Human Alu RNA Is a Modular Transacting Repressor
of mRNA Transcription during Heat Shock
Peter D. Mariner,1,2Ryan D. Walters,1,2Celso A. Espinoza,1Linda F. Drullinger,1Stacey D. Wagner,1Jennifer F. Kugel,1,*
and James A. Goodrich1,*
1Department of Chemistry and Biochemistry, University of Colorado at Boulder, 215 UCB, Boulder, CO 80309-0215, USA
2These authors contributed equally to this work.
*Correspondence: email@example.com (J.F.K.), firstname.lastname@example.org (J.A.G.)
Noncoding RNAs (ncRNAs) have recently been
discovered to regulate mRNA transcription in trans,
a role traditionally reserved for proteins. The breadth
of ncRNAs as transacting transcriptional regulators
and the diversity of signals to which they respond
are only now becoming recognized. Here we show
that human Alu RNA, transcribed from short inter-
spersed elements (SINEs), is a transacting transcrip-
tional repressor during the cellular heat shock
response. Alu RNA blocks transcription by binding
RNA polymerase II (Pol II) and entering complexes
at promoters in vitro and in human cells. Transcrip-
tional repression by Alu RNA involves two loosely
structured domains that are modular, a property
reminiscent of classical protein transcriptional regu-
lators. Two other SINE RNAs, human scAlu RNA and
mouse B1 RNA, also bind Pol II but do not repress
nation for why mouse cells harbor two major classes
of SINEs, whereas human cells contain only one.
Transcription of mRNA, the first step in expressing protein-
encoding genes, is tightly regulated to control cellular growth,
differentiation, and the response to external stimuli. The enzyme
that synthesizes mRNAs in eukaryotes is RNA polymerase II (Pol
general transcription factors (Thomas and Chiang, 2006). Much
work has been done to elucidate mechanisms by which eukary-
otic transcription is controlled, leading to the discovery of a wide
range of gene-specific activators and repressors, coactivators,
and chromatin modifying factors that work with the general
transcription machinery to control the initiation of transcription
at promoters. Recently, noncoding RNAs (ncRNAs) have been
2006). Compared to their protein counterparts, only a small
number of ncRNA transcriptional regulators have been identified
and characterized to date; however, their mechanisms of action
are diverse, ranging from regulating nuclear localization to
controlling the activity of general transcription factors.
We previously discovered that mouse B2 RNA is an ncRNA
transcriptional regulator that binds to Pol II and represses
mRNA synthesis (Allen et al., 2004; Espinoza et al., 2004). B2
RNA is transcribed by RNA polymerase III from short inter-
spersed elements (SINEs) (Kramerov et al., 1985). In response
to heat shock, the transcription of B2 SINEs is upregulated
(Fornace and Mitchell, 1986; Li et al., 1999; Liu et al., 1995).
Also upon heat shock, many mRNA genes are repressed (e.g.,
sive genes are highly activated (e.g., hsp70) (Findly and Peder-
son, 1981; Gilmour and Lis, 1985; O’Brien and Lis, 1993; Sonna
specific protein-encoding genesduring the heat shock response
in mouse cells (Allen et al., 2004). In vitro, B2 RNA binds Pol II
with high affinity, specificity, and kinetic stability (Espinoza
et al., 2004). Through its association with Pol II, B2 RNA is
recruited into preinitiation complexes at promoters in vitro where
it blocks all RNA synthesis.
A longstanding question in mammalian genomics is why
SINEs, which are retrotransposons, have been maintained and
more curiously why different species have different and appar-
ently unrelated SINEs (Kazazian, 2004; Schmid, 2003). For
example, human cells have a single predominant SINE, Alu,
whereas mouse cells have two predominant SINEs, B1 and
B2. Alu and B1 SINEs are derived from a 7SL-like precursor
and are not related in sequence to B2 SINEs, which are tRNA
derived (Kramerov and Vassetzky, 2005; Schmid, 1996). Like
B2 SINEs, Alu and B1 SINEs are transcribed by RNA polymerase
III to generate ncRNAs (Alu RNA and B1 RNA, respectively), and
levels of Alu RNA and B1 RNA increase during heat shock (Li
et al., 1999; Liu et al., 1995). Human Alu RNA is a tandem repeat
of two B1 RNA-like elements connected by an A-rich linker
(sequence alignments are shown in Figure S1 available online)
(Schmid, 1998; Schmid and Jelinek, 1982). In cells, two forms
of Alu RNA exist: the full-length transcript and a short-cytoplas-
mic transcript (scAlu RNA) consisting only of the 50B1-like
element, which likely results from processing full-length Alu
RNA (Maraia et al., 1993; Matera et al., 1990). Previous experi-
ments to determine the secondary structures of Alu RNA and
B1 RNA suggested that the basic folding pattern of each of the
two repeats, or arms, in Alu RNA is similar to that of B1 RNA
(Figure 1A) (Labuda et al., 1991; Maraia, 1991; Sinnett et al.,
Molecular Cell 29, 499–509, February 29, 2008 ª2008 Elsevier Inc. 499
1991). We recently determined a model for the secondary struc-
ture of B2 RNA (Figure 1A) and identified the minimal region that
is functional for binding Pol II and repressing transcription in vitro
(highlighted in gray) (Espinoza et al., 2007).
Here we show that Alu RNA acts as a transcriptional repressor
during heat shock in human cells, despite a lack of obvious
similarity to B2 RNA. Further analysis revealed that Alu RNA
binds two molecules of Pol II and has two loosely structured
regions that are required for transcriptional repression. We also
found that B1 RNA and scAlu RNA bind Pol II but do not repress
transcription in vitro. Interestingly, the repression domains of Alu
RNA are modular: they can be fused to B1 RNA to create
Figure 1. Alu RNA Represses mRNA Tran-
scription in Response to Heat Shock
(A) Comparison of the secondary structures of
human Alu RNA, mouse B1 RNA, and mouse B2
RNA. The 30B1-like element of Alu RNA, known
as the right arm, is designated Alu-RA RNA.
Models were generated with BayesFold software
(Knight et al., 2004) guided by structural probing
data for each RNA (Espinoza et al., 2007; Labuda
et al., 1991; Maraia, 1991; Sinnett et al., 1991).
The minimal region of B2 RNA that functions as
a transcriptional repressor in vitro is shaded in
gray (Espinoza et al., 2007).
(B) Nuclear transcripts from four protein-encoding
genes decrease upon heat shock in HEK293 cells,
while nuclear Alu RNA increases. As controls,
hsp70 RNA increases, while 18S rRNA and U2
RNA remain constant. HK II, hexokinase II; ACS
II, acyl-CoA synthetase II; GLUD-1, glutamate de-
hydrogenase-1; and hsp70,heat shock protein 70.
(C) An antisense oligonucleotide against Alu RNA
attenuates the repression of mRNA transcription
of four genes and the increase in Alu RNA after
heat shock in HEK293 cells. hsp70 RNA, 18S
rRNA, and U2 RNA are shown as controls.
(D) Alu RNA potently represses transcription in
a minimal reconstituted system. Relative amounts
of transcript compared to reactions without Alu
RNA are plotted versus Alu RNA concentration
and fit as described in the Experimental Proce-
chimeric ncRNA transcriptional repres-
Alu RNA represses Pol II transcription in
vitro and in cells, we found that Alu RNA
enters complexes with Pol II at promoters
of genes that are repressed during heat
shock and blocks RNA synthesis. To-
gether our studies show that SINE RNAs
in mouse and human cells, which are
unrelated in sequence, have evolved sim-
ilar functions as transacting transcription
factors. Moreover, Alu RNA has modular
repression domains (i.e., distinct regions
that, when fused to a transcriptionally
inert ncRNA that binds Pol II, turn that
ncRNA into a potent transcriptional repressor), a property remi-
niscent of protein transcriptional regulators.
Human Alu Is a Repressor of Pol II Transcription
during Heat Shock
When human cells are subjected to heat shock, levels of tran-
scription of some mRNA genes decrease (Sonna et al., 2002),
but the mechanism by which this repression occurs is not under-
stood. Figure 1B shows that levels of nuclear transcripts from
four housekeeping genes sharply decreased after heat shock
Alu RNA Is a Modular Transcriptional Repressor
500 Molecular Cell 29, 499–509, February 29, 2008 ª2008 Elsevier Inc.
of HEK293 cells. Under these same conditions, nuclear Alu RNA
levels substantially increased. As a control, we monitored levels
of transcript produced from the hsp70 gene, and as anticipated,
we observed a robust increase in the level of nuclear hsp70
transcript. In addition, we monitored levels of two abundant
ncRNAs, U2 snRNA and 18S rRNA, which are transcribed by
RNA polymerases II and III, respectively. Levels of 18S rRNA
and U2 RNA were similar before and after heat shock.
To determine whether Alu RNA functions in the repression of
Pol II transcription during heat shock, we transfected HEK293
cells with an antisense oligonucleotide directed against Alu
RNA or a control oligonucleotide with a scrambled sequence.
The antisense oligonucleotide attenuated the transcriptional
repression after heat shock of four mRNA genes, whereas it
had little effect on levels of 18S rRNA, U2 snRNA, or hsp70
mRNA (Figure 1C). In addition, the antisense oligonucleotide
substantially blocked the increase in Alu RNA that occurs upon
heat shock, although nuclear Alu RNA was not entirely elimi-
nated. Given the sequence diversity of SINEs in the human ge-
Figure 2. Both Arms of Alu RNA Bind Pol II,
although Only the Right Arm Represses
(A) Pol II shifts Alu RNA to two bands. EMSA with
32P-labeled Alu RNA (0.05 nM) and purified human
Pol II showed an apparent KD< 2 nM.
(B) Alu RNA binds two Pol II molecules. In the
EMSA shown, the concentration of Pol II was
held constant at 2 nM while Alu RNA was added
at the indicated concentrations.
(C) scAlu RNA and Alu-RA RNA each bind one
molecule of Pol II. EMSAs were performed with
32P-labeled RNAs (0.05 nM) and purified human
Pol II at the concentrations shown. B2 RNA(3-74)
was used as a negative control.
(D) B1 RNA binds Pol II. An EMSA was performed
with32P-labeled B1 RNA (0.05 nM) and purified
human Pol II at the concentrations shown.
(E) Alu-RA RNA represses Pol II transcription in
a reconstituted system, whereas scAlu RNA
does not. The 390 nt G-less transcript is shown.
nome, we do not know the fraction of dif-
ferent Alu RNAs in cells that was targeted
by the antisense oligonucleotide.
We next asked whether recombinant
Alu RNA could directly repress transcrip-
tion in a system reconstituted from highly
purified human components. As shown
in Figure 1D, Alu RNA potently repressed
Pol II transcription in this system (IC50=
2 nM). Therefore, Alu RNA acts as a re-
pressor of Pol II transcription in human
cells and in a well-defined in vitro system.
We previously found that mouse B2 RNA
could repress transcription in a human
nuclear extract. We now have found that
the reverse is also true; human Alu RNA
repressed transcription when added to
a mouse nuclear extract (Figure S2A).
We conclude that human Alu RNA and mouse B2 RNA share
a common biological function as transcriptional repressors dur-
ing heat shock, despite the fact that human Alu and mouse B2
SINEs are not evolutionarily related.
Each Arm of Alu RNA Binds Pol II, and the Right Arm
Is a Potent Transcriptional Repressor
Because B2 RNA binds directly to Pol II, we asked whether Alu
RNA also had this property. Pol II bound Alu RNA with high affin-
ity(appKD<2nM)butunexpectedly shiftedAluRNAtotwo bands
in electrophoretic mobility shift assays (EMSAs) (Figure 2A). At
lower concentrations of Pol II, the lower band predominated,
whereas at higher concentrations, the upper band predomi-
nated. This suggested that Alu RNA could bind two molecules
of Pol II. A control EMSA with B2 RNA and Pol II showed only
a single shifted band (Figure S2B). To further test the hypothesis
that Alu RNA can bind two molecules of Pol II, we performed
EMSAs in which the concentration of Pol II was held constant
while Alu RNA was titrated up. As can be seen in Figure 2B, at
Alu RNA Is a Modular Transcriptional Repressor
Molecular Cell 29, 499–509, February 29, 2008 ª2008 Elsevier Inc. 501
low concentrations of Alu RNA, when Pol II was in excess, two
distinct bands were observed. As the concentration of Alu RNA
was increased above that of Pol II, the upper band disappeared.
These results are consistent with the upper band containing
complexes composed of one molecule of Alu RNA and two
molecules of Pol II.
We hypothesized that each arm of Alu RNA can bind to Pol II.
of Alu RNA (Alu-RA RNA) bound Pol II with high affinity (appKD
control, we show that the 50region of B2 RNA, which folds into
a specific secondary structure, did not bind Pol II under the
same conditions (Espinoza et al., 2004; Espinoza et al., 2007).
These data are consistent with Alu RNA binding two polymer-
is related to both arms of Alu RNA in sequence and secondary
even though this ncRNA does not repress Pol II transcription in
vitro (Allen et al., 2004; Espinoza et al., 2004).
When each arm of Alu RNA was individually tested for the
ability to repress transcription in vitro, we found that Alu-RA
RNA repressed transcription, whereas scAlu RNA did not
(Figure 2E, B2 and B1 RNAs are shown as controls). Hence, in
vitro, Alu-RA RNA and full-length Alu RNA are functionally similar
to B2 RNA (i.e., they bind Pol II and repress transcription),
whereas scAlu RNA functions much like B1 RNA (i.e., it binds
Pol II but does not repress transcription). In contrast to Alu
RNA, B2 RNA does not appear to have two regions that can
independently bind Pol II: the 30region of B2 RNA (75–149) is
sufficient to bind Pol II and repress transcription, whereas the
50region of B2 RNA (3–74) does not bind Pol II nor repress
transcription (Espinoza et al., 2007). These data also show that
binding of a natural ncRNA to Pol II does not necessarily result
in transcriptional repression.
Two Regions of Alu RNA Mediate
The observation that Alu-RA RNA can repress transcription
whereas B1 RNA and scAlu RNA cannot suggests that the right
arm of Alu RNA possesses unique sequences and/or structural
features involved in repression. To begin to delimit the region(s)
of Alu-RA RNA that mediates transcriptional repression, we
constructed two truncated ncRNAs consisting of either the 50
region or the 30region of Alu-RA RNA (named Alu-RA50RNA
and Alu-RA30RNA, see the schematic in Figure 3A). (The func-
tional properties and sequences of these and subsequent
Alu RNA deletions and mutants are summarized in Figures S3
and S4.) As shown in the top two data panels of Figure
3A, Alu-RA30RNA repressed transcription in vitro, whereas
Alu-RA50RNA did not. Moreover, Alu-RA30RNA bound Pol II
(Figure 3B, middle panel), whereas Alu-RA50RNA did not
(Figure 3B, top panel). Together these results show that the 30
region of Alu-RA RNA is both necessary and sufficient for Pol II
binding and transcriptional repression.
To further delimit the region(s) of Alu-RA RNA involved in
transcriptional repression, we compared the sequence and
secondary structure of its 30region with those of scAlu RNA
and B1 RNA. We noted two features unique to the right arm of
Alu RNA (see Figure S1): (1) stem 9 is longer, and (2) the region
containing stem 5 is more loosely structured (i.e., less double
stranded). We hypothesized that transcriptional repression by
Alu-RA RNA involves one or both of these regions. To test this
hypothesis, we created deletions of these regions in the context
of Alu-RA30RNA (Alu-RA30-DS RNA and Alu-RA30-DL RNA, see
Figure 3A). When added to in vitro transcription assays, Alu-
RA30-DS RNA repressed transcription, whereas Alu-RA30-DL
RNA did not (Figure 3A, bottom two panels). Importantly,
Alu-RA30-DL RNA retained the ability to bind Pol II (Figure 3B,
bottom panel), demonstrating that the lack of transcriptional
repression was not due to lack of Pol II binding.
scriptional repression, we deleted the L region in the context of
full-length Alu RNA (Alu-DL RNA). Interestingly, we found that
Alu-DL RNA repressed transcription (Figure 3C, second panel),
suggesting that an additional region functions in transcriptional
repression. The only region contained in full-length Alu RNA
that is not a part of either scAlu RNA or Alu-RA RNA is the
A-rich single-stranded linker that connects the two arms. There-
10 nt from the 20 nt A-rich linker (Alu-DADL RNA, see the sche-
matic in Figure 3C). As shown in Figure 3C, Alu-DADL RNA did
not repress transcription. Importantly, this ncRNA retained the
ability to bind two molecules of Pol II (Figure 3D). We also
made a mutant Alu RNA lacking only the A region (Alu-DA
RNA) and found that it repressed transcription (Figure 3C,
bottom panel). The role of the A and L regions in mediating
repression was further tested by using transient transfection
assays. As shown in Figure 3E, transfection of HEK293 cells
luciferase reporter to a greater extent than did transfection of an
Alu-DADL RNA expression construct. Similarly, transfection of
an expression construct for B2 RNA, but not B1 RNA, repressed
reporter expression (Figure S2C). Together, the data presented
in Figure 3 show that two regions, which are not essential for
binding Pol II, within full-length Alu RNA function to repress
transcription. Moreover, the A region and the L region are each
sufficient to mediate transcriptional repression in the absence
of the other.
Alu RNA Contains Two Loosely Structured and Modular
The two regions of Alu RNA that mediate repression are both
loosely structured in the secondary structure model: the 20 nt
A region is devoid of secondary structure, and the L region
contains two short stems (2 bp each) that, if melted, would
give rise to a large bulge containing 13 nt on one strand and 14
nt on the other. To determine whether the loosely structured
nature of the L region is important for its function in transcrip-
tional repression, we made three mutant Alu-RA RNAs. First,
we converted a small bulge in the L region (indicated by the
arrow in the schematic in Figure 4A) into two base pairs, which
would allow a continuous 6 bp stem to form in the secondary
structure. This mutant RNA (Alu-RA-mt1 RNA) was unable to
repress transcription (Figure 4A); wild-type Alu-RA RNA and
Alu-RA-DL are shown as controls. Second, we eliminated base
pairing in the two short (2 bp) stems in the L region (on both sides
Alu RNA Is a Modular Transcriptional Repressor
502 Molecular Cell 29, 499–509, February 29, 2008 ª2008 Elsevier Inc.
of the arrow in the schematic in Figure 4A). This mutant RNA
(Alu-RA-mt2 RNA) repressed transcription (Figure 4A). Third,
the sequence of the L region was changed, but its loosely struc-
tured nature was maintained (Alu-RA-mt3 RNA). Alu-RA-mt3
RNA potently repressed transcription (Figure 4A, bottom panel).
Together these results indicate that the structure, but not the
sequence, of the L region is important for repression.
Because the regulatory domains found in transcriptional
if the same were true of the A and L regions of Alu RNA.To deter-
mine this, we took advantage of the fact that B1 RNA can bind
tightly to Pol II but does not repress transcription. First, we re-
Figure 3. Two Regions in Alu RNA Mediate
(A) The L region in the right arm of Alu RNA is
required for repression. Schematic representation
of Alu-RA RNA showing the 50and 30regions as
well as the S and L regions. The positions of stems
4–9 are indicated. Mutant ncRNAs were added to
in vitro transcription reactions. The 390 nt G-less
product is shown.
(B) Neither the 50region nor the L region of Alu-RA
is required for binding Pol II. EMSAs were
performed with32P-labeled Alu-RA50RNA (0.05
nM), Alu-RA30RNA (0.05 nM), or Alu-RA30-DL
RNA (0.05 nM) and purified human Pol II.
(C) The A-rich linker is a second region in Alu RNA
that mediates repression. The schematic shows
the positions of the deletions of the L and A-rich
regions in Alu RNA. Mutant ncRNAs were added
to in vitro transcription reactions. The 390 nt
G-less product is shown.
(D) Deletion of the A-rich and L regions of Alu RNA
does not impair Pol II binding. An EMSA was per-
formed with32P-labeled Alu-DADL RNA (0.05 nM)
and purified human Pol II.
(E) The A and L regions in Alu RNA mediate tran-
scriptional repression in human cells. HEK293
porter and either an expression construct for Alu
RNA, Alu-DADL RNA, or the parental pUC plasmid
asacontrol. Eachbaris the average of eight trans-
for each pairwise combination is shown.
placed a region of B1 RNA with the L re-
gion from the right arm of Alu RNA as dia-
grammed in Figure 4B (the boxed regions
were fused to create the chimeric B1-L
RNA shown in the middle). B1-L RNA
repressed transcription with a potency
similar to Alu-RA RNA (Figure 4C). Next,
we asked whether fusion of the 20 nt
A-rich linker from Alu RNA to B1 RNA
would turn the typically inert B1 RNA into
a transcriptional repressor (see the sche-
maticin Figure 4B).
Figure 4C, the fusion ncRNA (B1-A RNA)
fect on in vitro transcription (data not shown). In addition, we
tested whether the sequence or unstructured nature of the A
region is important for mediating repression. To do so we altered
with U, G, C, and A, respectively) in the context of the B1-A RNA,
ThisRNA (B1-URNA) potently
(Figure 4C). This indicates that it is the unstructured nature and
not the sequence of the A region that is important for transcrip-
tional repression. We conclude that the loosely structured A and
L regions of Alu RNA are modular transcriptional repression
Alu RNA Is a Modular Transcriptional Repressor
Molecular Cell 29, 499–509, February 29, 2008 ª2008 Elsevier Inc. 503
Alu RNA Enters Complexes at Promoters and Represses
All RNA Synthesis
To begin to determine the point in the transcription reaction that
is inhibited by Alu RNA, we asked if it could inhibit abortive
initiation, an assay that monitors the repetitive synthesis of 3 nt
RNA products. We found that Alu RNA caused a 10-fold
decrease in abortive initiation (Figure 5A). Hence, Alu RNA
represses transcription at a point prior to or at initiation. We
also found that to repress transcription Alu RNA must be added
to reactions prior to the assembly of preinitiation complexes on
promoter DNA (Figure 5B). To determine whether Alu RNA
affects theformation of preinitiation complexes, weusedEMSAs
in which promoter DNA was fluorescently labeled and Alu RNA
was32P labeled. Figure 5C shows both fluorimagery (left) and
phosphorimagery (right) scans of the same gel. Aligning the
two scans revealed that Alu RNA changed the migration of pre-
initiation complexes (compare lanes 2 and 3 in the fluorimagery
scan), and moreover, it comigrated with the promoter DNA
(compare lane 3 in the two scans). RNase treatment of preinitia-
tion complexes containing Alu RNA restored their migration to
the original position (lane 4). Lanes 5–7 show controls in which
the promoter DNA was omitted. We conclude that Alu RNA
assemblesinto preinitiationcomplexes andblocks alldetectable
RNA synthesis. This is similar to the mechanism of transcrip-
tional repression in vitro by B2 RNA (Espinoza et al., 2004).
Alu RNA or B2 RNA Occupies the Promoters
of Repressed Genes with Pol II after Heat Shock
in Human or Mouse Cells
The in vitro studies above lead to the prediction that inactive
transcription complexes containing Pol II and Alu RNA will
occupy the promoters of genes repressed during heat shock in
human cells. To test this we first performed ChIP assays with
an antibody against Pol II and found that the promoters of the
heatshock-repressed GLUD-1 andACS IIgeneshad substantial
Pol II occupancy after heat shock (Figure 6A). By contrast, Pol II
occupancy in the transcribed regions of these genes decreased
after heat shock, indicative of transcriptional repression. As
a control, we monitored Pol II occupancy at the heat shock-
activated hsp70 gene and found it to increase after heat shock
in both the promoter and downstream regions. In addition, we
found that in response to heat shock Pol II occupancy remained
constant at the U2 promoter and decreased in the downstream
region of the gene. This suggests that transcription of the U2
gene decreases after heat shock, but not to an extent that
influences the nuclear U2 snRNA pool (see Figure 1).
To determine the occupancy of Alu RNA on regions of the
genome, we developed a technique named chromatin oligo-
affinity precipitation (ChOP), which is a modified ChIP assay. In
the ChOP assays, a biotinylated antisense oligonucleotide was
used to affinity purify Alu RNA and associated biomolecules
Figure 4. Alu RNA Contains Two Loosely Structured and Modular Repression Domains
(A) The structure of the L region, and not its sequence, is critical for its function in transcriptional repression. The arrow in the schematic indicates the small bulge
that was converted into two base pairs in the Alu-RA-mt1 RNA. In the Alu-RA-mt2 RNA, the two base pair stems on both sides of the small bulge (indicated by the
arrow) were mutated to eliminate base pairing, thereby creating an RNA with a large bubble. In the Alu-RA-mt3 RNA, the L region (50-AAUGGCGUGAACCCGG
G-47nt-GGGUGACAGAGCGAGA-30) was replaced by a different sequence (50-UCUCGCUCUGUCACCC-47nt-CCCGGGUUCACGCCAUU-30) having the same
predicted secondary structure. Mutant ncRNAs were added to in vitro transcription reactions. The 390 nt G-less product is shown.
(B) The schematic shows the chimeric B1-L and B1-A RNAs.
(C) B1 RNA can be converted into a transcriptional repressor by addition of the L region, fusion to the A region, or fusion to a nonnatural unstructured U-rich RNA.
Relative amounts of transcript compared to reactions without ncRNAs are plotted versus ncRNA concentration and fit as described in the Experimental
Procedures. B1 and Alu-RA RNAs were used as controls.
Alu RNA Is a Modular Transcriptional Repressor
504 Molecular Cell 29, 499–509, February 29, 2008 ª2008 Elsevier Inc.
to determine whether Alu RNA was present at specific regions of
the genome. Using ChOP assays, we found that Alu RNA occu-
shock (Figure 6B). We did not observe Alu RNA at the heat
shock-activated hsp70 promoter before or after heat shock.
Alu RNA was also not detected at the U2 promoter, suggesting
that Alu RNA does not directly set the pattern of Pol II occupancy
in the U2 gene after heat shock. Because we had previously
found that mouse B2 RNA represses transcription in vitro by
assembling with Pol II into complexes at promoters (Espinoza
et al., 2004), we also performed ChIP and ChOP experiments
on mouse cells before and after heat shock. These experiments
showed that Pol II and B2 RNA co-occupy the promoters of the
Figure 5. Alu RNA Assembles into Preinitia-
tion Complexes In Vitro and Renders Them
(A) Alu RNA blocks abortive synthesis of 3 nt
RNAs. Three-nucleotide RNA synthesized in the
absence and presence of Alu RNA (50 nM) was
quantitated. Bars are the average of at least four
reactions, and errors are one standard deviation.
(B) Alu RNA represses transcription when added
to reactions prior to the formation of preinitiation
complexes, but not when added after they have
formed. The schematic shows a timeline for the
reactions. The 390 nt G-less transcript is shown.
(C) Alu RNA comigrates with promoter DNA in
preinitiation complexes resolved by EMSA and
changes the position of their migration. A single
native gel was scanned by using fluorimagery
(left image) to detect the fluorescently labeled
promoter DNA (F-DNA) and phosphorimagery
(right image) to detect32P-Alu RNA. Lanes 4 and
7 were moved from different regions of the same
gel in order to create the image shown.
Figure 6. Pol II and Either Alu RNA or B2
RNA Occupy the Promoters of Genes Re-
pressed during Heat Shock in Human and
Mouse Cells, Respectively
(A) ChIP assays using HEK293 cells show that af-
ter heat shock Pol II occupies the promoter, but
not the downstream region, of the repressed
GLUD-1 and ACS II genes.
(B) ChOP assays using HEK293 cells show that af-
ter heat shock Alu RNA occupies the promoter of
the repressed GLUD-1 and ACS II genes, but not
the promoters of the U2 gene and the activated
(C) In NIH 3T3 cells, Pol II occupies the promoters,
but not the downstream regions, of two repressed
genes after heat shock as determined by ChIP
(D) ChOP assays using NIH 3T3 cells show that af-
ter heat shock B2 RNA occupies the promoters of
two repressed genes, but not the promoter of the
activated hsp70 gene. Assays in all panels were
done two or more times and representative data
are shown. The percentages of input and eluate
samples used in PCRs are shown in Table S2.
Alu RNA Is a Modular Transcriptional Repressor
Molecular Cell 29, 499–509, February 29, 2008 ª2008 Elsevier Inc. 505
repressed actin and hexokinase II genes after heat shock in
mouse cells (Figures 6C and 6D). Together, the data in Figure 6
indicate that in cells Alu RNA and B2 RNA repress transcription
after heat shock via a mechanism similar to that which we deter-
mined in vitro (Figure 5; Espinoza et al., 2004).
Here we show that the SINE transcript Alu RNA represses mRNA
transcription in human cells during the response to heat shock.
Alu RNA binds directly to Pol II, with each arm being capable of
binding one polymerase molecule. Deletion analysis identified
tionalrepressionbutarenot requiredforPolIIbinding. Thesetwo
repression domainsarelooselystructuredand modular; combin-
ncRNA created a chimeric ncRNA that functioned as a potent
transcriptional repressor. Biochemical studies showed that Alu
RNA incorporates with Pol II into complexes at promoter DNA
and blocks RNA synthesis, a mechanism similar to that used by
in human and mouse cells, Alu RNA and B2 RNA, respectively,
occupy the promoters of repressed genes with Pol II, thereby
establishing a unique mechanism of biological regulation. In
addition, we discovered the modular nature of Alu RNA, thereby
highlighting its similarity to protein transcriptional regulators.
The general repression of transcriptional activity, as well as
repression of specific genes, upon cellular heat shock has been
previously observed in organisms ranging from Drosophila to
humans (Findly and Pederson, 1981; Gilmour and Lis, 1985;
O’Brien and Lis, 1993; Sonna et al., 2002). Here we show that
Alu RNA is a repressor of select mRNA genes in response to
heat shock in human cells. Given that B2 RNA represses mRNA
transcription in response to heat shock in mouse cells, we
conclude that two ncRNAs that are unrelated in sequence and
secondary structure share a similar biological function. Levels
of SINE transcripts, including Alu RNA, increase during cellular
stresses other than heat shock (Liu et al., 1995), increase during
cells (Tang et al., 2005). It is possible that Alu RNA modulates
transcription during a variety of biological responses in addition
to heat shock. Alu RNA has also been implicated in regulating
other aspects of gene expression, including alternative splicing,
RNA editing, translation, and miRNA expression and function
(Hasler et al., 2007; Hasler and Strub, 2006). It is likely that Alu
RNA, and perhaps other SINE transcripts, serves as a master
regulator of gene expression by targeting many different steps.
A longstanding question in evolutionary biology is why the
number of SINE families differs between species (Schmid,
to rodent B1 SINEs; however, primates lack sequences that
resemble the rodent B2 SINE family (Schmid, 1998). Our results
provide insight into why human cells do not have B2-like SINEs.
We found that human Alu RNA is functionally related to mouse
B2 RNA, thus providing a rationale for why the SINEs encoding
them have been maintained in mammalian genomes. In addition,
we found that the naturally occurring human scAlu RNA and
mouse B1 RNA both bind Pol II but do not repress transcription
in vitro. Therefore, our results lead us to propose that the human
the functions of both families of mouse SINEs (i.e., Alu RNA
contains the activities of both B1 and B2 RNAs). It is possible
their interaction with Pol II, using a yet to be determined mecha-
ical function(s) of B1 and scAlu RNAs and whether it involves
interaction with Pol II.
The mechanism by which Alu RNA represses transcription in
vitro and in cells involves its incorporation with Pol II into stable
complexes at promoters. The observation that Alu RNA does not
repress transcription after Pol II has formed preinitiation com-
plexes (Figure 5B) suggests that once Pol II is engaged with
promoter DNA it is resistant to repression by Alu RNA. It is our
hypothesis that the repression domains in Alu RNA target the
DNA binding channel of Pol II and interfere with the polymerase
forming proper contacts with promoter DNA, which are required
for transcription to occur. Consistent with this general model of
repression, we have found that B2 RNA is able to block the asso-
ciation of Pol II with a DNA/RNA hybrid (Espinoza et al., 2007).
Because the sequences of the A and L regions, as well as the
region of B2 RNA required for repression, are quite different
from one another, we believe that the contacts between Pol II
and the ncRNA repression domains are not sequence specific
functions to inhibit transcription only after being tethered to Pol II
Two other RNA transcriptional repressors, one bacterial and
another that functions in yeast, have previously been found to
interfere with contacts between an RNA polymerase and
promoter DNA. The natural bacterial 6S RNA represses tran-
scription by mimicking the DNA in an open complex, thereby
competingwiththepromoter forbinding thepolymerase (Barrick
et al., 2005; Gildehaus et al., 2007; Trotochaud and Wassarman,
ural RNA aptamer (FC) that represses transcription was found to
bind the DNA binding channel of Pol II and likely prevents the
polymerase from properly engaging promoter DNA (Ketten-
berger et al., 2006; Thomas et al., 1997). Recently, the FC
aptamer was used to design an RNA scaffold that yeast Pol II
could use as a template to perform RNA-dependent RNA
synthesis (Lehmann et al., 2007). It remains to be determined
how the mechanisms of transcriptional repression by 6S RNA
and the FC aptamer precisely relate to those of Alu and B2
RNAs and whether the SINE RNAs can serve as templates for
RNA-dependent RNA synthesis.
A hallmark of protein regulators of Pol II transcription is the
existence of modular domains (e.g., DNA binding domains and
regulatory domains) that can be mixed and matched to create
chimeric regulatory proteins. Surprisingly, we found that Alu
RNA contains modular repression domains, which can be fused
to an ncRNA that binds Pol II (B1 RNA) to create transcriptional
repressors. This similarity to protein transcriptional regulators
is striking and provides evidence for commonality in evolution
of ncRNA and protein transacting transcription factors. Given
the vast numbers of ncRNAs of unknown function that have
Alu RNA Is a Modular Transcriptional Repressor
506 Molecular Cell 29, 499–509, February 29, 2008 ª2008 Elsevier Inc.
only recently been discovered in mammalian cells (as reviewed
by Mattick, 2005), it is likely that other ncRNAs control transcrip-
tion of specific genes by targeting Pol II in trans.
Alu RNA and B2 RNA would be expected to function as
general transcriptional repressors, as opposed to gene-specific
repressors, because they bind directly to Pol II. Some genes,
such as hsp70, are transcriptionally activated in response to
heat shock. Therefore, mechanisms must exist to overcome
SINE RNA repression at specific genes in mammalian cells.
Because we found that Alu and B2 RNAs are not at the activated
hsp70 promoter, perhaps a factor that can displace Alu or B2
RNA from Pol II is recruited to the promoters of heat shock-
induced genes, for example, another ncRNA, an RNase, or an
RNA helicase. An ncRNA named HSR1 has been found to play
a role in the transcriptional upregulation of heat shock-induced
genes by stimulating the trimerization of the transcriptional
activator HSF (Shamovsky et al., 2006). Thus, at least two
ncRNAs function to regulate the transcriptional response to
heat shock in mammalian cells. Currently, we can only imagine
the breadth of ncRNAs that regulate mammalian transcription.
Plasmid Construction and RNA Preparation
Constructs used are described in the Supplemental Experimental Procedures.
The full-length Alu sequence was that of the cDNA clone TS 103 (Shaikh et al.,
1997) known to be expressed as a 281 nt transcript in human cells. Mutant Alu
RNAs and the region(s) of Alu RNA contained in deletions are described in
Figure S4. RNA preparation was performed as previously described (Allen
et al., 2004). Models for RNA secondary structures were generated with
BayesFold software (Knight et al., 2004) guided by structural probing data.
In Vitro Transcription and Electrophoretic Mobility Shift Assays
Transcription factors (TBP, TFIIB, TFIIF, and Pol II) were purified as described
elsewhere (Weaver et al., 2005). In vitro transcription and abortive initiation as-
says in the reconstituted transcription system were performed as described
previously (Espinoza et al., 2004; Kugel and Goodrich, 2003). Repression
curves were obtained by fitting data with the following equation: relative
transcription = 1? [ncRNA] / (IC50+ [ncRNA]). EMSAs were performed as
previously described (Espinoza et al., 2004). More detailed descriptions of
transcription assays and EMSAs are provided in the Supplemental Experi-
Heat Shock and Isolation of Nuclear RNAs
HEK293 cells and NIH 3T3 cells were maintained in 5% CO2at 37?C in DMEM
containing 10% FBS, 100 U/ml penicillin, 100 mg/ml streptomycin, and 2 mM
L-glutamate. HEK293 cells were heat shocked for 60 min at 43?C and recov-
ered for 30 min at 37?C. NIH 3T3 cells were heat shocked for 15 min at 45?C
and recoveredfor 15–45 min at 37?C. Nuclei from bothcell types were isolated
by resuspending cells in 80 ml of NP buffer (2 mM MgCl2, 10 mM Tris-HCl [pH
7.4],10 mMNaCl, and 0.5% [v/v] NP-40) per1 million cells and incubating for 5
min on ice. Nuclei were harvested by centrifugation and washed once in an
equal volume of NP buffer. Nuclear RNA was extracted with Trizol Reagent
Detection of Nuclear RNAs
For RT-PCR, RNA samples isolated as described above were treated with
DNase I (1–4 units) at 37?C for 30 min and then heat inactivated. The RNA
was added to reactions containing 6 mM of random decamer primer, 12 units
of RNA guard (GE Lifesciences), and Moloney Murine Leukemia Virus Reverse
Transcriptase in RT buffer (25 mM KCl, 50 mM Tris [pH 7.5], 10 mM DTT,
3.5 mM MgCl2, 100 mg/ml BSA, and 0.5 mM of each dNTP). Reactions were
incubated at 42?C for 1 hr, then heat inactivated at 95?C for 3 min. Parallel
reactions were performed in the absence of reverse transcriptase. cDNA
was then titrated into PCR reactions containing 0.5 mM each of the forward
and reverse primers (see Table S1). Titrations of the cDNA into the PCR
reactions were performed to ensure that signals were within the linear
For northern blotting, RNAs were resolved by 6% (w/v) denaturing PAGE
and subsequently transferred to Hybond N+ membrane (GE Lifesciences).
32P-labeled probes (1 3 107cpm) were hybridized to the membrane overnight
at 56?C. For Alu RNA, either Alu RNA probes B, C, and D or Alu Probe 21-mer
was used (see Table S1). Membranes were washed once with 53 SSC, then
incubated at 56?C in 53 SSC for 30–60 min.
For antisense experiments, oligonucleotides (Alu Antisense and Alu Control,
see Table S1) were purchased with phosphorothioate linkages at the terminal
three positions of both ends (Invitrogen). The antisense oligonucleotide was
designed to avoid chance targeting of protein-encoding transcripts, with the
exception of sense Alu RNA sequences that are contained in the UTRs of
some Pol II transcripts. HEK293 cells were split 1 day prior to transfection in
T-25 flasks so that cells would be 40%–50% confluent at the time of transfec-
tion. Cells were transfected with 800 pmole of either Alu antisense or Alu
control oligonucleotide using Oligofectamine reagent (Invitrogen) per the
manufacturer’s instructions. Transfections were performed in 5 ml total media
per flask in the absence of serum and antibiotics. After 4 hr, 2.5 ml of DMEM
with 30% serum was added to each flask to bring the final concentration of
serum to10%.Cells recovered for 19hrat whichpoint they wereheat shocked
as described above. For antisense experimentsthat were assayed by northern
blotting, reagents were increased 7-fold and transfections were performed in
293 cells were transfected at 70% confluency in 12-well plates with 2.5 ml of
Lipofectamine 2000 reagent (Invitrogen), 200 ng of p(AP-1)5-E1b-luc, and 1
mg of either pUC-T7-Alu, pUC-T7-AluDADL, or pUC. Transfections proceeded
in the absence of antibiotic for 16 hr at 37?C, then cell lysates were prepared
with 13 passive lysis buffer (Promega). Firefly luciferase activities were
measured by using the dual luciferase reporter assay system (Promega).
room temperature with gentle shaking. Glycine was added to a final concen-
tration of 0.125 M, and the cells were incubated at room temperature for 5
min with gentle shaking. Cells were harvested in PBS. Nuclei were isolated
by resuspending 7.5 million cells in 600 ml of buffer A (3 mM MgCl2, 10 mM
Tris-HCl [pH 7.4], 10 mM NaCl, and 0.5% [v/v] NP-40) and incubating for
5 min on ice. Nuclei were harvested by centrifugation and washed once in
an equal volume of buffer A. Nuclei were then resuspended in 200 ml of buffer
B (50 mM Tris [pH 7.9], 10 mM EDTA, 0.2 mM PMSF, 1% SDS, protease inhib-
itors [Complete cocktail tablets, Roche], and 100 units/ml SUPERase?In
[Ambion]) and incubated on ice for 10 min. Three-hundred microliters of buffer
C (15 mM Tris [pH 7.9], 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.2 mM
PMSF, protease inhibitors, and 100 units/ml SUPERase?In) was then added,
and samples were sonicated using a cup sonicator filled with ice water (five
1.5 min bursts at a setting of 4.5, with 1.75 min between bursts). After
centrifugation, the supernatants were diluted with buffer C to a final volume
of 1.5 ml and stored in 300 ml aliquots.
For each precipitation, one 300 ml aliquot of sonicated chromatin was
precleared with 24 ml of protein A/G beads (Santa Cruz Biotechnology) that
were equilibrated in buffer D (15 mM Tris [pH 7.9], 150 mM NaCl, 1 mM
EDTA, 0.5% NP-40, 0.2 mM PMSF, and protease inhibitors). Twenty-five
pmole of a biotinylated antisense oligonucleotide against B2 RNA or Alu
RNA, or a biotinylated control oligonucleotide was added (see sequences in
Table S1). In some cases, samples were heated to 95?C and slowly cooled
to room temperature. Samples were nutated overnight at 4?C. Protein A/G
beads (Santa Cruz Biotechnology) were blocked by nutating the beads 4 hr
at 4?C in an equal volume of buffer D containing 400 mg/ml yeast RNA and
Alu RNA Is a Modular Transcriptional Repressor
Molecular Cell 29, 499–509, February 29, 2008 ª2008 Elsevier Inc. 507
at 4?C with 1 ml of antibody against biotin (ab6643-100, Abcam).
Forty-five microliters of beads with bound antibody were added to the
sample containing the oligonucleotide and incubated at 4?C for 2 hr. Beads
were washed sequentially with 200 ml of low-salt buffer (20 mM Tris [pH 7.9],
150 mM NaCl, 2 mM EDTA, 1% Triton X-100, and 0.1% SDS), 200 ml of
high-salt buffer (20 mM Tris [pH 7.9], 500 mM NaCl, 2 mM EDTA, 1% Triton
X-100, and 0.1% SDS), 200 ml of LiCl buffer (10 mM Tris [pH 7.9], 1 mM
EDTA, 1% deoxycholine, 1% NP-40, and 250 mM LiCl), and twice with
200 ml of TE (10 mM Tris [pH 7.9], 1 mM EDTA). One-hundred microliters of
elution buffer containing 1% SDS and 0.1 MNaHCO3was added to the beads,
and the mixture was nutated for 15 min at room temperature. The beads were
precipitated and the supernatant was transferred to a new tube. NaCl was
added to 0.2 M, and crosslinks were reversed by incubation at 65?C for 4 hr.
Samples were treated with Proteinase K (10 mg) for 1 hr at 37?C. After phenol
extraction and ethanol precipitation, samples were resuspended in 50 ml TE
and subjected to PCR (see Table S1 for primer sequences). All PCR reactions
that showed a signal were confirmed to be in the linear range.
ChIP assays were performed with the method described for ChOP assays,
with the following modifications. One-hundred microliter aliquot of sonicated
chromatin and 15 ml of preblocked protein A/G beads were used for each
assay.Two microliters of a-Pol II antibody was added tosonicated, precleared
chromatin, and nutated overnight at 4?C (sc-9001X [Santa Cruz Biotech-
nology] for HEK293 cells and either sc899 [Santa Cruz Biotechnology] or
8WG16 for NIH 3T3 cells). Antisense oligonucleotides and SUPERase?In
were omitted. All PCR reactions that showed a signal were confirmed to be
in the linear range.
Supplemental Data include Supplemental Experimental Procedures, four
figures, and two tables and can be found with this article online at http://
We thank Tiffany Allen for technical assistance. This research was supported
by a Public Health Service grant (R01 GM068414) from the National Institutes
of General Medical Sciences.
Received: July 20, 2007
Revised: October 24, 2007
Accepted: December 20, 2007
Published: February 28, 2008
Allen, T.A., Von Kaenel, S., Goodrich, J.A., and Kugel, J.F. (2004). The
SINE-encoded mouse B2 RNA represses mRNA transcription in response to
heat shock. Nat. Struct. Mol. Biol. 11, 816–821.
Barrick, J.E., Sudarsan, N., Weinberg, Z., Ruzzo, W.L., and Breaker, R.R.
(2005). 6S RNA is a widespread regulator of eubacterial RNA polymerase
that resembles an open promoter. RNA 11, 774–784.
Espinoza, C.A., Allen, T.A., Hieb, A.R., Kugel, J.F., and Goodrich, J.A. (2004).
B2 RNA binds directly to RNA polymerase II to repress transcript synthesis.
Nat. Struct. Mol. Biol. 11, 822–829.
Espinoza, C.A., Goodrich, J.A., and Kugel, J.F. (2007). Characterization of the
structure, function, and mechanism of B2 RNA, an ncRNA repressor of RNA
polymerase II transcription. RNA 13, 583–596.
Findly, R.C., and Pederson, T. (1981). Regulated transcription of the genes for
actin and heat-shock proteins in cultured Drosophila cells. J. Cell Biol. 88,
Fornace, A.J., Jr.,and Mitchell, J.B.(1986). Inductionof B2RNA polymerase III
transcription by heat shock: enrichment for heat shock induced sequences in
rodent cells by hybridization subtraction. Nucleic Acids Res. 14, 5793–5811.
Gildehaus, N., Neusser, T., Wurm, R., and Wagner, R. (2007). Studies on the
function of the riboregulator 6S RNA from E. coli: RNA polymerase binding,
inhibition of in vitro transcription and synthesis of RNA-directed de novo
transcripts. Nucleic Acids Res. 35, 1885–1896.
Gilmour, D.S., and Lis, J.T. (1985). In vivo interactions of RNA polymerase II
with genes of Drosophila melanogaster. Mol. Cell. Biol. 5, 2009–2018.
Goodrich, J.A., and Kugel, J.F. (2006). Non-coding-RNA regulators of RNA
polymerase II transcription. Nat. Rev. Mol. Cell Biol. 7, 612–616.
Hasler, J., and Strub, K.(2006). Alu elements as regulators of gene expression.
Nucleic Acids Res. 34, 5491–5497.
Hasler, J., Samuelsson, T., and Strub, K. (2007). Useful ‘junk’: Alu RNAs in the
human transcriptome. Cell. Mol. Life Sci. 64, 1793–1800.
Kazazian, H.H., Jr. (2004). Mobile elements: drivers of genome evolution.
Science 303, 1626–1632.
Kettenberger, H., Eisenfuhr, A., Brueckner, F., Theis, M., Famulok, M., and
Cramer, P. (2006). Structure of an RNA polymerase II-RNA inhibitor complex
elucidates transcription regulation by noncoding RNAs. Nat. Struct. Mol.
Biol. 13, 44–48.
Knight, R., Birmingham, A., and Yarus, M. (2004). BayesFold: rational 2
degrees folds that combine thermodynamic, covariation, and chemical data
for aligned RNA sequences. RNA 10, 1323–1336.
Kramerov, D.A., and Vassetzky, N.S. (2005). Short retroposons in eukaryotic
genomes. Int. Rev. Cytol. 247, 165–221.
Kramerov, D.A., Tillib, S.V., Lekakh, I.V., Ryskov, A.P., and Georgiev, G.P.
(1985). Biosynthesis and cytoplasmic distribution of small poly(A)-containing
B2 RNA. Biochim. Biophys. Acta 824, 85–98.
Kugel, J.F., and Goodrich, J.A. (2003). In vitro studies of the early steps of RNA
synthesis by human RNA polymerase II. Methods Enzymol. 370, 687–701.
Labuda, D., Sinnett, D., Richer, C., Deragon, J.M., and Striker, G. (1991).
Evolution of mouse B1 repeats: 7SL RNA folding pattern conserved. J. Mol.
Evol. 32, 405–414.
Lehmann, E., Brueckner, F., and Cramer, P. (2007). Molecular basis of
RNA-dependent RNA polymerase II activity. Nature 450, 445–449.
Li, T., Spearow, J., Rubin, C.M., and Schmid, C.W. (1999). Physiological
stresses increase mouse short interspersed element (SINE) RNA expression
in vivo. Gene 239, 367–372.
Liu, W.M., Chu, W.M., Choudary, P.V., and Schmid, C.W. (1995). Cell stress
and translational inhibitors transiently increase the abundance of mammalian
SINE transcripts. Nucleic Acids Res. 23, 1758–1765.
Maraia, R.J. (1991). The subset of mouse B1 (Alu-equivalent) sequences
expressed as small processed cytoplasmic transcripts. Nucleic Acids Res.
Maraia, R.J., Driscoll, C.T., Bilyeu, T., Hsu, K., and Darlington, G.J. (1993).
Multiple dispersed loci produce small cytoplasmic Alu RNA. Mol. Cell. Biol.
Matera, A.G., Hellmann, U., and Schmid, C.W. (1990). A transpositionally and
transcriptionally competent Alu subfamily. Mol. Cell. Biol. 10, 5424–5432.
Mattick, J.S. (2005). The functional genomics of noncoding RNA. Science 309,
O’Brien, T., and Lis, J.T. (1993). Rapid changes in Drosophila transcription
after an instantaneous heat shock. Mol. Cell. Biol. 13, 3456–3463.
Schmid, C.W.(1996).Alu:structure,origin, evolution, significanceand function
of one-tenth of human DNA. Prog. Nucleic Acid Res. Mol. Biol. 53, 283–319.
Schmid, C.W. (1998). Does SINE evolution preclude Alu function? Nucleic
Acids Res. 26, 4541–4550.
Schmid, C.W. (2003). Alu: a parasite’s parasite? Nat. Genet. 35, 15–16.
Schmid, C.W., and Jelinek, W.R. (1982). The Alu family of dispersed repetitive
sequences. Science 216, 1065–1070.
Alu RNA Is a Modular Transcriptional Repressor
508 Molecular Cell 29, 499–509, February 29, 2008 ª2008 Elsevier Inc.
Shaikh, T.H., Roy, A.M., Kim, J., Batzer, M.A., and Deininger, P.L. (1997). Download full-text
cDNAs derived from primary and small cytoplasmic Alu (scAlu) transcripts.
J. Mol. Biol. 271, 222–234.
Shamovsky, I., Ivannikov, M., Kandel, E.S., Gershon, D., and Nudler, E. (2006).
RNA-mediated response to heat shock in mammalian cells. Nature 440,
Singh, K., Carey, M., Saragosti, S., and Botchan, M. (1985). Expression of
enhanced levels of small RNA polymerase III transcripts encoded by the B2
repeats in simian virus 40-transformed mouse cells. Nature 314, 553–556.
Sinnett, D., Richer, C., Deragon, J.M., and Labuda, D. (1991). Alu RNA
secondary structure consists of two independent 7 SL RNA-like folding units.
J. Biol. Chem. 266, 8675–8678.
Sonna, L.A., Gaffin, S.L., Pratt, R.E., Cullivan, M.L., Angel, K.C., and Lilly, C.M.
(2002). Effect of acute heat shock on gene expression by human peripheral
blood mononuclear cells. J. Appl. Physiol. 92, 2208–2220.
Tang, R.B., Wang, H.Y., Lu, H.Y., Xiong, J., Li, H.H., Qiu, X.H., and Liu, H.Q.
(2005). Increased level of polymerase III transcribed Alu RNA in hepatocellular
carcinoma tissue. Mol. Carcinog. 42, 93–96.
Thomas, M.C., and Chiang, C.M. (2006). The general transcription machinery
and general cofactors. Crit. Rev. Biochem. Mol. Biol. 41, 105–178.
Thomas, M., Chedin, S., Carles, C., Riva, M., Famulok, M., and Sentenac, A.
(1997). Selective targeting and inhibition of yeast RNA polymerase II by RNA
aptamers. J. Biol. Chem. 272, 27980–27986.
Trotochaud, A.E., and Wassarman, K.M. (2005). A highly conserved 6S RNA
structure is required for regulation of transcription. Nat. Struct. Mol. Biol. 12,
Wassarman, K.M., and Saecker, R.M. (2006). Synthesis-mediated release of
a small RNA inhibitor of RNA polymerase. Science 314, 1601–1603.
Weaver, J.R., Kugel, J.F., and Goodrich, J.A. (2005). The sequence at specific
positions in the early transcribed region sets the rate of transcript synthesis by
RNA polymerase II in vitro. J. Biol. Chem. 280, 39860–39869.
Alu RNA Is a Modular Transcriptional Repressor
Molecular Cell 29, 499–509, February 29, 2008 ª2008 Elsevier Inc. 509