chi18H8 cDNA sub-cloning

The nucleotide sequence of the chitinase Chi18H8 gene was previously deposited in the GenBank database under the accession number KC763366 [7]. The cDNA encoding for the chitinase was sub-cloned into the pET24b(+) expression plasmid (kanamycin resistance; Novagen Inc., Madison, USA) by using the primers chiEcoF (5′-ATAAAGAATTCCATGCGCCAGCTCACGCTTCTC-3′) and chiXhoR (5′-ATAAACTCGAGCTAATTGCCCCTATGCAGACTGG-3′), containing the underlined restriction sites for EcoRI and XhoI, respectively. The plasmid pGEX-6P-3::chi18H8, previously prepared [7], was used as DNA template. E. coli DH5α (Invitrogen-Life Technologies, Carlsbad, USA) was used as host for the sub-cloning procedures. The construct, following its control by DNA sequencing (BMR Genomics, Padua, Italy), was transformed into E. coli BL21 Star™(DE3) (Invitrogen-Life Technologies). Recombinant E. coli strains were maintained on Luria-Bertani broth (LB, Miller’s modification; Sigma-Aldrich, St. Louis, USA) agar plates supplemented with 50 µg/mL kanamycin.

Chi18H8 expression

All medium components and reagents were from Sigma-Aldrich, unless otherwise stated. Protein expression was carried out in the following media, supplemented with 50 µg/mL kanamycin: LB; terrific broth (TB); super broth (SB: 32 g/L tryptone, 20 g/L yeast extract, 5 g/L NaCl); autoinduction media A and B. Autoinduction medium A composition was based on [31]. Autoinduction medium B included: 10 g/L tryptone, 5 g/L yeast extract, 3.3 g/L (NH₄)₂SO₄, 6.8 g/L KH₂PO₄, 7.1 g/L Na₂HPO₄, 0.5 g/L glucose, 2 g/L α-lactose, 0.15 g/L MgSO₄, 2 mg/L CaCl₂, 2 mg/L MnSO₄ × H₂O, 2 mg/L ZnSO₄, 2 mg/L CoCl₂, 2 mg/L CuCl₂ × 2H₂O, 2 mg/L NiCl₂, 2 mg/L NH₄MoO₄, 2 mg/L FeCl₃. Trace element (MgSO₄, CaCl₂, MnSO₄ × H₂O, ZnSO₄, CoCl₂, CuCl₂ × 2H₂O, NiCl₂, NH₄MoO₄, FeCl₃) stock solutions were sterilized by filtration (0.2 µm) and stored at 4 °C.

Starter cultures were prepared from a single recombinant E. coli colony inoculated in 10 mL LB medium supplemented with 50 µg/mL kanamycin, grown overnight at 37 °C and 200 revolutions per minute (rpm). Baffled 300 mL Erlenmeyer flasks containing 50 mL of the different media were inoculated with the starter culture (initial OD_600nm = 0.1) and further incubated as above. For LB, TB and SB media, protein expression was induced by adding 0.4 mM isopropyl β-d-thiogalactopyranoside (IPTG) to cells at early- or late-exponential growth phase. After induction, cells were cultured at various temperatures (37, 25 or 20 °C, respectively) at 200 rpm and harvested by centrifugation (1900×g for 30 min at 4 °C) at different time intervals. Following centrifugation, total proteins in the supernatants (i.e. the cell-free fermentation broths) were concentrated by 10% (v/v) trichloroacetic acid precipitation and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE, see below). In parallel, cell pellets were sonicated on ice (3–5 cycles of 30 s each, with a 30-s interval, using a Branson Sonifier 250, Danbury, USA) in phosphate buffer saline (PBS) pH 7.3 (140 mM NaCl, 2.7 mM KCl, 10.1 mM Na₂HPO₄, 1.8 mM KH₂PO₄) containing 10 µg/mL deoxyribonuclease (DNase), 0.19 mg/mL phenylmethylsulfonylfluoride (PMSF) and 0.7 mg/mL pepstatin. Soluble (cytoplasmic) and insoluble (containing membranes and IBs) cell fractions were then separated by centrifugation at 20,000×g for 1 h at 4 °C. Insoluble fractions were re-suspended in a volume of PBS equal to the corresponding cytoplasmic soluble fraction (2 mL/g cell) for successive SDS-PAGE analyses. In every fraction, protein concentration was determined by the Biuret assay [32].

Chi18H8 production

Flask cultures of recombinant E. coli cells, grown overnight in LB medium supplemented with 50 µg/mL kanamycin, were used to inoculate (initial OD_600nm = 0.1) 2 L Erlenmeyer flasks (containing 750 mL medium), or 3 L P-100 Applikon glass reactors (Applikon Biotechnology, Delft, The Netherlands) equipped with a AD1030 biocontroller and AD1032 motor (containing 2 L medium), or 30 L Bioengineering (Bioengineering AG, Wald, Switzerland) stirred fermenter (containing 27 L medium). The medium used throughout was LB medium supplemented with 50 µg/mL kanamycin. 2 L Erlenmeyer flasks were incubated at 37 °C and 200 rpm. Growth in fermentors was conducted at 37 °C and stirring at 500 rpm for 3 L bioreactor or at 300 rpm for the 30 L bioreactor, respectively, at constant 1.0 vvm aeration rate and pressure control set at 0.5 bar. Dissolved oxygen (measured as % of the initial pO₂ value) and pH were monitored using an Ingold polarographic oxygen electrode and a pH meter, respectively. Foam production was monitored through an antifoam sensor and controlled by adding Antifoam SE-15. When cell culture reached an OD_600nm = 0.6, protein production was induced by adding 0.4 mM IPTG: cultivation was then prolonged at 20 °C for further 24 h.

Recovery of IBs and solubilization of Chi18H8

Escherichia coli cells were harvested by centrifugation at 3220×g for 20 min, washed with sodium chloride-Tris-EDTA (STE buffer: 10 mM Tris–HCl pH 8.0, 1 mM EDTA [ethylenediaminetetraacetic acid], 100 mM NaCl). The different protocols for IB recovery and solubilization are described in Additional file 1: Table S1. In the case of the acid solubilization method, after centrifugation, the cell pellet was re-suspended in 50 mM Tris–HCl pH 8.0, 25% (w/v) sucrose, 1 mM EDTA, and incubated for 30 min at room temperature under vigorous shaking. After sonication on ice (6 cycles of 30 s each, with a 30-s interval), 0.2 M NaCl, 1% (w/v) sodium deoxycholate (DOC) and 1% (v/v) Nonidet P-40 were added. The sample was further incubated as above and centrifuged (20,000×g at 4 °C for 30 min). The pellet was washed with 1% (v/v) Triton X-100 and 1 mM EDTA, followed by centrifugation at 12,000×g at 4 °C for 10 min; the procedure was repeated twice. IBs were then washed twice with 10 mL/g cell deionized water, with centrifugation at 12,000×g at 4 °C for 10 min in-between, and stored overnight at −20 °C. To solubilize Chi18H8 from IBs, the frozen pellet was re-suspended in 10 mL/g cell of 10 or 100 mM HCl or lactic acid and incubated at 37 °C and 200 rpm for 5 h. Insoluble material was removed by centrifugation at 1900×g at 4 °C for 5 min. The solubilized protein was then dialyzed overnight against 100 mM sodium acetate buffer pH 5.0 or 100 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) pH 5.6.

Chi18H8 purification

Enzyme samples following dialysis were used for testing different chromatographic methods, described in details in Additional file 1: Table S2. Chi18H8 purification/enrichment was achieved by loading the protein sample after dialysis against 100 mM HEPES pH 5.6 onto the weak anionic exchanger Diaion WA11 resin (Resindion s.r.l., Milan, Italy), previously activated with methanol/water (1:1) and then equilibrated with 100 mM HEPES pH 5.6. Pilot experiments were performed with 1 mL resin (wet volume) or, for large scale purification, with 40 mL of WA11 resin (in a 4.5 cm diameter Amicon column loaded at a flow rate of 20 mL/min). The chromatography, run under hydrophobic interaction (HIC) conditions, allowed to fully recover the Chi18H8 protein in the flow-through, whereas other proteins retained by the WA11 resin were then recovered by an isocratic elution using 50% (v/v) ethanol in 100 mM HEPES pH 5.6. The Chi18H8 concentration was estimated using the theoretical extinction coefficient at 280 nm (77,015 M^-1cm^-1), based on the amino acid sequence of the protein.

SDS-PAGE electrophoresis and zymogram analysis

Chi18H8 size, production and solubilization were analyzed by SDS-PAGE with 12% (w/v) polyacrylamide gels [33] and estimated by densitometric analysis (Quantity One, Bio-Rad Laboratories, Hercules, USA) using known amount of His₆-glycine oxidase (His₆-GO) from Bacillus subtilis [34] as a standard. Chitinolytic activity was detected through zymogram analysis (semi native-PAGE) using 10% (w/v) polyacrylamide gels containing 0.7 mg/mL of carboxymethyl-chitin-remazol brilliant violet (CM-chitin-RBV) (Loewe Biochemica, Suerlach, Germany), as described in [7]. Briefly, the protein samples were diluted in a sample buffer lacking any reducing agent and incubated for 10 min at room temperature. Electrophoresis was conducted at 4 °C in Tris-glycine-SDS running buffer according to standard running conditions. The gel was rinsed twice in 2.5% (v/v) Triton X-100 for 30 min at room temperature to remove SDS and incubated in 100 mM sodium acetate pH 5.0 at 37 °C until the appearance of clear zones indicating chitinolytic activity. The chitinase from Trichoderma viride (Sigma-Aldrich) was used as positive control.

Chi18H8 activity assay

Chi18H8 activity was assayed with the fluorimetric chitooligosaccharide analogues 4-methylumbelliferyl N-acetyl-β-d-glucosaminide (4-MU-GlcNAc), 4-methylumbelliferyl N,N’-diacetyl-β-d-chitobioside (4-MU-(GlcNAc)₂) and 4-methylumbelliferyl N,N’,N’’-triacetyl-β-d-chitotrioside (4-MU-(GlcNAc)₃) as described previously [7]. Unless otherwise stated, Chi18H8 activity was assayed at 37 °C in 100 mM sodium acetate pH 5.0 [7]. Chitinolytic activity was also determined on colloidal chitin, prepared from chitin flakes from shrimp shells (Sigma-Aldrich), by the colorimetric method previously described [8]. One unit (U) of Chi18H8 activity was defined as the amount of enzyme that released 1 μmol of 4-MU or GlcNAc per min at 37 °C.

Effect of pH and temperature on Chi18H8 activity

pH influence on Chi18H8 activity on the substrate 4-MU-(GlcNAc)₂ was determined by the fluorimetric assay in the following buffers (each 100 mM): glycine–HCl (pH 3.0), sodium acetate (pH 4.0 and 5.0), sodium phosphate (pH 6.0 and 7.0), Tris-HCl (pH 8.0) and sodium pyrophosphate (NaPPi, pH 9.0). Optimal temperature for the enzyme activity was determined by the same fluorescent assay, performing the reaction at various temperatures (from 5 to 70 °C). Stability of the enzyme was tested by incubating the recombinant protein at 30 °C at different pHs (in 100 mM sodium acetate pH 5.0 or 100 mM sodium phosphate pH 6.0 and 7.0): after different incubation times (from 0 to 144 h), the residual chitinolytic activity was assayed on 4-MU-(GlcNAc)₂.

Effect of metal ions, organic solvents, detergents and other compounds on Chi18H8 activity

The effect of metal ions [Ca²⁺ (CaCl₂ × 2H₂O), Cu²⁺ (CuCl₂ × 2H₂O), Fe³⁺ (FeCl₃ × 6H₂O), K⁺ (KCl), Mg²⁺ (MgCl₂ × 6H₂O), Mn²⁺ (MnCl₂ × 4H₂O), Ni²⁺ (NiCl₂ × 6H₂O), NH₄ (NH₄Cl), Zn²⁺ (ZnCl₂), Co²⁺ (CoCl₂ × 6H₂O)], reducing agents [dithiothreitol (DTT), 2-mercaptoethanol], the chelating agent EDTA, detergents [SDS, Triton X-100, Tween-20, DOC, N-lauroylsarcosine (NLS), Nonidet-40], organic solvents [ethanol, methanol, propanol, dimethyl sulfoxide (DMSO)], sugars (GlcNAc, chitobiose) and salt (NaCl) on Chi18H8 activity was investigated by adding each compound to the fluorimetric assay mixture. Final concentrations were: 20 mM for metal ions, EDTA and 2-mercaptoethanol, 10 mM for sugars and DTT, 10% (v/v) for organic solvents, and 1% (w/v) for detergents. In each case the residual activity was compared to the activity without additional compounds, set as 100%.

Circular dichroism

Far-UV CD spectra were recorded at 15 or 30 °C using a Jasco J-715 spectropolarimeter (Jasco, Cremello, Italy) equipped with temperature control, in the 190–250 nm wavelength range. Measurements were carried out in quartz cuvettes of path length 1 mm, employing a protein solution at 0.2 mg/mL in 10 mM HEPES pH 5.4, and corrected for buffer contribution. Secondary structure composition was calculated from deconvolution of the CD spectra with the program k2d3 (http://k2d3.ogic.ca//index.html) [35]. Binding of Ca²⁺ to purified Chi18H8 was investigated by recording the far-UV CD spectrum during the titration of the apoprotein form of the enzyme (obtained by incubation for 30 min with 20 mM EDTA), with increasing concentrations of CaCl₂ (5, 10, 100, 1000 and 10,000 μM). The change in CD signals at 219 nm versus CaCl₂ concentration were fitted to a classical 1:1 complex saturation equation to calculate the dissociation constant, K_d.

Determination of mode of action

A chitin-like substrate, i.e. a water-soluble high molecular weight chitosan with a degree of acetylation of 63%, was used as substrate for Chi18H8 to determine the enzyme’s mode of action as previously described [36]. The chitosan (30 mg) was dissolved in 3 mL of 80 mM sodium acetate, pH 5.5. To start the depolymerization reaction, 2.5, 7.5 or 26 µL of Chi18H8 (1.16 mg/mL) were added to 1 mL substrate. After incubating the mixture at 37 °C for 30 min, the reaction was stopped by adjusting the pH to 2.0 and boiling for 10 min. The reaction mixture containing the oligomers was injected and separated on three XK 26 columns in series, packed with Superdex™ 30 (Amersham Pharmacia Biotech, Piscataway, USA) using a refractive index (RI) detector (Shimadzu Schweiz GmbH, Reinach, Switzerland), as previously described [36].

Antifungal activity assay

Chi18H8 antifungal activity was evaluated by growth repression assays, in triplicate, on Fusarium graminearum ATCC 46779, Rhizoctonia solani ATCC 10183 and Botrytis cinerea ATCC 26943. Plate assays were carried out by centrally inoculating an agar plug, harboring an actively growing fungal culture, in 9-cm diameter Petri plates containing diluted potato dextrose agar (PDA). Sterilized paper discs (Munktell Filter ab, Falun, Sweden) 5 mm in diameter were soaked in 1000, 500, 100 or 10 µg/mL, respectively, of the solubilized Chi18H8 (in 100 mM sodium acetate, pH 5.0) and placed 25 mm around the fungal plug. A liquid assay system was also used. Wells (2 mL cellplate, Biofluidfocus, Largo, USA) containing fungal PDA agar plugs and 200 μL potato dextrose broth (PDB) were added with Chi18H8 (in 100 mM sodium acetate, pH 5.0) to a concentration of 1000, 100 or 10 µg/mL. Boiled enzyme in the same buffer was used as negative control in both assays. The systems were incubated at 23 °C in darkness for 48 h, and longer for F. graminearum due to slower mycelial growth rate. Growth of fungal hyphae was monitored daily using a microscope. For the liquid assay, dry weight biomass of the fungal plugs was measured seven days after inoculation.

Heterologous expression of the metagenome-sourced Chi18H8

The complete chi18H8 gene sequence (G + C ratio 64.4%) was previously identified by genetic and expression screening of a high-molecular-weight DNA metagenomic library constructed from a soil in Uppsala (Sweden), characterized as suppressive to club-root disease of cabbage [6, 7, 29]. The gene, consisting of 1275 nucleotides and encoding a protein of 424 amino acids (predicted molecular mass of 45.96 kDa and isoelectric point of 7.75), was obtained by primer walking technique of a fosmid clone [7]. For this work, it was cloned into the pET24b(+) expression plasmid in E. coli BL21 Star™(DE3) cells, under the control of the IPTG-inducible T7 promoter.

Recombinant E. coli cells harboring pET24b(+)::chi18H8 and the control strain carrying the empty vector were grown at the following conditions: (1) five different growth media (LB, TB, SB, autoinduction media 1 and 2); (2) adding 0.4 mM IPTG at the early or at the late exponential phase of growth; (3) keeping temperature after induction at 37 °C or reducing it to 25 or 20 °C; (4) growing cells after induction for different time intervals before harvesting. In all the above experimental conditions, an intense protein band at the expected molecular mass for Chi18H8 was observed in SDS-PAGE only in the insoluble fractions from the recombinant strain harboring the chi18H8 gene (no signal was detected in cytoplasmic soluble fractions, nor in cell-free culture broths after protein precipitation with trichloroacetic acid). Figure 1 shows the SDS-PAGE when (a) cells were grown in LB, (b) protein expression was induced in the early exponential phase, (c) and temperature was maintained at 37 °C after induction. Similar results were achieved in SB and TB media; lower protein production was observed using the two autoinduction broths [31] or by inducing cells during the late exponential phase of growth (data not shown).

As shown in Fig. 2, accumulation of Chi18H8 into the insoluble fraction increased by prolonging the incubation time after induction, reaching 2.7 mg/g cell (corresponding to 32 mg/L culture in LB medium) after 24 h at 37 °C. Decreasing the incubation temperature to 25 or 20 °C favored the accumulation of recombinant protein into insoluble fractions (more than 9 mg/g cell, corresponding to 55 mg/L culture after 24 h incubation in LB at 20 °C). Fluorimetric enzyme assay of these insoluble fractions revealed that no chitinolytic activity was detectable from cells grown at 37 °C, whereas a weak enzyme activity appeared when growth temperature decreased to 25 and 20 °C (Fig. 2).

Taking all together, these results suggest that Chi18H8 accumulated at 37 °C largely into IBs in an inactive and insoluble form, as often occurs when E. coli is used to express metagenomics-sourced enzymes [7, 8]. Indeed, decreasing the expression temperature favored the formation of so-called non-classical IBs, i.e. those IBs that are, at least in part, biologically active since they contain an increasing portion of recombinant active protein molecules [37–39].

Solubilization of IBs and purification of Chi18H8

Considering the above results and those previously reported [7] on E. coli pGEX-6P-3::chi18H8 producing cytoplasmic Chi18H8 at an extremely low yield, a new approach was tested to recover the chitinase at a higher yield by solubilization of IBs produced by E. coli BL21 Star™(DE3)/pET24b(+)::chi18H8 cells grown in LB and incubated for 24 h at 20 °C after adding IPTG. These IBs, that likely contain some protein molecules in native-like active conformation, were recovered by centrifugation from cell lysate after mild sonication conditions. See Additional file 1: Table S1 for the different solubilization methods that were tested. Briefly, first we tried classical methods, re-suspending IBs in solutions containing strong denaturing agents and chaotropes, such as urea, guanidium hydrochloride (GdnHCl) or guanidium thiocyanate (GdnTC), alone or combined with the detergent Triton X-100, the reducing agent DTT, the chelating agent EDTA or NaCl salt. By increasing the denaturing agent concentration, an increased protein solubilization (up to 85–95%) was observed through SDS-PAGE. However, during the following refolding steps (removing the denaturant agents by dialysis), Chi18H8 aggregation and precipitation occurred and consequently the protein samples were biologically inactive (as checked by the fluorimetric enzyme activity assay). Milder solubilization conditions of IBs (Additional file 1: Table S1) were then evaluated for better preserving the inferred existing native-like secondary structure and reducing the risk of protein aggregation during the purification process. These mild solubilization protocols were based on IB incubation with detergents (Triton X-100, NLS and CHAPS [3-[(3-cholamidopropyl)dimethylammonium]-1-propanesulfonate]), or alkyl alcohol (n-propanol), or acid pH solutions (HCl, lactic acid, formic acid). Using the protocols based on detergents and n-propanol, the solubilization yield was very low (ca. 5–10%, Additional file 1: Table S1). Recovery of active Chi18H8 was only achieved in the case of IB solubilization in aqueous acid solutions of hydrochloric or lactic acid (but not in formic acid). As reported in Additional file 1: Table S1, this protocol of acid solubilization involved an osmotic shock pre-treatment of cell pellet with 25% (w/v) sucrose for disrupting the outer membrane of E. coli cells and removing periplasmic proteins. Additionally, by introducing different and sequential washing steps (with low concentrations of detergents like DOC, Triton X-100 and Nonidet P-40) after sonication of the IBs and prior to protein solubilization, IB purity increased and contaminants were reduced during the following steps of Chi18H8 purification; actually the washing steps contributed to the removal of membrane fragments, see [37, 40].

Thus, by re-suspending IBs in 10 mM hydrochloric or lactic acid we achieved solubilization yields of more than 90 or 80%, respectively, as shown in Table 1. However, in the case of IB solubilization in 10 mM lactic acid, the specific activity of the recovered Chi18H8 (40.7 U/mg protein) was definitely higher than in hydrochloric acid (29.6 U/mg protein). Increasing the acid concentration to 100 mM led to a reduction of both solubilization yield and specific activity. SDS-PAGE analysis revealed that the solubilized Chi18H8 in 10 mM lactic acid was 70% pure (Fig. 3a; Table 2).

Solubilization agent	Concentration (mM)	Solubilization yield (%)	Specific activity (U/mgChi18H8)a
HCl	10	93 ± 4.0	29.6 ± 1.2
	100	20 ± 2.0	15.9 ± 0.9
Lactic acid	10	82 ± 3.5	40.7 ± 1.4
	100	60 ± 6.5	30.7 ± 1.3

Solubilization yield was estimated by densitometric analysis of proteins separated by SDS-PAGE. Enzyme activity was measured by fluorimetric assay on 4-MU-(GlcNAc)₂

^aValues represent the mean of three independent experiments (mean ± standard error)

	Volume	Activity	Protein concentration	Specific activity	Purification	Recovery
	(mL)	(U tot)	(mg/mL)	(U/mgprotein)	(fold)	(%)	(mg/gcell)
IBs	2.0	4.7	11.2	0.2	1.0	100.0	9.4
Soluble fraction from IBs	10.0	314.0	1.1	28.5	66.8	82.0	7.7
Purified Chi18H8	25.3	355.7	0.2	63.9	75.7	57.4	5.4

Enzymatic activity was measured by fluorimetric assay on 4-MU-(GlcNAc)₂. Protein concentration was determined by the Biuret assay

In order to further purify Chi18H8, different precipitation methods and chromatographic procedures were then evaluated. Salts, including NaCl, KCl, MgCl₂ and NH₄(SO₄)₂, were added to obtain a selective Chi18H8 precipitation, so were increasing concentrations of ethanol and the addition of CM-chitin-RBV. However, the enzyme co-precipitated with most of the contaminating proteins (data not shown). Different chromatographic methods were tested using sixteen resins, as listed in Additional file 1: Table S2. At most of the conditions, Chi18H8 did not interact with the resins and the chitinolytic activity was mainly found in the flow-through fractions, together with most of the contaminating proteins. In some cases the protein precipitated and, in general, it was inactive if the pH increased above 7.0. The weak cationic exchanger P11 and the hydrophobic resin HP20SS retained Chi18H8; however, from neither resin the enzyme could be eluted in a biologically active form (Additional file 1: Table S2). Interestingly, when the weak anionic exchanger WA11 resin was used at HIC conditions, most of the contaminating proteins were retained by the resin, allowing a partial purification of the chitinase from the flow-through in 100 mM HEPES pH 5.6. In this case, the purified protein was >95% pure (Fig. 3a) and the zymogram analysis in semi-denaturing conditions with CM-chitin-RBV confirmed the chitinolytic activity of the purified enzyme (Fig. 3b). Indeed, none chitinolytic activity was found when the contaminating proteins were eluted from WA11 resin by an isocratic elution using 50% (v/v) ethanol in 100 mM HEPES pH 5.6. As summarized in Table 2, the two-step lactic acid solubilization/HIC purification method allowed the recovery of more than 55% of the recombinant enzyme accumulated into IBs, with a 320-fold increase of specific activity (from 0.2 U/mg protein in the IBs, to 63.9 U/mg protein after HIC purification).

Chi18H8 production in 3 L bench- and 30 L industrial-bioreactors

Chi18H8 production and purification was scaled-up first in 3 L and then in 30 L bioreactors, where E. coli BL21 Star™(DE3)/pET24b(+)::chi18H8 cells were grown at the conditions optimized for flask fermentation. Figure 4 shows the time course of cell growth and Chi18H8 production in the 3 L bench-fermenter. Heterologous protein production was induced by adding IPTG to cells in the early stationary phase of growth. The maximum chitinase production in IBs after 24 h was 9.1 mg/g cell (corresponding to 75 mg/L) and 12.9 mg/g cell (equal to 80 mg/L) for the 3 L and the 30 L bioreactor, respectively. These amounts were comparable or even higher than what was obtained at flask level (Table 3).

	Cell weight (gcell/L)	Chi18H8 production (mg/gcell)	Solubilization yield (%)	Specific activity (U/mgprotein)
Flask	4.0	9.4	>80	40.7
3 L bench-bioreactor	5.3	9.1	>65	52.1
30 L industrial-bioreactor	6.2	12.9	>65	48.4

Protein expression was induced in the early exponential growth phase (OD_600nm ~0.6) with 0.4 mM IPTG and cells were harvested after 24 h of incubation. Chi18H8 production (mg protein/g cell) and solubilization yield in 10 mM lactic acid were estimated by densitometric analysis of proteins separated by SDS-PAGE. Specific activity of the solubilized enzyme was measured by the fluorimetric assay on 4-MU-(GlcNAc)₂

The following IB solubilization in 10 mM lactic acid allowed, from cells grown at both reactor sizes, the recovery of active Chi18H8 with a >65% yield and a specific activity of 52.1 and 48.4 U/mg protein, respectively (Table 3). Solubilized protein was about 70% pure, as revealed by SDS-PAGE densitometric analysis, similarly to the results obtained with the protein produced in flasks (not shown). Also purification using HIC with the weak anionic exchanger WA11 resin turned out to be scalable and suitable for the production of Chi18H8. A volume of 135 mL solubilized Chi18H8 in 100 mM HEPES pH 5.6, corresponding to 2.2 L of culture, was loaded onto a column of WA11 resin (40 mL). Also in this case, the Chi18H8 protein was found in the flow-through, whereas contaminating proteins were retained by the resin. Chi18H8 recovery yield was approximately 70% and the purity was above 95%.

Chi18H8 characterization

Mode of action

Size-exclusion chromatograms revealing the length distribution of the COs when increasing amounts of Chi18H8 were added to a water-soluble chitin-like substrate, i.e. a high-molecular weight chitosan with a degree of acetylation of 63%, are shown in Fig. 5. The molecules eluting first in the chromatograms (after 7 h) are the largest molecules (e.g. chitosan with more than 50 units). The signal at the other end of the chromatograms (18 h) is the salt peak. Furthermore, the chitin dimer (GlcNAc–GlcNAc, labelled AA in Fig. 5) is the peak eluting between 15 and 15.5 h, the peaks eluting at 14 and 13.4 h are respectively the tetramer (labelled 4 in the figure) and the pentamer, and so on [36]. With increasing amounts of Chi18H8 added to the substrate, the chromatograms show that the chitosan peaks (eluting at 7 h) are gradually reduced and a continuum of oligomers appears as a result of the hydrolysis of the polymeric chitosan chain. The continuum of COs in the chromatograms with no preference for even-numbered oligomers (Fig. 5) indicates that Chi18H8 operates according to an endo-mode of action [36].

pH and temperature profile

The optimum pH for the chitinolytic activity of the purified enzyme was between 5.0 and 6.0 and only 15.3% of the activity was maintained at pH 7.0, whereas no activity was detected at pH 3.0 and 4.0 and above pH 8.0 (Fig. 6a). Consistently, the chitinase was able to hydrolyze the CM-chitin-RBV only in the 5.0–7.0 pH range, and no activity was detectable in zymograms at pH 3.0, 4.0, 8.0 or 9.0 (Fig. 6c). The optimum temperature for chitobiosidase activity on 4-MU-(GlcNAc)₂ was between 35 and 40 °C. More than 55% of the activity was retained at 25 °C (Fig. 6b); at 50 °C the enzyme was inactive. The stability of Chi18H8 was investigated by measuring the residual activity following 144 h of incubation in buffers at different pHs (5.0, 6.0 and 7.0) at the temperature of 30 °C. When Chi18H8 was incubated at pH 7.0, a complete loss of activity was observed within 24 h. At pH 6.0, the activity dramatically decreased to 5% of the initial activity within 144 h. At pH 5.0, however, Chi18H8 retained about 70% of its initial activity after 144 h (Fig. 6d).