Variability in n-caprylate and n-caproate producing microbiomes in reactors with in-line product extraction

Abstract

Medium-chain carboxylates (MCCs) are used in various industrial applications. These chemicals are typically extracted from palm oil, which is deemed not sustainable. Recent research has focused on microbial chain elongation using reactors to produce MCCs, such as n-caproate (C6) and n-caprylate (C8), from organic substrates such as wastes. Even though the production of n-caproate is relatively well-characterized, bacteria and metabolic pathways that are responsible for n-caprylate production are not. Here, three 5 L reactors with continuous membrane-based liquid-liquid extraction (i.e., pertraction) were fed ethanol and acetate and operated for an operating period of 234 days with different operating conditions. Metagenomic and metaproteomic analyses were employed. n-Caprylate production rates and reactor microbiomes differed between reactors even when operated similarly due to differences in H₂ and O₂ between the reactors. The complete reverse β-oxidation (RBOX) pathway was present and expressed by several bacterial species in the Clostridia class. Several Oscillibacter spp., including Oscillibacter valericigenes, were positively correlated with n-caprylate production rates, while Clostridium kluyveri was positively correlated with n-caproate production. Pseudoclavibacter caeni, which is a strictly aerobic bacterium, was abundant across all the operating periods, regardless of n-caprylate production rates. This study provides insight into microbiota that are associated with n-caprylate production in open-culture reactors and provides ideas for further work.IMPORTANCEMicrobial chain elongation pathways in open-culture biotechnology systems can be utilized to convert organic waste and industrial side streams into valuable industrial chemicals. Here, we investigated the microbiota and metabolic pathways that produce medium-chain carboxylates (MCCs), including n-caproate (C6) and n-caprylate (C8), in reactors with in-line product extraction. Although the reactors in this study were operated similarly, different microbial communities dominated and were responsible for chain elongation. We found that different microbiota were responsible for n-caproate or n-caprylate production, and this can inform engineers on how to operate the systems better. We also observed which changes in operating conditions steered the production toward and away from n-caprylate, but more work is necessary to ascertain a mechanistic understanding that could be predictive. This study provides pertinent research questions for future work.

Keywords: bacteria microcompartments; caproate; caprylate; chain elongation; hexanoate; hydrogen; medium-chain carboxylate; octanoate; oxygen.

Fig 1

The RBOX pathway investigated in this study. The enzymes we examined in this study are highlighted in black boxes. The figure was modified with permission from Angenent et al. (21). RBOX pathway enzymes are ACAT, acetyl-CoA C-acyltransferase (Thiolase II); HAD, 3-hydroxy-acyl-CoA dehydrogenase; ECH, enoyl-CoA dehydratase; ACD, acyl-CoA dehydrogenase; EtfA/B, electron-transfer-flavoprotein subunit A/B; CoAT, acetyl CoA-transferase; TE, thioesterase; RNF, Rnf respiratory complex.

Fig 2

n-Caproate (A) and n-caprylate (B) total production rates (mmol C L⁻¹ d⁻¹) and the molar ratio of n-caprylate to n-caproate (C) in three reactors across three main periods (and 10 periods). Period divisions are explained in the methods. Error bars indicate the standard error for the measurements. *Legend indicates periods (Periods 1D, 3B, and 3C) in which biomass samples were collected from reactors for shotgun metagenomic analysis. R1–3 are Reactors 1–3.

Fig 3

Relative abundance of the top seven most dominant taxa of each reactor based on the Illumina 16S rRNA gene sequencing results on the genus level (A) and the species level (B) throughout the operating time. The first 75 days of the operating period were the startup period (light blue). The salmon, blue, and green shadings indicate Periods 1, 2, and 3, respectively; the stars indicate the metagenomic sampling time points.

Fig 4

The most abundant species (A) and genera (B) in the three reactors based on the shotgun metagenome analysis. After normalizing the read count for sample size, the heatmap shows the number of reads aligned to each taxon throughout different sampling points of the reactors. Only taxa with more than 12 k reads aligned are displayed. The names are ordered based on the NCBI taxonomy. The top of the plot shows the n-caprylate (blue) and n-caproate (green) volumetric production rates for Periods 1D, 3B, and 3C, respectively; color intensity is proportional to the production rates. R1–3 are Reactors 1–3.

Fig 5

Absence or presence of nine enzymes involved in the RBOX pathway (acronyms described previously) in reactor de-novo assembled metagenomes and proteomes as monitored by shotgun metagenomics analysis and proteomics. MAG taxonomy was assigned using gtdb-tk (r220). Enzyme acronyms were described in Fig. 1. A blue box denotes the presence in the metagenome, the letter P the presence in the metaproteome, and a white box without a P the absence in both the metagenome and metaproteome. MAGs identified to species level are depicted. A * indicates high-quality MAGs >90% complete and <5% contaminated (determined with CheckM).