BIL 250 - Lecture 11

Click HERE for your print copy.

Control of Gene Expression in Eukaryotes

As one might expect, control of gene expression in eukaryotes is more complex than in prokaryotes. There are more genes, and more cells, and in each cell, a different proportion of genes are activated and inactivated. This latter property determines the identity, morphology, physiology, etc. of the cell, as well as its part in the division of labor that is so critical to the function of multicellular organisms.

The mechanisms of gene expression control, while probably not homologous to those in prokaryotes, are at least analogous. And much of what we'll be doing today is comparing the mechanisms for gene expression control we've seen in E. coli to those found in various eukaryotes. Foremost among these will be Baker's Yeast (Saccharomyces cerevisiae, a fungus that's been so useful in the study of eukaryotic genetics that some have dubbed it "the E. coli of eukaryotes".

An Overview of Gene Expression Regulation: Prokaryotes vs. Eukaryotes

How are they similar, and how are they different?

I. How are Prokaryotic and Eukaryotic Gene Expression similar?

Both require the participation of regulatory proteins, some of which (transcription factors) attach directly to DNA sequences.

prokaryotes: activator and repressor proteins act on operators (DNA sequences just downstream of the promoter)

eukaryotes: activator proteins act on enhancer DNA sequences; repressor proteins act on silencer DNA sequences.
Enhancers can be found either upstream or downstream of the promoter. They can be found in introns, and can be located downstream of the polyadenylation site. The position of the enhancer has a profound effect on gene regulation.

In eukaryotes, as in prokaryotes, control of gene expression can occur at the level of

transcription
translation
protein activity (post-translation)

Unlike prokaryotes, eukaryotes (especially multicellular ones) must undergo an intricate sequence of gene expression events not only during regular metabolism, but also during development of cell types, tissues, organs and organ systems and general morphogenesis.

II. How do Prokaryotic and Eukaryotic Gene Expression differ?

Prokaryotes: About 1000 genes per genome

Eukaryotes: 25,000-30,000 genes per genome

Prokaryotes: Always On

In general, most of the genes of a bacterium are fixed in the "on" position. The majority of products are constantly made, and repressible systems are by far the most common gene regulation mechanisms in prokaryotes.

Eukaryotes: Mostly Off

off

Not only are genes turned "on" and "off" for specific functions during the organism's normal existence. Also, a cascade of gene activation and deactivation is responsible for early development and morphogenesis during ontogeny (embryo development). We'll focus on gene expression during ontogeny in the next lecture. For now, we'll stick to the every day function of a general eukaryotic cell.

Prokaryotes: Naked DNA

Prokaryotic DNA is largely unassociated with any proteins or other structures that could physically affect (or interfere with) gene expression.

Eukaryotes: Nucleosomes

chromatin

nucleosomes

Prokaryotes: One RNA polymerase

Prokaryotes use the same type of RNA polymerase to transcribe all genes.

Eukaryotes: Three RNA polymerases

RNA polymerase is a much larger molecule than prokaryotic RNA polymerase partly because

it must interact with more enzymatic "players" during transcription
it must house enzymatic machinery to perform functions such as co-transcriptional modification of the RNA transcript as it emerges from the enzyme.

Prokaryotes: No modification of mRNA transcript

What you see is what you get. Eukaryotes: Extensive modification of mRNA transcript

5' and 3' ends capped
introns removed and exons spliced

Prokaryotes:

Transcription and Translation take place simultaneously.

Eukaryotes:

Transcription in the nucleus; translation in the cytoplasm (e.r.).

Prokaryotes: Pribnow Box

Promoter is the Pribnow Box sequence.

Eukaryotes: TATA or Hogness Box

One promoter common to all eukaryotic proteome genes is known as the core promoter or basal promoter. This sequence of seven nucleotides--TATAAAA--is called the TATA Box or Hogness Box.

A transcription factor complex known as IID (TFIID) binds to the TATA box by means of a TATA-binding protein (TBP). Additional proteins in the complex interact with each other.

The entire complex acts as an activator.

gene enhancer sequences similar to the E. coli CAP site discussed previously may thus be involved in the Hogness Box control of eukaryotic gene expression.

Mechanisms of Eukaryotic Gene Expression Control

Transcriptional Control
Transcription can be

turned completely on or off
attenuated

There are three classes of control elements in eukaryotes:

The RNA polymerase II binding region (a.k.a. the core promoter)
cis-acting binding sequences that bind to proteins with RNA polymerase affinity, which in turn help bind RNA polymerase to a promoter.
enhancers and silencers

A cis-acting DNA element is a short DNA sequence that acts as a binding site for a protein that has an affinity for that specific sequence. The term "cis acting" means that the bound protein acts only upon DNA sequences on the same DNA molecule as the cis-acting sequence. (These are also found in prokaryotes. An example is the operator of the lac operon.) These elements:

include short consensus sequences
are usually located within 200 bp upstream of the transcription initiation site
can be included in a promoter or an enhancer
have an affinity for a specific protein that binds to it, serving as a regulator

A trans-acting DNA element is a DNA sequence that codes for a protein (a trans-acting factor) that controls the expression of a gene at a separate location by binding to its cis-acting element. Trans-acting factors:

are enzymes that interact with RNA polymerase
bind to RNA polymerase to stabilize the initiation complex
may bind to a few promoters and serve as positive regulators

A trans-acting factor can affect the expression of genes located on separate chromosomes. In order to control transcription, a regulatory protein must have

a domain (the protein's DNA binding site) that recognizes and binds to a gene's regulatory sequence
a domain that interacts with the domains of RNA polymerase active sites or proteins that interact with RNA polymerase
a domain that interacts with other proteins bound to regulatory sites for cooperative regulation
a domain that can influence the condensation of chromatin
a domain that can detect environmental/physiological conditions in the cell, and act as a trigger to appropriate gene expression in that environment.

In eukaryotes, enhancers and silencers are DNA sequences reminiscent of the CAP site in the E. coli lac operon:
- enhancers are located upstream or downstream of the promoter region, and have an affinity for specific activator proteins that make the initiation region more attractive to RNA polymerase.
- silencers are located downstream of a promoter, and have an affinity for specific repressor proteins that inhibit attachment of RNA polymerase to the initiation region.
(Recall that eukaryotes have three RNA polymerases, RNA p'ases I, II and III. RNA p'ase II is responsible for manufacture of all mRNA, so when we discuss proteome genes, the RNA polymerase involved is RNA polymerase II.)
An enhancer DNA sequence (or positive regulatory element) turns a gene ON. The mRNA-binding protein for which it has an affinity is called a positive regulatory protein, or activator. When the activator is bound to the enhancer, RNA polymerase is more highly attracted to the gene.
A silencer DNA sequence (or negative regulatory element) turns a gene OFF or reduces its rate of transcription. The mRNA-binding protein for which it has an affinity is called a negative regulatory protein, or repressor. When the repressor is bound to the silencer, RNA polymerase cannot attach and transcribe the gene.

Note that the genes encoding activators and repressors can be relatively far away (e.g., located on different chromosomes) from the enhancers and silencers they affect (trans-acting), or they may be located on the same chromosome (cis-acting).

It is believed that the action of enhancers and silencers involves their ability to bind downstream segments of DNA, causing them to loop upon themselves. This, in turn, can physically affect the rate of transcription. Here's an OVERVIEW.

Transcriptional control also can be exerted by the activity ofDNA-binding domains, proteins which bind directly to the DNA to affect its transcribability.

Nucleosomes and Eukaryotic Gene Expression

Some data suggest that a gene not actively being transcribed may have histones associated with its TATA box. The precise proximity of a gene's promoter to the nucleosome histones may affect how readilyi it is transcribed. But to fully understand this, let's recall the structure of chromatin.

Histones may block attachment of RNA polymerase to the TATA box, which makes transcription impossible.

Histone/nucleosome winding of the DNA molecule may change with the gene activity in a particular cell or tissue, or with the developmental stage of a particular cell.

When a gene needs to be transcribed but is unavailable because of histone proximity, moving the DNA strand can free the promoter and associated regulatory sequences so that the gene can be expressed.

Conversely, if an active gene associated with a nucleosome must be turned off, then moving the DNA so that repressors can attach to the gene's silences can also be achieved by rewinding the DNA (in the next cell cycle) on the nucleosomes in a different way.

Such changes are termed chromatin remodeling.

Histone Remodeling
Until the past couple of decades, histones were considered too similar and simple across species to be involved with gene expression. This view has changed.

It is now known that the eight proteins comprising the nucleosome "spool" each have their amino-terminal ends waving in the breeze beyond the tightly coiled main protein structure.

These histone tails may contain lysine residues that are readily modified by enzymes that attach acetyl and methyl groups to them after the histone has been translated, and even after it's already part of a nucleosome.

Other enzymes can phosphorylate or ubiquitinate the tails.

In all, at least 150 different modifications to the histone tails are known, and there may be many more.

These modifications are often reversible, and may be involved in a particular gene's ability to be expressed or shut down in different generations of the cell.

Covalent bonding of functional groups to the lysine residues modifying the histone tails can take many forms, and can carry a tremendous amount of information (there are 44 lysine residues per tail, each one available for modification). These modifications comprise what has been named a histone code.

The precise type of modification of a histone tail may be associated with very fine-tuned control of gene expression which is not yet fully understood.

For example, in some species, active genes have many histone tails modified with acetyl groups (they are hyperacetylated), whereas inactive genes have few acetyl groups on their histone tails (they are hypoacetylated).

The level of acetylation of the histone tail can cause

DNA to slide along the nucleosome and wind in a different pattern than the previous "incarnation"
Histone modification may affect the affinity of associated DNA for regulatory proteins.

Enhancer Activity

In eukaryotes, the participation of many regulatory proteins in a single gene activation event allows for greater variety in the gene's level of transcription.

Multiple regulatory proteins bound to binding sites in an enhancer can form a large, complex enhancesosome that has varying affinity for RNA polymerase, depending on its size and nature.

Highest levels of transcription occur only when a particular number of regulatory proteins are bound in a very specific way, and this--in turn--depends on the physiological environment of the cell.

The enhanceosome can both recruit additional co-activators and facilitate chromatin remodeling.

Post-Transcriptional Control of Gene Expression

Recall: an mRNA consists of a 5' UTR (untranslated region, or leader), an ORF (open reading frame, a.k.a. polypeptide coding sequence) and a 3' UTR (untranslated region, or trailer).

mRNA Lifespan

Certain sequences in the 3' UTR may promote very rapid degradation.

Other UTR sequences may not promote degradation, but can lead to lower levels of translation.

Mutations in these sequences can result in higher-than-normal translation of proteins encoded in the ORF portion of the gene, which is how they were discovered.

Hence, regulatory information is encoded not only in the protein-coding regions, but in the non-protein coding regions of mRNA

mRNA Modification

Modification or non-modification of the primary mRNA transcript may be used as a form of post-transcriptional control.

The spliceosome (remember the snurps) may serve as a form of post-transcriptional modification, altering gene expression via alternative splicing.

When associated with the snRNPs, mRNA cannot leave the nucleus (it can't bind and pass through the nuclear pores).

When intron-bound snurps release the mRNA transcript, the spliced mRNA can exit the nucleus, but the introns can't.

If the number of transcripts actually allowed to move out of the nucleus and into the cytoplasm is regulated, transport control is said to be in effect.

Post-Translational Control

Once a protein is made, how is its activity controlled?

activity can be controlled if the protein is allosteric.
(since proteins generally have a longer lifespan than mRNA in the cell, it is to the cell's advantage to have some sort of protein manaagement mechanism in place)
often, the end product of an operon binds to the allosteric site of the operon's enzyme, changing its affinities and activity. (e.g., tryptophan)
when the concentration of the end product decreases, it diffuses off the bound allosteric proteins, allowing that protein to once again become active.

Protein Lifespan
Proteins can last in the cell from a few seconds to longer than a cell cycle. How does the cell know which ones not to break down?

The N-end Rule: The specific aa at the N-(amino)-terminal signals the polypeptide's lifespan to various proteases in the cell.
(e.g.- arg lasts about 2 minutes; met lasts up to 20 hours)
The PEST Hypothesis: The # of regions rich in the four amino acids proline (P), glutamate (E), serine (S) and threonine (T) signal lifespan to the various proteases. Proteins rich in these regions have lifespans of 2 minuts or less; in almost all proteins with lifespans over 20 hours long (some as long as 220), the PEST sites are lacking.

And now for something almost, but not quite completely, different.

Genomic Imprinting

Mendel saw no difference in the inheritance of the traits he studied in peas whether the parent was male or female. Recently, however, twoto three dozen traits have been discovered that do appear to be expressed differently, depending on whether an allele is inherited from the mother or the father.

Genomic imprinting, or parental imprinting occurs when a gene inherited from either the father or the mother has been permanently inactivated in that parent, and is and passed on to offspring in the same inactivated condition.

The inactivated gene has not mutated: the DNA sequence is normal.

Inactivation is caused by the methylation of DNA in the regulatory sequences of the inactivated gene.

In general, highly methylated genes (usually methlyated at cytosine nucleotides) are inactive (not always, but usually).

During gamete formation in mammals, DNA is methylated differently between the sexes in genes that exhibit imprinting. This results in the silencing of an allele in either the male (paternally imprinted) or female (maternally imprinted) gamete--depending on the gene/trait.

maternal imprinting - the allele inherited from the mother is inactive. Only the allele inherited from the father is expressed.
paternal imprinting - the allele inherited from the father is inactive. Only the allele inherited from the mother is expressed.

Because only one allele of the imprinted gene is expressed, inheritance of such a gene is said to be monoallelic.

This silencing is normal, and is one mechanism by which dosage of gene products can be managed by the organism.

For example, in female mammals, one X chromosome is highly methylated and formed into a Barr Body. This leaves only one copy of each X-linked gene active in each cell, so both male and female mammals manufacture the same amount/concentration of any product on the X chromosome.

(Don't be misled by this example: most imprinted genes are on the autosomes.)

In any given species, certain traits are either maternally or paternally imprinted every time under normal circumstances, and are inherited that way in ever generation.

The Insulin-like Growth Factor Gene (Igf2) in Mice
One of the first imprinted genes discovered was the Igf2 in mice. The product of this gene is required for normal growth in the embryo, but only one allele--the paternal one--is expressed. The maternal copy of the same gene is highly methylated and silenced.

The Igf2 gene can mutate to become inactive in either gamete:

ovum: maternal mutation results in non-functional allele
sperm: paternal mutation results in non-functional allele

Because the maternal allele is silenced anyway, the inheritance of a mutant maternal allele will have no effect: the offspring grow to normal size.

But if a mutant paternal allele is inherited, the offspring lacks even one expressed copy of the gene, and will express dwarfing.

Human disorders involving erroneous imprinting inheritance

Heritable alteration of the DNA that does not involve a change in the nucleotide sequence is known as epigenetic inheritance. The alterations themselves (methylation of nucleotides; alternative histone wrapping) are called epigenetic markers.

Inheritance of imprinted traits is monoallelic, since only one version of the allele is ever expressed in any given individual. (Hemizygosity--the state of a gene being expressed as a result of only one allele's activity, as in X-linked traits in male mammals--can also be considered a form of monoallelic expression.)

The Clone Question
When Dolly the Sheep, the first cloned mammal, was introduced to the world in 1996, the scientific community was amazed. It had been believed that cloning of mammals was impossible because of the need for maternally and paternally imprinted homologous genes functioning normally to allow embryo development. (Dolly was derived from a somatic nucleus.)

Why the imprinting phenomenon has not interfered with successful mammal cloning in every case is still a mystery. But for every cloned mammal that survives to birth, thousands are aborted or simply fail to develop.

Dolly herself died at a very young age, and was exhibiting several disorders normally found in much older sheep. So cloning of mammals is still far from perfect, and the role of imprinting in normal embryo development is still not fully understood.

Chromatin Structure and Inheritance: Position Effect

When we study how genes can mutate, we will discuss a type of mutation in which a piece of a chromosome breaks away from its usual position and takes up residence at a different place on the same chromosome, or sometimes even on a different chromosome. Such mutations are called translocations. In a different type of mutation, a piece of chromosome breaks out, flips over, and reinserts in what is termed an inversion mutation. In any mutation involving rearrangement of chromosome segments, the genes carried on the moved segment will take up new positions in the chromosomes, and this can sometimes affect their expression. This is known as position effect.

In some translocation mutations, the new position of the gene can affect the rate of its transcription and translation.

For example, if a highly-transcribed gene is translocated to a region close to tightly coiled heterochromatin, it can sometimes be partially engulfed by that heterochromatin.

Heterochromatin coiling can sometimes inactivate genes, effecting a form of epigenetic silencing.

Position Effect Variegation Drosophila

Genes contained within stretches of condensed chromatin (heterochromatin) are usually transcriptionally inactive.

Heterochromatin is defined cytologically as the densely staining chromatin seen in the interphase nucleus. Many of the DNA sequences in heterochromatin are not transcribed. (The rRNA genes are a notable exception; they are abundantly transcribed! But most heterochromatic DNA is not.)

Several lines of evidence support the association of heterochromatic configuration with gene silencing. One of these is illustrated by position effect variegation (PEV).

If a gene is translocated from a position on the DNA which is highly transcribed to an area where transcription is muted, that gene may be transcribed at an abnormally low rate. If a gene is translocated to an area close to heterochromatin, the gene itself may not always be included in the heterochromatic complex, since its exact start and ending may differ slightly from cell to cell. This causes a phenotypic variation among cell fields in a tissue or population known as position effect variegation (PEV).

A classic example of PEV is seen in Drosophila. In some cases, a chromosomal inversion flips a segment of the X chromosome around so that the gene for eye color (which codes for red pigment in wild type (w⁺) comes into close apposition to a region of heterochromatin.

In cells where the gene ends up within the heterochromatin, it is silenced. The ommatidia derived from cells with silenced pigment genes will lack pigment, and appear white/colorless. In other cells, however, the gene is not quite included in the heterochromatin, and is expressed. The result is an eye that is variegated red and white.

Whatever other mutation might affect the eye color ("apricot", "cinnabar", etc.), the variegation will still affect the patterning of pigment deposition.

Bottom Line: Heterochromatin does not have a precise boundary, and its exact position in any given cell may differ slightly in a population of cells. So in some cell lines of the Drosophila eye, the red pigment will be normally expressed, whereas in cell lines in which the gene is included in the heterochromatin, a white (colorless) patch will result.

Expression of this trait does not depend on nucleotide sequence, but on DNA packaging.

Dosage Compensation: Inactivating an Entire Chromosome

In animals that determine sex with heteromorphic chromosomes, one sex or the other has two of the larger, more information-dense chromsomes and the other has two dissimilar sex chromsomes.

In mammals, which have about 1000 genes on the X chromosome, females have the potential to produce twice the amount of X-products as males. This would very likely cause problems. The mechanism that prevents this imbalance is known as dosage compensation.

The inactivated X chromosome (Barr body) has many of the epigenetic marks of heterochromatin, including hypermethylated cytosines and methylation of histone 3 (H3) at its lysine 9. Inactivated X chromsomes exhibit epigenetic inheritance: A cell's daughter cell will have all the same X chromsomes inactivated, though in any given organim one X or the other is randomly inactivated in any cell line. This results in mosaic expression of the encoded trait: the active allele in any given cell line will be expressed, even if it is the recessive allele.

Not all species that have XX females and XY males sequester an X as a Barr Body. For example:

Caenorhabditis elegans: Male X chromosome transcribes twice the amount of either of the single X chromosomes in the female (hypertranscription)
Drosophila melanogaster: The X chromosomes in the female each are transcribed at half the rate of the male's single X chromosome (hypotranscription)

Epigenetic Tales

As mentioned previously, Epigenesis is a heritable modification in gene function that occurs without change in the actual base sequence of the DNA code.

Paramutation: Anthocyanin pigment in corn plants

paramutation - the genetic activity of a particular allele is reduced when coupled with inheritance of a paramutagenic allele at the same locus.

Anthocyanin pigment is coded by an allele called "B"
A recessive mutant (b) results in a null allele: no pigment is produced
A paramutagenic allele at this locus (B') causes reduced pigment production.
BB x B'B' F1 heterozygotes are able to make only the reduced amount of pigment encoded by the B' allele. The BB' hybrid is indistinguishable from the B'B' homozygote.
So does that mean B' is dominant to B?
It's more complex than that.
If you cross the F1 plants, all offspring exhibit the weakly pigmented phenotype--and all subsequent generations exhibit only the B' phenotype.
The B allele is apparently permanently "crippled" in its activity by being inherited along with the B' allele. It has paramutated.
The exact mechanism is not fully understood, but may involve gene silencing by siRNA transcribed from the mutant pigment gene.

Re-thinking a Classic Tale of Natural Selection: Epigenesis in Pepper Moths?
Recall the tale of Biston betularia, the Pepper Moth, still a "poster child" for teaching the mechanism of natural selection.

...but is the classic tale really what happened in Nature?

Some researchers suspect that industrial melanism is actually induced by environmental factors and becomes heritable, making it a case of epigenesis. The exact mechanism of the pigment genes activity being altered in a heritable way is not yet known.

Epigenesis may simply be cases of gene expression gone (somewhat) awry. In most cases, the exact mechanism of epigenetic change that produces phenotypic change is not well understood.

Could epigenesis be a source of exaptation? An epigenetic character might allow the eukaryote exhibiting it to quickly alter the genome (using pre-existing material, and merely "repackaging" it) so that offspring are phenotypically different and (in some cases) better suited to a changed environment than the parental generation.

Stay tuned.

DNA Configuration and Eukaryotic Gene Expression

DNA can exist in one of several isomers.

The relationship of DNA configuration to gene expression is still not clear. But some forms (notably, B-, A- and Z-DNA are believed to be the forms in which DNA can be actively transcrbed.

B-DNA

This is the most common DNA conformation found in vivo

It is relatively narrow and more elongate than A or Z forms

phosphodiester backbone runs in a smooth curve

The base pairs are almost perpendicular to the axis of the double helix.

The wide major groove easily accessible to various polypeptides that affect level of transcription.

Small kinks along the length of a B-DNA strand may represent areas where a transcription factor is bound.

B-DNA form is believed to be the conformation of actively transcribed genes.

A-DNA

RNA-DNA hybrids and double-stranded RNA are most often found in this form.

the helix is shorter and of greater diameter than in B-DNA

phosphodiester backbone runs in a smooth curve

base pair H-bonds are slanted relative to the axis of the double helix

the major groove has become deep and narrow, making it less accessible to proteins that might affect transcription.

genes in this conformation, if transcribed, may be transcribed at a lower rate than genes in B-DNA configuration.

this form has not been observed in living organisms, and may be a laboratory artifact.

Z-DNA

helix spirals to the left, instead of the right, as it does in B and A DNA

this isomer can form in vivo, but function is not fully understood.

helix is more narrow and elongate than either A-DNA or B-DNA

major "groove" is so shallow as to no longer form a groove.

minor groove very narrow

base pairs nearly perpendicular to axis of double helix

phosphodiester backbone forms a zig-zaggiung pattern because of unusual sugar configuration.

some research suggests that Z DNA configuration is necessary for a gene to be transcriptionally active.

Z-DNA is stable in vitro when its cytosines are methylated (the 5-carbon H is replaced by CH₃)

methylation protects DNA from the action of endonucleases.

Genetic Response of Multicellular Organisms to Relatively Rapid Environmental Changes

How are certain cellular products produced so rapidly in response to sudden environmental changes?

Hormones may behave as regulators in multicellular eukaryotes in which short-term gene activation is necessary.

by definition, a hormone is a substance produced in one area, and used in a different area, to which it has been actively transported.
example: various steroid hormones (there are, of course, a number of different steroids) bind to specific receptors in the cytoplasm (SHR = "steroid hormone receptor). This receptor/steroid complex then binds directly to the DNA, regulating transcription.
another example: other hormones may bind to specific receptors in the plasma membrane. This binding activates adenyl cyclase (also embedded in the plasma membrane) to convert ATP-->cAMP. cAMP affects the binding of the hormone to the DNA, which can affect transcription rate.