Re-appraising the evidence for the source, regulation and function of p53-family isoforms

Abstract

The p53 family of proteins evolved from a common ancestor into three separate genes encoding proteins that act as transcription factors with distinct cellular roles. Isoforms of each member that lack specific regions or domains are suggested to result from alternative transcription start sites, alternative splicing or alternative translation initiation, and have the potential to exponentially increase the functional repertoire of each gene. However, evidence supporting the presence of individual protein variants at functional levels is often limited and is inferred by mRNA detection using highly sensitive amplification techniques. We provide a critical appraisal of the current evidence for the origins, expression, functions and regulation of p53-family isoforms. We conclude that despite the wealth of publications, several putative isoforms remain poorly established. Future research with improved technical approaches and the generation of isoform-specific protein detection reagents is required to establish the physiological relevance of p53-family isoforms in health and disease. In addition, our analyses suggest that p53-family variants evolved partly through convergent rather than divergent evolution from the ancestral gene.

Graphical Abstract

Figure 1.

p63 protein isoforms. (Top) Gene architecture of the human TP63 locus. Boxes represent exons and DNA cartoon represents introns. Exons are numbered according to canonical annotation and cryptic exons are named 3′, 10′ and 10″. Promoters P1 and P2 driving production of TAp63 and ΔNp63, respectively, are represented by arrows. Translation initiation codons (ATG) are indicated with arrowheads. Splicing patterns are signalled by dashed lines that connect the regions involved. Dashed lines at the 5' end (on the left) refer to splicing resulting from alternative promoter usage. For 3′ variants (on the right), top dashed lines are canonical (α variants) while bottom are alternative splicing patterns (β, γ, δ and ϵ). Shaded areas denote nucleotide sequences that correspond to different protein domains presented on the panel below. (Bottom) Putative p63 protein isoforms listed according to detection confidence from endogenous sources and showing the domains present. TA1 and TA2: transactivation domain. PRD, proline-rich domain; DBD, DNA-binding domain; OD, oligomerization domain; SAM, sterile alpha motif domain; TID, transactivation inhibitory domain. The unique peptides are shown for γ and ϵ. Sizes of exons and protein domains are to scale. Protein isoforms are grouped according to the evidence for their expression.

Figure 2.

The distribution of TP63 transcripts in human tissues. (A) Proportions of alternatively spliced TP63 mRNAs. Numbers inside the circles refer to the average number of reads for each tissue, and the shading of the circles represent the percentage of reads across 3′ exon–exon junctions that define the alternative splicing event. Data are normalized according to the number of reads of non-spliced exon–exon junctions using the approach described in (33). (B) Proportions of TAP63 and ΔNP63 mRNAs derived from alternative promoter usage. Note that absolute levels vary considerably among tissues. Data are taken from GTEx. Esoph., esophagus; EBV⁺ lym., Epstein–Barr virus-positive lymphocytes; skin-nexp., skin not exposed to sun; skin-sexp., skin exposed to sun.

Figure 3.

Distribution of TAp63 and ΔNp63 protein isoforms in normal human tissues. Paraffin sections (4 μm) of the indicated normal human tissues were immunostained for ΔNp63 (mouse monoclonal ΔNp63-1.1) or TAp63 (mouse monoclonal TAp63-4.1) (40). Antibody binding was detected using peroxidase polymer-conjugated anti-mouse immunoglobulin and 3,3′-diaminobenzidine (DAB, brown) as chromogen. Nuclei were counterstained with haematoxylin (blue).

Figure 4.

p53 protein isoforms.(Top) Gene architecture of the human TP53 locus. Boxes represent canonical and alternative exons and DNA cartoon represents introns. Exons are numbered according to canonical annotation and cryptic exons are named 2i, 9β and 9γ. Promoters P1 and P2 are represented by arrows and translation initiation codons (ATG) by arrowheads along with the position on the coding sequence of full-length p53. Splicing patterns are annotated by dashed lines that connect the regions involved. Top are canonical while bottom are alternative splicing patterns. Shaded areas denote nucleotide sequences that correspond to different protein domains presented on the panel below. (Bottom) Putative p53 isoforms listed according to detection confidence from endogenous sources and showing the domains present. TA1 and TA2: transactivation domain. PRR, proline-rich region; DBD, DNA-binding domain; HD, hinge domain; OD, oligomerization domain; C-t, regulatory C-terminal domain. The unique peptide sequences are shown for 9β and 9γ. Sizes of exons and protein domains are to scale, except for exon 11 and β and γ peptides. Protein isoforms are grouped according to the evidence supporting their expression from endogenous sources in different cell types and their regulation. *Endogenous expression of p53γ was inferred once in cancer cell lines under conditions affecting the activity of splicing factors, and with a complex detection approach based on specific siRNAs to alter the expression pattern of p53 isoforms combined with three Western blotting experiments: presence of signal at an expected position (∼47–48 KDa) using antibodies recognizing several p53 isoforms and absence of signal using both N-terminal and β-specific antibodies to discard p53/47 (Δ40p53α) and p53β, respectively (60).

Figure 5.

TP53 transcripts in human tissues and their relative polysome distribution. (A) Levels of alternatively spliced C-terminal TP53 mRNAs. Numbers inside the circles refer to the average number of reads for each tissue, and the shading of the circles represent the percentage of reads across 3′ exon–exon junctions that define the alternative splicing event. Data are normalized according to the number of reads of non-alternatively spliced exon–exon junctions using the approach described in (33). The tissues shown are those for which more than one mRNA variant is present. Tissues not shown contain only TP53α mRNA. Data are taken from GTEx. adi. sub., adipose-subcutaneous; adi. vis., adipose-visceral; col. tra., colon-transverse; EBV⁺ lym., Epstein–Barr virus-positive lymphocytes; esoph., esophagus; fallop. tube, fallopian tube; kid. med., kidney-medulla; sal. min., salivary gland-minor; si. ter., small intestine-terminal ileum; skin-nexp., skin not exposed to sun; skin-sexp., skin exposed to sun. (B) Quantification of TP53 mRNAs associated with ribosomal fractions (ribosome profiling) of HEK293 cells. Translation was halted with cycloheximide, cell extracts were generated and ribosomal fractions were isolated according to their sedimentation index on a continuous 10–50% sucrose gradient. Total RNA was precipitated, purified and used to create a library by polyA capture and reverse transcription (66,67). Samples were sequenced using Illumina PE150 technology. The figure shows a schematic representation of polysome fractionation (top) and the relative quantification of P1 versus P2 and α versus β TP53 mRNAs (bottom). Concentration of ribosomal fractions was monitored by absorbance at 254 nm.

Figure 6.

Gene architecture of the human TP73 locus. Boxes represent canonical and alternative exons and DNA cartoon represents introns. Exons are numbered according to canonical annotation and cryptic exon is denoted as 3′. Promoters P1 and P2 are represented by arrows and translation initiation codons (ATG) with arrowheads. Splicing patterns are annotated by dashed lines that connect the regions involved. Splicing at the 5' end (on the left) above the cartoon refers to splicing patterns resulting from alternative promoter usage, and below the cartoon refers to alternative splicing events. At the 3' end (on the right), canonical splicing is indicated above and alternative splicing patterns are indicated below. Shaded areas denote nucleotide sequences that correspond to different protein domains. Note that ΔTAp73 may originate in at least five ways; as a result of alternative splicing of Ex2, Ex2/3 and ΔN’, through use of P2 to transcribe exon 3′, as well as alternative translation initiation at an AUG located in exon 4.

Figure 7.

The distribution of TP73 transcripts in human tissues. (A) Proportions of alternatively spliced TP73 mRNAs. Numbers inside the circles refer to the average number of reads for each tissue and isoform data represent the percentage of reads across 3′ exon–exon junctions that define the alternative splicing event, normalized according to the number of reads of non-spliced exon–exon junctions, as described in (33). TP73α cannot be determined from junction reads since there is no specific junction to define it. Therefore, values were estimated by subtracting the reads assigned to other isoform from the total number of reads. β and ϵ cannot be uniquely distinguished; thus, the values represent their sum. The same approach was used for γ/ϵ. (B) Proportions of TAP73, ΔNP73 and ΔTAP73 mRNAs. ΔTAP73 levels were calculated as the number of TP73 reads that are not assigned to TAP73 or ΔNP73 (ΔEx2P73 and ΔEx2P73 are undetectable). Note that absolute TP73 mRNA levels vary considerably. Data are taken from GTEx. cer. hemi., cerebellar hemisphere; EBV⁺ lym., Epstein–Barr virus-positive lymphocytes; skin-nexp., skin not exposed to sun; skin-sexp., skin exposed to sun.