Purification, characterization and sequencing of the major β-1,3-glucanase from the midgut of Tenebrio molitor larvae

Insect Biochemistry and Molecular Biology 39 (2009) 861e874

Contents lists available at ScienceDirect

Insect Biochemistry and Molecular Biology journal homepage: www.elsevier.com/locate/ibmb

Purification, characterization and sequencing of the major b-1,3-glucanase from the midgut of Tenebrio molitor larvae
lia Ferreira a, * Fernando A. Genta a, b, Ivan Bragatto a, Walter R. Terra a, Cle a b

~o Paulo, C.P 26077, 05513-970, Sa ~o Paulo, Brazil Departamento de Bioqu
ımica, Instituto de Qu
ımica, Universidade de Sa Instituto Oswaldo Cruz, FIOCRUZ, Rio de Janeiro, Brazil

a r t i c l e i n f o

a b s t r a c t

Article history: Received 18 June 2009 Received in revised form 8 October 2009 Accepted 12 October 2009

The major b-1,3-glucanase from Tenebrio molitor (TLam) was purified to homogeneity (yield, 6%; enrichment, 113 fold; specific activity, 4.4 U/mg). TLam has a molecular weight of 50 kDa and a pH optimum of 6. It is an endoglucanase that hydrolyzes b-1,3-glucans as laminarin and yeast b-1,3-1,6glucan, but is inactive toward other polysaccharides (as unbranched b-1,3-glucans or mixed b-1,3-1,4glucan from cereals) or disaccharides. The enzyme is not inhibited by high substrate concentrations and has low processivity (0.6). TLam has two ionizable groups involved in catalysis, and His, Tyr and Arg residues plus a divalent ion at the active site. A Cys residue important for TLam activity is exposed after laminarin binding. The cDNA coding for this enzyme was cloned and sequenced. It belongs to glycoside hydrolase family 16, and is related to other insect glucanases and glucan-binding proteins. Sequence analysis and homology modeling allowed the identification of some residues (E174, E179, H204, Y304, R127 and R181) at the active site of the enzyme, which may be important for TLam activity. TLam efficiently lyses fungal cells, suggesting a role in making available walls and cell contents to digestion and in protecting the midgut from pathogen infections. Ó 2009 Elsevier Ltd. All rights reserved.

Keywords: b-1,3-glucanase Midgut Laminarinase Coleoptera Fungi digestion

1. Introduction

b-1,3-glucanases are enzymes that specifically hydrolyze b-1,3glycosidic bonds in b-1,3-glucans like laminarin (b-1,3-glucan from the brown algae Laminaria sp), callose (from higher plants) or pachyman (from the fungus Poria cocos). They are also active against the b-1,3-1,6-glucan from yeast and fungi cell walls. They are endo- (glucan-endo-1,3-b-D-glucosidase, E.C.3.2.1.39) or exo(glucan 1,3-b-glucosidase, E.C.3.2.1.58) enzymes. In a more recent classification of glycoside hydrolases based on hydrophobic amino acid clustering and sequence similarities, endo-b-1,3-glucanases from E.C.3.2.1.39 are present in families 16, 17, 55, 64 and 81, and exo-b-1,3-glucanases from E.C.3.2.1.58 are present in families 3, 5, 17 and 55 (Coutinho and Henrissat, 1999).

Abbreviations: TLam, Tenebrio molitor laminarinase; CFUs, colony-forming units; DPC, diethyl pyrocarbonate; EDC, N- (3-dimethylaminopropyl)-N0 -ethylcarbodiimide; EDTA, ethylenediaminetetraacetic acid; GGE, 2-glyceroyl-3-glucosyl ethylglycol mixed acetal; NBS, N-bromosuccinimide; or-laminarin, periodate-oxidized and reduced laminarin; PG, 1 phenylglyoxal; pHMB, 4-(hydroxymercuri) benzoic acid; TEMED, N, N, N0 , N0 - tetramethylethylenediamine; TNM, tetranitromethane. * Corresponding author. Fax: þ55 11 3818 2186. E-mail address: [email protected] (C. Ferreira). 0965-1748/$ e see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.ibmb.2009.10.003

b-1,3-glucanases has been described as abundant in the digestive tract of insects of the orders Collembola, Trichoptera, Dictyoptera, Orthoptera, Isoptera, Coleoptera and Diptera (Terra and Ferreira, 1994). In spite of this, their properties are poorly known. In insects, only two laminarinases from Periplaneta americana (Dictyoptera) saliva (Genta et al., 2003), and one from Abracris flavolineata (Orthoptera) regurgitate (Genta et al., 2007) were purified to homogeneity, and characterized. Some enzymological details are known also for Rhagium inquisitor (Coleoptera) enzymes (Chipoulet and Chararas, 1984). Up to now, no data are available regarding to which glycoside hydrolase family they pertain. As b-glucanases are not present in vertebrates, they can be used as a target for insect control. Tenebrio molitor (Coleoptera) is an important cosmopolitan pest of stored products (Richards and Davies, 1977). The properties and secretion of their b-glycosidases has already been studied (Ferreira et al., 2001, 2002, 2003). Insect digestive laminarinases putatively play a role in hemicellulose (callose or cereal b) digestion in herbivores (Terra and Ferreira, 1994) or in the digestion of b-1,3-1,6-glucans from fungi in detritivores (Genta et al., 2003). In this paper, we describe the purification, characterization and sequencing of T. molitor digestive laminarinase (TLam). The data on specificity, action pattern, chemical modification, protein and cDNA sequencing indicate that TLam is

862

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

a low processive b-1,3 endoglucanase and a member of glycoside hydrolase family 16 with a probable role in dietary fungal digestion.

One unit of enzyme (U) is defined as the amount that hydrolyzes 1 mmol of bonds/min.

2. Material and methods

2.4. T. molitor laminarinase (TLam) purification

2.1. Animals and chemicals

A sample of 2.5 mL of the midgut supernatant was applied onto a 5 mL EconoPac High Q ion-exchange chromatography column (EconoSystem, BioRad, USA) equilibrated with 20 mM imidazole buffer pH 7.0, containing 10 mM CaCl2. After passing 20 mL of the last buffer through the column, elution was accomplished with 0e1 M NaCl gradient in the same buffer (100 mL) plus 20 mL of buffer with 1 M NaCl. The flow was 2 mL/min and fractions of 2 mL were collected. The more active fractions against laminarin were pooled (30e40, Fig. 1A), diluted ten times with the imidazole buffer above and the material was loaded onto a 1 mL Resource Q ionexchange chromatography column (FPLC System, Pharmacia, Sweden) with a 150 mL Superloop (Pharmacia) and a flow of 4 mL/ min. The column was equilibrated and washed (5 mL) with the same buffer. Elution was achieved with 0e1 M NaCl gradient in the same buffer (60 mL) plus 5 mL of buffer with 1 M NaCl. The flow was 1 mL/min and fractions of 0.4 mL were collected. Fractions more active against laminarin (37e43, Fig. 1B) were combined, (NH4)2SO4 was added to attain a concentration of 1 M and the material was loaded onto a 1 mL Resource Phenyl hydrophobicinteraction chromatography column (FPLC) equilibrated with 50 mM MES buffer pH 6 containing 10 mM CaCl2 and 1 M (NH4)2SO4. After passing 5 mL of this buffer through the column, elution was carried out with 1e0 M (NH4)2SO4 gradient (60 mL) and 5 mL of buffer without saline. The flow was 1 mL/min and fractions of 0.4 mL were collected. Fractions more active against laminarin (137e142, Fig. 1C) were pooled and applied onto a HR10/30 Superdex 75 gel filtration column (FPLC), equilibrated with 100 mM citrate-sodium phosphate buffer pH 6 containing 10 mM CaCl2. Proteins were eluted with the same buffer (30 mL), with a flow of 1 mL/min, and fractions of 0.4 mL were collected. Fractions more active against laminarin (22e25, Fig. 1D) were combined and used as pure laminarinase (TLam). In order to determine the molecular weight of TLAM, its elution from Superdex 75 was compared with the elution of the following standards: ribonuclease A (13.7 kDa), soybean trypsin inhibitor (21.5 kDa), ovalbumin (45 kDa), bovine serum albumin (66 kDa) and thyroglobulin (67 kDa).

Stock cultures of T. molitor were maintained under natural photoregime conditions on wheat bran at 24e26  C and 70e75% relative humidity. Fully-grown larvae of both sexes (each weighing about 0.12 g), having midguts full of food, were used. Avicel was purchased from Merck (Darmstadt, Germany) and the other substrates were purchased from Sigma (USA). All chemical substances used were of analytical grade. 2.2. Preparation of samples Larvae were immobilized by placing them on ice, after which they were dissected in cold 342 mM NaCl. For determination of enzymes in gut sections, tissues were homogenized in cold MilliQ water with the aid of a PottereElvehjem homogenizer with 10 strokes and centrifuged at 10,000 g for 10 min at 4  C. The pellets were homogenized in cold MilliQ water using a micro tube homogenizer (Model Z 35, 997-1, Sigma, USA). The homogenates were stored at 20  C until use without noticeable changes in the activities. Both pellets and supernatants were assayed. For laminarinase purification, the midgut was pulled apart and homogenized as before in cold 20 mM imidazole buffer pH 7 containing 10 mM CaCl2 and 10 mM phenylthiocarbamide (PTC). The homogenate was centrifuged as before, the supernatant was filtered through a PVDF membrane with a 0.44 mm pore (Millipore) and used immediately. Wheat bran was homogenized in cold MilliQ water using a homogenizer model Skymsem TAR-02 (Siemsen, Brazil) at 10,000 rpm for 3 cycles of 30 s. The homogenate was sonicated with a Branson Sonifier 250, using three cycles of 30 s each (output 3) with 30 s intervals. The sample, after homogenization as described before, was centrifuged at 10,000 g for 10 min at 4  C. The supernatant was passed through glass wool to be freed of fat. 2.3. Protein determination and hydrolase assays Protein was determined with the silver method of Krystal et al. (1985), using ovalbumin as a standard. For this, samples were dialyzed for 3 h at room temperature against 20,000 volumes of 10 mM Tris, containing 10 mM Na2CO3, 0.75% Tween 20, with pH 10e12. Hydrolase activity was determined by measuring the release of reducing groups (Noelting and Bernfeld, 1948) from 0.25% (w/v) laminarin (from Laminaria digitata), 0.125% (w/v) lichenan (from Cetraria islandica), 0.25% (w/v) carboxymethylcellulose (CMC), 0.25% (w/v) xylan, 0.125% (w/v) b-1,3-1,4-glucan from Hordeum vulgare, 1% (w/v) pachyman (from P. cocos) and 1% (w/v) Avicel (microcrystalline cellulose). The Avicel and pachyman suspensions were maintained under stirring during the whole assaying time. The release of glucose during the assays was determined with the Triseglucoseeoxidase method of Dahlqvist (1968). b-Glucosidase activity was determined following the release of p-nitrophenolate (Terra et al., 1979) from 5 mM p-nitrophenyl b-D-glucoside (NPbGlu), or the release of glucose from 5 mM laminaribiose, gentiobiose or cellobiose. Unless otherwise specified, all substrates were assayed in 50 mM citrate-sodium phosphate pH 6.0 at 30  C under conditions such that activity was proportional to protein concentration and to time. Controls without enzyme or without substrate were included.

2.5. Sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (PAGE) and detection of proteins and enzyme activity in the gel SDS-PAGE was accomplished in gel slabs according to Laemmli (1970), as detailed in Ferreira et al. (2001). Staining for protein was done with the silver method of Blum et al. (1987). Molecular masses were calculated according to Shapiro et al. (1967). The following mass standards were used: lysozyme (14.4 kDa), soybean trypsin inhibitor (21.5 kDa), carbonic anhydrase (31 kDa), ovalbumin (45 kDa), bovine serum albumin (66 kDa) and phosphorylase b (97.4 kDa). In gel assays were done after PAGE in 12% acrylamide gel containing 1% laminarin and the samples were not heated and no mercaptoethanol was added. The other conditions were as in SDSPAGE. After electrophoresis, activity was detected according to a modification of the procedure of Pan et al. (1989) and Trudel et al. (1998), as described in Genta et al. (2003). For this, the gel slab was incubated in 20 mM 2-(N-morpholino) ethanesulfonic acid (MES) buffer pH 6.0 containing 10 mM CaCl2 for 15 min (buffer changes each 5 min). The gel was then incubated in a semi-dry system at 30  C with the same MES buffer. After 15 min, the gel slab was

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

0

20

40

0 60

0.3

0.035

3

-0.005 30

0.2 0.1

-0.005 120

140

0 160

Fractions

U/mL

D

0.3

0.015

40

50

60

0.2 0.1 80

70

Fractions

0.055 0.035

0.4

0.075

0.2

(NH4)2SO4, M

U/mL

2

NaCl, M

0.4

0.025

U/mL

0.6

Fractions

C

1

0.115

0.8

0.055

-0.005

B

1

NaCl, M

U/mL

A 0.085

863

0.025 0.015 0.005 -0.005 10

20

30

40

Fractions

Fig. 1. Purification of Tenebrio molitor midgut b-1,3-glucanase. Activity was determined with laminarin. (A) A sample of the soluble fraction of midgut contents was loaded onto an Econopac High Q ion-exchange chromatography column equilibrated with 20 mM imidazole buffer pH 7, 10 mM CaCl2 and eluted with 0e1 M NaCl gradient in the same buffer. One peak of laminarinase activity (Fractions 30e40) was collected, diluted ten fold and submitted (B) to an anion-exchange chromatography in Resource Q column in the buffer described above. After elution with a NaCl gradient (0e1 M) in the same buffer, the more active fractions (37e43) were pooled and after adding ammonium sulfate to 1 M, the pool was loaded (C) on a hydrophobic-interaction chromatography column (Resource Phenyl) in 50 mM MES buffer pH 6, with 10 mM CaCl2 and 1 M (NH4)2SO4. After elution with a (NH4)2SO4 gradient (1e0 M), the more active fractions (137e142) were pooled and applied (D) onto a gel filtration chromatography column (Superdex 75) in 100 mM citratesodium phosphate buffer pH 6, 10 mM CaCl2. The more active fractions (22e25) were combined and used as pure laminarinase (TLam) afterwards.

dipped in a 0.15% triphenyl tetrazolium chloride solution containing 1 M NaOH and was maintained in a microwave oven (20 s cycles in maximal power) until red bands appear against a clear pink background. 2.6. Effect of substrate concentration and pH on TLam activity The effect of substrate concentration on purified b-glucanase activity was determined by using at least 10 different substrate concentrations. KM and VM values (means and SEM) were obtained by least-squares fitting using the software Enzfitter (Elsevier, Biosoft). To determine TLAM stability in different pHs, enzyme samples were pre-incubated at 30  C for 120 min in different pHs before moving to pH 6 for standard assay. The activities were compared to a control assayed at pH 6 that was taken as 100. TLAM was incubated in the following buffers (25 mM): sodium acetate (pH 3.5e5.5), sodium phosphate (pH 6e8) and TriseHCl (pH 8e9), with ionic strength adjusted to 100 mM with NaCl. The buffer pH was adjusted at the temperature of the assay and the enzyme is stable in all pH values tested. Apparent V (Vapp) and apparent KM (KMapp) were determined at each pH from LineweavereBurk plots. The pKa values of the ionizable groups in the active site were determined from linear plots of KMapp/Vapp versus [Hþ] and of 1/Vapp versus 1/[Hþ] (Segel, 1975) according to the following equations:

h i ! Hþ KMapp KM K 1þ ¼ þ h E2i Vm KE1 Vmapp Hþ !  þ H 1 1 K 1þ ¼ þ  ES2  Vmapp Vm KES1 Hþ Where Ke1 is the ionization constant of the nucleophile and Ke2 of the proton donor.

2.7. Production of periodate-oxidized and reduced laminarin (or-laminarin) In order to obtain periodate-oxidized and reduced laminarin (laminarin with its non-reducing end modified into a glycerol derivative), laminarin was oxidized with periodate and then reduced with borohydride (Read et al., 1996). The concentration of modified laminarin was determined with the phenol-sulfuric procedure (Dubois et al., 1956). The yield of reduced laminarin was 100% and of or-laminarin was 36% (w/w) relative to the initial laminarin concentration. 2.8. Determination of the degree of multiple attack and rate of glucose/reducing group release The degree of multiple attack is the number of catalytic events, after the first, during the lifetime of a particular enzymeesubstrate complex (Robyt and French, 1967). It is calculated by the ratio between the rate of formation of low molecular weight saccharides (soluble in ethanol) and that of long chains (insoluble in ethanol). A sample of 0.4 mU of TLam (25 mL) was added to 25 mL of 2% laminarin in 10 mM MES pH 6.0 and 5 mM CaCl2. At different periods of time, the samples were placed in boiling water for 3 min and then 950 mL of ethanol were added. After standing for 1 h at 20  C, samples were centrifuged at 15,000 g for 10 min at 4  C. Supernatants were dried by vacuum centrifugation and 50 mL of water were added to pellets and dried supernatants, before the determination of reducing groups (Noelting and Bernfeld, 1948) in both samples and glucose (Dahlqvist, 1968) only in the supernatants. Glucose and dextrins give different molar reducing values with dinitrosalicylate reagent (Sengupta et al., 2000) and a correction of the values has to be done. For this, standard curves were prepared with glucose or laminaribiose using dinitrosalicylate reagent (Noelting and Bernfeld, 1948) and with glucose using Triseglucoseeoxidase reagent (Dahlqvist, 1968). In the fraction containing soluble carbohydrates, the amount of glucose determined with

864

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

Triseglucoseeoxidase reagent was used to calculate the correspondent absorbance that would be obtained with dinitrosalicylate reagent. This value was subtracted from the total absorbance obtained with dinitrosalicylate reagent in soluble fractions in order to calculate the reducing power of dextrins.

2.9. Determination of glucose, glycerol and GGE produced from or-laminarin TLam was incubated with or-laminarin. After assay interruption, the products were added to ethanol, centrifuged and the supernatants were lyophilized, as described in 2.8. Analysis of TLam products from or-laminarin was done as described in Genta et al. (2007). Briefly, samples were ressuspended in 100 mL of 100 mM citrate-phosphate buffer pH 5 with or without 50 mU of b-glucosidase (Sigma, G0395), and incubated for 1 h at 37  C. After that, glucose was determined as before and glycerol with the aid of the periodate/chromotropic acid method with some modifications of the method described by Lambert and Neish (1950). Briefly, 100 mL of water, 10 mL of 10 N H2SO4 and 50 mL of 0.1 M sodium periodate pH 5.0 were added to the samples. After mixing and waiting for 30 s, the reaction was interrupted with 50 mL of 1 M sodium arsenite. 1 mL of 2 mg/mL chromotropic acid in 80% (v/v), H2SO4 was added, samples were incubated in the dark, at 100  C for 30 min and finally the absorbance of the samples was read at 570 nm. GGE is composed of a glycerol derivative (from the modified non-reducing terminal of laminarin) b-linked to 3glucoside. GGE is calculated from total glycerol determined after b-glucosidase action minus glycerol found before b-glucosidase action.

2.10. Chemical modification studies TLam chemical inactivation was attempted with pHMB, DPC, TNM, NBS, PG, EDC or EDTA. pHMB and EDC react, respectively, with sulphydryl and carboxylate groups (Carraway and Koshland, 1972). DPC, TNM, NBS and PG react respectively with His (Miles, 1977), Tyr (Riordan and Vallee, 1972), Trp (Spande and Witkop, 1967) and Arg (Takahashi, 1968) side chains. EDTA chelates divalent cations. TLam remaining activity after different times of reaction was measured using 0.25% laminarin as substrate. The inactivation reactions were performed with: 3.75e30 mM EDC plus 100 mM glycine ethyl ester in 200 mM TEMED buffer pH 5; 0.1e10 mM TNM in 100 mM sodium phosphate buffer pH 8; 2e20 mM PG in 100 mM EPPS buffer containing CaCl2, pH 8; 10e100 mM EDTA in 100 mM MES buffer pH 6; 10 mM NBS (higher concentrations cause artifacts) in 50 mM citrate-sodium phosphate buffer pH 6, containing 5 mM CaCl2; and 0.25e1 mM pHMB in 4 mM EPPS buffer pH 8.0 containing 0.5 mM CaCl2. Prior to chemical modifications, the enzyme was dialyzed in the buffer to be used for the modification reaction. Reactions with EDC and TNM were stopped by the addition of sodium citrate (final concentration of 200 mM) and phenol (final concentration of 130 mM), respectively. Other reactions were stopped by a fifteen-fold dilution with 5 mM MES buffer pH 6.0 containing 0.5 mM CaCl2, followed by gel filtration chromatography in HiTrap Desalting column (Pharmacia) with the same buffer. Controls of enzyme activity in the absence of chemical reactants showed that the enzyme is stable in all the conditions used. Reactions with modifiers in presence of different concentrations of laminarin were also done. In this case, pseudo first-order rate constants were used to calculate dissociation constants of the enzymeelaminarin complex, according to the equation below:

ko 1 ¼ K D$ þ 1 ½S ko  k x Where ko and kx are the pseudo first-order rate constants of inactivation in the absence and presence, respectively, of laminarin, and KD is the dissociation constant of the enzymeelaminarin complex. KD can be calculated from a linear plot of ko/(ko-kx) against 1/[S]. When the modifier used was DPC, the remaining activity (Ra, %) after incubations stabilizes with time. The remaining percentage of native enzyme (Re) was calculated using the ratio between activities of modified and non-modified enzyme (R) and the remaining activity. The expression used for these calculations was Re ¼ (RaR)/(1R), and Re data were used to calculate the correspondent kobs values. 2.11. Determination of reaction order with respect to the chemical reactants The reaction orders of modification were determined from a plot of log kobs (the observed first-order rate constant for enzyme inactivation) against log of the reactant concentration. This type of plot should give a straight line with a slope equal to n, the apparent number of molecules reacting with each active site of the enzyme, to give an inactive enzymeereactant complex (Segel, 1975). 2.12. Cell lytic activity of TLam An aliquot of 60 mM TLam was incubated at 30  C with 450 colony-forming units (CFUs)/mL of Saccharomyces cerevisae S14 cells in 20 mM citrate-sodium phosphate buffer pH 6.0 containing 10 mM CaCl2. After different periods of time, samples were plated in YPDA medium (1% yeast extract, 1% peptone, 1% dextrose, 2% Agar) for counting the number of colonies after overnight incubation at 30  C. Controls performed without the enzyme showed that the cells are stable in the conditions used. 2.13. Cloning and sequencing of the cDNA that codes for TLAM Total RNA was extracted from midgut epithelium of T. molitor larvae with Trizol following the instructions of the manufacturer, Invitrogen, which are based on Chomczynski and Sacchi (1987), and sent to Stratagene (La Jolla, CA), in order to construct a cDNA library. At Stratagene the mRNAs were isolated, divided into two equal samples and used in cDNA synthesis with a poly-T and a random primer. Finally, the two cDNA pools were mixed (1:1) and non-directionally inserted in the vector l ZAPII. The library titer is 1.5.1010 pfu/mL. Laminarinase cDNA was obtained by PCR amplification of T. molitor library using a degenerate primer coding for a consensus sequence of glycoside hydrolases from family 16 (5'GGA TTT GGC CAG C(A/T)A T(C/T)T GG 30 ) and primers T3 or T7. The PCR reactions was done using TAQ Polimerase (Invitrogen, USA) (5 units) in 20 mM TriseHCl buffer, pH 8.4, with 50 mM KCl, 0.2 mM dNTP, and 4 mM MgCl2. The amplification was reached using 35 cycles at the following conditions: 45 s at 94  C, 30 s at 55  C and 90 s at 72  C. The products of these PCR reactions, with molecular sizes of 1000 and 500 bp for T3 and T7 reactions, respectively, were cloned in pGEM-T Easy Vector (Promega, USA) and sequenced. DNA sequencing was performed with the DNA kit Big Dye Terminator Cycle sequencing (PE Applied Biosystems). Each clone was sequenced in both DNA strands. The electropherograms of the sequenced clones were automatically processed for base calling, low quality detection and vector trimming, and assembled using the algorithm PhredePhrap (http://www.phrap.org/phredphrapconsed. html) (Ewin et al., 1998; Ewin and Green, 1998). The quality of the complete assembled laminarinase was above 50.

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

2.14. Sequence analysis and structural modeling The coding sequence of TLam obtained in 2.13 was analyzed with the algorithms compute pI/Mw tool (Gasteiger et al., 2005), signal IP (Emanuelsson et al., 2007), NetCGlyc (Julenius, 2007), NetOGlyc (Julenius et al., 2005) and NetNGlyc (Gupta et al., 2004) present at the EXPASY Proteomics Server (http://expasy.org/). Selected amino acid sequences of glycoside hydrolase family 16 were aligned with TLam using the ClustalW Multiple Alignment tool in BioEdit Sequence Aligment Editor (http://www.mbio.ncsu. edu/BioEdit/BioEdit.html). Phylogenetic analysis of these aligned sequences was done with SeqBoot, Protdist and Neighbor-Joining algorithms at PHYLIP package (Felsenstein, 2005; http://evolution. genetics.washington.edu/phylip.html). Regular bootstrap was done with 100 replicates, and Protdist was used with JoneseTayloreThornton matrix. Neighbor-Joining method was applied randomizing sequence order (jumble 5) and the consensus tree was built with the Consense program and drawn with Drawgram. Homology modeling was performed using the servers Phyre (Kelley and Sternberg, 2009), Swiss-Model (Arnold et al., 2006), 3DJigsaw (Bates et al., 2001), CPH Models (Lund et al., 2002), ESyPred3D (Lambert et al., 2002), Geno3D (Combet et al., 2002) and LOOPP (Teodorescu et al., 2004). The quality of the models generated was assessed by MOLPROBITY (Davis et al., 2007), PROCHECK (Laskowski et al., 1993), SSM (Krissinel and Henrick, 2004) and PROSA-web (Wiederstein and Sippl, 2007), and the results were analyzed with PDB Viewer (Guex and Peitsch, 1997) and PYMOL (DeLano Scientific LLC). 2.15. Semi-quantitative RT-PCR Larvae immobilized on ice were dissected in cold 342 mM NaCl with gloves, sterile forceps and glassware previously treated with diethyl pyrocarbonate. Total RNA was extracted from the gut epithelium or whole body minus gut of T. molitor larvae with Trizol following the instructions of the manufacturer, Invitrogen, which are based on Chomczynski and Sacchi (1987). Eventual genomic DNA was removed using Dnase I (Invitrogen). The total RNA was used to synthesize the corresponding cDNA with the aid of a reverse transcriptase present in the kit Superscript (Invitrogen). The resulting cDNAs were used as templates for amplifying sequences by PCR with primers specific for TLam. The primers used were: 50 - GTG AGA TTG ACA TAA TGG -30 (forward) and 50 - ATC GAG AAG CTG ATG TC -30 (reverse). PCR reaction was realized using TAQ DNA polymerase (Invitrogen) in reaction buffer containing 1.5 mM MgCl2. PCR conditions were: 35 cycles of 1 min at 94  C (denaturation), 1 min at 50  C (annealing) and 45 s at 72  C (synthesis). The number of cycles was chosen after several trials so that the amplification was in log phase and resulting in a clearly visible amplified band. 2.16. Sequencing of internal peptides from purified TLam Peptide sequencing was done at the Protein Core Facility of the Columbia University College of Physicians & Surgeons by M. A. Gawinowicz as described below. Gel strips containing purified TLam were transferred to clean tubes and 100 mL 0.01 M DTT/0.1 M TriseHCl, pH 8.5 were added. The tube was placed in a heating block at 55  C for 1e2 h. After cooling to room temperature, the liquid was removed from the tube and replaced with 100 mL 0.015 M iodoacetamide/0.1 M TriseHCl, pH 8.5. The mixture was allowed to react for 30 min in the dark after which the liquid was removed and the gel was washed as described below. Gel strips were washed once with 200 mL 0.05 M TriseHCl, pH 8.5, containing

865

25% (v/v) acetonitrile and twice with 200 mL of the same buffer containing 50% (v/v) acetonitrile for 20 min with shaking. After removing the washes, the gel pieces were dried for 30 min in a vacuum concentrator. TLam in gels were digested by adding 0.06e0.1 mg trypsin (modified, sequencing grade, Roche Molecular Biochemicals) in 13e15 mL 0.025 M TriseHCl, pH 8.5. The tubes were placed in a heating block at 32  C and left overnight. Peptides were extracted twice with 50 mL 50% (v/v) acetonitrile/2% (v/v) trifluoroacetic acid and the combined extracts were evaporated to w10 mL, then brought to 20 mL with 0.1% (v/v) formic acid and analyzed by LC-MS/MS. LC-MS/MS analysis was done on a Micromass Q-Tof hybrid quadrupole/time-of-flight mass spectrometer with a nanoelectrospray source. Capillary voltage was set at 1.8 kV and cone voltage 32 V; collision energy was set according to mass and charge of the ion, from 14 eV to 50 eV. Chromatography was performed on an LC Packings HPLC with a C18 PepMap column using a linear acetonitrile gradient with flow rate of 200 hL/min. Raw data files were processed using the MassLynx ProteinLynx software and pkl files were submitted for searching at http://www. matrixscience.com using the Mascot algorithm. 3. Results 3.1. b-glucanase activity in T. molitor gut and food We measured the b-glucanases present in T. molitor gut and food. Both have activities against laminarin and lichenan (Table 1). The optimum pH for laminarin hydrolysis in gut and food preparations is 6 and 4 (data not shown), respectively, indicating that laminarinase activity is not acquired from the food. Lichenase can have a food origin, since the optimum pH is equal in both samples (data not shown) and the difference in activity can be explained by enzyme concentration due to endo-ectoperitrophic circulation (Terra and Ferreira, 1994; Bolognesi et al., 2008). The distribution of all b-glucanase activities along the gut is similar and predominant in the first half of the midgut (Table 2). 3.2. Laminarinase purification Submitting the soluble fraction of T. molitor midgut to sequential ion-exchange chromatographies, at least four different activities against laminarin are resolved (Fig. 1A and B). The minor activity eluted from High Q column is derived from proteins that do not bind to the column at pH 7 (Fig. 1A), and was not studied further. The major activity eluted from High Q column was separated in three activity peaks after chromatography in a Resource Q column (Fig. 1B). Peaks 2 and 3 are much more active against b-glucosides as p-nitrophenyl-b-glucoside and laminaribiose than against laminarin (data not shown), and are probably T. molitor b-glycosidases. These enzymes have a significant activity against laminarin (Ferreira et al., 2001). Peak 1 from Fig. 1B is more active against laminarin than glycosides (data not shown). Hydrophobic interaction chromatography of this material, followed by gel filtration result in only one active peak against laminarin (Fig. 1C and D), without detectable

Table 1 b-Glucanase activities present in identical masses (wet weight) of wheat bran and Tenebrio molitor guts. Substrate

Wheat Bran, mU/20 mg

T. molitor mU/animal

Laminarin Lichenan

3.1  0.2 1.9  0.3

46  1 16  1

Results are means and SEM based on determinations carried out in four different preparations obtained from 10 g wheat bran or 10 animals each.

866

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

Table 2 Hydrolases and protein present in different gut sites of Tenebrio molitor larvae. Substrate

Laminarin Lichenan Protein, mg

Foregut

3 (50) 3 (41) 20

Midgut

Hindgut

Anterior

Posterior

80 (55) 79 (31) 670

16 (42) 17 (24) 180

1 (3) 2 (4) 130

Results are relative activities displayed as percentage of the sum of activities found in the different sections of the gut and specific activities (in parentheses) displayed as mU/mg protein. Figures are means based on determinations carried out in four different preparations obtained from ten insects each. SEM were found to be 5e20% of the means.

activity against glycosides (data not shown). After SDS-PAGE (Fig. 2), only one polypeptide with 50 kDa can be seen (TLam); this molecular mass is the same obtained from the gel filtration chromatography through Superdex 75 column (data not shown). TLam has the same migration in PAGE than the major laminarinase activity present in T. molitor midgut soluble fraction (Fig. 2), indicating that it is the only or most important activity against laminarin in this organ. The recovery and enrichment of laminarinase activity during the purification steps are shown in Table 3. 3.3. TLam properties Purified TLam has no activity against xylan, dextran, lichenan, CMC, avicel (microcrystalline cellulose), cellobiose, laminaribiose, gentiobiose, NPbGlu, pachyman (insoluble b-1,3-glucan) and H. vulgare b-glucan (b-1,3; 1,4-glucan). It hydrolyzes only b-1,3glucans as laminarin and Saccharomices cerevisiae b-1,3-1,6-glucan. The kinetic parameters of TLam acting on laminarin are presented in Table 4. TLam activity is not affected by treatment of reducing ends of laminarin with sodium borohydride, suggesting that TLam does not recognize the reducing end of the substrate. Furthermore, TLam activity decreases 50% by using as substrate laminarin modified at both ends by oxidation with periodate followed by reduction with borohydride. The laminarin used here has two non-reducing ends, and it contains one single branch one-glucose residue long per molecule. This means that TLam may recognize the non-reducing end of the substrate and/or the single glucose residue present at the only branch, that is also affected by periodate treatment. Forty percent of the products formed by TLam acting on laminarin are soluble (shown to be glucose, Table 4) and 60% insoluble

Table 3 Purification of the major laminarinase (TLam) from Tenebrio molitor larval midgut. Soluble fraction from 100 midguts was used as starting material. Fraction

Activity mU

Protein mg

Specific activity mU/mg

Yield (%)

Purification factor

Soluble fraction of midgut High Q eluate Resource Q eluate Resource phenyl eluate Superdex 75 eluate

4600 2530 644 276 276

118 46 11.3 0.2 0.063

39 55 57 1351 4400

100 55 14 6 6

1 1.4 1.5 35 113

in cold ethanol, meaning they are oligosaccharides with more than 10 residues long. These data suggest that TLam is an endoglucanase with low processivity. Based on the ratio between ethanol-soluble and insoluble products, TLam has a multiple attack degree of 0.6 (Table 4), with the soluble products generated being only or mainly glucose. Periodate-oxidation and borohydride reduction of laminarin does not change the ratio of products or the percentage of glucose generated by TLam (Table 4). Thus, the glucose generated by TLam is a product of multiple attacks, because the ends of this substrate were destroyed. Acting on Oxidized and reduced Laminarin (Or-Lam), TLam does not produce glycerol or GGE (data not shown), indicating that TLam does not recognize the non-reducing end of the substrate and does not perform exo-glucanasic activity. Nevertheless, TLam activity decreases 50% by using as substrate Or-Lam. This raises the possibility that the single glucose residue present at the single laminarin branch is important for polysaccharide recognition and binding, as TLam has a higher KM toward Or-Lam than upon laminarin, although the kcat values are similar for both substrates (Table 4). So, the lower activity observed with Or-Lam as a substrate may result from a decreased affinity to the polysaccharide due to the destruction of the sugar in the chain branch. Recognition of laminarin branch is also supported by the fact that TLam has no activity against the linear insoluble pachyman (data not shown), but hydrolyzes the insoluble and highly branched b-1,3-1,6-glucan from Saccharomices cerevisiae 1.7  0.1-fold faster than laminarin (concentration of both substrates: 0.25%). 3.4. Important groups for TLam activity TLam ionizable groups that are important for catalysis have pKa values of 5.8  0.1 and 6.6  0.1 in the free-enzyme (Fig. 3A) and 4.7  0.1 and 7.9  0.1 in the enzymeesubstrate complex (Fig. 3B). The values in free-enzyme are only estimates, since they are only one pH unit apart (Segel, 1975). TLam is unaffected by NBS, indicating that no Trp important to its activity is modified. Otherwise, TLam loses activity in the presence of several compounds detailed below. The maximal inactivation attained at each case was restrained by the stability of the enzyme at the reaction conditions without modifier and by the maximal concentration of modifier that causes no loss in modification specificity. PG reacts with an Arg residue in TLam according to pseudo firstorder kinetics (Fig. 4) with kobs ¼ 27.6  0.6 M2 min1 and Table 4 Kinetic parameters of TLam action against laminarin or periodate-oxidized and reduced laminarin (Or-Lam).

Fig. 2. Electrophoresis in 12% (w/v) polyacrylamide gel slabs. Lane 1, SDS-PAGE of purified TLam. Lanes 2, 3 and 4, PAGE containing 1% (w/v) laminarin. Lane 2, midgut soluble fraction; lanes 3 and 4, purified TLam. Lanes 1 and 4 were silver-stained for protein detection, and lanes 2 and 3 were treated with triphenyl tetrazolium chloride for revealing reducing groups formed by laminarinase activity.

Substrate KM, g/L

kcat, s1

kcat/KM Multiple attack % of products s1(g/L)1 degree as glucose

laminarin 1.5  0.1 1.40  0.03 0.97 Or-Lam 4.40  0.03 1.67  0.03 0.38

0.6 0.7

40 40

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

B

1.2

5

0.2 5

4

0

5

6

7

0 5

8

6

7

pH

8

pH

D

0.5

0.08

3

2

1

Vmax

Vmax /K M

C

-1 -1 k 2 (min .M )

Vmax

Vmax /K M

A

867

0

pKa = 6.40 ± 0.03 K2* = 4.71 ± 0.06

5

6

7

pH 0 5

6

7

pH

8

0 3

4

5

6

7

8

pH

Fig. 3. Effect of pH on the kinetic parameters of TLam and DPC-modified TLam (DPCTLam). The points are experimental and the curves are theoretical based on the constants (found by least-squares fitting with Enzfitter software) described in each case. (A) Effect of pH on TLam Vmax/KMapp ratio, pKE1 ¼ 5.8  0.1 and pKE2 ¼ 6.6  0.1; (B) Effect of pH on TLam Vmaxapp., pKES1 ¼ 4.7  0.1 and pKES2 ¼ 7.9  0.1.; (C) Effect of pH on DPC-TLam Vmax/KMapp ratio, pKE1 ¼ 5.6  0.2 and pKE2 ¼ 7.7  0.1; (D) Effect of pH on DPC-modified -TLam Vmaxapp., pKES1 ¼ 3.7  0.1 and pKES2 ¼ 7.7  0.1; DPCmodified TLam and TLam are stable during the assay time in the pH range 3.5e9.

a reaction order of 2.0 (Fig. 4). Reaction order of one indicates that only one modifier molecule and one enzyme residue react to inactivate the enzyme. It has been described that, depending on the condition of the reaction, one or two PG molecules bind to one Arg residue (see Mizohata et al., 2003). This work quantified incorporated PG, and did not determine the order of reaction. Marana et al. (2003) modified a b-glycosidase with PG and obtained a reaction order of 1.7. Nevertheless, the enzyme has only one Arg residue in its active site, since the substitution of Arg 97 by Met or Lys leads to

Fig. 5. Effect of pH on the chemical inactivation of TLam with EDC. The points are experimental data and the curve is theoretical based on the constants described (found by least-squares method with Enzfitter software). K2* Ko~bs - theoretical pH-independent value of the pseudo first-order constant of the inactivation reaction. TLam is stable during the assay time in the pH range 3.5e9.

an enzyme that is not affected by PG. It is possible that in the conditions employed in this paper and in experiments done by Marana et al. (2003), the binding of only one PG molecule is enough to inactivate the enzyme. If the complex of enzyme with one PG molecule is less stable than the complex with two PG molecules, the reaction order may be a number between 1 and 2, provided that on dissociating the enzyme-PG complex results in an active enzyme. Laminarin protects the inactivation of the enzyme by PG with a KD value of 0.11% that is equal to the KM value in the same reaction conditions (0.10%). This KM value was determined in the pH and ionic strength used in the modification reaction. The agreement of KD and KM values clearly indicates that the Arg residue is placed in TLAM active site. Taking into account the effect of EDTA and EDC, as detailed for PG, on TLam activity (maximal modification: EDTA, 60%, EDC, 80%),

Fig. 4. (A) Inactivation of TLam by PG at 30  C. Reaction mixtures contained (>) 2 mM, (,) 5 mM, (6) 10 mM, () 20 mM PG. (B) Determination of reaction order with respect to PG (r ¼ 1.00; slope ¼ 1.99). (C) Inactivation of TLam by PG at 30  C in presence of different concentrations of laminarin. Reaction mixtures contained 20 mM PG without laminarin (>) or with laminarin (,) 0.4%, (6) 0.5%, (B) 1%, (þ) 2%. (D) Replot of values presented in (C) used for calculation of KD of the TLamelaminarin complex using PG modification data. R ¼ 0.993; Y-axis intercept ¼ 0.97; KD ¼ 0.11%.

868

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

% Initial CFUs

100

50

0 0 Fig. 6. Inactivation of TLam by DPC at 30  C. Reaction mixtures contained (>) 0.1 mM, (,) 0.2 mM, (6) 0.5 mM, (B) 1 mM DPC or ( ) 1 mM DPC plus 1% laminarin.

the enzyme has one divalent cation (kobs ¼ 0.058  0.004 M1 min1, reaction order 1.1) and one carboxyl group (kobs ¼ 0.88  0.07 M1 min1, reaction order 1.0) important for its activity. As the modifications are protected by laminarin with binding constants as protector (KD values for EDTA, 0.18%; for EDC, 0.3%) like those as substrate (KM values determined in the conditions used for EDTA and EDC modification are, respectively, 0.18% and 0.3%), it is arguable that both a divalent cation and carboxyl group occur in the active site. The EDC-modified carboxyl group has a pKa value of 6.40  0.03, calculated from chemical modification data (Fig. 5). This figure quite well agrees with the kinetical pKES2 (Fig. 3). Treatment of TLAM with DPC causes a decrease of near 25% in TLAM activity with kobs ¼ 19  6 M1 min1 and a reaction order of 1.2 (Fig. 6). The modified enzyme has the same KM value of the native one (data not shown). The pKa values determined for the modified enzyme are: 5.6  0.2 and 7.7  0.1 in the free-enzyme (Fig. 3C) and 3.7  0.1 and 7.7  0.1 in the enzymeesubstrate complex (Fig. 3D). The pKa of the protonated group (pKa2) is increased relative to that of the native enzyme. TNM modifies (data not shown) a Tyr residue (maximal modification 80%) with

150

Fig. 8. Lysis of S. cerevisiae cells in hypotonic media after incubation with purified TLam. The cells were incubated in 20 mM citrate-sodium phosphate buffer pH 6 containing 10 mM CaCl2 with 60 mM of enzyme for the times indicated. Reaction mixtures contained (>) 42 fmol/CFU, (6) 2.5 fmol/CFU, (,) 1.3 fmol/CFU.

kobs ¼ 0.86  0.05 M1 min1 (reaction order not determined). The His and Tyr residues are probably at the enzyme active site, since laminarin protects the enzyme from inactivation by DPC and TNM, although KD values were not determined. Modification of Cys residue by pHMB reduces enzyme activity in a rate that increases at higher laminarin concentration (Fig. 7). This suggests that substrate binding induces a conformational change in TLam molecule that leads to higher exposure of the Cys residue, located outside the active site. In the absence of laminarin, the inactivation parameters are: kobs ¼ 0.083  0.007 M1 min1, reaction order 1.1. kobs increases in the presence of laminarin (Fig. 7D) and laminarin binding constant can be calculated from the plot in Fig. 7E as 0.45  0.06%. This value is equal to the KM value in the reaction conditions (0.49  0.04%). This agreement strongly suggests that laminarin binding at TLAM active site causes the exposure of the pHMB- modified Cys residue. Summarizing the data, evidence points to the occurrence in the TLam active site of Arg, His, Tyr, a divalent cation, and a carboxyl

96

- 4.4

- 4.8 -3.7

-3.35

D

1.25 1.00 0.75 0.50 0.25 0.00

-3

0

0.5

80

160

minutes

240

320

10

0

80

E

3 2 1

1

[lam],%

log [pHMB]

0

4

1/kobs liq.10-3 (min-1)

log kobs

97

B

kobs.103 (min-1)

-4

98

Remaining activity (%)

99

95

100

min

C 100

A 100 Remaining activity (%)

50

3

6

9

1/[lam],%-1

160

240

320

minutes

Fig. 7. Inactivation of TLam by pHMB at 30  C. (A) Reaction mixtures contained (>) 0.25 mM, (,) 0.5 mM or (6) 1 mM pHMB. (B) Determination of reaction order with respect to pHMB (r ¼ 0.98; slope ¼ 0.92). (C) Effect of laminarin concentration in the chemical inactivation of TLam with pHMB. Reaction mixtures contained (>) 0, (,) 0.1%, (6) 0.2%, ( ) 0.5% or (þ) 1% (w/v) laminarin. (D) Relationship between kobs (pseudo first-order constant of the inactivation reaction) and laminarin concentration in the reaction mixture. (E) Double reciprocal plot of kobsliq (kobs minus kobs in absence of laminarin) and laminarin concentration.

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

869

Fig. 9. Nucleotide and deduced amino acid sequences of Tenebrio molitor laminarinase (TLam). Residues conserved in at least 50% of family 16 insect proteins are with black background, boxed residues are directly involved in catalysis, and glycosylation sites are shaded. Signal peptide is double underlined. Nucleotide sequences corresponding to the primers used in RT-PCR reactions are shaded. Peptides that are underlined and numbered 1e7 correspond to sequences obtained after trypsin hydrolysis and mass spectrometric analysis. The sequence was deposited in the GenBankÔ under the ID ACS36221.

870

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

3.6. Cloning and sequencing of TLam cDNA

Fig. 10. Expression pattern of mRNAs for TLam in different tissues of Tenebrio molitor larvae as shown by semi-quantitative RT-PCR. Equal amounts (1 mg) of total RNA from different tissues were extracted and used as template in RT reactions. The product of this reaction was used in PCR reaction using specific primers for TLam-encoding cDNA and for ribosomal RNA. The product of amplification was analyzed in 1% agarose/TAE gel. The amplified band appears in the tissue when the corresponding mRNA is transcribed. MG, midgut; WBM, whole body minus midgut; RNA, ribosomal RNA. The number of amplification cycles was chosen after several trials so that the amplification was in log phase and resulting in a clearly visible amplified band.

group. The His residue seems to modulate the pKa of the protonated active group and a Cys residue becomes more exposed on laminarin binding. 3.5. Cell lytic activity of TLam Enzymes hydrolyzing b-1,3-glucosyl bonds may affect the b-1,3glucan chains that, with chitin fibers, form the cell wall of fungal cells (Magnelli et al., 2002). At low density, no S. cerevisiae cells remain viable after 15 min in the presence of TLam (Fig. 8), whereas at high cell densities, the cells are stable for the first 10 min, being completely lysed after 150 min of exposure (Fig. 8).

Using a degenerate primer that corresponds to a consensus sequence from glycoside hydrolases from family 16 and a T. molitor midgut cDNA library, we were able to amplify and sequence from a cDNA that codes for a putative b-glucanase precursor (pTLam, Fig. 9). The putative b-glucanase precursor cDNA has a signal peptide (corresponding to at least 21 amino acids) and codes for a mature protein with 359 amino acids, with an estimated molecular mass of 40,117 Da and an isoelectric point of 4.33 (Fig. 9). pTLam amino acid sequence has seven putative glycosylation sites, at residues T4, T8, T9, T13, T17, N188 and T323, the consensus sequence F(K/R/Q)(Y/F)G(K/R/E)X(E/V)(V/I)RAK(L/M)PX(G/A)(D/Q) W(L/I)XP and catalytic amino acids E195 and E200 (Fig. 9). Seven peptides were sequenced by LC-MS/MS from the purified TLam protein, showing complete identity with the sequence coded by the pTLam cDNA (Fig. 9). The data favor the view that the cDNA coding for the putative b-glucanase actually codes for TLam. Semi-quantitative RT-PCR showed that the mRNA coding for TLam is expressed only in the midgut (Fig. 10). The theoretical pI value of the predicted b-glucanase is consistent with the observed TLam binding to the HiTrap Q and Resource Q anionic exchange columns at pH 7 (Fig. 1). Phylogenetic comparisons between some proteins from glycoside hydrolase family 16 and TLam shows a clear separation between a group with insect glucanases (to which TLam pertains) and insect binding proteins (Fig. 12). It is noteworthy that the binding proteins described from Anopheles and Nasutitermes

Fig. 11. Amino acid sequence alignment of selected proteins from the glycoside hydrolase family 16. The sequences were retrieved from GenBankÔ. The listed proteins (without the signal peptide) are respectively from Tenebrio molitor (TLam, FJ864682; Tm BGRP, BAG14263.1 and Tm GNBP, BAC99308.1), Spodoptera frugiperda (Sf GLUC, ABR28478.1), Periplaneta americana (Pa GLUC, ABR28480.1), Nasutitermes comatus (Nc GLUC, AAZ08480.1) and Nocardiopsis sp. (No GLUC, BAE54302.1). Conserved residues in glucanases are with black background, consensus alternatives are shaded. Catalytic region in glucanases is boxed, and catalytic glutamates are marked with an asterisk. BGLUC - beta-glucanase, GNBP - Gramnegative Bacteria Binding Protein, BGRP - Beta-Glucan Recognition Protein.

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

Fig. 12. Neighbor-joining consensus tree (non-rooted) of selected insect protein sequences from glycoside hydrolase family 16. Sequences are from Tenebrio molitor (FJ864682, BAG14263.1 and BAC99308.1), Periplaneta americana (ABR28480.1), Spodoptera frugiperda (ABR28478.1), Helicoverpa armigera (ABU98621.1), Nasutitermes comatus (AAZ08480.1), Anopheles gambiae (ABU80032.1) and Drosophila melanogaster (AAF33849.1, AAF33851.1 and AAF33850.1). Bootstrap values (above 50, 100 replicates) for each branch point are given. Relevant information concerning this phylogenetic analysis is summarized in item 2.15.

871

branch with glucanases. All the sequences in this phylogenetic branch contain the two glutamate catalytic residues, which are absent from the binding proteins sequences (Fig. 11). TLam shows less identity to T. molitor binding proteins than to glucanases from the other insects described above, and also contains the catalytic residues and several amino acid residues conserved at the active site region in insect glucanases (Fig. 11). Homology modeling of TLam sequence resulted in an overall bjellyroll structure typical of glycoside hydrolase family 16 in all servers tested (data not shown). All models obtained were checked with the MOLPROBITY and PROCHECK packages, and only the results from the PHYRE server were of reasonable quality (data not shown). The match with the highest percentage of identity in the secondary structure alignment was the endo-beta-1,3-glucanase of Nocardiopsys (PDB code 2hyk; Fibriansah et al., 2007). In the Phyre server, this sequence resulted in a match with an e value of 9.5E-21 and a model with 100% precision. Analysis of this model with the SSM software revealed an estimated identity between model and target of 31%, with 76% identification of the secondary structure and a RMSD of 0.22  A. Besides that, ProSA-web analysis of the model revealed a structure with an overall quality in the range of the X-ray determined structures (Z-score 3.97), and MOLPROBITY and PROCHECK analysis showed that the structure of the model was of overall reasonable quality, with the exception of some sterical clashes between side-chains. In this respect, the overall structure of TLam probably is a b-jellyroll from glycoside hydrolase family 16 (Fig. 12)

Fig. 13. Homology model of the TLam structure (see details in the Results section) based on the crystal structure of endo-b-1,3-glucanase from Nocardiopsis (PDB code 2HYK). (A) Overview of the b-jellyroll general structure. Some residues putatively essential for TLam activity at the catalytical cleft are in evidence. (B) Detailed view of the active site region. Distances between the catalytic glutamates (E174 and E179) and some residues that are probably essential for TLam activity are shown. (C) and (D) General views of TLam surface. Catalytic glutamates are shown in red and other residues putatively involved in TLam activity are in blue. Figures were prepared with PYMOL. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

872

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

and, besides some loops that were not modeled (see Discussion), the structural position of 70% of its total backbone can be assumed to probably adopt this secondary arrangement Fig. 13. 4. Discussion 4.1. Occurrence, properties and action pattern of TLAM A 50 kDa laminarinase (TLam) was purified from T. molitor midguts. It is not acquired from food, which has a laminarinase activity different from TLam. The enzyme is not secreted by the intestinal microflora, because axenic larvae have laminarinase activity similar to control ones (Genta et al., 2006), being probably secreted by the anterior midgut epithelial cells. The pH optimum of TLam is around 6 in agreement with T. molitor anterior midgut pH value (Terra et al., 1985). TLam may be classified as EC 3.2.1.39, is not inhibited by high substrate concentrations nor is a highly processive enzyme, in contrast with other insect b-glucanases (Genta et al., 2003, 2007). Lytic capacity has been associated with high processivity (Rombouts and Phaff, 1976), but TLam is able to lyses S. cerevisiae cells, even not being remarkably processive. Chemical modifications showed that besides having a carboxyl group (probably from one of the catalytical residues, see Withers, 2001), TLam active site has one His, one Arg and one Tyr residues, plus a divalent cation. These features are present only in b-1,3glucanases from family 16 of glycoside hydrolases (Hahn et al., 1995; Tsai et al., 2005), suggesting that TLam is a member of this family. Recently, a beta-1,3-glucanase from Helicoverpa armigera larvae midgut was cloned and sequenced, but this enzyme apparently is related to immune recognition and does not lyse fungal cells (Pauchet et al., 2009). Besides that, an immune related family 16 glycosyl hydrolase was isolated from Nasutitermes corniger (Bulmer et al., 2009), being important for the control of the entomopathogenic fungi Metarhizium anisopliae. Members of glucoside hydrolase family 16 include two laminarinases from mollusks (Chlamys albidus and C. rosealbus), one from see urchin (Bachman and McClay, 1996) and many b-glucan-binding proteins from animal and insect origin. In the last mentioned proteins the catalytical residues are replaced by others (Yu et al., 2002). In the active site of family 16 enzymes, Tyr residues are hydrogen bonding with or hydrophobic stacking substrate glucose residues, and His and Arg residues form hydrogen bonds with the ring oxygen or hydroxyl groups of particular residues of the substrate (Tsai et al., 2005). Substrate binding changes the pKa values of the proton donor and the nucleophile of TLam. As DPC modification leads to a decrease in pKES1, His residue should stabilize the protonated nucleophile, shifting the optimum pH of TLam to a less acidic value in comparison with other endo-beta-1,3-glucanases from fungi and bacteria (Chang et al., 2009). Analysis of conserved residues in family 16 b-1,3-glucanases, and homology modeling of TLam sequence allowed the identification of the putative residues involved in substrate binding or catalysis in TLam active site (Fig. 12). From the three His residues conserved in b-1,3-glucanases from family 16 (H41, H204, H232), only one H204 is located at the active site cleft (Fig. 12A). In fact, H204 is in a position very close to the known catalytic glutamates (Fig. 12B), being probably the residue modified by DPC. With similar reasoning, we could point Y304, R127 and R181 (see Fig. 12A and B) as the residues that probably are important for TLam activity, being modified by TNM or PG. All the residues mentioned above are probably exposed to the solvent (Fig. 12C and D), being accessible to the chemical modifiers used in this study. Nevertheless, some caution must be taken, because some loops that seem to be typical

of insect b-1,3-glucanases (residues 1e35, 77e117, 262e281, 312e342 in TLam; see Fig. 10), are not present in bacterial b1,3-glucanases (as the Nocardiopsis enzyme that was used as a template) and were not included in the structural model. These regions can be responsible for particular properties of some insect b-1,3-glucanases, as the special recognition of the branched b-1,3-1,6-glucan from yeasts by TLam. In the case of the Cys residue modified by pHMB, only Cys 299 was included in the model, which makes the identification of this amino acid in TLam sequence very tentative. 4.2. TLAM specificity and function TLam is unable to act on b-1,3-1,4-glucans, abundant in cereals and only hydrolyze b-1,3-glucans, present in small amounts in plant phloem. T. molitor diet is rich in fungi and the coordinate secretion of laminarinase and chitinase (Genta et al., 2006) indicates that the role of TLam is the digestion of b-1,3-glucan present in fungal cell wall. The observation that TLam has strong lytic properties and that T. molitor intestinal flora has a low amount of fungi (Genta et al., 2006) favors this view. TLam specificity and pattern of action also point to digestion of fungal material, with the recognition of b-1,6 branchs, a structural feature typical of the highly branched fungal and yeast b-1,3-1,6-glucans. TLam lyses S. cerevisiae cells at levels above 50 CFU/mL (2.5 fmol/ CFU in Fig. 6), that is higher than the fungal cell densities reported for T. molitor midguts (2 CFU/mL, Genta et al., 2006), and within time periods that are in agreement with the passage of food bolus through the insect gut (135 min, Terra et al., 1985). This indicates that fungal cell lysis by TLam can occur at physiological conditions. In this way, the concerted action of TLam, T. molitor midgut chitinase (TmChi), b-glucosidases and b-N-acetyl-glucosaminidase might be responsible for the low densities (Genta et al., 2006) or complete absence of cultivatable fungal cells in most (60%) of T. molitor larvae studied (Genta, F.A., Dillon, R.J., Terra, W.R., Ferreira, C., unpublished data). Detritivores rely on bacterial and/or fungal cells for nutrient acquisition. b-1,3-Glucanases help in making available glucose from fungal cell wall, and in freeing cell contents to other digestive enzymes. The few insects where a highly active gut laminarinase was described are indeed detritivores (Terra and Ferreira, 1994) and its possible that all of them produce their own laminarinase. Entomopathogenic fungi have been described in the gut of insects from several orders (Slaymaker et al., 1998). Nevertheless, the preferred route for fungal infection in insects is the exoskeleton cuticle (Bidochka et al., 1997). The short time of exposure of the peritrophic membrane to dietary fungal cells caused by food movement (135 min in T. molitor larvae) may be an important factor that decreases the possibility of infection (Bidochka et al., 1997). Also, midgut bacterial flora can inhibit fungal development by competition or by production of phenolic antifungal secondary metabolites (Dillon and Dillon, 2004). Furthermore, the presence of highly lytic enzymes secreted by the insect midgut may help in the resistance to gut infections. The wide distribution of laminarinase in the gut of insects may be accounted by their different midgut roles. Experiments of suppression of TLam activity are needed to precisely establish the role and importance of this enzyme for T. molitor larvae. Acknowledgements This work was supported by the Brazilian research agencies FAPESP (Tematico and SMOLBnet programs) and CNPq. F.A. Genta was a post-doctoral fellow of FAPESP and is now a staff member of IOC-FIOCRUZ. I. Bragatto is a graduate fellow from CNPq W.R. Terra

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

References Arnold, K., Bordoli, L., Kopp, J., Schwede, T., 2006. The SWISS-MODEL workspace: a web-based environment for protein structure homology modeling. Bioinformatics 22, 195e201. Bachman, E.S., McClay, D.R., 1996. Molecular cloning of the first metazoan b-1,3 glucanase from eggs of the sea urchin Strongylocentrotus purpuratus. Proc. Natl. Acad. Sci. U.S.A. 93, 6808e6813. Bates, P.A., Kelley, L.A., MacCallum, R.M., Sternberg, M.J.E., 2001. Enhancement of protein modeling by human Intervention in applying the automatic programs 3D-JIGSAW and 3D-PSSM. Proteins: Structure, Funct. Genetics, Suppl. 5, 39e46. Bidochka, M.J., St. Leger, R.J., Roberts, D.W., 1997. Mechanisms of deuteromycete fungal infections in grasshoppers and locusts: an overview. Mem. Entomol. Soc. Can. 171, 213e224. Blum, H., Beir, H., Gross, H.J., 1987. Improved silver staining of plant proteins, RNA and DNA in polyacrilamide gels. Electrophoresis 8, 93e99. Bolognesi, R., Terra, W.R., Ferreira, C., 2008. Peritrophic membrane role in enhancing digestive efficiency: theoretical and experimental models. J. Insect Physiol. 54, 1413e1422. Bulmer, M.S., Bachelet, I., Raman, R., Rosengaus, R.B., Sasisekharan, R., 2009. Targeting an antimicrobial effector function in insect immunity as a pest control strategy. Proc. Natl. Acad. Sci. (Wash.) 106, 12652e12657. Carraway, K.L., Koshland Jr., D.E., 1972. Carbodiimide modification of proteins. Meth. Enzymol. 25, 616e623. Chang, A., Scheer, M., Grote, A., Schomburg, I., Schomburg, D., 2009. BRENDA, AMENDA and FRENDA the enzyme information system: new content and tools in 2009. Nucleic Acids Res. 37, D588eD592. Chipoulet, J.M., Chararas, C., 1984. Purification and partial characterization of a laminarinase from the larvae of Rhagium inquisitor. Comp. Biochem. Physiol. 77B, 699e706. Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinium triocyanateephenolechloroform extraction. Anal. Biochem. 162, 156e159.
age, G., Geourjon, C., 2002. Geno3D: automatic Combet, C., Jambon, M., Dele comparative molecular modeling of protein. Bioinformatics 18, 213e214. Coutinho, P.M., Henrissat, B.1999. Carbohydrate-active Enzymes Server at URL: http://afmb.cnrs-mrs.fr/_cazy/CAZY/index.html Dahlqvist, A., 1968. Assay of intestinal disaccharidases. Anal. Biochem. 22, 99e107. Davis, I.W., Leaver-Fay, A., Chen, V.B., Block, J.N., Kapral, G.J., Wang, X., Murray, L.W., Arendall III, W.B., Snoeyink, J., Richardson, J.S., Richardson, D.C., 2007. MolProbity: all-atom contacts and structure validation for proteins and nucleic acids. Nucleic Acids Res. 35, W375eW383. Dillon, R.J., Dillon, M.V., 2004. The gut bacteria of insects: nonpathogenic interactions. Annu. Rev. Entomol. 49, 71e92. Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F., 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350e356. Emanuelsson, O., Brunak, S., von Heijne, G., Nielsen, H., 2007. Locating proteins in the cell using TargetP, SignalP, and related tools. Nat. Protoc. 2, 953e971. Ewin, B., Green, P., 1998. Base calling of automated sequencer traces using phred. II. Error probabilities. Genome Res. 8, 186e194. Ewin, B., Hillier, L., Wendl, M., Green, P., 1998. Base calling of automated sequencer traces using phred. I. Accuracy assessment. Genome Res. 8, 175e185. Felsenstein, J., 2005. PHYLIP (Phylogeny Inference Package) Version 3.6. Distributed by the author. Department of Genome Sciences, University of Washington, Seattle. Ferreira, A.H.P., Marana, S.R., Terra, W.R., Ferreira, C., 2001. Purification, molecular cloning, and properties of a ß-glycosidase isolated from midgut lumen of Tenebrio molitor (Coleoptera) larvae. Insect Biochem. Mol. Biol. 31, 1065e1076. Ferreira, A.H.P., Ribeiro, A.F., Terra, W.R., Ferreira, C., 2002. Secretion of ß-glycosidase by middle midgut cells and its recycling in the midgut of Tenebrio molitor larvae. J. Insect Physiol. 48, 113e118. Ferreira, A.H.P., Terra, W.R., Ferreira, C., 2003. Characterization of a ß-glycosidase highly active on disaccharides and of a ß-galactosidase from Tenebrio molitor midgut lumen. Insect Biochem. Mol. Biol. 33, 253e265. Fibriansah, G., Masuda, S., Koizumi, N., Nakamura, S., Kumasaka, T., 2007. The 1.3 A crystal structure of a novel endo-b-1,3-glucanase of glycoside hydrolase family 16 from alkaliphilic Nocardiopsis sp. strain F96. Proteins: Struct. Funct. Bioinform. 69, 683e690. Gasteiger, E., Hoogland, C., Gattiker, A., Duvaud, S., Wilkins, M.R., Appel, R.D., Bairoch, A., 2005. In: Walker, John M. (Ed.), The Proteomics Protocols Handbook. Humana Press, New York, pp. 571e607. Genta, F.A., Terra, W.R., Ferreira, C., 2003. Action pattern, specificity, lytic activities, and physiological role of five digestive ß-glucanases isolated from Periplaneta americana. Insect Biochem. Mol. Biol. 33, 1085e1097. Genta, F.A., Dillon, R.J., Terra, W.R., Ferreira, C., 2006. Potential role for gut microbiota in cell wall digestion and glucoside detoxification in Tenebrio molitor larvae. J. Insect Physiol. 52, 593e601.

Genta, F.A., Dumont, A.F., Marana, S.R., Terra, W.R., Ferreira, C., 2007. The interplay of processivity, substrate inhibition and a secondary substrate binding site of an insect exo-b-1,3-glucanase. Biochim. Biophys. Acta 1774, 1079e1091. Guex, N., Peitsch, M.C., 1997. SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis 18, 2714e2723. Gupta, R., Jung, E., Brunak, S., 2004. Prediction of N-glycosylation sites in human proteins. http://www.cbs.dtu.dk/services/NetNGlyc/. Hahn, M., Pons, J., Planas, A., Querol, E., Heinemann, U., 1995. Crystal structure of A resolution. Bacillus licheniformis 1,3-1,4-b-D-glucan 4-glucanohydrolase at 1.8  FEBS Lett. 374, 221e224. Julenius, K., 2007. NetCGlyc 1.0: prediction of mammalian C-mannosylation sites. Glycobiol 17, 868e876. Julenius, K., Mlgaard, A., Gupta, R., Brunak, S., 2005. Prediction, conservation analysis and structural characterization of mammalian mucin-type O-glycosylation sites. Glycobiol 15, 153e164. Kelley, L.A., Sternberg, M.J.E., 2009. Protein structure prediction on the web: a case study using the Phyre server. Nat. Protoc. 4, 363e371. Krissinel, E., Henrick, K., 2004. Secondary-structure matching (SSM), a new tool for fast protein structure alignment in three dimensions. Acta Cryst D60, 2256e2268. Krystal, G., Macdonald, C., Munt, B., Ashwell, S., 1985. A method for quantitating nanogram amounts of soluble protein using the principle of silver binding. Anal. Biochem. 148, 451e460. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680e685. Lambert, M., Neish, A.C., 1950. Rapid method for estimation of glycerol in fermentation solutions. Can. J. Res. Sect. B e Chem. Sci. 28, 83e89. Lambert, C., Leonard, N., De Bolle, X., Depiereux, E., 2002. ESyPred3D: prediction of proteins 3D structures. Bioinformatics 18, 1250e1256. Laskowski, R.A., MacArthur, M.W., Moss, D.S., Thornton, J.M., 1993. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Cryst. 26, 283e291. Lund, O., Nielsen, M., Lundegaard, C., Worning, P.2002. CPHmodels 2.0: X3M a Computer program to extract 3D Models. CASP5 conference A102. Magnelli, P., Cippolo, J.L., Abeijom, C., 2002. A refined method for determination of Saccharomyces cerevisae cell wall composition and ß-1,6 glucan fine structure. Analyt. Biochem. 301, 136e150. Marana, S.R., Mendonc xa, L.M.F., Andrade, E.H.P., Terra, W.R., Ferreira, C., 2003. The role of residues R97 and Y331 in modulating the pH optimum of an insect b-glycosidase of family 1. Eur. J. Biochem. 270, 4866e4875. Miles, E.W., 1977. Modification of histidyl residues in proteins by diethylpyrocarbonate. Meth. Enzymol. 47, 431e442. Mizohata, E., Anwaruzzaman, M., Okuno, H., Tomizawa, K., Shigeoka, S., Kai, Y., Yokota, A., 2003. Chemical modification of arginine alleviates the decline in activity during catalysis of spinach rubisco. Biochem. Biophys. Res. Comm. 301, 591e597. Noelting, G., Bernfeld, P., 1948. Sur les enzymes amylolitiques. III. La beta-amilase:
et controle de l'absence d'alfa-amilase. Helv. Chim. Acta 31, dosage d'activite 286e290. Pan, S.Q., Ye, X.S., Kuc, J., 1989. Direct detection of ß-1,3-glucanase isozymes on polyacrylamide electrophoresis and isoelectrofocusing gels. Analyt. Biochem. 182, 136e140. Pauchet, Y., Freitak, D., Heidel-Fischer, H.M., Heckel, D.G., Vogel, H., 2009. Immunity or digestion: glucanase activity in a glucan-binding protein family from Lepidoptera. J. Biol. Chem. 284, 2214e2224. Read, S.M., Currie, G., Bacic, A., 1996. Analysis of the structural heterogeneity of laminarin by electrospray-ionisation-mass spectrometry. Carbohyd. Res. 281, 187e201. Richards, O.W., Davies, R.G., 1977. Imm's General Textbook of Entomology, tenth ed. Chapman, London. Riordan, J.F., Vallee, B.L., 1972. Nitration with tetranitromethane. Meth. Enzymol. XXV, 515e521. Robyt, J.F., French, D., 1967. Multiple attack hypothesis of amylase action. Action of porcine pancreatic, human salivary, and Aspergillus oryzae a-amylases. Arch. Biochem. Biophys. 122, 8e16. Rombouts, F.M., Phaff, H.J., 1976. Lysis of yeast cell walls. Lytic b-(1/3)-glucanases from Bacillus circulans WL-12. Eur. J. Biochem. 63, 121e130. Segel, I.H., 1975. Enzyme Kinetics. Behavior and Analysis of Rapid Equilibrium and Steady-state Enzyme Systems. John Wiley & Sons, Inc., New York, p. 957. Sengupta, S., Jana, M.L., Sengupta, D., Naskar, A.K., 2000. A note on the estimation of microbial glycosidase activities by dinitrosalicylic acid reagent. Appl. Microbiol. Biotechnol. 53, 732e735. Shapiro, A.L., Vinuela, E., Maizel, J.V., 1967. Molecular weight estimation of polypeptide chains by electrophoresis in SDS-polyacrylamide gels. Biochem. Biophys. Res. Commun. 28, 815e820. Slaymaker, A.K., Ferrington Jr., L.C., Lichtwardt, R.W., 1998. Chironomidae-Trichomycete associations: a literature review. J. Kans. Entomol. Soc. 71, 490e500. Spande, T.F., Witkop, B., 1967. Tryptophan involvement in the function of enzymes and protein hormones as determined by selective oxidation with N-bromosuccinimide. Meth. Enzymol. XI, 506e522. Takahashi, K., 1968. The reaction of pPhenylglyoxal with arginine residues in proteins. J. Biol. Chem. 243, 6171e6179. Teodorescu, O., Galor, T., Pillardy, J., Elber, R., 2004. Enriching the sequence substitution matrix by structural information. Proteins: Structure, Funct. Bioinform. 54, 41e48. Ï

and C. Ferreira are staff members of their respective departments and research fellows from CNPq.

873

874

F.A. Genta et al. / Insect Biochemistry and Molecular Biology 39 (2009) 861e874

Terra, W.R., Ferreira, C., 1994. Insect digestive enzymes: properties, compartmentalization and function. Comp. Biochem. Physiol. B 109, 1e62. Terra, W.R., Ferreira, C., Bastos, F., 1985. Phylogenetic considerations of insect digestion. Disaccharidases and the spatial organization of digestion in the Tenebrio molitor larvae. Insect Biochem. 15, 443e449. Terra, W.R., Ferreira, C., de Bianchi, A.G., 1979. Distribution of digestive enzymes among the endo and ectoperitrophic spaces and midgut cells of Rhynchosciara and its physiological significance. J. Insect Physiol. 25, 487e494. Trudel, J., Grenier, J., Asselin, A., 1998. Detection of enzymes active on various ß-1,3glucans after denaturating polyacrylamide gel electrophoresis. Elecrophoresis 19, 1788e1792.

Tsai, L.-C., Shyur, L.-F., Cheng, Y.-S., Lee, S.-H., 2005. Crystal structure of truncated Fibrobacter succinogenes 1,3-1,4-b-D-glucanase in complex with b-1,3-1,4-cellotriose. J. Mol. Biol. 354, 642e651. Wiederstein, M., Sippl, M.J., 2007. ProSA-web: interactive web service for the recognition of errors in three-dimensional structures of proteins. Nucleic Acids Res. 35, W407eW410. Withers, S.G., 2001. Mechanisms of glycosyl transferases and hydrolases. Carbohyd. Pol 44, 325e337. Yu, X.-Q., Zhu, Y.-F., Ma, C., Fabrick, J.A., Kanost, M.R., 2002. Pattern recognition proteins in Manduca sexta plasma. Insect Biochem. Mol. Biol. 32, 1287e1293.