When the NIH’s policy requiring consideration of Sex as a Biological Variable (SABV) was first being discussed, those of us who run behavioral experiments in laboratory animals raised a fleet of…concerns. Objections. Complaints, even. Some of these related to the perceived need to double up on certain equipment or facilities to avoid the potential for the mere presence of one sex to alter the behavior of the other sex. This might include concerns about where animals of each sex were housed and where they were subjected to behavioral experiments.
An article by Dalla and colleagues (2024) titled “Practical solutions for including Sex As a Biological Variable (SABV) in preclinical neuropsychopharmacological research” caught my eye. The vast majority of it is described as “guidelines” and the piece advances an agenda of going far beyond the initial NIH policy. However, it does address interesting questions about behavioral research in section 4, which is titled “Behavioral experiments: logistical considerations and sex-specific behavioral readouts“. In bullet points 5 and 6 under this section the authors suggest that cleaning behavioral equipment “becomes particularly important” due to odors, and add that using different sets for each sex is “advised“. For this, they cite three papers in which mouse behavior is altered by the urine of the other sex. Then they cite three papers to support the claim that “Two exceptions are operant testing and food motivation tasks where reward/motivation factors overcome these effects“. Game on. Unfortunately, all three papers are in mice and they do not involve drug self-administration, only food motivated behavior. So it touches on exactly the issue I was discussing with my colleague many years ago, and yet it is disappointingly light on data that would inform the discussion.
Things that seem truthy and are supported only by a thin and only indirectly relevant literature (or just lab lore*), such as these observations can be obstacles. Obstacles in the way of the greater goal, which in this case is to get more researchers doing a better job of including both male and female subjects in their research. Where a better job might mean more consistent inclusion of both sexes, but also more consistent treatment, which in behavioral experiments means shared treatment. Running males and females at the same time of day. Running males and females in the same equipment, in the same room, using the same staff to conduct experiments. If we don’t do that, can we ever be sure that an apparent sex difference in behavior is not in fact due to one of the minor procedural differences that were logistically necessary to keep the animals from being influenced by the other sex? Maybe one room tends to be slightly draftier than the other. Or noisier. Or hotter. Maybe the undergraduate research assistant available at 9am differ in some way from the one that comes in to work at 3pm.
I had discussions years ago with a colleague, who is in the same approximate area as my work, and we concluded that in an ideal world the field would test some of these lab lore truths as it adapted itself to SABV. Knowing what was likely to make a difference, and the likely magnitude and consistency of those differences would allow the field to work better and smarter. And maybe even more efficiently.
I have several obvious logistical questions for our usual work in the lab.
Such as, to make this brief, is it necessary to clean or separate the operant chambers used for intravenous self-administration (IVSA) experiments with male and female rats? Is it necessary to run them in different rooms? It is not all that unusual for a laboratory to use the same operant chamber to run more than one rat within a given day. This can be a logistic necessity due to limited space or equipment. These rats might be in the same study or they might be in different studies. Sure, we (most of us anyway) clean the chambers and refresh the bedding material in between each run but can we be absolutely SURE that there is no detectable smell of the animal in the chamber the prior run? Does the person running the rats need to change all of their clothes and PPE?
I decided to throw a tiny test of the hypothesis in the course of an ongoing study. We had both male and female rats doing stimulant drug IVSA, in this case involving initial training on either methamphetamine or alpha-PVP, the synthetic cathinone stimulant once called “flakka”. By the time we got around to this little test, they had about 60 prior sessions of IVSA including at least some experience with different doses of each drug. For this study we had two Pre-manipulation baseline sessions in which the rats were responding for infusions of alpha-PVP (0.05 mg/kg/infusion) in 1 hour sessions with a Fixed-Ratio 1 reward contingency (Figure 1). The key experiment was then to run a female rat IVSA session in the boxes prior to the males’ session without any cleaning or changing of the bedding in between the sessions. In this experiment, the number of infusions obtained after the females (After XX) was not significantly different from the baseline sessions. These data are very limited. And the eyeball test of the individual datapoints is slightly concerning. This may have failed to reach statistical significance but that is a very poor way to support a broad conclusion.
So, about a year later I had the opportunity to conduct another, very similar test with another cohort of male rats. In this case they’d been trained on methamphetamine (MA) or alpha-PPP (an analog of alpha-PVP) alternating with MA IVSA, had experience with both drugs in subsequent experiments and were again run in about 60 sessions before this. As above, Figure 2 shows two Pre-manipulation baseline sessions in which the rats were responding for infusions of alpha-PVP (0.05 mg/kg/infusion) in 1 hour sessions with a Fixed-Ratio 1 reward contingency. And again the key manipulation was a session on which female rats were run in the chambers prior the males’ sessions. In this case, there was a statistically significant increase in the number of infusions obtained.
There are all kinds of caveats and limitations and this is by no means a real study yet. It can always be the case that a one-time probe like this does not generalize to repeated exposure, session after session. Perhaps we would see a difference in initial acquisition over the first 10-20 sessions, something that would be critical for some approaches and experimental questions. Perhaps this is relevant for some drugs, but not other drugs. Maybe it depends on rat strain or the infusion dose? Maybe it only pertains to short IVSA sessions. Etc.
I may never find a good time to investigate this properly. But this certainly is interesting. We may try to delve into this a bit better in future.
This little test also speaks to the supposed “replication crisis”. Here are studies conducted pretty similarly that led, formally speaking, to different outcomes. A failure to replicate. One takeaway is “there is a statistically significant difference” and the other takeaway is that it doesn’t matter if you run males after females. There are reasons to prioritize the second experiment, since the sample size was 10 instead of 7. But I think this falls far short of proof. Maybe we got “lucky” with the second experiment. Sure the two experiments were similar but the training drug in one leads to higher IVSA rates and the two drugs in the second experiment lead to lower rates. Maybe it was the relatively novel experience of the “good” drug, alpha-PVP, in the second group that led them to be vulnerable to the impact of the females.
Or maybe it was something about the females themselves in the two respective groups that produced the difference?
So many questions.
*These are received wisdoms (aka tips and tricks, etc) about how to do experiments properly that may not appear consistently in methods descriptions. Or at all. All kinds of laboratory procedures and research methods may be subject to truths about best approaches that are passed down in laboratories. They are, quite often, very valuable to anyone trying to conduct the experiments. Best approaches may vary between labs doing more or less the same research…or received wisdom may be generalized across essentially the entire subfield.
TL neuro 