p12 represents the smallest, so far poorly characterized subunit of the mammalian DNA polymerase δ (polδ) heterotetramer. Previously, to gain a molecular understanding of endothelial cell activation by fibroblast growth factor-2 (FGF2), we identified an upregulated transcript in FGF2-overexpressing murine aortic endothelial cells (FGF2-T-MAE cells) showing 89% identity with human p12. Here, we cloned the open reading frame of the murine p12 cDNA and confirmed the capacity of overexpressed or exogenously added FGF2 to upregulate p12 mRNA and protein in endothelial and NIH3T3 cells with no effect on the other polδ subunits. p12 expression was instead unaffected by serum and different mitogens. Also, anti-p12 antibodies decorated FGF2-T-MAE cell nuclei and their chromosome outline during metaphase. Small interfering RNA-mediated knockdown of p12 caused a significant decrease in FGF2-driven proliferation rate of FGF2-T-MAE cells, in keeping with a modulatory role of p12 in polδ activity. Immunoistochemistry of FGF2-embedded Matrigel plugs and FGF2-overexpressing tumor xenografts demonstrated a nuclear p12 staining of angiogenic CD31+ endothelium. p12 immunoreactivity was also observed in the CD45+/CD11b+ inflammatory infiltrate. Thus, FGF2 upregulates p12 expression in endothelial cells in vitro and in vivo. p12 expression in infiltrating inflammatory cells may suggest additional, cell proliferation-unrelated functions for this polδ subunit.
The B family of DNA polymerases (pol), which comprises polα, polδ, and polɛ, is essential for the replication of the eucaryotic genome (Sugino, 1995; Waga and Stillman, 1998; Hubscher et al., 2000). Polα-primase is required for the synthesis of short primers that serve to initiate the synthesis of both the leading strand at the replication origin and the Okazaki fragments on the lagging strand (Lehman and Kaguni, 1989; Waga and Stillman, 1994). The elongation of the leading strand and completion of Okazaki fragments are catalysed by polδ (Hubscher et al., 2000). Also, polδ may displace downstream RNA–DNA primers and refill the gaps (Waga and Stillman, 1998). Polδ shares several features with polɛ: both polymerases are involved in DNA replication, repair, and recombination (Waga and Stillman, 1998) even though the two enzymes are specialized for different kinds of templates, like euchromatine vs heterochromatin, transcribed vs nontranscribed regions, near vs far from origins (Karthikeyan et al., 2000). Consistent with its role in lagging strand DNA synthesis, polδ is required for telomerase-mediated telomere addition in vivo (Diede and Gottschling, 1999).
The mammalian polδ is composed of at least four subunits (Liu et al., 2000). Two subunits form a tightly associated core heterodimer consisting of the catalytic p125 subunit, which has both 5′ to 3′ polymerase and 3′ to 5′ exonuclease activities, and the p50 subunit. The enzymatic activity of polδ is upregulated when quiescent cells are induced to proliferate (Yang et al., 1992) and p125 protein levels show a 2–3-fold increase at the G1/S phase of the cell cycle (Zeng et al., 1994). Also, p125 subunit interacts with and becomes phosphorylated by cyclin D3/Cdk4 and cyclin E/Cdk2 (Wu et al., 1998), thus suggesting a possible role in S phase checkpoint control. Additionally, p66/68 (Hughes et al., 1999) and p12 (Liu et al., 2000) were identified as part of the mammalian polδ complex. Whereas p66/68 binds PCNA, a specific role for p12 subunit has not been identified, although its addition to an in vitro assay enhances the DNA polymerizing activity of the enzyme (Podust et al., 2002).
Angiogenesis is characterized by increased microvessel endothelial cell proliferation, production and/or activation of matrix degrading enzymes, migration in the subendothelial matrix, and differentiation into functional new blood capillaries (Carmeliet, 2000). The local, uncontrolled release of angiogenic growth factors and/or alterations of the production of natural angiogenic inhibitors (Hanahan and Folkman, 1996) are thought to be responsible for the uncontrolled endothelial cell proliferation that takes place during tumor neovascularization and in angioproliferative diseases (Folkman, 1995). Thus, the identification of selective targets in angiogenic and/or transformed endothelium may have significant implications for the development of antiangiogenic therapies (Liekens et al., 2001). In this contest, gene expression profiles have been described for endothelial cells activated in vitro (Glienke et al., 2000; Kahn et al., 2000; Roland et al., 2000; Wang et al., 2000) or in vivo during tumor angiogenesis (St Croix et al., 2000). The results have allowed the identification of several known and unknown gene transcripts upregulated in endothelial cells under these conditions.
Fibroblast growth factor-2 (FGF2) represents a prototypic angiogenic growth factor (Basilico and Moscatelli, 1992). To gain a molecular understanding of FGF2-triggered endothelial cell activation, we compared the gene expression profiles between tumorigenic FGF2-overexpressing murine aortic endothelial cells (FGF2-T-MAE cells) and parental MAE cells by subtractive suppression hybridization (Dell’Era et al., 2002). Several genes were differentially expressed in FGF2-activated endothelial cells and, among them, we identified a transcript that showed 89% identity with the human polδ p12 subunit (Dell’Era et al., 2002).
Since the identified putative p12 transcript contained only a small portion of the gene, a cDNA library was constructed from FGF2-T-MAE cells and screened by PCR using 5′-IndexTermACCAAGAACTTCAGGACAGACATCA forward and 5′-IndexTermAACACTTGGTACACCTCTAGGGGG reverse primers to clone the full-length transcript. The positive isolated phage carried an 844 bp insert (GeneBank Accession number AF515709) with a translated coding sequence of 107 amino acids (Table 1 ). Blasting the sequence to the Ensembl Genome Browser (http://www.ensembl.org) showed that the p12 gene is located on mouse chromosome 19.
Northern blot analysis confirmed the upregulation of p12 transcript in FGF2-overexpressing FGF2-T-MAE cells, pZipFGF2-MAE cells, and retrovirally FGF2-infected MAE cells (Dell’Era et al., 2002) when compared to parental MAE cells (Figure 1a). Also, increased steady-state levels of p12 mRNA were observed in MAE cells transfected with FGF4 cDNA (pZipFGF4-MAE cells) (Dell’Era et al., 2001), suggesting that other members of the FGF family can mimic FGF2-mediated p12 upregulation. Also, Northern blotting (Mouse Multiple Tissue Northern Blot, BD-Bioscience) showed that p12 is widely expressed in murine adult tissues, the highest steady-state levels of p12 mRNA being detectable in heart and liver, followed by brain, spleen, and kidney. p12 expression is instead very low in lung and testis and below the limits of detection in skeletal muscle (data not shown).
To assess whether FGF2 is able to upregulate p12 expression in endothelial cells also when administered exogenously as a recombinant protein, MAE cells, microvascular murine brain endothelial (MBE) cells (Bastaki et al., 1997), and NIH3T3 cells were serum-starved for 48 h and stimulated with either recombinant FGF2 or 10% fetal calf serum. Then, steady-state p12 mRNA levels were evaluated by semiquantitative RT–PCR analysis. As illustrated in Figure 1b, FGF2 causes a significant upregulation of p12 expression in both endothelial cell types and in NIH3T3 cells. In contrast, serum exerted a very limited effect, if any, on p12 expression in all the cell lines tested. The capacity of FGF2 to cause p12 upregulation is specific since the growth factor does not affect the expression of the other polδ subunits p125, p50, and p66/68 (not shown). Also, other mitogenic stimuli (like vascular endothelial growth factor in MBE cells, epidermal growth factor in NIH3T3 cells, or phorbol-12-myristate-13-acetate in all the cell lines) were unable to affect p12 mRNA levels (data not shown).
To obtain a specific anti-p12 antiserum, we produced a recombinant glutathione-S-transferase (GST)-p12 chimeric protein using the GATEWAY™ cloning technology (Invitrogen). GST-p12 was purified from salt-induced BL21-SI™ competent cell (Invitrogen) extract by glutathione-affinity chromatography and injected in rabbits. Total IgG fraction was purified from the serum of immunized animals and used in a Western blot. As shown in Figure 1c, the anti-p12 antibody recognized a recombinant 6xHis-p12 chimera and an immunoreactive p12 protein of Mr 12 000 in the extract of control MAE cells, FGF2-T-MAE cells, control MBE cells, and FGF2-treated MBE cells. In agreement with transcript analysis, the levels of p12 protein were increased in FGF2-overexpressing and FGF2-treated cells when compared to control cells. Immunolocalization experiments demonstrate that p12 protein is localized in the nucleus of FGF2-T-MAE cells (Figure 3a) as well as of parental MAE cells (not shown). This is in keeping with the presence of a putative nuclear localization signal at amino-acid residues 3–19 in the p12 sequence (Table 1). Also, immunoreactive p12 protein coprecipitates with polδ p125 subunit when a cell extract was incubated with a specific anti-p125 antiserum (Figure 1d), confirming the in vivo association between the two proteins (Liu et al., 2000). Furthermore, p12 immunoreactivity decorates the chromosome outline during mitosis, underlying the role of p12 in nuclear DNA-linked processes (Figure 3b). Costaining of FGF2-T-MAE cells with antibodies against various cyclins (D1, E, B1) revealed that the levels p12 protein, at variance with p125 subunit (Zeng et al., 1994), are not modulated during the cell cycle (data not shown), in keeping with DNA microarray analysis of cycling HeLa cells (http://genome-www.stanford.edu/Human-CellCycle/Hela/index.shtml).
p12 subunit is not essential for DNA replication in yeast (Hubscher et al., 2000). On the other hand, p12 protein increases the DNA polymerizing activity of polδ in a cell-free assay (Podust et al., 2002), even though no data are currently available on the role of p12 in an in vivo mammalian system. On this basis, interference RNA was used to assess the impact of p12 subunit expression in FGF2-driven endothelial cell proliferation. As shown in Figure 2, transient transfection with diced p12 small interfering RNA (siRNA) effectively downregulated p12 protein levels and caused a significant decrease in the proliferation rate of FGF2-T-MAE cells whose proliferation is largely dependent upon the autocrine stimulation exerted by the overexpressed growth factor (Sola et al., 1997). In contrast, although polδ is required for telomerase-mediated telomere addition in vivo (Diede and Gottschling, 1999), the p12 silenced population did not show any reduction in telomerase activity (Figure 2c).
Given a possible role of p12 in FGF2-activated endothelial cells in vitro, we evaluated the capacity of FGF2 to cause p12 upregulation in an in vivo angiogenesis model. To this purpose, recombinant FGF2 was administered s.c. in C57BL/6 mice via a Matrigel plug (Passaniti et al., 1992). After 7 days, immunohistochemical analysis demonstrated a strong p12 immunoreactivity in the nucleus of FGF2-activated CD31+ endothelial cells infiltrating the plug (Figure 3c). No p12 immunoreactivity was observed in control plugs devoid of the growth factor nor in the resting endothelium of several adult murine organs, including heart, liver, kidney, and lung capillaries (data not shown).
To confirm the ability of FGF2 to upregulate p12 expression in endothelial cells in vivo, we performed immunohistochemical analysis of FGF2-overexpressing xenografts originated by the s.c. injection in nude mice of the human endometrial adenocarcinoma Tet-FGF2 cell line (Giavazzi et al., 2001). These cells originate tumor lesions whose growth and vascularization is driven by the transduced growth factor (Giavazzi et al., 2001). Again, nuclear p12 immunoreactivity was detectable in tumor endothelium (Figure 3d, inset). Interestingly, we also observed numerous p12+ cells surrounding the tumor blood vessels. Double immunostaining showed that these cells were positive for the pan-leukocyte marker CD45 (Figure 3e, inset), whereas α-smooth muscle actin+ mural cells did not show any p12 immunoractivity (Figure 3f). A similar nuclear p12 immunostaining was also seen in CD11b+ cells infiltrating the FGF2-treated Matrigel plugs (data not shown). CD11b antigen is expressed mainly on inflammatory cells (Larson and Springer, 1990), all of them being responsive to FGF2 (Takagi et al., 2000; Ohsaka et al., 2001). Taking together our data indicate that FGF2-activated endothelium as well as infiltrating inflammatory cells express p12 in vivo. Further experiments will be required to characterize unambiguously the p12+ leukocyte subpopulation(s).
In conclusion, in the present work, we demonstrate the capacity of FGF2 to upregulate polδ p12 subunit expression in endothelial cells in vitro and in vivo. To our knowledge, this represents the first experimental evidence for a modulation of p12 expression in mammalian cells. Also, p12 silencing causes a significant decrease in the proliferation rate of FGF2-activated endothelial cells, in keeping with the ability of p12 protein to increase the activity of the polδ enzyme in vitro (Podust et al., 2002). The capacity to increase p12 expression in endothelial cells appears to be restricted to FGF family members since other mitogenic stimuli, including vascular endothelial growth factor, epidermal growth factor, phorbol ester, and serum, are ineffective. Thus, p12 upregulation is not the mere consequence of an increase in the proliferation rate of endothelial cells but represents a more specific response to FGF-mediated activation of the endothelium. Finally, p12 expression in infiltrating inflammatory cells may suggest additional, cell proliferation-unrelated functions for this polδ subunit. In keeping with this hypothesis, preliminary observations indicate that p12 upregulation occurs also in pheochromocytoma PC12 cells during neuronal differentiation triggered by FGF2 or nerve growth factor, raising the possibility that the FRS2 signaling pathway shared by these growth factors is involved in the response (P Dell’Era, unpublished observations).
Basilico C and Moscatelli D . (1992). Adv. Cancer Res., 59, 115–165.
Bastaki M, Nelli EE, Dell'Era P, Rusnati M, Molinari-Tosatti MP, Parolini S, Auerbach R, Ruco LP, Possati L and Presta M . (1997). Arterioscl. Thromb. Vasc. Biol., 17, 454–464.
Carmeliet P . (2000). Nat. Med., 6, 389–395.
Dell'Era P, Belleri M, Stabile H, Massardi ML, Ribatti D and Presta M . (2001). Oncogene, 20, 2655–2663.
Dell'Era P, Coco L, Ronca R, Sennino B and Presta M . (2002). Oncogene, 21, 2433–2440.
Diede SJ and Gottschling DE . (1999). Cell, 99, 723–733.
Folkman J . (1995). Nat. Med., 1, 27–31.
Giavazzi R, Giuliani R, Coltrini D, Bani MR, Ferri C, Sennino B, Tosatti MP, Stoppacciaro A and Presta M . (2001). Cancer Res., 61, 309–317.
Glienke J, Schmitt AO, Pilarsky C, Hinzmann B, Weiss B, Rosenthal A and Thierauch KH . (2000). Eur. J. Biochem., 267, 2820–2830.
Hanahan D and Folkman J . (1996). Cell, 86, 353–364.
Hubscher U, Nasheuer HP and Syvaoja JE . (2000). Trends Biochem. Sci., 25, 143–147.
Hughes P, Tratner I, Ducoux M, Piard K and Baldacci G . (1999). Nucleic Acids Res., 27, 2108–2114.
Kahn J, Mehraban F, Ingle G, Xin X, Bryant JE, Vehar G, Schoenfeld J, Grimaldi CJ, Peale F, Draksharapu A, Lewin DA and Gerritsen ME . (2000). Am. J. Pathol., 156, 1887–1900.
Karthikeyan R, Vonarx EJ, Straffon AF, Simon M, Faye G and Kunz BA . (2000). J. Mol. Biol., 299, 405–419.
Larson RS and Springer TA . (1990). Immunol. Rev., 114, 181–217.
Lehman IR and Kaguni LS . (1989). J. Biol. Chem., 264, 4265–4268.
Liekens S, De Clercq E and Neyts J . (2001). Biochem. Pharmacol., 61, 253–270.
Liu L, Mo J, Rodriguez-Belmonte EM and Lee MY . (2000). J. Biol. Chem., 275, 18739–18744.
Ohsaka A, Takagi S, Takeda A, Katsura Y, Takahashi K and Matsuoka T . (2001). Inflamm. Res., 50, 270–274.
Passaniti A, Taylor RM, Pili R, Guo Y, Long PV, Haney JA, Pauly RR, Grant DS and Martin GR . (1992). Lab. Invest., 67, 519–528.
Podust VN, Chang LS, Ott R, Dianov GL and Fanning E . (2002). J. Biol. Chem., 277, 3894–3901.
Roland I, Minet E, Ernest I, Pascal T, Michel G, Remacle J and Michiels C . (2000). Eur. J. Biochem., 267, 3567–3574.
Sola F, Gualandris A, Belleri M, Giuliani R, Coltrini D, Bastaki M, Tosatti MP, Bonardi F, Vecchi A, Fioretti F, Ciomei M, Grandi M, Mantovani A and Presta M . (1997). Angiogenesis, 1, 102–116.
St Croix B, Rago C, Velculescu V, Traverso G, Romans KE, Montgomery E, Lal A, Riggins GJ, Lengauer C, Vogelstein B and Kinzler KW . (2000). Science, 289, 1197–1202.
Sugino A . (1995). Trends Biochem. Sci., 20, 319–323.
Takagi S, Takahashi K, Katsura Y, Matsuoka T and Ohsaka A . (2000). Acta Haematol., 103, 78–83.
Waga S and Stillman B . (1994). Nature, 369, 207–212.
Waga S and Stillman B . (1998). Annu. Rev. Biochem., 67, 721–751.
Wang JL, Liu YH, Lee MC, Nguyen TM, Lee C, Kim A and Nguyen M . (2000). Microvasc. Res., 59, 394–397.
Wu SM, Zhang P, Zeng XR, Zhang SJ, Mo J, Li BQ and Lee MY . (1998). J Biol. Chem., 273, 9561–9569.
Yang CL, Chang LS, Zhang P, Hao H, Zhu L, Toomey NL and Lee MY . (1992). Nucleic Acids Res., 20, 735–745.
Zeng XR, Hao H, Jiang Y and Lee MY . (1994). J. Biol. Chem., 269, 24027–24033.
This work was supported by grants from AIRC, MIUR (Centro di Eccellenza ‘IDET’, Firb 2001, and Cofin 2002), and ISS (Oncotechnological Program) to MP.
About this article
Cite this article
Dell'Era, P., Nicoli, S., Peri, G. et al. FGF2-induced upregulation of DNA polymerase-δ p12 subunit in endothelial cells. Oncogene 24, 1117–1121 (2005). https://doi.org/10.1038/sj.onc.1208359
- DNA polymerase