Skip Navigation

Core Promoter Mutations and Genotypes in Relation to Viral Replication and Liver Damage in East Asian Hepatitis B Virus Carriers

  1. Magnus Lindh,
  2. Charles Hannoun,
  3. Amar P. Dhillon,
  4. Gunnar Norkrans and
  5. Peter Horal
  1. Department of Clinical Virology and Infectious Diseases, Göteborg University, Göteborg, Sweden; Department of Histopathology, Royal Free Hospital, London, United Kingdom
  1. Reprints or correspondence: Dr. Magnus Lindh, Dept. of Clinical Virology, Göteborg University, Guldhedsgatan 10B, S-413 46 Göteborg, Sweden (magnus.lindh{at}microbio.gu.se).
 
Next Section

Abstract

Virus load and liver damage, as measured by quantitative polymerase chain reaction and histology activity index, were related to genotype and core promoter mutations in 43 chronic hepatitis B virus (HBV) carriers of East Asian origin. T-1762 mutants were more frequent in genotype C strains and were associated with more inflammation (P = .0036) and fibrosis (P = .0088) of the liver but not with hepatitis B e antigen (HBeAg) status or virus load. Conversely, precore mutations were associated with less liver inflammation (P =.08), which was linked to HBeAg negativity and lower viral replication. Carriers with genotype C were more often HBeAg positive (P = .03) with precore wild type strains and more-severe liver inflammation (P = .009) than were those with genotype B. These findings suggest that pathogenic differences between genotypes may exist and that the T-1762 mutation may be useful as a marker for progressive liver damage but seem to contradict that down-regulation of HBeAg production is the major effect of this mutation.

Chronic hepatitis B is a major health problem and may lead to cirrhosis and hepatocellular carcinoma. The liver damage in hepatitis B virus (HBV) infection is thought to result from the immune response and not from a cytopathic effect of the virus itself. Thus, the course of infection depends on several factors that influence the immune response, such as age at infection, genetic host factors (including HLA type), and probably the genetic variability of the virus, which influences the expression of viral antigens.

The genetic variability of HBV is observed both as the evolution of genotypes, that is, a divergence of the viral genome in the carrier population, and as the emergence of mutations in each infected subject. Several studies have investigated the effect of mutations on the clinical course of chronic hepatitis B. Mutations at nt 1762–1764 in the core promoter region have been associated with more-severe liver damage [1, 2] and with increased response to interferon [3]. However, the biologic effects of these mutations are not fully clarified. They have previously been linked to a “hepatitis B e antigen (HBeAg)-negative phenotype” [4] but have also been observed in the HBeAg-positive stage, in agreement with the experimental observation that they cause a relatively moderate down-regulation of the synthesis of HBeAg [5–7]. Some [5, 6, 8] but not all [7] studies have reported higher viral replication in strains with mutations at nt 1762–1764, possibly of importance for the induction of fulminant hepatitis [9] or the course of chronic HBV infection.

Six HBV genotypes (A–F) have been described [10, 11]. Genotypes B and C prevail in East Asia, where more than half of the estimated 300 million carriers in the world live. Unlike hepatitis C virus (HCV), genotype-related differences in the pathogenicity of HBV have not yet been described. However, in a large-scale Japanese study, more severe liver disease was found in carriers infected with HBV of serotype adr (i.e., genotype C) than in carriers with adw (mainly genotype B in Japan); moreover, this was associated with a difference in HBeAg status [12].

To investigate these associations further, we analyzed the liver damage in East Asian HBV carriers infected with genotypes B and C in relation to viral replication and mutations in the core promoter and precore regions.

Previous SectionNext Section

Methods

Patients

Serum samples from 48 chronic HBV carriers of East Asian origin were analyzed, and the liver damage in patients carrying HBV genotypes B (n = 23) or C (n = 20) was compared. These 43 subjects were from the following countries: Vietnam, 25; Thailand, 6; Korea, 6; China, 2; and 1 each from Indonesia, the Philippines, Malaysia, and Laos. All lacked antibodies to HCV and hepatitis delta virus.

Genotyping

Genotyping by restriction fragment length polymorphism of an S region [13] or a pre-S region amplicon [14] identified genotypes A, B, or C in 45 serum samples. In 2 cases, the genotype was established after sequencing (see below) by analysis of a few key nucleotides in the X region, which we have found to be strongly linked to genotypes; nt 1726–1727/1730 (CT/G specific for genotype B) and nt 1802 (T specific for genotypes B and C).

Sequencing

After polymerase chain reaction (PCR) using the outer primers 1603F (5′-GTTGCATGGAGACCACCGTGAAC) and 2058R (5′-GTATGGTGAGGTGAACAATG) or 2160R (5′-CTGACTACTAATTCCCTGGATGCTGGGTCT) and the inner primers 1603F or 1680F (5′-ATGTCGACAACCGACCTTGA) and 2058R, sequencing was done by the chain-termination method. Each amplicon was analyzed in both sense and antisense directions with 1680F or 1603F and 2058R, respectively, as primers in a cycle-sequencing reaction by use of fluorescent dye terminators. The sequence was read in an automated capillary sequence reader (Prism 310; Applied Biosystems, Foster City, CA) and analyzed by computer (Sequence Navigator software; Applied Biosystems).

Mutation analysis

The variability at nt positions 1858 and 1896 was analyzed with amplification-created restriction site methods as previously described [15] and by sequencing as described above. Mutations in the core promoter were analyzed by sequencing.

Serology

HBeAg and anti-HBe were analyzed by HBeAg IMX (Abbott Laboratories, Abbott Park, IL). HCV antibodies were analyzed by an anti-HCV system (Murex Diagnostics, Dartford, UK). Delta antibodies were analyzed by RIA (Abbott).

Quantitative PCR

Serum samples taken at the time of liver biopsy were analyzed by Amplicor HBV monitor (Roche Diagnostic Systems, Sommerville, NJ) according to the manufacturer's instructions. The detection range for this test is 103 to 107 copies/mL [16]. To extend the detection range, samples with HBV DNA above 107 copies/mL were reanalyzed after predilution 1:104 in negative serum.

Histopathology activity index (HAI) and alanine aminotransferase (ALT)

HAI was scored as described by Knodell et al. [17] in a blinded fashion. HAIinfl, that is, the sum of the component scores for piecemeal necrosis, lobular inflammation, and portal inflammation, and HAIfibr, the fibrosis score, were analyzed separately. ALT levels at the time of liver biopsy were recorded.

Statistics

HAI scores were compared by Mann-Whitney rank sum test and Fisher's exact test and ALT values by Mann-Whitney rank sum test.

Previous SectionNext Section

Results

Genotypes

The genotypes in the 48 carriers were as follows: B, 23; C, 20; A, 4; and 1 not typeable (negative PCR). The detailed results for patients with genotype B or C are shown in table 1. The mean age was 31.2 years (range, 18–56) in genotype B and 29.9 years (range, 13–53) in genotype C carriers. In the genotype B group, 5 of 23 patients were female compared with 8 of 20 in the genotype C group (P = .32). Nine (39%) of 23 persons with genotype B compared with 14 (70%) of 20 with genotype C were HBeAg-positive (P = .036). In agreement, HBV DNA levels were higher in subjects with genotype C, but not significantly: For genotypes B and C, respectively, the median HBV DNA levels were 105.5 and 106.5 copies/mL (P = .15). No significant difference was seen within groups of HBeAg-positive genotype B and C carriers (median: 109.1 and 107.6 copies/mL, respectively) and in genotype B and C HBeAg-negative subjects (median: 104.5 and 105.5 copies/mL, respectively; figure 1A).

Figure 1
View larger version:
Figure 1

Virus load and liver damage by histopathology activity index (HAI) scores in persons with HBV with genotypes B (n = 23) and C (n = 20). Box plots show 10th, 30th, 50th, 70th, and 90th percentiles and outliers. HBeAg, hepatitis B e antigen; HAIinfl, sum of component scores for necrosis, lobular inflammation, and portal inflammation.

The distribution of liver damage in the subjects with genotype B and C strains as measured by HAI scores is presented in figures 1B–1E and summarized in table 2. Persons with genotype C had higher inflammation scores than those with genotype B (P = .009). Consistently, the mean ALT was 3.1 times the upper reference value (URV) in the genotype C group compared with 1.2 × URV in the genotype B group (P = .0049). These findings reflect the higher virus load, and there was also a tendency for a more pronounced inflammation in relation to the virus load in persons with genotype C (P = .050; figure 1C). Fibrosis scores were not statistically different; a fibrosis score of 3 or 4 was seen in 55% of carriers with genotype C versus 35% with genotype B (table 2).

Mutations

The AGG→TGA mutation at nt 1762–1764 was the most common mutation in the core promoter region, appearing in 41.9% (18/43) of the subjects (figures 2, 3). Persons with wild type (AGG) strains had significantly higher HBV DNA levels in the HBeAg positive group (P = .0076; figure 2A). The TGA mutant was more common in persons with genotype C than with genotype B: 13 (68.4%) of 19 compared with 5 (21.7%) of 23, respectively (P = .0059). In addition, 1 genotype C strain showed an AGA mutant at nt 1762–1764. Presence of a TGA mutant was associated with more-severe inflammation and fibrosis (P = .0036 and .0088, respectively). This was seen in HBeAg-positive and -negative subjects (figure 2D). In fact, there was no association between the TGA mutation and HBeAg status: TGA mutants were found in 45% (10/22) of HBeAg-positive (in 1 as AGG/TGA mixture) and 33% (7/21) of HBeAg-negative subjects (in 2 as AGG/TGA mixtures).

Figure 2
View larger version:
Figure 2

Virus load and liver damage measured by histopathology activity index (HAI) scores in HBV carriers with T-1762 (mutant, n = 18) and A-1762 (wild type, n = 25) in core promoter. Box plots show 10th, 30th, 50th, 70th, and 90th percentiles and outliers. HBeAg, hepatitis B e antigen; HAIinfl, sum of component scores for necrosis, lobular inflammation, and portal inflammation.

Figure 3
View larger version:
Figure 3

Core promoter region of HBV in 43 East Asian chronic carriers with genotypes B (1–23) and C (24–43) infection in relation to hepatitis B e antigen (HBeAg) status and liver damage. HAIsum, sum of histology activity index scores for inflammation and fibrosis; B cons, consensus sequence of genotype B strains. K = G/T mixture, Y = C/T mixture, W = A/T mixture, R = A/G mixture. GenBank accession nos.: AF106089–AF106131.

Other mutations in the core promoter were more rare. Variation at nt 1751 was seen in 11 patients (6, G-1751; 5, T-1751) and was linked to genotype but not to liver damage. C-1752 was seen in 7 patients (6 with genotype C) and was associated with more-severe liver damage (P = .0073) but always occurred together with TGA at nt 1762–1764. A distinct genotype-related variation was observed at nt 1726–1730: All 23 genotype B strains had CTGAG, whereas 18 of 20 genotype C strains had AAGAC (2 had AGGAC). Less-consistent genotype-associated variation was seen at nt 1773 and 1799 (figure 3).

The GρA mutation at position 1896 that causes a TAG stop codon in the precore region was more frequent in genotype B (table 1; figure 4): Only 1 (5%) of 20 carriers with genotype C compared with 11 (48%) of 23 with genotype B were infected with precore TAG mutant strains only (P = .0052). Presence of a precore mutant strain was associated with HBeAg negativity (P = .004; P < .0001 for samples with isolated mutant infection) and lower HBV DNA levels (P = .001 and .0001, respectively), and carriers with such strains tended to have less-severe inflammation (P = .17 and .06, respectively) but not less fibrosis than associated with wild type infection (figure 4). In the HBeAg-negative phase, carriers with wild type infection had more-severe inflammation (P = .061), which was also related to virus load (P = .030).

Figure 4
View larger version:
Figure 4

Virus load and liver damage as measured by histopathology activity index (HAI) scores in HBV carriers with precore WT (wild type, n = 23) and MUT (stop mutant, n = 20). Box plots show 10th, 30th, 50th, 70th, and 90th percentiles and outliers. HBeAg, hepatitis B e antigen; HAIinfl, sum of component scores for necrosis, lobular inflammation, and portal inflammation.

There was no overall association between precore and core promoter mutations, although there was a tendency for an inverse relation: A precore mutant was found in all HBeAg-negative samples with wild type AGG (at 1762–1764) infection, as shown in table 1 (P = .030). All of the HBeAg-negative subjects had a precore or a core promoter mutant or a combination of both. However, in 2 (patients 15 and 22) a “double wild type” strain was present in mixture with the mutant strain.

Table 1
View larger version:
Table 1

Clinical and virological data on 43 hepatitis B virus carriers of East Asian origin.

Table 2
View larger version:
Table 2

HAIinfl and HAIfibr scores in 43 East Asian hepatitis B virus carriers infected with genotype B (n = 23) and genotype C (n = 20).

Five strains, all genotype C, had cytosine at position 1858 (C-1858). One of these, in a person with an HAI score of 4, had T-1857 and A-1897 (nucleotides that form a base pair in the pregenomic RNA stem loop), indicating a C→T/G→A double mutation at these sites, to create a TGA stop at codon 28.

Previous SectionNext Section

Discussion

In the present study, the AGG→TGA mutation at nt 1762–1764 in the core promoter region was associated with more-severe inflammation and fibrosis in chronic hepatitis B. The finding agrees with previous reports [1, 2] and indicates that this mutation may be useful as a prognostic marker. The association was seen in both HBeAg-positive and -negative subjects and was independent of virus load. Overall, HBV DNA levels were similar in carriers with and without TGA mutants. However, persons who were HBeAg positive with TGA mutants, which have been attributed a higher rate of replication [5, 6, 8], had a lower virus load (P = .007), showing that the host's immune response is probably more important for the HBV DNA level than inherent viral factors.

The TGA mutation has been associated with an “HBeAg-negative phenotype,” and there has been some experimental support for this idea. However, in only one study was the reduction of precore RNA transcription (and thus HBeAg synthesis) relatively prominent [6]; in the others it was absent [18] or moderate [5, 7, 8]. In the present series of patients, the TGA mutant was not associated with HBeAg negativity. In fact, the TGA mutation was more common in HBeAg-positive than in -negative persons, because of the high frequency of the mutation in HBeAg-positive genotype C subjects and wild type AGG in HBeAg-negative healthy carriers. These findings seem to contradict the suggestion in prior reports that effects on HBeAg expression are the major mechanism for the emergence of this mutant. Kidd-Ljunggren et al. [19] reported an overall association of the TGA mutation to HBeAg status, but this was not true for genotypes B and C: TGA mutants were found in 44% of HBeAg-positive subjects with these genotypes. Thus, other potential explanations for the frequent emergence of this mutation need to be investigated. For example, escape from proposed intracellular effects of cytokines on the core promoter [20] may be of importance, as may effects on the conformation of the 3′ RNA stem loop [21] or the lysine-valine to methionine-isoleucine exchange in the X protein.

Other mutations in the core promoter were rarer, and only the C-1752 mutation, which was always observed in conjunction with TGA at 1762–1764, was associated with liver damage, whereas variation at several other positions was genotype related. In particular, variation at nt 1726–1730 was strictly associated with genotype, and all genotype B strains had CTGAG. In agreement, we noted this variation in 9 genotype B GenBank sequences but not in any of 72 non-B genotype sequences (not shown). We also observed T at nt 1802 in all sequences of genotypes B, C, and F (the latter is restricted to the American continents) but not in the other genotypes. Thus, analysis of positions 1726–1730 and 1802 appears to be useful for identifying genotypes B and C.

An interesting finding was the association of genotype C with elevated inflammation scores. This agrees with the findings in a Japanese study of 1744 HBsAg carriers that found more liver damage in carriers with adr (invariably genotype C) compared with adw strains (which usually are genotype B in Japan) [12]. In that study, as in our present study, the more pronounced liver damage in adr carriers was related to a larger proportion being HBeAg-positive. Another Japanese study reported that adr carriers are more often HBeAg positive than are carriers infected with adw strains [22]. We have no explanation for this difference. However, as carriers with genotype B in the present study seemed more often to be in the early high-replicative (tolerance) or late low-active (surveillance) phase, genotype B could be associated with a faster transition through the immunoactive stage and a faster HBe seroconversion. This might to some extent be related to the fact that the precore G→A mutation at nt 1896 (which completely abrogates the synthesis and secretion of HBeAg) was more common in genotype B. Similarly, a cytosine at position 1858 (C-1858; seen in 5 patients) may delay HBeAg seroconversion in some persons with genotype C by preventing a precore TAG mutation. However, because HBe seroconversion in general implicates a reduction of HBV DNA levels of >99%, the finding seems to indicate that the viral replication in genotype B is more readily reduced by the immune response. This could be due to either genotype differences in antigen expression or regulation of viral replication, but this requires further study.

The low number of patients studied raises the possibility that the difference between genotypes is due to unrecognized selection bias or unknown factors, such as duration of infection. There was no difference in mean age among subjects with genotypes B and C. Although East Asia is a large region and possibly not homogeneous for HBV epidemiology, one might presume that a major proportion in both groups were infected in the perinatal period and thus that the difference in liver damage was not due to a difference in the duration of infection.

In summary, the present study shows that the AGG→TGA mutation at nt 1762–1764 is associated with more-severe liver damage in both HBeAg-positive and -negative subjects. Moreover, the results indicate that genotype C, compared with genotype B, is associated with a more pronounced liver inflammation, a lower frequency of precore mutants, and a higher frequency of 1762–1764 TGA mutants. The data point to the possibility of pathogenic differences among HBV genotypes and encourage further analysis of these associations in larger populations of HBV carriers.

Previous SectionNext Section

Acknowledgments

We thank Annkatrin Gusdal and Carolina Gustavson for expert technical assistance and Mats Fogelqvist at Roche Diagnostics for support.

Previous SectionNext Section

Footnotes

  • The study was approved by the local ethics committee. All patients gave informed consent.

  • Financial support: Swedish Medical Research Council, Göteborg Medical Society, Swedish Medical Society, and Claes Groschinsky Memorial Foundation.

  • Received August 27, 1998.
  • Revision received November 23, 1998.
Previous Section
 

References

  1. 1.↵
    1. Takahashi K,
    2. Aoyama K,
    3. Ohno N,
    4. et al
    . The precore/core promoter mutant (T1762A1764) of hepatitis B virus: clinical significance and an easy method for detection. J Gen Virol 1995;76:3159-64.
    Abstract/FREE Full Text
  2. 2.↵
    1. Lindh M,
    2. Gustavson C,
    3. Maårdberg K,
    4. Norkrans N,
    5. Dhillon P,
    6. Horal P
    . Mutation of nucleotide 1762 in the core promoter region during hepatitis B e seroconversion and its relation to liver damage in HBeAg+ carriers. J Med Virol 1998;55:185-90.
    CrossRefMedlineWeb of Science
  3. 3.↵
    1. Kanai K,
    2. Kako M,
    3. Aikawa T,
    4. et al
    . Core promoter mutations of hepatitis B virus for the response to interferon in e antigen-positive chronic hepatitis B. Am J Gastroenterol 1996;91:2150-6.
    MedlineWeb of Science
  4. 4.↵
    1. Okamoto H,
    2. Tsuda F,
    3. Akahane Y,
    4. et al
    . Hepatitis B virus with mutations in the core promoter for an e antigen-negative phenotype in carriers with antibody to e antigen. J Virol 1994;68:8102-10.
    Abstract/FREE Full Text
  5. 5.↵
    1. Buckwold V,
    2. Xu Z,
    3. Chen M,
    4. Yen SB,
    5. Ou JH
    . Effects of a naturally occurring mutation in the hepatitis B virus basal core promoter on precore gene expression and viral replication. J Virol 1996;70:5845-51.
    Abstract/FREE Full Text
  6. 6.↵
    1. Moriyama K,
    2. Okamoto H,
    3. Tsuda F
    . Mayumi M. Reduced precore transcription and enhanced core-pregenome transcription of hepatitis B virus DNA after replacement of the precore-core promoter with sequences associated with e antigen-seronegative persistent infections. Virology 1996;226:269-80.
    CrossRefMedlineWeb of Science
  7. 7.↵
    1. Gunther S,
    2. Piwon N,
    3. Will H
    . Wild-type levels of pregenomic RNA and replication but reduced pre-C RNA and e-antigen synthesis of hepatitis B virus with C(1653) → T, A(1762) → T and G(1764) → mutations in the core promoter. J Gen Virol 1998;79:375-80.
    Abstract/FREE Full Text
  8. 8.↵
    1. Buckwold VE,
    2. Xu Z,
    3. Yen TS,
    4. Ou JH
    . Effects of a frequent double-nucleotide basal core promoter mutation and its putative single-nucleotide precursor mutations on hepatitis B virus gene expression and replication. J Gen Virol 1997;78:2055-65.
    Abstract/FREE Full Text
  9. 9.↵
    1. Sato S,
    2. Suzuki K,
    3. Akahane Y,
    4. et al
    . Hepatitis B virus strains with mutations in the core promoter in patients with fulminant hepatitis. Ann Intern Med 1995;122:241-8.
    Abstract/FREE Full Text
  10. 10.↵
    1. Norder H,
    2. Courouce AM,
    3. Magnius LO
    . Complete genomes, phylogenetic relatedness, and structural proteins of six strains of the hepatitis B virus, four of which represent two new genotypes. Virology 1994;198:489-503.
    CrossRefMedlineWeb of Science
  11. 11.↵
    1. Okamoto H,
    2. Tsuda F,
    3. Sakugawa H,
    4. et al
    . Typing hepatitis B virus by homology in nucleotide sequence: comparison of surface antigen subtypes. J Gen Virol 1988;69:2575-83.
    Abstract/FREE Full Text
  12. 12.↵
    1. Shiina S,
    2. Fujino H,
    3. Uta Y,
    4. et al
    . Relationship of HBsAg subtypes with HBeAg/anti-HBe status and chronic liver disease. Part I. Analysis of 1744 HBsAg carriers. Am J Gastroenterol 1991;86:866-71.
    MedlineWeb of Science
  13. 13.↵
    1. Lindh M,
    2. Andersson AS,
    3. Gusdal A
    . Genotypes, nt 1858 variants, and geographic origin of hepatitis B virus—large-scale analysis using a new genotyping method. J Infect Dis 1997;175:1285-93.
    Abstract/FREE Full Text
  14. 14.↵
    1. Lindh M,
    2. Gonzalez JE,
    3. Norkrans G,
    4. Horal P
    . Genotyping of hepatitis B virus by restriction pattern analysis of a pre-S amplicon. J Virol Methods 1998;72:163-74.
    CrossRefMedlineWeb of Science
  15. 15.↵
    1. Lindh M,
    2. Horal P,
    3. Dhillon AP,
    4. Furuta Y,
    5. Norkrans G
    . Hepatitis B virus carriers without precore mutations in HBeAg-negative stage show more severe liver damage. Hepatology 1996;24:494-501.
    CrossRefMedlineWeb of Science
  16. 16.↵
    1. Lehtovaara P,
    2. Uusi-Oukari M,
    3. Laaksonen M,
    4. Bengtström M,
    5. Ranki M
    . Quantitative PCR for hepatitis B virus with colorimetric detection. PCR methods and applications 1993;3:169-75.
    MedlineWeb of Science
  17. 17.↵
    1. Knodell RG,
    2. Ishak KG,
    3. Black WC,
    4. et al
    . Formulation and application of a numerical scoring system for assessing histological activity in asymptomatic chronic active hepatitis. Hepatology 1981;1:431-5.
    MedlineWeb of Science
  18. 18.↵
    1. Nishizono A,
    2. Hiraga M,
    3. Kohno K,
    4. et al
    . Mutations in the core promoter/enhancer II regions of naturally occurring hepatitis B virus variants and analysis of the effects on transcription activities. Intervirology 1995;38:290-4.
    MedlineWeb of Science
  19. 19.↵
    1. Kidd-Ljunggren K,
    2. Oberg M,
    3. Kidd AH
    . Hepatitis B virus X gene 1751 to 1764 mutations: implications for HBeAg status and disease. J Gen Virol 1997;78:1469-78.
    Abstract/FREE Full Text
  20. 20.↵
    1. Romero R,
    2. Lavine JE
    . Cytokine inhibition of the hepatitis B virus core promoter. Hepatology 1996;23:17-23.
    CrossRefMedlineWeb of Science
  21. 21.↵
    1. Kidd A,
    2. Kidd-Ljunggren K
    . A revised secondary structure model for the 3-end of hepatitis B virus pregenomic RNA. Nucleic Acids Res 1996;24:3295-301.
    Abstract/FREE Full Text
  22. 22.↵
    1. Noguchi A,
    2. Hayashi J,
    3. Nakashima K,
    4. Hirata M,
    5. Ikematsu H,
    6. Kashiwagi S
    . HBsAg subtypes among HBsAg carriers in Okinawa, Japan. Evidence of an important relationship in seroconversion from HBeAg to anti-HBe. J Infect 1994;28:141-50.
    CrossRefMedlineWeb of Science
| Table of Contents

This Article

  1. J Infect Dis. (1999) 179 (4): 775-782. doi: 10.1086/314688
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most