Delivery, (7), pp. 237-243. Wang, X., Iqbal, J., Rahmat, D. and Bernkop-Schnürch, A. (2012) 'Preactivated thiomers: Permeation enhancing properties', International Journal of Pharmaceutics, 438(1–2), pp. 217-224. Yasukawa, T., Ogura, Y., Tabata, Y., Kimura, H., Wiedemann, P. and Honda, Y. (2004) 'Drug delivery systems for vitreoretinal diseases', Progress in Retinal and Eye Research, 23(3), pp. 253-281.
54
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
PT
Young, S., Wong, M., Tabata, Y. and Mikos, A. G. (2005) 'Gelatin as a delivery vehicle for the controlled release of bioactive molecules', Journal of Controlled Release, 109(1-3), pp. 256-274.
55
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
Figure 1 Structures of mucoadhesive polymers (A) chitosan, (B) polyacrylic acid (PAA), (C) polylactic-coglycolide (PLGA), (D) pectin, (E) gelatin (Elzoghby, 2013) and (F) polyallylamine in its (i) free base and (ii) salt forms (Vigl et al., 2009).
56
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
Figure 2 Mucoadhesive testing on thiolated PAA (black) and control PAA (white) samples, with and without the addition of 1% (m/m) cysteine (Verena M. Leitner et al., 2003)
57
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
Figure 3 (A) Structure of L-cysteine and (B) the proposed structure of PAA-cysteine conjugate (BernkopSchnürch and Steininger, 2000)
58
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
Figure 4 Structure of (A) Traut’s reagent (2-iminothiolane) and (B) the proposed structure of PAAm thiolated with Traut’s reagent
59
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
Figure 5 Mucoadhesive testing of a thiolated gelatin control (a native gelatin sample which had been directly thiolated with Traut’s reagent) and a thiolated gelatin sample following the two-step reaction (Duggan et al., 2015)
60
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
Figure 6 Structure of preactivated PAA (Iqbal et al., 2012a)
61
AC
CE
Graphical abstract
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
62