Carbonatogenic Bacteria from Corallium rubrum Colonies

Abstract

The precipitation of minerals, in particular carbonates, is a widespread phenomenon in all ecosystems, where it assumes a high relevance both from a geological and biogeochemical standpoint. Most carbonate rocks are of biological origin and made in an aquatic environment. In particular, bioprecipitation of carbonates is believed to have started in the Mesoproterozoic Era, thanks to a process often driven by photosynthetic microorganisms. Nevertheless, an important contribution to carbonate precipitation is also due to the metabolic activity of heterotrophic bacteria, which is not restricted to specific taxonomic groups or to specific environments, making this process a ubiquitous phenomenon. In this framework, the relationship between carbonatogenic microorganisms and other living organisms assumes a particular interest. This study aims to isolate and identify the culturable heterotrophic bacterial component associated with the coenosarc of Corallium rubrum in order to evaluate the occurrence of strains able to precipitate carbonates. In particular, the study was focused on the identification and characterisation of bacterial strains isolated from a coral coenosarc showing a high carbonatogenic capacity under laboratory conditions. Samples of C. rubrum were taken in the coastal waters of three Italian regions. The concentration of the aerobic heterotrophic microflora colonising C. rubrum coenosarc samples spanned from 3 to 6∙10⁶ CFU/cm². This variation in microbial populations colonising the C. rubrum coenosarc, spanning over 6 orders of magnitude, is not mirrored by a corresponding variability in the colony morphotypes recorded, with the mean being 5.1 (±2.1 sd). Among these bacteria, the carbonatogenic predominant species was Staphylococcus equorum (93% of the isolates), whereas Staphylococcus xylosus and Shewanella sp. accounted only for 3% of isolates each. All these strains showed a remarkable capacity of precipitating calcium carbonate, in the form of calcite crystals organised radially as well crystalised spherulites (S. equorum) or coalescing spherulites (Shewanella sp.). S. xylosus only produced amorphous precipitates of calcium carbonate. All bacterial strains identified were positive both for the production of urease and carbon anhydrase in vitro at 30 °C. It seems that they potentially possess the major biochemical abilities conducive to Ca²⁺ precipitation, as they showed in vitro. In addition, all our carbonatogenic isolates were able to hydrolyse the phytic acid calcium salt and then were potentially able to induce precipitation of calcium phosphates also through such a mechanism.

1. Introduction

The precipitation of minerals, in particular carbonates, mediated by microorganisms is a widespread process in terrestrial, freshwater, and marine environments [1,2] and it is referred as biomineralisation. Such a process is defined as the ability of certain organisms to form minerals by favouring their chemical precipitation as a consequence of their metabolism [3]. In particular, in a marine environment two main typologies of microbially induced carbonate precipitation (MICP) have been found. The first one is borne by microorganisms alone and mostly results in precipitation of calcium carbonate minerals, with microbialites being the common product [4]. Stromatolites are among the most ancient and studied forms of microbial precipitation of carbonates in marine environments. They are mainly produced by oxygenic phototrophs (cyanobacteria) and anoxygenic phototrophs (green and purple sulfur bacteria and green non-sulfur bacteria) [5,6] making microbial mats, which are communities typically containing thousands of microbial species and that can also harbour small molluscs and crustaceans. The second type of MICP takes place in the microbiota of sessile marine organisms that have an enclosing exoskeleton made by carbonatic minerals, such as corals.

Ivanov et al. (2017) [7] reported that concrete biocoating, with a layer of calcium carbonate crystals, can be obtained using urease-producing bacteria, calcium salt, and urea. Such a calcium carbonate biocoating could be able to protect concrete constructions in marine environment from reactions of magnesium and bicarbonate ions with the calcium silicate hydrate gel of concrete. More interestingly, this carbonate layer promotes colonisation of an epibiota mimicking a natural coral reef. Recent studies on the ecological relationships between microorganisms and corals have led to the birth of coral microbiology [8], mainly driven by the concern arising from the spread of diseases causing heavy losses in the coral populations [8,9] or the involvement of bacterial symbionts in host resilience to anthropogenic stress [10]. Indeed, coral microbiology should be considered from a wider standpoint, including all the microorganisms interacting with these sessile organisms, comprising those that play both an ecological and physiological role, as very little is known about the possible involvement of MICP in coral exoskeleton formation.

Corallium rubrum (L., 1758), a scleralcyonacea (Anthozoa, Octocorallia) belonging to the family of Coralliidae, is a coral of particular importance both from a commercial and historical point of view [11] and from an ecological perspective [12,13]. Despite all this, little is known about the microbiota associated with C. rubrum. It is known that corals are associated with a multitude of different microorganisms colonising their surface mucus layer, tissue, and calcium carbonate skeleton [8,14]. Koren and Rosenberg (2006) [15] described several bacteria belonging to the family of Vibrionaceae associated with the hexacoral Oculina patagonica. However, some bacterial species, particularly Vibrio corallilyticus, V. harveyi, V. splendidus, and V. shilonii, have been recognised as a possible cause of some coral diseases, such as coral bleaching [16,17]. Once again, the emphasis is put on putative virulence microorganisms that are able to harm the survival of coral populations.

During a survey of the coralligenous biocenosis surrounding the island of Procida (Southern Italy), carried out by Pasquale et al. (2011) [18], Actinomyces sp., Bacillus licheniformis, B. megaterium, B. mycoides, B. pumilus, Bacillus sp., Kocuria palustris, Kocuria sp., Pseudoalteromonas sp., Pseudomonas stutzeri, Pseudomonas sp., Staphylococcus equorum, S. vitulinus, Vibrio harvey, V. hepatarius, and V. shilonii were reported as the most representative bacteria belonging to the culturable heterotrophic bacterial population associated with C. rubrum. Such a complex ecological and physiological relationship between the coral and its microbiota is still to be fully understood.

The interplay between microorganisms and minerals at large includes both the weathering process, via dissolution of mineral crystals, and the process of making new minerals by the precipitation of ions occurring in microenvironments surrounding the microbial cells.

In this context, the phytic acid calcium salt hydrolysis ability of MICP bacteria in marine environment deserves particular attention, as the bacterial phytase activity has been closely associated with precipitation calcium phosphate compounds (CPCs) [19,20]. Since CPCs are more stable towards dissolution processes than CaCO_3, phytase activity takes on a certain importance.

According to Roeselers and Van Loosdrecht (2010) [19], the degradation of Ca-phytates, by microorganisms producing phytase, can be conducive to the formation of extremely stable calcium phosphate crystals [21].

One of the most studied mechanisms is the precipitation of CaCO₃ mediated by urea according to the following reactions [22]:

CO (NH₂)₂ + H₂O → NH₂COOH + NH₃

NH₂COOH + H₂O → NH₃ + H₂CO₃

H₂CO₃ ↔ HCO₃⁻ + H⁺

NH₃ + 2H₂O → 2NH₄⁺ + 2OH⁻

HCO₃⁻ + H⁺ + 2NH₄⁺ + 2OH⁻ → CO₃²⁻ + 2NH₄⁺ + 2H₂O

Ca²⁺ + Cell → Cell⁻Ca²⁺

Cell⁻Ca²⁺ + CO₃²⁻ → Cell⁻CaCO₃

This study aimed to isolate and identify the culturable heterotrophic bacterial component associated with the coenosarc of C. rubrum in order to evaluate the possible occurrence of strains that are able to precipitate carbonates. In particular, the study was focused on the identification and characterisation of bacterial strains showing a high carbonatogenic capacity under laboratory conditions.

2. Materials and Methods

2.1. Collection of C. rubrum Colonies and Sampling of Coenosarc

In May–June 2013, 35 colonies of C. rubrum were collected from the coralligenous communities of 3 Italian regions: Liguria (10 samples: L 1–10), Tuscany (9 samples: T 11–19), and Campania (16 samples: C 20–35) (Figure 1;

Table 1. Geographic coordinates of sampling stations and sampling depths.

Sampling Site	Sample	Geographic Coordinates	Depth of Sampling (m)
1	L 1–5 *	43°44.67′ N 7°40.917′ E	73
2	L 6–10 *	44°17.80′ N 9°12.140′ E	68
3	T 11–14 *	42°38.40′ N 10°6.49′ E	70
4	T 15–19 *	42°25.08′ N 10°04.81′ E	85
5	C 20–21 *	40°45.386′ N 14°10.849′ E	100
6	C 22–25 *	40°46.090′ N 14°16.165′ E	120
7	C-26	40°34.520′ N 14°25.022′ E	180
8	C 27–28 *	40°34.619′ N 14°24.688′ E	90
9	C 29–31 *	40°44.275′ N 14°01.097′ E	60
10	C 32	40°42.246′ N 13°50.944′ E	120
11	C 33	40°43.673′ N 13°49.493′ E	80
12	C 34–35 *	40°44.987′ N 14°01.589′ E	50

L: Liguria region; T: Tuscany region; C: Campania region. * Different samples of C. rubrum were collected from populations at the same site.

).

The coral samples were collected by means of a Remotely Operated Vehicle (ROV) equipped with an extendable arm. Immediately after sampling, fragments of C. rubrum were washed with sterile saline solution (NaCl 0.85%) to remove microorganisms not firmly attached to the coenosarc.

Sterile cotton swabs were gently rubbed on ca. 10 cm² of the middle part of the coral surface; transferred to sterile tubes, containing 2 mL of sterile saline solution; and then stored at 8–10 °C. The microbiological analyses were carried out within 24 h from sampling.

2.2. Isolation of Culturable Heterotrophic Bacteria Associated with C. rubrum Coenosarc

The saline solution inoculated tubes were vortexed for 1 min before carrying out the serial decimal dilutions to uniformly disperse the microorganisms in the liquid phase. culturable heterotrophic bacteria (CHB) were enumerated by the spread-plating method as follows. Aliquots (0.1 mL) of the dilutions were inoculated in triplicate in R2A Agar (Oxoid, UK) supplemented with NaCl (1%) (R2A-NaCl). The plates were incubated aerobically at 20 °C until 10 days. CHB were expressed as CFU/cm² of coral surface including mucus and tissue.

Within the 3 replicas of each sample, well-isolated colonies, representative of the different morphotypes observed on plates, were selected using a stereomicroscope. The selected colonies were purified in R2A-NaCl and then frozen at −80 °C in Tryptone Soya Broth (Oxoid) with glycerol (15%).

2.3. Screening of Carbonatogenic Strains

The frozen bacterial strains were revitalised in Tryptone Soya Agar (TSA) (Oxoid) and inoculated into plates containing Carbonate Bioformation Agar (CBA), consisting of TSA supplemented with CaCl₂·2H₂O (1% w/v). During the preparation of CBA, CaCl₂·2H₂O was sterilised separately in an autoclave. The plates were incubated aerobically at 30 °C for up to 10 days in humidified bags. The strains that during the incubation and upon observation with a stereomicroscope showed the formation of crystals were selected to evaluate their ability to precipitate Ca²⁺ in broth medium.

2.4. Carbonate Bioformation

Bacterial strains, showing crystal formation in CBA, were transferred to Carbonate Bioformation Broth (CBB) consisting of Nutrient Broth (Liofilchem, Italy) 3 g·L⁻¹, CaCl₂·2H₂O 36.682 g·L⁻¹, NaCl 10 g·L⁻¹, and urea 20 g·L⁻¹. The Ca²⁺ concentration was chosen according to Jarwar et al. (2023) [23] and Okwadha and Li (2010) [24]. Except for urea, which was sterilised by filtration at 0.2 µm, all the other ingredients were sterilised separately in an autoclave at 121 °C for 15 min; the final pH of the medium was 6.5.

In order to check the influence of pH on carbonate precipitation, the pH of the sterile reacting media containing 10 g L⁻¹ of Ca²⁺ was increased up to 8.5 by using 4% aqueous solution (w/v) of NaOH.

After growth in TSA, the bacterial strains were suspended in saline solution until they achieved a turbidity of the 0.5 MacFarland standard. From each suspension 1 mL was transferred in triplicate into two conical tubes containing 30 mL of CBB. The CBB tubes, with and without bacterial inoculum, were incubated aerobically at 18 and 30 °C on an orbital shaker (60 rpm) for up to 30 days. At 3, 10, and 30 days of incubation, 2 mL of broth was taken from the tubes and centrifuged at 6000 rpm for 10 min; the supernatants, after filtration at 0.2 µm, were analysed for the determination of Ca²⁺ by ion chromatography [25].

The ability of the different strains to reduce Ca²⁺ concentration in the liquid phase by precipitation was classified according to 3 arbitrary categories: slight (≤33% g∙L⁻¹ Ca²⁺), moderate (34–66% g∙L⁻¹ Ca²⁺), and high (≥67% g∙L⁻¹ Ca²⁺).

2.5. Molecular Identification of Bacterial Strains

Strains capable of removing at least 67% of the initial Ca²⁺ in the CBB at 18 or 30 °C within 30 days of incubation were selected and characterised molecularly and biochemically, and their ability to form crystals of CaCO₃ in vitro was checked.

The strains were identified using PCR amplification and direct Sanger sequencing of the 16S ribosomal RNA gene. DNA was extracted from colonies grown on TSA using a Lyses & PCR-GO Kit (Bacteria/Fungi) following manufacturer’s instructions. PCR amplification was carried out by sequencing the 16S rRNA gene (region V1–V6) using universal primers 63f (5′-CAG GCC TAA CAC ATG CAA GTC-3′) for the forward primer and 1387r (5′-GGG CGG WGT GTA CAA GGC-3′) for the reverse primer [26]. PCRs were performed with the following conditions: initial denaturation at 98 °C for 5s, followed by 35 cycles of denaturation at 98 °C for 5 s, annealing at 55 °C for 5 s and extension at 72 °C for 18 s, and final extension at 72 °C for 45 s.

Sanger sequencing was performed by using a 3130 Genetic Analyzer (Applied Biosystems, Carlsbad, CA, USA).

A sequence similarity search was performed using the BLAST algorithm against the GenBank database (http://blast.ncbi.nlm.nih.gov/Blast.cgi, accessed on 10 May 2025).

2.6. Biochemical Activity of Bacterial Strains

Urease activity was carried out according to Murray et al. (1995) [27] and modified as follows: qualitative urease activity was carried out by inoculating the highly carbon precipitation performers’ strains into 10 mL of Urea Indole Medium (Bio-Rad Laboratories, Hercules, CA, USA) and by incubating them at 30 °C for up to 3 days in aerobic conditions.

The microorganisms producing urease induce a colour change from yellow to red-fuchsia caused by the alkalinisation of the medium due to the release of ammonium from urea [23].

A qualitative carbonic anhydrase (CA) assay was performed according to Jarwar et al. (2023) [23]. Briefly, cellulose blank disks (diameter 5 mm) were soaked with 20 μL of p-nitrophenyl-acetate (p-NPA) solution, containing 0.67 mg of p-NPA (Sigma-Aldrich, St. Louis, MO, USA) dissolved in 3 mL of absolute ethanol (Oxoid), and then left to dry at 37 °C for 1 h and stored at −20 °C. The CA assay was performed by placing a disk near the colonies of the bacterial strains grown in TSA at 30 °C for 48 h. The CA-producing bacteria, by inducing the hydration of p-NPA to p-nitrophenol and acetate, produce the yellowing of the disk and the surrounding area within 5 min.

A phytic acid calcium salt hydrolysis assay was carried out according to the plate method described by Anastasio et al. (2010) [28], with some modifications. Briefly, selected strains were tested for their qualitative capability for hydrolyse phytic acid calcium salt by spot inoculation onto R2A-NaCl supplemented with dextrose (9.5 g·L⁻¹) and C₆H₁₆CaO₂₄P₆ (5 g·L⁻¹) (Sigma-Aldrich, St. Louis, MO, USA) at pH 7.2 and grown at 30 °C up to 10 days. Phytase-producing strains exhibited clear halo zones around the colony.

2.7. Carbonate Solubilisation

The calcium carbonate solubilisation was carried out according to Thakur et al. (2023) [29] with same modifications. Briefly, CaCO₃-bioforming strains were tested for their qualitative capability to solubilize carbonates by spot inoculation onto R2A-NaCl amended with dextrose (9.5 g·L⁻¹) and CaCO₃ (5 g·L⁻¹) (Sigma-Aldrich) at pH 7.2 and grown at 30 °C up to 10 days.

Carbonate-solubilising strains produce a clear halo zone around the colony in an opaque background on plates.

2.8. Micro X-Ray Diffraction and SEM Observation of Precipitated Crystals

The mineralogical composition of the dried precipitate (72 h at 60 °C) from CPB after 30 days of incubation was determined by powder X-ray microdiffraction (µ-XRD) in 28 samples.

X-ray microdiffraction (µ-XRD) data were collected using a Rigaku D-max rapid micro-diffractometer (Tokyo, Japan), operating at 40 kV and 30 mA. This instrument is equipped with a Cu-Kα source, curved-image-plate detector, flat graphite monochromator, variety of beam collimators, motorised stage, and a microscope for accurate positioning of the sample. The motorised stage allows two angular movements (rotation Ф and revolution Ω). The data were collected in reflection mode using a 0.3 mm collimator, with a collection time of 30 min. The µ-XRD data were collected as two-dimensional images and then converted into intensity-2θ profiles using the Rigaku R-Axis Display software version 1.0.8.3.

2.9. SEM Observation of Precipitated Crystals

The crystals precipitated on the bottom of the test tubes and those adhering to the tube walls were delicately detached with a small spatula, washed 3 times with ultrapure water, and dried at 60 °C for 72 h. SEM observation of precipitated crystals was carried out according to Pasquale et al. (2019) [25]. Morphological observations of air-dried precipitates formed in CBP were performed by an HR-FESEM (Zeiss Supra 40, Oberkochen, Germany) on carbon-coated samples dispersed on carbon stubs. The microscope was equipped with an energy dispersive X-ray spectrometer (EDS, Instruments INCA x-act, Abingdon, UK) for elemental analysis. Images were obtained with secondary electrons (SEs) acquired at low voltage using an in-lens detector and with a quadrant backscatter detector (QBSD). The SEM working conditions were set at an accelerating voltage of 5–15 kV, working distance of 3–8 mm, and aperture size of 30–60 µm.

3. Results and Discussion

3.1. Isolation of Culturable Heterotrophic Bacteria from C. rubrum

The concentration of the aerobic heterotrophic microflora colonising C. rubrum coenosarc samples spanned from 3 to 6·10⁶ CFU/cm² (

Table 2. Bacteriological spectrum of C. rubrum samples. L: Liguria region; T: Tuscany region; C: Campania region; CHB: Cultivable Heterotrophic Bacteria as unity-forming colonies; n.a.: not applicable; CBB: Carbonate Bioformation Broth.

C. rubrum Samples	CHB UFC/cm2	Number of Colonial Morphotypes	Strains Showing Ca++ Precipitation in CBB ≥ 67%	Carbonatogenic Strain Identification
				Code	Species (Accession Number)
L-1	150	8	0	n.a.	n.a.
L-2	18	6	2	L-21	Staphylococcus equorum (MH712951.1)
				L-22	Staphylococcus equorum (LN774671.1)
L-3	33	4	0	n.a.	n.a.
L-4	180	5	0	n.a.	n.a.
L-5	5,400,000	9	4	L-51	Staphylococcus equorum (LN774671.1)
				L-52	Staphylococcus equorum (KY940339.1)
				L-53	Staphylococcus equorum (LN774671.1)
				L-54	Staphylococcus equorum (LN774671.1)
L-6	9	2	1	L-61	Staphylococcus equorum (KY940339.1)
L-7	6	5	1	L-71	Staphylococcus equorum (LN774671.1)
L-8	18	6	2	L-81	Staphylococcus equorum (LN774671.1)
				L-82	Shewanella sp. (KC592373.1)
L-9	18	3	2	L-91	Staphylococcus xylosus (MK696228.1)
				L-92	Staphylococcus equorum (KY940339.1)
L-10	12	2	1	L-101	Staphylococcus equorum (LN774671.1)
T-11	210	5	1	T-111	Staphylococcus equorum (KY940339.1)
T-12	60	4	1	T-121	Staphylococcus equorum (KY940339.1)
T-13	360	2	0	n.a.	n.a.
T-14	600	6	1	T-141	Staphylococcus equorum (LN774671.1)
T-15	240	1	1	T-151	Staphylococcus equorum (LN774671.1)
T-16	600	7	1	T-161	Staphylococcus equorum (LN774671.1)
T-17	27,000	5	1	T-171	Staphylococcus equorum (LN774671.1)
T-18	24,000	5	1	T-181	Staphylococcus equorum (LN774671.1)
T-19	90,000	3	1	T-191	Staphylococcus equorum (LN774671.1)
C-20	360	6	1	C-201	Staphylococcus equorum (KY940339.1)
C-21	24	4	0	n.a.	n.a.
C-22	360	7	1	C-221	Staphylococcus equorum (KY940339.1)
C-23	1,200,000	4	1	C-231	Staphylococcus equorum (LN774671.1)
C-24	60	7	1	C-241	Staphylococcus equorum (KY940339.1)
C-25	120	5	0	n.a.	n.a.
C-26	150	5	0	n.a.	n.a
C-27	45,300	7	0	n.a.	n.a
C-28	210	4	0	n.a.	n.a.
C-29	3	3	0	n.a.	n.a.
C-30	9	4	0	n.a.	n.a.
C-31	60	4	0	n.a.	n.a.
C-32	10	3	0	n.a.	n.a.
C-33	15	4	1	C-331	Staphylococcus equorum (LN774671.1)
C-34	3,000,000	10	0	n.a.	n.a.
C-35	6,000,000	10	2	C-351	Staphylococcus equorum (KY940339.1)
				C-352	Staphylococcus equorum (LN774671.1)

). These results are in agreement with the findings of Pasquale et al. (2011) [18], who found a bacterial concentration of the C. rubrum coenosarc, sampled in the north, middle, and south Tyrrhenian Sea, spanning from 2 to 3.4∙10⁵ CFU/cm². Despite how the utilised substrate (R2A) allows the growth of both bacteria and fungi, no fungi were detected in any of the samples. In total, 180 strains were isolated, 151 of which were found as poorly precipitating Ca²⁺ and were not further identified.

The enormous quantitative variation in microbial populations colonising the C. rubrum coenosarc, spanning over 6 orders of magnitude, is not mirrored by a corresponding variability of the bacterial colony morphotypes recorded, with the mean being 5.1 (±2.1 sd). In Figure 2 the log₁₀ concentration of heterotrophic bacteria (CFU/cm²) versus the log₁₀ of the number of colonies morphotypes for each sample is shown. There is not a clear trend depicting the dependency of the number of morphotypes from the abundance of the microflora. The CFU/cm² values span from 3 (sample C-29) to 6·10⁶ (sample C-35), whereas the number of different colonials morphotypes span from 2 (samples L-6, L-10 and T-13) to 10 (samples C-34 and C-35).

Even the presence of strains that are high performers in Ca²⁺ removal is consistent neither with the amount of bacterial heterotrophic microflora nor with the number of different morphotypes displayed by the different samples (Table 2).

3.2. Screening of Carbonatogenic Strains

All the different colonies representative of the different morphotypes were selected from all the plates containing up to a maximum of 300 colonies. In total, 180 colonies were selected, purified on R2A plates, and further transferred on CBA plates. After incubation, stereomicroscopic observation showed that 117 strains formed crystals inside the colonies.

3.3. Carbonate Bioformation

After inoculation in tubes containing CBB and incubation at 18 and 30 °C up to 30 days, 88 out of 117 strains showed a limited capacity of Ca²⁺ depletion (≤33% g L⁻¹ Ca²⁺). All the remaining 28 strains were able to reduce at least 67% of Ca²⁺ from CBB within 30 days of incubation at 30 °C, whereas only 10 out of 28 of them were able to do the same when incubated at 18 °C. These 28 strains were further identified. Surprisingly, 26 strains were identified as S. equorum, whereas only 1 was identified as S. xylosus and 1 as Shewanella sp. The ability of those strains to utilize Ca²⁺ for the precipitation of calcium carbonate is shown in Figure 3.

The control samples, made only of sterile CBB, showed a Ca²⁺ concentration in a solution of 9.5 g·L⁻¹ (±0.1 sd) both at 18 °C and 30 °C, expressed as the mean of all incubation times. In order to check the influence of pH in carbonate precipitation, the pH value of the sterile reacting media, containing 10 g∙L⁻¹ of Ca²⁺, was increased up to 8.5. After 30 days of incubation at both temperatures, no precipitation carbonate crystals were observed.

In Figure 3 the Ca²⁺ concentrations in CBB for each tested strain are expressed as the percentage of the respective control values.

Almost all strains showed a poor ability to reduce the Ca²⁺ concentration in the liquid phase of the CBB at 3 days of incubation at 18 °C, as the values of Ca²⁺ left in the solution spanned from 72.3% of the control (strain L-2₁) to 99.1% (strain T-11₁). The mean value of all strains was 93.1% (± 6.8 sd). After 10 days of incubation, the Ca²⁺ left in the solution varied between 99.2% of the control (strain T-11₁) and 44.4% (strain L-8₂), with the mean value being 76.7% (± 15.3 sd). At 30 days of incubation, the rate of Ca²⁺ in the solution varied between 76.9% of the control value (strain L-9₁) and 15.1% (strain L-8₂). The mean value was 36.8% (± 12.6). The results related to the incubation at 30 °C showed an increased ability in MICP of all the strains tested. After 3 days of incubation, the Ca²⁺ left in the solution spanned from 94.7% (strain L-9₂) to 63.0% (strain T-18₁), whereas after 30 days the highest recorded value was 28.5% (strain L-5₃) and the lowest one was 0.1% (strain L-10₁). The control samples, made by sterile CBB, showed no significant losses of Ca²⁺ in the solution, with the overall mean value being equal to 95,0% (± 2 sd) of the starting concentration. It is worth noting that some strains express the maximum ability to remove Ca²⁺ from the solution only after 10 days at 30 °C. The strain L-9₂ (S. xylosus), for instance, when incubated at 30 °C, was a very poor performer at day 3 (94.7% of Ca²⁺ left in solution as compared with the control). This value was reduced to 89.4% at day 10 and further lowered to 2.1% at day 30. On the contrary, the strain L-8₂ was a rather good performer, with the values of Ca²⁺ left in the solution after 30 days of incubation being 15.1% and 0.3% at 18 °C and 30 °C, respectively.

In comparison with our results, the strains of S. equorum studied by Sepúlveda et al. (2021) [30] were able to precipitate, after 5 days of incubation at 30 °C, 3.5 g·L⁻¹ of CaCO₃, corresponding to 1.4 g·L⁻¹ of Ca²⁺. Those bacterial strains, therefore, precipitated only 17.95% of the Ca²⁺ initially present in the reacting media (7.8 g·L⁻¹). These data highlight the particular ability of the S. equorum strains we isolated in precipitating Ca²⁺, as our best performer was able to precipitate up to 99.9% of the Ca²⁺ (strain L-10₁) after 30 days of incubation at 30 °C. It should be also considered that the amount of precipitated calcium could be related to a number of factors, such as the composition of the culture medium and the incubation conditions of the cultures other than the specific microbial strain involved. Concerning Shewanella sp., Li et al. (2018) [31] found a marine strain of S. piezotolerans that was able to precipitate calcium carbonate as a mixture of calcite and vaterite. However, the authors did not make any attempt of measuring the quantity of calcium carbonate precipitated.

These results highlighted the importance of both time and incubation temperature to display the maximum MICP by the isolated strains.

The mean Mediterranean Sea temperatures, calculated from the data published by Manzella and Gambetta (2013) [32], span from 16.6 °C (mean of the upper 50 m in the central Mediterranean in November), to 15.1 °C (0–2 m in February), and to 23 °C (0–2 m in August) as the mean of the whole Mediterranean basin. The values of the sea temperature we quoted are the range of typical seasonal fluctuations, but it is not unusual to find higher summer temperatures in the Mediterranean surface water. Sparnocchia et al. (2006) [33] recorded a temperature of 28.6 °C in August in the Ligurian Sea, while the EU Programme Copernicus reported a temperature of 28.4 °C, in August 2024, as the highest in the Mediterranean (https://climate.copernicus.eu/c3s-seasonal-lookback-summer-2024; accessed on 20 October 2024). As a temperature of 30 °C is considered the physiological optimum for mesophilic bacteria, among which our isolated bacteria are listed, we choose it as the upper limit to check the enzyme production and MICP capabilities of our isolates. The ability of the strains we isolated to precipitate Ca²⁺ was significantly higher at a temperature representing the mesophilic optimum.

It could be argued that the ability of precipitating Ca²⁺ is not a prerogative for living and thriving in the particular environment constituted by the C. rubrum coenosarc. From an ecological standpoint it should be borne in mind that S. equorum, firstly described by Schleifer et al. (1984) [34] as the skin commensal of healthy horses, has been found in a variety of different environments: from human clinical specimens [35] to milk and cheese [36,37] and from naturally fermented sausages [38] to seafood [39].

Given that the salt tolerance of this bacterium is well known [39,40], as is its capacity of calcium removal through the MICP process [30], its presence in marine environments and its ability to precipitate Ca²⁺ are not surprising. The peculiarity of our findings lies in having isolated S. equorum as a largely predominant species that is able to effectively precipitate Ca²⁺.

The microbial community of the C. rubrum coenosarc, despite its concentration, spans from 3 to 6·10⁶ UFC/cm², and it seems quite poor in terms of the biodiversity of the culturable heterotrophic bacterial community.

The images shown in Figure 4 are consistent with the calcium spherulites found by Wu et al. (2020) [41] and Chafetz et al. (2018) [42] and show spherulite displaying an exposed core surrounded by radiating calcite crystals. EDX spot analysis on the spherulites and surrounding crystals (Figure 4D) showed only C, O, and Ca elements, which is coherent with the presence of CaCO₃ in the form of calcite (Figure 4B).

When crystals produced by Shewanella sp. are considered, a different morphology of coalescing spherulites, 2 to 6 μm in size, was observed (Figure 5). Fractured spherulites show needle-like microcrystals of calcite radiating from a centre that contains bacterial cells or empty holes similar in size and shape to the bacteria (Figure 5B). The development of spherulites with fibrillar crystals is attributed to the presence of organic molecules in their structure [43,44,45]. The occurrence of microbes and their by-products in the centre of spherulites favoured the nucleation of radially organised calcite crystals [42].

The precipitate made by Staphylococcus xylosus, when observed by SEM, did not show any crystal shape. The Ca²⁺ nature of the precipitate was inferred from the lowering of Ca²⁺ in the reacting solution.

3.3.1. Molecular Identification of Bacterial Strains

Of the 28 strains that were able to precipitate more than 66% of Ca²⁺ after 30 days of incubation at 30 °C, 26 were found to belong to the species S. equorum (Table 1). In the sample L-9 both S. equorum (L-9₁) and S. xylosus (L-9₂) were identified. S. equorum was isolated from 7 of the 10 coral colonies sampled in the Liguria region, from 8 of the 9 colonies sampled in Tuscany, and from 6 of the 16 colonies sampled in Campania. Shewanella sp. was isolated only in a sample from the Liguria region.

The isolation of S. equorum from C. rubrum samples has been reported for the first time by Pasquale et al. (2011, 2012) [18,46], who described it as the most frequent microbial species in coral samples collected from coralligenous biocenosis colonising hard substrates of the sea floor surrounding the island of Procida (Southern Italy).

To our best knowledge no other studies documenting such a widespread association between S. equorum and corals are available in the scientific literature. In 2010, Chiou et al. [47] reported the isolation, from a C. rubrum coenosarc, of bacterial species belonging to the genus Staphylococcus, Clostridium, and Legionella. The authors considered those species unusual, attributing this association to sea pollution caused by human activities. S. equorum was listed among the gorgonian-associated bacterial flora in the South China Sea [48]. S. equorum was associated with a number of different natural and artificial systems other than sea water [35,36,37,38,39], found as the causative agent of illness in marine invertebrates and fishes [49,50].

S. equorum was reported as phytate-degrading bacteria by Hill and Richardson [51]. These authors isolated S. equorum, together with Bacillus pumilus, Arthrobacter bergerei, and Pseudomonas marginalis, from a poultry-manure-enriched region in Chesapeake Bay (USA West Coast). With S. equorum being a known component of the poultry farm environment [52], the occurrence of such bacteria in a poultry-manure-polluted marine area indicates a terrestrial origin of the phytase positive S. equorum isolated by Hill and Richardson (2007) [51].

The genus Shewanella is a well-known component of coral microbiota: Sun et al. (2023) [53] and Shnit-Orland et al. (2010) [54] isolated a marine bacteria species nova named Shewanella corallii from a Red Sea coral. Despite the role of the genus Shewanella in marine environments, as a component of coral microbiota and as a promoter of precipitation of Ca²⁺ in conditions leading to iron reduction [55], to our best knowledge this bacterium was never associated with both C. rubrum microbiota and calcite precipitation.

3.3.2. Biochemical Characteristics of Carbonatogenic Isolates

De Muynck et al. (2010) [56] observed how the precipitation of calcium carbonate includes different factors, involving physical, chemical, and biological parameters. Among the biological parameters, apart from the chemical characteristics of bacterial cells making it a nucleation site for crystal nucleation [57], urea hydrolysis, induced by a series of reactions driven by urease and carbonic anhydrase, seems to play a major role [2,22].

According to Castro-Alonso et al. (2019) [22], the coordinated activity of urease and carbon anhydrase drives the cleavage of urea, the production of bicarbonate, and both ammonium and hydroxide, with the consequent alkalinisation of the microbial microenvironment facilitating the precipitation of Ca²⁺.

All S. equorum strains isolated here, as well S. xylosus and Shewanella sp., were positive both for the production of urease and carbon anhydrase in vitro at 30 °C. It seems that they potentially possess the major biochemical abilities conducive to Ca²⁺ precipitation, as they showed in vitro.

Carbonic anhydrase activity is related both to the precipitation of carbonates and to their dissolution in sea environments. Nevertheless, calcites’ reactivity to sea water chemistry is slow and poorly predictable, as it responds nonlinearly to the saturation state, implying the presence of multiple dissolution mechanisms. The geometry of the dissolving mineral surface also plays a role in the dissolving processes [58]. In addition, the calcite dissolution rates in freshwater under circumneutral pH conditions appear largely independent from the solution pH [59]. On the contrary, ammonia production from microbial degradation of organic matter makes the precipitation of calcite easier and faster, also thanks to the activity of enzymes as carbon anhydrase [60].

3.3.3. Carbonate and Calcium Phytate Solubilisation

Carbonate microbial dissolution is a well-known phenomenon included in the more general mechanism of mineral weathering [61]. According to these authors, the carbonate microbial weathering is a major form of interaction among rocks, soils, atmosphere, and organisms. Bennett et al. (2000) [62] described both carbonatic rock dissolution and calcite precipitation mediated by microorganisms in the same aquifer contaminated with crude oil, highlighting that the same microflora can act as carbonate makers and as carbonate dissolvers. Such processes have global impacts over the carbon biogeochemical cycle [63]. All our isolated carbonatogenic strains were also able to dissolve crystals of calcium carbonate.

All our carbonatogenic identified strains were able to hydrolyse the phytic acid calcium salt and then were potentially able to induce CPCs’ precipitation also through such a mechanism.

4. Conclusions

The C. rubrum coenosarc microflora of the samples coming from three Italian regions (Liguria, Toscana, and Campania) seems to be restricted to a few colony morphotypes, even if it can reach a six-digit concentration, expressed as CFU/cm². This apparent poorness of microbial biodiversity in this specific environment seems to be mirrored by a highly carbonatogenic bacterial flora, which is restricted to S. equorum, as the main actor in this process in our experiment conditions, as well as Shewanella sp. and S. xylosus.

Our findings add new insights on the multifaceted ecological role of S. equorum. The carbonatogenic ability of this bacterium was previously investigated in a terrestrial environment, whereas we found strains of S. equorum colonising C. rubrum that are also capable of Ca²⁺ precipitation, thanks to an enzymatic array comprising urease, carbon anhydrase, and phytase.

In our study we highlighted the presence of this bacterium as part of the natural microflora associated with the C. rubrum coenosarc.

The remarkable biochemical characteristics of our isolates deserve further study aimed at a better understanding of the overall ecological role played by S. equorum as a component of the C. rubrum coenosarc-associated microflora.