Making babies: how auxin regulates somatic embryogenesis in Arabidopsis tissue culture

In 1958, F. C. Steward, Marion O. Mapes, and Kathryn Mears released reports that outlined a profound discovery: plant cells could not only proliferate in culture, they could also be coaxed into regenerating entire plant bodies with roots and shoots (Steward et al., 1958). This work documented what we now call somatic embryogenesis. In the decades since this discovery, somatic embryogenesis has become commonplace in plant biology, since it can facilitate genetic transformation and clonal propagation. However, the mechanisms underlying somatic embryogenesis remain somewhat obscure. Auxin is involved somehow, since the most common means of inducing somatic embryogenesis is treatment of cultured tissues with exogenous auxins, especially the synthetic auxin 2,4-dichlorophenoxyacetic acid (2,4-D) (Wójcik et al., 2020). Kim Boutilier made a significant contribution to this field in 2002 when she discovered a gene that could, when ectopically expressed under the control of a constitutive promoter, induce the formation of somatic embryos in seedlings without the addition of exogenous hormones (Boutilier et al., 2002). She named this gene BABYBOOM (BBM) and determined that it encodes an AP2/ERF transcription factor that is normally expressed in developing seeds and embryos. Since its discovery, scientists have shown that BBM can be used to facilitate somatic embryogenesis in a wide range of plant species (Jha & Kumar, 2018). Remko Offringa was also involved in the discovery of BBM, as he was investigating the role of auxin in Arabidopsis development in his lab at Leiden University. The finding triggered his interest in the process of somatic embryogenesis, and he began to study another gene that could also induce somatic embryogenesis, called AT-HOOK NUCLEAR LOCALIZED 15 (AHL15). AHL15 is a member of a large gene family that encodes nuclear-localized proteins that bind to AT-rich regions of the genome. The exact function of AHL proteins remains somewhat mysterious but their mechanism of action might involve either direct transcriptional regulation of target genes or more global regulation by altering chromosome structure. A PhD student (and later post-doc) in the Offringa lab, Omid Karami, discovered that overexpression of AHL15 leads to some interesting phenotypes, including changes in ploidy levels (Karami et al., 2021) and even converting Arabidopsis into a woody, perennial plant (Karami et al., 2020; Rahimi et al., 2022). Auxin once again seemed to be involved in AHL15-triggered somatic embryogenesis, but the exact roles endogenous auxin production and auxin influx/efflux played in initiating and maintaining somatic embryogenesis remained in question. This is what Offringa and Karami set out to address in collaboration with Boutilier (Karami et al., 2022). First, they used an embryo identity visual marker transgene, pWOX2:NLS-YFP, for time-lapse imaging of embryo development in somatic embryos induced either by exogenous 2,4-D or by AHL15 overexpression. In both cases, pWOX2:NLS-YFP expression was not detected in the first four days of culture, but it was weakly expressed over the next two days in areas with embryogenic protrusions, and it was strongly expressed for about two days in dividing embryonic clusters. Expression was reduced as the embryos reached the globular stage and beyond. These observations led them to define three distinct developmental stages in somatic embryo induction (Figure 1). They then assessed the role of endogenous auxin production in the initiation and maintenance of somatic embryos in 2,4-D-treated tissues. To do this, they measured expression of the YUCCA (YUC) genes, which encode monooxygenases that convert indole-3-pyruvic acid into indole-3-acetic acid (IAA), the most important endogenous auxin in plants. Expression of YUC6, 7, 8, and 9 was strongly induced in somatic embryo forming tissues eight days after 2,4-D treatment. In order to further dissect the role of YUCs in somatic embryogenesis, they treated explant tissues with yucasin, an inhibitor of YUC activity, and found that this severely reduced the number of somatic embryos that were formed. Interestingly, pWOX2:NLS-YFP expression was detected in yucasin-treated tissue around day 6 of culture, but it disappeared two days later. This suggests that endogenous auxin is not required for the induction of somatic embryogenesis, but is required for the maintenance of embryonic cell identity. They then used the auxin-responsive reporter pDR5:GFP to visualize auxin dynamics in AHL15-overexpressing tissues. pDR5:GFP was strongly expressed after about six days in culture, just where somatic cells were beginning to acquire embryo identity. This expression pattern was not observed in the wild-type background. When they assessed YUC expression in these tissues, they observed elevated expression of YUC6, 7, 8, and 9 after about five days, which became even stronger after seven days. This suggests that AHL15 induces YUC transcription, which in turn elevates endogenous auxin levels and leads to somatic embryogenesis. When they were assessing the effect of yucasin treatment on AHL15-overexpressing tissue, they also noticed a significant reduction in the number of somatic embryos. Embryogenesis could be rescued, however, by exogenous IAA, confirming the observation that auxin is required for somatic embryogenesis. They also showed that expression of AHLs is required for 2,4-D-induced somatic embryogenesis, since inhibition of their expression prevented the production of somatic embryos even in the presence of exogenous 2,4-D. Furthermore, ahl knockout mutants had significantly reduced YUC expression, corroborating the previous conclusion that YUC genes act downstream of AHL15. Lastly, they applied either the auxin efflux inhibitor N-1-naphthylphthalamic acid (NPA) or the auxin influx inhibitor 1-naphthoxyacetic acid (1-NOA) to AHL15-overexpressing and 2,4-D-treated tissues. In both cases, auxin influx and efflux were not required for the acquisition of embryo identity, but were required for its maintenance and the successful completion of somatic embryogenesis. These findings clarified the role auxin plays in somatic embryogenesis, but also raised two new questions. First, if endogenous auxin is not involved in the acquisition of embryo identity, how can somatic embryogenesis be triggered by the auxin analog 2,4-D? The authors speculate that large-scale chromatin remodeling might also be necessary for acquisition of embryo identity, and this could be triggered by sufficiently high doses of 2,4-D. Second, why is 2,4-D itself incapable of maintaining embryonic cell identity? This might be due to a low cell-to-cell transport efficiency of 2,4-D, but both of these questions require additional experimentation. Boutilier's lab at Wageningen continues to explore the effects of transcription factors and chromatin remodeling on the promotion of embryo identity. Offringa's lab at Leiden University is further analyzing the broad role of AHL15 in plant development and trying to figure out why 2,4-D specifically is so good at inducing somatic embryogenesis.