From April 2023 to October 2023, during the peak season for ‘Ca. M. haemobos’ infections and tick activity in southern Henan Province (112°38′ ~ 113°24′ E, 33°04′ ~ 33°37′ N), the region is characterized by a diverse topography that includes mountains, hills, and flat or gently rolling plains, situated within the subtropical continental monsoon zone. To address the research question, 19 backyard farms affected by Ca. M. haemobos’ were selected as the study site. Previous research documented infections of ‘Ca. M. haemobos’ in goats, sheep, dogs, and cattle on these farms (14–16). Subsequently, investigations were conducted on cats, parasitic ticks, and ticks found in the environment at these locations. In total, 78 EDTA-anticoagulated blood and serum samples were collected from the femoral vein of the cats. Among the sampled animals, some exhibited clinical signs such as pale oral mucous, conjunctival infection, and hematuria. In addition, all ticks found on the skin surface of each cat were collected, resulting in a total of 284 ticks. These ticks were treated individually, following the methodology established in a previous study (14). Furthermore, questing ticks (n = 356) in the environment were collected using the drag–flag method (38), using a white cotton flannel cloth (1.2 m × 1 m). The flag was systematically dragged over low vegetation (1 ~ 30 cm in height) near the edges of paths and the borders of dense vegetation.

All blood samples underwent an initial examination as follows: the partial whole blood samples were analyzed using an automated hematology analyzer (DF55Vet, Dymind, China), and the results were compared to established reference ranges (red blood cell (RBC): 6.54–12.20; hematocrit (HCT): 30.30–52.30; hemoglobin (HGB): 9.80–16.20) in accordance with the provided instructions and the reference (39). Subsequently, based on the clinical symptoms documented earlier, the cats involved in this study were categorized into two groups. The remaining blood samples from the cats were preserved at −80°C for future analysis.

All adult tick samples were initially identified using morphological and taxonomic identification methods (40). Hard ticks and soft ticks were differentiated at the family level based on the position of the anal groove. Subsequently, at the genus level, identification was achieved by examining the morphology of the gnathosoma. Within the same genus, ticks exhibit considerable morphological similarity, necessitating the identification of distinguishing features such as the shape, color, and patterns of the scutum; the configuration of the spiracular plates; the presence of stripes on the tarsus; and the characteristics of the pulvilli. Finally, sexual dimorphism was assessed by comparing the size of the scutum between males and females (41).

The blood samples obtained from cats were subjected to DNA extraction using the EasyPure Blood Genomic DNA Kit (TransGen Biotech, China) according to the manufacturer’s instructions. Each tick sample was homogenized in 1 mL of phosphate-buffered saline, then placed on sterile filter paper to dry, and further put into tubes prefilled with ceramic beads (MagNA Lyser Green Beads, Roche, USA). A volume of 400 μL of PBS was added to each tube, and incubation was carried out for 5 h. After the tick bodies and scuta had softened, the tick tissues were homogenized at 7,000 rpm for 90 s by the MagNA Lyser Instrument (Roche, USA), after which 200 μL of each homogenate underwent DNA extraction utilizing the Universal Genomic DNA Kit (TIANGEN, China). The extracted genomic DNA was eluted in DEPC-treated water and subsequently used as a template in PCR reactions. For molecular identification, the conserved 12S rRNA gene of tick species was selected for amplification using primers T1B and T2A (42). In alignment with previous research (43), a specific pair of primers (H-MYC-F/R) was selected to identify potential hemotropic mycoplasmas. These primers have previously demonstrated efficacy in amplifying the partial 16S rRNA gene of various species, including M. haemofelis, Ca. M. haemominutum, M. haemocanis, Ca. M. haematoparvum, Mycoplasma ovis, Candidatus Mycoplasma haemovis, M. wenyonii, and ‘Ca. M. haemobos’ (14–16, 43). Furthermore, DNA from a strain of M. wenyonii served as a positive control, while DEPC-treated water was utilized as a negative control in all PCR reactions. The PCR procedures were conducted as described in previous studies (43, 44). In addition, nested primers Rick-out/Rick-in for R. felis (45), Apla-sense and ECB primers for Anaplasma (46), HAM-1F and HPF-2R primers for Hepatozoon (47), and BTH 18S 1st F/R and BTH 18S 2nd F/R primers for Babesia and Theileria (48) were used to detect potential tick-borne pathogens associated with anemia in cats. The extracted blood and tick DNA were reconstituted in 50 μL of double-distilled water, and the quantity and quality of DNA were assessed using a spectrophotometer (UV1000, Techcomp, China). The PCR amplification reactions were performed using an EasyTaq® PCR SuperMix kit (TransGen, Beijing, China) in a total reaction of 20 μL, which included 10 μL of 2× EasyTaq® PCR SuperMix, 0.4 μM of each primer, and 20 ng template DNA. The amplification conditions were as follows: predenaturation at 94°C for 5 min, followed by 30 cycles of 30 s at 94°C, 30 s at the appropriate annealing temperature determined by the specific primers, 2 min at 72°C, and a final extension at 72°C for 10 min. The primers and annealing temperature are shown in Table 1.

Following the initial molecular screening, all mycoplasma-positive amplicons were purified and sequenced, as previously described (14). The resulting sequences were then compared to relevant sequences available in the NCBI databases utilizing a BLAST search. Subsequently, longer fragments (1,393 bp) of the 16S rRNA gene were amplified from all samples that tested positive for ‘Ca. M. haemobos’ using the MHBforw and MHBrev primers (44). These fragments were also purified and sequenced following the aforementioned protocol.

The sequences of the long 16S rRNA gene were compared and aligned using the CLUSTALW program, incorporating strains isolated from cattle, buffalo, sheep, ticks, and dogs from various regions worldwide. Subsequently, a phylogenetic analysis was performed using MEGA6 (49), employing the neighbor-joining criterion and the Kimura two-parameter model (11, 14, 22, 44, 50). The robustness of the hypothesis was tested using bootstrap analysis with 1,000 replicates.

In this study, cats were divided into two groups: Group 1 comprised 57 anemic cats that exhibited clinical signs and had hematological values below the normal reference ranges, while Group 2 consisted of 21 healthy cats with hematological values within the normal reference ranges. Statistical analysis for significant differences was performed between groups using SPSS 17 software (IBM, USA) on the chi-square test. p-value<0.05 was considered the threshold for statistical significance.

In total, 640 ticks were collected from both cats and the environment. This collection included 284 adult ticks obtained from cats, along with 336 adult ticks and 20 nymph ticks collected from the environment. Following a thorough examination using morphological and taxonomic keys, a total of 119 male adult ticks and 501 female adult ticks were identified. Subsequent molecular identification through sequencing of the 12S rRNA gene revealed that these adult ticks comprised three species from two genera within the family Ixodidae: 407 H. longicornis (199 and 208 collected from cats and the environment, respectively), 155 R. microplus (58 and 97 collected from cats and the environment, respectively), and 78 Rhipicephalus sanguineus sensu lato (R. sanguineus sl) (27 collected from cats and 51 from the environment). Furthermore, both engorged and unengorged ticks were recorded among the three tick species collected from cats and the environment. Detailed information regarding the adult tick species, their source, sexes, life stages, and numbers is presented in Table 2.

As shown in Table 3, hemotropic mycoplasma infection rates were 54.4% (31 out of 57) and 14.3% (3 out of 21) in blood samples in Group 1 and Group 2, respectively; 55.4% (102 out of 184), 26.7% (4 out of 15), and 29.3% (61 out of 208) in H. longicornis in Group 1, Group 2, and the environment, respectively; 64.0% (32 out of 50), 25.0% (2 out of 8), and 35.0% (34 out of 97) in R. microplus in Group 1, Group 2, and the environment, respectively; and 68.2% (15 out of 22), 20.0% (1 out of 5), and 17.6% (9 out of 51) in R. sanguineus sl in Group 1, Group 2, and the environment, respectively. Significant differences in mycoplasma infection rates were observed between Group 1 and Group 2 in the blood samples (p = 0.002). All hemotropic mycoplasma-positive amplicons were sequenced and aligned using the BLAST search tool in GenBank. The results indicated the prevalence rates of ‘Ca. M. haemominutum’, M. haemofelis, and ‘Ca. M. haemobos’ in the blood samples, as well as in samples from H. longicornis, R. microplus, and R. sanguineus sl. In addition, screening for other pathogens revealed the presence of R. felis in cats and three tick species, with no other pathogens detected. Moreover, two types of co-infections (‘Ca. M. haemominutum’ + R. felis and ‘Ca. M. haemobos’ + R. felis) were identified in anemic cats, H. longicornis, and R. sanguineus sl.

Initially, the infection rates of ‘Ca. M. haemominutum’ were analyzed. ‘Ca. M. haemominutum’ infection rates were 31.6% (18 out of 57) and 14.3% (3 out of 21) in the blood samples in Group 1 and Group 2, respectively, and no significant differences (p = 0.127) were detected. In contrast, when comparing the infection rates in H. longicornis from the environment group (14.4%, 30 out of 208), a significant difference (p < 0.001) was noted in Group 1 (37.0%, 68 out of 184), whereas no significant difference (p = 1.000) was observed in Group 2 (13.3%, 2 out of 15). Furthermore, when compared to the infection rate of R. microplus in the environment group (13.4%, 13 out of 97), the infection rates of R. microplus in both Group 1 (22.0%, 11 out of 50) and Group 2 (0%, 0 out of 8) did not exhibit significant differences (p = 0.095 and p = 0.595, respectively). In addition, when comparing the infection rates of R. sanguineus sl in the environment group (7.8%, 4 out of 51), a significant difference (p = 0.023) was found in Group 1 (31.8%, 7 out of 22), whereas no significant difference (p = 1.000) was observed in Group 2 (0%, 0 out of 5).

In the second analysis, the infection rates of ‘Ca. M. haemobos’ were compared across different groups. ‘Ca. M. haemobos’ infection rates were 15.8% (9 out of 57) and 0% (0 out of 21) in the blood samples in Group 1 and Group 2, respectively, and no statistically significant differences (p = 0.124) were observed. Furthermore, when compared to the infection rate of H. longicornis in the environment group (8.7%, 18 out of 208), no significant differences (p = 0.281 and p = 1.000, respectively) were observed in Group 1 (12.0%, 22 out of 184) and Group 2 (6.7%, 1 out of 15). Similarly, the infection rates of R. microplus in Group 1 (24.0%, 12 out of 50) and Group 2 (25.0%, 2 out of 8) showed no significant differences (p = 0.638 and p = 1.000, respectively) compared to the environment group (20.6%, 20 out of 97). In addition, when comparing the infection rate of R. sanguineus sl in the environment group (9.8%, 5 out of 51), no significant differences (p = 0.119 and p = 0.445, respectively) were found in Group 1 (22.7%, 5 out of 22) and Group 2 (20.0%, 1 out of 5). Notably, ‘Ca. M. haemobos’-positive samples, which included one H. longicornis, two R. microplus, and one R. sanguineus sl, were collected from negative cats in Group 2.

M. haemofelis infection rates were 7.0% (4 out of 57) and 0% (0 out of 21) in the blood samples in Group 1 and Group 2, respectively; 6.5% (12 out of 184), 6.7% (1 out of 15), and 6.3% (13 out of 208) in H. longicornis in Group 1, Group 2, and the environment, respectively; 16.0% (8 out of 50), 0% (0 out of 8), and 1.0% (1 out of 97) in R. microplus in Group 1, Group 2, and environment, respectively; and 9.1% (2 out of 22), 0% (0 out of 5), and 0% (0 out of 51) in R. sanguineus sl in Group 1, Group 2, and the environment, respectively. Additional details are provided in Table 3. It is important to highlight that ‘Ca. M. haemominutum’, M. haemofelis, ‘Ca. M. haemobos’, and R. felis were detected in both engorged and un-engorged ticks across all three species, as presented in Table 3.

For further analysis, the longer amplicons from all positive samples of ‘Ca. M. haemobos’ were amplified, recovered, and sequenced. A total of six different sequences were identified among these positive samples. Six strains were selected as representatives for detailed analysis: the HN1807 strain (GenBank Accession Number MH388476) and the HN1841 strain (GenBank Accession Number MH388480), which have been previously described in sheep and R. microplus (14); the HN1921 strain (GenBank Accession Number MW463059) and the HN1933 strain (GenBank Accession Number MW463060), which have been previously described in dogs (16); and two novel sequence types: HN2318 (GenBank Accession Number OR818448) and HN2340 (GenBank Accession Number OR818449). In addition, new sequences were observed in those three tick species. Specifically, the following ‘Ca. M. haemobos’ strains were included in the analysis: clones 307 (EF616467) and 311 (EF616468) (cattle, Switzerland); no. 18 (EU367965) (cattle, Japan); C115, C080, C061, and B001 (MG948630) (cattle, Cuba); I924712 (KT985638) (cattle, Malaysia); HN1804 (MH388478) (tick, China); HN1807 (MH388476) (sheep, China); HN1823 (MH388475) (goat, China); HN1921 (MW463059) (dog, China); and CMboTWN03 (KJ883516) and CMboTWN04 (KJ883517) (cat, China). Furthermore, for the purpose of phylogenetic analysis, the following isolates were included: M. haemofelis isolates Australian no. 2 (AY150977) (Australia) and UK no. 5 (AY150984) (United Kingdom); M. coccoides (AY171918) (United Kingdom, laboratory mouse); Ca. M. turicensis 94–100 (DQ825454) (Tanzania, lion) and B3 (DQ464423) (Australia, cat); M. wenyonii Massachusetts (CP003703) (USA, cattle) and CGXD (EF221880) (China); and M. pneumoniae NBRC 14401 (NR 113659) (Japan, human). The two new representative isolates were compared to other ‘Ca. M. haemobos’ strains available in GenBank, revealing that the 16S rRNA sequences of the two new isolates exhibited 99.34–99.57% identity with those of other isolates. A phylogenetic tree was inferred based on the 16S rRNA sequence (Figure 1), indicating that the two new representative isolates clustered within the species of ‘Ca. M. haemobos’. Moreover, the two new isolates identified in the present study were found to be most closely related to an isolate (China-1) (EF424082) from buffalo in China, while being most distantly related to the isolate from Switzerland (clone 307).