Identification and manipulation of Neurospora crassa genes involved in sensitivity to furfural

Abstract

Background: Biofuels derived from lignocellulosic biomass are a viable alternative to fossil fuels required for transportation. Following plant biomass pretreatment, the furan derivative furfural is present at concentrations which are inhibitory to yeasts. Detoxification of furfural is thus important for efficient fermentation. Here, we searched for new genetic attributes in the fungus Neurospora crassa that may be linked to furfural tolerance. The fact that furfural is involved in the natural process of sexual spore germination of N. crassa and that this fungus is highly amenable to genetic manipulations makes it a rational candidate for this study.

Results: Both hypothesis-based and unbiased (random promotor mutagenesis) approaches were performed to identify N. crassa genes associated with the response to furfural. Changes in the transcriptional profile following exposure to furfural revealed that the affected processes were, overall, similar to those observed in Saccharomyces cerevisiae. N. crassa was more tolerant (by ~ 30%) to furfural when carboxymethyl cellulose was the main carbon source as opposed to sucrose, indicative of a link between carbohydrate metabolism and furfural tolerance. We also observed increased tolerance in a Δcre-1 mutant (CRE-1 is a key transcription factor that regulates the ability of fungi to utilize non-preferred carbon sources). In addition, analysis of aldehyde dehydrogenase mutants showed that ahd-2 (NCU00378) was involved in tolerance to furfural as well as the predicted membrane transporter NCU05580 (flr-1), a homolog of FLR1 in S. cerevisiae. Further to the rational screening, an unbiased approach revealed additional genes whose inactivation conferred increased tolerance to furfural: (i) NCU02488, which affected the abundance of the non-anchored cell wall protein NCW-1 (NCU05137), and (ii) the zinc finger protein NCU01407.

Conclusions: We identified attributes in N. crassa associated with tolerance or degradation of furfural, using complementary research approaches. The manipulation of the genes involved in furan sensitivity can provide a means for improving the production of biofuel producing strains. Similar research approaches can be utilized in N. crassa and other filamentous fungi to identify additional attributes relevant to other furans or toxic chemicals.

Keywords: Aldehyde dehydrogenases; CRE1; Furan; Furfural; Neurospora crassa; Pretreatment.

Fig. 1

Effect of furfural on growth of N. crassa wild-type and mutant strains in a 96-well plate assay, as measured at 490 nm. a Wells of a 96-well plate were inoculated with 4 × 10⁵ conidia of the wild-type strain in VgS media supplemented with different concentrations (15–75 mM) of furfural. b Delay in lag phase as expressed in the time required for detectable initiation of culture growth, compared to the control lacking furfural (15, 30 and 60 mM). c Inhibition in relative growth, as determined by O.D. of cultures exposed to furfural (15, 30 and 60 mM) relative to that of the control, 20 h after inoculation. Asterisks in b, c indicate a significant difference between the mutant strains and wild-type subjected to the same treatment (*P < 0.05, **P < 0.01, and ***P < 0.001 by Student’s t test). The wild-type control is shown in pink. The values represent the average of at least six biological replicates

Fig. 2

Conidial germination of wild-type and mutant strains as affected by exposure to furfural. Conidial germination of wild type, T112, T361 and ∆ncw-1, 4 (a) and 6 (b) h after inoculation. Asterisks in the figure indicate a significant difference between the mutant strains and wild type subjected to the same treatment (*P < 0.05, by Student’s t test). The values represent the average of three biological replicates

Fig. 3

In vitro depletion of furfural is dependent on carbon catabolic repression. Depletion of NADPH or NADH was monitored in vitro with free-cell extracts of N. crassa wild-type (a) or Δcre-1 (b) strains with 10 mM furfural as a substrate. The cultures were grown on Vogels media with 1.5% sucrose or CMC as a carbon source for 16 h, before proteins were extracted. Asterisks in the figure indicate a significant difference (*P < 0.05, by Student’s t test). The values represent the average of three biological replicates

Fig. 4

Summary of gene expression in N. crassa following exposure to furfural. Each sequence read was mapped to the Neurospora genome v12 using tophat v2.04 [67] and the expression of each gene was normalized as FPKM using cufflinks v2.02 [68] with a cutoff value of log2 ≥ 1.0 and an FDR-adjusted P value of ≤ 0.01. a Hierarchical clustering was performed on expression data from samples with cluster3.0 with blue representing low expression and yellow representing high expression. b Venn diagram of significantly differentially expressed genes in N. crassa following exposure to furfural and HMF. c All RNA-seq data sets were clustered by Euclidean distance using a variance-stabilizing transformation module in DEseq v1.14 [82]. Darker colors represent higher similarity between replicates

Fig. 5

GO terms associated with biological processes affected upon exposure to furfural in N. crassa. GO terms were associated with all protein coding genes using the program Blast2Go v 2.8 [70] and a multilevel pie chart generated from genes that are up regulated with a log2 value ≥ 1.0 and P-adjusted value ≤ 0.01 and a node score > 6.0. Only terminal nodes are presented in the pie chart with the node score for each biological process

Fig. 6

Schematic representation of plasmid insertion in transformants T112 (a) and T361 (b)