Molecular cloning and sequencing of the mexico isolate of hepatitis E virus (HEV

VIROLOGY 191, 550-558 (1992)

Molecular Cloning and Sequencing of the Mexico Isolate of Hepatitis E Virus (HEV)

CHIAO-CHAIN HUANG,**’ DAT NGUYEN,* JOHN FERNANDEZ,* KARYN Y. YUN,* KIRK E. FRY,* DANIEL W. BRADLEY,t ALBERT W. TAM,* AND GREGORY R. REYES*”

*Molecular Virology Group, Genelabs Incorporated, Redwood City, California 94063, and tHepatitis Branch, Division of Viral Disease, Center for Disease Control, Atlanta, Georgia 30333

Received June 3, 1992; accepted July 3 1, 1992

Hepatitis E virus (HEV) is the major causative agent of hepatitis E or what was formerly known as enterically transmitted non-A, non-B hepatitis. The disease has a worldwide distribution but occurs principally in developing countries in any of three forms: large epidemics, smaller outbreaks, or sporadic infections. Genetic variation of different HEV strains was previously noted and it will be important to determine the extent to which this variation may pose problems in the diagnosis and treatment of HEV infection. To analyze differences at the genetic level between HEV(Mexico; M) and the previously characterized HEV(Burma; B) and HEWPakistan; P) isolates, overlapping cDNAs were cloned from samples obtained from an infected human and an experimentally inoculated cynomolgus macaque. These cDNA clones, representing the nearly complete (7185-bp) genome of HEV(M), confirmed an expression strategy for the virus that involves the use of 3 forward open reading frames (ORFs). The HEV(M) strain has an overall 76 and 77% nucleic acid identity with the HEV(B) strain and HEV(P) strain, respectively; however, the degree of sequence variation was not uniform throughout the viral genome. A hypervariable region was identified in ORFl that exhibited a 58 and 54% nucleic acid sequence and 13% amino acid similarity with the Burma strain and the Pakistan strain, respectively. A large number of the nucleotide differences occurred at the third codon position, with the deduced amino acid sequences similarity of 83, 93, and 87% between HEV(M) and HEV(B) isolates in ORFl, ORF2, and ORF3, respectively, and with 84, 93, and 87% amino acid identities between HEV(M) and HEV(P) isolates in ORFl, ORF2, and ORF3, respectively. The nucleotide sequences derived from the highly conserved regions of HEV genome will be useful in developing polymerase chain reaction-based tests to confirm the viral infection. Knowledge of the extent of the sequence variation encountered with HEV will not only aid in the future development of diagnostic and vaccine reagents but also further our understanding of how HEV strain variation might impact the pathological outcome of infection. 0 1992 Academic Press, Inc.

INTRODUCTION

The term non-A, non-B (NANB) hepatitis came into use with the development of diagnostic tests for hepatitis A and hepatitis B infection. It soon became clear that there were two distinct types of NANB hepatitis that could be recognized with different epidemiological characteristics and pathological sequelae (Bradley, 1990b; Reyes and Baroudy, 1991). The first, mainly described in studies from North America and western Europe, is transmitted via blood and blood products and frequently results in chronic hepatitis. This paren- terally transmitted virus has recently been cloned and termed hepatitis C virus (HCV) (Choo er al., 1989). The second type of NANB hepatitis was referred to as the epidemic, waterborne, or enterically transmitted variety that, as its name implies, is transmitted by the fecal-oral route (Balayan et a/., 1983; Bradley and

’ To whom reprint requests should be addressed at Genelabs In- corporated, 505 Penobscot Drive, Redwood City, CA 94063.

’ Current address: Triplex Pharmaceuticals, 9391 Grogan’s Mill Road, The Woodlands, TX 77380.

Balayan, 1988; Bradley, 1990a; Khuroo, 1980; Wong et a/., 1980). Epidemiologically, this epidemic NANB hepatitis resembles hepatitis A but is distinctly different in the induction of fulminant hepatitis and resulting high mortality (10 to 20%) observed in pregnant women (Purcell and Ticehurst, 1988). The cause of this particular severe consequence of enterically transmitted NANB infection is unclear. Furthermore, like hepatitis A, no evidence of chronic disease is associated with this epidemic NANB hepatitis.

Recently, we have isolated a portion of the Burma strain of the virus, which we have termed hepatitis E virus (HEV), responsible for enterically transmitted NANB hepatitis (ET-NANBH) (Reyes et al., 1990a,b; Reyes, 1991). Utilizing the originally isolated clone, ETl-1, it was determined that the HEV genome con- sisted of a positive-sense, single-stranded, polyadenylated RNA molecule of approximately 7.5 kb (Reyes er a/., 1990a,b; Reyes, 1991). The initial characterization of the full-length genome of the Burma strain indicated that this agent was quite unlike any of the previously identified positive-sense RNA viruses of man (Tam et a/., 1991 a). More recently, a strain of HEV from Pakis-

0042.6822/92 $5.00 550

CopyrIght 0 1992 by Academic Press, Inc All rights of reptoducrrx I” any form reserved.

HEV GENOME STRUCTURE 551

tan [HEV(P)] has been molecularly cloned and sequenced (Tsarev et al., 1992). The nucleotide sequence of the Pakistan strain is highly related to that of the Burma strain except for a hypervariable region.

Studies with the first molecular clone ET1 .l , revealed that the majority of the enterically transmitted NANB hepatitis resulted from a group of agents similar to if not identical to the hepatitis E virus. It was therefore appropriate to refer to the disease as hepatitis E since it was now possible to differentiate hepatitis by both clinical and molecular criteria. Strain variation between viral isolates originating from Mexico and Burma [HEV(M) and HEV(B)], however, was first suggested when clone ETl-1 was used as a probe against RNA from the liver of a cynomolgus macaque (cyno) experimentally infected with HEV(M). The probe failed to de- tect the 7.5-kb viral transcript under standard (strin- gent) hybridization and wash conditions (Huang et al., unpublished data). The strain variation in the HEV genome may impose a serious problem in the diagnosis and treatment of HEV infection. It is therefore of biological and clinical importance to analyze and determine the sequence diversity in the viral genome of HEV(M) and HEV(B).

In this work, we have isolated and sequenced cDNA clones representing a nearly 7.2 kb viral genome of the HEV(M). Our results allow a direct sequence comparison between HEV(M), HEV(B), and HEV(P) revealing that HEV(B) is more closely related to HEV(P) than to HEV(M); permit a description of the genomic organiza- tion of HEV; confirm an expression strategy that involves multiple open reading frames; and confirm that HEV is the principle viral agent responsible for epidemic outbreaks of ET-NANBH in three geographically distinct regions (i.e., Mexico, Burma, and Pakistan).

MATERIALS AND METHODS

cDNA library construction

Three different HEV(M) cDNA libraries were prepared for cDNA clone isolation. The first cDNA library was made with RNA extracted from a human stool sam- ple collected during an HEV outbreak in Mexico (Brad- ley et a/., 1987). Briefly, a 10% fecal suspension was clarified by centrifugation and extracted for viral RNA using phenol:chloroform (1: 1). RNAwasethanol precip- itated prior to cDNA synthesis by random priming of first strands using a cDNA synthesis kit (Boehringer Mannheim Biochemicals). The blunt-ended cDNAs were modified for sequence-independent single primer amplification (SISPA) by the ligation of AB linker/ primers (Reyes et al., 1990a; Reyes and Kim, 1991). After Nrul cleavage (to remove any linker/primer dimers), an aliquot of the reaction was subjected to 30

cycles of Taq polymerase-mediated amplification (Saiki et a/., 1988). The reaction conditions used for melting (94”, 1 min), annealing (50”, 2 min), and extension (72”, 6 min) included 1 PA/I of “A” primer (5’- GGAATTCGCGGCCGCTCG-3’). After amplification, ali- quots of the SISPA cDNA were treated with EcoRI, column chromatographed on sephacryl S-300 column, concentrated by ethanol precipitation, and then ligated into lambda gtl0 phage vector.

Two additional cDNA libraries were also made using poly-(A)+ RNA isolated from the liver of a cynomolgus macaque (cyno) experimentally infected with HEV(M) (Bradley et a/., 1987). Briefly, total RNA was prepared from cyno liver tissue by solubilizing in 5 M guanidin- ium isothiocyanate followed by lithium chloride precipitation (Cathala et a/., 1983). After ethanol precipitation of total RNA, polyadenylated RNA was isolated using oligo-dT cellulose column chromatography and then converted to cDNA by random priming or oligo-dT priming of first strand. The resulting cDNAs were ligated into the EcoRl site of lambda gtl0 phage vector.

cDNA library screening

The phage libraries were screened by molecular hybridization using 32P-labeled DNA probes. Hybridiza- tion was performed in 50% formamide, 3 X SSC (1 x

SSC is 0.15 M sodium chloride, 0.015 M sodium ci- trate, pH 7.0) 3X Denhardt’s solution (1 X Denhardt’s solution is 0.2 mg/mI Ficoll, 0.2 mg/ml polyvinylpyrroli- done, 0.2 mg/ml bovine serum albumin), 50 mM HEPES (pH 7.0), and 200 pg/ml salmon sperm DNA at 42” for 12 hr. The filters were rinsed in 2~ SSC and 0.1% SDS at 25” for 5 min with one change of washing solution and then rinsed in 0.1 X SSC and 0.1 O/O SDS at 42” for 1 hr. Positive plaques were visualized by autora- diography.

Sequencing DNA

After isolating from the phage libraries, the HEV cDNA clones were subcloned into Bluescript plasmid vector and then sequenced with the chain-terminator technique (Sanger et a/., 1977). All portions of the reported sequence were obtained from both strands of the cloned DNAs.

Nucleotide sequence accession number

The nucleotide sequence data reported here will appear in the Genbank and EMBL nucleotide sequence data base under Accession Number M74506.

RESULTS

Molecular cloning of HEV(M)

We have identified by immunoscreening of a cDNA expression library two HEV(M) clones which encode

552 HUANG ET AL.

5’ -7.5 kb 3

I I AAAAA

135-3-6 ; 135-3-2 406-4-2 406-3-2

MRP-13-3V MRP-14-l

MRP-3-2 3LC1

MRP-4-25-1 liiGPAAAAA . -

MRP-57

b ORF2

- ORF3

FIG. 1. A schematic representation of HEV(M) cDNA clones and the three forward open reading frames (ORFs) of the virus. The relative position and overlap of the various clones along the -7.5.kb viral genome [including the 3’ stretch of adenosine residues, of which there are at least 150-200 in HEV(B)] are indicated. The MDT-l 2-1 clone contained a stretch of adenosine residues at its 3’ end, suggesting its location at the 3’terminus of the viral genome of HEV(M). Also shown are the two clones encoding epitopes, 406-4-2 and 406-3-2, which were previously identified by us (Yarbough et al., 1991a,b).

epitopes specifically recognized by antibodies from human and cynomolgus macaques infected with HEV (Reyes et a/., 1991 b; Skidmore et al., 1991; Yarbough et a/., 1991a,b). The discovery of these two HEV(M) clones (406-3-2 and 406-4-2) and the originally isolated HEV(B) clone (i.e., ET l-l) (Reyes et a/., 1990a,b) has allowed us to employ two different strategies to clone the viral genome of HEV(M).

The first cloning strategy is the use of ET l-l as a probe under reduced hybridization stringency, to isolate related HEV(M) clones. Two overlapping cDNA clones, 135-3-2 and 135-3-6 (Fig. l), were isolated from the SISPA amplified, random-primed stool cDNA library (see Materials and Methods). The end sequences of clones 135-3-2 and 135-3-6 were then syn- thesized for use as probes to screen a random-primed cyno liver cDNA library to yield a set of overlapping clones, two of which are shown in Fig. 1 (MRP-13-3V and MRP-14-l). Using this end-probe strategy, we obtained an ordered set of overlapping cDNA clones representing nearly the entire 5’ region of HEV(M), as illus- trated in Fig. 1.

Since several lines of evidence suggest that the immunoreactive clone 406-3-2 is located near the 3’ end of HEV(M), we used this clone as a probe to screen an oligo-dT-primed cyno liver cDNA library, yielding a clone MDT-l 2-l (Fig. 1). This cDNAclone possessed a

stretch of 15 adenosine residues at its 3’ end, as revealed by DNA sequencing (see below). On the basis of sequence comparison with HEV(B), we believe that MDT-l 2-l contains the very 3’ end of HEV(M).

The second epitope clone 406-4-2 was found to have homology with a sequence from HEV(B) strain that was located approximately 160 1 nucleotides 5’ to the 406-3-2 clone. The positioning of this second epitope clone gave rise to an alternative cloning strategy which was to use the polymerase chain reaction (PCR) to amplify the intervening genetic segment linking the two immunoreactive clones, 406-3-2 and 406-4-2. As shown in Fig. 1, a cDNA clone, 3LC1, was obtained by this approach. The details of the epitope-spanning approach is described elsewhere (Huang et al., unpublished data).

By these various procedures, a set of nine overlapping cDNA clones was isolated, representing an approximately 7.2 kb viral genome of HEV(M) (see below). A listing of these clones, their derivation, and the method of isolation are noted in Table 1.

Nucleotide sequence of HEV(M)

We sequenced the nine overlapping HEV(M) cDNA clones and assembled the resulting DNA sequences into a composite linear nucleotide sequence represent-


TABLE 1

HEV(M) RECOMBINANT cDNA CLONES

Clone Size (kb) Source@ Methodb

135-3-6 0.9 135-3-2 1.6 MRP-14-l 0.85 3LCl 1.3 MDT-12-l 0.93 MRP-13.3V 0.96 MRP-3-2 1.2 MRP-4-25-1 1.35 MRP-5-7 0.56

Feceslhulrp Feces/hu/rp Liverlcyfrp Liver/cy/dT Liver/cy/dT Liverlcylrp Liverlcyfrp Liver/cy/rp Liver/cy/rp

hyb/ETl .l hyb/ETl .l hyb/l35-3-2 PCR hyb/406.3-2 hyb/l35-3-6 hyb/MRP-13-3V hyb/MRP-3-2 hyb/MRP-4-25-l

a Biological source/species/priming of cDNA library; hu, human; cy, cynomolgus macaque; priming of cDNA by random primers (rp) or oligo dT (dT).

b Hybridization (hyb)/probe; PCR, polymerase chain reaction.

ing the nearly complete genome of HEV(M) (see below). The nucleotide sequence data have been depos- ited in the Genbank under Accession Number M74506. The nucleotide sequence is 7185 nucleotides including the 15 adenosine residues located at the very 3’ end. If the 3’ stretch of adenosine residues [at least 150-200 in HEV(B)] is included, the determined sequence is in agreement with the genome size previously estimated by Northern blot analysis (Reyes et a/., 1990). The percentage G + C content of the virus is 58% and is therefore clearly distinguished from that of HAV (37%; [Cohen et al., 19871). A survey of the GenBank version 72 nucleotide sequence library did not show similarity to any entries (except those noted here) when searched by FASTA program (Pearson and Lipman, 1988), indicating that HEV(M) is a unique viral entity.

Open reading frame analysis

Using the COD RNY sequence analysis program (In- telligenetics, Mountain View, CA), the 7.2 kb of HEV(M) sequence was analyzed for open reading frames (ORFs). Two large ORFs were found in the first and second reading frames in the positive polarity. As previously reported, the positive polarity was determined by strand-specific hybridization of genomic RNA extrated from virions present in fecal preparations (Reyes et a/., 1990a, 1991 a) and also by the existence of consensus motifs for nonstructural gene sequences (see below). The first large ORF begins with the very 5’ end of the viral genome and extends over 5 kb before termination at nucleotide position 5077 (Fig. 1). The second large ORF begins at nucleotide position 51 17, extends about 2 kb, and terminates at nucleotide position 7094. In addition, there also exists a 369 nucleotide

long ORF located from nucleotide position 5076 to position 5445 in the third open reading frame (Fig. 1).

The deduced amino acid sequences of these three ORFs are shown in Fig. 2. Inspection of the amino acid sequence of ORFl reveals that a portion of this reading frame appears to encode the RNA-dependent RNA polymerase of HEV. The canonical gly-asp-asp (GDD) tripeptide found in all positive-stranded animal and plant RNA viruses (Kamer and Argos, 1984) can be located in the deduced amino acid sequence. In addition, ORFl also was found to contain conserved motifs (GVPGSGKS and DEAP) associated with purine NTPase activity found in a variety of cellular and viral helicases (Gorbalenya et al., 1989). These helicases promote the unwinding of DNA, RNA, or RNA-DNA duplexes required for genome replication, transcrip- tion, recombination and repair (Gorbalenya et al., 1989). Both genetic elements are located in the exact same locations in the genome of the HEV(B) strain and HEV(P) strain.

The translation of ORF2 indicated a novel polypep- tide not present in the PIR protein database. This poly- peptide contains a high basic amino acid content similar to that seen with other viral capsid proteins (Tam et al., 1991 a), suggesting that it may be involved in the encapsidation of the genomic RNA by effectively neu- tralizing the negatively charged RNA. The second major epitope-encoding clone noted above (406-3-2) is located at the extreme carboxy terminus of ORF2. This reading frame would therefore appear to be the major ORF encoding the viral structural protein(s).

The 5’ portion of the HEV genome would therefore appear to encode the viral nonstructural and the 3’ portion to the viral structure gene(s). We have previously identified two broadly immunoreactive peptides by immunoscreening of an HEV(M) lambda gtl 1 expression library (Yarbough et a/., 1991 a). The data on antigen identification confirm the utilization of ORF2 and ORF3 and indicate that HEV utilizes all three positive-polarity reading frames to encode viral proteins.

Comparison of HEV(M), HEV(B), and HEV(P)

We have recently isolated and sequenced a contigu- ous set of overlapping cDNA clones representing the entire viral genome of HEV(B) (Tam et al., 1991 a,b). In that report, the 5’ end of the genome was established by a primer extension experiment and the 3’end of the genome was determined by the identification of a cDNA clone containing a long poly-(A) stretch of about 150-200 adenosine residues. Furthermore, the nearly entire genome of HEV(P) has been molecularly cloned and sequenced (Tsarev et al., 1992). We aligned the three nucleotide sequences (data not shown) and the

HUANG ET AL.

C Deduced amino acid sequence -. ORF3

*-Y EN-S

tel?ayF-SPPCALG‘FCCCSSCFCLCCPF.HRPvsRLAAw~AAAv

HEVY-P S ::

FIG. 2. The deduced amino acid sequence of HEV(M) and comparison to the predicted amino acid sequences of HEV(B) and HEV(P). The deduced amino acid sequences of HEV(M), HEV(B), and HEV(P) were aligned for comparison. The amino acids of the viral proteins encoded by HEV(B) and HEV(P) are shown only where they are different from the corresponding sites of the proteins encoded by HEV(M). Within open reading frame 1 (ORFl), the conserved motifs involved in purine NTPase activity found in a variety of cellular and viral helicases are highlighted. The canonical GDD (gly-asp-asp) tripeptide sequence found in all viral RNA-dependent RNA polymerase is marked with asterisks.

predicted amino acid sequences (Fig. 2) of HEV(M), HEV(B), and HEV(P) to analyze the molecular differences between the three viruses.‘These alignments reveal several pertinent points.

(1) Although the DNA sequences of HEV(M), HEV(B), and HEV(P) differed by about 24%, the rate of sequence variation is not uniform throughout the whole viral genome. The highest degree of relatedness is present at the extreme 5’ end of the genome (Fig. 3A). This same pattern of conservation has been observed with other RNA viruses such as polio (Svitkin et al.,

1990) HCV (Han et a/., 1991) and HAV (Emerson et al., 1991). It has been hypothesized that evolutionary constraints are placed on this portion of the genome by the role it plays in replication and translation of the virus. In the case of HAV, mutations in the 5’ region have been shown to play an important role in the adaptation of the virus to growth in tissue culture cells (Emerson et al., 1991).

Another indication of the structural importance of this region is suggested by examining the potential secondary structures found in the first 276 nucleotides of


A

BBV-X BE?-8 m-?-P

BBV-X Hw-B BBV-P

EBV-Y BBV-B BBV-P

BBV-M BE-?-B BBV-P

BBV-X HBV-B HBV-P

BBV-H EBV-B HBV-P

B

ACTCCGCCCTTGCGAATGCTGTGGTGGTCCGGCCmCCmC ACTCTOCCCTOGCGAATOCTAGTTAGGCCTTTTCTCTCTCAC~G ACTCTGCCCTTGCGAATGCTGGTAGTTAGGCCTTTTCTCTCTCACCAG *et* l **** l *tttt**tttttt tt ttttttt l * l * l * l **

CAGGTR?AGATCCTTATWAT~CATGCAACCTCGGCAGCTGGTG’l.l”K!G CAGATEAGATCCTCATPAACCTAA’IWJAACCTCGCCAG~~CCG CAGATIGAGATCCTTA’lTAACCTAATGCAACCTCGCCAGCTIGTl.FKCG tt* tttttttttt tt tt ** tttt***tttt ttttt tt tt tt

TCCn;AGGTTTTTEGMaTCACCCGATTCAACGTGTTATACATAATGAGC CCCCGAGGTTTTCTGGAATCATCCCATCCAGCGTGTCATCCATAACGAGC CCCCGAGGTTTTCTGGAACCATCCCATCCAGCGTGTTATCCATAATGAGC

** l ******* l **** tt l * t* tt ttttt tt *t**t tttt

TTGAGCAGTATTGCCGTGCTCGCTCGGGTCGCTGCCTTGAGATTGGAGCC TGGAGCmACTOCCGCGCCCGCTCCGGCCGCTGT~AAA-CGCC TGGAGCTITACTGTCGCGCCCGCTCCGGCCGCTGCCTCGAPJCC l tttt tt tt tt tt **tt* l * *t**t l * l * ttt*t l **

o--c /c-“--*-A-c,

c .U/C--G-C..Cy G-C- a--c’C A-.--u A -u-o,, U-A-0’ “-A A--u

/“-A\ y-c-0

A II u--o--c

‘icmol’ A-” c--o C-G A-” .-c-r, A---u.A c---o G--C

s n/G-c\n-

n-v-7 .*\ c--o

26 50 50

76 100 100

126 150 150

176 200 200

226 250 250

276 300 300

p--G- u-u-u,

2 G-“-c--o-“-u Y

c--o

FIG. 3. Nucleotide sequence comparison and potential secondary structures of HEV at the 5’ end. The extreme 5’ end of the genome of the three HEV isolates were aligned for comparison (A). Nucleotide sequences that are conserved among the three strains are indicated by asterisks. Since the 5’end of HEV(6) is 24 nucleotides longerthan that of HEV(M), these 24 nucleotides were added to the HEV(M) sequence to perform the secondary structure analysis (B).

the 5’end of HEV(M) as predicted by computer modeling. The algorithm of Zucker (Jacobson et a/., 1984) which is based on the minimization of free energy, pro- duced the model shown in Fig. 3B. Similar RNA secondary structures can also be found in the corresponding regions of HEV(B) and HEV(P) (data not shown).

This model is just one of many possible structures which can be predicted as there are several different algorithms for RNA secondary structure and even these do not take into account the potential interac- tions of the proteins that interact with the proposed structures (Major eta/., 1991; Zuker eta/., 1991). Mod- els of the analogous region of HAV also show similar amounts of secondary structure minimization of free energy.

The 5’ end of HEV(B) is 24 nucleotides longer than that of HEV(M) (Fig. 3A). Although we had extensively screened specifically primed cDNA libraries, no cDNA clones containing sequences that extended the current 5’ end of HEV(M) were isolated. It is conceiv- able that the current nucleotide sequence of HEV(M) is lacking the 24 nucleotides relative to that of HEV(B). However, it is also possible that the geographic vari- ants of HEV contain slightly different sequences at their extreme 5’ ends. Future studies of other HEV strains may help resolve this issue. A clear difference between the HEV(M) and HEV(B) strains was detected in the 3’ untranslated region (nucleotide position 7097 to position 7170). Within this region, HEV(M) contained 9 more nucleotides than HEV(B).

(2) Whereas the intergenic region between ORFl and ORF2 (nucleotide position 5080 to position 51 16) is 920/, identical between HEV(M), HEV(B), and HEV(P), the overall nucleotide homology decreases to about 679/o in the region located between nucleotide position 1500 and position 3000. In particular, we identified a hyper-variable region (position 2121 to position 2329)

XXV-X TAACCCAlTTTGCGGCGAGAGCAffiCTCTACACCCGCAC~TCCACA4 m-8 TAATCCA~CTGTGGCGRGAGCACAC~ACACCCGTAC-T~~A~ EEV-P TAATCCA~CTGTGGCGAGAGCACRCmACACCCG~C-TC~~

l ** t*t*t l * tttt*t****t l * l ******* tttttttt

BBV-X ~A-~CRCCCTTAACTGTC~CT~~C~T~~A~- - HEV-B TXATG-CCGTCTCTAGTC---CAGCCCGG---CCEACTTAGGTmAT BEV-P ~A~-CTGTTCCTAGTC---CAGCCCAG---CCCGACTTAGGTmAC

l * t t tt l l l l ***et t *

BBV-x --cTGcTcc--ccAcTc-- -CCACcTGerAepGCCAC~- -GCC REV-8 GTCTGAGCClTCTATACCTAGTAGWCCGCCA-CGCCTACCmCGGCC EBV-P ATCl-GAGCCTTCTATACCTAGTAGCGCCGCCA-CACCTACCCCGGCGGCC

t*t l * t t t l ttttt tt t tt l * * t**

mm-x ~~TROGCTCTCTC~ACC~ACCCTGACCCG W-8 C--CTCTACCCCCCCCE---CACCGGACCCITCCCCCCCCCT~CC BE'.'-P C--CTCTACCCCCCCCTG---CACCGGATCCTTCCCCTACTCTCTCWCT

l tt l * l t *t* t l * l * ttt ttt t tt

EBV-X ACAGATGGCTCACGC----CCCTC-iG-GGGCCCGTCCGGCTGG-CCCCAA m-8 ~~----G~G~~~~~~~~~~TTc~~~~cTAccGccGGGGccccGG BEY-P CC----GGCGOXGGTGAGCCGGCTCCTGGCGCTACCGCCAGGGCC

t ttt . * l * l * ttt t ttt l l * *t**

REV-M CC------CGAAlGGCGTTCCG-CAGCGCCGCTTACTACACACCTACCCT REV-8 CCATAACTCACCAGACGGCCCGGCACCGC~C~ CTCITCACCTACCCG BEV-P CCATAACCCACCAGA’XGCCCGGCATCGCCGCCTGCTCPrCCCG

l * l l t* l ** ** tttttt t tt tttt***t

--I GACGGCGCTAAGATCTATGTCGGCTCCA~CGAGTC REV-8 GATGGCTCTARGGTA~CGC~C~G~~G~~A~~~~~ Bw-P GATGGCTCTAAGGTGTTCGCCGGCTCGCTG~AGTCGACA~TACCTG

tt l ** l **.* * l l l ***** * l * l **** l * l * tt

BBV-M GClTGTCAACGCATCT?ACGCCGGCCACCGCCCTGGTGGCGGGCTMGTC XB”-8 GCTffi~~CGffiTC~ACCACCGCCCTGGCGCC EEL’-P GCTCGTTAACGCGTC~ACCACCGCCCTGGCGGTGGG~~TC

l ** l * l **** l **.* t l tttt*t+ttttt t, tttt* l * +

2125 2149 2149

2173 2192 2192

2215 2241 2241

2265 2286 2286

2309 2332 2332

2352 2382 2382

2402 2432 2432

2452 2482 2482

FIG. 4. Nucleotide sequence comparison of the three HEV isolates in the hypervariable region. The conserved nucleotide sequences are marked by asterisks.

556 HUANG ET AL.

TABLE 2

COMPARISON OF THE THREE OPEN READING FRAMES FOR MEXICO, BURMA, AND PAKISTAN STRAINS OF HEV

Identity (%)

Region

ORFlb

ORF2

ORF3

Size (nt)

5073/5079/5079

1977/l 980/l 980

369/369/369

HEV(M) HEV(P) HEV(M) HEV(P) HEV(M) HEV(P)

Nucleotides Amino acids

HEV(B) HEV(P) HEV(B) HEV(P)

74 75 83 84 93 100 99 100 81 81 93 93 94 100 99 100 90 91 87 87 99 100 100 100

a HEV(M)/HEV(B)/HEV(P); nt, nucleotides. * ORF, open reading frame.

that exhibited a high degree of sequence diversity [only 58% sequence similarity between HEV(M) and HEV(B) and 54% similarity between HEV(M) and HEV(P)]. In this hypervariable region, there were a number of gaps ranging from one to six nucleotides besides the exten- sive single nucleotide changes in the nucleotide sequences of the two viruses (Fig. 4). As a result of these genetic variations, there exists a stretch of 69 amino acids that are highly diverged between the three HEV viruses (Fig. 2). There is a net decrease of 2 amino acid residues encoded in the ORFl of HEV(M) compared to that observed in the other two isolates. The role for this region is unknown. Similarly, the protein encoded by the ORF2 of HEV(M) is one amino acid shorter than the protein encoded by HEV(B) or HEV(P).

(3) Whereas the DNA sequences of HEV(M) and HEV(B) differed by approximately 24% their predicted amino acid sequences were 83% identical in ORFl, 93% identical in ORF2, and 87% identical in ORF3. The predicted sizes and comparison of these ORFs for HEV(B) and HEV(M) is presented in Table 2. A large number of the nucleotide differences occurred at the third position of codons, reflecting the functional conservation of the encoded peptides. The ORFs for HEV(P) were also included in Table 2 for comparison. It appeared that HEV(B) is more closely related to HEV(P) than to HEV(M).

(4) Open reading frame analysis of the nucleotide sequence of HEV(M) revealed that the genomic organi- zation (5’-nonstructural/3’-structural) and the expression strategy (utilizing three different ORFs) are identical to those seen with HEV(B) (Tam et a/., 1991 a,b).

DISCUSSION

Enterically transmitted non-A, non-B hepatitis (ET- NANBH) is probably an old disease that was only re-

cently discovered (Purcell and Ticehurst, 1988). Epi- demics and suspected epidemics of this hepatitis have been recognized in Asia, Africa, the Middle East, Mex- ico, and Central America (Bradley and Maynard, 1986; Bradley et al., 1987; Bradley, 1990a; Purcell and Tice- hurst, 1988). In these areas, contaminated water appeared to be the vehicle of transmission and illness occurred principally in young adults. A putative etiologic agent(s) for ET-NANBH has been identified (Brad- ley and Maynard, 1986; Bradley et a/., 1987; Bradley, 1990a; Purcell and Ticehurst, 1988). The causative agent(s) is serologically distinct from all other recognized human hepatitis viruses and morphologically distinct from all the viruses that have been visualized, including HAV (Purcell and Ticehurst, 1988). The true geographic distribution of the etiologic agent(s) is not entirely clear. Immune electron microscopy studies have found that most, if not all, of ET-NANBH from different regions and different years appears to have been caused by serologically related agents (Purcell and Ticehurst, 1988). We have recently identified and cloned the Burma isolate of the virus, which we have termed hepatitis E virus (HEV), responsible for ET- NANBH (Reyes et al., 1990a; Tam et al., 1991a). The nearly entire genome of the Pakistan strain of HEV has also been cloned and sequenced (Tsarev et al., 1992). In this work, we have cloned and sequenced the Mex- ico strain of HEV [HEV(M)]. The results of the DNA sequencing allow us to compare the three different geographic isolates of HEV at the molecular level, illus- trating that the Burma, Pakistan, and Mexico strains of HEV are closely related. Furthermore, we have recently reported that a gene segment from the Burma strain of HEV [HEV(B)] specifically identified similar sequences in cDNA prepared from infected human fecal samples collected from five geographically distinct ET-NANBH outbreaks (Reyes et al., 1990a). Taken together, our


data strongly suggest that HEV is the principal causative agent responsible for the epidemic outbreaks of ET-NANBH that occurred worldwide.

Although HEV(M), HEV(B), and HEV(P) are closely related, their viral genomes show a considerable degree of sequence diversity as is common in other eu- karyotic viruses with RNA genomes. However, the pattern of nucleotide sequence variation is not consistent over the whole viral genome. For instance, while the 5’ end and the 3’ end of HEV are well conserved (about 85% identical), the gene segment located from nucleotide position 1500 to position 3000 is much less conserved than other gene segments of HEV (e.g., see Fig. 4). This hypervariable region is contained within the open reading frame 1 (Fig. 2) and is believed to encode nonstructural gene products of HEV (Tam et al., 1991 a). Since the nucleotide sequences in this region differ substantially including gaps and single nucleotide changes, its predicted amino acid sequence ex- hibits more variation than the amino acid sequences encoded by other gene segments of HEV (Fig. 2). The biological significance of the nucleotide sequence variation seen in the HEV genome will not be fully under- stood until the various viral encoded proteins are bio- chemically defined and characterized. However, nucleotide sequences derived from the highly conserved regions may be useful in developing PCR-based confir- matory tests of HEV infection.

Open reading frame analysis of the nucleotide sequence of HEV(M) revealed two large and one small positive-polarity ORFs (Fig. 2). The largest ORF (ORFl) extended approximately 5 kb from the 5’ end and contained the RNA-dependent RNA polymerase and a heli- case-like region. The second large open reading frame (ORF2) began 37 nucleotides downstream of ORFl and extended about 2 kb to the termination codon located 76 nucleotides from the 3’terminal stretch of 15 adenosine residues. This reading frame appears to encode a capsid-like peptide. A third open reading frame (ORF3) partially overlaps ORFl and ORF2 and contains 369 nucleotides. A previously identified HEV antigen (Yarbough et al., 1991 a,b) was found in the ORF3, es- tablishing the utilization of this open reading frame by the virus. Therefore, the expression strategy of HEV appears to involve the use of three different open reading frames. This pattern of gene expression is unique in that it has not been described in the various families of single-stranded positive-sense, nonenveloped RNA viruses infecting humans or animals.

Careful comparison of geographic isolates at the molecular level will determine the extent of the HEV genetic diversity. Further analysis will be required in order to ascertain how this diversity affects diagnostic tests based upon the recombinant antigens expressed from

any particular strain. Similar analyses will determine if serotypic differences exist that might have conse- quences for any future attempts at subunit vaccine development.

ACKNOWLEDGMENT

We thank Rick Cuevas for assistance in the preparation of the manuscript.

REFERENCES

BALAYAN, M. S., ANDZHAPANDZE, A. G., SAVINSKAYA. S. S., KETILADZE, E. S., BRAGINSKI, D. M., SAVINOZ, A. P., and POLESCHUK, V. F. (1983). Evidence for a virus in non-A, non-B hepatitis transmitted via the fecal oral route. Infervirology 20, 23-31.

BRADLEY, D. W. (1990a). Enterically-transmitted non-A, non-B hepatitis. Br. Med. Bull. 46, 442-461.

BRADLEY, D. W. (1990b). Hepatitis non-A, non-B viruses become identified as hepatitis C and E viruses. Prog. Med. Viral. 37, 1 Ol- 135.

BRADLEY. D. W.. and BALAYAN, M. S. (1988). Virus of enterically transmitted non-A, non-B hepatitis. Lancer 1, 819.

BRADLEY, D. W.. KRAWCNNSKI, K., COOK, E. H., MCCAUSTLAND, K. A., HUMPHREY, C. D. SPELBRING, J. E., MYINT, H., and MAYNARD, J. E. (1987). Enterically transmitted non-A, non-B hepatitis: Serial pas- sage of disease in cynomolgus macaque and tamarins and recov- ery of disease-associated 27-34 nm virus-like particles. Proc. Nat/. Acad. Sci. USA 84, 6277-6281.

BRADLEY, D. W., and MAYNARD, 1. E. (1986). Etiology and natural his- tory of post transfusion and enterically-transmitted non-A. non-B hepatitis. Semin. Liver Dis. 6, 56-66.

CATHALA, G., SAVOURET, J.-F., MENDEZ, B., WEST, B. L., KARIN, M.. MARTIAL, J. A.. and BAXTER, 1. D. (1983). A method for isolation of intact, translationally active ribonucleic acid. DNA 2, 329-335.

CHOO, Q. L., Kuo, G., WEINER, A. J., OVERBY, L. R., BRADLEY, D. W., and HOUGHTON, M. (1989). Isolation of a cDNA clone derived from a blood-borne non-A, non-B hepatitis genome. Science 244,359- 362.

COHEN, J. I., TICEHURST, J. R., PURCELL, R. H., BUCKLER-WHITE, A., and BAROUDY, B. (1987). Complete nucleotide sequence of wild-type hepatitis A virus: A comparison with different strains of hepatitis A virus and other picornaviruses. 1. Viral. 61, 50-59.

EMERSON, S. U., MCRILL, C., ROSENBLUM, B., FEINSTONE, S., and PUR- CELL, R. H. (1991). Mutations responsible for adaptation of hepatitis A virus to efficient growth in cell structure. J. Viral. 65, 4882- 4886.

GORBALENYA, A. E., KOONIN. E. V., DONCHENKO, A. P., and BLINOV, V. M. (1989). Two related super-families of putative helicases involved in replication, recombination, repair and expression of DNA and RNA genomes. Nucleic Acids Res. 17.4713-4730.

HAN, J. H., SHYAMALA, V., RICHMAN, K. H., BRAUER, M. J., IRVINE, B., URDEA, M. S.. TEKAMP-OLSON, P., Kuo. G., CHOO, Q.-L., and HOUGHTON, M. (1991). Characterization of the terminal regions of hepatitis C viral RNA: Identification of conserved sequences in the 5’ untranslated region and poly(A) tails at the 3’ end. Proc. Nat/. Acad. Sci. USA88, 1711-1715.

JACOBSON, A. B., GOOD, L., SIMMONE~I, J., and ZUCKER, M. (1984). Some simple computational methods to improve the folding of large RNAs. Nucleic Acids Res. 12, 45-52.

KAMER, G., and ARGOS, P. (1984). Primary structural comparison of RNA-dependent polymerases from plant, animal and bacterial viruses. Nucleic Acids Res. 12, 7269-7282.

558 HUANG ET AL.

KHUROO, M. S. (1980). Study of an epidemic of non-A, non-B hepatitis: Possibility of another human hepatitis virus distinct from post- transfusion non-A, non-B type. Am. J. Med. 68, 8 18-824.

MAJOR, F., TURCOIX, M., GAUTHERET, D., LAPALME, G., FILLION, E., and CEDERGREN, R. (1991). The combination of symbolic and nu- merical computation for three-dimensional modeling of RNA. Science 253, 1255-l 260.

PEARSON, W. R., and LIPMAN, D. J. (1988). Improved tools for biological sequence comparison. Proc. Nat/. Acad. Sci. USA 85, 2444- 2448.

PURCELL, R. H., and TICEHURST, J. R. (1988). Enterically transmitted non-A, non-B hepatitis: Epidemiology and clinical characteristics, In “Viral Hepatitis and Liver Disease” (A. J. Zuckerman, Ed.), pp. 131-137. A. R. Liss, New York.

REYES, G. R. (199 1). Molecular cloning of the hepatitis E virus. In “VII International Symposium on Viral Hepatitis and Liver Disease” (F. B. Hollinger, S. M. Lemon, and H. S. Margolis, Eds.), pp. 514- 5 17. Williams and Wilkins, Baltimore.

REYES, G. R., and BAROUDY, B. M. (1991). Molecular biology of non-A, non-B hepatitis agents: the hepatitis C and Hepatitis E viruses. ln “Advances in Virus Research” (K. Maramorosch, F. A. Murphy, and A. I. Shatkin, Eds.), Vol. 40, pp. 57-102. Academic Press, Orlando.

REYES, G. R., and KIM, J. P. (1991). Sequence-independent, single- primer amplification (SISPA) of complex DNA populations. Mol. Cell. Probes 5, 473-48 1.

REYES, G. R., HUANG, C.-C., YARBOUGH, P., YOUNG, L. M., TAM, A. W., MOECKLI. R. A., ET AL. (1991a). Hepatitis E virus (HEV): Epitope mapping and detection of strain variation. ln “Viral Hepatitis C, D, and E” (T. Shikata, R. H. Purcell, and T. Uchida, Eds.), pp. 237- 245. Elsevier Science, The Netherlands.

REYES. G. R., YARBOUGH, P. O., TAM, A. W. E., PURDY, M. A., HUANG, C.-C., KIM, J. P., BRADLEY, D. W., and FRY, K. E. (1991 b). Hepatitis E virus (HEV): The novel agent responsible for enterically transmitted non-A, non-B hepatitis. Gastroenferol. Jpn. 26, (suppl. 3), 142- 147.

REYES, G. R.. PURDY, M. A., KIM, J. P., LUK, K. C., YOUNG, L. M., FRY, K. E., and BRADLEY, D. W. (1990a). Isolation of a cDNA from the virus responsible for enterically transmitted non-A, non-B hepatitis. Science 247, 1335-l 339.

REYES, G. R., PURDY, M. A., KIM, 1. P., HUANG, C.-C.. BRADLEY, D. W., YOUNG, L. M.. FRY, K. E., LUK, K. C., and BRADLEY, D. W. (1990b). Enterically-transmitted non-A, non-B hepatitis (ET-NANB): Identifi- cation and characterization of a molecular clone. (J.-L. Sung and D.-S. Chen, Eds.), pp. 249-254. Excerpta Medica, Hong Kong.

SAIKI, R. K., GELFAND, D. H., STOFFEL, S., SCHARF. S. J., HIGUCHI, R., HORN, G. T., MULLIS, K. B.. and ERLICH, H. A. (1988). Primer-di- rected enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239, 487-491.

SANGER, F.. NICKLEN, S., and COULSON, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Nat/. Acad. SC;. USA 74, 5463-5467.

SKIDMORE, S. J., YARBOUGH, P. O., GABOR, K. A., TAM, A. W., REYES, G. R., and FLOWER, A. J. E. (1991). Imported hepatitis E in the U.K. Letter. Lancer 337, 1541.

SVITKIN, V. V., CAMMACK, N., MINOR, P. D., and ALMOND, J. W. (1990). Translational deficiency of the Sabin type 3 poliovirus genome: Association with an attenuating mutation C,,,+U. Virology 175, 103-109.

TAM, A. W., SMITH, M. M., GUERRA, M. E., HUANG, C.-C., BRADLEY, D. W., FRY, K. E., and REYES, G. R. (1991a). Hepatitis Evirus (HEV): Molecular cloning and sequencing of the full-length viral genome. tirology185, 120-131.

TAM, A. W., SMITH, M. M., GUERRA, M. E., BRADLEY, D. W., FRY, K. E., and REYES, G. R. (1991 b). Hepatitis E virus: cDNA isolation and sequencing. ln “VII International Symposium on Viral Hepatitis and Liver Disease” (F. B. Hollinger, S. M. Lemon, and H. S. Margo- lis. Eds.), pp. 521-524. Williams and Wilkins, Baltimore.

TSAREV, S. A., EMERSON, S. U., REYES, G. R.. TSAREVA, T. S.. LEGTERS, L. J., MALIK, I. A., IGBAL, M., and PURCELL, R. H. (1992). Character- ization of a prototype strain of hepatitis E virus. Proc. Nat/. Acad. Sci. USA 89, 559-563.

WONG, D. C., PURCELL, R. H., SREENIVASAN, M. A., PRASAD. S. R., and PAVRI, K. M. (1980). Epidemic and endemic hepatitis in India: Evi- dence for a non-A, non-B hepatitis virus aetiology. Lancer 2, 876- 879.

YARBOUGH. P. O., TAM, A. W., FRY, K. E., KRAWCZYNSKI, K., MCCAUST- LAND, K. A., BRADLEY, D. W., and REYES, G. R. (1991a). Hepatitis E virus: Identification of type-common epitopes. J. Viral. 65, 5790- 5796.

YARBOUGH, P. O., TAM, A. W., KRAWC~YNSKI, K., FRY, K. E., MCCAUST- LAND, K. A., MILLER, A., FERNANDEZ, J., HUANG, C.-C., BRADLEY, D. W., and REYES, G. R. (1991 b). Hepatitis E virus (HEV): Identifica- tion of immunoreactive clones. In “VII International Symposium on Viral Hepatitis and Liver Disease” (F. B. Hollinger, S. M. Lemon, and H. S. Margolis. Eds.), pp. 524-526. Williams and Wilkins, Bal- timore.

ZUKER, M., JAEGER, J. A., and TURNER, D. H. (1991). A comparison of optimal and suboptimal RNA secondary structures predicted by free energy minimization with structures determined by phyloge- netic comparison. Nucleic Acids Res. 19, 2707-2714.

Molecular cloning and sequencing of the mexico isolate of hepatitis E virus (HEV

Documents

Transcript of Molecular cloning and sequencing of the mexico isolate of hepatitis E virus (HEV