Orthobunyavirus Antibodies in Humans, Yucatan Peninsula, Mexico

We performed a serologic investigation to determine whether orthobunyaviruses commonly infect humans in the Yucatan Peninsula of Mexico. Orthobunyavirus-specific antibodies were detected by plaque reduction neutralization test in 146 (18%) of 823 persons tested. Further studies are needed to determine health risks for humans from this potentially deadly group of viruses.

W e previously reported the isolation of Cache Valley virus (CVV), Kairi virus (KRIV), Cholul virus (CHLV), and South River virus (SOURV) from mosquitoes in the Yucatan Peninsula of Mexico (1)(2)(3). Antibodies to CVV, CHLV, and SOURV were also detected in livestock in this region (4). These viruses belong to the genus Orthobunyavirus (5). All viruses in this genus possess a tripartite, single-stranded, negative-sense RNA genome.
CVV is a recognized human pathogen (5) that has been linked to severe encephalitis and multiorgan failure. KRIV has not been implicated as a cause of human disease, although antibodies to this virus have been detected in humans in Argentina (6). Recent data suggest that CHLV is a reassortant that acquired its small RNA segment from CVV and medium and large RNA segments from Potosi virus (POTV) (1). No clear evidence exists for human susceptibility to infection with CHLV or SOURV. However, diagnostic laboratories rarely test for orthobunyavirus infection; therefore, the true disease incidence and seroprevalence of these viruses remains to be determined. Because orthobunyaviruses comprise a neglected but potentially deadly group of viruses and recent studies have provided evidence of orthobunyavirus activity in the Yucatan Peninsula (1-4), we investigated whether orthobunyaviruses commonly infect humans in this region.

The Study
Serum samples were obtained from 823 febrile patients at the Secretaria de Salud de Yucatán and other health institutions in Merida during January-October 2007. The patients resided in all 3 states of the Yucatan Peninsula of Mexico: Yucatan (n = 809), Quintana Roo (n = 8) and Campeche (n = 6). The study was approved by the Institutional Biosafety Committees at Iowa State University (Ames, IA, USA) and the Universidad Autónoma de Yucatán (Mérida, Mexico).
Titers were expressed as the reciprocal of highest serum dilutions yielding >90% reduction in the number of plaques (PRNT 90 ). For etiologic diagnosis, the PRNT 90 antibody titer for each virus was required to be >4-fold greater than that to the other viruses tested.
Antibodies that neutralized CVV were detected in 146 (18%) of 823 study participants. The mean ages of patients with and without antibodies that neutralized CVV were 32.0 and 22.3 years, respectively. Logistic regression analysis showed that the risk for infection increased signifi cantly with age (p = 0.0001).
Serum samples from 50 seropositive patients were titrated and analyzed by comparative PRNT to identify the orthobunyaviruses responsible for these infections. Six persons were seropositive for CVV, 5 for CHLV, and 1 for SOURV or a SOURV-like virus; 38 had antibodies to an undetermined orthobunyavirus (Table). Because SOURV was the only CAL serogroup virus used in this study, and another CAL serogroup virus may have been responsible for the infection, the person who had a SOURV PRNT 90 titer >4-fold than that to the other viruses tested received a conservative PRNT diagnosis of seropositive for SOURV or a SOURV-like virus. Because interserogroup crossreactivity of neutralizing antibodies to viruses in the BUN and CAL serogroups has not been seen, the 17 persons with antibodies that neutralized SOURV and >1 of the BUN serogroup viruses might have been exposed to >1 viruses from each serogroup. CHLV and POTV share the same medium RNA segment, so antibodies for these viruses cannot be differentiated by PRNT. Furthermore, antibodies to CHLV and POTV cannot be differentiated by complement fi xation test (4). Thus, we cannot dismiss POTV as a possible cause of infection in some or all of the study participants who were seropositive for CHLV. However, it appears more likely that these persons had been infected with CHLV because this virus has been isolated in the Yucatan Peninsula, whereas no direct evidence has been found for POTV in this region.

Conclusions
We found 18% of the 823 Yucatan residents participating in our study had evidence of orthobunyavirus exposure. This number is presumably an underestimate; additional seropositive persons might have been identifi ed if the initial PRNTs had not been restricted to CVV. In particular, additional seropositive persons likely would have been detected if SOURV was used in the initial PRNTs, because a screening algorithm that includes only a BUN serogroup virus would likely miss many CAL serogroup virus infections. Nevertheless, we provide evidence that orthobunyaviruses commonly infect humans in the Yucatan Peninsula.
Previous serosurveys have provided information on the seroprevalence of orthobunyaviruses in humans in the United States. For example, antibodies that neutralized CVV were detected in 42/356 (12%) residents in Maryland and Virginia in 1961-1963 (8). Antibodies that neutralized Maguari virus or Tensaw virus were detected in 71/≈300 humans in Florida in the 1980s (9); as observed in our study, the highest PRNT titers for most of the seropositive persons in that study did not exceed 40.
All persons in our study cohort initially sought care for unspecifi ed fever; however, we could not determine whether any of these febrile illnesses were a direct consequence of orthobunyavirus infection. The detection of acute orthobunyavirus infections is limited because no IgM-capture ELISA for orthobunyavirus diagnosis exists. PRNTs can be used to identify recent orthobunyavirus infections when paired acute and convalescent serum samples are available, but for our study, only single serum samples were available from each participant. Orthobunyavirus viremias in humans are transient and of low magnitude, which makes reverse transcription PCR ineffective for the detection of orthobunyavirus RNA in serum samples. However, a duplex reverse transcription PCR was recently developed for the detection of CVV RNA in human cerebrospinal fl uid (10).
In conclusion, we provide evidence that orthobunyaviruses commonly infect humans in the Yucatan Peninsula. These viruses are also a common cause of infection of livestock in this region (4). Our fi ndings underscore the need to determine whether orthobunyaviruses represent an unrecognized cause of illness in humans and vertebrate animals in Mexico.