Sustained Administration of β-cell Mitogens to Intact Mouse Islets Ex Vivo Using Biodegradable Poly(lactic-co-glycolic acid) Microspheres

Summary

Here, we present methodology to generate and administer compound of interest-loaded poly(lactic-co-glycolic acid) (PLGA) microspheres to intact mouse islets in culture with subsequent immunofluorescence analysis of β-cell proliferation. This method is suitable for determining the efficacy of candidate β-cell mitogens.

Abstract

The development of biomaterials has significantly increased the potential for targeted drug delivery to a variety of cell and tissue types, including the pancreatic β-cells. In addition, biomaterial particles, hydrogels, and scaffolds also provide a unique opportunity to administer sustained, controllable drug delivery to β-cells in culture and in transplanted tissue models. These technologies allow the study of candidate β-cell proliferation factors using intact islets and a translationally relevant system. Moreover, determining the effectiveness and feasibility of candidate factors for stimulating β-cell proliferation in a culture system is critical before moving forward to in vivo models. Herein, we describe a method to co-culture intact mouse islets with biodegradable compound of interest (COI)-loaded poly(lactic-co-glycolic acid) (PLGA) microspheres for the purpose of assessing the effects of sustained in situ release of mitogenic factors on β-cell proliferation. This technique describes in detail how to generate PLGA microspheres containing a desired cargo using commercially available reagents. While the described technique uses recombinant human Connective tissue growth factor (rhCTGF) as an example, a wide variety of COI could readily be used. Additionally, this method utilizes 96-well plates to minimize the amount of reagents necessary to assess β-cell proliferation. This protocol can be readily adapted to use alternative biomaterials and other endocrine cell characteristics such as cell survival and differentiation status.

Introduction

Pancreatic β-cells are the only insulin-producing cells in the body and are critical to maintain blood glucose homeostasis. While healthy individuals have sufficient β-cell mass and function to properly regulate blood glucose, individuals with diabetes are characterized by insufficient β-cell mass and/or function^1,2. It has been proposed that inducing β-cell proliferation can ultimately increase β-cell mass and restore glucose homeostasis in individuals with diabetes³. However, evaluation and validation of potential β-cell proliferative compounds in intact islets is necessary before effective therapies can be developed. Transplantation of cadaveric human islets into individuals with diabetes restores blood glucose homeostasis for some time, but the availability and success of this experimental procedure is hindered by a shortage of human islets available for transplant and by β-cell death in the islets after transplant⁴. Even with the discovery of factors that induce multiplication of insulin-producing cells, a major challenge still exists in delivering these factors to relevant sites in vivo. One strategy for sustained local delivery of β-cell proliferative compounds is poly(lactic-co-glycolic) acid (PLGA). PLGA has a history of use in FDA approved drug delivery products owing to its high safety, biodegradability, and extended release kinetics⁵. Specifically, PLGA is a copolymer of lactide and glycolide that degrades via hydrolysis with water either in vivo or in culture into lactic acid and glycolic acid, which are naturally occurring metabolites in the body. The encapsulated drug compound can be released in the surrounding environment by both diffusion and/or degradation-controlled release mechanisms. Encapsulation of COI provides protection against enzymatic degradation, improving the bioavailability of the reagent compared to unencapsulated COI⁵. We suggest that PLGA microspheres can be used to administer candidate compounds to intact islets in culture, and ultimately in vivo. Testing the efficacy of PLGA to administer β-cell mitogens to islets ex vivo is critical before transplantation protocols are explored.

Currently, there is no technique to measure β-cell proliferation in live animals. Experiments to assess effectiveness of potential proliferative compounds in vivo therefore require administration of these compounds to live animals, with subsequent dissection and processing of pancreata for immunolabeling. Such protocols are expensive and laborious, and require the compound to be administered systemically, without any guarantee that they will reach the islets. Conversely, several immortalized β-cell lines are available for the study of insulin-producing cells in culture, but these cell lines lack the islet architecture and environment found in living organisms⁶. Immortalized β-cell lines are also characterized as having a much higher degree of replication than endogenous β-cells in vivo, thus complicating analysis of compounds that induce proliferation. In this study, we describe a protocol that uses intact islets isolated from adult mice. Unlike β-cell lines, intact islets retain normal islet architecture. Likewise, in contrast to experiments conducted in vivo, administering proliferative compounds directly to cultured intact islets significantly reduces the quantity of reagents that is necessary to accurately measure β-cell proliferation.

The current study utilizes PLGA to administer a COI, in this example, recombinant human Connective Tissue Growth Factor (rhCTGF). The method described here confers a significant advantage over the administration of raw compound to cultured islets as it allows for a continual release of compound into the media. Notably, this assay can be modified to administer a wide variety of proteins and antibodies of interest to intact islets. Effects on other endocrine cell types, including α-cells, may also be analyzed.

Protocol

All procedures were approved and performed in accordance with the Vanderbilt Institutional Animal Care and Use Committee.

1. Labeling COI with Fluorophore (Optional)

1. Choose a fluorescent dye that will react with a free primary amine (e.g., on a protein), such as succinimidyl esters or fluorescein derivatives, to visualize microsphere cargo. Dissolve 8x molar excess (relative to moles of COI) of fluorophore into 200 µl of dimethyl sulfoxide (DMSO).
2. Resuspend 50 mg of COI up to a final volume of 800 µl in a vehicle solution (final concentration of 62.5 ng/ µl). The vehicle solution will vary depending on the source of the COI.
   NOTE: Typical vehicle solutions include phosphate buffered saline (PBS) or DMSO.
3. Add all of the fluorophore/DMSO solution from step 1.1 and resuspend the COI. Vortex overnight at 4 °C. Alternatively, vortex the reaction mixture at room temperature (RT) for 4 hr.
4. Following the labeling reaction, remove excess fluorophore using a desalting column.
   1. Drain storage buffer from desalting column and rinse with 16 ml of deionized water. Discard all flow through.
   2. Add the fluorescently labeled COI to the desalting column. Elute with 1.2 ml of deionized water and collect the flow through.
   3. Repeat elution step an additional four times, each time adding 1.2 ml of deionized water and collecting the flow through as a separate sample.
   4. Freeze all collected flow through samples at -80 °C and lyophilize according to manufacturer's instructions for 24 hr.

2. COI-loaded Microsphere Preparation via the Water-in-oil-in-water Emulsion Solvent Evaporation Method

1. Add 1 mg of fluorescently labeled COI to 100 µl of deionized water to form the first water phase (W1).
2. Dissolve 65 mg of poly(lactic-co-glycolic acid) (50:50 Lactide:Glycolide, molecular weight 54,000 - 69,000) in 750 µl of dichloromethane in a microcentrifuge tube. Ultrasonicate for 10 - 30 sec (at 160 W) to completely dissolve the PLGA. This forms the oil phase (O).
3. Add all of the W1 phase generated in step 2.1 to the O phase in a drop-wise manner. Emulsify using a hand-held homogenizer at 20,000 rpm for 30 sec to form the W1/O phase.
4. Add all of the W1/O phase in a drop-wise manner to 15 ml of a 1% (weight/volume) aqueous poly(vinyl alcohol) (PVA) solution and emulsify using a hand-held homogenizer at 20,000 rpm for 30 sec.
5. Transfer all of the emulsion generated in step 2.4 to a 200 ml round-bottom flask and subject to a 635 mm Hg vacuum using a rotary evaporator for 1 hr to remove the solvent and generate the aqueous phase.
6. Aliquot 1 ml of the aqueous phase generated in step 2.5 to fourteen microcentrifuge tubes. Centrifuging the aqueous solution in the microcentrifuge tubes at 7,500 x g for 8 min.
   NOTE: At this step, the microspheres are present in the remaining aqueous solution and are concentrated to the bottom of the microcentrifuge tube.
7. Carefully remove 900 µl of aqueous solution from the microcentrifuge tubes with a micropipette, pipetting so as to not disturb the microspheres on the bottom of the microcentrifuge tube.
   1. Wash the microspheres of excess PVA by adding 1 ml deionized water to each microcentrifuge tube. Centrifuge at 7,500 x g for 8 min, and again carefully remove 900 µl of aqueous solution from the microcentrifuge tubes with a micropipette, pipetting so as to not disturb the microspheres on the bottom of the microcentrifuge tube.
      NOTE: The remaining 100 µl aqueous solution contains the generated microspheres.
8. Freeze the aqueous microsphere solution generated in step 2.7 at -80 °C.
9. Lyophilize microspheres using a lyophilizer according to manufacturer's instructions and store at -20 °C.
   NOTE: Measure loading efficiency of the COI by completely dissolving PLGA microspheres and determining the protein concentration relative to a standard curve of the free protein⁷.
10. As a negative control, generate a batch of "blank" microspheres (hereafter referred to as control microspheres).
    NOTE: The generation of control microspheres is identical to the generation of hydrophilic COI-loaded microspheres, with no COI addition to 800 µl of the vehicle solution in step 1.2. Match control particles in mass concentration of PLGA relative to COI-loaded PLGA microspheres for β-cell mitogen assay.

3. Preparation of Islet Culture Media and Pre-assay Media

1. Prepare 200 ml media for culture of intact mouse islets: RPMI 1640 media supplemented with 11 mM glucose, 10% horse serum, 100 U/ml penicillin G and 100 µg/ml streptomycin (hereafter referred to as islet culture media).
   NOTE 1: Unlike fetal bovine serum (FBS), horse serum lacks placental lactogen, an inducer of β-cell proliferation, which confounds analysis in the proliferation assay.
   NOTE 2: As the PLGA microspheres are not sterilized prior to use, the addition of antibiotics to the culture media is crucial to avoid microbiological contamination.
2. Remove 25 ml of islet culture media with an electronic pipettor and place in a 50 ml conical tube. Supplement the 25 ml of islet culture media in the 50 ml conical tube with a final concentration of 0.2 mM EGTA to generate the pre-assay media.
   NOTE: The addition of EGTA mildly loosens cell-cell contacts in the islets without altering islet architecture. This helps ensure the COI can reach the inner cells of the islet and helps prevent necrosis of the central islet.

4. Culturing of Intact Mouse Islets

1. Isolate intact islets from a pre-selected strain, sex, age, and genotype of mice following a routine collagenase digestion of the pancreas^8,9. Pool together islets from like samples.
2. Aliquot 40 islets into one well of a 96-well tissue culture plate in 200 µl of islet culture media. Visually size-match islets per well (a precise assessment of islet equivalency (IEQ) between samples is not necessary). Fill empty wells immediately adjacent to wells with islets with 200 µl sterile water to provide an evaporation buffer. Culture islets in media overnight at 37 °C under an atmosphere of 95% air and 5% CO₂.

5. Re-suspension of COI-loaded and Control PLGA Microspheres

1. After overnight islet recovery, resuspend microspheres by adding pre-assay media to one aliquot of lyophilized COI-loaded PLGA microspheres such that the final concentration of COI in the microcentrifuge tube is 10 ng/µl (the amount of COI present in a given amount of the generated microspheres was determined in step 2.9). Sonicate the resuspended COI-loaded microspheres in an ice water bath for 10 min at 160 W with 10 sec long pulses at 4 °C.
2. Resuspend control PLGA microspheres by adding the same volume of pre-assay media to an equal mass of lyophilized control PLGA microspheres. Sonicate the resuspended control PLGA microspheres in an ice water bath for 10 min at 160 W with 10 sec long pulses at 4 °C.
3. Visually confirm microspheres are dispersed by pipetting 2 µl of resuspended microspheres between two glass coverslips. Visualize microspheres using a 40X objective on an epifluorescence or brightfield microscope.
4. If significant aggregation of microspheres is still visible, sonicate in an ice water bath for an additional 10 min at 160 W with 10 sec long pulses at 4 °C and repeat visualization of resuspended microspheres.

6. Treatment of Islets with COI-loaded or Control PLGA Microspheres

1. For each well to be treated with COI-loaded microspheres, prepare assay media by diluting the 10 ng/µl of the resuspended COI microspheres with pre-assay media to a final volume of 100 µl and a pre-determined final concentration.
   NOTE: The optimal final concentration of the protein will vary for each COI used for this assay. Ideally, a range of final concentrations are tested.
2. For each well to be used as a control, prepare control assay media by diluting the resuspended control PLGA microspheres generated in step 5.2 with the same dilution volume used to dilute the COI-loaded microspheres in step 6.1.
   NOTE: Islets treated with control assay media will serve as a negative control to enable detection of any effect the blank microspheres may have on the experimental outcomes.
3. Carefully remove 100 µl of islet culture media from each well with a micropipette such that no islets are dislodged from the wells. Gently add 100 µl of assay media or 100 µl of control assay media to each well.
4. Incubate islets at 37 °C under an atmosphere of 95% air and 5% CO₂ for three days.

7. Dispersing Intact Islets onto Microscope Slides

1. After three days, use a micropipette to carefully remove and discard assay media from all wells so as not to dislodge islets. Gently add 200 µl PBS to islets to wash them of the assay media. Carefully remove and discard PBS with a micropipette and gently wash islets again with an additional 200 µl PBS.
2. Remove and discard PBS with a micropipette and add 100 µl of 0.025% trypsin, 2 mM EDTA solution. Incubate at RT for 3 min. Pipet islets in the trypsin-EDTA solution up and down every 2 min until islets are visually confirmed to be dispersed into single cells (note that some small clumps of cells may still be present) using a light microscope. Transfer the entire volume of each well individually into labeled- microcentrifuge tubes.
3. Add 400 µl of islet culture media to each microcentrifuge tube to stop the trypsin-mediated cell dissociation.
4. Centrifuge samples for 5 min at 100 x g at 4 °C to pellet islet cells to the bottom of microcentrifuge tubes.
5. Gently remove supernatant using a micropipette, and resuspend pellet in 200 µl fresh islet culture media.
6. Using a cyto-centrifuge, centrifuge dissociated islets at 140 x g for 3 min onto charged microscope slides.
7. Air dry slides at RT for 10 min, then draw a box around the dissociated islets with a hydrophobic marking pen.
8. Fix cells with 75 µl 4% paraformaldehyde (PFA) at RT for 10 min. Remove 4% PFA and gently wash cells in 75 µl PBS two times. After washing, permeabilize cells with 75 µl 0.2% Triton X-100 in PBS for 10 min. Then, wash cells with 75 µl PBS.
   Caution: PFA is known to be allergenic, carcinogenic, and toxic.

8. Immunofluorescence Labeling of Dissociated Islets for Insulin and the Cell Proliferation Marker, Ki67

1. Prepare a humid chamber by placing two wet paper towels in the bottom of a 100 slide capacity microscope slide box.
2. Place slides flat down (cells facing up) in humid chamber. Prepare samples for labeling by aspirating the PBS from the previous wash step using a micropipette and adding 75 µl of blocking solution (5% Normal Donkey Serum (NDS) in PBS) to each slide to block non-specific binding of antibodies to the specimens. Incubate slides in blocking solution for 1 hr at RT.
3. Gently aspirate blocking solution using a micropipette and add 75 µl of primary antibody solution containing guinea pig anti-insulin and rabbit anti-Ki67 antibodies, diluted 1 µg/ml in 5% NDS in PBS. Incubate at RT for 1 hr.
   NOTE: Alternative markers of cell proliferation include proliferating cell nuclear antigen (PCNA) and phosphohistone H3 (PH3),
4. Gently aspirate primary antibody solution using a micropipette and rinse the cells with 75 µl PBS. Aspirate PBS using a micropipette and rinse cells two more times, each time with 75 µL PBS. Incubate each sample with 75 µl anti-guinea pig Cy5 secondary antibody and anti-rabbit Cy3 secondary antibody diluted to 2 µg/ml in blocking solution for 1 hr at RT.
   NOTE: Do not use secondary antibodies labeled with the same or spectrally overlapping fluorophore that was already used to label the microspheres.
5. Aspirate secondary antibody solution using a micropipette and incubate in 75 µl of 300 nM DAPI in PBS for 3 min at RT. Aspirate DAPI using a micropipette and rinse in deionized water for 5 min.
6. Gently aspirate water from samples using a micropipette. Spot a drop (roughly 100 µl) of fast-drying mounting solution containing antifade reagent on a glass coverslip. Invert the microscope slide sample side down, onto the mounting solution. Gently press the coverslip against the slide using a pipet tip to remove bubbles and wipe away excess liquid with a soft cleaning tissue. Allow coverslips to dry overnight at RT.

9. Image Acquisition and Analysis

1. Use an epifluorescence microscope with an attached camera for image acquisition. For each fluorophore, determine the optimal exposure time (typically 20 msec for DAPI, 40 - 80 msec for insulin, and 250 msec for Ki67). Ensure image acquisition parameters are equal for all samples.
2. For each sample, manually count 3,000 insulin-positive cells and of those count how many are also Ki67-positive. Only count insulin-positive cells that have a well-defined and clearly visible nucleus. Calculate the percentage of proliferating β-cells for each sample by dividing the number of Ki67-positive/insulin-positive cells by the total number of insulin-positive cells and multiplying by 100.
3. After determining the percentage of proliferating β-cells for each sample, determine if a statistically significant difference in β-cell proliferation exists between control and experimental samples by analyzing results with a one-way ANOVA using commercially available statistical software.

Results

Figure 1 is a visual representation of the microspheres generated using the above protocol. The protocol described here yields rhCTGF-loaded microspheres of various sizes. The largest fraction of microspheres will be between 1 and 10 µm in diameter, though some microspheres may be larger (Figure 2). If desired, microsphere size can be tuned and optimized based on fabrication parameters such as homogenization speed and time, surfactant con...

Discussion

The study of β-cell proliferation in culture is typically hampered by several difficulties. First, immortalized β-cell lines are characterized by higher degrees of proliferation than what is found in endogenous β-cells in live islets. Additionally, these immortalized cell lines lack the normal architecture critical for normal β-cell function. These two facts make it difficult to determine if results obtained using immortalized β-cell lines will hold true when tested in vivo or in whole i...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Oregon Green 488 Carboxylic Acid, Succinimidyl Ester, 6-isomer	ThermoFisher Scientific	O6149	For labeling COI with fluorophore
DMSO Dimethyl Sulfoxide	Fisher BioReagents	BP231-1	For dissolving fluorophore in step 1
Disposable PD-10 Desalting Columns	GE Healthcare	17-0851-01	Desalting column used in step 1
Resomer RG 505, Poly(D,L-lactide-co-glycolide), ester terminated, molecular weight 54,000 - 69,000	Sigma-Aldrich	739960	Used in generation of microspheres in step 2
Poly(vinyl alcohol) molecular weight 89,000 - 98,000	Sigma-Aldrich	341584	Used in generation of microspheres in step 2
RPMI 1640	Thermo Scientific	11879-020	For culturing islets
Dextrose Anhydrous	Fisher BioReagents	200-075-1	Supplement for islet media
Penicillin-Streptomycin	Sigma-Aldrich	P4333	Antibiotics for islet media
Normal horse serum	Jackson ImmunoResearch	008-000-121	Supplement for islet media
96-well tissue culture plate	Corning	3603	For culturing islets
Ethylene glyco-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid	Sigma-Aldrich	E4378	Supplement for pre-assay islet media
Cytospin 4 Cytocentrifuge	Thermo Scientific	A78300003	For spinning cells onto microscope slides
EZ Single Cytofunnel	Thermo Scientific	A78710020	For spinning cells onto microscope slides
Ethylenediaminetetraacetic acid	Fisher BioReagents	BP118-500	Used in dissociating islets
paraformaldehyde	Sigma-Aldrich	P6148	For fixing cells
Triton  X-100	Fisher BioReagents	BP151	For permeabilizing cells
Normal donkey serum	Jackson ImmunoResearch	017-000-121	Blocking reagents for immunofluorescence
Anti-Ki67 antibody	abcam	ab15580	For Ki67 immunofluorescence
Polyclonal Guinea Pig Anti-Insulin	Dako	A0564	For insulin immunofluorescence
Cy3 AffiniPure Donkey Anti-Rabbit	Jackson ImmunoResearch	711-165-152	For Ki67 immunofluorescence
Cy5 AffiniPure Donkey Anti-Guinea Pig	Jackson ImmunoResearch	706-175-148	For insulin immunofluorescence
DAPI (4',6-Diamidino-2-Phenylindole, Dihydrochloride)	ThermoFisher Scientific	D1306	For nuclei visualization in immunofluorescence
Aqua-Mount	Lerner Laboratories	13800	Fast drying mounting media
FreeZone -105 °C 4.5 Liter Cascade Benchtop Freeze Dry System	Labconco	7382020	For lyophilization of microspheres