Reply to Wierzchos et al.: Microorganism-induced gypsum to anhydrite phase transformation

In response to Wierzchos et al. (1) regarding the mechanism of water extraction from gypsum rock by desert colonizing microorganisms (2), we provide details that refute their incorrect assessments.

We carefully selected areas without microorganism colonies that only contained gypsum, as confirmed by X-ray diffraction (XRD), for our culture experiments. Raman is useful for localized analysis of phase on surfaces, but here, it is complementary to XRD, which provides sample-wide detection (3, 4). Furthermore, selected area electron diffraction (SAED) confirms anhydrite, exhibiting arcs at 0.347 ± 0.003 nm and 0.278 ± 0.002 nm [(200) and (211) planes, respectively] with a [01−1] zone axis (Fig. 1 A and B). The closest reflection for gypsum, the (130) plane, was not observed. SAED and fast Fourier transform (FFT) show twofold symmetry of the arcs, while the gypsum (130) plane exhibits fourfold symmetry, which was not observed. Finally, the crystal morphology of gypsum (monoclinic) was observably different from anhydrite (orthorhombic) (Fig. 1 C−F). Thus, XRD, high-resolution transmission electron microscopy (TEM)/SAED, and scanning electron microscopy (SEM) confirm the phase transformation to anhydrite in our samples.

Fig. 1.

SEM and SAED of coupons in the control and experimental groups. (A and B) SAED and FFT of anhydrite crystals in the experimental group. In A, (200) and (211) planes are present in the SAED from the TEM image (Inset). Reference: The International Centre for Diffraction Data, ICDD #72–0916; American Mineralogist Crystal Structure Database, AMCSD 0005117. Twofold symmetry arcs indicate the aligned directions in two different mesocrystals marked with red dashed arrows. In contrast, the closest gypsum (130) plane has a d spacing of 0.3556 nm, which is not observed. Reference: ICDD #33-0311; AMCSD 0001809. (B) FFT of a single mesocrystal, showing the arcs corresponding to (200) planes and the twofold symmetry. (C and D) SEM micrographs show monoclinic gypsum crystals in the control group. (E and F) Mesocrystals of anhydrites in the groups cultured with cyanobacteria.

In our experiments, the size of the gypsum coupons were 5 mm × 8 mm × 0.5 mm and should be corrected as an erratum in ref. 2. Optical micrographs (Fig. 2 A and B) of coupons after culture experiments confirm the stability of the gypsum coupons. Based on the finite solubility of gypsum (∼2.5g/L), only ∼0.4 wt % of the coupon would dissolve in 100 µL of medium. Unlike highly soluble minerals, which can often undergo Ostwald ripening (i.e., those in reference 20 of ref. 2) to form large single crystals, the anhydrite crystals that formed as the result of microbial activity (2) were mesocrystalline. Here, the weakly soluble mineral crystallized via particle attachment at a lower energy barrier than via Ostwald ripening (5). In fact, the salinity, pH, temperature, and water activity within biofilms can affect the phase stability and transformations, even at room temperature (6, 7). Furthermore, microorganisms in biofilms are embedded in a matrix of hydrated extracellular polymeric substances, maintaining a highly hydrated microenvironment around the cells (8, 9). While the biofilm was desiccated in our experiment, because of its very nature, it retained water, as shown in figure 3F in ref. 2, enough for ion transfer. On biofilms, we did not explicitly state separation of cell aggregate and biofilm (figure 3D in ref. 2 might have led the authors of ref. 1 to be confused).

Fig. 2.

Optical and SEM micrographs of gypsum coupons after culturing with low and high concentrations of cyanobacteria. (A and B) Optical micrographs indicate coupons were not dissolved after culturing. The green cyanobacteria colonies are not uniformly distributed on the substrate. (C and D) Low-magnification SEM micrographs show cyanobacteria has preferred attaching planes. No cyanobacteria is observed in the yellow dashed circle.

The preferential attachment of cells to mineral facets and interactions at those specific interfaces have been discussed in previous research (10, 11). In our study, while geological samples can present challenges for data interpretation, our SEM micrographs from carefully sectioned coupons in cultured experiments clearly show the presence of cyanobacteria and biofilms only on specific crystallographic planes (Fig. 2). Interestingly, figure 1 C and D from ref. 1 also demonstrates facet-specific cell growth, refuting their own claim. We cite relevant literature regarding the study of gypsum endoliths, including several papers by the authors of ref. 1 (see references 6 through 9 and 10 in ref. 2). Thus, we reaffirm our results, which do provide insights into potential life in even more extreme environments, such as Mars.