| Principal Investigator : Ram A Vishwakarma
Project
Associates/Assistants
Technical
Assistants
Ph
D Students Collaborator The
project aims at the chemical synthesis and biosynthesis of GPI cell surface
molecules (lipophosphoglycan, proteophosphoglycans, and free GPIs) of Leishmania
parasite, and design and synthesis of structural and functional mimics and
inhibitors of the GPI biosynthetic pathway. GPIs are expressed abundantly on
the cell surface of Leishmania, Trypanosoma and Plasmodium
species, and have been implicated in the infectivity, transmission and
intracellular survival of the parasites. Parasite GPIs have distinct
structural features, compared to mammalian protein GPI-anchors, and are
interesting targets for chemical synthesis, biosynthesis and molecular design. The
objectives of the project are (i) design and synthesis of GPI anchor, LPG
structural domains (PI, GPI, conserved glycan-core, variable phosphoglycan
repeats and terminating cap), and GPI analogues, (ii) synthesis of labeled
biosynthetic precursors and their incorporation in the parasite to elucidate
the steps involved in LPG/GPI assembly, (iii) design and synthesis of
structural and functional mimics, and mechanism based inhibitors of the GPI
biosynthetic pathway, and (iv) role of parasite GPIs and their synthetic
analogues in PKC and PI-3-kinase mediated events in macrophages. Leishmania
species synthesize a unique class of molecules termed phosphoglycans,
including the membrane-bound LPG and secreted PPG. Their role(s) in parasite
virulence has been debated in recent years but now it has been established
that the principal virulent determinant
indeed consists of the PG repeat units, independent of their molecular
platform (lipid-linked in LPG and peptide-linked in PPG). In the preceding
years we have reported syntheses of all four individual structural domains of
LPG viz. the GPI anchor, conserved hexasaccharide core, phosphoglycan repeats,
and tetrasaccharide cap. Now the focus is on the assembly of these four
domains to construct full-length LPG of L.donovani. Efforts towards
this formidable synthetic target have continued this year, and
necessary amounts of suitably activated and structurally characterized
intermediates have been prepared. Particularly demanding problems of synthesis
of Galf
containing trisaccharide, its activation for coupling with the extended GPI
anchor, and linking of negatively charged PG domain with neutral cap have been
worked out. Although
LPG is expressed by all Leishmania species, there is one remarkable
difference between the structure of LPG from L.major and that from
other species; the 3-OH position of the galactose residue of the PG repeats is
highly substituted with 1-3b-galactosyl
branches in L.major (as shown in Figure-1), whereas this position
remains totally unsubstituted in L.donovani where the PGs are linear,
and obviously this is not without biological consequences.
Figure-1
: Structure
of LPG of L.major Last
year efforts towards the synthesis of the branched phosphoglycans of L.major
were initiated and now we have achieved a facile and bi-directional synthesis
(Scheme-1 and 2) of these complex molecules. Access to both the branched PG of
L.major and linear PG of L.donovani would allow us to address
structural and functional issues. The synthesis of the repeating PGs is
complicated, as compared to that of non-anomeric types such as DNA and RNA,
due to presence of anomeric phosphodiesters between each repeat unit. This
places additional requirements of anomeric stereo-control during the
synthesis, and also leads to instability due to the propensity of the glycosyl
ring to form a stabilized carbocation by expulsion of the anomeric phospho-monoester
leaving group. The branched L.major phosphoglycans presented specific
problems due to the presence of 3'-galactosyl branches in these labile
molecules. Key features of our approach include (a) remarkable dual
selectivity in protection of the easily available starting material, D-lactal,
with mutually compatible TBDMS and PMB groups at strategic positions, (b) a
gluco-manno transformation followed by 3' galactosylation to give the central
trisaccharide intermediate ready to serve either as the phosphate donor or
acceptor. Having secured efficient access to both the trisaccharide H-phosphonate
donor 9 and the trisaccharide acceptor 10, their coupling was
done by pivaloyl chloride followed by in situ iodine oxidation to give
the fully protected phosphohexa saccharide 11. The presence of (1®6)-phosphodiester
between the PG repeats was confirmed by 31P-13C
coupling (d, 5 and 7 Hz in 13C
NMR) for C-1 and C-2 of the Man units, and C-5 and C-6 of the corresponding
Gal units; these 13C
signals were shifted due to the a-
and b-
phosphorylation effect. The a-configuration
of mannosyl phosphate linkage was confirmed by NMR analysis that showed
characteristic coupling (5.45 ppm, JHH=1.9
and JHP=7
Hz) for the corresponding anomeric proton. This approach will now be extended
to a solid-phase method for the preparation of larger branched PGs for their
conjugation with suitable carrier antigenic proteins to probe these neo-glycoconjugates
for function.
Scheme-1
: Reagents
and conditions (a) Bu2SnO, MeOH, reflux, 4h; PMBCl, CsF, Kl, DMF, rt, 48h, 74%; (b)
NaH, 18-Crown-6, TBSCl, THF, 0oC-rt,
30 min, 65%; (c) (i) m-CPBA, Ether-NahCO3
buffer, 0oC, 3h, (ii) Ac2O,
py, rt, 16h, 80%; (d) DDQ, CH2Cl2-H2O,
rt, 12h, 85%; (e) Compd 5 and 6, Ether, TMSOTf, -20oC,
4h, 74%. Scheme-2
: Reagents
and conditions (a) Me2NH, CH3CN, -20oC,
3 h, 95%; (b) PCl3, imidazole, CH3CN, 0oC, 2 h, TEAB
work-up, 85%; (c) HF-CH3CN, 0oC,
2h, 85%; (d) compd 9 and 10, pivaloyl-Cl, pyridine, rt, 1h; I2
in aq py, 0.5h, TEAB work-up, 79%; (e) HF-CH3CN,
0oC, 2h; NaOMe-MeOH, 95%. The
above solution synthetic approach was extended to construct larger
phosphoglycans (19-22 repeats) by a polycondensation of a single bifunctional
PG-intermediate. Although this approach yielded a heterogeneous mixture of
19-22 repeats, availability of this unique material opened opportunity for us
to address functional aspects of the phosphoglycans. This year we have
evaluated the PG polycondensation product to address key biological question
regarding their role in modulating macrophage function. A series of
experiments led to the interesting result that synthetic PGs can induce
pro-inflammatory cytokines in monocytes in dose dependent manner, and activate
caspase-3 mediated pathway of apoptosis. This study addresses the
long-standing issue of whether the biological activity of LPG/PPG is due to
the glycan domain or the proteins associated with PG backbone. The result is
significant keeping in view that the PPGs are secreted in large amounts inside
the macrophages by the parasite. Flip-flop
of lipids across biomembranes is a fundamental aspect of membrane biogenesis.
Phospholipids, the building blocks of membrane bilayers, are made on the
cytoplasmic face of biogenic (self-synthesizing) membranes like the ER and
must be flipped to the opposite face (lumen) for bilayer expansion. Flipping
requires specific transporter proteins or flippases (distinct from ABC
transporters of plasma membrane) that facilitate bi-directional movement of
lipids across the membrane. Interestingly the major pathways of protein and
lipid glycosylation are postulated to require one or more (glyco)lipid
flipping steps. An example of this is the biosynthesis of GPI-anchored
proteins. Surprisingly, biosynthesis of GPIs begins on the cytoplasmic side of
the ER but the addition of protein occurs in the lumen. This implies that a
GPI intermediate must flip across the ER during GPI assembly but there is no
direct demonstration of this critical flipping step to date. To address this
question of GPI biosynthesis, we have synthesized the first generation of
functional fluorescent GPI probes and demonstrated their utility to show, for
the first time, that GlcNAc-PI and GlcN-PI both are able to flip across the ER
by an ATP-independent, protein-mediated process. The synthetic design
(Scheme-3) for the GPI probes involved three chiral building blocks: (a)
1-allyl-2,3,4,5-tetra-O-benzyl-D-myo-inositol 3 made in 7
steps from bis-cyclo hexylidene-D-inositol, (b) the 3,4,6-tri-O-acetyl-2-azido-2-deoxy-b-D-glucosyl
donor 4 and (c) the phosphatidyl lipid-donor 9 with a protected
terminal amine in the sn-1 acyl chain prepared from
1,2-isopropylidene-sn-glycerol. Glycosylation of 3 with 4 gave
the a-glucosaminyl(1®6)
Inositol 5 followed by deprotection of the allyl group and replacement
of the azido with NHBoc to enable the eventual selective coupling of the NBD
probe. Phospholipidation of 8 with 9 using H-phosphonate
chemistry gave the protected GPI intermediate 10. The next three steps,
removal of benzyls, coupling of the NBD probe, and deprotection of the NHBoc
group provided NBD-GlcN-PI (1), which on N-acetylation gave NBD-GlcNAc-PI
(2). For the flipping experiments (with Anant Menon) we have used a
biochemical reconstitution approach and assayed flippase activity in
proteoliposomes reconstituted from a Triton X-100 extract (TE) of ER. Trace
amounts of fluorescent NBD-labeled GPIs probes were included during
reconstitution of the proteoliposomes and flipping was assayed by exposing the
vesicles to dithionite, a dianion that destroys the fluorescence of lipids in
the outer leaflet of the vesicles (Figure-2). Phosphatidylinositols
possess two asymmetric components. All naturally occurring PI/GPI contain a myo-inositol
system substituted at D-1 position and a glycerol moiety acylated at sn-1
and sn-2 positions, although the fatty acid composition is variable.
Whether this stereochemical arrangement is important for their organization
and movement in the ER membrane is a question of fundamental importance for
understanding of intracellular signaling mechanisms and also as basis of
rational inhibitor design. We therefore have prepared all four possible
diastereoisomers (one natural and three un-natural) of the
phosphatidylinositols in their fluorescent forms (Figure-3). This set of four
PI probes will now be used in our membrane experiments.
Scheme-3
: Synthesis
of NBD-GlcNAc-PI and NBD-GlcN-PI. Reagents and conditions: (a) TMSOTf, Ch2Cl2,
0°C, 0.5h, 90%, (b) NaOMe, Ch2Cl2-MeOH,
rt, 24h; then NaH, BnBr, 80%, (c) PdCl2,
NaOAc, AcOH-H2O, rt, 48h, 64%, (d)
propanedithiol, Py-H2O, Et3N,
rt, 24h; then Boc2O, rt, 12h, 66%,
(e) Lipid-H-phosphonate 9, Py, Piv-Cl, rt, 0.5h; then I2
in Py-H2O, rt, 0.5h, 57%, (f)
Pd(OH)2, MeOH-CH2Cl2-H2O,
H2, 12h, 85%, (g) NBD-X,SE, DMF, Et3N,
2h, rt (h) TFA-CH2Cl2-CH3CN,
4:4:2, rt, 2h, 55%, (i) Ac2O, NaHCO3-MeOH,
rt, 0.5h, quant.
Figure-2
:
Flipping of GPI probes. (A) Fluorescence traces of assays with
liposomes (L) and proteoliposomes (P) containing the NBD GPIs. (B) Protein
dependence of the extent of dithionite reduction of NBD GPIs. (C) Model
showing 50% fluorescence loss on adding dithionite to liposomes (or inactive
proteoliposomes) (left), compared with 100% loss in flippase-active
proteoliposomes (right) due to flipping of NBD lipids. The
biosynthesis of Galactofuranose residues in LPG involves three putative
activities; UDP-Galp-mutase for conversion of UDP-Galp to
UDP-Galf, a membrane protein for transport of UDP-Galf from the
cytoplasm to the Golgi lumen, and a UDP-Galf-transferase for addition
of Galf residues to the growing LPG core. In order to study the process
by which Galf is incorporated into LPG, we designed two expression
systems for production of UDP-Galp-mutase, cloned from Mycobacterium
tuberculosis and E.coli K-12 in the form of GST-fusion and
histidine tagged proteins respectively. Availability of a good source of
UDP-Galf should allow us to make required amounts of radiolabeled
UDP-Galf for biosynthetic and transport activities in Leishmania
donovani. Towards
the synthesis of structural and function mimics of phosphoglycans such as the
hybrid LPG and multivalent glycodendrimers, a hexa-functional glycerol-inositol
core has been prepared in multi-gram scale, which can accommodate six
phosphoglycan branches. The availability of these GPI intermediates has set
the stage for linking them with the protected PG domain for construction of
hybrid LPGs. These hybrid LPGs have been designed to address the question of
domain specificity of LPG for PKC inhibition.
Figure-3
: All
possible diastereoisomers of phosphatidyl-inositol synthesized in fluorescent
forms. In
addition to our primary focus on Leishmania GPIs, we are also involved
in efforts towards chemical synthesis of the unique GPI anchor of the malarial
paprasite (P.falciparum) to address questions related to glycobiology
and immunology of these so-called malaria toxins (with DC Gowda). This year we
have made substantial progress towards synthesis of full length malarial GPI.
This required construction of top tetra-mannose domain and lower GPI anchor
with C-2 acylated myo-D-inositol. The tetrasaccharide fragment, with
Man-3 substituted with ethanolamine and conjugated to a carrier protein at
reducing end, is now being evaluated as an anti-disease vaccine candidate
against P.falciparum.
Publications Original peer-reviewed articles 1. Ruhela D and Vishwakarma RA (2003) Iterative synthesis of Leishmania phosphoglycans by solution, solid-phase and poly-condensation approaches without involving any glycosylation. J Org Chem 68:4446-4456. 2.
Ruhela D and Vishwakarma RA (2004) A facile and novel route to the antigenic
branched phosphoglycan of the protozoan Leishmania major parasite. Tetrahedron
Lett 45:2589-2592. Patents 1.
Balakrishnan A, Giridharan P, Vishwakarma RA and Nallankandy SN (2003) A
substituted furochromenone derivative exhibiting anti-cancer and anti-proliferative
properties. PCT Int Appl WO 2003022854. |
