The relationship between diet, plasma glucose, and cancer prevalence across vertebrates

Abstract

Birds have higher plasma glucose concentrations but lower cancer prevalence than other vertebrates. However, this inverse relationship between glucose and cancer may not hold within vertebrate groups. Given that diet affects blood sugar levels, and carnivores have higher cancer risk than herbivores, we also examined whether diet correlates with plasma glucose concentrations. We collected diet, mean plasma glucose concentration, and neoplasia data for up to 273 vertebrate species from existing databases. Across vertebrates, mean plasma glucose concentration negatively correlated with cancer prevalence, but that was mostly driven by differences in mean plasma glucose concentration and cancer prevalence between birds, mammals, and reptiles. Mean plasma glucose concentration was not correlated with diet across vertebrates nor with cancer prevalence within birds, mammals, or reptiles. Primary carnivores had higher neoplasia prevalence than herbivores when controlling for domestication. A hypothetical explanation for our results may be the evolutionary loss or downregulation of genes related to insulin-mediated glucose import in bird cells. This may have led to higher mean plasma glucose concentration, lower intracellular glucose concentrations in the form of glycogen, and production of fewer reactive oxygen species and inflammatory cytokines, potentially contributing to lower neoplasia prevalence in extant birds compared to mammals and reptiles.

Fig. 1. Variation in mean plasma glucose concentration in different vertebrate groups, and its relationship with overall cancer prevalence.

A Mean plasma glucose concentrations are relatively higher in Aves than Mammalia and Reptilia (Kruskal–Wallis test; P-value < 0.0001; degrees of freedom = 2). There is no correlation between mean plasma glucose concentration and malignancy prevalence across tissues for B 66 mammalian species, C 47 bird species, and D 33 reptile species. The statistical tests were two-sided. P-values have not been adjusted for FDR corrections. The P-values of pgls analyses that passed the FDR correction are noted with an asterisk in Supplementary Data 3–5. The outlier species (Rosner’s test) with the highest malignancy prevalence are: Didelphis virginiana in panel (B) Gallus gallus in panel (C) and Lampropeltis triangulum, Lampropeltis getula, and Pituophis catenifer in panel (D) from top to bottom. PGLS phylogenetic generalized least squares. For information on how to generate this figure, please see Source Data file.

Fig. 2. The relationship between mean plasma glucose concentration and diet across vertebrates.

A Trophic level is not correlated with mean plasma glucose concentrations across 242 species (PGLS: P-value ≥ 0.05; Supplementary Data 1). B There was no correlation between the percentage of fruit, plants, invertebrates, seeds, or endothermic vertebrates (Endotherms), and mean plasma glucose concentrations for 20, 34, 27, 17, or 10 species, respectively, after correcting for multiple testing. C The percentage of plant-based foods or animal-based foods in a species’ diet was not correlated with mean plasma glucose concentrations for 47 or 38 species, respectively (PGLS: P-value > 0.05). D Diet type is not correlated with mean plasma glucose levels for 67 species (PGLS: P-value > 0.05; Supplementary Data 1). The horizontal black line in each diet category shows the mean plasma glucose concentration in that trophic (plot A) or diet category (plot D). In D, herbivores refer to animals that have a 100% plant diet, omnivores refer to animals that have >0% plant and meat diet, and carnivores refer to animals that have a 100% meat (invertebrate or vertebrate) diet. The statistical tests were two-sided. P-values have not been adjusted for FDR corrections. The P-values of pgls analyses that passed the FDR correction are noted with an asterisk in Supplementary Data 1. Each dot shows the mean plasma glucose concentration and diet category of one species. N shows the number of species per diet category (plots A & D). We added minimal jitter in the plots in order to better visualize individual data points. PGLS phylogenetic generalized least squares. For information on how to generate this figure, please see Source Data file.

Fig. 3. Relationships between overall cancer prevalence and diet across vertebrates.

A Trophic level is not correlated with malignancy prevalence across tissues for 140 species, even when controlling for variations in glucose concentrations in their plasma, domestication, or both plasma glucose concentrations and domestication, and correcting for multiple testing (PGLS: Supplementary Data 7; P-values < 0.05 of between-trophic level comparisons are shown). B There is no correlation between the percentage of fruit, plants, invertebrates, seeds, or endothermic vertebrates (Endotherms) in a species’ diet and malignancy prevalence across tissues for 20, 34, 27, 17, or 10 species, respectively, after corrections for multiple testing. C The percentage of plant-based food in a species’ diet is not correlated with malignancy prevalence across tissues for 47 species (PGLS: P-value > 0.05). Also, there is no correlation in the percentage of animal-based food in a species’ diet and malignancy prevalence across their tissues for 38 species (PGLS: P-value > 0.05). D Diet type is not correlated with malignancy prevalence across 65 species (PGLS: P-value ≥ 0.05). The horizontal black line in each trophic level (plot A) or diet category (plot D) shows the median malignancy prevalence across tissues in that category. In D, herbivores refer to animals that have a 100% plant diet, omnivores refer to animals that have >0% plant and meat diet, and carnivores refer to animals that have a 100% meat (invertebrate or vertebrate) diet. Each dot shows the malignancy prevalence across tissues and diet category of a species. N shows the number of species per diet category (plots A & D). We added minimal jitter in the plots in order to better visualize individual data points. The statistical tests were two-sided. P-values have not been adjusted for FDR corrections. The P-values of pgls analyses that passed the FDR correction are noted with an asterisk in Supplementary Data 7. PGLS: phylogenetic generalized least squares. w/o without. For information on how to generate this figure, please see Source Data file.

Fig. 4. Relationship between gastrointestinal cancer prevalence and diet across vertebrates.

A Trophic level is not correlated with gastrointestinal malignancy prevalence across 94 species, with or without controlling for domestication, plasma glucose concentrations, both domestication and plasma glucose concentrations, and after correcting for multiple testing (Supplementary Data 7; P-values < 0.05 of between-trophic level comparisons are shown). B There is no correlation between the percentage of fruit, plants, invertebrates, or endothermic vertebrates (Endotherms) in a species’ diet and gastrointestinal malignancy prevalence for 15, 19, 18, or 10 species, respectively, after corrections for multiple testing (Supplementary Data 7). C There is no correlation in the percentage of plant-based or animal-based food in a species’ diet and gastrointestinal malignancy prevalence for 27 and 27 species, respectively, after corrections for multiple testing (Supplementary Data 7). D Diet type is not correlated with gastrointestinal malignancy prevalence across 42 species (PGLS: P-value ≥ 0.05) after corrections for multiple testing (Supplementary Data 7). The horizontal black line in each trophic level (plot A) and diet category (plot D) shows the median gastrointestinal malignancy prevalence in that category. In D, herbivores refer to animals that have a 100% plant diet, omnivores refer to animals that have >0% plant and meat diet, and carnivores refer to animals that have a 100% meat (invertebrate or vertebrate) diet. Each dot shows the gastrointestinal malignancy prevalence and diet category of a species. N shows the number of species per category (plots A & D). We added minimal jitter in the plots in order to better visualize individual data points. The statistical tests were two-sided. P-values have not been adjusted for FDR corrections. The P-values of pgls analyses that passed the FDR correction are noted with an asterisk in Supplementary Data 7. PGLS: phylogenetic generalized least squares. w/o without. For information on how to generate this figure, please see Source Data file.