|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
In the hantavirus-endemic area of northwestern Argentina, hantavirus disease was first diagnosed in 1994 in the subtropical region of Orán, in which HPS case-patients were associated with viruses of the ORN genotype. After the recognition of the disease in the neighboring province of Jujuy in 1997, an increasing number of HPS cases were prospectively diagnosed by enzyme-linked immunosorbent assay (ELISA). The largest number of HPS cases diagnosed in Jujuy Province originated in Ledesma Department (74.6%, 53 of 71); the main focus was located in the rural town of Yuto, with 62.3% (33 of 53) of the HPS cases recorded in this department (Ripoll C, unpublished data). Interestingly, a higher incidence of human hantavirus infection and a generally less severe clinical form of HPS has been observed in this location (Pini N and others, unpublished data and Ripoll C, unpublished data). This finding is different from the epidemiology of HPS in the rest of the hantavirus-endemic areas of Argentina. To determine the genetic diversity of hantaviruses causing HPS in northwestern Argentina, we studied the genetic types of hantaviruses found in human and rodent samples from this region.
| MATERIALS AND METHODS |
|
|
|---|
Patient samples. Human blood clots from serologically confirmed cases of acute HPS from Yuto and other locations of Jujuy Province were used as source for virus RNA extraction. Samples sent for clinical diagnosis followed the ethical procedures of the local hospital where patients were assisted. The protocol studies conducted in Yuto and surrounding areas were reviewed and approved by the institutional ethical review committee (Comité de Evaluación Ética de Investigaciones Biomédicas del Instituto Nacional de Enfermedades Virales Humanas).
Isolation of total RNA, RT-PCR, and sequence analysis. Total RNA was extracted from the lungs of hantavirus antibodypositive rodents and blood clots of hantavirus antibodypositive human by using the RNaid (PLUS) kit (Bio 101 Inc., La Jolla, CA). Amplification of virus RNA was done by a RT-PCR, followed by a nested PCR amplification as described previously.15 The DNA product of the nested PCR was excised after electrophoresis from an agarose gel, and bands of the predicted size were purified from gels slices using a GeneClean kit (Bio 101 Inc.). The nucleotide sequences of these products were determined by the automatic dydeoxy cycle sequencing technique (Applied Biosystems, Foster City, CA).3 Oligonucleotides primers were designed based on predicted virus-conserved sequence regions to amplify fragments of M genome segment glycoprotein-encoding region of Argentinian hantavirus genotypes and LN virus from Paraguay. Primers were designed to amplify fragments of the N protein of the S segment on the basis of conserved hantavirus sequences obtained from the GenBank database including LN virus, AF005727; AND virus strain AH-1, AF324901; ARA virus, AF AF307325; BMJ virus, AF482713; LEC virus, AF482714; ORN virus, AF482715; and PGM virus, AF482717.7,12 Extended sequence analysis was done with primers designed based on sequence data obtained in this study. Sequences were aligned with those of previously described hantaviruses using BioEdit version 5.0.6 (North Carolina State University, Raleigh, NC) and the computer software package Clustal W 1.4,16 followed by manual adjustment based on deduced amino acid sequences.
Phylogenetic analysis. Phylogenetic analysis was carried out on the multiple nucleotide (partial S and M genome segments) and amino acid (partial M-G2 genome segment) sequence alignments by using maximum parsimony (Phylogenetic Analysis Using Parsimony [PAUP] version 4.0b4a Macintosh computer software programs).17 Phylogenetic analysis by maximum parsimony was obtained by the heuristic search method. The bootstrap support for the results of the phylogenetic analysis was based on 500 replicates.
| RESULTS |
|
|
|---|
|
The second genotype identified was BMJ, previously described from a single O. chacoensis from Orán in Salta Province12; the virus sequence showed a 82.8%, 95.7%, and 94.2% nucleotide sequence identity in 273-, 328-, and 256-nucleotide fragments from the S, M-G1, and M-G2 genome segments, respectively, when compared with the virus sequence from one O. chacoensis from Salta province. This represents the first demonstration that BMJ virus is associated with HPS in Argentina. Finally, LN virus was recovered for the first time in Argentina from HPS cases and rodent samples. Comparison of the nucleotide sequences detected in five HPS cases, four C. callosus, and one A. simulator showed 83.684.2%, 80.285.1%, and 81.282.0% nucleotide sequence identity for the S, and M-G1, and M-G2 genome fragments, respectively, when compared with the LN virus from Paraguay. Mean sequence variation among the LN virus detected in the human and rodent samples was 0.6%, 2.7%, and 0.9% for the S, M-G1, and M-G2 genome segment fragments, respectively.
Phylogenetic analysis.
The phylogenetic analysis based on the M-G1 genome segment-encoding region using maximum parsimony (Figure 2
) showed that all samples from C. callosus, one A. simulator, and 5 HPS case samples fell into the LN lineage group; 6 HPS case samples fell into the ORN genetic group, and only one sample from an HPS patient grouped with BMJ virus. The nodes separating LN, ORN, and BMJ lineages had bootstraps values of 99%, 95%, and 99%, respectively; however, the exact branching order of the Argentinian hantaviruses relative to other South American lineages can not be resolved by the present phylogenetic analysis. ARA virus (host unknown) from Brazil is placed together with MAC virus (Akondontini-borne virus) from central Argentina, although with a low (52%) bootstrap support. A 1,371-nucleotide sequence of the M-G2 genome segment was generated from one representative strain of each of the three hantavirus lineages detected in Yuto. The phylogenetic tree showed a similar topology to the one obtained on the M-G1 genome segment fragment. The three hantavirus lineages (LN, ORN, and BMJ) are positioned together with the other South American hantaviruses, forming a well supported clade (90%) (Figure 3
); the nodes separating these lineages had bootstrap values of 100%, 60%, and 97%, respectively. The genetic analysis showed that the LN virus variant recovered from C. callosus (sample Cc29582) differed 16.8% at the nucleotide level and 2.9% at the deduced amino acid level relative to the Paraguayan LN virus; while representative virus strains of ORN (sample Hu50786) and BMJ (sample Hu52635) lineages, showed 7.6%, 4.3% nucleotide differences and 1.1%, 0.5% amino acid differences relative to the prototype ORN and BMJ viruses, respectively.
|
|
| DISCUSSION |
|
|
|---|
To identify the hantavirus(es) causing HPS in Jujuy Province, PCR amplification and sequencing of the appropriate PCR product was conducted in blood samples from seropositive HPS cases and rodent samples. Phylogenetic analysis of M and S genome fragments indicated a high genetic diversity of hantaviruses associated with HPS in the limited geographic area of Yuto: the circulation of three previously known hantaviruses (the Paraguayan LN virus, ORN virus, and BMJ virus) was documented. Phylogenetic analysis based on the sequence data of virus genome fragments of the S, M-G1, and M-G2 segments showed that the LN virus lineage, associated with C. laucha (vesper mouse) in Paraguay,7 was associated with C. callosus rodents and HPS cases from the same area of Yuto (more than 600 km from the area in Paraguay where the virus was described). The high sequence identity between human and rodent samples (> 97.3% nucleotide and 100.0% amino acid) implicated C. callosus as the primary rodent reservoir for LN virus circulating in Yuto. Our rodent trapping data indicated that C. callosus, as well as A. simulator, were the most abundant rodent species captured in the area (137 of 327, 38.0% and 140 of 327, 39%); however, C. callosus exhibited the highest prevalence of antibody to hantavirus (7 of 137, 5.1%). These observations correlate with one of the criteria currently accepted in identifying a primary rodent species as a primary rodent host for hantaviruses.18 The phylogenetic analysis places LN virus from Paraguay and Yuto together, with a 100% bootstrap support.
The fact that LN virus strains in Paraguay and Argentina (difference between them of 16.8% nucleotides and 2.9% amino acids) are associated with different species of the same rodent genus (C. laucha and C. callosus, respectively), suggests the coevolutionary relationship between the viruses and that of the corresponding primarily associated rodent reservoirs. Similarly, examples of a pattern of co-speciation was observed in North American rodent genus Peromyscus: SN-like viruses are associated to genetically distinct P. maniculatus and P. leucopus.1820 Our sequence data also indicated that a LN virus strain was derived from one A. simulator. Although the serologic data indicated that there were four positive A. simulator by ELISA, we detected virus sequences using the RT-PCR only in one of them. This could represent a LN virus spillover infection from the primary rodent host C. callosus into A. simulator. The pattern of unique hantavirus infection of a primary rodent host includes the chronic infection of the primary rodent reservoir species, as shown by the ability to consistently amplify hantavirus sequences from serologically positive rodents. Conversely, spill-over infections are usually charaterized by acute infection with subsequent clearance of virus.18 Thus, the possibility of an spillover from C. callosus into A. simulator cannot be excluded.
The other two genotypes recovered from HPS cases were the previously characterized Oligoryzomys-borne BMJ and ORN viruses.12 The recovery of a BMJ viral genotype from one fatal HPS case from Yuto constitutes the first evidence of its pathogenicity in Argentina; it was previously known only from one O. chacoensis in Orán. However, its association with human disease was predictable because the phylogenetic analysis had grouped the BMJ viral genotype into the sub-clade of the Oligoryzomys-borne viruses, most of them associated with HPS in Argentina. A recent report described the association of the BMJ viral genotype to HPS cases in Bolivia.14 The Oran virus genotype remains the most frequent cause of HPS, while previous studies showed that it was the main cause of HPS near Orán.12 Our sequence data identified ORN virus genotype in 50% (6 of 12) of the human samples that were positive by RT-PCR. Curiously, the high frequency of ORN virus genotype sequences in human case-patient samples does not correlate with low abundance of the rodent species that serves as its putative reservoir (3.3% of the total captures) or the lack of detectable hantavirus antibody prevalence among rodents from the genus Oligoryzomys. It must be pointed out that our rodent data correspond to a unique rodent capture study conducted in Yuto in 2000, while HPS case-patient samples included in the present study comprised a four-year period (19982001). Moreover, only two of six ORN viral genotype sequences were identified in human case-patient samples from Yuto; the remaining four ORN viral sequences came from HPS cases from Caimancito, Perico and Fraile Pintado, locations where no rodent trapping was conducted. As rodent community diversity is influenced by temporal, ecologic, and environmental factors, more intensive rodent serologic surveys in the region would serve to clarify this point.
Interestingly, one of the major differences in rodent behavior observed between Oligoryzomys and Calomys species is the relative capacity of C. callosus to invade human dwellings. Calomys callosus was the only wild rodent species captured inside houses in Yuto (Pini N and others, unpublished data), a fact that increases the potential exposure of humans to infected rodents. This also correlates with previous observations in Bolivian hemorrhagic fever outbreaks, caused by Machupo virus, which is harbored by C. callosus in Bolivia.21 On the other hand, a serologic and epidemiologic survey conducted in healthy people from Yuto (Pini N and others, unpublished data) simultaneously with our rodent trapping found one of the highest prevalences (6.5%) of antibody to hantavirus in the general population found in Argentina. These data correlate with previous observations from serologic surveys conducted in healthy communities of the Gran Chaco of Paraguay and Argentina (Salta Province) in which hantavirus antibody prevalences of 2040% had been found.7,22,23 Moreover, the particular epidemiologic pattern of higher incidence of human infection, low case-fatality rate (12.7%), and less clinical severity in this area would support the hypothesis that a less virulent hantaviruses may be circulating in the region under study. The fact that the two fatal cases were associated with Oligoryzomys-borne ORN and BMJ virus genotypes, while none was observed associated to LN virus might suggest that LN virus could be at least one of the viruses associated with a milder and/or inapparent form of infection in this region. Further ongoing comparative studies on the clinical pattern of infection associated with the different hantavirus lineages circulating in Jujuy will help to clarify this point.
Received August 5, 2003. Accepted for publication July 15, 2004.
Acknowledgments: We thank Dr. S. Morzunov and Dr. John Boone for critical review of the manuscript; Verónica Fasciani for technical assistance; and Horacio López, Germán ODuyer, Cesar Polidoro, Enrique Serrrano, Miguel Canchi, Alberto Segobia, Julio Gil, Bernardino Perez, Monica Diaz, and David Flores for their technical field work.
Financial support: This research was supported by the Administración Nacional de Laboratorios e Institutos de Salud (ANLIS), Ministerio de Salud Pública de la Nación, Argentina; Ministerio de Salud Pública de la Provincia de Buenos Aires, Argentina; and National Institutes of Health grant 1R01 AI-45059.
Authors addresses: Silvana Levis, Jorge Garcia, Noemí Pini, Gladys Calderón, and Delia Enria, Instituto Nacional de Enfermedades Virales Humanas Dr. Julio I. Maiztegui, Monteagudo 2510, 2700 Pergamino, Buenos Aires, Argentina, Telephone: 54-2477-433044, Fax: 54-2477-433045, E-mail: inevh{at}satlink.com. Josefina Ramírez and Elena Lozano, Hospital San Miguel, 4518 Yuto, Jujuy, Argentina, Telephone 54-3886-498015. Daniel Bravo, Hospital Oscar Orías, San Martin 66, 4512 Libertador Gral. San Martín, Jujuy, Argentina, Telephone 54-3886-42728. Stephen St. Jeor and Mariana Bego, Department of Microbiology, Howard Building 320, Room 146, University of Nevada, Reno, NV 89557, Telephone: 702-784-6161, Fax: 702-784-1620, E-mail: stjeor{at}med.unr.edu. Carlos Ripoll, Dirección de Epidemiología, San Martin 41, 4600 San Salvador de Jujuy, Jujuy, Argentina, Telephone 54-388-4221300. Rubén Barquez, Fundación Miguel Lillo, M. Lillo 205, 4000 Tucumán, Argentina, Telephone 54-381-4330888. Thomas G. Ksiazek, Special Pathogens Branch, Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, GA 30333, Telephone: 404-371-5329, Fax: 404-371-5449.
Reprint requests: Silvana Levis, Instituto Nacional de Enfermedades Virales Humanas Dr. Julio I. Maiztegui, Monteagudo 2510, 2700 Pergamino, Buenos Aires, Argentina.
| REFERENCES |
|
|
|---|
This article has been cited by other articles:
![]() |
M. L. MILAZZO, M. N. B. CAJIMAT, J. D. HANSON, R. D. BRADLEY, M. QUINTANA, C. SHERMAN, R. T. VELASQUEZ, and C. F. FULHORST CATACAMAS VIRUS, A HANTAVIRAL SPECIES NATURALLY ASSOCIATED WITH ORYZOMYS COUESI (COUES' ORYZOMYS) IN HONDURAS Am J Trop Med Hyg, November 1, 2006; 75(5): 1003 - 1010. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |