Introduction

With regard to α-amylase families of glycoside hydrolases (GHs) and the Carbohydrate-Active enZymes (CAZy) database, the family GH13 represents the main, in terms of chronology, the original α-amylase family, whereas the family GH57 is the second and smaller α-amylase family (Janecek et al 2014; Lombard et al. 2014). Established in 1991 (Henrissat 1991) and 1996 (Henrissat and Bairoch 1996), respectively, the former contains almost 77 thousand sequences from all the three domains of life—Bacteria, Archaea and Eucarya, while the latter consists more than 2300 sequences only lacking, in fact, any relevant member from eukaryotes (Lombard et al. 2014).

The main α-amylase family GH13 with about 30 different enzyme specificities belongs to the largest GH families within CAZy (Lombard et al. 2014). In addition to enzymes from hydrolases, transferases and isomerases, it contains also some non-enzymatic proteins involved in amino acid transport (Janecek et al. 1997; Gabrisko and Janecek 2009). The members of the α-amylase family GH13 (Matsuura et al. 1984; Svensson 1994; Kuriki and Imanaka 1999; MacGregor et al. 2001; Janecek 2002; van der Maarel et al. 2002; Janecek et al. 2014) employ a retaining reaction mechanism, share 4–7 conserved sequence regions (CSRs) and adopt a TIM-barrel domain with the GH13 catalytic machinery, i.e. the Asp206, Glu230 and Asp297 (Aspergillus oryzae α-amylase numbering) acting as catalytic nucleophile (CSR-II, strand β4), catalytic proton donor (CSR-III, strand β5) and transition-state stabiliser (CSR-IV, strand β7). Currently, the family is divided into 42 GH13 subfamilies (Stam et al. 2006), for which unique features in their amino acid sequences can be identified to be responsible for their individual specificities (Oslancova and Janecek 2002; Majzlova et al. 2013; Janecek et al. 2015; Kuchtova and Janecek 2016). Recently, a remarkable group of bacterial amylolytic enzymes has been described to possess aberrant catalytic machinery (Sarian et al 2017). The family GH13 forms, together with families GH70 and GH77, the clan GH-H; a remote homology of clan GH-H to family GH31, containing functionally related α-glucosidases, being also already suggested (Janecek et al. 2007).

In comparison with GH13, the second α-amylase family GH57 is a smaller family with only about 2000 members and less than 10 enzyme specificities (Lombard et al. 2014; Martinovicova and Janecek 2018). Although it also employs a retaining reaction mechanism (Palomo et al. 2011), it exhibits its own 5 CSRs and catalytic machinery different from GH13 within an incomplete TIM-barrel catalytic domain fold (Palomo et al. 2011; Imamura et al. 2003; Zona et al. 2004; Janecek and Blesak 2011; Blesak and Janecek 2012, 2013; Martinovicova and Janecek 2018). The individual enzyme specificities also possess unique features within their sequence logos that can be used as sequence-specificity fingerprints (Blesak and Janecek 2012; Martinovicova and Janecek 2018). A previous in silico analysis has suggested (Janecek and Kuchtova 2012) that the third α-amylase family GH119 should share with the family GH57 the CSRs, catalytic machinery and fold of the catalytic domain.

The family GH57 was established in 1996 (Henrissat and Bairoch 1996) based on the existence of two “amylases” originating from thermophilic bacterium Dictyoglomus thermophilum (Fukusumi et al. 1988) and hyperthermophilic archaeon Pyrococcus furiosus (Laderman et al. 1993b). They, although considered to be real α-amylases, were both sequentially different from sequences of the main α-amylase family GH13 members (Janecek 1998). Actually, they have been re-classified (Laderman et al. 1993a; Nakajima et al. 2004; Kaila et al. 2019) and thus become well-known as 4-α-glucanotransferases (4AGTs). Currently, the α-amylase family GH57 contains, in addition to α-amylase (Janecek et al. 2014), the specificities of 4AGT, amylopullulanase, α-glucan branching enzyme (GBE), dual-specificity of amylopullulanase–cyclomaltodextrinase, α-galactosidase, non-specified amylase, and maltogenic amylases (Comfort et al. 2008; Blesak and Janecek 2013; Jeon et al. 2014; Jung et al. 2014; Park et al. 2014; Martinovicova and Janecek 2018); the last one being still included only among the so-called non-classified sequences of the CAZy database (Lombard et al. 2014). Interestingly, the α-amylase specificity, best represented by the enzyme from Methanocaldococcus jannaschii (Bult et al. 1996), may rather be an amylopullulanase (Janecek et al. 2014), since that α-amylase was shown to be able to degrade not only starch but also a pullulan (Kim et al. 2001). With regard to solved tertiary structures, the first one determined by the X-ray crystallography was that of 4AGT from Thermococcus litoralis (Imamura et al. 2003), followed by the structures of the GBE from Thermotoga maritima (Dickmanns et al. 2006)—a bifunctional enzyme possessing also the α-amylase activity (Blesak and Janecek 2012; Zhang et al. 2019)—and GBEs from Thermus thermophilus (Palomo et al. 2011), Thermococcus kodakarensis (Santos et al. 2011) and Pyrococcus horikoshii (Na et al. 2017) as well as the maltogenic amylase from Pyrococcus sp. ST04 (Park et al. 2014). Of more than 2500 sequences available in the family GH57 (Lombard et al. 2014), only 27 members have already been characterised as real enzymes (Martinovicova and Janecek 2018).

As mentioned above, the catalytic domain consists of the so-called incomplete TIM-barrel, which is a (β/α)7-barrel domain, succeeded by a bundle of usually 3–4 α-helixes (Imamura et al. 2003; Dickmanns et al. 2006; Palomo et al. 2011; Santos et al. 2011; Park et al. 2014; Na et al. 2017), although some GH57 enzymes possess additional domains (Blesak and Janecek 2012, 2013). The catalytic machinery comprises the catalytic nucleophile—glutamic acid (Glu123 in the 4AGT from T. litoralis) and the catalytic proton donor—aspartic acid (Asp214), the former being positioned on the strand β4 (CSR-3) whereas the latter being placed on the strand β7 of the catalytic incomplete TIM-barrel (Imamura et al. 2003; Zona et al. 2004). The 5 CSRs were defined in 2004 (Zona et al. 2004), the first four of which (CSR-1–CSR-4) are located within the (β/α)7-barrel on strands β1, β3, β4 and β7, while the last fifth region CSR-5 is positioned within the α-helical bundle that, in fact, forms with the barrel the entire catalytic area of the family GH57 members (Blesak and Janecek 2012).

In the α-amylase family GH57, there has also been revealed a group of hypothetical proteins with sequences very closely related to those of the specificity of α-amylase, the so-called α-amylase-like proteins (Janecek and Blesak 2011). They may represent potential non-enzymatic members, somehow similar to those present in the main α-amylase family GH13, known as the rBAT and 4F2hc-antigen transport proteins that resemble the α-glucosidases from the oligo-1,6-glucosidase subfamily in both sequence and structure (Janecek et al. 1997; Fort et al. 2007; Gabrisko and Janecek 2009; Chillaron et al. 2010; Lee et al. 2019; Yan et al. 2019). In spite of their function obviously unrelated to amylolysis, the two groups have been assigned even their GH13 subfamily numbers, i.e. the GH13_34 for 4F2hc antigens and GH13_35 from rBAT proteins (Stam et al. 2006; Lombard et al. 2014; Janecek and Gabrisko 2016). For the family GH57 α-amylase-like proteins, a substitution in one or both catalytic residues is typical, although it has still not been revealed whether or not they exhibit any enzymatic and/or a protein function (Janecek and Blesak 2011; Martinovicova and Janecek 2018). From a taxonomic point of view, they originate mainly from the two bacterial genera of Bacteroides and Prevotella (Janecek and Blesak 2011).

The main goal of the present bioinformatics study was to deliver another story of protein homologues closely related to further two GH57 specificities—GBE and 4AGT. These so-called GBE-like and 4AGT-like proteins lack, similar to α-amylase-like proteins described previously (Janecek and Blesak 2011), one or both catalytic residues. Concerning the sequence, however, they are most similar to their “template” enzyme specificities. This in silico analysis could contribute to a better understanding and a more detailed knowledge of these “like” groups of proteins present in the α-amylase family GH57. It may thus indicate the way leading to predicting their possible functions and, eventually, to shortening the time necessary for experimental revealing their activities.

Materials and methods

Sequence collection and comparison

Sequences were extracted from our recent in silico analysis of the entire family GH57 comparing in total 1602 GH57 members (Martinovicova and Janecek 2018). From that pool of GH57 sequences, 55 GBE-like and 61 4AGT-like proteins that lacked one or both catalytic residues were collected; both groups of hypothetical proteins being completed by four and six experimentally characterised GBEs and 4AGTs, respectively (Fukusumi et al. 1988; Laderman et al. 1993a, b; Jeon et al. 1997; Tachibana et al. 1997; Imamura et al. 2003; Nakajima et al. 2004; Ballschmiter et al. 2006; Dickmanns et al. 2006; Murakami et al. 2006; Labes and Schonheit 2007; Palomo et al. 2011; Santos et al. 2011; Park et al. 2014; Paul et al. 2015; Na et al. 2017; Kaila and Guptasarma 2019; Kaila et al. 2019; Pang et al. 2019; Zhang et al. 2019). In order to make a relevant conclusion, 21 biochemically characterised enzymes representing the remaining specificities from the family GH57 were added to give a preliminary set of 147 studied sequences. For further detailed analysis, the numbers of GBE-like and 4AGT-like proteins were reduced from 55 to 13 and 61 to 18, respectively, in order to take into account only the sequences originated from unique organisms. This means that the identical GBE-like and 4AGT-like proteins from Treponema pallidum were eliminated and the final set consisted of 62 sequences (Fig. S1).

All studied sequences were retrieved from GenBank (Benson et al. 2018) and/or UniProt (The UniProt Consortium 2017) sequence databases. The sequence alignment of all 147 GH57 sequences was performed using the program Clustal-Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/; Sievers et al. 2011) with default parameters. The alignment spanned the substantial part of the family GH57 catalytic area including both the incomplete TIM-barrel and the bundle of 3–4 α-helices, exactly from the beginning of the CSR-1 to the end of CSR-5. The alignment was improved with some manual tuning in order to maximize similarities mainly with regard to correct correspondences within the five established CSRs (Zona et al. 2004).

Evolutionary analysis

The evolutionary tree was calculated as a maximum-likelihood tree (Jones et al. 1992) using the MEGA software (https://www.megasoftware.net/; Kumar et al. 2018) applying default programme parameters and bootstrapping procedure (the number of bootstrap trials used was 500). The tree was displayed with the program iTOL (https://itol.embl.de/; Letunic and Bork 2011).

Structure comparison

Three-dimensional structures were retrieved from the Protein Data Bank (PDB; Rose et al. 2015) for representatives of the family GH57, i.e. the GBE from T. kodakarensis (PDB code: 3N92; Santos et al. 2011) and the 4AGT from T. litoralis (PDB code: 1K1Y; Imamura et al. 2003). For making the three-dimensional models of GBE-like and 4AGT-like proteins, sequences of Planctomyces sp. SH-PL62 (UniProt Acc. No.: A0A142YKP4) and Treponema pedis (UniProt Acc. No.: S6A1V1) were selected as representatives of the former and latter group, respectively. The structural models were created with the Phyre2 server (https://www.sbg.bio.ic.ac.uk/phyre2/; Kelley and Sternberg 2009). The structures were superimposed using the program MultiProt (https://bioinfo3d.cs.tau.ac.il/MultiProt/; Shatsky et al. 2004) and the overlaps were displayed with the program WebLabViewerLite (Molecular Simulations, Inc.).

Results and discussion

Sequence comparison

The present in silico analysis represents a follow-up study of the original report on the newly identified group of the so-called α-amylase-like proteins from the α-amylase family GH57 that lack one or both catalytic residues although overall they exhibit an unambiguously close sequence similarity to its enzymatic counterpart (Janecek and Blesak 2011). In total, this study delivers 13 and 18 such “like” proteins (Fig. S1) from unique sources found in the evolutionary tree of the entire family GH57 (Martinovicova and Janecek 2018) in a close proximity to specificities of GBEs and 4AGTs, respectively. Interestingly, the taxonomical spectrum of source organisms of these “like” proteins is rather limited to a few genera in both cases, the genus Treponema being found dominant (Fig. S1). The alignment of all 62 selected sequences, i.e. 13 GBE-like and 18 4AGT-like proteins as well as 4 and 6 experimentally characterised GBEs and 4AGTs, respectively, completed with 21 remaining characterized GH57, has confirmed the extremely high sequence divergence typical for the family GH57 (Zona et al. 2004; Martinovicova and Janecek 2018), since the sequence section spanning the segment from the beginning of the CSR-1 to the end of CSR-5 reached the consensus length of 857 positions (Fig. S1). It is of note that it covers the so-called GH57 catalytic area (Janecek and Blesak 2011; Blesak and Janecek 2012, 2013) including both the incomplete TIM-barrel and the bundle of 3–4 α-helices, counting 458–463 residues for GBEs and 348–356 residues for 4AGTs. In order to focus on the so-called family GH57 sequence fingerprints (Blesak and Janecek 2013) bearing also the catalytic machinery, Fig. 1 shows the five CSRs well-established in the family (Zona et al. 2004) of all 62 studied sequences. It is evident that in all “like” proteins, there are substitutions present in either or in both catalytic residues.

Fig. 1
figure 1

Conserved sequence regions of family GH57 members focused on GBE-like and 4AGT-like proteins. The individual groups are distinguished from each other by different colours. The label of the protein source consists of: (i) the abbreviated name of the enzyme/protein; (ii) GenBank accession number; (iii) letter “A” or “B” for archeons and bacteria, respectively; and (iv) the name of the organism. The GH57 enzyme specificities, in addition to GBE (α-glucan branching enzyme) and 4AGT (4-α-glucanotransferase), are abbreviated as follows: AAMY α-amylase, AAMY like α-amylase-like protein, AGAL α-galactosidase, NSA non-specified amylase, MGA maltogenic amylase, APU amylopullulanase, APU–CMD amylopullulanase–cyclomaltodextrinase. The catalytic residues, i.e. a glutamic acid in CSR-3 (the catalytic nucleophile) and an aspartic acid in CSR-4 (the proton donor) are boxed. Note that the GBE-like and 4AGT-like proteins lack either one or both catalytic residues. Colour code for the individual residues: Trp, Phe, Tyr: yellow; Asp: blue, Glu: red; Val, Leu, Ile: green; Arg, Lys: cyan; His: brown; Cys: magenta; Gly, Pro: black

Full size image

With regard to GBE-like proteins, the catalytic nucleophile, the glutamic acid in CSR-3 is replaced by serine, tyrosine, phenylalanine, alanine, asparagine and even glycine, whereas the proton donor, the aspartic acid in CSR-4 is exchanged mostly by proline although also serine and asparagines are found there (Figs. 1 and S1). As far as the 4AGT-like proteins are concerned, the two catalytic residues are substituted predominantly by alanine and proline, respectively, but there are also arginine, asparagine, phenylalanine, tyrosine, histidine serine and glycine in the place of the catalytic nucleophile and phenylalanine, histidine, glutamic acid, asparagine, valine, serine and alanine replacing the proton donor (Figs. 1 and S1).

In addition to missing catalytic residues in CSR-3 and CSR-4, the former region seems to have been well conserved but the latter region has obviously undergone a more dramatic change in both “like” proteins. This is especially clear at the end of the CSR-4, which in both cases was defined as the region containing residues of sequence fingerprints of a given GH57 enzyme specificity (Janecek and Blesak 2011; Blesak and Janecek 2012, 2013). Thus in GBE-like proteins, the His-Trp signature unique for GBEs (Blesak and Janecek 2012) is no more seen; the two residues being replaced by various residues mostly by Val-Ala and/or combination of lysine and arginine (Fig. 1). The cysteine residue succeeding the catalytic nucleophile in CSR-3, typical for GBEs (Blesak and Janecek 2012; Martinovicova and Janecek 2018; Zhang et al. 2019) is also absent in GBE-like proteins (Fig. 1a). Concerning the 4AGT-like proteins, the tryptophan at the end of CSR-4 conserved in 4AGTs invariantly (Blesak and Janecek 2012) is substituted by different residues without any special preference; the only exception is represented by the “like” protein originating from Treponema azonutricium (Fig. 1).

Obviously, the protein chain segment spanning the two regions bearing the catalytic machinery, i.e. from CSR-3 to CSR-4, is longer for GBE-like proteins in comparison with 4AGT-like proteins (Fig. S1). This feature is, however, consistent with the known domain arrangement and overall sequence-structural organization within this part of both parent GBE and 4AGT enzymes (Imamura et al. 2003; Dickmanns et al. 2006; Palomo et al. 2011; Santos et al. 2011; Blesak and Janecek 2012; Na et al. 2017).

Evolutionary relationships

The evolutionary relationships of both the GBE-like and 4AGT-like proteins within the family GH57 are depicted in the evolutionary tree (Fig. 2). In the most recent and complete in silico analysis of this family comparing 1602 sequences (Martinovicova and Janecek 2018), four groups of the “like” proteins were identified in total—in addition to GBE-like and 4AGT-like proteins described in the present work, there were also the α-amylase-like and amylopullulanase–cyclomaltodextrinase-like proteins. Whereas the former group has already been well-known from previous studies (Janecek and Blesak 2011; Blesak and Janecek 2012, 2013), the latter one consisted of 5 proteins only (Martinovicova and Janecek 2018). Therefore, the main emphasis has been focused on the groups of GBE-like (55 members) and 4AGT-like (61 members) proteins.

Fig. 2
figure 2

Evolutionary tree of family GH57 members focused on GBE-like and 4AGT-like proteins. The tree is based on the alignment of sequences spanning the segments from beginning of CSR-1 to the end of CSR-5 (as presented in Fig. S1). The individual groups are distinguished from each other by different colours. The label of the protein source consists of: (i) the abbreviated name of the enzyme/protein; (ii) GenBank accession number; (iii) letter “A” or “B” for archeons and bacteria, respectively; and (iv) the name of the organism. The GH57 enzyme specificities, in addition to GBE (α-glucan branching enzyme) and 4AGT (4-α-glucanotransferase), are abbreviated as follows: AAMY α-amylase, AAMY like α-amylase-like protein, AGAL α-galactosidase, NSA non-specified amylase, MGA maltogenic amylase, APU amylopullulanase, APU–CMD amylopullulanase–cyclomaltodextrinase

Full size image

Overall view of the evolutionary tree (Fig. 2) clearly indicates the relatedness of selected GBE-like and 4AGT-like proteins to their enzymatic counterparts, i.e. GBEs and 4AGTs. Obviously, the two major subgroups of each of the two “like” protein groups, originating solely from the genus Treponema (for GBE-like proteins) or mostly from Treponema accompanied by a few sequences from Spirochaeta (for 4AGT-like proteins), form their own taxonomic clusters; the remaining representatives of each “like” group being placed a bit separately but still together with real GBEs and real 4AGTs (Fig. 2).

It is worth mentioning that all identified members of GBE-like and 4AGT-like proteins are of bacterial origin only, whereas GBEs and 4AGTs have been found in both Bacteria and Archaea (Martinovicova and Janecek 2018). Even if only the experimentally characterized GBEs and 4AGTs are considered, they originate from prokaryotes as follows (Fig. 2): (i) GBEs—two from bacteria (T. maritima and T. thermophilus) and two from archaeons (Pyrococcus horikoshi and T. kodakarensis); and (ii) 4AGTs—one from bacteria (D. thermophilum) and four from archaeons (Archaeoglobus fulgidus, P. furiosus, T. kodakarensis, T. litoralis and Thermococcus onnurineus).

With regard to rather limited spectrum of source organisms of these GBE-like and 4AGT-like proteins (Fig. S1), currently, it is not possible to explain it unambiguously because none of them has been cloned, expressed, and biochemically characterized. It is thus not possible to have a clue concerning, e.g., their properties. The taxonomic spectrum limited to a few organisms—mostly Treponema, Spirochaeta and Sphaerochaeta (all spirochetes)—seems thus to be a puzzle, too. It might be of interest that in the related family GH77 amylomaltases, those originated from borreliae have been revealed as possessing some unique amino acid substitutions in a very specific positions that are otherwise well-conserved in amylomaltases from non-borreliae sources (Godany et al. 2008; Kuchtova and Janecek 2015). In addition, for example, for the specificity of α-amylase, their “like” protein counterparts originate from both Archaea and Bacteria (Janecek and Blesak 2011), whereas, for the GBEs and 4AGTs, their “like” protein homologues have been found only in Bacteria (Fig. 2). Moreover, in the main α-amylase family GH13 (Janecek et al. 2014), which is also a polyspecific GH family (Lombard et al. 2014), there are not known such “like” proteins that would exist as counterparts to various enzyme specificities, like those recognized in the family GH57.

Structure comparison

The structural analysis has been focused on the GBE-like protein from Planctomyces sp. SH-PL62 (UniProt Acc. No.: A0A142YKP4) and the 4AGT-like protein from T. pedis (UniProt Acc. No.: S6A1V1). Despite the fact that they lack the catalytic machinery of the α-amylase family GH57 (Fig. 1), the models of their three-dimensional structures have been obtained unambiguously according to respective templates—the experimental structures of GBEs (Dickmanns et al. 2006; Palomo et al. 2011; Santos et al. 2011; Na et al. 2017) and 4AGT (Imamura et al. 2003) with adequate (no less than 95%) alignment coverage, respecting also the presence of the additional C-terminal β-stranded domain of T. litoralis 4AGT (Imamura et al. 2003). The details from the structural comparison are illustrated in Fig. 3. It is evident that each pair of the GH57 enzyme and its “like” counterpart shares the main structural features. This phenomenon was first observed in the family GH57 for the specificity of α-amylase and its α-amylase-like homologues (Janecek and Blesak 2011) and may be expected to be identified also for other, i.e. remaining enzyme specificities from the family (Martinovicova and Janecek 2018).

Fig. 3
figure 3

Structure comparison of family GH57 α-glucan branching enzyme and 4-α-glucanotransferase with their GBE-like and 4AGT-like proteins. aT. kodakarensis α-glucan branching enzyme (PDB code: 3N92); bPlanctomyces sp. SH-PL62 GBE-like protein (model); c their superimposition (444 Cα atoms; RMSD 0.00 Å; residues 4–453 aligned, i.e. 98% coverage, with 24% identity); d a close-up focused on catalytic residues of the α-glucan branching enzyme (red) and their eventual correspondences in the GBE-like protein (magenta); eThermococcus litoralis 4-α-glucanotransferase (PDB code: 1K1Y); fTreponema pedis 4AGT-like protein (model); (g) their superimposition (551 Cα atoms; RMSD 0.00 Å; residues 7–605 aligned, i.e. 95% coverage, with 21% identity); h a close-up focused on catalytic residues of the 4-α-glucanotransferase (blue) and their eventual correspondences in the 4AGT-like protein (cyan) with bound acarbose occupying the subsites from − 1 to + 3

Full size image

Based on the structural superimposition (Fig. 3), it is clear that both studied GBE-like and 4AGT-like proteins do not possess the catalytic residues confirmed for the family GH57. While the real GBE from T. kodakarensis (Santos et al. 2011) and 4AGT from T. litoralis (Imamura et al. 2003) have their catalytic machineries formed by the pairs of Glu123/Asp214 and Glu183/Asp354, respectively, at corresponding positions there are Phe163/Ser281 and Tyr127/Ser202 found in the structural models of GBE-like protein from Planctomyces sp. SH-PL62 (Fig. 3d) and the 4AGT-like protein from T. pedis (Fig. 3h), respectively. This observation can be extended to all GBE-like and 4AGT-like proteins of this study, as documented by their sequence alignment (Figs. 1 and S1). Of course, some of the “like” homologues may contain either the catalytic nucleophile or the proton donor, but not both residues simultaneously, i.e. similar to the above-mentioned α-amylase-like homologues (Janecek and Blesak 2011).

Conclusions

The present in silico study delivers two novel groups of hypothetical proteins, the members of the α-amylase family GH57, which are closely related to GBEs and 4AGTs enzyme specificities. Since the putative proteins lack either one or both residues of the family GH57 catalytic machinery, they are suggested to define the so-called GBE-like and 4AGT-like proteins, similar to the α-amylase-like homologues established previously. Although the exact function of these “like” proteins has been not known as yet, the existence of members of GH families with incomplete catalytic machinery is not exceptional. For example, even in the main α-amylase family GH13, there is a well-known group of animal heavy-chains of heteromeric amino acid transport proteins rBAT and 4F2hc antigens lacking the catalytic machinery and amylolytic activity. In each case of the three already revealed family GH57 “like” proteins—α-amylase-like homologues (identified previously) as well as GBE-like and 4AGT-like (the present study)—the experimental evidence concerning their eventual enzyme specificity or at least their exact protein activity would be of a special interest. The presented in silico analysis could accelerate the efforts focused on experimental studies of these unique proteins.