I had to kill a mouse this morning. It had invaded my daughter's bedroom, and our pathetic cat was doing his usual thing of responding with frantic confusion, chasing it and batting it around uselessly—so I coolly walked in, whacked it with a broom, scooped up the sad dead little beastie (which was so warm and soft), and deposited it in the Circular Funerary Receptacle.

If Science has its way, though, this commonplace occurrence could be more stressful.

First they made mutant Mighty Mice, with huge bulgy muscles. Then they figured out how to tweak their genes to get them to regenerate. And now they've lopped out a gene to make them completely fearless.

I swear, I'm going to have to scan the literature for sabre-toothed mice, tool-using mice, and mice with homicidal tendencies. Someday I'm going to try to get rid of a rodent, and it's going to turn around and whack me with a broom.

OK, that's a little bit of an overstatement. What researchers have done is identify a molecule involved in learning, which is active in a particular pathway that is involved in fear conditioning. It's a simple situation: if you give a mouse a conditioned stimulus, like a particular sound, and then give it a nasty unconditioned stimulus, like an electric shock, it will soon learn to react fearfully to just the sound alone. This is a good thing, a sensible thing, a smart thing for the mouse to do—if you're living in a cage surrounded by sadists in white lab coats, it's a good idea to learn how to react to warning signals.

The pathways consist of somatosensory information from the cortex and thalamus, an important way station in somatosensory integration, which all converge on a structure called the amygdala, which appears to be responsible for learning the association between different stimuli. The investigators screened the mouse brain for genes that are enriched in their expression in the amygdala, and found several: one of interest is stathmin/oncoprotein 18, a phosophoprotein that is found in the cytoplasm. Stathmin is an inhibitor of microtubule formation; it basically makes the skeleton of the cell a little less rigid, a little more unstable. It's this property that may be responsible for its role as a prerequisite for learning, since it may enable new synaptic contacts to be formed.

Major neural pathways transmitting tone CS and foot-shock US to the amygdala. Areas in red transmit CS to the LA; areas in black transmit US to the LA. Gray arrows show the direction of sensory information. CE, central nucleus of the amygdala; S1, primary somatosensory cortex; S2, secondary somatosensory cortex; PaIC, parietal insular cortex; TE3, temporal cortex area 1; LA, lateral nucleus of the amygdala; BLA, basolateral nucleus of the amygdala; PRh, perirhinal cortex; LP, lateral posterior thalamic nucleus; PV, paraventricular thalamic nucleus; Re, reuniens thalamic nucleus; SG, suprageniculate nucleus; MGd, dorsal division of the medial geniculate nucleus (MGN); MGm, medial division of the MGN; PIN, posterior intralaminar nucleus.

So Shumyatsky et al. did the obvious thing: they created knock-out mutations of this gene. The resulting mice look normal, right down to the density of synaptic spines in their brains, and they also aren't obviously brain damaged—they can still learn to run mazes like wild-type mice, so they have retained good spatial memory. Both respond to a shock in the same way, by freezing. Where they differ is in learned and innate fear. Wild type mice learn to associate a conditioned stimulus like a tone with a shock, and will freeze in response to the tone alone. The mutant mice are significantly different, and learn much more slowly. Mice also naturally avoid open spaces, and stereotypically stick to sheltered edges and corners; this behavior can be assessed with open field and elevated mazes, and again, the mutant mice were braver and spent more time in the open.

The mice also have deficits in long-term potentiation (LTP) in the relevant areas of the brain. LTP is a critical process in generating memory which involves changes in the strength of synaptic connections that can last from minutes to years.

They have a straightforward summary of what is going on in these mouse brains.

The current models of fear memory formation suggest that US and CS convergence in the LA leads to an increase in synaptic strength in the afferent inputs to the LA, thus providing the necessary cellular basis for memory encoding. This is accompanied by the recruitment of the numerous intracellular processes, some of which are dependent on the proper function of MTs [microtubules], including local events in the synapse that require protein and RNA transport to the synapse. This transport utilizes actin filaments (located in synapses) and MTs (located in axons and dendrites and maybe in synapses in limited quantities) as tracks by which cargo-transporting motor proteins move. Taken together, our work provides evidence for the possible role of MT dynamics in the regulation of innate and learned fear. Acting without the GRP/GRPR pathway, stathmin controls innate fear. However, acting in concert with the GRP/GRPR pathway, stathmin may control fear conditioning by regulating the ability of synapses in the neural circuitry of fear conditioning to undergo LTP.

These mutants are a tool to probe the molecular mechanisms underlying learning and memory, not a fiendish plan to generate super-mice.

I guess we should be too worried yet. Even if we find a new breed of mice with massive muscles and impressive powers of regeneration, if they are fearless and strut boldly across my kitchen floor, they're still going to succumb to a little broom-whacking. Fear is a good thing for mice; it keeps them alive and out of danger.

When they start tinkering with the genes involved in growth and produce mice 12 feet tall at the shoulder, that is when we should start to sweat.

Shumyatsky GP, Malleret G, Shin RM, Takizawa S, Tully K, Tsvetkov E, Zakharenko SS, Joseph J, Vronskaya S, Yin D, Schubart UK, Kandel ER, Bolshakov VY (2005) stathmin, a Gene Enriched in the Amygdala, Controls Both Learned and Innate Fear. Cell 123(4):697-709.