Molecular and Biochemical Parasitology, 9 (1983) 319-327 319

Elsevier

MBP 00369

STAGE SPECIFIC SECRETED AND SOMATIC ANTIGENS OF TRICHINELLA SPIRALIS

R. MICHAEL E. PARKHOUSE and NICHOLAS W.T. CLARK

National Institute.lbr Medical Research, Mill Hill, London NW7 IAA, England

Received 25 March 1983; accepted 9 May 1983)

Infective larvae, adult males and newborn larvae of Trichinella spiralis were cultured with [3SS]methionine

an vitro. Total secreted and total somatic (sodium deoxycholate-soluble) proteins were analyzed by

electrophoresis in the presence of sodium dodecyl sulphate (SDS-PAGE). Secreted proteins were relatively

few in number and were different for each stage, whereas somatic proteins gave an unresolved smear in all

cases. Immune precipitation with serum from infected mice revealed only one major antigen in secretions of

all stages. In a similar investigation of the solubilised somatic antigens, the complexity of antige0s ranged

from none in infective larvae, through few in the adult, to many in the newborn larvae. The total

concanavalin A-binding glycoproteins of each stage exhibited considerable individuality, and hence stage

specificity, when resolved by two dimensional gel analysis. These results extend our knowledge of stage

specific components of T. spiralis, and allow a rational approach towards the construction of diagnostic

procedures.

Key words: Trichinella spiralis; Stage specific; Antigens; Somatic; Secreted

INTRODUCTION

Many parasites, both unicellular and multicellular, pass through a series of discrete developmental life cycle stages. Such stages are normally characterised by stage specific surface antigens (ref. 1 for review), and frequently, but not always, accompany a change in the host environment. This has recently been shown to be so in the parasitic nematodes [2-6]. In these organisms the surface can be labelled with radioac- tive iodine to yield very few (one to five) different electrophoretic components, which are different for each stage of the parasite and which are usually glycosylated. The stage specific surface proteins of nematodes are also antigenic in conventionally infected hosts, and hence their contribution to parasite survival and /o r rejection, although presently not well understood, is likely to be crucial. Interestingly, in the nematode parasite Trichinella spiralis the surface can, in addition, change within a stage [3,7], as well as between stages. Finally, and as further evidence for the dynamic nature of the nematode surface, there is evidence for release, and hence turnover, of the stage specific surface antigens ([3] and our unpublished work) of T. spiralis.

0166-6851/83/$03.00 ~ 1983 Elsevier Science Publishers B.V.

320

In this paper, we have extended the investigation into stage specific antigens of two

other compartments of 1". spiralis: its secretions and total body constituents. Three stages (infective larvae, adults and newborn larvae) were biosynthetically labelled

with [35S]methionine in vitro to yield [35S]methionine-labelled secretions and whole

worms. The latter were detergent solubilised and then both secretions and solubilised

worms were analysed by electrophoresis in sodium dodecyl sulphate-containing

polyacrylamide gels (SDS-PAGE). Antigens present in secretions and solubilised

worms were identified by co-precipitation with sera from infected mice. As a further

analytical tool, the total (concanavalin A-binding) glycoproteins of each stage were analysed by two-dimensional gel analysis according to O'Farrell [8]. Substantial

differences were observed between the three developmental stages in all three ana- lyses.

MATERIALS AND METttODS

Collection of parasites was done as previously described [9]. Infective larvae were

recovered by digestion with pepsin of infected BALB/c mice. Adults were obtained

from Sprague-Dawley rats infected per os 6 days earlier with 6000 infective larvae.

Newborn larvae were collected by differential sedimentation from adult worms which

were incubated for 30 rain at 37°C; adult worms in 10 ml medium were allowed to

settle, leaving the newborn larvae in the supernatant.

Labelling of parasites with [35S]methionine in vitro. Parasites were incubated at 37°C in

2 ml tissue culture medium minus methionine (RPMI 1640, Gibco Ltd) containing penicillin (100 U ml-l), streptomycin (0.1 mg ml 1), 2 mM glutamine and 3% (v/v)

foetal calf serum) containing 200 btCi ml 1 of [35S]methionine at 650 Ci mmo1-1 (The

Radiochemical Centre, Amersham, England). The concentration of worms was

100-150 per ml of infective larvae or adults and 3000-4000 per ml of newborn larvae.

After 2 days at 37°C the medium and parasites were separated by centrifugation (400

X g for 5 rain).

The medium containing secreted counts was filtered through Sephadex G-25 equili-

brated with phosphate-buffered saline (PBS) (130 mM NaCI, 4 mM KCI, 8.1 mM

Na2HPO4, 1.9 mM NaH2PO4, pH 7.4)containing methionine (0.1 mg ml -~) in order to separate the high molecular weight secreted material from flee metbionine. The labelled worms were washed twice in PBS, and solubilised in sodium deoxycholate as previously described [3,4,7]. The mixture was freed of particulate material by centrifu-

gation (20 min, 30 000 X g) and the resultant supernatant, termed 'somatic' was retained. The total radioactivity present in the secreted and somatic compartments was estimated by precipitation of convenient aliquots with 10% (w/v) trichloroacetic

acid and subsequent scintillation counting.

Coprecipitation ofradiolabelled antigens. Soluble radioactive material was incubated

321

with serum (5- l0 gl) from mice infected with T. spiralis. Antigen-antibody complexes

were precipitated with an excess of goat anti-mouse IgG and washed 2-3 times with 0.13 M NaCI-0.01 M Tris-HC1, pH 8.0, containing 0.2% (w/v) Nonidet P. 40. Control

precipitates were similarly prepared using normal (uninfected) mouse serum.

Polyacrylamide gel electrophoresis. Samples were solubilised and electrophoresed exactly as previously described [4,7]. Samples for electrophoresis were first heated for

3 min at 100°C under reducing or non-reducing conditions. For effective reduction the sample was heated in an aqueous solution containing SDS ( 16.7 mg ml-1), glycerol (8.3 mg ml-l), trizma base, Sigma (0.75 mg ml-l), ethylenediaminetetraacetic acid diso-

dium salt (3.7 mg ml-l), bromophenol blue (1.7 lag ml TM) and !3-mercaptoethanol (8.3

mg ml 1). Polyacrylamide gel electrophoresis (PAGE) in the presence of sodium dodecyl

sulphate (SDS) was carried out on 7.5% gel slabs using the Pharmacia equipment (GS C-8) and the buffer gel systems suggested by the manufacturer. Standard radio-label- led proteins of known molecular weight (Pharmacia) were included on gels as mar-

kers. The gel slabs were fixed, dried down and autoradiographed.

Two-dimensional gel electrophoresis and radioactive concanavalin A overlay. This was done as previously described [8], and glycoproteins were subsequently revealed by immersing the gel in 125I-labelled concanavalin A, washing and autoradiography [ 10]. Worms were suspended at a concentration of 20% by volume in O'Farrell 's lysis buffer [8] and solubilisation was assisted by sonication for 3 rain, using a stainless steel probe. The resulting mixture was centrifuged (20 min, 30 000 >(g) and 30 lal of the supernatant was applied to the first (isoelectric focussing) gel. This was then electro-

phoresed in the second dimension on a standard SDS-PAGE slab. The gel was then fixed, washed, treated with 12SI-concanavalin A, washed, dried and autoradiographed

according to established procedures [8].

RESULTS

Secreted and somatic antigens. The amount of radioactive methionine incorporated by the three stages of T. spiralis is given in Table I. Adult males were selected from

mixed populations since females, if included, would liberate newborn larvae during the experiment and these would', of course, incorporate [35S]methionine. A considera- ble fraction of the total incorporated high molecular weight material is found in the secreted compartment - 51.4, 46.3 and 18.3% of total radioactivity incorporated into infective larvae, adult males and newborn larvae, respectively.

The complexity of the total somatic and secreted radioactive proteins was assessed by SDS-PAGE (Figs. 1,2). For all three stages examined the general results were the

same: a poorly resolved ' smear ' of labelled somatic antigens (Fig. la,d,g), and a considerably more restricted set of secreted antigens (Fig. 2a,d,g). Remarkably, the

322

T A B L E 1

Incorpora t ion of [~SS]methionine into ant igens of T. spiralis in vi tro

Parasi te s tage Total Somat ic

lncorpn.

(cpm)

Secreted

(:~ total Spec. ppt. ~: total Spec. ppt.

incorpn. (~.!,~) incorpn. (c)~)

Infective larvae I4.6 X 10 ~ 48.6 3.7 51.4 14.5

Adul t male 14.9 )< I(Y' 53.7 16.2 46.3 3.0

Newborn larvae 51.4 )< 106 81.7 33.7 18.3 9.4

Living worms were incubated in vitro in thc presence of [>S]meth ionine for 2 days at 37°C. Worms were

washed and solubi l ised to give somat ic proteins, and the cul ture superna tan t was freed of low molecu la r

weight rad ioac t iv i ty by passage over G-25-PBS con ta in ing meth ion ine (0.1 mg ml *). The counts given for

radioact iv i ty (total, somat ic or secreted) refl~r to t r ichloracet ic acid insoluble radioact ivi ty . The percentage

of rad ioac t iv i ty specifically precipi ta ted by infected mouse serum (Spec. ppt.) was calcula ted from the

difference in counts between precipi ta tes l~,rmed by normal mouse serum and infected mouse serum. All

de te rmina t ions were done in dupl ica te with a var iance of less than 5f7~, The samples were counted with an

efficiency of 40C~, and the results presented are of one typical exper iment . Detai ls as in Mater ia ls and

Methods.

a b c d e f g h i

M. Vlt. -3

x I0

.94

,67

43

30

Fig. 1, Total [3SS]methionine label led somat ic prote ins and ant igens of T. ,wirafis deve lopmenta l stages.

Analys is by S D S - P A G E under reducing condi t ions and coprec ip i ta t ion with immune serum. Worms were

incubated in [~SS]methioninc in vi t ro and total sod ium deoxychola te -so luble extracts were analysed by

S D S - P A G E : a, infective larvae: d, adul ts : g, newborn larvae. Extracts were incubated with control normal

nlonsc Scrunl or infected nlousc serum and then imfflune complexes were precipi ta ted with goat an t i -mouse

lg. The washed precipi ta tes were submi t t ed to S D S - P A G E analysis using 50(;~ of the final precipi ta te in all

cases. Cont ro l precipi tates: b, infective larvae: e, adults ; h, newborn lar'~ae, h n m u n e serum precipi tates: c,

inliective larvae: f. adulls , i, newborn larvae. Detai ls in Mater ia ls and Methods .

323

M. Wt. -3

x I0

'94

'67

' 4 3

30 a b c d e f g h i

Fig. 2. Secreted [3SSlmethionine-labelled secreted proteins and antigens of 7". spiralis developmental stages. Analysis by SDS-PAGE under reducing conditions and coprecipitation with immune serum. Worms were incubated with [3~S]methionine in vitro and the supernatants were analysed by SDS-PAGE: a, infective larvae; d, adults: g, newborn larvae. Precipitates formed using control normal mouse serum and goat anti-mouse Ig were also analysed. Control precipitates: b, infective larvae; e, adults; h, newborn larvae. Immune serum precipitates: c, infective larvae; f, adults; i, newborn larvae. Details in Materials and Methods.

pat tern of labelled secreted componen ts revealed by S D S - P A G E was different and

characteristic for each stage (Fig. 2a,d,g).

Since whole worms, whatever their developmental stage, are complex, multicellular

organisms, it is no surprise to discover their somatic [35S]methionine labelled proteins

yield such a smear; this simply indicates a complex and unresolved range of consti-

tuents.

In contrast , the gel profiles of the secreted material were considerably simpler. This,

and the evident motili ty of the worms at the end of the incuba t ion period, argue that

the medium conta ined a selective secretion rather than the accumulated soluble

residues of dead worms.

As might be expected, uninfected mouse serum did not precipitate significant

amount s of radioactivi ty from either secreted (Fig. 2b,e,h) or somatic ant igens (Fig.

lb,e,h).

With infected mouse serum, certain components were precipitated, both from the

secretions (Fig. 2c,f,i) and,- with the exception of the muscle larvae (Fig. lc) the

solubilised worm extracts (Fig. lf, i). For the secretions, the amount s specifically

precipitated were generally low (Table I), but the precipitated radioactivity was

largely localised in one main component , different in electrophoretic mobil i ty for

each stage (Fig. 2c,f,i).

A striking observat ion was the failure of infected mouse serum to react with any

componen t of the solubilised radiolabelled infective larvae (Fig. lc). In contrast ,

324

relatively large amounts of radiolabelled material were precipitated from adults and newborn larvae (Table I). Analysis of the precipitates by SDS-PAGE revealed that the host recognises a larger number of newborn larval somatic antigens {Fig. li) than adult antigens (Fig. If). By these criteria, then, stage specific proteins can be demons- trated in somatic and secreted compartments of all three developmental forms of T.

spiralis examined. Some, but not all of these, appear to be immunogenic in naturally infected murine hosts.

Glycoproteins. Total glycoproteins of whole fresh worms resolved by two dimensional gel electrophoresis, and then detected with an overlay of ~25I-concanavalin A and subsequent autoradiography. As can be seen (Fig. 3) each stage yielded a characteris- tic 'glycoprotein map' , consisting of stage specific and common, or shared, compo- nents. An analysis simply based on the concordance of radioactive glycoprotein spots is presented in Table II. Thus infective larvae, adults and newborn larvae were found

to have 32, 11 and 23 unique components out of totals ot69, 46 and 39, respectively. This type of analysis, depending solely on migratory properties on 2-dimensional gels, is likely to underestimate, rather than overestimate, differences. What is clear is that

there are many glycoproteins unique, and thus stage specific, for each of the three stages examined.

pH 4"5 7-5 4.5 7.5 4',5 7,5

_ &

- 9 4

- 6 7

- 4 3

- 3 0

INFECTIVE ADULT NEW- BORN

Fig. 3. Major glycoproteins of 72 spiralis developmental stages. Solubiliscd worms were submitted to

two-dimensional (O'Farrell) electrophoretic analysis and glycoproteins were revealed by ~251- concanavalin

A overlay. The direction of tile pH gradient in the first electrophoretic dimension and the position of

molecular weight markers co-run in the second. SDS-PAGE, dimensions are indicated. Details in Materials

and Melhods.

325

TABLE II

T, spiralis: glycoproteins of life cycle stages

Number of components Infective larvae Adults Newborn larvae

Total 69 46 39

Unique 32 11 23

Shared with: newborn 14 12 - adult 33 - - all 10 -

Comparative analysis of total glycoproteins of T. spiralis. Analysis of two-dimensional e[ectrophoresis and ~zSI-concanavalin A overlay. Whole worms were solubilised in O'Farrell's lysis buffer, electrophoresed and treated with ~25I-concanavalin A as described in Materials and Methods (result presented in Fig. 3). A tracing of each autoradiogram was prepared, and used in order to establish topographical identity of radioactive spots. Figures presented refer to the number of radioactive spots revealed by the radioactive concanavalin A. and then subsequently unique or shared by other stages.

DISCUSSION

The main point that emerges from this work is the existence of a large number of

stage specific proteins in the different life cycle forms of the parasitic nematode T.

spiral is . These have been previously demonst ra ted for surface proteins [3-5,7] and the

study has here been extended into the secretions and internal , or body, const i tuents of

three recoverable stages of the parasite. In fo rmat ion of this nature has part icular

relevance in the choice of parasite components suitable for use in diagnostic proce-

dures.

The profiles of secreted [3SS]methionine-labelled antigens on S D S - P A G E were

relatively simple, and are so completely different as to suggest that all secreted

components are stage specific. Not all, however, were antigens, that is recognised by

antibodies present in serum from convent ional ly infected mice. The secreted compo-

nents, therefore, can in principle be used either for the detection of antibodies or

parasites. In the former case the radiolabelled antigen would serve as the probe.

Detect ion of parasite antigefis is something of a contradic t ion in terms, since antigens

will combine with corresponding antibodies and become difficult, if not impossible, to

detect. More suitable for this purpose are parasite components which fail to elicit

an t ibody format ion in convent ional hosts, but which can do so in appropriately

chosen experimental animals. Such components should also be released by the para-

site in vivo to allow analysis in accessible host body fluids. In the model presented

above, then, one possibility would be to first experimentally prepare ant ibodies against

purified, or partially purified, secretory components that are not precipitated by sera

from normal parasite infections. At this stage, the choice of adjuvant , immunisa t ion

326

course and exper imenta l animal could be crucial. The resul tant an t i se rum would then

be used in a convent iona l inhibi t ion assay with homologous rad io labe l led secret ions

and a potent ia l source of parasi te antigen. A majo r advan tage of b iosynthet ica l ly

label led secreted antigen is that it is not con t amina t ed by debris a n d / o r , p roduc ts of

dying organisms. This is certainly not so when worms are ma in ta ined in vitro, in the

necessary absence of proteins , and the superna tan t s are rad io labe l led with iodine.

The total solubi l ised [3~S]methionine-labelled worms yielded, as might be expected.

a complex range of molecules which failed to resolve t idily upon S D S - P A G E analysis.

In was no tewor thy , however, that the two essential ly in t race l lu lar forms of the

parasi te , the infective larvae and adul ts conta ined none or few componen t s , respecti-

vely, recognised by an t ibodies from infected hosts. The newborn larvae, on the other

hand, which is a systemic parasi te , d id conta in many ant igenic componen t s prec ip i ta-

ted by an t ibodies in sera from infected mice. It is t empt ing to conclude from these

observa t ions that there is a cor re la t ion between accessibi l i ty of the paras i te and

frequency of complemen ta ry ant ibodies . These da ta also suggested the existence of

stage specific somat ic antigens. Interest ingly. we failed to demons t r a t e any [35S]me-

th ionine- label led internal ant igens c o m m o n to all three stages.

Analys is was then restr icted to concanava l in A-b ind ing g lycopro te ins in order to

reduce the complexi ty of somat ic ant igens examined. The result ing two-d imens iona l

(O 'Far re l l ) g lycopro te in maps were very different for each stage (Fig. 3), indicat ing a

large number of stage specific g lycoprote ins . Al though not to ta l ly unexpected, in view

of the very different morpho log ies of the three stages [1 1], nonetheless the result does

hold hope for the employment of such somat ic internal g lycopro te ins in d iagnos t ic

procedures .

In summary , both somat ic and secreted c o m p a r t m e n t s of T. spiralis can be shown to

conta in stage specific proteins. Some. but not all, o f these are immunogen ic in

convent iona l hosts. Both immunogen ic and non immunogen ic stage specific prote ins

can. in principle, be useful in d iagnos t ic procedures .

ACKNOWLEDGEMENTS

Part of this work was suppor t ed b\ ' a grant from the fi lariasis componen t of the

U N D P / W o r l d B a n k / W H O Special P rog ramme for Research and Tra in ing in Tropi -

cal Diseases (Gran t No. 02708434).

REFERENCES

1 Brox~n. K.N. ( 19761 Specificity in host-parasite interaction. In: Receptors and Recognition. Series A ICuatrecasas. P. and Grea',es. M.F.. eds.h pp. 119-175. Chapman and Hall. London.

2 Ogilx ie. B.M.. Philipp. M.. Junger>. M.. Maizels. R.M.. Worms. M.J. and Parkhouse. RM.E. (1980) The ~urfac¢ of nematodes and the immune response of the host. In: The Host-ln',ader lnterpla~ I,.an den Bossche. H.. ed.t. pp. 90-95. Elsexier/North-Holland. Amsterdam.

327

3 Philipp. M.. Parkhouse. R.M.E. and Ogilvie. B.M. (1980) Changing proteins on the surface of

parasitic nematodes. Nature 287. 538-540. 4 Parkhouse. R.M.E.. Philipp. M. and Ogilvie. B.M. [1981) Characterisation of surface antigens of

Trichinella spiralis infective larvae. Parasite Inlrnunol. 3. 339-352. 5 Clark. N.W.T.. Philipp. M. and Parkhouse. R.M.E. (1982) Non-covalent interactions result in

aggregation of surface antigens of the parasitic nematode Trichinella spiralis. Biochem. J. 206, 27-32.

6 Maizels. R.M.. Meghji. M. and Ogih'ie. B.M. (1983) Restricted sets of parasite antigens from the

surface of different stages and sexes of the nematode parasite Nippostrongylus brasiliensis. Immunolo-

g.~ 48. 107-121. 7 Jungery. M.. Clark. N.W.T. and Parkhouse. R.M.E. (1983) A major change in surface antigens during

the maturation of new born larvae of Trichinella spiralis. Mol. Biochem. Parasitol. 7, 101-109.

8 O'Farrell. P.H. (1975) High resolution tw'o-dirnensional electrophoresis of proteins. J. Biol. Chem.

250. 4007-4021. 9 Dennis. D.T.. Despommier. D. and Davis. N. (1970). Infectivity of the new born larvae of Trichinella

spiralis in the rat. J. Parasitol. 56, 974-977. 10 Burridge. K. (1978) Direct identification of specific glycoproteins and antigens in sodium dodecyl

sulphate gels. Methods Enzymol. 50. 54-56. 11 Gould. S.E. (1970) Trichinosis in Man and Animals, Charles C. Thomas, Springfield, II,,

U.S.A.