[gwern]

That is disappointing, but I'm also a little curious - why did this go all the way to field trials if the gene insertion was completely screwed up, to the point of being on the wrong chromosome?

    In November 2023, through molecular analyses performed by partners at the University of New England and University of Maine, TACF learned that the OxO gene of all Darling 58 trees was on a different chromosome than expected (chromosome 4 instead of chromosome 7). Upon further and additional independent investigation, scientists confirmed that the trees they had been researching were in fact descendants of a different event in the Darling line in which the OxO gene had been inserted into a coding region, causing a deletion in a known gene. That research has also indicated that the homozygous state (when an individual plant inherits the OxO gene from both parents, which occurs in 25% of offspring) is lethal, and that a majority of homozygous offspring die in the embryonic stage.

    TACF researchers suspect that the performance issues of Darling trees stem primarily from the placement of the OxO gene as well as the constitutive expression of the OxO gene which is always “switched on” via the 35S promoter. Somewhat like having a constant fever, that constitutive promotion appears to result in high metabolic costs for the trees. All events in the Darling line use this promoter. Therefore, TACF is no longer pursuing research efforts with any event in the Darling line.

If you are breeding to insert a single, specific, gene, into a single, specific place, it would seem like you'd of course do some sequencing to verify that it went in at the expected place and didn't do anything else off-target (like cause deletions in an unrelated gene), before you invested a year in field trials.

Is sequencing of tree genomes still so expensive that it makes sense to fail in the field first and only then go looking for problems?

[tamcap]

It's possible they checked before (and the article suggests they had an inclination where the gene should be) but the resultant backcrosses involved a spontaneous recombination of the relevant fragment. It's possible someone checked the wrong vial in lab, or mislabeled results. It's possible something funny "just happened".

Wet lab is hard, and everyone is chasing the eureka moment.

[hedora]

My reading of the article suggests they modified the genomes of multiple trees, and the “winners” that got into the field happened to have ancestors that had moved the target gene to a different chromosome.

They’re simultaneously trying to make precision modifications to the genome and also trying to preserve hundreds of wild strains.

If they succeed, the required technological advances will probably be broadly applicable.

I wish them luck.

[jszymborski]

That was my understanding too:

"Upon further and additional independent investigation, scientists confirmed that the trees they had been researching were in fact *descendants of a different event in the Darling line* in which the OxO gene had been inserted into a coding region, causing a deletion in a known gene. "

emphasis my own

[CuriousCosmic]

Well for comparison the sunflower has around 30% more base pairs in it's genome than the human genome, kentucky bluegrass has more than double, spruces have around 7x as many, pines have over 10x, but oak has less than a third.

So it's kind of all over the place. Even relatively similar plant species within the same family can have massively different genome sizes.

Chinese chestnuts have a comparatively small genome but they were only sequenced in the last 4 years and to my knowledge they are the only chestnut to have been sequenced.

[acomjean]

Plants are wierd.

“Polyploid organisms have more than two sets of homologous chromosomes. For example, humans have two sets of homologous chromosomes, meaning that a typical human will have 2 copies each of 23 different chromosomes, for a total of 46. Wheat on the other hand, while having only 7 distinct chromosomes, is considered a hexaploid and has 6 copies of each chromosome, for a total of 42.”

https://en.m.wikipedia.org/wiki/Plant_genetics

[robwwilliams]

This was often NOT done in many transgenic experiments 20 to 30 years ago. In mice transgenic insertions and inadvertent disruption of genes near the transgene insertion are only occasionally disruptive. The reelin gene disruption is a good example.

This work was started long before cheap whole genome sequencing.

However, comments are right—it was never too hard to examine the flanking sequence around the insertion site.

Just offered to help the team with telomere-to-telomere of the few rare wild American chestnut survivors. There are not that many.

[anonymouskimmer]

> There are not that many.

Ballpark how many is not that many?

[robwwilliams]

Under 20 is my understanding.

[anonymouskimmer]

You don't need to sequence the entire genome to know where copies of the gene of interest are located.

I'll edit that the research still looks worthwhile for the information they discovered. Using a highly expressing constitutive promoter showed that OxO is a good gene candidate for resistance, and now they can tailor it more with the wound-inducible promoter.