Is a constitutive red-shift an advantage for oxygenic photosynthesis under M dwarf starlight? Insights from Acaryochloris marina sp. str. Moss Beach

Abstract

In the next decades, space telescope missions will search for life evidence on exoplanets, focusing on robust biosignatures associated with oxygenic photosynthesis, including atmospheric oxygen accumulation and the Vegetation Red-Edge in surface reflectance spectra. Many habitable rocky exoplanets orbit M dwarf stars, whose spectral energy distribution may condition the rise and evolution of oxygenic photosynthesis. M dwarf stars emit predominantly far-red (700–750 nm) and near-infrared (750–1000 nm) light, and relatively little visible (400–700 nm) radiation, which on Earth predominantly drives photochemistry in most oxygenic phototrophs. Previous experiments proved some oxygenic phototrophs can photosynthesize under simulated M dwarf light but less efficiently than under solar radiation simulated in the range 365–780 nm. Indeed, tested organisms present photosynthetic apparatus evolved to harvest Sun’s visible light, however, M dwarfs’ irradiation might select adaptations optimized for harvesting far-red/near-infrared light. We measured sensitivities of a far-red/near-infrared-utilizing cyanobacterium, Acaryochloris marina sp. str. Moss Beach to simulated M dwarf spectrum and primeval anoxic, CO₂-rich atmosphere. This strain constitutively presents a high content of chlorophyll d, with in vivo absorption peak at 710 nm. Its permanently red-shifted photosynthetic apparatus required no acclimation to the stellar spectrum, maintaining strong growth and oxygen production, higher than that registered under simulated solar light. Moreover, abundant chlorophyll d caused a shift in whole-cell reflectance: the red-edge was beyond 700 nm, resulting in a Chl d-near-infrared-edge. Overall, a potentially similar metabolism on exoplanets orbiting M dwarfs could successfully produce both a gaseous biosignature and a characteristic surface biosignature.

A cyanobacterium evolved to photosynthesize with low energetic far-red light efficiently grows and produces oxygen in simulated planetary conditions, suggesting photosynthetic life on planets around red-dwarf stars could function evolved and generate characteristic and detectable signatures.Created in BioRender. Liistro, E. (2026) https://BioRender.com/j2de4ay

Introduction

In the last 20 years, space missions such as Kepler, the Transiting Exoplanet Survey Satellite (TESS, NASA) and the Global Astrometric Interferometer for Astrophysics (GAIA, ESA), have discovered a large number of exoplanets, over 6000 at the time of writing (From www.exoplanets.nasa.gov consulted on April 8, 2025). Some of these planets have been recognized as potentially habitable as their features and those of their host star(s) may support the presence and evolution of life (Kasting et al. 1993, Kaltenegger 2017, Méndez et al. 2021). Among these, rocky Earth-like exoplanets were found to be common around M dwarf stars (Kopparapu et al. 2013, Hsu et al. 2020). These stars, also referred to as M stars or red dwarfs, are considered viable targets for the search for life as not only do they frequently host rocky planets but they also have a lifespan long enough to allow for a biosphere to evolve (Shields et al. 2016). Moreover, their strong early stellar activity is considered to decline to solar levels within ∼3 billion years, thus not precluding habitability (Gale and Wandel 2016, O’Malley-James and Kaltenegger 2017, Schwieterman et al. 2018).

In the search for life on planets orbiting M dwarf stars, in situ detection is not feasible, as the closest M dwarf, Proxima Centauri b, is 4.2 light-years from Earth (Anglada-Escudé et al. 2016). Life detection must thus rely on remote sensing of biosignatures: patterns or molecules of unequivocal biological origin (Schwieterman et al. 2018, Claudi and Alei 2019). Missions such as the Atmospheric Remote-Sensing Infrared Exoplanet Large-survey (ARIEL, ESA) and the Habitable Worlds Observatory (HWO, NASA) will be launched in the coming years with the purpose of characterizing exoplanets atmospheres, looking for biosignatures (Tinetti et al. 2018, Feinberg et al. 2026). One of the strongest remotely detectable biosignatures on our planet is the presence of oxygen, O₂, and its byproduct ozone, O₃, in a planet’s atmosphere (Schwieterman et al. 2018, Cavalazzi and Westall 2019, Claudi and Alei 2019). Since on Earth the abundance of these molecules is caused by oxygenic photosynthesis, the metabolic process operated by cyanobacteria, algae, and plants (Cavalazzi and Westall 2019), scientists have been discussing on the feasibility of this metabolisms around M dwarf stars (Tinetti et al. 2006, Kiang et al. 2007a,b).

M dwarfs are cold and faint stars that emit mainly far-red (FR, 700–750 nm) and near-infrared (NIR, 750–1000 nm) light, and very low visible (VIS, 400–700 nm), in contrast to G stars, like the Sun (Chabrier et al. 1996, Pecaut and Mamajek 2013), whose emission is stronger in the visible. This emission spectrum challenges the possibility of canonical oxygenic photosynthesis to have evolved around such planets as this process on Earth relies mainly on VIS light, which is preferentially harvested by most common pigments and antenna complexes. Nonetheless, models indicate that the small amount of VIS light reaching a planet orbiting an M dwarf star could still allow for oxygenic photosynthesis to work efficiently, with a productivity estimated to be up to the 22% of that on Earth (Gale and Wandel 2016, Ritchie et al. 2018, Gray et al. 2025). If photosynthetic adaptations such as three or four photosystems in series have evolved around M dwarf stars, the productivity of oxygenic photosynthesis has been estimated to potentially increase to over the 75% of Earth’s (Kiang et al. 2007b). Additionally, oxygenic photosynthetic organisms generate another biosignature that is among the most studied: the Vegetation Red-Edge, VRE. VRE is a sharp increase in reflectance around 700 nm toward the NIR, as a consequence of the contrast between the high absorption in the visible, due to chlorophyll molecules, and the scattering of NIR due to cell structures (Gausman et al. 1975, Tucker et al. 1985, Grant 1987, Sagan et al. 1993, Turnbull et al. 2006, O’Malley-James and Kaltenegger 2018). It was proposed that, around an M dwarf, this feature of (eventual) oxygenic phototrophs could be altered because of the need to harvest longer wavelengths (Wolstencroft and Raven 2002). Due to the presence of FR or near-infra-red (NIR) harvesting pigments, the VRE might be shifted and become a “NIR” edge (Tinetti et al. 2006, Kiang et al. 2007a).

Recently, it was experimentally demonstrated that oxygenic photosynthesis and growth is possible under a simulated M dwarf spectrum in several Earth organisms, with different efficiencies (Boccia et al. 2026, Liistro et al. 2024, Battistuzzi et al. 2023a, 2023b). Cyanobacteria are ideal targets for these studies, as in recent decades some strains were found capable of photosynthesizing not only with visible light but also with solely FR light via the Far-Red Light Photoacclimation (FaRLiP) (Gan et al. 2014). In these species, the FaRLiP response is enabled by the presence of a specific gene cluster expressed in FR light which leads to the synthesis of (i) red-shifted pigments (chlorophyll d, Chl d, and chlorophyll f, Chl f), (ii) allophycocyanin proteins with red-shifted chromophores (far-red allophycocyanin, FR-APC), (iii) paralog subunits of photosystems that incorporate the pigments. With this remodeling of the photosynthetic apparatus, organisms present a shoulder of absorption in the FR and fluorescence emission shifted to longer wavelengths (Gan et al. 2014, Ho, Soulier, et al. 2017). Chlorogleopsis fritschii PCC6912 acclimated to the simulated M dwarf spectrum showed the activation of FaRLiP, leading to the use of both the visible and FR light available in a simulated stellar spectrum (Battistuzzi et al. 2023a).

These results are fundamental to understand the limits of oxygenic photosynthesis around M dwarfs, however, they all consider organisms evolved for photosynthesizing with visible light and that canonically harbor VIS photosystems, with some exception of strains that at most can acclimate to increase their FR absorption capacity through uphill energy transfer (Elias et al. 2024). It was hypothesized that the eventual evolution of oxygenic phototrophs under the irradiance of M dwarfs could have selected particular adaptations such as red-shifted pigments, modified antennae, or a higher number of photosystems working in series (Kiang et al. 2007a,b; Takizawa et al. 2017, Chitnavis et al. 2024).

Curiously, on Earth we have examples of organisms, adapted to FR-enriched light niches, with a unique photosynthetic apparatus that has constitutive and strongly red-shifted absorbance, namely in the cyanobacterial strains of the species Acaryochloris marina (Miyashita et al. 1996, Larkum and Kühl 2005). A. marina strains are characterized by the permanent large presence of Chl d in Photosystem I & II, accounting for more than 90% of the total chlorophyll content, and participating both in light harvesting and charge separation, with maximum absorption at 740 nm (Chen et al. 2002, Badshah et al. 2017). It is also interesting to notice how ancestral the divergence of this genus is in the evolution of oxygenic photosynthesis: the split of the Acaryochloris genus from the cyanobacterial species tree was found to precede the appearance of the FaRLiP phenotype, dated in the Paleoproterozoic (Antonaru et al. 2025).

When considering the eventual evolution of oxygenic photosynthesis around M dwarfs, A. marina is an excellent example of the evolution of a photosynthetic apparatus that mainly harvests FR wavelengths.

In this work, we evaluated the response of a strain of Acaryochloris marina to simulated light and atmospheric conditions. Experiments were designed to characterize basic ecophysiology of this strain, to distinguish whether this strain acclimates to a change in light spectra between FR and solar light, and to quantify its activity under simulated M dwarf light in an anoxic Archean-like atmosphere as a potential scenario for far-red photosynthesis as an early metabolism on a planet around an M dwarf.

Materials and methods

Tested organism and experimental plan

The organism tested in this study was Acaryochloris marina sp. str. Moss Beach. A. marina sp. str. Moss Beach was found as an epiphyte of a brown macroalga, in the rocky intertidal zone of the J.V. Fitzgerald Marine Reserve, California, and the Chl d in its photosynthetic apparatus can reach up to 99% of the total chlorophyll content (Kiang et al. 2022). Due to its Chl d-bearing photosystems, the in vivo Q_y absorption peak of A. marina sp. str. Moss Beach is at 705 nm, with its tail arriving at 750 nm. This strain lacks the accessory pigments phycobiliproteins, hence the light harvesting is only operated with chlorophylls and marginally with carotenoids, similarly to other strains of A. marina, as A. marina CCMEE 5410 and A. marina HICR111A (Chan et al. 2007, Mohr et al. 2010, Kiang et al. 2022).

As this strain is of recent discovery, we first identified the medium, temperature and light intensity suitable for the simulations. We then assessed whether A. marina sp. str. Moss Beach performed a physiological acclimation to a simulated M-dwarf spectrum, comparing the response with cells exposed to solely FR light and solar light. Finally, we proceeded by investigating its performances in simulated planetary conditions combining irradiation with a simulated anoxic primeval atmosphere.

Cultivation conditions for preliminary physiological sensitivity experiments

We obtained the initial inoculum of Acaryochloris marina sp. str. Moss Beach, isolated at Moss Beach, in the J.V. Fitzgerald Marine Reserve, California from co-authors Kiang and Parenteau (Kiang et al. 2022). The strain was maintained in liquid Marine-BG11 medium, MBG11, at 25°C under 15 µmol of photons m⁻²s−¹ of continuous white fluorescent light (L36W-840, OSRAM).

For the experiment testing sensitivity to growth media and temperature, 50 mL inocula with initial optical density at 750 nm, OD₇₅₀, of 0.2 were grown either in MBG11 medium or in ASNIII medium, a common medium for marine cyanobacteria. MBG11, rich in iron, is the medium suggested for A. marina, but the presence of small agglomerates hinders the maintenance of fully axenic cultures. ASNIII, on the other hand, is a common medium for marine strains and facilitates the maintenance of clean cultures but has a lower salinity and a lower iron content. For a more detailed composition of the two media Table S1 can be consulted. Concurrently three different growth temperatures were tested: 20°C, 25°C, and 30°C.

Cultures were exposed to these six different conditions–20°C, 25°C, 30°C in either MBG11 or ASNIII—under continuous irradiation of 15 µmol of photons m⁻²s⁻¹ of white light (TRUE-LIGHT 18T8, Lightfull). For each tested medium, four biological replicates were grown at 20°C, at 25°C, or at 30°C.

For the experiment testing sensitivity to light intensities, cultures with initial OD₇₅₀ of 0.2 were grown at 25°C in liquid MBG11 in a multi-cultivator (Multi-Cultivator MC 1000-OD, Photon Systems Instruments, Czech Republic). Three biological replicates were exposed to 10, 20, 40, 60, 80, 100, and 150 µmol of photons m⁻²s⁻¹ of continuous cold white light provided by built-in LEDs in the multi-cultivator.

Cultivation conditions for planetary simulations

Planetary conditions in terms of irradiance and atmosphere were simulated with a custom-made system extensively described in Battistuzzi et al. (2020, 2023a) and Claudi et al. (2021). Briefly, the set-up is comprised of three light sources and three growth chambers.

The light source for planetary simulation is a custom-made starlight simulator, SLS, which for this work was used to reproduce the irradiation of a quiescent (no or negligible UV) M dwarf (M7) in the range 350–850 nm (Fig. S1). For comparison, cultures were exposed also to a simulated solar light (SOL) and monochromatic FR light (FR). SOL light was provided by a custom-made visible-infrared light simulator with four independently manageable led channels in the range 365–800 nm. Channel 1 controls LEDs at 365 nm, 385 nm, 405 nm, 425 nm, and 500 nm; channel 2 controls SUNLIKE 5000 K and 690 nm LEDs; channel 3 (FR) controls LEDs at 730 and channel 4 (IR) controls LEDs at 750 nm and 780 nm. Channel intensity was regulated in order to accurately reproduce the Sun emission spectrum (Fig. S1). FR light peaked at 730 nm was provided with the illuminator described in Battistuzzi et al. (2023a) (Fig. S1a). For all light conditions, M7, SOL, FR, the total light intensity was 20 µmol of photons m⁻²s⁻¹, resulting respectively in 7.38, 16.92, and 1.47 µmol of photons m⁻²s⁻¹ of VIS light and 12.62, 3.08, and 18.53 µmol of photons m⁻²s⁻¹ of FR light. A more detailed repartition of light in the different wavebands is available in Fig. S1c.

For experiments investigating acclimation to the M7 starlight, 50 mL of fresh cultures at OD₇₅₀ 0.2 were transferred from white light to M7, SOL, and FR light. Growth and in vivo absorption were monitored periodically for 21 days, after which cultures were examined for physiological changes via in vivo absorption and fluorescence, pigment composition analysis and confocal microscopy.

For planetary simulations, where an anoxic primeval atmosphere was combined with stellar irradiation, atmosphere simulating chambers, ASCs, were used. Our ASCs are growth chambers that allow cultivation of organisms underneath both chosen light spectral irradiances—in this study M7, SOL and FR (Fig. S1b)—as well as prescribed gas composition. Inside the ASCs are one sensor for O₂ concentration (LOX-O2-S, SST sensing) and two sensors for CO₂ concentration (CO2M-20 and CO2M-100, SST sensing), which enable to monitor in real time and register the gas consumption and release during the experiment. In this work, the initial atmosphere inside the ASCs was composed of 95% nitrogen, N₂, and 5% CO₂, simulating an anoxic primeval planetary atmosphere. This composition approximates that of the Archean at ∼2.46 Ga just before the Great Oxidation Event (GOE), when atmospheric O₂ was negligible, CO₂ was 0.05–0.15 bar, and CH₄ had reduced to low levels >20 ppm (Catling and Zahnle 2020). We neglect the low CH₄ content for simplicity and note that it would not have affected the surface irradiance at this low ratio of CH₄ to the CO₂ (Trainer et al. 2006, Arney et al. 2016). The ASCs were then sealed over the course of experiments in order to monitor the impact of cyanobacteria on the initial atmosphere.

Raw data of gas concentrations in ppm registered during the experiments were converted in amount of µmol through Matlab R2018b (MathWorks) by applying the law of ideal gasses and considering the ASCs volume, 0.45 L (Battistuzzi et al. 2023a). Experiments with ASCs were run in parallel under M7, SOL and FR for four biological replicates.

Growth assessment

Growth of A. marina sp. str. Moss Beach was assessed through evaluation of optical density at 750 nm, OD₇₅₀, measured with a spectrophotometer (Cary100 UV-VIS, Agilent) against a blank (culture medium). OD₇₅₀ was measured in intervals of 2–3 days. Maximum growth rate µ was calculated on the steepest slope of growth curves as follows:

(1)

were OD_750f is the OD₇₅₀ value at the end of the interval, OD_750i is the value at the beginning of the interval and d is the number of days passed between the two points.

In vivo absorption, reflectance, and fluorescence

To record in vivo absorption spectra of cells, samples were centrifuged at 1400 g for 10 min at room temperature and the pellet was homogenized with a pestle. Six hundred microliter of fresh medium was added to solubilize the pellet and samples were analyzed with a Cary100 UV-VIS spectrophotometer (Agilent) inside quartz cuvettes. The opaque side of the cuvette was exposed to the light ray to correct the scattering (Shibata 1959, Amesz et al. 1961, Gan et al. 2014).

In vivo reflectance of cells was recorded with the set-up described in Liistro et al. 2026a). Briefly, a halogen lamp (Halostar Starlite 12 V 20 W G4, Osram) was used to irradiate 5 ml of liquid cultures with OD₇₅₀ of 5 in a petri dish and reflectance was collected in the range 400–1000 nm with an optical fiber connected to a spectrometer (Flame, OceanOptics) via the SpectraSuite software (OceanOptics). Acquired spectra were normalized against a white reference. To compare A. marina sp. str. Moss Beach acclimated to the M7 simulated starlight, we analyzed cultures of Synechocystis sp. PCC6803 and Chlorogleopsis fritschii PCC6912 acclimated to the same simulated stellar spectra for control reflectance spectra. Information on the acclimation that these two strains perform under this condition can be found in Battistuzzi et al. (2023a, 2023c).

Low temperature (77 K) in vivo fluorescence emission spectra were recorded with a Cary Eclipse Fluorescence Spectrometer (Agilent). Two milliliter of culture were centrifuged at 1500 g for 10 min at 4°C and the pellet resuspended in 1 mL of glycerol 60% (w/v) and 10 mM Hepes pH 7.5. Samples were frozen in liquid nitrogen and kept in dark at −80°C until analysis. For the analysis, samples were excited at 440 nm and emission spectra were recorded between 600 nm and 900 nm. Low temperature (77 K, −196.15°C) during recording of emission spectra was maintained with liquid nitrogen.

Pigment content

For the extraction of lipophilic pigments, samples were centrifuged for 10 min at 17 500 g at 4°C and to the pellet were added an equal volume of 150–212 µm acid washed glass beads (SIGMA) and 50 µl of HPLC grade methanol 100%. Three cycles of cell rupture were executed with a Bullet Blender Storm Pro Homogenizer (Next Advance) for 15 s at maximum speed alternated with 30 s in ice. One hundred fifty microliter of methanol were added to the lysate and a last cycle of rupture was executed. Samples were centrifuged at 20 000 g for 10 min at 4°C and the supernatant, containing extracted pigments, was transferred to a new tube and kept in ice. 200 µl of methanol were added to the pellet and, after vortexing for resuspension, another centrifugation was performed. The supernatant was pooled with the previously obtained extract. These last steps were repeated until the supernatant appeared transparent, indicating full extraction of pigments. Samples were kept in dark at −20°C until analysis. Pigment extracts were analyzed with a Cary100 UV-VIS spectrophotometer (Agilent). Concentration of chlorophyll a and chlorophyll d were obtained with the following equations (Ritchie 2006):

(2)

(3)

To evaluate the relative content of chlorophyll and carotenoids, RP-HPLC analyses were conducted on pigment extracts. Analyses were performed with an Agilent 1100 series LC with a Lichrospher 100 RP column (250 mm length, 4 mm diameter) containing 5 μm silica particles coated by C-18 atom chains (Merck) as the stationary phase. The mobile phase consisted in buffer A (methanol: acetonitrile: water in ratio 42:33:25) and buffer B (methanol: acetonitrile: ethyl acetate in ratio 50:20:30). Buffer A and B were eluted in the column according to the protocol optimized by Gan and collaborators (Gan et al. 2014). The relative content was assessed by integrating the area of each peak in chromatograms at 440 nm for four biological replicates.

Confocal imaging

A. marina sp. str. Moss Beach cells were imaged with a confocal microscope (STELLARIS 8, Leica) to check for eventual changes in dimension, morphology and chlorophyll localization. Cells were excited by a laser diode with emission at 405 nm and imaged with two different channels to separate the fluorescence signal from chlorophyll a (672–689 nm) and chlorophyll d (705–739 nm).

Statistical analyses

Statistical analyses, ANOVA analysis (one-way or two-way ANOVA depending on the dataset) and Tukey’s or Šídák’s multiple comparison tests, were performed with the software GraphPad Prism v10.1.0 (GraphPad).

Results

Responses to nutrients, temperatures and light intensities in A. marina sp. str. Moss Beach

Acaryochloris marina sp. str. Moss Beach has not previously been characterized for the sensitivity of its growth to environmental parameters. Prior to the planetary simulation, we performed preliminary experiments investigating how temperature, growth media and light intensities affected the growth rate of this strain. Being originally developed for experiments on cyanobacteria, the simulator currently has the lower limit of operational temperature set at 25°C, which is slightly higher than the environmental temperature at which A. marina sp. str. Moss Beach was isolated, which was around 19°C. To understand whether this difference could have a non-negligible impact on the growth and performances of cells during the simulation, we followed their growth at three different temperatures: 20°C, reproducing the natural environment, 25°C, the target temperature, and 30°C, the temperature at which many cyanobacteria strains are maintained and at which we performed planetary simulations on other strains in the past (Battistuzzi et al. 2023a, 2023c), Liistro et al. 2024).

Concurrently, we evaluated the growth in two different saline media, Marine BG11 (MBG11) and artificial seawater (ASNIII) in combination with the three tested temperatures (Fig. 1). Based on the maximum growth rate registered during the experiment, the only temperature that significantly affected cell growth was 30°C, in both MBG11 and ASNIII. Even though final values of OD₇₅₀ were higher in cells grown at 20°C than at 25°C, growth rates registered in the exponential phase did not differ significantly. With regard to the media, ASNIII supported a faster initial growth, although maximum growth rates at the same temperature were not significantly different from rates in MBG11. Nonetheless, after 7 days for cells at 25°C, and 12 days for cells at 20°C, a cessation in the growth was encountered, suggesting that scarcity of essential nutrients in the ASNIII medium was reached in a shorter time than in MBG11, where OD₇₅₀ kept gradually increasing.

Figure 1.

Growth of Acaryochloris marina sp. str. Moss Beach in different media and at different temperatures, in terms of values of OD₇₅₀ and of growth rate (µ, d⁻¹). Cells were grown in Marine BG11 (MBG11, blue shades) and in artificial seawater (ASNIII, red shades). For each medium, three temperatures were tested: 20°C (round), 25°C (square), and 30°C (triangle). OD₇₅₀ was registered every 2–3 days (left), and, from OD₇₅₀ values, the maximum growth rate (µ) for each condition was calculated (right). Data are presented as the mean of four biological replicates, for values of µ one-way ANOVA was employed, ***: P-val <.0005, **: P-val <.005, *: P-val <.05. ASNIII: artificial seawater medium; MBG11: Marine BG11 medium.

To assess whether the composition and organization of the photosynthetic apparatus was affected by the tested condition, the diagnostic trait of Chl a and d content was analyzed. No significant differences were registered in any of the conditions in terms of µg of pigment per OD₇₅₀ (Fig. 2), indicating cells maintain stable content of both Chl a and d. A slight difference occurred in cells grown at 30°C, which showed a higher percentage of Chl d both in ASNIII (98.23%) and in MBG11 (97.15%) than what registered in cells grown at 20°C and 25°C (around 94.50%).

Figure 2.

Pigment composition in Acaryochloris marina sp. str. Moss Beach grown in different media, ASNIII (red shades) and MBG11 (blue shades) and at different temperatures, 20°C (lighter shades), 25°C (normal shades), 30°C (darker shades). Chlorophyll a and d content are expressed as µg of pigment per OD₇₅₀, percentage of chlorophyll d on total chlorophyll content (rhombus); statistical significance was assessed with one-way ANOVA on at least three biological replicates. ASNIII: artificial seawater medium; MBG11: Marine BG11 medium; Chl a: chlorophyll a; Chl d: chlorophyll d.

Given the results of this preliminary screening, for further experiments cells were maintained in MBG11, which provides the appropriate amount of nutrients, at our target temperature of 25°C, since neither the growth nor the photosynthetic apparatus seem to be significantly affected.

With these parameters, experiments to identify the optimal light intensity for the acclimation to the M7 starlight were carried out. Indeed, prior to the simulation of planetary conditions, a step of acclimation to the stellar spectrum is needed, to understand whether a light with different spectral qualities triggers any change in the photosynthetic apparatus ascribable to any acclimation strategy. Acclimations may require time—days or weeks—to be implemented and detectable, and an evaluation of the optimal light intensity for the acclimation period is needed. In this case, the aim was to find a non-saturating light intensity to allow the identification of differences in growth capabilities and/or acclimation mechanisms when organisms are exposed to planetary simulations. Adding factors like a different spectrum and the anoxic atmosphere could make a saturating light become a stress factor and cells could incur photoinhibition.

To assess the most suitable light intensity, cultures at a starting OD₇₅₀ of 0.2 were grown in a wide range of intensities, namely 10, 20, 40, 60, 80, 100, and 150 µmol of photons m⁻²s⁻¹, maintaining the temperature at 25°C. Growth was followed by monitoring OD₇₅₀, from which the maximum growth rate µ was calculated (Fig. 3). The growth rate increased with increasing light intensities, up to 60 µmol of photons m⁻²s⁻¹, where final OD₇₅₀ reached the highest value of all conditions. With higher intensities, namely 80, 100, and 150 µmol of photons m⁻²s⁻¹, growth was instead reduced, reaching low final OD₇₅₀ values even though the maximum growth rates, registered in the first 2 days, was not significantly different from those exhibited at 10 and 20 µmol of photons m⁻²s⁻¹. This growth rate was however not sustained over time in cells exposed to 80, 100, and 150 µmol of photons m⁻²s⁻¹.

Figure 3.

Growth of Acaryochloris marina sp. str. Moss Beach at different light intensities, in terms of values of OD₇₅₀, and of growth rate (µ, d⁻¹). Cells were grown at 10 (black, round), 20 (red, square), 40 (orange, triangle), 60 (green, inverted triangle), 80 (cyan, rhombus), 100 (blue, square), and 150 (pink, triangle) µmol of photons m⁻²s⁻¹. OD₇₅₀ was registered every 2–3 days (left), and, from OD₇₅₀ values, the maximum growth rate (µ) for each condition was calculated (right). Data are presented as the mean of at least four biological replicates, for values of µ one-way ANOVA was carried out, ***: P-val <.0005, *: P-val <.05.

As for the previous experiment, diagnostic evaluation of the photosynthetic apparatus and pigment content was carried out on cells grown at the tested light intensities. In vivo absorption and low temperature fluorescence suggested no intensity had a major impact on cells—no marked differences were registered for any of the conditions (Fig. 4a, b). The highest mean values of Chl a and d content were registered at 10 and 20 µmol of photons m⁻²s⁻¹, a progressive decline of mean chlorophyll concentrations was observed at stronger intensities, starting from 40 µmol of photons m⁻²s⁻¹ and becoming statistically significant at 80, 100, and 150 µmol of photons m⁻²s⁻¹. Nonetheless, the percentage of Chl d of the total chlorophyll content remained stable, around 94%, under all light intensities, except for cells exposed to 150 µmol photons m⁻²s⁻¹, which showed a significant decrease (91.66%) (Fig. 4c). Relative carotenoid content was evaluated by integrating the area of peaks in HPLC chromatograms (Figs 4d and S2). In cells exposed to 10, 20, 40, and 60 µmol of photons m⁻²s⁻¹ the percentage of carotenoids on the total pigment extract was in the range 27%–35%, while in cultures exposed to 80, 100, and 150 µmol of photons m⁻²s⁻¹, the carotenoid percentage increased up to, respectively, 57.40%, 62.59%, and 76.02%.

Figure 4.

Pigment composition and photosynthetic apparatus organization in Acaryochloris marina sp. str. Moss Beach grown at different light intensities, 10 (black), 20 (red), 40 (orange), 60 (green), 80 (cyan), 100 (blue), and 150 (pink) µmol of photons m⁻²s⁻¹. (a) in vivo absorbance spectra of cells normalized at 710 nm, (b) low temperature (77 K) fluorescence emission spectra of cells, normalized at the maximum peak (748 nm), (c) chlorophyll a (full pattern) and d (stripe pattern) content expressed as µg of pigment per OD₇₅₀ and chlorophyll d percentage on total chlorophyll content (rhombus); statistical significance was checked with one-way ANOVA on at least four biological replicates, (d) percentage of total chlorophylls (full pattern) and total carotenoids (stripe pattern) in cells extracts. Chl a: chlorophyll a; Chl d: chlorophyll d.

Even though 40 and 60 µmol of photons m⁻²s⁻¹ produce the highest growth rates, these intensities are nearly saturating as shown by the initial decline of mean chlorophyll content, which is exacerbated with increasing intensities. Further experiments were hence performed at 20 µmol of photons m⁻²s⁻¹.

Planetary simulation: M dwarf starlight and primeval atmosphere

Different spectral qualities of light can trigger, in cyanobacteria, acclimations involving a remodeling of the photosynthetic apparatus (Sanfilippo et al. 2019). In order to assess whether this was the case for A. marina sp. str. Moss Beach under the simulated M dwarf spectrum, cells were transferred from white light to the simulated M dwarf starlight (M7), and, for comparison, to a simulated Solar starlight (SOL) and to monochromatic FR light (FR). Growth was followed for 21 days, a timing that canonically allows for light acclimations to be completely achieved (Ho et al. 2017, Liistro et al. 2026b). An initial lag phase of 6 days in growth was registered in all light conditions, after which cells started to grow. Values of OD₇₅₀ revealed a noticeable growth in both FR and M7 light, with a maximum rate respectively of 0.19 d⁻¹ and 0.18 d⁻¹. On the other hand, cells exposed to SOL light had a lower final OD₇₅₀ as well as a lower maximum growth rate (0.15 d⁻¹) (Fig. 5).

Figure 5.

Growth of A. marina sp. str. Moss Beach during acclimation to the simulated stellar spectra. Growth of cells exposed to solar (SOL, black), far-red (FR, red) and M dwarf (M7, blue) light was monitored with measures of OD₇₅₀ every 2–4 days (left), from these values, the maximum growth rate (µ) during the 21 days was measured (right). Data are presented as the mean of three biological replicates. SOL: simulated solar light; FR: far-red light; M7: simulated M dwarf starlight.

During acclimation to the different spectra, in vivo absorbance of cells was registered periodically, with no significant change of absorption spectra: after 21 days, there was no alteration from cell absorption spectrum at day zero, in any of the conditions (Fig. 6a). Comparably, chlorophyll content as well as the percentage of Chl d on the total chlorophyll content remained stable in all the tested conditions (Fig. 6b). Low temperature fluorescence coherently indicated that no modification involving the emission ranges of the photosynthetic apparatus were activated under M dwarf light, nor in any of the other light regimes: emission spectra were analogous, with the main peak being at 748 nm in all of the conditions (Fig. 6c). Finally, cells were imaged with confocal microscopy to evaluate eventual changes in morphology or in the localization of chlorophylls (Fig. 6d). Micrographs revealed similar shape and dimension of cells acclimated to all tested light, as well as comparable spatial distribution of Chl a and d autofluorescence.

Figure 6.

A. marina sp. str. Moss Beach features after acclimation to SOL, M7, FR light. (a) in vivo absorption spectra of cell before (day zero, d0) and after acclimation; (b) content of chlorophyll a (full pattern) and d (stripe pattern) in terms of µg of pigment per OD₇₅₀, percentage of chlorophyll d in cell extracts; statistical significance was assessed with one-way ANOVA on three biological replicates; (c) low temperature (77 K) fluorescence emission spectra of cells; (d) representative confocal images of respectively SOL-, FR- and M7-acclimated cells, red indicates the autofluorescence of chlorophyll a, cyan indicates the autofluorescence of chlorophyll d, scale bar corresponds to 2.5 µm. d0: day zero; SOL: simulated solar light; FR: far-red light; M7: simulated M dwarf starlight; Chl a: chlorophyll a; Chl d: chlorophyll d.

A. marina sp. str. Moss Beach cultures acclimated to SOL, FR, and M7 light were used to evaluate the response and performances in simulated planetary conditions comprehensive of a modified atmosphere. Liquid cultures of OD₇₅₀ 0.55 were placed in the atmosphere-simulating chambers, ASCs, supplied with an initial anoxic atmosphere, under the respective light source (SOL, FR, and M7) for 72 hours.

After 72 h of exposure, strong growth was registered in all conditions (Fig. 7a), indicating that the initial anoxic atmosphere was not preventing cell survival and reproduction. Interestingly, M7 light promoted the strongest growth, with cells in SOL and FR having a lower final OD₇₅₀.

Figure 7.

Growth and gas consumption of Acaryochloris marina sp. str. Moss Beach inside the ASCs under SOL, FR, and M7 light. (a) initial (0 h) and final (72 h) OD₇₅₀ values, (b) variation of CO₂ concentration, (c) variation of O₂ concentration, (d) CO₂ production rate, (e) O₂ production rate (negative values indicate consumption). (a) shows individual values (dots) and average values (horizontal line inside the box plot), *: P-val: ≤.05. For (b) and (c) T₀ was set at 5 h to exclude the initial period of gas equilibration inside the ASCs. For (d) and (e) the last 24 h of exposure were considered, to exclude the initial acclimation of organisms to the anoxic/microoxic environment. SOL: simulated solar light; FR: far-red light; M7: simulated M dwarf starlight; CO₂: carbon dioxide; O₂: oxygen.

Cells in M7 showed a higher CO₂ consumption and O₂ release, with final values of ΔCO₂ and ΔO₂ being respectively the lowest and the highest registered with respect to SOL and FR (Fig. 7b, c). As highlighted in Fig. 7d, e, under M7 and FR light the production rates of

O₂ and CO₂ respectively increase and decrease faster than in SOL light, where indeed the lowest final mean OD₇₅₀ value was encountered.

To assess the surface biosignature that an organism with the features of A. marina sp. str. Moss Beach could generate, the reflectance spectrum of cells was recorded, presented in Fig. 8 along with the reflectance spectra of two other cyanobacterial strains previously tested under the starlight simulator, Synechocystis sp. PCC6803 and Chlorogleopsis fritschii PCC6912 (Battistuzzi et al. 2023a, 2023c). Reflectance spectra were recorded on cultures acclimated to M7 light, in vivo absorption of S. sp. PCC6803 and C. fritschii PCC6912 acclimated to SOL and M7 light are available in Fig. S3.

Figure 8.

In vivo reflectance spectra of liquid cultures of A. marina sp. str. Moss Beach (black), Synechocystis sp. PCC6803 (green) and Chlorogleopsis fritschii PCC6912 (red) acclimated to the M7 light recorded with a Flame spectrometer (OceanOptics). Moss Beach: A. marina sp. str. Moss Beach; 6803: Synechococcus sp. PCC6803; 6912: Chlorogleopsis fritschii PCC6912; Chl a: chlorophyll a; Chl d: chlorophyll d.

The reflectance spectrum of oxygenic phototrophs can be quite diverse in the 400–650 nm region, because pigment types and concentration can vary severely. This was perceivable in the spectra of S. sp. PCC6803, C. fritschii PCC6912, and A. marina sp. str. Moss Beach. A decrease of reflectance due to Chl a (at 440 nm and 680 nm) and phycocyanin (630 nm) absorption is found in S. sp. PCC6803 and C. fritschii PCC6912 with different intensities, ascribable to a different concentration of these pigments in the two species. On the other hand, this pattern is not found in A. marina sp. str. Moss Beach, that lacks phycocyanin and has minor content of Chl a with respect to the predominant Chl d, resulting in reflectance depressions at different wavelengths (460 nm and 700 nm). Despite differences in this region, reflectance of oxygenic phototrophs is typically characterized by a steep increase in reflectance between 680 nm and 750 nm, as noticeable from the spectrum of S. sp. PCC6803. This rise originates from the contrast of the high absorption capacity of Chl a in the red and the scattering of infrared light due to the change in the refractive index between cellular membranes and the medium. This pattern, called the Vegetation Red-Edge, VRE, is detectable in the reflectance spectra of Earth surface areas covered by oxygenic photosynthetic organisms. This feature resulted altered in C. fritschii PCC6912 cells acclimated to M7 light, where a shift in reflectance was present above 700 nm, likely ascribable to the presence of small amounts of chlorophyll d and f. In vivo absorption of cultures used to obtain the reflectance spectra highlight indeed a shoulder of absorption beyond 700 nm in C. fritschii PCC6912, indicating the occurrence of the FaRLiP response in this strain under M7 irradiation (Fig. S3, Battistuzzi et al. 2023a). In A. marina sp. str. Moss Beach, due to the abundance of Chl d, the steep rise in reflectance was instead completely shifted toward longer wavelengths, between 710 nm and 780 nm, in what could be referred to as a Chl d-near-infra-red edge, or Chl d-NIR edge.

Discussion

Some of the planned space missions from NASA and ESA are aimed at characterizing the atmospheres of exoplanets, also in the search for biosignatures. In particular, HWO (NASA) will be able to detect the presence of O₂, that astrobiologists indicate as one of the strongest biosignatures to search for. As on Earth the accumulation of O₂ in the atmosphere is due to oxygenic photosynthesis, operated by cyanobacteria, algae and plants, for years the community has been debating on the feasibility and efficiency of this metabolism on exoplanets. In this work we focused on this possibility to occur around M dwarf stars, as they are considered good targets for their long lifespan and the abundancy of rocky exoplanets identified as habitable harbored in their stellar systems (Gale and Wandel 2016, Shields et al. 2016). The challenge for oxygenic photosynthesis to have evolved with an M dwarf spectrum resides in the long wavelength limit of this metabolism. On Earth, oxygenic photosynthetic organisms rely on visible light, and can only marginally harvest FR light, while M dwarfs emission spectra offer very low visible and high FR and IR irradiance (Chabrier et al. 1996).

Laboratory simulations have however demonstrated how an M dwarf spectrum can support strong growth in cyanobacteria and microalgae, and moderate growth in non-vascular plants that evolved to harvest visible light ((Battistuzzi et al. 2023a, 2023b, 2023c, Liistro et al. 2024, Boccia et al. 2026). Nonetheless, this does not imply that analogous organisms should be expected to have evolved under such spectrum: a red-shifted irradiance might have selected “red-shifted adaptations” as for instance modified antennae (Chitnavis et al. 2024) or two color reaction centers absorbing respectively visible and near-infra-red light. (Takizawa et al. 2017).

Considering the red-shifted adaptations of oxygenic photosynthesis that have evolved on Earth, in this work we tested the Moss Beach strain of Acaryochloris marina, a unique species with Chl d-bearing reaction centers and permanent red-shifted absorption, to assess whether its photosynthetic apparatus is a good fit under M dwarf-like irradiance, evaluating both time courses of OD₇₅₀ and O₂ production rates.

Chl d spectral type

While Kiang et al. (2022) observed A. marina sp. str. Moss Beach to have a spectral peak at 705 nm, our culture from a sample directly obtained from their lab exhibited a spectral peak at 710 nm. Shifts in the spectral peak in the same A. marina strain has been observed also for other strains (Min Chen, personal communication), which is likely due to photoacclimation to different growth conditions, the mechanisms for which are still being elucidated. Due to the divergence of their light harvesting systems and chlorophyll environment, A. marina strains exhibit a high variation in the fluorescence properties of Chl d and in the spectral features. Ulrich and colleagues (2024) identified three main spectral types depending on fluorescence emission features, which reflect differences in Chl energy levels and in chl-binding proteins. From the intermediate-wavelength spectral type, with emission maximum at 727–731 nm, diverged both the short-wavelength and the long-wavelength spectral types, with emission maxima respectively at 721–724 nm and 738–748 nm. Our analysis of low temperature fluorescence identified the spectral type of the strain Moss Beach, that, presenting a fluoresce peak at 748 nm, should be included in the longer wavelength group, to which also its close relative A. marina S15 belongs (Kiang et al. 2022, Ulrich et al. 2024). This locates A. marina sp. str. Moss Beach in the group of oxygenic photosynthetic organisms with the strongest red-shifted constitutive emission among the types differentiated by Ulrich et al. (2024), making it an excellent target to test Earth “red-shifted adaptions” in M dwarf simulated light. Very recently, an even further red-shifted strain was discovered, with the emission maximum at ∼760 nm (Oliver et al. 2026).

Absence of chromatic acclimation

Results of the preliminary acclimation to the simulated stellar spectra, M7, reproducing an M dwarf star, highlight how, differently from other species previously tested, A. marina sp. str. Moss Beach does not need to modulate the pigment content nor its photosynthetic apparatus in order to efficiently grow with the M7 light. When tested in the same set up, cyanobacteria like Chlorogleopsis fritschii sp. PCC6912 and Synechococcus sp. PCC7335, underwent severe remodeling of the photosynthetic apparatus, acclimating to harvest the abundant red and FR light available. C. fritschii PCC6912 exhibited the FaRLiP response (Battistuzzi et al. 2023a), maximizing FR light absorption, while S. PCC7335 strongly modulated the content of accessory pigments via chromatic acclimation type 3, maximizing red light absorption (Liistro et al. 2024). The unaltered absorption features and pigment content observed for A. marina sp. str. Moss Beach underlined, instead, how this strain is already well suited to photosynthesize under the simulated M7 stellar spectrum, growing even better than when exposed to the simulated solar light, SOL.

Growth rate and O₂ evolution under M dwarf simulated light

Interestingly, during acclimation, growth rates were higher than that exhibited in previous experiments conducted in white light: SOL, FR, and M7 lights all provide a portion of FR photons, in different amounts, which appears to be crucial for this strain to drive faster growth. Indeed, at the end of the acclimation experiment performed in terrestrial atmosphere, the highest OD₇₅₀ were observed in M7 light and monochromatic FR light, while in SOL light, which provided the lowest flux of FR photons of all three tested lights, the final mean OD₇₅₀ was the lowest. Studies conducted on A. marina sp. str. MBIC11017, found similar results: for this strain as well, a stronger growth was registered when cells were exposed to FR light rather than visible light. This was paired with a stronger photosynthetic rates, indicating longer wavelength drive photosynthesis more efficiently than visible ones (Behrendt et al. 2012).

Considering all the conditions tested in this study, A. marina sp. str. Moss Beach showed a maximal growth rate reaching 0.19 d⁻¹, comparable to that registered in the reference strain of A. marina, MBIC11017 (Swingley et al. 2005).

With A. marina sp. str. Moss Beach acclimated to SOL, FR and M7 spectra, we then proceeded with planetary simulation, combining the light treatment with a simulated planetary primeval atmosphere non-oxygenated (95% N₂ and 5% CO₂). Cells grew consistently in all the light conditions, suggesting the initial absence of oxygen had no major detrimental effects. As early oxygenic cyanobacteria in the Archean would have had to function in largely anoxic atmospheric conditions (Lyons et al. 2014); therefore, our experiment demonstrates the viability of growth at low O₂. Most importantly, in the anoxic atmosphere, the M7 starlight promoted the highest growth with respect with SOL and FR light. Cells were indeed efficiently sequestrating CO₂ and producing O₂, having the strongest rates under the simulated starlight when compared with cells in SOL and FR light.

All the cyanobacteria species examinated in our planetary experiments efficiently sequestered CO₂ and produced O₂, however, the grater O₂ productivity in A. marina sp. str. Moss Beach is noticeable if compared with the strains previously tested in this combination of stellar light and atmosphere simulation. After 72 h of exposure to simulated planetary conditions (starlight spectrum coupled with primeval atmosphere), we registered the production of ∼400 µmol of O₂ in A. marina sp. str. Moss Beach. C. fritschii PCC6912 and S. sp. PCC6803 reached similar O₂ production after 48 h under the same spectrum and in the same atmosphere tested in this work (Battistuzzi et al. 2023c), which is interesting considering they are both fast-growing strains and they were exposed to a higher total irradiance (30 µmol of photons m⁻²s⁻¹). S. sp. PCC7335, a relatively slow-growing organism, took 7 days to produce the same amount under the same spectrum—however, it must be considered that the total irradiance of the experiment was set to 10 µmol of photons m⁻²s⁻¹, half the one used in this study (Liistro et al. 2024).

Implication for exoplanet biosignature detection

If photosynthetic metabolisms like the one selected in A. marina sp. str. Moss Beach have evolved outside our Solar System, our data support the potential to generate atmospheric biosignature via the release of O₂. Of course, the planetary oxygenation time and the diversification and spread of oxygenic phototrophs would definitively control on the potential accumulation of this gas, hence on the possibility of its detection.

Moreover, the reflectance spectrum of A. marina sp. str. Moss Beach cells underlines how, if such metabolism is present on exoplanets orbiting M dwarf stars, and if it could be utilized in potentially far-red vegetation, its fingerprint would be highly characteristic, with a VRE profoundly shifted toward longer wavelength, becoming a near-infra-red-edge, NIRE.

Overall, A. marina sp. str. Moss Beach is an excellent example of how evolution on Earth has led oxygenic photosynthesis to red-shifted adaptations that can efficiently photosynthesize with an M dwarf like spectrum, underlining the possibility of natural selection to push highly functional metabolisms forward their energetic limits, as in the case of A. marina species.

Conclusions

The A. marina species constitute fine evidence of how evolutionary pressure can lead oxygenic photosynthesis not only toward visible-harvesting forms but also toward constitutive red-shifted forms.

In this work, the Moss Beach strain of A. marina was tested for the first time under a simulated M dwarf light, evaluating also its capacity to produce O₂ in an anoxic atmosphere enriched in CO₂.

The results of this work highlight how the photosynthetic apparatus of A. marina sp. str. Moss Beach represents a very good, constitutively red-shifted, fit for oxygenic photosynthesis under M dwarf stars, leading to the release of considerable amounts of O₂ with a strong impact on the initial anoxic atmosphere. Surprisingly the O₂ production rate of this Chl d cyanobacterium was comparable to that of Chl a-only cyanobacteria that have been previously studied. If evolved on exoplanets orbiting around an M dwarf, this kind of metabolism could hence generate a gaseous biosignature. The data of O₂ release here presented are of crucial interest for exoplanet habitability models and represent a reference to interpret the data that will eventually result from future space missions like ARIEL (ESA) and HWO (NASA), that are going to investigate exoplanets’ atmospheres. Regarding the possibility of such organisms to generate a surface biosignature, the pigment content of A. marina sp. str. Moss Beach is unaltered under the solar or the M dwarf spectra, resulting in a characteristic reflectance spectrum with a Vegetation Red-Edge sensibly shifted toward longer wavelengths, in a NIR edge.

The long wavelength limit for oxygenic photosynthesis remains an open question. Trade-offs in the efficiency and resilience of far-red phototrophy with lower energy photons have been identified in comparisons of enzyme turnover and electron transfers in Chl a, FaRLiP, and A. marina Chl d species (Viola et al. 2022), but even algae have been found tuned to FR light (Wolf and Blankenship 2019), and higher plants demonstrate oxygen evolution out to 750 nm or longer (Zhen 2020). The Chl d-NIR-edge we have observed in A. marina sp. str. Moss Beach is likely only one NIR edge along a continuum, which future work should continue to constrain relative to controls at the molecular and environmental scales.

Author contributions

Elisabetta Liistro (Conceptualization, Formal analysis, Investigation, Methodology, Data curation, Visualization, Writing—original draft, Writing—review and editing), Beatrice Boccia (Formal analysis, Data curation, Writing—review and editing), Mary Niki Parenteau (Resources, Conceptualization, Writing—review and editing), Nancy Y. Kiang (Resources, Conceptualization, Writing—review and editing), and Nicoletta La Rocca (Resources, Conceptualization, Investigation, Data Curation, Supervision, Writing—original draft, Writing—review and editing)

Conflicts of interest

None declared.

Funding

The research was funded by the Italian Space Agency through the “ASTERIA” project (ASI N. 2023-5-U.0) and by the Department of Biology of the University of Padova (Italy) via intramural grants.