Evidence for a functional role of Start, a long noncoding RNA, in mouse spermatocytes

Abstract

A mouse testis-specific long noncoding RNA (lncRNA), Start, is localized in the cytosol of Leydig cells and in the nucleus of pachytene spermatocytes. We previously showed that Start regulates steroidogenesis through controlling the expression of Star and Hsd3b1 genes in Leydig cells, but its function in germ cells was not known. Here we verified that a spermatocyte-specific protease gene, Prss43/Tessp-3, was downregulated in Start-knockout testes. To investigate the transcriptional regulatory activity of Start in spermatocytes, we first performed a series of reporter gene assays using a thymidine kinase promoter in spermatocyte-derived GC-2spd(ts) cells. A 5.4-kb genome sequence encompassing Start exhibited enhancer activity for this promoter, and the activity was decreased by knockdown of Start. Deletion of the Start promoter and replacement of the Start sequence abolished the enhancer activity and, consistently, the activity was detected in further experiments only when Start was actively transcribed. We then examined whether the Prss43/Tessp-3 gene could be a target of Start. A reporter gene assay demonstrated that the 5.4-kb sequence exhibited enhancer activity for a Prss43/Tessp-3 promoter in GC-2spd(ts) cells and that the activity was significantly decreased by knockdown of Start. These results suggest that Start functions in transcriptional activation of the Prss43/Tessp-3 gene in spermatocytes. Given that Start is presumed to regulate steroidogenic genes at the posttranscriptional level in Leydig cells, the function in spermatocytes is a novel role of Start. These findings provide an insight into multifunctionality of lncRNAs in the testis.

Fig 1. Expression of Prss/Tessp genes in Start-KO and wild-type testes.

(A) A schematic drawing of a 75-kb genomic region at mouse chromosome 9 (9: 110,625,000–110,700,000). Six white boxes specify protein-coding Prss/Tessp genes, whereas three black boxes are lncRNAs. Bent arrows indicate the transcriptional direction of each gene. An enlarged drawing at the bottom indicates positions of promoters and a BAC-derived sequence used in this study. Two grey boxes at the left-end of Prss43/Tessp-3 and Prss46 genes are regions of putative promoters, and a bold black bar is a region corresponding to a BAC-derived sequence that encompasses the full length of the Start sequence. (B) Expression of Prss/Tessp genes in the testis at 2.5 months determined by RNA-seq analysis. Total RNAs were purified from 2.5-month-old wild-type and Start-KO testes and used for RNA-seq analysis. The TPM values of Prss42/Tessp-2, Prss43/Tessp-3, Prss44/Tessp-4, Prss45/TESPL, Prss46, and Prss50/Tsp50 in Start-KO and wild-type testes were calculated. White bars represent the data from a wild-type testis, and black bars represent those from a Start-KO testis. RNA-seq was done with one set of littermates. (C) Expression of the Prss/Tessp genes in the testis at 2.5 months determined by qRT-PCR. Total RNAs were purified from 2.5-month-old wild-type and Start-KO testes and used for qRT-PCR. The Gapdh gene was used as an internal control to normalize the expression level of each gene. Data are presented as means ± S.D. from four sets of wild-type and KO littermates. Statistical significance was analyzed by Student’s t-test. **P < 0.01.

Fig 2. Effects of knockdown (KD) on enhancer activity of the BAC-5.4kb genome sequence encompassing Start.

(A) Reporter constructs and positions of sequences targeted by shRNAs. The BAC-5.4kb sequence shown at the top is derived from a BAC clone and carries the full length of Start (a black box). Two short lines represent sequences targeted by shRNAs for the knockdown (KD) experiment in (C). The “Control” vector contains the firefly luciferase gene (Luc) driven by the TK promoter. The “TK-BAC” construct has the BAC-5.4kb sequence downstream of Luc of the “Control” vector. (B) Enhancer activity of the BAC-5.4kb by a reporter gene assay. The “TK-BAC” and “Control” vectors were transfected into GC-2spd(ts) cells, and two days later, the cells were collected, and firefly luciferase activity was measured. Each value was normalized to that of Renilla luciferase activity, which was derived from a co-transfected construct. Data are presented as means ± S.D. from six independent experiments. Statistical significance was analyzed by Student’s t-test. **P < 0.01. (C) KD of Start decreased enhancer activity of TK-BAC-5.4kb. (Left) KD efficiency of two shRNAs (shRNA-v1 and shRNA-v2) was evaluated by qRT-PCR. The “TK-BAC” construct was stably transfected into GC-2spd(ts) cells, and two shRNA constructs and a control construct (Empty) were transiently transfected into these stable cell lines. After selection with hygromycin, total RNA was collected from each sample and used for qRT-PCR. The Gapdh gene was used as an internal control to normalize the expression level of each gene. (Right) The effect of Start-KD on enhancer activity of BAC-5.4kb was evaluated by a reporter gene assay. Firefly luciferase activity was measured and normalized to the amount of protein determined by the BCA assay. In both graphs, the value in the “Empty” group was set to 1.0. Data are presented as means ± S.D. from five independent experiments. Statistical significance was analyzed by one-way ANOVA followed by Dunnett’s test. **P < 0.01.

Fig 3. Attenuated enhancer activity caused by deletion of the Start promoter in the BAC-5.4kb sequence.

(A) Reporter constructs used in (B) and (C). “Control” and “TK-BAC” were described in the legend of Fig 2A. The “ΔProm” construct lacked a 354-bp sequence around the transcriptional start site of Start in “BAC-5.4kb”. This deletion is shown in detail in an enlarged circle on the right. The “λEco” construct carries a 2.3-kb λHindIII fragment in replacement with the Start sequence in “BAC-5.4kb”. The “λApa” construct lacks a 354-bp sequence around the transcriptional start site of Start in “λEco”. (B) Expression of Start in GC-2spd(ts) cells transfected with TK-BAC and ΔProm. Total RNA was collected from each sample and used for RT-PCR. The Gapdh gene was used as a positive control. Reverse transcription was done with (+) or without reverse transcriptase (-). The cycle numbers of PCR were 30 for Start and 25 for Gapdh. (C) Loss of enhancer activity in the BAC5.4-kb sequence by a deletion of the Start promoter and a replacement of the Start sequence. Each construct in (A) was separately transfected into GC-2spd(ts) cells. Two days later, firefly luciferase activity was measured, and each value was normalized to Renilla luciferase activity. The value in “Control” was set to 1.0. Data are presented as means ± S.D. from six independent experiments. Statistical significance was analyzed by one-way ANOVA followed by Dunnett’s test. Different letters represent the statistical significance (P < 0.01) among experimental groups.

Fig 4. Enhancer activity of parts of the BAC-5.4kb sequence.

(A) Reporter constructs used in (B) and (C). “Control” and “TK-BAC” were described in the legend of Fig 2A. The “Start-Forward” and “Start-Reverse” constructs contained the Start sequence downstream of the firefly luciferase gene (Luc) driven by the TK promoter in the forward and reverse directions, respectively. The “BAC-FH” and “BAC-SH” constructs contained the TK-driven Luc connected to the first half and the second half of the BAC-5.4kb genome sequence, respectively. (B) Start transcription from each construct in a reporter gene assay. Total RNA was collected from each sample and used for RT-PCR. The Gapdh gene was used as a positive control. Reverse transcription was done with (+) or without reverse transcriptase (-). The cycle numbers of PCR were 30 for Start and 25 for Gapdh. (C) Enhancer activity in each part of the BAC-5.4kb sequence encompassing Start. The constructs shown in (A) were separately transfected into GC-2spd(ts) cells. Two days later, firefly luciferase activity was measured and normalized to Renilla luciferase activity, which was derived from a co-transfected construct. The value in “Control” was set to 1.0. Data are presented as means ± S.D. from nine independent experiments. Statistical significance was analyzed by one-way ANOVA followed by post-hoc Tukey HSD test. Different letters represent the statistical significance (P < 0.01) among experimental groups.

Fig 5. The BAC-5.4kb sequence increased promoter activity of Prss43/Tessp-3 gene.

(A) Reporter constructs used in (B)-(E). The promoter sequence of each Prss/Tessp gene was inserted at the upstream of the firefly luciferase gene which was connected to the BAC-5.4kb sequence encompassing Start. (B, C) Enhancer activity of the BAC-5.4kb sequence for Prss43/Tessp-3 and Prss46 promoters by a reporter assay. Each construct depicted in (A) was transfected into GC-2spd(ts) cells, and two days later, firefly luciferase activity was measured. Each value was normalized to that of Renilla luciferase activity, which was derived from a co-transfected construct. Data are presented as means ± S.D. from three independent experiments. Statistical significance was analyzed by Student’s t-test. **P < 0.01. (D, E) Effects of Start-KD on enhancer activity of the 5.4-kb sequence. Each construct containing BAC-5.4kb in (A) was stably transfected into GC-2spd(ts) cells. The shRNA-v2 for Start was then transfected into the established cells, and after selection with hygromycin, firefly luciferase activity was measured. Each value was normalized to the amount of protein determined by the BCA assay. Data are presented as means ± S.D. from three independent experiments. Statistical significance was analyzed by Student’s t-test. **P < 0.01.