Novel Arenavirus Infection in Humans, United States

TOC Summary: North American Tacaribe serocomplex viruses cause acute central nervous system disease or undifferentiated febrile illnesses.


T he arenaviruses (family Arenaviridae, genus
Arenavirus) known to occur in North America include Whitewater Arroyo virus (WWAV), 7 other members of the Tacaribe serocomplex (Table 1), and lymphocytic choriomeningitis virus (LCMV, the prototypic member of the lymphocytic choriomeningitis-Lassa serocomplex). Specifi c members of the order Rodentia are the principal hosts of the arenaviruses, for which natural host relationships have been well characterized. For example, the hispid cotton rat (Sigmodon hispidus) in Florida is the principal host of Tamiami virus (6,7), and the ubiquitous house mouse (Mus musculus) is the principal host of LCMV (9).
Five South American members of the Tacaribe serocomplex, LCMV, and Lassa virus are etiologic agents of severe febrile illnesses in humans (10,11). The human health signifi cance of the North American Tacaribe serocomplex viruses has not been rigorously investigated (12).
Studies since the mid-1990s have shown that Tacaribe serocomplex viruses are widely distributed in the United States and Mexico and that woodrats (Neotoma spp.) and other members of the family Cricetidae are natural hosts of these viruses (1)(2)(3)(4)(5)8,13,14). The purpose of this study was to investigate whether humans have been infected with North American Tacaribe serocomplex viruses.

Materials and Methods
Samples of serum (n = 1,305), plasma (n = 2), and cerebrospinal fl uid (n = 70) from 1,185 persons in the United States with acute central nervous system disease or undifferentiated febrile illnesses were tested for immunoglobulin (Ig) G against the WWAV prototype strain AV 9310135 and LCMV strain Armstrong by using an ELISA as described (15 A 1:80 dilution and 1:320 dilution of each sample was tested against the WWAV antigen, LCMV antigen, and corresponding comparison (negative-control) antigens. The adjusted optical density (AOD) of a sample-antigen reaction was the optical density of the well coated with the test antigen minus the optical density of the well coated with the corresponding control antigen. A sample was considered positive if the AOD at 1:80 was >0.250, the AOD at 1:320 was >0.250, and the sum of the AOD at 1:80 and AOD at 1:320 was >0.750. Endpoint titers against each antigen were measured in the positive samples by using serial 2-fold dilutions from 1:320 through 1:40,960. The antibody titer of a positive sample was the reciprocal of the highest dilution for which the AOD was >0.250. Titers <320 were 160 in comparisons of titers to WWAV and LCMV in individual samples. The apparent homologous virus in an antibody-positive sample was the virus associated with the highest titer if the absolute value of the difference between the titers to WWAV and LCMV was >4-fold.

Results
We detected antibody against an arenavirus in 41 ( Twelve persons had positive test results for WWAV but not LCMV; 28 for LCMV but not WWAV; and 1 for WWAV and LCMV ( Table 2). In the positive samples, endpoint titers against WWAV and LCMV ranged from <320 to 10,240 and from <320 to 20,480, respectively. The apparent homologous virus was WWAV in 10, LCMV in 24, and indeterminate in 7 of antibody-positive persons ( Table 2).
Ages of the 10 persons in whom WWAV was the apparent homologous virus ranged from 5 to 70 years (median 43 years). Samples from these persons were submitted from Arizona, New Mexico, and North Carolina (1 sample each) and Florida and Wyoming (2 samples each); for 3 samples, state of submission was unknown.
The ELISA included paired samples from 8 antibodypositive persons. Time from onset of illness to the fi rst samples from these persons ranged from 0 to 47 days. In side-by-side tests, the endpoint titer to WWAV in the second sample was >4-fold higher than that to WWAV in the fi rst sample in paired samples from 2 persons, and the endpoint titer to LCMV in the second sample was >4-fold higher than that to LCMV in the fi rst sample in paired samples from 3 of the 6 other antibody-positive persons (Table 3).

Discussion
Previously, antibody to Tamiami virus was found in 5 (3.8%) of 131 Seminole Indians sampled in southern Florida (16), and antibody to a Tacaribe serocomplex virus was found in 2 (0.24%) of 829 persons who had worked with cricetid rodents in North America (15,17). The results of our current study strengthen the notion that Tacaribe serocomplex viruses enzootic in North America are infectious in humans. The increase in antibody titer against WWAV in cases 1 and 2 in this study (Table 3) suggests that a North American Tacaribe serocomplex virus caused the illnesses in these persons.
The WWAV strain AV 9310135 was originally isolated from a white-throated woodrat (N. albigula) captured in northwestern New Mexico (8). A recent study demonstrated a high level of diversity among the amino acid sequences of the structural proteins of the North American Tacaribe serocomplex viruses (5). Hypothetically, human IgG against some North American Tacaribe serocomplex viruses is not strongly reactive against WWAV in ELISA. If so, the prevalence of antibody to Tacaribe serocomplex viruses in this study actually might be >3.5%.
The severity of human disease caused by LCMV ranges from mild febrile illness to severe encephalitis and disseminated disease (18). The results of this study suggest that the illnesses in case-patients 4-6 ( Table 3) were caused by LCMV. Whether samples from these 3 persons were tested for anti-LCMV antibody (IgM or IgG) by clinical laboratories could not be determined from records maintained at CDC.
Specimens from 33 of the antibody-positive persons in this study were limited to single specimens. Perhaps these illnesses were caused by a North American Tacaribe serocomplex or by LCMV. The antibody titer to WWAV in the antibody-positive person from New Mexico was 10,240 in a serum sample collected on day 22 day after illness onset.
Future studies on the relevance to human health of the North American Tacaribe serocomplex viruses should include defi ning the clinical spectrum and epidemiology of human disease caused by these viruses. Some of these viruses may cause aseptic meningitis, encephalitis, or meningoencephalitis. Thus, human disease caused by North American Tacaribe serocomplex viruses may be confused with severe encephalitis caused by LCMV, especially in persons who report recent exposure to rodents.