Spt4 Is SelectivelyRequired for Transcriptionof Extended Trinucleotide RepeatsChia-Rung Liu,1 Chuang-Rung Chang,2 Yijuang Chern,3 Tzu-Han Wang,1 Wen-Chieh Hsieh,1 Wen-Chuan Shen,1

Chi-Yuan Chang,1 I-Chieh Chu,1 Ning Deng,4 Stanley N. Cohen,4,* and Tzu-Hao Cheng1,5,*1Institute of Biochemistry and Molecular Biology, National Yang-Ming University, No. 155, Section 2, Linong Street, Taipei, Taiwan,

Republic of China2Institute of Biotechnology, National Tsing-Hua University, Hsinchu, Taiwan, Republic of China3Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan, Republic of China4Department of Genetics, Stanford University School of Medicine, Stanford, CA 94305, USA5Brain Research Center, National Yang-Ming University, Taipei, Taiwan, Republic of China*Correspondence: thcheng@ym.edu.tw (T.-H.C.), sncohen@stanford.edu (S.N.C.)

DOI 10.1016/j.cell.2011.12.032

SUMMARY

Lengthy trinucleotide repeats encoding polyglut-amine (polyQ) stretches characterize the variantproteins of Huntington’s disease and certain otherinherited neurological disorders. Using a phenotypicscreen to identify events that restore functionality topolyQ proteins in S. cerevisiae, we discovered thattranscription elongation factor Spt4 is required totranscribe long trinucleotide repeats located eitherin ORFs or nonprotein-coding regions of DNAtemplates. Mutation of SPT4 selectively decreasedsynthesis of and restored enzymatic activity to ex-panded polyQ protein without affecting protein lack-ing long-polyQ stretches. RNA-seq analysis revealedlimited effects of Spt4 on overall gene expression.Inhibition of Supt4h, the mammalian ortholog ofSpt4, reduced mutant huntingtin protein in neuronalcells and decreased its aggregation and toxicitywhile not altering overall cellular mRNA syn-thesis. Our findings identify a cellular mechanismfor transcription through repeated trinucleotidesand a potential target for countermeasures againstneurological disorders attributable to expandedtrinucleotide regions.

INTRODUCTION

Abnormal expansion of trinucleotides in genes expressed in

neurons of the cerebral cortex and corpus striatum underlies

the pathological effects of Huntington’s Disease (HD) and

other inherited neurodegenerative disorders. (HD Collaborative

Research Group, 1993; Hands et al., 2008; Ross, 2002; Zoghbi

and Orr, 2000). The huntingtin (Htt) gene normally includes

8–25 CAG repeats in the first exon, but mutant alleles can

690 Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc.

contain 36 or more. As the CAG repeat mutation is dominant,

inheritance of one copy of the mutated huntingtin gene is

sufficient to produce HD. The frequency of the HD variant in

persons of western European descent is about 1 in 10,000.

Expansion of CAG repeats in protein-coding regions gener-

ates lengthy stretches of polyglutamine (polyQ), which promote

protein aggregation, inclusion body formation, and eventually

neuronal cell death in the brains of afflicted persons (Zoghbi

and Orr, 2000). While aggregation of the mutant protein is

thought to underlie HD pathogenesis (Davies et al., 1997;

DiFiglia et al., 1997; Mangiarini et al., 1996), aggregates have

also been viewed as a protective response against the toxicity

of misfolded huntingtin (e.g., Tydemers et al., 2010; Gutekunst

et al., 1999). Wild-type huntingtin, which is required for both

embryogenesis and normal neurological function in adulthood

(Dragatsis et al., 2000; Nasir et al., 1995), can be sequestered

by interaction with the mutant protein (Busch et al., 2003),

possibly contributing to the development of striatal neuropa-

thology (Van Raamsdonk et al., 2005).

Spt4 is a 12 kDa transcription elongation factor that can

aid RNA polymerase II processivity by reducing dissociation of

polymerase from the template (Rondon et al., 2003; Mason

and Struhl, 2005; Hirtreiter et al., 2010). It was first identified

in Saccharomyces cerevisiae (Winston et al., 1984) and is highly

conserved among eukaryotes (Guo et al., 2008; Hartzog et al.,

1996; Wenzel et al., 2010). Both Spt4 and Supt4h, its 14 kDa

human ortholog, contain an N-terminal zinc-finger domain that

at least in yeast is essential for biochemical activity (Malone

et al., 1993). Here we report that transcription of genes contain-

ing long repeats of CAG or other trinucleotides located in either

protein-coding or transcribed noncoding regions of templates

is selectively facilitated by the actions of Spt4 and Supt4h.

Diminished elongation of expanded-CAG-repeat transcripts in

yeast or animal cells defective in Spt4 or Supt4h leads to

reduced synthesis of proteins containing long polyQ stretches,

but not of proteins containing short polyQ regions. This event

decreases aggregation of long polyQ protein and restores

functionality.

C

D

E

WT

Glucose

hsp104

Galactose

ADE2 Vector

25Q-

ADE2

97Q-

ADE2

AIR

CAIR

Ade2p

FLAGADE2

25Q-

ADE2

97Q-

ADE2

Ade2

25Q

97Q

GAL1

A

B

WT

415

hsp104

415

spt4

415

spt4

415-SPT4

97Q-

ADE2

415-SPT4

WT

hsp104

spt4

spt4

NC CA

IB: -Tub FLAG

100%

4±5%

47±7%

96±2%

97Q-ADE2

Figure 1. spt4 Restores Enzymatic Activity to and Decreases Aggregation of 97Q-Ade2 Protein

(A) Principle of assay: Ade2 deficiency results in AIR accumulation and change in colony color from white to red.

(B) Location of polyQ regions in ADE2, 25Q-ADE2, and 97Q-ADE2 constructs. Fusion proteins contain an N-terminal FLAG epitope and are expressed under

GAL1 promoter control.

(C) Colony color phenotypes of cells expressing indicated constructs. Colonies formed by either wild-type (W303-1A) or isogenic hsp104D cells possess the

ade2-1 mutant allele and are red in glucose media.

(D) Colony color phenotypes of cells expressing 97Q-Ade2 in the absence or presence Spt4 function. The 97Q-ADE2 plasmid and either empty vector 415 or the

Spt4 expression construct 415-SPT4 were introduced by cotransformation. spt4D and hsp104D were derived from wild-type (WT) cells by deleting the entire

open reading frame. Cells were grown on medium containing galactose as the sole carbon source.

(E) Analysis of 97Q-Ade2 aggregation in spt4 mutant cells. Lysates were collected from cells shown in (D) and loaded onto a cellulose acetate CA membrane,

which traps only aggregated protein. PolyQ-Ade2 was detected by immunoblotting using anti-FLAG antibody. a-Tubulin (a-Tub) served as a loading control and

was assayed by slot blot using nitrocellulose NCmembranes. Percentage aggregation is shown relative toWT cells. The values shown aremeans ±SD from three

independent experiments. See also Figure S1.

RESULTS

Identification of Yeast Mutations that RestoreEnzymatic Activity to a Protein Containing Long PolyQExpansionsProteins containing polyQ expansions encoded by lengthy

repeats of the trinucleotide sequence CAG or CAA normally

are dysfunctional (Krobitsch and Lindquist, 2000; Ordway

et al., 1997; Scherzinger et al., 1999). To discover genetic events

that might restore enzymatic activity to polyQ proteins encoded

by lengthy repeats of the trinucleotides CAG or CAA, we estab-

lished a color-based assay that identifies colonies that can

convert p-ribosylamino imidazole (AIR) to p-ribosylamino imid-

azole carboxylate (CAIR) (Jones and Fink, 1982). In yeast cells

mutated the ADE2 gene, AIR accumulation results in red colored

colonies (Figure 1A). We found that introduction of a short polyQ

(25Q) segment into ectopic Ade2 protein produced in a chromo-

somally-mutated ade-2 strain under galactose promoter control

allowed enzymatic function, whereas a corresponding 97Q

protein was dysfunctional (Figures 1B and 1C), enabling us to

screen for transposon insertion mutants that restore enzymatic

activity to the 97Q protein.

Examination of�180,000 colonies identified nearly 200 candi-

dates in which colony color changed from red to pink or white.

Sequencing of PCR-amplified DNA determined that yeast cells

from most of these colonies contained contractions of the 97Q

repeat. In other isolates, a transposon was inserted into genes

encoding either Hsp104 or [PIN+] prion protein Rnq1, which

can alter folding and aggregation of polyQ proteins (Krobitsch

and Lindquist, 2000;Meriin et al., 2002) but have not been known

to affect polyQ protein function. Additionally, our screen identi-

fied a clone (clone 23–44) in which Hsp104 and Rnq1 were intact

and no contraction of 97Q-Ade2 had occurred. This clone con-

tained a transposon insertion in the SPT4 gene.

Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc. 691

A

B

C

D

E

F

G

H

Figure 2. Spt4 Regulates Expression and Aggregation of Proteins Containing Expanded PolyQ Repeats

(A) Expression of mRNA encoding 97Q- or 25Q-Ade2 was determined by Northern blotting using an oligonucleotide probe that recognizes the polyQ-coding

sequence in ADE2 mRNA. After normalization using 7SL RNA SCR1, ADE2 mRNA abundance was set as 100% in WT cells expressing 97Q-ADE2. Relative

mRNA abundance in each cell type is shown.

(B) Analysis of cellular 97Q- and 25Q-Ade2 protein expression and aggregation. Procedures are described in legend for Figure 1 and polyQ-Ade2 protein

abundance was determined by slot blot assay.

692 Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc.

The role of spt4 in resorting functionality to the 97Q-Ade2

protein was confirmed by expressing 97Q-Ade2 in an ade2-1

mutant strain containing an independently introduced spt4

deletion (spt4D). Like clone 23–44, which showed no contrac-

tion of the introduced trinucleotide repeat and normal expres-

sion of Hsp104 protein and [PIN+] prion protein (Figure S1

available online), this strain showed a change in colony color

and reduced formation of polyQ-dependent protein aggregates

(Figures 1D and 1E). Moreover, spt4 deletion was associated

with decreased abundance of 97Q-ADE2 mRNA (Figure 2A)

and protein (Figure 2B), which as mRNA half life determina-

tions indicated had not resulted from any alteration in tran-

script turnover (Figure 2C). Importantly, unlike yeast mutations

that result in a general reduction in mRNA synthesis by RNA

polymerase II (Rondon et al., 2003), mutation of SPT4 had little

effect on the production of Ade2 protein that contains only

a short polyQ stretch (i.e., 25Q-Ade2) (Figure 2B), or on the

color of yeast colonies expressing this protein (Figure 2H).

Chromatin immunoprecipitation assays showed that the

amount of RNA polymerase II bound to transcript sequences

downstream of a stretch of 99 contiguous CAG repeats was

reduced by spt4 deficiency, whereas occupancy of DNA seg-

ments upstream or downstream from a segment containing

only 29 CAG repeats was similar in SPT4+ and spt4D cells (Fig-

ure 3A). We conclude from this finding that Spt4 is required for

the transcriptional machinery to proceed through a long ex-

pansion of the CAG repeat sequence, but is dispensable for

transcription of a region containing a short stretch of CAG

repeats.

Decreased Production of Transcripts Encodingan Expanded PolyQ Region Accounts for Restorationof Ade2 Protein FunctionalityTo learn whether reduced expression of 97Q-Ade2 per se can

explain restoration of enzymatic function to this protein, we

tested the effects of varying 97Q-ADE2 expression using

glucose to inhibit transcription from the GAL1 promoter. In-

creasing glucose resulted in decreased redness of colonies

and a corresponding decrease in 97Q-Ade2 aggregates (Figures

2D–2F), arguing that the 97Q-Ade2 expression level modulates

both enzymatic activity and cellular aggregation of this protein.

Whereas it has been believed that dysfunction of proteins con-

taining long polyQ stretches results from an inherent propensity

to misfold (Krobitsch and Lindquist, 2000; Ordway et al., 1997;

Scherzinger et al., 1999), restoration of biological activity to

97Q-Ade2 by spt4 deletion suggests that this long-polyQ protein

can fold properly enough to function enzymatically if production

(C) Decay of polyQ-ADE2 mRNAs in WT and spt4D cells. Total RNA was isolated

expression. mRNA abundance was analyzed by Northern blotting and the data we

results from three separate experiments. Mean values and ± SDs are indicated.

(D) WT cells, precultured in raffinose, were grown in media containing the indi

expression under control of the glucose-repressed GAL1 promoter (Biggar and

analyzed by Northern blotting. Relative 97Q-ADE2 mRNA abundance is shown f

(E) Expression and aggregation of 97Q-Ade2 protein as shown in (D) were deter

(F) Colony color phenotypes of 97Q-ADE2-expressing cells cultured as indicated

cells containing the empty vector or galactose-inducible ADE2 were included as

(G) Slot blot assay of 97Q-Ade2 and 25Q-Ade2 protein expression in thp2D cells

(H) Colony color phenotypes of thp2D cells expressing either 97Q- or 25Q-Ade2

is simply decreased to a level sufficiently low to diminish its

aggregation. Our finding that reduced expression of 97Q-Ade2

leads to reduced aggregation of the protein in cells parallels the

observation by Scherzinger et al. (Scherzinger et al., 1999) that

in vitro aggregation of polyQ protein is affected by its concentra-

tion. The effect of Spt4 on elongation of transcripts having

lengthy tri-nucleotide repeats is not restricted to Ade2-encoding

transcripts, as deletion of SPT4 also resulted in delayed and

reduced aggregation of GFP protein encoded by a construct

containing 65 repeats of CAG (i.e., 65Q-eGFP) (Figure S2).

Effect of Spt4 on Transcription through LengthyStretches of Repeated Nucleotides in Nonprotein-Coding RegionsTrinucleotide repeat expansions can also occur in noncoding

regions of genes, leading to accumulation of toxic or aberrantly

localized transcripts associated with ‘‘nonpolyglutamine’’ human

neurodegenerative or neuromuscular disorders (Wojciechowska

and Krzyzosiak, 2011). We tested the ability of Spt4 to affect

synthesis of ADE2 transcripts containing multiple CAG repeats

in the 50UTR, as occurs, for example, in certain spinocerebellar

ataxias (Holmes et al., 1999; Zoghbi and Orr, 2000). As seen in

Figure 3B, deletion of SPT4 affected the cellular abundance of

mRNA transcripts containing 105 CAG repeats in the 50UTR but

had little effect on transcripts containing 29 repeats of CAG at

the same location. Reduction of mRNA by an expanded repeat

in the 50UTR was comparable to the reduction observed for

a similar number of repeats in the protein coding region (99Q-

Ade2, Figure S2). Moreover, consistent with evidence that cells

lacking Spt4 show a decline in RNA polymerase II processivity

as the transcription complex proceeds along the template

(Mason and Struhl, 2005), we observed that deletion of SPT4

had an even more prominent effect on transcription when a

CAG repeat was introduced into the 30UTR region of the tran-

script; there, short CAG repeats also affected the production

of transcripts (Figure 3C). Transcription of templates contain-

ing a lengthy stretch of CTG or CAA or long homopolymerA

sequences also was decreased by deletion of SPT4 (Figures

3D–3F).HomopolymerAhad thegreatest effect on theproduction

of ADE2 mRNA of any of the tested insertions; a sequence as

short as 72 adenine residues almost totally blocked transcription

in the spt4 deletion mutant. We were unable to stably maintain

constructs containing extended homopolymerG regions in vivo.

However, the results we obtained demonstrate that the require-

ment for Spt4 to mediate transcription of different kinds of trinu-

cleotide repeat expansions is affected by the nature of the

repeated sequence, its length, and its location in the transcript.

at the indicated times after the addition of glucose to stop polyQ-ADE2 gene

re normalized to SCR1. RNA decay profiles in the bottom panel show averaged

cated glucose concentrations plus 2% galactose to induce 97Q-ADE2 gene

Crabtree, 2001). Cells were collected after 12 hr and 97Q-ADE2 mRNA was

or WT and spt4D cells.

mined by slot blot (NC) and filter-trap (CA) assay, respectively.

in (D). W303-1A cells (labeled as WT) carry the ade2-1mutant allele and these

a control.

.

. See also Figures S2 and S3.

Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc. 693

± ±

± ±

± ±

± ±

± ±

±

A

B

C

D

E

F

Figure 3. spt4 Mutant Cells Show Impaired Transcription Elongation through a Long Stretch of CAG Repeats and Have TranscriptionalDefects in Other Highly Repetitive Sequences

(A) Chromatin immunoprecipitation was used to assay RNA polymerase II present on ADE2 containing either 99 or 29 CAG repeats. The fold change was

determined by qPCR and is relative to the values obtained inWT cells. Probe Awas used to detect theDNA fragment upstream of CAG repeats, whereas probes B

and C detected DNA segments 30 to CAG repeats. Error bars represent the SD.

(B) Effects of 105 or 29 CAG repeat insertions into 50UTR on transcript abundance were analyzed by Northern blotting using an oligonucleotide probe that

detected the ADE2 coding sequence. Percentages indicate ADE2 mRNA abundance in spt4D cells relative to WT cells normalized against SCR1.

(C) Same as (B) except that 101 or 29 CAG repeats were introduced into the 30UTR.(D) Same as (B) for 105 or 26 CTG repeats in Ade2 protein-coding region.

(E) Same as (D) for 90 or 24 CAA repeats.

(F) Same as (E) for 74, 24, or 10 AAA trinucleotide repeats.

Spt4 Deficiency Has Limited Effects on Overall GeneExpressionThe THO complex, which is formed by interaction of the Hpr1,

Mft1, Tho2, and Thp2 proteins, affects both elongation and proc-

694 Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc.

essivity by RNA polymerase II (Rondon et al., 2003; Mason and

Struhl, 2005) and is required globally for production of yeast

gene transcripts, including those that contain multiple tandemly

repeated segments (Voynov et al., 2006). However, in contrast to

what we observed for spt4D cells, yeast cells defective in the

THO complex (i.e., thp2D) showed severely impaired production

of Ade2 protein containing 25Q repeats as well as of the Ade2

variant containing 97Q repeats (Figure 2G), and failed to show

complementation of ade2 deficiency by either of the 97Q-

ADE2 or 25Q-ADE2 construct (Figure 2H). Thus, unlike Spt4,

THO does not differentiate between transcripts containing short

versus long stretches of a CAG trinucleotide repeat. Instead THO

deficiency results in a general impairment in protein synthesis

rather than selective reduction of synthesis of protein containing

an expanded polyQ region.

The ability of spt4 deletion to affect gene expression selectively

rather then generally was further evident from experiments that

employed RNA-seq analysis, which involves parallel en masse

sequencing of short cDNAs generated randomly from cellular

mRNAs (Cloonan and Grimmond, 2008; Nagalakshmi et al.,

2010). Transcription of most yeast genes was not significantly

affected by spt4 deletion, and only 11 genes were sufficiently

affected to reduce expression to one third of normal (p < 0.01)

(Table S1). 44 genes showed a prominent (>4-fold) increase in

RNA-seq reads in spt4D cells (p < 0.01) (Table S1). Whether the

effect of spt4 deletion on expression of these downregulated and

upregulated genes is direct or indirect has not been determined.

Reduced Aggregation of Long-PolyQ Ade2 Enhancesthe Function of Coexpressed Ade2-Containing ShortPolyQ StretchesInteraction of proteins containing an expanded polyQ region

with proteins that include a short nonpathogenic stretch of

polyQ results in coaggregation of both proteins in mammalian

cells and yeast (Duennwald et al., 2006; Kazantsev et al.,

1999). To learn whether spt4-mediated reduction of long-polyQ

protein aggregation affects the status of concomitantly-present

short polyQ protein, we expressed 99Q-Ade2 together with

29Q-Ade2 tagged at the C-terminal end with hemaglutinin

(29Q-Ade2-HA), and examined the effects of SPT4 mutation.

In wild-type W303-1A cells, such coexpression resulted in

phenotypic dominance of the dysfunctional 99Q-Ade2 protein

(Figure 4A) and the appearance of 29Q-Ade2-HA, which nor-

mally is soluble in the absence of the long polyQ protein, in

cellular aggregates (Figure 4B). However, Spt4 deficient cells,

which as seen in Figure S2 show reduced 99Q-Ade2 protein

aggregation and restoration of 99Q-Ade2 functionality, also

showed a decreased amount of 29Q-Ade2-containing protein

aggregates (Figure 4B), even though the amount of 29Q-Ade2

protein and mRNA was not itself affected by the lack of Spt4

(Figures 4B and 4C). These results demonstrate that downregu-

lation of 99Q-Ade2 protein production can affect the aggrega-

tion of protein encoded by a coexisting allele containing short

CAG repeats.

Role of Supt4h in Expression of Long PolyQ Proteinsin NeuronsSPT4 is highly conserved between yeast and mammalian

cells, and yeast spt4 mutations can be functionally comple-

mented by the mammalian cell ortholog Supt4h (Hartzog et al.,

1996). Using siRNA to reduce expression of Supt4h in the

murine striatal neuron cell line ST14A (Ehrlich et al., 2001) to

40% of normal, we found that the number of individual

cells showing foci of aggregation of an eGFP protein contain-

ing 81 repeats of glutamine was approximately one half the

number observed in a cell population that expresses Supt4h

normally (Figures 5A and 5B). Overall abundance of the 81Q-

eGFP was reduced in Supt4h-deficient cells, and the fraction

of eGFP protein present in aggregates was less than 30% of

what was observed in a cell population transfected with

negative control siRNA (Figure 5C). By contrast, expression

of 7Q-eGFP was affected only minimally by Supt4h knock-

down (Figures 5A–5C). Consistent with the above observa-

tions, neuronal cell toxicity resulting from accumulation of

protein containing expanded polyQ stretches (Li et al., 2000;

Varma et al., 2007) was reversed by Supt4h knockdown (Fig-

ure 5D). Conversely, siRNA directed against Supt4h had no

detectable effect on the viability of cells producing the 7Q-

eGFP protein.

Effects of Supt4h knockdown specifically on the production of

mutant huntingtin (Htt) were observed in striatal neurons derived

from homozygous HdhQ7/Q7 and HdhQ111/Q111 knock-in mice

(Trettel et al., 2000). Whereas RT-PCR and quantitative real-

time RT-PCR showed that HttQ7 mRNA and a-Tubulin mRNA

were only slightly affected in cells treated with siRNA directed

against Supt4h, the abundance of mutant HttQ111 mRNA was

substantially reduced (Figures 6A and 6B) in cells showing a

quantitatively similar reduction of Supt4h mRNA. Concurrent

with the observed reduction of HttQ111 mRNA was a decrease

in HttQ111 protein; however, production of HttQ7 and Tbp, which

contain only short stretches of glutamine, was not detectably

affected by Supt4h downregulation (Figure 6C). Importantly,

Supt4h knockdown in heterozygous HdhQ7/Q111 cells showed

that HttQ111 was decreased in cells transfected with Supt4h

siRNA, but that the same extent of Supt4h knockdown had little

effect on HttQ7 expression (Figure 6D). These findings indicate

that Supt4h is differentially required for expression of the mutant

Htt allele in mouse striatal neurons. Consistent with our finding

that deletion of SPT4 has very limited effects on the yeast tran-

scriptome, RNA-seq analysis of mRNAs from rat and mouse

neuronal cells showed no significant perturbation of the overall

gene transcription profiles (data available at http://www.ncbi.

nlm.nih.gov/geo/query/acc.cgi?acc=GSE33497). However, con-

sistent with earlier evidence that the Spt4/5 complex and other

proteins that facilitate transcription elongation also affect RNA

splicing (Lindstrom et al., 2003; Xiao et al., 2005; Oesterreich

et al., 2011), we observed that reduction of Supt4h expression

in mouse neurons altered representation of some exons in the

population of mature mRNAs we analyzed by RNA-seq (data

available at http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?

acc=GSE33497).

DISCUSSION

Our investigations reveal the role of transcription elongation

factors Spt4 and Supt4h in specifically promoting expression

of genes containing lengthy trinucleotide repeats. Whereas it

has been suggested that protein containing long polyQ tracts

encoded by CAG or CAA repeats is inherently prone to mis-

folding (Krobitsch and Lindquist, 2000; Ordway et al., 1997),

Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc. 695

±± ± ±

A

B

C

Figure 4. Coaggregation of Short PolyQ-Containing Proteins Is Reduced by spt4 Deletion

(A) Yeast colony color phenotypes. 29Q-ADE2-HA and the additional indicated plasmid constructs (see also Figure S2) introduced by cotransformation.

(B) Expression and aggregation of polyQ-Ade2 proteins were assessed by slot blot (NC) and filter-trap (CA) assay, respectively. Anti-FLAG antibody was used to

probe 29Q-Ade2-HA, 29Q-Ade2, and 99Q-Ade2, whereas anti-HA antibody detected only 29Q-Ade2-HA. a-Tubulin (a-Tub) served as loading control.

(C) Northern blot analysis of transcripts encoding expanded and short polyQ-Ade2 proteins. Transcript positions are indicated by arrows.

our results indicate that such protein can fold properly enough to

exhibit enzymatic activity if production is decreased. In yeast,

sufficient activity was restored to Ade2 protein containing

a long polyQ repeat by deletion of SPT4 to functionally comple-

ment ade2 deficiency in the colorimetric assay we designed.

Similarly, decrease of Supt4h reduced the production of protein

encoded by an eGFP variant containing a lengthy polyQ region,

and reduced the toxicity of Htt protein containing a long polyQ

segment. The effects of Spt4/Supt4h deficiency are allele

specific, and reduction of Supt4h decreased the expression of

a variant Htt allele without significantly affecting transcription

from coexpressed non mutant Htt. As decreased production of

variant Htt reduces its coaggregation with coexpressed wild-

type Htt protein (Duennwald et al., 2006), our data raise the pros-

pect that interference with the actions of Supt4h may enhance

functionality of wild-type Htt. Whether function can be restored

generally to long polyQ proteins decreasing their expression

and aggregation is not known.

696 Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc.

Although Spt4 commonly is viewed as a broadly-acting

transcription elongation factor, our results indicate that its tran-

script elongating actions—unlike those of the THO complex—

are highly specific. Consistent with this specificity, RNA-seq

analyses indicate that spt4 deletion affected only a small frac-

tion of the yeast transcriptome (Table S1). A line of Supt4h

heterozygous knockout mice found by quantitative RT-PCR

analysis to express Supt4h at half the normal level did

not show a detectable phenotypic abnormality (our unpub-

lished observations), and RNA-seq studies carried out in

mouse striatal neurons indicated that decreasing Supt4h

expression to a level shown by our experiments to reduce

the abundance of transcripts containing lengthy CAG repeats

affected overall gene transcription only minimally. Together

these observations argue that agents targeting Supt4h may

reduce the transcription of genes containing lengthy trinucleo-

tide repeats while having limited effects on normal mammalian

cell function.

±±

±

± ±

A B

C

D

Figure 5. Supt4h Regulates Expression, Aggregation, and Toxicity of 81Q-eGFP in Striatal Neural Cells(A) Left, murine ST14A striatal cells cotransfected with the 7Q-eGFP or 81Q-eGFP plasmid construct and siRNA directed against Supt4h (Supt4h si) or a negative

control siRNA (NC si) were monitored by fluorescence microscopy. Phase contrast images are shown at right. Lower panel, percentage of cells showing

81Q-eGFP foci under these conditions. The values shown in this panel and panel (D) are the mean ± SD for three independent experiments.

(B) Changes in Supt4h expression upon siRNA knockdown were analyzed by Western blotting. Protein abundance in cells treated with control siRNA was set as

100%. To ensure comparable transfection efficiency among samples, pRC-CMV-MnSOD was cotransfected and its protein expression determined.

(C) Expression and aggregation of polyQ-eGFP were assessed by slot blot (NC) and filter-trap (CA) assay, respectively. Relative expression and aggregation

levels are shown after a-Tubulin normalization.

(D) Viability of ST14A cells expressing 7Q-eGFP or 81Q-eGFP was determined in Supt4h siRNA knockdown cells and controls. After transfection cells were

cultured in media containing 0.5% serum and maintained at 39�C to induce neuronal cell differentiation. The number of viable cells expressing 7Q-eGFP in the

presence of control siRNA was set as 1, and the relative cell viability of other samples is indicated (*, **p < 0.05 by Student t test).

Our results indicate that the transcriptional block that Spt4 can

surmount is affected by the length of the trinucleotide repeat as

well as by its sequence. The ability of Spt4 to promote transcrip-

tion through lengthy repeats of different types of trinucleotides

inserted into noncoding regions of genes suggests that inhibition

of Supt4h function may have additional utility in the treatment of

nonpolyglutamine trinucleotide repeat diseases. Our data indi-

cate that in cells deficient in Spt4, lengthy trinucleotide repeats

lead to reduction of template-bound RNA polymerase distal to

the extended repeat, decreased transcript abundance, and

reduced production and aggregation of long polyQ protein.

This model is diagramed in Figure 7.

Complexes consisting of yeast Spt4 or human Supt4h with

their respective partners Spt5 and Supt5h (Guo et al., 2008; Hart-

zog et al., 1996; Wada et al., 1998; Wenzel et al., 2010) bind to

the coiled domain of RNA polymerase II and encircle the DNA

template to promote processivity (Klein et al., 2011; Martinez-

Rucobo et al., 2011). We observed that the effects of spt4 dele-

tion on expression of Ade2 containing a long polyQ region were

recapitulated by a spt5 mutation that uniquely prevents interac-

tion between Spt5 and Spt4 (Figure S3), implying complex

formation between Spt4/Supt4h and Spt5/Spt5h is required for

efficient transcription through lengthy trinucleotide repeats. As

Spt4 is not essential for cell viability (Malone et al., 1993) and

does not directly contact RNA polymerase, as revealed by

crystal structure analyses (Klein et al., 2011; Martinez-Rucobo

et al., 2011), this transcription elongation factor seems likely to

stabilize RNA polymerase/template complexes by binding to

Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc. 697

±

±

± ±

±

± ± ±

± ± ±

± ± ±

A

B D

C

Figure 6. Mutant Htt Expression Is Selectively Inhibited by Supt4h Downregulation

(A) HttmRNA abundance was examined by RT-PCR following Supt4h siRNA knockdown in striatal cells possessing homozygous wild-type (HdhQ7/Q7) or mutant

huntingtin alleles (HdhQ111/Q111). Detection of each indicated transcript was applied using specific primer set (Table S2) and PCR products were analyzed by

agarose gel electrophoresis. U6, which is transcribed by RNA polymerase III, served as a control to assess the amounts of RNA applied in the RT-PCR analysis.

Tuba1a was included to determine the effect of Supt4h on pol II-dependent transcription of housekeeping genes.

(B) Changes inHttmRNA abundance associated with Supt4h siRNA knockdownwere assessed by real-time qRT-PCR. EachmRNAwas normalized withU6 and

the transcript abundance in wild-type (HdhQ7/Q7) cells transfected with control siRNA was set as 1. Error bars represent the SD.

(C) Analysis of protein expression in cells transfected with Supt4h siRNA and in controls. Equal amounts of cell extracts were immunoblotted with antibodies that

detect Htt, Supt4h, Tbp, or a-Tubulin. Tbp, the TATA box binding protein, was included as a representative gene that contains a short CAG repeat. The protein

abundance in HdhQ7/Q7 cells treated with control siRNA was set as 1, and relative protein levels in other samples are indicated.

(D) Heterozygous striatal cells containing the indicated Htt alleles (HdhQ7/Q111) were transfected with Supt4h siRNA and protein expression was analyzed as

described in (C). The positions of HttQ7 and HttQ111 are indicated by triangles and circular dots respectively. Expression is shown relative to the HttQ7 in control.

See also Figure S4.

the template externally to the transcription bubble (Klein et al.,

2011; Martinez-Rucobo et al., 2011). Point mutations that

disable an Spt4 zinc-finger domain that aids interaction of the

RNA polymerase complex with DNA reduced the expression of

long polyQ protein to the same extent as deletion of the

entire SPT4 gene (data not shown), suggesting that the RNA

polymerase complex-stabilizing actions of Spt4 are important

for transcription of regions containing repeated trinucleotides.

In vitro, expanded CAG repeats cause RNA polymerase II to

pause and then dissociate from the template (Parsons et al.,

1998; Salinas-Rios et al., 2011), and such dissociation may

account for the reduced expression of variant Htt we observed

in neurons of knock-in mice as well as in fibroblasts from HD

patients (Figure 6 and Figure S4). CAG or CTG repeats in DNA

can form mismatched hairpin structures (Lenzmeier and Freu-

denreich, 2003), and Spt4/5-mediated transcription processivity

through CAG or CTG repeats may overcome such structural

barriers. Alternatively, our finding that deletion of SPT4 affects

698 Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc.

even transcripts that contain CAA repeats or homopolyA se-

quences suggests that a second type of DNA structure may

underlie the biological role of Spt4. Hybrids of DNA and RNA

form during the transcription process and such R loops are in-

creased during transcription of templates containing expanded

trinucleotide repeats (Grabczyk et al., 2007; Lin et al., 2010;

Reddy et al., 2011), or containing even an increased density of

a particular trinucleotide interrupted by disparate sequences

(Roy and Lieber, 2009). R-loop formation, which can hinder

transcription elongation (Belotserkovskii and Hanawalt, 2011),

is also prominent during transcription of homopolyA sequences

(Kiyama and Oishi, 1996).

Recently, RNAi has been used for allele-specific reduction of

mutant Htt abundance and aggregation (Hu et al., 2009; Pfister

et al., 2009; van Bilsen et al., 2008; Zhang et al., 2009). However,

there is also evidence that siRNA targeting of transcripts contain-

ing expanded trinucleotide repeats can enhance the toxicity of

these transcripts (Yu et al., 2011). Reduction of mutant Htt

Figure 7. Model Indicating the Effects of Spt4/Supt4h Downregulation on Expression of CAG-Containing Genes

Movement of RNA polymerase II along a DNA template containing only a short CAG repeat (indicated in red) does not require Spt4. However, expanded CAG

repeats form an impediment that requires the presence of Spt4 for efficient transcription elongation. Aggregation of proteins containing an expanded polyQ

repeat (squares) is concentration-dependent, and such aggregation renders long-polyQ protein dysfunctional. Reducing expression of these proteins decreases

aggregation and restores biochemical functionality.

protein by an engineered fusion protein that binds to polyglut-

amine has also been reported (Bauer et al., 2010). Whereas the

blood-brain barrier may prevent therapeutic application of

such approaches (Lichota et al., 2010), agents that reduce

Supt4h expression or its interaction with Supt5h may circumvent

this limitation.

EXPERIMENTAL PROCEDURES

Yeast Strains

All strains were constructed by the one-step gene disruption method and

confirmed by genomic PCR as described previously (Cheng and Gartenberg,

2000). hsp104D and spt4D were derived from parental W303-1A cells by re-

placing respective HSP104 and SPT4 genes with a PCR-amplified KanMX

gene. Similarly, thp2D was generated by replacing THP2 gene with a HIS3

marker gene.

Biochemical Analyses

To prepare yeast lysates, 10 OD600 units of cells were collected, suspended in

ESB buffer (2% SDS, 80 mM Tris-HCl [pH 6.8], 10% glycerol, 1.5% DTT) sup-

plemented with 1 mM Na3VO4, 1 mM DTT, 1 mM PMSF and a protease inhib-

itor cocktail (Sigma), followed by grinding in glass beads. The supernatant was

collected as crude lysate after centrifugation (600 g). All steps were performed

at 4�C. Cultured neurons were lysed using RIPA buffer (50 mM Tris-HCl

[pH 7.5], 150 mM NaCl, 1% NP-40, 0.1% SDS and 1% sodium deoxycholate)

with supplements mentioned above.

Crude lysates were diluted 100-fold with TBS buffer (20mMTris HCl [pH7.5],

500 mMNaCl), and then loaded onto a 0.2 mm cellulose acetate membrane for

the filter-trap assay or 0.45 mm nitrocellulose membranes for a slot blot assay.

Following two TBST (TBS with 0.05% Tween 20) rinses, membranes were

removed from the Bio-Dot (Bio-Rad) apparatus and blocked with PBS-T

(1X PBS, 0.2% Triton X-100) containing 5% nonfat dry milk, probed with

primary antibodies for 1 hr, washed three times with PBS-T for 15 min, and

incubated with appropriate secondary antibodies conjugated to horseradish

peroxidase for another 1 hr. After three 15 min PBS-T washes, signals were

detected using Western Lighting (PerkinElmer Life Sciences).

Equal quantities (10 or 15 mg) of protein lysate separated on SDS-

polyacrylamide gels, followed by transfer to nitrocellulose membranes.

Immunoblotting was performed essentially as described above. To analyze

huntingtin variants, Tris-acetate polyacrylamide gels were used to obtain

greater resolution (Hu et al., 2009).

Antibodies

Antibodies against a-tubulin (DM1A, Sigma), FLAG-epitope (F4042, Sigma),

Hsp104 (ab2924, Abcam), GFP (ab6556, Abcam), Supt4h (ab54350, Abcam),

superoxide dismutase MnSOD (DD-17, Sigma), Huntingtin (MAB2166,

Chemicon), TATA-binding protein (58C6, Sigma), and monoclonal anti-RNA

polymerase II antibody (4H8-ChIP grade, Abcam) used for chromatin immuno-

precipitation were purchased.

Northern Blotting

Yeast total RNA was isolated using the hot acid phenol method (Cheng and

Gartenberg, 2000). 15 mg RNA from each sample was loaded and separated

on 1% agarose gels, followed by transfer to a Nylon membrane (Schleicher

& Schuell). The membrane was hybridized as per the manufacturer’s sug-

gestions (DIG Northern Starter Kit, Roche) using polyQ, ADE2, or SCR1

probe.

Chromatin Immunoprecipitation

Chromatin immunoprecipitation (ChIP) was performed as described (Cheng

and Gartenberg, 2000) with the following modifications. Cells were harvested

(50 OD600) after culturing in 2% galactose medium for 12 hr. Following fixation,

cells were treated with zymolyase in 1.2 M sorbital 1 hr to facilitate cell lysis.

Chromatin was collected, washed with 0.1X TE buffer, and fragmented using

micrococcal nuclease (New England BioLabs). Chromatin associating with

RNA polymerase II was precipitated by anti-RNA polymerase antibody bound

to protein A agarose beads (Upstate). Washing buffer 3 (10 mM Tris-HCl [pH

8.0], 0.5 M LiCl, 1% NP-40, 0.5% sodium deoxycholate, 1 mM EDTA, and

0.05% SDS) was modified to increase stringency. qPCR primer sets (ChIP

prom-50/-30 as probe A, ChIP dn500-50/-30 as probe B, and ChIP dn1000-

50/-30 as probe C) were used to amplify respective DNA domains. Oligonucle-

otide sequences are listed in Table S2.

Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc. 699

Cell Culture and Transfection

Striatal cell lines ST14A (rat), HdhQ7/Q7 (mouse), HdhQ111/Q111 (mouse), and

HdhQ7/Q111 (mouse) were cultured in DMEM (HyClone) supplemented with

10% fetal bovine serum at 33�C with 5% CO2. ST14A was transfected with

pTet-Off plasmid (BD Biosciences) to establish the stable cell line ST14Atet,

which expressed pTRE2-(CAG)n-eGFP in the absence of tetracycline. DNA

and siRNA transfections were carried out using LipofectAMINE 2000 (Invitro-

gen). 100 nM of Supt4h siRNA (DHARMACON, ON-TARGET plus SMART

pool, L-048866-01) and (DHARMACON, J-086342-10, 50-UGGCCUACAA

AUCGAGAGAUU-30 and 50-UCUCUCGAUUUGUAGGCCAUU-30) were used

to inhibit expression of Supt4h in mice and rat cells, respectively. Transfection

of an equivalent amount of annealed double-stranded oligonucleotides

(50-UUCUCCGAACGUGUCACGUTT-30 and 50-ACGUGACACGUUCGGAGA

ATT-30) that do not target any gene served as a control.

Microscopy

Cells were visualized using the Axioskop 2 (Carl Zeiss) fluorescence micro-

scope with a 100X oil immersion objective. ST14Atet cells were seeded on

coverslips and cotransfected with siRNA and poly-eGFP plasmid constructs.

Imaging was performed at 24 hr.

RT-PCR and Quantitative RT-PCR

Total RNA (5 mg) wasmixedwith 5 mMoligo dT, 5 mMSnRNAU6 rt-PCR primer,

and 500 mM dNTPs. The mixture was incubated at 65�C for 5 min and then

chilled on ice. After addition of First-Strand Buffer, DTT (Cf = 10 mM) and

1 ml reverse transcriptase (Invitrogen), the reaction was carried out at 42�Cfor 1 hr. Equivalent volumes of cDNA products were amplified by PCR and

products were resolved on 2.5% agarose gels to determine relative transcript

levels with SnRNA U6 serving as control. Quantitative real-time PCR was per-

formed using a StepOnePlus Real Time PCR system (Applied Biosystems) and

analyzed using the Quantitative- Comparative CT (DDCT) program.

Procedures for preparing constructs containing tri-nucleotide repeats, for

genome-wide yeast cell screening, for RNA-seq analyses, and for others are

detailed in the Extended Experimental Procedures.

SUPPLEMENTAL INFORMATION

Supplemental Information includes Extended Experimental Procedures, two

tables, and four figures and can be found with this article online at doi:10.

1016/j.cell.2011.12.032.

ACKNOWLEDGMENTS

We thank Chih-Yen King, Fang-Jen Lee, Jing-Jer Lin, Michael Sherman,

Michael Snyder, Elena Cattaneo, Marcy MacDonald, and the HD Community

BioRepository (http://ccr.coriell.org) sponsored by CHDI Foundation for

providing experimental materials. We thank Cheng-Fu Kao for technical assis-

tance in the ChIP assay and Yong Cheng, Zhihai Ma, Mark McCreary, and

Richard Lin for assistance in compilation and analysis of RNA-seq data.

T.-H.C. received grants from National Science Council (NSC- 96-2311-

B-010-001 and NSC-98-2311-B-010-001-MY3), and the ‘‘Aim for the Top

University Plan’’ grant from the Ministry of Education of the Republic of China.

S.N. Cohen and N. Deng received support from Kwoh-Ting Li Professorship

funds endowed by the Friends of Stanford Foundation of Taiwan.

Received: June 2, 2011

Revised: October 3, 2011

Accepted: December 13, 2011

Published: February 16, 2012

REFERENCES

Bauer, P.O., Goswami, A., Wong, H.K., Okuno, M., Kurosawa, M., Yamada,

M., Miyazaki, H., Matsumoto, G., Kino, Y., Nagai, Y., et al. (2010). Harnessing

chaperone-mediated autophagy for the selective degradation of mutant hun-

tingtin protein. Nat. Biotechnol. 28, 256–263.

700 Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc.

Belotserkovskii, B.P., and Hanawalt, P.C. (2011). Anchoring nascent RNA to

the DNA template could interfere with transcription. Biophys. J. 100, 675–684.

Biggar, S.R., and Crabtree, G.R. (2001). Cell signaling can direct either binary

or graded transcriptional responses. EMBO J. 20, 3167–3176.

Busch, A., Engemann, S., Lurz, R., Okazawa, H., Lehrach, H., and Wanker,

E.E. (2003). Mutant huntingtin promotes the fibrillogenesis of wild-type hun-

tingtin: a potential mechanism for loss of huntingtin function in Huntington’s

disease. J. Biol. Chem. 278, 41452–41461.

Cheng, T.H., andGartenberg,M.R. (2000). Yeast heterochromatin is a dynamic

structure that requires silencers continuously. Genes Dev. 14, 452–463.

Cloonan, N., and Grimmond, S.M. (2008). Transcriptome content and

dynamics at single-nucleotide resolution. Genome Biol. 9, 234.

Davies, S.W., Turmaine, M., Cozens, B.A., DiFiglia, M., Sharp, A.H., Ross,

C.A., Scherzinger, E., Wanker, E.E., Mangiarini, L., and Bates, G.P. (1997).

Formation of neuronal intranuclear inclusions underlies the neurological

dysfunction in mice transgenic for the HD mutation. Cell 90, 537–548.

DiFiglia, M., Sapp, E., Chase, K.O., Davies, S.W., Bates, G.P., Vonsattel, J.P.,

and Aronin, N. (1997). Aggregation of huntingtin in neuronal intranuclear

inclusions and dystrophic neurites in brain. Science 277, 1990–1993.

Dragatsis, I., Levine, M.S., and Zeitlin, S. (2000). Inactivation of Hdh in the brain

and testis results in progressive neurodegeneration and sterility in mice. Nat.

Genet. 26, 300–306.

Duennwald, M.L., Jagadish, S., Giorgini, F., Muchowski, P.J., and Lindquist, S.

(2006). A network of protein interactions determines polyglutamine toxicity.

Proc. Natl. Acad. Sci. USA 103, 11051–11056.

Ehrlich, M.E., Conti, L., Toselli, M., Taglietti, L., Fiorillo, E., Taglietti, V., Ivkovic,

S., Guinea, B., Tranberg, A., Sipione, S., et al. (2001). ST14A cells have prop-

erties of a medium-size spiny neuron. Exp. Neurol. 167, 215–226.

Grabczyk, E., Mancuso, M., and Sammarco, M.C. (2007). A persistent

RNA.DNA hybrid formed by transcription of the Friedreich ataxia triplet repeat

in live bacteria, and by T7 RNAP in vitro. Nucleic Acids Res. 35, 5351–5359.

Group, H.s.D.C.R.. (1993). A novel gene containing a trinucleotide repeat that

is expanded and unstable on Huntington’s disease chromosomes. The

Huntington’s Disease Collaborative Research Group. Cell 72, 971–983.

Guo, M., Xu, F., Yamada, J., Egelhofer, T., Gao, Y., Hartzog, G.A., Teng, M.,

and Niu, L. (2008). Core structure of the yeast spt4-spt5 complex: a conserved

module for regulation of transcription elongation. Structure 16, 1649–1658.

Gutekunst, C.A., Li, S.H., Yi, H., Mulroy, J.S., Kuemmerle, S., Jones, R., Rye,

D., Ferrante, R.J., Hersch, S.M., and Li, X.J. (1999). Nuclear and neuropil

aggregates in Huntington’s disease: relationship to neuropathology. J. Neuro-

sci. 19, 2522–2534.

Hands, S., Sinadinos, C., andWyttenbach, A. (2008). Polyglutamine gene func-

tionanddysfunction in theageingbrain.Biochim.Biophys.Acta1779, 507–521.

Hartzog, G.A., Basrai, M.A., Ricupero-Hovasse, S.L., Hieter, P., and Winston,

F. (1996). Identification and analysis of a functional human homolog of the

SPT4 gene of Saccharomyces cerevisiae. Mol. Cell. Biol. 16, 2848–2856.

Hirtreiter, A., Damsma, G.E., Cheung, A.C., Klose, D., Grohmann, D., Vojnic,

E., Martin, A.C., Cramer, P., andWerner, F. (2010). Spt4/5 stimulates transcrip-

tion elongation through the RNA polymerase clamp coiled-coil motif. Nucleic

Acids Res. 38, 4040–4051.

Holmes, S.E., O’Hearn, E.E., McInnis, M.G., Gorelick-Feldman, D.A., Kleider-

lein, J.J., Callahan, C., Kwak, N.G., Ingersoll-Ashworth, R.G., Sherr, M.,

Sumner, A.J., et al. (1999). Expansion of a novel CAG trinucleotide repeat in

the 50 region of PPP2R2B is associated with SCA12. Nat. Genet. 23, 391–392.

Hu, J., Matsui, M., Gagnon, K.T., Schwartz, J.C., Gabillet, S., Arar, K., Wu, J.,

Bezprozvanny, I., and Corey, D.R. (2009). Allele-specific silencing of mutant

huntingtin and ataxin-3 genes by targeting expanded CAG repeats in mRNAs.

Nat. Biotechnol. 27, 478–484.

Jones, E.W., and Fink, G.R. (1982). Regulation of amino acid and nucleotide

biosynthesis in yeast. In The Molecular Biology of the Yeast Saccharomyces:

Metabolism and Gene Expression, E.W. Jones and J.R. Broach, eds. (Cold

Spring Harbor, NY: Cold Spring Harbor Laboratory Press), pp. 181–299.

Kazantsev, A., Preisinger, E., Dranovsky, A., Goldgaber, D., and Housman, D.

(1999). Insoluble detergent-resistant aggregates form between pathological

and nonpathological lengths of polyglutamine in mammalian cells. Proc.

Natl. Acad. Sci. USA 96, 11404–11409.

Kiyama, R., and Oishi, M. (1996). In vitro transcription of a poly(dA) x poly(dT)-

containing sequence is inhibited by interaction between the template and its

transcripts. Nucleic Acids Res. 24, 4577–4583.

Klein, B.J., Bose, D., Baker, K.J., Yusoff, Z.M., Zhang, X., and Murakami, K.S.

(2011). RNA polymerase and transcription elongation factor Spt4/5 complex

structure. Proc. Natl. Acad. Sci. USA 108, 546–550.

Krobitsch, S., and Lindquist, S. (2000). Aggregation of huntingtin in yeast

varies with the length of the polyglutamine expansion and the expression of

chaperone proteins. Proc. Natl. Acad. Sci. USA 97, 1589–1594.

Lenzmeier, B.A., and Freudenreich, C.H. (2003). Trinucleotide repeat insta-

bility: a hairpin curve at the crossroads of replication, recombination, and

repair. Cytogenet. Genome Res. 100, 7–24.

Li, H., Li, S.H., Johnston, H., Shelbourne, P.F., and Li, X.J. (2000). Amino-

terminal fragments of mutant huntingtin show selective accumulation in striatal

neurons and synaptic toxicity. Nat. Genet. 25, 385–389.

Lichota, J., Skjorringe, T., Thomsen, L.B., andMoos, T. (2010).Macromolecular

drug transport into the brain using targeted therapy. J. Neurochem. 113, 1–13.

Lin, Y., Dent, S.Y., Wilson, J.H., Wells, R.D., and Napierala, M. (2010). R loops

stimulate genetic instability of CTG.CAG repeats. Proc. Natl. Acad. Sci. USA

107, 692–697.

Lindstrom, D.L., Squazzo, S.L., Muster, N., Burckin, T.A., Wachter, K.C.,

Emigh, C.A., McCleery, J.A., Yates, J.R., III, and Hartzog, G.A. (2003). Dual

Roles for Spt5 in Pre-mRNAProcessing andTranscriptionElongationRevealed

by Identification of Spt5-Associated Proteins. Mol. Cell. Biol. 23, 1368–1378.

Malone, E.A., Fassler, J.S., and Winston, F. (1993). Molecular and genetic

characterization of SPT4, a gene important for transcription initiation in

Saccharomyces cerevisiae. Mol. Gen. Genet. 237, 449–459.

Mangiarini, L., Sathasivam, K., Seller, M., Cozens, B., Harper, A., Hethering-

ton, C., Lawton, M., Trottier, Y., Lehrach, H., Davies, S.W., et al. (1996).

Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause

a progressive neurological phenotype in transgenic mice. Cell 87, 493–506.

Martinez-Rucobo, F.W., Sainsbury, S., Cheung, A.C., and Cramer, P. (2011).

Architecture of the RNA polymerase-Spt4/5 complex and basis of universal

transcription processivity. EMBO J. 30, 1302–1310.

Mason, P.B., and Struhl, K. (2005). Distinction and relationship between elon-

gation rate andprocessivity ofRNApolymerase II in vivo.Mol. Cell 17, 831–840.

Meriin, A.B., Zhang, X., He, X., Newnam, G.P., Chernoff, Y.O., and Sherman,

M.Y. (2002). Huntington toxicity in yeast model depends on polyglutamine

aggregationmediated by a prion-like protein Rnq1. J. Cell Biol. 157, 997–1004.

Nagalakshmi, U., Waern, K., and Snyder, M. (2010). RNA-Seq: a method for

comprehensive transcriptome analysis. Curr. Protoc. Mol. Biol., Chapter 4,

Unit 4 11, 11–13.

Nasir, J., Floresco, S.B., O’Kusky, J.R., Diewert, V.M., Richman, J.M., Zeisler,

J., Borowski, A., Marth, J.D., Phillips, A.G., and Hayden, M.R. (1995). Targeted

disruption of the Huntington’s disease gene results in embryonic lethality and

behavioral and morphological changes in heterozygotes. Cell 81, 811–823.

Ordway, J.M., Tallaksen-Greene, S., Gutekunst, C.A., Bernstein, E.M., Cearley,

J.A., Wiener, H.W., Dure, L.S.t., Lindsey, R., Hersch, S.M., Jope, R.S., et al.

(1997). Ectopically expressed CAG repeats cause intranuclear inclusions and

aprogressive lateonset neurological phenotype in themouse.Cell91, 753–763.

Oesterreich, F.C., Bieberstein, N., and Neugebauer, K.M. (2011). Pause

locally, splice globally. Trends Cell Biol. 21, 328–335.

Parsons, M.A., Sinden, R.R., and Izban, M.G. (1998). Transcriptional proper-

ties of RNA polymerase II within triplet repeat-containing DNA from the human

myotonic dystrophy and fragile X loci. J. Biol. Chem. 273, 26998–27008.

Pfister, E.L., Kennington, L., Straubhaar, J., Wagh, S., Liu, W., DiFiglia, M.,

Landwehrmeyer, B., Vonsattel, J.P., Zamore, P.D., and Aronin, N. (2009).

Five siRNAs targeting three SNPs may provide therapy for three-quarters of

Huntington’s disease patients. Curr. Biol. 19, 774–778.

Reddy, K., Tam, M., Bowater, R.P., Barber, M., Tomlinson, M., Nichol

Edamura, K., Wang, Y.H., and Pearson, C.E. (2011). Determinants of R-loop

formation at convergent bidirectionally transcribed trinucleotide repeats.

Nucleic Acids Res. 39, 1749–1762.

Rondon, A.G., Garcia-Rubio,M., Gonzalez-Barrera, S., and Aguilera, A. (2003).

Molecular evidence for a positive role of Spt4 in transcription elongation.

EMBO J. 22, 612–620.

Ross, C.A. (2002). Polyglutamine pathogenesis: emergence of unifying mech-

anisms for Huntington’s disease and related disorders. Neuron 35, 819–822.

Roy, D., and Lieber, M.R. (2009). G clustering is important for the initiation of

transcription-induced R-loops in vitro, whereas high G density without clus-

tering is sufficient thereafter. Mol. Cell. Biol. 29, 3124–3133.

Salinas-Rios, V., Belotserkovskii, B.P., and Hanawalt, P.C. (2011). DNA slip-

outs cause RNA polymerase II arrest in vitro: potential implications for genetic

instability. Nucleic Acids Res. 39, 7444–7454.

Scherzinger, E., Sittler, A., Schweiger, K., Heiser, V., Lurz, R., Hasenbank, R.,

Bates, G.P., Lehrach, H., and Wanker, E.E. (1999). Self-assembly of polyglut-

amine-containing huntingtin fragments into amyloid-like fibrils: implications for

Huntington’s disease pathology. Proc. Natl. Acad. Sci. USA 96, 4604–4609.

Trettel, F., Rigamonti, D., Hilditch-Maguire, P., Wheeler, V.C., Sharp, A.H., Per-

sichetti, F., Cattaneo, E., and MacDonald, M.E. (2000). Dominant phenotypes

produced by the HD mutation in STHdh(Q111) striatal cells. Hum. Mol. Genet.

9, 2799–2809.

Tydemers, J., Mogk, A., andBukau, B. (2010). Cellular strategies for controlling

protein aggregation. Nat. Rev. Mol. Cell Biol. 11, 777–788.

van Bilsen, P.H., Jaspers, L., Lombardi, M.S., Odekerken, J.C., Burright, E.N.,

and Kaemmerer, W.F. (2008). Identification and allele-specific silencing of the

mutant huntingtin allele in Huntington’s disease patient-derived fibroblasts.

Hum. Gene Ther. 19, 710–719.

Van Raamsdonk, J.M., Pearson, J., Rogers, D.A., Bissada, N., Vogl, A.W.,

Hayden, M.R., and Leavitt, B.R. (2005). Loss of wild-type huntingtin influences

motor dysfunction and survival in the YAC128 mouse model of Huntington

disease. Hum. Mol. Genet. 14, 1379–1392.

Varma, H., Voisine, C., DeMarco, C.T., Cattaneo, E., Lo, D.C., Hart, A.C., and

Stockwell, B.R. (2007). Selective inhibitors of death in mutant huntingtin cells.

Nat. Chem. Biol. 3, 99–100.

Voynov, V., Verstrepen, K.J., Jansen, A., Runner, V.M., Buratowski, S., and

Fink, G.R. (2006). Genes with internal repeats require the THO complex for

transcription. Proc. Natl. Acad. Sci. USA 103, 14423–14428.

Wada, T., Takagi, T., Yamaguchi, Y., Ferdous, A., Imai, T., Hirose, S., Sugi-

moto, S., Yano, K., Hartzog, G.A., Winston, F., et al. (1998). DSIF, a novel tran-

scription elongation factor that regulates RNA polymerase II processivity, is

composed of human Spt4 and Spt5 homologs. Genes Dev. 12, 343–356.

Wenzel, S., Martins, B.M., Rosch, P., andWohrl, B.M. (2010). Crystal structure

of the human transcription elongation factor DSIF hSpt4 subunit in complex

with the hSpt5 dimerization interface. Biochem. J. 425, 373–380.

Winston, F., Chaleff, D.T., Valent, B., and Fink, G.R. (1984). Mutations affecting

Ty-mediated expression of the HIS4 gene of Saccharomyces cerevisiae.

Genetics 107, 179–197.

Wojciechowska, M., and Krzyzosiak, W.J. (2011). Cellular toxicity of expanded

RNA repeats: focus on RNA foci. Hum. Mol. Genet. 20, 3811–3821.

Xiao, Y., Yang, Y.H., Burckin, T.A., Shiue, L., Hartzog, G.A., and Segal, M.R.

(2005). Analysis of a Splice Array Experiment Elucidates Roles of Chromatin

Elongation Factor Spt4-5 in Splicing. PLoS Comput. Biol. 1, e39.

Yu, Z., Teng, X., and Bonini, N.M. (2011). Triplet repeat-derived siRNAs

enhance RNA-mediated toxicity in a Drosophila model for myotonic

dystrophy. PLoS Genet. 7, e1001340.

Zhang, Y., Engelman, J., and Friedlander, R.M. (2009). Allele-specific silencing

of mutant Huntington’s disease gene. J. Neurochem. 108, 82–90.

Zoghbi, H.Y., and Orr, H.T. (2000). Glutamine repeats and neurodegeneration.

Annu. Rev. Neurosci. 23, 217–247.

Cell 148, 690–701, February 17, 2012 ª2012 Elsevier Inc. 701