Transporting Biotherapeutics Across the Blood-Brain Barrier

October 15, 2020 Disclaimers

Forward Looking Statements This presentation contains forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. All statements other than statements of historical facts contained in this presentation, including, without limitation, statements regarding future results of operations and financial position of Denali Therapeutics Inc. (“Denali” or the “Company”); Denali’s business strategy, business plans, product candidates, planned preclinical studies and clinical trials; expectations regarding the timing of results of such studies and trials; plans, timelines and expectations related to DNL310, Denali’s TV technology platform and TV programs; the ability of the TV technology to effectively deliver large therapeutic molecules across the blood-brain barrier (“BBB”); plans, timelines and expectations related to DNL151 and other LRRK2 inhibitor molecules; plans, timelines and expectations related to DNL788, DNL758 and DNL343; expectations related to clinical trials to address Hunter Syndrome and the ability to establish biomarker proof-of-concept in patients by end of 2020; the potential benefits and results of the collaborations with Denali’s partners, including Biogen, Sanofi and Takeda; plans and expectations regarding patient recruitment, planned regulatory filings, long-term development plans and near-term pipeline milestones; and Denali’s priorities, regulatory approvals, timing and likelihood of success and expectations regarding collaborations, are forward-looking statements. Denali has based these forward-looking statements largely on its current expectations and projections about future events.

These forward-looking statements speak only as of the date of this presentation and are subject to a number of risks, uncertainties and assumptions, including but not limited to, risks related to: any and all risks to Denali’s business and operations caused directly or indirectly by the evolving COVID-19 pandemic; the risk of the occurrence of any event, change or other circumstance that could give rise to the termination of Denali’s collaboration agreements, including those with Biogen, Sanofi and Takeda; Denali’s early stages of clinical drug development; Denali’s ability to complete the development of, and if approved, commercialization of its product candidates; Denali’s dependence on successful development of its BBB platform technology, product candidates currently in its core program and biomarker strategy; expectations and potential benefits of strategic collaboration agreements may not be met and Denali may not be able to attract collaborators with development, regulatory and commercialization expertise; Denali’s ability to conduct or complete clinical trials on expected timelines; the risk that preclinical profiles of Denali’s product candidates, such as DNL 151 and DNL788, may not translate in clinical studies, and the uncertainty that any of Denali’s product candidates will receive regulatory approval necessary to be commercialized; Denali’s ability to obtain and maintain regulatory approval of its product candidates, and any related restrictions, limitations and/or warnings in the label of any approved product candidate; Denali’s ability to continue to create a pipeline of product candidates and develop commercially successful products; Denali’s ability to obtain, maintain, or protect intellectual property rights related to its product candidates and BBB platform technology; implementation of Denali’s strategic plans for its business, product candidates and BBB platform technology; Denali’s ability to obtain funding for its operations, including funding necessary to develop and commercialize its product candidates; and other risks. In light of these risks, uncertainties and assumptions, the forward-looking statements and events discussed in this presentation are inherently uncertain and may not occur, and actual results could differ materially and adversely from those anticipated or implied in the forward- looking statements. Accordingly, you should not rely upon forward-looking statements as predictions of future events. Information regarding additional risks and uncertainties may be found in Denali’s Annual Report on Form 10-K filed with the SEC on March 12, 2019, Denali’s Quarterly Report on Form 10-Q filed with the SEC on August 7, 2020 and Denali’s future reports to be filed with the SEC. Denali does not undertake any obligation to update or revise any forward-looking statements, to conform these statements to actual results or to make changes in Denali’s expectations, except as required by law.

Accuracy of Data This presentation contains statistical data based on independent industry publications or other publicly available information, as well as other information based on Denali’s internal sources. Denali has not independently verified the accuracy or completeness of the data contained in these industry publications and other publicly available information. Accordingly, Denali makes no representations as to the accuracy or completeness of that data.

2 AGENDA

TIME (ET) TOPIC SPEAKER

1:00 – 1:30 p.m. Introduction and Blood-Brain Barrier Transport Vehicle (TV) Overview Ryan Watts, PhD CEO, Denali

Simon Jones, MRCPCH, MBChB 1:30 – 1:50 p.m. Hunter Syndrome: Overview Willink Unit, Manchester Centre for Genomic Medicine

1:50 – 2:20 p.m. TV Flagship Program: ETV:IDS (DNL310) Carole Ho, MD CMO and Head of Development, Denali

2:20 – 2:40 p.m. Q&A

2:40 – 2:50 p.m. Break

2:50 – 3:20 p.m. TV Portfolio Programs Joe Lewcock, PhD CSO and Head of Discovery, Denali

3:20 – 3:35 p.m. Unlocking the TV Platform Potential Alex Schuth, MD COO, Denali

3:35 – 4:00 p.m. Q&A

3 Introduction

Ryan Watts, PhD, CEO OUR PURPOSE: DEFEAT DEGENERATION

RARE NEURODEGENERATIVE AMYOTROPHIC DISEASES LATERAL SCLEROSIS PARKINSON’S ALZHEIMER’S

Orphan 20,000+ (US) 1,000,000+ (US) 5,800,000+ (US)

>30 lysosomal storage diseases >45 known genetic associations >95 known genetic associations >35 known genetic associations

D n i o s i t e a a l s u e

a i

Normal PD Normal AD

Significant unmet medical need with few disease-modifying medicines 5 OUR PRINCIPLES SCIENTIFIC Genetic Pathway Potential Engineering Brain Delivery Biomarker-Driven Development

BUSINESS Broad Portfolio Parallel Investments Strategic Partnering

6 Number of Genetic Associations and Implicated Genes PILRA Degenogenes Degenogenes -related Degenogenes -related -related APOE4

Glial Biology Function Lysosomal Other Degenogenes Cellular Homeostasis Cellular

Disease Disease

ALS / FTD / ALS

Parkinson’s Parkinson’s Alzheimer’s Alzheimer’s DEGENOGENES DEFINE NEURODEGENERATION BIOLOGY DEGENOGENES DEFINE NEURODEGENERATION 7 OUR PORTFOLIO APPROACH

DIVERSE DIFFERENTIATED DATA-DRIVEN Therapeutic Pipeline Brain Delivery Technology Drug Development

D n i o s i t e a a l s u e

a i

Multiple therapeutic targets, Two Platforms: Focus on target engagement, modalities and indications in Degenogene Biology & pathway engagement, and neurodegeneration BBB Technologies patient phenotyping for clinical decisions

8 OUR PROGRESS: PAST TWO YEARS

• RIPK1 decision to advance 3Q DNL788; pause DNL747 3Q 1Q • Biomarker PoC LRRK2 inhibitors • Initiated P1/2 in Hunter syndrome (HV & PD for DNL201; HV for (ETV:IDS/DNL310) DNL151) • Sanofi initiated Ph1b COVID-19 of • IND accepted for ETV:IDS/DNL310 • Sanofi initiated Ph1 HV peripheral inhibitor for MPSII 1Q with peripheral inhibitor (RIPK1:DNL758) (RIPK1: DNL758) • First in human dosing of EIF2B • Selected DNL151 for late-stage • Initiated Ph1b in ALS activator (DNL343) testing in PD (LRRK2) (RIPK:DNL747) • Added PTV:PRGN & ETV:SGSH to • Initiated Ph1b in PD • Biogen partnership portfolio • Initiated Ph1b in AD (LRRK2: DNL151) $560M upfront, $465M equity (RIPK:DNL747) • $207M follow-on offering ($23/share) investment 2019 2020

• PoC data expected for • Orphan Drug & Rare Pediatric Disease ETV:IDS/DNL310 in Designation for ETV:IDS/DNL310 Back-to-back publications on 2Q 2Q TV Platform Technology in 4Q Hunter syndrome Science Translational Medicine

Clinical Regulatory Portfolio Corporate Milestones Milestones Transitions Strategy 9 OUR PORTFOLIO Large Molecule (TV Platform) Small Molecule AAV DRUG DEVELOPMENT PROGRAM TARGET DRUG CANDIDATE DISEASE INDICATION PARTNER Drug Discovery IND-Enabling Early Clinical Late Clinical Approved LYSOSOMAL FUNCTION PATHWAY

LRRK2 DNL151 Parkinson’s Biogen

Iduronate 2-sulfatase DNL310 MPS II (Hunter Syndrome)

PGRN DNL593 Frontotemporal Dementia Takeda

Alpha-Synuclein ATV:aSyn Parkinson’s, DLB, MSA

Sulfamidase ETV:SGSH MPS IIIA (Sanfilippo Syndrome)

Undisclosed ETV:LF1 LSD with Neurodegeneration

Undisclosed AAV:LF2 Parkinson’s GLIAL BIOLOGY PATHWAY

RIPK1 (CNS) DNL788 Alzheimer’s, ALS, MS Sanofi

TREM2 DNL919 Alzheimer’s Takeda

Undisclosed GB1 ALS CELLULAR HOMEOSTASIS

EIF2B DNL343 ALS, FTD

Tau ATV:Tau Alzheimer’s Takeda

Abeta ATV:Abeta Alzheimer’s Biogen

Undisclosed CH1 ALS, Parkinson’s OTHER COVID-19, Peripheral RIPK1 (Peripheral) DNL758 Sanofi Inflammatory Diseases LRRK2 (Peripheral) DNL975 Crohn’s Disease Biogen 10 HER2 ATV:HER2 Oncology 2020 CLINICAL PROGRESS AND PLANS

LRRK2 § Entered strategic collaboration with Biogen Parkinson’s § Advancing DNL151 into late-stage clinical trials in 2021

EIF2B § DNL343 Phase 1 in HV results to enable patient study – end of 2020 / early 2021 ALS

RIPK1 § IND submitted for DNL788 (Sanofi) CNS § First in human dosing by end of 2020 / early 2021 (Sanofi)

RIPK1 § DNL758 Phase 1b in COVID enrollment complete (Sanofi) Peripheral § Initiate DNL758 Phase 2 in cutaneous lupus early 2021 (Sanofi)

ETV:IDS § Establish biomarker proof-of-concept in patients by end of 2020 Hunter Syndrome

§ Establish TV Platform proof-of-concept in humans (ETV:IDS) TV Platform § Expanded portfolio with Biogen option programs § Initiate IND-enabling studies for additional programs 11 Blood-Brain Barrier Transport Vehicle

Ryan Watts, PhD, CEO LARGE OPPORTUNITY FOR BIOTHERAPEUTICS IN NEUROLOGICAL DISEASE

NEUROLOGICAL DISEASE APPROVED BIOTHERAPEUTICS

Cause of #1 Disability 4

Cause of Majority are intrathecal (IT) or #2 Death intracerebroventricular (ICV) delivery*

There is a significant need for blood-brain barrier (BBB) crossing biotherapeutic platforms

13 GBD 2016 Neurology Collaborators Lancet Neurology (2019) *Includes peptides, ASO, and AAV; Does not include multiple sclerosis KEY CONSIDERATIONS FOR BIOTHERAPEUTICS AND THE BRAIN

• Brains are surrounded by cerebrospinal fluid (CSF) that circulates and is turned over many times a day

• Directly injecting biotherapeutics into the brain faces major limitations • CSF flow into the brain is incomplete Mouse • Diffusion from a point of injection relies on Brownian motion (and is limited to a few mm) • Size matters: human brain is ~3,300x larger than mouse; direct injections are NOT SCALABLE

• The human brain has 400 miles of blood vessels Rat • Every neuron is near an associated capillary

• Blood vessels in the brain have a barrier (BBB)

• Ideal delivery approaches will utilize capillaries and Human cross the BBB for broad biodistribution to the brain • BBB crossing IS SCALABLE across species

14 WIDESPREAD BIODISTRIBUTION WITH BBB-CROSSING COMPARED TO LIMITED INTRATHECAL BIODISTRIBUTION

INTRATHECAL BIOTHERAPEUTIC INTRAVENOUS BBB-CROSSING BIOTHERAPEUTIC Limited distribution with sharp gradients at brain Widespread biodistribution dictated by capillary TfR and spinal cord CSF-contacting surfaces expression and CNS vascularity

Capillary TfR Sharp gradient expression limits brain facilitates biodistribution widespread biodistribution

CNS vascularity facilitates Sharp gradient widespread limits biodistribution biodistribution along the spinal cord

IV IDS Lumbar Region IT IDS

15 INCREASING BRAIN EXPOSURE BY TARGETING TRANSFERRIN RECEPTOR

TRANSFERRIN RECEPTOR (TfR)

Native biology • Actively transports iron into the brain • Plays critical role in iron homeostasis • Highly expressed on brain endothelial cells Target for brain delivery • High vascularization enables ubiquitous delivery • Constitutive transcytosis enables robust capacity • Efficient transcytosis is dependent on affinity

16 MILESTONES IN DEVELOPING TRANSFERRIN RECEPTOR (TfR) TECHNOLOGY

1980-1990 1990-2010 2010-2015 2015-2020

TfR binding transport vehicle (TV) platform brain delivery in mice and NHPs

Transport Vehicle TfR MAb brain Reduced affinity Monovalent single TfR bispecific brain TfR at the BBB delivery in rats TfR MAb bispecific chain Fab antibody delivery in mice and mice brain delivery fusion brain and NHPs TfR in mice delivery in mice binding

1500

#CNS drug delivery 1000 papers per year in Pubmed 500

1980 1990 2000 2010 2020 We have improved upon a long history of targeting TfR to develop an optimized BBB-crossing biotherapeutic platform 17 NHPs = nonhuman primates TV TECHNOLOGY DELIVERS BIOTHERAPEUTICS TO THE BRAIN

ATV TfR

Tf Tf

endothelial cell membrane

Antibody Transport Vehicle (ATV)

Published May 27, 2020

Brain regions (cynomolgus monkey) Cortex (cynomolgus monkey) ATV achieves high concentrations and broad distribution of antibodies in nonhuman primate brains 18 BROAD BIODISTRIBUTION OF ATV IN MOUSE BRAIN neocortex

huIgG1 NeuN huIgG1 NeuN DAPI huIgG1 NeuN huIgG1 NeuN DAPI

4 hours 4

24 hours 24

anti-BACE1 anti-BACE1

20 µm 20 µm

4 hours 4 24 hours 24

5 µm 5 µm

ATV35.21:BACE1 ATV35.21:BACE1

Time-dependent transfer of ATV antibodies from blood vessel (left) throughout the brain (right) 19 ETV TECHNOLOGY DELIVERS ENZYMES TO THE BRAIN

ETV:IDS crosses the BBB and reduces GAGs in all cell types in brain

Published May 27, 2020 ETV:IDS uptake in neurons

DAPI NeuN LAMP2 huIgG Colocalization

Denali Data ETV achieves high concentration and broad distribution of enzymes in the brain after IV administration 20 ETV:IDS SHOWS BROAD BRAIN BIODISTRIBUTION COMPARED TO IgG:IDS broad distribution ETV:IDS

broad biodistribution

IgG:IDS trapped vascular

vascular trapped

Denali Data; Neocortex, 10 mg/kg, n=3 per group (mean ± SEM); p values: two-way ANOVA w/ Sidak’s multiple comparison test; *** p £ 0.001 and **** p £ 0.0001.

ETV:IDS parenchymal biodistribution compared to much more limited vascular trapping with a bivalent, high affinity TfR binding IDS fusion protein (IgG:IDS) 21 MODALITY-BASED BRAIN DELIVERY PLATFORMS ENABLED BY TV

Antibody Transport Vehicle Enzyme Transport Vehicle § Deliver antibodies in bivalent or bispecific § Deliver enzymes to the brain format to the brain § Treat neuronopathic symptoms of § Natural IgG structure and properties lysosomal storage diseases § Examples: ATV:TREM2, ATV:Abeta § Examples: ETV:IDS, ETV:SGSH

ATV ETV

PTV OTV Protein Transport Vehicle Oligonucleotide Transport Vehicle § Deliver proteins to the brain § Deliver oligonucleotides to the brain § Replace or enhance function of natural § Protein knockdown with ASO or siRNA proteins § Examples: PTV:PGRN

22 TfR TfR binding binding

TfR TfR binding binding Standard Protein Antibody Fusion TfR binding

Standard Protein Antibody Fusion Bispecific Antibody TfR TfR TfR binding binding TV DIFFERENTIATION binding

Transport Vehicle

TfR binding TfR Bispecific binding Antibody Single Chain Standard Protein Antibody Fusion Antibody Fusion TfR Roche bindingJCR Pharma DENALI Genentech Key Biotherapeutic Properties Single Chain Standard Protein Fc fragment Bispecific Fusion Antibody Fusion Transport Vehicle TfR Natural IgG; no linkers, appendages Yesbinding No Yes Yes ARCHITECTURE Monovalent; avoids TfR degradation Yes Yes No Yes Single Chain Silence or modify effector function AntibodyYes Fusion No BispecificNo Yes Antibody Integrated TfR binding Yes No Yes Yes BIODISTRIBUTION Optimized TfR binding affinity Yes ? No Yes Cell-type specific rescue in brain ? ? ? Yes

Ability to deliver multiple modalities ? ? ? TfR Yes MODULARITY binding Retain bivalent/bispecific binding No Yes No Yes

Single Chain 23 Antibody Fusion EXPANDING THE THERAPEUTIC LANDSCAPE WITH TV DELIVERY

ENHANCE TARGETS UNLOCK TARGETS Improved efficacy with TV Creation of new opportunities

ATV:Aβ ATV:TREM2 ATV:HER2 ETV:IDS ETV:SGSH OTV PTV:PGRN (DNL310)

Increase biodistribution (10-30X) and enhance Unlock brain delivery of biotherapeutics activity by combining TfR with Target with systemic administration

24 Mucopolysaccharidosis type II (Hunter syndrome)

Simon Jones Willink Unit Manchester centre for Genomic Medicine Disclosures

• Consultant and investigator - Denali therapeutics • Consultant and investigator - Takeda • Consultant, investigator and shareholder - Orchard therapeutics • Consultant - Avrobio

26 Classification of MPS TYPE NAME ENZYME DEFICIENCY GAG MPS I Hurler / Scheie α-L-iduronidase HS DS

MPS II Hunter Iduronate-2-sulfatase HS DS

MPS IIIA Sanfilippo A Heparin sulfamidase HS

MPS IIIB Sanfilippo B N-acetyl-α-D-glucosaminidase HS

MPS IIIC Sanfilippo C Acetyl-CoA:α-glucosaminidase HS

MPS IIID Sanfilippo D N-acetylglucosamine-6-sulfatase HS

MPS IVA Morquio A N-acetylgalactosamine-6-sulfatase KS

MPS VI Maroteaux-Lamy N-acetylgalactosamine-4-sulfatase DS

MPS VII Sly β-Glucuronidase HS DS

MPS IX Natowicz Hyaluronidase HA

27 Classification of MPS TYPE NAME ENZYME DEFICIENCY GAG MPS I Hurler / Scheie α-L-iduronidase HS DS

MPS II Hunter Iduronate-2-sulfatase HS DS

MPS IIIA Sanfilippo A Heparin sulfamidase HS

MPS IIIB Sanfilippo B N-acetyl-α-D-glucosaminidase HS

MPS IIIC Sanfilippo C Acetyl-CoA:α-glucosaminidase HS

MPS IIID Sanfilippo D N-acetylglucosamine-6-sulfatase HS

MPS IVA Morquio A N-acetylgalactosamine-6-sulfatase KS

MPS VI Maroteaux-Lamy N-acetylgalactosamine-4-sulfatase DS

MPS VII Sly β-Glucuronidase HS DS

MPS IX Natowicz Hyaluronidase HA

28 Classification of MPS TYPE NAME ENZYME DEFICIENCY GAG MPS I Hurler / Scheie α-L-iduronidase HS DS

MPS II Hunter Iduronate-2-sulfatase HS DS

MPS IIIA Sanfilippo A Heparin sulfamidase HS

MPS IIIB Sanfilippo B N-acetyl-α-D-glucosaminidase HS

MPS IIIC Sanfilippo C Acetyl-CoA:α-glucosaminidase HS

MPS IIID Sanfilippo D N-acetylglucosamine-6-sulfatase HS

MPS IVA Morquio A N-acetylgalactosamine-6-sulfatase KS

MPS VI Maroteaux-Lamy N-acetylgalactosamine-4-sulfatase DS

MPS VII Sly β-Glucuronidase HS DS

MPS IX Natowicz Hyaluronidase HA

29 Classification of MPS TYPE NAME ENZYME DEFICIENCY GAG MPS I Hurler / Scheie α-L-iduronidase HS DS

MPS II Hunter Iduronate-2-sulfatase HS DS

MPS IIIA Sanfilippo A Heparin sulfamidase HS

MPS IIIB Sanfilippo B N-acetyl-α-D-glucosaminidase HS

MPS IIIC Sanfilippo C Acetyl-CoA:α-glucosaminidase HS

MPS IIID Sanfilippo D N-acetylglucosamine-6-sulfatase HS

MPS IVA Morquio A N-acetylgalactosamine-6-sulfatase KS

MPS VI Maroteaux-Lamy N-acetylgalactosamine-4-sulfatase DS

MPS VII Sly β-Glucuronidase HS DS

MPS IX Natowicz Hyaluronidase HA

Neuronopathic MPSII

• Between 1 in 100,000 and 1 in 150,000 • X-linked disorder • 2/3 of boys in some way neuronopathic • No current specific treatment but for very early HSCT - infrequent • IV Elaprase used in this group to improve QOL without much evidence base and varies from country to country • Remains the greatest unmet need in MPSII

32 Neuronopathic MPSII

• Most (around 50%) of cases have progressive CNS disease with developmental delay, plateau and regression • 20-30% of boys will have a static neurological disease • Behavioural and sleep disturbance common • Supportive measures relatively poor • Poor relationship between the severity of somatic disease and that of CNS disease in MPSII

33 Intravenous Elaprase in neuronopathic MPSII

• Usually improvements in somatic manifestations • Improved growth • Significant benefits in general well being (less infections, less pain/stiffness) • No discernible effect on cognition or prevention of dementia • Probable positive effect on survival

(Barbara K. Burton, Virginie Jego, Jaromir Mikl & Simon A. Jones, JIMD. 2017)

34 Unmet needs in MPSII

• Central nervous system • Skeletal system • Heart valves • Airway (tracheal) disease • Heterogeneity and incomplete response to Elaprase • Prediction of phenotype

35 Role of Biomarkers

• Obvious biomarker for somatic disease is the stored glycosaminoglycans themselves

36 Muenzer et al 2011 Role of Biomarkers

• Total GAGs in urine lack sensitivity but also an ability to reflect the CNS storage burden • Direct GAG species in blood are now possible to measure but still not the brain • Direct GAG measurement in CSF (DS and HS) likely the most useful • Interpretation of these in terms of an intervention critically depend on the nature of the intervention

37 Clinical endpoints

• Neurocognitive endpoints – Consensus statements important • Behavioural/PROM/QOL important but poorly validated this far • Somatic endpoints – Spirometry – 6 minute walk test

38 TV Flagship Program: ETV:IDS (DNL310)

Carole Ho, MD, CMO & Head of Development DEVELOPING A THERAPY FOR HUNTER SYNDROME

40 BRAIN DELIVERY IS A CRITICAL UNMET MEDICAL NEED

IDS ETV:IDS (DNL310) aims to treat the cognitive aspects

of Hunter syndrome while maintaining physical benefit transferrin receptor (TfR) in the entire body with IV administration epitope

Enzyme Transport Vehicle DENALI INVESTIGATIONAL APPROACH (ETV:IDS)

ETV:IDS has the potential to treat Hunter syndrome with a single weekly administered IV therapeutic

41 ETV:IDS (DNL310) FOR HUNTER SYNDROME

TARGET PATHWAY PATHOLOGY Loss of activity of IDS causes IDS deficiency reduces lysosomal Lysosomal dysfunction Hunter syndrome (MPS II) function causes inflammation and neuronal cell loss

Accumulation and increased Lysosomal dysfunction and increased lipids Increased neurodegeneration with Glycosaminoglycans (GAGs) in disease (Gangliosides, BMP) in disease increased Nf-L

Existing enzyme replacement therapies (ERTs) do not effectively cross the BBB and do not address neurodegeneration

42 ETV:IDS BIOMARKER-DRIVEN DEVELOPMENT STRATEGY

Target Engagement Pathway Engagement Patient Phenotyping Patient Impact

Early dose finding Increase confidence in dose selection Definitive clinical data

Ø Demonstrate clinical efficacy on neurocognitive endpoints

Ø Engaged regulatory • Reduce • Improve lysosomal function strategy accumulation of Ø Reduced abnormal accumulation of lipids IDS substrate • Reduce neuroaxonal injury (GAGs) Ø Nf-L

Phase 1/2 Phase 2/3

Denali’sSeptember biomarker 2019 -driven development strategy applied to development of ETV:IDS for Hunter syndromeJuly 2020

43 ETV:IDS DEVELOPMENT STRATEGY AND DATA OVERVIEW

Pre-clinical Efficacy Studies (Hunter Mouse) + Fluid Biomarker Analysis* (Hunter Patients) Data today (Oct 2020)

Supportive data for Clinical Observational Biomarker Study (n=28) future registration

Phase 1/2 (n=~16) Interim data (end 2020)

Ph2/3 Ph2/3 Phase 2/3 planning startup

UNGATE TV PLATFORM UNGATE Ph2/3 End 2020 Data Mid 2021 Data (GAG reduction) (Data: Lysosomal rescue + Nf-L)

Dose and endpoint selection for Phase 2/3 are informed by robust pre-clinical and clinical data set to maximize probability of success and patient impact *Bhalla and Ravi, et al 2020 44 ETV:IDS DEVELOPMENT STRATEGY AND DATA DETAIL

Data Today (Oct 2020) • Pre-clinical Efficacy • Target engagement (GAG reduction) • Correction of skeletal abnormalities Studies (Hunter Mouse) + • Improved lysosomal function • Improvement motor skills • Fluid Biomarker Analysis • Nf-L reduction • Rescued cognition (Hunter Patients)* Translatable fluid biomarkers of lysosomal function and neuroaxonal injury identified ✓

Clinical Observational • Understand correlation of biomarker changes and clinical endpoints Biomarker Study (n=28) Natural history data will provide support for effects on disease biology at filing ✓

Ungate TV Platform data (End 2020) Ungate Ph2/3 data (Mid 2021) Phase 1/2 (n=~16) • Safety • Safety • GAG reduction • Lysosomal lipid and Nf-L reduction TV platform validation and enables rigorous approach to dose selection ✓

Phase 2/3 • Improvement in cognition and function

Step-wise de-risking of development program *Bhalla and Ravi, et al 2020 45 ETV:IDS CORRECTS LYSOSOMAL FUNCTION AFTER SYSTEMIC ADMINISTRATION

Published May 27, 2020 ETV:IDS reduces lysosomal lipid accumulation in the brain TfR mu/hu IDS KO model

Vehicle Vehicle IDS ETV:IDS GM2 (d36:1) GM3 (d36:1)

4 week treatment, initiation of treatment at 2 months High dose PoC study with comparable enzyme activity for IDS and ETV:IDS IDS KO model (IDS KO; TfRmu/hu) Mouse model lacks IDS and has human TfR

ETV:IDS achieves high concentration and broad distribution of enzymes in the brain IDS does not reduce lysosomal lipid accumulation in the brain and CSF

46 PATIENT PHENOTYPING: DISCOVERY OF PATIENT BIOMARKERS

Published July, 2020

Clinical Status # Subjects Age (SD) Non-MPS Controls 25 5.8 (5.4) MPS II (on ERT) 6 3.3 (1.4) MPS II (pre and post Hematopoietic 2 1 (0) Stem Cell Transplant)

1st paper in Hunter patients to correlate abnormalities in GAGs with biomarkers of secondary lysosomal dysfunction (gangliosides, BMP), neuroaxonal injury (Nf-L), and inflammation

47 TARGET ENGAGEMENT: GAG REDUCTION INFORMED BY PATIENT DATA

Primary storage substrate (GAGS) metabolized ETV:IDS reduces accumulated by IDS is elevated in Hunter patients substrate in Hunter IDS KO Model

Serum CSF CSF Brain

<2x ~ 11x

CSF GAGs are elevated despite SOC Elaprase treatment 13 week treatment, initiation of treatment at 2 months

• ETV:IDS results in a dose dependent reduction in CSF and Brain GAGs in Hunter IDS KO Model • At 1 and 3mg/kg, ~50% reduction in GAGs in CSF is associated with downstream pathway effects *Bhalla and Ravi, et al 2020 48 *Ullman, Arguello, Getz et al 2020 PATHWAY ENGAGEMENT: GANGLIOSIDE REDUCTION INFORMED BY PATIENT DATA

Increased gangliosides due to lysosomal ETV:IDS corrects secondary storage of GM3 and dysfunction in CSF of Hunter patients other lysosomal lipids in Hunter IDS KO Model Brain Serum CSF CSF Brain

0.006 *** 80 **** *** ****

60 0.004 L) ~ 4x µ 40 (ng/ 0.002 (ng/mg) 20 CSF GM3 (d36:1) levels Brain GM3 (d36:1) levels 0.000 0 - - 1 3 - - 1 3

Vehicle ETV:IDS Vehicle ETV:IDS Increased CSF ganglioside levels in TfRmu/hu Ids KO;TfRmu/hu TfRmu/hu Ids KO;TfRmu/hu Hunter patients reflects lysosomal dysfunction in brain 13 week treatment, initiation of treatment at 2 months

• ETV:IDS corrects lysosomal dysfunction (ganglioside accumulation) in brain in a Hunter IDS KO model • At 1 and 3mg/kg, normalization of brain lysosomal lipid accumulation is correlated with CSF lipid reduction

*Bhalla and Ravi, et al 2020 49 *Ullman, Arguello, Getz et al 2020 PATHWAY ENGAGEMENT: BMP AND GLUCER REDUCTION INFORMED BY PATIENT DATA Increased BMP and glucosylceramide due ETV:IDS corrects lysosomal dysfunction to lysosomal dysfunction in CSF of Hunter (accumulation of BMP and Glucosylceramide) in patients Hunter IDS KO Model

* 0.3 **** 0.0008 CSF CSF * Brain ****

0.0006 0.2 l) µ 0.0004 (d18:1/16:0) (ng/mg) (ng/ GluCer (d18:1/16:0) 0.1 GluCer 0.0002 GluCer (ng/mg) (ng/mg) Brain GluCer (d18:1/16:0) CSF GluCer (d18:1/16:0)

GluCer 0.0

GluCer 0.0000 - - 1 3 - - 1 3 CSF CSF (area ratio)

Vehicle ETV:IDS Brain Vehicle ETV:IDS Brain TfRmu/hu Ids KO;TfRmu/hu TfRmu/hu Ids KO;TfRmu/hu **** 6 **** 0.0015 **** ****

4 0.0010 ) L BMP µ (ng/mg)

BMP (ng/ 2

0.0005 (ng/mg) (ng/mg)

Brain BMP (18:1/18:1) levels Brain BMP 0 CSF BMP(18:1/18:1) levels 0.0000

Brain BMP (18:1/18:1) - - 1 3 - - 1 3 CSF BMP (ng/mL) BMP CSF Brain BMP (18:1/18:1) Vehicle ETV:IDS Vehicle ETV:IDS TfRmu/hu Ids KO;TfRmu/hu TfRmu/hu Ids KO;TfRmu/hu 13 week treatment, initiation of treatment at 2 months • ETV:IDS corrects lysosomal dysfunction (GluCer and BMP) in brain in a Hunter IDS KO model • At 1 and 3mg/kg, normalization of brain lysosomal lipid accumulation is correlated with CSF lipid reduction *Bhalla and Ravi, et al 2020 50 *Ullman, Arguello, Getz et al 2020 BIOMARKER OF NEUROAXONAL INJURY INFORMED BY PATIENT DATA

Increased Nf-L reflects neuroaxonal injury Nf-L increases with age in Hunter patients in Hunter IDS KO Model

Serum CSF CSF

~ 6x ~ 5x

~ 5x

ETV:IDS ETV:IDS

Increased serum and CSF levels of Nf-L 1 2 reflects the lack of effect of SOC on neuroaxonal injury

• Nf-L, a marker of neuroaxonal injury, is elevated in Hunter patients and a Hunter IDS KO model • In Hunter KO animals, Nf-L increases and continues to rise with disease progression 51 *Bhalla and Ravi, et al 2020 ETV:IDS SLOWED NEUROAXONAL INJURY IN HUNTER KO MODEL

CSF Nf-L ETV:IDS prevents neuroaxonal injury with ETV:IDS prevents neuroaxonal injury with 1 treatment initiation at 2 months of age 2 treatment initiation at 4.5 months of age

6000 2500 ** ~ 5x *** 2000 * ETV:IDS ETV:IDS 4000 1 2 1500 CD 1000 2000 NfL (pg/mL) NfL NfL (pg/mL) NfL 500 CSF CSF 0 0 ~ 5x dose dose - - 3 - - 1 3 (mg/kg) (mg/kg) Vehicle ETV:IDS Vehicle ETV:IDS TfRmu/hu Ids KO;TfRmu/hu TfRmu/hu Ids KO;TfRmu/hu

13 week treatment, initiation of treatment at 2 months 17 week treatment, initiation of treatment at 4.5 months *Ullman, et al 2020 • ETV:IDS can slow neuroaxonal injury even at a stage where extensive neuroaxonal injury is ongoing • At 3mg/kg, at early & later disease stages, ETV:IDS completely and substantially reduces Nf-L 52 ETV:IDS IMPROVES MOTOR FUNCTION IN HUNTER KO MODEL

GROSS motor impairment in Hunter patients ETV:IDS improves motor skills in mouse model Treadmill Pole Test ) yr

h Developmental age ( Gross motor development motor Gross

Calendar age (yr) 17 week treatment, initiation of treatment at 4.5 months, 3mg/kg, Holt et al 2011 n=19-22 / group

At 3mg/kg, ETV:IDS improves locomotor performance and agility in a Hunter IDS KO model

53 ETV:IDS RESCUES COGNITIVE DEFICITS IN HUNTER KO MODEL

Loss of cognitive milestones ETV:IDS normalizes spatial learning and memory in mouse model In Hunter patients TRAINING TRIALS REINSTATEMENT TRIALS Worse ) yr

sh Cognitive performance Developmental age ( cognitive development Better

Calendar age (yr)

Holt et al 2011 17 week treatment, initiation of treatment at 4.5 months, 3mg/kg, n=19-22 / group

At 3 mg/kg, ETV:IDS normalizes spatial learning and memory deficits in a Hunter IDS KO model

54 ETV:IDS CORRECTS SKELETAL DISEASE IN HUNTER KO MODEL

Skeletal manifestations ETV:IDS improves skeletal disease manifestations in in Hunter syndrome mouse model Vehicle Vehicle ETV:IDS TfR mu/hu IDS KO, TfR mu/hu Femur ROI Trabecular

Peripheral ERT still does not fully address skeletal manifestations of disease including kyphosis, scoliosis,

and claw hand (left) & bone Femur ROI Cortical fragility (osteonecrosis of hip, above) 17 week treatment, initiation of treatment at 4.5 months, 3 mg/kg, n=5 / group Link et al 2010

At 3mg/kg, ETV:IDS corrects abnormal increased trabecular and cortical bone mass in the femur of Hunter IDS KO model imaged with micro-CT

55 ETV:IDS DIFFERENTIATION AND BEST IN CLASS ASSESSMENT

ENABLES NEURO- CSF GAG LYSOSOMAL SINGLE IV DATA ON DEGENERATION SUBSTRATE FUNCTION DELIVERY THERAPY CSF & BRAIN BIOMARKER REDUCTION IMPROVEMENT FOR PHYSICAL & CORRELATION IMPROVEMENT CNS DISEASE Denali ~50% @ 8 weeks *Animal model *Animal model (mouse) IV YES YES ETV:IDS GM3, BMP data (+) Nf-L data (+) (DNL310) Patient data in Q4 ‘20 JCR 32% @ 3 weeks No published No published IV YES NO (JR-141) (58% @ 26 weeks) data data Takeda/ Shire ~70% @ 8 weeks (GAGs measured from No published No published IT NO NO (TAK609/SHP location of IT data data 609) administration) Regenxbio 27-42% @ 8 weeks No published No published ICM NO NO (RGX-121) (3 patients) data data

Proof of concept with ~50% GAG reduction after ~8 weeks dosing in patients anticipated to be associated with subsequent lysosomal function and neurodegeneration biomarker improvement *Bhalla and Ravi, et al 2020 56 *Ullman, Arguello, Getz et al 2020 ENROLLING TWO CLINICAL TRIALS IN HUNTER SYNDROME

Observational study of Treatment Responsive Phase 1/2 study of DNL310 in children with Biomarkers in Hunter syndrome Hunter syndrome

An observational study of urine, blood, and CSF biomarkers Interventional study of DNL310 in ~16 pediatric patients with to explore correlations to clinical outcomes in ~28 patients Hunter syndrome (MPS II). The study includes a 24-week with Hunter syndrome (MPS II) aged 2-30. dose escalation phase with an 18-month safety extension.

Study Status: This study is currently recruiting participants. Study Status: This study is currently recruiting participants. For additional information, visit ClinicalTrials.gov For additional information, visit ClinicalTrials.gov (NCT04251026) (NCT04007536)

Interim PoC readout Aug 2020 expected late 2020

Dose escalation Safety Extension

Phase 1/2 data to demonstrate reduction in CSF GAGs expected by year end 2020

57 DNL310 PHASE 1/2 PATIENT STUDY

STUDY OVERVIEW PART 1 (DOSE ESCALATION)

• Neuronopathic and non-neuronopathic • Treatment naïve or on Elaprase > 4mo • Part 1 (dose escalation, 24 weeks) • Part 2 (safety extension, 18 months) 16 weeks, weekly Patients Cohort A Cohort A (5-10 yo, N= ~4 Neuronopathic) N= ~16 Dose D Dose C Cohort B (2-18 yo) Dose B 12 weeks, weekly Dose A (3mg/kg) o o Key 1 EP (Safety) Key 2 EP Dose TBD • Adverse events • CSF GAGs Dose TBD Cohort B Key • IRR • PK Dose TBD N= ~12 Endpoints • Urine GAGs (EP) Key PD / Exploratory EP CSF biomarkers of lysosomal dysfunction, Patients on SOC ERT switch to Weekly dosing Nf-L DNL310 on Day 1 of study CSF FPI August 2020

• Starting dose of 3mg / kg anticipated to reduce CSF GAGs and is ~3x enzyme activity compared to Elaprase • End 2020: Cohort A on track for Biomarker POC (CSF GAG reduction @ 8 weeks) • Mid 2021: demonstrate effects on lysosomal and neuronal injury (Nf-L) biomarkers in CSF +/- blood 58 ETV:IDS DEVELOPMENT STRATEGY AND DATA OVERVIEW

Preclinical Efficacy Studies (Hunter mouse) + Fluid Biomarker Analysis* (Hunter patients) Data today (Oct 2020)

Supportive data for Clinical Observational Biomarker Study (n=28) future registration

Phase 1/2 (n=~16) Interim data (end 2020)

Ph2/3 Ph2/3 Phase 2/3 planning startup

UNGATE TV PLATFORM UNGATE Ph2/3 End 2020 Data Mid 2021 Data (GAG reduction) (Data: Lysosomal rescue + Nf-L)

2/3 of lysosomal storage diseases ATV ETV have neuronopathic component

60 ETV:IDS CONCLUSIONS

• DNL310 has the potential to treat the neurologic and physical manifestations of Hunter syndrome with a single IV therapeutic, replacing current SOC

• Newly identified biomarkers in Hunter animal models and patients connects GAG reduction to lysosomal rescue (lipid reduction), reduced neurodegeneration (Nf-L), and motor / cognitive improvement – Brain and CSF GAG levels are highly correlated – 50% reduction in GAG results in reduction of lysosomal lipids, reduction in Nf-L, and motor and behavioral benefits

• Step 1: Phase 1/2 interim safety and biomarker data on GAG reduction (end 2020): • Provides proof of TV-enabled biotherapeutics in humans • Triggers Phase 2/3 planning and pre-Phase 3 global regulatory meetings

• Step 2: Phase 1/2 final safety and biomarker data showing rescue of lysosomal function and reduction in Nf-L enables dose selection and go decision for Phase 2/3 (mid 2021)

• Step 3: Positive neurocognitive data from Phase 2/3 enables registration

61 Q&A

62 Break TV Discovery Programs

Joe Lewcock, PhD, CSO OUR TV PORTFOLIO

DRUG DEVELOPMENT PROGRAM TARGET DRUG CANDIDATE DISEASE INDICATION PARTNER Drug Discovery IND-Enabling Early Clinical Late Clinical Approved

ATV – Antibody Transport Vehicle

TREM2 DNL919 Alzheimer’s Takeda

Abeta ATV:Abeta Alzheimer’s Biogen

Tau ATV:Tau Alzheimer’s Takeda

Alpha-Synuclein ATV:aSyn Parkinson’s, DLB, MSA

HER2 ATV:HER2 Oncology

ETV/PTV – Enzyme/Protein Transport Vehicle

Iduronate 2-sulfatase DNL310 MPS II (Hunter Syndrome)

PGRN DNL593 Frontotemporal Dementia Takeda

Sulfamidase ETV:SGSH MPS IIIA (Sanfilippo Syndrome)

Undisclosed ETV:LF1 LSD with Neurodegeneration

Undisclosed ETV:LF2 Parkinson’s

OTV – Oligonucleotide Transport Vehicle

Undisclosed OTV:CH2 Alzheimer’s

Undisclosed OTV:LF3 Parkinson’s

Undisclosed targets: LF - Lysosomal Function target; CH - Cellular Homeostasis target 65 MODALITY-BASED BRAIN DELIVERY PLATFORMS ENABLED BY TV

ATV ETV

66 EXPANDING THE THERAPEUTIC LANDSCAPE WITH TV DELIVERY

ENHANCE TARGETS UNLOCK TARGETS Improved efficacy with TV Creation of new opportunities

ATV:Aβ ATV:TREM2 ATV:HER2 ETV:IDS ETV:SGSH OTV PTV:PGRN (DNL310)

Increase biodistribution (10-30X) and enhance Unlock brain delivery of biotherapeutics activity by combining TfR with Target with systemic administration

67 ATV:Aβ SHOWS INCREASED BRAIN UPTAKE ACROSS LOW DOSES

TfRms/hu 1d

ms/hu : IV of 1, 3, or 10 mg/kg

DESIGN 5XFAD; TfR Brain 15 TfRms/hu 5XFAD; TfRms/hu ATV:Aβ No Aβ Plaques Aβ Plaques Present ****

10 **

cisLALA **** huIgG [nM] 5 ** ****

α-Aβ ATV α-Aβ ATV α-Aβ ATV α-Aβ ATV α-Aβ ATV α-Aβ ATV 1mpk 3mpk 10mpk 1mpk 3mpk 10mpk

ATV:Aβ significantly increases brain uptake across clinically-relevant doses

5xFAD = Alzheimer’s mouse model, TfRms/hu =human TfR knock-in Data represented as mean +/- SEM 68 cisLALA = LALA mutation to ablate effector function one side of Fc only One-way ANOVA, Sidak post-hoc multiple comparisons **p<0.01, ****p<0.0001 7.7X

4.7X

5.5X 6.7X 8.2X 4.8X ATV:Aβ SHOWS INCREASED PLAQUE DECORATION ACROSS LOW DOSES

TfRms/hu 1d

ms/hu : IV of 1, 3, or 10 mg/kg

DESIGN 5XFAD; TfR 10mg/kg Anti-Aβ 10mg/kg ATV:Aβ Aβ Plaques huIgG Aβ + huIgG Colocalization 0.4 TfRmu/hu 5XFAD; TfRmu/hu 17.0x ** 0.3

Colocalization 0.2 16.1x * huIgG

% Colocalization 0.1 15.6x ** % Aβ %

100 µm 0.0

α-Aβ ATV α-Aβ ATV α-Aβ ATV α-Aβ ATV 10 mpk 1 mpk 3 mpk 10 mpk ATV:Aβ significantly increases huIgG immunodecoration of plaques across clinically-

relevant doses Data represented as mean +/- SEM 69 One-way ANOVA, Sidak post-hoc multiple comparisons *p<0.05, **p<0.01 TREM2 THERAPEUTIC HYPOTHESIS: IMPROVE MICROGLIAL FUNCTION THROUGH ENHANCEMENT OF TREM2 ACTIVITY

Lipids & Apo-/Lipoproteins

mutations (e.g. R47H) TREM2

ADAM10/17 cleavage site

microglial cell membrane

DAP12 PIP2 OH DAG P P P SYK SYK P PLC2 IP3 P P

ITAM • Lipid processing • Phagocytosis • Transcription • Migration Activation State • Survival • Cytokine release

LOAD Gene Odds Ratio Disease effect Variant 2.08 TREM2 R47H risk (1.73 – 2.49) 1.67 TREM2 R62H risk (1.46 – 1.90)

PLCG2 P522R 0.68 protective 70 ATV:TREM2 INDUCES ROBUST PHARMACODYNAMIC RESPONSE IN BRAIN

Single Dose - 10 mg/kg

Brain huIgG ATV:Isotype TREM2 Antibody ATV:TREM2 4 Iba1 ATV:TREM2 3

1 Antibody concentration [nM] 0

TREM2

ATV:TREM2 Multi Dose – Matched Brain Exposure Brain huIgG Iba+ Microglia 100 8 **** **** 6 ATV:TREM2 has enhanced

10 4 activity compared to similar concentrations of anti-TREM2 /parenchyma 2 Relative Iba1+ area Brain concentration [nM] 1 0

Isotype TREM2 Isotype TREM2 Graphs represent data from 7 weeks dosing, 100mpk 100mpk 100mpk 100mpk 71 ATV:Isotype10mpk ATV:TREM210mpk ATV:Isotype10mpk ATV:TREM210mpk Data represented as mean +/- SEM , *p<0.05, **p<0.01, ****p<0.0001 TV HAS APPLICATIONS OUTSIDE OF NEURODEGENERATION

Brain huIgG Concentration Growth Inhibition 50 mg/kg, 24h post dose HER2+ OE19 cells 60 150 ATV:HER2 Untreated trastuzumab ATV:tras ****

40 **** 100

20 50 % Cell Viability Brain huIgG (nM)

0 0 0.01 0.1 1 10 100 1000

ATV:tras ATV:per Concentration (nM)

Trastuzumab Pertuzumab

TfRms/hu TfR binding of ATV:HER2 results in both enhanced brain uptake and improved activity in vitro

Data represented as mean +/- SEM Brain huIgG: One-way ANOVA, Sidak’s multiple comparison correction; ****p<0.0001; 72 xenogratft: Unpaired t-test,,**p<0.01, ****p<0.0001 TV HAS APPLICATIONS OUTSIDE OF NEURODEGENERATION

Brain huIgG Concentration HER2+ Xenograft (BT474) 50 mg/kg, 24h post dose trastuzumab + pertuzumab 60 ATV:HER2 150 ATV:tras + ATV:per ****

ns **** ** 40 **** 100

50 20 Brain huIgG (nM)

0 % Tumor Growth Inhibition Tumor % 0 3 + 3 10 + 10 20 + 20

ATV:tras ATV:per Dose of each test article (mg/kg)

Trastuzumab Pertuzumab

TfRms/hu ATV:HER2 displays enhanced anti-tumor efficacy in a xenograft model

Data represented as mean +/- SEM Brain huIgG: One-way ANOVA, Sidak’s multiple comparison correction; ****p<0.0001; 73 xenogratft: Unpaired t-test,,**p<0.01, ****p<0.0001 MODALITY-BASED BRAIN DELIVERY PLATFORMS ENABLED BY TV

ATV ETV

74 ETV ENABLES CNS DELIVERY OF ADDITIONAL LYSOSOMAL ENZYMES

Brain Heparan Sulfate CSF Heparan Sulfate **** **** 20 ETV:SGSH ns 9 **** 15 ** 6 **** 10 (D0A0, D0S0) 5 (D0A0, D0S0) 3 Total Heparan Sulfate (ng/ug) 0 Total Heparan Sulfate0 (ng/uL)

Vehicle Vehicle Vehicle Vehicle Fc:SGSH Fc:SGSH ETV:SGSH ETV:SGSH

TfRms/hu SgshD31N; TfRms/hu TfRms/hu SgshD31N; TfRms/hu

ETV:SGSH leads to >50% reduction in brain Heparan Sulfate (HS) and >80% reduction in CSF HS following a single dose in MPSIIIA mice Data shown as mean ± s.e.m.; **** p < 0.0001; ** p 0.001 One-way ANOVA, Tukey’s multiple comparison correction 75 Duration is 7d post single dose in ~3mo old mice @ treatment initiation ETV PLATFORM HAS BROAD POTENTIAL FOR MANY LYSOSOMAL STORAGE DISEASES

ETV:IDS (Hunter) ETV:SGSH (MPS IIIA) ETV:LF1

~ 2,500 – 4,000 Patients (US)

Other Lysosomal Storage Diseases with CNS Pathology ~ 10,000+ Patients (US) Denali has expanded portfolio with initiation of two additional ETV programs for treatment of LSDs 76 MODALITY-BASED BRAIN DELIVERY PLATFORMS ENABLED BY TV

ATV ETV

77 PROGRANULIN DEFICIENCY RESULTS IN LYSOSOMAL DYSFUNCTION GENETICS PATHWAY PATHOLOGY Mutations in GRN result in decreased PGRN deficiency dysregulates Lysosomal dysfunction causes levels of protein lysosomal function inflammation and neuronal cell loss

PGRN (human CSF) PGRN LAMP1 Nf-L (human CSF)

* * GRN WT iPSC-MG GRN KO iPSC-MG GRN HC GRN 78 HC FTD Patient CSF obtained and analyzed by Denali as part of LEFFTDS & ARTFL Consortium FTD LIPIDOMICS IDENTIFIES ALTERED BMP LEVELS IN GRN KO MOUSE BRAIN Brain BMP BRAIN: Grn KO vs. Grn WT

value (unadjusted) value 0.001 Brain Lysosomal Lipid - p 0.01 0.05

1 Lower in Grn KO Elevated in Grn KO 20 30 50 80 100 125 200 300 500 Percent WT (%) PGRN deficiency reveals lipid changes indicative of lysosomal dysfunction

79 BMP IS A UNIQUE PHOSPHOLIPID ENRICHED IN ENDOLYSOSOMES

Bis(monoacylglycero)phosphate (BMP)

• Phospholipid concentrated in the intraluminal vesicles (ILVs) of late endosomes and lysosomes

• Also found on exosomes and can therefore be detected in extracellular environment and biofluids

• Hypothesized to play a role in proper function of lysosomal enzymes

• Altered BMP levels observed in patients with lysosomal storage disorders

80 PTV:PGRN RESTORES LYSOSOMAL FUNCTION IN BRAIN

Brain BMP Liver BMP PTV:PGRN

PTV:PGRN

PTV:PGRN but not Fc:PGRN rescues Brain BMP levels in Grn KO mice

data shown as mean ± s.e.m.; **** p < 0.0001 81 One-way ANOVA, Sidak’s multiple comparison correction PTV:PGRN PROVIDES SUSTAINED RESCUE OF LYSOSOMAL FUNCTION AND MICROGLIAL ACTIVATION

High dose 1, 3, 7, 10 days

Brain huIgG Brain BMP Brain Trem2 Drug Concentration Lysosome Dysfunction Microglial Activation

PTV:PGRN Grn KO

Grn WT

Grn KO

Rescue of BMP and TREM2 levels maintained after PTV:PGRN is cleared

82 PTV:PGRN PROVIDES SUSTAINED RESCUE OF LYSOSOMAL FUNCTION AND MICROGLIAL ACTIVATION

High dose 1, 3, 7, 10 days Low dose 2, 3, 6 weeks

Brain huIgG Brain BMP Brain BMP Drug Concentration Lysosome Dysfunction Lysosome Dysfunction

PTV:PGRN

Grn WT Grn WT

Grn KO Grn KO

Weeks after IV Dosing

Rescue of BMP and TREM2 levels maintained after PTV:PGRN is cleared

83 CSF BMP IS A CLINICALLY TRANSLATABLE BIOMARKER IN FTD

CSF BMP correlated brain BMP in Grn KO CSF BMP levels are decreased in FTD patients and can be rescued with PTV:PGRN

Low dose 4, 24 hours di-18:1 BMP (CSF) di-22:6 BMP (CSF)

CSF BMP Brain:CSF BMP correlation 24 hours

HC Non-Grn Grn HC Non-Grn Grn FTD FTD FTD FTD

HC: Healthy Control

IND planned for PTV:PGRN planned in late 2021/early 2022

84 Patient CSF obtained and analyzed by Denali as part of LEFFTDS & ARTFL Consortium MODALITY-BASED BRAIN DELIVERY PLATFORMS ENABLED BY TV

ATV ETV

85 OLIGONUCLEOTIDE TRANSPORT VEHICLE (OTV) PLATFORM

SUCCESSFUL SITE-SPECIFIC OTV CONJUGATION

DAR = Drug Antibody Ratio Oligo Sequence / DAR 1 Chemistry

Fab Binding Target

Conjugation Site / Chemistry TfR Binding Affinity DAR 0 DAR 2 OTV

OTV is a modular platform with multiple properties that can be optimized for specific targets

86 OTV KNOCKS DOWN TARGET IN CNS WITH SYSTEMIC ADMINISTRATION

CortexCortex Spinal Cord 150 : Single dose 150 : Single dose : Multi dose : Multi dose OTV

100 100

50 50 Malat1 mRNA (% ATV:BACE1) (% Malat1 mRNA Malat1 mRNA (% ATV:BACE1) (% Malat1 mRNA Malat1 mRNA (% ATV Control) ATV (% Malat1mRNA ns ns Control) ATV (% Malat1mRNA ns ns 0 * **** 0 * **** ATVATV ASO mAb OTVOTV OTV ATVATV ASO mAb OTVOTV OTV ControlBACE1 ASO Single Multi ControlBACE1 ASO Single Multi

OTV selectively and cumulatively knocks down ASO target in CNS with peripheral delivery

Each IV dose = 2.5mg/kg ASO or molar equivalent Data represented as mean +/- SEM 87 Multi-dose = 4X 2.5mg/kg ASO or molar equivalent weekly One-way ANOVA, Sidak post-hoc multiple comparisons *p<0.05, ****p<0.0001 OTV HAS BROAD POTENTIAL IN A RANGE OF INDICATIONS

OTV Potential Spans Across: APP Indications LRRK2 • Alzheimer’s disease MAPT SNCA • Parkinson’s disease ApoE4 • Huntington’s disease SOD1 HTT • Amyotrophic lateral sclerosis (ALS) C9orf72 • Other Neurodegenerative/ FUS mHTT SNP1 Neuromuscular disorders ATXN2 mHTT SNP2 • Other Neurology disorders

PrP Targets GSY1 SMN • Range of additional targets possible GFAP OTV

UBE3A CD49D SCN1A

PMP22 SCN2A DMD

Select Existing ASO Targets 88 DISCOVERY TV PROGRAMS CONCLUSIONS

• The TV platform has potential for broad applicability and is capable of delivering a range of therapeutics to the CNS

• TV technology can enhance activity for antibody targets – ATV:Aβ displays significantly improved engagement of plaques in mouse model of Alzheimer’s disease – ATV:TREM2 shows enhanced brain uptake and microglial activation compared to anti-TREM2 – ATV:HER2 displays improved anti-tumor activity and increased brain concentrations; potential for treatment of CNS metastases

• TV technology unlocks systemic delivery of enzymes, proteins, and oligonucleotides – ETV:SGSH reduces heparan sulfate accumulation in brains of mice; potential to accelerate this and other enzymes following clinical proof of concept for ETV:IDS – PTV:PGRN delivers PGRN to the lysosome and corrects BMP deficits in mice; IND is planned for late 2021/early 2022 – Collaboration initiated with Secarna to enable development of two OTV targets

89 Unlocking the Potential of the TV Platform

Alex Schuth, MD, COO BROAD POTENTIAL FOR A WIDE RANGE OF INDICATIONS AND TARGETS

CURRENT FUTURE FOCUS OPPORTUNITIES

NEUROLOGY CURRENT PLATFORMS e.g., pain, epilepsy, NEURO- neuropsychiatry, DEGENERATION neuromuscular

e.g., AD, PD, ALS Enzymes Proteins Antibodies Oligos

Novel RMT Other ONCOLOGY Targets Gene Therapies Modalities e.g., CNS metastases POTENTIAL FUTURE PLATFORMS LYSOSOMAL STORAGE INFECTIOUS DISORDERS DISEASES e.g., Hunter Syndrome

91 OUR TV PORTFOLIO

DRUG DEVELOPMENT PROGRAM TARGET DRUG CANDIDATE DISEASE INDICATION PARTNER Drug Discovery IND-Enabling Early Clinical Late Clinical Approved

ATV – Antibody Transport Vehicle

TREM2 DNL919 Alzheimer’s Takeda

Abeta ATV:Abeta Alzheimer’s Biogen

Tau ATV:Tau Alzheimer’s Takeda

Alpha-Synuclein ATV:aSyn Parkinson’s, DLB, MSA

HER2 ATV:HER2 Oncology

ETV/PTV – Enzyme/Protein Transport Vehicle

Iduronate 2-sulfatase DNL310 MPS II (Hunter Syndrome)

PGRN DNL593 Frontotemporal Dementia Takeda

Sulfamidase ETV:SGSH MPS IIIA (Sanfilippo Syndrome)

Undisclosed ETV:LF1 LSD with Neurodegeneration

Undisclosed ETV:LF2 Parkinson’s

OTV – Oligonucleotide Transport Vehicle

Undisclosed OTV:CH2 Alzheimer’s

Undisclosed OTV:LF3 Parkinson’s

Undisclosed targets: LF - Lysosomal Function target; CH - Cellular Homeostasis target 92 KEY TV PARTNERSHIPS Enabling collaboration Development/commercial collaboration

TV engineering and platform IP Research collaboration on select TV-enabled targets

Therapeutic mAb discovery Discovery of novel AAV capsids to enable BBB-crossing therapeutics Two TV platform programs, including ATV:Abeta

Three TV platform programs: ATV:TREM2, PTV:PGRN, Discovery of novel ASO ATV:Tau therapeutics enabled by TV

ESTABLISHED NETWORK OF COLLABORATORS TO BUILD AND DEVELOP BROAD TV PORTFOLIO 93 COLLABORATIONS REALIZE VALUE AND ENABLE NEW OPPORTUNITIES

TAKEDA (JANUARY 2018) BIOGEN (AUGUST 2020)

• Exclusive option to three TV programs • LRRK2 co-development/co-commercialization • PTV:PRGN • TV Option agreement • ATV:TREM2 • Exclusive option to two TV programs (incl. • ATV:Tau ATV:Abeta) • Opt-in at IND • Right of first negotiation for two TV programs • Option rights limited to ATV/ETV programs for Alzheimer’s, Parkinson’s, ALS, MS

§ Upfront Payment: $150M (cash and equity) § Upfront Payment: $1.025B (cash and equity) § Downstream payments if Takeda exercises three options: § Downstream payments if Biogen exercises both TV options: § Up to $90M in pre-clinical milestones and opt-in payments § Up to $270M development milestones and opt-in payments § Up to $707.5M clinical and regulatory milestones § Up to $615M commercial milestones § Up to $225M commercial milestones § Mid-single digit to mid-teens royalties § Co-development and co-commercialization with 50/50 world- § Biogen responsible for development and commercialization wide cost and profit split § Downstream financials for ROFN programs to be discussed if/when Denali seeks a partner

94 VALUE UNLOCKED THROUGH DATA AND PARTNERSHIPS

Broader Neurological Disease

OTV

Alzheimer’s Bivalent ATV Parkinson’s Oncology DNL310 PTV (ETV:IDS) 10-30x Brain ETV

Impact & Value & Impact Exposure

Transport Vehicle Lysosomal Storage Diseases, FTD TfR binding Hunter Syndrome 2000 - 2015 2015 2017 End of 2020 2021+

BBB & RTM TV Technology Nonhuman Human Platform Expansion Basic Science Development primate POC POC and Acceleration

95 THANK YOU

96 Q&A