Polyhydroxyalkanoate synthesis by mixed microbial consortia cultured on fermented dairy manure: Effect of aeration on process rates/yields and the associated microbial ecology

Abstract

Polyhydroxyalkanoates (PHAs) are biodegradable polymers that can substitute for petroleum-based plastics in a variety of applications. One avenue to commercial PHA production involves coupling waste-based synthesis with the use of mixed microbial consortia (MMC). In this regard, production requires maximizing the enrichment of a MMC capable of feast-famine PHA synthesis, with the metabolic response induced through imposition of aerobic-dynamic feeding (ADF) conditions. However, the concept of PHA production in complex matrices remains unrefined; process operational improvements are needed, along with an enhanced understanding of the MMC. Research presented herein investigated the effect of aeration on feast-famine PHA synthesis, with four independent aeration state systems studied; MMC were fed volatile fatty acid (VFA)-rich fermented dairy manure. Regardless of the aeration state, all MMC exhibited a feast-famine response based on observed carbon cycling. Moreover, there was no statistical difference in PHA synthesis rates, with q_PHA ranging from 0.10 to 0.19 CmmolPHA gVSS^-1 min^-1; VFA uptake rates exhibited similar statistical indifferences. PHA production assessments on the enriched MMC resulted in maximum intracellular concentrations ranging from 22.5 to 90.7% (mgPHA mgVSS^-1); at maximum concentration, the mean hydroxyvalerate mol content was 73 ± 0.6%. While a typical feast-famine dissolved oxygen (DO) pattern was observed at maximum aeration, less resolution was observed at decreasing aeration rates, suggesting that DO may not be an optimal process monitoring parameter. At lower aeration states, nitrogen cycling patterns, supported by molecular investigations targeting AOBs and NOBs, indicate that NO₂ and NO₃ sustained feast-famine PHA synthesis. Next-generation sequencing analysis of the respective MMC revealed numerous and diverse genera exhibiting the potential to achieve PHA synthesis, suggesting functional redundancy embedded in the diverse MMC. Ultimately, results demonstrate that aeration can be controlled in waste-based ADF systems to sustain PHA production potential, while enriching for a diverse MMC that exhibits potential functional redundancy. Reduced aeration could also enhance cost competitiveness of waste-based PHA production, with potential further benefits associated with nitrogen treatment.

Keywords: ADF; Aerobic dynamic feeding; Next generation sequencing; Oxygen mass transfer coefficient; PHA; Polyhydroxyalkanoates; VFAs; Volatile fatty acids.

Figure 1

VFA and PHA profiles for the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Note that PHA reserves at the end of the SBR cycle were depleted.

Figure 1

VFA and PHA profiles for the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Note that PHA reserves at the end of the SBR cycle were depleted.

Figure 1

VFA and PHA profiles for the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Note that PHA reserves at the end of the SBR cycle were depleted.

Figure 1

VFA and PHA profiles for the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Note that PHA reserves at the end of the SBR cycle were depleted.

Figure 2

Residual dissolved oxygen profiles for the Enrichment reactors during the feast phase and into the famine phase for reactors operating at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d).

Figure 2

Residual dissolved oxygen profiles for the Enrichment reactors during the feast phase and into the famine phase for reactors operating at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d).

Figure 2

Residual dissolved oxygen profiles for the Enrichment reactors during the feast phase and into the famine phase for reactors operating at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d).

Figure 2

Residual dissolved oxygen profiles for the Enrichment reactors during the feast phase and into the famine phase for reactors operating at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d).

Figure 3

Example PHA production data (stage 1-B) for biomass from the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Operations were conducted to supply new substrate (VFA-rich fermenter liquor) to the biomass based on a sudden increase in dissolved oxygen.

Figure 3

Example PHA production data (stage 1-B) for biomass from the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Operations were conducted to supply new substrate (VFA-rich fermenter liquor) to the biomass based on a sudden increase in dissolved oxygen.

Figure 3

Example PHA production data (stage 1-B) for biomass from the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Operations were conducted to supply new substrate (VFA-rich fermenter liquor) to the biomass based on a sudden increase in dissolved oxygen.

Figure 3

Example PHA production data (stage 1-B) for biomass from the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Operations were conducted to supply new substrate (VFA-rich fermenter liquor) to the biomass based on a sudden increase in dissolved oxygen.

Figure 4

Example PHA Production data (stage 2-B) for biomass from the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Operations were conducted to supply new substrate (VFA-rich fermenter liquor) to the biomass every 30 minutes.

Figure 4

Example PHA Production data (stage 2-B) for biomass from the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Operations were conducted to supply new substrate (VFA-rich fermenter liquor) to the biomass every 30 minutes.

Figure 4

Example PHA Production data (stage 2-B) for biomass from the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Operations were conducted to supply new substrate (VFA-rich fermenter liquor) to the biomass every 30 minutes.

Figure 4

Example PHA Production data (stage 2-B) for biomass from the Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). Operations were conducted to supply new substrate (VFA-rich fermenter liquor) to the biomass every 30 minutes.

Figure 5

Nitrogen cycling in Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a) and 4 hr⁻¹ (b). Note that the abscissa scale begins at −10 minutes to reflect nitrogen concentrations before the cycle operations commenced.

Figure 5

Nitrogen cycling in Enrichment reactors operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a) and 4 hr⁻¹ (b). Note that the abscissa scale begins at −10 minutes to reflect nitrogen concentrations before the cycle operations commenced.

Figure 6

Relative abundance and taxonomic classification of the 16S rRNA gene sequencing results using the RDP for Enrichment reactors on operational day 271 and operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). The classified phylotypes are depicted in terms of the taxonomic hierarchy. Phylotypes which were not identified by the RDP or those whose identification at a specific taxonomic level was not statistically significant were aggregated, denoted “No ID”, and depicted in red. Identified phylotypes with less than 1% of the total relative abundance were aggregated, denoted “Minor”, and depicted in gray. Phylotypes with at least 1% relative abundance are labeled.

Figure 6

Relative abundance and taxonomic classification of the 16S rRNA gene sequencing results using the RDP for Enrichment reactors on operational day 271 and operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). The classified phylotypes are depicted in terms of the taxonomic hierarchy. Phylotypes which were not identified by the RDP or those whose identification at a specific taxonomic level was not statistically significant were aggregated, denoted “No ID”, and depicted in red. Identified phylotypes with less than 1% of the total relative abundance were aggregated, denoted “Minor”, and depicted in gray. Phylotypes with at least 1% relative abundance are labeled.

Figure 6

Relative abundance and taxonomic classification of the 16S rRNA gene sequencing results using the RDP for Enrichment reactors on operational day 271 and operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). The classified phylotypes are depicted in terms of the taxonomic hierarchy. Phylotypes which were not identified by the RDP or those whose identification at a specific taxonomic level was not statistically significant were aggregated, denoted “No ID”, and depicted in red. Identified phylotypes with less than 1% of the total relative abundance were aggregated, denoted “Minor”, and depicted in gray. Phylotypes with at least 1% relative abundance are labeled.

Figure 6

Relative abundance and taxonomic classification of the 16S rRNA gene sequencing results using the RDP for Enrichment reactors on operational day 271 and operated at oxygen mass transfer coefficients (k_La) of 20 hr⁻¹ (a), 12 hr⁻¹ (b), 8 hr⁻¹ (c), and 4 hr⁻¹ (d). The classified phylotypes are depicted in terms of the taxonomic hierarchy. Phylotypes which were not identified by the RDP or those whose identification at a specific taxonomic level was not statistically significant were aggregated, denoted “No ID”, and depicted in red. Identified phylotypes with less than 1% of the total relative abundance were aggregated, denoted “Minor”, and depicted in gray. Phylotypes with at least 1% relative abundance are labeled.