New York 1 and Sin Nombre Viruses Are Serotypically Distinct Viruses Associated with Hantavirus Pulmonary Syndrome

Abstract

New York 1 virus (NY-1) and Sin Nombre virus (SN) are associated with hantavirus pulmonary syndrome (HPS). NY-1 and SN are derived from unique mammalian hosts and geographic locations but have similar G1 and G2 surface proteins (93 and 97% identical, respectively). Focus reduction neutralization assays were used to define the serotypic relationship between NY-1 and SN. Sera from NY-1-positive Peromyscus leucopus neutralized NY-1 and SN at titers of ≥1/3,200 and ≤1/400, respectively (n = 12). Conversely, SN-specific rodent sera neutralized NY-1 and SN at titers of <1/400 and 1/6,400, respectively (n = 13). Acute-phase serum from a New York HPS patient neutralized NY-1 (1/640) but not SN (<1/20), while sera from HPS patients from the southwestern United States had 4- to >16-fold-lower neutralizing titers to NY-1 than to SN. Reference sera to Hantaan, Seoul, and Prospect Hill viruses also failed to neutralize NY-1. These results indicate that SN and NY-1 define unique hantavirus serotypes and implicate the presence of additional HPS-associated hantavirus serotypes in the Americas.

Hantaviruses are enveloped negative-stranded RNA viruses with a tripartite genome and comprise a distinct genus within the Bunyaviridae family (42). Hantaviruses are present worldwide and cause two human diseases: hemorrhagic fever with renal syndrome (HFRS) and hantavirus pulmonary syndrome (HPS) (41, 51). Each hantavirus is carried primarily by a specific small mammal host which is persistently infected (4, 5, 14, 19, 22, 27, 41). Hantaviruses are transmitted to humans through the inhalation of aerosolized excreta (29, 41).

In the Americas, hantaviruses are the cause of HPS, an acute respiratory distress syndrome with a 45% mortality rate. HPS was first recognized in patients in the southwestern United States in 1993 and has subsequently been identified in 28 states and Canada (18, 36). Recently identified HPS cases in South America indicate that HPS-associated hantaviruses are widely dispersed and that some HPS-associated hantaviruses may be transmitted from person to person (11, 13, 25).

Two integral membrane surface glycoproteins, G1 and G2, are present on the surface of hantaviruses (44, 50). Antibodies to G1 and G2 neutralize the virus, distinguish viral serotypes, and protect animals from hantavirus infection (1, 2, 6, 7, 9, 31). At present 11 distinct serotypes of hantavirus have been established: Hantaan (HTN), Puumala (PUU), Seoul (SEO), Dobrava, Khabarovsk, Thailand, Thottapalayam, Tula, Prospect Hill (PH), Sin Nombre (SN), and Black Creek Canal (BCC) (3–5, 8, 10, 23, 27, 28, 39, 40, 49). Thus far, HPS-associated viruses are represented by two serotypes, SN and BCC, with highly divergent G1 and G2 glycoproteins (38). SN is the prototype HPS-associated strain derived from the deer mouse, Peromyscus maniculatus, in the southwestern United States (10). In contrast, BCC was isolated from the Florida cotton rat, Sigmodon hispidus, and as a result BCC and SN are from discrete geographic locations and host species (6, 38, 39). Bayou virus (BAY) has also been associated with an HPS case in Louisiana (host, Oryzomys palustris), and El Moro Canyon virus (host, Reithrodontomys megalotis) exhibits similarity to HPS viruses but has yet to be associated with HPS (19, 26, 34, 48).

The New York 1 hantavirus (NY-1) is similarly derived from an isolated geographic location, an island off the coast of New York (47). NY-1 is also associated with HPS, and we isolated NY-1 from a unique host species, the white-footed mouse Peromyscus leucopus (47). NY-1 and BCC surface glycoproteins are also highly divergent. However, NY-1 and SN are more closely related and share 93 and 97% amino acid identities in their G1 and G2 proteins, respectively (22, 35).

In this study, we addressed the question of whether the 3 to 7% difference between NY-1 and SN glycoproteins specifies unique or common serotypic determinants. Reciprocal focus reduction neutralization (FRN) assays were performed on NY-1 and SN in order to determine their antigenic relationship. We report that serum neutralizing antibody titers to heterologous hantaviruses are 4- to 32-fold lower than those from animal or human sera to homologous hantaviruses. As a result, NY-1 and SN elicit unique neutralizing antibody responses and define discrete hantavirus serotypes. These findings indicate that 3 to 7% differences in hantavirus glycoproteins can confer serotypic differences between hantaviruses and further suggest that additional HPS-associated serotypes are likely to be identified in the Americas.

MATERIALS AND METHODS

Cells, media, and viruses.

Vero E6 cells were grown in Dulbecco’s Modified Eagle’s Medium (DMEM) containing 10% fetal calf serum (FCS, 56°C heat inactivated), penicillin (100 mg/ml), streptomycin sulfate (100 mg/ml), and amphotericin B (5 mg/ml). SN (CC107) (40), NY-1 (47), and PH (PH-1) (30) were grown on Vero E6 cells (ATCC CRL 1586) (12, 40, 43) in a biosafety level 3 facility. SN, PH, and NY-1 were adsorbed onto Vero E6 cell monolayers for 1 h, washed, and grown in maintenance medium (DMEM–2% FCS) (12, 40, 43). Maximal titers of NY-1, SN, and PH were between 5 × 10⁶ and 1 × 10⁷ focus forming units (FFU)/ml.

Sera.

Hyperimmune hamster reference sera to HTN, PUU, SEO, and PH were kindly provided by Ho Wang Lee at the World Health Organization Regional Center for HFRS, Asan Institute for Life Science, Seoul, Korea. The human sera used included one sample collected from a fatal case of HPS in New York (1995), six acute-phase serum samples from HPS patients in New Mexico, and eight convalescent-phase serum samples from HPS patients in New Mexico, California, and Texas. Rodent sera were collected from 12 P. leucopus from Long Island and Shelter Island, New York, and from small rodents (four species) from California, New Mexico, and Texas. Two human HPS cases have occurred in New York, and the infecting viruses were identified serologically and genetically as NY-1 (21, 22, 33) (Serum is currently available only from the second case, April 1995.) Human and animal sera were heat inactivated for 30 min at 56°C prior to neutralization assays. All sera tested reacted with the SN and NY-1 nucleocapsid proteins by immunofluorescence assay (IFA) and Western blotting (14, 21, 24, 33).

Expression of the SN nucleocapsid protein in baculovirus recombinants has been previously described (45). Rabbits were prebled and immunized subcutaneously with the SN N protein purified by preparative sodium dodecyl sulfate-polyacrylamide gel electrophoresis, electroelution from gel slices, and extensive dialysis against phosphate-buffered saline (PBS). The immunization scheme employed was four immunizations with TiterMax adjuvant (CytRx Corp.) at 10- to 14-day intervals. Approximately 50 μg of purified N protein was used for initial immunizations. Subsequent immunizations were performed with 20 μg of purified recombinant protein. Serum was collected 10 days after the final immunization and assayed for reactivity with bacterially expressed N protein as well as by immunofluorescence and immunoperoxidase staining of NY-1-infected Vero E6 monolayers (5 days postinfection [p.i.]).

IFA.

Virus-infected Vero E6 cells (NY-1, PH, PUU, or SEO) were bound to gelatin-coated glass slides (1 by 3 in.) and fixed with cold acetone, as previously described (27). All sera were tested in twofold dilutions on Vero E6 cells previously infected with the indicated viruses. For each primary virus-specific antibody, a species-specific secondary fluorescein-conjugated anti-immunoglobulin was used for the detection of virus proteins (Kirkegaard & Perry). Pooled human, preimmune rabbit, hamster, and negative P. leucopus sera served as negative controls.

Immunoperoxidase staining of NY-1-infected cells.

In order to establish the kinetics of NY-1 infection in Vero E6 cells we used an immunoperoxidase staining method for detecting hantavirus antigens in infected cells (15). Vero E6 cells were infected, the inocula were removed from cells, and at various times p.i. monolayers of infected cells were fixed with 100% methanol (−20 C) for 10 min. Following three washes in PBS, rabbit anti-nucleocapsid hyperimmune sera (made to pET30a-expressed SN N protein, 1/5,000) was added to cells in PBS–1% bovine serum albumin for 1 h. Cell monolayers were washed 5 times with PBS and were reacted with a horseradish peroxidase anti-rabbit conjugate (1/2,000) for 1 h. Monolayers were washed three times, and nucleocapsid protein-containing infected cells were quantitated following staining with 3-amino-9-ethylcarbazole (0.026%) in 0.1 M sodium acetate (pH 5.2)–0.03% H₂O₂ as previously described (15, 46).

FRN assay.

To evaluate neutralizing antibody titers, twofold serial dilutions of sera were mixed with approximately 200 FFU of NY-1, SN, or PH in 100 μl of PBS with 2% FCS for 1 h at 37°C. Virus was adsorbed to confluent monolayers of Vero E6 cells in duplicate wells of a 96-well plate. Viral inocula were removed, and following washing, cells were incubated in DMEM–2% FCS for 24 to 36 h. Viral antigen present in infected cells was detected by immunoperoxidase staining as described previously (15, 46). Infected cells were quantitated and compared to mock-neutralized controls. Neutralizing titers are the inverse of the maximum serum dilution that resulted in a >60% reduction in the number of infected foci. A cutoff of >80% neutralization uniformly resulted in a less than or equal to twofold reduction in serum neutralization titers.

RESULTS

Detection of hantavirus nucleocapsid protein in infected cells.

Hyperimmune rabbit sera to the baculovirus-expressed nucleocapsid protein of SN virus was used to detect NY-1-, SN-, and PH-infected cells by immunostaining of infected monolayers (Fig. 1) (15). Immunoperoxidase staining of the nucleocapsid protein within infected Vero E6 cells was detected as early as 12 to 24 h p.i. (Fig. 1B). The number of infected single cells increased from 24 to 48 h p.i. (Fig. 1C). Cell-to-cell spread of virus from initially infected cells resulted in an increase in the size of foci (two to three infected cells). Large foci of NY-1-infected cells were observed by 72 h p.i. along with additional single infected cells (Fig. 1D). Similar results were observed with SN- and PH-infected Vero E6 cells (not shown). The kinetics of N protein appearance in single cells suggested that antigen detected within the first 24 h p.i. was a result of input virus and that subsequent groups of infected cells resulted from amplification and spread of the virus within the monolayer. Single-step kinetics were similarly obtained at 24 h p.i. by endpoint dilution of NY-1 followed by immunoperoxidase staining as well as by titration of the virus and analysis by IFA. SN- or NY-1 sera (patient or animal) were indistinguishable by IFA from NY-1- or SN-infected cells, respectively (Table 1), through their N protein reactivity.

FIG. 1.

Immunoperoxidase staining of NY-1 proteins in Vero E6 cells. Vero E6 cells in 24-well plates were infected, the inocula were removed from cells, and at various times p.i. monolayers of infected cells were fixed with 100% methanol (−20°C) for 10 min. Following three washes in PBS, rabbit anti-nucleocapsid hyperimmune sera (made to the baculovirus-expressed SN nucleocapsid protein, 1/5,000) were added to cells in PBS–1% BSA for 1 h. Cell monolayers were washed five times with PBS and reacted with a horseradish peroxidase anti-rabbit conjugate (1/2,000) for 1 h. Monolayers were washed three times and were developed with the horseradish peroxidase substrate amino-ethyl-carbazole, producing a brown precipitate in cells (15, 46). Mock-infected (A) or NY-1-infected (B to D) monolayers were stained at 24 (panels A and B), 48 (panel C), or 72 (panel D) h p.i. Magnification, ×200.

TABLE 1.

NY-1 and SN neutralization antibody titers of sera from HPS patients

Serum no.	Time after onset of disease (days)	Type of virusa	FRN assay titerb
			NY-1	SN
1	5	NY-1	640	<20
2	1	SN	<20	160
3	1	SN	<20	160
4	1	SN	<20	80
4	16	SN	<20	320
5	19	SN	<20	80
6	22	SN	<20	160
8	269	SN	160	640
10	10c	SN	40	320
11	19c	SN	20	320
12	36c	SN	20	160

^aDenotes type of virus that the patient was presumably infected with as previously characterized by reverse transcriptase-PCR and protein blot analysis (18, 24, 36). 

^bFRN assay titers were determined as previously described (15, 46). FRN assay titers are reciprocals of the highest serum dilutions resulting in a 60% reduction in the number of foci of infected cells compared to those of virus pretreated with control serum. Serum FRN assay titers were uniformly reduced twofold by using an 80% FRN cutoff. Serum IFA titers to NY-1- and SN-infected cells were identical, indicating similar antibody responses to the cross-reactive N protein (14, 27, 28). All tested sera were positive by Western blotting to SN N protein as previously described (24). 

^cThese values are numbers of years. 

Serum neutralizing antibodies to NY-1, SN, and PH.

Hantavirus reference sera were studied by FRN assay for their ability to neutralize NY-1. Sera and viruses were preincubated, adsorbed to Vero E6 monolayers, and immunoperoxidase stained 24 to 36 h p.i. By using nucleocapsid protein-specific sera, infected cells were quantitated and compared to controls. HTN, PH, and SEO reference sera had no detectable neutralizing antibody titers against NY-1. PUU reference sera had a low-level neutralizing antibody titer (1/20) against NY-1.

HPS patient serum neutralization of NY-1 and SN.

HPS patient sera were assayed by FRN assay for their ability to neutralize NY-1, SN, and PH. None of the HPS patient sera tested had neutralizing antibody titers to PH virus (<1/20). Acute-phase patient serum from a fatal NY HPS case neutralized NY-1 at dilutions of up to 1/640 but did not neutralize SN (<1/20) (Table 1). In contrast, California- or New Mexico-derived HPS patient sera neutralized SN at 4- to >16-fold-higher titers than the respective neutralization titers to NY-1. Although there are no convalescent-phase NY-1 HPS sera, this demonstrates that convalescent- and acute-phase SN sera differ markedly in their ability to neutralize NY-1 and SN. Similarly, sera from New Mexico, which were retrospectively identified as being SN-positive convalescent-phase sera, had neutralizing antibody titers to SN which were 8- to 16-fold higher than to NY-1.

In order to further demonstrate the specificity of the HPS patient serum neutralization of SN and NY-1, the results of a reciprocal titration of NY-1- and SN-specific sera against NY-1 and SN are presented in Fig. 2. The same patient sera were used to neutralize NY-1 (Fig. 2A) and SN (Fig. 2B). Both sera had homologous neutralization titers of 1/640 but failed to neutralize the heterologous SN or NY-1 at dilutions of 1/20 to 1/2,560. Homologous neutralizing antibody titers were titratable, resulting in <20% neutralization at antibody dilutions of ≤1/2,560 (Fig. 2).

FIG. 2.

Titration of human serum neutralizing antibodies. Twofold serial dilutions of SN- or NY-1-specific sera were mixed with approximately 200 FFU of NY-1 (A) or SN (B) in 100 μl of PBS–2% FCS for 1 h at 37°C. Virus was adsorbed to confluent monolayers of Vero E6 cells in duplicate wells of a 96-well plate. Viral inocula were removed, and following washing, cells were incubated in DMEM–2% FCS for 24 to 36 h. Viral antigen present in infected cells was detected by immunoperoxidase staining and quantitated as described previously (15, 46). Experiments were repeated at least three times, and error bars reflect the range of values obtained from these experiments.

Rodent serum neutralization of SN and NY-1.

P. leucopus, trapped on Long Island in New York, were previously demonstrated to be PCR positive for NY-1 and hantavirus seropositive by IFA and Western blotting (20, 24, 33, 36, 47). Similarly, mammals from California, New Mexico, and Texas were previously demonstrated to be PCR positive and seropositive for SN, BAY, or ELMC (Table 2). Animal sera were tested for their ability to neutralize NY-1, SN, and PH. None of the animal sera neutralized PH (<1/32 to <1/400), while sera from BAY- and ELMC-positive animals had low-level or no neutralizing antibody titers to SN and failed to neutralize NY-1 (Table 2).

TABLE 2.

NY-1 and SN neutralization antibody titers of small mammal sera

Species (no. of anmals tested)	Location of animal	Type of virusa	Serum FRN assay titerb
			NY-1	SN
P. leucopus (12)	New York	NY-1	3,200–6,400c	400
P. maniculatus (7)	California	SN	<400, neg.	12,800
	California	SN	<400, neg.	3,200
	New Mexico	SN	<400, neg.	3,200
	California	SN	<400, neg.d	800
	California	SN	<200, neg.	<200, neg.
	New Mexico	SN	<400, neg.d	NDe
P. truei (1)	California	SN	<400, neg.	12,800
O. palustris (3)	Texas	BAY	<100, neg.	≤200
R. megalotis (2)	California	ELMC	<100, neg.	200–400

^aDenotes virus that animals were infected with as previously characterized by reverse transcriptase-PCR and protein blot analysis (18, 24, 36). 

^bFRN assay titers were determined as previously described (15, 46). FRN assay titers are reciprocals of the highest serum dilutions resulting in a 60% reduction in the number of foci of infected cells compared to those of virus pretreated with control serum. Serum FRN assay titers were uniformly reduced twofold by using an 80% FRN cutoff. neg., FRN assay endpoint titers were not determined below the noted serum dilution. 

^cTwo animals had serum neutralization titers of 1/3,200, and 10 animals had serum neutralization titers of 1/6,400. 

^dInsufficient amount of serum to test at higher concentrations. 

^eND, not determined. 

Twelve P. leucopus serum samples were evaluated for neutralizing antibody titers to NY-1 and SN. All the New York-derived P. leucopus had high neutralizing antibody titers to NY-1 and 8- to 16-fold-lower neutralizing antibody titers to SN (Table 2). In contrast, six animal serum samples from California or New Mexico neutralized SN at titers up to 1/12,800 but failed to neutralize NY-1 (Table 2). Two P. maniculatus serum samples which had low neutralizing antibody titers (1/800) to SN did not reach endpoints in the neutralization of NY-1. However, at the highest concentration in serum tested (1/400) we observed a <20% reduction in the number of NY-1 foci, suggesting these sera also have more than fourfold-lower neutralization titers against NY-1.

DISCUSSION

It is well established that neutralizing antibodies recognize hantavirus G1 and G2 surface glycoproteins (1, 6–8, 32, 37, 41). As a result, differences within G1 and G2 neutralization determinants differentiate hantavirus serotypes and define functional relationships between pathogenic hantavirus strains (41). HPS is now known to be caused by a variety of American hantaviruses; however, only two highly divergent HPS-associated viruses, SN and BCC, have been determined to define unique serotypes (6, 38, 39). With a number of closely related HPS-associated hantaviruses it is important to define serotypic relationships among these viruses and to establish whether common neutralization determinants are present within their G1 and G2 proteins.

In this study, NY-1 and SN, which differ in their G1 and G2 proteins by only 7 and 3%, respectively, were serologically evaluated for the presentation of unique neutralization determinants (21, 35). Sera from New York-derived P. leucopus had 8- to 16-fold-lower neutralizing antibody titers to SN than to NY-1. Conversely, sera from SN-infected animals, with high-titer neutralizing antibody titers to SN, still failed to neutralize NY-1 (<1/400). Sera from BAY- or ELMC-infected animals had no or low neutralizing antibody titers to SN and NY-1.

Serum from one of two fatal New York-derived HPS cases was available and was found to neutralize only NY-1 (16, 21). The fact that the single New York patient serum tested is an acute-phase serum (5 day post onset of disease) further emphasizes that differences in patient serum neutralizing antibody responses are evident early in HPS patients. Interestingly, acute-phase SN sera neutralize homologous SN but fail to neutralize NY-1. Convalescent-phase SN sera contain low-level cross-reactive neutralizing antibodies to NY-1 but have fourfold-lower neutralizing antibody titers to NY-1 than to SN. Since there is no convalescent-phase NY-1 patient serum it is not clear whether similar cross-reactive neutralizing antibody responses to SN are elicited following NY-1 infection.

These results suggest that NY-1 carried by P. leucopus in New York has substantially different neutralization determinants than those of SN or BAY HPS-associated viruses. Chu et al. (6) have previously demonstrated that the HPS-associated BCC and SN define unique serotypes with highly divergent G1 and G2 proteins (amino acid differences of 22 and 16%, respectively) (38). The FRN assay findings presented here also demonstrate the importance of the relatively small number of G1 and G2 amino acid differences between NY-1 and SN (7 and 3%, respectively) that define unique neutralization determinants. Although we have found that PUU sera neutralized NY-1 at a 1/20 dilution, low-level neutralizing serum titers (≤1/20) have previously been demonstrated against a number of heterologous hantaviruses (6, 8).

Interestingly, the potential for neutralization differences between SN and NY-1 were initially indicated by the failure of the New York patient serum to recognize a linear immunodominant epitope of the SN G1 protein (17, 24). In fact, G1 epitope reactivity could be useful for identifying specific hantaviruses during acute HPS. However, it remains to be tested whether the immunodominant G1 epitope is a clear serotypic marker.

Our results demonstrate that NY-1 and SN are distinct HPS-associated hantaviruses and thus bring to three the number of known serotypes of hantaviruses which cause HPS. These results also suggest that a number of additional hantavirus serotypes are likely to be identified even among presumably similar HPS-associated strains. However, these findings also indicate the presence of cross-reactive antigenic determinants on SN and NY-1 which need to be identified and considered for the development of vaccines or antibody-based therapeutics.