Methylation in maize mitochondrial DNA --Gracia Zabala, Carmen Oliver, Susan Gabay-Laughnan and John R. Laughnan Several researchers have reported the absence of methylation at the internal cytosine residue in 5'-C.C.G.G-3' sequences in mitochondrial DNA (mtDNA), based on the identical restriction patterns obtained with the pair of isoschizomers MspI (cuts 5'-C.C.G.G-3'and 5'-C.m5C.G.G-3') and HpaII (cuts 5'-C.C.G.G-3' but not 5'-C.m5C.G.G-3') (Groot and Kroon, Biochim. Biophys. Acta 564:355, 1979; Bonen et al., FEBS Let. 111:340, 1980; Borck and Walbot, Genetics 102:109, 1982). These results prompted the prevalent belief that mtDNA is, in general, unmethylated.

We have analyzed DNA methylation of mtDNA from a cytoplasmic male-sterile strain (cms-S) and a cytoplasmic revertant (CR) obtained from the sterile strain, using two pairs of m5C-sensitive and -insensitive restriction enzyme isoschizomers, HpaII and MspI, as well as EcoRII and BstNI. EcoRII cannot cleave the sequence 5'-C.m5C.N.G.G-3' (N = A or T), whereas BstNI can . Both enzymes digest the unmethylated sequence 5'-C.C.N.G.G-3'.

Figure 1. Ethidium bromide-stained agarose gel of EcoRII- (E) and BstNI-digested (B) mtDNA of sterile (S) and cytoplasmic revertant (CR) samples. The arrows to the left of the figure indicate major differences in the digestion patterns of the methylation sensitive enzyme (EcoRII) and its isoschizomer BstNI. Molecular weights are given in kb.

As in previous reports (Groot and Kroon, Biochim. Biophys. Acta 564:355, 1979; Bonen et al., FEBS Let. 111:340, 1980; Borck and Walbot, Genetics 102:109, 1982), we found no differences in the restriction patterns obtained with MspI and HpaII, indicating the absence of methylation of the CpG type. In contrast, the restriction patterns obtained with the pair of isoschizomers EcoRII and BstNI revealed several differences. Figure 1 shows three DNA fragments of 8.3, 6.2 and 5.3 kb present in the EcoRII (E) digests of both sterile (S) and cytoplasmic revertant (CR) mtDNA which are absent from BstNI digests of identical samples. A 3.0 kb fragment is also present in the BstNI digests, but is missing from the EcoRII digests. Since reactions were carried out in the presence of 1mM spermidine to facilitate DNA digestion, differences in restriction patterns between samples digested with these two enzymes are not the result of an artifactual partial digestion by EcoRII. Changes in DNA and enzyme concentration as well as incubation time did not alter the digestion patterns shown in Figure 1. We have repeated this experiment with two other sets of mtDNA samples and the results have been consistent.

In addition to differences in restriction patterns observed in ethidium bromide-stained agarose gels, Southern blot analyses with mtDNA probes have revealed further restriction fragment polymorphisms between the EcoRII and BstNI digests. Figure 2 shows three bands (3.8, 3.3 and 2.5 kb) hybridizing to the probe M2C1 (specific for the cox1 mitochondrial gene) in the EcoRII digests which are absent in the BstNI digests. The low abundance of these additional bands hybridizing to the M2C1 probe in the EcoRII digests compared to that of the 2.26 kb unique hybridization signal in the BstNI digests suggests that only a portion of cox1 or its flanking sequences is methylated.

Figure 2. Hybridization of M2C1 (specific probe for the mitochondrial cox1 gene) to a Southern blot of EcoRII- (E) and BstNI-digested (B) mtDNA from sterile (S) and cytoplasmic revertant (CR) samples. The arrows to the left of the figure indicate DNA fragments that the methylation sensitive enzyme EcoRII does not digest but that its isoschizomer BstNI does. Molecular weights are given in kb.

Southern hybridization analysis with the IS1E5 probe (a mtDNA sequence neighboring the cox1 sequence) also revealed incomplete digestion with EcoRII (data not shown). In contrast, no significant difference was observed when Southern blots were hybridized to a mtDNA probe containing the atpA gene (data not shown).

Our results suggest that the mtDNA of six different maize samples has methylated sequences of the CpNpG type. We have found no differences in methylation between mtDNA from sterile and revertant samples. It will be of interest to determine whether methylation plays a role in transcriptional regulation during mitochondrial development or stress, as has been shown in the chloroplast of sycamore and tomato (Ngernprasirtsiri et al., PNAS 85:4750, 1988; Ngernprasirtsiri et al., Plant. Physiol. 88:16, 1988).


Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors

Return to the MNL 64 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage