Influenza Virus Infection in Guinea Pigs Raised as Livestock, Ecuador

To determine whether guinea pigs are infected with influenza virus in nature, we conducted a serologic study in domestic guinea pigs in Ecuador. Detection of antibodies against influenza A and B raises the question about the role of guinea pigs in the ecology and epidemiology of influenza virus in the region.

To determine whether guinea pigs are infected with infl uenza virus in nature, we conducted a serologic study in domestic guinea pigs in Ecuador. Detection of antibodies against infl uenza A and B raises the question about the role of guinea pigs in the ecology and epidemiology of infl uenza virus in the region.

I nfl uenza A virus infection causes disease in humans and
domestic animals, including pigs, horses, and chickens, and seasonal epidemics among humans in the Northern and Southern Hemispheres result in hospitalizations and deaths worldwide. Infl uenza A virus transmission studies are often conducted in laboratory guinea pigs (1,2) because the virus can effi ciently spread from infected animals to naive guinea pigs by direct and indirect (short-range infectious aerosols) contact (3,4). However, whether guinea pigs are naturally infected with infl uenza virus outside the laboratory setting is not known.
In some regions of South America, guinea pigs are part of the traditional cuisine and are produced as livestock and sold commercially for human consumption. Guinea pigs are customarily raised on small rural farms in proximity to other livestock. Circulation of infl uenza virus in these populations has not been studied. Given the effect of infl uenza virus on human health and the susceptibility of guinea pigs to infl uenza virus infection in the laboratory, it is worthwhile to determine whether infl uenza virus can spread among guinea pigs in agricultural settings. As an initial step in this endeavor, we obtained serum samples from domestic guinea pigs in Ecuador and tested them for the presence of infl uenza antibodies to determine whether the guinea pigs had been infected with infl uenza virus.

The Study
We obtained serum samples from 40 guinea pigs from 3 different regions of Ecuador (Figure 1), 20 from Cuenca and 10 each from Guayaquil and the Manabí region. Cuenca is located in the Andes region, 2,500 m above sea level; it is one of the main producers of guinea pig meat. Guayaquil, the most populated city in Ecuador, is located at the head of the Gulf of Guayaquil on the Pacifi c Ocean. The Manabí region is located in western Ecuador on the Pacifi c Ocean coast. Serum samples were collected from adult guinea pigs that we purchased from local farms (Cuenca), where they had been raised as livestock, or from live animal markets (Guayaquil and Manabí). The samples were collected by heart puncture under general anesthesia (combination of ketamine and xylazine). Animals were euthanized after samples were obtained.
For antigens in the serologic analyses, we used whole viruses and recombinant hemagglutinin, nucleoprotein, and neuraminidase (subtypes N1 and N2) proteins produced in our laboratory as described (5,6).  . Three regions of Ecuador where guinea pig serum samples were obtained: Cuenca, Guayaquil, and Manabi. The country is bordered by Colombia to the north, Peru to the east and south, and the Pacifi c Ocean to the west. Cuenca is located in the Andes; the average annual mean temperature is 14.7°C, and the average annual relative humidity is 85%. Guayaquil is located at the head of the Gulf of Guayaquil; the mean temperature is 26.1°C, and relative humidity is 74%. The Manabí region is located on the Pacifi c Ocean coast; the mean temperature is 25.9°C, and relative humidity is 79%. ELISA was done as described (7), with slight modifications for the use of guinea pig serum. The cutoff for serum considered positive for infl uenza virus was the value of the negative control (naive guinea pig serum) +3 SD (from 3 repetitions). For Western blot (WB) analyses (8), serum samples were pooled into groups of 5 (groups 1-4, 5-6, and 7-8 were from Cuenca, Guayaquil, and Manabí region, respectively). Pooled samples were also used for the hemagglutination inhibition (HI) assay, as described (9), using infl uenza strains Brisbane07, Wisconsin05, and Vietnam04. As positive controls, we used serum from guinea pigs that we had infected with Cal09, Brisbane07, NewCal99, Wis-consin05, Vietnam04, or Yamagata88.
ELISA results for the 40 serum samples showed that 20, 18, and 14 were positive for infl uenza subtypes H1, H3, and H5, respectively ( Table 1). The samples were also tested for the presence of antibodies against the infl uenza virus nucleoprotein: results for 29 were positive. Samples with positive results to >1 hemagglutinin antigens and to nucleoprotein were also analyzed for the presence of antibodies against proteins of the neuraminidase subtypes N1 (Cal09, Vietnam04) and N2 (infl uenza A/Hong Kong/1/1968 [H3N2]) ( Table 1). Serum samples were also tested for the presence of infl uenza B virus by using whole virus, recombinant hemagglutinin, and recombinant nucleoprotein as described above. Samples from Cuenca showed the highest overall positivity to the 3 antigens (Table 2).
WB analysis of pooled samples from the 3 areas confi rmed the ELISA results that showed antibodies against the hemagglutinin and nucleoprotein antigens; however, samples from Manabí region (7,8) showed immunoreactivity to the H3 antigen only (Figure 2). WB results for serum samples from Cuenca and Guayaquil were positive for infl uenza B virus, but results were not positive for samples from Manabí when whole virus was used as antigen ( Figure  2). HI activity was observed in all pooled samples against all 3 viruses tested. Titers were highest in samples from Cuenca (320 for Brisbane07, 640 for Wisconsin05, and 80 for Vietnam04) and lowest for samples from Manabí (80 for Brisbane07 and 40 for Wisconsin05 and Vietnam04). Titers for samples from Guayaquil were 160 for Brisbane07 and 80 for Wisconsin05 and Vietnam04.

Conclusions
We showed the presence of infl uenza virus antibodies with HI activity to different virus subtypes in guinea pigs marketed through local farms and live animal markets in different regions of Ecuador. Seroprevalence was similar for infl uenza A virus subtypes H1 (50%) and H3 (45%). A study of infl uenza among humans in Ecuador also showed similar seroprevalences for subtypes H1 (5.1%) and H3 (5.5%) (10). This10-fold difference between seroprevalence rates among humans and the guinea pigs in our study may be explained by animal husbandry practices that facilitate infl uenza transmission (e.g., the crowding of animals in cages).
The presence of antibodies to infl uenza B virus in guinea pigs is notable because infection with this virus is thought to be restricted to humans (11). Recent studies in our laboratory demonstrated that guinea pigs can also be infected with infl uenza B viruses and that they readily transmit the virus to naive guinea pigs (12). These studies support our fi nding that infl uenza B virus has the potential to be transmitted from humans to other species. Further studies are needed to isolate and characterize the type B infl uenza virus present in the population of guinea pigs to determine if there has been an adaptation to the new host or if the guinea pig is only a transient reservoir for the human virus.
We found that several guinea pigs had antibodies against infl uenza A (H5). Ecuador is purported to be free of avian infl uenza (13); however, reports from other countries in South America demonstrate the presence of avian subtypes H3, H5N2, H7N3, and H13N9 in wild birds (14). In addition, chickens, turkeys, and guinea pigs that are later sold in local farmers' markets are raised together by some families, thereby facilitating contact of guinea pigs with avian infl uenza viruses. We tested only for seroreactivity to the H5 hemagglutinin and the N1 neuraminidase; therefore, further studies are needed to determine whether different avian origin infl uenza viruses are present in the guinea pig population. We did not determine whether guinea pigs are an incidental host for infl uenza virus infection or, if instead, the virus has been adapted to these animals or if guinea pigs are a natural reservoir for some infl uenza viruses. To this end, virus isolation and characterization would be necessary to determine the virus strains circulating in this population. In the laboratory, guinea pigs are infected and effi ciently transmit infl uenza viruses to naive hosts without showing any overt clinical signs of disease (1). Therefore, further studies are needed to address the specifi c role of guinea pigs raised as livestock in the ecology and epidemiology of infl uenza viruses in the region.
This work was partially funded by the Center of Excellence for Infl uenza Research and Surveillance, National Institutes of Health (grant HHSN266200700010C to P.P.) and the Keck Foundation (P.P.). F.K. was supported by an Erwin Schrödinger fellowship (J 3232) from the Austrian Science Fund.
Dr. Leyva-Grado is a postdoctoral fellow at Mount Sinai School of Medicine in New York. His primary research interests include the study of the pathophysiology of virus-induced respiratory infections in humans and domestic animals and the development of therapeutics for virus infections.