Ankit Basak (Chemistry/IMES)
When Ankit Basak arrived at MIT for graduate school, he came with a broad, interdisciplinary background and a keen interest in developing new tools for biological research. Raised in Kolkata, India, Basak pursued an integrated BS-MS program at the Indian Institute of Science Education and Research (IISER) Kolkata, where he studied chemistry and biology. From the start, he was drawn to technology development, and worked on wide-ranging projects, including synthesizing perovskite nanowires for optoelectronic applications, developing organic fluorescent sensors for detecting iron ions in water, and designing algorithms to accelerate MRI data acquisition.
That appetite for methodological innovation has carried through to his doctoral work at MIT, where Basak is a fifth-year PhD student, co-mentored by Professors Laura Kiessling (Chemistry) and Alex Shalek (IMES). His research now tackles a central question in microbiome science: how do human cells recognize different microbes and distinguish whether they are friend or foe?
Much of that recognition is mediated by glycans — complex sugar molecules decorating cell surfaces — and lectins, the glycan-binding proteins on host cells. However, the ability to study these interactions in their native environment — the mucosal barrier tissues that line our digestive, respiratory, and other tracts — has historically been out of reach. Traditional glycan profiling methods rely on ex vivo assays using cultured cells or small pieces of tissue, and measure bulk samples. They lack the single-cell, spatial detail needed to understand how recognition happens in the mucosal tissue environment.
Basak has addressed this gap by co-developing a platform called GOAT-seq (Glycan Outlining And Transcriptome sequencing). The method uses DNA-barcoded human lectins as molecular probes. When applied to single-cell or spatial transcriptomic workflows, the barcodes allow researchers to simultaneously capture glycan-binding profiles and gene expression readouts. “This means we can finally ask, in a tissue context, which host and microbial cells are binding to which lectins, and which transcriptional programs accompany that binding,” Basak explains.
Understanding glycan/lectin interactions between bacteria and mucosal tissues could shed light on why our bodies recognize some bacterial pathogens but others evade detection.
He has already have validated these methods for profiling host cell glycans and transcriptomes. The next step is to integrate the platform for microbial single cell and spatial transcriptomics. The biggest challenge, he says, “is that bacteria have so little mRNA. It’s very hard to get a robust readout of genetic information.” His strategy is to optimize workflows for expressing recombinant human lectins, which bind more effectively to microbial surfaces than the plant lectins typically used in research. By barcoding these human lectins and integrating them into single-cell and spatial sequencing platforms, he aims to capture not just the genetic programs of microbes but also their glycan signatures and their recognition by host immune proteins.
While the platform could be used in a wide range of disease areas, Basak is currently focusing on periodontitis, the chronic inflammatory gum disease that affects nearly one in five people worldwide. “Periodontitis is fascinating because it’s really about dysbiosis,” he says. “There’s a complex interplay between the microbial side and the host side that together drive the disease.” The implications extend well beyond oral health, he notes. Periodontitis is now linked to diseases including gastrointestinal cancers, type I diabetes, cardiovascular disease, rheumatoid arthritis, and even Alzheimer’s disease.
“Fusobacterium nucleatum is an oral bacterium but we now know it can cause colorectal cancer,” he explains. “And a lot of Alzheimer’s patients start with periodontitis. People have found bacterial proteins in their brain tissue — proteins that aren’t supposed to be there. It’s likely that when there’s inflammation, a leaky mucosal surface in the mouth allows bacteria to enter the bloodstream and migrate to these other sites.”
For Basak, these connections highlight the need for technologies that allow scientists to probe host/microbe interactions at high resolution, in their native tissue environment. Ultimately, he hopes to pursue an academic path that continues to blend technology development with biological discovery. For now, though, his focus is on unraveling the conversations between human cells and microbes in the mouth. “My goal is to use this new method to understand the microbial and host factors that mediate glycan/lectin interactions, so we can find therapeutic targets against this devastating disease.”
Helena Hu (Biological Engineering)
As a child in Inner Mongolia, Helena Hu spent most of her days immersed in nature. Living with her grandparents in a nomadic herding community with few other children, she was given free rein to explore the grasslands on her own. “I would wake up so excited because every day was so different,” she recalls. “Once I went out and found a little puppy and brought him back. But he was actually a wolf, and the mom came looking for him in our tent.” She laughs at the memory. “Every day was a new adventure.”
For Hu, those early days of exploration shaped her scientific curiosity. “I didn’t have the language for it then, but I was drawn to the idea that life is defined by dependence,” she says. “As herders, we moved with our animals over seasonal cycles. I saw how the grasslands recovered when we left them to rest, and how the rivers shaped life around them. That early awareness really influenced how I see biology today – as a study of systems, relationships, and the structures that hold them together.”
Today, Hu is a fourth-year PhD student in Biological Engineering at MIT, working jointly with Professors Ed Boyden and Bob Langer. Her focus is on systems and relationships at a much smaller scale than those of her childhood: the molecular interactions that shape the behavior of individual cells, which she sees as critical to understand larger-scale phenomena. “Every population-level response originates from the actions of individual cells,” she explains. “Microbes influence essential processes in the human body, but there are so many aspects of microbial function that we don’t understand because we can’t observe them directly, especially in intact biological contexts.”
“Really understanding the cell — the fundamental unit of life — I think that’s incredibly exciting,” Hu says — and critical to understand population-level phenomena.
Hu is developing a new version of expansion microscopy, a technique pioneered in Boyden’s lab a decade ago that physically magnifies biological samples by embedding them in a hydrogel that swells evenly with the addition of water. As the gel expands, it pulls biomolecules apart but maintains their relative positions, allowing scientists to resolve details that would otherwise be blurred together. Until now, expansion microscopy has offered up to about 20-fold magnification. Hu has pushed the limits of the method, developing an approach that can achieve up to 1000-fold expansion. Achieving this scale was critical, she notes, because it makes it possible to visualize the cell’s biomolecules and their interactions – the chemistry that drives key processes like cellular function, signaling, and drug binding.
“The first challenge was developing a gel that could give us this extreme stretch while maintaining structural integrity,” she explains. “Previous gels had problems with low mechanical robustness at high expansion.” Her method works sequentially: the expanded gel is re-embedded in a second gel, which is then expanded, and so on until the desired magnification is achieved. The result is the first expansion microscopy platform that can, in principle, map an entire proteome — and all the other biomolecules — in a single cell, as well as the interfaces where they interact.
Hu’s current challenge is building a robust imaging pipeline. “With a cellular sample, when it hasn’t been expanded, there are so many biomolecules in one area, the signal from the fluorophores is super bright. Once you expand the atomic units apart, the fluorophores are spread out, and it becomes very difficult to detect them because each one produces only a very weak signal,” she says. To address this, she is designing probes with highly amplified signals that can selectively bind their targets under the specific conditions required for high expansion. These probes will enable her to spatially barcode proteins (and ultimately other biomolecules) which can then be identified computationally.
Importantly, her approach will also allow researchers to capture variation rather than averages. With traditional single-particle electron microscopy, she notes, “you have to purify your proteins to get a 3D structure — so you’re losing biological context — and then you’re getting that structure by averaging millions of particles. You’re getting an average structure.” She gestures toward people walking past. “It’s like studying an ‘average human.’ There’s only so much you can learn.” More advanced methods like cryo-electron tomography (cryo-ET) and correlative light and electron microscopy preserve biological context and individual protein structures but come with significant technical challenges and require expensive equipment that is out of reach for most researchers. Ultimately, Hu says, her method should enable “Angstrom-level imaging of cell structures, in their spatial context, on a standard confocal microscope. No one’s ever been able to do that before.”
The potential applications are wide-ranging. Hu envisions her technology providing researchers across fields fundamental new insights on cell state, function, and interactions — as well as shedding light on potential drug targets by uncovering the molecular drivers of disease. “I think really understanding the cell, the fundamental unit of life — seeing it as it actually is, not as we’re imagining it – I think that’s incredibly exciting,” she says. “Without that, our knowledge is incomplete.”
