Comment on Pasini et al, page
The comprehensive analysis of the normal human red cell proteome reported by Pasini and colleagues provides an invaluable parts list that will contribute to diverse studies of normal red cell function and pathogenic disease mechanisms.
It has been nearly 40 years since Marchesi and Steers first isolated spectrin from human red cell membranes and initiated an era of intensive biochemical characterization of red cell membrane proteins.1 During the infancy of membrane biologic studies, red cells were a preferred source because the plasma membranes were easy to isolate in quantity, they appeared to be very simple, and they mediated numerous physiological processes that were poorly understood.
In this issue of Blood, Pasini and colleagues describe an in-depth analysis of the human red cell proteome. Although several prior reports used proteomics to characterize the red cell proteome, the current study is by far the most in-depth analysis to date. Of most importance, the high performance mass spectrometers and rigorous data analysis methods used in this study ensure that false-positive protein identifications have been minimized. Similar rigor has been applied to other aspects of the study, starting with the high purity of cell and membrane preparations and finishing with careful interpretation of the specific isoforms of gene families inferred from the peptide identifications resulting from evaluation of tandem mass spectrometry data. Hence, the results of this study, particularly the information contained in the supplemental tables, should provide valuable new insights into red cell biology and pathogenic conditions such as malarial infection because it defines a comprehensive list of protein components in different cellular compartments.
Perhaps the most surprising result from this study is the relatively high complexity of the red cell proteome. Red cells have often been regarded as a cytoplasm consisting primarily of hemoglobin encased by a very simple membrane containing only a handful of integral and peripheral membrane proteins. Indeed, early studies in the field named the red cell membrane proteins based upon their migration rate on SDS gels (ie, bands 1 to 7).2 Subsequent higher resolution gels detected a few more protein components leading to more complex nomenclature for some proteins, such as band 4, which was subsequently resolved into bands 4.1a, 4.1b, and 4.2. As protein analytic tools improved, a substantial number of additional proteins were identified in the red cell membrane and, to a lesser extent, within the cytoplasm. However, even the most current models of red cell membranes consider a relatively small number of proteins. In the current study, a total of 340 membrane proteins was identified, which indicates an unprecedented level of complexity for this simplest of mammalian membranes.FIG1
In-depth mining of the red cell membrane and cytoplasm proteomes provides a comprehensive protein parts list for future studies of membrane biology, pathogenic conditions, and parasite infection. Illustration by Kenneth Probst.
In-depth mining of the red cell membrane and cytoplasm proteomes provides a comprehensive protein parts list for future studies of membrane biology, pathogenic conditions, and parasite infection. Illustration by Kenneth Probst.
Although the total number of proteins identified may not seem to be impressive compared with those described in more complex mammalian cells and tissues, the challenge here is the extreme difference in protein abundances. Even with the most modern proteomics tools, it remains quite challenging to identify very low-abundance proteins in the presence of very high-abundance proteins such as hemoglobin and band 3 in the cytoplasm and membrane, respectively.
While this study provides a wealth of valuable data, there are 2 limitations of such protein lists that are inherent in any in-depth proteome analysis study. It is impossible in some instances to unambiguously distinguish between low-abundance specific proteins in a sample and low levels of contaminants from a variety of sources. Therefore, the specific association of some low-abundance proteins with red cell membrane or cytoplasmic fractions must be confirmed by independent methods. Also, although the current study is rigorous and as comprehensive as current technology permits, it is very likely that a few important low-abundance proteins have escaped detection.
In summary, although red cells are clearly the simplest of mammalian cells, they are far more complex than previously thought. Furthermore, our understanding of how these proteins interact and contribute to key cellular processes remains highly incomplete. ▪
