The Arabidopsis Pot1 and Pot2 proteins function in telomere length homeostasis and chromosome end protection

Abstract

Pot1 (protection of telomeres 1) is a single-stranded telomere binding protein that is essential for chromosome end protection and telomere length homeostasis. Arabidopsis encodes two Pot1-like proteins, dubbed AtPot1 and AtPot2. Here we show that telomeres in transgenic plants expressing a truncated AtPot1 allele lacking the N-terminal oligonucleotide/oligosaccharide binding fold (P1DeltaN) are 1 to 1.5 kb shorter than in the wild type, suggesting that AtPot1 contributes to the positive regulation of telomere length control. In contrast, telomere length is unperturbed in plants expressing the analogous region of AtPot2. A strikingly different phenotype is observed in plants overexpressing the AtPot2 N terminus (P2DeltaC) but not the corresponding region in AtPot1. Although bulk telomeres in P2DeltaC mutants are 1 to 2 kb shorter than in the wild type, these plants resemble late-generation telomerase-deficient mutants with severe growth defects, sterility, and massive genome instability, including bridged chromosomes and aneuploidy. The genome instability associated with P2DeltaC mutants implies that AtPot2 contributes to chromosome end protection. Thus, Arabidopsis has evolved two Pot genes that function differently in telomere biology. These findings provide unanticipated information about the evolution of single-stranded telomere binding proteins.

FIG. 1.

Two Pot genes in Arabidopsis. (A) Amino acid alignment of N-terminal portions of Pot1 proteins. Hs, Homo sapiens; Gg, Gallus gallus (chicken), Xt, Xenopus tropicalis (accession number NP_998876); At1, AtPot1; At2, AtPot2. Residues conserved in at least two sequences are highlighted in yellow. Residues conserved in at least four sequences are highlighted in red. Residues conserved in all five sequences are highlighted in pink. A consensus sequence is shown below the alignment. (B) RT-PCR of AtPot1 and AtPot2 mRNAs in different Arabidopsis tissues. M, molecular weight markers; F, flowers; S, stems; RS, rosette leaves; CL, cauline leaves; R, roots; C, callus. (C) Sequence similarity between AtPot1 and AtPot2 proteins and human Pot1 protein.

FIG. 2.

Overexpression of full-length AtPot1 and AtPot2 or the P1ΔC and P2ΔN derivatives in Arabidopsis does not alter telomere length. TRF analysis was performed on DNA extracted from whole primary transformants (T1) shown by RT-PCR to express full-length AtPot1 (A), full-length AtPot2 (B), P1ΔC (C), or P2ΔN (D). Telomere length is 2 to 5 kb in wild-type (WT) Arabidopsis plants of Columbia ecotype (40).

FIG. 3.

Overexpression of P1ΔN leads to telomere shortening. (A) Schematic diagram of the P1ΔN transgene and the corresponding endogenous AtPot1 allele (left); RT-PCR results of P1ΔN expression in transgenic plants 1 to 3 from panel B (right). Primers 1 and 2 are specific for the endogenous AtPot1 mRNA. After 40 cycles, AtPot1 mRNA is amplified in all plants equally well (lower panel, lanes 1 to 3, and WT). Primers 3 and 4 amplify both the P1ΔN transgene and the endogenous AtPot1 allele. After 20 PCR cycles, P1ΔN mRNA is amplified only in the transgenic plants (upper panel, lanes 1 to 3), but not in the wild-type plant (upper panel, WT), confirming that the P1ΔN transgene is overexpressed. (B and C) TRF analysis of P1ΔN mutants. Examples of P1ΔN primary transformants (T1) displaying wild-type telomere lengths (lanes 2 and 4) or shortened telomeres (lanes 1 and 3) are shown (B). Analysis of the second-generation (T2) progeny from transformants 1 (lanes 1 to 5) and 3 (lanes 6 to 9) from panel B (C). For most T2 plants, no additional telomere shortening was observed (lanes 1 to 3, 7, and 9), but in some siblings telomere tracts were extended to resemble the wild type (lanes 4 and 5 and lanes 6 and 8).

FIG. 4.

Morphological defects and telomere shortening in mutants overexpressing P2ΔC. (A) Schematic of the P2ΔC transgene and the corresponding endogenous AtPot2 allele (left); RT-PCR analysis of P2ΔC expression in transgenic plants 1 to 4 from panel D (right). RT-PCRs were conducted as described in the legend of Fig. 3. (B and C) Growth and developmental defects in P2ΔC mutants. (B) At 2 to 3 weeks of age, growth of several independent transgenic mutants (M) was delayed relative to wild-type (WT). (C) Such mutants displayed severe defects in leaf development. (D) TRF analysis of P2ΔC transformants. Primary transformants (T1) with a wild-type appearance displayed wild-type telomere length (lanes 2, 4, and 6), whereas plants that exhibited morphological defects had shortened telomeres (lanes 1, 3, and 5).

FIG. 5.

TRAP assays of wild-type and mutant Arabidopsis. Extracts from cauliflower inflorescence and wild-type Arabidopsis flowers serve as positive controls, and wild-type Arabidopsis leaves serve as a negative control for telomerase activity. P2ΔC and P1ΔN mutants have wild-type levels of telomerase activity in vitro.

FIG. 6.

Cytogenetic defects in mutants overexpressing P2ΔC. Cytogenetic analysis was performed on actively dividing mitotic tissues of pistils from wild-type (A) or P2ΔC mutants (B to H) using DAPI. Anaphase bridges, lagging chromosomes, and aneuploidy are evident in P2ΔC mutants.

FIG. 7.

Telomere fusion PCR analysis of P2ΔC and tert mutants. (A) Schematic diagram of telomere fusion PCR. Primers specific to unique subtelomeric sequences (white, black, and diagonally striped boxes) are directed toward telomeres and will only amplify a product if telomeres form covalent attachments with each other. (B) Primers specific for 4R and 3L, 4R and 5L, 4R and 5R, 3L and 3R, 3L and 5R, 5R and 5L, and 1L and 2R were used in the assay to amplify chromosome fusion products from DNA extracted from tert, wild-type, P2ΔC, and P1ΔN plants.