65 167 PiwiInteracting RNAs and Transposon Control

wea25324_ch16_471-521.indd Page 501
12/14/10
4:54 PM user-f469
/Volume/204/MHDQ268/wea25324_disk1of1/0073525324/wea25324_pagefile
16.7 Piwi-Interacting RNAs and Transposon Control
to be silenced. Polymerase V makes longer RNAs
from regions throughout the plant genome. These
longer RNAs attract siRNA-Ago4 complexes, but
only to regions that are targets for silencing, from
which these siRNAs were made. These complexes in
turn attract the enzymes required to methylate both
DNA and histones, which in turn leads to heterochromatization.
16.7 Piwi-Interacting RNAs and
Transposon Control
In Chapter 23 we will learn that DNA elements known as
transposons can transpose, or jump from place to place in
a genome. In doing so, they can interrupt and inactivate
genes, or even break chromosomes. Thus, transposition is a
dangerous process that can lead to cell death or disease,
such as cancer. Accordingly, it is important that cells be
able to control transposition. This is particularly true in
germ cells, which give rise to gametes that will pass genes
on to the next generation. The serious mutations or cell
death caused by transposition in germ cells reduce reproductive success and therefore threaten a species’s survival.
It is not surprising, therefore, that organisms have evolved
mechanisms for dealing with transposons, and that these can
be targeted to germ cells. In fact, germ cells produce another
class of small RNAs, 24 to 30 nt long, called Piwi-interacting
RNAs (piRNAs). Like siRNAs and miRNAs, piRNAs associate with Argonaute proteins, but these proteins belong to a
different branch, or clade, of the Argonaute superfamily than
3′-end
processing
3′
Transposon mRNA
the Ago proteins we have been discussing. The piRNAs bind
to members of the Piwi clade, while siRNAs and miRNAs
bind to members of the Ago clade.
The piRNAs of fruit flies and mammals tend to be complementary to either the sense or antisense strand of transposons from the same organism. These piRNAs derive
from clusters of piRNA genes, apparently via transcription
of a long cluster and subsequent processing of the precursor RNA into mature piRNAs. Some, if not most, of this
processing may actually occur simultaneously with inactivation of transposons, by a so-called ping-pong amplification loop, as follows (Figure 16.37):
In Drosophila, Piwi proteins such as Piwi and Aubergine
tend to associate with piRNAs that are complementary to
transposon mRNAs; these piRNAs usually have a U in the
first position. This piRNA-Piwi or -Aubergine complex can
associate through base-pairing with a transposon mRNA,
which triggers slicer cutting 10 nt upstream of an A that is
base-paired to the U at the 59-end of the piRNA. This cut,
together with processing at the 39-end of the transposon
mRNA, creates a short RNA that can associate with
another protein, Ago3, which preferentially binds to
RNAs that represent parts of transposon mRNAs. The RNAAgo3 complex can then bind to a piRNA precursor RNA
by base-pairing, and the slicer activity of Ago3 cuts just
upstream of the U of the A–U base pair. This cut, together
with end processing of the piRNA precursor, creates a mature piRNA that can bind to Piwi or Aubergine to start the
cycle over.
Note that this mechanism accomplishes two things: It
slices up transposon mRNA, thereby blocking transposition,
and it amplifies the amount of piRNA available, thus stimulating the process. Because the transcription of piRNA clusters is confined to germ cells, and somatic cells immediately
piRNA
5′ U
A
Piwi/Aub
3′
5′
Cut by slicer
AGO3
Piwi/Aub
5′ U
5′
3′
3′-end
processing
5′
3′
501
AGO3
A
U
3′
5′
piRNA cluster
Cut by slicer
Figure 16.37 Model for a ping-pong amplification loop for piRNAs. Details are in the text.
A
3′