Implant and method for producing an implant

Abstract

The invention relates to an implant derived from intestinal tissue and to an improved method for producing an implant from intestinal tissue. According to the invention, intestinal tissue comprising a tubular segment of intestine with at least part of its associated vasculature intact is processed by: perfusing the vasculature through a vessel thereof with at least one decellularizing medium; and separately perfusing the tubular segment of intestine through its lumen with at least one decellularizing medium. The new method greatly improves tissue processing as compared with the methods previously known in the art, in which decellularization solutions were perfused into the tissue via the vasculature only, relying on diffusion to decellularize and purify the small bowel tissue. In contrast, the invention provides for separate decellularization and purification protocols for the vascular and tubular intestinal components, which are designed to optimize decellularization and purification of each of these tissue structures while preserving certain tissue-specific three-dimensional structures. The resulting scaffold is substantially decellularized and provides an excellent implant for repair and regeneration of bowel tissue.

Claims

1. A method for producing an implant from intestinal tissue comprising a tubular segment of intestine with at least part of its associated vasculature intact, the method comprising: perfusing the vasculature through a vessel thereof with a first perfusion protocol, wherein the first perfusion protocol comprises perfusing the vasculature with a first series of decellularizing media for a first perfusion time period, wherein the first series of decellularizing media includes at least three decellularizing media comprising a first media including at least one detergent, a second media including at least one protease, and a third media including at least one DNase, and wherein the vasculature is washed in between each decellularizing medium with a first washing medium; separately perfusing the tubular segment of intestine through its lumen with a second perfusion protocol, wherein the second perfusion protocol comprises perfusing the tubular segment with a second series of decellularizing media for a second perfusion time period, wherein the second series comprises at least one detergent, at least one protease and at least one DNase, and wherein the tubular segment is washed in between each decellularizing medium with a second washing medium that is the same or different than the first washing medium; washing the implant after the second perfusion protocol with the first washing medium and/or the second washing medium; wherein perfusion of the vasculature is carried out according to a first perfusion protocol and perfusion of the tubular segment of intestine is carried out according to a second, different protocol; wherein the first and second perfusion protocols differ in at least one of: a) the detergent, b) the protease, c) the DNase, d) the second perfusion protocol comprises a greater number of perfusion steps comprising detergent, protease or DNase, than the first perfusion protocol, e) the second perfusion protocol is performed for a longer period than the first perfusion protocol, or f) combinations thereof.

2. A method according to claim 1, wherein the vasculature is perfused through a main artery thereof and wherein the at least one decellularising medium of the series of media perfusing the vasculature leaves the vasculature via a main vein.

3. A method according to claim 1, wherein cells are substantially removed from the intestinal tissue.

4. A method according to claim 1, wherein perfusion of the vasculature, perfusion of the tubular segment, or both, is carried out using serial perfusions of solutions of sodium dodecyl sulfate (SDS), trypsin and DNase.

5. A method according to claim 1, wherein any one or more of the following structures are at least partially retained in the implant: mucosal layer of the tubular component, submucosal layer of the tubular component, serosal layer of the tubular component, circular and/or longitudinal muscular layers of the tubular component, tunica adventitia of the vasculature, tunica media of the vasculature, or tunica intima of the vasculature.

6. A method according to claim 1, wherein the implant comprises an internal elastic lamella and an elastin layer of the internal elastic lamella is substantially preserved and forms an internal, luminal, surface in the vascular portion of the implant.

7. A method according to claim 1, wherein an input means and/or an output means are attached to a proximal end and/or a distal end of the tubular segment of intestine to facilitate perfusion of the tubular segment of intestine.

8. A method according to claim 1, wherein the washing media are free of any decellularizing agents.

Description

DETAILED DESCRIPTION OF THE DRAWINGS

(1) Embodiments of the present invention will now be further described with reference to the following non-limiting examples and accompanying figures in which:

(2) FIG. 1 is a diagrammatic representation of the anatomy of the small intestine and its associated vasculature;

(3) FIG. 2 is a schematic diagram of a perfusion system according to the present invention;

(4) FIG. 3 is a photograph showing a tubular segement of intestinal tissue with intact vasculature before treatment;

(5) FIG. 4 is photograph showing a representative implant according to the present invention, produced from the starting material shown in FIG. 3.

(6) FIG. 5A is a photomicrograph (40 final magnification) showing a section of a representative implant according to the present invention, stained with Picro Sirius Red and viewed under polarised light.

(7) FIG. 5B is a photomicrograph (100 final magnification) showing a section of a representative implant according to the present invention, stained with Picro-sirius red and Miller's elastin. The double arrow shows the mucosal layer and the small arrows point to preserved elastin within small-calibre submucosal vessels.

(8) FIG. 6A is a photograph showing a representative implant according to the present invention 1 hour post-implantation into a porcine recipient. The arrow heads point to clot-free mesenteric vessels.

(9) FIG. 6B is a photograph showing an alternative view of the implant of FIG. 6A. The arrows point to intestinal micro-connections with distinguishable decellularised vessels.

(10) FIG. 7A is a photomicrograph (20 final magnification) showing a section of a representative implant according to the present invention, stained with haematoxylin and eosin 1 hour post-implantation into a porcine recipient. This longitudinal section shows mesenteric vein and artery with clot-free lumens.

(11) FIG. 7B is a photomicrograph (20 final magnification) showing a section of a representative implant according to the present invention, stained with haematoxylin and eosin 24 hours post-implantation into a porcine recipient. This cross section of the mesenteric arcade shows clot-free vessels (dark arrows) and inflammatory cells infiltrating the outer parts of the graft (light arrows).

(12) FIG. 8A is a photomicrograph (100 final magnification) showing a section of a representative implant according to the present invention, stained with haematoxylin and eosin 1 hour post-implantation into a porcine recipient. This longitudinal section of the arterial anastomotie site of the graft shows host cells infiltrating the decellularised superior mesenteric artery (see arrows).

(13) FIG. 8B is a photomicrograph (200 final magnification) showing a section of a representative implant according to the present invention, stained with haematoxylin and eosin 24 hours post-implantation into a porcine recipient. The arrows point to CD133+ cells lining lumen of the decellularised artery of the graft.

DETAILED DESCRIPTION

(14) The small intestine and its associated vascular structures (10) are depicted in FIG. 1. The small intestine is a tubular structure (11) with a central lumen. The tubular structure is comprised of an outer serosal layer (12) covering two muscle layers comprising a longitudinal (13) and a circular (14) muscle layer. Under the muscle layers is the bowel submucosa (15) which is covered (in the internal aspect of the bowel lumen) by a mucosal layer (16) whose microstructure is comprised of villi, microvilli and crypts. The small intestine also has a blood supply comprising feeding (arterial) and draining (venous) vessels forming a vascular tree (20) ultimately becoming integral and associated with the small intestine tissue in the form of feeding arterioles and capillaries (arterial supply side) and draining capillaries and venules (venous return side).

(15) On the arterial (feeding) side, a feeding artery (21) has multiple sub-arterial branches (22) feeding various parts of the small intestine. The arteries and branches thereof are located in the vascular pedicle and surrounded and supported by connective tissue (the mesentery) (23). The arterial branches feed the small intestine associated/integral arterioles and capillaries (24). These vessels are located on and within the small intestine tissue and serve to perfuse the small intestine tissue with blood and other fluids and cells. Blood and fluids drain from the small intestine via integral venous capillaries and venules (25) which are located on and within the small intestine tissue. These vessels subsequently drain or return blood and fluids to the main venous return (vein) (26) via multiple venous branches (vein branches from the main draining vein) (27).

EXAMPLES

(16) All reagents were purchased from Sigma-Aldrich (UK) or Fisher Scientific (UK).

(17) All animals were maintained and handled in accordance with the Animals Scientific Procedures Act 1986 and studies were performed following guidelines stipulated and licensed by the UK Home Office. Experiments were performed using Large White Landrace crossbreed pigs. All animals were kepy under standard laboratory conditions and fed a commercial pelleted diet.

1. Preparation of Acellular Scaffold

(18) (a) Harvesting of Porcine Intestinal Tissue

(19) Following intravenous administration of heparin (7,000 U) in a 60-65 kg Large White Landrace crossbreed pig, a midline incision was created and segment of ileum was isolated together with its attached vascular supply approximately 20-30 cm in length. Residual lymph nodes in the mesentery were dissected; keeping the whole specimen leak proof; the distal end of the pedicle was tied off and isolated. The proximal end of the artery was cannulated with a 14 G Radiopaque I.V. cannula and flushed with 0.9% NaCl containing heparin (2,000 U) until there was no blood in the outflow at the proximal end of the vein. To maintain intraluminal pressure within the vasculature and maintain patency, a pulsatile pump perfusing heparinised (2000 U/I) 0.9% NaCl through the pedicle was attached to the arterial cannula. All vessels not involved in supplying the small intestine that was to be explanted were tied off, and the whole explant was removed from the animal. At the end of the procedure, the animal was terminated by intravenous overdose of anaesthesia. The vascular circuit was checked for any leakage and the intestinal lumen was flushed with 1 liter of 0.9% NaCl. At both ends of the ileal segment, silicone tubes were attached with purse-string sutures, and the segment of small intestine was attached to the second perfusion circuit. The harvested specimen was then ready for decellularisation. FIG. 3 shows a representative specimen prepared as described above. The bowel segment has been filled with saline and clamped at either end, aiding visualisation of the bowel structure, vascular system and fine capillaries overlying the serosal surface. The mesenteric arcade, in which the feeder arterial vessels and return venous vessels can clearly be seen, forms a fan-like structure in the inner portion of the bowel segment loop. The bowel segment has a pink appearance due to presence of cells and blood in the vessels and tissues.

(20) (b) Decellularisation of the Porcine Intestinal Tissue

(21) Immediately after explantation, two separate perfusion circuits were set up for decellularisation of the vascular component and the tubular intestinal component.

(22) As shown in schematic form in FIG. 2, a perfusion system (100) was set up for processing the intestinal tissue (101) comprising a vascular component (102) and a tubular intestinal component (103). Circuit I was attached via a cannula into the main artery (104) of the mesenteric arcade and exited via the main vein (105) and circuit II was attached to the silicone tubes tied to each end of the explanted tubular segment of ileum (103). Pumps (106, 107 respectively) were used to pump decellularising media separately around circuits I and II, the arrows representing the direction of flow.

(23) The decellularisation protocol used is presented in Table 1:

(24) TABLE-US-00001 TABLE 1 CIRCUIT I AND II Solution Time Temperature 1. 0.075% SDS solution 90 min 25 C. 2. Antibiotics solution 3 changes; 25 C. 15 min each 3. 0.05% Trypsin solution 90 min 37 C. 4. Antibiotics solution 3 changes; 25 C. 15 min each 5. DNase I solution 120 min 37 C. 6. Antibiotics solution Overnight 4 C. 7. 0.075% SDS solution 90 min 25 C. 8. Antibiotics solution 3 changes; 25 C. 15 min each 9. 0.05% Trypsin solution 90 min 37 C. 10. Antibiotics solution 3 changes; 25 C. 15 min each Circuit I 11. Antibiotics solution Till the end 25-37 C. of process Circuit II 11. 0.075% SDS solution 90 min 25 C. 12. Antibiotics solution 3 changes; 25 C. 15 min each 13. 0.05% Trypsin solution 90 min 37 C. 14. Antibiotics solution 3 changes; 25 C. 15 min each

(25) The intestinal tissue was submerged in the solutions used for perfusion through circuit I. Perfusion was conducted using a Watson Marlow 323S/D Pump (UK) at the rate of 30-60 rpm throughout the whole process. The cells, cytoplasmic components and finally deoxyribonucleic acids were broken down by serial perfusions of SDS, trypsin and DNase I solutions washed in between each step with PBS supplied with antibiotics. After the last stage of the decellularisation, the cannula was detached from the artery and the silicone tubes from the ileum. The acellular scaffold was sterilized by placing it in a solution of 0.1% peracetic acid in PBS on a horizontal shaker for three hours at room temperature. Subsequently, the specimen was washed in sterile PBS supplemented with antibiotics, and stored in sterile PBS supplied with antibiotics till further use.

(26) The resulting decellularised implant is shown in FIG. 4. The decellularised bowel implant has an off-white, semi-translucent appearance following removal of cells and blood. The tubular bowel component is shown clamped and filled with saline solution, allowing better observation of the preserved bowel structure, mesenteric arcade and the fine capillaries overlying the serosal surface. The mesenteric arcade forms a fan-like structure in the inner portion of the bowel segment loop. The feeding arterial vessel and venous return vessel are cannulated and clearly visible.

(27) (c) Histological and Molecular Analysis of Acellular Scaffold

(28) Three representative acellular scaffolds were subjected to histological and molecular analysis for the presence of nuclear material. Intestinal tissue from animals of the same age and weight were used for comparative purpose; samples from the entire length of the de-cellularised ileal and mesenteric arcade scaffold were fixed in 10% NBF and processed for wax embedding using routine laboratory procedures. Sections were cut at 5 m and stained with Haematoxylin and Eosin (H&E) and Picro-Sirius red with Miller's elastin (PSR-ME).

(29) 0.025 g from each scaffold was used for DNA isolation using GenElute Mammalian Genomic DNA Miniprep Kit (Sigma-Aldrich, UK) according to the manufacturer's instructions. The amount of DNA residual in the vascular and intestinal part of the scaffold-construct as well as DNA isolated from native ileum and associated vessels were quantified using spectrophotometer (AD=260). Electrophoretic separation of DNA was conducted in order to visualise the differences in the presence of DNA in the samples of acellular scaffold and native tissues. DNA was separated on 1% agarose gel in Tris-acetate Ethylendiaminetetraacetic Acid (TAE) buffer for 2 h at 100V. Subsequently, the gel was stained with ethidium bromide solution for 15 min.

(30) The quantity of GAGs in the acellular scaffolds was assessed using Blyscan Sulphated Glycosaminoglycan Assay (Biocolor; UK) according to the manufacturer's instructions. The assay was used to calculate the total sulphated glycans content according to spectrophotometric (650 nm) absorbance values of sulphation level (dye-binding capability) as calculated using the supplied assay standards (range 2-50 g/ml) of GAGs. The GAGs were isolated from three different batches of specimen, and the concentration was calculated using standard curve. The results are presented as g/g of wet weight. Values obtained for the acellular scaffold were compared to values for control porcine intestinal tissue.

(31) Histological analysis of the decellularised scaffold showed no intact cells in either the vascular mesenteric arcade or the ileal tissue. Analysis of the tissue under polarised light on PSR-ME stained sections showed good vascular preservation of collagen and elastin, with mainly yellow-to-orange collagen fibres of the submucosa and thin, green fibres of the mucosal layer appearing similar to native intestine (see FIG. 5A). Staining with H&E confirmed that elastin was preserved even within the small-calibre submucosal vessels (see FIG. 5B).

(32) Although some residual DNA was detected in both the vascular and intestinal part of the decellularised scaffold, it represented a significant decrease when compared with native tissue. The total amount of DNA found within the tubular intestinal component of the decellularised scaffold constituted only 0.75% of total DNA found in native tissue, whereas the decellularised vasculature had less than 1% of the total DNA found in native blood vessels (Table 2). There was decrease in the amount of GAGs remaining after the de-cellularisation process; overall there was a 42% retention of functional GAGs within the decellularised scaffold (Table 2).

(33) TABLE-US-00002 TABLE 2 DNA GAG's (standard (standard deviation) deviation) Specimen [ug/mg] [ug/g] Native ileal tubular 0.74 3.1 component (0.17) (0.05) Native vascular 0.36 2.9 component (0.08) (0.03) Decellularised ileal 0.005 (0.005) tubular component (0.002) Decellularised vascular 0.004 0.9 component (0.002) (0.03)

3. Implantation of the Acellular Scaffold into Porcine Recipient

(34) (a) Surgical Procedure

(35) Large White Landrace crossbreed pigs (n=7), (55-75 kg) underwent a right-sided nephrectomy via a midline incision. The vascular component (feeding artery and draining vein of ileal pedicle) of the scaffold was anastomosed in an end-to-end fashion to the renal artery and vein. The anastomotic site as well as the mesenteric vasculature of the implanted scaffold construct was checked for any bleeding. Mersilk ties were used to stem excess blood loss. Once blood was observed perfusing through the mesenteric arcade and the fine capillary network encasing the decellularised ileum, the entire scaffold implant was placed into the kidney cavity for one hour and the animal's physiological signs monitored. Thereafter, the animal was terminated with a lethal injection of sodium pentobarbitone (100 mg/kg).

(36) One scaffold was implanted for a recovery procedure for 24 hours. The scaffold was placed into the kidney cavity and the abdominal muscle layers were closed using intramuscular suture (3.0 Vicryl) and skin using horizontal mattress suture (3.0 Mersilk). Post-operatively, the animal received analgesia using intravenous Carprofen (4 mg/kg body weight). After 24 hours the animal was terminated and the graft immediately harvested.

(37) Prior to surgical implantation of the scaffold, an anti-coagulation protocol was applied. This comprised pre-treatment of the animal with anti-thrombotic drugs prior and during the surgical procedure as well as pre-conditioning the vascular part of the implant by injecting it with either neat or dissolved heparin sodium in saline solution, as shown in Table 3.

(38) TABLE-US-00003 TABLE 3 Additional anti- Heparin Total heparin coagulative/ Heparin sodium sodium applied sodium platelet- administered into scaffold received by formation Procedure systemically prior to animal/kg/h drugs received Animal (termimal/ Weight of during implantation of surgical by animal no. recovery) pig [kg] implantation [U] [U] procedure [U/kg/h] 1. Teminal 55 1. Post-incision: 1,000 55 7,000 2. Prior to releasing the vascular clamps: 1,000 2. Terminal 55 1. Post-incision: 1,000 51 7,000 2. Prior to releasing the vascular clamps: 2,000 3. Terminal 73 1. Post-incision: 1,000 82 1. Warfarin: 7,000 12 mg/day 2. Prior to releasing starting 3 days the vascular clamps: prior to surgery 2,000 3. Administered into graft's feeding artery: 2,000 4. Terminal 70 1. Post-incision: 30,000 260 1. Warfarin: 10,000 12 mg/day 2. Prior to releasing starting 3 days the vascular clamps: prior to surgery 10,000 5. Terminal 70 1. Post-incision: 30,000 260 1. Warfarin: 10,000 12 mg/day 2. Prior to releasing starting 3 days the vascular clamps: prior to surgery 10,000 6. Terminal 75 1. Post-incision: 35,000 298 1. Warfarin: 10,000 12 mg/day 2. Prior to releasing starting 2 days the vascular clamps: prior to surgery 10,000 2. Clexane: 120 mg on the day of surgery 7. Recovery 55 1. Post-incision: 30,000 255 1. Warfarin: (24 hours) 10,000 12 mg/day 2. Prior to releasing starting 3 days the vascular clamps: prior to surgery 2,000

(39) Successful end-to-end anastomosis of the de-cellularised artery and vein to the renal artery and vein of the recipient animal was achieved in all animals. In five out of seven implants, complete reperfusion of the scaffold (including vascular micro-connections within the decellularised bowel with the host's blood) was obtained within one hour of implantation (see FIG. 6), and the scaffolds remained patent for the duration of the experiment (Table 5). The seventh graft left in vivo for 24 hours was patent at the time of explantation.

(40) FIG. 6A shows the implant 1 hour post-implantation. The perfused bowel-derived segment has a healthy pink appearance typical of vital, perfused tissues. The mesenteric arcade shows good perfusion, as does the bowel implant. The arrow heads point to clot-free mesenteric vessels and complete perfusion of the microvessels on the serosal surface of the bowel segment is visible.

(41) FIG. 6B provides an alternative view of the same implant 1 hour post-implantation, at a higher magnification than that depicted in FIG. 6A. The arrows point to bowel micro-connections with distinguishable decellularised vessels. The image shows clot free and patent vascular system with arteries and veins clearly visible and perfused with the recipient's blood. The capillaries on the serosal surface are also clearly visible, perfused and structurally intact following decellularisation as evidenced by no leakage of blood. The perfused bowel segment has a healthy pink appearance typical of vital, perfused tissues.

(42) Successful reperfusion of the grafts with systemic blood was obtained in five out of the seven cases. In the two unsuccessful cases, reperfusion was prevented by clotting of the mesenteric vessels, which took place shortly after introducing the systemic blood to the scaffold. The most successful protocol combined oral pre-medication of the animals with 12 mg/day of Warfarin, pre-conditioning of the scaffold construct with 30.000 U of neat sodium heparin and intravenous infusion of 12,000-20,000 U of sodium heparin. Animals with a total heparin intake (calculated based on body weight and the length of the surgical procedure) of 255-260 U/kg/h were anticoagulated best. However, animals receiving 55 U/kg/h also had a satisfactory outcome with complete reperfusion and patency of the decellularised vessels, as shown in Table 4:

(43) TABLE-US-00004 TABLE 4 Outcome of the implantation Animal Successful Re-perfusion with Clot Excessive no. anastomosis blood formation bleeding 1. + + 2. + + 3. + + 4. + + 5. + + 6. + + +++ 7. + + +

(44) The length of the decellularised feeding artery and draining vein also contributed to whether reperfusion was likely or not, since scaffolds with shorter vessels performed better (see Table 5).

(45) TABLE-US-00005 TABLE 5 Total heparin recieved Length of the Average by animal pedicle length of regarding length Animal [cm] implanted of implanted vessels no. Artery Vein vessels [U/kg/h/cm] 1. 1 0.5 0.75 73 2. 3.0 2.5 2.75 30 3. 3.5 2.5 3 17 4. 3 2.35 2.675 30 5. 2 1.95 1.975 131 6. 1.9 1.9 1.9 137 7. 1.95 1.95 1.95 153 8. 2 1.85 1.925 132
(b) Histological and Immunohistochemical Analysis

(46) Following each implantation procedure, samples from the scaffold were removed, fixed and processed for routine histology. H&E stained sections were used to assess the degree of vascular leakage into the surrounding tissue, rate of perfusion and any signs of clotting within the blood vessels.

(47) Immuno-histochemical staining was incorporated in order to visualize any endothelial progenitor cells (CD133), smooth muscle actin (SMA), macrophages (CD68) and Von Willebrand Factor (vWF) within the grafts harvested 1 and 24 hours post-implantation. Prior to IHC staining all sections were de-waxed and rehydrated and any necessary antigen retrieval was carried out at this point. Endogenous peroxidise activity within the tissue was blocked using 3% hydrogen peroxidise in methanol for 30 minutes. Non-specific binding was prevented using normal horse serum (Impress Kit, Vector Laboratories, UK) for 30 minutes (CD68 SMA) and 2 hours (CD133, VWF). Sections where then incubated with the primary antibody diluted in PBS; for negative controls performed in parallel, only PBS was used. Sections were subsequently washed in PBS; 33 minute washes cycles and then incubated further with a biotinylated secondary antibody. This step was followed by Vectastain Elite ABC kit (Vector Laboratories, UK) and 33-minute cycles with PBS (CD133). Visualization was achieved using 3,3-diaminobenzidine tetrachloride (DAB) substrate (Vector Laboratories. UK), and 5 minutes wash in distilled water. Finally, sections were counterstained with Harris haematoxylin for 1 minute, dehydrated, cleared in xylene and cover slipped.

(48) All continuous data was expressed as the mean (n=3)Standard Error of the Mean (SEM) and p<0.05 was taken as significant. p Values were estimated using one-way ANOVA, and all computations were performed using GraphPad Prism 4 software.

(49) Histological analysis indicated that in five out of the seven implants (four short and one long-term study) the majority of vessels (including intestinal small diameter vessels and capillaries) were patent and contained non-coagulated blood (see FIG. 7A, B). Miller's elastin staining indicated that clotting was mostly triggered in places where the vascular wall or internal elastin was interrupted.

(50) As early as one hour post-implantation host cells were present in the anastomosed, acellular vessels of the scaffolds (FIG. 8A). The rate of cellular infiltration was reduced in the distal parts of the graft. Nevertheless, cells were present on the luminal surface of the mesenteric vessels. The rate of cellular infiltration of the scaffold ECM was very high 24 hours post-implantation, but no signs of intensive foreign body reaction could be characterised.

(51) The majority of cells infiltrating from the anastomotic side of the arteries at one hour were CD68+, however, 24 hours later more macrophages were present in the distal than in the proximal parts of the implanted vessels. Most of the cells present in the main artery (but not the vein) of the graft implanted for 24 hours were CD68. There were more macrophages within the clot-free parts of the scaffold-constructs. Few CD68+ cells were present in the host's renal vessels. Some of the cells lining the anastomosed artery were CD133+ as early as one hour post-implantaton, and their number increased within 24 hours (see FIG. 8B). Sections of anastomotic site showed similar amount of VWF+ staining in the donor and recipient main pedicle. However, no staining was detected when un-implanted decellularised main artery and vein were analyzed. No SMA+ cells could be seen in any of the implants.

(52) It is of course to be understood that the invention is not intended to be restricted by the details of the above specific embodiments, which are provided by way of example only.