Highlights
- •Analyzed gene expression in sex chromosome aneuploidy samples using linear models
- •Xi and Xa transcriptomes are modular
- •38% of X chromosome genes are affected by Xi copy number—in cis and in trans
- •10 X chromosome genes likely contribute to male-female differences in somatic tissues
Summary
The “inactive” X chromosome (Xi) has been assumed to have little impact, in trans, on the “active” X (Xa). To test this, we quantified Xi and Xa gene expression in individuals with one Xa and zero to three Xis. Our linear modeling revealed modular Xi and Xa transcriptomes and significant Xi-driven expression changes for 38% (162/423) of expressed X chromosome genes. By integrating allele-specific analyses, we found that modulation of Xa transcript levels by Xi contributes to many of these Xi-driven changes (≥121 genes). By incorporating metrics of evolutionary constraint, we identified 10 X chromosome genes most likely to drive sex differences in common disease and sex chromosome aneuploidy syndromes. We conclude that human X chromosomes are regulated both in cis, through Xi-wide transcriptional attenuation, and in trans, through positive or negative modulation of individual Xa genes by Xi. The sum of these cis and trans effects differs widely among genes.
Introduction
The X chromosome of eutherian mammals exists in two distinct epigenetic states that are referred to as “active” (Xa) and “inactive” (Xi).1,2,3 The “n−1” rule (where n is the number of X chromosomes per cell) states that all diploid human somatic cells possess one X chromosome in the active state (Xa), while all other (i.e., n−1) copies of chromosome (Chr) X4 are transcriptionally repressed through a mechanism known as X chromosome inactivation (XCI). Despite the name, Xi is functionally active, making critical contributions to human fitness and viability. For example, 99% of fetuses with only one sex chromosome (45,X) abort spontaneously, suggesting that viability hinges on gene expression from a second sex chromosome—either Xi or Y.5,6 The rare survivors likely have a mixture of 45,X cells and cells with a second sex chromosome, and they display a constellation of anatomic features known as Turner syndrome.7,8
Studies have revealed that as many as a quarter of X-linked genes are expressed from Xi in humans; such genes are said to “escape” X inactivation.9 Early studies demonstrated the expression of certain Chr X genes on Xi (“escape”) in human-rodent hybrid cell lines that had retained human Xi but had lost human Xa (for example, Mohandas et al., 1980; Brown et al., 1997; and Carrel et al., 1999).10,11,12 Subsequent allele-specific methods distinguished transcripts from Xa and Xi in human cell lines that exhibited skewed XCI or in single cells.13,14,15,16,17,18 While conceptually superior to hybrid cell lines, allele-specific methods yielded sparse data because they require the presence of heterozygous single-nucleotide polymorphisms (SNPs) to differentiate between alleles. Other studies approximated the contributions of Xi to X-linked gene expression by comparing samples with varying Xi copy numbers: in some cases, between 46,XY and 46,XX samples, and in others, between sex chromosome aneuploid and euploid samples.15,19,20,21,22,23,24,25,26 These studies employed analytic methods that made it difficult to separate the effect of Xi copy number from the potentially confounding effects of correlated factors such as Chr Y copy number, hormonal differences, or tissue composition. More importantly—as underscored by this study—previous work assumed, without directly testing, the independence and additivity of Xi and Xa expression. In particular, these studies assumed that any increase in expression observed with additional copies of Xi was due to expression from Xi, which may not always be the case. Given these limitations, we hypothesized that revisiting Xi gene expression with alternative experimental and analytic methods would reveal new insights.
Here, we used a series of quantitative approaches to investigate gene expression from Xi and Xa. Inspired by previous studies, we took advantage of the natural occurrence of diverse sex chromosome aneuploidies in the human population. We performed RNA sequencing (RNA-seq) on two cell types (lymphoblastoid cell lines and primary skin fibroblasts) from 176 individuals spanning 11 different sex chromosome constitutions—from 45,X (Turner syndrome) to 49,XXXXY. We analyzed the resulting data from these 176 individuals using linear regression models to identify significant changes in Chr X gene expression in identically cultured cells with zero, one, two, or three copies of Xi. 38% of Chr X genes displayed significant Xi-driven expression changes, which we quantified on a gene-by-gene basis using a novel metric that we developed called ΔEX. By combining ΔEX findings with allele-specific analyses performed in the same cell lines and comparing our results with published, independent annotations of genes subject to XCI, we found that Xi positively or negatively modulates steady-state levels of transcripts of at least 121 genes on Xa, in trans. Thus, Xi and Xa expression are highly interdependent. By combining ΔEX with published gene-wise metrics of evolutionary constraint, we identified a set of 10 Chr X genes most likely to drive phenotypes that are associated with natural variation in Xi copy number. These 10 candidate “drivers” can now be prioritized in studies of sex differences in common disease and in explorations of sex chromosome aneuploidy syndromes.
Results
Sampling gene expression across sex chromosome constitutions
To conduct a robust, quantitative analysis of Xi’s impacts on X-linked gene expression, we recruited individuals with a wide range of sex chromosome constitutions to provide blood samples and/or skin biopsies (Figure 1A). We generated or received Epstein Barr virus-transformed B cell lines (lymphoblastoid cell lines [LCLs]) and/or primary dermal fibroblast cultures from 176 individuals with one to four X chromosomes and zero to four Y chromosomes. After culturing cells under identical conditions, we profiled gene expression by RNA-seq in LCLs from 106 individuals and fibroblast cultures from 99 individuals (some individuals contributed both blood and skin samples; Tables 1 and S1). To enable analysis at both the gene and transcript isoform levels, we generated 100-bp paired-end RNA-seq reads to a median depth of 74 million reads per sample. A resampling (bootstrapping) analysis of our dataset indicated that including more individuals with sex chromosome aneuploidy would only marginally increase the number of differentially expressed genes detected in our analyses (Figure S1; STAR Methods)…
