HuR regulates telomerase activity through TERC methylation

Abstract

Telomerase consists of the catalytic protein TERT and the RNA TERC. Mutations in TERC are linked to human diseases, but the underlying mechanisms are poorly understood. Here we report that the RNA-binding protein HuR associates with TERC and promotes the assembly of the TERC/TERT complex by facilitating TERC C106 methylation. Dyskeratosis congenita (DC)-related TERC U100A mutation impair the association of HuR with TERC, thereby reducing C106 methylation. Two other TERC mutations linked to aplastic anemia and autosomal dominant DC, G107U, and GC107/108AG, likewise disrupt methylation at C106. Loss-of-HuR binding and hence lower TERC methylation leads to decreased telomerase activity and telomere shortening. Furthermore, HuR deficiency or mutation of mTERC HuR binding or methylation sites impair the renewal of mouse hematopoietic stem cells, recapitulating the bone marrow failure seen in DC. Collectively, our findings reveal a novel function of HuR, linking HuR to telomerase function and TERC-associated DC.

Fig. 1

HuR interacts with TERC in vitro. a RNA pull-down assays were performed using HeLa cell lysates and in vitro-transcribed RNAs depicted in Supplementary Fig. 1a. The presence of HuR in the pull-down materials was assessed by western blot analysis. p16 3′-UTR and CR (coding region) served as positive and negative controls, respectively. A 5-µg aliquot input (Inp.) and binding to GAPDH RNA were also assessed. b Purified his-HuR and in vitro-transcribed TERC was used for UV-crosslinking rEMSA assays. The covalently bound HuR was detected by western blotting. c Left, the association of HuR with TERC variants bearing mutations U40A, U100A, or U40A + U100A (Supplementary Fig. 1b) was determined by using RNA pull-down assays, as described in Fig. 1a. Right, quantification of the bands on the western blot (left); data are the means ± SD of the signals from three independent experiments and significance was analyzed by Student’s t-test (**p < 0.01). d In vitro ITC measurements of direct interactions between HuR and RNA oligonucleotides derived from human TERC, AUUUUUUGUCU (positions 37–47) and GUUUUUCUCG (positions 98–107), as described in Methods section. The dissociation constant (K_d) is calculated from 12 titrations and is indicated in the graphs

Fig. 2

HuR interacts with TERC in cells. a U2OS cells were co-transfected with a vector expressing flag-MS2-BP together with a vector expressing MS2-TERC or an empty vector. Forty-eight hours later, lysates were prepared and subjected to IP assays to assess the association of HuR with RNA MS2-TERC. Data are representative from three independent experiments. b The TriFC system was used for detecting the interactions between HuR and TERC in cells by flow cytometry. DKC1 and PABPC1 served as positive and negative controls, respectively

Fig. 3

HuR associates with hTERT in a TERC-dependent manner. a U2OS cells were co-transfected with a vector expressing flag-hTERT together with a vector expressing TERC for 48 h. IP assays were performed to assess the association of flag-hTERT and HuR. Data are representative from three independent experiments. b HeLa cells were transfected using a vector that expressed flag-hTERT. Forty-eight hours later, lysates were prepared and subjected to IP assays by using antibodies indicated. The IP materials were used to test the telomerase activity by using TRAP assays. Data are the means ± SD from three independent experiments

Fig. 4

HuR regulates telomerase activity. a, b HeLa cells were infected with a lentivirus vector expressing shHuR at the times indicated. Telomerase activity (a) and telomere length (b) were analyzed by TRAP assays and southern blot analysis (a and b, left), respectively. The means ± SD from three independent experiments were analyzed for significance by Student’s t-test (a and b, right) (*p < 0.05; **p < 0.01). c, d HeLa (c) or U2OS (d) cells were transfected with a vector expressing flag-TERT. Twenty-four hours later, cells were further transfected with a siRNA targeting HuR (c) or with a vector expressing TERC variants (d) (Supplementary Fig. 1b) and cultured for additional 48 h. UV-crosslinking followed by RNP IP assays were performed by using an anti-flag antibody. RNA isolated from IP materials was used for reverse transcription (RT) followed by real-time quantitative (q)PCR analysis to test the levels of flag-TERT-bound TERC (c) or its variants (d). Data in d were normalized against the levels of TERC or its variants. Data in c and d represent the means ± SD from three independent experiments; significance was analyzed by Student’s t-test (**p < 0.01)

Fig. 5

HuR regulates telomerase activity via TERC methylation. a Forty-eight hour after transfecting HeLa cells with a HuR siRNA (black) or a control siRNA (blank), RNA was isolated and used for bisulfite RNA sequencing analysis to measure the methylation of C106 and C323. b Forty-eight hour after transfecting HeLa cells with a vector expressing TERC variants or an empty vector (WT), RNA was isolated and used for bisulfite RNA sequencing analysis to measure C106 methylation in different variants (bearing point mutations were used for analysis). c U2OS cells were co-transfected with a vector expressing TERC variants C106G or U40A, U100A, or U40A + U100A, or C106G. Forty-eight hour later, TRAP assays were performed to determine the telomerase activity. d U2OS cells were co-transfected with a vector expressing flag-TERT together with a vector expressing TERC or its variant bearing C106G. UV crosslinking followed by RNP IP assays were performed to evaluate the association of flag-TERT with TERC and the variant bearing C106G. e HeLa cells were transfected with a vector expressing flag-hTERT or an empty vector (Input). Forty-eight hours later, lysates were prepared and subjected to UV crosslinking followed by IP assays using an anti-flag antibody. RNA prepared from the IP materials was further used for bisulfite RNA sequencing analysis to assess the methylation of C106. Data in c–d were normalized against the levels of TERC and its variants. Data in a, b, c, d, and e represent the means ± SD from three independent experiments; significance is analyzed by Student’s t-test (**p < 0.01)

Fig. 6

HuR regulates telomerase activity in mouse cells. a Schematic representation depicting the variants of mTERC. The point mutation sites are marked in red. b RNA pull-down assays were performed by using NIH3T3 cell lysates and in vitro-transcribed TERC variants depicted in Fig. 6a. The presence of HuR in the pull-down materials was assessed by western blot analysis. A 5-µg aliquot (input, Inp.) and proteins bound to GAPDH were included. Data represent the means ± SD of the band intensity from three independent experiments; significance was analyzed by Student’s t-test (*p < 0.05; **p < 0.01). c NIH3T3 cells were transfected with a siRNA targeting HuR. Forty-eight hours later, TRAP assays were performed to assess the telomerase activity. d RNA prepared from cells described in Fig. 6c was subjected to bisulfite RNA sequencing to analyze the methylation of C64 in mTERC. e A vector expressing mouse TERT (mTERT) was used for expressing mTERT in rabbit reticulocyte in vitro translation system. In vitro-transcribed mTERC or its variant bearing U15A + U58A or C64G was added into the system and used for TRAP assays to assess the telomerase activity. Data in (c, d), and (e) represent the means ± SD from three independent experiments; significance was analyzed by using Student’s t-test (**p < 0.01)

Fig. 7

HuR-telomerase axis impacts on the renewal of mHSCs. a Twelve weeks after transplanting, GFP⁺ donor-derived LT-HSCs (Lin⁻Sca1⁺cKit⁺FLT3⁺CD34⁻) were isolated from primary recipients of TERC^+/− group or G3TERC^−/− group (n = 5) and subjected to single-cell qPCR analysis to determine the telomere length. The ΔCT values of the qPCR data are represented as the means ± SD; significance was analyzed by using Student’s t-test (**p < 0.01). b, c Contribution of the GFP⁺CD45.2⁺ (with silenced HuR) cells to the indicated PB (b) and LT-HSC populations (c) at 12 weeks after transplantation, respectively. Data represent the means ± SD from five mice; significance was analyzed by using Student’s t-test (**p < 0.01). d–f G3TERC^−/− mHSC (LSK) cells were infected with lentiviruses expressing GFP together with TERC or its variants (U15A + U58A or C64G). The sorted cells were further treated and analyzed same as described in (a–c)

Fig. 8

A model summarizes the findings in this study. Association of HuR with TERC promotes TERT/TERC assembly through enhancing the methylation of TERC (m5C) in C106. The reported DC-related mutations impair HuR binding (U100A) or the methylation of TERC (U100A, G107U, and GC107/108AG)