Scribd
Upload
3 views30 pages

Chapter 8

Papers to read

Uploaded by

Monica Romero
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
3 views30 pages

Chapter 8

Papers to read

Uploaded by

Monica Romero
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
cHAPTER 8 Epigenetic Modifications of the Genome endetian genetics is one of the most successful explanatory frame- works in biology, but an early criticism was that it was concerned with the transmission of hereditary information and not with how the information was used in development. Indeed, the developmental biolo- gists thought that geneticists were so caught up with genes and chromo- somes that they ignored other mechanisms and processes that were essen tial for the proper development and functioning of organisms. For exam- ple, a key problem for embryology was that differentiation occurred even though all the cells are derived from the fertilized egg and have the same nuclear genome. It seemed self-evident to embryologists that there must be factors in the cytoplasm—maternally inherited—that contributed to differentiation. As well, examples of gene expression and inheritance were found that did not fit the Mendelian pattern—for example, Barbara ‘McClintock's observations of variegation in maize kernels and the inher- tance of traits associated with cytoplasmic organelles such as chloroplasts =}The term “epigenetics,” coined originally to describe how genetic information was used in development to produce an organism, came to cover a diverse set of phenomena but was regarded as peripheral to main- stream genetics. However, as the tools of recombinant DNA began to be applied, it became increasingly clear that these non-Mendelian patterns of inheritance and ae a expression play an extremely important role in the life of the cell” Bpigenctics is now taken to mean heritable mod- 189 190 » CHAPTER 8 iffcations to DNA that alter gene expression but not the genome DNA sequence (mutations). These modifications are typically inherited through mito- sis, and ofien through meiosis, and although they are heritable, they are not permanent and can be reversed, The field of epigenetics now includes some of the most fascinating biological phenomena. In this chapter, we will describe such epigenetic phe- nomena as X-chromosome inactivation and imprint- ing. The mechanisms of the control of gene expres: sion we consider here will point the way toward Chapter 9, where we will examine a newiy revealed world in which RNA plays wholly unsuspected roles. The Different Dosage of Sex Chromosomes Demands a Solution to the Problem of Gene Expression Having the right number of genes is vital to the well- being of many organisms, especially, it seems, of inamimals. Human beings are diploid (we have pairs of chromosomes) and chromosomal imbalances have severe consequences. These imbalances arise during. meiosis when chromosomes fail to separate (“nondis- junction”), thereby producing gametes missing a chromosome or having an extra copy of a chromo- some. In the fertilized egg, these result, respectively, in monosomy or trisomy of that chromosome. In humans, monosomy of any of the autosomes (the hnon-sex chromosomes) is lethal before birth, and although children with trisomies may be born, they suffer from severe deformities. Trisomy 21 (Down syndrome) is an exception and it is probably a conse quence of the very small number of genes (~330) present on chromosome 21. However, the sex chro~ mosomies are anomalous; human males are mono somic for two chromosomes (the X and Y), whereas human females ate diploid for the X chromosome and have no copy of the ¥. This situation reflects the evolutionary origin of the ¥ chtomosome and pro vides a remarkable example of epigenctics During male meiosis, even though the X and Y chromosomes are very different in size, they come together and pair, albeit only over a shore distance, just like the other chromosomes. Such pairing. requires substantial regions of similarity berween chromosomes, and this observation suggests that the X and Y chromosomes, despite their differences, are in fact homologs, which gradually began to diverge beginning some 300 million years ago. The two sexes are decribed as the homagametic sex and the heteregametic sex. The homegametic sex posses ¢s two identical chromosomes and so produces iden- tical gametes, all containing the same sex chromo- some, The heteragametic sex has one chromosome identical to those in the homogametic sex, as well as a second unique chromosome. When these two sex chromosomes are segregated into gametes, the result is two different types of gametes. This system of sex determination isa straightforward mechanism for producing equal numbers of male and female offspring. Mating the heterogametic sex (XY) to the homogametic sex (XX) produces four possible embryos, two each of XX and XY, so that the off spring have an overall 1:1 sex ratio (Table 8-1), Sexual differentiation requires a large number of genes, and localization of these genes on sex chro mosomes ensures that they segregate as a group dur- ing meiosis, So that these genes remain together, ctossing-over that normally occurs during meiosis between chromosome homologs is suppressed for the sex chromosomes in the heterogametic sex. In humans, there is a limited exception to this rule, which occurs in the pscudoautosomal region, where homology is sufficient to allow pairing and crossing- over between the X and the Y. In fact, ifa crossover event fails to occur in this region, the X and Y do not segregate properly during meiosis. ‘Sex determination using XX and XY has two major genetic costs. First, the heterogametic sex is haploid for both sex chromosomes, and so mutant alleles on a male X chromosome cannot be compen- sated by a normal allele on a second X chromosome. As a consequence there are many X-linked diseases, such as hemophilia and Duchenne muscular dystro- phy, that predominantly affect males because they have only a single copy of each gene on the X. ‘Women with a mutation on one X chromosome are termed carriers and generally have few or no symp- toms. Sccond, there is a difference in gene dosage between the homogametic and heterogametic sex. ‘How is it that the homogametic sex can have a dou- ble dase of each gene carried (in the ease of mam- ‘mals) on the two X chromosomes? TABLE 8-1. Genotypes of offspring produced in a system of heterogametic and homogametic sex determination Heterogametic parent x ¥ Homogamet x xx 7 parent x x xv ‘That there is a special mechanism to correct gene dosage is evident from the fact that human beings ‘with unusual numbers of sex chromosomes are viable. Women with Turner syndrome, who have only one X chromosome (typically called XO, where O represents the absence of a chromosome), are rel- atively normal, although they are sterile and have minor developmental changes. Individuals with an XXY chromosome complement (Klinefelter syn drome} are male and, although sterile, they develop fairly normally. Other X polysomies, including XXX and XXXX (females) and XXXY and XXXXY (males), have been observed. Males with XYY constitutions ‘occur as frequently as 1 in 1000 male births. These men develop normally and are fertile. Only One X Chromosome Is Active in Female Mammals Dosage compensation was noted first by Hermann ‘Muller for the w locus on the Drasophila X chromo- some. The wild-type 1" eye color is red, whereas the mutant allele 1" produces orange cyes. Muller noticed that females homozygous for w/w" had the same intensity of eye color as did males who had a single copy of w* (and a ¥ chromosome [Fig. 8-1]). ‘The implication was that there was some form of regulation that made one copy of m equivalent to two copies of a. The mechanism by which this ‘occurs is significantly different from that in mam mals, but the discovery of dosage compensation in Drosophila led the way for the explanation of sex chromosome dosage compensation in mammals. Tn 1949, Murray Barr and Ewart Bertram found a surprising difference between the nuclei of new- rons in male and female cats. In the nuclei of females, but not males, they found a heterochromat- ic body that stained deeply with DNA-sensitive dyes. ‘This observation of female-specific heterochromatin was corroborated in many other mammals, includ- ing humans, and became known as the “Barr body.” ‘ewas assumed that the Barr body represented an X chromosome, but it was Susumo Ohno’s studies of chromosomes in regenerating rat liver cells that demonstrated the identity of the Barr body as an X chromosome. In the prophase nuclei of female ani- ‘als, a very condensed X chromosome was observed paired to a less-condensed X. Particularly convincing evidence came from studies of nuclei from individu- als with differing numbers of X chromosomes—the umber of Barr bodies was always one less than the number of X chromosomes (Table 8:2). Epigenetic Modifications of the Genome = 191 Apricot Eye (w) 1 dose reg ihe FIGURE 8-1 Dosage compensation of genes on the Drosophila X chro- rmosome. (left column) The apricot allele (wl of the white gene (responsible for pigmentation ia the Drosophila eye! produces a pale orange color rather than the wild-type red color in females with a single copy ofthe gene. This gene is located on the Drosophila X chromosome, and females bear- ing a single copy can be prepared by crossing a line with a deficiency inthe region of the X carrying white with a fly car rying the w# allele, Sueprsingly, the eyes of males with one copy of w* (on the single X chromosome) area deeper shade cof orange, identical to females homozygous for w'/w*. Thus, the activity ofthe white gene is adjusted so that a single copy cof 1 in males is equivalent to two copies in females. (Right ccolurnn) This dosage effect on the X chromosome was further extended by studying male and female flies with an extra copy of w#, carried on a deleted X chromosome (lelX). This fragment of X was used because flies triploid for an entire X ‘are not viable and XXY fies are female.! The eyes of females ‘with three doses of w were nearly red, and those of males showed even more pigmentation, Careful measurements of the pigment demonstrated that the ratio of these three-dose females to two-dose males was approximately 1.5 to 2, as expected for a compensation mechanism that equalizes two doses of X-linked genes in females to one dose in males. ‘The mammalian X chromosome was not the first condensed chromosome observed. As long ago as 1928, the embryologist Emil Heitz developed tech- niques for cell staining that revealed that portions of chromosomes were highly condensed throughout the cell cycle. He termed this condensed material heterochromatin, to distinguish it from the euchro: matin that cycled between a condensed form during mitosis and a decondensed state during interphase. ‘Major portions of the X chromosome in Drosophits are permanently condensed, and i¢ was shown that the X-chromosome heterochromatin had a silencing 192 «© CHAPTER 8. ‘TABLE 8.2. Barr bodies and X chromosomes ‘Chromosome complement Number of XO, XY, XY HX RY, XXYY OK, XXY, XXAYY XXX, 200KKY XXX influence on genes translocated nearby (Fig. 8-2). For example, deletions and rearrangements that brought the white gene w* (making red eyes) close to the heterochromatic region of the X caused silencing, so that" was turned off in some cells. ‘The eyes of these flies were colored in a variegated pattern of red and white, These observations in Drosophila corrclated heterochromatin with silenc- ing and suggested how dosage compensation in ‘mammals might be brought about. If chromosomes fanctioned similarly in mammals, the heavily con- densed X chromosome that forms the Barr body was also likely to be inactive. In 1961, Mary Lyon advanced a hypothesis to ‘explain how X-chromosome dosage was controlled in mammals. While studying genes affecting coat color in mice, she observed that mutations in X-linked coat- color genes were neither dominant nor recessive. Instead, females heterozygous for these mutant coat- color genes had mottled or dappled coats (Fig, 8-3). @ Euchromatio. Ww gene etercetvomatin. Centromere oD we gene FIGURE #2 However, males never displayed these patterns, nor did rare XO females. Similar variegation of coat color ‘was also observed in animals with translocations between the X chromosome and autosomes carrying coat-color genes. This indicated that the X was influ: cencing the expression of the translocated genes in a ‘manner similar to that observed in Drosophila X chro- mosome translocations. Lyon proposed that one of | the two X chromosomes was randomly inactivated in female animals early during embryogenesis, so that in some cells one X chromosome was active and in some cells the other was active. The outcome of this ran- dom inactivation was a random distribution of hair cells expressing different pigments, which produced randomly patterned coat coloration in females. I also made the active X chromosome complements of male and female cells the same—both having a single active X chromosome. Lyon’s X-inactivation hypothesis was soon vali- dated by Ernest Beutler’s studies of another X- linked gene, the enzyme glucose-6-phosphate dehy- drogenase (G6PD). Deficiency of this enzyme pro- duces @ type of anemia, and the activity of the enzyme can be easily assayed. Women heterozygous for a G6PD mutation had two populations of red blood cells, one population expressing G6PD and the other deficient in the enzyme, This demonstrat- ‘ed that the X chromosome with the wild-type G6PD_ gene was active in the cells expressing G6PD, where- © Wi type Nal KO HK Heterochromatin on the Drosophila X chromosome silences nearby genes. Flies bearing a variegating eye color were discovered. (a) When their chromosomes were examined, it was apparent that the X chromosome: hhad been rearranged, so that the white gene had been brought closer to the highly condensed theterochio- ‘matic) portion of the X chromosome. (b} Similar rearrangements of the white gene near to the centromeres ©r ribosomal DNA locus of other chromosomes showed similar variegated silencing patter. This phenom tenon has been termed position effect variegation (PEV), because the marker gene (in this case, white) shows varying levels of gene expression when its placed near silent regions of the genome, @ ‘mottled gone Genotypes: 9 na XX ® brown gene 7 é 4 + + + jl “AN Wh | ti ti FIGURE 8-3 Epigenetic Modifications of the Genome ® 193 oma <— <2 Variegating coat-color patterns observed in female mice demonstrate random X inactivation in mammal. Mary Lyon analyzed mutations of linked coat-olor genes. (a) The wikt-lype X-linked pigmentation gene rotted produces black mice. Females homozygous forthe motled mutation (or mitant males, which Canty 2 single copy of mottled) are light gray, Surpsingly, female heterozygotes are variegated, with aeas of their ‘coat that are fully black and others that are light gray—apparently a mx of wildtype anad mutant! Rare Females with single X chromosome (XO) ate ako light ray. (5) Another coat-color gene, brown, has also heer Used to study coat-color variegation, Normally, the brown gene is Kocated on the eighth chromosome, and mice that are either homozygous forthe wild-type allelo—or ae heterozypous-—are blac, cegariss of sex, ‘whereas homozygous mutans are brown, However, females having a translocation between chromosomes 8 tnd X have a variegated coat color. No variegated males were found, even among offspring of the variegated females: These observations suggest that two copies of the X chromosome are requited fr the variegation effects, and thatthe X chromosome has the ability tigger variegation in other genes translocated ot as the X chromosome with mutant GOPD was active in the other cells. This finding was corroborated by analysis of G6PD in clonal cell lines produced from skin biopsies taken from six women heterozygous for G6PD isozymes Gd A and Gd B. The clones expressed only one or the other isozyme—none was AB. The experiment also demonstrated that the inactive state of a given X chromosome is maintained through mitosis (Fig. 8-4) In women heterozygous for X-linked mutations thar cause disease, 50% expression of a gene may or may not be sufficient to prevent illness depending ‘on whether the proteins must be made in each cell ‘or can be supplied from a limited population of cells. 194 = CHAPTER 8 ‘Culture of skin cells from heterozygous female Ss Cells muitipyy Isolate protein Fun protein gel and detact glucose-6-phosphate dohycrogenase For example, females heterozygous for G6PD defi- ciency suffer bouts of anemia hecause roughly half of their red blood cells are GOPD deficient. On the other hand, most females carrying a heterozygous mutation in the blood clotting factor VIII do not have hemophilia. These women can pass the muta- tion on to their sons, but they are rarely themselves affected, because their single good copy of factor VIMI supplies adequate levels of the enzyme. This is because the factor VIII protein is secreted from liver cells into the blood plasma, and factor VIII levels that are even just 50% of normal are still sufficient for normal blood clotting. However, about 10% of women heterozygous for factor VITT mutation hap- pen to have a high number of liver cells with an inac- tivated wild-type X, so that insufficient amounts of the clotting factor are produced, These women have mild symptoms of hemophilia. Below, we will learn FIGURE 8-4 ‘Glucose-6-phosphate dehydrogenase (G6PD) is used to demonstrate the ‘occurrence and stability of random X inactivation in humans, Skin samples ‘were taken fram women heterozy- gous for two different isoforms of the fenzyme, A and B. These two Isoforms hhave slightly different mobility cn a gel. The skin samples were disrupted into single cells, and clonal lines were allowed to grow up from each of these cell. Protein was isolated from these lines and analyzed. From each ‘woman, cell lines were recovered ‘expressing ether isoform A or isoform B but not both. mote about the mechanisms by which one of the two female X chromosomes is randomly inactivated. The Inactivated X Chromosome Is Coated with a Noncoding RNA Called Xist Some translocations between an X chromosome and aan autosome result in inactivation of translocated autosome genes that are adjacent to X chromosome sequences. Analysis of patients carrying transloca- ions between the X and autosomes demonstrated that there was a segment of the X, termed the X-inac- tivation center (Xie), that caused the silencing. The silencing influence spreads from the X for a limited distance along the attached autosomal segment. Ifa translocated autosome is attached to 2 segment of the X chromosome lacking the Xic, there is no silencing of the autosome. The discovery’ of the gene respon- sible for the activity of the Xic had to await the devel opment of methods for molecular cloning, The analysis of cDNAs made from the X chromo: some transcripts led to the discovery of a female-spe- cific CDNA clone. Expression of this transcript, called Xést, correlated with the presence of an inac tive X chromosome, so that Xist RNA was present only in females and not males. Cell lines derived from individuals with X polysomy, such as XXY, XXXXX, and XXXXY, had higher levels of expression of the Xist transcript than normal XX cells contain ing a single inactive X chromosome, The correlation berween the number of inactive X chromosomes and the amount of Xist transcript tan- talizingly suggested that Xist transcription originated from the inactive X chromosome, but definitive proof required study of active and inactive X chromosomes in isolation. Remarkably, itis possible to make hybrid mouse-human cell lines containing single human chromosomes as well as a full complement of mouse chromosomes. Murine and human cells ate mixed and treated with Sendai virus or a chemical such as polyethylene glycol, both of which induce fusion of cell membranes. ‘The resulting hybrid cells initially contain both mouse and human chromosomes but tend to spontancously lose most of the human chro- mosomes until only 2 few or even a single human chromosome remains. Fusing. cells in this way does not affect X inactivation—an inactive human X chro mosome remains inactive, The results of the experi- ment showed clearly that Xist transcription occurred only in cell lines containing an inactive X (Fig. 8-5) ‘Mouse-human fasion cell lines were also used to map Xisr to the Xic, Mouse-human hybrid cell lines were established ftom human females who carried translocations between the X chromosome and autosomes. Each cell contained a different part of the X chromosome. Analysis of Xisr transcription in cach of these cell lines (for which the X-inactivation status was known) revealed that Xist mapped to the location of the X-inactivation center. However, when the full-length Xist gene was is0- lated and sequenced, no protein-coding, sequence could be found, Intriguingly, in situ hybridization showed that the Xist transcript was present in the nucleus and was localized to the inactive X chromo- some. In cells with XXX and XXXXX constitutions, the Xist transcript “painted” each of the multiple Barr bodies in these cells (each Barr body corre- sponds to an inactive X chromosome). It seems that, in ways still not yet understood, Xist RNA coats the inactive X chromosome and silences genes. Fpigenetic Modifications of the Genome * 195 ne L_- Pa Heme toobiest oot g Mouse tumor cel cutee freee =¢o-- sett ra Sie ne os Seige =) & ) umber ot human chromosomes plus all ‘mouse chromosomes ray A ferner Seeaon ees os 1 2 14 ricune 0s Somatic cell fusion is used to demonstrate that Xist origi- nates from the inactive X. Mouse-human hybrid lines can be prepared by mixing cells and inducing membrane fusion, ‘Mest of the human chromosomes are fost at random, but lines can be found with one or moce chromesomes remain- ing. Lines containing aetive or inactive X chromosomes were prepared. RNA from these lines was analyzed by [Nosthern blot for Xst expression. Xist could be detected only in lines containing an inactive X (Xi hybrid). Note that the Dist transcript is heterogenous in size, Most genes produce ‘one or a few transcripts of defined langth—the heterogene- ity of Kistis atypical 196 = CHAPTER 8 “These experiments correlated Xist transcription ‘with the silent X chromosome, but was it required for silencing? Targeted deletions of Xist were made in mice, In these animals, an X chromosome with a dele tion of Xist was never silenced, thereby showing that Xist was required for silencing. But was Xie itself able 10 initiate silencing? This was demonstrated using, mouse embryonic stem (ES) cell lines containing Xist transgenes under the control of an inducible promot- er, When expression of the Xr transgene was induced, _Xist RNA was localized to the chromosome where the transgene was integrated and induced silencing of genes along that chromosome. When the transgene was inserted in the X chromosome of male cells and turned on, the cells died because the transgene caused inactivation of the single X chromosome present in ale cells, thereby silencing many essential genes. The silencing ability of Xist depends on the dif ferentiation state of the cells. When ES cells were allowed to differentiate before Xist expression was induced, the Xisr transgene no longer could induce silencing. In addition, in undifferentiated ES cells, the Xisr transgene silenced genes only while it was induced—if the Xistinducing drag was removed, normal gene expression resumed. Differentiation of ES cells triggers the process of X inactivation. In cells undergoing differentiation, there was a brief period in which removal of the Xist-inducing drug relieved silencing, but as differentiation progressed, the Xist-triggered silencing became irreversible. In fact, Xist can be deleted from an inactive X chromo: some withour disrupting the silencing of the rest of the genes on that chromosome. The Gene Xist, Its Complement, Tsix, and Other Nearby Sequences Are Required for X Inactivation Although an inducible Xist transgene can direct the inactivation of a chromosome, subsequent experi ments have revealed that other sequences on the X chromosome are necessary to ensure proper X inac- tivation. There are four separate steps in the process of X inactivation, First, the number of X chromo: somes present in the cell is in some way counted, so that all but one will be inactivated. Second, one of the X chromosomes is selected for inactivation. ‘Third, the process of silencing begins on the select- ed chromosome. Finally, the silencing signal spreads over the entire chromosome. The process of X inac- tivation has been intensively studied in mice using both targeted deletions on the X chromosome, and insertion of X chromosome segments into autoso- mal sites. Careful deletion of X-chromosome seg: ments helps to identify regions required for proper silencing, whereas insertion of X-chromosome sequences elsewhere in the genome has helped to identify the minimal sequences needed for silencing, ‘While characterizing the X-inactivation center, sev ral groups of investigators noticed that Xist was not the only transcript found. Surprisingly, long tran scripts were found in the antisense orientation to Xist and overlapping the Xist transcript entirely. ‘These transcripts originate from a gene that was named Tsix, in recognition ofits antisense orientation to Xist Like ist, Tax has n0 apparent open reading frame. ‘The expression patterns of Tair are largely comple- mentary to those of Xist In ES cells, Thivis expressed at high levels from both X chromosomes, whereas Xiit expression is low. As differentiation begins, one X chromosome is “chosen” for inactivation. Tir expres- sion quickly becomes limited to the fature active X, whereas Xisr expression from that chromosome declines. Simultaneously, Xist expression fiom the fature inactive chromosome rises, whereas Tsix expres- sion falls, Once X inactivation is established, ‘Thix is completely silent on the inactive X, whereas it contin- tues to be transcribed from the active X for some time. Female mice that are heterozygous for a Tsix deletion preferentially inactivate the X chromosome with the deletion, thereby showing that Tix tran- scription is required on the chromosome that remains active. As predicted, if Thix helps to inhibit silencing by Xist an X chromosome carrying an overexpressed Thix transgene is always active, However, in male cells, where X inactivation is never initiated, chromosomes with a Tsix promoter dele- sion remain active. Thus, Tsiv is needed only when an active X must be chosen. Further deletion experiments with the mouse X chromosome have identified another region, Xite, that promotes Tiix expression and so fosters the choice of the active X chromosome. The Xite region is polymorphic among different strains of mice. Because the activity of transcriptional start sites in the region correlates with the likelihood that an X chro- mosome will be chosen to remain active, it seems that polymorphism at the Xite region may influence which X chromosome outcompetes the other. The Xite region and the Tsix region can direct pairing of the X chromosomes in cells undergoing X inactiva- tion. This pairing seems to allow cross talk between X chromosomes throughout the process of counting. and the eventual choice of the active X. Not All Genes on the Mammalian X Chromosome Are Silenced ‘We have seen how one of the two X chromosomes in female mammals is randomly inactivated to achieve proper gene dosage, However, careful examination of ‘gene expression reveals that not all of the genes on the inactivated X are silenced. The Xist gene, for example, is actively transcribed from the inactive X. Another such gene is steroid sulfatase, the deficiency of which causes the X-linked skin disease ichthyosis, In contrast to the experiment with G6PD, cell clones deficient in steroid sulfatase could not be isolated from women whose sons were affected by X-linked ichthyosis. (These mothers must be heterozygous for steroid sul fatase deficiency.) Furthermore, all. mouse-human hybrid cell lines with single X chromosomes from these women expressed steroid suilfatase, whether the lines contained an inactive X or an active X. The steroid sulfatase gene is located close t0 the telomere of the short arm of the X chromosome, near the segment of the X that pairs with the ¥ chro- mosome at meiosis, This has been termed the psendoawtosomal region, pairing and crossovers occur there at meiosis in a fashion similar to the meiotic crossovers that occur between auto- somes. Other genes in the pseudoautosomal region are expressed from both the X and the Y in males and are expressed from both the active and inactive X in females. ‘The availability of the complete sequence of the X chromosome has permitted the detailed analysis of gene expression from the inactive X. This has revealed a much more complex pattern of gene expression in females than had been suspected. In mouse-human hybrid cel! lines carrying an inactive human X, about 75% of genes are silenced, while some 15% escape inactivation (Eig. 8-6). A few of the latter are expressed at levels nearly equivalent to expression observed from the active X. Sunprisingly, another 10% of genes escaping inactivation show differing levels of expression on inactive X chromosomes from different individuals—varying by as much as threefold. because Sex Chromosome Dosage in Caenorhabditis elegans and Drosophila Is Achieved by Altering Transcriptional Activation Measurements of phenotypic characteristics and enzymatic activities show that there is dosage com: pensation of many genes on the Draopbila X chro- Epigenetic Modifications of the Genome #197 mosome. This is not achieved by gene silencing— both X chromosomes in female flies are active. For example, female flies heterozygous for mutations in the X-linked eye color gene white (1) show an even color rather than patchy pigmentation observed in the coat colors of mice. However, cell-by-cell X-chro- mosome inactivation might have produced variega- tion too microscopic to observe. The definitive proof that both X chromosomes are aetive in fies came from analysis of the enzyme 6-phosphogluconate dehydrogenase, 6-PGD. In Drosophila there are two different alleles of 6-PGD, A and B, whose products differ in their electrophoretic mobility. The mature form of 6-PGD is a dimer containing two copies of the polypeptide; the composition of these forms is AA and BB. When flies of the two strains are mated and the 6-PGD of hybrid females analyzed, AA, BB, and a new form intermediate in mobility between A and B are found. The intermediate form is a dimer made up of A and B. Because assembly of dimers can only occur immediately after the protein is produced, the presence of dimers of intermediate size in het- erozygous females shows that both X chromosomes are active in each cell (Fig. 8-7). ‘There are two ways in which gene dosage in males and females could be equalized while keeping all X chromosomes active: Expression of male X chromo- some genes could be activated twofold or expression of genes on the two female X chromosomes could be halved. Drosophila has adopted the first strategy— transcriptional activity on the male X chromosome is owice that of each female X chromosome (Fig. 8-8). ‘This up-regulation is accomplished in part through the action of noncoding RNAs transcribed from and localized to the male X in a manner similar to that of Xist. However, in this case the RNAs direct activa: tion rather than inactivation. Analysis of mutations that were lethal only in the male revealed a number of genes that were essential for the survival of male flies. Two genes, roX Land roX2, were discovered on the X chromosome and, like Xist, these genes express noncoding RNAs that bind to the X from which they are transcribed. However, unlike Xiah, the roX/ and roX2 RNAs exert an activating influ- ence on the chromosome to which they bind, The nematode worm C. eleaans has adopted the other solution to dosage compensation—it turns down transcription on each of the X chromosomes of the XX sex so that it equals one-half the transcription from the single male X chromosome, (XX C. elegans are hermaphrodites—they produce both sperm and eggs. Hermaphrodites can selFfertilize to produce 198 = CHAPTER & Prepare somatic cal hybvids containing an Inaotive X chromosome rom dlferent woman —— — isolate ANA from somatic call hyorids | Psoudoautosoral region mt Expression was detected (SS Expression was not detectasie 150-4 Mo FIGURE 8-6 ‘The inactive X is not completely silent. Mouse-human hybrid cell lines were established containing the inac~ tive X from different women. Gene expression from genes on the X was analyzed in these lines; yellow indi «ates silent genes and blue indicates genes that are active, Note that the Xst gene is active on each inactive X chromosome, as expected, Also, there are many active genes at the pseudoautosomal region atthe tip of the X chromosome. Not all genes are inactive, and the pattem of expression varies between women. Drosophila Aromeagow — Bhomeygute ABheerengote eee Heterocimare form Extract protoin Epigenetic Modifications of the Genome = 199 offspring, ot they can mate with male worms.) This mechanism of dosage compensation was revealed by careful measurements of the levels of X-linked mRNA in males and hermaphrodites, Females carry ing two copies ofa gene had exactly the same level of mRNA as males that had a single copy, and females heterozygous for an X-linked gene cartied half the level of wild-type hermaphrodites or males. Furthermore, several mutations were found that blocked dosage compensation—in these mutants, hermaphrodites (XX) had twice the amount of mRNA from X-linked genes compared to. males (Table 8-3) Dosage compensation of sex chromosomes is an epigenetic control imposed during early develop- ment and is stably passed on from daughter cell to daughter cell during mitosis, There is also epigenet ic control of gene expression, which persists through meiosis but, remarkably, is reset, depending on which parent transmits the epigenetically regulated gene to the offspring, Parents Do Not Make Equa! Genetic Contributions to Their Offspring Each parent contributes one haploid set of chromo- somes to its offipring, and so it was long. assumed that the maternal and paternal genomes make an equal genetic contribution of autosomal genes to the offipring. However, experiments transferring nuclei between fertilized mouse eggs showed otherwise. There is a brief time after fertilization when the pronuclei contributed by the parents have not yet FIGURE 8-7 ‘6-phosphogluconate dehyrogenase reveals that both x chromosomes are active in female Drosophila. This enzyme has two isoforms that associate soon after translation to form a dimer. The dimes have differing gel mobility depending 1 their composition. AA dimers migrate more slowly than BBB dimers. The AB heterodimer has an intermediate mobili ty. Mixing AA and BB dimers after translation does not per- rit formation of AB heterodimers. Only partial refolding of the mixed proteins will permit heterodimer formation Thevefore, the presence of AB heterodimers in female fies indicates that both X chromosomes are simultaneously active in each cell 200 * CHAPTER 8 Femsle chromosomes ‘chromosomes ae paired} Fatio of grains on chvamasome 3 to X = 162 ‘Male chromosomes ato of grains on chromosome 8 to FIGURE 8-8 ‘Transcription on the Drosophila male X chromosome is twice as active as on each female X chromosome. Cells were isolated from Drosophila salivary glands and incubated with radioactive ribonucleotides to meas. lure transcriptional activity. Regions of high transcription are revealed by autoradiography. The autoradi- graph is superimposed over an image of the salivary glands, and the number of silver grains (corresponding to the amount of transcriptional activity) were counted. Note that homologous chromosomes are paired in the Drosophila salivary gland, so thatthe two X chromosomes in the female are paired. In comparing these chromesomes to the single male X, the difference in thickness is apparent. fused. Microsurgery can remove the male or female pronucleus fiom 2 fertilized egg and replace it with the male or female promucleus from another ferti ized egg. In this way, fertilized eggs can be pro- duced readily that contain only maternal (gyno: genetic) or paternal (androgenetic) genomes. More than 200 mouse embryos resulted from pronucleus transfer in embryos containing the genomes of both parents; of these, seven embryos developed normal- ly. This is a small number of successes; but no embryo composed of just maternal or paternal genomes survived. Thus, to complete embryogene- sis, a fertilized egg mast contain one female and one male pronucleus. If both maternal and paternal haploid genomes are required to complete development, they must make functionally distinct contributions to embryogenesis, This phenomenon of differential genetic contribu tions of parents to their offspring is called imprint- ing. The first observation of imprinting was in the fungus gnat Sciara, where the sex chromosome behaves in an unusual way. All embryos initially receive three copies of the sex chromosome: one from the mother and two from the father. During development of female individuals, a single paternal chromosome is eliminated, leaving an XX constitu tion, In males, both paternal chromsomes are elimi- nated. (The male germ cells lose just one paternal X, and in a strange meiosis produce XX sperm.) In sperm, the maternal and paternal sex chromosomes are somehow reset so that they behave paternally in the next generation, Mapping experiments showed 2 heterochromatic region on the X chromosome that directs the imprinted behavior. The function of this region is determined by the parent; passing the chro- mosome through the male sets the “imprinting mark,” whereas passage through the female resets i. ‘The concept of imprinting was developed from these observations that affect whole chromosomes, but similar effects were soon observed on single genes. For example, in maize, a transcription factor called 7, for red color, controls the genes that make pigment in the endosperm, the starchy part of the seed that nourishes the embryo. Only the maternal copy of the r gene is expressed. When several other endosperm genes are mutated in the mother, the Epigenetic Modifications of the Genome * 201 TABLE 8:3. Dosage compensation in Caenorhabditis elegans Tati pedcied| Gee —Reatne——withdonage Observed rato Comparison stulied dove _ compensation __of expression ‘ales xO 1 : : Hermapivodies XX ea ar i i les : 1 Hermaphrodites: me 2 ne i aos : ‘Hermaphrodites: ae z ° Ca foes : 1 Hermaphodies ioe 7 ih ee aes x0 ; 1 ; XX males. mye. 2 “ ne sas yon wet 10 ors as Mabe — wet 10 08s Males ; 1 4 a one 1 10 °. p28 hermaphrotte XX ‘Hermaphrodite XX. vor a ae dloy20 pemaproie aay ea Hernaphodite py 28 pemaphotie ayy ao “Hermaphrodite 2 10 26 doe20 male XO yy, fis Normal male XO ia T te a dip-20 male « 1 Normal male eet v ine uw ‘Based on data rom Mayer and Casson, Call. 871-881 (1985), Expression of faut cmosome genes fn the nematode worm [Link] was measured by Northern analysis and normalized to expression levels of autosomal genes. The gene ‘vas then compared Bere males XO) hzmapetes UX, and pectic XX males which ate XX bt have a musaton blocs hermapniodite development). the expres of mints uat and weet sexed evidences ‘dosage compensation, because expression levels were the same fo males, hermaphrodites, snd pheno- ‘ypie males he X une? didnot show evidence for dosage compensation; silat exceptions to Jossge compensation are shown by the genes in the pseudoautosoralrgion ofthe harman X Chromosore.) “The expression levels of Xlinked genes were then examined in worms cantying the dpy-28 mutation: doy 28 males are phenotypically normal bu dpy-28 hemmaphodites aesick and “dump.” Although gene feression in doy 28 males was unaected, the mutation appeared o dstupt the process of dosage com pensation in hermaphrodtes;doy-28 hemaghrodies expressed X

You might also like