Proteomic analysis of adipose tissue revealing differentially abundant proteins in highly efficient mid-lactating dairy cows

Abstract

The improvement of nutrient utilization efficiency in dairy cows represents an important task in view of the current rising demand for animal products and sustainable resource usage. In this perspective, the identification of appropriate markers to identify the most efficient animals for dairy production becomes a crucial factor. Residual feed intake (RFI), which represents the difference between predicted and actual intake, is used to define the efficiency of cows. In this study, subcutaneous adipose tissue (AT) was collected from five high efficient (HEF) and five low efficient (LEF) mid-lactation Holstein dairy cows, that represented subgroups of the 20% lowest RFI values (HEF) and highest 20% RFI values (LEF), out of a cohort of 155 cows that were examined for feed efficiency at the individual dairy barn at Volcani Institute, Israel. Adipose samples were examined for proteomic analysis by nano-LC/MS-MS and gene expression by RT-PCR. A total of 101 differential proteins (P ≤ 0.05 and fold change ± 1.5) and two protein networks related to feed efficiency were found between HEF and LEF cows. Among the enriched top canonical pathways, FAT10 signaling, EIF2 signaling, Sirtuin signaling, Acute phase response signaling, Protein ubiquitination and mTOR signaling pathways were related to feed efficiency in AT. Furthermore, abundance of transferrin (TF; FC = 78.35, P = 0.02) enriched pathways, including mTOR signaling, LXR/RXR and FXR/RXR activation was found in AT of HEF cows. Relative mRNA expression of RBM39, which is involved in energy metabolism, was decreased in AT of HEF versus LEF. The relationship found between the AT proteins and/or metabolic pathways and the feed efficiency demonstrates that AT may reflect metabolic adaptations to high efficiency, and suggests that these proteins together with their metabolic mechanisms are suitable candidates as biomarkers to identify efficient cows for dairy production.

Figure 1

Workflow used to study proteomic analysis of AT in HEF versus LEF dairy cows. AT was isolated from HEF and LEF groups, extracted proteins from samples. Proteins were digested and peptides were identified for the differential expression. Bioinformatic analysis and network analysis was performed to identify the FE related pathways generated by IPA (Qiagen). MS/MS spectra analysis confirmed the presence of peptides in the sequences of FE related proteins.

Figure 2

(A) Principle component analysis of HEF versus LEF dairy cows measured by IDEP9.1 server showing 14% variance. LEF samples represented in red circles whereas HEF samples represented in green triangles. (B) Volcano plot analysis of HEF versus LEF dairy cows generated by IDEP9.1 server with P value (< 0.05) on Y-axis and FDR (± 1.5) on X-axis. Each dot represents one protein and red color indicates more whereas blue color indicates less abundant proteins in AT. (C) Heatmap of AT analyzed by IDEP9.1 server where low peptide intensity represented by green color whereas high peptide intensity was measured in red color. Each cow in the study is numbered and represented in columns.

Figure 3

GO analysis of differentially abundant peptides of HEF versus LEF. (A) A pie chart of Biological process category (B) A pie chart of Molecular function category; (C) A pie chart of Cell component category.

Figure 4

GO terms of up and down regulating process of HEF versus LEF groups. (A) A pie chart of upregulated peptides involved process; (B) A pie chart of down regulated peptides involved process.

Figure 5

IPA analysis of top canonical enriched pathways in AT of HEF versus LEF groups. Top 18 canonical pathways related to feed efficiency of HEF versus LEF dairy cows on Y-axis and—log (P-value) on the X-axis.

Figure 6

Molecular interaction of Transferrin (TF) involved energy metabolism network generated by IPA (Qiagen). The upregulated peptides represented in red color and green color indicates downregulation. Direct interaction represented in lines and indirect interaction shown in dotted lines.

Figure 7

Identification of Transferrin (TF) peptides and structure. (A) MS spectra of top abundant TF peptides identified by LC–MS/MS; (B) Modelled structure of TF generated by MODELLER9V7 software showing sheets in blue color and helices in red color.