Tuesday, October 7, 2014

Genetics in Sunflowers.

Plant #1 (H. tuberosus).
Plant #1 leaf.
Last year I started a genetics experiment with sunflowers. I grew a perennial sunflower (Helianthus tuberosus) and an annual sunflower (H. annuus cv. "Russian Mammoth") and tried to cross them. I shook pollen from the very large H. annuus flower all over and around as many of the very small H. tuberosus flowers as I could. This cross has been done before ([1], [2], & [3]), but not necessarily with a giant variety like "Russian Mammoth" as the father variety. I was working under the assumption that I'd be able to recognize any hybrids.

At the end of the season, I dissected 12 seed heads and recovered approximately 70 seeds. I ended up not planting any seeds this year, as I knew I was moving to a new house and wouldn't be able to maintain the experimental plants.

As I was doing a final cleanup of my flat, I noticed a sunflower blooming out in the backyard. On closer examination, there were three sunflower plants poking up through the weed patch. The only sunflowers growing in the area last year were my experimental parents. The H. tuberosus mother was growing in a planter and the H. annuus father wasn't allowed to mature seeds in the garden, so I was pretty sure these sunflowers had grown from seeds dropped by the mother plant.

The first plant to bloom appears to be completely H. tuberosus, though it shows much more red in its stems than the mother.  The other two plants came into bloom later and appear to be hybrids.

There are several traits which distinguish H. tuberosus from H. annuus cv."Russian Mammoth":

Plant #3.
  • H. tuberosus leaf margins extend down the entire petiole, while in H. annuus the leaf margin ends at the top of the petiole. All three plants showed the H. tuberosus version of this trait.
  • H. tuberosus has an elongated leaf, while H. annuus has a wide leaf. The blooming plant showed the H. tuberosus version of this trait. The other two plants had leaves of an intermediate form.
  • H. tuberosus flowers are very small, while H. annuus cv."Russian Mammoth" flowers are very large. The first plant to bloom has flowers the size of those on H. tuberosus. The other two plants now have flowers which are about twice as wide as those on H. tuberosus.
  • H. tuberosus plants are highly branched with many flowers, while H. annuus cv."Russian Mammoth" has a single stem with a single terminal flower. The first plant to bloom has lots of side-branches and lots of flowers. The second plant to bloom has a few small side-branches and a few flowers. The third plant has no side-branches and a few additional flowers growing from leaf-axils.
  • H. tuberosus plants have thin stems, while H. annuus cv."Russian Mammoth" plants have a very thick stems. The first two plants to bloom have thin stems and need support to stay upright. The third plant to bloom has a thick stem, capable of keeping it upright without support.
  • H. tuberosus plants grow to near 6 feet tall, while H. annuus cv."Russian Mammoth" plants grow to near 11 feet tall. The first plants to bloom topped out below 5 feet tall. The second reached about 7 ft and the third reached over 8 feet tall before blooming. I'll get a better measurement of their height when they have died for the season.
  • H. tuberosus produces underground tubers, while H. annuus does not. I'm waiting until frost has killed the plants before digging them up, so they have the best chance to produce tubers.


Plant #3, showing winged petiole.
What predictions can we make about a cross between H. tuberosus and H. annuusH. tuberosus is hexaploid and H. annuus is diploid, so the resulting hybrid will be tetraploid.

H. tuberosus is usually propagated by tubers, so it can contain lots of hidden genomic diversity across it's six sets of chromosomes. The progeny plant which appears to be completely H. tuberosus has much higher levels of red pigment in its stems, indicating heterozygosity is present in the H. tuberosus parent. H. annuus cv. "Russian Mammoth" is an annual variety that has been stable since 1880, which means it is highly inbred and therefore highly homozygous.

With this information, the cross would look something like:

A1A2A3A4A5A6 x BB

The resulting F1 progeny would be:

AaAbAcB

There are 20 potential combinations of three alleles of of the possible six homologs at a each locus of the 17-chromosome sets found in the H. tuberosus parent, so it isn't surprising that the two observed hybrids aren't entirely alike.



Selfing the F1s...
AaAbAcB x AaAbAcB

…has all kinds of possible outcomes. If we simplify the calculation by assuming the H. tuberosus parent is homozygous, we get a cross...

AAAB x AAAB

…where half of the gametes are AA and the other half are AB. The Punnett square for this cross...


AA
AB
AA
AAAA AAAB
AB AAAB AABB

…shows there's no way to get an F2 which is homozygous for the alleles from the H. annuus parent. A goal of this project is to breed up a sunflower which has the tuber-generating trait of H. tuberosus and the super-sized growth of H. annuus cv. "Russian Mammoth". If I can isolate some F2s that appear AABB for some of the interesting traits, perhaps by isolating those that have the best tubers and the most overall growth, then I might later be able to select from F3 families which cover the whole range of allelic combinations at the loci important for these two traits. The Punnett square for selfing the AABB F2s...

AA
AB
AB
BB
AA AAAA AAAB AAAB AABB
AB AAAB AABB AABB ABBB
AB AAAB AABB AABB ABBB
BB AABB ABBB ABBB BBBB

Plant #3 flower buds.
…shows the very diverse combinations expected at each locus. From such an F3 family, I would then have the best odds of selecting out a plant with several alleles driving tuber formation and driving extreme plant growth. The combination of which, I hope, would then result in extreme tuber production.

In a typical diploid cross, the F2 generation is where the most combinations of alleles appear and where selection is most important. In this example, it wasn't obvious before calculating through the probabilities that the F3 generation would be where the most combinations of alleles would appear.

Hopefully, the deer at my new place will leave my sunflowers alone for the years it will take to complete this project.



References:
  1. Encheva, J., M. Christov, and P. Ivanov (2003). Characterization of Interspecific Hybrids Between Cultivated Sunflower H. annuus L. (cv. Albena) and Wild Species Helianthus tuberosus. Helia 26: 43-50.   (http://www.doiserbia.nb.rs/img/doi/1018-1806/2003/1018-18060339043E.pdf)
  2. http://bulbnrose.x10.mx/Heredity/sunflowerXchoke/sunflowerXchoke.html
  3. Kantar, M. B., K. Betts, J. Michno, J. J. Luby, P. L. Morrell, B. S. Hulke, R. M. Stupar, and D. L. Wyse (2014). Evaluating an interspecific Helianthus annuss x Helianthus tuberosus population  for use in a perennial sunflower breeding program. Field Crops Research 155: 254-264 (http://experts.umn.edu/pubDetail.asp?id=84888291084&o_id=199&t=pm)
Information about traits from references and observations.
  • Reference #2 indicates tuber formation is dominant in the F1s.
  • Winged petioles are dominant in F1.
  • Secondary branches are mostly dominant in F1. There seems to be allelic variation for this trait in the H. tuberosus parent.
  • Large flower size is mostly recessive in F1.
  • Tall growth is partly dominant in F1. There seems to be allelic variation for this trait in the H. tuberosus parent.