Can anyone point me to research on mitochondrial and chloroplast genetics of roses? In particular I would like to know if the mitochondrial and chloroplast genes of roses have been sequenced and, if so, the details.
Jeff Palmer at Indiana is the world’s expert in chloroplast and mitochondrial genetics. He’s in their Biology program, was head of it for a couple of sessions, has published quite a few reviews. He spoke at our department last autumn and I don’t recall much rosaceous in it. The thing is the mito genomes are so strongly conserved through phyla that you aren’t likely to find much difference even between widely divergent genera. and plastids have incredibly complex systems of rearrangement in some phyla, essentially no difference of many genera in others.Of course most of the stuff in mitos & plastids is actually nuclear encoded and may be different between species. But I don’t think the genomes of organelles per se can account for a whole lot of the species differences. Whatever it is in cytoplasmic genetics that accounts for the effects seen in crop plants, it is still too subtle to be well understood, except obvious things like cytoplasmic male sterility in maize where we know a single change in a single gene does it.
My curiosity has to do with whether roses plastid genomes are big enough that there might be regional mutants that cause meaningful differences in, say, vigor or pigmentation (carotenoids).
Most importantly, it’s the only way to know whether using a breeder maternally is any different than using it paternally. That, in turn, makes a great deal of difference in the effort of identifying F1 species hybrids. In that regard it’s way easier to use the species parent paternally but the risk is unknowingly dropping some important extrachromosomal genes.
Your point is well taken. So I went looking. Most recent paper, free at PLoS1 is by Dong, Liu, Yu, Wang and Zhou and entitled something like " Highly variable chloroplast markers for evaluation of plant phylogeny at low taxonomic…" They are very interested in peaches and have done sequencing of ct DNA for 5 species. They found enough variable regions that they can distinguish all species, with a combination of just two highly variable regions. So there is variation at the species level, but it is mostly in intergenic regions and introns, not within genes.
However, for us the pertinent point is in their more general method discussion. They collected from GenBank all the sequence info from genera where there were data available for multiple species. Rosa is not among them. The Prunus species are rosids which give you an idea of variability within this general kind of plants. But there doesn’t seem to be real data on say Hulthemia, R minutifolia, foetida, spinossisima, chinensis, multiflora etc. Great project for one of the grad students in plant breeding at the few existing programs dealing with roses.
I’ll keep looking, maybe there’s something hiding out there in the prickles.
Another paper, by Xi et al in 2013 PLoS Genetics incidentally lists higher plants for which mitochondrial DNA sequences have been done. No rose among them. So all we need is money and patience.
A+ for effort, Larry, thanks very much.
We can’t be the only rosacae breeders that would like an answer to this question. For example, I was thinking that the apple people at Cornell would be interested because they have collected a great many cultivars of Malus sieversii from the archaic population in Kazakhstan and are breeding with them. Given the generational lagtime and space requirements they would benefit even more than we would.
Maybe they would be able to shake some Rosbreed money loose for this purpose. Then again, maybe nobody has brought up the question before.
Don and Larry,
I don’t want to sound like the total dufus here, but as an engineer and a retired army officer, my background in botany is miniscule! Oh, I suppose a tad in some science class in college back in the 50’s. I grew up on a farm and know something about agriculture, and learned enuf about growing and hybridizing roses from Mitchie to pass, but could you tell us in plain English what you are looking for? I am sure that there are many others here that are in the same boat with me that are hesitant to ask this question. No problem if you prefer to ignore it.
John
I’ll take a stab at a better explaination but Larry’s the professor here so I hope he’s ready to correct me.
All cells from which plants and animals are made have within them specialized bodies called organelles. Each type of organelle does a particular job for the cell.
Plants and animal cells all have organelles (‘plastids’) called mitochondria. Mitochondria have the job of providing energy to the cell by ‘burning’ sugar in a highly controlled way and storing it in chemical bonds of a compound called ATP.
Plants (but not animals) have additional plastids called chloroplasts. Chloroplasts have the job of turning sunlight into energy and storing it in the chemical bonds of sucrose (table sugar). Put another way, chloroplasts are bodies within the plant cell where water and carbon dioxide are combined to form sugar, which is then used as building blocks for growing the plant.
As you learned in Bio 101, every cell in all living beings carry a nearly complete set of blueprints for the living being. These are what we call genes, made from bits of DNA strung together to form chromosomes. In all but the most simple living beings the DNA is stored in a special organelle called the cell nucleus that has the job of, well, storing and managing DNA.
However, a small but very important portion of the cell’s DNA is not stored in the nucleus. Mitochondria and chloroplasts are not found within the nucleus but each store within themselves some of the DNA needed for their own reproduction and operation.
Moreover, mitochondria and chloroplasts can contain additional DNA which is seemingly not related to their own structure or function but might be important to the cell for other reasons that we don’t yet understand.
Here’s the rub:
Offspring get half of their nuclear genes from momma and half from papa but…
Offspring get their mitochondrial genes and chloroplast genes only from mama.
The reason for this is that sperm (in animals) and pollen (in plants) don’t carry their mitochondria or chloroplasts along with them with they fertilize an ovum. Thus, only the mother’s mitochondria and chloroplasts are present in the fertilized ovum.
As breeders we usually assume that it makes no difference whether we cross the pollen of one cultivar with the ovum of another or vice versa. However, because of these differences of chloroplast and mitochondrial inheritance, this is a potentially false assumption.
If the mitochondria or chloroplasts genes are different between the mating partners then there will be a difference in mitochondrial or chloroplast genes in the offspring based on which parent is paternal and which is maternal.
So, that’s why I asked the question of whether anyone has sequenced the DNA of rose mitochondria and chloroplasts. We just do not know yet whether all such organelles are equal among all species of roses. From what little Larry was able to dig up it seems that there is good reason to think that they are not equal.
If these plastids do, in fact, have different DNA from species to species then the results of pairing mates one way could be profoundly different than pairing them in the other (‘directionality’).
The reason is that mitochondria and chloroplasts are both centers of energy metabolism. As such, all other cellular functions are fundamentally dependent on them. Size, growth rate, time to flowering, health, disease resistance and even less obvious traits like remontancy and rooting ability could all be affected by energetics.
There are also potential directional effects on pigmentation because the genes for the yellow pigments in petals are the same ones that make the green pigments (chlorophyll) in chloroplasts. IIRC at least some of these genes are already known to be present in the chloroplasts of some plants.
From a practical standpoint it makes a difference to breeders because it means that we need to make all crosses reciprocally - both ways - to ensure we get to see all possible variability in offspring from the cross.
It would also be a lot easier to work with species roses if we knew it didn’t matter which way we make the cross. In F1 crosses involving a species rose as maternal parent it is usually difficult to discriminate true hybrids from selflings because all the progeny tend to resemble mama. It is much easier to identify hybrids when the mother is a modern rose and the species rose is the pollen parent because only hybrids will look like the species parent.
The bottom line is that we really do not know whether or what difference it makes to use a particular rose as either the male or female parent and that there is real potential for there to be a profound difference. That difference depends on chromosomal and mitochondrial dna.
Don,
Thank you very much for giving me a start point. When you brought it to what we did study in Bio 101, the cell structure is the key to all living things and the makeup of that cell, the DNA and what they do are key. Now that I know what is behind your quest, I will go back, reread and study what you and Larry have been discussing. Not that I will be a genius or anything close, but at least have a rudimentary knowledge of it. I do remember a discussion that Mitchie and I had with Ralph Moore when he told her, when making her crosses, to first create the bush that she wanted and then hang a different face on it by using it as the mama plant. He could see that it does make a difference like you mention above in the part headed by Here’s the rub…
As I have always said, the forum is a wonderful learning tool. I was wondering what classroom I wandered into at first. Thanks again for your explanation. I hope there are others reading it that harkened back to their Bio 101 days as well for a point to start.
John
Don,
Not necessarily. It depends on the species, and sometimes on the individual. In geraniums (Pelargonium), chloroplasts are also transmitted in the pollen. That’s why crosses between green x variegated give more variegated offspring than the reciprocals.
Darlington (The Evolution of Genetic Systems. 1956) quoted the following breeding results:
G V W Total
G x W 77 13 10 280
W x G 72 28 – 93
Pelargoniums also transmit mitochondria through the pollen.
In the following list (Nagata, etal. Planta, 1999), m+ means mitochondria are carried in the pollen. p+ means plastids are carried.
Pelargonium zonale (geranium) m+ p+
Pharbitis nil (morning glory), Medicago sativa (alfalfa), Actinidia deliciosa (Kiwi fruit) m- p+
Triticum aestivum, Anthirrhinum majus, Lilium longiflorum, Arabidopsis thaliana m- p-
Musa acuminata m+ p-
The tables can be seen more clearly here.
Even species that do not usually transmit plastids or mitochondria through the male may do so occasionally.
Karl, you are correct about exceptions but that just strengthens the argument that it would really help us to know the relevant details about roses.
Pelargonium is interesting in another respect. It has the a huge mitochondrial genome, possibly the largest among plants.
Cytoplasmic effects (relating to plastids, mitochondria and what not) have been observed in the Rosaceae.
I don’t have the following paper, but I’m appending two comments from different sources.
Harland, S.C., and King, E.E.,“Inheritance of Mildew Resistance in Fragaria with special reference to Cytoplasmic Effects,” Heredity 11: 287 (1957). Jour. Bot. 30 (1943): pp. 311-314.
Harland and King (1957) found indications that powdery mildew resistance in diploid Fragaria vesca was dependent on two recessive genes and that its inheritance was complicated by cytoplasmic effects.
Working with mildew resistance in strawberries, Harland and King (1957) found reciprocal differences in crosses between commercial strawberries, Fragaria grandiflora, and a resistant diploid species, F. vesca. These differences persisted for several generations, indicating that true cytoplasmic inheritance was probably involved.
Pretty close Don. Minor point, the word plastid only applies to chloroplasts and their colorless relatives that may be storing starch (e.g. potato tubers) or making lipids in roots. Mitos are just organelles, like a bunch of other things in cells. So far as we know in higher plants there are DNA pieces in both mitos and plastids. Higher animals have DNA only in mitos. Some things closer to ameba have other organelles with DNA too, but we can ignore them for now.
A very important point about plastids is that they are the main site of synthesis of all kinds of interesting lipids, including carotenoids, vegetable oil in seeds etc. But parts of the pathways require reactions in the cytoplasm (outside the plastid) and most of the genes coding the proteins for the pathways, whether inside or outside the plastid are encoded in the nucleus. So we have this very interesting problem that pieces (enzymes and other proteins)) made in one place have to match those made in another. That is apparently where we get “cytoplasmic effects” in hybrids like maize.
Even though the mitos and plastids have very few genes of their own, it is kind of like when you’re fixing a car. Metric pieces may look real similar to english measure, but they sure don’t fit, or work. It only takes one cog or gear of the wrong size to screw it up totally. In mitochondria one gene always present in the mitochondrial DNA is part of the big enzyme complex called cytochrome oxidase. That is the last step in the energy-producing pathway of the cell. So mess with that and you’re dead, or male sterile at least.That’s what happened in maize. Very likely there are more subtle effects of each of the other mito genes.
Same for the chloroplasts. One gene for a part of the enzyme complex that fixes carbon dioxide into sugar, is encoded in the plastid genome. If it isn’t compatible with its partners that come from nuclear genes, the plant doesn’t do so well.
I hope this helps a little. Money would do more.
Money would do more.
How much, and for what purpose, exactly?
Don, I am guessing for research into what yourself and Larry have been discussing, roses and not other crops. It is over my head most of it. But it does make for some interesting reading along with Karls links.
Here’s some more interesting stuff that may not be immediately relevant. However, it is at least possible that specific genes that are nuclear in one species may be found only in the mitochondria or plastids of another species. In other words, some traits may be inherited through nuclear genes in one species, but only through the “cytoplasm” (i.e., exclusively maternal) in the other.
[hr]
Up until now, the three genetic systems were thought to be discrete, each going down its own pathway. But chloroplasts genes have now been found inside plant mitochondria, overturning conventional wisdom. To sum it all up, DNA seems promiscuous – no respecter of privacy and breaking down all isolating genetic barriers. (Ellis, John; “Promiscuous DNA – Chloroplast Genes inside Plant Mitochondria,” Nature, 299:678, 1982.)
It was a surprise when DNA sequences from mitochondria in yeast cells were discovered setting up shop in the nuclear genomes (i.e., the normal genetic endowment of the cell nucleus). Now biologists find that DNA sequences in many species regularly and frequently hop from one genome to another. Genetic material from cell chloroplasts mix with that of the mitochondria and that of the normal nucleus in what seems to be a free-for-all. This genome hopping has earned DNA the adjective “promiscuous.” (Lewin, Roger; “No Genome Barriers to Promiscuous DNA,” Science, 224:970, 1984.)
Michaelis, in several publications, reported on the great cytoplasmic differences among Epilobium species and races. Reciprocal crosses between Epilobium hirsutum and E. luteum were profoundly different. However, other differences were not so obvious.
When the two species were raised in poor, dry soil, one grew short and bushy, while the other was tall and spindly (reduced branching). I don’t recall which was which. However, after repeatedly pollinating the luteum-cytoplasm line by hirsutum, (i.e., luteum x hirsutum) x hirsutum) x hirsutum), etc.) until the nucleus was presumably pure hirsutum, the plants responded to poor, dry soil in the same way as luteum, rather than hirsutum. In other words, the cytoplasm determined the response.
Grant: Genetics of Flowering Plants (1975) wrote:
“The cytoplasmic-genetic differences between Epilobium hirsutum and E. luteum, and between different races of E. hirsutum, affect not only external morphology and pollen fertility of the plants, but also various physiological features. Among the later are temperature tolerance, sensitivity to poisons, susceptibility to parasites, permeability of protoplasm, viscosity of protoplasm, isoelectric point, redox potential, and oxidative enzyme activity. Differences in the activity of oxidative enzymes between different cytoplasms were considered to be a possible cause of the dwarfness of some Epilobium hybrids and the heterosis of others (Michaelis, 1953, 1954).”
So, even when reciprocal crosses look the same, we should not assume that they really are the same until both groups are exposed to various types of stress.
Sager (1972) gives more details, along with some pictures of the Epilobium hybrids, in case anyone is interested.
[quote=Don]
Can anyone point me to research on mitochondrial and chloroplast genetics of roses? In particular I would like to know if the mitochondrial and chloroplast genes of roses have been sequenced and, if so, the details.[/quote]
Systematic Botany (2007), 32(2): pp. 366–378
Phylogenetic Relationships in the Genus Rosa: New Evidence from Chloroplast DNA Sequences and an Appraisal of Current Knowledge
ANNE BRUNEAU, JULIAN R. STARR, and SIMON JOLY
ABSTRACT. The genus Rosa (roses) comprises approximately 190 shrub species distributed widely throughout the temperate and subtropical habitats of the northern hemisphere. Despite numerous recent studies examining phylogenetic relationships in the genus, relationships remain obscure due to problems such as poor identification of garden specimens, hybridization in nature and in the garden, and low levels of chloroplast and nuclear genome variation. Phylogenetic analyses of non-coding chloroplast sequences from the trnL-F region and psbA-trnH intergenic spacer for 70 taxa show slightly more variation than previous analyses of the genus. Bayesian and parsimony analyses suggest that subg. Rosa can be divided into two large clades, each with low internal resolution. One comprises species from sections Carolinae, Cinnamomeae and Pimpinellifoliae p.p., whilst the other consists of all of the remaining sections of subg. Rosa (Banksianae p.p., Bracteatae, Caninae, Indicae, Laevigatae, Rosa, Synstylae and Pimpinellifoliae p.p.). A fairly complete sampling of field-collected North American taxa has been incorporated in this analysis. Analyses indicate that migration into North America occurred at least twice within this primarily Old World genus. Most North American taxa, except R. setigera and R. minutifolia, fall into a single clade that includes Asian and European taxa. Analyses also are consistent with the notion that cultivated commercial roses have a relatively narrow genetic background. Six of the seven primary taxa believed to be involved in the creation of domesticated roses are found within the same large clade that mostly includes Asian and European taxa.
Sequence Analysis. The very low level of sequence divergence in Rosa detected in this chloroplast analysis confirms the conclusion of Matsumoto et al. (1998), based on a more limited data set, that the genus has a very narrow plastid genetic background. Studies based on nuclear, single-copy genes (Joly et al. 2006; Joly and Bruneau 2006) and ribosomal spacers (Wissemann and Ritz 2005; Ritz et al. 2005) indicate that this low-level of molecular divergence amongst rose species is not peculiar to the chloroplast, suggesting that most extant rose species have a very recent origin (Wissemann and Ritz 2005). For example, within North American taxa of sect. Cinnamomeae, no rose species differed in its sequence by more than 1.4% (15 absolute differences, 26 taxa). Thus it seems that the difficulty in distinguishing North American roses by morphology alone (e.g., Erlanson 1934; Lewis 1957; Erlanson-Macfarlane 1966; Joly 2006), where features represent more of a continuum than discrete characteristics, is mirrored by the paucity of differentiation between rose species at the molecular level. This low level of sequence divergence explains the difficulty encountered when reconstructing the phylogeny of this taxonomically difficult genus, whether based on non-coding chloroplast sequences or on nuclear ribosomal spacers. In part for this reason, but also because of potential problems with hybridization or introgression in garden material, inferences of phylogenetic relationships must at this time be based on a careful comparison of clades that are similarly resolved in this and all previous phylogenetic analyses.
Within Clade I it is interesting to note the varied phylogenetic positions of polyploid taxa for which more than one individual was sequenced. In particular, the position of the diploid R. acicularis var. nipponensis as sister to R. rugosa in an entirely Asian clade, and separate from polyploid R. acicularis may support Lewis’s (1959) contention that this taxon should be treated at the species level. The relatively well-supported grouping of R. rugosa, R. marrettii and R. acicularis var. nipponensis also is resolved in the ITS analyses of Wu et al. (2001). The remaining samples of R. acicularis occur in different positions in Clade I. Rosa acicularis has both hexaploid and octoploid populations, and for this species, the incongruence also may suggest that R. acicularis has multiple independent origins from different maternal parents as has been seen in many other polyploid species (e.g., Doyle et al. 1990; Soltis et al. 1995). A similar explanation is possible for two polyploid western North American species, R. nutkana (where two varieties were sampled) and R. californica. For other taxa, the incongruence may be the consequence of the non purity (hybridization, introgression) of the garden samples included in the analysis. For example, this may explain the conflicting positions of the two R. wichurana specimens as either nested within Clade II or as sister to it, where we used two gardencollected specimens, but from different gardens. This also may explain contradictory positions for the same species among the different chloroplast DNA analyses that have been published to date (present study; Matsumoto et al. 1998, 2001; Wissemann and Ritz 2005). For example, we suspect that contaminated botanical garden samples explains the position of R. californica within “Clade II” in the matK analyses of Matsumoto et al. (1998, 2001) and of R. palustris (sect. Carolinae) and R. rugosa (sect. Cinnamomeae) at the base of “Clade II” in the atpB-rbcL analyses of Wissemann and Ritz (2005). Likewise the position of R. bracteata (sect. Bracteatae) in our analyses as nested within Clade II, rather than as sister to R. cymosa (sect. Banksianae), contradicts the analyses of Matsumoto et al. (1998, 2001) and of Wissemann and Ritz (2005), and suggests that our sample of R. bracteata may represent a botanical garden contaminant.
Although poor clade support makes it difficult to draw strong conclusions, some interesting biogeographic patterns are apparent in our analyses. The presence of rose species in North America appears to be the consequence of multiple introductions in a genus that mostly is concentrated in the Old World (Europe and Asia). If we consider R. minutifolia as sister to all other Rosa species, then there would be a split at the base of the phylogeny between North American and Old World taxa. Based on these data it is therefore not possible to determine whether the genus Rosa has an Old or New World origin. An obvious introduction of Rosa in North America occurred in Clade II with R. setigera, the only species of this clade that is native to North America. Within Clade I, with the exception of R. nutkana var. hispida (a probable polyploid) and the polyploid R. acicularis (parsimony analysis), all of the first branching taxa are Asian in origin, whereas the later branching taxa are a mix of both North American and Asian species. It also is noteworthy that all diploid North American roses in Clade I are contained within a single subclade, although they are intermixed with Asian taxa. This suggests a single New World introduction for all diploid North American species of subg. Rosa, but because of the low resolution we cannot firmly conclude whether the Old World species in the mostly North American clade represent reintroductions in the Old World or an ancestral presence.
[PLANT SYSTEMATICS AND EVOLUTION
Volume 297, Numbers 3-4, 157-170,](http://www.springerlink.com/content/q483j74321165286/)
Untangling the hybrid origin of the Chinese tea roses: evidence from DNA sequences of single-copy nuclear and chloroplast genes
Jing Meng, Marie Fougère-Danezan, Li-Bing Zhang, De-Zhu Li and Ting-Shuang Yi
Abstract
The tea-scented China roses largely correspond to the three recognized double-petaled Rosa odorata (Andrews) Sweet (Rosoideae, Rosaceae) varieties, which are the ancestors of modern hybrid tea roses and had a definite and permanent influence on the evolution of modern garden roses. Here the hypothesis of a hybrid origin of the tea-scented China roses between R. odorata var. gigantea and R. chinensis was tested. Two single-copy nuclear genes of the cytosolic glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and the chloroplast-expressed glutamine synthetase (ncpGS) together with two plastid loci (trnL-F and psbA-trnH) were sequenced for representative accessions of four R. odorata varieties, R. chinensis, and 28 other Rosa species. Phylogenetic relationships were estimated from two nuclear loci using maximum parsimony and Bayesian analyses, and a haplotype network was constructed on the combined plastid data using NETWORK. For GAPDH and ncpGS loci, the clonal sequences of the three double-petaled varieties were clustered into two clades, one clade with R. odorata var. gigantea, and the other with partial sequences of R. chinensis, which suggested that the tea-scented China roses were hybrids between R. odorata var. gigantea and R. chinensis. Two plastid loci suggested that R. odorata var. gigantea could be the maternal parent and R. chinensis the paternal parent.
Simon Joly
Canadian (NSERC) funded postdoc
Allan Wilson Centre for Molecular Ecology and Evolution
Massey Univeristy
Palmerston North,
New Zealand
Evolution of North American Roses (Rosa: Rosaceae)
Roses are well known for their extensive morphological variation, which resulted in a notoriously complex taxonomy. Both hybridisation and polyploidy were thought to have played an important role in creating biodiversity in this genus, although the relative importance of each factor has rarely been addressed.
During my Ph.D. at the University of Montreal, I studied the species problem in native roses of eastern North America that itemise problems found at the genus level in terms of hybridisation and polyploidy. I defined species boundaries in this complex (Joly and Bruneau, submitted) and reconstructed the evolution of both diploids (Joly and Bruneau, 2006) and of polyploids (Joly et al. 2006). I also confirmed the species status of the three polyploids species that evolved repeatedly from different diploid progenitors (Joly et al. 2006). Overall, these studies showed that hybridisation occurred at both diploid and polyploid level, but also during the formation of polyploid species.
Publications
1 Joly, S. and A. Bruneau. Submitted. Delimiting species boundaries in Rosa sect. Cinnamomeae (Rosaceae) in eastern North America. Systematic Botany.
2 Wang, Y., S. Joly, and D. Morse. Submitted. Phylogeny of dinoflagellate plastid genes recently transferred to the nucleus supports a common ancestry with stramenopile plastid genes. Journal of Molecular Evolution.
3 Bruneau, A., J. R. Starr, and S. Joly. In press. Phylogenetic relationships in the genus Rosa: new evidence from chloroplast DNA sequences and appraisal of current knowledge. Systematic Botany.
4 Joly, S. and A. Bruneau. 2006. Incorporating allelic variation for reconstructing the evolutionary history of organisms from multiple genes: an example from Rosa in North America. Systematic Biology 55:623-636.
5 Joly, S. 2006. Évolution des roses (Rosa:Rosaceae) indigènes de la section Cinnamomeae à l’est des Montagnes Rocheuses. Ph.D. thesis, Département de Sciences biologiques, Université de Montréal, Montréal, Canada.
6 Joly, S., J. R. Starr, W. H. Lewis, and A. Bruneau. 2006. Polyploid and hybrid evolution in roses east of the Rocky Mountains. American Journal of Botany 93:412-425.
7 Bruneau, A., S. Joly, J. R. Starr, and J.-N. Drouin. 2005. Molecular markers indicate that the narrow Québec endemics Rosa rousseauiorum and Rosa williamsii are synonymous with the widespread Rosa blanda.Canadian Journal of Botany 83:386-398.
8 Joly, S., and A. Bruneau. 2004. Evolution of triploidy in Apios americana (Leguminosae) revealed by the genealogical analysis of the histone H3-D gene. Evolution 58:284-295.
9 Joly, S., J. T. Rauscher, S. L. Sherman-Broyles, A. H. D. Brown, and J. J. Doyle. 2004. Evolutionary dynamics and preferential expression of homeologous 18S-5.8S-26S nuclear ribosomal genes in natural and artificial Glycine allopolyploids. Molecular Biology and Evolution 21:1409-1421.
10 Joly, S., L. Brouillet, and A. Bruneau. 2001. Phylogenetic implications of the multiple losses of the mitochondrial coxII.i3 intron in the angiosperms. International Journal of Plant Science 162:359-373.