The considerable number of new worlds discovered so far is pushing scientists to provide evidence of life on other planets. The diversity in kind, composition, masses, and radii of these new worlds is so vast that almost all possible mass values are covered in continuity from Mars (
) up to super-Jupiters (>10
). Among all the planetary hosts, low mass stars, mainly M spectral type stars, are the main targets of the extrasolar planet surveys due to both their high density in the Galaxy and their small radii that provide higher amplitude transit signals than solar-like stars [1
]. Indeed, the most attractive characteristic of these systems is that 40% of M stars host super-Earths with a minimum mass between about 1 and 30 Earth masses, orbital periods shorter than 50 days, and radii between those of the Earth and Neptune (1–3.8 R
). Due to these high occurrence rates, super-Earths (1–10 M⊕) represent the most common type of components of planetary systems in the Galaxy [3
]. Even more striking, the frequency of super-Earths found in the habitable zone (HZ) of M dwarfs (with a period between 10 and 100 days) is about 50% [4
]. These results renew, with higher and more interdisciplinary efforts, the search for life as an astrophysical problem. In this framework, it is critical to determine the types of biosignatures (based on the so-far-recognized life signatures) for when designing the next generation of ground- and space-based instruments that will observe these planets at both high spectral and spatial resolutions (e.g., Reference [6
]). Among all the biosignatures, oxygen seems to be the most prominent signature that can reveal the existence of life on other planets [14
]. In this sense, it is our most robust and the most studied biosignature [15
]. Its presence, together with other gases, like CH
O, is the signal of a thermodynamic disequilibrium that, for a long time, has been considered as compelling evidence for life (e.g., Reference [16
]). Since the closure of of Seager & Bains (2015) [17
], many arguments countered this concept because using the thermodynamic disequilibrium as a biosignature cannot be easily generalized. Other authors have also argued that oxygen is not a suitable bioindicator due to photochemical reactions that have abiotic O
as a byproduct (e.g., H
O and CO
photoionization). A detailed discussion of several possible false positives is presented in Harman et al. (2015) [18
Today, in the the atmosphere of Earth, oxygen is highly abundant (21% by volume) due to the oxygenic photosynthesis of plants, algae, and cyanobacteria, the presence of which can be detected by remote observations, not only for the presence of O
in the atmosphere of the planet but also by discerning the red edge. The red edge is a feature associated with the high reflectance of photosynthetic organisms at near infra-red (NIR) in contrast with the absorption by chlorophyll in wavelengths shorter than about 700 nm. This phenomenon emerges due to the scattering of light at the interfaces between the cell walls and the air space inside the organism [19
In recent years, biologists have found species of cyanobacteria able to use far-red (FR) light for oxygenic photosynthesis due to the synthesis of chlorophylls d
], extending in vivo light absorption up to 750 nm, suggesting the possibility of exotic photosynthesis in planets around M stars. So far, a number of works have discussed the possibility of emerging oxygenic photosynthesis on a planet in the HZ of an M star (e.g., Reference [21
]) under favorable conditions.
Considering all these favorable observational and theoretical circumstances, it is important to assess, in an experimental way, the consequences of oxygenic photosynthesis on planets orbiting in the HZ of M stars. In this paper, we present the laboratory set up and experiments conducted with the aim of understanding the performance of photosynthetic organisms exposed to conditions similar to that of an Earth-like planet in the HZ of an M star. In particular, the laboratory set up simulates the exoplanetary surface temperature and radiations. In these experiments, we analyzed the growth and the photosynthetic efficiency of several cyanobacteria with different photosynthetic behaviors, based on chlorophyll fluorescence measurements. In particular, we selected a cyanobacterium unable to utilize FR light for carrying out oxygenic photosynthesis, two species able to exploit it and a species both able to use FR light and perform the so-called Chromatic Acclimation (CA), changing its color depending on the incident light spectrum.
This paper is organized as follows: In Section 2
, we describe the approach to the problem and the experimental plan. In Section 3
, the experimental set up and its validation is described; in Section 4
, we discuss the microorganisms we used and report the results of the experiment. Section 5
is allocated to the discussion and the conclusion.
3. Laboratory Set Up
The experimental set up is sketched in Figure 2
. It can be conceptually split into two parts: the stellar simulator on top, composed of the illuminator and a spectrometer, with the reaction cell on the bottom, where the microorganisms used for the experiments are hosted. The entire system is isolated in a dark container and is cooled by two fans, it is equipped with an anti-condensation system, and it is monitored through a webcam. A control PC allows to operate remotely the system without interfering with the experiment.
3.1. Star Irradiation Simulator
To achieve the described experimental aim, it was necessary to have an unconventional light source. In particular, the light sources used in photosynthetic study facilities are mainly metal halide, high-pressure sodium, fluorescent, and incandescent lamps. These lamps are commercially available and mostly emit solar or close to solar radiation, with a limited capability in adjusting the color temperature and the intensity of the output radiation. In our experiment, we need a light source able to reproduce the irradiation of stars other than the Sun in a quite simple and direct way. Furthermore, it should be operable without interfering with the experiment. To achieve our goal, we designed a completely different light source using light-emitting diodes (LEDs) controlled remotely by a computer.
LEDs have also been used in the laboratory as light sources for their efficiency in plants growth [55
]. For that purpose, the LED-based devices are built to illuminate the material in the red part (600–700 nm) of the PAR, while, typically, the blue part of PAR is covered by blue fluorescent lamps.
For our purposes, we need a completely different device. In fact, for Wien’s law ( nm K), different stars of different spectral types have the maximum of their emission at different wavelengths. In particular, while the Sun has a peak of emission at about 550 nm, an M star, which is about a factor of 2 cooler than the Sun, has its emission peak in the NIR range (at about 1000 nm). In order to appreciate the differences in the slope of the spectra of the stars of different spectral types, we need a collection of LEDs able to cover a slightly longer wavelength range than the PAR, between about 350–1000 nm. Moreover, this device shall be able to modulate the LEDs intensity in order to mimic, as close as possible, the flux variation of stars of different spectral types.
The available LEDs allow us to consider the wavelength range between 365 nm and 940 nm covered by 25 dimmable channels (see Table 1
). Because LEDs covering such wavelength range are manufactured with different technologies (from AlGaN/InGaN to GaAs/InGaP), their emitted luminosities are also different from each other, and each channel has a different number of LEDs to achieve the required optical power at a specific wavelength. Furthermore, we added a white LED with a correlated color temperature (CCT) of 2200–2780 K to fill the spectrum in the 630 nm region. We used 312 LEDs in total, arranged in five concentric rings on which the mosaic of circuit boards is arranged in a pie-chart shape, on the surface of which the diodes have been welded [56
]. Each channel is tunable enough to allow us to reproduce the radiation of stars of F, G, K, and M spectral types.
The modularity design of the board permits easy maintenance in case of damage, allowing us to remove only the problematic piece. The disposition of diodes on the board was designed to reduce the non-uniformity of the flux, due to the intrinsic light exit angle of each led. Moreover, a reflective cylinder and an optical diffusive foil were mounted to increase the uniformity. Since the thermal power of the system dominates its radiation power, the diodes are cooled by a fan set on the back of the board. A spectrometer collects the light through a slit head placed at a manually adjustable distance from the diffusive foil. The adopted spectrometer is a Component Off The Shelf (COTS) component. We selected the FLAME VIS-NIR by Ocean Optics, in which its pixel detector covers the wavelength range 190–1100 nm.
The illuminator is controlled by a custom control software [57
] that, by means of a graphical user interface (GUI), allows the user to select an appropriate spectrum chosen from a spectral library. For the input spectrum, the control software calculates the intensities of the 25 channels to best fit the spectrum. In any case, through the GUI, the user has access to each channel of the illuminator setting the output flux of the channel. The set spectrum is shown in a window of the GUI. The emitted spectrum is registered by the spectrograph and is superimposed on the input spectrum. Slight differences between the two can be fixed by adjusting the luminosity of each channel. The left panel of Figure 3
shows the simulated spectrum of a solar star (light SOL; see Section 4.2
), while the right panel of the same figure shows the simulated M star spectrum (light M7; see Section 4.2
). In both panels the input spectrum is represented in red color, and the emitted spectrum in blue. The input spectra are smoothed (e.g., see Figure S1 in the Supplementary Materials
for an M7 V star) due to the difficulties in reproducing the high resolution stellar spectra by the spectrum simulator.
3.2. The Reaction Cell
The incubator cell (see Figure 4
) is a steel cylinder of 0.5 l of volume in which the light enters through a thermally resistant Borofloat glass, with over 90% transmission in 365–940 nm wavelength range. The atmosphere in the cell can be flushed to change the initial O
, and N
levels. The cell is provided with pipe fittings and connected to an array of flow meters and needle valves (each for a different input gas: N
) to inject atmospheres of controlled and arbitrary compositions. Once the desired mixture is flushed through the cell, the input and output valves are closed to seal the inside environment and leave it to its evolution. Water vapor will quickly reach saturation value, due to the water-based medium in the sample Petri dish. When using high carbon dioxide levels, caution should be exercised to ensure a long enough flushing time to achieve equilibrium between the gas phase and dissolved CO
due to its high solubility. The base of the cell and the sample are kept at a constant temperature by controlling a Peltier cell on which the cylinder is leaned. The Peltier temperature set point is always kept
C lower than the surrounding environment (which its temperature is also controlled) to avoid any condensation on the upper glass window.
In the context of this work, the cell was always operating at ambient pressure and 30
C temperature with an initial composition of 75% vol. N
, 20% vol. O
, and 5% vol. CO
; this provides a high enough amount of carbon dioxide to be fixed into biomass throughout the experiment without excessively stressing the sample. Vital photosynthetic microorganisms in a liquid medium inside the cell are expected to produce oxygen. Hence, the cell is provided with a commercial fluorescence quenching oxygen sensor (Nomasense O
P300), while the CO
concentration is monitored via a custom Wavelength Modulation Spectroscopy (WMS) Tunable Diode Laser Absorption Spectroscopy (TDLAS, [58
]) set up. To monitor the gas, four wedged windows of 2.5 cm are pierced and paired two by two in opposite positions on the wall of the cell. Two of the windows are used by the CO
sensor for TDLAS, whereas one is used for the fluorescence quenching tablet, which is remotely sensed through an optical fiber. The reaction cell underwent several modifications for reducing the systematic errors and the human interferences during the experiments. In Battistuzzi et al. (2020) [59
], a complete description of the very last version of the reaction cell and the results obtained from a biological point of view is presented.
3.3. The Control Software
The starlight simulator (illuminator and spectrometer) and the incubator cell environment (gas sensors and Peltier cell) are controlled by two separate processes, running on the same computer. We wanted a stand-alone software to control the simulator because it could also be used also for other laboratory applications (e.g., photo-bioreactors, microscopy, yeast growth [60
3.4. Validation of the Experimental Set up
To validate the experimental set up, we positioned the cyanobacterium Synechocystis
sp. Pasteur Culture Collection (PCC) 6803 liquid cultures into the simulator chamber with an atmosphere consisting of a mixture of gasses in the following composition: 75% of N
, 20% of O
, and 5% of CO
. Eventually, we irradiated it by means of the star simulator with a solar (G2 V) spectrum with three different intensities: 30, 45, and 95
mole is used for
mole of photons). The organisms exposed to different light regimes grew with good photosynthetic efficiency. Description of the test and of the developed method to evaluate the growth of bacteria without any interferences by the operators are fully detailed in Battistuzzi et al. (2020) [59
5. Discussion and Conclusions
M stars are very popular in the astrobiology community due to their ubiquitous presence in the Galaxy and their small radii, which provide higher amplitude transit signals than solar-like stars. So far, they are recognized to be the most frequent hosts of super-Earths discovered orbiting in the HZ of a star. This sparked off a great theoretical debate about the possibility of having life, particularly photosynthetic life, on these planets. Several efforts have been spent aiming at modeling the upper wavelength limit of putative photoautotrophs on exoplanets. It has been hypothesized that oxygenic photosynthetic organisms could have developed pigments that do not utilize PAR light, but the more abundant NIR light, or employ photosystems using up to 3 or 4 photons per carbon fixed (instead of 2), as well as utilize more photosystems in series (3, 4, or even 6), allowing them to exploit photons of wavelengths up to 2100 nm [21
]. Moreover, the prospects for photosynthesis on habitable exomoons via reflected light from the giant planets that they orbit have been theoretically evaluated, suggesting that such photosynthetic biospheres are potentially sustainable on these moons except those around late-type M-dwarfs [78
]. However, up to now, no experimental data (survival, growth, and photosynthetic activity) about the behavior of oxygenic photosynthetic organisms exposed to simulated environmental conditions of exoplanets orbiting the HZ of M dwarfs, in particular, exposed to an M-dwarf light spectrum, have been produced. Numerous investigations (see Reference [67
], for a review) have been done, instead, in the field of the oxygenic photosynthesis “beyond 700 nm”, especially after the discoveries of cyanobacteria able to synthesize chl d
]. However, they were committed to understanding the molecular and biochemical mechanisms behind how the photosynthetic functioning of the photosynthetic apparatus under a FR light spectrum, rather than testing it under exotic light spectra [70
]. Here, to the best of the authors’ knowledge, for the first time, we present the experimental data obtained directly through exposing photosynthetic organisms to a simulated M dwarf spectrum. We compared the results to responses of those species under solar and FR simulated lights, using innovative laboratory instrumentation. As expected, in FR light, only the cyanobacteria able to synthesize chlorophyll d
could grow. Surprisingly, all strains, both able or unable to use FR light, grew and photosynthesized under the M dwarf generated spectrum in a similar way to the solar light and much more efficiently than under the FR one.
In particular, we compared the responses of strains able to have FarLiP and of the control microorganism PCC 6803 that does not. The growth estimated by F incremental ratio parameter obtained for all the cyanobacteria in our study shows a value that is very similar or equal, considering the error bars, to the value of F measured for those spots irradiated with the solar light. In the case of the irradiation with monochromatic light (far-red (FR)), only PCC 6803 is unable to acclimate itself to the FR light, while all the others show a normal photosynthetic efficiency under this light, as well. This suggests that PCC6803 grows very well under simulated M7 light by only using the visible part of the spectrum. The ability of the other organisms to exploit FR light does not seem be beneficial for growth under M7 simulated light. Furthermore, all the tested strains, except PCC 6803, have very similar values of the F/F under any kind of irradiation spectra. This highlights that they are able to acclimate to all the used lights. Our findings emphasize the importance of simulating both the visible and FR light components of an M dwarf spectrum to correctly evaluate the photosynthetic performances of oxygenic organisms exposed to such an exotic light condition.
Moreover, in a previous work [59
], we demonstrated that, with our experimental set up, we can measure the consumption of CO
and the production of O
of the PCC 6803 cyanobacterium under solar irradiation. This serves as a prelude to the future analysis of the cyanobacteria photosynthetic gas exchanges in real time during their growth under M star spectra irradiation.
Last but not least, we realized an experimental set up that allows us to reproduce, in the laboratory, an alien environment with the possibility to variate the thermal and physical conditions. In this way, we are carrying out experiments on photosynthetic organisms to verify their capacity of thriving and acclimating to extraterrestrial conditions. We developed new and original laboratory devices (e.g., the star irradiation simulator) and novel measurement methods (see Reference [59
]) that will allow new experiments in the future. To prepare for the next step of our research plan, we have already produced several models of stable super-Earth atmospheres to be used in the laboratory. We have started to monitor the evolution of oxygen and the fixation of carbon dioxide in the cyanobacteria exposed to very different irradiations and simulated atmospheres.
Hence, if the evolutionary tracks on a habitable planet in the HZ of an M star are quite similar to those on Earth, photosynthetic microorganisms could, as well, produce O and fix CO in organic matter on these planets orbiting such cold stars.
Will it be possible to observe the released oxygen in a remote way? The answer to this question is not simple because it depends not only on the efficiency in producing oxygen by photosynthetic organisms but also on the efficiency of the possible oxygen sinks that are at work on that planet. The reverse reaction to oxidize photosynthetic products depletes the atmospheric oxygen. The net release of oxygen in the atmosphere, due to this balance, is regulated by the sink of organics in the sediments. If the level of O
is low in the atmosphere, the reactions with reducing gases from vulcanism (H
S) and submarine weathering [86
] can deplete O
. If the O
production rate is greater than the depletion rate, its build-up in the atmosphere is possible [88
], and the Fe
oxidation process becomes an important one. Catling and Kasting (2017) and Kaltenegger et al. (2010) (Reference [86
], respectively, and References therein) discussed deeper on the build-up of oxygen in the atmosphere of a planet. Oxygen depletion is a time-dependent process. The atmospheric oxygen is recycled through respiration and photosynthesis in less than 10,000 years. In the case of total extinction of the biosphere of Earth, the atmospheric O
would disappear in a few million years [87
Thus, we conclude that only the observations can give us the right answer. So far, brand new ground- and space-based instruments are planned to be operative with the aim of finding and characterizing extrasolar planets. In the next future, dedicated space missions and space telescopes, like James Webb Space Telescope (JWST) and Origin Space Telescope (OST), and huge ground telescopes will be the right tools to search for life in other worlds.