THE JOURNAL OF EXPERIMENTAL ZOOLOGY 256242-254 (1990)

Concurrent Protein Synthesis Is Required for In Vivo Chitin Synthesis in Postmolt Blue Cralbs MICHAEL N. HORST Biochemistry Section, Division of Basic Science, School of Medicine, Mercer University, Macon, Georgia 31207 ABSTRACT

Chitin synthesis in crustaceans involves the deposition of a protein-polysaccharide complex at the apical surface of epithelial cells which secrete the cuticle or exoskeleton. The present study involves a n examination of in vivo incorporation of radiolabeled amino acids and amino sugars into the cuticle of postmolt blue crabs, Callinectes sapitlus. Rates of incorporation of both 3H leucine and 3H threonine were linear with respect to time of incubation. Incorporation of 3H threonine into the endocuticle was inhibited greater than 90% in the presence of the protein synthesis inhibitor, puromycin. Linear incorporation of 14C glucosandne into the cuticle was also demonstrated; a significant improvement of radiolabeling was achieved by using 14C-N-acetylglucosamine as the labeled precursor. Incorporation of 3H-N-acetylglucosamine into the cuticle of postmolt blue crabs was inhibited 89%by puromycin, indicating that concurrent protein synthesis is required for the deposition of chitin in the blue crab. Autoradiographic analysis of control vs. puromycin-treated crabs indicates that puromycin totally blocks labeling of the new endocuticle with 3H glucosamine. These results are consistent with the notion that crustacean chitin is synthesized a s a protein-polysaccharide complex. Analysis of the postmolt and intermolt blue crab cuticle indicates that the exoskeleton contains about 60% protein and 40% chitin. The predominant amino acids are arginine, glutamic acid, alanine, aspartic acid, and threonj ne.

Chitin is a polysaccharide polymer consisting of N-acetyl-D-glucosamine residues joined by beta 1 t o 4 linkages and is found in a variety of organisms, including protozoans, yeast, fungi, and nearly all invertebrates (Muzzarelli, '77). Arthropod chitin consists of a protein-polysaccharide complex which constitutes the bulk of the organic material in the exoskeleton; the protein components of the exoskeleton or cuticle may be crosslinked by phenolic tanning and tyrosine crosslinking, especially in insects (Hunt, '70). The biosynthesis of chitin is thought t o occur in the epithelial cells which lie just underneath the cuticle. Details of precursor assembly and incorporation into an extracellular site are not available. The hypothesis which we have been testing is that crustacean chitin synthesis begins in the rough endoplasmic reticulum with a structural protein which is glycosylated in a co-translational process, forming a primer which is then extended in the Golgi apparatus via the enzyme chitin synthetase. The completed protein-polysaccharide complex is then deposited into the extracellular space beneath the cuticle and covalently attached via transglycosylases, transpeptidases, and phenolic tanning. Implicit in this proposed pathway is the notion that concurrent protein synthesis is 0 1990 WILEY-LISS, INC.

required for the biosynthesis of crustacean chitin.

It has been shown that protein synthesis inhibitors block chitin synthesis in larval stages of the shrimp, PenZeus uannarnez (Horst, '89a). The present studies were initiated to determine the effects of protein synthesis inhibitors on a defined cuticle layer, the endocuticle, in an adult crustacean. Accordingly, freshly molted adult blue crabs were used as the model system t o examine the requiremenis for protein synthesis as a requisite t o chitin synthesis.

MATERIALS AND METHODS Chemicals 2,3,4,5 3H L-leucine (120 Ci/mmol), 6-3H Dglucosamine (40Ci/mmol), and 6-3H N-acetyl-Dglucosamine (32 Ci/mmol) were purchased from ICN Radiochemicals, Irvine, CA. D-(l-14C) glucosamine (7.0 mCi/mmol), UDP-(6-3H) Nacetyl-D-glucosamine (20.4 Ci/mmol), and Nacetyl-D-(l-14C) glucosamine (57.2 mCi/mmol) were obtained from New England Nuclear Corp., Boston, MA. Artificial seawater was prepared from reagent-grade chemicals as described Received Septembe. 22, 1989; revision accepted March 30, 1990.

CHITIN SYNTHESIS IN THE BLUE CRAB

(Cavanaugh, '75). Puromycin and cycloheximide were from Sigma Chemical Co., St. Louis, MO. HCl and OPA reagents were from Pierce Chemical Co., Rockford, IL. Sodium dodecyl sulfate (SDS) was a product of BDH and was obtained from Gallard Schlesinger. Omnifix was obtained from Xenetics Biomedical Inc., Tustin, CA. Ilford K.5D emulsion was obtained from Polysciences, Warrington, PA. Animals Premolt blue crabs were collected by hand from tidal creeks near the Florida State University Marine Laboratory, Turkey Point, FL, or were purchased from Mr. Steve Nordbrook, East Point, FL. Animals were held at 25°C in plastic dishpans containing aerated seawater and fed calf liver ad libitum; seawater was changed daily. Freshly molted animals were held for 4 t o 12 h prior to use in experiments. At this time, the animals synthesize one shell layer, the endocuticle (Skinner, '85). Extraction of radiolabeled tissues Radioisotopes such as 3H glucosamine may be incorporated into a variety of lipid-linked intermediates as well as soluble proteins and insoluble macromolecular chitin. The extraction procedure described below permits measurement of incorporation into each fraction. Thus, lipid intermediates are extracted with chloroform and methanol mixtures; soluble proteins are extracted with low salt buffer; certain insoluble proteins are extracted with sodium dodecyl sulfate (SDS). The final insoluble residue is defined as macromolecular chitin (Horst, '83, '89a).

243

ume of water (2 x ). Samples of the organic-soluble fraction were dried; after addition of scintillation fluid, radioactivity in the samples was measured. The material insoluble in chloroform :methanol was extracted with 2% (w/v) sodium dodecyl sulfate (SDS) in 0.1 M borate buffer, pH 9.0. After boiling for 3 min, samples were centrifuged and the soluble material was dialyzed and radioactivity measured. As an alternative t o the SDS extraction, samples were extracted with 9.5 M urea, pH 10.3, and centrifuged, and the soluble material was dialyzed. The two extraction procedures gave equivalent results. The final insoluble product was washed with water, extracted with acetone, and dried, and radioactivity was measured. Prior to scintillation counting, the residue was dispersed in water by use of a Heat Systems probe sonicator. Treatment of animals with antibiotics Freshly molted blue crabs (12-15 cm carapace width; 150-180 g) were held for 4 to 12 h prior to the start of the experiment. Crabs were injected with puromycin or cycloheximide (9 mg/injection) hourly for 5 h and radiolabeled precursor (3H threonine, 3H GlcNAc or 3H GlcN) was injected with the third dose of antibiotic; all injections were made into the dorsal sinus. All injection sites were sealed with a drop of 1%agarose to prevent leakage of injected material. After injection, animals were induced to autotomize a walking leg at various time points; legs were extracted and radioactivity was measured.

Tissue distribution of radioactivity in postmolt crabs Radiolabeling of postmolt blue crabs The distribution of radiolabeled precursors in Radioisotopes (24 x lo6 DPM in a volume of 5 control and puromycin-treated crabs was deterto 25 ~ 1were ) injected into the dorsal sinus. The mined as follows. Postmolt animals were injected injection site was sealed with 1%agarose. After as described above; after the incorporation period, various periods of time, animals were induced t o the dorsal and lateral portions of the carapace autotomize a walking leg to determine incorpora- were removed and freed of epithelial tissue. The tion of radioisotope. Autotomized legs were freed claws were freed of tissue and the cuticle was of tissue; the cuticle and epithelium were each saved. After the hepatopancreas was removed homogenized in artificial seawater buffered with and saved, the remaining tissue and appendages 20 mM HEPES buffer, pH 7.8. After centrifuga- were removed from the ventral carapace. Each tion (5,5OOg/30 min) the buffer-soluble fraction tissue was individually homogenized by using a was dialyzed against deionized water and radioac- Polytron homogenizer in 50-100 ml HBS buffer: tivity incorporated into the non-dialyzable mate- 0.05 M HEPES buffer, pH 7.1, containing 0.4 M rial was measured. The buffer-insoluble residue NaCl and 0.02 mM PMSF. The homogenate of the was resuspended in ch1oroform:methanol (2 :l), muscle and hepatopancreas was dialyzed. The homogenized, and centrifuged as before. The or- cleaned cuticle was homogenized in HBS; after ganic-soluble fraction was removed from the in- filtration through glass fiber filter paper, the solusoluble residue and extracted with an equal vol- ble material was dialyzed. The cuticle residue was

M.N. HORST

244

homogenized in 9.5 M urea, pH 10.3 (Horst, '89a); after filtration, the urea-soluble fraction was dialyzed. The final residue was washed with deionized water (3 x ) and with acetone and allowed to dry at room temperature. The epithelial cells were homogenized in HBS and the soluble material was dialyzed. The cell residue was extracted with chloroform :methanol (2 : 1);after filtration, the C :M (2 :1)-soluble material was removed, mixed with an equal volume of water, and centrifuged. The organic layer plus interface was saved and rendered uniphasic by addition of methanol, and an aliquot was placed in a scintillation vial and dried under a hood. The final epithelial cell residue was dried. Radioactivity in all fractions was determined by liquid scintillation counting for 5 or 10 min. Portions of solid samples were weighed and then dispersed by sonication in water; following addition of scintillation fluid, Cab-0-Sil, a thixotropic agent, was added to each vial to form a gel. For Table 1, three control and three puromycintreated crabs were used; for Table 2, five control and five puromycin-treated crabs were used. The data are presented as percent distribution -+ the standard error of the mean (SEMI. Tabular data were analyzed for significance by an independent Student T test; * indicates a P value of less than 0.05. Experiments depicted in Figures 1 through 4 were conducted three times and a representative figure is presented.

Autorad iographic procedures Postmolt crabs were injected as described above except that 25 pCi 3H GlcN was used per animal. After incorporation for 4 h, sections of dorsal carapace (1cm square) were removed from control and puromycin-treated animals. Tissue was fixed in Omnifix for 1 11 at room temperature followed by 15 h at 4°C. The tissue was then embedded in paraffin and 4 pni sections were placed on slides, deparaffinized, dried, and dipped with Ilford K.5D emulsion. After drying, the slides were stored dark at 4°C for 2-3 weeks; after development, slides were stained with hematoxylin and eosin or methylene blue and coverslipped. Photography was carried out by using a Nikon Biophot microscope. In vitro chitin synthesis assay The procedure for in vitro chitin synthesis assay is based upor. the standard assay for Artemia (Horst, '81, '83). Briefly, homogenates of blue crab tissue (5-20 mg protein) were incubated in 50 mM HEPES buffer, pH 7.1, containing 0.4 M NaC1, 30 mM MgC12, and 3p1 UDP-3H GlcNAc (75,000 dpm). Samples (0.5 ml) were incubated at 37°C for 1 h ilnd then extracted with chloroform :methanol (2 : 1). The insoluble material was extracted with chloroform :methanol :water (10 :10 :3); each organic soluble fraction was extracted with an equal volume of water, dried,

TABLE 1. Distribution of 3H threonine in control and puromycin-treated postmolt blue crabs' % of total

Fraction Blood Hepatopancreas Epithelial cells-buffer Epithelial cells-CH1: MeOH (2: 1) Epithelial cells-residue Cuticle-buffer Cuticle-urea Cuticle-residue Muscle tissue Tank Specific activity of cuticle residue (cpmimg)

Control radioactivity

S.E.M. Puromycin-treated radioactivity

0.92 1.75 2.46 0.021 0.32 10.71 4.51 2.23 37.21 40.31

f f t t f f f

0.47 0.001 0.08 1.78 0.82 0.08 3.09 1.95

0.62 f ,095 0.55 2 0.11" 0.80 f 0.13" 0.033 f 0.003* 0.42 f 0.03 5.09 t 0.35* 3.85 t 0.21 1.85 f 0.31 53.51 -t 2.97" 33.67 f 2.90

72.16

-t

20.8

12.67 t 2.00*

f 0.24 f 0.48 -t

'Three control postmolt crabs were injected with 25 million dpm 3H-L-threonine. After 3 h, tissues were dissected, extracted, and counted as described under Me ;hods. Three postmolt blue crabs were injected with puromycin (9 mgi0.5 ml seawater) at the fc,llowingtimes: 2 h and 1 h before isotope injection; concurrent with 3H threonine (25 million dpm); 1 h and 2 h after injection. After 3 h, tissue was dissected and extracted and radioactiv ty was determined as described under Methods. Tabular data were analyzed for significan:e by an independent Student T-test; * indicates a P value of less than 0.05.

CHITIN SYNTHESIS IN THE BLUE CRAB

245

TABLE 2 . Distribution of 3H N-acetylglucosamine in control and puromycin-treated postmolt blue crabs' % of total t S.E.M.

Fraction

Puromycin-treated radioactivity

Control radioactivity

Blood Hepatopancreas Epithelial Cells-buffer Epithelial Cells-CH1: MeOH (2: 1) Epithelial Cells-residue Cuticle-buffer Cuticle-urea Cuticle-residue Muscle tissue Tank Specific activity of cuticle residue (cpm/mg)

0.06 t .15 4.89 t .71 1.81 t .16 0.013 t .0016 0.407 ? .034 5.66 2 .53 1.40 t .24 32.87 i~ 3.88 24.29 & 1.62 28.06 t 3.06 1,001.34

?

0.53 2.55 2.97 0.016 0.36 7.91 2.15 6.30 37.54 39.77

54.92

t .093 ?

1.13*

f .53 t ,003 f .07 t .50* f .38 t 2.46* t 4.00* t 6.02

114.48 f 27.43*

'Five postmolt blue crabs were injected with 24 million dpm 3H-N-acetylglucosamine.After 3 h, tissues were dissected, extracted, and counted as described under Methods. Five postmolt blue crabs were injected with puromycin (9 mgl0.5 ml seawater) a t the following times: 2 h and 1h before isotope injection; concurrent with injection of 3H GlcNAc (24 million dpm); 1h and 2 h after injection. After 3 h, tissues were dissected and extracted and radioactivity was measured as described under Methods. Tabular data were analyzed for significance by a n independent Student T-test; * indicates a P value of less than 0.05.

and counted in scintillation vials. The C:M:W (10: 10:3) residue was extracted with boiling 2% SDS (3 min) and centrifuged. The radioactivity in the SDS-soluble and -insoluble fractions was measured. It is not known whether these assay conditions are optimal for blue crabs. However, it is not likely that the results would be substantially different if the buffer were modified t o more closely approximate the crab's natural environment.

Analptical methods

For amino sugar determinations, preliminary experiments indicated that maximal yield for hexosamine could be attained by hydrolyzing chitin samples in 6 M HC1 at 105°C for 12 h. After hydrolysis, samples were dried in vacuo and total hexosamine content was estimated by the method of Levvy and McAllan ('59). Neutral sugar was

60 1

C

0

1 0

1

2

3

0

1

2

3

0

1

2

3

Incubation Time (h) Fig. 1. Time course of 3H-L-leucine incorporation into the epidermis of postmolt blue crabs. The animal was injected with 3H leucine, legs were autotomized at various times; the epidermal tissue was removed and extracted as described

under Methods. Radioactivity in the extracted residue is shown in Panel A, buffer-soluble radioactivity in Panel B, and urea-soluble radioactivity in Panel C. Data for Panel A are reported as cpmimg product.

M.N. HORST

246

20

0

'0 F

X

z4

E

2 10

0

2

0 0

1

2

3

0 0

1

2

3

0

1

2

3

Incubation Time (h) Fig. 2. Time course of 3H-L-leucine incorporation into the cuticle of postmolt blue crabs. The animal was injected with 3H leucine; legs were autotomized at various times; the cuticle was freed of tissue and extracted as described under

Methods. Radioactivity in the extracted cuticle residue is shown in Panel A, buffer-soluble radioactivity in Panel B, and urea-soluble radioactivity in Panel C. The units reported in panel A are cpmirng product.

quantitated by the phenol sulfuric acid procedure (Dubois et al., '56). For amino acid analysis, samples (3-5 mg) were hydrolyzed under N2 in 6 M HC1 at 105°C for 20 h, dried over NaOH-pellets in vacuo, redissolved in water, and lyophilized. Samples were redissolved in water (2 ml) and aliquots (10-30 p1) were analyzed by HPLC on a n Altex 5 pm Ultrosphere ODS column (10 cm) using pre-column derivitization with o-phthalaldehyde (OPA) reagent (Bhown et al., '83). Values for aspartic acid and asparagine are reported as Asx, while glutamic acid and glutamine are given as Glx, since one does not know the percentage amide of each pair. Total protein was measured by a fluorescamine procedure using bovine serum albumin as standard (Horst, '81). HPLC analysis of GlcNAc and chitin oligosaccharides was carried out as described by Reyes et al. ('86). Briefly, samples (20-40 ~ 1in) 0.1 M potassium phosphate buffer, pH 6.3, were chromatographed on a Hewlett Packard NH2-column (10 pm; 4.6 x 200 mm). After application of the sample, the column was eluted isocratically with acetonitrile :water (70:30) at a flow rate of 1.5 ml/min. The effluent was monitored at 215 nm.

was released, indicating that the majority of the radioactivity in this fraction consisted of chitin oligosaccharides. Since cuticular chitin is highly crosslinked in crustaceans, the urea-insoluble material was acid solubilized and precipitated with 95% ethanol as described by Berger and Reynolds ('58). Treatment of such samples with Streptomyces chitinase showed that 80 to 89% of the 14C GlcN was solubilized, indicating that the UR radioactivity is in chitin. The material solubilized from the urea residue by chitinase digestion was analyzed by HPLC on a 10 pm NHZ column by the method of Reyes et al. ('86); 80% of the material chrom&ographed as chitobiose while 20% was identified as GlcNAc. During such experiments, greater than 98% of the radioactivity loaded on the HPLC column was recovered. Calibration of the HPLC column was achieved by using di-, tri-, and tetrasaccharides of chitin obtained by digest ion of reacetylated chitosan with wheat germ endochitinase (Molano et al., '79).

Characterization of radiolabeled products After injection of postmolt blue crabs with 14C glucosamine, radiolabeled products were extracted with 9.5 M urea, pH 10.3. Both the ureasoluble product (US) and the urea-insoluble residue (UR) were digested with chitinase from Streptomyces griseus (Molano et al., '77); 86% of the 14C glucosamine in the urea-soluble product

RESULTS Distribution of injected 14C-GlcN us. '4C-GlcN~ic in postmolt blue crabs When the tissue distribution of 14Cglucosamine in postmolt blue crabs was examined, the majority of the incoi-porated radioisotope (19%) was found in the exoskeleton residue after extraction with various solvents. The specific activity of the cuticle residue was 610 cpm/mg. Epithelial tissue contained the next highest level of I4C-GlcN which was eauallv divided between soluble and

CHITIN SYNTHESIS IN THE BLUE CRAB

-.C 0

c

e

600

n

247

1

2000

g.-r> c

.e

1000

m

.-0 U

U

0

I

.-ca, c

0

B

04 0

F

g

60

120

I80

Time of Incubation (min)

v

r .-c > .-c u

0 .U

U 0

-.-

120

60

180

Fig. 4. Effect of puromycin on incorporation of 3H-GlcN into the epidermis of postmolt blue crabs. Samples of epithelial tissue were removed from extirpated legs, homogenized in buffer, and centrifuged the soluble material was dialyzed and radioactivity was measured. Control, 0-0; puromycin-treated, 0- - -0).

C

a,

c

2

n

600

F 400 0

v

.-.-r> c

200

c

m

a 120

60

Time of Incubation (min)

Fig. 3. Effect of puromycin on incorporation of 3Hglucosamine into the cuticle of postmolt blue crabs. Animals were injected with 0.5 ml seawater plus or minus puromycin (9 mgiinjection) as described under Methods. After three injections, 3H glucosamine (12 x lo6 cpm) was injected, legs were extirpated at various times and extracted, and radioactivity was determined. A: Incorporation of radioactivity into the buffer-soluble fraction of the cuticle. B: Incorporation of 3H GlcN into the SDS-borate-soluble material. C: Incorporation of radioactivity into the SDS-borate-insoluble residue. Control, 0-0; puromycin-treated, -

+- -+.

insoluble fractions. Other tissues contained only traces of radioisotope (data not shown). Studies on the incorporation of radiolabeled GlcNAc were predicated on the fact that blue crabs and other arthropods resorb part of their old cuticle as GlcNAc, after digestion of chitin with

extracellular chitinases and chitobiase. Approximately 30% of the injected radioactivity was recovered in the cuticle residue; 6% of the radioactivity was obtained in the buffer-soluble fraction of the cuticle. A small amount of the total radiolabel was found in the blood (4%). The results indicate that labeled GlcNAc is incorporated to a much greater extent into the cuticle (two- to threefold) than is GlcN.

Incorporation of 3H leucine into postmolt blue crabs For studies on protein synthesis in postmolt crabs, 15 h postmolt animals were injected with 3H leucine and incorporation into cuticle and epidermal fractions was monitored. As shown in Figure 1, there is a biphasic incorporation of 3H leucine into the epidermal fraction; the majority of radiolabel is extracted with buffer, while a smaller amount is solubilized with 9.5 M urea. However, a significant amount of radioactivity is insoluble in both buffer and urea. An initial peak of labeling was observed at 30-60 min followed by a second phase of incorporation at 3 h (Fig. 1A). Incorporation of 3H leucine into the postmolt cuticle is linear in the case of the buffer- and ureasoluble fractions (Fig. 2). The specific activity of the buffer-soluble fraction of the cuticle decreases

248

M.N. HORST

after the initial peak at 1h. This may imply cross- duced to autotoniize a walking leg at various linking or turnover of the newly synthesized times; cuticle was freed of tissue and extracted as cuticular protein is occurring. described under Methods. The results of such an experiment are shown in Figure 3. Treatment of blue crabs with puromycin caused a 90% reducTissue distribution of 3H threonine in tion in labeling o Ithe buffer-soluble protein fracpostmolt blue crabs tion with 3H GlcN in comparison to the control Data presented in Figure 2 demonstrate that (Fig. 3A). Labeling in the presence of puromycin radioactive amino acids are incorporated into the was maximal at 30 min and then remained at the cuticle and epidermis of the postmolt blue crab. In same level for the next'2.5 h. Labeling in the SDSorder to determine the total tissue distribution of borate-soluble fraction was inhibited only 30% at such a labeled amino acid, postmolt blue crabs 30 min (Fig. 3B:, but after 3 h, inhibition had were injected with 3H threonine and tissue was reached 80%. Presumably, the activity observed sampled after 3 h incorporation. The majority of at 30 min represents transfer of 3H GlcN t o endogthe injected 3H threonine was found in the muscle enous peptides o r chitin acceptors. The effect of tissue and the holding water, while the epidermis puromycin on incorporation of 3H GlcN into the and cuticle accounted for nearly 20% of the radio- insoluble chitin .residue is shown in Figure 3C; activity (Table 1). The specific activity of the cuti- puromycin inhibits glucosamine incorporation by cle residue was 72 cpm/mg. 90% over the entire time course of the experiment. This resull, supports the data presented in Table 3; i.e., concurrent protein synthesis is reEffect of puromycin on protein synthesis in quired for the polymerization of crustacean chipostmolt blue crabs As a control experiment, it was necessary t o tin. When the incorporation of 3H GlcN into the show that puromycin blocks protein synthesis in epidermis of blue crabs in the presence and abpostmolt blue crabs. Accordingly, crabs were insence of puromycin was examined, rapid incorpojected with puromycin and 3H threonine as deration into the cclntrol buffer-soluble fraction was scribed under Methods. After 3 h incorporation, observed; purom ycin treatment blocked incorpotissues were collected and analyzed. As shown in ration into this fraction by 80% (Fig. 4).The level Table 1,reduced incorporation of injected radioacof radiolabeling in the epidermal buffer-soluble tivity was observed; the specific activity of the cuticle residue was 12.7 cpm/mg, indicating 83% fraction remained constant over the next 2.5 h. inhibition of protein synthesis in this fraction. Thus incorporation of 3H GlcN by the epithelial Thus, antibiotics such as puromycin can block cells is blocked by puromycin. Since turnover of enzymes necessary for chitin cuticular protein synthesis in the blue crab. synthesis may have occurred during the treatment with protein synthesis inhibitors, experiEffect of protein synthesis inhibitors on ments were carried out t o determine the specific chitin synthesis in the postmolt blue crab activities of several marker enzymes as well as Following injection of 3H GlcNAc the postmolt the enzymes of the chitin synthesis pathway. The blue crabs, the results presented in Table 2 were chitin synthesis assay measures incorporation of obtained. Nearly 33% of the 3H GlcNAc was 3H GlcNAc into two classes of lipid-linked interfound in the cuticle residue; specific activity of mediates and into two types of chitin products, the cuticle residue was 1,001 cpm/mg. When one urea (or SDS) soluble, the other an insoluble the crabs were injected with puromycin and 3H residue (Horst, '81, '83). As shown in Table 3, GlcNAc, the specific activity of the cuticle residue there was little difference between the control was 114 cpm/mg (Table 2). Thus, puromycin in- values and those obtained with cycloheximidehibits incorporation of 3H GlcNAc into the cuticle treated preparations indicating that the decrease residue by 89%. These results indicate that con- in chitin synthesis observed in the presence of incurrent protein synthesis is required for the depo- hibitors was not due to lack of enzymes. Rather, sition of crustacean chitin. the results indicate that the marker and chitin To examine the kinetics of chitin synthesis in synthesis enzymes are active and suggest that the the presence of protein synthesis inhibitors, blue reason for decreased synthesis is most likely due crabs were injected with 3H GlcN plus or minus t o lack of a protein acceptor which becomes glypuromycin (9 mg/injection). Animals were in- cosylated and serves as a primer molecule.

CHITIN SYNTHESIS IN THE BLUE CRAB

249

TABLE 3. Effect of cycloheximide treatment on activity of marker enzymes and in uitro chitin synthesis in postmolt blue crabs'

Enzyme (nmol/lO min/mg) Phosphodiesterase I Alkaline phosphatase 5' nucleotidase Chitin synthesis assay (cpm/h/mg) Chloroform :methanol (2 : 1) soluble Chloroform :methanol: water (10: 10:3) SDSlborate soluble SDSlborate residue (chitin)

Control activity

Cycloheximide-treated activity

7.8 7.5 7.2

6.0 7.6 6.6

186

45 1

140 41

155 56

22

19

'Two groups of three blue crabs were held for 12 h after molting injected with 0.5 ml seawater with or without 9 mghnjection cycloheximide every hour for 5 h. After waiting 1 additional h , the animals were sacrificed and epithelial tissues from each group were pooled and homogenized with Hepes-buffered seawater. Various marker enzymes were assayed (Horst, '81) in both control and cycloheximide-treated animals. Results are expressed as nanomoles substrate consumed per 10 min per milligram protein. Chitin synthesis activity is reported as cpm incorporated per hour per milligram protein.

Autoradiographic analysis of puromycin treatment When postmolt blue crabs were injected with 3H GlcNAc in the presence or absence of puromycin, a drastic change in incorporation of the radiolabel was observed by autoradiography. As shown in Figure 5A, intense labeling of newly synthesized endocuticle was observed in the control specimen. Labeling was also observed just below the apical membrane of the epithelial cells and in the perinuclear area (inset), consistent with the notion that the Golgi apparatus may be involved with 3H GlcNAc incorporation. When examined by electron microscopy, this region of the epithelial cell is rich in endoplasmic reticulum and Golgi (Horst, manuscript in preparation). When postmolt blue crabs were treated with puromycin plus 3H GlcNAc, there was little detectable incorporation of 3H GlcNAc into the endocuticle adjacent to the epithelial cells (Fig. 5B). Small amounts of radiolabel were observed in the apical portions of the epithelial cells. Finally, the morphology and thickness of the endocuticle appeared to be altered by puromycin treatment. These autoradiographic results are in good agreement with the biochemical finding that puromycin treatment of postmolt blue crabs blocks incorporation of radiolabeled GlcNAc into the cut(ic1e. Compositional analysis of blue crab cuticle Analysis of intermolt and pastmolt samples before and after decalcification indicated the cuticle

contains about 44% calcium salts (ash) and 56% organic material (data not shown). After decalcification of the cuticle with EDTA, the data shown in Table 4 were obtained. In the decalcified cuticle, amino acids comprise 59% to 62% of the weight, while chitin (as N-acetylglucosamine) contributes about 30%. Traces of neutral sugars are also found. Preliminary data indicate the presence of galactose with traces of fucose and mannose. Amino acid analysis of blue crab cuticle Intermolt and 15 h postmolt blue crab cuticle samples were subjected to amino acid analysis as described under Methods. As show in Table 5 , the predominant amino acids are aspartic acid, glutamic acid, serine, arginine, and glycine. The analysis of 15 h postmolt vs. intermolt crabs showed that the two amino acids used in the TABLE 4 . Comwosition of blue crab cuticle' wlw%

Component Hexosamine Protein Neutral sugar Ash

Intermolt cuticle

12 h postmolt cuticle

72 24 0.8 5

30 59 1.9 4

'Samples of cuticle were treated with EDTA to remove calcium salts before analysis. Ash weight was calculated after combustion of a sample in a muffle furnace at 1,lOo"C for 15 h. Analysis of the neutral sugars in both cuticles by gas-liquid chromatography revealed the presence of galactose plus traces of mannose and ficose.

250

M.N. HORST

Figure 5.

CHITIN SYNTHESIS IN THE BLUE CRAB

25 1

glycine or alanine; similar results have been reported for other crustacean cuticles by Austin et moles/1.000 residues al. ('81). Thus far, identification of the linkage Amino acid Intermolt 15 h Dostmolt between chitin and protein in the arthropod exoAsx 87 116 skeleton has eluded investigators. Glx 118 122 In the present study, some data on glucosamine Ser 14 21 incorporation were obtained by inducing animals His 89 100 t o autotomize appendages. Some question could GlY 71 84 be raised as to the use of this procedure. As reArg 156 132 Thr 80 58 viewed by Skinner et al. ('851, autotomization of Ala 102 91 legs has been used by numerous investigators, TYr 50 34 primarily t o accelerate ecdysis in intermolt or Met 18 3 premolt crabs. In such studies, animals missing 58 Val 74 as many as eight walking legs survive at a rate Phe 36 53 Ile 29 23 equal t o that of normal animals (Skinner and Leu 44 50 Graham, '70). In the present investigation, the LYS 31 51 remaining legs of experimental crabs were still 'Samples of cuticle were extracted with buffer, EDTA, urea, and moving just prior to removal, indicating an intact acetone as described under Methods. The residue was hydrolyzed and blood supply and nerve function. subjected to amino acid anlysis. The 14C-glucosamine-labeled products formed in the postmolt blue crab have been identified as radiolabeling studies, leucine and threonine, are chitin by digestion with Streptomyces chitinase. both present at significant levels in the postmolt The soluble products of the chitinase digestion cuticle. Interestingly, marked changes in the lev- were identified as chitobiose and GlcNAc when els of certain amino acids were detected, namely analyzed by HPLC. Since these were the only methionine, lysine, serine, and phenylalanine. In radiolabeled components detected, the results both intermolt and postmolt cuticle, amino acids suggest that the chitinase preparation was free of countributed about 60% of the total weight of the endoglycosidase H activity. In vivo studies of protein synthesis in the blue decalcified shell. crab show that radiolabeled amino acids are inDISCUSSION corporated into the epithelial cells and the cuticle The synthesis of the crustacean cuticle involves of the blue crab. Thus, 3H leucine is incorporated formation of both protein and polysaccharide com- into soluble proteins of the cuticle and epidermis ponents within the underlying epithelial cells. In in a linear manner with respect t o time of incubathe present study, the composition of the cuticle tion. There is also a rapid incorporation of 3H in the blue crab, Cullinectes supidus, was found t o leucine into the cuticle residue during the first be different during the intermolt and postmolt hour of incubation. However, there is a subsestages in terms of total protein and chitin content. quent loss of radioactivity from the cuticle over The amino acid composition of the cuticle at both the next 2 h. This result may represent restrucstages was also examined; notable differences in turing of the cuticle or turnover of some rapidly the levels of certain amino acids were detected, labeled component(s). Incorporation of radiolai.e., Met, Lys, Ser, and Phe. The major amino beled leucine into the insoluble fraction of the epiacids found in the blue crab cuticle were aspartic dermis was biphasic, with a rapid labeling obacid, glutamic acid, serine, arginine, and either served within 30 min and a second peak of labeling at 3 h. Taken together, these data suggest a vigorous incorporation of labeled amino Fig. 5. Autoradiography of control (A) vs. puromycin- acids into the cuticle of the postmolt blue crab. It was shown that protein synthesis inhibitors treated (B) blue crabs following injection of 3H GlcNAc. Crabs were injected, and the tissue was fixed and autoradiographed such as puromycin block incorporation of labeled (see Methods). A: Labeling of the new endocuticle is observed amino acids into products greater than 95%. Dur(arrow); the inset shows labeling in the perinuclear region of ing this study, no attempt wae .de t o determine epithelial cells (arrowheads). Following puromycin treatthe minimal dose of antibiotic required to observe ment, no labeling of the endocuticle is observed (B) in comparison to the control tissue. ep, epicuticle; ex, exocuticle; en, the effect. endocuticle. Original magnification, x 400. Incorporation of radiolabeled D-glucosamine TABLE 5 . Amino acid anlysis of blue crab cuticle'

M.N. HORST

252

observed. The major finding of this report is the observation that inhibition of protein synthesis with antibiotics such as puromycin blocks chitin synthesis in the blue crab. Since such experiments involve a 2 h preincubation period, it might be argued that all chitin synthetase in the epidermal cells had turned over during this period. Based upon data presented in Table 3, this appears t o be an unlikely possibility. Previous investigators have stated that the crustacean cuticle is deposited in a region at the apices of the epithelial cells, where chitin microfibrils and matrix proteins are assembled (Stevenson, '85). While this may be true for certain cuticular proteins, the results of the present study indicate that polymerization of chitin is dependent upon concurrent protein synthesis. This implies that chitin is initia1l:y synthesized as a glycoprotein complex. Several questions regarding the effect of puromycin treatment may be raised. First, why

and N-acetylglucosamine into the postmolt cuticle was examined t o determine if one compound might be a more suitable precursor than the other. The results of these studies indicate that GlcNAc is incorporated two- to threefold more readily than GlcN. The reason for this probably lies in the fact that glucosamine may be phosphorylated to yield glucosamine-6-phosphate and shunted into the glycolytic pathway rather than leading to the formation of UDP-GlcNAc. On the other hand, phosphorylation of GlcNAc produces GlcNAc-6-phosphate which leads directly to nucleotide sugar synthesis; there would appear t o be little or no deacetylation of this substrate by the crab epidermis. Incorporation of labeled glucosamine into the blue crab cuticle was linear with respect to time of incubation up to at least 3 h. Significantly, the majority of the radiolabeled glucosamine was incorporated into the cuticle residue, although some labeling of the soluble proteins of the cuticle was

Dolichol

CDP

UMP

HIPTP -

Dolichol P

-I

- GlcNac

NUDP

-

- -

Dolichol P P GlcNAc

4

/ UDP - GlcNac

UDP

- -

Dolichol P P- (GlcNac)

I

-

Dolichol P

Chito Protein

Chitin Protein (Cuticle) Complex

I

(GICNAC)n

1

GlcNAc

Phenolic tanning

IJDP-

ein (GlcNAc) n+x

Exocytosis

=

Microfibril formation

=

Transpeptidase

,)

A

Transglycosylase

Fig. 6. Proposed biosynthetic pathway for crustacean chitin synthesis. Lipid-linked steps are a t the top; chitin synthetase is at the lower left; and extraccllular crosslinking reactions are at the bottom. See text for details.

CHITIN SYNTHESIS IN THE BLUE CRAB

doesn't transfer of 3H glucosamine t o existing chitin oligosaccharides in the urea-soluble or residue fractions continue in the presence of puromycin? One possible explanation is that the 3 h preincubation period used in these experiments may allow sufficient time for completion of all existing oligosaccharide chains; thus there are no remaining oligosaccharides which may serve as acceptors. A second explanation is that the products are sequestered in a different cellular (or extracellular) compartment where they are no longer accessible to the glycosyltransferases. A second question regarding the puromycin experiment involves the data reported in Table 3. If puromycin treatment blocks the synthesis of a putative chitin primer, then why is in vitro synthesis of chitin observed after homogenization of puromycin-treated animals? The most likely explanation is that homogenization of epithelial cells makes a sequestered primer available for chitin synthetase. As summarized in Figure 6, the synthesis of crustacean chitin involves both intracellular and extracellular events. Synthesis is initiated by activation of dolichol; the crustacean dolichol kinase requires CTP and is calcium dependent (Horst, '89c). The initial glycosylation of dolichol is carried out by the GlcNAc-1-P transferase (Horst, '90). Further additions of GlcNAc occur until the lipid-linked oligosaccharide reaches three to eight GlcNAc residues (Horst, '83). Transfer of oligosaccharides from lipid to protein via the enzyme oligosaccharyl transferase (Horst, '89b) yields a chitoprotein. Such chitoproteins appear to serve as acceptors for chitin synthetase (Horst, '81, '89d), producing macromolecular chitin. Exocytosis of this material is followed by transglycosylase, transpeptidase, and protein tanning reactions which crosslink the material into the growing cuticle. Future studies will be directed at understanding these extracellular crosslinking reactions and their control by crustacean epithelial cells.

ACKNOWLEDGMENTS This work was supported by grant GM-30952 from the National Institutes of Health. I thank Ms. Lissa Jackson, Ms. Ely Klar, and Mr. J . West Hightower for technical assistance. Ms. Ginger Sanders typed the manuscript. Portions of this work were conducted at the Ed Ball Marine Labo-

253

ratory, Florida State University, Carrabelle, FL; my belated thanks to Dr. Robert Harriss for providing laboratory space. The encouragement and support of the late R.J. Winzler is gratefully acknowledged.

LITERATURE CITED Austin, P., C. Brine, J. Castle, and J. Zikakis (1981) Chitin. New facets of research. Science, 212:749-753. Berger, L., and D. Reynolds (1958) The chitinase system of a strain of Streptomyces griseus. Biochim. Biophys. Acta, 29:522-534. Bhown, A., T. Cornelius, and J. Bennett (1983) A rapid separation method for precolumn derivitized amino acids using reversed-phase HPLC. J. Liquid Chromatogr., 1 :50-52. Cavanaugh, G.M. (1975) Formulae and Methods VI. The Marine Biological Laboratory, Woods Hole, MA, 84 pp. Dubois, M., K.A., Gilles, J.K. Hamilton, P.A. Revers, and F. Smith (1956) Colorimetric method for determination of sugars and related substances. Anal. Chem., 28(3):350356. Horst, M.N. (1981) The biosynthesis of crustacean chitin by a microsomal enzyme from larval brine shrimp. J. Biol. Chem., 256:1412-1419. Horst, M.N. (1983) The biosynthesis of crustacean chitin. Isolation and characterization of polyprenol-linked intermediates from brine shrimp microsomes. Arch. Biochem. Biophys., 223:254-263. Horst, M.N. (1989a) Association between chitin synthesis and protein synthesis in the shrimp Penaeus vannamei. J. Crustacean Biol., 9:257-265. Horst, M.N. (198913) Glycosylation of exogenous peptide acceptors by larval brine shrimp microsomes. In: Synthetic Peptides. Approaches t o Biological Problems. J.P. Tam and T.E. Kaiser, eds. Alan R. Liss, New York, NY, pp. 51-62. Horst, M.N. (1989~)Dolichol phosphorylation occurs via a CTP-dependent reaction in Artemia larvae. J . Exp. Zool., 252:16-24. Horst, M.N. (19898 Molecular and cellular aspects of chitin synthesis in larval Artemia. In: Cell and Molecular Biology of Artemia Development. A.H. Warner, T. MacRae, and J. Bagshaw, eds. Plenum Press, New York, NY, pp. 59-76. Horst, M.N. (1990) Isolation of a crustacean N-acetyl-Dglucosamine-1-phosphate transferase and its activation by phospholipids. J. Comp. Physiol. [Bl, 159:777-778. Hunt, S. (1970). Protein-Polysaccharide Complexes in Invertebrates. Academic Press, New York, NY, 329 pp. Levvy, G., and A. McAllan (1959) N-acetylation and estimation of hexosamines. Biochem. J., 73:127-132. Molano, J., A. Duran, and E. Cabib (1977) A rapid and sensitive assay for chitinase using tritiated chitin. Anal. Biochem., 83:648-656. Molano, J., I. Polacheck, A. Duran, and E. Cabib (1979) An endochitinase from wheatgerm. Activity on nascent and preformed chitin. J. Biol. Chem., 254:4901-4907. Muzzarelli, R.A.A. (1977) Chitin. Pergamon Press, New York, N.Y., 329 pp. Reyes, F., M. Martinez, J. Calatayud, and R. Lahog (1986) Degradation of cell wall chitin from the fungus Aspergillus nidulans during autolysis. In: Chitin in Nature and Tech-

254

M.N. HORST

nology. R. Muzzarelli, C. Jeuniaux, and G. Gooday, eds. Plenum Press, New York, N.Y., pp. 99-102. Skinner, D. (1985) Molting and regeneration. In: The Biology of the Crustacea. D.E. Bliss and L.H. Mantel, eds. Academic Press, New York, NY, Vol. 9, pp. 43-146. Skinner, D., D. Graham, C. Holland, D. Mykles, C. Soumoff, and L. Yamaoka (1985) Control of molting in Crustacea. In:

Factors in Adult GrDwth. A. Wenner, ed. A. Balkema Publishers, Boston, MA, pp. 3-14. Skinner, D., and D. Graham (1970) Molting in land crabs: Stimulation by leg removal. Science, 169:383-385. Stevenson, J.R. (1985) Dynamics of the integument. In: Biology of the Crustaceit. D.E. Bliss and L.H. Mantel, eds. Academic Press, New Y ork, NY, Vol. 9, pp. 1-42.