Immunohistochemical Study of a New Experimental Model of Acute Cellular Xenograft Rejection A. Ramos, A. Vega, F. Val, G. Chavez, M. Lo ´ pez-Hoyos, J.C. Ruiz, A.L.M. de Francisco, J. Castillo, M.G. Fleitas, and M. Arias X ENOGRAFT rejection must be overcome before clin- ical application of xenotransplantation. However, the complete knowledge of all the humoral and cellular mech- anisms involved is not well defined. We recently described an experimental model of acute cellular rejection, 1 which should contribute to better understanding of those mecha- nisms. Here we present the immunohistochemical study of this ex vivo perfusion model of pig-to-human kidney xeno- transplantation, in which human blood was manipulated to achieve an immunomodulatory effect. MATERIALS AND METHODS Pig kidneys were connected to an ex vivo perfusion circuit and perfused with manipulated human blood as we previously de- scribed. 1 The number of experiments was five per group. The experiment took place for 180 minutes or until hyperacute rejec- tion was evident. Specimens were taken from all hemoperfused kidneys after the end of the experiment and frozen. Immunofluo- rescence assays included CD2, CD20, CD68, C5b9, C3, IgG, IgM, and NK cell deposits. RESULTS HAR appeared after 15 to 90 minutes in porcine kidneys perfused with no manipulated human blood, showing dif- fuse haemorrhage, thrombosis, and glomerular injury by light microscopy studies and important deposits of IgG and IgM by immunofluorescence assays. HAR was not present after 3 hours when manipulated human blood was used as perfusate. However, some biopsies of pig kidneys perfused with manipulated human blood showed signs of acute cellular rejection with different degrees of interstitial infil- trate of mononuclear cells, especially NK cells, with perivas- cular location. In the xenoantibody blood depleted group we found absence of C5b9 and IgG kidney deposit and intense infiltrates of IgM and NK cells (Table 1). DISCUSSION These studies further corroborate that HAR immune re- sponse involves antibody, coagulation system, complement cascade, and very early inflammatory cells. 2 This model may be valuable in the isolated evaluation of any of these effector arms simulating in vivo condition. 3 Perfusion blood depleted of these immunologic mediators substantially pro- longs the survival of kidney xenograft, avoiding the HAR, 4 and developing a fast acute cellular rejection. REFERENCES 1. Vega A, Ramos A, Val F, et al: Transplant Proc 31:2643, 1999 2. Daniels LJ, Platt JL: Kidney Int 58:S28, 1997 3. Kroshus TJ, Bolman RM, Dalmasso AP: Transplantation 62:5, 1996 4. Baldwin WM, Sanfilippo F: Curr Opin Organ Transplant 2:121, 1997 From the Transplantation Unit H.U.M. Valdecilla, Santander, Spain, and Imutran Pathology Lab, Cambridge, UK. Address reprint requests to Dr J Castillo, Hospital Valdecilla, Cirugia General II 9aT, 39008 Santander, Spain. Table 1. Immunofluorescence Findings in Porcine Kidney Perfused with Human Blood Perfusion Solution Immunofluorescence Stain CD2 CD20 CD68 Fibrina C3 C5b9 IgG IgM NK Pig blood control - - - - - - - - - Human blood not manipulated + - + + ++ + +++ +++ + Human blood complement depleted ++ - + +/- +/- +/- ++ ++ +++ Human blood platelet depleted + - + +/- ++ + ++ ++ + Human blood leukocytes depleted - - - - ++ ++ ++ ++ - Human blood xenoantibody depleted - - - - ++ +/- - +++ ++++ Note: Pig blood perfusion was used as control. 0041-1345/00/$–see front matter © 2000 by Elsevier Science Inc. PII S0041-1345(00)01060-5 655 Avenue of the Americas, New York, NY 10010 960 Transplantation Proceedings, 32, 960 (2000)