Octenidine selectively disrupts membrane-coupled photosynthetic functions in intact purple bacteria

Octenidine is a well-known dicationic amphiphile that perturbs membrane structure through bidentate electrostatic anchoring and hydrophobic partitioning. In this study, the effects of octenidine on intact cells of purple photosynthetic bacteria containing a cytochrome subunit (Rvx. gelatinosus) and lacking cytochromes (cycA, the cytochrome-less mutant of Rba. sphaeroides) were investigated. Laser-diode-induced absorption changes and prompt and delayed light were analyzed in the sub-millisecond to millisecond time range. The effects of octenidine on photosynthetic function were concentration-dependent and function-specific. (1) Octenidine had no detectable effect on the pigments or photochemistry of isolated reaction centers (RCs) in detergent solutions, indicating negligible binding to the RC surface or detergent belt. (2) RC photochemistry in intact cells was similarly resistant to octenidine. (3) Only a slight decrease in primary photochemical efficiency was observed at higher (>100 μM) concentrations. (4) Octenidine did not enhance fluorescence emission at 860 nm (LH2) relative to 900 nm (RC–LH1 core complex); instead, it increased non-radiative exciton losses relative to trapping, without significantly altering LH2→RC–LH1 energy transfer above 10 μM. (5) In contrast, octenidine strongly inhibited membrane-associated charge accumulation and energy transduction processes (Δ ψ, proton motive force). The quenching of millisecond delayed light and the inhibition of cytochrome turnover occurred at low (<10 μM) concentrations. (6) Divalent cations (e.g., Mg 2+) partially mitigated these effects, suggesting competitive membrane-binding interactions. (7) Overall, octenidine selectively impairs processes that depend on intact membrane organization, particularly those involving the lateral mobility of redox components, while leaving primary photochemistry largely unaffected. These findings provide mechanistic insight into how antiseptics in clinical use perturb bioenergetic membranes at the level of exciton dynamics and electron transport.

Purple non-sulfur photosynthetic bacteria are benchmark model systems for studying light-energy transduction in anoxygenic photosynthesis (Cogdell et al., 2006; Hunter et al., 2009; Sener et al., 2010; Blankenship, 2021; Fleming et al., 2023; Swainsbury et al., 2023; Fleming and Scholes, 2024; Maity and Kleinekathöfer, 2025). In these organisms, excitation captured by LH1/LH2 antenna complexes is efficiently funneled to the reaction center (RC), where ultrafast charge separation is initiated (Hu et al., 2002). Light-induced absorption changes and both prompt (PF) and delayed (DL) light are tightly coupled not only to excitonic dynamics but also to the redox and energetic state of the RC and the quinone/cytochrome pools (Lavergne et al., 2009; De Rivoyre et al., 2010).

Changes in membrane potential (Δ ψ), proton motive force (pmf), Q A /Q B occupancy, and donor-side electron supply can substantially reshape absorption and fluorescence signals by altering RC turnover and recombination pathways (Okamura et al., 2000; Wraight, 2004, Wraight, 2006). Protonophores and redox-active agents are classical tools for perturbing these parameters: they modify PF and DL by accelerating RC reopening, redirecting back-reaction pathways and shifting the free-energy gap between P + Q A − and P* (Nicholls and Ferguson, 2013; Onidas et al., 2013). These approaches provide sensitive, non-invasive probes of energy-transfer efficiency, RC trapping, and electron-transfer bottlenecks along both donor (cytochrome c/ bc 1) and acceptor (Q A /Q B quinone) branches (Turzó et al., 2000; Maróti et al., 2020; Maróti, 2024; Kis et al., 2024a; Kis et al., 2024b; Trotta and Maróti, 2024).

Octenidine dihydrochloride (OCT) is a widely used cationic bis-pyridine antiseptic with broad antimicrobial activity and good tolerability (Sedlock and Bailey, 1985; Assadian, 2016; Kholina et al., 2021). Mechanistically, OCT binds to negatively charged bacterial envelopes, perturbs lipid packing, depolarizes membranes, and disrupts barrier function in both Gram-negative and Gram-positive bacteria (Wiegand et al., 2008; Malanovic et al., 2020). Time-resolved SAXS studies on model membranes indicate that OCT strongly disrupts lipid hydrogen-bonding networks (McDowell et al., 2026). Such effects are particularly relevant in bacterial membranes, where hydrogen bonding between phosphatidylethanolamine (PE) and phosphatidylglycerol (PG) or cardiolipin (CL) contributes to membrane rigidity (Sevcsik et al., 2008). These perturbations arise from the detergent-like insertion of OCT into lipid bilayers, altering packing and permeability and ultimately leading to membrane depolarization and loss of viability (Gilbert and Moore, 2005; Epand and Epand, 2009).

Beyond antimicrobial lethality at high concentrations, OCT can directly influence pigment–protein photophysics. In oxygenic systems, OCT induces large increases in PSII fluorescence (≈5–10×), slows antenna-to-RC energy migration approximately threefold, and causes blue shifts in absorption and emission spectra—effects consistent with structural and dielectric perturbations of core antenna complexes (CP43/CP47) (Krause and Weis, 1991; Paschenko et al., 2023, 2024). These observations align with established principles whereby modifications of the protein–lipid environment alter excitonic coupling and energy-transfer kinetics within photosynthetic systems (Croce and van Amerongen, 2014). Similar effects have been reported for other membrane-active amphiphiles that perturb pigment organization, membrane fluidity, and supramolecular structure, thereby reshaping excitation-energy flow and fluorescence yield (Garab 2025). Similar disruption of pathways of energy transfer to LH1–RC complex (Knox et al., 2022) and electron transfer (Lukashev et al., 2025) has been observed in bacterial systems.

Recent syntheses of purple-bacterial electron-transfer networks emphasize that, in intact cells, fluorescence relaxation is governed not only by primary RC photochemistry but also by cyclic electron flow during multiple turnovers, including contributions from the cytochrome bc 1 complex and pools of mobile redox carriers. Donor- and acceptor-side limitations, quinone exchange, and diffusional constraints of periplasmic carriers all influence RC reopening and fluorescence relaxation dynamics (Asztalos et al., 2015; Swainsbury et al., 2023; Kis et al., 2024a).

Within this framework, OCT provides a unique opportunity to probe membrane–photophysics coupling. Specifically, does OCT primarily act by (i) weakening antenna–RC coupling and slowing energy migration, (ii) modifying RC back-reaction energetics governing DL yield and decay, or (iii) collapsing Δ ψ / pmf, thereby altering donor- and acceptor-side control of RC reopening? Addressing these questions using time-resolved PF and millisecond DL in intact cells enables a direct assessment of how a clinically relevant antiseptic perturbs exciton dynamics, charge separation, redox turnover, and membrane energetics.

Recent studies on bacterial chromatophores support this rationale, showing that OCT suppresses Δ ψ generation, accelerates pmf decay on the millisecond timescale, and disrupts acceptor-side electron/proton transfer in Rhodobacter sphaeroides [Mamedov et al., 2025; Lukashev et al., 2025).

The present study investigates the effects of OCT on intact purple photosynthetic bacteria by combining kinetic and spectral analyses of absorption and fluorescence over a wide concentration range. Its aim is to disentangle antenna/RC photophysical effects from membrane-electrochemical perturbations and to relate these changes to established mechanisms of photosynthetic energy transduction and OCT action.

More broadly, this study bridges antimicrobial membrane biophysics with fundamental photophysics. It addresses how membrane-active biocides reshape excitation energy transfer, RC trapping, quinone chemistry, and electron transport in vivo. Given the emerging role of purple non-sulfur bacteria in biotechnology—including in wastewater treatment, bioplastics production, and microbial electrosynthesis—understanding the photophysical footprint of OCT may also inform process stability and antimicrobial strategies in applied systems (Morrison and Bose, 2024; Amini et al., 2025). To my knowledge, this is the first study linking a clinically-used antiseptic to time-resolved photophysical signatures of bacterial photosynthesis in intact cells.

Bacterial strains, media, and chemicals

Wild-type strains of purple non-sulfur photosynthetic bacteria possessing tetraheme cytochrome subunits (Rubrivivax gelatinosus) and the cytochrome-less mutant of Rba. sphaeroides (cycA) were used. These strains have been characterized previously (Maróti et al., 2020; Kis et al., 2022, Kis et al., 2024a; Verméglio et al., 2016).

Rvx. gelatinosus cells were cultivated anaerobically in Siström minimal medium in screw-cap flasks under continuous illumination (∼13 W m −2) provided by 40 W tungsten lamps (Maróti and Wraight, 2008). Under these conditions, both cytochrome c 8 and HiPIP function as periplasmic electron donors (Lavergne et al., 2009).

As cycA cannot grow photosynthetically, it was cultivated under semi-aerobic conditions in half-filled Erlenmeyer flasks sealed with cotton plugs and shaken in the dark. Cultures reached the stationary phase approximately 3 days after inoculation.

Cells were harvested during exponential growth and bubbled with nitrogen for 15 min prior to measurements to maintain anoxic conditions.

The bacteriochlorophyll a (BChl a) content of the cells was determined by extraction into organic solvents, followed by absorbance measurements at characteristic wavelengths using published extinction coefficients (ε = 76 mM −1 cm −1 at 771 nm in acetone/ethanol, 7:2 v/v) (Akkerman et al., 2002).

Terbutryn (120 µM) was routinely applied to inhibit interquinone electron transfer from Q A − to Q B in the reaction center (RC). Octenidine dihydrochloride (OCT) was dissolved in ethanol and added to the cell suspension; the final ethanol concentration did not exceed 5%. Control experiments confirmed that ethanol alone (≤5%) had no measurable effect on the recorded signals. Measurements were conducted after 10 min incubation at room temperature.

Optical measurements

Absorption changes

Photochemical activity was monitored via light-induced absorbance changes using a home-built spectrophotometer with a 4 nm bandwidth (Maróti and Wraight, 1988). The oxidation of the RC special pair P (PQ A → P + Q A −) and its subsequent re-reduction—either via charge recombination (P + Q A − → PQ A) or by external donor D (P + D → PD +)—were initiated by a high-power (2 W) blue laser diode (450 nm; Roithner Lasertechnik) and detected at 790 nm.

The photooxidation of cytochrome c (cyt c 2+ → cyt c 3+) was induced using a 1 W red laser diode (802 nm; Roithner Lasertechnik) and monitored at 551 nm relative to 540 nm (Kis et al., 2022).

Prompt fluorescence

Fluorescence emission from antenna BChls was excited using a high-power, 2 W blue laser diode (450 nm; Roithner Lasertechnik) with variable pulse duration. Emission was detected either through a >850 nm high-pass filter or via an IR monochromator (Jobin Yvon H20), using a red-sensitive photomultiplier tube (R928, Hamamatsu) (Maróti, 2024). The excitation beam was oriented perpendicular to the detection axis. Samples were placed in a thermostated 1 × 1 cm quartz cuvette.

Delayed light

Delayed light (DL) measurements were performed using the same optical setup, with the photomultiplier tube Peltier-cooled and protected from intense prompt fluorescence (PF) using an electronically controlled mechanical shutter (Uniblitz VS25, Rochester, NY, United States). The photomultiplier anode current was recorded directly using a digital oscilloscope (Tektronix TDS 3032).

Each experiment was repeated on at least three independent cultures. Fluorescence and absorbance signals were normalized to BChl a content where appropriate. All measurements were performed at room temperature (24 °C) under thermostatic control.

Signal processing and data analysis

The signal-to-noise ratio was improved by averaging up to 32 traces at a repetition rate of 0.03 s −1, allowing the complete relaxation of flash-induced transients between acquisitions. The kinetics were analyzed by a least-squares (Marquardt) fit of a single exponential function. For numerical calculations and simulations, Mathcad 14.0 was used.

Tracking light-induced P/P + by absorption changes

In these experiments, light-induced absorption changes monitored at approximately 790 nm were used as an optical probe of primary donor (P → P +) oxidation in purple photosynthetic bacteria (Kirmaier and Holten, 1990; Maróti et al., 2020; Kis et al., 2024a, Kis et al., 2024b). Whole-cell suspensions were excited by rectangular illumination from a strong blue laser diode (450 nm), and the absorption change was detected during and after illumination (“in the light” and “in the dark,” respectively) (Figure 1). The slowly rising and decaying components were assigned to the generation and decay of P +, respectively and their characteristic rise and decay times were determined from the least-squares (Marquardt) fit of the monoexponential function to the measured kinetics.

A prompt step-like decrease in the signal was observed upon switching on the excitation light. This component was absent after illumination was terminated; this is attributed to light scattering detected perpendicular to the excitation beam. Thus, two overlapping signals of opposite sign contributed to the recorded traces: oxidation of P increased the apparent absorption, whereas light scattering caused an apparent decrease in the signal.

The cytochrome-less mutant of Rba. sphaeroides (cycA) treated with terbutryn provided the simplest system for analysis. Under these conditions, electron transfer was blocked on both the donor and acceptor sides, so the rise reflected primary photochemistry during illumination, whereas the decay followed the charge recombination of P + Q A − → PQ A in the dark. Under these conditions, P + Q A − was formed within approximately 1 ms and decayed on a timescale of approximately 100 ms. Qualitatively similar behavior was also observed when these restrictions were relaxed either on the acceptor side (cycA without terbutryn) or on the donor side (the pufC mutant of Rvx. gelatinosus), although the analysis became more complex (see Supplementary Figures S1, S2).

The effects of OCT on the extracted parameters were examined over a concentration range extending to several hundred μM, at which the suspensions became strongly turbid (Figure 2). The data obtained from whole cells were supplemented by data on charge recombination of isolated RC detected following flash excitation. The measured responses could be grouped into three categories. First, the charge-recombination phase was essentially unaffected by OCT: both the amplitude and the half-time changed only minimally in whole cells and the isolated RC protein. Second, the rising phase showed moderate sensitivity. Both the amplitude and the rise time changed only slightly below 100 μM OCT, whereas larger changes became apparent above this concentration. Third, the light-scattering component showed the largest OCT dependence. Relative to the OCT-free sample, the scattering signal approximately doubled at 50 μM OCT and increased steeply at higher concentrations, where it increasingly distorted the accompanying absorption signals.

Overall, the P/P + absorption changes indicated only limited OCT-dependent modification of excitation delivery to the RC and of primary photochemical trapping. Because the absorption measurements became increasingly affected by turbidity at high OCT concentrations, fluorescence induction was used as a complementary approach to probe OCT effects over a wider concentration range.

Induction of BChl fluorescence reveals OCT-dependent changes in the antenna system

Fluorescence induction provided a suitable set of observables for probing the functional state of the photosynthetic apparatus, especially in the cytochrome-less mutant cycA, in which electron donation to P + is blocked and the system behaves as a single-turnover RC preparation. The measured parameters report on the balance between excitation trapping and non-photochemical decay.

Upon the onset of rectangular excitation, fluorescence increased abruptly from the baseline to the initial level F 0, followed by a slower increase of the variable component (F v) to the maximum level F max (Figure 3). With increasing OCT concentration, F max decreased monotonically. In contrast, F 0 changed much less over the same concentration range. At low OCT concentrations, F 0 remained nearly constant, whereas at high OCT concentrations (>100 μM), it also decreased. As a result, the variable fluorescence, F v = F max − F 0, showed a marked decrease with increasing OCT.

The rise time of the fluorescence induction was affected much less strongly. Only at very high OCT concentrations, in the several-hundred-μM range, was a moderate slowing observed.

The F max spectrum was narrow and exhibited a peak at approximately 890 nm (Figure 4). After the addition of 50 μM OCT, the spectrum broadened slightly but remained consistent with a single dominant band, with no clear emergence of an additional component. Across the measured OCT range, the fluorescence spectra remained dominated by the long-wavelength emission component.

The fluorescence bands of LH1-RC and LH2 complexes are centered at 900 and 860 nm, respectively, and have substantial spectral overlap. To separate their contributions, the OCT dependence of fluorescence induction parameters measured at 860 nm and 900 nm wavelengths was compared (Figure 5). The trends measured at 860 nm closely followed those at 900 nm in the fluorescence induction. F max decreased progressively with increasing OCT at both emission wavelengths. The decrease was relatively greater at 900 nm than at 800 nm. The F v / F max ratio also declined with OCT concentration, with the decrease more pronounced at 860 nm than at 900 nm at high OCT concentrations. Even at a several hundred μM OCT concentration, F v / F max decreased from about 0.9 to about 0.7 at 900 nm. In contrast, the rise times changed only slightly and remained within the experimental error over most of the concentration range.

Multiple turnover under continuous illumination

When the closed RC state (P + Q A −) can be reopened by electron transfer from the donor side and onward transfer to the acceptor side, repeated charge separations may occur during continuous illumination. These multiple turnovers are accompanied by charge accumulation and the formation of a transmembrane proton motive force (pmf). OCT effects on such processes were examined by millisecond delayed light (DL) and cytochrome turnover measurements.

Millisecond delayed light

The decay of DL was measured in intact cells of Rvx. gelatinosus after 10 ms excitation with a high-intensity blue laser diode as a function of OCT concentration (Figure 6). At the end of the excitation, the opening of the mechanical shutter and acquisition of the DL signal were simultaneously triggered. Full opening of the shutter required approximately 3 ms. The DL amplitude and decay time were evaluated by accounting for the convolution of the shutter-opening kinetics with the exponential decay of DL.

The addition of OCT had little effect on the decay time but strongly reduced the DL amplitude—even at low concentrations where light scattering and major alterations of primary photosynthetic signals were not yet evident. In the cytochrome-less mutant cycA, no corresponding OCT-induced quenching of millisecond DL was observed (Supplementary Figure S3).

Cytochrome turnover

The kinetics of cytochrome oxidation were measured from absorption changes at 551 nm relative to 540 nm during intense rectangular excitation in intact cells of Rvx. gelatinosus (Figure 7). The accumulation of oxidized cytochromes increased rapidly upon illumination and reached saturation within a few milliseconds. After the addition of OCT, the kinetics changed substantially: the rise time remained nearly unchanged (approximately 500 μs), whereas the saturation level decreased strongly with increasing OCT concentration. Approximately 10 μM OCT reduced the saturating signal by about half.

The OCT dependence of the cytochrome signal was also examined under different ionic-strength conditions. In a standard Siström medium of low (≈10 mM) ionic strength, the half-maximal effect occurred at approximately [OCT] 1 / 2 ≈ 20 μM (Figure 7, right panel). In the presence of 200 mM NaCl, this value shifted to approximately 40 μM, whereas in 100 mM MgCl 2 (=300 mM ionic strength) it shifted to approximately 9–10 μM. Thus, NaCl and MgCl 2 affected the OCT dependence differently.

Differential sensitivity of photosynthetic functions to OCT

These experiments show that the structural and functional components of purple photosynthetic bacteria differ markedly in their sensitivity to OCT. The primary reaction-center photochemistry remained largely unaffected, and the LH1–RC antenna function showed only moderate sensitivity, whereas membrane-dependent processes associated with multiple turnovers, such as cytochrome photooxidation and membrane energization, were highly vulnerable. These observations indicate that OCT acts predominantly at the level of membrane organization and membrane-coupled function rather than by directly disrupting the intrinsic photochemistry of the RC.

Primary charge separation is structurally protected

The charge-recombination experiments demonstrated that OCT had essentially no effect on the decay kinetics of charge separation in the isolated RC protein. Despite the presence of pronounced charged domains on the RC surface and detergent belt around the isolated protein, no measurable perturbation of the primary photochemical reactions was observed. Similarly, RC photochemistry remained largely intact in the membrane environment of whole cells after OCT treatment.

These findings indicate that the primary charge-separated state is governed mainly by the local structural and dynamic properties of the cofactor environment, including fixed cofactor geometry, hydrogen-bonding interactions, and nuclear reorganization within the low-dielectric protein interior. Electrostatic perturbations caused by superficial binding of a cationic antiseptic are therefore strongly attenuated before they can influence the buried cofactors involved in charge recombination.

Two factors probably underlie this robustness. First, the protein matrix provides substantial dielectric shielding. The relevant cofactors are located well inside the RC, typically at distances of roughly 15–20 Å from the protein surface, where external charges are screened effectively. Second, the energetics of charge recombination are dominated by the large reorganization energy of the system. Any small OCT-induced electrostatic shift at the protein surface is expected to be negligible relative to the reorganization energy governing the electron-transfer event.

Thus, primary charge recombination in bacterial RCs appears to be a structurally protected quantum-mechanical process that is remarkably insensitive to perturbations at the membrane surface. The pronounced OCT effects observed in intact cells, therefore, most likely arise indirectly, through membrane disorder, collapse of electrochemical gradients, and the impairment of auxiliary electron-transfer processes, rather than from the direct modification of intrinsic RC photochemistry.

Light scattering is dominated mainly by non-cellular OCT species at higher concentrations

Upon the addition of OCT to bacterial suspensions, the light-scattering signal remained low below approximately 10 μM and then increased sharply at higher concentrations. At least three components may contribute to the observed scattering: (i) baseline scattering of untreated intact cells, (ii) OCT-dependent modification of cellular scattering caused by membrane-bound OCT, and (iii) scattering from OCT aggregates or other non-cellular OCT species in the aqueous phase.

The contribution of intact cells is expected to change only modestly. At the cell densities used here, scattering from individual bacteria should be approximately additive because multiple scattering and intercellular optical interactions are weak. OCT may alter cellular scattering by affecting cell aggregation, swelling, refractive index, or membrane packing. However, these effects are partly compensatory: aggregation increases the effective size of scattering objects but simultaneously reduces their number. Therefore, even if OCT modifies the optical properties of the cells, the resulting change in cell-derived scattering is not expected to be dominant.

In contrast, OCT molecules, and especially OCT aggregates, in the aqueous phase are expected to contribute increasingly at higher concentrations. Because OCT is amphiphilic and tends to self-associate, the bulk suspension likely contains an increasing population of non-cellular scattering species once the membrane-binding capacity of the cells begins to saturate. Order-of-magnitude estimates support this interpretation. Even if each cell binds a substantial number of OCT molecules, the total cell-bound OCT pool remains relatively small compared with the total OCT added at concentrations in the tens to hundreds of micromolar range. Accordingly, at higher OCT concentrations, the measured scattering signal is likely dominated by non-cellular OCT species rather than by OCT-modified cells.

This distinction is important when interpreting absorption measurements. In intact cells, the internal chromatophores remain optically nested within the bacterium and do not behave as an independently suspended population. Therefore, the scattering problem is much simpler than in isolated chromatophore suspensions, where the swelling, fragmentation, and aggregation of membrane vesicles can make the optical response substantially more complex.

OCT affects fluorescence mainly through the core LH1–RC domain

The fluorescence-induction experiments provide a useful functional window into how OCT affects the light-harvesting system. The main experimental observations were (i) marked decreases in F max and F v / F max and (ii) little or no major change in the F max (860)/ F max (900) ratio or in the closure time of the RC.

Taken together, these findings argue against a dominant inhibition of forward LH2→LH1 energy transfer. A major decrease in LH2→LH1 transfer would be expected to redistribute fluorescence toward LH2 emission and increase the F max (860)/ F max (900) ratio. However, this was not observed. Instead, the data are more consistent with increased excitation losses and reduced effective trapping within the LH1–RC core domain.

Within the framework of the simple antenna model (see Supplementary Notes), the decrease in F v / F max can be explained by an increase in the effective loss-to-trapping ratio, k 1 / k trap. This may result from increased dissipation in or near the LH1–RC complex, reduced efficiency of delivery to the RC trap, or both. The fluorescence data do not require a strong inhibition of LH2→LH1 connectivity and, if anything, argue against a major blockade of that pathway in intact cells.

This interpretation is biophysically plausible. OCT is a cationic amphipathic membrane-active compound that perturbs lipid organization, local dielectric properties, and protein–lipid interactions. Such perturbations are expected to affect the structurally integrated LH1–RC functional unit more readily than sever the LH2→LH1 transfer completely. In this view, OCT primarily alters the local membrane/protein environment of the core complex, thereby increasing energy loss or reducing trapping efficiency, while leaving the basic connectivity of LH2 and LH1 largely intact.

The RC closure time changed only slightly in our experiments. This is also consistent with the model predictions if the cells are well dark-adapted and the initial fraction of open RCs is close to unity. Under these conditions, moderate changes in loss or trapping efficiency can strongly influence fluorescence yield while only weakly affecting the overall time scale of RC closure.

Why do intact cells differ from chromatophore preparations?

Our results differ in part from reports on chromatophore preparations, in which OCT and related membrane-active compounds have been associated with stronger spectral redistribution and more pronounced evidence of antenna uncoupling (Strakhovskaya et al., 2021; Knox et al., 2022; Lukashev et al., 2025; Paschenko et al., 2024). Several factors may account for this difference.

First, the physical pathway of OCT to the photosynthetic membrane differs between intact cells and isolated chromatophores. Chromatophores are bare membrane vesicles, whereas intact cells contain outer membrane, periplasm, and intracellular membrane organization. However, the overall OCT sensitivity observed here suggests that membrane access itself is not the main limiting factor. OCT does not need to completely traverse the membrane; it can insert into membrane phases and redistribute laterally. Thus, the main determinant of its action is likely the membrane environment rather than simple penetration depth.

Second, the redox and structural state of chromatophore preparations is usually less well defined than that of inhibitor-treated intact cells. In terbutryn-treated cycA cells, donor- and acceptor-side reopening pathways are strongly restricted, and RCs are rapidly driven into a blocked state under illumination. Under such conditions, fluorescence induction reports on a relatively well-defined functional system. In chromatophore preparations, in contrast, heterogeneity in donor/acceptor composition, vesicle integrity, and membrane organization can produce more complex mixtures of states and kinetics.

Accordingly, the difference between chromatophore and whole-cell results may not reflect fundamentally different OCT chemistry but rather different physical organization and functional boundary conditions of the photosynthetic membrane system.

Membrane-associated turnover and energization are the most vulnerable targets

The strongest OCT effects were observed for millisecond delayed light and cytochrome turnover, both of which depend on intact membrane organization and on the mobility and connectivity of redox carriers. In contrast to primary photochemistry, these processes require coordinated multiple turnovers, quinone/cytochrome traffic, and maintenance of a transmembrane electrochemical gradient.

The strong sensitivity of delayed light to low OCT concentrations indicates that membrane energization is an early and particularly vulnerable target. The absence of a corresponding effect in the cytochrome-less mutant cycA further supports the conclusion that OCT mainly interferes with the membrane-associated processes required for repeated turnovers and charge accumulation rather than with the single primary photochemical event itself.

Similarly, the marked decrease in cytochrome photooxidation with increasing OCT concentration indicates a disruption of donor-side function. Because electron donation to the RC in intact cells depends on the structural integrity, spatial organization, and mobility of membrane-associated and periplasmic components, these data strongly support a membrane-level mode of action of OCT.

Based on some mechanistic aspects, the possibility should be addressed that OCT affects donor-side electron transfer not only indirectly via membrane energetics but also by perturbing interactions between membrane-bound and soluble redox partners. 1) Spatial organization effects. The RC–cytochrome subunit and the cytochrome bc 1 complex are embedded in the membrane, while soluble donors such as cytochrome c 8 or HiPIP shuttle electrons within the periplasmic space. The OCT-induced perturbation of membrane structure (e.g., altered lipid packing, increased disorder, or local curvature changes) may modify the relative spatial arrangement of these membrane protein complexes, thereby affecting the effective electron-transfer distances. 2) Diffusion constraints in the periplasm. Although the electron carriers diffuse in an aqueous phase, their motion is not purely free diffusion. It is influenced by transient binding interactions, electrostatic steering, and macromolecular crowding near the membrane surface. Changes in membrane organization can propagate into the adjacent periplasmic environment, potentially altering diffusion pathways and encounter frequencies. 3) Electrostatic and interfacial effects. OCT, as a cationic amphiphile, may modify the surface charge distribution of the membrane. This could influence docking interactions between soluble donors and their target complexes, which are often partially electrostatically guided. 4) Relative contribution compared to Δψ effects. Although these donor-side perturbations are plausible, our current data suggest that the dominant effect of OCT is the dissipation of the proton-motive force (particularly Δ ψ), as reflected in the strong suppression of delayed light and cyclic turnover. Any additional donor-side effects are, therefore, likely secondary but potentially synergistic.

The ionic strength experiments reinforce this picture. The shift of the OCT response in the presence of NaCl is broadly consistent with electrostatic screening. The different behavior observed with MgCl 2 indicates that simple Debye screening cannot fully explain OCT action. Divalent cations likely modify both the membrane surface and OCT speciation, thereby altering membrane susceptibility in ways not captured by a purely electrostatic model (Clifton et al., 2015).

A membrane-binding framework for OCT action

The overall behavior of OCT can be understood in terms of coupled electrostatic accumulation and hydrophobic membrane partitioning. Because OCT is dicationic and amphiphilic, negatively charged bacterial membranes create a strong local enrichment of the compound at the membrane interface. Once concentrated there, OCT can bind to and insert into the lipid matrix, perturbing packing, permeability, membrane potential, and lateral organization.

A Gouy–Chapman–Langmuir-type framework is, therefore, a useful phenomenological description of OCT action (see Supplementary Notes). In such a model, the apparent membrane affinity of OCT is enhanced by electrostatic attraction but becomes progressively self-limited as membrane binding neutralizes the surface charge. This provides a natural explanation of the extended concentration dependence of functional inhibition rather than a simple sharp threshold.

The membrane-binding model is especially useful for delayed light and cytochrome turnover because these observables report emergent membrane function rather than a single isolated molecular event. While this framework does not include all complications of OCT chemistry, such as self-association, aggregate formation, or multiple binding modes, it captures the essential principle that OCT action is amplified by preferential accumulation at negatively charged membranes.

This study provides a mechanistic picture of how octenidine disrupts photosynthetic membrane function in purple bacteria. The different photosynthetic components and processes display a clear hierarchy of sensitivity to OCT. Primary charge separation in the RC remains largely unaffected, and excitation harvesting and trapping in the LH1–RC core are only moderately impaired, whereas membrane-dependent multiple turnover and proton-motive-force generation are strongly inhibited.

These findings show that OCT primarily acts by compromising membrane-coupled energy conversion rather than by directly blocking the intrinsic photochemistry of the reaction center. The principal antimicrobial action of OCT on the photosynthetic apparatus is, thus, exerted through membrane destabilization, altered membrane organization, increased dissipation, and loss of effective electrochemical coupling.

More broadly, this study links the action of a clinically relevant antiseptic to fundamental photophysical and bioenergetic processes in intact photosynthetic bacteria. It also highlights the value of time-resolved absorption and fluorescence measurements for resolving how membrane-active compounds differentially affect primary photochemistry, excitation trapping, redox turnover, and membrane energization.