Article Figures & Data
Figures
- Fig. 1.
Nuclear localization of the PLAG1 protein. Immunofluorescence of Cos-1 cells cotransfected with the PLAG1 expression vector construct, pCAGGS-PLAG1 and with the pM3 expression vector expressing the DNA binding domain of Gal4 (16). The cells were costained with the PLAG1 antibody, 4′,6-diamidino-2-phenylindole and the GAL4 antibody (see “Materials and Methods”). This allows the visualization in the same cells of PLAG1 (A), nuclear DNA (B), GAL4 (C), or all three (D).
- Fig. 2.
Determination of the PLAG1 binding site. Alignment of the 23 oligonucleotides selected by seven cycles of CASTing using GST-PLAG1 (N2-C244). The frequency of each of the bases at each position is shown at the bottom of the figure, and N represents the number of oligonucleotides that carried a base at that position.
- Fig. 3.
The G-cluster is recognized by finger 3 and the core by fingers 6 and 7 of PLAG1. A, nucleotide sequences of the different oligonucleotides used in EMSA analysis. Mutations are underlined. B, EMSAs performed with equimolar amount (0.6 pmol) of bacterially expressed GST-PLAG1 proteins. GST-PLAG1 (N2-C244) (Lanes 1–5), (N84-C244) (Lanes 6–10), (N101-C244) (Lanes 11–15), (N159-C244) (Lanes 16–20) and (N2-C203) (Lanes 21–25) were incubated with the probes WT2 (Lanes 1, 6, 11, 16, and 21), mCLU2 (Lanes 2, 7, 12, 17, and 22), mCO2 (Lanes 3, 8, 13, 18, and 23), mCLUmCO2 (Lanes 4, 9, 14, 19, and 24), and WT2ml (Lanes 5, 10, 15, 20, and 25) as described in“ Materials and Methods”. The percentage of binding of all these mutants on the different probes were compared with the binding of F1–F7 to the probe WT2 and is the mean of at least six experiments. C, competition experiments performed on the probe WT2 in presence of increasing amounts of unlabeled double-strand oligonucleotides (10, 30, 100, 300, and 1000 ng) using recombinant full-length PLAG1 expressed in vitro in reticulocytes lysates; D, EMSAs performed with recombinant PLAG1 proteins produced in vitro in reticulocytes lysates. Wild-type PLAG1 (Lanes 1–5), F2mut (Lanes 6–10), F3mut (Lanes 11–15), and F7mut (Lanes 16–20) were incubated with the probes WT2 (Lanes 1, 6, 11, and 16), mCLU2 (Lanes 2, 7, 12, and 17), mCO2 (Lanes 3, 8, 13, and 18), mCLUmCO2 (Lanes 4, 9, 14, and 19), and WT2ml (Lanes 5, 10, 15, and 20) as described in “Materials and Methods”. Equal efficiency of protein expression was obtained for the different constructs as demonstrated by SDS-PAGE of proteins labeled with [35S]methionine (data not shown). The percentages of binding of all of these mutants on the different probes were compared to the binding of the wild-type PLAG1 to the probe WT2 and is the mean of at least three experiments.
- Fig. 4.
PLAG1 binds effectively to the P3 promoter of IGF-II. A, EMSAs performed with wild-type PLAG1 protein produced in vitro in reticulocyte lysates and incubated with the probes WT2 and P3-4. B, competition experiments performed on the probe WT2 in the presence of increasing amounts of unlabeled double-strand oligonucleotides (10, 30, 100, 300, and 1000 ng) using wild-type PLAG1 expressed in reticulocyte lysates.
- Fig. 5.
IGF-II P3 transcript is up-regulated in tumors with PLAG1 overexpression. A, Northern blot analysis of normal salivary gland (n.s.g.) tissues and pleomorphic adenomas hybridized sequentially with a 3.7-kb PLAG1 cDNA probe, an IGF-II exon 9 probe, and a 2-kb β-actin probe. RNAs tested included samples from three different normal salivary gland tissue specimens (Lanes 1, 5, and 8) and from adenomas c895 (Lane 2), c904 (Lane 3), cg650 (Lane 4), cg644 (Lane 6), and cg580 (Lane 7). B, recapitulation of the Northern blot analysis of normal salivary gland tissues and pleomorphic adenomas hybridized with probes specific for the genes encoding PLAG1, IGF-II, PDGF-B, GOS24, or c-Ha-Ras. The karyotype of the tumors has been described elsewhere (1, 2), and in tumors cg650 and cg601, the breakpoint occurs outside the PLAG1 region (20, 21).
- Fig. 6.
Prediction and schematic representation of PLAG1 binding consensus site. A, amino acids at position −1, 2, 3, and 6 (numbering with respect to the start of the α-helix) within the PLAG1 zinc fingers are shown in the first column. Bases predicted to be preferred for binding by these amino acids are shown in the second column (22, 23). The consensus found by CASTing is shown in the third column, and thick lines indicate that the predicted base matches with the selected one. B, comparison between the PLAG1 binding site and other reported consensus binding sites like the Zac1 consensus (27), one of the characterized WT-1 binding sites (33), the consensus sequence for Sp1 binding described as the decanucleotide 5′-(G/T)GGGCGG(G/A)(G/A)(C/T)-3′ (34) and the Egr-1/Zif268 consensus binding sequence (35).
Tables
- Table 1
PLAG1 can stimulate transcription through its consensus binding site
Two hundred ng of the indicated reporter luciferase (luc) construct were cotransfected into the fetal kidney cell line 293 together with 200 ng of the expression vector pCAGGS-PLAG1 or the empty vector pCAGGS. PLAG1 induction levels are expressed as the ratio of luciferase activity obtained in the cell transfected with pCAGGS-PLAG1 expression vector versus the activity obtained in cells transfected with the empty vector pCAGGS. The data are means ± SE of at least two independent transfection experiments, each performed in triplicate.
PLAG1 induction (WT)6-TK-luc 18.9 ± 2.6 (mCLU)6-TK-luc 1.0 ± 0.2 (mCO)6-TK-luc 0.9 ± 0.2 (mCLUmCO)6-TK-luc 0.6 ± 0.1 TK-luc 1.7 ± 0.6 - Table 2
Identification of human promoters with potential PLAG1 binding sites and determination of their capacity to be induced by PLAG1
The EPD (18) has been screened for the presence of the PLAG1 consensus GRGGC(N)7 RGGK with the pattern-matching algorithm implemented in the program findpatterns of the GCG software package (32). The list includes all of the human promoters present in the EPD containing at least two PLAG1 DNA binding consensus in their promoter. The ability of each promoter to be induced by PLAG1 has been estimated by cotransfection of the fetal kidney 293 cell line with pCAGGS-PLAG1 or pCAGGS expression vectors, together with reporter constructs in which each promoter has been cloned in front of a luciferase gene (see“ Materials and Methods”). PLAG1 induction levels are expressed as the ratio of luciferase activity obtained in the cell transfected with pCAGGS-PLAG1 expression vector versus the activity obtained in cells transfected with the empty vector pCAGGS. The data are means ± SE of at least two independent transfection experiments, each performed in triplicate.
PLAG1 induction 5 consensus : IGF-II promoter 3 7.9 ± 0.8 4 consensus : GOS24/TTP/Zfp36 3.0 ± 0.2 3 consensus : c-Ha-Ras promoter 3 0.9 ± 0.2 2 consensus : PDGF-B/c-sis 1.4 ± 0.6 TnI slow ND Gastrin ND VLDL ND SOD-1 ND Ia-ass. g′ ND H19 ND 0 consensus : pGL2 basic 0.7 ± 0.2 pSLA3 1.2 ± 0.1 RSV 1.1 ± 0.2 prohormone convertase 2 0.5 ± 0.03 somatostatin 1.0 ± 0.3
Volume 60, Issue 1
