Search

Animals produce ATP through oxidative phosphorylation using oxygen, but cellular energy can also be obtained through glycolysis when oxygen is not present at sufficient levels. Although most mammals of larger body mass have longer life spans, small dog breeds tend to outlive large breeds. Primary fibroblast cells from larger breeds of dogs have previously been shown to have increased dependency on glycolytic phenotypes across their lifespan. Different levels of activity of the glycolytic enzymes pyruvate kinase (PK), lactate dehydrogenase (LDH), and phosphoenolpyruvate carboxykinase (PEPCK) may provide insight to a mechanism that leads to the different metabolic phenotype observed in different sized breeds as they age. In this study, 1) we measured the activities of PK, LDH, and PEPCK in primary fibroblasts from dogs of different breed sizes and age classes and 2) measured the activities of PK and LDH in plasma from dogs of different breed sizes and age classes. We found that there was no significant relationship between body mass and PK, LDH and PEPCK activity in primary fibroblasts. Further, there were not significant differences with activity in these enzymes for old dogs compared to young dogs. In plasma, we found a negative correlation between PK activity and body mass and no relationship between LDH activity and body mass. There was a negative relationship between LDH activity and age in dogs. Further, while a negative correlational relationship between PK activity and age was only marginal, a best subsets regression model demonstrated a significant marginal effect of age on PK activity. PK and LDH may provide intermediates for other metabolic pathways in small breeds. However, large breed dogs may demonstrate a deficiency in metabolism at the PK level, a cellular metabolic pathway that may potentially aid in tumor progression.

Small breed dogs have longer lifespans than their large breed counterparts. Previous work demonstrated that primary fibroblast cells isolated from large breed young and old dogs have a persistent glycolytic metabolic profile compared with cells from small breed dogs. Here, we cultured primary fibroblast cells from small and large, young and old dogs and treated these cells with three commercially available drugs that show lifespan and health span benefits, and have been shown to reduce glycolytic rates: rapamycin (rapa), resveratrol (res) and metformin (met). We then measured aerobic and anaerobic cellular respiration in these cells. We found that rapa and res increased rates of non-glycolytic acidification in small and large breed puppies and basal oxygen consumption rates (OCR) in small and large breed puppies. Rapa increased proton leak and non-mitochondrial respiration in small and large breed puppies. Maximal respiration was significantly altered with rapa treatment but in opposing ways: large breed puppies showed a significant increase in maximal respiration when treated with rapa, and small old dogs demonstrated a significant decrease in maximal respiration when treated with rapa. In opposition to rapa treatments, met significantly decreased basal OCR levels in cells from small and large breed puppies. Our data suggest that rapa treatments may be metabolically beneficial to dogs when started early in life and more beneficial in larger breeds. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11357-021-00349-7.

We propose a generative classification model that extends Quadratic Discriminant Analysis (QDA) (Cox in J R Stat Soc Ser B (Methodol) 20:215–242, 1958) and Linear Discriminant Analysis (LDA) (Fisher in Ann Eugen 7:179–188, 1936; Rao in J R Stat Soc Ser B 10:159–203, 1948) to the Bayesian nonparametric setting, providing a competitor to MclustDA (Fraley and Raftery in Am Stat Assoc 97:611–631, 2002). This approach models the data distribution for each class using a multivariate Polya tree and realizes impressive results in simulations and real data analyses. The flexibility gained from further relaxing the distributional assumptions of QDA can greatly improve the ability to correctly classify new observations for models with severe deviations from parametric distributional assumptions, while still performing well when the assumptions hold. The proposed method is quite fast compared to other supervised classifiers and very simple to implement as there are no kernel tricks or initialization steps perhaps making it one of the more user-friendly approaches to supervised learning. This highlights a significant feature of the proposed methodology as suboptimal tuning can greatly hamper classification performance; e.g., SVMs fit with non-optimal kernels perform significantly worse.

A discrete approximation to the Polya tree prior suitable for latent data is proposed that enjoys surprisingly simple and efficient conjugate updating. This approximation is illustrated in two applied contexts: the implementation of a nonparametric meta-analysis involving studies on the relationship between alcohol consumption and breast cancer, and random intercept Poisson regression for Ache armadillo hunting treks. The discrete approximation is then smoothed with Gaussian kernels to provide a smooth density for use with continuous data; the smoothed approximation is illustrated on a classic dataset on galaxy velocities and on recent data involving breast cancer survival in Louisiana.