Evidence that P12, a specific variant of P16(INK4A), plays a suppressive role in human pancreatic carcinogenesis

Abstract

The INK4a-ARF locus plays a central role in the development of pancreatic tumors as evidenced by the fact that up to 98% of pancreatic tumor specimens harbored genetic alterations at the INK4a-ARF locus. Interestingly, in addition to the well-known P16(INK4A) (P16) and P14ARF tumor suppressors, the INK4a-ARF locus in pancreas encodes another protein, P12, whose structure, function, and contributions to pancreatic carcinogenesis remain to be elucidated. In the current study, we demonstrated that over-expression of p12 in human pancreatic cancer cells led to cell arrest at the G1 phase and such cell cycle arrest was related to down-regulation of a number of oncogenes, such as c-Jun, Fos, and SEI1. Furthermore, unlike P16, P12 did not retain any cyclin-dependent kinase 4 (CDK4)-inhibitory activity. Instead, P12 exhibited a transactivating activity not found in P16. We also examined the genetic status of p12 in a cohort of 40 pancreatic tumor specimens and found that p12 alteration was prevalent in pancreatic tumors with an incidence of 70% (28/40). These results support that P12 is a tumor suppressive protein distinct from P16, and its genetic inactivation is associated with pancreatic carcinogenesis.

Fig, 1

p12 gene, P12 protein, and its biochemical properties. (A) The structure of the INK4a-ARF locus. Rectangles represent DNA and mRNA, and cylinders represent proteins. e1, e2, in1: exons 1, 2, and intron 1 of p15^INK4B; E1β, E1α, E2, E3: exons 1β, 1α, 2, 2γ, and 3 of p16^INK4A; In1: intron 1 of p16^INK4A; kD, kilo Daltons. Sizes of coding regions and proteins are not in proportion strictly. (B) The amino acid sequence of P12. The novel intron-derived C-terminus is highlighted and underlined. (C) In vitro CDK4 inhibition assays. The reaction mixtures included 3 units of CDK4-cyclin D2, 50 ng of GSTRb379-928, 5μCi of [γ-32P] ATP, and varying amounts of P12 (•) and P16 (○) proteins. The incorporation of ³²P was quantitated using a PhosphoImager. Assays were performed in triplicate. (D) Transactivation assays. p12 cDNA was cloned into pM GAL4 DNA-binding domain vector in ORF, and pG5-Luc is used as a reporter vector. The empty pM vector and pM-p53 were used as negative and positive controls, respectively. The relative transactivation activity (%) was defined as the luciferase activity with the target / the luciferase activity with p53. Assays were performed in triplicate.

Fig. 2

Quantitative determination of mRNA expression of selected genes in cells. (A) Gene expression in p12-transfected Panc-1 cells. pcDNA3.1-transfected Panc-1 cells were used as control for normalization. (B) Gene expression in p12-transfected AsPc-1 cells. pcDNA3.1-transfected AsPc-1 cells were used as control for normalization. In both A and B, HPRT1 was used as an endogenous control. Values (y-axis) represent fold-changes (in the log form) calibrated to the expression levels of indicated genes in control cells. Error bars represent standard deviations. Dashed lines represent 2-fold increases (upper) and decreases (lower) in gene expression. Assays were conducted in triplicate.