PCR detection of Anaplasma phagocytophilum in goat flocks in an area endemic for tick-borne fever in Switzerland

Central Switzerland is a highly endemic region for tick-borne fever (TBF) in cattle, however, little is known about A. phagocytophilum in goats. In the present study, 72 animals from six goat flocks (373 EDTA blood-samples) in Central Switzerland were analysed for A. phagocytophilum DNA. A real-time PCR targeting the msp2 gene of A. phagocytophilum was performed and in positive samples the partial 16S rRNA, groEL and msp4 gene were amplified for sequence analysis. Four DNA extracts were positive. Different sequence types on basis of the amplified genes were found. For comparison, sequences of A. phagocytophilum from 12 cattle (originating from Switzerland and Southern Germany) were analysed. The 16S rRNA gene sequences from cattle were all identical amongst each other, but the groEL and msp4 gene differed depending on the origin of the cattle samples and differed from the variants from goats. This study clearly provides molecular evidence for the presence of different types of A. phagocytophilum in goat flocks in Switzerland, a fact which deserves more thorough attention in clinical studies.


INTRODUCTION
A naplasma (A.) phagocytophilum is an obligate tick-transmitted intracellular bacterium. It causes granulocytic anaplasmosis in dogs, cats, horses and humans and tick-borne fever (TBF) in ruminants (Rikihisa, 1991). TBF in cattle is characterized by high fever, respiratory symptoms, abortion, a sudden decrease in milk production, leucopenia and can cause severe economic losses in ruminants (Woldehiwet, 2006;Stuen, 2007). However, symptoms may vary depending on the infected species, the variant of A. phagocytophilum involved or the age, condition and immune status of the host (Woldehiwet, 2006;Stuen, 2007). A. phagocytophilum can show great genetic heterogeneity between different geographic regions or host species (reviewed in Woldehiwet, 2006). In sheep flocks, different genetic variants of A. phagocytophilum may circulate within one flock (Ladbury et al., 2008). In Switzerland, the first report of TBF in cattle was from the "Bernese Oberland" (Pfister et al., 1987) ( Fig. 1) and further evidence of A. phagocytophilum in cattle in Switzerland was given later (Hofmann-Lehmann et al., 2004). Even though there have been isolated reports on outbreaks of TBF in domestic and free-living feral goats, very little is known about the general presence of A. phagocytophilum in goats (Foster & Greig, 1969;Gray et al., 1988). Experimental infections of sheep and goats with A. phagocytophilum showed that all animals reacted with fever, rickettsiaemia, lymphocytopenia and reduced packed cell volume (PCV). However, fever in sheep set in earlier and lasted longer than in goats (Gokce & Woldehiwet, 1999). The main vector for A. phagocytophilum in central Europe is Ixodes (I.) ricinus (Linné 1758), which Original contribution Parasite, 2011, 18, 57-62 is the most abundant tick, also in Switzerland. There, PCR prevalence with A. phagocytophilum in I. ricinus ranges from 0.5 to 2.2 % (Leutenegger et al., 1999;Pusterla et al., 1999b;Liz et al., 2000;Wicki et al., 2000). In the present study, goat flocks in Central Switzerland were investigated with real-time PCR for the presence of A. phagocytophilum. In positive samples, sequence analysis was performed for partial 16S rRNA, groEL and msp4 gene, comparing A. phagocytophilum appearing in small and large ruminants.

MATERIALS AND METHODS
D uring a study on the implementation of the FAMACHA © test in goats suffering from gastrointestinal nematodes (GIN), faecal samples and EDTA-blood samples for determination of PCV value, were taken monthly from May to October 2008 from six goat flocks from Central Switzerland (Scheuerle et al., 2010). The EDTA-blood samples were kept frozen at -26 °C and were subsequently used in this study for the detection of A. phagocytophilum DNA in goats. Due to the frozen state of the samples, it was not possible to use them for serological examination. The six goat flocks (named A, B, G, J, M, R, according to the first letter of the name of their owners) were situated in the Cantons of Bern, Luzern and Zug (Fig. 1). Their pastures (no stream or wetland; altitude 500-1,100 m above sea level) were situated on steep hills surrounded by woodland, and hence were potential tick habitats. I. ricinus ticks had been identified previously by exemplary flagging close to the study sites. All animals spent the entire sampling period on the pastures. Each flock consisted of at least one buck, 10 to 100 does with ages ranging from one to 13 years and a lamb group. From each flock, eight to 13 animals were included in the study on GIN and subsequently for this investigation on A. phagocytophilum. For flock R in the month of July, blood samples were not available. Overall 373 samples from 72 individuals were tested. To compare the sequence variants involved, frozen EDTA-blood samples from 12 A. phagocytophilumpositive cattle were used. The TBF in cattle had been diagnosed positive previously by detection of morulae in the neutrophilic granulocytes in Giemsa-stained blood smears and the blood had been kept frozen. Altogether there were samples from nine cattle (2005), with typical symptoms of clinical TBF (apathy, high fever, sudden decrease in milk production), and three cattle from Southern Germany (2002), with unknown disease status. The Swiss cattle were from the region south of Interlaken, Switzerland and the German cattle from prealpine regions in Germany ( Fig. 1). This region in Switzerland is highly endemic for TBF in cattle (P.H. Boss, personal communication). DNA extraction was carried out with the High Pure PCR template preparation kit (Roche, Mannheim, Germany) according to the manufacturer's instruction. Quality and quantity of extracts was tested with a spectrophotometer (NanoDrop ® ND-1000, PeqLab, Erlangen, Germany). A real-time PCR targeting the msp2 gene of A. phagocytophilum was performed in an iCycler iQ (Bio-Rad, Munich, Germany) as described previously (Courtney et al., 2004;Silaghi et al., 2008). PCR-grade water served as negative, DNA-extracts of A. phagocytophilum-positive ticks as positive controls. For positive samples, further PCRs were carried out: a nested PCR targeting a part (530 bp) of the 16S rRNA gene (Massung et al., 1998) was performed in a GeneAmp 9700 thermocycler (Applied Biosystems, Darmstadt, Germany) using primers ge3a (5'-CACATGCAAGTCGAACGGATTATTC-3') and ge10r technelysium.com.au), sequence homology searches were made by BLASTn analysis of GenBank (www. ncbi.nlm.nih.gov) and multiple alignments performed with ClustalW (www.ebi.ac.uk).

RESULTS
I n general, the same goats were sampled over the entire period, for variations in the number of samples see Table I. Altogether, four of the 373 samples were positive for A. phagocytophilum in the real-time PCR. Four different goats were positive and each goat was positive only once. Positive goats were found in three of the six flocks (Fig. 1, Table I). The age of the positive goats ranged from one to five years; one was of unknown age. All positive goats were females. The PCV of the positive goats ranged from 23 to 35 %, the physiological range is from 28 to 40 % (Kraft & Dürr, 1999). Three different 16S rRNA gene sequence types differing in up to four nucleotide positions from each other were shown (Table II). In flock B, two different sequence types were identified within the same month. One sequence was identical to an A. phagocytophilum strain from ticks in Sweden and Germany (GenBank accession nos. AJ242783 and AF136714). The other nucleotide sequence obtained in flock B was also identified in flock M (Table II), identical with another one already obtained from ticks in Sweden and Germany, and also from roe deer in the Czech Republic and humans in Canada (GenBank accession nos. AJ242784, EU839847, AF311343 and AF136713). The sequence type identified in flock A was not 100 % identical with sequences in the GenBank database. Differences between 16S rRNA gene sequences from the present study and selected sequences from the GenBank are shown in Table II. All three differed, for example, in up to three nucleotide positions from the sequence obtained from cattle with TBF from  Switzerland, from cattle in Southern Germany and from variants previously shown to be pathogenic for sheep. All cattle were infected with the same 16S rRNA gene variant, regardless of the geographic origin. The sequences obtained in this study were deposited at GenBank with accession nos. FJ538288 to FJ538291. The 530 bp partial groEL gene was amplified in the four goats and 12 cattle (GenBank accession nos. GQ452225 to GQ452232). Three caprine variants differed in two nucleotide positions from each other and in 17 nucleotide positions from the fourth variant, which differed in only up to two positions from the variants of bovine origin. Altogether four different groEL variants of bovine origin were detected. All from Switzerland were the same, whereas the three from Germany were different amongst each other and differed from the Swiss one. Altogether seven different msp4 partial gene variants with differences in a total of 14 nucleotide positions were found. Three different variants were found in the amplified part of the msp4 gene of the goats and four variants were found in the cattle. All Swiss cattle had the same variant, whereas the three German cattle had three different variants. All variants of bovine origin differed from the goat variants (Genbank accession nos. HM028674, HM028675, HM028676, HM028677, HM028678, HM028679, HM028680)

T he results show the molecular presence of
A. phagocytophilum in goats in Central Switzerland. This adds to the knowledge on the epidemiology of A. phagocytophilum in Switzerland, which has been described in the tick I. ricinus (Leutenegger et al., 1999;Pusterla et al., 1999b;Liz et al., 2000;Wicki et al., 2000) and domestic animals like cattle, dogs and horses (Pfister et al., 1987;Pusterla et al., 1997a;Pusterla et al., 1997b;Pusterla et al., 1998a;Hofmann-Lehmann et al., 2004). Serologic evidence has additionally been given in dogs, cattle, red foxes and chamois (Pusterla et al., 1997c;Pusterla et al., 1998b;Pusterla et al., 1999a;Pusterla et al., 1999b;Liz et al., 2002) and also humans (Pusterla et al., 1998c;Weber et al., 2000). Serological or PCR presence of A. phagocytophilum has been described in goats in other countries as well, like Cyprus, Spain and the US (Massung et al., 1998;Amusategui et al., 2006;Chochlakis et al., 2009). None of the owners of the goat flocks observed any symptoms that could be attributed to A. phagocytophilum. However, symptoms of TBF can be very unspecific and the infections may also have been subclinical. The PCR presence of A. phagocytophilum without the development of clinical symptoms has been reported in dogs (Jensen et al., 2007, Galke et al., 2009. Therefore infection in the goats may have been unnoticed so far in the region, despite the endemic presence of A. phagocytophilum in ticks (Pusterla et al., 1999b) and in cattle (Pfister et al., 1987). Tickinfestation of the goats was very rarely observed by the owners. However, ticks are easily overlooked by farmers, as a regular examination of goats for ticks does not occur. In Norway, a flock of lambs showed reduced weight gain in connection with subclinical A. phagocytophilum infection, even though no apparent tick infestation on the pastures was observed (Stuen et al., 2002). Goats in the present study were massively infested with Accession no. a   64  76  84  170  175  376  461  463  464  465  467  468 Host Country  gastrointestinal nematodes (GIN) (Scheuerle et al., 2010) and therefore, reduced weight gain or any other clinical signs due to A. phagocytophilum could have been concealed. The PCV of the four goats was mostly in the physiological range. Here also, low PCV values could have been influenced by the detected infection with the GIN Haemonchus contortus (Scheuerle et al., 2010).

Nucleotide position
In this study, each positive goat was positive only once. Persistence of A. phagocytophilum infection in sheep and cattle has previously been shown, resulting in a cyclic parasitaemia, which has been shown in sheep (Stuen et al., 1998;Stuen et al., 2008).
Furthermore, the incubation period in an experimental infection of cattle with A. phagocytophilum is between five and nine days and the parasitaemic phase of neutrophils lasts up to two weeks (Pusterla & Braun, 1997). As samples were taken at monthly intervals, active infection may have been overlooked and the results do not allow conclusion to be drawn for the entire pasture period. The finding of A. phagocytophilum DNA in goats is important, as, even though no clinical signs were observed which could typically be connected with TBF, infection with A. phagocytophilum is known to increase the susceptibility for secondary infections and could thus contribute to productivity losses. Attention should be paid to keeping goats out of the reach of ticks.
In the investigated goat flocks, three different 16S rRNA gene variants were identified; one flock showed a mixed infection with two different variants in the same month. All these goat variants differed from strains which have previously been shown to be pathogenic for sheep. They also differed from variants from cattle with TBF from the same region in Switzerland.
All 16S rRNA gene sequences in cattle were the same, regardless of the country of origin and the development of clinical symptoms. The clinical cases in the Swiss cattle were very severe, whereas in Germany, no case of TBF in cattle has so far been reported. Therefore, it may well be that host specificity is more important than geographic origin. The groEL gene was amplified in the four goats and 12 cattle. Three caprine variants from three goats were very similar to each other and the fourth one from the fourth goat was very similar to the bovine ones. Four different groEL variants of bovine origin were detected; interestingly, all from Switzerland were the same, whereas the three from Germany differed amongst each other in one nucleotide position. The same was the case for the partial msp4 gene. Therefore, on the basis of these partial genes, there seems to be both geographic and interspecies variation.
The impact of these sequence variations needs further investigation, especially whether there is any influence on the pathogenicity in cattle or the pathogenic potential of A. phagocytophilum in goats. Furthermore, future studies need to be undertaken in monitoring active infections, including also the presence of serum antibodies and clinical parameters in goats.

Disclosure:
The authors declare that no conflict of interest exists.