An alternate splicing variant of the human telomerase catalytic subunit inhibits telomerase activity

Abstract

Telomerase, a cellular reverse transcriptase, adds telomeric repeats to chromosome ends. In normal human somatic cells, telomerase is repressed and telomeres progressively shorten, leading to proliferative senescence. Introduction of the telomerase (hTERT) cDNA is sufficient to produce telomerase activity and immortalize normal human cells, suggesting that the repression of telomerase activity is transcriptional. The telomerase transcript has been shown to have at least six alternate splicing sites (four insertion sites and two deletion sites), and variants containing both or either of the deletion sites are present during development and in a panel of cancer cell lines we surveyed. One deletion (beta site) and all four insertions cause premature translation terminations, whereas the other deletion (alpha site) is 36 bp and lies within reverse transcriptase (RT) motif A, suggesting that this deletion variant may be a candidate as a dominant-negative inhibitor of telomerase. We have cloned three alternately spliced hTERT variants that contain the alpha, beta or both alpha and beta deletion sites. These alternate splicing variants along with empty vector and wild-type hTERT were introduced into normal human fibroblasts and several telomerase-positive immortal and tumor cell lines. Expression of the alpha site deletion variant (hTERT alpha-) construct was confirmed by Western blotting. We found that none of the three alternate splicing variants reconstitutes telomerase activity in fibroblasts. However, hTERT alpha- inhibits telomerase activities in telomerase-positive cells, causes telomere shortening and eventually cell death. This alternately spliced dominant-negative variant may be important in understanding telomerase regulation during development, differentiation and in cancer progression.

Figure 1

Genomic organization of hTERT gene. (A) Solid boxes are exons; shaded boxes are 5′TR or 3′TR; the locations of the RT motifs (T, 1 and 2, A, B′, C, D and E) (8) are indicated on top; and the locations of alternate splicing sites (insertions 1–4 and deletions α and β are marked on the bottom (see Wick et al. [21] for an alternative mapping of the hTERT RT motif boundaries). The locations of termination codons that correspond to the variants containing insertions are marked with *, and ◆ marks the location of the termination codon for the variant containing deletion β; the deletion α does not cause a frameshift. (B) Four possible combinations of hTERT alternate splicing variants containing the deletions.

Figure 2

RT-PCR survey results of telomerase-negative (TRAP-) and telomerase-positive cell types (from BJ hTERT to H1299) as described in the Materials and Methods section. PCR products corresponding to four alternate splicing variants of hTERT containing the deletions are of different sizes and they are indicated on the

Figure 3

TRAP assay of telomerase-negative human fibroblast BJ cells infected with pBabe puro vector, pBabe puro hTERT α^- β^-, pBabe puro hTERT α^-, pBabe puro hTERT β^- and pBabe puro hTERT WT. Only pBabe puro hTERT WT produces telomerase activity in BJ cells.

Figure 4

Western blotting of SW39, DU145 and H1299 clones infected with the empty pBabe puro vector, pBabe puro hTERT α^- and pBabe puro hTERT WT. In vitro translated wild-type hTERT protein (TNT lane) is larger because it has a (His)⁶-Anti-Xpress epitope (pcDNA3.1HisC vector from Invitrogen, CA). Cell lysates of 250,000 cells are loaded in each lane, except for the positive control. Under these conditions, the endogenous levels of hTERT protein expressed by SW39, DU145 and H1299 are not detected.

Figure 5

Sample TRAP assays of H1299 clones infected with pBabe puro vector, pBabe puro hTERT α^-, pBabe puro hTERT β^-, pBabe puro hTERT α^-β^-, and pBabe puro hTERT WT during the course of cell culturing (250 cells assayed per lane). Telomerase activity levels in all clones have remained constant during the course of cell culturing.

Figure 6

Growth curves of clones of SW39 infected with pBabe puro vector, pBabe puro hTERT α^-, pBabe puro hTERT β^-, pBabe puro hTERT α^-β^- and pBabe puro hTERT WT All clones grow at the similar rates except at the end of their life-span, when the hTERT α^- clones show increased apoptosis.

Figure 7

Telomere length measurements (TRF) of SW39 clones and DU145 clones infected with pBabe puro vector, pBabe puro hTERT α^-, pBabe puro hTERT β^-, pBabe puro hTERT α^-β^- and pBabe puro hTERT WT during the course of cell culture. TRF lengths of all clones stay constant and within the range of clonal variation, except for the TRFs of the clones infected with hTERT α^-, which decrease over the time of cell culture.

Figure 8

The SW39 and DU145 clones that overexpress hTERT α^- variant eventually die by apoptosis. Levels of apoptosis are measured by staining with DAPI followed by microscopy. Shown are representative views of stained nuclei of SW39 clones (A) and DU145 clones (C). The apoptosis levels of the SW39 and DU145 clones that overexpressed hTERT α^- are 5–12%, whereas that of all other clones are below 1% (B and D).