Natural Host Relationships of Hantaviruses Native to Western Venezuela

Abstract

Strains of Caño Delgadito virus (CADV) and Maporal virus (MAPV) were isolated from 25 (8.9%) of the 280 rodents captured on farms in 1997 in western Venezuela. The results of analyses of laboratory and zoographic data indicated that Alston's cotton rat (Sigmodon alstoni) is the principal host of CADV, horizontal virus transmission is the dominant mode of CADV transmission in Alston's cotton rat in nature, a pygmy rice rat (Oligoryzomys sp.) is the principal host of MAPV, and the natural host relationships of CADV and MAPV are highly specific.

Introduction

Specific members of the order Rodentia are the principal hosts of the hantaviruses for which natural host relationships have been well characterized. For example, the deermouse (Peromyscus maniculatus) in the western United States and Canada is the principal host of Sin Nombre virus (SNV) (Childs et al. 1994, Drebot et al. 2001), the hispid cotton rat (Sigmodon hispidus) in southern Florida is the principal host of Black Creek Canal virus (BCCV) (Rollin et al. 1995, Glass et al. 1998), and the long-tailed pygmy rice rat (Oligoryzomys longicaudatus) in Argentina and Chile is the principal host of Andes virus (Levis et al. 1998, Calderon et al. 1999, Cantoni et al. 2001, Torres-Perez et al. 2004).

The hantaviruses known to occur in Venezuela are Caño Delgadito virus (CADV) and Maporal virus (MAPV). The results of analyses of nucleotide and amino acid sequence data in a previous study (Fulhorst et al. 2004) indicated that CADV and MAPV are different species in the virus family Bunyaviridae, genus Hantavirus.

The CADV prototype strain VHV-574 was originally isolated from an Alston's cotton rat (Sigmodon alstoni) captured in 1994 on Hato Maporal (HM), a farm near the town of Caño Delgadito in Portuguesa State (Fulhorst et al. 1997). Other evidence that CADV is naturally associated with Alston's cotton rat in western Venezuela is limited to the CADV-specific RNA from an Alston's cotton rat captured in Cojedes State and the CADV-specific RNA from an Alston's cotton rat captured in Barinas State (Fulhorst et al. 1997). The MAPV prototype strain HV 97021050 was originally isolated from a pygmy rice rat (Oligoryzomys sp.) captured in 1997 on HM (Fulhorst et al. 2004). The objective of this study was to extend our knowledge of the natural host relationships of CADV and MAPV in western Venezuela.

Materials and Methods

Study sites

Small rodents were trapped at three sites (CH-1, CH-2, and CH-3) on a farm near the town of Caño Hondo (CH) in Cojedes State (9°39′N, 68°35′W) and three sites (HM-1, HM-2, and HM-3) on HM (8°48′N, 69°27′W). The sites included sorghum fields, fallow fields, patches of thick herbaceous vegetation alongside fallow fields, tall grassy areas alongside a fallow field, and tall grassy areas alongside a grove of trees and unpaved road (Table 1).

Table 1.

Prevalence of Antibody to Caño delgadito Virus or Maporal Virus in Rodents Captured in 1997 in Western Venezula, by Trap Site and Transect^a

CH						Prevalence of antibody-positive rodents, by speciesb
Site	Transect	Ecotype	No. of traps	No. of trap-nights	Trap success rate	Osp	Sals	Zbre	Hano	Nsp	Psp
CH-1	CH-1A	Tall grass	50	100	29/100 (29.0%)	—	10/21	0/8	—	—	—
CH-1	CH-1B	Fallow field	50	100	22/100 (22.0%)	—	4/13	0/9	—	—	—
CH-2	CH-2	Thick herbaceous vegetation	140	280	78/280 (27.9%)	0/5	11/58	0/15	—	—	—
CH-3	CH-3A	Thick herbaceous vegetation	20	20	5/20 (25.0%)	—	—	0/1	0/4	—	—
CH-3	CH-3B	Sorghum field	33	33	3/33 (9.1%)	0/1	—	0/2	—	—	—
CH-3	CH-3C	Thick herbaceous vegetation	17	17	7/17 (41.2%)	0/5	—	0/1	0/1	—	—
CH-3	CH-3D	Thick herbaceous vegetation	10	10	2/10 (20.0%)	—	—	0/1	—	—	0/1
CH-3	CH-3E	Sorghum field	20	20	6/20 (30.0%)	0/1	0/2	0/1	0/1	0/1	—
		Total	340	580	152/580 (26.2%)	0/12	25/94	0/38	0/6	0/1	0/1

HM						Prevalence of antibody-positive rodents, by speciesb
Site	Transect	Ecotype	No. of traps	No. of trap-nights	Trap success rate	Osp	Sals	Zbre	Hano	Nsp	Psp
HM-1	HM-1	Tall grass	80	160	36/160 (22.5%)	1/2	5/14	0/20	—	—	—
HM-2	HM-2A	Tall grass	60	240	27/240 (11.3%)	0/1	1/13	0/13	—	—	—
HM-2	HM-2B	Fallow field	40	160	24/160 (15.0%)	1/3	4/12	0/9	—	—	—
HM-2	HM-2C	Fallow field	60	240	31/240 (12.9%)	—	0/16	0/15	—	—	—
HM-3	HM-3	Fallow field	100	200	10/200 (5.0%)	1/1	0/2	0/7	—	—	—
		Total	340	1000	128/1000 (12.8%)	3/7	10/57	0/64	—	—	—

^aTraps were set on two consecutive nights at CH-1 and CH-2, one night at CH-3 during the last week of January 1997, two consecutive nights at HM-1 and HM-2, one night at HM-3 during the first week of February 1997, two consecutive nights at HM-2, and one night at HM-3 during the second week of February 1997.

^bNumber antibody-positive/number tested. Osp, Oligoryzomys species (pygmy rice rat); Sals, Sigmodon alstoni (Alston's cotton rat); Zbre, Zygodontomys brevicauda (short-tailed cane mouse); Hano, Heteromys anomalus (Caribbean spiny pocket mouse); Nsp, Neacomys species (bristly mouse); Psp, Proechimys species (spiny rat).

CH, Caño Hondo; HM, Hato Maporal.

Safety

The rodents were captured and processed in the field in a manner described previously (Tesh et al. 1993). Personal protective equipment worn while processing the rodents included the Breathe-Easy^® powered air-purifying respiratory system with Breathe-Easy 10 headpiece (Racal Health and Safety, Frederick, MD), disposable gowns, and latex rubber gloves. All laboratory work with potentially infectious materials was done inside a biosafety level 3 laboratory or biosafety level 4 laboratory located on the campus of The University of Texas Medical Branch at Galveston.

Capture and processing of rodents

The rodents were captured in aluminum live-capture traps (model LFATDG; H. B. Sherman Traps, Tallahassee, FL) baited with small pieces of freshly cut pineapple. Traps were set on two consecutive nights at CH-1 and CH-2, one night at CH-3 during the last week of January 1997, two consecutive nights at HM-1 and HM-2, one night at HM-3 during the first week of February 1997, and two consecutive nights at HM-2 and one night at HM-3 during the second week of February 1997. The traps at each site were placed on transects at 8-m intervals, set 1 h before sunset, and checked at daybreak the following day. Each rodent was assigned a unique identification (FHV) number and then killed by exposure to a lethal dose of vaporized chloroform. The identification number, trap site and trap number, date of capture, species, sex, body weight, total length (nose to tip of tail), length of tail, and other information from each rodent were recorded on a standardized form. Samples of whole blood, lung, spleen, and kidney were collected from the animals captured on the farm near CH. These samples plus oropharyngeal (OP) swabs and samples of liver and urine were collected from the animals captured on HM. Secretions and other materials collected on an OP swab were mixed with 0.3 mL of 0.01 M phosphate-buffered saline (PBS; pH 7.4) containing 10% v/v heat-inactivated fetal bovine serum. Urine was collected by cystocentesis and mixed with 0.3 mL of PBS containing 10% v/v heat-inactivated fetal bovine serum. All samples were stored individually in cryovials in liquid nitrogen and subsequently shipped on dry ice to The University of Texas Medical Branch at Galveston. The carcasses of the pygmy rice rats, spiny rat (Proechimys sp.), bristly mouse (Neacomys sp.), and Caribbean spiny pocket mice (Heteromys anomalus) were deposited into Museo de Ciencias Naturales, Universidad Nacional Experimental de Los Llanos Occidentales “Ezequiel Zamora”, Guanare, Portuguesa. The carcasses of the Alston's cotton rats and short-tailed cane mice (Zygodontomys brevicauda) were burned at the field sites to minimize the risk of human exposure to Guanarito virus, which is the etiological agent of Venezuelan hemorrhagic fever (Salas et al. 1991) and naturally associated with Alston's cotton rat and the short-tailed cane mouse in Portuguesa State and Cojedes State (Fulhorst et al. 1999, Weaver et al. 2000).

Antibody assay

The blood samples were rendered noninfectious by gamma radiation (5 × 10⁶ rads, Co⁶⁰ source), diluted 1:20 v:v in PBS, and then tested for IgG against CADV strain VHV-574 and MAPV strain HV 97021050, using an indirect fluorescent antibody test (IFAT). The cell spots contained uninfected Vero E6 cells mixed 1:1 with Vero E6 cells infected with either VHV-574 or HV 97021050. Antibody (IgG) bound to cell-associated antigen was revealed using a fluorescein isothiocyanate–conjugated goat antibody raised against mouse (Mus musculus) IgG (Kirkegaard and Perry Laboratories, Gaithersburg, MD). Endpoint titers against VHV-574 and HV 97021050 were measured in the positive samples, using serial twofold dilutions beginning at 1:20 and ending at 1:640.

Virus assay

The samples from the OP swabs, crude 10% w/v homogenates of the samples of lung, spleen, and kidney in PBS, and samples of urine were assayed for hantavirus in a manner described previously (Fulhorst et al. 2002). Briefly, the samples were inoculated onto the monolayers of Vero E6 cells grown in 12.5-cm² plastic culture flasks; the monolayers were maintained under a fluid overlay at 37°C in a humidified atmosphere of 5% CO₂ in air; half of the overlay on each monolayer was replaced with fresh maintenance medium on day 7 postinoculation (PI); cell spots were prepared from the monolayers on day 13 or 14 PI; the cultures (passage history: E6+1) were stored at −80°C, and the cell spots were tested for hantavirus antigen, using an IFAT in which the primary antibody was a hyperimmune mouse ascitic fluid raised against SNV and the secondary antibody was a fluorescein isothiocyanate–conjugated goat antibody raised against mouse (M. musculus) IgG (Kirkegaard and Perry Laboratories). The IFAT-negative cultures (passage history: E6+1) from the antibody-positive rodents and three antibody-negative rodents were thawed and then tested for hantavirus by inoculation of 0.4 mL of cell culture material onto monolayers of Vero E6 cells grown in 12.5-cm² plastic culture flasks. Cell spots prepared from these cultures on day 13 or 14 PI were tested for hantavirus antigen as described above.

Genetic characterization of viruses

The nucleotide sequences of a 545- to 566-nt fragment of the small (S) genomic segments and 743- to 756-nt fragment of the medium (M) genomic segments of five viruses (Table 2) were determined to provide an assurance that the hantaviruses isolated from Alston's cotton rats in this study are strains of CADV and that the hantaviruses isolated from pygmy rice rats captured on HM are strains of MAPV. We note that the MAPV prototype strain HV 97021050 was isolated from the spleen of pygmy rice rat FHV-4083 (Fulhorst et al. 2004) and that this rodent was captured at HM-3 in this study (Table 3). The 545-nt sequence encoded a 181-aa fragment of the nucleocapsid (N) protein of CADV strain VHV-574, the 566-nt sequence encoded a 183-aa fragment of the N protein of strain HV 97021050, the 743-nt sequence encoded a 247-aa fragment of the glycoprotein precursor (GPC) of strain VHV-574, and the 756-nt sequence encoded a 251-aa fragment of the GPC of strain HV 97021050. Total RNA was isolated from monolayers of infected Vero E6 cells, using TRIzol^® Reagent (Invitrogen Life Technologies, Carlsbad, CA). The CADV S segment first-strand cDNA, MAPV S segment first-strand cDNA, CADV M segment first-strand cDNA, and MAPV M segment first-strand cDNA were synthesized using SuperScript II RNase H⁻ Reverse Transcriptase (Invitrogen Life Technologies) in conjunction with oligonucleotides HTS63 (5′-TAGTAGACTCCTTGAGAAGCTACTACGAC-3′), HTS90 (5′-TAGTAGTAGACTCCTTGAGAAGCTA-3′), HTM57 (5′-TAGTAGTAGACTCCGCACGAAGAAGC-3′), and HTM7 (5′-TAGTAGTAGACTCCGCAAGAAGAAGCA-3′), respectively. Subsequently, a 545-nt fragment of the CADV S segment first-strand cDNA, 566-nt fragment of the MAPV S segment first-strand cDNA, 743-nt fragment of the CADV M segment first-strand cDNA, and 756-nt fragment of the MAPV M segment first-strand cDNA were amplified using GoTaq^® DNA polymerase (Promega, Madison, WI) in conjunction with HTS12 (5′-GGTGTGATTTCATCTGCYTTCAT-3′) and HTS15 (5′-AAGCTGTAATGAGCACCCTCAAAG-3′), HTS42 (5′-AGCCTTCATAGTGGACTGTGC-3′) and HTS90, HTM30 (5′-TAGCCTGATACTGTGTTTCC-3′) and HTM42 (5′-GTTATGGAGCCWGGATGGACTG-3′), and HTM95 (5′-TGTAACCAGAAATAGTATTCC-3′) and HTM94 (5′-GTGCAGAAACAACACAAGTGGAGC-3′), respectively. Both strands of each gel-purified polymerase chain reaction product were sequenced directly, using the Big Dye^® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA). The nucleotide sequences of the S and M segments of HV 97021154, HV 97021192, HV 97021299, HV 97021320, and HV 97021290 were deposited into the GenBank nucleotide sequence database under accession nos. EU939919 through EU939922, FJ008979, EU867771, EU910960 through EU910962, and FJ008980, respectively.

Table 2.

Hantaviruses Isolated from Rodents Captured in 1997 in Western Venezuela and Selected for Genetic Characterization

Virus	Animal no.	Species	Date captured	Tissue	Locality	Site
HV 97021154	FHV-3907	Sals	Jan 29, 1997	Lung	CH	CH-1
HV 97021192	FHV-3945	Sals	Jan 29, 1997	Lung	CH	CH-2
HV 97021299	FHV-4052	Sals	Feb 4, 1997	Lung	HM	HM-2
HV 97021320	FHV-4073	Sals	Feb 5, 1997	Lung	HM	HM-1
HV 97021290	FHV-4043	Osp	Feb 4, 1997	Lung	HM	HM-1

Table 3.

Sex, Size class, and Antibody Titers of 21 Virus-Positive Alston's Cotton Rats and 4 Virus-Positive Pygmy Rice Rats Captured in 1997 in Western Venezuela

Animal no.	Speciesa	Sex	Size class (length)b	Trap site	Antibody titersc	Virus-positive specimen(s)
FHV-3874	Sals	M	IV (153)	CH-1	320, <20	L, K
FHV-3875	Sals	M	III (134)	CH-1	320, 80	L, K
FHV-3877	Sals	M	III (128)	CH-1	≥640, 160	L, K
FHV-3878	Sals	M	II (125)	CH-1	20, <20	L, S, K
FHV-3900	Sals	F	IV (149)	CH-1	320, 80	L, K
FHV-3902	Sals	F	III (126)	CH-1	≥640, 80	L, K
FHV-3903	Sals	M	III (128)	CH-1	160, 20	L, S, K
FHV-3907d	Sals	M	III (132)	CH-1	40, <20	L
FHV-3945d	Sals	M	III (141)	CH-2	320, 80	L, S, K
FHV-3962	Sals	F	IV (146)	CH-2	320, 160	L, K
FHV-3989	Sals	M	II (120)	CH-1	20, <20	L, S
FHV-3993	Sals	M	III (130)	CH-1	160, 40	L, S, K
FHV-4013	Sals	M	III (141)	CH-2	80, 20	L, S, K
FHV-4014	Sals	F	III (129)	CH-2	80, 20	L, K
FHV-4017	Sals	M	III (131)	CH-2	<20, <20	L
FHV-4052d	Sals	F	II (114)	HM-2	160, 20	L, S, K, OP swab, U
FHV-4061	Sals	F	III (139)	HM-2	320, 80	L, S, K
FHV-4063	Sals	M	IV (148)	HM-1	160, 20	K
FHV-4073d	Sals	M	III (143)	HM-1	160, 40	L, S, K
FHV-4074	Sals	M	IV (147)	HM-1	40, 20	L, S, OP swab, U
FHV-4133	Sals	M	IV (150)	HM-2	80, 20	K
FHV-4043d	Osp	M	— (82)	HM-1	<20, <20	L, S, K
FHV-4083e	Osp	F	— (83)	HM-3	80, 80	L, S, K, U
FHV-4129	Osp	M	— (83)	HM-2	320, 160	L, U
FHV-4134	Osp	M	— (99)	HM-2	<20, <20	L

^aThe Alston's cotton rats are arranged by locality, with CH followed by HM.

^bSize class (nose-to-rump length, measured in mm).

^cAntibody titers to CADV strain VHV-574 and MAPV strain HV 97021050, respectively.

^dThe nucleotide sequences of a fragment of the nucleocapsid protein gene and fragment of the glycoprotein precursor gene of the hantavirus isolated from lung tissue from this animal were determined in this study.

^eMAPV prototype strain 97021050 was originally isolated from the spleen of pygmy rice rat FHV-4083 (Fulhorst et al. 2004).

L, lung; S, spleen; K, kidney; OP swab, oropharyngeal swab; U, urine; CADV, Caño Delgadito virus; MAPV, Maporal virus.

Data analysis

Antibody titers <20 were considered 10 in comparisons of antibody titers to CADV and MAPV in individual blood samples. The homologous virus in an antibody-positive sample was assumed to be the virus that was associated with the highest titer if the absolute value of the difference between the titers to CADV and MAPV was ≥4-fold.

The N protein gene sequences and GPC gene sequences of HV 97021154, HV 97021192, HV 97021299, HV 97021320, and HV 97021290 were compared with the homologous sequences of CADV strain VHV-574 (GenBank accession nos. DQ285566 and DQ284451) and MAPV strain HV 97021050 (GenBank accession nos. AY267347 and AY363179). The alignments of the N protein gene sequences and GPC gene sequences were constructed using CLUSTAL W1.7 (Thompson et al. 1994). Identities between the GPC sequences and between the N protein gene sequences were calculated using MEGA, version 4.0 (Tamura et al. 2007).

The male Alston's cotton rats were assigned to four size classes based on their nose-to-rump lengths (measured in mm): I, 87–110 (n = 14); II, 111–126 (n = 28); III, 127–143 (n = 43); and IV, 144–153 (n = 13). Similarly, the female cotton rats were assigned to four size classes based on their nose-to-rump lengths (measured in mm): I, 83–108 (n = 8); II, 109–125 (n = 15); III, 126–143 (n = 23); and IV, 144–156 (n = 7). The upper boundary of class I was the mean length minus one standard deviation, the upper boundary of class II was the mean length, the upper boundary of class III was the mean length plus one standard deviation, and the upper boundary of class IV was the longest nose-to-rump length. The acceptable type I error in all statistical tests was α = 0.05.

Results

Two hundred and eighty rodents were captured in 1580 trap-nights on two farms in western Venezuela (Table 1), with an overall trap success rate of 17.7%. Antibody (IgG) to CADV or MAPV was found in 35 (23.2%) of the 151 Alston's cotton rats, 3 (15.8%) of the 19 pygmy rice rats, and none of the 110 other rodents (Table 1). The prevalence of antibody in the cotton rats captured on the farm near CH was not significantly different from the prevalence of antibody in the cotton rats captured on HM (Fisher's exact test, two tailed, p > 0.23). The three antibody-positive pygmy rice rats were captured on HM.

Antibody titers to CADV and MAPV in the antibody-positive cotton rats ranged from 20 to ≥640 and from <20 to 160, respectively. Likewise, antibody titers to CADV and MAPV in the three antibody-positive pygmy rice rats ranged from 40 to 320 and from 40 to 160, respectively. The apparent homologous virus in 30 cotton rats was CADV. The apparent homologous virus in the five other antibody-positive cotton rats and three antibody-positive pygmy rice rats could not be determined because the difference between the antibody titers to CADV and MAPV in each of these animals was less than fourfold.

Hantavirus was isolated from 20 (57.1%) of the 35 antibody-positive cotton rats, 1 (0.9%) of the 116 antibody-negative cotton rats, 2 (66.7%) of the 3 antibody-positive pygmy rice rats, 2 (12.5%) of the 16 antibody-negative pygmy rice rats, and none of the 110 other rodents. Hantavirus antigen was found in cell spots from 4 (30.8%) of the 13 Vero E6 cell cultures inoculated with original specimens from three antibody-negative animals and cell spots from 30 (45.4%) of the 66 Vero E6 cell cultures inoculated with original specimens from the 18 antibody-positive animals. Subsequently, hantavirus antigen was found in cell spots from 1 (11.1%) of the 9 Vero E6 cell cultures inoculated with IFAT-negative E6+1 materials from the 3 antibody-negative, culture-positive animals and cell spots from 25 (48.1%) of the 52 Vero E6 cell cultures inoculated with IFAT-negative E6+1 materials from the 38 antibody-positive animals in this study. Overall, the tests on the IFAT-negative E6+1 materials increased the number of culture-positive specimens from 34 to 60 and the number of culture-positive animals from 21 to 25. The culture-positive specimens included the samples of lung from 19 cotton rats and 4 pygmy rice rats, spleen from 10 cotton rats and 2 pygmy rice rats, kidney from 17 cotton rats and 2 pygmy rice rats, OP swabs from 2 cotton rats, and urine from 2 cotton rats and 2 pygmy rice rats (Table 3). The results of the pairwise comparisons of N protein amino acid sequences and pairwise comparisons of GPC amino acid sequences indicated that HV 97021154, HV 97021192, HV 97021299, and HV 97021320 are strains of CADV and that HV 97021290 is a strain of MAPV (Table 4).

Table 4.

Identities Between the Amino Sequences of a 177-AA Fragment of the Nucleocapsid Proteins and Between the Amino Acid Sequences of a 247-AA Fragment of the Glycoprotein Precursors of Five Strains of Caño Delgadito Virus and Two Strains of Maporal Virus Isolated from Rodents Captured in Western Venezuela

				Nucleocapsid protein (% amino acid sequence identity)
Virus	Straina	Host	Locality	VHV-574	HV 97021154	HV 97021192	HV 97021299	HV 97021320	HV 97021050	HV 97021290
CADV	VHV-574	Sals	HM	—	99.4	99.4	99.4	100.0	87.6	87.0
CADV	HV 97021154	Sals	CH	99.6	—	100.0	98.9	99.4	87.0	86.4
CADV	HV 97021192	Sals	CH	99.2	99.6	—	98.9	99.4	87.0	86.4
CADV	HV 97021299	Sals	HM	99.6	99.2	98.8	—	99.4	87.6	87.0
CADV	HV 97021320	Sals	HM	100.0	99.6	99.2	99.6	—	87.6	87.0
MAPV	HV 97021050	Osp	HM	86.2	86.2	85.8	85.8	86.2	—	98.9
MAPV	HV 97021290	Osp	HM	86.2	86.2	85.8	85.8	86.2	100.0	—
					Glycoprotein precursor (% amino acid sequence identity)

^aCADV strain VHV-574 was isolated from an Alston's cotton rat captured in 1994. The six other viruses were isolated from rodents captured in 1997.

Strains HV 97021154 and HV 97021192 were isolated from cotton rats captured on the farm near CH, and strains HV 97021299 and HV 97021320 were isolated from cotton rats captured on HM (Table 2). The antibody titers to CADV in these four cotton rats were at least fourfold higher than the corresponding antibody titers to MAPV. Similarly, the antibody titers to CADV in 26 of the 31 other antibody-positive cotton rats were at least fourfold higher than the corresponding antibody titers to MAPV. Collectively, these results indicate that a large majority (if not all) of the 35 antibody-positive cotton rats in this study were infected with CADV.

Twenty-six (27.7%) of the 94 cotton rats captured on the farm near CH and 10 (17.5%) of the 57 cotton rats captured on HM were infected with a hantavirus (i.e., antibody positive and/or culture positive) (Table 5). The prevalence of infection in the cotton rats captured on the farm near CH was not significantly different from the prevalence of infection in the cotton rats captured on HM (Fisher's exact test, two tailed, p > 0.17).

Table 5.

Prevalence of Hantaviral Infections in Alston's Cotton Rats Captured in 1997 in Western Venezuela, by Locality, Sex, and Size Class

	CH		HM
Size classa	Male	Female	Male	Female
I	1/8 (12.5%)	0/2 (0.0%)	0/6 (0.0%)	0/6 (0.0%)
II	5/23 (21.7%)	1/9 (11.1%)	0/12 (0.0%)	1/8 (12.5%)
III	13/29 (44.8%)	1/12 (8.3%)	2/10 (20.0%)	2/9 (22.2%)
IV	2/5 (40.0%)	3/6 (50.0%)	5/5 (100.0%)	0/1 (0.0%)
Total	21/65 (32.3%)	5/29 (17.2%)	7/33 (21.2%)	3/24 (12.5%)

^aSize class: I (smallest nose-to-rump lengths) to IV (largest nose-to-rump lengths).

Eight (10.8%) of the 74 small cotton rats (size classes I and II combined), 28 (36.4%) of the 77 large cotton rats (size classes III and IV combined), 22 (44.9%) of the 49 large male cotton rats (size classes III and IV combined), and 6 (21.4%) of the 28 large female cotton rats (size classes III and IV combined) were infected with a hantavirus (Table 5). The prevalence of infection in the 77 large cotton rats was significantly different from the prevalence of infection in the 74 small cotton rats (Fisher's exact test, two tailed, p < 0.01), but the prevalence of infection in the 49 large male cotton rats was not significantly different from the prevalence of infection in the 28 large female cotton rats (Fisher's exact test, two tailed, 0.05 < p < 0.06).

Hantavirus was isolated from 4 (57.1%) of the 7 pygmy rice rats captured on HM and none of 12 pygmy rice rats captured on the farm near CH. MAPV strain HV 97021290 was isolated from pygmy rice rat FHV-4129 and, as indicated previously, MAPV strain HV 97021050 was isolated from pygmy rice rat FHV-4083. Both of these rodents were antibody positive to CADV and MAPV (Table 3). The two other culture-positive pygmy rice rats (FHV-4043 and FHV-4134) were antibody negative to CADV and MAPV (Table 3). No attempt was made to determine the identities of the hantaviruses isolated from the antibody-negative pygmy rice rats. Thus, these rodents may have been infected with a hantavirus other than MAPV.

Discussion

Collectively, the isolation of CADV from 21 (13.9%) of the 151 Alston's cotton rats captured on farms in different states of western Venezuela and the absence of evidence for CADV infection in the 129 other rodents in this study indicate that Alston's cotton rat is the principal host of CADV. Similarly, the isolation of MAPV from 2 (28.6%) of the 7 pygmy rice rats and the absence of evidence for MAPV infection in the 121 other rodents captured on HM strengthen the notion that a pygmy rice rat (Oligoryzomys sp.) is the principal host of MAPV.

The results of published studies on BCCV, SNV, and other hantaviruses indicated that horizontal virus transmission is an important mode of intraspecific virus transmission in naturally infected rodent populations (e.g., Glass et al. 1998, Mills et al. 1998). The failure to isolate hantavirus from the 22 cotton rats in size class I in this study suggests that vertical transmission of CADV in Alston's cotton rat is an uncommon event. Together, the lack of evidence for vertical virus transmission and the positive association between prevalence of infection and cotton rat size in this study indicate that horizontal virus transmission is the dominant mode of CADV transmission in Alston's cotton rat and likely critical to the long-term maintenance of CADV in nature.

The antibody-positive male cotton rat in size class I (Table 5) had an anti-CADV IgG titer of 20 and was culture negative. Hypothetically, CADV infections in utero and in newborn cotton rats are lethal, the anti-CADV antibody in the male cotton rat in size class I was maternal in origin, and maternal antibody protects Alston's cotton rats against CADV infection during the first weeks of their lives (Zhang et al. 1989).

Allogrooming, mating, fighting, and other activities that entail close physical contact may facilitate horizontal transmission of CADV among Alston's cotton rats. Studies on the ecologies of SNV and other hantaviruses revealed a higher prevalence of anti-hantavirus antibody in adult male rodents than in adult female rodents (e.g., Mills et al. 1998, Glass et al. 1998). This association between prevalence of antibody and sex has been attributed to the more combative nature of adult male rodents relative to the intraspecific social behavior of their female counterparts. The lack of a statistical association between prevalence of infection and sex in this study could be because of the small sample size. Alternatively, adult male Alston's cotton rats are no more combative than adult female cotton rats and/or intraspecific aggression does not contribute significantly to CADV transmission among Alston's cotton rats in nature.

Thirty-two (84.2%) of the 38 short-tailed cane mice and 92 (97.8%) of the 94 cotton rats from the farm near CH were captured in 110 (45.8%) of the 240 traps on three transects (i.e., CH-1A, CH-1B, and CH-2) (Table 1). Similarly, 57 (89.1%) of the 64 short-tailed cane mice and 55 (96.5%) of the 57 cotton rats from HM were captured in 91 (37.9%) of the 240 traps on four transects (i.e., HM-1, HM-2A, HM-2B, and HM-2C) (Table 1). These observations suggest that the short-tailed cane mice lived in close physical association with Alston's cotton rats. Yet there was no evidence that any of the 102 short-tailed cane mice were infected with CADV. This may be because short-tailed cane mice are refractory to CADV infection. Alternatively, social interactions between short-tailed cane mice and Alston's cotton rats in nature are rare or not conducive to cotton rat-to-cane mouse transmission of CADV.

Some Oligoryzomys species are difficult to distinguish from other Oligoryzomys species on external morphological features alone. A previous publication indicated that rodent FHV-4083 in this study was a fulvous pygmy rice rat (Oligoryzomys fulvescens) (Fulhorst et al. 2004); however, the results of a recent analysis of cytochrome-b gene sequence data indicated that FHV-4083 is a pygmy rice rat but not a fulvous pygmy rice rat (John D. Hanson, unpublished results). Accordingly, FHV-4083, the 6 other oligoryzomine rodents captured on HM, and the 12 oligoryzomine rodents from the farm near CH were listed as pygmy rice rats (Oligoryzomys sp.) rather than fulvous pygmy rice rats (O. fulvescens) in this publication.

Hantavirus pulmonary syndrome (Koster and Levy 2006) is a potentially fatal zoonosis caused by SNV, Andes virus, and some of the other hantaviruses principally associated with members of the rodent family Cricetidae (Musser and Carleton 2005). It is generally accepted that humans usually become infected with hantaviruses by inhalation of aerosolized droplets of saliva, respiratory secretions, or urine from infected rodents or by inhalation of dust particles contaminated with virus. The isolation of CADV from OP swabs and urine from cotton rats in this study suggests that secretions and excretions from CADV-infected Alston's cotton rats are infectious to humans. Similarly, the isolation of MAPV from urine of the pygmy rice rats in this study suggests that urine from MAPV-infected pygmy rice rats is infectious to humans. Whether CADV or MAPV in humans is pathogenic has not yet been investigated.

Acknowledgments

National Institutes of Health grants AI-39800 (Mechanisms of emergence of zoonotic viral pathogens), AI-63235 (Animal models of hantavirus cardiopulmonary disease), and AI-67947 (Viral determinants of hantavirus pulmonary disease in the hamster) provided financial support for this study. James Meegan, Patricia Repik, and Christina Cassetti (National Institutes of Health) facilitated the grant support for this study. Robert B. Tesh (World Reference Center for Emerging Viruses and Arboviruses, Galveston, Texas) provided the anti-SNV hyperimmune mouse ascitic fluid.

Disclosure Statement

No competing financial interests exist.