Specific interaction of heterogeneous nuclear ribonucleoprotein A1 with the -219T allelic form modulates APOE promoter activity

Abstract

The polymorphic -219T/G variant in the APOE promoter has been associated with variations in basal transcriptional activity as well as with the risk of developing Alzheimer's disease, myocardial infarction and early-onset coronary heart disease. The molecular mechanisms underlying these effects are presently unknown. In this report, we show that nuclear extracts from Jurkat cells form a T-specific complex with a motif including the -219 site within the APOE promoter. By DNA-affinity chromatography and mass spectrometry, the human heterogeneous nuclear ribonucleoprotein hnRNPA1(A1) was identified as one component of the complex. In vitro binding analysis indicated that a fragment of A1 had a marked binding specificity for the T form. Interaction of A1 with this region is driven by an adjacent telomeric-like sequence; however, the presence of G, but not T, at -219 position inhibited this interaction. The differences in transcriptional activity between the -219T and -219G promoter allelic forms correlated with the expression levels of A1 in several cell lines; also, over-expression of A1 increased the activity of the T form relative to that of the G form. These results indicate that A1 transactivates APOE promoter activity by direct and specific interaction with the -219T site.

Figure 1

Analysis of the interaction of nuclear proteins from Jurkat cells with the –219 site of APOE promoter. Oligonucleotide probes spanning this site and having either T or G at –219 position (219T and 219G), or T at –219 and either A at –218 or C at –220 positions (218A and 220C) were analysed by EMSA using nuclear extracts from Jurkat cells. (A) Analysis of the interaction of 219T and 219G probes with proteins from nuclear extracts (N.E.). (B) Competition analysis of 219T binding by cold T- and G-containing probes. (C) Comparative analysis of the formation of the complex a with 219G, 218A and 220C probes (right panel) and competition analysis of –219T binding by cold 219T, 219G, 218A and 220C probes (left panel). In (B) and (C), cold probes (Cp) were pre-incubated at a 50-fold excess. The protein–DNA complexes are indicated at the left of each panel.

Figure 2

Evidence that A1 is part of the nuclear protein complexes formed by 219T probe. (A) EMSA analysis of the formation of complex a by the fractions eluted from the 219T affinity column. The column fractions were numbered by E1 to E8. (B) Identification of A1 in the 219T column eluates. Fractions were analyzed by SDS–PAGE followed by Coomassie staining; the indicated band was in gel digested and the resulting peptides analyzed by mass spectrometry. Bands identified as hnRNPs A2/B1 and A1 are indicated. (C) Evidence that A1 is specifically retained by 219T column. Fractions eluted from the 219T column and the other two control columns (NR1 and NR2) were assayed by western blot analysis using an anti-A1 antibody. The arrow indicates the position of A1. (D, E and F) Evidence that A1 forms part of the main nuclear complex formed by 219T in the presence of nuclear cell extracts. In (D), fraction E8 eluted from the 219T affinity column was incubated with 219T and 219G probes, subjected to UV-induced crosslinking and analyzed by SDS–PAGE followed by autoradiography (left panel). Nuclear extracts from Jurkat cells were incubated with 219T probe and subjected to immunoprecipitation using an anti-A1 (α-A1) or a control antibody (α-APP); the precipitated material was then subjected to UV-induced crosslinking, followed by SDS–PAGE analysis and autoradiography. The positions of the A1-containing, specific 219T-protein complex (p50) and of the non-specific complex (p52) are indicated. In (E), Jurkat nuclear extracts inmunodepleted with α-A1 and α-APP antibodies were incubated with 219T and 219G probes, subjected to UV- induced crosslinking, and analyzed by SDS–PAGE and autoradiography. In (F), Jurkat nuclear extracts (N.E.) inmunodepleted with α-A1 and α-APP antibodies were incubated with 219T probe and subjected to EMSAs. The position of complex a is indicated with an arrow. Molecular weight markers are indicated on the left in (B), (C) and (D). A non-specific band is indicated by an asterisk in (D) and (E).

Figure 3

Analysis of the interaction of A1 with –219T. (A) Sequence and denomination of the different DNA probes used for UV-crosslinking (B, C, D and E) and EMSA analysis (F). (B) Interaction of A1r with 219T and a non-related, double stranded DNA control probe. The A1r-219T complex was indicated by an arrow. (C) Interaction of A1r with 219T and 219G probes. In the indicated lanes, samples were previously incubated with an excess amount (50-fold) of the corresponding cold probes (Cp). The minor, –219T-specific protein complexes are indicated by arrows. (D) Interaction of UP-1r with 219T and 219G (upper panel). The 219T specific complex is indicated by an arrow. After autoradiography, the gel was reswollen and Coomassie Blue-stained (lower panel); the observed protein bands correspond to UP-1r. (E) Interaction of UP-1r with different, –219-related probes and competition binding analysis with an excess amount (100-fold) of cold oligonucleotides (Cp). (F) Formation of a complex in the presence of Jurkat nuclear cell extracts using different –219-related probes and competition analysis (50-fold) of 219T binding by 219T, 219G and Telmut cold probes (Cp). The complex a is indicated by an arrow.

Figure 4

Effect of A1 in the relative APOE promoter activity of the two –219 allelic forms. (A and C) Comparative analysis of the basal activity of the two –219 allelic forms of APOE promoter in different cell lines. CB3(A1) is the cell line CB3 stably transfected with A1. (B and D) Western blot analysis using an anti-A1 antibody of the expression of this protein in the different cell lines. Loading controls (LC) for equal protein amounts were performed by Ponceau S staining of the blot (B) and by immunostaining with anti-tubulin antibody (D). Data are representative of at least three independent experiments, performed in triplicate, and are expressed as mean ± SEM. Differences between the activity of –219T and G forms were analyzed by Student’s t-test. (* p < 0.04, ** p ≤ 0.01).