Elsevier

Bioresource Technology

Volume 100, Issue 11, June 2009, Pages 2823-2827
Bioresource Technology

A novel assay for rapid in vivo determination of phenotypic stability of recombinant ethanol-producing microorganisms

https://doi.org/10.1016/j.biortech.2009.01.013Get rights and content

Abstract

A rapid empirical assay is presented for assessing the phenotypic stability of continuous cultures of recombinant bacteria containing transposed pdc and adh genes for ethanol production. The method measures spectrophotometrically the rate of colour formation when cells oxidize added ethanol to acetaldehyde in the presence of Schiff’s reagent. During chemostat cultures of the recombinant ethanologen Escherichia coli KO11 on 20 g/l glucose, assay activities were stable and high at ca 8 × 10−4 ΔOD540/(s.OD550), reflecting the high, stable ethanol yield (ca 95%). On 20 g/l and 50 g/l xylose, ethanol yields declined rapidly to about 60% and this was closely mirrored by the assay activities which fell to ca 1.5 ΔOD540/(s.OD550), only slightly higher than those measured for the parent strain. Typically taking only about an hour to perform, the assay provides a faster means of gauging the phenotypic stability of ethanol production than is possible by conventional methods.

Introduction

The production of ethanol from fermentation of renewable lignocellulosic biomass such as wood has the potential to augment world liquid fuel supplies and reduce net CO2 generation without competing with food production (Gomez et al., 2008). However for the process to be economic, the entire carbohydrate content of the substrate needs to be fermentable to ethanol. This includes the hemicellulose sugars which, although accounting for as much as 40% of the total carbohydrate, are mainly unfermentable by traditional yeast strains (Galbe et al., 2007). Recently several different approaches have been adopted to construct recombinant microorganisms that can successfully ferment these sugars (Jarboe et al., 2007, Jeffries, 2006). This work focuses on one such approach, which has entailed inserting the pyruvate decarboxylase (pdc) and alcohol dehydrogenase II (adhB) genes from the highly ethanologenic bacterium Zymomonas mobilis into bacteria that do not normally produce ethanol at high-yield but which can take up the hemicellulose sugars. When expressed at high levels, the transposed genes divert pyruvate metabolism from acid production to high-yield ethanol production (Doran Peterson and Ingram, 2008). This technique has been successfully implemented with several different bacteria, including Escherichia coli (Dien et al., 2000, Ingram et al., 1987), Klesbiella oxytoca (Ohta et al., 1991a, Ohta et al., 1991b) and Erwinia strains (Beall and Ingram, 1993).

To increase productivity in fuel ethanol production, continuous fermentation is widely employed in large-scale starch-to-ethanol plants. In a continuous lignocellulose-to-ethanol process, the long-term stability of the recombinant strain is a matter for concern, particularly if there is a potential for the transposed ethanol-producing genes to be lost or for their expression to diminish. This raises the need for laboratory techniques capable of rapidly assessing the stability of a culture during fermentation. Existing attempts to evaluate stability in recombinant ethanol fermentations have either been based on plate-counting to determine the extent of retention of antibiotic markers (Dien et al., 2000, Ohta et al., 1991a, Ohta et al., 1991b, Yomano et al., 1998) or have relied on daily measurements of the fermenter ethanol concentration as an indicator of phenotypic stability (Dien et al., 2000, Dumsday et al., 1999, Martin et al., 2006). Both methods are slow, due to the long times required to incubate plate cultures and the usual need to monitor the ethanol concentration for several days before any trend to decreasing ethanologenicity can be definitively identified.

The aim of this work was to develop a rapid empirical method for assessing the phenotypic stability of continuous cultures of recombinant bacteria containing transposed pdc and adh genes for ethanol production. Ideally such a procedure would be able to predict any impending loss of ethanol-producing ability without the need for plate incubations, colony counting or protracted monitoring of the fermenter ethanol concentration and would be performable in as little as 1 h of assay time.

The method proposed (termed here the RBS or rosanilin bisulphite assay) is a quantitative spectrophotometric assay based on the chemistry of the aldehyde indicator plate technique used qualitatively by Ingram and colleagues to detect high ethanol-producing colonies of their ethanologenic E. coli strains (Conway et al., 1987, Ingram et al., 1990, Ingram and Conway, 1992, Ingram et al., 1987, Keshav et al., 1990, Ohta et al., 1991a, Ohta et al., 1991b). The method provides an empirical measure of the overall alcohol dehydrogenase activity of liquid culture samples, based on the rate of red colour formation when added ethanol is oxidized to acetaldehyde in the presence of Schiff’s reagent. The procedure was tested using the recombinant E. coli strain KO11 developed by Ingram and colleagues for commercial ethanol production from lignocellulose (Ohta et al., 1991a, Ohta et al., 1991b). This strain is well suited to this purpose since, while producing high ethanol yields in batch culture (Dien et al., 2003, Doran Peterson and Ingram, 2008, Hahn-Hägerdal et al., 1994, Lima et al., 2002, Rao et al., 2007), it shows progressively declining yields in continuous culture when cultivated on sugars other than glucose (Dumsday et al., 1997, Dumsday et al., 1999, Zhou et al., 2008), presumably due to reduced expression or loss of the transposed ethanol-producing genes (Dumsday et al., 1999).

Section snippets

Strain

Escherichia coli strains KO11 and ATCC 11303 were obtained from Professor L.O. Ingram (University of Florida, Gainesville, FL, USA) and the American Type Culture Collection (Rockland, MD), respectively, and were maintained lyophilised in ampoules. A fresh ampoule was used to start a new culture for each experiment.

Media

The base medium was modified Luria broth which contained: tryptone (Oxoid, Australia), 10 g/l; yeast extract (Oxoid, Australia), 5 g/l; and NaCl, 5 g/l. The carbon source was either d

Development of the liquid phase RBS assay

The chemistry of the RBS assay is based on the aldehyde indicator plate assay devised by Conway et al. (1987) to identify high ethanol-producing colonies of a recombinant E. coli strain containing the Z. mobilis pdc and adh genes. The method employs Schiff’s reagent, formed from pararosanilin and sodium bisulphite (Lillie, 1977). This produces an intense red colour in the presence of aldehydes and is a classic test for them (Vogel, 1989). In the method of Conway et al. (1987),

Discussion

The above data show that the RBS assay is successful as an empirical indicator of changes in ethanologenicity during long-term continuous growth by E. coli KO11. The assay can typically be performed in less than 1 h, providing a much more rapid indication of any change in phenotype than can be obtained using conventional methods. Unlike conventional assays of alcohol dehydrogenase activity that measure the reduction of NAD+ (e.g. Conway et al., 1987) the RBS method does not require cell lysis,

Conclusions

The RBS assay appears to be an effective empirical method for rapidly ascertaining changes in ethanologenicity of E. coli KO11. The assay should also be applicable to other strains containing recombined pdc and adh genes for ethanol production.

Acknowledgements

We are grateful to Dr Geoff Dumsday for helpful discussions, to Professor L.O. Ingram for the gift of the E. coli KO11 strain and to the Australian Energy Research and Development Corporation for financial support.

References (26)

  • T.W. Jeffries

    Engineering yeasts for xylose metabolism

    Curr. Opin. Biotechnol.

    (2006)
  • D.S. Beall et al.

    Genetic engineering of soft-rot bacteria for ethanol production from lignocellulose

    J. Ind. Microbiol.

    (1993)
  • T. Conway et al.

    Cloning and sequencing of the alcohol dehydrogenase II gene from Zymomonas mobilis

    J. Bacteriol.

    (1987)
  • B.S. Dien et al.

    Development of new ethanologenic Escherichia coli strains for fermentation of lignocellulosic biomass

    Appl. Biochem. Biotechnol.

    (2000)
  • B.S. Dien et al.

    Bacteria engineered for fuel ethanol production: current status

    Appl. Microbiol. Biotechnol.

    (2003)
  • J. Doran Peterson et al.

    Anaerobic respiration in engineered Escherichia coli with an internal electron acceptor to produce fuel ethanol

    Ann. N.Y. Acad. Sci.

    (2008)
  • G.J. Dumsday et al.

    Continuous ethanol production by Escherichia coli KO11 in continuous stirred tank and fluidized bed fermenters

    Australas. Biotechnol.

    (1997)
  • G.J. Dumsday et al.

    Comparative stability of ethanol production by Escherichia coli KO11 in batch and chemostat culture

    J. Ind. Microbiol. Biotechnol.

    (1999)
  • M. Galbe et al.

    Process engineering economics of bioethanol production

    Adv. Biochem. Eng./Biotechnol.

    (2007)
  • L.D. Gomez et al.

    Sustainable liquid biofuels from biomass: the writing’s on the walls

    New Phytol.

    (2008)
  • B. Hahn-Hägerdal et al.

    An interlaboratory comparison of the performance of ethanol-producing microorganisms in a xylose-rich acid hydrolyzate

    Appl. Microbiol. Biotechnol.

    (1994)
  • T.J. Hobley et al.

    Liquid injection gas chromatography and the ALDH assay overestimate free acetaldehyde in complex fermentation media

    Biotechnol. Tech.

    (1997)
  • Ingram, L.O., Conway, T., 1992. Cloning and sequencing of the alcohol dehydrogenase II from Zymomonas mobilis. U.S....
  • View full text