カイコ体液中のキモトリプシンインヒビターの生 理・生化学的研究

九州大学学術情報リポジトリ

Kyushu University Institutional Repository

カイコ体液中のキモトリプシンインヒビターの生 理・生化学的研究

白井, 孝治

九州大学農学研究科遺伝子資源工学専攻

https://doi.org/10.11501/3130902

出版情報：Kyushu University, 1997, 博士（農学）, 課程博士 バージョン：

権利関係：

Biochemical a.nd physiological stt1dies on chymotrypsin inhibitors of

Bombyx mori

Koji ^Shirai

1997

Contents

General Introduction 1

Chapter 1 Purification and characterization of three

humoral chymotrypsin inhibitors, Cls-3, 8 and 13' 5

Materials and Methods 8

Results 16

Discussion 44

Chapter 2 Biosynthesis, secretion and incorporation of CI-8 in the fat body at 5th instar and during larval-pupal

metamorphosis 50

Materials and Methods 51

Results 60

Discussion 89

Chapter 3 Involvement of CI-8 in apoptosis of the midgut

during the larval-Pupal metamorphosis 96

Materials and Methods 97

Results 102

Discussion 136

General Discussion 141

Summary 147

Acknowledgment 152

References 15 3

General Introduction

Insects have the open circulatory system which is different from the closed system of mammals including human. Blood or hemolymph is not confined to blood vessels but simply bathes all tissues. Its circulation is facilitated by a long tubular pulsatile organ, i.e., the heart or dorsal vessel.

Hemolymph proteins move in the body with the flow of hemolymph, reaching all organs. This system favors the massive transport of materials from issue to tissue in relation to metamorphosis and for the storage of proteins to be utilized later to provide the amino acids necessary for the formation of adult structures. In Bombyx mori, storage proteins

1

and 2 (SP

1

and SP2) are rapidly synthesized in the fat body at the last larval instar, occupying about

1

_Oo/o of the total humoral proteins. However, ^at the onset of spinning, these proteins begin to be suddenly sequestered into the fat body. Later, the fat body SPs are used for the adult development and become hardly detectable (Tojo ^{e t}

al., 1980).

It has been proposed that a functional shift of the insect fat body

takes place at the end of the last larval stage, from biosynthetic organ to storage organ (Dean ^et

al., 1985;

Palli and Locke,

1988).

The quality and quantity of humoral proteins mirror the overall

physiological changes during development and metamorphosis (Doira,

1968;

Banno ^et

al., 1993).

Although a lot of proteins are present in the hemolymph of B. mori, knowledge about functional proteins like enzymes is scarce except for protease inhibitors and prophenoloxidase. The fact that B. mori

hemolymph possesses many kinds of protease inhibitors (Nakamura,

1981)

_is

of interest, since protease inhibitors may act as biomodulators protecting other functional proteinaceous materials from degradation. Hemolymph

chymotrypsin inhibitors (Cis) of B. nwri have relatively strong inhibitory activities, and are present in a number of isoproteins separable by

electrophoresis (Eguchi eta/., 1979a; Eguchi and Kanbe, 1982; Fujii eta/., 1989, 1996a and b). Cis are detected in the hemolymph consistently

throughout the whole life cycle of B. nwri (Aratake, 1990), thus considered to be indispensable for development.

In mammals, humoral protease inhibitors have been known to participate in the self-defense system such as blood coagulation (Kurachi eta/., 1976), complement activation (Levy and Lepow, 1959; Nagaki et al., 1974) and inflammation (Lenny et al., 1982; Brizin et al., 1984). Also one of the B.

mori Cis natned inhibitor-d (Eguchi and Shomoto, 1985) was found to inhibit

a protease from a fungus, Aspergillus me Ileus (St Leger et al., 1988). The fungus is considered to invade the epidermal cells by melting. But this fungus can not infect B. mori. Moreover neither in vivo nor in vitro inhibition of

fungal invasion by humoral Cis has been demonstrated (moreover, it is said that this fungus hardly infects B. mori), CI components may intrinsically have protective ability from pathogenic invasion. A relatively low-molecular­

weight inhibitor, CI-13 (Fujii et al., 1989), can possibly regulate proteolysis triggering the prophenoloxidase activation (A so eta!., 1994 ), which in turn controls melanization, a reaction necessary for the recognition of alien substances. The activity of phenol oxidase in the hemolymph of Manduca sexta was inhibited by commercially available serine protease inhibitors (Saul

and Sugumaran, 1986, 1987). The hemolymph activity of phenol oxidase increases when an antibody raised against alaserpin, a humoral serpin-type elastase inhibitor of M. sexta (Kanost and Wells, 1989; Kanost, 1990a), is added to hemolymph (Kanost, 1990b). M. sexta alaserpin can also inhibit the proteolytic activity of hemocyte (Kanost, 1990b).

In B. mori, 10 Jll of hemolymph from the 3rd larva can inhibit 1 x 1 o-10

moles of chymotrypsin (Aratake, 1990). The concentration of Cis in

hemolymph tnay be high enough to inhibit quickly the target protease. As to M. sexta , alaserpin is at a titer of 430 #g/ml hemolymph on day 3 of the 5th instar and inhibits elastase in 0.01 sec (Kanost, 1990b). It can be inferred, therefore, that these inhibitors have high potency as regulators.

There are reports of the developmental changes in overall relative activity of ^B. rrwri Cis (Eguchi et al., 1986b; Aratake et al., 1990). It

increased steeply from the middle of the 5th instar, showing a maximum at the onset of spinning, when the level was 5 times that at the early 5th instar. Then the activity reduced gradually throughout the pupal life. If Cis concern only the self-defense system, the extensive up-and-down pattern of activity level during development would not be necessary. In fact, the inhibitor for fungal protease shows no drastic changes in activity (Eguchi and Kanbe, 1982). Thus Cis are supposed to have indispensable roles in normal development.

In holometabolous insects, almost all tissues undergo programmed cell death or structural and functional rearrangement during the transition from larva to adult via the pupal stage. In ^B. rrwri, the prolegs, silk glands, muscles etc. are disintegrated completely, whereas the fat body, midgut, malpighian tubules etc. are remolded. At the onset of spinning, ^B. rrwri larvae cease feeding and prepare to pupate, and the building materials and energy sources required ever after largely depend on the recycled materials. Thus proteolytic activities and their regulation have special significance during metamorphosis of insects. ^B. rrwri Cis which change in activity during metamorphosis may participate in the programmed cell death. Proteases in Sarcophaga hemocytes can attack and destroy the larval fat body during pupation and this process is possibly regulated by sarcocystatin A, a low-molecular-weight humoral protease inhibitor (Suzuki and Natori, 1985; Kurata et al., 1989, 1992).

The brief review described above implies intriguing biological functions

of B. mori Cis, and there is a need for detailed studies on their

characterization, in particular the dynamic states in relation to metamorphosis.

The present article deals with an attempt to assess the problem. In Chapter 1, the author purified and characterized Cis-3, 8 and 13' as representatives of the three CI groups (as will be mentioned below, at least one of the members of each group consistently appear in all B. mori strains examined so far and are considered to be indispensable for development). There were differences among the three Cis not only in the general property but also in the specificity of inhibition against proteases. In Chapter 2, further studies were done with CI-8 (this CI was selected because it was found to be a glycoprotein; it also inhibited digestive juice protease). The organ synthesizing CI-8 was the fat body, from which this inhibitor was secreted into the hemolymph, and the distribution of CI-8 in the fat body during development was investigated by immunohistochemical method. The hemolymph CI -8 was again sequestered into the fat body. Interestingly, the fat body which synthesizes CI -8 and that which sequester it were localized at different body regions. In Chapter 3, CI-8 was shown to be incorporated into the larval midgut epithelial cells, which undergo apoptosis during the larval-pupal metamorphosis. On the basis of these results, the author debated the behavior and possible physiological role of CI -8 in the fat body and midgut.

Chapter 1

Purification and characterization of three humoral chymotrypsin inhibitors, Cls-3, 8 and 13'

The hemolymph of ^{B. mori} has more than

16

isoforms of Cis, some of which have been purified and characterized. One called inhibitor-d had a relatively high molecular weight of about

43,000

and a pi of

5.06

(Eguchi and Shomoto,

1985).

Inhibitor-e and g were smaller polypeptides of

7,200

^and

8,400,

respectively

(Matsui an

^{d E}

guch

^i,

1991), with neutral to basic pis

judging from their electrophoretic mobilities. The antiserum for inhibitor-d reacted with inhibitors-d and c (the molecular weight of the latter was similar to that of inhibitor-d) but not with inhibitors-a, e and g (Eguchi and Shomoto,

1984 )

^. A relatively high-molecular-weight CI, named sw-Achy, was identified to be a serpin type (Sasaki and Kobayashi,

1984)

and its eDNA was cloned and the amino acid sequence was deduced (Narumi et

al., 1993).

This was a

distinct molecule in sequence from another high-molecular-weight serpin-type CI, called sw-Achy II (Sasaki,

1991),

which showed a high homology to sw­

AT, one of humoral trypsin inhibitors of ^{B. mori} (Takagi et

al., 1990).

^Also

three low-molecular-weight Kunitz type Cis, named SCis-I, II and III, were analyzed for structure (Sasaki,

1978, 1984

^and

1988).

Although SCis-I and II were basic proteins whereas that of SCI-III was acidic protein, their amino acid sequences exhibited very high homologies to each other.

On the other hand, genetic analyses of Cis were performed by

zymogram using hemolymph specimens from more than

400

stock strains kept in Kyushu University (Fujii et

al., 1989;

^Aratake,

1990;

^{Doira et}

al., 1992;

r06 w43 u05 w44 w31 c011 k03 c13 w45 e02

-

+

81=�

�/CI-2'

=- Cl-3 Cl-4

�/CI-6

� ^"CI-7

Cl-8

_. Cl-9

� Cl-10

_. Cl-13'

� Cl-13

Fig. 1-1. Polymorphism of hemolymph chymotrypsin inhibitors (Cis) analyzed by native PAGE (5-15o/o gradient).

The strains tested are drawn along the top margin. The activity on the gel was determined as described in text. The volume of each sample applied was 7.5 Jll.

Shinohara, 1993; Fujii eta!., 1996a, b). As seen in Fig. 1-1, the isoforms

were named Cls-1 to 13 (including 6' and 13'), and Cls-b1 and b2, according to the mobility on electrophoresis (Shinohara, 1993). All Cis except b1 and b2 (these are now under genetic analysis) are controlled by the genes named let­

A, Iet-B, /et-D, fet-E and let-H. The let-A gene mapped to 23.7 eM _on the second linkage group (2-23. 7) is responsible for the expression of Cls-13 and 13'. Also Iet-B and Iet-H were located on the 2nd linkage group (let-B responsible for Cls-9 and 10 while let-H for Cls-1, 2 and 2'). Up to date, no recombinants have been detected between let-A and Iet-B and between let-A and /ct-H, implying that these three genes are closely linked to each other.

The Ict-D gene (19-9.9) controls the expression of Cls-6, 6', 7 and 8, whereas the fet-E gene (22-0.0) is responsible for C_ls_-3 _and 4. Studies to date have shown that any silkworm strain consistently has one or two of the allelic CI components controlled by each of let-A (B, H), ^{/et-D and Jet-E,}

suggesting that the expression of these genes is indispensable for development.

Generally, isozymes seem to be evolved under diverse environmental conditions and to acquire specialized functions. For example, animal tissues have three aldolase isozymes, A (the muscle type), B (the liver type) and C (the brain type)(Horecker et al., 1972). Although sharing the same catalytic

reaction, these differ in relative affinity for the substrates fructose 1,6-

bisphosphate and fructose 1-phosphate. This difference reflects the biological function of the cell or tissue to which the isozyme belongs. Also each group of Cis is considered to perform some distinct roles. Fujii and his colleagues have purified and characterized Cis-1, 2 and 13 (Fujii et al., 1989; Deng et al., 1990). These are the Iet-H and let-A members, and such studies on the Cis controlled by the fet-E and /et-D genes are lacking. In addition, Cis

reported by other groups could not be specified for the gene loci. Therefore, Cis-3, 8 and 13' were chosen in this Chapter as representatives of Jet-E , let-

D and let-A members, respectively (CI-13' is an allelic counterpart of CI- 13), and subjected to isolation and characterization.

Results showed that CI-8 was a glycoprotein but others were not.

Oligosaccharide moieties of proteins have absorbed researcher's attention since these offer key roles for protein functions such as translocation. Consequently, the author mainly examined CI-8 in subsequent experiments. Also an

antiserum for CI-8 was raised and used to detect the relevant protein(s) in various strains and to measure its developmental changes in a strain.

Materials and Methods

Materials

Bovine pancreas a-chymotrypsin

( l

_{E. C. 3.} ^4. 21. 1], 3xcrystallized and lyophilized) and the synthetic substrate for chymotrypsin, N-acetyl-D, L­

phenylalanine-�-naphthylester and its solvent N,N' -dimethylformamide , and 3,3 '-diaminobenzidine were purchased from Sigma Chern. Co. Ltd. The molecular weight markers (L.M.M.) were products of Pharmacia Biotech Co.

Ltd. Coomassie Brilliant Blue R250 was from Fluka Chemika Co. Carrier Ampholine was purchased from Pharmacia Biotech Co. Ltd. DEAE-Sepharose CL 6B and Sephadex G-50 were both from Pharmacia LKB Biotech. Co. Ltd., concanavalin A (ConA) agarose were from Seikagaku Kogyo Co. and Butyl­

Toyopearl was from Toso Co. Peroxidase-labeled anti-rabbit lgG goat

antiserum was purchased from TAGO Burlingame, CA. HiTrap™ protein A­

Sepharose HP was a product of Pharmacia LKB Biotechnology. Alkaline phosphatase, ConA and peroxidase used were also commercial products of Toyobo Co. Ltd., Sigma Biochem. Co., Ltd. and Toyobo Co., Ltd.,

respectively. Peroxidase-lectin kit A was from Honen Co. Ltd. Trypsin, ficin, carboxypeptidase A, V8 proteinase and peptidase from Serratia species were purchased from Sigma Chern. Co., Ltd.

Animals

All strains tested were maintained in Laboratory of Insect Genetic Resources, Kyushu University. The cz strain of

B.

mori was used unless otherwise indicated. This strain has Cls-3, 8 and 13' in the hemolymph and was homozygous for each of

let-A, Iet-D

_and

Jet-E

genes. Antiserum

against CI-8 was raised also from this strain. The strains used for the detection of Cis in crude hemolymph preparations were t45, g60, g52, f03, w43, w44, u05, w32 and c60, covering Cls-1, 2 and 2'

(let-H),

Cls-3 and

4(/ct-E),

Cls- 6, 7 and 8

(let-D),

_CI-10

(let-B)

and Cls-13' and 13

(let-A)

_(Doira

et

al., 1992). Larvae were reared on mulberry leaves. Hemolymph was collected from larvae by cutting abdominal legs, freed from hemocytes by

centrifugation at 500 ^x g for 10 min, mixed with a small amount of phenylthiourea, and stored at -40°C until use.

Assay of inhibitory activity

The chymotrypsin inhibitory activity of CI was essentially assayed according to the method of Fujii

et

al. ( 1989). The standard reaction mixture (0.5 ml) contained 12.5 ]lg of a-chymotrypsin, 2 mg of casein, 0.04 M _sodium phosphate buffer, pH 7 .4, and CI. After incubation at 37°C for 10 min, the solution was rapidly mixed with trichloroacetic acid (TCA, final 5% ), allowed to stand at 37°C for 30 min and centrifuged at 5,000 ^x g for 10 min. The supernatant was measured for the absorbance at 280 nm (value T). Three controls were measured: the same procedure as above but casein was added after TCA (value C), the mixture was incubated without CI (value

E)

_{and also}

without CI but casein was added after TCA (value B). The inhibition of chymotrypsin activity was calibrated by the equation { (E-B)-(T -C)/

(E-B) }xi

00.

One unit of inhibitor was defined as the amount required for the complete inhibition of

4

_x

10-12

moles of a-chymotrypsin.

Determination of protein concentration

The protein concentration was determined by Folin method using filter paper tips with bovine serum albumin as a standard (Hayashi,

1983).

_During

column chromatography, proteins were measured by the absorbance at

280

nm.

Polyacrylamide gel electrophoresis

Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (SDS-PAGE) was done using a slab gel (Laemmili,

1970).

_The

molecular weight markers

(L.M.M.)

used were phosphorylase b

(94,000),

bovine serum albumin

(67,000),

_ovalbumin

(43,000),

carbonic anhydrase b

(30,000),

soybean trypsin inhibitor

(20,100),

and a-lactalbumin

(14,400).

_The

proteins were stained with Coomassie Brilliant Blue

R250.

Polyacrylamide gel electrophoresis under undenaturing conditions (native PAGE) was carried out using a slab gel as previously specified (Davis,

1964).

Detection of protease inhibitory activity on electrophoresis gel (zymogram) CI activity was detected after native PAGE (Uriel and Berges,

1968).

After electrophoresis, the gels were immersed in I

0

)lg/50 ml of a­

chymotrypsin solution made up in

0.1 M

sodium phosphate buffer, pH

8.0,

_and

incubated at

37°C

_for

10

min. Then the chymotrypsin solution was removed and the gel was rinsed with distilled water. After the gels were dried at

37°C

for 10 min, the staining sol uti on containing 10 mg/5 ml of N -acetyl-0, L ^- phenylalanine-

�

-naphthylester dissolved inN, N '-dimethylformamide and 50 mg/50 ml of tetrazotized orthodianisidine dissolved in sodium phosphate buffer, pH 8.0, was poured onto the gels, which were then incubated at 37°C for 15 min. After incubation the solution was discarded and the gels were washed with 2% acetic acid. CI was detected as an unstained zymogram band.

The inhibitory activity against trypsin was detected on the gels basically by the same procedure for CI except 10 mg/50 ml of trypsin solution was used instead of chymotrypsin solution.

Gel isoelectric focusing to estimate pis

Gel isoelectric focusing was done on 4% polyacrylamide gel containing 2o/o (v/v) carrier Ampholine.

Purification of Cis

All procedures except for ammonium sulfate precipitation were

performed at 4 °C. The hemocyte-free hemolymph of the cz strain was mixed with solid ammonium sulfate, and the precipitate between 30 and 70%

saturation of the salt was collected by centrifugation, taken up in buffer A (0.05 M Tris-HCl buffer, pH

6.8)

and dialyzed overnight against buffer A at 4°C. The solution was loaded onto a DEAE-Sepharose CL

6B

column (2.4 x 30 em), which was washed with buffer A and eluted with a linear gradient from 0.0 to 0.3 M NaCl made up in buffer A. The fractions containing CI activities were pooled and mixed with solid ammonium sulfate up to 70%

saturation. The precipitate was collected, taken up in buffer A and applied onto a ConA-agarose or Sephadex G-50 column equilibrated and eluted with buffer A. If necessary, the eluates containing CI activities were combined, concentrated as above and carried to the chromatography through a Butyl-

Toyopearl column equilibrated with buffer A. After the passed-through fraction was discarded, a linear gradient elution wa performed by decreasing concentration of ammonium sulfate from 15 to 0%.

pH and thermal stabilities of CI

To examine the pH stability, a CI solution (0.25 units/ Jll) was incubated at 37°C for 20 min at various pHs. The buffer systems used were citrate­

phosphate covering pHs from 3 to 7, Tris-HCl from 7 to 9 and glycine-NaOH from

9

to 1 1. The CI solution was assayed for its activity at pH 7.4 by using the reaction mixture containing 2.5 units of CI and 12.5 Jig of a-chymotrypsin and relative values to the control at pH 7.0 (Tris-HCl buffer) were obtained.

To examine the thermal stability,

a

CI solution (0.25 units/jtl) was incubated at various temperatures for 10 min, and its activity was assayed using the reaction mixture containing 2.5 units of CI and 5 Jlg of a­

chymotrypsin. The relative activity to that of the control at 4

OC

aws evaluated.

Preparation of antisera against Cis

To prepare antiserum for CI-8, purified preparation was injected subcutaneously into a rabbit with Freund's complete adjuvant (once a week, 4 times). A previously raised antiserum for CI- 13 (Deng, 1990) was also used.

Western blotting of Cis with antiserum against a CI

Proteins were resolved by native PAGE. After the electrophoresis, a gel was immersed in a transfer buffer (25 mM Tris and 192 mM glycine, pH 8.8) at room temperature for 30 min with gentle shaking, and proteins were

electrophoretically transferred onto PVDF membranes. The membrane was washed briefly in TBST llO mM Tris-HCI buffer, pH 7.4, containing 0.05%

Tween 20 (polyethylene 20 sorbitan monolaurate) and 150 mM NaClj.

Subsequently, the membrane was immersed in the blocking solution composed of TBST plus 5% skim milk to reduce a non-specific absorption. After the blocking solution was discarded, the membrane was incubated with antiserum solution [1/1,000 (v/v) for anti CI- 1 3 and 1 /10,000 for anti CI-8 diluted with TBST containing 0.5% skim milk] at room temperature for 1.5 hr, and washed 3 times with TBST for 1 5 min and soaked at room temperature for 1 hr in the second antibody solution (peroxidase-labeled anti-rabbit IgG goat antiserum diluted with TBST containing 0.5% skim milk). Finally, the membrane was treated with a staining solution ( 1 0 mg of 3,3 '-diaminobenzidine and 4 JLl of hydrogen peroxide in 30 ml of 0.1 M sodiu1n phosphate buffer, pH 7.4). After markedly brown bands had appeared, the membrane was thoroughly washed with distilled water.

Enzyme-linked immunosorbent assay (ELISA)

Enzyme-linked immunosorbent was essentially prepared according to a method previously reported(Voller et al., 1976). In brief, antiserum for CI was dialyzed overnight against 20 mM sodium phosphate buffer, pH 7.0, and loaded onto a column of Hi Trap™ protein A-Sepharose HP equilibrated with the same buffer. After washed with five column volumes of the same buffer, IgG fraction was eluted with 0.1 M citric acid, pH 4. IgG was concentrated using Centriplus™ and dialyzed against 20 mM sodium phosphate buffer, pH 7.0. An aliquot of the IgG solution containing 0.4 mg protein was mixed with an equal volume of 1 mg/ml alkaline phosphatase solution. The mixture was dialyzed for 18 hr against PBS (phosphate-buffered saline) at 4°C and mixed with 20 Jll of 20% (v/v) glutaraldehyde. After incubation at room temperature for 1 .5 hr, glutaraldehyde was removed by overnight dialysis against 0.05 M Tris-HCl buffer, pH 8.0, at 4°C. To the dialysate was added 0.05 M Tris-HCl buffer, pH 8.0, containing 1% (w/v) bovine serum albumin and 0.02% (w/v)

sodium azide to the total volume of 4.0 mi.

Enzyme-linked immunosorbent assay (ELISA) was performed in a flat­

bottom Limbro EIA plate (96 wells). Each well received 200 ;ll of antibody diluted in carbonate coating buffer, pH 9.6 (2 Jiglml), then the plates were incubated at 4 OC overnight. After coating, the wells were washed three times for 5 min with PBS containing 0.05% Tween 20 (PBST), then filled with antigen diluted in PBST containing 1% bovine serum albumin (PBST-BSA) and kept at 37°C for 2 hr. After the antigen solution was removed by suction, the wells were washed three times for 5 min with PBST, filled again with PBST

-BSA and kept at 37°C for 1

hr to block non-specific antibody binding on the plastic wells. The blocking solution (PBST -5o/o BSA) was removed and 200 Jil of alkaline phosphatase-labeled

anti-CI rabbit IgG

was added

to

each well, and the plates were then placed at 37°C for 2 hr. The solution was removed, and the wells were washed three times again. Substrate solution

( 10% diethanolamine buffer, pH 9.6, containing p-nitrophenyl phosphate at a concentration of 1 mg/ml) was added to the wells (200 Jillwell) and incubated at 25°C for 2 hr. The reaction was stopped by adding 50 Jil of 3 N NaOH to each well. Then the absorbance of each well was measured at 405 nm.

Detection, determination and structural analysis of carbohydrate moiety

Carbohydrate moiety was detected under previously specified conditions (Ochiai, 1984). After native PAGE, the gels were blotted electrophoretically onto PVDF membranes, which were blocked by TBST plus 5% skim milk as described for the Western analysis. Then the membranes were reacted with ConA (0.5 mg/ml of 15 mM sodium phosphate buffer, pH 6.7, containing 80 mM NaCI) at 37°C for 1.5 hr. After the solution was removed, the

membranes were washed three times with TBST, treated with peroxidase (0.1

mg/ml of 15 mM sodium phosphate buffer, pH 6.7) at 37°C for 60 min and

washed three times with TBST. Then the carbohydrate-containing protein bands were detected by the same staining method as described for the Western blotting.

Carbohydrate moiety was also detected by the phenol-sulfuric acid method. Solution of purified protein was diluted with distilled water to 42 jig/100 Jil and mixed with 100 Jil of 5% phenol solution and 500 Jil of

concentrated sulfuric acid. The mixture was kept at room temperature for 30 min, observed for the positive reaction (brown) and measured for the

absorbance at 490 nm. The content of oligosaccharide chain was calibrated using glucose as a standard.

The carbohydrate chain was analyzed by using a peroxidase-lectin kit A (Kijimoto-Ochiai et al., 1985� Katagiri et al., 1989). After SDS-PAGE, the gels were blotted onto PVDF membranes, which were immersed in a solution containing lectin conjugating house radish peroxidase (six kinds of lectins were used as indicated in Results). The membranes were washed to remove surplus lectin as far as possible. Lectin-bound proteins were visualized as brown bands. From the results, the possible structure of carbohydrate moiety was deduced.

Specificity of Cis against various proteases

Specificity of CI activity against various proteases were studied according to the method of Fujii et al.( 1989) using casein as a substrate.

Commercially available proteases including peptidase from Serratia species were used as specified in Results. Partially purified Beauveria bassiana protease (Shimizu et al., 1992) was kindly provided by Dr. S. Shimizu.

Digestive juice used in this study was collected from 5th instar larvae of the cz strain.

Results

Purification of Cls-3, ⁸ and 13'

The ammonium sulfate fraction of hemolymph of the cz strain, on day ⁴

of the 5th larval instar, was subjected to the chromatography with a DEAE­

Sepharose CL ^6B column. The eluates were assayed for CI activity, and aliquots of active fractions were submitted to zymogratn. CI activities were separated into three parts, which contained CI-3, ⁸ and

13';

these were eluted in this order (Fig.

1-2).

These Cis were further purified by the following procedure.

1.6

0.30

1.2 -

-

0

re

_o.8

<(

0

0.15

5

0

0.4

0.0

0 20 40 60 80

Fraction No.

0.00

100 120

0 ('(J

z

Fig.

1-2.

Elution profile of Cis on a DEAE-Sepharose CL ^6B column.

Elution profile of ammonium sulfate fraction (30-70%) of Cis on a DEAE-Sepharose CL ^6B column

(2.4

x 30 cm),equilibrated with 0.05

M Tris-HCI buffer, pH ^6.8, were eluted with a linear gradient of NaCI from 0.0 to 3.0 ^M. Each fraction contained 5.0 mi.

0 CX) N

<(

0.3

0.2

0.1

0.0

0 10

Fraction No.

20

Fig.1-3. Elution profile of CI-3 on a ConA-Agarose column.

Affinity chromatography of CI-3 on a ConA-Agarose column (0.8 x 20 em), equilibrated and eluted with 0.05 M Tris-HCl buffer, pH 6.8. Each fraction of 1.0 ml was collected.

CI-3 was in the washed-through fraction.

The CI-3 fractions without other Cis were combined and loaded onto a

ConA-Agarose column (Fig. 1-3 ). The passed-through fractions having CI -3 were collected and carried to Butyl-Toyopearl column chromatography. CI -3 was adsorbed under the conditions applied, indicating that it had weak affinity to the gel. Proteins were eluted by a negative gradient of ammonium sulfate (Fig. 1-4 ). The final preparation showed one zymogram band and one protein band on native PAGE, and one band on SDS-PAGE (Fig. 1-5A, B), indicating that CI-3 was pure.

The DEAE-Sepharose eluate containing Cl-8, slightly overlapping with

CI-13',

were collected and carried to ConA-Agarose chromatography.

Proteins made peaks(Fig. 1-6); the second one contained CI-8, indicating that

CI-8

had affinity to ConA as a ligand. Although being almost free of CI- 13' (the latter was in the first peak), the pooled fraction was further separated by Butyl-Toyopearl column chromatography applying a negative gradient elution as above. CI-8 was eluted as a single protein peak (Fig. 1-7), and found to be fairly pure upon analyses by native and

^SDS-PAGE

(Fig. 1-8A and B).

0.04

0 (X) C\1

<(

^0.02

0.00

0 20 40 60 80

Fraction No.

100

30-� 0 -

0 r::::

0 (.)

20-; Q)

-3

en

E ::l

r::::

10 E °

ct E

120

Fig.

^1-4.

Elution profile of CI-3 from a Butyl-Toyopearl column (

²

.

⁴ x 20

em).

The column was equilibrated with

^{0.05 M}

Tris-HCl buffer, pH 6.8, containing

^30%

saturation of ammonium sulfate and developed with a linear gradient of ammonium sulfate from 30 to

0%

saturation. Each fraction contained

^5.0

ml.

(A)

81 Cl-3 81 Cl-3

-94,000 -s7,ooo -42,000

Cl-3 � ..-- Cl-3

-29,000

-20,100 -14,400

Cl activity protein protein

Fig. 1-5. Gel electrophoretic patterns of purified CI-3.

Bl, CI-3 and L.M.M. indicate hemolymph, purified CI-3 and molecular weight markers, respectively. Arrows indicate the position of CI-3. (A) Native PAGE (10%) and (B) SDS-PAGE (12.5%).

0 00 C\1

<(

0.8

0.6

0.4

0.2

0.0

0 10 20 30

Fraction No.

40

Fig. 1-6. Affinity chromatography of CI-8 on a ConA-Agarose column.

Affinity chromatography of CI-8 on a ConA-Agarose column (1 x

15 em) equilibrated and eluted with 0.05 M Tris-HCI buffer, pH 6.8.

Fractions of 1 ml were collected.

0 CX) N

<

0.2

-� 15 0 ^-

0 c 0 0

.... Q)

10 m :!::

:::::J

0.1 U)

E 5 ::1 _c :::::J 0 E

� E

0.0

0 20 40 60 80 100

Fraction No.

Fig. 1-7. Elution profile of CI-8 on a Butyl-Toyopearl HW55 S _column.

Elution profile of CI-8 on a Butyl-Toyopearl HW55 S column (2.4 x 28 em) equilibrated with 0.05 M Tris-HCl buffer, pH 6.8, containing

15% saturation of ammonium sulfate and eluted with a linear gradient of ammonium sulfate from 15 to Oo/o. Each fraction contains 5.0 _mi.

(A)

Cl-8 �

Bl Cl-8

(B)

94,000- 67,000- 42,000-

29,000-

20,100-

14,400-

L.M.M. Cl-8

.._CI-8

Fig. 1-8. Gel electrophotetic pattern of purified

CI-8.

(A) Native PAGE

(10o/o). (B)

SDS-PAGE

(12.5%).

_For

Bl, CI -8

and L.M.M., see the legend to Fig.

1-5.

The CI -13' fractions at the DEAE-Sepharose step were pooled and further purified by a Sephadex G-50 column, from which two protein peaks appeared and CI-13' was in the second peak (Fig. 1-9). The final preparation ofCI-13' seemed to be fairly pure upon native and SDS-PAGE (patterns not shown).

0 CX) C\1

<C

0.6 -

-

0.4 -

-

0.2 -

-

0.0

0 10

� I 4

J

l

^I

20 30

Fraction No.

40

Cl-13'

...., - ... ,.

50 60

Fig 1-9. Gel filtration of CI-13' on a Sephadex G-50 column.

Gel filtration of CI-13' on a Sephadex G-50 column (2.8

x 75 em), equilibrated with 0.05 M Tris-HCl buffer, pH 6.8.

Fractions of 5.0 ml were collected.

Tables 1-1, 1-2 and 1-3 summarize the typical purification procedures for Cls-3, 8 and 13', respectively. These were isolated via the four or five steps with recoveries of 29.7, 0.4 and 4.7o/o, respectively.

Molecular weights and pis of purified Cls-3, 8 and 13'

The results of SDS-PAGE shown in Figs. 1-5B and 1-8B, wherein Cls-3 and 8 were co-electrophoresed with molecular weight markers, were replotted

Table 1-1. Summary of purification of CI-3.

The starting volume of hemolymph was 10 ml. The total protein contents and the total activities were measured after collecting the active fractions at each step during the purification procedures.

Total Total Specific

Recovery Step protein activity activity

(o/o)

^Fold

(mg) (units) (units/mg)

Blood

270.0 7200.0 26.7 100.0 1.0

30-70°/o

(NH4):z5� 201.0 6030.0 30.0 83.8 1.1

DEAE-

220.0 33.8 3.1 1.3

Sepharose

6.55

Con A-

130.0 55.6 1.8 2.1

Agarose

2.34

Butyl-

0.10 26.8 263.0 0.4 9.8

Toyopearl

Table 1-2. Summary of purification of CI-8.

The starting volume of hemolymph \Vas 10 ml. The total protein contents and the total activities were measured after collecting the active fractions at each steps during the purification procedures.

Total Total Specific

Recovery

Step protein activity activity Fold

(mg) (units) (units/mg) (o/o)

Blood

155.0 6120 39.8 100.0 1.0

30-70°/o

104.0 3350 32.3 54.6 0.8

(NH4)2S04 DEAE-

10.7 1030 96.7 16.9 2.4

Sepharose Con A-

0.90 425 474.0 6.9 11.9

Agarose Butyl-

0.26 287 1100.0 4.7 27.7

Toyopearl

Table 1-3. Summary of purification of CI-13'.

The starting volume of hemolymph was 10 ml. The total protein contents and the total activities were measured after collecting the fractions at each step during the purification procedures.

Total Total Specific

Recovery Step protein activity activity

(o/o)

^Fold

(mg) (units) (uniVmg)

Blood

213.0 6780 31.9 100.0 1.0

30-70°/o

(NH4)2S04

158.0 6360 40.4 93.7 1.2

DEAE-

2.19 2340 1070.0 34.3 24.9

Sepharose

Sephadex

0.22 2020

G-50 9190.0 29.7 294.0

as illustrated in Figs. 1- 10A and 1- 1 1A. Also the results for CI- 13' was seen

in Fig. 12A. The molecular weights of Cls-3, 8 and 13' were determined to be 40,000, 42,000 and 14,000, respectively. On the other hand, isoelectric

focusing analyses performed by the gel method indicated that pi of CI-3 was 5.5, that of CI-8 was 5.0 and that of CI-13' was 4.4 (Figs.

1-IOB,

1- 1 1B and

12B, respectively).

pH stability and thermostability of Cis-3, 8 and 13'

When incubated at various pHs for 20 min (at 37°C), the activity of CI-3 did not change largely between pHs 5 and 9, but lowered under alkaline (over pH 1 1) and acidic (under pH 3) conditions (Fig. 1- 13). CI-8 was also stable between pHs 6 and 1 1, but unstable at acidic pHs (Fig. 1-14 ). In contrast, CI-

13' kept its activity at all pHs tested (Fig. 1-15) , indicating that this inhibitor

was highly stable.

After heated at SSoC for

10

min at pH 7

(Figs. 1-16), Cls-3

and 8

(A) _1.2

a-lactalbumin

1.0 trypsin inhibiter

0.8 carbonic anhydrase

't-a:

0.6

0.4

phosphorylase b 0

10 4 10 5

Molecular weight

(B) 10

8

i

⁶

4

0 2 4 6 8 10 12

Distance from anode

Fig. 1-10. Measurement of molecular weight and estimation of pi of CI-3.

(A)Measurement of molecular weight of CI-3 by SDS-PAGE. For original patterns see Fig. 1-5B. (B) Estimation of pi of CI-3 by isoelectric focusing.

(A)

(B)

1.0 a-lactalbumin

carbonic anhydrase

0.5

ovalbumin

0

r

10 ₄

I I

10 ₅ Molecular weight

8

7 :I: Q. 6

5

4

o

T

^{I I I I I I I I I I I}

0 2 4 6 8 10 12 14

Distance from Anode

Fig. 1-11. Measurement of molecular weight and estimation of pi of CI -8.

Measurement of molecular weight of CI-8 by SDS-PAGE. For original patterns see Fig. 1-8. (B) Estimation of pi of CI-8 by isoelectric focusing.

(A)

1 .0

lt _0.5

myoglobin 2

trypsin inhibitor

ovalbumin

Cl-13'

carbonic anhydrase

albumin

phosphorylase b 0. 0 -4---.---.--...-r-r-.,....,....,..-r---r----r----r---T--r-T--r-r-.

(B)

:I: 0.

103 104 105

Molecular weight 8

7 6 5

4 0

0 2 4 ⁶ 8 10 12 14

Distance from anode

Fig. 1-12. Measurement of molecular weight and estimation of pi of CI-13 '.

(A) Measurement of molecular weight of Cl-1 3 ' by SDS-PAGE

performed by 15% gel (original patterns not shown).

(B)

Estimation of pi of CI -1 3' by gel isoelectric focusing.

120

� .-

0 -

�

⁸⁰

>

;::;

(.) (U

.� .... C1) _(U 40 4) a:

2

__... Citrate-phosphate buffer "

--o- Tris-HCI buffer Glycine-NaOH buffer

4 6 8 10

pH

Fig. 1-13. Effects of pH on stability of CI-3.

12

Each test used 25 units of CI-3. The CI activities were assayed after the CI solutions were kept at indicated pHs for 20 min at 37°C. The results were expressed in percentages to the activity at pH 7 (Tris-HCl buffer). The buffers (and pH ranges covered) were citrate-phosphate (pHs 3 to7), Tris-HCl (pHs 7 to

9)

and glycine-NaOH (pHs

9

^{to 10).}

120

-�

0 -

�

⁸⁰

>

;; 0 (U (I) >

;;

:

⁴⁰

a:

2 4

J

• Citrate-phosphate buffer --[}-Tris-HCI buffer

·· Glycine-NaOH buffer

6 pH

8 10 12

Fig. 1-14. Effect of pH on stability of CI -8.

Each test used 25 units of CI-8. The CI activities were assayed after the CI solutions were kept at indicated pHs for 20 min at 37°C. The results were expressed in percentages to the activity at pH 7.0 (Tris-HCI buffer). For buffers see the legend to Fig. 1-13.

120

-�

0 -

80

40

2 4

8 Citrate-phosphate buffer ...[].- Tris-HCI buffer

Glycine-NaOH buffer

6 8 10 12

pH

Fig. 1-15. Effect of pH on stability of CI-13'.

Each test used 25 units of CI-13'. The CI activities were assayed after the CI solutions were kept at indicated pHs for 20 min at 37 °C. The results were expressed in percentages to the activity at pH 7.0 (Tris-HCl buffer). For buffers see the legend to Fig. 1-13.

-�

0 -

100

50 --0- Cl-3

e Cl-8

1''"''''" Cl-13'

4 35 40 45 50 55 60 65 70 75 80 85 90 95 100

Temperature (°C)

Fig. 1-16. Thermostability of purified Cis.

Cis of 25 units were used in a test. The CI activities were measured after the reaction mixtures were heated for 10 min at indicated temperatures. The results were expressed in percentage to the activity of each inhibitor kept at 4 °C.

scarcely lost their activities. Heating at 65°C completely spoiled both of the CI activities. In contrast, CI-13' kept almost its activity even at 1 00°C, indicating that this inhibitor was highly stable also to heat.

Detection of carbohydrate in CI-3, 8 and 13'

Analysis for carbohydrate moiety using the PVDF membrane-peroxidase method after native PAGE gave a positive band at the position of CI-8 but not at those of CI-3 and 13' (Fig. 1-17). These results, together with the fact that CI-8 had marked affinity to ConA-Sepharose (see Fig. 1-6), indicated that CI-8 was aN -glycoprotein.

Fig. 1-17. Detection of oligosaccharide chains of the purified CI.

Cis were subjected to native PAGE ( 1

Oo/o)

and the gels was blotted to a PVDF membrane, which was immersed in ConA and then in peroxidase.

Analysis of carbohydrate chains of CI -8

The CI-8 preparation gave a clear color reaction by phenol-sulfuric acid method (Fig. 1-18A), supporting the notion that CI-8 was a glycoprotein. The absorbance at 490 nm when calibrated by a standard curve for glucose (Fig. 1-

18B) indicated that the molecular weight of carbohydrate moiety was about 1,000.

CI-8 was analyzed by the lectin-peroxidase method using six different lectins (Table 1-4 ). CI -8 reacted with ConA, lens culinaris agglutinin (LCA)

(A)

, ,

(B)

0.4

0.3

0.2

, CI-8

0.1

0 1 2 3 4 5 6

1Jg/tube

Fig. 1-18. Detection of carbohydrate in CI -8.

(A)

Detection of carbohydrate in CI-8 by treating with phenol and H2S04. Blank test was done by using distilled water instead of CI-8. (B) Measurement of the carbohydrate content in CI -8 by absorbancy at 490 nm after treatment with phenol-H2S04 and calibrated by using glucose as a standard.

and erythroagglutinating phytohemagglutinin E4 (PHA-E4), but not with peanut agglutinin (PNA), wheat germ agglutinin (WGA) and Ricinus communis agglutinin 120 ⁽R^CA 120). The latter three agglutinins gave negative reaction even after CI-8 was treated with sulfuric acid to remove possible sialic acid at the end of oligosaccharide chain.

Table. 1-4. Lectin-peroxidase reaction of CI-8.

After SDS-PAGE (10%), the gels were blotted to PVDF membranes, which were immersed in the indicated lectin labeled with peroxidase. Positive reaction (brown color) was indicated by+.

Lectin

Con A LCA PNA WGA PHA-E ₄ RCA120 PNA*

WGA*

RCA120*

+ +

+

*CI-8 was treated with H2S04 before analysis.

On the basis of the above results, it was speculated that the

oligosaccharide moiety is a biantennary chain of a complex type (Fig. 1-19).

The molecular weight of this putative oligosaccharide chain is again about 1 ,000, similar value to the total carbohydrate content (see Fig. 1-18B), suggesting the possibility that CI-8 has a single carbohydrate moiety per molecule.

·- - - -·

GlcNAc�1 Fuca1

� GlcNAc �1-l-2 Man a 1 I I

: : 6

4

⁶

: :

/

³

Man � 1 - 4GicNAc � 1 - 4GicNAc - Asn : GlcNAc �1

--:-₂

Man a 1

L ... �

Fig. 1-19. The putative structure of oligosaccharide chain of CI-8 on the basis of the lectin-peroxidase analysis (Table 1-4 ).

The square shows the region which was not determined by this studies. The overall structure was presumed to be biantennary chain of a complex type.

Relative inhibitory activities of CI-3, 8 and 13' against various proteases

All activities were measured with casein as a substrate. Cls-3, 8 and 13' inhibited the activity of a-chymotrypsin (EC 3.4.21.1) strongly (Table 1-5).

CI-3 and 8 also inhibited the proteolytic activity of crude digestive juice, but CI -13' did not. Moreover, Cls-8 and 13' had only a weak or no effects on activity of a protease partially purified from B. bassiana which is pathogenic to B. mori. All three Cis did not markedly affect the activities of trypsin (EC 3.4.21.4), ficin (EC 3.4.22.3), carboxypeptidase A (EC 3.4.2.1.), V8

proteinase (EC 3.4.21.19) and Serratia peptidase.

Table 1-5. Inhibitory specificity of Cis vs. different proteases.

All measurements were made in a linear range of proteolytic activity using casein as a substrate. The inhibitory activities were expressed in percentage to that against a-chymotrypsin.

Relative activity Cl-3 Cl-8 Cl-13'

a-chymotrypsin 51JQ 100.0 100.0 100.0

Trypsin 10 IJQ 0.0 3.3 0.0

Ficin 41JQ 2.9 11.1 0.0

Carboxypeptidase A 221JQ 0.0 7.1 0.0

VB proteinase 10 IJQ 0.0 7.8 0.0

Serratia _{peptidase 10 IJQ 0.0 0.0 1.2}

B. bassiana _protease Partially purified 13.2 0.0

Digestive juice Crude 100.0 100.0 0.0

-, not determined.

Reactivity of CI-3, 8 and 13' with antiserum for CI-13

CI-13' gave a clear band in the Western blotting analysis (after native

PAGE)

using an antiserum previously raised against CI-13, but CI-3 and 8 did not (Fig. 1-20). Therefore, Cls-3 and 8 had a low immunological homology to Cls-13 and 13'.

Detection of proteins immunologically related to CI-8 in hemolymph specimens

From larvae of nine B. mori strains covering almost all Cis controlled by let genes, hemolymph specimens were prepared and subjected to a native

PAGE

followed by the Western blotting using an anti-CI-8 serum raised in the present study (Fig. 1-21 ). This antiserum only reacted with the bands of Cls-

6, 7 and 8 (controlled by alleles of the Ict-D locus) but not with others.

Cl-3 Cl-13' Cl-8

Fig. 1-20. Immunological cross-reactivity of Cls-3, 8 and 13' vs.

antiserum for CI -13.

Immunological cross-reactivity was tested by the Western blotting method after native PAGE (7.4o/o). Arrow indicates the position of CI-13'.

Changes in total CI activity in hemolymph during development

Hemolymph specimens of the cz strain were daily taken from the 3rd ecdysis to the adult stage and assayed for the CI activity in toto. As shown Fig.

1-22, changes in activity in male were similar to those in female. The titer was low during the 4th in star: about 25 to 35 units/ 10 Jll of hemolymph. At the 4th molting, the level dipped to about 15 units/10 Jll, then increased gradually until day 3 of the 5th instar. Thereafter, the activity rose

(A)

t45 g60 g52 f03 w43 w44 uos W32 c60

(B)

145 g60 g52 f03 w43 w44 u05 W32 c60

Cl-10

Cl-13'

�--CJ-13

Cl-1

--CI-2

__/_CI-2'

---

81=�

Cl-6

�

^CI-7

Cl-8 Cl-10

Cl-13' --CI-13

Fig. 1-21. Immunological cross-reactivity of Cis in crude hemolymph specimens from nine B. mori strains vs. antiserum for CI -8 as tested by the

Western blotting method after native PAGE

(7.4%).

(A) Polymorphism of hemolymph Cis confinned by native PAGE

(10%)

followed by zymogram.

(B)

The results of the Western blotting. The strains used are drawn along the top margin.

>

== >

:;;

150

• Male activity ---0-- Female activity

�

⁵⁰

0

3M 4--1 4--2 4--3 4M 5-1 5-2 5-3 5-4 5-5 S.1 S.2 P-1 P-2 P-3 P-4 P-5 P-6 P-7 A

Stage

Fig. 1-22. Changes in activity of hemolymph Cis during larval, pupal and adult development.

Ordinate shows the units per 10 Jll of hemolymph. 3M, 4-x, 4M, 5-x, S-x, P-x and A indicate the 3rd molt, x day of the 4th instar, the 4th molt, x day of the 5th ins tar, x day of spinning, x day of pupa and the onset of emergence, respectively. The cz strain was used. This has a genetic marker to distinguish male(P

3)

and female (Ze).

steeply, peaking at the onset of spinning

(S-1,

a value of 110 to 120 units/10

Jd)

and then began to drop abruptly. The decrease became moderate after pupation, reaching at the level of 35 to 20 units/ 10 Jll.

The components of Cis as revealed by zymogram remained unchanged

throughout the whole developmental stages (Fig. 1-23), except that, in some individuals, the band of CI-3 was rarely seen at the mid-pupal stage.

Stage

3M 4-1 4-2 4-3 4M 5-1 5-2 5-3 5-4 5-5

Fig. 1-23. Electrophoretic patterns of activity of hemolymph Cis during larval, pupal and adult development.

The same samples (female) as those used in Fig. 1-22 were subjected to native PAGE (10

o/o)

followed by zymogram analysis.

For abscissa see the legend to Fig. 1-22.

Changes of CI-8 concentration in hemolymph during development

The daily changes of CI-8 concentration in hemolymph during the 5th in star and pupal stage were assayed by ELISA (Fig. 1-24) using the anti-CI-8

serum. The concentration increased markedly after day 2 of the 5th instar and attained a sharp peak at the onset of spinning: the concentration at the peek was about 1 mg/ml. Then it declined rapidly at a rate of about 500 )1g/ml/day.

Thereafter, the concentration decreased gradually until the late pupal stage.

This pattern was substantially consistent with that of the total CI activity of hemolymph.

.-...

�

�

.._...

c 0

;;

m -

... c Q) u c:

0 u

1000

500

e male blood

--0-- female blood 0

4M 5-1 5-2 5-3 5-4 5-5 8-1 8-2 P-1 P-2 P-3 P-4 P-5 P-6 P-7

Stage

Fig. 1-24. Developmental changes CI-8 concentration in hemolymph as measured by ELISA.

The values are expressed in 11 g per mi. The cz strain was used. For abscissa see the legend to Fig. 1-22.

Discussion

B. mori hemolymph has more than 16 kinds of Cis whose expression is controlled by the let genes located on different chromosomes. Possibly, Cis of each gene play distinct physiological roles. Therefore, comparative studies on purification and characterization of different Cis are necessary for the fuller understanding of complicated relationships of the hemolymph CI system.

CI-13', 8 and 3 were examined in this chapter as respective representatives of let-A, Jet-E and Iet-D gene members, respectively. The cz strain used for the purification has an advantage in that is homologous for these three loci and does not produce any allelic counterparts probably with similar biochemical properties. The characteristics of these Cis revealed in the present chapter are summarized in Table 1-6.

Table. 1-6. Summary of properties of the purified Cis-3, 8 and 13'.

Cl-3 Cl-8 Cl-13'

Gene lct-E lct-D let-A

Molecular size 40,000 42,000 14,000

pi 5.5 5.0 4.4

Oligosaccharide N-linked type

Type Serpin Serpin Kunitz

Thermostability <55°C <55°C -1oooc

pH stability 5-9 6-11 3-11

Inhibition

to a-chymotrypsin + + +

to DJ _{protease* + +}

*DJ protease =digestive juice protease

CI-13' was purified by only two column chromatographies (DEAE­

Sepharose and Toyopearl HW55) and had a low molecular weight and an acidic pl. This was considered to be a Kunitz type in hi bi tor. CI -13' was very stable for heat treatment and to wide range of pHs like other Kunitz type inhibitors CI-1,2and 13 (Fujii etal., 1989;Deng, 1990; Dengetal., 1990).

Therefore, heating or keeping under extreme pH conditions is probably efficient for the isolation of CI -13' and other Kunitz type inhibitors. The characteristics of CI -13' were very similar to those of its allelic counterpart, CI-13 (Fujii et al., 1989); e.g., the molecular size of CI-13 was also 14,000 (Fujii et al., 1989). eDNA of CI-13 has been isolated from the library of fat body (Yakiyama et al., unpublished data); the deduced amino acid sequence indicated that CI -13 has a molecular weight of about 7 ,000, much smaller than the size obtained by SDS-PAGE. The correct size of low-molecular-weight protein is often difficult to be measured by its relative mobility, and further studies about CI-13' will be needed. SCI-III isolated from the hemolymph of B. mori by Sasaki ( 1978) was similar to CI -13 and 13' (in particular to the former) with a molecular weight of 7,000 and a pi of 4.0. However, the

amino acid sequence of SCI-III (Sasaki, 1984) slightly differs from those of Cls-13 and 13' (Shinohara, 1993; Yakiyama et al., unpublished data). SCI-III has not been identified in the 400 stock strains kept in Kyushu University.

This is probably because a size and pi of SCI-III is similar to those of CI-13' and both polypeptides may by co-migrated (the migration rate of CI-13 is as fast as the dye marker in native PAGE). Thus it is not obvious at present whether SCI-III is controlled by the let-A gene or not.

Also previously purified and characterized were CI-1 and 2 (Deng e t al., 1990), which are controlled by the Iet-H gene, closely linked with let-A.

The properties of CI -1 and 2 were very similar to those of CI -13', except that CI-1 and 2 are neutral proteins (Deng et al., 1990). In addition to these, the

hemolymph of B. mori has basic Cis named SCI-I and II (Sasaki, 1978), each comprising 62 amino acid residues with differences from each other at only two positions (Sasaki,

1988).

There were other basic Cis called b

1

and b2 (Shinohara, 1993), whose N -terminal amino acid sequences showed a high homology to CI-13, although the gene loci for these novel Cis are uncertain.

All hitherto-known low-molecular-weight Kunitz type Cis of B. mori showed homology in amino acid sequence to each other, suggesting that these Cis would be evolved by duplication from a common ancestor gene as

considered to occur in the ovoinhibitor gene (Kato, 1988). As to the latter, the putative duplicated domains link like ^a praying beads to make a multi-headed inhibitor and to exert multi-functions in one molecules. In contrast, Cis in B.

mori hemolymph are present in multiple isoforms each of which would have a specific role, e.g., judged from the different specificities against various

proteases.

Cis-3 and 8 (serpin type inhibitors, controlled by Jet-E and Ict-D

genes, respectively) were purified by essentially the similar methods to each other, and these Cis are comparable to each other in molecular weight, pi and thermostability (Table 1-6). However, CI-3 and 8 indicated quite different profiles in chromatography, especially in that with ConA-Agarose. As to the pH stability, CI-3 lost its activity under alkaline conditions, while CI-8 was stable even at pH 11. Serpin-type inhibitors similar to Cls-3 and 8 in characteristics including molecular weight (about

40,000)

and pi (about 5) have also been purified from B. mori hemolymph by other groups [e.g., sw­

Achy (Sasaki and Kobayashi, 1984; Narumi et al., 1993) and inhibitor-d (Eguchi and Shomoto, 1985); the similarity of latter to the present ones is partial]. However, CI-8 reacted with ConA. The presence of sugar moiety has not been described in previous reports of the silkworm protease inhibitors except for Cis-6 and 7, which are allelic counterparts of CI-8 (Shinohara,

weight (one carbohydrate chain per molecule). Its structure was proposed to be a complex type with a biantennary chain.

At present, a eDNA for CI-8 was isolated and analyzed for its nucleotide

sequence (Tomoi et al., 1995). The deduced amino acid sequence of CI-8 was highly similar to that of sw-Achy (Narumi et al., 1993 ), with a homology of about 95%. These are inferred to be allelic components. However, sw-Achy has no oligosaccharide, and its N-terminal amino acid sequence (Narumi e t al., 1993) is not completely identical to all Cis controlled by let-D. Without

further information discussion with respect to the relationships between these Cis is impossible.

In the present immunoblotting analysis, the antiserum raised against CI-8

reacted only to Cls-6, 7 and 8, i.e., the members of the let-D gene. Similarly, the antiserum for CI-13 showed cross-reaction only to CI-13', a co-dominant allelic counterpart of CI-13. These facts suggested that the Iet-D gene has a rather remote phylogenetic relationship with the let-A and Jet-E genes. A serpin type CI from B. mori hemolymph named sw-Achy II was quite different from sw-Achy (another serpin type CI) in amino acid sequence (Sasaki, 1991). However, the sequence of sw-Achy II from the N-terminus to residue 336 is completely identical to the corresponding region of the

silkworm anti-trypsin sw-AT, whereas the degree of similarity between sw­

Achy II and sw-AT from residue 337 to the C-terminus was only 46% (Takagi et al., 1990). Sasaki ( 1991) proposed a mosaic structure for sw-Achy II and sw-AT (with two independent genes for two proteins or one gene for the conserved region and two genes for the variable region). FromM. sexta, 38 eDNA clones for serpins were isolated; these were identical in sequence except for a region encoding the C-terminal 40 to 45 residues, which includes the reactive center (Jiang eta!., 1994 ). There are 11 variants of the catalytic center region, each encoded by a different version of the 9th exon in the serpin

gene. The evolution of this insect serpin gene has resulted from duplication and sequence divergence of only the exon encoding the reactive site.

Alternative pre-mRNA splic,ing then generates variety of proteases, using the same protein framework joined to different reactive site region cassettes (Jiang et al., 1994 ). Thus, an antiserum for alaserpin, one of the serpin group of M. sexta, reacted with humoral 10 proteins (Kanost, 1990b ). In B. mori, homologous members of Cis (e.g. Cis-1 and 2, sw-Achy II and sw-AT) may possibly be produced via similar processes to those of M. sexta.

TheN-terminal amino acid sequence of CI-8 (Tomoi et al., 1995) and that of CI-3 (Shinohara, 1993) both differ from the corresponding region of sw-Achy II. Thus Cis-8 and 3 might have not trudged the same way of evolution as the sw-Achy II gene. Eguchi's group did not find Cis of the let­

E. type. Taken together, the author proposes that CI-3 is a novel

chymotrypsin inhibitor, which no previous reports about hemolymph protease inhibitors of B. mori have described.

Cis-3, 8 and 13' all exhibited anti-chymotryptic activity. Cis-3 and 8 inhibited the proteolytic activity of digestive juice, although CI-13' did not.

Detailed inhibitory mechanisms revealed in some protease inhibitors (Huber and Carrell, 1989; Sasaki, 1985) are yet to be examined in the present Cis.

The question arises as to whether the differential inhibitory spectra might have relevance to the type of Cis (Kunitz type or serpin type) and to the way of contribution to the proteolytic control systems in vivo. As a clue to the study on these problems, CI-8 was analyzed for its behavior in the following two chapters.

The overall patterns for the developmental changes in the relative

activity of hemolymph Cis as measured in toto, and those in the concentration of CI-8 as determined by ELISA, were in agreement with previously reported fluctuation patterns of hemolymph inhibitory activity (Eguchi et al., 1986b;

Aratake ^et al., 1990). The increase and decrease of hemolymph Cis making a sharp peak at the onset of spinning might reflect some drastic physiological switchover occurring at this point of development. Interestingly, the patterns were roughly reminiscent of the mode of changes of SPs 1 and 2. Examination of allatectomized larvae (Eguchi ^et al., 1986b) suggested that the synthesis of protease inhibitors accelerates when juvenile hormone is absent in the

hemolymph of early 5th instar larvae, a feature similar to that known for SPs (Izumi ^et al., 1984 )^. This problem wi 11 be discussed again in the next chapter.

In summary, studies in this chapter established the principal procedures for purification of representative Cis and clarified their basic properties.

Subsequently, these will be utilized to approach the dynamic aspects of Cis, via the mode of transportation of CI-8 in r^el^ati^on to development and

metamorphosis.