Mosaic Loss of Y Chromosome in White Blood Cells: Its Impact on Men’s Health

Abstract

We present a brief introduction of loss of Y chromosome (LOY) in blood and describe the known risk factors for this condition. We then overview the associations between LOY and age-related disease traits. Finally, we discuss murine models and the potential mechanisms by which LOY contributes to disease.

Mosaic Loss of Y Chromosome in Leukocytes

Humans have X and Y sex chromosomes; females have two X chromosomes and males have single X and Y chromosomes. The X and Y are almost completely different; however, early in human evolution, the X and Y chromosomes began as a pair of autosomes that began to separate in identity ∼180 million years ago. Over the course of millions of years, the X chromosome retained most of its ancestral genes, whereas the Y chromosome lost almost all of the genes that it once shared with the X chromosome. Recognition of these events has led to the popular notion that the Y chromosome may eventually vanish completely. However, comparisons of Y chromosomes between mammalian species have revealed that gene loss on the Y chromosome occurred very early in evolution and that the current Y chromosome has been extremely stable for at least the past 25 million years (1, 2). However, the Y chromosome does in fact vanish, but this happens in the blood cells of males as they age and it is very prevalent.

Mosaic loss of Y chromosome (LOY) in leukocytes is the phenomenon where males lose their Y chromosome in a portion of blood cells. LOY is the most common postzygotic mutation in humans, and there is a very close association between aging and LOY (3). LOY is commonly described as a form of “clonal hematopoiesis,” an age-related phenomenon that is characterized by clonal expansion of blood stem cells and their progeny that harbor one or more somatic mutations in “driver” genes that confer a competitive advantage to the cell. In contrast to conventional clonal hematopoiesis driver genes, LOY generates larger size clones, suggesting that LOY can confer a considerable fitness advantage to the cell. In addition, multiple LOY events can occur in an individual where independent blood stem cells can lose the Y chromosome in parallel. In an extreme case, an 81-yr-old male was estimated to have at least 62 independent LOY clones (4). All else being equal, aging is the strongest risk factor for LOY, and LOY is reported to be detectable in 2.5% of 40-yr-old men, 40% of 70-yr-old men, and 57% of 93-yr-old men (5, 6). In a separate study of 29 men, all individuals (ages 64 to 94) had detectable LOY cells based on single-cell analysis (7). In an individual, the size of a LOY clone also increases with age (8). These findings indicate LOY is a nearly ubiquitous condition in the elderly male population.

Although measurable LOY events have been thought to be uncommon in children and young adults, recent studies suggest that LOY may also occur in infants and as early as embryogenesis (4, 9). While LOY clones are relatively small in young men, the increase in clone size and prevalence of LOY in the population increase exponentially with age. In view of this, it can be speculated that there is a “selective pressure” that suppresses LOY clone expansion during youth, but this somatic maintenance declines with older age when reproductive success becomes less likely. Thus, it would follow that the expansion of LOY clones may be harmful to men’s health. In fact, researchers have recently come to appreciate that LOY in the blood is associated with poor survival as well as increased incidence of various disease traits in men.

Genetic Basis of LOY in Leukocytes

Measurements of LOY in the blood can be made with data derived from single nucleotide polymorphism (SNP) arrays, whole-genome sequencing, and droplet digital PCR (ddPCR). SNP array is a high-throughput, relatively cost-efficient genotyping approach that has been employed to query LOY in large biobanks. Since SNP arrays can encompass the Y chromosome, it is possible to compare the observed intensity of signals from Y chromosome SNPs to a reference intensity. The degree of the observed intensity reduction has been utilized as a proxy for the proportion of cells in the samples that lack the Y chromosome. Because the Y chromosome consists of the male-specific region of Y (MSY: located only on the Y chromosome) and the pseudoautosomal regions (PARs: small regions at the ends of the X and Y chromosomes with identical sequences) and SNPs may occur throughout both regions, there are several methods for assessing LOY depending on which portion of the Y chromosome is used (mLLR-Y and PAR-Y, respectively).

Additionally, SNP-genotyping of autosomes has been employed to elucidate potential mechanisms that regulate the occurrence and progression of LOY. Although LOY is a nonheritable somatic mutation, the genetic predisposition for LOY is heritable with an estimated total heritability of ∼32% (10). Large-scale genome-wide association studies (GWAS) have enhanced our understanding of germline contributions to LOY. The first genome-wide significant SNP signal was mapped near the TCL1A locus (3). TCL1A encodes T-cell leukemia/lymphoma protein 1A, and it functions to increase the activity of the AKT protein kinase. Single-cell transcriptomics revealed that TCL1A is specifically expressed in B cells, and men with LOY express greater amounts of TCL1A. How TCL1A variants lead to the loss of the Y chromosome is unknown. Similarly, TCL1A variants are also associated with other forms of clonal hematopoiesis that are caused by genes that are recurrently mutated in leukemia (i.e., clonal hematopoiesis with indeterminate potential or CHIP) (11), implicating TCL1A as a general regulator of clonal expansions in the hematopoietic system. Another study of 85,542 individuals identified 19 genomic regions that are associated with LOY, and many of these loci are associated with genes involved in cell cycle regulation (5). Intriguingly, these 19 loci are also associated with loss of X chromosome (LOX) in women, suggesting that these genetic variants reflect a general mechanism of aneuploidy. A larger GWAS study of 205,011 individuals, using a more sensitive detection method, discovered 156 genetic variants associated with LOY (10). Many of these variants occur in genes involved in cell cycle fidelity and the DNA-damage response, and some of which are associated with the regulation of hematopoietic stem cells. Similar results were observed in an analysis of 95,380 individuals in the Biobank Japan project, demonstrating the robustness of the identified genetic variants (12). In addition, an exploration of rare genetic variants (minimal allele frequency less than 0.5%) using exome sequencing data from 82,277 individuals resulted in the discovery of the loss of function mutations in the CHEK2 and GIFGY1 genes, which confer a higher risk for LOY (13). In aggregate, more than 100 genetic variants may be acting in hematopoietic stem cells to promote the development of LOY via aberrant cell cycle regulation and/or impaired DNA damage response. Potentially, increased proliferation with a higher rate of chromosomal missegregation may promote LOY within a cell, which may be tolerated due to an impaired DNA damage response (14). Further mechanistic studies are warranted to validate these speculations and to elucidate how these genetic variants promote the formation of LOY.

In addition to aging and genetic predisposition, smoking is a third risk factor that significantly increases the risk of developing LOY. Compared to nonsmokers, smokers are more likely to lose the Y chromosome in their blood cells in a manner dependent on the number of smoking years (odds ratio = 2.4–4.3) (3). Mendelian randomization studies suggest that the association between smoking and hematopoietic LOY may be causal (5, 15). Interestingly, the effect of smoking seems reversible. After smoking cessation, men who have ever smoked are no more likely to have LOY than those who have never smoked (3). Like smoking, acute respiratory infection may influence LOY clone growth. A recent study demonstrated that LOY is enriched in patients with severe COVID-19, and 6 months after the patients were discharged from intensive care, the proportion of cells with LOY had substantially decreased (16).

Clinical Consequences of LOY in Leukocytes

LOY was first reported more than half a century ago, and it had generally been regarded as a biomarker of aging or a clinically benign age-related alteration in the blood (17, 18). Because the function of the Y chromosome was generally believed to be exclusive to male determination (i.e., development of an anatomical male) and sperm production, it was assumed that the Y chromosome was unimportant in blood cells. Consistent with this notion, the Y chromosome only contains a limited number of genes (∼45), and most are reported to be testis specific. However, recent epidemiological studies have cast doubt on these beliefs by showing that LOY in the blood is associated with an elevated risk of all-cause mortality (6, 19, 20) and a variety of clinical conditions, including hematological cancers (21–23), solid cancers (3, 21, 22, 24–27), Alzheimer’s disease (28, 29), and cardiovascular disease (19, 20, 30) (FIGURE 1). Intriguingly, GWAS revealed that genetic variants for LOY overlapped with known cancer susceptibility variants for several nonhematological cancers (31), suggesting the association between LOY and cancers could be partially mediated through shared genetic variants between both traits (i.e., the “common soil” theory). However, this finding does not rule out the possibility that immune cells lacking the Y chromosome directly impact cancer progression.

FIGURE 1.

Causes and consequences of mosaic loss of Y (LOY) chromosome in leukocytes in men Aging, environmental factors, and genetic predisposition are the risks for acquisition of LOY in leukocytes. LOY can lead to functional change of monocytes/macrophages and may be associated with age-related pathologies including heart failure. AD, Alzheimer’s disease; HSCs, hematopoietic stem cells.

The first noncancerous disease to be associated with LOY in blood was Alzheimer’s disease (AD), a finding that has been supported in multiple studies (28, 29, 32, 33). Dumanski et al. (28) found that individuals with a higher proportion of LOY blood cells have a 2.8-fold greater likelihood of AD diagnosis than those with a lower proportion of LOY. Furthermore, Mendelian randomization analysis indicates that the association between LOY and AD might be causal (34). These findings suggest the possibility that peripheral LOY immune cells infiltrate the brain and accelerate neurodegeneration. On the other hand, a recent single-cell analysis of the brain revealed that LOY occurs in 12–30% of microglia, a tissue-resident cell type that is maintained locally throughout life (35, 36). LOY in microglia was also shown to increase with age (37). Considering microglia are not replaced by blood-derived macrophages, microglial LOY might represent an additional mechanism by which LOY accelerates AD. Potentially, locally proliferating microglia lose the Y chromosome and contribute to neurodegeneration through altered functions (37). These findings indicate that LOY in nonhematologic cells may also contribute to disease. How frequently LOY occurs in different nonhematologic cell types and the role those mutations play in disease pathogenesis are both intriguing lines of inquiry and will require further experimental research. LOY in blood has also been linked to cardiovascular (CV) disease (19, 20, 30). The first was association was made with a composite of CV events, including sudden CV death, hemorrhagic or ischemic stroke, myocardial infarction, fatal heart failure, and fatal aneurysm rupture, in patients with severe atherosclerosis after adjustment for age and smoking status (hazard ratio = 2.28; n = 366, 3-yr follow-up) (30). Intriguingly, carotid endarterectomy samples from individuals with and without LOY did not show detectable differences in histological analyses or the expression of inflammatory chemokines and cytokines, suggesting that LOY accelerates CV mortality through other mechanisms. Recently, three analyses of data compiled from the UK Biobank (n = 223,550) demonstrated that hematopoietic LOY is associated with a wide variety of CV diseases and heart failure (20). In this study it was found that increasing levels of LOY corresponded with a greater risk of CV death in an ∼11 year follow-up. More specifically, for males with LOY > 40% (meaning more than 40% of an individual’s blood cells lack the Y chromosome), the death from CV disease increased 1.31-fold. Within the broad category of CV diseases, LOY was associated with a 3.48-fold increased risk for mortality from hypertensive heart disease, 2.76-fold for aortic aneurysm and dissection, and 1.76-fold for heart failure (2.42-fold for congestive heart failure) (20). More recently, a study has uncovered an association between mLOY in leukocytes and impaired survival in patients who underwent successful transcatheter aortic valve replacement (TAVR) during a 3-yr follow-up (38). Receiver operating characteristic curve analysis revealed a 2.2-fold increase in mortality in men with LOY >40% and LOY remained an independent predictor of death after multivariate analysis.

In addition to cancer, AD and CVD, LOY is also reported to be associated with diverse diseases, including infectious disease (39), type 2 diabetes (19), obesity (19), autoimmune diseases (40, 41), age-related macular degeneration (42), suicide (32), and COVID-19 (16). Notably, single-cell transcriptome data revealed that when compared to euploid leukocytes, leukocytes without the Y chromosome had different expression patterns of not just genes located on the Y-chromosome but also genes on autosomes (7, 43). The composite evidence from numerous single-cell RNAseq studies suggests that functionally altered leukocytes, resulting from LOY, may have a direct impact on disease processes.

Modeling LOY in Leukocytes of Mice

Although epidemiological studies have shown that LOY is associated with increased risk for hematological malignancies and nonhematological diseases, these studies are unable to assess cause-and-effect relationships or define mechanistic details. Thus investigators have employed experimental models to examine the causal associations between LOY in leukocytes and age-related diseases.

Two studies have used CRISPR/Cas9 technology to model LOY in mice and explore the functional effects of this somatic mutation (20, 44). These studies differ in the gRNA target sequences in the mouse Y chromosome as well as the disease states under investigation (FIGURE 2). To ablate the Y-chromosome, Zhang et al. (44) targeted the amplified genes on the Y chromosome, such as Ssty1 (spermatogenesis-specific transcript on Y 1) and Ssty2. Since there are more than 300 copies of Ssty1 and Ssty2 on the mouse Y chromosome (FIGURE 2A), multiple cleavages induced by a single gRNA that targets these repeated genes led to the removal of the Y chromosome from the cells (44, 45). In the second approach, Sano et al. (20) and Adikusuma et al. (46) modeled LOY by targeting repetitive sequences in the centromere of the Y chromosome with the CRISPR guide RNA delivered by a lentivirus vector. In this study, the bone marrow from young male mice was ablated via lethal irradiation and replaced by transplantation of Y chromosome-depleted hematopoietic stem cells (FIGURE 2B). Assessing the peripheral blood of these mice, the CRISPR/Cas9 method successfully removed the Y chromosome from 49% to 81% of white blood cells, a level of LOY that can be observed in many humans. Moreover, greater than 95% of lentivirally transduced cells displayed a loss of the Y chromosome, representing a high efficiency of Y chromosome depletion by targeting the centromere. In this study, the authors found no upregulation of DNA damage response and repair (DDR) gene transcript levels in hematopoietic cells of LOY mice (unpublished observations), in contrast to the results in models that targeted Ssty genes (44). While both gRNAs used in these studies have been reported to have minimal off-target effects (20, 45, 46), increasing the depth of sequencing may detect rarer mutations and uncover off-target effects.

FIGURE 2.

Modeling of hematopoietic mosaic loss of X (LOX) in mice Experimental conditions to model loss of Y (LOY) in the studies of Zhang et al. (44) (A) and Sano et al. (20) (B). These two studies used different gRNAs to target the Y chromosome. Recipient mice were sublethally irradiated females in study of Zhang et al. (44) and were lethally irradiated males in study of Sano et al. (20).

Zhang et al. (44) showed that the proportion of LOY cells increased over time and that LOY accelerated the development of acute myeloid leukemia in the background of female Trp53-deficient, AML1-ETO-fusion gene-expressing cells (in effect, the model examines LOX). On the other hand, Sano et al. (20) examined the causality between LOY in leukocytes and nonhematological diseases. In this study, it was observed that male LOY mice exhibited reduced survival and displayed multimorbidity at 15 months following bone marrow transplantation. These mice displayed worse cardiac function and elevated fibrosis in the lungs, kidneys, and heart, indicating that LOY might contribute to the age-related fibrotic decline of organ function in males. The authors also found that macrophages from mice lacking the Y chromosome promote fibrosis by stimulating fibroblasts in the heart to produce more connective tissue (FIGURE 1). Notably, the LOY cardiac macrophages in mice (20), as well as LOY monocytes observed in men with impaired long-term survival after TAVR (38), displayed increased expression of transforming growth factor β-associated signaling. These findings suggest mechanistic link between monocytes/macrophages lacking the Y chromosome and cardiac fibrosis (20, 38). Intriguingly, LOY occurs more frequently in macrophages and monocytes compared to other blood cell populations (16), indicating that LOY in myeloid cells may confer these disease traits.

Although the Y chromosome lost most of its ancestral genes during evolution, a few genes have persisted on the human (and mouse) Y chromosome that are shared with the X chromosome. For this subset of genes, women have two copies encoded by each X chromosome, and men have a total of two copies, one encoded by the X and one encoded by the Y chromosome. These genes are ubiquitously expressed throughout the body and have no obvious connection to male sex determination or sperm production, suggesting that they may have functional roles in autosomal cells. Eif2s3y, Kdm5d, Uty, and Ddx3y of the Y chromosome are expressed in immune cells of mice (20), and each has a homolog on the X chromosome (Eif2s3x, Kdm5c, Utx, and Ddx3x, respectively). Among these four genes, Kdm5d and Uty genes are of particular interest as they encode epigenetic regulators. Furthermore, previous studies have suggested that the Y chromosome serves a role beyond reproductive tissues, regulating autosomal gene expression via epigenetic mechanisms (47). KDM5D is an H3K4 demethylase and is implicated in sex-biased methylation and gene expression of nonreproductive cells in humans and mice (48, 49). Zhang et al. (44) and Batdorj et al. (47) demonstrated that Kdm5d deficiency contributes to the clonal expansion of HSCs and leukemogenesis, although this may be mediated via a demethylase-independent mechanism. On the other hand, the Y chromosome gene(s) responsible for the development of nonhematological disease and the profibrotic characteristics of LOY are unknown. Thus future studies are warranted.

Conclusions

The underlying mechanism of sex biases in disease incidence or severity is commonly attributed to sex hormones, and there is little question that estrogens and androgens play significant roles in pathological processes. However, recent studies suggest that sex differences can also be attributed to X and Y chromosomal differences that are independent of sex hormones. Further, the disruption of the balance of X and Y chromosomes in men could contribute to male-specific disease predisposition, and further studies are warranted to elucidate more fully the roles of LOY in gender-biased diseases.