for neurodegenerative disorders D R D AV I D C H AM B E R S P R I - - PowerPoint PPT Presentation

D R D AV I D C H AM B E R S P R I N C I PAL I N V E S T I G ATO R & L E C T U R E R I N F U N C T I O N AL G E N O M I C S G E N O M I C S D R U G D I S C O V E RY U N I T W O L F S O N C E N T R E F O R AG E - R E L AT E D D I S E AS E S ( C AR D ) K I N G ’ S C O L L E G E L O N D O N S E 1 1 U L U K

Integrated genomic approaches to repositioning drugs for neurodegenerative disorders

Declaration

 Thermo Fisher Scientific and its affiliates are not

endorsing, recommending, or promoting any use or application of Thermo Fisher Scientific products presented by third parties during this seminar. Information and materials presented or provided by third parties are provided as-is and without warranty of any kind, including regarding intellectual property rights and reported results. Parties presenting images, text and material represent they have the rights to do so. If applicable; modify as appropriate, (for which you must have a signed speaker engagement agreement) including if free products/support were provided: Speaker was provided travel and hotel support by Thermo Fisher Scientific for this presentation, but no remuneration.

Lab themes & research areas

Investigative biomarker approaches Drug discovery

Drug repurposing: CMAP & candidate approaches Pathway ID & validation: RA, RTKs, GPCRs Large-scale high-resolution biomarkers: Single cell FFPE Emerging Pathways: miRNA & exosomal signaling Genomics-based drug repositioning

AD PD NDD Regen Pain

Alzheimer’s Disease: the unmet need

• 35 million people worldwide with dementia
• 78 million by 2040
• >60% have Alzheimer’s Disease (AD)
• Huge human and financial cost: Global cost

estimated > $600 billion

• Symptomatic treatments give modest but important

benefit

• Disease-modifying drugs are urgently needed to:
• Delay the onset of Alzheimer’s disease
• Improve long term outcomes

‘Current drugs help mask the symptoms of Alzheimer's, but do not treat the underlying disease or delay its progression’: Alzheimer’s Association 2016

 Polyproteinopathies (Ab, NFT, aSyn)  Synaptodendritic rarefaction  Inflammation  Mitochondrial dysfunction  Multiple transmitter deficits  Aberrant neural network activity  Reduced neurogenesis  Degeneration of specific neuronal cells  Epigenetics  Lysosomal proteolysis  Dysregulation intracellular Ca2+ Levels  Oxidative damage  Perpetuated cell-cell spread



Why is it so difficult to find a drug with multiple disease targets?

AD: biological targets for drug discovery

The drug discovery pipeline: why aren’t there many new drugs?

High costs: $1.5 billion to bring a new compound to market Long timelines: around 12 years; patents valid for 25 years Cumulative number of new drugs (NMEs) approved by FDA circa 2013 Circa 2012: 1700 CT cancer vs 30 CT AD

Can we use other drugs?

 Drug repositioning or repurposing.  Identifies already existing compounds which may

have benefit in treating target disease

 Benefits include saving time and money: $5-10m

making it accessible for research charities

 The dosage, tolerability & side-effects are known  Potential new delivery mechanisms

How do we go about repurposing studies?

Sildenafil Specific enzyme (PDE5) inhibitor Unsuccessful for angina Successful for male impotence

‘On’ target approach: reiterated mechanism of action of the drug ‘Off’ target approach: identify novel targets for existing drugs

Amantadine Licensed for influenza: M2 Proton channel blocker Discovered NMDA receptor antagonist Used for Parkinson’s Disease

How do we go about repurposing studies?

A genomic approach to drug repurposing The Connectivity Map (CMAP) [Broad Institute] An ‘Off’ target approach

The CMAP in a nutshell

Disease gene expression signature Disease gene expression signature 1. Generate via Array or NGS 2. Generate via GWAS, WES 3. Generate manual list 4. Generate via metadata (Spied) 5. Efficacious Drug Mimetic Drug Gene Expression profile 1. Generated by Affymetrix Array 2. Non parametric ‘ranking’ 3. Generated by Bead studio (LINCs) 4. Cancer cell focussed Drug Gene Expression profile

Connectivity Mapping

Justin Lamb et al. Science 2006;313:1929-1935

Accordingly, the Cmap resource has the potential to connect human diseases or degenerative states with the genes that underlie them and the drugs that treat them

CMAP: key parameters

All treatments of cells are 6h Original CMAP: dosing distribution Human: MCF7 breast adenocarcinoma cell line

Justin Lamb et al. Science 2006;313:1929-1935

CMAP = >1300 FDA approved compound profiles in MCF7

Does CMAP work: cancer proof of concept

Cimetidine

Human lung adenocarcinoma signature generated Experimental validation of cimetidine for lung adenocarcinoma

Sirota, M., et al.,. Sci Transl Med, 2011. 3(96): p. 96ra77

Can we do better than CMAP for NDD? A Systematic Approach to Develop and Evaluate the Best Candidate Treatments for Repositioning as Therapies for Alzheimer’s Disease: SMART-AD

Prof Clive Ballard Prof Pat Doherty Prof Jonathan Corcoran Dr Gareth Williams Dr Anne Corbett Dr David Chambers Prof Paul Francis Prof Simon Lovestone

SMART AD

SMART AD is driven by human genetics: SPIED

 Searchable platform-independent expression database

(SPIED)

 SPIED uses deposited profiles as surrogates for biology

comparison across all platforms and species

 Can query SPIED to identify all experiments relevant to

specific questions and then generate consensus signature:

 Generate gene expression signatures for different

classifications of AD

 Human: Early  Human: Moderate  Human: Severe  Mouse: most representative AD model to human AD

SMART AD

Query CMAP with Human Early AD signature: anticorrelates

Approximately 200 drugs significantly anti correlate with early AD signature

SMART AD

CMAP Drug Candidates from multiple independent drug classes

Heatmap: transcriptional similarity of the 200 SMART AD candidates to each other reveals distinct classes of drugs including: anti-inflammatory, anti-bacterial, analgesics & anti-depressives correlation

SMART_AD: Cell type

Human: cerebral cortical iPSC* neurons

Rat: hippocampal neurons Human: MCF7 breast adenocarcinoma cell line

Increasing relevance for SMART AD initiative

Do candidate drugs generate an anti correlating profile in human neuronal cells?: NMAP & ApoE4 NMAP

NMAP CMAP

Human: cerebral cortical iPSC* neurons: ApoE4

ApoE4: NMAP

SMART AD

The distribution of significantly altered gene expression values over the assayed drugs is shown [left] The distributions are relatively symmetric, with ~1000 up and down regulated genes on average, shown right. SMART_AD candidate compounds induce robust and genome-wide gene expression changes in neurons (hyCCN IPSCs) : Affymetrix U133 2.0

Generate an AD-relevant neuronal connectivity map: NMAP

The effects are not necessarily mediated by classic ligand- receptor pharmacology

SMART AD

SMART AD: NMAP summary

1300 CMAP Candidates 200 CMAP hits –ve AD 160 NMAP –ve AD 40 retain –ve AD Systematic review & Steering Panel Triage ~ 1000 Transcriptomic profiles generated: NMAP SPIED: AD ‘Early’ Signature

In vitro assays for AD Candidates

Ab 1-42 P Tau Cell Death H202 Neuro genesis Wnt

SMART AD

Ab 1-42 Cell Death Assay characterisation in mouse cortical neurons: 3 Day

10uM 3uM 1uM diluent untreated 0.0 0.2 0.4 0.6

Abeta42 titre

** *** *** ** **

3 day absorbance (570nm)

Plate 3

Abeta42 only Diluent C18 C34 C37 50 100 *** ***

3 day % of control

C18 > Abeta42: p=0.0003

SMART AD

SMART AD: NMAP summary

1300 CMAP Candidates 200 CMAP hits –ve AD 160 NMAP –ve AD 40 retain –ve AD 12 pass in vitro Systematic review & Steering Panel Triage ~ 1000 Transcriptomic profiles generated

SMART AD: What classes of Drugs?

 SMART AD: select hits to progress based upon diverse

drug classes and interaction with different pathways

 Antibiotics  NSAIDS  Receptor antagonists  Histone deacetylase inhibitors  Naturally-occurring compounds

SMART AD

NMAP data predicts the pathways candidates target in human neuronal cells: Drug F

Pathways enriched in the top 500 responders. Immune system, WNT signalling and Amyloids are notable pathways.

PATHWAY p N n IMMUNE SYSTEM 0.025892 868 31 HEMOSTASIS 0.041398 445 17 METABOLISM OF PROTEINS 0.027839 414 17 CELL CYCLE 0.014612 375 17 CELL CYCLE MITOTIC 0.046736 290 12 CYTOKINE SIGNALING IN IMMUNE SYSTEM 0.045332 256 11 TRANSCRIPTION 0.010066 191 11 POST TRANSLATIONAL PROTEIN MODIFICATION 0.00603 176 11 JAK STAT SIGNALING PATHWAY 0.002425 154 11 CLASS I MHC MEDIATED ANTIGEN PROCESSING PRESENTATION 0.046481 226 10 FOCAL ADHESION 0.021307 190 10 FATTY ACID TRIACYLGLYCEROL AND KETONE BODY METABOLISM 0.007743 158 10 SYSTEMIC LUPUS ERYTHEMATOSUS 0.002735 134 10 RNA POL I RNA POL III AND MITOCHONDRIAL TRANSCRIPTION 0.000943 115 10 ENDOCYTOSIS 0.022208 164 9 FACTORS INVOLVED IN MEGAKARYOCYTE DEVELOPMENT AND PLATELET PRODUCTION 0.005261 125 9 RNA POL I TRANSCRIPTION 0.000329 82 9 WNT SIGNALING PATHWAY 0.029008 146 8 SIGNALING BY NOTCH 0.003218 94 8 CELL ADHESION MOLECULES CAMS 0.043728 133 7 CELL CYCLE 0.025335 115 7 HYPERTROPHIC CARDIOMYOPATHY HCM 0.005809 83 7 SIGNALING BY NOTCH1 0.001344 63 7 MEIOSIS 0.049235 109 6 DNA REPAIR 0.039739 102 6 MEIOTIC RECOMBINATION 0.017968 82 6 ST INTEGRIN SIGNALING PATHWAY 0.016297 80 6 AMYLOIDS 0.015497 79 6

SMART AD

SMART AD: NMAP summary & current stage

1300 CMAP Candidates 200 CMAP hits –ve AD 160 NMAP –ve AD 40 retain –ve AD 12 pass in vitro

in vivo

6

5 x FAD

Apply genomic profiling to determine BBB penetration Systematic review & Steering Panel Triage ~ 1000 Transcriptomic profiles generated Determine histopathology and behavioural impact of candidates

Milestones 1 - 4

Proof of Concept II: Retinoid signalling and ageing



Generate signature of efficacious drug with undesirable off-target effects:



Reduction in adult neurogenesis is concomitant with decline in atRA levels



Exogenous retinoid signalling can reverse age-related decline in hippocampal neurogenesis



Target process neuroprotection and neurogenesis



Target Disease: Neurodegeneration (AD)



ApoE4 impairs adult hippocampal neurogenesis



Internally Generated Query Signature



Use CMAP to find Drugs that correlate: mimetic



Validation: in vitro assay of neuronal cell death

RETINOIDS AND AD

 Retinoid-based compounds: drugs of gene

expression

 Retinoids signal via steroid hormone-like

receptors (RARabg & RXRabg) to directly modulate gene expression in target cells

 Accordingly they are ideal targets for CMAP-

based approach as they control defined cohorts of genes that can be correlated with disease signature

Using CMAP to find RA bio-mimetics

Exposure to RAR agonists over a period of 28 days: stable signature

Chambers & Maden CRC Press 2017

Using CMAP to find bio-mimetics

David Chambers & Malcolm Maden

Co-ordinating compounds across models

Drug P Drug I Drug N Drug P2 Drug U Drug M Drug SP some are concentration dependent e.g. Drug P, but many

• f them work across

concentrations from 100 mM to 1 nM

Chambers, Maden & SMART AD

SMART AD Proprietary cell to cell signaling RAR a/b/g Neurogenesis 2-6

Acknowledgements



SMART AD: Clive Ballard (Lead PI), Patrick Doherty (PI), Gareth Williams (PI), Jonathon Cocoran (PI), Paul Francis (PI), Anne Corbett



RA a/b/g: Malcolm Maden, University of Florida



EGFR NSC SPIED: Pat Doherty, Gareth William & Phil Sutterlin



Exosomal Signalling & Regeneration: Ketan Patel Micregen & University of Reading



Exosomal miRNA-21-5p & Pain: Marcia Malcangio



Funding



Wellcome Trust, BBSRC, KHP, Micregen, PE & PF