Dr. Fiona Marshall: Three Decades Inside GPCR Drug Discovery

This conversation traces how structure-based GPCR drug discovery moved from a stubborn crystallography problem to a generative platform — through the career of a scientist who helped engineer that shift. Dr. Fiona Marshall walks through the decade when membrane proteins kept failing to crystallize in detergent, and how Chris Tate's thermostabilization strategy finally made GPCR structures routine, the insight that seeded Heptares Therapeutics. The conversation moves across discoveries that shaped a generation of GPCR pharmacology: the GABA-B heterodimer, the free fatty acid receptors, the unexpected allosteric pockets that emerged once structures became available. It also confronts where the field still struggles — translating biased agonism from cell line to clinic, drugging orphan receptors without obvious phenotypes, and expanding the voices that define the field's agenda. For Dr. Marshall, the throughline is personal: she remembers the exact lecture at Bath where she first heard about the β-adrenergic receptor, and has spent three decades inside the rooms where GPCR drug discovery actually changed.

About the Guest

Dr. Fiona Marshall is VP and Head of Neuroscience Discovery at MSD, leading teams in West Point, Boston, and London. She co-founded Heptares Therapeutics and served as Chief Scientific Officer for twelve years, through the company's acquisition by Sosei, helping establish structure-based drug design as a productive strategy across GPCR families. Earlier, at GSK, she contributed to the identification of RAMP proteins, the GABA-B heterodimer, and ligands for orphan receptors including what are now known as FFA2 and FFA3. Her work spans GPCR structural biology, membrane protein stabilization, allosteric modulation, and the pharmacology of receptor families implicated in neuroscience, immunology, and metabolism.

Scientific Themes of the Conversation

• Structure-based GPCR drug design and the thermostabilization breakthrough

• GPCR heterodimerization and the logic of accessory proteins (RAMPs, GABA-B1/B2)

• Allosteric modulation and non-orthosteric binding sites on GPCRs

• Biased agonism, target validation, and animal-to-human translation failure

• Orphan GPCRs and the missing pharmacology of ~100 unassigned receptors

• GPCRs in oncology, immunology, and microbiome-host signaling

Key Insights from the Conversation

Why Membrane Proteins Wouldn't Crystallize The bottleneck wasn't expression level, as most groups had assumed — it was the instability of the receptor in detergent. Chris Tate's thermostabilization strategy, which introduced stabilizing mutations selected by thermal assay, reframed the problem and opened the door to routine GPCR crystallography. Dr. Marshall is clear that the conceptual reframing, not a brute-force technical push, is what unlocked the field.

The GABA-B Heterodimer Was Hiding in Plain Sight Dr. Marshall's team at GSK cloned GABA-B1, tried to get it to signal, and couldn't. A screen run to find an intracellular scaffolding partner pulled up something unexpected: a second GABA-B receptor. The moment baclofen activated ion channels in Xenopus oocytes expressing both — with Dr. Marshall in the room — a new category of GPCR biology opened, with heterodimers as obligate signaling units.

The Ligand That Wasn't the Ligand Screening orphan receptors in yeast, the team chased peptide hits that made no SAR sense. The realization — triggered by Andrew Brown sitting with a table of actives and inactives — was that the active condition was the acetate buffer, not the peptide. The orphan was a free fatty acid receptor. Years of assumed peptide biology dissolved into a simpler truth hidden in the solvent.

Structures Reveal Pockets the Field Didn't Know Existed Once GPCR structures became routine, unexpected binding sites emerged: extra-helical pockets on the glucagon receptor, intracellular sites for C5a antagonists, allosteric modulators bound far outside canonical regions. These discoveries rewrote what counts as "druggable" on receptors once considered fully mapped.

Biased Agonism Needs Human Biology, Not Just Pathway Dials The promise of biased ligands — pulling one signaling arm while sparing another — only survives translation if the coupling profile in the relevant human cells matches the cell-line model. Dr. Marshall's caution is hard-won: many animal-model-validated GPCR drugs have failed in the clinic precisely because this step was skipped.

Oncology Ignored GPCRs for Decades — and Is Paying Attention Now Twenty years ago, cancer biologists dismissed GPCRs: they weren't oncogenic and didn't kill tumor cells directly. The shift came through the tumor microenvironment — immune cells, chemokine receptors, adenosine signaling — where GPCRs now sit at the center of immuno-oncology strategies that tumor-intrinsic target lists missed.

Thirty Years, and the Same Names Keep the Mic One of Dr. Marshall's most pointed observations: the people speaking at GPCR conferences today are, in large part, the people who were speaking at them twenty-five years ago. She frames this as a scientific problem — a narrowing of perspectives on hard problems — not only a social one, and argues the old guard should actively make room for the next cohort.

Episode Timeline

Timestamps were generated using AI for readability.

00:00 Welcome and guest introduction

02:32 How an undergraduate lecture on β-adrenergic signaling set a career in motion

07:07 A venture capital visit to the LMB and the founding of Heptares

11:28 Building a membrane protein team from scratch

15:29 Unexpected allosteric pockets and the next wave of GPCR drug design

18:47 Which GPCR families pharma is chasing now — and why

21:23 Why animal model validation keeps breaking in clinical trials

27:33 How oncology finally stopped ignoring GPCRs

36:39 The GABA-B heterodimer — the moment baclofen finally worked

40:06 The acetate-in-the-buffer discovery

41:47 Thirty years in, and the same speakers keep getting the mic

Selected Quotes

About this episode

Fiona Marshall got fascinated with GPCRs after attending a lecture on how the beta-adrenergic receptor in the heart is activated by adrenaline, during her undergraduate studies at Bath University. She then pursued her Ph.D. in neuroscience at Cambridge University. An expert in GPCR biology, Fiona published the first description of the cloning and structural requirements of the GABAB receptor. One of her career path-defining moments came when she visited Dr. Chris Tate and Dr. Richard Henderson at the Laboratory of Molecular Biology in Cambridge, UK. As a co-founder of Heptares Therapeutics, now called Sosei Heptares, a GPCR-focused drug discovery and development biotechnology company, Fiona and her team made considerable breakthroughs in the field of GPCR stabilization and structure-based drug design.

Today, Dr. Marshall is the VP Head of Neuroscience Discovery and Head of Discovery UK, Global Head of Neuroscience discovery research leading teams in West Point, Boston, and London at MSD.

Join me and learn more about her fascinating career trajectory.

Dr. Fiona Marshall on the web

• LinkedIn

• Twitter

• Google Scholar

• MSD UK

• Dr. GPCR Ecosystem