Abstract

UCRs are a subset of conserved sequences (100% between orthologous regions of human, rat and mouse genomes) located in both intra- and intergenic regions. There are 481 UCRs described and more than 50% of all the UCRs (256/481) has been classified as non exonic (with no evidence of coding protein), while the other 47% have been designated either exonic (overlap mRNAs of known genes), or possible exonic (with no clear evidence of overlap with protein coding genes) (Bejerano et al, Science 2004). Recently, our group reported that a subset of transcribed UCRs is aberrantly expressed in leukemia (Calin et al, Cancer Cell 2007). Because of the high degree of conservation, the UCRs are likely to be functional and may be involved in the phylogeny of mammals. Since hematopoiesis is regulated by signaling pathways and transcription factors that are highly conserved throughout phylogeny, we hypothesize that UCRs may be differentially expressed in hematopoietic tissues and may play a role in the regulation of this process. In this study we analyzed the UCRs expression in normal hematopoiesis by performing microarray analysis of hematopoietic precursors (HPCs) obtained by culturing human bone marrow (non-mobilized) CD34+ selected cells with different cytokine combinations for 2 weeks to stimulate differentiation to the erythrocyte (E), megakaryocyte (MK), monocyte (M) and granulocyte lineages (G). The following cytokine combinations were used: E (TPO/SCF/IL-3); G (G-CSF, GM-CSF, SCF, IL-3, IL-3); MK (TPO, SCF, IL-3) and M (M-CSF, GM-CSF, IL-6, IL-3 and SCF). Differentiation to the selected lineages was monitored every 3 days using morphology, special staining (benzidine) and flow cytometry analysis using appropriate lineage specific antibodies. Human bone marrow CD34+ and peripheral blood (PB) CD3+ (pan-T cells) and CD19+ (B-lymphocytes) selected cells were obtained from 3 different donors. Total RNA was obtained from the E and M cultures at days 7, 9, 11 and 14; G (days 10 and 14) and MK at day 14 and was hybridized in duplicate to the UCRs microarray chip (OSU version 4). After normalization of the array data with quantiles we analyzed the data using the BRB tools.

Unsupervised analyses of the data revealed that samples segregated mainly in 6 main clusters corresponding to the M, E, G and Mk lineages, PB lymphocytes and CD34+ cells. Next we performed a series of two class analysis where we compared the UCRs expression of CD34+ cells vs. differentiated cells (E, G, MK, M, T and B-lymphocytes) one at a time (i.e. CD34+ cells vs. E). In table 1 we showed the top UCRs differentially expressed between in vitro differentiated HPCs/PB lymphocytes and CD34+ cells (+ or − sign reflects if the UCRs is in the sense (+) or antisense orientation (−).

Table 1

 Erythrocyte Megak Monocyte Granulocyte Pan T-Lym CD19+ Lym 
Up-regulated uc.283+ uc.285+ uc.132− uc.161+ uc.419− uc.469− 
 uc.285+ uc.350+ uc.33+ uc.145− uc.145− uc.132− 
 uc.10+ uc.73+ uc.420− uc.262− uc.382− uc.145− 
 uc.188− uc.356− uc.269− uc.170− uc.170− uc.170− 
Down-regulated uc.43− uc.309+ uc.188− uc.160+ uc.356− uc.331+ 
 uc.183− uc.43− uc.160+ uc.356− uc.477+ uc.183− 
 Erythrocyte Megak Monocyte Granulocyte Pan T-Lym CD19+ Lym 
Up-regulated uc.283+ uc.285+ uc.132− uc.161+ uc.419− uc.469− 
 uc.285+ uc.350+ uc.33+ uc.145− uc.145− uc.132− 
 uc.10+ uc.73+ uc.420− uc.262− uc.382− uc.145− 
 uc.188− uc.356− uc.269− uc.170− uc.170− uc.170− 
Down-regulated uc.43− uc.309+ uc.188− uc.160+ uc.356− uc.331+ 
 uc.183− uc.43− uc.160+ uc.356− uc.477+ uc.183− 

These signatures were highly predictive since only few UCRs (mean 24) were able to predict the lineage of origin with no error (mean percentage of correct classification: 100%) after cross validation using multiples algorhytm (Class prediction within BRB). We validated the array results for uc.283+, uc.285+ and uc.161+ in a panel of differentiated CD34+ HPCs and CD34+ cells using qRT-PCR. In summary we have characterized the UCRs expression during hematopoietic differentiation of HSC and identified distinctive signatures associated with particular lineages. This research has the potential to identify novel regulators of hematopoiesis and may give insights into basic biology of gene expression and cell fate determination.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author