Mur Ligase Online Research Meeting July 2022

[mattodd]

Date: July 12th 2022
Time: 2pm UK time (other times)
Place: https://ucl.zoom.us/j/91379419977
Recording: https://youtu.be/EEwsGkOniHU
Previous Meeting: #76

Apologies:
Decks: Please @opensourceantibiotics/murligase remember that if you share slides/info, to drag and drop those into a comment on this page, below.

Agenda:

1) Elaborated Fragments Inhibiting Two Murs

• @AJLloyd105 @chrisdowson1 @LauraDS1 to update on further screening of the Warwick Enamine Collection. In meeting update was provided. Screenshots shown below. @chrisdowson1 mentioned in meeting that a new data analysis method has come online that will help with determination of Km etc, which should speed things up. Millimolar concentrations had been used, starting at 2 mM then diluting. Some compounds were down in the 20-100 uM inhibitors so @eyermanncj suggested trying lower concs. @chrisdowson1 suggested trying co-cryst rather than just soaking (since structural change is happening). Competitive swap out with the ADP experiments had been tried. @eyermanncj suggested UMA without ADP was the best one to go for. Exchange needs a high conc of compound, which @LauraDS1 is doing, and the presence of magnesium ions. @eyermanncj asked whether the Mg2+ is getting in the way? @AJLloyd105: there is a known optimum for Mg2+ conc. If too high may start to generate interfering chelates. @AJLloyd105 reported that compound N15 (see slides) appeared to be occluding, not completely inhibiting. Warwick project students are pursuing the next hits. (Can we get them on Github so that they can work with us on design?)

• @eyermanncj @LauraDS1 to update on any additional compounds that have consequently been ordered/evaluated

• @LauraDS1 to update on screening (inhibition, SPR, soak) of elaborated fragments sent by @edwintse in Compound to Warwick Inhibition Assay Jun 2022 #84 Enzymatic evaluation scheduled.

• @LauraDS1 @LizbeK to update on any further fragment soaking and/or co-crystallisation experiments. Comments in meeting: MurD and MurE from E coli had been used in crystallisation experiments with no ligands bound: giving open structures. Hit rate with small molecules is low. To improve: co-cryst with other ligands, then soak. e.g. MurD with ADP, ADCDP, UMA. Jan Abendroth also got structure of Psae and Ace MurD with UMA. Working on further structures (below). Laura has already ordered some compounds from the assays, and those have arrived, so plates are being set up for structural work. Psae MurC (being purified this week), D and E. @edwintse will ship 14 elaborated fragments to Warwick.

2) Atomwise Hits

• @mattodd to follow up with Denzil Bernard about potential additional suggestions based on what has been found to date.
• @LauraDS1 to complete SPR experiments (MurC, MurD in presence and absence of ADP/AMPPNP) on Atomwise compounds (SPR for mur ligases #55 and on wiki), following the previous crystal structures of several compounds bound (see also Last XChem EcMurE results uploaded #52). Hasn't this been done and uploaded @LauraDS1 ?
• @LauraDS1 @Rebecca-Steventon @chrisdowson1 to obtain inhibition data with the Atomwise compounds, to verify whether binding gives inhibition. Done? Not yet. Scheduled.
• Residual item to do: @LauraDS1 @LizbeK to clarify, on the wiki here, whether the Atomwise compounds that have been co-crystallised have a binding site that overlaps with the "target site" Atomwise used @LauraDS1 addressed this here and in the March meeting there was discussion that the Atomwise compounds did bind in a partial overlap with the intended binding site. This has been clarified as part of Predictive Modelling Competition to Design Binders of MurD, an Antibacterial Target #69, and all we need now is @LauraDS1 to summarise briefly on wiki what we know.

3) Variants of AZ Compounds

• @Yuhang-CADD to update on the synthesis of amino-AZ5595 (completed) and the amino derivative of that compound (also completed - OSA_001095, or "1095"). Needs to be evaluated vs MurC and vs. bacteria. Report provided in-meeting from 24:23. So the set of OSA1071, OSA1095 and OSA1044 is at Warwick and needs to be evaluated enzymatically.
• @Yuhang-CADD to post a summary of which of his molecules have been shipped where, along with their local and OSA codes. This is important since there may be some confusion about which molecule went where, and we'd like to be sure that the molecules have been evaluated. In meeting: Jan to chase what happened with OSA_001944.
• @Yuhang-CADD to update on structural determination using microED of several OSA compounds sent to the Rosalind Franklin Institute. In meeting: several structures have been sent.
• @Yuhang-CADD to consider alternative/additional strategies for improving accumulation, and suggestions are welcome.
• @Rebecca-Steventon to report on the assay @Yuhang-CADD 's compound here.
• @Yuhang-CADD compound structure (AZ cmpd 4) to be overlayed with AZ5595 (PDB 6X9N) and AZ8074 (PDB 6X9F) <-- @Yuhang-CADD to clarify what has been done and what still needs doing. Not clear. @Yuhang-CADD to resolve with @eyermanncj and Jan Abendroth and report back.
• @Yuhang-CADD to ship AZ5595 to SSGCID. This is as a control, for crystallisation work. Paused: we ideally ship other compounds at the same time, e.g. the compounds mentioned here, if @eyermanncj can show they dock well.
• @chrisdowson1 to work on compound transfer from SSGCID to Warwick and update the group. Does this need an update, or can it be mothballed?

4) New Protein Structures

As per update from Jan Abendroth in #80, new structures obtained for MurD + UMA? Others?

Update provided by Jan in meeting. Acet MurD with UMA. Got number of crystal forms, all in apo conformation without UMA. Surprising since had worked previously for Psae.
Psae MurC - got one molecule per asymmetric unit. Goal to re-grow for soaking with AZ compounds. Was going to get those re-screened and do soaking (had seen good thermal shifts for).

Mothball this? --> Do we need to consider the importance of the observed carbamoylation of a residue (Lys198 (-> KCX198) in a structure (7TI7). Does this influence inhibitor design? See @eyermanncj's analysis, or can this be mothballed?

A discussion of the possible value of truncated structures was had as part of the Soak-In at #80

5) De Novo Computational Modelling

• @edwintse @danielgedder to complete synthesis/procurement of molecules suggested in Predictive Modelling Competition July Summary #85 Weak binders expected, so SPR/soaking. @LauraDS1 and team to keep all molecules, since we'd like to keep all molecules for in-house library at UCL. DMSO stock solutions likely to be best.
• @Yuhang-CADD to report on progress towards @jhjensen2's originally-suggested compounds (last update was here). In meeting: Has been started, being pursued.

6) Other Potential Starting Points

• @chrisdowson1 to report on progress towards securing the Merck Open Global Health Library.
• @eyermanncj and @chrisdowson1 to speak to the benefits of ordering the Enamine kinase library. Should this be mothballed?
• @chrisdowson1 to update on access to compounds developed/evaluated in the LifeARC project. In meeting: agreement signed, compounds in the post. All structures/SI sent. To update next time after receipt of materials.

7) Misc/AOB

• @edwintse to clone Open Source Malaria Logo and Poster OpenSourceMalaria/TechOps#4 for OSA so that people can "advertise" their participation.
• Future involvement with SSGCID. @bartrum potentially to report.
• We need a volunteer to reach out to CC4CARB to assess whether we are suitable for a proposal. Wait until we have more data. Paused.
• In meeting: @mattodd will submit something on OSA to the UNESCO call for exemplars of open science projects.

8) List of Raw Data That Needs Posting Online

Suitable locations: ideally an electronic lab notebook, another suitable repository (Zenodo, Figshare) or Github itself.

• @Rebecca-Steventon - dose-response data for potential dual inhibitors OSA742, OSA754, OSA759, OSA789 that was summarised in this graphic.

9) Mothballs (if no actions then these need to be linked in wiki and closed)

• @Rebecca-Steventon has posted data of @LizbeK fragment screen (https://github.com/opensourceantibiotics/murligase/wiki/Pocket-binding-site - but @drc007 would like structures to be added).
• @LauraDS1 to liaise with Peter and @LizbeK to see if a soak of the @eyermanncj Enamine compounds is possible at Diamond.
• Comparative analysis was done by @ZigBu of mur ligase ATP binding sites (Comparison of the ATP Sites/Structural Similarity Between Murs #56). Any learnings?
• @ZigBu also posted (UCL Mur Ligase Local Team Meetings November 2021  #57) on a paper highlighting that mur domain movement may not be dependent on substrate binding. Again, anything we need to consider? @eyermanncj posted "As a reminder the dynamics of E. coli murD is well established. Here is a “recent” NMR study on E. coli murD."
• @mattodd @chrisdowson1 have parked the idea of a Euro Lead Factory screen.

Next Meeting

Aug 9th 2pm UK time for those of us not on vacation.

[Yuhang-CADD]

General Progress To Date

Synthesis of Amino AZ5595

Amino AZ5595 Verified by HRMS and NMR

Shipment of Compounds

Micro ED Results

OSA_001041 (WYH5-4-P) The structure was resolved

OSA_001048 (WYH13-4-P) The structure was not resolved due to the lack of suitable crystals (also might be deterioration of the amine, will send new batch to them later on)

OSA_001042 (WYH6-2-P) The structure was attempted and failed due to the presence of multiple lattices and lack of reflections. (Weird, as this structure had been resolved by X-ray crystallography already and was sent as a positive control)

AZ cmpd 4 (OSA_001044, WYH9-2-P)

PDB structure was submitted by Peter last month but not yet shown on the database, I am guessing it has not yet been through the publication process?

De Novo Computational Modelling_Jan's Predictions

Suzuki reaction started for ConfGen17
Testing reported conditions, LCMS showed no product peak.

Potential reasons:

1. catalyst expired
2. ring-opening of the substrate 2 and completely transformed into other byproducts (may need to couple the amine substrate 1 with Boc-Glycine first)
3. conditions reported not suitable for the system.

[edwintse]

These are the four multitargeting inhibitors from Adrian's slides last meeting

[Yuhang-CADD]

Alternative/additional strategies for improving accumulation

Incorporating the structural features of known Efflux Pump Inhibitors (EPI) into the amine product?

[mattodd]

Jan/Peter have submitted the structure of @Yuhang-CADD compound (OSA_001044) bound to Pae MurC:
Title : Pseudomonas aeruginosa MurC with WYH9-2-P - OSA_001044
Corresponding authors : Peter Horanyi, Jan Abendroth, Wang Yuhang, Matthew Todd, Isabelle Phan, Peter Myler, Donald Lorimer, Tom Edwards
Your deposition D_1000267042 is assigned the following accession code(s): PDB ID 8DOF
Thanks Jan.

[mattodd]

@Yuhang-CADD I'm listening to a talk by Helen Zgurskaya
Described here I think: https://journals.asm.org/doi/10.1128/mBio.02785-20
Take a look and see if there are some design criteria we can use?

[Yuhang-CADD]

Hi Matt,

Thanks for sharing this wonderful paper!

According to the paper, the rules may vary as the bacterial species-specific differences in the composition of the OM and efflux pump plays a role.

Such as the eNTRy way rules only applies to E.Coli and other enterobacteria due to the permeability properties of general porins dominate the OM and eject molecules based on their size, shape and electrostatic properties.

In P.ae, substrate specificities of the transporters (of efflux pumps) and efflux constant are the major drivers of efflux avoidance rules. The paper claimed a developed and validated model that describes the avoidance and inhibition of active efflux in P.ae.

From my shallow understanding of this paper, I think our focus should be on the rules/features for efflux avoidance instead of inhibition considering the way our hits (such as AZ5595 amine derivative) were expected to behave.

Thus, as the paper listed (also useful/applicable in our system), an increase in the number of hydrogen bond acceptors (HBAs) is correlated with strong efflux avoidance (our hits like AZ5595 or its amine derivative do have more than 5 N or O atoms within the structure). Also, a reduced acylindricity and anisotropic polarizability of the substrate's structure may also help reduce the efflux effects towards it (our compounds also have this feature, so no worries for now I assume).

In summary, based on the structural information of our compounds (AZ derivatives), I think they do match the criteria we extracted from the paper (mentioned above).

But please let me know if my perspective needs to be corrected.

Many thanks!
Yuhang

[mattodd]

Laura and Adrian Slides.

Slide 1, from Laura:

There was discussion of what the best concentration of compounds might be - micromolar to millimolar. Also soaking vs co-cryst; e.g. UMA system without ADP with/without Mg2+ (some discussion of optimal Mg2+ conc at 11 mins), with relatively high concentration of test compound. Crystal samples (EcMurD:UMA, and Apo EcMurE and apo EcMurD both in the presence of all four Enamine compounds) have been sent to Diamond.

Slide 2, from Adrian (starts 12:55 ends 16:35)

Enamine library Pae MurE hit N15. Data possible suggesting N15 binds in a way that incompletely occludes MurE active site. Two undergrads are starting at Warwick to pursue Enamine hits (they can be invited to the meetings...) - let's connect them up to everyone else.