Updated January 27, 2021:
March 15, 2020 Report
Our focus over the past three months has been on selecting for and propagating homozygous T2 a-CT overexpression and BADC RNAi soybean lines. To that end, we have now produced 7 probable homozygous a-CT lines from independent transformation events (5 in Thorne genetic background and 2 in Williams 82), with additional screening in progress. We are aiming to obtain 6 independent stable a-CT lines per genetic background. As mentioned previously, based on preliminary batch-wise FT-NIR oil content measurements, many of these a-CT lines display increased oil content, while also displaying a pronounced increase in 18:1 to 18:3 fatty acid ratio (increased oleic relative to linolenic acid). We are in the process of verifying a-CT heterologous overexpression in these lines and of obtaining reference oil content measurements, after which we will commence mapping of transgene insert locations via Illumina and/or Nanopore sequencing as described in our proposal.
Unfortunately, recent evidence suggests that some BADC proteins may actually function as activators, rather than inhibitors, of ACCase enzyme activity (Shivaiah et al. 2020). Based on a shrunken and wrinkled seed phenotype, low germination rate, and possible lower oil content in our T1/T2 BADC RNAi soybean seeds (based on preliminary QSorter NIR measurements that require further validation), it appears that in soybean, BADC repression may reduce, rather than promote, seed oil accumulation. We are currently undertaking work to verify successful BADC silencing in these lines, which must be completed before any final conclusions can be made. Due largely to the low germination rate of our T2 BADC RNAi seeds we are experiencing difficulties in recovering stable homozygous BADC RNAi lines. We will be resorting to seed germination on sterile sucrose-containing media in order to try to recover homozygous BADC RNAi lines, which will cause significant project delays. While our BADC RNAi soybean lines do not, unfortunately, thus far display the high-oil phenotype that we originally hypothesized would be present, these lines, if verified to have successful BADC silencing, will nonetheless be informative in better understanding the regulation of seed oil accumulation in soybean.
Unfortunately, we must correct several erroneous claims from our previous USB grant applications and reports. First, our current USB grant proposal claimed that Arabidopsis CTI knockouts have up to 20% higher seed oil content. In fact, no clear evidence of an effect on seed oil exists. Second, we previously claimed that overexpression of a-CT raised seed oil content in Arabidopsis thaliana and Camelina sativa by up to 60%. It is in fact lower. We apologize for these errors.
Updated January 27, 2021:
September 15, 2020 Report
Continuing the work described in our previous report, we have completed collection of T3 seeds from putative homozygous a-CT overexpression lines. We have collected seeds from 7 of 7 independent putative homozygous a-CT lines (5 of 5 lines in Thorne genetic background and 2 of 2 lines in Williams 82 background), and from 3 of 3 high-oil segregating lines (which were propagated due to elevated preliminary oil measurements in spite of not reaching homozygosity in the T2 generation). In addition, we collected T2 seeds from additional independent Williams 82 a-CT lines, which will be used to meet our revised goal of producing 5 independent homozygous a-CT overexpression lines per genetic background. We also collected T2 seeds from additional BADC RNAi lines to compensate for low germination efficiency and for the difficulty in recovering homozygous BADC RNAi lines that we described previously. We have arranged for a repeat visit to Corteva’s Johnston, Iowa campus, now tentatively scheduled for either this month or next (depending on equipment availability and travel permissions), during which we will obtain oil and protein measurements for all of our latest seeds via Bruker Tango FT-NIR.
Due to the University Covid-19 shutdown, we had to forego most a-CT and BADC transcript/protein expression profiling (and some genotyping) in the previous generation of developing T3 and T2 seeds, and this profiling has been pushed back to the following generation (see previous report for more details). We are currently starting the next generation of plants and anticipate that developing seeds will be ready for sampling around February of 2021. Developing seed samples will be collected in excess from all plants and stored at -80 °C, to allow for follow-up transcriptomic and proteomic profiling on the most interesting lines. Based upon preliminary immunoblotting conducted before the shutdown, it is unclear whether the putative higher oil content observed in transgenic a-CT OE soybeans is due to bona fide heterologous pea a-CT overexpression or whether it may actually be due to unintentional silencing of native soybean a-CT(s). Ascertaining which of these two scenarios, if either, is actually occurring in vivo will be one of the main priorities of our expression profiling. For most lines, we anticipate that this next generation will be the last grown in the greenhouse, although a subsequent generation will be necessary to provide an additional 3 independent event replicates for Williams 82 a-CT overexpression lines and may also be necessary for various BADC RNAi lines (TBD). Depending on the oil measurements obtained from FT-NIR, and depending on the success of our next funding proposal, we may still attempt to grow high-oil a-CT OE lines in the field next summer, although Covid-19-related delays and an ongoing Covid-19 personnel shortage may make that original objective unrealistic.
As we mentioned previously, we are experiencing difficulties in recovering stable homozygous BADC RNAi lines. We have tentatively ruled out seed abortion since we have not observed an obvious increase in aborted seeds in BADC RNAi pods. While we initially hypothesized that the difficulty in homozygote recovery might be due to the low germination rate of severely wrinkled BADC RNAi seeds, in which homozygous transformants might be over-represented, it is difficult to rationalize such copy-number-dependent lethality in an RNAi line. Furthermore, skewed segregation ratios in surviving seedlings suggest that the lack of homozygote recovery could instead be due to reduced pollen viability, survival, or fertility. Although the BADC RNAi cassette in our plants is driven by a seed-specific glycinin (Gly1) promoter, we are unaware of any published examination of Gly1 expression in pollen (and there are no pollen data points in public soybean expression atlases), so it is possible that the Gly1 promoter may be ‘leaky’ and that some BADC RNAi silencing may in fact occur in pollen. Alterations of ACCase subunit expression have previously been reported to mediate nuclear-cytoplasmic conflict in pea, manifesting in decreased pollen viability, survival, or fertility (Bogdanova et al. 2015), so it seems plausible that other ACCase alterations may manifest in a similar manner. To test whether the presence of the BADC RNAi cassette influences pollen viability or survival, we will quantitatively compare the morphology and germination of heterozygous BADC RNAi and wild-type pollen. We will also quantify seed number per pod and the number of aborted seeds per pod in the T3 generation to test our initial assessment regarding the absence of excess BADC RNAi seed abortion. Finally, while we previously discussed our plan to attempt germination of BADC RNAi seeds on sucrose media (in case the difficulty in homozygote recovery does in fact arise from impaired germination of homozygous seeds), we have decided to delay such efforts due to the time required as well as their possible inutility. We will be able to better assess their potential after we have obtained additional BADC RNAi segregation ratio data and assessed BADC RNAi pollen viability. The latter data will also provide an indication of whether it would be worthwhile to attempt the direct profiling of BADC and glycinin transcript expression in BADC RNAi and WT soybean pollen (which could be technically challenging without resorting to single-cell sequencing technology).