A hidden Markov model to predict early mastitis from test-day somatic cell scores.

In many countries, high somatic cell scores (SCS) in milk are used as an indicator for mastitis because they are collected on a routine basis. However, individual test-day SCS are not very accurate in identifying infected cows. Mathematical models may improve the accuracy of the biological marker by making better use of the information contained in the available data. Here, a simple hidden Markov model (HMM) is described mathematically and applied to SCS recorded monthly on cows with or without clinical mastitis to evaluate its accuracy in estimating parameters (mean, variance and transition probabilities) under healthy or diseased states. The SCS means were estimated at 1.96 (s.d. = 0.16) and 4.73 (s.d. = 0.71) for the hidden healthy and infected states, and the common variance at 0.83 (s.d. = 0.11). The probability of remaining uninfected, recovering from infection, getting newly infected and remaining infected between consecutive test days was estimated at 78.84%, 60.49%, 11.70% and 15%, respectively. Three different health-related states were compared: clinical stages observed by farmers, subclinical cases defined for somatic cell counts below or above 250 000 cells/ml and infected stages obtained from the HMM. The results showed that HMM identifies infected cows before the appearance of clinical and subclinical signs, which may critically improve the power of the studies on the genetic determinants of SCS and reduce biases in predicting breeding values for SCS.

Genetic factors affecting susceptibility to udder pathogens.

Many studies have identified genetic factors underlying resistance or susceptibility to mastitis in dairy cows and heifers. Some authors focused on polygenic variation while others searched for genes and/or quantitative trait loci with major effects on mastitis. Classical traits related to mastitis include somatic cell counts, electrical conductivity and clinical cases of the disease. With the development of automatic milking devices and '-omics' technologies, new traits are considered, such as acute phase proteins, immunological assays, and milk flow patterns, and new biological pathways are discovered, for example the role of mammary epithelium and the nervous system. The usefulness of these traits for the identification of resistant cows is discussed in relation to the biological mechanisms underlying the development of the disease. In addition, the utility of these traits for genetic improvement is reviewed. Finally, the problem of durability in resistance is addressed, including co-evolution and the cost of resistance.

Neutrophils in the war against Staphylococcus aureus: predator-prey models to the rescue.

To address the question of whether a minimum concentration of blood neutrophils is necessary to decrease Staphylococcus aureus concentration in mastitic milk, literature was searched for studies in which neutrophils were incubated with Staph. aureus. Different mathematical models that describe the changes in Staph. aureus population as a function of neutrophilic concentrations were applied to the collected data. The best fitted model established (1) that the rate of bacterial killing depended on the ratio of neutrophils to bacteria with neutrophilic attack rate accelerating at first before decelerating as the ratio increases, and (2) that neutrophil concentration should be within a limited range to trigger a decline in the bacterial population. Outcomes of this model are supported by what is known about neutrophilic functions and laboratory findings in bovine and human neutrophils. These results may be of assistance in setting selection goals for a better resilience to Staph. aureus mastitis in dairy cattle. Indeed, an optimal neutrophilic concentration appears to exist for successful clearance of Staph. aureus infection, which is neither the lowest nor the highest one.

Genetic factors affecting susceptibility of dairy cows to udder pathogens.

Bovine mastitis remains the most costly disease in dairy cattle. Breeding for resistance to udder pathogens has been proposed as a complementary tool to therapeutic and prophylactic measures not totally effective against the disease. This paper reviews factors affecting cows' susceptibility to pathogens at the animal, cellular/hormonal and DNA levels. Such factors will be useful in achieving genetic improvement for resistance only if they have desirable properties at the genetic and immunological levels. Because such properties are not always of significant magnitude, further research is necessary to identify characteristics of resistance in cows, considering the constant and complex interactions that occur between hosts and pathogens.

Pregnancy-associated glycoprotein and decreased polymorphonuclear leukocyte function in early post-partum dairy cows.

Phagocytosis and oxidative burst activity of polymorphonuclear neutrophil leukocytes (PMN) isolated from blood and pregnancy-associated glycoprotein (bPAG) concentrations in plasma were evaluated in two longitudinal studies in dairy cows from 3 weeks before until 5 weeks after calving, carried out in the United States and in Europe. Ingestion of Staphylococcus aureus by blood PMN increased during the first week after calving and normalised 3 weeks post-partum. Phagocytosis of Escherichia coli did not change in the early post-partum period. In both studies, a significant decrease in oxidative burst activity of PMN was observed between 1 and 3 weeks after calving. In all cows, a very significant increase in plasma bPAG concentration was found between 1 week before and 2 weeks after calving. The peak of bPAG concentration in plasma immediately preceded the alterations of blood PMN functions. These results suggest that bPAG may be associated with inhibition of PMN function of dairy cows during the early post-partum period.

Genetic association between parameters of inmate immunity and measures of mastitis in periparturient Holstein cattle.

Relationships between genetic measures of mastitis (somatic cell score, score for clinical mastitis, and scores for IMI with major or minor pathogens) and immunological parameters (physiological and molecular markers) were examined for periparturient Holstein cows. Physiological markers included 11 in vitro immunological assays. Molecular markers included the second exon of the DRB3 locus of the bovine major histocompatibility complex, the IgG2 isotype genotype, and the CD18 genotype (the locus responsible for bovine leukocyte adhesion deficiency). A gene substitution model was used to estimate the additive genetic effects of alleles of the three molecular markers on estimated breeding value (EBV) for mastitis measures. Pearson correlation coefficients between EBV for immunological assays and EBV for mastitis measures were computed. Molecular markers explained up to 40% of the variation in EBV for measures of mastitis. The presence of allele DRB3.2*16 was associated with higher EBV for SCS. Allele DRB3.2*8 was associated with increased EBV for clinical mastitis, as was the IgG2b allele and the normal CD18 allele. Alleles DRB3.2*11, *23, IgG2a, and the recessive allele for bovine leukocyte adhesion deficiency were associated with decreased clinical mastitis. A positive genetic association was found between allele DRB3.2*24 and EBV for IMI by major pathogens and between DRB3.2*3 and IMI by minor pathogens. Several correlations between EBV for immunological assays and EBV for mastitis measures were significantly different from 0. Cows with low EBV for SCS tended to have neutrophils that had greater functional ability at maximal immunosuppression, low serum IgG1, and high numbers of circulating mononuclear cells. Immunological parameters, including physiological and molecular markers, are useful aids to understand the genetics of resistance to mastitis.

Genetic association of bovine lymphocyte antigen DRB3 alleles with immunological traits of Holstein cattle.

The associations between alleles at the BoLA (bovine lymphocyte antigen) DRB3 locus and 20 indicator traits of innate and adaptive immunity were investigated. Periparturient Holsteins (n = 127) were genotyped at the BoLA DRB3 locus using polymerase chain reaction and restriction fragment length polymorphism. Twenty-two alleles were observed in the study population, and frequencies ranged from 21 to < 1%. The same cattle were tested for a total of 20 innate and adaptive immunity traits, including lymphocyte response to mitogens (proliferative responses and Ig secretion), serum Ig, complement and conglutinin concentrations, total leukocyte count, and selected assays for neutrophil function. Models with gene substitution effects were used to investigate associations between BoLA DRB3 alleles and each of the immunological variables. Significant associations were found with 13 of the leukocyte functions tested. The number of immune parameters with significant associations with any allele ranged from 0 (with alleles DRB3.2*23 and DRB3.2*27) to 7 (with DRB3.2*8). The immunological parameter that had the most associations with alleles was serum IgG2 concentration with 6 alleles. One group of 4 alleles (representing 46% of the total allele frequency) was uniformly associated with increased IgM and complement and decreased mononuclear cell numbers. Thus, we demonstrated that the BoLA DRB3 genotype can influence measures of innate and adaptive immunity.

Effects of left displaced abomasum on test day milk yields of Holstein cows.

A linear model was used to estimate the effects of left displaced abomasum on 57,383 test day milk yields collected between January 1, 1992 and December 31, 1993 in New York State from 12,572 Holstein cows in parity < or = 6. Twenty-five index categories were created to differentiate among milk collected between calving and up to 60 d after diagnosis of left displaced abomasum, milk collected > 60 d after diagnosis of left displaced abomasum, milk collected before or after diagnosis of cows with any diseases other than left displaced abomasum, and milk collected from cows that remained healthy for the entire lactation. For each parity, the statistical model included fixed effects (management, age at calving, length of dry period, previous 305-d milk yield, stage of lactation, year and season of sampling date, disease index, and DHIA SCS at milk sampling) and random effects (permanent and temporary environments) on test day milk yields. From calving to 60 d after diagnosis, cows with left displaced abomasum yielded on average 557 kg less milk than did cows without left displaced abomasum; 30% of losses occurred before diagnosis. Milk loss increased as parity and productivity increased. Cows with left displaced abomasum were nearly twice as likely to have another disease than were cows without left displaced abomasum.

Effects of parity, breed, herd-year, age, and month of kidding on the milk yield and composition of dairy goats in Belgium.

SUMMARY: Lactation records, collected from 1989 to 1995, for Belgian dairy goats were examined for effects of parity, breed, herd-year, age at kidding, and month of kidding on milk, fat and protein yields, and on fat and protein percentages. Total lactation yields were computed from test-day yields, using Fleischmann's method. For lactations longer than 250 days, yields were truncated at 250 days. Partial lactations (25 days minimum) were extended to 250 days, and fat and protein percentages were deducted from the yields. After editing a total of 2997 lactation records of 2562 does remained. The parity classifications were first and second or greater; kidding seasons were defined as December, January, February, March, April to July, and August to November. The breed groups were Anglo-Nubian, Chamoisee, Saanen, and crossbreeds. The data were analysed within parity with a fixed-effect model that included effects of herd-year, breed, month of kidding, and age at Kidding. All effects in the model were significant on the traits measured for parities, except the age effects on fat percentage and of month of kidding on protein percentage at first parity. Saanen had the highest milk yield at first parity; at later parities, the Chamoisee produced most. Yield increases were large between first and later parities. Does freshening in February produced more milk than those that freshened at other months. Monthly variations in fat and protein percentages across parities, although mostly significant, were small. Yields increased with age at first parity; for later parities, they increased until 25-30 months for milk and until 40 months for fat and protein, and then decreased. Fat and protein percentages increased up to 15-17 months of age, and then decreased to the lowest values around 25 months; this was followed by an increase around 32 months, which tended to stabilize until 50 months. The ranges of R(2) associated with effects were: for herd-year, 43-52% and 22-36%; for breed, 3-11% and 4-12%; for month of kidding 6-10% and 2-5%; for age at kidding 3-7% and 1-17%-for first and second, and later parities, respectively. RÉSUMÉ: L'influence des effets numéro de lactation, race, troupeau-année, mois et âge à la mise bas a été étudiée sur la production des chèvres laitières de Belgique, enregistrées de 1989 à 1995. Les productions totales réalisées ont été calculées à partir des résultats aux contrôles. Pour les lactations supérieures à 250 j, les productions ont été tronquées à 250 j. Les lactations partielles (minimum 25 j) ont été extrapolées à 250 j. Au total, 2997 lactations à 250 jours étaient disponibles, pour 2562 chèvres des races Anglo-nubienne, Chamoisée, Saanen et croisées. Les données ont été réparties en 2 parités (1 et ≥ 2), 6 mois de mise bas (décembre, janvier, février, mars, avril à juillet et août à novembre). Le modèle d'analyse intra-parité comprenait les effets fixes troupeau-année, race, mois et âge à la mise bas. Tous les effets ont été significatifs sur les paramètres étudiès à l'exception de l'effet de l'âge sur les taux de matière grasse et de l'effet du mois de mise bas sur les taux de protéine, en parité 1. En première lactation, les Saanen ont enregistré les plus hautes productions de lait tandis qu'en seconde lactation, ce furent les Chamoisées. L'accroissement des productions a été important entre première et seconde lactation. Les mises bas de février ont enregistré les productions les plus élevées. Les taux de matières grasse et protéique ont accusé de faibles variations saisonnières. A l'exception de la parité 1 oú elles augmentaient, les productions de lait, matière grasse et protéine tendaient à diminuer avec l'âge. Une diminution des taux avec l'âge a été observée. Les données ont permis d'établir de nettes différences de production entre la première et la seconde parité à âges identiques; cependant leur nombre encore limité n'a pas permis d'apprécier les effets d'intéractions âge × parité, âge × mois de mise bas on troupeau-année × mois de mise bas. Les valeurs de R(2) associées aux effets ont été: pour troupeau-année 43 à 52% et 22 à 36%, pour race 3 à 11% et 4 à 12%, pour mois de mise bas 6 à 10% et 2 à 5%, et pour l'âge à la mise bas 3 à 7% et 1 à 17%, respectivement pour les primipares et les multipares. ZUSAMMENFASSUNG: Einfluss von Laktationsnummer, Rasse, Herde-Jahr, Wurfmonat und -jahr auf Milch-, Fett- und Proteinertag und Fett- und Milchprozente bei belgischen Milchziegen Die Gesamtlaktationsleistungen wurden aus den Stichtagsleistungen mit der Methode von Fleischmann berechnet. Laktationsleistungen von mehr als 250 Tagen Dauer wurden auf 250 Tage standardisiert, Teillaktationen (Minimum 25 Tage) auf 250 Tage extrapoliert und Fett- und Proteingehalte aus den Leistungen abgeleitet. Von 2562 Ziegen der Rassen Anglo-Nubier, Gemsfarbige Gebirgsziege, Saanen sowie Kreuzungen standen Daten von 2997 Laktationen zu 250 Tagen zur Verfügung. Die Daten wurden in erste und folgende Laktationen sowie nach Wurfmonat (Dezember, Januar, Februar, März, April bis Juli, August bis November) gruppiert. Die Analyse der Daten wurde mit einem Modell fixer Effekte vorgenommen, das Herde-Jahr, Rasse, Wurfmonat und- alter beinhaltete. Ausser den Effekten des Wurfalters auf Fettprozente und des Wurfmonats auf Proteinprozente in der ersten Laktation waren alle untersuchten Parameter signifikant. In der ersten Laktation erzielten die Saanenziegen, in den folgenden Laktationen die Gemsfarbigen die höchste Milchleistung. Zwischen der ersten und den folgenden Laktationen stieg die Milchleistung stark an. Würfe im Februar verzeichneten die höchste Milchleistung. Monatliche Variationen in Fett- und Proteinprozenten über alle Laktationen waren zwar meist signifikant, aber von geringem Ausmass. Die Leistungen erhöhten sich in der ersten Laktation mit steigendem Alter. In den folgenden Laktationen stiegen die Milchleistung bis zum Alter von 25 bis 30 Monaten und die Fett- und Proteinerträge bis 40 Monate; danach nahmen die Leistungen ab. Fett- und Proteinprozente erhöhten sich bis zu einem Alter von 15 bis 17 Monaten und nahmen dann ab, um bei 25 Monaten die tiefsten Werte zu erreichen. In der Folge nahmen die Werte ab 32 Monaten wieder zu, um sich bei 50 Monaten zu stabilisieren. Die Bereiche der R(2) der verschiedenen Effekte lagen für Herde-Jahr bei 43 bis 52%, die Rasse bei 3 bis 11%, den Wurfmonat bei 6 bis 10% bzw. 2 bis 5% (erste und folgende Laktationen) und das Wurfalter bei 3 bis 7% bzw, 1 bis 17%.

Mastitis of periparturient Holstein cattle: a phenotypic and genetic study.

Environmental and genetic factors affecting somatic cell scores, clinical mastitis, and IMI by minor and major pathogens were studied on 137 periparturient Holstein cows selected for milk production. Environmental effects were obtained by generalized least squares and logistic regression. Genetic parameters were from BLUP and threshold animal models. Lactation number affected the number of quarters with clinical mastitis and the number of quarters infected with minor pathogens. The DIM affected somatic cell score and number of quarters infected with major pathogens. Heritabilities for all mastitis indicators averaged 10%, but differences occurred among the indicators. Correlations between breeding values of the number of quarters infected with minor pathogens and the number infected with major pathogens were antagonistic and statistically significant.

Two retroviral infections of periparturient Holstein cattle: a phenotypic and genetic study.

Environmental and genetic factors affecting prevalences of antibodies to bovine leukosis virus and bovine immunodeficiency-like virus were studied on 137 periparturient Holstein cows selected for milk production. Environmental effects were obtained by logistic regression, and genetic parameters were determined using threshold animal models. Cows selected for high predicted transmitting ability for kilograms of milk fat plus protein had the highest prevalence of antibodies to bovine immunodeficiency-like virus and the lowest prevalence of antibodies to bovine leukosis virus. Heritability estimates for susceptibility to retroviral infections were close to zero.

Study of immunological dysfunction in periparturient Holstein cattle selected for high and average milk production.

Data from twenty assays of traits associated with innate and adaptive immunity were evaluated from 137 periparturient Holstein cows. These cows had been selected through planned matings for four different levels of milk production (high and average pounds of milk, and high and average pounds of milk fat plus protein). For up to seven generations, the genetic lines were produced by mating females of each line to sires of corresponding merit. With the exceptions of neutrophil ingestion of Staphylococcus aureus and directed migration, all assays measuring neutrophil functions were depressed beginning 2 to 3 weeks before calving through 3 weeks after calving. Serum concentrations of immunoglobulin G1 decreased while those of immunoglobulin G2 increased around calving time. Serum complement and conglutinin concentrations decreased before calving and reached a minimum around calving time. Cows selected for high milk production (pounds of milk and pounds of milk fat plus proteins) had significantly higher (P < 0.10) numbers of circulating neutrophils and mononuclear cells, had higher (P < 0.10) neutrophil resting chemiluminescence and higher (P < 0.10) neutrophil directed migration than cows with average production potentials. There were significant (P < 0.001) sire progeny group differences for most traits associated with the immune system that we tested. These results can be considered encouraging, in that selection for high milk yield did not produce unfavorable correlated responses in the functional capacity of immune function traits, and that there is sufficient genetic variation in these immunological traits among sires of high genetic merit for milk production to potentially improve the immunocompetence of periparturient cows through planned mating experiments.

Effects of clinical ketosis on test day milk yields in Finnish Ayrshire cattle.

A linear model for repeated measurements was used to estimate the effects of clinical ketosis on 722,198 test day milk yields collected from September 1, 1985 to January 31, 1988 on 60,851 Finnish Ayrshire cows of parity < 7. An index was created to differentiate among milk collected within 17 d following diagnosis of ketosis, milk collected before or > 17 d after diagnosis, and milk collected on nonketotic cows. For each parity separately, the statistical model included fixed effects (ketosis, calving season, year and season of milk sampling, and stage of lactation) and random effects (herd and permanent and temporary environments) on test day milk yields. The pattern underlying correlations between temporary environmental effects was accommodated in the statistical model. Compared with those for nonketotic cows, lactation curves of cows with ketosis showed a depression in early lactation; estimated milk loss was 44.3 kg for 17 d after diagnosis. The 305-d milk yield of cows diagnosed with ketosis was estimated to be 141.1 kg higher than that of cows free of ketosis. Although milk losses occurred after ketosis, ketotic cows yielded more milk over the entire lactation than did nonketotic cows; and yields would have been even higher if cows had not had ketosis.

Immunological parameters of periparturient Holstein cattle: genetic variation.

The genetic variability of blood neutrophil functions, lymphocyte blastogenic responses to mitogens, serum Ig concentrations, and serum complement and conglutinin activities was investigated from 35 d prepartum to 35 d postpartum for 137 Holstein cows. Periparturient cows experience an immunosuppression of various immunologic parameters at calving. Heritability estimates were obtained before, during, and after the episodes of immunosuppression. Significant genetic variability occurred in the periparturient changes for total number of neutrophils, neutrophil chemokinesis, assays of the neutrophil respiratory burst associated with phagocytosis (cytochrome c reduction, chemiluminescence, and iodination), serum concentrations of IgG1, IgG2, and IgM, and serum homolytic complement activity. This variability implies that immune profiles could be used for the selection of cattle with improved innate immune response without adverse effects on milk productivity. These results should be considered tentative, however, because the number of observations included in the data were limited.

Effect of season, genetic line, and sire on growth concentrations of somatotropin in serum of Holstein cows in early lactation.

To determine the effect of selection for milk yield on somatotropin concentrations, blood samples were collected from 128 Holstein cows whose sires had either high or average predicted differences for milk. The phenotypic difference in milk yield between the high and average yielding groups was 1726 kg on a 305-d mature equivalent basis. At 37 +/- 7 d postpartum, four blood samples were taken from each cow at 1-h intervals beginning at 0800 h. The statistical model contained genetic line, sires within line, and season as whole-plot effects and time of sampling as a subplot effect. The concentration of growth hormone was significantly higher in the high yielding group (1.89) than in the average yielding group (1.49). Cows sampled in summer had the highest concentrations of growth hormone, whereas cows sampled in spring had the lowest concentrations. Sires did not significantly influence the somatotropin concentration of their progeny. In a second analysis with seven extreme outliers (peak values) deleted, the variation of somatotropin concentrations in the cows was significantly accounted for by their sires. Repeatability of somatotropin concentration was .49 within cows.

Comparison of natural transmission of bovine leukemia virus in Holstein cows of two genetic lines selected for high and average milk production.

One hundred and fifty lactating Holstein cows from 2 genetic lines selected for high and average milk production were used in the study. Sera from 6 annual herd tests were analyzed by agar-gel immunodiffusion test for antibodies to bovine leukemia virus. Odds of being seropositive were analyzed by use of stepwise and backward logistic regression procedures. Analysis within birth year revealed that estimated in odds increased by 0.19/year of age among cows of the high genetic line and by 0.43 among cows of the average genetic line. This was accompanied by a more important cohort effect among high producers than among average producers.