**3. Discussion**

Seaweed cell-wall is made of different components namely cellulose, xylan, mannan, among others [19], although important constituents are alginates and fucans for brown algae, ulvans and other sulphated glycans for green algae, and agar and carrageenan for red algae [20]. The latter complex group of components are also sulphated polysaccharides which are recognised by their pharmacological and industrial uses [21]. In Rhodophyta, agar and carrageenan account for 40–50% of the dry weight, but can be as much as 70–80% [20]. Focusing on carrageenans, the backbone structure comprises linear chains of repeating D-galactose sugars and 3,6-anhydro-galactose units, with a different number and location of the sulphate groups attached. Sulphation degree and molecular conformation allow classifying carrageenans in three main types with commercial importance, namely κ-, ι- and λ-carrageenan, all containing 15–40% ester sulphate; although seaweeds can also contain hybrids carrageenans [22]. Sulphation takes place in the cell wall by the action of carbohydrate sulfotransferase whilst desulphation by galactose-6 sulfurylase also occurs in the cell wall, although its role is not fully clarified. Galactose-6 sulfurylase catalyses the conversion of μ-carrageenan into κ-carrageenan, but λ-carrageenan seems to not be susceptible to its action [16,23,24]. The number of sulphate groups is, in turn, one of the features that affects the properties of carrageenan types as carrageenans are gel-forming and viscosifying polysaccharides [25]. In addition, variations in carrageenan composition, i.e., sulphation degree, have been associated to life cycles stages of algae, alga species and environment conditions.

With the goal of revealing changes in sulphation degree of galactan of carrageenan during carposporogenesis in *G. imbricata*, expression levels of two genes involved in the biosynthesis of carrageenan has been reported. One of main achievements of this work lies on the fact that changes in transcript levels of these genes, i.e., *carbohydrate sulfotransferase* and *galactose-6-sulfurylase*, are associated at different stages of thalli development and the well-recognised stages of cystocarp maturation. Thus, expressions of *sulfotransferase* were reported in all stages of thalli development (Figure 4A). Results indicated that (1) the *ST1* and *ST2* could encode proteins responsible for similar mechanisms

and metabolic pathways during thalli development, and (2) *sulfotransferase* could be involved in housekeeping activity as sulphated galactan is component of cell wall of red seaweeds. Hence, transcript expression suggests that the addition of sulphate groups occurs during all stages of thalli development.

It is also worth drawing attention to the behaviour of both *sulfurylase* sequences (Figure 4B). The expression behaviour of *sulfurylase* (*SY1* and *SY2*) indicated that they may play several roles and that they are time-regulated. It was also evident that *SY1* was induced to a larger extent when the thalli were fertilised (Figure 4B). Noticeably, overexpression was higher in *sulfurylase SY1* than *sulfurylase SY2* for fertilised thalli, suggesting a concrete development-specific expression role for the *sulfurylase SY1* gene during thalli maturation. Apart from that, *sulfurylase SY2* expressions may also depend on the stage of the thalli, as differential transcript expression was also reported, albeit with different expression levels that were significantly lower than those for *sulfurylase SY1* (Figure 4B). The exhibition of these expression patterns for *SY1* and *SY2* in the red seaweed *G. imbricata* opens up an interesting pathway for determining whether the differential expressions of *SY1* and *SY2* could be a consequence of specific transcription factors of each gene copy. Interestingly, it could infer the presence of specific transcription factors, which would only function efficiently at specific stages of thalli maturation. Hence, the different stages of seaweed thalli development could contribute to the expression of different regulatory proteins. In higher plants, different proteins can participate as elements of regulatory mechanisms and co-ordinate cell activity [26]. Furthermore, if the transcription factors were activated and regulatory proteins were synthesised, this would reinforce the theory of a time-gene regulation of two sulfurylase as thalli development proceeds. All in all, genes that encode proteins responsible for the desulphation of the galactan skeleton of the cell wall will allow *G. imbricata* thalli to soften and enable the development of reproductive structures (cystocarps) in thalli.

Cystocarp disclosure is elicited by complex multiple factors, encompassing the period from when cystocarps first become visible (well-developed cystocarps), through to the adequate development of cystocarps (fully-developed cystocarps). Thus, cystocarps could achieve optimal maturity when elicitation factors and the mechanisms that weaken and soften the cell wall act in co-ordination [8]. In line with this, the behaviour of the genes that encode the adding and elimination of the sulphate group during maturation of reproductive structures must be appraised, along with the genes that encode oxidative-stress-related proteins, which also potentially contribute to softening the thallus.

Little has been reported to date about genes that control sulphate addition. This is one of the few studies aimed at characterising *sulfotransferase* expression levels at different stages of maturity in red seaweed cystocarps. Although monitoring the expressions of two types of *sulfotransferase* (*ST1* and *ST2*) showed similar behaviour and transcript expression levels, significant transcript expression was shown in fully-developed cystocarps of *G. imbricata* (Figure 5A,B). Our results seem to sugges<sup>t</sup> that genes encoding proteins that add the sulphate group are working to reconstitute the cell wall after disclosure and maturity of cystocarps and even early release of spores, as this is an unsynchronised process. Thus, it would not be misconceived to think that once fertilisation has occurred in *G. imbricata*, the re-arrangemen<sup>t</sup> of carrageenan constituents proceeds to improve and recover the organisation of the cell wall.

Remarkably, what occurs with *sulfurylase* differs from the expected. Following the line of argument, the thalli would soften prior to the appearance of cystocarps. Hence, one would expect transcript overexpression of *sulfurylase* to be reported when cystocarps are first developing (well-developed cystocarps) and always prior to their complete maturity (fully-developed cystocarps). However, *SY2* showed transcript expression is significantly higher in the presence of fully-developed cystocarps than in thalli with well-developed cystocarps (Figure 6B). Thus, different functionality of *SY1* and *SY2* can be assumed. To gain greater insight into what could be occurring, it is worth considering that fertilisation in *G. imbricata* takes place when a spermatium fertilises a carpogonium on the female gametophyte. The fertilised carpogonium develops into a structure called a cystocarp that will contain spores [27]. This cystocarp develops in the auxiliary ampulla after the auxiliary cell receives the diploid

nuclei from the fertilised carpogonial cell [28]. Thus, carrying the argumen<sup>t</sup> of *SY* di fferent functionality a step further, *SY1* may be responsible for weakening the cell wall to embrace the development of auxiliary ampulla cells, and *SY2* allows the cystocarps to grow in size and develop their spores within. These di fferentiated functions should show a co-ordination of the di fferent kinds of *sulfurylase* over time and cell type during the cystocarp maturation. In short, this time–transcript expression could orchestrate specific adaptation and protective responses during cystocarps development. This would mean that *sulfurylase* genes are closely related to reproductive development and that maturity stages of cystocarps require genes expressed in di fferent ways from early stages.

Beyond the sulphation and de-sulphation of galactan backbone as a consequence of thalli development and cystocarp maturation, the gene encoding *WD40* has also been reported to potentially play a role in cystocarp maturing, as the gene transcripts were significantly overexpressed in fully-developed cystocarps, compared with expression in developing cystocarps (Figure 7). To some extent, these results are to be expected as earlier studies on gene expression during carposporogenesis of *G. imbricata* showed *WD40* and *cytochrome P450* transcript expressions are limited to di fferent signals related to both cystocarp disclosure and development of the red seaweed *G. imbricata* [9,10]. Specifically, *WD40* plays an active role in the processes of regulation and response to damage [29], so *WD40* gene expression in *G. imbricata* could sugges<sup>t</sup> cross talk between cystocarp maturation and a reduction of oxidative damage.

In summary, gene expressions involved in the sulphation and desulphation of galactan backbone are associated with alterations in thalli development and cystocarp maturation in the red seaweed *G. imbricata*. This opens an interesting framework to gain insight into gene mechanisms involved in carrageenan synthesis.
