The major goal of our Flight Definition is to identify and characterize plasma proteins in the blood plasma of astronauts that can be used as predictive biomarkers of immunological dysfunction due to space flight. Specifically, we characterize the proteome of blood plasma collected from the same astronauts at different times, i.e., pre-, in-, and post-flight.
In the last annual report, we reported that a total of 453 unique and non-redundant proteins were identified at >99% confidence. Changes in protein concentrations during the pre-, in-, and post-flight timeline were determined by Students’ t-test analysis of comparisons between groups (p<0.1 is considered significant). Our data demonstrate that there are 14 proteins with significant changes, i.e., increased or decreased, in expression levels in plasma samples collected in-flight as compared to those collected pre-flight. We also reported that there were 16 proteins identified with significant changes, i.e., increased in green or decreased in red, in expression levels found astronauts’ plasma collected post-flight, in relation to those collected in-flight. Last January, we presented our results at the Annual NASA Human Research Program (HRP) workshop in Galveston.
During the past year, we use Principal Component Analysis (PCA) to analyze these 453 unique and non-redundant proteins identified by LTQ Orbitrap XL Ion trap mass spectrometer. The PCA is frequently used in the global analysis of the “omic” datasets. It provides fully unsupervised information on the dominant directions of highest variability in the data and can, therefore, be used to investigate similarities between individual samples, or the formation of clusters. The PCA is a dimensionality reduction method used to reduce the dimensionality of large data sets by transforming a large set of variables into a smaller one while preserving as much information as possible. The new variables are called the principal components and they are linear combinations of the actual variables. The first principal component is a linear combination of all the actual variables that have maintained the greatest amount of variation. The second principal component is a linear combination of the remaining variables to give the second greatest amount of variation and this can continue for third, fourth, and so on components but usually the first and second components carry the most important information. A scatter plot of the first and second components (a score plot) will often display samples sharing similar characteristics being grouped together apart from samples with different characteristics (i.e., it shows clustering based on similarity). Another plot is also generated in this analysis called a loading plot. The loading plot shows how strongly each original variable influences the principal component. Often, most points in a loading plot will be clustered around the center and the outlier points may be associated with the variables making the largest contribution to the data variation. In summary, the PCA indicates the differences in the pattern of protein expression profiles between samples collected pre- and in-flight.
Moreover, we constructed a heatmap to visualize the result of a hierarchical clustering calculation of the 14 proteins differentially expressed in samples collected at pre- or in-flight from each astronaut. The results from the heatmap show that the intensities (level of expression) of tropomodulin-3 (TMOD3), SERPINA7, SERPING1, and SERPRINC1 were higher in samples collected in-flight, in relation to those collected pre-flight. Notably, the protein in the SERPIN (serine protease inhibitor) family, i.e., SERPINA7, SERPINC1, and ISERPING1, is the majority of those with increased intensities (expression levels). However, the data also demonstrated individual variability in the intensities of these proteins. It should be noted that SERPIN protease inhibitors comprise a large family of molecules involved in inflammation, immune response, blood clotting, hormone transport, and complement activation, dementia, and tumorigenesis. Our heatmap data also showed high intensities of CP, FLG2, KRT2, F13A1, DMKN, LUM in samples collected pre-flight. Subsequently, the intensities of expression of these proteins were declined in samples collected in-flights. Likewise, the intensities of another set of proteins (i.e., CFB, FBLN1, AHSG, and ECM1) were high in samples collected pre-flight; this followed by a reduction in intensities in samples collected in-flight, with one exception of increased intensity of FBLN1 protein in sample4s collected mid-flight from one astronaut.
Overall, the heatmap of our dataset demonstrated that the intensities of not only proteins in the SERPIN family but also TMOD3 (an actin filament pointed-end capping protein with multifunctional roles, including cell proliferation, cell migration, inflammation, and carcinogenesis) are consistently high in the in-flight samples. In contrast, low intensities of lumican (LUM) have been repeatedly detected in the in-flight samples. Of note, LUM is a major keratan sulfate proteoglycan of the cornea responsible for the circumferential growth, corneal transparency, epithelial cell migration, and tissue repair. Hence, it is possible to speculate that a reduction in LUM level may be associated with vision impairment that has been observed in many astronauts after spaceflight. In summary, our findings suggest that dysregulation of the SERPIN, TMOD3, and LUM may affect cell/tissue integrity and homeostasis, leading to late occurring health risks. Thus, our results may represent a foundation for the identification of countermeasures against the harmful effects of spaceflights.