Hardwood species are valuable biological resources that have an important role in the economy and ecology of ecosystems worldwide. Non-zygotic or somatic embryogenesis (SE) is a powerful tool in plant biotechnology as it is a form of clonal propagation, amenable to cryopreservation of valuable germplasm and genetic transformation including gene editing. The SE process involves five steps and includes somatic embryo induction, proliferation, maturation, plantlet conversion, and subsequent plant acclimatization. This review aims to provide a general overview of these steps in different SE systems developed for hardwood species. Factors that influence the induction stage such as the age of the donor plant, genotype and culture media are discussed. The role of different explant types, i.e. zygotic embryos and non-zygotic tissues, such as roots, flower tissues, nodes, internodes, leaves or shoot apices, in SE induction are especially emphasized. Histological studies of the origin of somatic embryos and the sequence of events leading to their development from initial explants are assessed. Maintenance of embryogenic capacity carried out by subculture of embryogenic inocula on semisolid or liquid media through cell suspension cultures or by temporary immersion systems is described. At present, the main concerns associated with the application of SE for large-scale propagation of elite hardwoods are related to the embryo maturation, germination, and plantlet conversion steps, and these are highlighted in this review. Finally, molecular aspects associated with somatic embryo induction and development are also described. Attempts to overcome the hurdles identified in the embryogenic process, and future lines of research are proposed.

Experiments were performed to determine the influence of proliferation medium on the maintenance of embryogenic competence and on repetitive embryogenesis in Castanea sativa Mill. somatic embryos derived from leaf explants. Somatic embryo proliferation was carried out by both direct secondary embryogenesis and by the culture of nodular callus tissue originated from cotyledons of somatic embryos. Both systems led to the production of cotyledonary somatic embryos on Murashige and Skoog proliferation medium supplemented with 0·1 mg l^–1 benzyladenine and 0·1 mg l^–1 naphthaleneacetic acid. Carbon source and concentration had a marked influence on maturation and subsequent germination ability of chestnut somatic embryos. Plantlet conversion was achieved in embryos matured on media with 6 % sucrose, and on 3 or 6 % maltose, whereas mean shoot length, root length and leaf number of produced plants were not significantly affected by these maturation media. Overall, the best results were obtained with 3 % maltose‐matured somatic embryos, giving rise to 6 % plant recovery in addition to 33 % of embryos exhibiting only shoot development. The application of a 2‐month cold treatment at 4 °C to somatic embryos matured on medium with 3 % maltose was necessary for achieving plant conversion, while partial desiccation did not appear to influence this response. A total of 39 % of embryos eventually produced plants either through conversion to plantlets or indirectly through rooting of shoots. Shoots formed by somatic embryos could be excised, multiplied and rooted following the micropropagation procedures previously developed for chestnut.

A reproducible procedure for induction of somatic embryogenesis (SE) from adult trees of Eucalyptus globulus Labill. and the hybrid E. saligna Smith × E. maidenii has been developed for the first time. Somatic embryos were obtained from both shoot apex and leaf explants of all three genotypes evaluated, although embryogenic frequencies were significantly influenced by the species/genotype, auxin and explant type. Picloram was more efficient for somatic embryo induction than naphthaleneacetic acid (NAA), with the highest frequency of induction being obtained in Murashige and Skoog medium containing 40 µM picloram and 40 mg l⁻¹ gum Arabic, in which 64% of the shoot apex explants and 68.8% of the leaf explants yielded somatic embryos. The embryogenic response of the hybrid was higher than that of the E. globulus, especially when NAA was used. The cultures initiated on picloram-containing medium consisted of nodular embryogenic structures surrounded by a mucilaginous coating layer that emerged from a watery callus developed from the initial explants. Cotyledonary somatic embryos were differentiated after subculture of these nodular embryogenic structures on a medium lacking plant growth regulators. Histological analysis confirmed the bipolar organization of the somatic embryos, with shoot and root meristems and closed procambial tissue that bifurcated into small cotyledons. The root pole was more differentiated than the shoot pole, which appeared to be formed by a few meristematic layers. Maintenance of the embryogenic lines by secondary SE was attained by subculturing individual cotyledonary embryos or small clusters of globular and torpedo embryos on medium with 16.11 µM NAA at 4- to 5-week intervals. Somatic embryos converted into plantlets after being transferred to liquid germination medium although plant regeneration remained poor.

Somatic embryogenesis is considered a convenient tool for investigating the regulating mechanisms of embryo formation; it is also a feasible system for in vitro regeneration procedures, with many advantages in woody species. Nevertheless, trees have shown recalcitrance to somatic embryogenesis, and its efficiency remains very low in many cases. Consequently, despite the clear potential of somatic embryogenesis in tree breeding programs, its application is limited since factors responsible for embryogenesis initiation have not yet been completely elucidated. In the present work, we investigated key cellular factors involved in the change of developmental program during leaf somatic embryogenesis initiation of white oak (Quercus alba), aiming to identify early markers of the process. The results revealed that pectin esterification, auxin accumulation and DNA demethylation were induced during embryogenesis initiation and differentially found in embryogenic cells, while they were not present in leaf cells before induction or in non-embryogenic cells after embryogenesis initiation. These three factors constitute early markers of leaf embryogenesis and represent processes that could be interconnected and involved in the regulation of cell reprogramming and embryogenesis initiation. These findings provide new insights into the mechanisms underlying plant cell reprogramming, totipotency and embryogenic competence acquisition, especially in tree species for which information is scarce, thus opening up the possibility of efficient manipulation of somatic embryogenesis induction.