Identification of Piroplasmosis in Patients

We conducted a retrospective investigation among participants experiencing unexplained fever or anemia across 10 counties in Yunnan Province, China, during May 2017–June 2020. Demographic data, medical history, and epidemiologic exposure history had been collected through a structured questionnaire. We retrieved data on clinical manifestations, underlying medical conditions, laboratory test results, treatments, and outcomes from medical records; 2 investigators cross-validated the data. The patients’ blood samples were collected at various time points during their hospital stay; a portion of each patient’s blood samples were immediately processed for blood smear preparation, and the residual blood was stored at −80°C until nucleic acid extraction for batch PCR amplification and downstream analysis. The Ethics Committee of the Yunnan Provincial Institute of Endemic Disease Control approved the study (approval no. 2016-005), which was conducted in accordance with medical research regulations in China. All participants provided written informed consent before their inclusion in the study.

Source Tracing Investigation

As part of the study, we conducted retrospective testing on ticks, livestock, and small mammals from areas surrounding the participants. Wild small mammals were captured using snap traps, and aseptic tissue samples, including liver and spleen, were collected and stored at −80°C for subsequent analysis. We taxonomically identified the captured animals to the species level on the basis of external morphology, coloration, measurements, and dental characteristics. We collected whole-blood samples from livestock via jugular vein puncture using EDTA anticoagulant tubes and stored them at −20°C until DNA extraction. We manually removed ticks from livestock and collected host-seeking ticks by flag-sweeping vegetation at the same sampling sites. An entomologist identified tick species. We preserved tick samples at −80°C before DNA extraction.

PCR Amplification and Sequencing

We extracted DNA from human blood, livestock blood, small mammal tissues, and tick samples using the DNeasy Blood & Tissue Kit (QIAGEN, https://www.qiagen.com) in accordance with manufacturer’s instructions. To detect Theileria and Babesia parasites, we performed nested PCR targeting the 18S rRNA gene using outer primers Piro0F/Piro6R and inner primers Piro1F/Piro5.5R (13,14), followed by agarose gel electrophoresis and Sanger sequencing. We amplified additional genetic loci, including the 5.8S rRNA (303 bp), internal transcribed spacer region (1,300 bp), P32 immunodominant protein gene (875 bp), and cytochrome oxidase subunit I (1,200 bp). We conducted concurrent testing for other potential infections with Rickettsia spp., Borrelia burgdorferi, B. recurrentis, and Bartonella spp., which could potentially be transmitted by similar transmission routes or cause similar symptoms, to exclude differential diagnoses (Appendix 1 Table 1).

Phylogenetic Analysis

We performed phylogenetic analysis using sequences assembled with the CLC Main Workbench version5.5 (QIAGEN). We conducted comparative analysis against sequences in GenBank using BLAST (https://blast.ncbi.nlm.nih.gov). We conducted phylogenetic analysis of all sequences using MEGA version 11.0 software (https://www.megasoftware.net). We constructed phylogenetic trees using the neighbor-joining method with the p-distance model based on 1,000 bootstrap replicates.

Morphologic and Serologic Testing

We stained thin peripheral blood smears collected from the participants with Giemsa and examined under a light microscope (Olympus, https://evidentscientific.com). We used the recombinant T. uilenbergi immunodominant protein (rTuIP) as the diagnostic antigen for Western blot analysis (15,16). Anhui Global Gene Technology Company (Chuzhou, China) conducted protein expression and purification using the pQE31 vector as the expression vector and BL21 (DE3) as the expression host. Tengchong People’s Hospital (Tengchong County, China) provided serum samples collected for rTuIP antibody detection from some patients with suspected protozoan infections and used serum samples from healthy persons as negative controls.

We separated recombinantly expressed TuIP protein (20 μg) using 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred it onto a polyvinylidene difluoride membrane (0.45 μm pore size). To block nonspecific binding, we incubated the membrane with 5% skimmed milk powder in Tris-buffered saline containing Tween 20 for 2 hours. We used patient serum samples diluted at various ratios (1:25–1:800) as primary antibodies; serum samples from healthy persons diluted at 1:100 served as negative controls. We then incubated the membrane overnight at 4°C. We applied horseradish peroxidase–labeled human IgG secondary antibody, diluted at 1:5,000, for a 1-hour incubation at room temperature. After washes in Tris-buffered saline containing Tween 20, we detected the signal using an enhanced chemiluminescence developing reagent.

Identification of Piroplasmosis in Participants

Figure 1

Figure 1. Photomicrographs from peripheral blood smears of a patient (patient 2) who tested positive for Theileria luwenshuniparasite in study of piroplasmorida in humans, southwest China, May 2017–June 2020. Arrows...

We screened a total of 1,721 participants from 10 counties in Yunnan Province for protozoan infections; 1,362 participants were experiencing fever, and 359 had anemia, (Appendix 1 Table 2). Among those, we identified 18 inpatients and 1 outpatient to have suspected protozoan infections by 18S rRNA sequencing of blood samples; amplification for additional T. luwenshuni genetic loci was not successful. Meanwhile, molecular testing for Rickettsia spp., B. burgdorferi sensu lato, B. recurrentis, and Bartonella spp. yielded uniformly negative results. We examined peripheral blood smears from the 19 suspected cases for intraerythrocytic parasites using Giemsa staining; patient 2 tested positive for piroplasmosis parasites (Figure 1). No parasites were detected in the blood smears of the remaining patients.

Figure 2

Figure 2. Phylogenetic analysis of 18s rRNA (1,600-bp) gene sequences of Theileria and Babesiaspp. isolates in study of piroplasmorida in humans, southwest China, May 2017–June 2020. Colored shading...

We successfully amplified and sequenced the 18S rRNA gene of T. luwenshuni parasites from 13 blood samples, yielding sequences ≈1,600 bp long. Those sequences exhibited a similarity range of 99.28%–100% among the samples; we deposited them into GenBank (accession nos. PQ720759–71). Similarity analysis indicated that the T. luwenshuni samples shared 99.69% identity with a reference sequence isolated from a Haemaphysalis longicornis tick (GenBank accession no. OQ540587.1) and 98.60% identity with a T. luwenshuni isolate from goats (GenBank accession no. OQ134905.1), both collected in Shandong Province, China. Phylogenetic analysis further showed that the sequences clustered within the same monophyletic branch (Figure 2; Appendix 1 Table 3).

We successfully amplified and sequenced the 18S rRNA gene of Babesia spp. from 6 blood samples, yielding sequences ≈1,600 bp long. We submitted those sequences to GenBank (accession nos. PQ720772–5, PQ722837, PQ722838). Phylogenetic analysis demonstrated that samples DQ-H1-33, TC-H4-76, and TC-H4-54 clustered with multiple B. microti strains from reference sequences (GenBank accession nos. KF410825.1 and AB050732.1), exhibiting a high genetic similarity (Figure 2). In addition, samples TC-H4-11 and TC-H4-56 clustered with Babesia Tianzhu–like subgroups that were identified in white yaks in China (14). Meanwhile, sample TC-H4-53 clustered with another subgroup, which suggested that TC-H4-53 may represent an uncharacterized or novel Babesia variant or strain (Figure 2).

Figure 3

Figure 3. Western blot analysis of the target protein (55-kDa) expression in the serum samples from 5 patients with suspected protozoan infection in study of piroplasmorida in humans, southwest China, May 2017–June...

Western blot analysis confirmed the diagnostic utility of the rTuIP protein for detecting T. luwenshuni infections. Of 11 serum samples from participants with suspected hemoprotozoan infections, 6 samples (from 5 patients) exhibited characteristic rTuIP-specific bands at a 1:25 dilution (Figure 3). Of note, longitudinal samples from patient 1 (P1-1 and P1-3) both exhibited positive reactivity. Control samples from healthy participants showed no target protein–specific bands.

Analysis of longitudinal antibody responses also revealed dynamic seroconversion patterns over time (Appendix 2 Figure, panel A). Sample P1-1 exhibited faint reactivity at 1:25 dilution but was negative at >1:50. In contrast, subsequent samples from the same patient (P1-2 and P1-3) displayed strong immunoreactivity across all consecutive doubling tested dilutions (1:50–1:800) (Appendix 2 Figure, panel A), indicating the maturation of antibody titers. We saw similar consistently high-titer responses in patient 2 at 2 timepoints, with clear bands visible from 1:50–1:800 dilutions (Appendix 2 Figure, panel B).

Epidemiologic and Clinical Characteristics of the Patients

Among the 13 patients with suspected T. luwenshuni infection (Appendix 1 Table 4), the median age was 40 (interquartile range [IQR] 32.5–68.0) years; 10 (77%) patients were male and 13 (23%) female. The median hospital stay was 13 (IQR 7.5–17.0) days. All patients were engaged in farming. Underlying conditions noted at admission included renal failure in 5 (38%) patients, upper gastrointestinal bleeding in 5 (38%) patients, and a history of cancer in 3 (23%) patients; 1 patient reported prior blood transfusion history. The primary clinical manifestations were anorexia (8 [62%] patients), listlessness (8 [62%] patients), malaise (5 [38%] patients), and melena (4 [31%] patients). Other reported symptoms included vomiting (4 [31%] patients), dizziness (3 [23%] patients), and palpitations (3 [23%] patients); cough, jaundice, splenomegaly, and weight loss were each observed in 1 patient (8%). One patient reported neurologic symptoms (Appendix 1 Table 4). Laboratory findings revealed anemia in all 13 patients and leukocytosis in 2 (15%) patients. Elevated serum levels of aspartate aminotransferase, alanine aminotransferase, or gamma-glutamyl transferase were present in 1 patient (8%). All patients received general supportive and symptomatic treatment; 4 (31%) patients underwent blood transfusions, and 5 (38%) patients required hemodialysis (Table 1).

Among the 5 patients with suspected Babesia infections (Appendix 1 Table 5), median age was 56 years (IQR 25.5–59.5 years); 4 (80%) patients were male and 1 (20%) female. Similar to the T. luwenshuni cohort, all patients were engaged in farming. Underlying conditions were renal failure in 1 patient (20%) and a history of upper gastrointestinal bleeding in 3 patients (60%). Median hospitalization duration was relatively short at 10 days (IQR 5.5–10.5 days). Two patients (40%) had a history of blood transfusion; 1 (20%) patient had a history of gastric hemorrhage or hypertension. The most frequently reported symptoms were nonspecific. Four (80%) patients experienced dizziness and malaise, 3 (60%) patients melena, and 2 (40%) listlessness. Bloating, chest discomfort, palpitations, and dyspnea were all present in 1 (20%) patient. Of note, none of the patients in this group reported vomiting, jaundice, or splenomegaly. Laboratory tests revealed anemia in all patients and leucopenia in 1 (20%) patient. Elevated serum levels of aspartate aminotransferase, alanine aminotransferase, or gamma-glutamyl transferase were not detected in any case. In terms of treatment, 3 (60%) patients received blood transfusions, and 1 patient (20%) required hemodialysis (Table 1).

Traceability Survey

Figure 4

Figure 4. Phylogenetic analysis of 18s rRNA (1,600-bp) gene sequences of Theileria and Babesiaspp. isolates in study of piroplasmorida in humans, southwest China, May 2017–June 2020. Tree on...

Figure 5

Figure 5. Characterization of host and vector associations determined from phylogenetic analysis of Theileria and Babesia spp. isolates in study of piroplasmorida in humans, southwest China, May 2017–June 2020....

We performed PCR analyses on 2,100 livestock, 1,530 wild small mammals, and 1,516 ticks collected from 10 counties in Yunnan Province to detect Piroplasmorida infections (Table 2). We used the same protocol previously optimized for human diagnostics. Among the livestock samples, 147 (7.00%) tested positive. We constructed a phylogenetic tree using 18S rRNA sequences (≈1,600 bp) to classify the positive samples into 5 species: B. microti, Babesia sp. Tianzhu-like, B. bigemina, B. bovis, and T. luwenshuni (Figures 4, 5). We detected 1 strain of B. microti and 2 strains of T. luwenshuni parasites in livestock from Tengchong County. The overall positivity rate for small mammals was 2.42%. We identified B. microti as the predominant species. Phylogenetic analysis identified a variant, tentatively named Babesia sp. YN-2, in small mammals (Figure 4). Among tick samples, 37 (2.44%) tested positive for Piroplasmorida; detected species including B. microti, B. bigemina, Babesia sp. Tianzhu-like, B. bovis, Babesia sp., Colpoda sp., and T. luwenshuni. Phylogenetic analysis further classified the novel Babesia sp. into 2 clades: Babesia sp. YN-2 and Babesia sp. YN-3.

Genetic analysis of the Babesia sp. Tianzhu-like isolates identified in patients, livestock, and ticks yielded 2 major clades with high homology (96.11%–98.67%) but distinct variations of 21–62 bp. We tentatively designated those clades as Babesia sp. isolate Tianzhu-like 1 and Babesia sp. isolate Tianzhu-like 2. The phylogenetic tree depicted the relationships among Babesia and Theileria species collected from various hosts and vectors. Of note, we identified B. microti in multiple host and vector types, indicating its high adaptability across diverse environments (Figure 5). In addition, we detected the novel Babesia sp. YN-2 in both small mammals and ticks, suggesting the existence of a potential transmission cycle between wildlife and arthropod vectors. The detection of Babesia sp. isolate Tianzhu-like parasites across humans, livestock, and ticks highlights its broad host range and geographic distribution. Similarly, the widespread occurrence of T. luwenshuni parasites in humans, livestock, and ticks underscores the species’ potential for transmission among multiple hosts.