Gene expression in chromosomal Ridge domains : influence on transcription, mRNA stability, codon usage, and evolution - 6: Discussion : mechanism of Ridges and implications for evolution

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Gene expression in chromosomal Ridge domains : influence on transcription,

mRNA stability, codon usage, and evolution

Gierman, H.J.

Publication date

2010

Link to publication

Citation for published version (APA):

Gierman, H. J. (2010). Gene expression in chromosomal Ridge domains : influence on

transcription, mRNA stability, codon usage, and evolution.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.

6

Discussion: Mechanism of Ridges and

Implications for Evolution

6

Discussion: Mechanism of Ridges and Implications for

Evolution

6.1 Ridges Form Chromosomal Domains That Facilitate High and Broad Expression of Genes

In the introduction, we have discussed that highly expressed genes cluster in the human genome in Regions of Increased Gene Expression (Ridges) (Caron 2001). Ridges are characterized by high gene density and high GC content and poorly expressed genes cluster in anti-Ridges (Versteeg 2003). Genes in Ridges tend to be broadly expressed throughout different tissue types (Lercher 2002). However, the maximal expression of genes is also higher in Ridges (Versteeg 2003). This is explained by the genome-wide correspondence between maximal expression and breadth of expression (Lercher 2002; Eisenberg 2003; Versteeg 2003; Singer 2005). For example, the well-known housekeeping genes beta-tubulin (TUBB) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) are highly expressed throughout many different tissues and located in Ridges on chromosome 6p21 and 12p13 respectively. Conversely, the cluster of globin genes (HBA1, HBA2, HBQ1, HBM) on chromosome 16p13 is very tissue-specific, but also highly expressed and also located in a Ridge. Ridges thus form chromosomal domains that facilitate both high expression and broad expression. In chapter 2 we have shown that Ridges indeed provide domain-wide up-regulation of a reporter gene compared to anti-Ridges (Gierman 2007). Here, we will first discuss known mechanisms that may mediate domain-wide up-regulation of gene expression, and propose a novel hypothesis.

6.2 Spreading of Histone Modifications

Chromatin in Ridges is consistently more open than in anti-Ridges (Gilbert 2004; Goetze 2007), and appears a likely mechanism for mediating a domain-wide effect on expression. Histone modifications are strongly implicated in both chromatin formation and the regulation of gene expression. They are also capable of spreading over large genomic distances as seen with spreading of methylation on histone 3 lysine 9 (H3K9) or H3K27 during heterochromatin formation (Weiler 1995), inactivation of the X-chromosome (Plath 2002), or silencing of the Hox cluster (Gould 1997). Furthermore, various genome-wide studies have shown that profiles of some histone marks are associated with active transcription (H3 acetylation and H3K4 methylation) and correlate with the transcriptome profile of Ridges (Roh 2005; Barski 2007). However, acetylation and H3K4 methylation were found to be mainly restricted to promoters and regulatory elements of genes and were thus concluded not to represent domain-wide modifications (Roh 2005; Barski 2007). Indeed, examples of domain-wide spreading of histone marks mentioned above only represent repressive marks at specialized domains of related genes. In general, the spreading of histone modifications is blocked by boundary elements, where insulator proteins bind (West 2002). Currently, two insulator proteins are known: the ubiquitous CCCTC-binding factor (CTCF) (Lobanenkov 1990; Bell 1999) and the germ-line-specific CTCF-like (CTCFL, a.k.a. BORIS) (Loukinov 2002). Genome-wide profiling of CTCF revealed that in general, CTCF binding occurs every 2–3 genes, suggesting a dense

6

compartmentalization of chromatin domains in the human genome (Kim 2007; Barski 2007). The exceptions mostly appear to be specialized clusters of co-expressed genes such as olfactory genes and histone genes (Kim 2007). Also, CTCF binding sites demarcate chromosomal regions that escape X inactivation (Filipova 2005). It is possible that active histone marks do not spread along the chromatin fiber, but rather that domain-wide deacetylation occurs outside of Ridges. This would predict that treatment of cells with histone deacetylase (HDAC) inhibitors provides a general up-regulation of expression outside of Ridges. This is not the case however, and generally only a few hundred genes are found to be regulated after treatment with HDAC inhibitors (de Ruijter 2003; de Ruijter 2005). As Ridges are hundreds of genes long, we conclude that the lack of spreading of active marks and the ubiquitous binding of CTCF appears to make a spreading of histone modifications a less likely mechanism to explain domain-wide up-regulation of gene expression in Ridges.

6.3 Nuclear Localization

Another mechanism underlying domain-wide high gene expression in Ridges might be targeted recruitment of Ridge-sequences. Targeting of the chromatin fiber to specific sites in the nucleus together with histone modifying enzymes, would explain the correspondence of nuclear localization and Ridges (Goetze 2007) and the occurrence of open chromatin fibers throughout gene dense (i.e. Ridge) domains as reported by Gilbert et al. (Gilbert 2004). However, as discussed in the introduction, nuclear localization is often a consequence rather than cause of transcription (Volpi 2000; Zink 2004; Pickersgill 2006; Morey 2009). It is conceivable though, that for example the Lamina Associated Domains (LADs), which are characterized by interaction with the nuclear lamina, low gene expression, inactive histone marks, and low gene density, are to some extent responsible for the low expression in anti-Ridges (Guelen 2008; Wen 2009). If LADs are the only organizing mechanisms underlying Ridges and anti-Ridges, this would predict a dichotomy of the genome, where any gene residing in the roughly 60% of the genome not consisting of LADs (Guelen 2008), is only influenced by local effects such as nearby enhancers etc. This is not in agreement with our own data, where we still see a clear difference in expression of a reporter construct between Ridges and intermediate domains (Gierman 2007). Even within Ridges, reporter genes embedded within the most active Ridges showed a 2-fold higher expression than the average for all Ridge integrations together (Gierman 2007). Although we can not rule out other potential mechanisms which might recruit e.g. histone acetyltransferases to Ridges, nuclear localization or LADs do not provide sufficient explanation for the existence of chromosomal Ridge domains.

6.4 Gene Density and Bystander Effect

The striking correlation between gene density and the expression profile of the human transcriptome map suggest an alternative simple mechanism: the increased proximity of transcriptionally active genes in Ridges could provide a synergistic advantage and increases expression of genes through the increased availability of transcription factors, enhancers or other regulatory elements. For example, when the murine globin locus control region (which has a strong enhancer activity), is

6

up-regulated (Noordermeer 2008). Also, a reporter construct integrated into the Arabidopsis Flowering Locus C, is up-regulated together with induction of the surrounding genes (Finnegan 2004). Similarly, induction of immediate-early genes in mouse cells was shown to induce transcription of adjacent genes (Ebisuya 2008). This is not necessarily a general phenomenon, as there are also reports where genes fail to be up-regulated upon activation of their adjacent genes (Zink 2004; Morey 2009). Importantly, in the study by Morey et al., the genes that did not show up-regulation were already active, and the lack of a bystander effect can not be attributed to a lack of the necessary transcription factors (Morey 2009).

In our study of the expression of integrated reporter constructs and their relation to domain-wide expression levels, we also considered the possibility that gene density or a bystander effect was responsible for the higher reporter activity in Ridges (Gierman 2007). This would imply that not a domain-wide property of Ridges, but rather the increased occurrence of nearby highly expressed genes would have up-regulated the reporter constructs in Ridges. Although Ridges are enriched for highly expressed genes, roughly two thirds of highly expressed genes are located outside of Ridges. Thus, many clones with integrations outside of Ridges were in fact flanked by highly expressed genes. Conversely, we also had clones with integrations in Ridges flanked by poorly expressed genes. We found no significant difference in reporter activity when comparing reporter constructs with highly expressed neighboring genes versus those with poorly expressed neighboring genes. However, when we split up these groups according to their chromosomal expression domain, we again found that reporters integrated in active expression domains showed a higher activity, regardless the expression of their neighboring genes (Gierman 2007). It is clear that regulatory elements in the local genomic environment such as enhancers can affect the transcription of nearby genes. However, the analyses described above do not provide evidence for either gene density or a bystander effect of neighboring genes, as underlying mechanism for Ridges.

6.5 GC Content

We have concluded that gene density alone is not likely to mediate the up-regulation of expression by Ridges. GC content has also been reported to affect transcription directly. Kudla et al. reported that GC-rich isoforms of Hsp70, IL2 or GFP genes were more highly expressed due to their high GC content (Kudla 2006). GC content has also been proposed to influence transcription via Z-DNA formation (Khuu 2007; Rich 2003). However, in our study we observed a clear effect of Ridges on transcription, whilst using an identical reporter construct (with the same GC content) (Gierman 2007). It does not seem likely that the GC content of the sequences flanking the reporter up-regulate transcription. Also, correlation of reporter activity with surrounding GC content produced weaker correlations than with the transcriptome map or gene density (Gierman 2007, data not shown).

There is however another genomic feature directly implicated in transcription that shows a strong correlation with the Human Transcriptome Map: occurrence of CpG

6

islands (see Figure 1). CpG islands are GC-rich stretches of hundreds to thousands of bases that are enriched for CpG dinucleotides compared to the genomic average (Gardiner-Garden 1987; Antequera 2003; Han 2008). CpG islands are only found in vertebrates and often occur within regulatory regions and promoters of especially housekeeping genes (Gardiner-Garden 1987; Antequera 2003; Elango 2008). Below, we will discuss the relation between CpG islands and genomic organization and propose how CpG islands might provide a mechanism that explains the mode of action of Ridges.

6.6 Introduction to CpG Islands and Transcription

CpG islands arose due to genome-wide methylation of CpGs. In vertebrates, methylation normally occurs only at a cytosine adjacent to a guanine (CpG). As discussed in the introduction, most CpGs tend to disappear over time, as methylated

Figure 1. CpG island density and the Human Transcriptome Map for chromosome 16. (A) Physically

mapped transcriptome profile of the human chromosome 16. Giemsa banding is illustrated below the tran-scriptome map (centromere/heterochromatic region is green and marked ‘cen’). Ridges are shaded red and marked with an ‘R’. Black vertical bars represent genes and their height indicates domain activity for a moving median window of 49 genes (MM49) in 133 pooled SAGE libraries from different tissues. Below is the chromosomal position in megabases (UCSC Genome build HG18). Illustration adapted from Gierman et al. (Figure 1; Gierman 2007). The two lower panels show the density of (B) RefSeq genes and (C) CpG islands (UCSC Genome browser, build HG18; http://genome.ucsc.edu/). CpG islands were predicted us-ing the followus-ing criteria: a length of 200–300 bp (light green) or greater than 300 bp (dark green) with a GC content of 50% or greater and a ratio greater than 0.6 of observed number of CpG dinucleotides to the expected number on the basis of the number of Gs and Cs in the segment. Note: The globin cluster is located in the first Mb of chromosome 16 (16p13.3).

6

dinucleotides present in regulatory DNA elements such as promoters, are protected from methylation as these sequences are occupied by transcription factors and other proteins. Thus, promoters and other DNA elements occupied with transcription machinery proteins in the germ-line become relatively enriched for CpGs. The average CpG island density of the human genome is 13 per Mb (Han 2008). However, there is roughly a 5–20 fold increase of CpG density in Ridges compared to the genomic average. For example, there are 129 CpG islands in the megabase surrounding the globin gene cluster on chromosome 16p13 (Figure 1). This means that in the case of the globin cluster, every 8 kb a CpG island is encountered. The translocation of a gene into a Ridge, will thus very likely bring its promoter in close vicinity of one or more CpG islands.

Until recently, the transcriptional start sites of most promoters were assumed to be demarcated by TATA boxes (Kornberg 2007). Transcriptional start sites can be found by mapping the position of transcription factor IID, which together with other general transcription factors and Polymerase II, makes up the pre-initiation complex (PIC). Genome-wide mapping of TFIID showed that the vast majority (88%) of transcriptional start sites were associated with CpG islands (Kim 2005). These sites were not enriched for TATA boxes compared to the genomic average. Instead, less than 10% of active promoters had a TATA box (Kim 2005). Other core promoter elements like the DPE and INR make up a part of the remaining PIC binding sites (Kim 2005). There are thus two classes of genes: Those who have their transcriptional start sites occupied with transcriptional machinery in the germ-line (i.e. CpG island genes), and those which depend on core promoter elements like the TATA box for transcription (see also Antequera 2003). Importantly, the PIC is also found at transcriptionally inactive genes (Kim 2005). In addition, CpG islands are usually acetylated, even when the corresponding gene is inactive (Roh 2005).

Initiation of transcription also differs between TATA box promoters and CpG island promoters. The PIC binds directly to the TATA box, defining a strict site for transcription initiation (Sandelin 2007). Conversely, CpG island promoters use a broad region of up to several hundred base pairs or more, over which transcription starts. This indicates that transcription initiation in CpG islands is more relaxed and differs from that of TATA-box promoters.

Promoters with CpG islands appear to be constitutively occupied with transcription factors and other components of the transcriptional machinery, including histone acetyltransferases. TATA box promoters however, constitute a minority in the genome and are more tissue-specific. Here we explore the possibility that the high density of CpG islands in Ridges provides a transcriptional synergy: By constitutively recruiting general transcription factors and other transcriptional machinery proteins (e.g. histone acetyltransferases), to CpG islands in Ridges. This would facilitate the initiation of transcription and/or transcription itself.

6

6.7 Emergence of CpG Islands and Chromatin Domains: Prevertebrate Evolution

Early on in evolution and prior to multicellular life, DNA methylation and histones came into existence. The chemical addition of a methyl group to either adenosine or cytosine (i.e. DNA methylation) can be found throughout all kingdoms (bacteria, archaea, fungi, plants, invertebrates and vertebrates). DNA methylation provides many functions including the silencing of foreign DNA, regulation of gene expression, and dosage compensation for sex chromosomes (i.e. X inactivation). Histones and histone modifying enzymes arose in the archaea and eukaryotes, probably as structural components to organize DNA. When multicellular invertebrates arose, the ‘energetic cost’ of the genomic load was greatly reduced. Unicellular organisms direct a relatively large amount of resources towards the replication of their genome. For invertebrates like e.g. flies, this energetic cost has become very low, allowing them to increase the size of their genomes to expand their regulatory and evolutionary potential. During invertebrate evolution, the spatial regulation of histone modifications in chromosomal domains created specialized chromatin domains such as the Polycomb-group regulated Hox cluster and repressive chromatin domains like Lamina Associated Domains (LADs) (Pickersgill 2006; Guelen 2008; Wen 2009; see also Chapter 1.5 and de Wit 2009). These chromatin domains might have played a role in forming Ridges, as outlined below.

The protovertebrate ancestor (i.e. the common ancestor of all vertebrates, see Figure 2), probably had genome-wide CpG methylation established, as vertebrates do today. Most likely, the main function of protovertebrate DNA methylation was the silencing of genomic repeats and transposons. Due to CpG hypermutability, CpG dinucleotides started to become increasingly rare throughout the genome (Brown 1987; Brown 1988; Brown 1989). However, CpGs present in DNA sequences (often promoters) that were consistently occupied by the transcriptional machinery in the germ-line, were protected from methylation and therefore not lost by mutation. Thus, CpG islands slowly emerged, often coinciding with promoter elements of housekeeping genes. In turn, the conversion of methylated CpGs into T/G mispairs gave rise to a biased base excision repair system that favored guanine over thymine in a T/G mismatch (see also Chapter 1.4).

As the protovertebrate genome became depleted of ‘solitary CpGs’, the remaining CpG islands became functional DNA elements that recruit e.g. transcription factors. Recognition motifs of (general) transcription factors might thus have become biased towards using CpGs (i.e. C/G or G/C motifs), as this decreases the likelihood of binding non-promoter sequences (see also Antequera 2003). For example, the general transcription factors NFI and Sp1 and the family of Sp1-like transcription factors bind to G-rich consensus sequences (Suske 1999; Roulet 2000; Letovsky 1989; Zhao 2005).

The constitutive recruitment of the transcriptional machinery by CpG islands, would have made CpG islands mutually exclusive with repressive chromatin domains such

6

Figure 2. Tree of life showing emergence of isochores and Ridges. The figure shows a simplified

approxi-mation of the phylogenetic relationships between major or well-known classes of species. The vertebrates are indicated by gray branches and all have CpG islands, which do not occur outside of vertebrates. Indicated by gray letters is the weak isochore structure (wi) or normal isochore structure (i). Black arrows mark the protovertebrate common ancestor (P; ~525 million years ago) and archosaur common ancestor (A; ~250 million years ago). Ridges are indicated with a gray ‘R’: Clustering of highly expressed genes (i.e. Ridges) has been reported in humans (Caron 2001; Versteeg 2003), mice (Carninci 2005; own obser-vation (unpublished data)), and birds (Nie 2010). Organisms for which clustering of co-expressed genes has been reported, but evidence of Ridges is lacking are marked with a gray ‘N’: Fruit fly (Spellman 2002), Nematode (Roy 2002), Budding yeast (Velculescu 1997; Cohen 2000; Burhans 2006), Plants (Williams 2004; Ren 2005). The figure is based on data from the ‘Tree of Life Webproject’ (Maddison 2007). Note: Body temperature and isochore structure of dinosaurs is unknown.

6

as LADs. For example, histone acetyltransferases might counteract the effect of histone deacetylase activity required for maintaining the lower expression levels in LADs. Indeed, LADs are depleted of CpGs (Guelen 2008). Thus, the existence of repressive chromatin domains and CpG islands can create a dichotomy in the genome, where negative selection against disruption of a repressive chromatin domain can prevent the translocation or formation of CpG islands in these domains. This creates a bias that directs translocated or newly created CpG islands outside of repressive chromatin domains.

Also, negative selection can occur against translocation to repressive chromatin domains, but from the viewpoint of the gene (i.e. regardless of possible disruption of the repressive chromatin domain). Those genes, for which up-regulation of expression would provide a benefit for the protovertebrate during evolution, will most likely move outside of repressive chromatin domains or even from a repressive chromatin domain to the ‘normal’ CpG island-containing regions of the genome. Conversely, genes for which a lower expression is beneficial might benefit from relocating into a repressive chromatin domain. However, lowering expression is relatively easily achieved by e.g. mutation of the promoter sequence (most mutations will either be neutral or deleterious). It can be expected, that (partial) inactivation of a gene by mutation or deletion is far more efficient than up-regulation by random events. These mechanisms might thus have given rise to a non-uniform distribution of genes (i.e. a gene density distribution). Where over time repressive chromatin domains would become depleted of genes, CpG island-containing regions would generally increase in gene density. We hypothesize that with the generation of clusters of CpG island-containing genes, these CpG islands may exert a synergistic effect on transcription of genes in these clusters. To some extent, a bystander effect between two different genes can always be expected. However, as different CpG islands always have in common the occupation in the germ-line, they are more likely to share certain factors. This is true in general for housekeeping genes, but also for non-housekeeping genes which are expressed in the germ-line. We propose that the shared use of general transcription machinery allows CpG islands to mediate a general bystander effect.

6.8 Emergence of Ridges: Effect on Genomic Organization of Warm-blooded Vertebrates

Once clusters of genes in the (proto)vertebrate genome reach the required CpG island density threshold to exert a domain-wide ‘CpG island bystander effect’, the earliest Ridges might have arisen. Possibly, this occurred in a vertebrate amniotic (warm-blooded) common ancestor. The emergence of Ridges, regardless of the underlying mechanism, would have driven gene density distribution: Now positive selection can occur for translocation to a Ridge, for those genes for which up-regulation is beneficial (see also chapter 4). Note that there is a potential positive feedback for the creation of Ridges: Within a certain range, an increase in CpG island density can increase the ‘CpG island bystander effect’. This means, that as the up-regulatory properties of Ridges increase, it becomes easier during evolution

6

point, equilibrium will be reached as a further integration of more genes will become increasingly difficult due to the risk of disrupting an existing gene.

Another possible effect of CpG islands on transcription is the chromatin remodeling activity of Polymerase II. Apart from transcription itself, the permanent presence of PICs on CpG islands throughout Ridges, and perhaps to some extent their spreading, might aid (the initiation of) transcription and/or contribute to the open chromatin structure found in Ridges. This might also explain the observation that non-expressed genes in Ridges can also possess open chromatin structure (Gilbert 2004).

By now, the genome of the (proto)vertebrate has become highly non-uniform: Several types of (repressive) chromatin domains are present, as are Ridges and a gene density distribution has emerged. To some extent, these differences between genomic regions could provoke differences in the recombination rate (Montoya-Burgos 2003; Duret 2008; Duret 2009). Together with the already present biased repair system, this non-uniform organization might influence isochore structure (see also Duret 2002). Equally so, recombination rates might drive the repositioning of genes and therefore, the creation of Ridges. For example, (sub)telomeric regions show increased recombination (Duret 2008), and Ridges indeed often occur at telomeres. Once isochores emerge, they might in turn also affect the distribution of CpG islands. CpGs will arise in regions with increased recombination rates, potentially converting existing regulatory elements into CpG islands. It is worth mentioning here, that the rate of cytosine deamination increases strongly with (body) temperature. A rise in temperature of 10 ºC causes a 5-fold increase in the rate of deamination (Fryxell 2000). This means that the emergence of warm-blooded vertebrates (i.e. mammals, birds and possibly archosaurs, see Figure 2) has likely had a strong impact on the establishment of isochore structure. Indeed, the warm-blooded mammals and birds have a strong isochore structure, whereas cold-blooded fish have much weaker isochores (Costantini 2007). Still, fish and other vertebrates all appear to have a non-uniform distribution of gene density correlated with CpG island density. Also, genomic GC content (together with recombination), can affect gene density, as deletions might increase together with GC content and recombination rate (Montoya-Burgos 2003). Finally, the distribution of genomic repeats and transposons might, in first instance, purely be a consequence of the existing genomic organization. But once present, repeats themselves can impact genomic organization by increasing recombination, like for example Alu repeats (Witherspoon 2009). Together, the combined action of the mechanisms described above could have provided the vertebrate genome with Ridges and a complex genomic organization that can greatly facilitate the regulation of gene expression.

6.9 Impact of Ridges on Evolution

The emergence of (invertebrate) multicellular organisms required complex regulatory systems such as the Hox genes to help guide development. Vertebrates are larger and more complex and require even more regulation for highly specialized systems

6

such as adaptive immunity. However, we propose that the markedly different genomic organization of especially the warm-blooded vertebrates does not primarily reflect an increasing need for regulation. Instead, we propose that Ridges and the genomic organization in warm-blooded vertebrates compensate for the loss of natural selection to drive regulation.

The force of natural selection within a population is directly dependent on the size of the breeding population (i.e. effective population size) (Wright 1931; Wright 1937). Natural selection can in theory select for any change with a minimal fitness advantage, as long as the effective population size is large enough. This ensures that the specific mutation will occur frequently enough to become fixed. This works very well for unicellular organisms which have extremely large population sizes. Invertebrates such as flies and nematodes are much larger and as a result, have populations that are smaller than bacteria, but still large enough to allow selection for e.g. preferred codons (see chapter 4). However, the emergence of vertebrates created animals so large, that effective population sizes decrease with many magnitudes compared to invertebrates. For example, although the current world population of humans is close to 7*109_{, and was probably less than 1*10}7 _several

thousand years ago, this still is far less than the microbial population of the gut of one single human being (1*1013_{) (Hooper 2001). The small population size of humans}

and other vertebrates means that the gradual adaptation of genes (i.e. the fixing of small fitness gains through natural selection), will be impaired. Indeed, selection for preferred codons, as seen in bacteria and invertebrates, does not occur in humans or is extremely weak (see chapter 4). This might have created an evolutionary block for vertebrates. Whereas bacteria and invertebrates can ‘wait’ for the right mutations to become fixed, this might not function well in vertebrate evolution. In line with this, the sequence divergence between human and chimp promoters shows no correlation with expression divergence, indicating that evolutionary changes in expression have no relation to the amount of mutations (Khaitovich 2005).

We propose that the translocation of genes in or out of Ridges provides a mechanism by which gene expression can be regulated in several ways, without the need for ‘mutation’. We have shown that Ridges provide an immediate up-regulation of expression of 4 to 8-fold compared to anti-Ridges (Gierman 2007). In contrast to mutations with small fitness gains, such a strong effect is easier to fix during evolution. In addition, translocation to a Ridge might also convert a tissue-specific gene into a ubiquitously expressed housekeeping gene or vice-versa. Finally, the correspondence with the isochore structure would allow genes to substantially increase their expression levels further, as the high GC content of Ridges improves mRNA stability and increases the usage of preferred codons (Chapters 3 and 4).

6.10 Conclusion

We propose the model outlined above as an explanation for the emergence and mode of action of Ridges. The interaction of the mechanisms described in this chapter might to some degree explain the complex genomic organization of the warm-blooded vertebrate genome. Importantly, the mode of action of Ridges

6

several mechanisms contribute to this genomic organization. The idea of CpG island promoters as a large group of non-stringent synergistic promoters, might also explain some poorly understood experimental findings. For example, recent years have produced large quantities of genome-wide mappings of various transcription factors. Surprisingly, many of these factors are often found to be bound ubiquitously at the promoters of thousands of different genes (van Steensel 2005). Very often, knockdown of a factor will only affect the expression of a subset of the genes it directly binds (Farnham 2009; Suzuki 2009). Synergistic CpG promoters would bind many factors (as observed), but not be necessarily dependent on any specific one (explaining the lack of strong regulation after knockdown).

Here, we have outlined our hypothesis that describes a mode of action for Ridges. Ridges, combined with genomic organization, thus might allow for a strong acceleration of evolutionary change in vertebrate evolution. We hypothesize that the extremely small population sizes of vertebrates, require mechanisms such as Ridges to overcome the paucity of natural selection.

6

References

Antequera F. Structure, function and evolution of CpG island promoters. Cell Mol Life Sci. 2003 Aug;60(8):1647-58.

Barski A, Cuddapah S, Cui K, Roh TY, Schones DE, Wang Z, Wei G, Chepelev I, Zhao K. High-resolution profiling of histone methylations in the human genome. Cell. 2007 May 18;129(4):823-37.

Bell AC, West AG, Felsenfeld G. The protein CTCF is required for the enhancer blocking activity of vertebrate insulators. Cell. 1999 Aug 6;98(3):387-96.

Brown TC, Jiricny J. A specific mismatch repair event protects mammalian cells from loss of 5-methylcytosine. Cell. 1987 Sep 11;50(6):945-50.

Brown TC, Jiricny J. Different base/base mispairs are corrected with different efficiencies and specificities in monkey kidney cells. Cell. 1988 Aug 26;54(5):705-11.

Brown TC, Jiricny J. Repair of base-base mismatches in simian and human cells. Genome. 1989;31(2):578-83.

Burhans DT, Ramachandran L, Wang J, Liang P, Patterton HG, Breitenbach M, Burhans WC. Non-random clustering of stress-related genes during evolution of the S. cerevisiae genome. BMC Evol Biol. 2006 Jul 21;6:58.

Carninci P, Kasukawa T, Katayama S, Gough J, Frith MC, Maeda N, Oyama R, Ravasi T, Lenhard B, Wells C, et al., FANTOM Consortium; RIKEN Genome Exploration Research Group and Genome Science Group (Genome Network Project Core Group). The transcriptional landscape of the mammalian genome. Science. 2005 Sep 2;309(5740):1559-63.

Caron H, van Schaik B, van der Mee M, Baas F, Riggins G, van Sluis P, Hermus MC, van Asperen R, Boon K, Voûte PA, et al. The human transcriptome map: clustering of highly expressed genes in chromosomal domains. Science. 2001 Feb 16;291(5507):1289-92.

Cohen BA, Mitra RD, Hughes JD, Church GM. A computational analysis of whole-genome expression data reveals chromosomal domains of gene expression. Nat Genet. 2000 Oct;26(2):183-6

Costantini M, Auletta F, Bernardi G. Isochore patterns and gene distributions in fish genomes. Genomics. 2007 Sep;90(3):364-71

de Ruijter AJ, Meinsma RJ, Bosma P, Kemp S, Caron HN, van Kuilenburg AB. Gene expression profiling in response to the histone deacetylase inhibitor BL1521 in neuroblastoma. Exp Cell Res. 2005 Oct 1;309(2):451-67.

de Ruijter AJ, van Gennip AH, Caron HN, Kemp S, van Kuilenburg AB. Histone deacetylases (HDACs): characterization of the classical HDAC family. Biochem J. 2003 Mar 15;370(Pt 3):737-49.

de Wit E, van Steensel B. Chromatin domains in higher eukaryotes: insights from genome-wide mapping studies. Chromosoma. 2009 Feb;118(1):25-36.

Duret L, Arndt PF. The impact of recombination on nucleotide substitutions in the human genome. PLoS Genet. 2008 May 9;4(5):e1000071.

Duret L, Galtier N. Biased gene conversion and the evolution of mammalian genomic landscapes. Annu Rev Genomics Hum Genet. 2009;10:285-311.

Duret L, Semon M, Piganeau G, Mouchiroud D, Galtier N. Vanishing GC-rich isochores in mammalian genomes. Genetics. 2002 Dec;162(4):1837-47.

Ebisuya M, Yamamoto T, Nakajima M, Nishida E. Ripples from neighbouring transcription. Nat Cell Biol. 2008 Sep;10(9):1106-13.

Eisenberg E, Levanon EY. Human housekeeping genes are compact. Trends Genet. 2003 Jul;19(7):362-5.

Elango N, Yi SV. DNA methylation and structural and functional bimodality of vertebrate promoters. Mol Biol Evol. 2008 Aug;25(8):1602-8.

Farnham PJ. Insights from genomic profiling of transcription factors. Nat Rev Genet. 2009 Sep;10(9):605-16.

Filippova GN, Cheng MK, Moore JM, Truong JP, Hu YJ, Nguyen DK, Tsuchiya KD, Disteche CM. Boundaries between chromosomal domains of X inactivation and escape bind CTCF and lack CpG methylation during early development. Dev Cell. 2005 Jan;8(1):31-42.

Finnegan EJ, Sheldon CC, Jardinaud F, Peacock WJ, Dennis ES. A cluster of Arabidopsis genes with a coordinate response to an environmental stimulus. Curr Biol. 2004 May 25;14(10):911-6.

6

isochores. Mol Biol Evol. 2000 Sep;17(9):1371-83.

Gardiner-Garden M, Frommer M. CpG islands in vertebrate genomes. J Mol Biol. 1987 Jul 20;196(2):261-82.

Gierman HJ, Indemans MH, Koster J, Goetze S, Seppen J, Geerts D, van Driel R, Versteeg R. Domain-wide regulation of gene expression in the human genome. Genome Res. 2007 Sep;17(9):1286-95. Gilbert N, Boyle S, Fiegler H, Woodfine K, Carter NP, Bickmore W. Chromatin architecture of the human

genome: gene-rich domains are enriched in open chromatin fibers. Cell. 2004 Sep 3;118(5):555-66. Goetze S, Mateos-Langerak J, Gierman HJ, de Leeuw W, Giromus O, Indemans MH, Koster J, Ondrej

V, Versteeg R, van Driel R. The three-dimensional structure of human interphase chromosomes is related to the transcriptome map. Mol Cell Biol. 2007 Jun;27(12):4475-87.

Gould A. Functions of mammalian Polycomb group and trithorax group related genes. Curr Opin Genet Dev. 1997 Aug;7(4):488-94.

Guelen L, Pagie L, Brasset E, Meuleman W, Faza MB, Talhout W, Eussen BH, de Klein A, Wessels L, de Laat W, et al. Domain organization of human chromosomes revealed by mapping of nuclear lamina interactions. Nature. 2008 Jun 12;453(7197):948-51.

Han L, Su B, Li WH, Zhao Z. CpG island density and its correlations with genomic features in mammalian genomes. Genome Biol. 2008;9(5):R79.

Hooper LV, Gordon JI. Commensal host-bacterial relationships in the gut. Science. 2001 May 11;292(5519):1115-8.

Khaitovich P, Hellmann I, Enard W, Nowick K, Leinweber M, Franz H, Weiss G, Lachmann M, Pääbo S. Parallel patterns of evolution in the genomes and transcriptomes of humans and chimpanzees. Science. 2005 Sep 16;309(5742):1850-4.

Khuu P, Sandor M, DeYoung J, Ho PS. Phylogenomic analysis of the emergence of GC-rich transcription elements. Proc Natl Acad Sci U S A. 2007 Oct 16;104(42):16528-33.

Kim TH, Abdullaev ZK, Smith AD, Ching KA, Loukinov DI, Green RD, Zhang MQ, Lobanenkov VV, Ren B. Analysis of the vertebrate insulator protein CTCF-binding sites in the human genome. Cell. 2007 Mar 23;128(6):1231-45.

Kim TH, Barrera LO, Zheng M, Qu C, Singer MA, Richmond TA, Wu Y, Green RD, Ren B. A high-resolution map of active promoters in the human genome. Nature. 2005 Aug 11;436(7052):876-80.

Kornberg RD. The molecular basis of eukaryotic transcription. Proc Natl Acad Sci U S A. 2007 Aug 7;104(32):12955-61

Kudla G, Lipinski L, Caffin F, Helwak A, Zylicz M. High guanine and cytosine content increases mRNA levels in mammalian cells. PLoS Biol. 2006 Jun;4(6):e180.

Lercher MJ, Urrutia AO, Hurst LD. Clustering of housekeeping genes provides a unified model of gene order in the human genome. Nat Genet. 2002 Jun;31(2):180-3.

Letovsky J, Dynan WS. Measurement of the binding of transcription factor Sp1 to a single GC box recognition sequence. Nucleic Acids Res. 1989 Apr 11;17(7):2639-53.

Lobanenkov VV, Nicolas RH, Adler VV, Paterson H, Klenova EM, Polotskaja AV, Goodwin GH. A novel sequence-specific DNA binding protein which interacts with three regularly spaced direct repeats of the CCCTC-motif in the 5’-flanking sequence of the chicken c-myc gene. Oncogene. 1990 Dec;5(12):1743-53.

Loukinov DI, Pugacheva E, Vatolin S, Pack SD, Moon H, Chernukhin I, Mannan P, Larsson E, Kanduri C, Vostrov AA, et al. BORIS, a novel male germ-line-specific protein associated with epigenetic reprogramming events, shares the same 11-zinc-finger domain with CTCF, the insulator protein involved in reading imprinting marks in the soma. Proc Natl Acad Sci U S A. 2002 May 14;99(10):6806-11.

Maddison DR, Schulz KS. 2007. The Tree of Life Web Project. http://www.tolweb.org/.

Montoya-Burgos JI, Boursot P, Galtier N. Recombination explains isochores in mammalian genomes. Trends Genet. 2003 Mar;19(3):128-30.

Morey C, Kress C, Bickmore WA. Lack of bystander activation shows that localization exterior to chromosome territories is not sufficient to up-regulate gene expression. Genome Res. 2009 Jul;19(7):1184-94.

Nie H, Crooijmans RP, Bastiaansen JW, Megens HJ, Groenen MA. Regional regulation of transcription in the chicken genome. BMC Genomics. 2010 Jan 14;11:28.

6

der Spek P, Pombo A, de Laat W. Transcription and chromatin organization of a housekeeping gene cluster containing an integrated beta-globin locus control region. PLoS Genet. 2008 Mar 7;4(3):e1000016.

Pickersgill H, Kalverda B, de Wit E, Talhout W, Fornerod M, van Steensel B. Characterization of the Drosophila melanogaster genome at the nuclear lamina. Nat Genet. 2006 Sep;38(9):1005-14. Plath K, Mlynarczyk-Evans S, Nusinow DA, Panning B. Xist RNA and the mechanism of X chromosome

inactivation. Annu Rev Genet. 2002;36:233-78.

Ren XY, Fiers MW, Stiekema WJ, Nap JP. Local coexpression domains of two to four genes in the genome of Arabidopsis. Plant Physiol. 2005 Jun;138(2):923-34.

Rich A, Zhang S. Timeline: Z-DNA: the long road to biological function. Nat Rev Genet. 2003 Jul;4(7):566-72.

Roulet E, Bucher P, Schneider R, Wingender E, Dusserre Y, Werner T, Mermod N. Experimental analysis and computer prediction of CTF/NFI transcription factor DNA binding sites. J Mol Biol. 2000 Apr 7;297(4):833-48.

Roh TY, Cuddapah S, Zhao K. Active chromatin domains are defined by acetylation islands revealed by genome-wide mapping. Genes Dev. 2005 Mar 1;19(5):542-52.

Roy PJ, Stuart JM, Lund J, Kim SK. Chromosomal clustering of muscle-expressed genes in Caenorhabditis elegans. Nature. 2002 Aug 29;418(6901):975-9.

Sandelin A, Carninci P, Lenhard B, Ponjavic J, Hayashizaki Y, Hume DA. Mammalian RNA polymerase II core promoters: insights from genome-wide studies. Nat Rev Genet. 2007 Jun;8(6):424-36. Singer GA, Lloyd AT, Huminiecki LB, Wolfe KH. Clusters of co-expressed genes in mammalian genomes

are conserved by natural selection. Mol Biol Evol. 2005 Mar;22(3):767-75.

Spellman PT, Rubin GM. Evidence for large domains of similarly expressed genes in the Drosophila genome. J Biol. 2002;1(1):5.

Suske G. The Sp-family of transcription factors. Gene. 1999 Oct 1;238(2):291-300.

Suzuki H, Forrest AR, van Nimwegen E, Daub CO, Balwierz PJ, Irvine KM, Lassmann T, Ravasi T, Hasegawa Y, de Hoon MJ, et al. The transcriptional network that controls growth arrest and differentiation in a human myeloid leukemia cell line. Nat Genet. 2009 May;41(5):553-62.

van Steensel B. Mapping of genetic and epigenetic regulatory networks using microarrays. Nat Genet. 2005 Jun;37 Suppl:S18-24.

Velculescu VE, Zhang L, Zhou W, Vogelstein J, Basrai MA, Bassett DE Jr, Hieter P, Vogelstein B, Kinzler KW. Characterization of the yeast transcriptome. Cell. 1997 Jan 24;88(2):243-51.

Versteeg R, van Schaik BD, van Batenburg MF, Roos M, Monajemi R, Caron H, Bussemaker HJ, van Kampen AH. The human transcriptome map reveals extremes in gene density, intron length, GC content, and repeat pattern for domains of highly and weakly expressed genes. Genome Res. 2003 Sep;13(9):1998-2004.

Volpi EV, Chevret E, Jones T, Vatcheva R, Williamson J, Beck S, Campbell RD, Goldsworthy M, Powis SH, Ragoussis J, et al. Large-scale chromatin organization of the major histocompatibility complex and other regions of human chromosome 6 and its response to interferon in interphase nuclei. J Cell Sci. 2000 May;113 ( Pt 9):1565-76.

Weiler KS, Wakimoto BT. Heterochromatin and gene expression in Drosophila. Annu Rev Genet. 1995;29:577-605.

Wen B, Wu H, Shinkai Y, Irizarry RA, Feinberg AP. Large histone H3 lysine 9 dimethylated chromatin blocks distinguish differentiated from embryonic stem cells. Nat Genet. 2009 Feb;41(2):246-50. West AG, Gaszner M, Felsenfeld G. Insulators: many functions, many mechanisms. Genes Dev. 2002

Feb 1;16(3):271-88.

Williams EJ, Bowles DJ. Coexpression of neighboring genes in the genome of Arabidopsis thaliana. Genome Res. 2004 Jun;14(6):1060-7.

Witherspoon DJ, Watkins WS, Zhang Y, Xing J, Tolpinrud WL, Hedges DJ, Batzer MA, Jorde LB. Alu repeats increase local recombination rates. BMC Genomics. 2009 Nov 16;10:530.

Wright S. Evolution in Mendelian Populations. Genetics. 1931 Mar;16(2):97-159.

Wright S. The distribution of gene frequencies in populations. Science. 1937 May 21;85(2212):504. Zhao C, Meng A. Sp1-like transcription factors are regulators of embryonic development in vertebrates.

Dev Growth Differ. 2005 May;47(4):201-11.

6

Rosenecker J, Schindelhauer D. Transcription-dependent spatial arrangements of CFTR and adjacent genes in human cell nuclei. J Cell Biol. 2004 Sep 13;166(6):815-25.