Ganciclovir Prophylaxis Improves Late Murine Cytomegalovirus-Induced Renal Allograft Damage : Transplantation

Journal Logo

Basic and Experimental Research

Ganciclovir Prophylaxis Improves Late Murine Cytomegalovirus-Induced Renal Allograft Damage

Shimamura, Masako1,6; Seleme, Maria C.1; Guo, Lingling2; Saunders, Ute1,5; Schoeb, Trenton R.3; George, James F.2; Britt, William J.1,4

Author Information
Transplantation Journal 95(1):p 48-53, January 15, 2013. | DOI: 10.1097/TP.0b013e3182782efc
  • Free


Human cytomegalovirus (CMV) infection is associated with adverse direct and indirect effects in renal transplantation ranging from acute rejection to inferior long-term allograft outcome (1–4). Ganciclovir antiviral prophylaxis has been correlated with improved survival in clinical studies (5). In animal models, murine CMV (MCMV) and rat CMV accelerate both acute and late kidney transplant rejection (6–10).

Prior studies in animal models have demonstrated accelerated and intensified recruitment of leukocytes into CMV-infected renal allografts compared with uninfected grafts. Early lymphocytic infiltration is described in both uninfected and CMV-infected grafts, consistent with an allogeneic response, but is more abundant in the CMV-infected animals, suggesting a potential role for antiviral lymphocyte activation. Mononuclear infiltrates are recruited preferentially to CMV-infected grafts, compared with uninfected grafts at both early and late times after transplantation, as well as NK cells, Gr-1+ myeloid cells, and antigen-presenting CD11c+ cells at early times after transplantation (6, 7, 9, 10). CMV infection is also associated with augmented induction of adhesion molecules, lymphoid activation markers, proinflammatory chemokine profiles, and fibrogenic molecules within infected allografts.

In cardiac and renal transplantation models of CMV infection, treatment with ganciclovir improves histopathologic manifestations of rejection (10, 11). In the cardiac allograft model, 30 days of ganciclovir prophylaxis improves cardiac allograft vasculopathy at 90 days after transplantation. Ganciclovir prophylaxis in the renal allograft model also is associated with decreased graft infiltration by myeloid, antigen-presenting, mononuclear, and NK cells compared with untreated CMV-infected grafts. The late effects of ganciclovir prophylaxis on renal allograft histology have not been previously examined in the animal model. We undertook the following study to investigate whether short-course (14 days) ganciclovir prophylaxis affects late (42 days) histology and inflammation using the murine renal transplantation model.


Late Allograft Histology Is Improved by Ganciclovir Prophylaxis

MCMV-infected BALB/c kidneys were transplanted into MCMV naïve C57BL/6 mice (D+/R- combination) with cyclosporine immunosuppression. Control transplants were performed between uninfected donors and recipients (D-/R-). For the experimental ganciclovir prophylaxis group, D+/R- animals were treated with ganciclovir for 14 days starting immediately after transplantation. Animals were sacrificed at day 42 and organs were harvested for pathologic and flow cytometric analysis. Pathology was analyzed by a veterinary pathologist blinded to sample identity. Histology was scored according to a grading scale devised by the pathologist based on the criteria used for grading of clinical renal transplant biopsies and histopathologic features of renal allograft rejection in rodents (9, 12–14).

At day 42 after transplantation, MCMV-infected allografts were more severely damaged than control uninfected grafts (Fig. 1). The total histology score for uninfected grafts was 8.3±0.9, whereas the score for MCMV-infected grafts was 15.6±0.6 (P<0.01). The greatest differences were seen in categories of interstitial inflammation, edema, and perivascular inflammation. Also notable was the global disruption of cytoarchitecture observed in MCMV-infected grafts, compared with a patchy distribution of inflammation and tissue destruction in the uninfected grafts, in which areas of relatively preserved glomeruli and tubules were still observable. MCMV-infected grafts receiving ganciclovir prophylaxis showed significantly less severe histologic injury compared with MCMV-infected grafts without prophylaxis, with a total histology score of 9.5±1.4 (P<0.01). Ganciclovir-treated grafts had a histologic appearance more similar to uninfected grafts than infected grafts, including the patchy distribution of leukocytic infiltrates in a peritubular distribution, and the focality of tissue damage with interspersed preservation of glomeruli and tubules; however, ganciclovir-treated grafts did contain areas of perivascular leukocytic infiltrates resembling those in MCMV-infected grafts without ganciclovir prophylaxis.

Improvement of histologic features in MCMV-infected allografts after ganciclovir prophylaxis. At day 42 after transplantation, uninfected renal allografts (No CMV), MCMV-infected grafts (CMV), and MCMV-infected grafts with ganciclovir prophylaxis (CMV+GCV) were analyzed for histologic damage using a grading scale resembling that used for human renal transplant biopsies. Criteria were graded from 0 (none) to 3 (severe). Total score for CMV grafts was significantly different (**P<0.01) from scores of both uninfected and ganciclovir-treated grafts. CMV, cytomegalovirus; MCMV, murine cytomegalovirus.

MCMV Infection Induces NK Cells and Myeloid Infiltrates That Are Diminished With Ganciclovir Prophylaxis

To define the nature of the leukocyte infiltrates, flow cytometric analysis of allograft infiltrating CD45+ leukocytes was performed. NK cells were quantitated as CD45+/CD3-/CD49b+ cells and compared between experimental groups (Fig. 2A). MCMV-infected grafts contained significantly greater frequencies of NK cells compared with uninfected grafts (P<0.01). Ganciclovir-treated grafts exhibited a reduction in NK cell infiltrates toward levels displayed by uninfected grafts, such that the NK cell frequency became statistically similar to both uninfected and MCMV-infected grafts. In addition, the frequency of myeloid cells expressing the surface markers CD45, CD11b, and Gr-1 (Fig. 2B) were also statistically greater in MCMV-infected grafts compared with uninfected grafts, with diminution in ganciclovir-treated grafts toward frequencies statistically similar to uninfected grafts (P>0.05). In contrast, the frequencies of CD4+ and CD8+ T lymphocytes (Fig. 2C) in the CD45+ leukocyte population did not show statistically significant differences among uninfected, MCMV-infected, and ganciclovir-treated grafts at day 42 after transplantation. These results resemble those found in allografts at 14 days during ganciclovir prophylaxis (10), with an induction of NK cells and myeloid infiltrates in MCMV-infected grafts compared with uninfected grafts, and attenuation of these infiltrates (but not T cells) by ganciclovir treatment. These results indicate that the reduction in NK and myeloid cells in the ganciclovir-treated grafts is durable after the period of antiviral prophylaxis.

MCMV-induced infiltration of NK and myeloid cells decreases after ganciclovir prophylaxis. CD45+ cells from allografts at day 42 after transplantation were analyzed by flow cytometry for (A) NK cells (CD3-/CD49b+ cells), (B) myeloid cells (CD11b+/Gr-1+ cells), and (C) CD4+ and CD8+ T lymphocytes. CMV-infected grafts showed significantly greater NK cell (**P<0.01) and myeloid (*P<0.05) infiltrates than uninfected grafts, and ganciclovir prophylaxis decreased these infiltrates to levels statistically indistinguishable from uninfected grafts. CD4+ and CD8+ lymphocyte frequencies were not significantly different between groups. CD45+ cells in spleen (D) and liver (E) were analyzed for frequencies of CD4+, CD8+, CD3-/CD49b+, and Gr-1+ cells in transplant recipients receiving MCMV- grafts (No CMV), MCMV+ grafts (CMV), and MCMV+ grafts treated with ganciclovir (CMV+GCV). The frequencies of all cell types were similar between all groups, with the exception of splenic CD8+ cells that were statistically fewer in frequency in the spleens of the CMV+GCV group compared with the spleens from the No CMV group (*P<0.05). CMV, cytomegalovirus; MCMV, murine cytomegalovirus; NK cells, natural killer cells.

Next, systemic immune responses were interrogated by analyzing leukocyte populations in the spleens (Fig. 2D) and livers (Fig. 2E) from transplant recipients receiving uninfected transplants (No CMV, white bars), MCMV-infected transplants (CMV, black bars), and MCMV-infected transplants with ganciclovir prophylaxis (CMV+GCV, checkered bars). CD45+ populations were analyzed for frequencies of CD4+, CD8+, CD3-/CD49b+ (NK cells), and Gr-1+ cells. In the spleen (Fig. 2D), only CD8+ cells showed any statistically significantly different results: ganciclovir prophylaxis was associated with a statistically lower frequency of splenic CD8+ cells compared with uninfected transplants (asterisk) but was similar to the MCMV-infected recipients without ganciclovir prophylaxis. The frequencies of CD8+ cells in the liver were similar for all experimental groups. No differences in frequencies of CD4+, CD3-/CD49b+, and Gr-1+ cells were found in spleens or livers among the three experimental groups. These results indicate that the differences among the intragraft NK cell and Gr-1 responses were specific to the allograft and were not manifested in the systemic responses at day 42 after transplantation.

NK Cell Depletion Improves MCMV-Associated Allograft Injury

Next, to examine whether the quantity of NK cell infiltrates directly influences the degree of graft injury in MCMV-infected allografts, NK cells were depleted from D+/R- MCMV-infected transplants via treatment with anti-NK1.1 antibodies at days 0 and 7, and histology was analyzed at day 14 after transplantation. Animals undergoing NK cell depletion remained healthy, similar to animals not receiving anti-NK1.1 antibodies, and had no clinical symptoms of illness from viral infection. The efficacy of depletion was confirmed by flow cytometric analysis for CD45+/CD3-/NKp46+ cells (Fig. 3A,B), which showed a virtually complete absence of NK cells (Fig. 3B, left, checkered bar) in anti-NK1.1–treated allografts at day 14 after transplantation (7 days after the last antibody treatment) compared with untreated allografts that showed robust infiltration of NK cells (Fig. 3B, left, black bar). Allograft histology showed improved damage scores in anti-NK1.1–treated grafts (total score, 7.5±1.0) compared with untreated grafts (total score, 11.83±1.4; P<0.05; Fig. 3C). The frequencies of myeloid cells were similar in untreated and NK cell–depleted allografts (Fig. 3B, right), indicating that NK cell depletion did not influence the myeloid infiltrates into MCMV-infected allografts. These results indicate that direct depletion of NK cells can ameliorate virus-associated allograft injury.

Depletion of NK cells results in improved allograft histologic scores. Recipients of MCMV-infected allografts were treated with anti-NK1.1 antibodies to deplete NK cells (anti-NK1.1) and compared with recipients without depletion (CMV) at day 14 after transplantation. A, NK cell depletion was confirmed via flow cytometry of allograft infiltrates for CD3-/NKp46+ cells, gated on CD45+ cells. B, (left) frequencies of CD3-/NKp46+ cells in allograft were significantly lower (*P<0.05) in untreated grafts (CMV) compared with treated grafts (anti-NK1.1); (right) frequencies of myeloid cells were similar in untreated and NK cell–depleted allografts. C, histopathologic analysis of allografts showed improved damage scores in NK cell–depleted animals (anti-NK1.1) compared with undepleted animals (CMV; *P<0.05). CMV, cytomegalovirus; MCMV, murine cytomegalovirus; NK cells, natural killer cells.


This study shows that short-term ganciclovir prophylaxis improves late allograft tissue damage in the murine renal transplant model. MCMV-infected grafts without antiviral treatment showed intense leukocytic infiltrates in the peritubular and perivascular areas of cortex and medulla, resulting in global tubular destruction with secondary loss of glomeruli by day 42 after transplantation. In contrast, uninfected grafts demonstrated patchy leukocyte infiltrates located in a peritubular and perivascular distribution, with relative preservation of glomeruli and interspersed with areas of normal tubules without significant peritubular leukocytic infiltration. Flow cytometric analysis suggested that these infiltrates in the uninfected grafts were largely CD4+ and CD8+ lymphocytes, which were also present in the MCMV-infected grafts. The major difference in the cell types found in MCMV-infected and uninfected grafts was found by flow cytometry to consist of NK cells and Gr-1+ myeloid cells. The analysis of grafts after ganciclovir prophylaxis showed histologic findings resembling the uninfected grafts, with patchy leukocyte infiltrates and areas of relatively preserved tubules and glomeruli, and the presence of CD4+ and CD8+ lymphocytes but continued modulation of NK cells and myeloid infiltrates after ganciclovir prophylaxis. These results are consistent with findings described previously at day 14 after transplantation using this same model, in that ganciclovir prophylaxis did not reduce early CD4+ and CD8+ infiltration into MCMV-infected allografts (10) but was associated with reduced NK cells and Gr-1+ myeloid infiltrates during prophylaxis and a moderate increase in these cell types at 1 week after the cessation of ganciclovir prophylaxis (10). It is now shown in the current study that the infiltration of NK and myeloid cells into the infected allograft was still moderated in the grafts receiving ganciclovir prophylaxis, even at late times after the cessation of antiviral administration. Systemic splenic and liver responses did not reflect these intragraft differences in NK and myeloid cell recruitment, suggesting that local factors within the allografts might contribute to these differences in intragraft immune responses. The mechanism by which these differences in intragraft leukocyte recruitment are induced remains undefined.

Because ganciclovir prophylaxis was associated with decreased NK cell infiltrates even after the cessation of prophylaxis, the impact of NK cell depletion on MCMV-induced allograft injury was also interrogated. The NK cell–depleted MCMV-infected allografts showed decreased allograft injury despite the concurrent presence of myeloid infiltrates. This result suggests that NK cells may contribute to virus-associated allograft injury by direct or indirect mechanisms. It remains undefined whether ganciclovir prophylaxis directly influences NK cells or whether the reduction in NK cells serves as a surrogate marker for other functions of antiviral therapy in ameliorating virus-associated graft injury. Further studies could elucidate mechanisms by which antivirals modulate tissue injury in CMV-infected allografts and define more clearly the significance of the NK cells in this pathogenesis.

In this animal model, MCMV-infected allografts contained greater NK cell infiltrates compared with uninfected grafts at both early and late times after transplantation. CMV infection is well described to activate NK cells, which are critical for the early control of CMV infections (15). In renal transplant patients, peripheral blood NK cells increase during human CMV infection and demonstrate an activated phenotype (16, 17). In clinical biopsies from renal transplant patients, NK cells are found by immunohistochemical staining and transcriptome analysis at early times after transplantation in kidneys, which subsequently demonstrate late rejection, suggesting that NK cell infiltration into allografts may influence late graft outcomes (18, 19). To date, intragraft NK cells have not been examined in the context of human CMV infection in clinical populations and might provide relevant insight into the mechanisms of indirect effects attributed to human CMV infection in renal transplant patients.

An alternative explanation for the described findings, not explored in these studies, includes the role of NKT cells. Although interest has been developing in this cell population in renal transplantation, NKT cells have not been well studied in the context of CMV infections and transplantation but could constitute a population depleted by the anti-NK1.1 antibody. Future studies may dissect the respective roles of NKT cells from NK cells in MCMV-associated graft injury by using this murine renal transplantation model before analyzing these leukocyte populations in clinical transplantation.

In summary, studies in the murine model demonstrate that MCMV-infected renal allografts experience more severe tissue injury at late times after transplantation in association with persistent NK cell infiltrates. Ganciclovir prophylaxis results in decreased NK cell infiltration and improved allograft histology even at 28 days after the cessation of prophylaxis. Direct NK cell depletion also ameliorates MCMV-associated allograft damage. These results indicate that ganciclovir exerts modulatory effects on late tissue damage in MCMV-infected allografts and suggest a potential role of NK cells in the pathogenesis of MCMV infection in renal transplantation.


Virus and Animals

MCMV strain Smith with an ORF m157 deletion (MCMVSmithΔ157) was propagated, prepared, and stored as described previously (10). Female BALB/cJ or C57BL/6J mice (The Jackson Laboratory, Bar Harbor, ME) were maintained in American Association for the Accreditation of Laboratory Animal Care–approved animal facilities maintained by the Animal Resources Program of the University of Alabama at Birmingham (Birmingham, AL) under specific pathogen-free conditions (health surveillance protocol available at Mouse maintenance and experimental protocols were approved by the University of Alabama at Birmingham Institutional Animal Care and Use Committee.

Renal Transplantation Surgery

For MCMV-infected transplants, donor BALB/cJ (H-2d) mice were infected by intraperitoneal injection with 104 plaque-forming units of MCMVSmithÄm157 strain virus at least 12 weeks before renal transplantation (D+/R- transplants). Donors without MCMV infection (D-/R- transplants) were used as controls. Allogeneic orthotopic kidney transplantation from donor BALB/cJ mice into recipient C57BL/6J (H-2b) mice (R-) was performed as described previously (10, 20). The contralateral native kidney of the recipient was left intact because life-sustaining transplantation was not required for these experiments. Recipients were treated with cyclosporine (Novartis Pharmaceuticals, East Hanover, NJ) at 10 mg/kg per day subcutaneously once daily starting immediately postoperatively for 14 days (21). One experimental group of D+/R- animals was treated with ganciclovir prophylaxis (Roche Laboratories, Nutley, NJ) at 15 mg/kg per day subcutaneously once daily starting immediately postoperatively for 14 days (22). A separate experimental group of D+/R- recipients was treated with anti-NK1.1 antibodies (clone PK136; eBioscience, San Diego, CA) at 200 µg/dose intraperitoneally at days 0 and 7 after transplantation (23). Results from four to five animals were analyzed per experimental group.

Flow Cytometric Analysis of Allograft Immune Infiltrates

Organs were harvested at day 14 or 42 after transplantation and processed for flow cytometric analysis as described previously (10). After blocking Fc receptors with anti-mouse CD16/CD32 (clone 93; eBioscience), flow cytometric analysis of single-cell suspensions was performed using a combination of the following monoclonal anti-mouse antibodies (eBioscience): fluorescein isothiocyanate (FITC)– or phycoerythrin-conjugated CD45 (30-F11), FITC-conjugated Gr-1 (RB6-8C5), FITC- or peridinin chlorophyll protein–conjugated CD3e (145-2C11), phycoerythrin-conjugated CD8α (53-6.7), allophycocyanin-conjugated CD4 (GK1.5), allophycocyanin-conjugated CD49b (DX5), and FITC-conjugated NKp46 (29A1.4). Flow cytometry studies were performed using a dual laser FACSCalibur and analyzed using FlowJo software (BD Biosciences, San Jose, CA). Results were quantitated as frequencies after gating on live cells expressing the pan-leukocyte marker CD45.

Histology and Scoring

Allografts were perfused with saline to organ pallor, and portions were fixed for 24 hr in 10% neutral buffered formalin (Sigma, St. Louis MO), processed routinely for paraffin embedding and sectioning, and stained with hematoxylin-eosin. Sections were evaluated by a veterinary pathologist (T.R.S.) blinded to sample identity using a scale devised for this study, based on the clinical Banff criteria for renal allograft histology scoring as well as scales published for grading of rodent renal allografts (9, 13, 14). The Banff criteria used for patient biopsy grading could not be used directly in this study because the kinetics of rejection in the animal model is more rapid compared with that observed in patients. Criteria in the grading scale included glomerular changes (primarily sclerosis and, to a lesser extent, proliferation); tubular degeneration, including atrophy and epithelial cell necrosis; interstitial inflammatory cells; interstitial fibrosis; edema; perivascular inflammatory cell accumulation; arteritis (primarily endarteritis with occasional focally necrotizing arteritis); necrosis (foci of complete necrosis of multiple tubules); and capsulitis. Each was scored 0 to 3 for absent, mild, moderate, or severe, respectively, and individual scores were summed to yield an overall score with a maximum possible value of 27.

Statistical Analysis

All assays were analyzed using four to five animals for each experimental group. Groups were analyzed using analysis of variance and pairs were compared using the Student’s t test using Prism 5.0 software, accepting statistically significant differences at P<0.05 (GraphPad, San Diego, CA).


1. Freeman RB Jr. The ‘indirect’ effects of cytomegalovirus infection. Am J Transplant 2009; 9: 2453.
2. Rubin RH, Tolkoff-Rubin NE, Oliver D, et al.. Multicenter seroepidemiologic study of the impact of cytomegalovirus infection on renal transplantation. Transplantation 1985; 40: 243.
3. Gerstenkorn C, Balupuri S, Mohamed MA, et al.. The impact of cytomegalovirus serology for 7-year graft survival in cadaveric kidney transplantation—the Newcastle experience. Transplant Int 2000; 13: S372.
4. Fitzgerald JT, Gallay B, Taranto SE, et al.. Pretransplant recipient cytomegalovirus seropositivity and hemodialysis are associated with decreased renal allograft and patient survival. Transplantation 2004; 77: 1405.
5. Kliem V, Fricke L, Wollbrink T, et al.. Improvement in long-term renal graft survival due to CMV prophylaxis with oral ganciclovir: results of a randomized clinical trial. Am J Transplant 2008; 8: 975.
6. Lautenschlager I, Soots A, Krogerus L, et al.. Effect of cytomegalovirus on an experimental model of chronic renal allograft rejection under triple-drug treatment in the rat. Transplantation 1997; 64: 391.
7. Lautenschlager I, Soots A, Krogerus L, et al.. Time-related effects of cytomegalovirus infection on the development of chronic renal allograft rejection in a rat model. Intervirology 1999; 42: 279.
8. Inkinen K, Holma K, Soots A, et al.. Expression of TGF-beta and PDGF-AA antigens and corresponding mRNAs in cytomegalovirus-infected rat kidney allografts. Transplant Proc 2003; 35: 804.
9. Soule JL, Streblow DN, Andoh TF, et al.. Cytomegalovirus accelerates chronic allograft nephropathy in a rat renal transplant model with associated provocative chemokine profiles. Transplant Proc 2006; 38: 3214.
10. Shimamura M, Saunders U, Rha B, et al.. Ganciclovir transiently attenuates murine cytomegalovirus-associated renal allograft inflammation. Transplantation 2011; 92: 759.
11. Lemstrom K, Sihvola R, Bruggeman C, et al.. Cytomegalovirus infection-enhanced cardiac allograft vasculopathy is abolished by DHPG prophylaxis in the rat. Circulation 1997; 95: 2614.
12. Mannon RB, Kopp JB, Ruiz P, et al.. Chronic rejection of mouse kidney allografts. Kidney Int 1999; 55: 1935.
13. Solez K, Colvin RB, Racusen LC, et al.. Banff ‘05 Meeting Report: differential diagnosis of chronic allograft injury and elimination of chronic allograft nephropathy (‘CAN’). Am J Transplant 2007; 7: 518.
14. Solez K, Colvin RB, Racusen LC, et al.. Banff 07 classification of renal allograft pathology: updates and future directions. Am J Transplant 2008; 8: 753.
15. Bukowski JF, Woda BA, Welsh RM. Pathogenesis of murine cytomegalovirus infection in natural killer cell-depleted mice. J Virol 1984; 52: 119.
16. Venema H, Van Den Berg AP, Van Zanten C, et al.. Natural killer cell responses in renal transplant patients with cytomegalovirus infection. J Med Virol 1994; 42: 188.
17. van Dam JG, Damoiseaux JG, Christiaans MH, et al.. Acute primary infection with cytomegalovirus (CMV) in kidney transplant recipients results in the appearance of a phenotypically aberrant CD8+ T cell population. Microbiol Immunol 2000; 44: 1011.
18. Hancock WW, Gee D, De Moerloose P, et al.. Immunohistological analysis of serial biopsies taken during human renal allograft rejection. Changing profile of infiltrating cells and activation of the coagulation system. Transplantation 1985; 39: 430.
19. Vitalone MJ, Naesens M, Sigdel T, et al.. The dual role of epithelial-to-mesenchymal transition in chronic allograft injury in pediatric renal transplantation. Transplantation 2011; 92: 787.
20. Zhang Z, Schlachta C, Duff J, et al.. Improved techniques for kidney transplantation in mice. Microsurgery 1995; 16: 103.
21. Andoh TF, Gardner MP, Bennett WM. Protective effects of dietary L-arginine supplementation on chronic cyclosporine nephrotoxicity. Transplantation 1997; 64: 1236.
22. Kern ER, Bidanset DJ, Hartline CB, et al.. Oral activity of a methylenecyclopropane analog, cyclopropavir, in animal models for cytomegalovirus infections. Antimicrob Agents Chemother 2004; 48: 4745.
23. Hirohashi T, Chase CM, Della Pelle P, et al.. A novel pathway of chronic allograft rejection mediated by NK cells and alloantibody. Am J Transplant 2011; 12: 313.

Cytomegalovirus; Kidney; Transplantation; Ganciclovir

© 2013 Lippincott Williams & Wilkins, Inc.