The Hox-genes and the larval development

The boundaries between polychaete’s larval and postlarval ontogenesis not always are obvious. There is relatively clear trochophore stage – an unsegmented ciliated larva. Such type of larva is called primary. These larvae consist of few cells, which specialize early and belong not to many cell types (Peterson et al., 1997). Trochophore larvae of studied polychaetes (Chaetopterus, Platynereis, Alitta) don’t use Hox-cluster for body patterning. The similar situation has been described for sea urchin and for nemerteans, which activate Hox-cluster axial transcription only during metamorphosis and only in presumptive tissues of developing definitive body (Arenas-Mena et al., 2000; Hiebert and Maslakova, 2015). Eric H. Davidson supposes (Davidson, 2001), that regulation programs of the development for primary Bilaterian larvae may emerge at that period when Hox-cluster yet included few genes and these genes didn’t take part in regionalization, but took part in tissue-specific differentiation.

Besides primary larva, there is secondary segmented larva with an organization that is similar to definitive one. Secondary larvae of all studied polychaetes use Hox-cluster for the regionalization of body parts. Both Capitella and nereid polychaetes colinearly turn on Hox-genes at the segmented larva stage. However larval and postlarval ontogenesis of these phylogenetically distant polychaetes isn’t directly relevant. The Capitella hasn’t a trochophore and its secondary larva consists of many segments from two growth zones (Seaver et al., 2005). A part of these segments forms almost simultaneously from lateral growth zone, and other part derives from subterminal growth zone and forms consecutively. The nereid’s secondary larva (the nectochaete) consists of few segments. All of them derive from ventral somatic plate.

During the comparison of Hox-genes expression patterns of juvenile Alitta and late Capitella ’s larva, an unexpected similarity is revealed. In both cases genes transcribe by wide gradients. Much of these genes mark a growth zone. There is visible parallel between ortholog genes expression patterns (Fig.2bc). There is an impression about heterochronic shift, which had moved a nereid-like definitive development program to the larval period of Capitella development. In this case a postlarval Capitella ’s program looks like an independent add-on, which is absent in nereids. There is other possiblility: a definitive stage was closer to Capitella ’s postlarval stage. In that case Hox-genes took part in single neural ganglions specification, and the time of their wider latitude was shifted to larval stage. If so, then nereid’s ancestors had excluded a definitive stage and had shifted breeding time at the earlier stage. This version is supported by the fact that in Rotifera and Chaetognatha (basal Spiralia) Hox-genes involved almost exclusively in linear neuron specification (Papillon et al., 2005; Frobius and Funch, 2017). In this light a Capitella ’s complex secondary larva and a juvenile nereid are homologous life cycle stages, which had intercalated between primary larva and definitive stage, and in nereids a definitive stage disappeared during the evolution.

We can see at the examples of studied polychaetes, that same animal during its ontogenesis may use different programs of development: an embryonic program for primary larva formation, a larval one – for secondary larva’s generation, and a postlarval program – for definitive body formation. All above-mentioned polychaetes ontogenesis features point at high modularity of their development programs. Polychaete’s ontogenetic modules consist of Gene Regulatory Networks (GRNs), which interact weakly and can evolve independently. In fact, they are separate epigenomes, packed into common genome. Unlike epigenomes from different cell lines (for example those which work in mammal’s fibroblasts), they have complete hierarchic structure, which isn’t chained by epigenetic limitations. This exactly explains wonderful somatic embryogenesis of polychaetes. In polychaetes, heterochronies and heterotopies of development program involve all levels of GRNs hierarchy and therefore can shift whole ontogenetic modules. If these shifts occur in the space, then that is a good prerequisite for somatic embryogenesis. If shift occurs in time, then there become possible an appearance of new ontogenesis stage or loss old one.

The vast morphological variety of Spiralia is a direct sequence of this multi-level ontogenesis, which is evolutionary flexible at each level. The marks of Hox-cluster genes structural and functional evolution are well-tracked and the wider an array of studied objects the truer our ideas about evolution’s ways.

* Corresponding Author address

Email: nereisvi@gmail.com