"In a discovery that raises questions about the validity of a huge volume of animal research…a study has found that mice dislike the smell of men," The Times reports. A study found that “man smell” may cause a stress response in mice…
"Eek! Why mice are unafraid of women,” is the bizarre and not entirely accurate headline in The Daily Telegraph today. The Telegraph and other papers have reported on a lab study that suggests “man smell” may cause a much higher stress response in mice than the smell of women.
If these findings are accurate, and the presence of male researchers does in fact influence rodent behaviour, it could cast doubt on the validity of decades of research using rodents.
The study measured the response of mice to pain in a variety of conditions. When mice have heightened stress levels, it is thought that pain-killing chemicals kick in as a response. Counterintuitively, however, decreased physical pain can be a sign of increased mental trauma.
The researchers found that mice did not appear to be in as much pain if a man, a T-shirt that had recently been worn by a man or the bedding of non-castrated male animals were placed near them. Women or T-shirts recently worn by women did not have any effect. Levels of a stress hormone also significantly increased when male odour was near, but not when female odour was in close proximity.
The researchers suggest that the mice are stressed by the presence of male odour and that either they consciously pretend not to be in pain, or it happens as a natural response to stress. They believe that the sex of experimenters in laboratory studies had an impact on the results and should be taken into account in the future.
The study was carried out by researchers from McGill University, Quebec; the University of Montreal; the University of Alabama; the Karolinska Institute, Stockholm; and Harvard College, Pennsylvania. It was funded by the Louise and Alan Edwards Foundation, the Natural Sciences and Engineering Research Council of Canada, and the US National Science Foundation.
The study was published in the peer-reviewed journal Nature Methods.
The UK’s media reporting was reasonably accurate. However, despite the study finding that mice appear to be more stressed by males than females, it didn’t examine whether mice become “more timid” in the presence of males and “bolder” around women. This means the MailOnline’s headline question: “Is this why women are scared of mice…?” can be answered with a categorical “no”.
The Times’ reporting of the study was the most useful, as it grasped the wider implications of the research: that previous work involving mice, particularly research studying stress responses, may have been influenced by the presence of male researchers.
This was a laboratory study of mice and rats, which aimed to see if their behaviour was affected by the gender of lab technicians. Lab technicians had thought that mice behaved differently when they were in the room with them and wanted to conduct an experiment to see if this true, as it may affect the results of other laboratory research.
The researchers measured the response of mice to pain under a variety of conditions, to see if it was affected by men, women, the male or female smell and the smell of other male mammals.
There were four male and four female researchers, and they used between eight and twelve mice per experiment, only using each mouse once. The mice were looked after by men except in one study, where they were looked after by women.
The researchers injected both hind legs of the mice with a solution that would cause pain and inflammation.
After the injections, the mice were either left in an empty room, or a male or female researcher sat in the room about half a metre from the cages.
Facial expressions of pain were recorded using a technique called the Mouse Grimace Score (MGS). This is calculated by looking at a series of still images and scoring each on a scale of no pain (0), moderate pain (1) and severe pain (2) compared to their usual expression. The results are totalled and averaged (see this manual for a brief summary of the MGS (PDF, 208kb)). The researchers also measured the level of corticosteroid that the mice produced, as this is a hormone known to increase in response to stress.
The researchers repeated the experiment by placing a T-shirt that had been worn by the male or female researchers on the chair. They then repeated it with gauze soaked in chemicals that are thought to be secreted in higher concentrations in men than women (human pheromones).
Further experiments used bedding material from unfamiliar male mice, guinea pigs, rats, cats and dogs. They compared results for animals that had been castrated.
Some of these experiments were repeated with rats.
The researchers then reanalysed data they had collected from other experiments, with each experiment using between 226 and 610 mice, to see if there was a difference in the level of pain expressed, depending on whether the laboratory researcher had been male or female.
In one experiment, facial grimacing of the mice was significantly reduced in the presence of each of the four men compared to an empty room, by an average of 36%.
There was no effect if any of the four women were in the room compared to an empty room.
The researchers were able to replicate the results by placing T-shirts worn by the male researchers half a metre away from the mice. This reduced facial grimacing for 30 to 60 minutes. However, also placing a female worn T-shirt next to the male worn T-shirt stopped the effect. There was also no effect if just a T-shirt worn by a female researcher was placed near the cage.
Three chemicals thought to be secreted more by men reduced facial grimacing.
Facial grimacing was also reduced if bedding from unfamiliar animals that had not been castrated was used. Bedding from other animals who they were familiar with or who had been castrated did not reduce facial grimacing.
The level of the stress hormone, corticosteroid, increased when the mice were exposed to T-shirts worn by men, but not by women. It increased to the same level as when mice are restrained for 15 minutes in a tube or forced to swim for three minutes.
When reanalysing previous research, they found that the pain thresholds of mice appeared to be higher if the laboratory experimenter was male.
The researchers suggested two explanations for the results. The first is that the mice might be consciously pretending not to be in pain when they can smell unfamiliar males nearby. The second is “stress-induced analgesia”, which is an innate (natural) response where pain processing in the spinal cord is prevented by the stress.
The researchers concluded that “experimenter sex can thus affect apparent baseline responses in behavioural testing".
“Although it is short-lasting, stress caused by male experimenters may represent a confound of much existing animal research extending even to nonbehavioural studies in which tissues were obtained from live rodents euthanised by either male or female personnel.” They say that the “findings strongly suggest that standard laboratory practice should account for experimenter sex when investigating any phenomenon possibly affected by stress”.
This interesting laboratory experiment suggests that mice have a higher stress response to males than females. However, it does not prove that mice would be more or less timid towards either sex, as reported by the media.
The implications of this study are that the sex of the laboratory experimenter may have affected the results of tests using rodents.
Importantly, the research report does not make it clear if the washing habits and use of deodorants and perfumes was taken into account by the researchers.
It is not clear from this study how big this difference could be and whether it would have any impact on whether a drug or technique should progress to human clinical trials.
One method that could be useful in the future would be to run two identical mice studies: one using just male researchers and the second using female researchers. The results could then be compared to see if there were any significant differences.