The evolution of seed plants is still a contentious topic in plant biology. In particular, the phylogeny of gymnosperms and their relationship to flowering plants remains debated as morphological and molecular analyses contradict each other on key relationships. This on-going conflict can hinder the unambiguous assessment of homology through comparative morphological studies. Luckily, developmental genetics can provide another line of evidence for the homology of structures. Similar developmental mechanisms can yield additional indications for the shared ancestry of particular morphologies. In this issue of New Phytologist, Carlsbecker et al. (pp. 261–275) present their study on the molecular control of female reproductive development in Picea abies. By combining their genetic results with paleobotanical data, they are able to shed new light on the complicated evolution of the pine ovuliferous scale. Even more intriguingly, the authors provide rare genetic insights into the regulation of reproductive organ development in gymnosperms by studying the naturally occurring mutant of Norway spruce, Picea abies var. acrocona.
‘… the aberrant ovuliferous organization in acrocona mutant cones is remarkably similar to some Paleozoic conifer genera … what biological interpretation can be given to this observation?’
Conifer ovuliferous scales: an evolution of reduction
Historically, understanding the morphological evolution of conifers largely relied on comparing female reproductive organs of extinct and extant conifers, as male fossil cones are remarkably similar to modern cones. It was Rudolph Florin's seminal study on the ovuliferous organs from modern and fossil conifers and their putative predecessors, the cordaitales, that resulted in the first thorough classification of the Paleozoic conifers (summarized in Florin, 1951). Furthermore, Florin hypothesized that the evolution of the ovuliferous organs of conifers from a cordaite-like ancestor could be explained by reduction processes as used in Zimmermann's telome theory (Florin, 1951). The telome theory describes the evolution of land plant architecture through a sequential process of overtopping, planation and webbing of branches or ‘telomes’ (Zimmermann, 1952). Indeed, when examining the female reproductive organs of extinct and modern conifers in a phylogenetic context, a similar reduction trend is apparent (Florin, 1951; Clement-Westerhoff, 1988).
Among the early paleozoic conifers, those belonging to the genus Thuringiostrobus can be viewed as one of the most ancestral conifers (Fig. 1). Thuringiostrobus florinii from the Lower Permian possessed radially symmetrical ovuliferous organs, composed of several intertwined sterile and fertile scales (Fig. 1a; Clement-Westerhoff, 1988). Through the process of planation (‘the positioning of structures in one plane’), bilaterally symmetrical ovuliferous organs would subsequently have evolved from the ovuliferous organs of a Thuringiostrobus-like conifer. Such a transition can be exemplified by the fossil species Pseudovoltzia liebeana. This late Permian conifer possessed bilaterally symmetrical female reproductive organs consisting of five scales, two of which were sterile while the other three major scales bore the seeds (Fig. 1b). Further planation of the ovuliferous organs followed by the fusion and reduction of its individual scales could finally lead up to the uniform ovuliferous scale observed in most extant conifers, as, for example, in P. abies (Fig. 1c).
Carlsbecker et al. utilize beautiful in situ hybridization data to study the expression patterns of several genes – including five novel MADS-box genes – throughout the development of female cones of P. abies. In doing so, the authors discern distinct sectors within the ovuliferous scale as indicated by the distinct expression patterns of several genes. They hypothesize that the different patterns may indicate functionally and evolutionary independent sectors within the P. abies ovuliferous scale. Furthermore, they elegantly link this rather complex genetic organization of the, at first glance, simple ovuliferous scale, to Florin's hypothesis describing the evolutionary reduction of female reproductive organs of conifers. Thus the different sectors strengthen the notion that ovuliferous scales of P. abies are derived from a more compound structure that was gradually reduced throughout conifer evolution.
Genetic insights into ovuliferous scale development
When studying a nonmodel species, researchers are constantly faced with the daunting task of getting clear functional data. This specifically applies to gymnosperms, as there are almost no functional approaches available to study the genetic mechanisms underlying gymnosperm development. Yet major evolutionary questions in plant biology would benefit from such functional studies. As would, for example, the longstanding questions concerning the origin and early evolution of the angiosperm flower. To solve this problem, it has been proposed to study naturally occurring mutants that possess specific abnormalities of interest (Rudall et al., 2011). In this way, it would be possible to identify the genes involved in certain developmental mechanisms without the need for transgenic approaches. One example of such a mutant with several interesting characteristic is the naturally occurring acrocona variety of P. abies. This early-cone setting variety of Norway spruce produces, in addition to wild type cones, morphologically atypical female cones, that form when vegetative shoots gradually convert into female-cone like structures (Acheré et al., 2004; Uddenberg et al., 2013). Given these defects in cone-setting and cone morphology, the study of acrocona could answer pivotal questions concerning the mechanisms of reproductive organ development of gymnosperms. Carlsbecker et al. show that acrocona mutant cones develop aberrant ovuliferous scales composed of three lobes, rather than a single structure as in wild type (compare Fig. 1c and d). Although the exact genetic mutation or mutations underlying the acrocona phenotypes have not yet been identified (Acheré et al., 2004), changes in gene expression have been described. Recently the upregulation of DAL19, a SOC1-like gene, during cone-setting of acrocona has been suggested to be involved in the early cone-setting phenotype (Table 1; Uddenberg et al., 2013). Similarly, Carlsbecker et al. report an absence of DAL14 expression during the development of acrocona mutant cones. Although many other genes may exhibit a differential expression pattern in ovuliferous scale development, it is tempting to interpret the acrocona ovuliferous abnormalities in light of the absence of DAL14 expression. In angiosperms, the orthologs of DAL14, AGL6-like genes, function in reproductive organ specification and determination, in part, redundantly with the canonical E-class genes belonging to the SEPALLATA subfamily of MADS-box genes (Table 1; Ohmori et al., 2009; Rijpkema et al., 2009; Smaczniak et al., 2012). Intriguingly, the acrocona ovuliferous phenotype seems to exhibit characteristics consistent with such a function. Ovuliferous organ development in acrocona seems to be more shoot-like and thus acrocona may lack proper determinate development of the ovuliferous scale. Another noteworthy phenotype in which DAL14 might also be involved relates to the ovules. Carlsbecker et al. notice that acrocona ovuliferous scales often display erect ovules meaning that the micropyle is pointed away from cone axis. By contrast, ovules of P. abies are inverted. Together, this study of the acrocona mutant suggests a possible function of DAL14 in reproductive development by determining organ and meristem identities in gymnosperms, similar to SEP-like genes in angiosperms.
Table 1. MADS-box genes of Picea abies and the function of their orthologs in Arabidopsis thaliana
Functions in Arabidopsis
DAL3, DAL4, DAL9, DAL19
Activator of vegetative to reproductive transition
Sepal, petal, stamen, carpel and ovule development
Activator of vegetative to reproductive transition
Carpel, ovule and stamen development
DAL11, DAL12, DAL13
Petal and stamen specification
DAL23, DAL6, DAL22
Fruit, seed and ovule development
Carlsbecker et al. also notice that the aberrant ovuliferous organization in acrocona mutant cones is remarkably similar to some Paleozoic conifer genera, for example, Pseudovoltzia (compare Fig. 1b and d). What biological interpretation can be given to this observation? If we understand the mechanism regulating specific morphological changes in extant species and we observe similar changes in the fossil record, it is possible that the mechanism responsible for these evolutionary changes is also similar (Sanders et al., 2007). In the case of acrocona, this implies that the genetic mechanism involved in the transitional series of fossil ovuliferous organs might be similar to that responsible for the acrocona ovuliferous phenotypes. Carlsbecker et al. speculate that the evolutionary reduction of ovuliferous organs might involve the gradual expansion of DAL2, an AGAMOUS/SEEDSTICK-like gene, because of its function in meristem determinacy in Arabidopsis (Smaczniak et al., 2012). Yet given the differential expression of DAL14 in acrocona and the resemblance of its ovuliferous organs to those of Pseudovoltzia, an alternative interpretation involving DAL14-like genes could also be given. Rather than a DAL2-like gene, we speculate that expression changes of an ancestral DAL14-like might be responsible for the ovuliferous reductions observed throughout the fossil record. For instance, through a spatial expansion of DAL14 expression from a central area to more lateral and apical areas of the ovuliferous organs, reproductive organ determination might have accelerated and prematurely arrested the development of more shoot-like scales. Yet both interpretations, involving DAL2- or DAL14-like genes, are not necessarily mutually exclusive, as the gain of determinacy throughout the ovuliferous scale could comprise an interaction of these genes.
There is an important caveat, however. Although the ovuliferous organs of acrocona are indeed strikingly similar to Pseudovoltzia, this elegant evolutionary interpretation requires more molecular evidence in acrocona and rigourous genetic studies could strengthen or refute the aforementioned evolutionary hypotheses. In particular, functional characterization of the genes involved in reproductive organ development could be of major importance. This, though, is exactly where the problem lies, as functional research in gymnosperms remains limited without transgenic approaches. In this context, the early cone-setting phenotype of the acrocona mutant and the advent of the P. abies genome may speed up the development of an efficient transformation protocol in gymnosperms (Nystedt et al., 2013; Uddenberg et al., 2013). Until that time, well executed studies in naturally occurring mutants of gymnosperms, like these described in Carlsbecker et al., can provide us with those precious glimpses of the genetics of gymnosperm cone development.