--Pam Cooper, Ed Butler and Kathy Newton

One means of studying the influence of nuclear genes on mitochondrial gene expression is through the use of interspecific hybrids. Maternally transmitted variations in mitochondrial gene expression may arise because of an "incompatibility" between nuclear and mitochondrial genomes which have experienced separate selective pressures.

We are interested in determining whether nuclear background influences the production of transcripts coming from known mitochondrial genes. For our molecular analyses of nuclear-cytoplasmic interactions, we have been using teosinte-maize hybrids. The cytoplasms from the teosintes Zea luxurians (Z.l.), Zea perennis (Z.p.), and Zea diploperennis (Z.d.) were introduced into maize nuclear backgrounds by recurrent backcrosses with the maize inbreds W23 or A619. These materials were generously provided to us by Jerry Kermicle. The inbreds were used as pollen parents 6-9 times before the plants were analyzed.

We have examined the effect of nuclear background on transcript levels for the three mitochondrially encoded subunits of the cytochrome c oxidase complex, subunits 1,2,and 3. Mitochondria were purified from immature ear shoots (cobs) and the RNA extracted (Stern and Newton, Methods Enzymol. 118:488). The RNA was separated on 1.2% agarose-formaldehyde gels and then blotted onto nylon. The Northern blots were then probed with 32P-labelled, cloned cytochrome c oxidase subunit genes, cox1 (Isaac et al., EMBO J 4:1617), cox2 (Fox and Leaver, Cell 26:315), or cox3 (Hiesel et al., EMBO J. 6:29).

Nuclear background influences both the size and number of transcripts from the cox2 gene when Z.p. and Z.d. are the cytoplasm sources. In the A619 background, three major transcripts of 2.2, 1.7, and 1.4kb are detectable, whereas in the W23 background, two transcripts of 2.2 and 2.0kb are visible. No effect of nuclear background is observed in plants possessing Z.l. cytoplasm. In contrast, no differences between nuclear backgrounds were detectable for transcripts from the cox1 and cox3 genes for any of the teosinte cytoplasms.

We conducted genetic analyses to characterize the nuclear gene(s) causing the cox2 transcript differences. The F1 hybrids Z.p.-A619 X W23 and Z.p.-W23 X A619 both possess the three transcripts found in Z.p.-A619. The same is true if Z.d. is the cytoplasm source. When Z.p.-A619 X W23 was self pollinated, 4/13 (30.7%) of F2 individuals tested expressed two transcripts of the sizes found in Z.p.-W23. Thus A619 carries a single nuclear gene responsible for the dominant three-transcript pattern.

Southern analysis of mitochondrial DNA from the teosinte-maize hybrids revealed no differences between the two nuclear backgrounds when probed with cox2, suggesting that all the transcripts originate from a single mitochondrial gene. When Northern blots were probed with different regions of the cloned cox2 gene, both exons of the gene were present in all the major transcripts. The intron of the gene was not present in any of the transcripts; thus intron splicing cannot account for the differences. We are currently exploring whether the differences might arise because of: A) differential processing of regions 5' and/or 3' to the gene, or B) multiple transcription initiation sites for the gene.

The use of teosinte-maize hybrids has enabled us to identify a nuclear gene which regulates cox2 transcript production in the mitochondrion. This gene is specific for cox2, as we could identify no transcript differences for the other two mitochondrial cox genes.

Return to the MNL 63 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage