Sex
John C Bradford 1
and Jennifer L Robinson 1,2
1 Bioengineering Graduate Program, University of Kansas, Lawrence, KS, United States of America
2 Department of Chemical and Petroleum Engineering, University of Kansas, Lawrence, KS, United States of America
For functional tissue regeneration post injury, congenital defect, or tissue resection, it is critical to understand the driving mechanisms that promote regeneration rather than fibrotic scarring to reduce the onset of degenerative disease [ 161 ]. Functional tissue healing after injury follows a temporal process of controlled inflammation to clear tissue debris, cell migration to the injury site, proliferation, and new ECM production to close the wound. This process primarily involves immune cells, resident and circulating progenitor cells, and resident FB-like cells [ 162 ]. Historically, research conducted on mechanisms of tissue repair and regeneration focused on male samples and pathologies that were assumed to be directly translated to female tissue and disease. However, an individual’s biological sex significantly affects their ability to regenerate functional tissue [ 163 ]. Examples include the reduced ability for women to heal and regenerate new, healthy epidermis [ 164 ], cartilage [ 165 ], fibrocartilage [ 166 ], bone [ 167 ], skeletal muscle [ 168 , 169 ], smooth muscle [ 168 , 169 ], cardiac muscle [ 168 , 169 ], adipose [ 170 ], and neural tissue [ 171 ] after menopause, which results from a significant loss of sex hormone signaling and disproportionately enhances the risk for many degenerative diseases including osteoporosis, osteoarthritis, cardiovascular disease, and Alzheimer’s disease. The repair and subsequent regeneration of these tissues is modulated by macrophage polarization which is directly impacted by estrogen signaling [ 164 , 168 – 170 ]. Based on the NIH Office of Research on Women’s Health’s 2020 Strategic Plan [ 172 ], there is a push to ‘incorporate findings of sex/gender differences in the design and application of new technologies, medical devices, and therapeutic drugs’. However, in a PubMed search of TE and regenerative medicine publications from 2019, only 28.4% of the 10 651 publications reported subject sex at all [ 163 ] ( figure 15(A) ). Of that subset of studies, only 38% reported using both male and female samples. Such issues highlight the need for including sex as a variable in preclinical studies, specifically those focused on regenerative therapies. Also, the clinical and epidemiological data and the mechanisms that drive sex differences in tissue regeneration have lagged.
While it is exciting to see increased focus interrogating sex differences in tissue repair and regeneration such as with bone in the context of osteoporosis, this is still uncharted territory. Until there are sex-specific diagnosis considerations, treatment options, and therapy modalities for tissue regeneration in diseases with female predispositions, there is still much to do. In a recent review, we compiled the current knowledge on estrogen’s role in maintaining musculoskeletal cell stemness and reducing senescence [ 163 ]. Surprisingly, not many studies have been conducted to glean much understanding ( figure 15(B) ). Thus, a mechanistic understanding of what drives sex differences in tissue repair and regeneration is needed.
The main challenges limiting advancement in strategies to account for sex differences in tissue repair and regeneration are deciphering a mechanistic understanding from both clinical and in vitro studies, ensuring reproducibility of the results, and the understudied mechanisms of sex differences in repair and regeneration. First, it is difficult to determine causality from epidemiological data due to confounding variables including fluctuating sex hormone levels with wide patient variability, hormone receptor levels, and environmental factors [ 173 ]. To address this, in vitro/ex vivo and/or animal studies are conducted to determine the sex-dependent factors responsible for the observed effect. The first area of focus is relevant models for differentiating and understanding sex hormone signaling response in biological male and female cells based on phenotype and source. Based on current knowledge, there are many unknowns about the receptors involved, receptor-mediated signaling, sensitivity of the receptor based on source, and resulting signaling pathway involved in transcriptional effects. Second, how these parameters change based on the cell host source and the cell–material interactions based on differences in structural and mechanical cues of the cell microenvironment have not been investigated. Finally, to understand the biochemical, structural, and mechanical pathways involved in tissue repair and regeneration, multiple cell types, hormones, and cell microenvironments must be considered. This is where TE models using biomaterials engineered to mimic native tissue are critical.
The second major challenge is reproducibility given the large number of variables involved in the process including age of cells, hormone levels, socioeconomic background, ethnicity, and other genetic and epigenetic factors. For example, while 17 β -estradiol is often studied as it is the main circulating estrogen, additional sex hormones such as progesterone, relaxin, and testosterone play a role in tissue repair, regeneration, and homeostasis and naturally occur from sub-nano to micromolar concentrations in vivo . In addition, exogenous sex hormones are commonly taken as contraceptives, and HRT to offset symptoms of menopause and aging in males; however, these patients are excluded in clinical trials. Technological advancements must be used to methodically determine the driving parameters that dictate sex differences in tissue repair and regeneration in models that mimic native tissue to generate clinically relevant data.
Excitingly, the field is at the cusp of addressing the challenges in determining sex differences in tissue repair and regeneration. In our opinion, there are four major advancements that are critical to achieve this endeavor: the -omics revolution, physiologically relevant tissue engineered models, machine learning/artificial intelligence, and high throughput in vitro models.
Using transcriptomics, proteomics, metabolomics, lipidomics, etc, at the single cell scale, more holistic, large data sets are generated to better understand differences in male and female samples. Figure 16 highlights the tissue-specific genes that are likely responsible for sex different responses in meniscal fibrochondrocytes dosed with estrogen ( figures 16(A) – (C) ) [ 163 ], in liver and adipose tissue [ 174 – 176 ] ( figure 16(D) ), and a systemic review of differential gene transcripts over multiple tissue types [ 175 ] ( figure 16(E) ) using transcriptomics. As another example, proteomics using techniques such as tandem mass spectroscopy provides full investigation of changes to the ECM proteins, information used to mimic the chemistry and structure of the native ECM in biomaterial design. Physiologically relevant TE models using biomaterials are needed to fundamentally interrogate the components and pathways driving these sex differences. It has rapidly become apparent that 2D models do not accurately recapitulate the native structural, mechanical, and biochemical microenvironment of native tissues and fail to provide the signaling that cells recognize in vivo . Comparing the microenvironment of male and female tissue, there is evidence that collagen fiber structure, orientation, and resulting mechanical properties differ [ 5 , 177 – 179 ]. A 3D scaffolds and cellular aggregates provide solutions to study crosstalk between multiple cell types and allow for precise and controlled mechanical testing. Knowledge gained from these in vitro models will be imperative for design of biomaterial products that promote sex-specific repair including the ability to control release of sex hormones based on promising pre-clinical data for skin and bone regeneration [ 179 ]. TE incorporates recent advances in genetic engineering in CRIPSR/CAS9 and lentiviral vector systems to interrogate and differentially regulate genes of interest as revealed by the transcriptomics. Machine learning and artificial intelligence can be used to determine novel and optimal combinations of biomaterial properties and cell–material interactions for more targeted synthesis and testing with decreased time and cost. Further, microfluidics allows for precise, nano-to-micron scale and high throughput control of a cell’s microenvironment. One example of their usage is in generating concentration gradient devices that allow the investigation of a large range of biologic concentrations with less time and cost.
It is becoming more widely known that biological sex plays a role in tissue repair and regeneration as guided by the immune and stem cells that orchestrate these processes. However, the factors that guide this response and effect are unknown. For advancements to occur, researchers and clinicians must consider sex differences as a variable in diagnosis and treatment and remove pre-conceived assumptions from previous understanding based on studies prior to sex as a variable being required. Information gleaned from the discussed studies are critical for both biological male and female health as sex hormone signaling and cell response to microenvironmental changes is important for physiological development, tissue repair, regeneration, and homeostasis in both sexes. Increasing this understanding is imperative for biomaterial product design to promote sex specific regenerative pathways.
J B and J R acknowledge support from NIH NIGMS R35GM143081 and KU Startup Funds. The content is solely the responsibility of the authors and does not necessarily represent the office views of NIH or KU.
The
Patricia K Thomas 1,2,3
and Anthony G Lau 1
1 The College of New Jersey, Department of Biomedical Engineering, Ewing, NJ, United States of America
2 Department of Biomedical Engineering, Wake Forest University School of Medicine, Winston-Salem, NC, United States of America
3 Virginia Tech-Wake Forest University School of Biomedical Engineering and Sciences, NC, United States of America
Hormones fluctuate throughout a female’s life cycle. There is a monthly menstrual cycle of hormones that occurs unless pregnancy takes place. Pregnancy has its own cycle of hormones throughout the 40 weeks, followed by a period of lactation. In states of pregnancy, postpartum, and non-pregnancy, there are hormone level changes which include progesterone, estrogens, LH, and FSH ( figure 17 ). These changing hormone levels throughout each cycle affect all tissues of the body, whether it is increasing joint laxity or affecting bone strength [ 176 , 181 – 183 ]. For example, the rise in estrogen seen in the menstrual cycle is associated with increased joint laxity in sports medicine [ 176 ]. However, these sex-specific hormones are not the only hormones affecting these tissues. Other hormone levels, such as calcitonin and parathyroid hormone (PTH), which both regulate osteoclast activity in bone, have been shown to change over the course of pregnancy [ 183 , 184 ]. Most female bone health research has been focused on the post-menopausal period of life [ 185 ], with limited studies on the bone changes that occur during pregnancy. The bone changes around pregnancy are important to understand, as they set the trajectory of bone health leading up to the onset of menopause.
During pregnancy, there are many physiological and hormonal changes that can impact bone health. It is well known that typical pregnancies affect both total calcium and bone metabolism hormone levels. While ionized calcium levels and phosphate levels in the blood are constant throughout the pregnancy in humans, the total serum calcium, which includes ionized, complexed, and albumin bound fractions of calcium, decreases [ 182 , 184 , 186 – 189 ]. In contrast, total vitamin D levels increase early in pregnancy, and free vitamin D levels are found to increase during the third trimester [ 182 , 184 , 186 , 188 ]. The increased vitamin D leads to an increase in calcium absorption in the intestines, allowing for the mother to maintain the ionized calcium levels. Hormonally, calcitonin levels are increased, promoting bone formation [ 183 , 184 ]. However, it is difficult to determine the state of bone remodeling because serum markers are an indirect measurement of bone remodeling, and there are confounding effects with other hormones.
One study in humans found that urinary cross-linked Type I collagen N-telopeptide (NTx), a biomarker for bone resorption, peaked in the third trimester and postpartum [ 189 ]. The study also noted that estrogens also increased throughout pregnancy with further increases after birth. Along with estrogen increase, there was a decrease in total calcium levels in the serum. In humans, bone mineral density (BMD) in the spine and trochanter of the femur was also found to decrease significantly between baseline and postpartum [ 189 ]. The hip and femoral neck BMD did not change significantly, likely due to Wolff’s law compensating for the increased weight on these bones.
In pregnant rats, no significant differences in torsional mechanical properties of femora were found until day 20 of a 22 d gestation [ 182 ]. However, the pregnant rats had smaller fracture angles, greater stiffness, and lower shear moduli compared to controls. Histologically, there were larger bone crystals at day 20 than controls. This leads to fewer crystals in the same volume, allowing for fractures to propagate more easily, even though the mineral content was not significantly different. There was also an increase in vasculature in the cortical bone of pregnant samples, which would also affect the material properties that were not directly tested [ 182 ].
Postpartum, bone appears to temporarily demineralize to meet calcium requirements during lactation. This is suspected to be caused by low estrogen levels but might also be caused by PTH-related protein (PTH-rP), which increases during pregnancy and stays elevated during lactation [ 183 , 184 ]. The mean ionized calcium level appears to increase along with serum phosphorous levels, but PTH, free and bound vitamin D levels fall after birth [ 184 ]. Calcitonin increases at the beginning of lactation, likely to either counteract or cause the increased calcium and phosphorous levels.
This increased bone turnover results in losses of BMD in animal models and humans. In humans, most of the mineral loss is in the trabecular bone, where bone density falls 3%–10%, with smaller losses in the cortical bone. This is compared to women of reproductive age with low estrogen levels who have 1%–4% losses in trabecular bone density and none in cortical bone [ 190 , 191 ]. While estrogen likely contributes to the decreased bone density, it is possible that PTH-rP is also contributing to the accelerated loss of BMD. Similarly, to humans with low estrogen levels, rats with low estrogen levels did not experience as much bone loss as the pregnant rats [ 183 ].
During lactation, but not directly after birth, the maximum load on lumbar vertebrae and the vertebral flexural rigidity decreases 64% and 56% respectively, in rats. This is followed by a recovery at about 8 weeks after weaning [ 183 ]. In contrast, the maximum load of the femur was not different at birth but was 26% lower at weaning. This could be due to the loss of the weight from the birth of the offspring, causing bone resorption. At 8 weeks recovery after weaning, the max load on the femur had increased compared to at weaning but was still 16% lower than controls. However, geometrically, the average cross-sectional area of the cortical bone decreased during pregnancy and did not recover at 8 weeks after weaning. Overall bone strength was compromised during both pregnancy and lactation but showed recovery after weaning. This suggests that the material properties were compromised during this period and were recovered, while the structural properties of bone did not recover. It is important to note the integrity of both the bone material and structure are important for maintaining overall bone health and strength. If either one remains deficient after pregnancy, there could be long-term bone health risks.
Despite the potential for long term bone health risks after pregnancy, there is some evidence on the contrary. One study found pregnancy reduced risk of hip and vertebral fractures in post-menopausal women, with each additional pregnancy reducing the risk of hip fracture by 9%. The women who had children had the same BMD as women who did not have children, making the diminished risk likely due to geometrical changes in the bone structure [ 192 ].
Existing studies show holes in knowledge focusing on material properties of tissue and how they change throughout and after pregnancy. There are also gaps with regards to how hormones and specific pathways affect the bone tissue during and after pregnancy. While there is a better understanding of the short-term effects on both material and mechanical properties and fracture risk than on the long-term effects, there is a lack of understanding for the effects of one singleton pregnancy compared to multiparity pregnancies or multiple pregnancies. One major challenge for addressing these knowledge gaps is the lack of tools to perform robust longitudinal assessments of bone health beyond measures of bone density, especially without exposing the subject to significant doses of ionizing radiation, which is known to be detrimental to bone health.
Recent developments in peripheral quantitative computed tomography (pQCT) systems are yielding higher image resolution, while lowering radiation doses, and are becoming more common for clinical/research use in humans. However, how these lower dose radiation exposures affect both short- and long-term bone health in humans is still not well understood. There have also been developments for using high resolution MRI to quantify bone microarchitecture, however, there are still some limitations for image resolution at the available magnetic field strengths and the length of time for acquiring these scans.
The microstructural, geometric, and material property changes all need to be evaluated during specific periods of an individual’s life, particularly during the reproductive cycle (i.e. menstruation, pregnancy, lactation, menopause), where there are significant hormonal changes. An understanding of how the hormonal changes affect the components that contribute to overall bone strength can be used to evaluate injury risks as well as novel biomaterials for bone in women.
Guest
Kaitlin Fogg 1
and Michelle L Oyen 2
1 School of Chemical, Biological and Environmental Engineering, Oregon State University, United States of America
2 Center for Women’s Health Engineering and Department of Biomedical Engineering, Washington University in St. Louis, St. Louis, MO, United States of America
Women’s health has been a historically underserved area within medicine. Recent analysis demonstrated that inventions associated with women’s health are disproportionately designed by female inventors [ 1 ]. This reveals twin problems in the development pipeline, the lack of female engineers and inventors, and a lack of basic science knowledge in the field. Recognition is growing that there are unique aspects of women’s health that have been understudied in the medical research community, also likely related to the underrepresentation of women within the STEM fields. Recent analysis has shown that funding for research into women’s health lags that of diseases that are gender neutral or that solely or disproportionately affect men [ 2 ].
The field of Biomaterials—including sub-fields such as drug delivery, tissue engineering (TE), biomimetic model materials, and biofabrication techniques—has seen rapid growth in applications in women’s health in the last decade. Recent special journal issues [ 3 – 6 ] and review articles [ 7 – 11 ] have highlighted these growing applications and have all been published within the last 3 years. This demonstrates the exciting growth of a new emerging field of scientific and engineering endeavors with the potential to make dramatic basic science discoveries that translate to clinical practice in the service of improving women’s health. This Roadmap article covers a sample of overviews on how biomaterials are being used in Women’s Health research. The articles are grouped into four topics: breast cancer and breast reconstruction; non-pregnant reproductive organs; pregnancy; and sex differences in non-reproductive organs and systems.
The first section contains two articles on biomaterials related to the breast. Tseng and Payton review developments leading to biofidelic in vitro breast cancer models based on naturally derived and synthetic biomaterials. They discuss how these systems model the tumor microenvironment (TME) and consider the effects of matrix stiffness on cell behavior. They further consider recent advances in the use of microfluidic devices and three-dimensional (3D) bioprinting techniques. Next, Holeman, McLoughlin and Fisher consider biomaterials in reconstruction of the nipple-areolar complex (NAC) following total mastectomies. They note that long-term satisfaction rates following breast reconstruction surgery (BRS) are reasonably low, and they consider TE alternatives to existing approaches. Their section highlights the importance of emotional well-being in patients following the trauma of major cancer surgery. Biomaterials approaches thus have potential for saving lives through fundamental cancer research and for improving saved lives by improving body image and self-confidence of post-op patients.
The second section of this Roadmap is comprised of three articles describing how biomaterials can be used to either develop in vitro tissue models of gynecological tissue or designed as clinical interventions for gynecological diseases. Sutton and Shikanov provide an overview of how natural and synthetic biomaterials can support follicle development in vitro and how current 3D culture systems have improved our understanding of folliculogenesis. They then outline the current engineering challenges that must be overcome to develop an artificial ovary, which could restore fertility in survivors of pediatric cancers with premature ovarian insufficiency (POI). Next, a review by Gnecco details the biomaterials and in vitro culture systems that have been used to model the endometrium. They highlight the pros and cons of each approach and conclude by summarizing future limitations that must be addressed to continue to advance the field of reproductive health. The third article by Knight is a call to action for interdisciplinary research to address pelvic organ prolapse (POP), a condition currently affecting one in eight women. The previous synthetic meshes that were used had unacceptably high complication rates, leaving patients with POP with limited options. Thus, this article describes how biomaterial based clinical interventions could either promote tissue regeneration or provide mechanical support to pelvic organs.
The third section of this Roadmap considers pregnancy and continues the theme of examining both in vitro research and possible future in vivo developments. Pregnancy is complicated by the co-existence of two (or more) genetically distinct individuals within the maternal reproductive tract. The maternal uterus interfaces directly with the fetal placenta and membranes; the fetus itself has no direct contact with the maternal tissues. There are ethical limits within the field of pregnancy research that have led to significant interest in the development of in vitro tools for studying the complexities of the maternal–fetal interface. In the first of two sections about placenta, Slaby and Weaver consider the need for biomimetic 3D materials systems that recapitulate placental architecture. They note that mechanistic studies of implantation and placental development would be enabled both for normal and pathological organ architectures. Continuing with the placenta, Hashemi considers placenta-on-a-chip microfluidic studies, which have permitted in vitro investigations of the placenta’s barrier function. These chip-based devices have provided opportunities to consider pharmacokinetics, an important consideration in pregnancy since drugs taken by the mother may be transmitted to the developing fetus. Returning the focus to maternal tissues, Zhang and House consider both in vitro and in vivo applications of novel silk-based biomaterials technologies to address the clinical problem of maternal cervical insufficiency in pregnancy. Current clinical practice involves insertion of a suture to keep the cervix closed until term birth; this practice has remained largely unchanged since the 1950s. The authors note that TE an in vitro cervix using a cell-seeded scaffold allows for study of the mechanisms of softening, while injectable hydrogel biomaterials present a novel and minimally invasive treatment paradigm for cervical insufficiency.
The last section of this Roadmap details how biomaterials can be used to deepen our understanding of how biological sex differences affect non-gynecological tissues. Sex as a biological variable has been historically excluded from in vitro studies, animal studies, and clinical trials, resulting in a knowledge gap regarding how tissue properties change with regards to hormones and how specific treatments must be tailored to account for hormonal changes. This is still a relatively new field, as it was not until 2016 that the National Institute of Health began requiring sex as a biological variable to be included in study designs. Vogt and Aguado detail how including sex as a biological variable has brough to light sex dimorphisms in cardiovascular disease. They outline how biomaterials can play a unique role in characterizing sex-specific mechanisms, either by using 3D cell culture platforms to study mechanisms in vitro or designing biomaterial implants that take sex-specific tissue properties into account. Next, Bradford and Robinson provide a broad overview of how the field of TE has begun to account for sex as a variable and outline the current and future challenges. Specifically, they describe how biomaterial-based tissue engineered models of disease using sex-specific cells can fill the knowledge gap of how sex hormone signaling drives disease, tissue repair, and cell–material interactions. Third, Thomas and Lau review how hormone fluctuation during pregnancy and lactation directly bone mechanical properties. They point out existing gaps in knowledge and provide a framework for how understanding the effects of hormones on tissue material properties can be used to identify novel treatments for bone repair.
Overall, these articles detail how interdisciplinary research combining basic science, engineering, biomaterials, and clinical research can have profound effects to improve the lives of patients that are currently drastically underserved. These article chapters represent a sample of a quickly evolving field and are not intended to provide an exhaustive review. However, they highlight different scientific and engineering approaches, different applications—both in vitro and in vivo —and illustrate the potential for further development of the research area. We hope that this Roadmap motivates researchers to engage with this important and understudied specialism as it emerges. Further, we hope this encourages funders and entrepreneurs to consider the possibilities in translating research into products designed to improve women’s lives worldwide.
Precision
Brandon J Vogt 1,2 and Brian A Aguado 1,2
1 Department of Bioengineering, University of California San Diego, La Jolla, CA, United States of America
2 Sanford Consortium for Regenerative Medicine, La Jolla, CA, United States of America
Cardiovascular disease remains the leading cause of death in the United States in both men and women [ 135 ]. Across a variety of cardiovascular diseases, men and women show sex dimorphisms in disease progression, morphology, response to treatments, and outcomes. For example, adult women with ST-segment elevation myocardial infarcts are nearly twice as likely as men to die within 30 d of receiving in-hospital care [ 136 ]. A historical failure to account for sex dimorphisms in the majority of biomedical research has likely contributed to the observed sex differences and inequities in cardiovascular disease treatment outcomes. Women have been significantly underrepresented in cardiovascular disease clinical trials, leading to results that are biased towards the clinical outcomes of men [ 137 ]. Additionally, failing to identify the sex of the cells and/or animals used in most research studies has been detrimental to understanding female pathophysiology, since most research studies use male cells and animal subjects [ 138 ]. To change course, the National Institute of Health began requiring sex as a biological variable to be considered in pre-clinical studies in 2016, with calls to integrate sex and gender more effectively throughout biomedical research and device design [ 139 ].
Biomedical devices have also been traditionally designed and engineered using male-centric approaches [ 140 ], which underscores the need for engineers to join the movement to include sex as a biological variable in research and device development. For instance, biomaterials serve as enabling tools for in vitro and in vivo disease modeling, medical devices, and TE. However, biomaterials research still lags in accounting for sex dimorphisms at all length scales [ 141 ], often leading to male-biased responses to biomaterial implants. For example, women undergoing transcatheter aortic valve replacement have a 13% higher standardized in-hospital mortality rate relative to men [ 142 ]. Sex differences in valve replacement outcomes may be due in part to differences in tissue biomechanical properties, as male carotid arteries have been shown to have a higher failure strain than female arteries [ 143 ]. Given the significant sex dimorphisms observed in cardiovascular diseases and recent advances in biomaterial design, biomaterial engineers have a unique opportunity to characterize sex-specific mechanisms of cardiovascular disease progression, potentially leading to more targeted, sex-specific treatments. Ongoing areas of biomaterials research include (a) investigating intracellular sex differences using in vitro cell culture platforms and (b) engineering biomaterial implants that interact appropriately with the host by accounting for sex-specific tissue heterogeneities and immune responses.
To create improved in vitro models, precision biomaterials [ 144 ] may be used to probe intracellular sex differences, including hormonal, chromosomal, and transcriptomic differences ( figure 14 ). Hormone effects on cardiovascular disease have been well documented, with studies finding that testosterone loss in aging men and estrogen loss in women during menopause lead to increased risk of coronary artery disease and myocardial infarction [ 145 ]. Hormone effects have also been explored within the transgender community, with studies showing that transgender women receiving estrogen hormone therapy were at an increased risk for ischemic heart disease, whereas transgender men receiving testosterone hormone therapy experienced elevated hypertension risk [ 146 ]. Outside of hormone biology, the genetic and epigenetic effects of sex chromosomes on cellular phenotypes must also be considered. For example, men with rare cases of Y polysomy (two Y chromosomes) have increased cardiovascular mortality, while women with X monosomy (one X chromosome) have higher incidence of ischemic heart disease, suggesting the X and Y chromosomes play a role in cardiovascular disease mechanisms [ 147 ]. Significant sex differences have also been observed in the cardiac transcriptome, with previous studies finding that male cardiac myocytes have increased Rho/Rho-associated protein kinase signaling activity relative to female myocytes [ 148 , 149 ]. Open questions remain regarding how genetic and epigenetic modifiers expressed on sex chromosomes impact how cells in culture respond to engineered microenvironments.
In addition to cell culture, precision biomaterials can also be designed as implantable devices to explore sex dimorphisms at the organism scale. For example, female human hearts have a significantly higher proportion of ventricular cardiomyocytes relative to other cells than male hearts, which increases heart contractility and stroke volume in women [ 150 , 151 ]. As such, implantable engineered cardiac tissues may need to be designed to reflect heterogeneous populations of cardiac cells. Another biomaterial design challenge is accounting for sex differences in the composition of the cardiac ECM. Exploring microscopic sex differences in the ECM could offer insights into observed macroscopic sex differences in heart anatomy, such as increased left ventricle diameter and posterior wall thickness in men [ 152 ]. For example, female mice experience significant increases in lysyl oxidase, an enzyme involved in collagen fibril cross-linking, offering a possible mechanism for the increased ventricular stiffness observed in women [ 153 ]. Beyond sex differences in cellular and ECM composition, additional studies focused on how male and female immune systems respond to biomaterial implants or particle-based therapeutics are needed [ 154 ].
Simple changes can be made to improve sex-specific cardiovascular research, beginning with quality control practices such as reporting the sex of all cells, serum, and animal models used in future studies. Additionally, culturing cells in phenol red free media with charcoal stripped serum will help reduce the confounding effects of estrogens and androgens in cell culture studies. Cell culture may be further improved by optimizing sex-specific media formulations containing physiologically relevant hormone levels.
A diverse set of in vitro and in vivo tools are being developed to more accurately mimic components of cardiovascular tissues and may be leveraged to reveal sex differences in cellular behavior and tissue/organ function. For example, models of matured human cardiac tissues generated from pluripotent stem cells and fibrin gels can be used to model sex-specific responses to drugs, hormones, or other small molecules [ 155 ]. Furthermore, biomaterials with tunable stiffness show promise for exploring the synergistic effects of mechanical (e.g. tissue stiffness) and biochemical cues (e.g. inflammatory factors) on gene expression. For instance, valvular interstitial cells have been cultured on hydrogels in the presence of serum from aortic valve stenosis patients to recapitulate sex-specific phenotypes in valve disease patients, which is partially regulated by genes that escape X chromosome inactivation [ 156 , 157 ]. Moreover, computational models have been developed to predict patient-specific gene expression and drug responses in valve myofibroblasts on hydrogels [ 158 ]. Additionally, tissue engineered vascular grafts are being used as implantable scaffolds to study sex differences in cellularity, inflammatory response, and biomaterial degradation [ 159 ]. Taking a different approach, other researchers have turned to using decellularized heart valves to preserve the sex-specific tissue microenvironment and explore sex differences in inflammatory responses to the ECM after implantation [ 160 ]. As biomaterials are engineered as implantable devices, an improved understanding of how material degradation impacts sex-specific, systemic inflammatory responses will be needed.
Given that sex dimorphisms exist in nearly every aspect of the cardiovascular system in health and disease, in vitro and in vivo models that seek to understand mechanisms behind cardiovascular disease must no longer ignore sex as a biological variable. Models that recapitulate sex-specific differences apparent in human disease will become increasingly important to identify mechanisms of disease more accurately. Moving forward, precision biomaterials will serve as important tools to engineer more accurate models and help identify the sex-specific mechanisms that drive cardiovascular disease progression. As these sex-specific mechanisms are better characterized, treatments will be developed that are better targeted to the individual, with the goal of achieving equity in treatment outcomes irrespective of sex.
B A A acknowledges funding from the NIH (R00 HL148542) and the Burroughs Wellcome Fund Postdoctoral Enrichment Program.
Biomedical
Yali Zhang 1
and Michael D House 1,2
1 Mother Infant Research Institute, Tufts Medical Center, Boston, MA, United States of America
2 Department of Obstetrics and Gynecology, Division of Maternal Fetal Medicine, Tufts Medical Center, United States of America
Cervical insufficiency is a well-known complication of pregnancy that leads to preterm birth. The cervix forms the lower portion of the uterus ( figure 13(A) ). Normally, the cervix remains closed as the fetus grows. At term, the cervix dilates under the influence of uterine contractions. Cervical insufficiency describes a clinical condition in which the cervix dilates prematurely. In cervical insufficiency, the cervix dilates as the fetus grows, which can lead to a pre-viable or a peri-viable birth. Deliveries prior to viability are a miscarriage. Infants born in the peri-viable period can suffer long-term health consequences such as chronic lung disease, blindness and cerebral palsy.
The static function of the cervix is critically important for a full-term pregnancy [ 117 ]. During fetal growth, multiple static stresses act to dilate the cervix. Static stresses arise from uterine distension, the hydrostatic pressure of the amniotic cavity, and anatomic variations of the pelvic structures. Countering these static stresses are the strength of the fibrous connective tissue of the cervical stroma, and the adhesion of the fetal membranes on the cervix [ 118 ]. When static stresses exceed cervical strength, the cervix shortens and dilates, which can lead to cervical insufficiency.
The current standard of care treatment for cervical insufficiency is cerclage suture ( figure 13(B) ) [ 119 ]. Under regional anesthesia, a non-absorbable suture is placed around the cervix, with 4–6 passes of the needle through cervical tissue. After traveling around the cervix in a purse string fashion, a knot is tied. A cerclage suture is placed in the first or second trimester of pregnancy, and it is removed near the due date. The most common type of suture used for cerclage is polyester braided tape (Mersilene, Ethicon RS23). The rationale of cerclage suture is to support the cervix and prevent preterm dilation. The cerclage is removed when there are signs of labor or at 37 weeks of gestational age.
This section will discuss our recent efforts to use TE to study the connective tissue of the cervical stroma and to explore a novel injectable hydrogel to treat cervical insufficiency.
The cervical stroma is the load bearing tissue of the cervix and impaired load bearing properties could lead to cervical insufficiency. The cervical stroma is primarily composed of collagen (dry weight 80%–90%) in humans [ 120 ]. During pregnancy, the stroma softens significantly in preparation for childbirth. Mechanical testing of human cervical tissue in vitro demonstrated peak stresses in compression are an order of magnitude less for cervical tissue from pregnancy compared with non-pregnant tissue [ 121 ]. Aspiration testing of human cervical tissue during pregnancy confirmed significant softening as pregnancy advances [ 122 ]. Although cervical softening is readily measured both in vitro and in vivo , the biochemical mechanisms that cause cervical softening are incompletely understood.
Animal models have shed light on how changes in cervical ECM lead to cervical softening [ 123 , 124 ]. Cervical softening is presumed to arise from changes in the fibrous collagen of the cervix. Studies of pregnant mice demonstrated significant decreases in collagen cross-link density during pregnancy, which correlates with increased collagen solubility and decreased collagen organization on transmission electron microscopy [ 125 , 126 ]. Changes in collagen cross-links, increased collagen solubility and decreased collagen organization is also seen in biopsy studies of the human cervix [ 120 , 127 ].
Model systems for investigating cervical softening have important limitations. Studies of the cervix in animals may not reflect cervical changes in humans [ 124 ]. Studies of the human cervix in vivo are limited by the difficulty of obtaining biopsy samples. Challenges of traditional model systems have motivated our efforts to develop an engineered model of human cervical tissue (discussed below).
In addition to studying engineered cervical tissue, we are investigating an alternate treatment for cervical insufficiency. The cerclage procedure was originally described by Drs Shirodkar and McDonald in the 1950s and has remained unchanged since that time. There have been few attempts to reconsider treatment for cervical insufficiency with the goal of improving efficacy and decreasing adverse effects. In terms of efficacy, the largest randomized trial of cerclage placed in the first trimester ( n = 1292 patients) showed a small benefit of cerclage. Delivery prior to 33 weeks occurred less frequently in the cerclage group (13%) compared with the control group (17%, p = 0.03) [ 128 ]. For second trimester patients with a short cervix, cerclage placement showed clinical benefit but was still associated with a 15% risk of delivery prior to 28 weeks [ 129 ]. In addition, adverse effects can be seen with cerclage suture such as cervical laceration, which occurs in 4.8%–7.9% of cases [ 130 , 131 ].
A significant challenge of improving cerclage treatment is that the biomechanical environment of the cervix during pregnancy is understudied. Only in recent years have groups attempted to work out the 3D stress state of the cervix during pregnancy [ 118 , 132 ]. An improved understanding of cervical biomechanics is critically needed to understand the mechanism of cerclage efficacy and rationally design improvements for cervical insufficiency.
To address the limitations of current models of cervical softening during pregnancy, engineered models of cervical tissue are being investigated [ 133 ]. During pregnancy, it is known that cervical tissue is exposed to multiple factors that could decrease cervical stiffness including changes in hormone concentrations, inflammatory conditions, and mechanical loading. To study cervical tissue without the complexity of the environment in vivo , an engineered cervix was developed. Human cervical FBs were isolated from explants of non-pregnant cervical tissue. FBs were seeded on porous silk scaffolds and cultured in spinner flasks. After 4 weeks, the seeded scaffolds produced an ECM with biochemical components and histological appearance similar to native cervical stroma. Using the engineered cervix model, the influence of steroid hormones on ECM properties was studied [ 133 ]. The engineered cervix model suggested estradiol promoted ECM growth and progesterone had a softening effect. We expect the 3D engineered cervix model will permit studies not possible with animal models or human biopsy studies.
To study a potential alternative to cerclage surgery, injectable hydrogels for cervical augmentation are being investigated ( figures 13(C) and (D) ) [ 134 ]. Cerclage requires regional anesthesia for placement. Also, cerclage removal can be challenging, especially if labor is present. An injectable treatment is a non-surgical approach that avoids surgical risks and avoids the need to remove the cerclage. An injectable hydrogel creates a composite tissue which could have improved properties (e.g. increased stiffness) for better function. Also, in contrast to a cerclage, injectable hydrogels increase cervical volume, which could improve barrier properties between the vaginal and uterine environments. In a recent study, purified silk fibroin protein was crosslinked with horseradish peroxidase and hydrogen peroxide to create an elastic hydrogel with mechanical properties similar to human cervical tissue [ 134 ]. In a rabbit model of pregnancy, the silk-based hydrogel was injected into the cervix. The hydrogel demonstrated cervical augmentation, biocompatibility, and biodegradation. Compared to controls, no adverse effect was seen in terms of kit delivery and kit viability. Future studies will compare an injectable treatment to cerclage in terms of improved barrier properties and mechanical support.
Biomedical engineering offers rigorous and innovative strategies to study cervical function in pregnancy. Patients with cervical insufficiency suffer structural failure of the cervix. Engineers have the tools and training to assess structural failure and improve structural support. In addition, TE strategies are yielding new insights into cell–matrix interactions of cervical tissue. These studies show the considerable potential for using a biomedical engineering approach for cervical insufficiency in pregnancy.
We gratefully acknowledge support from the following NIH Grants 5R01EB021264, 5P41EB027062 and 5K12HD000849. We also acknowledge support from the Bridge Funding Award from the Society for Maternal Fetal Medicine.
Biomimetic
Emily M Slaby and Jessica D Weaver
School of Biological and Health Systems Engineering, Arizona State University, AZ, United States of America
The placenta is a dynamic, complex, and temporary organ that develops rapidly throughout pregnancy ( figure 11(A) ), continuously adapting to provide nutrients to the developing fetus and protect it from immunological attack. Placenta research is vital to understanding pregnancy complications and the mechanisms underlying successful pregnancies. The human placenta is challenging to study pre-parturition due to potential risks to the fetus, and substantial differences in placental physiology and development between species make animal models of limited use to investigate human placental dynamics. The vast majority of in vitro research with placental cells (e.g. cell lines or primary placenta cell isolates) or villous explants takes place in 2D culture systems, which fail to replicate the complexity and architecture of a multicellular, 3D tissue ( figure 11(B) ). Next-generation in vitro culture systems that replicate the placental microenvironment will be pivotal to advancing our knowledge of human placental physiology and immunology.
Biomaterials have been used for decades to generate artificial 3D cellular environments in vitro for a wide variety of tissues, though they have only permeated placenta research within the last decade. Biomimetic materials, biomaterials engineered to mimic a tissue microenvironment, can be designed to exploit and manipulate material properties such as mechanical stiffness [ 73 , 96 ], ECM signaling [ 97 – 101 ], and soluble or matrix-tethered signaling gradients [ 102 , 103 ] ( figure 11(C) ). Several 2D in vitro placenta culture systems have used surface modification to generate biomimetic synthetic materials using substrates such as polyacrylamide [ 97 , 101 ], polydimethylsiloxane [ 98 , 102 ], and polycarbonate [ 99 , 100 ], coated with ECM components such as collagen, fibronectin, and decellularized human placenta. While these 2D systems can approximate late gestation placental transport in on-a-chip platforms ( figure 11(A-iii) ), they are poor models for early gestation transport, placenta remodeling and development, and immunomodulation ( figure 11(A-i) ); therefore, 3D biomimetic culture platforms are necessary to fully replicate the placenta microenvironment through all stages of gestation.
Within the past decade, commercially available 3D culture platforms such as Matrigel [ 104 ] and Geltrex [ 105 ] have been used to generate biomimetic culture environments to study placenta formation and cell–cell interactions. Recently, researchers have used synthetic biomimetic 3D culture systems such as GelMA hydrogels [ 73 , 103 , 106 , 107 ] to study invasion and cell signaling in response to altered matrix stiffness. These recent advancements in 3D in vitro placenta culture systems demonstrate promise in generating physiologically relevant biomimetic materials with a high potential for impact in placenta research.
With a dearth of healthy placental cell and villous explant samples available from early and mid-gestation pregnancies, the development of 3D biomimetic in vitro culture models is critical to advancing early gestational stage human placenta research. Non-primary tissue cell sources for these models include trophoblast cell lines, established trophoblast progenitor stem cell lines, and induced pluripotent stem cell-derived trophoblasts. The vast majority of research on these cells has been conducted on 2D surfaces [ 97 , 101 , 102 ] or in commercially available 3D matrices [ 104 , 105 ] like Matrigel, which are poor replicas of the native placenta microenvironment. Matrigel, which is derived from decellularized murine tumors, provides a rich environment of ECM and soluble factors to encapsulated cells; however, the substantial batch-to-batch variability in these platforms leads to poor experiment reproducibility, which is a common challenge with naturally-derived materials. While 2D systems like placenta-on-a-chip [ 98 – 100 ] models are effective for studying the transport of drugs and molecules across the placenta, these systems primarily replicate late gestation transport and fail to reproduce the complexity of early and mid-gestation placenta architecture ( figure 11(A) ).
Recapitulation of the multi-cellular placenta architecture in vitro has remained elusive, and organoid generating techniques often generate an inverted structure. Future challenges in human placenta research necessitate the development and application of biomaterial systems with precise control over biomimetic signaling that mimic the dynamic changes and signaling of the fetal–maternal interface ( figure 11(C) ). Through the course of gestation, the placenta exhibits gradients of ECM composition, nutrients, metabolites, and hormones that drive the development and infiltration of the maternal decidua. Simulating these gradients in synthetic culture systems will be critical to accurately model the placenta. Additionally, biomimetic systems will require control over matrix stiffness and porosity and should facilitate cell-responsive remodeling of the matrix. Finally, the chemistry used to fabricate such a matrix must be minimally cytotoxic to avoid altering the encapsulated cell phenotype. Surmounting these simultaneous challenges would yield effective biomimetic material platforms for placenta research.
To advance human placenta research, rationally designed and dynamic biomimetic material platforms are necessary to replicate the early and mid-gestational placenta microenvironment. Current placenta-on-a-chip designs [ 98 – 100 ] have generated increasingly complex models of late gestation placenta with biomimetic materials and multiple cell types; however, in vitro models of early- to mid-gestation placenta will need to exploit 3D biomimetic platforms to replicate the complex multi-cellular placenta architecture. Future biomimetic material platforms must recapitulate placenta mechanical stiffness, ECM composition, dynamic remodeling, and soluble and immobilized signaling gradients to approximate normal and disordered placenta microenvironment states. For example, modifying ECM composition and matrix stiffness could enable the mechanistic study of disorders of placental development and implantation [ 97 , 101 ], such as placenta accreta, increta, and percreta. Similarly, modulation of local nutrients and soluble and immobilized signaling molecules can facilitate studies of placental mechanisms of immune tolerance and associated disorders, such as preeclampsia.
Rapid biomaterial advancements over the past few decades have resulted in a plethora of biomimetic material platforms. Naturally-derived matrices such as decellularized ECM-based hydrogels can be used to provide matrix signaling to embedded cells, though mechanical stiffness is challenging to control in these matrices. Conversely, synthetic polymer-based hydrogels offer a high degree of control over matrix mechanical properties, but they must be modified to imbue biomimetic signaling. Additionally, synthetic hydrogels have less batch-to-batch variability than natural matrices, which could result in more reproducible experimental outcomes. The available range of biomimetic material platforms continues to grow, with an increasing capacity to control spatiotemporal factors within matrices. Expanded libraries of cytocompatible, biomimetic platforms will be necessary to accelerate studies of normal and disordered placenta development in the next decade.
Biomimetic materials are a powerful emerging tool in the field of placenta research, enabling the generation of 3D culture systems that better recapitulate the native placenta microenvironment. Advanced biomaterial-based in vitro culture systems may enable the study of human placenta development, physiology, and immunology in an unprecedented manner, and expanding this area of research could have broad impacts on maternal mortality, diseases of pregnancy, and understanding of immunological tolerance mechanisms.
The authors are supported by the Juvenile Diabetes Research Foundation Innovator Award (1-INO-2020–915A-N), the Arizona Biomedical Research Centre New Investigator Award, and the National Institutes of Health Director’s New Innovator Award (DP2AI169476).
Biomaterials
Nicole N Hashemi
Department of Mechanical Engineering, Iowa State University, Ames, IA, United States of America
The human placenta is a critical and complex organ that performs various functions. These features include several of the following: safe circulation and flow of fetal and maternal blood, nutrients, and oxygen; elimination of waste and carbon dioxide; and protection from diseases, infections, and xenobiotics from mother to fetus. Performing these actions supports fetal development and maturation [ 108 – 110 ]. The placenta is composed of the syncytiotrophoblasts and the fetal capillary endothelial layer. The placental barrier separates the fetal circulation from the maternal blood so that the two bloodstreams do not mix [ 99 ]. Placental examination and monitoring are complicated given the considerable changes in placental structure during pregnancy [ 111 ]. Needless to say, as pregnancy progresses, the mother’s placental barrier changes more dramatically, making the monitoring process of the fetus and placenta itself even more difficult [ 111 ]. Although placenta research and monitoring are complex, the knowledge already gained from the placenta has proven to be an essential source of information to understand the overall performance of the human placenta. Therefore, placental physiological function and structure play an important role in investigating, monitoring, and understanding the characteristics of the human placental barrier. Understanding placenta physiology lays the foundation for researchers to assess the human placenta and create models for the scientists and pharmaceutical industries to understand it better. However, conducting such studies is time-consuming and could pose a risk to the fetus in in-vivo. Ex-vivo studies are limited regarding the reliable assessment of transport mechanisms in the human body. Placenta-on-a-chip models provide opportunities to test and understand the effects of medicines on the placenta while eliminating the risk of fetal and maternal rupture and destruction [ 99 , 112 , 113 ].
The placenta-on-chip has made significant strides in testing drugs across microfabricated placental barriers, but this model has only recently been developed compared to in-vivo and ex-vivo models. One of the main challenges in the placenta-on-a-chip models is to make sure that the models can perform under a few efficient drug intake tests; this is partly owing to the fact that researchers have devoted most of their time in recent years to modifying the procedure for producing placental chips. Scientists have devoted a significant amount of work to improve the fabrication and operation of the placenta-on-a-chip; yet, researchers are unable to replicate the crucial physiological properties of the placental barrier. For instance, placental chips lack assays for lipids and amino acid transport through the placental barrier, which are crucial qualities for sustaining chip performance and placenta responsiveness and replicating the human placenta’s physiology [ 99 ]. In contrast to in-vivo and ex-vivo models, the in-vitro technique has been tested more thoroughly throughout model development than earlier models. There are concerns regarding the expertise of pharmaceutical study models as well as the capability of the chip to undertake complex tests and monitor polydimethylsiloxane chips. For instance, as the functionality and complexity of the 3D co-cultured cell model rises, microarray analysis for high-resolution tissue screening and visualization of tissue location becomes more challenging to conduct [ 114 ]. In the biological response assay, the limited number of cells on the chip compared to the in vivo placenta impedes the detection sensitivity of drug uptake and cellular interactions. As a result, the presentation and assessment of the chip’s clinical results make it difficult for pharmacists to communicate data to patients [ 115 ]. Moreover, it has been reported that due to the size of cell culture chambers and channels, the shear stress developed is significantly lower than the shear stress developed in fetal capillaries [ 108 ]. In addition, further studies are required to assess the effect of shear stress on the formation of syncytiotrophoblasts since It is possible for trophoblasts to detect flow shear stress via mechanotransduction; blood flow shear stress rates through invading uterine spirals range from 1 to 10 dyn cm −2 ; however, the shear stress lowers to less than 0.1 dyn cm −2 in the intervillous area, according to studies [ 116 ]. The cells subjected to shear stress after differentiation into trophoblast-like cells showed proof of the formation of syncytium compared to those not subjected to shear stress when trophectoderm-like cells from human naive induced pluripotent stem cells were placed under a shear stress of 10 dyn cm −2 . Consequently, shear stress has been found to be essential for the syncytialization of trophoblasts [ 116 ]. Additionally, multiple cell lines of the human placenta should be used to better mimic the physiology and anatomy of the human placenta developed for placenta-on-a-chip studies [ 108 ]. Figure 12 shows an in vivo -like microsystem with HUVECs representing the endothelium, trophoblasts (BeWo) cells representing the epithelium, and a semipermeable membrane representing the placental barrier.
As in-vitro approaches continue to be developed, the capability to perform a variety of investigation on micro-engineered barriers for pharmaceutical trials are realized. However, more disease-related studies are required to advance the mimicking of the placental barriers. Also, by using these in-vitro models, scientists may gain new insights into the developmental stages of foetation. For example, one will gain knowledge of placental development and how the transport of compounds affects fetal growth. However, the applicability of the universal model has been thoroughly discussed, and outlined that the model must simulate the physiological properties of the human placenta in-vivo while ensuring that procedures and methods are performed in the same way for all. Developing a working model for conducting drug intake studies is vital for a better assessment of the influences of drugs on the maternal–fetal barriers. Moreover, including genetic studies to assess the placental expression of genes could help researchers in examining their models to achieve a model which is closer to the actual in-vivo studies [ 116 ]. Creating truly representative placenta models that accurately mimic the human placenta’s significant changes during different pregnancy stages remains a challenge. This has inspired the scientific community to exploit technological advances to overcome these obstacles. In addition, maintaining the confluency and functionality of trophoblasts for a prolonged time are critical challenges to address. The continued funding and support for the placenta-on-a-chip models will aid scientists in advancing the device’s overall functionality. Scientists will be able to conduct various experiments, other than pharmaceutical and drug testing, across the micro-engineered barrier in the future as researchers continue to innovate on the in vitro technique.
The application and further development of placenta-on-a-chip for pharmacokinetic/pharmacodynamic studies of the human placenta will aid the reduction of drug-induced congenital disabilities and a better understanding of the effects of these drugs on the human placenta. The significance of the placental barrier during pregnancy has been investigated for decades, but the newly emerged techniques have enabled new in-vitro technology that accurately mimics the anatomical structure and physiological function of the placenta and the barrier. Multiple placenta-on-a-chip [ 99 , 108 , 116 ] platforms have appeared from the imitation of both placental barrier anatomy and physiology and deepens the understanding of this vital organ system. With the advent of more advanced placenta-on-a-chip models of the placental barrier, such models are expected to substantially change how research and development are done in this field.
This work was partially supported by the Office of Naval Research Grant N000141712620 and National Science Foundation Award 2014346.
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.