Caitlin J. Andrews’s research while affiliated with The University of Sydney and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (4)


of global protein supplies (kcal/capita/day) and correlations within the available data
A Average log ratio of animal-based to plant-based protein supplies for each country, from 1961–1990, and B 1991 –2018, using all available data; grey shows an absence of data. (C) Supplies of plant- and animal-based protein as a function of year, D GDP per capita, (E) fat supplies (kcal/capita/day), and (F) carbohydrate supplies (kcal/capita/day). C–F The individual data points show the raw data values. The red, green, and blue smoothed lines represent the predicted values for ABP, PBP, and total protein, respectively, based on year (C), GDP (D), fat supplies (E), and carbohydrate supplies (F). Predictions were obtained using generalized additive mixed models, with random effects for countries. The shaded areas around the lines represent the 95% confidence intervals centred on these predictions. The smooth term significance is detailed in Supplementary Table 1 to Supplementary Table 12. GDP gross domestic product per capita (expressed in international-$ at 2011 prices). A,B Map data from Natural Earth, facilitated by rnaturalearth R package⁹⁴.
Predicted effects of macronutrient supplies (in terms of kcal/capita/day) on life table survivorship to age 5 (l_5)
Survivorship outcomes were predicted based on 2017 global nutrient supply, GDP, and population data, using the GAMM selected as the best approximating model for the entire dataset. A–C shows the effects of animal-based protein, plant-based protein, and fat supplies on l5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${l}_{5}$$\end{document} in females. D–F shows the effects of animal-based protein, plant-based protein, and fat supplies on l5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${l}_{5}$$\end{document} in males. The colour map and contour lines on each surface shows predicted l5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${l}_{5}$$\end{document} associated with the aligning proportions of protein supplies on the x\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$x$$\end{document} and y\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$y$$\end{document} axes. Survivorship values increase from blue to red, as shown by the colour bar scale. Moving from (A–C) and from (D–F) across each row of surfaces, the quantity of fat supplies increases. The surfaces for low (A, D), median (B, E) and high (C, F) fat show predicted effects at the first, second, and third quartiles of global fat supplies in 2017, respectively. A–F assume the median carbohydrate supply and GDP for 2017 (1571 kcal/capita/day; 2011-Intl$30902). Total non-protein energy (NPE) is the sum of energy supplies from fat and carbohydrate for each surface. The grey dashed isoline indicates a constant total protein energy (PE) supply at 400 kcal/capita/day on all surfaces (A–F). The smooth terms for macronutrient supply were statistically significant in both sexes (Supplementary Table 16). GAMM generalized additive mixed model, GDP gross domestic product per capita (in international-$ at 2011 prices).
Predicted effects of macronutrient supplies (in terms of kcal/capita/day) on life table survivorship to age 60 (l60)
Survivorship outcomes were predicted across the 2017 global nutrient supply space, as described in Fig. 2. A–C and D–F show female and male l60\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${l}_{60}$$\end{document} outcomes, respectively. The colour map and contour lines show predicted l60\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${l}_{60}$$\end{document} survivorship, as described in Fig. 2. Survivorship values increase from blue to red, as specified by the colour bar legend (min = 0.849, max = 0.959). Fat supplies are fixed at low, median and high for surfaces (A, D), (B, E) and C, F, respectively, based on the first, second and third quartiles of global fat supplies in 2017. Carbohydrate supplies and GDP are held constant across all surfaces (A–F), as specified in Fig. 2. Total non-protein energy (NPE) held constant as indicated on each surface, as specified in Fig. 2. The grey dashed isoline on each surface indicates total protein energy (PE) supplies of 400 kcal/capita/day. The smooth terms for macronutrient supply were statistically significant in both sexes (Supplementary Table 17). GDP gross domestic product per capita (in international-$ at 2011 prices).
Predicted effects of macronutrient supplies (in terms of kcal/capita/day) on life expectancy at birth (e0)
e0 values were derived from pairs of corresponding l\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$l$$\end{document}₅ and l\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$l$$\end{document}₆₀, as shown in Figs. 2, 3, and visualised across the same nutrient supply space. A–C and (D–F) show female and male e0\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${e}_{0}$$\end{document} outcomes, respectively. The colour map and contour lines in (A–F) show the e0\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${e}_{0}$$\end{document} in years, corresponding to the aligning proportions of protein supplies given by the (x,y)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(x,y)$$\end{document} coordinates. In (A–F), life expectancy values increase from blue to red, as specified by the colour bars legend (min = 70 years, max = 85 years). Moving from (A–C) and from (D–F) across each row of surfaces, the quantity of fat and total non-protein energy (NPE) supplies increase. The surfaces for low (A, D), median (B, E) and high (C, F) fat supplies show predicted effects at the first, second, and third quartiles of global fat supplies in 2017, respectively. (A–F) assume the median carbohydrate supply and GDP for 2017 (1571 kcal/capita/day; $30902). Total non-protein energy (NPE) is the sum of energy supplies from fat and carbohydrate for each surface. Total protein energy (PE) supplies are held constant at 400 kcal/capita/day along the grey dashed isoline shown on each surface. l5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${l}_{5}$$\end{document} = survivorship to age 5; l60\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${l}_{60}$$\end{document}= survivorship to age 60. GDP gross domestic product per capita (in international-$ at 2011 prices).
Effects of proportion animal- vs plant- based protein supplies on probability of mortality (qx), across different age classes, in 2017
Optimal macronutrient supply ratios for sex-specific survival at ages 5 and 60 were computed using a general-purpose optimization and selected as animal- and plant- based model environments (detailed in Supplementary Table 18). The qx\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${q}_{x}$$\end{document} schedules from birth to 95 years in each model environment were then derived. The log hazard ratio for the animal- based nutritional environment was taken as the log of the ratios of qx\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${q}_{x}$$\end{document} for animal- vs plant-based nutritional environments. The total energy for the primarily animal-based composition is 3150 kcal for males and 3110 kcal for females. Total energy for the primarily plant-based composition is 3710 kcal for males and 3700 kcal for females. Within model environments, the percentage of carbohydrates and fats differed by <0.5% across sexes (Supplementary Table 18). qx\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${q}_{x}$$\end{document} = probability of mortality at age x\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$x$$\end{document}.
Associations between national plant-based vs animal-based protein supplies and age-specific mortality in human populations
  • Article
  • Full-text available

April 2025

·

140 Reads

Caitlin J. Andrews

·

·

Stephen J. Simpson

·

Alistair M. Senior

Transitions to sustainable food systems require shifts in food production and availability, particularly the replacement of animal-based protein with plant-based protein. To explore how this transition may relate to demographic patterns, we undertake an ecological analysis of global associations between age-specific mortality, total national macronutrient distributions, and protein substitution. Our dataset includes per capita daily food supply and demographic data for 101 countries from 1961–2018. After adjusting for time, population size, and economic factors, we find associations between low total protein supplies and higher mortality rates across all age groups. Early-life survivorship improves with higher animal-based protein and fat supplies, while later-life survival improves with increased plant-based protein and lower fat supplies. Here, we show that the optimal balance of protein and fat in national food supplies, which correlates with minimal mortality, varies with age, suggesting that reductions in dietary protein, especially from animal sources, may need to be managed with age-specific redistributions to balance health and environmental benefits.

Download

Global association of macronutrient supplies and multiple sclerosis (MS) disease burden. A,B) Age‐standardized MS prevalence (blue) and incidence (red) of both sexes A), global GDP per capita (U.S. dollar, black) and supplies of carbohydrate (blue), fat (light yellow) and protein (red) as functions of year B). C,D) Predicted effects of macronutrient supplies on MS prevalence rate C) and incidence rate D) (See Supporting Information for statistics and interpretation and Figure S1 and Tables S1–S4).
High carbohydrate (HC) feeding aggravated CNS autoimmunity while high fat (HF) was protective. A) Study timeline. Mice were fed on high protein (HP), high carbohydrate (HC) and high fat (HF) diets for 6–7 weeks before experimental autoimmune encephalomyelitis (EAE) induction and were kept on the same diet throughout the EAE clinical course. B) EAE clinical course for HP (red), HC (blue) and HF (light yellow) groups (n = 10/group). C) Histological analysis of neuroinflammation and demyelination of the spinal cord from HP, HC and HF mice isolated on D28 of EAE, paraffin‐sectioned and stained with H&E and Luxol fast blue (LFB) and their corresponding quantification results. Scale bars = 20 µm. D) Heatmap comparisons for the gene expression of Ifng, Tnfa, Nlrp3, Ccl2, and Cd68 in the spinal cord of HP, HC and HF mice isolated on D28 of EAE analyzed by qPCR. E) Representative flow cytometric plots of CNS‐infiltrating T cells of HP, HC and HF mice on D28 of EAE. F) Proportions of infiltrating immune cells and activated microglia (MHChi Mic) within the CNS of HP, HC and HF mice on D28 of EAE determined by flow cytometry. G) Representative flow cytometric plots of splenic Th1 (IFNγ‐producing CD4⁺ T cells) and Th17 (IL‐17‐producing CD4⁺ T cells) cells of HP, HC and HF mice on D28 of EAE. H) Proportions of pro‐inflammatory cytokine‐producing T cells in the spleens of HP, HC and HF mice were determined by cytometry. I) Total number of Treg in draining lymph nodes (dLNs) of HP, HC and HF mice on D7 of EAE determined by flow cytometry. J) IL‐10 production in immune cells from dLNs of HP, HC and HF mice quantified by flow cytometry. K,L) Representative flow cytometric plots K) and scattering dot plots L) of dLN Th1 and Th17 cells. M–O) Proliferation of dLN Th1 M), Th17 N), and IFNγ‐producing CD8⁺ T cells O) of HP, HC and HF mice, with (filled dots) and without (hollow dots) stimulation, quantified by Ki67 staining. P–R) IFNγ production in dLN CD4⁺ P), CD8⁺ Q) and γδT cells R) of HP, HC and HF mice upon MOG antigen stimulation. S,T) Proliferation of dLN Treg of HP, HC and HF mice, with (filled dots) and without (hollow dots) stimulation S) and their corresponding proliferation indices T), quantified by Ki67 staining, with the dashed line denoting proliferation index of 1. N = 6‐10, data are represented as mean ± S.E.M., with *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, by one‐way ANOVA for most analyses; # p < 0.05 by unpaired t‐test; two‐way ANOVA for the clinical curve analysis, and paired t‐test for the proliferation assay analysis. (See also Figures S2–S6, Supporting Information).
High fat feeding modulated immunometabolism to induce tolerance during CNS autoimmunity. A) Violin plots of the gene set scores for glycolysis and fatty acid metabolism for CNS‐infiltrating immune cells in Ctrl (cyan) and EAE (red) mice analyzed by scRNA‐seq. B) Gene set enrichment analysis (GSEA) showing the enrichment of glycolysis pathway in the peripheral blood mononuclear cells (PBMCs) from MS patients versus Ctrl. C,D) Violin plots of the gene set scores for glycolysis and fatty acid metabolism of T cells in PBMCs (C) and cerebrospinal fluid (D, CSF) from Ctrl (pink) and MS patients (blue) analyzed by scRNA‐seq. E–G) Levels of HK1 (E), CPT1A (F) and lipid‐staining BODIPY 493/503 (G) in dLN T cells and conventional dendritic cells (cDC) of HP, HC and HF mice on D7 of EAE were quantified by flow cytometry. H) IL‐10 production (IL‐10⁺% of cells) in different immune cell subsets (CD4⁺ T cells (green), CD8⁺ T cells (brown), Treg (cyan) and cDC (dark green)) was significantly correlated with their corresponding lipid content (BODIPY⁺% of cells), as quantified by BODIPY 493/503 staining. (See also Figures S5, S7–S9, Supporting Information). N = 6‐10, data are represented as mean ± S.E.M., with *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, by one‐way ANOVA.
High fat feeding preconditioned T cell toward Treg differentiation. A) Experimental design. Mice were fed on HP, HC and HF diets for 6–7 weeks before sorting splenic naïve CD4⁺ T cells for RNA‐sequencing (RNA‐seq) and reduced‐representation bisulfite sequencing (RRBS‐seq). B,C) GSEA showing the enrichment of mTORC1 signaling pathway comparing HF naïve T cells with HP (B) and HC (C). D–F) GSEA showing the enrichment of signals upregulated in Treg versus conventional T cells comparing HF naïve T cells with HP (D,E) and HC (F). G,H) Venn diagrams showing the hypo‐methylated (G) and hyper‐methylated (H) regions comparing HF naïve T cells with HP (red) and HC (blue) and their overlapping genes Map3k11 (hypo‐methylated) and Ube2q1 (hyper‐methylated), and the violin plots for the expression of Map3k11 and Ube2q1 analyzed by RNA‐seq. I) Proportions of Treg differentiated from splenic naïve CD4⁺ T cell from HP (red), HC (blue) and HF (light yellow) mice quantified by flow cytometry. J) Proportion of splenic Treg from HP, HC and HF mice. K) Proportion of splenic Treg from mice fed on HF diet and mice first fed on HF diet and then switched to HC diet (HF→HC), and mice fed on HP diet and mice fed on HP diet then switched to HF diet (HP→HF). L) Visualization of the compositions of diets used in this study. Each circle represents one diet and their relative locations on the x, y and hypotenuse axes denote the proportion of protein, fat and carbohydrate (carb). The proportions of carbohydrate are also reflected by the color range. M–O) Contributions of macronutrient compositions to Treg proportions in mesenteric lymph node (M, mLN) and thymus (N) and to in vitro Treg differentiation experiment (O), were modelled by mixture modelling and mapped on right‐angled mixture triangles, consisting of protein (x‐axis), fat (y‐axis) and carbohydrate (hypotenuse). (See Supporting Information for statistics and interpretation and Table S5–S8). N = 6‐8, data are represented as mean ± S.E.M., with *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, by one‐way ANOVA for most analyses and unpaired t‐test for the Treg dynamics analysis.
Overview of the potential mechanisms underlying the multifaceted anti‐inflammatory effects of an isocaloric high fat (HF) diet. HF reprogrammed the immunometabolism, reducing the pro‐inflammatory glycolysis while promoting the anti‐inflammatory lipid storage in T cells and dendritic cells (DC), which led to a more tolerogenic phenotype as reflected by higher IL‐10 production. HF also modulated naïve CD4⁺ T cell at epigenetic and transcriptomic levels. HF reduced the promoter methylation of Map3k11 and increased the gene expression, which was linked to dampened T cell activation and reduced Th1 and possibly Th17 responses. Higher Map3k11 expression might associate with increased IL‐10 production and enhanced Treg generation, via activation of AP1/JunD signals, which warranted future investigations. HF also upregulated the mTORC1 signaling in naïve CD4⁺ T cells and skewed them toward a transcriptomic profile more similar to Treg, promoting Treg generation. Together, through these multi‐factorial mechanisms, HF feeding was linked to protection against experimental autoimmune encephalomyelitis (EAE).
High Fat Low Carbohydrate Diet Is Linked to CNS Autoimmunity Protection

March 2025

·

93 Reads

·

1 Citation

·

·

Julen Reyes

·

[...]

·

Multiple sclerosis (MS) is an inflammatory and neurodegenerative disease of the central nervous system (CNS) believed to be driven by autoimmune mechanisms. Genetic and environmental factors are implicated in MS development, and among the latter, diets and nutrients are emerging as potential critical contributors. However, a comprehensive understanding of their impacts and the underlying mechanisms involved is lacking. Harnessing state‐of‐the‐art nutritional geometry analytical methods, it is first revealed that globally, increased carbohydrate supply is associated with increased MS disease burden, while fat supply has an opposite effect. Furthermore, in a MS mouse model, experimental autoimmune encephalomyelitis (EAE), it is found that an isocaloric diet high in carbohydrate aggravated EAE, while a diet enriched in fat (HF) is fully protective. This is reflected by reduced neuroinflammation and skewing toward anti‐inflammatory phenotypes. The protective effects from the HF diet are multifaceted. Metabolically, HF increased lipid storage in immune cells, correlating with their increased anti‐inflammatory IL‐10 production. Transcriptionally and epigenetically, HF feeding preprogrammed naïve T cells toward a less activated but more tolerogenic phenotype. It is showcased that manipulating diets is a potentially efficient and cost‐effective approach to prevent and/or ameliorate EAE. This exhibits translational potentials for prevention/intervention of MS and possibly other autoimmune diseases.



High fat low carbohydrate diet is linked to protection against CNS autoimmunity

July 2024

·

57 Reads

·

1 Citation

Multiple sclerosis (MS) is a common central nervous system (CNS) autoimmune disease, and diets and nutrients are emerging as critical contributing factors. However, a comprehensive understanding of their impacts and the underlying mechanisms involved is lacking. Harnessing state-of-the-art nutritional geometry analytical methods, we first revealed that globally, increased carbohydrate supply was associated with increased MS disease burden, while fat supply had an opposite effect. Furthermore, in a preclinical MS mouse model, experimental autoimmune encephalomyelitis (EAE), we found that an isocaloric diet high in carbohydrate aggravated EAE, while a diet enriched in fat was fully protective. This was reflected by reduced neuroinflammation and skewing towards anti-inflammatory phenotypes, which involved transcriptomic, epigenetic and immunometabolic changes. We showcased that manipulating diets is a potentially efficient and cost-effective approach to prevent and/or ameliorate EAE. This exhibits translational potentials for intervention/prevention of MS and possibly other autoimmune diseases.

Citations (2)


... replying to: A. Senior et al.; Scientific Reports https://doi.org/10.1038/s41598-025-90267-x (2025. ...

Reference:

Reply to: A caveat about the use of trigonometric functions in statistical tests of Nutritional Geometry models
A caveat about the use of trigonometric functions in statistical tests of nutritional geometry models

... Therefore, it would be of interest to inspect which threshold levels of ketosis are required for epilepsy treatment or prevention. In addition, a high-fat, low-carbohydrate diet is likely to influence the gut-brain axis [20] and immunometabolism [21], which are also critical for the physiology and pathology of the CNS system. ...

High fat low carbohydrate diet is linked to protection against CNS autoimmunity