We have cloned a SOMATIC EMBRYOGENESIS RECEPTOR KINASE (SERK) gene from Medicago truncatula (MtSERK1) and examined its expression in culture using real time PCR. In the presence of the auxin 1-naphthaleneacetic acid (NAA) alone, root differentiation occurs from the proliferating calli in both the cultured highly embryogenic seed line (2HA) and a low to nonembryogenic seed line (M. truncatula cv Jemalong). Auxin stimulated MtSERK1 expression in both 2HA and M. truncatula cv Jemalong. Embryo induction in proliferating calli requires a cytokinin in M. truncatula and unlike root formation is substantively induced in 2HA, not M. truncatula cv Jemalong. On embryo induction medium containing NAA and the cytokinin 6-benzylaminopurine (BAP), expression of MtSERK1 is elevated within 2 d of initiation of culture in both M. truncatula cv Jemalong and 2HA. However, MtSERK1 expression is much higher when both NAA and BAP are in the medium. BAP potentiates the NAA induction because MtSERK1 expression is not up-regulated by BAP alone. The 2HA genotype is able to increase its embryo formation because of the way it responds to cytokinin, but not because of the cytokinin effect on MtSERK1. Although the studies with M. truncatula indicate that somatic embryogenesis is associated with high SERK expression, auxin alone does not induce somatic embryogenesis as in carrot (Daucus carota) and Arabidopsis. Auxin in M. truncatula induces roots, and there is a clear up-regulation of MtSERK1. Although our analyses suggest that MtSERK1 is orthologous to AtSERK1, which in Arabidopsis is involved in somatic embryogenesis, in legumes, MtSERK1 may have a broader role in morphogenesis in cultured tissue rather than being specific to somatic embryogenesis.

Transcriptional profiling of embryogenic callus produced from Medicago truncatula mesophyll protoplasts indicated up-regulation of ethylene biosynthesis and ethylene response genes. Using inhibitors of ethylene biosynthesis and perception, it was shown that ethylene was necessary for somatic embryogenesis (SE) in this model legume. We chose several genes involved in ethylene biosynthesis and response for subsequent molecular analyses. One of these genes is a gene encoding a transcription factor that belongs to the AP2/ERF superfamily and ERF subfamily of transcription factors. We demonstrate that this gene, designated M. truncatula SOMATIC EMBRYO RELATED FACTOR1 (MtSERF1), is induced by ethylene and is expressed in embryogenic calli. MtSERF1 is strongly expressed in the globular somatic embryo and there is high expression in a small group of cells in the developing shoot meristem of the heart-stage embryo. RNA interference knockdown of this gene causes strong inhibition of SE. We also provide evidence that MtSERF1 is expressed in zygotic embryos. MtSERF1 appears to be essential for SE and may enable a connection between stress and development.

Summary: Mitochondrial fusion in plants and its role in development are poorly understood. Cultured tobacco mesophyll protoplasts provide an excellent experimental system for visualizing mitochondrial dynamics. Before protoplasts first divide, mitochondria undergo a phase of extensive elongation before fission causes an increase in number, followed by actin filament (AF)‐dependent dispersion that distributes mitochondria uniformly throughout the cytoplasm. Here, by fusing protoplasts containing either green fluorescent protein‐ or MitoTracker‐labelled mitochondria, we show that elongation results from fusion during early (4–8 h) protoplast culture. This massive mitochondrial fusion (MMF) leads to near‐complete mixing of the mitochondrial population within 24 h. Staining isolated mitochondria with 4′,6‐diamidino‐2‐phenylindole (DAPI) revealed that in freshly prepared protoplasts mitochondrial nucleoids were unequally distributed, with many mitochondria failing to stain with DAPI, suggesting the presence of an incomplete mitochondrial genome. Following MMF, nucleoids were distributed evenly throughout the population, thereby ensuring continuity of the mitochondrial genome in daughter cells. Massive mitochondrial fusion appears to be specific to dedifferentiation, since it also occurs in mesophyll protoplasts of Arabidopsis and Medicago but not in protoplasts from already dedifferentiated cells such as BY‐2 or callus cultures. Efficient MMF requires an inner membrane electrical gradient, cytoplasmic protein synthesis, microtubules and functional kinesin but not ATP or AFs, indicating fundamental differences from mitochondrial fusion in non‐plant systems. Our studies reveal that individual mitochondria are connected over time by fusion events, a finding that allows a clearer interpretation of how novel mitochondrial genotypes develop following cell fusion, and indicates that developmentally regulated fusion ensures continuity of the mitochondrial genome.

Somatic embryogenesis (SE) is induced in vitro in Medicago truncatula 2HA by auxin and cytokinin but rarely in wild type Jemalong. The putative WUSCHEL (MtWUS), CLAVATA3 (MtCLV3) and the WUSCHEL-related homeobox gene WOX5 (MtWOX5) were investigated in M. truncatula (Mt) and identified by the similarity to Arabidopsis WUS, CLV3 and WOX5 in amino acid sequence, phylogeny and in planta and in vitro expression patterns. MtWUS was induced throughout embryogenic cultures by cytokinin after 24–48 h and maximum expression occurred after 1 week, which coincides with the induction of totipotent stem cells. During this period there was no MtCLV3 expression to suppress MtWUS. MtWUS expression, as illustrated by promoter-GUS studies, subsequently localised to the embryo, and there was then the onset of MtCLV3 expression. This suggests that the expression of the putative MtCLV3 coincides with the WUS-CLAVATA feedback loop becoming operational. RNAi studies showed that MtWUS expression is essential for callus and somatic embryo production. Based on the presence of MtWUS promoter binding sites, MtWUS may be required for the induction of MtSERF1, postulated to have a key role in the signalling required for SE induced in 2HA. MtWOX5 expressed in auxin-induced root primordia and root meristems and appears to be involved in pluripotent stem cell induction. The evidence is discussed that the homeobox genes MtWUS and MtWOX5 are “hijacked” for stem cell induction, which is key to somatic embryo and de novo root induction. In relation to SE, a role for WUS in the signalling involved in induction is discussed.

SOMATIC EMBRYOGENESIS RECEPTOR-LIKE KINASE (SERK) genes have been demonstrated to play a role in somatic embryogenesis in several plant species. As more is learnt about these genes, the view of their role in plant development has broadened. The Medicago truncatula MtSERK1 gene has been associated with somatic embryogenesis and in vitro root formation. In order to study the role of MtSERK1 in development further, the MtSERK1 promoter sequence has been isolated and cloned into a promoter–GUS analysis vector. SERK1 promoter-driven GUS expression was studied in A. tumefaciens-transformed cultures and regenerated plants, in A. rhizogenes-transformed root clones, and in nodulation. In embryogenic cultures, GUS staining is detected after 2 d of culture at the edge of the explant and around vascular tissue. Expression at the explant edge intensifies over subsequent days and then is lost from the edge as callus formation moves inward. MtSERK1 expression appears to be associated with new callus formation. When somatic embryos form, GUS staining occurs throughout embryo development. Zygotic embryos show expression until the heart stage. The in planta studies reveal a number of interesting expression patterns. There appear to be three types. (i) Expression associated with the primary meristems of the root and shoot and the newly formed meristems of the lateral roots and nodule. (ii) Expression at the junction between one type of tissue or organ and another. (iii) Expression associated with the vascular tissue procambial cells. The data led us to conclude that MtSERK1 expression is associated with developmental change, possibly reflecting cellular reprogramming.

Medicago truncatula Gaertn. cv. Jemalong, a pasture species used in Australian agriculture, was first proposed as a model legume in 1990. Since that time M. truncatula, along with Lotus japonicus (Regal) Larsen, has contributed to major advances in understanding rhizobia Nod factor perception and the signalling pathway involved in nodule formation. Research using M. truncatula as a model has expanded beyond nodulation and the allied mycorrhizal research to investigate interactions with insect pests, plant pathogens and nematodes. In addition to biotic stresses the genetic mechanisms to ameliorate abiotic stresses such as salinity and drought are being investigated. Furthermore, M. truncatula is being used to increase understanding of plant development and cellular differentiation, with nodule differentiation providing a different perspective to organogenesis and meristem biology. This legume plant represents one of the major evolutionary success stories of plant adaptation to its environment, and it is particularly in understanding the capacity to integrate biotic and abiotic plant responses with plant growth and development that M. truncatula has an important role to play. The expanding genomic and genetic toolkit available with M. truncatula provides many opportunities for integrative biological research with a plant which is both a model for functional genomics and important in agricultural sustainability.