Developing sex-accurate cell culture environments

Abstract

The extracellular matrix remains under-recognized as a sex-dependent entity. Designing culture environments that account for sex as a biological variable requires considering not only cellular sex, but also the usage of sex-specific scaffolds to create a holistically sex-accurate platform.

The prevalence of many diseases varies with biological sex, and within the last 20 years it has been recognized that sex-specific phenomena occur outside the context of the full organism, as cells themselves exhibit sexual dimorphic behaviour. Evidence of sex-dependent cell function has been found in almost every tissue examined, with implications for understanding disease progression, drug toxicity and treatment efficacy.

Although advancements have been made to account for sex differences following implementation of the National Institutes of Health (NIH) requirement to consider sex as a biological variable, as well as similar recommendations by the European Commission and Canadian Institutes for Health Research, this progress is primarily in the design of in vivo animal experiments and clinical trials. Cellular sex in in vitro studies remains under-investigated and unevenly reported; an analysis in the biomaterials field revealed that approximately 10% of papers disclosed the sex of primary cell cultures¹. However, this is still far ahead of where we stand in representing sex differences in the physical structure that surrounds these cells: the extracellular matrix (ECM).

Sexual dimorphisms in the ECM

The recognition that cell behaviour is a function of cellular sex has not translated into viewing the ECM as a sex-dependent entity. However, male and female organs exhibit differences in their ECM composition, crosslinking and degradation^2–4, and these sexual dimorphisms have implications for both the biochemical and biophysical environment of cells.

For example, healthy, young (age 17–40) male hearts contain more collagen type I, III and IV than healthy, young female hearts, with the male hearts correspondingly exhibiting greater stiffness⁵. Differential expression of multiple molecules related to ECM production and homeostasis probably underlie this compositional difference. Transforming growth factor-β (TGFβ) stimulates collagen production, and TGFβ signalling is upregulated in young male hearts compared with female hearts. The collagen crosslinking enzyme lysyl oxidase (LOX) is also increased in young male hearts relative to female counterparts. Similarly, male hearts have increased production of tissue inhibitors of metalloproteinases (TIMPs), which inhibit ECM degradation⁵. Together, these ECM regulators paint a picture of increased ECM production and crosslinking, combined with decreased ECM degradation, which creates the stiffer, collagen-enriched ventricular tissue observed in healthy, young male hearts. However, almost all of these sex-specific trends are reversed with ageing; after age 50, female hearts exhibit levels of LOX and TIMP production, TGFβ signalling and presence of collagens I, III and IV that are equivalent to (or greater than) those seen in older male hearts.

ECM sexual dimorphisms have been described in other young, healthy tissues, most notably in the cartilage and brain. In cartilage, the mechanical strength of male bovine articular cartilage is over threefold than that of females, probably owing to increased expression of pathways and proteins related to the stabilization of ECM structures in males³. In the brain, sex differences in anatomy have long been recognized, but sex differences in ECM composition and tissue mechanical properties also exist. Healthy cortical brain tissue in female mice have increased expression of laminin and collagen IV content in comparison with male brains, which have increased fibronectin content⁶. Moreover, both human and animal studies have noted differences in brain mechanical properties, as several regions are softer in male brains compared with female brains⁷.

These ECM differences have been observed in healthy people, suggesting that female and male individuals do not share the same ‘starting point’ when it comes to the onset of disease. Moreover, in various cardiovascular diseases (such as calcific aortic valve disease and aneurysms),joint pathologies (such as osteoarthritis) and neurodegenerative diseases (such as multiple sclerosis), sex-related ECM differences become more pronounced with disease progression. Although it may seem contrary to our scientific training to use two different culture environments as our ‘healthy’ condition in experiments, this is what the biology calls for. That said, the field lacks a complete characterization of ECM sex differences for most tissues, healthy or diseased. Historically, ECM analyses have not separated tissues by sex. This information is needed to design sex-specific culture environments that account for the role of ECM sex dimorphisms in tissue function.

Reporting cell sex

The NIH policy on sex as a biological variable does not extend to in vitro studies. Thus, cellular sex remains unreported in a majority of published papers¹. This omission creates challenges with respect to data reproducibility and raises concerns about how to properly interpret and apply study results. We concur with the recommendations made by others in this field¹, noting that journals should require authors to report cellular sex in manuscripts, and vendors of biological materials should readily provide information about the sex, age and demographics of commercially available cells. Moreover, many researchers are limited in their ability to account for sex differences simply because cells of different sexes are often difficult to obtain from existing biobanks or vendors. Therefore, greater sex diversity in the creation of new cell lines is needed, including for human induced pluripotent stem cells, as male cells are over-represented by approximately 20% in several major biobanks.

There is also the question of whether the reported cell sex is based upon sex as assigned by external anatomy, or if cells have undergone chromosomal testing. Because sex chromosome aneuploidies are often not reflected by anatomical appearances, the method by which patient or specimen sex was determined should be disclosed across all types of studies, ranging from clinical to cellular. Finally, serum and hormones should be sex-matched when aiming to recapitulate sex-specific environments in vitro, as these factors carry critical sex-specific cues that influence how cells interact with their ECM⁸.

The risk of providing mixed signals

To understand how scaffold properties control stem cell differentiation, several studies have explored culturing cells in an intentionally ‘confusing’ environment by providing multipotent cells with mismatched cues; for example, a substrate stiffness meant to direct cells towards one differentiation path combined with a media formulation that targets a different path⁹. By placing cells in a ‘one-size-fits-all’ ECM environment, we are perhaps unintentionally creating cell confusion experiments. Although it is common practice to use serum of unknown sex and a unisex scaffold, culturing a female cell in a male environment (male serum and a scaffold with ECM composition and biomechanics associated with male tissues) may create a circumstance analogous to placing a muscle cell on a stiff substrate with osteogenic medium. If the intent of the research is to examine or account for biological differences between male and female individuals (generally defined chromosomally as XY and XX, respectively, in humans), then it is important to use culture environments where cellular sex is matched to the ECM environment and serum source.

To account for sex-specific cellular behaviours in three-dimensional in vitro models, researchers must pursue the creation of sex-specific scaffold environments. Over the years, many studies have clearly demonstrated that even small changes in ECM composition, architecture and stiffness can alter cell function; it is time to recognize that these microenvironmental variables also matter for providing sex-accurate culture environments. Similarly, these considerations extend to the creation of holistically sex-accurate models of disease. Failure to account for both cellular and extracellular sex differences in our experimental design can impede our ability to investigate disease mechanisms and treatments in vitro.

Outlook

Knowledge gaps in our understanding of ECM sex differences affect our ability to design sex-accurate cell culture environments and understand sex-specific cell and tissue behaviours. Such sex-accurate culture environments are important, as they can form the basis for investigating disease progression mechanisms, identifying biomarkers and testing new treatments. Accounting for sex differences could alter both the type and timing of therapeutic treatments offered to patients. Ultimately, progress towards sex-accurate culture environments will require efforts by scientists, clinicians, vendors and journals to characterize sexual dimorphisms, improve the availability of biological materials from different sexes, apply descriptive nomenclature to define sex of these materials and encourage reporting of sex in published research.