Discovery of naturally occurring transmissible chronic hepatitis B virus infection among Macaca fascicularis from mauritius island


  • Tatiana Dupinay,

    1. Université de Lyon, Lyon, France
    2. INSERM U1052, Centre de Recherche en Cancérologie de Lyon, Lyon, France
    3. CNRS UMR5286, Centre de Recherche en Cancérologie de Lyon, Lyon, France
    4. Ecole pratique des hautes études, Paris, France
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Tarik Gheit,

    1. International Agency for Research on Cancer, Lyon, France
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Pierre Roques,

    1. Division of ImmunoVirology (SIV), Institute of Emerging Diseases and Innovative Therapies, Centre d'Energie Atomique, Fontenay-aux-Roses, France
    2. Université Paris-Sud 11, Orsay, France
    Search for more papers by this author
  • Lucyna Cova,

    1. Université de Lyon, Lyon, France
    2. INSERM U1052, Centre de Recherche en Cancérologie de Lyon, Lyon, France
    3. CNRS UMR5286, Centre de Recherche en Cancérologie de Lyon, Lyon, France
    Search for more papers by this author
  • Philippe Chevallier-Queyron,

    1. Université de Lyon, Lyon, France
    2. INSERM U1052, Centre de Recherche en Cancérologie de Lyon, Lyon, France
    3. Hospices civiles de Lyon, Lyon, France
    Search for more papers by this author
  • Shin-i Tasahsu,

    1. The Research Center for Hepatitis and Immunology, National Center for Global Health and Medicine, Ichikawa, Japan
    Search for more papers by this author
  • Roger Le Grand,

    1. Division of ImmunoVirology (SIV), Institute of Emerging Diseases and Innovative Therapies, Centre d'Energie Atomique, Fontenay-aux-Roses, France
    2. Université Paris-Sud 11, Orsay, France
    Search for more papers by this author
  • François Simon,

    1. Hopital Saint Louis, Paris, France
    Search for more papers by this author
  • Geneviève Cordier,

    1. Université Lyon, , Université Lyon I, INRA UMR 754, IFR128 Biosciences Lyon-Gerland, France
    Search for more papers by this author
  • Lahcen Wakrim,

    1. Virology Unit, Viral Hepatitis Laboratory, Pasteur Institute of Morocco, Casablanca, Morocco
    Search for more papers by this author
  • Soumaya Benjelloun,

    1. Virology Unit, Viral Hepatitis Laboratory, Pasteur Institute of Morocco, Casablanca, Morocco
    Search for more papers by this author
  • Christian Trépo,

    1. Université de Lyon, Lyon, France
    2. INSERM U1052, Centre de Recherche en Cancérologie de Lyon, Lyon, France
    3. CNRS UMR5286, Centre de Recherche en Cancérologie de Lyon, Lyon, France
    4. Hospices civiles de Lyon, Lyon, France
    Search for more papers by this author
  • Isabelle Chemin

    Corresponding author
    1. Université de Lyon, Lyon, France
    2. INSERM U1052, Centre de Recherche en Cancérologie de Lyon, Lyon, France
    3. CNRS UMR5286, Centre de Recherche en Cancérologie de Lyon, Lyon, France
    • Address reprint requests to: Isabelle Chemin, Ph.D., CRCL-INSERM 1052, Team 16, 151 Cours Albert Thomas, 69003 Lyon, France. E-mail:; fax: 0033472681971.

    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

  • This work was supported by an “ANRS” Grant.

  • See Editorial on Page 1533


Despite a high prevalence of hepatitis B virus (HBV) infection in endangered apes, no HBV infection has been reported in small, old-world monkeys. In search for a small, nonhuman primate model, we investigated the prevalence of HBV infection in 260 macaque (Cercopithecidae) sera of various geographical origins (i.e., Morocco, Mauritius Island, and Asia). HBV-positive markers were detected in cynomolgus macaques (Macaca fascicularis) from Mauritius Island only, and, remarkably, HBV DNA was positive in 25.8% (31 of 120) and 42% (21 of 50) of serum and liver samples, respectively. Strong liver expression of hepatitis B surface antigen and hepatitis B core antigen was detected in approximately 20%-30% of hepatocytes. Furthermore, chronic infection with persisting HBV DNA was documented in all 6 infected macaques during an 8-month follow-up period. Whole HBV genome-sequencing data revealed that it was genotype D subtype ayw3 carrying substitution in position 67 of preS1. To confirm infectivity of this isolate, 3 Macaca sylvanus were inoculated with a pool of M. fascicularis serum and developed an acute HBV infection with 100% sequence homology, compared with HBV inoculum. We demonstrated the presence of a chronic HBV infection in M. fascicularis from Mauritius Island. This closely human-related HBV might have been transmitted from humans, because the initial breeding colony originated from very few ancestors 300 years ago when it was implemented by Portuguese who imported a handful of macaques from Java to Mauritius Island. Conclusion: This report on natural, persisting HBV infection among cynomolgus macaques provides the first evidence for the existence of a novel, small simian model of chronic HBV infection, immunologically close to humans, that should be most valuable for the study of immunotherapeutic approaches against chronic hepatitis B. (Hepatology 2013;58:1610–1620)




alanine aminotransferase


covalently closed circular DNA


Centre d'Energie Atomique


chronic hepatitis B


hepatitis B core


hepatitis B core antigen


hepatitis B surface antigen


hepatitis B virus


human T-lymphotropic virus




nonhuman primates


Na+/taurocholate cotransporting polypeptide


phosphate-buffered saline


polymerase chain reaction


simian immunodeficiency virus


University of California Davis


viral genome equivalents

Despite the existence of an effective vaccine, chronic hepatitis B virus (HBV) infection remains a major public health problem, responsible for 55% of hepatocellular carcinomas worldwide. Current chronic hepatitis B (CHB) treatments (e.g., interferon and nucleos(t)ide analogs) remain long lasting, expensive, partially efficient (25%), and frequently lead to the emergence of resistant variants.[1] Because chronic HBV carriers are crippled by weak, functionally impaired immune responses, immunotherapeutic approaches that are able to stimulate or restore humoral and cellular virus-specific immune responses are currently considered as a priority goal for CHB therapy.[2, 3] However, major hurdles are the lack of suitable in vivo models of HBV infection close to humans, with the exception of chimpanzees, which are now a protected and unaffordable species. Therefore, there is an urgent need for the development of a novel primate model for CHB infection studies that should be immunologically very close to humans, regarding innate and cellular responses, and that will permit accurate evaluation of new immunotherapeutic anti-HBV approaches.

In the last 20 years, HBV transmission to old-world primates maintained in captivity has been reported. An “HBV-like” virus was also found in nonhuman primates (NHPs), including Hominidae (chimpanzee, gorilla, and orangutan),[4-11] Hylobatidae (gibbon),[4, 12-15] and Atelidae (woolly monkey).[16] Those species are distributed over Africa (chimpanzee and gorilla), Southeast Asia (orangutan and gibbon), and South America (woolly monkey). HBV-like viruses infecting various NHP species or subspecies are genetically distinct from one another as well as from human HBV genotypes. Their clustering in specific groups suggests that they could represent indigenous virus populations.[17] However, the origin of these hepatitis B–like viruses remains controversial. The hypothesis of cross-species transmission from humans cannot be ruled out, because the various HBV genotypes from chimpanzees from different geographical regions cluster with human HBV genotype D, although the refined phylogenic analysis data indicate the presence of distinct viral isolates,[18] supporting the concept that these viruses may have evolved with their hosts. Previous epidemiological studies have reported a high prevalence of HBV infections in great apes that was comparable to human populations in Gabon and Congo.[19] However, the presence of natural HBV infection among small monkeys has hitherto never been demonstrated.

Our previous studies already opened the possibility of using macaques for HBV studies, which are the NHPs most commonly used in biomedical research. We have demonstrated both successful in vivo HBV transfection and in vitro HBV transduction with baculovirus vector in macaques, although only transient viral infection could be generated by this method in these animals.[20, 21]

In the current study, we therefore searched for the presence of a natural HBV infection among macaques of various geographical origins by analyzing sera and liver samples from macaques (Cercopithecidae) originating from Asia (China, Indonesia, and the Philippines), Morocco, and Mauritius Island.

Materials and Methods

Macaques and HBV Infection

Mauritius cynomolgus macaques

Mauritius adult cynomolgus macaques (Macaca fascicularis), 4-5 years old with body weight >5 kg, were first quarantined and maintained in international accredited breeding facilities in Mauritius Island and were imported from Mauritius and housed at the Centre d'Energie Atomique (CEA; Fontenay-aux-Roses, France). A permit (FR0803100893-I) was obtained from CITES to import the adult M. fascicularis from Mauritius to France. NHPs are used at the CEA in accord with French national regulations, and CEA facilities are fully authorized (under no. B-92-032-02) for animal use and for NHP breeding (under no. 2005-69). Animals were used under supervision of veterinarians in charge of the animal facility, and the protocols employed were reviewed and approved by the ethical animal committee of the CEA.

Asian cynomolgus macaques

Asian M. fascicularis macaques were housed at The California National Primate Research Center, University of California Davis (UCD; Davis, CA). Sera were collected during routine veterinary procedures and stored at −70°C until they were tested for HBV markers. Animal work was approved under UCD Institutional Animal Care and Use Committee protocol 10665 (Hepatitis B-Like Virus Infection in Nonhuman Primates).

Macaque sylvanus from Morocco

Macaque sylvanus (Macaca sylvanus) were captured in the wild (Middle Atlas Mountains) and were quarantined and maintained at the Pasteur Institute of Casablanca (Morocco) under conditions that met or exceeded all requirements needed for the physical and psychological well-being of such animals. These macaque sylvanus had not been exposed to any hepatotropic viruses before in vivo inoculation of HBV DNA, and all animals were negative for serological markers of infection with hepatitis A, B, and C and human T-lymphotropic virus (HTLV)-I and HTLV-II viruses.

Animals were kept in the Pasteur Institute individual cage during quarantine. The Pasteur Institute from Morocco is fully authorized by the Moroccan Ministry of Agriculture to capture NHPs for experimental purposes (authorization for animal capture no. 14062 EF/CPN/PN).

M. sylvanus Inoculation by Mauritius HBV Isolate

Three M. sylvanus (BL12, BL13, and BL14) were intravenously inoculated with 1 mL of cynomolgus macaques–positive HBV DNA serum (103 particles/mL). After inoculation, animals were bled weekly to test for HBV surface antigens (HBsAgs), anti-HBc (hepatitis B core) antibodies (Abs), and alanine aminotransferase (ALT) and aspartate aminotransferase levels. For HBV infection follow-up, monkeys were anesthetized by an intramuscular injection of ketamine (1 mg/kg) before collection of blood. At the end of follow-up, monkeys were anesthetized with ketamine and then sacrificed with an intracardiac injection of KCl.

Serum DNA Extraction, Polymerase Chain Reaction Assay, and Southern Blotting Hybridization

Nucleic acids were extracted from 140 µL of serum using a nucleic acid extraction kit (Qiagen, Courtaboeuf, France). Presence of HBV DNA was tested in macaque serum using polymerase chain reaction (PCR), followed by southern blotting analysis. Primers for PCR amplification were selected from sequences overlapping the core and surface genes that are highly conserved among all human HBV genotypes and NHP HBV-like viruses.[20]

Hepatitis B Infection Marker

HBsAg detection was performed with the VIDAS HBsAg Ultradetection kit (bioMérieux, Marcy l'Etoile, France) and the Ortho Antibody to HBsAg ELISA Test System 3 (Ortho Clinical Diagnostics, Inc., Raritan, NJ). Total anti-HBc Ab detection was performed with the VIDAS Anti-HBc Total II kit (bioMérieux).

Liver Nucleic Acid Extraction

We also tested for the presence of HBV DNA in livers from experimentally inoculated M. sylvanus. Nucleic acids were extracted from 10 mg of liver tissue with the MasterPure Complete DNA and RNA Purification Kit (Epicentre Biotechnologies, Le Perray en Yvelines, France) or by a procedure described in detail by Jilbert et al.[22]

Quantification of Viral Load by Real-Time PCR

Quantitative analysis of viral load was performed by real-time PCR (Light Cycler; Roche, Grenoble, France).[23] HBV DNA was also quantified by real-time PCR using the primers, 5'-GCTGACGCAACCCCCACT-3' (forward) and 5'-AGGAGTTCCGCAGTATGG-3' (reverse). An iCycler MyiO thermocycler (96-well format; Bio-Rad, Hercules, CA) was used with an iQ SYBR Green Supermix kit (Bio-Rad, Marnes-la-Coquette, France). This quantitative PCR was validated for a detection limit of 50 copies of HBV/genome/mL of serum.

Quantification of Covalently Closed Circular DNA in Liver by Real-Time PCR

A real-time PCR assay was previously validated for the specific detection of covalently closed circular DNA (cccDNA) and total intracellular HBV DNA in liver biopsy specimens.[24] cccDNA and total intracellular HBV DNA were measured and normalized to per-cell values, using the cellular β-globin gene, ultimately providing median intrahepatic cccDNA levels. Serial dilutions of a plasmid containing HBV monomer (pHBVEcoRI) were used as quantification standards.

Immunofluorescence on Liver Sections for HBsAg and HBc Antigen

Five-micrometer-thick frozen liver tissue sections were fixed in acetone for 10 minutes at −20°C, blocked for 10 minutes with phosphate-buffered saline (PBS)/3% bovine serum albumin, and incubated for 45 minutes at room temperature with rabbit polyclonal Abs raised against hepatitis B core antigen (HBcAg; AbD Serotec, Colmar, France) or with mouse monoclonal Abs raised against HBsAg (Dako, Trappes, France). After this incubation, sections were washed in PBS and labeled with fluorescein-conjugated goat antimouse or -rabbit secondary Ab (1:100; Bio-Rad, France). Sections were stained with Evans blue, mounted, and finally examined with a Leica DM RXE confocal microscope (Leica Microsystems, Wetzlar, Germany).

Phylogenetic Analysis

Sequences were edited using the SeqMan program in the LASERGENE package (DNASTAR, Inc., Madison, WI). Sequences were thereafter aligned with the corresponding region in sequences retrieved from GenBank. Phylogenetic analysis was carried out with the ClustalX program package version 2. Phylogenetic trees were constructed using neighbor joining in the ClustalX package. Genotypes and subgenotypes were determined by analysis of the amplified fragments of the S gene with sequences from previously genotyped and subgenotyped strains.[25] The deduced amino acid sequence of the S gene region was used to determine the serotype, which was assessed from the amino acids at codons 122, 127, and 160.[26]

Examination of HBV Isolates Recombination

Assessment of possible recombination was investigated by using the software packages, Simmonic 2005 v1.6 and SimPlot v3.5.1, both implementing PHYLIP (Phylogeny Inference Package v3.68; J. Felsenstein, Department of Genome Sciences, University of Washington, Seattle, WA[27]).


Detection of HBV Markers Among Macaques

We investigated the natural HBV infection in sera samples from two macaque species, the M. sylvanus and M. fascicularis, belonging to the Cercopithecidae family. Two hundred and sixty serum samples from macaques were tested for HBV DNA by PCR (Table 1). Of the 120 Asian M. fascicularis sera and 20 Moroccan M. sylvanus sera, all were HBV negative. By contrast, 25.8% (31 of 120) Mauritius M. fascicularis sera showed HBV DNA positivity with a viral load ranging from 101 to 106 HBV DNA copies/mL (mean viral load: 8.62 × 103 viral genome equivalents [VGE]/mL). Viremia subsequently could be performed for 6 HBV DNA–positive macaques, and after an 8-month period, all 6 animals maintained viral DNA levels between 101 and 103, peaking at 106 HBV DNA copies/mL for 1 animal (Fig. 1). The majority of animals exhibited only modest viremia variations over 8 months of follow-up. In addition, each quantitative PCR for HBV DNA detection was performed in triplicate and exhibited only limited variations.

Table 1. Prevalence of HBV Infection in Macaques From Diverse Geographical Origins
Common NameSystemic NameOriginType of SampleTotal AnimalsHBV DNA Positivea (%)HBV DNA QuantificationbMean of Viremia (VGE/mL)
  1. Mean of 22 macaque sera tested.

  2. Abbreviation: ND, not determined.

  3. a

    Presence of HBV DNA was tested using PCR in the S and C genes at least twice, with amplification products run on agarose gels, followed by southern blotting hybridization.

  4. b

    Quantitative analysis of viral load in serum or liver of macaques was performed by real-time PCR (Light Cycler; Roche, Grenoble, France). Quantification was further validated by branched DNA (bDNA; Test VERSANT HBV bDNA 3.0) or Amplicor Monitor Roche assays for a number of animal samples.

Cynomolgus macaqueM. fascicularisMauritius IslandLiver5021/50 (42)0.2-102 copies/hepatocyte 
Cynomolgus macaqueM. fascicularisMauritius IslandSerum12031/120 (26)101−106 copies/mL8.62 × 103
Cynomolgus macaqueM. fascicularisAsiaSerum1200ND0
Barbary macaqueM. sylvanusMoroccoSerum200ND0
Figure 1.

Schematic representation of the 8-month follow-up of viremia in 6 M. fascicularis from Mauritius Island. HBV DNA quantification was performed with a quantitative PCR assay. Viral loads were expressed in VGE/mL. Each symbol represents the viral copy number of 1 animal.

Next, we analyzed liver biopsies from Mauritius Island M. fascicularis and demonstrated the presence of HBV DNA sequences in 21 of 50 (42%) analyzed samples (Table 1). In addition, HBsAg and HBcAg was investigated by immunostaining in liver tissue of 9 HBV DNA–positive Mauritius macaques and showed, for all these animals, 20%-30% of strongly stained hepatocytes (Fig. 2). Liver histological examination did not reveal any significant pathological changes (data not shown). Liver HBV DNA quantification performed by real-time PCR assays[24] showed the presence of 2 × 10−1−102 total HBV DNA copies/hepatocyte, and 1 × 10−1−101 cccDNA copies/hepatocyte (Table 1 and data not shown). Among these 21 HBV DNA-positive M. fascicularis, 4 were also HBsAg positive in serum using a commercially available HBsAg test. The most positive Mauritius macaque for HBsAg (positive in the Ortho HBsAg test and VIDAS HBsAg Ultra) was estimated in cobas HBsAg II quant to approximately 1.4 IU/mL and it gave us a positive HBeAg detection with cobas Elecsys immunoassay, with a value of 0.284 Paul Erhlich Institute standard units/mL.[28]

Figure 2.

IF analysis of HBcAg and HBsAg in liver tissues from HBV DNA–positive M. fascicularis. Arrows show HBsAg and HBcAg positivity in hepatocytes. (A) HBc cytoplasmic expression. (C) Same slide tested without primary Ab: Only some weak auto fluorescent dots are detected. (B) Cytoplasmic expression of HBsAg in hepatocytes. (D) Same slide tested without primary anti-HBs Ab: No fluorescent spots in the cytoplasm of hepatocytes are detected.

Phylogenetic Analysis of Complete Genome From Mauritius M. fascicularis HBV Isolate Provide Evidence of Genotype D

Phylogenetic analysis of HBV isolates from Mauritius Island, based on S gene, showed that all 12 sequences clustered together in a unique clade and revealed that all sequences analyzed belonged to genotype D, subgenotype D3, and serotype ayw3. In the phylogenetic tree, these isolates segregated into one clade, sharing similarity with human HBV genotype D isolates from Europe and the United States. The phylogenetic tree of the C-gene analysis demonstrated strong clustering of M. fascicularis HBV sequence into human HBV genotype D (data not shown). After the successful amplification of the complete genome, the sequencing data revealed that it was of 3,182 base pairs in length (data not shown). Phylogenetic analysis showed that this complete genome clustered with human HBV subgenotype D3 (Fig. 3) because it was also the case when subgenomic regions C and PreS2/S were analyzed (data not shown). Moreover, the complete genome M. Fascicularis HBV sequence was 98%-99% identical to previously published human HBV sequences (Fig. 3). To get better insight into the similarity of macaque, nonhuman, and human HBV, amino acid sequences were deduced from different genes of the viral genomes and aligned with previously published sequences. One substitution (P67S in the pre-S1 domain) was interesting because it was located in a key region for viral entry (Fig. 4). Thus, databases indicated that this proline residue within preS1 is strongly conserved among all HBV genotypes, and only a few sequences with this mutation were found in published HBV sequences. Among these particular amino acid sequences harboring the P67 mutation, six were found to be associated with human HBV and the remaining among chimpanzees or gibbons (as illustrated in Fig. 4). A number of changes along the genome can be noticed, as compared to prototypes of the different known HBV genotypes (as illustrated in Supporting Fig. 1).

Figure 3.

Phylogenetic analysis was realized with ClustalX and using the neighbor-joining method (Kimura 2 parameters model; R = 2). Dendogram based on phylogenetic analysis of whole genome in M. fascicularis, compared with reference human and nonhuman HBV strains (accession numbers are indicated on the tree).

Figure 4.

Alignment of amino acid sequences of the fragment of pre-S1 domain from the M. fascicularis and chimpanzee HBV isolates and with human HBV genotype A and D. Dashes indicated deleted amino acids. Red boxes indicate the deletion of pre-S1 and the presence of the substitution P67S.

Finally, the complete Mauritius M. fascicularis HBV genome sequenced was examined for the presence of recombination with other HBV genotypes using the previously described, Bootscan analysis implemented in the SimPlot software program.[27] Bootscan analysis showed no evidence of recombination between HBV DNA from M. fascicularis and other genotypes (Fig. 5).

Figure 5.

SimPlot analysis demonstrating the absence of recombination between human HBV and M. fascicularis strain. This strain was subjected to Bootscan analysis over the entire genome using the SimPlot program.

Transmission of the Mauritius HBV Isolate to M. sylvanus

To explore the infectivity of HBV from Mauritius M. fascicularis, we inoculated 3 M. sylvanus with serum pool (103 particles/mL) from an HBV DNA–positive M.fascicularis. Serum viral DNA, HBsAg, and ALT levels were followed for 9 months. All 3 animals exhibited a similar pattern with serum HBsAg peaking between 4 and 7 weeks postinoculation, followed by an elevation of ALT up to approximately 250 U/L. Moreover, HBV viremia paralleled HBsAg patterns (Fig. 6A-D). Liver histopathology, performed at necropsy (i.e., 9 months postinfection) showed a resolving acute hepatitis pattern, including mild sinusoidal dilatation, presence of inflammatory cells, and histopathological modifications, indicating resolving acute hepatitis (Fig. 6E,F), whereas histological analysis of liver sections from control animals did not show such pathology (data not shown). Expression of intracellular HBV antigens investigated by immunofluorescence (IF) on frozen liver sections showed an expression of HBV core and surface antigens in approximately 20%-30% of hepatocytes (data not shown).

Figure 6.

Schematic representation of HBV infection markers follow-up in 3 M. sylvanus (BL12, BL13, and BL14) inoculated with HBV isolated from Mauritius M. fascicularis. (A) Transaminase variations after HBV infection. (B) HBsAg detection postinfection by Ortho Antibody to HBsAg ELISA Test System 3 (Ortho Clinical Diagnostics, Inc., Raritan, NJ). (C) HBV DNA quantification in genome Eq/mL in serum of macaque BL14 after infection; *viremia was further confirmed peaking to 1.4 × 103 VGE/mL at this time point (day 63) by an Amplicor Monitor (Roche, Grenoble, France). (D) HBV DNA detection in serum of macaque BL14 by PCR with primers located in the S region, followed by southern blotting hybridization of HBV DNA amplification products. (E and F) Pathological analysis for macaque BL14 liver 9 months postinfection through hematoxylin phloxine saffron (×20) and KP1 (×10) (anti-CD68) staining, respectively; arrows show few macrophage infiltrations (brown cells).


To develop a novel small simian model for the study of novel therapeutic approaches of CHB, we extensively searched for the presence of natural HBV infection in NHPs currently used for biomedical research, especially among macaques (Cercopithecidae) of various geographical origin. After investigation in M. fascicularis from China, Indonesia, the Philippines, and Mauritius Island, as well as M. sylvanus from Morocco, we report here the detection of HBV infection in macaques from Mauritius Island only. To date, occurrence of HBV infections was reported in approximately 16% of great-ape populations, based on PCR positivity and HBsAg detection,[4, 17, 29] and it was shown that human HBV isolates could also infect gibbons or chimpanzees.[30, 31] Here, we provide the first description of chronic HBV infection in small monkeys that can be used under laboratory conditions.

HBV DNA was found positive in 25.8% and 42% of Mauritius M. fascicularis sera and livers, respectively. Interestingly, 6 macaques were repeatedly positive for serum HBV DNA over an 8-month follow-up period, indicating the presence of chronic infection, and the majority of them exhibited only modest viremia variations. By contrast, the viremia of 1 animal (OGD6) varied greatly from relatively high (month 1) to undetectable values (month 8). Similarly, we and others have also observed and reported on important variations in viremia overtime in some cases of occult hepatitis B patients.[32]

Importantly, phylogenetic analysis of a complete viral genome showed that it was HBV genotype D and, more specifically, subgenotype D3, serotype ayw3. The detailed analysis of the pre-S1 sequence revealed proline-to-serine substitution at position 67, which seems to be more specific to NHP HBVs. This substitution is located within the pre-S region that is known to play a crucial role in viral entry,[33] suggesting a possible effect on viral pre-S1 domain conformation and subsequently on the species specificity of this isolate.

Recently, the article by Yan et al.[34] described a receptor for HBV in humans. It does also report on the few differences in amino acid between the human Na+/taurocholate cotransporting polypeptide (NCTP) and the macaque cynomolgus NCTP, therefore underlying that the species sensitivity to HBV infection may rely on those amino acids. Moreover, Yan et al. demonstrated a specific interaction between NCTP and the pre-S1 domain, which mediated the binding of the virions. Because we were not able to infect primary Mauritius macaques hepatocytes with human HBV, but only with the HBV Mauritian isolate (data not shown), we may suppose that minor changes in the pre-S1 domain of the Mauritian HBV may have possibly been adapted to the Mauritian cynomolgus NCTP receptor.

Detection of HBsAg and HBcAg in liver sections from Mauritius Island's M. fascicularis showed approximately 30% of strongly stained hepatocytes. In addition, to confirm the infectivity of this isolate, 3 naïve M. sylvanus were inoculated with a pool of sera from HBV-positive Mauritius Island macaques. We have used the M. sylvanus macaques for this transmission study becausee most of Mauritius Island's M. fascicularis macaques have either anti-HBsAg Abs or were HBV carriers (data not shown). All 3 M. sylvanus macaques presented a parallel rise in HBsAg levels and HBV DNA with increasing ALT values and histopathological signs of acute hepatitis, which were observed in serum of all these animals for several weeks postinoculation, thus confirming the infectivity and pathogenicity of this inoculum.

The occurrence of HBV zoonosis still remains poorly documented. Zoonotic infection of HBV has been suggested in great apes because their HBV genotypes tend to cocluster according to the environmental geographic distribution of genotypes in Africa and Southeast Asia.[17, 35] The sequence homology between HBV DNA isolated from Mauritius M. fascicularis and human HBV is probably related to the introduction of a few animals approximately 400 years ago by Portuguese sailors from Java to Mauritius island.[36] Since then, animals may have expanded from an initial effective of 10-15 individuals and remained isolated in the island for approximately 80-100 generations.[37, 38] The initial event leading to HBV infection of macaques by humans may have occurred at the time of capture and importation by the Portuguese that may have been infected by HBV genotype D. Genotype D is widespread all over the world, with accounts in India, Asia, Europe, and North America.[25, 39, 40] Whether the existence of this HBV infection among the Mauritius M. fascicularis population could be a potential source of infection transmission to humans who come in contact with them, as demonstrated for simian immunodeficiency virus (SIV),[41] is, at present, speculative and the precise risk remains to be assessed.

An understanding of HBV evolution in humans could greatly benefit from better knowledge of its predecessor, simian HBV, in NHPs. Whereas HBV causes liver disease in humans, this virus generally produces only a benign infection in primates. This avirulence is often attributed to coevolution between the virus and its host, possibly resulting from codivergence over millions of years. We now provide a temporal reference to further test this hypothesis about the evolution of HBV in its natural small primate host.

In addition to data presented here, we have previously reported that M. sylvanus species can be transfected with an HBV-coding plasmid.[20] Moreover, in vitro transduction of primary macaque hepatocyte with baculovirus-HBV constructs secreted HBV particles.[21] Those results had already opened up the possibility of using the macaque model for HBV study, although chronic viral infection was not achieved in these experimental studies.

Our current data support the evidence of chronic HBV infection in some macaques and the possible transmission of the HBV strain to M. sylvanus from Morocco. Moreover, our screening evaluation performed on 120 animals demonstrates, for the first time, that a high proportion of wild-living M. fascicularis in Mauritius Island is naturally infected with HBV. However, to date, most of the HBV infections in M. fascicularis were found to be cryptic, with very low titer or undetectable markers in the absence of liver disease. To date, we have no hypothesis how this prevalence can be maintened in natura, because very low peripheral viral load implies a poor transmission efficiency either within the macaque group or from macaque to human animal-care team. Such natural, low-grade infections in Mauritius macaques are similar to occult HBV infections in humans, which present a major concern in the hepatitis B field[42-45] and may therefore represent an excellent model and unique opportunity for their study and therapy for which no recommendations exist.

The Mauritius M. fascicularis could represent a particularly valuable model for novel anti-HBV immunotherapeutic approaches for evaluation because of their limited genetic diversity, notably in the Mhc class I region, that could influence the immune cellular response as it has been demonstrated for SIV infection.[46-48] Further characterization of this novel small primate close to humans will make it possible, for the first time, to test innovative immunotherapeutical approaches, including therapeutic DNA vaccination,[49] alone or in combination with cytokines, possibly as an adjuvant to antiviral nucleos(t)ides for their ability to cure chronic HBV infection.


Many thanks to B. Chomel for his interest and support. The authors are grateful to Nick Lerche and Souad Sekkat for their invaluable guidance and help in this study.