J. Biochem. 101, 545-551 (1987)

Purification of a Lectin from the Hemolymph of Chinese Oak ( pernyi) Pupae1

Xian-Ming QU,2 Chun-Fa ZHANG,3 Hiroto KOMANO, and Shunji NATORI

Faculty of Pharmaceutical Sciences, The University of Tokyo, Bunkyo-ku, Tokyo 113

Received for publication, November 7, 1986

A lectin with affinity to galactose was purified to homogeneity from the hemolymph of diapausing pupae of the Chinese oak silk moth, Antheraea pernyi. The molecular mass of this lectin was 380,000 and it formed an oligomeric structure of a subunit with a molecular mass of 38,000. The hemagglutinating activity in the hemolymph was found to increase with time after immunization with E. coli. Studies with antibody against the purified lectin showed that increase in the hemagglutinating activity was due to the same lectin, suggesting that the amount of the lectin increased in response to intrusion of foreign substances. The function of this lectin in the defence mechanism is discussed.

The hemolymphs of many invertebrates are known have no definite humoral immune system like that to contain agglutinin (1-7). Most of these agglu in vertebrates, one of the biological roles of inver tinins are proteins binding carbohydrates, so they tebrate lectins has been suggested to be in the can be defined as invertebrate lectins. These lec defence mechanism (8-11). However, no conclu tins differ greatly in molecular masses, subunit sive evidence to support this idea has yet been structures, hapten sugars, and ionic requirements obtained. (7). Probably, these lectins participate in many Previously, we purified a lectin from the aspects of fundamental biological events, such as hemolymph of Sarcophaga peregrina (flesh-fly) development, differentiation, recognition of self larvae (12). This Sarcophaga lectin is peculiar in and non-self, and so forth. Since invertebrates the following two points: 1) Normal larvae do not contain this lectin, but it is promptly induced in 1 This work was supported in part by a Grant-in-Aid the hemolymph on injury of the body wall (12, 13). for Scientific Research from the Ministry of Education, Namely, this lectin is induced under conditions Science and Culture of Japan. when the defence system of this is supposed 2 Present address: Shanghai Institute of Biochemistry , to be enhanced. Consistent with this possibility, Academica Sinica, 320 Yue-Yang Road, Shanghai, we showed that this lectin participates in the elim China. ination of sheep red blood cells introduced into the 3 On leave of absence from the Silkworm Research In abdominal cavity of larvae (14). 2) This lectin is stitute, Fengcheng, Liaoning Province, China. Abbreviations: SDS, sodium dodecyl sulfate; kDa, synthesized not only on injury of the body wall of larvae, but also in the early embryonic stage and

Vol. 101, No. 3, 1987 545 546 X.-M. QU, C.-F. ZHANG, H . KOMANO, and S. NATORI

in the early pupal stage in normal development of buffer, pH 6.4, containing 0.13 M NaCl, 5 mM this insect (15). Thus, it may have functions in KCl, and 1 mM CaCl2 by extensive washing. The both the defence mechanism and development. If final preparation was stored at 4•Ž.

such lectins are essential for development, they are Molecular Weight Determination by Gel Filtra probably present in all holometabolous . tion•\A column of Toyopearl HW-55 (Toyo Soda, We found a similar lectin in the hemolymph Japan, 1 •~ 80 cm) was equilibrated with 10 mM

of pupae of the Chinese oak silk moth, Antheraea phosphate buffer, pH 7.4, containing 130 mM NaCl

pernyi. This paper describes the purification and , 3 mM KCl, and 0.2 M glucose. The pro some characteristics of this Antheraea lectin. tein sample (180 ƒÊg/0.5 ml) was applied to the

column, and chromatography was carried out at

MATERIALS AND METHODS room temperature at a flow rate of 2 ml/h. Frac tions of 1 ml were collected and assayed for hemag and Collection of Hemolymph•\Di glutinating activity. The column was calibrated apausing pupae of the Chinese oak silk moth, A. with bovine serum albumin (68,000), aldolase pernyi, were supplied from the Liaoning Province (158,000), ferritin (450,000), and thyroglobulin Silkworm Research Institute, China. Pupae were (670,000). stored at 4•Ž. Under these conditions, pupae Polyacrylamide Gel Electrophoresis•\Electro could be kept for at least 6 months without devel phoresis on polyacrylamide SDS-slab gel was car opment. For collection of hemolymph, the poste ried out by the method of Laemmli (16). Proteins

rior tip of the pupae was cut off with fine scissors were denatured by heating them in 1% SDS solu and the hemolymph that exuded was collected in tion containing 2% 2-mercaptoethanol for 20 min

an ice-cooled test tube containing a few crystals at 75•Ž. The stacking gel contained 3% acryl

of phenylthiourea to inhibit the activity of phenol amide and the separating gel contained 12.5% oxidase. Hemocytes were removed by centrifuga acrylamide. After electrophoresis, the gel was

tion for 30 min at 10,000 •~ g, and the resulting stained by the method of Fairbanks et al. (17).

clear supernatant was stored at -20•Ž. Usually, For estimation of the molecular mass, the gel was

200 ml of hemolymph was collected from 100 calibrated with bovine serum albumin (69,000), pupae. ovalbumin (45,000), and chymotrypsinogen

Assay of Hemagglutinating Activity•\Hemag (25,000). glutinating activity was assayed essentially as de Antibody against Antheraea Lectin•\Antibody

scribed before (12), with commercially available against Antheraea lectin was raised by injecting 500 rabbit red blood cells as an indicator. Unagglu ƒÊg of purified lectin mixed with complete Freund's tinated cells formed a clear dot, whereas agglu adjuvant into a male albino rabbit. Injections tinated cells formed a diffuse mat on the bottom were made into the footpads and also to five to of wells of microtiter V-plates. The end point of six sites in the back. A booster injection of the transition from a diffuse mat to a dot was distinct same amount of the lectin mixed with incompleted on dilution of the hemolymph. Hemagglutinating Freund's adjuvant was given two months later, activity was defined as the reciprocal of the maxi and the was bled one week after the booster mum dilution of the test sample causing hemag injection. The reactivity of the antiserum was glutination (titer). determined by the double gel diffusion method of

Acid-Treated Sepharose 6B•\Acid-treated Se Ouchterlony (18). Immunoglobulin G (IgG) was pharose 6B was prepared as follows: About 150 purified from the serum by the method of McCauley ml of Sepharose 6B (Pharmacia) was suspended in and Racker (19). Protein was determined by the

500 ml of 0.2 N HCl and washed well with distilled method of Lowry et al. (20). water. The resulting paste was suspended in 240 ml of 0.2 N HCI and kept at 53•Ž for 2 h with RESULTS gentle shaking. The acid-treated Sepharose 6B thus obtained was washed extensively with distilled Hemagglutinating Activity of Hemolymph of water, transferred to a column (4 •~ 27 cm), and Antheraea Pupae•\Significant hemagglutinating ac equilibrated with 10 mM CH3COOH/CH3CONH2 tivity was detected in the hemolymph of pupae of

J. Biochem . PURIFICATION OF Antheraea LECTIN 547

TABLE I. Effects of sugars on hemagglutinating activity.

Fig. 1. Enhancement of the hemagglutinating activity in the hemolymph of Antheraea pupae on injection of E. coll. A suspension of 105 E. coli in a volume of 50 ƒÊl was injected into each pupa cells at time 0. Pupae were kept at 4•Ž. As controls, normal larvae were kept under the same conditions. Hemolymph samples a 20-Fold diluted hemolymph was used . were collected at the indicated times and their hemag

glutinating activity was measured. Each point repre sents the average value for four pupae. •œ immunized sugar specificity or to multiple lectins each with a

pupae; •›, normal pupae. different sugar specificity. To clarify this point, we tried to purify this lectin(s). the Chinese oak silk moth, A. 1pernvi, when assayed Purification of Antheraea Lectin-Since the with rabbit red blood cells as an indicator. This hemagglutinating activity in the hemolymph was activity increased markedly with time after injec inhibited by n-galactose, the lectin could be puri tion of a suspension of E. coli into the pupae (105 fied by affinity chromatography on galactose. cells/animal), as shown in Fig. 1. Similar increase About 235 ml of hemolymph collected from 100 in the production of antibacterial proteins termed pupae was heated at 70•Ž for 15 min to reduce cecropins was observed in the pupae after the same its viscosity, and then cooled and centrifuged for treatment (21). Thus, this hemagglutinin may well 20 min at 10,000 •~ g to remove denatured proteins. be a sort of defence protein like Sarcophaga lectin The resulting clear supernatant was applied to a found in the hemolymph of S. peregrina larvae column of acid-treated Sepharose 6B (4x 27 cm) when their body wall is injured with a hypodermic equilibrated with 10 mM acetate buffer. Under needle, and it may function in the elimination of these conditions most proteins did not bind to the foreign substances introduced into the abdominal column and were recovered in the flow-through

cavity. fraction. But this fraction did not contain ap

This hemagglutinating activity was found to preciable hemagglutinating activity. The bound be inhibited by a-methyl-D-glucoside and D-galac material was eluted with 0.2 M galactose. Each tose at concentrations of 5 and 25 mM, respec fraction of eluate was dialyzed extensively against tively, as shown in Table I. At much higher 10 mM phosphate buffer, pH 7.4, containing 130 concentrations, lactose and N-acetyl-D-galacto mM NaCl and 3 mM KCI, and then its hemag

samine also inhibited this activity, but the other glutinating activity was assayed. As shown in sugars tested had no effect. The hemagglutinating Fig. 2a, a single peak of protein with hemagglu activity in the hemolymph of immunized pupae tinating activity was eluted under these conditions. showed the same sugar specificity. From these The recovery of the activity from the column was results, we concluded that this hemagglutinin is a usually 30%. The lectin was almost pure at this lectin. However, these results do not show whether stage, because on SDS-polyacrylamide gel electro the activity was due to a single lectin with multiple phoresis, each fraction gave a main band of protein

Vol. 101, No. 3, 1987 548 X.-M. QU, C.-F. ZHANG , H. KOMANO, and S. NATORI

Fig. 2. (a) Affinity chromatography of hemolymph lectin on Sepharose 6B. Hemolymph collected from normal pupae was heated at 70•Ž for 15 min and then centrifuged at 10,000 ; g for 20 min. The clear super Fig. 3. (a) Gel filtration of Antheraea lectin on Toyo natant (190 ml) was diluted 2-fold with 10 mM acetate pearl HW-55. Fractions from acid-treated Sepharose buffer, pH 6.4, containing 0.13 M NaCl, 5 mM KCl, 6B were pooled and lyophilized. Dried material was and 1 mM CaClz and applied to a column of acid dissolved in 10 mM phosphate buffer, pH 7.4, con treated Sepharose 6B (4 •~ 27 cm) at room temperature. taining 130 mM NaCl, 3 mM KCl, and 0.2 M glucose The column was washed with 5 volumes of the buffer, and applied to a column of Toyopearl HW-55 (1 •~55 and then adsorbed material was eluted with 0.2 M cm) equilibrated with the same buffer. Fractions of galactose in the same buffer, changing the buffer at the 2 ml were collected and their hemagglutinating activity fraction 1. Fractions of 4 ml were collected and was assayed. The arrow indicates the position of dialyzed extensively before measurement of their hem aldolase (Mr 158,000). •›, optical density at 280 run; •œ agglutinating activity. •›, optical density at 280 nm; •œ , hemagglutinating activity. (b) SDS-polyacrylamide , hemagglutinating activity. (b) SDS-polyacrylamide gel electrophoresis of fractions eluted from Toyopearl gel electrophoresis of fractions eluted from acid-treated HW-55. Numbers correspond to those in (a). Sepharose 6B. Numbers correspond to fraction num bers in (a).

protein peak. This peak was eluted from the with a molecular mass of 38,000 (38-kDa protein) column much earlier than aldolase, which has a coinciding with the hemagglutinating activity, as molecular mass of 158,000, clearly indicating that shown in Fig. 2b. The two distinct upper bands it was larger than 38 kDa. However, as shown in may be artefacts of electrophoresis, because they Fig. 3b, on SDS-polyacrylamide gel electropho were obtained with all the fractions tested, and resis each fraction gave a single band of 38-kDa were not detected when pooled 38-kDa protein protein coinciding with the hem agglutinating ac fractions were subjected to electrophoresis under tivity. Therefore, the 38-kDa protein was a sub

the same conditions. unit of the intact lectin.

For determination of whether the 38-kDa A summary of the purification is given in protein was the active lectin itself or a subunit of Table ‡U. We purified the Antheraea lectin from the lectin, the fractions from Sepharose 6B were the pupal hemolymph to near homogeneity essen combined and subjected to gel filtration through tially by two steps: heat treatment and affinity Toyopearl HW-55. As is evident from Fig. 3a, chromatography on acid-treated Sepharose 6B . the hemagglutinating activity was recovered from The overall recovery of activity in this procedure the column as a single peak coinciding with a was about 30%.

J. Biochem . PURIFICATION OF Antheraea LECTIN 549

TABLE ‡U. Summary of purification of Antheraea lectin .

a Protein was determined by the method of Lowry et al . (20).

TABLE ‡V. Blood type specificity of agglutinating activity.

Fig. 4. Determination of the molecular mass of intact lectin by gel filtration on Toyopearl HW-55. The col umn was calibrated with (a) thyroglobulin, (b) ferritin, (c) aldolase, and (d) bovine serum albumin. The open circle shows the position of intact lectin.

Characterization of Purified Lectin•\The mo

lecular mass of Antheraea lectin was determined

by gel filtration through Toyopearl HW-55 with

various marker proteins. As shown in Fig. 4,

the molecular mass of the intact lectin was 380,000,

suggesting that it consists of about 10 subunits of Fig. 5. Inhibition of hemagglutinating activity in the 38-kDa protein. The sugar specificity of the puri hemolymph by antibody against purified lectin. In fied lectin was exactly the same as that of crude creasing amounts of antibody or normal IgG were added

material in the hemolymph shown in Table I, to a fixed amount of hemolymph prepared from normal

indicating that the hemolymph of Antheraea pupae pupae or immunized pupae, and remaining hemaggluti nating activity was measured. •£, normal hemolymph+ contains only one lectin with specificity to a normal IgG; •¢, immunized hemolymph+normal IgG; methyl-D-glucoside, galactose, lactose, and N •œ, normal hemolymph+antibody; •›, immunized acetyl-D-galactosamine. This lectin agglutinated hemolymph + antibody. rabbit red blood cells, but not sheep or human

red blood cells irrespective of their blood type, as

shown in Table ‡V. Similar results were obtained present specifically on the surface of rabbit red for the hemagglutinating activity in the hemolymph. blood cells.

Probably, this lectin recognizes a complex struc To determine whether the hemagglutinating ture containing the above sugar moieties that is activity induced in the hemolymph of pupae im

Vol. 101, No. 3, 1987 550 X.-M. QU, C.-F. ZHANG, H. KOMANO, and S. NATORI munized with E. coli is due to the same lectin as that the hemolymph of Antheraea larvae do not that present in normal pupae, we raised antibody contain this lectin, and that it is induced as a against purified lectin. As is evident from Fig. defence response on injection of foreign substances 5, this antibody completely inhibited the hemag into the larvae. However, we could not test this glutinating activity present in the hemolymph of possibility directly, because Antheraea larvae were immunized pupae as well as that of normal pupae. not available. Since the hemagglutinating activity in the hemol Sarcophaga lectin was found to be synthesized ymph increased after injection of E. coli, this result in the pupal stage, and thus to be present in pupal suggests that the intrusion of foreign substances hemolymph. However, it is not certain whether such as E. coli cells increased the production of additional induction of the same lectin occurs the lectin present in the hemolymph of normal after injection of foreign substance into Sarco pupae to above its basal level. phaga pupae, as observed in Antheraea pupae. From the present results and those obtained

DISCUSSION on Sarcophaga lectin, we speculate that this type of humoral lectin is commonly present in holo This paper describes the purification of Antheraea metabolous insects, and that its synthesis is regu lectin to near homogeneity from the hemolymph lated by both the developmental stage and the of pupae. On comparison of this lectin with that defence mechanism. The main function of such of Sarcophaga, which we purified previously, several lectins induced in response to the intrusion of points are noteworthy. The two lectins both have foreign substances is to eliminate them. In fact, an oligomeric structure composed of essentially the we demonstrated that Sarcophaga lectin is essential same subunits, although the molecular mass of for the elimination of sheep red blood cells intro Sarcophaga lectin is much less, being about 190,000. duced into the abdominal cavity of the larvae (14). The sugar specificities of the two lectins are also Then what is the function of these lectins in the similar, since both have strong affinity to n-galac pupae? Probably, these lectins play a role in tose. However, unlike Sarcophaga lectin, Anther scavenging unnecessary cells or tissue fragments aea lectin was found to have affinity to a-methyl produced during metamorphosis, but further stud D-glucoside as well as galactose moieties and of ies are needed to test this possibility. the cells tested it agglutinated only rabbit red blood cells. REFERENCES The hemagglutinating activity in the hemol ymph increased with time after injection of E. coli 1. Marchalonis, J.J. & Edelman, G.M. (1968) J. Mol. into the abdominal cavity of pupae. Experiments Biol. 32, 453-465 2. Scott, M.T. (1971) Arch. Zoos. Exp. Gen. 112, with antibody against Antheraea lectin showed 73-80 that the hemagglutinin that increased after the 3. Suzuki, T. & Natori, S. (1983) J. Biochem. 93, intrusion of E. coli was immunologically the same 583-590 as that originally present in the hemolymph. Prob 4. Umetsu, K., Kosaka, S., & Suzuki, T. (1984) J. ably, de novo synthesis of this lectin is promoted Biochem. 95, 239-245 by injection of E. coli. This situation is similar 5. Ravindranath, M.H., Higa, H.H., Cooper, E.L., to that with Sarcophaga lectin: the hemolymph of & Paulson, J.C. (1985) J. Biol. Chem. 260, 8850 third instar larvae of Sarcophaga does not contain 8856 Sarcophaga lectin, but the lectin is promptly in 6. Giga, Y., Sutoh, K., & Ikai, A. (1985) Biochemistry duced when the body wall is injured (12). We 24,4461-4467 have found that Sarcophaga lectin is synthesized 7. Yeaton, R.W. (1981) Dev. Comp. Immunol. 5, 391 402 in the fat body, that the induction occurs at the 8. Lackey, A.M. (1981) J. Insect Physiol. 27, 139-143 transcription level, and that transcription of the 9. Stebbins, M.R. & Hapner, K.D. (1985) Insect Sarcophaga lectin gene is enhanced more by intru Biochem. 15, 451-462 sion of foreign substances into abdominal cavity 10. Parish, C.R. (1977) Nature 267, 711-713 of larvae than by mere injury of body wall (15, 11. Yeaton, R.W. (1981) Dev. Comp. Immunol. 5, 535 22). By analogy with Sarcophaga, we expect 545

J. Biochem. PURIFICATION OF Antheraea LECTIN 551

12. Komano, H., Mizuno, D., & Natori, S. (1980) 18. Ouchterlony, O. (1958) Prog. Allergy 5, 1-78 J. Biol. Chem. 255, 2919-2924 19. McCauley, R. & Racket, E. (1973) Mol. Cell. Bio 13. Komano, H., Nozawa, R., Mizuno, D., & Natori, chem. 1, 73-81 S. (1983) J. Biol. Chem. 258, 2143-2147 20. Lowry, O.H., Rosebrough, N.J., Farr, A.L., & 14. Komano, H. & Natori, S. (1985) Dev. Comp. hn Randall, R.J. (1951) J. Biol. Chem. 193, 265-275 munol. 9, 31-40 21. Qu, X.-M., Steiner, H., Engstrom, A., Bennich, H., 15. Takahashi, H., Komano, H., & Natori, S. (1986) & Boman, O.H. (1982) Eur. J. Biochem. 127, J. Insect Physiol. 32, 771-779 219-224 16. Laemmli, U.K. (1970) Nature 227, 680-685 22. Takahashi, H., Komano, H., Kawaguchi, N., 17. Fairbanks, G., Steck, T.L., & Wallach, D.F.H. Obinata, M., & Natori, S. (1984) Insect Biochem. (1971) Biochemistry 10, 2606-2617 14,713-717

Vol. 101, No. 3, 1987