Evolution of Enzyme Superfamilies: Comprehensive Exploration of

Nov 1, 2016 - Enzyme evolution: innovation is easy, optimization is complicated. Matilda S .Newton , Vickery L Arcus , Monica L Gerth , Wayne M Patric...
5 downloads 0 Views 4MB Size
Current Topic pubs.acs.org/biochemistry

Evolution of Enzyme Superfamilies: Comprehensive Exploration of Sequence−Function Relationships F. Baier, J. N. Copp, and N. Tokuriki* Michael Smith Laboratories, University of British Columbia, Vancouver, BC V6T 1Z4, Canada ABSTRACT: The sequence and functional diversity of enzyme superfamilies have expanded through billions of years of evolution from a common ancestor. Understanding how protein sequence and functional “space” have expanded, at both the evolutionary and molecular level, is central to biochemistry, molecular biology, and evolutionary biology. Integrative approaches that examine protein sequence, structure, and function have begun to provide comprehensive views of the functional diversity and evolutionary relationships within enzyme superfamilies. In this review, we outline the recent advances in our understanding of enzyme evolution and superfamily functional diversity. We describe the tools that have been used to comprehensively analyze sequence relationships and to characterize sequence and function relationships. We also highlight recent large-scale experimental approaches that systematically determine the activity profiles across enzyme superfamilies. We identify several intriguing insights from this recent body of work. First, promiscuous activities are prevalent among extant enzymes. Second, many divergent proteins retain “function connectivity” via enzyme promiscuity, which can be used to probe the evolutionary potential and history of enzyme superfamilies. Finally, we discuss open questions regarding the intricacies of enzyme divergence, as well as potential research directions that will deepen our understanding of enzyme superfamily evolution.

E

1).5,7,11 They provide dramatic, but not unusual, examples of evolutionarily related functional diversity.2 Recent large-scale function profiling, combined with extensive bioinformatic studies, has revealed important details about the functional diversity and evolution of enzyme superfamilies.15−18 In this review, we summarize recent advances that have expanded our understanding of sequence and function diversity, as well as the evolutionary relationships within enzyme superfamilies. We highlight new observations of several enzyme superfamilies that come from exhaustive bioinformatics and large-scale function-profiling efforts. We analyze previously published large-scale experimental data from superfamily-wide enzyme function-profiling studies and discuss the implications in the context of evolutionary divergence of enzyme superfamilies. In particular, we note that seemingly distant enzyme functions are commonly connected through enzyme promiscuity across a superfamily. Furthermore, we suggest that these efforts should be integrated with new approaches to gain further insights into how enzyme superfamilies diverge and to reveal the molecular details underlying enzyme evolution.

nzymes have evolved remarkable functional diversity, and understanding how new enzyme functions evolve is a central goal in biological science. The initial steps toward understanding enzyme evolution were taken more than 80 years ago when enzymes were first structurally and biochemically characterized.1 An enormous body of work since that time has profoundly deepened our knowledge of enzyme functional diversity and their sequence−structure−function relationships.2−7 It has become clear that the diversity of structural folds and mechanistic features in the active site (catalytic residues and cofactors) are critical determinants of the extraordinarily diverse functional repertoire observed in modern enzymes.8−12 At present, the CATH (class, architecture, topology and homologous superfamilies) database classifies 53 million structural domains that fall within 2737 superfamilies, in which proteins share a common structural fold, a sequence motif, and a set of catalytic features.13 Even within a single superfamily exist enzymes that are capable of catalyzing a diverse range of chemical reactions, to an astonishing degree.2,14 Enzymes in the same superfamily often act on different substrates (i.e., altered size, shape, and electrostatics) and/or catalyze different chemical reactions (i.e., altering the chemical bond that is broken or formed, or even utilizing completely different chemistry, such as hydrolase, isomerase, and oxidase reactions).2,14 Examples of functionally diverse enzyme superfamilies include the haloacid dehalogenase (HAD), enolase, cytosolic glutathione transferase (cytGST), metallo-β-lactamase (MBL), and amidohydrolase superfamilies, each of which contains members that catalyze a wide variety of distinct chemical reactions, spanning all six EC classes (Figure © 2016 American Chemical Society



PROMISCUITY AS A ROUTE TO FUNCTIONAL DIVERGENCE WITHIN PROTEIN SUPERFAMILIES The modern theory of how enzymes evolve was established in 1970s by Jacob, who postulated that “Evolution does not Received: July 16, 2016 Revised: October 8, 2016 Published: November 1, 2016 6375

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388

Current Topic

Biochemistry

Figure 1. Sequence and function diversity within selected enzyme superfamilies: HAD (haloacid dehalogenase), cytGST (cytosolic glutathione transferase), amidohydrolase, MBL (metallo-β-lactamase), and enolase. Sequence diversity and structural fold information was retrieved from the SFLD31 and PFAM33 databases. Function diversity was retrieved from the CATH database.13

produce innovations from scratch. It works on what already exists, either transforming a system to give it a new function or combining several systems to produce a more complex one”.19 In this view, a new enzyme function evolves from a preexisting enzyme that exhibits promiscuous functions, which we define here as any latent and secondary functions additional to the enzyme’s native and physiological functions. Concurrently, Jensen conceptualized enzyme evolution as a process whereby substrate promiscuity (or ambiguity; ability to turn over nonnative substrates that represent the same enzymatic reaction to the native substrate) and catalytic promiscuity (ability to catalyze non-native reactions) are the foundation for the evolution of new functions.20 Thus, ancestral enzymes may have been multifunctional (nonspecialized) or promiscuous, and with a change in the environmental selection pressure, a promiscuous activity could become the target of selection and be further improved through the accumulation of adaptive mutations that yield higher catalytic efficiencies for that function (Figure 2). Gene duplication (before or after adaptive mutations) eventually leads to the emergence of a new enzyme function.21 Indeed, this classical view is supported by several recent enzyme evolution studies.22−27 For example, several xenobiotic degrading enzymes, such as organophosphate hydrolase and atrazine chlorohydrolase, evolved from precursor enzymes that possessed those functions as latent promiscuous activities.28,29 Additionally, in the course of sequence and functional expansions, new catalytic activities may emerge, which subsequently allow further functional divergence, ultimately leading to the broad range of enzymatic functions observed within contemporary enzyme superfamilies (Figure 1).

Figure 2. Schematic representation of the evolutionary process by which functional divergence occurs within a theoretical enzyme superfamily. Circles represent a single sequence (enzyme), and colors represent the native physiological function. The inner circle represents promiscuous activities. The functional divergence from a common ancestor (light blue) occurs via the recruitment of promiscuous activities and evolutionary optimization of these functions to generate new specialized enzymes (gray, deep blue, and olive). During the adaptive process or genetic drift, a new promiscuous function may subsequently arise in a derived family and lead to further expansion of the functional repertoire in the superfamily (green, pale blue, purple, and magenta).



capture global sequence relationships within protein superfamilies.33 Traditional phylogenetic analyses can be used to identify sequence-based clusters that can be classified on the basis of the early branches in the superfamily’s phylogeny.32 While justified, this approach can be problematic because a single superfamily can easily exceed 10000 sequences, and amino acid sequence identities between enzymes belonging to different functional families can be extremely low, sometimes with only a few catalytically important residues being conserved.33 Thus, performing phylogenetic analysis with all

EXHAUSTIVE SEQUENCE CHARACTERIZATION FACILITATES COMPREHENSIVE CLASSIFICATION OF FUNCTIONAL FAMILIES The first step toward understanding the sequence and functional diversity within a superfamily, as well as the evolutionary relationships between its extant members, is to establish rigorous classification systems that assign sequences into functional families.30−32 The rapid increase in the volume of sequencing data over the past two decades allowed us to 6376

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388

Current Topic

Biochemistry

Figure 3. Representative sequence similarity network (SSN) of the MBL superfamily members reproduced from ref 15; 33843 protein sequences are represented by 6233 nodes, which share 60% of the enzymes that were tested (50 of 80) displayed activity for more than one β-keto acid substrate, whereas 30 of those 80 enzymes were specific. The FCN of the BKACE family leads to conclusions similar to those reached about the FCNs of the MBL and cytGST superfamilies (Figure 6A), in which all the functions that were tested, which correspond to different substrates of β-keto acid cleavage reaction, are connected through promiscuous enzymes. Furthermore, substrates with distinct chemical features, e.g., anionic, cationic, nonionic polar, and apolar, form separate clusters. In general, nonionic polar and apolar substrates are more connected through promiscuous enzymes, as highlighted by the β-ketohexanoate substrate [S11 (Figure 6A)], which is connected to all other substrates. In contrast, charged substrates [green border, cationic S1 and S2; red border, anionic S3 and S4 (Figure 6A)] form distinct clusters, which are located at the peripheries of the FCNs and are overall less widely connected. Haloacid Dehalogenase Superfamily. The HAD superfamily is comprised of 120000 sequences, which share a Rossmann fold “core” domain and a fused “cap” domain.17,69 The majority of HAD enzymes require Mg2+ and a conserved active site aspartic acid.69 Although the superfamily is named after the haloacid dehalogenase (C−Cl bond cleavage) enzyme,

Figure 4. Schematic depiction of superfamily-wide function profiling and function connectivity network (FCN). (A) Selected enzymes from a superfamily are screened against a set of enzymatic functions to reveal native (high activity, dark shading) and promiscuous activities (lower activities, lighter shading). The activity data are typically represented in a table or heat map format. (B) Same activity data visualized as a “function connectivity network” (FCN) using Cytoscape.134 Enzymes (circles) are connected to a function (large square) if they exhibit activity (edges). In Figures 5 and 7, multiple substrates are used to assay some of the enzymatic functions; in this case, an edge is drawn when an enzyme is active toward at least one substrate. Enzyme nodes are qualitatively shaded depending on the number of activities they catalyze, from white being specific for one function to black being highly promiscuous (at least seven functions). Enzymes and functions cluster qualitatively depending on their connectivity, with highly interconnected nodes clustering together.

hydrolytic reactions.15 The enzymes were highly specialized toward their native reactions (kcat/KM ≥ 104 M−1 s−1). However, most MBL enzymes exhibited some degree of promiscuous activity, albeit with a comparatively low catalytic efficiency (kcat/KM ≤ 102 M−1 s−1). Overall, 18 of the 24 enzymes catalyzed multiple reactions, and on average, each enzyme catalyzed 2.5 of the 10 reactions tested. The prevalence of promiscuity in this superfamily means there is high connectivity between functions, with 9 of 10 functions being connected in the FCN (Figure 5A). Interestingly, we observe a subclustering of enzymatic reactions in the FCN; reactions that are related to anionic substrates (phosphodieterase, sulfatase, and phosphonate hydrolase) and neutral substrates (βlactamase, glyoxalase II, and lactonase) form isolated clusters (Figure 5A). A similar clustering of sequences (enzymes) associated with reactions using either anionic or neutral substrates was also observed in SSNs and phylogenetic analysis, which suggests that the FCN clustering reflects functional divergence in evolutionary history (Figure 3).15,66 Cytosolic Glutathione Transferase Superfamily. The cytosolic glutathione transferase (cytGST) superfamily contains more than 10000 known sequences. Its most common functions involve the utilization of glutathione as a cofactor to metabolize endogenous compounds, detoxify chemicals, and prevent oxidative stress.16,67 CytGSTs structurally consist of two domains: a smaller N-terminal thioredoxin-like fold domain that binds glutathione and a larger C-terminal domain that is responsible for substrate recognition and binding.67 The superfamily catalyzes a diverse range of functions (more than 140 EC terms) that can be categorized mechanistically into three groups based on the usage of glutathione: “oxidized”, “consumed” or “not consumed”.16 Furthermore, the three groups can be subclassified into 15 functions according to their reaction chemistry.16 Mashiyama et al. recently performed a large-scale function-profiling analysis and characterized 82 enzymes in this superfamily for 15 different functions, which 6379

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388

Current Topic

Biochemistry

Figure 5. FCNs constructed on the basis of the function-profiling analysis of the MBL and cytGST superfamilies. (A and B) Large square nodes represent functions; small nodes represent enzymes, and edges indicate the existence of enzymatic activity. The numbers within each function node indicate how many other functions are directly connected via promiscuous enzymes. Enzyme nodes are quantitatively shaded depending on the number of activities they catalyze, from white being specific for one function (connected to one function) to black being highly promiscuous (connected to at least seven functions). (A) The FCN of the MBL superfamily represents the function-profiling analysis of 24 enzymes against 10 enzymatic functions (using 12 different substrates) that represent native functions of 10 families:15 BLA, β-lactamase; TPN, chlorothalonyl dehalogenase; AKS, alkylsulfatase; SLG, glyoxalase II; ARS, arylsulfatase; PDE, phosphodiesterase; PPP, phosphonate monoester hydrolase; PCE, phosphorylcholine-esterase; PTE, phosphotriesterase; HSL, homoserine lactonase. The thickness of the edge represents the level of enzymatic activity. (B) The FCN of the cytGST superfamily represents the function-profiling analysis of the catalytic activity of 256 enzymes for 15 enzymatic functions using 175 different substrates.16 Triangle nodes represent enzymes of the AMPS sequence cluster, which is shown in panel C. The border color of square nodes indicates enzymatic function: glutathione oxidized (blue border), glutathione consumed (red border), or not consumed (green border). Enzymatic functions: NAS, nucleophilic aromatic substitution; NS, nucleophilic substitution; CA, conjugate addition; NA, nucleophilic addition; ERO, epoxide ring opening; TL, thiolysis; IMZ, isomerization; HD, hydrolytic dehalogenation; DSBR, disulfide bond reductase; PO, peroxidase; RD, reductive dehalogenation; DG, deglutathionylation; DHAR, dehydroascorbate reductase; AAR, alkylarsenate reductase. The numbers above each enzymatic function indicate how many other functions are directly connected via a promiscuous enzyme. (C) SSN of the 54 AMPS cluster proteins that are included in panel B, generated using the enzyme function initiative (EFI) enzyme similarity tool3 and visualized in Cytoscape.134 The shading of each node denotes the level of promiscuity as shown in panel B.

(small box in Figure 6B). We believe that the high connectivity is associated with the extensive number of experimental characterizations (>200 enzymes against 167 substrates). In addition, the set of substrates is less diverse compared to those of the other studies; they generally involve the same chemical reactions [P−O (98%) and P−C (2%) bond cleavage] and thus almost exclusively differ by their leaving group. To obtain a better separation and clustering of functions (substrate classes), we re-analyzed the data using a more stringent OD650 = 0.5 cutoff (the highest OD650 observed in the assay was ∼1.0), which gave rise to a clustering of the distinct functions (substrate classes). For example, we observe that all five monophosphate sugar substrates (acid, alcohol, aldose, amine, and ketose sugars) cluster together, whereas other functions are separated. On the other hand, phosphonate substrates, which involve cleavage of P−C bonds instead of P−O bonds (all other substrates), are the least connected substrate class. Thus, employing an activity cutoff, similar to the BLAST E-value cutoff in SSNs, can be a useful method for observing relationships between functions and enzymes in FCNs.

the majority of enzymes are phosphate hydrolases such as phosphoesterases, ATPases, sugar phosphomutases, and other phosphatases and phosphonatases involved in P−O and P−C bond cleavage.69 The cellular functions of HAD phosphatases include primary metabolism of amino acids and sugars, secondary metabolism, dNTP pool regulation, cellular housekeeping, and nutrient uptake.69 Huang and co-workers measured the activity profile of more than 200 enzymes against a diverse library of 167 phosphatase (98%) and phosphonatase (2%) substrates (grouped into 13 substrate classes).17 The authors considered an enzyme−substrate pair as active if a 30 min incubation of the reaction mixture (5 μM enzyme and 1 mM substrate) provides an OD650 of ≥0.2, using a chemical dye to detect the release of inorganic phosphate. Overall, most of the tested enzymes were highly promiscuous; 75% of the enzymes act on more than five substrates, and 23% are active for more than 41 substrates (up to 143 substrates).17 Indeed, the FCN of the HAD superfamily generated from this data set revealed very dense connectivity between functions (substrate classes) and complete connection among all 13 functions, in which each function is connected to every other functional class 6380

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388

Current Topic

Biochemistry

Figure 6. FCN representation of the function-profiling analysis of the BKACE and HAD superfamilies, in which there is divergent substrate specificity but generally conserved chemical reaction. (A and B) Large square nodes represent functions; small nodes represent enzymes, and edges indicate the existence of enzymatic activity. The numbers within each function node indicate how many other functions are directly connected via promiscuous enzymes. Enzyme nodes are quantitatively shaded depending on the number of activities they catalyze, from white being specific for one function (connected to one function) to black being highly promiscuous (connected to at least seven functions). (A) FCN of the BKACE family, which represents the function-profiling analysis of 80 enzymes against 15 substrates.18 Substrates are catalyzed by condensation with acetylCoA to produce a CoA ester and acetoacetate. The border color of square nodes indicates the substrate property: anionic (green border), cationic (red border), nonionic polar (blue border), or apolar (black border). Substrates (only for forward reaction) are indicated by the following: S1, SKAH; S2, dehydrocarnitine; S3, β-ketoadipate; S4, β-ketoglutarate; S5, 3,5-dioxohexanoate; S6, 5-hydroxy-β-ketohexanoate; S7, 6-acetoamido-βketohexanoate; S8, β-ketopentanoate; S9, β-ketoisocaproate; S10, (E)-β-ketohex-4-enoate; S11, β-ketohexanoate; S12, 7-methyl-β-ketooct-6-enoate; S13, β-ketooctanate; S14, β-ketododecanoate; S15, benzoylacetate. (B) FCN of the HAD superfamily, which represents the function-profiling analysis of 204 enzymes against 167 substrates, which were classified into 13 different substrate classes: PP, phosphonates; DS, disaccharides; ALS, alcohol sugars; ACS, acid sugars; ADS, aldol sugars; NDT, nucleotide di- and triphosphates; NM, nucleotide monophosphates; KS, ketose sugars; BPS, bisphosphate sugars; AA, amino acids; AS, amine sugars; EH, easily hydrolyzed; others (two-, seven-, eight-, and nine-carbon sugars).17 The green border color of square nodes indicates monophosphate sugar substrate classes. The numbers within each substrate class node indicate how many other substrate classes are directly connected via promiscuous enzymes. The activity cutoffs correspond to the corrected absorbance of the end point assay employed by Huang et al. and considered OD650 ≥ 0.2 for their analysis. Note that the FCN with the OD650 = 0.5 cutoff contains only 145 enzymes, because activities of 59 enzymes were below the more stringent 0.5 cutoff.



EVOLUTIONARY PERSPECTIVES ON FUNCTION CONNECTIVITY THROUGH PROMISCUITY There are common features and general trends that can be extracted from these function-profiling analyses, which we discuss in the following sections. The FCNs provide intriguing insights into the function connectivity within superfamilies through enzymes with promiscuous functions. However, we would like to note some caveats in our analyses. First, each study features a different range of functions and sequences that were tested, i.e., the number and type of functions (chemical reactions, substrate classes, or substrates) and the number and sequence divergence of enzymes. Second, the assays for each study were different and had different levels of sensitivity, which can also affect how many functions per enzyme are revealed. For example, a less sensitive assay would mask activities that are below the detection limit, resulting in fewer promiscuous activities being identified. Most Functions Are Connected via Promiscuous Enzymes. The FCNs reveal that most functions are connected through enzyme promiscuity (Figures 5 and 6). Although these connections may not directly relate to the historical divergence of functions within the superfamilies, the high connectivity among these functions provides experimental support for the

notion that diverse functions could evolve from a common ancestor. Adaptive evolution proceeds through functional intermediates, and new functions are recruited from existing promiscuous enzymes and subsequently optimized through adaptive mutations and gene duplication events (Figure 2). Biased Promiscuity Underlies Indirect Function Connectivity. Despite the overall function connectivity, many functions are only indirectly connected through “intermediate functions”; i.e., two functions are not connected by the promiscuity of a single enzyme, but through at least one other function and two other enzymes (Figures 5 and 6). The most prominent example is seen in the BKACE family, where cationic and anionic substrates have no direct connections but are connected through only neutral substrates (Figure 6A). It is rational that an enzyme adapted to a cationic substrate would possess a negatively charged active site cavity that would not be appropriate for anionic substrates, and vice versa. Enzymes that have evolved toward neutral substrates, however, may be able to promiscuously act on both anionic and cationic substrates to some extent. Similar trends are observed in the other superfamilies that we examined (Figures 5 and 6). Therefore, the emergence of a new function may be associated with only distinct functions within a superfamily, and its evolution could be restricted to particular progenitor enzymes. In other words, 6381

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388

Current Topic

Biochemistry

Figure 7. FCN representation of the metal-dependent activity profiles for five MBL superfamily enzymes. (A−D) FCNs represent the functionprofiling analysis of five enzymes against eight enzymatic functions with six different metal ions (Fe2+, Zn2+, Co2+, Cd2+, Ni2+, and Mn2+). Large square nodes represent functions; small nodes represent enzymes, and edges indicate the existence of enzymatic activity. Enzyme nodes are quantitatively shaded depending on the number of activities they catalyze, from white being specific for one function (connected to one function) to black being highly promiscuous (connected to at least seven functions). (A) FCN of the activity profiles of all six reconstituted metal isoforms. (B) FCN of activity profiles of the Ni2+-reconstituted metal isoforms. (C) FCN of activity profiles of the Mn2+-reconstituted metal isoforms. (D) FCN of activity profiles of the Zn2+-reconstituted metal isoforms; the corresponding FCNs of Co2+, Cd2+, and Fe2+ are not shown because of space limitations. Enzymatic functions: BLA, β-lactamase; SLG, glyoxalase II; ARS, arylsulfatase; PDE, phosphodiesterase; PCE, phosphorylcholineesterase; PTE, phosphotriesterase; LAC, lactonase; EST, esterase/lipase. The numbers within each function node indicates how many other functions are directly connected via a promiscuous enzyme.

whereas others catalyze up to six reactions (Figure 5B). Interestingly, specificity and promiscuity levels are not correlated with particular sequence clusters within the AMPS subgroup but are instead scattered throughout (Figure 5C). Such cryptic genetic variation has also been observed in laboratory evolution studies where enzymes were subjected to “neutral drift” (accumulation of mutations under a purifying selection pressure for the native function), which produced notable changes in their activity promiscuous functional profiles.75,76 Thus, cryptic genetic variation can result in sequences that already have higher activities toward a new substrate and thus support a role for neutral genetic diversity in driving the innovation and evolution of new enzyme functions.72−74 This observation also indicates, however, that some sequences are less “evolvable” than others (in some cases, appearing entirely unevolvable) because they encode enzymes that do not exhibit enough promiscuous activity to provide a selective advantage when the environment changes to favor a new function.

functional expansion from a common ancestor may be constrained to one or a few specific trajectories to reach functions that are chemically distant from the ancestral functions (Figure 2). In contrast, however, some functions are frequently observed as being linked by promiscuous enzymes and thus are more likely to be connected to many other functions (Figures 5 and 6). These functions might easily evolve from various progenitor enzymes within a superfamily. Indeed, there are notable examples of convergent evolution in the literature: β-lactamase activity has arisen at least twice within the MBL superfamily.61,66 In the HAD superfamily, phosphomutase activity evolved independently more than once.69 Additional cases of convergent evolution have been observed in many other superfamilies, including phosphatidylinositol phosphodiesterases,6 enolases,37 and Zn-dependent peptidases.70 The Degree of Promiscuity Is Enzyme-Dependent. Many of the enzymes assayed across these studies were promiscuous, and some exhibited activity toward a remarkable variety of substrates [dark and black enzyme nodes in the FCNs (Figures 5 and 6)]. In contrast, some enzymes were specific to only one function [white enzyme nodes in the FCNs (Figures 5 and 6)]. Thus, the level of promiscuity is enzyme-dependent and varies significantly. It is possible, however, that expanding the scope of chemical reactions and substrates that are assayed may reveal promiscuous activities in enzymes that currently appear to be highly specific. The range and type of promiscuity also vary substantially among enzymes within the same functional family. Different members of the same functional family are, generally, orthologous enzymes that play the same physiological role in different species but possess diverse sequences due to genetic drift.71 These sequence changes are relatively neutral in terms of the native function but can alter promiscuous functions and thereby result in “cryptic genetic variation”.72−76 For example, among five B1 β-lactamases from the MBL superfamily, two enzymes catalyze only their native reaction while each of the remaining three enzymes is capable of up to three additional reactions (Figure 5A). Additionally, members of the AMPS (alpha-, mu-, pi-, and sigma-like) subgroup of the cytGST superfamily also exhibit different levels of promiscuity; some AMPS enzymes catalyze one reaction,



FUNCTION CONNECTIVITY DEPENDS ON THE ENVIRONMENT AND COFACTOR AVAILABILITY Function profiles and enzyme activity levels can vary depending on the environment. A simple example is that chemical engineers used “enzyme condition promiscuity” to achieve catalysis in water-limited environments with hydrolytic enzymes; such environs favor ester synthesis instead of hydrolysis.77 Similarly, other environmental changes can promote the appearance of new promiscuous functions. For example, a temperature shift can induce changes in substrate specificity in a bacterial thymidine kinase for various nucleoside analogues.78 Cofactor exchange, e.g., metal ions, flavins, and hemes, can cause condition-specific promiscuity and lead to different activity profiles depending on the conditions under which the enzymes were assayed.56,79−83 For example, a recent study by Lapalikar et al. showed that the substitution of F420dependent reductases with FMN instead of F420 allows them to catalyze both oxidation and reduction of the same substrate.84 Furthermore, a recent study by Reynolds et al. evolved a cytochrome P450 enzyme to incorporate a nonproteinogenic cofactor, iron deuteroporphyrin IX, which allows the enzyme to 6382

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388

Current Topic

Biochemistry

cap) or C1 and C2 (large cap). Huang et al. found that C1 and C2 enzymes exhibit generally more promiscuity than C0 enzymes do. For the cytGST superfamily, Mashiyama et al. could not identify any specific active site property or feature that is associated with the catalysis of particular reactions. Besides these rough tendencies, no study has successfully provided explicit molecular explanations for different types of enzyme functionality across a superfamily, let alone for the existence of promiscuous activities. This difficulty reflects the challenges faced when predicting and annotating the functions for new and/or uncharacterized sequences. Further efforts to characterize structural and functional details, including structural dynamics, are necessary for us to advance our understanding of the structure−function relationships that determine the catalytic function(s) that different enzymes perform.

catalyze the non-natural carbenoid-mediated olefin cyclopropanation reaction, which is not observed with the native cofactor heme.85 In the MBL superfamily, most enzymes are assumed to incorporate Zn2+ in the active site; however, some have been shown to prefer Fe2+, Ni2+, Mn2+, and Co2+.80 We previously characterized the effect of exchanging the active site metal ions of five enzymes from the MBL superfamily (six different metals and eight different reactions). This systematic analysis revealed that the function profile of these enzymes, and thus their function connectivity, varies significantly across different environments [in this case environments with different metal ion availabilities (Figure 7)].80 Interestingly, individual metal ions display limited connectivity; however, when the activity profiles of all metals are superimposed, all reactions are connected (Figure 7). Although the concentration of metal ions in the cellular milieu is generally regulated, metalloenzymes do not always achieve perfect metalation with a specific, most active metal ion(s), but rather exist as various metal isoforms in the cell.86−91 Also, environmental variation in metal availability or cellular stress can lead to severe mismetalation of metalloenzymes.86−88,92 Thus, cofactor-dependent activity profile changes that were observed in these in vitro studies may also reflect functional variation in the natural environment. To date, there has been no direct evidence to support the idea that environmentally dependent functional promiscuity fosters the expansion of enzyme superfamily diversity. The diversity of cofactor utilities observed in enzymes within a single enzyme superfamily, however, may imply that very evolutionary scenario.93 For example, a recent study by Goldman et al. suggests that the functional diversity of TIM barrel enzymes may stem from the incorporation of different cofactors, including FMN, NAD, NADP, and various metal ions.93 Furthermore, Ahmed et al. described new families of split β-barrel fold enzymes, which display surprising cofactor diversity, including the binding and utilization of F420, FMN, FAD, and heme.48 Interestingly, some of these enzymes promiscuously bind multiple cofactors with considerably high affinity in vitro.48



FUTURE PERSPECTIVES Over the past decade, the integration of high-throughput sequencing, large-scale sequence and evolutionary analysis, and experimental characterizations has allowed new global views of functional diversity for many enzyme superfamilies. However, our understanding of enzyme functions and the evolutionary processes that led to their expansion is far from complete. Many open questions remain, and further efforts are required to develop a better understanding of how this remarkable functional diversity has evolved. Continuous efforts to establish accurate superfamily classifications that are based on exhaustive sequence characterization in conjunction with experimental data are extremely important, and the integration of metagenomic (environmental DNA) surveys is imperative, if we are to explore the unrecognized sequence clusters and potentially important groups of these superfamilies.31,42,95 The development of tools that can accurately predict enzyme function(s) is also essential to support the experimental exploration of sequence space; experimental efforts to characterize functional and structural properties are unlikely to ever meet the exponential rate at which the volume of sequence information continues to grow.96 Furthermore, there are strong biases of sequence, biochemical, and structural information that need to be recognized and accounted for. The focus of many experimental approaches has been generally narrow and traditionally confined to relatively few protein folds and superfamilies, or even isolated to selected functional groups, partly because of biomedical and industrial interests and the experimental assays that are available.97 As a result, the majority of functional families and superfamilies have yet to be fully, or even partially, characterized.37 In addition, the misannotation of gene functions is common in public databases that use automated computational annotation predictions based upon homology with existing entries; a study published by Schnoes et al. observed that one-third of the 37 investigated superfamilies contained misannotation levels of >80%.98 Sequencing bias may also hinder functional annotation and delineation of sequence clusters, as most sequences deposited to genome databases are from relatively few model organisms and isolated microbial strains.97,99 Thus, experimental investigations guided by exhaustive sequence analysis are essential for the effective and comprehensive characterization of enzyme superfamilies. Highthroughput experimental platforms capable of systematically characterizing a large number of sequences would also be highly beneficial.39,100,101



STRUCTURAL FEATURES THAT DETERMINE ACTIVITY PROFILES In general, the members of a single enzyme superfamily share the same mechanistic features in the active site. Similarly, it has been shown that the native and promiscuous activities within a single enzyme are catalyzed in the same active site.94 Elucidating the molecular and structural basis for enzyme chemical and substrate specificity, however, remains extremely challenging. For example, in our previous study of 24 MBL enzymes, we found that many enzymes, despite large differences in active site shape, volume, and hydrophobicity, could catalyze the same reactions, such as the native functions of phosphodiesterases or β-lactamases.15 The only tendency we found was that enzymes that hydrolyze charged substrates as their native function possess more hydrophilic active sites. Bastard et al. also found a structure−function correlation for enzymes of the BKACE superfamily. In particular, they found that enzymes with more hydrophobic and noncharged active sites turn over hydrophobic and noncharged polar substrates. In contrast, enzymes that have negatively or positively charged residues in the active site turn over generally positively and negatively charged substrates, respectively. Members of the HAD superfamily are structurally classified depending on the cap domain above the active site, as being either C0 (minimal 6383

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388

Current Topic

Biochemistry

Moreover, most enzymes are tightly regulated in terms of expression (transcription and translation) and cellular localization (e.g., nuclear, intracellular, and extracellular).119 Several studies have recently shown that expression and localization were co-optimized with function during the course of evolution to ensure that each enzyme appears at the right time and space within the cell.26,120,121 Therefore, molecular evolutionary studies that integrate in vivo parameters, such as spatial and temporal expression or metabolite flux, are of particular interest and will likely provide new insights into the evolutionary dynamics permitting functional diversification beyond enzyme promiscuity.115,122 Many questions regarding the molecular details of enzyme specificity and promiscuity remain unanswered. What are the molecular determinants that differentiate distinct chemical and substrate specificities? What are the molecular distinctions between specialized enzymes and promiscuous enzymes? How many mutations are generally necessary to switch one function to another? What are the molecular mechanisms that underlie functional transitions? Can all promiscuous functions act as a sufficient foundation for the evolution of efficient new enzymes? Some promiscuous activities may not provide a selective advantage simply because their activity level is too low.123 In addition, even if the initial activity is sufficient to provide a selective advantage, all sequences may not be optimal evolutionary starting points, because of genetic and structural constraints.114 Indeed, growing evidence indicates that epistasis (nonadditive interactions between mutations) is prevalent, plays a significant role in shaping fitness landscapes, and constrains evolutionary trajectories.124−126 Seemingly neutral mutations can positively and negatively affect the evolution of a new function by acting as an “epistatic ratchet”, which alters the effect of adaptive mutations and therefore the evolvability of enzymes.114,127,128 For instance, a few recent studies have shown that the same mutations have very different effects on the activity levels of orthologous enzymes, being beneficial for some and detrimental for others.127,129 Deciphering the molecular causes of “evolvability” of enzymes toward new functions is one of the most intriguing aspects of this exploratory process. By characterizing resurrected ancestral sequences and using comparative laboratory evolution experiments, we may yield further insights into the molecular details of epistasis and how it constrains the evolution of enzymes.58,130−133 In summary, we are beginning to understand the evolution of functional diversity in enzyme superfamilies. Utilizing the diverse and powerful tools now available, we can tackle longstanding questions in biology and achieve new insights into the evolutionary process that has created such vast functional diversity. Continuous efforts to explore and integrate highresolution studies (e.g., characterizing the molecular details of evolutionary trajectories) and low-resolution analyses (e.g., superfamily-wide function profiling) will be essential to promoting our understanding of enzyme evolution and enhancing our ability to design and engineer novel enzymes.

In the four comprehensive enzyme characterization studies discussed here, the scope of the functions and substrates that were investigated was relatively narrow compared to the large diversity of functions that can exist in a single superfamily (Figure 1). The question of whether significantly different functions, e.g., oxidoreductases, lyases, and hydrolases, can also diverge from each other in the same manner via promiscuous functions, therefore, remains. A recent study showed that, whereas functional expansion typically occurs within each of the six major EC classes (i.e., oxidoreductase, transferase, hydrolase, lysase, isomerase, and ligase), evolutionary transitions between these classes, while also observed, are far less common.2 Highly distinct chemical reactions can be catalytically exclusive, as the chemical and structural requirements may be incompatible, and thus, it might be rare to observe that a single enzyme catalyzes very distinct functions. However, there are a few anecdotal cases that have been described. For example, a reductive dehalogenase of the cytGST superfamily that is involved in the biodegradation pathway of xenobiotic polychlorinated compounds in Sphingomonas chlorophenolica exhibits high levels of isomerase activity, which is suggested to be its ancestral activity.102 Also, decarboxylases from Escherichia coli and Pseudomonas pavonaceae have been shown to promiscuously function as a synthase103 and hydratases, respectively.104 Interestingly, Yew et al. also showed that the synthase activity of the E. coli decarboxylase could be improved by 260-fold with only four mutations in the active site. Despite these examples, we suggest that in some instances “hopeful monsters” may be required for the emergence of new catalytic functions, as large structural and mechanistic reorganizations, including the insertion of domains with different catalytic properties, are often observed between functionally distant enzymes.105 Indeed, laboratory evolution studies demonstrated that loop and domain insertions and/or deletions can cause the emergence of new catalytic functions, which can be subsequently optimized by further adaptive mutations.29,106 At the same time, a recent work by Aravind and co-workers highlights examples of the drastic structural divergence of proteins, via domain incorporation, topological rewiring, loop modifications, and mutations of catalytic residues, that can occur even when their molecular functions remain conserved.107 Such cases of dramatic structural rearrangements that are neutral for the native function lead us to speculate that proteins may, in some cases, be able to explore new regions of sequence and structural space in which they can catalyze new promiscuous activities without necessarily changing the ancestral function or impairing organismal fitness.107 Another challenge for the functional exploration of enzyme superfamilies is to establish a link between an enzyme’s in vitro activity profile and its in vivo functions.108 For example, to what extent are promiscuous activities associated with cellular functions?109−113 The existence of both native and promiscuous substrates in the same cellular compartment may cause cross-inhibition. Thus, the kinetic balance between the two activities would affect the organism’s fitness and may be particularly relevant during the evolution toward a new function.114,115 In addition, environmental changes could significantly affect the observed in vivo levels of native, as well as promiscuous, activities. For example, organisms consistently face environmental fluctuations such as variation of nutrition and cofactor availability,89,116,117 pH (in particular periplasm and extracellular matrix),118 and temperature,78 thus potentially exposing enzymes to many types of environmental variation.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Funding

This work was supported by a Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant (RGPIN 418262-12). N.T. is a CIHR new investigator 6384

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388

Current Topic

Biochemistry

(16) Mashiyama, S. T., Malabanan, M. M., Akiva, E., Bhosle, R., Branch, M. C., Hillerich, B., Jagessar, K., Kim, J., Patskovsky, Y., Seidel, R. D., Stead, M., Toro, R., Vetting, M. W., Almo, S. C., Armstrong, R. N., and Babbitt, P. C. (2014) Large-Scale Determination of Sequence, Structure, and Function Relationships in Cytosolic Glutathione Transferases across the Biosphere. PLoS Biol. 12, e1001843. (17) Huang, H., Pandya, C., Liu, C., Al-Obaidi, N. F., Wang, M., Zheng, L., Toews Keating, S., Aono, M., Love, J. D., Evans, B., Seidel, R. D., Hillerich, B. S., Garforth, S. J., Almo, S. C., Mariano, P. S., Dunaway-Mariano, D., Allen, K. N., and Farelli, J. D. (2015) Panoramic view of a superfamily of phosphatases through substrate profiling. Proc. Natl. Acad. Sci. U. S. A. 112, E1974−83. (18) Bastard, K., Smith, A. A. T., Vergne-Vaxelaire, C., Perret, A., Zaparucha, A., De Melo-Minardi, R., Mariage, A., Boutard, M., Debard, A., Lechaplais, C., Pelle, C., Pellouin, V., Perchat, N., Petit, J.-L., Kreimeyer, A., Medigue, C., Weissenbach, J., Artiguenave, F., De Berardinis, V., Vallenet, D., and Salanoubat, M. (2013) Revealing the hidden functional diversity of an enzyme family. Nat. Chem. Biol. 10, 42−49. (19) Jacob, F. (1977) Evolution and tinkering. Science 196, 1161− 1166. (20) Jensen, R. A. (1976) Enzyme recruitment in evolution of new function. Annu. Rev. Microbiol. 30, 409−425. (21) Innan, H., and Kondrashov, F. (2010) The evolution of gene duplications: classifying and distinguishing between models. Nat. Rev. Genet. 11, 97−108. (22) Voordeckers, K., Brown, C. A., Vanneste, K., van der Zande, E., Voet, A., Maere, S., and Verstrepen, K. J. (2012) Reconstruction of Ancestral Metabolic Enzymes Reveals Molecular Mechanisms Underlying Evolutionary Innovation through Gene Duplication. PLoS Biol. 10, e1001446. (23) Noor, S., Taylor, M. C., Russell, R. J., Jermiin, L. S., Jackson, C. J., Oakeshott, J. G., and Scott, C. (2012) Intramolecular Epistasis and the Evolution of a New Enzymatic Function. PLoS One 7, e39822. (24) Mohamed, M. F., and Hollfelder, F. (2013) Efficient, crosswise catalytic promiscuity among enzymes that catalyze phosphoryl transfer. Biochim. Biophys. Acta, Proteins Proteomics 1834, 417−424. (25) Copley, S. D. (2009) Evolution of efficient pathways for degradation of anthropogenic chemicals. Nat. Chem. Biol. 5, 559−566. (26) Ngaki, M. N., Louie, G. V., Philippe, R. N., Manning, G., Pojer, F., Bowman, M. E., Li, L., Larsen, E., Wurtele, E. S., and Noel, J. P. (2012) Evolution of the chalcone-isomerase fold from fatty-acid binding to stereospecific catalysis. Nature 485, 530−533. (27) Huang, R., Hippauf, F., Rohrbeck, D., Haustein, M., Wenke, K., Feike, J., Sorrelle, N., Piechulla, B., and Barkman, T. J. (2012) Enzyme functional evolution through improved catalysis of ancestrally nonpreferred substrates. Proc. Natl. Acad. Sci. U. S. A. 109, 2966−2971. (28) Seffernick, J. L., de Souza, M. L., Sadowsky, M. J., and Wackett, L. P. (2001) Melamine deaminase and atrazine chlorohydrolase: 98% identical but functionally different. J. Bacteriol. 183, 2405−2410. (29) Afriat-Jurnou, L., Jackson, C. J., and Tawfik, D. S. (2012) Reconstructing a Missing Link in the Evolution of a Recently Diverged Phosphotriesterase by Active-Site Loop Remodeling. Biochemistry 51, 6047−6055. (30) Brown, S. D., and Babbitt, P. C. (2012) Inference of functional properties from large-scale analysis of enzyme superfamilies. J. Biol. Chem. 287, 35−42. (31) Akiva, E., Brown, S., Almonacid, D. E., Barber, A. E., Custer, A. F., Hicks, M. A., Huang, C. C., Lauck, F., Mashiyama, S. T., Meng, E. C., Mischel, D., Morris, J. H., Ojha, S., Schnoes, A. M., Stryke, D., Yunes, J. M., Ferrin, T. E., Holliday, G. L., and Babbitt, P. C. (2014) The Structure-Function Linkage Database. Nucleic Acids Res. 42, D521−30. (32) Dunwell, J. M., Culham, A., Carter, C. E., Sosa-Aguirre, C. R., and Goodenough, P. W. (2001) Evolution of functional diversity in the cupin superfamily. Trends Biochem. Sci. 26, 740−746. (33) Punta, M., Coggill, P. C., Eberhardt, R. Y., Mistry, J., Tate, J., Boursnell, C., Pang, N., Forslund, K., Ceric, G., Clements, J., Heger, A., Holm, L., Sonnhammer, E. L. L., Eddy, S. R., Bateman, A., and Finn, R.

and a Michael Smith Foundation of Health Research (MSFHR) career investigator. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank members of the Tokuriki lab for comments on the manuscript and in particular Dave W. Anderson for editing the manuscript.



ABBREVIATIONS SSN, sequence similarity network; FCN, function connectivity network; MBL, metallo-β-lactamase; HAD, haloacid dehalogenase; cytGST, cytosolic glutathione transferase; BKACE, β-keto acid cleavage enzyme; AMPS, alpha-, mu-, pi-, and sigma-like.



REFERENCES

(1) Blow, D. (2000) So do we understand how enzymes work? Structure 8, R77−81. (2) Furnham, N., Dawson, N. L., Rahman, S. A., Thornton, J. M., and Orengo, C. A. (2016) Large-Scale Analysis Exploring Evolution of Catalytic Machineries and Mechanisms in Enzyme Superfamilies. J. Mol. Biol. 428, 253−267. (3) Gerlt, J. A., Bouvier, J. T., Davidson, D. B., Imker, H. J., Sadkhin, B., Slater, D. R., and Whalen, K. L. (2015) Enzyme Function InitiativeEnzyme Similarity Tool (EFI-EST): A web tool for generating protein sequence similarity networks. Biochim. Biophys. Acta, Proteins Proteomics 1854, 1019−1037. (4) Glasner, M. E., Gerlt, J. A., and Babbitt, P. C. (2006) Evolution of enzyme superfamilies. Curr. Opin. Chem. Biol. 10, 492−497. (5) Gerlt, J. A., Babbitt, P. C., Jacobson, M. P., and Almo, S. C. (2012) Divergent Evolution in Enolase Superfamily: Strategies for Assigning Functions. J. Biol. Chem. 287, 29−34. (6) Furnham, N., Sillitoe, I., Holliday, G. L., Cuff, A. L., Laskowski, R. A., Orengo, C. A., and Thornton, J. M. (2012) Exploring the evolution of novel enzyme functions within structurally defined protein superfamilies. PLoS Comput. Biol. 8, e1002403. (7) Seibert, C. M., and Raushel, F. M. (2005) Structural and Catalytic Diversity within the Amidohydrolase Superfamily. Biochemistry 44, 6383−6391. (8) Das, S., Dawson, N. L., and Orengo, C. A. (2015) Diversity in protein domain superfamilies. Curr. Opin. Genet. Dev. 35, 40−49. (9) Todd, A. E., Orengo, C. A., and Thornton, J. M. (2001) Evolution of function in protein superfamilies, from a structural perspective. J. Mol. Biol. 307, 1113−1143. (10) Aloy, P., Oliva, B., Querol, E., Aviles, F. X., and Russell, R. B. (2002) Structural similarity to link sequence space: new potential superfamilies and implications for structural genomics. Protein Sci. 11, 1101−1116. (11) Meng, E. C., and Babbitt, P. C. (2011) Topological variation in the evolution of new reactions in functionally diverse enzyme superfamilies. Curr. Opin. Struct. Biol. 21, 391−397. (12) Farías-Rico, J. A., Schmidt, S., and Höcker, B. (2014) Evolutionary relationship of two ancient protein superfolds. Nat. Chem. Biol. 10, 710−715. (13) Sillitoe, I., Lewis, T. E., Cuff, A., Das, S., Ashford, P., Dawson, N. L., Furnham, N., Laskowski, R. A., Lee, D., Lees, J. G., Lehtinen, S., Studer, R. A., Thornton, J., and Orengo, C. A. (2015) CATH: comprehensive structural and functional annotations for genome sequences. Nucleic Acids Res. 43, D376−81. (14) Brown, S. D., Gerlt, J. A., Seffernick, J. L., and Babbitt, P. C. (2006) A gold standard set of mechanistically diverse enzyme superfamilies. Genome Biol. 7, R8. (15) Baier, F., and Tokuriki, N. (2014) Connectivity between catalytic landscapes of the metallo-β-lactamase superfamily. J. Mol. Biol. 426, 2442−2456. 6385

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388

Current Topic

Biochemistry D. (2012) The Pfam protein families database. Nucleic Acids Res. 40, D290−D301. (34) Atkinson, H. J., Morris, J. H., Ferrin, T. E., and Babbitt, P. C. (2009) Using Sequence Similarity Networks for Visualization of Relationships Across Diverse Protein Superfamilies. PLoS One 4, e4345. (35) Barber, A. E., and Babbitt, P. C. (2012) Pythoscape: a framework for generation of large protein similarity networks. Bioinformatics 28, 2845−2846. (36) Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. (1990) Basic local alignment search tool. J. Mol. Biol. 215, 403− 410. (37) Brown, S. D., and Babbitt, P. C. (2014) New Insights about Enzyme Evolution from Large Scale Studies of Sequence and Structure Relationships. J. Biol. Chem. 289, 30221−30228. (38) Cravatt, B. F., Wright, A. T., and Kozarich, J. W. (2008) Activity-based protein profiling: from enzyme chemistry to proteomic chemistry. Annu. Rev. Biochem. 77, 383−414. (39) Davids, T., Schmidt, M., Böttcher, D., and Bornscheuer, U. T. (2013) Strategies for the discovery and engineering of enzymes for biocatalysis. Curr. Opin. Chem. Biol. 17, 215−220. (40) Prosser, G. A., Larrouy-Maumus, G., and de Carvalho, L. P. S. (2014) Metabolomic strategies for the identification of new enzyme functions and metabolic pathways. EMBO Rep. 15, 657−669. (41) Carpenter, A. E., and Sabatini, D. M. (2004) Systematic genome-wide screens of gene function. Nat. Rev. Genet. 5, 11−22. (42) Gerlt, J. A., Allen, K. N., Almo, S. C., Armstrong, R. N., Babbitt, P. C., Cronan, J. E., Dunaway-Mariano, D., Imker, H. J., Jacobson, M. P., Minor, W., Poulter, C. D., Raushel, F. M., Sali, A., Shoichet, B. K., and Sweedler, J. V. (2011) The Enzyme Function Initiative. Biochemistry 50, 9950−9962. (43) Pieper, U., Chiang, R., Seffernick, J. J., Brown, S. D., Glasner, M. E., Kelly, L., Eswar, N., Sauder, J. M., Bonanno, J. B., Swaminathan, S., Burley, S. K., Zheng, X., Chance, M. R., Almo, S. C., Gerlt, J. A., Raushel, F. M., Jacobson, M. P., Babbitt, P. C., and Sali, A. (2009) Target selection and annotation for the structural genomics of the amidohydrolase and enolase superfamilies. J. Struct. Funct. Genomics 10, 107−125. (44) Lukk, T., Sakai, A., Kalyanaraman, C., Brown, S. D., Imker, H. J., Song, L., Fedorov, A. A., Fedorov, E. V., Toro, R., Hillerich, B., Seidel, R., Patskovsky, Y., Vetting, M. W., Nair, S. K., Babbitt, P. C., Almo, S. C., Gerlt, J. A., and Jacobson, M. P. (2012) Homology models guide discovery of diverse enzyme specificities among dipeptide epimerases in the enolase superfamily. Proc. Natl. Acad. Sci. U. S. A. 109, 4122− 4127. (45) Korbel, J. O., Jensen, L. J., von Mering, C., and Bork, P. (2004) Analysis of genomic context: prediction of functional associations from conserved bidirectionally transcribed gene pairs. Nat. Biotechnol. 22, 911−917. (46) Zhao, S., Kumar, R., Sakai, A., Vetting, M. W., Wood, B. M., Brown, S., Bonanno, J. B., Hillerich, B. S., Seidel, R. D., Babbitt, P. C., Almo, S. C., Sweedler, J. V., Gerlt, J. A., Cronan, J. E., and Jacobson, M. P. (2013) Discovery of new enzymes and metabolic pathways by using structure and genome context. Nature 502, 698−702. (47) Hasnain, G., Roje, S., Sa, N., Zallot, R., Ziemak, M. J., de CrécyLagard, V., Gregory, J. F., and Hanson, A. D. (2016) Bacterial and plant HAD enzymes catalyse a missing phosphatase step in thiamin diphosphate biosynthesis. Biochem. J. 473, 157−166. (48) Ahmed, F. H., Carr, P. D., Lee, B. M., Afriat-Jurnou, L., Mohamed, A. E., Hong, N.-S., Flanagan, J., Taylor, M. C., Greening, C., and Jackson, C. J. (2015) Sequence-Structure-Function Classification of a Catalytically Diverse Oxidoreductase Superfamily in Mycobacteria. J. Mol. Biol. 427, 3554−3571. (49) Xiang, D. F., Kolb, P., Fedorov, A. A., Xu, C., Fedorov, E. V., Narindoshivili, T., Williams, H. J., Shoichet, B. K., Almo, S. C., and Raushel, F. M. (2012) Structure-based function discovery of an enzyme for the hydrolysis of phosphorylated sugar lactones. Biochemistry 51, 1762−1773.

(50) Hitchcock, D. S., Fan, H., Kim, J., Vetting, M., Hillerich, B., Seidel, R. D., Almo, S. C., Shoichet, B. K., Sali, A., and Raushel, F. M. (2013) Structure-guided discovery of new deaminase enzymes. J. Am. Chem. Soc. 135, 13927−13933. (51) Kolb, P., Ferreira, R. S., Irwin, J. J., and Shoichet, B. K. (2009) Docking and chemoinformatic screens for new ligands and targets. Curr. Opin. Biotechnol. 20, 429−436. (52) Roodveldt, C., and Tawfik, D. S. (2005) Shared Promiscuous Activities and Evolutionary Features in Various Members of the Amidohydrolase Superfamily. Biochemistry 44, 12728−12736. (53) Weng, J.-K., Philippe, R. N., and Noel, J. P. (2012) The rise of chemodiversity in plants. Science 336, 1667−1670. (54) Copley, S. D. (2012) Moonlighting is mainstream: Paradigm adjustment required. BioEssays 34, 578−588. (55) Tokuriki, N., and Tawfik, D. S. (2009) Protein dynamism and evolvability. Science 324, 203−207. (56) Nobeli, I., Favia, A. D., and Thornton, J. M. (2009) Protein promiscuity and its implications for biotechnology. Nat. Biotechnol. 27, 157−167. (57) Babtie, A., Tokuriki, N., and Hollfelder, F. (2010) What makes an enzyme promiscuous? Curr. Opin. Chem. Biol. 14, 200−207. (58) Renata, H., Wang, Z. J., and Arnold, F. H. (2015) Expanding the enzyme universe: accessing non-natural reactions by mechanismguided directed evolution. Angew. Chem., Int. Ed. 54, 3351−3367. (59) Romero, P. A., and Arnold, F. H. (2009) Exploring protein fitness landscapes by directed evolution. Nat. Rev. Mol. Cell Biol. 10, 866−876. (60) Weng, J.-K. (2014) The evolutionary paths towards complexity: a metabolic perspective. New Phytol. 201, 1141−1149. (61) Bebrone, C. (2007) Metallo-β-lactamases (classification, activity, genetic organization, structure, zinc coordination) and their superfamily. Biochem. Pharmacol. 74, 1686−1701. (62) Daiyasu, H., Osaka, K., Ishino, Y., and Toh, H. (2001) Expansion of the zinc metallo-hydrolase family of the beta-lactamase fold. FEBS Lett. 503, 1−6. (63) Puehringer, S., Metlitzky, M., and Schwarzenbacher, R. (2008) The pyrroloquinoline quinone biosynthesis pathway revisited: a structural approach. BMC Biochem 9, 8. (64) Holdorf, M. M., Owen, H. A., Lieber, S. R., Yuan, L., Adams, N., Dabney-Smith, C., and Makaroff, C. A. (2012) Arabidopsis ETHE1 encodes a sulfur dioxygenase that is essential for embryo and endosperm development. Plant Physiol. 160, 226−236. (65) Silaghi-Dumitrescu, R., Kurtz, D. M., Ljungdahl, L. G., and Lanzilotta, W. N. (2005) X-ray crystal structures of Moorella thermoacetica FprA. Novel diiron site structure and mechanistic insights into a scavenging nitric oxide reductase. Biochemistry 44, 6492−6501. (66) Aravind, L. (1999) An evolutionary classification of the metallobeta-lactamase fold proteins. In Silico Biol. 1, 69−91. (67) Armstrong, R. N. (1997) Structure, catalytic mechanism, and evolution of the glutathione transferases. Chem. Res. Toxicol. 10, 2−18. (68) Bellinzoni, M., Bastard, K., Perret, A., Zaparucha, A., Perchat, N., Vergne, C., Wagner, T., de Melo-Minardi, R. C., Artiguenave, F., Cohen, G. N., Weissenbach, J., Salanoubat, M., and Alzari, P. M. (2011) 3-Keto-5-aminohexanoate cleavage enzyme: a common fold for an uncommon Claisen-type condensation. J. Biol. Chem. 286, 27399− 27405. (69) Burroughs, A. M., Allen, K. N., Dunaway-Mariano, D., and Aravind, L. (2006) Evolutionary genomics of the HAD superfamily: understanding the structural adaptations and catalytic diversity in a superfamily of phosphoesterases and allied enzymes. J. Mol. Biol. 361, 1003−1034. (70) Makarova, K. S., and Grishin, N. V. (1999) The Zn-peptidase superfamily: functional convergence after evolutionary divergence. J. Mol. Biol. 292, 11−17. (71) Gabaldón, T., and Koonin, E. V. (2013) Functional and evolutionary implications of gene orthology. Nat. Rev. Genet. 14, 360− 366. 6386

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388

Current Topic

Biochemistry (72) Wagner, A. (2008) Neutralism and selectionism: a networkbased reconciliation. Nat. Rev. Genet. 9, 965−974. (73) Paaby, A. B., and Rockman, M. V. (2014) Cryptic genetic variation: evolution’s hidden substrate. Nat. Rev. Genet. 15, 247−258. (74) Masel, J., and Trotter, M. V. (2010) Robustness and evolvability. Trends Genet. 26, 406−414. (75) Amitai, G., Gupta, R. D., and Tawfik, D. S. (2007) Latent evolutionary potentials under the neutral mutational drift of an enzyme. HFSP J. 1, 67−78. (76) Bloom, J. D., Romero, P. A., Lu, Z., and Arnold, F. H. (2007) Neutral genetic drift can alter promiscuous protein functions, potentially aiding functional evolution. Biol. Direct 2, 17. (77) Hult, K., and Berglund, P. (2007) Enzyme promiscuity: mechanism and applications. Trends Biotechnol. 25, 231−238. (78) Lutz, S., Lichter, J., and Liu, L. (2007) Exploiting temperaturedependent substrate promiscuity for nucleoside analogue activation by thymidine kinase from Thermotoga maritima. J. Am. Chem. Soc. 129, 8714−8715. (79) Sánchez-Moreno, I., Iturrate, L., Martín-Hoyos, R., Jimeno, M. L., Mena, M., Bastida, A., and García-Junceda, E. (2009) From kinase to cyclase: an unusual example of catalytic promiscuity modulated by metal switching. ChemBioChem 10, 225−229. (80) Baier, F., Chen, J., Solomonson, M., Strynadka, N. C. J., and Tokuriki, N. (2015) Distinct Metal Isoforms Underlie Promiscuous Activity Profiles of Metalloenzymes. ACS Chem. Biol. 10, 1684−1693. (81) Nielsen, M. M., Suits, M. D. L., Yang, M., Barry, C. S., MartinezFleites, C., Tailford, L. E., Flint, J. E., Dumon, C., Davis, B. G., Gilbert, H. J., and Davies, G. J. (2011) Substrate and metal ion promiscuity in mannosylglycerate synthase. J. Biol. Chem. 286, 15155−15164. (82) Badarau, A., and Page, M. I. (2006) The variation of catalytic efficiency of Bacillus cereus metallo-beta-lactamase with different active site metal ions. Biochemistry 45, 10654−10666. (83) Dimitrov, J. D., and Vassilev, T. L. (2009) Cofactor-mediated protein promiscuity. Nat. Biotechnol. 27, 892. (84) Lapalikar, G. V., Taylor, M. C., Warden, A. C., Onagi, H., Hennessy, J. E., Mulder, R. J., Scott, C., Brown, S. E., Russell, R. J., Easton, C. J., and Oakeshott, J. G. (2012) Cofactor promiscuity among F420-dependent reductases enables them to catalyse both oxidation and reduction of the same substrate. Catal. Sci. Technol. 2, 1560. (85) Reynolds, E. W., McHenry, M. W., Cannac, F., Gober, J. G., Snow, C. D., and Brustad, E. M. (2016) An evolved orthogonal enzyme/cofactor pair. J. Am. Chem. Soc. 138, 12451−12458. (86) Carter, E. L., Tronrud, D. E., Taber, S. R., Karplus, P. A., and Hausinger, R. P. (2011) Iron-containing urease in a pathogenic bacterium. Proc. Natl. Acad. Sci. U. S. A. 108, 13095−13099. (87) Xu, Y., Feng, L., Jeffrey, P. D., Shi, Y., and Morel, F. M. M. (2008) Structure and metal exchange in the cadmium carbonic anhydrase of marine diatoms. Nature 452, 56−61. (88) Culotta, V. C., Yang, M., and O’Halloran, T. V. (2006) Activation of superoxide dismutases: putting the metal to the pedal. Biochim. Biophys. Acta, Mol. Cell Res. 1763, 747−758. (89) Foster, A. W., Osman, D., and Robinson, N. J. (2014) Metal Preferences and Metallation. J. Biol. Chem. 289, 28095−28103. (90) Clugston, S., Yajima, R., and Honek, J. (2004) Investigation of metal binding and activation of Escherichia coli glyoxalase I: kinetic, thermodynamic and mutagenesis studies. Biochem. J. 377, 309−316. (91) Waldron, K. J., and Robinson, N. J. (2009) How do bacterial cells ensure that metalloproteins get the correct metal? Nat. Rev. Microbiol. 7, 25−35. (92) Imlay, J. A. (2014) The mismetallation of enzymes during oxidative stress. J. Biol. Chem. 289, 28121−28128. (93) Goldman, A. D., Beatty, J. T., and Landweber, L. F. (2016) The TIM Barrel Architecture Facilitated the Early Evolution of ProteinMediated Metabolism. J. Mol. Evol. 82, 17−26. (94) Tokuriki, N., Jackson, C. J., Afriat-Jurnou, L., Wyganowski, K. T., Tang, R., and Tawfik, D. S. (2012) Diminishing returns and tradeoffs constrain the laboratory optimization of an enzyme. Nat. Commun. 3, 1257.

(95) Cantarel, B. L., Coutinho, P. M., Rancurel, C., Bernard, T., Lombard, V., and Henrissat, B. (2009) The Carbohydrate-Active EnZymes database (CAZy): an expert resource for Glycogenomics. Nucleic Acids Res. 37, D233−8. (96) Radivojac, P., Clark, W. T., Oron, T. R., Schnoes, A. M., Wittkop, T., Sokolov, A., Graim, K., Funk, C., Verspoor, K., Ben-Hur, A., Pandey, G., Yunes, J. M., Talwalkar, A. S., Repo, S., Souza, M. L., Piovesan, D., Casadio, R., Wang, Z., Cheng, J., Fang, H., Gough, J., Koskinen, P., Törönen, P., Nokso-Koivisto, J., Holm, L., Cozzetto, D., Buchan, D. W. A., Bryson, K., Jones, D. T., Limaye, B., Inamdar, H., Datta, A., Manjari, S. K., Joshi, R., Chitale, M., Kihara, D., Lisewski, A. M., Erdin, S., Venner, E., Lichtarge, O., Rentzsch, R., Yang, H., Romero, A. E., Bhat, P., Paccanaro, A., Hamp, T., Kaßner, R., Seemayer, S., Vicedo, E., Schaefer, C., Achten, D., Auer, F., Boehm, A., Braun, T., Hecht, M., Heron, M., Hönigschmid, P., Hopf, T. A., Kaufmann, S., Kiening, M., Krompass, D., Landerer, C., Mahlich, Y., Roos, M., Björne, J., Salakoski, T., Wong, A., Shatkay, H., Gatzmann, F., Sommer, I., Wass, M. N., Sternberg, M. J. E., Škunca, N., Supek, F., Bošnjak, M., Panov, P., Džeroski, S., Šmuc, T., Kourmpetis, Y. A. I., van Dijk, A. D. J., Braak, C. J. F. t., Zhou, Y., Gong, Q., Dong, X., Tian, W., Falda, M., Fontana, P., Lavezzo, E., Di Camillo, B., Toppo, S., Lan, L., Djuric, N., Guo, Y., Vucetic, S., Bairoch, A., Linial, M., Babbitt, P. C., Brenner, S. E., Orengo, C., Rost, B., Mooney, S. D., and Friedberg, I. (2013) A large-scale evaluation of computational protein function prediction. Nat. Methods 10, 221−227. (97) Schnoes, A. M., Ream, D. C., Thorman, A. W., Babbitt, P. C., and Friedberg, I. (2013) Biases in the Experimental Annotations of Protein Function and Their Effect on Our Understanding of Protein Function Space. PLoS Comput. Biol. 9, e1003063. (98) Schnoes, A. M., Brown, S. D., Dodevski, I., and Babbitt, P. C. (2009) Annotation error in public databases: misannotation of molecular function in enzyme superfamilies. PLoS Comput. Biol. 5, e1000605. (99) Hug, L. A., Baker, B. J., Anantharaman, K., Brown, C. T., Probst, A. J., Castelle, C. J., Butterfield, C. N., Hernsdorf, A. W., Amano, Y., Ise, K., Suzuki, Y., Dudek, N., Relman, D. A., Finstad, K. M., Amundson, R., Thomas, B. C., and Banfield, J. F. (2016) A new view of the tree of life. Nature 1, 16048. (100) Colin, P.-Y., Kintses, B., Gielen, F., Miton, C. M., Fischer, G., Mohamed, M. F., Hyvönen, M., Morgavi, D. P., Janssen, D. B., and Hollfelder, F. (2015) Ultrahigh-throughput discovery of promiscuous enzymes by picodroplet functional metagenomics. Nat. Commun. 6, 10008. (101) Bachovchin, D. A., Koblan, L. W., Wu, W., Liu, Y., Li, Y., Zhao, P., Woznica, I., Shu, Y., Lai, J. H., Poplawski, S. E., Kiritsy, C. P., Healey, S. E., DiMare, M., Sanford, D. G., Munford, R. S., Bachovchin, W. W., and Golub, T. R. (2014) A high-throughput, multiplexed assay for superfamily-wide profiling of enzyme activity. Nat. Chem. Biol. 10, 656−663. (102) Anandarajah, K., Kiefer, P. M., Donohoe, B. S., and Copley, S. D. (2000) Recruitment of a double bond isomerase to serve as a reductive dehalogenase during biodegradation of pentachlorophenol. Biochemistry 39, 5303−5311. (103) Yew, W. S., Akana, J., Wise, E. L., Rayment, I., and Gerlt, J. A. (2005) Evolution of enzymatic activities in the orotidine 5′monophosphate decarboxylase suprafamily: enhancing the promiscuous D-arabino-hex-3-ulose 6-phosphate synthase reaction catalyzed by 3-keto-L-gulonate 6-phosphate decarboxylase. Biochemistry 44, 1807− 1815. (104) Poelarends, G. J., Serrano, H., Johnson, W. H., Hoffman, D. W., and Whitman, C. P. (2004) The hydratase activity of malonate semialdehyde decarboxylase: mechanistic and evolutionary implications. J. Am. Chem. Soc. 126, 15658−15659. (105) Toth-Petroczy, A., and Tawfik, D. S. (2014) Hopeful (protein InDel) monsters? Structure 22, 803−804. (106) Park, H. S. (2006) Design and Evolution of New Catalytic Activity with an Existing Protein Scaffold. Science 311, 535−538. 6387

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388

Current Topic

Biochemistry (107) Zhang, D., Iyer, L. M., Burroughs, A. M., and Aravind, L. (2014) Resilience of biochemical activity in protein domains in the face of structural divergence. Curr. Opin. Struct. Biol. 26, 92−103. (108) Davidi, D., Noor, E., Liebermeister, W., Bar-Even, A., Flamholz, A., Tummler, K., Barenholz, U., Goldenfeld, M., Shlomi, T., and Milo, R. (2016) Global characterization of in vivo enzyme catalytic rates and their correspondence to in vitro kcat measurements. Proc. Natl. Acad. Sci. U. S. A. 113, 3401−3406. (109) Oberhardt, M. A., Zarecki, R., Reshef, L., Xia, F., DuranFrigola, M., Schreiber, R., Henry, C. S., Ben-Tal, N., Dwyer, D. J., Gophna, U., and Ruppin, E. (2016) Systems-Wide Prediction of Enzyme Promiscuity Reveals a New Underground Alternative Route for Pyridoxal 5′-Phosphate Production in E. coli. PLoS Comput. Biol. 12, e1004705. (110) Soo, V. W. C., Hanson-Manful, P., and Patrick, W. M. (2011) Artificial gene amplification reveals an abundance of promiscuous resistance determinants in Escherichia coli. Proc. Natl. Acad. Sci. U. S. A. 108, 1484−1489. (111) Piedrafita, G., Keller, M. A., and Ralser, M. (2015) The Impact of Non-Enzymatic Reactions and Enzyme Promiscuity on Cellular Metabolism during (Oxidative) Stress Conditions. Biomolecules 5, 2101−2122. (112) Guzmán, G. I., Utrilla, J., Nurk, S., Brunk, E., Monk, J. M., Ebrahim, A., Palsson, B. O., and Feist, A. M. (2015) Model-driven discovery of underground metabolic functions in Escherichia coli. Proc. Natl. Acad. Sci. U. S. A. 112, 929−934. (113) Notebaart, R. A., Szappanos, B., Kintses, B., Pál, F., Györkei, Á ., Bogos, B., Lázár, V., Spohn, R., Csörgő , B., Wagner, A., Ruppin, E., Pál, C., and Papp, B. (2014) Network-level architecture and the evolutionary potential of underground metabolism. Proc. Natl. Acad. Sci. U. S. A. 111, 11762−11767. (114) Khanal, A., Yu McLoughlin, S., Kershner, J. P., and Copley, S. D. (2015) Differential effects of a mutation on the normal and promiscuous activities of orthologs: implications for natural and directed evolution. Mol. Biol. Evol. 32, 100−108. (115) Copley, S. D. (2012) Toward a Systems Biology Perspective on Enzyme Evolution. J. Biol. Chem. 287, 3−10. (116) Cotruvo, J. A., Jr., and Stubbe, J. (2012) Metallation and mismetallation of iron and manganese proteins in vitro and in vivo: the class I ribonucleotide reductases as a case study. Metallomics 4, 1020− 1036. (117) Ellington, A. D., and Bull, J. J. (2005) Evolution. Changing the cofactor diet of an enzyme. Science 310, 454−455. (118) Wilks, J. C., and Slonczewski, J. L. (2007) pH of the cytoplasm and periplasm of Escherichia coli: rapid measurement by green fluorescent protein fluorimetry. J. Bacteriol. 189, 5601−5607. (119) Dixon, D. P., Hawkins, T., Hussey, P. J., and Edwards, R. (2009) Enzyme activities and subcellular localization of members of the Arabidopsis glutathione transferase superfamily. J. Exp. Bot. 60, 1207−1218. (120) Pougach, K., Voet, A., Kondrashov, F. A., Voordeckers, K., Christiaens, J. F., Baying, B., Benes, V., Sakai, R., Aerts, J., Zhu, B., Van Dijck, P., and Verstrepen, K. J. (2014) Duplication of a promiscuous transcription factor drives the emergence of a new regulatory network. Nat. Commun. 5, 4868. (121) Voordeckers, K., Pougach, K., and Verstrepen, K. J. (2015) How do regulatory networks evolve and expand throughout evolution? Curr. Opin. Biotechnol. 34, 180−188. (122) Nam, H., Lewis, N. E., Lerman, J. A., Lee, D. H., Chang, R. L., Kim, D., and Palsson, B. O. (2012) Network Context and Selection in the Evolution to Enzyme Specificity. Science 337, 1101−1104. (123) McLoughlin, S. Y., and Copley, S. D. (2008) A compromise required by gene sharing enables survival: Implications for evolution of new enzyme activities. Proc. Natl. Acad. Sci. U. S. A. 105, 13497− 13502. (124) DePristo, M. A., Weinreich, D. M., and Hartl, D. L. (2005) Missense meanderings in sequence space: a biophysical view of protein evolution. Nat. Rev. Genet. 6, 678−687.

(125) Kaltenbach, M., and Tokuriki, N. (2014) Dynamics and constraints of enzyme evolution. J. Exp. Zool., Part B 322, 468−487. (126) Miton, C. M., and Tokuriki, N. (2016) How mutational epistasis impairs predictability in protein evolution and design. Protein Sci. 25, 1260−1272. (127) Lunzer, M., Golding, G. B., and Dean, A. M. (2010) Pervasive cryptic epistasis in molecular evolution. PLoS Genet. 6, e1001162. (128) Bridgham, J. T., Ortlund, E. A., and Thornton, J. W. (2009) An epistatic ratchet constrains the direction of glucocorticoid receptor evolution. Nature 461, 515−519. (129) Parera, M., and Martinez, M. A. (2014) Strong Epistatic Interactions within a Single Protein. Mol. Biol. Evol. 31, 1546−1553. (130) Harms, M. J., and Thornton, J. W. (2010) Analyzing protein structure and function using ancestral gene reconstruction. Curr. Opin. Struct. Biol. 20, 360−366. (131) Harms, M. J., and Thornton, J. W. (2013) Evolutionary biochemistry: revealing the historical and physical causes of protein properties. Nat. Rev. Genet. 14, 559−571. (132) Khersonsky, O., and Tawfik, D. S. (2010) Enzyme Promiscuity: A Mechanistic and Evolutionary Perspective. Annu. Rev. Biochem. 79, 471−505. (133) Hartl, D. L. (2014) What can we learn from fitness landscapes? Curr. Opin. Microbiol. 21, 51−57. (134) Shannon, P., Markiel, A., Ozier, O., Baliga, N. S., Wang, J. T., Ramage, D., Amin, N., Schwikowski, B., and Ideker, T. (2003) Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 13, 2498−2504.

6388

DOI: 10.1021/acs.biochem.6b00723 Biochemistry 2016, 55, 6375−6388