PRODUCTION AND CHARACTERIZATION OF MONOCLONAL ANTIBODIES TO ANOPHELES TESSELLATUS MIDGUT

The production of monoclonal antibodies (Mabs) against antigen derived from the midgut ofAnopheles tessellatus is described. Three cloned Mabs examined were found to be directed against conformational epitopes on midgut antigens. Ingestion of these Mabs in a bloodmeal did not affect mosquito mortality or fecundity. The intake of the Mabs, when compared to normal mouse IgG, with a bloodmeal containingPlasmodium v i v a gametocytes did not reduce the suscept~bility of the mosquito to parasite infection.


INTRODUCTION
The midgut of the mosquito is composed of a single layer of epithelial cells and plays a n important role in mosquito-pathogen interactions.The ingested bloodmeal is digested in the lumen ofthe midgut by proteolytic enzymes secreted by midgut epithelial cells.'The sexual stages and ookinetes of the malaria parasite Plasmodium can be damaged by mosquito t r y p ~i n .~ The formation of a peritrophic membrane by secretions from the epithelium and the presence of receptors on the epithelium for ookinetes also have a function in establishing a Plasmodium infection in the vector m o ~q u i t o .~ Cellular and humoral immune responses against arthropod vector antigens in vertebrate hosts influence vector physiology and modulate the transmission of pathogen^.^Midgut antigens have been implicated as the targets for such immune mechanism^.^Immune sera produced in rabbits5 and mice6 against mosquito tissues (including midguts from non-blood fed Anopheles tessellatus Theobald) reduced mosquito fecundity and its susceptibility to P. vivax.' Fecundity reduction has been attributed to antibodies binding to target antigens in the gut, causing it to become leaky and thereby permitting,other antibodies to enter the haemocoel to interfere with normal physiological processe~.~Mon~clonal antibodies (Mabs) produced by hybridoma cultures8 represent one of the many different antibodies produced by conventionally immunized animals.Mabs have been used as probes for investigatingmosquito trypsins and yolk protein^.^^-'^It is possible that Mabs to midgut antigens are more effective than polyclonal sera in influencing mosquito fecundity and its infectivity to malaria parasites because they can be used a t high concentrations against specific target molecules.Once biological activity is established characterization of target antigens is also more easily done with Mabs.This paper describes the production and characterization of three Mabs against midgut antigens of An. tessellatus.

METHODS AND MATERIALS
Midgut antigen: Midguts from 2-3 day old sugar fed female An.tessellatus maintained in the laboratory13 were dissected in phosphate buffered saline (PBS) pH 7.2 a t 4OC and stored a t -20°C until used for preparation of antigen.Midguts from 50 mosquitoes were pooled, homogenized and then lyophilized.

Immunization of mice:
Lyophilized antigen was reconstituted i n distilled water a t a concentration of approximately 0.5 mglml protein.Balblcmice were injected intra-peritoneally with 0.2 ml antigen as a 1:l emulsion with Freund's complete adjuvant.A second immunization with 0.2 ml antigen was performed as a 1:l suspension of Freund's incomplete adjuvant 3 wks later.The mice were boosted 12 wks later with 0.15 ml antigen without adjuvant and the spleen removed for fusion 3d later.
-Monoclonal antibody production and assay: Spleen cells from two immunized mice were pooled and fused with cells of the non-immunoglobulin secreting myeloma cell line, Ag8-653 in 38% polyethylene glycol according to established t e c h n i q ~e s .~J ~ The cells were cultured in medium containing hypoxanthine, aminopterin, thymidine (HAT) in 96 and 24 well plates.A dot immunobinding assay modified from Hawked5 was carried out to screen hybridoma culture supernatants for antibodies.Midgut antigen was applied directly (at 1-2 midgut equivalents/ spot) onto a nitrocellulose filter marked with a grid, air dried and then treated with 3% bovine serum albumin for 3h to block non-specific binding sites on the filter.Hybridoma culture supernatants (first antibody) were added individually onto each antigen spot for 2h a t 23OC; the filters were washed in PBS and thenincubated with peroxidase conjugated sheepanti-mouse immunoglobulin (Silenus, Australia) for l h a t 23OC.The reaction was visualized with chloronaphthol and H,0,.16Four hybridomas identified as positive in the immunobinding assay were cloned by limiting dilution in 96 well plates in a medium containing spleen cells.When clones were well grown, the culture supernatants were tested again by the immunobinding assay for specific antibodies.Strongly positive clones each derived from hybridomas PlG4,24C6 and 24D4 were grown in 24 well plates and then propagated in tissue culture flasks.
Ascites were produced by injecting 5 x lo5-lo6 cloned hybridoma cells into pristane or Freund's incomplete adjuvant14 primed Balblc mice.Control ascites were produced by injectingAg8-653 cells.Ascites fluid was tapped, and proteins .precipitated with ammonium sulphate, dialysed against 3 changes of 5mM sodium phosphate (pH 6.5) and immunoglobulin IgG purified by DEAE chromatography.14

Characterization of antigens:
Midguts from 300 mosquitoes in 3 0 0 ~1 PBS was mixed with 5 0 0 ~1 of 2 x concentrated Laemmli sample buffer17 and placed in boiling water for 3 min and then subjected to electrophoresis on a 10% polyacrylamide gel containing sodium dodecyl sulphate (SDS-PAGE).17Moleculer weight standards (Sigma, USA) were subjected to electrophoresis in parallel lanes.Following electrophoresis, the proteins were transferred to nitrocellulose membranes18 and strips of the blotted proteins probedlg with 1:5 dilutions of hybridoma culture supernatants, purified IgG a t 0.05 mg ml-l, immune mouse serum a t 1:500 dilution (prepared from the midgut immunized mice from which spleens were removed for fusion), rabbit anti-midgut serum5 (1:500 dilution), and unimmunized mouse and rabbit sera (at 1:50 dilution).Peroxidase conjugated rabbit anti-mouse IgG or goat anti-rabbit IgG (Sigma, U.S.A.) a t 1500 dilution was used as the second antibody.Antigenantibody binding was visualized using a chemiluminescent detection system (Amersharn, UK) or chloronaphthol and H,O,.

Assays for effects ofMabs on mosquito fecundity, mosquito mortality and parasite transmissibility:
Mice producing ascites were restrained and used to blood feed groups of 3-4 day old An.tessellatus on two consecutive days.These mosquitoes were held for 72h post bloodmeal and dissected to determine the numbers of mature oocytes (eggs) pr~duced.~Mortality was recorded daily during this period.
Mab I g G (100 p1) was reconstituted in 200 pl normal rabbit serum, mixed a t 50% haematocrit with human erythrocytes parasitized with P. v i v a garnatocytes and fed to 3-4d old An.tessellatus through a membrane feeder as described previ~usly.~The finai concentration of IgG in the bloodmeal was 0.7 mg mI-'.The susceptibility of P. vivax to An. tessellatus was determined by counting the numbers of oocysts on the midgut of the mosquito, 10d post-infection.

RESULTS
After fusion, hybrid cells were observed in 58% of wells.Hybridomas PlG4, 24D4,24C6,24C5, PlC11, PlF10, P2C8 produced antibodies that reacted with midgut antigens in immunobinding assays.Of these, the 3 hybridomas that gave the strongest antibody responses, PlG4,24D4 and 24C6 were cloned by limiting dilution.Over time, the many other hybridomas lost their reactivity with midgut antigen.A typical result obtained in an immunobinding assay with PlG4,24C6 and other Mabs is shown in Figure 1.
The presence of IgG from the Mabs, 24D4C4 and 24C6E9 in the bloodmeal reduced the susceptibility of An. tessellatus to P. vivax (Table 2) ; in these experiments, the presence of IgG antibodies from Ag8-653 ascites also reduced the susceptibility of the vector, when compared to the absence of mouse IgG.Immunoglobulin (IgG) from the Mab PlG4E7 did not have a significant effect on the susceptibility of the vector to P. vivax (data not shown).The mortality of An. tessellatus was not increased by ingesting Mabs -(either from ascites or IgG) in a bloodmeal (data not shown).
the target antigens was made difficult by the conformational nature of the epitopes involved.Identifying biologically useful Mabs involves quite difficult technical procedures and also may require the screening of a large number of Mabs for biological activity.The production of monospecific, polyclonal sera directed against specific purified proteins of the gut may be a n alternative approach towards the same objective.

Figure 2 :
Figure 2 : Western blots of An. tessellatus midgut antigens reacting with A. rabbit anti-midgut serum,B.non-immune rabbit serum, C. mouse anti-midgut serum and D. non-immune mouse serum Migration positions of pre-stained molecular weight markers (Sigma, USA) are indicated in kDa.df=bromophenol blue dye front.

Table 1 : Immunobinding capacity of denatured midgut antigens.
Denaturation was carried out for l5min a t each temperature.Ascites fluid was used as the first antibody.RTroom temperature.Intensity of colour development: 0 (no colour) to +4 (maximum).