Abstract
Background information
Recent reports and developments on sex bias in basic and preclinical research prompted the organization of this symposium. Different points were addressed by different specialists and are briefly summarized below.
PD Dr Birgit Ledermann, President SGV, Basel, Switzerland:
Summary of oral presentations
Dr Marcel Gyger, Lausanne, Switzerland:
It is often said that the estrous cycle of female mice leads to greater variability in data than in males. The recent papers of Zucker et al., have challenged this view quite drastically; males can be as variable than females. This raises the question of whether sex of the animals used in the experiments should be appraised more clearly in Form A by the local ethical review board? Form A is a mandatory form to be filed for seeking permission to perform animal experimentation in Switzerland. A qualitative assessment of several researchers’ answers to the question of which sex they selected in their research was done. Concerns of researchers were of three kinds besides the fact that some vital processes or diseases are clearly sex-linked. The first concern was not to increase the number of animals by using additional females for 3Rs reasons. Second, females are preferred because they are less aggressive than males, and third often researchers do not have the necessary human and financial resources to include both genders in their studies. It is obvious that many studies include only one sex and that this choice is often not scientifically relevant. This sex-biased research should be reevaluated and measures should be implemented to have a sex-balanced approach in research.
Professor Irving Zucker, University of California, Berkley, USA:
Animal models in the early part of the 20th century relied on cats and dogs. An increase in the use of rats was observed in the 1950s, a peak in the use of non-human primates in the 1980s, and a sharp increase in the use of mice in 1989 that was related to transgenic technology. The rise in the use of males in experimentation in the 1960s is explained by the fact that studies prior to the 1950s, where no sex was indicated, were mostly done on males.1 There is a long list of sex differences, running into the hundreds – for example, anxiety, pain, aggressive behavior, anesthesia, neurotransmitters, and brain structure. But are researchers aware of such differences? A meta-analysis included 1600 articles from 42 journals in 2009.2 Most research in many biological disciplines is conducted primarily on males. Sex bias is most pronounced in neuroscience and pharmacology. Several recommendations are given to avoid sex bias, both sexes should be studied in sufficient numbers. Funding agencies should take actions to favor studies on females.
Professor Neil Bradbury, University of Chicago, Illinois, USA:
Why should a researcher care about the sex of the cells under investigation unless they work with specific sex-dependent diseases? The American Journal of Physiology indicates that species, sex, strain and race should be stated, whereas 75% of papers published in this journal do not state the sex of the cells. Genetic differences are intrinsic and related to sex chromosomes (5% of the genome is encoded by sex chromosomes), i.e. the zinc finger genes ZFY and ZFY. X-linked diseases are more prevalent in males, while hemizygosity in males can be lethal. There is a selective X-inactivation in females, but not all genes are subject to X-inactivation. The sex of mammary and prostate cells is obvious. Cell line sex determination can be used to identify X or Y genotype, based on a six base pair insertion in ZFY, by the presence of two or a single peak for XY or XX genotype, while variability indicates XXY or XYY karyotypes. Unfortunately, in many transformed cell lines the Y chromosome gets degraded and presents an XO karyotype. Moving towards personalized medicine it becomes important to mention the sex of animals and also that of cell lines when reporting biomedical data.
Professor Eleanor N Fish, University of Toronto, Toronto, Canada:
Sex influences multiple aspects of the immune-phenotype. Sex effects on genes are related to various factors such as X-linked activation and skewing, to a higher number of miRNAs, hormonal variations, ischemic stroke, X-chromosome linked receptors such as G-proteins, interleukin and Toll-like receptors and variations in transcriptional and translational factors, thus explaining differences in the immune-response between sexes. A sex bias is found in infection and sepsis and sex-related to immune response to vaccination. Major autoimmune diseases also show a sex-related distribution in females and males. Regulatory inflammatory pathways are dependent on estrogen, androgen and testosterone levels. Sex-related diversity is found in human microbiota (skin, gut, lung, etc). There is complex regulation of the microbiome, hormones and genes, by estrogen, and this influences immune responses at many levels.
Professor Kathryn Sandberg, Georgetown University, Washington, DC, USA:
In mid January 2014, political awareness rose on gender bias in preclinical and basic research. In 2015 all stakeholders discussed and agreed that grant submissions to the National Institutes of Health (NIH) should also include sex as a biological variable. On 19 October 2015, J A Clayton published an article on guiding principles to consider sex as a biological variable in FASEB J. 3 This will mostly affect preclinical research studying both sexes. In addition, the Georgetown consensus group with experts from different fields has proposed different strategies to promote research on sex and gender, to foster research on gender differences and establish new research institutes and promote educational training.
Dr Michael F W Festing, University of Leicester, Leicester, UK:
Many studies use only male animals, but recent developments require grant applicants to balance male and female animals in preclinical and basic research projects. Several factors need to be considered such as the 3Rs, replication, randomization and the need for enough statistical power. Several possibilities are proposed: a factorial design allowing the use of 10 females and 10 males, rather than 20 animals of a single sex, with little difference in statistical power; use of analysis of variance (ANOVA) analysis and determine two-way interactions by a randomized block design or a preferential design in little blocks that allows interpretation using a multi-way ANOVA.
PD Dr Beat M Riederer, Editor-in-Chief, Laboratory Animals, UK:
Females have been neglected in biomedical research for a long time, leading to withdrawal of prescription drugs due to a sex bias in pain research and a bias in publishing scientific data. Two important points which may promote research that includes both genders are economics and ethics with regard to the 3Rs. Breeding only one sex versus breeding both females and males costs double and uses double the number of cages, which is especially true for transgenic animals. Publishing biomedical research in Laboratory Animals requires that the ARRIVE guidelines are respected and that species, strain and sex are mentioned and the number of animals used is justified. In addition, the use of only one sex needs to be mentioned and justified. Such information may provide valuable indications for the planning of preclinical studies. A recent editorial has addressed this issue. 4
Professor Richard Svanbäck, Uppsala University, Uppsala, Sweden:
Vertebrates harbor diverse communities of symbiotic gut microbes. It has been shown that microbiota composition depends on interactions between host diet and sex within populations of wild and laboratory fish, laboratory mice and humans. Experimental diet manipulations in laboratory stickleback and mice have confirmed that diet affects microbiota differently in males versus females. The prevalence of such genotype and environment (sex by diet) interactions implies that therapies to treat dysbiosis might have sex-specific effects.
Professor Robert E Sorge, University of Alabama at Birmingham, Birmingham, USA:
Pain is subjective and sex-dependent. Women greatly outnumber men in reports of chronic pain. There are known differences in receptor function and location between the sexes in addition to hormonal differences. There are clear sex differences with respect to immune system involvement in pain in animals, suggesting that the same might be true in humans. There is a significant sex difference in neuropathic pain in mice. The mechanisms of chronic pain in male mice involving microglia in a testosterone-dependent manner, while female mice utilize T-cells to mediate pain instead. These data translate to a variance in presentation and prevalence of chronic pain between men and women.