Biogeography of Tick-Borne Bhanja Virus (Bunyaviridae) in Europe

Bhanja virus (BHAV) is pathogenic for young domestic ruminants and also for humans, causing fever and affections of the central nervous system. This generally neglected arbovirus of the family Bunyaviridae is transmitted by metastriate ticks of the genera Haemaphysalis, Dermacentor, Hyalomma, Rhipicephalus, Boophilus, and Amblyomma. Geographic distribution of BHAV covers southern and Central Asia, Africa, and southern (partially also central) Europe. Comparative biogeographic study of eight known natural foci of BHAV infections in Europe (in Italy, Croatia, Bulgaria, Slovakia) has revealed their common features. (1) submediterranean climatic pattern with dry growing season and wet mild winter (or microlimatically similar conditions, e.g., limestone karst areas in central Europe), (2) xerothermic woodland-grassland ecosystem, with plant alliances Quercetalia pubescentis, Festucetalia valesiacae, and Brometalia erecti, involving pastoral areas, (3) presence of at least one of the tick species Haemaphysalis punctata, Dermacentor marginatus, Rhipicephalus bursa, and/or Hyalomma marginatum, and (4) presence of ≥60% of the 180 BHAV bioindicator (157 plant, 4 ixodid tick, and 19 vertebrate spp.). On that basis, Greece, France (southern, including Corsica), Albania, Spain, Hungary, European Turkey, Ukraine (southern), Switzerland (southern), Austria (southeastern), Germany (southern), Moldova, and European Russia (southern) have been predicted as additional European regions where BHAV might occur.

Modern predictive methods for potential geographic distribution of organisms, for example, of ixodid ticks or pathogens they transmit, are usually based on "remote sensing" satellite imagery combined with the use of geographic information system [21].In this contribution, another approach is presented that is based on a "close (ground) sensing" of data.The procedure is characterized by  comparing known natural foci of BHAV within Europe to extract their typical features and select those characters that can be used as factors predicting potential presence of the virus in other geographic areas [22].

Description of Natural Foci of Bhanja Virus.
Eight natural foci of BHAV in Europe were selected (Figure 2) being defined as those areas where the virus was either isolated from ixodid ticks or its presence indicated by a very high seroprevalence rate (≥50%) among local domestic  Description of biota focused on terrestrial vascular plants (Tracheophyta), ixodid ticks (Ixodidae-Table 1) and vertebrates (Vertebrata), and based on checklists of species present in the eight European areas: a total of 2517 spp.terrestrial plants, 28 spp.ixodid ticks (Table 1) and 96 mammalian, 172 avian, and 40 reptilian spp.were included, each species occurring in at least one of the areas.A great number of resources were utilized for the survey: Flora Europaea [28][29][30][31][32] plus national and regional floral checklists, monographs [33][34][35][36][37][38][39][40] for geographic distribution of ixodid ticks, and an extensive collection of literature for vertebrates [22].

Numerical Comparative Analysis of the Foci.
The data recorded as + or (+) were evaluated as the presence, − as the absence, while the variables with a doubtfull occurrence (?) in at least one of the two areas compared pairwise were omitted from the similarity calculations of the particular pair of areas.Numerical comparative pairwise analysis among the BHAV foci was carried out using the Jaccard similarity coefficient for binary (presence-absence) qualitative data, and Euclidean distance for normalized quantitative climatic data [41].Matrices of both the similarity coefficient and Euclidean distance values were then subjected to the average linkage UPGMA cluster analysis [42], and the results were expressed as dendrograms of relationships (similarity) among the areas.

Selection of Bioindicators.
The procedure for selection of "bioindicators" of Bhanja virus (i.e., those species indicating potential presence of BHAV in an area) consisted in the following steps at scanning the checklists of biota (i.e., terrestrial flora, ixodid, and vertebrate faunas): (1) "constant species" were considered those species present in all (8) or nearly all (6-7) BHAV areas; (2) "bioindicators" were then selected as those constant species having a relatively narrow ecological amplitude and a restricted distribution over Europe (i.e., ubiquitous and/or generalist/euryoic species were omitted).

Physiographic, Climatic, and Biotic Similarity of the
European Foci of Bhanja Virus.The results of topographical, geological, pedological, and gross vegetational characters (102 in total) revealed a great similarity between the foci C and S on the one hand, among B, F, and K (all three are karst areas) on the other, while only a partial similarity between V and G and a marked dissimilarity of A (Figure 6).
Figure 7 presents climadiagrams of annual course of temperature and precipitation, while Figure 6(a) shows the results of cluster analysis of the eight BHAV natural foci based on 39 climatic variables.The latter analysis indicates marked similarity among F, G, and B in their climate; other The species is -, absent; +, present; (+), rare; ?, dubious in the area.For letter abbreviations of areas, see Section 2.
Italian foci (S, V, C) are also related, and Bulgarian A as well.
On the other hand, Slovak K is considerably different from all other foci in its climate.Cluster analysis of terrestrial flora among the foci resulted in a group consisting of three Central-Italian foci (F, G, V), related floristically to B, South-Italian foci (C, S) formed another group, while Bulgarian A is less similar, and Slovak K the least related focus (Figure 6).When the plant species restricted to Eumediterranean and littoral zone ("−m") were excluded from the set, the result was slightly different: Central-Italian foci were floristically more similar to South-Italian foci than to Croatian B. Again, A and K deviated, the latter more than the former.The topology of dendrograms of avifaunal and herpetofaunal similarity of the foci was identical, and the dendrogram based on mammalian faunas deviated only negligibly (Figure 6).All these three It means that there is a notable congruence between the similarity of foci based on fauna of Ixodidae (a number of them being vectors of BHAV) and that based on terrestrial flora, whereas vertebrate faunas yielded a different pattern [22].
Calculations of the congruence among the biotic and abiotic components of the foci by Pearson correlation coefficient [42] revealed that ixodid fauna reflects best the terrestrial flora and also climate, much less correlates with vertebrate fauna, whereas physiographic features do not correlate with the other components.In other words, there is an evident, statistically significant relation between climate, terrestrial vascular plants, and ixodid ticks, where the climate is the ultimate causative factor; vegetation reflects these abiotic weather conditions and affects (shelter, etc.) or indicates the formation of tick synusies, while much less closely that of vertebrate communities.Climate has the Mediterranean pattern, with peak rainfall in the winter (Figures 3 and 4), and a dry, hot summer period or, in central Europe (K, Slovak Karst-Figure 5), there are extrazonal xerothermic habitats on karstified limestone approaching meso-and microclimatically those in southern Europe.Mean annual precipitation is 600-1100 mm, air temperature 8-15 (1) submediterranean climatic pattern with dry growing season and wet mild winter (or mesoclimatically similar, e.g., karstic conditions in Central Europe),
Table 2 shows suggested bioindicators of BHAV natural foci in Europe.Predicted countries and areas in Europe where BHAV could potentially occur have been based on the presence of a sufficiently high (≥60%) proportion of species of the bioindicator set (Figure 8).The countries (areas) with predicted occurrence of Bhanja virus include the following (Table 3): Italy, former Yugoslavia (present Slovenia, Croatia, Serbia, Monte Negro, Macedonia, Bosnia), Bulgaria, Greece, France (southern), Albania, Romania, Spain, Hungary, European part of Turkey, Ukraine (southern), Slovakia (southern), Switzerland (Tessin), Austria (southeastern), Corsica, Sardinia, Germany (southern), Moldova, European Russia (southern), Portugal, and possibly Crete.On the other hand, European countries with improbable occurrence of Bhanja virus are Belgium, Czechland (Czech Republic), Poland, Great Britain, The Netherlands, Belarus, Sweden, Denmark, Ireland, Lithuania, Norway, Latvia, Estonia, Finland, and Iceland.

Discussion
Eight areas in Europe were selected for comparison and called natural foci.In general, a natural focus (nidus) of a disease is an area where the agent circulates in an ecosystem between vertebrates and haematophagous arthropod vectors for many years [43,45].In a natural focus, the agent should be therefore found repeatedly (in consecutive years) either directly (by isolation from arhropod vectors or vertebrate hosts) or indirectly, by detection of antibodies in vertebrates.All eight areas selected in this study fulfil this criterion of a natural focus.For instance, even in the most "deviating" and northernmost area, Slovak Karst, antibodies to BHAV were first detected in goats and sheep in 1976 [46], then repeatedly in the years 1982 and 1983 [22], and BHAV was isolated from local Dermacentor marginatus ticks in 1987 [9].In general, groups of species or whole communities are often more reliable as bioindicators than individual species.Therefore only the presence of substantial proportion of the proposed set of indicator species can have a sufficient predictive power.In this study, the cut-off level was established tentatively at 60%.
A unique opportunity to verify the set of predictors of BHAV occurrence established in this study is a comparison of two close and in many respects very similar Dalmatian islands Brač and Hvar (Croatia).There is, namely, a known, potent natural focus of BHAV infections on Brač (B in this study), whereas no known BHAV focus on Hvar.Vesenjak-Hirjan et al. [27] detected antibodies to BHAV in 31.5% of inhabitants (875 examined) on Brač but only in 1.0% (512 examined) on Hvar, though antibodies to, for example, sandfly fever were distributed homogeneously among inhabitants of both islands: 62.1% and 59.4%, respectively.The authors noted that "the underlying reasons for such a great difference on two neighbouring islands have not as yet been explained."However, the suggested set of predictors of BHAV foci does enable clear differentiation of both islands in the following crucial factors: (1) plant community Ostryo-Carpinion adriaticum (order Quercetalia pubescentis), occurring in the central elevated part of Brač, does not appear on Hvar; (2) very limited numbers of domestic ruminants graze on Hvar (typical extensive pastures are completely lacking) contrasting to their large numbers (even for centuries) on Brač; (3) population density of BHAV vector ticks (Haemaphysalis punctata, Dermacentor marginatus) is (consequently) much lower on Hvar than on Brač.It is therefore probable that this predictive approach could well be used to test the probability of potential BHAV occurrence even in individual districts within a particular country.
The possibility of indication of natural foci of tickborne encephalitis (TBE) by the presence of specific plant communities was proposed by parasitologists and botanists more than four decades ago [47][48][49].However, Ixodes ricinus, the vector of TBE, is a generalist (euryoic) tick species with a very wide ecological valence and therefore much less amenable to predictive analysis.On the other hand, prediction is relatively easier in the specialist (stenoic) species like some metastriate ticks, including vectors of BHAV, having a rather narrow ecological valence.These species are generally bound to certain specific habitats and the analysis of their distribution can thus be more straightforward and effective.
The present paper shows that a "close/ground sensing," although not so sophisticated, is an alternative to remote sensing and applicable for prediction of geographic distribution of ixodid ticks and pathogens they transmit.A prerequisite for this alternative approach is existence of corresponding databases and/or checklists of abiotic and biotic variables in respective areas or a complex field survey.An anonymous reviewer of this manuscript suggested that BHAV infection could have followed humans with their domestic animals (Bos indicus) infested with one-host vector ticks such as Boophilus spp.from Asia to Africa and a mixture of African and Asian cattle (Bos taurus) into the Mediterranean.Additionally, primaeval movements of tickinfested sheep and goats from Asia to Africa and Europe could play an important role in the dispersion of BHAV.

Figure 2 :
Figure 2: Bhanja virus in Europe at the time of the study start: dot: BHAV isolated; circle with four short lines: antibody positivity ≥50% of domestic ruminants; circle: seropositivity <50%.The letters indicate the selected foci; for their abbreviations see Section 2.
dendrograms showed the formation of three clusters of foci: (a) all Italian, (b) Croatian B, (c) A and K. On the other hand, ixodid faunas yielded a dendrogram (Figure 6(b)) very similar to that based on Tracheophyta, with formation of two clusters and two singletons: (a) Central-Italian foci (F, G, V) and Croatian B; (b) South-Italian foci (C, S), (c) Bulgarian A, and (d) Slovak K.

Figure 6 :
Figure 6: Dendrograms of similarity of the eight Bhanja virus areas in Europe, based on Jaccard coefficient (C J ) or Euclidean distance (d jkclimate), and cluster analysis of the values.(a) Physiographic, climatic, and floral similarity; "−m," Eumediterranean plus maritime littoral plant species were omitted (b) Faunal (ixodid and vertebrate) similarity.

Figure 7 :
Figure 7: Climate characteristics of the eight Bhanja virus areas in Europe: climadiagrams sensu Walter (monthly course of mean air temperature and precipitation; mean annual temperature and mean total precipitation).

Figure 8 :
Figure 8: Prediction of potential Bhanja virus area in Europe, based on proportion of the bioindicator species present in particular countries and areas.

Table 1 :
Checklist of ixodid fauna in eight European BHAV areas (example of one of the three checklists of biota, viz Ixodidae, Vertebrata, and Tracheophyta).

Table 2 :
Bioindicator species of the potential Bhanja virus occurrence in Europe.

Table 3 :
Bhanja virus-prediction of occurrence in European countries.