LARP3, LARP7, and MePCE are involved in the early stage of human telomerase RNA biogenesis

Abstract

Human telomerase assembly is a highly dynamic process. Using biochemical approaches, we find that LARP3 and LARP7/MePCE are involved in the early stage of human telomerase RNA (hTR) and that their binding to RNA is destabilized when the mature form is produced. LARP3 plays a negative role in preventing the processing of the 3'-extended long (exL) form and the binding of LARP7 and MePCE. Interestingly, the tertiary structure of the exL form prevents LARP3 binding and facilitates hTR biogenesis. Furthermore, low levels of LARP3 promote hTR maturation, increase telomerase activity, and elongate telomeres. LARP7 and MePCE depletion inhibits the conversion of the 3'-extended short (exS) form into mature hTR and the cytoplasmic accumulation of hTR, resulting in telomere shortening. Taken together our data suggest that LARP3 and LARP7/MePCE mediate the processing of hTR precursors and regulate the production of functional telomerase.

Fig. 1. The establishment of in vitro systems to examine the biogenesis of human telomerase.

a The in vitro 3′ end processing assay was carried out in 293T cell extracts at 37 °C for the indicated times. RNA was purified and resolved on a 6% polyacrylamide gel containing 8 M urea. Actin served as the loading control. b The exL and mature forms of hTR signals were normalized to 0 min. The bars are presented as mean values +/− SD calculated from triplicate experiments of three technical replicates. Dots represent data points from individual experiments. The significance was calculated with a two-sided Student’s t test; *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.001. c Western blotting analysis of telomerase assembled on biotin-labelled hTR pulled down with streptavidin beads for the indicated times. The signals from purified telomerase (lanes 7–12) were normalized to the peak signal. Input (IN) represents 10% of the purified telomerase (PT) samples. d Western blotting and Northern blotting analysis of the in vitro purified telomerase assembled on exL and mature forms of hTR. e Telomerase activity of the in vitro purified telomerase assembled on exL and mature forms of hTR. f Telomerase activity at the indicated time points of telomerase assembly on the mature form was normalized to that of telomerase assembly on the exL form. The mean values +/− SD were calculated from triplicate experiments of three biological replicates. Dots represent data points from individual experiments. The significance of the change in telomerase activity between samples was calculated with a two-sided Student’s t test; *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.001. g Telomerase processivity quantitation of in vitro-purified telomerase assembled on exL and mature forms of hTR. The intensity of each major band (+4, +10, +16, +22, +28, and so on) from the telomerase activity assay in e was quantitated by phosphorimager analysis. Source data are provided as a Source data file.

Fig. 2. LARP3, LARP7, and MePCE are involved in the early stage of telomerase assembly.

a Western blot analysis of in vitro-assembled telomerase purified at the indicated times. The signals from purified telomerase (lanes 7–12) were normalized to the peak signal. Input (IN) represents 10% of the purified telomerase (PT) samples. The samples derive from the same experiment and that gels/blots were processed in parallel. b Western blotting analysis of in vitro assembled telomerase assembled on the different hTR species as shown in the schematic (exL, exS, mature, 3′ stem loop-deleted, and pseudoknot). c Western blot analysis of 293T cell extracts after immunoprecipitation with antibodies against LARP3, LARP7, MePCE, and DKC1. d RT‒qPCR quantification of hTR recovered from immunoprecipitations for LARP3, LARP7, MePCE, and DKC1 normalized to the IgG control. Bar graph of the mean fold change in the hTR relative to the control samples. The mean values +/− SEM were calculated from triplicate qRT‒PCR experiments of three biological replicates. Dots represent data points from individual experiments. The significance of the change in hTR between samples was calculated with a two-sided Student’s t test; *p < 0.05, **p < 0.01, ***p < 0.005. e Endogenous LARP3, LARP7, MePCE, and DKC1 were immunoprecipitated and subjected to a telomerase activity assay. f Bar graph of the mean fold change in telomerase activity relative to that in the DKC1-associated telomerase samples. The mean values +/− SEM were calculated from triplicate telomerase activity experiments of three biological replicates. Dots represent data points from individual experiments. The significance of the change in telomerase activity between samples was calculated with a two-sided Student’s t test; *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.001. Source data are provided as a Source data file.

Fig. 3. LARP3 binding competes with tertiary structure formation.

a Schematic showing the proposed tertiary structure of exL with or without a mutation. b Western blot analysis of DHX36 and LARP3 pulled down with biotinylated wild-type, U460C, GG375/6AU, or 375-377GGA-deleted mutant hTR. c Western blotting analysis of telomerase assembled on biotin-labelled wild-type, U460C, or 375-377GGA-deleted mutant hTR pulled down with streptavidin beads for the indicated times. d The in vitro 3′ end processing assay with ³²P-labelled wild-type, U460C, or 375-377GGA-deleted mutant hTR fragments was carried out in 293T cell extracts at 37 °C for the indicated times. RNA was purified and resolved on a 6% polyacrylamide gel containing 8 M urea. Actin served as the loading control. Data are presented as mean values +/− SD calculated from triplicate experiments of three technical replicates. Dots represent data points from individual experiments. The significance was calculated with a two-sided Student’s t test; *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.001. Source data are provided as a Source data file.

Fig. 4. LARP3 plays a negative role in telomerase biogenesis.

a Western blots of cell extracts prepared from 293 T cells treated with either shRNA targeting LARP3 or transfected with an LARP3 plasmid. Endogenous tubulin served as a loading control. b, c Total RNA from 293 T cells treated with either shRNA targeting LARP3 (b) or transfected with an LARP3 plasmid (c) was subjected to qRT‒PCR for total hTR, 3′-extended hTR, GAPDH, ATP5β, and HPRT. Bar graph of the mean fold change in the hTR level relative to that of the control samples normalized to that of GAPDH, ATP5β, and HPRT. The mean values +/− SEMA were calculated from triplicate qRT‒PCR experiments of three biological replicates. Dots represent data points from individual experiments. The significance of changes between samples was calculated with a two-sided Student’s t test; *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.001. d Western blotting analysis of telomerase assembled on biotin-labelled hTR in the indicated extracts, followed by pulldown with streptavidin beads for the indicated times. e LARP3 was immunoprecipitated from cell extracts prepared from 293T cells either treated with shRNA targeting LARP3 or transfected with an LARP3 plasmid and subjected to a telomerase activity assay. f, g The in vitro 3′ end processing assay with ³²P-labelled hTR fragments was carried out in cell extracts prepared from 293T cells either treated with either shRNA targeting LARP3 (f) or transfected with an LARP3 plasmid (g) at 37 °C for the indicated times. RNA was resolved on a 6% polyacrylamide gel containing 8 M urea. Actin served as the loading control. The mean values +/− SD were calculated from triplicate experiments of three technical replicates. Dots represent data points from individual experiments. The significance was calculated with a two-sided Student’s t test; *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.001. Source data are provided as a Source data file.

Fig. 5. Reducing the expression level of LARP3 increases telomerase function and causes telomere elongation.

a Western blots of cell extracts prepared from K562 cells treated with shRNAs targeting luciferase or LARP3. Endogenous tubulin served as a loading control. b Total RNA from LARP3 knockdown K562 cells was subjected to qRT‒PCR to measure the levels of total hTR, 3′-extended hTR, GAPDH, ATP5β, and HPRT. Bar graph of the mean fold change for 3′-extended hTR relative to that of the control samples normalized to GAPDH, ATP5β, and HPRT. The mean values +/− SEM were calculated from triplicate qRT‒PCR experiments of three biological replicates. Dots represent data points from individual experiments. The significance of changes between samples was calculated with a two-sided Student’s t test; *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.001. c An in vitro 3′ end processing assay with ³²P-labelled hTR fragments was carried out in the indicated cell extracts. RNA was resolved on a 6% polyacrylamide gel containing 8 M urea. Actin served as the loading control. The mean values +/− SD were calculated from triplicate experiments of three technical replicates. Dots represent data points from individual experiments. The significance was calculated with a two-sided Student’s t test; *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.001. d Western blotting analysis of telomerase assembled on biotin-labelled hTR in the indicated extracts, followed by pulldown with streptavidin beads for the indicated times. e Endogenous DKC1 was immunoprecipitated and subjected to a telomerase activity assay. f Bar graph of the mean fold change in telomerase activity relative to the control group. The mean values +/− SEM were calculated from three biological replicates with a two-sided Student’s t test. g Telomere lengths determined by TRF analysis of gDNA prepared from K562 cells treated with shRNAs targeting luciferase or LARP3. Source data are provided as a Source data file.

Fig. 6. LARP7 and MePCE knockdown impairs the hTR processing and cellular localization.

a Western blots of cell extracts prepared from the indicated shRNA-treated HeLa cells. b TRF analysis of gDNA prepared from the indicated shRNA-treated HeLa cells. c Telomerase activity assay of endogenous DKC1 immunoprecipitates from the indicated shRNA-treated HeLa cell extracts. d Bar graph of the mean fold change in telomerase activity relative to the control group. The mean values +/− SEM were calculated from three biological replicates with a two-sided Student’s t test. e Northern blots of total RNA prepared from the indicated shRNA-treated HeLa cells. Bar graph of the mean fold change in hTR levels relative to the control and normalized to U1 snRNA levels. The mean values +/− SEM were calculated from three biological replicates with a two-sided Student’s t test. f Bar graph showing the distribution of hTR 3′ end positions mapped using 3′ RACE present on a subset of the transcripts. The numbers of reads analysed were as follows: sh-luciferase, 5,827,715; sh-PARN, 4,725,159; sh-LARP7, 5,008,122; and sh-MePCE, 6,105,815. g In situ hybridization and IF in the indicated shRNA-treated HeLa cells. Coilin served as a Cajal body marker. The scale bar represents 5 µm. h Bar graph illustrating the distribution of hTR shown in (g) in the cytosolic and nuclear fractions. The mean values +/− SEM were calculated from three biological replicates with a two-sided Student’s t test. i Total protein and RNA prepared from nuclear or cytosolic fractions from the indicated shRNA-treated HeLa cells were subjected to Western blots and Northern blots, respectively. Lamin A/C and GAPDH served as a nuclear marker and a cytosolic marker, respectively. The fold change in the hTR levels relative to those in the control samples was normalized to the levels of Lamin A/C or GAPDH, respectively. Dots shown in the bar graph represent data points from individual experiments. p values; *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.001. Source data are provided as a Source data file.

Fig. 7. The proposed model of the role of LARP3, LARP7, and MePCE during the stepwise assembly of human telomerase.

After transcription of human telomerase RNA (hTR), the 3′-extended long (exL) form without triple helix formation is preferentially recognized by LARP3 and is subsequently degraded. The triple helix conformation prevents LARP3 binding, which protects the exL form of hTR from rapid degradation and facilitates hTR biogenesis. PARN, LARP7, and MePCE function during the conversion of the 3′-extended short (exS) form into mature hTR. The absence of PARN, LARP7, and MePCE impairs the processing of exS and causes cytoplasmic localization of hTR.