Antigen stripping from the nematode epicuticle using the cationic detergent cetyltrimethylammonium bromide (CTAB)

Summary The cuticular antigens of adult Nematospiroides dubius were selectively removed using the cationic detergent cetyltrimethylammonium bromide (CTAB). Nonionic, zwitterionic or anionic detergents were ineffective in comparison. The biochemical profile of the antigens removed by detergent was identical to that of surface antigens removed by homogenization, with the added advantage that detergent‐stripped antigens lacked many of the background antigens (excretory/secretory‐ES and somatic) seen in homogenates. In addition, the detergent was shown to act in a non‐invasive manner as electron micrographs failed to reveal any gross damage to the nematode outer cuticle. The observed selective release of significant quantities of relatively clean nematode surface antigen by CTAB in a non‐invasive or destructive manner provides the impetus for definitive studies on the relevance of surface antigens (in the absence of ES or somatic antigens) to the overall immunogenicity of this and other parasites.


Introduction
The outermost surface of the nematode cuticle consists of a polyanionic glycocalyx overlying a trilamellate layer; together. they form the epicuticle complex (Lee 1977, Lumsden 1975, Inglis 1983. The net surface charge of the epicuticle is negative. and it has been shown to bind polyvalent cationic complexes such as cationized ferritin and ruthenium red (Himmelhoch & Zuckernian 1983. Murrell & Graham 1983.
Correspondence: Dr D.I.Pritchard. M RC Experimental Parasitology Research Group, Department of Zoology, University of Nottingham, Nottingham NG7 2RD. UK.

D. I. Pritchnrd et al.
It is also apparent that the nematode epicuticle expresses molecules which are highly immunogenic to the parasitized host (Philipp, Parkhouse & Ogilvie 1980), and that these surface antigens can be shed into the external milieu (Vetter & Klaver-Wesseling 1978, Smith et nl. 1981. These observations suggest that surface antigens may be relevant to the generation and expression of immunity, and recent data. albeit indirect and utilizing protective monoclonal antibodies specific for surface antigens, supports this possibility (Ortega-Pierres. Mackenzie & Parkhouse 1984). However, there are as yet no examples of successful vaccination against helminth infections using purified preparations of antigens identified by surface-labelling methods (Philipp & Rumjaneck 1984).
Therefore it was decided to attempt to accelerate the relatively slow turnover of surface-expressed molecules in a controlled manner using a technique of selective antigenstripping from the parasite surface by detergents. Detergents have been widely used to solubilize nematode surface antigens (reviewed by Philipp & Rumjaneck 1984), but apart from a few unpublished reports do not seem to have been used in a controlled manner. In the present study, advantage was taken of the polyanionic surface charge of the nematode, and the fact that cationic detergents exist and have been successfully used to remove surface-determinants from viruses (Bachmayer 1975, Laver & Webster 1976. The selective removal of surface antigens was demonstrated, and the potential of the technique to provide sufficient relatively pure immunogenic cuticular material for immunization studies discussed.

NEMATODE RECOVERY AND ANTISERA
Neinatospiroides dubius adults were harvested from CFLP mice as described previously (Pritchard et al. 1983). Immune sera (IMS) were raised in CFLP mice using a multiple infection schedule (Behnke & Parish 1979). Normal mouse sera (NMS) were obtained from uninfected controls.

SURFACE-LABELLING, IMMUNOPRECIPITATION A N D SDS-PAGE
Parasite antigens were analysed biochemically following the methods of Philipp et 01. (1980) andPritchard et al. (1984). Briefly, freshly recovered parasite material was washed extensively before labelling with I'25 using Iodogen (Fraker & Speck 1978). Labelled nematodes were homogenized in the presence of protease inhibitors and centrifuged for 30 min at 11 000 g (4'C) to obtain a source of solubilized surface-labelled material (*SAg).
Otherwise, labelled nematodes were washed extensively in Eagle's MEM containing 50 pg penicillin/ml, streptomycin/ml 50 pg and fungizone (2.5 pg/ml) until the washings were 'cold'. The washed, labelled worms were then cultured in the same medium at 37°C for up to 28 h. At regular time intervals, aliquots of the culture supernatant were aseptically removed for analysis by TCA precipitation (10 pl+50 gl 10% TCA+50 p1 Fetal Calf Serum (FCS) and immunoprecipitation (50 ,LA + 50 p1 IMS or NMS).

DETERGENTS
Detergents were incorporated into the Eagle's M E M in an attempt to selectively strip surface antigens. The selection of detergents was determined by their surfactant properties (Helenius & Simons 1975), and the following were used: Triton X-100. nonionic: sodium deoxycholate (NaDOC). anionic; CTAB, cationic and Empigen BB (zwitterionic). The release in the presence of detergent was compared with that of medium alone.

ELECTRON MICROSCOPY
Adult N . dubius were transferred to 2.5'; glutaraldehyde in 0.1 M phosphate buffer, pH 7.2 for 2 h, and postfixed in I : { osmium tetroxide in 0.1 M phosphate buffer for 1 h (at room temperature). Following dehydration. the worms were transferred to propylene oxide before being infiltrated and embedded in Spurr resin. For light microscopy autoradiography (LMARG) 0.5 pm sections were coated by dipping in a 1 : 1 mixture of Ilford K2 Nuclear Emulsion and O.l:/, glycerol a t 47°C. LMARGs were developed in D19 developer, fixed in 30% sodium thiosulphate, and examined and photographed unstained under Nomarski optics o n a Zeiss Photomicroscope 11. Unlabelled material and LMARGs exposed to light showed no evidence of positive or negative chemography.

Antigen stripping by detergents during culture in vitro
The cationic detergent CTAB selectively stripped appreciably more TCA-precipitable '251-labelled material from the surface of adult N . dubius when compared with Triton X-  dramtically by titation to optimal proportions. P 100, Empigen BB and medium alone ( Figure I). Release was seen to occur very rapidly over the initial 4 h of culture, during which time the nematodes were still motile and appeared morphologically intact. By 20 h of culture, 67';; of the total available '2SI-labelled material had been released compared with 19-22% released in medium by the other detergents used in these experiments ( Table 1).
Precipitaton of the same supernatants with IMS revealed that the pattern of release of immunogenic surface-labelled material was identical to the release of TCA-precipitable material (therefore graph not shown). Although immunogenic material was gradually released in medium alone over the 20 h incubation period, this process was rapidly accelerated by CTAB but not by the other detergents. The quantity of labelled material precipitated by normal mouse serum (NMS) was negligible in comparison (Table 1).
When CTAB (cationic) was compared directly with N a D O C (anionic), the cationic detergent was far superior in releasing appreciable quantities of surface-labelled material into the culture supernatants (Table 2). Even concentrations as low as 0,0259/, CTAB were more effective than 17; NaDOC, which might be surprising considering the more stringent mode of action of the latter detergent (Helenius & Simons 1975).
It was also demonstrated that surface-labelled material was effectively released from larval (d6L4) N . duhius, adult Necutor unzericunus and the infective muscle larvae of T. spiralis (data not shown).

SDS-page analysis of the C T A B-stripped untigeris (Figure 2 )
Immunoprecipitates manufactured between IMS/NMS and the supernatants of CTAB cultures and supernatants of control (medium) cultures were analysed. The resulting profiles on autoradiography were compared with those of CTAB supernatants alone (stripped antigen alone after 1 h incubation-lane 9), radiolabelled whole surface antigen released by homogenization (*SAg-lane 1) and an immunoprecipitate (IP) manufactured between immune mouse serum (IMS) and *SAg (lane 6). It can be seen (Figure 2) that the major surface antigens of adult N . dubius (c.f, *SAg, lanes 1 and 6) were released into the supernatant of cultures within 1 h in the presence of CTAB (Supernatant-lane 9) and that these antigens were immunoprecipitable (IP) by IMS (lane 8). The antigens present in the CTAB supernatant after 1 h were comparable in molecular size to those released from labelled adults by homogenization (*SAg). The major immunogenic moiety which has been detected on the adult surface resolves at 65000 on SDS-PAGE (Pritchard et al. 1984) and it can be seen from Figure 2 that this antigen was released into medium alone during the course of incubation (immunoprecipi- tates M I , M20, M28-lanes 2,3,7) and that an antigen of identical molecular weight was released by CTAB within 1 h (immunoprecipitate CTAB I-lane 4) but that little else was released by hour 20 of the incubation (CTAB 20-lane 5). It would therefore appear that CTAB strips the majority of the immunogenic surface-labelled material of N . diihiiis within 1 h o f incubation. It is also pertinent that the major immunogenic band (65 000) released into culture is also prominent in immunoprecipitates manufactured between IMS and *SAg.
Finally here, the major surface-labelled proteins of Neeutor atneriecmis and Triehinellu spiralis (see inset- Figure 2) were similarly removed by CTAB after I h, indicating that CTAB is indeed a useful agent for the rapid removal of surface-labelled material from the nematode cuticle.
However, it does not suffice that a method should be rapid without being clean. Therefore, it was decided to compare the protein content of CTAB-stripped material with that of *SAg released by homogenization. Labelled preparations of each type were equalized in terms ofct/min precipitable by TCA before analysis by SDS-PAGE, and the resultant gel stained with Coomassie brilliant blue. If CTAB is preferentially stripping surface antigen, then the track containing this preparation should contain significantly less protein than that track containing the homogenization-prepared *SAg. despite the fact that identical ct/min were loaded onto the gel. It can be seen from Figure 3 that this was indeed the case, as material from worms homogenized to release their surface antigens both before (lanes 3,4) and after (lanes 1,2) CTAB treatment shows the characteristic profile on SDS-PAGE of adult worm homogenate (Pritchard et ul. 1983) whereas the CTAB preparation contained none of these proteins despite having an equivalent quantity of radiolabelled surface antigen (in ct/min). The autoradiograph corresponding to these lanes is also shown to demonstrate this point. It is also interesting to note here that the CTAB-stripped preparation was devoid of labelled material in the 'ES-fraction' of adult N . dzrbiza antigens. The ES antigens of N . dubius are known to resolve between molecular weight 14000 and 20000 under the conditions imposed (l2-l5'V0 running gel-Pritchard et al. 1983) and can be seen to present as four distinct bands in the homogenized preparations (lanes 1 4 , cold AH lane 7).

Electron microscopy
Electron microscopical analysis was conducted on parasitic worms in the presence or absence of CTAB. Gross analysis ( Figure 4A) revealed that labelling was largely restricted to the surface of the worm and that significant quantities of label were removed by CTAB (Figure 4 B 4 . f . Figure I). At the EM level, CTAB was shown to cause disruption of the outermost layer (glycocalyx?) of the outer cortex ( Figures 4C, D) without appearing to cause significant disruption of the cortical layers themselves.

Discussion
There is little doubt that parasite surface antigens generate an immune response (Philipp et al. 1980) and that the resulting immune response can be effective against parasites in tritro and in vivo. For example, the infectivity of T. spiralis newborn larvae to mice is significantly reduced when worms are pre-incubated with an IgG, monoclonal antibody directed against a surface antigen of molecular weight 64 000 (Ortega-Pierres et al. 1984). In addition, monoclonal antibodies directed against surface determinants of S. nzansoni schistosomula have been obtained by several workers (Grzych et al. 1982;Tavares et al. 1983) and shown to mediate in vitro killing of larvae and passive protection against in vivo challenge. Despite these elegant studies, there are as yet no examples of successful vaccination against helminth infections using purified preparations of antigens identified by surface-labelling methods (Philip & Rumjaneck 1984). The limiting factor in the quest to assess directly the biological significance of surface antigens has been the lack of sufficient purifiable material for bioassay. In the present study, surface antigens have been selectively harvested from a number of species of parasitic nematode using a cationic detergent cetyltrimethylammonium bromide. Approximately 10 times as much surfacelabelled material was released by incubation in the presence of CTAB compared with medium alone or anionic, zwitterionic or nonionic detergents, suggesting that a charge interaction is occurring between CTAB and the polyanionic 'glycocalyx' of the parasite surface (Inglis 1984).
The surface-labelled material released by CTAB retained its antigenicity, as shown by immunoprecipitation with postinfection sera, and the released material appeared biochemically identical on SDS-PAGE to those antigens released by the conventional homogenization techniques. CTAB-stripped material was further shown to lack the background proteins released by homogenization, in particular the protein bands usually associated with ES-products (Pritchard et al. 1983).
Significantly, worms incubated in CTAB were visually viable during the first 4 h of incubation (during which time the majority of releasable material appeared in culture supernatants) although infections following incubation will verify viability absolutely. For example, preliminary data using T. spiralis has revealed that infective larvae incubated for 1 h in CTAB retain both infectivity and immunogenicity. In the same context, electron microscopical analysis of adult N . dzibius failed to reveal any gross damage to the nematode surface following incubation for 1 h in CTAB despite the removal of labelled material (as assessed by a decrease in the number of silver grains on autoradiography). The most noticeable effect of CTAB was a disruption of the outermost surface ('glycocalyx') of the outer cuticle. It is possible that the formations seen at the outer cuticle following CTAB-treatment represent the aggregated clusters described above.
In conclusion, treatment of some species of live nematodes with a cationic detergent releases significant quantities of relatively clean surface antigens in a non-invasive or grossly destructive manner. This observation provides the impetus for definitive studies on the absolute relevance of surface antigens (c.f. ES, somatic antigens) to the overall immunogenicity o f the parasite-in terms of antibody binding, antibody-mediated cellular cytotoxicity and immunization.