Last week, Nature issued a new paper. The paper used two different strains of mice, one lacking all serotonin neurons (called Lmx1b knockouts), and one lacking the rate limiting enzyme for the production of serotonin (called TPH2 knockouts).

The authors demonstrated that these mice, lacking serotonin, did not distinguish between sexual partners, mounting male and female mice with seemingly equal fervor. Not only that, the mice showed the same scent preferences for male and female mice and vocalized for male and female mice equally. When the serotonin in these animals was replaced, they preferred female mice over male mice. The authors concluded that serotonin regulates sexual preference in mammals.

When I heard about this paper, I knew immediately how the media were going to take this one up. Sure enough, pretty soon I saw "Scientists turn mice gay by depriving them of serotonin," "Sexual preference chemical found in mice" and "Did Chinese scientists turn mice gay?". Most of the titles are misleading, but the results of the study are enough to make anyone question. DID these scientists find the key to sexual preference? Or is it more complicated than that?

While Ed Yong gave a fantastic overview of the paper and main issues over at Not Exactly Rocket Science, I’d like to talk a little bit about the history behind this paper, the literature on serotonin and sexual behavior in animals, and with that in mind, look at this paper again with different eyes.

Many people are aware, to some extent, that serotonin has something to do with sexual behavior. We’ve all heard about the sexual side effects of antidepressants. Many of these drugs are known as selective serotonin reuptake inhibitors (SSRIs).

In the brain, serotonin is recycled back into the cell after release by serotonin transporters, so that it can be broken down or used again. When antidepressants like Prozac block these transporters, serotonin remains in the synapses between cells, and it builds up, resulting in more serotonin signaling in the brain.

This affects sexual behavior. Generally speaking, high levels of serotonin (either administered directly into the brain, or as the result of genetic absence of the transporter), inhibit sexual behavior. This means that antidepressants which increase serotonin cause problems like longer time to ejaculate in men.

Of course, if HIGH levels of serotonin cause decreases in sexual behavior, what do LOW levels of serotonin do? Pretty much what you’d expect. Stimulating an inhibitory receptor that decreases serotonin neuron signaling will increase sexual behavior. Stopping the synthesis of serotonin (resulting in a massive serotonin decrease) also increases sexual behavior. The literature seems pretty clear. In general: high serotonin = low sexual behavior, and low serotonin = high sexual behavior. This is all in males, though some research in females suggests that similar effects may be in play.

But that’s sexual behavior OVERALL. What does that have to do with male or female preference? So far, it’s hard to tell. While this is the first paper demonstrating a lack of sexual preference, Wainberg et al. looked at the effects of SSRI antidepressants on gay and bisexual men with compulsive sexual behaviors.

While the current Nature paper held this study up in support of their studies on sexual preference, I don’t think it’s as close as they say. After all, these were not just gay and bisexual men, these were specifically men with compulsive sexual behaviors. That’s not a difference in sexual preference, rather they had what would be classified by some as sexual addiction-like behaviors (excessive porn use, excessive masturbation, intrusive thoughts, etc).

Those men had decreases in masturbation and sexual desire when they got antidepressants, because antidepressants increase serotonin. Increased serotonin generally leads to less sexual behavior. Though the study was done in gay and bisexual men, it’s possible they would have observed similar findings in heterosexual men with similarly compulsive behaviors.

The big question is: how does this paper differ? As I mentioned above, they took mice with no serotonin neurons and mice with no ability to produce serotonin and looked at their sexual behavior. As we can tell from the literature, higher serotonin generally = less sexual behavior, and lower serotonin generally = more sexual behavior. These mice had extremely low serotonin. While the work looking at sexual preference in the mice is in the main body of the paper, the work on general sexual behavior is in the supplemental data.

Presumably, with really low serotonin, we’d be looking for increases in sexual behavior. But it’s difficult to detect increases in sexual behavior in male mice. Male mice will, on average, mount susceptible female mice upon contact 90% of the time. In the mice that had no serotonin, the mounting of females increased to almost 100%. So while it does look higher, this is what you call a ceiling effect, and there’s no significant difference.

They also looked at other behaviors, like how long the mice spent mounting female mice. Here again, there is what looks like an increase, but the result is non-significant.

(Supplemental Figure 1D from the Nature paper)

The interesting data is when the animals are exposed to male mice. While normal mice show basically no mounting of other males, the mice with low serotonin (either due to absence of serotonin neurons or the absence of the rate limiting enzyme to make serotonin), showed large amounts of mounting. Again the numbers are almost at ceiling, with the low serotonin animals mounting almost 100% of the time. In fact, low serotonin animals of both types mounted males and females with roughly equal preference.

(Figure 1a from the Nature paper)

The question then becomes whether we are seeing a switch in sexual preference (not to preferring males, but to having no preference between sexes), an inability to distinguish the difference between male and female mice, or simply an increase in sexual activity. There is presumably an increase in sexual activity present, though it appears to hit the ceiling. The authors have to determine whether the mice with low serotonin can tell the difference between male and female mice.

First they ran a large number of odor tests, for things like food substances, aversive smells (like predators) or other unfamiliar mice. The low serotonin animals passed all of those tests. But when given male or female genital scents or urine, they show no preference for the male or female. But what is also interesting is that they show large decreases in sniffing.

(Figure 3 from the Nature paper)

So it appears that the the mice with reduced serotonin show a lack of sexual preference. The authors then found that they could RESTORE the sexual preference for females by administered serotonin to the low serotonin animals (you’ll notice below that this increase in serotonin produced a drastic decrease in sexual behavior in normal mice).

(Supplemental figure 10 from the Nature paper)

The authors conclude that the presence or absence of serotonin in these mice controls sexual preference. But based on the work in this paper and the literature on serotonin and sexual behavior, I think this conclusion may be premature. The proposed mechanism is a good hypothesis, but is what we are seeing in these data really a change in sexual PREFERENCE? Or is it merely a drastic increase in sexual behavior in low serotonin mice, one so profound that sexual activity will occur regardless of the target? Perhaps when the authors increased serotonin in these mice, they were decreasing sexual behavior to normal levels rather than changing preference.

Further, the mice with decreases in serotonin either lacked all serotonin neurons or lacked the enzyme to produce serotonin entirely. The deletion of serotonin in this manner is a drastic change, involving many different areas of the brain and a wide variety of behaviors, including, but not limited to: aggression, social affiliation behavior, sexual behavior, vulnerability to depressive-like states, memory formation and others. The authors note that the change in sexual behavior could be due to changes in social behaviors. Some mouse models with low serotonin also show very low aggressive activity. Perhaps they are just being really friendly.

Finally, there’s the question of smell. While the mice passed tests for scents associated with food, predators and other mice in a social context, I did notice that these mice show low sniffing between male and female urine. Is it a lack of preference? Or is it, rather, a lack of ability to detect the difference? Serotonin modulates odor inputs in the olfactory bulb and can play a role in learning. So it is possible that these mice with low serotonin could be experiencing subtle effects in how they process the odors from males and females.

This is a possible issue to test. Scientists would have to change serotonin levels just in the olfactory bulb, as well as in other regions, to determine how this could impact sexual behavior. But mice rely on smell to a very large extent, especially in social behaviors (which can include mating), and the role of serotonin in smell hasn’t been ruled out yet.

So does this paper prove that there are drastic increases in sexual behavior associated with low serotonin? Absolutely. Does it show that low levels of serotonin change sexual PREFERENCE? Well, that’s difficult to say. As with anything we do with serotonin, the answer seems more complicated than that.


1) Liu et al. "Molecular regulation of sexual preference revealed by genetic studies of 5-HT in the brains of male mice" Nature, 2011. doi: 10.1038/nature09822

2) Gawker, "Scientists turn mice gay by depriving them of serotonin."

3) BBC, "Sexual preference chemical found in mice"

4) AOL, "Did Chinese Scientists Turn Mice Gay?"

5) Not Exactly Rocket Science, "Low Serotonin Mice are Less Choosy about the Sex of Partners"

6) Neurotic Physiology, "Depression post 4: the serotonin system"

7) Neurotic Physiology, "Depression post 2: current therapies"

8) Dominguez and Hull, "Serotonin impairs copulation and attenuates ejaculation-induced glutamate activity in the preoptic area" Behavioral Neuroscience, 2010.

9) Madeo et al., "The effects of citalopram and fluoxetine on sexual behavior in healthy men: evidence of delayed ejaculation and unaffected sexual desire. A randomized, placebo-controlled, double-blind, double-dummy, parallel group study." Journal of Sexual Medicine, 2008.

10) Chan et al. "The serotonin transporter plays an important role in male sexual behavior: a study in serotonin transporter knockout rats." Journal of Sexual Medicine, 2011 DOI: 10.1111/j.1743-6109.2010.01961

11) Fernández-Guasti et al. "Stimulation of 5-HT1A and 5-HT1B receptors in brain regions and its effects on male rat sexual behaviour." European Journal of Pharmacology, 1992.

12) Yamanouchi and Kakeyama. "Effect of medullary raphe lesions on sexual behavior in male rats with or without treatments of p-chlorophenylalanine." Physiology and Behavior, 1992.

13) Wainberg et al. "A double-blind study of citalopram versus placebo in the treatment of compulsive sexual behaviors in gay and bisexual men." Journal of Clinical Psychiatry, 2006.

14) Petzold, Hagiwara and Murthy. "Serotonergic modulation of odor input to the mammalian olfactory bulb." Nature Neuroscience, 2009.

About the Author: SciCurious has a Ph.D. in Physiology, a Bachelor of Arts in Philosophy and a Bachelor of Science in Biology. She is currently a postdoctoral researcher. Her professional interests are in neurophysiology, specifically the interactions of neurotransmitter systems. Her writing has appeared at ScienceBlogs, the Guardian and the New York Times, and she has been published in two editions of The Open Laboratory: The Best of Science Blogging. She blogs at Neurotic Physiology (and can also be found on Twitter) .

The views expressed are those of the author and are not necessarily those of Scientific American.