Proteomics unveil a central role for peroxisomes in butyrate assimilation of the heterotrophic Chlorophyte alga Polytomella sp

Abstract

Volatile fatty acids found in effluents of the dark fermentation of biowastes can be used for mixotrophic growth of microalgae, improving productivity and reducing the cost of the feedstock. Microalgae can use the acetate in the effluents very well, but butyrate is poorly assimilated and can inhibit growth above 1 gC.L^-1. The non-photosynthetic chlorophyte alga Polytomella sp. SAG 198.80 was found to be able to assimilate butyrate fast. To decipher the metabolic pathways implicated in butyrate assimilation, quantitative proteomics study was developed comparing Polytomella sp. cells grown on acetate and butyrate at 1 gC.L^-1. After statistical analysis, a total of 1772 proteins were retained, of which 119 proteins were found to be overaccumulated on butyrate vs. only 46 on acetate, indicating that butyrate assimilation necessitates additional metabolic steps. The data show that butyrate assimilation occurs in the peroxisome via the β-oxidation pathway to produce acetyl-CoA and further tri/dicarboxylic acids in the glyoxylate cycle. Concomitantly, reactive oxygen species defense enzymes as well as the branched amino acid degradation pathway were strongly induced. Although no clear dedicated butyrate transport mechanism could be inferred, several membrane transporters induced on butyrate are identified as potential condidates. Metabolic responses correspond globally to the increased needs for central cofactors NAD, ATP and CoA, especially in the peroxisome and the cytosol.

Keywords: heterotrophy; metabolic pathways; microalgae; quantitative proteomics; volatile fatty acids.

Figure 1

Parameters of the growth of Polytomella sp. on acetate and butyrate. (A, B) Growth curves (g_X.L⁻¹) and substrate consumption (g_CS.L⁻¹) in presence of 1 g_CS.L⁻¹ acetate or butyrate, (C) biomass yields (g_X.g_CS⁻¹) and growth rates (d⁻¹) derived from these growth curves and (D) the concentrations of sugars and lipids are plotted on the left (g.L⁻¹) while their proportions to the total biomass are plotted on the right (g.g_X⁻¹). Arrows indicate when biomass for further proteomics analysis was sampled. X, dry weight; CS, dissolved organic carbon. Error bars correspond to standard deviations based on 3 biological replicates.

Figure 2

Comparison of global proteomes of Polytomella sp. grown on acetate and butyrate. (A) Total proteins (30 μg) from exponentially growing cells, resolved in a 12% SDS-polyacrylamide gel stained with Coomassie Blue G250. (B) Volcano plot displaying the differential abundance of proteins of both conditions analysed by MS-based quantitative proteomics. The volcano plot represents the -log10 (limma value of p, cut off 0.004) on y-axis plotted against the log2 (FoldChange acetate/butyrate) on the x-axis. Green and red dots represent proteins more abundant in, respectively, the acetate or butyrate conditions (Benjamini-Hochberg FDR < 1%). The Venn diagram indicates that of the total of 1,772 proteins detected for acetate and butyrate combined, 46 were significantly induced on acetate and 119 on butyrate.

Figure 3

Overview of the Polytomella proteome revealed by the differential approach, represented per metabolic category as determined by Mercator. (A) Total number of proteins identified in both acetate and butyrate-growing cells. (B) Cumulative iBAQ values of total acetate and butyrate proteomes give an indication of the total protein abundance per category. The values associated to the graphs are indicated in the legend-table.

Figure 4

Number of proteins per category that are significantly more induced in one condition. Categories with a significant difference in the number of proteins between the acetate or the butyrate condition with respect to the background in the Fisher test are indicated with one asterisk (value of p<0.05), two asterisks (p < 0.01) or three asterisks (p < 0.001). Note that some categories do not contain any differentially expressed proteins.

Figure 5

Schematic representation of cellular metabolism as a function of metabolic (sub) categories and log2 FoldChange values using the program MapMan, based on the results of Mercator. For all entries with a FC score a value of p <0.004 applies. Note that for increased visibility of the lower range of the log2 FC scale was set from 5 to −5 whereas a few proteins on butyrate actually show higher FC values.

Figure 6

Proposed metabolic reconstruction of the assimilation pathway of butyrate based on FC values and targeting predictions. The log2 FC acetate/butyrate value is indicated by the color codes. All di/tricarboxylic acids that may be imported into the mitochondria are indicated in blue. Enzymes with a red halo are involved in antioxidant defense. Enzymes codes and further information can be found in Table 1 and the Supplementary Table 1.

Figure 7

Metabolic reconstruction of the branched amino acid degradation pathway in the peroxisome and proposed interactions with the mitochondria. The log2 FC acetate/butyrate is indicated by the color codes. Further information can be found in Table 1. Arrows in grey represent less likely or hypothetical pathways, pathways that are proposed use black arrows. ACSF exhibits an FC value with p > 0.004. ALD color gradient indicates different isoforms with FC values between 0 and 6.