Antigen processing in earthworms

Immunology Letters, 41 (1994) 273-277

Elsevier ScienceB.V. IMLET 2196

Antigen processing in earthworms Ludmila Tu~kov~ * and Martin Bilej Department of lmmunology, Institute of Microbiology, Academy of Sciences of the Czech Republic, Videhskdt 1083, 142 20 Prague 4, Czech Republic

(Received10 August 1993;revised27 October1993; accepted27 May 1994) Key words: Lumbricus terrestris; Eiseria foetida; Antigen-bindingprotein;Proteolyticactivity

1. Summary The administration of protein antigens into earthworms Lumbricus terrestris and Eisenia foetida induces the formation of antigen-binding protein (ABP) with the maximum response occurring between days 4 and 8. High proteolytic activities observed both in coelomocytes and in coelomic fluids cause rapid antigen degradation; the majority of antigen is digested during the first 24 h. To analyze the role of proteolytic processing of antigen in ABP response in vitro, the intact antigen (ARS-HSA) as well as its proteolytic fragments were added to tissue explant cultures and ABP formation followed by indirect ELISA using mAb to ABP. The response to small fragments was comparable to that induced by intact ARS-HSA. Furthermore, the response to intact antigen was almost completely blocked by the non-toxic serine protease inhibitor Pefabloc, while the response to small fragments was only slightly reduced. The kinetics of response to intact antigen significantly differed from that induced by small ( < 3 kDa) fragments. We suggest that proteolytic processing is involved in the stimulation of ABP formation.

2. Introduction Analyses of defence mechanisms of earthworm Lumbricus terrestris and Eisenia foetida have revealed that administration of protein antigens into the coelomic cavity leads to a marked increase in coelomic fluid protein concentration and to the formation of 56 kDa antigen-binding protein (ABP) [1,2]. Its binding specificity is rather low since it is able to bind different * Corresponding author: Ludmila Tu~kov~, Department of Im-

munology, Institute of Microbiology,Academy of Sciences of the Czech Republic, Videfiskfi1083, 142 20 Prague 4, Czech Republic. SSDI 0165 -2478(94)00112 -5

antigens, although the most efficient binding was always obtained with the antigen used for stimulation. The ABP was found not only in the coelomic fluid but also on the surface of certain coelomocytes and its highest concentration and the highest number of ABPbearing cells were detected between the 4th and 8th days after primary stimulation. After a secondary challenge the response was faster and more pronounced [2-4]. Moreover, in both coelomic fluids and in coelomocytes a powerful proteolytic system was present that neither affected self-proteins nor those of related species but digested foreign (vertebrate) antigens so efficiently that, as early as 4 hr after administration, no intact (undigested) molecules were detectable [5-7]. These results led us to determine whether the rapid digestion of administered protein antigen is related to the ABP response, i.e., whether the response to antigenic stimulation in earthworms also occurs after processing the stimulating antigens. The aim of these experiments was to test the effect of proteolytic fragments of protein antigen on stimulation and subsequent ABP formation. 3. Materials and Methods 3.1. Earthworms

Adult earthworms Lumbricus terrestris kept at 15°C in soil and Eisenia foetida kept at 20°C in mold were used in all experiments. 3.2. Harvesting o f coelomic fluid

Coelomic fluid containing free coelomocytes was obtained by puncturing the coelomic cavity with a glass micropipette. The suspension, pooled from 5-10 earthworms, was centrifuged (100 × g for 10 min), the coelomic fluid was collected, recentrifuged (1000 × g for 10 min), and the supematant stored at -70°C. 273

3.3. Estimation of the coelomic fluid proteolytic activity Coelomic fluid (15 ~1) was incubated for 24 h with 50/zl of the mixture of cold ARS-HSA (5/xg/sample) and 12SI-labelled ARS-HSA (100000 cpm) at room temperature. After incubation, 100 ~1 of 20% human serum albumin (HSA) was added to increase the total protein concentration. Proteins were precipitated with 200 /zl of 10% trichloracetic acid (TCA) and the samples centrifuged. The supernatants were collected, the sediments solubilized with 1 M NaOH, and radioactivity in both fractions measured using a gamma-counter. To test the effect of Pefabloc (4-(2-aminoethyl)-benzenesulphonyl fluoride; Boehringer), the inhibitor of serine proteases, 1, 5, 10, or 50 /zl of the 0.1 M stock solution of Pefabloc was added to each sample.

3.4. Preparation of tissue explants The modified method of Janda and Bohuslav [8] was used. To minimalize possible contamination, L. terrestris were maintained in cotton wool soaked with 2% antibiotic solution (antibiotic antimycotic solution, Sigma, St. Louis, MO: containing 10 000 IU penicillin, 10 mg streptomycin, and 25 ~g amphotericin B per ml in 0.9% sodium chloride) and tetracycline (60 /zg/ml) for 2 days before beginning the experiment. Earthworms were anesthesized in 4°C cold 5% ethanol and washed in sterile PBS. The body wall was opened with sterile scissors on the ventral side and the gut wall covered with chloragogen tissue was prepared. The gut wall was cut into small fragments (2 × 2 mm) that were washed 10 times in sterile isotonic (220 mOsmol) RPMI-1640 medium supplemented with glutamine (1 mM), HEPES (25 mM), and antibiotics (1% of antibiotic antimycotic solution, and tetracycline, 60 /xg/ml). The fragments of gut wall were cultivated in 24-well plates (in 1 ml of medium) at room temperature.

3.5. Stimulation of tissue cultures Tissue explants were stimulated with 10/xl of A R S HSA or its proteolytic fragments (see below). To exclude proteolytic activity 50 ~1 of Pefabloc stock solution was added to parallel samples.

3.6. Estimation of ABP formation The formation of ABP was detected by ELISA performed as described earlier [9]. Briefly, the 96-well ELISA plates (Koh-i-noor, Hardtmuth, Czech Republic) were coated overnight at 4°C with 100/.tl of ARS-HSA (1 m g / m l ) in PBS. After washing with PBS containing 0.05% Tween-20 (T-PBS) and with PBS alone the wells 274

were saturated with 1% bovine serum albumin (BSA). After another washing the wells were filled in triplicate with 100 /xl of tissue culture supernatants, 100 /zl of coelomic fluid (positive control), or 1% BSA (negative control) and incubated for 2 h at room temperature. The wells were then washed and 50 /xl of a monoclonal antibody 6 D 4 / 2 (to ABP: Ig fraction) diluted 1:30 in PBS added to wells for 90 min at room temperature. After incubation the plates were washed and 50 /zl of pig anti-mouse Ig conjugated to peroxidase (SwAM-Px, Institute of Sera and Vaccines, Prague, Czech Republic) diluted 1:500 in T-PBS applied and the microplates incubated for 2 h at room temperature. Wells were then washed and the enzyme reaction developed by substrate o-phenyldiamine (Lachema, Brno, Czech Republic) diluted in 0.01 M phosphate buffer, pH 6.0, containing H202, stopped by H2SO 4 and read on Microelisa Minireader MR (Dynatech) at 470 nm.

3.7. In vitro digestion of the antigen and gel chromatography After 24-h digestion with coelomic fluid (of L. terrestris or E. foetida) mixed in a ratio of 1:1 (1 ml and 1 ml), ARS-HSA (10 mg) was separated on Centricon 10 or Centricon 3 (Amicon). Low molecular weight fragments ( < 10 kDa) were applied to a Sephadex G-25 column in 0.5% NHaHCO 3. Five main fractions were separated and concentrated by lyophilization.

4. Results and Discussion

4.1. Tissue explant cultures To determine if the proteolytic fragments have similar stimulating effects as whole intact antigen (ARSHSA), we used the culture technique of L. terrestris tissue explants for assaying the ABP response in vitro. Small tissue explants of L. terrestris gut wall were cultured either alone or in the presence of whole intact ARS-HSA or its proteolytic fragments. Cultures performed in the presence of coelomic fluid, employed as a source of proteolytic enzymes (used for antigen digestion), were used as controls. After cultivation, the level of ABP was tested in the culture media by the ELISA technique using monoclonal antibodies specific to ABP [9].

4.2. Stimulation of tissue explant cultures As shown in Fig. 1, the highest stimulating effect, on ABP formation was exerted by the whole intact ARS-

~

0.22 *- 0.05

coelomic fluid ~

0.24 + 0.04

control

ARS-HSA ~

% of non-precipitable fragments 801

~

0.56#. 0.06" SOt

,20 kDa ~

0.23 -* 0.03

10-20 kDa ~ 3-10 kDa

0.51 *- 0 . 0 5 *

~

i

0 . 4 9 * _ 0.07"

~

,3ko.

40'

o7o.oo6. _ 0

_ 0.2

_ r 0.4

J 0.6

018

O.D.

Fig. 1. ABP formation in L. terrestris tissue explant cultures 7 days after stimulation with ARS-HSA or its small proteolytic fragments followed by ELISA. Non-stimulated cultures and cultures stimulated with L. terrestris coelomic fluid used for digestion were employed as controls. Indicated means+SD; * significant at P < 0.05 level. 0

10

20

30

40

50

IJI of Pefabloc

H S A and its smallest proteolytic fragments ( < 20 kDa). Surprisingly, the effect of the large proteolytic fragments ( > 20 kDa) was significantly lower and reached the value corresponding to that obtained in experiments where cultivation was performed without any antigenic stimulation. Since proteolytic fragments were prepared by splitting the antigen ( A R S - H S A ) with coelomic fluid, the experiments were also carried out in the presence of coelomic fluid alone. Again, no significant effect on A B P formation was observed.

•

L. t e r r e a t r i s

I

E. f o e t i d a

Fig. 2. The inhibition of proteolytic activity of L. terrestris and E. foetida coelomic fluids by Pefabloc.

of A B P formation in earthworm tissue cultures, we have compared responses in the presence or absence of an inhibitor of serine proteases, Pefabloc, which, in contrast to PMSF, exerts no damaging effect on cultured cells. The effect of Pefabloc on proteolytic activity of coelomic fluid was comparable to that of P M S F and reached almost 100% (Fig. 2). A s shown in Fig. 3, the inhibitor almost completely blocked responses to intact antigen but only slightly reduced A B P formation

4.3. Effect o f the serine protease inhibitor To confirm the results showing that proteolytic processing of the protein antigen is involved in induction

% of control

350 300 250 200 150

m

controls

ARS-HSA

710 kDa

without inhibitor

~

3-10 kDa

<3 kDa

with inhibitor

Fig. 3. ABP response of tissue explant cultures (L. terrestris, day 7) after stimulation with ARS-HSA and its proteolytic fragments in the absence and presence of Pefabloc inhibitor followed by ELISA. Indicated as values relative to those for non-stimulated controls (100%, day 0). 275

% of control 200

2O0

% of control

ARS-HSA 180

180

160

160

140

140

120

120

100' 1

2

3

4

6

6

7

8

Small proteolytic fragments < 3 kDa

100 0

1

2

3

days without Pefabloc

- - I ~ with Pefabloc

4

5

6

7

days ~

without Pefabloc

--4-- with Pefabloc

Fig. 4. The kinetics of ABP formation in the presence and absence of Pefabloc inhibitor measured by ELISA. Indicated as values relative to those for non-stimulated controls (100%, day 0).

induced by small proteolytic fragments. These results confirmed that proteolytic processing of antigen represents part of the ABP response. 4.4. Kinetics of the ABP response Small fragments ( < 10 kDa) were separated by gel filtration using a Sephadex G-25 column into 5 fractions and the stimulating effects of the smallest ( < 3 kDa) tested in the presence or absence of Pefabloc inhibitor. The kinetics of ABP formation after stimulation of tissue cultures with whole intact antigen (ARSHSA) and its small proteolytic fragments during the first 7 days after stimulation (Fig. 4) showed that the maximal response to intact antigen was observed on day 5 and decreased on day 7 while the response to the smallest proteolytic fragments was continuously increasing reaching the highest value on day 7. Again, ABP formation was significantly decreased by Pefabloc only when intact antigen was used for stimulation. There were no significant differences in results of experiments in which coelomic fluid of L. terrestris or E. foetida were used for digestion of antigen.

detectable in coelomic fluid, we strongly suggest that proteolytic processing of antigen represents part of the mechanisms of ABP formation which is first detectable around day 4 after stimulation [1,2]. The question is why non-stimulating large fragments were not further split during cultivation (pronounced proteolytic activity is present in the culture medium) [5] into smaller fragments and did not stimulate ABP formation. In fact, the amount of small TCA-non-precipitable fragments increased during cultivation [5]. The only plausible explanation we can consider is that binding of antigen on ABP molecules present on coelomocytes may not always be activated by the proper signal that leads to cell stimulation and ABP formation. What may occur is only accommodation of peptides of a certain size and structural motifs that can mediate activation signals that lead to ABP formation.

Acknowledgements We wish to thank Dr. Jaroslav Rejnek for stimulating discussion and critical suggestions during all of our studies.

5. Conclusion

References

At present we cannot explain why only intact antigen and its small proteolytic fragments stimulated the ABP response while large fragments did not. Since shortly after antigen administration (during the first 4 h) all antigen was digested and no intact molecules were

[1] Tu~kovfi, L., Rejnek, J. and Sima, P. (1988) Dev. Comp. Immunol. 12, 287. [2] Tu~kovfi, L., Rejnek, J., Bilej, M. Posplgil, R. (1991) Dev. Comp. Immunol. 15, 263. [3] Bilej, M., Tu~kovfi, L, Rejnek, J. and V~tvi~ka, V. (1990) Immunol. Lett. 26, 183.

276

[4] Bilej, M., Rossmann, P., VandenDriessche, T., Scheerlinck, J.-P., De Baetsellir, P., Tufikov~, L., V~tvi~ka, V. and Rejnek, J. (1991) Immunol. Lett. 29, 241. [5] Bilej, M., Tu6kov~, L. and Rejnek, J. (1993) Immunol. Lett. 35, 1. [6] Rejnek, J., Tu6kov~i,L., Lima, P. and Bilej, M. (1993) Immunol. Lett. 36, 131.

[7] Leipner, C., Tu~kov~i, L., Rejnek, J. and Langner, J. (1993) Comp. Biochem. Physiol. 105B, 637. [8] Janda, V. and Bohuslav, P. (1934) Publ. Fac. Sci. Univ. Charles 133, 1. [9] Tu6kov~i,L., Rejnek, J., Bilej, M., H~jkovL H., and Romanovsk~, A. (1991) Comp. Biochem. Physiol. 100B, 19.

277