Transcriptional regulation of mouse L-selectin

Abstract

L-selectin mediates the initial tethering and rolling of lymphocytes in high endothelial venules. To study the transcriptional regulation of mouse L-selectin, we cloned 4.5 kb 5'-flanking sequences of the mouse sell. Luciferase analysis of serial 5'-deletion mutants showed that the first 285 bp was sufficient to drive high promoter activity in EL4 cells, but not in Sell-negative HeLa cells, suggesting that this fragment harbors the minimal mouse sell promoter and contains cis-elements for lymphocyte-specific expression. Site-directed mutagenesis and chromatin immunoprecipitation showed that Mzf1, Klf2, Sp1, Ets1, and Irf1 bind to and activate the mouse sell promoter. Over expression of these transcription factors in EL4 cells increased expression of sell mRNA. Silencing the expression of Sp1 by siRNA significantly decreased sell promoter activity in EL4 cells. We conclude that sell transcription is regulated by Mzf1, Klf2, Sp1, Ets1, and Irf1.

Figure 1

Mapping of the transcription initiation sites. The sequence shown is from nucleotide - 131 to +3. The translation start codon ATG is shown in italic. The -50 and -100 positions are labeled with a star underneath each. Identified transcription initiation sites indicated by arrows on top, numbers represent the frequency of the given end in twenty sequenced transformants.

Figure 2. Promoter activity of the 5’ flanking sequence of mouse sell

(A) Schematic representation of the serial 5’ deletion reporter constructs. (B) Luciferase activity from transiently transfected mouse EL4 (open bars) or HeLa (black bars) cells. Luciferase activity expressed as percentage of pGL3-Promoter (a SV40 promoter driven reporter represented as SV40) transfected EL4 or HeLa cells. Data presented are Mean ± SD of at least three independent experiments in triplicate. (C) Alignment of 5’-flanking sequences of mouse, rat, chimpanzee, and human sell. Annotated sequences of the first ~260 bp of 5’-flanking sequences. The numbers shown on the top of the arrowheads are based on the mouse sequence. Putative transcription factor binding sites are underlined.

Figure 3. Ets1, Mzf1, Klf2, Sp1, and Irf1 activate mouse sell promoter activity

(A) Luciferase activity from transiently transfected EL4 cells with mSell285 (285), 285E, 285M, 285S, 285I. (B) EL4 cells were co-transfected with mSell285 and pcDNA3.1-Ets1, -Mzf1, -Klf2, -Sp1, and -Irf1, respectively, as shown in black bars or co-transfected with mSell285 or 285E, 285M, 285S, and 285I and its corresponding transcription factor expression plasmids (open bars). pcDNA3.1 (shown as Vec) was used as a control to exclude the promoter effects of the vector backbone. Luciferase activity of wild-type mSell285 alone or cotransfected with Vec was set at 100. Data presented are Mean ± SD of at least three independent experiments in triplicate; mean values are compared by unpaired t-test. P<0.05 is considered significant.

Figure 4. Ets1, Irf1, Klf2, Mzf1, and Sp1 bind L-selectin promoter in vivo, and upregulate sell expression

(A) Representitive gel of ChIP analysis showing the in vivo binding of transcription factors to the sell promoter. The eluted genomic DNA was amplified by PCR and visualized with ethidium bromide on a 1.5% agarose gel. Lane 1, 1 kb plus ladder; lane 2, 1% input; lane 3, no antibody; lane 4, normal rabbit IgG; lane 5, Ets1 antibody; lane 6, Irf1 antibody; lane 7, Klf2 antibody; lane 8, Mzf1 antibody; and lane 9, Sp1 antibody. (B) Representative gel of RT-PCR showing Ets1, Irf1, Klf2, Mzf1, and Sp1 upregulate sell mRNA expression. Lane M, 1.0 kb plus ladder; lane 1, basal level of L-selectin expression, lane 2 to lane 7 are L-selectin expression in EL4 cells transfected with Ets1, Irf1, Klf2, Mzf1, and Sp1 expressing plasmids, respectively. (C) Ratio of sell to GAPDH mRNA expression in each group. Data presented are Mean ± SD of at least three independent experiments in triplicate; mean values are compared by unpaired t-test. P<0.05 is considered significant.

Figure 5. Ets1 contributes to lymphocyte-specific sell promoter activity

(A) Luciferase activity from HeLa cells transiently co-transfected with mSell285 and one of the transcription factor expression plasmids were analyzed and graphed as percentage of SV40. (B) Luciferase activity from HeLa cells transiently co-transfected with mSell285 and an increasing amount of Ets1 expression plasmid shown in the figure. Luciferase activity of SV40 was set at 100. Data presented are Mean ± SD of at least three independent experiments in triplicate.

Figure 6. Silencing of Sp1 down-regulates mouse sell promoter activity

(A) EL4 cells were transiently transfected with 0.8 μg (lane 1) of Si-RNA-A (a non-targeting 20-25 nt siRNA designated as a negative control), or 0.4 or 0.8μg of Sp1-SiRNA (lanes 2 and 3) respectively and incubated for 36 hours. Ten μg of cell lysates were resolved on 10% SDS-PAGE, and Sp1 was detected by Western blot. Molecular weight markers indicated with numbers. β-actin was used as loading control (bottom panel). (B) EL 4 cells were transiently co-transfected with mSell105, mSell185, and mSell285 and either 0.8 μg of Sp1-SiRNA or 0.8 μg of Si-RNA-A respectively. Luciferase activity was analyzed 36 hours after co-transfection and the luciferase activity from Si-RNA-A co-transfected group was set at 100. Data presented are Mean ± SD of at least three independent experiments in triplicate; mean values are compared by unpaired t-test. P<0.05 is considered significant.