Restricted use of VH4 germline segments in an acute multiple sclerosis brain.

Multiple sclerosis (MS) cerebrospinal fluid and brain contain increased IgG and oligoclonal bands. Whether this oligoclonal and polyclonal IgG is directed against a disease-relevant antigen remains unknown. To distinguish between random activation versus a targeted B-cell response, we analyzed the IgG heavy chain variable region (VH) repertoire expressed in different lesions of an acute MS brain. To obtain a representative sample of the VH repertoire, we constructed directional complementary DNA libraries from plaque-periplaque messenger RNA and amplified VH regions from the library by nested polymerase chain reaction. When MS VH sequences were aligned to germline segments, about 60% of different VH sequences in the acute MS brain were VH4 germline segments, significantly greater than the known approximately 20% VH4 germline prevalence. Specific VH sequences were overrepresented and expressed at multiple plaque sites. Within some overexpressed populations, there were distinct sequence differences (clonal variants) indicative of clonal expansion. Alignment of VH sequences to their closest germline counterparts revealed extensive somatic mutation and the preferential accumulation of amino acid replacement mutations in complementarity determining regions. These observations suggest the limited B-cell response found in this acute MS brain was antigen driven.

Extraction and purification of active IgG from SSPE and MS brain.

Immunoglobulin (Ig) G was purified from soluble and membrane fractions of postmortem subacute sclerosing panencephalitis (SSPE) brain, multiple sclerosis (MS) brain plaque-periplaque white matter, and normal human brain (NHB) white matter. After homogenization in 0.32 M sucrose and removal of cell debris and nuclei by low-speed centrifugation, soluble and crude membrane fractions were separated by ultracentrifugation. After removal of sucrose by dialysis, IgG was isolated from the soluble fraction by protein A affinity chromatography. IgG was obtained from the membrane fraction by elution at low pH and purification from the eluate by protein A chromatography. Whereas very little IgG was in NHB white matter, significant levels of IgG were recovered from both SSPE and MS brain. Both immunocytochemical staining of measles virus-infected cells in tissue culture and protein immunoblotting of virus-infected cell lysates showed that the IgG from SSPE brain contained activity specific for measles virus protein. The abundance, purity and functional activity of IgG extracted from SSPE and MS brain indicate that IgG extracted from the brain of humans with an inflammatory disease of unknown etiology can be used to identify its corresponding antigen.

Strategies to identify sequences or antigens unique to multiple sclerosis.

Chronic inflammatory and infectious diseases of the central nervous system (CNS) are characterized by increased IgG and oligoclonal bands (OGBs) in the brain and cerebrospinal fluid (CSF). The OGBs in CNS infectious diseases of known cause have been shown to be directed against the pathogenic agent. In multiple sclerosis (MS), the antigenic specificity of the OGBs is unknown, but could be directed against an infectious agent, an autoantigen, or both. In a molecular approach to identify antigens specific for MS, we constructed directional cDNA expression libraries with mRNA extracted from chronic and acute MS plaques and periplaque white matter. The libraries were: (1) screened to identify clones whose expression products react with MS CSF, but not with CSF from other infectious and inflammatory diseases of the CNS; (2) subtracted by hybridization to mRNA from normal human brain white matter and differentially screened to detect unique MS transcripts; and (3) used as template in polymerase chain reactions to amplify, clone, and sequence IgG heavy and light chain variable regions (VH and VL, respectively) expressed in MS plaques. Analysis of the VH and VL IgG repertoire in MS brain may identify disease-relevant IgG sequences that can be assembled into functional antibodies using recombinant phage technology. Such recombinant antibodies will be useful to probe brain tissue to identify antigens unique to MS.

The search for virus in multiple sclerosis brain.

Plaque-periplaque areas from MS brain tissue were explanted and propagated in tissue culture. The same in vitro techniques that successfully rescued measles virus from SSPE brain, papovavirus from PML brain, and HSV from normal human trigeminal ganglia, were applied. MS brain cells were also inoculated into chimpanzees, multiple rodent species, and embryonated hens eggs. No neurologic disease developed in experimentally infected animals, and no cytopathic effect was observed in explanted cells, or after cocultivation or fusion of MS brain cells with indicator cells. Further analysis of explanted and cocultivated cells by indirect immunofluorescence with various antiviral antisera prepared against viruses associated with post-infectious encephalomyelitis, as well as antisera to other ubiquitous viruses, failed to detect viral antigen. Finally, attempts to detect a latent enveloped virus in MS brain cells by 'superinfecting' MS brain cells in culture with vesicular stomatitis virus (VSV) did not reveal a VSV non-neutralizable fraction. Nevertheless, since oligoclonal bands (OGBs) in the CSF of patients with chronic infectious diseases of the CNS are directed against the causative agent, it is likely that OGBs in MS CSF are antibody directed against the agent or antigen that triggered disease. Although the relevant antibody may be scarce relative to irrelevant antibody in MS CSF, and only small amounts of an MS-specific antigen may be present in brain, this report provides a rationale for strategies proposed in our companion report by Owens et al which will allow detection of an MS-specific antigen or its cognate RNA in brain.

Varicella-zoster virus (VZV) transcription during latency in human ganglia: construction of a cDNA library from latently infected human trigeminal ganglia and detection of a VZV transcript.

The entire varicella-zoster virus (VZV) genome appears to be present in latently infected human ganglia, but the extent of virus DNA transcription is unknown. Conventional methods to study virus gene transcripts by Northern (RNA) blotting are not feasible, since ganglia are small and VZV DNA is not abundant. To circumvent this problem, we prepared radiolabeled cDNA from ganglionic RNA, hybridized it to Southern blots containing VZV DNA, and demonstrated the presence of a transcript within the SalI C fragment of the virus genome (R. Cohrs, R. Mahalingam, A. N. Dueland, W. Wolf, M. Wellish, and D. H. Gilden, J. Infect. Dis. 166:S24-S29, 1992). To further map VZV transcripts, in the work described here we constructed a cDNA library from poly(A)+ RNA obtained from latently infected human ganglia. Phage DNA isolated from the library was used in PCR amplifications to detect VZV-specific inserts. The specificity of the PCRs was provided by selection of a primer specific for VZV gene 17, 18, 19, 20, or 21 and a second vector-specific primer. VZV gene 21-specific sequences were identified by PCR amplification. The PCR product contained the XhoI cloning site and poly(A)+ sequences between vector and VZV gene 21 sequences. The sequence motif at the 3' end of VZV gene 21, determined by cloning and sequencing of the PCR product, consisted of 49 to 51 nucleotide bases of 3'-untranslated DNA, the termination codon for the VZV gene 21 open reading frame, and DNA sequences reading into the VZV gene 21 open reading frame.

Varicella-zoster virus reactivation without rash.

Reactivation of varicella-zoster virus (VZV) leads to localized zoster (shingles), a syndrome characterized by pain and a vesicular rash. Rarely, patients experience radicular pain without zosteriform rash, cases that have been regarded as zoster sine herpete (zoster without rash). Virologic evidence for zoster sine herpete is sparse. However, VZV can produce other neurologic and visceral diseases in the absence of rash or radicular pain. The clinical and virologic features of zoster sine herpete and other disorders produced by VZV without rash are reviewed. Evidence is also presented for the detection of VZV DNA in human blood mononuclear cells of elderly individuals in the absence of skin lesions or other VZV-associated neurologic or systemic disease.

Acute simian varicella infection. Clinical, laboratory, pathologic, and virologic features.

Five African green monkeys inoculated intratracheally with 7.5 x 10(3) to 1.4 x 10(5) plaque-forming units of simian varicella virus (SVV) were subjected to clinical, laboratory, pathologic, and virologic analyses to study the pathogenesis of acute varicella. All animals developed viremia and rash and were sacrificed 8 to 11 days post-infection. No serum was available for postmortem serologic studies. Examination of multiple organs for pathologic changes and for SVV-specific antigen and nucleic acid revealed inflammation, hemorrhagic necrosis, and intranuclear Cowdry A inclusions in liver, lung, lymph node, and spleen; mild inflammation without necrosis in adrenal gland, kidney, and bone marrow, and SVV-specific antigen and nucleic acids in all viscera examined. No pathologic changes, SVV antigen or nucleic acids were detected in the spinal cord or in the brain from any of the monkeys. Ganglia revealed mild inflammation but no necrosis, and intranuclear inclusion bodies in non-neuronal cells of one trigeminal ganglion; SVV antigen and nucleic acids were detected in both non-neuronal and neuronal cells in ganglia. The pathologic and virologic findings in viscera are consistent with those described in viscera of humans with disseminated zoster, but the mild inflammatory changes in ganglia during acute simian varicella infection contrast with the extensive hemorrhagic necrosis and intranuclear inclusion bodies seen in human ganglia after disseminated varicella or zoster. Nevertheless, these studies show that ganglia become infected with varicella virus during primary infection, although the route of primary ganglionic infection remains to be determined, and indicate the possible usefulness of the SVV model to study varicella pathogenesis in humans.

Peripheral blood mononuclear cells of the elderly contain varicella-zoster virus DNA.

Peripheral blood mononuclear cells (PBMC) from humans of different ages were analyzed for DNA sequences specific for varicella-zoster virus (VZV) genes 29 and 62 by polymerase chain reaction (PCR). Neither VZV gene was detected in DNA from umbilical cord blood PBMC of 10 infants or from blood PBMC of two 3-year-old children. In 22 humans less than 60 years old, gene 29 was not detected, and gene 62 was detected in only one subject. In 33 humans greater than 60 years old, including patients with postherpetic neuralgia, PBMC from 4 subjects contained gene 29, 4 contained gene 62, and 1 contained both genes. The presence of VZV DNA correlated significantly with age (P less than .05, chi 2 and logistic regression analysis), but not with gender or postherpetic neuralgia.