19 Mar 2001
Cytoplasm, inheritance and mutations
Karl King

From what I can gather from various sources, there is no uniform pattern to what cytoplasmic inheritance may influence, or whether it will be observed at all. There have been some speculations about the male parent influencing this, and the female parent influencing that, but I don't see it as a widespread pattern.

We must get one thing out of the way before we can understand how cytoplasmic inheritance works. Ova are big, and pollen nuclei are small. Nevertheless, in some cases the paternal (pollen) influence is greater than the maternal (seed).

One trick to getting a bigger bang to the pollen grain is to supercharge the organelles. (Nagata, etal. Planta, 1999) In those species that do transmit plastids or mitochondria or both in pollen, the quantity of organellar DNA increases just before the pollen is formed. The *number* of organelles may be small, but they are packed with extra DNA. Thus, when they find their way to ova or endosperm, the organelles can reproduce quickly — without waiting for DNA replication. The examples cited differ in which components are transmitted, mitochondria = m, plastids = p; + indicates presence, - for absence in pollen:

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-

In other instances cytoplasmic inheritance is biparental, or may vary from maternal-only through biparental to primarily-paternal in the same species under genetic control. If cytoplasmic factors can influence the expression of nuclear genes, it is only appropriate that nuclear genes should also regulate some aspects of cytoplasmic inheritance. (I have to wonder how a species that does not ordinarily receive mitochondria or plastids through pollen would react to these when introduced by foreign pollen.)

When a normal Coastal Redwood was pollinated by an albino sport, more than 2-thirds of the seedlings were at least partially albino. Some were white or pale green, and soon died. Others were variegated, or produced white branches. There were relatively few albino plastids (chloroplasts), but in some seedlings these albino plastids actually dominated.

In geraniums (Pelargonium), this segregation can be more orderly, producing neatly layered chimeras. The outer layer of tissue contains exclusively albino plastids, while the green plastids are restricted to the inner layers. In crosses of green and variegated geraniums, the following percentages were found:

% green
% variegated
% white
total
G x W
77
13
10
280
W x G
72
28
-
93

Darlington (Evolution of Genetic Systems) wrote, "Thus the male germ cell makes a greater contribution of these cytoplasmic particles. And the female cell sometimes fails to contribute anything at all."

Cytoplasm and nuclear mutations

One of the more surprising influences of cytoplasm is its ability to alter the rate of mutation in chromosomes. This was demonstrated decades ago in Epilobium, and more recently in Brassica (Song, etal, Rapid genome change in synthetic polyploids of Brassica and its implications for polyploid evolution.Proc. Natl. Acad. Sci. August 1995). In the recent experiment, Brassica nigra and B. rapa were crossed in both directions, and the seedlings were treated with colchicine to induce chromosome doubling. The two groups of tetraploids carried essentially the same chromosome sets, and both quickly gave rise to structural modifications (mutations), but they differed greatly in *which* chromosomes were being modified.

In the cross Rapa x Nigra, the bulk of the cytoplasm was of the Rapa type. In these tetraploids the Nigra chromosomes mutated more rapidly than the Rapa chromosomes. In the reciprocal cross, the Rapa chromosomes mutated more rapidly in the Nigra cytoplasm.

Hybrids are interesting, but not always desirable because they are uncertain in their progenies. Large-scale breeding and careful selection are necessary to produce stable, true-breeding strains. But if we want to increase the variability of a species, we might use that species as pollen parent in crosses with some moderately remote relatives. After a few generations of repeated pollination, most of the qualities of the desired species will be transferred to the new cytoplasm, and may "vary" (mutate) more rapidly than they did before. These traits might then be stabilized by backcrossing (as pollen parent) to the original species, to stop further structural changes. The new traits would add to the range of variation in the original species.

Graft Hybrids

This is another interesting puzzle. There was a debate about graft hybrids a century ago that has never really finished. Mendelists, of course, insisted that heredity is in the nucleus, and arranged their experiments to "prove" that graft hybrids do not occur, or only as chimeras. Less dogmatic researchers (or dogmatic in other ways) repeatedly demonstrated that something was being transmitted across graft unions.

Burbank did much grafting, but usually with seedlings of mixed ancestry. In one instance, however, there is an indication of something odd. He received seeds of a purple-leafed plum from China, and grafted the seedlings to a tree already hosting hundreds of other seedlings. None of the purple-leafed scions bloomed, and no such specimens were grown in the area. Yet a seed from another scion on that tree grew into a purple-leafed tree with fruit shaped like the very distinctive 'Wickson' plum (inverted pear-shape). Burbank had never before seen a purple-leaf "mutant" among his many plum seedlings, so it may be stretching coincidence to suggest that "random chance" was playing with him.

Genetics, 1962, has an article by Frankel dealing with the transmission of cytoplasmic sterility between stock and scion in Petunias.

Petunia 'Rosy Morn' was a male-fertile strain that gave only the occasional male-sterile seedling. It was grafted to a male-sterile clone as stock or scion, and produced a proportion of male-sterile progeny. Interestingly, more male-sterile seedlings were produced from seeds collected in winter (31.7%) than summer (26.4%).

Something certainly happened, and the obvious conclusion is that a cytoplasmic factor (whatever it might be) passed between stock and scion. But that 25% looks suspiciously like segregation of a nuclear gene. So, perhaps the male-sterility involved both a cytoplasmic factor and a nuclear-gene. In most cases, the tendency to produce male-sterile progeny faded soon after the graft was removed from male-sterile clone, but in one case the effect continued to express for a year after separation.

This case reminded me of Ivan Michurin's warning to plant breeders. He advised against grafting cultivated Apple varieties to stocks of wild types when the cultivated varieties were to be used for breeding. He claimed to have observed undesirable qualities of the wild types showing up — not in the scions themselves, but in their seedlings. This is precisely what happened in the Petunia example. The grafted 'Rosy Morn' did not become male-sterile, but a substantial number of its self-seedlings did.

Taller etal (Theoretical Applied Genetics, 1998) discussed graft-induced variants in peppers (Capsicum). In some instances, a new "gene" was apparently introduced into the grafts, which then segregated in the ordinary way. In others, new traits — not found in either graft component — appeared. Finally, some grafts and their progeny revealed previously hidden genetic components of known traits.

"The effect of the stock is also detectable in segregating progeny derived from crosses of G [graft induced variant strain] with Sp [Spanish Paprika] and Y [Yatsubusa], showing deviation toward the stock characteristics." Again, this supports Michurin's warning against wild apples as stocks.

Grafts are occasionally found "in the wild", but we may doubt that these play a significant role in evolution. However, another sort of graft is more common.

Endosperm and embryo are attached in seeds, but they are not derived from the same cells thanks to double pollination. The details vary, but the general pattern is that two cells are contributed by a single grain of pollen. One finds its way to the ovum, while the other merges with an accessory cell (sister-cell to the ovum). The fertilized ovum grows into the embryo, the other fertilized cell develops into the endosperm.

Sometimes two accessory cells fuse into one, which is then "fertilized". In this case, the endosperm would be triploid while the embryo is merely diploid (assuming diploid parents). This might account for some cases where a cross may give many viable seeds, but only one or a few survive. A triploid endosperm might compensate for the weakness in the ordinary diploid endosperms.

The usual pattern of double pollination can be disrupted when the pollen is of a different species (or genus) than the seed parent. Michurin reported the case of a Peach seed, pollinated by an Almond, that contained four kernels — each of which sprouted. Bradley reported similarly: "Calostemma luteum x Pancratium maritimum gave a quantity of fertile seed, and the seedlings are growing strongly, but none have flowered yet, so it is uncertain if the cross has taken. Curiously, several of the seeds developed two plants, and one seed gave three plants from the one seed. Apparently the foreign pollen stimulated more than one maternal cell to develop, possibly without combining with the ovum. Whether the pollen contributed to the endosperm is open to further experiment.

Herbert also mention an attempted cross (of Pancratium ?) that resulted in large, lumpy seeds without embryos. Again, it is unknown whether the pollen participated in endosperm development, or merely stimulated more than one maternal cell to grow as endosperm.

I mention these cases only as examples of the weirdness that can occur following wide-crossing. A more significant case was recorded by Burbank of a cross between the Wren's Egg x Lima bean. He got only one pod for his efforts, and just four seeds. All odd:

"... the upper part of their cotyledons (varying from one-quarter to three-quarters of their length in different specimens) were indubitably those of the lima bean; while the lower part of each cotyledon was precisely that of the horticultural pole-bean.

"These parts were connected with serrated edges, which at last separated, allowing the lima bean part to drop away. Such separation, however, did not occur until the vines had made a foot or more of growth.

"The cotyledons on each side were divided uniformly in every case."

The plants appeared to be purely maternal, except that they were unusually vigorous. Seedlings from these plants reverted to the ordinary habit and vigor of the Wren's Egg bean.

There are other similar cases, such as Naudin's cross of Datura stramonium x D. ceratocaula. The two surviving seedlings (of 60 seeds) were entirely maternal, but nearly twice as tall. They failed to set fruit on the lower forks, but recovered their fertility later in the season. The seedlings from these reverted to ordinarly D. stramonium.

Burbank's example is the most important because we can see that the pollen definitely participated in the formation of the endosperm, if not of the embryo. The reaction of cytoplasm to foreign chromosomes (or to excess chromosomes) is one aspect of hybrid vigor and new-polyploid gigantism. In this case there must have been cytoplasmic factors from the pollen parent that entered the embryo from the endosperm (before the Lima-like parts of the cotyledons fell off). Otherwise there is no obvious explanation for the increased vigor of the pseudo-hybrids, or its subsequent loss.

Cases like these have me wondering what sort of influences might follow seed grafts — grafting a developing embryo onto the endosperm of a distant relative.

Micrografting is a revelation. Years ago, the first time I raised roses from seeds, all my precious seedlings damped off. With nothing to lose, I found a rusty old single edged razor blade and cut the seedlings. Some had good cotyledons and shoots but dead roots. Others seemed to have died upwards, leaving the roots. I cut them on the diagonal, and brought together a root of one seedling with the cotyledons and tiny leaves of another. I wrapped the stem with thread and returned the plants to the terrarium (not the best place to sprout rose seeds, but appropriate for grafted seedlings).

Wonder of wonders, some of the grafts took. I got two plants composed of four original seedlings.

Embryonic and juvenile tissues are much more accomodating than mature tissues. Micrografting has been used to propagate choice specimens of Blue Spruce — small shoots are grafted to seedlings, and the seedling tissue finds its way into/onto the little scion. I'd just read about this when my rose seedlings presented me with the opportunity discussed above.

So, what I have in mind is grafting Cape Belladonna, Nerine, Brunsvigia and other embryos onto Crinum seeds.

Michurin and his followers have used grafting — sometimes repeated grafting — to bring incompatible into "vegetative approximation", whatever that means. The technique has worked often enough to remain in use, so it might be used to bring plants of different genera a little closer together to improve the chances of successful crossing.

I can't guarantee that a Cape Belladonna embryo grafted to a Crinum seed would mature into a plant that would readily accept Crinum pollen, but it's worth the experiment.

Cytoplasm and Disease

Resistance or susceptibility to disease may be genetic (nuclear genes) or cytoplasmic, or a combination of both. Cytoplasmic resistance/susceptibility to mildew has been observed in Fragaria (Strawberries) and Epilobium, to mention just two. Another instance of extreme susceptibility to infection had Corn growers sweating in the 1970s.

Early in the 20th century a female corn plant (male sterile) was studied. All the seedlings inherited the cytoplasmic male sterility — until a strain was found that carried a gene that restored male fertility. As "F1 Hybrid" corn became more popular, various types of Corn were bred into the male-sterile cytoplasm. This reduced the cost of "hybrid" corn since it wasn't necessary to de-tassle the plants used as seed parents. Field, flint, dent and other breeds of Corn were bred to exploit male sterility, and the male fertility restoring gene.

If any future gene-archaeologists happen to study Corn from the 1960s and early 1970s, they might conclude that all the standard varieties "evolved from" a single female ancestor. The proof would be in the shared mitochondrial DNA.

But in the early 1970s a side-effect of the male-sterile cytoplasm was discovered — plants carrying it were unusually susceptible to Southern Corn Blight. Unusually hot and humid weather allowed the Blight to spread farther north, and corn growers were worried. There was talk of a "mutant" strain of the blight, but soon the truth was discovered. Despite appearances, the numerous strains of corn were a virtual monoculture. Whatever other characteristics the breeds might have, most shared the same cytoplasm.

I think about Corn and Blight whenever I read about DNA experts contradicting the opinions of paleontologists. The DNA evidence indicates that modern humans spread around the world displacing whatever other humanoid species might have been there before them. But the paleontological evidence reveals nothing of the sort. No new species appeared at the time the DNA guys say it did.

So far I haven't read any discussion of the possibility that some disease was lethal or merely annoying according to the type of mitochondria the people carried. Other hereditary factors would have been of little or no consequence, but such a hypothetical disease would act as a filter — or "selective pressure" — weakening or killing all the folks who inherited the "bad" mitochondria, or whatever other cytoplasmic factor may have been involved.

Selective elimination of cytoplasm-types would effectively "reset the evolutionary clock". Just as the evidence of mitochondrial DNA in corn (late 1960s) could give a mistaken idea of the evolutionary age of the varieties grown at that time.

Cytoplasm and stress

Some cytoplasmic influence isn't expressed unless the organism is stressed. In crosses of Epilobium luteum x E. hirsutum, a variety of cytoplasmic effects were observed (resistance to mildew, male sterility, albinism, dwarfism). The hybrids were backcrossed to the pollen parent 13 times, to give a Hirsutum nucleus in an approximately Luteum cytoplasm.

When grown in reduced cultural conditions (starved), one of the species becomes dwarf and bushy; the other tall and spindly. I forget which is which. At any rate, the plants with a Hirsutum nucleus in Luteum cytoplasm behaved like pure Luteum rather than like pure Hirsutum. Under better conditions, the two forms of Hirsutum were more similar.

Some species, like those of Epilobium, can be polymorphic for cytoplasm type. In other genera, several allied species may share essentially the same cytoplasm. The reasons for this polymorphism, or lack of it, are probably as varied as for any polymorphism of nuclear genes. Some species exploit the polymorphism to some advantage, others manage to get along without it.

Cytoplasmic male sterility

Male sterility is a fairly frequent result of crossing plants with different types of cytoplasm. In these cases, the male sterility may occur only in one of the cytoplasm types. Epilobium luteum-hirsutum remained male-sterile after 13 generations of Hirsutum pollination. Vicia faba minor x major hybrids segregate about 25% male sterile plants in the F2 generation. The reciprocal cross does not give male sterile offspring. In this case 6 genes were identified, all linked together, that produced male serility only when homozygous and only in the cytoplasm of Vicia faba minor. Onions and Flax offer other examples of cytoplasmic male sterility following hybridization.

Male sterile cabbage has been produced by pollinating the radish 'Early Scarlet Globe' with Cabbage, then backcrossing to the Cabbage until a "pure bred" Cabbage was produced with Radish cytoplasm. Since Cabbage is usually not grown for seed (except by breeders) it isn't necessary to use a fertility restoring gene to produce useful "F1 hybrid" Cabbage.

There's quite a bit more to the cytoplasm story, but this should do for a start. There is also the matter of plasmagenes, possibly strands of RNA capable of replicating in the cells. In other cases, these RNA units may be involved in gene silencing and other aspects of epigenetic regulation. If an existing gene is turned off, the effect may be the same as a "new gene" being introduced, or an old one mutating.

There are a number of examples of tissue selection that may fit a model involving segregation of cytoplasmic units. Some instances of white/green variegation of leaves do not conform to what would be expected if albino and green plastids were segregating. Something finer grained would be required — plasmagenes. Selection of rose canes with fewer than normal thorns, repeated several times, has led to the production of thornless varieties. These cannot be due to simple gene-mutations, since the change was gradual rather than abrupt. Likewise for selection of dependably reblooming forms from varieties that were originally once-blooming or uncertain. (The rose General Jacqueminot was an unreliable rebloomer that was improved substantially by bud selection. Commandante Beaurepaire was changed from once-bloomer to the fairly reliable rebloomer.)

Herbert insisted that when breeding for double camellias, pollen from petaloid anthers was to be preferred. McNab claimed that dwarf rhododendrons could be produced by using pollen from short stamens. Bradley reported that different anthers of Hippeastrums could give different sorts of progeny.

If we cling to the simplistic Mendelian model, we have no hope for understanding what Herbert, McNab, and Bradley were writing about. But if differential development of stamens has a physiological basis, and if the pollen is capable of capturing the factors involved, then it is not unreasonable to suppose that these authors and others were making valid observations.

Some refs:

Sirks: Plasmatic Inheritance
Maternal cytoplasmic effects in patrogenesis
Non-nuclear heredity in fusion hybrids
Traumatic Pollination

Karl King