Immobilization of dextransucrase on functionalized TiO2 supports

Accepted Manuscript Immobilization of dextransucrase on functionalized TiO2 supports

Miona Miljković, Vesna Lazić, Katarina Banjanac, Slađana Davidović, Dejan Bezbradica, Aleksandar Marinković, Dušan Sredojević, Jovan M. Nedeljković, Suzana Dimitrijević Branković PII: DOI: Reference:

S0141-8130(18)30295-2 doi:10.1016/j.ijbiomac.2018.04.027 BIOMAC 9435

To appear in: Received date: Revised date: Accepted date:

17 January 2018 30 March 2018 5 April 2018

Please cite this article as: Miona Miljković, Vesna Lazić, Katarina Banjanac, Slađana Davidović, Dejan Bezbradica, Aleksandar Marinković, Dušan Sredojević, Jovan M. Nedeljković, Suzana Dimitrijević Branković , Immobilization of dextransucrase on functionalized TiO2 supports. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Biomac(2017), doi:10.1016/ j.ijbiomac.2018.04.027

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 Immobilization of dextransucrase on functionalized TiO2 supports

Miona Miljković,a Vesna Lazić,b* Katarina Banjanac,c Slađana Davidović,a Dejan Bezbradica,a Aleksandar Marinković,d Dušan Sredojević,b Jovan M. Nedeljković,b Suzana Dimitrijević

Department of Biochemical Engineering and Biotechnology, Faculty of Technology and

CR

a

IP

T

Brankovića

Metallurgy, University of Belgrade, Karnegijeva 4, 11120 Belgrade, Serbia Vinča Institute of Nuclear Sciences, University of Belgrade, P.O. Box 522, 11001 Belgrade,

US

b

Center of Innovation, Faculty of Technology and Metallurgy, University of Belgrade,

Karnegijeva 4, 11120 Belgrade, Serbia

Department of Organic Chemistry, Faculty of Technology and Metallurgy, University of

ED

d

M

c

AN

Serbia

Correspondence to: Vesna Lazić, Vinča Institute of Nuclear Sciences, University of Belgrade,

CE

*

PT

Belgrade, Karnegijeva 4, 11120 Belgrade, Serbia

P.O. Box 522, 11001 Belgrade, Serbia.

AC

Phone: +38163 8260878; E-mail address: [email protected]

Abstract

The TiO2 based hybrid supports with different functional groups (amino, glutaraldehyde or epoxy) were prepared and their influence on immobilization of dextransucrase (DS) was studied. Novel synthetic route for surface modification of TiO2 with amino and glutaraldehyde groups 1

ACCEPTED MANUSCRIPT was developed taking advantage of charge transfer complex (CTC) formation between surface Ti atoms and salicylate-type of ligands (5-aminosalicylic acid (5-ASA)). The proposed coordination of 5-ASA to the surface of TiO2 powder and optical properties of CTC was presented. The pristine TiO2 and amino functionalized TiO2 have higher sorption capacity for DS (12.6 and 12.0

T

mg/g, respectively) compared to glutaraldehyde and epoxy activated supports (9.6 and 9.8 mg/g,

IP

respectively). However, immobilized enzyme to either glutaraldehyde or epoxy functionalized

CR

TiO2 have almost two times higher expressed activities compared to pristine TiO2 support (258, 235 and 142 IU g-1, respectively). Thermal stability of enzyme immobilized on glutaraldehyde

US

and epoxy functionalized supports was studied at 40 °C, as well as operational stability under

AN

long-run working conditions in repeated cycles. After five cycles, DS imobilized on glutaraldehyde activated support retained almost 70% of its initial expresssed activity, while,

M

after five cycles, performance of DS immobilized on epoxy activated support was significantly

ED

lower (15%).

AC

CE

PT

Keyword: Surface functionalization of TiO2; Dextransucrase; Enzyme immobilization.

2

ACCEPTED MANUSCRIPT 1. Introduction

Dextransucrase (EC 2.4.1.5) (DS), obtained from Leuconostoc mesenteroides T3, is valuable biocatalyst, in particular, for the production of the soluble dextran chains by

T

polymerization of glucose molecules, as well as variety of oligosaccharides [1]. However, the

IP

most important application of DS is its catalytic action in synthesis of dextran, essential

CR

component in a blood volume expander. Dextran is a polysaccharide whose molecular structure consists of D-glucans with contiguous α-1,6 glycosidic linkages in the main chains and α-1,2, α-

US

1,3 and α-1,4 branch glycosidic linkages [2, 3]. In addition, beside medical, dextran also has

AN

industrial application in food, pharmaceutical and chemical industries as an adjuvant, emulsifier, carrier and stabilizer [4, 5]. However, the industrial application of DS prepared by “green” bio-

M

approach is limited due to small yield of enzyme production and its low catalytic activity. Thus,

ED

there is great interest for development of new technologies that provide improved performance of biocatalyst. The immobilization of the enzymes onto solid supports may ensure their extensive

PT

usage on industrial level. In general, during immobilization of enzymes, the stabilization of

CE

secondary structure followed with improved activity and/or stability of enzymes can occur. Thus, if properly designed, the immobilization becomes an effective way to produce biocatalyst with

AC

desired properties such as high activity, stability, selectivity, reduced inhibition by medium or product, even purity [6, 7]. Also, it should be mentioned that orientation of immobilized enzyme molecules on support has direct impact on their activity [8]. In order to improve enzyme activity, enzyme should be orientated on such way that the substrate can easily access the active sites. On the contrary, the interaction between enzyme and support can cause distortion of enzyme molecule and thus decrease its activity and/or stability.

3

ACCEPTED MANUSCRIPT The most common methods proposed to immobilize DS are based on adsorption, covalent coupling and/or encapsulation on different support materials, but, most of them, results in low specific activities, as well as low activity upon recovery [1, 2, 9-13]. Extensive research in the field of nanoscience led to development of synthetic procedures that provide nanomaterials with

T

desired properties. Immobilization of enzymes is novel prospective field of usage of

IP

nanomaterials. Recent studies suggest that immobilization of enzymes onto nanomaterials

CR

improves their operational, thermal and pH stability [14]. Size, specific surface area and pore size are essential characteristics of nanoparticles for the immobilization process. Nonporous

US

nanomaterials provide reduced diffusional limitation for the substrate and high enzyme loadings

AN

per mass of support due to large specific surface area. For successful application of porous nanomaterials in enzyme immobilization, two main criteria are imperative in achieving high

M

enzyme activity and stability: size matching between pore size and molecular diameter of

ED

enzyme, and undisturbed diffusion of substrate/product into/from pores of nanomaterial [15]. The immobilization of enzymes into pores of support prevent formation of intermolecular interactions

PT

(such as aggregation and proteolysis) and shield enzyme from interaction with external interfaces

CE

(air, oxygen, organic solvent) which can cause unwanted conformational changes in enzyme structure and hence deactivation [16].

AC

Magnetic and silica nanoparticles have been the most frequently used nanoparticles for immobilization of enzymes [17-21]. Also, in smaller extent, Au, SnO2, ZrO2 and chitosan nanoparticles have been used as support for enzymes’ immobilization [22]. Compared to other materials, TiO2 has antimicrobial ability, good mechanical strength, and high corrosion resistance [23, 24]. Because of that, nano-sized TiO2 has been considered as suitable support for enzyme immobilization in several recent reports [25-28]. However, only Yang et al. [26] took advantage of charge transfer complex (CTC) formation between surface Ti atoms and catecholate-type of 4

ACCEPTED MANUSCRIPT ligand (dopamine) for functionalization of TiO2 in order to improve sorption capacity of hybrid support for immobilization of enzymes. It should be noted that, to the best of our knowledge, immobilization of DS on either pristine TiO2 or functionalized TiO2 has not been reported in literature.

T

The main focus of this study is functionalization of commercial TiO2 nano-powder

IP

(Degussa P25) in order to obtain high-capacity support for immobilization of DS. Two different

CR

synthetic routes were used for activation of TiO2 support with different functional groups. The first, based on well-known silylation reaction, led to epoxy-functionalized TiO2 powders. Epoxy

US

activation of support surface is one of the most used modification methods due to high reactivity

AN

and stability of epoxy groups in broad pH range providing possibility for formation of covalent bond with the enzyme molecules. Mechanism of enzyme immobilization on epoxy-functionalized

M

supports occurs in three steps [7, 29]. The first step consists of the enzyme adsorption onto

ED

support surface. In the second step formation of covalent bond between reactive groups (amino, hydroxyl, thiol and phenolic) of the adsorbed enzyme and neighboring epoxy groups takes place,

PT

while blocking of remaining epoxy groups on the support occurs in the third step. The second

CE

route is based on surface modification of TiO2 with 5-aminosalicylic acid (5-ASA), i.e. the CTC formation between surface Ti atoms and salycilate-type of ligand [30-34]. This methodology

AC

leads to amino-functionalized TiO2 powder. So far, the studies concerning CTC formation, accompanied with the red-shift of optical absorption, have been primarily motivated by potential improvement of efficiency of photo-driven processes [30-34]. Beside epoxy activation, the treatment of support or enzyme with glutaraldehyde is one of the most used techniques to obtain immobilized enzyme via covalent attachment [35]. There are at least 3 ways for enzyme immobilization via glutaraldehyde [35, 36]. First one is treatment of enzyme with gluataraldehyde, which is previously immobilized via ion exchange on the amino modified 5

ACCEPTED MANUSCRIPT support. The two other utilize glutaraldehyde activated amino modified support: (i) with glutaraldehyde monomer, or (ii) glutaraldehyde dimer per primary amino group on support [37]. The monomer and dimer glutaraldehyde activated support have very different reactivity toward proteins. However, both monomer and dimer are able to rapidly react with the nucleophiles of the

T

protein molecule after the protein adsorption via electrostatic interaction. In this study, amino-

IP

modified TiO2 nanoparticles served as a precursor for preparation of glutaraldehyde-activated

CR

support.

Performance of immobilized enzyme on TiO2 supports activated with different functional

US

groups was thoroughly examined. The parameters including sorption capacity, immobilization

AN

yield, expressed activity, specific activity and activity immobilization yield, as well as kinetic data were used to compare abilities of different hybrid supports to immobilize DS. Finally,

M

special attention was paid to operational stability of immobilized enzyme, and, its performance

CE

PT

2. Material and methods

ED

was studied in repeated cycles.

AC

2.1. Functionalization and characterization of TiO2 powders

All chemicals (Degussa P-25 TiO2 powder, 5-amino salicylic acid (5-ASA), glutaraldehyde (GA) and anhydrous (3-glycildyloxypropyl)trimethoxysilane (GOPTMS)) were of the highest purity available (Evonik Industries (Essen, Germany) or Fluka Analytical (USA)), and they were used without further purification. The amino functionalized TiO2 support (TiO2/5-ASA) was prepared as described elsewhere [23] by taking advantage of CTC formation between surface Ti atoms and 5-ASA. 6

ACCEPTED MANUSCRIPT Surface modification of TiO2 nano-powder (Degussa P-25) was performed by dispersing 100 mg of powder in 20 mL of deionized water containing 30.6 mg 5-ASA (molar ratio between TiO2 powder and 5-ASA was 6.25:1). Formation of CTC was indicated by immediate coloration of dispersion. Mixture was stirred overnight, and then modified TiO2 powder was separated by

T

centrifugation, washed several times with water and dried at 40 °C in vacuum oven.

IP

The aldehyde-functionalized TiO2 support (TiO2/5-ASA/GA) was prepared in reaction

CR

between TiO2/5-ASA and GA using method described by Bezbradica et al. [38] Mixture consisting of 1 g of TiO2/5-ASA and 56 µL of GA containing solution (25% of GA in 1.12 mL of

US

0.2 M sodium phosphate buffer) was stirred 1 h at 25 °C. After that, the functionalized support

AN

was thoroughly washed with buffer.

The epoxy-modified TiO2 powder (TiO2/GOPTMS) was prepared using procedure

M

described for silanization of silica nanoparticles [19, 20]. Briefly, 300 mg of TiO2 was dispersed

ED

in 36 mL of anhydrous toluene containing 1.2 ml of (3-glycildyloxypropyl)trimethoxysilane (GOPTMS) under inert atmosphere (molar ratio between TiO2 and GOPTMS was 1:2). Reaction

PT

mixture was stirred for 1 h at 25 °C, but, during that time, every 15 min, reaction mixture was

CE

sonicated for 5 min. Then, epoxy-activated TiO2 powder (TiO2/GOPTMS) was filtrated, rinsed twice with fresh toluene and sonicated for 5 min to remove physically adsorbed GOPTMS.

AC

Finally, the epoxy-functionalized TiO2 powder was dried in the vacuum oven at 40 °C for 24 h. Reflection spectra of pristine and surface-modified TiO2 powder with 5-ASA were measured using Shimadzu UV-Visible UV-2600 spectrophotometer equipped with an integrated sphere ISR-2600 Plus. The amount of epoxy groups introduced on the TiO2 surface was quantified by a titration procedure as described elsewhere [39]. Briefly, 25 mg of TiO2/GOPTMS was suspended in mixture of benzene and acetic acid. The suspension was titrated with HBr/CH3COOH solution 7

ACCEPTED MANUSCRIPT and its volume at equilibrium was used to calculate the amount of epoxy groups (μmol g−1) on the support’s surface.

T

2.2. Numerical calculations

IP

The DFT calculations with the periodic boundary conditions (PBC) were carried out.

CR

Gaussian 09 suite of programs was employed for all calculations [40]. The Heyd-ScuseriaErnzerhof screened hybrid functional (HSE06) [41] was used together with the Pople 6-31G(d,p)

US

valence double-zeta polarized basis set [42]. The geometric optimization of all atomic

AN

coordinates was carried out within the unit cell (7.57×10.54 Å), while the lattice constants were fixed. These models imply infinite vacuum space along the z-direction. The ultrafine integration

M

grid has been specified for all calculations. The unit cells that were used for the periodic

ED

calculations have been constructed based on the defect-free crystal structure of anatase TiO2 (101), which are believed to take part in the photocatalytic reactions [43]. By using this slab

CE

PT

model, 5-ASA was anchored to TiO2 surface to form CTC.

AC

2.3. Production and purification of dextransucrase (DS)

For DS production microorganism Leuconostoc mesenteroides T3, isolated from water kefir grain as described elsewhere [44], was grown using medium developed by Miljković et al. [45].

Briefly, sucrose, yeast extract, phosphate and salts were sterilized separately. Cultures were grown in Erlenmeyer flasks at 23 °C under microaerofilic conditions. The culture broth was 8

ACCEPTED MANUSCRIPT centrifuged at 7000 rpm for 10 min at 4 °C to separate the cells. Than DS was precipitated from cell free supernatant by addition of PEG 400 in ratio 3:1 [46]. The enzyme was allowed to precipitate at 4 °C for 24 h. The mixture was centrifuged at 10000 rpm for 15 min at 4 °C to separate the DS fraction. The pellet was dissolved in 20 mM sodium acetate buffer (pH 5.4).

T

Dissolved enzyme was dialyzed against water using 10 kDa cut-off membrane (Thermo

IP

scientific, USA) in order to purify samples from PEG 400 and other impurities smaller than 10

US

AN

2.4. Enzyme immobilization and activity assay

CR

kDa. The DS solutions were further analyzed for enzyme activity and protein estimation.

The immobilization of DS was performed on pristine TiO2 as well as functionalized TiO2

M

supports (TiO2/5-ASA, TiO2/5-ASA/GA and TiO2/GOPTMS). Briefly, 5 mg of support was

ED

incubated with 0.5 mL of DS solution in 20 mM acetate buffer (pH=5.4). The activity of free enzyme and expressed activity of immobilized enzyme were

PT

determined by incubating 50 μL of enzyme solution or 5 mg of dry immobilized enzyme,

CE

respectively, in 500 μL of reaction mixture (10 % (w/v) sucrose in 20 mM sodium acetate buffer, pH=5.4), by measuring the initial rate of fructose production using the dinitrosalicylic acid

AC

(DNS) method [47]. The stated values are the average of at least three sets of measurements. The activity of free enzyme and expressed activity of immobilized enzyme are expressed in IU mL-1 and IU g-1 units, respectively. The activity immobilization yield (IYa) was calculated using the following equation:

𝐼𝑌𝑎(%) =

𝑒𝑥𝑝𝑟𝑒𝑠𝑠𝑒𝑑 𝑎𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑜𝑓 𝑖𝑚𝑚𝑜𝑏𝑖𝑙𝑖𝑧𝑒𝑑 𝑒𝑛𝑧𝑦𝑚𝑒 𝑎𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑜𝑓 𝑓𝑟𝑒𝑒 𝑒𝑛𝑧𝑦𝑚𝑒

× 100

(1)

9

ACCEPTED MANUSCRIPT

The enzyme concentration was determined according to Lowry’s method.[48] The concentration of bound enzyme (mg of DS per 1 g of support) was determined indirectly as difference between the initial protein concentration and the protein concentration in the

T

supernatant after immobilization. The protein immobilization yield (IYp) was calculated using the

CR

𝑏𝑜𝑢𝑛𝑑 𝑝𝑟𝑜𝑡𝑒𝑖𝑛 𝑐𝑜𝑛𝑐𝑒𝑛𝑡𝑟𝑎𝑡𝑖𝑜𝑛

𝑝𝑟𝑜𝑡𝑒𝑖𝑛 𝑐𝑜𝑛𝑐𝑒𝑛𝑡𝑟𝑎𝑡𝑖𝑜𝑛 𝑖𝑛𝑡𝑟𝑜𝑑𝑢𝑐𝑒𝑑 𝑖𝑛𝑡𝑜 𝑖𝑚𝑚𝑜𝑏𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 𝑝𝑟𝑜𝑐𝑒𝑠𝑠

× 100

(2)

US

𝐼𝑌𝑝(%) =

IP

following equation:

M

AN

2.5. Desorption, thermal and operational stability of immobilized enzyme

The nature of interaction between enzyme and supports, i.e. the amount of covalently

ED

immobilized enzyme on functionalized TiO2 supports (TiO2/5-ASA/GA and TiO2/GOPTMS)

PT

was evaluated using following procedure. In the first step, after 3 h of immobilization, physically adsorbed enzyme was peeled-off from the support using 1 mL of 1 mol L−1 NaCl. The dispersion

CE

was stirred in an orbital shaker (150 rpm) for 30 min at 25 °C. In the second step, to desorb DS

AC

immobilized by hydrophobic interaction between enzyme and support, the samples were treated with 1 mL of 1 % Triton X-100 (continuous shaking at 150 rpm for 30 min). Finally, the samples were centrifuged, washed with distilled water and dried in a vacuum oven at 30 °C for 1 h. The activities of immobilized enzyme were measured before and after treatment with NaCl and Triton X-100. The operational stability of immobilized DS on TiO2/5-ASA/GA and TiO2/GOPTMS supports was tested in five repeated cycles using sucrose as the substrate. For each cycle 5 mg of

10

ACCEPTED MANUSCRIPT immobilized enzyme was incubated with 0.5 mL of substrate at 30 °C. After 5 min the reaction mixture was centrifuged for 30 s at 12000 rpm and the supernatant was quantitatively transferred in tubes containing 0.5 mL of DNS (reagent used for determination of enzyme activity). Activity of immobilized enzyme was determined by measuring the absorbance at 540 nm. Then, the

T

reaction mixture was centrifuged for 2 min in order to separate immobilized enzyme. The

IP

separated immobilized enzyme was washed with 2 mL of sodium acetate buffer and used in the

CR

next cycle on the same manner. The activities of immobilized enzyme in consequent cycles were normalized to the activity obtained in the first run.

US

Influence of temperature on free and immobilized DS on TiO2/5-ASA/GA and

AN

TiO2/GOPTMS was examined by preheating the DS enzyme at 40 oC in different time intervals (15, 45, 120 and 360 minutes, 12 and 24 h). The residual activity was analyzed as described

M

previously. The relative activities of immobilized enzyme were normalized to the activity

CE

PT

3. Results and Discussion

ED

obtained using thermally untreated sample.

AC

3.1. Functionalization of TiO2 nano-powders

Commercially available TiO2 powder (Degussa P25) was chosen for immobilization of dextransucrase (DS) enzymes synthesized by “green” route from Leuconostoc mesenteroides T3. The surface of TiO2 was functionalized with various functional groups and their influence on the efficiency of DS immobilization was compared with pristine support. Microstructural characteristics of Degussa P25 powder are well-established in literature (size: 20‒30 nm; specific surface area: 50‒60 m2g-1; pore diameter: 17.5 nm) [49, 50]. Also, it is well-known that TiO2 is 11

ACCEPTED MANUSCRIPT amphoteric and the zero point charge is at pHZPC=5.9 [51]. It should be noted that in all experiments the pH was 5.4, and consequently the most of the surface hydroxyl groups were protonated. In order to functionalize surface of TiO2 nano-powder with amino group surface-

T

modification with 5-ASA was performed. The Kubelka-Munk transformations of diffuse

IP

reflection data for the unmodified and surface-modified TiO2 nano-powder with 5-ASA are

CR

shown in Fig. 1. The appearance of absorption in visible spectral range upon surface-modification of TiO2 with 5-ASA is consequence of the CTC formation. The extent of red-shift of absorption

US

onset (from about 400 to 700 nm) is consistent with literature data concerning surface-

AN

modification of TiO2 with salicylate-type of ligands [23, 30-32]. It is well-established in literature that binding of 5-ASA to the surface of TiO2 takes place

M

over neighboring hydroxyl and carboxyl groups leaving amino group free [23]. The proposed

ED

coordination is supported by detailed FTIR analysis of free and bound 5-ASA onto TiO2 powder. The presence of vibrations that belong to amino group [52] has been observed upon adsorption of

PT

5-ASA onto TiO2 powders at slightly shifted positions [23]. Having in mind reactivity of amino

CE

group, its availability was important for further, more complex functionalization of TiO2 support. The quantum chemical calculations based on DFT were performed in order to estimate the

AC

energy of charge transfer transition. The optimized structures of 5-ASA molecule anchored onto the (101) surface of anatase-TiO2 is presented as inset to Fig. 1, and, it is identical to the proposed surface structure based on literature data [23, 30-32]. Since, analysis of optical properties of TiO2/5-ASA organic/inorganic hybrid is not the main focuss of this study, the data obtained by DFT calculations are in details presented in Supporting Information (position of energy levels is tabulated and frontier crystal orbitals for TiO2/5-ASA system are graphically presented). At this point, it should be emphasized a reasonable good agreement between 12

ACCEPTED MANUSCRIPT experimentally determined and theoretically calculated band gap values of TiO2/5-ASA hybrid (1.77 and 2.17 eV, respectivelly) confirming in addition proposed coordination of 5-ASA onto TiO2. The TiO2/5-ASA hybrid served as a starting material for preparation of inorganic support

T

functionalized with aldehyde groups in the reaction with glutaraldehyde (GA). Following well-

IP

established procedure in literature [37, 38], the treatment of TiO2/5-ASA with GA was based on

CR

activation of primary amino group introduced onto surface of nanoparticles by reaction with glutaraldehyde (Scheme 1) [35].

US

The different synthetic route was used for functionalization of TiO2 support with epoxy

AN

groups. The silylation reaction between (3-glycildyloxypropyl)trimethoxysilane (GOPTMS) and pristine TiO2 powder, i.e. the condensation reaction between silanol and surface hydroxyl groups

M

that belong to TiO2 was carried out on the same manner as epoxy-silanization of silica

ED

nanoparticles (Scheme 1) [19, 20]. The concentration of epoxy groups was found to be 500 μmol g−1, similar to the value of commercial epoxy-activated carriers such as Eupergit® [53]. Surface

PT

concentration of epoxy groups was determined prior to use of TiO2/GOPTMS for DS

CE

immobilization, since the high density of reactive groups can diminish enzyme activity due to multi-point attachment to functionalized support [20, 54-56].

AC

For the sake of clarity, the synthetic pathways and surface structure of inorganic/organic TiO2 hybrids, prepared to study influence of various functional groups on the immobilization of dextransucrase (DS), are presented in Scheme 1.

3.2. Immobilization of dextransucrase by functionalized TiO2 powders

13

ACCEPTED MANUSCRIPT Kinetic data concerning immobilization of DS onto functionalized TiO2 supports with hydroxyl, amino, aldehyde and epoxy groups are presented in Fig. 2. Kinetic data were fitted using the Lagergren's first-order equation for adsorption kinetics, and the amounts of adsorbed DS at equilibrium and the first-order rate constants are collected and presented in Table 1.

T

Enzyme activity and protein content of DS used in immobilization experiments were 3.8 U mL-1

IP

and 0.16 mg mL-1, respectively. Based on obtained adsorption data, some general features can be

CR

recognized. Firstly, the immobilization of DS onto four studied sorbents obeys to the Lagergren's pseudo first-order equation for adsorption kinetics; corresponding fitted curves have high

US

correlation coefficients (≥0.990). Secondly, the maximum amount of immobilized enzyme is

AN

reached faster when pristine TiO2 nano-powder was used as a sorbent instead of functionalized ones (approximately after 1 and 3 h, respectively). Consequently, the adsorption rate constant for

M

immobilization of protein onto pristine TiO2 is significantly higher compared to the rate constants

ED

for TiO2/5-ASA, TiO2/5-ASA/GA and TiO2/GOPTMS (see Table 1). Thirdly, the sorption capacity of the pristine TiO2 towards DS was found to be the highest (see Table 1). Slightly

PT

smaller sorption capacity has TiO2 powder functionalized with 5-ASA. This result is in

CE

agreement with literature data reported by Zhuang et al. [27] concerning immobilization of enzyme adenosine deaminase onto bare TiO2 and TiO2 functionalized with amino group.

AC

However, the amounts of protein bound to TiO2/5-ASA/GA and TiO2/GOPTMS are about 20% smaller compared to unmodified TiO2 powder. Fourtly, the data concerning protein immobilization yield (Fig. 2 (B)) complement/support kinetic adsorption data. For example, eighty percent of initially available proteins are adsorbed onto unmodified TiO2 powder during the first hour of immobilization. Prolonged adsorption process did not induce increase of the protein immobilization yield. On the other hand, protein immobilization yield onto TiO2/5-ASA reached value close to the pristine TiO2 after three hours. Finally, as expected, protein 14

ACCEPTED MANUSCRIPT immobilization yields onto TiO2/5-ASA/GA and TiO2/GOPTMS after three hours are significantly smaller compared to unmodified TiO2. Enzyme immobilization onto TiO2/5ASA/GA support is faster compared to TiO2/GOPTMS support for the same final amount of enzyme on support.

T

Most likely, the obtained results can be explained in terms of electrostatic interaction

IP

between TiO2 nano-powder and enzyme. As mentioned earlier, the zero point charge of TiO2 is at

CR

pHZPC=5.9 [51], while the isoelectric point of DS is around 4 [57, 58]. Because of that, under stated experimental conditions (pH=5.4), electrostatic attraction between positively charged

US

surface of TiO2 nanoparticles and global anionic character of enzyme accelerates adsorption

AN

process. Inorganic/organic hybrids, synthesized using two different pathways (see Scheme 1), were designed with purpose to facilitate covalent attachment of enzymes to support. However,

M

DS immobilization onto TiO2/5-ASA/GA and TiO2/GOPTMS also follows the Lagergren's

ED

pseudo first-order adsorption kinetics indicating that prior to covalent binding of the enzyme nonspecific adsorption, i.e. association between macromolecule and support takes place [1]. The

PT

lower sorption capacities of TiO2/5-ASA/GA and TiO2/GOPTMS in comparison to pristine TiO2

CE

and TiO2/5-ASA can be explained by smaller number of active sites – functional groups ‒ upon surface modification, compared to the number of positevely charged hydroxyl groups on the

AC

pristine TiO2 or free amino groups from TiO2/5-ASA hybrid. During immobilization of DS on TiO2/5-ASA and TiO2/5-ASA/GA, rapid adsorption via electrostatic interactions between enzyme and support takes place [59]. However, in case of glutaraldehyde activated support rapid adsorption is followed by “intramolecular” reaction between nucleophiles of the enzyme and adjacent introduced glutaraldehyde groups in the support. Slower immobilization on TiO2/5ASA/GA compared to TiO2/5-ASA can be explained as result of hindered approach of large DS molecule to secondary amine obtained after activation with GA [60]. Adsorption comprises all 15

ACCEPTED MANUSCRIPT weak interactions between enzyme and support, such as hydrogen bonding, van der Waals attraction and electrostatic interactions (pristine TiO2 and TiO2/5-ASA supports). This method induces lower degree of conformational changes and activity loss of immobilized enzyme compared to covalent binding (TiO2/5-ASA/GA and TiO2/GOPTMS supports). However, the

T

weak interaction does not prevent the enzyme from leaking. In order to get desired orientation of

IP

enzyme with the best activity and stability, covalent binding between enzyme and support with

CR

desired functionality has to can be established.

Often the activity of immobilized enzymes is not directly proportional to their loading,

US

since their specific activities, dependent on microenvironment and conformational flexibility, can

AN

be modified at high protein loading [19]. To assess applicability and optimization level of TiO2 based supports, the expressed and specific activity of immobilized DS were measured as a

M

function of time (Fig. 3; (A) and (B), respectively), while the data concerning the activity

ED

immobilization yield are presented in Fig. 3 (C).

First, it should be noted that expressed activities of immobilized enzyme increase in a

PT

reverse order compared to their concentration on four different supports (pristine TiO2, TiO2/5-

CE

ASA, TiO2/5-ASA/GA, and TiO2/GOPTMS). For example, pristine TiO2 has the highest sorption capacity (12.4 mg g-1), but immobilized DS has the smallest expressed activity (142 IU g-1). On

AC

the other hand, amount of bound DS is the smallest on TiO2/GOPTMS (9.8 mg g-1), but its expressed activity is the highest (258 IU g-1) among all studied hybrids. The differences between pristine and functionalized TiO2 sorbents are more pronounced when data are normalized to amount of immobilized DS, i.e. when data are expressed as time-dependent specific activity of immobilized enzyme. The shape of kinetic curves for functionalized TiO2 supports (TiO2/5-ASA, TiO2/5-ASA/GA and TiO2/GOPTMS) is consequence of significantly slower immobilization of enzyme compared to the pristine TiO2. Also, at equilibrium, inverse order between protein 16

ACCEPTED MANUSCRIPT immobilization yield values and activity immobilization yield values for pristine and functionalized TiO2 powders was observed (compare Fig. 2 (B) and Fig. 3 (C)). The highest value of activity immobilization yield (64%) was found for TiO2/5-ASA/GA, and, to the best of our knowledge, this value is higher than previously reported data in literature for immobilization

T

of DS isolated from microorganism [1, 61]. There are a limited number of studies concerning DS

IP

immobilization on supports activated with GA [61, 62]. Alcalde et al. [61] compared

CR

immobilization of native and dextran-free DS from L. mesenteroides NRRL B-512F on GA activated amino-silica particles. The reported activity was considerable lower than the activity

US

obtained in this study. It should be emphasized that activity immobilization yield for

AN

TiO2/GOPTMS is close to that for TiO2/5-ASA/GA, while for pristine TiO2 support and TiO2/5ASA activity immobilization yield values were smaller. Most likely, these results can be

M

explained in terms of orientation of enzyme molecules during immobilization on support’s

ED

surface [8] indicating that active centers are probably fully exposed to the medium in the case of DS immobilization on both, TiO2/GOPTMS and TiO2/5-ASA/GA.

PT

In addition, the high yield values for TiO2/5-ASA/GA and TiO2/GOPTMS most likely

CE

can be explained by covalent linkage of DS to hybrid supports. Two steps procedure was applied to confirm this assumption. In the first step, treatment with electrolyte (NaCl) was used to

AC

terminate electrostatic interaction established between the enzyme and support, while, in the second step, treatment with Triton X-100 was applied to eliminate hydrophobic interactions between enzyme and support. Consequently, remaining activity originates from covalently bound enzyme. In particular, approximately 23% of enzyme is linked to TiO2/5-ASA/GA via electrostatic interactions, while hydrophobic interaction between enzyme and support does not take place. Similar results were obtained for immobilization of enzyme on TiO2/GOPTMS. After treatment with NaCl and Triton X-100, enzyme retained 75% of its initial activity. These data 17

ACCEPTED MANUSCRIPT strongly support assumption that DS is prevalently bound via strong covalent bonds to TiO2/5ASA/GA and TiO2/GOPTMS hybrid supports. Also, it seems that covalent immobilization of enzyme over different functional groups (aldehyde and epoxy) exhibited beneficial influence on its activitiy, since the highest specific activities of immobilized DS were achieved with these

T

supports (Fig. 3B).

IP

The most common way for DS immobilization, reported by different group of authors, is

CR

immobilization by entrapment in calcium alginate beads [9, 56, 62]. However, although formation of covalent bond between enzyme and support provide stable system preventing

US

enzyme’s leaching, literature data concerning covalent immobilization of DS are limited [1, 63].

AN

Gomez de Segura et al. [1] reported covalent immobilization of DS on different Eupergit supports. These authors obtained slightly higher activity of immobilized DS in comparison to this

M

study, but applying larger number of purification steps that consequently increases

ED

immobilization cost.

In order to evaluate which one of two functionalized supports, TiO2/5-ASA/GA or

PT

TiO2/GOPTMS, is more suitable for DS immobilization, their thermal and operational stabilities

CE

under long-running working conditions were examined. Thermal stability of free and covalently immobilized DS on TiO2/5-ASA/GA and TiO2/GOPTMS supports at 40 °C are shown in Fig 4A.

AC

Almost the same stability is observed for immobilized DS on both examined supports, but it is much better compared to free enzyme. It is clear that due to formation of covalent bonds enzyme structure is stabilized protecting enzyme from denaturation which can be caused by heating. Expressed activities of immobilized DS on TiO2/5-ASA/GA and TiO2/GOPTMS supports in five repeated cycles are presented in Fig. 4B. After five cycles, DS imobilized on TiO2/5-ASA/GA retained almost 70% of its initial expresssed activity, while, on the other hand, DS immobilized on TiO2/GOPTMS support preserved only 15% of its initial expressed activity. Bassed on 18

ACCEPTED MANUSCRIPT obtained results concerning expressed activity, thermal and operational stability of immobilized DS, TiO2/GOPTMS and specialy TiO2/5-ASA/GA seems to be promissing supports for synthesis of dextran and/or oligosaccharides.

IP

T

4. Conclusion

CR

To conclude, attempt was made to improve immobilization of dextransucrase (DS) on commercial TiO2 catalyst (Degussa P25) using surface modification as a tool to activate

US

inorganic support with various functional groups (amino, glutaraldehyde and epoxy). The

AN

functionalization of TiO2 with amino group led to increased expressed activity of DS in comparison to unmodified powder, but interaction between the enzyme and hybrid support

M

remained electrostatic in its nature. The high values of expressed activity and immobilization

ED

yield were achieved for covalently immobilized DS on TiO2 functionalized with either glutaraldehyde or epoxy groups. Preliminary experiments, performed under long-run working

PT

conditions, indicated potential applicability of DS immobilized on glutaraldehyde-activated TiO2

CE

support for dextran production, since, after five cycles, about 70% of its initial activity was preserved. Also, better thermal stability was achieved for covalently immobilized enzyme in

AC

comparison to free one. In addition, it is important to point out that novel synthetic route for preparation of TiO2 support functionalyzed with glutaraldehyde groups is introduced taking advantage of charge transfer complex formation between surface Ti atoms and 5-aminosalycilic acid. Having in mind diversity of bidentate benzene derivatives ‒ small organic molecules suitable for functionalization of wide band gap oxides – further extension of this synthetic approach seems to be worth of exploring.

19

ACCEPTED MANUSCRIPT Acknowledgements Financial support for this study was granted by the Ministry of Education, Science and

AC

CE

PT

ED

M

AN

US

CR

IP

T

Technological Development of the Republic of Serbia (Projects III 45020 and TR 31035).

20

ACCEPTED MANUSCRIPT References [1] A.G. de Segura, M. Alcalde, M. Yates, M.L. Rojas-Cervantes, N. López-Cortés, A. Ballesteros, F.J. Plou, Immobilization of Dextransucrase from Leuconostocmesenteroides NRRL B-512F on Eupergit C Supports, Biotechnology Progress 20(5) (2004) 1414-1420. [2] Ž. Petronijević, S. Ristić, D. Pešić, A. Šmelcerović, Immobilization of dextransucrase on regenerated benzoyl cellulose carriers, Enzyme and Microbial Technology 40(4) (2007) 763-768.

T

[3] R.K. Purama, A. Goyal, Screening and optimization of nutritional factors for higher dextransucrase

IP

production by Leuconostocmesenteroides NRRL B-640 using statistical approach, Bioresource

CR

Technology 99(15) (2008) 7108-7114.

[4] A.K. Goulas, D.A. Fisher, G.K. Grimble, A.S. Grandison, R.A. Rastall, Synthesis of isomaltooligosaccharides and oligodextrans by the combined use of dextransucrase and dextranase,

US

Enzyme and Microbial Technology 35(4) (2004) 327-338.

[5] S. Davidović, M. Miljković, V. Lazić, D. Jović, B. Jokić, S. Dimitrijević, M. Radetić, Impregnation of

AN

cotton fabric with silver nanoparticles synthesized by dextran isolated from bacterial species Leuconostoc mesenteroides T3, Carbohydrate Polymers 131 (2015) 331-336.

M

[6] O. Barbosa, C. Ortiz, Á. Berenguer-Murcia, R. Torres, R.C. Rodrigues, R. Fernandez-Lafuente, Strategies for the one-step immobilization-purification of enzymes as industrial biocatalysts,

ED

Biotechnology Advances 33(5) (2015) 435-456.

[7] C. Mateo, J.M. Palomo, G. Fernandez-Lorente, J.M. Guisan, R. Fernandez-Lafuente, Improvement of enzyme activity, stability and selectivity via immobilization techniques, Enzyme and Microbial

PT

Technology 40(6) (2007) 1451-1463.

[8] K. Hernandez, R. Fernandez-Lafuente, Control of protein immobilization: Coupling immobilization

CE

and site-directed mutagenesis to improve biocatalyst or biosensor performance, Enzyme and Microbial Technology 48(2) (2011) 107-122.

AC

[9] Z. Ölçer, A. Tanriseven, Co-immobilization of dextransucrase and dextranase in alginate, Process Biochemistry 45(10) (2010) 1645-1651. [10] A. Gupta, K.A. Prabhu, Immobilization and properties of dextransucrase from Leuconostoc mesenteroides culture LM The Journal of General and Applied Microbiology 41(5) (1995) 399-407. [11] C. Kubik, B. Sikora, S. Bielecki, Immobilization of dextransucrase and its use with soluble dextranase for glucooligosaccharides synthesis, Enzyme and Microbial Technology 34(6) (2004) 555-560. [12] A. Reischwitz, K.-D. Reh, K. Buchholz, Unconventional immobilization of dextransucrase with alginate, Enzyme and Microbial Technology 17(5) (1995) 457-461.

21

ACCEPTED MANUSCRIPT [13] S.A.U. Qader, A. Aman, A. Azhar, Continuous Production of Dextran from Immobilized Cells of Leuconostoc mesenteroides KIBGE HA1 Using Acrylamide as a Support, Indian Journal of Microbiology 51(3) (2011) 279-282. [14] J. Kim, J.W. Grate, P. Wang, Nanostructures for enzyme stabilization, Chemical Engineering Science 61(3) (2006) 1017-1026. [15] C. Garcia-Galan, A. Berenguer-Murcia, R. Fernandez-Lafuente, R.C. Rodrigues, Potential of

T

different enzyme immobilization strategies to improve enzyme performance, Advanced Synthesis and

IP

Catalysis 353(16) (2011) 2885-2904.

[16] L. Betancor, F. López-Gallego, A. Hidalgo, M. Fuentes, O. Podrasky, G. Kuncova, J.M. Guisán, R.

CR

Fernández-Lafuente, Advantages of the Pre-Immobilization of Enzymes on Porous Supports for Their Entrapment in Sol−Gels, Biomacromolecules 6(2) (2005) 1027-1030.

US

[17] A.A. Gokhale, J. Lu, I. Lee, Immobilization of cellulase on magnetoresponsive graphene nanosupports, Journal of Molecular Catalysis B: Enzymatic 90 (2013) 76-86. [18] M.I. Kim, H.O. Ham, S.-D. Oh, H.G. Park, H.N. Chang, S.-H. Choi, Immobilization of Mucor

AN

javanicus lipase on effectively functionalized silica nanoparticles, Journal of Molecular Catalysis B: Enzymatic 39(1) (2006) 62-68.

M

[19] K. Banjanac, M. Mihailović, N. Prlainović, M. Stojanović, M. Carević, A. Marinković, D. Bezbradica, Cyanuric chloride functionalized silica nanoparticles for covalent immobilization of lipase,

ED

Journal of Chemical Technology & Biotechnology 91(2) (2016) 439-448. [20] K. Banjanac, M. Mihailović, N. Prlainović, M. Ćorović, M. Carević, A. Marinković, D. Bezbradica,

PT

Epoxy-silanization – tool for improvement of silica nanoparticles as support for lipase immobilization with respect to esterification activity, Journal of Chemical Technology & Biotechnology 91(10) (2016)

CE

2654-2663.

[21] K. Pashangeh, M. Akhond, H.R. Karbalaei-Heidari, G. Absalan, Biochemical characterization and

AC

stability assessment of Rhizopus oryzae lipase covalently immobilized on amino-functionalized magnetic nanoparticles, International Journal of Biological Macromolecules 105 (2017) 300-307. [22] E.P. Cipolatti, M.J.A. Silva, M. Klein, V. Feddern, M.M.C. Feltes, J.V. Oliveira, J.L. Ninow, D. de Oliveira, Current status and trends in enzymatic nanoimmobilization, Journal of Molecular Catalysis B: Enzymatic 99 (2014) 56-67. [23] B. Milićević, V. Đorđević, D. Lončarević, S.P. Ahrenkiel, M.D. Dramićanin, J.M. Nedeljković, Visible light adsorption of surface modified TiO2 powder with bidentate benzene derivatives, Micropor. Mesopor. Mat. 217 (2015) 184-189.

22

ACCEPTED MANUSCRIPT [24] D. Mihailović, Z. Šaponjić, R. Molina, N. Puač, P. Jovančić, J. Nedeljković, M. Radetić, Improved Properties of Oxygen and Argon RF Plasma-Activated Polyester Fabrics Loaded with TiO2 Nanoparticles, ACS Applied Materials & Interfaces 2(6) (2010) 1700-1706. [25] S.H. Bang, A. Jang, J. Yoon, P. Kim, J.S. Kim, Y.-H. Kim, J. Min, Evaluation of whole lysosomal enzymes directly immobilized on titanium (IV) oxide used in the development of antimicrobial agents, Enzyme and Microbial Technology 49(3) (2011) 260-265.

T

[26] C. Yang, H. Wu, J. Shi, X. Wang, J. Xie, Z. Jiang, Preparation of Dopamine/Titania Hybrid

IP

Nanoparticles through Biomimetic Mineralization and Titanium(IV)–Catecholate Coordination for Enzyme Immobilization, Industrial & Engineering Chemistry Research 53(32) (2014) 12665-12672.

CR

[27] W. Zhuang, Y. Zhang, L. He, R. An, B. Li, H. Ying, J. Wu, Y. Chen, J. Zhou, X. Lu, Facile synthesis of amino-functionalized mesoporous TiO2 microparticles for adenosine deaminase immobilization,

US

Microporous and Mesoporous Materials 239 (2017) 158-166.

[28] C. Ji, L.N. Nguyen, J. Hou, F.I. Hai, V. Chen, Direct immobilization of laccase on titania nanoparticles from crude enzyme extracts of P. ostreatus culture for micro-pollutant degradation,

AN

Separation and Purification Technology 178 (2017) 215-223.

[29] C. Mateo, G.F. Fernandez-Lorente, O. Abian, R. Fernandez-Lafuente, J.M. Guisan, Multifunctional

M

epoxy supports: a new tool to improve the covalent immobilization of proteins. The promotion of physical adsorptions of proteins on the supports before their covalent linkage, (1525-7797 (Print)) (2000).

ED

[30] T.D. Savić, M.I. Čomor, N.D. Abrazović, Z.V. Šaponjić, M.T. Marinović-Cincović, D.Ž. Veljković, S.D. Zarić, I.A. Janković, Anatase nanoparticles surface modified with fused ring salicylate-type ligands

PT

(1-hydroxy- 2-naphthoic acids): a combined DFT and experimental study, J. Alloy Compd. 630 (2015) 226-235.

CE

[31] T.D. Savić, Z.V. Šaponjić, M.I. Čomor, J.M. Nedeljković, M.D. Dramićanin, M.G. Nikolić, D.Ž. Veljković, S.D. Zarić, I.A. Janković, Surface modification of anatase nanoparticles fused ring salicylate-

AC

type ligands (3-hydroxy- 2-naphthoic acids): a combined DFT and experimental study of optical properties, Nanoscale 5 (2013) 7601-7612. [32] I.A. Janković, Z.V. Šaponjić, M.I. Čomor, J.M. Nedeljković, Surface modification of colloidal TiO 2 nanoparticles with bidentate benzene derivatives, J. Phys. Chem. C 113 (2009) 12645-12652. [33] I.M. Dugandžić, D.J. Jovanović, L.T. Mančić, N. Zheng, S.P. Ahrenkiel, O.B. Milošević, Z.V. Šaponjić, J.M. Nedeljković, Surface modification of submicronic TiO2 particles prepared by ultrasonic spray pyrolysis for visible light absorption, Journal of Nanoparticle Research 14(10) (2012) 1157. [34] I.M. Dugandžić, D.J. Jovanović, L.T. Mančić, O.B. Milošević, S.P. Ahrenkiel, Z.V. Šaponjić, J.M. Nedeljković, Ultrasonic spray pyrolysis of surface modified TiO2 nanoparticles with dopamine, Materials Chemistry and Physics 143(1) (2013) 233-239. 23

ACCEPTED MANUSCRIPT [35] P.G. Vazquez-Ortega, M.T. Alcaraz-Fructuoso, J.A. Rojas-Contreras, J. López-Miranda, R. Fernandez-Lafuente, Stabilization of dimeric β-glucosidase from Aspergillus niger via glutaraldehyde immobilization under different conditions, Enzyme and Microbial Technology 110 (2018) 38-45. [36] H. Zaak, L. Fernandez-Lopez, C. Otero, M. Sassi, R. Fernandez-Lafuente, Improved stability of immobilized lipases via modification with polyethylenimine and glutaraldehyde, Enzyme and Microbial Technology 106 (2017) 67-74.

T

[37] L. Betancor, F. López-Gallego, A. Hidalgo, N. Alonso-Morales, G.D.-O.C. Mateo, R. Fernández-

IP

Lafuente, J.M. Guisán, Different mechanisms of protein immobilization on glutaraldehyde activated supports: Effect of support activation and immobilization conditions, Enzyme and Microbial Technology

CR

39(4) (2006) 877-882.

[38] D.I. Bezbradica, C. Mateo, J.M. Guisan, Novel support for enzyme immobilization prepared by

US

chemical activation with cysteine and glutaraldehyde, Journal of Molecular Catalysis B: Enzymatic 102 (2014) 218-224.

[39] M. Mihailović, M. Stojanović, K. Banjanac, M. Carević, N. Prlainović, N. Milosavić, D. Bezbradica,

Process Biochemistry 49(4) (2014) 637-646.

AN

Immobilization of lipase on epoxy-activated Purolite® A109 and its post-immobilization stabilization,

M

[40] M.J. Frisch, Gaussian 09, Revision B.01, , Gaussian, Inc., Wallingford CT, 2009. [41] J. Heyd, G.E. Scuseria, M. Ernzerhof, Hybrid functionals based on a screened Coulomb potential, J.

ED

Chem. Phys. 118 (2003) 8207-8211.

[42] W.J. Hehre, R. Ditchfield, J.A. Pople, Self—Consistent Molecular Orbital Methods. XII. Further

PT

Extensions of Gaussian—Type Basis Sets for Use in Molecular Orbital Studies of Organic Molecules, J. Chem. Phys. 56 (1972) 2257-2262.

CE

[43] S. Higashimoto, T. Nishi, M. Yasukawa, M. Azuma, Y. Sakata, H. Kobayashi, Photocatalysis of Titanium Dioxide Modified by Catechol-Type Interfacial Surface Complexes (Isc) with Different

AC

Substituted Groups, J. Catal. 329 (2015) 286-291. [44] S. Davidovic, M. Miljkovic, D. Antonovic, M. Rajilic-Stojanovic, S. Dimitrijevic-Brankovic, Water Kefir grain as a source of potent dextran producing lactic acid bacteria, Hemijska industrija 69(6) (2015) 595-604.

[45] M. Miljković, S. Davidović, M. Rajlić-Stojanović, S. Šiler-Marinković, S. Dimitrijević-Branković, Screening for factors affecting dextransucrase production from Leuconostoc mesenteroides isolated from water kefir graints using statistical approach 8th-International Conference of the Chemical Societies of the South-East European Countries - ICOSECS 8, Serbia, Belgrade, 2013, p. 264.

24

ACCEPTED MANUSCRIPT [46] A. Goyal, S.S. Katiyar, Fractionation of Leuconostoc mesenteroides NRRL B-512F dextran sucrase by polyethylene glycol: a simple and effective method purification, Journal of Microbiological Methods 20(3) (1994) 225-231. [47] G. Miller, Use of dinitrosalicylic acid reagent for determination of reducing sugars, Anal Chem 31 (1959) 426-429. [48] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the Folin phenol

T

reagent, J Biol Chem 193 (1951) 265–275.

IP

[49] C.B. Almquist, P. Biswas, Role of Synthesis Method and Particle Size of Nanostructured TiO2 on Its Photoactivity, J Catal 212 (2002) 145-156.

CR

[50] K. Suttiponparnit, J. Jiang, M. Sahu, S. Suvachittanont, T. Charinpanitkul, P. Biswas, Role of Surface Area, Primary Particle Size, and Crystal Phase on Titanium Dioxide Nanoparticle Dispersion Properties,

US

Nanoscale Res Lett 6 (2011) 27-34.

[51] M. Kosmulski, The significance of the difference in the point of zero charge between rutile and anatase, Advances in colloid and interface science 99(3) (2002) 255-64.

AN

[52] S. Mourdikoudis, L.M. Liz-Marzan, Oleylamine in Nanoparticle Synthesis, Chem. Mater. 25 (2013) 1465-1476.

M

[53] E. Katchalski-Katzir, D.M. Kraemer, Eupergit® C, a carrier for immobilization of enzymes of industrial potential, Journal of Molecular Catalysis B: Enzymatic 10(1) (2000) 157-176.

ED

[54] C. Mateo, O. Abian, R. Fernandez–Lafuente, J.M. Guisan, Increase in conformational stability of enzymes immobilized on epoxy-activated supports by favoring additional multipoint covalent attachment

PT

☆, Enzyme and Microbial Technology 26(7) (2000) 509-515. [55] V. Grazu, F. López-Gallego, J.M. Guisán, Tailor-made design of penicillin G acylase surface enables

CE

its site-directed immobilization and stabilization onto commercial mono-functional epoxy supports, Process Biochemistry 47(12) (2012) 2538-2541.

AC

[56] D.-H. Zhang, L.-J. Peng, Y. Wang, Y.-Q. Li, Lipase immobilization on epoxy-activated poly(vinyl acetate-acrylamide) microspheres, Colloids Surf B Biointerfaces 129 (2015) 206-210. [57] A.M. Chludzinski, G.R. Germaine, C.F. Schachtele, Purification and properties of dextransucrase from Streptococcus mutans, Journal of Bacteriology 118(1) (1974) 1-7. [58] K. Itaya, T. Yamamoto, Dextransucrase as an enzyme associating with alkaline earth metal ions, Agr. Biol. Chem. 39(6) (1975) 1187-1192. [59] O. Barbosa, C. Ortiz, A. Berenguer-Murcia, R. Torres, R.C. Rodrigues, R. Fernandez-Lafuente, Glutaraldehyde in bio-catalysts design: a useful crosslinker and a versatile tool in enzyme immobilization, RSC Advances 4(4) (2014) 1583-1600. [60] J.M. Guisan, Immobilization of Enzymes and Cells, Human Press, Totowa, New Jersey, 2006. 25

ACCEPTED MANUSCRIPT [61] M. Alcalde, F.J. Plou, A. Gómez de Segura, M. Remaud-Simeon, R.M. Willemot, P. Monsan, A. Ballesteros, Immobilization of native and dextran-free dextransucrases from Leuconostoc mesenteroides NRRL B-512F for the synthesis of glucooligosaccharides, Biotechnology Techniques 13(11) (1999) 749755. [62] A.M. Hashem, M.A. El-Refaei, H.M. Gebril, F. Abdel-Fattah, Immobilization of Leuconostoc paramesenteroides Dextransucrase Enyzme and Characteriyation of its Enzyme Properties, Journal of

T

Basic & Applied Sciences 8 (2012) 344-352.

IP

[63] M. Parlak, D. Ustek, A. Tanriseven, A novel method for covalent immobilization of dextransucrase,

AC

CE

PT

ED

M

AN

US

CR

Journal of Molecular Catalysis B: Enzymatic 89(Supplement C) (2013) 52-60.

26

ACCEPTED MANUSCRIPT Scheme 1. Schematic presentation of synthetic pathways for preparation of TiO2 support functionalized with four different groups: (a) protonated hydroxyl group, (b) amino group introduced by functionalization of TiO2 with 5-ASA, (c) epoxy group introduced by silylation of TiO2 with GOPTMS, and (d) aldehyde group introduced by reaction between amino group and

AC

CE

PT

ED

M

AN

US

CR

IP

T

glutaraldehyde.

27

ACCEPTED MANUSCRIPT Table 1. Analysis of the adsorption of DS on functionalized TiO2 support using the Lagergren's first-order equation for adsorption kinetics. The sorption

The adsorption

group

capacity (mg g-1

constant

support)

(h-1)

Hydroxyl

12.6

3.80

TiO2/5-ASA

Amino

12.0

1.30

TiO2/5-ASA/GA

Aldehyde

9.6

TiO2/GOPTMS

Epoxy

9.8

CR

TiO2

T

Functional

IP

Support

R2

0.992 0.990 0.996

3.80

0.992

AC

CE

PT

ED

M

AN

US

1.85

28

ACCEPTED MANUSCRIPT Figure Captions

Fig. 1. Kubelka-Munk transformations of diffuse reflection data of pristine TiO2 powder (a) and surface-modified TiO2 powder with 5-ASA (b); inset: the optimized geometry, used in DFT

IP

T

calculations, of 5-ASA anchored onto the (101) surface of anatase-TiO2.

CR

Fig. 2. (A) Adsorption kinetic data of DS on pristine TiO2 (black), TiO2/5-ASA (red), TiO2/5ASA/GA (blue), and TiO2/GOPTMS (green), as well as corresponding fitted curves obtained

US

using the Lagergren's first-order equation for adsorption kinetics. (B) Protein immobilization

AN

yields, IYp(%), as a function of time.

M

Fig. 3. (A) Expressed activity (IU per g of support), (B) specific activity (IU per mg of protein)

ED

and (C) activity immobilization yields of immobilized DS on TiO2, TiO2/5-ASA, TiO2/5-

PT

ASA/GA, and TiO2/GOPTMS supports.

CE

Fig. 4. (A) Thermal stability of free and immobilized DS on TiO2/5-ASA/GA and TiO2/GOPTMS supports at 40 oC, and (B) operation stability of immobilized DS at 30 °C in five

AC

repeated cycles.

29

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T

Fig. 1.

30

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T

Fig. 2.

31

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T

Fig. 3.

32

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T

Fig. 4.

33

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T

Graphical abstract

34

ACCEPTED MANUSCRIPT Highlights

CE

PT

ED

M

AN

US

CR

IP

T

Novel approach for synthesis of amino- and aldehyde-activated TiO2 supports. Improved immobilization of dextransucrase (DS) onto surface-modified TiO2 powders. The high enzyme activity of covalently bound DS on functionalized TiO2 powders. Efficient dextran production under long-running experimental conditions. Comparison of experimental data and quantum chemical calculations.

AC

    

35