1
2
3
The regulation of gene expression during development may occur at different levels.
Most of the regulations occurs at the transcriptional levels, however, the later
(“downstream”) type of posttranscriptional regulations e.g. regulation of translation,
posttranslational modifications of the proteins, their transport etc. are also
important.
Recently, very important type of posttranscriptional regulation was found to be
mediated by small RNAs. The molecular mechanisms and their importance for the
diverse developmental processes in eukaryotes with emphasis on the development
regulation in plants will be discussed in the second part of this lesson.
The regulatory events are diagrammed here as a linear sequence, which, however, is
rarely the case. More often, the individual regulatory events occur in a maze of
networking interactions.
4
5
Regulation of the chromatin structure represents one of the very basal gene
expression regulatory levels. Chromatin is a substrate for DNA-dependent RNA
polymerases that transcript the DNA encoded information into the “words and
sentences” of RNA.
Regulation of chromatin structure and its accessibility to DNA-dependent RNA
polymerases depends on many factors, one of the most important is the regulation of
chromatin binding to nucleosomes and chromatin methylation.
Regulation of chromatin interaction with histones, the positively charged proteins
forming the core of nucleosomes, is performed via modification of acetylation status
of the N-terminal portion of histones, especially histones H3 and H4. This occurs via
action of histone acetyl transferases or histone deacteylases.
6
Modification of the chromatin methylation is performed via DNA methyltransferases.
Interestingly, there is difference in the methylation in animals and in plants.
In animals, the methylation occurs mostly on the cytosine that occurs next to
guanosine (the sequence is denoted as CpG). In mammals, 60-90% of all CpGs are
methylated.
In plants, cytosines are methylated both symmetrically (CpG or CpNpG) and
asymmetrically (CpNpNp), where N is any nucleotide.
Methylation status is usually “reset” in the zygote and is reconstituted during
development again. E.g. the methylation is very low in the mouse embryo at the
blastula stage, however, DNA derived from later stages when organogenesis is
initiated is substantially more modified by methylation.
DNA methylation also stably alters the gene expression pattern in cells such that cells
can "remember where they have been"; in other words, cells programmed to be
pancreatic islets during embryonic development remain pancreatic islets throughout
the life of the organism without continuing signals telling them that they need to
remain islets.
7
DNA methylation is involved in the genomic imprinting, i.e. the genes originating
from both parents are often diversely methylated, which results into differential
expression of parental genomes (for the importance of the imprinting in the parental
conflict and epigenetics, see the lecture “Bi0580 Developmental genetics” by prof.
Vyskot).
Up to know it is not clear how methylation regulates transcription. Possibly,
methylation status affects chromatin configuration or binding general repressor
factors.
7
8
Regulation of transcription occurs via specific interaction of both general and tissue
specific transcription factors (TFs) with promoter and/or enhancer sequences.
The scheme above shows simplified subsequent formation of the complex of TFs
involved in the regulation of transcription. Interaction of general TFIID with the TATA
box induces distortion of the DNA structure (see the next slide).
9
Induction of structural changes upon interaction of TFIID with DNA. This may be
important for the assembly of other TFs involved in the formation of transcription
initiation complex.
This change of confirmation provides a kind of “signature” that is recognized by other
proteins and NA polymerase to recognize the proper binding site. However, there are
also TATA box-less promoter, where probably other types of “signatures” occur.
10
The scheme showing the formation of the transcription initiation complex and the
interaction of both positive (open symbols) and negative (solid symbols) factors.
These proteins bind to the regulatory sequences that might be hundreds or even
thousands of base pairs away from the promoter. These protein interact with each
other and with the RNA polymerase, integrating thus many signals into a “yes” or
“no” response of the basal promoter, i.e. the region adjacent to the TATA box and
recognized by the RNA polymerase.
The individual positive or negative factors are complex and their activity might be
regulated by their phosphorylation status or via their interaction with other proteins
(i.e. monomeric or dimeric) etc..
11
There is a whole family of transcription activating factors (TAFs) that interact with
signalling molecules, e.g. steroid hormones, thyroid hormones or retinoic acid and in
a response to the signal translocate to the nucleus and activate transcription.
One of the type of TAF are leucine zipper or bZIP type TAFs. These TAFs are dimeric,
with leucine-rich hydrophobic face formed by the Leu that occurs every 7th aa.
That allows the factor to take the proper configuration, which provides the dimer
with the ability to bind DNA via charged aa.
12
An example of the “microprocessor”-like acting promoter is a promoter of the
endo16 gene from the sea urchin.
Endo16 is a single copy gene that codes for the RGD-containing calcium-binding
protein of the cells of invaginating archenteron during gastrulation.
As visualized in experiments, the Endo16 protein may be an adhesion molecule,
involved in the gastrulation of the embryo (Nocente-McGrath C et al, 1989).
There have been identified several gene regulatory modules in the endo16 gene that
have positive or negative regulatory role. These modules were identified via
formation of deletion mutants of the transcriptional fusions with reporter gene.
The analysis has revealed that the module A has a positive function and must interact
with its cognate TAFs for transcription to occur.
Module G enhances the expression when the A and B are active.
C, D, E and F are responsible for the specificity of the expression of endo16 during sea
urchin development.
Each of the modules has several protein interaction sites, some of them general,
other unique. Site for the protein SpGCF1 is present in many modules and is probably
responsible for looping of chromatin, allowing thus bringing of distal regulatory
13
modules close to the basal promoter.
This type of regulation, i.e. based on the different activities of diverse regulatory
sequences is sometimes called combinatorial and is common for development of
many living creatures.
In the combinatorial type of regulations, some modules may act synergistically, some
of them antagonistically, some may have both positive and negative roles (e.g. the
module B, see the figure). This variability allows very precise and responsive
regulations towards changing environmental conditions.
13
An example of the combinatorial gene regulation is the regulation of β-globin type of
hemoglobin chains of humans.
As discussed in the Lesson 5, the type of hemoglobin produced by the fetus changes
during development. The hemoglobin present in the liver-produced hemoglobin is
composed of two α- and two β-type chains. The β-type hemoglobin chains are of
several developmental types, produced by ε, γ1, γ2 and β (in this order). In addition,
there is minor adult type of β-type hemoglobin, called δ globin.
The genes for the β-type chains are aligned on the chromosome in the order, in which
they are expressed during development (se the figure).
For the expression of individual cell types is distinctive an upstream regulatory
sequence called locus control region (LCR). LCR is located about 50 kbp away from the
most proximal ε gene.
The LCR structure is different in erytrocyte precursor cells in comparison to other
cells that could be demonstrated by the changes in the sensitivity to low
concentrations of DNase, suggesting low amount of nucleosomes bound.
For the expression of the particular genes, the interaction of their regulatory
sequences with LCRs is necessary. Because of LCR can interact only with one
regulatory sequence at a time, only one type of genes for the particular β-type chain
is activated (the first interaction of LCR with ε gene, which is later in development
14
replaced by the other one, is shown by the double-headed arrow).
The underlying molecular mechanisms of the specific pattern of the LCR movement
from the most proximal towards the most distal gene cannot be satisfactory
explained.
Probably, acetylation of H3 histones might play a role and possibly, other genes
outside of the β-type chain family are involved in the regulation of LCR activity. That
seems to be confirmed by the identification of other human genes with similar
structure, suggesting common regulatory mechanisms via LCRs.
14
15
Posttranscriptional modifications takes place after transcription and these are a
matter of regulation that has developmental consequences, too.
The most important posttranscriptional modification in eukaryotes in splicing of
introns. Immediately aftre transcription, the hnRNA is polyadenylated and interacts
with proteins and small nuclear RNAs (snRNAs) that are part of small nuclear
nucleoproteins. These interact with specific recognition sites and allow splicing of
introns and connection of exons into final mRNA.
16
An example of the developmental importance is an alternative, sex-specific splicing of
DOUBLE SEX (DSX) gene in Drosophila. DSX is a last member of the cascade involved
in the sex specification of Drosophila.
In females, exons 1 through 4 are connected, while in males, exons 1-3, 5 and 6 are
connected into final mRNA that is translated.
Male form of DSX blocks the female-type of differentiation and vice versa, the female
form of DSX blocks the male-type of differentiation.
17
18
Initiation of translation, elongation of the peptide chain and protein release are all
the steps being regulated during development.
An example is translation initiation after fertilization e.g. in the sea urchin eggs. On
the figure above, there is shown a dramatic increase of polyribosomes formation
(polyribosome is a string of ribosomes on mRNA) and increase in the protein
synthesis after urchin egg fertilization.
The polyribosomes form on the preexisting mRNA, originating from the egg.
19
3’ untranslated region (3’ UTR) of mRNA could interact with specific proteins. This
interaction then leads to the inhibition of translation initiation and the process is
called masking of mRNAs.
This mechanism occurs in many organisms, including Xenopus, sea urchin, mouse and
Coenorhabditis.
The masking regulates also polyA tail length. Once unmasked, the mRNA could be
intensely polyadenylated, leading to increase of the polyA tail.
The length of the polyA tail also affects translation initiation and stability of mRNA.
On the figure above, there is shown a mechanism of the sex determination in
Coenorhabditis, where posttranslational regulation of expression plays important
role.
Most C. elegans have a female body but are hermaphroditic, producing both sperm
and eggs at different times.
20
The first germ cells to differentiate in the nematode become sperm, which are stored
in the uterus for later use. After the fourth molt (from larva to adult), the germ cells
cease making sperm and begin to make eggs. These eggs will eventually become
fertilized by the stored sperm.
The process determining which path the germ cell follows—to sperm or to egg—
depends on the translational repression of different messages.
The initiation of sperm formation is achieved by the repression of the tra-2 message.
The Tra-2 protein is essential for the development of eggs and female body cells, and
repression of tra-2 mRNA translation in germ cells causes them to become sperm.
The 3' UTR of this message contains two regions of 28 nucleotides, each of which
appears to bind a putative repressor protein that is synthesized during the larval
stages associated with spermatogenesis. If these regions are mutated, the translation
of the tra-2 mRNA is not repressed, no sperm is made, and the nematode is
functionally female instead of hermaphroditic .
The switchover from spermatogenesis to oogenesis also requires suppressing the
translation of the fem-3 gene through its 3' UTR.
The Fem-3 protein is critical for specifying male body cells and sperm production.
Transcription of the fem-3 gene is inhibited by Tra-2 protein, but the repression of
existing fem-3 messages is also needed. This translational repression appears to be
effected by the binding of a translational inhibitor by the 3' UTR of the fem-3 mRNA.
Thus, the initiation of spermatogenesis in hermaphroditic nematodes and the
transition from spermatogenesis to oogenesis appears to be regulated by
translational repression through the 3' UTR.
20
21
Messenger mRNA may be localized to the specific positions in the cell. Specific mRNA
localization was discussed in the Lesson 2, when during syncitial stage of the
Drosophila embryo formation, BICOID mRNA is located to the dorsoanterior portion
of the developing embryo.
Again, as just discussed, 3’ UTR plays an important role in the tethering of BICOID
mRNA.
Transcription of the bicoid message occurs in the nurse cells (during stages 4 and 5 of
oogenesis) and the mRNA is transported immediately into the oocyte.
BICOID mRNA is localized to the anterior end of the oocyte with the aid of STAUFEN
protein. STAUFEN recognizes regions of double stranded mRNA, which are present in
arm III of the BICOID 3' UTR (see left-hand figure). The STAUFEN-BICOID mRNA
complex associates with the minus ends of the microtubules, thereby localizing
BICOID to the anterior.
On the right-hand figure, there is a secondary structure of the BICOID mRNA 3'UTR.
22
STAUFEN protein binds to specific sequences of arms III, IVb and Vb shown here in
blue lettering.
Helix III is essential for the binding to STAUFEN protein and accumulation in the
anterior of the oocyte. The insert shows the interactions between helices III of two
BICOID mRNAs. This structure is recognized by the STAUFEN protein.
22
Another example represents localization of NANOS mRNA (see the figure above).
NANOS mRNA is localized posteriorly and acts as a translational suppressor of the
HUNCHBACK protein (recall the Lesson 2 and next slides).
HUNCHBACK is a TF that further acts as a regulator of other donwnstream genes, e.g.
KRUPPEL (KR), KNIRPS (KNI) and GIANT (GT) (see next slides).
23
The levels of BCD and NANOS at the time of fertilization.
24
After fertlization, the proteins are translated. NANOS inhibits HUNCHBACK
translation posteriorly.
25
Later in development, HUNCHBACK is induced anteriorly by BCD.
The gradient of HUNCHBACK regulates the position of expression of KRUPPEL
(KR), KNIRPS (KNI) and GIANT (GT).
HUNCHBACK upregulates KR and KNI, where KR requires higher than maternal
level of HUNCHBACK .
GT is repressed by HUNCHBACK.
26
NANOS mRNA requires for its function in a translational repression of HUNCHBACK
another protein, PUMILO.
PUMILO binds to the 3’UTR of HUNCHBACK mRNA and allows thus binding of NANOS
and subsequent deadenylation of the HUNCHBACK and its degradation.
Localization of NANOS mRNA is under control of several genes (e.g. OSKAR, STAUFEN,
VALOIS, VASA and TUDOR). The proteins of these genes bind top the 3’UTR of NANOS
mRNA and allow its posterior localization.
Interestingly, the localization of NANOS mRNA is predominantly at the posterior pole,
but not exclusively. There is still some NANOS located in the middle or even anterior
portion of the embryo. However, only posteriorly located NANOS is translated.
That is ensured by the action of protein SMAUG. SMAUG binds to the specific
sequence in the 3’UTR of NANOS mRNA and inhibits its translation. The localization of
mRNA to the posterior domain relieves this translational repression (see the figure).
27
28
Autocatalytic cleavage and esterification represent another example of
posttrransaltional modifications during development.
HEDGEHOG (HD) protein from Drosophila is important signalling ligand involved in
the formation of boundaries from body segments and other patterned elements
during development.
HD contains the signal peptide that directs the HD protein to the ER and later to the
Golgi apparatus. After leaving the Golgi apparatus, the protein is cleaved into two
parts: The N-terminal (19 kD) and C-terminal portion (25 kD).
The N-terminal portion is later modified via esterification and cholesterol is added to
its C-terminus. Only the N-terminal part acts as an active protein that bids to its
receptor.
Modification of HD via cholesterol binding has an important role in the regulation of
the HD diffusion from the HD expressing cells. Genes like DISPATCHED or TOUT VELU
do affect mobility of HD that is important for HD-mediated signalling.
29
Similar function of cholesterol as in case of HD protein seems to be applied also in
the case of its vertebrate homologue, SONIC HEDGEHOG (SHH), discussed previously
(see Lesson 4) during the formation of dorsolateral axis in the neural tube
development.
Notochord acts as a source of the SHH production that leads to the differentiation of
floor plate in the ventral portion of the neural tube while motor neurons differentiate
in the dorsolateral poprtion, where SHH concentration is much lower.
Probably, the cholesterol modification is involved in the regulation of SHH
distribution.
30
31
RNA interference or RNAi, is a mechanism of targeted posttranscriptional regulation
of gene expression, what is also called posttranscriptional gene silencing (PTGS).
The molecular mechanism o PTGS was discovered in experiments with gene silencing
in Coenorhabditis elegans. It was found that surprisingly both sense and anti-sense
mRNA when injected into Coenorhabditis was able to silence the gene of interest.
One of possible hypothesis predicted that contamination during in vitro transcription
might lead to the formation of dsRNa that would be responsible for the observed
phenomenon.
The hypothesis was confirmed by the introducing dsRNA in Coenorhbditis and it was
found that the dsRNa induced gene silencing by about an order or two better then
anti sense mRNA.
Importantly, in then forward genetic screens were identified mutants that were
affected in the dsRNA-induced gen silencing. This suggested that the mechanism of
dsRNA-mediated gene silencing is an intrinsic regulatory mechanism for regulation
of gene expression.
That finding triggered an explosion of subsequent discoveries and led to the
32
identification of many members of the entire pathway.
32
It has been found that dsRNA might be either an intermediate or a trigger in PTGS.
In the first case, dsRNA is formed by the action of RNA-dependent RNA polymerases (RdRPs), which use specific transcripts as a template. It is still
not clear, how these transcripts are recognized, but it might be e.g. abundant RNA that is a result of viral amplification or transcription of foreign
DNA.
It not clear, how the foreign DNA might be recognized, possibly, lack of bound proteins on the foreign “naked” DNA and its subsequent
“signature” (e.g. by specific methylation pattern) during packing of the foreign DNA into the chromatin structure might be involved.
The highly abundant transcripts might be recruited to the RdRPs by the defects in the RNA processing, e.g. lack of polyadenylation.
In the case when dsRNA is a direct trigger, there are two major RNA molecules involved in the process: Short interference RNA (siRNA) and micro
RNA (miRNA), both encoded by the endogenous DNA.
These two functionally similar molecules differ in their origin:
siRNAs are dominantly product of the cleavage of the long dsRNA that are produced by the action of cellular or viral RdRPs. However, there are
also endogenous genes, e.g. short hairpin RNAs (shRNAs) allowing production of the siRNA (see the figure).
miRNAs are involved in the developmental-specific regulations and are product of transcription of endogenous genes encoding for small dsRNAs
with specific structure (see the figure).
In addition to siRNAs, there are trans-acting siRNAs (tasiRNAs) that are a special class of siRNAs that appear to function in development (much like
miRNAs) but have a unique mode of origin involving components of both miRNA and siRNA pathways.
Developmental regulations via miRNAs are more often used in animals then in plants.
The dsRNAs of all origins and pre miRNAs are cleaved by DICER or DICER-like (DCL) enzyme complexes with RNAse activity, leading to production
of siRNAs and miRNA, respectively.
These small RNAs are of 21-24 bp long and bind either to RNA-induced transcriptional silencing complex (RITS) or RNA-induced silencing
komplex (RISC).
33
In siRNA and miRNA biogenesis, DICER or DICER-like (DCL) proteins cleave
long dsRNA or foldback (hairpin) RNA into ~ 21 – 25 nt fragments.
Dicer’s structure allows it to measure the RNA it is cleaving. Like a cook who
“dices” a carrot, DICER chops RNA into uniformly-sized pieces.
Note the two strands of the RNA molecule. The cleavage sites are indicated by yellow
arrows.
34
ARGONAUTE proteins bind small RNAs and their targets.
ARGONAUTE proteins are named after the argonaute1 mutant of Arabidopsis;
ago1 has thin radial leaves and was named for the octopus Argonauta which it
resembles (see the figure).
ARGONAUTE proteins were originally described as being important for plant
development and for germline stem-cell division in Drosophila melanogaster.
ARGONAUTE proteins are classified into three paralogous groups: Argonaute-like
proteins, which are similar to Arabidopsis thaliana AGO1; Piwi-like proteins, which
are closely related to D. melanogaster PIWI (P-element induced wimpy testis); and
the recently identified Caenorhabditis elegans-specific group 3 Argonautes.
Members of a new family of proteins that are involved in RNA silencing mediated by
Argonaute-like and Piwi-like proteins are present in bacteria, archaea and eukaryotes,
which implies that both groups of proteins have an ancient origin.
The number of Argonaute genes that are present in different species varies. There are
35
8 Argonaute genes in humans (4 Argonaute-like and 4 Piwi-like), 5 in the D.
melanogaster genome (2 Argonaute-like and 3 Piwi-like), 10 Argonaute-like in A.
thaliana, only 1 Argonaute-like in Schizosaccharomyces pombe and at least 26
Argonaute genes in C. elegans (5 Argonaute-like, 3 Piwi-like and 18 group 3
Argonautes).
http://youdpreferanargonaute.com/2009/06/
35
MicroRNAs are encoded by MIR genes, fold into hairpin structures that are
recognized and cleaved by DCL (Dicer-like) proteins.
In summary, siRNAs-mediates silencing via post-transcriptional and
transcriptional gene silencing, while miRNAs -mediate slicing of mRNA and
translational repression.
36
In 2006, Andrwe Z. Fire and Craig C. Mello were honored by the Nobel prize “for their
discovery of RNA interference - gene silencing by double-stranded RNA“.
37
38
Small RNAs can initiate gene silencing through covalent modifications of the
DNA or its associated histone proteins, interfering with transcription.
This form of silencing is frequently associated with stably silenced DNA
including centromeres and transposons, but also occurs at genes.
This figure represents chromatin (DNA wrapped around histones) in an open and
closed conformation. Chromatin modifications include covalent modifications to DNA
and the histone proteins.
39
siRNAs can target DNA for silencing by cytosine methylation or histone
modifying enzymes.
DNA can be covalently modified by cytosine methylation, carried out by DNA
methyltransferases.
The precise mechanisms by which siRNAs target DNA for silencing are not
known, but involve the action of two plant-specific RNA-polymerase
complexes, RNA Polymerase IV (Pol IV) and RNA Polymerase V (Pol V).
40
Plants have additional RNA Polymerase complexes that contribute to
silencing.
Besides RNA polymerases I-III that occur also in other eucaryotes, plant
posses RNA polymerase IV that is involved in the siRNAs and RNA
polymerase V that is necessary for the recruitment of AGO to DNA (see the
table above).
41
In the wild-type background, the red indicates endogenous chlorophyll fluorescence
with no GFP expression; it is silenced.
NRPD1A encodes a subunit of RNA Polymerase IV. In the nrpd1a mutant, RNA
Polymerase IV isn’t produced, interfering thus with silencing.
Green signal (arrow) indicates GFP is expressed, showing that Pol IV is
required for gene silencing.
42
RNA Pol IV contributes to siRNA production. Non-coding RNAs produced by
RNA Pol V direct silencing machinery to target sites.
43
Most of the cellular siRNAs are derived from transposons and other repetitive
sequences. In Arabidopsis, as shown above, there is a high density of these
repeats in the pericentromeric regions of the chromosome.
Deep sequencing methods allow the population of siRNAs in a cell to be mapped
onto their regions of homology. The abundance of small RNAs and
transposons/retrotransposons along each of the five Arabidopsis chromosomes is
shown relative to the centromere (circle on each line representing the
chromosomes). In Arabidopsis these repetitive elements occur primarily in
pericentromeric regions.
Transposons and repetitive elements are more dispersed in organisms with larger
genomes.
44
Transcriptional silencing on the example of crossing of two lines of transgenic plants,
each of them being heterozygous (hemizygous) for a single copy transgene with
different resistance marker.
The green ovals represent cotyledons of healthy, antibiotic resistant seedlings. The
parental lines carry one copy of the transgene which acts as a dominant trait.
Therefore, 50% of the progeny from the cross should carry any one dominant trait,
and 25% should carry both. Note that these transgenes insert at different loci – they
are independent. They are not two alleles of a single locus.
45
Sometimes one of the transgenes was silenced in the progeny carrying both
genes.
That is reflected in the observed ratios of observed frequency of plant resistant
to both on of the used antibiotics. In this case, the gene for resistance to
hygromycin was silenced.
46
The mechanism of silencing in this case is promoter methylation.
47
48
49
microRNAs slice mRNAs or interfere with their translation. Thus, in contrast to
siRNAs, miRNAs do not induce transcriptional silencing.
miRNAs are thought to have evolved from siRNAs, and are produced and
processed somewhat similarly. Plants have a small number of highly
conserved miRNAs, and a large number of non-conserved miRNAs.
miRNAs are encoded by specific MIR genes but act on other genes – they are
trans-acting regulatory factors. miRNAs in plants regulate developmental and
physiological events.
Plant miRNA bind preferentially to the 5’UTR or coding sequence and require
almost complete complementarity with their targets.
In contrast to that, in animals there is much higher proportion of genes being
regulated by miRNAs. The most of animal miRNA bind to the 3’UTR and
extensive sequence complementarity is not required.
In animals, dominant regulatory mechanism mediated by miRNAs is
50
translational repression, while in plants, most of the miRNAs initiate slicing of
target mRNAs.
50
Plants have 4 or more DCL proteins, more than found in other organisms. The
amplification of DCL proteins is thought to allow plants great flexibility in
pathogen defence responses.
Note that mammals make do with one dicer, and insects and fungi with two. Like
most components of the siRNA pathway, dicer-like genes are amplified in plants.
51
Arabidopsis has 10 AGO proteins. They are not all well characterized and there is
some functional overlap.
52
miRNAs are encoded by MIR genes. MIR genes are transcribed into long
RNAs called primary miRNA (pri-miRNA) with partial complementarity.
The pri-miRNA transcript folds back into a partially double-stranded stem-loop
or hairpin structure, which is processed by DCL1. The product of DCL
cleavage dsRNA with a 2-nucleotide 3´ overhang is called a miRNA-miRNA*
duplex.
The miRNA-miRNA* duplex is transported from the nucleus into the cytoplasm
by the homologue of exportin 5 HASTY (will be discussed later).
Subsequently, the miRNA* strand is degraded.
53
In spite of there are quite few genes regulated by miRNAs in plants, there is
high number of MIR genes.
However, most of them are evolutionary non-conserved and probably provide
no competitive advantage. It has been suggested that there is high rate of birth
and death and only few of them became stabilized in the genome.
Nearly half of the targets of conserved miRNAs are transcription factors.
54
Among the evolutionary conserved, i.e. important regulators belong miRNAs
that target TF. E.g. miR156 family is highly conserved within the plant kingdom
(it is found in angiosperms as well as mosses)
miR156 is encoded by six or more genes in Arabidopsis and targets
transcription factors that control developmental phase changes (see further
slides).
The red sequence indicates the miRNA produced from each of the six Arabidopsis
MIR156 genes (MIR156A – MIR156F). The boxed sequence shows the miRNA*.
55
The table shows targets of some of the evolutionary conserved MIR genes (or rather
respective miRNAs). As discussed previously, these are mostly MIR genes encoding
miRNAs recognizing TFs.
56
Plant miRNAs are thought to be derived from their target sequences following
gene duplication, inverted duplication and divergence.
Only some miRNAs confer selective advantage and are retained and further duplicated.
57
Vegetative phase change is the transition from juvenile to adult growth in
plants.
The transition includes changes in the leaf shape and arrangenment, internode
length, and the accumulation of epidermal hairs (trichomes) and waxes.
58
Some cacti have very different juvenile and adult growth patterns.
59
Eucalyptus leaves are strongly dimorphic, as are leaves of holly and ivy. In
other plants including Arabidopsis and maize the change is more subtle.
60
In Arabidopsis, juvenile leaves are rounder, less serrated, and have trichomes
only on the upper (adaxial) surface; adult leaves also have trichomes on the
lower (abaxial) surface.
In the next few slides the leaves are colour coded as juvenile (olive), adult (green) or
reproductive, also sometimes called cauline leaves (red) based on their morphology
and anatomy. The boxes above the leaves schematically indicate the relative duration
of each phase (for comparing wild-type and mutant plants).
61
HASTY, with a shortened juvenile phase, encodes a protein needed for miRNA
export from nucleus to cytoplasm.
It’s easy to see from these figures that hasty has a shortened juvenile phase relative
to wild-type.
62
Loss-of-function zippy mutants prematurely express adult vegetative traits.
ZIPPY encodes an ARGONAUTE protein, AGO7.
Zippy prematurely expresses adult traits; this is a different effect that hasty, in which
the total non-reproductive phase was shortened as well.
63
As could be clearly seen from the phenotyope of transgenic line with ectopic
overexpression of miR156 gene, there is dramatic elongation of the juvenile phase.
Highly conserved miR156 targets SQUAMOSA PROMOTER BINDING PROTEIN-box
(SBP-box) genes in maize, called SPL genes in Arabidopsis.
SBP-box genes are TFs that regulate expression of key regulators of flowering TF,
SOC1, LFY, and AP1 and thus regulate the transition from juvenile to
adult/reproductive growth phase.
64
The SQUAMOSA PROMOTER BINDING PROTEIN-LIKE (SPL) genes are a family of
SBP-box transcription factors that are miR156 targets.
In wild-type plants, miR156 expression decreases with plant age, allowing SPL
to accumulate and promote phase change.
The graph indicates the relative activity of miR156 and SPL, showing that as miR156
levels decrease SPL levels increase. When miR156 levels are high, SPL mRNA is
cleaved and the protein can’t accumulate.
65
The miR156 effect can be eliminated by deletion of the 3’ UTR or mutations that
interfere with miR156 binding, as shown schematically.
In either case, SPL protein accumulates early (it is indifferent to the presence of
mir156), promoting phase change. Note that the miR156 resistant plants are
flowering when the control plants are still vegetative.
66
Similarly, interfering with miR156 production allows SPL to accumulate early and
promotes precocious phase change.
67
Interestingly, in Arabidopsis, MIR172 gene is induced by one of the SPL transcription
factors, suggesting that a cascade of miRNA-regulated genes controls phase
transition.
68
miRNAs were discovered in studies of developmental progressions in the nematode C.
elegans. miRNA encoded by lin-4 is required for proper larval development.
Genes lin-14 and lin-4 promote or repress developmental progression in the nematode C. elegans.
Genetic studies showed that lin-4 acts antagonistically to lin-14 but that lin-4 does not encode any
protein.
The functional domain of lin-4 was identified and found to be complementary to several regions in the
3´ UTR of lin-14 (see the figure, red).
Subsequently, the interaction between these complementary sequences was shown to be necessary
for lin-4’s repression of lin-14, revealing the basis for gene regulation by miRNAs.
69
For proper developmental progression in C. elegans, the juvenile-promoting activity
of the lin-14 gene has to be repressed by the action of the miRNA encoded by lin-4.
70
71
72