Decreased mitochondrial OGG1 expression is linked to mitochondrial defects and delayed hepatoma cell growth

Abstract

Many solid tumor cells exhibit mitochondrial respiratory impairment; however, the mechanisms of such impairment in cancer development remain unclear. Here, we demonstrate that SNU human hepatoma cells with declined mitochondrial respiratory activity showed decreased expression of mitochondrial 8-oxoguanine DNA glycosylase/lyase (mtOGG1), a mitochondrial DNA repair enzyme; similar results were obtained with human hepatocellular carcinoma tissues. Among several OGG1-2 variants with a mitochondrial-targeting sequence (OGG1-2a, -2b, -2c, -2d, and -2e), OGG1-2a was the major mitochondrial isoform in all examined hepatoma cells. Interestingly, hepatoma cells with low mtOGG1 levels showed delayed cell growth and increased intracellular reactive oxygen species (ROS) levels. Knockdown of OGG1-2 isoforms in Chang-L cells, which have active mitochondrial respiration with high mtOGG1 levels, significantly decreased cellular respiration and cell growth, and increased intracellular ROS. Overexpression of OGG1-2a in SNU423 cells, which have low mtOGG1 levels, effectively recovered cellular respiration and cell growth activities, and decreased intracellular ROS. Taken together, our results suggest that mtOGG1 plays an important role in maintaining mitochondrial respiration, thereby contributing to cell growth of hepatoma cells.

Fig. 1

Mitochondrial respiratory defects may be associated with decreased mtOGG1 expression. (A, B) Chang- L cells (Ch-L) and four different SNU hepatoma cell lines (SNU354, SNU387, SNU423) were cultured for 2 days to maintain exponentially growing state. (A) Cell lysates were subjected to Western blot analysis. (B) Cellular oxygen consumption rate (OCR) was measured using XF analyzer as described in “Materials and Methods”. ^**, < 0.01 vs. Chang-L. (C) Six human HCC tumor samples (T) and their surrounding tissues (S) were applied to Western blot analysis.

Fig. 2

Decreased mtOGG1 expression occurs with mtDNA damage. (A) Schematic diagram of mtDNA. Bold lines with alphabets indicate the regions sequenced. Dotted line indicates the site for ND2 probe used for Southern blot analysis. ^* is the unique Nhe I site and # indicates the newly found Nhe I site in Chang-L and SNU387 cells. (B) Total genomic DNA was subjected to Southern blot analysis as described in “Materials and Methods” and integrity of mtDNA was detected with a mtDNA-specific probe which was generated against ND2 region (4851–5364 of mtDNA). (C) Several regions [bold lines a, b, c and d of (A)] of mtDNA were amplified and sequenced to examine mutation of mtDNA. Right panel comparison of representative sequence profiles between Chang-L and SNU423.

Fig. 3

Decreased expression of mtOGG1 is mainly due to decreased OGG1-2a mRNA expression. (A) Schematic cDNA structure of OGG1 variants which were generated by alternative splicing. Each box represents exons and dotted lines indicate alternatively spliced joining between OGG1 variants. Blunted arrow pairs on the top of each cDNA structure indicate coding regions. Arrows on the box represent primer positions used for RTPCR. (B) Expression levels (a) of mRNAs of OGG1-2 isoforms in Chang-L cells were examined by RT-PCR with F1/R2 primer sets (upper panel) and F2/R2 primer sets (lower panel). Quantifications of the bands are shown (b). (C) Total mRNAs were isolated from Chang-L and SNU hepatoma cells and subjected to RT-PCR using F1/R2 primer set and primer set for β-actin and mRNA levels of three OGG1-2 (mtOGG1) isoforms were quantitated (a). Right panel shows a representative RT-PCR result. OGG1-2a mRNA levels of Chang- L and SNU hepatoma cells were compared by RTPCR with F2/R2 primer sets (b). (D) mRNA levels of OGG1-1 (ncOGG1) isoforms were detected by RT-PCR using F1/R1 primer set.

Fig. 4

Decreased mtOGG1 expression-associated mitochondrial dysfunction may also be associated to delayed cell growth and increased intracellular ROS. Chang-L cells (Ch-L) and four different SNU hepatoma cell lines (SNU354, SNU387, SNU423) were cultured for 2 days to maintain exponentially growing state. (A) Cell growth rates were measured by counting trypan blue positive cells. No clear dead cells were observed. (B) Western blot analysis. (C) Intracellular ROS levels were monitored by flow cytometric analysis after staining cells with DCFH-DA. (D) Mitochondrial ROS levels were monitored by flow cytometric analysis after staining cells with MitoSOX fluorescence dye. ^**, < 0.01 vs. Chang-L cell.

Fig. 5

Knockdown of mtOGG1 expression decreases cellular respiration and cell growth. Chang-L cell was treated with the indicated amounts of siRNAs for OGG1-2a for 48 h. (A) mRNA levels were monitored by RT-PCR with the F1/R2 primer set (shown in Fig. 3A). (B) Cellular oxygen consumption rates were measured using Mitocell respirometer as described in “Materials and Methods”. (C) Cell growth rates were monitored by counting trypan blue positive cells. No clear dead cells were observed. (D) Intracellular ROS levels. (E) Western blot analysis. ^**, < 0.01 vs. negative control siRNA (si-NC).

Fig. 6

Overexpression of mtOGG1 recovers cellular respiration and cell growth. OGG1-2a (mtOGG1) was stably expressed in SNU423 that has low level of mtOGG1 and a missense mutation on mtDNA. Three clones with different expression levels of mtOGG1 were selected. (A) Western blot analysis. To detect complex II flavoprotein (Fp), more sensitive antibody (#MS204, Mitoscience, USA) was used. (B) Total cell homogenates of mtOGG1 clone #11 was subjected to subcellular fractionation. Lamin B (nucleus), α-tubulin (cytosol) and COX2 (mitochondria) proteins were used as organellar markers (T, total; N, nuclear; C, cytosol; and M, mitochondria). (C) Cellular oxygen consumption rate were monitored using Mitocell respirometer. (D) Cell growth rates were measured by counting trypan blue positive cells. No clear dead cells were observed. (E) Intracellular ROS levels with DCFH-DA staining. ^**, < 0.01; ^*, < 0.05 vs. mock.