Proteomic Landscape of Human Sperm in Patients with Different Spermatogenic Impairments

Abstract

Although the proteome of sperm has been characterized, there is still a lack of high-throughput studies on dysregulated proteins in sperm from subfertile men, with only a few studies on the sperm proteome in asthenozoospermic and oligoasthenozoospermic men. Using liquid chromatography-mass spectrometry (LC-MS/MS) along with bioinformatics analyses, we investigated the proteomic landscape of sperm collected from subfertile men (n = 22), i.e., asthenozoospermic men (n = 13), oligoasthenozoospermic men (n = 9) and normozoospermic controls (n = 31). We identified 4412 proteins in human sperm. Out of these, 1336 differentially abundant proteins were identified in 70% of the samples. In subfertile men, 32 proteins showed a lower abundance level and 34 showed a higher abundance level when compared with normozoospermic men. Compared to normozoospermic controls, 95 and 8 proteins showed a lower abundance level, and 86 and 1 proteins showed a higher abundance level in asthenozoospermic and oligoasthenozoospermic men, respectively. Sperm motility and count were negatively correlated with 13 and 35 and positively correlated with 37 and 20 differentially abundant proteins in asthenozoospermic and oligoasthenozoospermic men, respectively. The combination of the proteins APCS, APOE, and FLOT1 discriminates subfertile males from normozoospermic controls with an AUC value of 0.95. Combined APOE and FN1 proteins discriminate asthenozoospermic men form controls with an AUC of 1, and combined RUVBL1 and TFKC oligoasthenozoospermic men with an AUC of 0.93. Using a proteomic approach, we revealed the proteomic landscape of sperm collected from asthenozoospermic or oligoasthenozoospermic men. Identified abundance changes of several specific proteins are likely to impact sperm function leading to subfertility. The data also provide evidence for the usefulness of specific proteins or protein combinations to support future diagnosis of male subfertility.

Keywords: asthenozoospermia; male subfertility; oligoasthenozoospermia; proteome; sperm.

Figure 1

(A) Schematic of the study population and their classification into normozoospermic men (N), subfertile men (abnormal, AN) and their specific phenotypes, asthenozoospermic men (A), and oligoasthenozoospermic men (OA). (B–D): Volcano plot showing the differential abundance levels, i.e., the log2 Fold Change (FC) plotted against the −log10 p-value of proteins in sperm collected from (B) the subfertile men (n = 22) compared to the normozoospermic controls (n = 31, AN vs. N), (C) the oligoasthenozoospermic men (n = 9) compared to the normozoospermic controls (n = 31, OA vs. N), and (D) the asthenozoospermic men (n = 13) compared to the normozoospermic controls (n = 31, A vs. N). Significantly abundant proteins were highlighted by color (adjusted p-value < 0.05).

Figure 2

(A) Venn diagram and (B) Upset plot of proteins that showed significantly different abundance levels (adjusted p-value < 0.05) in the comparison of subfertile men compared to the normozoospermic controls (AN vs. N), the oligoasthenozoospermic men compared to the normozoospermic controls (OA vs. N), and the asthenozoospermic men compared to the normozoospermic controls (A vs. N).

Figure 3

(A,C): Volcano plots of correlations between protein abundance levels of all identified proteins and sperm motility or sperm count plotted against the −log10 p-value, respectively. Correlated proteins were highlighted by color (r > 0.5 and r < −0.5). (B,D): Heatmaps representing vertically hierarchical clustering of the abundance levels of proteins that were correlated with sperm motility or sperm count, respectively (r > 0.5 and r < −0.5). Horizontally, the proteins were sorted by increasing sperm motility or sperm count, respectively. The colour of rectangles represents z-scored protein abundance levels (white, lower abundant proteins; dark blue, higher abundant proteins; light blue, no change).

Figure 4

(A–C): Scatterplot displaying the direction of regulation (log2 fold change) of proteins that showed significantly different abundance levels (p-value < 0.05) in the (A): AN vs. N and/or OA vs. N, (B): AN vs. N and/or A vs. N, and (C): A vs. N and/or OA vs. N. (D): Scatterplot displaying the direction of regulation (log2 fold change) of proteins that showed significantly different abundance levels (p-value < 0.05) in both A vs. N and OA vs. N. The color represents the area under the curve (AUC) value of proteins that were oppositely regulated.

Figure 5

Receiver operating characteristic (ROC) curves comparing sensitivity and specificity of single proteins and protein combinations in predicting (A): subfertility (AN), (B): asthenozoospermia (A,C): oligoasthenozoospermia (OA) as compared to normozoospermic controls (N). Area under the curve (AUC) values, the confidence interval (CI), and the p-values of single proteins and protein combinations were indicated.