Risk factors for purulent vaginal discharge and its association with reproductive performance of lactating Jersey cows

Despite the rapid growth in popularity of the Jersey breed, most research on dairy cows in the United States has been done with the Holstein breed. Postpartum uterine diseases negatively influence reproductive performance of dairy cows and limited data are available regarding predisposing factors for uterine diseases in Jersey cows. Our objectives were to determine the prevalence and risk factors for purulent vaginal discharge (PVD) and its effect on fertility of lactating Jersey cows. This was a retrospective observational study with data collected from 3,822 Jersey cows. The Metricheck device was used for PVD diagnosis, and positive cases (≥50% of pus in exudate) were further classified for severity using the following 4 categories based on the amount of pus observed: 50 to 60% pus in exudate, 60 to 90% pus in exudate, 90 to 100% pus in exudate, and 90 to 100% pus in exudate + uterine fluid detected by palpation per rectum. Univariable and multivariable regression analyzes were conducted to dissect the risk factors for PVD and severity of PVD in Jerseys cows. The major risk factors for PVD were calving-related problems, retained fetal membrane, metritis, and days in the close-up diet. A subgroup of cows (n = 740) was scored for body condition and locomotion scores and had blood sampled in the peripartum for determination of plasma concentrations of nonesterified fatty acids and β-hydroxybutyrate. In the multivariable analysis that considered data collected for the subgroup of cows, peripartum nonesterified fatty acids, postpartum β-hydroxybutyrate, and peripartum locomotion and body condition scores were not retained in the reduced model of predictors of PVD. Not surprisingly, pregnancy per artificial insemination following the first and second services was reduced in cows diagnosed with PVD compared with cows without PVD. In addition, PVD was associated with increased odds of pregnancy loss after the first service and reduced hazard of pregnancy by 305 d in milk. The stratification of PVD severity according to the amount of pus observed and the consistency of the uterus was meaningful, as observed by the differences in reproductive outcomes between cows with diverging amounts of pus in the retrieved exudate. Risk factors for PVD in Jersey cows was similar to previously reported for Holstein cows, and a strong detrimental effect of PVD on fertility was also observed in Jersey cows.

Short communication: Simultaneous measurements of estrus behavior and plasma concentrations of estradiol during estrus in lactating and nonlactating dairy cows

Estrus is an important behavior that can potentially be subjected to genomic selection. Circulating estradiol concentrations at estrus may be a useful phenotype if the absolute concentrations of estradiol are associated with overt phenotypes for estrus (activity, rump touches, or both; e.g., mounts, chinrests) that can be easily observed. The objective was to measure plasma estradiol concentrations at estrus and associate these measurements with the increase in activity (steps per hour) and rump touches received at estrus. We also tested the effect of lactation on the estrus traits that we measured. Cows (n = 11 lactating and n = 9 nonlactating) were treated with PGF_2α to synchronize estrus. A jugular vein was cannulated to collect blood every 2 h for plasma estradiol measurement. Plasma LH was measured during the periestrual period to determine the time of the LH surge. Cows were fitted with an accelerometer to measure activity (steps per hour) and a capacitive touch sensing device to measure the number of rump touches and total touch time. Plasma estradiol concentrations were poorly correlated with overt signs of estrus during the period leading up to maximum estrus activity. After peak estrus activity (when cows were going out of estrus and plasma estradiol concentrations were decreasing), a stronger correlation was detected between overt signs of estrus and plasma estradiol concentrations. Effective selection for improved estrus expression based on plasma estradiol concentrations will depend on whether the cow is coming into or going out of estrus at the time of blood sampling. An association existed between lactation and fewer number of hours in estrus when estrus was defined by an increase in activity (steps per hour). Lactating cows had a shorter interval from the onset of estrus to the LH surge, and the shorter interval to the LH surge may have reduced the period of elevated estradiol during estrus in the lactating cows. Understanding mechanisms that control the sensitivity of the cow to estradiol and making appropriate selection decisions based on these mechanisms will likely increase overt signs of estrus in dairy cows.

Scanning electron microscopy of the surface epithelium of the bovine endometrium

The surface epithelium of the bovine endometrium comprises at least 2 cell types (ciliated cells and secretory cells with microvilli), but their distribution and morphological changes over the estrous cycle are poorly understood. The objective was to quantify the number of ciliated cells and assess morphological changes in secretory cells on the uterine surface epithelium during the estrous cycle. Caruncular endometrium (CAR) and intercaruncular endometrium (ICAR) samples were collected from the uterine body, the horn ipsilateral to the corpus luteum or dominant follicle (H-CL/DF), and the horn contralateral to the corpus luteum or dominant follicle (H-NCL/NDF) from heifers following slaughter on d 0 (estrus; n = 5) or d 14 (mid-luteal phase; n = 5) of the estrous cycle. Samples were prepared for scanning electron microscopy at 1,000× magnification. Four to 10 fields (256 × 225 µm) for each sample were examined (n = 567 images). The number of ciliated cells was counted and the surface was scored for the morphology of the secretory cells (0 = absence of microvilli on surface; 3 = 100% of surface covered with microvilli). Ciliated cells were present in both the CAR and ICAR regions. The number of ciliated cells per field increased from d 0 to 14 in CAR and decreased from d 0 to14 in ICAR. The scanning electron microscopy revealed a general lack of uniformity in the lawn of microvilli on the surface of the endometrium. Based on the scores, approximately 25% of the fields had a surface that was <50% covered by microvilli. Depletion of microvilli may be explained by a normal process where apical protrusions are formed and either regress back into the cell surface or break to release their contents into the uterine lumen. These studies support the hypothesis that the surface of the luminal epithelium changes during the estrous cycle through a process that involves remodeling of the apical surface. The morphology of the apical surface may have a key role in governing pregnancy establishment.

Pregnancy associated glycoproteins (PAGs) and pregnancy loss in high vs sub fertility heifers

Reproductive inefficiency and infertility are major financial burdens to domestic livestock. Variables associated with these reproductive losses during early gestation include contributions from the oocyte, uterus, sperm, embryo and placenta. Bovine pregnancy associated glycoproteins (PAG) are produced by the binucleate cells of the ruminant placenta and can be used to diagnose pregnancy. Increased circulating concentrations of PAG early in gestation have been correlated with pregnancy success and decreased concentrations are predictive of impending embryonic mortality in both beef and dairy cattle. The objectives of the current study were to determine whether: 1) heifer fertility status is associated with circulating concentrations of PAG and pregnancy loss; and 2) PAG concentrations within the same animal are repeatable across multiple pregnancies. We hypothesized maternal PAG concentrations would be increased in high fertility compared to subfertile heifers but not repeatable across subsequent pregnancies in the same heifer. Serial embryo transfer (ET; n = 4 rounds) was used to classify predominately Angus heifers (n = 92) as highly fertile (HF = 30; 100% pregnancy success) or subfertile (SF = 62; average = 33%; range = 25-75% pregnancy success) based on day 28 ultrasound diagnosis. Blood samples were collected at both day 28 and 44 for quantification of circulating PAG concentrations by an in house PAG ELISA with antibodies raised against early secreted PAGs. Pregnancy was terminated at day 44 of gestation and heifers were allowed 30 days recovery before synchronization for the next ET. Only heifers that were diagnosed pregnant by ultrasound were used in this study (HF: n = 30, SF: n = 62). Serum concentrations of PAGs were not different between HF (5.90 ± 0.27 ng/mL) and SF (5.56 ± 0.31 ng/mL; P = 0.16) heifers at day 28 of gestation nor was there a difference at day 44 of gestation (P = 0.32). Subfertile heifers had increased pregnancy loss between days 28 and 44 of gestation. Based on odds ratio analysis, SF heifers had a 2.41 times chance to undergo pregnancy loss between day 28-44 compared to HF heifers (P < 0.05). There was no correlation (P > 0.05) in maternal circulating concentrations of PAG between pregnancies on day 28 or 44 of gestation in samples obtained from HF heifers. In summary, circulating concentrations of PAG are not different between HF and SF heifers; however, HF classified heifers have decreased pregnancy loss between days 28 and 44 of gestation.

Genome-wide association analysis and gene set enrichment analysis with SNP data identify genes associated with 305-day milk yield in Holstein dairy cows

Milk production traits, such as 305-day milk yield (305MY), have been under direct selection to improve production in dairy cows. Over the past 50 years, the average milk yield has nearly doubled, and over 56% of the increase is attributable to genetic improvement. As such, additional improvements in milk yield are still possible as new loci are identified. The objectives of this study were to detect SNPs and gene sets associated with 305MY in order to identify new candidate genes contributing to variation in milk production. A population of 781 primiparous Holstein cows from six central Washington dairies with records of 305MY and energy corrected milk were used to perform a genome-wide association analysis (GWAA) using the Illumina BovineHD BeadChip (777 962 SNPs) to identify QTL associated with 305MY (P < 1.0 × 10^-5 ). A gene set enrichment analysis with SNP data (GSEA-SNP) was performed to identify gene sets (normalized enrichment score > 3.0) and leading edge genes (LEGs) influencing 305MY. The GWAA identified three QTL comprising 34 SNPs and 30 positional candidate genes. In the GSEA-SNP, five gene sets with 58 unique and 24 shared LEGs contributed to 305MY. Identification of QTL and LEGs associated with 305MY can provide additional targets for genomic selection to continue to improve 305MY in dairy cattle.

Effects of recombinant bovine somatotropin during the periparturient period on innate and adaptive immune responses, systemic inflammation, and metabolism of dairy cows

The aim of this experiment was to determine effects of treating peripartum dairy cows with body condition score ≥3.75 with recombinant bovine somatotropin (rbST) on immune, inflammatory, and metabolic responses. Holstein cows (253±1d of gestation) were assigned randomly to 1 of 3 treatments: untreated control (n=53), rbST87.5 (n=56; 87.5mg of rbST), and rbST125 (n=57; 125mg of rbST). Cows in the rbST87.5 and rbST125 treatments received rbST weekly from -21 to 28d relative to calving. Growth hormone, insulin-like growth factor 1, haptoglobin, tumor necrosis factor α, nonesterified fatty acids, β-hydroxybutyrate, glucose, and cortisol concentrations were determined weekly from -21 to 21d relative to calving. Blood sampled weekly from -14 to 21d relative to calving was used for hemogram and polymorphonuclear leukocyte (PMNL) expression of adhesion molecules, phagocytosis, and oxidative burst. Cows were vaccinated with ovalbumin at -21, -7, and 7d relative to calving, and blood was collected weekly from -21 to 21d relative to calving to determine IgG anti-ovalbumin concentrations. A subsample of cows had liver biopsied -21, -7, and 7d relative to calving to determine total lipids, triglycerides, and glycogen content. Growth hormone concentrations prepartum (control=11.0±1.2, rbST87.5=14.1±1.2, rbST125=15.1±1.3ng/mL) and postpartum (control=14.4±1.1, rbST87.5=17.8±1.2, rbST125=21.8±1.1ng/mL) were highest for rbST125 cows. Cows treated with rbST had higher insulin-like growth factor 1 concentrations than control cows (control=110.5±4.5, rbST87.5=126.2±4.5, rbST125=127.2±4.5ng/mL) only prepartum. Intensity of L-selectin expression was higher for rbST125 than for control and rbST87.5 cows [control=3,590±270, rbST87.5=3,279±271, rbST125=4,371±279 geometric mean fluorescence intensity (GMFI)] in the prepartum period. The PMNL intensities of phagocytosis (control=3,131±130, rbST87.5=3,391±133, rbST125=3,673±137 GMFI) and oxidative burst (control=9,588±746, rbST87.5=11,238±761, rbST125=12,724±781 GMFI) were higher for rbST125 cows than for control cows during the prepartum period. Concentrations of serum IgG anti-ovalbumin tended to be higher for rbST125 cows than for control cows (control=0.75±0.11, rbST87.5=0.94±0.10, rbST125=1.11±0.11 optical density) in the prepartum period. Haptoglobin concentration was significantly reduced 7d postpartum for rbST125 treatment compared with control and rbST87.5 treatments (control=2.74±0.28, rbST87.5=2.81±0.28, rbST125=1.87±0.28 optical density). Although treatment tended to affect postpartum β-hydroxybutyrate (control=747.5±40.2, rbST87.5=753.2±40.1, rbST125=648.8±39.7 µmol/L), it did not affect liver contents of total lipids, triglycerides, or glycogen. Incidence of metritis among rbST125 cows was reduced compared with that in control cows (control=23.1, rbST87.5=18.0, rbST125=7.8%). Treatment of dairy cows with 125mg of rbST improved innate immune responses and IgG concentration, with limited effects on metabolism.

Prepartum stocking density: effects on metabolic, health, reproductive, and productive responses

The objectives of the current experiment were to determine the effects of 2 prepartum stocking densities on milk yield, concentration of metabolites during the peripartum period, and health and reproductive parameters of dairy cows. Jersey cows enrolled in the experiment at 254±3 d of gestation were balanced for parity (nulliparous vs. parous) and previous lactation projected 305-d mature equivalent milk yield (parous) and assigned to 1 of 2 treatments: 80% headlock stocking density (80SD; 38 animals/48 headlocks) and 100% headlock stocking density (100SD; 48 animals/48 headlocks). The number of experimental units was 8 (4 replicates and 2 pens/treatment per replicate). In total, 154 nulliparous and 184 parous animals were enrolled in the 80SD treatment and 186 nulliparous and 232 parous animals were enrolled in the 100SD treatment. At the start of each replicate, treatments were switched within pen. Cows were milked thrice daily and monthly milk yield, fat and protein content, and somatic cell count data were recorded up to 155 d postpartum. Plasma nonesterified fatty acid concentration was measured weekly, from -18±3 to 17±3 d relative to calving, and plasma β-hydroxybutyrate was measured weekly, from 1±2 to 17±3 d relative to calving. Cows were examined 1, 4±1, 7±1, 10±1, and 13±1 d relative to calving for diagnosis of uterine diseases. Blood was sampled for determination of progesterone concentration and resumption of ovarian cycles 35±3 and 45±3 d relative to calving. Average headlock (74.1±0.4 vs. 94.5±0.3%) and stall (80.8±0.4 vs. 103.1±0.4%) stocking density was lower for the 80SD treatment compared with the 100SD treatment. Treatment did not affect incidence of retained fetal membranes (80SD=5.1, 100SD=7.8%), metritis (80SD=21.2, 100SD=16.7%), acute metritis (80SD=9.9, 100SD=9.4%), and vaginal purulent discharge (80SD=5.8, 100SD=7.9%). Concentrations of nonesterified fatty acids (80SD=251.5±6.1, 100SD=245.9±5.6μmol/L) and β-hydroxybutyrate (80SD=508.2±14.3, 100SD=490.9±13.6μmol/L) were not different between treatments. Treatment had no effect on percentage of cows removed from the herd on the first 60 d postpartum (80SD=6.1, 100SD=5.1%) and on rate of removal from the herd up to 305 d postpartum 80SD=referent, 100SD [adjusted hazard ratio (95% confidence interval)]=1.02 (0.75, 1.38). Percentages of cows pregnant to first (80SD=41.9, 100SD=48.4%) and second (80SD=49.3, 100SD=42.0%) postpartum AI were not different between treatments. Finally, treatment did not affect energy-corrected milk yield up to 155 d postpartum (80SD=33.8±0.5, 100SD=33.4±0.5kg/d). In herds with weekly or twice weekly movement of new cows to the prepartum pen and separate housing of nulliparous and parous animals, a target stocking density of 100% of headlocks on the day of movement is not expected to affect health, metabolic, reproductive, and productive parameters.

Effects of weekly regrouping of prepartum dairy cows on metabolic, health, reproductive, and productive parameters

The objectives of the current experiment were to determine the effect of 2 prepartum grouping strategies on the health, metabolic, reproductive, and productive parameters of dairy cows. Jersey cows enrolled in the experiment at 253±3 d of gestation (d 0=calving) were balanced for parity and projected 305-d mature equivalent and assigned to 1 of 2 treatments. Cows assigned to the traditional (TRD; n=6 replicates with a total of 308 cows) treatment were moved to the study pen as a group of 44 cows and weekly thereafter groups of 2 to 15 cows were moved to the study pen to reestablish stocking density. Cows assigned to the all-in-all-out (AIAO; n=6 replicates with a total of 259 cows) treatment were moved to the study pen in groups of 44 cows, but no new cows entered the AIAO pen until the end of the replicate. At the end of each replicate, a new TRD and AIAO group started but pens were switched. Cows were milked thrice daily and monthly milk yield, fat and protein contents, and somatic cell count data were recorded up to 305 d postpartum. Plasma nonesterified fatty acid concentration was measured weekly from d -18±3 to 24±3 and plasma β-hydroxybutyrate was measured weekly from d 3±3 to 24±3. Cows were examined on d 1, 4±1, 7±1, 10±1, and 13±1 for diagnosis of uterine diseases and had their ovaries scanned by ultrasound on d 39±3 and 53±3 to determine resumption of ovarian cycles. Average stocking density was reduced for the AIAO (71.9%) treatment compared with the TRD (86.9%) treatment. Treatment did not affect the incidences of retained fetal membranes (TRD=10.9, AIAO=11.6%), metritis (TRD=16.7, AIAO=19.8%), and acute metritis (TRD=1.7, AIAO=3.6%). Concentrations of nonesterified fatty acids (TRD=80.4±8.2, AIAO=62.9±8.5 µmol/L) and β-hydroxybutyrate (TRD=454.4±10.9, AIAO=446.1±11.1 µmol/L) were not different between treatments. Percentages of cows that resumed ovarian cycles by d 39±3 (TRD=70.8, AIAO=63.1%) and 53±3 (TRD=90.1, AIAO=90.2%) were not different between treatments. Similarly, treatment had no effect on rate of removal from the herd {TRD=referent, AIAO [(adjusted hazard ratio (95% confidence interval)]=0.85 (0.63, 1.15)} or rate of pregnancy [TRD=referent, AIAO=1.07 (0.88, 1.30)]. Finally, treatment did not affect energy-corrected milk yield (TRD=34.4±0.6, AIAO=34.3±0.7 kg/d). In conditions of adequate feed bunk space, the AIAO treatment did not improve health, metabolic, reproductive, or productive parameters compared with the TRD treatment.

Evaluation of presynchronized resynchronization protocols for lactating dairy cows

The objectives of this experiment were to determine the speed at which cows that had their estrous cycle presynchronized with a GnRH or PGF(2α) injection are reinseminated and become pregnant. Furthermore, this experiment aimed to determine whether treatment with a controlled internal drug-releasing (CIDR) insert during the timed artificial insemination (AI) protocol improves pregnancy per AI (P/AI) of cows that had their estrous cycle presynchronized with GnRH or PGF(2α). Lactating cows from 2 herds were assigned to 1 of 2 presynchronization treatments at 32 ± 4 d after AI: GGPG (n=452)--GnRH injection at enrollment (d 0), 7d before the start of the timed AI protocol, and P11GPG (n=466)--PGF(2α) injection on d 3, 11 d before the start of the timed AI protocol. Cows observed in estrus at any interval after enrollment were reinseminated on the same day. Cows not observed in estrus by d 7 were paired by presynchronization treatment and assigned to receive or not receive a CIDR insert during the timed AI protocol (CIDR = 240, no CIDR = 317). Timed AI protocols were the Ovsynch56 at site A and the Cosynch48 at site B. A subsample of cows from site A had their ovaries scanned by ultrasound at enrollment and on the day of the first GnRH and PGF(2α) injections of the timed AI protocol and had blood sampled at each injection of the timed AI protocol for determination of progesterone concentration. Cows were examined for pregnancy 32 ± 4 and 67 ± 4 d after reinsemination. Cows in the P11GPG treatment had a faster reinsemination rate [adjusted hazard ratio = 1.24 (95% CI = 1.07, 1.45)] and were less likely to be submitted to the timed AI protocol (40.3 vs. 89.8%) and to be reinseminated at a fixed time (38.6 vs. 83.9%). The interval from enrollment to reinsemination was shorter for cows in the P11GPG group (13.0 ± 0.4 vs. 15.0 ± 0.2d). Presynchronization treatment did not affect P/AI 32 ± 4 d (GGPG = 42.3%, P11GPG = 39.3%) and 67 ± 4 d (GGPG = 37.0%, P11GPG = 35.4%) after reinsemination. Pregnancy rate from d 0 to 7 (GGPG = 3.6%, P11GPG = 17.7%) and from d 8 to 14 (GGPG = 1.6%, P11GPG = 5.7%) were greater for cows in the P11GPG treatment. Treatment with the CIDR insert during the timed AI protocol did not affect P/AI 32 ± 4 d (CIDR = 41.7%, no CIDR = 41.4%) and 67 ± 4 d (CIDR = 36.5%, no CIDR = 35.3%) after reinsemination. A greater percentage of cows in the GGPG treatment had progesterone concentration ≥ 1 ng/mL on the day of the first GnRH injection of the timed AI protocol (83.8 vs. 51.5%), but a greater percentage of cows in the P11GPG treatment ovulated in response to the first GnRH injection of the timed AI protocol (66.1 vs. 46.8%). We conclude that the P/AI of cows that had their estrous cycle presynchronized with GnRH or PGF(2α) was not different, but in herds with adequate estrous detection efficiency and accuracy, presynchronization with PGF(2α) may reduce the interval to the establishment of pregnancy.