Progressive multifocal leukoencephalopathy

Objective-Progressive multifocal leukoencephalopathy is caused by polyomavirus JC in immunosuppressed patients. IC virus genotypes are identified by sequence analysis of the viral genome. Despite the prevalence of acquired immunodeficiency syndrome in sub-Saharan Africa, few cases of progressive multifocal leukoencephalopathy have been reported from this region. Here we describe 4 African cases and provide an analysis of viral genotypes.

Methods.-Immunohistochemical staining by labeled streptavidin-biotin for capsid protein antigen was performed on all cases. Polymerase chain reaction amplification of viral genomic DNA was followed by direct cycle sequencing.

Results.-JC virus type 3 was identified in 2 cases, and type 6 was isolated in 1 case. The viral regulatory region from 1 case showed an uncommon rearrangement pattern.

Conclusions.-Progressive multifocal leukoencephalopathy in West African patients with acquired immunodeficiency syndrome is caused by African genotypes of JC virus (types 3 and 6). The prevalence of disease in this autopsy series from sub-Saharan Africa (1.5%) was less than has been reported from Europe and the United States (4% to 10%) and may be partly due to biological differences in IC virus genotypes. Further studies will be needed to confirm this observation.

(Arch Pathol Lab Med. 1999;123:395-403)

Progressive multifocal leukoencephalopathy (PML) is a fatal demyelinating disease of the central nervous system caused by the human polyomavirus JC (JCV). Prior to the acquired immunodeficiency syndrome (AIDS) pandemic, PML was described in rare cases of immunocompromised patients with hematologic malignancies, such as chronic lymphocytic leukemia and Hodgkin's disease.1 With the advent of human immunodeficiency virus (HIV) infection and AIDS, the reported incidence of PML in the AIDS patient population worldwide has increased, ranging from 0.8% of an autopsy series from Sao Paulo, Brazil,2 to 10% in a European study of 215 AIDS cases.3 Some of this disparity in incidence rates may be related to increased reporting and the application of advanced molecular biologic techniques to diagnosis. However, in view of the increased incidence of PML in the AIDS patient population, this subacute demyelinating disease can no longer be regarded as a rare condition.4,5

Acquired immunodeficiency syndrome and HIV infection are highly prevalent in sub-Saharan Africa,6 with estimated seropositivity in 16 million people, accounting for approximately 3% of the entire population of the subcontinent.7 Despite the high incidence of AIDS and HIV in this environment and the reported association of PML and immunodeficiency, very few cases of PML have been described in Africa.8-10

Human polyomavirus JC was first cultured from the brain of a patient with PML in 1971 by Padget and coworkers.11 Subsequently, the entire genome of the virus was sequenced in 1984 by Frisque et al.12 The genome of JCV (Mad-1) is a circular molecule of dsDNA and is 5.1 kilobases in length. It codes for the early proteins, large T and small t antigens, which regulate transcription of the late region proteins VP1-3 and agnoprotein, after viral DNA replication. Viral genes are transcribed bidirectionally from the origin of DNA replication within the regulatory region (ori). The JCV regulatory region shows an "archetypal" non-rearranged configuration when excreted in urine and displays unique rearrangements of the archetypal form, consisting of duplications and deletions within the promoter/enhancer when sequenced from PML brain tissue.13,14 Progressive multifocal leukoencephalopathy-type regulatory regions are derived from the archetype and are hypervariable and unique in each case, but can be classified into several patterns based on a scheme of deletions and duplications within certain domains identifiable in the JCV regulatory region.13-15The regulatory regions of polyomaviruses, including JCV, are responsible for control of viral transcription and replication,16 and changes in the nucleotide sequence within the viral regulatory region may influence the efficiency of viral transcription.17

Based on sequence analysis of segments of the VPI and large T-antigen genes18-20 and a phylogenetic analysis of complete genomes,21 JCV can be divided into several geographically based genotypes. Types 1 22,23 and 2 24,25 are of European and Asian origin, respectively, while types 3 and 6 2-30 are African in origin. Type 4 appears to be a recombinant form of genotypes 1 and 3, and so far has been found only within the United States.19 The biological significance of JCV genotypes in relation to PML is not clearly understood, although studies have shown that JCV type 2B is more frequently found in the brains of patients with PML than in JCV strains excreted in the urine of a control population.20,31

The pathognomonic feature of PML is a productive infection of oligodendrocytes in the brain, leading to an inability to maintain myelin and subsequent demyelination. This breakdown is usually accompanied by abortive infection of astrocytes with development of a neoplastic phenotype manifesting as giant and bizarre astrocytes, which may be described as "neoplastoid." Astrocytes may occasionally be infected permissively by JCV.32-34 There is associated reactive astrocytosis with occasional chronic inflammatory cell infiltrates. There are no significant morphologic differences between AIDS-related and non-- AIDS-related PML, except for an apparent increase in inflammatory infiltrates, necrosis, and focal encephalitis in AIDS-related cases, whereas non-AIDS-related PML cases show increased numbers of bizarre astrocytes.3,35,36 We describe the pathologic features and characterize the JC virus genotypes with regulatory region rearrangements in 4 African patients infected with HIV and PML who died in Cote d'Ivoire, West Africa.

MATERIALS AND METHODS

Brain sections were obtained from the 4 patients diagnosed postmortem with PML during an HIV autopsy study in Abidjan, Cote d'Ivoire, West Africa.6 These 4 patients represent 1.5% of the 266 cadavers of symptomatic HIV-positive adults who were autopsied with systematic sampling of fixed brain (8-12 blocks per case). The patients included 3 men and 1 woman; 2 had HIV-1, 1 had HIV-2, and 1 was dually seroreactive (Table 1). A total of 8 unstained, formalin-fixed, paraffin-embedded sections from representative PML lesions from each case were selected for further analysis. One section from each case was stained with hematoxylin-eosin and Luxol fast blue. Other diagnoses in the cerebral tissues were noted from the main study pathology database. Immunohistochemistry

After deparaffinization in xylene, sections were rehydrated in graded ethanol to water and then transferred to antigen retrieval solution Citra (BioGenex, San Ramon, Calif). Microwave antigen retrieval was carried out using published procedures.37 Endogenous peroxidases were neutralized with 0.3% hydrogen peroxide in methanol. Sections were washed in phosphate-buffered saline containing 0.05% Tween 20. Background staining was blocked using 10% normal goat serum for 30 minutes. The following primary antibodies were then applied to selected sections after dilution. Rabbit polyclonal antibody against simian virus 40 (SV 40), which cross-reacts with JCV capsid protein, was obtained from Lee Biomolecular Research Inc (San Diego, Calif; dilution, 1:50). Monoclonal antibodies to glial fibrillary acidic protein, dilution 1:50; T cells (CD45RO, UCHL-1), dilution 1:50; and B cells (CD20, L26), dilution 1:50, were all obtained from Dako Corporation (Carpinteria, Calif). Double-labeled immunohistochemistry was carried out on each case by a streptavidin-biotin amplification method using the Histostain-DS kit (Zymed Laboratories, San Francisco, Calif), following the recommended protocol. Staining for JCV capsid protein antigen and glial fibrillary acidic protein was performed on each case, while T and B cells were stained in cases 2 and 3 with moderate to severe inflammation. Positive control slides were obtained from a known AIDS-related PML case. Negative control slides were obtained from the cerebellum and cerebral cortex of an autopsy case without neurological disease. Both positive and negative control slides were processed in tandem with the cases under study.

Polymerase Chain Reaction

A 215-bp fragment was amplified from the VP1 major capsid protein gene by a two-step polymerase chain reaction (PCR) with the JCV-specific primers JLP-15 and -16. The method utilized a hot-start technique followed by 63 degC for annealing and elongation and 94 degC for denaturation for 1 minute each.14 After 50 cycles and a final extension period at 72 degC for 10 minutes, the reaction was stopped at 4 degC. Type determination in the VP1 gene was confirmed in a 141-bp fragment of the large T-antigen gene using primers JTP-1 and -2. In case 1, a 372-bp fragment of the VT-- intergenic region was amplified with primers VPV-5 and -6. Primer sequences are given in Table 2.

To determine the regulatory region sequence right of ori, primers JRR-15 and -34 were designed to anneal downstream of the TATA-box. Previous methods yielding longer PCR products did not result in DNA amplification, probably owing to extensive formalin fixation of the tissue. The 50-cycle program included steps for denaturation at 94 degC, annealing at 60 degC, and elongation at 72 degC for 1 minute each, followed by final elongation at 72 degC for 10 minutes. The time for denaturation was reduced to 30 seconds after 10 cycles. All PCRs were performed using UlTma DNA polymerase with 3'-5' exonuclease activity (Perkin Elmer Cetus, Norwalk, Conn).

Cycle Sequencing

Following purification by preparative gel electrophoresis, PCR products were directly sequenced using the primers shown in Table 2. Primers were end-labeled with ^sup 33^P-ATP (Amersham, Arlington Heights, Ill) mixed with 2% to 5% of the cleaned template, chain-terminating nucleotides, and a thermostable DNA polymerase (SequiTherm EXCEL, Epicentre Technologies, Madison, Wis). Initial denaturation for 1 minute at 95 degC was followed by 30 cycles of 30 seconds at 95 degC for denaturation and 1 minute at 63 degC for annealing and elongation. The products were analyzed on a 6% polyacrylamide gel containing 50% (w/v) urea (National Diagnostics, Atlanta, Ga). Gels were fixed with 12% (v/v) methanol plus 10% acetic acid and transferred to 3MM chromatography paper (Whatman, Maidstone, England). After drying under vacuum, the gel was exposed to Bio-Max x-ray film (Kodak, Rochester, NY).

JC virus genotypes were identified as previously described.26,27,29,30) Sequence analysis utilized GCG version-8 UNIX programs (Genetics Computer Group Inc, Madison, Wis). Primer design was assisted with OLIGO version 5 (NBI, Plymouth, Minn).

RESULTS

The clinical diagnoses in the 4 patients with PML (Table 1) were based on simple clinical examination only. No neuroradiologic studies were available. The 4 patients stayed in hospital 8, 2, 6, and 9 days prior to death.

Light Microscopy

Hematoxylin-eosin- and Luxol fast blue-stained sections from the 4 cases showed characteristic features of PML, including focal demyelination with infiltration by sheets of foamy macrophages, some of which were myelin-laden, exhibiting the characteristic feature of "gitterzellen." Oligodendrocyte nuclei were enlarged with eosinophilic, basophilic, and amphophilic inclusions in some cells, typical of infection with JCV (Figure 1). There was a paucity of infected oligodendrocytes in the central demyelinated foci and increased numbers toward the periphery of lesions. Foci with early demyelination had greater numbers of infected glial cells. Reactive astrocytes were numerous at the margins of lesions, and a few astrocytes with viral inclusions were seen in older lesions. Some of the reactive astrocytes were gemistocytic and had plump nuclei and abundant cytoplasm. Neoplastoid giant astrocytes were seen at the cortex-white matter junction (Figure 2, D) and within the central demyelinated foci of older lesions (Figure 2, B). Some were large cells with hyperchromatic pleomorphic nuclei and scanty cytoplasm. A bizarre astrocyte with a dividing or kidney-shaped nucleus is shown in Figure 2, E. Giant and bizarre astrocytes were not seen in the cerebellar lesion (case 4). The neuropil surrounding early demyelinative foci showed prominent vacuolation (Figure 2, H).

All 4 cases showed mild to severe mononuclear cell infiltrates (Table 3) which are mostly perivascular with sprinkled cells in the stroma. There was active invasion of blood vessel walls by chronic inflammatory cells in case

1. The cortex was uninvolved by PML lesions, except for mild reactive astrocytosis and occasional infected glial cells in case 2.

Other cerebral lesions and systemic diseases in these 4 patients are shown in Table 1. In case 2 there were small foci of cerebral toxoplasmosis, and in case 4 scanty cytomegalovirus inclusions were found within microglial nodules in the cerebrum and cerebellum.

Immunohistochemistry

Immunohistochemical staining for JCV capsid protein antigen was positive in all cases. Oligodendrocytes and occasional astrocytes were positive; however, bizarre and giant astrocytes were nonreactive. Antigen-positive glial cells were more commonly seen in early lesions (Figure 2, C) and distributed around the margins of older lesions. Neurons both in the cerebrum and cerebellum were negative for viral capsid protein antigen. Reactive astrocytes were positive for anti-glial fibrillary acidic protein and were mostly seen at the margins of lesions. Inflammatory infiltrates were strongly reactive for antibodies to T and B cells and showed a T-cell predominance (Figure 2, G) in the 2 cases examined by immunohistochemistry.

JC Virus Genotypes

JC virus DNA was amplified from the cerebral cortex of cases 1, 2, and 3, but not from the cerebellar lesion of case 4 (Table 3). This was probably due to prolonged formalin fixation and the small size of the lesion in this case. As shown in Figures 3 and 4, the partial sequence of the VP1 gene (JLP-15 and -16) and T-antigen gene (JTP-1 and -2) was identical to the type 3 sequence in cases 2 and 3. This genotype was originally detected in urine of Tanzanian patients positive for HIV37and in a Gambian AIDS patient.9 The viral DNA sequence from case 1 showed a different genotype. The case 1 sequence was similar to that of JCV from an African-American PML patient (strain 601), termed type 6.? Both are related to JCV genotypes recently confirmed as ancient African strains by identification in the urine of aboriginal African Pygmies and Bantu.26 The identity of the JCV genotype in case 1 was further confirmed in a larger segment of the VT-intergenic region (VPV-5 and -6 fragment, Figure 5).

IC Virus Regulatory Region

The predicted length of the PCR product based on the archetypal, non-rearranged regulatory region using primers JRR-15 and -34 is 241 bp. Polymerase chain reaction resulted in a single band of about 185 and 200 bp in cases 1 and 3, respectively. Double bands in case 2 were separated by about 80 bp, with the major product of 260 bp indicating that one of the primers annealed in a duplicated region. These sequencing results are shown in Figure 6 in comparison with the rearrangement patterns found in 40 cases of PML from the United States.13,14 A possibly new rearrangement pattern was identified in a Gambian HIV-2-positive patient with PML caused by a JCV type 3 strain, strain 313. 9 The regulatory region of strain 601 30 shows the long-duplicate pattern. Of the 3 Abidjan strains, strain 314 (case 2) and strain 602 (case 1) had the D region deleted, which is the most common rearrangement in PML strains. However, strain 315 (case 3) retained this portion, including position 133, where cytidine is characteristic of type 3 strains. This retained region corresponds to that duplicated in the Gambian strain 313 (Figure 6).

COMMENT

JC virus has a worldwide distribution and exists as geographically based genotypes,l8 21 perhaps due to coevolution with the human species.25,26 In this article we highlight the pathology and JCV genotypes from 4 cases of PML occurring in African AIDS patients. Despite the high prevalence of HIV infection and AIDS in sub-Saharan Africa, very few cases of PML have been reported from Africa,8,9 and the pathology has never been characterized in detail. The apparent dearth of reported cases was thought to be due to a lack of neuroradiologic facilities for antemortem diagnosis and a low autopsy rate.8,10 Recent reports, however, indicate that human polyomavirus JC, the causative agent for PML has distinct genotypes of African origin.23-30 Is it possible that African genotypes represent a form of JCV with diminished virulence? The 4 cases analyzed in detail here were part of a previously reported study that included the brains from 266 sampled patients with symptomatic HIV disease. Of these, 212 (80%) had AIDS by the 1987 Centers for Disease Control and Prevention criteria.6 The prevalence of PML in this cohort was 1.5% (4/266). This is considerably lower than the rate of 4% to 7% from the United States4,5 and the reported occurrence in autopsy studies from Europe.38 Only a Brazilian study has reported a lower prevalence (0.8%).2 The JCV genotypes circulating in Brazil have not been studied.

Possible viral influences on the pathogenicity and neurotropism of JCV in Africa may include not only the genotype of the viral coding region (type 3 or 6), but also the rearrangement of the archetypal viral regulatory region (promoter/enhancer) occurring in PML brain isolates of these genotypes.16,17 In general, JCV regulatory region rearrangements are each unique, but they appear to be variations on several recognizable patterns. These patterns have been termed "long duplicate," "short triplicate," and "D-retaining." 14 The "D" region of the archetypal regulatory region, usually deleted, comprises approximately 66 bp (positions 115-180) in the middle of the 267-bp archetypal regulatory region. Of these 3 patterns, the D-retaining pattern is the least common, representing about 15% of the strains studied.13,14 Recently, a fourth variant (strain 313) was observed in a single Gambian case9 and may represent a new rearrangement pattern of JCV regulatory regions in African PML patients.

Of the total of 5 type 3 and type 6 JCV strains from African or African-American PML brains studied to date, 3 have shown one of the less common rearrangements. Strain 313's is the only strain with region D duplication, whereas strains 315 (case 3) and 601 30 have partial or complete retention of region D (Figure 6). If the African host with PML rearranges the JCV regulatory region differently than European hosts, it may be mechanisms impacting the D region that are involved. The distinctiveness of these African rearrangements ensures that they represent new virus strains and not laboratory contaminants.

The other possibility could be that the rapid mortality associated with AIDS in sub-Saharan Africa allows few patients to survive long enough to develop this chronic progressive disease.6,39 This observation may be validated by a report from Europe, in which 6 PML cases were identified in 116 African AIDS patients (5.2%) (H. Taelman, written communication, March 1988).40 We may speculate that this group of patients has a higher incidence of PML because better medical care for AIDS, which is arguably available in Europe, would allow longer survival and increased opportunity to develop PML. However, the JCV genotypes associated with these cases are not known.

Pathologic examination of the Abidjan cases revealed the classic triad of multifocal demyelination, oligodendrocytopathy, and neoplastoid astrocytes in 3 of the 4 cases. Giant astrocytes were seen in greater frequency in case 2 (type 3) and were more prominent in the cortex-white matter junction area than within the center of demyelinating lesions, confirming reports by other authors.41,42 Immunohistochemical staining for JCV capsid protein antigen was not demonstrated in neoplastoid astrocytes, but was positive in a few infected astrocytes as previously reported.3243 There were no evident genotype-specific differences in morphology, although PML due to type 3 JCV (cases 2 and 3) showed an increased number of neoplastoid astrocytes when compared with type 6 (case 1). In addition, there was an increased number of perivascular chronic inflammatory infiltrates in the brains with type 3 strains when compared to the brain with type 6, although the latter had more extensive reactive astro(-ytosis. It seems unlikely that these differences represent JCV ge notype-specific effects, although a larger series should be studied before final conclusions are drawn.

Inflammatory infiltrates are a possible distinguishing feature between AIDS- and non-AIDS-related PML, inasmuch as increased incidence of inflammation is sometimes seen in AIDS-related cases.36 Reports have suggested that increases in the inflammatory infiltrates in PMI may be associated with prolonged survival. The 4 cases reported here exhibited various levels of chronic inflammation, ranging from mild to severe (Table 3). No data on survival from onset of PML or HIV disease were available for these patients. The CD4 counts indicated verv late IV disease in patients 1 and 4. These 2 patients had only mild inflammation (Table 3). Immunohistochemical characterization of infiltrates in cases 2 and 3 is consistent with findings in another series, which showed a mixed infiltrate of T and B cells with a T-cell predominance.?3

Opportunistic infections are a common occurrence in AIDS. Tuberculosis due to Mycobacterium tuberculosis and bacteremia caused by gram-negative organisms were the most common findings in this African series.6 The role of opportunistic infectious agents must be taken into consideration when determining the significance of inflammation in AIDS-related PML. Case 2, with mild focal toxoplasmosis, showed moderate inflammatory infiltrates, whereas in case 4, with mild cytomegalovirus encephalitis, there was little inflammation. Case 3, which showed the most severe inflammation, had no detected intracranial infection apart from PML.

Gray mater involvement by PML has been reported and is more common in AIDS-related cases.38 In this study, gray matter involvement was limited to a focus of a few infected glial cells observed in the cerebral cortex of case 2.

Vacuolation of white matter has been described in 2 cases of early PML,34 with another report suggesting morphologic similarity to Creutzfeld-Jakob disease.3 All of our cases revealed various stages of white matter vacuolation; the most severe changes were found in case 3, which had a majority of early and moderately aged lesions. There were no similarities in the vacuolar changes observed here to those of Creutzfeld-Jakob disease, since vacuolation or spongiform changes in Creutzfeld-Jakob disease are usually confined to the cortex and gray matter with involvement of neuronal and astrocytic processes. Although the etiology of white matter vacuolation in PML is uncertain, it has been suggested that lipid-laden macrophages may loose their fat content during processing, with secondary development of artifactual vacuoles in routinely processed histologic sections.44

Human immunodeficiency virus-2 is endemic to the West African subregion. Progressive multifocal leukoencephalopathy occurring in HIV-2-infected patients has been reported in the literature.3,9 One of the cases in our study was infected with HIV-2 alone (case 4), and another was dually infected with HIV-1 and HIV-2 (case 1). Lesions of PML in case 4 were confined to the cerebellum and appeared mild when compared with the other cases in our series. JC virus could not be amplified from this case by PCR, possibly owing to the small size of the lesion or the effects of fixation in formalin. However, in another PML case from a Gambian HIV-2-infected patient, JCV type 3 was amplified.9 Epidemiologic studies on HIV-2 in West Africa indicate that these patients survive longer than HIV-1-infected patients.45 It is difficult to predict whether PML in HIV-2-infected patients would be more common owing to longer overall survival time in which to develop PML, or less common owing to less severe immunosuppression. In addition, it is not known how possible transactivating effects of HIV-2 (eg, by way of the Tat protein) may compare with those of HIV-1.46

In conclusion, we report here the pathologic findings as well as the JCV genotypes and regulatory region sequences from 4 cases of PML occurring in Africans. Progressive multifocal leukoencephalopathy in Africa is caused by African genotypes of JCV There were no recognizable differences in the pathomorphology of African and non-African PML, although African genotypes of JCV do show distinct sequence variations in the coding regions for early and late proteins and in the regulatory region sequences (promoter/enhancer) when compared with non-African genotypes. It appears that PML in African patients is not rare, but may be less common in this environment than in European settings. To what extent differences may be explained by host genetic influences, JC viral genotype differences, or differences in the natural history of HIV in Africa remains to be explored.

We gratefully acknowledge the assistance of M. Honde, MD, and A. Kadio, MD, of the University Hospital in Treichville, Abidjan, Cote d'Ivoire, in the performance of the autopsy study.

References

1. Walker DL. Progressive multifocal leukoencephalopathy. In: Vinken PJ, Bruyn GW, Klawans FIL, Koetsier JC, eds. Demyelinating Diseases. Amsterdam, The Netherlands: Elsevier; 1985:503-524. Handbook of Clinical Neurology; vol 47.

2. Chimelli L, Rosemberg S, Hahn MD, Lopes MBS, Barretto Netto M. Pathology of the central nervous system in patients infected with the human immunodeficiency virus (HIV): a report of 252 autopsy cases from Brazil. Neuropathol Appl Neurobiol. 1992;18:478-488.

3. Kuchelmeister K, Gullotta F, Bergman M, Angeli G, Masini T. Progressive multifocal leukoencephalopathy in acquired immunodeficiency syndrome: a neuropathologic;il study. Pathol Res Pract. 1991:189:163 173. 4. Berger JR, Concha M. Progressive multifocal leukoencephalopathy: the evolution of a disease once considered rare. I Neurovirol. 1995;1:5-18. 5. Berger JR, Kaszovitz B, Post MI, Dickinson G. Progressive multifocal leukoencephalopathy associated with human immunodeficiency virus infection: a review of the literature with a report of sixteen cases. Ann Intern Med.1987;107: 78 87.

6. Lucas SB, Hounnou A, Peacock C, et al. The mortality and pathology of HIV infection in a West African city. AIDS. 1993;7:1569-1579. 7. Quinn TC. Global burden of HIV pandemic. Lancet.1996;348:99-106. 8. Lucas SB, Sewankambo N, Nambuya A, et al. The morbid anatomy of Af rican AIDS. In: Giraldo G, Beth-Giraldo E. Clumeck N, Gharbi MR, Ky,aIN,azi SK, de The G, eds. AIDS and Associated Cancers in Africa. Basel, Switzerland: Karger; 1988:124-133.

9. Stoner GL, Agostini HT, Ryschkewitsch CF, et al. Two cases of progressive multifocal leukoencephalopathy (PML) due to IC virus: detection of JCV type 3 in a Gambian AIDS patient. J Med Microbiol. 1997;47:733-742. 10. Williams AO. Pathology of HIV infection and AIDS in Africa. In: Williams AO, ed. AIDS: An African Perspective. Boca Raton, Fla: CRC Press; 1992:162232.

11. Padgett BL, Walker DL, ZuRhein GM, Eckroade RJ, Dessel BH. Cultivation of papova-like virus from human brain with progressive multifocal leucoencephalopathy. Lancet. 1971;1:1257-1260.

12. Frisque RJ, Bream GL, Cannella MT. Human polyomavirus JC virus genome. J Virol. 1984;51:458-469.

13. Ault GS, Stoner GL. Human polyomavirus JC promoter/enhancer rearrangement patterns from progressive multifocal leukoencephalopathy brain are unique derivatives of a single archetypal structure. J Gen Virol. 1993;74:14991507.

14. Agostini HT, Ryschkewitsch CF, Singer EJ, Stoner GL. JC virus regulatory region rearrangements and genotypes in progressive multifocal leukoencephalopathy: two independent aspects of virus variation. J Gen Virol. 1997;78:659-664. 15. Sugimoto C, Ito D, Tanaka K, et al. Amplification of JC virus regulatory DNA sequences from cerebrospinal fluid: diagnostic value for progressive multifocal leukoencephalopathy. Arch Virol. 1998:143:249-262.

16. Walker DL, Frisque RJ. The biology and molecular biology of JC virus. In: Salzman NP, ed. The Polyomaviruses. New York, NY: Plenum Press: 1986:327 377. The Papovaviridae; vol 1.

17. Roos RP. The regulation of gene expression in neurotrophic viruses. JNeurovirol. 1997;3:527.

18. Sugimoto C, Kitamura T, Guo I, et al. Typing of urinary IC virus DNA offers a novel means of tracing human migrations. Proc Natl Acad Sci U S A. 1997;94:9191-9196.

19. Agostini HT, Ryschkewitsch CF, Stoner GL. Genotype profile of human polyomavirus JC excreted in urine of immunocompetent individuals. J Clin Microbiol. 1996;34:159-164.

20. Agostini HT, Ryschkewitsch CF, Mory R, Singer El, Stoner GL. IC virus (ICV) genotypes in brain tissue from patients with progressive multifocal leukoencephalopathy (PML) and in urine from controls without PML: increased frequency of JCV type 2 in PML. J Infect Dis. 1997;176:1-8. 21. lobes DV, Chima SC, Rvschkewitsch CF, Stoner GL. Phylogenetic analysis of 22 complete genomes of the human polyomavirus IC. J Gen Virol. 1998;10: 2491-2498.

22. Agostini HT, Ryschkewitsch CF, Stoner GL. JC virus type 1 has multiple subtypes: three new complete genomes. I Gen Virol. 1998;79:801 805.

23. Stoner GL, Agostini HT, Rvschkewitsch CF, Komoly S. IC virus excreted by multiple sclerosis patients and paired controls from Hungary. Mult Scler. 1 998;4:45-48.

24. Agostini HT, Shishido-Hara Y, Baumhefner RW, Singer El, Ryschkewitsch CF, Stoner GL. IC virus type 2: definition of subtypes based on analysis of ten complete genomes. I Gen Virol. 1998;79:1143-1151.

25. Agostini HT, Ryschkewitsch CF, Yanagihara R, Davis V, Stoner GL. Asian genotype of IC virus (JCV) in Native Americans and in a Pacific Island population:

markers of human evolution and migration. Proc Natl Acad Sci U S A. 1997;94: 14542-14546.

26. Chima SC, Ryschkewitsch CF, Stoner GL. Molecular epidemiology of human polyomavirus JC in the Biaka pygmies and Bantu of central Africa. Mem Inst Oswaldo Cruz. 1998;93:615-624.

27. Agostini HT, Brubaker GR, Shao J, et al. BK virus and a new type of )C virus excreted by HIV-1 positive patients in rural Tanzania. Arch ViroL 1995;140: 1919-1934.

28. Guo J, Kitamura T, Ebihara H, et al. Geographical distribution of the human polyomavirus IC virus type A and B and isolation of a new type from Ghana. I Gen Virol. 1996;77:919-927.

29. Agostini HT, Ryschkewitsch CF, Brubaker GR, Shao J, Stoner GL. Five complete genomes of JC virus type 3 from Africans and African Americans. Arch Virol.1997;142:637-655.

30. Agostini HT, Ryschkewitsch CF, Stoner GL. The complete genome of JC virus type 6 from the brain of an African-American with progressive multifocal leukoencephalopathy (PML). J Hum Virol. 1998;1:267-272. 31. Agostini HT, Ryschkewitsch CF, Singer EJ, Baumhefner RW, Stoner GL. IC virus type 2B is found more frequently in brain tissue of progressive multifocal leukoencephalopathy patients than in urine from controls. J Hum Virol. 1998;1: 200206.

32. Mazlo M, Tariska I. Are astrocytes infected in progressive multifocal leukoencephalopathy (PML)? Acta Neuropathol (Berl). 1982;56:45-51.

33. Teo CG, Wong SY, Best PV. IC virus genomes in progressive multifocal leukoencephalopathy: detection using a sensitive non-radioisotopic in situ hybridization method. J Pathol. 1989;157:135-140.

34. Astrom KE, Stoner GL. Early pathological changes in progressive multifocal leukoencephalopathy: a report of two asymptomatic cases occurring prior to the AIDS epidemic. Acta Neuropathol (Berl). 1994;88:93-105.

35. Aksamit AJ, Gendelman HE, Orenstein JM, Pezeshkpour GH. AIDS-associated progressive multifocal leukoencephalopathy (PML): comparison to non-- AIDS PML with in situ hybridization and immunohistochemistry. Neurology. 1990;40:1073-1078.

36. Schmidbauer M, Budka H, Shah KV. Progressive multifocal leukoencephalopathy (PML) in AIDS and in the pre-AIDS era: a neuropathological comparison using immunocytochemistry and in situ DNA hybridization for virus detection. Acta Neuropathol (Berl). 1990;80:375-380.

37. Leong AS, Milios J, Leong FJ. Epitope retrieval with microwaves: a comparison of citrate buffer and EDTA with three commercial retrieval solutions. Appl Immunohistochem. 1996;4:201-207.

38. Martinez AJ, Sell M, Mitrovics T, et al. The neuropathology and epidemiology of AIDS: a Berlin experience: a review of 200 cases. Pathol Res Pract. 1995;191:427-443.

39. Grant AD, Djomand G, De Cock KM. Natural history and spectrum of disease in adults with HIV/AIDS in Africa. AIDS. 1997;11:S43-S54.

40. Sonnet J, Taelman H. Clinical and biologic profile of African AIDS: a study of 42 patients. In: Stacquet M, Hemmer R, Baert A, eds. Clinical Aspects of AIDS and AIDS-Related Complex. Oxford, England: Oxford University Press; 1986:7879.

41. Weinstein VF, Woolf AL, Meynell MJ. Progressive multifocal leukoencephalopathy and primary hypersplenism. J Clin Pathol. 1963;26:405418.

42. Bhatia KP, Morris JH, Frackowiak RS. Primary progressive multifocal leukoencephalopathy presenting as an extrapyramidal syndrome. I Neurol. 1996;243:91-95.

43. Aksamit AJ, Sever IL, Major EO. Progressive multifocal leukoencephalopathy: JC virus detection by in situ hybridization compared with immunohistochemistry. Neurology. 1986;36:499-504.

44. Wiley CA, Grafe M, Kennedy C, Nelson IA. Human immunodeficiency virus (HIV) and JC virus in acquired immune deficiency syndrome (AIDS) patients with progressive multifocal leukoencephalopathy. Acta Neuropathol (Berl). 1988;76:338-346.

45. Whittle H, Morris I, Todd J, et al. HIV-2-infected patients survive longer than HIV-1-infected patients. AIDS. 1994;8:1617-1620. 46. Krachmarov CP, Chepenik LG, Barr-Vagell S, Khalili K, Johnson EM. Activation of the IC virus Tat-responsive transcriptional control element by association of the Tat protein of human immunodeficiency virus 1 with cellular protein Pura. Proc Natl Acad Sci U S A. 1996;93:14112-14117.

Accepted for publication November 18, 1998.

From the Neurotoxicology Section, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Md (Drs Chima, Agostini, and Stoner and Ms Ryschkewitsch), and UMDS Guy's and St Thomas' Medical and Dental School, Department of Histopathology, St Thomas Hospital, London, United Kingdom (Dr Lucas). Dr Agostini is now with the Department of Ophthalmology, University of Freiburg, Germany.

Reprints: Gerald L. Stoner, PhD, National Institutes of Health, Bldg 36, Room 4A-27, Bethesda, MD 20892-4126.

Copyright College of American Pathologists May 1999
Provided by ProQuest Information and Learning Company. All rights Reserved