Comparative Analysis of the Exo-Erythrocytic Development of Five Lineages of Haemoproteus majoris, a Common Haemosporidian Parasite of European Passeriform Birds

Haemoproteus parasites (Apicomplexa, Haemosporida) are widespread pathogens of birds, with a rich genetic (about 1900 lineages) and morphospecies (178 species) diversity. Nonetheless, their life cycles are poorly understood. The exo-erythrocytic stages of three Haemoproteus majoris (widespread generalist parasite) lineages have been previously reported, each in a different bird species. We aimed to further study and compare the development of five H. majoris lineages—hCCF5, hCWT4, hPARUS1, hPHSIB1, and hWW2—in a wider selection of natural avian hosts. A total of 42 individuals belonging to 14 bird species were sampled. Morphospecies and parasitemia were determined by microscopy of blood films, lineages by DNA-barcoding a 478 bp section of the cytochrome b gene, and exo-erythrocytic stages by histology and chromogenic in situ hybridization. The lineage hCWT4 was morphologically characterized as H. majoris for the first time. All lineage infections exclusively featured megalomeronts. The exo-erythrocytic stages found in all examined bird species were similar, particularly for the lineages hCCF5, hPARUS1, and hPHSIB1. Megalomeronts of the lineages hWW2 and hCWT4 were more similar to each other than to the former three lineages. The kidneys and gizzard were most often affected, followed by lungs and intestines; the site of development showed variation depending on the lineage.


Introduction
Haemoproteus species (Apicomplexa, Haemosporida, Haemoproteidae) are parasites of birds and found on all continents except Antarctica [1,2]. These pathogens multiply asexually and produce gametocytes in their avian hosts before being transmitted to their dipteran vectors (biting midges, Ceratopogonidae), in which the fusion of the gametes and the development of sporozoites occur [1,3].
Anemia is an acknowledged symptom caused by Haemoproteus parasites, with the gametocytes developing in the erythrocytes [4,5]. However, the exo-erythrocytic development, which occurs in different organs, was initially assessed to be less harmful for their avian hosts compared with that of haemosporidian parasites of the genera Plasmodium and Leucocytozoon [6], and research on Haemoproteus tissue stages remained scarce. Recent studies reported exo-erythrocytic stages of several Haemoproteus species [7][8][9][10][11][12][13][14][15], not only increasing and changing traditional knowledge of these parasites but also pointing out how little is still known about their development in avian hosts.
Most reports of exo-erythrocytic stages are either from hosts found naturally dead [7,8,10,11,15] or from individuals that have been purposely targeted for their parasite infection collected from birds collected in 2019 to 2022. The organs were fixed in 10% formalin and later embedded in paraffin wax.
In July 2021, one black redstart (Phoenicurus ochruros) found dead in Lithuania (54 • 52 58.4 N 25 • 25 38.6 E) with a skull fracture was frozen and brought to the P. B. Šivickis Laboratory of Parasitology, Nature Research Centre, Lithuania. In October 2021, one long-tailed tit (Aegithalos caudatus) found dead at the Ventes Ragas Ornithological Station, was frozen until dissection. Both birds were dissected, the heart was pressed to a glass slide to prepare a blood film, little pieces of organs were collected in 96% EtOH and stored at +4 • C, and the organs were fixed in formalin for 24 h, washed in distilled water for 1 h, transferred to 70% EtOH, and processed as the other samples.

Blood Film Microscopy
Haemosporidian parasite species were determined by screening the blood films with an Olympus BX61 microscope (Olympus, Tokyo, Japan) at ×1000 magnification [1]. Parasitemia intensity was calculated as the number of infected erythrocytes per 1000 erythrocytes or per 10,000 erythrocytes in case of low infection and expressed as percentage [24].
Images of the parasites were taken using the same microscope equipped with an Olympus DP70 digital camera and AnalySIS FIVE software (Olympus, Tokyo, Japan). The software Adobe InDesign CS6 (Adobe, https://www.adobe.com/products/indesign.html (accessed on 25 June 2023)) was used to prepare the gametocytes figure.
Chromogenic in situ hybridization (CISH) was applied on additional sections, at least one per individual, following the original protocol [8,25]. A genus-specific oligonucleotide probe (Haemo18S), which targets the 18S ribosomal RNA of Haemoproteus parasites, was used to confirm the generic origin of the observed tissue stages [8].
Pictures of the tissue stages were taken using the camera DP12 with the software DP-SOFT or the camera UC90 with the software cellSens (Olympus, Tokyo, Japan), respectively. The software CorelDraw 2019 (RRID:SCR_014235, https://www.coreldraw.com/en/) was used to prepare the figures. (accessed on 9 November 2022) For all bird individuals single-infected with H. majoris hPARUS1, the number of sections investigated and the number of sections positive for exo-erythrocytic stages were counted and reported. The number of very young megalomeronts and other megalomeronts (referring to growing, mature, and ruptured megalomeronts) were also counted and reported. It is to be noted that very young megalomeronts can be found on up to three sequential sections of organs, but most often, one very young megalomeront was observed in only one section. Other, bigger megalomeronts are most often observed on several sequential sections. However, not all sections investigated were sequential. Therefore, the detection of megalomeronts is biased greatly for big megalomeronts compared with very young megalomeronts. Due to the fact that the actual shape (oval, round, long, or slender) of megalomeronts is difficult to predict, the choice was made to not try to distinguish the real number of the bigger megalomeronts.

Molecular Analysis
DNA extractions were done using the SET-buffer-stored blood following the ammonium acetate protocol [26] or the DNeasy Blood & Tissue Kit (QIAGEN, Venlo, The Netherlands). The extracted DNA was used as template in a nested PCR following the original protocols [23,27] with the outer primer pair HAEMNFI/HAEMNR3 and inner primer pairs HAEMF/HAEMR2 for detection of Haemoproteus and Plasmodium parasites and HAEMFL/HAEMR2L for Leucocytozoon parasites. A negative control (ddH 2 O) and a positive control (sample known to be infected with Plasmodium and Leucocytozoon parasites) were used in all PCRs to check for possible contamination and efficiency of the PCR. PCR products were run on 2% agarose gel. Positive PCR products were sequenced in both directions with the Big Dye Terminator V3.1 Cycle Sequencing Kit and the ABI PRISM TM 3100 capillary sequencing robot (Applied Biosystems, Foster City, California) or sent for sequencing to Microsynth (Microsynth, Austria). One sample was subjected to cloning of the 18S ribosomal RNA of parasites following the protocol of a previous study [28].
Chromatograms were checked for possible mixed infections (more than one peak for a base) using the software Geneious Prime 2022.0.2 (https://www.geneious.com, last accessed on 9 November 2022) or Bioedit (https://bioedit.software.informer.com last accessed on 9 November 2022). Sequences were subjected to BLAST searches in MalAvi database [16] and NCBI GenBank database (National Library of Medicine, Bethesda, Maryland, https://www.ncbi.nlm.nih.gov/genbank/, last accessed on 9 November 2022) to determine the lineages. Sequences obtained from infections with no more than two lineages from the same genus were deposited in GenBank OR143042-OR143096 (i.e., sequences from the two bird individuals with mixed infections of possibly four lineages of H. majoris were not deposited).
Genetic pairwise distances between the lineages were calculated in the software MEGA-X: Molecular Evolutionary Genetics Analysis across computing platforms [29].
The gametocytes present in the hCWT4 infection display the main H. majoris characteristics ( Figure 1E-H): gametocytes grow around the erythrocyte nuclei but do not encircle them; dumbbell-shaped growing gametocytes are present; and pigment granules are roundish, sometimes oval, of medium to small size, and are randomly scattered. The parasite morphology was the same as described before, and its description is not repeated here [1]. Molecular analysis confirmed the H. majoris infections of the dissected birds and revealed five co-infections of different H. majoris lineages, two co-infections with other Haemoproteus lineages, and eleven co-infections with Leucocytozoon lineages (Tables 1 and S1). In total, seven birds were found infected only with the hCCF5 lineage, eight with hCWT4, twelve with hPARUS1, three with hPHSIB1, and seven with hWW2; these numbers exclude the samples with mixed infection of different H. majoris lineages (Tables 1 and S1). Nine of the identified co-infections by microscopy were not picked up by PCR, and only one lineage was identified in those infections.
Haemoproteus species were identified by microscopic examination in samples with co-infections where exo-erythrocytic stages were found (Table 1). The gametocytes present in the hCWT4 infection display the main H. majoris characteristics ( Figure 1E-H): gametocytes grow around the erythrocyte nuclei but do not encircle them; dumbbell-shaped growing gametocytes are present; and pigment granules are roundish, sometimes oval, of medium to small size, and are randomly scattered. The parasite morphology was the same as described before, and its description is not repeated here [1].  Information is organized by lineage of infection and data are provided for the season of collection, parasitemia, host species, parasite identification (through microscopy and PCR), and organs in which the exo-erythrocytic stages were found both in haematoxylin and eosin (H&E)-stained sections and the sections treated with chromogenic in situ hybridization (CISH) with the probe Haemo18S.  Birds negative for exo-erythrocytic stages were excluded from this table and are present in Table S1.

Haemoproteus majoris Exo-Erythrocytic Stages
Only megalomeronts were found during all single H. majoris infections. Meronts were only present in samples with mixed infections of H. majoris and other Haemoproteus species. Megalomeronts were found in 25 of the 42 investigated birds. However, not all organs were collected from 10 of the 17 individuals that were negative for exo-erythrocytic stages (Tables 1 and S1).

Exo-Erythrocytic Stages Morphology
hPARUS1, hCCF5, and hPHSIB1 lineages of H. majoris displayed similar morphology to that of megalomeronts (Figures 2A-D,J-S and S1, S3), which were surrounded by a thick capsular-like wall and possessed variously shaped interconnected cytomeres in which merozoites developed. Some megalomeronts in hCWT4 and hWW2 infections were slightly different in their morphology ( Figure 2E-I,T-X). Mainly, the cytomeres were present, but they were denser and interconnected more tightly (compare Figure 2F,G,W,X with Figure 2A-C,J). Interestingly, their host species also differed, with hPARUS1 found predominantly in Parus spp., hCCF5 in Fringilla coelebs, and hCWT4 and hWW2 in Sylvia spp. (Tables 1 and S1).
Some of the megalomeronts found in the gizzard with hCWT4 infection ( Figure 2E) developed in the koilin layer and looked degenerated, reacting poorly with the Haemo18S probe, while other megalomeronts at a similar stage of growth showed a deep purple CISH signal. Other parasite tissue structures were found in the gizzard with varying shapes, some of them appearing ruptured with very poor to no CISH signal ( Figures S2 and S4).
Megalomeront morphology of the same H. majoris lineages' infection did not differ between host species. For example, megalomeronts of hPARUS1 infections found in Cyanistes caeruleus, Parus major, and Parus montanus were of the same morphology ( Figures 2J-N and S3).
Very young megalomeronts (less than 20 µm in diameter) were found in 10 birds, 1 in hCCF5, 1 in hCWT4, 5 in hPARUS1, 1 in hPHSIB1 lineage infections, and 2 in H. majoris co-infections. The host cell nucleus was present and slightly enlarged (Figures 2D,I,N,S, 3B-D and S3). The host cell nucleus was not seen in more advanced, developing, and mature megalomeronts ( Inflammatory reactions were observed around some megalomeronts. The inflammatory infiltrates were predominantly lymphohistiocytic, and only occasionally were single heterophils seen. Inflammation was present around seemingly intact megalomeronts, where the intensity could vary between single inflammatory cells ( Figure 2B) and several layers of them ( Figure 2O-R,T). Adjacent to damaged or ruptured parasitic structures the inflammation was generally severe.
Individuals infected with the same parasite lineage and a similar intensity of parasitemia often featured megalomeronts (Table 1) but not always (Table S1).

Exo-Erythrocytic Stages Site of Development
Considering all the examined lineages of H. majoris, the kidneys and the gizzard were most commonly affected by megalomeronts, followed by the lungs and intestine, even though 10 birds had not their intestine and gizzard collected and investigated (Tables 1 and S1). By considering each lineage of the infection, the lungs, gizzard, and intestine were most often affected in hCCF5 infections, the gizzard in hCWT4 infections, the kidneys and lungs in hPARUS1 infections, and the heart and gizzard in hWW2 infections. In hPHSIB1 infections, only one individual was positive for megalomeronts, with the kidneys affected. In co-infections with several H. majoris lineages, the intestine and the brain were most commonly affected (Tables 1 and S1).
The number of H. majoris exo-erythrocytic stages present per organ varied per histological section analyzed. In most organs, megalomeronts were only seen sporadically, whereas, in the kidneys, pancreas, and gizzard, megalomeronts were found in greater numbers (up to 8 megalomeronts were seen in a single section) (Figures 2A,T, 3E, S2C and S3K,AD).

Seasonal Occurrence of Exo-Erythrocytic Stages of H. majoris hPARUS1
hPARUS1 lineage was sampled in spring, summer, and autumn and displayed exoerythrocytic stages differently. Megalomeronts were seen in all seasons, however, with different frequencies (Table S2). In spring, the kidneys, lungs, liver, heart, intestine, and gizzard were found affected in 4 to 21 sections out of 13 to 31 sections for the 3 individuals investigated, while in summer megalomeronts were found in 1 to 5 sections out of 9 sections examined for one individual (Table S2). In autumn, mainly the kidneys were found to be affected (with the spleen found to be affected once in 1 bird only), with 21 sections positive for exo-erythrocytic stages out of 304 sections of kidneys investigated from 8 individuals (Table S2). Ten times more slides were thus investigated to find approximately the same number of positive sections for the kidneys between individuals collected in spring and autumn (Table S2). The other organs were not found affected by megalomeronts in autumn, even after investigating more than 100 sections, while in spring and summer, around 10 investigated sections reported a high number of exo-erythrocytic stages (Table S2).
In hPARUS1 single infections of Haemoproteus, very young megalomeronts were found in two individuals in spring and in four individuals in autumn. More of these very young megalomeronts were found in individuals in autumn (26 found in 21 positive sections out of 304 sections of the kidneys investigated) than in spring (3 found in 20 positive sections out of 31 sections of the kidneys investigated). In spring, more than 1 big megalomeront was found per section, adding up to more than 90 megalomeronts (without distinguishing if it is a different megalomeront from the previous section) in the kidneys in 20 positive sections out of 31, while in autumn, 11 big megalomeronts were found in the 304 investigated sections of kidneys.

Exo-Erythrocytic Stages in H. majoris Co-Infections with Several of Its Lineages
Five samples were found to be co-infected with several lineages of H. majoris, as determined by PCR-based testing and the obtained cyt b sequences (Tables 1 and S1). The megalomeront found in the brain of one individual co-infected with several H. majoris lineages ( Figure 3A) could not be confirmed as Haemoproteus as the structure was absent in the CISH-tested section. Only eosinophilic staining and no basophilic staining was observed, which could indicate the occurrence of a degenerating megalomeront.
Very young megalomeronts were found in the brain, kidneys, and intestine of two individuals, each co-infected with hWW2 and hPHSIB1. The very young megalomeront found in the intestine was bigger (~96 µm in largest diameter) than the other very young megalomeronts (<20 µm in largest diameter). However, the slightly enlarged host cell nucleus was still visible in the former ( Figure 3D).
Some of the megalomeronts found in sample '20 Sp' co-infected with the lineages hPHSIB1 and hWW2 ( Figure 3E) looked similar to those found in single infections with hCWT4 and hWW2 ( Figure 2F,G,W,X).

Exo-Erythrocytic Stages of Other Haemoproteus species
Small meronts (up to 25 µm) were found in a Fringilla coelebs (sample '1 Sp') where microscopic examination identified the presence of H. fringillae in blood films. Positive CISH signals with the Haemo18S probe were observed for these meronts, confirming their Haemoproteus origin (Figure 3G insert). However, as no other sample investigated with single infection of H. majoris revealed small meronts, these stages likely belong to another Haemoproteus species, most likely H. fringillae, and not H. majoris.
Big meronts (up to 60 µm) were found in F. coelebs ('3 Sp') in a blood vessel of the lungs, and positive CISH signals were observed for the corresponding parasites seen in H&E-stained sections ( Figure 3H,I). The meronts were tightly grouped in a blood vessel, forming a cluster well visible under the microscope. Individual meronts of this cluster were at different stages of development, with merozoites well visible in some meronts, which showed little CISH signal, whereas other developing meronts had deep purple CISH signals ( Figure 3I insert). This indicates the asynchronous maturation of meronts and the asynchronous release of merozoites. Microscopic examination identified the presence of three other Haemoproteus species in the blood films of the F. coelebs individual '3 Sp' (H. fringillae, H. magnus, and H. sp.-Tables 1 and S1). It was not possible to determine the species identity of the Haemoproteus meronts found. However, as none of the H. majoris single-infected birds were found with such meronts, the latter are most likely not H. majoris but belong to one of the other Haemoproteus species present in the co-infection.

Discussion
The main result of this study is the report and comparative analysis of the exoerythrocytic stages (megalomeronts) for five lineages (hCCF5, hCWT4, hPARUS1, hPHSIB1, and hWW2) of H. majoris in different avian host species. Megalomeronts developed in the examined bird individuals during single infections. This is the first comparative study that addresses the exo-erythrocytic development of different lineages of one Haemoproteus species in different avian hosts, providing opportunities to address possible patterns of development in avian haemoproteids. This study also identified the lineage hCWT4 as another lineage of H. majoris; this contributes to better understanding the intra-species genetic diversity of this widespread Haemoproteus species.
The lineages hPARUS1, hPHSIB1, hCCF5, and hWW2 were identified as H. majoris by Križanauskienė et al. [19], but the parasitemia intensity was too low in the sample infected with hCWT4 to describe and identify the species [19]. In our study, the microscopic examination of blood films of all birds revealed H. majoris infections, including all hCWT4 infections and the co-infections with several other Haemoproteus species.
Twelve birds were infected with H. majoris lineage hPARUS1, and megalomeronts were found to develop in the kidneys of eight, in the lungs of three, in the liver of two, and in the heart, spleen, pancreas, and gizzard of one bird (Tables 1 and S1). The megalomeronts were found in different bird species (P. major, C. caeruleus, and P. montanus) and their structure was similar to that of the megalomeronts previously reported in the kidneys, liver, lungs, and spleen of a single P. major [9]. This confirms the kidneys as the main site of development of H. majoris hPARUS1.
A previous report identified the kidneys and the intestine as sites of development of H. majoris lineage hPHSIB1 [12]. In our study, the lineage hPHSIB1 was found in three bird species, with megalomeronts observed only in the kidneys of one P. sibilatrix (Tables 1 and S1). It should be noted that the intestine and the gizzard were not collected for histological purposes for two individuals (S. atricapilla and P. sibilatrix), while the third individual (Aegithalos caudatus) was found dead and no information on gametocytes was available. It is thus difficult to know if the intestine is one of the organs most commonly affected by hPHSIB1 infections or if this is only the case for the kidneys. No individuals were found infected with the lineage hPHYBOR04 of H. majoris, for which megalomeronts were described in the kidneys of a Turdus pilaris [9]. As was the case with lineage hPARUS1, only megalomeronts were found in all examined birds infected with the lineages hPHSIB1 (this study, [12]) and hPHYBOR04 [9].
H. majoris lineage hCCF5 was mostly found in Fringilla coelebs. Megalomeronts were found to develop mainly in the lungs, pancreas, and gizzard (Tables 1 and S1). Only one out of seven individuals showed megalomeronts in the kidneys. In this regard, hCCF5 seems different from the lineages hPARUS1, hPHSIB1, or hPHYBOR04, whose tissue stages mainly developed in the kidneys (this study, [9]). The morphology of the hCCF5 exo-erythrocytic stages was highly similar to the megalomeronts of the lineages hPARUS1, hPHYBOR04, and hPHSIB1 found in this and previous studies [9,12]. In all these megalomeronts, a capsular-like wall surrounded the parasite, in which interconnected cytomeres were seen to develop (Figure 2A-C).
H. majoris parasites of the lineages hCWT4 and hWW2 were mostly found affecting the gizzard (found in 5 individuals), even though the gizzard and intestine were not collected in 9 out of 15 birds. The kidneys were sampled and examined from all birds, but no megalomeronts were found during hCWT4 infections (eight individuals) and were seen only in one out of seven hWW2 infections. This differs compared with the previously reported sites of development of megalomeronts in hPARUS1, hPHSIB1, and hPHYBOR04 infections (this study, [9,12]). The morphology of some megalomeronts found in hCWT4 and hWW2 infections was slightly different from megalomeronts of other H majoris lineages. Mainly, the cytomeres were interconnected more tightly (similar to a spiderweb), while megalomeronts of hPARUS1, hPHSIB1, hCCF5, and hPHYBOR04 infections possessed interconnected cytomeres of a bigger size (compare Figure  2F,G,W,X with Figure 2A-C, Figures S2,S4 vs. S1,S3). This could reflect a lineage variation or a difference in the stage of development of the megalomeronts. The megalomeronts found in hCWT4 and hWW2 infections were more irregularly shaped compared with the typical roundish form of megalomeronts (compare Figure 2E-G,T-X with Figure 2A-C,J-M,O-R and Figures S2,S4 vs. S1,S3). This difference in shape could be due to the different locations in the organs. hCWT4 megalomeronts were found in the muscular layers of the gizzard, where they are exposed to varying pressures during peristaltic contractions of the organ. This could explain the irregular shapes of the megalomeronts observed. Megalomeronts in hCWT4 infections were also found to develop in the koilin cuticle of the gizzard, a thick protective layer on the inside of the gizzard that is continuously worn out and where no nutrients are delivered. This could be detrimental to the development of megalomeronts, as their growth might be altered. It is to be noted that none of the megalomeronts found in the koilin cuticle were mature ( Figure 2E), and the CISH signals were only light purple, suggesting that the development of the parasite had stopped (low expression of RNA while the stage was not mature either).
Megalomeronts with fully developed merozoites showed barely any CISH signal ( Figure S3P,AO), compared with the young and developing stages (Figures 2 and 3 vs. Figures S1-S5), a phenomenon already observed in other Haemoproteus parasites and suggesting low ribosomal content in mature stages [30]. The very young megalomeronts were found more easily due to their deep purple CISH signals, facilitating their search in the corresponding H&Estained sections. Their small sizes and location in host cells with the host cell nuclei present but slightly enlarged were striking compared with previous reports of Haemoproteus parasites [7][8][9][10][11][12][13]15]. These very young megalomeronts were found in different host species infected with different H. majoris lineages ( Figure 2D,I,N,S; Figures S3I,R-T,W-AA, S5H,I). In the infections with hPARUS1, very young megalomeronts were found in the kidneys of several individuals collected in different seasons. They were more often observed in autumn than in spring in positive sections, while developing and mature megalomeronts were more often found in spring.
Megalomeronts of H. majoris hPARUS1 were found in three bird species, two of which were found at different times of the year. Parus major were found infected with hPARUS1 twice in spring (one negative for megalomeronts in this study and one positive in [9]), once in summer (positive in this study), and twice in autumn (positive in this study), while Cyanistes caeruleus was sampled twice in the spring (positive in this study) and five times in autumn (two positive and three negative in this study). The individuals sampled in spring contained the largest number of megalomeronts, with up to four organs affected, while fewer individuals were affected by megalomeronts in autumn, with only one to two organs affected (and at a lesser intensity). These seasonal differences in the maturation of the tissue stages might result from a different parasite-load strategy in the host. In Europe, spring is characterized by the active transmission of Haemoproteus parasites [1], and the increase in parasitemia in blood, which is readily visible in blood films. Mature megalomeronts are expected to be more numerous in spring, as merozoites are actively produced and released into circulation. In comparison, parasitemia usually markedly decreases in autumn and is barely present during winter. This seasonality in parasitemia is directly related to exo-erythrocytic development in haemoproteids and should result in less numerous megalomeronts in organs, lessening the burden on the host until the relapse of infection in the following spring [1]. It is still unknown under which stage (unicellular hypnozoite-like parasites or dormant megalomeronts at early development or advanced stage) Haemoproteus parasites are present in their host during winter, as gametocytes are rarely seen in blood films. A small number of young megalomeronts might be enough to initiate exo-erythrocytic development and relapse in spring, but their size and presence might also harm the host as inflammatory reactions are observed. Very young megalomeronts, due to their smaller size, might have a lesser influence on the host, but as their CISH signals were deep purple, active development and growth still occur. It is unknown whether the parasite can slow down its development and persist during winter.
Haemoproteus majoris is a generalist parasite [21] with six lineages differing by 1 to 6 bp (0.2-1.3 %) in the cyt b barcoding sequence. The lineage hCWT4 is closer to hWW2 and hPHYBOR04, with a 1 bp difference (0.2 %), whereas hPARUS1, hPHSIB1, and CCF5 have more differences from the other lineages (differences range from 2 to 6 bp). It seems probable that megalomeronts are the main and probably only tissue stage during H. majoris infections. The morphological similarities of megalomeronts and the sites of their development seem to reflect their genetic distances (hCWT4/hWW2 vs. hCCF5/hPARUS1). This is the first study which analyzed more than two lineages belonging to one Haemoproteus species in regard to their exo-erythrocytic development. It is thus difficult to speculate if the slight differences in morphology and site of development between the H. majoris lineages are due to the generalist features of this parasite or are due to lineagespecific features, or both. It would be interesting to study additional multi-lineage species, such as H. tartakovskyi, H. lanii, and H. balmorali, to explore intraspecific variation of exoerythrocytic development. Only one parasite closely related to H. balmorali, H. attenuatus, has been reported to develop meronts in the lungs of its host [14]. An investigation into the exo-erythrocytic stages of other closely related species, which are genetically closely related, would help to better understand if the genetic distances can predict patterns of the exo-erythrocytic development or if the development is only species-specific.
Differences in parasitemia intensity were small in most sampled birds, with parasitemia intensity mostly around 1% (Table 1). Among individuals infected with the same lineage and showing the same intensity of parasitemia, some contained megalomeronts but others did not (Table S1). Similar results were previously reported during H. pastoris infection in Sturnus vulgaris [13]. Mainly, no trend for megalomeront development was determined among the individuals with parasitemia ranging from 1 to 26%, with the exception of one individual with 10% parasitemia, for which no megalomeronts were found [13]. Parasitemia intensity was not associated either with the number of affected organs by megalomeronts. In H. pastoris, nine organs were reportedly affected by the parasites [13], while for all examined H. majoris lineages and hosts, megalomeronts were found in ten organs (this study, [9,12]).
Finally, two birds with H. majoris infections were found to be co-infected with other Haemoproteus spp., and other exo-erythrocytic stages were found. These exo-erythrocytic stages were meronts of small sizes following the capillaries of the lungs (Figure 3G), and they were similar to meronts found in H. dumbbellus [30] of one individual or bigger and clustered together in a blood vessel of the lungs ( Figure 3H,I) of the second individual. No meronts were previously reported in H. majoris infections, only megalomeronts (this study, [9,12]). Deep purple in situ CISH signals were observed in the meronts in the two individuals, confirming their Haemoproteus identity ( Figure 3G-I inserts), and the microscopic examination of the blood smears identified co-infections in both cases with H. fringillae. It should be noted that the second individual was also infected with other Haemoproteus spp., aside from H. majoris and H. fringillae. It is most likely that these meronts are not of H. majoris but might very well be of H. fringillae. This observation shows the difficulties in the research on the exo-erythrocytic development of haemosporidians during co-infections, which predominate in wildlife and calls for the development of more specific in situ diagnostic probes (e.g., on a species or even lineage level), which would increase the resolution of exo-erythrocytic stages belonging to different parasite species.

Conclusions
This study extends knowledge about the lineage diversity of the widespread H. majoris haemosporidian parasite by adding the lineage hCWT4 to this morphospecies. Only megalomeronts were observed developing in all examined single H. majoris infections, strongly suggesting that meronts might be absent in this parasite. Megalomeronts were also found in mixed infections of several H. majoris lineages, while meronts were only found in cases of mixed infections of H. majoris and other Haemoproteus species. The following findings are worth more attention.
The organs were found to be differently affected by H. majoris megalomeronts, with hPARUS1 megalomeronts most commonly found in the kidneys; hCCF5 in the lungs, intestine, and kidneys; hCWT4 in the gizzard; and hWW2 in the heart and gizzard. Megalomeronts were also found in those organs for the other lineages (but at a lower intensity). The morphology of all reported H. majoris megalomeronts was mostly similar in all lineages, with some minor differences in hCWT4 and hWW2 infections. Very young megalomeronts were found for four of the H. majoris lineages. The parasitemia intensity could not be associated with the presence of megalomeronts in organs. However, megalomeronts were more often found in birds from spring than from autumn, with more developing and mature megalomeronts in spring and more very young stages in autumn. It would be interesting to investigate other generalist and specialist parasite species in different avian hosts across different seasons to better understand the pattern of their development and to make sure how the exo-erythrocytic development depends on the parasite lineage, the host species, and the season. It would also be interesting to see if very young megalomeronts can be found in the host during the winter period, possibly as dormant stages responsible for relapses in spring.

Supplementary Materials:
The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/pathogens12070898/s1. Figure S1: Megalomeronts of Haemoproteus majoris lineage hCCF5 in Fringilla coelebs; Figure S2: Megalomeronts of Haemoproteus majoris lineage hCWT4 in Sylvia communis and Sylvia curruca; Figure S3: Megalomeronts of Haemoproteus majoris lineage hPARUS1 in Cyanistes caeruleus, Parus major, and Parus montanus; Figure S4: Megalomeronts of Haemoproteus majoris lineage hWW2 in Sylvia atricapilla; Figure S5: Megalomeronts of Haemoproteus majoris found in co-infections with several of its lineages; Table S1: Data on all individuals with Haemoproteus majoris infections investigated for exo-erythrocytic stages; Table S2: Data on all individuals with Haemoproteus majoris hPARUS1 infections investigated for exo-erythrocytic stages numbers. Data Availability Statement: The generated sequence data were deposited in the NCBI GenBank database. The parasite voucher preparations (blood and histological preparations) are available at Nature Research Centre, Vilnius, Lithuania, and at the Queensland Museum, Brisbane, Australia, upon request.