A glance at a star-nosed mole (Condylura cristata) is enough to convince most people that something very strange has evolved in the bogs and wetlands of North America. There’s nothing else on the planet quite like this little palm-sized mammal. Its nose is ringed by 22 fleshy appendages, called rays, which are engorged with blood and in a constant flurry of motion when the animal searches for food. What is this star? How did it evolve and what is it for? What advantage would be worth sporting such an ungainly structure? To a neuroscientist interested in sensory systems, this kind of biological anomaly represents an irresistible mystery. I first began studying star-nosed moles in the early ’90s in an attempt to answer some of these basic questions. But I soon discovered that this unusual animal, like many other specialized species, could reveal general principles about how brains process and represent sensory information. In fact, star-nosed moles have been a gold mine for discoveries about brains and behavior in general—and an unending source of surprises. The most obvious place to start the investigation was with that bizarre star.
Exquisitely sensitive touch
The star-nosed mole’s “nose” is not an olfactory organ, but a skin surface that mediates touch. Innervated by more than 100,000 sensory neurons, the star is probably the most sensitive and highly acute touch organ found on any mammal. Under a scanning electron microscope, the skin surface resolves into a cobbled landscape covered with tens of thousands of tiny epidermal domes. Each is about 60 µm in diameter, and each contains a circular disk in its center. Known as Eimer’s organs, these sensory protrusions cover the entire surface of the star’s 22 appendages. In total, a single star contains about 25,000 domed Eimer’s organs, each one served by four or so myelinated nerve fibers and probably about as many unmyelinated fibers.[1. K.C. Catania, J.H. Kaas, “Areal and callosal connections in the somatosensory cortex of the star-nosed mole,” Somatosens Mot Res, 18:303-11, 2001.] This adds up to many times more than the total number of touch fibers (17,000) found in the human hand—yet the entire star is smaller than a human fingertip.
In 2006, to investigate how Eimer’s organs function, a student of mine, Paul Marasco, recorded signals from nerve fibers that supply the star, and found that the organs respond vigorously to compression and vibration.[2. P.D. Marasco, K.C. Catania, “Response properties of primary afferents supplying Eimer’s organ,” J Exp Biol, 210:765-80, 2007.] But several features were remarkable. For example, the skin areas monitored by the nerve fibers (i.e., their receptive fields on the star) were so small we had to use a microscope and thin, hairlike probes only a few hundred microns in diameter (von Frey hairs) to delineate them. This means the star provides exceptional resolution, much like a high-density photoreceptive chip in a camera that produces a hi-res image with many pixels. In addition, fibers were responsive to tiny calibrated forces applied to the skin, indicating the organs are sensitive to the slightest pressure. Finally, many of these sensory, or afferent, neurons were directionally sensitive, meaning they responded to only one direction of sweeping movement across the star. This suggests that when the star is pressed against something, the Eimer’s organs might act together to detect the deflections caused by microscopic textures and surface features. Many of the mole’s favorite prey items, such as aquatic insect larvae and other small invertebrates, have distinctive ridges and spines—just the kinds of features that would differentially deflect the Eimer’s organs upon contact.
In addition to exploring the anatomy and electrophysiology of Eimer’s organs, an early part of the investigation consisted of a detailed survey of the star. This was done to determine whether it might host any other kinds of sensory organs. I found no other structures, but did notice that the Eimer’s organs for each ray were segregated from those in the other rays by a thin strip of epidermis at the ray’s base; each ray was a separate sensory unit. This seemingly inconsequential observation intrigued me, because it reminded me of a landmark finding in rodent brains related to touch. In 1970, Thomas Woolsey and Hendrik Van der Loos (working at Johns Hopkins University School of Medicine in Baltimore) found that each separate whisker on the face of a mouse projects to a separate anatomically visible unit in the neocortex.[3. T.A. Woolsey, H. Van der Loos, “The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex. The description of a cortical field composed of discrete cytoarchitectonic units,” Brain Research, 17:205-42, 1970.] Could such modules exist in the star-nosed mole brain, and correspond to the different rays of the star?
Stars and stripes
To investigate brain organization in star-nosed moles, I teamed up with Jon Kaas here at Vanderbilt University to map the touch areas of the mole’s neocortex. Using electrodes, we identified a large neocortical area that responded to signals from the star, seemingly divided into three distinct, but interconnected, regions. When we flattened and sectioned the neocortex to look for anatomical correlates that might relate to the different nasal rays, we were not disappointed. Each of the three distinct star maps contained two halves—one in each hemisphere of the brain, and each mapping to the 11 star rays on the opposite side of the nose. When properly stained, the neocortex lit up in a beautiful and complex series of pinwheels, one set of 11 stripes in each half of the three different maps.[4. K.C. Catania, “Cortical organization in moles: Evidence of new areas and a specialized S2," Somatosens Mot Res, 17:335-47, 2000.] No other species with three anatomically visible maps of a sensory surface has been described. This is a testament to the complexity of the neural networks that process touch in star-nosed moles. It also provides important advantages for studying the brain by allowing for precise measurements, neuronal recordings, and identification of connections between readily identified areas.
Subsequent studies have taken advantage of these features to reveal, for example, that the three half-star maps within a single hemisphere are precisely interconnected at topographically identical points. In other words, neurons representing the same ray in each of the maps are interconnected. The repeated maps are each representing the same area of touch “space,” but are likely processing different components of touch for that region.1
The three half-star maps of the left hemisphere are connected with those of the right hemisphere along the midline of the star representations (at the base of the map between ray representations 1 and 11) in a collection of nerve fibers called the corpus callosum. Connections along the midline of the maps likely explain how the moles maintain a fused perception of the world despite the mapping of only half of the body in each cerebral hemisphere—an anatomical oddity seen across the animal kingdom, including in humans. Indeed, half a century ago University of Edinburgh researchers made a similar observation for visual cortical representations in mammals, showing they were also heavily joined along the sensory representation of the midline of vision (the vertical meridian).[5. B.P. Choudhury et al., “The function of the callosal connections of the visual cortex,” Q J Exp Physiol, 50:214-19, 1965.] This idea has since been termed the “midline fusion hypothesis.”
We have also been exploring response properties of individual neurons in each module of the three maps to determine any differences in how the maps process touch. But one feature of the brain maps in particular drew our attention and raised the next, most obvious question. In each hemisphere, ray number 11, one of the smallest rays of the star, has by far the largest representation in the brain, taking up roughly 25 percent of the primary somatosensory map. Why would this be the case?
A somatosensory fovea
To investigate this question, we began videotaping star-nosed moles while they foraged and searched for food. Initial observations were made using a regular video camera at 30 frames per second, but this provided poor resolution of the fast-moving mole. So we switched to a high-speed camera and recorded moles at 500 frames per second. A consistent pattern emerged in every recording that we analyzed: the moles were using the 11th rays for all of their detailed investigations. Whenever the mole touched something of interest, particularly potential food, with any of the other rays (1–10), it made a sudden movement to position the smallest rays over the object for repeated rapid touches.[6. K.C. Catania, F.E. Remple, “Asymptotic prey profitability drives star-nosed moles to the foraging speed limit,” Nature, 433:519-22, 2005.] The similarities with vision were striking. The star movements resembled saccadic eye movements—quick movements of the eyes from one focus point to another—in their speed and time-course. The two 11th rays are over-represented in primary somatosensory cortex relative to their size, just as the small visual fovea in primates—a small region in the center of the eye that yields the sharpest vision—is over-represented in primary visual cortex. Interestingly, some bats also have an auditory fovea for processing important echolocation frequencies. It appears that evolution has repeatedly come to the same solution for constructing a high-acuity sensory system: subdivide the sensory surface into a large, lower-resolution periphery for scanning a wide range of stimuli, and a small, high-resolution area that can be focused on objects of importance.
Another question was whether the large representation of the tactile fovea in moles reflects a larger number of nerve fibers from rays 11 compared to the other fibers. This question was comparatively easy to address by simply counting fibers in moles (if counting 200,000 fibers can be called easy). It turned out that the 11th rays have more fibers per Eimer’s organ compared to other rays. But because these rays are smaller in size and thus have fewer sensory organs, the total number of fibers does not account for the larger representation in each hemisphere of the brain. The left and right 11th rays host 10 percent of the fibers in each side of the star, but take up 25 percent of the star representation in each hemisphere. The results therefore suggest two tiers of specialization: more nerve fibers in each sensory organ on the skin surface, and greater cortical territory devoted to each nerve fiber in the brain maps. A similar relationship has been reported for the primate visual system,[7. P. Azzopardi, A. Cowey, “Preferential representation of the fovea in the primary visual cortex,” Nature, 361:719-21, 1993.] underscoring how evolution has repeatedly created high-resolution sensory systems in similar ways.
But the results did not answer the question of how the behaviorally important inputs from the 11th rays may be allocated more brain territory during development. We are continuing to investigate this question by examining the early development of sensory representations in star-nosed moles. But in the course of our video recordings of ray-11 foveation movements, we realized something else that was interesting about star-nosed moles—they are extremely fast. Star-nosed moles can identify and eat a small prey item in as little as 120 msec, with an average time of 230 msec.6 This finding literally placed them in the Guinness Book of World Records as the fastest foragers among mammals.
Star-nosed mole diets
The key to making sense of the star-nosed mole’s speed turned out to be the habitat where we collect the animals for our studies. I had to look at the problem through the lens of a behavioral ecologist. In particular, optimal foraging theory provides a framework for considering foraging speed (or more specifically, prey-handling time) in relationship to the energy content of food. In essence, predators are assumed to try to maximize their rate of energy intake by reducing the time they spend searching for and handling prey. Individual prey items are ranked according to their profitability, with those higher in energy content worth more of the predator’s time and energy. In this model, prey would only be added to the diet if, based on their profitability, they increase the animal’s average energy intake during foraging.
That’s where things get interesting for star-nosed moles. With incredibly short handling times for eating very small prey, the mole can profitably consume a range of foods, including small aquatic insect larvae and annelids, that would not be worth the time or effort of slower animals. Could this be the main purpose of the star? Star-nosed moles live in wetlands, where they compete with other insectivores, especially shrews, for food, so having a prey category to themselves would be especially useful. More direct evidence can be found in the work of William John Hamilton, a biologist who worked at Cornell in the 1930s. He examined the gut contents of many star-nosed moles and showed that they do indeed include many small invertebrates in their diet.
But the real smoking gun for the mole’s feeding strategy can be found literally right under its nose. Just behind the 11th ray of the star, the mole has a unique set of modified front teeth that form the dental equivalent of a pair of fine tweezers. High-speed video shows these specialized teeth are used to efficiently pluck tiny prey items from the ground. It is also clear from the behavior that the teeth and the star act as an integrated unit—the 11th rays, located directly in front on the teeth, spread apart as the teeth move forward to grasp small food. Thus, tweezer-like teeth and the exquisitely sensitive star likely evolved together as a means to better find and handle small prey quickly. Of course, star-nosed moles also eat larger profitable prey whenever it is available (we feed them large night crawlers in captivity), and they have larger rear teeth for this purpose. Nevertheless, it appears that the ability to rapidly detect and consume small prey was the major selective advantage that drove the evolution of the star.
We continue to investigate these remarkable animals and to uncover new and unusual abilities. Star-nosed moles are semiaquatic, and we recently discovered they can “sniff” while submerged. Underwater sniffing is accomplished by exhaling air bubbles that spread over objects, and then re-inhaling the same air bubbles. This sensory ability was completely unanticipated, as it was previously (and logically) thought that mammals could not use olfaction while underwater. Another avenue of recent investigation, led by Diana Bautista of the University of California, Berkeley, takes advantage of the high density of mechanosensory neurons that innervate the star to search for molecules that mediate touch. The ion channels responsible for mechanosensation have been difficult to identify, and sensory specialists like the star-nosed mole may provide important clues. In addition, the star-nosed mole’s genome has recently been sequenced, and this resource may provide new insights into the molecular biology of touch and the development and evolution of biological novelties such as the star. I predict this unusual species has many more surprises in store.
Kenneth C. Catania is a Stevenson Professor of Biological Sciences at Vanderbilt University.