This is your fetal brain on drugs.

We interrupt this series on "junk DNA" and rampant folk science to bring you a months-overdue Journal Club.

I wonder how many of my readers remember this little tidbit of American genius:



I remember some very funny spoofs, mostly on T-shirts. (Back then, I think the Internet was still a toy for geeks at the NCSA.) "This is your brain. This is your brain on drugs. This is your brain on drugs with a side of bacon. Any questions?"

Marijuana, as I recall, was typically included as one of the frying pans that could turn your central nervous system into a not-very-heart-healthy staple at Denny's. It was – and probably still is – easy to get the impression that smoking pot would hollow out your skull and make you into the inspiration for a character played by Keanu Reeves.

But that's baloney. Long-term marijuana use is certainly not without effects on the brain (duh), but its most abundantly-documented pathological outcome is, well, stupidity. (Mild stupidity. How such an effect is detected in an American population is not so clear to me.) And gosh, if we intend to stamp out stupidity-enhancing behavior through legal action, we'd better send the Marines to Hollywood right now. Seriously, there are few well-established long-term negative effects of using cannabis, and most of those are associated with smoking marijuana and not with the neurological impact of cannabis itself. (Full disclosure: I have never had a joint to my lips, and the closest I've come to inhaling is second-hand at the occasional concert. It would seem that my stupidity has a different cause.)

The rules are different, though, when developing brains are the subject, and it doesn't matter whether the neuroactive substance is legal or not. Maybe pot doesn't mess up a young adult's brain, but that doesn't mean it won't affect a fetal brain. And in fact, some recent studies indicate that we should pay close attention to the possibility that fetal brain development is affected by cannabis. One of those studies, "Hardwiring the Brain: Endocannabinoids Shape Neuronal Connectivity" by Paul Berghuis and colleagues, published in Science last May, suggests that mammalian prenatal brain development is likely to be significantly impacted by cannabis. It's an interesting paper for that reason, and because it deals with two of the subjects of my own research: neuronal growth cones and Rho GTPase signaling. I'll briefly explain those terms later.

The active ingredient in pot is a chemical called Delta(9)-tetrahydrocannabinol, or THC. THC affects the brain by activating receptors on particular types of neurons in the brain, causing these neurons to release less of their neurotransmitters (the normal chemical signals used for communication among neurons). While a serious intelligent design proponent might need to claim that the "purpose" of these receptors is to help people respond to pot (to suppress nausea while on chemotherapy, for example), scientists instead sought and found the chemicals within the brain that normally act on these receptors. These chemicals are called endocannabinoids, signifying that they are cannabis-like but originate from within. (After biologists discovered the endocannabinoids, they subsequently discovered the receptors, but that's not an issue here.)

This means that a first step toward discovering the potential roles of endocannabinoids in brain development is the identification of the parts of the developing brain that display the receptors. If you know where the receptors are, then you know where the chemicals are likely to act. And those are the areas that are likely to be affected by cannabinoids like THC, that come from outside.

Neurons are the brain cells that send and receive electrical signals. A typical neuron has many (perhaps thousands) of dendrites, which receive signals from other neurons, and one axon, which transmits signals to other cells, often a great distance away.
	A typical neuron. Image credit: NIH, NIDA

During brain development, neurons have to develop their magnificent and specific architectures. Beginning as a boring little round ball, a neuron has to sprout and extend dendrites and (typically) a single axon. The axon must somehow migrate to its final position, which may be in a completely different part of the body or right next door.

When Berghuis et al. looked for endocannabinoid receptors in the developing brain, they found them in the cerebral cortex, and specifically they found them in the growing axons of the cerebral cortex. In case you haven't been introduced to the cerebral cortex, it is thought to be responsible for "all forms of conscious experience."

Layers of the developing cerebral cortex of a mouse. The red streaks are developing axons that are displaying endocannabinoid receptors. From Berghuis et al., Figure 1D.

They found the receptors in other developing brain regions, too, and they showed that the endocannabinoids are likely to be produced in those regions at those times. The somewhat surprising result raises the possibility that cannabinoids affect how the brain develops, by affecting how the axons develop.

What might these effects be? The authors found that the receptors were clustered right at the growing tips of these developing axons. This region is called the growth cone, and it's one focus of my own research, because it's obviously the place where the axon is continuously elongating, and it's a place where the skeleton of the cell must be always remodeling.

The growth cone of a mouse neuron. The red indicates structural elements of the growth cone; the green blobs are endocannabinoid receptors, and the yellow smudges indicate where the red and green overlap. From Berghuis et al., Figure 2C.

If endocannabinoid receptors are located right on the growth cone, then they are positioned to influence speed and direction of axon outgrowth. Yikes!

Okay, so endocannabinoids (and, of course, THC from pot smoke) are uniquely positioned to affect growing axons in the brain. But what's the effect? The authors show that one effect is the inhibition of steering mechanisms in the growth cone. In my favorite experiment, they put neurons into an electric field, where the growth cones tend to steer toward the negative pole. When the neurons were treated with an endocannabinoid, they failed to show this preference.

Axon growth in an electric field. Each black tracing represents the behavior of one axon. On the left, notice that untreated axons tend to grow toward the negative pole (left side), and many of those that are growing toward the positive pole are turning away from it. On the far right, notice that axons treated with the endocannabinoid grow in every direction and don't care about the electric field; the center shows how they grow when there's no electric field at all. From Berghuis et al., Figure 3D .

The authors went on to show that this effect seems to result from the activation of a well-known signaling system inside cells, mediated by a protein called RhoA. RhoA is a Rho GTPase, and I'll spare you the details since you've probably read all my papers already. :-) What matters is this: Rho signaling is known to be involved in axon growth, and is generally a negative influence on axon growth. In fact, some attempts to stimulate axon growth in the spinal cord after injury (and paralysis) are focused on the inactivation of RhoA and its partners. So this connection between endocannabinoids and Rho GTPases is further evidence of a specific – and likely negative – influence of cannabinoids on axon outgrowth in the developing brain.

But is there any evidence of a specific effect on brain development, in an animal? The final experiment presented in the paper is a genetic experiment, in which the authors examined the brains of mice in which the endocannabinoid receptor (one in particular) was genetically deleted in certain parts of the brain. And they found that certain neurons in the cerebral cortex of these mutant mice had lost almost half of their inputs, presumably due to the inability of the incoming axons to find their way to the recipient neurons. In other words, when the receptors were deleted from a subpopulation of neurons, those neurons evidently had trouble making their normal connections.

What this means is that to whatever extent the human brain resembles the mouse brain with regard to expression of cannabinoid receptors and their function in growth cones, the developing human brain is potentially vulnerable to damage, or at least alteration, by exposure to THC. And as the authors note, this may partly explain recent findings (in rats) that point to permanent alterations in brain function in pot users – alterations that may predispose these people to much more serious addictions.

I've long been inclined to skepticism regarding anti-pot hysteria, and I strongly support efforts to legalize and legitimize medical use of cannabis. But these data should make us look hard at the potential implications of cannabis exposure during human development.

Article(s) discussed in this post: Berghuis, P., et al. (2007) Hardwiring the Brain: Endocannabinoids Shape Neuronal Connectivity. Science 316:1212-1216.