What makes for resilience?

Say you take two seemingly similar mice and humiliate them. One becomes submissive and anxious. The other continues to behave normally. What distinguishes the two?

Researchers have performed this experiment, taking normal small mice and exposing them to large, aggressive mice in a model called “social defeat.” The defeated mice are then housed next to the aggressive mice. Most of the small mice show signs of what looks like anxiety, depression, and low status. For the rest of their lives, they will defer even to companion mice they had engaged with before.

But some of the small mice don’t change, even after days of exposure to the large bullies. And the small mice — the anxious and the unflappable — are not just similar, they are genetically identical.

Of course, genetically identical twins are not indistinguishable. Through the vagaries of very early development — the random movement of molecules, changes related to position in the womb, differences in nutrition — by the time of birth, twins differ slightly. Their DNA segments show varying levels of activity.

Scientists can now measure these differences, using microarray DNA chips that compare thousands of genes, gene sections, or indicators of gene activity. The research shows that mice with the same sequence of DNA on each chromosome are epigenetically distinct.

Epigenetics refers to gene expression that derives from experience. I promised in a previous post to convey some of what I had learned about epigenetics at a session held at the recent annual meeting of the American Psychiatric Association. Much of what was reported on was not new; results of the key experiment were published two years ago in Nature Neuroscience, by Nadia Tsankova, working in Eric Nestler’s laboratory at the University of Texas Southwestern Medical Center in Dallas. At the meetings, Nestler reported on subsequent progress.

What follows is reasonably difficult material — in part because I am compressing complex information; in part because I am not an expert in this field and so lack the ability to simplify. (Readers are welcome to point out any mistakes I have made.) If you skip the sentences you don’t understand, you’ll likely still get the gist. So here goes:

Looking at DNA only, we seem to be simple creatures. Relative to worms, mammals have only one-and-a-half times the DNA. But part of what makes us (and rodents) distinctive is that we have many “junk,” or non-coding, sequences, on the order of a hundred times more than worms do. This excess allows for more folding of DNA and the associated chromatin proteins that compose our chromosomes. We display some genes so that messengers can attach to them. Other genes remain hidden in folds.

In some cases, this expression or repression is regulated by the attachment of small chemicals to the exposed parts of the DNA complex — for the chemists in the group, through deacetylation and methylation of histone tails. These alterations affect whether the genes are active or dormant.

In the mouse model, methylation of the histone tail prevents the cell from producing factors that allow for the making of new cells and the formation of new cell connections. For those are familiar with the brain-derived neurotrophic factor (BDNF) theory of depression (I outline it in Against Depression), the methylation suppresses the production of BDNF and thus acts against resilience.

Scientists knew to look at this process because when they compared gene arrays in the defeated and resilient mice, researchers found differences in the methylation of a part of a gene that regulates the production of BDNF.

In theory, if you can prevent or reverse the methylation of key parts of the DNA complex, you prevent the effects of intimidation or make the timid mouse ordinarily bold again. Conventional antidepressants have this effect — almost. If you treat the intimidated mice with imipramine, one of the oldest antidepressants, you get a return of BDNF production and, with it, normal boldness. (Similar results in other experiments occur with the newer antidepressants, the SSRIs, like Paxil and Prozac.) But imipramine does not fully undo the initial damage; instead, it induces a neurochemical compensation. On an epigenetic level, the antidepressant-treated mice still bear the marks of social defeat.

And epigenetic change may be heritable change within the mouse brain. When the mouse makes new nerve cells, they, too, will have DNA folded in a form that sustains timidity. The social defeat is an environmental change that has a genetic effect — within the given mouse, though, of course, not in its sperm or eggs. The early experience appears to mark the brain forever.

At the meetings, Nestler and others reported on attempts to induce more direct antidepressant (or anti-timidity) effects. Instead of imipramine, researchers looked at “histone deacetylation (HDAC) inhibitors.” Some of these medications are used in cancer treatment. Some psychiatric medications, like valproic acid (Depakote), used in bipolar disorder, are HDAC inhibitors as well. But some of the HDAC inhibitors infused into mouse brains are keyed to the specific injury induced by social defeat. The HDAC inhibitors appear to work as antidepressants and in some cases, more effectively than conventional medications; in particular, combining Prozac and an HDAC inhibitor was more restorative than giving Prozac alone. Direct genetic changes (through genes introduced via viruses) can have similar results, creating resilient mice.

I have written at length about the frustration in psychiatry that for decades we have not seen a truly novel approach to depression treatment, one that goes beyond neurotransmitters. The epigenetic research points in a new direction, looking inside the cell and even beyond the gene, to quite simple gene modulation — the addition or subtraction of a chemical at one or two sites.