We Contain Multitudes
The Dusan Bogunovic lab closely analyzes cells from patients with genetic conditions to understand how disease severity is influenced by cells’ decisions about which copy of a gene to express.
By Alan Dove | Portraits by Jörg Meyer
A patient walks into a doctor’s office with symptoms of an inherited disease, but the diagnosis doesn’t add up. According to a quick genetic test, they lack the mutation that’s supposed to cause the condition, which also developed later in life than it should have. What’s going on?
Traditionally, these cases would be lumped under labels such as “idiotypic” or “unknown etiology.” However, as genome sequencing and gene expression profiling have gotten cheaper and faster, researchers have found a surprising new explanation: somatic mosaicism.
Genetics textbooks describe the genome as a single blueprint, and many of us grew up learning that every cell in our body (sperm and eggs excepting) contains the same genome. But that’s turning out to be more of a starting point. Somatic mutations accumulate as cells divide, even during embryogenesis, and are wellknown triggers for cancer development. But recent discoveries show that acquired genetic changes are more common than previously believed and may also drive many more diseases never thought to have genetic underpinnings.
“If you are considering a variant that occurred at a certain point in a person’s development, that variant could be isolated to cells in the brain, or it could be isolated to a person’s germline, in sperm or egg, in which case it might not impact that person at all—but it might impact their future children,” says Jennifer Posey, MD, PhD, chief genomics officer at CUIMC and associate professor of pediatrics and medicine.
No One Is Immune
For doctors unaccustomed to thinking about genetic disease, recent findings on somatic mosaicism may come as a shock.
Jason Liebowitz, MD, a rheumatologist and assistant professor of medicine at VP&S, had that experience in 2020, when a team at the National Institutes of Health (NIH) published their discovery of VEXAS, a new inflammatory condition caused by mutations that appear in some blood stem cells. “VEXAS was a landmark paradigm shift in the way we think about a disease that arises due to somatic mutations and has features that look like our classical autoimmune diseases,” says Dr. Liebowitz.
Patients with VEXAS often present with odd constellations of autoinflammatory symptoms. Without sequencing the affected blood cells (not all possess the mutation), physicians struggle to diagnose them. Dr. Liebowitz points to a patient his team saw recently: Though her initial symptoms resembled lupus, she went on to develop features of spondyloarthritis and vasculitis. “Now this was starting to be weird. Why would this person have features of many different distinct autoimmune conditions that don’t typically swim in the same pool?” says Dr. Liebowitz.
Sequencing her blood cells’ DNA, however, suggested a VEXAS diagnosis. “Ten years ago, we would’ve tried to fit her into one of our clinical boxes, but now we have a molecular diagnosis for her, and it is going to inform her treatment and what we can offer her,” says Dr. Liebowitz, who now advocates for widespread genetic testing in medicine. “I’m definitely incorporating genetic sequencing more and more into my daily practice. Sometimes it’s leading to a specific diagnosis, sometimes not, but it’s still worthwhile,” he says.
Deeper genetic testing, including performing genome sequencing and testing multiple cell types, could also help researchers find better explanations for the heterogeneity of some autoimmune diseases. While many, such as rheumatoid arthritis, have clear disease courses, others remain frustratingly variable. Lupus, for example, can cause anything from a mild rash and joint pain to multiple organ failure. “Lupus feels strongly like a disease where we’ve grouped a large heterogeneous group of patients in one bucket, but maybe genetics will help us tease apart the many subtypes,” says Dr. Liebowitz.
Clear diagnoses can also be deeply meaningful to patients, he adds. “The number one question patients have for rheumatologists is ‘Why did I develop this disease to begin with, and why now?’” Somatic mosaicism may provide answers.
Better diagnoses, in turn, should lead to more targeted therapies. Dr. Liebowitz points to oncology as an example, where treatments tailored to the genetics of a patient’s tumor are proving far more effective than older, more general chemotherapies. “In rheumatology, we have these very broad-based immunosuppressive regimens that very non-specifically suppress the immune system,” says Dr. Liebowitz. “If there’s a genetic disease that really affects one pathway, you can either develop new drugs or use existing drugs and target that specific pathway.”
An Ever-Expanding Mosaic
Mosaicism itself isn’t a new discovery; geneticists have long known of examples such as calico cats, whose patchwork coats come from the random inactivation of X chromosomes. Recently, however, new DNA and RNA sequencing tools are showing that the phenomenon is much more widespread than previously thought.
“We are not universally good at detecting somatic mosaicism, especially when we don’t know to suspect it,” says Dr. Posey, who is also chief of the Division of Clinical Genetics in the Department of Pediatrics. “I think it explains far more than we realize.”
The NIH and other research agencies have started ramping up efforts to plumb the depths of somatic mosaicism, with large consortia now enrolling participants for extensive genetic testing. “They’re doing sequencing in multiple different tissue types so they can really understand: OK, if we compare blood to skin to another tissue, what are the variants that we’re observing, and which variants are specific to a particular tissue?” says Dr. Posey.
Those projects should help to reveal the scope of somatic mosaicism, but scientists are also probing the diversity of its effects. “You could have a very damaging variant that’s present in 1% of cells and that has no impact on health; you could have another, less damaging variant that’s present in 5% of cells, and that could be enough to push against the threshold of disease,” says Dr. Posey. Specific mutations will likely have different critical thresholds for causing disease, which may also vary in different tissues; a mutation that’s harmless in skin might be deadly in cardiac muscle.
Understanding the impacts of somatic mutations will also require more basic genetics research. Most human protein-coding genes still have unclear or unknown roles in health and disease, and even wellcharacterized genes can fail in multiple ways. Dr. Posey adds that because gene products work in concert, some mutations may only matter in the context of variations in other genes: “Most likely, the mosaic variants across the genome interact with one another, collectively shaping the clinical features observed in that individual.”
Though much of that research will take years to bear fruit, Dr. Posey is already working to bring a more nuanced understanding of genetics to patients. As one of the leaders of the NIH-funded GREGoR Consortium (Genomics Research to Elucidate the Genetics of Rare diseases), she hopes to find better ways to diagnose and treat genetic conditions. “What we’re really looking at is how do we go beyond the current standard of care and think about how to get families a better genetic or molecular diagnosis?” she says.
Products of the Past
Somatic mutations also present a major challenge for one of the hottest technologies in regenerative medicine: induced pluripotent stem cells (iPSCs). Over the past two decades, researchers have discovered and refined techniques for “reprogramming” fully differentiated cells, such as fibroblasts, to turn back their developmental clocks. The reprogrammed cells resemble embryonic stem cells and can then be induced to differentiate into other tissues.
In principle, iPSCs derived from a patient’s own cells could be used to generate replacement tissues, or even organs, that could then be transplanted back into the patient without triggering immune rejection. “My group has been interested in whether or not that process would carry along any mutations that those individual cells accumulated on their way to becoming your skin or your blood or even your brain,” says Kristin Baldwin, PhD, professor of genetics and development.
The answer turns out to be “yes,” and Dr. Baldwin’s team has been characterizing the extent and potential impact of those mutations. “There is no avoiding this; they will be there,” she says. “So, present efforts are screening through the mutations that we know are very bad, and I think in the future, people will basically look at the whole genome and have a book of variations that are harmful and things that are neutral.”
Besides screening for dangerous mutations, iPSC developers may also be able to avoid them in the first place, by picking the right somatic cells to reprogram. “If you knew which cells in your body had the fewest mutations, I think that would be a great place to start,” says Dr. Baldwin. She adds that “some people are using umbilical cord blood, which should have a lower number of mutations.” Others have proposed using cells that differentiate early in development and don’t divide, such as neurons. But both cell types are hard to obtain.
Fortunately, Dr. Baldwin’s lab has discovered a surprising source of low-mutation cells that’s more widespread: old blood stem cells. The finding arose when the group was studying the blood of healthy centenarians. “As you might guess, you get more and more mutations as you age, but then we saw at about 80, the number went down again, and we didn’t know why,” says Dr. Baldwin. Her lab then realized that as people age, their blood comes from smaller populations of stem cells, reducing the number of potential mutation sources.
That finding hints at one of the other challenges of diagnosing conditions driven by somatic mosaicism: finding the right cells to sequence. “If you’ve got someone with strong symptoms and the blood doesn’t tell you that the gene is there, you might want to do a follow-up of a broader set of their tissues,” says Dr. Baldwin.
Transcriptional Mosaicism
As strange as the new data on somatic mosaicism may seem, the field is poised to get even stranger. Besides accumulating random mutations, cells may also make independent decisions about which copy of a gene to express. The phenomenon, called autosomal random monoallelic expression, is turning some long-standing assumptions about genetics on their heads.
Every biology student learns that each cell in our body (except sperm and eggs) contains two copies of each gene, one from each parent, and each copy plays an equal part in the cell. That’s why Dusan Bogunovic, PhD, professor of pediatric immunology in the Department of Pediatrics, was so surprised when he analyzed gene expression in cells from a patient with one mutant copy and one wild-type copy of a critical gene.
“We expected to see about 50% of the RNAs from the mutant gene and 50% from the wild-type gene, but all the RNA we found came from the mutant gene,” he says.
To determine whether that was a general phenomenon, he and a team of collaborators profiled gene expression in a set of immune cells from 10 healthy New Yorkers. They found that those cells had inactivated either the maternal copy or the paternal copy of about 5% of their genes. “That means that some cells in your body can be more Mom and less Dad, or vice versa, depending on which copies are inactivated,” says Dr. Bogunovic. “To make things even more complicated, the inactivated copies differ from cell to cell and can perhaps change with time.”
The researchers did a similar analysis in several families with different genetic disorders affecting their immune systems and found that the disease-causing allele was more likely to be active in sick patients and suppressed in healthy relatives with the same gene.
“We don’t see a preference for immune genes or any other class of genes, so we think this phenomenon can explain the wide variability in disease severity we see with many other genetic conditions,” says Dr. Bogunovic. “This could be just the tip of the iceberg.” His lab is now studying the mechanisms behind selective gene inactivation, with an eye toward finding ways to redirect it to suppress expression of the undesirable copy of a gene.
In addition to testing for somatic DNA mutations, physicians may soon be ordering tests for RNA expression patterns to characterize the full extent of a patient’s illness—paving the way for whole new approaches to naming and treating disease.
“I think that more and more, we’re going to move from clinical diagnoses to molecular diagnoses,” Dr. Liebowitz says. “And then I think there will be a huge shift to developing new medications or applying existing medications in a much more targeted way.”