Integration of New Embryonic Nephrons Into the Kidney

Integration of New Embryonic Nephrons Into the Kidney Adrian S. Woolf, MD, A. Hornbruch, PhD, and Leon G. Fine, MD • The current report summarizes our experiments exploring the feasibility of creathig a chimeric kidney, that is, an organ constituted by cells deJived from more than one fertilized ovum. The overall strategy has been to obtain donor renal tissue from avian and murine embryos and to implant this into the host avian mesonephric mesoderm or into the cortex of murine neonatal kidney. In both models, donor cells were distinguished from the host by the presence of characteristic nuclear or cytoplasmic markers. Examination of quail to chick transplants showed the tandem development of mesonephric tissue in the form of bilobed organ. In the mouse chimeric kidney, examined 2 to 4 weeks postnatally, transplanted metanephric tissue grew and developed glomeruli, proximal tubules, and cords of cells, which extended into the medulla of the host kidney. Before death, intravenous FITC-dextran was administered to the host mouse. Some transplanted tubules were connected to filtering glomeruli, as judged by the presence of fluorescein within their lumens. These experimental models provide novel means with which to study nephrogenesls in vivo. Finally, If the embryonic donor tissue could be genetically engineered before implantation, the prospect of "nephron therapy" 'arises, in which altered implanted nephrons could deliver therapeutically useful molecules into the urine or kidney Interstitium. © 1991 by the National Kidney Foundation, Inc. INDEX WORDS: Chimeric kidney; quail/chick mesonephros; mouse metanephros.

W

E EXAMINED the possibility of adding new nephrons to the avian and mammalian kidney. In order to achieve this, we have exploited the growth potential of the embryo by implanting mesonephric tissue into chick embryos and metanephric tissue into the kidneys of neonatal recipient mice. In the latter species, nephrogenesis continues for a few days after birth and it was reasoned that the use of neonatal renal cortex as the host tissue would provide a suitable environment to encourage growth, development, and integration of the "microtransplants. l " The formation of chimeric kidneys by this technique not only provides a new in vivo model with which to study renal development, but also introduces the possibility of ultimately modifying kidney function by the introduction of whole nephron units that can filter plasma. After genetic modification of renal cells, such implants could theoretically be tailored to secrete biologically active molecules into the urine or peritubular circulations. To distinguish donor from recipient tissue, the implanted cells were chosen on the basis of easily identifiable cellular markers. Quail cells, when subjected to a particular staining process, have prominent nucleoli that distinguish them easily from chick cells. Mammalian metanephric tissue was obtained from the homozygous beige mouse 2 and a mouse line transgenic for the (j-globin gene, 3 which provided cytoplasmic and nuclear markers, respectively, that could be identified in the chimeric kidney on paraffin sections.l

THE FORMATION OF AN EMBRYONIC CHICK-QUAIL CHIMERIC MESONEPHROS

Tissue from the mesonephric ridge (Fig 1), three somites in width, was obtained from stage-18 quail embryos, which had 35 pairs of somites. The tissue was surgically removed from somite levels 22 to 24, 25 to 27, and 28 to 30, and grafted into stage-16 to -18 chick embryos just lateral to one of the mesonephric ridges at the level of the peloic girdle. Embryos were killed 24 to 72 hours after transplantation and fixed for histological examination. Since the donor tissue was of quail origin, these cells were easily identifiable by virtue of their prominent nucleoli. In a number of embryos, successful transplantation was achieved. In each case, the undifferentiated mesonephric mesenchymal cells of the quail developed into a well-defined mesonephros, which formed a bilobed organ toFrom the Division of Nephrology, Department of Medicine, UCLA School of Medicine , Los Angeles, CA; and the Department of Anatomy and Developmental Biology, University College and Middlesex School of Medicine , London, England. A.S. W. was supported by the Berkeley Fellowship (University College and Middlesex Hospital, London, England), and L. G. F. by a Senior Fellowship from the Fogarty International Centre , by a Grant-in-Aid from the American Heart Association by National Institutes of Health Grant No. R01 DK34049. Address reprint requests to Adrian S. Woolf, MD, Division of Nephrology, UCLA School of Medicine , 10833 Le Conte Ave, Los Angeles, CA 90024-1689. © 1991 by the National Kidney Foundation, 1nc.

0272-6386/91/J 706-0005$3. 00/0

American Journal of Kidney Diseases, Vol XVII, No 6 (June), 1991: pp 611-614

611

612

WOOLF, HORNBRUCH , AND FINE

quail v chick), but the epithelial cells external to the glomerular capillary basement membrane were of quail origin (Fig 1). Previous experiments by a number of investigators had demonstrated the existence of chimeric glomeruli when mesonephric tissue is implanted into chorioallantoic membranes, but the embryoembryo organ chimerism that we achieved had, to our knowledge, been described in only one previous communication. 4 Rodemer et al excised part of the Wolffian duct and underlying mesoderm and transplanted quail tissue into the corresponding chick region and also observed a hybrid organ. 4 Of importance is that they observed hybrid tubules, ie, chick and quail cells in the same section of a single tubule, whereas in our experiments, the tubules that formed from the quail mesenchyme were entirely of quail origin. Encouraged by these embryo-embryo transplantation results, we proceeded to embryo-neonatal transplants in the mouse. THE FORMATION OF A FUNCTIONING CHIMERIC MAMMALIAN KIDNEY

Fig 1. Quail-chick chimeric mesonephros. (A) Arrow indicates the mesonephric ridge mesoderm (mm), the site of undifferentiated quail donor tissue lateral to the somite (s). (8) A chimeric mesonephros in the chick embryo comprised of adjoining quail- (q) and chick-derived (c) organs. (C) High-power view of a quail-derived glomerulus. Cells of quail origin are identified by the prominent nucleolus (arrowhead).

gether with its adjacent chick counterpart (Fig 1) . Well-differentiated tubules, as well as occasional glomeruli of donor origin, were seen. The level of resolution of light microscopy did not allow for a definitive determination of the origin of the endothelial cells of the glomerular capillaries (ie,

Donor tissue was obtained from two mouse models. (1) The ,8-globin transgenic mouse line 83 was created and provided by Cecilia Lo (University of Pennsylvania, Philadelphia, PA). Nuclei contain 1,000 randomly repeated copies of a ,8globin-containing plasmid. These DNA inserts are not expressed, but can be identified by in situ hybridization. 3 The line was derived from SJL x C57BLl6 hybrids and, because fertility is low, they were crossed with CBA mice to provide donor metanephric tissue. The tissue was implanted into the outbred Q strain (MRC Mammalian Development Unit, London, England) . Despite the lack of histocompatibility, we found no evidence of rejection during short-term studies ( < 30 days). (2) The homozygous beige mouse C57BL6 bgj/bgi was obtained from Jackson Laboratories, Maine. This model of the Chediak-Higashi syndrome has kidneys that become pigmented with age. Giant lysosomes within the pars recta of the proximal tubule can be identified by autofluorescence 2 and enzyme histochemistry. 1 The congenic C57BLl6J mouse was used as a recipient. Pieces of metanephric tissue derived from 13 to 16 days' postcoital embryos were implanted into a tunnel fashioned in the renal cortex of the recipient

613

CHIMERIC KIDNEYS

mouse within 24 hours of its birth. At 2 to 4 weeks of age, the animals were killed and the chimeric kidneys examined. Fifteen seconds before death and removal of the kidney FITC-dextran (molecular weight, 10,000) was injected into the inferior vena cava. This molecule is filtered by the glomerulus,5 and on paraffin sections viewed under ultraviolet excitation, the appearance within the lumen of a tubule of bright green fluorescence was taken as evidence that the tubule had received glomerular ultrafiltrate.! The transgenic signal labels all donor cells, but using the beige marker, only proximal tubules can be firmly assigned to the donor. However, the advantage of the latter technique is that the autofluorescent granules and fluorescent filtered FITC-dextran can be viewed simultaneously. Approximately half of the mice that survived the neonatal operation were found to have evidence of transplanted tissue within the host kidney. The tissue appeared as a nodule within the cortex of the host kidney, which was estimated to not exceed 5 % of the total kidney volume. Transgenic donor glomeruli were identified, some with open capillary loops, while others appeared avascular. Tubules were also present and ranged from mature phenotypes (with periodic acid-Schiff [PAS]-positive brush borders, open lumens, and giant intracellular granules) in beige transplants to structures that lacked specific features. Although further ultrastructural analysis with electron microscopy would be desirable, this has not yet been per-

Fig 2. Chimeric murine kidney. Donor beige proximal tubules are identified by the presence of giant Iysosomes, which appear yellow under ultraviolet excitation (arrowheads). The presence of bright green fluorescence from FITC-dextran within their lumens (arrow) suggests that the transplanted tubules are patent and connected to filtering glomeruli. (Original magnification x 320.)

formed because of the difficulty in identifying the transplant by the naked eye. In two chimeric kidneys, we found clear evidence of filtration of fluorescein into the lumen of beige proximal tubules, indicating that those tubules were connected to a filtering glomerulus (Fig 2). In contrast, metanephric implants into adult kidneys grew as poorly differentiated, nonfiltering masses localized under the renal capsule. DISCUSSION

These preliminary studies demonstrate the feasibility of creating a chimeric kidney. Although fetal pancreatic islets 6 and neurons? have been transplanted into the adult, the current study probably describes the most complex embryonic tis~ue that has formed a functioning organ after implantation in vivo. The presence within the mouse implants of beige tubules containing giant granules showed that differentiation and growth had occurred, because these markers were absent in fetal tissue. In contrast, tubule formation was relatively poor in other studies that have transplanted the rodent metanephros onto the avian chorioallantoic membrane. 8 Although highly advanced proximal tubular differentiation can occur in metanephric organ culture, in that model glomeruli remain avascular. 9 In the mouse chimeric kidneys described, the capillary loops within some transplanted glomeruli suggested continuity with the host vasculature. Further studies incorporating electron microscopy should provide a useful tool for studying glomeru-

614

WOOLF, HORNBRUCH, AND FINE

logenesis. The appearance of FITC-dextran within the lumen of some transplanted proximal tubules argues that they are connected to filtering glomeruli. Theoretically, it might be possible for transplanted glomeruli to filter at a low rate into blind-ended tubules and definitive evidence of connections between donor nephrons and host connecting tubules can probably only be obtained by careful microdissection. The chimeric kidney represents a novel in vivo model of nephrogenesis. However, further studies will need to be performed to address the question of long-term viability of the implant, because survival of the tissue might eventually be compromised by ischemia or by lack of histocompatibility. If these problems can be resolved, it becomes possible to envisage the modification of the host kid-

ney by the implant. Although the addition of the filtering units to the host is unlikely to make a large impact on total glomerular filtration rate, it might be possible to implant a genetically engineered metanephros that would target a new or an altered gene product into the urine or the peritubular circulation. This might be achieved by implanting metanephric tissue transgenic for a secreted gene product or by first genetically modifying metanephros in organ culture before implantation. Preliminary evidence that the latter course is possible is provided by the successful transduction of the bacterial /3-galactosidase gene into both the epitheloid NRK cell line and primary cultures of mouse metanephros using a replication-
REFERENCES I. Woolf AS , Palmer SJ, Snow ML, et a1 : Creation of a

functioning chimeric mammalian kidney. Kidney Int 38:991997, 1990 2. Nakatsuji N: Beige granules as a cell marker for clonal analysis in kidney and liver of mouse aggregation chimeras, and three dimensional reconstruction for serial paraffin sections. Development 104:383-390, 1988 3. Varmuza S, Prideaux V, Kothary R, et al: Polytene chromosomes in mouse trophoblast giant cells. Development 102: 127-134, 1988 4. Rodemer ES, Ihmer A, Wartenberg H: Gonadal development of the chick embryo following microsurgically caused agenesis of the mesonephros and using interspecific quail-chick chimeras. J Embryol Exp Morphol 98 :269-285 , 1986 5. Steinhausen M, Wayland H, Fox JR: Renal dye tests V PfIiigers Arch 369:273-279, 1977

6. Mullen Y, Clare-Salzler M, Stein E, et al: Islet transplantation for the cure of diabetes. Pancreas 4:123-135, 1989 7. Lindvall 0, Brundin P. Widner H, et al: Grafts of fetal dopamine neurons survive and improve motor function in Parkinson 'S disease. Science 247:574-577, 1990 8. Sariola H, Ekblom p. Lehtonen E, et al : Differentiation and vascularization of the metanephric kidney grafted on the chorioallantoic membrane. Dev Bioi 96:427-435 , 1983 9. Avner ED, Sweeney WE, Ellis D: Serum free organ culture of embryonic mouse metanephros, in Methods for SerumFree Culture of Epithelial and Fibroblastic Cells. New York, NY, Liss, 1984, pp 33-41 10. Price J, Turner D, Cepko C: Lineage analysis in the vertebrate nervous system by retrovirus-mediated gene transfer. Proc Nat! Acad Sci USA 84:156-160, 1987