Project description:Frailty is defined as a clinical syndrome. It reflects a decrease in physiological reserve capacities that alters the mechanisms of adaptation to stress. Its clinical expression is modulated by biological, physiological, social and environmental changes. Thus, the accumulation of declines in these multiple systems leads to vulnerability to adverse outcomes (Fried et al., 2001; Nourhashémi et al., 2001). Indeed, frailty increases the risk of disability, falls, hospitalization and death. Age is currently a major determinant of frailty but does not explain this syndrome alone. It has been estimated that the prevalence of frailty is 4% in people aged 65-69 years and increases to 26% in people aged over 85 years (Clegg et al., 2013). However, this parameter is not sufficient to understand why long-lived people can preserve their physical function even at extreme ages (Ayers et al., 2017). The characterization of frailty has therefore become a priority in order to put in place actions to reduce or delay its harmful consequences. Early diagnostic methods are needed to identify those at risk. The search for biomarkers is part of this approach. As a result, the search for biomarkers of frailty has led to the identification of several types of molecules linked to the inflammatory response (e.g. IL-6, CRP, TNF-α and homocysteine), to nutritional status (e.g. vitamin D, albumin) or even endocrine. metabolism (e.g. IGF-1 and testosterone) (Picca et al., 2022; Mailliez et al., 2020; Alvarez-Sanchez et al, 2020). The development of new analytical techniques now allows multiplexed analysis rather than an individual biomarker to characterize a biological profile of this complex and dynamic frailty syndrome (Picca et al., 2022). Proteomic analysis is the technique of choice for studying the complete protein content of a biological sample. This technique has already proven itself in the discovery of new biomarkers in age-related diseases such as Parkinson's disease (Raghunathan et al., 2022) and Alzheimer's disease (Bai et al., 2021). Several studies have already described the plasma proteomic profile of frailty (Shamsi et al., 2012; Lin et al., 2017; Landino et al., 2021; Sathyan et al., 2020; Verghese et al., 2021), highlighting the importance of biological pathways associated with frailty: immune response, coagulation pathway or lipid metabolism. On the other hand, the determination of the level of frailty of the patients thanks to a proteomic analysis could also make it possible to improve the prediction of this syndrome as well as its phenotyping. In a first work on plasma samples, Sathyan et al. constructed a frailty prediction model that was more strongly correlated with chronological age than observed clinical frailty (Sathyan et al., 2020). Cognitive impairment plays a key role in determining vulnerability to adverse outcomes (Cesari et al., 2013), leading to the proposal in 2013 of a concept and operational definition of “cognitive frailty”. (Sugimoto et al., 2022). Proteomic analysis of cerebrospinal fluid (CSF) is therefore an approach of choice for understanding brain responses to aging and frailty syndrome. Its analysis could allow the identification of new biomarkers linked to the frailty syndrome. Indeed, many of the CSF proteins are released by the various cells of the central nervous system (CNS). This biofluid thus reflects the molecular content of neuronal and glial cells. For this reason, any change in the composition of the CSF reflects the pathophysiological state of the brain (inflammation, infection, neurodegeneration...). CSF proteomics is therefore a technique of choice to comprehensively understand the cerebral pathogenic pathways linked to frailty. The PROLIPHYC cohort is composed of patients suspected of normal pressure hydrocephalus (NPH). We have previously shown that there is an association in these patients between an alteration of the biomechanical properties of the CNS and the frailty index, and that homocysteine is the only plasma macromolecule associated with these two clinical and biomechanical parameters (Vallet et al., 2020; Guillotin et al., 2022). In this new study, we sought to characterize the protein composition of CSF related to frailty by quantitative proteomic analysis. The second objective was to develop a model for predicting frailty from the abundance of CSF proteins. This model was then compared to the frailty estimated by the frailty index. Together, our results enabled us to highlight a possible link between inhibition of the expression of proteins involved in neurogenesis and the frailty index, then to propose a first proteomic profile of frailty in the CSF and a model for predicting frailty based on the biological signature of the CNS.

Project description:Purpose: Obesity is a global health issue. To investigate if protein and fat contents of the diets had effects on energy balance via the canonical hunger signaling pathways in the hypothalamus, RNAseq was performed on RNA extracted from the hypothalami of mice exposed to the different diets. A suggested mechanism by which animals may avoid obesity is by burning off excess energy via upregulation of white adipose tissue (WAT) browning. To investigate if protein and fat content of the diet had effects on energy balance via the browning related signaling pathways in the WAT, RNAseq was performed on RNA extracted from the subcutaneous WAT (sWAT) and epididymal WAT (eWAT) of mice exposed to the different diets. Methods: C57BL/6 male mice were used in this work. All mice were fed a standard diet with 10% fat and 20% protein (D12450B, Research Diets Ltd) for 2 weeks as the baseline period. Following 2 weeks of baseline monitoring (at age 12 weeks), all mice were randomly allocated to different groups and switched to the experimental diets for 12 weeks. After 12 weeks all mice were sacrificed and dissected. Methods: In total, mice were fed on 4 diet series, each series consisting of 6 different diets (total = 24 diets). In the first two series (Series 1: D14071601–D14071606 and series 2: D14071607 – D14071612) we fixed the level of fat by energy, and varied the protein content. The protein source was casein. The balance was made up by carbohydrate (roughly equal mix of corn starch and maltodextrose). The source of fat was a mix of cocoa butter, coconut oil, menhaden oil, palm oil and sunflower oil. This mix was designed to match the balance of saturated, mono-unsaturated and polyunsaturated fats (ratio 47.5: 36.8: 15.8) and the n-6: n-3 ratio (14.7: 1) in the typical western diet. The proportions of the different fat constituents and hence fatty acid distributions did not change as the total fat content changed. Sucrose and cellulose were both fixed 5% by energy and weight respectively, and all diets were supplemented with a standard vitamin and mineral mix. In the second two series of diets (series 3: D14071613 – D14071618 and series 4: D14071619 – D14071624) we fixed the level of protein by energy and then allowed the fat content to vary. In these diets the sucrose, cellulose and vitamin and mineral contents were the same as the diets in series 1 and 2. All these diets can be ordered direct from research diets (www.researchdiets.com) using the diet codes provided. Methods: From each diet group, the hypothalami of 8/20 individuals were collected. The left halves of two, and the right halves of another two, were pooled together as one sample, and the same was performed with the other 4 hypothalami, resulting in each diet group having 2 pooled samples of 4 hypothalami (n = 48 samples in total across 24 diets). From each diet group, the sWAT and eWAT of 12/20 individuals were also collected. A small piece from each of six sWAT collections were pooled together as one sample, and the same was performed with the other six eWAT collections. In this way each diet group had one pooled sWAT sample and one pooled eWAT sample (also n = 48 across 24 diets). Methods: The total RNA of the hypothalamus and WAT was isolated using the RNeasy Mini Kit (QIAGEN, 74104) according to manufacturer's protocol. All sequencing was carried out using the Illumina NextSeq 500 sequencer. RNA fragments were sequenced by 75 bp long reads from paired ends (PE 2 x 75 bp, 150 bp per fragment). Quality control checks for raw data FASTQ files were done by using FASTQC (a quality control tool for high throughput sequence data; http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Paired-end reads were mapped to the Mus musculus genome (GRCm38) using Bowtie 2-2.1.0, TopHat-2.0.10, and Samtools-0.1.19; uniquely mapped reads for each gene were counted against the GTF file of GRCm38 provided by Ensembl (release 83) using HTSeq-0.6.1p1 using the strand = reverse; after read count data were obtained from the TopHat-HTSeq pipeline, counts per million (CPM) value for each gene was calculated by using the R package ‘edgeR’ (version 3.12.0, R version 3.2.2) to normalize the count data by the size of the library of each sample. Genes with the CPM value ≥ 1 in at least one of the 24 diets group were retained (Anders et al., 2013). Generalized Linear Modelling (GLM) was applied by R (version 3.2.2). The GLM model used here was: ~p+f+p:f, which regresses gene expression (CPM) against the protein (p) and fat contents (f) of diets, as well as their interaction (p:f). However, when the effect of the interaction was not significant (p value ≥ 0.05), the interaction term was dropped and a revised model (~p+f) was utilized. Results: With TopHat-HTSeq pipeline, reads of each sample were mapped to 46,078 genes. In hypothalamus there were 15,371 genes with the counts per million (CPM) value ≥ 1 in at least one of the 24 diets group; in white adipose tissue there were 18,202 genes with the CPM value ≥ 1 in at least one of the 24 diets group. No major changes in hypothalamic gene expression levels were found in relation to different dietary protein levels at fixed fat contents, however hypothalamic gene expression showed increase in expression of genes in reward pathways in relation to dietary fat, while Agrp and Npy were both downregulated in relation to dietary fat levels. WAT gene expression showed decrease in expression of general thermogenic related genes and WAT browning related genes in relation to both dietary protein and dietary fat, while Tgfb1, Pdk4 and Fgf1 were all upregulated in relation to dietary fat levels. Conclusions: Significant positive associations were evident between the fat levels of the diet and the main hedonic signaling systems linked to food intake. Significant negative associations were found between both protein and fat levels of the diet and WAT browning or general thermogenic signalings linked to energy expenditure.

Dataset Information

Effect of dietary nutrition on frailty in patients with colorectal cancer undergoing chemotherapy

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets