Saaremaa Hantavirus Should Not Be Confused with Its Dangerous Relative, Dobrava Virus

In a recent paper, Klempa and coauthors described an isolate from a striped field mouse (Apodemus agrarius) captured in Slovakia (4). They concluded that this hantavirus, which they call Dobrava virus (DOBV)-Aa, is responsible for most of the DOBV-caused cases of hemorrhagic fever with renal syndrome (HFRS) in central Europe. While we appreciate this contribution, we also feel that in that paper some opinions on the controversial issue of DOBV and Saaremaa virus (SAAV) have not been represented. We wish to correct these in the following paragraphs.

First, we will discuss nomenclature. The name SAAV, first suggested in 2000 (1), is in line with the old tradition in arbovirus and robovirus (for rodent borne) taxonomy; discoverers of a new virus name it after the geographic area where it was first found. Had we named the newly discovered virus properly at the beginning, i.e., in 1997 (13), there would not be any problem. But the name SAAV was not coined until 1999, when, in a collaborative study, we detected the cocirculation of DOBV and SAAV in Slovenia in association with their respective rodent hosts, A. flavicollis and A. agrarius. Reproductive isolation of these two hantaviruses—one is allowed to use such a term after the virus species concept has been introduced (see, e.g., reference 16)—was a crucial piece of information to draw a demarcation line. These results were published in 2000 (1). By that time, the first SAAV isolate had been reported from Estonia (9) and the virus was also found in Russia (12). In a paper published in the next year (14), SAAV was reported in Slovakia and it was suggested to call it “DOBV-Aa” (from A. agrarius). We feel that this name is not correct, and not for priority reasons only; it is also misleading. DOBV is known to be one of the most dangerous European viral pathogens, with a fatality rate of associated HFRS of around 10% (Table 1). In contrast, no fatalities have so far been associated with SAAV, not even in Estonia and Latvia, where human anti-SAAV seroprevalence is about 3% (6, 17). In addition to the paper by Klempa et al. and other studies from the same laboratory (e.g., reference 14), our unpublished results on HFRS patients in Germany, Slovakia, and Estonia confirm the low pathogenicity of SAAV for humans. Clinical data indicate that SAAV infections are much less severe than DOBV infections, perhaps even milder than Puumala virus infections. In line with this, our studies with animal models clearly distinguished SAAV from DOBV; i.e., SAAV infections did not harm mice, while DOBV killed 100% of the animals infected (5). It obviously makes a major difference for patients to be diagnosed with SAAV and not DOBV infection.

TABLE 1.

Differences between SAAV and DOBV

Parameter	SAAV	DOBV	Reference(s)
Rodent carrier	Apodemus agrarius	Apodemus flavicollis	1, 3, 9, 13, 14
Where first discovered	Estonia (Saaremaa Island)	Slovenia (Dobrava Village)	3, 13
Other areas where found	European Russia, Slovakia, Slovenia, Hungary, Germany, Denmark	Albania, Greece, Serbia, Croatia, Hungary, Slovakia	17 (review)
Reproductive isolation documented	Slovenia (Prekmurje), Slovakia (Kosice region)	Slovenia (Prekmurje), Slovakia (Kosice region)	1, 14
Lethality for humans (%)	None reported	9-12	2, 10
Lethality for suckling mice	No	Yes	5
Antigenic distinction by rabbit sera	Yesa	Yesa	9
Antigenic distinction by human sera	Yesb	Yesb	3a

^aA two- to fourfold or greater difference.

^bA greater-than-fourfold difference in the majority of human convalescent-phase sera.

In their paper, Klempa et al. stated that the SAAV isolate from Estonia is “apparently a reassortant.” This conclusion is based on the conflicting phylogenies inferred for the S/L and M segment sequences of the SAAV (DOBV-Aa) and DOBV (DOBV-Af) strains. We have discussed this specific issue in detail elsewhere (15). Here we wish to emphasize that there is an alternative point of view (7, 11, 15) which is based on the results of in-depth phylogenetic analyses using not only the TreePuzzle method (which Klempa et al. used) but also the classic maximum-likelihood, maximum-parsimony, and distance matrix methods. The important conclusion from these studies is that all SAAV (DOBV-Aa) S sequences are monophyletic; i.e., they share a most recent common ancestor which is different from the most recent common ancestor shared by all DOBV strains (DOBV-Af).

The main reason for difficulties in inferring correct S segment-based phylogeny might be a host switch which probably occurred in the evolution of DOBV and SAAV (8, 11). A likely scenario is that pre-DOBV colonized another host, A. agrarius, establishing pre-SAAV. In the course of evolution, the housekeeping N and L proteins (and the encoding S and L genome RNA segments) of the two viruses have been diverging more slowly than surface glycoproteins G1 and G2 (and the encoding M segment), which are involved in the recognition of a host cell receptor(s). Consequently, M/G1G2 sequences have accumulated more mutations than S/N and Lsegm/Lprot sequences, making phylogenetic reconstructions easier. By the way, the L segment-based phylogeny inferred by Klempa et al. is currently based on a short part of the sequence (541 out of 6,500 nucleotides) and therefore awaits further investigation.

In spite of this controversy, it seems that we may agree that SAAV (DOBV-Aa) and DOBV (DOBV-Af) are distinct entities. We have summarized their differences in Table 1. It is important to make the distinction since the two viruses differ drastically in pathogenicity and circulate in the same geographic regions in central and southeastern Europe.