- Email: [email protected]
SEN virus infection and its relationship to transfusion-associated hepatitis
SEN Virus Infection and Its Relationship to Transfusion-Associated Hepatitis TAKEJI UMEMURA,1 ANTHONY E. T. YEO,1 ALESSANDRA SOTTINI,2 DANIELE MORATTO,2 YASUHITO TANAKA,1 RICHARD Y.-H. WANG,1 J. WAI-KUO SHIH,1 PETER DONAHUE,3 DANIELE PRIMI,2 AND HARVEY J. ALTER1
SEE EDITORIAL ON PAGE 1334
SEN virus (SEN-V) is a recently identified singlestranded, circular DNA virus. Two SEN-V variants (SENV-D and SENV-H) were assayed by polymerase chain reaction (PCR) to investigate their role in the causation of transfusion-associated non–A to E hepatitis. The incidence of SEN-V infection after transfusion was 30% (86 of 286) compared with 3% (3 of 97) among nontransfused controls (P < .001). Transfusion risk increased with the number of units transfused (P < .0001) and donor-recipient linkage for SEN-V was shown by sequence homology. The prevalence of SEN-V in 436 volunteer donors was 1.8%. Among patients with transfusion-associated non–A to E hepatitis, 11 of 12 (92%) were infected with SEN-V at the time of transfusion compared with 55 of 225 (24%) identically followed recipients who did not develop hepatitis (P < .001). No effect of SEN-V on the severity or persistence of coexistent hepatitis C virus (HCV) infection was observed. In 31 infected recipients, SEN-V persisted for greater than 1 year in 45% and for up to 12 years in 13%. SEN-V–specific RNA (a possible replicative intermediate) was recovered from liver tissue. In summary, SENV-D and -H were present in nearly 2% of US donors, and were unequivocally transmitted by transfusion and frequently persisted. The strong association of SEN-V with transfusion-associated non–A to E hepatitis compared with controls raises the possibility, but does not establish that SEN-V might be a causative agent of posttransfusion hepatitis. The vast majority of SEN-V–infected recipients did not develop hepatitis. (HEPATOLOGY 2001;33:1303-1311.) After cloning of the hepatitis C virus (HCV)1 and development of sensitive serologic and molecular assays for this pathogen, a dramatic decline in the incidence of transfusionAbbreviations: HCV, hepatitis C virus; HGV, hepatitis G virus; GBV-C, GB virus type C; TTV, TT virus; SEN-V, SEN virus; PCR, polymerase chain reaction; cDNA, complementary DNA; ALT, alanine aminotransferase; CI, confidence interval. From the 1The Department of Transfusion Medicine, Warren Grant Magnuson Clinical Center, National Institutes of Health, Bethesda, MD; 2DiaSorin Inc., Brescia, Italy; and 3DiaSorin Inc., Stillwater, MN. Received September 27, 2000; accepted February 27, 2001. The sequences reported in this paper have been deposited in the DDBJ/EMBL/GeneBank data base (accession no. AB049170-AB049183). Address reprint requests to: Harvey J. Alter, M.D., Department of Transfusion Medicine, Building 10, Room 1C711, National Institutes of Health, Bethesda, MD 208921184. E-mail: [email protected]; fax: 301-402-2965. Copyright © 2001 by the American Association for the Study of Liver Diseases. 0270-9139/01/3305-0036$35.00/0 doi:10.1053/jhep.2001.24268
associated hepatitis occurred.2,3 However, approximately 10% of transfusion-associated hepatitis cases4 and 20% of community-acquired hepatitis cases5 do not have a defined etiology suggesting the existence of additional causative agents. Hepatitis G virus (HGV),6 also known as GB virus C (GBV-C),7 was initially suggested as a causative agent of non–A to E hepatitis, but this was not confirmed in further studies.8-10 The originally described TT virus (TTV), discovered in 1997 by representational difference analysis, was detected in 3 of 5 cases of transfusion-associated hepatitis and was proposed as a potential cause of non–A to E hepatitis.11 Using TTV primers identical to those reported, non–A to E hepatitis cases and controls in the NIH prospective series were tested but no association with transfusion-associated hepatitis was found.12 Subsequently, primers to more conserved regions of the TTV genome were used in studies in Japan, and the agent was found in greater than 90% of Japanese blood donors.13 This further diminished the likelihood that TTV was the cause of transfusion-associated hepatitis and precluded the possibility of a practical screening assay. Other studies have also confirmed the high prevalence of TTV and its lack of association with transfusion-associated hepatitis.14,15 Recently, a novel DNA virus designated SEN virus (SEN-V) was discovered in the serum of an intravenous drug abuser also infected with human immunodeficiency virus.16,17 SEN-V was initially described as a single-stranded DNA virus of approximately 3,800 nucleotides.16 Phylogenetic analysis of SEN-V has shown the existence of 8 strains.18 Although structurally similar to TTV, SEN-V has less than 55% sequence homology and less than 37% amino acid homology with the TTV prototype.18 Preliminary studies of SEN-V variants were conducted by Dr. Primi (DiaSorin Inc., Brescia, Italy). The prevalence of 5 SEN-V strains (A, B, H [former C], D, and E) and a consensus sequence designated total SEN-V were measured in various donor and patient populations. It was shown that measuring total SEN-V was not practical because the prevalence in donors was 13% and the rate in all transfused populations exceeded 70%. SENV-B was also excluded as a useful screening assay because it was present in 10% of donors and only 8% of patients with transfusion-associated non–A to E hepatitis. SENV-A and SENV-E were found in low prevalence in donors, but did not appear to be associated with non–A to E hepatitis. In contrast, SENV-D and SENV-H had favorable prevalence ratios being found in less than 1% of donors and more than 50% of transfusion-associated non–A to E cases. These initial polymerase chain reaction (PCR) data were confirmed by cloning and sequencing which showed that SENV-D and SENV-H sequences were found in a high proportion of non–A to E hepatitis cases. This prelimi-
1303
1304 UMEMURA ET AL.
nary effort served as the basis for performing a systematic search for SENV-D and H in blood recipients who did and did not develop transfusion-associated hepatitis and in nontransfused controls. This systematic investigation is the foundation of the current study. PATIENTS AND METHODS Selection of Patients and Donors. Stored serum samples from donors and recipients were obtained from subjects enrolled in prospective studies of transfusion-associated hepatitis at the NIH conducted from October 1972 to December 1997. The methods of enrollment, frequency, and duration of patient follow-up and the criteria for the diagnosis of hepatitis have been described previously.3 The study protocols were reviewed and approved by the appropriate institutional review boards, and all study subjects gave written informed consent. SEN-V positivity was determined in 13 patients who developed transfusion-associated non–A to E hepatitis after open-heart surgery, 49 patients who developed transfusion-associated hepatitis C, 232 transfused patients who did not develop hepatitis, and 100 patients who had heart surgery but did not require transfusion or develop hepatitis. For patients identified as having transfusion-associated SEN-V infection, all available linked donor samples were tested for SEN-V DNA (n ⫽ 213). The prevalence of SEN-V was determined in 436 current volunteer blood donors and 387 randomly selected donors whose samples were obtained prior to 1990. Samples were stored at ⫺70°C before testing for SEN-V DNA. Detection of SEN-V DNA and HCV RNA by the PCR. Nucleic acids were extracted from 100 L of serum by QIAamp DNA Blood Mini Kit (Qiagen, Valencia, CA). The extracted DNA was eluted in 160 L buffer containing 10 mmol/L Tris-HCl, 0.5 mmol/L ethylenediaminetetraacetic acid, pH 9.0. SENV-D DNA and SENV-H DNA were determined by PCR. For SENV-D detection, primer D10S (sense primer 5⬘-GTAACTTTGCGGTCAACTGCC-3⬘) and primer L2AS (antisense primer, 5⬘-CCTCGGTTKSAAAKGTYTGATAGT-3⬘ [K ⫽ G or T, S ⫽ C or G, Y ⫽ C or T]) were used in a 40-L PCR mixture containing PCR buffer, 0.25 mmol/L dNTPs, 2.5 mmol/L magnesium chloride, 1.75 U AmpliTaq DNA Polymerase (Perkin-Elmer Applied Biosystems, Foster City, CA), and 10 L of extracted DNA. PCR involved 40 cycles (94°C for 1 minute, 58°C for 1 minute, and 72°C for 1 minute for each cycle) followed by the extension reaction at 72°C for 9 minutes in a Perkin Elmer 9700 thermal cycler. For SENV-H detection, primer C5S (sense primer 5⬘-GGTGCCCCTWGTYAGTTGGCGGTT-3⬘ [W ⫽ A or T]) and the same antisense primer L2AS were used. Forty cycles (94°C for 1 minute, 62°C for 1 minute, and 72°C for 1 minute for each cycle) were again performed, followed by the extension reaction at 72°C for 9 minutes. PCR products were analyzed by DNA enzyme immunoassay (DiaSorin, Saluggia, Italy) with SENV-D (5⬘-ATGATAGGCTTCCCYTTTAACTATAACCCA-3⬘) or SENV-H (5⬘-CCCCTTCCAGGTATTGCATGAAGAGTATTAC-3⬘) specific 5⬘-end biotinylated probes. Specimens with optical density (OD) values ⱖ0.350 were considered positive for SENV-D or SENV-H. The estimated lowest detection level for SEN-V DNA is 10 copies per each test based on a dilution series of plasmid DNA containing an SEN-V PCR target insert. This was equivalent to 1,600 copies/mL in the original serum being extracted. All positive samples were confirmed by duplicate retesting. Procedures for HCV RNA PCR were performed as described elsewhere.19 Quantitation of SEN-V DNA. SEN-V DNA was quantitated with external standards containing known concentrations of SEN-V DNA. These standards were constructed by inserting SEN-V–positive PCR products into TOPO TA cloning vector pCR 2.1 (Invitrogen, Carlsbad, CA). Digoxigenin labeling mix (200 mol/L of dATP, dCTP, and dGTP each, 190 mol/L dTTP, and 10 mol/L digoxigenin-11dUTP) was used in the PCR reaction mixture. Streptavidin-coated black enzyme-linked immunosorbent assay plates (Roche Molecular Biochemicals, Indianapolis, IN) were incubated overnight at 2°C to
HEPATOLOGY May 2001
8°C with biotinylated SENV-D or SENV-H probes. The plates were washed and 90 L of hybridization solution and 10 L of PCR amplified products denatured at 95°C for 5 minutes were added. The hybridization was performed at 50°C for 1 hour for SENV-D or at 45°C for 1 hour for SENV-H. The enzyme-linked immunosorbent assay plates were then washed again and anti-digoxigenin antibodies conjugated to alkaline phosphatase were added. The mixture was incubated for 30 minutes at 37°C. Reactivity was detected using 0.25 mmol/L disodium 3-(4-methoxyspiro {1, 2-dioxetane-3, 2⬘-(5⬘chloro) tricyclo [3.3.1.13,7] decan}-4-yl) phenyl phosphate as substrate in 0.1 mol/L Tris-HCl, 0.1 mol/L NaCl, pH 9.5. The reaction was incubated at 37°C for 15 minutes. Chemiluminescence reading was obtained by using a 1450 Micro Beta PLUS scintillation counter (Wallac Inc, Turku, Finland). The total chemiluminescence reading of each standard was plotted against the SEN-V DNA copy number to establish a standard curve; the counts of chemiluminescence and the copy number correlated linearly from 5 to 10,000 copies per reaction (P ⬍ .001; Pearson’s correlation coefficient, r ⫽ .79). The number of viral particles was determined in relation to the standard curve and expressed as copies per milliliter. Sequence and Phylogenetic Analysis of SEN-V DNA. DNA extracted from the sera of 1 SENV-D– and 1 SENV-H–positive patient with transfusion-associated non–A to E hepatitis and 3 implicated donors (1 SENV-D and 2 SENV-H) was amplified using SEN-V–specific primers. All samples produced bands visible on agarose gel. To determine the nature of the differently sized bands, amplicons containing poly A tails were excised from the agarose gel and the DNA content was purified and ligated to a TOPO TA cloning vector pCR 2.1. Using DNA extracted from transformed Escherichia coli, both strands were sequenced with AutoRead sequencing Kit (Amersham Pharmacia Biotech, Piscataway, NJ) using fluorescent M13 universal primer and fluorescent M13 reverse primer. Sequences were determined on 4 to 6 clones each for respective PCR products. The sequences, excluding primers sequences (D, 179 bp; H, 178 bp), were aligned with clustal W (version 1.8). Using the computer program ODEN version 1.1.1,20 the number of nucleotide substitutions per site (genetic distance) between the isolates was estimated by the 6-parameter method.21 Based on these values, a phylogenetic tree was constructed by the neighbor-joining method.22 Detection of SEN-V in Liver Tissue. To determine the presence of SEN-V replication in the liver, assays were conducted on liver tissue surgically extracted from 2 SEN-V–positive patients who had hepatocellular carcinoma. RNA was extracted from 25 mg of liver tissue using the S.N.A.P. Total RNA isolation kit (Invitrogen) with slight modification to the manufacturer’s instructions. The modification consisted of incubating eluted nucleic acids with RNase-free DNase at 37°C for 30 minutes instead of the recommended 10 minutes. The eluted RNAs (125 L) were precipitated with 2 volumes of ethanol and 1/10th volume of sodium acetate pH 5.2, 3 mol/L for 1 hour at ⫺20°C. The pelleted RNA was washed with 500 L of 75% ethanol, centrifuged, and resuspended in 25 L of RNase-free water. Ten L of RNA were retrotranscribed to single-stranded DNA using the SUPERSCRIPT II RNase H⫺ reverse transcriptase (Gibco BRL, Rockville, MD) and 500 ng of primer L3AS (5⬘-GCCTATAAAAATGTKASWGWACCAKCC-3⬘) in a final volume of 20 L. Five microliters of complementary DNA (cDNA) and 2.5 L of RNA (as control) were amplified using primers NEW BCD 1S (sense primer 5⬘CCYAARCTMTTTGAAGACMA-3⬘ [M ⫽ A or C, R ⫽ A or G]) in combination with primer L2AS, respectively, derived from conserved sequences among SENV-D and SENV-H. The PCR reactions were performed with a 50-L mixture containing template, 5 L of cDNA, PCR buffer, 300 ng of a single PCR primer, a 200 mol/L concentration of each dNTPs, and 1.25 U of Taq polymerase (PerkinElmer Applied Biosystems, Foster City, CA). The reaction was performed in a DNA thermal cycler with mineral oil. PCR consisted of preheating at 94°C for 5 minutes, 40 cycles of 94°C for 1 minute, 54°C for 1 minute, and 72°C for 1 minute, with a terminal extension incubation at 72°C for 7 minutes. To verify the specificity of the amplified products, aliquots of 5 L of each amplified product were
HEPATOLOGY Vol. 33, No. 5, 2001
UMEMURA ET AL.
1305
TABLE 1. Prevalence of SEN-V DNA in Volunteer Donors and Patients SEN-V DNA Positivity Group
Total n
n (%)
(95% CI)
Volunteer donors (current) Volunteer donors (pre-1990) Patients*
436 387 394
8 (1.8%)† 9 (2.3%)‡ 11 (2.8%)§
(0.6-3.1) (0.8-3.8) (1.2-4.4)
*Sampled before surgery/transfusion. † vs. ‡: P ⫽ .80; † vs. §: P ⫽ .49; ‡ vs. §: P ⫽ .85.
hybridized in a DNA enzyme immunoassay with the probe (5⬘-TCCGTTCTGCTCACCACAAACGTAGACAACCC-3⬘) specific for a conserved sequence within the amplified DNA. Serological and Biochemical Assays. Antibodies to HCV were measured by a second-version enzyme immunoassay according to the directions of the manufacturer (Abbott HCV EIA 2.0; Abbott Laboratories, North Chicago, IL). Antibody to hepatitis E virus was measured by the method of Tsarev et al.23 Alanine aminotransferase (ALT) was measured by a 3-point kinetic assay with an automated photometric analyzer (model 917; Hitachi–Boehringer Mannheim, Indianapolis, IN), and the values were expressed in international units (normal range, 5-41 U/L). Statistical Analysis. Student’s t test was used to analyze continuous variables. 2 test with Yates’s correction was used in the analysis of categorical data; when the number of the subjects was less than 5, Fisher’s exact test was used. The Mantel-Haenszel 2 test was used for trend analysis. Pearson’s correlation coefficient was calculated to assess the robustness of the quantitative SEN-V assay. A P value of ⱕ.05 was considered significant. Statistical analyses were performed using SigmaStat (version 2.03; SPSS Inc., Chicago, IL) and SAS 6.12 (SAS Institute, Cary, NC). RESULTS Prevalence of SEN-V in Blood Donors and Patients Prior to Transfusion. SEN-V DNA was detected in 8 of 436 (1.8% [95% con-
fidence interval (CI): 0.6-3.1]) randomly selected recent (1999) blood donors and 9 of 387 (2.3% [95% CI: 0.8-3.8]) randomly selected donors sampled prior to HCV screening (pre-1990). There was no statistically significant difference in donor prevalence between these time frames (P ⫽ .80) (Table 1). The prevalence of SEN-V in a sample obtained from patients before surgery or transfusion was 2.8% (11 of 394 [95% CI: 1.2-4.4]). This was not significantly different from the prevalence observed in recent blood donors (P ⫽ .49) or pre1990 donors (P ⫽ .85) (Table 1). Newly Acquired SEN-V Infections in Relationship to Blood Transfusion. Excluding the 11 patients who were SEN-V positive
before surgery, SEN-V DNA (strain D and/or H) was detected
FIG. 1. The relationship between the prevalence of SEN-V infection and the number of units of blood transfused. A step-wise increase in the risk of infection was observed with increased transfusion volume (P ⬍ .0001).
after surgery in 86 of 286 (30% [95% CI: 25-35]) patients who were transfused compared with 3 of 97 (3% [95% CI: 0.6-9]) patients who were not transfused (P ⬍ .001) (Table 2). The strong association between blood transfusion and SEN-V infection was similar when the data were analyzed for combined SENV-D and SENV-H infection or for SENV-D (P ⫽ .001) or SENV-H (P ⬍ .001) individually (Table 2). The association with blood transfusion was also indicated when SEN-V incidence was stratified according to transfusion volume (Fig. 1); increasing prevalence of SEN-V infections was observed as the transfusion volume increased from 0 to greater than 13 units (P ⬍ .0001). The 86 recipients who developed an acute SEN-V infection after transfusion received a mean of 13.8 (95% CI: 10.1-17.5) units of blood compared with a mean of 9.4 (95% CI: 8.510.3) units transfused to the 200 recipients who were not SEN-V infected (P ⬍ .01). Donor-Recipient Linkage. Samples were available from all donors in only 19 of the 86 (22%) recipients who developed SEN-V infection after transfusion. Those for whom complete donor sampling was available received a mean of 10.0 (95% CI: 6.8-13.2) units of blood compared with a mean of 14.8 (95% CI: 10.2-19.5) units transfused to the 67 SEN-V–positive recipients for whom some or all donor samples were missing (P ⫽ .28). Thirteen of the 19 (68%) SEN-V–infected recipients for whom all donor samples were available had at least 1 SEN-V– positive donor (7 with 1 positive donor, 5 with 2 positive donors, and 1 with 4 positive donors). For the 12 evaluable non–A to E hepatitis cases, 3 had a full complement of donor
TABLE 2. Relationship of Blood Transfusion to New SEN-V Infections Proportion of SEN-V Infections by Strain New SEN-V Infections* Group
Nontransfused Transfused
n
97† 286‡
SENV-D
SENV-H
n
%
(95% CI)
n
%
(95% CI)
n
%
(95% CI)
3 86
3%§ 30%㛳
(0.6-9) (25-35)
1 37
1%¶ 6%#
(0.03-6) (3-8)
2 70
2%** 17%††
(0.3-7) (13-21)
*Sampled presurgery and at 4 to 8 weeks after surgery. †Excludes 3 patients SEN-V DNA⫹ before surgery. ‡Excludes 8 patients SEN-V DNA⫹ before surgery. § vs. 㛳: P ⬍ .001; ¶ vs. #: P ⫽ .001; ** vs. ††: P ⬍ .001. #, ††: 21 patients had combined SENV-D and SENV-H infections and were included in both columns.
1306 UMEMURA ET AL.
HEPATOLOGY May 2001
bored an SENV-H variant identical to 1 strain recovered from the recipient, and the other donor (donor 2) showed 94.4% sequence homology with the second strain in this recipient (Fig. 2). Sequence homology between donor and recipient of greater than 99% for both SENV-D and SENV-H in these 2 cases unequivocally confirms transmission of these agents by blood transfusion. Incidence of SEN-V Infections in Transfused Patients in Relationship to Hepatitis Outcome. Excluding 8 transfused patients with
FIG. 2. Phylogenetic tree analysis comparing SEN-V sequences from a patient with transfusion-associated non–A to E hepatitis and those from the implicated donors. Patient F.G. received blood from 2 SEN-V–infected donors, designated in the figure as donor 1 and donor 2. Cloning and sequencing of serum from patient F.G. revealed that he was infected with 2 diverse strains of SENV-H. Each rectangle in the figure represents a clone. Two patient clones (Pat. FG-1 and FG-2) were identical to 3 clones from donor 1 (Don. 1-4, 1-5, and 1-6) and 94% homologous with 3 other clones from donor 1 (Don. 1-1, 1-2, and 1-3). Patient clones (Pat. FG-3 and FG-4) showed 94% homology with clones from donor 2 (Don. 2-1, 2-2, 2-3, and 2-4). The horizontal bar indicates the number of nucleotide substitutions per site.
samples, 3 had only partial donor sampling, and 6 had no available donor samples. An SEN-V positive donor was identified in 4 of the 6 (67%) evaluable non–A to E hepatitis cases (2 with complete donor samples and 2 with partial samples). In each of the 13 infections for which there was a linked SEN-V positive donor and recipient sample, the SEN-V strain (D or H) in the donor was identical to that in the recipient. However, one patient was acutely infected with SENV-D and SENV-H, but only SENV-D was identified in the implicated donor. Cloning and sequencing was performed to determine the extent of homology between the agents found in the donor and recipient in 2 cases of non–A to E hepatitis. In 1 patient (K.S.), sequencing revealed 99.4% nucleotide homology between the SENV-D strain recovered from the donor and the strain found in the recipient when measured 7 weeks after transfusion (data not shown). A second patient (F.G.) with non–A to E hepatitis was SENV-H positive and received blood from 2 different SENV-H–positive donors. Cloning and sequencing of the recipient revealed 2 distinct SENV-H populations (Fig. 2). One SENV-H–positive donor (donor 1) har-
preexisting SEN-V DNA, newly acquired SEN-V infections were observed during posttransfusion follow-up in 11 of 12 (92%) patients with non–A to E hepatitis compared with 55 of 225 (24%) patients who were transfused, but did not develop hepatitis (P ⬍ .001; Table 3). Of the 8 patients with preexisting SEN-V DNA, 1 developed non–A to E hepatitis after transfusion and 7 showed no biochemical or clinical evidence of hepatitis during follow-up. New SEN-V infections were also observed in 20 of 49 (41%) patients who developed hepatitis C (Table 3); this incidence in patients with hepatitis C, although high, was significantly lower than the 92% incidence of SEN-V infection in patients with non–A to E hepatitis (P ⫽ .005). Of the 11 recipients with SEN-V–positive, non–A to E hepatitis, 2 were infected with SENV-D alone, 7 with SENV-H alone, and 2 with both SENV-D and SENV-H (Table 3). In total, 4 of 12 (33%) patients with non–A to E hepatitis were infected with SENV-D and 9 of 12 (75%) with SENV-H (Table 3). Among the 55 recipients who were SEN-V infected, but did not develop hepatitis, 14 (25%) were infected with SENV-D alone, 29 (53%) with SENV-H alone, and 12 (22%) with both SENV-D and SENV-H. In total, 26 of 225 (12%) patients without hepatitis were infected with SENV-D and 41 of 225 (18%) with SENV-H (Table 3). The frequency of SENV-D infections in patients with non–A to E hepatitis (33%) versus the frequency in recipients without hepatitis (12%) was statistically significant (P ⫽ .05 by Fisher’s exact test). SENV-H was more strongly associated with non–A to E hepatitis being present in 75% of cases compared with 18% of recipients without hepatitis (P ⬍ .001 by Fisher’s exact test; Table 3). Hence, SENV-H was the primary agent associated with non–A to E hepatitis, but 2 patients with non–A to E hepatitis had SENV-D infection alone and 2 patients were coinfected with SENV-H and -D. Relationship Between Viremia and ALT Level. Of the 11 patients with non–A to E hepatitis who were acutely infected with SEN-V, none were icteric and the mean peak ALT level
TABLE 3. Prevalence of New SEN-V Infections in Relation to Transfusion and Hepatitis Outcome
Disease Outcomes
Total n
Non–A to E hepatitis HCV hepatitis No hepatitis
12 49 225
Mean No. Transfusions
Proportion of SEN-V Infections by Strain Total SEN-V Infections*
SENV-D
SENV-H
(95% CI)
No.
(%)
[95% CI]
No.
(%)
[95% CI]
No.
(%)
[95% CI]
11.8 (5.8-17.9) 16.8 (10.5-23.1) 9.3 (8.5-10.1)
11 20 55
(92%)† (41%)‡ (24%)§
[62-100] [27-56] [19-30]
4 7 26
(33%)㛳 (14%)¶ (12%)#
[10-65] [6-27] [7-16]
9 18 41
(75%)** (37%)†† (18%)§§
[43-94] [23-52] [13-23]
*Excludes 11 patients SEN-V DNA⫹ before surgery. † vs. ‡: P ⫽ .005; † vs. §: P ⬍ .001 (Fisher’s exact test); ‡ vs. §: P ⫽ .031. 㛳 vs. ¶: P ⫽ .13 (Fisher’s exact test); 㛳 vs. #: P ⫽ .05 (Fisher’s exact test); ¶ vs. #: P ⫽ .77. ** vs. ††: P ⫽ .039; ** vs. §§: P ⬍ .001 (Fisher’s exact test); †† vs. §§: P ⫽ .008. 㛳, **: 2 patients had combined SENV-D and SENV-H infections. ¶, ††: 5 patients had combined SENV-D and SENV-H infections. #, §§: 12 patients had combined SENV-D and SENV-H infections.
HEPATOLOGY Vol. 33, No. 5, 2001
UMEMURA ET AL.
1307
FIG. 3. SEN-V infection in 10 patients with transfusion-associated non–A to E hepatitis. Levels of ALT (shaded areas), SENV-D DNA (dashed line), and SENV-H DNA (solid line) in 6 patients infected with SENV-H only (patients F.G., R.W., H.S., W.B., J.C., and J.S.), in 2 patients infected with SENV-D only (patients A.B. and W.S.), and in 2 patients infected with both SENV-D and SENV-H (patients K.S. and M.B.) are shown plotted against the time after transfusion. Qualitative PCR results of SEN-V DNA (positive, ⫹; negative, ⫺) are shown above each panel. The horizontal lines indicate the normal ALT level. Patients R.W., W.B., J.S., and A.B. received 14, 32, 3, and 9 units of blood, respectively; no samples from these donations were available for testing. For the remaining patients, some or all of the donor samples were available for testing and the results are depicted as the number of donors/ the number of SEN-V tested/the number of SEN-V⫹: F.G., 12/12/2; H.S., 27/10/0; J.C., 2/2/0; W.S., 4/2/1; K.S., 15/15/1 (the donor was SENVD⫹; the recipient was coinfected with SENV-D and SENV-H); and M.B., 11/10/1 (the donor was SENVH⫹; the recipient was coinfected with SENV-D and SENV-H).
was 396 U/L (95% CI: 111-681). Sufficient serial samples to correlate ALT level with the level of viremia were available in 10 of the 11 patients with non–A to E hepatitis (Fig. 3). In every case, the virus was absent pretransfusion and became detectable before or at the same time as the first ALT elevation
that signaled the onset of hepatitis. There was general concordance between the ALT level and the level of viremia, as seen particularly in cases F.G., R.W., and W.B., but there were many time points in which these levels were discordant, as exemplified in dually infected patients K.S. and M.B.
1308 UMEMURA ET AL.
HEPATOLOGY May 2001
TABLE 4. Relationship of SEN-V Infection to Severity and Persistence of Hepatitis Mean Peak ALT Group
n
(U/L; 95% CI)
Posttransfusion non-A to E hepatitis SEN-V⫹ 11 396 (111-681) SEN-V⫺ 1 200 Posttransfusion HCV hepatitis SEN-V⫹ 20 604 (420-788)§ SEN-V⫺ 29 649 (493-804)㛳
Chronicity* n
(%)
(95% CI)
2 1
(18%)† (100%)‡
(2-52)
18 24
(90%)¶ (83%)#
(68-99) (64-94)
*ALT elevations persisting ⬎1 year. † vs. ‡: P ⫽ .25 (Fisher’s exact test). § vs. 㛳: P ⫽ .60. ¶ vs. #: P ⫽ .69 (Fisher’s exact test).
Association of SEN-V With Severity and Chronicity of Liver Disease. In patients with non–A to E hepatitis, the mean peak
ALT levels for 11 subjects with SEN-V infection was 396 U/L (95% CI: 111-681) and for the 1 subject without SEN-V infection was 200 U/L (Table 4). Patients with HCV had significantly higher mean ALT levels (630 U/L [95% CI: 513-748]) than patients with non–A to E hepatitis (380 U/L [95% CI: 118-642]) (P ⬍ .001). In patients with hepatitis C, the mean peak ALT level for the 20 subjects coinfected with SEN-V was 604 U/L (95% CI: 420-788) compared with 649 U/L (95% CI: 493-804) for the 29 subjects with HCV infection alone (P ⫽ .60). Thus, hepatitis C was more severe than non–A to E hepatitis and the severity in hepatitis C cases was not influenced by coexistent SEN-V infection (Table 4). Chronic hepatitis C (duration ⬎1 year) developed in 18 of 20 (90% [95% CI: 68-99]) patients who were coinfected with HCV and SEN-V compared with 24 of 29 (83% [95% CI: 64-94]) patients who had transfusion-associated HCV without SEN-V infection (Table 4). This difference in the rates of HCV chronicity was not statistically significant (P ⫽ .69). Persistence of SEN-V Infection. Serial samples sufficient to assess the persistence of SEN-V infection were available from 31 patients, including 11 patients with non–A to E hepatitis and 20 patients coinfected with hepatitis C. Clearance of viremia,
was defined as at least 2 consecutive negative PCR determinations with no subsequent recurrence of viremia. All 31 patients were SEN-V DNA negative before transfusion and SEN-V positive 4 to 8 weeks posttransfusion; overall, 24 of 31 recipients (77%) cleared SEN-V DNA during the course of follow-up, 4 (13%) were chronic carriers for from 4 to 12 years (Fig. 4; cases 27, 29, 30, and 31) and in 3 (10%) cases (Fig. 4; cases 18, 19, and 21) there was insufficient sampling to determine the clearance status. Seventeen (55%) recipients (cases 1-17) became SEN-V DNA negative within 6 months of exposure and an additional 3 became negative within 2 years of exposure (cases 20, 22, and 23) bringing the 2-year clearance rate to 65%. By 5 years, 3 additional cases became SEN-V negative (cases 24, 25, and 26) for a 5-year clearance rate of 74%; a final patient cleared the virus in year 7 (case 28) for the overall clearance rate of 77%. The rate of viral clearance was the same in patients classified as non–A to E hepatitis (nonABC in Fig. 4) as in those classified as hepatitis C. Hence, there was no evidence that coinfection with HCV either prolonged or shortened the duration of SEN-V viremia. In 2 patients (cases 27 and 29) with non–A to E hepatitis, SEN-V viremia persisted for at least 4 and 8 years, respectively; both these patients had chronic ALT elevations suggesting that persistent SEN-V infection may have been associated with chronic hepatitis, but neither patient had a liver biopsy to verify chronic hepatitis. Detection of SEN-V in the Liver. To study whether SEN-V replicates in the liver, we obtained liver tissue from 2 patients with hepatocellular carcinoma. These patients were not included in the cohort of prospectively followed open-heart surgery patients that served as the primary population for this study, but were selected because of the availability of a largevolume of liver tissue removed at the time of cancer surgery. After extracting and destroying DNA with DNase, RNA was extracted from 25 mg of both the tumor and the surrounding normal liver. Half of the extracted RNA was reverse-transcribed to cDNA. Aliquots of both RNA and cDNA were tested in a PCR reaction using SEN-V primers. Figure 5 shows that the cDNA fractions from both tumor and nontumor tissue specifically amplified with SEN-V primers. In contrast, the FIG. 4. Persistence of SEN-V infection. Eleven patients with SENV–positive non–A to E (non-ABC) hepatitis and 20 patients with combined HCV and SEN-V infections were followed sufficiently long term to assess the rate of SEN-V clearance. All recipients were SEN-V negative before transfusion and SEN-V positive 4 to 8 weeks posttransfusion. Patients 1 to 17 (55%) cleared the virus within 6 months as defined by 2 consecutive negative PCR determinations without recurrence. Six additional patients (20 and 22-26) cleared the virus within 2 to 5 years for a 5-year clearance rate of 74%. Four (13%) patients (27 and 29-31) remained carriers for 4 to 12 years. Insufficient sampling was available to determine viral clearance in cases 18, 19, and 21. The clearance rate was similar in patients with SEN-V infection alone and those with combined SEN-V and HCV infection.
HEPATOLOGY Vol. 33, No. 5, 2001
FIG. 5. Detection of SEN-V in the liver of 2 patients with hepatocellular carcinoma. Nucleic acid was extracted from the hepatic tumor and from surrounding nontumorous tissue in both patients. The extract was treated with DNase to destroy viral and cellular DNA and the RNA was reversetranscribed and then amplified with SEN-V–specific primers. The cDNA from both tumor tissue and normal liver was specifically amplified whereas RNA was not amplified. Because DNA had been destroyed by DNase and RNA did not amplify, the restricted amplification of cDNA suggested that this was derived from an SEN-V RNA replicative intermediate.
RNA did not amplify. Since SEN-V is a DNA virus, the specific amplification of cDNA suggests that the liver contained a SEN-V RNA replicative intermediate. Since the RNA could not be amplified by the SEN-V primers used, it is unlikely that the positive reactions obtained in the cDNA fractions were due to residual SEN-V DNA in the RNA preparation. No statistical analysis was performed because only a single experiment was conducted. DISCUSSION
After the discovery of the hepatitis viruses A through E, several lines of evidence suggested that an additional hepatitis agent(s) might exist.4 By exclusion, these have tentatively been designated non–A to E hepatitis or sometimes non-ABC hepatitis. The evidence for the existence of one or more unrecognized hepatitis agents includes the observations that 10% to 20% of both transfusion-associated and communityacquired hepatitis cases test negative for all established hepatitis viruses,4,5 as do approximately 30% of cases of cryptogenic chronic liver disease and cirrhosis,24 most cases of hepatitis-associated aplastic anemia25 and a large proportion of fulminant hepatitis cases.26 Armed with these observations, many investigators, particularly those linked to industry, embarked on viral discovery programs. These led sequentially to the discoveries of GBV-C,7 HGV,6 and TTV.11 None of these agents proved to be the cause of non–A to E hepatitis. More recently, SEN-V was described16,17 and the TTV family was expanded to include SANBAN,27 YONBAN,28 and TUS01.29 Physical characterization and phylogenetic analysis now suggests that TTV, SANBAN, YONBAN, TUS01, and SEN-V are all members of the Circoviridae family, a group of small, single stranded, nonenveloped circular DNA viruses previously associated only with diseases in animals. These agents in the human circovirus family are highly diverse, having sequence differences that classify each member into subtypes and distinguish members into subfamilies showing sequence homology of only 40% to 60%. As the number of SEN-V variants and TTV variants expanded in independent investigations conducted in Japan,13,29,30 Italy,16 and the United States,12 it became apparent
UMEMURA ET AL.
1309
that there was overlap among these agents and that SEN-V and TTV and their widely divergent variants were members of what might be considered a super-family of human circoviruses.18 In this investigation, we studied 2 SEN-V variants designated SENV-D and SENV-H. A separate study18 suggested that SENV-D and SENV-H resemble TTV variants genotypes 12 and 16, respectively.29 SENV-D and SENV-H were chosen for additional study from among the 8 known variants of SEN-V because cloning and sequencing revealed their presence in the majority of transfusion-associated non–A to E hepatitis cases, and molecular screening showed them to have relatively low prevalence in the general population. The full investigation involved all our 13 known transfusion-associated non–A to E cases, 232 control patients who were transfused and did not develop hepatitis, 49 patients with transfusion-associated hepatitis C, 100 nontransfused controls, and 823 donors sampled before and after routine HCV screening was initiated. Eleven patients who were SEN-V positive before their index surgical procedure were excluded from analysis. SENV-D and SENV-H infections were of moderate prevalence among current volunteer blood donors (1.8%), and were not significantly higher in donors before HCV screening (pre1990). The prevalence was similar in hospitalized patients before transfusion (2.8%) suggesting that the general population prevalence is approximately 2% to 3%. In the United States, the prevalence of these SEN-V variants is at least 5 times higher than the prevalence of HCV viremia (0.3%) and hepatitis B surface antigenemia (0.1%) among blood donors before screening for these agents. In this study, nearly 15% of SEN-V–infected recipients became long-term carriers; 12 years of persistent infection was documented in 2 recipients. In contrast, 55% of infected subjects cleared the virus within 6 months and 74% within 5 years of onset. This suggests that the number of exposures to SEN-V considerably exceed the number of active infections detected by PCR. There is need to develop an assay to detect antibodies to SEN-V so as to estimate better the overall frequency and potential burden of this infection. Our findings strongly suggest that SEN-V is transmitted by transfusion. A significant association with blood transfusion was observed among patients who were prospectively followed after open-heart surgery; 30% of transfused patients were acutely infected with SENV-D and/or -H compared with 3% of patients who were not transfused (P ⬍ .001). In addition, there was a significant association between transfusion volume and SEN-V infections with the risk increasing stepwise to 45% in those who received 13 or more units of blood (P ⬍ .0001). Donor samples were available in only 6 of the 11 SEN-V– infected non–A to E hepatitis cases. An SEN-V positive donor was detected in 4 of these 6 cases (67%) and in a similar proportion of donors to SEN-V–infected recipients who did not develop non–A to E hepatitis. The failure to detect SEN-V in 100% of implicated donors probably reflected the generally low level of viremia in this infection and the potential degradation of nucleic acid to nondetectable levels in these samples that had been stored up to 20 years. However, one cannot rule out that the SEN-V infections in some of these cases were unrelated to transfusion because 3% of patients undergoing open-heart surgery were acutely infected with SEN-V variants in the perioperative period in the absence of transfusion. This suggests either nosocomial transmission of SEN-V, as has
1310 UMEMURA ET AL.
been previously reported for TTV31 or reactivation of latent SEN-V infection. In the absence of an established antibody assay for SEN-V, it was not possible to directly assess prior exposure and the potential for reactivation. The strongest evidence for transmission of SEN-V by blood transfusion derives from molecular linkage experiments. Cloning and sequencing of SEN-V was performed in 2 non–A to E cases. In both, virtually identical SEN-V variants were noted in the donor and recipient and, in 1 recipient 2 distinct SEN-V variants were transmitted by different donors. The most significant finding in this study was the strong association between acute infection with SEN-V and the development of transfusion-associated non–A to E hepatitis. Eleven of 12 patients (92%) with non–A to E hepatitis were SEN-V negative before transfusion and became SEN-V positive in temporal association with the onset of hepatitis. Further, the level of viremia tended to parallel the level of ALT. Although the observed association with transfusion-associated non–A to E hepatitis is striking, the association does not prove causality. There are several considerations that must be addressed before a causal relationship can be established. First, the number of non–A to E cases studied is small. We have tested all the prospectively followed cases at our hospital, but these cases are relatively rare compared with transfusion-associated hepatitis C, and the total number available was only 13, one of whom was preinfected with SEN-V and could not be analyzed. Hence, additional non–A to E hepatitis cases need to be studied, and it is essential that these findings be confirmed in other prospectively followed cohorts. Second, 24% of 225 patients who did not develop hepatitis were also acutely infected with SEN-V. Because the number of non–A to E hepatitis cases was small and the number of identically followed patients who did not develop hepatitis was large, it would appear that the vast majority of SEN-V–infected patients do not manifest biochemical or clinical evidence of hepatitis. Although the proportion of hepatitis cases who were infected with SEN-V (92%) was significantly (P ⬍ .001) greater than nonhepatitis controls (24%), the low overall percentage of SEN-V–infected patients who actually develop disease is unsettling in terms of establishing a causal association. Nonetheless, it is not unusual in virology that the majority of immunocompetent subjects who are infected with a given agent fail to manifest evident disease. Indeed, this phenomenon is quite common with cell-associated herpes viruses such as cytomegalovirus and Epstein-Barr virus and even common with hepatitis viruses A and B. Thus, whereas a lower background rate would have increased the likelihood of a causal association, the observed low frequency of hepatitis among SEN-V–infected subjects does not negate the possibility that SEN-V is the responsible agent in some patients. Third, before establishing SEN-V as a cause of non–A to E hepatitis, it is essential to prove that SEN-V is hepatotropic and replicates within hepatocytes. Such proof is currently lacking although preliminary evidence for intrahepatic replication of SEN-V was found in this study. In our study, liver tissue was obtained from 2 patients with hepatocellular carcinoma that were SEN-V DNA positive. Nucleic acid was extracted from the tumor and from the surrounding normal tissue. After DNA inactivation, the residual RNA was reverse transcribed and amplified using SEN-V–specific primers. SEN-V was detected in both tumor and nontumor tissue in both cases. Since this experiment detected RNA specific to a DNA virus, it is un-
HEPATOLOGY May 2001
likely to have been a blood contaminant. Because DNA had been destroyed, the experiment was interpreted to show the presence of an RNA replicative intermediate of SEN-V within liver tissue. Supporting intrahepatic localization is a recent report that the related TT virus has been found in the liver by in situ hybridization.32 These data are not sufficient to prove hepatotropism and intrahepatic replication, but if confirmed would lend further support to the hypothesis that SEN-V can cause hepatitis in some infected individuals. The sophistication and sensitivity of molecular approaches to viral discovery virtually ensures that new viruses will continue to be discovered. In the hepatitis arena alone, the past 5 years has seen the independent discovery of GBV-C and its variant, HGV and now the enlarging family of circoviruses that includes TTV, SANBAN, YONBAN, and SEN-V. Each time a novel sequence is discovered, a search for disease associations ensues. The process of establishing a causal relationship with viruses becomes very difficult when the prevalence of the agent is relatively high and the proportion of infected individuals who manifest diseases relatively low. Thus far, no disease association has been established for GBV-C/HGV or TTV. The current study suggests a disease association for SEN-V, but there are several caveats to this interpretation as described above. The sensitivity of molecular amplification and subtractive cloning techniques may uncover agents that are merely innocent bystanders, a “normal viral flora” similar to that found for bacteria. These organisms may not be pathogenic or may be pathogenic only under special circumstances such as an acquired immunodeficiency. Such molecular discoveries generally will not fulfill Koch’s postulates and thus there is need to establish new guidelines for causality to meet the exigencies of the molecular age. Where SEN-V will fall on the pathogenic scale is difficult to establish at this time, but it is unequivocal that SEN-V and the new “super-family” of human circoviruses are novel infectious agents that are readily transmitted by transfusion and that can lead to persistent infection. At issue is whether these agents cause non–A to E hepatitis or some extrahepatic disease and, if so, with what frequency and with what degree of clinical relevance. If causality is established, then secondary issues such as the need to screen the donor population will have to be addressed. It is premature to consider blood screening at this time. An additional consideration is that members of the human circovirus family are so diverse (only 40% to 60% sequence homology) that they might have very different pathogenicity and each phylogenetic grouping may have to be studied independently to assess clinical relevance. Such investigation of disease associations should not be limited to hepatitis. REFERENCES 1. Choo Q-L, Kuo G, Weiner AJ, Overby LR, Bradley DW, Houghton M. Isolation of a cDNA clone derived from a blood-borne non-A, non-B viral hepatitis genome. Science 1989;244:359-362. 2. Kuo G, Choo Q-L, Alter HJ, Gitnick GL, Redeker AG, Purcell RH, Miyamura T, et al. An assay for circulating antibodies to a major etiologic virus of human non-A, non-B hepatitis. Science 1989;244:362-364. 3. Alter HJ, Purcell RH, Shih JW, Melpolder JC, Houghton M, Choo Q-L, Kuo G. Detection of antibody to hepatitis C virus in prospectively followed transfusion recipients with acute and chronic non-A, non-B hepatitis. N Engl J Med 1989;321:1494-1500. 4. Alter HJ, Bradley DW. Non-A, non-B hepatitis unrelated to the hepatitis C virus (non-ABC). Semin Liver Dis 1995;15:110-120. 5. Alter MJ, Margolis HS, Krawczynski K, Judson FN, Mares A, Alexander WJ, Hu PY, et al. The natural history of community-acquired hepatitis C in the United States. N Engl J Med 1992;327:1899-1905.
HEPATOLOGY Vol. 33, No. 5, 2001 6. Linnen J, Wages J Jr, Zhang-Keck ZY, Fry KE, Krawczynski KZ, Alter H, Koonin E, et al. Molecular cloning and disease association of hepatitis G virus: a transfusion-transmissible agent. Science 1996;271:505-508. 7. Simons JN, Leary TP, Dawson GJ, Pilot-Matias TJ, Muerhoff AS, Schlauder GG, Desai SM, et al. Isolation of novel virus-like sequences associated with human hepatitis. Nat Med 1995;1:564-569. 8. Alter HJ, Nakatsuji Y, Melpolder J, Wages J, Wesley R, Shih JW-K, Kim JP. The incidence of transfusion-associated hepatitis G virus infection and its relation to liver disease. N Eng J Med 1997;336:747-754. 9. Tanaka E, Alter HJ, Nakatsuji Y, Shih JWK, Kim JP, Matsumoto A, Kobayashi M, et al. Effect of hepatitis G virus infection on chronic hepatitis C. Ann Intern Med 1996;125:740-743. 10. Alter HJ. G-pers creepers, where’d you get those papers? A reassessment of the literature on the hepatitis G virus. Transfusion 1997;37:569-572. 11. Nishizawa T, Okamoto H, Konisi K, Yoshizawa H, Miyakawa Y, Mayumi M. A novel DNA virus (TTV) associated with elevated transaminase levels in posttransfusion hepatitis of unknown etiology. Biochem Biophys Res Commun 1997;241:92-97. 12. Matsumoto A, Yeo AET, Shih JWK, Tanaka E, Kiyosawa K, Alter HJ. Transfusion-associated TT virus infection and its relationship to liver disease. HEPATOLOGY 1999;30:283-288. 13. Takahashi K, Hoshino H, Ohta Y, Yoshida N, Mishiro S. Very high prevalence of TT virus (TTV) infection in general population of Japan revealed by a new set of PCR primers. Hepatol Res 1998;12:233-239. 14. Simmonds P, Davidson F, Lycett C, Prescott LE, MacDonald DM, Ellender J, Yap PL, et al. Detection of a novel DNA virus (TT virus) in blood donors and blood products. Lancet 1998;352:191-195. 15. Naoumov NV, Petrova EP, Thomas MG, Williams R. Presence of a newly described human DNA virus (TTV) in patients with liver disease. Lancet 1998;352:195-197. 16. Primi D, Fiordalisi G, Mantero JL, Mattioli S, Sottini A, Bonelli F, Vaglini L, et al. Identification of SENV genotypes. International patent number WO0028039 (international application published under the patent cooperation treaty). Internet address: http://ep.espacenet.com/. 17. Sottini A, Mattioli S, Fiordalisi G, Mantero G, Imberti L, Moratto D, Primi D. Molecular and biological characterization of SEN viruses: a family of viruses remotely related to the original TTV isolates. Proceedings of the 10th International Symposium on Viral Hepatitis and Liver Disease. H. Margolis (Ed), Meditech Media, Atlanta. 2001 (in press). 18. Tanaka Y, Primi D, Wang RYH, Umemura T, Yeo AET, Mizokami M, Alter HJ, et al. Genomic and molecular evolutionary analysis of a newly identified infectious agent (SEN virus) and its relationship to the TT virus family. J Infec Dis 2001;183:359-367. 19. Alter HJ, Sanchez-Pescador R, Urdea MS, Wilber JC, Lagier RJ, Di Bisceglie AM, Shih JW, et al. Evaluation of branched DNA signal amplification for the detection of hepatitis C virus RNA. J Viral Hepat 1995;2:121-132.
UMEMURA ET AL.
1311
20. Ina Y. ODEN: a program package for molecular evolutionary analysis and database search of DNA and amino acid sequences. Comput Appl Biosci 1994;10:11-12. 21. Gojobori T, Ishii K, Nei M. Estimation of average number of nucleotide substitutions when the rate of substitution varies with nucleotide. J Mol Evol 1982;18:414-423. 22. Saitou N, Nei M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 1987;4:406-425. 23. Tsarev SA, Tsareva TS, Emerson SU, Kapikian AZ, Ticehurst J, London W, Purcell RH. ELISA for antibody to hepatitis E virus (HEV) based on complete open-reading frame-2 protein expressed in insect cells: identification of HEV infection in primates. J Infect Dis 1993;168:369-378. 24. Kodali VP, Gordon SC, Silverman AL, McCray DG. Cryptogenic liver disease in the United States: further evidence for non-A, non-B, and non-C hepatitis. Am J Gastroenterol 1994;89:1836-1839. 25. Brown KE, Tisdale J, Barrett AJ, Dunbar CE, Young NS. Hepatitis-associated aplastic anemia. N Engl J Med 1997;336:1059-1064. 26. Ferraz ML, Silva AE, Macdonald GA, Tsarev SA, Di Bisceglie AM, Lucey MR. Fulminant hepatitis in patients undergoing liver transplantation: evidence for a non-A, non-B, non-C, non-D, and non-E syndrome. Liver Transpl Surg 1996;2:60-66. 27. Hijikata M, Takahashi K, Mishiro S. Complete circular DNA genome of a TT virus variant (isolate name SANBAN) and 44 partial ORF2 sequences implicating a great degree of diversity beyond genotypes. Virology 1999; 260:17-22. 28. Takahashi K, Hijikata M, Samokhvalov EI, Mishiro S. Full or near full length nucleotide sequences of TT virus variants (type SANBAN and YONBAN) and the TT virus-like mini virus. Intervirology 2000;43:119123. 29. Okamoto H, Takahashi M, Nishizawa T, Ukita M, Fukuda M, Tsuda F, Miyakawa Y, et al. Marked genomic heterogeneity and frequent mixed infection of TT virus demonstrated by PCR with primers from coding and noncoding regions. Virology 1999;259:428-436. 30. Umemura T, Yeo AET, Shih JWK, Matsumoto A, Orii K, Tanaka E, Kiyosawa K, et al. The prevalence of SEN virus infection in Japanese patients with viral hepatitis and liver disease [Abstract]. HEPATOLOGY 2000; 32(suppl):381A. 31. Okamoto H, Akahane Y, Ukita M, Fukuda M, Tsuda F, Miyakawa Y, Mayumi M. Fecal excretion of a nonenveloped DNA virus (TTV) associated with posttransfusion non-A-G hepatitis. J Med Virol 1998;56:128132. 32. Rodriguez-In˜igo E, Casqueiro M, Bartolome´ J, Oritz-Movilla N, Lo´pezAlcorocho JM, Herrero M, Manzarbeitia F, et al. Detection of TT virus DNA in liver biopsies by in situ hybridization. Am J Pathol 2000;156: 1227-1234.
SEE EDITORIAL ON PAGE 1334
SEN virus (SEN-V) is a recently identified singlestranded, circular DNA virus. Two SEN-V variants (SENV-D and SENV-H) were assayed by polymerase chain reaction (PCR) to investigate their role in the causation of transfusion-associated non–A to E hepatitis. The incidence of SEN-V infection after transfusion was 30% (86 of 286) compared with 3% (3 of 97) among nontransfused controls (P < .001). Transfusion risk increased with the number of units transfused (P < .0001) and donor-recipient linkage for SEN-V was shown by sequence homology. The prevalence of SEN-V in 436 volunteer donors was 1.8%. Among patients with transfusion-associated non–A to E hepatitis, 11 of 12 (92%) were infected with SEN-V at the time of transfusion compared with 55 of 225 (24%) identically followed recipients who did not develop hepatitis (P < .001). No effect of SEN-V on the severity or persistence of coexistent hepatitis C virus (HCV) infection was observed. In 31 infected recipients, SEN-V persisted for greater than 1 year in 45% and for up to 12 years in 13%. SEN-V–specific RNA (a possible replicative intermediate) was recovered from liver tissue. In summary, SENV-D and -H were present in nearly 2% of US donors, and were unequivocally transmitted by transfusion and frequently persisted. The strong association of SEN-V with transfusion-associated non–A to E hepatitis compared with controls raises the possibility, but does not establish that SEN-V might be a causative agent of posttransfusion hepatitis. The vast majority of SEN-V–infected recipients did not develop hepatitis. (HEPATOLOGY 2001;33:1303-1311.) After cloning of the hepatitis C virus (HCV)1 and development of sensitive serologic and molecular assays for this pathogen, a dramatic decline in the incidence of transfusionAbbreviations: HCV, hepatitis C virus; HGV, hepatitis G virus; GBV-C, GB virus type C; TTV, TT virus; SEN-V, SEN virus; PCR, polymerase chain reaction; cDNA, complementary DNA; ALT, alanine aminotransferase; CI, confidence interval. From the 1The Department of Transfusion Medicine, Warren Grant Magnuson Clinical Center, National Institutes of Health, Bethesda, MD; 2DiaSorin Inc., Brescia, Italy; and 3DiaSorin Inc., Stillwater, MN. Received September 27, 2000; accepted February 27, 2001. The sequences reported in this paper have been deposited in the DDBJ/EMBL/GeneBank data base (accession no. AB049170-AB049183). Address reprint requests to: Harvey J. Alter, M.D., Department of Transfusion Medicine, Building 10, Room 1C711, National Institutes of Health, Bethesda, MD 208921184. E-mail: [email protected]; fax: 301-402-2965. Copyright © 2001 by the American Association for the Study of Liver Diseases. 0270-9139/01/3305-0036$35.00/0 doi:10.1053/jhep.2001.24268
associated hepatitis occurred.2,3 However, approximately 10% of transfusion-associated hepatitis cases4 and 20% of community-acquired hepatitis cases5 do not have a defined etiology suggesting the existence of additional causative agents. Hepatitis G virus (HGV),6 also known as GB virus C (GBV-C),7 was initially suggested as a causative agent of non–A to E hepatitis, but this was not confirmed in further studies.8-10 The originally described TT virus (TTV), discovered in 1997 by representational difference analysis, was detected in 3 of 5 cases of transfusion-associated hepatitis and was proposed as a potential cause of non–A to E hepatitis.11 Using TTV primers identical to those reported, non–A to E hepatitis cases and controls in the NIH prospective series were tested but no association with transfusion-associated hepatitis was found.12 Subsequently, primers to more conserved regions of the TTV genome were used in studies in Japan, and the agent was found in greater than 90% of Japanese blood donors.13 This further diminished the likelihood that TTV was the cause of transfusion-associated hepatitis and precluded the possibility of a practical screening assay. Other studies have also confirmed the high prevalence of TTV and its lack of association with transfusion-associated hepatitis.14,15 Recently, a novel DNA virus designated SEN virus (SEN-V) was discovered in the serum of an intravenous drug abuser also infected with human immunodeficiency virus.16,17 SEN-V was initially described as a single-stranded DNA virus of approximately 3,800 nucleotides.16 Phylogenetic analysis of SEN-V has shown the existence of 8 strains.18 Although structurally similar to TTV, SEN-V has less than 55% sequence homology and less than 37% amino acid homology with the TTV prototype.18 Preliminary studies of SEN-V variants were conducted by Dr. Primi (DiaSorin Inc., Brescia, Italy). The prevalence of 5 SEN-V strains (A, B, H [former C], D, and E) and a consensus sequence designated total SEN-V were measured in various donor and patient populations. It was shown that measuring total SEN-V was not practical because the prevalence in donors was 13% and the rate in all transfused populations exceeded 70%. SENV-B was also excluded as a useful screening assay because it was present in 10% of donors and only 8% of patients with transfusion-associated non–A to E hepatitis. SENV-A and SENV-E were found in low prevalence in donors, but did not appear to be associated with non–A to E hepatitis. In contrast, SENV-D and SENV-H had favorable prevalence ratios being found in less than 1% of donors and more than 50% of transfusion-associated non–A to E cases. These initial polymerase chain reaction (PCR) data were confirmed by cloning and sequencing which showed that SENV-D and SENV-H sequences were found in a high proportion of non–A to E hepatitis cases. This prelimi-
1303
1304 UMEMURA ET AL.
nary effort served as the basis for performing a systematic search for SENV-D and H in blood recipients who did and did not develop transfusion-associated hepatitis and in nontransfused controls. This systematic investigation is the foundation of the current study. PATIENTS AND METHODS Selection of Patients and Donors. Stored serum samples from donors and recipients were obtained from subjects enrolled in prospective studies of transfusion-associated hepatitis at the NIH conducted from October 1972 to December 1997. The methods of enrollment, frequency, and duration of patient follow-up and the criteria for the diagnosis of hepatitis have been described previously.3 The study protocols were reviewed and approved by the appropriate institutional review boards, and all study subjects gave written informed consent. SEN-V positivity was determined in 13 patients who developed transfusion-associated non–A to E hepatitis after open-heart surgery, 49 patients who developed transfusion-associated hepatitis C, 232 transfused patients who did not develop hepatitis, and 100 patients who had heart surgery but did not require transfusion or develop hepatitis. For patients identified as having transfusion-associated SEN-V infection, all available linked donor samples were tested for SEN-V DNA (n ⫽ 213). The prevalence of SEN-V was determined in 436 current volunteer blood donors and 387 randomly selected donors whose samples were obtained prior to 1990. Samples were stored at ⫺70°C before testing for SEN-V DNA. Detection of SEN-V DNA and HCV RNA by the PCR. Nucleic acids were extracted from 100 L of serum by QIAamp DNA Blood Mini Kit (Qiagen, Valencia, CA). The extracted DNA was eluted in 160 L buffer containing 10 mmol/L Tris-HCl, 0.5 mmol/L ethylenediaminetetraacetic acid, pH 9.0. SENV-D DNA and SENV-H DNA were determined by PCR. For SENV-D detection, primer D10S (sense primer 5⬘-GTAACTTTGCGGTCAACTGCC-3⬘) and primer L2AS (antisense primer, 5⬘-CCTCGGTTKSAAAKGTYTGATAGT-3⬘ [K ⫽ G or T, S ⫽ C or G, Y ⫽ C or T]) were used in a 40-L PCR mixture containing PCR buffer, 0.25 mmol/L dNTPs, 2.5 mmol/L magnesium chloride, 1.75 U AmpliTaq DNA Polymerase (Perkin-Elmer Applied Biosystems, Foster City, CA), and 10 L of extracted DNA. PCR involved 40 cycles (94°C for 1 minute, 58°C for 1 minute, and 72°C for 1 minute for each cycle) followed by the extension reaction at 72°C for 9 minutes in a Perkin Elmer 9700 thermal cycler. For SENV-H detection, primer C5S (sense primer 5⬘-GGTGCCCCTWGTYAGTTGGCGGTT-3⬘ [W ⫽ A or T]) and the same antisense primer L2AS were used. Forty cycles (94°C for 1 minute, 62°C for 1 minute, and 72°C for 1 minute for each cycle) were again performed, followed by the extension reaction at 72°C for 9 minutes. PCR products were analyzed by DNA enzyme immunoassay (DiaSorin, Saluggia, Italy) with SENV-D (5⬘-ATGATAGGCTTCCCYTTTAACTATAACCCA-3⬘) or SENV-H (5⬘-CCCCTTCCAGGTATTGCATGAAGAGTATTAC-3⬘) specific 5⬘-end biotinylated probes. Specimens with optical density (OD) values ⱖ0.350 were considered positive for SENV-D or SENV-H. The estimated lowest detection level for SEN-V DNA is 10 copies per each test based on a dilution series of plasmid DNA containing an SEN-V PCR target insert. This was equivalent to 1,600 copies/mL in the original serum being extracted. All positive samples were confirmed by duplicate retesting. Procedures for HCV RNA PCR were performed as described elsewhere.19 Quantitation of SEN-V DNA. SEN-V DNA was quantitated with external standards containing known concentrations of SEN-V DNA. These standards were constructed by inserting SEN-V–positive PCR products into TOPO TA cloning vector pCR 2.1 (Invitrogen, Carlsbad, CA). Digoxigenin labeling mix (200 mol/L of dATP, dCTP, and dGTP each, 190 mol/L dTTP, and 10 mol/L digoxigenin-11dUTP) was used in the PCR reaction mixture. Streptavidin-coated black enzyme-linked immunosorbent assay plates (Roche Molecular Biochemicals, Indianapolis, IN) were incubated overnight at 2°C to
HEPATOLOGY May 2001
8°C with biotinylated SENV-D or SENV-H probes. The plates were washed and 90 L of hybridization solution and 10 L of PCR amplified products denatured at 95°C for 5 minutes were added. The hybridization was performed at 50°C for 1 hour for SENV-D or at 45°C for 1 hour for SENV-H. The enzyme-linked immunosorbent assay plates were then washed again and anti-digoxigenin antibodies conjugated to alkaline phosphatase were added. The mixture was incubated for 30 minutes at 37°C. Reactivity was detected using 0.25 mmol/L disodium 3-(4-methoxyspiro {1, 2-dioxetane-3, 2⬘-(5⬘chloro) tricyclo [3.3.1.13,7] decan}-4-yl) phenyl phosphate as substrate in 0.1 mol/L Tris-HCl, 0.1 mol/L NaCl, pH 9.5. The reaction was incubated at 37°C for 15 minutes. Chemiluminescence reading was obtained by using a 1450 Micro Beta PLUS scintillation counter (Wallac Inc, Turku, Finland). The total chemiluminescence reading of each standard was plotted against the SEN-V DNA copy number to establish a standard curve; the counts of chemiluminescence and the copy number correlated linearly from 5 to 10,000 copies per reaction (P ⬍ .001; Pearson’s correlation coefficient, r ⫽ .79). The number of viral particles was determined in relation to the standard curve and expressed as copies per milliliter. Sequence and Phylogenetic Analysis of SEN-V DNA. DNA extracted from the sera of 1 SENV-D– and 1 SENV-H–positive patient with transfusion-associated non–A to E hepatitis and 3 implicated donors (1 SENV-D and 2 SENV-H) was amplified using SEN-V–specific primers. All samples produced bands visible on agarose gel. To determine the nature of the differently sized bands, amplicons containing poly A tails were excised from the agarose gel and the DNA content was purified and ligated to a TOPO TA cloning vector pCR 2.1. Using DNA extracted from transformed Escherichia coli, both strands were sequenced with AutoRead sequencing Kit (Amersham Pharmacia Biotech, Piscataway, NJ) using fluorescent M13 universal primer and fluorescent M13 reverse primer. Sequences were determined on 4 to 6 clones each for respective PCR products. The sequences, excluding primers sequences (D, 179 bp; H, 178 bp), were aligned with clustal W (version 1.8). Using the computer program ODEN version 1.1.1,20 the number of nucleotide substitutions per site (genetic distance) between the isolates was estimated by the 6-parameter method.21 Based on these values, a phylogenetic tree was constructed by the neighbor-joining method.22 Detection of SEN-V in Liver Tissue. To determine the presence of SEN-V replication in the liver, assays were conducted on liver tissue surgically extracted from 2 SEN-V–positive patients who had hepatocellular carcinoma. RNA was extracted from 25 mg of liver tissue using the S.N.A.P. Total RNA isolation kit (Invitrogen) with slight modification to the manufacturer’s instructions. The modification consisted of incubating eluted nucleic acids with RNase-free DNase at 37°C for 30 minutes instead of the recommended 10 minutes. The eluted RNAs (125 L) were precipitated with 2 volumes of ethanol and 1/10th volume of sodium acetate pH 5.2, 3 mol/L for 1 hour at ⫺20°C. The pelleted RNA was washed with 500 L of 75% ethanol, centrifuged, and resuspended in 25 L of RNase-free water. Ten L of RNA were retrotranscribed to single-stranded DNA using the SUPERSCRIPT II RNase H⫺ reverse transcriptase (Gibco BRL, Rockville, MD) and 500 ng of primer L3AS (5⬘-GCCTATAAAAATGTKASWGWACCAKCC-3⬘) in a final volume of 20 L. Five microliters of complementary DNA (cDNA) and 2.5 L of RNA (as control) were amplified using primers NEW BCD 1S (sense primer 5⬘CCYAARCTMTTTGAAGACMA-3⬘ [M ⫽ A or C, R ⫽ A or G]) in combination with primer L2AS, respectively, derived from conserved sequences among SENV-D and SENV-H. The PCR reactions were performed with a 50-L mixture containing template, 5 L of cDNA, PCR buffer, 300 ng of a single PCR primer, a 200 mol/L concentration of each dNTPs, and 1.25 U of Taq polymerase (PerkinElmer Applied Biosystems, Foster City, CA). The reaction was performed in a DNA thermal cycler with mineral oil. PCR consisted of preheating at 94°C for 5 minutes, 40 cycles of 94°C for 1 minute, 54°C for 1 minute, and 72°C for 1 minute, with a terminal extension incubation at 72°C for 7 minutes. To verify the specificity of the amplified products, aliquots of 5 L of each amplified product were
HEPATOLOGY Vol. 33, No. 5, 2001
UMEMURA ET AL.
1305
TABLE 1. Prevalence of SEN-V DNA in Volunteer Donors and Patients SEN-V DNA Positivity Group
Total n
n (%)
(95% CI)
Volunteer donors (current) Volunteer donors (pre-1990) Patients*
436 387 394
8 (1.8%)† 9 (2.3%)‡ 11 (2.8%)§
(0.6-3.1) (0.8-3.8) (1.2-4.4)
*Sampled before surgery/transfusion. † vs. ‡: P ⫽ .80; † vs. §: P ⫽ .49; ‡ vs. §: P ⫽ .85.
hybridized in a DNA enzyme immunoassay with the probe (5⬘-TCCGTTCTGCTCACCACAAACGTAGACAACCC-3⬘) specific for a conserved sequence within the amplified DNA. Serological and Biochemical Assays. Antibodies to HCV were measured by a second-version enzyme immunoassay according to the directions of the manufacturer (Abbott HCV EIA 2.0; Abbott Laboratories, North Chicago, IL). Antibody to hepatitis E virus was measured by the method of Tsarev et al.23 Alanine aminotransferase (ALT) was measured by a 3-point kinetic assay with an automated photometric analyzer (model 917; Hitachi–Boehringer Mannheim, Indianapolis, IN), and the values were expressed in international units (normal range, 5-41 U/L). Statistical Analysis. Student’s t test was used to analyze continuous variables. 2 test with Yates’s correction was used in the analysis of categorical data; when the number of the subjects was less than 5, Fisher’s exact test was used. The Mantel-Haenszel 2 test was used for trend analysis. Pearson’s correlation coefficient was calculated to assess the robustness of the quantitative SEN-V assay. A P value of ⱕ.05 was considered significant. Statistical analyses were performed using SigmaStat (version 2.03; SPSS Inc., Chicago, IL) and SAS 6.12 (SAS Institute, Cary, NC). RESULTS Prevalence of SEN-V in Blood Donors and Patients Prior to Transfusion. SEN-V DNA was detected in 8 of 436 (1.8% [95% con-
fidence interval (CI): 0.6-3.1]) randomly selected recent (1999) blood donors and 9 of 387 (2.3% [95% CI: 0.8-3.8]) randomly selected donors sampled prior to HCV screening (pre-1990). There was no statistically significant difference in donor prevalence between these time frames (P ⫽ .80) (Table 1). The prevalence of SEN-V in a sample obtained from patients before surgery or transfusion was 2.8% (11 of 394 [95% CI: 1.2-4.4]). This was not significantly different from the prevalence observed in recent blood donors (P ⫽ .49) or pre1990 donors (P ⫽ .85) (Table 1). Newly Acquired SEN-V Infections in Relationship to Blood Transfusion. Excluding the 11 patients who were SEN-V positive
before surgery, SEN-V DNA (strain D and/or H) was detected
FIG. 1. The relationship between the prevalence of SEN-V infection and the number of units of blood transfused. A step-wise increase in the risk of infection was observed with increased transfusion volume (P ⬍ .0001).
after surgery in 86 of 286 (30% [95% CI: 25-35]) patients who were transfused compared with 3 of 97 (3% [95% CI: 0.6-9]) patients who were not transfused (P ⬍ .001) (Table 2). The strong association between blood transfusion and SEN-V infection was similar when the data were analyzed for combined SENV-D and SENV-H infection or for SENV-D (P ⫽ .001) or SENV-H (P ⬍ .001) individually (Table 2). The association with blood transfusion was also indicated when SEN-V incidence was stratified according to transfusion volume (Fig. 1); increasing prevalence of SEN-V infections was observed as the transfusion volume increased from 0 to greater than 13 units (P ⬍ .0001). The 86 recipients who developed an acute SEN-V infection after transfusion received a mean of 13.8 (95% CI: 10.1-17.5) units of blood compared with a mean of 9.4 (95% CI: 8.510.3) units transfused to the 200 recipients who were not SEN-V infected (P ⬍ .01). Donor-Recipient Linkage. Samples were available from all donors in only 19 of the 86 (22%) recipients who developed SEN-V infection after transfusion. Those for whom complete donor sampling was available received a mean of 10.0 (95% CI: 6.8-13.2) units of blood compared with a mean of 14.8 (95% CI: 10.2-19.5) units transfused to the 67 SEN-V–positive recipients for whom some or all donor samples were missing (P ⫽ .28). Thirteen of the 19 (68%) SEN-V–infected recipients for whom all donor samples were available had at least 1 SEN-V– positive donor (7 with 1 positive donor, 5 with 2 positive donors, and 1 with 4 positive donors). For the 12 evaluable non–A to E hepatitis cases, 3 had a full complement of donor
TABLE 2. Relationship of Blood Transfusion to New SEN-V Infections Proportion of SEN-V Infections by Strain New SEN-V Infections* Group
Nontransfused Transfused
n
97† 286‡
SENV-D
SENV-H
n
%
(95% CI)
n
%
(95% CI)
n
%
(95% CI)
3 86
3%§ 30%㛳
(0.6-9) (25-35)
1 37
1%¶ 6%#
(0.03-6) (3-8)
2 70
2%** 17%††
(0.3-7) (13-21)
*Sampled presurgery and at 4 to 8 weeks after surgery. †Excludes 3 patients SEN-V DNA⫹ before surgery. ‡Excludes 8 patients SEN-V DNA⫹ before surgery. § vs. 㛳: P ⬍ .001; ¶ vs. #: P ⫽ .001; ** vs. ††: P ⬍ .001. #, ††: 21 patients had combined SENV-D and SENV-H infections and were included in both columns.
1306 UMEMURA ET AL.
HEPATOLOGY May 2001
bored an SENV-H variant identical to 1 strain recovered from the recipient, and the other donor (donor 2) showed 94.4% sequence homology with the second strain in this recipient (Fig. 2). Sequence homology between donor and recipient of greater than 99% for both SENV-D and SENV-H in these 2 cases unequivocally confirms transmission of these agents by blood transfusion. Incidence of SEN-V Infections in Transfused Patients in Relationship to Hepatitis Outcome. Excluding 8 transfused patients with
FIG. 2. Phylogenetic tree analysis comparing SEN-V sequences from a patient with transfusion-associated non–A to E hepatitis and those from the implicated donors. Patient F.G. received blood from 2 SEN-V–infected donors, designated in the figure as donor 1 and donor 2. Cloning and sequencing of serum from patient F.G. revealed that he was infected with 2 diverse strains of SENV-H. Each rectangle in the figure represents a clone. Two patient clones (Pat. FG-1 and FG-2) were identical to 3 clones from donor 1 (Don. 1-4, 1-5, and 1-6) and 94% homologous with 3 other clones from donor 1 (Don. 1-1, 1-2, and 1-3). Patient clones (Pat. FG-3 and FG-4) showed 94% homology with clones from donor 2 (Don. 2-1, 2-2, 2-3, and 2-4). The horizontal bar indicates the number of nucleotide substitutions per site.
samples, 3 had only partial donor sampling, and 6 had no available donor samples. An SEN-V positive donor was identified in 4 of the 6 (67%) evaluable non–A to E hepatitis cases (2 with complete donor samples and 2 with partial samples). In each of the 13 infections for which there was a linked SEN-V positive donor and recipient sample, the SEN-V strain (D or H) in the donor was identical to that in the recipient. However, one patient was acutely infected with SENV-D and SENV-H, but only SENV-D was identified in the implicated donor. Cloning and sequencing was performed to determine the extent of homology between the agents found in the donor and recipient in 2 cases of non–A to E hepatitis. In 1 patient (K.S.), sequencing revealed 99.4% nucleotide homology between the SENV-D strain recovered from the donor and the strain found in the recipient when measured 7 weeks after transfusion (data not shown). A second patient (F.G.) with non–A to E hepatitis was SENV-H positive and received blood from 2 different SENV-H–positive donors. Cloning and sequencing of the recipient revealed 2 distinct SENV-H populations (Fig. 2). One SENV-H–positive donor (donor 1) har-
preexisting SEN-V DNA, newly acquired SEN-V infections were observed during posttransfusion follow-up in 11 of 12 (92%) patients with non–A to E hepatitis compared with 55 of 225 (24%) patients who were transfused, but did not develop hepatitis (P ⬍ .001; Table 3). Of the 8 patients with preexisting SEN-V DNA, 1 developed non–A to E hepatitis after transfusion and 7 showed no biochemical or clinical evidence of hepatitis during follow-up. New SEN-V infections were also observed in 20 of 49 (41%) patients who developed hepatitis C (Table 3); this incidence in patients with hepatitis C, although high, was significantly lower than the 92% incidence of SEN-V infection in patients with non–A to E hepatitis (P ⫽ .005). Of the 11 recipients with SEN-V–positive, non–A to E hepatitis, 2 were infected with SENV-D alone, 7 with SENV-H alone, and 2 with both SENV-D and SENV-H (Table 3). In total, 4 of 12 (33%) patients with non–A to E hepatitis were infected with SENV-D and 9 of 12 (75%) with SENV-H (Table 3). Among the 55 recipients who were SEN-V infected, but did not develop hepatitis, 14 (25%) were infected with SENV-D alone, 29 (53%) with SENV-H alone, and 12 (22%) with both SENV-D and SENV-H. In total, 26 of 225 (12%) patients without hepatitis were infected with SENV-D and 41 of 225 (18%) with SENV-H (Table 3). The frequency of SENV-D infections in patients with non–A to E hepatitis (33%) versus the frequency in recipients without hepatitis (12%) was statistically significant (P ⫽ .05 by Fisher’s exact test). SENV-H was more strongly associated with non–A to E hepatitis being present in 75% of cases compared with 18% of recipients without hepatitis (P ⬍ .001 by Fisher’s exact test; Table 3). Hence, SENV-H was the primary agent associated with non–A to E hepatitis, but 2 patients with non–A to E hepatitis had SENV-D infection alone and 2 patients were coinfected with SENV-H and -D. Relationship Between Viremia and ALT Level. Of the 11 patients with non–A to E hepatitis who were acutely infected with SEN-V, none were icteric and the mean peak ALT level
TABLE 3. Prevalence of New SEN-V Infections in Relation to Transfusion and Hepatitis Outcome
Disease Outcomes
Total n
Non–A to E hepatitis HCV hepatitis No hepatitis
12 49 225
Mean No. Transfusions
Proportion of SEN-V Infections by Strain Total SEN-V Infections*
SENV-D
SENV-H
(95% CI)
No.
(%)
[95% CI]
No.
(%)
[95% CI]
No.
(%)
[95% CI]
11.8 (5.8-17.9) 16.8 (10.5-23.1) 9.3 (8.5-10.1)
11 20 55
(92%)† (41%)‡ (24%)§
[62-100] [27-56] [19-30]
4 7 26
(33%)㛳 (14%)¶ (12%)#
[10-65] [6-27] [7-16]
9 18 41
(75%)** (37%)†† (18%)§§
[43-94] [23-52] [13-23]
*Excludes 11 patients SEN-V DNA⫹ before surgery. † vs. ‡: P ⫽ .005; † vs. §: P ⬍ .001 (Fisher’s exact test); ‡ vs. §: P ⫽ .031. 㛳 vs. ¶: P ⫽ .13 (Fisher’s exact test); 㛳 vs. #: P ⫽ .05 (Fisher’s exact test); ¶ vs. #: P ⫽ .77. ** vs. ††: P ⫽ .039; ** vs. §§: P ⬍ .001 (Fisher’s exact test); †† vs. §§: P ⫽ .008. 㛳, **: 2 patients had combined SENV-D and SENV-H infections. ¶, ††: 5 patients had combined SENV-D and SENV-H infections. #, §§: 12 patients had combined SENV-D and SENV-H infections.
HEPATOLOGY Vol. 33, No. 5, 2001
UMEMURA ET AL.
1307
FIG. 3. SEN-V infection in 10 patients with transfusion-associated non–A to E hepatitis. Levels of ALT (shaded areas), SENV-D DNA (dashed line), and SENV-H DNA (solid line) in 6 patients infected with SENV-H only (patients F.G., R.W., H.S., W.B., J.C., and J.S.), in 2 patients infected with SENV-D only (patients A.B. and W.S.), and in 2 patients infected with both SENV-D and SENV-H (patients K.S. and M.B.) are shown plotted against the time after transfusion. Qualitative PCR results of SEN-V DNA (positive, ⫹; negative, ⫺) are shown above each panel. The horizontal lines indicate the normal ALT level. Patients R.W., W.B., J.S., and A.B. received 14, 32, 3, and 9 units of blood, respectively; no samples from these donations were available for testing. For the remaining patients, some or all of the donor samples were available for testing and the results are depicted as the number of donors/ the number of SEN-V tested/the number of SEN-V⫹: F.G., 12/12/2; H.S., 27/10/0; J.C., 2/2/0; W.S., 4/2/1; K.S., 15/15/1 (the donor was SENVD⫹; the recipient was coinfected with SENV-D and SENV-H); and M.B., 11/10/1 (the donor was SENVH⫹; the recipient was coinfected with SENV-D and SENV-H).
was 396 U/L (95% CI: 111-681). Sufficient serial samples to correlate ALT level with the level of viremia were available in 10 of the 11 patients with non–A to E hepatitis (Fig. 3). In every case, the virus was absent pretransfusion and became detectable before or at the same time as the first ALT elevation
that signaled the onset of hepatitis. There was general concordance between the ALT level and the level of viremia, as seen particularly in cases F.G., R.W., and W.B., but there were many time points in which these levels were discordant, as exemplified in dually infected patients K.S. and M.B.
1308 UMEMURA ET AL.
HEPATOLOGY May 2001
TABLE 4. Relationship of SEN-V Infection to Severity and Persistence of Hepatitis Mean Peak ALT Group
n
(U/L; 95% CI)
Posttransfusion non-A to E hepatitis SEN-V⫹ 11 396 (111-681) SEN-V⫺ 1 200 Posttransfusion HCV hepatitis SEN-V⫹ 20 604 (420-788)§ SEN-V⫺ 29 649 (493-804)㛳
Chronicity* n
(%)
(95% CI)
2 1
(18%)† (100%)‡
(2-52)
18 24
(90%)¶ (83%)#
(68-99) (64-94)
*ALT elevations persisting ⬎1 year. † vs. ‡: P ⫽ .25 (Fisher’s exact test). § vs. 㛳: P ⫽ .60. ¶ vs. #: P ⫽ .69 (Fisher’s exact test).
Association of SEN-V With Severity and Chronicity of Liver Disease. In patients with non–A to E hepatitis, the mean peak
ALT levels for 11 subjects with SEN-V infection was 396 U/L (95% CI: 111-681) and for the 1 subject without SEN-V infection was 200 U/L (Table 4). Patients with HCV had significantly higher mean ALT levels (630 U/L [95% CI: 513-748]) than patients with non–A to E hepatitis (380 U/L [95% CI: 118-642]) (P ⬍ .001). In patients with hepatitis C, the mean peak ALT level for the 20 subjects coinfected with SEN-V was 604 U/L (95% CI: 420-788) compared with 649 U/L (95% CI: 493-804) for the 29 subjects with HCV infection alone (P ⫽ .60). Thus, hepatitis C was more severe than non–A to E hepatitis and the severity in hepatitis C cases was not influenced by coexistent SEN-V infection (Table 4). Chronic hepatitis C (duration ⬎1 year) developed in 18 of 20 (90% [95% CI: 68-99]) patients who were coinfected with HCV and SEN-V compared with 24 of 29 (83% [95% CI: 64-94]) patients who had transfusion-associated HCV without SEN-V infection (Table 4). This difference in the rates of HCV chronicity was not statistically significant (P ⫽ .69). Persistence of SEN-V Infection. Serial samples sufficient to assess the persistence of SEN-V infection were available from 31 patients, including 11 patients with non–A to E hepatitis and 20 patients coinfected with hepatitis C. Clearance of viremia,
was defined as at least 2 consecutive negative PCR determinations with no subsequent recurrence of viremia. All 31 patients were SEN-V DNA negative before transfusion and SEN-V positive 4 to 8 weeks posttransfusion; overall, 24 of 31 recipients (77%) cleared SEN-V DNA during the course of follow-up, 4 (13%) were chronic carriers for from 4 to 12 years (Fig. 4; cases 27, 29, 30, and 31) and in 3 (10%) cases (Fig. 4; cases 18, 19, and 21) there was insufficient sampling to determine the clearance status. Seventeen (55%) recipients (cases 1-17) became SEN-V DNA negative within 6 months of exposure and an additional 3 became negative within 2 years of exposure (cases 20, 22, and 23) bringing the 2-year clearance rate to 65%. By 5 years, 3 additional cases became SEN-V negative (cases 24, 25, and 26) for a 5-year clearance rate of 74%; a final patient cleared the virus in year 7 (case 28) for the overall clearance rate of 77%. The rate of viral clearance was the same in patients classified as non–A to E hepatitis (nonABC in Fig. 4) as in those classified as hepatitis C. Hence, there was no evidence that coinfection with HCV either prolonged or shortened the duration of SEN-V viremia. In 2 patients (cases 27 and 29) with non–A to E hepatitis, SEN-V viremia persisted for at least 4 and 8 years, respectively; both these patients had chronic ALT elevations suggesting that persistent SEN-V infection may have been associated with chronic hepatitis, but neither patient had a liver biopsy to verify chronic hepatitis. Detection of SEN-V in the Liver. To study whether SEN-V replicates in the liver, we obtained liver tissue from 2 patients with hepatocellular carcinoma. These patients were not included in the cohort of prospectively followed open-heart surgery patients that served as the primary population for this study, but were selected because of the availability of a largevolume of liver tissue removed at the time of cancer surgery. After extracting and destroying DNA with DNase, RNA was extracted from 25 mg of both the tumor and the surrounding normal liver. Half of the extracted RNA was reverse-transcribed to cDNA. Aliquots of both RNA and cDNA were tested in a PCR reaction using SEN-V primers. Figure 5 shows that the cDNA fractions from both tumor and nontumor tissue specifically amplified with SEN-V primers. In contrast, the FIG. 4. Persistence of SEN-V infection. Eleven patients with SENV–positive non–A to E (non-ABC) hepatitis and 20 patients with combined HCV and SEN-V infections were followed sufficiently long term to assess the rate of SEN-V clearance. All recipients were SEN-V negative before transfusion and SEN-V positive 4 to 8 weeks posttransfusion. Patients 1 to 17 (55%) cleared the virus within 6 months as defined by 2 consecutive negative PCR determinations without recurrence. Six additional patients (20 and 22-26) cleared the virus within 2 to 5 years for a 5-year clearance rate of 74%. Four (13%) patients (27 and 29-31) remained carriers for 4 to 12 years. Insufficient sampling was available to determine viral clearance in cases 18, 19, and 21. The clearance rate was similar in patients with SEN-V infection alone and those with combined SEN-V and HCV infection.
HEPATOLOGY Vol. 33, No. 5, 2001
FIG. 5. Detection of SEN-V in the liver of 2 patients with hepatocellular carcinoma. Nucleic acid was extracted from the hepatic tumor and from surrounding nontumorous tissue in both patients. The extract was treated with DNase to destroy viral and cellular DNA and the RNA was reversetranscribed and then amplified with SEN-V–specific primers. The cDNA from both tumor tissue and normal liver was specifically amplified whereas RNA was not amplified. Because DNA had been destroyed by DNase and RNA did not amplify, the restricted amplification of cDNA suggested that this was derived from an SEN-V RNA replicative intermediate.
RNA did not amplify. Since SEN-V is a DNA virus, the specific amplification of cDNA suggests that the liver contained a SEN-V RNA replicative intermediate. Since the RNA could not be amplified by the SEN-V primers used, it is unlikely that the positive reactions obtained in the cDNA fractions were due to residual SEN-V DNA in the RNA preparation. No statistical analysis was performed because only a single experiment was conducted. DISCUSSION
After the discovery of the hepatitis viruses A through E, several lines of evidence suggested that an additional hepatitis agent(s) might exist.4 By exclusion, these have tentatively been designated non–A to E hepatitis or sometimes non-ABC hepatitis. The evidence for the existence of one or more unrecognized hepatitis agents includes the observations that 10% to 20% of both transfusion-associated and communityacquired hepatitis cases test negative for all established hepatitis viruses,4,5 as do approximately 30% of cases of cryptogenic chronic liver disease and cirrhosis,24 most cases of hepatitis-associated aplastic anemia25 and a large proportion of fulminant hepatitis cases.26 Armed with these observations, many investigators, particularly those linked to industry, embarked on viral discovery programs. These led sequentially to the discoveries of GBV-C,7 HGV,6 and TTV.11 None of these agents proved to be the cause of non–A to E hepatitis. More recently, SEN-V was described16,17 and the TTV family was expanded to include SANBAN,27 YONBAN,28 and TUS01.29 Physical characterization and phylogenetic analysis now suggests that TTV, SANBAN, YONBAN, TUS01, and SEN-V are all members of the Circoviridae family, a group of small, single stranded, nonenveloped circular DNA viruses previously associated only with diseases in animals. These agents in the human circovirus family are highly diverse, having sequence differences that classify each member into subtypes and distinguish members into subfamilies showing sequence homology of only 40% to 60%. As the number of SEN-V variants and TTV variants expanded in independent investigations conducted in Japan,13,29,30 Italy,16 and the United States,12 it became apparent
UMEMURA ET AL.
1309
that there was overlap among these agents and that SEN-V and TTV and their widely divergent variants were members of what might be considered a super-family of human circoviruses.18 In this investigation, we studied 2 SEN-V variants designated SENV-D and SENV-H. A separate study18 suggested that SENV-D and SENV-H resemble TTV variants genotypes 12 and 16, respectively.29 SENV-D and SENV-H were chosen for additional study from among the 8 known variants of SEN-V because cloning and sequencing revealed their presence in the majority of transfusion-associated non–A to E hepatitis cases, and molecular screening showed them to have relatively low prevalence in the general population. The full investigation involved all our 13 known transfusion-associated non–A to E cases, 232 control patients who were transfused and did not develop hepatitis, 49 patients with transfusion-associated hepatitis C, 100 nontransfused controls, and 823 donors sampled before and after routine HCV screening was initiated. Eleven patients who were SEN-V positive before their index surgical procedure were excluded from analysis. SENV-D and SENV-H infections were of moderate prevalence among current volunteer blood donors (1.8%), and were not significantly higher in donors before HCV screening (pre1990). The prevalence was similar in hospitalized patients before transfusion (2.8%) suggesting that the general population prevalence is approximately 2% to 3%. In the United States, the prevalence of these SEN-V variants is at least 5 times higher than the prevalence of HCV viremia (0.3%) and hepatitis B surface antigenemia (0.1%) among blood donors before screening for these agents. In this study, nearly 15% of SEN-V–infected recipients became long-term carriers; 12 years of persistent infection was documented in 2 recipients. In contrast, 55% of infected subjects cleared the virus within 6 months and 74% within 5 years of onset. This suggests that the number of exposures to SEN-V considerably exceed the number of active infections detected by PCR. There is need to develop an assay to detect antibodies to SEN-V so as to estimate better the overall frequency and potential burden of this infection. Our findings strongly suggest that SEN-V is transmitted by transfusion. A significant association with blood transfusion was observed among patients who were prospectively followed after open-heart surgery; 30% of transfused patients were acutely infected with SENV-D and/or -H compared with 3% of patients who were not transfused (P ⬍ .001). In addition, there was a significant association between transfusion volume and SEN-V infections with the risk increasing stepwise to 45% in those who received 13 or more units of blood (P ⬍ .0001). Donor samples were available in only 6 of the 11 SEN-V– infected non–A to E hepatitis cases. An SEN-V positive donor was detected in 4 of these 6 cases (67%) and in a similar proportion of donors to SEN-V–infected recipients who did not develop non–A to E hepatitis. The failure to detect SEN-V in 100% of implicated donors probably reflected the generally low level of viremia in this infection and the potential degradation of nucleic acid to nondetectable levels in these samples that had been stored up to 20 years. However, one cannot rule out that the SEN-V infections in some of these cases were unrelated to transfusion because 3% of patients undergoing open-heart surgery were acutely infected with SEN-V variants in the perioperative period in the absence of transfusion. This suggests either nosocomial transmission of SEN-V, as has
1310 UMEMURA ET AL.
been previously reported for TTV31 or reactivation of latent SEN-V infection. In the absence of an established antibody assay for SEN-V, it was not possible to directly assess prior exposure and the potential for reactivation. The strongest evidence for transmission of SEN-V by blood transfusion derives from molecular linkage experiments. Cloning and sequencing of SEN-V was performed in 2 non–A to E cases. In both, virtually identical SEN-V variants were noted in the donor and recipient and, in 1 recipient 2 distinct SEN-V variants were transmitted by different donors. The most significant finding in this study was the strong association between acute infection with SEN-V and the development of transfusion-associated non–A to E hepatitis. Eleven of 12 patients (92%) with non–A to E hepatitis were SEN-V negative before transfusion and became SEN-V positive in temporal association with the onset of hepatitis. Further, the level of viremia tended to parallel the level of ALT. Although the observed association with transfusion-associated non–A to E hepatitis is striking, the association does not prove causality. There are several considerations that must be addressed before a causal relationship can be established. First, the number of non–A to E cases studied is small. We have tested all the prospectively followed cases at our hospital, but these cases are relatively rare compared with transfusion-associated hepatitis C, and the total number available was only 13, one of whom was preinfected with SEN-V and could not be analyzed. Hence, additional non–A to E hepatitis cases need to be studied, and it is essential that these findings be confirmed in other prospectively followed cohorts. Second, 24% of 225 patients who did not develop hepatitis were also acutely infected with SEN-V. Because the number of non–A to E hepatitis cases was small and the number of identically followed patients who did not develop hepatitis was large, it would appear that the vast majority of SEN-V–infected patients do not manifest biochemical or clinical evidence of hepatitis. Although the proportion of hepatitis cases who were infected with SEN-V (92%) was significantly (P ⬍ .001) greater than nonhepatitis controls (24%), the low overall percentage of SEN-V–infected patients who actually develop disease is unsettling in terms of establishing a causal association. Nonetheless, it is not unusual in virology that the majority of immunocompetent subjects who are infected with a given agent fail to manifest evident disease. Indeed, this phenomenon is quite common with cell-associated herpes viruses such as cytomegalovirus and Epstein-Barr virus and even common with hepatitis viruses A and B. Thus, whereas a lower background rate would have increased the likelihood of a causal association, the observed low frequency of hepatitis among SEN-V–infected subjects does not negate the possibility that SEN-V is the responsible agent in some patients. Third, before establishing SEN-V as a cause of non–A to E hepatitis, it is essential to prove that SEN-V is hepatotropic and replicates within hepatocytes. Such proof is currently lacking although preliminary evidence for intrahepatic replication of SEN-V was found in this study. In our study, liver tissue was obtained from 2 patients with hepatocellular carcinoma that were SEN-V DNA positive. Nucleic acid was extracted from the tumor and from the surrounding normal tissue. After DNA inactivation, the residual RNA was reverse transcribed and amplified using SEN-V–specific primers. SEN-V was detected in both tumor and nontumor tissue in both cases. Since this experiment detected RNA specific to a DNA virus, it is un-
HEPATOLOGY May 2001
likely to have been a blood contaminant. Because DNA had been destroyed, the experiment was interpreted to show the presence of an RNA replicative intermediate of SEN-V within liver tissue. Supporting intrahepatic localization is a recent report that the related TT virus has been found in the liver by in situ hybridization.32 These data are not sufficient to prove hepatotropism and intrahepatic replication, but if confirmed would lend further support to the hypothesis that SEN-V can cause hepatitis in some infected individuals. The sophistication and sensitivity of molecular approaches to viral discovery virtually ensures that new viruses will continue to be discovered. In the hepatitis arena alone, the past 5 years has seen the independent discovery of GBV-C and its variant, HGV and now the enlarging family of circoviruses that includes TTV, SANBAN, YONBAN, and SEN-V. Each time a novel sequence is discovered, a search for disease associations ensues. The process of establishing a causal relationship with viruses becomes very difficult when the prevalence of the agent is relatively high and the proportion of infected individuals who manifest diseases relatively low. Thus far, no disease association has been established for GBV-C/HGV or TTV. The current study suggests a disease association for SEN-V, but there are several caveats to this interpretation as described above. The sensitivity of molecular amplification and subtractive cloning techniques may uncover agents that are merely innocent bystanders, a “normal viral flora” similar to that found for bacteria. These organisms may not be pathogenic or may be pathogenic only under special circumstances such as an acquired immunodeficiency. Such molecular discoveries generally will not fulfill Koch’s postulates and thus there is need to establish new guidelines for causality to meet the exigencies of the molecular age. Where SEN-V will fall on the pathogenic scale is difficult to establish at this time, but it is unequivocal that SEN-V and the new “super-family” of human circoviruses are novel infectious agents that are readily transmitted by transfusion and that can lead to persistent infection. At issue is whether these agents cause non–A to E hepatitis or some extrahepatic disease and, if so, with what frequency and with what degree of clinical relevance. If causality is established, then secondary issues such as the need to screen the donor population will have to be addressed. It is premature to consider blood screening at this time. An additional consideration is that members of the human circovirus family are so diverse (only 40% to 60% sequence homology) that they might have very different pathogenicity and each phylogenetic grouping may have to be studied independently to assess clinical relevance. Such investigation of disease associations should not be limited to hepatitis. REFERENCES 1. Choo Q-L, Kuo G, Weiner AJ, Overby LR, Bradley DW, Houghton M. Isolation of a cDNA clone derived from a blood-borne non-A, non-B viral hepatitis genome. Science 1989;244:359-362. 2. Kuo G, Choo Q-L, Alter HJ, Gitnick GL, Redeker AG, Purcell RH, Miyamura T, et al. An assay for circulating antibodies to a major etiologic virus of human non-A, non-B hepatitis. Science 1989;244:362-364. 3. Alter HJ, Purcell RH, Shih JW, Melpolder JC, Houghton M, Choo Q-L, Kuo G. Detection of antibody to hepatitis C virus in prospectively followed transfusion recipients with acute and chronic non-A, non-B hepatitis. N Engl J Med 1989;321:1494-1500. 4. Alter HJ, Bradley DW. Non-A, non-B hepatitis unrelated to the hepatitis C virus (non-ABC). Semin Liver Dis 1995;15:110-120. 5. Alter MJ, Margolis HS, Krawczynski K, Judson FN, Mares A, Alexander WJ, Hu PY, et al. The natural history of community-acquired hepatitis C in the United States. N Engl J Med 1992;327:1899-1905.
HEPATOLOGY Vol. 33, No. 5, 2001 6. Linnen J, Wages J Jr, Zhang-Keck ZY, Fry KE, Krawczynski KZ, Alter H, Koonin E, et al. Molecular cloning and disease association of hepatitis G virus: a transfusion-transmissible agent. Science 1996;271:505-508. 7. Simons JN, Leary TP, Dawson GJ, Pilot-Matias TJ, Muerhoff AS, Schlauder GG, Desai SM, et al. Isolation of novel virus-like sequences associated with human hepatitis. Nat Med 1995;1:564-569. 8. Alter HJ, Nakatsuji Y, Melpolder J, Wages J, Wesley R, Shih JW-K, Kim JP. The incidence of transfusion-associated hepatitis G virus infection and its relation to liver disease. N Eng J Med 1997;336:747-754. 9. Tanaka E, Alter HJ, Nakatsuji Y, Shih JWK, Kim JP, Matsumoto A, Kobayashi M, et al. Effect of hepatitis G virus infection on chronic hepatitis C. Ann Intern Med 1996;125:740-743. 10. Alter HJ. G-pers creepers, where’d you get those papers? A reassessment of the literature on the hepatitis G virus. Transfusion 1997;37:569-572. 11. Nishizawa T, Okamoto H, Konisi K, Yoshizawa H, Miyakawa Y, Mayumi M. A novel DNA virus (TTV) associated with elevated transaminase levels in posttransfusion hepatitis of unknown etiology. Biochem Biophys Res Commun 1997;241:92-97. 12. Matsumoto A, Yeo AET, Shih JWK, Tanaka E, Kiyosawa K, Alter HJ. Transfusion-associated TT virus infection and its relationship to liver disease. HEPATOLOGY 1999;30:283-288. 13. Takahashi K, Hoshino H, Ohta Y, Yoshida N, Mishiro S. Very high prevalence of TT virus (TTV) infection in general population of Japan revealed by a new set of PCR primers. Hepatol Res 1998;12:233-239. 14. Simmonds P, Davidson F, Lycett C, Prescott LE, MacDonald DM, Ellender J, Yap PL, et al. Detection of a novel DNA virus (TT virus) in blood donors and blood products. Lancet 1998;352:191-195. 15. Naoumov NV, Petrova EP, Thomas MG, Williams R. Presence of a newly described human DNA virus (TTV) in patients with liver disease. Lancet 1998;352:195-197. 16. Primi D, Fiordalisi G, Mantero JL, Mattioli S, Sottini A, Bonelli F, Vaglini L, et al. Identification of SENV genotypes. International patent number WO0028039 (international application published under the patent cooperation treaty). Internet address: http://ep.espacenet.com/. 17. Sottini A, Mattioli S, Fiordalisi G, Mantero G, Imberti L, Moratto D, Primi D. Molecular and biological characterization of SEN viruses: a family of viruses remotely related to the original TTV isolates. Proceedings of the 10th International Symposium on Viral Hepatitis and Liver Disease. H. Margolis (Ed), Meditech Media, Atlanta. 2001 (in press). 18. Tanaka Y, Primi D, Wang RYH, Umemura T, Yeo AET, Mizokami M, Alter HJ, et al. Genomic and molecular evolutionary analysis of a newly identified infectious agent (SEN virus) and its relationship to the TT virus family. J Infec Dis 2001;183:359-367. 19. Alter HJ, Sanchez-Pescador R, Urdea MS, Wilber JC, Lagier RJ, Di Bisceglie AM, Shih JW, et al. Evaluation of branched DNA signal amplification for the detection of hepatitis C virus RNA. J Viral Hepat 1995;2:121-132.
UMEMURA ET AL.
1311
20. Ina Y. ODEN: a program package for molecular evolutionary analysis and database search of DNA and amino acid sequences. Comput Appl Biosci 1994;10:11-12. 21. Gojobori T, Ishii K, Nei M. Estimation of average number of nucleotide substitutions when the rate of substitution varies with nucleotide. J Mol Evol 1982;18:414-423. 22. Saitou N, Nei M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 1987;4:406-425. 23. Tsarev SA, Tsareva TS, Emerson SU, Kapikian AZ, Ticehurst J, London W, Purcell RH. ELISA for antibody to hepatitis E virus (HEV) based on complete open-reading frame-2 protein expressed in insect cells: identification of HEV infection in primates. J Infect Dis 1993;168:369-378. 24. Kodali VP, Gordon SC, Silverman AL, McCray DG. Cryptogenic liver disease in the United States: further evidence for non-A, non-B, and non-C hepatitis. Am J Gastroenterol 1994;89:1836-1839. 25. Brown KE, Tisdale J, Barrett AJ, Dunbar CE, Young NS. Hepatitis-associated aplastic anemia. N Engl J Med 1997;336:1059-1064. 26. Ferraz ML, Silva AE, Macdonald GA, Tsarev SA, Di Bisceglie AM, Lucey MR. Fulminant hepatitis in patients undergoing liver transplantation: evidence for a non-A, non-B, non-C, non-D, and non-E syndrome. Liver Transpl Surg 1996;2:60-66. 27. Hijikata M, Takahashi K, Mishiro S. Complete circular DNA genome of a TT virus variant (isolate name SANBAN) and 44 partial ORF2 sequences implicating a great degree of diversity beyond genotypes. Virology 1999; 260:17-22. 28. Takahashi K, Hijikata M, Samokhvalov EI, Mishiro S. Full or near full length nucleotide sequences of TT virus variants (type SANBAN and YONBAN) and the TT virus-like mini virus. Intervirology 2000;43:119123. 29. Okamoto H, Takahashi M, Nishizawa T, Ukita M, Fukuda M, Tsuda F, Miyakawa Y, et al. Marked genomic heterogeneity and frequent mixed infection of TT virus demonstrated by PCR with primers from coding and noncoding regions. Virology 1999;259:428-436. 30. Umemura T, Yeo AET, Shih JWK, Matsumoto A, Orii K, Tanaka E, Kiyosawa K, et al. The prevalence of SEN virus infection in Japanese patients with viral hepatitis and liver disease [Abstract]. HEPATOLOGY 2000; 32(suppl):381A. 31. Okamoto H, Akahane Y, Ukita M, Fukuda M, Tsuda F, Miyakawa Y, Mayumi M. Fecal excretion of a nonenveloped DNA virus (TTV) associated with posttransfusion non-A-G hepatitis. J Med Virol 1998;56:128132. 32. Rodriguez-In˜igo E, Casqueiro M, Bartolome´ J, Oritz-Movilla N, Lo´pezAlcorocho JM, Herrero M, Manzarbeitia F, et al. Detection of TT virus DNA in liver biopsies by in situ hybridization. Am J Pathol 2000;156: 1227-1234.