Serological study of hantavirus in man in the Autonomous Community of Madrid, Spain

The samples were obtained according to the ethical standards of the Alcala University committee on human experimentation and in accordance with the Helsinki declaration of 1975, as revised in 1983.

Indirect immunofluorescence assay (IFA)
Sera were tested by IFA, as described previously by Lee et al. [10]. Puumala (strain Hallnäs-B1), Hantaan (strain 76/118) and Seoul (strain 80/39) viruses were used. Viruses were propagated in Vero E6 cells (ATCC CRL 1586) and fixed on spot slides. The fluorescein-labelled conjugate used was a rabbit anti-human IgA, IgG and IgM serum (Sigma), diluted 1 in 128 in PBS containing Evan's blue. Sera showing a typical pattern of fluorescence at titres 32 were considered positive.

Western blots (WB)
These were performed according to the protocols described by Jenison et al. [11]. A recombinant plasmid containing the Seoul N gene expressed from pET 23b vector (Novagen, Madison, WI, USA) was provided by B. Hjelle (Department of Pathology, University of New Mexico, USA). The recombinant nucleocapsid protein was expressed with Escherichia coli BL21 (Novagen). Briefly, the viral antigen was applied to SDS-polyacrylamide 412% gels (BioRad, Hercules, CA, USA) and transferred electrophoretically to nitrocellulose membranes. The membranes were allowed to dry and then cut lengthwise into 2-mm-wide strips with a hand-held paper shredder. The strips were stored at 4°C in PBS buffer containing non-fat dried milk 5%. Serum samples were pre-incubated for 6 h in a Western blot tray (BioRad) with a blocking reagent that consisted of a detergent lysate of E. coli in PBS buffer containing milk 5%, at 1 in 400 dilution [11]. The antigen-containing strips were then placed into the wells in the tray and incubated overnight at 4°C. The strips were rinsed in detergent-PBS buffer (10 mM sodium phosphate, pH 7.4, deoxycholic acid 0.5%, Triton X-100 0.5%, 0.1 M NaCl), and incubated with a 1 in 1000 dilution of alkaline phosphatase-conjugated goat anti-human IgG (Sigma) for 4 h and then rinsed again. Nitroblue tetrazolium and 5-bromo-4-chloro-3-indoyl-phosphate substrate (Sigma) were added for 10 min and finally the strips were rinsed with distilled water.

Statistical analysis
Differences in proportions in two-way tables were made, by the S² or the Fisher's exact test, with Stat View^® software in Apple^TM format.

Seven sera were from male (0.38%) and five from female (0.24%) subjects, age range 1678 years with a mean of 50.1 years. The age distribution of seroprevalence is shown in Fig. 3. No statistical differences were found according to sex or age.

Three sera from the north-western area were positive (0.19%), eight from the south-eastern area (0.69%) and one (0.08%) from the central area. The prevalence in the south-eastern area was significantly higher than in the other areas (p <0.05).

Nine (0.23%) sera reacted in IFA with Puumala virus, nine (0.23%) with Seoul virus and seven (0.18%) with Hantaan virus. Six serological patterns were distinguished, the most frequent being reactivity to all three viruses (33.3%). Titres of the positive sera ranged from 32 to 512.

In Spain, epidemiological studies have been performed in the mainly rural populations of Soria [6] and Guadalajara [17] provinces, with seroprevalence rates of 2.2% and 1.8%, respectively, by IFA. In Cataluña, rates of 0.32.3% were reported [7, 18]. Therefore, the seroprevalence found in the ACM is very similar to those found in other regions of Spain.

The present study found no significant difference in seroprevalence by sex, similar to a number of other reports, although the male:female ratio of clinical epidemic nephropathy cases may not be 1 : 1, and in Russia was reported as 5 : 1 [6, 17, 19, 20].

The pattern of age-related seroprevalence rates reported here is similar to those from other European studies, e.g., Sweden [19], with the highest prevalence at age 5574 years. Antibodies against hantavirus remain detectable for decades after infection [16, 21] and, therefore, a progressive increase in seroprevalence is to be expected. A seroprevalence of 30% has been found in elderly subjects in Finland [22].

The highest seroprevalence in this study was in the south-eastern area. This region is dotted with numerous rural municipalities dedicated principally to agriculture; in comparison, the central area is typically urban, and the north-western area combines both urban and rural municipalities, with cattle-raising being an important activity. In general, hantavirus infection affects rural populations more than urban ones [1].

Six serological patterns of reactivity were identified by IFA. The most frequent pattern in the central and north-western areas was reactivity against all three viruses used. In comparison, studies in Italy and Germany suggest that seroreactivity is commonest against Puumala virus [12, 14]. Clethrionomys glareolus, the natural reservoir for this virus in Europe, is not present in the ACM, which may explain the decreased importance of Puumala-like virus in this area. Also, the present study was unable to confirm seroreactivity against Puumala virus by Western blot, which may have led to an underestimate of seroreactivity against viruses antigenically related to Puumala. The nucleocapsid protein of Seoul virus may not provide a good substrate for detection of cross-reactivity with Puumala virus [23].

The wild rats Rattus rattus and R. norvergicus are the natural reservoir for Seoul virus. These have a world-wide distribution, including Europe, and are also present in the ACM [24], suggesting that the seroreactivity detected against this virus in the present study and a study in Soria [6] may be real. Antibodies specific to Seoul virus were detected by IFA among sera from the general population, patients with acute renal failure and in rodents, in a serological study from Ireland [25]. Seoul antigen was detected in rodent lungs from Portugal [26].

In summary, this study confirms the presence of infection by hantavirus in the ACM. It appears that there is more than one circulating serotype; however, further study, by plaque reduction neutralisation tests, will be necessary to define these precisely. The sensitivity of serological tests may also be improved by use of local strains, when they become available. A better understanding of these infections, including investigation of natural viral reservoirs and of clinical manifestations of infection, is an important goal in Spain.