Gene Regulation II

Despite the sequence diversity among promoters, genes transcribed by RNA polymerase II can be classified in two different and mutually exclusive groups according to the distribution of CpG dinucleotides across their 5' ends. In one class, the frequency of CpGs is the same as the genome average, which is roughly one every 100 nucleotides. This class invariably includes genes whose expression is restricted to a limited number of cell types... In contrast, the 5' end of the genes belonging to the other group is surrounded by a region ~1 kb long where the frequency of CpGs is approximately 10 times higher than the genome average. These regions were very appropriately called CpG islands ... and show such a conspicuous clustering of CpG dinucleotides that it can be readily detected by visual inspection of the CpG plot. The consistent association of CpG islands with the upstream region of many genes immediately suggested a possible involvement in transcriptional regulation ....

What is so special about CpGs relative to the other 15 possible dinucleotides in DNA? CpGs are the sites where methylation takes place, and ~80 % of them are methylated at position 5 of the cytosine ring in humans and mice. Somewhat paradoxically, CpGs remain nonmethylated at CpG islands, despite their abundance, whereas the majority of the remaining CpGs scattered across the genome are mostly methylated. In addition to the lack of methylation, human and mouse CpG islands have a G+C content of 67 and 64 %, approximately, while the genome averages are 41 % and 42 %, respectively...

These distinctive features of the CpG islands in terms of a lack of methylation and an elevated G+C content are accompanied by an equally distinctive chromatin organization. Chromatin analysis at global genomic level has revealed that CpG islands show the properties usually ascribed to "open" or "active" chromatin. This includes hyperacetylation of histones H3 and H4, a deficiency in histone H1, positioned nucleosomes and nucleosome-free regions that coincide with enhanced sensitivity to nucleases relative to bulk DNA... These properties highlight CpG islands as regions that are particularly well suited for direct access to DNA, which is consistent with their co-localization with the promoters of many genes. Previous studies based on the biochemical isolation of the CpG island fraction estimated approximately 45,000 and 37,000 in the human and mouse genomes, respectively..., although recent computational predictions have lowered these figures to about 27,000 and 15,500...

Regardless of the absolute number of CpG islands in the genome, a more relevant issue is what kind of genes are associated with them, since only ~60% of all human genes are associated with CpG isands. This includes all the housekeeping genes--those expressed in all cell types--and about half of the tissue- specific genes...

Transcription from CpG-rich and CpG-poor promoters
How does methylation affect transcription from CpG island and non-island promoters? Since CpG islands are nonmethylated in sperm and remain consistently devoid of methylation in somatic tissues, regardless of the expression of the genes associated with them, it is unlikely that DNA methylation would play any role in their regulation. Exceptions to this rule are the CpG islands of imprinted genes, those in the mammalian X inactive chromosome and those associated with the MAGE genes that become methylated during normal mammalian development... Even in this case, methylation is not the primary inactivating signal but takes place at a stage when transcription has been switched off by other means...

Transcription is strongly repressed upon unscheduled de novo methylation of CpG islands in cell lines and tumour cells, a phenomenon that occurs at high frequency in these situations but never in the organism under normal physiological conditions with the exceptions mentioned above...

In contrast with CpG islands, CpG-poor promoters are methylated in sperm and are always associated with tissue-specific genes. A direct role of DNA methylation in the regulation of this class of promoters predicts a correlation between their methylation profile and their level of expression. Many examples and also several exceptions to this correlation have been described, suggesting that although DNA methylation affects gene expression, it is unlikely to play a general role as a transcriptional regulator... Detailed analysis of the kinetics of demethylation and gene expression has shown that in some cases where demethylation correlates with expression, the former fol-lows the binding of transcription factors rather than being a prerequisite for it. For example, binding of nuclear factor kappa B (NF-kB) transcription factor to an intronic enhancer of a k-chain gene is required for demethylation in B cells...

That DNA methylation, despite its prominent presence in the mammalian genome, is not a general regulator of gene expression is not surprising since proper gene regulation takes place in invertebrates, many of whose genomes have a low or undetectable level of methylation.

Transcriptional Controls: The main mechanism of turning genes on and off occurs at the level of transcription. We have already seen at least one site that although it is outside the coding sequence of the gene, is important in proper proper gene expression: the promoter (RNA holoenzyme binding site). However, other sites are also involved. Transcriptional Regulation in E. coli: One Example: The regulation of the genes involved with lactose catabolism in E. coli was discovered by Jacob and Monod in 1961. E. coli only produces (in any significant quantity) a set of enzymes necessary to break down lactose when lactose is present. The lac Operon: The genes that encode the three protein needed for lactose catabolism lie adjacent to each other. These three genes plus the regulatory sequences nearby constitute the lac operon. The three genes are transcribed as one polycistronic message, therefore if one is turned on, all three are turned on. (This kind of clustering of functionally related genes is seen often in prokaryotes.)
The Structural Genes: The enzymes are β-galactosidase (cleaves lactose), lactose permease (facilitates the entry of lactose into the cell), and transacetylase (gets rid of a tag-along toxin that enters with lactose by the action of lactose permease). The genes that code for these three proteins are designated z, y, and a, respectively.
The Promoter: This is the RNA polymerase holoenzyme binding site (as described earlier). The Operator: This site is a sequence to which the repressor protein binds. It is just downstream from the promoter. When repressor protein is bound, RNA polymerase is prevented from binding to the promoter. This sequence is designed o. The operator was discover by finding cis-acting constitutive mutants (always produced the three gene produces, regardless of the presence or absence of lactose or repressor protein). The Repressor: This is a protein coding sequence which produces the repressor protein, capable of binding to the operator (and turning off transcription). This gene is designated i. It was discovered as a trans-acting gene that apparently made a diffusible substance which could affect an operator on a different "chromosome." (merodiploid, or merozygote F' cells) CAP Binding Site: This site is about 60 nucleotides upstream from the transcription start site. It binds a protein called catabolite activator protein (CAP). The binding of cAMP to CAP stimulates it to bind to the CAP binding site. This binding then facilitates the binding of RNA polymerase to the promoter (enabling transcription).
Negative Control in the lac Operon: When lactose is not present, the level of transcription of these three genes is negligible. However, in the presence of lactose, their transcription is greatly ramped up. This process is an example of negative control, since there is something that normally turns the expression of this gene off (which can be regulated). In the "normal" state (no lactose present), the i gene makes repressor protein which binds to the o site, preventing RNA polymerase binding, thereby turning off transcription. However, in the presence of lactose, an isomer of lactose (allolactose) binds allosterically to the repressor, preventing it from binding to o. Therefore, RNA polymerase can bind to the promoter and transcription of these three genes is turned on. (Protein binding turns off the operon = negative control) lac operon problems: Lac Operon Problems: I⁺ is a normal repressor gene I- is a mutant repressor gene that does not make any functional repressor protein. I^S is a mutant of the Lac repressor gene which makes repressor protein that represses the Lac operon even in the presence of inducer (lactose)(it makes a repressor protein to which allolactose cannot bind). O⁺ is a normal operator sequence O^- is an mutant operator to which repressor protein cannot bind. P⁺ is a normal promoter sequence P^- is a promoter sequence to which RNA polymerase cannot bind. Z⁺ makes normal, functional β-galactosidase Z^- is a mutant that makes no functional β-galactosidase Y⁺ makes normal, functional lactose permease Y^- is a mutant that makes no functional lactose permease For each of these genotypes, indicate whether β-galactosidase and lactose permease would be produced constitutively, inducibly, or not at all. (What does constitutively and inducibly mean?) I^- P^- O^- Z⁺ Y⁺ / I⁺ P⁺ O⁺ Z⁺ Y^-I^- P⁺ O^- Z⁺ Y⁺ / I⁺ P⁺ O⁺ Z^- Y^- I^- P⁺ O^- Z⁺ Y⁺ / I⁺ P⁺ O⁺ Z^- Y^- I^- P⁺ O^- Z⁺ Y⁺ / I⁺ P⁺ O⁺ Z⁺ Y⁺ I⁺ P⁺ O⁺ Z⁺ Y⁺ / I⁺ P⁺ O⁺ Z^- Y⁺ I^- P⁺ O⁺ Z^- Y⁺ / I⁺ P⁺ O⁺ Z⁺ Y⁺ I⁺ P⁺ O⁺ Z⁺ Y^- / I^- P⁺ O⁺ Z⁺ Y⁺ I^S P⁺ O⁺ Z⁺ Y⁺ / I⁺ P⁺ O⁺ Z⁺ Y⁺ I⁺ P⁺ O^- Z⁺ Y^- / I⁺ P⁺ O⁺ Z^- Y⁺ I⁺ P^- O⁺ Z⁺ Y⁺ / I⁺ P⁺ O⁺ Z⁺ Y⁺ Know about the "Lactose Paradox" for the exam: http://sandwalk.blogspot.it/2008/10/lactose-paradox.html
Positive Control in the lac Operon: One wrinkle in this situation is that if both glucose and lactose are present, the cell apparently "prefers" to use glucose over lactose, and the lac operon is turned off. This is positive control because the event involved is the facilitation of transcription. When the concentration of glucose is high, the concentration of cAMP is low (due to the regulation of another gene: adenylyl cyclase, which converts ATP to cAMP). When glucose levels drop, cAMP levels rise. Therefore, when there is a high concentration of glucose, there is little cAMP, so it cannot bind to CAP. Without cAMP, CAP does not bind to the CAP binding site, and therefore, RNA polymerase binding is not facilitated. Transcription is off. But, if there is little glucose, then cAMP levels are high, so CAP-cAMP levels are high, and CAP-cAMP's binding to the CAP binding site facilitates RNA polymerase binding to the promoter, turning transcription on. (Binding turns on.) So, for the lac operon to be turned on, lactose must be present and glucose absent (or in low concentration). That is, the negative control mechanism has be inactivated and the positive control mechanisms has to be activated. Many similarly regulated mechanisms have been found in prokaryotes. (About how CAP-cAMP facilitates RNA polymerase binding to the promoter: CAP "recruits" RNA polymerase to bind to the promoter. It interacts with a few amino acids on the RNA polymerase α-chain CTD and causes them to bind to the promoter, thus causing the whole RNA polymerase to bind to the promoter. The reference is here. Start reading at "Transcription activation at class I CAP-dependent promoters.")
Transcriptional Control in Eukaryotes: While there are many common features between prokaryotic and eukaryotic gene regulation (like the presence of a promoter and the presence of controlling sequences upstream from the structural gene), there are some major differences. One is that eukaryotes do not have polycistronic message and we usually do not see clusters of functionally related genes that are turned on by turning on the transcription of a single RNA. Functionally related genes are often scattered across the genome and therefore co-regulation is more complex than in prokaryotes. Enhancers: Besides the promoters, other gene-specific sequences called enhancers are important in regulating gene expression in eukaryotes. These sequences may be 10 of thousands of nucleotides away from the gene, or even on a different chromosome. A specific transcription factor binds to the enhancer and others may bind near the promoter, enhancing the binding of RNA polymerase to the promoter (turning the gene on). (Adding an enhancer changes fish fins into limb-like appendages. Enhancer involved in being blond.) These important DNA binding proteins have similarities in structure: zinc fingers, helix-turn-helix, leucine zipper, helix-loop-helix.
Repressors: Proteins may also bind to sites near the promoter and interfere with RNA polymerase binding. They may also interact directly with RNA polymerase to prevent transcription.
Chromatin Structure and Transcription: Transcriptionally active chromatin is not highly condense (it is not heterochromatin), but apparently in the 30 nm fiber condition. The nucleosomal histones must be removed for transcription. This may be by HMGN protein binding (similar to H1) causing decondensation, or by acetylation of histones.
Noncoding RNAs: These small RNAs can regulate gene expression post-transcriptionally, but they can also inhibit transcription by causing methylation of histones and thus the formation of heterochromatin. Methylation and Epigenetics: In vertebrates, the addition of a methyl group to a base (especially cytosine forming 5-methylcytosine) is often involved in transcriptional inactivation. Enzymes that add methyl groups to DNA are called DNA methyltransferases (DNMTs). Methylation patterns are generally preserved after replication by DNMTs. Methylation of DNA can turn off transcription by altering histones (causing chromatin compacting) and may also affect the binding of specific transcription factors. Unmethylated CpG islands (see "Part 3" video above) show hyperacetylation of histones H3 and H4, a deficiency in histone H1, with nucleosome-free regions promoters (transcription is on). (Methylation Video details will not be on the test--except the summary covered in the "Part 3" video above.) Epigenetics: 2010 article. An interesting example is seen in the rare phenomenon of gene imprinting (unusual event of methylations being NOT removed during zygote formation) as occur in the H19 gene. (Imprinting/methylation is an example of epigenetic changes to the gene: 2010 article) This gene produces only an RNA product with function yet unknown (but may related to cancer: 2010 article)(Methylation and resetting the genome) Excerpt from "Structure, function and evolution of CpG island promoters" by F. Antequera (details of the following will not be on the test--except the summary covered in the "Part 3" video above.)
CpG density defines two classes of RNA polymerase II promoters Despite the sequence diversity among promoters, genes transcribed by RNA polymerase II can be classified in two different and mutually exclusive groups according to the distribution of CpG dinucleotides across their 5' ends. In one class, the frequency of CpGs is the same as the genome average, which is roughly one every 100 nucleotides. This class invariably includes genes whose expression is restricted to a limited number of cell types... In contrast, the 5' end of the genes belonging to the other group is surrounded by a region ~1 kb long where the frequency of CpGs is approximately 10 times higher than the genome average. These regions were very appropriately called CpG islands ... and show such a conspicuous clustering of CpG dinucleotides that it can be readily detected by visual inspection of the CpG plot. The consistent association of CpG islands with the upstream region of many genes immediately suggested a possible involvement in transcriptional regulation .... What is so special about CpGs relative to the other 15 possible dinucleotides in DNA? CpGs are the sites where methylation takes place, and ~80 % of them are methylated at position 5 of the cytosine ring in humans and mice. Somewhat paradoxically, CpGs remain nonmethylated at CpG islands, despite their abundance, whereas the majority of the remaining CpGs scattered across the genome are mostly methylated. In addition to the lack of methylation, human and mouse CpG islands have a G+C content of 67 and 64 %, approximately, while the genome averages are 41 % and 42 %, respectively... These distinctive features of the CpG islands in terms of a lack of methylation and an elevated G+C content are accompanied by an equally distinctive chromatin organization. Chromatin analysis at global genomic level has revealed that CpG islands show the properties usually ascribed to "open" or "active" chromatin. This includes hyperacetylation of histones H3 and H4, a deficiency in histone H1, positioned nucleosomes and nucleosome-free regions that coincide with enhanced sensitivity to nucleases relative to bulk DNA... These properties highlight CpG islands as regions that are particularly well suited for direct access to DNA, which is consistent with their co-localization with the promoters of many genes. Previous studies based on the biochemical isolation of the CpG island fraction estimated approximately 45,000 and 37,000 in the human and mouse genomes, respectively..., although recent computational predictions have lowered these figures to about 27,000 and 15,500... Regardless of the absolute number of CpG islands in the genome, a more relevant issue is what kind of genes are associated with them, since only ~60% of all human genes are associated with CpG isands. This includes all the housekeeping genes--those expressed in all cell types--and about half of the tissue- specific genes... Transcription from CpG-rich and CpG-poor promoters How does methylation affect transcription from CpG island and non-island promoters? Since CpG islands are nonmethylated in sperm and remain consistently devoid of methylation in somatic tissues, regardless of the expression of the genes associated with them, it is unlikely that DNA methylation would play any role in their regulation. Exceptions to this rule are the CpG islands of imprinted genes, those in the mammalian X inactive chromosome and those associated with the MAGE genes that become methylated during normal mammalian development... Even in this case, methylation is not the primary inactivating signal but takes place at a stage when transcription has been switched off by other means... Transcription is strongly repressed upon unscheduled de novo methylation of CpG islands in cell lines and tumour cells, a phenomenon that occurs at high frequency in these situations but never in the organism under normal physiological conditions with the exceptions mentioned above... In contrast with CpG islands, CpG-poor promoters are methylated in sperm and are always associated with tissue-specific genes. A direct role of DNA methylation in the regulation of this class of promoters predicts a correlation between their methylation profile and their level of expression. Many examples and also several exceptions to this correlation have been described, suggesting that although DNA methylation affects gene expression, it is unlikely to play a general role as a transcriptional regulator... Detailed analysis of the kinetics of demethylation and gene expression has shown that in some cases where demethylation correlates with expression, the former fol-lows the binding of transcription factors rather than being a prerequisite for it. For example, binding of nuclear factor kappa B (NF-kB) transcription factor to an intronic enhancer of a k-chain gene is required for demethylation in B cells... That DNA methylation, despite its prominent presence in the mammalian genome, is not a general regulator of gene expression is not surprising since proper gene regulation takes place in invertebrates, many of whose genomes have a low or undetectable level of methylation.