Do anti-angiogenic VEGF (VEGFxxxb) isoforms exist? A cautionary tale

Abstract

Splicing of the human vascular endothelial growth factor-A (VEGF-A) gene has been reported to generate angiogenic (VEGFxxx) and anti-angiogenic (VEGFxxxb) isoforms. Corresponding VEGFxxxb isoforms have also been reported in rat and mouse. We examined VEGFxxxb expression in mouse fibrosarcoma cell lines expressing all or individual VEGF isoforms (VEGF120, 164 or 188), grown in vitro and in vivo, and compared results with those from normal mouse and human tissues. Importantly, genetic construction of VEGF164 and VEGF188 expressing fibrosarcomas, in which exon 7 is fused to the conventional exon 8, precludes VEGFxxxb splicing from occurring. Thus, these two fibrosarcoma cell lines provided endogenous negative controls. Using RT-PCR we show that primers designed to simultaneously amplify VEGFxxx and VEGFxxxb isoforms amplified only VEGFxxx variants in both species. Moreover, only VEGFxxx species were generated when mouse podocytes were treated with TGFβ-1, a reported activator of VEGFxxxb splice selection in human podocytes. A VEGF164/120 heteroduplex species was identified as a PCR artefact, specifically in mouse. VEGFxxxb isoform-specific PCR did amplify putative VEGFxxxb species in mouse and human tissues, but unexpectedly also in VEGF188 and VEGF164 fibrosarcoma cells and tumours, where splicing to produce true VEGFxxxb isoforms cannot occur. Moreover, these products were only consistently generated using reverse primers spanning more than 5 bases across the 8b/7 or 8b/5 splice junctions. Primer annealing to VEGFxxx transcripts and amplification of exon 8b primer 'tails' explained the artefactual generation of VEGFxxxb products, since the same products were generated when the PCR reactions were performed with cDNA from VEGF164/VEGF188 'knock-in' vectors used in the generation of single VEGF isoform-expressing transgenic mice from which the fibrosarcoma lines were developed. Collectively, our results highlight important pitfalls in data interpretation associated with detecting VEGFxxxb isoforms using current methods, and demonstrate that anti-angiogenic isoforms are not commonly expressed in mouse or human tissues.

Figure 1. Murine VEGF-A splice variants.

Several splice variants of VEGF are produced, by alternative splicing of exons 6 & 7. Exons 3 & 4 encode dimerisation and VEGF receptor binding sites, exon 7 encodes neuropilin binding sites and exons 6 & 7 encode heparin binding sites.

Figure 2. C-terminal exon sequences of murine and human VEGF-A genes.

Top two panels show the predicted splicing reactions for murine VEGF188b, 164b and 120b (sequence in bold and arrowed), highlighting the exon 7/exon 8b and exon 5/exon 8b splice junction sequences. These splicing reactions would generate putative VEGFxxxb protein isoforms with a PLTGKTD C-terminus. The reverse PCR primers designed for this study are indicated below the exon sequences. The lower panel shows the corresponding sequence of the human VEGF-A gene C-terminus highlighting the exon 7/exon 8b splice junction that generates VEGFxxxb isoforms with a SLTRKD C-terminus. The reverse PCR primers used in the amplification reactions are indicated below the exon sequences and DB 165b/188b-1 and DB 3'UTR are as previously published . The bases highlighted in bold exhibit mismatches from the published human VEGF-A sequence.

Figure 3. A general RT-PCR approach with primers to exon 7a and 3'UTR failed to detect VEGFxxxb isoforms in human and mouse cDNA extracts.

RT-PCR reactions were performed using forward/reverse primers designed to amplify both VEGFxxx and VEGFxxxb isoforms from cDNAs extracted from mouse fibrosarcoma cell lines & tumours and normal mouse tissues (heart, lung, liver and kidney) as well as commercially sourced human tissue cDNAs. A, A single product corresponding to VEGF164/188 (194 bp) is evident in the panel of mouse cDNAs using forward and reverse primers designed to exon 7 and the 3'UTR (exon7/3′UTR) with no evidence of VEGF164b/188b (128 bp) products. B & C, A single product corresponding to VEGF165/189 (194 bp) is evident in human brain, bladder and kidney (AMS Biotechnology Ltd), and prostate and kidney (Primer Design Ltd) using forward and reverse primers designed to exon 7 (DB exon7a) and the 3'UTR (DB 3'UTR) respectively. No products corresponding to VEGF165b/189b (128 bp) are evident in any of these samples.

Figure 4. A general RT-PCR approach using primers to exon 3 or 4 and 3'UTR failed to detect VEGFxxxb isoforms in mouse and human cDNA extracts respectively.

A, PCR products corresponding to VEGF188 (474 bp), VEGF164 (402 bp), VEGF164/120 heteroduplex (arrowed) and VEGF120 (270 bp) are evident in the panel of mouse cDNAs amplified using the 3′UTR reverse primer and a forward primer to exon 3. B & C, PCR products corresponding to VEGF189 (371 bp), VEGF165 (299 bp) and VEGF121 (167 bp) are evident in the commercial human tissue cDNAs amplified using the 3′UTR reverse primer (DB 3'UTR) and a forward primer to exon 4 (DB exon4). D, Amplification of the heteroduplex species (arrowed) a lso occurred when VEGF164 and VEGF120 tumour cDNAs were pooled and analysed by RT-PCR using exon3/3'UTR primers (lanes labelled 120+164). E, The same heteroduplex species was generated when cDNAs from VEGF164 and VEGF188 expressing fibrosarcoma cells were pooled and amplified as above (lanes labelled 120+164). M corresponds to a 100 bp ladder, whilst -tem represents a control PCR reaction in which water was used instead of cDNA template. The Figure contains data we published previously .

Figure 5. Effect of IGF-1 and TGFβ-1 on expression of VEGF isoforms in mouse podocytes.

Cell lysates were prepared from untreated control (C) podocyte cells or cells treated with either 100 nM IGF-1 (IGF) or 1 nM TGFβ-1 (TGF) for 12 hours in serum free media. Results from three independent experiments are shown. RT-PCR using general primers designed to simultaneously amplify both VEGFxxx and VEGFxxxb isoforms (exon7a/3'UTR) revealed only VEGFxxx (194 bp). Qualitative assessment of the PCR products suggests that treatment with either growth factor increased VEGFxxx expression. The same RT-PCR strategy using three different extracts (1, 2, 3) from HEK 293 cells similarly revealed only VEGFxxx (194 bp). (-tem) corresponds to reactions containing water instead of cDNA. (-RT) corresponds to reactions containing water instead of reverse transcriptase.

Figure 6. Characterisation of VEGF isoform expression in mouse and human cDNA extracts using VEGFxxxb isoform-specific RT-PCR.

A, RT-PCR using an exon 4 forward primer (DB exon4) and an exon8 b/7 isoform-specific reverse primer (DB165b/189b-1) failed to detect any PCR products in commercial human tissue cDNAs (AMS Biotechnology Ltd.). B & C, RT-PCR using an exon 3 forward primer and a reverse primer designed to amplify VEGF164/188 (164b/188b-1) detected a single PCR product in mouse fibrosarcoma cell lines, as well as a number of PCR products in solid fibrosarcomas and normal CBA and SCID mouse tissues. Product a represents a truncated VEGF species containing exons 3, 4 & 8b. Product b represents a product identical in sequence to VEGF164b. Product c represents a product exhibiting 100% identity to murine acetyl co-A acetyl transferase. D, No products corresponding to VEGF120b were detected in mouse cDNAs when RT-PCR was performed using the same exon 3 forward primer but a reverse primer designed to amplify VEGF120b (120b-1).

Figure 7. Discriminating VEGF isoform-specific RT-PCR highlights the importance of primer design in influencing detection of putative VEGFxxxb products.

RT-PCR was performed using an exon 3 forward primer together with different VEGFxxxb isoform specific reverse primers designed to detect VEGF188b and/or VEGF164b or VEGF120b. A, RT-PCR using a reverse primer spanning 13 bases across the exon8b/exon7 junction (164b/188b-2) amplified a putative VEGF164b (311 bp) product in wt, VEGF164 & VEGF120-expressing tumours and a putative VEGF188b product (383 bp) in VEGF188-expressing tumours. B, More discriminating RT-PCR using a 164b/188b-3 reverse primer spanning only 4 bases across the exon8b/7 splice-junction failed to reveal VEGFxxxb PCR products. The ability to detect VEGFxxx products in reactions using exon8 and 3′UTR-C reverse primers highlights efficient PCR conditions. C, No products corresponding to VEGFxxxb isoforms were generated by RT-PCR using cDNA from mouse fibrosarcoma cells and a reverse primer spanning 5 bases across the exon8b/7 junction (164b/188b-1); VEGF188b (1209 bp) and VEGF164b (1137 bp) products were amplified from pPNTVEGF¹⁶⁴ (V¹⁶⁴) & pPNTVEGF¹⁸⁸ (V¹⁸⁸) ‘knock-in’ plasmid vector cDNA templates . D & E, Putative VEGF188b and/or VEGF164b products were detected in RT-PCR reactions using reverse primers spanning 13 bases across the exon8b/7 junction (164b/188b-2 & 164b/188b-4) in fibrosarcoma cells and pPNTVEGF¹⁶⁴ & pPNTVEGF¹⁸⁸ plasmids. F & G, Products corresponding to VEGF120b (179 bp) species were readily detected in all fibrosarcoma tumour and cell extracts with a VEGF120-specific reverse primer spanning 16 bases across the exon8b/5 junction (120b-2), whilst no VEGF120b products were generated when the reverse primer spanned only 5 bases across the 8b/5 junction (120b-1).

Figure 8. Exon splice junction sequences in the murine and human VEGF-A genes.

Exon junction sequences are shown for the murine VEGF-A gene (A) and human VEGF-A gene (B), highlighting the sequence TGCAG, which exists at the end of exons 3, 4 & 7 (in-frame stop codons are asterisked). Since any VEGF164b/188b reverse primer sequence will contain sequence complementary to these bases, this highlights the possibility of obtaining truncated VEGF PCR products. The sequence CAG also occurs within the 3′UTR, immediately prior to the start of the predicted exon8b sequence (shown in italics), demonstrating the need for VEGF164b/18b isoform-specific primers to contain a minimum of four bases that span the exon8b/exon7 junction.

Figure 9. Discriminating VEGF isoform-specific RT-PCR using human cDNA extracts further highlights the importance of primer design.

A, Human cDNAs from colon, prostate and kidney (Primer Design Ltd) were amplified in RT-PCR reactions using an exon 4 forward primer together with either DB165b/189b-1 (lane 1), h165b/189b-2 (lane 2), h165b/189b-3 (lane 3), h165b/189b-4 (lane 4) or h165b/189b-5 (lane 5). Primer sequences are highlighted in Table 1 & Fig 2. A putative VEGF165b (∼212 bp) product is evident in lanes 2 & 3, but absent in lane 1. All three reverse primers spanned 5 bases across the 8b/7 junction, however differed in their melting temperatures (see main text), with h165b/189b-2 & 3 exhibiting more compatible Tms with the forward primer. A similar 165b product was more abundantly amplified in lanes 4 & 5 using reverse primers spanning more bases (13 or 14 respectively) across the 8b/7 splice junction. In addition, these latter reverse primers also amplified a product corresponding to 188b (∼283 bp). However DNA sequencing revealed that the product amplified in lanes 2 & 3 was not VEGF but LIM domain only, protein isoform 7, confirming that detection of apparent VEGFxxxb species was only evident using 8b/7 reverse primers with increasing complementary sequence to exon7. B, Similar results were obtained in RT-PCR reactions using cDNA extracted from a normal kidney patient biopsy and amplified with the exon 4 forward primer and the DB165b/189b-1 (lane1), h165b/189b-2 (lane 2) and h165b/189b-4 (lane 4) reverse primers. C, RT-PCR using the DBexon7/DB3'UTR forward/reverse primers capable of simultaneously amplifying VEGFxxx & VEGFxxxb, confirmed only VEGFxxx (194 bp) amplification in the human samples tested.