Immunostimulatory Phosphatidylmonogalactosyldiacylglycerols (PGDG) from the marine diatom Thalassiosira weissflogii: Inspiration for a novel synthetic toll-like receptor 4 agonist

Article abstract of DOI:10.3390/md17020103

An unprecedented phosphatidylmonogalactosyldiacylglycerol pool (PGDG, 1) rich in polyunsaturated fatty acids was isolated from the marine diatoms Thalassiosira weissflogii. Here we report for the first time the NMR characterization of this rare lipid from

Full text of DOI:10.3390/md17020103

Article
Immunostimulatory
Phosphatidylmonogalactosyldiacylglycerols (PGDG)
from the Marine Diatom Thalassiosira weissflogii:
Inspiration for a Novel Synthetic Toll-Like Receptor
4 Agonist
Emiliano Manzo 1,†, Carmela Gallo 1,†, Rossella Sartorius 2, Genoveffa Nuzzo 1, Angela Sardo 1,
Piergiuseppe De Berardinis 2, Angelo Fontana 1 and Adele Cutignano 1,*
1
CNR-Institute of Biomolecular Chemistry, via Campi Flegrei, 34 - 80078 Pozzuoli (Napoli), Italy;
emanzo@icb.cnr.it (E.M.); carmen.gallo@icb.cnr.it (C.G.); nuzzo.genoveffa@icb.cnr.it (G.N.);
angela.sardo@icb.cnr.it (A.S.); afontana@icb.cnr.it (A.F.)
2
CNR-Institute of Protein Biochemistry, via Pietro Castellino, 111 - 80131 Napoli, Italy;
r.sartorius@ibp.cnr.it (R.S.); p.deberardinis@ibp.cnr.it (P.D.B.)
* Correspondence: acutignano@icb.cnr.it (A.C.); Tel.: +39 081 8675313
†
These authors contributed equally to this work.
Received: 16 January 2019; Accepted: 5 February 2019; Published: 9 February 2019
Abstract: An unprecedented phosphatidylmonogalactosyldiacylglycerol pool (PGDG, 1) rich in
polyunsaturated fatty acids was isolated from the marine diatoms Thalassiosira weissflogii. Here we
report for the first time the NMR characterization of this rare lipid from marine organisms along
with a synthetic strategy for the preparation of a PGDG analog (2). PGDG 1 exhibited
immunostimulatory activity in human dendritic cells (DCs) and the synthetic PGDG 2 was prepared
to explore its mechanism of action. A Toll-like receptor-4 (TLR-4) agonistic activity was evidenced
in human and murine DCs underlying the antigen-specific T-cell activation of this class of
molecules.
Keywords: marine diatom; Thalassiosira weissflogii; phosphogalactolipids; Toll-like receptor 4;
dendritic cells; immunostimulatory lipid; phosphoglycolipid synthesis
1. Introduction
Glycerolipids containing both a carbohydrate and a phosphate group as structural components
are uncommon in nature. They have been classified in two groups named ‘glycophospholipids’ and
‘phosphoglycolipids’, respectively. Members of the first group include phosphatidylglucosides
reported from bacterial and, more recently, from mammalian sources [1,2]. They are typically
characterized by saturated fatty acids and are minor lipids in quantitative terms although they are
thought to play important biological functions [3,4]. On the other hand, phosphoglycolipids are
typical bacterial products, for the first time identified in the 1960s [5] and afterwards characterized
only by chemical degradation [6–10] and mass spectrometric techniques [11]. In phosphoglycolipids,
the phosphate moiety is located on position 6 of the sugar while in glycophospholipids typically
occurs on the anomeric carbon 1. Many glycolipids exhibited immunomodulatory functions as
previously reported [12–14] and immunostimulants represent an emerging class of drugs for the
treatment of infectious disorders and cancer. In the frame of a biological screening of marine extracts
on Peripheral Blood Mononuclear Cells (PBMC) from human donors, the diatom Thalassiosira
weissflogii (CCMP 1336) was selected due to its immunostimulatory properties, measured as
production of IL-6, a cytokine with pleiotropic activity in immune regulation, hematopoiesis,
Mar. Drugs 2019, 17, 103; doi:10.3390/md17020103
www.mdpi.com/journal/marinedrugs

Mar. Drugs 2019, 12 of 21
inflammation and oncogenesis [15]. A bioassay guided fractionation of the active extract on LH-20
resin afforded a fraction featured by low abundance polar lipid species exhibiting a molecular weight
(MW) between 1300 and 1600 amu in HR-ESI-MS. This mass range was significantly higher than MW
of major glycerol-based lipids typically reported from T. weissflogii as well as from other microalgae
and prompted us to further investigation [16–18]. Here we report molecular characterization,
synthesis of a structural analog and immunostimulatory properties of a novel pool of
phosphoglycolipids originally isolated from this marine diatom.
2. Results and Discussion
Preliminary NMR data on the bioactive LH-20 fraction from T. weissflogii extract suggested the
presence of a monoglycoglycerolipid backbone. A further purification step by HPLC was carried out
and a homogeneous bioactive fraction was subjected to spectroscopic (NMR) and spectrometric (ESI-
MS/MS) analyses. Spectral data, along with information from chemical degradation allowed to
characterize the complex pool as unusual phosphatidylmonogalactosyldiacylglycerols (PGDG, 1,
Figure 1).
Acyl
Acyl: EPA
O
C16:3
O
1''
C16:2
C16:1
C16:0
Acyl
O
3''
O
Gly2
P
HO
-O
O
6'
Acyl
O
O
4'
1'
O
O
1
HO
Acyl
OH
3'
3
-Galp
Gly1
1
Figure 1. Core structure of PGDG 1 from Thalassiora weissflogii. Main acyl chains are reported.
Structural characterization of this class of lipids was not straightforward due to the minute
amount recovered and severe overlapping of proton NMR signals in the region between δ 3.5 and 4.2
of the two glycerols and the sugar protons along with lack of corresponding NMR data reference in
the literature for this class of lipids. The monosaccharide residue was identified by both NMR data
(Table 1) and acidic degradation as a β-galactopyranose. In fact, by 2D NMR experiments an extended
spin system was reconstructed, which connected the anomeric proton at δ 4.23 (C1’, 104.5 ppm) to
methine protons at δ 3.53 (C2’, 71.9 ppm), 3.51 (C3’, 73.7 ppm), 3.99 (C4’, 68.3 ppm), 3.63 (C5’, 74.2
ppm) and to methylene at 4.09/3.92 (C6’, 63.4 ppm). In particular, the coupling constants J = 9.7 and
3.3 Hz measured for proton H-3’, were indicative of axial/axial and axial/equatorial orientations of
the vicinal protons H2’/H3’ and H3’/H4’, respectively. Furthermore the coupling constant between
H-1’ and H-2’ (J1,2 = 7.2 Hz) along with the chemical shift of the carbon C1’ (104.5 ppm) suggested a
β-configuration of the anomeric carbon of the sugar which was therefore assigned as β-
galactopyranose. Its identity was confirmed by ion exchange chromatography after TFA hydrolysis,
in comparison with authentic standards. The anomeric carbon at 104.5 ppm showed HMBC
connection through a glycosidic bond with the methylene protons of a glycerol residue at δ 3.99 and
3.68 (Gly1-C3, 68.0 ppm) which in turn were coupled with the oxymethine proton at δ 5.25 (Gly1-C2,
71.0 ppm) and the oxymethylene protons at δ 4.40 and 4.23 (Gly1-C1, 63.3 ppm). The downfield
chemical shifts of the glycerol protons at C1 and C2 were in agreement with the presence of acyls at
these positions.
However, HSQC data indicated the presence of another glycerol residue (Gly2), whose protons
resonated at almost the same values of Gly1 but distinguishable by the carbon chemical shifts values
at 63.4 (Gly2-C1’’), 71.2 (Gly2-C2’’) and 64.1 ppm (Gly2-C3’’). In particular, the 13C NMR signal

Mar. Drugs 2019, 13 of 21
centered at 71.2 ppm appeared clearly split in a doublet with a constant of 8.2 Hz likely due to a 3JC-P
coupling (Figure 2) [19]. Another doublet was observed for carbon C5’ of the galactose residue at 74.3
ppm (3JC-P = 8.2 Hz) thus suggesting that the second glycerol moiety was joined through a phosphate
group at Gly2-C3” to the C6’ of the sugar (Figure 2). As a confirmation, signal at δ 4.09 (H-6’a),
resonating in a non-crowded region of the protonic spectrum, showed a complex multiplicity due to
3JH-P coupling (Supporting Information, Figure S1) [20].
Table 1. 1H and 13C NMR data of natural (1) and synthetic PGDG (2) (600 MHz, CDCl3:CD3OD, 1:1).
1
2
1
2
13C,
1H, , m,
13C,
1H, , m,
J (Hz)
13C,
1H, , m ,
J (Hz)
13C,
1H, , m,
ppm
J (Hz)
ppm
ppm
ppm
J (Hz)
Gal
1’
Fatty acyl
C=O
174.1–
174.4
34.0–
34.8
174.2,
174.6
34.6,
34.8
104.5
71.8
73.7
68.3
74.3
63.5
4.23, d, 7.2
3.53, dd, 9.7, 7.2
3.51, dd, 9.7, 3.3
3.99, m
104.5
71.9
73.7
68.0
74.3
63.3
4.23, d, 7.1
2’
3’
4’
5’
3.53, dd, 9.8, 7.1
3.51, dd, 9.8, 3.2
3.99, brd, 3.2
α-CH2
β-CH2
2.34, m
2.33, m
1.61, m
1.27, m
25.1–
25.5
1.62–1.69, m
1.27–1.32, m
5.36, m
25.5
29.6–
30.3
29.7-
30.3
CH2 (chain)
HC=CH
128.4–
130.6
27.1–
27.7
3.63, dd, 8.4, 5.7
3.64, dd, 8.4, 5.8
6’a
6’b
Gly1
1a
4.09, m
3.92, m
4.08, m
3.93, m
=CHCH2CH2-
=CHCH2CH=
-CH2CH3
2.02–2.11, m
2.82–2.85, m
1.31, m
26.2
4.40, brd, 12.1
4.23, m
4.39, dd, 12.2. 2.9
4.24, m
23.2–
23.3
63.3
71.0
68.1
63.2
71.1
68.2
23.2
14.3
1.30, m
1b
2
5.25, m
5.25, m
=CHCH2CH2CH3
=CHCH2CH3
23.5
1.38, m
3a
3.99, m
3.96, dd, 10.7, 5.5
3.68, dd, 10.7, 5.3;
20.9
2.07, m
3b
3.68, dd, 10.5, 5.6;
14.0–
14.4
Gly2
-(CH2)nCH3
=CHCH2CH2CH3
=CHCH2CH3
0.88, t, 7.1
0.92, t, 7.2
0.97, t, 7.5
0.89, t, 7.1
1’’a
1”b
4.44, dd, 12.1, 3.0
4.20, m;
4.44, dd, 12.1, 3.2
4.21, m;
14.0–
14.4
63.4
63.3
14.0–
14.4
2’’
3’’
71.2
64.1
5.23, m
4.00, m
71.0
64.3
5.25, m
3.99, m
Figure 2. 13C NMR spectrum enlargement of 1 showing C-P couplings.

Mar. Drugs 2019, 14 of 21
According to deshielded 1H and 13C NMR resonances (Table 1) and LC-ESI-MS/MS analysis, both
glycerols were 1,2-acylated by fatty acids, whose relative composition was inferred by GC-MS
analysis of the hydrolyzed PGDG pool (Supporting Information, Table S1). Furthermore, detailed
fatty acids (FA) composition of main lipid species was inferred by extensive HR-ESI-MS/MS analysis
(Supporting information, Figure S19 and Table S2). The natural PGDG pool resulted mainly
composed by unsaturated fatty acids, including eicosapentaenoic acid (EPA) and analogs from the
C16 series, mainly C16:3 and C16:2.
As put forward for bacterial phosphatidylmonoglucosylglycerides, the origin of PGDG lipids
can be biosynthetically traced back to a transphosphatidylation reaction between a phosphatyl group
donor such as phosphatidylglycerol (PG) and a monogalactosyldiacylglycerol (MGDG) [21,22].
Indeed, the fatty acid profile of PGDG 1 (Supporting Information, Table S2) matches that of MGDG
and PG from marine diatoms and in particular from this species as already reported in our previous
paper, both highly rich in polyunsaturated fatty acids [16–18]. Following this line of reasoning, both
glycerol moieties should belong to the 1,2-diacyl-3-X-sn-glycerol series, reflecting the stereochemistry
of the biosynthetic precursors as depicted in Scheme 1. The fatty acid position on 1,2-glycerols could
not be established in our LC-MS conditions due to existence of isobaric species and poor
chromatographic resolution, which hampered specific fragmentation studies. Nevertheless, the
plastidial origin of both MGDG and PG precursors typically carrying C16 fatty acids at position 2
and EPA at position 1 of glycerol, as previously reported [16], suggests the same fatty acid relative
distribution on position 1 and 2 of glycerol moieties in PGDG lipids.
AcylO
OAcyl
AcylO
PG
OAcyl
O
O
P
-O
OH
O
P
O
O
O
transphosphatidylation
OH
-O
OH
+
HO
HO
+
OH
O
OAcyl
OAcyl
OH
OH
O
HO
O
OH
OAcyl
OAcyl
HO
O
OH
PGDG
MGDG
Scheme 1. Possible biogenetic origin of phosphatidylmonogalactosyldiacylglycerol (PGDG) from
monogalactosyldiacylglycerol (MGDG) and phosphatidylglycerol (PG) by trans-phosphatidylation.
In the literature, there are very few examples of bacterial galactose-based phosphoglycolipids
[9] and a unique report from marine bacteria of a glucose analog [11]; in all cases, these lipids bear
saturated fatty acids and have been characterized only by mass spectrometry and degradative
chemistry. Here we document the first finding, and provide full NMR structural data, of
galactopyranose-based phosphoglycolipids, decorated with polyunsaturated fatty acids, from
marine environment.
PGDG are very minor lipids in T. weissflogii compared to other related galactolipids like MGDG
but also with respect to other lipid classes, such as sulfoquinovosides, phospholipids and
triacylglycerols [16,23]. The biological function of PGDG in diatoms is still unknown: a structural role
in membranes or as storage lipids seems unlikely due to their low abundance, but cannot be ruled
out. Interestingly, we found an increased amount of these lipids in the late stationary-initial senescent
phase of the growth curve with respect to exponential/stationary phase (data not shown). Different
levels of metabolites during the cell growth may have a biological significance, implying for example
a signaling or regulatory role, or a stress response. The latter aspect is particularly invoked when an
increase of metabolite levels is observed in coincidence with reduced availability of nutrients as
occurs in the late phase of a culture growth. This observation opens intriguing perspectives to further
biological studies aiming at explore the PGDG function.

Mar. Drugs 2019, 15 of 21
In order to characterize the immunological properties of PGDG 1 we used an in vitro tool based
on monocyte-derived dendritic cells (moDCs) [24]. DCs are potent innate immune cells and represent
highly specialized antigen-presenting cells (APC) that play pivotal roles in the induction of protective
immunity against potentially harmful antigens [25]. DCs are characterized by an “immature” state
followed by maturation upon activation [25]. Maturation of moDCs was assessed by measuring the
expression of major histocompatibility complex (MHC)-Class II (HLA-DR), intercellular adhesion
molecule (ICAM, CD54), and the co-stimulatory molecules CD83 and CD86 by flow cytometry
[26,27]. The bioactive fraction from T. weissflogii extract was tested on moDCs in the range of 10–50
µg/mL and showed a saturating effect on upregulation of surface markers already at the lowest
concentration (Figure S20). We therefore selected 1–10 µg/mL as the best range to test the purified
PGDG pool. When assayed on human moDCs in this range, PGDG 1 displayed immunostimulatory
activity at 10 µg/mL inducing moDCs maturation (Figure 3a). All the phenotypic markers were
upregulated after treatment with PGDG 1. The compounds also elicited the mRNA overexpression
of interleukins IL-6, IL-8 and IL-12p40 from 5 µg/mL (Figure 3b). These results are not totally
surprising since glycoconjugates and in particular glycolipids are involved in cell-cell recognition
mechanisms and immunological response [28–31].
Figure 3. (a) Phenotypic surface markers analysis of monocyte-derived dendritic cells (moDCs): isotypic
control (light grey), control cells (ctrl) (dark grey), treated cells with PGDG (1) (pink); (b) Cytokine gene
expression upon stimulation of moDCs with PAM2CSK4 (PAM, 1 µg/mL) as positive control, with natural
PGDG (1) at concentrations of 1, 5 and 10 µg/mL compared to cells treated only with vehicle (ctrl). Data are
expressed as fold change +/− SD after 24 hours exposition. One representative experiment of two is shown.

Mar. Drugs 2019, 16 of 21
Asterisks indicate significant differences from ctrl; p < 0.001 by one-way ANOVA and the Tukey multiple
comparison test. ** p < 0.01, *** p < 0.001, **** p < 0.0001.
With the aim of exploring the molecular target and to get more insight into the immunological
potential of PGDGs, considering the low amount and the inherent instability of these natural
products, we developed a synthetic strategy for the preparation of a structural analog with two
stearoyl and two palmitoyl fatty acids onto the Gly1 and Gly2 glycerol subunits, respectively (2, Figure
4 and Scheme 2).
O
O
O
1''
O
Gly2
3''
O
O
P
O
-O
O
HO
4'
6'
O
O
1'
O
HO
1
3'
OH
3
O
Gly1
2
O
Figure 4. Chemical structure of synthetic PGDG 2.

Mar. Drugs 2019, 17 of 21
O
1-Trityl chloride (1.3 eq)
Pyridine-90°C/2h
2-Ac2O
OMe
OMe
HO
AcO
OAc
OP
OH
AcO
OTr
OH
P(OMe)3/pyridine
CBr4
HAc/HBr 33%
79%
O
O
O
O
58%
HO
AcO
AcO
AcO
84%
OH
OH
OAc
OAc
OAc OAc
II OAc OAc
I
III
D-galactose
CH2Cl2/BF3OEt2
-10°C
O
O
O
OMe
OMe
OMe
AcO
AcO
OMe
OMe
OP
OP
Thrichloroacetonitrile
DBU/CH2Cl2/0°-2h
OMe
HO
O
AcO
OP
BnNH2(1.5 eq)/THF
67%
O
O
O
O
O
O
O
86%
AcO
AcO
66%
AcO
OAc
V OAc
IV OAc
OH
O
VI
NH
Cl3C
O
O
OMe
OMe
OMe
OMe
Stearic acid (2.0 eq)
DMAP (1 eq)/DCC
(2.0 eq)
AcO
OP
AcO
OP
DOWEX H+
MeOH/H2O 95/5
O
O
OH
O
O
O
O
OH
AcO
Stearoyl
Stearoyl
AcO
94%
59%
OAc
OAc
VII
VIII
O
CCl3CN (10 eq)/ pyridine
75°C overnight
O
O
O
OH
OH
O
AcO
OP
AcO
OP
TMS-Cl (1.3 eq) NaI
(1.3 eq)/CH3CN
HO
O
OH
O
O
O
O
O
O
O
O
AcO
O
Stearoyl
Stearoyl
AcO
85%
Stearoyl
65%
OAc
OAc
IX
Stearoyl
X
O
OH
O
O
O
O
DOWEX H+
O
OH
MeOH/H2O/CHCl3
AcO
AcO
MeI/Ag2O/CH3CN
40°C overnight
OP
OP
OMe
OMe
O
97/3/30 - 60°C
O
O
O
O
O
O
O
66%
AcO
AcO
Stearoyl
Stearoyl
Stearoyl
53%
OAc
OAc
Stearoyl
XII
XI
O
Palmitoyl
O
Palmitoyl
O
Palmitoyl
O
Palmitoyl
O
Palmitic acid (1.5 eq)
DMAP (1 eq)/DCC
(1.5 eq) 45°C
O
O
O
AcO
AcO
OP
OP
TMS-Cl (2.6 eq) NaI
(2.0 eq)/CH3CN
OH
OMe
O
overnight
O
O
O
O
O
O
O
AcO
AcO
Stearoyl
Stearoyl
66%
68%
OAc
OAc
XIV Stearoyl
Stearoyl
XIII
1''
O
Palmitoyl
Palmitoyl
1
3''
O
O
3
O
HO
OP
OH
NH2NH2(H2O) (5 eq)
EtOH 85% 48°C
6'
O
O
O
O
HO
Stearoyl
Stearoyl
1'
44%
3'
OH
2
Scheme 2. Synthetic procedure for 3-O-[6’-O-(1’’,2’’-dipalmitoyl-3’’-phosphoryl-sn-glycerol)-1’-β-D-
galactopyranosyl]-1,2-distearoyl-sn-glycerol (2).
To the best of our knowledge only one chemical synthesis of phosphatidyl-β-glycosyl-
diacylglycerol was developed by Van Boeckel et al. [32] in the 80s; in the reported procedure the β-
glycosidic bond was formed by Koenigs-Knorr conditions using glucopyranosyl-bromide as
precursor and mercuric salts as catalysts, but problems linked to the stability of one intermediate
forced the authors to make the strategy longer and complicate. Here we adopted a new and versatile
strategy for the preparation of 2 (scheme 2). In particular, the phosphate group in 6’ position of the
sugar unit was introduced by trimethyl phosphite/carbon tetrabromide reaction of 1’,2’,3’,4’-tetra-O-
acetyl-galactose (II), prepared from commercial D-galactose; after deacetylation at anomeric carbon
of III with benzylamine and preparation of trichloroacetimidate derivative (V), the coupling with (S)-
1,2-O-isopropylidene glycerol aglycon led to 6’-O-dimethylphosphoryl-1’-β-D-2’,3’,4’-triacetyl-
galactopyranosyl-sn-isopropylidene-glycerol (VI); the choice of the acetate as hydroxyl protecting
group favored the formation of β-glycosidic linkage [33–35]. Hydrolysis of acetonide, condensation
with stearic acid and demethylation of phosphate group with trimethyl-silyl-chloride led to 6’-O-
phosphoryl-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl-1,2-distearoyl-sn-glycerol
(IX)
then
functionalized on phosphate group with (S)-1,2-O-isopropylidene glycerol by phosphate

Mar. Drugs 2019, 18 of 21
trichloroacetimidate activation to get 3-O-[6’-O-(3’’-phosphoryl-sn-isopropylidene-glycerol)-1’-β-D-
2’,3’,4’-triacetyl-galactopyranosyl]-1,2-distearoyl-sn-glycerol (X), subsequently methylated at
phosphate (XI). Deprotection of acetonide and condensation with palmitic acid gave 3-O-[6’-O-
(1’’,2’’-dipalmitoyl-3’’-methylphosphoryl-sn-glycerol)-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl]-1,2-
distearoyl-sn-glycerol (XIII) that was finally de-O-methylated by trimethyl-silyl-chloride and
deacetylated by hydrazine monohydrate [35] obtaining 3-O-[6’-O-(1’’,2’’-dipalmitoyl-3’’-phosphoryl-
sn-glycerol)-1’-β-D-galactopyranosyl]-1,2-distearoyl-sn-glycerol (2). NMR data of the final synthetic
product were superimposable to those of 1 (Table 1), thus corroborating the proposed core structure
together with glycerol moieties stereochemistry of 1.
We tested the activity of synthetic PGDG 2 on moDCs cells and we actually found an increase
of the effect on cells maturation at 10 µg/mL. The synthetic compound elicited a more remarkable
effect on the upregulation of characteristic moDCs markers, in particular on the co-stimulatory
molecules CD83 and CD86 (Figure 5a) as well as on IL12p40 gene expression although there was a
slight reduction on IL-6 and IL-8 expression with respect to the natural pool 1 (Figure 5b). In general,
upon activation of moDCs, the production of IL-8 and IL-6 shows a faster kinetic rate than IL-12 and
the up-regulation of co-stimulatory molecules, and can be already observed after 2 hours from
stimulation [36–38]. We may interpret the different expression of IL-6 and IL-8 induced by the two
stimulants 1 and 2 as the result of different kinetics, with a faster, more potent effect of the synthetic
analog 2, acting earlier in the moDCs assay with a consequent anticipated decline, as observed at 24
hours, of the expression level of the above interleukins compared to natural PGDG 1. The differences
in the stimulation profile could be due to the saturated acyl residues in PGDG 2 which may modulate
inflammatory gene expression in moDCs to a different extent with respect to unsaturated acyl fatty
acids of PGDG 1 and/or to different molecular targets involved in the mechanism of
immunostimulation [38].
The effect of many immunostimulant lipids exhibiting phosphate groups on acylated
aminosugars in their molecular architecture is mediated by Toll-like receptors (TLRs) [39,40]. Among
these lipids, bacterial lipopolysaccharide (LPS) and its synthetic derivative monophosphoryl lipid A
(MPLA) are the most widely studied agonists of TLR-4 [41]. Glycerol-based lipids with phosphate
and carbohydrate in their structure represent another group of lipids for which an immunomodulant
activity has been analyzed, even though only in a limited number of cases [12,14,42,43]. The best-
known member of this group is lipotheicoic acid (LTA), composed by a glycoglycerolipid bound to
a hydrophilic polymer of glycerophosphate, for which a specific TLR-2 agonistic activity has been
documented [12,43].
To verify whether PGDG 1 and 2 activity were TLR-2/TLR-4 mediated, we tested their effects in
a cell-based assay with HEK-293 cells stably co-transfected to express full-length human TLR-2 or
TLR-4 and the secreted alkaline phosphatase (SEAP) reporter gene under the transcriptional control
of a NF-kB response element [44]. The activity of these compounds is independent from TLR-2
induction. In fact, when both molecules were assayed for TLR-2 linked NF-kB SEAP activity, results
were negative (Figure 6a). On the other hand, as shown in Figure 6b, natural PGDG 1 activated TLR-
4 reporter cells at concentrations that stimulate maturation and cytokine expression in moDCs (10
µg/mL). Synthetic PGDG 2 has a comparable effect at the lowest concentrations, but doubled the
SEAP production at 10 µg/mL.

Mar. Drugs 2019, 19 of 21
Figure 5. (a) Phenotypic surface markers analysis of moDCs: isotypic control (light grey), control cells
(dark grey), treated cells with PGDG (2) (pink); (b) Cytokine gene expression upon stimulation of
moDCs with PAM2CSK4 (PAM, 1 µg/mL) as positive control, with synthetic PGDG (2) at
concentrations of 1, 5 and 10 µg/mL compared to cells treated only with vehicle (ctrl). Data are
expressed as fold change +/− SD after 24 hours exposition. One representative experiment of two is
shown. Asterisks indicate significant differences from the ctrl; p < 0.001 by one-way ANOVA and the
Tukey multiple comparison test. * p < 0.5, ** p < 0.01, *** p < 0.001, **** p < 0.0001.
Figure 6. (a) TLR-2 NF-kB/SEAP reporter cell line assay with increasing concentrations of natural
PGDG 1 and synthetic PGDG 2; PAM2CSK4 (PAM) was used as positive control; (b) TLR-4 activation

Mar. Drugs 2019, 110 of 21
by increasing concentrations of natural PGDG 1 and synthetic PGDG 2 in TLR-4 NF-kB/SEAP reporter
cell lines; LPS was used as positive control. Asterisks indicate significant differences from the cells
treated only with vehicle (ctrl); p < 0.001 by one-way ANOVA and the Tukey multiple comparison
test. * p < 0.05, ** p < 0.01, **** p < 0.0001.
To confirm the involvement of TLR-4 receptor, we applied a neutralizing antibody to block TLR-
4 on moDCs. We used an anti TLR-4 antibody previously reported as blocking agent in in vitro
experiments [45], to evaluate the effect after treatment with synthetic PGDG 2 by measuring CD83
on moDCs surface. High expression of this marker is a peculiarity of fully matured human DCs and
critical for the in vivo priming of naïve T cells[46]. The concentration of 1 µg/mL was estimated as the
dose of the antibody to cover all TLR-4 sites on moDCs surface. As shown in Figure 7a, in comparison
with the stimulation with the phosphoglycolipid alone, CD83 surface expression of cells stimulated
with PGDG 2 after treatment with TLR-4 antibody is nearly null. The residual effect could be due to
partial blockage or to other mechanisms involving PGDG immunostimulation. Therefore we
evaluated the upregulation of IL-12p40 and IFN- gene expression as more characteristic cytokines
of DCs activation after TLR ligation [47]. The overexpression of both mediators disappeared in cells
treated with PGDG and TLR-4 antibody (Figure 7b).
Figure 7. (a) moDCs CD83 surface expression: control cells (grey), treated cells (pink); (b) cytokine
gene expression upon stimulation of moDCs with PGDG 2 at 10 µg/mL or with ab anti-TLR4 (AB) at
1 µg/mL, compared to the cells treated only with vehicle (ctrl). Data are expressed as fold change +/−
SD after 24 hours exposition. One representative experiment of two is shown. Asterisks indicate
significant differences from the cells treated only with vehicle (ctrl); p < 0.001 by one-way ANOVA
and the Tukey multiple comparison test. ** p < 0.01, *** p < 0.001.
Antigen presentation function of DCs is tightly regulated by the maturation process. A
functional assay was set up to investigate the immunological properties of 2 in improving DCs
antigen presentation and T cell activation. To this aim, mouse bone marrow-derived dendritic cells

Mar. Drugs 2019, 111 of 21
(BM-DCs) were incubated for 3 hours with the synthetic peptide OVA257–264 in presence of 2 or LPS or
MPLA as maturation stimuli. DCs were then assessed for their capability to triggering activation of
OTI (OVA257–264 specific) hybridoma T cells B3Z, by evaluating interleukin-2 (IL-2) release in the
culture supernatants in response to OVA257–264. B3Z CD8+ T cell line is known to produce high levels
of IL-2 exclusively upon activation of T cell receptor specific for the OVA257–264 SIINFEKL peptide in
the context of H-2Kb MHC class I complex expressed on antigen presenting cells [48].
We found that synthetic PGDG 2 at 1 µg/mL significantly improves DCs capability of activating
an antigen-specific T cell response, especially when low doses (0.004 and 0.04 µM) of OVA257–264
peptide were used. In particular, at 0.04 µM OVA257–264 concentration, the IL-2 release by B3Z cells
stimulated with PGDG-matured DCs yet reached a plateau with respect to LPS or MPLA matured
DCs (Figure 8).
Figure 8. IL-2 release of mouse OTI hybridoma cell line B3Z in response to OVA257–264 SIINFEKL
peptide. Bone marrow-derived dendritic cells (BM-DCs) were incubated with graded doses of
synthetic OVA257–264 peptide in combination with 1µg/mL LPS, MPLA or 2 and co-cultured with B3Z
hybridoma cells for 40 hours. Supernatants were assayed in duplicate. Mean ± SEM is reported, one
representative experiment of two is shown. p < 0.001 by two-way ANOVA and the Bonferroni
multiple comparison test. ** p < 0.01, *** p < 0.001.
Microalgal PGDG emerged as a novel class of TLR-4 agonists. Since structurally related bacterial
phosphatidylglycosylglycerides with phosphate at carbon 6’ of the sugar have never been tested for
their immunomodulant potential or, more in general, for any biological activity [6,11,21], this
represents
the
first
report
on
immunostimulant
activity
of
bioactive
phosphatidylmonoglycosyldiacylglycerol lipids. Based on the chemical structure and similarity with
LPS and LTA it is hard to find any structure/activity relationship suggesting that many chemical
features can modulate ligand-receptor interaction. One of these is represented by fatty acyl
composition that is likely responsible of the more potent effect of synthetic (C16:0/C18:0) PGDG 2
versus the natural pool of PGDG 1 rich in C20/C16 unsaturated fatty acids [49].
3. Materials and Methods
3.1. General
NMR spectra were recorded in CDCl3:CD3OD (1:1) on a 600 MHz Bruker Avance III
spectrometer (Bruker BioSpin GmbH, Rheinstetten, Germany) equipped with a CryoProbeTM at
600.15 MHz. Chemical shifts values are reported in ppm (δ) and referenced to internal signals of
residual protons (CD3OD 1H δ 3.34, 13C 49.0 ppm). High resolution mass spectra were acquired on a
Q-Exactive Hybrid Quadrupole-Orbitrap Mass Spectrometer (Thermo Scientific, Milan, Italy) on-line
with a UHPLC apparatus Infinity 1290 (Agilent Technologies, Santa Clara, CA, USA). HPLC analyses
were performed on a Jasco system (PU-2089 Plus-Quaternary Gradient Pump equipped with a Jasco
MD-2018 Plus Photodiode Array Detector (Jasco, Cremella, Italy). Solid Phase Extraction (SPE) was

Mar. Drugs 2019, 112 of 21
carried out using polystyrene-divinyl benzene resin (CHROMABOND® HR-X, Macherey-Nagel,
Düren, Germany). GCMS analysis was performed by an ion-trap MS instrument in EI mode (70eV)
(Thermo, Polaris Q) connected with a GC system (Thermo, GCQ) by a 5% phenyl/methyl
polysiloxane column (30m x 0.25mm x 0.25µm, Agilent, VF-5ms) using Helium as gas carrier. Sugar
analysis was achieved on a HPAE-PAD (Dionex-Thermo Fisher Scientific, Milan, Italy). Chemicals
were of analytical reagent grade and solvents of HPLC grade (Sigma-Aldrich, Milan, Italy) and were
used without any further purifications. The synthetic peptide OVA257–264 (SIINFEKL) was purchased
from Primm (Naples, Italy). Lipopolysaccharides from E. coli (serotype 055:B5) and MPLA were from
Sigma-Aldrich.
3.2. Culture Condition
Thalassiosira weissflogii (Grun.) Fryxell et Hasle (CCMP 1336) was purchased from National
Center for Marine Algae and Microbiota (Bigelow Laboratory for Ocean Sciences, Maine, USA).
The axenic strain was grown in a 10 L polycarbonate carboys, in f/2 culture medium, at room
temperature (22.0 ± 1.0 °C) and gently bubbled with sterile air. The initial cell density was about 10000
cells mL−1, and the inoculum was taken from 1 L of healthy exponentially growing culture. Artificial
light intensity (200 µmol m-2 s-1) was provided by daylight fluorescent tubes with a 14 h:10 h light:
dark photoperiod.
Cell growth was estimated by daily using a Burker counting chamber 0,1 mm depth (Merck,
Leuven, Belgium;) under an inverted microscope AxioVision SE64 (Carl Zeiss, Thornwood, NY,
USA).
Microalgae were harvested in the late stationary/beginning senescent phase by centrifugation in
a swing out Allegra 12XR (Beckman Coulter Inc., Palo Alto, CA, USA) at 2300 g for 10 minutes, and
stored at −80 °C until analyses.
3.3. Extraction, Fractionation and Identification of PGDG (1)
The cell pellet of T. weissflogii (10.2 g) obtained as above described was extracted with methanol
(3 × 50 mL), sonicated and centrifuged to remove cell debris. The methanol phase was filtered through
paper and concentrated under vacuum without heating by rotavapor (R-300, Buchi Italia, Cornaredo,
Italy). The raw extract (550 mg) was fractionated on LH-20 resin eluted with CHCl3:CH3OH 1:1
affording 13 samples A-O. Fraction B (15 mg) was further purified by HPLC on a Luna Phenyl-hexyl
column (Phenomenex, 250 × 10 mm, 5µm, 3 mL/min), by CH3OH/H2O elution gradient from 80% to
100% CH3OH in 10 min, holding at 100% for 10 min. PGDG 1 (2 mg) was isolated as an unresolved
pool of species eluting in the range Tr = 14–15.8 min.
PGDG (1). Yellowish amorphous solid. NMR data see Table 1. HR-ESI MS data see Supporting
Information Figure S19 and Table S1; Tables S2 for Fatty acid analysis.
3.4. Fatty Acid Analysis
200 µg of PGDG pool (1) were dissolved in MeOH (0.5 mL), treated with Na2CO3 and heated to
53 °C for 20 h. The reaction mixture was diluted with a solution of HCl 0.2 M and extracted with Et2O
(3 × 2 mL); the organic solvent was evaporated under reduced pressure and then analyzed in GCMS
by using the following gradient: Initial 160 °C holding for 3 min; then 5 °C/min up to 260 °C followed
by 30 °C/min up to 310 °C, holding for 3 min at 310 °C. Inlet temp 260 °C; Split flow 10 mL/min. Full
scan m/z 50–450. Results are reported in Supporting Information, Table S1.
3.5. LC-ESI-/MS/MS Analysis
LC-MS/MS analysis was carried out according to [18]. Briefly, 200 µg of PGDG pool (1) were
dissolved in MeOH (1 mL) and analysed in negative ionization mode in the mass range 1000–1800
m/z on a Q-Exactive mass spectrometer coupled to a UHPLC system Infinity 1290 equipped with a
Kinetex Biphenyl 2.6 µm, 150 × 2.1 mm column, (Phenomenex, Castel Maggiore, Bologna, Italy).
Eluent A: water and eluent B: MeOH. The elution program consisted of a gradient from 40% to 80%

Mar. Drugs 2019, 113 of 21
B in 2 min, then to 100% B in 13 min, holding at 100% B for 7 min. Flow rate was 0.3 mL/min. The
injection volume was 10 μL, the autosampler was maintained at 10°C and the column at 28 °C. Results
are reported in Supporting Information, Table S2.
3.6. Monosaccharide Analysis
The aqueous phase from previous extraction was transferred in a reaction vial and subjected to
hydrolysis with TFA 6 M, at 110 °C for 4 h. The reaction mixture was evaporated under nitrogen
stream, redissolved with distilled water and directly analysed by High Pressure Anion Exchange
column (Carbopac PA1) connected with a Pulsed Amperometric Detector (HPAE-PAD, eluting with
18 mM NaOH, at 0.25 mL/min. Sugar identification was achieved by co-injection with commercial
standards.
3.7. Synthetic Procedure for PGDG 2
6’-O-trityl-D-1’,2’,3’,4’-Tetracetyl-galactopyranose (I): D-galactose (0.06 g, 0.334 mmol) was
dissolved in pyridine (0.6 mL); 1.3 equivalents of trityl chloride (0.121 g, 0.434 mmol) was added
slowly in different portions; the reaction mixture was stirred for 2 h at 90 °C; subsequently, acetic
anhydride (0.3 mL) was added and after stirring overnight the mixture was evaporated under a
stream of nitrogen and purified by silica gel chromatography using a light petroleum ether/diethyl
ether gradient to give 6’-O-trityl-D-1’,2’,3’,4’-tetracetyl-galactopyranose (I) (0.165 g, 0.28 mmol, 84%);
1
Rf (light petroleum ether/diethyl ether 3:7) = 0.49; H-NMR (400 MHz, CDCl3): (mix of - and -
anomers)  7.52-7.12 (overlapped, aromatic protons, 15H), 6.39 (bs, 1H), 5.71 (d, J= 8.2 Hz, 1H), 5.62
(bs, 1H), 5.53 (m, 1H), 5.39 (m, 1H), 5.28 (m, 1H), 5.09 (m, 1H), 4.32 (m, 1H), 3.39 (m, 1H), 3.18 (m, 1H)
2.31-1.98 (overlapped, acetate methyls); HRESIMS m/z 613.2073 [M+Na]+ (calcd for C33H34O10Na+,
613.2044).
D-1’,2’,3’,4’-Tetracetyl-galactopyranose (II): 6’-O-trityl-D-1’,2’,3’,4’-tetracetyl-galactopyranose (I)
(0.165 g, 0.28 mmol) was dissolved in acetic acid (0.64 mL) at 10 °C and bromidric acid (33% in acetic
acid, 0.056 mL) was added; after one minute of stirring the mixture was filtered and washed with
water (0.4 mL); the filtrate was diluted with 2 mL of water and extracted three times with
dichloromethane (1.3 mL). The organic phase was purified by silica gel chromatography using a light
petroleum ether/diethyl ether gradient to give D-1’,2’,3’,4’-tetracetyl-galactopyranose (II) (0.077 g,
0.22 mmol 79%); Rf (light petroleum ether/diethyl ether 3:7) = 0.20; 1H-NMR (400 MHz, CDCl3) (mix
of - and -anomers): 6.31 (bs, 1H), 5.68 (d, J= 7.7 Hz, 1H), 5.41 (bd, J= 2.9 Hz, 1H), 5.31 (m, 1H), 5.12
(m, 1H), 3.69 (m, 1H), 3.49 (m, 1H), 2.16-1.91 (overlapped, acetate methyls); HRESIMS m/z 371.0957
[M+Na]+ (calcd for C14H20O10Na+, 371.0949).
6’-O-dimethylphosphoryl-D-1’,2’,3’,4’-Tetracetyl-galactopyranose (III): D-1’,2’,3’,4’-tetracetyl-
galactopyranose (II) (0.077 g, 0.22 mmol) was dissolved in pyridine (0.28 mL); trimethylphosphite
(0.033 mL, 0.288 mmol) and carbon tetrabromide (0.084 g, 0.266 mmol) were added at 0°C. The
mixture was stirred in argon atmosphere for 2 hours at room temperature. The solvent was
evaporated and the residue was purified by by silica gel chromatography using a chloroform
/methanol gradient to give 6’-O-dimethylphosphoryl-D-1’,2’,3’,4’-tetracetyl-galactopyranose (III)
1
(0.059 g, 0.128 mmol 58%); Rf (diethyl ether) = 0.15; H-NMR (400 MHz, CDCl3) (mix of - and -
anomers): 6.34 (bs, 1H), 5.68 (d, J= 7.8 Hz, 1H), 5.4 (bd, J= 2.7 Hz, 1H), 5.29 (m, 1H), 5.08 (m, 1H),
4.15 (m, 1H), 4.10-4.02 (m, 2H), 3.86-3.70 (overlapped, phosphate O-methyls), 2.19-1.95 (overlapped,
acetate methyls); HRESIMS m/z 479.0926 [M+Na]+ (calcd for C16H25O13PNa+, 479.0925).
6’-O-dimethylphosphoryl-D-2’,3’,4’-triacetyl-galactopyranose (IV): 6’-O-dimethylphosphoryl-
D-1’,2’,3’,4’-tetracetyl-galactopyranose (III) (0.059 g, 0.128 mmol) was dissolved in tetrahydrofurane
(0.36 mL); benzylamine (0.0176 mL, 0.192 mmol) was added; the mixture was stirred overnight at
room temperature and after evaporation of solvent was purified by silica gel chromatography using
a
chloroform/methanol
gradient
to
give
6’-O-dimethylphosphoryl-D-2’,3’,4’-triacetyl-
galactopyranose (IV) (0.0356 g, 0.086 mmol 67%); Rf (chloroform:methanol 95:5) = 0.49; 1H-NMR (400
MHz, CDCl3) (mix of - and -anomers): 5.49 (bd, J= 3.0 Hz, 1H), 5.43 (bs, 1H), 5.13 (d, J= 7.7 Hz,
1H), 5.29 (m, 1H), 5.08 (m, 1H), 4.95 (m, 1H), 4.51 (m, 1H), 4.12-4.01 (m, 2H), 3.88–3.69 (overlapped,

Mar. Drugs 2019, 114 of 21
phosphate O-methyls), 2.20-1.95 (overlapped, acetate methyls); HRESIMS m/z 437.0832 [M+Na]+
(calcd for C14H23O12PNa+, 437.0819).
6’-O-dimethylphosphoryl-1’--D-2’,3’,4’-triacetyl-galactopyranosyl-trichloroacetimidate (V): 6’-
O-dimethylphosphoryl-D-2’,3’,4’-triacetyl-galactopyranose (IV) (0.0356 g, 0.086 mmol) was dissolved
in dichloromethane (0.38 mL); 1,8-Diazabicyclo[5.4.0]undec-7-ene (DBU) (0.00266 mL, 0.0172 mmol)
and trichloroacetonitrile (0.087 mL, 0.86 mmol) were added and the mixture was stirred on molecular
sieves for 2 h at 0 °C; after evaporation of solvent the mixture was purified by silica gel
chromatography using a dichloromethane / methanol gradient to give 6’-O-dimethylphosphoryl-1’-
β-D-2’,3’,4’-triacetyl-galactopyranosyl-trichloroacetimidate (V) (0.0412 g, 0.074 mmol 86%); Rf
(chloroform:methanol 95:5) = 0.70; 1H-NMR (400 MHz, CDCl3): s, NH), 6.59 (bd, J= 3.7 Hz, 1H),
5.58 (bd, J = 2.3 Hz, 1H), 5.40 (dd, J= 2.4, 10.6 Hz, 1H), 5.33 (dd, J = 3.7, 10.6 Hz, 1H), 4.42 (m, 1H),
4.12-3.96 (m, 2H), 3.77-3.68 (overlapped, phosphate O-methyls, 6H), 2.21–1.96 (acetate methyls, 9H);
HRESIMS m/z 579.9928 [M+Na]+ (calcd for C16H23Cl3NO12PNa+, 579.9916).
6’-O-dimethylphosphoryl-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl-sn-isopropylidene-glycerol
(VI): 6’-O-dimethylphosphoryl-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl-trichloroacetimidate (V)
(0.0412 g, 0.074 mmol) was dissolved in anhydrous dichloromethane (0.44 mL) and 1.5 equivalents of
1,2-(S)-O-isopropylidene glycerol was added; the reaction mixture was kept at -20 °C on activated 3
Å molecular sieves under argon; boron trifluoride etherate (0.00254 mL, 0.0103 mmol) was added and
stirring was maintained for 2 h; after another addition of boron trifluoride etherate (0.00254 mL,
0.0103 mmol), the temperature was raised to −10 °C and the reaction mixture was stirred overnight
at that temperature. After neutralization with triethylamine (4L) and filtration on celite, the
mixture was evaporated under reduced pressure and purified by silica gel chromatography using a
dichloromethane/methanol gradient to give 6’-O-dimethylphosphoryl-1’-β-D-2’,3’,4’-triacetyl-
galactopyranosyl-sn-isopropylidene-glycerol (VI) (0.0258 g, 0.048 mmol, 66%);
Rf
(chloroform:methanol 95:5) = 0.65; 1H-NMR (400 MHz, CDCl3):  (bd, J = 3.3 Hz, 1H), 5.14 (dd, J
= 8.04, 10.5 Hz, 1H), 4.97 (dd, J = 3.3, 10.5 Hz, 1H), 4.55 (d, J = 8.0 Hz, 1H), 4.34 (m, 1H), 4.14–3.78 (m,
7H), 3.75–3.68 (overlapped, phosphate O-methyls, 6H), 2.11 (s, acetate methyl, 3H), 2.02 (s, acetate
methyl, 3H), 1.93 (s, acetate methyl, 3H), 1.36 (s, acetonide methyl, 3H), 1.29 (s, acetonide methyl, 3H);
HRESIMS m/z 551.1512 [M+Na]+ (calcd for C20H33O14PNa+, 551.1500).
6’-O-dimethylphosphoryl-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl-sn-glycerol (VII): 6’-O-
dimethylphosphoryl-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl-sn-isopropylidene-glycerol
(VI)
(0.0258 g, 0.048 mmol) was dissolved in methanol-water solution (0.66 mL, 9/1) and Dowex 50WX8-
H+ resin (0.4 g) was added; after 3 h under stirring, the reaction mixture was filtered, evaporated and
purified by silica gel chromatography using a gradient of dichloromethane/methanol to give 6’-O-
dimethylphosphoryl-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl-sn-glycerol (VII) (0.0138 g, 0.028
mmol, 59%); Rf (chloroform:methanol 95:5) = 0.30; 1H-NMR (400 MHz, CDCl3):  (bd, J = 3.1 Hz,
1H), 5.20 (dd, J = 7.4, 10.1 Hz, 1H), 5.03 (dd, J = 3.2, 10.1 Hz, 1H), 4.53 (d, J = 7.4 Hz, 1H), 4.17–4.10 (m,
1H), 4.08–4.01 (m, 1H), 3.97 (m, 1H), 3.88–3.84 (m, 2H), 3.79–3.75 (overlapped, phosphate methyls,
6H), 3.70 (m, 2H), 3.59 (dd, J = 4.8, 11.4 Hz, 1H), 2.16 (s, acetate methyl, 3H), 2.07 (s, acetate methyl,
3H), 1.98 (s, acetate methyl, 3H); HRESIMS m/z 511.1197 [M+Na]+ (calcd for C17H29O14PNa+, 511.1187).
6’-O-dimethyl-phosphoryl-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl-1,2-distearoyl-sn-glycerol
(VIII): 6’-O-dimethylphosphoryl-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl-sn-glycerol (VII) (0.0138
g, 0.028 mmol) was dissolved in anhydrous dichloromethane (0.4 mL); stearic acid (0.032 g, 0.112
mmol), dicyclohexylcarbodiimide (0.0232 g, 0.112 mmol) and DMAP (0.0069 g, 0.028 mmol) were
added under argon; the reaction mixture was stirred overnight at room temperature; after
evaporation under reduced pressure, the mixture was purified by silica gel chromatography using a
gradient of dichloromethane/methanol to give 6’-O-dimethyl-phosphoryl-1’-β-D-2’,3’,4’-triacetyl-
galactopyranosyl-1,2-distearoyl-sn-glycerol (VIII) (0.0264 g, 0.026 mmol, 94%);
Rf
(chloroform:methanol 95:5) = 0.80; 1H-NMR (400 MHz, CDCl3):  (bd, J = 2.6 Hz, 1H), 5.19–5.14
(overlapped, 2H), 5.02 (dd, J = 3.8, 10.5 Hz, 1H), 4.48 (d, J = 7.3 Hz, 1H), 4.13 (dd, J = 2.7, 11.7 Hz, 1H),
4.16–4.07 (m, 3H), 3.95 (m, 1H), 3.75 (d, J =11.4 Hz, phosphate methyls, 6H), 3.65 (dd, J = 5.7, 11.7Hz,
1H), 2.30 (t, J = 7.1 Hz,-methylenes of acyl portions, 4H), 2.16 (s, acetate methyl, 3H), 2.06 (s, acetate

Mar. Drugs 2019, 115 of 21
methyl, 3H), 1.98 (s, acetate methyl, 3H), 1.61 (overlapped,-methylenes of acyl portions, 4H), 1.37–
1.20 (m, aliphatic protons), 0.88 (t, J=7.0 Hz, methyls, 6H); HRESIMS m/z 1043.6418 [M+Na]+ (calcd for
C53H97O16PNa+, 1043.6406).
6’-O-phosphoryl-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl-1,2-distearoyl-sn-glycerol (IX): 6’-O-
dimethyl-phosphoryl-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl-1,2-distearoyl-sn-glycerol
(VIII)
(0.0264 g, 0.026 mmol) was dissolved in acetonitrile (1.36 mL); TMS-Cl (0.008 mL, 0.062 mmol) and
sodium iodide (0.0097 g, 0.068 mmol) were added and after stirring in the dark at room temperature
overnight, the mixture was filtered and washed with dichlorometane; the filtrate was evaporated to
get 6’-O-phosphoryl-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl-1,2-distearoyl-sn-glycerol (IX) (0.022 g,
0.022 mmol, 85%); Rf (chloroform:methanol 85:15) = 0.10; 1H-NMR (400 MHz, CDCl3):  (bs, 1H),
5.22–5.15 (overlapped, 2H), 5.02 (bd, J = 10.3 Hz, 1H), 4.50 (d, J = 7.1 Hz, 1H), 4.13–3.92 (overlapped,
5H), 3.68 (m, 1H), 2.30 (t, J = 7.1 Hz,-methylenes of acyl portions, 4H), 2.16 (s, 3H, acetate methyl),
2.05 (s, 3H, acetate methyl), 1.98 (s, 3H, acetate methyl), 1.66–1.54 (overlapped,-methylenes of acyl
portions, 4H), 1.36–1.18 (m, aliphatic protons), 0.88 (t, J = 6.8 Hz, methyls, 6H); HRESIMS m/z 991.6127
[M−H]− (calcd for C51H92O16P−, 991.6129).
3-O-[6’-O-(3’’-phosphoryl-sn-isopropylidene-glycerol)-1’-β-D-2’,3’,4’-triacetyl-
galactopyranosyl]-1,2-distearoyl-sn-glycerol
(X):
6’-O-phosphoryl-1’-β-D-2’,3’,4’-triacetyl-
galactopyranosyl]-1,2-distearoyl-sn-glycerol (IX) (0.022 g, 0.022 mmol) and 1,2-(S)-O-isopropylidene
glycerol (0.0058 g, 0.044 mmol) were dissolved in small quantities of pyridine and evaporated
different times by nitrogen flow and finally were dissolved in anhydrous pyridine (0.3 mL) and
trichloroacetonitrile (0.0226 mL, 0.22 mmol) was added; after stirring at 75 °C under argon
atmosphere overnight, the mixture was evaporated and purified by silica gel chromatography using
a gradient of dichloromethane/methanol to give 3-O-[6’-O-(3’’-phosphoryl-sn-isopropylidene-
glycerol)-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl]-1,2-distearoyl-sn-glycerol (X) (0.0158 g, 0.0143
mmol, 65%); Rf (chloroform:methanol 85:15) = 0.25; 1H-NMR (400 MHz, MeOD):  (bs, 1H), 5.24
(m, 1H), 5.16-5.09 (overlapped, 2H), 4.70 (d, J = 7.0 Hz, 1H), 4.38-4.23 (overlapped, 2H), 4.15 (dd, J =
7.0, 12.2 Hz, 1H), 4.15–4.10 (overlapped, 2H), 4.02 (m, 1H), 3.96–3.89 (overlapped, 3H), 3.84 (m, 1H),
3.75 (m, 1H), 2.39–2.31 (overlapped,-methylenes of acyl portions, 4H), 2.19 (s, 3H, acetate methyl),
2.09 (s, 3H, acetate methyl), 1.99 (s, 3H, acetate methyl), 1.68–1.60 (overlapped,-methylenes of acyl
portions, 4H), 1.43 (s, methyl of acetonide, 3H), 1.37 (s, methyl of acetonide, 3H), 1.34–1.15 (m,
aliphatic protons), 0.93 (t, J = 6.9 Hz, methyls, 6H); HRESIMS m/z 1105.6798 [M−H]− (calcd for
C57H102O18P−, 1105.6809).
3-O-[6’-O-(3’’-methylphosphoryl-sn-isopropylidene-glycerol)-1’-β-D-2’,3’,4’-triacetyl-
galactopyranosyl]-1,2-distearoyl-sn-glycerol (XI): 3-O-[6’-O-(3’’-phosphoryl-sn-isopropylidene-
glycerol)-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl]-1,2-distearoyl-sn-glycerol (X) (0.0158 g, 0.0143
mmol) was dissolved in anhydrous acetonitrile (0.54 mL); silver (I) oxide (0.00692 g, 0.03 mmol) and
iodomethane (0.0286 mL, 0.46 mmol) were added; after stirring at 40°C overnight, the mixture was
evaporated and purified by silica gel chromatography using a gradient of dichloromethane/methanol
to
give
3-O-[6’-O-(3’’-methylphosphoryl-sn-isopropylidene-glycerol)-1’-β-D-2’,3’,4’-triacetyl-
Rf
galactopyranosyl]-1,2-distearoyl-sn-glycerol (XI) (0.0085 g, 0.00758 mmol, 53%);
(chloroform:methanol 85:15) = 0.85; 1H-NMR (400 MHz, CDCl3):  (bd, J = 2.2 Hz, 1H), 5.22-5.16
(overlapped, 2H), 5.01 (m, 1H), 4.48 (d, J = 7.5 Hz, 1H), 4.35-4.28 (overlapped, 2H), 4.16–4.02
(overlapped, 4H), 3.99–3.92 (overlapped, 2H), 3.82-3.72 (overlapped, 7H), 2.30 (t, J= 7.1 Hz,-
methylenes of acyl portions, 4H), 2.16 (s, 3H, acetate methyl), 2.06 (s, 3H, acetate methyl), 1.98 (s, 3H,
acetate methyl), 1.64–1.56 (overlapped,-methylenes of acyl portions, 4H), 1.43 (s, methyl of
acetonide, 3H), 1.36 (s, methyl of acetonide, 3H), 1.33–1.21 (m, aliphatic protons), 0.88 (t, J = 7.0 Hz,
methyls, 6H); HRESIMS m/z 1143.6910 [M+Na]+ (calcd for C58H105O18PNa+, 1143.6931).
3-O-[6’-O-(3’’-methylphosphoryl-sn-glycerol)-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl]-1,2-
distearoyl-sn-glycerol (XII): 3-O-[6’-O-(3’’-methylphosphoryl-sn-isopropylidene-glycerol)-1’-β-D-
2’,3’,4’-triacetyl-galactopyranosyl]-1,2-distearoyl-sn-glycerol (XI) (0.0085 g, 0.00758 mmol) was
dissolved in methanol-water-chloroform suspension (0.72 mL, 97/3/30) and Dowex 50WX8-H+ resin
(0.17 g) was added; after 7 h under stirring at 60 °C, the reaction mixture was filtered, evaporated and

Mar. Drugs 2019, 116 of 21
purified by silica gel chromatography using a gradient of dichloromethane/methanol to give 3-O-[6’-
O-(3’’-methylphosphoryl-sn-glycerol)-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl]-1,2-distearoyl-sn-
glycerol (XII) (0.0052 g, 0.005 mmol, 66%); Rf (chloroform:methanol 95:5) = 0.45; 1H-NMR (400 MHz,
CDCl3):  (bs, 1H), 5.24–5.14 (overlapped, 2H), 5.02 (m, 1H), 4.50 (d, J = 7.9 Hz, 1H), 4.31 (dd, J =
3.0, 11.9 Hz), 4.17-4.09 (overlapped, 3H), 4.06 (m, 1H), 3.99–3.91 (overlapped, 2H), 3.84–3.74
(overlapped, 6H), 3.71–3.63 (overlapped, 3H), 2.30 (t, J = 7.2 Hz,-methylenes of acyl portions, 4H),
2.18 (s, 3H, acetate methyl), 2.06 (s, 3H, acetate methyl), 1.99 (s, 3H, acetate methyl), 1.68–1.53
(overlapped,-methylenes of acyl portions, 4H), 1.36–1.27 (m, aliphatic protons), 0.88 (t, J = 6.8 Hz,
methyls, 6H); HRESIMS m/z 1103.6619 [M+Na]+ (calcd for C55H101O18PNa+, 1103.6618).
3-O-[6’-O-(1’’,2’’-dipalmitoyl-3’’-methylphosphoryl-sn-glycerol)-1’-β-D-2’,3’,4’-triacetyl-
galactopyranosyl]-1,2-distearoyl-sn-glycerol (XIII): 3-O-[6’-O-(3’’-methylphosphoryl-sn-glycerol)-1’-
β-D-2’,3’,4’-triacetyl-galactopyranosyl]-1,2-distearoyl-sn-glycerol (XII) (0.0052 g, 0.005 mmol) was
dissolved in anhydrous dichloromethane (0.3 mL); palmitic acid (0.00532 g, 0.022 mmol),
dicyclohexylcarbodiimide (0.0051 g, 0.022 mmol) and DMAP (0.00194 g, 0.0152 mmol) were added;
the reaction mixture was stirred overnight at 45 °C; after evaporation of solvent under reduced
pressure, the mixture was purified by silica gel chromatography using a gradient of
chloroform/methanol to give 3-O-[6’-O-(1’’,2’’-dipalmitoyl-3’’-methylphosphoryl-sn-glycerol)-1’-β-
D-2’,3’,4’-triacetyl-galactopyranosyl]-1,2-distearoyl-sn-glycerol (XIII) (0.0078 g, 0.005 mmol, 66%); Rf
(chloroform:methanol 95:5) = 0.90; 1H-NMR (400 MHz, CDCl3):  (bs, 1H), 5.25–5.14 (overlapped,
2H), 5.02 (dd, J = 2.7, 10.8 Hz, 1H), 4.49 (d, J = 7.4 Hz, 1H), 4.36–4.25 (overlapped, 3H), 4.20–4.08
(overlapped, 4H), 4.04 (m, 1H), 3.98–3.93 (overlapped, 2H), 3.75 (d, J = 11.1 Hz, phosphate methyl,
3H), 3.65 (dd, J = 6.1, 10.6 Hz, 1H), 2.36–2.25 (overlapped,-methylenes of acyl portions, 8H), 2.16 (s,
3H, acetate methyl), 2.06 (s, 3H, acetate methyl), 1.98 (s, 3H, acetate methyl), 1.64–1.55 (overlapped,-
methylenes of acyl portions, 8H), 1.34–1.17 (m, aliphatic protons), 0.88 (t, J = 6.9 Hz, methyls, 12H);
HRESIMS m/z 1580.1205 [M+Na]+ (calcd for C87H161O20PNa+, 1580.1211).
3-O-[6’-O-(1’’,2’’-dipalmitoyl-3’’-phosphoryl-sn-glycerol)-1’-β-D-2’,3’,4’-triacetyl-
galactopyranosyl]-1,2-distearoyl-sn-glycerol
(XIV):
3-O-[6’-O-(1’’,2’’-dipalmitoyl-3’’-
methylphosphoryl-sn-glycerol)-1’-β-D-2’,3’,4’-triacetyl-galactopyranosyl]-1,2-distearoyl-sn-glycerol
(XIII) (0.0078 g, 0.005 mmol) was dissolved in acetonitrile (0.4 mL); TMS-Cl (0.0015 mL, 0.01 mmol)
and sodium iodide (0.0014 g, 0.01 mmol) were added and after stirring in the dark at room
temperature overnight, the mixture was filtered and washed with dichlorometane; the filtrate was
evaporated to get 3-O-[6’-O-(1’’,2’’-dipalmitoyl-3’’-phosphoryl-sn-glycerol)-1’-β-D-2’,3’,4’-triacetyl-
galactopyranosyl]-1,2-distearoyl-sn-glycerol (XIV) (0.00524 g, 0.0034 mmol, 68%); Rf
(chloroform:methanol 95:5) = 0.30; 1H-NMR (400 MHz, CDCl3):  (bs, 1H), 5.24–5.11 (overlapped,
2H), 5.00 (dd, J = 2.6, 10.3 Hz, 1H), 4.53 (d, J = 7.0 Hz, 1H), 4.35-4.24 (overlapped, 3H), 4.15–4.24
(overlapped, 5H), 4.02–3.87 (overlapped, 3H), 3.68 (dd, J = 4.7, 10.9 Hz, 1H), 2.35–2.22 (overlapped,-
methylenes of acyl portions, 8H), 2.14 (s, 3H, acetate methyl), 2.04 (s, 3H, acetate methyl), 1.95 (s, 3H,
acetate methyl), 1.63–1.51 (overlapped,-methylenes of acyl portions, 8H), 1.32–1.15 (m, aliphatic
protons), 0.86 (t, J = 6.5 Hz, methyls, 12H); HRESIMS m/z 1542.1059 [M−H]− (calcd for C86H158O20P−,
1542.1090).
3-O-[6’-O-(1’’,2’’-dipalmitoyl-3’’-phosphoryl-sn-glycerol)-1’-β-D-galactopyranosyl]-1,2-
distearoyl-sn-glycerol (2): 3-O-[6’-O-(1’’,2’’-dipalmitoyl-3’’-phosphoryl-sn-glycerol)-1’-β-D-2’,3’,4’-
triacetyl-galactopyranosyl]-1,2-distearoyl-sn-glycerol (XIV) (0.00524 g, 0.0034 mmol) was dissolved
in aq. ethanol (85%) (0.4 mL), hydrazine monohydrate (0.0025 g, 0.05 mmol) was added, and the
reaction mixture was stirred for 7h at 48°C. After evaporation of solvent under a stream of nitrogen,
the residue was purified by silica gel chromatography using a gradient of chloroform/methanol to
give
3-O-[6’-O-(1’’,2’’-dipalmitoyl-3’’-phosphoryl-sn-glycerol)-1’-β-D-galactopyranosyl]-1,2-
distearoyl-sn-glycerol (2) (2.1 mg, 0.0015 mmol, 44%); Rf (chloroform/methanol 8:2) = 0.40; See Table
1 and Supporting information for NMR data. HRESIMS m/z 1416.0792 (calcd for C80H152O17P-,
1416.0773).

Mar. Drugs 2019, 117 of 21
3.8. Immunological Assays
3.8.1. moDCs Generation
For each assay, human peripheral blood mononuclear cells (PBMC) were isolated from two
healthy donors by routine Ficoll density gradient centrifugation. Monocytes were purified from
human peripheral blood mononuclear cells using MACS CD14 microbeads (Miltenyi Biotech,
Auburn, CA, USA) according to the manufacturer’s instructions. Purity was checked by staining with
a FITC-conjugated anti-CD14 antibody (Milteny Biotech, Auburn, CA, USA) and FACS analysis.
Immature DCs were obtained by incubating monocytes at 8 x∙105/mL in RPMI 1640 medium
supplemented with 10% fetal calf serum (FCS), 1% L-glutamine 2 mM, 1% penicillin and
streptomycin, human IL-4 (5 ng/mL) and human GM-CSF (100 ng/mL) for five days.
3.8.2. Cells Staining and Stimulation
After five days in culture, surface staining was performed on monocyte-derived dendritic cells
(MoDCs) for flow cytometry analysis with mouse anti-human HLA-DR FITC, CD83 PE, CD86 APC,
CD54 PeVio770, CD14 FITC, CD3 PercP (all from Miltenyi Biotech) and analyzed by flow-cytometer
LSRFortessa™ X-20 cell analyzer (BD Bioscience, Frankin Lake, NJ, USA) according to standard
protocol. MoDC were then incubated with natural and synthetic compound in 12-well plates at
concentration of 8 × 105 cell/mL. Stimulation with PAM2CSK4 1 µg/mL (Invivogen, San Diego, CA,
USA) was used as positive control. Cells treated only with vehicle (DMSO) were used as the control.
For the experiment with the anti TLR-4 neutralizing antibody, cells were incubated 30 minutes with
only the antibody (1 µg/mL) at 37 °C before addition of PGDG. After 24 hours, expression of all
surface markers was estimated again by staining with fluorochrome-conjugated antibodies.
3.8.3. Real Time PCR analysis
Total RNA was isolated using Trizol Reagent, according to the manufacturer’s protocol. RNA
quantity and purity were measured with a NanoDrop 2000 spectrophotometer (Thermo Scientific,
Waltham, MA, USA). Sample purity was checked by A260/A280 ratios between 1.80 and 2.00.
Extracted RNAs from all preparations were in this range. Cytokines mRNA expression was measured
by quantitative Real Time-PCR. Results are expressed as means ± SD.
3.8.4. TLR-2 and TLR-4 Assays
Human TLR-4/NF-kB/SEAP and TLR-2/NF-kB/SEAP (SEAPorter™) HEK 293 reporter cells
(Novus Biologicals, Littleton, CO, USA) were plated in 24-well plates at 2 × 105 cells/well. After 16
hours TLR4 cells for functional assays were transiently transfected with MD2/CD14 plasmid vector
(Invivogen, San Diego, CA, USA) for 24 hours. Cells were then stimulated with compounds for 24
hours. LPS for TLR-4 cell line and PAM2CSK4 for TLR-2 cell line were used as positive control under
the same conditions. SEAP (Secreted Alkaline Phosphatase) was analyzed using the SEAPorter Assay
(Novus Biologicals, Littleton, CO, USA) in according to manufacturer’s instructions. Quantitative
data (ng/mL) were obtained by a standard curve for the SEAP protein.
3.8.5. BM-DCs Generation
Eight/nine-week-old female C57BL/6 mice were purchased from Charles River (Lecco, Italy) and
housed in IGB “A. Buzzati-Traverso” Animal House Facility under standard pathogen-free
conditions abiding institutional guidelines. Bone marrow-derived dendritic cells (BM-DCs) were
produced from precursors isolated from the bone marrow of C57BL/6 mice by culturing them with
200 U/mL recombinant murine granulocyte/macrophage colony-stimulating factor (GM-CSF)
(Peprotech, Rocky Hill, NJ, USA) in RPMI 1640 (Lonza, Basel, Switzerland) medium supplemented
with 10% FCS, 100 U/mL penicillin, 100 µg/mL streptomycin, 1 mM sodium pyruvate and 50 µM 2-
mercaptoethanol. Immature DCs were collected at day 7 of culture and were assayed for dendritic
cell phenotype by staining with the monoclonal antibody anti-CD11c-PE-Cy7 (HL3, BD Biosciences).

Mar. Drugs 2019, 118 of 21
3.8.6. B3Z CD8+ T Cell Activation Assay
1 × 106/mL immature BM-DCs were incubated for 3 hours with different concentrations of
OVA257–264 synthetic peptide (from 0.0004 to 4 µM) in presence of 1µg/mL of LPS, MPLA, or PGDG 2.
As control, DCs incubated only with medium were used. After the incubation, cells were washed and
co-cultured (50,000/well) with the OTI hybridoma cell line B3Z (50,000/well) for 40 hours. The IL-2
released into cell co-culture supernatants was measured by ELISA. Supernatants (0.1 mL/well) were
assayed in duplicate using mouse IL-2 ELISA MAX™ Standard (Biolegend, San Diego, CA, USA),
according to the manufacturer’s instructions. B3Z OTI hybridoma cell line, recognizing the OVA257–
264 SIINFEKL determinant, was growth in complete RPMI 1640 (10% FCS 100 U/mL penicillin, 100
μg/mL streptomycin 1% Glutamine, 1% NEM, 1% Sodium Pyruvate, 50 µM 2-Mercaptoethanol) and
was a kind gift of Dr. Bénédicte Manoury (INEM, INSERM U1151-CNRS UMR 8253, Hôpital Necker,
Paris, France.).
3.8.7. Statistical Analysis
Comparative analysis was performed using the analysis of variance (ANOVA) followed by the
Bonferroni post hoc comparisons or the Tukey test for multiple comparison test. A p-value less than
0.05 was considered as statistically significant. All analyses were performed using the GraphPad
Prism 5.00 for Windows software (GraphPad Software, San Diego, CA, USA).
4. Conclusions
A rare pool of lipids, namely phosphatidylmonogalactosyl diacylglycerols PGDG (1), was
isolated and characterized for the first time from the marine diatom T. weissflogii.
Phosphogalactolipid 1 is a tetra-acyl derivative featured by (poly)unsaturated fatty acid chains on a
3-O-[6’-(sn-glycero-3’’-phosphoryl)-β-D-galactopyranosyl]-sn-glycerol. Its molecular core structure
was defined by spectroscopic methods and confirmed by total synthesis of the analog 2, performed
according to a versatile and general approach that gives access to a series of regio- and stereo-pure
phosphatidyl-β-glycosyl-diacylglycerol derivatives for biological and pharmacological development.
PGDG lipids gave immunostimolatory activity on DCs in the range 1–10 µg/mL and TLR-based
assays evidenced a TLR-4 agonistic activity: this represents the first report of an immunomodulant
activity for these uncommon lipids. Functional studies with hybridoma T cells revealed an adjuvant
potential of synthetic PGDG 2, significantly higher than standard lipid adjuvants such as LPS and
MPLA at 1 µg/mL in presence of OVA257-264 antigen. These results encourage further studies to define
the specific intracellular signaling pathway mediated by TLR-4, to assess structure-activity
relationship through preparation of molecular analogs and to explore the full pharmacological
potential of this class of bioactive lipids.
Supplementary Materials: 1D (1H and 13C) and 2D NMR spectra of 1. 1H and 13C NMR spectra of 2, 1H NMR of
synthetic intermediates V-XIV. Mass spectrometric (LCMS and GCMS) data of 1. Immunostimulant assay on
bioactive fraction from T. weissflogii.
Author Contributions: Conceptualization, A.C.; Methodology, E.M., C.G., R.S., P.D.B. and A.C.; Investigation,
E.M., C.G., R.S., G.N., A.S. and A.C.; Resources, E.M., C.G., P.D.B., A.F., and A.C.; Validation, C.G., R.S. and
A.C.; Writing – Original Draft Preparation, E.M., C.G. and A.C.; Writing – Review & Editing, R.S., A.F. and P.D.B;
Supervision, A.C.; Project Administration, A.C.; Funding Acquisition, A.F.
Acknowledgments: Authors are grateful to Raffaele de Palma for his support for immunological assays, and to
Ugo d’Oro for preliminary biological screening carried out in the frame of the MIUR Project “Antigens and
adjuvant for vaccines and immunotherapy” (PON01_00117) that partially funded this research. Funding was
also from PO-FESR 2014-2020 Project "SATIN". Servizio NMR of ICB is kindly acknowledged for recording
spectra.
Conflicts of Interest: The authors declare no conflict of interest.
References

Mar. Drugs 2019, 119 of 21
1.
2.
Short, S.A.; White, D.C. Metabolism of the glucosyl diglycerides and phosphatidylglucose of Staphylococcus
aureus. J. Bacteriol. 1970, 104, 126–132.
Nagatsuka, Y.; Kasama, T.; Ohashi, Y.; Uzawa, J.; Ono, Y.; Shimizu, K.; Hirabayashi Y. A new
phosphoglycerolipid, ‘phosphatidylglucose’, found in human cord red cells by multi‐reactive monoclonal
anti‐i cold agglutinin, mAb GL‐1/GL‐2. FEBS Lett. 2001, 497, 141–147.
3.
4.
Nagatsuka, Y.; Hirabayashi Y. Phosphatidylglucoside: A new marker for lipid rafts. Biochim. Biophys. Acta
2008, 1780, 405–409.
Takahashi, H.; Hayakawa, T; Murate, M.; Greimel, P.; Nagatsuda, Y.; Kobayashi, T.; Hirabayashi, Y.
Phosphatidylglucoside: Its structure, thermal behavior, and domain formation in plasma membranes.
Chem. Phys. Lipids 2012, 165, 197–206.
5.
6.
Smith, P.F.; Henrickson, C.V. Glucose-containing phospholipids in Mycoplasma laidlawii, strain B. J. Lip. Res.
1965, 6, 106–111.
(a) Wilkinson, S.G. Lipids of Pseudomonas diminuta. Biochim. Biophys. Acta 1969, 187, 492–500; (b) Wilkinson,
S.G.; Bell, M.E. The phosphoglucolipid from Pseudomonas diminuta. Biochim. Biophys. Acta 1971, 248, 293-
299.
7.
8.
Shaw, N.; Smith, P.F.; Verheij, H.M. The structure of 'phosphatidylglucose'. Biochem. J. 1970, 120, 439–441.
Fischer, W.; Ishizuka, I.; Landgraft, H.R.; Herrmann, J. Glycerophosphoryl diglucosyl diglyceride, a new
phosphoglycolipid from Streptococci. Biochim. Biophys. Acta 1973, 296, 527–545.
Veerkamp, J.H.; Van Schaik, F.W. Biochim. Biochemical changes in Bifidobacterium bifidum var.
Pennsylvanicus after cell wall inhibition VII. Structure of the phosphogalactolipids. Biochim. Biophys. Acta
1974, 348, 370–387.
9.
10. Fischer, W.; Landgraft, H.R. Glycerophosphoryl phosphatidyl kojibiosyl diacylglycerol, a novel
phosphoglucolipid from Streptococcus faecalis. Biochim. Biophys. Acta, 1975, 380, 227–244.
11. Abraham, W.; Meyer, H.; Lindholst, S.; Vancanneyt, M.; Smit, Phospho- and sulfolipids as biomarkers of
Caulobacter sensu lato, Brevundimonas and Hyphomonas. J. Appl. Microbiol. 1997, 20, 522–539.
12. Wicken, A.J.; Knox K.W. Lipoteichoic Acids: a new class of bacterial antigen. Science 1975, 187, 1161–1162.
13. Godfrey, D.; Hammon, K.; Poulton, L.; Smyth, M.; Baxtera, A. NKT cell: facts functions and fallacies.
Immunol. Today 2000, 21, 573–583.
14. Fujimoto, Y.; Mitsunobe, K.; Fujiwara, S.; Mori, M.; Hashimoto, M.; Suda, Y.; Kusumoto, S.; Fukase, K.
Synthesis and biological activity of phosphoglycolipids from Thermus thermophilus. Org. Biomol. Chem. 2013,
11, 5034–5041.
15. Cutignano, A.; Nuzzo, G.; Ianora, A.; Luongo, E.; Romano, G.; Gallo, C.; Sansone, C.; Aprea, S.; Mancini,
F.; D’Oro, U.; Fontana A. Development and application of a novel SPE-method for bioassay-guided
fractionation of marine extracts. Mar. Drugs 2015, 13, 5736–5749.
16. Cutignano, A.; Luongo, E.; Nuzzo, G.; Pagano, D.; Manzo, E.; Sardo, A.; Fontana, A. Profiling of complex
lipids in marine microalgae by UHPLC/tandem mass spectrometry. Algal Res. 2016, 17, 348–358.
17. Cutignano, A; d’Ippolito, G.; Romano, G.; Lamari, N.; Cimino, G.; Febbraio, F.; Nucci, R.; Fontana, A.
Chloroplastic glycolipids fuel aldehyde biosynthesis in the marine diatom Thalassiosira rotula. ChemBioChem
2006, 7, 450–456.
18. d’Ippolito, G.; Tucci, S.; Cutignano, A.; Romano, R.; Cimino, G.; Miralto, A.; Fontana A. The role of complex
lipids in the synthesis of bioactive aldehydes of the marine diatom Skeletonema costatum” Biochim. Biophys.
Acta/ Mol. Cell Biol. Lipids 2004, 1686, 100–107.
19. Murari, R.; Abd El-Rahman, M.M.A.; Wedmir, Y.; Parthasarathy, S.; Baumann, W.J. Carbon-13 nuclear
magnetic resonance of phospholipids in solution. Spectral and stereochemical assignments based on 13C-
31P and 13C-14N couplings. J.Org. Chem. 1982, 47, 2158–2163.
20. Williamson, M.P.; Griffin, C.E. Three- and four-bond 31P-1H coupling constants and germinal protons
nonequivalence in ethyl esters of phosphorus acids. J. Phys. Chem. 1968, 72, 4043–4047.
21. Pieringer, R.A.; Ganfield, M. W. Phosphatidylkojibiosyl diglyceride: Metabolism and function as an anchor
in bacterial cell membranes. Lipids 1975, 10, 421–426.
22. Shaw, J.M.; Pieringer, R.A. Phosphatidylmonoglucosyl diacylglycerol of Pseudomonas diminuta ATCC 11568
J. Biol. Chem. 1977, 252, 4395–4401.
23. Nuzzo, G.; Gallo, C; d’Ippolito, G.; Cutignano, A.; Fontana, A. Composition and quantitation of microalgal
lipids by ERETIC 1H NMR method. Mar. Drugs 2013, 11, 3742–3753.

Mar. Drugs 2019, 120 of 21
24. Norikatsu M.; Jimin G.; Hironori M.; Yasushi O.; Hiroaki T.; Akira T. Discovery of novel immunostimulants
by dendritic-cell–based functional screening, Blood 2005, 106, 3082–3089.
25. Rossi M.; Young JW. Human dendritic cells: potent antigen-presenting cells at the crossroads of innate and
adaptive immunity. J Immunol. 2005, 175, 1373–81.
26. Dudek A.M.; Martin S.; Garg A.D.; Agostinis P. Immature, semi-mature, and fully mature dendritic cells:
toward a DC-cancer cells interface, that augments anticancer immunity. Front Immunol. 2013, 4, 438.
27. Rovati, B.; Mariucci, S.; Manzoni, M.; Bencardino, K.; Danova, M. Flow cytometric detection of circulating
dendritic cells in healthy subjects. Eur. J. Histochem. 2008, 52, 45–52.
28. Cummings, R. D. The repertoire of glycan determinants in the human glycome. Mol. Biosys. 2009, 5, 1087–
1104.
29. Vartabedian, V.F.; Savage, P.B.; Teyton, L. The processing and presentation of lipids and glycolipids to the
immune system. Immun. Rev. 2016, 272, 109–119.
30. Dowling, J.K.; Mansell, A. Toll-like receptors: the swiss army knife of immunity and vaccine development.
Clin. Trasl. Immunol., 2016, 5, e85.
31. Manzo, E.; Cutignano, A.; Pagano, D.; Gallo, C.; Barra, G.; Nuzzo, G.; Sansone, C.; Ianora, A.; Urbanek, K.;
Fenoglio, D.; Ferrera, F.; Bernardi, C.; Parodi, A.; Leonardi, A.; Filaci, G.; De Palma, R.; Fontana A. A new
marine-derived sulfoglycolipid triggers dendritic cell activation and immune adjuvant response. Sci. Rep.
2017, 7, 6286.
32. Van Boeckel, C.A.A.; Visser, G.M.; Van Boom, J.H. Synthesis of phosphatidyl-β-glucosyl glycerol
containing a dioleoyl diglyceride moiety: Application of the tetraisopropyldisiloxane-1,3-diyl (tips)
protecting group in sugar chemistry. part IV. Tetrahedron, 1985, 41, 4557–4565.
33. Schmidt, R.R.; Michel, J. Facile synthesis of α‐ and β‐O‐glycosyl imidates; preparation of glycosides and
disaccharides. Angew. Chem. Int. Ed. 1980, 19, 731–732.
34. Schmidt, R.R. New methods for the synthesis of glycosides and oligosaccharides-Are there alternatives to
the Koenigs‐Knorr method? Angew. Chem. Int. Ed. 1986, 25, 212–235.
35. Manzo, E.; Ciavatta, M.L.; Pagano, D.; Fontana, A. An efficient and versatile chemical synthesis of bioactive
glyco-glycerolipids. Tetrahedron Lett. 2012, 53, 879–881.
36. Sallusto, F.; Palermo, B.; Lenig, D.; Miettinen, M.; Matikainen, S.; Julkunen, I.; Forster, R.; Burgstahler, R.;
Lipp, M.; Lanzavecchia, A. Distinct patterns and kinetics of chemokine production regulate dendritic cell
function. Eur. J. Immunol. 1999, 29: 1617–1625.
37. Hellman, P.; Eriksson, H. Early activation markers of human peripheral dendritic cells. Hum. Immunol.
2007, 68, 324–333.
38. Wang, J.; Wang, X.; Hussain, S.; Zheng, Y.; Sanjabi, S.; Ouaaz, F.; Beg A. Distinct roles of different NF-kB
subunits in regulating inflammatory and T cell stimulatory gene expression in dendritic cells. J. Immunol.
2007, 178, 6777–6788.
39. Steinhagen, F.; Kino, T.; Bode, C.; Klinman, D.M. TLR-based immune adjuvants. Vaccine 2011, 17, 3341–
3355.
40. Takeuchi, O.; Akira, S. Pattern recognition receptors and inflammation. Cell 2010, 140, 805–820.
41. Chow, J.C.; Young, D.W.; GolenBock, D.T.; Gusovsky, F. Toll-like receptor-4 mediates lipopolysaccharide-
induced signal transduction, J. Biol. Chem. 1999, 274, 10689–10692.
42. Yang, F.L.; Hua, K.F.; Yang, Y.L.; Zou, W.; Chen. Y.P.; Liang, S.M.; Hsu, H.Y.; Wu, S.H. TLR-independent
induction of human monocyte- IL-1 by phosphoglycolipids from thermophilic bacteria. Glycoconj. J. 2008,
25, 427–439.
43. Ginsburg, I. Role of lipoteichoic acid in infection and inflammation. Lancet 2002, 2, 171–179.
44. Burger-Kentischer, A.; Abele, I. S.; Finkelmeier, D.; Wiesmüller, K. H.; Rupp, S. A new cell-based innate
immune receptor assay for the examination of receptor activity, ligand specificity, signaling pathways and
the detection of pyrogens. J. Immunol. Methods 2010, 358, 93–103.
45. Song, R.; Fullerton, D.A; AL, Zheng D; Zhao, KS; Meng, X. BMP-2 and TGF-β1 mediate biglycan-induced
pro-osteogenic reprogramming in aortic valve interstitial cells. J. Mol. Med. 2015, 93, 403–412.
46. Hirano, N.; Butler, M.O.; Xia, Z.; Ansén, S.; von Bergwelt-Baildon, M.S.; Neuberg, D.;. Freeman, G. J.;
Nadler L. M. Engagement of CD83 ligand induces prolonged expansion of CD8+ T cells and preferential
enrichment for antigen specificity. Blood 2006, 107, 1528–1536.
47. Sheng, K.C.; Day, S.; Wright, M.D.; Stojanovska, L.; Apostolopoulos, Y. Enhanced dendritic cell-mediated
antigen-specific CD4+ T cell responses: IFN-gamma aids TLR stimulation. J. Drug Deliv. 2013, 2013, 516749.

Mar. Drugs 2019, 121 of 21
48. Karttunen, J., S. Sanderson; Shastri, N. Detection of rare antigen-presenting cells by the lacZ T-cell
activation assay suggests an expression cloning strategy for T-cell antigens. Proc. Natl. Acad. Sci. USA 1992,
89, 6020–6024.
49. Stöver, A.G.; Da Silva Correia, J.; Evans, J.T.; Cluff, C.W.; Elliott, M.W.; Jeffery, E.W.; , Johnson, D.A.; Lacy,
M.J.; Baldridge, J.R.; Probst, P.; Ulevitch, R.J.; Persing, D.H.; Hershberg, R. M.; Structure-activity
relationship of synthetic Toll-like receptor 4 agonists. J Biol Chem. 2004, 279, 4440–4449.
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article
is an open access article distributed under the terms and conditions of
the
Creative
Commons
Attribution
(CC
BY)
license
(http://creativecommons.org/licenses/by/4.0/).