In Press, 1995, Transplantation Proceedings copyright by Appleton and Lange.

CD45 Isoforms and Lymphocyte Activation Antigens in Acute Renal Allograft Rejection

A D Battochio, K Solez, S Poppema, S Cockfield+ and D C Rayner

Department of Laboratory Medicine and Pathology, and +Department of Medicine, University of Alberta Faculty of Medicine, Edmonton, Canada T6G 2R7

Running titles: Battochio et al. CD45 isoforms in renal graft rejection


Lymphocyte surface marker analysis is a powerful tool for correlating cell phenotype and function. In kidney allograft rejection, a primary role of CD8+ cells in the inflammatory infiltrate has been apparent for some time [1], and the expression of other antigens (such as CD57) [2] also appears to be correlated with acute rejection.

The CD45 molecule (leukocyte common antigen) exists as multiple isoforms which are expressed by different cell types in different functional states [3]. The CD45RO isoform, in conjunction with CD8, has predictive value in the diagnosis of acute rejection [4]. New monoclonal antibodies provide an even more refined tool for the study of CD45 variants; these reagents recognize additional epitopes on the different CD45 isoforms, some of which are glycosylation-dependent [5].

In this study, our aim was to determine whether the extended phenotype of the inflammatory infiltrate, based on expression of CD45 isoforms and other membrane antigens, is a useful adjunct in evaluation of the kidney allograft biopsy. Our findings confirm that an interstitial infiltrate rich in CD8+ and CD45RO+ cells is characteristic of acute rejection. In addition, we show that these antigens have considerably more diagnostic power than other T cell activation markers, including CD25, CD26 and CD69.




Frozen renal transplant biopsies were selected from archival tissue. All biopsies were performed locally, from first allografts less than eight weeks post-transplantation. Patients were maintained on standard triple immunosuppressant therapy of prednisone, cyclosporine and azothioprine; most (in both the rejection and non-rejection groups) were treated with supplementary immunosuppression before biopsy. Paraffin sections of six biopsies showed moderate acute rejection: four with Class 2A rejection changes (t3 tubulitis) and two with Class 2B (v1 vasculitic) rejection. In addition, three non-rejection biopsies were selected based on the absence of significant tubulitis or arteritis lesions (t0 v0), but sufficient interstitial inflammation (i1) to permit assessment of the phenotypic characteristics of the infiltrating leukocytes. These three patients had diagnoses of acute tubular necrosis, urine leakage, and ureteric obstruction.



Monoclonal antibodies specific for CD2 (Leu5b), CD3 (Leu4), CD4 (Leu3a + 3b), CD8 (Leu2a), CD25 (anti-IL-2 receptor), CD38 (Leu17), CD45 (HLe-1), CD69 (Leu23), and HLA-DR were all obtained from Becton Dickinson. OPD4, specific for CD45RO, was obtained from DAKO. Monoclonal antibodies to CD26 (65-2A6), CD45RO (UCHL1), CD45RA (MT2 and MB1) and CD45RB (MT3 and 6B6) were derived from hybridoma culture supernatants maintained in the Department of Laboratory Medicine, Cross Cancer Institute, Edmonton, Canada.

OCT-embedded frozen sections were cut at 4Ám, and stained using the panel of monoclonal antibodies and a standard two-step indirect immunoperoxidase method. In each condition, 200-500 infiltrating mononuclear cells were scored as positive or negative by an observer blinded as to the diagnosis, and the results expressed as a proportion of the total CD45+ cells.



Biopsies from acute rejection cases showed an interstitial leukocyte infiltrate which was approximately four times as dense as in the non-rejection biopsies. In addition, the infiltrate showed qualitative differences: the rejection biopsies had a markedly greater proportion of T cells (CD2+/CD3+), largely attributable to a fourfold increase in CD8+ cells (Table 1).

Expression of the activation antigen CD38 showed a similar 3.5-fold increase, from 6 to 21%, in acute rejection. In both groups, approximately 30% of the CD38+ population was recognizably plasmacytoid with very strong CD38 expression (denoted CD38 bright cells). The activation antigens CD25, CD26 and CD69 were present at low levels in both groups, and were not significantly increased in rejection. HLA-DR was also expressed to a comparable degree in the non-rejection biopsies (35 ▒ 10% of interstitial cells) and in the rejection group (45 ▒ 3%).

Expression of the different CD45 isoforms differed between the two groups (Table 2). The proportion of CD45RO+ cells, mainly antigen-experienced T cells, increased markedly (5 to 11-fold) in the rejection biopsies. Expression of the isoforms CD45RA and CD45RB, which have a wider cellular distribution, generally increased to a smaller degree (2 to 4-fold). An exception was the subset of CD45RB+ cells detected by the antibody MT3, which increased by a factor of 8 in the rejection biopsies.



Consistent with previous work [1], CD8+ T cells formed a large proportion of the rejection infiltrate. Surprisingly, the T cell activation antigens CD25, CD26 and CD69 were found in only a small minority of cells (0.6 - 5.5%), although this proportion was increased in the rejection biopsies. CD38 expression had greater utility as a marker of rejection, although the CD38+ cell population was clearly heterogeneous and in some cases included numerous plasma cells. The CD38 antigen corresponds to a glycoprotein involved in lymphocyte activation, and is expressed on a range of cells including most monocytes and natural killer cells, as well as plasma cells and activated T cells [6].

In T cells, the CD45RO isoform is believed to correlate with antigen experience or immunologic memory, and is expressed reciprocally to the 'naive' isoform CD45RA. We found expression of CD45RO was markedly augmented in acute rejection, with a greater increase observed when the antibody OPD4, rather than UCHL1, was used for detection. This difference may be due to the different specificities of the two antibodies: the epitope recognized by OPD4 is T cell specific, whereas UCHL1 also detects a population of monocytes [7].

The observed increase in CD45RA+ cells in the rejection infiltrates may reflect greater overall numbers of T cells, as well as expression of CD45RA by other cell populations including B cells and mononuclear phagocytes. Lower levels of CD45RA expression were observed using the antibody MB1.

CD45RB was detected at much lower levels using the antibody MT3 rather than 6B6. This is consistent with the known specificity of MT3 for a restricted CD45RB epitope (present only in the absence of the sequence encoded by the C exon) [8]. In contrast, the CD45RB epitope detected by 6B6 has a wide cellular distribution and is present on most lymphocytes and mononuclear phagocytes. Using 6B6, we observed CD45RB expression in only 68% of the infiltrating leukocytes, suggesting that only a strongly staining population was detected under our conditions. In T cells, there is evidence that a strongly CD45RB+ subpopulation is responsible for secretion of the Th1-type inflammatory cytokines (reviewed in Ref. 9).

Our results support the view that the T cell antigens CD8 and CD45RO may be useful markers of acute rejection [4]. Observations in an additional patient indicate that the surface antigen profile has independent diagnostic power. This 22-year-old woman had deteriorating renal function eight weeks after kidney transplantation; a marginally adequate biopsy was obtained which showed no definite tubulitis or vasculitis. Following further deterioration, a second biopsy was obtained 12 days later which showed mild acute rejection. Retrospective immunohistochemical staining of both biopsies showed a CD45 isoform profile typical of acute rejection (Table 3). These results suggest that the immunophenotype may provide a valuable supplement to morphology in evaluating the difficult or atypical biopsy.



  1. Sanfilippo F, Kolbeck PC, Vaughn WK, et al.: Transplantation 40:679, 1985.


  2. Beeschorner WE, Burdick JF, Williams GM, et al.: Transplant Proc 17:618, 1985.


  3. Trowbridge IS, Thomas ML: Annu Rev Immunol 12:85, 1994.


  4. Ibrahim S, Dawson DV, Sanfilippo F: Transplantation 59:724, 1995.


  5. Lai R, Visser L, Poppema S: Lab Invest 64:844, 1991.


  6. Malavasi F, Funaro A, Roggero S et al.: Immunol Today 15:95, 1994.


  7. Poppema S, Lai R, Visser L: Am J Pathol 139:725, 1991.


  8. Poppema S, Lai R, Visser L: J Immunol 147:218, 1991.


  9. Poppema S, Lai R, Visser L et al.: Leuk Lymphoma, in press, 1995.




Please note your browser must be capable of handling Tables. Newer versions of the Netscape browser are recommended



Table 1: T cell phenotypic markers and activation antigens in kidney graft rejection.

Antigens Non-rejection
(mean ▒SEM)
(mean ▒SEM)
CD2 + 3 17 ▒3 59 ▒8
CD4 19 ▒3 23 ▒3
CD8 10 ▒1 39 ▒2
CD25 2 ▒1 3 ▒2
CD26 3 ▒3 6 ▒2
CD38 dim 4 ▒2 15 ▒3
CD38 bright 2 ▒1 6 ▒2
CD69 0.3 ▒0.3 1.2 ▒0.6

Click here to go back.


Table 2: CD45 isoform expression in kidney graft rejection.

Specificity McAb Non-rejection
(mean ▒SEM)
(mean ▒SEM)
CD45RO OPD4 1.5 ▒1 17 ▒5
CD45RO UCHL1 5 ▒4 22 ▒5
CD45RA MT2 14 ▒7 26 ▒5
CD45RA MB1 3 ▒3 11 ▒4
CD45RB MT3 4 ▒1 31 ▒6
CD45RB 6B6 36 ▒8 68 ▒8

Click here to go back.


Table 3: Immunophenotype at two time points in a rejecting allograft.

Days post-transplant 58 70
Creatinine (ÁM) 130-190 250
Banff score g0 i2 t0 v0 g0 i2 t2 v0
CD2 + 3 57% 61%
CD4 23 16
CD8 39 47
CD45RO (OPD4 22 13
CD45RA (MT2) 33 24
CD45RB (6B6) 54 38

Click here to go back.


Return to Banff Presentations


Last Modified: April 03, 1996 11:00:03 AM