Elsevier

Bioresource Technology

Volume 281, June 2019, Pages 260-268
Bioresource Technology

Modeling growth and fermentation inhibition during bioethanol production using component profiles obtained by performing comprehensive targeted and non-targeted analyses

https://doi.org/10.1016/j.biortech.2019.02.081Get rights and content

Highlights

  • A metabolomics approach estimated bioethanol production in lignocellulosic hydrolysates.

  • The calculation revealed behaviors of minor inhibitory compounds in hydrolysates.

  • Minor inhibitory compounds were observed in the hydrolysates.

  • The methodology can clue understanding behaviors of inhibitors in hydrolysates.

Abstract

Corn cob and corn stover hydrolysates are forms of lignocellulosic biomass that can be used in second generation bioethanol production and biorefinery processes. Growth and fermentation inhibitors generated during physicochemical and enzymatic hydrolysis decrease ethanol and biomaterial production during the subsequent biological processes. Here, estimates of growth and fermentation inhibition during bioethanol fermentation were made using component profiles of corn cobs and corn stover at different degrees of hydrolysis. The component profiles were acquired by non-targeted gas chromatography mass spectrometry and targeted high-performance liquid chromatography. Correlations between the comprehensive analysis results and yeast growth and ethanol production were modeled very accurately by partial-least-squares regression analysis. Acetate, apocynin, butyrovanillone, furfural, furyl hydroxymethyl ketone, m-methoxyacetophenone, palmitic acid, syringaldehyde, and xylose, were compounds with very variable importance in projection values and had negative correlation coefficients in the model. In fact, methoxyacetophenone, apocynin, and syringaldehyde inhibited fermentation more than furfural in equivalent concentration.

Introduction

Lignocellulosic biomass is a promising feedstock for biofuel and other biochemical production to improve sustainability and the bioeconomy (Devappa et al., 2015, Macerelli et al., 2012, Philp, 2018), and consists of cellulose and hemicellulose encapsulated in a lignin matrix (Sannigrahi et al., 2010). The high degree of crystallinity prevents lignocellulosic biomass being converted into biofuel or other biochemicals through physicochemical and microbial processes. Various physical, physicochemical, chemical, and biological pretreatments have been used to attempt to allow lignocellulosic biomass to be converted into biofuel (Zhong et al., 2018, Moizer et al., 2005, Konishi et al., 2015). Hydrolysis of lignocellulosic biomass generates sugars that can be used by microorganisms but also toxic compounds (depending on the intensity of the hydrolysis pretreatment) that can inhibit subsequent fermentation (Modenbach and Nokes, 2012). The hydrolysates of lignocellulosic biomass include numerous toxic compounds, including weak acids, aromatic compounds, and furan aldehydes, and different mixtures of compounds are produced under different pretreatment conditions and using different feedstocks (Konishi et al., 2015, Zha and Punt, 2013). Furfural and 5-hydroxymetyl furfural (5-HMF), categorized in furan aldehydes that inhibit microbes, are produced in relatively large amounts during the hydrolysis of xylose and glucose, respectively, during saccharification (Jönsson et al., 2013, Palmqvist and Hahn-Hägerdal, 2000b). 5-HMF can be further hydrolyzed to formate, a weak acid (Jönsson et al., 2013, Palmqvist and Hahn-Hägerdal, 2000b). Another weak acid, acetate, is derived from acetylated saccharides, which are major constituents of hemicellulose. Phenolic compounds such as catechol, 4-hydroxybenzoic acid, and vanillin are derived from lignin (Palmqvist and Hahn-Hägerdal, 2000b). These compounds and other compounds affect, together or separately, microbe growth and metabolic pathways. Few attempts have been made to investigate the actual behaviors of inhibitory materials produced through the hydrolysis of lignocellulosic biomass (Zha and Punt, 2013).

Second-generation biofuel production and biorefining using lignocellulosic biomass is important because lignocellulosic biomass (produced in large quantities during forestry, agriculture, and agroindustry activities) is the most abundant, cheap, and renewable resource that does not interfere with the food supply (Capolupo and Faraco, 2016, Macerelli et al., 2012). However, individual lignocellulosic biomass feedstocks cannot meet future demand, so biofuel and biorefining facilities need to be able to use various feedstocks, depending on feedstock availability including seasonal and geographical effects (Ahorsu et al., 2018, Zhang et al., 2018).

Metabolomics is a powerful tool for differentially analyzing intracellular metabolites and fingerprinting natural materials, including foods. Metabolomics has been used to predict the geographical origins of hazelnuts (Klockmann et al., 2017), to rank Japanese green tea (Ikeda et al., 2007, Jumtee et al., 2011) and sake (Mimura et al., 2014), and to identify markers to authenticate Asian palm civet coffee (Jumhawan et al., 2013). The effects of yeast and lactic acid bacteria on soy sauce fermentation have been studied by non-targeted metabolic profiling using gas chromatography mass spectrometry (GC–MS) (Harada et al., 2017). The approach can be applied to estimating the microbial growth from the components of media.

In this study, it was investigated that the effects of inhibitory hydrolysates on yeast growth and fermentation during bioethanol production (as a model process) using a metabolomics-like approach involving comprehensive instrumental and multivariate statistical analyses. It was assessed the behaviors of major inhibitory materials and the additive effects of minor inhibitory materials in actual lignocellulosic biomass hydrolysates.

Section snippets

Microorganisms and chemicals

Saccharomyces cerevisiae NBRC1136 (synonym S288c) was purchased from the National Institute of Technology and Evaluation Biological Resource Center, Japan. Glycerol stock was prepared by cultivating the strain for 48 h in YM medium (containing 10 g/L Glc, 3 g/L yeast extract, 3 g/L malt extract, and 5 g/L peptone (all from BD Biosciences, Franklin Lakes, NJ, USA)) at 28 °C. The culture was then stored in 20% glycerol at −80 °C. The glycerol stock was used in all the experiments.

Dried and

Biomass compositions

The sugar, lignin, and ash contents of the lignocellulosic biomass samples used are shown in Table 2. The xylan and ash (possibly derived from hemicellulose) contents were slightly higher for the Chinese corn cobs than the other corn cobs. The glucan content was slightly higher for the Indonesian corn cobs than the other corn cobs. The corn stover contained large proportions (up to 54% w/w) of glucan. It may be affect that fresh green biomass had been harvested and immediately dried. The

Conclusions

It was demonstrated that a method combining comprehensive non-targeted and targeted chemical analyses and PLS-R modeling successfully estimated the results of culture experiments. VIP scores and correlation coefficients were used to assess the behaviors of the hydrolysate components. The results suggested that inhibitory compounds in hydrolysates synergistically inhibit growth and fermentation during bioethanol production, and low concentrations but important inhibitors such as apocynin and m-

Acknowledgements

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. We thank Gareth Thomas, PhD, from Edanz Group (www.edanzediting.com/ac) for editing a draft of this manuscript.

References (34)

  • R. Ahorsu et al.

    Significance and challenges of biomass as a suitable feed stock for bioenergy and biochemical production: a review

    Energies

    (2018)
  • L. Capolupo et al.

    Green methods of lignocellulose pretreatment for biorefinery development

    Appl. Microbial. Biotechol.

    (2016)
  • T. Ikeda et al.

    Prediction of Japanese green tea ranking by Fourier transform near-infrared reflectance spectroscopy

    J. Agric. Food Chem.

    (2007)
  • R. Harada et al.

    Microbe participation in aroma production during soy sauce fermentation

    J. Biosci. Bioeng.

    (2017)
  • L.J. Jönsson et al.

    Bioconversion of lignocellulose: inhibitors and detoxification

    Biotechnol. Biofuels

    (2013)
  • U. Jumhawan et al.

    Selection of discriminant markers for authentication of Asian palm civet coffee (Kopi Luwak): a metabolomics approach

    J. Agric. Food Chem.

    (2013)
  • D. Kim

    Physico-chemical conversion of lignocellulose: inhibitor effects and detoxification strategies: a mini review

    Molecules

    (2018)
  • Cited by (14)

    • Recent advances in lignocellulosic biomass white biotechnology for bioplastics

      2022, Bioresource Technology
      Citation Excerpt :

      In microbial production from lignocellulosic biomass, fermentation inhibition is a huge problem to be solved. Lignocellulosic hydrolysate contains different contaminants, and some of them inhibit cell growth or fermentation, causing reduced target products (Watanabe et al., 2019). Although the detoxification of the hydrolysates is effective in attenuating fermentation inhibition (Liu et al., 2020), the additional treatment increases the total cost of microbial production.

    • Enhancement of bioethanol production by a waste biomass-based adsorbent from enzymatic hydrolysis

      2021, Journal of Cleaner Production
      Citation Excerpt :

      The pretreatment of lignocellulosic biomass is necessary to reduce its recalcitrance to subsequent enzymatic hydrolysis and microbial fermentation (Zhang et al., 2019). The hydrolysis of pretreated biomass results in a large number of byproducts, including sugar-derived furans, lignin-derived phenols, and some weak acids (Watanabe et al., 2019), with inhibitory effects on the subsequent fermentation process. Their toxicity depends on the lignocellulose type and pretreatment conditions (Nandal et al., 2020).

    • Bioethanol production estimated from volatile compositions in hydrolysates of lignocellulosic biomass by deep learning

      2020, Journal of Bioscience and Bioengineering
      Citation Excerpt :

      For generating the score plots, all data were used for the modeling. In the first step, Indonesia corncobs hydrolysates were used as validation data, and the others as supervised data for DNN modeling because good accuracy has been observed by PLS (12). To search the practical DNN structure, several DNN structures were tested for estimating overall yeast growth and ethanol production.

    View all citing articles on Scopus
    View full text