The human platelet: strong transcriptome correlations among individuals associate weakly with the platelet proteome

Abstract

Background: For the anucleate platelet it has been unclear how well platelet transcriptomes correlate among different donors or across different RNA profiling platforms, and what the transcriptomes' relationship is with the platelet proteome. We profiled the platelet transcriptome of 10 healthy young males (5 white and 5 black) with no notable clinical history using RNA sequencing and by Affymetrix microarray.

Results: We found that the abundance of platelet mRNA transcripts was highly correlated across the 10 individuals, independently of race and of the employed technology. Our RNA-seq data showed that these high inter-individual correlations extend beyond mRNAs to several categories of non-coding RNAs. Pseudogenes represented a notable exception by exhibiting a difference in expression by race. Comparison of our mRNA signatures to a publicly available quantitative platelet proteome showed that most (87.5%) identified platelet proteins had a detectable corresponding mRNA. However, a high number of mRNAs that were present in the transcriptomes of all 10 individuals had no representation in the proteome. Spearman correlations of the relative abundances for those genes represented by both an mRNA and a protein showed a weak (~0.3) connection. Further analysis of the overlapping and non-overlapping platelet mRNAs and proteins identified gene groups corresponding to distinct cellular processes.

Conclusions: The results of our analyses provide novel insights for platelet biology, show only a weak connection between the platelet transcriptome and proteome, and indicate that it is feasible to assemble a platelet mRNA-ome that can serve as a reference for future platelet transcriptomic studies of human health and disease.

Figure 1

Reads, genes and the genome. A) Percentage of mapped reads across annotated genomic regions. Shown is the average percentage of uniquely mapped reads (long RNAs) that land on different genomic regions for the 10 samples. B) Table showing how many individuals share how many of the sequenced mRNAs (RNA-seq) and proteins (proteome reported in Burkhart et al [15]).

Figure 2

Inter- and intra-individual correlations. A) Heatmap of the inter-individual correlation of all mRNA transcripts (RNA-seq). B) Heatmap of the inter-individual correlation of all mRNA transcripts (microarray). C) Heatmap of the intra-individual correlation of all mRNA transcripts (RNA-seq vs microarray). The sample IDs are labeled with a W (White) or B (Black).

Figure 3

Enriched elements and pseudogenes. A) Enrichment analysis of the expressed genomic elements. Shown is the average enrichment for the 10 sequenced samples for various categories of annotated transcripts. The x-axis is the genomic element and the y-axis is the average enrichment value (log2) for each category. Values are averaged across all ten samples. Those categories reaching significant enrichments (P-value < = 0.05) are indicated with a “*”. B) Pseudogenes. Heatmap of the inter-individual Pearson correlations of pseudogene transcripts (RNA-seq). The sample IDs are labeled with a W (White) or B (Black).

Figure 4

Three groups of platelet genes. A) Venn iagram showing the number of genes contained in each of the three shown categories of genes. Note that the five categories are non-overlapping. B) Top entries of DAVID analysis for the GO, KEGG pathway, and UP_TISSUE terms corresponding to the genes contained in each of the categories shown in the A) panel.

Figure 5

Relationships between the platelet transcriptome (left) and proteome (right). The entries comprise some of the known causes that may underlie the observed discordance between platelets mRNAs and platelet proteins