To investigate the behaviour of donor myoblasts at the vesico-ureteric junction (VUJ) and to evaluate their potential as an autologous bulking agent, as myoblast transplantation has been shown to regenerate damaged or degenerated tissue, and it was postulated that they could be used to treat vesico-ureteric reflux.
MATERIALS AND METHODS
Muscle biopsies were obtained from the lower limb muscles of 10 pigs. The quality of the cells was evaluated by electrophysiological and immunohistochemical tests. The cell membranes of myoblasts were labelled with PKH26, a fluorescent dye. Six weeks after taking of the muscle biopsies all pigs underwent cell transplantation; 30 × 106 cells suspended in transplantation medium (in 1-mL syringes) were injected at the VUJ, into the proximal urethra and the rhabdosphincter. At the VUJ volumes of 1 mL were injected, whereas in the urethra and rhabdosphincter small cell depots (0.1 mL) were injected. All the pigs were killed 8 weeks later, and the myoblasts and newly formed myofibres were identified using fluorescence microscopy, with a histological evaluation and investigation of potential local inflammatory reaction.
Two to three intact layers of autologous myoblasts were found in the outer aspects of the large cell depots in the VUJ. Immunohistochemistry further showed that the myoblasts were only viable at these outermost borders of the large bulking areas, whereas necrosis with red fluorescent cell detritus was visible in the remaining central aspects of the large bulk of cells. By contrast, cells survived and formed myotubes in the wall of the proximal urethra and the rhabdosphincter where the small cell depots had been injected.
In small depots, transplanted autologous myoblasts can survive and differentiate into myofibres, while in a large bulk of cells the vast majority of cells become necrotic. The present results show that myoblasts cannot be used for augmentation of large volumes of tissue or as a bulking agent.
The aim of modern minimally invasive strategies for treating VUR in children is to reduce the diagnostic and therapeutic comorbidity. In recent years endoscopic therapy of VUR has gained widespread popularity due to its technical simplicity, short anaesthesia time and good efficiency rates, particularly in children with low-grade VUR . In many paediatric urology centres the injection of non-animal stabilized hyaluronic acid suspended in dextranomer gel as a bulking agent is considered as the first option in treating VUR in children . One of the proposed mechanisms by which this endoscopic therapy might lead to good success rates is improved fixation of the ureter to the vesical trigone, and creation of a solid support dorsal to the intravesical portion of the ureter. Additionally, coaptation of the vesico-ureteric junction (VUJ) and distal portion of the ureter during the storage phase and during high-pressure micturition has been postulated to be responsible for the successful therapy of VUR. In the last few years the development of cell-culture techniques and biodegradable substances have paved the way for the use of autologous cells in treating urological diseases . Faced with the problem of biocompatibility, risks of particle migration, granuloma formation or long-term stability of artificial bulking agents, it was postulated that autologous cells would be particularly suitable as a bulking agent. Autologous cells would potentially provide an ideal material for new tissue-engineering techniques for treating VUR, as they would be biocompatible, non-immunological and non-allergenic. Clinical trials have shown that autologous cell transplantation can be used to create new functional and non-immunogenic tissue . In addition, it was shown that autologous cells can be used as a bulking agent for a short period .
The endoscopic therapy of VUR with mammalian autologous injectable chondrocytes was reportedly successful to some degree . Autologous chondrocytes, which can form viable cartilage, have the potential advantage that they form a stable submucosal bulk of tissue at the VUJ . Overall success rates after one or two injections of chondrocyte suspensions harvested from articular cartilage were 79% for ureters after 3 months ; at 1 year after injection there was persistent correction of VUR in 70% of ureters and 65% of patients . Based on the primary low success rate of 60% after one injection, the original mean volume of 0.91 mL was increased, creating a ‘volcano-like’ endoscopic appearance at the VUJ. However, a histological evaluation of the excised injected bulk in four patients with persistent reflux, and hence had open ureteric reimplantation, revealed only calcified alginate with no evidence of viable cells. A causal relationship between the amount and volume of injected chondrocytes and the success rates has not been established, and the potential influence of injected volumes has not been investigated.
Myoblast transplantation has been shown to regenerate damaged and degenerated tissue. The efficiency and tolerability of ultrasonography-guided transurethral injections of autologous myoblasts and fibroblasts in the treatment of urinary incontinence were reported recently . It was shown that small depots of cells can be injected very precisely. The injected cells integrate into the rhabdosphincter and the urethral submucosa, and increase the thickness and contractility of the rhabdosphincter, the striated urinary sphincter. Autologous myoblasts (or satellite cells) obtained from skeletal muscle are muscle progenitor cells. They can restore damaged muscular integrity of striated muscles. Therefore, the question is whether these cells could be used as an autologous bulking agent for treating VUR at the VUJ. Potentially, they could be injected in large cell depots to create a bolus that increases the submucosal length of the ureter, and might also act as a fixation point. Myoblasts obtained from skeletal muscle biopsies can be cultured routinely with no problems. Therefore, it is possible to isolate and proliferate muscle-derived progenitor cell cultures which consist of viable, non-fibroblast, desmin-expressing myoblasts .
To date, the viability and physiological behaviour of donor myoblasts injected at the VUJ have not been investigated. In the present study the survival of muscle-derived progenitor cells after transplantation into the urinary bladder was explored in a pig model, the main aim being to investigate if autologous myoblasts can be used as an injectable bulking compound for treating VUR.
MATERIALS AND METHODS
The experiments on pigs were conducted after approval by the Federal Ministry for Education, Science and Culture of the Republic of Austria. Good Laboratory Practice conditions and state-of-the-art cell-culture techniques were used to ensure the sterility and reproducibility of the cell cultures.
Primary skeletal myoblasts were isolated from the lower limb of 10 Tyrolean domestic pigs. The muscle biopsies were placed into MEM-Medium (Invitrogen, Paisley, UK) with gentamicin 5 mg/L. The muscle tissue was dissected from connective tissue, minced into small tissue pieces (2 mm) and washed in PBS. Muscle samples were enzymatically dissociated according to the cell dispersion technique described by Blau and Webster . To obtain monoclonal cell lines, single cells were selected from the primary cell cultures using Celltram Vario (Eppendorf, Hamburg, Germany) and transferred into gelatine-coated wells (96-well plate) containing proliferation medium (MEM medium supplemented with 15% fetal calf serum and 0.1 µg/mL gentamicin). The cells were cultured in growth media and maintained in a proliferating state. After the myoblasts had reached confluence, the cultures were re-plated at a lower density. Desmin was used as a marker to identify clones of myoblasts.
The myoblasts were labelled with PKH26 fluorescence dye (Sigma, Austria) according to the manufacturer’s instruction. The cells were washed with PBS, centrifuged, and 250 µL of Diluent C (Sigma) were subsequently added to the cell suspension; 15 µL of the PKH dye was then added to 250 µL of Diluent C. After a 6-min incubation at room temperature, fresh culture medium was added to stop further incorporation of the fluorescent dye into the cell membranes. The cell suspension was washed and centrifuged three times.
At 6 weeks after taking of the muscle biopsies, all pigs were again placed under general anaesthesia. The urinary bladder was incised to expose the trigone with both ureteric orifices and the bladder neck. Cells were transplanted (30 × 106 cells suspended in transplantation medium, loaded in 1-mL syringes) at both VUJ and in the proximal urethra (rhabdosphincter). A stay suture was placed distal to the ureteric orifice to provide stabilization of the VUJ during the injection with the cultured myoblasts (Fig. 1). The volume of the injected cell suspension that was placed beneath the ureteric orifice to achieve coaptation was 1 mL. In the posterior urethra cell depots of 0.1 mL with equal cell concentrations were injected submucosally and into the rhabdosphincter at five different sites.
At 8 weeks after injecting the cells, the bladder and urethra were removed in all pigs to investigate the sites where the cell depots had been injected. The histological evaluation focused on the survival and differentiation of the injected cells and tissue, as well as the tissue reaction at the donor site. The specimens were embedded in paraffin (15–18 paraffin blocks per pig), and 12 serial sections (6 µm) were obtained from each block. Sections were dried overnight, fixed in 5% paraformaldehyde and stabilized in 1% Tween. Histological sections were examined by fluorescence microscopy, and the co-localization of transplanted myoblasts with immunohistochemistry and trichrome-Masson-Goldner staining. Myogenicity was confirmed by immunohistochemical staining for desmin. The injected donor cells were identified using fluorescence microscopy to visualize the PKH26 membrane staining.
The histological evaluation of the posterior urethra showed engraftment of the cells at the different sites of injection in the wall of the proximal urethra and the rhabdosphincter in all pigs. Survival, integration of the injected myoblasts into existing striated muscle groups, and the formation of multinucleated myotubes was shown by immunofluorescence and the standard histological examination after injection of myoblasts into the rhabdosphincter (Fig. 2). Positivity of immunocytochemical staining for desmin confirmed the myogenicity of these cells. Highly magnified histological sections of the proximal urethra showed that mononucleated myoblasts, seen as eosinophilic cells with central nuclei, were present in the urethral wall. In the urethral wall the myoblasts did not fuse and form myofibres. No signs of inflammation, infection, or scar formation could be detected in the specimens.
By contrast with the findings in the urethral wall and the rhabdosphincter, viable cells were detected only at the border area of the cell depots at the submucosal ureteric orifices after injection of myoblasts (Fig. 3). Only two to three intact layers of autologous myoblasts derived could be found in histological sections at the periphery of the large bulking depots (Fig. 4). Immunohistochemistry showed the viability of the outermost transplanted myoblasts, by contrast with widespread central necrosis of the injected cells, apparent as red fluorescent cell detritus (Fig. 5). The loss of the myogenic phenotype in these bulk depots was further shown by scant or negative immunostaining for desmin in all pigs. Furthermore, there were no marked signs of an inflammatory response or fibrotic scars at the sites of injection.
The endoscopic therapy of VUR in children has gained widespread use as because is a minimally invasive procedure. The preferred bulking substances, dextranomer and hyaluronic acid, have been considered to be harmless agents for therapeutic use in humans. The search for alternative substances is based on autologous cells. These are biocompatible and provide the opportunity to regenerate congenitally damaged tissue by means of specific tissue engineering. It was shown that primary skeletal muscle-derived stem cells can differentiate into myotubes and myofibres in the smooth muscle layers of the bladder wall . In all of the present pigs the transplanted cells that had been injected in small cell depots into the posterior urethra and the striated rhabdosphincter survived. The myoblasts that had been injected into the rhabdosphincter fused and formed new myofibres, whereas the myoblasts remained as mononuclear cells in the urethral wall. These observations confirm studies which show that transplanted myoblasts are myogenic stem cells, and can survive and mediate the formation of new muscle tissue when the cells are injected into striated musculature . Myoblasts are characterized by their slow growth in tissue culture, and rapid differentiation and formation of myotubes after transplantation .
That the cells survived and started to differentiate myogenically supports the concept of the use of cultured muscle-derived stem cells as an injectable bio-implant for regenerating the rhabdosphincter. The most important finding of the present study was myoblasts survived and differentiated only when small cell depots (0.1 mL) were injected. The problem and the challenge in treating VUR children is to provide a solid volume underneath the mucosa of the ureteric orifice, to partly occlude and fix the VUJ. To achieve a sufficient bulk, significantly more autologous cells and a larger bulk of injectable agent were injected at each side of the ureteric orifices (1 mL containing 30 × 106 myoblasts). Histological evaluation of the VUJ in all pigs showed that the vast majority of injected cells in the centre underwent necrosis. Therefore, the main aspect of the submucosal bulk consisted only of necrotic cell detritus 8 weeks after implantation. Only at the outermost margins of the large cell bulk were viable autologous myoblasts identified.
In 2001 Caldamone and Diamond  published the 1-year results of injecting autologous chondrocytes in children treated for VUR. Four children who had persistent VUR had an open surgical ureteric reimplantation after implantation of chondrocytes. Histological examination of the excised surgical specimens showed calcified alginate with a granulomatous reaction, but no evidence of viable chondrocytes. Injected volumes of autologous chondrocytes were 0.91 mL for primary therapy, and up to 1.2 mL in repeated treatment, with 20 × 106 cells/mL . Although it is difficult to compare distinct functional forms of cells, the results were similar in the present study. In all pigs comparable concentrations of transplanted autologous cells and volumes of injectable agents were applied. As in the reported four children, it appears to be impossible for autologous cells to survive in a large cell bulk.
Similar experience was reported over the last decade with the use of autologous fat cells for augmentation of tissue and treatment of urinary incontinence. It was shown that injected fat undergoes necrosis, reabsorption and replacement with fibrotic tissue within 3 months after an injection of cells. Survival of the cells was shown to depend mostly on the vascularity of the host tissue, in particular microvascular density. The same effects were reported after injection of fat tissue for cosmetic regeneration in facial areas; therefore, repeated injections are typically required. Moreover, it was shown that the use of autologous lipo-injection for VUR in renal transplant patients was entirely inappropriate . Histological evaluation showed no neovascularization in the centre of this bulk and hence the injected cells in the centre of the bulk had no chance to survive, as was found in the present study. It must be assumed that injecting many cells and large volumes of cell depot results in very limited overall survival of cells and cell bulk .
In a recent study the functional and histological changes after myoblast injection in the porcine rhabdosphincter were evaluated. Into two different areas of the rhabdosphincter, different cell numbers were injected (total volume 1.5 mL) . In each area, 10 depots with volumes of 0.15 mL were injected along the rhabdosphincter. Histological examination of the specimens showed that the injected cells had survived at the injection site and had formed new myofibres. There was a statistically significantly greater urethral closure pressure, of up to >300%, in all cases where more cells were injected. Pressures were even decreased after injecting fewer cells. Therefore, the effects after applying myoblasts into the rhabdosphincter are dose-dependent. The small diameter of the injected cell depots provided the basis for survival of the injected cells, differentiation and regeneration of the rhabdosphincter. By contrast, the size of the injected cells depots was much larger in the present study.
The implications of the present data are far reaching; they strongly support earlier publications that myoblasts cannot be used as bulking agents or for forming large volumes of new muscle tissue, as the vast majority of cells do not survive the transplantation procedure when large volumes of injection are used. This finding is important for all therapeutic concepts using autologous myoblasts as a cellular therapeutic agent, e.g. for treating urinary incontinence, VUR, heart failure, myocardial infarction or muscular dystrophy.
In addition, it must be considered that the VUJ consists of smooth muscle cells and urothelium, and therefore does not represent an ideal site for injection of striated muscle myoblasts either. Instead, it would be interesting and more promising to cultivate smooth muscle cells. As the successful and routine cultivation of smooth muscle has not been possible so far, skeletal muscle myoblasts were used in the present study. Some of the injected cells showed distinct smooth muscle behaviour, including expression of the contractile filament smooth-muscle actin. These findings suggest that subpopulations of skeletal muscle myoblasts might have the potential to differentiate into smooth muscle cells . A recent report  showed the induction of pluripotent stem cells from adult human fibroblasts by defined factors, and therefore supports this hypothesis further. Another question that remains to be answered in further experiments is the fate of injected myoblasts after applying small depots at the VUJ. If the myoblasts are injected as small depots, they should be able to survive as they do after injecting small depots into the proximal urethra. Further studies will elucidate if injection of small depots of transplanted autologous myoblasts provides the basis for regeneration of the ureteric orifices in treatment of VUR.
In conclusion, myoblasts cannot be used as an autologous bulking agent at the VUJ. Despite the obvious advantage of an autologous material with no immunological reaction, applying a large cell bulk apparently does not enable the survival of the vast majority of cells. Only at the outermost borders of the cell depots can injected myoblasts survive. By contrast, myoblasts survive and differentiate into muscle tissue when they are injected in small depots. Therefore, myoblasts can be used for regenerating small muscular structures but not for creating a large bulk or for forming larger volumes of muscle tissue.
CONFLICT OF INTEREST
Hannes Strasser and Rainer Marksteiner are founders and co-owners of Innovacell Biotechnologie GmbH, where the autologous cells were grown. Wolfgang Schwaiger is an employee of Innovacell and was responsible for the cell cultures.