After carbon, oxygen, and nitrogen, sulfur is the most abundant element in living organisms. Sulfur participates in the composition of cysteine and methionine, which are key amino acids for protein folding (e.g., disulfide bonds) and essential for the reactivity of many enzymes as catalytic amino acids or complexing cofactors (e.g., iron-sulfur cofactors). Sulfur is also found in vitamins, steroids, and many carbohydrates including polysaccharides, proteoglycans, and glycolipids. Sulfur is assimilated by living organisms in the form of inorganic sulfate ions, which are the second most abundant anions in seawater. The concentration of sulfate ions in the oceans ranges from 25 to 28 mM; these values are very high compared with those found in freshwater and soil (10–50 μM) (Wright and Colling, 1995; Friedlander, 2001; Bochenek et al., 2013).

The abundance of sulfate ions and its ubiquity has resulted in a wide distribution of sulfated molecules and more especially, sulfated polysaccharides in marine organisms. In contrast to land polysaccharides, a huge number marine polysaccharides are decorated with sulfate ester groups. Except for sulfated glycosaminoglycans such as chondroitin sulfate or herapan sulfate found in the matrix of animal cells, anionic polysaccharides found in terrestrial organisms carry uronic groups (i.e., pectins, hyaluronic acid, polysialic acids). The occurrence of sulfated polysaccharides in macroalgae, marine angiosperms, and seagrasses has been proposed to be an adaptation to marine environments (Aquino et al., 2005; Olsen et al., 2016), suggesting that the sulfation of polysaccharides is a marker of marine origins. Nevertheless, in all cases, the biosynthesis of sulfated polysaccharides requires the activation of sulfate in PAPS (3′-phosphoadenosine-5′-phosphosulfate), catalyzed by PAPS synthase, and then the grafting of a sulfate ester group onto the carbohydrate via a specific sulfotransferase (Klaassen and Boles, 1997; Zhang et al., 1998).

The most studied sulfated marine polysaccharides are those extracted from the cell walls of macroalgae due to the industrial interest in their hydrocolloid properties. Agarans and carrageenans are the most exploited polysaccharides from red algae and the wide diversity of red algae polysaccharides has been reviewed recently (Usov, 2011). Similarly, the sulfated polysaccharides extracted from brown and green algae also show high diversity. Several reviews have highlighted their structural complexity and their potential applications (see, for example, Kloareg and Quatrano, 1988; Berteau and Mulloy, 2003; Lahaye and Robic, 2007; Pomin and Mourão, 2008). The diversity of sulfated polysaccharides biosynthesized by marine microorganisms, such as bacteria and microalgae, has been less explored so far. However, sulfate groups were found to decorate the carbohydrate backbone of several secreted polysaccharides and matrix lipo-polysaccharides biosynthesized by marine bacteria (Mancuso Nichols et al., 2005; Nazarenko et al., 2011). Composition analyses attested also the structural diversity of sulfated polysaccharides biosynthesized in microalgae but only a few polysaccharide structures have been resolved (Hoagland et al., 1993; Gügi et al., 2015). For example, studies and structural analyses of the main cell-wall polysaccharide in diatoms—a group of ecologically important marine organisms—suggest that it is a sulfated glucuronmannan (Percival and McDowell, 1967; Willis et al., 2013).

Marine polysaccharides make up a large part of algal biomass and are a food source for heterotrophic organisms. This reservoir of sulfated polysaccharides are produced by concomitant fixation of sulfur and carbon by photosynthetic organisms. Enzymatic degradation of complex sulfated polysaccharides requires a set of enzymes that can cleave the glycosidic bond and remove any decorations, such as sulfate groups, from the carbohydrate backbone. Specific glycoside hydrolases or polysaccharide lyases depolymerize marine polysaccharides break the glycosidic bonds via hydrolysis and β-elimination mechanisms, respectively. Concerted action with sulfatases can result in the production of neutral mono- and oligosaccharides then used for energy consumption.

The avalanche of gene sequences produced during the numerous marine genomics and metagenomics programs have revealed the important role of sulfate in marine biology (Glöckner et al., 2003; Teeling et al., 2012). Genomes of polysaccharide-degrading bacteria reveal the co-occurrence of glycoside hydrolases and polysaccharide lyases with sulfatases. More interestingly, in the bacterial genus Bacteroidetes known to include many polysaccharide-degrading strains, all the genes coding for polysaccharide-degrading enzymes leading to the saccharification of sulfated polysaccharides are clustered in the genome in so-called “polysaccharide utilization loci” (PUL) (Sonnenburg et al., 2006; Flint et al., 2008; White et al., 2014). More recently, the discovery of marine polysaccharide-degrading enzymes in Bacteroidetes strains found in the microbiota of the human gut—likely acquired by gene transfer from marine organisms associated with seafood (Hehemann et al., 2010)—suggest that the investigation of the enzymatic degradation of marine polysaccharides goes beyond the ecological understanding of the ocean cycle, and also involves human health. Likewise, genomic analyses of human microbiota have revealed numerous PULs with sulfatase genes involved in the degradation of many sulfated polysaccharides, probably of marine origin.

Therefore, the purpose of this review is to summarize the work dealing with marine polysaccharide sulfatases starting with a short introduction to sulfatase classification, followed by a highlight on the diversity of marine polysaccharides sulfatases revealed by genomic data mining and then a description of the marine polysaccharide sulfatases that have been biochemically characterized to date. Glycosaminoglycan sulfatases will not be considered here, although several chondroitin or heparin sulfatases have been identified in marine prokaryotes (Han et al., 2014; Wang et al., 2015a).

Sulfatases are grouped into four classes based on sequence homology, crystallographic structure and mechanisms. Type I sulfatase—or formylglycine-dependent sulfatase—encompasses the vast majority of the known sulfatases and to date, all the biochemically characterized carbohydrate sulfatases (Hanson et al., 2004). These sulfatases require post-translational conversion of a cysteine or a serine into a formylglycine amino acid residue essential for catalysis. The oxidation of amino acids is catalyzed by a formylglycine-generating enzyme (FGE), also named sulfatase-modifying factor 1 (SUMF1), which recognizes the consensus amino acid sequence C/S-X-P-X-R (Cosma et al., 2003; Dierks et al., 2003; Sardiello et al., 2005; Bojarová and Williams, 2008).

In anaerobic bacteria, cysteine and serine residues can be modified by another family of enzymes called anaerobic sulfatase-maturating enzymes (anSMEs) (Berteau et al., 2006). Crystallographic structure of FGE (Bond et al., 1997; von Bülow et al., 2001) and anSME (Goldman et al., 2013) demonstrate that their mechanisms differ in the use or non-use of oxygen molecules, respectively. A recent review covers the state of knowledge on these enzymes as well as their potential biotechnological applications (Appel and Bertozzi, 2015).

Based on inhibition experiments and the analysis of the crystal structure of sulfatase complexes, two mechanisms of desulfation have been proposed, differing essentially in the hydration state of the formylglycine residue at the beginning of the reaction (Bond et al., 1997; Lukatela et al., 1998). However, both mechanisms generally agree that the last steps of the reaction involve a covalent bond between the sulfate anion and the formylglycine amino acid, and the regeneration of formylglycine by hydrolysis (Figure 1). The reaction results in the retention of the carbon configuration, the chirality of the carbon center being unaffected by the desulfation catalysis (Williams et al., 2014; Figure 2).

Type II sulfatase encompasses Fe(II) α-ketoglutarate-dependent alkylsulfatases, which belong to the dioxygenase superfamily (Müller et al., 2004). The catalysis requires molecular oxygen (O₂) and α-keto-glutarate as a co-substrate and leads to the oxidation of the primary sulfate ester group in the aldehyde and concomitantly to the oxidative decarboxylation of the α-keto-glutarate into succinate. Type III sulfatases are represented by enzymes related to Zn²⁺- or Mn²⁺-dependent metallo-β-lactamases. The cleavage of the C-OS bond occurs via a nucleophilic attack of the carbon by a molecule of water activated by the binuclear metal ion complex located in the active site (Hagelueken et al., 2006). The loss of the sulfate group results in the inversion of the configuration of the carbon (Figure 2). No carbohydrate sulfatases fall in the type II or type III sulfatase categories.

Two other sets of enzymes represent another potential sulfatase families. One sulfatase belonging to the superfamily of amidohydrolases has been purified from the marine bacteria Pseudoalteromonas carrageenovora; the recombinantly expressed enzyme is active on a synthetic substrate (i.e., methylemberliferyl sulfate), but showed less activity than the native purified enzyme (Genicot et al., 2014). The galactose-6-sulfurylase (discussed below in Section Predicted marine polysaccharide sulfatases) catalyzes the formation of anhydrogalactose concomitantly to the release of the sulfate group (Rees, 1961a,b).

The massive compilation of gene sequences through the numerous genome and metagenome sequencing programs has led to an exponential increase in the number of putative proteins whose functions have, for the most part, not yet been determined (Hanson et al., 2010). Despite the development of bioinformatics tools, prediction of protein function requires experimental validation, including a comprehensive overview of metabolic pathways and functional characterization of the proteins. Like other classes of proteins, sulfatases have seen the number of predicted genes increase exponentially in databases, particularly owing to the massive sequencing of marine organisms. The precise function of the sulfatases, such as the nature of their sulfated substrates (carbohydrate, lipids, metabolites, etc.), cannot be predicted based on sequence comparison with biochemically characterized sulfatases due to the lack of data. Based on the surge in sulfatase gene sequences and their genomic environment, probable function and preferred substrates of carbohydrate sulfatases can be suspected.

Sequencing of marine life has shed light on the diversity of carbohydrate sulfatases. The modularity of sulfatases and the genome organization of Bacteroidetes (i.e., PUL organization) have the potential to drive future functional characterization of the large amount of sulfatase sequences. Ultimately, the comprehensive classification of sulfatases based on sequence homology, allowing prediction of function is a challenging task for future work, calling for bioinformatics analyses (Wegner et al., 2013; Barbeyron et al., 2016) and crystallographic description of catalytic sites to identify the key amino acids involved in substrate recognition.

Functional characterization can benefit from genetic tools, such as gene mutation, to inactivate encoded proteins, and from more classic approaches, such as biochemical analyses of the enzymes purified from the organism in which they have been detected or heterologous expression in E. coli. However, only very few carbohydrate sulfatases have been biochemically characterized, underlining the difficulties in preparing them to purity. Recombinant marine bacterial sulfatases have been obtained, but always with a low level of activity. This low activity can be explained by the low rate of post-translational maturation of sulfatases in classic proteins expression systems that are naturally unable to convert cysteine and serine into formylglycine amino acid. Co-expression of sulfatases with known maturating enzymes (i.e., FGE, AnSME) can result in the production of a low amount of active enzymes, but better knowledge of the sulfatase maturating system—which probably involves several yet-to-be-discovered proteins—is a bottleneck to the production of active recombinant sulfatases.

The diversity of sulfatases lies in the diversity of sulfated polysaccharides present in the oceans. The composition and structure are simply unknown for most of them. Therefore, a comprehensive analysis of sulfatase function requires, in parallel, novel data on marine polysaccharides. Macro- and micro-algae, the primary producers of the oceans, harbor an unexplored and massive source of sulfated polysaccharides that contain both carbon and inorganic sulfate ions. Deciphering the turnover of marine polysaccharides involving sulfatases and other enzyme partners, will enhance our molecular understanding of the biogeochemistry of the ocean.

The author confirms being the sole contributor of this work and approved it for publication.