So much of science and medicine comes down to taking pictures. Biology, especially, has long been considered an observational enterprise.
Biology is an attempt to establish coherence across billions of images. Hence, the types of questions we ask are limited to the methods we have for observation, while the theories we construct are constrained by what we see.
There are manifold ways to take a picture of a living thing. Among these, two broad and separable categories are structural imaging and functional imaging.
Structural imaging is more widely used, and is impressive in scope. One familiar modality is x-ray imaging, whereby we can bombard a thing with x-rays and watch how they diffract. In the clinic, this can detect a stress fracture in a runner’s foot. In the lab, this same strategy revealed the double-helix structure of DNA and is used daily to tease out the intricate, nanometer-sized folds of proteins.
However, functional imaging is coming of age as well. In functional imaging, what you see is not a still-life, but a readout of a dynamic process occurring in a living thing. Biological molecules are not like concrete building blocks, but like dancers, and their choreography is constantly vibrating, twisting, interacting, disentangling, and cutting.
Two of the most common types of functional imaging, at least in the medical arena, are positron emission tomography (PET) scans and functional magnetic resonance imaging (fMRIs). Both track dynamic bodily processes, tying functionality to location.
In PET scans, a fluorescent derivative of glucose is injected and allows the imaging of glucose uptake. Brains and livers take up a lot of glucose, and show up bright on a PET scan, but tumors also take up a disproportionate share and are visible by PET.
fMRI visualizes blood flow, and is commonly used to show which areas of the brain are triggered during certain activities or when presented with certain stimuli. This works since blood flow can be robustly correlated with neuronal activation.
Beyond the clinically ubiquitous examples, biologists and doctors are constantly looking for new ways to look at life. New methods are especially relevant for functional imaging, as the imaging strategy is specific to the function being detected.
While generally each function demands its own novel methodology, the Craik Lab at UCSF recently published a strategy for imaging thrombin activity that may be modular enough to apply to a variety of biological functions. The research, led by former junior faculty member Dr. Michael Page, was published in the October 1st issue of Nature Communications.
Thrombin is a protein essential to the formation of blood clots. Sometimes these blood clots are desirable — as in the case of wound healing — but blood clots in the wrong places can cause complications and often these can be lethal.
For instance, pulmonary embolism is the blockage of the lung’s main artery by a blood clot. Being able to image thrombin activity would help doctors detect and grade a raft of complications, including stroke, myocardial infarction, and pulmonary embolism.
To image thrombin, Page and colleagues looked to its function. Thrombin is in a family of proteins called proteases, enzymes that cut other proteins. Specifically, thrombin cuts a protein called fibrinogen, and the resulting pieces are sticky enough to self-assemble into the hydrogel-like structure that constitutes a blood clot (think about the consistency of jello).
So, Page and company created a synthetic peptide (a shorter piece of a longer protein) that matched the thrombin recognition site. Thrombin only recognizes certain proteins for cutting because it recognizes a specific peptide sequence and always cuts where it sees that sequence. The research team embedded this sequence as part of a larger synthetic peptide so that thrombin would cut at the recognition sequence and release a fragment that sticks into cell membranes.
Furthermore, this fragment was coupled with a variety of fluorescent dyes or radioisotopes. This provided the means to actually see the probe as it gets deposited near sites of thrombin activity.
Using mice, the group demonstrated robust localization of their probe to pulmonary embolisms generated by injecting thromboplastin into either the inferior vena cava or the tail vein. Non-specific binding is one of the biggest obstacles to creating successful probes for protease activity, but the imaging was both bright and specific.
Finally, the approach appears adaptable to imaging many other proteases. The probe contains three parts: the fluorescent dye, the membrane-binding portion, and the thrombin-specific sequence. Only this last part would need to be changed to make this strategy useful for seeing the activity of other proteases. Proteases are critical enzymes and have differential activity in a variety of diseases, including atherosclerosis and cancer.
These thrombin activity probes will be useful for both detecting and diagnosing clotting pathologies, as well as studying the mechanisms of blood clot formation in a model organism like a mouse. Functional imaging in its breadth will continue to shape which biological activities we can see, and thus how we describe the mechanisms of life.
