The Genomic Analysis of Erythrocyte microRNA Expression in Sickle Cell Diseases

Citation: Chen S-Y, Wang Y, Telen MJ, Chi J-T ( The Genomic Analysis of Erythrocyte microRNA Expression in Sickle Cell Diseases Shao-Yin Chen 0 Yulei Wang 0 Marilyn J. Telen 0 Jen-Tsan Chi 0 Baohong Zhang, East Carolina University, United States of America 0 1 The Institute for Genome Sciences and Policy, Duke University School of Medicine , Durham , North Carolina, United States of America, 2 Department of Molecular Genetics and Microbiology, Duke University School of Medicine , Durham , North Carolina, United States of America , 3 Applied Biosystems, Foster City , California, United States of America, 4 Division of Hematology, Department of Medicine, Duke University School of Medicine , Durham, North Carolina , United States of America Background: Since mature erythrocytes are terminally differentiated cells without nuclei and organelles, it is commonly thought that they do not contain nucleic acids. In this study, we have re-examined this issue by analyzing the transcriptome of a purified population of human mature erythrocytes from individuals with normal hemoglobin (HbAA) and homozygous sickle cell disease (HbSS). Methods and Findings: Using a combination of microarray analysis, real-time RT-PCR and Northern blots, we found that mature erythrocytes, while lacking ribosomal and large-sized RNAs, contain abundant and diverse microRNAs. MicroRNA expression of erythrocytes was different from that of reticulocytes and leukocytes, and contributed the majority of the microRNA expression in whole blood. When we used microRNA microarrays to analyze erythrocytes from HbAA and HbSS individuals, we noted a dramatic difference in their microRNA expression pattern. We found that miR-320 played an important role for the down-regulation of its target gene, CD71 during reticulocyte terminal differentiation. Further investigation revealed that poor expression of miR-320 in HbSS cells was associated with their defective downregulation CD71 during terminal differentiation. Conclusions: In summary, we have discovered significant microRNA expression in human mature erythrocytes, which is dramatically altered in HbSS erythrocytes and their defect in terminal differentiation. Thus, the global analysis of microRNA expression in circulating erythrocytes can provide mechanistic insights into the disease phenotypes of erythrocyte diseases. - Erythrocytes constitute more than 90% of the cell population in the peripheral blood and are responsible for efficient gas exchange in the human body. Mature erythrocytes are the end-products of a highly regulated differentiation process that involves the gradual loss of cellular organelles, a decline in nucleic acid content, and a step-wise acquisition of erythrocyte characteristics [1]. One striking feature of erythroid differentiation is that the nuclei are extruded from cells as they differentiate into reticulocytes, the immediate precursor of mature erythrocytes. While cytoplasmic RNA and translation activities are still detectable in CD71+ reticulocytes, they fall below the detection limit as cells terminally differentiate to become CD71- mature erythrocytes [2]. Since hemoglobin makes up the majority of erythrocyte cellular proteins, these cells are frequently thought to serve merely as inert and passive containers of hemoglobins. However, their more dynamic nature was suggested by the recent finding that erythrocytes can mount cellular signaling and trigger functional responses during physiological and pathological stresses [3,4]. This suggests that erythrocytes have a more sophisticated intracellular environment than previously appreciated. Processes disrupting erythrocyte homeostasis lead to human diseases that present huge economic and social challenges. Even though erythrocyte diseases have been studied for a long time, our current understanding of the erythrocyte still cannot fully explain the multitude of clinical effects or the enormous clinical variation seen in patients with these diseases. Global analysis of gene expression using microarray technology holds great potential for advancing our understanding of erythrocyte diseases. This technology has led to an explosion of knowledge in understanding pathogenic mechanisms and clinical heterogeneity in human cancers. However, its application to erythrocyte diseases has been limited by the long-held belief that mature erythrocytes lack RNA expression. This belief is supported by the ability some RNAbinding dyes (such as thiazole orange or methylene blue) to stain reticulocytes but not erythrocytes, an observation which forms the basis of the clinical utility of these dyes in distinguishing reticulocytes from mature erythrocytes [5]. Furthermore, analysis of RNA isolated from whole blood by PAXgene technology [6] revealed no detectable contribution from erythrocytes. Only gene signatures from neutrophils, lymphocytes and reticulocytes were detected, even though erythrocytes are the predominant cell type in the blood [7]. Given the potential limitations of sensitivity and size bias for these traditional means of isolating and characterizing erythrocyte RNA, it remained possible that erythrocytes may contain RNA species not previously identified. Here, we provide several lines of evidence that human mature erythrocytes, although lacking in ribosomal and large-sized RNAs, contain diverse and abundant microRNAs. Our findings confirm the results of a recent study on small RNA expression in Plasmodium falciparum that found several microRNA species in peripheral erythrocytes [8]. Importantly, the discovery of these microRNAs allowed us to use microarrays to compare their expression in mature erythrocytes from normal (HbAA) and homozygous sickle cell (HbSS) individuals and to identify a link between the dysregulation of microRNA and the pathogenesis of sickle cell disease (SCD). Purification of Human Mature Erythrocytes To test for the possibility that mature erythrocytes contain previously undetected RNAs, we developed a protocol to obtain a pure population of mature erythrocytes by removing other blood cells through a series of purification procedures (Fig 1A). Blood collected from healthy volunteers first went through a PALL PurecellNeo filter to remove most leukocytes and was then placed in a Ficoll-hypaque density gradient to separate erythrocytes from leukocytes and platelets. Finally, any remaining leukocytes (CD45+) and reticulocytes (CD71+) in the packed erythrocyte pellets were removed by magnetic immuno-depletion with AutoMacs. The purity of the erythrocytes obtained by this method was assessed with lineage-specific markers. Purified cells were 99.8% positive for the erythrocyte marker CD235a, while negative for leukocyte (CD45, CD16) and reticulocyte (CD71) markers (Fig 1B). These results correlated well with molecular evidence for leukocyte depletion, in that no CD45 mRNA was detected by RT-PCR (Fig 1C). The absence of reticulocytes in our erythrocyte preparations was validated (...truncated)