Transcriptional profiling of hematopoietic stem cells by high-throughput sequencing.

Microarray analysis has made it feasible to carry out extensive gene expression profiling in a single assay. Various hematopoietic stem cell (HSC) populations have been subjected to microarray analyses and their profiles of gene expression have been reported. However, this approach is not suitable to identify novel transcripts or for profiling of genes with low expression levels. To obtain a detailed gene expression profile of CD34(-)c-Kit(+)Sca-1(+)lineage marker-negative (Lin(-)) (CD34(-)KSL) HSCs, we constructed a CD34(-)KSL cDNA library, performed high-throughput sequencing, and compared the generated profile with that of another HSC fraction, side population (SP) Lin(-) (SP Lin(-)) cells. Sequencing of the 5'-termini of about 9,500 cDNAs from each HSC library identified 1,424 and 2,078 different genes from the CD34(-)KSL and SP Lin(-) libraries, respectively. To exclude ubiquitously expressed genes including housekeeping genes, digital subtraction was successfully performed against EST databases of other organs, leaving 25 HSC-specific genes including five novel genes. Among 4,450 transcripts from the CD34(-)KSL cDNA library that showed no homology to the presumable protein-coding genes, 29 were identified as strong candidates for mRNA-like non-coding RNAs by in silico analyses. Our cyclopedic approaches may contribute to understanding of novel molecular aspects of HSC function.