INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The ability to degrade nitriles is quite common among microorganisms. The potential of nitrile-degrading enzymes for biotransformations, waste treatment, and the production of herbicide-resistant plants has been assessed (14, 29). Stereoselective hydrolysis of nitriles and amides with whole cells or isolated enzymes has been reported for a number of strains, such as Pseudomonas putida (6, 28), Rhodococcus erythropolis (10), Rhodococcus equi (19), and Rhodococcus rhodochrous (13, 21). In Japan, enzymes from Pseudomonas chlororaphis and R. rhodochrous are used for the production of low-cost chemicals, such as acrylamide (23).

In nature, three different groups of enzymes are involved in the microbial hydrolysis of nitriles. Nitrilases (EC 3.5.5.1 and 3.5.5.7) hydrolyze nitriles to the corresponding carboxylic acids, forming ammonia; nitrile hydratases (EC 4.2.1.84) form amides from nitriles which can be subsequently hydrolyzed by amidases (EC 3.5.1.4). The nocardiaform actinomycete R. rhodochrous has been reported to produce both the nitrilase and the nitrile hydratase/amidase system, depending on the inducer used (32).

Various nitrilases that hydrolyze aromatic and aliphatic substrates have been described for R. rhodochrous. Some of these enzymes were selectively induced with benzonitrile and propionitrile, respectively (11). Nitrilases from R. rhodochrous have been used for the production of acrylic and methacrylic acid (23). "Aliphatic" nitrilases, such as those from various R. rhodochrous strains (NCIMB 11216, K22, and J1), seem to be quite unusual among microorganisms (11, 15, 23). Formerly, nitrilases have been thought exclusively to hydrolyze aromatic substances while aliphatic nitriles have been believed to be degraded by a nitrile hydratase/amidase enzyme system (15).

The reaction mechanism, regulation, and photoactivation of nitrile hydratases, which usually consist of  and  subunits containing either nonheme iron or cobalt atoms, have been studied in detail (14, 27). R. rhodochrous has been reported to produce a high-molecular-weight nitrile hydratase and a low-molecular-weight nitrile hydratase, which were selectively induced by the reaction products (amides) and by urea, respectively (16, 22, 32). The high-molecular-weight nitrile hydratase from R. rhodochrous is used for industrial production (30,000 tons/year) of acrylamide (14, 26, 32).

In this paper, we report on the modification of polyacrylonitrile (PAN) fibers and granulates using enzymes from R. rhodochrous NCIMB 11216, which has previously been used for chemoselective hydrolysis of nitriles at our university (13). Although acrylonitrile seems to be one of most suitable substrates for both nitrile hydratase (25) and nitrilase from R. rhodochrous (15), there are no reports in the literature on the enzymatic hydrolysis of PANs.

PAN fibers have a slightly increasing share of 2,700 tons/year in 1997, holding approximately 10% of the global synthetic fiber market. Several attempts have been made in the last few years to increase the ecoefficiency of the production processes, including the subsequent dyeing of the fibers. However, various chemical methods for the modification of PAN fibers to make them more hydrophilic and thereby enhance dye uptake have not been very successful. Partial hydrolysis of surfacial nitrile groups into amides and acids, which resist strong acids and alkali, leads to irreversible yellowing of the fabrics. Additionally, elevated reaction temperatures, aggressive chemicals, and higher concentrations of dimethyl sulfoxide (12) would lead to unwanted changes in the macroscopic behavior of the fibers. Long-term mild alkali treatment led to lower dye uptake, which was explained by hydrolysis of ester groups of the copolymers to less hydrophilic hydroxyl (-OH) groups. Thus, selective enzymatic hydrolysis of surfacial nitrile groups of PAN fibers offers a promising alternative to chemical processes. The utilization of endoglucanases to enhance the properties of cellulosic fibers has been studied extensively, including contributions from our laboratories (2, 3, 9, 18), and several cellulase-based processes have been introduced into the textile industry. Similarly, nitrile-degrading enzymes could have an immense potential in this area.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Organism and culture conditions. R. rhodochrous NCIMB 11216 was inoculated from agar stock cultures on potato dextrose agar and grown in three steps using two rich media (media 1 and 2) and a minimal medium containing propionitrile as described previously (13). Additionally, a trace element solution containing the following (concentrations in milligrams per liter are in parentheses) was added: ZnSO₄ · 7H₂O (100), MnCl₂ · 4H₂O (30), H₃BO₃ (300), CoCl₂ · 6H₂O (200), CuSO₄ · 5H₂O (10), NiCl₂ · 6H₂O (20), and NaMoO₄ · 2H₂O (30). Cultures were grown for 24 h in each medium at 30°C and 180 rpm. A quarter of the cell harvest was used for inoculation of medium 2, and the total cell harvest was used for inoculation of the minimal medium. Harvesting of cells was performed at 2,500 × g for 10 min at 4°C. Cells were disrupted by ultrasonic treatment on ice and centrifuged again to obtain the cell-free supernatant (enzyme preparation).

Enzyme assay. The incubation mixture for the determination of nitrile-degrading enzyme activity contained 40 mg of the enzyme preparation (protein) per liter, 25 mM acrylonitrile or acrylamide, and 20 mM K₂HPO₄/KH₂PO₄ as a buffer (pH 6.5). The mixture was incubated at 30°C and 200 rpm, and samples were taken every 5 min. The reaction was stopped after 60 min by thermal inactivation of the enzymes, which were subsequently removed by centrifugation. Gas chromatography was used for analysis of the reaction products. The measurement was carried out using an HP5890 SerII gas chromatograph equipped with an HP-Innowax cross-linked polyethylene glycol column (30 m by 0.25 mm). The column temperature was raised to 220°C in accordance with the following profile: 0 to 1 min, 120°C; 1 to 11 min, increase to 220°C; 11 to 13 min, 220 °C.

Protein assay. Protein adsorbed to the fabric was measured by the Lowry method using bovine serum albumin as the standard (8). A 150-mg fabric sample was incubated with 5 ml of a solution containing (grams per liter) 16.6 Na₂CO₃, 4.0 NaOH, 0.2 Na-K tartrate, and 0.1 CuSO₄5H₂O while shaking for 10 min at 22°C. Thereafter, 10% (vol/vol) Folin Ciocalteau phenol reagent (1 N; Merck) was added and the A₇₅₀ was read after 30 min of incubation at 22°C. The protein content in solution was determined by the Bradford method (1). Specific enzyme activities were calculated based on the protein content of the enzyme preparation.

Enzymatic treatment of PAN. Two granular PAN standards (40 and 190 kDa; PSS Polymer Standards Service GmbH, Mainz, Germany) were treated with the crude enzyme preparation. The incubation mixture contained 1% (wt/vol) PAN and 0.36 or 0.0036% (wt/vol) enzyme preparation (protein) in 1 ml of 57 mM phosphate buffer (pH 7.0). Experiments were carried out in 1.5-ml screw-cap plastic flasks, which were shaken at 300 rpm at 25°C for 72 h. The reaction was stopped by centrifugation, and ammonia that had been released into the supernatant was monitored as described previously during incubation for 15 min at 50°C (4).

Enzymatic treatment of acrylic fibers. Commercial acrylic fibers consisting of PAN and 7% (wt/wt) vinyl acetate as a copolymer were obtained from Fisipe SA, Lavradio, Portugal. The acrylic fibers were washed twice with tap water at 60°C for 1 h each time. Enzymatic treatment was performed in screw-cap plastic flasks (100 ml) at 30°C on a Linitest machine (horizontal shaker commonly used for treatment of fabrics) rotating at 30 rpm for 3 days. The incubation mixture contained 1.0 g of acrylic fibers, 5.0 mg of enzyme preparation (protein) or heat-inactivated enzymes (control), and 0.5% (vol/vol) dimethyl formamide (DMF) in 40 ml of 57 mM phosphate buffer set to different pH values as indicated below. Subsequently, the fabrics were washed with 1% (wt/vol) Na₂CO₃ at pH 11 for 30 s, rinsed with distilled water for 10 s, and dried for 1 h at 60°C. The release of acetic acid was monitored using an enzyme-based assay kit from Boehringer.

Dyeing experiments. Dyeing of enzymatically treated fabrics was carried out using the Ahiba Spectradye system from Datacolor International (Lucerne, Switzerland). A 2.0-g fabric sample, 80 mg of methylene blue (C.I. 52015 Basic Blue 9; Sigma), and 20 mg of Coomassie brilliant blue G (C.I. 42655 Acid Blue 90; Sigma) were incubated in 40 ml of distilled water (bath ratio, 1:20) at 50°C and 60 rpm for 60 min (temperature increase to 40°C from 0 to 5 min, addition of dye from 5 to 10 min at 40°C, increase to 50°C from 10 to 20 min, 50°C from 20 to 80 min, and cooling down to 20°C from 80 to 85 min). Dye uptake was quantified by measuring the K/S value, which correlates to the reflectance (R in percent) in the following way:
K/S values of the fabrics were measured using an ACS reflectance spectrometer.

XPS. X-ray photoelectron spectroscopy (XPS) analysis was carried out on an ESCALAB 200A (VG Scientific, West Sussex, United Kingdom). The X-ray tube had Mg K radiation, and the analyzing mode was CAE. The following parameters were set for region spectra: maximum count rate, 29,905/s; analyze, 20 eV; step size, 0.10 eV; dwell time, 200 ms; number of channels, 201; number of scans, 10; time for region, 402 s.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Production of nitrile-degrading enzymes. R. rhodochrous NCIMB 11216 produced nitrile-degrading enzymes (nitrilase and/or the nitrile hydratase/amidase enzyme system), as indicated by the formation of propionic acid when the bacteria were grown on culture medium containing propionitrile. During cultivation (24 h), 10% of the propionitrile present was transformed into propionic acid.

1

1

1

1

Enzymatic treatment of PANs. The R. rhodochrous enzymes were able to hydrolyze nitrile groups of granular PAN, as indicated by the release of ammonia. Less ammonia was released from a high-molecular-mass PAN (190 kDa; PAN190) (Fig. 1).

View larger version (11K): [in this window] [in a new window]   FIG. 1.   Hydrolysis of granular PANs with R. rhodochrous enzymes at different concentrations after 72 h of incubation (expressed as a percentage of total possible conversion; 0.36 and 0.0036 g of enzyme g of PAN1; control, 0.36 g of heat-inactivated enzyme g of PAN1).

View larger version (21K): [in this window] [in a new window]   FIG. 2.   Dyeing of enzymatically pretreated acrylic fibers with methylene blue and Coomassie brilliant blue G and effect of posttreatment with Na2CO3 (control, heat-inactivated enzyme).

1

1

1

View larger version (27K): [in this window] [in a new window]   FIG. 3.   Changes in the elementary composition of enzymatically treated acrylic fibers (determined by XPS analysis; error bars indicate values from two separate experiments).

View larger version (14K): [in this window] [in a new window]   FIG. 4.   XPS analysis of acrylic fibers. A typical peak survey diagram is shown. KLL peaks are due to the Auger effect.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

R. rhodochrous NCIMB 11216 can transform a wide range of nitriles (11, 13). In this paper, we report for the first time that the enzymes of R. rhodochrous can also hydrolyze surfacial nitrile groups of PAN fibers and granulates. R. rhodochrous hydrolyzed monomeric acrylonitrile and acrylamide with nitrile hydratase and amidase, respectively, when grown in the presence of propionitrile. Recently, a nitrilase from R. rhodochrous NCIMB 11216 hydrolyzing aliphatic nitriles has been described which was produced during cultivation of the organism in the presence of propionitrile and benzonitrile (11). Previously, other authors reported the existence of both a nitrile hydratase/amidase system and a nitrilase in R. rhodochrous NCIMB 11216 (13).

On acrylonitrile, the nitrile hydratase activity of the R. rhodochrous enzyme preparation was 14.2 nkat mg¹ (cell dry weight). Previously, on the same substrate, nitrilase activities of 11 and 63 nkat mg¹ were measured for R. rhodochrous NCIMB 11216 and J1, respectively, when these strains had been cultivated in the presence of caprolactam (24). However, this "nitrilase" activity on acrylonitrile could also result from the cooperative action of nitrile hydratase and amidase because the intermediately formed amides cannot be detected if instantly hydrolyzed by amidase. In this case, the presence of nitrile hydratase can be determined if amidase is selectively inhibited (17). We found that the nitrile hydratase activity of R. rhodochrous NCIMB was significantly higher than amidase activity (1.7 nkat mg¹ on acrylamide), which is in contrast to observations for enzymes from R. erythropolis (17).

Hydrolysis of nitrile groups of PANs with enzymes from R. rhodochrous was studied with two granular PAN standards (40 and 190 kDa) and commercial acrylic fibers containing vinyl acetate as a copolymer. Interestingly, nitrile groups of both PAN40 and PAN190 were partially converted to the corresponding acid while nitrile groups of acrylic fibers were only hydrolyzed to the amide. Thus, nitrile hydratases hydrolyzed surfacial nitrile groups of acrylic fibers but the resulting amides were obviously not accessible to amidases. In agreement with these results, amidases from R. rhodochrous NCIMB 11216 and AJ270 have been reported to be generally more sensitive to the geometry of the substrate than nitrile hydratases (5, 20). Esterases from R. rhodochrous have previously caused problems during biotransformations of nitriles with ester function (13). Nevertheless, we have not observed any concurrent hydrolysis of vinyl acetate by esterases during hydrolysis of surfacial nitrile groups of acrylic fibers.

Pretreatment of acrylic fibers with R. rhodochrous enzymes improved fabric-dyeing efficiency, as indicated by K/S value increases. However, in the case of methylene blue, posttreatment of the fabrics with sodium carbonate to remove adsorbed enzymes seemed to be necessary to gain leveling. Nonleveling during dyeing is caused by adsorption of dye to the fiber-bound enzymes. This phenomenon depends both on the type of enzyme and on the dyeing conditions, as we have previously shown for indigo backstaining during cellulose washing (3).

Surfacial nitrile groups of acrylic fibers were hydrolyzed by the R. rhodochrous enzymes to a maximum of only 16%, although acrylonitrile is known to be a very good substrate for both nitrilases and nitrile hydratases from these organism (15, 25). However, the enzyme reaction on PAN is restricted by several factors related to the properties of the polymer. The structure of PAN is assumed to have a rigid, irregularly helical conformation of the polymer chain including planar zigzag packing (Fig. 5). The cylinders, with a diameter of 6 nm, can bind to each other through the antiparallel orientation of the side groups. In the model of Warner, fibers are composed of fibrillar subunits containing distinct regions of amorphous and partially ordered material (30).

View larger version (16K): [in this window] [in a new window]   FIG. 5.   Model of PAN structure. Reprinted with permission from the original publisher (7).

Enzyme adsorption to and desorption from the polymer affect the hydrolysis rate. Additionally, the crystallinity and hydrophobicity of PAN fibers may limit the accessibility of nitrile groups to the enzyme. Similarly, the structure and properties of cellulose influence its enzymatic degradation. It has been suggested that endoglucanases randomly cleave cellulose into smaller fragments, generating new ends, which are then hydrolyzed endwise by the action of cellobiohydrolases. These latter enzymes are also thought to erode crystalline regions of cellulose, making them more susceptible to endoglucanase attack (31).

In contrast to that of synthetic PAN polymers, enzymatic hydrolysis of natural polymers like cellulose has been investigated for decades. The architecture of cellulose-degrading enzymes has been studied in detail, and the function of the active sites and binding domains related to specificities for cellulose has been elucidated. Based on this knowledge, these enzymes and genetically improved products have been successfully applied in the textile industry. Similarly, nitrile-degrading enzymes could have an immense potential for the improvement of PAN fibers. Thus, future investigations could focus on the reaction mechanisms of nitrilases and nitrile hydratases in relation to the structure and properties of PANs.