Interactions of Mitochondrial and Nuclear Genes That Affect Male Gametophyte Development

Apart from their agronomic importance in hybrid seed production, mutations that encode cytoplasmic male sterility (CMS) provide a means to probe the role of the mitochondrion in reproductive development. Fertility restorers are examples of nuclear genes that affect cytoplasmic gene expression, and their identification can illuminate the interactions between the two genomes. In this review, we consider what is known about the nature of mutant mitochondrial loci that disrupt pollen development, focusing on their creation through recombination events that have often involved ATP synthase subunit genes. We also discuss how the mutant mitochondrial genes' function or expression is affected by the presence of nuclear fertility restorers and the information gained about these nuclear genes through recent map-based cloning efforts. We also describe the evidence that mitochondrial gene expression can affect the function of nuclear gene products that control floral development. In natural plant populations, the most widespread manifestation of disturbed mitochondrial–nuclear interaction is altered floral development, particularly the loss of the male gametophyte. A substantial fraction of species exhibit populations that include both hermaphroditic (male-fertile) and functional female (male-sterile) individuals, a situation termed gynodioecy by Darwin (discussed by Budar and Pelletier, 2001; Charlesworth, 2002). Although the genetics of the male sterility has not yet been determined in many known gynodioecious species (reviewed by Kaul, 1988), a typical finding is cytoplasmic inheritance of male sterility and restoration of fertility by one or more nuclear alleles. Although non-Mendelian traits in plants could be encoded by either cytoplasmic genome, in all cases to date in which such a sterility-encoding gene has been identified in an organelle, the mitochondrion was the location of the mutation that disrupts pollen development. CMS forces outcrossing; however, the potential advantage of outcrossing does not appear to explain completely the success of the gynodioecious reproductive strategy. Quite possibly, the selective advantages and disadvantages of a mixture of female and hermaphroditic plants may vary between individual genera (Kaul, 1988; Budar et al., 2003). For some species, there is sound evidence that female plants produce more seeds, perhaps because of reduced energetic investment in male floral organs. Indeed, in many species, CMS halts pollen development at a very early developmental stage, potentially saving considerable output of resources. However, no increase in seed set in females has been detected in some species. Furthermore, in other species, the disruption occurs late in development, after considerable energy has been expended. Such a finding raises the question of whether cytoplasmic genes that confer male sterility may somehow confer a survival advantage to individuals, so that more CMS plants reproduce, even if individuals do not produce more seeds. The potential adaptive value of gynodioecy and its impact on the population structure and evolution of species remain topics for further inquiry (recently reviewed by Charlesworth, 2002; Budar et al., 2003; Saur Jacobs and Wade, 2003). Understanding the nature of CMS-encoding cytoplasmic genomes and the nuclear genes that suppress male sterility may be instructive in creating models for the maintenance of the balance of female and hermaphroditic plants in natural plant populations. In addition to the naturally occurring CMS that has been observed in wild plant populations, the trait has been synthesized a of in which the nuclear of one species has been the cytoplasmic of The of pollen development that a of is termed male Quite the finding of CMS after from the of cytoplasmic genes expression is by nuclear genes in the species. However, there also is evidence that the of can be perhaps creating cytoplasmic a of in cytoplasmic and (reviewed by and CMS individuals that and also have after individuals often are plant populations in and are detected The of CMS and fertility restoration have been by plant to hybrid of a of species and seed from a plant from For species in which the is a or a the of does not the or value of the hybrid and the of pollen may even be or seed is the from a hybrid however, the presence of a nuclear fertility is to confer on the hybrid on the for species, the of these genes are to of or The of CMS-encoding and fertility restorers a in plant Furthermore, to the of genes can the to for hybrid the that encode an is to cytoplasmic genomes in and CMS a the has not been between two often may and have no is is to CMS and are of that have from known CMS such to the of of and is to in or events have been A for mitochondrial is to the of a the In most species, and mitochondrial genomes are and both are through the has a means to the of and and to mitochondrial genomes (reviewed by of between CMS and has that fertility does not the In and that mitochondrial genomes have been to for to the identification of loci Budar and Pelletier, The genetics is to species in which and plant are and if mitochondrial genomes are in so identified can be further by of gene on the expression of genes that A in which synthesized by in CMS and are may a if is synthesized in in to be For et to a by this strategy. the development of may be to in the by the mitochondrial in CMS and the most is to the mitochondrial or genomes in and CMS to for by of the of nuclear fertility restorers on expression of the genes. is not because a may affect the of not a gene also that of a a gene could be in CMS if to be affected by a The evidence from a of altered expression in either CMS in or of to fertility A that known CMS-encoding genes are is their from recombination between mitochondrial genes and or between mitochondrial of the for recombination in plant mitochondrial the identification of a gene that has been by recombination is not to that is a in can genes that have from recombination and et al., may not be and the expression of may have no For if an is to the or of a gene is or the be to the A of have in which an gene or has been detected in a of CMS not in a of some these be to be to because may the of mitochondrial genomes. Indeed, the a of a between a gene and known to be loci for which there is evidence for a role in CMS are are by the of of for loci to be a are in the genes for of ATP of of genes for of genes. of in plant of the can be in in the The to be identified completely also a gene that from a of recombination In a was identified that CMS in et al., to the finding of a the gene which is of of the of and an termed and The also the of the and genes from an in et al., The of evidence this in CMS was the of altered gene expression in the presence of the of that at the are reduced in and and the of the products of the gene is reduced and a that is to a that a of on and et al., The to be and is in is the The identification of and two of both loci from recombination events that the was not in the two genes et al., was perhaps for if there been between the two for genes in CMS of other species may have that CMS have been is how often the events ATP synthase subunit gene and of if the gene does not ATP synthase often the is an In the of the subunit is in the In not does the gene provide the for the the and gene was a of the subunit of ATP the ATP synthase was to a to was to be in the of the and to exhibit some of the of from et al., 2003). For of this review, genes are to the for the A gene has been to encode subunit et al., 2003). to this gene are in many other plant mitochondrial genomes and have been termed in Furthermore, a a to that of was identified in ATP synthase et al., 2003). a such genes are to The gene in CMS in and to the in the of the of the and genomes of CMS and this for the a was not between the two detected and 1988; et al., the genetics in to the identification of the of and the nuclear from the an that was CMS of a gene termed to the gene et al., to in which the gene was or not et al., evidence for its role in pollen The also the detected in et al., The also is reduced in et al., and et al., however, no in the was detected in CMS plants and et al., the in was a gene to was detected to The gene in and altered in was not the was observed not to in plants et al., ATP synthase genes also in the of two other and was in the because the of the and exhibit one and in between CMS and detected by for known mitochondrial genes to of altered on the the a gene was is of the of to of and to and et al., The presence of the the of the in CMS and The of between the and that of the in the the be of the in the CMS in the identification of a gene termed of is not is the of and a gene termed The of is not to of the gene are in other mitochondrial genomes. the to the of the of the and genes are The is affected by the presence of the a to CMS Although most of CMS has the and two CMS have been for the presence of mitochondrial genes. is either or the of from CMS a in CMS et al., of the a gene to by a to a of the A was observed in CMS not in in et al., the gene the is not genes of the CMS by between and also have been that encode mitochondrial genes in CMS and for in expression and to the identification of a a to that of to an The products of this are not yet could include a encoded by both the and because and can one of et al., 2003). The also is in a of an between and to A CMS from termed is the most and mitochondrial of the in CMS and plants a that in between in the CMS plant and that of The of between the two and also was affected by the presence of a fertility and this the was to the of a that of et al., et al., The of are to of et al., 2003). In of the gene are reduced in not in et al., termed by the a to in of CMS not et al., et al., is the ATP synthase gene in the and most other CMS fertility restoration in is the are and pollen are of is that cytoplasmic in populations. of in on a two and are affected by the presence of the et al., The pollen population is at a that a potentially in the of a the from et al., 2002). The of the gene is is a gene that of the of is of from the et al., 2002). products of the have not been fertility restoration of the is A gene termed a an to the of et al., The of has of to of by a in which of to of the The in the from that in CMS is an finding of altered was not plants and an in CMS and is not The gene in the for to be to a gene that the of of an of known to be However, also occurs in et al., The of and in the genes could be to in the of the of the two genes and of have also been to be in recombination events that to the which have of the recombination events of other genes may to the of the genes. has in recombination events in two In most is known at the about the CMS from either or a nuclear Such can be to fertility by nuclear genes from of the a gene termed which has at its and is an from a in the CMS not in the or and et al., 2002). also is in one of Although CMS are for the of hybrid one of these has been at the the which is to fertility by a gene termed plants mitochondrial genomes to that a of was CMS et al., The CMS an of termed which is to at its the of are to of in the et al., 2002). is the of the of to the in et al., The of in this are altered in the presence of et al., and the presence of the of perhaps a of et al., the products of and in the CMS have not been is not known whether their or is altered in The has the known that have in recombination events to produce genes and in CMS A to all is the of that have no to plant The of these a because have no to or plant nuclear the of completely genomes. In one species, the of the gene is of of A was identified in by CMS and and a that was termed and was to a The is in after in the a CMS et al., Furthermore, which does not affect the of the The of was to in reproductive of CMS plants et al., et al., was further the of pollen disruption in plants the gene from a nuclear to be male et al., other genes from other species, such and et al., et al., et al., have not in perhaps because the expression or the mitochondrial not the expression of the mitochondrial The may have in CMS plants the gene is to the the of pollen et al., The of a mitochondrial that male sterility has not been determined for species. that many to reproductive in CMS and is of reproductive for and such have been on that the however, whether the disruption occurs in the or a for further in many species. and have some to the disruption that in pollen development. The has been to because not CMS also to by of the and to the In the presence of the a in the mitochondrial et al., from the nuclear and to the mitochondrion plants to be to both and the plants not male et al., perhaps because of the of the which is in a in to pollen is not is in pollen is A by an perhaps some in in a to or to and of the of genes CMS has that many of and a of the encoded have been to be or the mitochondrial However, whether the disrupt a the or perhaps a by the is not mitochondrial have been between of CMS and of a of species (reviewed by and For in have been detected in from and CMS et al., these could from a disruption in some other expression in CMS to from the of the mitochondrial et al., Although these and other have in the of of between CMS and whether these are the or of other mitochondrial is not of the genes in that many of include of ATP synthase are to ATP synthase subunit genes. raises the that ATP synthase could be a in the pollen development in CMS in a of species. CMS was to have a in floral et al., et that mitochondrial ATP synthase in from CMS was that in to of and not has been on the disruption of a mitochondrial function by the expression of an or However, for at some of the CMS that the presence of the of the expression of a mitochondrial gene that has a In this the or of and of the in does not pollen development by of a in expression of the mitochondrial are a of examples of in of species that in the of a of the mitochondrial In a CMS that genes the the expression of a such and et al., et al., has been at the and not to be of the expression of the genes in many of the in to be A CMS is an disrupt pollen development. In some such the is the is in reproductive In the gene is by a in of et al., A for pollen could be of a in or of mitochondrial and mitochondrial between in the plant and et al., so could in development. However, is to whether the of a a mitochondrion has in or whether of a mitochondrial are an of an increase in mitochondrial which has been in the and of and the for most of CMS is that have a role to in the male this in development, for energy or mitochondrial products may be so that of mitochondrial function this is the finding that of or in the in et al., 2003). Furthermore, plants that a in a gene that a mitochondrial subunit are male et al., in mitochondrial function is by a its may be of either the or was a many evidence of of the and mitochondrial in (reviewed by and Kaul, in plant appear to a role in for et al., and evidence that in CMS exhibit of is that be to in CMS of many other species the are in CMS and For exhibit the For exhibit altered For CMS and are for and a of pollen in the For in CMS the pollen on the For in CMS have been to or For CMS have no which are of and have that the floral in these CMS the in and in the of a for floral and the expression of of the known floral genes has been in CMS or In of the reduced in et al., 2002). In reduced expression of of and detected et al., 2003). The of the that disturbed expression of the known floral genes could be for floral was by et a CMS the of in the CMS to the plants and some functional pollen et al., 2002). gene can produce such in the floral developmental is an for further Although are known to affect all of the genes in the species the of has not been determined for the of all restorers are known to affect either the or the of the and some have been observed to affect both and A in the of of a that a is the population often for a plant mitochondrial such genes exhibit and be which from not by to how the was of the of the and has that in of a so that two increase in et al., in is both the of a of and a expression of from a that and are involved in of the In the presence of the of and et al., on these and have that the may be genes et al., about loci through map-based cloning (discussed are this of a more one from the in The the presence of of not the also of and and The on could be the of the of the on or perhaps the of two or more genes that on of of the gene has been to be the In addition to the presence of either of two genes that confer fertility and has been the of et al., events in the presence of of the restorers in the exhibit some however, is that the events produce from which be the that could encode no in to CMS et al., A has been in which that could encode the gene even to the are in and in is not known is reduced in the presence of the a in does not that the at the of or is evidence for of the of the in the and in male and in could be the of In mitochondrial that a for a the is be and to whether a is is to whether is on the mitochondrial disruption to CMS often occurs very early in pollen development, are very and is to of for Furthermore, the of CMS plants in many species has that disruption of mitochondrial in the is the of development and may be the of and Kaul, the a of the of the expression of in CMS may not be by of and For such species, in may be In CMS or could be to et advantage of a an altered that in the of for and the in which the two genes are The was to from that could no in floral of the et al., that to affect all other genes to date at the the in in the loss of a of a on the of such a has been detected in other species. the is not is reduced to a et al., the expression of by the of the the or the has not yet been is not known how the of this is known about how the of the mitochondrial is and affected by nuclear gene Indeed, even the of the plant mitochondrial in is not the of a and a of from recombination have been for a of species (reviewed by and However, of a a mixture of and and A to the of of from the of mutations in the which in the loss of et al., of that a gene which is involved in and recombination et al., 2003). The have that may be for recombination or that mutation could in of the of such a to the maintenance of the the question of the of of genes be by their cloning and a gene is a can be that is completely a for the presence of a which to between of CMS and The of the genes to date are not from the of The nuclear fertility gene to be was of which in to fertility to The by was observed to exhibit to et al., of the encoded that has the et al., is a which by a that of are and exhibit no in expression from plants that are at both and are for pollen development in the of may a that can be for by the mitochondrial The role of could be to the of a a of is of both and however, the of its in is not known and 2002). Although all of the other in this review, which a are known to have an on the expression of the is that some of the in some species (reviewed by and Kaul, may include genes that by the pollen through a for the presence of a its The and of the in the was identified by and are at of the have an or for the to be in the are in and are in The of the two are and the are the of the in and the The at in the of in by and The is in that a gene a is in to The function of is not A gene was from an and to a in the to and to exhibit et al., 2002). gene is not in reproductive The gene to be identified was the in and have that the to a that genes et al., 2003; et al., 2003; et al., 2003; et al., 2003). cloning of the which fertility to CMS the to two genes. A gene we have termed fertility to CMS the other gene not et al., 2003). reduced of the A gene from a a in the and a to et al., 2003). et between and to that the to a which was and to genes. further to a of these genes the nuclear of CMS A the gene identified in was to confer fertility and their et et genes to the of the The other two genes of in and was observed to fertility in plants et al., 2003). In on the from these we that the and are and that a is to A gene has been to the expression of a in whether this gene is a is not yet The gene identified by and the of a gene in the that confer fertility in the do not the male sterility of the The CMS is known to the expression of a in from CMS et al., and the CMS and for genes mitochondrial the the genes and the expression of the in observed that a gene termed in the that in CMS plants and 2003). a that was the was to be a to a and for genes. Such genes can be either by or by the information to for of the gene to the The was a in of genes in the nuclear et al., and Although genes also can be in genomes. A was by from in by and is in The most occurring at of the and the are observed to that from the at the in the of these also are a is on The in the and genes also exhibit a of that to the for the have been in gene expression in and in and plant and The and the are for the of and its et al., et al., The gene is for of the and for the of and et al., gene is for the of and 2003). A mutant in which the of the is has an in the gene that the et al., in the gene exhibit and have in the of of the et al., 2003). In the has been in a of et al., in to which is not in a et al., 2003). the encoded by and to or is not yet A have been to to to et al., 2003). A of function was identified in a for which may is not known et al., and have also been to et al., 2001; and 2003). the is to the which is known to and to et al., is that some may and that may a that could or of the of their fertility restorers are to have from genes involved in the of mitochondrial gene expression, a the identification of of a gene often appear to from recombination we on how CMS and fertility restorers may and in mitochondrial genomes appear to be to of from other genomes. from and nuclear considerable of of the of the mitochondrial genomes from plants that a of to genes that encode either known mitochondrial genes or mitochondrial of function et al., et al., et al., 2002). the of the in of the genomes many and not in either of the other two genomes. The for recombination that is of plant may the of that to the an to gene or a of an of a mitochondrial the may an the in an is not at all or at a a recombination may so that the a more for Furthermore, the may remain at a in some in the nuclear genes that control and recombination an increase in the of the for this of from of et to the in and that of the the at in other the expression of the is to the there be to its However, if the the mitochondrial gene genes for mitochondrial expression of the be reduced the expression of genes. is the of a nuclear gene that or or the by of one to on the products of the is the finding of genes at the nuclear loci in and that have been to affect the expression of et al., 2002; et al., 2003; and 2003). also is the finding that of restorers in affect the expression of mitochondrial genes and that the presence of the of a known mitochondrial genes and A may on also to mitochondrial genes. of these may on a the the population and of the nuclear gene may not have to the of genes from the the and mitochondrial genes may be on the function of the nuclear gene that a mitochondrial gene A mutation in such a gene could in fertility restoration be a of the expression of mitochondrial genes in to the in which mitochondrial genes. is the genetics of restoration in the mutations have been identified that the in in male of these mutations are because also are for the expression of mitochondrial genes et al., a was to the of both the and its may to a nuclear gene for for genes in may provide a of mutations in nuclear genes that are for mitochondrial gene expression et al., 2003). more nuclear loci that affect the expression of CMS-encoding are and we how evolution has plants to a mitochondrial that often on the of information in many we can that loci the be of genes that the expression of mitochondrial A of genes that affect the of The genes that by the expression of or the expression of mitochondrial genes. We and for and for and for and and for the The on has been by the of and the