- Email: [email protected]
Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles
COREL-08201; No of Pages 7 Journal of Controlled Release xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Controlled Release journal homepage: www.elsevier.com/locate/jconrel
Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles Ilia Rivkin a,1, Yifat Galnoy-Glucksam a,1, Inbar Elron-Gross a, Amichay Afriat b, Arik Eisenkraft b,c,d, Rimona Margalit a,⁎ a
Department of Biochemistry and Molecular Biology, The George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv, Israel NBC Protection Division, IMOD, The Hebrew University, Jerusalem, Israel IDF Medical Corps, The Hebrew University, Jerusalem, Israel d The Institute for Research in Military Medicine, The Faculty of Medicine, The Hebrew University, Jerusalem, Israel b c
a r t i c l e
i n f o
Article history: Received 29 December 2015 Received in revised form 28 March 2016 Accepted 28 March 2016 Available online xxxx Keywords: TIC exposure Mass casualty events Aerosols Respiratory damage
a b s t r a c t The purpose of this study was to develop a treatment for respiratory damage caused by exposure to toxic industrial chemicals (TICs), including mass casualty events, by aerosols of dexamethasone and/or N-acetyl cysteine formulated in targeted lipid-based particles. Good encapsulation, performance as slow-release drug depots, conservation of matter, and retention of biological activity were obtained for the three drug-carrier formulations, pre- and post-aerosolization. Weight changes over a 2 week period were applied, deliberately, as a non-invasive clinical parameter. Control mice gained weight continuously, whereas a non-lethal 30 minute exposure of mice to 300 ppm Cl2 in air showed a two-trend response. Weight loss over the first two days, reversing thereafter to weight gain, but at a rate and level significantly slower and smaller than those of the control mice, indicating the chlorine damage was long-term. The weight changes of Cl2-exposed mice given the inhalational treatments also showed the two-trend response, but the weight gain rates and levels were similar to those of the control mice, reaching the weight-gain range of the control mice. Following this proof of concept, studies are now extended to include additional TICs, and biochemical markers of injury and recovery. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Respiratory damage is among the major injuries caused by exposure, whether accidental or deliberate, to toxic industrial chemicals (TICs), such as chlorine, ammonia, hydrogen chlorine, and many others [1–3]. Treatment of such damage – addressing the surge of reactive oxygen species, the pulmonary inflammation and pulmonary edema – is still in the category of an unmet therapeutic need, despite the availability of effective drugs [1–3]. The problem is not with the drugs themselves, but with their delivery. In the available formulations the drugs are free, hence prone to the well-known deficiencies of treatment with free drugs that too often result in poor therapeutic responses, treatment failure and safety limitations [4]. The approach we devised, to turn the situation – including in cases of mass casualty events – from the current unmet, to a met, therapeutic need is composed of three linked elements: (1) to replace treatment with free drugs by drugs formulated in a targeted lipidbased carrier that can, furthermore, perform as a slow-release drug ⁎ Corresponding author at: Department of Biochemistry and Molecular Biology, Tel Aviv University, Tel Aviv 69978, Israel. E-mail address: [email protected] (R. Margalit). 1 Equal contribution.
depot (2) to deliver the drug-carrier formulations in an aerosol form, directly to the airways and lungs, using a clinical portable inhalation device, and (3) to have both formulations and devices in the vehicles of the first responders so that treatment can be initiated at the field and continued thereafter in the patient's home or in a medical facility. The carrier we selected for the task is a specialized multilamellar liposome, surface-modified by hyaluronan anchored covalently to its surface (denoted HA-L). These liposomes were shown to have active targeting to macrophages and can also bind with high affinity to the extra cellular matrix (ECM) [5–6]. The drugs selected were the antioxidant N-Acetyl-cysteine (NAC) and the anti-inflammatory corticosteroid dexamethasone (Dex). A portable nebulizer was the inhalation device. We first formulated the drugs in the liposomes, pursued and optimized physicochemical properties, verified that the drug-carrier formulations were stable to the nebulization process and simulated the deposition of the aerosol in human airways and lungs using an Anderson Cascade Impactor (ACI). We next conducted the first step in feasibility studies, in mice exposed to a chlorine–air mixture and treated with an aerosol of our novel drug-carrier formulations. The respiratory damage in mice, exposed to chlorine–air mixtures either by whole body or nose only, has been well established [7–13]. Similar respiratory damage in mice,
http://dx.doi.org/10.1016/j.jconrel.2016.03.039 0168-3659/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
2
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx
caused by a host of other agents, was also reported [17–21]. Both invasive and non-invasive approaches have been applied to evaluate the respiratory damage and the impact of therapy. In most of those cases, where invasive approaches were applied, the experiments were for short durations, typically 1–3 days from insult. The non-invasive approach, following mice weight changes, has been applied quite extensively, mostly for periods of 3–21 days [18–21]. Defining, moreover, weight loss as a traditional indicator of an acute pulmonary bleomycin insult, Limjunyawong et al. followed mice weight changes even up to 200 days [17]. Given that our objective was to follow-up each animal for a relatively prolonged period (2–3 weeks) from exposure with/without treatment we opted for this first step in feasibility studies, to apply a noninvasive approach, such as continuous monitoring of animal weight [11,13–21]. Our goal was to have the weight changes of the chlorineexposed and treated mice on a par with those of control mice (exposed to air alone).
2.2.2. Drug encapsulation 2.2.2.1. N-acetyl cysteine (NAC). Buffer-free, salt-free lyophilized HA-L powders (see Section 2.2.1) were brought to room temperature and rehydrated back to original pre-lyophilization volume by the NAC stock solution (20 mg NAC/ml in PBS containing 12.5 mM EDTA at the final pH of 7.5) incubating the system at 37 °C for 24 h. 2.2.2.2. Dexamethasone (Dex). Dex-encapsulating HA-L were prepared essentially as described in Section 2.2.1 above, except the drug was added to the ethanolic lipid solution at the concentration range of 5– 10 mg Dex/ml. 2.2.2.3. Co-encapsulation of Dex and NAC in the same liposome. Dexencapsulating HA-L were prepared as in Section 2.2.2.2, and lyophilized from an aqueous suspension that was buffer-free and salt-free. These lyophilized liposomes were rehydrated in the NAC solution, as described in Section 2.2.2.1, above.
2. Materials and methods 2.3. Kinetics of drug efflux and determination of encapsulation efficiency 2.1. Materials Phospholipon 90G (high purity Soybean phosphatidylcholine (SPC)) was a kind gift from Nattermann Phospholipid GmbH (Cologne Germany). Dexamethasone, NAC, dipalmitoylphosphatidylethanolamine (DPPE), cholesterol (CH) and EDC (ethyl-dimethyl-aminopropylcarbodiimide) were from Sigma Chemical Co. (St. Louis, USA). Hyaluronan (HA) 1.5 MDa used for the liposomes was a kind gift from Genzyme (Cambridge, MA, USA). The chlorine gas was in a mini-cylinder (Portacyl®, Specialty Gases of America Inc., The American Gas Group, purchased in Israel from Maxima Company) 300 ppm chlorine balance air gas mixture. Total gas quantity per cylinder was 58 l, and the rate of gas flow was 0.5 l/min supplied via series fixed flow regulator 70 (Ashcroft®, inlet: 1000 PSIG, flow: 0.5 SLPM, purchased in Israel from Maxima Company). Liquid Scintillation cocktail, Ultima Gold™, was from PerkinElmer Life and Analytical Sciences Inc. (USA). Dialysis tubing (molecular weight cutoff of 12,000–14,000) was from Spectrum Medical Industries (Los Angeles, CA). All other reagents were of analytical grade. Ultracentrifugation was performed with a Sorval Discovery M120 SE micro ultracentrifuge (TN, USA). Lyophilization was performed with a HETO Drywinner 3 (Alleraod, Denmark). The nebulizer was DeVILBISS's PulmoAide Compact Compressor 3655. The Anderson Cascade Impactor was from ThermoFisher Scientific (Franklin, MA, USA). 2.2. Preparation of drug-free and of drug-encapsulating HA-L 2.2.1. Drug-free HA-L The lipid composition was SPC:DPPE:CH 75:5:20 (mole ratios) and the total lipid concentration was 100 mg/ml. Liposome preparation was essentially as described under [21,23] (and the references within). The first step was preparation of regular multilamellar liposomes (RL). The lipids were dissolved in ethanol, and evaporated to obtain a dry lipid film in a rotary evaporator under reduced pressure. The swelling solution (0.1 M borate buffer at pH 9) was added to the lipid film and the system was incubated (in a shaker bath) for 2 h at 65 °C. To obtain HA-L, HA was dissolved in acetate buffer (0.1 M, pH 4.5) at the concentration of 2 mg/ml. It was pre-activated by incubation with EDC (20 mg per 1 mg HA) for 2 h at 37 °C and added to the RL suspension, at the ratio of 1:1 (v/v). This reaction mixture was incubated in a shaker bath for 24 h at 37 °C. The HA-L were freed from excess materials and byproducts by centrifugation for 30 min at 4 °C and a g force of 160,850 followed by several successive washes and re-centrifugations in 0.1 M NH4HCO3, suspending the final pellets in this salt solution. Aliquots of 1 ml of these liposomes were frozen for 2 h at −80 °C, followed by lyophilization. The resultant liposome powders were stored at − 18 °C until further use.
2.3.1. NAC/HA-L Kinetics of drug efflux from the liposomes was studied according to our previously-developed experimental set up and data processing [21–24] (and the references within). Briefly, a suspension of drugencapsulating liposomes (see Section 2.2.2.1 above) was placed in a dialysis sac that was immersed in a continuously-stirred receiver vessel containing drug-free buffer (PBS, 12.5 mM EDTA, pH 7.5), receiver to liposome volume ratio was 15:1. At designated time points, the dialysis sac was transferred from one receiver vessel to another containing fresh drug-free buffer. NAC concentration was determined in each dialysate and in the sac (at the beginning and end of the run). The data were analyzed according to a previously derived multi-pool kinetic model, expressed in Eq. (1) below. ft ¼
n X f i 1− exp−ki t
ð1Þ
i¼1
where ft is the fraction of the total drug in the liposomal system that diffused out of the sac at time t, n is the number of independent drug pools in the liposomal system, fi is the fraction of the total drug in the system that occupied the ith pool at time = 0, ki is the rate constant for drug efflux from the ith pool, and t is time, the free variable. Data processing was done by computer-aided non-linear regression analysis using the KaleidaGraph software. These kinetic experiments also yield the efficiency of drug encapsulation (fi, for the encapsulateddrug pool, at time = 0) which is defined as the ratio of liposomeencapsulated drug to the total drug in the system. 2.3.2. Dex/HA-L All methods were essentially as listed above for NAC, with the following exceptions. The lyophilized dry powder of Dex/HA-L (see Section 2.2.2.2 above) was rehydrated back to original volume with drug-free water or buffer. The dialysis and data processing were performed as in Section 2.3.1 above. 2.3.3. NAC + DEX/H-L The dialysis and data processing were performed as in Section 2.3.1 above, except that the liposomal formulation in the dialysis sac was as detailed in Section 2.2.2.3 above. 2.4. Drug assays in solution and in liposomal samples Dex was assayed in free and in liposomal formulations as previously described, by inclusion of trace 3H-dexamethasone in all Dexcontaining systems [23]. NAC was assayed by a colorimetric assay
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx
3
Scheme 1. The inhalation chamber.
based on the Ellman reaction for determination of thiols [25].We have previously adapted this assay for liposomal samples and 96-well plates [26]. 2.5. Making and characterizing drug-liposome aerosols 2.5.1. Nebulizer set up and operation The nebulizer set up developed for aerosol collection and the optimized experimental conditions are listed together with the relevant results (Section 3.2). The devise was operated according to the following manufacturer's specifications: Compressor free air flow ≥ 8 l/pm, nebulization rate ≥ 0.15 ml/min, supplied nebulizer operating pressure 0.65 bar, and supplied nebulizer liter flow 5.5 l/min. 2.5.2. Physicochemical characterization of the aerosolized liposomes The following parameters were measured to assess quantitative, post nebulization, recovery: (i) the volume of sample put into the nebulization cup, and the volume of the collected aerosol; (ii) lipid and drug concentrations for each sample, pre- and post-nebulization. Dex and NAC were assayed as described in Section 2.4 above, and the lipid was assayed in drug-free liposomes by inclusion of trace 3H-cholesterol in the formulations. To evaluate the impact of nebulization on system properties, test samples were assayed - before and after nebulization for drug encapsulation efficiency and drug efflux kinetics, as described in Section 2.3 above. 2.6. Simulating lung deposition, using the Anderson Cascade Impactor (ACI) The nebulizer was connected to the input of the device, and a water trap was connected to the device output. The nebulizer cup was filled with 7 ml of the test material, and the nebulizer was operated as described in Section 2.5.1 above. At the end of the run, the nebulizer was disconnected from the ACI, the ACI disassembled, the material collected from each device jet stage and assayed for lipid and each drug, as detailed in Section 2.5.2 above.
2.7. The animal model studies 2.7.1. The inhalation chamber The inhalation chamber, shown in Scheme 1, was manufactured in the Tel Aviv University machine shop according to our specifications, and can accommodate up to 10 mice. It is made of Plexiglas, and its dimensions are 13 ∗ 13 ∗ 24 cm. The complete system (i.e., the inhalation chamber(s), the nebulizer and the chlorine/air cylinder) was placed in a chemical hood, within the animal facility of Tel Aviv University.
2.7.2. The experimental design and protocol Male Balb/C mice 10–12 weeks old, average weight at day = 0 in the range of 20 g were from the animal facility of TAU. The Tel Aviv University Institutional Animal Care and Use Committee approved all animal procedures according to the guidelines of the Office of Laboratory Animal Welfare (permit # L-12049). Animals were group-housed (7 per cage) in solid-bottomed plastic cages designed to allow easy access to standard laboratory food and water ad lib. The mice were kept in a 12:12 light–dark cycle in a controlled temperature chamber (24 ± 1 °C). Typically, 5 mice were put into the inhalation chamber to receive the following treatments and there were usually 2–3 repeats (see legends to figures for specific animal numbers/group): (1) Control group: mice Table 1 Physicochemical properties of NAC + Dexamethasone formulated in HA-L, pre- and postnebulization. Drug
Encapsulation and efflux of encapsulated drug % Encapsulation
k2 ∗ 1000 (h−1)
Half-life (h)
NAC Pre-nebulization Post-nebulization
31 (±2) 26 (±2)
0.020 (±0.005) 0.022 (±0.005)
35 32
DEX Pre-nebulization Post-nebulization
93 (±1) 91 (±2)
0.013 (±0.0005) 0.024 (±0.0009)
53 29
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
4
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx
2.7.3. Data processing and statistics The % change in weight (from the initial weight) was calculated for each mouse at each experimental run, for each weighing day. The % change in weight was then averaged for each group (i.e. mice receiving air, chlorine or chlorine and treatment) for each day, and the relevant SEM was calculated.
3. Results and discussion 3.1. Physicochemical characterizations of the drug-liposome formulations Fig. 1. Lipid and drug recoveries post-aerosolization. L, L-D, L-N and L-ND denote drug-free (“empty”) liposomes, dexamethasone-encapsulating liposomes, NAC-encapsulating liposomes and dexamethasone + NAC encapsulated in the same liposome, respectively. Solid white bar: lipids' recovery (liposomes, L) light-gray bars: dexamethasone recovery in L-D (left bar) and in L-ND (right bar) dark-gray bars: NAC recovery in L-N (left bar) and in L-ND (right bar). Each bar is the average of at least three determinations, and the error bars represent the standard deviations. The bars were drawn with depth to be consistent with Figs. 5 and 6, where the depth assists in visualizing data values close to zero.
receiving air only (through the nebulizer, as illustrated in Scheme 1), for 30 to 50 min; (2) chlorine-exposed un-treated mice: mice receiving the chlorine–air mixture (300 ppm chlorine in air) for 30 min; (3) chlorineexposed then treated mice: mice exposed to chlorine as in group 2 above, immediately followed by exposure to the aerosol of a selected treatment (given through the nebulizer). Treatments were selected from free and liposomal Dex, NAC or a combination of both drugs. Further details of the treatments including doses are provided with the results. Each mouse was tagged, and was weighed before being put into the inhalation chamber, and thereafter every 2–3 days for a period of 2–3 weeks. The mice were monitored throughout the run period for survival and behavior.
Three drug-carrier formulations were made and characterized. Two were single drug formulations, namely Dex in HA-L, and NAC in HA-L, denoted L-D and L-N, respectively. The third formulation, denoted LND, contained both drugs in the same liposome, the lipophilic Dex occupying the lipid regions of the liposome and the hydrophilic NAC occupying the aqueous regions of the liposome, especially in the inner aqueous core. Determination of the encapsulation efficiency and kinetics of drug efflux was performed for each formulation, as described under Section 2.3, and the data were processed according to Eq. (1). For all systems the data fit two initial drug pools, one encapsulated and the other the excess unencapsulated drug. Typical results, for the L-ND system are listed in Table 1. Looking, at this stage, at the pre-nebulization data listed in Table 1, the encapsulated doses for each of the drugs were at therapeutic levels, with high and moderate encapsulation efficiencies obtained for Dex and NAC, respectively. The half-lives of drug efflux that range from 35 to 50 h, are clear indications that this system performs as the desired slow-release drug depot for each of the drugs. Results for the other two formulations, L-D and L-N were quite similar.
Fig. 2. Simulated deposition of the hyaluronan liposomes (HA-L) in human airways and lungs, using the Anderson Cascade Impactor (ACI) device. Right-hand side: Scheme of the corresponding aerosol-particle size ranges (μm) and localization in human airways and lungs. Left-hand side: the liposome deposition. Each bar is an average of 3 independent experiments and the error bars represent the standard deviations.
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx
5
Fig. 3. Mice in the inhalation chamber — behavioral responses as function of exposure type. A. Control mice exposed to air only. B. Mice exposed to the chlorine/air gas mixture (300 ppm chlorine in air).
3.2. The liposomal aerosols and retention of pre-aerosolization properties Nebulization is not a gentle process, which raised the concern that the aerosolized liposomes may be damaged, losing material (lipids and/or drugs), losing encapsulated matter (i.e., encapsulated drug moving out of the liposome) or losing the slow-release nature. To evaluate whether such damages occurred and if so, their extent, each liposome formulation was aerosolized and the aerosol collected as described in Section 2.5. Typically 7 ml of the desired liposome formulation was put into the nebulization cup, 80% of which were recovered. Material recovery (Fig. 1) was quite satisfactory. There were no lipid losses, nor NAC losses – whether this drug was encapsulated alone (i.e. L-N) or together with Dex (i.e., L-ND). There was some loss of Dex (whether encapsulated alone or with NAC), but at a tolerable level, leaving a sufficient level of encapsulated dose for the therapeutic effect. The post-aerosolization encapsulation efficiencies and kinetics of drug efflux, also shown in Table 1, indicate that, for each drug, there was no significant change in the distribution between encapsulated and unencapsulated drug. The nebulization made no significant changes in the half-life of NAC efflux but did change that of Dex from 53 to 29 h. It can be that during the nebulization process Dex's organization within the lipid bilayers has undergone some changes that reduced the energetic cost of its release. Regardless of the molecular processes involved, we argue that this change in Dex efflux is not harmful and could actually be beneficial since in the aerosol the efflux rates of both drugs have become quite similar. In addition, as will be shown in Section 3.3, this change in Dex had no effect on its therapeutic activity. In summary, the liposomal formulations were stable to the aerosolization process, retaining their original pre-aerosolization properties, which allowed us to put the concerns to rest, and proceed to further characterization of the liposomal aerosols and to proof of concept studies in animal models. To further characterize the liposomal aerosols, we simulated their deposition in human airways and lungs, using the ACI device as described in Section 2.6. The results (Fig. 2) show that, as desired for the intended therapy, the liposomes were deposited along the airways and lungs.
The chlorine-exposed mice (4B) are clearly affected — they cease moving and huddle together in the chamber as far as possible from the chlorine inlet, showing signs of stress and rubbing their noses. This behavior was observed only during the exposure to the chlorine. Once this exposure ceased, whether the mice were given air or an aerosol of the liposomal treatment, their behavior became similar to that of the control mice (4A). 3.3.2. Mouse weight changes as a function of time and treatment Weight changes determined for mice exposed to chlorine with and without L-ND treatment, as well as those of the control mice, are shown in Fig. 4. As expected, control mice receiving air only continued their normal growth over the 14 day period. The mice exposed to chlorine alone showed three distinct differences from the control mice: (i) a two-trend response: a drop in weight during the first two days after which the trend reversed to weight gain (ii) a slower rate of weight gain, over the entire period from day = 2 and on and (iii) lower levels of weight gain over the entire period. These differences indicate that the chlorine exposure generated a clinical damage that was, furthermore, long term. As indicated in the Introduction, noninvasive follow-up of mice weight changes post-exposures to chlorine or other agents was a major measure applied to investigate the respiratory damage [11, 13–21]. Most of these studies reported weight loss during the early days post-exposure, as well as the trend reversal to weight gain in those studies that went beyond early days. Given that the experimental designs and conditions were quite different than those of the present study, we find the similarities in trends quite encouraging. The
3.3. Animal model studies 3.3.1. Behavioral responses All mice survived, irrespective of their exposure to air, chlorine, liposomes or combinations of. The behavior of the mice in their cages after the exposures and during the 2–3 weeks of observation was normal, except the weight losses presented and discussed in Section 3.3.2 below. Typical behaviors of the control and chlorine-exposed mice in the inhalation chamber are shown in Fig. 3. The control mice (4A), exposed to air only, are seen to occupy the whole chamber moving around freely.
Fig. 4. Mice weight changes as a function of time and treatment. Each data point is an average of all mice in the treatment group (n = 14, n = 21 and n = 7 for mice receiving air, chlorine, and chlorine + treatment, respectively). Treatment formulation was HA-L encapsulating both NAC and Dex (i.e. L-ND), at the drug doses of 4 mg/ml and 0.12 mg/ml, respectively. The error bars represent the SEM and the lines are nontheoretical drawn to emphasize the trends in the data.
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
6
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx
Fig. 5. Weight changes of chlorine-exposed mice at day = 15 as a function of drug formulations comparing free vs. liposomal drugs. White-bars are free drug and lightgray bars are liposomal drug. A. Mice treated by aerosols of free or liposomal NAC (i.e. LN), each at the dose of 4 mg/ml. B. Mice treated by aerosols of free or liposomal Dex (i.e. L-D), at the respective doses of 1.2 mg/ml and 0.12 mg/ml. C. Mice treated by aerosols of free or liposomal NAC + Dex (i.e. L-ND), at the doses of 4 mg/ml NAC and 1.2 mg/ml Dex. Each bar is an average of all mice in the treatment group (n = 5) and the error bars represent the SEM. The bars were drawn with depth in order to assist visualizing data values close to zero.
limitations invasive measures pose for long-term follow-up have already been discussed in the Introduction section. We furthermore wish to point out that, in studies of long-term weight change followup, that also reported results of invasive measures, the latter were performed at few selected time points (deduced from the weight follow-up) on additional groups of mice [17], or upon termination of the experiment [20]. The mice exposed to chlorine followed by treatment with liposomal aerosol of the formulation containing both drugs (i.e. L-ND) also show the two-trend response (Fig. 4), albeit with three distinct differences compared to the mice receiving chlorine without treatment. The treatment mitigated the weight drop over the first two days. The treatment increased both rate and levels of weight gain, bringing it quite close to that of the control mice – which was the goal of the treatment, as discussed in the Introduction section. Drugs, routes of administration and experimental protocols applied in the field for the treatment of respiratory damage, were quite different than ours, allowing comparisons of trends alone [14,16,27,28]. Yet, we find those results encouraging in support of the approach we
implemented although none of the previous studies applied drugcarrier formulations and few administered the free drugs in aerosol form. IP administration of a free antioxidant or a free prostaglandin mitigated the early weight drop [14,16] and increased the rate of weight gain [14]. Free ascorbate and deferoxamine, administered first IM, then several doses via aerosol reduced mortality and lung injury in chlorine exposed mice [27]. Combination treatment, by IP administration of free dexamethasone phosphate and free NAC, to chlorine exposed mice 1 h post-exposure was evaluated 24 h post-exposure and was concluded to have therapeutic potential [28]. To evaluate the benefits of the liposomal formulations over those of free drugs we compared (for chlorine-exposed mice) treatments with free drugs to those of the corresponding liposomal formulations. The results, for day 15 from exposure and treatment are shown in Fig. 5. Starting with NAC, the data (Fig. 5A) show that treatment with the liposomal NAC was better than the same dose (4 mg/ml) of free NAC. The liposomal Dex (0.12 mg/ml) is quite better than 10 fold higher (1.2 mg/ml) free Dex (Fig. 5B), and a similar trend is seen for the treatment with both drugs — 0.12 mg/ml Dex + 4 mg/ml NAC, free or liposomal (Fig. 5C). We would also like to point out that in all cases the scatter in the treatment with the free drugs is much higher than with the liposomal formulations. Thus, the liposomal formulations show two benefits over the free drugs: higher efficacy, and the potential for lower variability among patients. Results for the three liposomal formulations are shown in Fig. 6, comparing the weight changes of days 2 and 15. This complements the data shown in Fig. 4 for one liposomal treatment. All three drugliposome formulations were efficacious, achieving the goal of reaching the weight changes of the control mice (the slightly higher weight gains of the liposomal formulations compared to the control mice were not statistically significant). There are, however, some differences at the early time (day 2). Both the L-N and the L-ND formulations mitigated the chlorine-induced weight drop on day 2, while the formulation with Dex alone (i.e., L-D) did not, giving the former two a slight advantage over the latter one.
4. Conclusions The results reported here allow drawing the overall conclusion that the novel liposomal formulations, based on the hyaluronan liposomes and encapsulating one or more drugs with anti-oxidant and antiinflammatory activities, have high potential to successfully resolve unmet therapeutic needs due to respiratory damage caused by chlorine exposure. We would like to emphasize that these findings were obtained under the conditions in which the treatment was dispersed (in aerosol form) throughout the chamber. Each mouse received, therefore, the treatment by breathing the atmosphere in the chamber, which was furthermore shared with the other mice in the chamber. This, in our opinion, strengthens the impact of our results, since in a real-life situation a treated individual would receive the treatment personally and directly to his airways. We furthermore postulate that inhalational therapy with these liposomal formulations is not restricted to chlorine-induced respiratory damage. It also has potential for respiratory damage due to exposure to other TICs, as well as to cases of ALI and ARDS caused by other factors. Studies extended towards these goals are underway.
References Fig. 6. The impact of liposomal treatment on chlorine-exposed mice at days 2 and 15. White-bars: day 2, light-gray bars: day 15. Drug and liposome formulations, drug doses, as well as animal numbers/group are similar to those listed in the legend to Fig. 4. The error bars represent the SEM. The bars were drawn with depth in order to assist visualizing data values close to zero. Statistical significance is of control and of each treatment, compared to the group exposed to chlorine alone. **p b 0.005, ***p b 0.001.
[1] A. Eisenkraft, A. Falk, Toxic inhalation injury — management and medical treatment, in: H. Salem, S.A. Katz (Eds.), Inhalation Toxicology, third ed.CRC Press 2014, pp. 357–421. [2] W.C. White, J.C. Martin, Chlorine gas inhalation. Human clinical evidence of toxicity and experience in animal models, Proc. Am. Thorac. Soc. 7 (2010) 257–263. [3] B.J.A. Jugg, A.J. Smith, S.J. Rudall, P. Rice, The injured lung: clinical issues and experimental models, Philos. Trans. R. Soc. B 366 (2011) 306–309.
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx [4] R. Margalit, Biomaterial-based particulate drug carriers, in: D. Peer (Ed.), Handbook of Harnessing Biomaterials in Nanomedicine, Pan Stanford Publishing Pte. Ltd., Singapore 2012, pp. 1–16. [5] Y. Glucksam-Galnoy, T. Zor, R. Margalit, Hyaluronan-modified and regular multilamellar liposomes provide sub-cellular targeting to macrophages, without eliciting a pro-inflammatory response, J. Control. Release 160 (2012) 388–393. [6] F.E. Lennon, P.A. Singleton, Role of hyaluronan and hyaluronan-binding proteins in lung pathobiology, Am. J. Phys. Lung Cell. Mol. Phys. 301 (2011) L137–L147. [7] J.G. Martin, et al., Chlorine-induced injury to the airways in mice, Am. J. Respir. Crit. Care Med. 168 (2003) 568–574. [8] G.W. Hoyle, et al., Deviations from Haber's Law for multiple measures of acute lung injury in chlorine-exposed mice, Toxicol. Sci. 118 (2010) 696–703. [9] X. Tian, et al., Acute lung injury induced by chlorine inhalation in C57BL/6 and FVB/ N mice, Inhal. Toxicol. 20 (2008) 783–893. [10] W. Song, et al., Inhibition of lung fluid clearance and epithelial Na+ channels by chlorine, hypochlorous acid, and chloramines, J.B.C. 285 (2010) (99916-97820). [11] B.V. Patel, M.R. Wilson, M. Takata, Resolution of acute lung injury and inflammation: a translational mouse model, Eur. Respir. J. 39 (2012) 1162–1170. [12] S.G. Zarogiannis, et al., Postexposure aerosolized heparin reduces lung injury in chlorine-exposed mice, Am. J. Lung. Cell. Mol. Physiol. 307 (2014) 1347–1354. [13] C.B. Massa, et al., Acute chlorine gas exposure produces transient inflammation and a progressive alteration in surfactant composition with accompanying mechanical dysfunction, Toxicol. Appl. Pharmacol. 278 (2014) 53–64. [14] S.M. Castro, et al., Antioxidant treatment ameliorates respiratory syncytial virus-induced disease and lung inflammation, Am. J. Respir. Crit. Care Med. 174 (2006) 1361–1369. [15] J.S. LEE, et al., Priming with endotoxin increases acute lung injury in mice by enhancing the severity of lung endothelial injury, Anat. Rec. 294 (2011) 165–172. [16] Z. Guo, et al., Prevention of LPS-Induced acute lung injury in mice by progranulin, Mediat. Inflamm. 9 (2012) 01–10.
7
[17] N. Limjunyawong, W. Mitzner, M.R. Horton, A mouse model of chronic idiopathic pulmonary fibrosis, Phys. Rep. 2 (2014) 1–8. [18] C. Li, et al., IL-17 response mediates acute lung injury induced by the 2009 pandemic influenza A (H1N1) virus, Cell Res. 22 (2012) 528–538. [19] S. Lax, M.R. Wilson, M. Takata, D.R. Thickett, Using a non-invasive assessment of lung injury in a murine model of acute lung injury, BMJ Open Respir. Res. 1 (2014) 1–8. [20] S. Hayashi, et al., Establishment of an animal model for pulmonary fibrosis in mice using Monocrotaline, Toxicol. Phathol. 23 (1995) 63–71. [21] L. Morales-Nebreda, et al., Intratracheal administration of influenza virus is superior to intranasal administration as a model of acute lung injury, J. Virol. Methods 209 (2014) 116–120. [22] D. Peer, R. Margalit, Physicochemical evaluation of a stability-driven approach to drug entrapment in regular and in surface-modified liposomes, Arch. Biochem. Biophys. 383 (2000) 185–190. [23] I. Elron-Gross, Y. Glucksam, R. Margalit, Liposomal dexamethasone–diclofenac combinations for local osteoarthritis treatment, Int. J. Pharm. 376 (2009) 84–91. [24] R. Margalit, et al., Liposomal drug delivery: thermodynamic and chemical kinetic consideration, J. Control. Release 17 (1991) 285–296. [25] S. Fischer, A. Arad, R. Margalit, Liposome-formulated enzymes for organophosphate scavenging: butyrylcholinesterase and demeton-S, Arch. Biochem. Biophys. 434 (2005) 108–115. [26] G.L. Ellman, et al., A new and rapid colorimetric determination of acetylcholinesterase, Biochem. Pharmacol. 7 (1961) 88–95. [27] S.G. Zarogiannis, et al., Ascorbate and deferoxamine administration after chlorine exposure decrease mortality and lung injury in mice, Am. J. Respir. Cell Mol. Biol. 45 (2011) 386–392. [28] E. Wigenstam, B. Koch, A. Bucht, S. Jonasson, N-acetyl cysteine improves the effects of corticosteroids in a mouse model of chlorine-induced acute lung injury, Toxicology 328 (2015) 40–47.
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
Contents lists available at ScienceDirect
Journal of Controlled Release journal homepage: www.elsevier.com/locate/jconrel
Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles Ilia Rivkin a,1, Yifat Galnoy-Glucksam a,1, Inbar Elron-Gross a, Amichay Afriat b, Arik Eisenkraft b,c,d, Rimona Margalit a,⁎ a
Department of Biochemistry and Molecular Biology, The George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv, Israel NBC Protection Division, IMOD, The Hebrew University, Jerusalem, Israel IDF Medical Corps, The Hebrew University, Jerusalem, Israel d The Institute for Research in Military Medicine, The Faculty of Medicine, The Hebrew University, Jerusalem, Israel b c
a r t i c l e
i n f o
Article history: Received 29 December 2015 Received in revised form 28 March 2016 Accepted 28 March 2016 Available online xxxx Keywords: TIC exposure Mass casualty events Aerosols Respiratory damage
a b s t r a c t The purpose of this study was to develop a treatment for respiratory damage caused by exposure to toxic industrial chemicals (TICs), including mass casualty events, by aerosols of dexamethasone and/or N-acetyl cysteine formulated in targeted lipid-based particles. Good encapsulation, performance as slow-release drug depots, conservation of matter, and retention of biological activity were obtained for the three drug-carrier formulations, pre- and post-aerosolization. Weight changes over a 2 week period were applied, deliberately, as a non-invasive clinical parameter. Control mice gained weight continuously, whereas a non-lethal 30 minute exposure of mice to 300 ppm Cl2 in air showed a two-trend response. Weight loss over the first two days, reversing thereafter to weight gain, but at a rate and level significantly slower and smaller than those of the control mice, indicating the chlorine damage was long-term. The weight changes of Cl2-exposed mice given the inhalational treatments also showed the two-trend response, but the weight gain rates and levels were similar to those of the control mice, reaching the weight-gain range of the control mice. Following this proof of concept, studies are now extended to include additional TICs, and biochemical markers of injury and recovery. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Respiratory damage is among the major injuries caused by exposure, whether accidental or deliberate, to toxic industrial chemicals (TICs), such as chlorine, ammonia, hydrogen chlorine, and many others [1–3]. Treatment of such damage – addressing the surge of reactive oxygen species, the pulmonary inflammation and pulmonary edema – is still in the category of an unmet therapeutic need, despite the availability of effective drugs [1–3]. The problem is not with the drugs themselves, but with their delivery. In the available formulations the drugs are free, hence prone to the well-known deficiencies of treatment with free drugs that too often result in poor therapeutic responses, treatment failure and safety limitations [4]. The approach we devised, to turn the situation – including in cases of mass casualty events – from the current unmet, to a met, therapeutic need is composed of three linked elements: (1) to replace treatment with free drugs by drugs formulated in a targeted lipidbased carrier that can, furthermore, perform as a slow-release drug ⁎ Corresponding author at: Department of Biochemistry and Molecular Biology, Tel Aviv University, Tel Aviv 69978, Israel. E-mail address: [email protected] (R. Margalit). 1 Equal contribution.
depot (2) to deliver the drug-carrier formulations in an aerosol form, directly to the airways and lungs, using a clinical portable inhalation device, and (3) to have both formulations and devices in the vehicles of the first responders so that treatment can be initiated at the field and continued thereafter in the patient's home or in a medical facility. The carrier we selected for the task is a specialized multilamellar liposome, surface-modified by hyaluronan anchored covalently to its surface (denoted HA-L). These liposomes were shown to have active targeting to macrophages and can also bind with high affinity to the extra cellular matrix (ECM) [5–6]. The drugs selected were the antioxidant N-Acetyl-cysteine (NAC) and the anti-inflammatory corticosteroid dexamethasone (Dex). A portable nebulizer was the inhalation device. We first formulated the drugs in the liposomes, pursued and optimized physicochemical properties, verified that the drug-carrier formulations were stable to the nebulization process and simulated the deposition of the aerosol in human airways and lungs using an Anderson Cascade Impactor (ACI). We next conducted the first step in feasibility studies, in mice exposed to a chlorine–air mixture and treated with an aerosol of our novel drug-carrier formulations. The respiratory damage in mice, exposed to chlorine–air mixtures either by whole body or nose only, has been well established [7–13]. Similar respiratory damage in mice,
http://dx.doi.org/10.1016/j.jconrel.2016.03.039 0168-3659/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
2
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx
caused by a host of other agents, was also reported [17–21]. Both invasive and non-invasive approaches have been applied to evaluate the respiratory damage and the impact of therapy. In most of those cases, where invasive approaches were applied, the experiments were for short durations, typically 1–3 days from insult. The non-invasive approach, following mice weight changes, has been applied quite extensively, mostly for periods of 3–21 days [18–21]. Defining, moreover, weight loss as a traditional indicator of an acute pulmonary bleomycin insult, Limjunyawong et al. followed mice weight changes even up to 200 days [17]. Given that our objective was to follow-up each animal for a relatively prolonged period (2–3 weeks) from exposure with/without treatment we opted for this first step in feasibility studies, to apply a noninvasive approach, such as continuous monitoring of animal weight [11,13–21]. Our goal was to have the weight changes of the chlorineexposed and treated mice on a par with those of control mice (exposed to air alone).
2.2.2. Drug encapsulation 2.2.2.1. N-acetyl cysteine (NAC). Buffer-free, salt-free lyophilized HA-L powders (see Section 2.2.1) were brought to room temperature and rehydrated back to original pre-lyophilization volume by the NAC stock solution (20 mg NAC/ml in PBS containing 12.5 mM EDTA at the final pH of 7.5) incubating the system at 37 °C for 24 h. 2.2.2.2. Dexamethasone (Dex). Dex-encapsulating HA-L were prepared essentially as described in Section 2.2.1 above, except the drug was added to the ethanolic lipid solution at the concentration range of 5– 10 mg Dex/ml. 2.2.2.3. Co-encapsulation of Dex and NAC in the same liposome. Dexencapsulating HA-L were prepared as in Section 2.2.2.2, and lyophilized from an aqueous suspension that was buffer-free and salt-free. These lyophilized liposomes were rehydrated in the NAC solution, as described in Section 2.2.2.1, above.
2. Materials and methods 2.3. Kinetics of drug efflux and determination of encapsulation efficiency 2.1. Materials Phospholipon 90G (high purity Soybean phosphatidylcholine (SPC)) was a kind gift from Nattermann Phospholipid GmbH (Cologne Germany). Dexamethasone, NAC, dipalmitoylphosphatidylethanolamine (DPPE), cholesterol (CH) and EDC (ethyl-dimethyl-aminopropylcarbodiimide) were from Sigma Chemical Co. (St. Louis, USA). Hyaluronan (HA) 1.5 MDa used for the liposomes was a kind gift from Genzyme (Cambridge, MA, USA). The chlorine gas was in a mini-cylinder (Portacyl®, Specialty Gases of America Inc., The American Gas Group, purchased in Israel from Maxima Company) 300 ppm chlorine balance air gas mixture. Total gas quantity per cylinder was 58 l, and the rate of gas flow was 0.5 l/min supplied via series fixed flow regulator 70 (Ashcroft®, inlet: 1000 PSIG, flow: 0.5 SLPM, purchased in Israel from Maxima Company). Liquid Scintillation cocktail, Ultima Gold™, was from PerkinElmer Life and Analytical Sciences Inc. (USA). Dialysis tubing (molecular weight cutoff of 12,000–14,000) was from Spectrum Medical Industries (Los Angeles, CA). All other reagents were of analytical grade. Ultracentrifugation was performed with a Sorval Discovery M120 SE micro ultracentrifuge (TN, USA). Lyophilization was performed with a HETO Drywinner 3 (Alleraod, Denmark). The nebulizer was DeVILBISS's PulmoAide Compact Compressor 3655. The Anderson Cascade Impactor was from ThermoFisher Scientific (Franklin, MA, USA). 2.2. Preparation of drug-free and of drug-encapsulating HA-L 2.2.1. Drug-free HA-L The lipid composition was SPC:DPPE:CH 75:5:20 (mole ratios) and the total lipid concentration was 100 mg/ml. Liposome preparation was essentially as described under [21,23] (and the references within). The first step was preparation of regular multilamellar liposomes (RL). The lipids were dissolved in ethanol, and evaporated to obtain a dry lipid film in a rotary evaporator under reduced pressure. The swelling solution (0.1 M borate buffer at pH 9) was added to the lipid film and the system was incubated (in a shaker bath) for 2 h at 65 °C. To obtain HA-L, HA was dissolved in acetate buffer (0.1 M, pH 4.5) at the concentration of 2 mg/ml. It was pre-activated by incubation with EDC (20 mg per 1 mg HA) for 2 h at 37 °C and added to the RL suspension, at the ratio of 1:1 (v/v). This reaction mixture was incubated in a shaker bath for 24 h at 37 °C. The HA-L were freed from excess materials and byproducts by centrifugation for 30 min at 4 °C and a g force of 160,850 followed by several successive washes and re-centrifugations in 0.1 M NH4HCO3, suspending the final pellets in this salt solution. Aliquots of 1 ml of these liposomes were frozen for 2 h at −80 °C, followed by lyophilization. The resultant liposome powders were stored at − 18 °C until further use.
2.3.1. NAC/HA-L Kinetics of drug efflux from the liposomes was studied according to our previously-developed experimental set up and data processing [21–24] (and the references within). Briefly, a suspension of drugencapsulating liposomes (see Section 2.2.2.1 above) was placed in a dialysis sac that was immersed in a continuously-stirred receiver vessel containing drug-free buffer (PBS, 12.5 mM EDTA, pH 7.5), receiver to liposome volume ratio was 15:1. At designated time points, the dialysis sac was transferred from one receiver vessel to another containing fresh drug-free buffer. NAC concentration was determined in each dialysate and in the sac (at the beginning and end of the run). The data were analyzed according to a previously derived multi-pool kinetic model, expressed in Eq. (1) below. ft ¼
n X f i 1− exp−ki t
ð1Þ
i¼1
where ft is the fraction of the total drug in the liposomal system that diffused out of the sac at time t, n is the number of independent drug pools in the liposomal system, fi is the fraction of the total drug in the system that occupied the ith pool at time = 0, ki is the rate constant for drug efflux from the ith pool, and t is time, the free variable. Data processing was done by computer-aided non-linear regression analysis using the KaleidaGraph software. These kinetic experiments also yield the efficiency of drug encapsulation (fi, for the encapsulateddrug pool, at time = 0) which is defined as the ratio of liposomeencapsulated drug to the total drug in the system. 2.3.2. Dex/HA-L All methods were essentially as listed above for NAC, with the following exceptions. The lyophilized dry powder of Dex/HA-L (see Section 2.2.2.2 above) was rehydrated back to original volume with drug-free water or buffer. The dialysis and data processing were performed as in Section 2.3.1 above. 2.3.3. NAC + DEX/H-L The dialysis and data processing were performed as in Section 2.3.1 above, except that the liposomal formulation in the dialysis sac was as detailed in Section 2.2.2.3 above. 2.4. Drug assays in solution and in liposomal samples Dex was assayed in free and in liposomal formulations as previously described, by inclusion of trace 3H-dexamethasone in all Dexcontaining systems [23]. NAC was assayed by a colorimetric assay
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx
3
Scheme 1. The inhalation chamber.
based on the Ellman reaction for determination of thiols [25].We have previously adapted this assay for liposomal samples and 96-well plates [26]. 2.5. Making and characterizing drug-liposome aerosols 2.5.1. Nebulizer set up and operation The nebulizer set up developed for aerosol collection and the optimized experimental conditions are listed together with the relevant results (Section 3.2). The devise was operated according to the following manufacturer's specifications: Compressor free air flow ≥ 8 l/pm, nebulization rate ≥ 0.15 ml/min, supplied nebulizer operating pressure 0.65 bar, and supplied nebulizer liter flow 5.5 l/min. 2.5.2. Physicochemical characterization of the aerosolized liposomes The following parameters were measured to assess quantitative, post nebulization, recovery: (i) the volume of sample put into the nebulization cup, and the volume of the collected aerosol; (ii) lipid and drug concentrations for each sample, pre- and post-nebulization. Dex and NAC were assayed as described in Section 2.4 above, and the lipid was assayed in drug-free liposomes by inclusion of trace 3H-cholesterol in the formulations. To evaluate the impact of nebulization on system properties, test samples were assayed - before and after nebulization for drug encapsulation efficiency and drug efflux kinetics, as described in Section 2.3 above. 2.6. Simulating lung deposition, using the Anderson Cascade Impactor (ACI) The nebulizer was connected to the input of the device, and a water trap was connected to the device output. The nebulizer cup was filled with 7 ml of the test material, and the nebulizer was operated as described in Section 2.5.1 above. At the end of the run, the nebulizer was disconnected from the ACI, the ACI disassembled, the material collected from each device jet stage and assayed for lipid and each drug, as detailed in Section 2.5.2 above.
2.7. The animal model studies 2.7.1. The inhalation chamber The inhalation chamber, shown in Scheme 1, was manufactured in the Tel Aviv University machine shop according to our specifications, and can accommodate up to 10 mice. It is made of Plexiglas, and its dimensions are 13 ∗ 13 ∗ 24 cm. The complete system (i.e., the inhalation chamber(s), the nebulizer and the chlorine/air cylinder) was placed in a chemical hood, within the animal facility of Tel Aviv University.
2.7.2. The experimental design and protocol Male Balb/C mice 10–12 weeks old, average weight at day = 0 in the range of 20 g were from the animal facility of TAU. The Tel Aviv University Institutional Animal Care and Use Committee approved all animal procedures according to the guidelines of the Office of Laboratory Animal Welfare (permit # L-12049). Animals were group-housed (7 per cage) in solid-bottomed plastic cages designed to allow easy access to standard laboratory food and water ad lib. The mice were kept in a 12:12 light–dark cycle in a controlled temperature chamber (24 ± 1 °C). Typically, 5 mice were put into the inhalation chamber to receive the following treatments and there were usually 2–3 repeats (see legends to figures for specific animal numbers/group): (1) Control group: mice Table 1 Physicochemical properties of NAC + Dexamethasone formulated in HA-L, pre- and postnebulization. Drug
Encapsulation and efflux of encapsulated drug % Encapsulation
k2 ∗ 1000 (h−1)
Half-life (h)
NAC Pre-nebulization Post-nebulization
31 (±2) 26 (±2)
0.020 (±0.005) 0.022 (±0.005)
35 32
DEX Pre-nebulization Post-nebulization
93 (±1) 91 (±2)
0.013 (±0.0005) 0.024 (±0.0009)
53 29
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
4
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx
2.7.3. Data processing and statistics The % change in weight (from the initial weight) was calculated for each mouse at each experimental run, for each weighing day. The % change in weight was then averaged for each group (i.e. mice receiving air, chlorine or chlorine and treatment) for each day, and the relevant SEM was calculated.
3. Results and discussion 3.1. Physicochemical characterizations of the drug-liposome formulations Fig. 1. Lipid and drug recoveries post-aerosolization. L, L-D, L-N and L-ND denote drug-free (“empty”) liposomes, dexamethasone-encapsulating liposomes, NAC-encapsulating liposomes and dexamethasone + NAC encapsulated in the same liposome, respectively. Solid white bar: lipids' recovery (liposomes, L) light-gray bars: dexamethasone recovery in L-D (left bar) and in L-ND (right bar) dark-gray bars: NAC recovery in L-N (left bar) and in L-ND (right bar). Each bar is the average of at least three determinations, and the error bars represent the standard deviations. The bars were drawn with depth to be consistent with Figs. 5 and 6, where the depth assists in visualizing data values close to zero.
receiving air only (through the nebulizer, as illustrated in Scheme 1), for 30 to 50 min; (2) chlorine-exposed un-treated mice: mice receiving the chlorine–air mixture (300 ppm chlorine in air) for 30 min; (3) chlorineexposed then treated mice: mice exposed to chlorine as in group 2 above, immediately followed by exposure to the aerosol of a selected treatment (given through the nebulizer). Treatments were selected from free and liposomal Dex, NAC or a combination of both drugs. Further details of the treatments including doses are provided with the results. Each mouse was tagged, and was weighed before being put into the inhalation chamber, and thereafter every 2–3 days for a period of 2–3 weeks. The mice were monitored throughout the run period for survival and behavior.
Three drug-carrier formulations were made and characterized. Two were single drug formulations, namely Dex in HA-L, and NAC in HA-L, denoted L-D and L-N, respectively. The third formulation, denoted LND, contained both drugs in the same liposome, the lipophilic Dex occupying the lipid regions of the liposome and the hydrophilic NAC occupying the aqueous regions of the liposome, especially in the inner aqueous core. Determination of the encapsulation efficiency and kinetics of drug efflux was performed for each formulation, as described under Section 2.3, and the data were processed according to Eq. (1). For all systems the data fit two initial drug pools, one encapsulated and the other the excess unencapsulated drug. Typical results, for the L-ND system are listed in Table 1. Looking, at this stage, at the pre-nebulization data listed in Table 1, the encapsulated doses for each of the drugs were at therapeutic levels, with high and moderate encapsulation efficiencies obtained for Dex and NAC, respectively. The half-lives of drug efflux that range from 35 to 50 h, are clear indications that this system performs as the desired slow-release drug depot for each of the drugs. Results for the other two formulations, L-D and L-N were quite similar.
Fig. 2. Simulated deposition of the hyaluronan liposomes (HA-L) in human airways and lungs, using the Anderson Cascade Impactor (ACI) device. Right-hand side: Scheme of the corresponding aerosol-particle size ranges (μm) and localization in human airways and lungs. Left-hand side: the liposome deposition. Each bar is an average of 3 independent experiments and the error bars represent the standard deviations.
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx
5
Fig. 3. Mice in the inhalation chamber — behavioral responses as function of exposure type. A. Control mice exposed to air only. B. Mice exposed to the chlorine/air gas mixture (300 ppm chlorine in air).
3.2. The liposomal aerosols and retention of pre-aerosolization properties Nebulization is not a gentle process, which raised the concern that the aerosolized liposomes may be damaged, losing material (lipids and/or drugs), losing encapsulated matter (i.e., encapsulated drug moving out of the liposome) or losing the slow-release nature. To evaluate whether such damages occurred and if so, their extent, each liposome formulation was aerosolized and the aerosol collected as described in Section 2.5. Typically 7 ml of the desired liposome formulation was put into the nebulization cup, 80% of which were recovered. Material recovery (Fig. 1) was quite satisfactory. There were no lipid losses, nor NAC losses – whether this drug was encapsulated alone (i.e. L-N) or together with Dex (i.e., L-ND). There was some loss of Dex (whether encapsulated alone or with NAC), but at a tolerable level, leaving a sufficient level of encapsulated dose for the therapeutic effect. The post-aerosolization encapsulation efficiencies and kinetics of drug efflux, also shown in Table 1, indicate that, for each drug, there was no significant change in the distribution between encapsulated and unencapsulated drug. The nebulization made no significant changes in the half-life of NAC efflux but did change that of Dex from 53 to 29 h. It can be that during the nebulization process Dex's organization within the lipid bilayers has undergone some changes that reduced the energetic cost of its release. Regardless of the molecular processes involved, we argue that this change in Dex efflux is not harmful and could actually be beneficial since in the aerosol the efflux rates of both drugs have become quite similar. In addition, as will be shown in Section 3.3, this change in Dex had no effect on its therapeutic activity. In summary, the liposomal formulations were stable to the aerosolization process, retaining their original pre-aerosolization properties, which allowed us to put the concerns to rest, and proceed to further characterization of the liposomal aerosols and to proof of concept studies in animal models. To further characterize the liposomal aerosols, we simulated their deposition in human airways and lungs, using the ACI device as described in Section 2.6. The results (Fig. 2) show that, as desired for the intended therapy, the liposomes were deposited along the airways and lungs.
The chlorine-exposed mice (4B) are clearly affected — they cease moving and huddle together in the chamber as far as possible from the chlorine inlet, showing signs of stress and rubbing their noses. This behavior was observed only during the exposure to the chlorine. Once this exposure ceased, whether the mice were given air or an aerosol of the liposomal treatment, their behavior became similar to that of the control mice (4A). 3.3.2. Mouse weight changes as a function of time and treatment Weight changes determined for mice exposed to chlorine with and without L-ND treatment, as well as those of the control mice, are shown in Fig. 4. As expected, control mice receiving air only continued their normal growth over the 14 day period. The mice exposed to chlorine alone showed three distinct differences from the control mice: (i) a two-trend response: a drop in weight during the first two days after which the trend reversed to weight gain (ii) a slower rate of weight gain, over the entire period from day = 2 and on and (iii) lower levels of weight gain over the entire period. These differences indicate that the chlorine exposure generated a clinical damage that was, furthermore, long term. As indicated in the Introduction, noninvasive follow-up of mice weight changes post-exposures to chlorine or other agents was a major measure applied to investigate the respiratory damage [11, 13–21]. Most of these studies reported weight loss during the early days post-exposure, as well as the trend reversal to weight gain in those studies that went beyond early days. Given that the experimental designs and conditions were quite different than those of the present study, we find the similarities in trends quite encouraging. The
3.3. Animal model studies 3.3.1. Behavioral responses All mice survived, irrespective of their exposure to air, chlorine, liposomes or combinations of. The behavior of the mice in their cages after the exposures and during the 2–3 weeks of observation was normal, except the weight losses presented and discussed in Section 3.3.2 below. Typical behaviors of the control and chlorine-exposed mice in the inhalation chamber are shown in Fig. 3. The control mice (4A), exposed to air only, are seen to occupy the whole chamber moving around freely.
Fig. 4. Mice weight changes as a function of time and treatment. Each data point is an average of all mice in the treatment group (n = 14, n = 21 and n = 7 for mice receiving air, chlorine, and chlorine + treatment, respectively). Treatment formulation was HA-L encapsulating both NAC and Dex (i.e. L-ND), at the drug doses of 4 mg/ml and 0.12 mg/ml, respectively. The error bars represent the SEM and the lines are nontheoretical drawn to emphasize the trends in the data.
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
6
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx
Fig. 5. Weight changes of chlorine-exposed mice at day = 15 as a function of drug formulations comparing free vs. liposomal drugs. White-bars are free drug and lightgray bars are liposomal drug. A. Mice treated by aerosols of free or liposomal NAC (i.e. LN), each at the dose of 4 mg/ml. B. Mice treated by aerosols of free or liposomal Dex (i.e. L-D), at the respective doses of 1.2 mg/ml and 0.12 mg/ml. C. Mice treated by aerosols of free or liposomal NAC + Dex (i.e. L-ND), at the doses of 4 mg/ml NAC and 1.2 mg/ml Dex. Each bar is an average of all mice in the treatment group (n = 5) and the error bars represent the SEM. The bars were drawn with depth in order to assist visualizing data values close to zero.
limitations invasive measures pose for long-term follow-up have already been discussed in the Introduction section. We furthermore wish to point out that, in studies of long-term weight change followup, that also reported results of invasive measures, the latter were performed at few selected time points (deduced from the weight follow-up) on additional groups of mice [17], or upon termination of the experiment [20]. The mice exposed to chlorine followed by treatment with liposomal aerosol of the formulation containing both drugs (i.e. L-ND) also show the two-trend response (Fig. 4), albeit with three distinct differences compared to the mice receiving chlorine without treatment. The treatment mitigated the weight drop over the first two days. The treatment increased both rate and levels of weight gain, bringing it quite close to that of the control mice – which was the goal of the treatment, as discussed in the Introduction section. Drugs, routes of administration and experimental protocols applied in the field for the treatment of respiratory damage, were quite different than ours, allowing comparisons of trends alone [14,16,27,28]. Yet, we find those results encouraging in support of the approach we
implemented although none of the previous studies applied drugcarrier formulations and few administered the free drugs in aerosol form. IP administration of a free antioxidant or a free prostaglandin mitigated the early weight drop [14,16] and increased the rate of weight gain [14]. Free ascorbate and deferoxamine, administered first IM, then several doses via aerosol reduced mortality and lung injury in chlorine exposed mice [27]. Combination treatment, by IP administration of free dexamethasone phosphate and free NAC, to chlorine exposed mice 1 h post-exposure was evaluated 24 h post-exposure and was concluded to have therapeutic potential [28]. To evaluate the benefits of the liposomal formulations over those of free drugs we compared (for chlorine-exposed mice) treatments with free drugs to those of the corresponding liposomal formulations. The results, for day 15 from exposure and treatment are shown in Fig. 5. Starting with NAC, the data (Fig. 5A) show that treatment with the liposomal NAC was better than the same dose (4 mg/ml) of free NAC. The liposomal Dex (0.12 mg/ml) is quite better than 10 fold higher (1.2 mg/ml) free Dex (Fig. 5B), and a similar trend is seen for the treatment with both drugs — 0.12 mg/ml Dex + 4 mg/ml NAC, free or liposomal (Fig. 5C). We would also like to point out that in all cases the scatter in the treatment with the free drugs is much higher than with the liposomal formulations. Thus, the liposomal formulations show two benefits over the free drugs: higher efficacy, and the potential for lower variability among patients. Results for the three liposomal formulations are shown in Fig. 6, comparing the weight changes of days 2 and 15. This complements the data shown in Fig. 4 for one liposomal treatment. All three drugliposome formulations were efficacious, achieving the goal of reaching the weight changes of the control mice (the slightly higher weight gains of the liposomal formulations compared to the control mice were not statistically significant). There are, however, some differences at the early time (day 2). Both the L-N and the L-ND formulations mitigated the chlorine-induced weight drop on day 2, while the formulation with Dex alone (i.e., L-D) did not, giving the former two a slight advantage over the latter one.
4. Conclusions The results reported here allow drawing the overall conclusion that the novel liposomal formulations, based on the hyaluronan liposomes and encapsulating one or more drugs with anti-oxidant and antiinflammatory activities, have high potential to successfully resolve unmet therapeutic needs due to respiratory damage caused by chlorine exposure. We would like to emphasize that these findings were obtained under the conditions in which the treatment was dispersed (in aerosol form) throughout the chamber. Each mouse received, therefore, the treatment by breathing the atmosphere in the chamber, which was furthermore shared with the other mice in the chamber. This, in our opinion, strengthens the impact of our results, since in a real-life situation a treated individual would receive the treatment personally and directly to his airways. We furthermore postulate that inhalational therapy with these liposomal formulations is not restricted to chlorine-induced respiratory damage. It also has potential for respiratory damage due to exposure to other TICs, as well as to cases of ALI and ARDS caused by other factors. Studies extended towards these goals are underway.
References Fig. 6. The impact of liposomal treatment on chlorine-exposed mice at days 2 and 15. White-bars: day 2, light-gray bars: day 15. Drug and liposome formulations, drug doses, as well as animal numbers/group are similar to those listed in the legend to Fig. 4. The error bars represent the SEM. The bars were drawn with depth in order to assist visualizing data values close to zero. Statistical significance is of control and of each treatment, compared to the group exposed to chlorine alone. **p b 0.005, ***p b 0.001.
[1] A. Eisenkraft, A. Falk, Toxic inhalation injury — management and medical treatment, in: H. Salem, S.A. Katz (Eds.), Inhalation Toxicology, third ed.CRC Press 2014, pp. 357–421. [2] W.C. White, J.C. Martin, Chlorine gas inhalation. Human clinical evidence of toxicity and experience in animal models, Proc. Am. Thorac. Soc. 7 (2010) 257–263. [3] B.J.A. Jugg, A.J. Smith, S.J. Rudall, P. Rice, The injured lung: clinical issues and experimental models, Philos. Trans. R. Soc. B 366 (2011) 306–309.
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039
I. Rivkin et al. / Journal of Controlled Release xxx (2015) xxx–xxx [4] R. Margalit, Biomaterial-based particulate drug carriers, in: D. Peer (Ed.), Handbook of Harnessing Biomaterials in Nanomedicine, Pan Stanford Publishing Pte. Ltd., Singapore 2012, pp. 1–16. [5] Y. Glucksam-Galnoy, T. Zor, R. Margalit, Hyaluronan-modified and regular multilamellar liposomes provide sub-cellular targeting to macrophages, without eliciting a pro-inflammatory response, J. Control. Release 160 (2012) 388–393. [6] F.E. Lennon, P.A. Singleton, Role of hyaluronan and hyaluronan-binding proteins in lung pathobiology, Am. J. Phys. Lung Cell. Mol. Phys. 301 (2011) L137–L147. [7] J.G. Martin, et al., Chlorine-induced injury to the airways in mice, Am. J. Respir. Crit. Care Med. 168 (2003) 568–574. [8] G.W. Hoyle, et al., Deviations from Haber's Law for multiple measures of acute lung injury in chlorine-exposed mice, Toxicol. Sci. 118 (2010) 696–703. [9] X. Tian, et al., Acute lung injury induced by chlorine inhalation in C57BL/6 and FVB/ N mice, Inhal. Toxicol. 20 (2008) 783–893. [10] W. Song, et al., Inhibition of lung fluid clearance and epithelial Na+ channels by chlorine, hypochlorous acid, and chloramines, J.B.C. 285 (2010) (99916-97820). [11] B.V. Patel, M.R. Wilson, M. Takata, Resolution of acute lung injury and inflammation: a translational mouse model, Eur. Respir. J. 39 (2012) 1162–1170. [12] S.G. Zarogiannis, et al., Postexposure aerosolized heparin reduces lung injury in chlorine-exposed mice, Am. J. Lung. Cell. Mol. Physiol. 307 (2014) 1347–1354. [13] C.B. Massa, et al., Acute chlorine gas exposure produces transient inflammation and a progressive alteration in surfactant composition with accompanying mechanical dysfunction, Toxicol. Appl. Pharmacol. 278 (2014) 53–64. [14] S.M. Castro, et al., Antioxidant treatment ameliorates respiratory syncytial virus-induced disease and lung inflammation, Am. J. Respir. Crit. Care Med. 174 (2006) 1361–1369. [15] J.S. LEE, et al., Priming with endotoxin increases acute lung injury in mice by enhancing the severity of lung endothelial injury, Anat. Rec. 294 (2011) 165–172. [16] Z. Guo, et al., Prevention of LPS-Induced acute lung injury in mice by progranulin, Mediat. Inflamm. 9 (2012) 01–10.
7
[17] N. Limjunyawong, W. Mitzner, M.R. Horton, A mouse model of chronic idiopathic pulmonary fibrosis, Phys. Rep. 2 (2014) 1–8. [18] C. Li, et al., IL-17 response mediates acute lung injury induced by the 2009 pandemic influenza A (H1N1) virus, Cell Res. 22 (2012) 528–538. [19] S. Lax, M.R. Wilson, M. Takata, D.R. Thickett, Using a non-invasive assessment of lung injury in a murine model of acute lung injury, BMJ Open Respir. Res. 1 (2014) 1–8. [20] S. Hayashi, et al., Establishment of an animal model for pulmonary fibrosis in mice using Monocrotaline, Toxicol. Phathol. 23 (1995) 63–71. [21] L. Morales-Nebreda, et al., Intratracheal administration of influenza virus is superior to intranasal administration as a model of acute lung injury, J. Virol. Methods 209 (2014) 116–120. [22] D. Peer, R. Margalit, Physicochemical evaluation of a stability-driven approach to drug entrapment in regular and in surface-modified liposomes, Arch. Biochem. Biophys. 383 (2000) 185–190. [23] I. Elron-Gross, Y. Glucksam, R. Margalit, Liposomal dexamethasone–diclofenac combinations for local osteoarthritis treatment, Int. J. Pharm. 376 (2009) 84–91. [24] R. Margalit, et al., Liposomal drug delivery: thermodynamic and chemical kinetic consideration, J. Control. Release 17 (1991) 285–296. [25] S. Fischer, A. Arad, R. Margalit, Liposome-formulated enzymes for organophosphate scavenging: butyrylcholinesterase and demeton-S, Arch. Biochem. Biophys. 434 (2005) 108–115. [26] G.L. Ellman, et al., A new and rapid colorimetric determination of acetylcholinesterase, Biochem. Pharmacol. 7 (1961) 88–95. [27] S.G. Zarogiannis, et al., Ascorbate and deferoxamine administration after chlorine exposure decrease mortality and lung injury in mice, Am. J. Respir. Cell Mol. Biol. 45 (2011) 386–392. [28] E. Wigenstam, B. Koch, A. Bucht, S. Jonasson, N-acetyl cysteine improves the effects of corticosteroids in a mouse model of chlorine-induced acute lung injury, Toxicology 328 (2015) 40–47.
Please cite this article as: I. Rivkin, et al., Treatment of respiratory damage in mice by aerosols of drug-encapsulating targeted lipid-based particles, J. Control. Release (2015), http://dx.doi.org/10.1016/j.jconrel.2016.03.039