Cell associated scaffolds for delivery of agents

Abstract

The present invention relates to the use of scaffolds to enhance the viability of cells implanted in the integumentary system such that the cell may release an agent. The scaffold is capable of protecting the cell, as well as allowing for adequate nutrient delivery at the implant site through vascularisation in and around the scaffold.

Claims

1. A method for delivering an agent to a subject consisting of associating cells that produce said agent with a scaffold, and implanting said scaffold-associated cells subcutaneously in said subject, wherein the scaffold increases vascularisation at the site of subcutaneous implantation when compared to subcutaneous implantation of the same cell type without said scaffold and enhances the viability of the scaffold-associated cell after implantation by promoting vascularization at the site of implantation, and wherein the cells are encapsulated, wherein the scaffold comprises a lid region or a base region comprising cubic pores with maximum dimensions ranging from 5 μm to 25 μm, and a centre region comprising cubic pores with maximum dimension of 2000 μm and a minimum dimension of at least 500 μm, wherein the pores of the centre region are large enough to allow a single encapsulated cell to be housed.

2. The method of claim 1, wherein the scaffold comprises biocompatible fibres, which are 10-50 μm thick.

3. The method of claim 2, wherein the fibres are produced by 3D-printing by melt electrospin writing.

4. The method of claim 1, wherein the scaffold comprises pores that become traversed by blood vessels within at least 10 weeks after being implanted subcutaneously.

5. The method of claim 3, wherein the scaffold reduces or delays the host cell response of the subject to the scaffold-associated cells when compared to the host cell response of the subject to the same cell type implanted subcutaneously without said scaffold or said encapsulation.

6. The method of claim 5, wherein the cells are encapsulated in alginate.

7. The method of claim 5, wherein a deficiency of said agent is a causative factor of a disease or disorder selected from the group consisting of diabetes, liver failure, and Parkinson's disease.

8. The method of claim 7, wherein the disease or disorder is diabetes.

9. The method of claim 8, wherein the agent is selected from the group consisting of insulin, an insulin analog, or a precursor of insulin, such as preproinsulin or proinsulin.

10. The method of claim 9, wherein the scaffold-associated cells are pancreatic cells.

11. The method of claim 1, wherein the scaffold comprises a lid region and a base region arranged at opposite sides of the scaffold.

12. The method of claim 1, wherein the scaffold comprises a boxed configuration.

13. A method of treating diabetes, consisting of implanting an encapsulated cell associated with a scaffold subcutaneously, where said cell releases insulin, an insulin analog, or a precursor of insulin, wherein the scaffold increases vascularisation at the site of subcutaneous implantation when compared to subcutaneous implantation of the same cell type without said scaffold and enhances the viability of the scaffold-associated cell after implantation by promoting vascularization at the site of implantation, and wherein the cells are encapsulated, wherein the scaffold comprises a lid region or a base region comprising cubic pores with maximum dimensions ranging from 5 μm to 25 μm, and a centre region comprising cubic pores with maximum dimension of 2000 μm and a minimum dimension of at least 500 μm, wherein the pores of the centre region are large enough to allow a single encapsulated cell to be housed.

14. The method of claim 13, wherein the cell is encapsulated in alginate before being associated with said scaffold.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings as follows.

(2) FIG. 1: MIN6 pseudoislet formation, encapsulation and simulation by glucose. (A) MIN6 pseudoislets were formed by growing cells in suspension culture. (B) Capsules formed by passing air through 2.2% alginate solution, followed by gelling in a BaCl.sub.2 bath. (C) MIN6 pseudoislets encapsulated in alginate, and (D) their responsiveness to change in glucose concentration to secrete insulin into the media measured. Data: mean±s.d., p<0.05, Mann-Whitney U-test.

(3) FIG. 2: PCL scaffold created by melt electrospin writing. (A) Scaffold of height 1.5 mm containing multiple cubic pores of dimension 0.9×0.9×0.9 mm. (B) Each alginate microcapsule slotting into a cubic pore. (C) The body of scaffold together with less porous base region prior to fusing it with the body and (D) a high power view of the complete device with microcapsules visible inside it. (E) Mouse islets retain their viability in microcapsules when placed in PCL scaffolds kept in tissue culture. Green fluorescent staining of CFDA shows the cells are alive; there was no staining of the red dye PI, which would have indicated cell death.

(4) FIG. 3: Potential of PCL scaffolds for vascularisation. (A) HUVEC attach (arrows) to fibres of PCL scaffolds within 24 hours of seeding. (B, C) In cultures of >7 days the associated cells remain viable as observed by fluorescence of Calcein-AM, and retain expression of vascular endothelial cell marker CD31.

(5) FIG. 4: Effect of implantation of PCL scaffold in diabetic mice. (A) PCL scaffold containing encapsulated MIN6 pseudo-islets were subcutaneously implanted in diabetic NOD/SCID mice and retained for 4 weeks. (B) Hyperglycaemia in diabetic mice implanted with scaffolds containing encapsulated insulin-producing cells (red) was normalized to blood glucose levels of non-diabetic mice (blue). When the scaffolds were coated with a factor that promotes blood vessel formation (VEGF, purple) normalization of blood glucose was faster (red vs purple). In contrast, encapsulated insulin-producing cells implanted without a scaffold in diabetic did not normalize glucose levels (green). (C) Scaffold removed from mice post implantation showed formation of blood vessel (arrow).

(6) FIG. 5: Histology assessment of explanted scaffolds. (A) H&E staining of explanted scaffold containing encapsulated MIN-6 pseudoislets. (B) Sirius red stain showing areas of collagen deposition (pink) forming an ECM around the alginate capsules and throughout the scaffold. (C) Higher magnification of Sirius red stain showing attachment of cells to PCL fibres and collagen deposition crosslinking them. (D) CD31 staining showing blood vessel formation, (E) DAPI stain of the same region and (F) merged image of CD31 and DAPI.

(7) FIG. 6: Insulin production from implants. (A) Encapsulated cells within the scaffolds continue to respond to glucose stimulation ex vivo when scaffolds were incubated in 25 mM glucose (16 hr) followed by 2.8 and 20 mM glucose (1-hour each). Culture media collected after each incubation to measure insulin by ELISA, and scaffolds washed (arrow) in between change of media. Data: mean±range (B) Insulin staining within pseudoislets placed in the PCL scaffold. (C) Cell population stained with DAPI and (D) Merged image showing insulin production from encapsulated cells.

(8) FIG. 7: Detection of mouse insulin in diabetic NOD/SCID mice implanted with either sham scaffolds or scaffolds associated with MIN6 cell clusters. (A) Circulating concentration of mouse insulin at termination. (B) Rate of insulin release from grafts in culture. P<0.01 using Mann-Whitney Utest.

(9) FIG. 8. Effect of implanted scaffold-associated cells on (A) blood glucose and (B) body weight of immune competent BALB/c mice.

(10) FIG. 9: Insulin production from implanted scaffold-associated cells (A) Circulating concentration of mouse insulin in diabetic mice f*P<0.01 vs sham, by ANOVA. Dotted line: assay detection limit; Shaded area: circulating range of insulin in healthy mice. (B) Ex vivo immune staining of insulin (green) located within MIN6 pseudo-islets inside a microcapsule.

(11) FIG. 10: Vascularisation of implanted scaffold. Blood vessels were clearly visible (A) macroscopically on the side of the implant proximal to the skin and (B) on the side distal to the skin. (C) Blood vessels in red and nuclei in blue inside the implant as visualised by multiphoton intravital microscopy.

(12) FIG. 11: Host cell response to implantation of scaffold-associated cells in immune-competent mice (A) H&E stain shows infiltration of cells in the scaffolds and foci of inflammatory recruitment (arrows) on predominantly on scaffold, but not around microcapsule (*). (B) H&E stain on device under higher magnification showing no foci of inflammatory cells around microcapsule. (C) NF-κB p65+ pro-inflammatory cells localised on scaffolds and not around microcapsule. Similarly (D) MPO+ neutrophils, (E) CD68+ monocyte/macrophages and (F) very few CD19+ B-lymphocytes were all localised to scaffold and not around microcapsules. CD4+ T-lymphocytes were not detectable. Sirius red stain on implant shows ECM deposition of collagen (pink) (G) interconnecting the strands of the scaffold, and (H) around microcapsules. (I) α-SMA+ myofibroblasts showing area of active fibrogenesis around microcapsules and in the scaffold.

(13) FIG. 12: Schematic showing MEW printed scaffolds that may include barrier membranes to prevent poential escape of cells from within devices. (A) Boxed configuration of device with a body, lid and base. Barrier membrane may be placed between the body and lid/base (B) Cylindrical configuration of device with a hollow core where a barrier membrane can be inserted.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

(14) Preferred embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings.

EXAMPLE 1

(15) Methods

(16) Cell culture:

(17) Monolayers of MIN6 cells (immortalised mouse pancreatic β cell line) were cultured to 50-70% confluency at 37° C. with 5% CO.sub.2 in Dulbecco's modified Eagle's medium (DMEM) containing 25 mM glucose supplemented with 10% (v/v) heat-inactivated fetal bovine serum (FBS), 2 mM L-glutamine, 1 mM sodiμm pyruvate, 100 μM non-essential amino acids, 100 units/mL penicillin and 100 μg/mL streptomycin. Pseudoislets were formed by seeding 1×10.sup.6 MIN6 cells in ultra-low attachment 10 cm dishes (Corning Incorp., NY, USA) and culturing for 5 days at 37° C. in 5% CO2. During this time, the cells came together to self-form pseudoislets.

(18) Alginate Encapsulation:

(19) Pseudoislets were pelleted by low speed centrifugation (10×g, 2 min), the culture media was discarded and the loosely pelleted pseudoislets suspended in a highly purified 2.2% alginate solution (60:40 guluronic acid: mannuronic acid, UP MVG PRONOVA, FMC Biopolymer). The pseudoislet/alginate suspension (1:6 packed cell volμme: alginate) was transferred to a syringe and inserted into the encapsulation device which consisted of an air-driven droplet generator with a tubing connected to a cylinder of medical grade air. The suspension was infused through the encapsulation device at a rate of 0.67 mL/min with airflow rate of 7 L/min. The flowthrough of droplets was collected in a 20 mM BaCl.sub.2 bath for 2 minutes to allow gelling of the capsules. The capsules (˜700 μm) containing 1-3 pseudoislets were washed in phosphate buffered saline (PBS) and cultured overnight at 37° C. in 5 % CO.sub.2 in DMEM as described above.

(20) Melt Electrospin Writing (MEW):

(21) Biological grade polycaprolactone (PCL) scaffolds were created by MEW as described previously using a translating x-y stage. Briefly, PCL (Perstorp, M.sub.n ˜41 kDa, PDI=1.78) pellets were placed in a Luer lock syringe and heated to 60° C. in an oven overnight to remove air bubbles. A blunt 23 G needle acting as the spinneret was attached to the syringe and placed into an electrical heating system heated to 65° C. The PCL melt was then electrospun using a collector distance of 15 mm, and flow rate of 20 mL/h with a voltage of 10 kV applied to the spinneret. Writing with the PCL fibres (20 μm) was achieved by collecting on a grounded plate connected to a programmable x-y stage, controlled using Mach 3 software.

(22) Preparation of Implants:

(23) For each device, 24 alginate capsules were individually picked and each placed inside a cubic pore (900×900×900 μm) of the PCL scaffold of size 1 cm.sup.2. The capsules were secured by placing a PCL lid and base (1 cm.sup.2) which were heat sealed to the scaffold from four sides. The complete device was placed in DMEM media.

(24) Animal Model:

(25) All animal experiments were approved by University of Sydney Animal Ethics Committee (Protocols 2015/879 and 2017/1237). Immune-deficient NOD/SCID or immune-competent BALB/c mice were housed under normal acclimatising conditions (12 hours dark/light cycle, chow rodent diet and drinking water ad libitum). Some animals (weighing >20 g) were made diabetic by multiple low dose intraperitoneal injections of streptozotocin [STZ; NOD/SCID: 5×40 mg/kg; BALB/c 4×75 mg/kg body weight)]. A subset of the diabetic mice was anesthetised (2-5% isoflurane) and PCL scaffold implanted by subcutaneous surgery on the back near the scapula. Analgesia (0.1 mg/kg buprenorphine and 0.5 mg/kg meloxicam) was administered for 3 days after surgery. Body weight and random blood glucose (tail vein prick) were measured 3 times a week for 4 weeks after which animals were anesthetised, blood collected by cardiac puncture, implanted device and pancreas removed. The implanted device was placed in DMEM (25 mM glucose) and cultured overnight at 37° C. in 5% CO.sub.2for measurement of insulin in the media.

(26) Static Glucose Stimulation:

(27) Incubation in glucose for insulin secretion was conducted on encapsulated pseudoislets and on implanted devices after removal from the host. DMEM (25 mM glucose) was removed from encapsulated pseudoislets/implants, and samples washed with Kreb's Ringer Bicarbonate (KRB) buffer (115 mM NaCl; 4.7 mM KCl; 1.28 mM CaCl.sub.2 and 1.2 mM MgSO.sub.4) supplemented with 0.1% BSA. This step was followed by 1-hour incubation with KRB with 2.8 mM glucose. Samples were washed in KRB before 1-hour incubation with KRB with 20 mM glucose. All incubations were at 37° C. in 5% CO.sub.2. Supernatants were carefully removed after each incubation period and stored at −20° C. Insulin release was measured by mouse insulin ELISA kit (Mercodia, USA) as per manufacturer's instructions.

(28) Histological Assessment:

(29) After implants were removed from mice, they were embedded in OCT medium and stored at −80° C. Sections (10 μm) were placed on slides and fixed with methanol for 5 min. Samples were placedin 100% and 70% and followed by immersion in water. Haematoxylin and eosin (H&E) staining performed as per routine established protocol. To demonstrate collagen deposition, Sirius red staining was performed. Sections were stained 1-hour with picric acid containing 0.1% fast green and 0.1% direct red.

(30) Immunofluorescent Staining:

(31) For localisation of specific proteins OCT embedded samples were sectioned (10 μm) and fixed with methanol for 5 min following 3 washes in Tris-Buffered Saline and Tween 20 (TBST), non-specific sites were blocked by incubation with 1% BSA in TBS (blocking buffer) for 1-hour. The slides were then washed off with TBST prior to overnight incubation in primary antibodies (insulin 1:400, CD31 1:100, NF-κB p65 1:200, MPO 1:100, CD68 1:50, CD19 1:100, α-SMA 1:200) diluted in blocking buffer. The sections were then washed in TBST and finally fluorescent-labelled secondary antibodies (1:1000) diluted in the blocking buffer were added and the slides incubated for 1 hour in the dark. The sections were washed in TBST, briefly dried and mounted in Prolong-Gold mounting media containing DAPI (Molecular Probes, OR, USA).

(32) Results

(33) Pseudoislet Formation and Encapsulation:

(34) MIN6 cells were grown as a suspension culture for 5 days to generate self-forming aggregates or pseudoislets (FIG. 1A). These structures are similar in size (135±44 μm diameter, n=8) to normal mouse islets. Alginate capsules of 755±49 μm diameter (FIG. 1B) were generated. Between 1 and 3 pseudoislets were contained within these capsules (FIG. 10). A sample of 10 randomly selected capsules (containing 1-3 islets) were tested for insulin secretion in response to change in glucose concentration. As shown in FIG. 1D increasing glucose concentration from 2.8 to 20 mM caused ˜3-fold increase (p=0.02, n=4/treatment) in insulin secretion by the encapsulated MIN6 pseudoislets.

(35) Scaffold Printing:

(36) Scaffolds were created by the process of MEW using PCL. The structure consists of a lattice of cubic pores. Each cubic pores has dimensions 0.9×0.9×0.9 mm, as shown by scanning electron micrograph (FIG. 2A), and is sufficient to contain a spherical alginate capsule which can be slotted into the cubes. The scaffold and the associated alginate capsules were secured by a base and lid structure (FIG. 2C) on the top and bottom. The base and lid were also printed using MEW but contained smaller cubic pores. The base region and the centre region were printed together, whereas the lid was separately printed and heat sealed to the rest of the scaffold. The completed scaffold is shown in FIG. 2D. The PCL scaffold enhanced cell viability, evident from mouse islet cells staining green (viable) with CFDA and not red (dead) with PI after placing in the scaffold and incubation in vitro (FIG. 2E).

(37) PCL Scaffold Encourage Endothelial Cell Attachment:

(38) The main purpose of the scaffold was to promote angiogenesis when implanted in vivo. To enhance angiogenesis, it is necessary for endothelial cells to adhere to the strands of the scaffold. Adhesiveness to PCL scaffolds was tested in vitro with human umbilical vein endothelial cells (HUVEC). These cells readily attached to PCL, maintain viability and expressed endothelial cell marker CD31 (FIG. 3A-C).

(39) Normalisation of Blood Glucose by Subcutaneous Implantation:

(40) PCL scaffolds containing either encapsulated MIN6 pseudoislet or empty alginate capsules were implanted subcutaneously in STZ-treated diabetic NOD/SCID mice (FIG. 4A). The wound site was closed and mice followed for approximately 4 weeks. Body weight and random BGL were measured 3 times a week. Non-STZ treated mice served as non-diabetic controls (NDC). BGL ranged between 6-9 mmol/L in NDC mice over the study period, whereas in diabetic mice implanted (DI) with encapsulated MIN6, mean BGL reached ˜30 mmol/L at the time of implantation and gradually started to decline 5-7 days post implantation. The decreasing trend continued, resulting in a lowering of BGL to 4 mmol/L (FIG. 4B). Implants from these mice were removed as shown in FIG. 4C. There was growth of fibro-fatty tissue in and around the scaffold, and formation of blood vessels were also observed.

(41) PCL Scaffold Promotes Cellular Infiltration and ECM Formation:

(42) Histological assessment of explanted scaffolds by H&E staining (FIG. 5A) showed large populations of infiltrated cell throughout the implanted device. These were not inflammatory infiltrates as confirmed by negative staining of myeloperoxidase expressing polymorphonuclear cells and F4/80 expressing monocyte/macrophages (not shown). Rather, they were mostly fibroblast as evident from the large amount of collagen deposition around the PCL as shown by pink areas of Sirius red staining (FIG. 5 B and C). Cells not only infiltrated the scaffolds, but attached to the PCL strands as observed under high magnification (FIG. 5C).

(43) The formation of blood vessels was confirmed within the scaffolds, with staining for the vascular endothelial cells marker CD31. FIG. 5D illustrates classical chicken-wire like expression of CD31, as seen on blood vessels (arrow). CD31 expression was also seen on the cross-section through a blood vessel (circle).

(44) Encapsulated MIN6 Pseudoislets Maintain Expression and Secretion of Insulin:

(45) The grafts were explanted at 4 weeks, and incubated overnight in 25 mM glucose, followed by 1-hour static stimulation using 2.8 or 20 mM glucose. Insulin in the media was measured. FIG. 6A shows mouse insulin was released from the explants, in a glucose concentration (and time) dependent manner. Using immunostaining it was confirmed that the encapsulated MIN6 cells were producing insulin. As shown in FIG. 6B-D, the encapsulated pseudoislets stained positively for insulin.

EXAMPLE 2

(46) PCL Scaffold Associated Insulin Cells Produce and Release Insulin

(47) Diabetic NOD/SCID mice were implanted subcutaneously with either a sham implant or PCL scaffold containing MIN6 clusters (as previously described in Example 1). At termination of experiment, blood was collected from mice and plasma extracted to measure circulating concentration of mouse insulin using an ELISA (Mercodia, Uppsala, Sweden). FIG. 7A shows trace amounts of endogenous mouse insulin in sham mice, but significantly higher concentrations of insulin in animals implanted with PCL scaffold containing MIN6 cells.

(48) The implants were taken out from animals and cultured for 16 hours in serum free DMEM at 37° C., 5% CO.sub.2 To determine if the implants secreted insulin, the media was assayed for mouse insulin. As shown in FIG. 7B sham implants secreted no insulin, whereas PCL scaffold implants containing MIN6 cells in the device secreted insulin at a mean rate of 0.35 ng/hr.

(49) Implants Produce Insulin in Immune Competent BALB/c Mice

(50) We tested if the scaffold-associated insulin-producing MIN-6 cells implant could function in BALB/c mice that have a normal immune system (immune competent) and with a major histocompatibility difference from the cells (H-2Kd and H-2Kd respectively). BALB/c mice were made diabetic using streptozotocin (4×75 mg/kg) and either a sham implant or the scaffold-associated MIN6 cell implant (approximately 50 MIN6 clusters in 24 microcapsules) was implanted subcutaneously. Blood glucose level (BGL) and body weights were monitored and compared to healthy non-diabetic control mice (FIG. 8, green line). FIG. 8A shows, in healthy mice mean blood glucose remained ˜8 mmol/L throughout experimental duration. In diabetic mice BGLs were >25 mmol/L on the day of implantation. In mice with sham implants BGL remained high (black line), but those implanted with scaffolds containing MIN6 cells (blue line) BGL was gradually lowered to normal levels between 33 and 39 days post-implantation.

(51) All mice had similar body weights at start of experiment and showed increase in body weight during experimental duration (FIG. 8B). Weight gain was highest in non-diabetic control mice (green), and least in diabetic mice with sham implant (black). In diabetic mice implanted with MIN6 cells (blue) weight gain was greater compared to sham.

(52) Similar to the NOD/SCID mice, elevated levels of insulin in plasma of mice receiving treatment was measured (FIG. 9A) and confirmed ex vivo presence of insulin within encapsulated cells of the implant (FIG. 9B).

(53) PCL Scaffold Facilitates Vascularisation in BALB/c Mice

(54) Blood vessels were clearly visible macroscopically on implants on the side proximal (FIG. 10A) and on the side distal to the skin (FIG. 10B). Blood vessels (red) within the scaffold was visualised by multiphoton intravital microscopy (FIG. 10C). Dextran (70 kDa)-Texas Red conjugated and DAPI were injected into mice via the tail vein and device visualised after 1 hour.

(55) PCL Scaffold Associated Cells have Reduced Host Cell Response

(56) To examine FBR (foreign body reaction) ex vivo implants were used. Data is shown from immune-competent BALB/c mice. Histology of the implants was studied by haematoxylin and eosin (H&E) stain, which showed abundant cellular infiltration into the device (FIG. 11A). Often clusters of cells typical of inflammatory foci were seen along or adjacent to PCL scaffolds (FIG. 11A arrows), though these foci were not seen around the micro-capsules (FIG. 11A*,B). NF-kB p65 immuno-stain was used to confirm pro-inflammatory cells were predominantly localised to the scaffolds (FIG. 11C). These inflammatory infiltrates were comprised mainly of neutrophils and monocytes/macrophages (FIG. 11D,E respectively). Occasionally, small numbers of B-lymphocytes (FIG. 11F) were observed, but CD4+ T-lymphocytes were undetectable (not shown). Consistent with the histology and NF-kB p65 expression, the cell specific markers were localised to the scaffolds. Sirius red stain on device shows ECM deposition of collagen (pink) (G) interconnecting the strands of the scaffold, and (H) around microcapsules. (I) α-SMA+ myofibroblasts showing area of active fibrogenesis around microcapsules and in the scaffold.

EXAMPLE 3

(57) An implant was designed to include a barrier region with pore size 0.5-20 μm to further stop encapsulated cells from leaving the centre region after implantation. The barrier region may be manufactured from the same or different materials as the rest of the scaffold using MEW by depositing the fibres with narrow spacing (<100 μm) in a geometric pattern to allow for the 0.5-20 μm porosity. Alternatively, a barrier region in the form of a membrane may be commercially available or one may be made using established techniques (e.g. fibre weaving, ion beam or particle ablation, templating, phase separation). Examples of scaffolds comprising barrier regions are shown in FIG. 12.