Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Feb 10:10:e61818.
doi: 10.7554/eLife.61818.

Structure and mechanism of a phage-encoded SAM lyase revises catalytic function of enzyme family

Xiaohu Guo #  1 , Annika Söderholm #  1 , Sandesh Kanchugal P #  1 , Geir V Isaksen #  1   2 , Omar Warsi  3 , Ulrich Eckhard  1 , Silvia Trigüis  1 , Adolf Gogoll  4 , Jon Jerlström-Hultqvist  1   3 , Johan Åqvist  1 , Dan I Andersson  3 , Maria Selmer  1
Affiliations

Structure and mechanism of a phage-encoded SAM lyase revises catalytic function of enzyme family

Xiaohu Guo et al. Elife. .

Abstract

The first S-adenosyl methionine (SAM) degrading enzyme (SAMase) was discovered in bacteriophage T3, as a counter-defense against the bacterial restriction-modification system, and annotated as a SAM hydrolase forming 5'-methyl-thioadenosine (MTA) and L-homoserine. From environmental phages, we recently discovered three SAMases with barely detectable sequence similarity to T3 SAMase and without homology to proteins of known structure. Here, we present the very first phage SAMase structures, in complex with a substrate analogue and the product MTA. The structure shows a trimer of alpha-beta sandwiches similar to the GlnB-like superfamily, with active sites formed at the trimer interfaces. Quantum-mechanical calculations, thin-layer chromatography, and nuclear magnetic resonance spectroscopy demonstrate that this family of enzymes are not hydrolases but lyases forming MTA and L-homoserine lactone in a unimolecular reaction mechanism. Sequence analysis and in vitro and in vivo mutagenesis support that T3 SAMase belongs to the same structural family and utilizes the same reaction mechanism.

Keywords: E. coli; S-adenosyl methionine; bacteriophage; biochemistry; chemical biology; lyase; molecular biophysics; structural biology.

Plain language summary

Bacteria can be infected by viruses known as bacteriophages. These viruses inject their genetic material into bacterial cells and use the bacteria’s own machinery to build the proteins they need to survive and infect other cells. To protect themselves, bacteria produce a molecule called S-adenosyl methionine, or SAM for short, which deposits marks on the bacteria’s DNA. These marks help the bacteria distinguish their own genetic material from the genetic material of foreign invaders: any DNA not bearing the mark from SAM will be immediately broken down by the bacterial cell. This system helps to block many types of bacteriophage infections, but not all. Some bacteriophages carry genes that code for enzymes called SAMases, which can break down SAM, switching off the bacteria’s defenses. The most well-known SAMase was first discovered in the 1960s in a bacteriophage called T3. Chemical studies of this SAMase suggested that it works as a 'hydrolase', meaning that it uses water to break SAM apart. New SAMases have since been discovered in bacteriophages from environmental water samples, which, despite being able to degrade SAM, are genetically dissimilar to one another and the SAMase in T3. This brings into question whether these enzymes all use the same mechanism to break SAM down. To gain a better understanding of how these SAMases work, Guo, Söderholm, Kanchugal, Isaksen et al. solved the crystal structure of one of the newly discovered enzymes called Svi3-3. This revealed three copies of the Svi3-3 enzyme join together to form a unit that SAM binds to at the border between two of the enzymes. Computer simulations of this structure suggested that Svi3-3 holds SAM in a position where it cannot interact with water, and that once in the grip of the SAMase, SAM instead reacts with itself and splits into two. Experiments confirmed these predictions for Svi3-3 and the other tested SAMases. Furthermore, the SAMase from bacteriophage T3 was also found to degrade SAM using the same mechanism. This shows that this group of SAMases are not hydrolases as originally thought, but in fact ‘lyases’: enzymes that break molecules apart without using water. These findings form a starting point for further investigations into how SAM lyases help bacteriophages evade detection. SAM has various different functions in other living organisms, and these lyases could be used to modulate the levels of SAM in future studies investigating its role.

PubMed Disclaimer

Conflict of interest statement

XG, AS, SK, GI, OW, UE, ST, AG, JJ, JÅ, DA, MS No competing interests declared

Figures

Figure 1.
Figure 1.. SAMase reaction.
Top: Hypothetical S-adenosyl methionine (SAM) hydrolase reaction previously suggested to be catalyzed by T3 SAMase. Bottom, green: SAM lyase reaction shown in this study to be catalyzed by all tested bacteriophage SAMases.
Figure 1—figure supplement 1.
Figure 1—figure supplement 1.. Proposed mechanism of rescue of an ilvA auxotrophic mutant by SAMases (Jerlström Hultqvist et al., 2018).
Expression of enzymes in the methionine synthesis pathway is repressed by MetJ in complex with S-adenosyl methionine (SAM). SAM degradation leads to increased expression of the enzymes in the metJ regulon. At high expression level, the promiscuous function of MetB, production of alpha-ketobutyrate from O-succinyl homoserine (green arrow), provides rescue of isoleucine biosynthesis.
Figure 2.
Figure 2.. Structure of SAMase Svi3-3.
(A) Structure of the Svi3-3 trimer in complex with S-adenosyl homocysteine (SAH). (B) Monomer structure of Svi3-3. (C) Interactions of SAH (cyan sticks) at the trimer interface, (D) Fo–Fc omit map for SAH contoured at three sigma. (E) 5’-Methyl-thioadenosine (MTA; cyan sticks) and Pro (green sticks), overlay of SAH is shown in transparent light gray sticks. (F) Unbiased Arcimboldo (Rodríguez et al., 2009) electron density map for MTA bound to Svi3-3, contoured at two sigma (0.39 e-/Å3).
Figure 3.
Figure 3.. Comparison of the apo structure of Svi3-3 (gray) and the 5’-methyl-thioadenosine (MTA) complex structure (colored as in Figure 2A, transparent).
Arrows indicate conformational differences between the N-terminal regions in the two structures.
Figure 3—figure supplement 1.
Figure 3—figure supplement 1.. Size-exclusion chromatography coupled to small angle X-ray scattering (SEC-SAXS) of Svi3-3.
(A) Overlay of signal plots from SEC-SAXS of Svi3-3 in the presence and absence of 5 mM S-adenosyl methionine (SAM; plotted with an offset of 0.05 for clarity). (B and C) Zoom in of signal plot for Svi3-3 apo (B) and Svi3-3 in the presence of SAM (C). Black markers indicate the radii of gyration (Rg) calculated from the individual scattering curves. Gray bars indicate the data frames used for further analysis. (D). Scattering curves for Svi3-3 in the presence and absence of SAM. Colors as in A. The blue curve is for clarity plotted with an offset of +1. (E). Guinier plot for Svi3-3 apo. (F). Guinier plot for Svi3-3 in the presence of SAM.
Figure 4.
Figure 4.. Svi3-3 assays.
(A) Relative enzymatic activity of Svi3-3 variants. The data points are based on technical duplicates from two different protein purifications, where activity is related to wild type (WT) purified at the same time and the average is represented by a black line. (B) Differential scanning fluorimetry (DSF) data for Svi3-3 variants with different concentrations of S-adenosyl methionine (SAM) and S-adenosyl homocysteine (SAH). Error bars correspond to ±1 standard deviation based on triplicate data.
Figure 4—figure supplement 1.
Figure 4—figure supplement 1.. Absolute enzymatic activity of Svi3-3 variants.
Data is from technical duplicates from two different protein purifications (ordered from left to right) with the average represented by a black line. Calculated average activities ±1 standard deviation: wild type (WT): 9.5 ± 1.7 s−1, Y58F: 1.9 ± 0.7 s−1, E69Q: 0.002 ± 0.0006 s−1, E69A: 0.007 ± 0.0016 s−1, E105Q: 3.3 ± 0.99 s−1.
Figure 5.
Figure 5.. Structure-guided sequence alignment of SAMases with demonstrated activity (Jerlström Hultqvist et al., 2018).
Secondary structure of Svi3-3 is displayed above the alignment. Red boxes with white letters indicate conserved residues and red letters in white boxes show conservatively substituted residues. Figure was prepared using ESPript (Gouet et al., 2003).
Figure 6.
Figure 6.. Molecular dynamics (MD) simulations of Svi3-3.
(A) Time evolution of root mean square deviation (RMSD) to the Svi3-3 crystal structure for each monomer during the 100 ns MD simulation in the presence of S-adenosyl methionine (SAM). (B) Average backbone root mean square fluctuation (RMSF) over 100 ns MD simulation of apo Svi3-3 and its complex with SAM. The largest difference is seen for active-site residues 104–106.
Figure 6—figure supplement 1.
Figure 6—figure supplement 1.. Protein–ligand interactions obtained from 100 ns molecular dynamics (MD) simulations illustrating the percentage of interaction from the total simulation time.
Figure 6—figure supplement 2.
Figure 6—figure supplement 2.. Active-site structure of Svi3-3 as observed from MD simulations.
(A) Snapshot from 100 ns molecular dynamics (MD) simulation illustrating the dominant interactions for the methionine part of the ligand in the active site. (B) Density functional theory (DFT) cluster model utilized for the DFT calculations. Atoms marked with an asterisk are kept fixed during the calculations.
Figure 7.
Figure 7.. Density functional theory (DFT) calculations on the S-adenosyl methionine (SAM) lyase reaction of Svi3-3.
(A) Optimized DFT structures for the reactant (left), transition (middle), and product (right) state in the SAM lyase reaction mechanism. (B) Calculated free energy profiles for reactant (1), transition state (1–2) and product state (2) for the SAM lyase reaction mechanism. Further conversion from homoserine lactone (2) to the tetrahedral intermediate (3) that would form homoserine is not supported by the high DFT energies.
Figure 8.
Figure 8.. Representative chromatogram from ion exchange chromatography of decarboxylated S-adenosyl methionine (dcSAM) reactions and controls.
0.32 mM dcSAM mix (77% dcSAM) was incubated ± Svi3-3 for 0–23 hr.
Figure 8—figure supplement 1.
Figure 8—figure supplement 1.. Differential scanning fluorimetry (DSF) duplicate data for Svi3-3 in the presence of different concentrations of decarboxylated S-adenosyl methionine (dcSAM) mix (72% dcSAM, 28% S-adenosyl methionine [SAM] + 5’-methyl-thioadenosine [MTA], orange plus signs) and SAM (open blue squares).
The dcSAM mix Tm values are also plotted as function of the concentration of SAM + MTA in the mix (cyan triangles). If dcSAM would bind to Svi3-3 and increase its thermal stability to the same extent as SAM and MTA, the orange plus signs and blue squares would overlap. If dcSAM would at all contribute to thermal stabilization, the cyan triangles would be above the open blue squares. The data indicate that dcSAM in a mixture with 28% SAM + MTA does not bind to Svi3-3.
Figure 9.
Figure 9.. Characterization of SAMase reaction products.
(A) TLC separation of enzymatic reactions and controls, shown in color and gray scale for clarity (1: Svi3-3, 2: T3 SAMase, 3: Orf1, 4: S-adenosyl methionine [SAM; 17 nmol], 5: L-homoserine [20 nmol], 6: L-homoserine lactone [400 nmol]). (B) 1H nuclear magnetic resonance (NMR) spectra at 600 MHz in sodium phosphate buffer, D2O, pH 7.4. Top: Enzymatic degradation of SAM (4 mM) (500 nM enzyme) after 45 min. Middle: Homoserine lactone (59 mM) showing onset of hydrolysis after 10 min. Bottom: Homoserine (57 mM).
Figure 9—figure supplement 1.
Figure 9—figure supplement 1.. 1H nuclear magnetic resonance (NMR) spectra of a reference sample of homoserine lactone (600 MHz, 59 mM in sodium phosphate buffer, D2O, pH 7.4).
Figure 9—figure supplement 2.
Figure 9—figure supplement 2.. 1H nuclear magnetic resonance (NMR) spectra of a reference sample of homoserine (600 MHz, 57 mM in sodium phosphate buffer, D2O, pH 7.4).
Figure 9—figure supplement 3.
Figure 9—figure supplement 3.. 1H nuclear magnetic resonance (NMR) spectra of a reference sample of homoserine lactone at various concentrations and exposure times to sodium phosphate buffer showing progressive hydrolysis to homoserine (600 MHz, sodium phosphate buffer, D2O, pH 7.4).
Spectrum 1: 70 mM (t = 0); spectrum 2: 17.5 mM (t = 50 min); spectrum 3: 4.4 mM (t = 547 min); spectrum 4: 1.1 mM (t = 579 min); spectrum 5: 0.3 mM (t = 603 min).
Figure 10.
Figure 10.. Mechanism of S-adenosyl methionine (SAM) degradation.
(A) The first step is catalyzed by the SAM lyase, releasing the products homoserine lactone and 5’-methyl-thioadenosine (MTA). Hydrogen bonds between the enzyme and the reacting part of SAM are shown as dashed lines. The two subunits contributing to one active site are indicated in slate and salmon. (B) Homoserine lactone is spontaneously hydrolyzed to homoserine in solution.
See this image and copyright information in PMC

References

    1. Afonine PV, Grosse-Kunstleve RW, Echols N, Headd JJ, Moriarty NW, Mustyakimov M, Terwilliger TC, Urzhumtsev A, Zwart PH, Adams PD. Towards automated crystallographic structure refinement with phenix.refine. Acta Crystallographica Section D Biological Crystallography. 2012;68:352–367. doi: 10.1107/S0907444912001308. - DOI - PMC - PubMed
    1. Basak B, Bandyopadhyay D, Patra M, Banerji A, Chatterjee A, Banerji J. Role of sulfur compounds in the detection of amino acids by ninhydrin on TLC plate. Journal of Chromatographic Science. 2005;43:104–105. doi: 10.1093/chromsci/43.2.104. - DOI - PubMed
    1. Becke AD. Density‐functional thermochemistry. III. the role of exact exchange. The Journal of Chemical Physics. 1993;98:5648–5652. doi: 10.1063/1.464913. - DOI
    1. Bowers KJ, Chow E, Xu H, Dror RO, Eastwood MP, Gregersen BA, Klepeis JL, Kolossvary I, Moraes MA, Sacerdoti FD, Salmon JK, Shan Y, Shaw DE. Scalable algorithms for molecular dynamics simulations on commodity clusters. Proceedings of the 2006 ACM/IEEE Conference on Supercomputing, SC; 2006. - DOI
    1. Capitani G, Hohenester E, Feng L, Storici P, Kirsch JF, Jansonius JN. Structure of 1-aminocyclopropane-1-carboxylate synthase, a key enzyme in the biosynthesis of the plant hormone ethylene. Journal of Molecular Biology. 1999;294:745–756. doi: 10.1006/jmbi.1999.3255. - DOI - PubMed

Publication types

MeSH terms

LinkOut - more resources