OUP user menu

Histochemical Localization of N-Acetyl-Galactosamine in the Midgut of Lutzomyia longipalpis (Diptera: Psychodidae)

L. G. Evangelista, A.C.R. Leite
DOI: http://dx.doi.org/10.1603/0022-2585-39.3.432 432-439 First published online: 1 May 2002


The binding of lectins to the midgut of the female sand fly Lutzomyia longipalpis (Lutz & Neiva) was investigated using lectin-gold conjugates. Midguts from laboratory-reared flies provided fructose solution and/or blood fed on hamster were dissected at 6, 24, and 48 h and at 5 and 7 d after feeding. Before examination by transmission electron microscopy, each midgut was sectioned, incubated with lectins from four sources (Canavalia ensiformis [ConA], Helix pomata agglutinin [HPA], peanut agglutinin [PNA], and wheat germ agglutinin [WGA]), then conjugated with colloidal gold. Only HPA, which is specific for N-acetyl-galactosamine (GalNAc), bound to the midgut. Binding sites were cytoplasmic secretory granules and microvilli throughout the length of the midgut epithelium. Binding occurred in sand flies fed fructose as well as in flies receiving a blood meal. The presence of GalNAc on the midgut microvilli of sand flies before, during, and after blood feeding indicates this amino-sugar is not altered by digestion. As a structural component, GalNAc may represent a terminal on a receptor molecule. The failure of the sand fly peritrophic matrix to bind WGA by N-acetyl-glucosamine may be caused by the complex composition of the membrane, which renders N-glycan inaccessible to the lectin-gold conjugate.

  • Lutzomyia longipalpis
  • Phlebotominae
  • midgut
  • N-acetyl-galactosamine
  • histochemistry

The phlebotomine sand fly Lutzomyia longipalpis (Lutz & Neiva) has considerable medical and veterinary importance in the Neotropics as the main vector of Leishmania chagasi Cunha & Chagas, a causative agent of American visceral leishmaniasis in mammals, including humans (Tesh 1995, Uribe 1999). To transmit L. chagasi, the sand fly must take a blood meal from an infected host. The amastigote form of the parasite multiplies and develops in the midgut lumen to the infective promastigote form, which is inoculated into a new host when the insect takes a second blood meal (Elnaiem et al. 1994). Several attempts have been made to elucidate the role of the molecules that are important in the vector-parasite relationship, using L. longipalpis or other phlebotomines (Schlein and Romano 1986, Borovsky and Schlein 1987, Davies et al. 1990, Ingram and Molyneux 1991, Lang et al. 1991, Schlein et al. 1991, Pimenta et al. 1992, Schlein et al. 1992, Dillon and Lane 1993a, b, Pimenta et al. 1994, Sacks et al. 1994, Sacks et al. 1995, Mohareb et al. 1998, Dillon and Lane 1999, Jacobson and Schlein 1999, Mahoney et al. 1999, Stierhof et al. 1999, Sacks et al. 2000). These molecules include the lectins (agglutinin). Lectins are proteins or glycoproteins that occur in plants, animals and micro-organisms. They have no immune origin but specifically bind carbohydrates or glycocomplexes and mediate various biological processes (Lis and Sharon 1986, Rudiger 1998, Vijayan and Chandra 1999). Based on agglutination studies, endogenous lectins, or lectin-like molecules, have been detected in the hemolymph and tissue lysates of phlebotomine sand flies, including L. longipalpis (Wallbanks et al. 1986, Ingram and Molyneux 1991, Volf 1993, Volf et al. 1994, Svobodová et al. 1996, Volf and Killick-Kendrick 1996, Volf et al. 1998).

Lectins conjugated with colloidal gold particles have been used in electron microscopy studies to probe for sugar molecules in different cellular compartments and extracellular structures (Roth 1983). Although a few ultrastructural studies have used lectin-gold to bind molecules of medical and veterinary important insects (Peters et al. 1983, Dorner and Peters 1988, Rudin and Hecker 1989), none of these studies have probed the sand fly midgut. In the current study lectin-gold conjugates were used to bind sugars in the midgut of L. longipalpis, and the binding sites studied by transmission electron microscopy (TEM).

Materials and Methods

A colony of L. longipalpis was maintained as described by Modi and Tesh (1983) in the Department of Parasitology, Institute of Biological Sciences of the Federal University of Minas Gerais. Newly emerged female sand flies were provided 30% fructose solution ad libitum for 3 d, then allowed to blood feed on hamsters. The engorged females were maintained at 25°C and 95% RH and fed 30% fructose solution ad libitum. Groups of 10 females that were fed on fructose or had taken a single blood meal were dissected after 6, 24, and 48 h and at 5 and 7 d after feeding and the midgut removed. The midguts were fixed in a mixture of 3% paraformaldehyde and 0.5% glutaraldehyde in 0.1 M phosphate buffer (PBS, pH 7.4) at 4°C for 3 h, dehydrated in an ethanol series, and embedded in L.R. White acrylic resin (London Resin Company, Basingstoke, UK) as described by Lang et al. (1991). Sections were cut with a Reichert-Jung microtome and mounted on glass slides and nickel grids. Semithin sections of midguts were mounted on glass slides and examined to select material suitable for ultra-thin sectioning. Photomicrographs were taken using a Leica DC 100 digital camera. Each lectin-gold conjugate (Table 1) was diluted in a 0.15 M PBS solution to which 0.5% bovine serum albumin (BSA) and 0.05% Tween 20 was added. Grids with ultra-thin sections were incubated by floating them face down in PBS/BSA/Tween for 30 min and then in lectin-gold for 60 min. The sections were washed five times in Milli-Q water, and stained for 10 min each in successive solutions of 4% osmium tetroxide, saturated uranyl acetate, and lead citrate. As a negative control, lectin-gold was neutralized (premixed) for 60 min with the corresponding anti-lectin (Sigma, St. Louis, MO) that had been diluted 1:100. Sand fly cuticles were processed as described above and incubated with the WGA-gold to serve as a positive control. Samples were examined and electro-micrographed using a Zeiss EM-10 TEM.

View this table:
Table 1


The midgut of unfed L. longipalpis before (Fig. 1A) and after excision (Fig. 1B), is thin and inconspicuous. TEM of semithin sections revealed a monolayer of epithelial cells surrounding a lumen filled with meconium (Fig. 1C). Six hours after engorgement, the midgut epithelium remained in contact with the blood and the peritrophic matrix could not be detected (Fig. 1D). The midgut with blood 48 h after feeding is shown in Fig. 1E. A semithin section of the midgut 24 h after blood feeding exhibits a peritrophic matrix layer thicker than the epithelial tissue layer (Fig. 1F). Five to 7 d after blood feeding, the midguts are thin and transparent (Fig. 71G) and have a thin monolayer of epithelial cells (Fig. 1H).

Fig. 1

Photomicrographs of midguts from adult female Lutzomyia longipalpis. (A) Inconspicuous midgut (arrow) inside sand fly that had fed only on fructose (mounted in Canada balsam). (B) Idem, after removal from sand fly (in fixative solution). (C) Longitudinal semithin section from sand fly that had not taken a blood meal (stained with toluidine blue). (D) Transverse semithin section 6 h after taking a blood meal (stained with toluidine blue). (E) Midgut removed from sand fly 48 h after taking a blood meal (in fixative solution). (F) Transverse semithin section of midgut 24 h after taking a blood meal (stained with toluidine blue).(G)Midgut removed from sandfly 5 d after taking a blood meal, containing air bubble (arrow) (in fixative solution). (H) Longitudinal semithin section of sand fly midgut 5 d after taking a blood meal (stained with toluidine blue). Bm, blood meal; Ep, epithelium; Mgl, midgut lumen; Me, meconium; Pm, peritrophic matrix; Mt, Malpighian tubule. Scale bar = mm.

When ultrathin midgut sections were incubated with lectin-gold conjugates of Con A, PNA, WGA, and HPA at different dilutions, only HPA bound to the epithelium of the fructose-feeding L. longipalpis. GalNAc was observed in cytoplasm and, perhaps, within secretory granules (Fig. 2A) and microvilli (Fig. 2B). In blood fed L. longipalpis after 6 h (results not shown), 24 h (Fig. 3A), and 48 h (results not shown) and 5 (results not shown) and 7 d (Fig. 3B), the cytoplasmic granules and microvilli were bound only by HPA-gold. The peritrophic matrix that formed in the insect midgut 24–48 h after blood feeding did not bind any lectins. An untreated midgut epithelium is shown in Fig. 4. The positive control (Fig. 5) showed a strong affinity for N-acetyl-glucosamine present in the fly cuticle.

Fig. 2

Electron micrographs of the epithelium midgut of Lutzomyia longipalpis that had fed only on fructose, bound by HPA-gold. (A) Cytoplasmatic secretory granules, transverse section; 69,740×. (B) Microvilli (arrowhead), transverse section; 69,740×. Mi, mitochondrion; Mv, microvilli; Nu, nucleus.

Fig. 3

Electron micrographs of the midgut epithelium of Lutzomyia longipalpis after feeding on blood, bound by HPA-gold. (A) Cytoplasmatic granules and microvilli after 24 h, longitudinal section; 55,440×. (B) Microvilli bound after 7 d, longitudinal section; 88,220×.

Fig. 4

Electron micrograph of the midgut epithelium of Lutzomyia longipalpis (negative control), longitudinal section; 22,040×.

Fig. 5

Electron micrograph of Lutzomyia longipalpis adult female cuticle bound by WGA-gold (positive control); 69,740×.


Midguts from female L. longipalpis fed fructose or blood, bound only HPA indicating the presence of GalNAc in the epithelial midgut. This finding indicates that expression of the GalNAc is not linked to the particular food source ingested by L. longipalpis. The GalNAc may be a component of secretory molecules in the cytoplasm, and also in the microvillar cytoskeleton of L. longipalpis, and is probably associated with the insect glycocalyx (Lane et al. 1996). Although the control mechanism of GalNAc is not known, there is evidence that sand fly midgut enzymes (Schlein et al. 1983, Dillon and Lane 1993a, Mahmood and Borovsky 1993) and lectins (Volf and Palanová 1996) are regulated according to the food type and the time since ingestion and that such molecules directly or indirectly influence the survival and forward migration of Leishmania within the vector (Schlein et al. 1983, Borovsky and Schlein 1987, Volf et al. 1998, Svobodová 2000). A possible linkage of GalNAc synthesis with L. longipalpis digestive enzymes biosynthesis cannot be disregarded. Trypsin has been demonstrated by electro-immunocytochemistry in the epithelium and lumen of the midgut of Stomoxys calcitrans (L.) (Jordão et al. 1996). In contrast, the lepidopteran larval midgut expresses GalNAc as part of the glycoprotein aminopeptidase, which is thought to be a receptor for the CrylAc toxin of the bioinsecticide Bacillus thurigiensis Berliner (Burton et al. 1999). GalNAc may be not be regarded a component of lectin-like molecules in midgut extracts of sand flies, because the amino-sugar was not an effective inhibitor in several tests, including those using L. longipalpis (Volf et al. 1994).

Lipophosphoglycan (LPG), the major surface glycoconjugate of the promastigotes, is responsible for attachment to, and detachment from, the microvillar midgut epithelium during metacyclogenesis after blood meal digestion (Davies et al. 1990, Lang et al. 1991, Sacks et al. 1994, Mahoney et al. 1999, Sacks et al. 2000). The LPG of noninfective procyclic promastigotes of Leishmania major Yakimoff & Soskor bears repeated galactose terminal branches that act as ligands and attach to the microvilli of Phlebotomus papatasi (Scopoli). Lengthening of the phosphoglycan chain reduces accessibility to terminal galactose units and substitution by arabinose results in detachment of the promastigotes, which results in infective metacyclic forms (Pimenta et al. 1992; Pimenta et al. 1994; Sacks et al. 1994, 1995; Saraiva et al. 1995). Midgut receptors for the stage- and species-specific Leishmania tropica (Wright) procyclic LPG have been described as polypeptides of MW 65 kDa, or less, in P. papatasi (Dillon and Lane 1999). Intraspecific differences in midgut attachment between Sudanese and Indian strains of L. donovani Ross in P. argentipes Annandale & Brunetti are caused by polymorphism of LPG; terminal mannose and galactose groups were lost because of clustering and folding of the longer metacyclic molecule in parasites from the Sudanese strain (Sacks et al. 1995). In the Indian strains, glucose terminal branches are lost during metacyclogenesis (Mahoney et al. 1999). The midgut receptors of P. argentipes are unknown. If the polymorphism of LPG in L. chagasi (Sacks et al. 1994) is similar to that of L. donovani from the Sudan (Sacks et al. 1995), it is possible that the midgut epithelium of L. longipalpis also has receptors for procyclic promastigotes similar to those in P. argentipes.

Dorner and Peters (1988) did not find sugar molecules on the midgut epithelium of larval Aedes aegypti (L.), Anopheles stephensi (Liston), Culex quinquefasciatus Say and Odagmia ornata Meigen, using lectin-gold conjugates, or Ulex europaeus agglutinin (DBA, specific for fucose). Based on a range of techniques, chitin was described as a minor component (7.2%) of the weight of peritrophic matrix from Lucilia cuprina (Wiedemann) larvae (Tellam and Eisemann 2000). In contrast, lectin-gold conjugates, with affinity to Man or Glc, Gal, GalNAc, and GlcNAc, were observed on the epithelium (secretory vesicle and glycocalyx) of adult mosquitoes, with preferential binding of GalNAc in An. stephensi and GlcNAc in Ae. aegypti (Rudin and Hecker 1989); Con A was also observed in the epithelial intercellular spaces of both mosquitoes. Although the mesh-like peritrophic matrix is extracellular, (Peters 1992, Eisemann and Binnington 1994), it was not marked by any lectins in L. longipalpis 24 or 48 h after a blood meal. Peters et al. (1983) located sugar residues in invertebrates using lectin-gold conjugates and identified Man or Glc on the peritrophic matrix of Calliphora erythrocephala Meigen larvae by Con A; this lectin also bound the peritrophic matrix of larval Nematocera (Dorner and Peters 1988). Although the peritrophic matrix of mosquitoes has Gal, GalNAc and GlcNAc (terminal sugars that are unbound by Con A), An. stephensi expressed more GalNAc than Ae. aegypti, with the latter molecule being preferentially marked (Rudin and Hecker 1989). Chitin, which comprises holopolymers of GlcNAc units, has been described in P. chinensis Newstead, P. morgolensis Sinton and P. squamirostris Newstead (Feng 1951); P. papatasi (Blackburn et al. 1988); P. perniciosus Newstead (Walters et al. 1993) and L. spinicrassa Morales-Alarcon et al. (Walters et al. 1995) based on studies that used other than gold conjugates. A negative reaction to chitin was observed in the peritrophic matrix of P. longipes Parrot & Martin (Gemetchu 1974). The peritrophic matrix of L. longipalpis may be inaccessible to labeling by WGA-gold, due to its complex chemical structure and composition, as well as the probable interaction of chitin with other molecules (Tellam et al. 1999). In our study, this was tested, and found to be true, after exposing WGA lectin to sand fly cuticle as a positive control.

Further ultrastructural studies similar to ours need to be made to complement biochemical investigations. Ultimately, knowledge gained in this way will help us find ways to disrupt the sand fly vector-Leishmania parasite relationship.


To Bruce Alexander for revision of the original manuscript. This work was partially supported by "Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)" and "Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG)."

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/), which permits non-commercial reuse, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions{at}oup.com

References Cited

View Abstract