The methyl phosphate capping enzyme Bmc1/Bin3 is a stable component of the fission yeast telomerase holoenzyme

Abstract

The telomerase holoenzyme is critical for maintaining eukaryotic genome integrity. In addition to a reverse transcriptase and an RNA template, telomerase contains additional proteins that protect the telomerase RNA and promote holoenzyme assembly. Here we report that the methyl phosphate capping enzyme (MePCE) Bmc1/Bin3 is a stable component of the S. pombe telomerase holoenzyme. Bmc1 associates with the telomerase holoenzyme and U6 snRNA through an interaction with the recently described LARP7 family member Pof8, and we demonstrate that these two factors are evolutionarily linked in fungi. Our data suggest that the association of Bmc1 with telomerase is independent of its methyltransferase activity, but rather that Bmc1 functions in telomerase holoenzyme assembly by promoting TER1 accumulation and Pof8 recruitment to TER1. Taken together, this work yields new insight into the composition, assembly, and regulation of the telomerase holoenzyme in fission yeast as well as the breadth of its evolutionary conservation.

Fig. 1. Bmc1 interacts with U6 snRNA and the telomerase RNA TER1.

A Enrichment of Bmc1 PrA immunoprecipitated transcripts compared to an untagged control (biological replicates = 3). Axes represent log2 of fold change (FC) and negative log of false discovery rate (FDR) (Benjamini–Hochberg adjusted P value ≤0.05). B Northern blot analysis of the mature and intron-containing U6 from total RNA and PrA immunoprecipitates from untagged (wild type, wt) and PrA-tagged Bmc1 strains. C Northern blot of the telomerase RNA TER1, U6, and U5 from total RNA and PrA immunoprecipitates from untagged and PrA-tagged Bmc1 strains, and semi-quantitative RT-PCR analysis of TER1. Source data are provided as a Source Data file.

Fig. 2. Bmc1 interacts with the same TER1 species as well-established components of the telomerase holoenzyme.

A The 5’ and 3’ ends of Bmc1-, Trt1-, Pof8-, and Lsm3-associated TER1 were identified by cRACE. The results of ten independent clones per immunoprecipitation are shown below a schematic of the architecture of TER1. B RNase H northern blots of RNase H-generated 5’ and 3’ ends of TER1 immunoprecipitated by various telomerase components. The same blot was stripped and reprobed for U6. C, D Northern blot of α-TMG flow through (FT) and immunoprecipitated (IP) transcripts from total RNA (C) and Bmc1-associated RNA (D). Source data are provided as a Source Data file.

Fig. 3. Bmc1 interacts with components of the mature telomerase holoenzyme.

A Gene ontology analysis (biological process) of top 50 Bmc1 protein interactors. B, C Examination and nucleic acid-dependence of interactions between Bmc1 and Pof8 (B) and Bmc1 and Trt1 (C) by co-immunoprecipitation and western blotting. Blots were reprobed for beta-actin as a loading control. Source data are provided as a Source Data file.

Fig. 4. Bmc1 is recruited to the active telomerase holoenzyme by Pof8.

A Northern blot and semi-quantitative RT-PCR of TER1 and U6 in PrA immunoprecipitates for an untagged wild type (wt), PrA-tagged (Bmc1 PrA), and PrA-tagged Pof8 knockout strain (Bmc1 PrA ∆Pof8). Bmc1 PrA was detected in input and immunoprecipitated samples by western blots probing for PrA (bottom panel). Possible cleavage products are indicated with an asterisk. B qRT-PCR of TER1 and U6 in Bmc1 PrA immunoprecipitates from wild-type and pof8∆ strains normalized to input RNA. Relative TER1 and U6 IP was calculated by comparing percent immunoprecipitation of TER1 or U6 to immunoprecipitation from an untagged strain (mean ± standard error, two-tailed unpaired t test *P  <  0.05, **P  <  0.01, ***P  <  0.001, and ****P < 0.0001) (n = 3 biological replicates). C Northern blot of mature and intron-containing U6 in PrA- and myc-tagged immunoprecipitated RNA. D Telomerase assay of PrA-tagged Bmc1 in a wild-type and pof8∆ strain. A ³²P-labeled 100-mer oligonucleotide was used as a loading control. Telomerase extension products were compared to a terminal transferase ladder, with +1 and +4 extension products indicated. Western blot probing for PrA following PrA immunoprecipitation is shown in the panel below. E Proposed model of the fission yeast telomerase holoenzyme. TER1 structure and binding locations are based on models constructed by Hu et al. and Mennie et al.. Source data are provided as a Source Data file.

Fig. 5. Bmc1 promotes TER1 accumulation and Pof8 recruitment to telomerase.

A Quantitation of TER1 and U6 in total RNA by qRT-PCR, normalized to act1 mRNA. P values over bars represent comparison to a wild-type strain (mean ± standard error, two-tailed unpaired t test *P  <  0.05, **P  <  0.01, ***P  <  0.001, and ****P < 0.0001) (n = 3 biological replicates). B Southern blot comparing telomere length in following three restreaks on rich media (one restreak = 20–25 generations). Genomic DNA was digested with EcoRI (left) or ApaI (right) to yield different sized telomere restriction fragments. C qRT-PCR of TER1 in Pof8-myc immunoprecipitates from a wild-type and bmc1∆ strain normalized to input RNA. Relative TER1 IP was calculated by comparing percent immunoprecipitation of TER1 to immunoprecipitation from an untagged strain (mean ± standard error, two-tailed unpaired t test *P  <  0.05, **P  <  0.01, ***P  <  0.001, and ****P < 0.0001) (n = 3 biological replicates). P values over each bar represent results of a two-tailed unpaired t test with Pof8-myc. D Telomerase assay of myc-tagged Pof8 in a wild-type and bmc1∆ strain. A ³²P-labeled 100-mer oligonucleotide was used as a loading control. Telomerase extension products were compared to a terminal transferase ladder, with +1 and +4 extension products indicated. Western blot probing for myc following myc immunoprecipitation is shown in the panel below. E, F Relative telomerase extension activity for myc-tagged Pof8 immunoprecipitates in a wild-type and bmc1∆ strain. The intensity of the +4 extension product was normalized to a precipitation loading control, then further normalized to TER1 expression (E) or the amount of TER1 immunoprecipitated with Pof8 (F). P values over each bar represent the results of a two-tailed unpaired t test with Pof8-myc (n = 4 biological replicates, mean ± standard error, *P  < 0.05, **P < 0.01, ***P <  0.001, and ****P < 0.0001). Source data are provided as a Source Data file.

Fig. 6. Phylogenic distribution of Bmc1/Bin3 and Pof8 in fungi.

Consensus cladogram describing the phylogenic relationships of 472 species representative of fungi phylum and classes and highlighting (using a color code) the distribution of Bin3 and Pof8 in these species. The cladogram is a consensus tree of 5328 distinct protein coding gene trees resulting from a genome-wide, against all, protein comparison (see “Methods”). Only posterior probabilities inferior to 1 are shown. The Bin3 and Pof8 distribution is recapitulated in Supplemental Data 3 with corresponding protein sequences. A cartoon presenting structural domains of Pof8 is presented. Pof8 La module and RRM1 are only inferred from secondary structure predictions and are represented by La* and RRM1* in the corresponding cartoon. In four species (colored in orange in the cladogram), only an N-terminal truncated version of Pof8 (Pof8*) that cannot accommodate the La module is present. Full cladogram with species names and statistical supports of the different nodes is presented in Supplementary Fig. 7.