Aspartyl-tRNA Synthetase
Created by Brian Ciomei
DNA translation is one of the three processes that comprise the central dogma of molecular biology. Ribosomes receive messenger RNA produced during transcription and translate it to an amino acid sequence, which will form into a protein. Various transfer RNAs deliver the correct amino acid to the ribosome which is determined by its anticodon. Aspartyl-tRNA synthetase (pdb ID=1B8A) belongs to a group of proteins called aminoacyl-tRNA synthetases (aaRSs), which catalyze bonding of an amino acid to its corresponding tRNA. The most important function of these synthetases is the fidelity of the genetic code; aminoacyl-tRNA synthetases only make a mistake one in every ten thousand couplings (1). This coupling is an aminoacylation reaction that occurs in the catalytic domain of the enzyme. The molecular weight of aspartyl-tRNA synthetase (AspRS) from Pyrococcus kodakaraensis is 101,803.03 Da and its isoelectric point is 5.3. Its sequence length is 438 residues, with ATP and Mg ligands (3).
AspRS is a homodimer like most of the other class II aminoacyl-tRNA synthetases (2). The two subunits are connected by a helical hinge domain, comprising residues 100-131 (3,4). One end of this hinge domain is a helix-loop-helix motif, which allows connection to both monomers, while the other end helps stabilize interactions with the tRNA (2). AspRSs may also contain an insertion domain that is specific to either eukaryotes or prokaryotes. Pyrococcus kodakaraensis, however, is an archaeon and does not contain this domain (3). In prokaryotes, the insertion domain is a curved anti-parallel beta-sheet with alpha helices on each side. The insertion domain helps stabilize bonding to the cognate tRNA, and may also be necessary for dimerization (2,4). While these domains are necessary for the domain to function, the catalytic and anti-codon recognition domains are characteristic of aminoacyl-tRNA synthetases.
The anti-codon recognition domain forms the N-terminal domain of the enzyme, and its function is generally the same across all of the aaRSs. As its name implies, the main function of this domain is to specifically bind to the AspRS based partially on the anticodons of the cognate tRNA. The anticodon sequence on each type of tRNA corresponds to one or more codon sequences that are encoded into messenger RNA. In many organisms, there is one type tRNA for each amino acid, each specific type of tRNA delivering the corresponding amino acid to the ribosome unit to form polypeptides. Before delivery, the tRNA must first be “charged” with the correct amino acid, which is the function of the aminoacyl-tRNA synthetases(1).
For correct translation of the genetic code, it is important that the correct cognate amino acid is used to charge a tRNA. aaRSs are very consistent due to the three-dimensional, stereospecific nature of the coupling of the enzyme, tRNA, and amino acids (1). The N-terminal domain of each aaRS is shaped differently enough so that binding of the incorrect tRNA is nearly impossible. This specificity is one reason that drug interaction with an aaRS is not normally feasible; change of one amino acid in the enzyme can completely compromise its function (4).
For aspartyl-tRNA synthetase, there are three main sites of recognition for the aspartic acid tRNA (tRNAAsp). Like most class II aaRSs, this takes advantage of both the unique acceptor stem of the tRNA, where aspartic acid will bind, and the unique anticodon sequence. The discriminator base G73 also helps identify tRNAAsp but is not unique to this tRNA (3,5).
The AspRS anticodon recognition domain is a set of six mostly anti-parallel strands, with exception to the two middle strands, spanning residues 26-163 (3). An alpha helix also reaches between strands two and four. This type of structure is not uncommon in biochemistry, and is also found in staphylococcal nuclease, heat-labile enterotoxin, and verotoxin-1, all of which bind either nucleotides or oligosaccharides (5). AspRS binds nucleotides on tRNAAsp, so this function is consistent. 14 residues directly bond to tRNAAsp, including Phe-127, Gln-138 (both interact with the base U35), Arg-119 (interacts with both U35 and C36), and Glu-188 (G34) (5). The tRNAAsp and AspRS both undergo conformational changes when forming the bound complex, which also helps with specificity. Complexion occurs after the AspRS approaches the tRNAAsp on the variable loop side and interacts with the acceptor stem. This interaction causes bending of the acceptor stem and severe deformation of the anticodon loop, which extends into the anticodon recognition domain (6). Interactions then occur that are unique to this pairing.
While most aaRSs are specific to one tRNA, there are some that are less discriminatory. Some archaeal and bacterial forms of aspartyl-tRNA synthetase, like the variant in P. kodakaraensis, exhibit this lack of specificity. In this case, AspRS will recognize and bind either tRNAAsp or tRNAAsn, the asparagine transfer RNA (3). Usually, this occurs when the organism is lacking one or more aaRSs, in this case asparaginyl-tRNA synthetase (7). These two tRNAs are especially susceptible for two reasons: firstly, all tRNAs share very similar structure; secondly, asparagine is very similar in structure to aspartic acid. Both tRNAs have the same G73 discriminator base, and differ by only one anticodon residue – they are otherwise identical (3). This allows for the formation of an Asp-tRNAAsn complex. Asp-tRNAAsn/Glu-tRNAGln amidotransferase (Asp/Glu-Adt) then converts the misacylated Asp-tRNAAsn to Asn-tRNAAsn, which can be used by the ribosome for polypeptide creation (7).
Several simple changes to the AspRS enzyme have been correlated to this tRNA insensitivity although the comprehensive list is unknown. In yeast AspRS, one loop of the enzyme in particular shows sensitivity to the anticodon sequence for aspartic acid. If the loop is shortened, it loses specificity and will accept the asparagine anticodon as well (3). In Escherichia coli, replacing Leu-31 with histidine inactivates the asparagine discrimination (8). However, this does have consequence for organisms without Asp/Glu-Adt,. AspRS lacking Asp/Asn specificity in these cases can be toxic. In one case, Asp-tRNAAsn is formed, but not transaminated, to Asn-tRNAAsn. This causes an aspartic acid to be attached to the polypeptide instead of the correct asparagine, leading to dysfunction or failure of the protein. Depletion of tRNAAsn due to use by both AspRS and AsnRS can also lead to malformed polypeptides (8).
The second large domain of aspartyl-tRNA synthetase (and the other aaRSs) is the catalytic domain. While the anticodon recognition domain plays a large role in translation fidelity, the catalytic domain is the actual functional site . The core of the active site is composed of a seven-stranded beta-sheet, with six antiparallel strands and an additional parallel strand. The site is divided into “pockets” by insertion domains and alpha helices (9). Before the reaction can occur, the substrates must bind to the active site.
ATP blocks the amino acid binding site, so the amino acid must be bound to the enzyme first. Several key residues mediate binding of the amino acid in this first step. The side chain of Glu-170 binds to the amino group of aspartic acid, causing closure of the “flipping loop” portion of the enzyme. Hydrogen bonding occurs between the carboxylate group of aspartic acid and the side chain of Ser-364, in addition to one of the amino groups of Arg 214 (3). The side chain of the amino acid is also bonded to Gln-303, Arg-485, and Gly-526. The first two of these three residues also form a hydrogen bond and salt bridge network with Glu-344, Lys-306, Gln-307, and Asp-342, which also interacts with aspartic acid (9). After this stable aspartyl-tRNA is formed, ATP can bind to the complex and provide energy for the aminoacylation reaction.
To enhance the stability of the bound complex, the ATP substrate conforms to a U-shape, which is common for class II aaRSs. Hydrogen bonds involving the amino groups of Arg-412 and His-223 bind to a gamma-phosphoryl group of the ATP and hold it in place. Magnesium ions, another AspRS substrate, help maintain the bent shape of the ATP and preserve the stability of the entire structure (3). When the ATP is successfully bound, the aspartic acid becomes blocked into its binding pocket (9).
The aminoacylation reaction is a two-step reaction: first, the amino acid is activated; second, the amino acid is transferred to the tRNA. The alpha-phosphate of the bound ATP is attacked nucleophilically by the alpha-carboxylate group of the amino acid, forming the aspartyl-adenylate intermediate (Asp-AMP). Asp-342 in the AspRS helps increase the nucleophilic character of the carboxylate, allowing this attack to occur. Lys-306, Glu-344, Arg-485, and Gln-303 maintain the position of the carboxyl group, which needs to be moved from its normal location (9). The ATP and aspartic acid molecules are superimposable over their reactionary counterparts in this moved position, separated only by about 2.4 Å (3). Next, the phosphate bond is hydrolyzed and an inorganic pyrophosphate group is released (3,9). Mg2+, an AspRS metal ion ligand, plays a critical role during the first step of this reaction . The magnesium ion facilitates nucleophilic attack by withdrawing electrons away from the phosphate atoms (3). The magnesium ion, as well as Arg-531, then catalyzes the hydrolysis of the phosphate group (3,9). For completeness, it must be noted that Mn2+ is listed as a ligand on the AspRS PBD entry, however manganese is only used for testing purposes in vitro, and is functionally identical to Mg2+, which occurs in vivo (3).
The first step of the reaction leads to a minor conformational change, causing the terminal ribose on the acceptor stem of the tRNA to be positioned above the enzyme complex. The 3’ hydroxyl group of the ribose then commences a nucleophilic attack on the carboxylic group of aspartic acid, by way of a cyclic intermediate. The alpha-phosphate group facilitates this nucleophilic attack; a free oxygen atom from the phosphate draws the hydroxyl proton, allowing the intermediate to form (9). The charged tRNA can then be released by AspRS for function in the ribosome. The overall reaction can be summarized as:
(1) Asp + Atp Asp-AMP + PPi
(2) Asp-AMP + tRNAAsp Asp-tRNAAsp + AMP
Aspartyl-tRNA synthetase and other aminoacyl-tRNA synthetases
There is a significant amount of variation between AspRSs of different organisms, but all perform the same function. The other aaRSs are similar in function, but correspond to a different tRNA – one for each of the 20 basic amino acids. It is helpful to break down the aaRSs into classes and subgroups based on structure. These classes help trace homology and the origins of life, while also identifying aaRSs that may be or may become nonspecific due to mutation.
The most obvious difference is the method by which the aaRS interacts with its cognate tRNA. This difference is the basis of the separation between class I and class II aaRSs. Class I aaRSs approach the tRNA along the minor groove of its acceptor stem. Approaching in this way allows access to the 2’ hydroxyl group of the terminal adenosine of the tRNA. This is due to the common structure of class I synthetases which are bulky and have hydrophobic amino acid specificities (10). The desired target for aminoacylation is the 3’ OH, but this is not possible for class I enzymes. Aminoacylation occurs at the 2’ hydroxyl, and the group is then transferred to the 3’ hydroxyl by way of a transesterification reaction. While initially targeting the 2’ hydroxyl may be the defining property of this class, these aaRSs share many other properties. All of the catalytic domains of class I enzymes are based upon a Rossmann fold , which projects from a beta sheet and is used in nucleotide binding (1,3).
Class II aaRSs are generally less bulky and are characterized by aminoacylation at the 3’ hydroxyl group of the cognate tRNA. These aaRSs approach from the major groove on tRNA, a mirror image of the class I method (10). While all class I enzyme catalytic subunits are built around a Rossmann fold, class II aaRSs have an antiparallel beta sheet surrounded by alpha helices in their catalytic domain and share three common motifs (3,13). In some cases, class II is subdivided even further, with different subgroup arrangements for archaeal and bacterial organisms (10). It is also worth noting that class II contains 10 amino acids, as does class I, which is histologically interesting.
The Dali server is a useful tool for comparing aaRSs in the same class or between classes. After inputting a query protein, the Dali server searches through the protein data bank and compares each item with the 3D structure of the query protein. The results are assigned a Z-score and returned, with a higher Z-score indicating a better match. Unsurprisingly, most of the top results are aspartyl-tRNA synthetases from other organisms (the query AspRS is from P. kodakaraensis), and the next highest are all aminoacyl-tRNA synthetases (11). This reiterates the similarity of structure between the aaRSs.
The top two matches with P. kodakaraensis aspartyl-tRNA synthetase (Z-score: 59.3) as the query protein (that were not AspRS from a different organism) areasparaginyl-tRNA synthetase (PDB ID: 1X54, Z-score: 43.6) and lysyl-tRNA synthetase (PDB ID: 1LYL, Z-score: 30.4) . Based on the analysis done previously, it follows that asparaginyl-tRNA synthetase (AsnRS, from Pyrococcus horikoshii) is the closest match structurally. Simply switching one residue or shortening one loop in the tertiary structure, depending on the organism, allows AspRS to bond tRNAAsn. AsnRS is a class II aaRS and therefore shares many properties, including its subdomain structure, ligands, and metal ions. The PDB lists different ligands, but in practice the ligands are the same. (4S)-2-Methyl-2,4-pentanediol does not interact with AsnRS in vivo, but it is used experimentally to help with crystallography (12). AMP is listed instead of ATP, but this is only a semantic difference. Both are involved in the reaction; the aminoacylation reaction activates the amino acid by reacting with ATP, which forms the Asn-AMP (or Asp-AMP) intermediate (9). The structures of AsnRS and AspRS differ only very slightly, in some cases even less so than between two AspRS from different organisms. Comparing the two enzymes using the PSI-BLAST program from NIH returns an E value of 4e-145. A value of 0 is a perfect match, so between this and the Dali Z value, it is clear the two aaRSs are almost identical.
The case is almost the same with lysyl-tRNA synthetase (LysRS, from E. Coli), with both aaRSs sharing most structural and functional properties. Again, the quantitative Dali and PSI-BLAST values support this notion. The E-value for LysRS compared to AspRS is 5e-130, which is again very close to a perfect match of 0. While most of the two molecules are the same, there is key difference in LysRS which helps maintain tRNALys specificity. Both enzymes bond to their cognate tRNA acceptor stems and use the properties of the stem as an identifier. In LysRS, one of the structural loops in the tRNA binding site has a different conformation which decreases its active surface. This leads to steric effects that maintain binding specificity (13). This particular LysRS enzyme is of additional interest, as it is one of two forms of LysRS formed in E. coli. LysS is normally produced, but under extreme physiological conditions this organism forms this type of enzyme, LysU. The reason for LysU production and the mechanism by which it is regulated are not understood, but it is hypothesized that LysU is more thermostable, and therefore survives better in extreme conditions (13).
No drugs currently exist that target aspartyl-tRNA synthetase, nor are there drugs that target any other aminoacyl-tRNA synthetase. However, they are cited many times as possible drug targets. If successful, aaRSs would provide narrow range and effective targets for anti-protozoan drugs. Their usefulness involves their specificity and the variety of aaRSs. Disruption of as few as one residue in some aaRSs cripples or destroys their function. DNA replication cannot occur without all the required aaRSs, so this would in turn cause death of the organism. Since there are 20 variations of aaRSs, each with slightly different structure among the two classes, this also increases the likelihood of finding a targeted, narrow-range drug (4). Success in this area could potentially create a new type of safer, more effective drug targeted at many varieties of ailments.