Deficiency in beta1,3-galactosyltransferase of a Leishmania major lipophosphoglycan mutant adversely influences the Leishmania-sand fly interaction.

To study the function of side chain oligosaccharides of the cell-surface lipophosphoglycan (LPG), mutagenized Leishmania major defective in side chain biosynthesis were negatively selected by agglutination with the monoclonal antibody WIC79.3, which recognizes the galactose-containing side chains of L. major LPG. One such mutant, called Spock, lacked the ability to bind significantly to midguts of the natural L. major vector, Phlebotomus papatasi, and to maintain infection in the sand fly after excretion of the digested bloodmeal. Biochemical characterization of Spock LPG revealed its structural similarity to the LPG of Leishmania donovani, a species whose inability to bind to and maintain infections in P. papatasi midguts has been strongly correlated with the expression of a surface LPG lacking galactose-terminated oligosaccharide side chains. An in vitro galactosyltransferase assay using wild-type or Spock membranes was used to determine that the defect in Spock LPG biosynthesis is a result of defective β1,3-galactosyltransferase activity as opposed to a modification of LPG, which would prevent it from serving as a competent substrate for galactose addition. The results of these experiments show that Spock lacks the β1,3-galactosyltransferase for side chain addition and that the LPG side chains are required for L. major to bind to and to produce transmissible infection in P. papatasi.

Protozoan parasites of the genus Leishmania spend the extracellular phase of their life cycle as flagellated promastigotes within the gut of their sand fly vectors. The lipophosphoglycan (LPG) 1 is the most abundant molecule on the surface of Leishmania promastigotes. The molecule consists of a neutral oligosaccharide cap followed by a backbone structure made up of repeating PO 4 -6-Gal(␤1, 4)Man␣1 units, which are linked by a phosphosaccharide core to a phosphoinositide lipid anchor. Species-specific polymorphisms in LPG structure may occur in the structure of the cap and in the composition or number of oligosaccharide side chains that branch from the repeat units (1).
Three types of LPG have been described, based upon the nature of side chain substitutions (2). Type-1 LPG bears no side chains and is represented by the LPGs of the East African Leishmania donovani strains ( Fig. 1) (3). Type-2 LPGs have glycan chains linked to the C-3 of galactose in the repeat units. Included in this group is Leishmania mexicana LPG, which bears glucose-containing side chains on approximately 30% of its repeats; Leishmania tropica LPG, which is highly substituted with over 19 different oligosaccharides (3,4); and Leishmania major LPG, in which virtually every repeat is substituted with ␤1,3-galactose-terminated side chains (Fig. 1). Type-3 LPG is exemplified by Leishmania aethiopica. In this case, the C-2 position of mannose is substituted with a single ␣-mannose residue on approximately 35% of the repeat units.
LPG has been implicated as an adhesion molecule that mediates the interaction of procyclic promastigotes with the midgut epithelium of the sand fly vector (5). In order for Leishmania promastigotes to maintain infection in their sand fly vector, they must be anchored to the midgut epithelium during the passage of the digested bloodmeal. We previously demonstrated the significance of LPG polymorphisms for the establishment and maintenance of infection in Leishmania:vector pairs (6). Specifically, we showed that the ␤1,3-galactose-terminated side chains of L. major are critical for both in vitro binding to and in vivo infection of Phlebotomus papatasi midguts. The finding that only L. major, and not species which lack terminal ␤-galactose-containing side chains, is capable of this interaction implies the species-specific co-evolution of receptor: ligand pairs.
One approach to understanding how the structure of LPG dictates its function is to generate parasites that are deficient in distinct steps of LPG synthesis (6 -10). The structural mutations of LPG can then be correlated with phenotypic characteristics such as infectivity for the vector (5, 6) and host (7). Using this strategy, the present report describes the biochemical and biological characterization of the L. major mutant, Spock. This represents the first description of a defect in LPG side chain biosynthesis resulting from a deficiency in ␤1,3galactosyltransferase activity. The impact of this lesion, an impaired Leishmania-sand fly interaction, provides genetic evidence for previously published data that specific LPG oligosaccharides mediate the binding of Leishmania promastigotes to the sand fly midgut and that this trait alone can affect the ability to produce transmissible infections.

EXPERIMENTAL PROCEDURES
Materials-Materials were obtained as follows: medium 199, Dulbecco's modified Eagle's medium, glutamine, penicillin/streptomycin, and * This work was supported in part by Grant R37 AI20491 (to S. J. T.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Generation of LPG Mutants-Procyclic parasites were mutagenized according to the protocol of Iovannisci et al. (11). Approximately 10 8 parasites (10 6 /ml) were incubated in DMEM containing 8 g/ml Nmethyl-N-nitroso-NЈ-nitroguanidine (4 h, room temperature). The parasites were washed twice with fresh DMEM and cultured for 1 week in complete medium before selection. Parasites mutant in lipophosphoglycan side chain biosynthesis were negatively selected by agglutination with the monoclonal antibody WIC79.3 (12). Promastigotes from log phase cultures were washed twice with cold DMEM and resuspended at 10 7 /ml DMEM containing a 1:100 dilution of WIC79.3 ascites fluid. The suspension was incubated for 0.5 h at room temperature, and the agglutinated parasites were pelleted by low speed centrifugation (1000 rpm, 10 min). The supernatant was collected and centrifuged (3000 rpm, 10 min) to pellet non-agglutinated (i.e. WIC79.3 negative) parasites that were then placed into complete medium and further subjected to 10 rounds of selection. The WIC79.3 negative parasites were cloned by limiting dilution and analyzed by FACS for WIC79.3 reactivity.
FACS Analysis of LPG Expression-Log phase promastigotes were washed once with FACS buffer (phosphate-buffered saline, pH 7.2, 5% fetal calf serum). Washed parasites (10 6 /ml) were incubated (0.5 h, on ice) with 3 g/ml WIC79.3 FabЈ fragments in round-bottom microtiter plates. The parasites were washed twice with FACS buffer and incubated (0.5 h, on ice) with fluorescein-conjugated goat anti-mouse IgG FabЈ (Jackson ImmunoResearch Laboratories). After incubation with the second stage antibody, the parasites were washed twice and resuspended in FACS buffer containing 2% formaldehyde. Stained parasites were analyzed for fluorescence using a Becton Dickinson FACScan (San Jose, CA) and CellQuest software.
Agglutination Assays-Antibodies or lectins (50-l aliquots) were serially diluted with DMEM containing 2% bovine serum albumin into flat-bottomed Immulon-2 microtiter plates. Washed parasites were added in equal volume (final density, 2-3 ϫ 10 6 cells/ml) and gently mixed. The plates were incubated for 0.5 h at room temperature, and the percentage of free parasites compared to control wells (without lectin or antibody) was determined.
Metabolic Labeling and Extraction of LPG-Exponentially growing cells (1-3 ϫ 10 9 ) were metabolically labeled with [ 3 H]galactose (100 Ci) in Dulbecco's modified Eagle's medium for 2 h at room temperature. LPG was extracted and solubilized in Solvent E (H 2 O:ethanol/ diethylether/pyridine/NH 4 OH, 15:15:5:1:0.017) as described elsewhere (13). The LPG extract was evaporated to dryness under N 2 , resuspended in 3 ml of 0.1 M NaCl, 0.1 M acetic acid, and applied to a 1-ml phenyl-Sepharose column equilibrated in the same buffer. The column was washed sequentially with 3 ml of equilibration buffer, 2 ml of 0.1 M acetic acid, and 1 ml of H 2 O. LPG was eluted with Solvent E and evaporated to dryness (13).
Gel Filtration of LPG-The lipid anchor of LPG was removed by resuspending lyophilized LPG in CHAPS buffer (50 mM HEPES, 5 mM EDTA, 0.2% CHAPS, pH 7.4) and incubating with phosphatidylinositolspecific phospholipase C (37°C, 24 h). The resultant phosphoglycan (PG) was dialyzed against water, lyophilized, and resuspended in 40 mM NH 4 OH/1 mM EDTA. The PG was applied to a Sephadex G-150 column (1 ϫ 85 cm) and eluted with 40 mM NH 4 OH, 1 mM EDTA. Fractions were collected (0.6 ml) and counted by liquid scintillation. Void and retention volumes were determined with blue dextran and [ 3 H]mannose, respectively.
Quantification of Repeat Units per LPG-Purified LPG (0.5 mg) was deaminated by incubation (18 h, 37°C) with 300 l of 0.5 M NaNO 2 and 300 l of 0.25 M sodium acetate to remove the phosphatidylinositol anchor and generate the phosphoglycan derivative containing 2,5-anhydromannose at the reducing end. The 2,5-anhydromannose residue was reduced to 2,5-anhydromannitol with the addition of 2 mg/ml NaBH 4 in 1 M NH 4 OH, and then the PG was hydrolyzed with 2 N trifluoroacetic acid to generate phosphorylated and neutral monosaccharides. Phosphorylated monosaccharides (primarily galactose 6-phosphate) were removed by anion exchange chromatography, and the neutral monosaccharides were subjected to alditol acetate derivatization in preparation for gas chromatography-mass spectroscopy. Briefly, monosaccharides were reduced with 2 mg/ml NaBH 4 in 1 M NH 4 OH (2 h, room temperature), and borate was removed by the repeated addition of methanol:acetic acid (9:1, v/v) and evaporation. Dried samples were O-acetylated by incubation (20 min, 121°C) with 0.1 ml of acetic anhydride and 0.1 ml of pyridine and dried under nitrogen. The derivatized monosaccharides were dissolved in 1 ml of dichloromethane and extracted three times with water. The lower phase was dried under nitrogen at room temperature and resuspended in methanol for gas chromatography using an SP2380 column.
In Vitro LPG Biosynthesis-Parasites were harvested at a density of 3-5 ϫ 10 7 cells/ml, and membranes were prepared as described previously (14). Standard incubation mixtures (400 l) contained 0.4 mg of membrane protein, 50 mM HEPES (pH 7.4), 25 mM KCl, 5 mM MnCl 2 , 5 mM MgCl 2 , 0.1 mM N ␣ -p-tosyl-L-lysine chloromethyl ketone, 1 g/ml leupeptin, 1 mM ATP, 0.5 mM dithiothreitol, and 4 Ci UDP-[ 3 H]galactose (specific activity, 15.3 Ci/mmol). For experiments in which LPG was used as an exogenous acceptor for side chain addition, 200 g of purified LPG from either L. donovani or Spock was added. To determine the ability of isolated membranes to synthesize complete LPG, 10 M GDP-mannose was included in the assay. The mixture was incubated for 1 h at 26°C. The reaction was terminated with the addition of 3 volumes of distilled water, and the membranes were pelleted in a microcentrifuge. LPG was released from the membrane pellet and depolymerized by hydrolysis with 40 mM trifluoroacetic acid (100°C, 8 min). The hydrolysate was lyophilized to remove trifluoroacetic acid, and samples were resuspended in 15 mM Tris (pH 9) and dephosphorylated with alkaline phosphatase. The samples were desalted over Bio-Rad AG1-X8 and AG50-X12 columns and resuspended in MeOH: H 2 O (1:1) for application to a Whatman 3-mm chromatography paper and run for 36 h in a descending paper chromatography tank preequilibrated with 1-butanol:pyridine:H 2 0 (6:4:3). The paper was cut into 1-cm segments and counted by liquid scintillation. Migration distances were compared to oligosaccharide standards, which were visualized using the periodic acid/silver nitrate method (15).
Sand Fly Infection and Dissection-P. papatasi sand flies were reared and maintained in the Department of Entomology, Walter Reed Army Institute of Research. 3-5-day-old female sand flies were fed through a chick skin membrane (16) on a mixture of heparinized mouse blood containing 10 6 procyclic promastigotes per milliliter. For infections using procyclic promastigotes, the red blood cells were washed twice in 0.86% NaCl and added back to heat-inactivated (56°C, 45 min) plasma. Blood-engorged sand flies were separated and maintained at 28°C with 30% sucrose. At various times after feeding, the flies were anesthetized with CO 2 , and their midguts were dissected. The number of midgut promastigotes in infected flies was determined by placing individual midguts into a microcentrifuge tube containing 30 l of phosphate-buffered saline, pH 7.4. Each gut was homogenized with a Teflon-coated microtissue grinder, and released promastigotes were counted in a hemocytometer.
In Vitro Assay for Promastigote Binding to Sand Fly Midguts-Binding of promastigotes to sand fly midguts was quantitated by a modification of an in vitro technique (6). 3-5-day-old nonfed female sand flies, maintained on 30% sucrose, were dissected in phosphatebuffered saline. Heads, crops, hindguts, and Malpighian tubules were removed, and the isolated midguts were opened along the length of the abdominal segment with a fine needle. Midguts (7-10 per group) were placed in the concave wells of a microscope chamber slide. Leishmania promastigotes (2.5 ϫ 10 7 ) in 40 l were added to the guts and incubated for 45 min at room temperature. The guts were then individually washed by placing them in successive drops of phosphate-buffered saline. Guts were homogenized, and released promastigotes were counted as described above.

RESULTS
Generation of LPG-defective Mutants-L. major procyclic promastigotes were mutagenized with nitrosoguanidine and subjected to several rounds of negative selection with the monoclonal antibody WIC79.3. This antibody recognizes the mono-and di-␤1,3-galactose side chains of wild-type LPG (Fig.  1) (12, 17). Selection enrichment was monitored by FACS analysis, and WIC79.3 negative parasites were cloned by limiting dilution. Fig. 2 compares by FACS analysis the WIC79.3 reactivity of the wild-type L. major ( Fig. 2A) to that of one mutant, Spock (Fig. 2B), which was chosen for further analysis. Spock has maintained its WIC79.3 negative phenotype through daily passage in laboratory culture. Species typing analyses by polymerase chain reaction amplification of the mini-exon repeats and by isoenzymes validated that Spock was derived from the parental L. major strain.
Lectin and Antibody Agglutination-Spock's lack of WIC79.3 reactivity could be explained either by a complete absence of LPG or by the loss of the specific WIC79.3 epitope on LPG. To discriminate between these possibilities, an agglutination assay with the monoclonal antibody CA7AE was performed (Fig.  3A). CA7AE was raised against L. donovani LPG (Fig. 1) and recognized phosphorylated disaccharide repeats that were un-modified by side-chain oligosaccharides (18). As expected, L. donovani promastigotes were agglutinated at all CA7AE dilutions tested. Appreciable agglutination of wild-type L. major parasites did not occur because the abundant, ␤1,3-galactosecontaining side chains of its LPG mask the CA7AE epitope. In contrast, Spock is significantly agglutinated by high dilutions (1:5000) of CA7AE ascites. This result indicates that Spock synthesizes LPG that is relatively unmodified by side-chain oligosaccharides, similar to that of L. donovani LPG. Fig. 3B shows the results of an agglutination assay with peanut lectin (peanut agglutinin). Peanut agglutinin recognizes terminal galactose residues in a ␤1,3-linkage to the penultimate sugar (galactose, in the case of L. major LPG) (19). While wild-type L. major parasites were completely agglutinated at all lectin concentrations tested, Spock parasites remained unagglutinated at concentrations as high as 50 g/ml, indicating the absence of terminally exposed ␤-galactose residues. This is in contrast to the sensitivity of L. donovani to peanut lectin, which agglutinates the parasite at concentrations above 6 g/ml, reflecting the presence of a terminal ␤-galactose residue in the neutral cap.
Characterization of in Vivo Labeled LPG from Spock-The relative ability of Spock to synthesize LPG was examined by metabolic labeling of the cells with  mannose, extraction of LPG by organic solvents, and purification of the glycoconjugate on a column of phenyl-Sepharose. Incorporation of radioactive galactose and mannose precursors showed a slight decrease and increase, respectively, into Spock LPG compared to wild type (data not shown). To obtain information regarding the relative sizes of the radioactive LPGs, wild-type L. major, Spock, and L. donovani LPGs were delipidated by nitrous acid deamination and then were subjected to gel filtration on Sephadex G-150 (Fig. 4). Wild-type L. major PG eluted first, followed by Spock and then L. donovani, indicating that the wild-type L. major synthesizes a slightly larger glycoconjugate than either the mutant or L. donovani. In addition, the length of each PG was assessed by determining the ratio of mannose:2,5-anhydromannose (27.4 (wild-type L. major), 20.0 (Spock), and 16.4 (L. donovani)). These ratios reflect the relative number of repeats in LPG assuming a similar number of mannose residues in the glycan core and cap domains.
Characterization of the Repeating Units of Spock LPG-To confirm that Spock produces an LPG defective in the synthesis of side chain sugars recognized by WIC79.3 and peanut agglutinin, [ 3 H]galactose-labeled LPGs from wild-type and Spock cells were depolymerized with mild acid to selectively cleave the sugar-1-phosphate bonds of the repeating units of LPG, generating phosphooligosaccharides. After dephosphorylation of the sugars with alkaline phosphatase, the neutral oligosaccharide fragments were resolved by descending paper chromatography along with appropriate standards (Fig. 5). Based on published data (3, 20 -22), the major 3 H-labeled fragments (i.e. repeat units) from the wild-type LPG migrated in a manner consistent with the trisaccharide Gal-Gal-Man (Fig. 5A, fractions 14 -16) and a mixture of tetrasaccharides, Gal-Gal-Gal-Man and Ara-Gal-Gal-Man (Fig. 5A, fractions 7-10). There were lesser amounts of other oligosaccharides. In contrast, Spock LPG revealed a single major [ 3 H]galactose-labeled fragment that co-eluted with the standard Gal-Man disaccharide (Fig. 5B, fractions 25-32).
Confirmatory information about the repeating units was obtained by FACE analysis. LPGs from wild-type L. major, L. donovani, and Spock cells were subjected to mild acid hydrolysis. Aliquots of the hydrolysate were dephosphorylated with alkaline phosphatase. The fragments were derivatized at the reducing ends with a fluorophore and subjected to electrophoresis. Fluorescent bands were visualized by digital UV imaging (Fig. 6). In lane 1, the most prominent fragment is the trisaccharide from the wild-type L. major LPG, with lesser amounts of two tetrasaccharides and the Gal-Man disaccharide. The characteristic disaccharide repeating unit (Gal-Man) of the L. donovani LPG is shown in lane 3. The pattern of fragments from Spock LPG (lane 2) is virtually identical to that of L. donovani, indicating the lack of side chains. There was no change in the pattern of Spock LPG fragments as a function of the growth phase (data not shown).
Analysis of the Defect in a Cell-free System of LPG Biosynthesis-The results from experiments using intact Spock cells suggested that the LPG from these mutants did not have the characteristic side chain sugars initiated by a ␤1,3-galactose addition to the PO 4 -Gal-Man backbone units. To investigate if the defect was due to the absence of ␤1,3-galactosyltransferase activity or a possible modification of the LPG that precludes side chain addition, a cell-free glycosylation assay was used (14). Membranes from wild-type L. major or Spock were incubated with exogenous Spock or L.donovani LPG and UDP-[ 3 H]galactose. The repeating units of the prelabeled LPG were analyzed for the presence of oligosaccharide side chains by subjecting the material to mild acid hydrolysis and dephosphorylation with alkaline phosphatase. The fragments generated were resolved by descending paper chromatography.
When wild-type or Spock membranes were incubated in the absence of exogenous LPG (Fig. 7, A and D, respectively), no incorporation of [ 3 H]galactose from UDP-[ 3 H]galactose was observed. This experiment confirms the lack of de novo LPG biosynthesis in the absence of exogenously added GDP-mannose. Analysis of fragments from both exogenously supplied L. donovani and Spock LPG (Fig. 7, B and C, respectively) that had been incubated with wild-type L. major membranes revealed the presence of trisaccharide fragments, indicating that both LPGs are competent acceptors for ␤1,3-galactose addition. In contrast, analysis of LPG fragments generated from the incubation of Spock and L. donovani LPG with Spock membranes showed no trisaccharides, indicating that the Spock membranes lack the enzyme for side chain addition (Fig. 7, E and F, respectively). However, Spock membranes are enzymatically active and capable of in vitro assembly of the PO 4 -Gal-Man repeat units when both UDP-[ 3 H]galactose and GDPmannose are included in the incubation medium. Incorporation of the radiolabel into wild-type and Spock LPG was similar (44,600 cpm versus 66,680 cpm, respectively). Analysis of the wild-type LPG fragments showed the presence of tri-and tetrasaccharides (Fig. 8A), while fragments generated with Spock membranes were predominantly disaccharides (Fig. 8B).
In Vitro Binding of Spock to Sand Fly Midguts-Isolated P. papatasi midguts were incubated with wild-type L. major or Spock procyclic promastigotes. After washing, the guts were homogenized, and released parasites were counted. The results are shown in Fig. 9. The average number of wild-type promastigotes per midgut was approximately 7,500, while Spock exhibited significantly less binding (approximately 1,800 parasites/midgut).
Survival of Spock in P. papatasi-2 days after sand flies were membrane-fed on bloodmeals containing wild-type L. major or Spock, all flies in each group harbored similar numbers of viable promastigotes (data not shown). By day 5 postinfection, after passage of the digested bloodmeal, the parasite burden differed substantially in the three groups (wild type, 43,000 parasites/midgut; Spock, 800 parasites/midgut; L. donovani, 28 parasites/midgut), indicating that Spock, like L. donovani, is unable to maintain significant infection in P. papatasi vector (Fig. 10). DISCUSSION The ability of Leishmania parasites to bind to the sand fly midgut during excretion of the digested bloodmeal is essential for the development of transmissible infections. Sand fly vectors can, in some instances, transmit only certain species of Leishmania. Such species-specific differences in vectorial competence have been directly correlated with the ability of promastigotes to attach to the sand fly midgut, the variable outcomes of which are controlled by structural polymorphisms in LPG. The leishmanial receptor of P. papatasi is highly specific for the galactose-terminated oligosaccharide branches of L. major LPG (6). L. major, being the only Old World Leishmania to express abundant side chains terminating in galactose, appears to have yielded to a selective pressure that enables it to exploit a widespread phlebotomine species that is otherwise refractory to parasites expressing relatively unsubstituted LPGs.
The present report describes the interaction of a L. major LPG mutant, defective in ␤1,3-galactose addition, with the sand fly vector. This work furnishes genetic evidence for the previously published contention that the binding of L. major promastigotes to the midgut epithelium is mediated by the extensive side chain oligosaccharides (5,6). The mutant, Spock, was generated by nitrosoguanidine treatment of wild-type L. major and negative selection with the monoclonal antibody WIC79.3. This monoclonal antibody is specific for the galactose-terminated oligosaccharide side chains of L. major LPG (7) and does not recognize this moiety on other glycoconjugates such as glycoproteins (23). Therefore, the use of WIC79.3 allowed us to specifically isolate parasites defective in LPG sidechain biosynthesis. The lack of these side chains diminished Spock's ability to bind to P. papatasi midguts in in vitro binding assays and to maintain infections in sand flies after passage of the bloodmeal. Thus, the mutant L. major Spock displayed a biological phenotype similar to that of L. donovani. The data provide strong evidence that the inability of L. donovani to utilize P. papatasi as natural vector can be explained solely on the basis of its LPG lacking appropriate side-chain additions, regardless of any other interspecies molecular differences that might exist.
Analysis of depolymerized, dephosphorylated LPG fragments from Spock indicated that the defect in Spock LPG biosynthesis lies in the ␤1,3 addition of galactose to the 3-hydroxyl of the galactose residue in the disaccharide repeats of the LPG backbone. Membranes from wild-type L. major but not from Spock were able to add galactose-initiated side chains to Spock LPG, demonstrating that the molecule is a competent substrate for oligosaccharide modification. It follows that the genetic defect of Spock involves the deficiency of a glycosylation enzyme (i.e. the absence of a ␤1,3-galactosyltransferase) rather than a covalent modification of the Gal-Man repeat unit, which would prevent side chain addition to the galactose residue. The lack of ␤1,3-galactosyltransferase activity in L. donovani has been demonstrated previously by Ng and co-workers (14).
A mammalian ␤1,3-galactosyltransferase has been de- FIG. 8. Ability of wild-type and Spock membranes to synthesize LPG in vitro. LPG was extracted from wild-type (A) and Spock membranes (B), depolymerized, and alkaline phosphatase-treated prior to paper chromatography. Glc6 to Glc2 indicate migration distance of oligoglucose standards.
FIG. 9. Attachment of wild-type and Spock promastigotes to P. papatasi midguts in vitro. P. papatasi midguts were dissected and incubated with promastigotes, and binding was determined as described under "Experimental Procedures." scribed, the deficiency of which leads to Tn-syndrome (24,25). When the Tn antigen (O-linked ␣N-galactosamine) or its sialylated derivative is not ␤1,3-galactosylated, hematopoietic cells bearing the antigen bind serum anti-␣-GalNAc antibodies, causing the mild hemolytic anemia and thrombocytopenia observed in patients with this rare acquired disorder. Preliminary studies indicate that the transferase deficiency in Tn-syndrome results from the silencing of the galactosyltransferase gene or possibly its transcriptional control unit by DNA methylation. However, the enzyme has not been isolated nor has its gene been cloned, so molecular studies of the deficiency have not been performed.
Spock represents the first example of a Leishmania mutant defective in assembly of side chain sugars of LPG. Thus, Spock provides an important tool for the delineation of the biosynthetic pathway and functional significance of LPG. Experi-ments are currently underway to isolate the galactosyltransferase gene by functional complementation of Spock with wildtype L. major DNA using methods that enabled the isolation and characterization of the putative galactofuranosyltransferase gene of L. donovani (10).