A high-throughput screening assay for distinguishing nitrile hydratases from nitrilases.

A modified colorimetric high-throughput screen based on pH changes combined with an amidase inhibitor capable of distinguishing between nitrilases and nitrile hydratases. This enzymatic screening is based on a binary response and is suitable for the first step of hierarchical screening projects.

Nitrile metabolizing enzymes are of outstanding synthetic importance because of their ability to afford amides, carboxylic acids, amines and cyanohydrins, compounds used as building blocks in the pharma and chemical industries ( van Pelt 2011 , Chen 2009 , Lin 2012 , Yamada 2001 ). They can also be potent biocatalysts for detoxification of anthropogenic toxic nitrile pollutants that are widespread throughout the world ( Vesela 2012 , Vesela 2010 ). Among them, nitrile hydratase is an interesting commercial enzyme already used successfully on a ton scale for the production of commodities and pharmaceuticals such as acrylamide and nicotinamide ( Prasad 2010 , Kobayashi 2000 , Yamada 2001 ). Nitrile hydratase was first discovered in 1980 by Yasano and collaborators and it is believed that for the most part, they are involved in a cascade reaction with amidases, affording carboxylic acids from nitriles passing through an amide intermediate ( Yasano 1980 ). Nitrilases are also present in many different species and afford a carboxylic acid directly from a nitrile compound ( Prasad 2010 ) ( Figure 1 ).

Figure 1

Natural pathways for enzymatic conversion of nitriles to carboxylic acids.

A number of screening assays for nitrile-converting enzymes based on continuous and stopped methods are well documented in the literature ( Asano 2002 , Martinkova 2008 ; Reisinger 2006 , Santoshkumar 2010 ; He 2011 ; Zheng 2011, Yazbeck 2006 , Wang 2012 ). However, as nitrilases and nitrile hydratase-amidases afford the same final product, it is important to design a screening assay able to distinguish between the two enzymatic pathways.

Herein, we describe a colorimetric high-throughput screening assay based on pH changes coupled with the use of an amidase inhibitor. This screen is based on a binary response allowing differentiation between nitrilases and nitrile hydratase-amidases enzymatic systems and is suitable for the first step of hierarchical screening projects.

A Banerjee-modified colorimetric and pH sensitive assay coupled with an amidase inhibitor was performed for screening nitrilase and nitrile hydratase-amidase enzymes. Commercially available microorganisms potentially containing the nitrile hydratase and nitrilase enzymatic systems were used as positive controls. All nitriles and their corresponding amides and carboxylic acids and an amidase inhibitor were evaluated to detect any possible color change interferences within the enzymatic assay system. It was assumed that the strains that did not accumulate the amide during nitrile degradation indicated nitrilase activity. The intermediate accumulation of the corresponding amide during nitrile metabolism combined with carboxylic acid formation was taken as an indication of the existence of a nitrile hydratase-amidase system ( Layh 1997 ). The expression of nitrile hydratases was induced by acetonitrile or benzonitrile for aliphatic and aromatic nitriles, respectively. Mandelonitrile was too toxic to the microorganisms prior to enzymatic induction therefore it was not used as an inducing agent. The well-known amidase inhibitor diethyl phosphoramidate, DEPA ( Bauer 1998 ), was chosen for this screening since its color did not affect the assay readout and also it is not influenced by pH changes during the course of the assay. The use of an amidase inhibitor permitted the accumulation of the amide intermediate, thus allowing the discrimination between nitrile hydratase-amidases and nitrilases when only one of these enzymatic systems was present ( Brady 2004 ). However, when the microorganism had both enzymatic systems, it was not possible to reach a definitive conclusion. Moreover, a microbial control experiment is important since the production and/or excretion of acidic metabolites into the extracellular media in concentrations high enough to cause color changes in the pH indicator may compromise the assay validity. The screening assays could be monitored by simple microtiter plate visual inspection ( Figure 2 ). Additionally, a colorimetric master plate was used as reference color scale.

Figure 2

Screening for nitrile hydratase and nitrilase producing strains using mandelonitrile as a substrate in a microplate. Row A: control experiments: A1–A3 mandelonitrile, A4–A6: mandelamide, A7–A9: mandelic acid; A10–A12: DEPA. A: time zero; B: time = 36 h. * Photo was taken from the bottom up. Rows B, C, D, F, G and H: different isolated microorganisms were checked for their ability to hydrolyze mandelonitrile

As expected, Pseudomonas putida CCT 2357 and Pseudomonas fluorescens CCT 3178 are exclusively nitrilase producing strains ( Table 1 ). This result is supported by evidence from the literature ( Chen 2009 , Prasad 2010 ) and can be rationalized by the maintenance of yellow color in the presence or absence of amidase inhibitor. On the other hand, Nocardia simplex CCT 3022 produces only nitrile hydratase-amidase enzymes since the carboxylic acid formation was detected in the experiment without amidase inhibitor but not in the assay with the addition of DEPA. If a nitrilase was present, color change would be expected in the experiment with DEPA addition, however, no color change was observed. The strains Rhodococcus ruber CCT 1879, Rhodococcus equi CCT 0541, Rhodococcus erythropolis CCT 1878 and Nocardia brasiliensis CCT 3439 produce both a nitrilase and nitrile hydratase-amidase. In this case, the screening assay cannot give a conclusive response. Experiments were completed in the presence of a nitrilase inhibitor (AgNO3), however, they were unsuccessful because of the high sensitivity of AgNO3 to air and the formation of colored precipitate. The results obtained with Arthrobacter sp CCT 1875 show the importance of microbial control. This screening assay is not suited to cases where acidic metabolites are produced or excreted into extracellular media in concentrations high enough to cause a pH change above the buffer capacity.

Table 1

Enzymatic assay results for detection of nitrilase, nitrile hydratase and amidase activities.

Microorganism	Microbial control	Assay without DEPA	Assay with DEPA
Pseudomonas putida CCT 2357	blue	yellow	yellow
Pseudomonas fluorescens CCT 3178	blue	yellow	yellow
Arthrobacter sp CCT 1875	yellow	yellow	yellow
Nocardia brasiliensis CCT 3439	blue	blue	yellow
Nocardia simplex CCT 3022	blue	yellow	greenish
Rhodococcus erythropolis CCT 1878	blue	blue	yellow
Rhodococcus equi CCT 0541	blue	blue	yellow
Rhodococcus ruber CCT 1879	blue	blue	greenish

The control experiments with acetonitrile, DEPA and acetamide compounds as well as with the microbial cultures alone are blue. Acetic acid and other carboxylic acids are yellow. A greenish tint to the blue or yellow indicates a small drop in pH due to formation of acetic acid.

The reactions were performed on a 10 mL scale in an Erlenmeyer flask and monitored by GC-FID to confirm the results obtained with the microtiter plate assay ( Figures S1 and S2: Supporting Information ). No false positives or false negatives were detected.

To confirm that this method can be used for screening of nitrile hydratase-producing microorganisms, we used it to screen our in house wild type microorganism library. These microorganisms were previously isolated from different sources (cassava waste water, cassava, benzonitrile herbicide impacted soils and UNESP garden) by enrichment techniques. From a library of 130 microorganisms isolated from nitrile-impacted and non-impacted areas, 19 positive hits were detected, of which 12 were nitrilase producers, 3 were NHase-amidase producers and 5 produced both enzymatic systems.

Interestingly, all microbial cultures which afforded positive hits came from nitrile-impacted areas. These results suggest the importance of natural microbial acclimation and also suggest a strategy for enzyme identification. The enzymatic assay was capable of detecting and distinguishing nitrile hydratases from nitrilases when only one was produced in a bacterial strain. The enzymatic assay was validated by GC-FID monitoring. Nitrile maximum inhibitory concentrations (MIC) were determined using a MTT assay. The MIC values varied for each tested nitrile and microbial culture. The wild type microorganisms are currently being subjected to 16S RNA identification and show promise as whole cell biocatalysts.