Created By: Dylan Campbell

Yeast N-Acetyltransferase (PDB ID: 3W6S) from the Mpr1 gene of Saccharomyces cerevisiae is a transferase enzyme. This protein has a molecular weight of 84,700.60 Da, an isoelectric point of 6.86 (1), and it exists as a homotrimer where the three subunits form a closed triangle. Each subunit has 229 residues and provides about 27,910.93 Da of the total molecular weight (1, 2). Each subunit has 6 α helices and 8 β sheets, which is a typical composition for acetyltransferases (3). The helices tend to form near the surface of the protein and the sheets tend to lie in a plane with each other, cutting directly through the middle of the subunit with some sheets on the ends of the plane forming active sites. Each subunit of the protein has very few residues with sulfur in them because this maintains stability of the molecule even in highly toxic conditions (4). In each subunit, there is only one methionine residue and two cysteine residues, the latter of which are involved in a disulfide bond within each subunit.

For crystallization, a solution with magnesium ions and hexaethylene glycol was used. The magnesium ion rests over the Asp-26 residue (4). This residue has a carboxylic acid group pointing into the solution which has a negative charge at neutral pH that can easily be stabilized by the magnesium cation. The hexaethylene glycol molecules rest in pockets near the active sites on each of the subunits of the enzyme. These pockets include a Lys-123 residue that is trapped in a nonpolar region (2). This residue has a polar nitrogen side chain that would not usually be favored to fold towards the hydrophobic center of the protein. Hexaethylene glycol encircles the Lys-123 side chain, directing its electronegative oxygen atoms towards the polar lysine tail and its hydrocarbons towards the nonpolar pocket, thus stabilizing the lysine residue within its pocket. X-ray diffraction was used determine the structure of the protein to a resolution of 1.9 Å (4).

Yeast N-acetyltransferase is a very useful protein for Saccharomyces cerevisiae because it protects the yeast cells from damage due to environmental stress that results in production of reactive oxygen species (ROS) (5). Cells naturally produce these ROS during apoptosis. During these processes, superoxide anions, hydrogen peroxide, and other oxygen-free radicals are produced to effectively kill the cell. The addition of ethanol or excessive heating of cells can also cause production of these toxins. Enzymes like yeast N-acetyltransferase provide tolerance for these stress factors by reducing the level of ROS (4). Yeast N-acetyltransferase accomplishes this by making the species less reactive by acetylating the amine group of the substrate, which is present in some ROS. The active site of yeast N-acetyltransferase has multiple critical residues that are responsible for binding to a molecule of acetyl coenzyme A as a source of an acetyl group and for binding to some specific toxic substrate (5).

Look at cis-4-hydroxy-L-proline (CHOP) as an example of a proline-derived ROS. Proline-derived toxins resemble true proline and therefore are able to trick the cell into using them as a replacement for proline in protein production. If new proteins are built with these derivatives instead of proline, the proteins will not be able to fold properly and therefore cells will not be able to grow or reproduce (6). The main interactions include the side chain of Asn-135 and the peptide backbone of Leu-173 and Asn-172 forming hydrogen bonds with the carboxylic acid tail of CHOP. In order to ensure that Leu-173 and Asn-172 are oriented properly to face the active site, the side chains of Asn-135 and Asn-125 also hydrogen bond with the side chain of Asn-172. This twists the side chains of Leu-173 and Asn-172 outwards from the active site, thereby leaving the backbone open to the pocket. By recognizing carboxylic acid groups, the active site of yeast N-acetyltransferase can identify target ROS. Similarly, the nitrogen in the backbone of Phe-138 and in the ring of Trp-185 both form hydrogen bonds with water molecules near the active site. This allows for hydrogen bonding with both the substrate, in this case CHOP, and the acetyl coenzyme A. For the substrate, a hydrogen bond is formed with the cyclic amine and for acetyl coenzyme A, a hydrogen bond forms with the carboxyl group of the acyl group to be cleaved. The hydrogen bonds help to catalyze the reaction and add selectivity because the residues involved help to identify only the toxins with a cyclic amine. Asn-178 also uses its side chain to hydrogen bond with a nearby water molecule so that, at the end of the acetylation reaction, the anionic sulfur group of coenzyme A is protonated and can be released from the enzyme. Until the sulfur atom is protonated, the hydrogen bonding from Asn-178 can also act on the sulfur atom to help stabilize the negative charge (3, Figure 1).

Although yeast N-acetyltransferase does detoxify CHOP, it is highly unlikely that this is the primary substrate for this enzyme. Despite that, the highly conserved nature of the Mpr1 gene in differing species argues that yeast N-acetyltransferase serves a vital role somewhere else (5). It is possible yeast N-acetyltransferase is more of an emergency backup for other antioxidant enzymes that are found in cells. For instance, yeast N-acetyltransferase shows a high affinity for pyrroline-5-carboxylate (P5C) which is found in equilibrium with glutamate-semialdehyde in the mitochondria of cells. Perforations can arise in the mitochondrial walls and P5C can be released into the cell, acting as a toxin (5). Since yeast N-acetyltransferase is found in the cytoplasm of cells and not the mitochondria, this could support the theory that yeast N-acetyltransferase is present to manage emergencies like a P5C leak.

Other aspects of yeast N-acetyltransferase also make this enzyme very suitable for its function. For instance, the sturdy nature of this protein allows it to survive in the very toxic conditions it is meant to rectify. For instance, ROS species can easily oxidize target molecules such as sulfur. If yeast N-acetyltransferase had more residues with sulfur in it, there would be a higher likelihood that, as ROS levels rose, more of the enzyme’s sulfur atoms would be oxidized which could result in poor folding and a loss of function (7). However, each subunit of yeast N-acetyltransferase only has three total sulfur atoms (2). Two of those atoms in each subunit are from cysteine residues that are involved in disulfide bridging with each other. Because the sulfur atoms are already involved in a bridge, oxygen is less likely to bind to them and cause their structure to change (7). In some experiments, Cys-130 and Cys-134 were completely replaced with other residues. Even though this did alter the trimerization of yeast N-acetyltransferase because these cysteines are near the interface between the subunits, function was preserved, which would imply a similar maintenance of function even if these side chains were oxidized (8). The final sulfur is present in Met-228. Because this is so far from the active site, even changing it to methionine sulfoxide would not inactivate of the protein (7).

Further research could also illuminate aspects of yeast N-acetyltransferase’s structure and function that are not yet known. As of now, yeast N-acetyltransferase has been placed in the superfamily gen5-related N-acetyltransferases (GNAT) because all of those enzymes acetylate substrates and are very structurally conserved throughout species. The only region where most of these enzymes tend to differ is at the residues on their tail ends (3). Similar to yeast N-acetyltransferase, these enzymes tend to use water molecules to stabilize charges that are formed in acetylation reactions as well as to help lock substrates into the active site. Another protein in this family is serotonin N-acetyltransferase (SNAT) found in Homo sapiens (PDB ID: 1IB1). This protein is responsible for acetylating serotonin to form melatonin. Serotonin is responsible for making people feel energetic and awake whereas melatonin initiates the sleep cycle every night by making the person feel drowsy. Due to the roles of serotonin and melatonin, SNAT plays an enormous role in maintaining circadian rhythms in the human body (3). While SNAT may not detoxify like yeast N-acetyltransferase did, they use similar processes of acetylation. A key difference between the GNAT superfamily and yeast N-acetyltransferase is that the GNAT proteins tend to have a β-bulge that protrudes into the active site. Because yeast N-acetyltransferase lacks this feature, the active site is able to accept larger molecules without too much steric interference (3). This would explain why yeast N-acetyltransferase is one of the few enzymes in that family that acetylates proline derivatives directly on their ring structures (4). Most of the other GNAT proteins acetylate a nitrogen on the tail end of a straight chain, making it easier to fit into tight active sites (3).

Other relatives of yeast N-acetyltransferase were found using structural comparison servers. Two of these servers that are popular are Dali and PSI-BLAST. Dali compares proteins based upon differences in intramolecular distances and similarities in tertiary structures. A Z score is generated, and any Z score greater than 2 shows that the proteins that were analyzed had very similar tertiary structures. PSI-BLAST works by comparing the primary structure, or sequence, of multiple proteins. The sequences are aligned and gaps are added where needed to make sure the strands line up properly. The similarity of the two sequences is then determined and written as an E value. E values smaller than 0.05 are deemed significant, therefore the proteins must have similar sequences. A good comparison protein for yeast N-acetyltransferase is bacterial N-acetyltransferase (PDB ID: 4H89) from Kribbella flavida. This new protein had a Z score of 19.2 and an E score of 6e-62, which are both significant values for comparison (9, 10). The gene Gcn5 encodes bacterial N-acetyltransferase, and the protein itself is 59 residues shorter than yeast N-acetyltransferase while also showing no binding sites for ligands. The bacterial enzyme also does not tend to trimerize. However, the general tertiary and secondary structures of the bacterial protein are very similar to those of the yeast protein. Just like other acetyltransferases, the bacterial version shows α-helixes on the periphery of the molecule with a plane of β-sheets extending from one side to the other and slightly bent on either side (2). This provides some similarities in the active site as well, including the presence of Asn-131 and Trp-138 residues, both of which were instrumental in binding acetyl coenzyme A for the acetyltransferase reaction in yeast. While the bacterial protein may have been slightly modified to fit into the environment of the bacterial cell, most of the function of this protein is probably similar to that of the yeast protein and therefore the protein is highly conserved between the species.

These aspects of yeast N-acetyltransferase make it an interesting subject for research. Yeast N-acetyltransferase’s high tolerance for ROS and other environmental stress factors such as high heat and the presence of ethanol would make it perfect for yeast cells expressing Mpr1 to be used in the fermentation process of producing alcohol (4). Engineering of the yeast cells desired for fermentation could improve the overall efficiency of the process and maximize output. Also, there are multiple homologues of the Mpr1 gene coding for N-acetyltransferase present in certain fungi. If this gene was inhibited, it would be much easier to kill fungal cells, which means this route could be very promising for antifungal medications (4).