香京julia种子在线播放

The studied gypsum was sampled from the Sidi Boutbal quarry (SB), (35°42′36.5″N 0°21′57.4″W), located in the Oran district, Algeria. This quarry is situated at the eastern extremity of the Lower Chelif basin (Mediterranean Basin) (Figure 1A). The Chelif Basin is one of the largest Messinian peripheral sub-basins, characterized by an elongated and ENE–WSW oriented structure spanning over 260 km in length and 35 km in width. It is situated between the sabkha of Oran to the west and El Asnam city to the east (Rouchy et al., 2007).

The SB quarry is predominantly composed of an Upper Miocene to Plio-Quaternary succession, exhibiting is an anticlinal structure that exposes a 60 m thick evaporitic succession at its core. This succession consists of bottom grown gypsum beds comprising of twinned selenite crystals. Four gypsum beds (Gb1, Gb2, Gb3, and Gb4), interbedded with thin clay and sandstone layers are recognized, as illustrated in Figures 1C–E. However, only the Gb2, Gb3, and Gb4 are visibly exposed in the outcrop. About 30 kg of gypsum, containing preserved filaments, were collected from the 15 m thick third gypsum bed (Gb3) using a hammer and chisel (asterisk symbol in Figures 1C–E). The sampled gypsum rock was subsequently packed in plastic bags for further preparation and to minimize contaminations. The gypsum rock exhibited a light gray to brown coloration, manifesting as either small crystal selenite (0.5 – 2 cm) (SC, Figure 2A) or 10 cm-sized twinned arrow-head selenite crystals (TS, Figure 2A). It displayed an alternation of millimeter-thick turbid and limpid laminae, visible by eye. The limpid laminae exhibited rare to absent visible solid inclusions, while the turbid laminae contained a concentrated network of up to millimeter-long brownish filaments within the re-entrant angle of the twinned selenite crystals (Figure 2B).

During sample preparation, measures were implemented to prevent contamination from the rock’s surface and external sources. A hand-sized gypsum rock was meticulously cleaved to unveil a pristine, uncontaminated surface. A 1 mm in thick segment of gypsum was meticulously extracted from this surface using a sterilized spatula. No additional processing was conducted. An initial examination using optical microscopy was performed on several thin sections to select the most suitable one for the study presented here. A small fraction of the representative thin section (Figure 2B) was subsequently prepared for analysis with SEM and EDX (Department of Chemistry, Biochemistry and Pharmaceutical Sciences, University of Bern, Switzerland) and LIMS by fixing the thin section on a LIMS-specific sample holder using conductive copper tape (Figure 2C). Finally, the sample was gold sputter coated (∼10 nm gold layer) to enhance the sample surface for both SEM and LIMS analyses (Figure 2D; Gruchola et al., 2022). Through the use of optical microscopy, SEM and EDX measurements, one well-preserved filament was identified and selected for the subsequent, detailed LIMS chemical analysis.

A detailed description of the space-prototype LIMS instrument used in this study can be found in previous publications, such as (Grimaudo et al., 2020) and references therein. Only a brief overview of the mass spectrometric system and its principle of operation are given in the following.

The LIMS system consists of a miniature reflectron-type time-of-flight (R-TOF) mass spectrometer to which a laser ablation ion source is coupled. The latter is a pulsed femtosecond laser system, with a pulse repetition rate of 1 kHz, a pulse width of τ ∼ 190 fs, and emits a fundamental laser beam at wavelength of λ = 775 nm. This beam is then converted to 258 nm using a harmonic generator unit. The 258 nm output is further split into two laser beams of similar pulse energy, delayed by about 40 ps, and recombined collinearly in a double-pulse (DP) unit (Tulej et al., 2018; Riedo et al., 2021). Two laser beam attenuators allow the precise adjustment of both pulse energies before recombination. After the DP unit, a beam guiding system guides the laser beam through a beam expander, through the entrance window of the vacuum chamber and through the mass analyzer towards the sample surface. A 3D translation stage allows accurate positioning of the sample surface relative to the entrance electrode of the mass analyzer (Figure 3).

Each laser pulse ablates and ionizes a distinct layer from the sample material. The ion-optical system of the mass analyzer collects the positively charged ions and guides them towards the detection unit (MCP detector system). Note, in this study the signal was only collected from one of the four anodes (Riedo et al., 2017). Species arrive sequentially at the detector system according to their mass to charge ratios (m/q). The TOF signal is recorded with a sampling speed of 3.2 GS/s using a high-speed ADC data acquisition system. For the automation and control of the entire instrument (laser system, data acquisition, sample stage, and pulse delay stage), in-house written software is used.

In all measurement campaigns, the applied voltage difference over the MCPs in the detector unit was set to 1700 V. LIMS measurements on the gypsum sample were conducted using pulse energies of 3.0 μJ and 3.2 μJ with a pulse energy stability of about ±0.1 μJ for the first and second pulse, respectively. These values refer to pulse energies measured at the laser focus on the sample surface. LIMS measurements covering both gypsum and the selected fossil filament regions (Figure 4) were conducted in a grid of 17 × 25 spots (Figures 5A,C), (the data used for the sample analysis covered an area of 11 × 12 craters (Figure 5D)).

In this measurement, 800 laser bursts were applied to each sampled location. Each burst consisted of 50 laser shots, for a total of 40,000 laser shots per sample. For each burst, the mass spectra of the individual laser shots were histogrammed into a single spectrum. This allows the analysis of the chemical composition of the ablated layers with a resolution at the μm level and lower. Subsequent to data collection, 2D atomic fraction maps were produced for the analyzed area. Depth profile analyses were conducted for the individual locations to understand the spatial heterogeneity of the sample at the μm spatial scale. Through element intensity correlations, the mineralogical composition of the sample can be derived. Atomic Force Microscopy (AFM) measurements were performed to correlate between the number of laser shots and crater depth, allowing for an accurate inclusion size measurement. The description and the results of the AFM measurements are summarized in the supplementary material section.

Optical microscopy analysis of the SB Gb3 sample under transmitted plane-polarized light showed that turbid laminae consist of a densely interwoven network (Figure 4A) of brownish, sinuous, and curved fossil filaments (Figure 4B) of various sizes (ranging from µm to mm length and 20–80 µm width) (Figure 4A, white triangles). Under reflected plane-polarized light, the fossil filaments appear to be partially hollow (arrows in Figure 4C) and exhibit different mineralogical compositions (Figures 4C,E,F). The fossil filaments are syngenetic and embedded within gypsum, as they are situated on the side opposite the focal plane of a secondary veinlet (arrows in Figures 4B,E,F). Close observation of the fossil filament under transmitted cross-polarized light reveals the presence of spherical to subspherical grains measuring 5–10 µm in size, typically with sharp edges, high relief, and high birefringence (Figures 4D,J). Additionally, within the filament, scattered opaque black grains of varying sizes are present (arrows in Figure 4K), while the area surrounding the fossil filament appears dark. Fluid inclusions, mainly rectangular in shape, were observed in all the examined thin sections. A close-up SEM analysis of a hollow fossil filament (∼80 µm width) revealed an inner layer (∼15 µm) composed of two main minerals (Figure 4G). The predominant mineral is formed by spherical to subspherical crystals, often hollow (∼5–10 µm), with EDX spectra indicating the presence of C, Mg, Ca, and traces of Si, corresponding to Mg-Ca carbonate minerals (Figure 4I, red spectrum). These crystals are intercalated by a flaky, wavy subhedral thin to thick layered aluminosilicate mineral (clay minerals) composed of C, O, Mg, Al, Si, Fe, and traces of K (EDX spectra, Figure 4I, blue spectrum). The inner layer of the filament is morphologically and compositionally distinct from the gypsum, mainly composed of Ca, S, O, and traces of Si (Figure 4I, yellow spectrum).

The area selected for LIMS measurements was sampled at 425 single surface positions (Figures 5B,C). The abundance of elements related to gypsum (O, S, Ca) and the fossil filament were analyzed in detail for the 425 positions (Figure 6). The gypsum element intensities are uniformly distributed across the sampled area, except in or around the filament. There, O, S, and Ca intensities drop significantly (Figure 6A, rows 2–7). Al, Si, C, and Mg intensities are at the background signal level (Figure 6A, middle and bottom panels). However, a significant increase in their intensities is observed in and around the filament (rows 2–10), with the presence of noticeable double peaks in rows 3–9 that align with the filament edges and decrease in intensity that corresponds to the hollow area (Figure 5D, rows 3–9). This pattern is evident as the measurement’s direction intersects the fossil filament (see black arrows in Figure 5C). The chemical composition difference is further illustrated in Figure 6B, where the intensity distribution of several elements across locations 91–101 are shown (Figure 6A row 2, see red rectangle in Figure 5D for location on the sample). At locations 92 to 99, corresponding to locations at and around the filament, the intensities of K, Li, Mg, Al, Si, C, B, and Cl increase significantly compared to gypsum locations 101 and 102, where only O, Na, Ca, and S were detected above the noise level, as expected for gypsum. A smaller increase is observed for P, F, V, Cr, Mn, and Co. Given the distance of 20 μm between single laser ablation craters, the length of the fossil filament in the horizontal direction is approximately 100 μm.

Figure 7 shows a comparison of mass spectra recorded from two distinct locations on the filament (location 110 and 144) and one location on gypsum (location 230) (Figure 7A). The mass spectrum of the filament at location 110 reveals the presence of numerous metallic and non-metallic minor elements, including Li, B, C, F, Al, P, S, Cl, K, Ti, V, Cr, Mn, Fe, Co., and Sr, alongside the major elements O, Mg, Si, and Ca. In contrast, the spectrum of the filament at location 144 (Figure 7B) exhibits distinct elemental abundances: F, Na, P, and Mg are less abundant, and significantly lower intensities are observed for Li, Al, and Si. Furthermore, B, F, Cl, V, Cr, Mn, Fe, and Co. are not detected at location 144. Consequently, the presence of these elements is correlated with the abundance of Al and Si, and similar relative abundances of these elements can be observed on average at each ablated location. A more pronounced variation in elemental abundance is evident when analyzing depth profiles, where micrometer-sized features (though less abundant) with varying elemental composition can be identified (e.g., carbonaceous deposits and other mineralogical grains like e.g., dolomite and clay minerals), as discussed below. The gypsum mass spectrum reveals the presence of H, O, S, Ca, and Na as major elements along with B, C, K, and Sr as minor trace elements, with concentrations at the ppm level. Some of the isotopes of Ti, Fe, Ni, Cu, Zn, As, and Se (Figure 7A) exhibit isobaric interference with several polyatomic species that correspond primarily to molecular species with compositions of Ca_xS_yO_z. It is noteworthy that an intensity increase for Ti is observed as well for its isotopes (⁴⁶Ti, ⁴⁷Ti, ⁴⁹Ti, ⁵⁰Ti), while an isobaric interference of ⁴⁸Ti with ³²S¹⁶O, and ⁵⁰Ti with ³⁴S¹⁶O or ³²S¹⁸O cannot be ruled out. Nevertheless, the observation of mass peaks from other isotopes confirms the presence of Ti. For the observation of ⁵⁶Fe (Figure 7A), the interference with ²⁸Si₂, Al₂, and ⁴⁰Ca¹⁶O cannot be ruled out. However, given that the ⁵⁴Fe/⁵⁶Fe isotope ratio is close to the expected value, the isobaric interferences appear to be minor.

The presence metallic and non-metallic elements at the filament location (Figure 7, location 110) C, H, N, O, P, S, Ca, Mg, Na, K, Cl, Al, Li, Cr, B, Si, F, V, Mn, and Co. can be attributed to two distinct groups 1) Biologically relevant elements, and/or 2) Elements involved in the mineral formation. This attribution is due to the faith of these elements before, during, and after the filament’s fossilization process. The biologically relevant elements are chemical elements essential for living organisms’ structure and function. The minerals within the fossil filament could have been controlled, induced, or influenced by the presence of living organisms or their organic matter. During the fossilization process, minerals that had been formed previously interact with the organic matter, resulting in its complete or partial replacement by a mineral phase. The elements within the organic matter could either be involved in the mineral formation, adsorbed, or preserved as organic inclusions. Thus, some detected elements within the fossil filament, such as C, H, O, Ca, Mg, Al, and Si, are attributed to both groups.

The biologically relevant elements C, H, N, O, P, and S (CHNOPS) constitute the bulk of major macromolecules. Ca, Mg, Na, K, and Cl are the major ions within procaryotic cells, while the main transition metals detected are V, Cr, Mn, and Co. The N peaks at m/z 14 show isobaric interference with Si⁺⁺ and is excluded from further analysis. B, Si, and F are involved in specialized metabolic processes (Kabiraj et al., 2022; Wackett et al., 2004; Ikeda, 2021). Lastly, Al, Li, and Cr are reduced and/or methylated by some prokaryotes. These elements were not found in the host, except for H, O, S, and Ca, typical gypsum components. B, C, Na, and K can be attributed to fluid inclusions within gypsum. Sr was also detected at the gypsum location and its presence is due to a substitution with Ca in the lattice during gypsum formation (Fei et al., 2022). Fe, Ni, Cu, Zn, As, and Se are also biologically relevant elements, but because of potential isobaric interferences with polyatomic species, they will not be discussed further.

High intensities of elements including H, O, Na, Mg, Al, Si, Ca, and K, and the difference of composition in locations 110 and 144, suggest that the inner filament layer is likely composed of a mixture of mineralogical species. A grid of 11 × 12 positions (in total 132 positions) (Figure 5D) was selected for further analysis, and atomic fraction maps were generated for selected elements, categorized into groups of biologically relevant elements and elements attributed to the mineral formation (Figure 8). For this analysis, the measured element intensities were corrected using relative sensitivity factors (RSCs) to obtain quantitative concentrations. The RSC factors were obtained from the mass spectrometric analysis of the geological standard reference material NIST SRM 610 and are summarized in the supplementary material section. The RSCs for H and B were assigned a value of 1 (no correction), as they are not certified within SRM 610. The same applies to C and F, but the RSC for C was assumed similar to Si and the RSC for F similar to Cl (similar physical and chemical properties). During the ablation and ionization process, some elements are easier to ionize (e.g., alkali elements) than others (e.g., F, Cl, O, S, C, Si). With the additional study of the SRM 610 geological standard, the determination of calibration factors to correct for the nonstoichiometric ionization efficiency becomes feasible (Neuland et al., 2016).

In Figure 8, the filament’s position within the studied area is indicated by significant increases or decreases in the intensities of specific elements observed at the corresponding locations (refer to, for instance, the intensity maps of ²⁷Al and ²⁸Si). This indicates a mineralogical composition distinct from the host material. The CHNOPS elements (excluding N, as explained earlier) and elements pertinent to microbial building blocks and metabolism can be identified in both the filament core and its surrounding layer. The H and O content in gypsum (CaSO₄·2H₂O) is notably lower compared to the fossil filament. The low H and O content is potentially attributed to the partial desiccation of gypsum to hemi-hydrate sulfate bassanite or the dehydrated anhydrite accruing in the vacuum chamber before LIMS measurements were started, although gypsum is considered quite stable at temperatures below 300 K (Tang et al., 2019). In contrast, the observed high H and O content within the fossil filament suggests the presence of hydrated swelling minerals and/or organic inclusions. In gypsum, S, Ca, and Na exhibit higher concentrations compared to the filament commensurate with the gypsum composition. The fossil filament is distinguished by a distinct composition evident by substantial enrichment of C, H, O, P, Mg, Al, Si, Li, K, Cr, V, Co., Mn, and Cl. This zone is of significance because it emphasizes the unique composition by presence of biologically relevant elements and mineralogical characteristics found in the inner layers of the fossil filament.

Figure 9A shows the chemical depth profile analysis of the gypsum mineral (location 90). There, the intensity of Ca, S, and O decays smoothly with increasing ablation layer, as expected for depth profile studies of a homogeneous composition (Grimaudo et al., 2020). The Si, Al, C, and Mg intensities show similar behavior and appear to be at background level over the 800 ablated layers. Figure 9B shows the chemical depth profile for the elements ¹²C, ²⁴Mg, ²⁷Al, ²⁸Si, ⁴⁰Ca, ³²S, ¹⁶O and H of the fossil filament (location 94) for the first 300 ablation layers. In the first 35 layers, an increase of the ¹²C intensity is anti-correlating with a decrease in ²⁴Mg, ²⁷Al, ²⁸Si, ⁴⁰Ca, ¹⁶O, and H intensities. At the ablation layers 35 to 50, a significant drop of the ¹²C intensity is observed, and anti-correlated with a considerable intensity increase of ²⁴Mg, ²⁷Al, ²⁸Si, ⁴⁰Ca, and H. This is followed by a second intensity peak of ¹²C and an intensity drop of ²⁴Mg, ²⁷Al, ²⁸Si, ⁴⁰Ca, and H at the ablation layers between 50 and 60, which anti-correlate again, indicating layers of carbonaceous deposits. At the ablation layers 110 to 140 and 190 to 210, two double peaks of ¹²C and ²⁴Mg are perfectly correlating with a smooth increase of ⁴⁰Ca and ¹⁶O at that depth. An interesting intensity drop of ³²S at the ablation layers from 170 to 210 anti-correlating with the second double peaks of ¹²C and ²⁴Mg, and the smooth increase of ⁴⁰Ca and ¹⁶O, suggests an inclusion of a mineral formed from Mg, Ca, C and O. Based on the AFM results, the size of the Mg, Ca, C, and O forming mineral, spanning 10 to 15 ablation layers, can be estimated to be approximately 8–10 μm in thickness. This measurement correlates with the size and mineralogical composition of the subspherical mineral observed in Figure 4.

As can be observed by eye and optical microscopy in Figures 2C, 4A–E, the Algerian PLG is characterized by long, hollow, sinuous, interwoven, pristine fossil filaments embedded in gypsum. Similar filamentous structures were documented in the PLG units at various locations across the Mediterranean Basin such as Italy (Panieri et al., 2010; Dela Pierre et al., 2015; Guibourdenche et al., 2022; Schopf et al., 2012), Spain, Greece (Natalicchio et al., 2022) and Cyprus (Allwood et al., 2013b).

The LIMS results enable the localization and investigation of the fossil filament’s chemical composition (Figures 6, 7), which reveals a complex and distinguished mineralogical composition from the host. The high intensity of the major elements at the edges of the filament and the significant intensity decrease in the middle (Figure 6A) are due to the hollowness of the filament, as observed under optical microscopy and confirmed with SEM (Figures 4F,G).

Considering the shape, color, and optical mineralogy of the different minerals observed within the filament (Figure 4) along with the LIMS results, the dominant mineralogical species are 1) Dolomite, 2) Clay minerals, and 3) Pyrite. Dolomite (CaMg (CO₃)₂), include its precursor phases such as high-magnesium calcite ((Ca,Mg) CO₃), disordered dolomite, and protodolomite (Fang et al., 2021), which exhibits variable Mg and Ca ratios across different locations in the filament. Calcium carbonates (CaCO₃) and magnesium carbonates (MgCO₃) might also be present but in minor concentrations. Clay minerals are identified by the heightened intensities of Si, Al, O, and H (Figure 7A), which are key components of the basic framework of clay minerals. Pyrite observed as black, opaque framboidal grains of rare and scattered occurrence (Figure 4K). The detailed analysis of pyrites using LIMS would require a specialized measurement campaign (Lukmanov et al., 2022), which is beyond the scope of this study.

Dolomite may exhibit different Mg/Ca ratios in distinct crystals within the measured spot and between the different locations, accounting for the stoichiometry variability, as depicted in the atomic fraction maps in Figure 8 across the filament. Additionally, the composition of clay minerals is diverse, potentially incorporating additional elements present in the depositional environment (Brigatti et al., 2006).

Clay minerals are composed of cations coordinated with oxygen atoms. Common cations include Si⁴⁺, Al³⁺, Fe³⁺, and Mg²⁺, but other cations, such as Li⁺, Mn²⁺, Co²⁺, Ni²⁺, Cu²⁺, Zn²⁺, V³⁺, Cr³⁺, and Ti⁴⁺ have also been detected within clay minerals (Brigatti et al., 2006). Cation exchange in clay minerals is reversible, diffusion-controlled, stoichiometric, and often selective for one cation over another (Brigatti et al., 2006). About 90% of a clay mineral’s surface consists of nanometric interlayer spaces accessible to water molecules and other compounds. Clay minerals are swelling hydrated minerals (Borrego-Sánchez and Sainz-Díaz, 2022; Chen et al., 2022), thus the presence of clay could explain the observed high intensities of H and O detected in the filament compared to gypsum (Figure 8). This explains the presence of the remaining elements at lower intensities, which are likely integrated into the composition or adsorbed (Figure 7A). The correlation is strongly supported by the absence or the significant decrease of most trace elements and biologically relevant chemical species in a clay-poor location, where only small Si and Al intensities are detected (Figure 7B).

Based on the structure, texture, and form of the clay minerals observed under SEM (Figures 4G,H), the clay mineral is most likely montmorillonite (a member of the smectite group), that forms as microscopic crystal precipitating from aqueous solutions (Christidis, 2012). Similar characteristics of hollow fossil filaments with traces of dolomite, carbonates and clay minerals were found in Messinian PLG at various locations across the Mediterranean Basin (Panieri et al., 2008; Natalicchio et al., 2022). The non-uniform horizontal and vertical distributions of the constituent chemical species, as depicted in Figures 8, 9B, are explained by the physical and chemical properties of dolomite and clay minerals.

The fossil filament also contains biologically relevant elements, which are primarily sequestered, adsorbed, or incorporated within the clay mineral structure due to its high capacity for adsorbing surrounding elements and organic molecules. The mass spectrum of the filament, presented in Figure 7, shows the presence of 19 metallic and non-metallic elements that are either directly, indirectly, or distantly related to life. For some of these elements, 2D atomic fraction maps are shown in Figure 8, including the CHNOPS elements (¹⁴N map is not presented because of the possible interference with the doubly charged ²⁸Si ion). The major cations Mg, Na, K and Ca are also detected (Figure 7). K cations are essential for many key cellular processes, including protein synthesis. The Na⁺/K⁺ transmembrane proton gradient plays a crucial role in energy generation and maintaining cellular integrity in bacteria when facing osmotic stress (Dibrova et al., 2015). Both Mg and Ca play crucial roles as cations for all prokaryotes, though Ca is less abundant in bacteria compared to Mg (Wackett et al., 2004). These elements are integral to various prokaryotic processes and the formation of biofilms (Wang et al., 2019). Chlorine is also identified in the mass spectrum of the filament at location 110 (Figure 7A). The anion Cl⁻ is essential for the survival of some marine bacteria, such as the halophilic bacterium Halobacillus halophilus (Roeßler and Müller, 2002), acting as a key factor in signal transduction, gene expression modulation, and protein synthesis (Sewald et al., 2007). Which may have played a vital role in the survival of extremophiles during the MSC, when salinity levels rose to 130‰–160‰ (Warren, 2016).

Some transition metals including V, Mn, and Co were detected in the fossil filament. Some of these elements are essential for growth and survival, acting as growth-limiting factors for prokaryotes even at relatively low concentrations (Andreini et al., 2008; Bradley et al., 2020; Buccella et al., 2019). Cobalt, for instance, is deemed essential in more than 20 cobalamin-dependent enzymes in prokaryotes (Xiu et al., 2021). Vanadium is involved in respiration, central metabolism, and DNA repair (Frawley and Fang, 2014; Argüello et al., 2013; Rehder, 2015). Mn plays a critical role in protecting bacteria from cellular damage induced by various stressors, including UV-C radiation, gamma-irradiation, wet and dry heat, as well as hydrogen peroxide (H₂O₂). In the deposition environment of the PLG, these elements would have been vital for the living extremophile prokaryotes. Chromium was also detected; however, it is considered toxic, and some prokaryotes employ survival strategies such as adsorption, absorption, accumulation, and transformation within their cells (Zahoor and Rehman, 2009). In contrast, Ti’s biological essentiality has not been discovered yet, although some marine bacteria have been observed adhering to TiO₂ particles (Hurst, 2022), which may be beneficial as a shield from UV radiations (Remick and Helmann, 2023).

The presence of Si detected in the fossil filament layers can be attributed to extracellular and intracellular bio-silicification by prokaryotes (Ikeda, 2021; Moore et al., 2020). This process allows cyanobacteria and other marine prokaryotes to survive harmful UV radiation by using siliceous deposits as a protective shield (Konhauser, 2006). It has also been demonstrated that silicon concentrations in some marine cyanobacteria are significantly higher than in sea water, suggesting an active silicon accumulation, often exceeding that of P and S (Ikeda, 2021). Considering the high salinity of the water column during the MSC–up to five times that of today’s seawater (Warren, 2016) – the studied filament may have experienced both extracellular and intracellular Si accumulation (Baines et al., 2012). This accumulation may have caused cellular lysis during entombment within the gypsum matrix, indicated by the observed 10 µm long opening that connects the two holes, as shown in Figures 4G,H. It is also observed that the same mineralogical composition is observed in the two holes and at the opening connecting them, suggesting a mineral formation from the same primary solution, which might be the cell cytoplasm. The presence of gypsum only at the two sides of the opening, without effectively separating the holes, strongly indicates that the structural forces involved did not allow for an even distribution of the gypsum (Figure 4H).

Boron was also found to be essential for N fixation and growth under N-deficient conditions in some cyanobacteria (Miwa and Fujiwara, 2009). However, F, which was also detected within the fossil filament, primarily occurs as fluoride (F⁻) in seawater and, along with certain transition metals, can be lethal to microorganisms when reaching certain concentrations (Pal et al., 2022). Lithium, despite its toxic nature, was also detected and is known to occur in certain marine bacteria living in salars (Remick and Helmann, 2023). The high-intensity peak of Al within the filament structure suggests its presence in the water column at elevated concentrations. However, Al lacks biological functionality and has low bioavailability. Its primary toxic effects in microbes arise from competition with Fe and Mg, as well as binding to DNA, membranes, or cell walls (Exley and Mold, 2015). Aluminum may have contributed to the formation of clay minerals through interaction with Si, O, and H. The presence of F, Al, and other toxic elements in the water column may have been responsible for cellular death, leading to the rapid entombment of filaments within the emerging PLG crystals.

Dolomite is rare in modern sedimentary environments, although seawater is generally oversaturated with respect to dolomite (Gregg et al., 2015; Pina et al., 2022). Abiotic synthesis of dolomite has been unsuccessful under Earth’s surface conditions of 25°C and 1 bar (Braithwaite et al., 2004) and has only been achieved under specific conditions at temperatures of 95°C (Yang et al., 2022). In contrast, gypsum can be easily precipitated in laboratory experiments at ambient temperature (Disi et al., 2023). Furthermore, it was also proven that under diagenesis or metamorphic conditions (temperatures higher than 55°C) gypsum transforms into anhydrite (Engel et al., 2021). This transformation could rule out the possible abiotic formation of dolomite within gypsum, which requires high temperatures and pressure conditions (Yang et al., 2022).

The dolomitization reaction is endothermic and influenced by the Ca²⁺ and Mg²⁺ ratio. Moreover, water molecules reduce the availability of Mg²⁺ for dolomite crystal growth (Petrash et al., 2017). The synthesis of true stoichiometric dolomite without prokaryotes involvement at room temperature and pressure in vitro remains elusive (Yang et al., 2022; Pina et al., 2022; Han et al., 2022; Yasumoto-Hirose et al., 2022). During PLG crystallization, the elevated concentration of sulfates (SO₄ ^2–) does not inhibit dolomite formation (Sánchez-Román et al., 2009). Prokaryotes contribute to dolomite precipitation in two ways: (1) By increasing the surrounding alkalinity through various metabolic pathways and/or (2) Concentrating ions within the cell envelope and extracellular polymeric substances (EPS). This induces localized supersaturation of Ca²⁺ and Mg²⁺ ions, overcoming low-temperature kinetic barriers to dolomite formation (Sánchez-Román et al., 2008; Sánchez-Román et al., 2009; Chan et al., 2004; Fowle and Fein, 2001; Roberts et al., 2013; Moreira et al., 2004). This process could be applicable to sulfate reducing bacteria and in some environments to sulfide-oxidizing bacterial filaments thriving in the PLG depositional water column (Dela Pierre et al., 2015; Moreira et al., 2004).

Prokaryotes significantly contribute to clay mineral formation, primarily by supplying Fe, which interacts with dissolved Si and Al. Additionally, their EPS serve as nucleation sites for cation species, influencing mineral precipitation and cell encrustation (Joo et al., 2023; Konhauser, 2006; Konhauser and Urrutia, 1999; Alimova et al., 2009; Jaisi et al., 2007; Cuadros, 2017; Mueller, 2015). After cell death and during the mineral replacement process, the organic structure can act as a template and may preserve organic matter (Banfield et al., 2001). Furthermore, clay minerals promote bacterial biofilm formation by retaining water, as well as organic and inorganic nutrients, while also forming a mineral barrier shielding the biofilm from external environmental conditions (Cuadros, 2017). This explains the detection of clay minerals within the fossil filament layer associated with biologically relevant elements and organic inclusions.

Given the context of this study, determining the biogenicity and syngeneity of the filament is crucial for drawing meaningful conclusions about the discovery of biosignatures, especially in the context of Mars exploration. Several criteria have been proposed to assess the biogenicity of microbial fossils. These include evaluating the habitability of the paleoenvironment where the organism was fossilized, the presence of carbonaceous material, evidence of growth into a fluid-filled fracture, the non-uniformity of filament diameter, hollowness, sinuous and curved morphologies, Fe-oxide or clay composition, and semi-parallel growth of filament, (McMahon et al., 2020; McMahon et al., 2021; Oehler and Cady, 2014). These criteria align with the observations made in our study and those reported in other fossil bearing PLG across the Mediterranean Sea. Additionally, the observed alternations between turbid and limpid laminae in the studied PLG indicate periodic phases of microbial mat development and gypsum precipitation (Dela Pierre et al., 2012; Dela Pierre et al., 2014).

The paleontological and environmental context of the PLG and the studied fossil filament meets all the mentioned criteria, in addition to an observed potential indication of lysis within the filament. Moreover, the presence of dolomite and clay minerals serves as an indirect biosignature of life, as their formation is induced and/or influenced by prokaryotes (Cosmidis and Benzerara, 2022). Notably, the presence of an inner layer within the fossil filament strongly indicate that the filament was trapped during the primary crystallization of the PLG and is not a product of a secondary diagenesis of gypsum. Additionally, the general orientation of the fossil filaments, primarily parallel to the crystal growth plane, strongly suggest their syngenetic deposition as the PLG crystal grew.

The MSC was a brief yet impactful palaeoceanographic event that left a lasting imprint on the depositional environments of the Mediterranean. The sedimentary sequences linked to the MSC reflect rapid and profound environmental fluctuations, which challenged the adaptive capabilities of life forms in regions experiencing extreme conditions such as rapid evaporation, high salinity, anoxia, and desiccation leading to a notable dominance of halophilic archaea and bacteria. This shift is evidenced by the observed pattern in lipid biomarkers (Natalicchio et al., 2019), as well as variations in morphological and mineralogical diversity (Lugli et al., 2010; Natalicchio et al., 2013; Dela Pierre et al., 2015; Wang et al., 2023; Bailey et al., 2009). The presence of these biosignatures provides insights into the redox conditions at the seafloor and indicates an active biogeochemical sulfur cycle within Messinian brines (Schulz and Jørgensen, 2001). It is well documented that bacteria and archaea play a crucial role in the sulfur cycle during the crystallization of gypsum. This phenomenon has been observed at Gypsum Hill Spring in the Canadian High Arctic, where cold, hypersaline, anoxic conditions, and highly reducing brines rich in sulfate and sulfide characterize the environment. This environment serves as a Mars analog (Magnuson et al., 2023).

The shift in Martian climate from the wet Noachian era to the dry Hesperian period (4.1–3.0 billion years ago) likely led to the presence of saline surface waters rich in sulfur species (Macey et al., 2020b). As previously stated, this climatic transition resulted in the deposition of various sulfate minerals at different locations on Mars (Kounaves et al., 2010; Liu et al., 2024; Wilk et al., 2024; Rapin et al., 2019; Ettehadi et al., 2023). Recently, NASA’s Perseverance rover has detected hydrated Ca-and Mg-sulfate minerals in the Noachian-age Jezero crater, which hosts a fan-delta system formed by a river draining into a paleolake. This lake was likely precipitated from a low temperature sulfate-rich fluid of moderate pH during multiple episodes (Siljeström et al., 2024).

The formation of hydrated gypsum on Mars suggests an origin in arid, oxidizing, saline, and acidic shallow surface groundwaters (Michalski et al., 2022; Benison and Bowen, 2006; Benison and LaClair, 2003; Squyres and Knoll, 2005). Given that, on Earth, the formation of carbonates (dolomite) depends on a thick atmosphere that moderates surface temperatures and a hydrosphere with a neutral to moderately alkaline nature (Fernández-Remolar et al., 2012), the occurrence of carbonate deposition on a cold and acidic Mars would have not been possible, conditions not favorable to maintain a relatively high activity of CO₃ ²⁻ anions in solution. In these low-pH depositional environments on Mars the abiotic origin of dolomite within gypsum could be ruled out and, prokaryotes, if ever existed, would have played a crucial role in forming dolomite by elevating overall alkalinity. Thus, if dolomite or carbonate minerals are discovered in Martian gypsum or hydrated sulfates in general, a prokaryotic origin should be considered.

It is important to emphasize that proto-dolomite formation may also be mediated by clay minerals, such as montmorillonite, at pH 9.7 (Liu et al., 2019). However, considering the acidic depositional environment associated with sulfate minerals on Mars, the potential mediation of protodolomite by montmorillonite is unlikely. Furthermore, given the low temperature and pressure deposition conditions of ancient sulfate minerals on Mars, the abiotic formation of dolomites within gypsum, which requires high temperatures (95°C) and pressure conditions (Yang et al., 2022), could be ruled out. Also, if dolomite is formed abiotically under these high temperatures, the host gypsum containing dolomite, would have undergo diagenesis at earlier temperatures levels of 55°C and transforms to anhydrite (Engel et al., 2021). Despite this, the presence of dolomite should not be automatically interpreted as evidence of biological activity unless alternative origins can be conclusively ruled out (Banfield et al., 2001).

The presence of several other biosignatures detected with LIMS and optical microscopy in our study, which validate the biogenicity of the fossil filament, should be considered when searching for life in sulfate deposits on Mars. More intriguingly, the identification of clay minerals within the fossil filament has significant implications for investigating biosignatures on Mars. Firstly, the induced and influenced clay mineralization on the outer prokaryotic cells can serve as indirect evidence of biomineralization (Jaisi et al., 2007). Secondly, clay minerals demonstrate robust absorption capabilities, enhancing their potential for bio-preservation (Broz, 2020). Hence, it is highly probable that the clay minerals present significantly influenced the preservation of morphological and elemental biosignatures in the examined fossil filament and potentially in Martian hydrated sulfate minerals.

Figure 10 is a flowchart summarizing the complimentary measurements made in this study and, techniques used for each measurement, the interpretation that led to the conclusion about the biogenicity (or abiotic) origin of fossil filament.