*2.6. 3D Modeling of CamPhoD*

A theoretical model for the PhoD-like phosphodiesterase/phosphatase from the marine bacterium *C. amphilecti* KMM 296 (GenBank ID: WP\_043333989) was generated using structural bioinformatics methods (Figure 8A–C). Among the 3D structures currently available in the protein database (PDB), the low identical alkaline phosphatase D from *B. subtilis* (PDB ID: 2YEQ), which has a new architecture of the phosphatase active site based on Fe3<sup>+</sup> and two Ca2<sup>+</sup> ions, is apparently a single homologous template for modeling CamPhoD, which is a new member of the PhoD family inherent in the marine bacteria (data not shown). The amino acid sequences of CamPhoD and the template possess 20.5% identity and 38% similarity. (Figure 8A). The aa residues of the CamPhoD binding Ca<sup>2</sup>+/Co2<sup>+</sup> atoms in the active center are highly conserved. However, the bonds of the iron atom in the CamPhoD active center differ from the template bonds in replacing Cys 124 with Gly 117 (Figure 8A). In comparison with the template, the phosphate molecule in the modeled CamPhoD active site interacts with Tyr 158, Asp 221, and with one Fe3<sup>+</sup> atom, two Ca2<sup>+</sup> atoms, and three water molecules (Figure 8C).

A similar interaction with phosphate is observed when two Ca2<sup>+</sup> atoms are replaced with Co2<sup>+</sup> (Figure 9). The superimposition of the CamPhoD model and template showed that the root mean square deviation (RMSD) for 473 Cα atoms is 1 Å. The structural differences are in the structure of some of the loops on the outer surface of the protein (Figure 8C). In the CamPhoD structure, there is no α-helix at the C-terminus of the molecule, which is presented in the template, which possibly regulates the availability of the active center to accept the substrate (Figure 8A,C).

The analysis of contacts in the CamPhoD complexes with the reaction product PO4 <sup>3</sup><sup>−</sup> showed that the enzyme forms fewer contacts with the product than the template due to the shortened C-terminal region (Figure 9). The model of the CamPhoD complex with the substrate molecule has made it possible to determine the amino acid residues of the enzyme associated with the substrate binding (Figure 10).

**Figure 8.** (**A**) Modeling of the CamPhoD 3D structure. An alignment of the amino acid sequences of the alkaline phosphatase/phosphodiesterase CamPhoD from the marine bacterium *C. amphilecti* KMM 296 (GenBank ID: WP\_043333989) and alkaline phosphatase (phosphodiesterase) D from *B. subtilis* (PDB ID: 2YEQ). The amino acid sequences identity and similarity (color boxed) and the secondary structure of the template are highlighted. Note: α-helixes = red sticks; β-structure = blue arrows; the binding of amino acid (aa) residues (Ca2+/Co2+) = \*; the binding of conserved aa residues (Fe3<sup>+</sup>) = o; and the residue Cys 124 of the template = •. (**B**) The 3D structure model of CamPhoD with the reaction product Pi and metal ions in the active center (the protein structure is a ribbon diagram, Pi is in stick form, and Ca2<sup>+</sup> is shown as spheres). (**C**) The 3D superimposition of the CamPhoD model (orange) and the template (PDB ID: 2YEQ) (shown in pink, with the blue C-terminal part).

## *2.7. E*ff*ect of CamPhoD on Bacterial Biofilms*

In order to study the effect of alkaline phosphatase/phosphodiesterase CamPhoD on the inhibition of biofilm formation or on their destruction, bacterial biofilms of both individual and mixed species were grown. The study found that CamPhoD (0.1 U/mg) had a slight inhibitory effect on the biofilm formation of three species of *Bacillus*, namely *B. licheniformis*, *B. aegricola*, and *B. berkelogi* (18–32%), and dispersed the already formed biofilms of these species by 8–15% (Table 4). At the same time, CamPhoD did not inhibit the formation of biofilms in *B. subtilis* and *Pseudomonas aeruginosa* and did not degrade them.

**Figure 9.** A 2D diagram of the contacts of the CamPhoD active center (**A**) and the template (PDB ID: 2YEQ) (**B**).

**Figure 10.** A model of the CamPhoD complex with the substrate 5 -pNP-TMP (**A**)**,** and a 2D diagram of the contacts of 5 -pNP-TMP in the CamPhoD active center (**B**).

**Table 4.** Effect of the alkaline phosphatase/phosphodiesterase CamPhoD from *Cobetia amphilecti* KMM 296 on the bacterial biofilms.


\* K-control strains grown without treatment with the enzyme.

Under natural conditions, biofilms are most often formed by not just one but by several types of bacteria [56]. In view of this, the study of such mixed biofilms is of great fundamental and practical importance. We investigated the formation of biofilms by mono and mixed cultures of *Yersinia pseudotuberculosis* and *Salmonella enteritidis*, as well as the effect of CamPhoD during the 3 day incubation with the enzyme (Table 4). The destruction of mature biofilms by the studied enzyme ranged from 11% to 24% depending on the strain. For comparison, DNase I degraded about 30% of the biofilm formed by *Y. pseudotuberculosis* and more than half of the *B. subtilis* biofilm [57].

The alkaline phosphatase CmAP from *C. amphilecti* KMM 296 was also shown to effectively inhibit the growth of the new biofilms and degradation of the mature biofilms of *S. enteritidis,* as well as *P. aeruginosa* and *B. subtilis* [7], in contrast to CamPhoD. The antibacterial activity of the alkaline phosphatase against *P. aeruginosa* has been found in *E. coli*, which is known as a causative agent of diarrhea [58]. The effect of alkaline phosphatases on pathogens has been studied by exploring the gut microbiota modulation ability in the alkaline phosphatase of the intestinal PhoA enzymes, in which the level of decrease or dysfunction is associated with intestinal inflammation, dysbiosis, bacterial translocation, and subsequently systemic inflammation [33–35].

The presence of extracellular alkaline phosphatases of the structural families PhoA (GenBank ID: WP\_084589490.1) and PhoD (GenBank ID: WP\_043333989.1) in the marine gamma-proteobacterium *C. amphilecti* KMM 296 may indicate either their distinct or cooperative functions for the hydrolysis of various phosphorus-containing organic molecules, depending on the environmental conditions and cell lifestyle [7,36]. The analogous PhoD enzyme from *B. subtilis* is thought to target specific phosphate-containing molecules, such as teichoic acids linked to the wall peptidoglycan via phosphodiester bonds. It is possible that other PhoD family members also have specific biological roles rather than operating as general phosphatases, such as members of the PhoA and PhoX families [15]. In spite of the absence of both CamPhoD and CmAP ability to cleave an important signaling molecule c-di-GMP, they may provide a level of Pi, acting on other extracellular phosphate-containing substrates and playing a crucial role in the bacterial behavior. For example, the phosphorylated lipid A or the linear intermediate product of the c-di-GMP hydrolysis, pGpG, have also been recently found as substrates for alkaline phosphatases and phosphodiesterases and as participants in cell signaling [32,59]. The mechanism of the putative participation of any alkaline phosphatase from *C. amphilecti* KMM 296 in the bacterial cell signaling has yet to be clarified, including regulation of biofilms or the species content in the bacterial consortium, with which the marine bacterium has to coexist in the mollusk digestive system.
