A thymus-specific noncoding RNA, Thy-ncR1, is a cytoplasmic riboregulator of MFAP4 mRNA in immature T-cell lines

Abstract

Background: Postgenomic transcriptome analyses have identified large numbers of noncoding (nc)RNAs in mammalian cells. However, the biological function of long ncRNAs in mammalian cells remains largely unknown. Our recent expression profiling of selected human long ncRNAs revealed that a majority were expressed in an organ-specific manner, suggesting their function was linked to specific physiological phenomena in each organ. We investigated the characteristics and function of ncRNAs that were specifically expressed in the thymus, the site of T-cell selection and maturation.

Results: Expression profiling of 10 thymus-specific ncRNAs in 17 T-cell leukemia cell lines derived from various stages of T-cell maturation revealed that HIT14168 ncRNA, named Thy-ncR1, was specifically expressed in cell lines derived from stage III immature T cells in which the neighbouring CD1 gene cluster is also specifically activated. The Thy-ncR1 precursor exhibited complex alternative splicing patterns and differential usage of the 5' terminus leading to the production of an estimated 24 isoforms, which were predominantly located in the cytoplasm. Selective RNAi knockdown of each Thy-ncR1 isoform demonstrated that microfibril-associated glycoprotein 4 (MFAP4) mRNA was negatively regulated by two major Thy-ncR1 isoforms. Intriguingly, the MFAP4 mRNA level was controlled by a hUPF1-dependent mRNA degradation pathway in the cytoplasm distinct from nonsense-mediated decay.

Conclusions: This study identified Thy-ncR1 ncRNA to be specifically expressed in stage III immature T cells in which the neighbouring CD1 gene cluster was activated. Complex alternative splicing produces multiple Thy-ncR1 isoforms. Two major Thy-ncR1 isoforms are cytoplasmic riboregulators that suppress the expression of MFAP4 mRNA, which is degraded by an uncharacterized hUPF1-dependent pathway.

Figure 1

Expression profiles of thymus-specific ncRNAs in T-cell leukemia cell lines. Quantitation of each ncRNA by qRT-PCR is shown in the graph. The expression level of the cell line showing the highest expression was defined as 100%. Bars represent the means ± SD of three measurements. The following T-cell leukemia cell lines were used: 1, RPMI-8402; 2, H-SB2; 3, CCRF-CEM; 4, DND41; 5, HPB-ALL; 6, Jurkat; 7, MOLT-3; 8, TALL-1; 9, MOLT-13; 10, MOLT-16; 11, PEER; 12, HUT-78; 13, MOTN-1; 14, MT-1; 15, SKW-3; 16, HUT-102; and 17, MT-2. The cell lines are categorized into four types according to EGIL criteria (7). Type-T-II includes cell lines #1 and 2 (open graph bars); type-T-III includes lines #3 to #8 (black graph bars); type-T-IV includes lines #9 to #11 (light gray graph bars); and mature T includes lines #12 to #17 (dark gray graph bars).

Figure 2

Coordinated expression of Thy-ncR1 and CD1 gene cluster. A. Genomic location of Thy-ncR1 on chromosome 1 (q23.1). The CD1 gene cluster is adjacent to Thy-ncR1. The distance between each gene is shown below the diagram in kbp. B. The expression of the CD1 gene family is highly correlated to Thy-ncR1 expression. Quantitation of each gene transcript by qRT-PCR is shown in the graph. The highest expression level among the 17 cell lines was defined as 100% for each transcript. Bars represent means ± SD for three independent experiments. C. Thy-ncR1 expression is cell-lineage specific. RT-PCR to detect Thy-ncR1, the CD1a and 1b gene transcripts, and the DC differentiation markers CD14, CD80, and CD86.

Figure 3

Complex alternative splicing to produce multiple Thy-ncR1 isoforms. A. Schematics of Thy-ncR1 isoform formation by alternative splicing. The polyadenylation sites are shown. The olfactory receptor gene OR10R2 and a pseudogene (OR10K1) encoded in the antisense strand are shown by the bold arrows. The exon numbers are shown below the diagram. B. The relative amounts of Thy-ncR1 isoforms. An RNase protection assay using riboprobes spanning the exon-exon junction of each isoform was conducted using total RNA from thymus (T) and Jurkat (J) cells. The negative control was yeast RNA (Y). The bands corresponding to each spliced isoform and the other isoforms in terms of each riboprobe are shown by closed and open arrows, respectively. The ratio of each isoform in thymus and Jurkat cells, which was calculated by dividing the band intensity of each isoform (closed arrow) by the sum of the band intensities of all isoforms (closed arrow + opened arrow), is shown below. C. Northern hybridization analyses to detect Thy-ncR1 isoforms. The DNA probes used are shown above. D. Additional splicing variants within exon 1. A schematic diagram of alternative splicing is shown above. RNase protection assay probe I derived from exon 1 is indicated by an arrow. The patterns obtained by RNase protection assays are shown in the lower panel with a diagram for each isoform (Ex 1a, 1b, and 1c). E. The combination formed by each exon 1 isoform spliced to other exons. RT-PCR using the exon 1 primer that detects all exon 1 isoforms, and an exon 2, 3, or 4 primer was carried out with RNA prepared from thymus (T), HPB-ALL (H), Jurkat (J), and SKW3 (S) cells. Three symbols (circle, triangle, and square) correspond to the exon 1 isoforms shown in D.

Figure 4

Thy-ncR1 transcripts are cytoplasmic but are not susceptible to NMD. A. Subcellular distribution of Thy-ncR1 isoforms. HPB-ALL cells were fractionated into nuclear and cytoplasmic fractions. RNA prepared from each fraction was used for qRT-PCR analyses. The levels of Thy-ncR1 isoforms (exon 1-3, 2-3, and 4-5) and precursor retaining intron 1 (int1-ex2) were quantified in the nuclear and cytoplasmic fractions. GAPDH and β-actin mRNA are the controls for cytoplasmic RNA, and β-actin pre-mRNA (actin intron) is the control for nuclear RNA. Thy-ncR1 is not susceptible to NMD. B. Potential ORFs (closed boxes) in Thy-ncR1 exons. The scales (nt) are shown below the diagram. C. Depletion of hUPF1 from Jurkat cells treated with siRNA (si-hUPF1) was confirmed by immunoblot analysis with anti-hUPF1 antibody. D. qRT-PCR analyses of Thy-ncR1 isoforms and known NMD targets (gas5 and UHG) in control (NC, adjusted to 100%) and hUPF1-depleted cells (si-hUPF1) as shown in B. Bars represent means ± SD for three independent experiments.

Figure 5

Thy-ncR1 controls MFAP4 mRNA levels. A. Isoform-specific knockdown of Thy-ncR1. The positions of siRNAs designed to knockdown specific Thy-ncR1 isoforms are shown. Isoforms that could be degraded by each siRNA are shown below the arrow. B. Confirmation of Thy-ncR1 knockdown. The levels of Thy-ncR1 isoforms were determined by qRT-PCR. C. Increase in MFAP4 mRNA upon Thy-ncR1 knockdown. The MFAP4 mRNA level in Thy-ncR1 knockdown cells was determined by qRT-PCR. D. RNA hybridization to detect MFAP4 mRNA in HPB-ALL cells treated with control and siRNA 1 of Thy-ncR1.

Figure 6

MFAP4 mRNA level is regulated by a unique hUPF1-dependent pathway. A, B. Effects of the knockdown of human NMD factors on MFAP4 mRNA. RNAi was conducted using control siRNA, hUPF1-targeted siRNA (si-hUPF1, in A), or hUPF2-targeted siRNA (si-hUPF2-1, in B) in Jurkat cells. C. Co-immunoprecipitation of MFAP4 mRNA from HPB-ALL cell extracts with anti-hUPF1 antibody. Known NMD targets (gas5 and UHG) and SMD targets (Arf1 and c-jun) were used as positive controls. D, E. Double knockdown of Thy-ncR1 and hUPF1. Immunoblot confirmed the depletion of hUPF1 upon treatment with si-hUPF1 in the presence or absence of si-RNA1 against Thy-ncR1 (D). qRT-PCR of MFAP4, gas5, and Thy-ncR1 in Jurkat cells that were treated with each combination of siRNAs shown