Luke Trimmer-Smith, Ph.D., began his doctoral studies in 2012 at the University of Florida College of Liberal Arts and Sciences’ Department of Biology, specializing in virology and cross-species transmission. Then-Emerging Pathogens Institute faculty member and biology professor Derek Cummings, Ph.D., hired Trimmer-Smith as a postdoctoral associate. At the time, Cummings was trying to start a wet lab at the EPI. Trimmer-Smith, now an assistant scientist with the EPI, built the lab into what it is today, actively running assays such as plaque-reduction neutralization tests. That foundation set the stage for a research study that won him first place in the EPI Research Day 2026 early-stage investigator poster competition.
Trimmer-Smith’s research centers on dengue, a mosquito-borne illness caused by four related but distinct serotypes of the same virus. The human immune system can recognize and respond to each type, but immunity to a single serotype doesn’t necessarily protect against the others. In fact, a second dengue infection can actually be more severe than the first. Developing effective dengue prevention requires understanding not just how the viruses look today, but also how they’ve evolved over decades. Additionally, it is crucial to know how well our immune systems recognize both old and new strains.
To answer those questions, Trimmer-Smith and his colleagues turned to antigenic cartography: a technique that maps viruses in space based on how the immune system “sees” them. Viruses that the immune system treats as similar cluster together, while those that look very different are located farther apart.
“The more distantly related viruses are, the greater the distance between them on the map,” Trimmer-Smith explained.
Previous research had only mapped dengue back to the 1990s. Trimmer-Smith and his team extended that timeline to the mid-1970s by using serum samples from African green monkeys. Individual monkeys were infected with a single dengue strain. Their serum was collected to test how well antibodies against a single virus could neutralize others across four decades of viral history.
What they found was striking. Dengue viruses from the 1970s and 1980s sat far out on the edges of the map and had neutralization profiles quite different from one another and from those of more recent strains. Viruses from the 1990s onward moved closer to the center, becoming more antigenically similar. But the movement wasn’t a straight-line march inward. Instead, new strains appeared first at the outer edges, moved toward the center, then spread back outward, a push-and-pull dynamic shaped by the immune landscape of their hosts.
Trimmer-Smith produced the raw data himself, counted plaques, and generated neutralization titers. Titers are laboratory measurements of the number of antibodies your body has produced to fight a particular disease. The higher the titer, the more effectively a serum inactivated the virus. Together, thousands of those data points became the coordinates on the antigenic map.
“This study is a stepping stone,” he said. “People designing a vaccine can look at this map and choose viruses that give the best chance of broad neutralization.”
Examining the relationships among different strains of a virus over time can inform disease prevention and mitigation strategies and shed light on how viruses evolve. If the virus changes over time and becomes more similar to the older virus again, immunity may not be as protective. A vaccine that is effective for current viruses may not be as effective with a strain that’s been out of human circulation for several decades or future versions of the virus.
“The flu is a good example, although it’s not the same as dengue. We try to predict flu outbreaks based on what we’ve seen in the past. By knowing how they evolve over time, we can take predictive measures.”
The team plans to expand the work to other arboviruses, such as Zika, which caused outbreaks across Latin America in 2015 and 2016. Someone who has had dengue may have some neutralizing antibodies against Zika, and vice versa. Related viruses, such as yellow fever, could also be incorporated to help explain their relationship to dengue and Zika.
For Trimmer-Smith, the open questions are what make the work exciting. “With every new discovery, more questions arise,” he said. “By understanding how these viruses change over time, there are endless possibilities of where you can go with the research.”
Antigenic relationships of dengue viruses from 1975-2016 reveal early convergence and subsequent divergence in antigenic space
Collaborators
- Luke Trimmer-Smith – Emerging Pathogens Institute, University of Florida
- Angkana T. Huang – NUS Saw Swee Hock School of Public Health
- Leah Katzelnick – US NIH
- Bernardo Garcia Carreras – Consultant
- Kellie Goodnight – Emerging Pathogens Institute, University of Florida
- Matt Hitchings – Department of Public Health, Center for Statistics and Quantitative Infectious Diseases (CSQUID), Emerging Pathogens Institute, College of Public Health and Health Professions, University of Florida
- Derek Cummings – Department of Epidemiology, Bloomberg School of Public Health, Department of Biomedical Engineering, Whiting School of Engineering, Johns Hopkins University
Abstract
Understanding long-term antigenic evolution of dengue viruses (DENV) is critical for interpreting population immunity and guiding vaccine strategies. Prior work using a standardized panel of sera demonstrated that DENV isolates from 1994–2016 initially converged toward the center of antigenic space before diverging, consistent with decreasing cross-reactivity among circulating viruses. Here, we extend this framework by characterizing antigenic relationships of earlier DENV isolates from 1975–1990.
We measured neutralization titers for 74 viruses (9 DENV1, 35 DENV2, 27 DENV3, 3 DENV4) using a panel of sera derived from 20 African green monkeys, each infected with a single DENV strain (5 per serotype). This same serum panel was previously used to characterize 348 viruses from 1994–2016, enabling direct integration into a unified antigenic map spanning four decades. Antigenic cartography was used to position all viruses within a common space and to quantify patterns of cross-reactivity and temporal movement.
Viruses from 1975–1990 occupy positions further from the center of the antigenic map relative to later isolates, indicating reduced cross-reactivity both within and between serotypes during this earlier period. Analyses of temporal trajectories suggest that antigenic evolution is not solely characterized by gradual movement toward the center of antigenic space; rather, viruses appear to emerge in more antigenically central regions before subsequent convergence and later divergence.
Together, these findings extend the temporal scope of dengue antigenic evolution and suggest a dynamic, non-monotonic process in which antigenic positioning reflects both emergence in peripheral space and later population-level pressures shaping convergence and divergence. These results have implications for understanding long-term immune landscapes and the design and evaluation of dengue vaccines.