We are using structural biology as a tool to determine the mechanism of proteins, particularly those involved with bacterial cell wall synthesis. This basic understating can inform and contribute towards many fields such as the design of better drugs, the way proteins evolve and the mechanisms by which enzymes operate. Structural biology is used along with microbiology and molecular biology to test new antimicrobials and develop enzymes for industrial purposes.
Antimicrobial compound development
As part of the response to the current rise in antibiotic resistance there is a pressing need to develop new antimicrobials. Within the drug design and delivery and Microbiology and Biotechnology research groups we are developing two classes of antimicrobial which prevents the bacterial cell wall from being formed. These are teixobactin and Moenomycin A derivatives. These compounds prevent the bacterial wall from forming properly and this causes the cell to burst under its own internal pressure and die. Fig 1 shows how teixobactin and moenomycin block the cell wall synthesis. Teixobactin works by binding tightly to the lipid II substrate preventing it from being incorporated into the cell wall. Moenomycin works by binding tightly to the active site of an enzyme called GT51, part of the PB1a synthetic protein. The PB1a protein incorporates lipid II into the growing peptidoglycan chain.
Fig 1. The attachment of lipid II to glycan polymer and cross linking of glycan chains performed by GT51 and TPase respectively (Structure of bifunctional PBP1a). This shows the approximate positioning within the inner membrane and the lipid II binding role of the antibacterials Teixobactin (2) and Moenomycin.
The research team and rational design approach
Based in the Joseph Banks Laboratory (JBL), the research is performed as a collaborative effort between a number of groups from the Schools of Life Sciences, Pharmacy and Chemistry as well as external collaborators. We use a “rational design approach” where novel compounds are produced, tested for efficacy, toxicity and protein binding (see Fig 2). The Singh group (http://staff.lincoln.ac.uk/isingh) uses organic synthesis, bioconjugations and peptide based delivery systems to produce novel derivatives molecules based on the naturally occurring forms of Teixobactin and Moenomycin A. The Taylor group (http://staff.lincoln.ac.uk/etaylor) along with Dr. S. Prior (http://staff.lincoln.ac.uk/sprior) then test these compounds using microbiology, structural biology (X-ray crystlography and NMR) and biochemical techniques to assess how well these molecules inhibit the target and work against the bacteria. We are also working with Prof. Mathew Goddard (http://staff.lincoln.ac.uk/mgoddard) to investigate the way bacteria evolve resistance to these new antimicrobials (This work has been kindly supported by the rosetrees trust). This approach will inform the design of the next generation of compounds to try and help combat antimicrobial resistance.
Fig.2. The rational design approach for developing lead antimicrobial compounds.
Teixobactin: Key Challenges
Teixobactin is a natural product produced from a soil bacteria which is difficult to grow. The constraint is the ability to grow these microbes and/or produce enough of the compounds as well as produce compounds which vary slightly from the natural form (known as analogues). An inability to produce or synthesise this novel antibiotic is a significant hurdle to evaluating and driving further development.
Teixobactin: The solutions to the challenge
Produce active synthetic teixobactin analogues w1th more “drug like” propteries to allow the exploitation of this new class of antibiotics.
Fig 3. The newly discovered antibacterial teixobactin
Moenomycin A: Key Challenges
Moenomycin poor pharmacokinetic properties, when given intravenously the lipid tail binds to plasma proteins rendering it inactive. When given orally, the negatively charged phosphate are carboxylate groups prevent passage through intestinal cell membrane preventing Moenomycin A from reacting the infection site. Gram negative bacteria show intrinsic resistance, the presence of the outer membrane ensures that Moenomycin A cannot reach GT51 which is located between the inner and outer membranes of Gram negative bacteria.
Fig 4. The antibacterial MoenomycinA
The results for far…..
Thanks to the hard work of many PhD, Masters, MBiol and project students we have produce and tested some promising derivative molecules (see publication 1-3 below). Particular thanks go to Charlie Vincent, Dan Lloyd, Ryan Packer and Peter Alexander who have been involved this testing.
We have helped to decode which residues / groups in the antimicrobials are important for activity. The Singh group have developed methods to make sufficient quantities to test further and weare working towards developing these molecules as therapeutic agents. We have developed and tested new teixobactin analouges ( which have been protected by a UK patent application (NO.1703753.2) and Moenomycin derivatives (see patent http://www.google.com/patents/WO2016034894A2?cl=en).
Some of these molecules have shown improved activity against Gram negative bacteria. This is an important step forward as some of these bacteria are particularly challenging in terms of resistance development.
The Structural and Functional Analysis of a Medically Important Bacteriophage.
My other interests lie in the interplay between certain pathogenic bacteria and their compatible bacteriophage. In some cases a complex chain of molecular events may result in the expression of bacteriophage encoded toxin genes or virulence factors. These have a direct effect, causing the symptoms of diseases such as diphtheria, cholera, dysentery, botulism, necrotizing pneumonia, toxic shock, food poisoning and scarlet fever. We have focused on the pathogenic bacterium Streptococcus pyogenes SF370 M1 (GAS) and one of its associated Siphovirida prophages SF370.1.
Bacteriophages are viruses that infect bacteria. They do this by “hijacking” the bacterial metabolism in order
to reproduce; the virus injects its own DNA in the bacteria. This DNA contains all the necessary information to produce new phage. Here the DNA is decoded by the bacteria to produce new virus particles. Finally the bacteria are burst open by enzymes called endolysins (Fig 5.) and the new virus particles are released. S. pyogenes M1 (GAS) is a gram-positive human pathogen responsible for a wide range of disease variants from mild infection such as cellulites, strep sore throat (pharyngitis) to the more invasive life threatening forms such as toxic shock and necrotizing fasciitis (flesh eating disease). A fascinating three way interplay exists between the lysogenized “dormant” SF370.1 prophage, the bacterium and the human host. Phage induction results in phage proliferation and the expression of the two key phage encoded virulence factors the SpeC toxin and the Spd1 DNase (Fig 6.).
- Fig.6 .The phage mediated transfer of SpeC toxin and the Spd1 Dnase1 genes (red) in the Lysogenic (genome integrated) and lytic (none integrated) variants of the phage reproductive cycle. The blue arrow indicated the point in the cycle where the activity of Hylp1 helps in the transfection process. The green arrow indicated the point in the cycle where the SpeC toxin, Dnase1 and endolysin genes are expressed.
These proteins go on to interact with the human immune system. The SpeC toxin has a well documented action, which is to crosslink the major histocompatibility complex class ii with antigen presenting cells, causing massive cytokine production and many of the symptoms associated with scarlet fever and toxic shock syndrome. The virulence role of the streptococcal DNase, is less well defined, but they have been shown to protect GAS against neutrophil killing by degrading the DNA in neutrophil extracellular traps. This promotes S. pyogenes to adopt a much invasive and dangerous necrotizing mode of growth as well as probably helping the phage progeny to escape the immune system.
Our aim is to obtain the 3 dimensional structures of key proteins encoded by the SF 370.1 phage genome (Fig 7.) and using molecular biology, biochemistry and microbiological techniques look at their interactions.
A typical study involves the cloning and expression of the target gene. The protein is then purified, screened and crystallized.
How do we “Look” at the 3-D Structure of Protein molecules?
Proteins are in fact “Chains” of amino acid molecules joined together by a common linkage. There are 20 different types of amino acid all with different shapes and charges. These chains fold up to give a three dimensional shape which is very important and gives the protein its function. Unfortunately protein molecules are just too small to be detected using visible light so we cannot use a microscope. In order to “see” them we need to use electromagnetic radiation with a shorter wavelength such as X-rays. However there is a problem with this, as there are no lenses suitable to focus X-rays as in a light microscope and the Intensity of the radiation required would destroy the protein molecule. So we use X-ray Crystallography!
When a protein crystallises (Fig.8) the molecules arrange themselves in a three dimensional lattice. When shot with an X-ray beam the electrons in different lattice layers “reflect” the beam causing X-ray diffraction. We are able to measure this on a detector (Fig 9. a) and by a series of mathematical calculations and experiments we are able to map the electron density within a protein. Once the electron density map is produced we are able to model in the protein “chain” (Fig 9. b) which is great fun and like doing a 3-D jigsaw! We can use this model to explain the biological function of the protein. There are different ways of showing the structure including a cartoon (Fig 9. c).
The crystals above are grown from a phage regulatory protein and were used to solve its structure. We can tell from the structure (c) it probably functions by binding DNA, but more biochemical work is needed before we know if that for certain.
- Fig. 9. The sequence of milestones leading to the structural solution of a phage regulatory protein. Single protein crystals causes X-rays to be diffracted (a) a series of these diffraction patterns are collected as a data set. This structure (c) was solved in combination with a derivative data set, and is shown here as a cartoon.
Elements within the structure may be seen to account for the proteins function such as catalysis. Several themes have emerged from this work, which are the study of hyaluronate lyase which help the phage transcend the bacterial hyaluronic acid capsule, the mechanisms of bacterial endolysins and the Spd1 DNase1 virulence factor (Figure 10.a and b).
Our general interests include the structure and mechanisms of carbohydrate active enzymes, particularly protein domains which bind to “non classical signature sugars”. These are typically associated with the cell walls of some problematic bacteria such as S. pyogenes. Specific interactions between these sugars may be used as the basis for a rapid identification technique, but little is known about how it works on a molecular level. Other past work includes the carbon fixing enzymes of the ribulose monophosphate pathway, enzymes which break down plant polysaccharides and enzymes involved in the deglycosylation of proteins.
- Parmar, A., Iyer, A., et al. (2017) “Syntheses of potent teixobactin analogues against methicillin-resistant Staphylococcus aureus (MRSA) through the replacement of L-allo-enduracididine with its isosteres”. Chemical Communications
- Parmar, A., Prior, S. H., et al. (2017) Defining the molecular structure of teixobactin analogues and understanding their role in antibacterial activities. Chemical Communications 53, 2016-2019
- Parmar, A., Iyer,A., et al. (2016) Efficient total syntheses and biological activities of two teixobactin analogues. Chemical Communications 52, 6060-6063
- Korczynska, J.E., Turkenburg, J.P. and Taylor, E.J. (2011) The structural characterisation of a prophage encoded extracellular DNase from Streptococcus pyogenes. Nucleic Acids Research, doi:10.1093/nar/gkr789.
- Hemsworth, G., Taylor, E.J et al. (2013). The copper active site of CBM33 polysaccharide oxygenases. J Am Chem Soc. 135 (16), 6069–6077.