Apparent prevalence of hemotropic mycoplasma in dairy calves and replacement heifers on Michigan farms

The bovine hemoplasmas include Mycoplasma wen-yonii and Candidatus Mycoplasma hemobos which are increasingly recognized as infecting cattle throughout the world. Infection with hemotropic mycoplasma has been reported to be widespread in mature dairy cows, but little is known about prevalence in calves and heifers. The objective of this study was to investigate the prevalence and dynamics of infection with M. wenyonii and C.M. hemobos infections in calves and replacement heifers on Michigan dairy farms and assess potential associations between infection status and hematological values. The study was designed as a prospective cross-sectional study with a longitudinal component. A convenience sample of 11 farms agreed to participate and were visited twice between March and September 2022. During the first farm visit, researchers collected blood samples from up to 94 animals per farm distributed among newborn and pre-weaned calves (n ≤ 31), weaned calves (n = 21), pre-breeding heifers (n = 21), and pregnant heifers (n = 21). During the first visit, blood samples (n = 174) were also collected from a convenience sample of mature cows to confirm the herd infection status. The same calves and heifers were sampled again about 95 d (±3.0) later. During the first visit, blood samples were collected from 797 calves and replacement heifers, while 675 samples were collected during the second visit due to inability to locate some animals. Detection of M. wenyonii and C. M. hemobos were based on results of real time PCR. The hematocrit was determined using microcentrifugation and the concentration of leukocytes using an automated cell counter. In all herds, most mature cows that were sampled tested positive for infection. The within herd apparent prevalence of hemoplasma in calves and replacement heifers was 100% for both M. wenyonii and C. M. hemobos . The apparent prevalence of hemoplasma in youngstock was associated with age. In calves that were

Although hemoplasmas have been reported to infect cattle in several countries (dos Santos et al., 2012;Girotto et al.2012;Tatsukawa et al. 2021;Díaz-Sánchez et al., 2019), changes in the nomenclature and diagnostic methods have contributed toa the lack of knowledge about the epidemiology and pathogenesis of these organisms.In recent years, there has been growing recognition of the scope of hemoplasma infections in cattle, particularly in countries in the northern hemisphere (Nishizawa et al., 2010;Tagawa et al., 2010Tagawa et al., , 2012Tagawa et al., , 2013;;Fujihara et al., 2011;Schambow et al., 2021).These studies have documented widespread prevalence of infection and highlighted their significance as an emerging pathogen in cattle populations.While previous research has focused on the prevalence of these infections in adult cattle, limited information is available about infection in calves and replacement heifers.In a recent study that tested adult cows in 82 herds located in Wisconsin or Michigan, they reported 100% herdlevel apparent prevalence for both M. wenyonii and C. M. hemobos (Schambow et al., 2021).The prevalence of infection was >70% in first lactation cows, which suggested that infection occurred before initial calving (Schambow et al., 2021).Unfortunately, no studies have been conducted to investigate the rate of hemoplasma infection in calves and replacement heifers.Given the limited understanding of hemoplasma epidemiology in cattle, understanding the rate of infection in calves can provide insights on the timing of infection, which is essential for the development of effective preventive measures.
The objective of this study was to determine the prevalence and dynamics of infection of with M. wenyonii and C.M. hemobos in calves and replacement heifers on Michigan dairy farms and assess potential associations between infection status and hematological values.We hypothesized that infections occurred before first calving, varied with age, and would impact some hematological values.

Farm Eligibility and Recruitment
Based on 100% herd-level prevalence and > 75% within-herd apparent prevalence of adult cows reported in a previous study of herds in this region (Schambow, et al. 2021), a convenience sample of herds was used in this study.Herds that previously participated in a prevalence study of adult cows in Michigan (n = 18) (Schambow et al., 2021) were mailed a recruitment letter and then contacted by phone to invite them to participate in this study.Additional herds in Michigan were recruited through announcements at regional conferences.All herds volunteered to participate and were chosen based on their location in Mid-Michigan, as well as having sufficient numbers of animals, including pre-weaned calves to pregnant heifers.Farmers that expressed interest in the study were contacted by phone to schedule a sampling visit.

Animal Eligibility and Enrollment Criteria
All calves and replacement heifers between 1 d old and 27 mo old were eligible to be tested.To detect at least 5% prevalence of infection with 95% confidence we used a standard formula to estimate sampling required for detection of disease (Dohoo et al., 2009) and estimated that we needed to sample ≥15 calves or heifers within each age group.
Based on the sample size estimate, we planned to collect blood samples from up to 94 calves and replacement heifers per farm distributed as: newborn and preweaned calves < 60 d of age (n < 31), weaned calves (2-8 mo of age, n = 21), pre-breeding heifers (9-12 mo of age; n = 21), and pregnant heifers (>12 mo of age; n = 21).When herds contained less than our target number of animals in an age group, all animals in that group were sampled.Based on logistical challenges of identification and restraint of group housed replacement animals, when greater than the target number of animals were present in a group, convenience sampling de Souza Ferreira et al.: Apparent prevalence of hemotropic… was used to select animals.While our primary objective was to estimate prevalence in calves and replacement heifers, to verify that all enrolled farms contained infected mature cows, during the first farm visit blood samples were also collected from a convenience sample of approximately 15 mature cows, regardless of age or parity and regardless of herd-size.

Sample Collection and Questionnaire
Each farm was visited twice.The first visits took place between March and June 2022 and second visits occurred between June and September 2022.In Michigan, during the first visit period (March to June), the climate typically experiences a transition from winter to spring, with temperatures ranging from cool to mild.In contrast, the second visit period (June to September) corresponds to summer months, characterized by warmer temperatures.During the first visit, researchers collected 2, 5 mL vials of blood into EDTA tubes using venipuncture of the jugular veins or coccygeal vessels.A unique ear tag was attached to each animal to facilitate identification during the second farm visit and demographic data (age, breed, and housing) was collected.During the second visit, a brief questionnaire (available from the authors) was given to an owner or manager of each farm to collect information about housing, husbandry, vaccination protocols, treatment protocols, and reproductive management.During the second farm visit, blood samples were collected from animals that had been tested previously and were easily located.All blood samples were promptly cooled to 4°C.One of the duplicate samples was sent to the Michigan State University's Veterinary Diagnostic Laboratory (located in Lansing, MI) to undergo real-time PCR (rt-PCR) testing, while the other was used to determine hematocrit and leukocyte differential count in the Ruegg Laboratory in the College of Veterinary Medicine at Michigan State University.

Real-time PCR Detection (rt-PCR)
Whole blood in evacuated tubes containing EDTA were stored at 4°C for 24 to 48 h before DNA extraction.Immediately before extraction, the tubes of blood were inverted several times to mix.A magnetic bead assisted DNA extraction method was performed using the KingFisher Flex Purification System (Thermo Fisher Scientific, Waltham MA) with the MagMAX Core Nucleic Acid Purification Kit (Thermo Fisher Scientific, Waltham, MA), following the manufacturer's instructions.Briefly, 200 µL whole blood from a sample was added to a well of King Fisher Flex Deep Well 96 Plate which contained 20 ul of MagMA CORE magnetic bead solution and 10 ul of MagMAX CORE proteinase.This was mixed gently by pipetting up and down several times and was followed by a 2-min incubation at room temperature.Then 700 µL of a 1:1 mixture of MagMAX Core Lysis Solution with MagMAX CORE Binding Solution was added to the well.The plate of blood samples in extraction reagents were loaded onto the King Fisher Flex platform that had been preloaded with a Pharma KingFisher Flex 96 Deep-Well Tip Comb plate, appropriate plates of wash solution, and a plate of elution solution.The wash and elution solutions were supplied in the kit.The MagMAX Core Flex program was used for automated DNA extraction and the extracted DNA was eluted into 90 µL of elution solution.Plates of eluted DNA were stored frozen at −20° until use.
The PCR reaction mixture was 10 µL of 2X PerfeCTa qPCR ToughMix Low Rox (Quanta Bio, Beverly, MA), 400 nmol each of forward and reverse PCR primers, and 250 nmol of hydrolysis probe.Molecular grade water was added to the reaction mix to bring the reaction mix volume to 18 µL.Finally, 2 µL of extracted sample DNA or positive control DNA was added to the reaction mixture.The negative control was 2 µL of molecular grade water added to the reaction mixture.To detect and quantitate copies of DNA from M. wenyonii or C. M. hemobos, real-time qPCR assays specific to each organism were used.The PCR primers and probes used targeted the 16S ribosomal RNA gene.A common set of PCR primers was used for both M. wenyonii and C. M. hemobos.The forward primer, 5′-GAAAGYCTGATGGAGCAATA-3′, had a predicted melting temperature (Tm) range of 56/59°C and the reverse primer, 5′-SCTTTACGCCCAATAAATC-3′, had a predicted Tm range of 55/56°C.The hydrolysis probe for C. M. hemobos, FAM-TGAGGTACT/ZEN/ ATCAGTTGTTATCCCTC-3IABkFq), had a predicted Tm of 65°C.The hydrolysis probe for M. wenyonii, JOEN-CGCGCCTTG/ZEN/ATGGTACTAATTGA-3IABkFq, had a Tm of 66°C.The PCR primers and probes were obtained from Integrated DNA Technologies, Coralville, IA.Reaction conditions were optimized using a temperature gradient to determine optimal annealing temperature(s) for the reaction.
The PCR reaction used 2x PerfeCTa qPCR Tough-Mix Low Rox (Quanta Bio, Beverly, MA).The positive controls for the PCR assays were separate synthetic DNAs that included the PCR primers and probe for M. wenyonii or for C. M. hemobos (Integrated DNA Technologies, Coralville IA).Those positive controls were included in duplicates of 10-fold dilutions of known concentration of synthetic DNA to produce a standard curve for quantitation of organism DNA copy number.Two negative controls included for each assay consisted de Souza Ferreira et al.: Apparent prevalence of hemotropic… of reagent mix plus molecular grade water.The PCR assays for detection of M. wenyonii or C. M. hemobos consisted of a single initial polymerase enzyme activation step of 95°C for 10 min, followed by 40 cycles of denaturation at 95°C for 15 s and annealing/extension at 55°C for 1 min.The target amplification was monitored in real time using an ABI 7500 Fast Real Time PCR System (Applied Biosystems, Foster City, CA) and software that was supplied by the manufacturer.Threshold crossing (Ct value) ranges were provided to interpret the PCR results, classifying positive as Ct <35.4,suspect as Ct 35.5 -37.0, or negative as Ct >37.0.To reduce risk of cross-contamination during the PCR essays, reagent preparation, nucleic acid extraction, addition of sample DNA to the PCR reactions tubes, and the PCR essay all were done in separate rooms with dedicated supplies and laboratory gowns.
Samples falling within the suspect range were confirmed as positive or negative using an in-house diagnostic gel-based PCR assay targeting the 16S ribosomal RNA gene.The gel-based PCR primers were minor modification of the 3-prime ends of primers previously described (Jensen et al., 2001).The forward PCR primer for the assay was 5′-ACGAAAGTCTGATG-GAGCAATAC-3′ and the reverse PCR primer was 5′-ACGCCCAATAAATCCGRATAATG-3′.The reaction mixture concentration was 12.5 µL AmpliTaq Gold 360 Master Mix (Thermo Fisher Scientific, Waltham MA), 400 nmol each for the forward and reverse PCR primer, and molecular biology grade water added to a final volume of 18 µL.A 2 µL aliquot of either, sample DNA, positive control DNA, or negative control molecular biology grade water were added to complete the reaction mixture.The positive control DNA was the same as that used for the qPCR assay.The reaction conditions were 1 cycle of 95°C for 5 min then 40 cycles of 56°C for 30 s and 72°C for 30 s, followed by a final extension step of 72°C for 5 min.The PCR product was electrophoresed through a 1.5% agarose gel and visualized with ethidium bromide staining.Detection of a PCR amplicons of 170 bp and/or 193 bp indicated the sample was positive.

Hematological Analysis
The hematocrit (PCV) of each blood sample was determined using microcentrifugation (Hawksley and Sons, Ltd., Sussex, UK) at 18,600 × g for 5 min.The hematocrit value was measured using a micro-capillary reader (Damon/IEC Division, Needham Heights, MA, USA).To ensure precision and accuracy, triplicate capillary tubes were used for each blood sample, and the mean value was calculated.The leukocyte differential (count of eosinophils (EC), lymphocytes, monocytes (LC), neutrophils (NC), and total leukocytes (TLC) was measured using an automated cell counter (QScout, Advanced Animal Diagnostics, Morrisville, NC) according to the manufacturer's protocol.

Study Approval
This study was approved by the Institutional Animal care and Use Committee at Michigan State University (PROTO202100154) and was deemed exempt from human subjects' review by the Institutional Review Board at Michigan State University (Study ID 0008451).

Statistical Procedures
Calculations.The prevalence of hemoplasmas was calculated as the number of PCR-positive animals divided by the total number of animals sampled on the first visit.
Cumulative incidence was calculated at herd level as follows: Cumulative Incidence = number of new cases occurred during observation period / total calves and replacement heifers at risk at the start of the study.Calves and replacement heifers infected at enrollment were not included in the calculation of incidence risk.The numerator was the number of animals that were PCR-negative at first visit and became positive at the second test, and the denominator was the number of calves that were PCR-negative on the first sampling.Animals that tested negative during the first sampling but were lost to follow up, were excluded from the denominator.
As the time to follow up varied among farms, the incidence density rate was used for comparing incidence among different age groups and farms.Incidence density was calculated as the number of new PCR positive cases divided by the number of calf-days at risk.Calf-days at risk were calculated by multiplying the number of calves that tested PCR-negative during the first sampling by the number of days between the 2 visits.Incidence density was expressed as the number of new cases per 100 calf-months at risk.
The persistence of infection was calculated at the herd level as the total number of animals that were PCR positive on both sampling visits divided by the total number of animals that tested positive on the first visit.Animals that tested positive on the first visit but were lost to follow-up, were excluded from the denominator.

Statistical methods.
Animals were the unit of analysis.Statistical analyses were performed using SAS version 9.4 (SAS Institute, de Souza Ferreira et al.: Apparent prevalence of hemotropic… Cary, NC) and statistical significance was defined as P ≤ 0.05.Normality of the data was evaluated using normal probability and box plots with PROC UNI-VARIATE.Descriptive statistics were performed using PROC MEANS and used to characterize the participating herds, animals, and hematological findings.For categorical variables (i.e., age groups, herds) the x 2 test or Fisher's exact test (cell frequencies of ≤ 5) was used to compare prevalence and cumulative incidence.
PROC GLIMIX was used to compare the incidence density rate among the different age groups, including farm as a random effect.Group comparisons were conducted using the Tukey adjustment for multiple comparisons.The normality of residuals was assessed by employing PROC UNIVARIATE and visually explored through residual plots.
Separate analyses were performed for both sampling visits to examine the association between infection status for M. wenyonii, C. M. hemobos or co-infection, and dependent hematological variables.PROC GLIMMIX was used to test the hypothesis that each dependent variable (PCV, Monocytes count, Neutrophil count, Lymphocytes count, or Total Leukocyte count) was associated with infection status (0 = PCR negative, 1 = PCR positive).Each dependent variable was analyzed independently, and each model included herd as random effect and age as a fixed effect.Neutrophil count was log 10 transformed to achieve normality.Because the eosinophil count was not normally distributed, the Mann-Whitney U test was used to assess the hypothesis that eosinophil count was associated with infection status.
To compare the hematological values, animals were classified into 2 groups based on infection status during each of the 2 visits.The "negative group" consisted of animals that tested negative on both visits, while "new infection" comprised animals that tested negative initially but became positive in the subsequent visit.An independent 2-sample t-test was employed to test the hypothesis that there was an association between the mean of each dependent variable (PCV, LC, MC, NC, and TLC) and infection status (negative, new infection) groups.To ensure normality, the NC and MC variable were log 10 transformed.As the EC was not normally distributed, the Mann-Whitney U test was performed to test the hypothesis that there was a difference between the distribution of the EC among the 2 groups (0 = negative, 1 = new infection).PROC MIXED was used to test the hypothesis that the dependent variable PCV was associated with the nucleic acid load (cycle threshold (CT)) of the rt-PCR for each hemoplasma infection (CT for M. wenyonii and CT for C. M. hemobos) and the model included age as fixed effect and herd as random effect.

Herd Characteristics and General Management Practices
Of 18 farms that had participated in a previous study (Schambow et al., 2021), 5 (27.2%) were enrolled in this study.An additional 6 farms that had not been previously sampled were recruited based on contacts at industry meetings.The herds were located in 9 counties in central Michigan (Ingham, Ionia, Isabella, Barry, Washtenaw, Osceola, Gratiot, Clinton, and Montcalm).Enrolled herds together contained a total of approximately 6,850 lactating and dry cows, 773 pre-weaned calves, 1,254 weaned calves, 1,448 pre-breeding heifers and 2,401 pregnant heifers.The median number of mature cows per herd was 260 but ranged in size from 108 to 3,900 mature cows (mean = 622 ± 1102 SE).The average bulk tank SCC was 140,000 cells/mL ranging from 60,000 to 275,000 cells/mL.All mature cows were housed in freestalls while pre-weaned calves were mostly housed in individual hutches (73.7% of farms) with 26.2% of farms using group housing.All farms housed weaned, pre-breeding and pregnant heifers in groups.Typical of dairy farms in Michigan, animals were predominantly Holstein (98.9%).About 45% of farms provided occasional access to pasture throughout the year for some age groups.Fly control measures were used on 81.8% of the farms.About 50% of farmers stated that they used needles on more than one heifer, with an estimated average of 6.5 heifers per needle.Furthermore, 36% reported reusing palpation sleeves on multiple animals.
Each herd was visited twice, with an average interval of 95 d (SE ± 3.0) between visits ranging from 85 to 120 d.During the first visit, blood samples (n = 797) were collected from calves and replacement heifers with approximately 73 ± 6 animals per farm tested (range was 41 to 95 animals per farm).Not all farms contained all age groups of calves.During the second visit, 122 (15%) animals could not be located resulting in blood collection from 675 calves and replacement heifers.The number of animals per farm tested during the second visit was 61 ± 6 and ranged from 32 to 81 animals.At de Souza Ferreira et al.: Apparent prevalence of hemotropic… the first visit, blood samples (n = 174) were also collected from mature cows (median of 15 per farm, ranging from 8 to 30) with the purpose of simply confirming that the herds contained infected mature cows.
The apparent prevalence and odds of infection with hemoplasmas increased as calves aged (Table 3).The infection rate remained relatively stable at 6-8% during the first 6 mo of life.A sharp increase occurred by about 8 mo, and thereafter continued increasing to almost 100% prevalence after 17 mo of age (Figure 1).Among all farms, the least proportion of positive tests were observed in calves aged 1 to 3 mo, with proportions of positive tests ranging from 0.0% to 18.1% (Table 3).For calves aged 4 to 6 mo positive animals ranged from 0.0% to 25.0%.As compared with prevalence in the first 3 mo of age, the likelihood of infection was similar 1.0 (95% CI: 0.3 -2.7) for animals 4-6 mo of age.The proportion of infection increased for animals aged 7 -9 mo ranging from 0% to 66.6% and the likelihood of infection was 4.7 (95% CI: 2.3 -9.7) times greater as compared with animals in the first 3 mo of age.Both proportion of infection and the likelihood of being infected increased considerably in animals >10 mo.
Compared with calves aged 1 to 3 mo, the likelihood of infection increased dramatically during the second year of life (Table 3).The apparent prevalence for calves and replacement heifers showed a positive association with age group for infection with C. M. hemobos only (P < 0.01), M. wenyonii only (P < 0.01), and co-infection with both hemoplasmas (P < 0.01) (Table 4).Notably, only one calf in the 1 -3 mo age group was co-infected with both hemoplasmas, while 6 (1.6%) tested positive for C. M. hemobos only, 18 (5.0%)for M. wenyonii only, and 334 (93.0%) were PCR-negative for hemoplasmas (Table 4).The proportion of single infections for animals aged 4-6 mo was < 5% for both organisms (Table 4) and no calves were co-infected.Co-infection became more prevalent in calves aged 7-9 mo as compared with younger animals (Table 4), but the proportion of single infections did not vary between C. M. hemobos and M. wenyonii (Table 4).For animals >10 mo of age, coinfection was more common as compared with single infections and most calves were positive (Table 4).
Mycoplasma wenyonii and C. M. hemobos was confirmed using rt-PCR in blood samples obtained from at least 8 cows in all herds (herd-level apparent prevalence in cows of 100%).While the sampling scheme was not designed to estimate prevalence in this age group, of the 11 herds, 5 herds had a 100% apparent prevalence of infection with M. wenyonii, or C. M. hemobos as none of the animals tested negative (Table 2).Among all farms only 4.8% ± 1.8% of the cows tested negative.Within-herd apparent prevalence of M. wenyonii in mature cows differed among farms (P = 0.01), but no difference was observed for infection with C. M. hemobos only (P = 0.25), or co-infection with both hemoplasmas (P = 0.12).

Incidence and Persistence of Infection
Among herds, cumulative incidence of infection during the 95 d period between sampling visits varied from 3.7% to 96.0% (36.0 ± 9.7) (Table 5).A greater cumulative incidence was observed for farms 10 and 11, which were sampled in late summer, compared with the other farms (P < 0.001; Table 5).Most of the calves that were positive on the first visit remained positive for hemoplasmas at the second visit, (89.6% ± 3.6).Apparent clearance (indicating animals were initially PCR-positive but became negative when tested at the second visit) by herd, varied from 0% to 40.0% (10.18 ± 3.6).

Hematological Findings
Of the 797 blood samples collected on the first visit, leukocyte counts (including eosinophils, lymphocytes, monocytes, neutrophils, and total leukocytes), were performed on 728 and hematocrit were performed on 789 samples.On the second visit, of 675 animals sampled, leukocyte differentials were processed for 627 samples, while hematocrits were performed on 672 samples (Table 7).Not all blood samples that were collected were able to be analyzed due to technical issues encountered during sample processing.The mean hematocrit and leukocytes counts were within normal reference ranges on both visits.There were no differences observed in PCV, counts of lymphocytes, monocytes, neutrophils, and total leukocytes based on rt-PCR test results for hemotropic mycoplasma (P > 0.06).Greater eosinophil count (P < 0.001) in calves and replacement heifers were observed on both visits for PCR-positive animals compared with PCR-negative.
The mean value of all blood tests were within their reference ranges for animals that tested negative at both visits and for animals that became positive at the second visit.Except for greater eosinophil counts (P < 0.01) and lower monocytes counts (P = 0.01), animals that developed new infections did not have differences in hematological values as compared with animals that remained PCR-negative (Table 8).
There were no associations between PCV and CT values of animals infected either with M. wenyonii (P > 0.08) or C. M. hemobos (P > 0.16) on either the first or second sampling visits.

DISCUSSION
To the best of our knowledge, this study is the first to report the prevalence of hemoplasma infection in US dairy calves.In agreement with Schambow et al. (2021) herd-level apparent prevalence of infection in mature cows was 100% for both hemoplasmas.Similar to mature cows, all farms contained calves and replacement heifers that tested positive for hemoplasmas although the apparent prevalence and incidence appeared to vary among farms.Taken together, the results of Schambow et al. (2021) and this study suggest that hemoplasma infection is common in dairy cattle in Michigan.Although the herds were a convenience sample, and some were the same as those in the previous report, which could potentially bias results as the farmers were already familiar with the disease (Schambow et al., 2021), we included 6 herds in this study that had not been tested previously.The consistent prevalence of hemoplasma infection in these new herds further supports that hemoplasma infections are widespread among dairy farms in Michigan.Most reports on hemoplasma infection in cattle have focused on assessing prevalence at a specific point in time, using a cross-sectional study design (Fujihara et al., 2011;Congli et al., 2011;Schambow et al., 2021).Consequently, little is known about the rate at which new cases of hemoplasmas emerge.Understanding the incidence is crucial, as it provides insights into the dynamics of the affected population and helps identify individuals at risk of contracting the disease (Smith 2012).While our study also has a cross-sectional approach, we incorporated a longitudinal component to examine the occurrence of new cases, particularly in younger animals.
The overall animal-level prevalence of individual hemoplasma in our study was similar to the proportions reported in a Japanese study that tested 124 beef cattle aged 1-4 yr (Tatsukawa et al., 2021).They reported that 25.8% were infected with M. wenyonii only, 8.1% tested positive for C. M. hemobos, and 62.1% had coinfections with both hemoplasmas (Tatsukawa et al., 2021).In another study that tested 19 animals <1 yr of age in Malaysia, 31.5% were infected with M. wenyonii, 10.5% were positive for C. M. hemobos, and 36.8% had co-infections with both hemoplasmas (Mohd Hasan et al., 2017).In contrast, another study conducted in Japan that tested 128 animals <1 yr old found that infection with M. wenyonii was more prevalent (27.3%) compared with co-infection with both hemoplasmas (2.3%) (Tagawa et al., 2012).It has been suggested that differences in cattle breeds and geographical variations may influence the prevalence of each hemoplasma (Tatsukawa et al., 2021).
Interestingly, our findings suggest that as animals age, co-infection with both bovine hemoplasmas becomes more common than single infections.This could be a cumulative result of the prolonged exposure to both pathogens.As older animals spend more time within the herd compared with younger animals, they accumulate contact with potential sources of hemoplasmas infection.However, the prevalence of each hemoplasma infection among older animals has varied among studies.Some researchers have reported greater prevalence of single M. wenyonii infections (Mohd Hasan et al., 2017;Niethammer et al., 2018;Erol et al., 2022), while others have observed a greater prevalence of single C. M. hemobos infection (Fujihara et al., 2011) (as compared with co-infections).Currently, there is not enough evidence to determine potential risk factors associated with infection with individual hemoplasmas, as most studies that have assessed prevalence have included relatively few herds and were not designed to identify differences among species.
The variation in apparent prevalence and the cumulative incidence we observed among herds may be de Souza Ferreira et al.: Apparent prevalence of hemotropic… partially attributed to the time of the year the animals were sampled.While our study was not designed to be able to determine differences in prevalence based on season, in agreement with Strugnell and McAuliffe (2012) the farms that were sampled in late summer appeared to have greater prevalence as compared with farms we sampled in earlier months.Although potential modes of transmission of hemoplasmas in cattle remain poorly understood, mechanical transmission through blood sucking insects has been reported as a possible route (Hofmann-Lehmann et al., 2004).Increased activity of arthropod vectors during warmer months might account for our results, suggesting that vectors may play a role in transmission, but studies need to be conducted to specifically address this issue.Nevertheless, we also observed new infections during colder months which is unfavorable for arthropod vectors in our region .This suggests that other transmission mechanisms may be involved.Transplacental transmission has been reported but its overall significance is unknown (Fujihara et al., 2011;Sasaoka et al., 2015).In a prevalence study conducted in Japan, dairy calves were tested using PCR during their first week of their life.Of 71 calves, 14% were infected with hemoplasmas, while 86% tested negative (Tagawa et al., 2013).Our results were similar as few young calves were infected.We hypothesize that transplacental transmission may be another route of infection.However, as reported by Tagawa et al. (2013), we could not differentiate between transplacental or transmission in early life as calves were not tested immediately after delivery.
Given that age is a common risk factor for many diseases, determining the timing of infection with hemoplasmas can contribute to a better understanding of risk factors and aid development of interventions to reduce the risk of infection.Our results indicate that the apparent prevalence of infection with hemoplasmas remains relatively stable within the first months of life, with a sharp increase around 8 mo, and a consistently high prevalence thereafter.While we did not evaluate modes of transmission, possible risk factors could be related to routine practices that facilitate exposure.Transmission of Bovine Leukemia Virus (BLV) has been associated with exposure to infected blood by reuse of needles and rectal palpation sleeves as well as failure to adequately remove blood from instruments used for dehorning and hoof-trimming (Divers et al., 1995;Kuczewski et al., 2021).Re-use of needles for multiple injections is common in Michigan and likely in other regions.In 2018, the National Animal Health Monitoring System (NAHMS) reported that >51% of operations administered 2 to 10 injections pre needle (USDA, 2018).Producers interviewed by Schambow et al. (2021) estimated that a typical mature cow had  received approximately 65 injections from birth and the same needle was used to inject about 15 animals.In our current study, about half of farmers indicated that they used the same needle to inject about 7 animals.
Adoption of practices that reduce potential exposure to blood borne pathogens (use of single-use needles and obstetrical sleeves) has been shown to reduce the prevalence of BLV in a dairy herd from 44% to 17% over a 2-year period, (Sprecher et al., 1991).Similar practices could reduce transmission of hemoplasmas.
There is limited knowledge about variation in bacterial load of hemoplasmas in the bloodstream of cows (Keeton and Jain, 1973;Strugnell et al., 2011).In an older study that tested blood obtained from a splenectomized bull and used scanning electron microscopy for diagnosis, many RBC were initially infected with E. wenyonii but were not diagnosed in blood samples collected 10 d later (Keeton and Jain, 1973).Apparent clearance of infection with M. wenyonii was reported in 9 of 12 cows that were tested 40 d apart using PCR and Denaturing Gradient Gel Electrophoresis (DGGE) (Strugnell et al., 2011).In contrast, in our study very few test-positive animals later tested negative.Molecular diagnostics are preferred for diagnosing hemoplasma infection (Tagawa et al., 2008;Ritzmann et al., 2009;Hoelzle et al., 2011;Girotto-Soares et al., 2016;Niethammer et al., 2018).We used rt-PCR for testing but could not determine if animals that initially tested positive and then became negative achieved bacteriological clearance or if infection was not detected due to a low bacterial load.However, most animals that tested positive remained positive and our results are similar to previous observations that cats infected with hemoplasmas become persistent carriers (Messick, 2004).Our results suggest that infections occur during the first year of life and remain persistent, thus potentially providing a reservoir for infections of herdmates.
Hemotropic Mycoplasmas attach to the surface of red blood cells and can cause hemolysis, resulting in anemia (Gladden et al., 2016).Although infection with C. M. hemobos and M. wenyonii have been associated with this condition (Hoelzle et al., 2011;Hofmann-Lehmann et al., 2004;Purnell et al., 1976;Gladden et al., 2016), the evidence of bovine hemoplasmas leading to anemia is conflicting (Gladden et al., 2016).Unlike Tagawa et al., (2010Tagawa et al., ( , 2012)), we were not able to identify an association between infection status and anemia.However, our results were consistent with other researchers who used PCR for testing (Tatsukawa et al., 2021;McFadden et al., 2016).Animals that we sampled appeared clinically healthy, and previous researchers have reported that when anemia is present in infected animals it is usually accompanied by other clinical signs such as fever (Smith et al., 1990), malaise, or edema of the hind limbs, udder (Genova et al., 2011), or scrotum (Montes et al., 1994).At least one researcher has reported the presence of anemia in the absence of clinical signs in a mature cow infected with M. wenyonii and bovine hemoplasmas should be considered in the differential diagnosis for any cow presenting with anemia (Gladden et al., 2016).
Only a few researchers have investigated the impact of hemoplasma infection on WBC of cattle (Tagawa et al., 2010;Niethammer et al., 2018;Tatsukawa et al., 2021).In one study, greater WBC was reported for cows co-infected with both bovine hemoplasmas as compared with PCR-negative cows (Tatsukawa et al., 2021).Infection with M. wenyonii (Tagawa et al., 2010) or C. M. hemobos in cows (Niethammer et al., 2018) have both been associated with greater WBC as compared with PCR-negative cows.Although our results did not agree with these studies, we did observe an interaction between hemoplasma infection and eosinophils.While the eosinophils were within the normal range, animals infected with hemoplasmas had a greater eosinophil concentration than non-infected animals.The differences in eosinophils concentration between infected and non-infected cattle were unexpected, as an increase in eosinophils is typically associated with parasitic infections (Kramer, 2000).We also compared leukocyte counts of animals that remained PCR-negative on both visits with animals that changed from PCR-negative to positive.While an increase in monocytes could be anticipated due to the involvement of monocytes in immune defense against bacterial pathogens, the observed trend was rather unexpected.Animals that tested positive for the infection exhibited slightly lower mean MC values compared with those that tested negative.While relatively few differences seen in hematological values, previous researchers have reported that hemoplasmas may stimulate the immune system resulting in changes in leukocyte counts (Niethammer et al., 2018;Tatsukawa et al., 2021).

CONCLUSIONS
Our results provide novel information about the prevalence and incidence of infection with hemotropic mycoplasmas in dairy calves and heifers.The increase in prevalence as the heifers age may be attributed to the chronic nature of hemoplasma infections.The greater cumulative incidence observed on farms sampled during the summer may further support the hypothesis that vectors play a significant role in hemoplasmas transmission.Our findings agree with previous observations that infection with hemoplasmas alone does not usually lead to anemia and most animals chronically infected with hemoplasmas remained apparently healthy.The association between infectious status and levels of eosinophils and monocytes may indicate an attempt of the immune system to combat the infection.However, further stud-ies are necessary to gain a better understanding of this association and its underlying mechanisms.Group of heifers that tested negative for hemotropic mycoplasma on both visits.
2 Group of heifers that tested negative for hemotropic mycoplasma on the first and became PCR positive on the second visit. 3 The statistics of the hematological parameters of hemoplasma-positive group were compared with those of the negative group by t-test.

de
Souza Ferreira et al.: Apparent prevalence of hemotropic…Table 3. Apparent prevalence of infection with either Mycoplasma wenyonii or Candidatus Mycoplasma hemobos in calves and replacement heifers by farm and months of age at the first sampling visit detected using real-time PCR on 11 farms in

de
Figure 1.-Apparent prevalence of infection with either Mycoplasma wenyonii or Candidatus Mycoplasma hemobos by month of age at the first sampling visit for calves and replacement heifers on 11 farms in central Michigan during March-June 2022.Bars represent SE.

de
Souza Ferreira et al.: Apparent prevalence of hemotropic…

Table 1 .
de Souza Ferreira et al.: Apparent prevalence of hemotropic… Within herd apparent prevalence of infection with Candidatus Mycoplasma hemobos and Mycoplasma wenyonii using rt-PCR for 797 calves and heifers aged between 1 d and 27 mo old, tested on 11 farms in Michigan in March -June 2022 (initial sampling)

Table 2 .
Within herd apparent prevalence of infection with Candidatus Mycoplasma hemobos or Mycoplasma wenyonii using rt-PCR for 174 mature cows tested on 11 farms in Michigan in March -June 2022 (initial sampling) 1Prevalence differed among farms (P = 0.01).

Table 4 .
Apparent prevalence of infection with Mycoplasma wenyonii, Candidatus Mycoplasma hemobos and co-infection by months of age at the first sampling visit tested using rt-PCR on blood samples collected from calves and replacement heifers on 11 farms in central Michigan during March-June 2022

Table 5 .
Herd-level apparent prevalence, persistence infection, apparent absence, and cumulative incidence of Candidatus Mycoplasma hemobos and Mycoplasma wenyonii tested using rt-PCR on blood samples collected twice from calves and replacement heifers on 11 farms in central Michigan during March-September 2022

Table 8 .
t-test results comparing hematological parameters in calves and replacement heifers negative for infection with Mycoplasma wenyonii or Candidatus Mycoplasma hemobos at both sampling visits or negative at the first sampling visit but PCR positive on the second sampling visit for M. wenyonii and/or C. M. hemobos tested using rt-PCR on 11 farms in central Michigan during March-September 2022 1