Symbiotic cooperation between freshwater rock-boring bivalves and microorganisms promotes silicate bioerosion

Overall, the characterizations detailed above point out that the contact between the bottom of the macroborings and the shells of the bivalves likely have represented a hotspot of microbial activity, which was not observed elsewhere at the surface of the siltstone devoid of macroborings. In the next two sections, we discuss how the association between Lignopholas fluminalis and microorganisms may have acted symbiotically to facilitate boring in siltstone.

A possible strengthening of the mechanical abrasion thanks to microbial EPS

As mentioned above, macroborings resulting from bioerosion are most often observed either in calcareous rocks, which are highly sensitive to bioweathering, or in soft substrates such as peat or clays, which are readily drilled through bioabrasion. Here, the siltstone is both chemically much more resistant than carbonates and harder than the substrates commonly subjected to bioabrasion.

Bolotov et al.⁸ have reported that the mean hardness of the siltstone was 62 kgf mm⁻², i.e., twice as much as clayey materials. In comparison, the compilation of Yang et al.¹⁷ indicates that the hardness of Bivalvia shell is an order of magnitude lower than that of quartz, and only slightly greater than that of the siltstone, ranging between 110 and 270 kgf mm⁻². In addition, both the structure and hardness of the siltstone were found to be homogeneous, such that it is unlikely that Lignopholas fluminalis took advantage of any local weakness to bore into the rock. Finally, the macroboring walls did not exhibit any marks, as opposed to the experimental results obtained by Nederlof and Muller¹⁰ using the piddock Barnea candida, which is a close relative of Lignopholas fluminalis⁸. However, such scrap marks resulting from the abrasion of the substrate by the denticles of the piddocks were obtained by rotating the shells in a soft materials (wax). The bioabrasion ability of the shells of Barnea candida is thought to be limited to soft substrata such as clays or peat and most likely, they cannot abrade harder substrata such as chalk¹⁰.

Similarly, we argue that the various features collected here suggest that there is no clear evidence that the direct contact between the shells of Lignopholas fluminalis and the substrate is responsible for the bioabrasion of the siltstone. Instead, single grains excavated from the borehole partly remained trapped at the surface of the shells, embedded into an organic matrix that we interpreted as a biofilm. It can reasonably be assumed that these single grains, which are essentially hard minerals such as quartz and feldspars, acted like abrasive materials that contributed to drill the siltstone through the rotation of the shells. Therefore, the presence of microorganisms in the interfacial region between the substrate and the borers possibly strengthened their boring ability, although at that point, it remains impossible to state whether this interaction is obligatory or facultative. In any case, from a mechanical standpoint, Lignopholas fluminalis bivalves likely took advantage of the biofilm attached to the surface of their shells to increase their boring ability.

Enhancing the weakening of the rocks through microbially-induced weathering

In addition to bioabrasion, some macroborers are also known for their ability to promote bioweathering⁵. This mechanism of bioerosion is suggested to be limited to calcareous substrates and not significant for substrates such as siltstones, whose rock-forming minerals have a dissolution rate that is between 6 and 8 orders of magnitude lower than that of calcite at circum-neutral pH conditions (according to rate data from¹⁸ for quartz¹⁹, for albite²⁰, for chlorite and²¹ for calcite).

Notwithstanding, we argue that mass transfer did occur during the process of boring discussed in the present study. We detail below the reasons why we think that this mass transfer cannot result from the abiotic dissolution of the grains by the bulk fluid, and suggest that microorganisms were responsible for the dissolution of the siltstone, which ultimately facilitated the formation of borings by Lignopholas fluminalis.

The strongest evidence for mass transfer is the occurrence of secondary Mn-rich crystals found embedded in an organic matrix at the bottom of the macroborings. Because such minerals were not found elsewhere in the rock sample, this finding indicates that the contact region between the bivalves and the siltstone was not simply mechanically eroded, but also chemically weathered. The source of Mn is most likely chlorite, which represents the richest source of Mn among the rock-forming minerals (0.2 to 0.7 wt% according to quantitative EDX analyses). In addition, the location of the minerals (specifically embedded in the organic matrix) indirectly suggests that microbes were responsible for the dissolution of chlorite. This latter assertion can be further supported by comparing the residence time of a chlorite grain at the bottom of a pit to the time required to dissolve chlorite with a bulk aqueous fluid:

First, several studies estimated the lifespan of bivalve piddocks of the family Pholadidae (to which Lignopholas fluminalis belongs) to be on the order of 10 years²². The deepest macroborings that we observed, possibly corresponding to the oldest bivalves, were on the order of 1 cm, leading to a mean erosion rate of R_erosion = 1 mm yr⁻¹.

Second, the grain size of the siltstone is comprised between 0.2 and 50 µm, with an average value around Ø = 10 µm⁸. The average time (t) required for a 10-µm grain to be excavated from the bottom of the pit and released to the environment can thus be estimated following:

yielding t = 10^–2 year. This value indicates that Mn must be efficiently released from chlorite over a time interval as short as 10^–2 year (~ 3.7 days) to be incorporated into secondary minerals.

Finally, the radial retreat (∆h) of a hypothetical spherical grain of chlorite dissolved over a time interval of 3.7 days, can be calculated using:

where M , ρ, and R_chlorite stand for the molar mass, the density and the dissolution rate of chlorite, respectively. Considering the rate data from Lowson et al.²⁰, the far-from-equilibrium dissolution rate of chlorite at room temperature and circum-neutral pH conditions can be estimated to be on the order of 10^–17 mol cm⁻² s⁻¹. Considering a typical value of ρ = 3.0 g cm⁻³ for chlorite and a molar mass of M = 697 g mol⁻¹, ∆h is on the order of 0.1 Å, i.e., much less than an atomic monolayer at the chlorite surface. These crude calculations illustrate that Mn mobilization through the dissolution of chlorite with a circum-neutral pH fluid is highly unlikely. Therefore, an alternative mechanism to explain this mass transfer requires the existence of a microenvironment with greater weathering properties, such as that provided by microbial biofilm.

Several studies have demonstrated that microenvironments can be generated at the silicate-microbe contact²³, where the local conditions in terms of pH and saturation state strongly differ from the bulk conditions^24,25, with the development of surface biofilms further intensifying this effect through hydraulic decoupling²⁶. Although the large-scale impact of chemical compounds secreted by microbes on silicate weathering rates remains an open and controversial question (e.g.^27,28,29,30), several studies showed that chemically aggressive conditions (low pH, high concentration of organic acids) can result in a significant increase of silicate weathering rates, at least locally^25,31. Here, an increase of the dissolution rate of chlorite by up to two orders of magnitude would have been required to get an appreciable release of Mn. According to the dissolution rate law developed by Lowson et al.²⁰, such an increase can be reached if the local pH conditions in the vicinity of chlorite are on the order of 3, a value that is fully compatible with pH measured in some microbial biofilms in previous studies²⁴.

The microorganisms are the major catalysts of manganese cycling in the natural environment³² and manganese is a micronutrient essential for the development of microbial communities, for which rocks represent the main source³³. As such, it might have been targeted by microbes for several reasons, which include Mn oxidation by chemolithoautotrophs^32,33,34 or incorporation as enzyme cofactor³⁵.

One can wonder whether (i) the borers specifically targeted areas where microbes were already thriving at the surface of the siltstone and actively dissolving the crystals, or (ii) whether attachment of macroborers was a prerequisite to the establishment of microbial communities dissolving the siltstone. Supporting the first assertion, a few studies have proposed that microborings supposedly attributed to microbial weathering (e.g.,³⁶) might weaken rocky substrates, eventually facilitating the subsequent drilling of microborings by bivalves¹⁴. However, all occurrence of silicate microborings that we are aware of dealt with volcanic rocks and more specifically, pre-fissured basalt glass^15,36,37. As a matter of fact, our multiscale investigation of the rock substrate did not reveal the presence of any tubular microchannels, and biofilms were not observed anywhere other than in macroborings. As a consequence, we speculate that a nascent bioabrasion of the substrate by the bivalves was required to allow for the establishment of microbial communities and trigger the onset of microbial weathering. Supporting this assertion, freshwater mussels are known to concentrate limiting nutrients such as C, N and P in the benthos and stimulate biofilm growth (³⁸ and references therein). In turn, microbially-induced rock weathering likely contributed to a greater dissolution along grain boundaries, ultimately facilitating grain detachment and rock-boring by Lignopholas fluminalis. Of note, this mechanism would be the biotic equivalent of the abiotic erosion and weathering of limestone³⁹.

To conclude, our study sheds new light on the possible mechanisms of silicate bioerosion by macroborers. On the one hand, we suggest that microorganisms likely benefited from the early stages of siltstone drilling by macroborers to thrive at the bottom of macroborings. On the other hand, we provide evidence that microbes contributed to bioerosion by actively dissolving minerals, while hard minerals (quartz and feldspars) trapped in biofilms at the surface of the shells further facilitated the development of macroborings via mechanical abrasion. Therefore, the association between Lignopholas fluminalis and microbes has the main characteristics of what is commonly defined as a symbiotic action. Finally, this finding also raises three main concluding remarks:

1. (i)

   In addition to the increase in macrofaunal diversity previously reported⁷, the development of macroborings also likely contributed to an unexpected increase of microbial diversity that remains largely unexplored;

2. (ii)

   Our study underlines that preventive strategies to mitigate bioerosion might have to target on suppression of bacterial biofilm development in order to achieve effective solutions;

3. (iii)

   Finally, although the contribution of microbes to silicate weathering at large space and time scales remains unknown and debated, the present study suggests that this impact is far from negligible when coupled to macroborers in what appears as a symbiotic relation. As suggested here, such microbial communities may contain specific microorganisms with efficient weathering-ability, which would be worth investigating to possibly identify efficient bioinspired strategies of silicate weathering, of prime importance for a range of industrial and societal concerns including CO₂ sequestration.

Source: Ecology - nature.com