| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Parasitology, University of São Paulo, São Paulo, SP, Brazil; and
Department of Biochemistry and Immunology, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil
Correspondence: Dr. Igor C. Almeida, Department of Parasitology, Institute of Biomedical Sciences, University of São Paulo, Av. Prof. Lineu Prestes 1374, São Paulo, SP 05508-900, Brazil. E-mail: ialmeida{at}icb.usp.br
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGPI ANCHORS: CORRELATION BETWEEN...CHARACTERIZATION OF THE...IDENTIFICATION OF MACROPHAGE...COMPARATIVE STRUCTURAL ANALYSIS...IMPLICATIONS OF THE RESISTANCE...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
Key Words: protozoan parasites • macrophages • innate immunity • cytokine • Toll-like receptors
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGPI ANCHORS: CORRELATION BETWEEN...CHARACTERIZATION OF THE...IDENTIFICATION OF MACROPHAGE...COMPARATIVE STRUCTURAL ANALYSIS...IMPLICATIONS OF THE RESISTANCE...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
The transmission of T. cruzi by natural vectors, the hematophagous triatomine bugs, occurs in a large area from Southern regions in the United States to Argentina and Chile. Chagas’ disease or American trypanosomiasis, caused by infection with T. cruzi, is a long-lived infection that affects approximately 18 million individuals in Latin America [1 ]. The infection with this trypanosomatid protozoan begins with a short acute phase characterized by high parasitemia and various clinical symptoms including myocarditis. With development of host immunity against T. cruzi, most parasites are cleared from the bloodstream and tissues, the clinical symptoms subside, and an asymptomatic chronic phase is maintained with a scarce number of parasites, which persist for life in most Chagasic patients. However, symptomatic forms may emerge in some chronic Chagasic patients, usually affecting the heart and/or digestive organs [2 ]. Despite the controversy on the mechanism of immunopathogenesis in chronic Chagasic patients [3 ], recent studies indicate that during chronic infection with T. cruzi, parasites are physically associated with the sites of inflammation on the heart [4 , 5 ] and esophagus [6 ]. During infection with T. cruzi, the activation of the immune system is likely to be involved in at least two aspects of Chagas’ disease pathophysiology, which are the control of parasite replication and spread in the host tissues and the inflammatory reaction in infected host tissues. The latter is regarded as a major cause of tissue damage leading to organ dysfunction during acute and chronic phases of the disease.
In fact, early studies demonstrated that host resistance/susceptibility
to infection is, at least in part, determined at the very early stages
of infection, before the development of adaptive immunity
[7
8
9
10
]. It is believed that the cell functions involved
on innate immunity, which are triggered by T. cruzi
parasites, may have important consequences on different aspects of
T. cruzi infection, such as load of tissue parasitism,
tissue tropism, and the pathogenesis of Chagas’ disease. At the onset
of infection, T. cruzi activates cells from macrophage
lineage to produce high levels of chemokines and proinflammatory
cytokines [11
]. Most notably, the cells exposed to
protozoan products will produce interleukin (IL)-12, which is
responsible for initiation of interferon-
(IFN-) synthesis by
natural killer (NK) cells and which favors the differentiation of Th0
lymphocytes into the Th1 phenotype [11
12
13
]. Thus far,
IFN- has been shown to be the crucial cytokine mediating resistance
during acute infection with this protozoan. It is noteworthy that
different studies show that Trypanosoma products may also
serve as a second signal to stimulate the synthesis of reactive
nitrogen intermediates (RNI), such as nitric oxide (NO), by
IFN--primed macrophages, a phenomenon that is essential for host
resistance during early stages of infection with this parasite
[8
].
Little is known about the parasite molecules that trigger functions of the innate immune system during the early stages of protozoan infection. In the last few years, the identification and structural characterization of parasite molecules that initiate the synthesis of proinflammatory cytokines and RNI by macrophages has been the focus of many studies [14 15 16 17 18 19 20 21 22 23 24 25 26 ]. Glycosylphosphatidylinositol (GPI)-anchored molecules are the predominant antigens present on the plasma membrane of parasitic protozoa such as T. cruzi (Fig. 1 ). Experimental evidence strongly suggests that GPI anchors derived from Plasmodium falciparum, Trypanosoma brucei, and T. cruzi play an important role on the activation of the innate immune system during protozoan infection. Among the biological properties of these protozoan-derived glycolipoconjugates are their ability to elicit the synthesis of proinflammatory cytokines and chemokines, generation of RNI, as well as the expression of adhesion molecules by host macrophages and endothelial cells.
|
|
GPI ANCHORS: CORRELATION BETWEEN STRUCTURE AND ACTIVITY ON CELLS FROM MACROPHAGE LINEAGE |
|---|
|
TOPABSTRACTINTRODUCTIONGPI ANCHORS: CORRELATION BETWEEN...CHARACTERIZATION OF THE...IDENTIFICATION OF MACROPHAGE...COMPARATIVE STRUCTURAL ANALYSIS...IMPLICATIONS OF THE RESISTANCE...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
1-2Man1-6Man1-4GlcN1-6myo-inositol-1-PO4-.
A great variety of closely related structures arise from this conserved
motif by the addition of carbohydrate residues (e.g.,
Manp, Galp, ßGalp,
ßGalf, GlcNAc, and sialic acid) and phosphorylated
substituents [i.e., ethanolaminephosphate (EtNP),
2-aminoethylphosphonate (AEP), and
Glc-1-PO4-]. Such hydrophilic moiety of
the GPI anchor is covalently attached, through the
myo-inositol-linked phosphate residue to a hydrophobic
moiety composed of a glycerolipid or a ceramide. In addition, the
myo-inositol ring can be further modified by acylation with
palmitic acid (C16:0) or, more rarely, myristic acid (C14:0). In free
and protein-associated GPIs, an extensive diversity of structures is
also generated at the hydrophobic moiety by qualitative and
quantitative changes on the lipid component(s). Many other structural
and evolutionary aspects of GPI structures have been reviewed
extensively elsewhere [30
31
32
].
In previous studies conducted in our laboratories, in collaboration
with other groups, we have demonstrated that tGPI-mucins and highly
purified tGPI elicit a potent proinflammatory response in
lipopolysaccharide (LPS)-resistant C3H/HeJ and LPS-sensitive C3H/HeN
murine macrophages [20
, 23
]. The intact
tGPI anchor is a remarkably potent activator of murine macrophages,
inducing proinflammatory cytokines and NO synthesis in the 0.1–10 nM
range. At this level, the tGPI is an inflammatory agent as potent as
the bacterial LPS and Mycoplasma lipopeptide. It is
interesting that mucin-derived GPIs and free GPIs, the
glycoinositolphospholipids (GIPLs) isolated from the insect-derived
(epimastigote and metacyclic) T. cruzi developmental stages,
are rather poor activators of macrophages when compared with the
trypomastigote-derived molecules (tGPI). When used in the 0.1–5.0 µM
range, most of the GIPLs and GPI-mucins derived from epimastigotes are
able to induce tumor necrosis factor (TNF-), IL-12, and NO at
levels comparable to those observed for P. falciparum and
T. brucei GPIs [14
, 17
,
22
, 24
, 25
]. Considering the
extremely short time that the insect-derived stages from T.
cruzi persist in the vertebrate host, probably the high
concentrations of 0.1–5.0 µM GPI-anchored mucins and/or GIPLs will
never be achieved in the case of American trypanosomiasis. Thus, most
of our work has been carried out using GPI-anchored mucins derived from
the mammalian cell-derived trypomastigote stage of T. cruzi.
Nonetheless, structural analytical studies of mucin-derived GPIs and
GIPLs, from distinct developmental stages and strains of T.
cruzi, have been helpful to establish a correlation between
structure and proinflammatory activity in protozoan-derived GPI
anchors.
Analysis of tGPI from T. cruzi trypomastigote mucins by
electrospray ionization mass spectrometry (ESI-MS) and gas
chromatography-mass spectrometry (GC-MS) revealed the presence of four
to eight hexoses (Man4 and Gal0–4, in an
average ratio of 4:2.5) in the GPI glycan core, as well as EtNP and AEP
as phosphorylated substituents [23
]. The anomeric
structure and linkage positions of the Man and Gal residues have not
been determined, but it has been assumed that at least the Man residues
are arranged in the glycan core as in the conserved motif
Man1-2Man1-2Man1-6Man1-4GlcN1-6 myo-inositol-1-PO4-,
commonly shared by all T. cruzi GPI anchors thus far
characterized [33
34
35
36
37
38
39
]. The most striking finding was
that the tGPI phosphatidylinositol (PI) moiety is composed of
sn-1-O-(C16:0)alkyl-2-O-acylglycerol,
containing mainly unsaturated (C18:1 and C18:2) fatty acids at the
sn-2 position [20
, 23
]. Thus
far, no other parasite-derived GPI anchor had been shown to contain
unsaturated fatty acids [30
, 31
]. Chemical
removal, by alkaline hydrolysis, of the unsaturated C18:1- or
C18:2-fatty acid from the sn-2 position of tGPI abrogates
most, but not all, of its inflammatory activity. Furthermore, periodate
oxidation, which breaks the linkage between carbon atoms containing
vicinal hydroxyl groups located at the GPI glycan, and nitrous
deamination, which selectively releases the PI moiety from the GPI,
leads to the complete inactivation of tGPI. The
alkylacylglycerol-phosphoinositol moiety alone, up to 25 nM, was not
able to induce cytokines (TNF- and IL-12) or NO [20
,
21
, 23
].
Structural analysis of the GPI anchor isolated from mucin-like
glycoproteins of T. cruzi, noninfective insect-derived
epimastigotes (eGPI) revealed a conserved glycan sequence
(Man1-2Man1-2Man1-6Man1-4GlcN) and a PI composed
exclusively of 1-O-(C16:0)alkyl-2-O-(C16:0 or
C18:0)acylglycerol-3-phosphate-1-myo-inositol
[23
, 37
, 38
]. The GPI anchor
of mucin-like glycoproteins from the infective, insect-derived
metacyclic trypomastigote stage (mGPI) contains the same conserved
Man4-GlcN glycan sequence, but the
myo-inositol-phosphate-lipid moiety contains predominantly
(70%) inositol-phosphoceramides (IPCs) composed of a
(d18:0)-sphinganine long-chain base and mainly lignoceric acid (C24:0;
48%) or palmitic acid (C16:0; 15%), and to a lesser extent (30%),
1-O-(C16:0)alkyl-2-O-(C16:0 or C18:0)
acylglycerol, also found in eGPIs [38
]. Aside from the
abundantly expressed GPI-mucins [31
, 40
,
41
], the GIPLs, formerly known as
lipopeptidophosphoglycan (LPPG) [42
], are found on the
plasma membrane of all T. cruzi stages and strains thus far
investigated [31
, 33
, 34
,
39
, 42
43
44
45
]. However, they are only
abundantly expressed on the insect vector-derived epimastigote and
metacyclic stages [33
, 34
, 39
,
42
43
44
45
]. The GIPLs are also present on other protozoan
parasites but not in higher eukaryotic cells [30
,
31
]. As with any other GPI-anchor, the GIPLs share the
conserved hydrophilic motif of
Man1-4GlcN1-6-myo-inositol-1-PO4-.
However, various modifications are found in the lipid moiety and
carbohydrate branches linked to this conserved core, depending on the
parasite species and strain origin. In T. cruzi, for
instance, the GIPL lipid moiety contains mainly ceramide
([d18:0]-sphinganine or [d18:1]-sphingosine linked to lignoceric
acid or palmitic acid) or
sn-1-O-(C16:0)alkyl-2-O-(C16:0 or
C18:0)acylglycerol [23
, 39
,
43
].
eGPI and mGPI-mucins and GIPLs were found to be at least 100- to 1000-fold less active than the tGPI-mucins in the induction of proinflammatory cytokines and NO by murine macrophages [20 , 21 , 23 ]. The primary structure of GPIs from the major T. cruzi mucins and GIPLs are shown in Figure 2A . Comparing the structures of eGPI, mGPI, and tGPI, we notice that the latter has a longer glycan core, due to the extra Gal residues, and a more fluid lipid moiety, because of the presence of an unsaturated fatty acid at the sn-2 position. Most probably, this unique combination makes the tGPI more soluble in aqueous media/phases and with a final tridimensional conformation, which allows it to be more adequately presented to a putative macrophage receptor.
|
and IL-1 production by murine
macrophages. Removal of the fatty acids linked to the glycerol portion
by specific phospholipases completely abolished the cytokine induction,
indicating that the lipid moiety could play an essential role in the
macrophage-activation process [14
]. Similar data were
obtained with GPI anchors purified from the variant surface
glycoprotein (VSG) of T. brucei trypomastigotes
[16
].
Further, comparing the cytokine- and NO-inducing activity of GPIs
isolated from P. falciparum MSP and T. brucei
VSG, Tachado et al. [19
] demonstrated that the GPI
glycosyl-inositolphosphate sequence could activate protein tyrosine
kinase (PTK). Conversely, the diacylglycerol moieties of the P.
falciparum and T. brucei GPIs were able to activate
protein kinase C-
(PKC-). Together, the GPI
glycosyl-inositolphosphate and the diacylglycerol moieties from these
protozoan GPIs had a synergistic effect in triggering the synthesis of
TNF-. Magez et al. [22
] have also demonstrated that
highly purified, GPI-anchored VSG from T. brucei is able to
induce TNF- and IL-1 production by LPS-low-reponsive (C3H/HeJ)
and LPS-responsive (C3H/HeN) murine macrophages. The TNF--inducing
activity is mapped to the
Man1-2Man1-6-(Gal2–4)Man1-4GlcN1-6myo-inositol-1-HPO4
moiety of the VSG-GPI, and the IL-1-inducing activity is found to be
associated with the sn-1,2-O-dimiristoylglycerol
moiety of the GPI anchor. It is interesting that the Gal residues
were found to be critical for optimal induction of TNF-.
Additional efforts have been made to characterize the GPI anchors from
the MSP family members of P. falciparum intraerythrocytic
stages [15
, 46
]. It has been proposed that
the MSP contains two GPI species: a major one, Pfg1, with a
sequence of
EtNP-(Man1-2)-Man1-2Man1-6Man1-4GlcN1-6(myristoyl)-myo-inositol-1-PO4--sn-1,2-(C16:0/C16:0)diacylglycerol,
and a minor one, Pfg1ß, lacking the fourth Man residue distal from
GlcN [15
, 46
]. In a more recent study, Naik
et al. [24
] were able to elegantly perform, for the
first time in the malaria research field, a direct structural analysis
by chemical and enzymatic methods as well as mass spectrometry of free
and MSP-derived GPIs from P. falciparum erythrocytic stages.
The glycan core structures of MSP-1, MSP-2, and MSP-4 were found to be
identical to those previously described [15
,
46
47
48
]. In contrast, the major lipid constituents of the
PI moiety differ from those observed by metabolic labeling
[46
]. For instance, at the C-2 of the
myo-inositol ring, myristic acid (C14:0) only accounted for
10% of the total acylation, whereas the major (90%) substituent was
palmitic acid (C16:0). The PI diacylglycerol moiety contained at the
sn-1 position mainly C18:0 (80%), C16:0 (12%), and small
amounts of C14:0 (4%), C20:0 (2%), and C22:0 (2%). Conversely, at
the sn-2 position, C18:1 (two isomers, oleic and
cis-vaccenic acids, 88%) and C18:2 (12%) were
characterized as the main substituents [24
]. These fatty
acids have been previously found at the sn-2 position of the
alkylacylglycerol portion of the potent inflammatory GPI anchors from
T. cruzi tGPI-mucins [20
, 23
].
Free GPIs or GIPLs are also present in P. falciparum and are
similar to MSP-derived GPIs, because oleic acid
(
9-cis, 
85%), linoleic acid (9%), and
cis-vaccenic (11-cis, 6%) are
the main fatty acids at the sn-2 position, whereas stearic
acid and palmitic acid are the main substituents at the sn-1
position [24
]. It was also shown that highly purified
free or protein-linked P. falciparum GPIs, in the 0.03–0.5
µM range, could elicit TNF- by murine macrophages, confirming
previous data obtained by Schofield and collaborators
[14
, 16
]. The structure of the major
P. falciparum-derived MSP-GPI is shown in Figure 2B
.
In a more recent study, Vijaykumar et al. [25
] have
shown that the TNF--inducing activity of P.
falciparum-free GPIs is dependent on the recognition of the fourth
distal Man residue (see Fig. 2B ), because the removal of this residue
by jack bean -mannosidase completely abolished the cytokine
secretion. Preincubation of macrophages with
methyl--mannopyranoside, mannobiose (Man1-2Man) and
Man4-GlcN-myo-Ins-PO4-
led to full inhibition of the native GPI activity. The authors
suggested that the mechanism of macrophage activation by P.
falciparum GPI could involve its recognition by a putative
macrophage mannose receptor, rather than endocytosis or membrane
insertion of the GPI. It is worth noting that most GPI anchors from
T. cruzi contain the fourth Man residue distal to GlcN
[23
, 30
, 31
,
33
34
35
36
37
38
39
, 41
] (Fig. 2A
and 2B) . Whether the
presence of this Man residue is critical for the
cytokine-/NO-inducing activity of these GPIs is still an open question.
The TNF--inducing activity of P. falciparum GPIs is also
dependent on the presence of an intact GPI structure. Chemical
treatments with fluoridic acid, which breaks the linkage between the
phosphate and the myo-inositol residue, and with nitrous
acid, which breaks the bond between the GlcN and
myo-inositol residues, completely abolished the GPI
TNF--inducing activity. Removal of both fatty acids at
sn-1 and sn-2 positions and the extra fatty acid
at the C-2 of myo-inositol by alkaline hydrolysis with
methanolic ammonia also completely abrogated TNF- induction.
Surprisingly, the fatty acid (oleic or linoleic) at sn-2
does not seem to be essential for eliciting TNF-, because the
sn-2 lyso-GPI species in the 250–1000 nM range has the same
activity as the native, fully-acylated Plasmodium GPI
[25
]. This result is in apparent contradiction with
previous observations with T. cruzi-derived tGPIs, in which
the unsaturated fatty acid at the sn-2 position was critical
for the potent induction of TNF-, IL-12, and NO [20
,
23
]. It is important to mention that the sn-2
lyso-GPI species from P. falciparum still contains two fatty
acid chains, one at the C-2 of the myo-inositol ring and the
other at the sn-1 position of the glycerol moiety (Fig. 2B) .
The latter might serve as a replacement for the absent fatty acid at
the sn-2 position in the lyso-GPI species.
It is interesting that branched iM4-GIPLs
[Man1-2Man1-6(Man1-3)Man1-4GlcN1-6-PI] from infective
promastigote forms of Leishmania mexicana
and lipophosphoglycan (LPG) from the infective promastigote or
amastigote form of L. donovani and L. major
[30
, 31
, 49
, 50
]
at concentrations up to 5 µM did not elicit TNF- or NO
production by murine macrophages [18
, 21
].
We carefully looked at the primary structure of Leishmania
LPG and noticed that it contains a glycan motif distinct from other
protozoan GPIs (Fig. 2)
. None of Leishmania GIPLs or LPG
species thus far described contain the fourth Man1-2-linked residue
distal from GlcN [30
, 31
]. Maybe the lack
of this residue, as in the case of Plasmodium
Man3-GlcN-PI [25
], makes the
Leishmania LPG and GIPLs completely inactive as a macrophage
activator. In addition, the LPG and GIPL species from Leishmania
spp. have longer (C18:0–C26:0) alkyl chains at sn-1
position [30
, 31
, 51
], which
may turn them less soluble and, perhaps, with a conformation not so
adequate for effectively interacting with the putative macrophage
receptor.
|
CHARACTERIZATION OF THE SIGNALING PATHWAY IN MACROPHAGES TRIGGERED BY T. CRUZI-DERIVED GPI ANCHORS |
|---|
|
TOPABSTRACTINTRODUCTIONGPI ANCHORS: CORRELATION BETWEEN...CHARACTERIZATION OF THE...IDENTIFICATION OF MACROPHAGE...COMPARATIVE STRUCTURAL ANALYSIS...IMPLICATIONS OF THE RESISTANCE...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
and IL-12 synthesis by macrophages exposed to
tGPI-mucin or tGPI. Furthermore, tGPI and tGPI-mucin were able to
induce phosphorylation of I
B and the use of SN50 peptide, an
inhibitor of nuclear factor (NF)-B translocation, resulted in 70%
of TNF- synthesis, only marginally affecting the production of IL-12
by macrophages exposed to tGPI-mucin or tGPI [28
].
In addition, we have shown that cAMP-mimetic or -enhancing agents are
efficient inhibitors of TNF- and IL-12 synthesis while increasing
IL-10 production in macrophages exposed to tGPI-mucin. The inhibitory
effect cAMP-mimetic or -enhancing agents on TNF- and IL-12 synthesis
was partially reversed by the PKA inhibitor H-89. Exogenous IL-10 was
also shown to effectively inhibit the synthesis of proinflammatory
cytokines by macrophages exposed to tGPI-mucin or tGPI. It is
interesting that in macrophages pre-treated with cAMP-mimetic or
-enhancing agents, endogenous IL-10 is apparently involved in the
regulation of IL-12 but not of TNF- synthesis stimulated by T.
cruzi glycolipids [26
].
|
IDENTIFICATION OF MACROPHAGE COUNTERPART RECEPTORS FOR T. CRUZI-DERIVED GPI ANCHORS |
|---|
|
TOPABSTRACTINTRODUCTIONGPI ANCHORS: CORRELATION BETWEEN...CHARACTERIZATION OF THE...IDENTIFICATION OF MACROPHAGE...COMPARATIVE STRUCTURAL ANALYSIS...IMPLICATIONS OF THE RESISTANCE...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
B [28
]. In addition, similar
IC50 values of inhibitors specific for different MAPKs,
NF-B, and cAMP-mimetic or -enhancing agents are required to inhibit
the synthesis of different cytokines (i.e., TNF- and IL-12) in
macrophages exposed to LPS, tGPI-mucin, or tGPI [26
,
28
]. Furthermore, our studies demonstrate that
pretreatment with LPS or tGPI-mucins effectively induces a state of
cross-tolerance between these microbial stimuli [28
].
Considering the recent studies describing the involvement of the
Toll-like-receptors (TLR) on macrophage responsiveness to bacterial
glycolipids/lipopeptides [52
53
54
55
56
57
58
59
], we have investigated
the role of TLR2 and TLR4 on induction of cytokine and NO synthesis by
macrophages exposed to GPI-mucins and GIPLs derived from different
T. cruzi developmental stages. Activation of TLRs leads to
translocation of the transcription factor NF-B. Therefore, we used
Chinese hamster ovary (CHO) cells transfected with TLR2 and/or TLR4 as
well as the reporter CD25 (IL-2 receptor) gene, under a promoter that
is highly responsive to NF-B [60
], to test the
activity of the T. cruzi glycolipids [29
].
Our results show that tGPIs or tGPI-mucins, in the 1–10 nM range,
trigger maximal surface expression of CD25 in CHO cells transfected
with TLR2 but not with TLR4. Consistent with our findings in
macrophages, GPI-mucins and GIPLs derived from the epimastigote stage
of T. cruzi in the 100 nM–1 µM range, were able to
trigger significant CD25 expression in the surface of CHO cells
transfected with TLR2. More importantly, by using cells from TLR2 or
TLR4 knock-out mice, our results indicate the essential involvement of
TLR2, but not TLR4, in the induction of IL-12, TNF-, and NO by
murine-inflammatory macrophages activated by tGPI or tGPI-mucins
[29
].
Thus, the parasite GPI-anchored molecules share with other microbial lipoconjugates, such as bacterial LPS/lipid A (from Gram-negative bacteria); triacylated lipoproteins (LP) and lipoteichoic acid (LTA; from Gram-positive bacteria); peptidoglycan (PG); Mycoplasma-diacylated macrophage-activating lipoprotein/lipopeptide (MALP-2); and Mycobacterium lipoarabinomannan (LAM), the ability to efficiently trigger various functions in cells from macrophage lineage [54 ]. However, we are intrigued by the fact that the specificity of TLR2 allows the recognition of molecules with such distinct, primary structure. Thus, in the next section, we will carry out a brief molecular comparative analysis between bacterial lipoconjugates and protozoal GPI anchors aiming to visualize some of their common structural features and individual characteristics, which could explain, at least in part, why these molecules use a similar class of receptors.
|
COMPARATIVE STRUCTURAL ANALYSIS OF PROTOZOAN AND BACTERIAL GLYCOLIPIDS/LIPOPEPTIDES THAT ACTIVATE CELLS FROM INNATE IMMUNE SYSTEM |
|---|
|
TOPABSTRACTINTRODUCTIONGPI ANCHORS: CORRELATION BETWEEN...CHARACTERIZATION OF THE...IDENTIFICATION OF MACROPHAGE...COMPARATIVE STRUCTURAL ANALYSIS...IMPLICATIONS OF THE RESISTANCE...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
In the last few years, an increasing number of studies have clearly described the involvement of TLRs in the recognition of lipid-containing PAMPs, such as LPS/lipid A, LTA, LP, MALP, and LAM. The lipid-containing PAMPs and protozoal GPI anchors share two basic motifs or features: one hydrophobic moiety, consisting of at least one lipid substituent; and a hydrophylic cap, comprising water-soluble or polar components, such as carbohydrates, amino acids, cyclic alcohols (e.g., myo-inositol), and phosphorylated substituent groups (e.g., PO4-, EtNP, and AEP). The overall structures of the major lipid-containing PAMPs, most of them known to be recognized by TLRs, are shown in Figure 3 [23 , 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 ].
|
It is also interesting that in some PAMPs that activate TLR2, the lipid
moiety is connected to the glycan portion through a phosphoinositol
moiety, as in GPI and LAM, but not in LTA and MALP. The
myo-inositol ring can be further modified by adding mainly
C16:0-fatty acid at C-2, as in P. falciparum GPI
[24
] (Fig. 2B)
; or an Man residue and a fatty acid
(of unknown nature) at C-2 and C-3, respectively, as in M.
bovis bacillus Calmette-Guerin (BCG) cellular LAM
[66
]. In this type of LAM the
myo-inositol-linked Man residue can be even further
substituted by an acyl chain not yet characterized. The presence of two
to four fatty acids in the cellular LAM makes this compound a poor
inducer of proinflammatory cytokines by human dendritic cells (DCs),
comparable in activity to the complete, deacylated LAM. In contrast,
the parietal LAM, apparently comprising only one single fatty acid at
C-1, namely
12-O-(methoxypropanoyl)-12-O-hydroxy-stearic
acid, can potently induce IL-8 and TNF- by human DCs
[86
87
88
89
].
On the hydrophylic moiety, most lipid-containing PAMPs, except MALP, share a glycan core, whose primary function might be to improve the solubility of the whole molecule. It is interesting that MALPs isolated from different Mycoplasma species and containing distinct N-terminal hydrophilic sequences show similar cytokine-inducing activity [73 , 74 ]. This result is strong evidence that the hydrophylic core of a PAMP primarily serves just to enable an adequate solubility to the lipopeptide during presentation to the TLR-mediated macrophage-activating system. The second function of the hydrophylic moiety of a PAMP, regardless of whether it is a glycan or peptide, might be to provide specific, structural motifs, which could be recognized by putative co-receptor(s) on the plasma membrane of cells of the innate immune system. This happens, for instance, in the case of two distinct forms of LAM, the mannose-capped LAM (ManLAM) and the arabinofuranosyl-terminated LAM (AraLAM), which differ from each other in the number and linkage positions of carbohydrate and other substituents linked to the lipomannan (LM) core containing the PI moiety. As a result, AraLAM isolated from avirulent mycobacteria is a much better activator of murine macrophages and human monocytes than ManLAM, purified from virulent Mycobacterium [88 , 89 ]. The precise importance of the hydrophilic core of most PAMPs in this interaction, however, is still unknown and soon should be the focus of exciting structural and functional studies.
|
IMPLICATIONS OF THE RESISTANCE TO AND PATHOGENESIS OF T. CRUZI INFECTION |
|---|
|
TOPABSTRACTINTRODUCTIONGPI ANCHORS: CORRELATION BETWEEN...CHARACTERIZATION OF THE...IDENTIFICATION OF MACROPHAGE...COMPARATIVE STRUCTURAL ANALYSIS...IMPLICATIONS OF THE RESISTANCE...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
T. cruzi can infect and replicate inside any kind of
nucleated cell in the vertebrate host. If not controlled by the immune
system, the T. cruzi parasites become highly virulent and
leads to generalized infection, which is always fatal in few days. The
fast elimination of the definitive host is an obvious limitation for
parasite persistence in its life cycle. In contrast, the long-living,
chronically infected host will increase the chances of T.
cruzi transmission to the insect vector, thus maintaining the
parasite life cycle in nature. As mentioned above, the early resistance
to T. cruzi is highly dependent on the endogenous synthesis
of IL-12, TNF- and IFN-, which will culminate on the release of
RNI by macrophages and control of parasite replication during the early
stages of infection [8
9
10
]. Therefore, our in vitro
studies suggest a possible role for the tGPI-mucins (or tGPI) in the
activation of cells from macrophage lineage [20
,
21
, 23
], triggering an early resistance
mediated by the innate immunity during the acute phase of infection
with T. cruzi, before the establishment of acquired
immunity. The ability of T. cruzi parasites to evade various
effector functions displayed by cells from the innate immune system has
also been described. We believe therefore that the balance between
early activation of innate immunity by tGPI-mucins and parasite evasion
is an essential step in the T. cruzi life cycle, which
determines the survival of the definitive host by limiting parasite
replication; however, this is not sufficient to eliminate the parasite
from host tissues.
Another possible role for tGPI-mucins (or tGPI) in the pathophysiology of Chagas’ disease is the potentiation of the inflammatory reaction at the site of parasitic infection. It has been shown that infection with T. cruzi trypomastigotes (or exposure to tGPI-mucins) triggers macrophages [27 , 92 , 93 ] or cardiomyocytes [94 ] to release chemothatic factors such as the serum amyloid A protein and various chemokines. Therefore, we speculate that tGPI-mucins from the trypomastigote stage of T. cruzi could act as an adjuvant, promoting the inflammatory reaction observed in the infected host tissue, which is regarded as the major cause of tissue damage and organ dysfunction during acute and chronic phases of the disease.
|
CONCLUSIONS AND FUTURE DIRECTIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGPI ANCHORS: CORRELATION BETWEEN...CHARACTERIZATION OF THE...IDENTIFICATION OF MACROPHAGE...COMPARATIVE STRUCTURAL ANALYSIS...IMPLICATIONS OF THE RESISTANCE...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
In regard to the counterpart receptors involved on macrophage activation by GPI anchors from T. cruzi, our data clearly show the involvement of TLR2. It has been suggested that different TLRs work in a combined form [54 , 95 ] so that it will be important to investigate the involvement of other TLRs on macrophage activation by GPI anchors from T. cruzi parasites. The involvement of other co-receptors such as the CD14 or the mannose receptor, involved on macrophage activation by bacterial glycolipids [54 , 95 ] or P. falciparum-derived GPI anchors [25 ], remains to be investigated in the context of macrophage activation by T. cruzi-derived GPI anchors. In addition, the investigation of TLR activation by GPI anchors from other protozoan parasites is an important issue. These and other questions should help to further elucidate the molecular interactions involved on macrophage activation by protozoan parasites and provide new insights to avoid excessive inflammatory responses during infection with these parasites.
Finally, the importance of the proinflammatory activity of the T. cruzi-derived GPI anchors in the pathophysiology of Chagas’ disease remains to be clarified. The identification of TLRs as important receptors for macrophage activation by T. cruzi-derived GPI anchors and the generation of various knock-out mice for such receptors should help to evaluate the importance of these dominant protozoan-derived glycolipids in the parasite:vertebrate host interaction. In addition, TLRs appear to be highly conserved, being ancient receptors that confer a certain degree of specificity to the innate immune system from invertebrate and vertebrate hosts [96 ]. Therefore, the interaction of TLRs and GPI anchors from parasitic protozoa may have the important role of determining the fate of parasitism in insects as well. Further understanding of the interaction of GPI anchors and related structures with receptors from the TLR family may also be helpful to develop new prophylactic strategies to fight the debilitating and often fatal diseases caused by distinct protozoan parasites.
|
ACKNOWLEDGEMENTS |
|---|
Received May 31, 2001; revised July 3, 2001; accepted July 5, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGPI ANCHORS: CORRELATION BETWEEN...CHARACTERIZATION OF THE...IDENTIFICATION OF MACROPHAGE...COMPARATIVE STRUCTURAL ANALYSIS...IMPLICATIONS OF THE RESISTANCE...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
-activated macrophages Infect. Immun. 63,4862-4867[Abstract]
and IL-12 synthesis by macrophages exposed to microbial stimuli Br. J. Pharmacol. 127,1195-1205[Medline]
B phosphorylation for induction of proinflammatory cytokines by macrophages indicates functional similarity of receptors triggered by glycosylphosphatidylinositol anchors from parasitic protozoa and bacterial LPS J. Immunol. 166,3423-3431
