Metatranscriptomics supports mechanism for biocathode electroautotrophy by “Candidatus Tenderia electrophaga”

Biocathodes provide a stable electron source to drive reduction reactions in electrotrophic microbial electrochemical systems. Electroautotrophic biocathode communities may be more robust than monocultures in environmentally relevant settings, but some members are not easily cultivated outside of the electrode environment. We previously used metagenomics and metaproteomics to propose a pathway for coupling extracellular electron transfer (EET) to carbon fixation in “Candidatus Tenderia electrophaga”, an uncultivated but dominant member of the Biocathode-MCL electroautotrophic community. Here we validate and refine this proposed pathway using differential metatranscriptomics of replicate MCL reactors poised at the growth potential 310 mV and the suboptimal 470 mV (vs. standard hydrogen electrode). At both potentials, transcripts from “Ca. Tenderia electrophaga” were more abundant than from any other organism and its relative activity was positively correlated with current. Several genes encoding key components of the proposed “Ca. Tenderia electrophaga” EET pathway were more highly expressed at 470 mV, consistent with a need for cells to acquire more electrons to obtain the same amount of energy as at 310 mV. These included cyc2, encoding a homolog of a protein known to be involved in iron oxidation, confirmed to be differentially expressed by droplet digital PCR of independent biological replicates. Average expression of all CO2 fixation related genes is 1.23-fold higher at 310 mV, indicating that reduced energy availability at 470 mV decreased CO2 fixation. Our results substantiate the claim that “Ca. Tenderia electrophaga” is the key MCL electroautotroph, which will help guide further development of this community for microbial electrosynthesis. IMPORTANCE Bacteria that directly use electrodes as metabolic electron donors (biocathodes) have been proposed for applications ranging from microbial electrosynthesis to advanced bioelectronics for cellular communication with machines. However, just as we understand very little about oxidation of analogous natural insoluble electron donors, such as iron oxide, the organisms and extracellular electron transfer (EET) pathways underlying the electrode-cell direct electron transfer processes are almost completely unknown. Biocathodes are a stable biofilm cultivation platform to interrogate both the rate and mechanism of EET using electrochemistry and study the electroautotrophic organisms that catalyze these reactions. Here we provide new evidence supporting the hypothesis that the uncultured bacterium “Candidatus Tenderia electrophaga” directly couples extracellular electron transfer to CO2 fixation. Our results provide insight into developing biocathode technology, such as microbial electrosynthesis, as well as advancing our understanding of chemolithoautotrophy.

been demonstrated for Mariprofundus ferrooxydans PV-1 and Acidithiobacillus ferrooxidans (9,76 10), while other autotrophs require supplemental energy from light or hydrogen for initial growth 77 on a cathode (11,12). However, recent reports indicate that communities containing 78 Gammaproteobacteria using a high potential biocathode as an energy source can be reproducibly 79 enriched (13,14). 80 Previous metagenomic and proteomics studies of MCL have led to the identification of 81 putative EET and CO2 fixation mechanisms (8,15) although efforts at cultivation have not 82 yielded the proposed electroautroph, "Candidatus Tenderia electrophaga" (16). Other work has 83 resulted in isolates from biocathode enrichments, but these have not been autotrophic (17,18). 84 Metaproteomic analysis suggests that proteins that may be involved in EET are expressed at high 85 5 levels in the biofilm, including a homolog of Cyc2, a protein known to be involved in Fe +2 86 oxidation in A. ferrooxidans (8). Subsequent metaproteomic analysis of the biofilm at two 87 different electrode potentials showed that some components of the electron transport chain 88 (ETC) are differentially expressed including an ortholog of Cyc1, thought to be involved in Fe +2 89 oxidation in M. ferrooxydans , and a hypothetical protein homologous to the terminal 90 oxidase cytochrome cbb3 subunit CcoO (15). More precise quantification of the changes in gene 91 expression obtained using RNA-seq to quantify the relative abundance of transcripts (20) can be 92 used to understand the molecular mechanisms used for growth on the cathode.

93
Building upon our previous work (4, 8, 15), we applied RNA-seq to MCL to assess gene 94 expression of proposed EET and CO2 fixation pathways for "Ca. Tenderia electrophaga" when 95 the applied electrode potential is adjusted from 310 mV vs. SHE (optimal for growth) to 470 mV 96 (suboptimal for growth). This 160 mV increase in potential decreases the ΔG 0' for the reduction 97 of O2 by about one third, and from previous metaproteomic experiments run under identical 98 conditions is expected to result in changes in gene expression to compensate for the change in 99 energy availability. These results support previously hypothesized roles of some protein 100 complexes (8,15,19), and reveal possible EET roles for other proteins. We also provide further 101 evidence that "Ca. Tenderia electrophaga" is the keystone species in this community and is 102 strongly associated with current production. These results provide further understanding of the 103 molecular mechanisms involved in electroautotrophic growth on a cathode, enabling the use of 104 members of this community to develop biocathodes for applications including synthesis of value 105 added compounds or biofuels as well as potential ET components for microbial bioelectronics.

107
Biocathode-MCL Metatranscriptome: Eight BES were grown under previously described standard conditions at an applied 109 electrode potential of 310 mV SHE. Seven of the eight biological replicates reached maximum 110 current production within five days (Table 1, Fig. S1). Replicate S2A took two days longer. The 111 midpoint potential (EM) of each BES measured by cyclic voltammetry (CV) once steady-state 112 current was achieved was identical to that previously reported (4,8,15) at ca. 440 mV vs. SHE 113 (Table 1). Following CV, the electrode potential was either returned to 310 mV or adjusted to 114 470 mV, a potential at which we expect the biofilm to recover less energy per electron based on 115 previous experiments.

116
MiSeq mRNA sequencing yielded between 4.5 -16.2 million reads per replicate sample 117 that passed quality control steps and were used for read alignment to the metagenome, of which 118 37.9-87.5% could be unambiguously aligned (Table 1). The relative activity of MCL constituents 119 was assessed as the proportion of reads mapping to a metagenome bin normalized by relative 120 length of bin. Using this metric, MCL constituents previously implicated in key roles in the 121 biocathode -"Ca. Tenderia electrophaga", Marinobacter sp. strain CP1, and Labrenzia sp. strain 122 CP4.were highly active in all samples at both potentials. Ten other metagenome bins were 123 active in all eight samples using a cutoff of 0.01% of relative activity. Two of these, 124 Parvibaculum sp. and Kordiimonas sp. had activity similar to that of Labrenzia sp. and

125
Marinobacter sp., indicating that they may also have significant roles in the biocathode 126 community. "Ca. Tenderia electrophaga" was the most active organism in all eight samples, 127 comprising 53% to 84.3% of activity (Fig. 1). Marinobacter sp., Labrenzia sp., Kordiimonas 128 sp., and Parvibaculum sp. average 24.6% of the remaining activity. These five organisms make 129 up the core of the biofilm community, on average accounting for 93.6% of community activity. Two-tailed Student's T-tests indicated that no organism has a significant change in activity 131 between the two potentials tested (p > 0.09 for all bins).

132
Correlation of "Ca. Tenderia electrophaga" transcriptional activity to current density 133 Activity of "Ca. Tenderia electrophaga" was strongly correlated to current density at both  For this reason, we concentrated our analysis of metatranscriptomics data on "Ca. Tenderia 141 electrophaga" to focus specifically on how direct EET is linked to CO2 fixation.

Different Pathways for Electron Transport Are Used at Different Potentials:
143 Previous metagenomic and metaproteomic analysis of Biocathode-MCL revealed 144 proteins for carbon fixation through the CBB cycle and a putative EET pathway in "Ca. Tenderia 145 electrophaga" (8). Statistical analysis of protein expression revealed that nine proteins in "Ca.

146
Tenderia electrophaga" were detected more often at one potential versus another, but 147 methodological limitations meant that this was likely to be a small fraction of the response to 148 changing potential (15). Efforts to cultivate a representative isolate of "Ca. Tenderia 149 electrophaga" from MCL have thus far been unsuccessful, suggesting that other community 150 constituents satisfy unknown requirements for its growth. We therefore used metatranscriptomics 151 to obtain higher resolution analysis of differentially expressed mRNA levels for proteins 152 8 suspected to be involved in electroautotrophic growth. This allowed us to test the hypothesis that 153 expression of protein coding genes involved in respiration and energy flux in the primary 154 electroautotroph is directly influenced by the potential of electrons supplied by the cathode. A 155 complete list of genes we propose to be associated with these pathways is presented in Table S1.

156
Electrochemical measurements indicate redox dependent direct electron transfer (DET) is 157 occurring between the electrode and "Ca. Tenderia electrophaga" (4, 5), therefore, we searched 158 the metatranscriptome for evidence of an EET conduit whose expression may be affected by the 159 change in electrode potential. The cyc2 homolog in "Ca. Tenderia electrophaga" (Tel_03480) 160 was predicted to have a role in EET (8) based on its proposed involvement in iron oxidation and 161 was significantly differentially expressed, with relative transcript levels 1.8-fold higher at 470 162 mV. The gene for a predicted hexaheme lipoprotein (Tel_04230) was significantly more highly 163 expressed at 470 mV, raising the possibility that this protein is involved in electron transfer. An 164 undecaheme c-cyt (Tel_16545) was previously identified as a possible route for EET in "Ca.

165
Tenderia electrophaga" due to the large number of predicted heme binding sites, known to be 166 important for EET in Shewanella and Geobacter spp., and its conservation among other EET 167 capable organisms (8). The gene for this protein was not differentially expressed.

168
It is thought that soluble, periplasmic cytochromes mediate ET between the outer 169 membrane and the cytoplasmic membrane bound ETC (21, 22). Several genes previously noted 170 from "Ca. Tenderia electrophaga" which may encode proteins involved in transferring electrons 171 across the periplasm to ETC were examined for changes in expression between the two electrode 172 potentials. Expression was higher for three tri-heme cytochromes (Tel_16515, Tel_16520, 173 Tel_16530) at 310 mV, but was not statistically significant (p > 0.01). Overall, these genes are 174 among the most highly expressed in "Ca. Tenderia electrophaga", indicating their importance for 175 9 growth at the cathode. They also appear to be co-transcribed with a gene encoding a 176 tetratricopeptide repeat (Tel_16525), which are known to be involved in protein:protein 177 interaction and may be involved in aligning cytochromes in a bridge configuration as proposed 178 for other EET capable bacteria (23). Peptides from this tetratricopeptide repeat protein were 179 previously found to be significantly more abundant at the suboptimal potential (15), although in 180 the metatranscriptomic data, this cluster of genes is slightly more highly transcribed at 310 mV.

181
A di-heme cytochrome c4, Cyc1, has been proposed to be a periplasmic electron shuttle 182 in the iron oxidizing bacterium M. ferrooxydans (19). The gene encoding this protein is found in 183 "Ca. Tenderia electrophaga", in a region displaying synteny with two contigs from the M. 184 ferrooxydans genome ( Figure S2). Like the tri-heme cytochromes, it is very highly expressed at 185 both potentials, but is not significantly differentially expressed. Other genes for potential 186 periplasmic electron carriers identified from the metatranscriptome that were not previously  (Table S1). Only one operon in "Ca. Tenderia electrophaga" contains all four of the canonical genes 213 for a bacterial cytochrome cbb3 oxidase (ccoNOPQ), but it has relatively low expression.

214
However, the genome also encodes several proteins homologous to terminal oxidases (Complex 215 IV) that could potentially reduce O2 and generate proton motive force (PMF). Of these, the 216 genes encoding Complex IV-2 had some of the highest total expression of any genes, and 217 clustered with the undecaheme cytochrome and two additional copies of ccoO (Fig. 3) which 218 suggests that this represents a way for electrons to enter the cell and be used to reduce O2.

219
Among genes with a large change in expression, cyc2 clustered with a gene cluster 220 encoding a possible nitrate reductase and a complex including a ccoO homolog and a copper 221 11 containing plastocyanin homolog labeled as Putative Complex IV. However, this cluster of 222 genes lacks a protein that could function as an analog of the membrane associated proton 223 pumping CcoN. Thus it is unlikely to make a substantial contribution to PMF. All of the genes 224 encoding components of the two ATP synthase operons were more highly expressed at 310 mV, 225 many with a false discovery rate (FDR) of <0.01, indicating that energy levels in "Ca. Tenderia 226 electrophaga" are dependent upon the electrode potential.

227
Expression of 19 genes was quantified using droplet digital-PCR (ddPCR) to verify 228 differential expression of several genes hypothesized to be important for growth on the cathode 229 (Table S2). These measurements we performed on an independent set of eight MCL functional roles that could not be assigned to a single pathway were excluded from this analysis.

244
Most components of central carbon metabolism, including the CBB cycle were more highly 245 expressed at 310 mV, although the absolute magnitude of the change was relatively small (Fig. 4, 246 Table S3). Only four genes were differentially expressed, ribose 5-phosphate isomerase A and 247 ribulose-phosphate 3-epimerase which catalyze anaplerotic reactions in the CBB cycle, and 6-248 phosphogluconate dehydrogenase and glucose-6-phosphate 1-dehydrogenase which catalyze 249 components of the pentose phosphate pathway that supply metabolic precursors for biosynthesis.   Based on a prior analysis of Biocathode-MCL using low scan rate CV, we predicted 287 current to be approximately halved when the electrode potential was switched from 310 mV to 288 470 mV (15). However, as previously observed (15), for biocathode-MCL, switching to and 289 maintaining a more positive potential (470 mV) over a much longer time period (>50 h) resulted 290 in an increase in current attributed to O2 reduction after the initial adjustment period. This 291 increase in current is interpreted as the result of the need to make up for the decrease in energy 292 available per electron at the higher potential. Using a derivation of the Nernst Equation (ΔG = 293 nFΔE 0' ) to estimate the theoretical yield of O2 reduction to H2O using an electron donor at a 294 potential of 310 mV, yields -47 kJ/mol e -, requiring an additional ~61 kJ/mol eto reduce 295 NAD(P) + . Assuming that the energy required to pump protons across the membrane is ~21 296 kJ/mol, at the optimal potential "Ca. Tenderia electrophaga" could export two H + per e -, and 297 would need to use three H + to reduce NAD(P) + . This results in a theoretical balance for electron 298 utilization by the forward vs. reverse electron transport pathways of ~ 60%/40% to produce 299 NAD(P)H. More electrons would need to go to the forward path to generate a proton gradient 300 for ATP production. In comparison, using an electron donor at a potential of 470 mV reduces 301 the ΔG to -32 kJ/mol e -, lowering the theoretical yield to ~1.5 H + exported per e -, and the larger 302 ΔE 0' between the electrode and NAD(P)H requires at least four H + translocated per NAD(P) + 303 reduced. This results in a balance of at least 80% and 20% for the electron utilization by the 304 forward and reverse pathways. Thus, theoretically, at least twice as many electrons are needed to 305 generate the same amount of reducing equivalents for CO2 fixation at the more positive potential.

306
During growth in standard culture conditions, the acidophilic iron oxidizer A. ferrooxidans is 307 predicted to have a ratio closer to 90/10% for the forward vs. reverse pathways, including the 308 proton gradient necessary to generate ATP (30). Recent experimental evidence suggests that 309 during growth on an electrode, the ratio of electron utilization by the forward to reverse 310 pathways is close to 15:1 in A. ferrooxidans (10). EET and the ETC. Several proteins could potentially make up this link, but their roles are 333 presently unclear. Hierarchical clustering of potential electron transfer components by 334 expression leads to several distinct clusters (Fig. 3). One cluster contains the ACIII and Complex 335 I genes, which suggests that these complexes may be functionally linked. This cluster also 336 contains the Cyc1 homolog which suggests that it may facilitate the transfer of electrons to the 337 "uphill" ETC, rather than to the terminal oxidase as proposed elsewhere (19,30). The canonical 338 cytochrome cbb3 oxidase is part of a cluster with succinate dehydrogenase (Complex II) and the 339 cytochrome bc1 complex. This may indicate that they form part of a forward ETC, which would 340 enable the use of stored reserves in the form of glycogen.

341
Metatranscriptomics supports previous work by our group that identified "Ca. Tenderia 342 electrophaga" as the electroautotroph. It appears to be the key member of the community and its 343 predicted lifestyle as an electroautotroph is supported by a several key findings. "Ca. Tenderia 344 electrophaga" was more active than any other organism and its activity was positively correlated 345 with current density. Genes for proteins previously predicted to be involved in EET in "Ca.

346
Tenderia electrophaga", including Cyc2, were more highly expressed at 470 mV, when more 347 electrons are needed to generate the same amount of PMF. The membrane bound ETC that is 348 necessary to generate the ATP and reducing power for CO2 fixation was active at both potentials, 349 but a potential complex IV analog and several that may be involved in the "downhill" branch of 350 the ETC showed increased expression in response to changes in electrode potential. These uptake is approximately half of that at the "optimal" potential. After 52 hours, the biofilm was 375 scraped from ~3 x 9 cm portions of both sides of the cathode with a sterile razor blade,  Correlations between current density and relative abundance of mRNA from each 425 metagenomic genome or bin were calculated as Pearson's product moment correlation 426 coefficient (44). The critical value for a significant correlation for a two-tailed test with a sample 427 size of n = 4 is 0.9 for a p-value of 0.1. A higher p-value cutoff for the correlation is justified 428 because this is a different kind of statistical test and it assumes a linear relationship between the 429 data, which is likely not the case for most genes. Furthermore, there is a large variability 430 between samples of the same conditions (current magnitude, current increasing/decreasing, and 431 community composition) that would prevent the near perfect correlation a p-value of 0.01 would 432 require.

433
Validation of differential expression with ddPCR: 434 RNA samples from eight additional biological replicates were obtained from the same 435 inoculum following the protocol above (four optimal and four suboptimal). After RNA 436 extraction, a whole transcriptome amplification protocol was applied using the WTA-2 kit 437 (Sigma) according to the manufactures instructions, performing a no template negative control.

438
Individual ddPCR reactions for each primer set listed in Table S5 were set up using QX200 439 ddPCR EvaGreen Supermix (BioRad) and QX200 Droplet Generation Oil for EvaGreen 440 (BioRad). Copy numbers of mRNAs for 19 genes were assessed as copy number per µl of 441 reaction minus calculated copy number per µl of corresponding negative control reaction.

442
Samples were normalized using a trimmed means approach to account for variation in the 443 proportion of "Ca. Tenderia electrophaga" in the community, and significance of differential 444 expression was assessed using two-tailed t-tests.   Figure 1. Relative activity of the five major constituents of Biocathode-MCL. Proportion of 594 reads matching a genome or bin normalized by the length of the genome or bin was taken as a 595 proxy for activity. Sample labels are as in Table 1.

608
Expression is normalized to 0 between biological replicates to focus on change in expression 609 between 310 mV and 470 mV. Details about genes in each complex are found in Table S1.

610
Sample identifiers are as in Table 1.  Table S3.   Figure S1. Chromoamperometry of biofilms grown at the 310 mv and either switched to 470 mV 632 or continued at the optimal potential. Vertical lines indicate the time that the CV was measured

638
The M. ferrooxydans genes are found on three contigs in the NCBI database, but two 639