Experimental Parasitology 148 (2015) 1–16

Contents lists available at ScienceDirect

Experimental Parasitology

journal homepage: www.elsevier.com/locate/yexpr

Full length article spp.: An experimental study on host susceptibility to avian malaria Dimitar Dimitrov a,b,*, Vaidas Palinauskas a, Tatjana A. Iezhova a, Rasa Bernotiene˙ a, Mikas Ilgu¯ nas a, Dovile Bukauskaite˙ a, Pavel Zehtindjiev b, Mihaela Ilieva b,c, Anatoly P. Shapoval d, Casimir V. Bolshakov d, Mikhail Yu Markovets d, Staffan Bensch c, Gediminas Valkiu¯ nas a a Institute of Ecology, Nature Research Centre, Akademijos 2, Vilnius 21, LT-08412, Lithuania b Institute of Biodiversity and Ecosystem Research, Bulgarian Academy of Sciences, 2 Gagarin Street, Sofia 1113, Bulgaria c Department of Biology, Lund University, Ecology Building, Lund S-22362, Sweden d Biological Station Rybachy of the Zoological Institute, Russian Academy of Sciences, Rybachy, Kaliningrad Region 238535, Russia

HIGHLIGHTS GRAPHICAL ABSTRACT

• Species of Haemamoeba and have broad avian host range. • Species of and Huffia have restricted avian host range. • Specificity of different cyt b lineages of the same parasite species is variable. • Susceptibility of the same avian host to various Plasmodium isolates is variable. • are more susceptible to co- infections, which are more virulent.

ARTICLE INFO ABSTRACT

Article history: The interest in experimental studies on avian malaria caused by Plasmodium species has increased re- Received 4 August 2014 cently due to the need of direct information about host–parasite interactions. Numerous important issues Received in revised form 20 October 2014 (host susceptibility, development of infection, the resistance and tolerance to avian malaria) can be an- Accepted 13 November 2014 swered using experimental infections. However, specificity of genetically different lineages of malaria Available online 18 November 2014 parasites and their isolates is largely unknown. This study reviews recent experimental studies and offers additional data about susceptibility of birds to several widespread cytochrome b (cyt b) lineages of Plas- Keywords: modium species belonging to four subgenera. We exposed two domesticated avian hosts (canaries Serinus Avian malaria Plasmodium canaria and ducklings Anas platyrhynchos) and also 16 species of common wild European birds to malaria Experimental infection infections by intramuscular injection of infected blood and then tested them by microscopic examina- susceptibility tion and PCR-based methods. Our study confirms former field and experimental observations about low Parasite specificity specificity and wide host-range of Plasmodium relictum (lineages SGS1 and GRW11) and P. circumflexum

* Corresponding author. Fax: +370 5 272 93 52. E-mail address: [email protected] (D. Dimitrov). http://dx.doi.org/10.1016/j.exppara.2014.11.005 0014-4894/© 2014 Elsevier Inc. All rights reserved. 2 D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16

(lineage TURDUS1) belonging to the subgenera Haemamoeba and Giovannolaia, respectively. However, the specificity of different lineages and isolates of the same parasite lineage differed between species of exposed hosts. Several tested Novyella lineages were species specific, with a few cases of successful development in experimentally exposed birds. The majority of reported cases of mortality and high parasitaemia were observed during parasite co-infections. Canaries were susceptible mainly for the species of Haemamoeba and Giovannolaia, but were refractory to the majority of Novyella isolates. Ducklings were susceptible to three malaria infections (SGS1, TURDUS1 and COLL4), but parasitaemia was light (<0.01%) and transient in all exposed birds. This study provides novel information about susceptibility of avian hosts to a wide array of malaria parasite lineages, outlining directions for future experimental research on various aspects of biology and epidemiology of avian malaria. © 2014 Elsevier Inc. All rights reserved.

1. Introduction ways and might be highly virulent and even lethal in some birds (Atkinson and LaPointe, 2009; Palinauskas et al., 2011). Important Blood parasites of the genus Plasmodium (, question about tolerance and resistance to avian malaria can be an- ) are vector borne pathogens, which cause malaria and swered only using experimental infections (Atkinson et al., 2013). have complicated life cycles, involving female mosquitoes as de- Bird immunity play a key role in host–parasite interaction pro- finitive host and vertebrate animals including humans as cesses, and it is likely to have an effect on parasite evolution (Sorci, intermediate hosts. The vertebrate host serves as reservoir of the 2013). However, the existing knowledge about bird mortality during infection, with parasite development in the blood and other tissues. malaria infection is mainly due to observations of infections inci- A characteristic feature of Plasmodium spp. is the erythrocytic me- dentally reported in exotic birds that were moved from their natural rogony, which takes place in the peripheral blood. This feature allows distribution areas and kept in aviaries, zoos and private collec- initiation of controlled experimental infections without participa- tions (Cranfield et al., 1994; Ferrell et al., 2007; Fix et al., 1988). The tion of the vectors. This method of experimental research was challenges to identify emerging infection diseases in nature are developed in the end of the 19th century and has since been used mainly due the prevailing methods of sampling wild birds, i.e. dif- in studies of virulence, pathogenicity and other aspects of biology ferent methods for mist-netting and trapping. Such methods of avian malaria, and is still a frequently used method (Iezhova et al., preferably select actively moving healthy birds, but individuals suf- 2005; Larcombe et al., 2013; Palinauskas et al., 2007; Valkiu¯ nas, 2005; fering severe malaria, which leads to lowering of the locomotion Zehtindjiev et al., 2008). activity, remain underestimated (Valkiu¯ nas, 2005). For these reasons From the controlled experiments, huge knowledge about malaria the experimental studies constitute a powerful tool in determin- parasites’ pathogenicity, specificity and epidemiology have been re- ing the impact of malarial infections on bird individuals and vealed. Experiments with avian hosts as model organisms were populations (Larcombe et al., 2013; Palinauskas et al., 2008, 2011; particularly important in the beginning of human malaria re- Yorinks and Atkinson, 2000; Zehtindjiev et al., 2008). search; they contributed to the better understanding of Plasmodium Species identification of avian malaria and related haemospo- parasites’ biology and unravelling their complicated life cycles ridian parasites is currently based on microscopic examination of (Marzal, 2012). Currently, the experiments with bird malaria para- morphological features of the parasites, mainly during erythro- sites are performed for a variety of purposes aimed at better cytic stages; experimental data about specificity of the parasites to understanding of parasite diversity, host–parasites interactions, life- avian host and vectors are also considered (Garnham, 1966; history traits, interactions during co-infections, clarification of the Valkiu¯ nas, 2005). Application of molecular markers (DNA bar- genetic variation and the analysis of gene expression during primary codes) provides additional opportunities for parasite species infections (Garnham, 1966; Iezhova et al., 2005; Larcombe et al., identification. Initial molecular surveys revealed a much larger 2013; Palinauskas et al., 2008, 2009, 2011; Permin and Juhl, 2002; genetic diversity than existing morphospecies suggested by tradi- Valkiu¯ nas et al., 2013; Williams, 2005; Yorinks and Atkinson, 2000; tional (Beadell et al., 2004, 2006; Bensch et al., 2000, 2004; Zehtindjiev et al., 2008). Hellgren et al., 2004; Martinsen et al., 2008; Perkins and Schall, 2002). Malaria parasites are widespread in birds, and the prevalence This information raises questions regarding validity of both the of infection often exceeds 20% in many bird populations. However, current systematic of haemosporidians and the established host rather low malaria infection prevalence (<10%) often is reported in specificity knowledge. For example, several mitochondrial cyto- some avian host species too, for example the birds belonging to the chrome b (cyt b) lineages (SGS1, GRW04, GRW11 and LZFUS01), families Hirundinidae, Icteridae, Prunellidae, Regulidae, Sittidae, which likely represent intraspecific genetic variation, have been de- Troglodytidae and some others (Valkiu¯ nas, 2005). Due to anthro- termined within the P. relictum morphospecies (Ilgu¯ nas et al., 2013; pogenic influence and the global climate and environmental changes, Palinauskas et al., 2007; Valkiu¯ nas et al., 2007, 2008). These and many widespread host-generalist malaria pathogens expand in new areas other cyt b lineages might have a different ability to infect differ- and bird populations. An example is P. relictum, a virulent invasive ent bird species, but this remains largely unknown. In addition, avian malaria parasite (Garamszegi, 2011; Loiseau et al., 2012a), variability of biological features of the same cyt b lineage isolated which is the most prevalent and widely distributed parasite among from different donor species (different isolates) might also affect the haemosporidians (Garnham, 1966). specificity and pathogenicity of certain parasites, but there is no in- Specificity is a key character determining parasite virulence, formation on this issue. disease epidemiology, host-range and parasite geographic distri- In the present study, we review published information and bution (Bensch et al., 2009; Garnham, 1966). Introduction of P. provide new experimental data on the infection of birds (mainly relictum and its vector quinquefasciatus on Hawaii remains one of Passeriformes) with several of the widespread and prevalent avian of the most remarkable examples for impact caused by avian malaria malaria species and their cyt b lineages in Western Palearctic. All on the non-adapted bird community (Van Riper et al., 1986; Warner, experimental recipient birds were exposed by inoculation of in- 1968). Many cases of mortality of wild birds due to malaria have fected blood. Because some donor birds were migratory and were been accumulated in the literature (Valkiu¯ nas, 2005). Recent studies sampled after arrival from tropical regions, they likely were in- also show that avian malaria affects their hosts in many different fected with parasites of tropical origin. We believe that the obtained D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16 3 results are valuable for planning further experimental research on 2.3. DNA extraction, PCR screening and sequencing avian malaria specificity, resistance – tolerance trade-offs, patho- genicity and other issues. Importantly, we included the dataset about Total DNA was extracted using phenol-chloroform or ammonium- negative results of experimental inoculations, which often remain acetate protocols (Richardson et al., 2001; Sambrook et al., 1989). unpublished, but are important because they might indicate the pres- Quantification of the DNA was performed in the samples by using ence of innately refractory avian hosts. spectrophotometer or nanodrop (SpectromeGenesys 10 Bio; IMPLEN Nanophotometer P330). Samples with high DNA concentration were resolved to approximately of 25 ng/μl before proceeding with PCR. 2. Materials and methods For molecular screening, we used nested-PCR protocol as de- scribed by Hellgren et al. (2004), except primer pair used for partial 2.1. Study site and collection of the material amplification of cytochrome b (cyt b) of parasites of the genus (HaemFL and HaemR2L). The result of the PCR am- Experiments were performed between 2005 and 2013. Birds were plification was visualised by running of 1.5 μl of the final PCR product caught and maintained at the Biological Station “Rybachy” of the on 2% agarose gel. Samples with bands approximately of 500 base Zoological Institute of the Russian Academy of Sciences on the pairs (bp) were considered positive for parasites. Because good- Curonian Spit in the Baltic Sea (55°09′N, 20°51′E) and Kalimok Bi- quality sequences of already described lineages of parasites were ological Station of the Institute of Biodiversity and Ecosystem obtained, all fragments were sequenced only with forward primer Research at the Bulgarian Academy of Sciences (44°00′N, 26°26′E). (HaemF) from 5′ end as described by Bensch et al. (2000). Experiments with canaries were performed both at Kalimok Bio- The obtained sequences with length of 478 bp (excluding primers) logical Station and at the Institute of Ecology, Nature Research Centre, were edited and aligned using BioEdit software (Hall, 1999). Se- Vilnius, Lithuania. Experimental procedures of this study were ap- quences with double peaks on the ABI (Applied BioSystems file proved by the Lithuanian State Food and Veterinary Office (protocol format) chromatograms were considered as co-infections. The no. 0221, 4-01-2012), Ministry of Environment and Water of Bul- samples with ambiguous chromatograms or co-infections, as de- garia (permission nos. 28, 55, 76, 91, 213, 243 and 427 between 2005 termined by microscopic examination of blood films were and 2013) and Biological Station “Rybachy” of the Zoological Insti- reamplified and resequenced in order to determine stable double tute of the Russian Academy of Sciences (agreement #1 of 25-05- peaks positions. 2010). All efforts were made to minimise handling time and potential PCR screening of the multiple samples from the individual re- suffering of animals. None of the experimental birds suffered ap- cipient bird was used to assist in distinguishing co-infections. We parent injury due to handling or maintaining during experiments. applied both microscopic and molecular identification at repeat- Because of license agreements, the numbers of used birds were small ed time points in the course of the infections to minimise the risk in some experiments (Tables 1 and 2). of not detecting possible hidden co-infections in the experimental Wild birds were captured by mist nets or large “Rybachy” traps, birds. identified, ringed and bled after puncturing the brachial vein. Ap- The ‘Basic Local Alignment Search Tool’ (BLAST) implemented proximately 30 μl of blood was obtained in heparinised in MalAvi database (http://mbio-serv2.mbioekol.lu.se/Malavi/ microcapillaries and stored in SET-buffer (0.15 M NaCl, 0.05 M Tris, blast.html) and in the GenBank (http://blast.ncbi.nlm.nih.gov/ 0.001 M EDTA, pH 8.0) for further DNA extraction and PCR screen- Blast.cgi) was used to determine the lineage of detected parasite ing. Right away, several drops were used for preparation of two or sequences. three thin blood films. 2.4. Experimental design

2.2. Preparation of blood films and microscopic examination We isolated Plasmodium parasites from donor birds of 30 species (mainly passerines) belonging to 16 families (Tables 1 and 2). In all, Blood films were air dried and fixed in absolute methanol for 1 287 individual birds were inoculated experimentally by infected min. To speed-up examination of wild-caught birds, one blood film blood. Among them 222 were wild birds and 65 domesticated birds was stained according the fast method, i.e. 30% Giemsa solution was (52 canaries and 13 ducklings). The donors of infections were wild used and staining lasted 15 min; such slides were examined within birds tested for parasitaemia using microscopic examination of blood 1–2 h after preparation. That provided an opportunity to collect suf- films. All of them survived to the end of the study and then were ficient donor birds in experiments and quickly release non-infected released. We reported one case of natural infection in a domestic individuals with minimum handling time. The rest of the comple- canary (Serinus canaria). This bird was commercially purchased and mentary slides were stained according to traditional protocol (10% used as a donor of malarial infection too; it was hatched and grown Giemsa solution for 1 h) within 1–7 days after their preparation; in captivity in Bulgaria. that provide opportunity to reach better quality of staining. The recipient species of the wild birds belonged to 16 species Blood films were examined under light microscope, first at low and 6 families (Tables 1 and 2). Domestic canaries and two day- magnification (×400) for 10–15 min and then at least 100 fields were old domestic ducklings (Anas platyrhynchos) were purchased from examined at high magnification (×1000). The intensity of the markets and/or poultry. Seven of the recipients were 12–20 days parasitaemia was calculated as percentage by actual counting of in- old great tit (Parus major) nestlings, which were taken from nests fected erythrocytes per 1000 erythrocytes when infection was high and raised by hand in the laboratory. All recipient birds were tested or per 10,000 erythrocytes, when infection was light (<0.1%). Species both by microscopic examination and PCR, and were proved to be of parasites were identified according to Valkiu¯ nas (2005), Valkiu¯ nas parasite-free by these tools before experiments. Canaries and wild et al. (2007, 2008) and Palinauskas et al. (2007). birds were kept in quarantine with the aim to adapt to captivity con- Voucher blood slides from the donor and recipient birds are ditions for 10 days before the experiments; these birds were tested present in the collection of Institute of Ecology, Nature Research for possible presence of natural infections as described above. Most Centre, Vilnius, Lithuania (with curator Dr T. Iezhova) and Insti- of the recipients were confirmed to be malaria negative at least two tute of Biodiversity and Ecosystem Research, Sofia, Bulgaria (with times. The first test was on the day of capture of wild birds and pur- curator Dr D. Dimitrov) and are available on request from these col- chasing of canaries, and the second test was done after the lections. quarantine period right before inoculation. Additionally, negative 4 D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16 controls were used in most of the experiments (see Palinauskas et al., 1st + 2nd + 3rd + Noncoding. All positions containing gaps and missing 2008, 2009, 2011). data were eliminated. All experimental birds were kept in vector protected aviaries and cages, with suitable size for every species, perches and vegetation. 2.6. Statistical analyses Insectivorous birds were fed with meal worms, larva of flies and mixture of eggs ad libitum. Seedeaters were supplied with com- The analyses were carried out using ‘Statistica 7’ package. A two- mercial mixture of seeds for canaries and bits of apples ad libitum. tailed Fisher’s exact test was used for estimation of differences All cages were equipped with watering troughs. Birds were kept between numbers of susceptible avian hosts exposed to different under natural light–dark photoperiod (17:7 at Rybachy and 15:9 at cyt b lineages. A p value of 0.05 or less was considered significant. Kalimok). Depending on the weight of donor birds, we obtained between 3. Results 100 and 300 μl of blood by puncturing the brachial vein. Freshly pre- pared mixture containing the infected blood, 3.7% solution of sodium 3.1. Phylogenetic analyses citrate and 0.9% saline (4 parts of blood to 1 part of sodium citrate and 5 parts of saline) in amount of 150–300 μl (approximately 7 μl/g We used seven readily identified Plasmodium species from 15 according to weight of recipients) was immediately injected in the cyt b lineages in experiments (Fig. 1, Tables 1 and 2). The majority pectoral muscle of recipient birds, as described by Palinauskas et al. of the cyt b lineages, which were used in the Bayesian phylogeny (2007). implementation, cluster according to their subgeneric taxonomic In some experiments, infected blood was obtained from identity (Fig. 1). Exceptions are lineages of Giovannoliaia parasites, cryopreserved material. The tube with infected blood was thawed; which appeared paraphyletic. All lineages of P. relictum, P. the blood was mixed with 12% NaCl (1/3 of the original sample), circumflexum and P. elongatum form well supported clades with high equilibrated for 5 min at room temperature, mixed with one volume posterior probabilities (Fig. 1). of 1.6% NaCl and centrifuged at 1400 rpm for 5 min. After centrifu- Plasmodium (Giovannolaia) homocircumflexum (COLL4) and one gation, the supernatant was removed, 1.6% of NaCl (1/3 of the original lineage of unknown species identity (GRW10) were placed outside sample) was added and centrifuged as before. After removing the the clusters of lineages assigned to certain subgenera. The lineage supernatant, the same procedure was repeated three times with 0.9% COLL4 belongs to subgenus Giovannolaia, but it did not cluster with NaCl solution. The final mixture was diluted with 0.9% NaCl and sub- the other parasites of this subgenus clade in the phylogenetic in- inoculated into one or two canaries for isolate multiplication, as ference (Fig. 1). The genetic divergence between the lineages described by Palinauskas et al. (2008). TURDUS1 and SW5 is 3.8% and both lineages form well supported The recipient birds were monitored for parasitaemia by taking clade (Fig. 1), which is also in accordance with their assignment to blood every third day for at least 20 days post inoculation (dpi). Two the same morphological species of P. circumflexum. blood films were prepared and examined microscopically, as de- The lineage AEMO1 of Plasmodium sp. differed significantly from scribed above. Blood samples for PCR screening were collected every GRW06 and PADOM06 of P. elongatum (the genetic difference is 5.6% 6–10 dpi. and 5.8% respectively), but these lineages clustered well on the phy- Samples with high parasitaemia (>1%) were cryopreserved. About logenetic tree and likely belong to the subgenus Huffia (Fig 1). 200 μl of citrated blood was mixed with an equal amount of Although the cyt b lineages of Novyella species formed a mono- glycerolyte 57 [57.1 g Glycerol, 1.23 ml DL-Lactic Acid, sodium salt, phyletic group, the posterior probability support for this clade was 0.17 g Sodium phosphate (dehydrate), 0.086 g Sodium phosphate low. The lineage GRW02 of P. ashfordi was relatively distant from (anhydrous), 0.030 g Potassium Chloride, making the volume to other taxa of subgenus Novyella with genetic difference from the 100 ml with dH2O]. The mixture was equilibrated for 15 min at room type species of this subgenus, P. vaughani (lineage SYAT05), of 8.4%. temperature and mixed again with 200 μl of glycerolyte. The freez- ing vial with blood mixture was immediately placed in liquid 3.2. Susceptibility of recipients to Haemamoeba and Giovannolaia nitrogen (Garnham, 1966). parasite infections

2.5. Phylogenetic analyses Different infections of P. relictum predominated in our experi- ments (Tables 1 and 2). We exposed 14 species of recipient birds Bayesian phylogeny was constructed based on alignment of 27 to four cyt b lineages of this parasite, i.e., SGS1, GRW04, GRW11 and cyt b lineages (23 Plasmodium spp. and 4 spp.) using LZFUS01. Seven species of birds were exposed to the cyt b lin- MrBayes version 3.1.2 (Ronquist and Huelsenbeck, 2003). We used eages of closely related parasites of unidentified species, i.e., the General Time Reversible model including invariable sites and PADOM02, YWT4 and COLL1 (Fig. 1, Table 1). The lineage SGS1 of variation among sites (GTR + I + G) selected by the software P. relictum was the most prevalent; it was isolated from 14 differ- jModelTest 2 (Darriba et al., 2012) as best-fit model under Bayes- ent donors belonging to six families (Tables 1 and 2). This lineage ian Information Criterion. Gaps and missing data in the alignment developed parasitaemia in 78 of 102 (including co-infections) in- were discarded prior to analyses. Two simultaneous runs were con- fected individual birds (77.6% successful infection). Of 10 recipient ducted with a sample frequency of every 100th generation over 10 bird species exposed to the lineage SGS1, the following hosts were million generations. The 25% of the trees were discarded as burn- refractory: (1) common starlings (Sturnus vulgaris) infected with the in period. The remaining trees were used to construct a majority isolate from reed warbler (Acrocephalus scirpaceus), (2) canary in- rule consensus tree. The phylogenies were visualised using Tree fected with the isolate from great tit (Parus major), and (3) ducklings View 1.6.6. (software available from http://evolution.genetics infected with the isolate from red crossbill (Loxia curvirostra) .washington.edu/phylip/software.html). The abbreviations (codes) (Table 1). The highest parasitaemia (50%) of P. relictum developed of cyt b lineages are given according to MalAvi database; GenBank in one of six exposed red crossbills infected with reed warbler isolate, accession numbers of the lineages are shown in parenthesis in Fig. 1. with no cases of mortality reported in this experiment (Table 1). The sequence divergence between different lineages was calcu- Specificity of P. relictum, lineage GRW11 was relatively high in lated using Jukes–Cantor model of substitution, with all substitutions comparison with SGS1. However, the difference was not signifi- weighted equally (Uniform rates), implemented in the programme cant (Fisher’s exact test, p = 0.3103). From 28 exposed individual birds MEGA6 (Tamura et al., 2013). Codon positions included were belonging to 8 species, 15 individuals of 5 species developed D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16 5

Fig. 1. Bayesian phylogeny of 23 cytochrome b (cyt b) lineages of avian malaria parasites (Plasmodium spp.) and specificity of the lineages used in the experiments (given in bold) to the exposed bird genera shortened as follows: Serinus (Ser), Anas (Ana), Acrocephalus (Acr), Carduelis (Car), Fringilla (Fri), Loxia (Lox), Motacilla (Mot), Parus (Par), Passer (Pas), Sturnus (Stu), Sylvia (Syl) and Turdus (Tur). Four cyt b lineages of Haemoproteus (Haemoproteus) species were used as an outgroup. Codes of lineages are given, followed by the Latin name of morphologically identified parasites. GenBank accession numbers can be found in parenthesises. Posterior probability values are given next to the nods. Grey boxes show the closely related cyt b lineages of parasites belonging to different subgenera.

parasitaemia (including exposed birds to co-infections with GRW11). susceptible to the lineage GRW04 was significantly lower in It is worth mentioning that 11 birds were inoculated with single comparison with SGS1 (Fisher’s exact test, p = 0.0459), indicating isolate from garden warbler (Sylvia borin). This isolate developed higher specificity of the former lineage. Canary (Fringillidae), chaf- only in recipients belonging to the same bird species and in one of finch (Fringilla coelebs, Fringillidae), house sparrow (Passer domesticus, three exposed siskins (Carduelis spinus). Interestingly, the highest Passeridae) and blackcap (Sylvia atricapilla, Sylviidae) were refrac- parasitaemia developed in experiments, in which donor and recip- tory (Fig. 1, Table 1). Again, the highest parasitaemia developed in ient birds belonged to the same species (see Table 1, Sylvia borin, experiments, in which recipient and donor birds belonged to the parasitaemia 10% and Carduelis chloris, 6.7%), suggesting the higher same species; however, light parasitaemia developed in one reed virulence of the cyt b lineage GRW11 in bird species, which have warbler (0.001%) even in this case, indicating existence of been formerly exposed to the same lineage. lower susceptibility to this isolate of phylogenetically related hosts Nineteen individual birds were exposed to the lineage GRW04 (Table 1). of P. relictum, but parasitaemia developed only in five individuals, The lineage PADOM02, isolated from the house sparrow suc- which belong to the same species as the donor bird, the great reed cessfully developed in canaries, with a short prepatent period (5 warbler (Acrocephalus arundinaceus) or closely related member dpi) and relatively high maximum parasitaemia (1.4%). Another of the same genus, reed warbler (Table 1). Number of birds sister cyt b lineage YWT4 appeared to be species-specific and 6

Table 1 Susceptibility of birds exposed to single Plasmodium sp. infection by inoculation of infected blood. Names of domesticated experimental birds are given in bold font.

Subgenus, species and cytochrome b lineage of Plasmodium Species of donor bird (parasitaemia, %) Species of recipient bird Results

No. of First record Maximum Period of exposed/no. in blood, parasitaemia, % observation, of infected days days

Haemamoeba Plasmodium relictum SGS1 Acrocephalus arundinaceus (0.01) Serinus canaria 1/1 12 0.50 30 Acrocephalus scirpaceus (0.1)a Carduelis chloris 6/6 3 5 30 Carduelis spinus 6/6 6 30 30 Fringilla coelebs 6/6 12 0.1 30 Loxia curvirostra 6/6 3 50 30 Passer domesticus 6/3 9 0.04 30 Sturnus vulgaris 6/0 30 Coccothraustes coccothraustes (0.01)a Loxia curvirostra 2/2 8 20 65 Passer domesticus 2/2 14 0.1 50 Passer hispaniolensis (0.14) Passer hispaniolensis 5/4 3 2.35 36 1–16 (2015) 148 Parasitology al./Experimental et Dimitrov D. Luscinia megarhynchos (0.001) Serinus canaria 2/2 7 0.4 34 Carduelis chloris (0.03) Serinus canaria 2/2 9 1.20 25 Coccothraustes coccothraustes (0.02) Serinus canaria 1/1 8 0.20 32 Delichon urbicum (0.01–0.06)b Anas platyrhynchos 1/1 11 0.001 47 Serinus canaria 2/2 9 0.13 33 Emberiza calandra (0.01–0.02) Serinus canaria 3/3 5 0.53 28 Fringilla coelebs (0.04) Serinus canaria 2/2 10 1.3 31 Sylvia curruca (0.01–0.03) Serinus canaria 1/1 20 0.03 35 Parus major (0.01) Serinus canaria 2/0 31 Parus major 8/8 6 2.0 30 Loxia curvirostra (0.1) Anas platyrhynchos 2/0 21 Plasmodium relictum GRW11 Carduelis chloris(0.03) Serinus canaria 1/1 <9d 0.40 30 Carduelis chloris 1/1 <9 6.70 45 Passer domesticus (0.02) Anas platyrhynchos 2/0 21 Passer montanus (0.01) Serinus canaria 4/4 7 1.1 34 Sylvia borin (0.05) Carduelis spinus 3/1 <14 0.01 34 Loxia curvirostra 2/0 50 Passer domesticus 2/0 40 Sturnus vulgaris 2/0 30 Sylvia borin 2/2 <12 10.0 100 Plasmodium relictum GRW04 Acrocephalus arundinaceus (0.01–0.03) Serinus canaria 5/0 33 Acrocephalus arundinaceus 4/4 7 0.5 31 Acrocephalus scirpaceus 1/1 <11 0.001 28 Fringilla coelebs 2/0 28 Sylvia atricapilla 3/0 30 Passer domesticus 4/0 52 Plasmodium sp. YWT4 Motacilla flava (0.01–0.5) Serinus canaria 2/0 30 Motacilla flava 2/2 6 16.5 22 Acrocephalus arundinaceus 2/0 105 Passer domesticus 4/0 29 Plasmodium sp. PADOM02 Passer domesticus (0.01) Serinus canaria 2/2 5 1.4 34 Plasmoudium relictum (unknown lineage) Acrocephalus arundinaceus (0.03) Passer domesticus 6/0 50 Giovannolaia Plasmodium circumflexum TURDUS1 Troglodytes troglodytes (0.01) Anas platyrhynchos 2/1 12 0.01 21 Carduelis spinus 5/4 9 30 47 Loxia curvirostra 1/1 <18 2.0 93 (continued on next page) Table 1 (Continued)

Subgenus, species and cytochrome b lineage of Plasmodium Species of donor bird (parasitaemia, %) Species of recipient bird Results

No. of First record Maximum Period of exposed/no. in blood, parasitaemia, % observation, of infected days days

Novyella Plasmodium ashfordi GRW02 Cuculus canorus (0.01) Serinus canariae 2/0 116

Parus major 7/7 12 0.00 30 1–16 (2015) 148 Parasitology al./Experimental et Dimitrov D. Fringilla coelebs 2/0 45 Carduelis spinus 22/22 6 82 105 Carduelis chloris 2/0 45 Acrocephalus arundinaceus (0–3.2) Acrocephalus arundinaceus 10/10 25 4.1 82 Passer domesticus 6/0 87 Passer montanus 6/0 83 Turdus philomelos 4/0 53 Acrocephalus scirpaceus (0.01) Fringilla coelebs 2/0 35 Plasmodium homonucleophilum SW2 Locustella naevia (0.02) Carduelis spinus 2/0 45 Plasmodium vaughani SYAT05 Sturnus vulgaris (0.8–1) Serinus canaria 3/0 34 Sturnus vulgaris 1/1 12 0.60 45 Turdus merula (0.001–1.5) Anas platyrhynchos 2/0 31 Serinus canaria 5/0 51 Serinus canariae 3/0 Turdus merula 1/1 16 17.7 53 Turdus merulae 1/1 <19 0.01 27 Carduelis spinus 2/0 31 Huffia Plasmodium elongatum GRW06 Acrocephalus arundinaceus (0c)a Acrocephalus arundinaceus 2/1 <15 2 50 Passer domesticus (0) Passer domesticus 5/0 37 Plasmodium elongatum (unknown lineage) Acrocephalus arundinaceus (0.01) Serinus canaria 4/4 23 0.08 47 Plasmodium sp. Plasmodium sp. GRW10 Luscinia luscinia (0.02) Carduelis spinus 2/0 41 Plasmodium sp. AEMO01 Muscicapa striata (0.01) Sylvia atricapilla 3/0 35

a Isolates according to published data (see Palinauskas et al., 2008; Valkiu¯ nas et al., 2008). b Variation of parasitaemia in different individual birds when more than one donor was used in experiments. c Negative result by microscopic examination. d Certain day is unknown, but it is less than indicated. e Inoculation with cryopreserved blood. 7 8 D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16 developed only in avian host, from which it was isolated, i.e., yellow wagtail (Motacilla flava)(Table 1). Despite of the close phyloge- netic relationship with the host-generalist lineages of the P. relictum clade (Fig. 1), the lineages YWT4 showed restricted host-range and high specificity. In one donor bird with light parasitaemia of P. relictum, we failed to amplify parasite DNA probably due to poor quality of the DNA template. This isolate from great reed warbler did not develop in any of the six infected house sparrows (Table 1). Plasmodium circumflexum (lineage TURDUS1) was the most common species of the subgenus Giovannolaia tested experimentally for its specificity. Our results (Tables 1 and 2) indicate high suscepti- bility of all recipients (five bird species of four families) to the isolates from both Eurasian wren (Troglodytes troglodytes) and hawfinch. The highest parasitaemia (18%) developed in siskins infected with single infection isolated from the Eurasian wren (Table 1). Red crossbills were susceptible, but maximum parasitaemia was light (0.3–2%). Duck- lings, house sparrows and starlings were also susceptible, but parasitaemia was extremely light (<0.01%) and transit in these birds. Fig. 2. Average parasitaemia of Plasmodium relictum (lineage SGS1) in experimen- tally infected red crossbills (Loxia curvirostra) during single infection and co- 3.3. Susceptibility of recipients to Novyella and Huffia parasite infection with Plasmodium ashfordi (lineage GRW02). Vertical bars show standard infections deviations. Number of exposed birds was six in both experimental groups.

We exposed birds belonging to 12 species to three different malaria parasites of subgenus Novyella: Plasmodium ashfordi, Plas- parasite present in co-infections was P. relictum (lineage SGS1); it modium homonucleophilum and Plasmodium vaughani (the lineages was present in three of seven donor birds. Infection of the lin- GRW02, SW2 and SYAT05, respectively). The majority of these par- eages SGS1 and COLL1, isolated from a single hawfinch, developed asite species and their different isolates were highly specific, i.e., in three recipient species (siskins, red crossbills and house spar- parasitaemia developed only in recipients belonging to the same rows), but only the lineage COLL1 developed in one individual of species as the donor birds, with two exceptions (Table 1). First, P. four exposed starlings (Table 2). The intensity was light in the house ashfordi (lineage GRW02) isolated from common cuckoo (Cuculus sparrows and starlings and only part of the recipients developed canorus) developed light parasitaemia in great tits, which was de- parasitaemia. tectable only by PCR screening, but not by the microscopic Co-infection of P. relictum (lineage GRW11) and P. circumflexum examination. Importantly, this parasite was not detected in the neg- (lineage TURDUS1), isolated from a wild hawfinch developed in ative control nestlings used in this experiment. Second, siskins were siskins, red crossbills and house sparrows. However, only one of three highly susceptible to P. ashfordi and are good experimental hosts infected starlings was susceptible, and only P. circumflexum devel- for this parasite. Plasmodium homonucleophilum (lineage SW2) iso- oped (Table 2). The highest parasitaemia of P. relictum reached 70% lated from grasshopper warbler (Locustella naevia) was not infective in one of the infected siskins. for the exposed siskins. All the recipient bird species that differ from The co-infection of P. relictum (SGS1) and P. ashfordi (GRW02) the donors were refractory to P. vaughani (lineage SYAT05) iso- was isolated from the garden warbler. Both parasites developed lates from the and blackbird (Turdus merula)(Fig. 1, parasitaemia in siskins and red crossbills, but the starlings were re- Table 1). fractory to both infections. The maximal parasitaemia was high, Plasmodium elongatum was the single species of the subgenus i.e., > 50% of P. relictum and approximately 20% of P. ashfordi (Table 2). Huffia used in this study. This parasite was isolated from great reed Plasmodium relictum reached markedly higher parasitaemia when warbler, house sparrow and, as a co-infection with P. relictum from recipient birds were exposed to both parasites. For example, in red a hawfinch (Tables 1 and 2). Only the isolates from the great reed crossbills, infected with single infection of P. relictum (lineage SGS1) warbler and hawfinch developed parasitaemia in recipient birds. The the average intensity was markedly lower (Fig. 2). isolate from great reed warbler was detectable only by PCR in the A co-infection of P. homocircumflexum (lineage COLL4) and P. donors; parasitaemia of approximately 2% developed in the recip- relictum (LZFUS01) was isolated from the red-backed shrike (Lanius ient from the same species (Table 1). In the canaries exposed to collurio)(Table 2). Only lineage COLL4 developed in all exposed red isolates from the great reed warbler, light parasitaemia devel- crossbills and siskins. One of the ducklings was susceptible to the oped, but it was short-lasting (transit, detectible for 3–9 days). We same lineage, but developed light transit parasitaemia (Fig. 1, Table 2). also could not detect the parasite in canaries by PCR; it was de- The relatively closely related cyt b lineages (PADOM02 and tectable by microscopic observation only. GRW11 with genetic divergence of 2.9%) were detected in one of We detected malaria parasites of the lineages GRW10 and the purchased domestic canaries; this bird was used as a donor. AEMO01 at light parasitaemia in the thrush nightingale (Luscinia Single infection of P. relictum (lineage GRW11) developed in all re- luscinia) and in spotted flycatcher (Muscicapa striata), respectively cipient canaries (Table 2). Parasitaemia was relatively high in the (Table 1). The lineages GRW10 and AEMO01 did not develop in exposed canaries (about 12%) and light in the ducklings (0.04%). siskins and blackcaps, respectively (Fig. 1, Table 1). The parasitaemias We failed to detect the cyt b lineage of Plasmodium (Novyella) in the donors were low and that preclude identification of these par- sp. by PCR in golden oriole (Oriolus oriolus) in spite of the fact that asite species. the intensity of parasitaemia was at least 200 times higher than of P. relictum. Canaries were susceptible only for P. relictum (SGS1); 3.4. Susceptibility of recipients to co-infections maximum parasitaemia was > 90% in one bird. Natural co-infection of Haemoproteus tartakovskyi, Haemopro- Several experiments were performed using donors co-infected teus concavocentralis and Plasmodium sp. was identified in one with two different malaria parasites (Table 2). The most common hawfinch (data are not shown in Table 2). Molecular screening de- Table 2 Susceptibility of birds to co-infection of two Plasmodium species after inoculation of infected blood. Names of domesticated experimental birds are given in bold font. The first record of parasites in the blood (the fourth column) and maximum parasitaemia (the fifth column) in the recipients are given in the same order as the parasites mentioned in the first column, and data of each parasite are separated with the sign “&”.

Parasite species and its cytochrome b lineage reported in Species of recipient Results donor bird; intensity of parasitaemia indicated in host No. of exposed/no. First record Maximum Period of References parenthesis of infected in blood, parasitaemia, % observation, days days .Dmto ta.Eprmna aaiooy18(05 1–16 (2015) 148 Parasitology al./Experimental et Dimitrov D. Plasmodium (Haemamoeba) relictum SGS1 Carduelis spinus 2/2 <6a 12 37 – P. (H.) sp. COLL1* Loxia curvirostra 2/2 <4 8 100 – Donor: Coccothraustes coccothraustes (0.02, overall Passer domesticus 2/1 <11 0.03 49 – intensity of both lineages) Sturnus vulgaris 4/1* <12 0.01 14 –

Plasmodium (H.) relictum GRW11 (0.001) Carduelis spinus 2/2 12 & 24* 70 & 0.80* 28 – P. (Giovannolaia) circumflexum TURDUS1 (0.001)* Loxia curvirostra 2/2 17 & 20* 2.5 & 0.30* 65 Palinauskas et al., 2007 Donor: Coccothraustes coccothraustes Passer domesticus 2/2 25 & 20* 0.04 & 0.10* 60 Palinauskas et al., 2007 Sturnus vulgaris 3/1* 14 0.02 20 –

Plasmodium (H.) relictum SGS1 (0.001) Carduelis spinus 6/6 6 & 9* 58&19* 33 Palinauskas et al., 2011 P. (Novyella) ashfordi GRW02 (0.001)* Loxia curvirostra 6/6 6 & 12* 90&23* 33 Palinauskas et al., 2011 Donor: Sylvia borin Sturnus vulgaris 6/0 33 Palinauskas et al., 2011

P. (H.) relictum LZFUS01 (0.5) Loxia curvirostra 4/4* 84 24Palinauskas et al., 2014 P. (G.) homocircumflexum COLL4 (0.01)* Carduelis spinus 2/2* <84 65Palinauskas et al., 2014 Donor: Lanius collurio Anas platyrhynchos 2/1* <8 0.001 21

Plasmodium (H.) sp. PADOM02 Serinus canaria 2/2* 6 12.1 56 – P. (H.) relictum GRW11* Anas platyrhynchos 2/2* 12 0.04 21 – Donor: Serinus canaria (0.02)

Plasmodium (N.) sp. – unknown lineage (0.2) Serinus canaria 2/2* 79424– P. (H.) relictum SGS1 (0.001)* Donor: Oriolus oriolus

Plasmodium (H.) relictum – unknown lineage Serinus canaria 1/1 9 & 15* 0.07 & 6.50* 42 – P.(Huffia) elongatum – unknown lineage* Donor: Coccothraustes coccothraustes (0.02)

a Certain day is unknown, but it is less than indicated. * An asterisk indicates parameters of an infection, which is marked with the same symbol in the first column. 9 10 D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16 tected only the cyt b lineage HAWF2 of H. concavocentralis in this 4.1. Specificity of Haemamoeba and Giovannolaia parasites donor bird. The single canary exposed to this co-infection devel- oped high parasitaemia of P. elongatum (6.5%) and light parasitaemia This study confirmed former results from molecular and mor- of P. relictum (0.07%) detected and identified by microscopy. The PCR phological surveys about broad host-range of the cyt b lineage SGS1 screening detected co-infection of relatively divergent Plasmo- of P. relictum (Beadell et al., 2006; Ewen et al., 2012; Garnham, 1966; dium spp. cyt b lineages in the recipient bird 9 dpi (first molecular Hellgren et al., 2007; Kim and Tsuda, 2010; Loiseau et al., 2012a; sampling), which is in accordance with the morphological identi- Marzal et al., 2011; Valkiu¯ nas, 2005; Waldenström et al., 2002). Of fication of both parasites in the smears, but does not allow the wild birds tested, only the starling was unsusceptible to this in- differentiation of the cyt b lineages in the recipient. fection (Table 1). Interestingly, the susceptibility of birds to the lineages SGS1, GRW11 and GRW04 of P. relictum was different (Tables 1 and 2). The lineages GRW04 and GRW11 were more spe- 3.5. Pathogenicity cific than SGS1 (Table 1), despite all having wide geographical distribution and host-range (MalAvi database: http://mbio- The virulence of different lineages and their isolates of Plasmo- serv2.mbioekol.lu.se/Malavi/index.html). Genetic divergence between dium spp. varied markedly. We reported several cases of severe SGS1 and GRW11 in cyt b gene is negligible (0.2% or 1 nucleotide pathology and even mortality in exposed birds, probably due to high out of 479 base pair in the MalAvi barcoding region); these lin- parasitaemia. It is worth noting that we did not examine tissue eages have similar and completely synchronous sporogony in the samples from the organs, and possible pathology caused by exo- pipiens mosquitoes (Kazlauskiene˙ et al., 2013). Al- erythrocytic stages (phanerozoites) remains unknown. The majority though the lineage GRW11 has been naturally found in the house of mortality cases (75%, 6 out of 8) were reported in birds exposed sparrow (Marzal et al., 2011), we were not able to infect this species to co-infections. One of the siskins exposed to co-infection of the with our isolate of GRW11 from a garden warbler (Table 1). It is lineages SGS1 and COLL1 died probably due to high parasitaemia worth noting this isolate of GRW11 was species specific in general. (12%) and anaemia caused by both these parasites. These lineages Ducklings were not susceptible to the single infection of a GRW11 were difficult to distinguish by morphology of their blood stages isolate from the house sparrow, but it developed when these birds and we could not identify parasitaemia of each lineage. were exposed to co-infection of GRW11 and PADOM02 isolated from We observed that the virulence of the host-generalist P. relictum domestic canary. It remains unclear if the ducklings become sus- (lineages SGS1 and GRW11) increase during co-infections. Two of ceptible because of different isolates of lineage GRW11, presence six red crossbills co-infected with P. relictum (SGS1) and P. ashfordi of co-infection with PADOM02 or due to the individual differences (GRW02) died between 30 and 33 dpi due to high parasitaemia of in the immune system in recipients. A similar pattern of develop- P. relictum, which reached over 50% (Fig. 2, Table 2). During co- ment was observed also in the birds exposed to co-infection of P. infection of P. relictum (GRW11) and P. circumflexum (TURDUS1), relictum and P. circumflexum (the lineages GRW11 and TURDUS1) parasitaemia of the former parasite reached 70% (Table 2) and one isolated from hawfinch. The siskin and red crossbill were highly sus- of the infected siskins died 28 dpi. A similar case was observed with ceptible to the lineage GRW11 in passages with co-infection of one of two canaries exposed to the co-infection of Plasmodium TURDUS1 (Table 2), in contrast to when they were exposed to single (Novyella) sp. (unknown lineage) and P. relictum (SGS1). This canary infections of GRW11 (Table 1). Red crossbills did not develop single died 24 dpi, likely due to P. relictum parasitaemia, which exceeded infection at all, and only one of three siskins was susceptible, but 90%. parasitaemia was light and transient. Contrary, during co-infection Light infection of an unidentified Plasmodium sp. was isolated all red crossbills were susceptible and parasitaemia of P. relictum from a hawfinch. Initially, the infected canary developed P. relictum (lineage GRW11) in exposed siskins reached 70% (Table 2). One of (parasitaemia, 0.07%), but after 15 dpi a second infection of P. three exposed starlings developed light parasitaemia (0.02%) of P. elongatum appeared and reached higher intensity (6.5%). The re- circumflexum (lineage TURDUS1) 14 dpi. However, starlings were cipient died 42 dpi, and numerous destroyed blood cells and severe refractory to the single and co-infections of lineage GRW11 (Tables 1 anaemia were recorded in smears before death. and 2). It seems probably that either the exposure to different iso- In single infection experiments, mortality of exposed birds was lates or/and co-infection influence avian host-susceptibility, but more rare. Two of 22 infected siskins (9%) died after exposure to P. ashfordi detailed studies with larger samples size are needed to confirm these (the lineage GRW02 isolated from common cuckoo). suggestions. In all, mortality of birds during co-infection was 10-fold higher The lineage GRW04 of P. relictum is responsible for severe pa- than during single infections in this study (Fisher’s exact test, thology in Drepanidinae birds on Hawaiian Islands (Beadell et al., p = 0.0008). 2006). Atkinson et al. (1995, 2000) multiplied this parasite in do- mestic canaries for the purpose of laboratory investigation of 4. Discussion pathogenicity in endemic Hawaiian birds. Surprisingly, domestic ca- naries were not susceptible to the cyt b lineage GRW04 in our Since the discovery of avian malaria and related haemosporid- experiments that we isolated from an African migrant, the great reed ian parasites, experimental infection is an essential technique in warbler. A possible explanation might be that the isolates from studies of the host–parasite interactions (Garnham, 1966; Marzal, Hawaii and tropical Africa differ in their specificity, but are 2012). In the present study we combined the microscopic and mo- undistinguishable by their morphology and mitochondrial cyt b gene. lecular examination with experimental infection of birds with the Recently, the next-generation 454 sequencing revealed marked aim to test for the specificity of the most prevalent lineages of avian cryptic mitochondrial DNA diversity of P. relictum on Hawaii (Jarvi Plasmodium spp. For the malaria infection experiments we used et al., 2013). It is possible that we also deal with different parasite widely distributed European bird species (Tables 1 and 2) that are clones, which have their own specificity and pathogenicity char- relatively easy to maintain in captivity, domestic canaries and duck- acters, but are undistinguishable both by morphology of the blood lings. The latter have been widely used as experimental hosts in early stages and mitochondrial cyt b gene sequences. However, there is and recent studies addressing issues relating to life-history, mor- no information about this issue in relation to our material. Inter- phology, taxonomy, specificity and pathogenicity of Plasmodium spp. estingly, Marzal et al. (2011) showed that the lineage GRW04 was (Atkinson et al., 1995, 2000; Cornet et al., 2014; Garnham, 1966; widespread in the house sparrows in the temperate regions of North Kazlauskiene˙ et al., 2013; Valkiu¯ nas, 2005). America, but it was absent from this species in Europe. Results of D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16 11 our experiments showed that some European birds (house spar- number of exposed individual birds did not allow clarifying rows, chaffinches, blackcaps and domestic canaries) are not these differences; additional studies are needed to answer this susceptible to the lineage GRW04 isolated from the great reed question. warbler. These findings partly explain the reported different dis- Plasmodium (Giovannolaia) homocircumflexum used in our ex- tribution, host-range and specificity of the lineage GRW04 in the periments has been recently described as cryptic species of P. wildlife of the Old World and the New World (Marzal et al., 2011). circumflexum. Detailed morphological description of P. It has been reported that several subsequent passages of P. homocircumflexum (lineage COLL4) based on phylogenetic analy- relictum (GRW04) in the 7 day old ducklings resulted in succes- ses, examination of blood and tissue stages and additional biological sively increased parasitaemia in the recipients (LaPointe et al., 2005). characters indicates first cryptic speciation recorded in an avian We did not use this method and perhaps that explains why the ma- malaria parasite (Palinauskas et al., 2014). jority of the exposed ducklings were unsusceptible to the malarial infections in this study (Table 1). It is worth noting that the in- 4.2. Specificity of Novyella and Huffia parasites creased virulence of Plasmodium chabaudi (rodent malaria parasite) was also observed in the experiments with several passages In the present study we found that malaria parasites of the sub- (Mackinnon and Read, 2004). The same study suggests that mos- genus Novyella were highly specific as only one isolate developed quito transmission reduces the virulence of the parasite. To our in recipients of another bird species than the donor bird. Siskins were knowledge, such comparative studies are lacking in avian malaria highly susceptible to P. ashfordi (lineage GRW02) isolated from a research. If the same pattern is also valid for the haemosporidian common cuckoo; all 22 infected birds developed high parasitaemia. parasites in birds, vector transmission would be essential to measure Additionally, two of the exposed siskins died due to high true virulence in the laboratory experiments. parasitaemia (over 80%). Plasmodium ashfordi has a tropical origin The lineage YWT4 is a sister lineage to the SGS1 and GRW11 and is prevalent in adult great reed warblers that have been at least (Fig. 1), but its specificity was different. This parasite appears to be once in Africa where they gain this infection (Valkiu¯ nas et al., 2007; species-specific; it has been recorded only in the yellow wagtail Zehtindjiev et al., 2008). Transmission of P. ashfordi has not been (Motacilla flava) (MalAvi database: http://mbio-serv2.mbioekol.lu.se/ reported in the temperate regions. Hence this parasite is novel for Malavi/index.html). Our experiments confirmed the high specificity European siskins and it is associated with high virulence. This is an of the lineage YWT4 (Table 1). Domestic canaries, great reed war- example of what might happen if malaria parasite of tropical origin blers and house sparrow were refractory to the lineage YWT4 of changes transmission area and infects naïve bird populations in Plasmodium sp. We were unable to identify the species of this lineage Europe. because of insufficient morphological material. Based on available Seven nestlings of the great tit were susceptible to the same morphological data, we suspect that our isolate contained a co- isolate of P. ashfordi (GRW02), but developed transient and ex- infection of at least two species of Plasmodium, but PCR amplification tremely light parasitaemias (0.0001%) without presence of showed single infection of the cyt b lineage YWT4. Because general gametocytes. Mortality was not reported in the exposed nestlings. primers often underestimated co-infections of haemosporidian para- It is worth mentioning that similar cases of positive PCR diagnos- sites (Dimitrov et al., 2013; Martínez et al., 2009), additional studies tics without gametocytes present in the blood were observed during should be performed in order to identify species identity of the abortive development of some of Haemoproteus parasites (family lineage YWT4. ) (Cannell et al., 2013; Olias et al., 2011), but these In Europe lineage PADOM02 is restricted to southern countries infections were lethal for the hosts. If the parasites did not com- and mainly in house sparrows and it is not prevalent based on PCR plete their development to gametocyte stage in the peripheral blood testing (Marzal et al., 2011). Domestic canaries successfully devel- their natural transmission through the vectors is impossible oped the infection in our experiment (Table 1). It might be that (Valkiu¯ nas, 2005). Recent reports of Plasmodium spp. in the Gala- distribution of the lineage PADOM02 is limited due to specificity pagos penguins (Spheniscus mendiculus) without detection of of mosquito vectors; that need additional investigation. gametocytes are additional examples of abortive development of Plasmodium circumflexum is a type species of the subgenus avian malaria parasites (Levin et al., 2013). Molecular diagnostic Giovannolaia and has worldwide distribution. It infects numerous studies based solely on PCR screening might record such infec- birds of the orders Anseriformes, Columbiformes, Coraciiformes, tions as positives and might include them in further evolutionary Charadriiformes, Falconiformes, Strigiformes, Galliformes, and some and co-evolutionary analyses, but they are dead ends of transmis- others, but particularly of the Passeriformes (Valkiu¯ nas, 2005). Mo- sion. To avoid misleading interferences we recommend applying PCR- lecular screening showed similar results for the cyt b lineages based methods in parallel with microscopic examination, which can TURDUS1 and SW5 (MalAvi database: http://mbio-serv2 demonstrate if gametocytes are present in the circulation. .mbioekol.lu.se/Malavi/index.html), which both are considered to The lineage SW2 of P. homonucleophilum has been recently de- be P. circumflexum (Palinauskas et al., 2007; Valkiu¯ nas et al., 2014). scribed as a new species from the common grasshopper warbler Our results (Tables 1 and 2) confirm low specificity of P. circumflexum (Ilgu¯ nas et al., 2013). This lineage has been formerly attributed to and are in accordance with these molecular studies; that can explain P. polare (see Beadell et al., 2006), but without proof of the mor- the wide geographical and host distribution of this parasite. In our phological identification. Several molecular studies have shown that experiments this parasite was isolated from two donor species, i.e., the lineage SW2 of P. homonucleophilum is widespread in different the Eurasian wren and the hawfinch. Interestingly, both isolates dif- populations of great tit (Beadell et al., 2006; Glaizot et al., 2012; fered in their development and intensity in red crossbills and siskins. Van Rooyen et al., 2013) and blue tit (Cyanistes caeruleus)(Szöllo˝si It is worth noting that the P. circumflexum isolate from hawfinch was et al., 2011) in Europe. Since both bird species are resident in the co-infected with P. relictum (lineage GRW11). When in co-infection, temperate regions of Western Palearctic (Cramp, 2004), P. much lower parasitaemia of P. circumflexum (over 10-fold less) de- homonucleophilum is certainly transmitted in Europe. In this study, veloped in recipient birds than during single infection (compare siskins were refractory to the isolate from the common grasshopper- Tables 1 and 2). This result was unexpected and it is opposite to what warbler (Table 1). we observed in the cases of P. relictum (lineages SGS1 and GRW11) Interestingly, we recorded P. vaughani (lineage SYAT05) in a wild- (Fig. 2 and see above). Again, it remains unclear if different sus- caught common starling. This bird was refractory to host-generalist ceptibility and parasitaemias in these cases are due to different parasites like P. relictum (lineages SGS1 and GRW11) and P. ashfordi isolates of the parasites or to co-infection. However, the small (GRW02) after experimental exposure (Palinauskas et al., 2008, 2011, 12 D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16 this study). The lineage SYAT05 is common in many resident and with traditional microscopic studies, which revealed broad, but scat- short-distance migrant passerine species in Europe (MalAvi data- tered distribution of P. elongatum (Valkiu¯ nas, 2005). base: http://mbio-serv2.mbioekol.lu.se/Malavi/index.html; Cramp, We detected the lineage AEMO01 for the first time in spotted 2004). Most probably, P. vaughani (lineage SYAT05) is actively trans- flycatcher on the Curonian Spit in the Baltic Sea (MalAvi database: mitted in the temperate regions of Europe. In our experiments, http://mbio-serv2.mbioekol.lu.se/Malavi/index.html), a long dis- domestic canaries and ducklings were refractory to the lineage tance migrant wintering in central Africa (Cramp, 2004). It is unclear SYAT05. Hence, these birds are not suitable hosts for further ex- where this bird gained the infection. Further studies on the trans- perimental research with this parasite lineage. mission and distribution of lineage AEMO01 are needed to answer This study shows that malaria parasites of the subgenus Novyella this question. The lineage GRW10 was recorded for the first time are usually strictly specific to certain bird species and do not develop in the thrush nightingale during this study. The range of transmis- in the recipient birds belonging to other species than the donor birds sion of this parasite is also unknown but it has been recorded at (Fig. 1, Table 1). These observations are in accordance with former low prevalence in two African migrants, i.e., the great reed warbler field studies reporting that many species of Novyella are host spe- and the collared flycatcher (Ficedula albicollis)inSweden(Bensch cialists in African birds (Loiseau et al., 2012b). et al., 2007; Kulma et al., 2013; MalAvi database: http://mbio- Plasmodium elongatum is a type species of the subgenus Huffia serv2.mbioekol.lu.se/Malavi/index.html). and this infection is widespread all over the world and causes lethal disease in some naturally and experimentally infected birds 4.3. Pathogenicity (Garnham, 1966; Valkiu¯ nas, 2005). The cyt b lineage GRW06 cer- tainly belongs to P. elongatum (Valkiu¯ nas et al., 2008). Using PCR The majority of mortality cases in birds were observed during amplification, this lineage has been recorded al low frequencies in the exposure of recipients to co-infections, which have been con- house sparrows in southeast Europe (Kalimok Biological Station) and sidered more virulent in avian hosts also in other studies (Palinauskas New Zealand (Marzal et al., 2011). Our attempt to observe et al., 2011; Zehtindjiev et al., 2008). Interestingly, mortality was parasitaemia in house sparrows was unsuccessful, but birds were rarely associated with development of high parasitaemia of both PCR positive (Table 1). This is probably because the intensity was parasites in parallel (Table 2). Probably the immune response too low in the donor bird. Of two great reed warblers used as donors, managed to control one of the infections, but not the second one, one of the infections developed at sufficient intensity allowing mor- which developed at high parasitaemia and killed the host, as it is phological identification and molecular characterisation of P. the case with one infected canary exposed to two malarial para- elongatum in our former study (Valkiu¯ nas et al., 2008). sites isolated from the golden oriole. The infection of Plasmodium Interestingly, two donor great reed warblers were infected with (Novyella) sp. was present in the donor at much higher intensity than P. elongatum according to microscopy examination, but the PCR P. (Haemamoeba) relictum (lineage SGS1), but only the latter para- method showed negative results in detection these infections. In site developed in the exposed canary (Table 2). However, the intensity spite of clear presence of this parasite in the donors and the re- of parasitaemia significantly differed among the individual recipi- cipients according microscopic examination, the PCR was negative ents, and it reached approximately 3% and then dropped to chronic in all tested complementary samples. levels in the second infected canary. This result is in accordance with In former studies, canaries were considered as excellent exper- conclusions of previous studies regarding the dynamics of imental hosts to unknown lineages of P. elongatum (Garnham, 1966; parasitaemia in experimentally infected birds (Larcombe et al., 2013; Valkiu¯ nas, 2005), but our study showed light and short-lasting Palinauskas et al., 2008, 2011) and some of the alleles of immune parasitaemia in the canaries in the majority of experiments after system associated with infectious diseases (Westerdahl et al., 2012). exposure to the lineage GRW06 (Table 1). It might be that differ- One canary died after infection with P. elongatum isolate ob- ent isolates of P. elongatum markedly vary in their development and tained from the hawfinch. However, a co-infection of P. relictum and virulence. The lineage GRW06 of P. elongatum is locally transmit- P. elongatum was observed in the recipient in this case. It is prob- ted at Kalimok Biological Station, but it is rarely found in the local able that the co-infection changed the development pattern of P. wild birds (Dimitrov et al., 2010). The development of this para- elongatum in this bird. Only one individual bird was infected with site in naturally infected birds has been often reported in the internal this isolate (Table 2) and we cannot certainly conclude about severe organs and only a few erythrocytic meronts and gametocytes pathology caused by this combination of parasites; that warrants usually can be observed in the peripheral blood (Garnham, future research. The present study mainly deals with the host sus- 1966). That is an obstacle for microscopic detection of this infec- ceptibility of birds exposed to various avian malaria parasite species, tion in wildlife. cyt b lineages and their isolates. Broad generalisation about pat- We isolated two cyt b lineages of Plasmodium (GRW10 and terns of virulence of particular co-infections of malaria species, cyt AEMO01), which were not identified to species level due to light b lineages or isolates would be speculative due to the small sample parasitaemia and these were detected in the thrush nightingale and size in many experiments of this study. However, (1) the several cases the spotted flycatcher, respectively. Refractoriness of the recipient of severe pathology that we observed after exposure of the recipi- bird species to these lineages (Table 1) did not provide the oppor- ents to co-infections and (2) available published information tunity to determine their species identity. The lineage AEMO01 (Palinauskas et al., 2011; Zehtindjiev et al., 2008) testify the higher clustered with the lineages of P. elongatum and might belong to virulence of co-infections and should be considered as an impor- species of the subgenus Huffia (Fig. 1). The lineage GRW10 is with tant goal in future experiments aiming better understanding uncertain position in the phylogenetic tree and it is therefore dif- pathogenicity of avian malaria parasites. ficult to speculate about its identity based on our data set. The transmission of the lineage AEMO01 probably takes place in trop- 4.4. Importance of the experiments with avian malaria ical areas of Africa (Hellgren et al., 2007) and the Mediterranean region (Iberian Peninsula) (MalAvi database: http://mbio- In majority of the experiments we used wild donors with dif- serv2.mbioekol.lu.se/Malavi/index.html). It was found in the black ferent chronic levels of parasitaemia (Table 1). However, we are vulture (Aegypus monachus), which is a resident species in Spain confident that this fact did not affect the experiments for avian host (Cramp, 2004) and also in the African paradise-flycatcher (Terpsiphone specificity and maximum parasitaemia developed in the recipi- viridis), a bird species that is sedentary in tropical Africa (BirdLife ents. According to several previous studies, the ability to infect birds International, 2014). Recent PCR-based screening is in accordance and the dynamics of parasitaemia does not depend on the dose of D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16 13 meronts in the inoculum (Palinauskas et al., 2008; Permin and Juhl, tribution and host-range of many Plasmodium spp. lineages in 2002). However, the dose negatively correlates with the duration wildlife. However, the emergence of some malarial infections in new of the prepatent period (Permin and Juhl, 2002) and the latter pa- areas increases (Garamszegi, 2011). Malaria infected birds should rameter is also markedly depending on host individuals (Palinauskas be considered as possible source of new infections due to the rapidly et al., 2008) and differs between the exposure to single-infection changing environment, climate change and accidental or deliber- and co-infection (Palinauskas et al., 2011). ated anthropogenic introduction of birds and/or mosquito vectors It is worth mentioning, that we used canaries, which were bred in regions where they formerly were absent (Atkinson and LaPointe, in captivity and commercially purchased. Most of them were 2009; Loiseau et al., 2012a). Each year, billions of migrating birds uninfected at the time of purchase, except one that carried a co- move between the Palearctic and the tropical regions of Africa and infection with the lineages GRW11 and PADOM02. The lineage Asia. That creates favourable conditions for spreading avian malaria GRW11 developed relatively high parasitaemia in two exposed ca- parasites. We detected several cyt b lineages of Plasmodium spp., naries (Table 1). This finding indicates that it is important to test which infect birds during wintering in the tropics and subtropics. purchased birds for possible natural infection before experiments, These are GRW04, GRW02, AEMO01, LZFUS01, COLL4 (Hellgren et al., particularly in countries where active malaria transmission occurs. 2007; Waldenström et al., 2002; MalAvi database: http://mbio- Both lineages (GRW11 and PADOM02) are common parasites in- serv2.mbioekol.lu.se/Malavi/index.html). Natural infections of these fecting house sparrows worldwide (MalAvi database, see Bensch parasites were found only in adult birds that had done seasonal long- et al., 2009, Marzal et al., 2011). The house sparrow is a synan- distance migrations at least once, thus they likely got the infections thropic species and its close association with human settlements in the tropics. Such individuals are reservoirs of tropical malarial might explain host-switching of malaria infections from house spar- parasites in Europe and it is possible that these parasites might be rows to domestic canaries. transmitted to sedentary birds if an appropriate vector appears, par- The majority of our experiments involved wild birds as recipi- ticularly during warming of climate. Our data (Table 1) can be used ents of malaria parasites. Although all birds were confirmed to be for predicting the outcome of some tropical infections in seden- negative both by PCR-based and microscopic examination before tary bird populations in temperate regions. Previous studies showed the experiments, it is difficult to completely rule out that some that some parasites of tropical origin (the lineages GRW02) are highly natural infections were undetected due to possible latent infec- virulent for European birds (Palinauskas et al., 2011). More re- tion stage, which cannot be detected by PCR or microscopy. search is needed for better understanding how to prevent possible Immunological test are sensitive diagnostics of latent infections (Jarvi emergence of such infections conditioned by climate changes and et al., 2002), but are difficult to apply in case of markedly diverse anthropogenic factors in Europe. avian malaria parasites due to requirement of specific antigens, which Well-controlled experiments also provide new information about are absent for all tested parasite lineages (Tables 1 and 2). In our patterns of dynamics of parasitaemia and influence of malaria para- study we detected several cases of infection only by PCR, but mi- sites on birds (Zehtindjiev et al., 2008) as these features is difficult croscopy was more sensitive than PCR in parasite diagnostics in many to estimate in wild-caught and single-examined bird individuals. cases as well (Tables 1 and 2). We suppose that if the recipients were For example, parasitaemias are chronic in the majority of the samples confirmed negative both by PCR and microscopy multiple times of wild-caught birds by netting and/or trapping (see the Introduc- before inoculation it is unlikely that all used recipients will have tion). An experimental study on the dynamic of parasitaemia in been exposed formerly and developed antibodies in the wild to the naturally and experimentally infected birds showed that chronic exactly the same parasite lineages that have been used for expo- parasitaemia of P. ashfordi is positively correlated with primary peak sure during our experiments (Tables 1 and 2). To our knowledge, of parasitaemia in great reed warblers (Asghar et al., 2012). This few studies investigated the immune response to avian malaria in finding provides a new tool in studies of avian malaria in wildlife, immunologically experienced birds (Atkinson et al., 2001; particularly for parasitaemia studies, which are related to patho- Cellier-Holzem et al., 2010). Indeed, re-exposure of avian hosts to genicity. Because most of the surveys on avian malaria have been the same or similar malaria parasite might affect prepatent period based on single sampling of wild-caught birds with light and developed parasitaemia (Atkinson et al., 2001). However, avail- parasitaemia, actual estimate of true impact of the parasites during able data on the reciprocal immunity showed that chronic infection high intensity on bird populations remains hardly detectable. If the of P. relictum do not influence experimental infections with P. acute level of the parasitaemia might be predicted based on the level elongatum, P. hexamerium,orP. rouxi (Draper, 1953). In other words, of chronic parasitaemia, the records not only of the prevalence, but cross-immunity to different parasite species is typically low. also the intensity of infections are valuable in natural population Because natural vectors of the majority of parasite lineages used studies. In this way, the possible maximum parasitaemia and pos- in this study remain unknown and some lineages are likely of trop- sible virulence effect on the avian hosts can be appropriately ical origin, the experimental exposure provides valuable information evaluated even using single sampling of birds in wildlife. This method about parasite specificity and also can be used for morphological, should be of particular interest for conservation biologists, espe- taxonomical and virulence studies (Iezhova et al., 2005; Palinauskas cially in the parasitological studies dealing with endangered bird et al., 2007, 2011; Valkiu¯ nas, 2005; Valkiu¯ nas et al., 2007, 2008; species or under the process of their reintroduction. However, Zehtindjiev et al., 2008). However, such mode of infection is not because Asghar et al.’s (2012) study was based on data from single natural as it eliminates development of primary exoerythrocytic host–parasite system, the authors encourage similar experimental meronts, which are induced only by sporozoites. That is why the observations in additional host–parasite systems for further devel- data of susceptibility should be carefully used in theoretical opment of a quantitative predictive model for estimation of generalisations. It is worth mentioning that the available experi- parasitaemia level in wild birds using data about chronic mental data indicate that parasitaemia and virulence of the Hawaiian parasitaemia. strain of P. relictum was similar in blood and sporozoite induced in- From the taxonomical point of view, the experimental infec- fections (Atkinson et al., 1995, 2000; R. Žiegyte personal observation); tion of birds is valuable because it provides an opportunity for thus, our experimental data on specificity (Tables 1 and 2) likely detailed studies of morphological characters of various blood stages reflect the true situation in wildlife. of Plasmodium parasites (Palinauskas et al., 2007; Valkiu¯ nas et al., Because many European birds are susceptible to malaria para- 2007, 2008). That is important for the malaria parasite species iden- sites of African origin (Tables 1 and 2), the absence of the susceptible tification and the development of molecular characterisation (DNA vectors might be one of the explanations for restricted spatial dis- barcoding) because when the parasitaemia is synchronised, which 14 D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16 is the case for some Plasmodium spp., not all stages necessary for acknowledged for their assistance in the laboratory. The experi- species identification are present in smears (Valkiu¯ nas, 2005). Plas- ments described herein comply with the current laws of Russia, modium species identification is possible mainly after investigation Lithuania and Bulgaria. DD was supported by European Union Struc- of parasitaemia in its dynamics and observation of all blood stages tural Funds project “Postdoctoral Fellowship Implementation in (trophozoites, meronts and gametocytes). Multiple sampling of the Lithuania” (VP-3.1-ŠMM-01-V-02-004). same individuals of wild birds requires recaptures, which is diffi- cult to do in wildlife. Experimental infections provide opportunities References for morphological parasite identification and development of their molecular markers (Palinauskas et al., 2007; Valkiu¯ nas et al., 2007, 2008). Asghar, M., Westerdahl, H., Zehtindjiev, P., Ilieva, M., Hasselquist, D., Bensch, S., 2012. Primary peak and chronic malaria infection levels are correlated in experimentally The PCR-based molecular techniques opened new opportuni- infected great reed warblers. Parasitol. 139, 1246–1252. doi:10.1017/ ties to identify haemosporidian species during light intensity of S0031182012000510. parasitaemia and revealed great genetic diversity of avian malaria Atkinson, C.T., LaPointe, D.A., 2009. Introduced avian diseases, climate change, and the future of Hawaiian honeycreepers. J. Avian Med. Surg. 23, 53–63. doi:10.1647/ parasites (Beadell et al., 2006; Bensch et al., 2000, 2004, 2009; 2008-059.1. Hellgren et al., 2007; Martinsen et al., 2008; Marzal et al., 2011; Atkinson, C.T., Woods, K.L., Dusek, R.J., Sileo, L.S., Iko, W.M., 1995. Wildlife disease Perkins and Schall, 2002). However, the DNA barcoding is still in and conservation in Hawaii: pathogenicity of avian malaria (Plasmodium relictum) in experimentally infected iiwi (Vestiaria coccinea). Parasitol. 111, 59–69. its infancy in wildlife haemosporidian parasites and only a small doi:10.1017/S003118200007582X. number of lineages (sequences) have been linked with their mor- Atkinson, C.T., Dusek, R.J., Woods, K.L., Iko, W.M., 2000. Pathogenicity of avian malaria phological species. Of approximately 500 deposited cyt b lineages in experimentally-infected Hawaii amakihi. J. Wildl. Dis. 36, 197–204. doi:10.7589/0090-3558-36.2.197. of Plasmodium, 27 have been linked with their morphological species Atkinson, C.T., Dusek, R.J., Lease, J.K., 2001. Serological responses and immunity to (MalAvi database, http://mbio-serv2.mbioekol.lu.se/Malavi/ superinfection with avian malaria in experimentally-infected Hawaii amakihi. index.html accessed on 25.03.2014). Taxonomic experimental studies J. Wildl. Dis. 37, 20–27. doi:10.7589/0090-3558-37.1.20. Atkinson, C.T., Saili, K.S., Utzurrum, R.B., Jarvi, S.I., 2013. Experimental evidence for supplemented with DNA sequence information are essential for de- evolved tolerance to avian malaria in a wild population of low elevation Hawai’i’ veloping barcodes for avian malaria parasites. Amakihi (Hemignathus virens). EcoHealth 10, 366–375. doi:10.1007/s10393-013- 0899-2. 4.5. Conclusions Beadell, J.S., Gering, E., Austin, J., Dumbacher, J.P., Peirce, M.A., Pratt, T.K., et al., 2004. Prevalence and differential host-specificity of two avian blood parasite genera in the Australo-Papuan region. Mol. Ecol. 13, 3829–3844. doi:10.1111/j.1365- Former experimental studies showed that the same Plasmo- 294X.2004.02363.x. dium lineages can cause markedly different pathology in different Beadell, J.S., Ishtiaq, F., Covas, R., Melo, M., Warren, B.H., Atkinson, C.T., et al., 2006. Global phylogeographic limits of Hawaii’s avian malaria. Proc. R. Soc. Lond. B. bird species (Palinauskas et al., 2008). This study indicates (Tables 1 Biol Sci 273, 2935–2944. doi:10.1098/rspb.2006.3671. and 2) that there might be differences in the avian host-susceptibility Bensch, S., Stjernman, M., Hasselquist, D., Ostman, O., Hansson, B., Westerdahl, H., not only among different parasite species, but also among differ- et al., 2000. Host specificity in avian blood parasites: a study of Plasmodium and Haemoproteus mitochondrial DNA amplified from birds. Proc. R. Soc. Lond. B. Biol ent cyt b lineages of the same morphospecies and even different Sci 267, 1583–1589. doi:10.1098/rspb.2000.1181. isolates of the same cyt b lineage. This finding is in accordance with Bensch, S., Pérez-Tris, J., Waldenström, J., Hellgren, O., 2004. Linkage between nuclear former experimental observations about marked difference of vir- and mitochondrial DNA sequences in avian malaria parasites: multiple cases of cryptic speciation? Evolution 58, 1617–1621. doi:10.1111/j.0014-3820.2004 ulence in different strains of Plasmodium juxtanucleare in domestic .tb01742.x. chickens (Garnham, 1980; Valkiu¯ nas, 2005). Bensch, S., Waldenström, J., Jonzén, N., Westerdahl, H., Hansson, B., Sejberg, D., The key results of this experimental study are: (1) malaria para- et al., 2007. Temporal dynamics and diversity of avian malaria parasites in a single host species. J. Anim. Ecol. 76, 112–122. doi:10.1111/j.1365-2656.2006 sites of the subgenera Novyella and Huffia are more vertebrate host .01176.x. specific than species of Haemamoeba and Giovannolaia; (2) suscep- Bensch, S., Hellgren, O., Pérez-Tris, J., 2009. MalAvi: a public database of malaria tibility of the same avian host varies between cyt b lineages of the parasites and related haemosporidians in avian hosts based on mitochondrial same parasite species; (3) susceptibility of the same avian host is cytochrome b lineages. Mol. Ecol. Resour. 9, 1353–1358. doi:10.1111/j.1755- 0998.2009.02692.x. also different to the various isolates of the same parasite lineage; BirdLife International, 2014. Species factsheet: terpsiphone viridis. Downloaded. (4) birds are more susceptible to co-infections, which are more vir- Available from: (accessed 14.05.14.). ulent than single infections; (5) canaries are good experimental hosts Cannell, B.L., Krasnec, K.V., Campbell, K., Jones, H.I., Miller, R.D., Stephens, N., 2013. The pathology and pathogenicity of a novel Haemoproteus spp. infection in wild for many lineages of the subgenera Haemamoeba and Giovannolaia, Little Penguins (Eudyptula minor). Vet. Parasitol. 197, 74–84. doi:10.1016/ but not of Novyella; (6) different isolates of Huffia species develop j.vetpar.2013.04.025. in canaries, but parasitaemia is usually low and transient; (7) Cellier-Holzem, E., Esparza-Salas, R., Garnier, S., Sorci, G., 2010. Effect of repeated exposure to Plasmodium relictum (lineage SGS1) on infection dynamics in ducklings are mainly unsusceptible to almost all lineages of domestic canaries. Int. J. Parasitol. 40, 1447–1453. doi:10.1016/ malaria parasites, but light and transient parasitaemia usually j.ijpara.2010.04.014. develops. Cornet, S., Bichet, C., Larcombe, S., Faivre, B., Sorci, G., 2014. Impact of host nutritional status on infection dynamics and parasite virulence in a bird-malaria system. J. This study offers novel information and indicates possible di- Anim. Ecol. 83, 256–265. doi:10.1111/1365-2656.12113. rections for the design of future experimental investigations on the Cramp, S., 2004. Birds of the Western Palearctic Interactive. Oxford University Press. variety of host–parasite–vector interactions, including specificity and Cranfield, M., Graczyk, T., Beall, F., Ialeggio, D., Shaw, M., Skjoldager, M., 1994. Subclinical avian malaria infections in African black-footed penguins (Spheniscus virulence studies in the wildlife. Such research is essential for es- demersus) and induction of parasite recrudescence. J. Wildl. Dis. 30, 372–376. timation of the true impact of the pathogens on bird populations doi:10.7589/0090-3558-30.3.372. in the conditions of rapidly changing environment. Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2012. jModelTest 2: more models, new heuristics and parallel computing. Nat. Methods 9, 772. doi:10.1038/ nmeth.2109. Acknowledgments Dimitrov, D., Zehtindjiev, P., Bensch, S., 2010. Genetic diversity of avian blood parasites in SE Europe: cytochrome b lineages of the genera Plasmodium and Haemoproteus We would like to acknowledge the ringers and the staff of the (Haemosporida) from Bulgaria. Acta Parasitol. 55, 201–209. doi:10.2478/s11686- 010-0029-z. “Rybachy” and Kalimok Biological Stations (Report no 57) for their Dimitrov, D., Valkiu¯ nas, G., Zehtindjiev, P., Ilieva, M., Bensch, S., 2013. Molecular help in catching, handling and maintaining the birds in captivity. characterization of haemosporidian parasites (Haemosporida) in yellow wagtail We are grateful to Aneliya Bobeva, Vilius Armalis and Martin P. (Motacilla flava), with description of in vitro ookinetes of Haemoproteus motacillae. Zootaxa 3666, 369–381. doi:10.11646/zootaxa.3666.3.7. Marinov for their help during field work. Violeta Skukauskaite˙- Draper, C.C., 1953. Observations on the reciprocal immunity between some avian Kokiniene˙ , Asghar Mohammed and Asta Križanauskiene˙are plasmodia. Parasitol. 43, 139–142. doi:10.1017/S0031182000018412. D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16 15

Ewen, J.G., Bensch, S., Blackburn, T.M., Bonneaud, C., Brown, R., Cassey, P., et al., 2012. Marzal, A., Ricklefs, R.E., Valkiu¯ nas, G., Albayrak, T., Arriero, E., Bonneaud, C., et al., Establishment of exotic parasites: the origins and characteristics of an avian 2011. Diversity, loss, and gain of malaria parasites in a globally invasive bird. malaria community in an isolated island avifauna. Ecol. Lett. 15, 1112–1119. PLoS ONE 6, e21905. doi:10.1371/journal.pone.0021905. doi:10.1111/j.1461-0248.2012.01833.x. Olias, P., Wegelin, M., Zenker, W., Freter, S., Gruber, A.D., Klopfleisch, R., 2011. Avian Ferrell, S.T., Snowden, K., Marlar, A.B., Garner, M., Lung, N.P., 2007. Fatal hemoprotozoal malaria deaths in parrots, Europe. Emerg. Infect. Dis. 17l, 950–952. doi:10.3201/ infections in multiple avian species in a zoological park. J. Zoo Wildl. Med. 38, eid1705.101618. 309–316. doi:10.1638/1042-7260(2007)038[0309:FHIIMA]2.0.CO;2. Palinauskas, V., Kosarev, V., Shapoval, A., Bensch, S., Valkiu¯ nas, G., 2007. Comparison Fix, A.S., Waterhouse, C., Greiner, E.C., Stoskopf, M.K., 1988. Plasmodium relictum as of mitochondrial cytochrome b lineages and morphospecies of two avian malaria a cause of avian malaria in wild-cought magellanic penguins (Sphenicus parasites of the subgenera Haemamoeba and Giovannolaia (Haemosporida: magellanicus). J. Wildl. Dis. 24, 610–619. doi:10.7589/0090-3558-24.4.610. Plasmodiidae). Zootaxa 1626, 39–50. Garamszegi, L.Z., 2011. Climate change increases the risk of malaria in birds. Glob. Palinauskas, V., Valkiu¯ nas, G., Bolshakov, C., V, Bensch, S., 2008. Plasmodium relictum Change Biol. 17, 1751–1759. doi:10.1111/j.1365-2486.2010.02346.x. (lineage P-SGS1): effects on experimentally infected passerine birds. Exp. Parasitol. Garnham, P.C.C., 1966. Malaria Parasites and Other Haemosporidia. Blackwell Scientific 120, 372–380. doi:10.1016/j.exppara.2008.09.001. Publications Ltd., Oxford. Palinauskas, V., Valkiu¯ nas, G., Krizanauskiene, A., Bensch, S., Bolshakov, C.V., Garnham, P.C.C., 1980. Malaria in its various vertebrate hosts. In: Kreier, J.P. (Ed.), 2009. Plasmodium relictum (lineage P-SGS1): further observation of effects Malaria. Part 1. Epidemiology, Chemotherapy, Morphology and Metabolism. on experimentally infected passeriform birds, with remarks on treatment Academic Press, New York, pp. 95–144. with Malarone. Exp. Parasitol. 123, 134–139. doi:10.1016/j.exppara.2009.06 Glaizot, O., Fumagalli, L., Iritano, K., Lalubin, F., Van Rooyen, J., Christe, P., 2012. High .012. prevalence and lineage diversity of avian malaria in wild populations of great Palinauskas, V., Valkiu¯ nas, G., Bolshakov, C.V., Bensch, S., 2011. Plasmodium relictum tits (Parus major) and mosquitoes (Culex pipiens). PLoS ONE 7, e34964. (lineage SGS1) and Plasmodium ashfordi (lineage GRW2): the effects of the doi:10.1371/journal.pone.0034964. co-infection on experimentally infected passerine birds. Exp. Parasitol. 127, Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor 527–533. doi:10.1016/j.exppara.2010.10.007. and analysis program for Windows95/98/NT. Nucleic Acids Symp. Ser. 41, 95– Palinauskas, V., Žiegyte˙ , R., Ilgu¯ nas, M., Iezhova, T.A., Bernotiene˙ , R., Bolshakov, C., 98. et al., 2014. Description of the first cryptic avian malaria parasite, Plasmodium Hellgren, O., Waldenström, J., Bensch, S., 2004. A new PCR assay for simultaneous homocircumflexum n. sp., with experimental data on its virulence and studies of Leucocytozoon, Plasmodium and Haemoproteus from avian blood. J. development in avian hosts and mosquitoes. Int. J. Parasitol. doi:10.1016/ Parasitol. 90, 797–802. doi:10.1645/GE-184R1. j.ijpara.2014.08.012. Hellgren, O., Waldenström, J., Peréz-Tris, J., Szöll, E., Hasselquist, D., Križanauskiene˙, Perkins, S.L., Schall, J.J., 2002. A molecular phylogeny of malarial parasites recovered A., et al., 2007. Detecting shifts of transmission areas in avian blood parasites: from cytochrome b gene sequences. J. Parasitol. 88, 972–978. doi:10.1645/0022- a phylogenetic approach. Mol. Ecol. 16, 1281–1290. doi:10.1111/j.1365- 3395(2002)088[0972:AMPOMP]2.0.CO;2. 294X.2007.03227.x. Permin, A., Juhl, J., 2002. The development of Plasmodium gallinaceum infections in Iezhova, T.A., Valkiu¯ nas, G., Bairlein, F., 2005. Vertebrate host specificity of two avian chickens following single infections with three different dose levels. Vet. Parasitol. malaria parasites of the subgenus Novyella: Plasmodium nucleophilum and 105, 1–10. doi:10.1016/S0304-4017(01)00645-8. Plasmodium vaughani. J. Parasitol. 91, 458–461. doi:10.1645/GE-3377RN. Richardson, D.S., Jury, F.L., Blaakmeer, K., Komdeur, J., Burke, T., 2001. Parentage Ilgu¯ nas, M., Palinauskas, V., Iezhova, T.A., Valkiu¯ nas, G., 2013. Molecular and assignment and extra-group paternity in a cooperative breeder: the Seychelles morphological characterization of two avian malaria parasites (Haemosporida: warbler (Acrocephalus sechellensis). Mol. Ecol. 10, 2263–2273. doi:10.1046/j.0962- Plasmodiidae), with description of Plasmodium homonucleophilum n. sp. Zootaxa 1083.2001.01355.x. 3666, 49–61. doi:10.11646/zootaxa.3666.1.5. Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes 3: bayesian phylogenetic inference under Jarvi, S.I., Schultz, J.J., Atkinson, C.T., 2002. PCR diagnostics underestimate the mixed models. Bioinformatics 19, 1572–1574. doi:10.1093/bioinformatics/btg180. prevalence of avian malaria (Plasmodium relictum) in experimentally-infected Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning, a Laboratory Manual. passerines. J. Parasitol. 88, 153–158. doi:10.1645/0022-3395(2002)088 Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. [0153:PDUTPO]2.0.CO;2. Sorci, G., 2013. Immunity, resistance and tolerance in bird-parasite interactions. Jarvi, S.I., Farias, M.E., LaPointe, D.A., Belcaid, M., Atkinson, C.T., 2013. Next-generation Parasite Immunol. 35, 350–361. doi:10.1111/pim.12047. sequencing reveals cryptic mtDNA diversity of Plasmodium relictum Szöllo˝ si, E., Cichon, M., Eens, M., Hasselquist, D., Kempenaers, B., Merino, S., et al., in the Hawaiian Islands. Parasitol. 140, 1741–1750. doi:10.1017/ 2011. Determinants of distribution and prevalence of avian malaria in blue tit S0031182013000905. populations across Europe: separating host and parasite effects. J. Evol. Biol. 24, Kazlauskiene˙ , R., Bernotiene, R., Palinauskas, V., Iezhova, T.A., Valkiu¯ nas, G., 2013. 2014–2024. doi:10.1111/j.1420-9101.2011.02339.x. Plasmodium relictum (lineages pSGS1 and pGRW11): complete synchronous Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013. MEGA6: molecular sporogony in mosquitoes Culex pipiens pipiens. Exp. Parasitol. 133, 454–461. evolutionary genetics analysis version 6.0. Mol. Biol. Evol. 30, 2725–2729. doi:10.1016/j.exppara.2013.01.008. doi:10.1093/molbev/mst197. Kim, K.S., Tsuda, Y., 2010. Seasonal changes in the feeding pattern of Culex pipiens Valkiu¯ nas, G., 2005. Avian Malaria Parasites and Other Haemosporidia. CRC Press, pallens govern the transmission dynamics of multiple lineages of avian malaria Boca Raton, FL. parasites in Japanese wild bird community. Mol. Ecol. 19, 5545–5554. Valkiu¯ nas, G., Zehtindjiev, P., Hellgren, O., Ilieva, M., Iezhova, T.A., Bensch, S., 2007. doi:10.1111/j.1365-294X.2010.04897.x. Linkage between mitochondrial cytochrome b lineages and morphospecies of Kulma, K., Low, M., Bensch, S., Qvarnström, A., 2013. Malaria infections reinforce two avian malaria parasites, with a description of Plasmodium (Novyella) ashfordi competitive asymmetry between two Ficedula flycatchers in a recent contact zone. sp. nov. Parasitol. Res. 100, 1311–1322. doi:10.1007/s00436-006-0409-3. Mol. Ecol. 22, 4591–4601. doi:10.1111/mec.12409. Valkiu¯ nas, G., Zehtindjiev, P., Dimitrov, D., Krizanauskiene, A., Iezhova, T.A., Bensch, LaPointe, D.A., Goff, M.L., Atkinson, C.T., 2005. Comparative susceptibility of introduced S., 2008. Polymerase chain reaction-based identification of Plasmodium (Huffia) forest-dwelling mosquitoes in Hawai’i to avian malaria, Plasmodium relictum.J. elongatum, with remarks on species identity of haemosporidian lineages Parasitol. 91, 843–849. doi:10.1645/GE-3431.1. deposited in GenBank. Parasitol. Res. 102, 1185–1193. doi:10.1007/s00436-008- Larcombe, S., Bichet, C., Cornet, S., Faivre, B., Sorci, G., 2013. Food availability and 0892-9. competition do not modulate the costs of Plasmodium infection in dominant male Valkiu¯ nas, G., Kazlauskiene˙ , R., Bernotiene˙ , R., Palinauskas, V., Iezhova, T.A., 2013. canaries. Exp. Parasitol. 135, 708–714. doi:10.1016/j.exppara.2013.10.012. Abortive long-lasting sporogony of two Haemoproteus species (Haemosporida, Levin, I.I., Zwiers, P., Deem, S.L., Geest, E.A., Higashiguchi, J.M., Iezhova, T.A., et al., Haemoproteidae) in the mosquito Ochlerotatus cantans, with perspectives on 2013. Multiple lineages of avian malaria parasites (Plasmodium) in the Galapagos haemosporidian vector research. Parasitol. Res. 112, 2159–2169. doi:10.1007/ islands and evidence for arrival via migratory birds. Conserv. Biol. 27, 1366–1377. s00436-013-3375-6. doi:10.1111/cobi.12127. Valkiu¯ nas, G., Palinauskas, V., Ilgu¯ nas, M., Bukauskaite˙ , D., Dimitrov, D., Bernotiene˙, Loiseau, C., Harrigan, R.J., Cornel, A.J., Guers, S.L., Dodge, M., Marzec, T., et al., 2012a. R., et al., 2014. Molecular characterization of five widespread avian First evidence and predictions of Plasmodium transmission in Alaskan bird haemosporidian parasites (Haemosporida), with perspectives on the PCR-based populations. PLoS ONE 7, e44729. doi:10.1371/journal.pone.0044729. detection of haemosporidians in wildlife. Parasitol. Res. 113, 2251–2263. Loiseau, C., Harrigan, R.J., Robert, A., Bowie, R.C.K., Thomassen, H.A., Smith, T.B., et al., doi:10.1007/s00436-014-3880-2. 2012b. Host and habitat specialization of avian malaria in Africa. Mol. Ecol. 21, Van Riper, C., III, Van Riper, S.G., Goff, M.L., Laird, M., 1986. The epizootiology and 431–441. doi:10.1111/j.1365-294X.2011.05341.x. ecological significance of malaria in Hawaiian land birds. Ecol. Monogr. 56, Mackinnon, M.J., Read, A.F., 2004. Immunity promotes virulence evolution in a malaria 327–344. doi:10.2307/1942550. model. PLoS Biol. 2, E230. doi:10.1371/journal.pbio.0020230. Van Rooyen, J., Lalubin, F., Glaizot, O., Christe, P., 2013. Avian haemosporidian Martínez, J., Martínez-De La Puente, J., Herrero, J., Del Cerro, S., Lobato, E., persistence and co-infection in great tits at the individual level. Malar. J. 12, 40. Rivero-De Aguilar, J., et al., 2009. A restriction site to differentiate Plasmodium doi:10.1186/1475-2875-12-40. and Haemoproteus infections in birds: on the inefficiency of general primers for Waldenström, J., Bensch, S., Kiboi, S., Hasselquist, D., Ottusson, U., 2002. Cross-species detection of mixed infections. Parasitol. 136, 713–722. doi:10.1017/ infection of blood parasites between resident and migratory songbirds in Africa. S0031182009006118. Mol. Biol. Evol. 11, 1545–1554. doi:10.1046/j.1365-294X.2002.01523.x. Martinsen, E.S., Perkins, S.L., Schall, J.J., 2008. A three-genome phylogeny of malaria Warner, R.E., 1968. The role of introduced diseases in the extinction of the endemic parasites (Plasmodium and closely related genera): evolution of life-history traits Hawaiian avifauna. Condor 70, 101–120. and host switches. Mol. Phylogenet. Evol. 47, 261–273. doi:10.1016/ Westerdahl, H., Asghar, M., Hasselquist, D., Bensch, S., 2012. Quantitative disease j.ympev.2007.11.012. resistance: to better understand parasite-mediated selection on major Marzal, A., 2012. Recent advances in studies on avian malaria parasites. In: Okwa, histocompatibility complex. Proc. R. Soc. Lond. B. Biol Sci 279, 577–584. O.O. (Ed.), Malaria Parasites. INTECH, pp. 135–158. doi:10.1098/rspb.2011.0917. 16 D. Dimitrov et al./Experimental Parasitology 148 (2015) 1–16

Williams, R.B., 2005. Avian malaria: clinical and chemical pathology of Plasmodium Zehtindjiev, P., Ilieva, M., Westerdahl, H., Hansson, B., Valkiu¯ nas, G., Bensch, S., gallinaceum in the domesticated fowl Gallus gallus. Avian Pathol. 34, 29–47. 2008. Dynamics of parasitemia of malaria parasites in a naturally and doi:10.1080/03079450400025430. experimentally infected migratory songbird, the great reed warbler Acro- Yorinks, N., Atkinson, C.T., 2000. Effects of malaria on activity budgets of cephalus arundinaceus. Exp. Parasitol. 119, 99–110. doi:10.1016/j.exppara experimentally infected juvenile apapane (Himatione sanguinea). The Auk 117, .2007.12.018. 731–738. doi:10.1642/0004-8038(2000)117[0731:EOMOAB]2.0.CO;2.