Bayesian structural equation models for inferring relationships between phenotypes: a review of methodology, identifiability, and applications

Structural equation models provide a general statistical modelling technique for estimating and testing relationships among variables. Such relationships are often not revealed by standard linear models, but are of importance for understanding mechanisms underlying e.g., production-related diseases, such as mastitis. This paper gives a review of Bayesian structural equation models concerning methodology and identifiability, focused on animal breeding and genetics modelling. Applications of this type of methods in animal breeding are also reviewed critically, with discussion on advantages and disadvantages of these approaches.     

Effectiveness of aquaflor (50% florfenicol) to control mortality associated with Streptococcus iniae in freshwater-reared subadult sunshine bass

We conducted a field trial to evaluate the effectiveness of Aquaflor (50% florfenicol) for controlling mortality associated with Streptococcus iniae in freshwater-reared subadult sunshine bass (female white bass Morone chrysops X male striped bass M. saxatilis). Bacterial samples collected from moribund fish representing a reference population were presumptively identified microbiologically and were later confirmed to be S. iniae by biochemical characterization and polymerase chain reaction. The trial comprised a 1-d acclimation period, 10-d treatment period, and 14-d posttreatment period. During the treatment period, Aquaflor-medicated feed was administered to treated tanks (N = 3) at a target dose of 10 mg of florfenicol x kg of fish(-1) x d(-1), and nonmedicated feed was administered to control tanks (N = 3). At the end of the posttreatment period, mean (+/- SD) cumulative mortality in treated tanks (9 +/- 11%) was significantly (P = 0.040) less than that in control tanks (52 +/- 13%). Analysis of medicated feed samples revealed that treated tanks had received an actual dose of 8.3 mg florfenicol x kg fish(-1) x d(-1) (83% of target). No florfenicol was detected in control feed samples. Although the actual florfenicol dose administered to treated tanks was less than the target dose, the trial was accepted by the U.S. Food and Drug Administration Center for Veterinary Medicine as demonstrating the efficacy of Aquaflor to control mortality associated with S. iniae in cultured sunshine bass populations.     

Age-related alterations to immune parameters in Labrador retriever dogs

In order to assess age-related changes in the immune status of Labrador retriever dogs, leukocyte phenotypes, lymphocyte proliferative capacity, and serum antibody levels were measured in four cohorts of dogs, ranging from 2 to 10 years of age. Absolute numbers of white blood cells, lymphocytes, monocytes, granulocytes, and CD3+, CD4+, CD8+ and CD21+ lymphocytes significantly decreased with increasing age. Relative percentages of lymphocytes and CD4 cells were significantly decreased, and relative percentages of granulocytes and CD8 cells significantly increased, with age. The CD4:CD8 ratio showed a significant age-related decrease. Proliferative responses of T-cells to mitogens in whole-blood cultures either increased (Concanavalin A) or remained the same (phytohemagglutinin) with age when data was normalised to allow for differences in responding cell number. Similarly, normalised data of proliferative response to anti-CD3 stimulation together with phorbol 12-myristate 13-acetate showed an age-related increase. Serum levels of total IgA significantly increased with age whereas total IgG levels remained unchanged. These observations illustrate a significant change to a number of immune parameters with age. However, further work is required to determine whether the differences reported here are sufficient to cause overt or functional immune senescence in Labrador retriever dogs.     

Lack of glucokinase regulatory protein expression may contribute to low glucokinase activity in feline liver

In most mammals, glucokinase (GK) acts as a hepatic “glucose sensor” that permits hepatic metabolism to respond appropriately to changes in plasma glucose concentrations. GK activity is potently regulated by the glucokinase regulatory protein (GKRP), which is encoded by the GCKR gene. GKRP binds GK in the nucleus and inhibits its activity. GK becomes active when it is released from GKRP and translocates to the cytosol. Low glucokinase (GK) activity is reported to be a principal feature of feline hepatic carbohydrate metabolism but the molecular pathways that regulate GK activity are not known. This study examined the hypothesis that species-specific differences in GKRP expression parallel the low GK activity observed in feline liver. Hepatic GKRP expression was examined using RT-PCR, immunoblot, and confocal immunomicroscopy. The results show that the GCKR gene is present in the feline genome but GCKR mRNA and the GKRP protein were absent in feline liver. The lack of GKRP expression in feline liver indicates that the low GK activity cannot be the result of GKRP-mediated inhibition of the GK enzyme. However, the absence of the permissive effects of GCKR expression on GK expression and activity may contribute to reduced GK enzyme activity in feline liver. The study results show that the cat is a natural model for GCKR knockout and may be useful to study regulation of GCKR expression and its role in hepatic glucose-sensing and carbohydrate metabolism.     

The use of eutectic mixture of lidocaine and prilocaine in mice (Mus musculus) for tail vein injections

ObjectiveTo investigate a topical local anesthesia technique as a means to prevent and/or diminish pain in mice in a laboratory setting associated with tail vein injections performed by personnel in training.Study designProspective, randomized experimental trial.AnimalsThirty six adult female, 23–28 g CD-1 mice from an in-house training colony. They were acclimated to routine training and handling classes.MethodsEutectic mixture of local anesthetics (EMLA) cream (2.5% lidocaine/2.5% prilocaine) or a bland ointment control (n = 18) was applied on the tail prior to intravenous injection. The injections were performed by novices, who had never attempted the procedure, and experienced personnel. All participants were blinded to treatment groups. Three injection attempts were allowed per animal. The mice were observed and scored by blinded evaluators for behavioral and physiological changes, including respiratory rate, vocalization, tail flick, and escape behaviors, during and after the injection.ResultsThis study demonstrates that aversive behaviors induced by lateral tail vein injection were not changed by the preemptive application of EMLA cream. The aversive behaviors associated with lateral tail vein injection were significantly affected by the number of injection attempts and the individual's experience level.Conclusions and clinical relevanceTopical EMLA cream did not reduce signs of aversive reaction to tail vein injection and thus we did not find support for its use in mouse training programs for tail vein injections.     

Investigation of the Role of Androstenedione-19-oic Acid in the Presence of 19-Norandrostenedione in Intact Male Horse Plasma Using Liquid Chromatography–Tandem Mass Spectrometry

A liquid chromatography–tandem mass spectrometry method was developed to confirm the presence of androstenedione-19-oic acid in intact male equine plasma and to show the source of 19-norandrostenedione in equine plasma. Androstenedione-19-oic acid was recovered from acidified plasma by liquid–liquid extraction using methyl tert-butyl ether and separated on an Ace 5 C₈ column. A triple quadrupole mass spectrometer was used to detect the analytes in negative electrospray ionization mode. Limits of detection, quantification, and confirmation of the method were 0.1, 0.5, and 1.0 ng/mL, respectively. The linear dynamic range of quantification was 0.5–50 ng/mL. The presence of androstenedione-19-oic acid was confirmed in all plasma samples obtained from intact male horses but not those from gelded and female horses; the average concentration was 3.1 ± 1.6 ng/mL, suggesting androstenedione-19-oic acid is an endogenous compound only in intact male horse plasma samples. The conversion of androstenedione-19-oic acid to 19-norandrostenedione in equine plasma was demonstrated by spiking androstenedione-19-oic acid into blank plasma and monitoring the generation of 19-norandrostenedione and its increase in concentration during storage. Results indicated that androstenedione-19-oic acid was readily converted into 19-norandrostenedione; the higher the storage temperature, the faster the conversion. The conversion was not affected by the types of plasma samples collected from gelded and female horses or by anticoagulants used in blood collection to harvest plasma. Compared with other matrices such as water, methanol, and phosphate-buffered saline, the conversion of androstenedione-19-oic to 19-norandrostenedione in equine plasma was faster, suggesting that there is an unknown factor(s) in equine plasma that enhances the conversion.     

Effect of post-insemination progesterone supplementation on pregnancy rate in dairy cows

Progesterone plays an important role in maintenance of pregnancy. It is hypothesized that insufficient progesterone early in pregnancy may result in embryonic loss, and that supplemental progesterone would decrease pregnancy loss in dairy cows. In Experiment 1, 84 cows and 16 heifers from a single dairy operation were selected randomly. Within each age category, controlled internal drug release (CIDR) devices were inserted into the vagina of every other female on Day 4 post-insemination and removed on Day 18 post-insemination. Transrectal ultrasonography was performed to determine pregnancy at 4 time periods [days 30 to 37 (week 5), days 44 to 51 (week 7), days 58 to 65 (week 9), and days 86 to 93 (week 13)]. Progesterone supplementation had no effect on pregnancy rate. In Experiment 2, there were no differences in progesterone concentrations between cows that did and did not receive a CIDR. Further, cows receiving CIDR devices did not have an increase in circulating progesterone concentrations 30 min or 1 h after CIDR insertion. It appears that progesterone supplementation does not increase circulating levels of progesterone in the early pregnant lactating dairy cow. Alterative methods to influence progesterone concentrations and/or early embryonic loss need to be investigated.