Sustained Release Strategy Designed for Lixisenatide Delivery to

Subscriber access provided by UNIV OF SOUTHERN INDIANA

Biological and Medical Applications of Materials and Interfaces

Sustained Release Strategy Designed for Lixisenatide Delivery to Synchronously Treat Diabetes and Associated Complications Yaping Zhuang, Xiaowei Yang, Yamin Li, Yipei Chen, Xiaochun Peng, Lin Yu, and Jiandong Ding ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b10346 • Publication Date (Web): 30 Jul 2019 Downloaded from pubs.acs.org on July 31, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Sustained Release Strategy Designed for Lixisenatide Delivery to Synchronously Treat Diabetes and Associated Complications

Yaping Zhuang,† Xiaowei Yang,

†

Yamin Li,‡ Yipei Chen, †

†State

†

Xiaochun Peng,

‡

Lin Yu,*

Jiandong Ding†

Key Laboratory of Molecular Engineering of Polymers, Department of

Macromolecular Science, Fudan University, Shanghai, 200438, China. ‡Department

of Orthopaedic Surgery, Shanghai Jiaotong University Affiliated Sixth People's Hospital, Shanghai, 200233, China.

KEYWORDS: Lixisenatide; injectable hydrogel; type 2 diabetes mellitus; blood sugar control; diabetic complication

1

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT: Diabetes and its complications have become a global challenge of public health. Herein, we aimed to develop a long-acting delivery system of lixisenatide (Lixi), a glucose-dependent antidiabetic peptide, based on an injectable hydrogel for the synchronous treatment of type 2 diabetic mellitus (T2DM) and associated complications. Two triblock copolymers, poly(ε-caprolactone-co-glycolic acid)-poly(ethylene glycol)poly(ε-caprolactone-co-glycolic acid) (PCGA-PEG-PCGA) and poly(D, L-lactic acid-coglycolic acid)-poly(ethylene glycol)-poly(D, L-lactic acid-co-glycolic acid) (PLGA-PEGPLGA) possessing temperature-induced sol-gel transitions, were synthesized by us. Compared to the two single-component hydrogels, their 1/1 mixture hydrogel not only maintained the temperature-induced gelation but also exhibited a steadier degradation profile in vivo. Both in vitro and in vivo release studies demonstrated that the mixture hydrogel provided the sustained release of Lixi for up to 9 days, which was attributed to balanced electrostatic interactions between the positive charges in the peptide and the negative charges in the polymer carrier. The hypoglycemic efficacy of Lixi delivered from the mixture hydrogel after a single subcutaneous injection into diabetic db/db mice was comparable to that of twice-daily administrations of Lixi solution for up to 9 days. Furthermore, three successive administrations of the abovementioned gel system within a month significantly increased the plasma insulin level, lowered glycosylated hemoglobin and improved the pancreatic function of the animals. These results were superior or equivalent to those of twice-daily injections of Lixi solution for 30 days, but the number of injections was markedly reduced from 60 to 3. Finally, an improvement in hyperlipidemia,

2


Page 2 of 64

Page 3 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


augmentation of nerve fiber density and enhancement of motor nerve conduction velocity in the gel formulation-treated db/db mice indicated that the sustained delivery of Lixi arrested and even ameliorated diabetic complications. These findings suggested that the Lixi-loaded mixture hydrogel has great potential for the treatment of T2DM with significant improvements in the health and quality of life of patients.

3



Page 4 of 64

1. INTRODUCTION Diabetes mellitus, which is characterized by chronic hyperglycemia that results from relative insulin insufficiency and/or insulin resistance, has become a global epidemic, and the incidence of diabetes mellitus has more than doubled globally in the past three decades.1 According to an International Diabetes Federation (IDF) estimation, there are 451 million diabetics worldwide in 2017, and type 2 diabetes mellitus (T2DM) accounts for over 90% of diagnosed cases.2-4 In particular, China has emerged as the global diabetes epicenter, with nearly 114 million adults living with diabetes, which equates to 11.6% of the total adult population.3 Diabetes not only places a substantial economic burden on individuals but also consumes an enormous amount of public health resources.4 Meanwhile, persistent hyperglycemia, which is a hallmark of all forms of diabetes, can cause a variety of complications, including microvascular damage, cardiovascular disease, retinopathy, neuropathy, amputations and kidney failure.3,

5

Most diabetics

develop at least one complication, and cardiovascular complications are the greatest cause of diabetic mortality.3 To effectively manage glucose, slow the progression of diabetes and minimize the risk of diabetic complications, new and comprehensive therapies are constantly being investigated.6-11 Nevertheless, few of therapies are able to achieve the simultaneous treatment of diabetes and the associated complications. Over the past decade, incretinbased therapies using glucagon-like peptide-1 (GLP-1) receptor agonists have opened up a new avenue for the treatment of diabetes.12-17 Compared with conventional 4


Page 5 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


antidiabetic drugs, such as metformin and insulin, the insulinotropic effects of GLP-1 receptor agonists are glucose-dependent, providing a minimal risk of hypoglycemia.13, 18 Meanwhile, clinical trials have demonstrated the administration of GLP-1 receptor agonists is beneficial for attenuating the risk of cardiovascular complications of patients with T2DM.13, 19 Lixisenatide (Lixi), which is manufactured by Sanofi Aventis, was licensed by the U.S. Food and Drug Administration (FDA) in July 2016 for the treatment of T2DM and became the sixth licensed drug in the family of GLP-1 receptor agonists.13, 19 Lixi is a synthetic analog of exendin-4 that contains an addition of 6 alkaline lysine residues at the Cterminus and a deletion of a proline residue. These structural modifications markedly improve the stability of Lixi and increase its binding affinity for the GLP-1 receptor.19-20 Similar to other GLP-1 receptor agonists, Lixi can stimulate glucose-dependent insulin release, decrease glucagon secretion and delay gastric emptying.19-20 Although the halflife of Lixi is only 3 h, its increased binding affinity combined with the ability to slow gastric emptying allows it to be administered once per day.19 As a chronic disease, treatment of diabetes accompanies the rest lifetime of a patient and patient adherence with devised treatment regimens is crucial to achieve the desired therapeutic efficacy. Apparently, the frequent daily injection does not satisfy patient compliance. Consequently, the development of long-acting delivery systems of Lixi to realize the synchronous treatment of diabetes and associated complications and significantly reduce the frequency of administration is very valuable and meaningful for

5



the amelioration of the health and quality of life of patients with T2DM. However, as far as we know, no such long-acting delivery system of Lixi has been reported thus far. Recently, injectable in situ forming hydrogels have acquired extensive attention because of simple fabrication and nonsurgical implantation procedures.21-26 In particular, injectable and thermosensitive hydrogels that are aqueous polymer solutions with low viscosities at low or room temperature and exhibit sol-gel transitions upon heating have been employed as one of the promising implantable systems for minimally invasive drug delivery and injectable tissue engineering.27-34 Therapeutic agents or cells can be loaded by simple blending with the aqueous polymer solution at a low temperature, and this type of loading process offers a very high loading efficiency.35-36 The drug-containing gel depot is spontaneously formed after the injection of the aqueous polymer solution into a target area, typically in the subcutaneous layer of a warm-blooded mammal, by triggering a temperature-responsive sol-gel transition at body temperature. To date, many thermosensitive and biodegradable polymer hydrogel systems have been opened up, including copolymers of poly(ethylene glycol) (PEG)/polyester,30, 37-40 PEG/polypeptide,28, 41-42

and poly(phosphazenes).43-44 Among these materials, thermosensitive hydrogels

composed of PEG/polyester copolymers are well received for various biomedical applications due to convenient one-pot synthesis and tunable gelation properties.37, 45-46 Meanwhile, their in vitro cytocompatibility and in vivo biocompatibility have been well verified.36-37, 45-46Additionally, PEG and many polyesters have been approved by FDA and utilized in clinic for many years.

6


Page 6 of 64

Page 7 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Desirable sustained delivery of hydrophobic small-molecule drugs has been achieved using thermosensitive hydrogels.27, 47-48 However, protein/peptide drugs are often rapidly released from hydrogel depots by diffusion due to their commonalities, such as hydrophilicity.44, 49 The severe initial burst release of hydrophilic protein/peptide agents and instability remain the crucial obstacles for developing controlled and sustained protein/peptide delivery systems; however, once excessive measures are employed to suppress the initial burst release, the problem that the drug can be not fully released at the advanced stage may become another obstacle, and the two opposing issues often influence each other. In this study, a sustained delivery system of Lixi was developed using thermosensitive hydrogels as the carrier for the first time. Unlike exendin-4 and insulin with negative charges, Lixi has a positive charge at physiological pH due to the C-terminal addition of 6 lysine residues.19 In contrast, PEG/polyester block copolymers have slightly negative charges.50 Therefore, we hypothesized that the positive charges in Lixi can form electrostatic linkages with the negative charges in the polymer carrier and that an appropriate electrostatic interaction between the two components might allow a predictable and customizable release of Lixi from the hydrogel matrix through charge dissociation. To validate the assumption, two thermosensitive triblock copolymers, PCGA-PEGPCGA and PLGA-PEG-PLGA, were synthesized by our group. The two single-component thermosensitive hydrogels as well as their 1/1 mixture hydrogel were used to deliver Lixi.

7



Their physicochemical properties and temperature-induced gelation behaviours in water were measured. The electrostatic interactions between Lixi and the carrier polymers were studied. To confirm the optimal Lixi-loaded gel system, the in vitro release behaviours of Lixi from the different hydrogels as well as the degradation of the hydrogels in vivo were evaluated in Institute of Cancer Research (ICR) mice. Subsequently, we conducted a pharmacokinetic study in Sprague Dawley (SD) rats. Meanwhile, the in vivo glucoselowering efficacy was also confirmed following a single subcutaneous injection of the optimal gel formulation into diabetic db/db mice. Furthermore, the glycosylated hemoglobin (HbA1c) level, plasma insulin concentration and pancreatic islet morphology were detected to confirm the long-term blood sugar control of db/db mice after three administrations of the optimal system. Finally, the lipid concentration in plasma, motor nerve function and histological morphologies of sciatic nerves and kidneys of animals were also assayed to verify whether the treatment with the optimal formulation is beneficial for reducing or arresting the progression of diabetic complications. Figure 1 schematically shows the current intention of the pilot study to achieve an injectable and thermosensitive hydrogel system containing Lixi for the synchronous treatment of T2DM and associated complications.

8


Page 8 of 64

Page 9 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. A schematic showing the development of a long-acting antidiabetic peptide delivery system using a thermosensitive hydrogel for the synchronous treatment of T2DM and associated complications. (a) The amino acid sequence of Lixi. Red represents acidic residues, blue represents alkaline residues, and green represents neutral residues. (b) The fabrication of a Lixi-loaded thermosensitive hydrogel system. The micelles formed by self-assembly of amphiphilic PEG/polyester block copolymers in water gradually 9



aggregate into a percolation micelle network with the rise of temperature, which is called sol-gel transition. (c) Achievement of long-term blood sugar control and amelioration of diabetic complications with the administration of the Lixi-loaded thermosensitive hydrogel system.

2. MATERIALS AND METHODS 2.1. Materials. PEG with the molecular weight (MW) of 1500, ε-caprolactone (CL) and stannous 2-ethyl-hexanoate (stannous octoate, 95%) were obtained from Sigma-Aldrich (USA). D, L-lactide (LA) and glycolide (GA) were purchased from Hangzhou Medzone Biotech Ltd. (Zhejiang, China). Lixi was provided by GL Biochem Ltd. (Shanghai, China). Cyanine5.5 N-hydroxysuccinimide ester (Cy5.5-NHS) was bought from Lumiprobe (USA). An ultrasensitive mouse insulin enzyme-linked immunosorbent assay (ELISA) kit and a mouse HbA1c kit were purchased from Crystal Chem. A Lixi Enzyme Immunoassay Kit was purchased from Phoenix Pharmaceuticals, Inc. (Burlingame, CA, USA). Lowdensity lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglyceride (TG) assay kit were purchased from Nanjing Jiancheng Bioengineering Institute. 2.2. Experimental animals. SLAC Laboratory Animal Co., Ltd. (Shanghai, China) provided ICR mice (female, 25 ± 2 g) and SD rats (male, 400 ± 25 g). The Model Animal Research Center of Nanjing University (Jiangsu, China) provided male diabetic db/db mice (BKS genetic background, 6 weeks old). All animals were fed in a temperature10


Page 10 of 64

Page 11 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


controlled room under a controlled photoperiod (12 h of light/12 h darkness) and had free access to water and food. The animals were housed for at least one week to acclimate to the environment prior to the experiments. The protocols for the animal experiments conformed to the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals (NIH Publications No. 8023, revised 1978) and achieved the approval of the ethics evaluation committee of Fudan University. 2.3. Synthesis and characterization of triblock copolymers. We synthesized two triblock copolymers, PCGA-PEG-PCGA and PLGA-PEG-PLGA, based on a bulk ring-opening copolymerization method. The detailed synthetic procedure has been described in previous publications.51 Considering the synthesis of PCGA-PEG-PCGA as an example, 15 g of PEG (0.01 mol) was transferred into a flask and then dehydrated at 130 °C under vacuum with continuous mechanical stirring for 3 h. After the flask was cooled to 50 °C, a fixed amount of CL and GA was added under the protection of argon. Next, the reaction system was stirred at 80 °C under reduced pressure, and argon was replaced three times in the flask to eliminate the residual moisture in the monomers. After all the monomers were melted, stannous octoate (0.2 wt% of monomers) was added, and then the reaction system was stirred at 150 °C under an argon atmosphere for 12 h. Crude polymers were washed at least three times using 80 °C water to remove the residual monomers and lowMW products. The final products were collected after lyophilization and stored at -20 °C until use. The PLGA-PEG-PLGA triblock copolymer was obtained by a similar procedure but using LA and GA monomers.

11



The chemical structures and compositions of the resultant polymers were confirmed by a 400 MHz 1H NMR spectrometer (Bruker, AVANCE III HD). The 1H NMR spectra of the two triblock copolymers were obtained at 25 °C using CDCl3 as the solvent. The MWs and values of molar mass dispersity (ÐMs) of the copolymers were determined by gel permeation chromatography (GPC, Agilent 1260) equipped with a differential refractometer as the detector. The eluent was tetrahydrofuran with a flow rate of 1.0 mL/min, and the measurements were carried out at 35 °C. The MWs were calibrated with monodispersed polystyrene (PS) standards. 2.4. Differential scanning calorimetry (DSC) measurements. DSC measurements were carried out on a TA Instruments Q2000 with a temperature and thermal sensitivity accuracy of ±0.1 °C and 0.2 μW, respectively. The experiments were conducted in an insert atmosphere using nitrogen with a flow of 50 mL/min. An empty pan was used as a reference. 10 mg of triblock copolymer was measured using the procedure as follows: ramping to 80 °C, isothermal annealing at 80 °C for 5 min to erase any thermal history, and then cooling to -80 °C and keeping for 5 min. Finally, the system was ramped to 80 °C again, and the data of first cooling cycle and second heating cycle were collected. The rates of cooling and heating were 20 °C/min. 2.5. Rheological study. The sol-gel transition of the aqueous copolymer solutions was investigated using a strain-controlled rheometer (Kinexus, Malvern) equipped with a Peltier cone-plate (conical: 1°, diameter: 60 mm, gap: 0.03 mm). The viscosity (η), storage modulus (G') and loss modulus (G'') were collected under an angular oscillation frequency

12


Page 12 of 64

Page 13 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of 10 rad/s and strain (0.5%) within the linear viscoelastic range as a function of temperature. The heating rate was fixed at 0.5 °C /min. 2.6. Dynamic laser scattering (DLS) and zeta potential analysis. The copolymers were first dialyzed, and then their hydrodynamic diameters and zeta potentials in water in the presence or absence of Lixi were analyzed by a DLS apparatus (Zetasizer Nano ZS90, Malvern) at 25 °C. An argon ion laser operating at a wavelength of 532 nm was used as a light source, and the measurements were performed at a scattering angle of 90°. The specimens were filtered using 0.45 μm filters to remove possible dust and equilibrated at 25 °C for 10 min before the measurement. 2.7. Circular dichroism (CD) spectroscopy. A CD instrument (Bio-Logic MOS-450) was used to analyze changes in the secondary structure of Lixi in the presence of a copolymer. CD spectra were recorded in a wavelength range of 190-260 nm with a band width of 1 nm and scanning speed of 0.5 nm/s at 25 °C. Each spectrum was corrected by deducting the corresponding deionized water background signal. 2.8. In vitro drug release. First, the stability of Lixi in phosphate-buffered saline (PBS) at 37 °C was detected. A concentration of 4 mg/mL of Lixi in PBS (pH 7.4) was immersed into a water bath shaker at 37 °C. One milliliter of Lixi solution was removed at the same time on each day. The stability of the Lixi solution was analyzed by a ultraviolet (UV)-vis spectrophotometer (Malvern, TU-1950) and a CD spectrometer. 25 wt% aqueous solutions of PCGA-PEG-PCGA, PLGA-PEG-PLGA and their 1/1 mixture, which is defined as the mixture, were prepared by dissolving the polymers in PBS, and then Lixi was added

13



into the polymeric aqueous solutions with continuous stirring at 4 °C for 12 h. Subsequently, 0.5 g of the aqueous polymer solution containing Lixi was accurately transferred into 10 mL vials (inner diameter: 15 mm). Then, the vials containing the samples were incubated in a water bath at 37 °C with an indicated shaking rate (50 rpm) for 15 min to form physical hydrogels. 5 mL of PBS (37 °C, pH 7.4) containing 0.025% NaN3 was gently added to the vials as the release medium. The release medium was completely withdrawn and replaced by the same amount of fresh buffer solution (37 °C) at the same time point every day. The amount of released Lixi in the medium was determined via a UV-vis assay at a wavelength of 227 nm. The UV absorbance of the release media of hydrogels without Lixi obtained at the same time intervals was used as the blank control. All release experiments were conducted in triplicate. 2.9. In vivo hydrogel formation and degradation. In situ hydrogel formation and subsequent in vivo degradation were performed in female ICR mice. Aqueous solutions of PCGA-PEG-PCGA, PLGA-PEG-PLGA and the mixture in PBS without NaN3 (25 wt%) were prepared and then subcutaneously injected into the backs of mice using a 1 mL syringe. The injection volume of the hydrogel was 0.2 mL per mouse. At each predetermined time point, five mice were euthanized via cervical dislocation, the injection regions were carefully cut open, and the remaining hydrogels were imaged. Finally, the gels were removed, the fibrous capsules on the surface were removed carefully, and the residual gels were weighed.

14


Page 14 of 64

Page 15 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2.10. In vivo imaging. Noninvasive in vivo imaging was conducted in ICR mice to trace the release of Lixi. Cy5.5-NHS is a near-infrared reactive dye with an excitation wavelength of 690 nm that was used to label Lixi. Briefly, 1 µmol of Cy5.5-NHS was dissolved in dimethyl sulfoxide (DMSO) and then added into an aqueous solution (pH 8.4) containing 4 µmol of Lixi. The system was stirred at 25 °C for 12 h. Next, dialysis and freeze-drying were carried out to obtain the Cy5.5 labeled Lixi (Cy5.5-Lixi). 0.2 mL of aqueous mixture solution (25 wt%) loaded Cy5.5-Lixi was injected into the ICR mice subcutaneously and then imaged at predetermined time intervals using an in vivo imaging system (In-Vivo Xtreme, Bruker). 2.11. Pharmacokinetic study of Lixi in SD rats. The pharmacokinetics was examined in male SD rats. One randomly divided group (labeled as free Lixi, n = 4) was administered a single subcutaneous injection of Lixi-containing PBS solution (4 mg/mL, 0.4 mL), and the other group (labeled as Lixi/Gel, n = 4) received a single subcutaneous injection of the mixture hydrogel containing Lixi (4 mg/mL, 0.4 mL). At predetermined time points (free Lixi group: 0.25, 0.5, 1, 3, 5, 8, 12, and 24 h; Lixi/Gel group: 0.25, 0.5, 1, 3, 5, 8, 12, 24, 72, 120, 168, 216, and 264 h), 0.25 mL blood samples were drawn from the tail vein and centrifuged (1000 ×g, 10 min, 4 °C) to separate plasma. The obtained serums were stored at -80 °C until the assay. The concentration of Lixi in blood serum was analyzed using a Lixi Enzyme Immunoassay Kit. 2.12. Hypoglycemic efficacy in db/db mice. The experimental diabetic db/db mice (8 weeks old) were randomly assigned to three groups (n = 8 mice per group) as follows: a

15



control group (PBS), a free Lixi group (twice-daily injections of Lixi solution), and a Lixi/Gel group (a single injection of gel system). The control group was administered twice-daily subcutaneous injections of 7.5 mL/kg PBS (pH 7.4). The free Lixi group received twicedaily subcutaneous injections of Lixi solution at a dose of 1.5 mg/7.5 mL/kg per injection for 10 days. The Lixi/Gel group received a single subcutaneous administration of the aqueous mixture solution (25 wt%) containing 4 mg/mL Lixi at a dose of 30 mg/7.5 mL/kg. The total doses in the two groups were equal over the 10-day testing period. Before the start of the experiment, blood samples were collected from all animals by the tail tip, and blood glucose levels were assayed using a blood glucose meter (AccuCheck Active, Roche Diagnostics, USA). The mean value was used as the initial blood glucose level. During the entire experiment, the mice had free access to water and food, and the blood glucose levels were assayed daily. After the first injection of Lixi/Gel system for ten days, the Lixi/Gel group received the other two doses of the Lixi/Gel formulation on days 10 and 20. Correspondingly, the mice in the control and free Lixi groups received twice-daily subcutaneous injections of PBS and Lixi solution, respectively, over the following 20 days. One month later, all the mice were euthanized, and blood samples obtained from the eyes were transferred into anticoagulant tubes with ethylenediaminetetraacetic acid (EDTA) and stored at -80 °C. The blood samples obtained were divided into two parts. One part was used to detect serum insulin levels using an insulin ELISA kit, and HDL-C, LDL-C and TG in serum were confirmed by commercial assay kits. The other part was used to determine blood HbA1c

16


Page 16 of 64

Page 17 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


concentrations using a mouse HbA1c kit. All measurements were performed and calculated according to the instructions in the assay kits. 2.13. Hematoxylin and eosin (HE) and immunofluorescence staining of the pancreas. After the euthanasia of the db/db mice, the pancreases were carefully dissected and harvested, and then sections were prepared. According to the standard procedures, paraffinembedded sections were stained with HE. The pancreas morphology was observed and imaged using a ZEISS microscope (Axiovert 200). Insulin immunofluorescence staining was also carried out to assess the pancreatic β-cell proliferation. The sections were immersed in blocking buffer for 30 min at ambient temperature followed by incubation overnight at 4 °C with an insulin antibody (1:200 dilution). On the following day, the sections were washed three times with PBS (pH 7.4) and incubated with a FITCconjugated secondary antibody (1:400 dilution) for 50 min at ambient temperature in the dark. Finally, the sections were incubated with medium containing 4′,6-diamidino-2phenylindole (DAPI) as the nuclear dye. All antibody and fluorescent molecules were provided by Servicebio (Wuhan, China). A panoramic scan of the tissue on each slice was performed using a Nikon Eclipse C1 (Nikon DS-U3) microscope, and Image J software was used to count the total green fluorescence intensity on each slide. 2.14. Sciatic motor nerve conduction velocity (MNCV). The electrophysiological study was performed immediately prior to animal sacrifice at one month post-treatment. The diabetic db/db mice were anesthetized using chloral hydrate, and a normal body temperature was maintained to ease animal stress from the anesthetic. The nerve

17



examination lasted less than 30 min for each mouse. The sciatic nerve was exposed, and the stimulating needle electrode was placed in the sciatic nerve proximal between the femoral trochanter and the ischial tuberosity. The recording needle was placed at the ankle joint at the sciatic nerve passing site, and the reference electrode was placed between the simulating electrode and the recording electrode, 1 cm away from the recording electrode. Stimulation with a single-pulse square wave for 0.1 ms at 1.5 Hz was conducted. The time from the start of the stimulation to the onset of an evoked potential in the muscle was recorded, and the distance between the stimulating electrode and the recording electrode was measured to calculate the MNCV. 2.15. Luxol fast blue (LFB) staining of the sciatic nerve. After the completion of the electrophysiological study, the animals were sacrificed. The exposed sciatic nerves were collected and evaluated by LFB staining. First, the sciatic nerve tissues were fixed with 4% paraformaldehyde at 4 °C overnight. Subsequently, the samples were dehydrated in ethanol with a graded concentration (70-100%) and embedded in paraffin. Sections with a thickness of 5 µm were prepared and stained with LFB. The nerve morphology was observed and imaged using a ZEISS microscope (Axiovert 200). 2.16. HE staining of the kidney. The kidney tissues of the db/db mice were also harvested and fixed with 4% paraformaldehyde after the mice were sacrificed. According to the standard procedures, paraffin-embedded sections were stained with HE, and the kidney morphology was observed and imaged by a ZEISS microscope (Axiovert 200).

18


Page 18 of 64

Page 19 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2.17. Statistical analysis. Data are displayed as the mean ± SD. A one-way t-test was used for comparisons between the groups. The differences between the groups were considered statistically significant when p-values were less than 0.05.

3. RESULTS 3.1. Characterization of the PCGA-PEG-PCGA and PLGA-PEG-PLGA triblock copolymers. The PCGA-PEG-PCGA and PLGA-PEG-PLGA triblock copolymers were obtained by ring-opening copolymerization of different monomers in the presence of PEG with terminal hydroxyl groups as the macroinitiator and stannous octoate as the catalyst. Figure 2 shows the 1H NMR spectra of the two triblock copolymers with their chemical structures. The number-average MW, Mn, of the PCGA-PEG-PCGA triblock copolymer was calculated based on the integration of the peaks at 4.60 (-COCH2O-), 3.65 (CH2CH2O-), and 1.39 (-COCH2CH2CH2CH2CH2O-) ppm. For PLGA-PEG-PLGA, the peaks at approximately 4.80 (-COCH2O-), 3.65 (-CH2CH2O-), and 1.55 (-COCH(CH3)O-) ppm were used to determine the Mn. GPC analysis showed that the two specimens displayed a unimodal pattern with ÐMs less than 1.30, indicating the successful synthesis of the desired polymers. Table 1 summarizes the molecular parameters of the two triblock copolymers used in this study.

19



Page 20 of 64

Figure 2. 1H NMR spectra of the (a) PCGA-PEG-PCGA and (b) PLGA-PEG-PLGA triblock copolymers in CDCl3. Table 1. Molecular parameters of the PCGA-PEG-PCGA and PLGA-PEG-PLGA triblock copolymers Sample

M na

Monomer ratioa

M nb

ÐMb

8000

1.30

(mol mol-1) PCGA-PEG-PCGA

1750-1500-1750

CL/GA = 12.4/1

20


Page 21 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


PLGA-PEG-PLGA a)

1850-1500-1850

LA/GA = 2.8/1

5100

1.15

The Mn of the middle block PEG was provided by Sigma-Aldrich, and the Mn of each

polyester block was calculated by 1H NMR. b) Confirmed by GPC.

3.2. Thermal properties and aqueous solution stabilities of the copolymers and their mixture. DSC thermograms of the two copolymers are presented in Figure 3(a). The melting of the PCGA segment and the subsequent melting of recrystallized PCGA domain led to a bimodal melting peak at 35 and 41 °C in the heating curve of PCGA-PEG-PCGA. Additionally, a weak melting transition detected at 15 °C was ascribed to the PEG constituent. Two crystallization peaks at -13 °C and 12 °C in the cooling process were assigned to the PEG and PCGA blocks, respectively. In contrast. the PLGA-PEG-PLGA copolymer exhibited an amorphous state due to the absence of melting peak and crystallization peak during the DSC measurement. Table S1 details the DSC data. The stabilities of the aqueous solutions of the two polymers and their 1/1 mixture, which is defined as the mixture, were also detected at ambient temperature. When the PCGAPEG-PCGA aqueous solution was placed at ambient temperature for one day, it spontaneously transformed into a translucent gel due to the crystallization of PCGA segments, as shown in Figure 3(b). In contrast, the aqueous solutions of PLGA-PEGPLGA and the mixture maintained a free-flowing liquid state, even after one week of

21



storage under the same conditions. Apparently, the stable sol state could offer great convenience in practical applications.

Figure 3. (a) DSC thermograms of the PCGA-PEG-PCGA and PLGA-PEG-PLGA polymers in the second heating cycle and the first cooling cycle. The heating and cooling rates were 20 °C/min. (b) The stabilities of the aqueous polymer solutions at ambient temperature. “1” represents the PCGA-PEG-PCGA/PBS system, “2” represents the PLGA-PEG-PLGA/PBS system, and “3” represents the mixture/PBS system. The polymer concentration was fixed at 25 wt%.

22


Page 22 of 64

Page 23 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3.3. Sol-gel transitions of the aqueous polymer solutions. In this study, 25 wt% aqueous solutions of PCGA-PEG-PCGA, PLGA-PEG-PLGA and the mixture were prepared in PBS, and these solutions were free-flowing liquids at room or low temperature and exhibited sol-gel transitions in response to an increase in temperature. The sol-gel transitions of these aqueous polymer solutions over temperature were investigated using a strain-controlled rheometer, and the results are presented in Figure 4. The viscosity of the polymer solutions underwent an abrupt increase as the temperature went up, indicating in situ gelation. The changes in the G' and G'' of the indicated systems over temperature are also shown in Figure S1. In general, the temperature of the crossover point that G' equals G'' is denoted as the sol-gel transition temperature.39 According to this criterion, the PCGA-PEG-PCGA and PLGA-PEG-PLGA solutions transformed into semisolid gels at 37 and 34 °C, respectively, while the sol-gel transition temperature of the mixture/PBS system was 35 °C. Meanwhile, it is well-known that amphiphilic PEG/polyester block copolymers easily self-assemble into core-corona micelles in aqueous medium.52 The temperature-induced gelation of the aqueous solutions of PEG/polyester copolymers was owe to the formation of a percolation micelle network via micellar aggregation.39, 52 Herein, all three systems maintained a physical hydrogel state under physiological conditions, indicating their potential as injectable biomaterials for in

vivo applications.

23



Figure 4. The viscosity () of the indicated samples in PBS with the rise of temperature. The polymer concentration was 25 wt%. Heating rate: 0.5 °C/min; oscillatory frequency: 10 rad/s.

3.4. Interactions between Lixi and the copolymers. DLS was employed to explore the interaction of Lixi with the copolymers in aqueous medium. Figure 5(a) shows the micellar sizes of the two copolymers and the mixture in water before and after the loading of Lixi. All the three polymer systems at a low concentration formed uniform micelles and their sizes were 35, 35, and 38 nm, respectively. The addition of Lixi did not affect the unimodal pattern of the polymer micelles but increased the particle sizes to some extent. The surface charge of the micelles was measured by zeta potential analysis to determine whether the introduction of Lixi changes the surface charge (Figure 5(b)). The pure copolymer micelles have a slightly negative surface charge in the neutral pH condition. In contrast, Lixi is positively charged (+34 mV) due to the presence of positively charged amine groups on the peptide backbone. Considering the mixture micelles as an 24


Page 24 of 64

Page 25 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


example, the complex of Lixi with the mixture micelles resulted in a decrease in the negative surface charge. The surface charge of the micelles approached electric neutrality as the concentration of Lixi increased. CD is the most widely used tool for detecting the secondary structure, including the folding and binding properties of proteins/peptides, at low concentrations. As shown in Figure 5(c), native Lixi had a negative bimodal band at approximately 208 and 222 nm, suggesting a typical α-helix secondary structure of the peptide. After loading Lixi into the aqueous copolymer solutions, the secondary structure of Lixi changed significantly. The negative signals at 208 and 222 nm significantly decreased in magnitude and showed a redshift as the copolymers were added. Increasing amounts of PLGA-PEG-PLGA in the polymers led to more substantial changes. Combined with the abovementioned measurements, we speculated that the electrostatic interactions between Lixi and the copolymers facilitate the loading of the drug into the polymer micelles and lead to an increase in micellar sizes and the transition of the Lixi conformation.

25



Figure 5. Interactions between Lixi and the copolymers. The Lixi concentration was 0.008 mg/mL, and the polymer concentration was 0.05 wt%. (a) The hydrodynamic diameters of the indicated systems. (b) The zeta potentials of the indicated systems. The polymer concentration was fixed at 0.05 wt%, the Lixi concentration was 0.008 mg/mL when the weight ratio of polymer and Lixi was 250:1. The concentration of pure Lixi was 0.008 mg/mL. (c) CD spectra of native Lixi and Lixi in the indicated systems.

26


Page 26 of 64

Page 27 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3.5. In vitro Lixi release from the hydrogels. First, the stability of Lixi in aqueous medium was evaluated. As shown in Figure S2, both the intensity of UV-vis absorption of the characteristic peak and the CD spectra of Lixi did not obviously change, even after Lixi was dissolved in PBS at 37 °C for 3 days. These results indicated that compared to the stability of other peptides such as exendin-4,53 the stability of the Lixi solution was significantly enhanced due to the C-terminal addition of 6 alkaline lysine residues. Next, we evaluated the in vitro release behaviours of Lixi. As displayed in Figure 6(a), Lixi was released quickly from the PCGA-PEG-PCGA hydrogel matrix, and the cumulative released amount reached 84.3 ± 2.3% in the initial 5 days. In contrast, the PLGA-PEGPLGA hydrogel formulation presented a slow release pattern, and only 58.2 ± 0.4% of Lixi was liberated within 10 days. Unlike the two single-component hydrogel systems, the mixture hydrogel provided sustained release of Lixi for up to 10 days, and 93.1 ± 2.8% of the encapsulated drug was released. Clearly, the mixture hydrogel is the optimized vehicle for the sustained delivery of Lixi in the present study. Furthermore, we evaluated the effects of drug loading amounts on the release profiles using the mixture hydrogel as the carrier, as shown in Figure 6(b). All the hydrogel systems containing various amounts of Lixi had a similar release profile, and more than 80% of the loaded Lixi was released. Compared to the other two systems, the 4 mg/mL Lixi-loaded mixture hydrogel system showed a slightly faster release rate. Finally, the release data were fitted by the Higuchi model, a diffusion-controlled mechanism,33 and the first order model that represents a mechanism of diffusion

27



combined with carrier erosion. 35 As displayed in Table S2, the release data more closely matched with the first order model and good squared correlation coefficients were achieved (R2 > 0.98). This finding manifested that the release of Lixi from the hydrogel matrix was dependent on the drug diffusion coupled with the erosion of hydrogel.

Figure 6. (a) In vitro Lixi release profiles from the different hydrogels. The concentration of Lixi was 4 mg/mL. (b) In vitro Lixi release profiles from the mixture hydrogel with the different drug concentrations. The polymer concentration was 25 wt%. The data are displayed as the mean ± standard deviation (SD) (n = 3). The lines are just visual guides. 3.6. In vivo gelation and degradation of the hydrogels. ICR mice were utilized to evaluate in vivo gelation and degradation of the three kinds of hydrogels. After subcutaneous injection 28


Page 28 of 64

Page 29 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of the aqueous polymer solutions, in situ hydrogels with a spherical morphology rapidly formed under the skin within 30 seconds due to contact with the body heat. Figure 7(a) shows representative images of the remaining gels at the predetermined time points. The sizes of the three hydrogels gradually decreased over the degradation time due to hydrolysis of the polyester segments. The PCGA-PEG-PCGA hydrogel maintained its integrity in vivo for over 6 weeks, while both the in vivo persistence of PLGA-PEG-PLGA hydrogel and mixture hydrogel lasted for up to 5 weeks. Furthermore, the remaining gels were separated from the adjacent tissues and weighed, and the variation curves of the weights as a function of degradation time are shown in Figure 7(b). The three hydrogel systems first experienced a swelling process on the first day and subsequently exhibited a gradual decrease in weight. The PCGA-PEG-PCGA hydrogel showed the slowest degradation rate, and only 62% of the mass was lost within 6 weeks. This finding was attributed to the crystallization of the PCGA blocks, which hindered the entry of water molecules and thus inhibited the hydrolysis of the polyester segments. In contrast, both the PLGA-PEG-PLGA hydrogel and mixture hydrogel nearly lost 100% of the mass during the same degradation period. Interestingly, compared to the PLGA-PEG-PLGA system, the mixture hydrogel showed a steadier degradation rate, which is beneficial for in vivo applications. Additionally, no tissue necrosis, congestion, edema, and redness were observed at the injection sites of the mice during the whole period of in vivo degradation test, which manifested the good biocompatibility of the hydrogels.

29



Figure 7. Research on degradation of the hydrogels in vivo. (a) Optical images of the indicated hydrogels at the predetermined time points after the subcutaneous injection of the aqueous polymer solutions (25 wt%) into the ICR mice. (b) The mass of the remaining hydrogels as a function of degradation time in the ICR mice. The results are presented as the mean ± SD (n = 5).

3.7. In vivo imaging of Lixi from the mixture hydrogel. Considering that the Lixi-loaded mixture hydrogel (Lixi/Gel) exhibited a satisfactory release dynamic in vitro and that the carrier matrix showed a steadier degradation curve in vivo, we chose it as the optimal formulation for the in vivo assessment. For real-time and noninvasive monitoring of the release of Lixi from the mixture hydrogel in vivo, Lixi was labeled with Cy5.5, which is a commonly used near-infrared fluorescent dye, and then loaded into the hydrogel matrix. Figure 8 shows the changes in the fluorescence signal of Cy5.5-Lixi as a function of time after the subcutaneous injection of the mixture hydrogel into the ICR mice. The 30


Page 30 of 64

Page 31 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fluorescence intensity first increased and then gradually decreased with time, indicating the sustained release of the drug from the hydrogel matrix. Compared to the in vitro release curve, the introduction of Cy5.5, a hydrophobic molecule, markedly enhanced the hydrophobicity of Lixi, resulting in a slower drug release rate in vivo from the hydrogel matrix.

Figure 8. Fluorescent imaging to trace the in vivo release of Cy5.5-Lixi from the mixture hydrogel after subcutaneous injection. All of the merged X-ray and fluorescent images are from the same ICR mouse.

3.8. Pharmacokinetic study. We also carried out a pharmacokinetic study in SD rats to detect the in vivo release pattern of Lixi from the mixture hydrogel. Both the Lixi/Gel and a solution of Lixi alone (free Lixi) at the same dose were subcutaneously administered to the backs of the rats. Subsequently, blood samples were obtained from the tail vein at different time points and measured by a Lixi Enzyme Immunoassay Kit. The results are presented in Figure 9, and the corresponding pharmacokinetic parameters were calculated, as shown in Table 2.

31



As expected, in the free Lixi group, the plasma level of Lixi abruptly increased in the initial 1 h post-injection and then decreased rapidly. The plasma half elimination time (t1/2z) of Lixi was only 2.2 h, and the drug was nearly undetectable in plasma at 12 h postinjection. In contrast, the Lixi that was entrapped in the hydrogel showed a prolonged circulation time in plasma. The initial burst release was significantly suppressed, and the concentration of Lixi in plasma remained stable until 9 days due to the constant release of the drug from the depot. The maximum plasma concentration (Cmax) of the Lixi/Gel was 24.5 ng/mL, which was remarkably lower than that of free Lixi (106.4 ng/mL), but the t1/2z of the Lixi/Gel was 14 times greater than that of free Lixi. Additionally, the Lixi/Gel had 7.6-fold higher area under the curve (AUC) than free Lixi, indicating significantly increased bioavailability.

Figure 9. Plasma Lixi concentrations in SD rats following a single injection of the optimal Lixi-loaded hydrogel system or free Lixi. The data are displayed as the mean ± SD (n = 4).

32


Page 32 of 64

Page 33 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 2. Pharmacokinetic parameters of the optimal Lixi-loaded hydrogel formulation and free Lixi in SD rats following the subcutaneous injection. Sample

Cmax (ng/mL)

Tmax (h)

t1/2z (h)

AUC(0-last)

MRT

(h·ng/mL) Free Lixi

106.4

1.0

2.2

378.9

2.8

Lixi/Gel

24.5

0.5

30.3

2891.6

94.6

AUC (0-last): area under the curve from time 0 until the final data point. MRT: mean retention time.

3.9. In vivo hypoglycemic efficacy. After successful demonstrations of the sustained release feature of Lixi from the mixture hydrogel both in vitro and in vivo, the pharmacodynamic profile of the Lixi/Gel system was detected using nonfasting diabetic db/db mice with a mutant leptin receptor gene as a model of T2DM. First, we evaluated the hypoglycemic efficacy of the free Lixi system. As shown in Figure S3, because metabolism in mice is faster than that in humans, the glucose-lowering effect lasted only a few hours after a single subcutaneous administration of free Lixi into the db/db mice. Therefore, to continuously control blood glucose levels, the free Lixi group (the positive control group) received twice-daily injections at a dose of 1.5 mg/7.5 mL/kg per injection in this study.

33



As shown in Figure 10(a), the mice receiving the PBS treatment remained at a hyperglycemic level, while the blood glucose levels of the animals were efficiently reduced and kept steady during the whole experimental period after twice-daily administrations of free Lixi. A single subcutaneous injection of the Lixi/Gel system at a dose of 30 mg/7.5 mL/kg also resulted in a significant improvement in the blood glucose level, and the hypoglycemic efficacy was maintained until 9 days, similar to the results from the twicedaily administrations of free Lixi. This continuous glucose-lowering effect demonstrated that the active Lixi was steadily released from the mixture hydrogel depot. Furthermore, no significant difference in the AUC values of the blood glucose between the Lixi/Gel group and the free Lixi group was observed (Figure 10(b)), indicating that Lixi administered in the different forms at the same total dose had the same bioavailability, while the patient compliance was greatly improved by a single injection of the Lixi/Gel system.

34


Page 34 of 64

Page 35 of 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 10. (a) Changes in the blood glucose level after the subcutaneous injection of the different formulations into nonfasting db/db mice. (b) AUC values of the blood glucose calculated for 10 days. The data are displayed as the mean ± SD (n = 8). The free Lixi group received twice-daily injections of Lixi solution at a Lixi dose of 0.75 mg/3.75 mL/kg per injection. The Lixi/Gel group was administered by a single subcutaneous injection of the Lixi/Gel system at a Lixi dose of 15 mg/3.75 mL/kg. The total doses in the two groups were equivalent over the whole experiment period. Significant differences between any two groups are specifically indicated with *** for p

Sustained Release Strategy Designed for Lixisenatide Delivery to

Recommend Documents