The presence of antibodies against the AD2 epitope of cytomegalovirus glycoprotein B is associated with acute rejection after renal transplantation



The aim of this study was to evaluate the association between antibodies against cytomegalovirus (CMV) glycoprotein B (gB) and acute rejection after transplantation. Seventy-seven consecutive renal transplant recipients in a D + /R+ setting were studied. Biopsy-proven rejection occurred in 35% of the recipients. Among these recipients, 85% had antibodies against CMV gB. The rate of acute rejection was significantly higher in recipients with antibodies against gB than in those without them. Antibodies against gB can be a useful predictor of acute rejection in renal transplant recipients in a D + /R+ setting.

List of Abbreviations

amino acid


antigen domain 2




both donor and recipient positive for CMV


glycoprotein B


glycoprotein H




glutathione S-transferase

Renal transplantation is a most valuable treatment for patients with end-stage renal disease, offering a long-term survival benefit compared with patients on dialysis [1]. However, acute rejection episodes are an important risk factor for functional deterioration of solid-organ transplants [2]. Although novel immunosuppressive regimens have reduced graft loss, susceptibility to infections has increased. Viral replication after transplantation may contribute to reduced graft function and survival through the associated inflammation and cytokine release [3]. Uncontrolled replication of viruses such as adenovirus, CMV, polyomavirus BK, John Cunningham virus, parvovirus B19 and human herpes virus-6 and -7 triggers direct and/or indirect effect in transplant recipients [4]. Among these viruses, CMV is the most important pathogen affecting kidney allograft recipients. When CMV is reactivated under immunosuppressive conditions, it has both direct effects, such as development of CMV disease, and indirect effects on transplantation, including increased incidence of allograft rejection [5]. The role of CMV infection in acute rejection after renal transplantation remains controversial; several studies have suggested that it can lead to allograft rejection [6, 7]. Because investigation of strategies for preventing CMV replication and acute rejection is of ongoing interest [8], we have concentrated on this matter in our series of our studies.

Cytomegalovirus, a member of the herpesvirus family, has a large genome which encodes over 65 unique glycoproteins [9]. It is well known that some of the glycoproteins encoded by CMV induce strong immune responses, as do other viral components. Among the glycoproteins gB, one of the most abundant envelope components, is essential for viral replication and considered one of the major target molecules for neutralizing antibodies as well as for cellular immune response [10]. Three linear antibody-binding sites have been described: it is well known that the AD2 site I epitope of gB is conserved in CMV isolates and is the major epitope for neutralization [9, 11, 12]. The antibody-binding site on AD2 is located between a.a. 28 and 84 of gB [9, 11]. gB is also a target for CMV-specific T-cell immunity. Although little is known about any association between gB AD2 and CMV-specific T-cells, Elkington et al. isolated CD4+ cytotoxic T lymphocytes [13], which recognize epitopes from CMV gB in association with HLA-DR7 and DR11 antigens. In addition to gB, gH has been used to identify preexisting strain-specific antibodies [14, 15]. Previously, we found that reinfection of seropositive recipients with a different type of CMV is also associated with acute rejection and CMV disease in renal transplant patients [15]. A study which reevaluated the previous study has also indicated that the absence of antibodies against gB in transplantation recipients is a good indicator of CMV disease [16].

In this study, we investigated whether, in addition to CMV disease, antibodies against gB AD2 contribute to prediction of acute rejection in renal transplantation in D + R+ setting, irrespective of gH serological matching.

This study investigated 77 CMV seropositive renal transplant recipients whose donors were also CMV seropositive (D + /R+ setting) and in whom antibodies against amino-terminal regions of CMV-gH had been detected; these recipients were enrolled at Fukushima Medical University and Tokyo Women's Medical University and have been described previously [15]. All study recipients had received hemodialysis treatment before transplantation and had received living-related renal transplants. This study was approved by the Institutional Ethics Committee and written informed consent was obtained from all subjects. All serum specimens were obtained before transplantation.

To detect antibody against CMV gB AD2 site I, which is located between a.a. positions 68 and 77 of gB of AD169 strain, GST protein fused with the AD2 site I was expressed and purified as described previously [17]. DNA cassette encoding the conserved epitope in CMV AD2 site I was cloned into the expression vector pGEX-5X (Amersham Bioscience [now GE Healthcare], Piscataway, NJ, USA). GST fusion proteins containing the gH epitopes from the AD169 and Towne strain were used to detect CMV gH type-specific antibodies as previously reported [15]. OD values specific to each antigen were obtained by subtracting the OD values for GST as described previously [15]. An arbitrary cutoff for ELISA (OD = 0.25) was defined as the mean plus two standard deviations of OD values of a panel of healthy CMV seronegative volunteers [15]. Detection of strain-specific gH-antibodies in the recipients' serum samples, which matched those of their donors, was considered gH-m antibody positivity.

The basic characteristics of the renal transplant recipients are summarized in Table 1. Fifty-two of the 77 recipients had antibodies against gB. There were no differences between patients with and without gB antibodies in other relevant variables, namely age, sex, number of HLA mismatches and immunosuppression protocols. The transplant recipients were followed up for 6 months after transplantation. Rejection was suspected when serum creatinine concentrations increased more than 25% above the basal level in the absence of urinary tract obstruction or renal graft artery stenosis, as described previously [15]. The first rejection episode was confirmed histologically by biopsying the grafts. Preemptive therapy was employed when CMV infection and/or CMV end-organ disease were diagnosed, as described previously [15].

Table 1. Baseline characteristics
CMV gB serostatusgB (+)gB (−)
No. of pairings5225
Age (years; mean ± SD)41.4 ± 12.642.9 ± 14.8
HLA mismatch (A, B, DR)2.9 ± 1.32.6 ± 1.1
No. of immunosuppression (%)
Tacrolimus-based regimen50 (96)24 (96)
Cyclosporin-based regimen2 (4)1 (4)

Using StatView 5.0, Fisher's exact test was used to evaluate the rate of acute rejection in different gB serostatus groups. Statistical significance was set at P < 0.05. The incidence of biopsy-proven acute rejection was calculated using the Kaplan–Mayer method, and comparisons were carried out by the log-rank test using SPSS.

Subsequent to their entry into the study, 27/77 recipients (35%) in a D + /R+ setting experienced biopsy-proven rejection during the 6 months after transplantation. Among these 27 D + /R+ patients with rejection, 23 (85%) had antibodies against CMV gB. The incidences of acute rejection among recipients with (gB+) and without (gB−) antibodies against gB AD2 were 44% and 16%, respectively. The rate of acute rejection was significantly higher in gB+ recipients than in gB− recipients (Table 2). Figure 1 shows Kaplan–Meier curves for the cumulative probability of freedom from biopsy-proven acute rejection. There were significant differences between the gB+ group and the gB− group according to the log-rank test (P = 0.025).

Table 2. Comparison of incidence of acute rejection according to antibody response against CMB gB AD2
CMV gB serostatusgB (+)gB (−)P
No. of recipients52250.021
No. of recipients with acute rejection (%)23 (44)4 (16)
Figure 1.

Kaplan–Meier curves of the cumulative probability of freedom from biopsy-proven acute rejection. The incidence of acute rejection in the gB + group is significantly higher than that in the gB− group (P = 0.025).

Our previous study, which employed multivariate logistic regression analysis to identify factors associated with acute rejection, including age, sex, number of HLA mismatch, immunosuppression protocols, demonstrated that reinfection of seropositive recipients with a different gH-serotype of CMV is associated with adverse outcomes in renal transplant recipients [15]. On the other hand, in the present study, there was no significant difference in the gB antibody-positive rate between gH-m+ and gH-m− recipients with acute rejection (Table 3), suggesting that presence of antibodies against gB is a risk factor irrespective of gH serological matching.

Table 3. gB AD2 seroprevalence according to gH-type matching in the 27 recipients with acute rejection
 gH type matchP
No. of recipients with rejection1017ns
No. of gB (+) recipients with acute rejection (%)9 (90)14 (82)

Many studies have reported a relationship between CMV and allograft rejection in renal transplant recipients. Previously, we reported that mismatch of gH antibody types between donors and recipients of renal transplantation in a D + /R+ setting, which probably indicates reinfection with a strain different from the original CMV strain, is associated with acute rejection after transplantation [15]. In this study, we revisited the risk of acute rejection in the same cases and found that 23 of the 27 recipients who experienced biopsy-proven acute rejection during the 6 months follow up after transplantation had antibodies against CMV gB AD2, indicating that the presence of antibodies against the gB AD2 may be a good predictor of rejection in recipients in a D + /R+ setting.

About 30–70% of CMV positive subjects have antibodies against gB AD2 [9, 17], which is one of the major epitopes for neutralizing antibodies [9, 11]. That the prevalence of antibodies against gB is similar in gH-matched and -mismatched recipients with acute rejection, suggests that the presence of gB antibodies is a risk factor, independent of mismatch of gH serotypes. Because of the limited number of recipients with acute rejection, further study of a larger patient group is required to confirm this finding. Nevertheless, we postulate that immune responses against CMV gB, which our ELISA system detected, may be associated with acute rejection. Although CMV-specific cellular immunity provides protection by limiting CMV reactivation and replication, it is plausible that acute rejection is a consequence of strong cell-mediated responses against ongoing CMV activity. Because gB is one of the significant targets for CMV-specific CD8+ and CD4+ T-cell immunity [10, 18], it would be interesting to ascertain whether CMV-specific T-cell activity against CMV-gB correlates with the outcome of our ELISA findings concerning gB AD2. Endogenous CMV-gB is presented efficiently by MHC Class II molecules of endothelial, epithelial and glial cells and can promote CD4+ T-cell recognition [19].

In conclusion, this study, which reevaluated a previous study, indicates that the presence of antibodies against gB in transplantation recipients may be a good indicator of possible acute rejection. Further study are needed to evaluate the association between antibody responses against gB and cellular immune responses in renal transplant recipients.


We thank all the subjects who participated in this study. This work was supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (No. 16591609).


No authors have any conflicts of interest to declare.