Sterol Sulfates and Sulfotransferases in Marine Diatoms

Article abstract of DOI:10.1016/bs.mie.2018.03.003

Sterol sulfates are widely occurring molecules in marine organisms. Their importance has been so far underestimated although many of these compounds are crucial mediators of physiological and ecological functions in other organisms. Biosynthesis of sterol sulfates is controlled by cytosolic sulfotransferases (SULTs), a varied family of enzymes that catalyze the transfer of a sulfo residue (–SO₃H) from the universal donor 3′-phosphoadenosine-5′-phosphosulfate to the hydroxyl function at C-3 of the steroid skeleton. The absence of molecular tools has been the main impediment to the development of a biosynthetic study of this class of compounds in marine organisms. In fact, there is very limited information about these enzymes in marine environments. SULT activity has, however, been reported in several marine species, and, recently, the production of sterol sulfates has been linked to the control of growth in marine diatoms. In this chapter, we describe methods for the study of sterol sulfates in this lineage of marine microalgae. The main aim is to provide the tools useful to deal with the biosynthesis and regulation of these compounds and to circumvent the bottleneck of the lack of molecular information. The protocols have been designed for marine diatoms, but most of the procedures can be used for other marine organisms.

Full text of DOI:10.1016/bs.mie.2018.03.003

CHAPTER FIVE
Sterol Sulfates and
Sulfotransferases in
Marine Diatoms
Carmela Gallo, Genoveffa Nuzzo, Giuliana d’Ippolito1,
Emiliano Manzo, Angela Sardo, Angelo Fontana1
National Research Council of Italy, Institute of Biomolecular Chemistry, Pozzuoli, Naples, Italy
1Corresponding authors: e-mail address: gdippolito@icb.cnr.it; afontana@icb.cnr.it
Contents
1. Introduction
2. Sulfotransferases
3. SULTs and Sterol Sulfates in Marine Diatoms
4. Methods
4.1 Isolation and Purification of Sterol Sulfates
4.2 Identification of Sterol Sulfates
4.3 UPLC-MS Analysis of Sterol Sulfates
4.4 Inhibition of Sterol Sulfate Biosynthesis in Marine Diatoms
4.5 13C Labeling of the Carbon Skeleton of Diatom Sterols
5. Conclusions
102
103
107
111
112
118
120
122
127
129
130
References
Abstract
Sterol sulfates are widely occurring molecules in marine organisms. Their importance
has been so far underestimated although many of these compounds are crucial medi-
ators of physiological and ecological functions in other organisms. Biosynthesis of sterol
sulfates is controlled by cytosolic sulfotransferases (SULTs), a varied family of enzymes
that catalyze the transfer of a sulfo residue (ꢀSO3H) from the universal donor 30-phos-
phoadenosine-50-phosphosulfate to the hydroxyl function at C-3 of the steroid skeleton.
The absence of molecular tools has been the main impediment to the development of a
biosynthetic study of this class of compounds in marine organisms. In fact, there is very
limited information about these enzymes in marine environments. SULT activity has,
however, been reported in several marine species, and, recently, the production of sterol
sulfates has been linked to the control of growth in marine diatoms. In this chapter, we
describe methods for the study of sterol sulfates in this lineage of marine microalgae.
The main aim is to provide the tools useful to deal with the biosynthesis and regulation
of these compounds and to circumvent the bottleneck of the lack of molecular infor-
mation. The protocols have been designed for marine diatoms, but most of the proce-
dures can be used for other marine organisms.
#
Methods in Enzymology, Volume 605
ISSN 0076-6879
https://doi.org/10.1016/bs.mie.2018.03.003
2018 Elsevier Inc.
All rights reserved.
101

102
Carmela Gallo et al.
1. INTRODUCTION
Sulfation (also found as sulfurylation) is an essential posttranslational
modification of many molecules including proteins, carbohydrates, and lipids.
The reaction is also one of the major conjugating pathways responsible for the
detoxification and subsequent elimination of xenobiotics and endogenous
small molecules in eukaryotes. The modification of metabolites by the addi-
tion of a sulfo group can have a profound influence on the biological prop-
erties of the products. In general, sulfation facilitates the water solubility of
acceptor substrates, thereby leading to inactivation and excretion of biologi-
cally active compounds from the body. Thus, sulfation plays an important role
in the metabolism and regulation of key endogenous compounds and the
detoxification of xenobiotics. However, this is not the only biological func-
tion of this reaction, and there are many molecules that, when sulfated, either
retain or acquire other biological activities. In some cases, the presence of
the sulfate group on some molecules can also be a prerequisite for their
biological function. For example, pregnenolone sulfate is one of the most
biologically important neurosteroids in mammals, and its decrease is associated
with certain brain diseases (Korinek et al., 2011; Zheng, 2009). In humans,
sulfation provides an excellent avenue for the homeostatic control of hydro-
phobic signal molecules such as oxysterols and steroids (Javitt, Lee, Shimizu,
Fuda, & Strott, 2001; Ren et al., 2014; Strott & Higashi, 2003; Zheng, 2009),
hormones (Visser, van Buuren, Rutgers, Eelkman Rooda, & de Herder, 1990),
peptides (Farzan et al., 1999), and catecholamines (Coughtrie, Taskinen, &
Hood, 2002).
The roles of sulfated compounds in organisms other than mammals are
less clear. In bacteria, different studies suggest that sulfation of specific sub-
strates is important in ecological and pathological interactions (Longo,
Motta, Mauri, Landini, & Rossi, 2016). A historical example is the establish-
ment of the specific symbiosis between the nitrogen-fixing bacterium
Sinorhizobium meliloti and the plant Medicago sativa by sulfation of the
lipooligosaccharide signals named Nod factors (Lerouge et al., 1990; Roche
et al., 1991). S. meliloti mutants lacking the sulfation pathway are unable to
induce root nodulation in alfalfa but gain the ability to colonize the roots
of vetch (Ehrhardt et al., 1995). In plants, several sulfated brassinosteroids have
been related to the response to pathogens and synthesis of pathogen-related
signals methyl jasmonate and salicylic acid (Jez, Ravilious, & Herrmann,
2016; Marsolais et al., 2007). It has been also reported that sulfation inactivates

Sterol Sulfates and Sulfotransferases
103
the biological activity of 24-epibrassinolide or flavonoids, suggesting that
plants, like animals, use the addition of the sulfo moiety to regulate the pools
of these secondary metabolites (Hashiguchi et al., 2013, 2014; Rouleau
et al., 1999).
Since the isolation of cholesterol sulfate from the starfish Asterias rubens
(Bjo€rkman, Karlsson, & Nilsson, 1972; Godfellow & Goad, 1973), several
sulfated sterols have been also described from a wide variety of marine
organisms, particularly sponges (Aiello, Fattorusso, & Menna, 1999) and
echinoderms (D’Auria, Fontana, Minale, & Riccio, 1990). Several of
these molecules have a broad spectrum of biological activities and potential
therapeutic applications. Of particular note has been their activity against
HIV-1 (Tziveleka, Vagias, & Roussis, 2003) and pathogenic fungi
(Boonlarppradab & Faulkner, 2007; DiGirolamo, Li, Jacob, Clark, &
Ferreira, 2009). The latest reports include modulation of receptors or
regulatory proteins such as sirtuins (Nakamura et al., 2018), protein kinase
Cζ (Whitson et al., 2009, 2008), pregnane-X-receptor (Festa et al., 2011),
and farnesoid-X-receptor (Sepe et al., 2011).
2. SULFOTRANSFERASES
Sulfation is catalyzed by a broad family of enzymes named
sulfotransferases that transfer a sulfo group (ꢀSO3H) or its anion form from
the biochemical donor 30-phosphoadenosine-50-phosphosulfate (PAPS) to
hydroxy or amine moieties of various substrates (Fig. 1; Chapman, Best,
Fig. 1 Sulfation reaction of sterol.

104
Carmela Gallo et al.
Hanson, & Wong, 2004; Cook et al., 2016; James, 2014; Klaassen &
Boles, 1997; Negishi et al., 2001). The reaction with small acceptor
molecules is mostly described in eukaryotic organisms (Hattori, Inoue,
Inoue, Arai, & Tamura, 2006; Herna´ndez-Sebastià, Varin, & Marsolais,
2008; Hirschmann, Krause, & Papenbrock, 2014; James, 2011; Langford,
Hurrion, & Dawson, 2017; Liu, Yang, Sugahara, Yasuda, & Liu, 2005;
Venkatachalam, Llanos, Karami, & Malinovskii, 2004) even if PAPS-
dependent transfer has been also reported in prokaryotes, including the
previous cited example of Nod and the direct role in polyketide biosynthesis
(Ehrhardt et al., 1995; Funabashi et al., 2010; Gehret et al., 2011; Gu et al.,
2009; Han et al., 2012; McCarthy et al., 2012; Purchartova´ et al., 2015;
Tang et al., 2013). In bacteria, a PAPS-independent mechanism of sulfation
has been also established (Malojciꢀc & Glockshuber, 2010; Malojcˇiꢀc,
Owen, & Glockshuber, 2014; Malojciꢀc et al., 2008).
Structure, dynamics, and systematic of sulfotransferases have been well
detailed only in vertebrates and most of our knowledge of this family of
enzymes derive by investigations in humans where sulfation affects criti-
cally physiological and immunological processes (Cook et al., 2013, 2016;
Cook, Wang, & Leyh, 2015; Coughtrie, 2016; Gamage et al., 2006;
Leyh, Cook, & Wang, 2013; Negishi et al., 2001; Rath, Verdugo, &
Hemmerich, 2004; Runge-Morris, Kocarek, & Falany, 2013; Tibbs,
Rohn-Glowacki, Crittenden, Guidry, & Falany, 2015).
The family is divided in cytosolic and membrane-associated proteins
although this organization can be misleading since in most cases the subcel-
lular localization of theses enzymes is unknown. In fact, the term cytos-
olic refers to the ability of the enzymes to be extracted from cells in a
soluble form. Membrane-associated sulfotransferases add sulfo groups to
large biomolecules such as complex polysaccharides and proteins, whereas
the cytosolic forms accept as substrate small organic molecules including
sterols, flavonoids, glucosinolates, phenols, and catecholamines (Chapman
et al., 2004). According to the HUGO Gene Nomenclature Committee
(http://www.gene.ucl.ac.uk/nomenclature/), SULT is used as the abbrevia-
tion for the cytosolic sulfotransferase superfamily and the different isoforms
are classified according to the amino acid sequence (Allali-Hassani et al.,
2007; Blanchard, Freimuth, Buck, Weinshilboum, & Coughtrie, 2004;
Coughtrie, 2012, 2016). SULTs sharing greater than 45% sequence
identity are grouped into a single family denoted by an Arabic numeral
(e.g., SULT1), while subfamily members have a minimum of 60% sequence
identity are designed by alphabetic categories (i.e., SULT1A, SULT1B, etc.).

Sterol Sulfates and Sulfotransferases
105
SULT isoforms sharing higher amino acid sequence identity are delineated by
a second number after the subfamily letter (e.g., SULT1C3). All gene names
follow the enzyme name except that the gene descriptor is in italics (e.g.,
Sult1a1) according to widely accepted conventions. In vertebrates, there
are six distinct SULT families (although not all species express members of
each family), with at least an additional five related families found in inverte-
brates and bacteria (SULTs are particularly abundant in cyanobacteria) and in
plants.
The human SULTs, that include 13 different isoforms belonging to four
families (Almazroo, Miah, & Venkataramanan, 2017; Coughtrie, 2016),
have distinguishing biochemical features and substrate specificity despite
their high level of conservation. Individual isoforms of the same family tend
to show similar substrate preferences. The SULT2 family is primarily
involved in the sulfoconjugation of sterols and alcohols and embraces two
subfamilies, SULT2A1 and SULT2B1. The latter one is additionally divided
into two isoforms, SULT2B1a and SULT2B1b, with overlapping substrate
specificity on the prototypic hydroxysteroid substrate, dehydroepiandroster-
one, but high selectivity for pregnenolone or cholesterol, respectively (Fuda,
Lee, Shimizu, Javitt, & Strott, 2002; Her et al., 1998).
SULTs show highly conserved regions (called I, II, III, IV) (Fig. 2) that
have been identified through sequence alignments of SULTs from plants,
animals, and bacteria (Allali-Hassani et al., 2007; Chapman et al., 2004;
Herna´ndez-Sebastià et al., 2008; Hirschmann et al., 2014; Marsolais &
Varin, 1995, 1998). The N-terminus contains Region I that interacts with
the 50-phosphate of PAPS (PSB loop). Region II is also essential for binding
Fig. 2 Representation of conserved Regions I–IV of SULTs. The binding regions of
PAPs—50-phosphosulfate-binding site (PSB loop), the 30-phosphate-binding site (PB
loop), and the GxxGxxK motif—are labeled. This figure was adapted from research orig-
inally published by Papenbrock and coworkers (Hirschmann, F., Krause, F., & Papenbrock, J.
(2014). The multi-protein family of sulfotransferases in plants: Composition, occurrence,
substrate specificity, and functions. Frontiers in Plant Science, 5(October), 1–13. https://
doi.org/10.3389/fpls.2014.00556).

106
Carmela Gallo et al.
of the cofactor and starts with His108 that is responsible for proton accep-
tance during the sulfuryl transfer (Kakuta, Pedersen, Carter, Negishi, &
Pedersen, 1997). In human SULT2A1, conformational changes in two
cysteine residues affect the binding of both the cofactor and hydroxysteroid
substrate (Qin, Teesch, & Duffel, 2013; Rehse, Zhou, & Lin, 2002a,
2002b). This Region II also contains the residues Arg130 and Ser138 that
are shared in many SULTs and are involved in hydrogen bonding with
the cofactor (Hirschmann et al., 2014; Shimizu, Fuda, Yanai, & Strott,
2003). Three amino acids located in Region IV bind the 30-phosphate of
PAPS and form the so-called PB loop. The motif GxxGxxK located after
the PB loop is essential for the binding of PAPS and estradiol in the estrogen
SULT from mammals (Driscoll, Komatsu, & Strott, 1995).
Although much is known of the way in which sulfation regulates metab-
olism, a clear consensus has not been reached on the biology of transfer of
the sulfo group and the mechanistic aspects of the reaction. Sulfation is pro-
posed to involve an SN2 nucleophilic attack by the alcohol on the sulfate of
PAPS (Kakuta, Petrotchenko, Pedersen, & Negishi, 1998; Negishi et al.,
2001) but more recent investigations of human SULT2A1, according to
partial evidence from other SULTs, delineate a process involving several
transient intermediates with the release of the nucleotide as the rate-limiting
step (Wang, Cook, Falany, & Leyh, 2014). Furthermore, consistent with the
structural data, molecular dynamics simulations predict that human
SULT1A1 and SULT2A1 use a gating mechanism to select substrates: in
silico acceptor-binding pockets of these enzymes open and close in response
to the binding of the cofactor (Cook et al., 2013).
In plants, there is no general agreement on the classification of SULTs on
the basis of sequence identity/similarity or substrate specificity (Hirschmann
et al., 2014; Klein & Papenbrock, 2004; Koprivova & Kopriva, 2016). In
2014 Hirschmann et al. reported more than five hundred putative plant
sulfotransferase sequences in the Pfam database (Hirschmann et al.,
2014). Among these, 459 have a Sulfotransfer_1 domain (PF00685), 49 a
Sulfotransfer_2 domain (PF03567), and 16 a Sulfotransfer_3 domain
(PF13469). To date 22 putative SULTs have been identified in Arabidopsis
thaliana, and 35 genes coding proteins with a sulfotransferase domain
are reported in Oryza sativa (Chen et al., 2012). Several phenol SULTs have
been also described in Asteraceae (Marsolais & Varin, 1998), and a
brassinosteroid SULT has been characterized in Brassica sp. (Rouleau
et al., 1999).

Sterol Sulfates and Sulfotransferases
107
3. SULTS AND STEROL SULFATES IN MARINE DIATOMS
Hundreds of sulfated compounds have been described in marine spon-
ges and echinoderms, but very little is known about their biosynthesis
(Kornprobst, Sallenave, & Barnathan, 1998). Sulfotransferase activity has
been also described in cnidarians (Cormier, Hori, & Karkhanis, 1970) and
sea urchins (Malins & Roubal, 1982), and sulfation has been reported in
algae, microalgae, and other phototrophic organisms (Herna´ndez-Sebastià
et al., 2008). Ho et al. found 84 sulfotransferases of the superfamily 1 and
2 by mining the genomes of green microalgae and diatoms (Ho, 2015).
However, like for the results of the transcriptome profiling of Gracilaria
changii and Gracilaria salicornia (Rhodophyta) (Ho, 2015), most of these
enzymes are related to carbohydrate metabolism. Four genes coding for
putative SULTs have been very recently identified by transcriptome analysis
in the diatom Skeletonema marinoi (Gallo, D’Ippolito, Nuzzo, Sardo, &
Fontana, 2017). The amino acid sequence for the highly conserved regions
(Table 1) underlines the phylogenetic relationship of these proteins with
SULTs from different groups of organisms (other diatoms, cyanobacteria,
green algae, plants, and mammalians).
As shown in Fig. 3, at least one of the sequences of S. marinoi codes for a
putative protein with 30% similarity with human SULT2B1, SULT2A1,
and SULT1C that highly prefer cholesterol and other 3β-hydroxysteroids
as substrates. Despite the large number of putative SULTs sequences in
microalgae, none have been isolated and characterized to date.
Diatoms are unicellular eukaryotic phytoplankton responsible for
nearly one-half of total marine primary productivity (Allen et al., 2005;
Falkowski, Barber, & Smetacek, 1998; Smetacek, 2000). Evolutionary
and ecological success of these aquatic protists depends on mechanisms to
monitor and adapt to changes in environmental conditions (Armbrust,
2009; Falciatore, Ribera D’Alcalà, Croot, & Bowler, 2000; Vardi et al.,
2008). Determining the factors that regulate the succession of diatom
populations has important implications in the study of energy flow and nutri-
ent cycling in marine ecosystems. In addition to sedimentation and grazing
by herbivores, programmed cell death is also considered one of the major
causes for control of algal blooms (Bidle, 2016; Bidle & Falkowski, 2004;
Brownlee, 2008; Vardi et al., 2009). Recently a few authors have also put
forward that these mechanisms of death confer a competitive advantage

108
Carmela Gallo et al.
Table 1 Reference Sequence Number of Proteins Used for Phylogenetic Analysis
Organism
Protein Reference Number
Ancylostoma duodenale
Arabidopsis lyrata
KIH65753.1
XP_002890763.1
OAP14364.1 SOT7
XP_013723193.1
XP_010425175.1
XP_006293532.1
KOO25185.1
Arabidopsis thaliana
Brassica rapa
Camelina sativa
Capsella rubella
Chrysochromulina sp.
Crocosphaera watsonii
Cyanothece sp. ATCC51142
Cyanothece sp. CCY0110
Cyanothece sp. PCC7424
Cyanothece sp. PCC8801
Daucus carota
WP_007308169.1
ACB53617.1
WP_008278423.1
WP_012598684.1
ACK66184.1
XP_017233697.1
XP_005757376.1
OEU18550.1
Emiliana huxleyi
Fragilariopsis cylindrus (FRAGPS-1)
Homo sapiens 1B1
NP_055280.2
Homo sapiens 1B2
BAA24547.1
Homo sapience 1C
AAC95519.1
Homo sapiens 1E
AAH27956.1
Homo sapiens 2B1A
Homo sapiens 2B1B
Ixodes scapularis
1Q1Z_A
NP_004596.2
XP_002409755.1
XP_013794057.1
XP_002509199.1
EWM21324.1
Limulus polyphemus
Micromonas commoda
Nannochloropsis gaditana
Necator americanus
XP_013308151.1
XP_002178901.1
Phaeodactylum tricornutum (PHATR-1)

Sterol Sulfates and Sulfotransferases
109
Table 1 Reference Sequence Number of Proteins Used for Phylogenetic Analysis—
cont’d
Organism
Protein Reference Number
XP_002178554.1
XP_002177379.1
XP_002179849.1
P49890.1
Phaeodactylum tricornutum (PHATR-2)
Phaeodactylum tricornutum (PHATR-3)
Phaeodactylum tricornutum (PHATR-4)
Rattus sp.
Skeletonema marinoi Fe7-1 (SMFE7-1)
MMETSP1039-
20121108j3029_1
Skeletonema marinoi Fe7-2 (SMFE7-2)
Skeletonema marinoi Fe7-3 (SMFE7-3)
Thalassiosira oceanica (THAOC-1)
Thalassiosira oceanica (THAOC-2)
MMETSP1039-20121108j3249
MMETSP1039-20121108j14798
EJK47508.1
EJK45887.1
Thalassiosira pseudonana-1
(THAPSDRAFT-1)
XP_002287189.1
Thalassiosira pseudonana-2
(THAPSDRAFT-2)
XP_002288711.1
XP_002295934.1
XP_002293257.1
XP_002296120.1
XP_002294792.1
XP_002290864.1
XP_002292441.1
Thalassiosira pseudonana-3
(THAPSDRAFT-3)
Thalassiosira pseudonana-4
(THAPSDRAFT-4)
Thalassiosira pseudonana-5
(THAPSDRAFT-5)
Thalassiosira pseudonana-6
(THAPSDRAFT-6)
Thalassiosira pseudonana-7
(THAPSDRAFT-7)
Thalassiosira pseudonana-8
(THAPSDRAFT-8)
Thiothrix nivea DSM 5205
Trichodesmium erythraeum
WP_002710664.1
WP_011609991.1
SMAR-1, -2, -3, -4 in phylogenetic tree refer to the putative proteins reported from transcripts ID:
TR14735i7, TR5935i1, TR11343i2, TR8670i1, respectively (RNA-sequencing data have been depos-
ited under accession code SRP108217 in the NCBI SRA database) (Gallo et al., 2017).

Trichodesmium erythraeum
Thiothrix nivea DSM5205
87
44
Cyanothece sp. PCC7424
47
98
Cyanothece sp. PCC8801
Cyanothece sp. CCY0110
62
80
Crocosphaera watsonii
Cyanothece sp. ATCC51142
84
THAPSDRAFT-8
100
100
SMAR-1
SMFE7-1
88
62
Limulus polyphemus
Ixodes scapularis
Ancylostoma duodenale
48
100
100
Homo sapiens 2B1B
Homo sapiens 2B1A
Homo sapiens 1C
Homo sapiens 1B1
Homo sapiens 1B2
99
100
99
70
99
Rattus sp.
Arabidopsis lyrata
Arabidopsis thaliana
Daucus carota subsp. sativus
80
99
65
Brassica rapa
Capsella rubella
Camelina sativa
99
SMAR-2
PHATR-4
47
43
60
Emiliana huxleyi
SMAR-4
SMFE7-2
100
100
77
THAPSDRAFT-7
Nannochloropsis gaditana
Micromonas commoda
Chrysochromulina sp.
54
THAPSDRAFT-1
THAOC-2
FRAGPS-1
99
90
SMAR-3
90
SMFE7-3
THAPSDRAFT-3
THAOC-1
THAPSDRAFT-6
PHATR-3
THAPSDRAFT-2
THAPSDRAFT-5
45
PHATR-1
PHATR-2
THAPSDRAFT-4
99
0.1
Fig. 3 Phylogenetic relationship of SULT proteins of diatoms. The evolutionary history was inferred by using the maximum likelihood method
based on the Poisson correction model. The tree with the highest log likelihood (ꢀ17549.1328) is shown. The percentage of trees in which the
associated taxa clustered together (1000 bootstraps) is shown next to the branches. Branches corresponding to partitions reproduced in
<40% bootstrap replicates are collapsed. A discrete gamma distribution was used to model evolutionary rate differences among sites
(two categories (+G, parameter¼8.1260)). The rate variation model allowed for some sites to be evolutionarily invariable ([+I], 0.0000% sites).
The analysis involved 50 amino acid sequences. All positions with less than 80% site coverage were eliminated. That is, fewer than 20% align-
ment gaps, missing data, and ambiguous bases were allowed at any position. There were a total of 203 positions in the final dataset. Evo-
lutionary analyses were conducted in MEGA7. Scale bar represents 0.1 substitutions per amino acid position.

Sterol Sulfates and Sulfotransferases
111
under nutrient limitation and increase the resilience of the species in diverse
oceanic regimes (Thamatrakoln, Korenovska, Niheu, & Bidle, 2012). Sterol
sulfates of diatoms induce cell death of the diatom S. marinoi by a threshold
effect (Gallo et al., 2017). The molecules trigger activation of NO pathways
and oxidative burst that ultimately sensitize cells to programmed cell death
activation by apoptotic-like processes. In analogy with the terminal part of a
marine bloom, cell death is followed by massive formation of cellular debris.
Intracellular concentrations of sterol sulfates linearly correlate with culture
aging and reach autoinhibitory levels only in the declining phase, which
is in agreement with previous reports that correlate aging to bloom
demise in phytoplankton communities (Bidle, 2016; Bidle & Bender,
2008; Bidle & Falkowski, 2004; Chung, Hwang, & Chang, 2005). On
the contrary, impairment of sterol sulfate biosynthesis by inhibiting the
key sulfation step protects the diatom and reduces the susceptibility of the
cell cultures to crash. On the whole, these results imply participation of
these small molecules in the regulatory mechanisms of growth and prompt
the idea that they can act alone or in connection with other molecules
(D’Ippolito et al., 2009; Fontana, d’Ippolito, Cutignano, Miralto, et al.,
2007; Fontana, D’Ippolito, Cutignano, Romano, et al., 2007; Lamari
et al., 2013; Vardi et al., 2008, 2006) as intracellular mediators in signal
transduction pathways. In this view, dissection and manipulation of their
biosynthesis can offer a new view of cellular regulation in the oceans and
may find direct and general application in the development of diatom feed-
stocks for biotech applications (Daboussi et al., 2014; d’Ippolito et al., 2015).
We used diverse methods including NMR, MS profiling, mass cultivation,
transcriptomic, and cell biology tools to give a direct support of the role of
sterols sulfates in diatoms. Remarkably, similar pathways are present in every
diatom so far sequenced, and it is likely that these molecules have an univer-
sal role as death signals in the lineage thus playing a major role in bloom
termination.
4. METHODS
In the following section, the approach for the analysis of sterol sulfates
in marine diatoms is illustrated. If not otherwise specified, standard protocols
were used. All of the described reagents and chromatographic supports were
purchased from commercial vendors. It is important to note that ratio-
nalization based on sequence alignments of SULTs is of limited value in
predicting biochemical properties of unknown members of this protein

112
Carmela Gallo et al.
family (Hirschmann et al., 2014). This is due to the fact that minor changes
of SULT sequences lead to differences in substrate affinities and enzyme
kinetics, i.e., SULTs with more than 85% sequence homology do not display
similar enzymatic characteristics. In general, experiments of enzyme substrate
specificities and phylogenetic analysis seem do not correspond. For example,
the in silico assignment of A. thaliana SULT13 as a brassinosteroid SULT did
not correspond with experimental data that showed flavonoids as preferred
substrates (Hirschmann et al., 2014). Thus, enzymatic assays with additional
functional studies are necessary to give reliable information about substrate
specificity and role of these enzymes in diatoms.
4.1 Isolation and Purification of Sterol Sulfates
4.1.1 Algal Culture and Sample Preparation
4.1.1.1 Equipment
•
•
•
•
Centrifuge with a swing-out rotor (Allegra X12R—Beckman Coulter)
Rotary evaporator
Climate chamber
2-L polycarbonate autoclavable bottles, transparent with polypropylene
screw cap, heat resistant up to +135°C, autoclavable
Fill venting closures for 2-L carboys, autoclavable, diameter: 53mm
Sartorius 0.22-μm Midisart filters
Micropipette
•
•
•
•
•
•
•
•
•
•
•
•
•
Glass beaker
Hot plate
50-mL Falcon tubes
Conical centrifuge tubes
Solvent-resistant pipette tips
Pasteur pipettes
15- and 7-mL Glass screw-top vials
Magnetic stirrer
Magnetic stirrer bars
Vortex mixer
4.1.1.2 Buffers and Reagents
•
f/2 medium—filtered (0.22μM) sterile sea water supplemented with:
106μM Na2SiO3ꢁ9H2O, 882μM NaNO3, 36μM NaH2PO4ꢁH2O,
2.05nM biotin, 0.369nM cyanocobalamin (B12), 0.296μM Thiamine-
HCl, 117μM Na2 EDTA, 117μM FeCl3ꢁ6H2O, 0.91μM MnCl2ꢁ4H2O,

Sterol Sulfates and Sulfotransferases
113
39.3nM CuSO4ꢁ5H2O, 76.5nM ZnSO4ꢁ7H2O, 42nM CoCl2ꢁ6H2O,
26nM NaMoO4ꢁ2H2O, 0.4M Trizma buffer and adjust pH at 7.8 by
98% H2SO4 or 37% HCl.
•
•
•
•
MeOH HPLC grade (ꢂ99%)
Trizma® (2-amino-2-(hydroxymethyl)-1,3-propanediol) hydrochloride.
MOPS (3-(N-morpholino)-propanesulfonic acid)
Milli-Q® deionized water
4.1.1.3 Procedure
1. Set up diatom cultures in prefiltered sterile (0.22μm) f/2 medium at an
initial concentration of 1 cellsmLꢀ1.
2. Add 0.4 M TRIS–HCl buffer (pH 8) and adjust pH to 7.8–8.1 with
HCl if necessary.
3. Incubate cultures in the climate chamber (T¼22.0ꢃ1.0°C) at
200 μmol photons mꢀ2 sꢀ1, with 14:10h (light:dark) photoperiod.
4. Harvest diatoms in the first day of the declining growth phase by gentle
centrifugation in conical tubes at 2300ꢄg for 10min at 12°C.
5. Decant supernatant in the sink and immediately determine the wet
weight of the pellet.
6. Add MeOH (5mL/g wet biomass) to the pellet.
7. Vortex for 10s and sonicate for 30s.
8. Centrifuge the suspension at 2300ꢄg for 5min at 5°C.
9. Transfer the supernatant to a recovery flask by a Pasteur pipette.
10. Repeat twice the extraction of the residual pellet with MeOH (5mL/g
wet biomass).
11. Combine the hydroalcoholic solutions and dry by a rotary evaporator.
12. Dissolve the dry extract in a minimal volume of MeOH/CH2Cl2 2:1
(v/v).
13. Transfer the solution to a glass vial by Pasteur pipette.
14. Rinse with 1.5mL MeOH/CH2Cl2 2:1 (v/v) and repeat if necessary.
15. Remove solvents under nitrogen.
16. Store under argon and at ꢀ80°C until analysis.
4.1.1.4 Notes
1. The extraction procedure can be readily used for other marine samples,
including invertebrates and algae. In those cases, it is recommended to
freeze-dry and grind the sample before the addition of MeOH.

114
Carmela Gallo et al.
2. With a few diatoms (e.g., S. marinoi) and microalgae, it is convenient to
replace TRIS–HCl with 0.4M MOPS buffer and adjust the pH to 7.8
with NaOH.
4.1.2 Solid-Phase Extraction
4.1.2.1 Equipment
•
CHROMABOND® HR-X cartridge (plastic tube size¼6mL; sorbent
weight¼500mg) (Sorbent Technologies, Inc.)
GX-271 ASPEC® apparatus (Gilson, Inc.) equipped with Trilution® LH
software (Fig. 4)
•
•
•
•
•
Rotary evaporator
10-mL Borosilicate glass tubes
7-mL Glass screw-top vials
15-mL Conical centrifuge tubes (corning, Falcon®)
4.1.2.2 Solvents and Reagents
•
•
•
•
CH3CN HPLC grade (ꢂ99%)
MeOH HPLC grade (ꢂ99%)
CH2Cl2 HPLC grade (ꢂ99%)
Milli-Q® deionized water
Fig. 4 Image of GX-271 ASPEC Gilson liquid handling apparatus. The inset (left corner)
shows (A) needle probe, (B) sealing caps 6mL, and (C) Prepacked Chromabond
HR-column.

Sterol Sulfates and Sulfotransferases
115
4.1.2.3 Procedure
1. Suspend 20mg of dry extract in 1mL distilled water.
2. Transfer the suspension to a 15-mL conical tube.
3. Turn on GX-271 ASPEC Gilson apparatus.
4. Prepare eluents: H2O 100%, CH3OH:H2O 1:1 (v/v), CH3CN:H2O
7:3 (v/v), CH3CN 100%, CH2Cl2:CH3OH 9:1 (v/v).
5. Set bed layout positions according to Fig. 5:
Place solvents as described.
Place HRX-cartridge in the well “DEC Mobile Rack 1.”
Place the sample to load in the well “Sample 1.”
Place 5 glass tubes in the wells 1 of the “Collector” from 1 to 5.
6. Set liquid handling operations as follows:
6.1 Condition, 6mL MeOH.
6.2 Condition, 6mL H2O.
6.3 Repeat condition step, 6 mL H2O.
Fig. 5 Bed layout of GX-271 ASPEC Gilson liquid handling apparatus. The picture shows
in detail how to accommodate solvents, samples, column, and vials (see main text).

116
Carmela Gallo et al.
6.4 Load and collect, 1mL (1mL of sample solution is withdrawn
and loaded onto column).
6.5 Dispenser, 2mL H2O.
6.6 Mix, 1mL H2O.
6.7 Load and collect, 2mL.
6.8 Fractionate, 6mL H2O (to collect in well 1 of Collector 1).
6.9 Fractionation, 9mL CH3OH/H2O 1:1 (v/v) (to collect in well 1
of Collector 2).
6.10 Fractionation, 9mL CH3CN/H2O 7:3 (v/v) (to collect in well 1
of Collector 3).
6.11 Fractionation, 9mL 100% CH3CN (to collect in well 1 of Col-
lector 4).
6.12 Fractionation, 9mL CH2Cl2/CH3OH 9:1 (v/v) (to collect in
well 1 of Collector 5).
7. Transfer fraction eluted by CH3CN/H2O 7:3 (v/v) to an evaporating
flask.
8. Remove solvent by rotary evaporation.
9. Dissolve the dry fraction in a minimal volume of MeOH.
10. Transfer the solution to a glass vial by Pasteur pipette.
11. Rinse with 1.5mL MeOH.
12. Combine and remove solvents under nitrogen.
13. Store fraction under argon at ꢀ20°C until analysis.
4.1.2.4 Notes
1. Between each step rinse needle by MeOH.
2. For further information check the instruction manual of Trilution® LH.
4.1.3 HPLC Purification
4.1.3.1 Equipment
•
•
•
JASCO LC-2000Plus HPLC composed of PU-2089 Plus-quaternary
gradient pump and MD-2018 Plus photodiode array detector (PDA).
Sedex 85 low-temperature-evaporative light scattering detector
(LT-ELSD) (SEDERE, Inc.).
Reversed-phase column (Phenomenex, C-18 Luna), 4.6ꢄ250mm,
˚
100A.
•
•
•
•
Rotary evaporator (Buchi, Rotavapor R-200).
Borosilicate glass tubes.
7-mL Glass screw-top vials.
Pasteur pipettes.

Sterol Sulfates and Sulfotransferases
117
4.1.3.2 Solvents and Reagents
•
•
MeOH HPLC grade (ꢂ99%).
Milli-Q deionized water.
4.1.3.3 Procedure
1. Checking the initial settings of the HPLC system:
1.1. Confirm that the waste line is in a waste container.
1.2. Verify that the minimum and maximum pressure and the flow
rate are set to the correct values on the front panel of the solvent
delivery system (the pump).
1.3. Press “zero” on the LT-ELSD front panel in order to set the
blank.
1.4. Rinse a 100-μL syringe with deionized water, then with
100μL MeOH.
2. Prepare the mobile phases:
Deionized water (Pump A)
MeOH (Pump B)
3. Connect reversed-phase column.
4. Wash column with 100% MeOH for 10min at a flow of 2mL/min.
5. Set the flow rate at 1mL/min.
6. Equilibrate the HPLC system in MeOH:H2O 85:15 (v/v) for 10min.
7. Set gradient conditions:
MeOH:H2O 85:15 (v/v) for 10min
MeOH:H2O 85:15 (v/v) to MeOH:H2O 90:10 (v/v) in 30min
Return to MeOH:H2O 85:15 (v/v)
8. Set evaporative light-scattering detector in order to have:
Evaporation temperature 45°C
Nebulization temperature 60°C
N2 flow 1.4mLminꢀ1
9. While system is equilibrating, dissolve fraction in MeOH at a final con-
centration of 5mg/mL.
10. With the injector handle in the load position, slowly inject 100–150μL
sample solution through the septum port.
11. Verify that the data collection program is set to collect data for 40min.
12. When ready to start, rotate the injector handle to the inject position
(which injects the sample into the mobile phase and start on the com-
puter data collection program immediately).
13. Leave the syringe in the septum.
14. Collect peaks between 10 and 25min in glass vials.
15. Dry at rotary evaporator and store samples under argon at ꢀ20°C.

118
Carmela Gallo et al.
4.2 Identification of Sterol Sulfates
The spectroscopic data about sterol sulfates are rather limited, and unambig-
uous identification of the molecules is more challenging and time-
consuming than one could expect on the basis of the large literature about
identification and characterization of sterols. The weakness is the absence of
NMR data in solvents that are commonly used with both classes of com-
pounds. In fact, most of the NMR data of sterols are reported in deuterated
chloroform (or other chlorinated solvents) in which sterol sulfates are little
soluble. On the other hand, the best solvent for sterol sulfates is deuterated
methanol in which sterols are not soluble. In Sections 4.2.1 and 4.2.2, we
report two simple methods that allow for a comparison of sterol sulfates with
the corresponding sterols (Fig. 6) in order to obtain a rapid and clear-cut
identification of the molecules.
4.2.1 Chemical Conversion of Sterols to Sterol Sulfates
4.2.1.1 Equipment
•
•
•
•
•
•
•
•
Rotary evaporator
Heating magnetic stirrer
Magnetic stirrer bars
Borosilicate glass tubes
7-mL Glass screw-top vials
Pasteur pipettes
Glass syringes
25-mL Evaporating flask
4.2.1.2 Solvents and Reagents
•
•
•
•
•
•
•
Sulfur trioxide triethylamine complex (Sigma-Aldrich)
Dowex 50WX8, sodium form
Anhydrous N,N-dimethylformamide
Silica gel 60 (0.063–0.2mm) (Merck)
Deionized water
Glass TLC silica plate 60G F254 (Merck)
Argon
Fig. 6 Derivatization of sterols to sterol sulfates.

Sterol Sulfates and Sulfotransferases
119
4.2.1.3 Procedure
1. Mix triethylamine-sulfur trioxide complex (0.04mmol) and sterol
(0.01mmol) in anhydrous N,N-dimethylformamide (0.5mL) under
argon atmosphere.
2. Stir at 35°C for 1h.
3. Add water (0.2mL).
4. Pour the mixture over a silica gel column to remove the excess of
triethylamine-sulfur trioxide.
5. Elute with 4mL diethyl ether:dichloromethane 1:1 (v/v) and discard.
6. Elute with 4mL methanol/dichloromethane 1:1 (v/v) and collect the
sample.
7. Transfer to an evaporating flask.
8. Dry by rotary evaporation.
9. Add 5mL methanol.
10. Add 0.5g Na+ Dowex 50WX8.
11. Stir for 5h at room temperature.
12. Filtrate on paper and concentrate the filtrate by rotary evaporator.
13. Repeat the exchange reaction with 0.5g Na+ Dowex 50WX8.
14. Filtrate the mixture.
15. Combine filtrates and transfer to an evaporating flask.
16. Dry at rotary evaporator and store samples under argon at ꢀ20°C.
4.2.2 NMR Analysis
4.2.2.1 Equipment
•
AMX 600 spectrometer equipped with an inverse TCI Cryoprobe®
(Bruker)
•
•
•
5mm Precision NMR tubes
Pasteur pipette
7-mL Glass screw-top vials
4.2.2.2 Solvents and Reagents
•
•
CDCl3 ꢂ99.8 atom % D, anhydrous
CD3OD ꢂ99.8 atom % D
4.2.2.3 Procedure
1. Prepare the mixture of CD3OD/CDCl3 1:1 (v/v) in a glass vial.
2. Dissolve dry sterol sulfates in the deuterated mixture and transfer to
NMR tube.

120
Carmela Gallo et al.
A
B
ppm
1.00
0.95
0.90
0.85
0.80
0.75
0.70
Fig. 7 1H-NMR comparison of the region of the methyl groups of dihydrobrassicasterol
(A) and dihydrobrassicasterol sulfate (B) in CD3OD/CDCl3 1:1 (v/v) (CH3OD, 3.34ppm).
3. Acquire 1H-NMR spectrum (Pulprog: zg) with ns¼8 (d1¼1s,
SW¼14, O1¼5ppm).
4. Acquire 13C NMR spectrum (Pulprog: zggppgse.t1.be) with ns¼30000
(d1¼1s; SW¼300, O1¼100ppm).
5. Compare NMR spectra of sterols and sterol sulfates.
4.2.2.4 Notes
1. Prepare CDCl3:CD3OD (1:1) mixture prior to use and do not store.
2. NMR data in CDCl3:CD3OD 1:1 are comparable but not identical to
literature values of sterols in CDCl3.
3. Despite slight changes of chemical shift, the distribution of the methyl
1
groups in the H-NMR region between 0.5 and 1.2ppm (“methyl”
region) in CDCl3:CD3OD 1:1 is reliable for the identification of sterols
with alkyl-branched chains, as exemplified for dihydrobrassicasterol and
dihydrobrassicasterol sulfate in the below Fig. 7.
4.3 UPLC-MS Analysis of Sterol Sulfates
4.3.1 Equipment
•
Q-Exactive Hybrid Quadrupole-Orbitrap mass spectrometer (Thermo
Fisher Scientific) equipped with an electrospray ionization (ESI) source
and Infinity 1290 UHPLC (Agilent Technologies)

Sterol Sulfates and Sulfotransferases
121
•
•
Xcalibur™ software (Thermo Fisher Scientific)
Reversed-phase column, Kinetex Biphenyl (Phenomenex) 2.1ꢄ150mm,
2.6μm
•
Verex™ Vial (10mm Screw Top, 2mL) with Blue (12ꢄ32mm) Screw
Neck Cap with preslit PTFE/silicone septum
Micropipettes
•
•
•
Solvent-resistant pipette tips
7-mL Glass screw-top vials
4.3.2 Solvents and Reagents
•
•
•
•
MeOH, LCMS grade
Water, LCMS grade
Sodium cholesteryl sulfate (Sigma-Aldrich)
[25,26,26,26,27,27,27-D7]-Cholesterol sulfate (C/D/N Isotopes)
4.3.3 Procedure
1. Prepare the mobile phases:
Water (Pump A)
MeOH (Pump B)
2. Set autosampler at 10°C and column oven at 28°C.
3. Set flow rate a 0.3-mLminꢀ1.
4. Equilibrate the HPLC system for 5min in MeOH:H2O 7:3 (v/v).
5. Set gradient conditions as follows:
MeOH:H2O 70:30 (v/v) per 2min
MeOH:H2O 70:30 (v/v) to MeOH:H2O 85:15 (v/v) in 8min
MeOH:H2O 85:15 (v/v) to 100% MeOH in 5min
100% MeOH for 5min
Return to MeOH:H2O 70:30 (v/v) and reequilibrate for 5min
6. Set parameters of Q-Exactive Hybrid Quadrupole-Orbitrap mass
spectrometer as follows:
a. Spray voltage negative polarity 3.0kV, capillary temperature 320°C,
S-lens RF level 60, auxiliary gas temperature 320°C, sheath gas flow
rate 50L/min, auxiliary gas flow rate 30L/min.
b. Full MS scans range 200–600, mass resolution of 70,000.
c. Target value (AGC) 1ꢄ106, maximum to allow accumulation
time (IT) at 100ms.
d. For data-dependent MS/MS (ddMS2) analysis, a Top10 method is
used. The 10 most intense peaks are selected for fragmentation with
a stepped normalized energy of 20–25–30eV. AGC is 1ꢄ105 with
IT 75ms and 17,500 mass resolution.

122
Carmela Gallo et al.
7. Set run time at 20min.
8. Prepare triplicates of blank and standard cholesterol sulfate(s) at six dif-
ferent concentrations between 1ng and 200ngmLꢀ1.
9. Build the calibration curve by sequential injection of blank and standard
solutions. Run each concentration point in triplicate.
10. While the analysis of standard is running, add d7-cholesterol-sulfate to
the diatom pellet (1μg/10mg of wet diatom biomass) and extract
according to procedure described in Section 4.1.1.3.
11. Dissolve the extracts in MeOH at a final concentration of 0.05μgμLꢀ1.
12. Inject 5μL sample.
13. Extract peaks of sterol sulfates (Table 2) and standard (m/z 472.3).
14. Verify the presence of the diagnostic fragment at m/z 97 in the MS/MS
spectra due to the loss of sulfate.
15. Integrate areas by Xcalibur™ software.
16. Build the calibration curve by plotting peak areas of standard against
concentrations.
17. Use the calibration curve equation of the standard to calculate sterol
sulfates concentration in the extracts.
4.3.4 Notes
1. This method has been tested for sterol sulfates in the diatoms S. marinoi
and Cyclotella cryptica (Gallo et al., 2017). As reported in the below Fig. 8,
sterol sulfates in these species show the following composition and reten-
tion times: 5.83min for cholesterol sulfate (CHOS), 6.41min for
dihydrobrassicasterol (DHBS), and 7.01min for β-sitosterol sulfate
(βSITS).
4.4 Inhibition of Sterol Sulfate Biosynthesis in Marine Diatoms
The involvement of SULTs in the development of various diseases and alter-
ation of metabolic processes has led to a search for inhibitors of these enzymes
as potential therapeutics (Cook et al., 2016; Marto, Morello, Monteiro, &
Pereira, 2017; Rath et al., 2004). These compounds belong to different chem-
ical classes including environmental chemicals (e.g., hydroxylated poly-
chlorinated biphenyls), steroids (e.g., danazol), drugs (e.g., ibuprofen), and
natural products (e.g., curcumin) (Wang & James, 2006). In this section, a suc-
cessful strategy for the inhibition of the biosynthesis of sterol sulfates in diatoms
by use of the SULT inhibitor quercetin is described. Flavonoids are recog-
nized inhibitors of different SULTs (Alhusainy et al., 2012; Ghazali &
Waring, 1999), and quercetin, one of the most abundant flavonoids,

Sterol Sulfates and Sulfotransferases
123
Table 2 Structures and ESIꢀ Molecular Ion Peak of Selected Sterol Sulfates
Compound
m/z
Chemical Structure
Cholesterol
465.3
O
–O
O
S
O
d7-Cholesterol
472.3
505.3
CD3
D
D3C
O
O
S
–O
O
Lanosterol
O
O
S
–O
O
Desmosterol
463.3
477.3
479.3
477.3
O
O
S
S
S
S
–O
O
24-Methylene cholesterol
O
O
–O
O
Campesterol
O
O
–O
O
Brassicasterol
O
O
–O
O
Continued

124
Carmela Gallo et al.
Table 2 Structures and ESIꢀ Molecular Ion Peak of Selected Sterol Sulfates—cont’d
Compound
m/z
Chemical Structure
Dihydrobrassicasterol
479.3
O
O
S
S
S
–O
O
β-Sitosterol
Stigmasterol
Fucosterol
493.3
491.3
491.3
O
O
–O
O
O
O
–O
O
O
O
S
–O
O
is able to inhibit sulfation of different substrates, including sterols by human
SULT2A1 (Mes´ıa-Vela & Kauffman, 2003).
4.4.1 Equipment
•
•
•
24 Multiple well plates
96 black Multiple well plates
Microscope (AxioVertA1; magnification of ꢄ200 and ꢄ400) (Carl
Zeiss)
•
•
•
Bu€rker chamber
Centrifuge with a swing-out rotor (Allegra X12R) (Beckman Coulter)
FP-8300 fluorescence spectrophotometer (JASCO) equipped with fluo-
rescence microplate reader

100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
DHBS
m/z 479.3193
βSITS
m/z 493.3354
CHOS
m/z 465.3037
0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
Time (min)
Fig. 8 Composition and elution order of sterol sulfates of the marine diatoms Skeletonema marinoi and Cyclotella cryptica. CHOS, cholesterol
sulfate; DHBS, dihydrobrassicasterol sulfate; βSITS, β-sitosterol sulfate.

126
Carmela Gallo et al.
4.4.2 Buffers and Reagents
•
f/2 medium: filtered (0.22μM) sterile sea water supplemented with:
Na2SiO3ꢁ9H2O, final concentration: 106μM; NaNO3 final concentra-
tion: 882μM; NaH2PO4ꢁH2O, final concentration: 36μM; Biotin (vit.
H), final concentration: 2.05ꢄ10ꢀ9 M; Cyanocobalamin (B12), final
concentration: 3.69ꢄ10ꢀ10 M; Thiamine-HCl, final concentration:
2.96ꢄ10ꢀ7 M; Na2EDTA, final concentration: 1.17ꢄ10ꢀ5 M;
FeCl3ꢁ6H2O, final concentration: 1.17ꢄ10ꢀ5 M; MnCl2ꢁ4H2O, final
concentration: 9.10ꢄ10ꢀ7 M; CuSO4ꢁ5H2O, final concentration:
3.93ꢄ10ꢀ8 M; ZnSO4ꢁ7H2O, final concentration: 7.65ꢄ10ꢀ8 M;
CoCl2ꢁ6H2O, final concentration: 4.20ꢄ10ꢀ8 M; NaMoO4ꢁ2H2O,
final concentration: 2.60ꢄ10ꢀ8 M; 0.4M Trizma buffer and adjust
pH up at 7.8 by H2SO4 98% or HCl 37%.
•
•
DMSO
Quercetin, >95% (HPLC) (Sigma-Aldrich)
4.4.3 Procedure
1. Inoculate 24-multiwell plate with 1mL/well of exponentially growing
diatom culture (6ꢄ104 cellsmLꢀ1) in f/2 medium.
2. Prepare scalar concentrations of quercetin between 0.25 and 40μM in
10μL of DMSO.
3. Add compound to cells in the multiwell plate. Load each concentration
point in triplicate.
4. Leave three wells with cells without treatment as blank and three wells
only with 10μL DMSO as control.
5. Incubate plates at 200μmol photons mꢀ2 sꢀ1, at 20ꢃ2°C, with 14:10h
light:dark photoperiod.
6. Each day recover 100μL of sample from each well after gentle mixing
of the plate by micropipette.
7. Count and monitor each sample by Bu€rker chamber (10μL/sample).
See note 1.
8. Pipette 80μL of each sample in a black 96-well plate.
9. Pipette solution up and down to mix.
10. Measure the fluorescence emission at 679/5nm by fluorescence spec-
trophotometer plate reader. Set 451–455nm as excitation wavelength.
See note 2.
11. Select samples affected by treatment and scale up the cultures to 10L
volume in triplicate.
12. Recover cells by centrifugation at 2300ꢄg for 10min at 12°C.
13. Carry out quantitation of sterol sulfates according to the LC-MS pro-
tocol described in Section 4.3.

Sterol Sulfates and Sulfotransferases
127
4.4.4 Notes
1. For cell counting, use at least three squares (dilute samples if the count is
ꢂ100 cells/square), calculate the median value, and determine cell con-
centration (cell/mL) by the following calculation:
Cell=mL ¼ average countper squareꢄdilution factor ꢄ104
2. Chlorophyll a fluorescence is used as measure of the total amount of
intact cells. Calibration curve is to repeat before each fluorescence assay
by using different dilution and the same volume of the diatom culture
used for the assay.
3. If the LC-MS analysis is not carried out immediately after cell collection,
freeze pellet in liquid nitrogen and store at ꢀ80°C until analysis.
4.5 13C Labeling of the Carbon Skeleton of Diatom Sterols
Labeling studies are a standard technique to probe cellular metabolism and to
elucidate in vivo biosynthetic pathways. With sterols, the use of 1-13C-
glucose is also conclusive to differentiate between mevalonate and non-
mevalonate metabolic pathways. Current methods are restricted to few
species, and there is a pressing need for new rational approaches in order
to systematically use labeled probes in complex systems such as cells of
marine microalgae. The following section describes the labeling of sterol
skeletons after use of 1-13C-glucose with culture of autotrophic microalgae.
4.5.1 Equipment
•
Climate chamber
1-L sterile ventilated culture flasks (TPP25-300cm2) with filter
•
screw caps
•
Savan Micro Modulyo freeze dryer (Thermo Scientific, Austin, TX,
USA)
•
•
•
•
•
•
•
•
•
•
•
Rotary evaporator
Micropipette
Vortex
Orbital agitator
Magnetic stirrer
Magnetic stir bars
Conical centrifuge tubes
Solvent-resistant pipette tips
Pasteur pipettes
7-mL Glass screw-top vials
15mL Evaporating flask with conical ground

128
Carmela Gallo et al.
4.5.2 Buffers and Reagents
•
f/2 medium—filtered (0.22μM) sterile sea water supplemented
with: 106μM Na2SiO3ꢁ9H2O, 882μM NaNO3, 36μM NaH2PO4ꢁH2O,
2.05nM biotin, 0.369nM cyanocobalamin (B12), 0.296μM Thiamine-
HCl, 117μM Na2 EDTA, 117μM FeCl3ꢁ6H2O, 0.91μM MnCl2ꢁ
4H2O, 39.3nM CuSO4ꢁ5H2O, 76.5nM ZnSO4ꢁ7H2O, 42nM CoCl2ꢁ
6H2O, 26nM NaMoO4ꢁ2H2O, 0.4M Trizma buffer and adjust pH at
7.8 by 98% H2SO4 or 37% HCl
•
Stock solution in f/2 medium of antibiotics: penicillin, streptomycin,
ampicillin, and kanamycin
•
•
Methanol, HPLC grade
D-Glucose-1-13C, 99 atom % 13C (Sigma-Aldrich)
4.5.3 Procedure
1. Set up diatom cultures (1ꢄ104 cellmLꢀ1) in 0.4L of prefiltered
(0.22μm) sterile f/2 medium.
2. Add 1.5μgmLꢀ1 penicillin, 1μgmLꢀ1 streptomycin, 1μgmLꢀ1 ampi-
cillin, and 0.5μgmLꢀ1 kanamicin.
3. Supplement the culture with 2gLꢀ1 [1-13C]-glucose.
4. Incubate cultures in ventilated 1-L sterile culture flask in the dark, at
20ꢃ2°C.
5. Maintain culture flask under gentle agitation by orbital agitator for
2 weeks.
6. Recover cells by centrifugation at 2300ꢄg for 10min at 12°C.
7. Freeze pellets in liquid nitrogen and store at ꢀ80°C until analysis.
8. Add MeOH (5mL/g wet biomass) to the pellet.
9. Vortex for 10s and sonicate for 30s.
10. Centrifuge the suspension at 2300ꢄg for 5min at 5°C.
11. Transfer the supernatant to a recovery flask by a Pasteur pipette.
12. Repeat twice the extraction of the residual pellet with MeOH (5mL/g
wet biomass).
13. Combine the hydroalcoholic solutions and dry by a rotary evaporator.
14. Dissolve the dry extract in a minimal volume of MeOH/CH2Cl2
2:1 (v/v).
15. Transfer the solution to a glass vial by Pasteur pipette.
16. Rinse with 1.5mL MeOH/CH2Cl2 2:1 (v/v) and repeat if necessary.
17. Remove solvents under nitrogen.
18. Fractionate dry extract according to SPE protocol reported in
Section 4.1.2.

Sterol Sulfates and Sulfotransferases
129
19. Collect fraction eluted by CH3CN/H2O 7:3 (v/v) for sterol sulfates.
20. Collect fraction eluted by 100% CH3CN for sterol.
21. Dry fractions under nitrogen and store at ꢀ80°C until analysis or use.
4.5.4 Notes
1. Establishment of heterotrophic conditions depend on strain and culture
parameters, whereas consumption of carbon depends on the transport or
diffusion of the carbon source across the membrane, and on the enzy-
matic processes required for its incorporation into the central carbon
metabolism. Active transport appears to be the primary means by which
algae acquire organic carbon substrates from the environment. Active
transport of carbon sources has been documented in several species of
diatoms and is likely the most common mechanism for sequestering
these compounds. The diatom C. cryptica has been used as model species
in this protocol.
2. Standard quantity and solvent volumes of pellet extraction refer to 50mg
of dry weight biomass.
5. CONCLUSIONS
Sterol sulfates are very common in marine organisms, but, despite
several reports about occurrence and potential function, their biosynthesis
has not been addressed so far. In this chapter we have shown analytical tech-
niques and experimental protocols using whole, living diatoms that allow
identification and function of sterol sulfates, as well as confirmation of the
function assigned to sulfation of this class of terpenes. The chapter also
includes the labeling of the sterol skeleton by stable isotopes in order to test
the synthetic activity and elucidate the biosynthetic route. The methods are
tailored on products of diatoms, but, with little modifications and careful
evaluation, they are also suitable for the study of this family of natural prod-
ucts in other marine organisms.
The rationale for the use of these techniques is the lack of biochemical
and molecular information to anchor the functional characterization of
SULT in marine environments. In diatoms, transformation approaches, sta-
ble generation of strains with modulated gene expression and genetic engi-
neering, are slowly becoming accessible and their use is so far restricted to
very few species (mainly, Phaeodactylum tricornutum and Thalassiosira
pseudonana) among the thousands that are predicted to live in the aquatic
environments (Barrera & Mayfield, 2013; Buhmann et al., 2014;

130
Carmela Gallo et al.
Daboussi et al., 2014; Hopes, Nekrasov, Kamoun, & Mock, 2016; Kadono
et al., 2015, 2016; Kira, Ohnishi, Miyagawa-Yamaguchi, Kadono, &
Adachi, 2016; Niu et al., 2012; Nymark et al., 2017; Nymark, Sharma,
Sparstad, Bones, & Winge, 2016; Radakovits, Eduafo, & Posewitz, 2011;
Sabatino et al., 2015; Uhlig et al., 2011). Furthermore, to the best of our
knowledge, there are only a few reports of characterization of marine
diatom proteins by heterologous expression (Ast et al., 2009; Domergue
et al., 2003; Meyer et al., 2004; Uhlig et al., 2011). Paradoxically, correlation
of functions to genes and gene products is even more important in the light
of the advent of the modern sequencing techniques that are quickly expan-
ding the genetic information of diatoms. Currently, most of the information
on eukaryotic genomes come from mammals, fungi, and plants (Del Campo
et al., 2014; Kuhlisch, Deicke, Ueberschaar, Wichard, & Pohnert, 2017).
The lack of appropriate reference sequences risks to lead to erroneous con-
clusions with those organisms, such as marine invertebrates or microalgae,
that fall outside these groups. The hazard is even higher with taxonomic
groups such as diatoms that have genomes apparently deriving from the
fusion of genetically distant ancestors or with enzymes such as SULTs that
have affinity with a wide number of diverse substrates but have relatively
high sequence similarity and great plasticity. The tools reported in this chap-
ter aim to provide functional support in the presence of little molecular
information about the biosynthesis of diatom sterol sulfates and, whenever
possible, to overcome the bias arising by the asymmetric acquisition of geno-
mic data among eukaryotes. It is easy to predict that attribution of function
to elements of cell structure such genes and proteins will be one of the major
themes in molecular and cellular biology of this lineage of microalgae in the
next future.
REFERENCES
Aiello, A., Fattorusso, E., & Menna, M. (1999). Steroids from sponges: Recent reports.
Steroids, 64, 687–714.
Alhusainy, W., van den Berg, S. J. P. L., Paini, A., Campana, A., Asselman, M.,
Spenkelink, A., et al. (2012). Matrix modulation of the bioactivation of estragole by
constituents of different alkenylbenzene-containing herbs and spices and physiologically
based biokinetic modeling of possible in vivo effects. Toxicological Sciences, 129(1),
174–187. https://doi.org/10.1093/toxsci/kfs196.
Allali-Hassani, A., Pan, P. W., Dombrovski, L., Najmanovich, R., Tempel, W., Dong, A.,
et al. (2007). Structural and chemical profiling of the human cytosolic sulfotransferases.
PLoS Biology, 5(5), e97. https://doi.org/10.1371/journal.pbio.0050097.
Allen, J. T., Brown, L., Sanders, R., Moore, C. M., Mustard, A., Fielding, S., et al. (2005).
Diatom carbon export enhanced by silicate upwelling in the Northeast Atlantic. Nature,
437(7059), 728–732. https://doi.org/10.1038/nature03948.