Anion and cation fluxes in Escherichia coli during glucose fermentation and their biochemical implications Authors: Liana Vanyan, Heghine Gevorgyan, Liza Jorkntsyan, Karen Trchounian, Sunil Nath Published: 2026 ABSTRACT The end products of glucose fermentation by Escherichia coli include organic acids such as formic, acetic, lactic, and succinic acid. Anions are transported out of the cell along with protons by a symport mechanism mediated by the FOF1-ATPase operating in the hydrolysis mode. The present work measures and analyzes new data on H+ efflux, K+ influx, and anionic formate, acetate, lactate, and succinate efflux rates in whole cells of E. coli grown at 2 g L-1 and 8 g L-1 glucose under fermentative conditions at pH 7.5, with 2 g L-1 added glucose. The experiments are performed both in the presence and absence of 0.2 mM N N’-dicyclohexylcarbodiimide (DCCD), and in the presence and absence of external formate. At 2 g L-1 glucose in the growth medium, and addition of 2 g L-1 glucose, the mean DCCD-sensitive H+ efflux rate of (1.83 ± 0.20) × 10-3 mmol min-1 is almost exactly compensated by a (succinate + K+) transport rate of (1.94 ± 0.12) × 10-3 mmol min-1. This equivalence also holds in other experimental conditions investigated, e.g. in the presence of 10 mM external formate, and with 8 g L-1 glucose in the growth medium and 2 g L-1 added glucose. Comparison of ion transport rates between wild type and mutant atpB (a) and atpE (c) E. coli cells at a growth pH of 7.5 with 2 g L-1 added glucose are also made. Biochemical implications of these findings are discussed, and a structural model involving cotransport and supercomplex formation with the FOF1-ATPase that is consistent with the experimental observations is presented. Finally, the great scope for future work, and the innovations needed for further progress on these key aspects of bioenergetics and transport across biomembranes are discussed. Introduction Escherichia coli utilizes sugars such as glucose and xylose, alcohols such as glycerol, and other carbon sources and produces various products under anaerobic conditions1‒3. The end products of glucose fermentation include organic acids such as formic, acetic, lactic, and succinic acid4. The succinate formed during anaerobic sugar fermentation cannot be further metabolized and is exported out of the cell. The succinate/organic acids are transported along with protons by a symport mechanism5‒9. A major proton transporting enzyme is the FOF1-ATP synthase, one of the key enzymes in bioenergetics, which catalyzes the synthesis of ATP in the terminal step of oxidative phosphorylation and photophosphorylation10‒12. ATP synthesis/hydrolysis occur in the F1 portion of the enzyme, whereas ion transport occurs in the FO part. It is well known that the E. coli FO consists of three subunits with mismatched ab2cn stoichiometry with respect to F113‒19. Recently, cryo-EM studies have provided unprecedented information on how rotation of the FO c-ring leads to conformational changes in F1, suggesting how the FO and F1 motors are coupled20,21. In energy-utilizing processes19, and during fermentative conditions the FOF1-ATPase works in the reverse ATP hydrolysis mode and pumps H+ out of the cell. Moreover, its regulation is different compared to that in respiratory conditions, and data suggest that there are important links between FOF1-ATPase and solute secondary transporters or membrane-bound enzymes22,23. The FOF1-ATPase is regulated by external pH, potassium ions, and other factors. Transport of different fermentation weak acids e.g. lactic, acetic, succinic acids) have been shown to generate membrane potential24,25. Moreover, lactic, acetic and other acids can also dissipate proton gradients in a concentration-dependent way26. Succinate is transported via various Dcu transporters or carriers that are mainly expressed under anaerobic conditions and have been known to be important for succinate/H+ efflux during fermentation25,27. External formate and FocA/FocB formate channels have also been implicated recently in regulation28,29. A simultaneous measurement of transport rates of cations (H+ and K+) and organic anions (formate, acetate, lactate, and succinate) under fermentative conditions in cells and through the ATPase has not been made. This research gap has been addressed in the present study. Given the above, it is important to measure cation and anion transport rates during glucose fermentation. The current paper records and analyzes new data on H+ efflux, K+ influx, and anionic formate, acetate, lactate, and succinate efflux in E. coli cells grown at 2 g L‒1 and 8 g L-1 glucose under fermentative conditions at pH 7.5, both in the presence and absence of DCCD, and in the presence and absence of external formate. The biochemical implications and scope for future work on bioenergetic aspects and on key aspects of transport across biomembranes have also been discussed. Materials and methods Bacteria, growth conditions, and membrane vesicles The wild type E. coli BW25113 rrnB ΔlacZ4787 HsdR514 Δ(araBAD)567 Δ(rhaBAD)568 rph-1 was used in this study30,31. Bacteria were grown under anaerobic conditions at 37º C for 22–24 h in highly buffered peptone medium with the following composition: for studies at pH 7.5–20 g L−1 peptone, 15 g L−1 K2HPO4, 1.08 g L−1 KH2PO4, 5 g L−1 NaCl supplemented 2 g L−1 and 8 g L−1 glucose31,32. Anaerobic conditions were achieved by using glass bottles with plastic press-caps. O2 was removed from the medium by autoclaving, after which the bottles were closed by press-caps, as described previously31,32. Measurement of H+ and K+ ion fluxes H+ and K+ ion fluxes by whole cells were determined in the assays using Tris–phosphate buffer pH 7.5 (150 mM Tris–phosphate containing 0.4 mM MgSO4, 1 mM NaCl, 1 mM KCl upon addition of glucose (2 g L-1) to the bacterial suspension, by registering the changes in H+ (ΔaH+) or K+ (ΔaK+) fluxes in the external medium using selective pH (HI1131B, Hanna Instruments, Portugal) or potassium electrodes (HI4114, Hanna Instruments, Portugal)33,34. The electrode readings were calibrated by titration of the medium with 0.01 M HCl or with 0.1 M KCl. ΔaH+ and ΔaK+ were expressed in mmol min-1 per 109 cells in 1 unit of volume (mL). The whole cells were incubated with 0.2 mM N N’-dicyclohexylcarbodiimide (DCCD) where indicated, as in earlier work33,34. Determination of organic acids The organic acids formate, acetate, lactate, and succinate were determined by HPLC. The HPLC system Agilent 1260 Infinity II LC Bio-inert was used with a refractive index detector (Agilent RID, G1362A, set on positive polarity and optical unit temperature of 55 °C). Data processing was carried out using Agilent Open LAB CDS. A Macherey–Nagel EC 250/4.6 NUCLEOSIL 120–5 C18 column (250 × 4.6 mm, MN720041.46, Düren, Germany) was used for analysis. The column was purged with acetonitrile/water (80/20) at 60 °C overnight and then regenerated with a mobile phase (5 mM sulfuric acid in ddH2O) for a few hours to reach experimental conditions. For regular column maintenance, 20 mm inlet filter (5041–2168, Agilent, Germany) was used. Samples were injected by multi-sampler (Agilent Bio, G5668A). The total analysis time of a single injection was 42 min, injection volume was 10 μL. Column temperature was kept at 60 °C using a thermostatically controlled column compartment (Agilent MCT, G7116A). Analytes were eluted with 5 mM H2SO4 at 0.4 mL min-1. Concentrations of dissolved substances were calculated according to standard procedures35. In experiments agar, peptone, glucose, Tris (Carl Roth GmbH, Germany), ATP (Tris salt), DCCD, lysozyme, formate, acetate, lactate, succinate, (Sigma, USA) and other chemicals of analytical grade were used. Data obtained from 3 independent assays were averaged, and standard deviations of values were calculated, they did not exceed 5% if not shown. Other procedures were as described previously31,33. Results and discussion H+ fluxes in the presence and absence of DCCD by wild type E. coli cells during glucose fermentation at pH 7.5 It is well known that during fermentative conditions the ATP synthase functions in the hydrolysis mode. Ion pumping out of the cells is mainly performed by the FOF1-ATPase, which has been discussed at length previously11,12,25,28,29,36,37. During fermentation, besides FOF1, it has been experimentally demonstrated earlier that lactate or succinate transport generates local electrical potential, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta \psi$$\end{document} 24,27. It has been reviewed that with FOF1, fermentation weak acids contribute to maintain overall ionic driving forces25. Since we expect anionic transport during fermentation to occur along with proton symport, the cation and anion fluxes need to be measured in wild type E. coli cells. Hence, first, total and DCCD-sensitive H+ efflux was determined during glucose (2 g L-1) fermentation at pH 7.5. For cells were grown at pH 7.5 and 2 g L-1 glucose, with addition of 2 g L-1, total H+ efflux in wild type cell assays in the absence of DCCD measured 5.82 × 10-3 mmol min-1 under these conditions (Fig. 1). In order to understand the contribution of FOF1-ATPase to the total H+ flux, DCCD, a specific inhibitor of the FOF1-ATPase in E. coli38,39 was added. The H+ efflux decreased by ∼ 30% in the presence of 0.2 mM DCCD (to 3.99 × 10-3 mmol min-1). It has been shown in earlier studies that the H+ flux in wild type cells in the presence of DCCD is identical to the flux in a DK8 mutant lacking FOF1 with and without addition of DCCD to this mutant40. Thus the DCCD-sensitive H+ flux occurs due to the activity of FOF1 and measured 1.83 × 10-3 mmol min-1 under our conditions (Fig. 1).Fig. 1H+ efflux and K+ influx by whole cells of E. coli wild type strain BW25113 in the absence and presence of 0.2 mM DCCD. In assays at pH 7.5, and 1 mM added K+ for potassium in the K+ influx assays, glucose was added at a concentration of 2 g L-1 for both 2 g L-1 glucose and 8 g L-1 glucose in the growth medium. These are shown by the solid green line for 2 g L-1 glucose and by the chained blue line for 8 g L-1 glucose in the growth medium in the absence of DCCD, and by the dashed red line and the black dotted line in the presence of 0.2 mM DCCD for the 2 g L-1 and 8 g L-1 glucose respectively. Bacteria and glucose were added at the times indicated. Bacteria were grown under fermentative conditions at pH 7.5 in the presence of 2 g L-1 or 8 g L-1 glucose at 37 °C. The assay pH was identical with the growth pH. See Materials and Methods. H+ efflux and K+ influx by whole cells of E. coli wild type strain BW25113 in the absence and presence of 0.2 mM DCCD. In assays at pH 7.5, and 1 mM added K+ for potassium in the K+ influx assays, glucose was added at a concentration of 2 g L-1 for both 2 g L-1 glucose and 8 g L-1 glucose in the growth medium. These are shown by the solid green line for 2 g L-1 glucose and by the chained blue line for 8 g L-1 glucose in the growth medium in the absence of DCCD, and by the dashed red line and the black dotted line in the presence of 0.2 mM DCCD for the 2 g L-1 and 8 g L-1 glucose respectively. Bacteria and glucose were added at the times indicated. Bacteria were grown under fermentative conditions at pH 7.5 in the presence of 2 g L-1 or 8 g L-1 glucose at 37 °C. The assay pH was identical with the growth pH. See Materials and Methods. K+ uptake and anion efflux in the presence and absence of DCCD by wild type E. coli cells during glucose fermentation at pH 7.5 The hypothesis that FOF1 is associated with various secondary transport systems has been advanced in both aerobic metabolism, e.g. through the enzyme’s coupled a-subunit transport system7,41‒45 as well as in fermentative metabolism, e.g. through Trk or Dcu transporters22,23,25,34. Hence we determined the uptake or extrusion of K+ ions and anions in the presence and absence of DCCD during fermentation of glucose in wild type E. coli cells (Figs. 1, 2, 3). K+ uptake When 2 g L-1 glucose was added in the assays containing external K+ at a concentration of 1 mM, wild type cells revealed an uptake of potassium ions with a rate of 0.95 × 10-3 mmol min-1 (Figs. 1, 4a). The addition of 0.2 DCCD greatly perturbed K+ uptake and led to a substantial diminution of its influx rate in the absence of inhibitor by > 90% to 0.08 × 10-3 mmol min-1 (Fig. 4b). A- efflux The production, transport, and role of different organic acids, e.g. formate, acetate, lactate, and succinate during anaerobic fermentation is still unclear. Depending on external factors such as pH, glucose concentration, the amount of acids produced and subsequently transported across the membrane is variable46. Hence the concentration of these various organic anions was measured as a function of time for E. coli cells during glucose fermentation at pH 7.5. For addition of glucose at low concentration (2 g L-1), same as in the growth medium, the data are shown in Figs. 2, 3.Fig. 2Anion efflux by whole E. coli wild type cells in the absence of DCCD. Glucose was used in the assays at the same concentration (2 g L-1) as in the growth medium. The assay pH of 7.5 was the same as the growth pH. The concentration of formate (solid light blue squares, ■), acetate (solid orange circles, ●), lactate (solid grey triangles, ▲), and succinate anions (solid yellow rhombuses, ♦) are illustrated as a function of time. Change in glucose concentration with time is shown by open dark blue squares, □. For other conditions see legend to Fig. 1 and Materials and Methods.Fig. 3Anion efflux by whole E. coli cells in the presence of 0.2 mM DCCD. Glucose was employed in assays at concentration of 2 g L-1, the same as in the growth medium. The assay pH of 7.5 was also the same as the growth pH. The concentration of formate (solid light blue squares, ■), acetate (solid orange circles, ●), lactate (solid grey triangles, ▲), and succinate anions (solid yellow rhombuses, ♦) are shown as a function of time. Variation in glucose concentration with time is shown by open dark blue squares, □. For other conditions see legend to Fig. 1 and Materials and Methods. Anion efflux by whole E. coli wild type cells in the absence of DCCD. Glucose was used in the assays at the same concentration (2 g L-1) as in the growth medium. The assay pH of 7.5 was the same as the growth pH. The concentration of formate (solid light blue squares, ■), acetate (solid orange circles, ●), lactate (solid grey triangles, ▲), and succinate anions (solid yellow rhombuses, ♦) are illustrated as a function of time. Change in glucose concentration with time is shown by open dark blue squares, □. For other conditions see legend to Fig. 1 and Materials and Methods. Anion efflux by whole E. coli cells in the presence of 0.2 mM DCCD. Glucose was employed in assays at concentration of 2 g L-1, the same as in the growth medium. The assay pH of 7.5 was also the same as the growth pH. The concentration of formate (solid light blue squares, ■), acetate (solid orange circles, ●), lactate (solid grey triangles, ▲), and succinate anions (solid yellow rhombuses, ♦) are shown as a function of time. Variation in glucose concentration with time is shown by open dark blue squares, □. For other conditions see legend to Fig. 1 and Materials and Methods. In the absence of DCCD, the overall kinetics is nonlinear and biphasic and can be analyzed based on enzymological principles47. The initial rates of formate, acetate, lactate, and succinate efflux measured 3.62 × 10-3, 2.50 × 10-3, 3.67 × 10-3, and 1.19 × 10-3 mmol min-1 respectively for the data plotted in Fig. 2. The maximal translocation rates are depicted by the bar chart in Fig. 4a. These rapid initial rates of anion efflux are sustained for the time period in which the glucose consumption rate is high, i.e. at short times, when the external glucose concentration decreases at a fast rate (Fig. 2). Thereafter, the glucose concentration reduces to almost zero, and moreover, the rate of its disappearance is very small and close to zero, which leads to declining rates of anion transport (Fig. 2).Fig. 4Anion and cation flux in whole cells of E. coli for the data represented in Figs. 1, 2, 3. Symbols on the x-axis denote the fluxes of the following species: H+: proton efflux; f: formate efflux; a: acetate efflux; l: lactate efflux; s: succinate efflux, K+: potassium ion influx. Y-axis shows values of ion fluxes in mmol min-1 × 103 (mean ± SD; n=3). a): flux of various species in the absence of DCCD, calculated based on their initial rates. ****p < 0.0001. b): flux of species in the presence of 0.2 mM DCCD. For conditions, see legend to Figs. 2, 3. ****p < 0.0001. Anion and cation flux in whole cells of E. coli for the data represented in Figs. 1, 2, 3. Symbols on the x-axis denote the fluxes of the following species: H+: proton efflux; f: formate efflux; a: acetate efflux; l: lactate efflux; s: succinate efflux, K+: potassium ion influx. Y-axis shows values of ion fluxes in mmol min-1 × 103 (mean ± SD; n=3). a): flux of various species in the absence of DCCD, calculated based on their initial rates. ****p < 0.0001. b): flux of species in the presence of 0.2 mM DCCD. For conditions, see legend to Figs. 2, 3. ****p < 0.0001. By contrast, in the presence of 0.2 mM DCCD, the rates of glucose consumption and efflux of formate, acetate, lactate, and succinate anions are all considerably slowed down. Moreover, the increase in their concentration outside cells shows an almost perfectly linear trend over a time period of seven minutes (Fig. 3). The rates of anion efflux of formate, acetate, lactate, and succinate during this time measured 0.42 × 10-3, 0.18 × 10-3, 0.44 × 10-3, and 0.12 × 10-3 mmol min-1 respectively for the data in Fig. 3. These anion efflux rates in the presence of DCCD are shown in the bar chart of Fig. 4b. The results of Figs. 4a, b in the absence and presence of DCCD respectively reveal the lack of ionic balance of proton efflux by either K+ influx or by efflux of any particular anion, —or any combinations thereof—, pointing to the presence of unknown cotransporters or exchangers in the whole cell. The difference map between Figs. 4a and 4b based on the flux data of Figs. 1, 2, 3 plotted in Fig. 5 showing the DCCD-sensitive H+ and other exchanges tell a more interesting story. The DCCD-sensitive H+ exchange rate is not balanced by either the rate of formate, acetate, or lactate anion efflux, which are too large compared to the net H+ efflux rate, nor by either the succinate efflux rate or the K+ influx rate, which are individually too small compared to the rate of the DCCD-sensitive H+ efflux (Fig. 5). Analysis of the data shown in Figs. 1 ‒ 3 reveal that only the sum of the rates of net succinate efflux and net K+ influx, s + K+, drawn in the last panel of Fig. 5 almost exactly balance—to ∼ 5%, and within the error—the rate of the DCCD-sensitive H+ efflux tabulated in the first panel. The biochemical implications of this finding shall be discussed in a subsequent section.Fig. 5Analysis of DCCD-sensitive component of ionic fluxes based on transport data on whole cells of E. coli in Figs. 1 ‒ 4. DCCD-sensitive flux of ionic species based on the difference between fluxes in Figs. 4a and 4b are shown on the y-axis (mean ± SD; n=3). H+: proton efflux; f: formate efflux; a: acetate efflux; l: lactate efflux; s: succinate efflux, K+: potassium ion influx; s + K+: sum of succinate efflux and potassium ion influx. ns: not significant, p > 0.05; **p < 0.01, ****p < 0.0001. Analysis of DCCD-sensitive component of ionic fluxes based on transport data on whole cells of E. coli in Figs. 1 ‒ 4. DCCD-sensitive flux of ionic species based on the difference between fluxes in Figs. 4a and 4b are shown on the y-axis (mean ± SD; n=3). H+: proton efflux; f: formate efflux; a: acetate efflux; l: lactate efflux; s: succinate efflux, K+: potassium ion influx; s + K+: sum of succinate efflux and potassium ion influx. ns: not significant, p > 0.05; **p < 0.01, ****p < 0.0001. Given the above results from the analysis of fluxes of Figs. 1, 2, 3, 4, the mean ± SD values for the efflux rates of protons and the various organic anions and K+ based on quantitative differences for our pooled data are of interest. Figure 5 tabulates the mean ± SD values of the DCCD-sensitive H+ efflux and the anion/cation fluxes—both individually and for addition of the efflux rate of the relevant anion (succinate) and the influx rate of K+. Figure 5 also tests the statistical significance of the data. Mean transport rates of 4.07 × 10-3, 3.19 × 10-3, 4.10 × 10-3 mmol min-1 for (formate + K+), (acetate + K+), and (lactate + K+) in Fig. 5 were ∼ two-fold higher compared to the mean DCCD-sensitive H+ rates of 1.83 × 10-3 mmol min-1. Even the individual formate, acetate, and lactate transport rates exceeded the DCCD-sensitive H+ rate by > 25% for acetate and > 75% for formate and lactate (Fig. 5). Only the mean (succinate + K+) transport rates of 1.94 × 10-3 mmol min-1, denoted by s + K+, perfectly matched the mean DCCD-sensitive proton transport rates within the error (Fig. 5). No other permutation or combination of the data could equal the DCCD-sensitive difference map of H+ rates, which is a specific measure of transport flux through the ATPase. The above results were tested further. Formate has been implicated in modulating H+ and K+ transport rates in recent work28,29. Experiments with conditions identical to those described in this Sect. (2 g L-1 glucose in both growth and assay media, see Figs. 1, 2, 3) were performed in the absence and presence of 0.2 mM DCCD but now with 10 mM formate added in the external medium. Similar results to those shown in Figs. 1 ‒ 3 were obtained. The DCCD-sensitive H+ efflux rate now measured 2.10 × 10-3 mmol min-1, while the DCCD-sensitive (succinate + K+) transport in the presence of 10 mM formate measured between 1.82 × 10-3 ‒ 2.68 × 10-3 mmol min-1, once again matching the DCCD-sensitive proton translocation rate. Cation and anion fluxes at high glucose concentration The above aspects were further explored at different glucose concentrations. Results are shown in Fig. 6 for whole cells of E. coli grown at pH 7.5 in the presence of high glucose concentration of 8 g L-1, and addition of 2 g L-1 glucose. The DCCD-sensitive H+ rates of 3.20 × 10-3 mmol min-1 are balanced by the (succinate + K+) transport rate with a mean ± SD of (2.85 ± 0.20) × 10-3 mmol min-1 (Fig. 6), again reflecting good agreement. However, when experiments were performed with cells grown at either 2 g L-1 or 8 g L-1, with added high glucose concentration of 8 g L-1, the glucose consumption rates were very high, and an automated or improved methodology for fast measurement of anion transport rates with a sampling time of ∼ 15 s needs to be developed. These aspects shall be taken up in our future work and are further discussed in the Conclusion and future outlook section.Fig. 6Mean ± SD for the DCCD-sensitive component of proton and (succinate anion + K+) ion fluxes (n=4), i.e. for the sum of (succinate efflux and K+ influx), denoted by s + K+, for whole cells of E. coli grown under fermentative conditions at pH 7.5 in the presence of 8 g L-1 glucose at 37 °C, and addition of 2 g L-1 glucose. The assay pH was identical with the growth pH. Mean ± SD for the DCCD-sensitive component of proton and (succinate anion + K+) ion fluxes (n=4), i.e. for the sum of (succinate efflux and K+ influx), denoted by s + K+, for whole cells of E. coli grown under fermentative conditions at pH 7.5 in the presence of 8 g L-1 glucose at 37 °C, and addition of 2 g L-1 glucose. The assay pH was identical with the growth pH. Table 1 presents the main results of this work on cation and anion fluxes determined in the absence and presence of DCCD for glucose fermentation by wild type Escherichia coli cells. New transport data on mutant Escherichia coli cells are summarized in Table 2. Table 2 compares proton transport data on wild type cells in the absence and presence of 0.2 and 0.5 mM DCCD with atpB (a) and atpE (c) mutants at a growth pH of 7.5 with 0.2% glucose. A maximal proton transport of 2.4 mM min-1 in wild type BW25113 strain in the absence of DCCD—corresponding to a proton transport rate of 0.012 mmol min-1—is assigned a value of 100%, and transport rates in atp mutant cells are tabulated with respect to this reference.Table 1Summary of H+, K+, and various anion transport fluxes (mmol min-1 × 103) in the absence and presence of DCCD determined in this work for E. coli under different conditions.NoConditionH+ effluxformateacetatelactatesuccinate∑A − effluxDCCD-sensitive H+ effluxK+ influx(succinate + K+)12g L-1 BW + 2 g L‒1 glucose5.82 ± 0.133.62 ± 0.122.50 ± 0.063.67 ± 0.151.19 ± 0.1210.98 ± 0.210.95 ± 0.032.14 ± 0.1222 g L-1 BW + 2 g L‒ 1 glucose + 0.2 mMDCCD3.99 ± 0.150.42 ± 0.020.18 ± 0.010.44 ± 0.020.12 ± 0.011.16 ± 0.040.08 ± 0.00.2 ± 0.01Difference component, 1 ‒ 21.83 ± 0.203.20 ± 0.122.32 ± 0.063.23 ± 0.151.07 ± 0.129.82 ± 0.221.83 ± 0.200.87 ± 0.031.94 ± 0.1232 g L-1 BW + 2 g L‒ 1 glucose + 10 mM formate6.252.882.062.412.169.510-1.042 g L-1 BW + 2 g L‒ 1 glucose + 10 mM formate + 0.2 mM DCCD4.150.680.320.920.342.260‒0.09Difference component, 3 ‒ 42.102.201.741.491.827.252.100‒0.861.82‒2.6858 g L-1 BW + 2 g L‒1 glucose7.002.011.046.351.4610.861.05-1.7568 g L-1 BW + 2 g L‒ 1 glucose + 0.2 mMDCCD3.800.360.261.620.052.29-0.55‒0.47Difference component, 5 ‒ 63.201.650.784.741.418.583.201.28-1.602.85 ± 0.20Table 2Comparison of proton ion transport rates JH (%) between wild type and atpB (a) and atpE (c) mutant E. coli cells at a growth pH of 7.5 with 0.2% glucose. A maximal H+ transport rate of 0.012 mmol min-1 in the absence of DCCD is assigned as 100% for ease in comparison of fluxes. Measurements were made every min and the measured transport rates were found to be almost constant between 1 and 5 min. Errors lie within ± 2% points.StrainJH in absence ofDCCDJH in presence of 0.2 mMDCCDJH in presence of 0.5 mMDCCDDCCD-sensitiveJH at 0.5 mM DCCDBW25113 (wt)100663367atpB (a)3333atpE (c)3333 Table 1Summary of H+, K+, and various anion transport fluxes (mmol min-1 × 103) in the absence and presence of DCCD determined in this work for E. coli under different conditions.NoConditionH+ effluxformateacetatelactatesuccinate∑A − effluxDCCD-sensitive H+ effluxK+ influx(succinate + K+)12g L-1 BW + 2 g L‒1 glucose5.82 ± 0.133.62 ± 0.122.50 ± 0.063.67 ± 0.151.19 ± 0.1210.98 ± 0.210.95 ± 0.032.14 ± 0.1222 g L-1 BW + 2 g L‒ 1 glucose + 0.2 mMDCCD3.99 ± 0.150.42 ± 0.020.18 ± 0.010.44 ± 0.020.12 ± 0.011.16 ± 0.040.08 ± 0.00.2 ± 0.01Difference component, 1 ‒ 21.83 ± 0.203.20 ± 0.122.32 ± 0.063.23 ± 0.151.07 ± 0.129.82 ± 0.221.83 ± 0.200.87 ± 0.031.94 ± 0.1232 g L-1 BW + 2 g L‒ 1 glucose + 10 mM formate6.252.882.062.412.169.510-1.042 g L-1 BW + 2 g L‒ 1 glucose + 10 mM formate + 0.2 mM DCCD4.150.680.320.920.342.260‒0.09Difference component, 3 ‒ 42.102.201.741.491.827.252.100‒0.861.82‒2.6858 g L-1 BW + 2 g L‒1 glucose7.002.011.046.351.4610.861.05-1.7568 g L-1 BW + 2 g L‒ 1 glucose + 0.2 mMDCCD3.800.360.261.620.052.29-0.55‒0.47Difference component, 5 ‒ 63.201.650.784.741.418.583.201.28-1.602.85 ± 0.20 Summary of H+, K+, and various anion transport fluxes (mmol min-1 × 103) in the absence and presence of DCCD determined in this work for E. coli under different conditions. 2g L-1 BW + 2 g L‒ 1 glucose 2 g L-1 BW + 2 g L‒ 1 glucose + 0.2 mM DCCD 8 g L-1 BW + 2 g L‒ 1 glucose 8 g L-1 BW + 2 g L‒ 1 glucose + 0.2 mM DCCD Table 2Comparison of proton ion transport rates JH (%) between wild type and atpB (a) and atpE (c) mutant E. coli cells at a growth pH of 7.5 with 0.2% glucose. A maximal H+ transport rate of 0.012 mmol min-1 in the absence of DCCD is assigned as 100% for ease in comparison of fluxes. Measurements were made every min and the measured transport rates were found to be almost constant between 1 and 5 min. Errors lie within ± 2% points.StrainJH in absence ofDCCDJH in presence of 0.2 mMDCCDJH in presence of 0.5 mMDCCDDCCD-sensitiveJH at 0.5 mM DCCDBW25113 (wt)100663367atpB (a)3333atpE (c)3333 Comparison of proton ion transport rates JH (%) between wild type and atpB (a) and atpE (c) mutant E. coli cells at a growth pH of 7.5 with 0.2% glucose. A maximal H+ transport rate of 0.012 mmol min-1 in the absence of DCCD is assigned as 100% for ease in comparison of fluxes. Measurements were made every min and the measured transport rates were found to be almost constant between 1 and 5 min. Errors lie within ± 2% points. There was no appreciable difference in the proton translocation rates obtained in atp mutants atpB and atpE in the absence and presence of 0.2 mM DCCD, as expected. The data in Table 2 lead to the tentative hypothesis that atp mutant strains catalyze a lower H+ transport rate in the presence or absence of DCCD corresponding to the difference in proton rates between the measured rates in the absence of DCCD and the DCCD-sensitive proton transport rates at high DCCD concentrations (≧0.5 mM). This interesting hypothesis shall be tested in further biochemical studies. The use of fast data sampling times (see Future outlook section) of 15 s—compared to ∼ 1 min—was attempted successfully for measurement of potassium ion transport rates. The results (not shown) revealed a very rapid inhibition of K+ transport rates in both atpB (a) and atpE (c) mutants in the absence of DCCD within 15 s. It suggests an interesting a‒c communication and coupling between the counter-transport of the two cations, and a possible role of the a-subunit in the counter-transport of K+. These important mechanistic aspects merit further experimental investigation. Biochemical implications Overall, analysis of the experimental data of Figs. 1, 2, 3 and results shown in Figs. 5, 6 (summarized in Table 1) lead to the conclusion that the DCCD-sensitive H+ efflux mediated by the FOF1-ATPase is balanced by the sum of succinate efflux and K+ influx in E. coli whole cells during glucose fermentation. This general result had been worked out quantitatively and mechanistically (see Eqs. (1) ‒ (3) and Fig. 4c of ref. 48) previously. A simplified schematic model of the working mechanism that summarizes these findings, emphasizing a 1:1 proton-anion symsequenceport or proton-K+ antisequenceport operating under our conditions is given in Fig. 7.Fig. 7Simplified schematic representation of a working mechanism of coupled proton and anion (e.g. succinate) or counter-cation (e.g. K+) fluxes under ATP hydrolytic conditions. Bold arrows represent primary translocations, and dashed arrows show secondary translocations. Adapted and modified from Nath and Villadsen (2015)51. See text for discussion. Simplified schematic representation of a working mechanism of coupled proton and anion (e.g. succinate) or counter-cation (e.g. K+) fluxes under ATP hydrolytic conditions. Bold arrows represent primary translocations, and dashed arrows show secondary translocations. Adapted and modified from Nath and Villadsen (2015)51. See text for discussion. How can the above ion co- or counter-transport processes be rationalized in structural terms? We suggest a likely structural model in Fig. 8 based on our experimental results. Traditionally, under respiratory conditions in aerobic metabolism, the ATPase enzyme with interacting a-c subunits in the membrane is taken together to constitute the H+-transporting FO sector of the FOF1-ATPase/ATP synthase, and is named as such11,12. However, it is well known that protons bind and unbind from their binding site located entirely in the c-subunit at the a-c interface. Suppose, in a thought experiment, we were to coin a nomenclature in which we term the FOF1-ATPase/ATP synthase minus the a-subunit as the H+-translocating complex. Then the a-subunit, through its interactions and association with the c-subunit, could complex with the c-subunit and form an a-c supercomplex in the FO part of the enzyme. The a-subunit then translocates the coanion/counterion, and together, the a-c supercomplex within FO—and by extension, the FOF1-ATPase/ATP synthase—catalyzes the nearly electroneutral transport of H+ and succinate‒ (or other anions/cations) under physiological conditions in aerobic systems (Fig. 8).Fig. 8Realization of the working model by physical association and structural union between FOF1-ATPase and DcuD and/or TrkA transport systems for efficient energy transduction in biosystems. The image on the left with the a-c supercomplex in the FO portion of FOF1 represents the case under respiratory conditions in aerobic metabolism (adapted and modified from Sobti et al. (2020)21. The diagrams on the right show two cases for anaerobic fermentation, with the a-subunit replaced by the DcuD or the TrkA system, with possibly both coexisting in different fractions, depending on conditions. In principle, in another organism inhabiting a different environment, e.g. alkaliphilic bacteria in soda lakes, these subunits can be conceived to be substituted by a subunit with an Na+-translocating function, working jointly with the c-subunit involved in the antiport of H+ ions51. Together the model presents a unified view of bioenergetic processes, but for nomenclature. See the text for details. Realization of the working model by physical association and structural union between FOF1-ATPase and DcuD and/or TrkA transport systems for efficient energy transduction in biosystems. The image on the left with the a-c supercomplex in the FO portion of FOF1 represents the case under respiratory conditions in aerobic metabolism (adapted and modified from Sobti et al. (2020)21. The diagrams on the right show two cases for anaerobic fermentation, with the a-subunit replaced by the DcuD or the TrkA system, with possibly both coexisting in different fractions, depending on conditions. In principle, in another organism inhabiting a different environment, e.g. alkaliphilic bacteria in soda lakes, these subunits can be conceived to be substituted by a subunit with an Na+-translocating function, working jointly with the c-subunit involved in the antiport of H+ ions51. Together the model presents a unified view of bioenergetic processes, but for nomenclature. See the text for details. It is readily conceivable that the “a-subunit” under respiratory conditions transporting weak organic anions could be replaced by its analog Dcu complex (transporting succinate) or by its country cousin, the TrkA complex (counter-transporting K+ ions) under fermentative conditions, or in aerobic systems in the presence of valinomycin. In this scenario, a chimera or combination of such complexes would coexist in the energy-transducing membrane for addition and joint utilization of energy (Fig. 8). Such a model would reconcile a large amount of information in the field of ATP synthesis/hydrolysis collected by both the Trchounian-Nath groups during the past two decades. Conclusion The concentrations of cations and anions have been measured as a function of time in the presence and absence of DCCD by wild type E. coli cells during glucose fermentation at pH 7.5 under different conditions. These include the presence of glucose at a concentration of 2 g L-1 glucose and 8 g L-1 glucose in the growth medium, and addition of 2 g L-1 glucose in the the absence and presence of 10 mM formate in the external medium. The rates of H+ efflux, K+ uptake, and formate, acetate, lactate, and succinate efflux have been calculated under the various experimental conditions. It has been found that the DCCD-sensitive efflux of H+ through the FOF1-ATPase is almost exactly compensated by the sum of the efflux of succinate anion and the influx of K+ ions. A working mechanism summarizing these findings has been presented. An integrating structural rationale and model involving the FOF1-ATPase, the succinate (e.g. DcuD) and K+ (e.g. TrkA) systems involving association of transporters that together catalyze efficient energy transduction has been formulated. Such a model should prove useful in a number of biological contexts. Future outlook A prognosis for the future looks interesting. Experimentally, the work can be extended on several lines, for example by including assays of ATP hydrolysis activity by FOF1-ATPase and investigation of various mutants along with measurement of transport rates at neutral and acidic pHs during glucose fermentation. New data along these lines on atpB (a) and atpE (c) mutants is presented in Table 2. It is also important to look at cation and anion transport rates at high added glucose concentration, e.g. 8 g L-1. As discussed, glucose consumption and anion transport under these experimental conditions occur very rapidly—within a minute—and fast methods for kinetic measurement need to be innovated. These include either the development of an automated, continuous monitoring protocol, or use of a micromixing device—as in stopped flow equipment—coupled thereafter with HPLC measurements at short sampling times on the order of 15 s. The use of fast 15 s sampling time (compared to ∼ 1 min) was tried out for measurement of potassium ion transport rates on the atp mutants. The results revealed a very rapid inhibition of K+ transport rates in both atpB (a) and atpE (c) mutants in the absence of DCCD—within the sampling time of 15 s, and thereafter. These approaches shall be further explored in future in our collaborative Indo-Armenian research project. Theoretically, it seems very important to understand how charge balance can be maintained by a one-ion-at-a-time, single molecule mode of pumping and function by the FOF1-ATPase against a large, long-range, bulk-to-bulk negative-inside electrical potential in the cell measuring ∼ - 150 mV believed to be created by the cumulative pumping of a large number of H+ ions. The latter would be expected to override the influence of ATPases working in single molecule mode, unless the enzyme molecules are perfectly synchronized. However, the ATPase motors operate independently, and no mechanism for their synchronization has been put forth until today. This is problematic, especially given that the H+ is the primary (first) ion pumped from inside to outside by the FOF1 working in the ATP hydrolysis mode, either electrogenically, or in a cotransport process. The origin and thermodynamic basis for a delocalized potential across biomembranes needs to be revisited. A thermodynamic theory based on fundamental electrochemical and physical chemistry principles, built for example using the pioneering work of J. W. Gibbs49,50, needs to be developed to provide insights into this challenging biological problem.