Polyester Fleeces used as an Artificial Interstitium Influence the Spatial Growth of Regenerating Tubules

In regenerative medicine the use of stem/progenitor cells is a valuable therapeutical option for the regeneration of diseased tissues and organs. However, the secure application of cell-based therapies for the treatment of renal failure requires exact information regarding the mechanisms of parenchyma development in combination with different kinds of biomaterials. Recently, we demonstrated that application of stem/progenitor cells in combination with a polyester fleece I-7 as an artificial interstitium supported the spatial generation of tubules during perfusion culture in chemically defined Iscove`s Modified Dulbecco’s Medium containing aldosterone (1x10-7M). In the present experiments we investigated if the use of different polyester fleeces (Posi-4 and Posi-5 in comparison to I-7) has any effect on generation, differentiation and spatial development of tubules. In consequence, immunohistochemistry, transmission electron microscopy and scanning electron microscopy were performed. Since the specimens were not coated by extracellular matrix proteins, unique insights in the contact zone between the basal lamina, interstitial cells and surrounding polyester fibers can be obtained. Analyzing the specimens developing in I-7, Posi-4 and Posi-5 polyester fleeces by means of immune histochemistry and transmission electron microscopy illuminated that no cell biological differences could be observed. In contrast, scanning electron microscopy of generated tubules demonstrated that a difference in spatial distribution and different diameters of tubules could be registered. It is concluded that tested I-7, Posi-4 and Posi-5 polyester fleeces appear as promising candidates to shelter stem/progenitor cells after implantation although they exhibit different spatial growth pattern. Citation: Glashauser A, Denk L, Minuth WW (2011) Polyester Fleeces used as an Artificial Interstitium Influence the Spatial Growth of Regenerating Tubules. J Tissue Sci Eng 2:105. doi:10.4172/2157-7552.1000105


Introduction
Since several years biomedical and clinical research is focusing on the use of stem/progenitor cells (s/pC) in renal injury and repair [1][2][3][4][5][6]. Two characteristic features, self-renewal and the ability to differentiate, make s/pC ideally adequate for cellular therapy of acute and chronic renal failures [7]. However, the success of this strategy is subject to a proper integration of the s/pC in the diseased organ at the site of damage [8][9][10].
In the meantime different methods for s/pC application were elaborated. The infusion over the capillary system is a frequently used method to administer these cells Figure 1a [8,12]. However, poor cell survival after application and no cell concentration at the site of necessary regeneration are the disadvantages of this application method [11,13]. Further contributions for kidney repair are the casual injection into diseased parenchyma Figure 1b [14] and the subcapsular injection of s/pC Figure 1c [15][16][17]. Cells for repair are either injected as suspension [18] or after coating with extracellular matrix proteins [19]. These methods share the disadvantage that numerous foci of degenerated parenchyma cannot be reached by spot application. Thus, the amount of injected or implanted s/pC in a diseased organ abides labile and cell concentration is not obtained at sites of need.
Independent of therapeutic application the s/pC have to migrate from the site of implantation through the interstitial space into diseased areas of the kidney in order to terminate the degradation process and turn it in regeneration [20]. Thus, to support the repair of diseased parenchyma one has to elaborate the optimal site of implantation and to learn about model systems leading to renal tubular regeneration in combination with s/pC [3,22].
In consequence, due to these needs we decided to investigate the adhesion of s/pC to a polyester fleece which can be implanted at the site of the earlier renal stem/progenitor cell niche located between the renal capsule and the outer cortex of parenchyma Figure 1d. Following this strategy it is possible to concentrate s/pC within the fleece structures. This innovative delivery system is preferred because the concentrated cells can stay at their site of implantation between the capsule and the outer renal cortex during the initial phase of adaption. Further an optimal ratio between surface area and volume can be achieved in order to provide the diseased organ with a high density of s/pC [23]. Another advantage is the possibility to incorporate later growth factors in the fleece fibers playing a key role in tissue restoration and regeneration [24][25][26].
However, before the idea can be realized to implant s/pC within a fleece, numerous problems have to be considered. The correct selection of polyester fleece as support material for the adhesion of s/ pC is dependent on several factors. The fleece fibers have to reflect a structured environment with tissue-specific mechanical properties. In In consequence, immunohistochemistry, transmission electron microscopy and scanning electron microscopy were performed. Since the specimens were not coated by extracellular matrix proteins, unique insights in the contact zone between the basal lamina, interstitial cells and surrounding polyester fibers can be obtained. Analyzing the specimens developing in I-7, Posi-4 and Posi-5 polyester fleeces by means of immune histochemistry and transmission electron microscopy illuminated that no cell biological differences could be observed. In contrast, scanning electron microscopy of generated tubules demonstrated that a difference in spatial distribution and different diameters of tubules could be registered. addition, the fleece material must be biocompatible to s/pC, to diseased, regenerating and healthy renal parenchyma. Further cell adhesion, cellular spreading and proliferation are influenced by porosity such as pore shape, size, interconnectivity and orientation, surface chemistry, topology and architecture [27]. Even the diffusion of respiratory gas, small molecules, nutrients and metabolic products must be promoted [28]. Finally the mechanical features of the used biomaterial have to withstand in vivo forces until the regenerated parenchyma has sufficient stability to protect itself [10].
Previous and present culture experiments demonstrated that numerous renal tubules derived from s/pC can be generated between two layers of polyester fleece I-7 [29][30][31], Posi-4 or Posi-5 [32]. Since the generation of tubules was performed without coating by extracellular matrix, it is possible to investigate the outer surface of the tubules, the interstitial space and the neighborhood to polyester fibers.
In consequence, the focus of interest in this study was directed to the formation of renal tubules within an artificial polyester interstitium needed for the future implantation of s/pC. Immunhistochemistry was performed to analyze the cell biological degree of differentiation, while transmission electron microscopy and scanning electron microscopy were made to obtain new information about the spatial distribution of tubules.

Isolation of stem/progenitor cells (s/pC)
Both kidneys of one-day old anesthesized and sacrificed New Zealand rabbits were removed and cut into two parts as described earlier [33]. Tissue containing renal s/pC was harvested from the outer cortex by stripping off the capsula fibrosa with fine forceps Figure 2a.

Biomaterials
To investigate the generation of tubules within an artificial interstitium a set of polyester fleeces, established I-7 (Walraf, Grevenbroich, Germany) versus Posi-4 and Posi-5 fleeces (Positech, Hallwil, Suisse), was tested.  . By stripping off the capsula fibrosa (CF) from neonatal rabbit kidney renal stem/progenitor cells within mesenchyme and collecting duct ampullae can be isolated (a). To create an artificial interstitium renal stem/progenitor cells are mounted between layers of polyester fleece within a tissue carrier (b). During perfusion culture always fresh medium is transported (arrow) by a peristaltic pump. To maintain a constant temperature of 37°C, the culture container is placed on a thermoplate and covered with a lid.

Histochemical labeling
After perfusion culture was terminated the sandwich set-ups containing renal tissue within two layers of polyester fleece (5 mm diameter) were embedded in 1 % agarose (Serva, Heidelberg, Germany), surrounded by TissueTek (O.C.T. TM COMPOUND, Sakura Finetek, Zoeterwoude, Netherlands) and frozen at -80°C. To analyze cell biological features 20 µm thick cryosections were made, fixed in ice cold ethanol, washed several times with phosphate buffered saline (PBS) and incubated for 30 minutes with blocking solution (PBS, pH 7.5, 10% horse serum, GIBCO, 1% bovine serum albumin, Serva). For soybean agglutinin-labeling (SBA, Vector, Burlingame, USA) the samples were exposed to fluorescein-isothiocyanate (FITC)-conjugated lectin diluted 1:2000 in blocking solution for 45 minutes. For immunohistochemical label anti-laminin γ1 (kindly provided by Dr. L. Sorokin, Lund, Sweden) was used undiluted, anti-cingulin (Progen Biotechnik, Heidelberg, Germany) was applied 1:10 and anti-collagen type III (III-53, Calbiochem, Schwalbach, Germany) was added 1:250 in blocking solution for one hour. After washing several times with 1 % BSA in PBS the specimens were incubated for 45 minutes with goatanti-rat-IgG-rhodamine, donkey-anti-guinea-pig-IgG-fluoresceinisothiocyanate or donkey-anti-mouse-IgG-fluorescein-isothiocya-nate (Jackson Immunoresearch Laboratories, West Grove, USA) diluted 1:50 in PBS containing 1% BSA. Following several washes with PBS the sections were embedded with Slow Fade Light Antifade Kit (Molecular Probes, Eugene, USA) and then analyzed using an Axioskop 2 plus microscope (Zeiss, Oberkochen, Germany). Fluorescence images were taken with a digital camera at a standard exposure time of 1.3 seconds and thereafter processed with Corel DRAW Graphic Suite X5 (Corel Corporation, Otawa, Canada).

Scanning Electron Microscopy (SEM)
To analyze the spatial distribution of generated tubules by SEM, specimens were fixed in 70 % ethanol, dehydrated in a graded series of ethanols, transferred to acetone, and critical-point dried in carbon dioxide. Finally, samples were sputter-coated with gold (Polaron E 5100, Watford, GB) and examined in a scanning electron microscope DSM 940 A (Zeiss, Oberkochen, Germany). Images of the screen were taken by a Pentax SLR Digital camera and thereafter processed with Adobe Photoshop (Adobe, California, USA) and Corel DRAW Graphic Suite X5.

Transmission Electron Microscopy (TEM)
For transmission electron microscopy (TEM) the specimens were fixed in 2% glutaraldehyde containing 0.1 M sucrose in 0.1 M  cacodylate buffer (pH 7,5) over night at 4°C. After several washes with PBS tissue was post fixed in 1 M PBS containing 1% osmium tetroxide. After rinsing in PBS specimens were dehydrated in graded series of ethanols and embedded in Epon. Polymerization was performed at 60°C for 48 h. Ultrathin sections were obtained with a diamond knife on an ultramicrotome EM UC6 (Leica GmbH, Wetzlar, Germany). Sections were collected onto grids (200 mesh) and contrasted using 2% uranyl acetate and lead citrate. For TEM studies, grids were analyzed by a TEM Zeiss 902 (Zeiss, Oberkochen, Germany). Electron micrographs were recorded digitally using a slow scan CCD camera and thereafter processed with Adobe Photoshop (Adobe, California, USA) and Corel DRAW Graphic Suite X5.

Pattern of the fleece surfaces
The present experiments were performed to analyze the spatial development of renal tubules within three different polyester fleeces.

Development of tubules within polyester fleeces
To analyze the spatial distribution of generated tubules frozen sections were made vertically to the fleece surface. Staining with toluidine blue showed that generated specimens contained numerous tubules at the interface of the artificial interstitium made with polyester fleece I-7 Figure 4a Figure 4c did not cover the generated tubules at the upper and lower side but are found to be widely integrated within the tissue.
In the next series of experiments generated tissue was labelled

Transmission electron microscopical features of generated tubules
Label for laminin γ1 and cingulin showed that polarization of epithelial cells was established during generation of tubules. To obtain more information transmission electron microscopy (TEM) of generated tubules at the interface of polyester fleeces I-7 Figure 5a

Scanning electron microscopy at the interface of an artificial interstitium
During generation of renal tubules the surrounding is not stacked by proteins derived from a coating process. This situation allowed the investigation of the outer tubule surface by scanning electron microscopy (SEM) Figure 6 Of special interest is the interface between the basal lamina of tubules and the surrounding polyester fibers, the occurrence of interstitial cells and the deposition of synthesized interstitial matrix.

SEM analysis of generated tubules within polyester fleece I-7
SEM shows that generated tubules were growing in a more or less parallel fashion to the polyester fibers of the fleece Figure 6a, d, g. The outer diameter of generated tubules was found to range between 14 µm and 49 µm. The basal aspect of each tubule was covered by a continuously developed basal lamina. Looking at the outer surface of the tubules interstitial cells and numerous fibers obviously consisting of newly synthesized extracellular matrix proteins could be detected Figure 6a, d, g. The synthesized fibers of extracellular matrix were interposed between generated tubules and single I-7 polyester fleece fibers as well as between contiguous tubules Figure 6d, g. Although the tubules grew in the vicinity of the polyester fibers they had only loose and astonishingly irregular contact to the fleece fibers.

SEM analysis of generated tubules within polyester fleece Posi-4
Scanning electron microscopy of Posi-4 fleece depicts generated tubules in a parallel growth Figure 6b

Focus to the outer surface of generated tubules
Since the specimens were not coated by extracellular matrix proteins the interface between tubules, interstitial cells and polyester fleece fibers now can be investigated by ultrastructural methods. SEM of I-7, Posi-5 and Posi-4 specimens depicts that the outer surface of generated tubules is covered by interstitial cells and a network of conspicuous filigreed fibers consisting obviously of newly synthesized extracellular matrix Figure 6. Thus, more information about the connection between interstitial cells and the outer surface of tubules was obtained by higher resolution. SEM images of generated tubules within polyester fleece Posi-5, for example, illlustrated interstitial cells (IC) on their surface Figure 7. These cells exhibit a diameter of 7-8 µm and were distributed in irregular distances Figure 7a, b. They contacted the tubule by cellular protrusions. The length of these protrusions ranged from 4 µm to 10 µm. The diameter spanned from 80 nm up to 1 µm depending on the distance of the interstitial cell to the basal lamina of the tubule. During extension the diameter of the protrusions became smaller. Before reaching the surface of the tubule the interstitial cell processes showed several dichotomous branchings and became thinner Figure 7b Looking over all, the tested I-7, Posi-4 and Posi-5 fleeces appear as promising candidates to harbour s/pC during regeneration of parenchyma in acute and chronic renal failure, although they show a different spatial growth pattern of generated tubules.

Discussion
The present experiments were performed to investigate by cell biological Figure 4 and ultrastructural methods Figure 5,6,7,8 the spatial growth of tubules within three selected polyester fleeces. All applied polyester fleeces I-7, Posi-4 and Posi-5 exhibit biocompatibility, although they differ in crosslinking, thickness, porosity, fiber diameter and appearance of the pores [32]. However, these structural differences do not influence the generation of tubules and the polarization of epithelial cells. This could be recognized based on the formation of a tight junctional complex separating the luminal and lateral plasma membrane of tubule cells.
Regarding the spatial growth pattern a difference in tubule growth could be observed. The ingrowth of tubules between polyester fleece fibers is rather high in Posi-4 Figure 4b but lower in Posi-5 Figure 4c polyester fleece. In the case of I-7 polyester fleece a tubule ingrowth could not be observed Figure 4a. These findings seem to correlate with the water porosity of the used fleeces [32], which implied that a higher porosity (Posi-4) supports the development and ingrowth of tubules. This observation is sustained by several studies emphasizing a high porosity for ensuring attraction for cells [10,34,35]. According to Zippel et al. interconnected pores allow cell ingrowth and vascularization, a higher porosity results in greater cell ingrowth [27]. The above mentioned interconnected pores between fleece fibers were observed in Posi-4 and Posi-5 polyester fleece and consequently they can be a reason for the demonstrated integration and spatial distribution of tubules.
Both light and electron microscopy showed that the development of tubules occurs independently if I-7, Posi-4 or Posi-5 polyester fleece were used as artificial interstitium Figure 4. Immunohistochemistry of these specimens further demonstrated that the generated tubules exhibit the same cell biological differentiation profile. In all three cases tubules with a completely developed basal lamina and tight junctional complexes could be demonstrated. Within the cytoplasm numerous lysosomal elements respectively vacuoles were detected on transmission electron micrographs Figure 5. The increased occurrence of lysosomal elements and vacuoles within in the cytoplasm could illuminate a possible implication of endo-and/or exocytosis. At the basolateral plasma membrane the degree of protrusions and infoldings indicates a high transport capacity Figure 5g, h, i [36,37]. A higher transport capacity gives reason to expect a high cell activity which could be correlated with a high endo-and/or exocytosis rate. Consequently many vacuoles and lysosomal elements can be observed.
Since the interstitial space between the generated tubules was not stacked by proteins from a coating process, it was possible to explore the outer surface of generated tubules by SEM. Hence, the observation of generated tubules within different polyester fleeces showed a difference in the diameter of the tubules Figure 6. Within I-7 Figure 6a, d,g polyester fleece the outer tubule diameter ranged between 14 µm and 49 µm, within Posi-4 Figure 6b, e, h polyester fleece between 20 µm and 46 µm and within Posi-5 Figure 6c, f, i polyester fleece between 17 µm and 33 µm. These wide and different ranges of diameters within the used polyester fleeces cannot be explained by differences during culture or in the fixation procedure because all of the tissues were treated the same. Most probably, the different diameters of generated tubules within the artificial interstitium might be explained by the occurrence of various tubules. In the kidney of two weeks old Albino rats a proximal tubular diameter of 42,7 µm and a distal tubular diameter of 26,5 µm was measured [38]. As described by Knepper et al. the mean proximal tubular diameter in rabbit kidney was 37 µm and the average distal tubular diameter amounted to 25 µm [39]. Cole et al. have demonstrated that during development of rabbit kidney an increase of the size of proximal tubule cells arose. If the size of proximal tubule cells increases, the outer diameter of the proximal tubule will also increase [40]. However, the immunohistochemical profile militates for the development of the same kind of tubules. Further the applied markers indicate the formation of collecting tubules. In consequence, the various tubule diameters and the different stage of tubule development seemed to be caused by surface features of Posi-4, Posi-5 and I-7 polyester fleece.
In this study the basal lamina of the generated tubules was covered by bright particles most probably extracellular matrix with different shape Figure 6, 7. These findings are in conformity with previous studies [41]. As yet the nature of these particles remains to be elaborated. Whereas nephrogenesis and terminal differentiation of renal tubules are dependent on extracellular matrix molecules like for example cell adhesion molecules [42][43][44][45][46], it seems reasonable to presume that these molecules are found on the surface of generated tubules and interstitial cells.
In the present study the discovered interstitial cells showed cellular processes with numerous bifurcations which contacted the basal lamina of generated tubules and which seem to be interwoven with the lamina fibroreticularis Figure 7. These observations are in accordance with previous experiments. For example, Lewis et al. showed that interstitial, fibroblast-like cells have an irregular, stellate shape with numerous long and thin processes, which appear to contact and to adhere to the basement membrane of adjacent tubules [47]. Depending on their location within the kidney and on their functional state the fibroblast phenotype shows some variations [48,49]. The observed cells were thought to represent "fibroblast-like" cells because they adhere to the generated tubules by cellular protrusions and their changed shape could be based on the location or the functional state [47][48][49][50]. The previous report by McAuliffe, who found in DI rats an almost total absence of interstitial cell processes, could not be confirmed in our study [51]. By SEM, Takahashi-Iwanaga et al. described cellular processes with a length of 10 µm and a decrease of the diameter from the cellular primary, to secondary and to tertiary processes [52]. These facts coincide with our results Figure 7c, d. Processes of interstitial cells with tapering ends or processes corresponding to flat pierced leaves could not be observed [52,53]. In the present study the processes do not seem to end but rather are interwoven with the lamina fibroreticularis of the generated tubules. It has been shown by transmission electron microscopy that the lamina fibroreticularis acts as connecting element to the interstitial space [41]. These adhesions and connections to the basal lamina of the generated tubules suggest that cross-talk between epithelial cell and interstitial, "fibroblast-like" cell might be mediated by mechanical forces [53,54].
The present SEM observations stated that generated tubules are fastened to the polyester fibers by extracellular matrix fibers resulting in a net-like structure Figure 8. These fibers were not existent after isolation of s/pC, but they were synthesized during perfusion culture [55]. A feature of the fibers is collagen type III [55,56]. This component is also found on tubules within the kidney and is part of the renal interstitium [57][58][59][60]. It is established that collagen synthesis is induced by aldosterone [61][62][63]. Therefore the appearance of collagen fibers in our cultures can be a consequence of the administered aldosterone. As described by Lemley and Kriz the collagen type III fibers form a network enveloping the tubules [64]. A net-like structure of the extracellular matrix at the surface of the tubules could partly be observed in our study. Collagen fibers are known to be part of scaffolding function [53]. These findings make it reasonable to assume that the observed net-like structure of the connecting element between generated tubules and a polyester fleece is part of the above mentioned scaffolding function.

Conclusion
In this study we demonstrated that the development and biological differentiation of renal tubules is not dependent on the chosen polyester fleece used as artificial interstitium. But ingrowth of tubules in the polyester fleece layers can only be observed if polyester fleeces with interconnected pores like Posi-4 or Posi-5 polyester fleece are used as artificial interstitium. In contrast, I-7 polyester fleece prevented in growth of tubules in the space between the fibers. Further interstitial cells lining with long processes to the tubules and extracellular matrix fibers seem to act as connecting elements. This important observation was possible, because the interface between tubules and artificial interstitium is not stacked by proteins derived from a coating process.