動物実験2：セックスとジェンダーの相互作用の分析

The Challenge
Method: Analyzing How Sex and Gender Interact in Animal Research
Method 1: Analyzing Sex
Method 2: Analyzing Factors Intersecting with Sex
Method 3: Analyzing How Sex and Gender Interact
 

Method: Analyzing How Sex and Gender Interact in Animal Research

How can we best design animal studies to take into account sex (biological characteristics) that interact with gender (sociocultural or environmental factors and processes)? The figure below shows the complex interdependency of sex and gender throughout the rodent life cycle.

Method: Analyzing Sex

• 1. Sex differences must be investigated before they can be ruled out (see Not Considering Sex Difference as a Problem).
• 2. Research can be done stepwise. Male and female animals should be strain- (or strain and genotype) and age-matched, and reared under identical conditions (cages, bedding, diet). Females should not be breeders unless required for assessment of the phenotype.
  • Step 1. Total sample size (based on power calculations): Adopting a strategy of both female and male animals or cells seems likely to allow detection of at least some sex influences, namely the largest ones that presumably researchers would first want to detect, with no impact on sample size or cost.
  • Step 2. Sex-based powering: tests hypothesis in both males and females and power each to determine effect.
  • Step 3. Comparison between sexes: power study to determine the actual “sex effect.” Testing for sex effects has a financial cost. However, a demonstrated sex difference justifies sex-specific research because harm in one sex is costly to society and individual patients. Overall, it is less expensive to understanding sex in the basic science phase than during the more costly clinical trial phase. This may decrease the number of drugs that fail in development and also help companies avoid being forced to remove drugs from the market due to adverse events in one sex.
• 3. To appreciate the presence/absence of sex effects, researchers should also evaluate overlap between groups (similarities between males and females) and difference within groups (differences among males or among females). Overemphasizing sex differences should be avoided.
• 4. Finding no sex effect should also be reported. To reduce publication bias, researchers should report when sex differences (main or interaction effects) are not detected or when data regarding sex differences are statistically inconclusive (Wizemann, 2012). Reporting null results is crucial for meta-analysis.

For phenoytpes that do not display sex difference, future experiments should be sex inclusive, that is include equal numbers of randomly selected males and females for each test group studied. Not every experiment needs to be designed to evaluate sex differences. However, for every experiment, the sex of the animal test subjects should be noted in the article and reported in the methods section to ensure that experiments are reproducible and findings (in one sex) are not over-generalized (to the other sex) (Wizemann, 2012).

View General Method

 

Method: Analyzing Factors Intersecting with Sex:

1. Considering the estrous cycle (Byers et al., 2012)
• a. Becker et al. (2005) recommend that researchers record the stage of the estrous cycle: Testing can include two groups of females in two specific stages of the rat estrus cycle or four groups of females, representing the four days of the cycle. Becker et al. caution against testing females on random days of the estrous cycle, because a sex difference that varies with the estrous cycle might be overlooked. They also caution that a small group of females in close contact may synchronize cycles, leading researchers to miss variations that occur with cycles.
• b. In a meta-analysis of nearly 10,000 traits, Prendergast et al. (2014) found that, for most biological measurements, females are no more variable than males. Other factors, including group versus single animal housing, can have a greater impact on variability of a trait than stages of the estrous cycle.

2. Menopause Models
Menopause is an emerging area of research in animal modeling studies. One study reported that immunological changes accompany this hormonal transition. Ovariectomized mice undergo "acute menopause," and exhibit "reduced lymphocyte chemotaxis, mitogen-induced T cell proliferation responses, and [Interleukin-2] production" (Marriott et al., 2006).

3. Pregnancy or Pseudopregnancy
Less than 10% of medications approved by the U.S. Food and Drug Administration since 1980 has enough information to determine risks for birth defects (Adam et al., 2011; Mishra & Mohanty, 2010). New animal research that evaluates drug safety should assess effects on the dam and the fetus during pregnancy and lactation (McDonnell-Dowling & Kelly, 2015).

4. Pharmacokinetics
The estrous cycle can also affect pharmacokinetics. Kulkarni et al. (2012) found that the oral bioavailability of genistein, a soy isoflavone with antioxidant properties, was inversely correlated with estrogen level (which regulates hepatic disposition of a drug).

When can experiments be done in only one sex?

• 1. When studying a sex-specific phenomenon, such as ovarian cancer or prostate cancer.
• 2. To address inadequate of published data for one sex in a particular area.
• 3. Where there is strong, statistically robust evidence that sex does not influence a trait or outcome.

In diseases where one sex predominates, such as breast cancer, both sexes may still need to be included but researchers may choose not to use them in equal numbers. Things become more complicated in instances where a disease differs profoundly in men and woman but an animal model has been developed for only one sex. For instance, rodent models of hypertension do not accurately reflect the physiology of heart disease in women. In these instances, new models will need to be developed.

Method: Analyzing How Sex and Gender Interact

Animal research includes the interaction between sex (biological characteristics, such as genes, hormones, age, reproductive phase, strain, etc.) and gender (socio-cultural or environmental process, such as caging practices, attitudes and behaviors of researchers, room temperature, diet, etc.). The double-ended arrows represent interactions between sex and gender

Gender interacts with sex, and vice versa. Environmental processes may impact male and female animals differently, such as caging practices or differential handling. These processes may include gender. Researchers should not identify an effect as dependent on sex (or a biological trait) when, in fact, it depends on an environmental condition.

View General Method

 

Animal research includes the interaction between sex (biological characteristics, such as genes, hormones, age, reproductive phase, strain, etc.) and cultural or environmental processes (such as caging practices, attitudes and behaviors of researchers, room temperature, diet, etc.). The double-ended arrows represent interactions between the biological sex characteristics of the animal and lab environmental factors. Environmental processes, such as caging practices or differential handling (which may include gender assumptions and practices on the part of researchers), may impact male and female animals differently. Investigators should not identify an effect as dependent on sex (or a biological trait) when, in fact, it is influenced by an environmental (lab) condition.

One case in point is researcher sex. This example focuses on pain research in the lab. Researchers induce pain in rats and mice. They find that rats and mice don’t show their pain to men researchers. Animals don’t show their pain when a man is in the room, as compared to an empty room, but they do show their pain when a woman is in the room. The researchers identified this as the “male-observer effect.”

What’s going on? It’s not how the researchers act or how they handle the animals. The animals smell the men’s pheromones. According to Jeffrey Mogil at McGill University, this phenomenon may throw into question all prior results from pain research.