Advanced 3D imaging reveals hidden nerve networks inside joints

Temporomandibular disorders (TMDs) are a group of more than 30 conditions that cause pain and dysfunction in the jaw. So what could a small tissue sample from a rat's knee have to do with treating them? 

A new publication from University of Pittsburgh researchers offers some answers. In "Advanced Tissue Clearing and Three-Dimensional Imaging Approaches to Visualize Neural Innervation in the Rat Knee Joints" (doi.org/10.1038/s44303-026-00167-6), Alejandro Almarza, professor of oral and craniofacial sciences in the School of Dental Medicine with a secondary appointment in the Swanson School of Engineering's Department of Bioengineering, used specialized imaging techniques to map the architecture of nerves inside knee joint tissue. For Almarza, this research lays critical groundwork for visualizing how nerve patterns in densely innervated joints are related to pain, allowing him to better understand disorders of the temporomandibular joint (TMJ).

Most of us go through life without much pain in the face that isn't related to a tooth, but the TMD umbrella is very broad, and the cause behind that pain is relatively unknown. For the vast majority of TMDs, we're dealing with muscle-based or joint-related problems, and this work could help us understand why these occur." 

Alejandro Almarza, professor of oral and craniofacial sciences, School of Dental Medicine, University of Pittsburgh

A joint effort 

TMJs on both sides of the face connect the jawbone to the skull and act like a sliding hinge, allowing us to talk, chew and yawn. The relationship between nerve density and pain in joints like the TMJ is relatively unknown, and the traditional method for studying these joint nerves involves slicing tissue into thin slivers and staining them with dyes to make nerve cells visible under a microscope.

Cutting tissue apart, however, destroys its three-dimensional structure, making it impossible to see how the nerves branch throughout a joint. To get a clear picture of these nerve structures in 3D, Almarza partnered with two professors from Pitt's Center for Biologic Imaging (CBI) to use both light sheet fluorescence microscopy and an imaging technique known as tissue clearing. 

"Tissue clearing makes an entire piece of tissue transparent for 3D imaging so you can visualize the nerves inside, and the microscope we used works like a wall of light sweeping through the volume of tissue all at once, making it faster than a traditional microscope while still achieving near-confocal resolution with minimal tissue damage," Almarza said. "Some of the best of these systems in the world are custom-built here at Pitt by Simon Watkins, and the clearing methods have been developed by Alan Watson."

Watkins, distinguished professor of cell biology and immunology, founded the CBI in 1991. Unlike a typical fee-based core facility, CBI faculty collaborate directly with researchers to design specialized microscopes and imaging techniques from the ground up. While Watkins is the expert in building the scopes themselves, his colleague Alan Watson, associate professor of cell biology, provides the other half of the equation: the computing infrastructure, tissue clearing protocols, and programming expertise to store and analyze the enormous volumes of data these systems produce. Because no current commercial solution exists for imaging nerves inside of large, dense tissue, the team built one.

"Clearing joint tissue isn't entirely new, but it presents some really interesting challenges. Alejandro came to us with a problem that was hard to deal with, one we'd also struggled with for years, and as a group we were able to work together and find a solution," Watson said. "And these high-speed imaging techniques generate enormous amounts of data, so we've developed high-performance computing systems to store, process and visualize it all." 

Clearing the way for understanding pain

The team ultimately compared two tissue clearing methods: PEGASOS, a previously established protocol for bone-containing tissue, and c-Clear, developed in-house at the CBI. PEGASOS left behind autofluorescence protein that both blocked the microscope's laser from fully penetrating the tissue and caused high background, but c-Clear introduced a 24-hour photobleaching step that inactivated those molecules before staining, allowing fluorescent antibodies to bind to neurofilament and produce a complete three-dimensional map of the joint's nerves. 

"The c-Clear method takes about six to eight weeks to obtain an image, making it far more labor and time-intensive than normal histological methods, but the result is an extremely powerful and clear representation of how these nerves branch," Almarza said.

C-Clear does come with one significant caveat: the sheer size of the data it generates. A single three-dimensional nerve map of the knee contains about one terabyte of information, and the full collection from the project runs about 16 terabytes. Luckily, supporting that feat is the CBI's computing infrastructure: seven petabytes of storage and an H200 GPU cluster used to stitch, clean and analyze every dataset, making it possible to deposit the full collection publicly for anyone to access and download on the National Institute of Health's SPARC Portal. 

"I believe we're the first to publish this new type of imaging dataset on the portal," Almarza said. "The photos and videos are amazing, and our next challenge is quantification and figuring out the computational pipelines to really analyze what we're seeing."

Ultimately, looking at a rat's knee may seem far removed from the joint that helps humans chew and talk, but the connection is deliberate. Through the NIH HEAL Initiative, Almarza is part of the ReJoin Consortium, a $50 million project aimed at mapping nerve architecture across joints, species and disease states to expand understanding of pain signaling. With c-Clear now validated on some of the most challenging tissue the consortium has yet encountered, Almarza can turn his attention to the structure he set out to study all along. 

"There are a lot of people whose radiographs look like they should have pain in their TMJ, but they're actually talking just fine," Almarza said. "Is it because of the type of nerves in there? And why is it different from people with pain? That's the type of question this research is hoping to answer."