Publications

The work, in plain language

Protein Science · 2025

Proline variants in the BRCA1 coiled-coil domain disrupt folding and binding to PALB2

Baker CNS, Pajela PGC, Martin DE, Dzyuba SV, Stewart MD. Protein Science. 2025;34(1):e5240.

What we asked

When BRCA1 carries a variant of unknown significance, can we tell whether it still binds its repair partner PALB2 — quickly and cheaply?

What we found

A simple heat-based binding test reliably separates harmful variants from harmless ones, flags two new likely-harmful ones, and reveals that a proline anywhere in the region collapses the helix and breaks binding.

Why it matters

The test takes about four days and can be run by undergraduates — a realistic first-pass screen to help classify the variants actually found in patients.

Read the paper (open access) →

Biomolecules · 2024

Installation of an indole on the BRCA1 disordered domain using triazine chemistry

Claton LE, Baker C, Martin H, Dzyuba SV, Zaman K, Prokai L, Stewart MD, Simanek EE. Biomolecules. 2024;14(12):1625.

What we asked

The stretch of BRCA1 that grips PALB2 carries no built-in fluorescent handle, which makes it hard to watch. Could we clip one on without changing how the protein behaves?

What we found

A simple, inexpensive labeling reaction attaches a glowing indole tag at a single site, and the tagged protein keeps both its shape and its normal grip on PALB2.

Why it matters

It gives the lab a fast way to watch the BRCA1–PALB2 grip and to screen large sets of variants for the ones that break it.

Read the paper (open access) →

microPublication Biology · 2024

One of these strains is not like the others: C. elegans DW102 has an altered dependence on brc-1 and brd-1 for regulation of cyp gene transcription

Thapa I, Sellin Jeffries MK, Stewart MD. microPublication Biology. 2024.

What we asked

Does a widely-used worm strain really behave like a double mutant of the two BRCA1-like genes, as the field assumes?

What we found

No. Most of its expected gene-silencing changes don't appear, and one gene shifts the opposite direction — suggesting the strain carries hidden, unreported differences.

Why it matters

Many labs build experiments on this strain. Flagging the discrepancy protects everyone's conclusions and keeps a shared research tool honest.

Read the paper (open access) →

Nucleic Acids Research · 2023

Conservation of transcriptional regulation by BRCA1 and BARD1 in Caenorhabditis elegans

Thapa I, Vahrenkamp R, Witus SR, Lightle C, Falkenberg O, Sellin Jeffries MK, Klevit RE, Stewart MD. Nucleic Acids Research. 2023;51(5):2108–2116.

What we asked

BRCA1 and BARD1 silence certain genes in human cells. Do their counterparts in C. elegans — a worm we can study throughout life — do the same job?

What we found

Yes. The worm proteins tag histone H2A and switch off the same family of estrogen-processing genes, even though one partner grips the spool a different way.

Why it matters

It establishes the worm as a living model for BRCA1/BARD1, opening the door to studying these tumor-suppressor functions in a whole animal, across development.

Read the paper (open access) →

Biochemical Journal · 2021

The BRCA1/BARD1 ubiquitin ligase and its substrates

Witus SR, Stewart MD, Klevit RE. Biochemical Journal. 2021;478(18):3467–3483.

What we asked

BRCA1/BARD1’s one enzymatic job is to clip ubiquitin tags onto other proteins — but which proteins, what does each tag do, and where has the field gone wrong studying it?

What we found

A working map of BRCA1/BARD1’s known targets and the different tags it leaves — some marking proteins for destruction, others switching genes off — plus the experimental traps to avoid and the structure that shows how it singles out histone H2A.

Why it matters

It’s a roadmap. Pinning each tag to a function is how the thousands of “unknown” BRCA1 and BARD1 variants might one day be read as harmful or harmless.

Read the paper →

Nature Structural & Molecular Biology · 2021

BRCA1/BARD1 site-specific ubiquitylation of nucleosomal H2A is directed by BARD1

Witus SR, Burrell AL, Farrell DP, Kang J, Wang M, Hansen JM, Pravat A, Tuttle LM, Stewart MD, Brzovic PS, Chatterjee C, Zhao W, DiMaio F, Kollman JM, Klevit RE. Nature Structural & Molecular Biology. 2021;28(3):268–277.

What we asked

BARD1 lets BRCA1 tag the right spot on histone H2A — but what does that actually look like? We set out to see the machine gripping its target.

What we found

The first structure of the BRCA1/BARD1 complex docked on a nucleosome. BARD1 makes a previously unseen contact that tilts the enzyme and aims the tag at the correct lysines — and cancer-linked BARD1 changes sit right at that contact.

Why it matters

Seeing exactly how BARD1 directs the tag explains why specific inherited changes break it, sharpening which variants should be treated as dangerous.

Read the paper →

PNAS · 2018

De novo mutation in RING1 with epigenetic effects on neurodevelopment

Pierce SB, Stewart MD, Gulsuner S, Walsh T, Dhall A, McClellan JM, Klevit RE, King M-C. PNAS. 2018;115(7):1558–1563.

What we asked

A child had a neurodevelopmental syndrome with no known cause. Was a single new change in RING1 — a protein that tags histone H2A — to blame, and how?

What we found

The change left the enzyme able to do its chemistry but unable to tag H2A on the DNA spool. Recreated in a worm, the same change disrupted how neurons migrated and wired — even with a healthy second copy, matching the patient.

Why it matters

It tied a specific epigenetic tag to building a nervous system, and pointed to disrupted histone tagging as a cause of neurodevelopmental disorders.

Read the paper →

PNAS · 2018

BARD1 is necessary for ubiquitylation of nucleosomal histone H2A and for transcriptional regulation of estrogen metabolism genes

Stewart MD, Zelin E, Dhall A, Walsh T, Upadhyay E, Corn JE, Chatterjee C, King M-C, Klevit RE. PNAS. 2018;115(6):1316–1321.

What we asked

BRCA1’s partner BARD1 also carries inherited breast-cancer mutations — but what does its half of the machine actually do?

What we found

BARD1 is the piece that lets the pair grip the DNA spool. Cancer-linked BARD1 changes leave the enzyme otherwise working, yet specifically stop it from tagging histone H2A — and breast cells then over-produce estrogen-processing enzymes whose byproducts damage DNA.

Why it matters

It gave the BARD1 RING domain a job for the first time and singled out H2A tagging as a function worth watching when judging whether a variant is dangerous.

Read the paper →

More

Full publication list

The complete, always-current list lives on ORCID.

View ORCID profile →