What happens if you try to breed Triploid citrus

[SoCal2warm]

Some of you may be aware that Oroblanco and Melogold grapefruits are triploid, meaning they have 3 sets of chromosomes rather than the normal 2.
So a natural question that arises is what would happen if you tried to breed a triploid citrus with something else? What would you get.
Although Oroblanco is normally considered seedless, I found several fully formed large seeds inside one the fruits I just bought. I'm assuming all/most of these seeds are nucellar (clones of the parent) but it got me thinking.

I've been doing a lot of research, and wanted to present to you what I've found so far, in an attempt to answer the question. I'm no expert in this area.

Basically, when a triploid like Oroblanco pollinates another citrus variety (generally a monoembryonic variety) it induces the formation of seed. These seeds will be haploid, having only a single set of chromosomes from the female fruit parent. Most of the gametes produced from a triploid will have unevenly split chromosomes during meiosis, but a smaller portion will also be diploid and haploid. That means that pollen from a triploid parent could contribute either one or two sets of chromosomes. Assuming that the female parent is diploid, this will result in the seed either having 2 or 3 sets of chromosomes. So you could get a normal zygotic seedling, or you could get another triploid. The triploids will have a lower probability of having fully formed seeds, so the majority will not survive. (Embryo rescue techiques are usually employed to be able to grow these underdeveloped seeds, but in a small number of cases the seeds will be well formed enough to be able to germinate on their own)

It is also possible for a triploid parent to produce an unreduced gamete but this comparatively much more rare, so in that case the gamete would be able to contribute 3 sets of chromosomes.

Haploid citrus plants are normally feeble and slower growing, generally sterile, and right now I cannot find any references to a haploid citrus plant ever having produced fruit. But for breeding purposes, they can be reconstituted back to normal diploids by treating the growth on bud shoots with colchicine, and then later taking a graft from that offshoot.

This still leaves so many questions but it's a start to understanding.

Triploid plants can and do go through meiosis (cite). Of course, their fertility is lowered with most gametes being aneuploids (uneven number of chromosomes), however, triploid plants also produce small numbers of euploids (1x, 2x, 3x)(Fig 1 of the above cited source).

This should make you wonder why we don't often observe seeds in triploid fruit since they can make viable gametes (1x, 2x, 3x). This seems to be purely driven by the number of female gametes (ovules) produced vs. the number of male gametes (pollen grains) produced. Plants produce many more pollen grains than ovules making it much more likely that a euploid pollen grain will be produced. Contrasted by the small number of ovules produce making it very unlikely that a euploid ovule will be produced.

The fertility of triploid plants is the foundation for one of the most used mechanisms for polyploid formation called the triploid bridge. The triploid plant produces 1x, 2x, and 3x gametes that combine with the 1x gametes produced by a diploid plant to form 2x, 3x, and 4x offspring
 

I could well be wrong about most of the seeds in Oroblanco being nucellar. That would be interesting.

[Marcin]

I used to plant Sweetie seeds (Sweetie is a triploid cross similar to Oroblanco). All seeds were monoembryonic, zygotic. Some grew into nice, vigorous plants, but some seedlings were weak and kind of defective. So my idea is you can use a triploid as a mother plant, you will just need to screen the weak seedlings out, and keep the healthy ones.

An interesting paper:
https://ejournal.sinica.edu.tw/bbas/content/2009/3/Bot503-06.pdf

[Ynk88]

I also found some seeds in a triploid hybrid. But it was a Tacle mandarin and not an Oroblanco or Melogold grapefruit.
I found 2 seeds after eating about 25 Tacle   

So I have sown them at beginning of february this year, because I love this hybrid, for me it's the best tasting mandarin I have eaten so far.
The first seed was monoembryonic and the second was polyembryonic.
The monoembryonic seed was a nice big one and the poly was not fully formed.

The plant from the monoembryonic seed is groing much faster than the two plants from the other seed ,and I wonder if it will give something good to taste. I'll have to wait a long time before tasting some fruits, if they survive till there 

[Marcin]

Ynk88, congrats on finding these seeds

This year I've also found my first seed in a Persian lime (a triploid too). The seed was small but luckily sprouted and gave this one seedling. It's about 1 month old now. I hope it will get more vigorous soon.

[SoCal2warm]

After several hours of research I came across something.
P. Ollitroult found half diploid and half triploid among the seeds of 'Oroblanco'.
Citrus genetics, breeding and biotechnology, Iqrar A. Khan, p205

Only diploids were found in the progeny of clementine fertilized by Oroblanco pollen, the same source says. But I wonder if they checked to see if the next generation were even hybrids, since very likely the triploid Oroblanco pollen might just have been inducing parthenocarpy. Edit: oops, if that was the case, one would expect haploid offspring (see gynogenesis). Another potential explanation, I did read in some citrus cultivars (particularly many oranges) the surrounding nucellar tissue in the embryo is highly sensitive to triploidy and will abort.

Also, while it is true triploid seeds are most often found in shrivelled up smaller seed (which also causes low viability), this is not necessarily true when the fruit parent happens to also be a triploid. The poor seed development is caused by differences in ploidy between the developing megagametophyte and surrounding nucellar tissue.

[SoCal2warm]

So a quick summary, here is what I gathered from the information
(disclaimer: this could be completely wrong)

When Oroblanco pollinates a normal diploid, you get a haploid (at least the vast majority of the time), or in other circumstances apparently, maybe depending on the cultivar, you will get all diploid
When Oroblanco self-pollinates, you get half diploid and half triploids (and doesn't seem to cause gynogenesis probably because the ploidy of the fruit parent is the same as the surrounding pollen tissue)
There is probably a tiny ratio of tetraploid but those seeds are very underdeveloped and will not germinate on their own.
When a normal diploid pollinates Oroblanco, you probably get diploid and triploid

[Ynk88]

Hey ! Nice seedling Marcin 

Now I understand where this topic goes.
Sorry if I was off-topic SoCal2warm 
It's very interesting, I wonder what will be the conclusion.

[Ilya11]

So a quick summary, here is what I gathered from the information
(disclaimer: this could be completely wrong)

When Oroblanco pollinates a normal diploid, you get a haploid (at least the vast majority of the time)
When a normal diploid pollinates Oroblanco, you probably get diploid and triploid

Disclaimer is right, it is wrong.
Triploids never give haploid plants ; theoretically  their gametes  are only  haploid, diploid and triploid.
Pollination of Oroblanco with diploid clementine pollen is giving only diploid plants due to unfavorable completion of diploid gametes   with haploid ones.
(Citrus Genetics, Breeding and Biotechnology,edited by Iqrar Ahmad Khan, p.205)

[Marcin]

You may also get aneuploids. This paper mentions them in the case when triploid is a mother.

https://ejournal.sinica.edu.tw/bbas/content/2009/3/Bot503-06.pdf

Ynk88, you weren't off-topic. I'd like to see the photos of these seedlings to compare them with my experiences.

[SoCal2warm]

Disclaimer is right, it is wrong.
Triploids never give haploid plants ; theoretically  their gametes  are only  haploid, diploid and triploid.

I believe you were misunderstanding. Triploid pollen can induce the formation of haploid embryo from the female fruit parent; the single chromosome set would always come from the female parent in such a case.

This study reports haploid plantlet regeneration through gynogenesis in Citrus clementina Hort. ex Tan., cv. Nules, induced by in vitro pollination with pollen grains of Oroblanco, a triploid cultivar of grapefruit. It indicates that parthenogenesis induced in vitro by triploid pollen can be an alternative method to obtain haploids in monoembryonic cultivars of Citrus.

https://link.springer.com/article/10.1023/A:1010627310808


There are a few possible interpretations to these findings (not discussing the link above now). Unfavorable competition of one pollen type was only listed as a possibility, as well as a possibility all the triploid seeds (when they form in Oroblanco fruits) didn't result from pollination.

[Ynk88]

Ok Marcin so here they are.
The first left is from the monoembryonic seed and the two right are from the polyembryonic one.
The one from the left sprouted about two month after sowing, and the two others about the end of June.

[Ilya11]

Disclaimer is right, it is wrong.
Triploids never give haploid plants ; theoretically  their gametes  are only  haploid, diploid and triploid.

I believe you were misunderstanding. Triploid pollen can induce the formation of haploid embryo from the female fruit parent; the single chromosome set would always come from the female parent in such a case.

It is not triploid gamete that is giving haploid mother alike plants, it can be easily replaced by irradiated diploid pollen.

[Marcin]

Thanks Ynk88. I hope they all will grow well.

[SoCal2warm]

An interesting paper:
https://ejournal.sinica.edu.tw/bbas/content/2009/3/Bot503-06.pdf

Thank you so much.

If I can summarize the results here: They looked at the seeds of triploid citrus (this particular cultivar having resulted from a tetraploid cross between Hamlin orange x rough lemon, which then was crossed with diploid tangerine) which was open-pollinated, and out of 25 well-developed seeds, 13 were triploid, 10 were diploid, 1 was tetraploid, and 1 was aneuploid. What was particularly interesting was that all of the triploids appeared to be nucellar. This suggests that the likelihood of triploid meiosis being able to produce a complete haploid gamete is far higher than a diploid gamete.
Assuming this is the case, then presumably the single tetraploid seed must have resulted from an unreduced triploid gamete being sexually combined with a normal haploid gamete.
It seems then that the probability of a complete whole diploid chromosome set being able to escape from triploid meiosis is very unfavorable, all the triploids resulting from a triploid fruit will be clones of the parent. (The triploids will be the offspring with very few seeds)

[Heinrich]

What was particularly interesting was that all of the triploids appeared to be nucellar. This suggests that the likelihood of triploid meiosis being able to produce a complete haploid gamete is far higher than a diploid gamete.

Nucellar seedlings are never derived from meiosis. Therefore, you can´t conclude from nucellar seedlings anything about meiosis.

There is a lot of misunderstanding and there are many wrong statements about citrus genetics.
 
Ilya, thank you for some corrections and your comments.

[SoCal2warm]

Nucellar seedlings are never derived from meiosis. Therefore, you can´t conclude from nucellar seedlings anything about meiosis.

I meant that, because all the triploids identified were nucellar, that seems to indicate no triploids resulted from meiosis, and by implication it seems meiosis of a triploid set is unable to produce a diploid gamete (or at least it's very rare). So if there were no triploids resulting from meiosis but they identified one tetraploid, it's logical to conclude that that tetraploid was very unlikely to have been a result of triploid meiosis.

For that to happen, there would have had to have been two different triploid sets that each split into an intact diploid, and then those diploid gametes would have had to have fused together to produce the final tetraploid. Not a very high probability of happening, especially since presumably no triploids formed from diploid gametes.

[SoCal2warm]

Heinrich and Ilya11, you are both being illogical and drawing faulty conclusions.

Heinrich, you said "Nucellar seedlings are never derived from meiosis". I did not imply that they were.
In fact by the very fact that they were (presumably) not derived from meiosis, you can conclude something about meiosis.
If there were no triploids resulting from meiosis, what does that tell us? And it would certainly seem to indicate something about the origins of that tetraploid they found.

Ilya11, go back and read over everything I posted. I never said that triploid plants gave rise to haploid offspring that were alike. You seem to be assuming I said something that I did not.

Ilya11 has still failed to give any reasoning why what I wrote is wrong.

Pollination of Oroblanco with diploid clementine pollen is giving only diploid plants due to unfavorable completion of diploid gametes   with haploid ones.
(Citrus Genetics, Breeding and Biotechnology,edited by Iqrar Ahmad Khan, p.205)

That's not necessarily true. That's just an assumption they made. 

My alternative theory offers an explanation. If Oroblanco is not producing any diploid gametes then of course there would not be triploid offspring. No haploids are resulting because the Clementine pollen is not triploid and is therefore unable to induce gynogenesis in the Oroblanco, and besides, the female parent being triploid might have something to do with it (apparently it may take triploid pollen pollinating a diploid plant for this effect to happen). Sometimes, in certain species, nucellar seed formation can be induced by pollination, when the resulting embryo is not very strong as a result of chromosomal differences/abnormalities, so if the majority of the pollen is not triploid, that could be an explanation for the lack of nucellar formation.

[Ilya11]

You should read more attentively the articles you cited. The story about gynogenetic haploids induced by triploid pollen has nothing to do with real world pollination and fruiting. All was done in vitro.

You are stretching your conclusions and are very reiterative and unclear.

[Heinrich]

SoCal2warm, an argument like “There were no triploids resulting from meiosis. This suggests…” certainly will be accepted. It is also your style of argumentation which leads to dissent and disagreement.  I don´t want to discourage you. I am sure, the more you learn about genetics the better will be your conclusions. Your arguments will improve and will be than more widely accepted.

[SoCal2warm]

Well yes I may be jumping to conclusions.

[SoCal2warm]

Here's four seedlings of Oroblanco:



They have a bit of a winged petiole, unlike the grapefruit seedlings I'm growing.