| |||||||||||||||||||||||||||||||||||||||
Syncytial giant-cell hepatitis is a rare but severe form of hepatitis that is associated with autoimmune diseases, drug reactions, and viral infections. We used serologic, molecular, and immunohistochemical methods to search for an infectious cause in a case of syncytial giant-cell hepatitis that developed in a liver-transplant recipient who had latent infection with variant B of human herpesvirus 6 (HHV-6B) and who had received the organ from a donor with variant A latent infection (HHV-6A). At the onset of the disease, the detection of HHV-6A (but not HHV-6B) DNA in plasma, in affected liver tissue, and in single micromanipulated syncytial giant cells with the use of two different polymerase-chain-reaction (PCR) assays indicated the presence of active HHV-6A infection in the patient. Expression of the HHV-6A–specific early protein, p41/38, but not of the HHV-6B–specific late protein, p101, was demonstrated only in liver syncytial giant cells in the absence of other infectious pathogens. The same markers of HHV-6A active infection were documented in serial follow-up samples from the patient and disappeared only at the resolution of syncytial giant-cell hepatitis. Neither HHV-6B DNA nor late protein was identified in the same follow-up samples from the patient. Thus, HHV-6A may be a cause of syncytial giant-cell hepatitis.
HHV-6, a ubiquitous beta-herpesvirus with two distinct variants, A and B, is the causative agent of sixth disease of childhood.6,7 After primary infection, HHV-6 establishes a latent state in the host, but it may reactivate and cause illness in immunocompromised patients.6,8 HHV-6 reactivation occurs in 40 to 50% of recipients of bone marrow and solid-organ transplants, mostly due to variant B, with variant A accounting for less than 2 to 3% of infections. Active infection (or reactivation) has been associated with bone marrow suppression, encephalitis, pneumonitis, and hepatitis.8 Occasionally, HHV-6 infection may result from primary infection as a consequence of virus transmission from the donor to the recipient through a transplanted organ.6,8 We describe a case of severe post-transplantation syncytial giant-cell hepatitis in an adult patient who was latently infected with HHV-6B.
Case Report
A 43-year-old man with Caroli's disease underwent orthotopic liver transplantation because of recurrent cholangitis. Serologic analyses that were performed before transplantation were positive for cytomegalovirus (CMV), Epstein–Barr virus (EBV), human herpes simplex viruses 1 and 2 (HSV-1 and HSV-2), varicella–zoster virus (VZV), and parvovirus B19 and were negative for human immunodeficiency virus (HIV) and hepatitis virus types A, B, and C. Serologic analyses of the donor were positive for CMV and EBV and negative for HIV and hepatitis virus types A, B, and C.
Induction immunosuppression consisted of a triple-drug schedule with tacrolimus, corticosteroids, and mycophenolate mofetil. Maintenance immunosuppression consisted of tacrolimus with trough blood levels of 10 to 15 ng per milliliter after postoperative day 2. On day 10, fever developed. The physical examination was unremarkable. A full blood count showed mild neutrophil leukocytosis, and liver-function tests revealed elevated levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) (Figure 1).
|
High-dose corticosteroid pulse treatment, followed by the administration of the monoclonal antibody muromonab-CD3 (OKT3), was undertaken for 5 days. Despite gradual improvement in AST and ALT levels, the cholestatic indexes increased and fever persisted (Figure 1). On day 13, the results of another liver biopsy showed improvement in the acute graft rejection (indeterminate, according to Banff grading) and the appearance of few giant-cell hepatocytes. On day 16, microbiologic and molecular examinations again were normal, including for the aforementioned viruses, except for HHV-6, which was positive on PCR. Intravenous ganciclovir was started at a dose of 5 mg per kilogram of body weight twice daily. On day 18, the results of a third liver biopsy showed complete resolution of acute graft rejection and diffuse giant transformation of hepatocytes, consistent with syncytial giant-cell hepatitis (Figure 2A). On day 34, after continued ganciclovir therapy, the fever resolved, but the cholestatic indexes remained elevated, and the full blood count disclosed anemia (hemoglobin, 8.5 g per deciliter), with an elevated reticulocyte count (40%) and lactate dehydrogenase level (2140 IU per liter), thrombocytopenia (platelet count, 45,000 per cubic millimeter), and mild renal failure (creatinine level, 2.1 mg per deciliter [186 µmol per liter]). The Coombs test and analyses for antiplatelet antibodies were negative. HHV-6 viremia was still detectable in plasma, in the absence of all the other pathogens, either by culture or molecular examination, and the results of a new liver biopsy showed the persistence of syncytial giant-cell hepatitis. A diagnosis of thrombotic microangiopathy was suspected.
|
Methods
Serologic Studies
The serologic assays that were performed have been described previously.10 The sources of virus antigens were the GS strain (variant A) of HHV-6 passaged in the T-cell acute lymphoblastic leukemia–derived cell line, HSB-2, and the Z29 strain (variant B) passaged in Molt3 cells. The serologic results were evaluated with sample identities masked.
Histopathological and Immunohistochemical Studies
Formalin-fixed and paraffin-embedded liver specimens were obtained from fine-needle biopsy from the patient on postoperative days 11, 13, 18, 40, and 44 and at 3 and 6 months and 1 year after transplantation. The specimens were stained with hematoxylin and eosin for routine histologic studies.
In situ hybridization and immunohistochemical analyses with standard immunoperoxidase-staining procedures were performed, as described previously,11,12 with the use of a mouse monoclonal antibody against the early protein (p41/38) of HHV-6A (Advanced Biotechnologies); a mouse monoclonal antibody against the late protein (p101K) of HHV-6B (Chemicon International); monoclonal or polyclonal antibodies against HSV-1 and HSV-2 (code numbers, B0114 and B0116, respectively) at a dilution of 1:200, against CMV (clone CCH2) at a dilution of 1:50, against EBV (EBER, PNA probe/Fluorescein), and against human papillomavirus (clone K1H8) at a dilution of 1:100 (all from Dakopatts); and monoclonal antibody against adenovirus (clone 20/11) at a dilution of 1:100 (Chemicon).11,12 Both HHV-6 monoclonal antibodies were applied for 1 hour at room temperature (dilution 1:200) after antigen retrieval in a microwavable pressure cooker (Nordic Ware) containing 1.5 liters of 10 mM citrate buffer (pH 6.0) with 0.1% Tween 20 and were microwaved (Model R4A80, Sharp Electronics) for 8 minutes at 700 W. They were visualized with the use of EnvisionPlus (Dako), according to the manufacturer's instructions.11,12
Molecular Studies and Micromanipulation and Single-Cell PCR
Molecular studies and micromanipulation and single-cell PCR were performed as described previously.13,14 A quantitative real-time PCR assay was performed with the use of a commercially available diagnostic kit (Nanogen Advanced Diagnostics) in order to quantify the HHV-6 load in the donor's peripheral-blood leukocytes, as well as in the five available liver-biopsy specimens from the patient. Results by quantitative PCR were confirmed by an independent laboratory. The presence of DNA from herpesvirus (EBV, CMV, HHV-6, HHV-8, HHV-7, HSV-1, HSV-2, and VZV), adenovirus, and polyomavirus (either JC or BK) was confirmed in micromanipulated single giant-cell hepatocytes by PCR analysis with the use of protocols that have been described previously.13,14,15,16,17 Results were confirmed by an independent laboratory in a double-blinded manner.
To obtain efficient PCR amplification of HHV-6 DNA in single cells from formalin-fixed, paraffin-embedded tissues, two new primers were designed to amplify a 306-bp fragment (left primer, 5'ATCACGATCGGCGTGCTAT3'; right primer, 5'ATGGATTTCCGTGGAAGAAA3'). The number of cycles was 35 at an annealing temperature of 55°C. The PCR conditions were the same as those reported previously.13
The characterization of HHV-6 variants was performed by restriction analysis of the HHV-6 PCR product13 and also by a highly sensitive primer-specific PCR assay, as described previously.18 Moreover, in order to obtain efficient PCR amplification of HHV-6 in single cells from formalin-fixed, paraffin-embedded tissues with the primer-specific PCR assay, the probe, as originally described by Boutolleau et al.,18 was used as an internal primer for a second round of PCR to amplify a 113-bp fragment.
Results
Serologic Studies
On the day of transplantation, serum from the donor was negative for IgM and positive for IgG antibodies against HHV-6 at a titer of 1:80, and serum from the patient was negative for IgM and positive for IgG antibodies against HHV-6 at a titer of 1:40 (Figure 1). After urea treatment of the samples, the HHV-6 IgG titers, from both the donor and the patient, were reduced by a factor of 4 or less (log2 reduction, 
2), indicative of high-avidity antibodies and consistent with the HHV-6 seropositive status.
On postoperative day 16, the patient had both a positive IgM titer (1:40) and a rising titer of IgG antibodies against HHV-6. The avidity of the latter was low, as shown by the reduction in the IgG titer by a factor of 8 after washing with urea (from 1:2560 to 1:320; log2 reduction, 3). A reduction in the antibody titer by a factor of 8 after urea treatment is mild and not conclusive for the presence of low-avidity IgG in our patient, although a similar finding in a transplant recipient was described previously.10 Since the serologic assay cannot discriminate between the two HHV-6 variants, we hypothesized that this borderline drop in IgG after urea treatment could be ascribed to the presence of a mixture of low-avidity IgG antibodies against HHV-6A and high-avidity IgG antibodies against HHV-6B so that this background of high-avidity antibodies would have prevented a sharper decrease in the IgG titers. These serologic findings, which suggested possible active viral infection, spurred us to investigate whether HHV-6A was the replicating virus in our patient with latent HHV-6B infection (Figure 1).
Molecular Studies
The identification of HHV-6 variants A and B in all samples from the donor and the patient was confirmed by two different assays: a primer-specific PCR assay (Figure 3)13 and a restriction analysis of the HHV-6 PCR product (Figure 4E and 4F).18 At the time of transplantation, HHV-6A (but not HHV-6B) DNA was detected in peripheral-blood leukocytes from the donor, with a viral load of 4500 genome equivalents (gEq) per microgram of DNA. At the same time, HHV-6B (but not HHV-6A) DNA was detected in peripheral-blood leukocytes from the patient on PCR (Figure 3A). On PCR assay, HHV-6A DNA was detected in the patient's plasma at a relatively low titer (20 to 40 genomes per milliliter), starting on postoperative day 16 and persisting at similar levels until the sixth month after transplantation, whereas HHV-6B DNA was never detected in the same plasma samples (Figure 1).
|
|
Single-cell PCR that was performed on micromanipulated giant-cell hepatocytes from the liver-biopsy specimen on day 18 confirmed the presence of HHV-6A DNA, at an average level of 500 to 850 genomes per cell, but not HHV-6B or other viruses mentioned above. HHV-6A DNA was found only in giant-cell hepatocytes with an amplification efficiency ranging from 40 to 50%14 (Figure 3B and Figure 4A through 4F). HHV-6A DNA was not detected in any of the 90 to 100 normal uninuclear hepatocytes, whereas Y-chromosome DNA could be amplified in 10 to 20% of the same cells (Figure 3C and 3D).
Immunohistochemical Studies
Immunohistochemical analysis of the liver-biopsy specimen obtained on day 18 with the use of p41/38 monoclonal antibody against HHV-6A highlighted cytoplasmic reactivity only in giant-cell hepatocytes, confirming their productive infection by HHV-6A (Figure 2B). The result of such analysis was negative with antibodies against the other viruses (Figure 2C). The same analysis for HHV-6A had negative results on a liver-biopsy sample without syncytial giant-cell hepatitis, collected 1 year after transplantation (Figure 2D). Immunohistochemical analysis with the HHV-6B p101K monoclonal antibody on a liver-biopsy specimen obtained on day 18 was negative (Figure 2E). Such analysis showed intense staining of follicular dendritic cells of a reactive germinal center in a patient with HHV-6–related lymphadenopathy, who was tested as a positive control subject and was found to have HHV-6B DNA11,12 (Figure 2F).
Discussion
In this patient, active HHV-6 infection was manifested as syncytial giant-cell hepatitis. These findings implicate HHV-6 as a possible new cause of this disease. In most cases, syncytial giant-cell hepatitis that is caused by an infectious agent has been related to a paramyxoviral source, as determined either by the finding of structures resembling nucleocapsids of paramyxoviruses in hepatocytes or by the theory that the fusion of cells to form syncytial giant cells is a pathologic feature of paramyxovirus infection.1,2,3,4,5
A possible relationship between HHV-6 and syncytial giant-cell hepatitis has been suggested previously on the basis of serologic positivity to HHV-6 in a single patient, in whom both an acute drug reaction and an autoimmune disease might have been other causative factors.19 The demonstration of HHV-6A active infection by several methods in plasma, liver tissue, and single giant-cell hepatocytes, which coincided with the onset of the patient's symptoms and histologic diagnosis, provides strong evidence that HHV-6A had a pathogenic role in this case. Extensive and repeated microbiologic testing during the entire course of the disease did not identify any other potential cause. Of interest, HHV-6 has been reported to induce cell–cell fusion in lymphocytes and epithelial cells both in vitro and in vivo in a manner similar to that of paramyxoviruses.20,21,22
We have also documented the transmission of HHV-6A from a liver donor to a recipient with latent HHV-6B infection by showing the presence of either HHV-6A DNA or antigen in plasma and affected liver tissue from the recipient after transplantation. It has been reported that the detection of HHV-6 DNA in plasma, even at low titers, is correlated with active viral replication in immunosuppressed patients.23,24 HHV-6 chromosomal integration may occur in vivo, either in patients with various diseases or in healthy subjects, and is associated with the persistence of a high viral load in infected cells.6,25 The viral load in a sample of peripheral-blood leukocytes from the donor at the time of transplantation was elevated but not at a level typically seen with HHV-6 chromosomal integration.25
Of note, it has been reported that HHV-6 is frequently reactivated in critically ill patients as a result of stress-related mechanisms, without affecting the severity of disease.26 It remains to be established whether such a reactivation is critical to the viral transmission and whether it could enhance pathogenicity in organ recipients. Furthermore, the progressive decrease in the viral load in serial liver-biopsy and plasma samples from the patient makes it unlikely that integrated HHV-6 was transmitted through liver transplantation and stably retained in the recipient's liver. It is also unlikely that HHV-6 DNA that was detected in the patient's plasma was simply the result of the shedding of stably infected cells. On the other hand, the absence of HHV-6B DNA and antigen in plasma and affected liver tissue in a patient with latent HHV-6B infection strongly suggests that HHV-6B did not have a pathogenic role.
These findings also raise issues about the immune response to the two HHV-6 variants and provide further arguments in support of systematic virologic screening of organ donors. Although HHV-6A and HHV-6B are supposed to stimulate cross-reactive T-cell responses because they share more than 88% sequence homology,27,28 it has been reported that at least 7% of the T-cell clones that are reactive to HHV-6 show specific and distinct patterns of proliferation either to variant A or variant B in vitro.29 The HHV-6A active infection in our patient suggests that the immune response that was mounted against HHV-6B may not have been protective against HHV-6A. Thus, more extensive screening of organ donors, on the basis of a wider panel of tests for both variants, may specifically address the infectious risks of the recipients and have consequences in designing prophylactic or preemptive therapeutic strategies, as in the case of CMV.8 In our patient, we continued ganciclovir for a prolonged time, because a few cases of syncytial giant-cell hepatitis were reported to have resolved with prolonged courses of ribavirin.30,31 It is unclear which factor — the use of antiviral therapy or the patient's recovery of immune function — led to the resolution of disease in this patient over several months.32,33 In conclusion, HHV-6 should be included among the possible causes of syncytial giant-cell hepatitis, at least in liver-transplant recipients.
Supported by grants from the Italian Ministry for Education, Universities, and Research (to Dr. Torelli), the Associazione Italiana per la Ricerca sul Cancro (to Dr. Luppi), the European Commission's Sixth FrameWork Programme, INCA project (2005-018704) (to Drs. Luppi and Schulz), the Associazione Italiana Lotta alle Leucemie, Linfoma e Mieloma, Modena Omlus (to Dr. Potenza), and the Programma di Ricerca Regione Università 2007–2009 (to Dr. Torelli).
No potential conflict of interest relevant to this article was reported.
We thank Claudio Cermelli and Gianandrea Pasquinelli for their technical advice and Alberto Bagni and Antonio Daniele Pinna for their clinical follow-up of the patient.
Source Information
From the Department of Oncology and Hematology (L.P., M.L., P.B., G.T.), the Integrated Department of Diagnostic and Laboratory Services and Legal Medicine (G.R., M. Pecorari, W.G., M. Portolani), the Department of Internal Medicine and Medical Specialties (S.C., M.C., G.G.), and the Liver and Multivisceral Transplant Center (M.M., F.D.B., G.E.G.) — all at the University of Modena and Reggio Emilia, Azienda Ospedaliera Policlinico, Modena; and the Department of Clinical and Experimental Medicine, University of Bologna, and St. Orsola Malpighi General Hospital, Bologna (T.L., M.P.L.) — all in Italy; and the Department of Virology, Hannover Medical School, Hannover, Germany (T.F.S.).
Drs. Potenza, Luppi, and Barozzi contributed equally to this article.
Address reprint requests to Dr. Luppi at the Department of Oncology and Hematology, University of Modena and Reggio Emilia, Azienda Ospedaliera Policlinico, Via del Pozzo 71, 41100 Modena, Italy, or at mario.luppi{at}unimore.it.
References
| |||||||||||||||||||||||||||||||||||||||
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
-GT 
