INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

WE HAVE SHOWN PREVIOUSLY THAT nonvascularized embryonic day 15 (E15) metanephroi from outbred rat embryos transplanted into the omentum of nonimmunosuppressed adult rat hosts undergo growth and differentiation and are vascularized by major vessels arising from the host omentum (15, 17). In contrast, developed kidneys transplanted from one rat into another undergo acute rejection within 7 days (15).

With the use of inbred rat strains (PVG-RT1^cPVG-RT1^avl), we demonstrated identical findings after transplantation across the rat major histocompatibility complex (MHC) RT1. Transplanted metanephroi (that contain no mature dendritic cells) from PVG donors into PVG-RT1^avl hosts were rejected only after subsequent transplantation of dendritic cell-containing skin from PVG rats. This observation indicates that a state of peripheral immune tolerance secondary to T cell ignorance is permissive, at least in part, of the survival of transplanted metanephroi in nonimmunosuppressed RT1-disparate rats (14).

Because of the state of peripheral tolerance that exists posttransplantation of metanephroi (14) and because, unlike the case after transplantation of developed kidneys, the vasculature of transplanted developed metanephroi originates, at least in part, from the host (15), there are theoretical advantages to transplanting metanephroi relative to developed kidneys. In the case of xenotransplantation, the presence of host endothelium in a transplanted organ would obviate the problem of antigen presentation by donor endothelial cells and, in the case of pighuman metanephric xenografts, hyperacute rejection (3, 18).

To shed light on the feasibility of metanephros xenotransplantation, we implanted rat metanephroi into the omentum of mice. Transplanted rat metanephroi grow and differentiate under these conditions in host mice that receive hCTLA4Ig, anti-CD45RB, and anti-CD154 but not without the use of these agents. The glomerular capillary loops of developed transplanted rat glomeruli consist, at least in part, of cells originating from mice.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Metanephroi were surgically dissected from E15 Lewis rat embryos (Harlan, Indianapolis, IN) under a dissecting microscope using previously described techniques (15) and placed immediately into ice-cold Ham's F12 DMEM (DMEM:HF12) solution. The following growth factors, previously shown to enhance the growth of metanephroi grown in vitro (10) or of transplanted metanephroi (7-9, 17) were added to the DMEM:HF12: recombinant human insulin-like growth factor (IGF)-I (Genentech, San Francisco, CA), 10⁷ M; recombinant human IGF-II (Bachem, Torrance, CA), 10⁷ M; recombinant human transforming growth factor- (Upstate Biotechnology Lake Placid, NY), 10⁸ M; recombinant human hepatocyte growth factor (R&D Systems, Minneapolis, MN), 10⁸ M; recombinant human vascular endothelial growth factor (VEGF; Genentech), 5 ug/ml; recombinant human basic fibroblast growth factor (R&D Systems), 5 ug/ml; recombinant human nerve growth factor (Boehringer Mannheim, Indianapolis, IN), 5 ug/ml; retinoic acid (Sigma Chemicals, St. Louis, MO), 10⁶ M; corticotropin-releasing hormone (Sigma Chemicals) 1 ug/ml; Tamm Horsfall protein (Biomedical Technologies, Stoughton, MA), 1 ug/ml; 25 mM prostaglandin E1 and iron-saturated transferrin (5 ug/ml). Optimal concentrations of growth factors (pharmacological) were determined by trial and error using a Sprague-DawleySprague-Dawley metanephros transplantation model (8, 15).

Metanephroi were implanted by placing them into a pouch of the omentum of anesthetized 10- to 14-wk-old female (host) C57Bl/6J mice (Jackson Laboratories, Mount Desert Island, ME) after 45 min of incubation in the DMEM:HF12 solution containing the growth factors (8, 9). All host mice were subjected to identical dark/light cycles (12:12 h) and identical diets for 7 days preimplantation of metanephroi and postimplantation.

For control experiments only (Fig. 1), metanephroi from E14 C57Bl/6J mouse embryos were transplanted into 10- to 14-wk-old C57Bl/6J mice, or metanephroi from E15 Lewis rat embryos were transplanted into 12-wk-old Lewis rats and removed from hosts 2 wk later.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Delineation of experimental groups.

When stated in the text, mouse hosts were treated with tolerance-inducing agents. Control animals received injections of vehicle. For CTLA4Ig, we used a modification of the regimen described by Larsen et al. (12) in a mouse cardiac allotransplantation model [0.2 mg on the day of transplantation (day 0) and on days 2, 4, and 6 posttransplantation]. For anti-CD45RB, we used a modification of the regimen described by Zhang et al. (22) in a rat-to-mouse cardiac xenograft model (0.1 mg iv on day 1 before transplantation and on the day of transplantation and 0.1 mg ip on days 1-10 posttransplantation). For anti-CD154, we used the exact regimen described by Larsen et al. (12) in combination with CTLA4Ig.

Our regimen was hCTLA4Ig (Genetics Institute, Cambridge, MA), 0.2 mg ip on the day of transplantation (day 0), and on days 2 and 4 posttransplantation; anti-CD45RB (Clone 23G2, Pharmingen, San Diego, CA) 0.1 mg iv on day 3 before transplantation (day 3) through day 0 and 0.1 mg ip on days 1-10 posttransplantation; and anti-CD154 (Clone MR1, Pharmingen), 0.25 mg ip on days 0, 2, and 4 posttransplantation.

Metanephroi were removed from mice, fixed, embedded in paraffin, sectioned, and stained with hematoxylin and eosin exactly as in previous studies (15). Immunohistochemistry was performed on zinc-fixed tissues (Pharmingen) using previously described techniques (16). A monoclonal antibody specific for mouse CD31 (Mec13.3 clone Pharmingen) (20, 21) was used to detect endothelial cells (10 ug/ml dilution). The secondary antibody was a biotinylated goat anti-rat IgG (Pharmingen). As a control, rat IgG_2a (Pharmingen) was substituted for anti CD31.

All figures are representative of >5 transplants.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A delineation of experimental groups designated a-d is shown in Fig. 1.

We transplanted metanephroi from an E15 rat embryo into the peritoneum of 10-wk-old mice (Fig. 1A). Two weeks later, either no trace of the metanephros could be found in mice that received no tolerance-inducing agents, or a yellowish piece of tissue, too small to embed, was observed in the omentum.

In contrast, in mice that received hCTLA4Ig, anti-CD45RB, and anti-CD154 (Fig. 1B), the metanephros had grown into a kidneylike structure (Fig. 2A). Hematoxylin- and eosin-stained sections of developed metanephroi revealed a nephrogenic zone (NZ), cortex, and medulla (Fig. 2B). Developing nephrons were observed in the NZ (Fig. 2C). Developed glomeruli were observed deeper within the cortex (Fig. 2D).

View larger version (116K): [in this window] [in a new window]   Fig. 2.   Photograph (A) and photomicrographs (B-D) of hematoxylin- and eosin-stained sections of rat metanephroi 2 wk posttransplantation into a mouse omentum. A: developed metanephros (m) is labeled. B: a nephrogenic zone (NZ), cortex (C) and medulla (M) are labeled. C: developing nephron (arrowhead) in the NZ. D: developed glomerulus (g) deeper within the cortex. Magnifications are shown for A and B and for C and D (d).

As was the case in Sprague-DawleySprague-Dawley metanephros transplants (15), occasional nests of lymphocytes were observed in transplanted developed metanephroi (not shown). However, no evidence of tubulitis was observed (Fig. 2).

To gain insight into the origin of the vasculature (donor vs. host) of metanephros xenografts, we stained developed metanephroi using anti-mouse CD31. Shown in Fig. 3A is a photomicrograph of a paraffin-embedded section of a developing rat metanephros 2 wk posttransplantation into the peritoneum of a mouse (Fig. 1B). CD31-positive mouse (host) cells are shown adjacent to a developing rat (donor) nephron or comma-shaped body in the NZ (Fig. 3A). Sections stained using anti-mouse IgG_2a in place of anti-CD31 show no positive staining (Fig. 3B). As a positive control (Fig. 1C), CD31-positive mouse endothelial cells are shown in a mousemouse transplant (Fig. 3C).

View larger version (87K): [in this window] [in a new window]   Fig. 3.   Photomicrographs of stained sections of rat (A, B) or mouse (C) metanephroi 2 wk posttransplantation into a mouse omentum. A: stained using anti-CD31. Arrows delineate endothelial cells. cb, comma-shaped body. B: stained using control antibodies. C: A stained using anti-CD31. Arrows delineate endothelial cells. Magnification is shown.

Shown in Fig. 4A is a photomicrograph of a paraffin-embedded section containing a developed glomerulus in a rat metanephros 2 wk after transplantation into a mouse (Fig. 1b), stained using anti-mouse CD31. A positively staining glomerular capillary loop is delineated. As a negative control (Fig. 1D), no CD31-positive mouse endothelial cells are shown in a ratrat transplant (Fig. 4B). As a positive control (Fig. 1C), CD31-positive mouse endothelial cells are shown in a mousemouse transplant (Fig. 4C).

View larger version (87K): [in this window] [in a new window]   Fig. 4.   Photomicrographs of stained sections of rat (A, B) or mouse (C) metanephroi 2 wk posttransplantation into a mouse omentum (A, C) or a rat omentum (B) stained using anti-CD31. Arrows show positively staining glomerular capillary loops. Glomeruli (g) are labeled. Magnification is shown (A).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

T cell-mediated rejection is a major threat to the long-term survival of xenografts. In the case of pighuman transplantation, the T cell anti-pig response is a significant one and is thought to be mediated by recognition of xenoantigens presented both directly and indirectly (3). Although direct presentation may not be a threat to metanephric xenograft survival because of the absence of professional antigen-presenting cells in transplanted metanephroi (14), indirect presentation remains operative.

For a number of reasons, including the induction of unacceptable side effects, overcoming T cell-mediated xenograft rejection through conventional systemic immunosuppression may not be possible (3). It has been suggested that the use of tolerance-inducing agents such as CTLA4Ig, anti-CD154 (CD40 ligand) (1, 3, 12), and anti-CD45 (22) could provide an effective and acceptable alternative to the use of conventional immunosuppression. Our data are consistent with the utility of their use in metanephros xenotransplantation (ratmouse).

Ours (Figs. 2-4) are not the first metanephroi xenografts. We previously transplanted mouse metanephroi beneath the capsule of rat kidneys. Under these conditions, development of mouserat transplants was observed only in hosts treated with cyclosporine (6). In addition, Dekel et al. (2) have implanted human metanephroi beneath the renal capsule of irradiated rats.

However, to our knowledge, Figs. 2-4 illustrate the first ratmouse metanephroi xenografts and the first metanephros xenografts transplanted into the omentum. In addition, ours represent the first use of tolerance-inducing agents for metanephros transplantation.

It has been suggested that the shortage of human kidneys available for transplantation could be alleviated by the use of animal kidney xenografts (18). In many ways, pigs represent the ideal renal organ donor for humans. This is because, relative to more closely related nonhuman primates, pigs are plentiful and their size, digestive, circulatory, respiratory, and renal physiologies are very similar to those of humans. Unfortunately, the transplantation of porcine vascularized organs, including kidneys, into humans is rendered problematic, in part because of the reaction of preformed antibodies against antigens present on the vascular endothelium of the pig (hyperacute rejection) (3, 18).

Unlike a developed kidney, the E15 rat metanephros is a nonvascularized organ (15). Insight into the origin of the blood supply for developed transplanted metanephroi is provided by experiments in which developing kidneys are transplanted to ectopic sites. In the case of 11-day-old mouse or chick metanephroi grafted onto the chorioallantoic membrane of the quail, the vasculature is derived entirely from the host (19). In the case of 11- to 12-day-old mouse metanephroi grafted into the anterior chamber of the eye, the glomerular microvascular endothelium derives from both donor and host (11). In either case (11, 19), and in the case of rat metanephroi transplanted into the rat omentum (15), large external vessels derive from the host.

Tufro (20) has demonstrated that rat metanephroi, cocultured in collagen gels with endothelial cells of mouse origin, are invaded by the endothelial cells that form capillarylike structures within and surrounding developing rat nephrons. This process can be prevented by the addition of anti-VEGF neutralizing antibodies to cultures. Tufro concludes that when exogenous endothelial cells are available, they are capable of invading rat metanephros explants in an organized manner similar to that occurring during normal development. She suggests that VEGF produced by differentiating nephrons acts as a chemoattractant for developing capillaries. In addition, Tufro's data show that developing rat kidneys can attract mouse endothelial cells (20).

The data shown in Figs. 3 and 4 support and extend Tufro's suggestion. They show that at least one component of the developing metanephros vasculature, including many glomerular capillary loops in ratmouse chimeric kidneys that develop in vivo, originates from the mouse host.

Perspectives