trans

236 TRYPANOSOMATID CARBOHYDRATES
schaudinni, Herpetomonas samuelpessoai and
Crithidia fasciculata. The Leptomonas glycophosphosphingolipid
resembles that of
Leishmania in the -linked Man in the GPI
core, and resembles that of T. cruzi in the presence
of aminoethylphosphonate and ceramide
lipids. The Endotrypanum GIPLs are similar
to Leishmania Type II and hybrid GIPLs but
are sphingolipids rather than glycerol lipids.
Several Endotrypanum GIPLs contain D-arabinose,
and the terminal saccharide chains
resemble the phosphoglycan side chains of
L. major LPG. The Leishmania GIPLs contain
predominantly -linked Gal while Endotrypanum
contains -linked Gal. Such similarities
are consistent with the close evolutionary
relationship between Leishmania, trypanosomes,
Leptomonas and Endotrypanum.
GP63
GP63 (Figure 10.4B) is the major cell surface
glycoprotein of Leishmania promastigotes with
500 000 copies per cell, accounting for 1% of
all protein. In amastigotes, GP63 is expressed
to a lower level and is found in the flagellar
pocket as opposed to covering the entire surface,
as in promastigotes. GP63 is a 63 kDa zinc
metalloprotease and is GPI-anchored to the
cell surface via a myristic acid-containing GPI
anchor. The amastigote GP63 subpopulation
found in the flagellar pocket lacks a membrane
anchor. An active site structural motif
found in other zinc proteases has been identified
in structural studies that may aid design
of specific inhibitors. GP63 contains three
potential glycosylation sites, and the N-linked
glycans have been characterized in L. mexicana
and L. major. The glycans are biantennary
high-mannose type, and some bear a terminal
glucose in 1,3 linkage. Man 6 GlcNAc 2 and Glc
Man 6 GlcNAc 2 are present in all promastigotes
species examined. In amastigotes the structures
are more variable, while in L. donovani no N-
linked glycans appear to be present. The terminal
glucose in the GP63 glycan is highly unusual
with respect to oligomannose structures
found in glycoproteins. Whether the stagespecific
changes in glycan structure affect parasite
infectivity and development is unknown.
The abundance and location of GP63 suggest
that it may play a role in the infection
process. GP63 expression is decreased in
avirulent compared to virulent promastigotes,
and GP63 is found in increased amounts in
stationary phase L. braziliensis. By virtue of
its activity, GP63 may interact with and proteolytically
degrade host macromolecules.
Indeed, solubilized GP63 molecules bind and
cleave the human plasma protease inhibitor
2-macroglobulin. The formation of 2- macroglobulin-GP63
complexes results in a dosedependent
inhibition of the proteolytic
activity of soluble GP63 on azocasein. However,
2-macroglobulin may not interact with promastigotes
in the bloodstream, since GP63 at
the surface of live promastigotes does not recognize
2-macroglobulin. GP63 may also participate
in the attachment of promastigotes
to macrophage surface receptors via complement
components and protect the parasite
against complement-mediated lysis. Attempts
to obtain mutants that are defective in GP63
by targeted gene deletion have been hindered
by the fact that GP63 is encoded by a multigene
family. Targeted deletion of six out of
seven GP63 genes did not affect parasite growth
in vitro or prevent disease in mice. Knockouts
of GPI8, the GPI:protein transaminidase which
eliminates the expression of GP63 along with
other GPI-anchored proteins, grew normally
in culture and macrophage infectivity in vitro
was unaffected. More importantly, the GPI8
mutant was able to establish infection in mice,
suggesting that GP63 is not essential for growth
or infectivity in mammals.
BIOCHEMISTRY AND CELL BIOLOGY: PROTOZOA