Comparison of Candida Albicans Fatty Acid Amide Hydrolase Structure with Homologous Amidase Signature Family Enzymes

Fatty acid amide hydrolase (FAAH) is a well-characterized member of the amidase signature (AS) family of serine hydrolases. The membrane-bound FAAH protein is responsible for the catabolism of neuromodulatory fatty acid amides, including anandamide and oleamide, that regulate a wide range of mammalian behaviors, including pain perception, inflammation, sleep, and cognitive/emotional state. To date, limited crystal structures of FAAH and non-mammalian AS family proteins have been determined and used for structure-based inhibitor design. In order to provide broader structural information, the crystal structure of FAAH from the pathogenic fungus Candida albicans was determined at a resolution of 2.2 Å. A structural comparison with a brown rat Rattus norvegicus FAAH as well as with other bacterial AS family members, MAE2 and PAM, showed overall similarities but there were several discriminative regions found: the transmembrane domain and the hydrophobic cap of the brown rat FAAH were completely absent in the fungal FAAH structure. Along with these results, a phylogenetic analysis of 19 species within the AS family showed that fungal FAAHs diverged from a common ancestor before the separation of eukarya and prokarya. Taken together, this study provides insights into developing more potent inhibitors of FAAH as well as expanding our knowledge of the relationships between AS family members.


Introduction
Fatty acid amide hydrolase (FAAH) is a well-characterized member of the amidase signature (AS) family containing a highly conserved sequence called the amidase consensus sequence, which is rich in serine and glycine residues [1]. There are five members of the AS family as peptide amidase (PAM), malonamidase E2 (MAE2), subunit A of Glu-tRNA Gln amidotransferase, 6-aminohexanoate-cyclic-dimer hydrolase, and FAAH [2][3][4][5][6]. Members of the AS family include more than 200 proteins from over 90 different organisms from bacteria to humans and are involved in a variety of important biological functions. This family of enzymes is evolutionarily distinct but has diverged to develop a wide range of substrate specificities.
FAAH is a membrane-bound enzyme expressed in the brain, small intestine, pancreas, and muscle (skeletal and cardiac) that is responsible for the catabolism of neuro-modulatory fatty acid amide molecules including anandamide (AEA) and oleamide [7]. These signaling molecules regulate a wide range of mammalian behaviors including appetite, motility, sleep, and cognitive/emotional state, in addition to being involved in pain perception and inflammatory responses. Catabolism of AEA by FAAH results in the production of ethanolamine and arachidonic acid (AA) [8]. The endogenous cannabinoid (i.e.; endocannabinoid) system is comprised of two cannabinoid receptors (CB1 and CB2) and the endocannabinoids, AEA and 2-arachidonylglycerol (2-AG) [9]. Other compounds have been shown to bind to CB1 and CB2 but these may have limited effects in organisms.
There is currently a great deal of interest in developing FAAH inhibitors to disrupt endocannabinoid signaling [10,11]. Since FAAH was identified as a drug target, numerous potent and selective FAAH inhibitors (e.g., URB, PF, OL, BIA, macamides, etc.) have been synthesized and reported [12][13][14][15][16][17][18][19]. Inhibiting the degradation of endogenous signaling molecules provides an attractive approach for therapeutic intervention since this approach may prevent side effects related to direct cannabinoid receptor agonism by synthetic molecules [20]. Because FAAH blockade only potentiates an activated signaling pathway by raising the endogenous concentration of the lipid-signaling molecule at its site of action, this blockade provides a temporal and spatial pharmacological control that is not typically available to a more classic agonist directly acting to a receptor. Mice completely lacking FAAH sense less pain and have less motility [21].
Only a few crystal structures are currently available for AS family members. These include the structures of FAAH from Rattus norvegicus, malonamidase E2 (MAE2) from Bradyrhizobium japonicum, peptide amidase (Pam) from Stenotrophomonas maltophilia, glutamyl-tRNA Gln amidotransferases from Staphylococcus aureus, and a 6-aminohexanoate cyclic dimer hydrolase from Arthrobacter sp. [2][3][4][5][6]. Structural studies of the FAAH protein elucidate the action mechanism of the inhibitor using a series of inhibitors and FAAH from Rattus norvegicus (RnFAAH). However, no other structural reports, such as comparisons with other orthologous structures have been reported. Moreover, the functional variety and the wide evolutionary distribution of AS family enzymes underscore the importance of determining the three-dimensional structure of these enzymes in order to understand the structure-function relationship.
In order to provide a broader view of the three-dimensional structure and to suggest a structural basis for the widely varying individual substrate specificities of AS family enzymes, we determined the crystal structure of FAAH from the pathogenic fungus Candida albicans at a resolution of 2.2 Å. Based on structural analysis of Candida albicans FAAH (CaFAAH), we compared the newly determined structure with that of RnFAAH as well as those of the Bradyrhizobium japonicum MAE2 and Stenotrophomonas maltophilia peptide amidase (Pam). Furthermore, using both structure and sequence comparisons, the evolutionary relationship between AS family members was investigated based on their phylogenetic relationships. Ultimately, this study may provide insight into the development of more potent or widely used inhibitors for FAAH.

Cloning of FAAH from Candida Albicans
The CaFAAH gene was amplified from a Candida albicans genomic DNA library using the polymerase chain reaction (PCR), as described in a previous study [22]. Briefly, the amplified fragment was digested with the restriction enzymes NheI and XhoI (R016S and R0075, respectively, Enzynomics, Daejeon, Republic of Korea), and was ligated into pET26b vectors using T4 DNA ligase (M0202S, Roche, Basel, Switzerland). The plasmid was then transformed into Escherichia coli (E. coli) strain DH5α, and the transformants were confirmed using colony PCR. All oligonucleotides used in the study were purchased from Cosmogenetech (Seoul, Republic of Korea).
To overcome the phasing problem, we produced a selenomethionine-substituted protein. Plasmids encoding CaFAAH were transformed into the methionine auxotrophic E. coli strain B834 (DE3) and cultured using a synthetic M9 minimal medium [23]. Cells were grown and induced with the same procedure as in the LB medium. His-tag affinity chromatography using Ni-NTA agarose and SEC was performed using the same method as the native CaFAAH purification process. To preserve the oxidation of the Se-Met crystals, the buffer used contained 150 mM NaCl, 10 mM dithiothreitol (DTT; 233155, Calbiochem, USA), and 25 mM Tris/HCl at pH 7.5 (Sigma-Aldrich, USA).

Crystallization
All crystallization trials were performed at 7 • C using either sitting-drop or hanging-drop vapor diffusion methods. CaFAAH crystals grew within a day in the crystallization condition of 30% w/v polyethylene glycol (PEG) 4000, 200 mM lithium sulfate, and 100 mM Tris/HCl, pH 8.5. The selenomethionine-derivatized CaFAAH crystals were tested under similar conditions to that of the native crystals of CaFAAH but conditions included the presence of 10 mM DTT. Prior to flash cooling all crystals in liquid nitrogen, 20% glycerol was added to the reservoir solution as a cryoprotectant.

Data Collection and Structure Determination
All diffraction datasets were collected at 100 K on the beamline 7A at the Pohang Accelerator Laboratory (PAL; Pohang, Republic of Korea) using a Quantum 270 CCD detector (San Jose, CA, USA). Data were processed using the HKL-2000 suite [24]. Crystals of the selenomethione-derivatized CaFAAH belonged to space group P2 1 and they diffracted to a resolution of 2.2 Å. The crystal structures were solved with single-wavelength anomalous diffraction phasing methods using the PHENIX.autoSol program in the PHENIX package version 1.9-1692 [25]. Model building was performed using the Wincoot program [26]. The structural models were refined using the PHENIX refine program. Data collection details and statistics are provided in Table 1.
where |F o | and |F c | are the observed and calculated structure factor amplitudes, respectively; d R free was calculated using 5% of the data.

Phylogenetic Analysis
The amino acid sequences of CaFAAH and other relevant species were obtained from the National Center for Biotechnology Information (NCBI) database (Table 2). Amino acid sequences were aligned using ClustalW [27]. Only well-aligned and conserved alignment sites were extracted from each alignment subset using Gblock (v0.91b) [28]. The phylogenetic tree was reconstructed with the maximum likelihood method using the PhyML program (v3.1/3.0) [29]. aLRT graphical representation of the phylogenetic tree (cladogram) was produced using TreeDyn from the PHYLIP package (v3.66), which is available at the website (http://www.phylogeny.fr/).

Overall Structure of CaFAAH
Two molecules of CaFAAH were found in the asymmetric unit and each monomer in the dimeric CaFAAH was positioned with two-fold symmetry ( Figure 1A). There are two functional regions in the dimeric structure. In the first region, both the α13-β7 and β5-β6 loops are mainly involved in dimerization through interactions with α18 helices in their partner molecules ( Figure 1A). In the second functional region, each of the α1-α2 loops is located on the same side of the dimer at a 90 • rotation along the X-axis. As was also observed in the RnFAAH structure [4], this parallel orientation of each monomer has a role in substrate recruitment from the same side of the membrane.
The overall structure of the CaFAAH monomer consists of 11 central twisted β-sheets surrounded by 22 α-helices ( Figure 1B). This structural arrangement is also present in previously reported FAAH structures from different species [14]. Furthermore, the structural fold could be presented as a feature of a broad range of AS family proteins. In the vicinity of the α3, α11, and β4 helices, the central deep pocket contains a unique Ser-Ser-Lys triad, which plays an important role in the enzyme's hydrolase activity ( Figure 1B,C). In addition, similar to other FAAHs, the pocket, also referred to as the active site, is located in the amidase signature domain. The oligomerization domain, which is involved in dimerization of the CaFAAH dimer and comprised of helix α13 and part of loop α1-α2, is located outside of the central region ( Figure 1B,C).  Figure 1C. Each of the active sites and oligomerization domains are highlighted with a yellow and a green box, respectively.

Active Site
The catalytic triad signature comprised of residues Ser233, Ser257, and Lys158 is a main component of the active site in the CaFAAH structure ( Figure 2A). These residues are referred to as the AS sequence and are highly conserved among the amidase family members [30]. The Ser257 residue acts as the primary nucleophile that attacks the substrate and the Lys158 plays a role as a general acid/base ( Figure 2B). Ser217 serves as a proton-shuttle between Lys158 and Ser257. In addition, four residues, namely Lys210, Ile254, Asn397, and Asn422 are expected to interact with the substrate and therefore are critical for the amidase activity of FAAH.  Figure 1C. Each of the active sites and oligomerization domains are highlighted with a yellow and a green box, respectively.

Active Site
The catalytic triad signature comprised of residues Ser233, Ser257, and Lys158 is a main component of the active site in the CaFAAH structure ( Figure 2A). These residues are referred to as the AS sequence and are highly conserved among the amidase family members [30]. The Ser257 residue acts as the primary nucleophile that attacks the substrate and the Lys158 plays a role as a general acid/base ( Figure 2B). Ser217 serves as a proton-shuttle between Lys158 and Ser257. In addition, four residues, namely Lys210, Ile254, Asn397, and Asn422 are expected to interact with the substrate and therefore are critical for the amidase activity of FAAH.

Comparison of CaFAAH Structure with the AS Family Protein RnFAAH
To assess the structural relationship between CaFAAH and other AS family proteins, we searched for homologous structures using the DaLi server [31]. The output suggested that the structure with the greatest homology to CaFAAH is FAAH from Rattus norvegicus (RnFAAH; PDB code, 1MT5) with a Z-score of 45.5 and a root mean square deviation (r.m.s.d) value of 2.0 Å. A structure-based sequence comparison of CaFAAH with RnFAAH showed they were 26.74% identical with a high degree of conservation being especially noted for the AS sequence. However, there are several interesting differences in both the secondary structures and sequences (Supplementary Materials Figure 1).
While the overall structures of both CaFAAH and RnFAAH are similar, there are five discriminative regions in the overlaid structures. First, the α20 helix in CaFAAH, covering residues 511-518, is positioned ~8.8 Å outward compared to that of the equivalent helix in RnFAAH ( Figure  3A; box I). Second, a major difference is that the hydrophobic cap, composed of two helices (α18 and α19) connected by a hairpin loop covering residues 404-433 in RnFAAH, is absent in the CaFAAH structure ( Figure 3A; box II). The proposed role of the hydrophobic cap in RnFAAH is to assist in membrane insertion of the N-terminal region [4]. Third, the secondary structure covering residues 330~335 in CaFAAH shows an α13-β2 loop but the equivalent region in RnFAAH forms an α-helix (α11) instead of a loop ( Figure 3B; box III). Fourth, the oligomerization domain α13 helix covering

Comparison of CaFAAH Structure with the AS Family Protein RnFAAH
To assess the structural relationship between CaFAAH and other AS family proteins, we searched for homologous structures using the DaLi server [31]. The output suggested that the structure with the greatest homology to CaFAAH is FAAH from Rattus norvegicus (RnFAAH; PDB code, 1MT5) with a Z-score of 45.5 and a root mean square deviation (r.m.s.d) value of 2.0 Å. A structure-based sequence comparison of CaFAAH with RnFAAH showed they were 26.74% identical with a high degree of conservation being especially noted for the AS sequence. However, there are several interesting differences in both the secondary structures and sequences (Supplementary Materials Figure S1).
While the overall structures of both CaFAAH and RnFAAH are similar, there are five discriminative regions in the overlaid structures. First, the α20 helix in CaFAAH, covering residues 511-518, is positioned~8.8 Å outward compared to that of the equivalent helix in RnFAAH ( Figure 3A; box I). Second, a major difference is that the hydrophobic cap, composed of two helices (α18 and α19) connected by a hairpin loop covering residues 404-433 in RnFAAH, is absent in the CaFAAH structure ( Figure 3A; box II). The proposed role of the hydrophobic cap in RnFAAH is to assist in membrane insertion of the N-terminal region [4]. Third, the secondary structure covering residues 330~335 in CaFAAH shows an α13-β2 loop but the equivalent region in RnFAAH forms an α-helix (α11) instead of a loop ( Figure 3B; box III). Fourth, the oligomerization domain α13 helix covering residues 317-330 in CaFAAH has a different amino acid sequence but the same secondary structure as the α10 helix in the RnFAAH structure ( Figure 3B; box IV, Supplementary Materials Figure S1). Therefore, the mode of dimerization should be similar in both structures. Finally, the N-terminal region covering residues 1-29 of RnFAAH is predicted to be a monotopic transmembrane domain. Therefore, the N-terminal part was excluded from the reported structure in order to improve the purification and crystallization of the protein [4]. However, the amino acid sequence of the equivalent region in CaFAAH is very different. Furthermore, since the full-length CaFAAH was crystallized, the structure of CaFAAH contains an N-terminal region comprising 25 amino acid residues that form two α-helices; these are not present in the RnFAAH structure ( Figure 3B; box V, Supplementary Materials Figure S1).
Taken together, CaFAAH has several distinct structural features compared to RnFAAH. Of importance, two major hydrophobic patches that form the N-terminal transmembrane domain and the hydrophobic cap, respectively, are not present in the CaFAAH structure. These features suggest that CaFAAH has unique functions or may indicate evolutionary changes reflective of fungal FAAHs within the broader AS family. region in CaFAAH is very different. Furthermore, since the full-length CaFAAH was crystallized, the structure of CaFAAH contains an N-terminal region comprising 25 amino acid residues that form two α-helices; these are not present in the RnFAAH structure ( Figure 3B; box V, Supplementary Materials Figure 1). Taken together, CaFAAH has several distinct structural features compared to RnFAAH. Of importance, two major hydrophobic patches that form the N-terminal transmembrane domain and the hydrophobic cap, respectively, are not present in the CaFAAH structure. These features suggest that CaFAAH has unique functions or may indicate evolutionary changes reflective of fungal FAAHs within the broader AS family.

Comparison of CaFAAH Structure with AS Family Proteins BjMAE2 and SmFAM
Other than RnFAAH, two more crystal structures in the AS family have been reported, namely malonamidase (MAE2) from Bradyrhizobium japonicum (BjMAE2; PDB code, 1OCK) and peptide amidase (PAM) from Stenotrophomonas maltophilia (SmFAM; PDB code, 1M22) [3,32]. We also compared these two structures with that of CaFAAH using a pair-wise analysis of the DaLi server

Comparison of CaFAAH Structure with AS Family Proteins BjMAE2 and SmFAM
Other than RnFAAH, two more crystal structures in the AS family have been reported, namely malonamidase (MAE2) from Bradyrhizobium japonicum (BjMAE2; PDB code, 1OCK) and peptide amidase (PAM) from Stenotrophomonas maltophilia (SmFAM; PDB code, 1M22) [3,32]. We also compared these two structures with that of CaFAAH using a pair-wise analysis of the DaLi server [31].
A comparison of the structure-based sequence alignment between CaFAAH and BjMAE2 showed the two proteins were 23.18% identical (Supplementary Materials Figure S2), with a Z-score of 40.4 and an r.m.s.d of 2.3 Å. However, there were three regions of difference in the overlaid structures ( Figure 3C). First, the CaFAAH structure contained five additional α-helices (α1~α5) in its N-terminus, as well as two α-helices (α20-α21) near the C-terminus; no equivalent regions were found in BjMAE2 ( Figure 3C; boxes I, II). In addition, the β1-β2 loop covering 20 residues in BjMAE2 was substituted with an α-helix (α8) in the CaFAAH structure ( Figure 3C; dashed region III).
A comparison of the structure-based sequence alignment between CaFAAH and SmPAM showed the two proteins were 24.74% identical (Supplementary Materials Figure S3), with a Z-score of 36.6 and an r.m.s.d. of 2.4 Å. These values indicated that the similarity of SmPAM to CaFAAH was somewhat lower than that of BjMAE2. There were also two regions of significant difference in the superposed structures ( Figure 3D). First, similar to the overlaid CaFAAH and BjMAE2 structures, five N-terminal α-helices (α1-α5) in CaFAAH were absent in the structure of SmPAM ( Figure 3D; dashed region I) and second, the α8-α9 loop of SmPAM was present as the α13 helix in the CaFAAH structure ( Figure 3D; box II).
A comparison of CaFAAH with the RnFAAH-PF-750 complex (PDB code: 2VYA) showed that Leu210, Met208, and His393 of CaFAAH may interact with the pyridine ring of PF-750 in addition to the residues of the catalytic triad ( Figure 4A). However, Trp481 from CaFAAH, which is equivalent to Ile491 from RnFAAH, clashed with the PF-750 inhibitor, suggesting there may be a different conformation for PF-750 in CaFAAH. Furthermore, residues Phe432 and Thr488 in RnFAAH, which interact with the quinolone ring of PF-750, were not found in the structure of CaFAAH ( Figure 4A). These observations suggest that the binding affinity of PF-750 for CaFAAH may be lower than that for RnFAAH.
A comparison of CaFAAH with the RnFAAH-URB597 complex (PDB code: 3LJ6) indicated that residues Met208 and Leu210 from CaFAAH may interact with the pyridine ring of URB597 ( Figure 4B). However, His393 from CaFAAH (which has no equivalent residue in RnFAAH) was incompatible with URB597, and it is therefore presumed that the URB597 inhibitor may have an altered conformation at the active site of CaFAAH. Moreover, RnFAAH's residues Phe432 and Tyr488, which interact with the pyridine and phenol rings of URB597, do not exist in the CaFAAH structure ( Figure 4B).
The comparison of CaFAAH with the RnFAAH-MAFP complex (PDB code: 1MT5) showed that the Met208 and Leu210 residues from CaFAAH may interact with the aliphatic chain of MAFP in a manner similar to URB597 ( Figure 4C). However, in this model, both His393 and Trp481 in CaFAAH are also incompatible with MAFP, which is likely to affect MAFP localization in CaFAAH. As with the PF-750 and URB597 inhibitors, RnFAAH residues Phe432 and Thr488, which are responsible for forming an interaction with the aliphatic chain of MAFP, do not exist in CaFAAH ( Figure 4C). Interestingly, MAFP contains a fluorophosphonate group where the position of the oxygen atom in the group is equivalent to that of the water molecule (W1) shown in the PF-750 and URB597 complex structure ( Figure 4). Therefore, the Ser233 residue from CaFAAH may interact with the oxygen atom in the fluorophosphonate group of MAFP instead of a water molecule as was seen in the two other complexes ( Figure 4C).

Phylogenetic Analysis of AS Family Proteins
Based on the local structural differences found, we investigated the phylogenetic relationships and evolutionary distances among 19 members of the AS family (Table 2). CaFAAH is phylogenetically closely related to both Kluyveromyces lacts FAAH and Debaryomyces hansenii FAAH, indicating that fungal FAAH could be categorized to a monophyletic taxon ( Figure 5). Likewise, bacteria, fungi, and higher eukaryotes converge to a polyphyletic taxon. This result indicates that AScontaining genes already existed in common ancestors prior to the divergence of prokarya and eukarya [33]. In addition, for the three AS family members that we previously compared using structural analysis, CaFAAH is more closely related to RnFAAH than either BjMAE2 or SmPAM based on the phylogenetic analysis.

Phylogenetic Analysis of AS Family Proteins
Based on the local structural differences found, we investigated the phylogenetic relationships and evolutionary distances among 19 members of the AS family (Table 2). CaFAAH is phylogenetically closely related to both Kluyveromyces lacts FAAH and Debaryomyces hansenii FAAH, indicating that fungal FAAH could be categorized to a monophyletic taxon ( Figure 5). Likewise, bacteria, fungi, and higher eukaryotes converge to a polyphyletic taxon. This result indicates that AS-containing genes already existed in common ancestors prior to the divergence of prokarya and eukarya [33]. In addition, for the three AS family members that we previously compared using structural analysis, CaFAAH is more closely related to RnFAAH than either BjMAE2 or SmPAM based on the phylogenetic analysis. using ClustalW [27] and the phylogenetic tree was produced using the PHYLIP package program.

Discussion
The AS family of proteins, which contains the amidase consensus sequence, plays important roles in biological processes. They are bifunctional enzymes with both intrinsic amidase and esterase/peptidase activities, able to cleave ester, amide, and nitrile bonds depending on their enzyme-specific ratios [34][35][36][37]. Currently, a large number of proteins containing AS sequences has been identified, including amidases and aspartic proteinases that may form part of this family [38]. Among them, FAAH, MAE2, PAM, subunit A of Glu-tRNA Gln amidotransferase, and 6aminohexanoate cyclic dimer hydrolase are considered representative members of the AS family that share somewhat conserved active sites [38].
Among the five members of the AS family, FAAH, which contains a common catalytic triad (Ser-Ser-Lys), is characterized as being able to hydrolyze amide bonds [1]. Moreover, because FAAH is highly involved in the neuronal signal transduction pathways that relay pain, many groups have focused on developing potent chemical inhibitors using RnFAAH [10]. Therefore, there have been several reports on the crystal structures of RnFAAH in complexes with inhibitors [16,[39][40][41][42][43]. Because there is limited structural information about FAAHs beyond RnFAAH, it is necessary to compare structures homologous with RnFAAH to expand our knowledge of the functional aspects of FAAH. Such a comparison may also provide new information for drug development.
Comparisons with the structures of AS family members revealed several critical differences, such as the presence of an N-terminal non-hydrophobic helical region corresponding to the transmembrane domain of RnFAAH (Supplementary Materials Figure 1). In addition, the hydrophobic cap in RnFAAH was not present in the CaFAAH, BjMAE2, and SmPAM structures. Furthermore, the N-terminal region of CaFAAH was completely absent in BjMAE2 and SmPAM (Supplementary Materials Figure 2; Figure 3). These discrepancies in the structure/sequence may be indicative of the evolutional developments/diversity of FAAH proteins. Our phylogenetic analysis showed that AS family members including FAAH proteins were clearly converged to a polyphyletic taxon from a common ancestor. Moreover, although limited structural reports are available, there may be an evolutionary correlation between the sequence and structure of AS family members.
Taken together, the structural information presented here and the phylogenetic analysis of CaFAAH may provide insights into developing broader, more effective, potent drugs for the treatment of pain in addition to having implications in the divergence of enzymes. However, further studies with fungal FAAHs are required to uncover their physiological substrates as well as to understand their biological relationships with higher eukaryotic FAAH proteins.

Discussion
The AS family of proteins, which contains the amidase consensus sequence, plays important roles in biological processes. They are bifunctional enzymes with both intrinsic amidase and esterase/peptidase activities, able to cleave ester, amide, and nitrile bonds depending on their enzyme-specific ratios [34][35][36][37]. Currently, a large number of proteins containing AS sequences has been identified, including amidases and aspartic proteinases that may form part of this family [38]. Among them, FAAH, MAE2, PAM, subunit A of Glu-tRNA Gln amidotransferase, and 6-aminohexanoate cyclic dimer hydrolase are considered representative members of the AS family that share somewhat conserved active sites [38].
Among the five members of the AS family, FAAH, which contains a common catalytic triad (Ser-Ser-Lys), is characterized as being able to hydrolyze amide bonds [1]. Moreover, because FAAH is highly involved in the neuronal signal transduction pathways that relay pain, many groups have focused on developing potent chemical inhibitors using RnFAAH [10]. Therefore, there have been several reports on the crystal structures of RnFAAH in complexes with inhibitors [16,[39][40][41][42][43]. Because there is limited structural information about FAAHs beyond RnFAAH, it is necessary to compare structures homologous with RnFAAH to expand our knowledge of the functional aspects of FAAH. Such a comparison may also provide new information for drug development.
Comparisons with the structures of AS family members revealed several critical differences, such as the presence of an N-terminal non-hydrophobic helical region corresponding to the transmembrane domain of RnFAAH (Supplementary Materials Figure S1). In addition, the hydrophobic cap in RnFAAH was not present in the CaFAAH, BjMAE2, and SmPAM structures. Furthermore, the N-terminal region of CaFAAH was completely absent in BjMAE2 and SmPAM (Supplementary Materials Figures  S2 and S3). These discrepancies in the structure/sequence may be indicative of the evolutional developments/diversity of FAAH proteins. Our phylogenetic analysis showed that AS family members including FAAH proteins were clearly converged to a polyphyletic taxon from a common ancestor. Moreover, although limited structural reports are available, there may be an evolutionary correlation between the sequence and structure of AS family members.
Taken together, the structural information presented here and the phylogenetic analysis of CaFAAH may provide insights into developing broader, more effective, potent drugs for the treatment of pain in addition to having implications in the divergence of enzymes. However, further studies with fungal FAAHs are required to uncover their physiological substrates as well as to understand their biological relationships with higher eukaryotic FAAH proteins.