But Significantly Different From, Human Herpesvirus 6 and Human Cytomegalovirus (Chronic Fatigue Syndrome/T Lymphocytes) Zwi N

Home , Human herpesvirus 6

Proc. Natd. Acad. Sci. USA Vol. 89, pp. 10552-10556, November 1992 Medical Sciences Human herpesvirus 7 is a T-lymphotropic virus and is related to, but significantly different from, human herpesvirus 6 and human cytomegalovirus (chronic fatigue syndrome/T lymphocytes) Zwi N. BERNEMAN*, DHARAM V. ABLASHIt, GE LI*, MAUREEN EGER-FLETCHER*, MARVIN S. REITZ, JR.*, CHIA-LING HUNGO§, IRENA BRUS¶, ANTHONY L. KOMAROFFII, AND ROBERT C. GALLO*,** *Laboratory of Tumor Cell Biology and tLaboratory of Cellular and Molecular Biology, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892; TPharmacia, Columbia, MD 21045; NMount Sinai School of Medicine, New York, NY 10029; and IlDepartment of Internal Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, MA 02115 Communicated by Maurice Hilleman, August 6, 1992

ABSTRACT An independent strain (JI) of human herpes- for 1-2 hr at 370C with virus-containing culture supernatant. virus 7 (HV-7) was isolated from a patient with chronic Afterward, the cell lines were cultured in RPMI 1640 medi- fatigue syndrome (CFS). No s cant association could be um/2-10%o fetal calf serum, whereas the CBMC were grown established by seroepidemiology between HHV-7 and CFS. in RPMI 1640 medium/5-20%o fetal calf serum with or with- HHV-7 is a T-lymphotropic virus, infecting CD4+ and CD8+ out 10% (vol/vol) interleukin 2 (Advanced Biotechnologies, primary lymphocytes. HHV-7 can also infect SUP-T1, an Columbia, MD). Infection was assessed by the appearance of immature T-cell line, with variable success. Southern blot large cells, immunofluorescence, and EM. analysis with DNA probes scanning 58.8% of the human Immunofluorescence. Indirect immunofluorescence with herpesvirus 6 (HHV-6) genome and hybridizing to all HHV-6 human polyclonal and murine monoclonal antisera was done strains tested so far revealed homology to HHV-7 with only and uninfected cells on 8- or 10-well slides 37.4% ofthe total probe length. HHV-7 contains the GGGTTA on infected control repetitive sequence, as do HHV-6 and Marek's disease chicken (1). The following HHV-6 monoclonal antibodies (mAbs) herpesvirus. DNA sequencing of a 186-base-pair fragment of were used: 9A5D12 (p41), 12B3G4 (p135), 6A5G3 (gpll6/ HHV-7(JI) revealed an identity with HHV-6 and human cyto- 64/54), 2D6 (gp82/105), 7A2 (gp102) (9) (from Bala Chan- megalovirus of 57.5% and 36%, respectively. Oligonudeotide dran, University of Kansas Medical Center, Kansas City); primers derived from this sequence (HV7/HV8, HV10/HV11) H-JG 16 (gplSO), H-AR 9 (gpl90), and H-AR 1-5 (gpll0/60) amplified HHV-7 DNA only and did not amplfy DNA from (from Janos Luka, University of Nebraska Medical Center, other human herpesviruses, incuding 12 different HHV-6 Omaha); C-5 (p38/41) (10) (from Gary Pearson, Georgetown strains. Southern blot analysis with the p43L3 probe containing University, Washington). The nature of infected cells was the 186-base-pair HHV-7 DNA fragment hybridized to HHV-7 determined by double immunofluorescence as described for DNA only. The molecular divergence between human cyto- Fig. 1. megalovirus, on the one hand, and HHV-6 and HHV-7, on the Southern Blot Hybridization. High-molecular-weight DNA other, is greater than between HHV-6 and HHV-7, which, in from cells infected with the following HHV-6 strains was turn, is greater than the difference between HHV-6 strains. examined: GS (1), CO1, C02, C03, C05 (11), U1102 (2), Z29 This study supports the classification ofHHV-7 as an additional (4), OK (12), KF, BA (13), SF (14), and DA (15). DNA from member of the human f3-herpesviruses. cells infected with the other human herpesviruses-herpes simplex virus type 1, Epstein-Barr virus, varicella-zoster An additional lymphotropic herpesvirus, human herpesvirus virus, and HCMV-was also analyzed (16). After digestion 6 (HHV-6), was reported a few years ago by several labora- with HindIII or BamHI, Southern blot analysis was done as tories (1-4). HHV-6 productively infects T lymphocytes in described for Fig. 3. The HHV-6 probes (16-21) used are vitro (5) and in vivo (6). Recently, another T-lymphotropic listed in Table 1. herpesvirus, human herpesvirus 7 (HHV-7), was described PCR Amplification, Cloning, and Sequencing of HHV-7 by Frenkel et al. (7). Independently, we have isolated HHV-7 DNA. HHV-7 sequence and clone were obtained as follows: (strain JI) from peripheral blood mononuclear cells (PBMC) Initial PCR amplification was in a 100-iul reaction, containing of a patient with chronic fatigue syndrome (CFS) (8). We 2 pug of HHV-7-infected SUP-T1 cell DNA; 10 mM Tris HCI describe here the serological, immunological, and molecular (pH 8.3); 50 mM KCl; 1.5, 2.0, or 2.5 mM MgCl2; 0.01% characteristics of HHV-7(JI). We also show that HHV-7 has gelatin; 0.2 mM (each) dATP, dCTP, dGTP, and dTTP; 4 ILM a tropism for primary human T lymphocytes and for the SUP-T1 T-cell line. A sequence was obtained of HHV-7 Abbreviations: CBMC, cord blood mononuclear cells; CFS, chronic DNA, which shows a homology with HHV-6 and with human fatigue syndrome; HCMV, human cytomegalovirus; HHV-6 and cytomegalovirus (HCMV).tt This HHV-7 sequence enabled HHV-7, human herpesvirus 6 and 7, respectively; PBMC, peripheral us to establish molecular reagents for identification of the blood mononuclear cells; mAb, monoclonal antibody. §Present address: Advanced BioScience Laboratories, Kensington, virus.# MD 20895. **To whom reprint requests should be addressed at: Laboratory of MATERIALS AND METHODS Tumor Cell Biology, National Cancer Institute, National Institutes of Health, Building 37, Room 6A09, Bethesda, MD 20892. In Vitro Infection. Cell lines or phytohemagglutinin A-stim- ttThe sequences reported in this paper have been deposited in the ulated cord blood mononuclear cells (CBMC) were incubated GenBank data base (accession no. L03525 for HHV-7 and L03526 for HHV-6). #lThis work was presented, in part, at the Annual Meetings of the The publication costs of this article were defrayed in part by page charge Laboratory ofTumorCell Biology, Aug. 17,1990, and Sept. 3, 1991, payment. This article must therefore be hereby marked "advertisement" National Cancer Institute, Bethesda, MD, and at the 15th Inter- in accordance with 18 U.S.C. §1734 solely to indicate this fact. national Herpesvirus Workshop, July 9, 1991, Pacific Grove, CA. 10552 Downloaded by guest on October 2, 2021 Medical Sciences: Bernernan et A Proc. Natl. Acad. Sci. USA 89 (1992) 10553 Table 1. Southern blot analysis of HHV-7(JI) DNA with Table 2. Sequence of DNA oligonucleotides used HHV-6 probes Name Sequence HHV-6 Insert Hybridization Primer for PCR probe size, kb to HHV-7(JI) Source (ref.) HV3.1 5'-GCNCCNTAYGAYATYCAYTTC-3' pZVH14 8.7 - (16,17) HV3.2 5'-GCNCCNTAYGAYATYCAYTTT-3' pZVB70 22.3 + (0.1x SSC) (18,19) HV3.3 5'-GCNCCNTAYGAYATACAYTTC-3' pZVB9 11.0 - (18) HV3.4 5'-GCNCCNTAYGAYATACAYTTT-3' pZVB43 8.0 - (17,18) HV4.1 5'-RAANARNGTNGCRATNGGNGT-3' pZVB10 6.2 + (0.1x SSC) (18) HV4.2 5'-RTANARNGTNGCRATNGGNGT-3' pZVB12 6.0 - (18) HV4.3 5'-RAANARNGTNGCTATNGGNGT-3' pZVB54 5.5 - (18) HV4.4 5'-RTANARNGTNGCTATNGGNGT-3' pZVB15 3.4 - (18) HV7 5'-TATCCCAGCTGTTTTCATATAGTAAC-3' pZVB16 2.5 - (18) HV8 5'-GCCTTGCGGTAGCACTAGATTTTTTG-3' pZVB2 2.0 - (18) HV10 5'-CAGAAATGATAGACAGATGTTGG-3' pZVB85 1.9 (18) HV11 5'-TAGATTTTTTGAAAAAGATTTAATAAC-3' pHD5 5.5 - (20, 21) Probe for pSMD6a 5.6 - (21) hybridization pH6Z-101 6.9 + (0.5x SSC) P. Pellett HV1 5'-(GGGTTA)6-3' The pHD5 and pSMD6a probes were from R. Honess (National HV9 5'-CCTAATGAAGGCTACTTTGAAGTACAAATG-3' Institute of Medical Research, London), and the pH6Z-101 probe HV12 5'-AGAATTCTGTACCCATGGGCACATTTGTAC-3' was from P. Pellett (Centers for Disease Control, Atlanta). SSC N, any nucleotide (A, C, G, T); Y, pyrimidine nucleotide (C, T); concentrations in parentheses indicate the most stringent wash after and R, purine nucleotide (A, G). Based on regions of amino acid hybridization at which signal could still be detected. identity (see Fig. 4), degenerate oligonucleotide primers were synthesized for PCR analysis: HV3.1, HV3.2, HV3.3, and HV3.4 for the primers HV3.1, HV3.2, HV3.3, HV3.4, HV4.1, HV4.2, APYDIHF peptide motif in a sense orientation, and HV4.1, HV4.2, HV4.3, and HV4.4; and 5 units of Taq DNA polymerase. HV4.3, and HV4.4 for the TPIATLF/Y motif, in an antisense Forty cycles ofPCR amplification were done according to the direction. Localization ofthe oligonucleotides on the 43L3 sequence following program: 94°C for 1 min, 45°C for 2 min, 72°C for were as follows: HV7, 1-26 (sense); HV8, 186-161 (antisense); HV9, 2 min with an increase of 2 sec an 82-111 (sense); HV10, 48-70 (sense); HV11, 171-145 (antisense); per cycle, and additional and HV12, 132-103 (antisense). 7 min at the end at 72°C. The PCR products were analyzed as described elsewhere (22). The polyacrylamide gel piece con- RESULTS taining the 228-base-pair (bp) PCR product was cut out, and the amplified DNA was eluted out of the gel piece in 0.5 M Cell Tropism of HIHV-7(JI). Double immunofluorescence ammonium acetate/i mM EDTA at 37°C overnight. After showed that infected CBMC are predominantly T lympho- phenol/chloroform extraction and ethanol precipitation, the cytes (Fig. 1). Infected cells, as determined by immunoflu- amplified fragment was redissolved in distilled water, treated orescence with patient's serum, showed the following char- with T4 polynucleotide kinase, extracted with phenol/ acteristics: 97% CD2+, 93% CD3+, 95% CD7+, 42% CD4+, chloroform, ethanol-precipitated, redissolved in distilled wa- 4% CD8+, 0%6 CD1+, 0%o CD19+, 0%6 membrane and cyto- ter, and blunt-end-ligated overnight at room temperature into plasmic-immunoglobulin+, 0o CDllb+, and 0%o CD16+. We EcoRV-digested and phosphatased pBluescript plasmid were unable to productively infect the following cell lines: (Stratagene). Recombinant bacterial colonies were screened CEM, CEM(SS), MOLT-3, MOLT-4, MOLT-4/8, HSB-2, with the 32P-labeled amplified HHV-7 DNA fragment. The HSB-8, Jurkat, VDSO, H9, H.11.8, GB, Alex, Ramos, Raji, insert of one selected recombinant B-jab, SB, U937, HL-60, HeLa, A-204, HFS, and Vero. Only plasmid (designated 3L25) an was sequenced by using the Sequenase version 2.0 DNA SUP-T1, immature T-cell line derived from a childhood kit T-cell non-Hodgkin's lymphoma (23), has to date been shown sequencing (United States Biochemical). Based on the to DNA sequence located to the support an infection with HHV-7, as demonstrated by internally regions covered by immunofluorescence with patient's serum the degenerate the HV7 and HV8 were and EM. How- primers, primers ever, not all our efforts to productively infect SUP-T1 with synthesized (Table 2). Thirty cycles of PCR amplification HHV-7(JI) have been successful. Infection of were done in a final volume of 100 ,l containing 1 ,ug of SUP-T1 with HHV-7-infected CBMC DNA, 10 mM Tris HCl HHV-7(JI) resulted in the appearance of large cells, some of (pH 8.3), 50 which adhere to the bottom of the culture flask. The propor- mM KCI, 1.5 mM MgCl2, 0.01% gelatin, 0.2 mM (each) tion of infected cells, as determined by immunofluorescence dNTP, 0.5 AM of the HV7 and HV8 primers, and 2.5 units of with patient's serum, ranged between a few percent to 80% Taq DNA polymerase, with the following program: 1 min at 2 but was generally <20%. In some SUP-T1 cultures, there was 94°C, min at 60°C, 2 min at 72°C with an increase of 2 sec a persistence of infection with HHV-7(JI) for at least 2 mo per cycle, and an additional final extension at 72°C for 7 min. (Fig. 2), before overgrowth of infected cells by uninfected The amplified product was cloned using the TA cloning cells. system version 1.0 (Invitrogen). Transformed bacterial col- Immunological Relatedness of HHV-7(JI) to HHV-6. In- onies were screened with the 32P-end-labeled HV9 oligonu- fected CBMC and SUP-T1 cells only showed positive immu- cleotide probe (Table 2), resulting in selection of the p43L3 nofluorescence with mAbs 9A5D12, 12B3G4, and C-5 (Fig. clone. The HHV-7 sequences in Figs. 4 and 5 were deter- 2), which recognize early-late antigens of HHV-6 (10, 24). mined on the 3L25 clone and correspond to the complete Occasionally, the mAb 9A5D12 immunofluorescence signal insert of the p43L3 clone. PCR amplification ofHHV-7 DNA in infected CBMC was negative, whereas it was strong in sequences, using primers HV10 and HVii (Table 2), was HHV-7(JI)-infected SUP-T1 cells. done by using the same protocol as described for primer pair Seroepidemiology of HIHV-7. Immunofluorescence studies HV7/HV8, with annealing at 55°C instead of 60°C. with human sera revealed that 14 of 17 healthy blood donors Comparison and alignment of DNA and protein sequences had IgG antibodies against HHV-7. In a blind seroepidemi- were done by using the CLUSTAL and PALIGN programs in the ological survey comparing 30 CFS patients and 17 age- and PC/GENE software (IntelliGenetics). sex-matched healthy adult controls, geometric mean IgG Downloaded by guest on October 2, 2021 10554 Medical Sciences: Berneman et al. Proc. Nad. Acad Sci. USA 89 (1992) ZVH14 ZVB70 ZVB10 HHV-7 HHV-6 HHV-7 HHV-6 HHV-7 HHV-6 BL GS Z29 Ml1 C02Ok 410 j CBL GSE 29C01 CO2 CK *; .J CBL GS Z29 C1 C02 a A I WC E Kb 23 - 94- 66- 4 4 6-

kb D 23 - 94- 66- 44- 3 =

p43L3 pH6Z-1O1 HHV-7 HHV-6 HHV-6 HHV-71 C1o C05 Z S9S$wT1 j CXL _T CBL'GS*4w-129= OM OIK G* GS C02 U1 1O DoM OK iO -W*-at-I H 23 23.1- ,.:- 44- . ^ 6.6- 4 4- w1 4 Bid=~~~~~~~~~4* FIG. 1. Double immunofluorescence of HHV-7-infected CBMC. I 5 (A) Phase contrast: one large cell (left) and several smaller ones. (B) .~~ CD2 membrane immunofluorescence: all cells are T lymphocytes. FIG. 3. Hybridization of the same Southern blot (A-G) with the (C) Intracellular immunofluorescence with patient's serum: the large inserts of HHV-6 probes pZVH14, pZVB70, and pZVB10 and with cell at left and the smaller one at right are infected with HHV-7(JI). the HHV-7 probe p43L3 and hybridization of another.. Southern blot Live infected CBMC were incubated with murine mAbs directed (H) with the HHV-6 probe pH6Z-101. (A-G) Five micrograms of against CD1, CD2, CD3, CD4, CD8, CD11b, CD36, CD7, and CD19, DNA in each lane (CBL, uninfected CBMC DNA). (H) The seven as well as with mouse IgG as a control; the cells were then treated lanes on left contain 0.33 ,ug; the other lanes contain 4 ,ug of DNA. with F(ab')2 fragments of a rhodamine-conjugated goat anti-mouse DA/H and DA/M, HSB-2 and MOLT-3 cells, respectively, infected IgG antiserum and centrifuged on a glass slide in a Cytospin with HHV-6 strain DA; SUP-T1 and CBL, uninfected SUP-T1 cells cytocentrifuge (Shandon, Pittsburgh). The slides were immersed for and CBMC, respectively; and JI, HHV-7 strain JI propagated in 10 min in acetone at -200C, dried, and stained with a 1/20 dilution SUP-T1. HindIII-digested DNA was electrophoresed in an 0.8% of patient's serum and then treated with a fluorescein-conjugated agarose gel and transferred to a nylon membrane by Southern goat anti-human IgG antiserum. blotting. Prehybridization of nylon filters for at least 2 hr at 3rC in Hybrisol I solution (Oncor, Gaithersburg, MD) was followed by titers for patients vs. controls were 87.1 ± 3.9 vs. 38.9 ± 5.5 addition ofthe DNA probe at 106 cpm/ml that was 32P-labeled by nick (P = 0.078, two-tailed t test). translation. Hybridization was overnight at 37C. Filters were Southern Blot Analysis. Although all HHV-6 probes hy- washed with 6x standard saline/citrate (SSC)/0.5% SDS twice for 15 bridized to DNA ofall HHV-6 strains tested so far, only three min at 60TC and autoradiographed. Afterward the filters were washed and to with increased stringency, generally ending with 0.1x SSC/0.5% of them (ZVB70, ZVB10, pH6Z-101) hybridized SDS. Washing conditions were as follows: probes ZVH14 (A and B), HHV-7 DNA (Table 1, Fig. 3). Because the ZVB70 probe p43L3 (G), and pH6Z-101 (H), 6x SSC/0.5% SDS at 60TC; probes contains the GGGTTA tandem repeat also present in ZVB70 (C and D) and ZVB10 (E and F), 0.1x SSC/0.5% SDS at Marek's disease herpesvirus (19), we examined whether that 60TC. Exposure time was as follows: A, C, and E, 4 hr; B and F, 3 sequence is also present in HHV-7. Hybridization with the days; D and G, 2 days; and H, overnight. HV1 oligonucleotide probe (Table 2) showed that this was so and that the ZVB70 hybridization bands coincided with the (18). The ZVB16 sequence (L. Rosenthal, personal commu- (GGGTTA)" ones. The HHV-6 probes used in this study scan nication) and 2 kb of the ZVB70 clone are represented twice 100.0 kilobases (kb)-i.e., =58.8% of the HHV-6 genome in the HHV-6 genome, in the terminal direct repeats (25), and were counted twice. Ofthe 58.8% ofthe HHV-6 genome thus scanned, at most 37.4% showed a strong homology to HHV-7. PCR Amplification, Cloning, and Sequencing of HHV-7 DNA. Because the hybridization studies showed that HHV- 7(JI) was related to HHV-6, we tried to amplify DNA from HHV-7(JI), using primers designed after comparing sequences from HHV-6(GS) and HCMV. In a study of the ZVB70 clone (unpublished results), we encountered a homology between the putative HCMV UL31 protein (26, 27) and a putative HHV-6 protein, including two short, separate stretches with an amino acid identity (Fig. 4). Based on the hypothesis that this identity would also be conserved in HHV-7 DNA, we synthesized degenerate oligonucleotide primers covering the two regions of amino acid identity (Fig. 4, Table 2) and amplified, cloned, and sequenced an HHV-7 DNA 5). Sequence alignment showed a nu- FIG. 2. Immunofluorescence staining of a 12-week-old HHV-7- fragment (Fig. infected SUP-T1 culture with HHV-6 mAbs. (a) Negative control cleotide homology of 57.5% between HHV-7 and HHV-6, of (treatment with the second, fluorescein-conjugated goat anti-mouse 36% between HHV-7 and HCMV, and of 39.8% between IgG antiserum only and counterstained with Evans blue). (b) mAb HHV-6 and HCMV (Fig. 5). At the amino acid level there is, 9A5D12 (p41). (c) mAb 12B3G4 (p135). (d) mAb C-5 (p38/41). respectively, an identity and a similarity of 51.6% and 19.4% Downloaded by guest on October 2, 2021 Medical Sciences: Berneman et al. Proc. Natl. Acad. Sci. USA 89 (1992) 10555

HHV-7 (43L3) - YPSCFHIVTLPIQFSSRNDRQML - 23 HHV-6 HHV-6 HHV-7 HHV-6 (ZVB70) - APYDIHIYPSRCHIVILPIRYFTKPDKQIL - 30 HCMV (UL31) - APYDIXFGVQPRQTVELDLRYVQITDRCFL - 30 HSV-1 VZV GS C02 C 5 U1102DA/M BA OK SF M *..** . .. *. . * D M Jl JICBL M M KF SF I HHV-7 (43L3) - VSSYPNEGYFEVQMCPWVQNSPLQIVIKSF - 53 EBVHCM8COO C031 I29 HHV-6 (ZVB70) - ISGYQNEGFFETQVMLWAPGTPLHITLRSF - 60 bp HCMV (UL31) - VANLPHEDAFYTGLSVWRGGEPLKVTLWTR - 60 1000- . . - *.** * * * * * .** 500- HHV-7 (43L3) - SKNLVLPQG - 62 200- HHV-6 (ZVB70) - SPNLILPQSTPIATLF - 76 100- HCMV (UL31) - TRSIVIPQGTPIATLY - 76 50-

FIG. 4. Alignment of homologous putative protein sequences of HHV-7, HHV-6, and HCMV. Name of original sequence or clone is in parentheses. Amino acid identity between the three viruses is indicated by a star; similarity is indicated by a dot. Identical parts of HHV-6 and HCMV, printed in boldface type, served to design the degenerate primers used to amplify and clone the HHV-7 sequence (see Table 2). FIG. 6. PCR amplification of HHV-7 DNA sequences, using primers HV10 and HV11 (Table 2). PCR amplification was done by between HHV-7 and HHV-6, of 25.8% and 17.7% between using the same protocol as described for primer pair HV7/HV8, with HHV-7 and HCMV, and of 27.4% and 19.4% between annealing at 55C instead of 60°C. (Upper) Ethidium bromide strain HHV-6 and HCMV (Fig. 4). of the _6% polyacrylamide gels. (Lower) Hybridization with probe Molecular Diagnostic Reagents Specific for HHV-7. The HV12 (Table 2) of the PCR products transferred by electroblotting p43L3 plasmid probe, containing the amplified HHV-7 DNA from these gels to nylon filters (washing conditions: 6x SSC/0.5% SDS at 60°C, twice for 10 mmn; autoradiography after 4 hr at -80°C). fragment without the region covered by the degenerate M, DNA Mr standards (GelMarker I, Research Genetics, Huntsville, primers, only hybridized to HHV-7 DNA (Fig. 3G, arrow) AL); DA/H and DA/M, HHV-6(DA) propagated in the HSB-2 and and not to any other human herpesvirus DNA tested (herpes MOLT-3 cell lines, respectively; and CBL, cord blood lymphocytes. simplex virus 1, Epstein-Barr virus, varicella-zoster virus, HCMV), including the 12 HHV-6 strains mentioned in this DNA was amplified, and DNA from the other human herpes- study. Based on the HHV-7 sequence, oligonucleotide prim- viruses tested was not amplified (Fig. 6). Initially DNA from ers and probes were synthesized (Table 2) for PCR analysis. two HHV-6(SF) cultures was amplified with the HHV-7- With primer pairs HV7/HV8 or HV10/HV11, only HHV-7 specific primers. The first of those two cultures was propagated in adult PBMC, the supernatant from which passed the HHV-7 - TATCCCAGCTGTTTTCATATAGTAACATTACCAATTCAGT -40 virus to a CBMC culture. However, when we infected other HHV-6 - TATCCAAGCCGATGCCACATAGTAATTCTACCTATCCGAT -40 CBMC with another HHV-6(SF) virus stock and used the supernatant ofthat infected culture to infect a second CBMC HCMV - GGGGTGCAACCGCGGCAGACGGTGGAGTTGGACTTGCGCT -40 culture, no HHV-7-specific DNA could be amplified from HHV-7 - TATCCCAGCTGTTTTCATATAGTAACATTACCAATTCAGT -40 either culture (Fig. 6, SF lanes), strongly suggesting that the HHV-7-specific amplification in the other cultures was due to HHV-7 - TTTCATCCAGAAATGATAGACAGATGTTGGTGTCAAGCTA -80 HHV-7 in propagation of ii liii i i I ll the adult PBMC that were used for II III, II I I I Bll HHV-6(SF). HHV-6 - ATTTCACAAAGCCCGATAAGCAAATCCTCATATCCGGCTAACTCGTACGCGTT TGiGCAT -80 I II I II I II HCOV - ACvNTGCAGAF~TCACAG ~eACGTGTTTCPeTTGGTGGCCAAC~TT -80 IA G T111111 IA I 111111 DISCUSSION HHV-7 - TTTCATCCAGAAATGATAGACAGATGTTGGTGTCAAGCTA -80 HHV-6 and HHV-7, two recently discovered viruses, are

HHV-7 - -120 T-lymphotropic human herpesviruses. The initial HHV- 7(RK) isolate was cultured from activated CD4+ T lympho- HHV-6 - CCAGAACGAAGGATTCTTCGAAACCCAGGTGATGCTGTGG -120 cytes of a healthy adult (7). In this report, we show that l l HCOV - GCCACACGAGGACGCCTTTTACACGGGGCTCAGCGTGTGG -120 CBMC infected in vitro were >90%o positive for the pan-T l Ill

HHV-7 - TCCTAATGAAGGCTACTTTGAAGTACAAATGTGCCCATGG -120 markers CD2, CD3, and CD7. Infected cells were primarily, but not exclusively, CD4+; some infected CD8+ cells were also seen. This result is reminiscent of the situation with HHV-7 - GTACAGAATTCTCCTCTTCAAATTGTTATTAAATCTTTTT -160 Ill HHV-6, which can infect CD4+, as well as CD8+, cells (28). HHV-6 - GCCCCGGGAACACCTTTGCACATTACGCTGCGTTCATTTT -160 In addition, the only cell line that we have been able to infect

HCOV - CGCGGCGGCGAGCCGCTCAAAGTCACGCTGTGGACGCGCA -160 with HHV-7(JI) is the immature T-cell line SUP-T1. Although HHV-7(JI) was isolated from a patient with CFS, HHV-7 - GTACAGAATTCTCCTCTTCAAATTGTTATTAAATCTTTTT -160 no statistically significant difference emerged between the IgG titers of CFS patients and healthy controls. At present, HHV-7 - CAAAAAATCTAGTGCTACCGCAAGGC -186 there is no clear association of elevated HHV-7 IgG levels HHV-6 - CTCCAAATCTGATCCTGCCTCAAAGC -186 and CFS. This result does not exclude the association of HHV-7 and some cases of CFS, but serial serum sample HCOV - CGCGTTCCATCGTGATCCCGCAGGGC -186 analysis would be required to demonstrate this association. HHV-7 - CAAAAAATCTAGTGCTACCGCAAGGC -186 Most healthy adults have IgG antibodies against HHV-7 in their serum, demonstrating prior infection; this result is FIG. 5. Alignment of homologous DNA sequences of HHV-7, HHV-6, and HCMV. Identity is indicated by vertical hatched bars. consistent with another recent report showing seropositivity The HHV-7 sequence does not include the region covered by in >90% of healthy adults (28). The association of HHV-7 degenerate primers. Origin of sequences is the same as indicated in with disease remains to be determined. Fig. 4. The HHV-7 sequences above and in Fig. 4 were determined Molecularly, immunologically, and biologically, HHV-7 is on the 3L25 clone and correspond to the complete insert ofthe p43L3 related to HHV-6; yet the two viruses differ. Although all clone. HHV-6 probes used hybridized to DNA of all HHV-6 strains Downloaded by guest on October 2, 2021 10556 Medical Sciences: Berneman et al. Proc. Nad. Acad. Sci. USA 89 (1992) tested so far, only some of them hybridized to HHV-7 DNA Katsafanas, G., Kramarsky, B. & Brus, I. (1992) J. Irffect. Dis. (ref. 7, this study), showing that the molecular divergence 166, 690-691. between HHV-6 and HHV-7 is greater than that between 9. Balachandran, N., Amelse, R. E., Zhou, W. W. & Chang, different HHV-6 strains and strongly suggesting that HHV-7 C. K. (1989) J. Virol. 63, 2835-2840. is divergent enough from HHV-6 to be considered another 10. Iyengar, S., Levine, P. H., Ablashi, D., Neequaye, J. & This con- Pearson, G. R. (1991) Int. J. Cancer 49, 551-557. herpesvirus rather than a "subtype" of HHV-6. 11. Krueger, G. R. F., Sander, C., Hoffmann, A., Barth, A., clusion is further strengthened by the sequence of a 186-bp Koch, B. & Braun, M. (1991) In Vivo 5, 217-226. stretch ofHHV-7 DNA, which was only 57.5% identical with 12. Kikuta, H., Lu, H., Matsumoto, S., Josephs, S. F. & Gallo, HHV-6 DNA. This difference between HHV-7 and HHV-6 R. C. (1989) J. Infect. Dis. 160, 550-551. DNAs allowed us to develop specific molecular reagents for 13. Carrigan, D. R., Knox, K. K. & Tapper, M. A. (1990) J. Infect. the diagnosis of HHV-7, both at the level of Southern blot Dis. 162, 844-851. analysis and PCR amplification. It is possible that other 14. Levy, J. A., Ferro, F., Lennette, E. T., Oshiro, L. & Poulin, regions of the HHV-7 genome have a higher level of homol- L. (1990) Virology 178, 113-121. ogy with HHV-6, which could account for the recognition of 15. Ablashi, D. V., Josephs, S. F., Buchbinder, A., Hellman, K., certain HHV-7 epitopes by mAbs raised against HHV-6 (ref. Nakamura, S., Llana, T., Lusso, P., Kaplan, M., Dahlberg, J., caveat for the molecular diagnosis of Memon, S., Imam, F., Ablashi, K. L., Markham, P. D., Kra- 29 and this study). One marsky, B., Krueger, G. R. F., Biberfeld, P., Wong-Staal, F., HHV-7 is the potential presence ofboth HHV-6 and HHV-7, Salahuddin, S. Z. & Gallo, R. C. (1988) J. Virol. Methods 21, especially when one of these viruses is propagated in adult 29-48. PBMC. This combination is probably due to the reactivation 16. Josephs, S. F., Salahuddin, S. Z., Ablashi, D. V., Schachter, of latent HHV-6 or HHV-7 in cultured adult PBMC (7). F., Wong-Staal, F. & Gallo, R. C. (1986) Science 234, 601-603. Therefore, it is advisable to avoid propagating HHV-6 or 17. Josephs, S. F., Ablashi, D. V., Salahuddin, S. Z., Jagodzinski, HHV-7 in adult PBMC. L. L., Wong-Staal, F. & Gallo, R. C. (1991) J. Virol. 65, Although both HHV-6 and HHV-7 are T-lymphotropic 5597-5604. viruses, sequence data show that these viruses do not belong 18. Josephs, S. F., Ablashi, D. V., Salahuddin, S. Z., Kramarsky, virus B., Franza, B. R., Jr., Pellett, P., Buchbinder, A., Memon, S., to the y-herpesvirus group that includes Epstein-Barr Wong-Staal, F. & Gallo, R. C. (1988) J. Virol. Methods 21, (30). Instead, the clear sequence homology with HCMV (refs. 179-190. 17, 20, 21, and this study) strongly supports the classification 19. Kishi, M., Harada, H., Takahashi, M., Tanaka, A., Hayashi, of HHV-6 and HHV-7 as members of the ,B-herpesviruses. M., Nonoyama, M., Josephs, S. F., Buchbinder, A., Schach- ter, F., Ablashi, D. V., Wong-Staal, F., Salahuddin, S. Z. & We thank Drs. N. Balachandran, J. Luka, and G. Pearson for their Gallo, R. C. (1988) J. Virol. 62, 4824-4827. gifts of antibodies; Dr. P. Pellett and the late Dr. R. Honess for the 20. Efstathiou, S., Gompels, U. A., Craxton, M. A., Honess, probes generated in their laboratory; Drs. R. Downing, G. Krueger, R. W. & Ward, K. (1988) Lancet 1, 63-64. P. Pellett, H. Kikuta, D. Carrigan, and J. Levy for their HHV-6 21. Lawrence, G. L., Chee, M., Craxton, M. A., Gompels, U. A., strains; and Dr. L. Rosenthal (Georgetown University, Washington) Honess, R. W. & Barrell, B. G. (1990) J. Virol. 64, 287-299. for allowing us to refer to his unpublished observations regarding the 22. Berneman, Z. N., Gartenhaus, R. B., Reitz, M. S., Blattner, pZVB16 clone. We thank Dr. M. E. Klotman for constructive W. A., Manns, A., Hanchard, B., Ikehara, O., Gallo, R. C. & comments on the manuscript. National Institutes of Health Grants Klotman, M. E. (1992) Proc. Natl. Acad. Sci. USA 89, 3005- R01AI26788, R01AI27314, U01AI32246 to A.L.K. supported the 3009. research. 23. Smith, S. D., Shatsky, M., Cohen, P. S., Warnke, R., Link, M. P. & Glader, B. E. (1984) Cancer Res. 44, 5657-5660. 1. Salahuddin, S. Z., Ablashi, D. V., Markham, P. D., Josephs, 24. Balachandran, N. (1992) in Human Herpesvirus 6: Epidemiol- S. F., Sturzenegger, S., Kaplan, M., Halligan, G., Biberfeld, ogy, Molecular Biology and Clinical Pathology, eds. Ablashi, P., Wong-Staal, F., Kramarsky, B. & Gallo, R. C. (1986) D. V., Krueger, G. R. F. & Salahuddin, S. Z. (Elsevier Sci- Science 234, 596-601. ence, Amsterdam), pp. 97-120. 2. Downing, R. G., Sewankambo, N., Serwadda, D., Honess, R., 25. Martin, M. E. D., Thomson, B. J., Honess, R. W., Craxton, Crawford, D., Jarrett, R. & Griffin, B. E. (1987) Lancet li, 390. M. A., Gompels, U. A., Liu, M. Y., Littler, E., Arrand, J. R., 3. Tedder, R. S., Briggs, M., Carmon, C. H., Honess, R., Rob- Teo, I. & Jones, M. D. (1991) J. Gen. Virol. 72, 157-168. ertson, D. & Whittle, H. (1987) Lancet Hi, 390-392. 26. Jahn, G., Kouzarides, T., Mach, M., Scholl, B.-C., Plachter, 4. Lopez, C., Pellett, P., Stewart, J., Goldsmith, C., Sanderlin, B., Traupe, B., Preddie, E., Satchwell, S. C., Fleckenstein, B. K., Black, J., Warfield, D. & Feorino, P. (1988) J. Infect. Dis. & Barrell, B. G. (1987) J. Virol. 61, 1358-1367. 157, 1271-1273. 27. Chee, M. S., Bankier, A. T., Beck, S., Bohni, R., Brown, 5. Lusso, P., Markham, P. D., Tschachler, E., di Marzo- C. M., Cerny, R., Horsnell, T., Hutchison, C. A., III, Veronese, F., Salahuddin, S. Z., Ablashi, D. V., Pahwa, S., Kouzarides, T., Martignetti, J. A., Preddie, E., Satchwell, Krohn, K. & Gallo, R. C. (1987) J. Exp. Med. 167, 1659-1670. S. C., Tomlinson, P., Weston, K. M. & Barrell, B. G. (1990) 6. Takahashi, K., Sonoda, S., Higashi, K., Kondo, T., Takahashi, Curr. Top. Microbiol. Immunol. 154,125-169. H., Takahashi, M. & Yamanishi, K. (1989) J. Virol. 63, 28. Lusso, P., Malnati, M., DeMaria, A., Balotta, C., DeRocco, 3161-3163. S. E., Markham, P. D. & Gallo, R. C. (1991) J. Immunol. 147, 7. Frenkel, N., Schirmer, E. C., Wyatt, L. S., Katsafanas, G., 685-691. Roffman, E., Danovich, R. M. & June, C. H. (1990) Proc. Nati. 29. Wyatt, L. S., Rodriguez, W. J., Balachandran, N. & Frenkel, Acad. Sci. USA 87, 748-752. N. (1991) J. Virol. 65, 6260-6265. 8. Berneman, Z. N., Gallo, R. C., Ablashi, D. V., Frenkel, N., 30. Chee, M. S. & Barrell, B. (1990) Trends Genet. 6, 86-91. Downloaded by guest on October 2, 2021