Training associated alterations in equine respiratory immunity using a multiomics comparative approach

Abstract

Neutrophilic airway inflammation is highly prevalent in racehorses in training, with the term mild to moderate equine asthma (MMEA) being applied to the majority of such cases. Our proposed study is largely derived from the strong association between MMEA in racehorses and their entry into a race training program. The objectives of this study are to characterise the effect of training on the local pulmonary immune system by defining the gene and protein expression of tracheal wash (TW) derived samples from Thoroughbred racehorses prior to and following commencement of race training. Multiomics analysis detected 2138 differentially expressed genes and 260 proteins during the training period. Gene and protein sets were enriched for biological processes related to acute phase response, oxidative stress, haemopoietic processes, as well as to immune response and inflammation. This study demonstrated TW samples to represent a rich source of airway cells, protein and RNA to study airway immunity in the horse and highlighted the benefits of a multiomics methodological approach to studying the dynamics of equine airway immunity. Findings likely reflect the known associations between race-training and both airway inflammation and bleeding, offering further insight into the potential mechanisms which underpin training associated airway inflammation.

Figure 1

Tracheal wash cytology. Percentages (mean + SEM %) of macrophages, haemosiderophages, lymphocytes, neutrophils and eosinophils in the TW samples of 16 Thoroughbred racehorses during resting (T0) and training (T1) period. Note, the macrophage ratio represented the ratio of macrophages without phagocytosed haemosiderin; consequently, the decrease in the ratio of macrophages at T1 reflected the increase in the proportion of macrophages containing haemosiderin (i.e. haemosiderophages) at this time point.

Figure 2

MA plot. The gene expression data visualised as a two-dimensional scatter plot of the log2 ratio of expression values between the two timepoints. MA plots show the log-fold change (y axis) (M-values, i.e. the log of the ratio of level counts for each gene between T1 and T0) against the log-average (x axis) (A-values, i.e. the average level counts for each gene across the samples). Each dot represents one gene, and the red colour indicates the 2138 genes identified to be differentially expressed between the two time points using a false discovery rate (FDR) of < 0.05.

Figure 3

Gene expression of tracheal wash derived cells is modified during intense training. Hierarchical clustering based on normalized gene counts of differentially expressed genes in samples derived from 16 Thoroughbred racehorses at two time points using a false discovery rate (FDR) of < 0.05 and a fold change of 1.5.

Figure 4

Ingenuity^® Pathway Analysis predicted signs of activation of IFN signalling pathway. Ingenuity^® Pathway Analysis (IPA) predicted signs of activation of IFN signalling pathway in relation to intense training (z-score = 1.897 and p value = 7.28E−05). A canonical pathway network was derived using IPA by screening 2138 differentially expressed genes in tracheal wash (TW) samples derived from Thoroughbred racehorses before and during training for their impact on IFN signalling. Green shades indicate relative downregulation of genes during training, while red shades indicate relative upregulation. More intense (darker) colours indicate greater increases or decreases.

Figure 5

Comparative analysis of main diseases and biofunctions associated with differentially expressed genes and proteins. Comparative analysis of main diseases and biofunctions associated with upregulated and downregulated genes and proteins in response to intense training. Venn diagrams were generated based on the IPA Disease and Biofunction results derived from the upregulated (left panel) and downregulated (right panel) transcriptomic and proteomic datasets.

Figure 6

Venn diagram of genes and proteins regulated by IFN types. The Venn diagram shows the number of genes or proteins regulated by one or more IFN type (Type I, II or III). Venn diagrams were generated based on the upregulated genes (A), proteins (B) and the downregulated genes (C), proteins (D) during the training period (interferome database). Note there are very few datasets for type III interferons—only twenty datasets from just two human experiments; thus any interpretation of type III interferons should be done with caution (http://www.interferome.org/interferome/home.jspx).

Figure 7

Transcriptomic profile of racehorses with high neutrophil count on TW samples. (A) Hierarchical clustering was based on normalized gene counts of differentially expressed genes in samples derived from horses with and without a neutrophil count (High_N) exceeding publicised acceptable limits (54 were upregulated and 3 were downregulated). (B) Immune response-related biofunctions were enriched in airway cells derived from Thoroughbreds with high neutrophil count.