- Email: [email protected]
Accumulation of p53 in infectious mononucleosis tissues
Accumulation of p53 in Infectious Mononucleosis Tissues AAMIR EHSAN, MD, HONGXIN FAN, MD, PHYLLIS A. EAGAN, BS, HUMA A. SIDDIQUI, MD, AND MARGARET L. GULLEY, MD Epstein-Barr virus (EBV) infects lymphocytes, where it persists indefinitely for the life of the host; whether the virus interacts with p53 to maintain itself in these cells is unknown. Lymphoid biopsy samples from 10 patients with infectious mononucleosis (IM) were examined for expression of p53 by immunohistochemistry. Accumulation of p53 was detected in all 10 cases, primarily in large lymphocytes of the expanded paracortex. T h e presence of EBV was confirmed in all 10 cases by EBER1 (EBV-encoded RNA) in situ hybridization, whereas 11 non-IM control samples lacked significant EBER1 and did not express p53 in paracortical lymphocytes. Interestingly, EBV infection alone does not cause accumulation of intracellular p53, because many more cells expressed EBER1 than p53 in the IM tissues. To determine whether p53 was confined to the subset of infected cells in which viral replication was occurring, BZLF1 immunostains were performed. Viral BZLF1 was detected in 8 of 10 IM tissues; however, the paucity and small size of the BZLFl-expressing lymphocytes suggests that they are not the same cells overexpressing p53. To further examine the relationship between p53 and EBV
Epstein-Barr virus (EBV) is a lymphotropic herpes virus that infects more than 90% of humans worldwide. Primary EBV infection in early childhood is generally asymptomatic, but when infection is delayed until adolescence or adulthood it causes the clinical syndrome of infectious mononucleosis (IM). IM represents an EBVdriven proliferation of benign lymphocytes that is eventually controlled by humoral and cellular immune responses. When EBV infects human B lymphocytes, there are 3 possible cellular responses: latent persistence of the EBV genome, replicative production of new virions, and cell death secondary to the host response to infection. In the latent phase of infection, viral DNA persists indefinitely inside the cell and is passed on to cellular progeny during mitosis. In the replicative phase of infection, a cascade of viral and host proteins are expressed to mediate viral genome replication and packaging into infectious virions. The master switch controlling the transition from latent to replicative EBV
From the Department of Pathology, University of Texas Health Science Center at San Antonio, and the Audie L. Murphy Memorial Veteran's Hospital, San Antonio, TX. Accepted for publication August 8, 2000. Supported by a grant from the Veteran's Administration. Address correspondence and reprint requests to Aamir Ehsan, MD, Department of Pathology, MSC 7750, The University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Dr, San Antonio, TX 78229-3900. Copyright © 2000 by W.B. Saunders Company 0046-8177/00/3111-0010510.00/0 doi: 10.1053/hupa.2000.19447
gene expression, the tissues were studied for latent membrane protein 1 (LMP1) expression by immunohistochemistry. Viral LMP1 was observed in the large paracortical lymphocytes of all 10 cases of IM,
indicating co-localization of p53 and LMP1 in these cells. Our findings confirm that p53 overexpression is not specific for nodal malignancy and that p53 accumulation is characteristic of IM. Because p53 was not coexpressed in the same cells as BZLF1, it appears that BZLF1 is not directly responsible for p53 accumulation. Nevertheless, co-localization of p53 and LMP1 in activated-appearing lymphocytes suggests that EBV infection is responsible for p53 accumulation. HUM PATHOL 31:1397-1403. Copyright © 2000 by W.B. Saunders Company
Key words: Epstein-Barr virus, infections mononucleosis, EBER1, p53, BZLF1, LMP1. Abbreviations: EBV, Epstein-Barr virus; IM, infections mononucleosis; EBER1, EBV-encoded RNA; LMP1, Latent membrane protein 1; BZLF1, BamH1 Z leftward reading frame 1
infection is the BZLF1 protein, an EBV-encoded protein that is responsive to cellular differentiation signals. Through modulation of BZLF1, EBV persists indefinitely in resting B lymphocytes, and is triggered to replicate in cells undergoing terminal differentiation.1 Expression of EBV replicative proteins has been examined in IM tissues3-6 Anagnostopoulos et al5 detected BZLF1 in rare lymphoid cells of IM tonsils, including lymphocytes interspersed in the surface epithelium. Other replicative factors (early antigen [EA] and viral capsid antigen [VCA] were not detected by Laytragoon-Lewin et al 6 in 2 IM tonsils. Niedobitek et al ~ found BZLF1 and, much less frequently, early antigen, in scattered small lymphocytes and plasmacyfic cells in or near the surface epithelium. In vitro studies by Zhang et al7 showed that EBVencoded BZLF1 could bind to and inactivate p53. The possible relationship between viral replication and p53 expression has not yet been explored in IM tissues. The p53 protein has been called the "guardian of the genome" because it transmits signals to halt cell proliferation or trigger cell death in the event of DNA damage. Several other DNA viruses, most notably human papillomavirus, interfere with p53 function as a mechanism of viral pathogenesis. Whether EBV likewise circumvents p53-mediated checkpoints is speculative, but there is evidence of abnormal p53 accumulation in several EBV-related tumors, including Burkitt's lymphoma and nasopharyngeal carcinoma. 8,9 Interestingly, the mechanism of p53 overexpression may differ by tumor type, because mutation of the TP53 gene is common in Burkitt's lymphoma1°,1x but not in nasopharyngeal carcinoma? 2-x4 Presumably mutations are not common in benign reactive processes such as IM.
1397
HUMAN PATHOLOGY Volume31, No. 11 (November 2000) It has been shown that normal B cells transfected with EBV proteins (LMP1 or EBNA2) overexpress p53 t h r o u g h induction of NF-kB. 15 W h e t h e r p53 is overexpressed in vivo during acute EBV infection has not yet been studied. Immunohistochemical analysis of 10 cases of reactive lymph node hyperplasia f o u n d no evidence of p53 accumulationt6; however, it is not clear whether any of these cases represented IM. To our knowledge, expression of p53 in naturally infected IM tissues has not yet been reported. To investigate the association between EBV and p53 expression in benign lymphoid tissue, we evaluated lymph node biopsy specimens from 10 IM patients and 11 non-IM controls. Serologic data were used to support the clinical diagnosis of IM, and the presence of EBV was confirmed in all cases of IM by in situ hybridization to EBV-encoded RNA (EBER1). Immunohistochemistry was used to detect p53, as well as viral LMP1 and BZLF1, to determine whether these proteins are co-localized in IM lesions.
MATERIALS AND METHODS Formalin-fixed, paraffin-embedded lymphoid tissues from IM patients were collected from the archival tissue files of the hematopathology consultation service at the University of Texas Health Science Center at San Antonio. These tissues included 6 tonsils, 3 lymph nodes, and 1 nasopharyngeal mass (Table 1). The histology and EBER1 in situ hybridization of 8 cases was previously reported. 17 Diagnosis of IM was made by using standard clinicopathologic criteria, TM including clinical history, morphology, and serologic detection of IgM heterophile antibodies (Color Slide II, Seradyn Clinical Diagnostics, Indianapolis, IN) or serologic evidence of primary EBV infection (defined as VCA IgM above 1:10, VCA total antibody above 1:160, EA-D above 1:40, EA-R less than 1:10, and EBNA less than 1:10). A control cohort consisted of 11 non-IM reactive lymphoid tissues (3 tonsils and 8 lymph node biopsy specimens) obtained from 11 patients ranging in age from 8 to 55 years (mean, 27). Eight cases had a clinical and morphologic diagnosis of reactive lymphadenitis, including 3 tonsils diagnosed as chronic tonsillitis and 5 lymph nodes with reactive follicular hyperplasia. The remaining 3 lymph nodes were removed as part of non-neoplastic surgical conditions (2 cases with
jugular nodes adjacent to carotid atheromatous plaques, and 1 case of a perisplenic node in a motor vehicle accident victim). Serologic studies done in 8 of the 11 control patients (3 tonsils and 5 reactive lymph nodes) showed negative heterophile antibody and no serologic evidence of primary EBV infection.
Detection of EBV by In Situ Hybridization EBER1 in situ hybridization is a sensitive means of detecting EBV because virtually all latently infected cells abundantly express it. 19 In situ hybridization to EBER1 transcripts was performed by using complimentary digoxigenin-labeled riboprobes as previously described. 2° Tissue sections were placed on silane-coated slides, where they were dewaxed, rehydrated, digested with proteinase K, and hybridized. The digoxigenin signal was detected using the Genius system (Boehringer-Mannheim, Indianapolis, IN), and the tissue was counterstained with 1% methyl green. To insure the integrity of template RNA in all tissues, a control riboprobe targeting ubiquitous cellular U6 RNA was applied in parallel reactions. Positive tissue controls consisted of cases of EBV-associated Hodgkin's disease. Probe controls showed that antisense but not sense EBER1 riboprobe targets latently infected cells. Probe templates for EBER1 (RA386) and U6 (RA390) were donated by Richard Ambinder, MD, PhD, of Johns Hopkins University. Light microscopy was used to identify intranuclear EBER1 and U6 transcripts and to semiquantitatively assess the proportion of cells expressing EBER1 in each of the following cell populations: paracortical lymphocytes further subcategorized as small lymphocytes and large lymphocytes, Reed Sternberg-like cells (RS-like), plasma cells, apoptotic cells, germinal center cells, and tonsillar epithelial cells.
Detection of p53 by Immunohistochemistry Immunohistochemical analysis was performed on paraffin tissue sections by using the DO7 antibody (Novacastra, Newcastle-upon-Tyne, UK). In brief, sections were deparaffinized in xylene, rehydrated, and washed in phosphate-buffered saline. Antigen availability was enhanced by pretreating sections for 5 minutes in a pressure cooker in boiling 0.01 mol/L Na citrate buffer. Endogenous peroxidase was quenched with sodium azide and H202, nonspecific binding was blocked with ovalbumin, and sections were incubated in DO7 antibody (diluted at 1:1,000) overnight. Bound antibody was detected with biotinylated rabbit anti-mouse secondary
Table 1. Clinical and Serologic Features of Infectious Mononucleosis Patients Case
Age/Gender
1. 2. 3. 4. 5.
17/M 25/M 17/F 32/M 15 m o / M
Cervical node Tonsils Tonsils Tonsils Cervical node
Lymphadenopathy URT obstruction URT obstruction URT obstruction Lymphadenopathy
6. 7. 8.
80/M 26/M 16/F
Tonsils Cervical node Tonsils
Pharyngitis Lymphadenopathy URT obstruction
9.
38/M
Nasopharyngeal mass
Hearing loss
21 m o / M
Tonsils
URT obstruction
10.
Tissue
Clinical Presentation
Abbreviations: URT, upper respiratory tract; EBV, Epstein-Barr virus.
1398
Serology Heterophile antibody positive
Heterophile antibody positive Heterophile antibody positive EBV titers positive Heterophile antibody positive Heterophile antibody positive EBV titers positive Heterophile antibody positive EBV titers positive Heterophile antibody positive EBV titers positive Heterophile antibody positive EBV titers positive
p53 IN INFECTIOUSMONONUCLEOSIS(Ehsanet al) antibody (1:100) followed by peroxidase-labeled streptavidin (1:100) and diaminobenzidine chromogen. Tissues were counterstained with 1% methyl green. Positive controls included carcinomas of colon, stomach, and breast that were proved to contain mutant p53 by DNA sequencing. Semiquantitative measurement of intranuclear p53 expression was performed according to method ofAllred et al. m The expanded paracortex was evaluated microscopically, and each case was assigned a proportion score based on the proportion of cells expressing p53 (0 = none, 1 = 1% or less, 2 = 2%-10%, 3 = 11%-33%, 4 = 34%-66%, 5 = >66% of cells) and an intensity score based on the average intensity of positive staining (0 = none, 1 = weak, 2 = intermediate, 3 = strong). These 2 scores were added to obtain a total score ranging from 0 to 8, which serves as a semiquantitative measure of p53 expression. In addition to evaluating p53 scores in the population of paracortical cells, proportion scores were also evaluated in the subsets of cell populations described above for EBER1 stains.
Detection of EBV-BZLF1 Protein by Immunohistochemistry The EBV-BZLF1 antibody (clone BZ1, DAKO, Carpenteria, CA) was used for immunohistochemical detection and localization of the BZLF1 protein. Sections were deparaffinized in xylene, and antigen availability was enhanced by pretreatment with 1 m g / m L pronase E for 5 minutes. Nonspecific binding was blocked with ovalbumin, diluted BZLF1 antibody was applied (1:50), and sections were incubated with a secondary biotinylated antiserum against mouse immunoglobulin followed by peroxidase-labeled streptavidin and diamineobenzidine chromogen. Sections were counterstained with 1% methyl green. An oral hairy leukoplakia sample was used as a positive control. Only nuclear staining was considered to be a true positive. (False-positive cytoplasmic staining was commonly seen in plasma cells and eosinophils in both IM cases and non-IM controls). The staining pattern of BZLF1 protein was assessed in the same cell populations described above for EBER1 stains, using the same scoring system described for p53 interpretation.
Detection of LMP1 by Immunohistochemistry Immunohistochemical analysis was performed on paraffin tissue sections by using EBV-LMP antibody (Dako, Carpenteria, CA). The EBV-LMP1 antibody was diluted in 1:50 antisera for 1 hour. Sections were treated in the same manner as described above for BZLF1. Positive controls included lymph nodes from known EBV-related cases of Hodgkin's disease. Only membranous and cytoplasmic staining was interpreted as a true positive. The staining pattern of LMP1 was assessed in the same cell populations described for EBER1 stains.
sinuses, a n d focal geographic necrosis. T h e paracortical zone contained a p o l y m o r p h o u s infiltrate of small to medium-sized lymphocytes; large lymphocytes including immunoblasts, histiocytes, scattered apoptotic cells; and few eosinophils or plasma cells. T h e n u m b e r o f atypical giant RS-like cells varied a m o n g cases, but when present these cells were often located adjacent to zones of necrosis. A few mitotic figures were seen in all cases. Four of the six tonsil biopsy specimens contained stratified squamous m u c o s a containing occasional small lymphocytes within the epithelium. Non-IM cases were characterized by follicular and paracortical hyperplasia without necrosis. T h e interfollicular zone contained a p o l y m o r p h o u s infiltrate of small, m e d i u m , and large lymphocytes with rare eosinophils; however, no apoptotic cells or RS-like cells were identified. Plasma cells were often located n e a r the sinuses. Tingible body m a c r o p h a g e s and m o d e r a t e mitotic figures were n o t e d in the germinal centers, whereas rare mitotic activity was seen in the paracortex.
EBV Localization by In situ Hybridization In all IM cases, n u m e r o u s EBV-infected cells were detected by EBER1 in situ hybridization. EBERI-expressing cells were p r e d o m i n a n t l y paracortical lymphocytes ranging f r o m small to large in size, including nearly all immunoblasts a n d RS-like cells (Fig 1A). T h e p r o p o r t i o n of EBERl-positive paracorfical cells varied by area in a given case, with some areas having a proportion score of 3 (11% to 33%) and o t h e r areas app r o a c h i n g a m a x i m u m p r o p o r t i o n score of 5 ( > 6 6 % ) . Minorities of lymphocytes within the surface epithelium were positive, particularly those f o u n d d e e p in the tonsillar crypts. In contrast, the epithelial cells were consistently negative for EBER1. Rare lymphoid cells in germinal centers were positive, but no EBER1 was detectable in plasma cells. EBER1 was f o u n d in some apoptotic cells; however, m a n y apoptotic cells were U6negative, suggesting p o o r preservation of RNA and the possibility of false-negative EBER1 results. In the non-IM control cases, EBER1 stains showed expression in rare scattered small to medium-sized lymphocytes, estimated at fewer than 0.1% of cells. U6 stain was positive, indicating that RNA was preserved and available for hybridization. T h e negative EBER1 results further c o n f i r m e d the clinical, morphologic, a n d serologic impression that these were not EBV-related lymp h o i d hyperplasias. EBER1 was completely absent in the surface epithelium of those control cases in which mucosa was present.
p53 Expression
RESULTS All cases classified as IM had histomorphologic features and clinical and serologic findings consistent with a diagnosis of IM. C o m m o n histomorphologic features included distortion or c o m p l e t e e f f a c e m e n t of lymph n o d e architecture by an e x p a n d e d paracortex, little or no follicular hyperplasia, focal preservation of
T h e D07 m o n o c l o n a l antibody was used for i m m u nohistochemical detection and localization of p53 protein. Lymphocytes expressing p53 were f o u n d in m a n y parts of the tissue in all 10 cases of IM, although the pattern of positivity was h e t e r o g e n e o u s a m o n g cases. In all cases, p53 was expressed p r e d o m i n a n t l y in large lymphocytes and also was expressed in a minority of
1399
HUMAN PATHOLOGY
Volume 31, No, 11 (November 2000)
FIOURE 1. EBER1 in situ hybridization (A; case 3) shows nuclear staining of many small and large lymphocytes, Immunohistochemical stains for p53 (B; case 3) shows intranuclear accumulation of p53 protein in large lymphocytes of the paracortical zone. Immunostain for BZLF1 (C; case 6) shows subepithelial small lymphocyte with intranuclear BZLF1 protein. The white arrow points to the tonsillar epithelium. Immunostain for LMP1 (D; case 3) shows membranous and cytoplasmic staining of large lymphocytes and RS-like cells. (Original magnifications: A, x 1,000; B, x500; C, × 1,000; D, x500).
RS-like cells and only rarely in small lymphocytes (Fig 1B). The p53-positive lymphocytes were distributed t h r o u g h o u t the paracortex, including the areas near tonsillar crypts. In addition, basal epithelial cells were positive for p53 protein in 4 tonsils. Expression of p53 was not seen in plasma cells, in most atypical RS-like cells, or in apoptotic cells. In the e x p a n d e d paracortical zones of the IM cases, proportion scores a m o n g cases varied from 2 to 3 and, a m o n g cases with at least 2% of cells expressing p53, intensity scores varied from 2 to 3 (Table 2). In any given case, intensity also varied from cell to cell. T h r e e cases had a total p53 score of 6, 3 cases had a total score of 5, and 4 cases had a total score of 4. No p53 was detected in the paracortical cells of the non-IM control cases. However, p53 was detected in occasional basal epithelial cells of the stratified squamous epithelium of tonsils, and in scattered lymphoid cells within germinal centers of IM and non-IM control cases. These p53-expressing basal epithelial cells and germinal center cells have been described by other investigators 22 and are likely to be in the G1 phase of the cell cycle when p53 is physiologically upregulated to detectable levels.
Viral BZLF1 Protein Expression
Immunohistochemistry was used to detect and localize EBV-BZLF1 protein in the IM and non-IM control tissues. Sensitivity and specificity of the stain was validated in oral hairy leukoplakia specimens, where BZLF1 was strongly expressed in the stratum spinosum layer of hyperplastic stratified squamous epithelium. BZLF1 was expressed in 8 of 10 cases of IM. In the 8 positive cases, BZLF1 was restricted to scattered small lymphocytes in the paracortical areas. In IM tonsils, small lymphocytes staining positive for BZLF1 were often present in the subepithelial paracortical regions near the tonsillar crypts, and BZLF1 was also expressed in occasional small lymphocytes that were interspersed with epithelial cells of the surface epithelium (Fig 1C). Proportion and intensity scores for paracortical cells show the following patterns of BZLF1 expression: 2 cases had high scores (proportion score 3, intensity score 3), 4 cases were intermediate (proportion score 2, intensity score 2), and 2 cases had low scores (proportion score 2, intensity score 1). These scores are shown in Table 2. BZLF1 expression was not seen in epithelial cells,
1400
p53 IN INFECTIOUSMONONUCLEOSIS (Ehsan et al)
Table 2. Propoffion and Intensity Scores of p53 and BZLF-1 Protein Expression in All 10 Cases of IM BZLF-1 Expression
p53 Expression
Cases
Proportion Score
Intensity Score
Proportion Score
Intensity Score
1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
3 3 3 2 2 3 2 3 3 2
2 3 3 2 2 3 2 2 2 2
0 2 2 0 2 3 2 2 2 3
0 2 2 0 1 3 2 1 2 3
germinal center cells, RS-like cells, or apoptotic cells of IM cases. The non-IM cases had no BZLF1 except for cytoplasmic staining of plasma cells that was also observed in IM cases and was interpreted as false positive. In the positive control tissues, true BZLF1 staining was intranuclear. LMP1 Expression In the IM cases, immunohistochemical stains for EBV LMP1 showed membranous and cytoplasmic staining in most paracortical large lymphocytes, immunoblasts, and RS-like cells (proportion scores 4 or 5; Fig 1D). In tonsils, the distribution of these LMPl-expressing cells was often enriched near the tonsillar crypts. The intensity score for paracortical lymphocytes varied among cases, ranging from 1 to 3 (weak to strong). Only occasional small lymphocytes and apoptotic cells were LMPl-positive (proportion scores 2 or less). No staining was seen in germinal center cells, plasma cells, or epithelial cells. The non-IM controls were completely negative for LMP1, except for cytoplasmic staining of eosinophils and rare plasma cells, also seen in IM cases and interpreted as false positive.
DISCUSSION This is the first report showing p53 expression in lymphoid tissues of primary EBV infection. In contrast, p53 was not detected in 11 benign, reactive lymph nodes. All IM tissues were confirmed to be EBV-related using serology and EBER1 in situ hybridization, whereas all non-IM patients were confirmed as unrelated to acute EBV infection by the same methods. The striking difference in p53 expression between these 2 cohorts suggests that p53 is specifically upregulated in EBV-related lymphadenitis, and it raises the issue of whether EBV infection causes p53 accumulation in IM tissues. Apparently EBV infection alone is not sufficient to cause a cell to express detectable p53, because the number of cells expressing EBER1 far exceeded the number of cells expressing p53 in every case of IM. To specifically evaluate whether p53 is altered in a subset
of infected cells undergoing viral replication, BZLF1 was assayed immunohistochemically. BZLF1 is the critical immediate-early viral protein controlling entry into the lytic phase of replication and acts as a transcription factor to initiate upregulation of a cascade of proteins needed for virion production, a Our results are in agreement with those of Niedobitek et al~ who found that cells expressing lyric cycle antigens were mainly small lymphocytes in paracortical areas and near crypt epithelium. But, in contrast to their findings, we did not demonstrate BZLF1 expression in plasma cell nuclei. It seems likely that BZLF1 expression varies with stage of cellular differentiation, and there is a narrow window during which BZLF1 is upregulated. In vitro studies have shown that BZLF1 protein not only binds p53 but also functionally inactivates it. 7,31 A major goal of this study was to determine whether this mechanism operates in vivo to sequester p53 in EBVreplicating lymphocytes. Although most IM tissues coexpressed both markers in a small fraction of lymphocytes, expression was not localized to the same cell types, suggesting that there is no direct interaction between BZLF1 and p53 in vivo. This finding does not exclude indirect interactions. Indeed, indirect effects are supported by co-localization of cells expressing either of these proteins to the same anatomic zones within lymphoid tissue. Expression of p53 in IM tissues indicates that p53 overexpression is not specific for malignancy but rather can be found in benign, reactive lymphoproliferations. This finding refutes the theory that p53 immunostains might help discriminate between benign and neoplastic lymph nodes, a6 Furthermore, overexpression of p53 in IM but not in the other reactive lymphadenopathies that we examined suggests a role for EBV in altering p53 expression. In vitro studies have shown that EBV-encoded LMP1 mediates upregulation of p53 levels by activation of NF-kB.15 This pathway represents another possible mechanism of p53 induction during acute EBV infection in vivo. To explore the possible relationship of LMP1 and p53 proteins in acute EBV infection, we studied expression of these two factors in our IM cases. We showed that LMP1 was expressed predominantly in large paracortical lymphocytes including RS-like cells, and that these cells were enriched near the tonsillar crypts. Expression of p53 was seen in the same cell types and in the same anatomic distribution, suggesting the possibility that p53 is upregulated by the LMPl-triggered NF-kB pathway described in vitro. In vitro studies have also shown that the concentration of p53 relative to that of LMP1 determines whether EBV-infected cells undergo apoptosis. ~,34 Although we could not measure protein concentrations by the methods employed in this study, we observed rampant lymphoid proliferation as well as focal apoptosis among infected cells, suggesting the possibility that p53 function is altered by viral infection. With current technology, it is impossible to determine whether p53 is functional or dysfunctional in naturally infected IM tissues. The p53 protein is an important negative regulator
1401
HUMAN PATHOLOGY
Volume 31, No. 11 (November 2000)
of cell proliferation that is inactivated and concomitantly overexpressed in many malignancies. 24-27In some of these tumors, overexpression of p53 protein in the absence of TP53 gene mutation suggests that the protein is rendered dysfunctional at a post-translational level. 2s-~° Presumably IM tissues do not harbor TP53 gene mutations, so the accumulation of p53 in these reactive lymphocytes implicates a post-translational effect of viral pathogenesis, or else a physiologic upregulation of p53 in response to EBV infection. Further analysis of the staining patterns of EBER1 and BZLF1 in IM tissues shows additional information that could shed light on viral pathogenesis. First, neither EBER] nor BZLF1 were expressed in tonsillar epithelial cells, confirming reports by several others 5,2~ that EBV is absent in normal epithelium. These data support the hypothesis that EBV shed in saliva from normal individuals probably originates from mucosal lymphocytes rather than epithelial cells. In fact, our identification of BZLFl-expressing lymphocytes localized in or near the surface mucosa of IM tonsils suggests that these lymphocytes could be the source of salivary virions. LMP1 is known to upregulate the intracellular levels of several antiapoptotic proteins, including bcl-2, A20, and Mcl-1. 32,3s It has been hypothesized that LMPl-mediated anti-apoptotic effects during the lyric cycle of viral replication lengthen cellular lifespan to permit even more virions to be produced. 32 In our study, LMP1 was not expressed in the same cell population as BZLF1. Our findings confirm those of Niedobitek et al, 3 indicating that BZLFl-positive cells are consistently LMPl-negative. This result implies that LMP1 is absent in naturally infected cells that are entering into the lytic cycle of viral replication. Based on our observations in IM cases, we propose that LMP1 expression is upregulated later than BZLF1 during lytic replication, once cells have transitioned from small resting lymphocytes into larger activated-appearing lymphocytes. The sequential expression of BZLF1 and then LMP1 during lyric replication would explain the co-localization of BZLF]- and LMPl-expressing lymphocytes in the mucosal region of the tonsil where virion shedding is thought to originate. Although it is tempting to speculate that LMPl-triggered NFkB activation is responsible for p53 upregulation in these lymphocytes, we cannot exclude the possibility some of the 50 or so other EBV gene products that are upregulated during lyric viral replication trigger p53 expression or inactivation. RS-like cells in IM were predominantly EBER1positive, LMPl-positive, and BZLFl-negative, which is the same phenotype as true RS cells of Hodgkin's disease. 17,~5-37Another cell type c o m m o n to both IM and Hodgkin's tissues is apoptotic cells, also referred to as mummified cells. The precise histogenesis of apoptotic cells in IM is difficult to determine because morphology is obscured, but in some cases their large size suggests derivation from RS-like cells. Their appearance and expression pattern combine to suggest that the apoptotic cells of IM are lymphocytes whose repli-
cative infection terminated in cell death. It is unclear whether cell death results from p53-mediated apoptosis versus the end product of virion production. -~s In conclusion, this study reveals several new observations. First, p53 accumulation is not specific for malignancy in lymph nodes. In fact, p53 expression is quite characteristic of the benign lymphadenitis caused by acute EBV infection. In contrast, the EBV-negative benign lymphadenopathies in this study did not express p53, implicating EBV infection as responsible for p53 accumulation. Viral BZLF1 co-localizes to the same areas of the node as p53, but not to the same cell types, suggesting no direct interaction between these factors. Coexpression of p53 and LMP1 in the same subset of paracortical large lymphocytes supports the concept that these cells are activated to produce new virions or die, perhaps depending on whether p53 is allowed to trigger apoptosis. Finally, our data support the growing body of literature suggesting that lymphocytes rather than oropharyngeal epithelial cells are the origin of infectious virions shed in saliva.
REFERENCES 1. Miller G: The switch between latency and replication of EBV. J Infect Dis 161:833-844, 1990 2. Tierney RJ, Steven N, Young LS, et al: Epstein-Barr virus latency in blood mononuclear cells: Analysis of viral gene transcription during primary infection and in the carrier state.J Viro168:73747385, 1994 3. Niedobitek G, Agathanggelou A, Herbst H, et al: Epstein-Barr virus infection in infectious mononucleosis: Virus latency, replication and phenotype of EBV-infected cells. J Pathol 182:151-159, 1997 4. Prang NS, Hornef MW, Jager M, et al: Lytic replication of Epstein-Barr virus in the peripheral blood: Analysis of viral gene expression in B lymphocytes during infectious mononucleosis and in the normal carrier state. Blood 89:1665-1677, 1997 5. Anagnostopoulos I, Hummel M, Kreschel C, et al: Morphology, immunophenotype, and distribution of latently a n d / o r productively Epstein-Barr virus-infected cells in acute infectious mononucleosis: Implications for the interindividual infection route of EpsteinBarr virus. Blood 85:744-750, 1995 6. Laytragoon-Lewin N, Chen F, Avila-CainoJ, et al: Epstein-Barr virus gene expression in lymphoid B-cells during acute infectious mononucleosis and clonality of the directly growing cell lines. I n t J Cancer 71:345-349, 1997 7. Zhang Q, Gutsch D, Kenney S: Functional and physical interaction between p53 and BZLFI: Implication for Epstein-Barr virus latency. Mol Cell Biol 14:1929-1938, 1994 8. Gulley ML, Burton MP, Allred DC, et al: Epstein-Barr virus infection is associated with p53 accumulation in nasopharyngeal carcinoma. HVM PATHOL 29:252-259, 1998 9. Lehtinen T, IsolaJ, Aine R, et ah Accumulation of p53 protein correlates with tumor proliferative activity in EBV positive Burkitt's lymphoma. Hematol Oncol 10:273-279, 1992 10. Balint E, Reisman D: Increased rate of transcription contributes to elevated expression of the mutant p53 gene in Burkitt's lymphoma cells. Cancer Res 56:1648-1653, 1996 11. Cherney BW, Bhatia KG, Sgadari C, et al: Role of the p53 tumor suppressor gene in the tumorigenicity of Burkitt's lymphoma cells. Cancer Res 57:2508-2515, 1997 12. Edwards RH, Raab-Traub N: Alterations of the p53 gene in Epstein-Barr virus associated immunodeficiency-related lymphomas. J Virol 68:1309-1315, 1994 13. Farrell PJ, Allan GJ, Shanahan F, et ah p53 is fi'equently mutated in Burkitt's lymphoma cell lines. EMBOJ 10:2879-2887, 1991 14. Lo KW, Mok CH, Huang DP, et al: p53 mutation in human nasopharyngeal carcinomas. Anticancer Res 12:195%1963, 1992
1402
p53 IN INFECTIOUSMONONUCLEOSIS (Ehsan et al) 15. Chen W, Cooper NR: Epstein-Barr virus nuclear antigen 2 and latent membrane protein independently transactivate p53 through induction of NF-kappaB activity. J Virol 70:4849-4853, 1996 16. Nakamura H, Said JW, Miller CW, et al: Mutation and protein expression of p53 in acquired immunodeficiency syndromerelated lymphomas. Blood 82:920-926, 1993 17. Reynolds DJ, Banks PM, Gulley ML: New characterization of infectious mononucleosis and a phenotypic comparison with Hodgkin's disease. A m J Pathol 146:379-388, 1995 18. StricklerJG, Fedeli F, Horwitz CA, et al: Infectious mononucleosis in lymphoid tissue: Histopathology, in situ hybridization, and differential diagnosis. Arch Pathol Lab Med 117:269-278, 1993 19. Ambinder RF, Mann RB: Detection and characterization of Epstein-Barr virus in clinical specimens. Am J Pathol 145:239-252, 1994 20. GuUey ML, Eagan PA, Quintanilla-Martinez L, et al: EpsteinBarr virus DNA is abundant and monoclonal in the Reed-Sternberg ceils of Hodgkin's disease: Association with mixed cellularity subtype and Hispanic American ethnicity. Blood 83:1595-1602, 1994 21. Allred DC, Clark GM, Elledge R, et al: Accumulation of mutant p53 is associated with increased proliferation and poor clinical outcome in node-negative breast cancer. J Natl Cancer Inst 85:200-206, 1993 22. Sauter ER, Cleveland D, Trock B, et al: p53 is expressed in fifty percent of pre-invasive lesions of the head and neck epithelium. Carcinogenesis 15:2269-2274, 1994 23. Karajannis MA, Hummel M, Anagnostopoulos I, et al: Strict lymphotropism of Epstein-Barr virus during acute infectious mononucleosis in nonimmunocompromised individuals. Blood 89:28562862, 1997 24. Levine A, Momand J, Finlay C: The p53 tumor suppressor gene. Nature 351:453-456, 1991 25. Vogelstein B, Kinzler K: p53 function and dysfunction. Cell 70:523-526, 1992 26. Hollstein M, Sidransky D, Vogelstein B, et al: p53 mutations in human cancers. Science 253:49-53, 1991 27. Martin A, Flaman JM, Frebourg T, et al: Functional analysis
of the p53 protein in AIDS-related non-Hodgkin's lymphomas and polymorphic lymphoproliferations. BrJ Haematol 101:311-317, 1998 28. Chilosi M, Doglioni C, Menestrina F, et al: Abnormal expression of the p53-binding protein MDM2 in Hodgkin's disease. Blood 84:4295-4300, 1994 29. Matsushima AY, Cesarman E, Chadburn A, et al: Post-thymic T cell lymphomas frequently overexpress p53 protein but infrequently exhibit p53 gene mutations. A m J Pathol 144:573-584, 1994 30. Villuendas R, Piris MA, Algara P, et al: The expression of p53 protein in non-Hodgkin's lymphomas is not always dependent on p53 gene mutations. Blood 82:3151-3156, 1993 31. Cayrol C, Flemington EK: Identification of cellular target genes of the EBV transactivator Zta: Activation of TGF-Bigh3 and TGF-B1. J Virol 69:4206-4212, 1995 32. Henderson S, Rowe M, Gregory C, et al: Induction of bcl-2 expression by Epstein-Barr virus latent membrane protein 1 protects infected B-cells from programmed cell death. Cell 65:110%1115, 1991 33. Chen Weiping, Huang Shuang, Cooper Neil R: Levels of p53 in Epstein-Barr virus-infected cells determine cell fate: Apoptosis, cell cycle arrest at the G1/S boundary without apoptosis, cell cycle arrest at the G2/M boundary without apoptosis, or unrestricted proliferation. Virology 251:217-226, 1998 34. Okan I, Wang Y, Chen F, et al: The EBV-encoded LMP1 protein inhibits p53-triggered apoptosis but not growth arrest. Oncogene 11:1027-1031, 1995 35. Siebert JD, Ambinder RF, Napoli VM, et al: Human immunodeficiency virus-associated Hodgkin's disease contains latent, not replicative, Epstein-Barr virus. HUM PATHOL26:1191-1195, 1995 36. Brousset P, Knecht H, Rubin B, et al: Demonstration of Epstein-Barr virus replication in Reed-Sternberg cells of Hodgkin's disease. Blood 82:619-624, 1993 37. Pallesen G, Sandvej Ig, Hamilton-Dutoit SJ, et al: Activation of Epstein-Barr virus replication in Hodgkin and Reed-Sternberg cells. Blood 78:1162-1165, 1991 38. Allday MJ: Apoptosis and Epstein-Barr virus in B cells. EBV Report 3:55-65, 1996
1403
Epstein-Barr virus (EBV) is a lymphotropic herpes virus that infects more than 90% of humans worldwide. Primary EBV infection in early childhood is generally asymptomatic, but when infection is delayed until adolescence or adulthood it causes the clinical syndrome of infectious mononucleosis (IM). IM represents an EBVdriven proliferation of benign lymphocytes that is eventually controlled by humoral and cellular immune responses. When EBV infects human B lymphocytes, there are 3 possible cellular responses: latent persistence of the EBV genome, replicative production of new virions, and cell death secondary to the host response to infection. In the latent phase of infection, viral DNA persists indefinitely inside the cell and is passed on to cellular progeny during mitosis. In the replicative phase of infection, a cascade of viral and host proteins are expressed to mediate viral genome replication and packaging into infectious virions. The master switch controlling the transition from latent to replicative EBV
From the Department of Pathology, University of Texas Health Science Center at San Antonio, and the Audie L. Murphy Memorial Veteran's Hospital, San Antonio, TX. Accepted for publication August 8, 2000. Supported by a grant from the Veteran's Administration. Address correspondence and reprint requests to Aamir Ehsan, MD, Department of Pathology, MSC 7750, The University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Dr, San Antonio, TX 78229-3900. Copyright © 2000 by W.B. Saunders Company 0046-8177/00/3111-0010510.00/0 doi: 10.1053/hupa.2000.19447
gene expression, the tissues were studied for latent membrane protein 1 (LMP1) expression by immunohistochemistry. Viral LMP1 was observed in the large paracortical lymphocytes of all 10 cases of IM,
indicating co-localization of p53 and LMP1 in these cells. Our findings confirm that p53 overexpression is not specific for nodal malignancy and that p53 accumulation is characteristic of IM. Because p53 was not coexpressed in the same cells as BZLF1, it appears that BZLF1 is not directly responsible for p53 accumulation. Nevertheless, co-localization of p53 and LMP1 in activated-appearing lymphocytes suggests that EBV infection is responsible for p53 accumulation. HUM PATHOL 31:1397-1403. Copyright © 2000 by W.B. Saunders Company
Key words: Epstein-Barr virus, infections mononucleosis, EBER1, p53, BZLF1, LMP1. Abbreviations: EBV, Epstein-Barr virus; IM, infections mononucleosis; EBER1, EBV-encoded RNA; LMP1, Latent membrane protein 1; BZLF1, BamH1 Z leftward reading frame 1
infection is the BZLF1 protein, an EBV-encoded protein that is responsive to cellular differentiation signals. Through modulation of BZLF1, EBV persists indefinitely in resting B lymphocytes, and is triggered to replicate in cells undergoing terminal differentiation.1 Expression of EBV replicative proteins has been examined in IM tissues3-6 Anagnostopoulos et al5 detected BZLF1 in rare lymphoid cells of IM tonsils, including lymphocytes interspersed in the surface epithelium. Other replicative factors (early antigen [EA] and viral capsid antigen [VCA] were not detected by Laytragoon-Lewin et al 6 in 2 IM tonsils. Niedobitek et al ~ found BZLF1 and, much less frequently, early antigen, in scattered small lymphocytes and plasmacyfic cells in or near the surface epithelium. In vitro studies by Zhang et al7 showed that EBVencoded BZLF1 could bind to and inactivate p53. The possible relationship between viral replication and p53 expression has not yet been explored in IM tissues. The p53 protein has been called the "guardian of the genome" because it transmits signals to halt cell proliferation or trigger cell death in the event of DNA damage. Several other DNA viruses, most notably human papillomavirus, interfere with p53 function as a mechanism of viral pathogenesis. Whether EBV likewise circumvents p53-mediated checkpoints is speculative, but there is evidence of abnormal p53 accumulation in several EBV-related tumors, including Burkitt's lymphoma and nasopharyngeal carcinoma. 8,9 Interestingly, the mechanism of p53 overexpression may differ by tumor type, because mutation of the TP53 gene is common in Burkitt's lymphoma1°,1x but not in nasopharyngeal carcinoma? 2-x4 Presumably mutations are not common in benign reactive processes such as IM.
1397
HUMAN PATHOLOGY Volume31, No. 11 (November 2000) It has been shown that normal B cells transfected with EBV proteins (LMP1 or EBNA2) overexpress p53 t h r o u g h induction of NF-kB. 15 W h e t h e r p53 is overexpressed in vivo during acute EBV infection has not yet been studied. Immunohistochemical analysis of 10 cases of reactive lymph node hyperplasia f o u n d no evidence of p53 accumulationt6; however, it is not clear whether any of these cases represented IM. To our knowledge, expression of p53 in naturally infected IM tissues has not yet been reported. To investigate the association between EBV and p53 expression in benign lymphoid tissue, we evaluated lymph node biopsy specimens from 10 IM patients and 11 non-IM controls. Serologic data were used to support the clinical diagnosis of IM, and the presence of EBV was confirmed in all cases of IM by in situ hybridization to EBV-encoded RNA (EBER1). Immunohistochemistry was used to detect p53, as well as viral LMP1 and BZLF1, to determine whether these proteins are co-localized in IM lesions.
MATERIALS AND METHODS Formalin-fixed, paraffin-embedded lymphoid tissues from IM patients were collected from the archival tissue files of the hematopathology consultation service at the University of Texas Health Science Center at San Antonio. These tissues included 6 tonsils, 3 lymph nodes, and 1 nasopharyngeal mass (Table 1). The histology and EBER1 in situ hybridization of 8 cases was previously reported. 17 Diagnosis of IM was made by using standard clinicopathologic criteria, TM including clinical history, morphology, and serologic detection of IgM heterophile antibodies (Color Slide II, Seradyn Clinical Diagnostics, Indianapolis, IN) or serologic evidence of primary EBV infection (defined as VCA IgM above 1:10, VCA total antibody above 1:160, EA-D above 1:40, EA-R less than 1:10, and EBNA less than 1:10). A control cohort consisted of 11 non-IM reactive lymphoid tissues (3 tonsils and 8 lymph node biopsy specimens) obtained from 11 patients ranging in age from 8 to 55 years (mean, 27). Eight cases had a clinical and morphologic diagnosis of reactive lymphadenitis, including 3 tonsils diagnosed as chronic tonsillitis and 5 lymph nodes with reactive follicular hyperplasia. The remaining 3 lymph nodes were removed as part of non-neoplastic surgical conditions (2 cases with
jugular nodes adjacent to carotid atheromatous plaques, and 1 case of a perisplenic node in a motor vehicle accident victim). Serologic studies done in 8 of the 11 control patients (3 tonsils and 5 reactive lymph nodes) showed negative heterophile antibody and no serologic evidence of primary EBV infection.
Detection of EBV by In Situ Hybridization EBER1 in situ hybridization is a sensitive means of detecting EBV because virtually all latently infected cells abundantly express it. 19 In situ hybridization to EBER1 transcripts was performed by using complimentary digoxigenin-labeled riboprobes as previously described. 2° Tissue sections were placed on silane-coated slides, where they were dewaxed, rehydrated, digested with proteinase K, and hybridized. The digoxigenin signal was detected using the Genius system (Boehringer-Mannheim, Indianapolis, IN), and the tissue was counterstained with 1% methyl green. To insure the integrity of template RNA in all tissues, a control riboprobe targeting ubiquitous cellular U6 RNA was applied in parallel reactions. Positive tissue controls consisted of cases of EBV-associated Hodgkin's disease. Probe controls showed that antisense but not sense EBER1 riboprobe targets latently infected cells. Probe templates for EBER1 (RA386) and U6 (RA390) were donated by Richard Ambinder, MD, PhD, of Johns Hopkins University. Light microscopy was used to identify intranuclear EBER1 and U6 transcripts and to semiquantitatively assess the proportion of cells expressing EBER1 in each of the following cell populations: paracortical lymphocytes further subcategorized as small lymphocytes and large lymphocytes, Reed Sternberg-like cells (RS-like), plasma cells, apoptotic cells, germinal center cells, and tonsillar epithelial cells.
Detection of p53 by Immunohistochemistry Immunohistochemical analysis was performed on paraffin tissue sections by using the DO7 antibody (Novacastra, Newcastle-upon-Tyne, UK). In brief, sections were deparaffinized in xylene, rehydrated, and washed in phosphate-buffered saline. Antigen availability was enhanced by pretreating sections for 5 minutes in a pressure cooker in boiling 0.01 mol/L Na citrate buffer. Endogenous peroxidase was quenched with sodium azide and H202, nonspecific binding was blocked with ovalbumin, and sections were incubated in DO7 antibody (diluted at 1:1,000) overnight. Bound antibody was detected with biotinylated rabbit anti-mouse secondary
Table 1. Clinical and Serologic Features of Infectious Mononucleosis Patients Case
Age/Gender
1. 2. 3. 4. 5.
17/M 25/M 17/F 32/M 15 m o / M
Cervical node Tonsils Tonsils Tonsils Cervical node
Lymphadenopathy URT obstruction URT obstruction URT obstruction Lymphadenopathy
6. 7. 8.
80/M 26/M 16/F
Tonsils Cervical node Tonsils
Pharyngitis Lymphadenopathy URT obstruction
9.
38/M
Nasopharyngeal mass
Hearing loss
21 m o / M
Tonsils
URT obstruction
10.
Tissue
Clinical Presentation
Abbreviations: URT, upper respiratory tract; EBV, Epstein-Barr virus.
1398
Serology Heterophile antibody positive
Heterophile antibody positive Heterophile antibody positive EBV titers positive Heterophile antibody positive Heterophile antibody positive EBV titers positive Heterophile antibody positive EBV titers positive Heterophile antibody positive EBV titers positive Heterophile antibody positive EBV titers positive
p53 IN INFECTIOUSMONONUCLEOSIS(Ehsanet al) antibody (1:100) followed by peroxidase-labeled streptavidin (1:100) and diaminobenzidine chromogen. Tissues were counterstained with 1% methyl green. Positive controls included carcinomas of colon, stomach, and breast that were proved to contain mutant p53 by DNA sequencing. Semiquantitative measurement of intranuclear p53 expression was performed according to method ofAllred et al. m The expanded paracortex was evaluated microscopically, and each case was assigned a proportion score based on the proportion of cells expressing p53 (0 = none, 1 = 1% or less, 2 = 2%-10%, 3 = 11%-33%, 4 = 34%-66%, 5 = >66% of cells) and an intensity score based on the average intensity of positive staining (0 = none, 1 = weak, 2 = intermediate, 3 = strong). These 2 scores were added to obtain a total score ranging from 0 to 8, which serves as a semiquantitative measure of p53 expression. In addition to evaluating p53 scores in the population of paracortical cells, proportion scores were also evaluated in the subsets of cell populations described above for EBER1 stains.
Detection of EBV-BZLF1 Protein by Immunohistochemistry The EBV-BZLF1 antibody (clone BZ1, DAKO, Carpenteria, CA) was used for immunohistochemical detection and localization of the BZLF1 protein. Sections were deparaffinized in xylene, and antigen availability was enhanced by pretreatment with 1 m g / m L pronase E for 5 minutes. Nonspecific binding was blocked with ovalbumin, diluted BZLF1 antibody was applied (1:50), and sections were incubated with a secondary biotinylated antiserum against mouse immunoglobulin followed by peroxidase-labeled streptavidin and diamineobenzidine chromogen. Sections were counterstained with 1% methyl green. An oral hairy leukoplakia sample was used as a positive control. Only nuclear staining was considered to be a true positive. (False-positive cytoplasmic staining was commonly seen in plasma cells and eosinophils in both IM cases and non-IM controls). The staining pattern of BZLF1 protein was assessed in the same cell populations described above for EBER1 stains, using the same scoring system described for p53 interpretation.
Detection of LMP1 by Immunohistochemistry Immunohistochemical analysis was performed on paraffin tissue sections by using EBV-LMP antibody (Dako, Carpenteria, CA). The EBV-LMP1 antibody was diluted in 1:50 antisera for 1 hour. Sections were treated in the same manner as described above for BZLF1. Positive controls included lymph nodes from known EBV-related cases of Hodgkin's disease. Only membranous and cytoplasmic staining was interpreted as a true positive. The staining pattern of LMP1 was assessed in the same cell populations described for EBER1 stains.
sinuses, a n d focal geographic necrosis. T h e paracortical zone contained a p o l y m o r p h o u s infiltrate of small to medium-sized lymphocytes; large lymphocytes including immunoblasts, histiocytes, scattered apoptotic cells; and few eosinophils or plasma cells. T h e n u m b e r o f atypical giant RS-like cells varied a m o n g cases, but when present these cells were often located adjacent to zones of necrosis. A few mitotic figures were seen in all cases. Four of the six tonsil biopsy specimens contained stratified squamous m u c o s a containing occasional small lymphocytes within the epithelium. Non-IM cases were characterized by follicular and paracortical hyperplasia without necrosis. T h e interfollicular zone contained a p o l y m o r p h o u s infiltrate of small, m e d i u m , and large lymphocytes with rare eosinophils; however, no apoptotic cells or RS-like cells were identified. Plasma cells were often located n e a r the sinuses. Tingible body m a c r o p h a g e s and m o d e r a t e mitotic figures were n o t e d in the germinal centers, whereas rare mitotic activity was seen in the paracortex.
EBV Localization by In situ Hybridization In all IM cases, n u m e r o u s EBV-infected cells were detected by EBER1 in situ hybridization. EBERI-expressing cells were p r e d o m i n a n t l y paracortical lymphocytes ranging f r o m small to large in size, including nearly all immunoblasts a n d RS-like cells (Fig 1A). T h e p r o p o r t i o n of EBERl-positive paracorfical cells varied by area in a given case, with some areas having a proportion score of 3 (11% to 33%) and o t h e r areas app r o a c h i n g a m a x i m u m p r o p o r t i o n score of 5 ( > 6 6 % ) . Minorities of lymphocytes within the surface epithelium were positive, particularly those f o u n d d e e p in the tonsillar crypts. In contrast, the epithelial cells were consistently negative for EBER1. Rare lymphoid cells in germinal centers were positive, but no EBER1 was detectable in plasma cells. EBER1 was f o u n d in some apoptotic cells; however, m a n y apoptotic cells were U6negative, suggesting p o o r preservation of RNA and the possibility of false-negative EBER1 results. In the non-IM control cases, EBER1 stains showed expression in rare scattered small to medium-sized lymphocytes, estimated at fewer than 0.1% of cells. U6 stain was positive, indicating that RNA was preserved and available for hybridization. T h e negative EBER1 results further c o n f i r m e d the clinical, morphologic, a n d serologic impression that these were not EBV-related lymp h o i d hyperplasias. EBER1 was completely absent in the surface epithelium of those control cases in which mucosa was present.
p53 Expression
RESULTS All cases classified as IM had histomorphologic features and clinical and serologic findings consistent with a diagnosis of IM. C o m m o n histomorphologic features included distortion or c o m p l e t e e f f a c e m e n t of lymph n o d e architecture by an e x p a n d e d paracortex, little or no follicular hyperplasia, focal preservation of
T h e D07 m o n o c l o n a l antibody was used for i m m u nohistochemical detection and localization of p53 protein. Lymphocytes expressing p53 were f o u n d in m a n y parts of the tissue in all 10 cases of IM, although the pattern of positivity was h e t e r o g e n e o u s a m o n g cases. In all cases, p53 was expressed p r e d o m i n a n t l y in large lymphocytes and also was expressed in a minority of
1399
HUMAN PATHOLOGY
Volume 31, No, 11 (November 2000)
FIOURE 1. EBER1 in situ hybridization (A; case 3) shows nuclear staining of many small and large lymphocytes, Immunohistochemical stains for p53 (B; case 3) shows intranuclear accumulation of p53 protein in large lymphocytes of the paracortical zone. Immunostain for BZLF1 (C; case 6) shows subepithelial small lymphocyte with intranuclear BZLF1 protein. The white arrow points to the tonsillar epithelium. Immunostain for LMP1 (D; case 3) shows membranous and cytoplasmic staining of large lymphocytes and RS-like cells. (Original magnifications: A, x 1,000; B, x500; C, × 1,000; D, x500).
RS-like cells and only rarely in small lymphocytes (Fig 1B). The p53-positive lymphocytes were distributed t h r o u g h o u t the paracortex, including the areas near tonsillar crypts. In addition, basal epithelial cells were positive for p53 protein in 4 tonsils. Expression of p53 was not seen in plasma cells, in most atypical RS-like cells, or in apoptotic cells. In the e x p a n d e d paracortical zones of the IM cases, proportion scores a m o n g cases varied from 2 to 3 and, a m o n g cases with at least 2% of cells expressing p53, intensity scores varied from 2 to 3 (Table 2). In any given case, intensity also varied from cell to cell. T h r e e cases had a total p53 score of 6, 3 cases had a total score of 5, and 4 cases had a total score of 4. No p53 was detected in the paracortical cells of the non-IM control cases. However, p53 was detected in occasional basal epithelial cells of the stratified squamous epithelium of tonsils, and in scattered lymphoid cells within germinal centers of IM and non-IM control cases. These p53-expressing basal epithelial cells and germinal center cells have been described by other investigators 22 and are likely to be in the G1 phase of the cell cycle when p53 is physiologically upregulated to detectable levels.
Viral BZLF1 Protein Expression
Immunohistochemistry was used to detect and localize EBV-BZLF1 protein in the IM and non-IM control tissues. Sensitivity and specificity of the stain was validated in oral hairy leukoplakia specimens, where BZLF1 was strongly expressed in the stratum spinosum layer of hyperplastic stratified squamous epithelium. BZLF1 was expressed in 8 of 10 cases of IM. In the 8 positive cases, BZLF1 was restricted to scattered small lymphocytes in the paracortical areas. In IM tonsils, small lymphocytes staining positive for BZLF1 were often present in the subepithelial paracortical regions near the tonsillar crypts, and BZLF1 was also expressed in occasional small lymphocytes that were interspersed with epithelial cells of the surface epithelium (Fig 1C). Proportion and intensity scores for paracortical cells show the following patterns of BZLF1 expression: 2 cases had high scores (proportion score 3, intensity score 3), 4 cases were intermediate (proportion score 2, intensity score 2), and 2 cases had low scores (proportion score 2, intensity score 1). These scores are shown in Table 2. BZLF1 expression was not seen in epithelial cells,
1400
p53 IN INFECTIOUSMONONUCLEOSIS (Ehsan et al)
Table 2. Propoffion and Intensity Scores of p53 and BZLF-1 Protein Expression in All 10 Cases of IM BZLF-1 Expression
p53 Expression
Cases
Proportion Score
Intensity Score
Proportion Score
Intensity Score
1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
3 3 3 2 2 3 2 3 3 2
2 3 3 2 2 3 2 2 2 2
0 2 2 0 2 3 2 2 2 3
0 2 2 0 1 3 2 1 2 3
germinal center cells, RS-like cells, or apoptotic cells of IM cases. The non-IM cases had no BZLF1 except for cytoplasmic staining of plasma cells that was also observed in IM cases and was interpreted as false positive. In the positive control tissues, true BZLF1 staining was intranuclear. LMP1 Expression In the IM cases, immunohistochemical stains for EBV LMP1 showed membranous and cytoplasmic staining in most paracortical large lymphocytes, immunoblasts, and RS-like cells (proportion scores 4 or 5; Fig 1D). In tonsils, the distribution of these LMPl-expressing cells was often enriched near the tonsillar crypts. The intensity score for paracortical lymphocytes varied among cases, ranging from 1 to 3 (weak to strong). Only occasional small lymphocytes and apoptotic cells were LMPl-positive (proportion scores 2 or less). No staining was seen in germinal center cells, plasma cells, or epithelial cells. The non-IM controls were completely negative for LMP1, except for cytoplasmic staining of eosinophils and rare plasma cells, also seen in IM cases and interpreted as false positive.
DISCUSSION This is the first report showing p53 expression in lymphoid tissues of primary EBV infection. In contrast, p53 was not detected in 11 benign, reactive lymph nodes. All IM tissues were confirmed to be EBV-related using serology and EBER1 in situ hybridization, whereas all non-IM patients were confirmed as unrelated to acute EBV infection by the same methods. The striking difference in p53 expression between these 2 cohorts suggests that p53 is specifically upregulated in EBV-related lymphadenitis, and it raises the issue of whether EBV infection causes p53 accumulation in IM tissues. Apparently EBV infection alone is not sufficient to cause a cell to express detectable p53, because the number of cells expressing EBER1 far exceeded the number of cells expressing p53 in every case of IM. To specifically evaluate whether p53 is altered in a subset
of infected cells undergoing viral replication, BZLF1 was assayed immunohistochemically. BZLF1 is the critical immediate-early viral protein controlling entry into the lytic phase of replication and acts as a transcription factor to initiate upregulation of a cascade of proteins needed for virion production, a Our results are in agreement with those of Niedobitek et al~ who found that cells expressing lyric cycle antigens were mainly small lymphocytes in paracortical areas and near crypt epithelium. But, in contrast to their findings, we did not demonstrate BZLF1 expression in plasma cell nuclei. It seems likely that BZLF1 expression varies with stage of cellular differentiation, and there is a narrow window during which BZLF1 is upregulated. In vitro studies have shown that BZLF1 protein not only binds p53 but also functionally inactivates it. 7,31 A major goal of this study was to determine whether this mechanism operates in vivo to sequester p53 in EBVreplicating lymphocytes. Although most IM tissues coexpressed both markers in a small fraction of lymphocytes, expression was not localized to the same cell types, suggesting that there is no direct interaction between BZLF1 and p53 in vivo. This finding does not exclude indirect interactions. Indeed, indirect effects are supported by co-localization of cells expressing either of these proteins to the same anatomic zones within lymphoid tissue. Expression of p53 in IM tissues indicates that p53 overexpression is not specific for malignancy but rather can be found in benign, reactive lymphoproliferations. This finding refutes the theory that p53 immunostains might help discriminate between benign and neoplastic lymph nodes, a6 Furthermore, overexpression of p53 in IM but not in the other reactive lymphadenopathies that we examined suggests a role for EBV in altering p53 expression. In vitro studies have shown that EBV-encoded LMP1 mediates upregulation of p53 levels by activation of NF-kB.15 This pathway represents another possible mechanism of p53 induction during acute EBV infection in vivo. To explore the possible relationship of LMP1 and p53 proteins in acute EBV infection, we studied expression of these two factors in our IM cases. We showed that LMP1 was expressed predominantly in large paracortical lymphocytes including RS-like cells, and that these cells were enriched near the tonsillar crypts. Expression of p53 was seen in the same cell types and in the same anatomic distribution, suggesting the possibility that p53 is upregulated by the LMPl-triggered NF-kB pathway described in vitro. In vitro studies have also shown that the concentration of p53 relative to that of LMP1 determines whether EBV-infected cells undergo apoptosis. ~,34 Although we could not measure protein concentrations by the methods employed in this study, we observed rampant lymphoid proliferation as well as focal apoptosis among infected cells, suggesting the possibility that p53 function is altered by viral infection. With current technology, it is impossible to determine whether p53 is functional or dysfunctional in naturally infected IM tissues. The p53 protein is an important negative regulator
1401
HUMAN PATHOLOGY
Volume 31, No. 11 (November 2000)
of cell proliferation that is inactivated and concomitantly overexpressed in many malignancies. 24-27In some of these tumors, overexpression of p53 protein in the absence of TP53 gene mutation suggests that the protein is rendered dysfunctional at a post-translational level. 2s-~° Presumably IM tissues do not harbor TP53 gene mutations, so the accumulation of p53 in these reactive lymphocytes implicates a post-translational effect of viral pathogenesis, or else a physiologic upregulation of p53 in response to EBV infection. Further analysis of the staining patterns of EBER1 and BZLF1 in IM tissues shows additional information that could shed light on viral pathogenesis. First, neither EBER] nor BZLF1 were expressed in tonsillar epithelial cells, confirming reports by several others 5,2~ that EBV is absent in normal epithelium. These data support the hypothesis that EBV shed in saliva from normal individuals probably originates from mucosal lymphocytes rather than epithelial cells. In fact, our identification of BZLFl-expressing lymphocytes localized in or near the surface mucosa of IM tonsils suggests that these lymphocytes could be the source of salivary virions. LMP1 is known to upregulate the intracellular levels of several antiapoptotic proteins, including bcl-2, A20, and Mcl-1. 32,3s It has been hypothesized that LMPl-mediated anti-apoptotic effects during the lyric cycle of viral replication lengthen cellular lifespan to permit even more virions to be produced. 32 In our study, LMP1 was not expressed in the same cell population as BZLF1. Our findings confirm those of Niedobitek et al, 3 indicating that BZLFl-positive cells are consistently LMPl-negative. This result implies that LMP1 is absent in naturally infected cells that are entering into the lytic cycle of viral replication. Based on our observations in IM cases, we propose that LMP1 expression is upregulated later than BZLF1 during lytic replication, once cells have transitioned from small resting lymphocytes into larger activated-appearing lymphocytes. The sequential expression of BZLF1 and then LMP1 during lyric replication would explain the co-localization of BZLF]- and LMPl-expressing lymphocytes in the mucosal region of the tonsil where virion shedding is thought to originate. Although it is tempting to speculate that LMPl-triggered NFkB activation is responsible for p53 upregulation in these lymphocytes, we cannot exclude the possibility some of the 50 or so other EBV gene products that are upregulated during lyric viral replication trigger p53 expression or inactivation. RS-like cells in IM were predominantly EBER1positive, LMPl-positive, and BZLFl-negative, which is the same phenotype as true RS cells of Hodgkin's disease. 17,~5-37Another cell type c o m m o n to both IM and Hodgkin's tissues is apoptotic cells, also referred to as mummified cells. The precise histogenesis of apoptotic cells in IM is difficult to determine because morphology is obscured, but in some cases their large size suggests derivation from RS-like cells. Their appearance and expression pattern combine to suggest that the apoptotic cells of IM are lymphocytes whose repli-
cative infection terminated in cell death. It is unclear whether cell death results from p53-mediated apoptosis versus the end product of virion production. -~s In conclusion, this study reveals several new observations. First, p53 accumulation is not specific for malignancy in lymph nodes. In fact, p53 expression is quite characteristic of the benign lymphadenitis caused by acute EBV infection. In contrast, the EBV-negative benign lymphadenopathies in this study did not express p53, implicating EBV infection as responsible for p53 accumulation. Viral BZLF1 co-localizes to the same areas of the node as p53, but not to the same cell types, suggesting no direct interaction between these factors. Coexpression of p53 and LMP1 in the same subset of paracortical large lymphocytes supports the concept that these cells are activated to produce new virions or die, perhaps depending on whether p53 is allowed to trigger apoptosis. Finally, our data support the growing body of literature suggesting that lymphocytes rather than oropharyngeal epithelial cells are the origin of infectious virions shed in saliva.
REFERENCES 1. Miller G: The switch between latency and replication of EBV. J Infect Dis 161:833-844, 1990 2. Tierney RJ, Steven N, Young LS, et al: Epstein-Barr virus latency in blood mononuclear cells: Analysis of viral gene transcription during primary infection and in the carrier state.J Viro168:73747385, 1994 3. Niedobitek G, Agathanggelou A, Herbst H, et al: Epstein-Barr virus infection in infectious mononucleosis: Virus latency, replication and phenotype of EBV-infected cells. J Pathol 182:151-159, 1997 4. Prang NS, Hornef MW, Jager M, et al: Lytic replication of Epstein-Barr virus in the peripheral blood: Analysis of viral gene expression in B lymphocytes during infectious mononucleosis and in the normal carrier state. Blood 89:1665-1677, 1997 5. Anagnostopoulos I, Hummel M, Kreschel C, et al: Morphology, immunophenotype, and distribution of latently a n d / o r productively Epstein-Barr virus-infected cells in acute infectious mononucleosis: Implications for the interindividual infection route of EpsteinBarr virus. Blood 85:744-750, 1995 6. Laytragoon-Lewin N, Chen F, Avila-CainoJ, et al: Epstein-Barr virus gene expression in lymphoid B-cells during acute infectious mononucleosis and clonality of the directly growing cell lines. I n t J Cancer 71:345-349, 1997 7. Zhang Q, Gutsch D, Kenney S: Functional and physical interaction between p53 and BZLFI: Implication for Epstein-Barr virus latency. Mol Cell Biol 14:1929-1938, 1994 8. Gulley ML, Burton MP, Allred DC, et al: Epstein-Barr virus infection is associated with p53 accumulation in nasopharyngeal carcinoma. HVM PATHOL 29:252-259, 1998 9. Lehtinen T, IsolaJ, Aine R, et ah Accumulation of p53 protein correlates with tumor proliferative activity in EBV positive Burkitt's lymphoma. Hematol Oncol 10:273-279, 1992 10. Balint E, Reisman D: Increased rate of transcription contributes to elevated expression of the mutant p53 gene in Burkitt's lymphoma cells. Cancer Res 56:1648-1653, 1996 11. Cherney BW, Bhatia KG, Sgadari C, et al: Role of the p53 tumor suppressor gene in the tumorigenicity of Burkitt's lymphoma cells. Cancer Res 57:2508-2515, 1997 12. Edwards RH, Raab-Traub N: Alterations of the p53 gene in Epstein-Barr virus associated immunodeficiency-related lymphomas. J Virol 68:1309-1315, 1994 13. Farrell PJ, Allan GJ, Shanahan F, et ah p53 is fi'equently mutated in Burkitt's lymphoma cell lines. EMBOJ 10:2879-2887, 1991 14. Lo KW, Mok CH, Huang DP, et al: p53 mutation in human nasopharyngeal carcinomas. Anticancer Res 12:195%1963, 1992
1402
p53 IN INFECTIOUSMONONUCLEOSIS (Ehsan et al) 15. Chen W, Cooper NR: Epstein-Barr virus nuclear antigen 2 and latent membrane protein independently transactivate p53 through induction of NF-kappaB activity. J Virol 70:4849-4853, 1996 16. Nakamura H, Said JW, Miller CW, et al: Mutation and protein expression of p53 in acquired immunodeficiency syndromerelated lymphomas. Blood 82:920-926, 1993 17. Reynolds DJ, Banks PM, Gulley ML: New characterization of infectious mononucleosis and a phenotypic comparison with Hodgkin's disease. A m J Pathol 146:379-388, 1995 18. StricklerJG, Fedeli F, Horwitz CA, et al: Infectious mononucleosis in lymphoid tissue: Histopathology, in situ hybridization, and differential diagnosis. Arch Pathol Lab Med 117:269-278, 1993 19. Ambinder RF, Mann RB: Detection and characterization of Epstein-Barr virus in clinical specimens. Am J Pathol 145:239-252, 1994 20. GuUey ML, Eagan PA, Quintanilla-Martinez L, et al: EpsteinBarr virus DNA is abundant and monoclonal in the Reed-Sternberg ceils of Hodgkin's disease: Association with mixed cellularity subtype and Hispanic American ethnicity. Blood 83:1595-1602, 1994 21. Allred DC, Clark GM, Elledge R, et al: Accumulation of mutant p53 is associated with increased proliferation and poor clinical outcome in node-negative breast cancer. J Natl Cancer Inst 85:200-206, 1993 22. Sauter ER, Cleveland D, Trock B, et al: p53 is expressed in fifty percent of pre-invasive lesions of the head and neck epithelium. Carcinogenesis 15:2269-2274, 1994 23. Karajannis MA, Hummel M, Anagnostopoulos I, et al: Strict lymphotropism of Epstein-Barr virus during acute infectious mononucleosis in nonimmunocompromised individuals. Blood 89:28562862, 1997 24. Levine A, Momand J, Finlay C: The p53 tumor suppressor gene. Nature 351:453-456, 1991 25. Vogelstein B, Kinzler K: p53 function and dysfunction. Cell 70:523-526, 1992 26. Hollstein M, Sidransky D, Vogelstein B, et al: p53 mutations in human cancers. Science 253:49-53, 1991 27. Martin A, Flaman JM, Frebourg T, et al: Functional analysis
of the p53 protein in AIDS-related non-Hodgkin's lymphomas and polymorphic lymphoproliferations. BrJ Haematol 101:311-317, 1998 28. Chilosi M, Doglioni C, Menestrina F, et al: Abnormal expression of the p53-binding protein MDM2 in Hodgkin's disease. Blood 84:4295-4300, 1994 29. Matsushima AY, Cesarman E, Chadburn A, et al: Post-thymic T cell lymphomas frequently overexpress p53 protein but infrequently exhibit p53 gene mutations. A m J Pathol 144:573-584, 1994 30. Villuendas R, Piris MA, Algara P, et al: The expression of p53 protein in non-Hodgkin's lymphomas is not always dependent on p53 gene mutations. Blood 82:3151-3156, 1993 31. Cayrol C, Flemington EK: Identification of cellular target genes of the EBV transactivator Zta: Activation of TGF-Bigh3 and TGF-B1. J Virol 69:4206-4212, 1995 32. Henderson S, Rowe M, Gregory C, et al: Induction of bcl-2 expression by Epstein-Barr virus latent membrane protein 1 protects infected B-cells from programmed cell death. Cell 65:110%1115, 1991 33. Chen Weiping, Huang Shuang, Cooper Neil R: Levels of p53 in Epstein-Barr virus-infected cells determine cell fate: Apoptosis, cell cycle arrest at the G1/S boundary without apoptosis, cell cycle arrest at the G2/M boundary without apoptosis, or unrestricted proliferation. Virology 251:217-226, 1998 34. Okan I, Wang Y, Chen F, et al: The EBV-encoded LMP1 protein inhibits p53-triggered apoptosis but not growth arrest. Oncogene 11:1027-1031, 1995 35. Siebert JD, Ambinder RF, Napoli VM, et al: Human immunodeficiency virus-associated Hodgkin's disease contains latent, not replicative, Epstein-Barr virus. HUM PATHOL26:1191-1195, 1995 36. Brousset P, Knecht H, Rubin B, et al: Demonstration of Epstein-Barr virus replication in Reed-Sternberg cells of Hodgkin's disease. Blood 82:619-624, 1993 37. Pallesen G, Sandvej Ig, Hamilton-Dutoit SJ, et al: Activation of Epstein-Barr virus replication in Hodgkin and Reed-Sternberg cells. Blood 78:1162-1165, 1991 38. Allday MJ: Apoptosis and Epstein-Barr virus in B cells. EBV Report 3:55-65, 1996
1403