Interaction between the RNA-dependent ATPase and poly(A) polymerase subunits of the TRAMP complex is mediated by short peptides and important for snoRNA processing
Interaction between the RNA-dependent ATPase and poly(A) polymerase subunits of the TRAMP complex is mediated by short peptides and important for snoRNA processing
Jillian S. Losh 1 2
Alejandra Klauer King 1 2
Jeremy Bakelar 0
Lacy Taylor 0
John Loomis 2
Jason A. Rosenzweig 3
Sean J. Johnson 0
Ambro van Hoof 1 2
0 Department of Chemistry and Biochemistry, Utah State University , Logan, UT 84322-0300 , USA
1 The University of Texas Graduate School of Biomedical Sciences at Houston , Houston, TX , USA
2 Department of Microbiology and Molecular Genetics, University of Texas Health Science Center-Houston , Houston, TX 77030 , USA
3 Department of Biology and Department of Environmental and Interdisciplinary Sciences, Texas Southern University , Houston, TX 77004 , USA
The RNA exosome is one of the main 3 to 5 exoribonucleases in eukaryotic cells. Although it is responsible for degradation or processing of a wide variety of substrate RNAs, it is very specific and distinguishes between substrate and non-substrate RNAs as well as between substrates that need to be 3 processed and those that need to be completely degraded. This specificity does not appear to be determined by the exosome itself but rather by about a dozen other proteins. Four of these exosome cofactors have enzymatic activity, namely, the nuclear RNA-dependent ATPase Mtr4, its cytoplasmic paralog Ski2 and the nuclear non-canonical poly(A) polymerases, Trf4 and Trf5. Mtr4 and either Trf4 or Trf5 assemble into a TRAMP complex. However, how these enzymes assemble into a TRAMP complex and the functional consequences of TRAMP complex assembly remain unknown. Here, we identify an important interaction site between Mtr4 and Trf5, and show that disrupting the Mtr4/Trf interaction disrupts specific TRAMP and exosome functions, including snoRNA processing.
Almost all cellular RNAs undergo extensive processing
before becoming fully mature and functional RNAs. The
pathways involved in these post-transcriptional
modifications, although tightly regulated, are not error-free.
Therefore, cells have evolved a variety of surveillance mechanisms
to ensure the fidelity of this RNA maturation process and
gene expression. The RNA exosome complex is a major
player in both RNA processing and surveillance
mechanisms. For example, the RNA exosome has 3 to 5
exoribonuclease activity that is required for removal of 3
extensions from precursor RNAs (e.g. 5.8S rRNA; 1,2) and to
degrade misprocessed RNAs (e.g. hypomethylated tRNAi;
3). Many exosome substrates have been identified, but how
the exosome distinguishes between substrates that require
processing, substrates that require degradation and
nonsubstrates is poorly understood.
One determinant of exosome specificity appears to be
provided by over a dozen proteins that are collectively
called exosome cofactors. These proteins are thought to
initiate interaction with substrates and deliver them to the
exosome by poorly defined mechanisms. Most prominent
among these cofactors are two RNA-dependent ATPases
(also called RNA helicases), Mtr4 and Ski2 (4,5). Unlike
other exosome cofactors that appear to be specific for a
limited number of exosome functions, Mtr4 is required for all
known nuclear exosome functions, while Ski2 is required for
all known cytoplasmic exosome functions.
Mtr4 is a subunit of the heterotrimeric Trf4/5 Air1/2
Mtr4 polyadenylation (TRAMP) complex. In the yeast
Saccharomyces cerevisiae, the other two subunits are one
noncanonical poly(A) polymerase (either Trf4 or Trf5), and one
zinc knuckle RNA binding protein (either Air1 or Air2;
6,7). The duplicated TRF and AIR genes arose during a
whole genome duplication in an ancestor of S. cerevisiae
(http://ygob.ucd.ie/; 8,9) and, therefore, other eukaryotes
have only one ortholog of each, which in humans are named
*To whom correspondence should be addressed. Tel: +1 713 500 5234; Email:
These authors contributed equally to the paper as first authors.
PAPD5 and ZCCHC7, respectively. Furthermore, it has
been shown that the hMTR4, PAPD5 and ZCCHC7
proteins interact (10), suggesting that TRAMP complex
formation is conserved between fungi and animals. In yeast,
Mtr4 is encoded by an essential gene, but TRF4, TRF5,
AIR1 and AIR2 are not individually essential (4,11).
However, a trf4 trf5 double mutant is inviable (12), and an
air1 air2 double mutant is extremely slow growing (13).
These growth phenotypes suggest that all three subunits of
the TRAMP complex perform critical functions, although
the importance of assembly into a TRAMP complex is not
Importantly, Mtr4 is more abundant than the other
TRAMP subunits, indicating that it exists outside TRAMP,
and some functions of the exosome require Mtr4 but not
the other TRAMP subunits (6). The TRAMP complex
is thought to be required for degrading incorrectly
processed RNAs, including rRNAs and tRNAs, as well as
non-coding RNAs with unknown functions (cryptic
unstable transcripts and stable unannotated transcripts; 14,15).
An example of a reaction that requires Mtr4, but not
other TRAMP subunits is the processing of 5.8S rRNA
from a 3 extended precursor (6). One hypothesis is that
TRAMP is involved in exosome-mediated RNA
degradation, while TRAMP-independent Mtr4 is involved in
exosome-mediated RNA processing. Definitive proof for
this hypothesis would require showing whether or not 3
extended and/or polyadenylated species that accumulate in
TRAMP mutants are precursors in the normal processing
pathway, or aberrant transcripts marked for degradation.
Adding nucleotides to the 3 end of an RNA to
facilitate removal of nucleotides in the 3 to 5 direction is
somewhat counterintuitive, but several roles of
TRAMPmediated polyadenylation in exosome-mediated
degradation have been offered. First, the main catalytic subunit of
the RNA exosome (Rrp44) is accessed through a long
narrow central channel formed by the other nine subunits of
the exosome (16,17). Therefore, exosome-mediated
degradation is thought to require a long unstructured region,
which may be provided by TRAMP-mediated
polyadenylation. Under this hypothesis, the unstructured tail would
have to be long enough to traverse the channel (30 nts).
Such long TRAMP-dependent tails have been detected in
exosome mutants and TRAMP can synthesize long A-tails
in vitro (6,7,18,19). However, a typical TRAMP-synthesized
tail in wild-type cells is thought to be only 34 nts (20,21)
and, thus, not long enough to completely pass through the
exosome central channel. One caveat is that the short
Atails seen in wild-type cells are necessarily the ones that have
not (yet) been degraded by the exosome. RNAs that receive
longer A-tails may be more rapidly degraded by the
exosome. Therefore, the short A-tails seen at steady state may
not be representative of all the tails synthesized by TRAMP.
Another indication that long, unstructured 3 tails are not
required for exosome-mediated degradation comes from the
observation that the exosome appears to be fully capable
of degrading cytoplasmic substrate RNAs independently
of a poly(A) polymerase, including substrates that
contain very stable secondary structures (e.g. G-quadruplexes;
22,23). An alternative hypothesis is based on the
observation that Mtr4 specifically interacts with oligo-adenylated
RNAs (2426), and thus TRAMP-synthesized tails may
target substrates to Mtr4 rather than directly to the RNA
exosome. Under both of these hypotheses, substrate RNAs
would initially interact with the poly(A) polymerase
subunit of TRAMP, and subsequently be handed off to the
exosome. Strikingly, while these hypotheses explain the role
of the poly(A) tail, neither of these hypotheses readily
explain why Mtr4 and a poly(A) polymerase assemble into a
TRAMP complex. A third possibility is that
polyadenylation by Trf4/5 occurs in response to a block or stall
during normal processing by the exosome. Physical interaction
between the Mtr4/exosome machinery and Trf4/5 may
facilitate polyadenylation of such degradation product and
polyadenylation may enhance subsequent re-engagement
of the Mtr4/exosome machinery. Consistent with the idea
that TRAMP can act on partially degraded RNAs is that
TRAMP-dependent tails are added at multiple positions,
including the region corresponding to the mature RNA
(27). Thus, it is not clear whether stable association of the
poly(A) polymerase with Mtr4 facilitates substrate
handoff from the Mtr4/exosome machinery to the poly(A)
polymerase or vice versa.
Resolving these questions on TRAMP and Mtr4
function requires a better understanding of how the subunits
of TRAMP interact and of the consequences of
impairing TRAMP complex assembly. Here, we identify a short
peptide on the poly(A) polymerase subunit that is
important for TRAMP complex assembly. This short peptide is
essentially the only conserved sequence in the N-terminus
of the poly(A) polymerase subunit. Furthermore, this
Nterminus is predicted to be largely disordered. We also show
that deletion of this peptide impairs assembly of TRAMP
and has a specific effect on the accumulation of 3 extended
snoRNAs. We anticipate that the strain we generated will
be an important tool in future characterization of TRAMP
MATERIALS AND METHODS
The yeast two-hybrid assays used plasmids pOBD2 and
pOAD that were previously described (28). The yeast
two-hybrid plasmids for Mtr4 (pAV745), Mtr4-archless
(pAV746) and Trf553199 (pAV744) have been
previously described as have the positive controls that contain
MEC3 and RAD17 (29). Yeast two-hybrid plasmids with
Trf5 residues 68199 (pAV759), 83199 (pAV760), 98199
(pAV761), 118199 (pAV766), 53184 (pAV762), 53169
(pAV763), 53154 (pAV764), 53139 (pAV767) and 53124
(pAV768) were generated similarly to pAV744. Briefly, these
regions of TRF5 were PCR amplified with primers that
contain PvuII and PstI sites, digested with those restriction
enzymes and cloned into pOAD. The yeast two-hybrid
plasmids with Trf5 residues 98117 were generated by cloning
oligonucleotides for this region into the PvuII and PstI sites
The Trf5- 98117-TAP and Trf5-TAP plasmids are low
copy (CEN) plasmids derived from pRS415 (30) with the
endogenous TRF5 promoter driving expression of Trf5 or
Trf5-98117. The trf5-98117-TAP plasmid (pAV854)
was generated by ligation of three DNA fragments. The
first fragment contained the promoter and 5 end of
TRF5 and was generated by PCR with oligonucleotides
(TATTATGCGGCCGCCCACAAAGTACTACATCTATGGTCT) and oAV870
digestion with NotI and SpeI. The second fragment
contained the 3 end of TRF5 and was generated by PCR
with oligonucleotides oAV871
(GCGGCGACTAGTGAACAAATAAAGGAAGATGATGATG) and oAV869
(GCGGCGCTGCAGCAAGAGCCTGGCCTTTAGAGAGCC) and digestion with SpeI and PstI. The third
fragment was pAV476 (31) digested with NotI and PstI.
The TRF5-TAP plasmid (pAV885) was subsequently
generated by replacing an XbaI to BstAPI fragment that
contained the internal deletion with the corresponding
Cloning of full-length Mtr4WT (pSJJ004) and Mtr4archless
(pSJJ009) expression constructs was performed previously
(32). Additional truncated variants used in this study
include Mtr4 74 (pSJJ012, comprising residues 751073),
Mtr41614 (pSJJ014) and Mtr4665815 (pSJJ020). Each of the
Mtr4 constructs was cloned into a pET151/D-TOPO vector
with a TEV protease-cleavable 6-His tag on the N-terminus.
Strain trf4/trf5/pRS416-TRF4 has been previously
described (MATa, leu2-0, ura3-0, his3-1, met15-0,
trf4::NAT, trf5::KAN; 33). The TRF5 and trf5-98
117 plasmids were introduced by transformation and
selection on media lacking leucine. Transformants were then
plated on media containing 5FOA to select for cells that
had lost the pRS416-TRF4 plasmid. The rrp44-exo,
rrp44endo and rrp6::KAN strains are isogenic to this strain
and have been previously described (31). Yeast two-hybrid
strains PJ694 and PJ694a have been previously
described (28,34). They are MAT and MATa, respectively,
and in addition are both trp1901, leu23,112, ura3
52, his3200, gal4, gal80, LYS2::GAL1-HIS3,
Yeast two-hybrid assays
Yeast two-hybrid assays were carried out as previously
described (29). Briefly, the MTR4 TRP1 plasmids were
transformed into PJ694 and the TRF5 LEU2 plasmids were
transformed into PJ694a. Transformants were crossed to
each other and diploids that contain both plasmids were
selected on SC-TRP-LEU. These diploids were then serially
diluted and spotted on SC-TRP-LEU, SC-ADE and
The purification of TAP-tagged proteins was done as
previously described (22). Briefly, yeast cells were lysed by
vortexing in the presence of glass beads and IP50 buffer (50
mM TRIS HCl pH7.5, 50 mM KCl, 2 mM MgCl2, 0.1%
triton X100 and cOmplete EDTA-free protease inhibitors;
Roche). The lysate was clarified by centrifugation and
incubated with IgG Sepharose (GE Healthcare) for 2 h at
4C. The beads were pelleted, washed twice with IP50 and
twice with IP150 (same as IP50, except 150 mM KCl
instead of 50 mM), before being mixed with Laemmli loading
buffer. Western blot analysis used antibodies against
protein A (Sigma), Mtr4 (Gift from Patrick Linder) and Pgk1
RNA isolation and northern blotting was performed
essentially as previously described (32). Quantitative reverse
transcriptase-PCR (qRT-PCR) was carried out using the
SYBR Green RNA-to-Ct 1-Step kit from Applied
Biosystems as per the manufacturers instructions with primers
that were previously described (15). Transcriptome
sequencing was performed by LC Sciences (Houston, TX,
USA) on a HiSeq 2500. Briefly, poly(A)+ RNA was isolated
from duplicate cultures of the trf4, trf5 deletion strain
carrying either the TRF5 or trf5-98117 plasmid and
converted to a sequencing library. Each library was sequenced,
yielding between 10 and 14 million 50 nt reads that were
mapped to the annotated yeast genes (ftp://ftp.ensembl.org/
pub/release-77/fasta/saccharomyces cerevisiae/dna/) using
Bowtie. Genes that were significantly up- or down-regulated
in the trf5-98117 mutant were identified using edgeR
Recombinant protein expression and purification
The expression and purification of Mtr4WT and truncated
Mtr4 proteins was carried out as performed previously (32).
All proteins were recombinantly expressed in an Escherichia
coli BL21-codon+-(DE3)-RIL cell line (Stratagene). Cell
lysis was performed by lysozyme treatment and sonication of
frozen cell pellets. Nickel affinity, Heparin affinity and
Superdex 200 (GE) gel filtration was used to purify Mtr4 and
truncated Mtr4 variants. Final purification buffer consisted
of 50 mM HEPES (pH 7.5), 160 mM NaCl, 5% glycerol and
2 mM -mercapto-ethanol.
Binding analysis of Mtr4 to Trf5 and Air2 was carried out
using fluorescence anisotropy. A Trf5 peptide comprising
residues 98124 with an N-terminal fluorescein label was
purchased through the Keck lab at Yale University.
Binding reactions were buffered in 50 mM HEPES (pH 7.5), 140
mM NaCl and 12% glycerol. Concentration of the
fluorescently labeled peptide was held constant at 50 nM, with
increasing concentrations of Mtr4. Mtr4 was incubated with
the fluorescein-labeled Trf5 peptide for 30 min at 30C prior
to collection of fluorescence anisotropy data. Anisotropy
was measured using a Synergy H4 Hybrid Multi-Mode
Microplate Reader (BioTek) with an excitation at 485 20 nm
and an emission at 528 20 nm (the appropriate filters for
detection of fluorescein) at 30C.
For Air2 binding studies, an Air2 peptide comprising
residues 129 with an N-terminal fluorescein label was
obtained from Dr. Joshua Price at Brigham Young
University. Binding reactions were buffered in 50 mM HEPES (pH
7.5), 100 mM NaCl, 0.1 mg/ml bovine serum albumin and
12% glycerol. Concentration of the labeled Air2129
peptide was held constant at 80 nM and titrated with
increasing concentrations of Mtr4. Samples were incubated for
3 min after each titration, as changes in anisotropy were
not observed beyond this incubation time. Anisotropy at
each titration point was measured 10 times and averaged.
Anisotropy measurements were obtained using a
steadystate-photon counting spectrofluorometer, PC1 with Vinci
software, from ISS Instruments. Excitation and emission
slits were adjusted to 0.5 nm and temperature was
maintained at 25C. The excitation wavelength was 495 nm and
emission anisotropy was measured at 521 nm.
A conserved 20 amino acid peptide in the N-terminus of Trf5
is sufficient for Mtr4 interaction
The TRAMP complex was initially identified in a yeast
twohybrid screen for Mtr4 interacting proteins (6). All of the
Trf5 clones identified included amino acid residues 53199,
suggesting that this part of Trf5 contains the Mtr4
interacting site. Bioinformatic analysis did not reveal any
recognizable domains contained within this region, but PONDR
(Predictor of Naturally Disordered Regions; http://www.
pondr.com/index) suggested that this region was largely
intrinsically disordered. Multiple sequence alignment showed
that, similar to other intrinsically disordered regions, the
sequence was generally poorly conserved, except for residues
98117 of Trf5. As shown in Figure 1A and Supplementary
Figure S1, this 20 amino acid region is conserved in yeast
Trf4 and Trf5 proteins. Since small conserved motifs in
intrinsically disordered regions are often proteinprotein
interaction motifs (reviewed in 36), we hypothesized that this
may be a major Mtr4 interacting site.
To test whether amino acids 98117 are sufficient for a
yeast two-hybrid interaction with Mtr4, we generated
further truncations starting with the Trf5 53199 fragment
identified previously (6). Deleting 15, 30 or 45 amino acids
from the N-terminus of the 53199 region did not affect
yeast two-hybrid interaction with Mtr4 (Figure 1B),
suggesting the main interaction site was C-terminal of amino
acid 98. In contrast, deleting 60 amino acids from the
Nterminus abrogated the yeast two-hybrid interaction. In
similar C-terminal deletions, deleting 15, 30, 45, 60 or 75
amino acids from the 53199 region did not affect yeast
twohybrid interaction with Mtr4, suggesting the main
interaction site was N-terminal of amino acid 124 (Figure 1B).
Thus, these results are consistent with the conserved
peptide of amino acid residues 98117 forming a major Mtr4
To more directly test whether amino acids 98117 of Trf5
form an Mtr4-interaction site, we cloned just these residues
into a yeast two-hybrid vector. As shown in Figure 1C, this
peptide indeed interacted with Mtr4. We and others have
previously solved the crystal structure of Mtr4, which
revealed that it consists of a core helicase fold shared with
the Ski2-like family of RNA and DNA helicases and an
arch domain that is present only in Mtr4, Ski2 and their
orthologs (32,37). Deleting this arch domain did not disrupt
the interaction with other TRAMP subunits (32,37).
Similarly, the 98117 peptide of Mtr4 interacts both with
fulllength and archless Mtr4 (Figure 1C). In fact, in the yeast
two-hybrid analysis, the Mtr4-archless interaction with Trf5
98117 allowed for more robust growth, as we have
previously reported for the longer Trf5 53199 fragment. These
results identify an interaction of a small peptide in the
unstructured N-terminus of Trf4/5 with the helicase core of
Mtr4 that may be important for TRAMP complex
To test whether Trf5 residues 98117 directly bind to
purified recombinant Mtr4, we used fluorescence anisotropy.
Figure 2A shows that indeed, a fluorescently labeled Trf5
peptide directly bound to recombinant Mtr4 with a Kd
of 10.7 M. This binding was specific as no binding was
observed of this peptide to purified recombinant Rrp47
(another exosome cofactor; data not shown). We further
probed the Mtr4-Trf5 peptide interaction with truncated
forms of Mtr4. An N-terminal truncation (Mtr4 74;
Figure 2B), and an Mtr4 arch deletion (Mtr4archless; Figure 2C)
both had binding affinities for the Trf5 peptide similar to
full-length Mtr4. We conclude that the Trf5 peptide binds to
the core region of Mtr4, consistent with a recent co-crystal
structure (see Discussion; 38).
Although these fluorescence anisotropy data indicate that
Trf5 residues 98117 directly bind to Mtr4, the modest Kd
of 10.7 M suggests other residues may also contribute.
One candidate is the very N-terminus of the Air1/2 subunit,
since previous co-immunoprecipitation experiments have
implicated it in Mtr4 interaction (39). We tested a peptide
corresponding to Air2 residues 129 for binding to Mtr4
by fluorescence anisotropy and observed an apparent Kd of
6.9 M (Figure 2D). The Air2 peptide binds with similar
affinity to an N-terminal truncation (Mtr4 74; Figure 2E)
or an Mtr4 arch deletion (Mtr4archless; Figure 2F). Thus,
while the Trf598117 peptide can bind Mtr4, an Air1/2
peptide may also contribute to TRAMP complex assembly
(see Discussion; 38,39).
A 20 amino acid peptide in the N-terminus of Trf5 is
important for TRAMP complex formation
Although the above analysis suggests two peptides
mediate the interaction between Trf/Air and Mtr4, it does not
address whether both are required. Since it has previously
been shown that the Air N-terminal peptide is required for
co-immunoprecipitating Mtr4, but not for viability (39), we
focused on testing whether the Trf5 peptide is also
important. Initially, we compared the binding of Mtr4 to either
the Trf5 98117 peptide or Trf5 deleted for this peptide by
yeast two-hybrid analysis. Figure 3A shows that unlike Trf5
98117, the Trf5 construct lacking 98117 (Trf5- 98117)
fails to interact.
Although a positive interaction in yeast two-hybrid
analysis is often informative, the negative interaction between
Trf5- 98117 and Mtr4 is not definitive. We therefore
sought to analyze whether residues 98117 are indeed
required for Mtr4 interaction in the endogenous TRAMP
complex. To this end, we generated TAP-tagged versions of
both full-length Trf5 and Trf5- 98117 and expressed these
in yeast from their endogenous promoters. The TAP-tagged
Trf5 proteins were purified and tested for co-purification
itsb012 ENDG89RKHNDSGAE99KDSQAL010WVI E110KNDSATE201KNDMGAT310NE410HNDSQA510DF610Y710VI E810PKSGAFL910VF011DSGAFLE111DSMAWL211NSTE311DSTE411PDSGTE511DGAE611DAV711
vector + vector
Mtr4+ trf5 98-117
Mtr4+ trf5 98-117
Mtr4-archless+ trf5 98-117
Mtr4-archless+ trf5 98-117
with endogenous Mtr4 by western blot with antibodies
raised against Mtr4. As shown in Figure 3B, endogenous
Mtr4 was readily detectable in the purification of full-length
Trf5, but not in the purification of Trf5- 98117. We
conclude that residues 98117 of Trf5 form a major interaction
site for Mtr4, and that preventing this interaction impairs
formation of TRAMP in vivo.
The Mtr4/Trf interaction is not required for viability, but is
required for specific TRAMP functions
To explore the role of assembly of Trf4/5 with Mtr4 in a
stable TRAMP complex, we tested whether the trf5-98117
plasmid could function as the sole source of Trf4/5 through
a plasmid shuffle assay. Surprisingly, the trf5-98117
allele fully complemented the lethality of a trf4, trf5 strain
(Figure 4A). Furthermore, the steady-state protein levels of
Trf5-98117 were similar to those for wild-type Trf5 (data
not shown). We conclude that stable association with Mtr4
is not needed for protein stability or for the essential
function of Trf4/5.
Having generated a strain impaired for TRAMP complex
formation, we tested whether it affected specific TRAMP
functions. As their name implies, cryptic unstable
transcripts (CUTs) are not readily detectable in wild-type cells,
but they accumulate in mutants impaired in TRAMP or
exosome activity (15). TRAMP is thought to be responsible
for both their polyadenylation and subsequent degradation.
We therefore used qRT-PCR with gene-specific primers
during the reverse transcription step such that defects in
either polyadenylation or subsequent degradation would be
detectable. As expected, we detected accumulation of
several different CUTs in the absence of either the
exoribonuclease activity of Rrp44 or Rrp6 (rrp44-exo and rrp6,
respectively; Figure 4B; data not shown for other CUTs;
15,40). In contrast, a strain lacking the endonuclease
activity of Rrp44 (rrp44-endo) did not accumulate increased
CUT levels. Similarly, as previously reported (15,41,42), we
detected CUT accumulation in the absence of Trf4, that is
in trf4, trf5 complemented with a full-length Trf5
plasmid. Importantly, the CUT steady-state levels in the trf4,
trf5 strain complemented with a Trf5- 98117 plasmid
were similar to those in the same strain complemented with
full-length Trf5. Therefore, impairing the Mtr4/Trf
interaction does not affect the steady-state level of CUTs.
Mutations in TRAMP complex or exosome subunits
have also been shown to result in 3 extended and
polyadenylated snoRNA species. Thus, we next examined
the effect of trf5-98117 on two representative snoRNAs,
the C/D box snoRNA snR128 and the H/ACA snoRNA
snR33. As previously reported (18,19,29,43), 3 extended
species were detectable by northern blotting in rrp6 and
rrp44-exo strains. These species are detected as smears,
instead of discreet products, indicative of their
polyadenylation. We did not detect an increase in these species in
the rrp44-endo mutant. These polyadenylated snoRNA
species have also previously been reported in trf4 mutants,
and consistent with this, we detected them in a trf4, trf5
double mutant complemented with a wild-type TRF5
plasmid. However, these species were less abundant in this
Trf598-124 binding to Mtr4
Mtr4WT ( M)
Air21-29 binding to Mtr4
Mtr474 ( M)
K = 6.9 M 0.6
Kdd = 6.9 M 0.6
Mtr4WT ( M)
Mtr474 ( M)
10 mM 3-AT
trf4, trf5 [TRF5] strain than in the exonuclease strains.
Importantly, the trf4, trf5 strain complemented with
the trf5-98117 plasmid reproducibly accumulated more
3 extended snoRNAs than the same strain complemented
with the wild-type TRF5 plasmid. As previously described
for exosome mutants (18,19), the steady-state level of
mature snoRNAs was not altered in the trf5-98117 strain
(Figure 4C; 18,19). Thus, impairing the Mtr4/Trf
interaction interferes with normal processing or degradation of
these 3 extended snoRNA species.
To study the effects of disrupting the Trf5-Mtr4
interaction more globally, we also analyzed duplicate samples
of the TRF5 and trf5-98117 strains by transcriptome
sequencing of poly(A)+ RNA. This analysis revealed that
the set of most significantly affected genes (false discovery
rate (FDR) = 0.01) included 71 that were overexpressed
in trf5-98117 (Table 1) and only five that were
downregulated. The set of 71 overexpressed genes was
predominated by known TRAMP substrates, including 43 snoRNA
genes. The snoRNAs detected as RNA-seq hits included
both C/D box and H/ACA box snoRNAs. snoRNAs are
processed from primary transcripts in a variety of ways.
Figure 5 shows examples of monocistronically encoded
snoRNAs that are either only 3 processed (snR8; Figure 5A) or,
3' extended snoRNA snR128
3' extended snoRNA snR33
SRP (loading control)
in addition, are 5 processed by Rnt1 and Rat1 (snR87;
Figure 5B). Also shown is an example of a snoRNA that is
processed from a spliced intron (snR18; Figure 5C). For this
and other intron-encoded snoRNAs there was a clear
increase in reads that mapped to the snoRNA and the part
of the intron that is 3 of the snoRNA, but there was no
effect on the flanking protein-coding exons. Similarly, other
mRNAs (such as RPL11A; Figure 5E) were not affected.
Finally, Figure 5D shows an example of seven snoRNAs
(snR7278) that are transcribed as one polycistronic
precursor. For each snoRNA there is a clear increase in the
read density for both the snoRNA and the region just 3
of the mature snoRNA. Strikingly, among the other genes
that were detected as overexpressed are seven genes that are
just 3 of one of the overexpressed snoRNA genes (Table 1).
The inclusion of these genes in the RNA-seq hits is likely
due to the presence of 3 extended polyadenylated
snoRNAs. All of these changes were clearly reproducible in the
duplicate transcriptome sequencing samples (e.g.
Supplementary Figure S2). Thus, no particular kind of snoRNA
appeared to be overrepresented among the RNA-seq hits.
Although 23 other snoRNAs were not in the list of hits at
the 0.01 FDR, most of these were significantly up-regulated
at reduced stringency (P-values 0.004 to 0.05), and these
snoRNAs also were not enriched for any particular kind
of snoRNA. This set of snoRNAs detected as up-regulated
at reduced stringency included snR33, which we had
arbitrarily chosen to analyze by northern blot (Figure 4C;
P < 0.004 for snR33). Finally, the RNA-seq hits included
six other non-coding RNA loci that have previously been
shown to be TRAMP and/or RNA exosome substrates
(rRNA, U1 and U6 snRNA and the RNA subunits of the
signal recognition particle, RNase P and RNase MRP; 18
20,33). These hits include RNAs transcribed by RNA
polymerase I, II and III. Thus, poly(A)+ transcriptome
sequencing data confirmed and extended our northern blot
analysis. Overall, the RNA analyses of the trf5-98117 strain
indicate that many polyadenylated TRAMP substrates
accumulate if Mtr4/Trf interaction is disrupted.
Based on our new and previously published results we
conclude that the TRAMP complex is composed of two
wellfolded catalytic cores that are brought together by two
short peptide motifs. The majority of Mtr4 is well folded
and forms an RNA-dependent ATPase core (32,37), while
the polyadenylation catalytic core is assembled from
wellfolded domains of the Trf4/5 and Air1/2 subunits (44). In
addition to these well-folded domains, each TRAMP
subunit appears to have intrinsically disordered regions that
function to mediate proteinprotein interactions.
Specifically, the N-termini of both Trf4/5 and Air1/2 contain
small peptides that directly interact with Mtr4 (Figure
2; 38). Deletion of either one of these peptides impairs
TRAMP complex formation such that Mtr4 is no longer
immunoprecipitated with the Trf or Air subunit (Figure 3;
38,39). In vitro, each of these peptides establish a relatively
low affinity interaction but combine for a high affinity
interaction between the cores (38). We suspect that deleting
one of the interaction sites eliminates formation of a
snR78 snR77 snR76 snR75 snR74 snR73 snR72
ble TRAMP complex in vivo, since we and others have not
been able to detect any TRAMP complex under such
conditions (Figure 3; 38,39). However, we cannot exclude that
a less stable or transient interaction mediated through the
other peptide occurs in each of these experiments. In
addition to these peptide motifs that mediate TRAMP
complex assembly, the Mtr4 N-terminus contains a small motif
that mediates interaction with the exosome cofactors Rrp6
and Rrp47 (45), the C-terminus of Trf4 contains a small
motif that mediates interaction with Nrd1 (46), and the
Ctermini of Trf5, Air1, and Air2 contain short conserved
motifs that may mediate interactions with additional factors
(Supplementary Figure S1 and unpublished observations).
Whether all of these interactions occur simultaneously or
whether TRAMP dynamically associates with these factors
remains to be determined.
While we were preparing this manuscript, Falk et al.
independently identified similar Trf4 and Air2 regions that can
directly interact with Mtr4 (38). Specifically, they showed
that a complex of recombinant Trf4 and Air2 that
contained either one of the peptide motifs interacted with Mtr4
with affinities very similar to those we determined. The
agreement is remarkable considering the difference in
methods (isothermal titration calorimetry/ITC versus
fluorescence anisotropy), Mtr4 construct (Mtr4- N80 and
N74) and Trf/Air binding partner (dimeric protein
complexes versus isolated peptides). Furthermore, both the ITC
and anisotropy measurements suggest that neither the
Nterminus nor the arch domain of Mtr4 is required for these
interactions. Overall, the in vitro data presented here and
by Falk et al. suggest that the two peptides make equivalent
contributions to TRAMP assembly.
The Falk et al. paper also presented a crystal structure
of a Trf4-Air2 fusion peptide bound to Mtr4 (38). In this
structure, Air2 residues 518 bind to the fist domain of Mtr4
(which is part of the arch domain). However, we detected no
significant binding of Air2 residues 129 to the isolated fist
domain in solution (Supplementary Figure S3), and both
Falk et al. and our data show that Air2 binds with equal
affinity to full-length and archless Mtr4 in solution,
suggesting that the interactions with the fist do not make critical
contributions to TRAMP assembly. Furthermore, the Falk
et al. structure indicates that Air2 residues 129 make
additional contacts with the helical bundle (also named ratchet)
domain of Mtr4, but not the RecA domains, while our data
indicate that the RecA domains of Mtr4 are sufficient for
binding of Air2 residues 129 in solution (Supplementary
Figure S3). Thus, the solution binding assays and crystal
structure do not fully agree on the binding site on Mtr4 for
Air2. One explanation for the discrepancy between the
solution binding assays and crystal structure is that the
interactions observed in the crystal structure may be influenced
by the artificial tethering of the Air2 and Trf4 peptides that
was needed to facilitate crystallization (38). Additional
biochemical and structural characterization will be needed to
fully define the Air2-Mtr4 interface. Importantly, our
studies and the published crystal structure agree on the critical
region in Trf4/5 that mediates Mtr4 interaction.
By deleting residues 98117 of Trf5, we show that
impairing TRAMP complex formation does not result in an
obvious growth defect. Falk et al. showed that mutating residues
in Mtr4-GFP residues critical for Trf4/5 interaction caused
a slow growth phenotype (38). There are at least two
explanations that can unite these observations. First, as
discussed by Falk et al., it is possible that the Mtr4 residues
that interact with Trf4/5 are also important for some other
aspect of Mtr4 function. Alternatively, it is possible that
disrupting the Mtr4-Tr4/5 interaction is tolerated in a
wildtype Mtr4 strain but not in an Mtr4-GFP strain.
Consistent with this latter possibility is the observation that some
other mutations in Mtr4 that have no growth phenotype in a
wild-type Mtr4 strain, cause slow growth in the Mtr4-GFP
strain (45). In either case, our results show that the Mtr4/Trf
interaction is not required for viability. Both northern blot
and transcriptome sequencing indicates that this strain has
a defect in snoRNA processing and some other TRAMP
functions. Most, if not all, snoRNAs accumulate as 3
extended species in the poly(A)+ fraction. This suggests that
these RNAs can still be polyadenylated by Trf5 but then fail
to be degraded by the exosome. The simplest interpretation
appears to be that stable TRAMP complex assembly is
required to efficiently hand off substrates from the Trf subunit
to the Mtr4 subunit and subsequently to the exosome.
However, many alternative explanations cannot be excluded. For
example, it has been shown that Mtr4 can modify the in vitro
activity of Trf4 and vice versa (21,47), and thus impairing
the interaction between Mtr4 and Trf4/5 may have effects
beyond simply handing off substrates.
Our initial sequence analysis identified residues 98117
to be conserved in orthologs of Trf4 and Trf5 from other
yeast species, but conservation in other species was not
readily detectable. However, it is notable that a multiple
sequence alignment of PAPD5/hTFR4 orthologs from
vertebrates also identifies a small conserved region in an
otherwise poorly conserved N-terminus. The conserved sequence
in vertebrates (EQxDFi/lP) is similar to the sequence
conserved in yeast (d/nNxDFIxf/l). We therefore suspect that
the interaction we describe between the yeast proteins is
conserved in animals, even though standard sequence
analysis tools fail to detect sequence conservation.
Supplementary Data are available at NAR Online.
We thank Drs. Stanley Fields and Stepanka Vanacova for
generously providing plasmids and strains, Dr. Patrick
Linder for -Mtr4 antibodies and Dr. Joshua Price for synthesis
of the Air2 peptide. We thank members of the Johnson and
van Hoof labs for insightful comments.
NIH [R01GM099790 to A.v.H.]; Welch foundation
[AU1773 to A.v.H.]; NSF [MCB-1020739 to A.v.H.]; NSF
[MCB-0952920 to S.J.]. Graduate Dissertation
Enhancement Award (Utah State University) [to J.B.]. Funding for
open access charge: NIH grant R01GM099790.
Conflict of interest statement. None declared.
1. Mitchell , P. , Petfalski , E. and Tollervey , D. ( 1996 ) The 3 end of yeast 5.8S rRNA is generated by an exonuclease processing mechanism . Genes Dev. , 10 , 502 - 513 .
2. Mitchell , P. , Petfalski , E. , Shevchenko , A. , Mann , M. and Tollervey , D. ( 1997 ) The exosome: a conserved eukaryotic RNA processing complex containing multiple 3 ->5 exoribonucleases . Cell , 91 , 457 - 466 .
3. Kadaba , S. , Krueger , A. , Trice , T. , Krecic , A.M. , Hinnebusch , A.G. and Anderson , J. ( 2004 ) Nuclear surveillance and degradation of hypomodified initiator tRNAMet in S. cerevisiae. Genes Dev. , 18 , 1227 - 1240 .
4. de la Cruz , J. , Kressler , D. , Tollervey , D. and Linder , P. ( 1998 ) Dob1p (Mtr4p) is a putative ATP-dependent RNA helicase required for the 3 end formation of 5.8S rRNA in Saccharomyces cerevisiae . EMBO J ., 17 , 1128 - 1140 .
5. Jacobs Anderson , J.S. and Parker , R. ( 1998 ) The 3 to 5 degradation of yeast mRNAs is a general mechanism for mRNA turnover that requires the SKI2 DEVH box protein and 3 to 5 exonucleases of the exosome complex . EMBO J ., 17 , 1497 - 1506 .
6. LaCava , J. , Houseley , J. , Saveanu , C. , Petfalski , E. , Thompson , E. , Jacquier , A. and Tollervey , D. ( 2005 ) RNA degradation by the exosome is promoted by a nuclear polyadenylation complex . Cell , 121 , 713 - 724 .
7. Vanacova , S. , Wolf , J. , Martin , G. , Blank , D. , Dettwiler , S. , Friedlein , A. , Langen , H. , Keith , G. and Keller , W. ( 2005 ) A new yeast poly(A) polymerase complex involved in RNA quality control . PLoS Biol ., 3 , e189 .
8. Byrne , K.P. and Wolfe , K.H. ( 2005 ) The Yeast Gene Order Browser: combining curated homology and syntenic context reveals gene fate in polyploid species . Genome Res ., 15 , 1456 - 1461 .
9. Kellis , M. , Birren , B.W. and Lander , E.S. ( 2004 ) Proof and evolutionary analysis of ancient genome duplication in the yeast Saccharomyces cerevisiae . Nature , 428 , 617 - 624 .
10. Lubas , M. , Christensen , M.S. , Kristiansen , M.S. , Domanski , M. , Falkenby , L.G. , Lykke-Andersen , S. , Andersen , J.S. , Dziembowski , A. and Jensen , T.H. ( 2011 ) Interaction profiling identifies the human nuclear exosome targeting complex . Mol. Cell , 43 , 624 - 637 .
11. Giaever , G. , Chu , A.M. , Ni , L. , Connelly , C. , Riles , L. , Veronneau , S. , Dow , S. , Lucau-Danila , A. , Anderson , K. , Andre , B. et al. ( 2002 ) Functional profiling of the Saccharomyces cerevisiae genome . Nature , 418 , 387 - 391 .
12. Castano , I.B. , Heath-Pagliuso , S. , Sadoff ,B.U., Fitzhugh , D.J. and Christman , M.F. ( 1996 ) A novel family of TRF (DNA topoisomerase I-related function) genes required for proper nuclear segregation . Nucleic Acids Res ., 24 , 2404 - 2410 .
13. Inoue , K. , Mizuno , T. , Wada , K. and Hagiwara , M. ( 2000 ) Novel RING finger proteins, Air1p and Air2p, interact with Hmt1p and inhibit the arginine methylation of Npl3p . J. Biol. Chem., 275 , 32793 - 32799 .
14. David , L. , Huber , W. , Granovskaia , M. , Toedling , J. , Palm , C.J. , Bofkin , L. , Jones, T. , Davis , R.W. and Steinmetz , L.M. ( 2006 ) A high-resolution map of transcription in the yeast genome . Proc. Natl. Acad. Sci. U.S.A. , 103 , 5320 - 5325 .
15. Wyers , F. , Rougemaille , M. , Badis , G. , Rousselle , J.C. , Dufour , M.E. , Boulay , J. , Regnault , B. , Devaux , F. , Namane , A. , Seraphin , B. et al. ( 2005 ) Cryptic pol II transcripts are degraded by a nuclear quality control pathway involving a new poly(A) polymerase . Cell , 121 , 725 - 737 .
16. Malet , H. , Topf , M. , Clare , D.K. , Ebert , J. , Bonneau , F. , Basquin , J. , Drazkowska , K. , Tomecki , R. , Dziembowski , A. , Conti , E. et al. ( 2010 ) RNA channelling by the eukaryotic exosome . EMBO Rep ., 11 , 936 - 942 .
17. Bonneau , F. , Basquin , J. , Ebert , J. , Lorentzen , E. and Conti , E. ( 2009 ) The yeast exosome functions as a macromolecular cage to channel RNA substrates for degradation . Cell , 139 , 547 - 559 .
18. Allmang , C. , Kufel , J. , Chanfreau , G. , Mitchell , P. , Petfalski , E. and Tollervey , D. ( 1999 ) Functions of the exosome in rRNA, snoRNA and snRNA synthesis . EMBO J ., 18 , 5399 - 5410 .
19. van Hoof, A. , Lennertz , P. and Parker , R. ( 2000 ) Yeast exosome mutants accumulate 3 -extended polyadenylated forms of U4 small nuclear RNA and small nucleolar RNAs . Mol. Cell. Biol ., 20 , 441 - 452 .
20. Wlotzka ,W., Kudla , G. , Granneman , S. and Tollervey , D. ( 2011 ) The nuclear RNA polymerase II surveillance system targets polymerase III transcripts . EMBO J ., 30 , 1790 - 1803 .
21. Jia , H. , Wang , X. , Liu , F. , Guenther , U.P. , Srinivasan , S. , Anderson , J.T. and Jankowsky , E. ( 2011 ) The RNA helicase Mtr4p modulates polyadenylation in the TRAMP complex . Cell , 145 , 890 - 901 .
22. van Hoof, A. , Frischmeyer , P.A. , Dietz , H.C. and Parker , R. ( 2002 ) Exosome-mediated recognition and degradation of mRNAs lacking a termination codon . Science , 295 , 2262 - 2264 .
23. Meaux , S. and van Hoof , A. ( 2006 ) Yeast transcripts cleaved by an internal ribozyme provide new insight into the role of the cap and poly(A) tail in translation and mRNA decay . RNA , 12 , 1323 - 1337 .
24. Bernstein , J. , Ballin , J.D. , Patterson , D.N. , Wilson , G.M. and Toth , E.A. ( 2010 ) Unique properties of the Mtr4p-poly(A) complex suggest a role in substrate targeting . Biochemistry , 49 , 10357 - 10370 .
25. Jia , H. , Wang , X. , Anderson , J.T. and Jankowsky , E. ( 2012 ) RNA unwinding by the Trf4/Air2/Mtr4 polyadenylation (TRAMP) complex . Proc. Natl. Acad. Sci. U.S.A. , 109 , 7292 - 7297 .
26. Taylor , L.J. , Jackson , R.N. , Rexhepaj , M. , King , A. , Lott , L. , van Hoof , A. and Johnson , S.J. ( 2014 ) The Mtr4 ratchet helix and arch domain both function to promote RNA unwinding . Nucleic Acids Res ., 42 , 13861 - 13872 .
27. Schneider , C. , Kudla , G. , Wlotzka , W. , Tuck , A. and Tollervey , D. ( 2012 ) Transcriptome-wide analysis of exosome targets . Mol. Cell , 48 , 422 - 433 .
28. Uetz , P. , Giot , L. , Cagney , G. , Mansfield , T.A. , Judson , R.S. , Knight , J.R. , Lockshon , D. , Narayan , V. , Srinivasan , M. , Pochart , P. et al. ( 2000 ) A comprehensive analysis of protein-protein interactions in Saccharomyces cerevisiae . Nature , 403 , 623 - 627 .
29. Klauer ,A.A. and van Hoof , A. ( 2013 ) Genetic interactions suggest multiple distinct roles of the arch and core helicase domains of Mtr4 in Rrp6 and exosome function . Nucleic Acids Res ., 41 , 533 - 541 .
30. Sikorski , R.S. and Hieter , P. ( 1989 ) A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae . Genetics , 122 , 19 - 27 .
31. Schaeffer , D. , Tsanova , B. , Barbas , A. , Reis , F.P. , Dastidar , E.G. , Sanchez-Rotunno , M. , Arraiano , C.M. and van Hoof , A. ( 2009 ) The exosome contains domains with specific endoribonuclease, exoribonuclease and cytoplasmic mRNA decay activities . Nat. Struct. Mol. Biol ., 16 , 56 - 62 .
32. Jackson , R.N. , Klauer , A.A. , Hintze , B.J. , Robinson , H. , van Hoof , A. and Johnson , S.J. ( 2010 ) The crystal structure of Mtr4 reveals a novel arch domain required for rRNA processing . EMBO J ., 29 , 2205 - 2216 .
33. San Paolo ,S., Vanacova , S. , Schenk , L. , Scherrer, T. , Blank , D. , Keller , W. and Gerber , A.P. ( 2009 ) Distinct roles of non-canonical poly(A) polymerases in RNA metabolism . PLoS Genet ., 5 , e1000555 .
34. James , P. , Halladay , J. and Craig , E.A. ( 1996 ) Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast . Genetics , 144 , 1425 - 1436 .
35. Robinson , M.D. , McCarthy , D.J. and Smyth , G.K. ( 2010 ) edgeR: a Bioconductor package for differential expression analysis of digital gene expression data . Bioinformatics , 26 , 139 - 140 .
36. Tompa , P. , Davey , N.E. , Gibson , T.J. and Babu , M.M. ( 2014 ) A million peptide motifs for the molecular biologist . Mol. Cell , 55 , 161 - 169 .
37. Weir , J.R. , Bonneau , F. , Hentschel , J. and Conti , E. ( 2010 ) Structural analysis reveals the characteristic features of Mtr4, a DExH helicase involved in nuclear RNA processing and surveillance . Proc. Natl. Acad. Sci. U.S.A. , 107 , 12139 - 12144 .
38. Falk , S. , Weir , J.R. , Hentschel , J. , Reichelt , P. , Bonneau , F. and Conti , E. ( 2014 ) The molecular architecture of the TRAMP complex reveals the organization and interplay of its two catalytic activities . Mol. Cell , 55 , 856 - 867 .
39. Holub , P. , Lalakova , J. , Cerna , H. , Pasulka , J. , Sarazova , M. , Hrazdilova , K. , Arce , M.S. , Hobor , F. , Stefl , R. and Vanacova , S. ( 2012 ) Air2p is critical for the assembly and RNA-binding of the TRAMP complex and the KOW domain of Mtr4p is crucial for exosome activation . Nucleic Acids Res , 40 , 5679 - 5693 .
40. Lebreton , A. , Tomecki , R. , Dziembowski , A. and Seraphin , B. ( 2008 ) Endonucleolytic RNA cleavage by a eukaryotic exosome . Nature , 456 , 993 - 996 .
41. Houseley , J. , Kotovic , K. , El Hage , A. and Tollervey , D. ( 2007 ) Trf4 targets ncRNAs from telomeric and rDNA spacer regions and functions in rDNA copy number control . EMBO J ., 26 , 4996 - 5006 .
42. Fasken , M.B. , Leung , S.W. , Banerjee , A. , Kodani , M.O. , Chavez , R. , Bowman , E.A. , Purohit , M.K. , Rubinson , M.E. , Rubinson , E.H. and Corbett , A.H. ( 2011 ) Air1 zinc knuckles 4 and 5 and a conserved IWRXY motif are critical for the function and integrity of the Trf4/5-Air1/2-Mtr4 polyadenylation (TRAMP) RNA quality control complex . J. Biol. Chem. , 286 , 37429 - 37445 .
43. Schneider , C. , Leung , E. , Brown , J. and Tollervey , D. ( 2009 ) The N-terminal PIN domain of the exosome subunit Rrp44 harbors endonuclease activity and tethers Rrp44 to the yeast core exosome . Nucleic Acids Res ., 37 , 1127 - 1140 .
44. Hamill , S. , Wolin , S.L. and Reinisch , K.M. ( 2010 ) Structure and function of the polymerase core of TRAMP, a RNA surveillance complex . Proc. Natl. Acad. Sci. U.S.A. , 107 , 15045 - 15050 .
45. Schuch , B. , Feigenbutz , M. , Makino , D.L. , Falk , S. , Basquin ,C., Mitchell , P. and Conti , E. ( 2014 ) The exosome-binding factors Rrp6 and Rrp47 form a composite surface for recruiting the Mtr4 helicase . EMBO J ., 33 , 2829 - 2846 .
46. Tudek , A. , Porrua , O. , Kabzinski , T. , Lidschreiber , M. , Kubicek , K. , Fortova , A. , Lacroute , F. , Vanacova , S. , Cramer , P. , Stefl , R. et al. ( 2014 ) Molecular basis for coordinating transcription termination with noncoding RNA degradation . Mol. Cell , 55 , 467 - 481 .
47. Jia , H. , Wang , X. , Anderson , J.T. and Jankowsky , E. ( 2012 ) RNA unwinding by the Trf4/Air2/Mtr4 polyadenylation (TRAMP) complex . Proc. Natl. Acad. Sci. U.S.A. , 109 , 7292 - 7297 .