Sox2 targets ATG10 to induce autophagy.

<p>(<i>A</i>) Microarray. <i>Top panel</i>, HCT116 colorectal cancer cells stably expressing <i>mock</i> or <i>Sox2</i> were subjected to microarray analysis as described in “Materials and Methods”. The yellow color indicates the number of genes that did not show a significant difference between <i>mock</i> and <i>Sox2</i> expression. The red and green colors indicate the number of genes up- or down-regulated, respectively, in HCT116 cells stably expressing <i>Sox2</i> compared with cells expressing the <i>mock</i> control. <i>Bottom table</i>, summary of autophagy-related genes obtained from the microarray analysis. Up- and down-regulation are denoted in Log2 values. (<i>B</i>) Expression of <i>ATG10</i> induced by Sox2. Total RNA (1 µg) from HCT116 cells infected with <i>mock</i> or <i>Sox2</i> was reverse transcribed and <i>ATG10</i> was amplified by cycle-dependent PCR and the <i>ATG10</i> expression level was visualized by agarose gel electrophoresis at the 21^st cycle. ß-Actin was used as an internal control to verify equal utilization of cDNA for PCR. (<i>C</i>) Up-regulation of ATG10 and LC3b protein levels induced by <i>Sox2</i> expression. Proteins were extracted from HCT116 cells stably expressing <i>mock</i> or <i>Sox2</i> and ATG10 and LC3b proteins were visualized by Western blotting using specific antibodies. ß-Actin was used as an internal control to verify equal protein loading. (<i>D</i>) Confirmation of Sox2-induced ATG10 protein level. HCT116 colorectal cancer cells were infected with <i>mock</i> or <i>Sox2</i> and cultured 5 days. The cells were fixed, hybridized with an ATG10 specific primary antibody and Alexa 488-conjugated secondary antibody. The ATG10 protein levels were observed by fluorescence microscopy. L.M. indicates the same area of light microscopy corresponding to fluorescence microscopy (X200). (<i>E</i>) Construction of the <i>ATG10</i> promoter <i>luciferase</i> reporter plasmid. The nucleotide sequences of the human <i>ATG10</i> promoter region were downloaded from Ensemble (<a href="http://uswest.ensemble.org" target="_blank">http://uswest.ensemble.org</a>). The putative promoter analysis was conducted using TFSEARCH (v1.3) (<a href="http://cbrc.jp/htbin/nph-tfsearch" target="_blank">http://cbrc.jp/htbin/nph-tfsearch</a>). The putative SRY and Sox binding consensus sequences are denoted in red and the polymerase III binding site is boxed. (<i>F</i>) The BAC clone (RP11-111B20) was purchased from Empire Genomics (Buffalo, NY) and 1528 and 711 bp of the <i>ATG10</i> promoter region were amplified by PCR. The PCR fragment was recombined with the <i>pGL3</i> basic vector to construct pGL3-AT10-1582 and pGL3-ATG10-711 <i>luciferase</i> reporter plasmids. The <i>pGL3-ATG10-luc</i> reporter plasmids were confirmed by DNA sequencing. The <i>ATG10</i> promoter <i>luciferase</i> reporter plasmids, <i>pGL3-AT10-1582</i> and <i>pGL3-ATG10-71-</i>luc, were transiently transfected into HCT116 cells and the firefly luciferase activity was analyzed after 24 h. The <i>phRL-SV40 renilla</i> luciferase reporter plasmid was co-transfected as an internal control to verify equal transfection and normalization of firefly luciferase activity (*p<0.001). (<i>G</i>) The <i>pGL3-ATG10-1528</i> luciferase reporter plasmid was transiently co-transfected with a <i>mock</i> or <i>Sox2</i> viral expression vector into HCT116 cells and the firefly luciferase activity was analyzed after 24 h. The <i>phRL-SV40 renilla</i> luciferase reporter plasmid was co-transfected as an internal control to verify equal transfection and normalization of firefly luciferase activity (*p<0.001).