Ectopic cambia: Connections between natural and experimental vascular mutants

One of the most remarkable innovations in the evolution of vascular plants is secondary growth: the developmental process by which plants grow thicker. The textbook illustration of secondary growth—a core of secondary xylem surrounded by a sheath of secondary phloem—is generated through a cylindrical meristem called the “vascular cambium” (Figure 1A). This modality is conserved across thousands of species, providing mechanical support for ever-elongating shoots, and a regenerative source of vascular tissues to feed the ever-expanding crown of branches and leaves. However, like all “rules” in biology, exceptions exist and some are rather extreme. In this essay, we discuss the phenomenon of “ectopic cambia” (Cunha Neto, 2023), broadly defined as cases where additional vascular cambia arise in atypical locations, in addition to the first-formed original vascular cambium (Figure 1B). The diversity of ectopic cambia has been studied exclusively from an anatomical perspective (e.g., Carlquist, 2007), thus, nothing is currently known about the molecular underpinnings of this convergent phenotype in naturally occurring plants. Within the experimental literature on model plants, however, we find promising clues that ectopic cambia can be induced through the misexpression of conserved genes related to vascular development. Here, we draw parallels between the independent, yet synergistic, insights from naturally occurring and transgenic phenotypes to draw hypotheses on the evolution of development of ectopic cambia across plants. In addition to the original vascular cambium, additional cambia can emerge in atypical locations in taxa from at least 17 seed plant lineages (Figure 1C). Several terms have been adopted to describe these developmental variants. In general, “successive cambia” refers to concentric rings of cambium and its products encircling the stem that superficially resemble growth rings (e.g., Tamaio et al., 2009; Figure 1B), but this term has also been used to describe discrete secondary vascular units (e.g., Nejapa et al., 2022). By contrast, “neoformations” describes round vascular units, i.e., each ectopic cambium is cylindrical (Leal et al., 2020; Figure 1B). Lastly, “corded” was introduced exclusively to describe the anatomy of Thinouia (Sapindaceae) stems consisting of exactly five equidistant circular neoformations (Tamaio and Somner, 2010); however, variations on this pattern have been reported more recently (Acevedo-Rodríguez, 2015 onwards). The proliferation of terms has proven useful for species descriptions and carries systematic value for species identification in the field because they are visible to the naked eye. However, this anatomy-based classification system has obscured our ability to see the remarkable similarities between forms that collectively represent cases of de novo cambia. From a developmental perspective, “successive cambia,” “neoformations,” and “corded” all arise through the same developmental steps—existing parenchymatous cells become meristematic and give rise to a localized cambium that will generate secondary xylem and/or phloem (Figure 2). Hence, here we adopt a development-based term—“ectopic cambia”—that is unifying in representing all forms of de novo cambia formation (Cunha Neto, 2023) and is consistent with terminology in molecular biology (Zhu et al., 2018). The meaning of “ectopic” differs depending on the context. In evolutionary biology, “ectopic” in ectopic cambia refers to the formation of new cambia in atypical locations relative to the putative ancestral state (Cunha Neto, 2023). By contrast, in studies that characterize the phenotypic outcomes of aberrant gene expression of conserved vascular genes, “ectopic” refers to a new location with respect to the typical development of a given plant. For example, overexpression of the conserved Class III homeodomain-leucine zipper (HD-ZIP III) gene, popREVOLUTA (PRE), generated ectopic cambia in the cortex of Populus (Robischon et al., 2011; Figure 3A). Thus, these are considered “ectopic” because wild type Populus only develops a single cambium. Can experimentally induced phenotypes such as this be leveraged to understand the repeated evolution of naturally occurring ectopic cambia? Not only do transgenic phenotypes highlight candidate genes involved in ectopic cambia formation, they also potentially hint at why they arise where they do. In naturally occurring plants, ectopic cambia always arise from parenchyma, but depending on the species, parenchyma will be from vascular tissue in the stem, i.e., pericycle (Tamaio et al., 2009) and phloem (Leal et al., 2020), or from the ground tissue, i.e., the cortex (Leal et al., 2020) or the pith (Cunha Neto, 2023). Examples of this diversity are illustrated in Figure 2. Rarely do ectopic cambia arise in both cortical and vascular parenchyma (Rajput et al., 2006). Interestingly, the spatial specificity of ectopic cambia is recapitulated in the experimental mutants. Four scenarios have been reported to produce experimentally induced ectopic cambia in model species, all of which involve misexpression of conserved genes related to vascular development: (1) overexpression of HD-ZIP III members in the stems of poplar (Robischon et al., 2011, Zhu et al., 2013, Zhu et al., 2018); (2) overexpression of phloem-derived CLAVATA3/Embryo Surrounding Region (CLE) genes in the stems of poplar (Kucukoglu et al., 2017); (3) the removal of putative negative regulators (PTL), which enhanced the effects of overexpressing WUSCHEL HOMEOBOX RELATED 4 (WOX4) in the stems and roots of Arabidopsis (Zhang et al., 2019); and (4) high levels of auxin signaling in roots of Arabidopsis (Smetana et al., 2019). HD-ZIP III member, REVOLUTA (REV) has been long known to play a critical role in cambium activity and xylary fiber differentiation, and this was further validated by the findings that REV was independently lost in Nymphaea (Povilus et al., 2020) and the aquatic sea grass Zostera marina (Roodt et al., 2019)—two lineages that evolutionarily lost their cambium. The loss of REV in cambiumless lineages compliments the finding that the overexpression of the REV ortholog PRE produces cortical ectopic cambia in Populus trichocarpa (Nisqually1) (Figure 3A; Robischon et al., 2011). Intriguingly, overexpression of other HD-ZIP III members, PtrHB4 (Figure 3B; Zhu et al., 2018) and PtrHB7 (Figure 3C; Zhu et al., 2013) both produce ectopic cambia and secondary vascular tissue in the cortex of Populus × euramericana cv. ‘Nanlin895’. Although these works suggest overexpression of HD-ZIP III genes can induce cortex-derived ectopic cambia, this alone is not sufficient to hypothesize the molecular origins of the vast majority of ectopic cambia, which instead arise from vascular tissues (pericycle or phloem parenchyma). CLE genes present a missing puzzle piece in the origins of vascular-derived ectopic cambia. CLE 41/42/44 comprise the TRACHEARY ELEMENT DIFFERENTIATION INHIBITORY FACTOR (TDIF) peptide, generated in the phloem, which interacts with the receptor, PHLOEM INTERCALATED WITH XYLEM (PXY). In turn, PXY regulates WOX4 and WOX14 to promote procambium and cambium proliferation in a conserved manner across vascular plants (Hirakawa and Bowman, 2015; Etchells et al., 2016). Interestingly, in lineages that evolutionarily lost their cambium, CLE genes have unique substitutions at conserved amino acids, indicating pseudogenization (Povilus et al., 2020). Ubiquitous overexpression of members of the phloem-derived CLE family in poplar (Kucukoglu et al., 2017) have led to ectopic cambia in the phloem (Figure 3D). The above-mentioned findings suggest that the misexpression of conserved-vascular genes in atypical locations, may lead to localized developmental aberrations that create ectopic cambia. Building on this working hypothesis, we also note that auxin, as AUXIN RESPONSE FACTOR (ARF), has been shown to have the capacity to generate ectopic expression of WOX4, PXY, and AtHB8 (Smetana et al., 2019). Auxin is further implicated because it plays a role in wound-induced cambium formation by inducing Arabidopsis NAC DOMAIN CONTAINING PROTEIN (ANAC) in incised Arabidopsis shoots (reviewed in Shimadzu et al., 2023). Therefore, localized auxin signaling may define the site of an ectopic cambium through misexpression of conserved vascular genes. Ectopic cambia present a unique opportunity to investigate how additional vascular cambia are generated de novo; a phenomenon that is apparently evolutionarily labile, given the repeated evolution of this developmental variant across plants. To understand which genes underly the convergent evolution of ectopic cambia and whether they are controlled through similar or diverent mechanisms, comparative transcriptomics and functional genetic studies in nonmodel species are needed. Moreover, finescale spatial-temporal transcriptomics has the power to reveal if de novo cambia arise by parenchyma de-differentiation followed by re-differentation or transdifferentiation (a cell directly shifts identity to another cell type without returning to previous developmental state). In addition to differential gene expression, it is possible that the evolutionary loss of genes and/or gene duplication may have played a role. We would like to emphasize that the terms “successive cambia,” “neoformations,” and “corded” remain useful and should persist in species descriptions because they are precise in defining the anatomy of a species. Yet, to take this research from descriptive to a mechanistic understanding, we believe it is best to adopt the unifying concept of ectopic cambia grounded in an evolutionary and developmental framework, because this facilitates comparative studies between distantly related groups. By adopting the concept of ectopic cambia, we more readily see the links with transgenic phenotypes in model species that also use the term “ectopic cambia” (e.g., Kucukoglu et al., 2017) as a starting place to understand the naturally occuring ectopic cambia across plants. J.G.O. wrote the first draft, and I.L.C.N edited and heavily contributed to subsequent drafts and the final manuscript. The authors thank Dr. Mariane Sousa-Baena and Yanã Campos Rizzieri for discussions that influenced the perspectives in this article. We would also like to thank our reviewers, and Dr. Aman Husbands for reading and advising us on figure design and nomenclature. This work was supported by Lab Startup Funds from Cornell University and New York University to J.G.O. All data are presented within the text with citations.