De novo shoot organogenesis: from art to science Je ´ ro ˆ me Duclercq 1, 2 , Brigitte Sangwan-Norreel 1 , Manuella Catterou 1 and Rajbir S. Sangwan 1 1 Universite ´ de Picardie Jules Verne, Unite ´ de Recherche EA3900 – Laboratoire Androgene ` se et Biotechnologie, Faculte ´ des Sciences, 33 Rue Saint-Leu, 80039 Amiens, France 2 Hormonal Crosstalk Group, VIB Department of Plant Systems Biology, Ghent University, Technologiepark 927, B-9052 Ghent, Belgium In vitro shoot organogenesis and plant regeneration are crucial for both plant biotechnology and the fundamen- tal study of plant biology. Although the importance of auxin and cytokinin has been known for more than six decades, the underlying molecular mechanisms of their function have only been revealed recently. Advances in identifying new Arabidopsis genes, implementing live- imaging tools and understanding cellular and molecular networks regulating de novo shoot organogenesis have helped to redefine the empirical models of shoot organ- ogenesis and plant regeneration. Here, we review the functions and interactions of genes that control key steps in two distinct developmental processes: de novo shoot organogenesis and lateral root formation. De novo shoot and lateral root organogenesis De novo shoot organogenesis (DNSO) relies on somatic cell totipotency (i.e. the capacity to regenerate in vitro the entire plant from single somatic cells) and it is the most common pathway leading to in vitro plant regeneration [1– 3]. It has been demonstrated that in vitro plant regenera- tion occurs via two major pathways, de novo organogenesis and somatic embryogenesis, both of which are dependent on phytohormone perception, cell division and dedifferen- tiation to acquire organogenetic competence, organ initi- ation and development [1,4,5]. Both regeneration pathways are extensively used, either for research or practical applications [3]. However, in contrast to the embryogenic pathway, the de novo organogenetic pathway is more often used in biotechnological breeding methods (i.e. in vitro micropropagation, haploid production and genetic engineering), particularly in dicotyledonous plants, mainly because the plant explants and in vitro conditions are relatively simple and more robust. Here, we review recent studies that have advanced our understand- ing of the molecular networks involved in the DNSO process. Furthermore, we will compare this process to de novo lateral root (LR) organogenesis. LR formation has been used extensively as a model system to study the molecular mechanisms of phytohormonal interactions regulating de novo organogenesis [6–9]. The molecular and cellular basis of LR formation has been most exten- sively studied in Arabidopsis (Arabidopsis thaliana) and new genes that regulate LR initiation, patterning and emergence processes have been identified [7,8,10–16]. LRs are initiated when root pericycle founder cells under- go anticlinal divisions to create a dome-shaped root pri- mordium [17]. Auxin plays a crucial role in regulating this process [7,8,18,19]. Recently, molecular data on the mech- anism of auxin transport, auxin distribution and/or gra- dients and its regulation have become available, providing significant insights into the regulation of LR developmen- tal processes, and thus shedding new light on the molecu- lar control of DNSO [7,8,20–25]. Temporal dissection of the organogenetic processes De novo organogenesis here refers to the in vitro formation of shoots and roots from cultured explants. This organoge- netic process is influenced by the type of explant used as well as environmental and chemical factors (i.e. phytohor- mones) [2,26–29]. The classical finding of Skoog and Miller [29], showing the importance between the ratio of auxin and cytokinin (CK), is still the guiding principle of in vitro organogenesis. A high CK to auxin ratio induces shoot organogenesis, whereas opposite low ratio results in root development. This chemical regeneration pattern has been demonstrated over a wide range of plant species [1,3,30] and it is illustrated in Figure 1a, which shows the forma- tion of shoots, roots and callus from cultured leaf explants. Based on the function and requirements of the exoge- nous phytohormones [31–34], DNSO is generally divided into three morphological stages: (i) morphogenic compe- tence acquisition (cell dedifferentiation), (ii) induction (cell determination for specific organ formation in response to exogenous phytohormones), and (iii) morphological differ- entiation (organ morphogenesis proceeding independently of exogenous phytohormones). Unfortunately, in vitro re- generation of most important crop plants does not strictly follow these stages; hence, the development of predictable and routine DNSO remains a challenge. In particular, the common elements behind plant cell totipotency that pro- vide varied plant cells with their remarkable in vitro regeneration ability are still not well known. However, during the past few years, molecular geneticists have begun taking up the greater challenge of the molecular dissection of this process, using A. thaliana regeneration systems [27,35–42]. Review Corresponding author: Sangwan, R.S. (rajbir.sangwan@u-picardie.fr). 1360-1385/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.tplants.2011.08.004 Trends in Plant Science, November 2011, Vol. 16, No. 11 597