Like a lightbulb illuminating the moment you flip a switch, the brain pays an immediate energy cost every time a neuron fires. Bistra Iordanova has built her career studying brain function, but over time, she kept returning to a question her field hadn't fully investigated: how does this “cost” of the brain's metabolism impact how we age?
“I’ve collected lots of data about blood flow and the brain’s neuronal activity,” Iordanova said. “I eventually included data on glucose, lactate, creatine, and other brain metabolites in relation to aging, and then one morning, I found myself with so much information that I really had no clue what was going on. Simple linear models no longer worked, and dimensionality reduction approaches were not as useful as I hoped.”
For Iordanova , assistant professor of bioengineering at the University of Pittsburgh’s Swanson School of Engineering, that influx of data set the stage for an interdisciplinary collaboration with Liang Zhan , associate professor of electrical and computer engineering, to build integrative neuro-metabolic models capable of making predictions about brain health. Now, the duo is co-investigating a five-year, $3.3M R01 NIH project , “Multiscale Models of Age-Specific Neurometabolic Coupling," to create a whole-brain theory on how the brain's metabolic processes affect cognition in aging.
Looking beyond blood flow
While most research has typically focused on amyloid plaques and hemodynamics as early warning signs of Alzheimer's Disease (AD), Iordanova and Zhan are focused on the metabolic changes that happen in the brain’s networks by looking at the impact of specific metabolites like glucose, lactate, and creatine on brain activity.
“The brain requires large amounts of glucose and oxygen to function as billions of interconnected cells work together,” Iordanova said. “But old age brings decline in metabolic efficiency, and our brain cells have to maintain networks by adapting their metabolic processing.”
When that metabolic adaptation fails, it can lead to cognitive decline and dementia, and some individuals may be especially vulnerable due to genetics, lifestyle, or other factors. Ultimately, the goal of this work is to support the development of metabolic screening and therapies for at-risk individuals years before energy metabolism begins to affect cognition. First, however, the team must use advanced modeling strategies to make sense of the vast amount of metabolic and neural brain data they will collect.
A multi-scale, multi-species model
This approach will span three distinct levels of biological scale. At the smallest level, they will use two-photon microscopy to quantify the relationship between red blood cell velocity, neural activity, and lactate transients in late-onset AD mouse models. Next, wide-field imaging will capture how mitochondrial activity ripples across cortical networks in mice. At the largest scale, the team will explore the impact of metabolism on functional whole-brain connectivity by integrating data from brain MRI in both animal models and human cohorts.
“We want to connect what we know in mice on a cellular level to what we know from non-invasive imaging in humans,” Iordanova said. “And to build a comprehensive brain network theory, we need a lot of robust information at different scales.”
Once the data is collected at each step, Zhan will use his expertise in brain network modeling and graph theory to build the computational architecture that ties these complicated layers of brain data together.
“The structures of mouse and human brains differ, and of course, we know that the brain is highly complex,” Zhan said. “Because many brain regions function together, diseases such as Alzheimer’s affect not just a single region but the entire brain network, which we aim to characterize and understand.”
Toward better treatments through interdisciplinary collaboration
The ultimate goal of the project is to bridge the long-standing gap between discoveries in the laboratory and treatments that benefit patients. Although researchers have plenty of AD data from mouse models, translating those insights to humans remains a challenge. By studying how metabolic factors interact with genetics, sex-differences and aging, the team hopes to identify biological signals that could eventually guide more personalized approaches to preventing cognitive decline.
“I’m obsessed with figuring out how to translate what we learn in mouse models of Alzheimer’s into something meaningful for humans, because translation is not trivial,” Iordanova said. “A frequent quip on the translation from mouse to human is that scientists have cured Alzheimer’s in mice many times, but patients and families are still living through the disease every day. If we can improve on the cross-species approach to identify common metabolic vulnerabilities combined with certain genetic risk factors, then perhaps we could tailor interventions at the right time to the people who would benefit most.”
Although they now collaborate closely, Iordanova and Zhan were not previously familiar with each other’s work or the potential of their combined expertise until a chance meeting at a radiology event just a few years ago. While the research itself is promising, the team also hopes their partnership will encourage more engineers and scientists from different fields to pursue collaborative projects together.
“Engineers and scientists from different fields often speak very different technical languages, and it’s rare to see those perspectives truly come together,” Iordanova said. “There’s real value when you are actually in the room with someone from another field and working to bridge that gap, and I think that’s what’s really valuable about this collaboration.”
The interdisciplinary team also includes co-investigators Alberto Vazquez , Tao Jin and Alex Poplawsky from the School of Medicine and Nicholas Fitz and Rebecca Deek from the School of Public Health at University of Pittsburgh. This project is supported by the National Institute on Aging (R01AG092661) for the period January 2026 through December 2030.