Characterization of LINE-1 Ribonucleoprotein Particles
Aure ´lien J. Doucet1,2*, Amy E. Hulme2, Elodie Sahinovic1, Deanna A. Kulpa2, John B. Moldovan3, Huira C.
Kopera2, Jyoti N. Athanikar2, Manel Hasnaoui1, Alain Bucheton1, John V. Moran2,3,4,5*, Nicolas Gilbert1*
1Institut de Ge ´ne ´tique Humaine, CNRS, UPR 1142, Montpellier, France, 2Department of Human Genetics, University of Michigan Medical School, Ann Arbor, Michigan,
United States of America, 3Cellular and Molecular Biology Program, University of Michigan Medical School, Ann Arbor, Michigan, United States of America, 4Department
of Internal Medicine, University of Michigan Medical School, Ann Arbor, Michigan, United States of America, 5Howard Hughes Medical Institute, Chevy Chase, Maryland,
United States of America
The average human genome contains a small cohort of active L1 retrotransposons that encode two proteins (ORF1p and
ORF2p) required for their mobility (i.e., retrotransposition). Prior studies demonstrated that human ORF1p, L1 RNA, and an
ORF2p-encoded reverse transcriptase activity are present in ribonucleoprotein (RNP) complexes. However, the inability to
physically detect ORF2p from engineered human L1 constructs has remained a technical challenge in the field. Here, we
have employed an epitope/RNA tagging strategy with engineered human L1 retrotransposons to identify ORF1p, ORF2p,
and L1 RNA in a RNP complex. We next used this system to assess how mutations in ORF1p and/or ORF2p impact RNP
formation. Importantly, we demonstrate that mutations in the coiled-coil domain and RNA recognition motif of ORF1p, as
well as the cysteine-rich domain of ORF2p, reduce the levels of ORF1p and/or ORF2p in L1 RNPs. Finally, we used this
tagging strategy to localize the L1–encoded proteins and L1 RNA to cytoplasmic foci that often were associated with stress
granules. Thus, we conclude that a precise interplay among ORF1p, ORF2p, and L1 RNA is critical for L1 RNP assembly,
function, and L1 retrotransposition.
Citation: Doucet AJ, Hulme AE, Sahinovic E, Kulpa DA, Moldovan JB, et al. (2010) Characterization of LINE-1 Ribonucleoprotein Particles. PLoS Genet 6(10):
Editor: Gregory S. Barsh, Stanford University, United States of America
Received February 10, 2010; Accepted September 3, 2010; Published October 7, 2010
Copyright: ? 2010 Doucet et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: AJD was the recipient of fellowships from the French government (Ministe `re de l’Enseignement Supe ´rieur et de la Recherche), from the Association
pour la Recherche contre le Cancer (ARC), and from the Fondation Recherche Medicale (FRM). AEH and DAK were supported in part by a Michigan Predoctoral
Training Grant from the National Institutes of Health (5T32GM07544). AEH also was supported by a Rackham Predoctoral Fellowship. JNA was supported in part
by a NRSA postdoctoral fellowship (F32GM20859). HCK was supported in part by an American Cancer Society postdoctoral fellowship (PF-07-059-01-GMC). Work
in the laboratory of NG is supported by the Centre National de Recherche Scientifique and the Agence Nationale de la Recherche (ANR-05-JCJC-0120). Work in the
laboratory of JVM is supported by an NIH grant (GM060518). The University of Michigan Cancer Center Support Grant (5P30CA46592) helped defray some of the
sequencing costs incurred during this study. JVM is an investigator of the Howard Hughes Medical Institute. The funders had no role in study design, data
collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: John V. Moran is listed as an inventor of the following patent: ‘‘Compositions and methods of use of human retrotransposons:
Application No. 60/006,831, Issued November, 2000.’’ Moran has received no money from the patent and the patent does not influence any of the results/
interpretations in the paper. This should not represent a conflict of interest, but is being voluntarily disclosed.
* E-mail: Nicolas.Gilbert@igh.cnrs.fr (NG); firstname.lastname@example.org (JVM); email@example.com (AJD)
LongInterspersed Element-1 (LINE-1or L1)sequences comprise
17% of human DNA and represent the predominant class of
autonomous retrotransposon-derived sequences in the genome .
Greater than 99.9% of L1 elements are molecular fossils that are no
longer capable of mobilization (i.e., retrotransposition) [1–3].
However, the average human genome still harbors a small cohort
(approximately 80–100) of retrotransposition-competent L1s (RC-
L1s) [4,5]. A wealth of experimental evidence suggests that ongoing
RC-L1retrotranspositionhasthe potential toimpactthe genomeby
a myriad of mechanisms (reviewed in [6–8]).
A human RC-L1 is approximately 6 kb in length; it begins with
a ,910 bp 59 untranslated region (UTR) that harbors an internal
RNA polymerase II promoter [9–11], two non-overlapping open
reading frames (ORF1 and ORF2), and ends with a 39 UTR that
is followed by either a polyadenylic acid (poly A) or A-rich
sequence (Figure 1A) [12,13]. Genetic and biochemical evidence
suggest that the ORF1 and ORF2-encoded proteins (ORF1p and
ORF2p, respectively) preferentially associate with their encoding
mRNA in cis to form a ribonucleoprotein particle (RNP) that
probably is an intermediate in the retrotransposition process
[14–19]. The resultant RNP then gains access to the nucleus,
where L1 integration presumably occurs by target-site primed
reverse transcription (TPRT) [20–23].
Studies conducted with mouse and human RC-L1s have
uncovered a number of conserved domains within ORF1p that
are important for retrotransposition. The amino acid sequence of
the ORF1p amino-terminus is poorly conserved among mamma-
lian L1s, but it is predicted to form a coiled-coil or a-helical
domain that is important for ORF1p multimerization [15,24–27].
In human ORF1p, this region contains a putative leucine zipper
(LZ) domain that is absent from other mammalian L1s, although a
similar motif is present in the L1-like Swimmer element of teleosts
[15,24,27–29]. The coiled-coil domain of ORF1p is followed by a
RNA recognition motif (RRM) , and experiments in cultured
human cells have shown that mutations in conserved residues of
the RRM domain (e.g., a N157A/R159A double mutant) adversely
affect L1 retrotransposition and the formation of cytoplasmic
structures known as ORF1 cytoplasmic foci .
The carboxyl-terminus of ORF1p contains amino acid residues
that are conserved among mammalian L1s [24,27,32]. Biochem-
PLoS Genetics | www.plosgenetics.org1October 2010 | Volume 6 | Issue 10 | e1001150
ical analyses have shown that mouse ORF1p homotrimers bind L1
RNA in a sequence independent manner [33,34]. Mutations of a
conserved di-arginine motif (RR261–262in human L1) can decrease
ORF1p RNA binding or mouse ORF1p nucleic acid chaperone
activity [33,35]. Similarly, studies using human L1s revealed that
alanine mutations in conserved amino acid residues in the
carboxyl terminus of ORF1p (RR261–262, and YPAKLS282–287,
respectively) both compromise the ability of ORF1p to localize to
RNPs and severely reduce L1 retrotransposition efficiency [16,32].
Thus, ORF1p is postulated to have critical functions at discrete
steps in the retrotransposition pathway.
Biochemical and genetic studies have revealed that human and
mouse ORF2 are translated by an unconventional mechanism
[36–39]. It is hypothesized that as few as one or two molecules of
ORF2p are translated per L1 RNA molecule, which could explain
why it has been difficult to detect ORF2p produced from
engineered L1s in cultured cells . ORF2p contains endonu-
clease (EN) and reverse transcriptase (RT) activities that are
critical for the target-site cleavage and reverse transcription steps
of TPRT [22,23,32,40]. ORF2p also contains a conserved
cysteine-rich (C) domain near its carboxyl-terminus [27,41].
Mutations in the C-domain adversely affect L1 retrotransposition
; however, the biochemical role of the C-domain in L1
retrotransposition remains poorly understood.
Epitope-tagging systems and enzymatic assays have been
developed to facilitate detection of L1 ORF1p and ORF2p RT
activity from engineered wild-type and mutant human L1s [16,17].
However, the inability to reliably and directly detect ORF2p from
engineered human L1s in transfected cultured human cells has
hindered progress in the field [37,42]. Here, we have devised an
epitope and/or RNA-tagging system to show that ORF1p, ORF2p,
and L1 RNA form a ribonucleoprotein complex, which may
represent a minimal RNP retrotransposition intermediate. Consis-
tent with previous studies, transient transfection/immunofluores-
cence-basedexperimentsrevealed that the L1-encodedproteinsand
L1 mRNA often form discrete cytoplasmic foci, and that many of
these foci associate with stress granules . Finally, we have
extended previous analyses [16,17] and demonstrate that mutations
in conserved functional domains of ORF1p and/or ORF2p
adversely affect L1 RNP formation, the reverse transcription of
L1 RNA, and L1 cytoplasmic foci formation. Thus, we have
developed a system that should allow a greater understanding of the
L1 retrotransposition mechanism at the molecular level.
A system to detect L1 ORF2p in cultured cells
Previous studies have examined the co-localization of L1
ORF1p and L1 RNA in RNPs derived from cells transfected
with epitope-tagged wild-type or mutant human L1 expression
constructs [16,17]. To physically detect L1 ORF2p, we modified
existing L1 expression vectors (pJM101/L1.3 or pDK101) to
contain either a 530 bp TAP tag or a 72 bp FLAG-HA tag on the
pES2TE1) [43,44]. To facilitate the identification of L1 RNA,
we also introduced a 1312 bp DNA fragment that contains 24
copies of a stem-loop sequence that can bind the phage MS2
protein into the L1 39UTR (Figure 1A; pAD3TE1) [45,46]. As a
control, we generated a plasmid that expresses TAP-tagged
ORF2p from a monocistronic transcript (Figure 1A; pAD500).
L1 constructs were equipped with a retrotransposition indicator
cassette (mneoI), subcloned into a pCEP4 episomal expression
vector, and were assayed for retrotransposition in cultured human
HeLa cells [32,47,48]. Inclusion of either the TAP or FLAG-HA
epitope tag onto the carboxyl-terminus of ORF2p had little effect
on the L1 retrotransposition efficiency when compared to a wild-
type control construct lacking the tag (Figure 1B; pADO2Tt,
pAD2TE1, and pES2TE1 vs. pJM101/L1.3). Similarly, the
inclusion of the MS2 stem loop sequences into the L1 39UTR
did not dramatically affect L1 retrotransposition efficiency
(Figure 1B; pADL1MT vs. pJM101/L1.3), although we did
observe an approximate 2.7 fold reduction in L1 retrotransposition
efficiency from a construct containing both the protein and MS2
tags (Figure 1B; pAD3TE1 vs. pJM101/L1.3). As a negative
control, we demonstrated that a construct containing a missense
mutation in the putative L1 RT active site (pAD135; D702A) was
defective for retrotransposition (Figure 1B). Thus, engineering
epitope and/or RNA tags into the L1 expression vectors is
compatible with retrotransposition in cultured cells.
(Figure 1A; pAD2TE1and
Physical detection of ORF2p in HeLa cells
To detect the L1-encoded proteins from the engineered
plasmids, we transfected each construct into HeLa cells and
selected for cells containing the respective L1 expression vectors by
exploiting the hygromycin B selectable marker on the pCEP4
episome (Figure 1A; see Materials and Methods). Consistent with
previous studies [16,17,49], western blot analyses of whole cell
lysates using antibodies directed against the ORF1p T7-epitope
tag revealed the presence of a ,40 kDa protein from constructs
containing the tag (Figure 2A, middle panel (aT7); pAD2TE1,
pAD3TE1, and pES2TE1), but not from controls lacking the tag
(Figure 2A; pJM101/L1.3 and pADO2Tt). We also could detect
the ,40 kDa protein with polyclonal antibodies against endoge-
nous human ORF1p (Figure 2B, aORF1 panels; pAD2TE1,
pJM101/L1.3, and pDK101) . Notably, we observed a slight
difference in the mobility of T7-tagged and untagged ORF1p
(Figure 2B; right panel (aORF1), pJM101/L1.3 vs. pDK101),
which most likely is due to the additional amino acids imparted by
the T7 epitope tag. Controls revealed that ORF1p was not
detected from a construct that lacks ORF1 (Figure 2A and 2B;
pAD500) or from a construct that contains a premature stop
Long Interspersed Element-1 (LINE-1 or L1) sequences are
the predominant class of autonomous retrotransposons in
the human genome and comprise an astounding 17% of
human DNA. Although the majority of L1s are considered
to be ‘‘dead,’’ an average human genome contains ,80–
100 active L1s. Active L1s encode two proteins (ORF1p and
ORF2p) that are required for mobility (retrotransposition)
by a ‘‘copy and paste’’ mechanism termed target-site
primed reverse transcription. Prior experiments suggested
that ORF1p, ORF2p reverse transcriptase activity, and L1
mRNA associate in ribonucleoprotein (RNP) particles and
that RNP formation is a necessary step in L1 retrotranspo-
sition. However, the difficulty in detecting ORF2p from
engineered human L1s has prevented a thorough under-
standing of its role in L1 retrotransposition. Here, we have
exploited epitope and/or RNA–tagging strategies to detect
and characterize a ‘‘basal’’ RNP complex from engineered
human L1s. We also expanded on previous studies and
characterized how mutations in conserved functional
domains of ORF1p and ORF2p can adversely affect L1
RNP formation/function. Finally, our strategy allowed us to
detect the L1–encoded proteins and L1 RNA in cytoplas-
mic foci. Thus, we have developed and employed a system
to gain greater understanding of LINE-1 retrotransposition
at the molecular level.
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org 2October 2010 | Volume 6 | Issue 10 | e1001150
Figure 1. The retrotransposition efficiency of engineered L1s used in this study. A. A diagram of L1 plasmids used in this study: Each plasmid
is a derivative of pJM101/L1.3 or pDK101 [4,16]. The constructs were tagged with the mneoI retrotransposition indicator cassette [32,47], and are
expressed from the pCEP4 episomal vector (Invitrogen). Labeled rectangles indicate the relative positions of the L1 59UTR, ORF1p and ORF2p. Labeled
flags at the 39 ends of ORF1 and/or ORF2 are used to denote the epitope tag in the respective constructs. pAD3TE1 also contains 24 copies of a stem-
loop sequence that can bind the phage MS2 protein (light rectangle labeled MS2 24x) . pAD500 is a monocistronic ORF2p expression vector that
lacks ORF1 as well as the inter-ORF spacer sequence. B. Representative results of the retrotransposition assay: L1 retrotransposition efficiency was
assayed as described previously [32,48]. HeLa cells transfected with pJM101/L1.3 serve as a positive control. Untransfected HeLa cells and HeLa cells
transfected with an RT mutant (pAD135; D702A) serve as negative controls. Cartoons of constructs used in the experiment are indicated in the figure.
All constructs contain the mneoI retrotransposition indicator cassette.
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org3 October 2010 | Volume 6 | Issue 10 | e1001150
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org4 October 2010 | Volume 6 | Issue 10 | e1001150
codon in ORF1 (Figure 2A; pADO1S). Qualitative reverse
transcriptase-PCR (RT-PCR) experiments further confirmed that
L1 RNA was expressed from each of the transfected constructs
(Figure 2C; see Oligonucleotides and RT-PCR sections in
Materials and Methods for details).
To detect ORF2p expression, we conducted western blot
analyses on whole cell lysates derived from transfected cells using
antibodies directed against either the TAP or FLAG-HA epitope
tag (Figure 2A). A ,170 kDa protein was detected from L1
constructs containing TAP-tagged ORF2p, but not from an
untagged wild-type control (Figure 2A; left panel (aTAP);
pADO2Tt, pAD2TE1, pAD3TE1, and pAD500 vs. pJM101/
L1.3). The ,170 kDa product corresponds to the predicted size of
ORF2p (,150 kDa) plus the predicted size of the TAP tag
(,19 kDa) [12,43]. A ,170 kDa protein also was detected using
antibodies against endogenous ORF2p (Figure 2B; left panel
(aORF2)) . Similarly, a ,155 kDa protein was detected from
L1 constructs containing FLAG-HA-tagged ORF2p, but not from
the untagged wild-type control (Figure 2A; right panel (aHA):
pES2TE1 vs. pJM101/L1.3). Consistent with previous genetic
studies, TAP-tagged ORF2p expression was greatly diminished by
introducing a stop codon in ORF1 (Figure 2A; pADO1S) and was
most abundant when expressed from an ORF2p monocistronic
expression vector (Figure 2A and 2B; pAD500) .
Epitope-tagged ORF2p localizes to ribonucleoprotein
To test whether ORF2p localizes to ribonucleoprotein particles
(RNPs), we transfected HeLa cells with pAD2TE1, selected for
transfected cells, and isolated RNPs by ultracentrifugation (see
Materials and Methods) [16,17]. Western blotting revealed that
ORF1p and ORF2p were readily detected in the RNP fraction
(Figure 3A, top panel).
We next used the L1 Element Amplification Protocol (LEAP)
assay to determine whether the RNP preparations contained an L1-
specific reverse transcriptase activity . Consistent with previous
studies, a diffuse set of LEAP products that ranged in size from
,220 to ,400 bp was detected in pAD2TE1-derived RNPs, but
not from pAD135-derived (D702A; RT mutant) RNPs (Figure 3A,
lower panel). Cloning and sequencing of the pAD2TE1-derived
LEAP products confirmed that L1 reverse transcription generally
initiated at variable sites within the L1 poly (A) tail, which accounts
for variably-sized LEAP products (data not shown ).
Epitope-tagged ORF2p form a complex with ORF1p and
their encoding RNA
To further verify that ORF1p, ORF2p, and L1 mRNA form an
RNP, HeLa cells were transfected with either pES2TE1 or
pDK101. Whole cell extracts then were subjected to immunopre-
cipitation using an anti-FLAG M2 antibody fused to agarose beads
(Figure 3B). Incubation of the beads with a FLAG peptide followed
by western blot analysis revealed an enrichment of ORF1p and
ORF2p in the pES2TE1, but not in the pDK101 immunoprecip-
itated reactions (Figure 3C). We sometimes detected a faint band
of ,40 kDa in the pDK101 immunoprecipitated reactions upon
longer film exposures, suggesting that some T7-tagged ORF1p
may bind non-specifically to the anti-FLAG M2 agarose beads
(data not shown). However, subsequent experiments/product
characterization determined that the pES2TE1 immunoprecipi-
tated fraction contained LEAP activity, whereas the pDK101
immunoprecipitated fraction lacked a readily detectable LEAP
activity (Figure 3D).
Interestingly, we consistently observed less ORF1p associated
with RNPs in immunoprecipitation experiments when compared to
experiments conducted with whole cell lysates or crude RNPs
(Figure 2A and Figure 3A). These data suggest either that ORF1p is
less tightly associated with L1 mRNA than ORF2p in RNPs (which
is consistent with previous observations ) and/or that a fraction
of ORF1p is dissociated from L1 RNA during the immunoprecip-
itation process. Regardless, whereas previous studies showed that
ORF1p, ORF2p RT activity, and L1 RNA co-localize to RNPs
[16,17], we were able to demonstrate the physical association of
these components in immunoprecipitation experiments.
Mutations in both ORF1p and ORF2p affect L1 RNP
formation and/or function
Previous studies identified activities associated with ORF1p and
ORF2p that are critical for L1 retrotransposition [22,30,32,35].
Here, we expanded on these analyses to determine whether
mutations in the L1-encoded proteins affect their ability to localize
to RNPs and/or impact L1 reverse transcriptase activity in the
LEAP assay. We first tested mutants in the following functional
domains of ORF1p: 1) the putative leucine zipper domain
(pAD113; NLR157–159ALA); 3) the carboxyl-terminal nucleic acid
binding domain (pAD105; RR261–262AA); 4) an ORF1p mutation
that affects mouse nucleic acid chaperone activity (pAD106;
RR261–262KK); and 5) a double mutant in the putative leucine
zipper domain and carboxyl-terminal nucleic acid binding domain
(pADL/R; L93,100,107,114V/RR261–262AA) (Figure 4A; see Materi-
als and Methods) [16,30–32,35]. Each of these mutations,
including the LZC mutation (pADLZC; L93,100,107,114V), severely
compromise L1 retrotransposition efficiency in HeLa cells (Figure
S1A, S1B). The LZC mutant data are in agreement with a
published report, which demonstrated a L93/100/114A triple
mutation inactivates L1 retrotransposition .
Figure 2. Detection of ORF1p and ORF2p from engineered L1 constructs. A. Representative results from western blot analyses: Whole cell
lysates derived from untransfected HeLa cells or HeLa cells transfected with the indicated L1 expression constructs were subjected to western blot
analyses. Top panels: western blots conducted with anti-TAP antibodies (aTAP; left side) or anti-HA antibodies (aHA; right side) to detect epitope-
tagged ORF2p. Middle panels: western blots conducted with anti-T7 antibodies (aT7) to detect epitope-tagged ORF1p. Lower panels: western blots
conducted with anti-tubulin antibodies (aTubulin) served as a loading control. Molecular weight standards (Invitrogen, left side, and New England
Biolabs, right side) are listed at the left of each series of gels. B. Protein detection specificity: Whole cell lysates derived from HeLa cells transfected with
the indicated constructs were subjected to western blot analyses using antibodies against either endogenous ORF1p (aORF1) or endogenous ORF2p
(aORF2). TAP-tagged ORF2p also was detected using an anti-TAP antibody. T7-tagged ORF1p also was detected with an anti-T7 antibody. Tubulin
was detected using an anti-tubulin (aTubulin) antibody and served as a loading control. Molecular weight standards (Invitrogen) are listed at the left
of each series of gels. C. RT-PCR analyses: RT-PCR reactions using RNAs isolated from whole cell lysates derived from transfected cells revealed that L1
RNA was expressed from each of the constructs. GAPDH mRNA detection was used to assess the quality of the RNA preparations and as a loading
control. Reactions without template (PCR control) or reverse transcriptase (RT control) were used as negative controls. DNA size markers (Invitrogen)
are indicated at the left of the gel. Colored cartoons of the constructs used in the experiments are indicated next to their respective names. The black
lines indicate the 59 and 39 UTRs. The green and red boxes indicate ORF1 and ORF2p respectively. When present, epitope tags are indicated. All
constructs contain the mneoI retrotransposition indicator cassette.
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org5October 2010 | Volume 6 | Issue 10 | e1001150
Figure 3. Biochemical identification of a basal L1 RNP complex. A. L1 RNPs contain ORF1p, ORF2p, L1 RNA, and L1 reverse transcriptase activity:
RNP pellets were obtained from untransfected HeLa cells, or from HeLa cells transfected with wild-type (pAD2TE1) and reverse transcriptase deficient
(pAD135) L1 constructs. As in Figure 2, tagged ORF1p and ORF2p were detected using anti-T7 (aT7) and anti-TAP (aTAP) respectively. Ribosomal S6
protein was detected using an anti-S6 (aS6) antibody and was used as an RNP loading control. Reverse transcriptase activity was detected using the
LEAP assay as described previously . Reactions without template (No Template) or RNPs (No RNP/RNA) were used as negative controls. Top panel:
LEAP reactions (LEAP-L1). Middle panel: L1 RT-PCR reactions conducted with M-MLV reverse transcriptase control for the presence of L1 RNA in RNPs
(M-MLV-L1). Bottom panel: GAPDH RT-PCR reactions conducted with M-MLV reverse transcriptase assess RNP RNA quality and serve as a RT-PCR
internal control (M-MLV-GAPDH). B. Flow chart of the L1 RNP immunoprecipitation reaction: Whole cell extracts were prepared from HeLa cells
transfected with either pDK101 or pES2TE1. Immunoprecipitation reactions were conducted by incubating the resultant lysates with agarose beads
fused to an anti-FLAG M2 antibody. The elution of ORF2p from the beads was performed by FLAG peptide competition. Western blotting and LEAP
assays were performed on aliquots of the whole cell extracts or the elution fractions to detect the L1-encoded proteins and L1-specific reverse
transcriptase activity, respectively. C. Co-immunoprecipitation of ORF1p and ORF2p: Whole cell extract (input) and immunoprecipitated (elution)
products from pDK101 or pES2TE1 transfected cells were subjected to western blotting to identify ORF2p (aHA; top panel), ORF1p (aT7 middle panel)
or tubulin (aTubulin, bottom panel). The femto ECL substrate (Pierce) was used to detect ORF1p and ORF2p. D. A basal L1 RNP complex contains L1
RNA and retains L1 reverse transcriptase specific activity: LEAP was performed on whole cell extracts (input) or immunoprecipitated (elution) products
from pDK101 or pES2TE1 transfected cells. Reactions conducted without template (No Template) or without RNPs (No RNP) were used as negative
controls. As in Figure 2, colored cartoons of the constructs are indicated in panels A, C and D. Molecular weight/DNA size markers (Invitrogen) are
indicated at the left of the images. All constructs contain the mneoI retrotransposition indicator cassette.
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org6 October 2010 | Volume 6 | Issue 10 | e1001150
Figure 4. TAP tagged ORF2p and RT activity detection in RNP preparation. A. Schematic representation of the amino acid mutation positions
in L1 sequence: The names of plasmids containing L1s with mutations in the ORF1p coiled-coil domain (CC-LZ), the ORF1p RNA recognition motif
(RRM), and the ORF1p carboxyl-terminal (CTD) domain are indicated below the schematic. The names of plasmids containing mutations in the ORF2p
endonuclease domain (EN), reverse transcriptase domain (RT) or cysteine-rich domain (C) also are shown. pADL/R is a double mutant that contains a
putative leucine zipper mutation and a carboxyl-terminal domain mutation in ORF1p. pADL/C is a double mutant that contains a putative leucine
zipper mutation in ORF1p and a C-domain mutation in ORF2p. The flags indicate the epitope tag present on ORF1 and ORF2. B. Detection of ORF1p
and ORF2p from mutant L1 constructs: RNPs from HeLa cells transfected with a RC-L1 (pAD2TE1) or the indicated mutant L1 constructs (see Figure 4A)
were analyzed by western blotting . Tagged L1 proteins were detected as in Figure 3; ORF2p (top panel), ORF1p (middle panel). Ribosomal S6
protein detection was used as a loading control (bottom panel). Molecular weight markers (Invitrogen) are indicated at the left of the image. C. L1 RT
activity of RNP fractions detected by LEAP: An aliquot from each of the indicated RNP preparations noted above was used to perform LEAP assays (see
Figure 3) . RNPs from pAD2TE1 served as a positive control. RNPs from untransfected HeLa cells or pAD135 (D702A; RT mutant) transfected cells
served as negative controls. Reactions without RNPs (No RNP/RNA) or template (No Template) also were used as negative controls. Top panel: LEAP
reactions (LEAP-L1). Middle panel: L1 RT-PCR reactions conducted with M-MLV reverse transcriptase control for the presence of L1 RNA in the RNP
fractions (M-MLV-L1). Bottom panel: GAPDH RT-PCR reactions conducted with M-MLV reverse transcriptase assess RNP RNA quality and serve as a RT-
PCR internal control (M-MLV-GAPDH). DNA size markers (Invitrogen) are indicated at the left of the image. All constructs in panel B and C contain the
mneoI retrotransposition indicator cassette.
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org7October 2010 | Volume 6 | Issue 10 | e1001150
Multiple independent RNP preparations derived from cells
transfected with each of the respective mutants were analyzed by
western blotting to examine the presence and abundance of both
ORF1p and ORF2p (Figure 4B). LEAP assays then were used to
determine whether those RNPs contained an L1-specific reverse
transcriptase activity (Figure 4C). Once again, control MLV RT-
PCR-based experiments, using the same oligonucleotide primers
employed in the LEAP assay, indicated that L1 RNA was present
at roughly comparable levels in the RNP fraction of HeLa cells
transfected with the mutant constructs (Figure 4C).
Consistent with previous data , a mutation in the ORF1p
carboxyl-terminal domain (pAD105; RR261–262AA) led to a severe
reduction in the ability of ORF1p, but not ORF2p, to localize to
the RNP fraction (Figure 4B). RNPs derived from pAD105-
transfected cells had a readily detectable LEAP activity, although
the constellation of LEAP products differed from those in the wild-
type control, pAD2TE1, because they frequently initiated reverse
transcription from within the 39 end of the L1 mRNA (Figure 4B
and 4C; Figure S2A, S2B, S2C and S2D; pDK105 RR261–262AA;
data not shown). Similar data also were observed for an L1
containing a mutation in the carboxyl-terminal domain (pDK116;
YPAKLS282–287AAAALA) as well as for pAD500, a TAP-tagged
ORF2p construct that lacks ORF1 (Figure 4B and Figure S2A and
S2B). These findings support the hypothesis that ORF2p can
preferentially associate with its encoding RNA independent of
ORF1p binding and that the resultant RNPs retain LEAP activity
. Indeed, the constellation of LEAP products observed in the
RR261–262AA, YPAKLS282–287AAAALA, and pAD500 mutants
support our previous hypothesis that ORF1p binding to L1
mRNA possibly may restrict hybridization of the LEAP primer to
the L1 poly (A) tail .
Mutations that affect mouse nucleic acid chaperone activity
(pAD106; RR261–262KK) had little effect on the ability of ORF1p
and ORF2p to localize to RNPs or on LEAP activity (Figure 4B;
[16,35]). We occasionally observed a greater abundance of the
lower molecular weight LEAP products, when compared to our
wild-type control, pAD2TE1 (Figure 4C). Indeed, closer inspec-
tion consistently revealed slightly higher levels of the major LEAP
products (,220 to ,400 bp) and a slightly lower level of the
shorter LEAP products from the RR261–262KK mutant (pAD106
and pDK106; Figure 4C, Figure S2B and S2C) when compared to
LEAP products derived from the RR261–262AA (pAD105 and
pDK105; Figure 4C, Figure S2B and S2C) and YPAKLS282–
287AAAALA mutants (pDK116; Figure S2B). Thus, although the
L1 RT activity detected in the LEAP assay does not appear to
require ORF1p, it is clear that specific mutations in ORF1p can
affect the constellation of products observed in these assays.
Mutations in the putative ORF1p leucine zipper-binding
domain (pADLZC; L93,100,107,114V) reduced ORF1p and ORF2p
localization in the RNP fraction and consistently exhibited lower
qualitative levels of LEAP activity when compared to the wild-type
control, pAD2TE1 (Figure 4B and 4C). Indeed, quantitative
LEAP experiments conducted with pLZC-derived RNPs (a
L93,100,107,114V mutant that lacks an epitope tag on ORF2p)
revealed a five to seven-fold reduction in LEAP activity when
compared to a corresponding wild-type control (pDK101; Figure
S2E). Subsequent data from LEAP experiments designed to detect
variable length L1 cDNA products further suggest that the LZC
mutation adversely affects early steps in the reverse transcription of
L1 RNA and does not appear to affect L1 RT elongation (Figure
The putative leucine zipper domain-carboxyl terminal domain
double mutant (ADL/R; L93,100,107,114V/RR261–262AA) shared
biochemical characteristics of each single mutant. Similar to
pAD105; RR261–262AA, ORF1p levels were severely reduced in
pADL/R-derived RNPs. However, similar to the putative leucine
zipper domain (pADLZC; L93,100,107,114V) mutant, ORF2p levels,
as well as LEAP activity, were reduced in pADL/R-derived RNPs
when compared to the wild-type control, pAD2TE1. Moreover,
the LEAP product profile in the double mutant resembled that in
the pAD105 mutant (Figure 4B and 4C; Figure S2). Thus, the
above data suggest that the LZC mutant adversely affects the
accumulation and/or stability of L1 RNPs and that the reduction
of ORF2p in RNPs likely contributes to the observed decrease in
Mutations in the ORF1p RRM domain (pAD113; NLR157–159
ALA) also led to a severe reduction in the ability of ORF1p and
ORF2p to localize to the RNP fraction of transfected cells
(Figure 4B). Indeed, ORF2p only was observed upon over-
exposure of the resultant western blots (data not shown). The
reduced level of ORF2p in pAD113-derived RNPs also correlated
with a decrease in LEAP activity when compared to the
pAD2TE1 wild-type control (Figure 4B and 4C). Notably, it is
unlikely that the NLR157–159ALA mutation dramatically affects
ORF2 translation because we can detect ORF2p from this mutant
by immunofluorescence (see below). Moreover, preliminary data
(n=4 independent experiments) indicates that the NLR157–159
ALA mutant can serve as a ‘‘driver’’ in a genetic-based trans-
complementation assay to mobilize a reporter gene (ORF1mneoI;
) at roughly 60 to 80% the level of the wild-type control,
pAD2TE1-NT (Doucet et al., preliminary data). These data are
consistent with previous genetic studies, which suggested that
ORF1p binding to L1 RNA is not required for ORF2 translation
. Moreover, the data suggest that the NLR157–159ALA
mutations severely compromise the accumulation and/or stability
of L1 RNPs (see Discussion).
We next tested mutants in the following functional domains of
ORF2p for their effect on L1 RNP formation and L1 reverse
transcriptase activity: 1) the L1 endonuclease domain (pAD136;
H230A); 2) the L1 reverse transcriptase domain (pAD135; D702A); 3)
the cysteine-rich domain (pAD162; CWWDC1143–1147SWWDS)
(Figure 4A) [19,22,32]. As expected, the L1 RT mutant (pAD135;
D702A) did not dramatically affect the ability of ORF1p or ORF2p
and 4C) . These data are consistent with previous suppositions
that the D702A mutant likely blocks the reverse transcription step in
We repeatedly observed a slight reduction of ORF2p in RNPs
derived from the tested endonuclease mutant, and this reduction
correlates with a reproducible decrease in LEAP activity
(Figure 4A; pAD136; H230A). We also observed a severe reduction
of ORF2p, as it was only detected upon longer film exposures
(data not shown), and a strong decrease of LEAP activity in RNPs
derived from the tested cysteine-rich domain mutant (pAD162;
CWWDC1143–1147SWWDS). Finally, the leucine zipper/C-domain
SWWDS) displayed both a reduction of ORF1p in RNPs and a
concomitant decrease in LEAP activity (Figure 4B and 4C).
As additional controls for the above experiments, we demon-
strated that mutant constructs containing a T7-epitope tag on
ORF1p, but lacking an ORF2p epitope tag exhibited similar
qualitative LEAP activities as the pAD2TE1 mutant based
constructs (Figure S2). We also demonstrated that the amount of
T7-tagged ORF1p and TAP-tagged ORF2p in whole cell lysates is
similar to that in the RNP fraction for each of the pAD2TE1
mutant constructs, and that these proteins were not enriched in
insoluble aggregates in the pellet obtained after cell lysis (data not
shown). Thus, we conclude that mutations within discrete
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org8October 2010 | Volume 6 | Issue 10 | e1001150
functional domains of ORF1p and ORF2p have differential effects
on L1 RNP formation/function.
Immunofluorescence detection of the L1–encoded
proteins and L1 mRNA
Previous studies have shown that ORF1p often aggregates in
cytoplasmic structures termed cytoplasmic foci . Unlike the
RNP assays described above (which detect the steady state amount
of ORF1p and ORF2p in the RNP fraction of hygromycin
resistant cells ,9 days post-transfection), the L1 cytoplasmic foci
formation assays allows the opportunity to visually detect the L1-
encoded proteins and/orL1
,48 hours post-transfection.
To test whether the ORF1p cytoplasmic foci also contain
ORF2p and L1 RNA, we conducted immunofluorescence-based
localization experiments in a U-2 OS human osteosarcoma cell
line that can support the retrotransposition of engineered human
L1 constructs (Figure S3A). Initial experiments conducted with
pAD2TE1 revealed that ORF1p and ORF2p generally co-
localized to discrete cytoplasmic foci 48 hours post-transfection,
and that many foci were located near the periphery of the nucleus
(Figure 5A). Time course analyses further demonstrated that
cytoplasmic foci were apparent in ,50% of transfected cells as
early as 12 hours post-transfection, and that ,90% of transfected
cells displayed cytoplasmic foci 72 hours post-transfection (Figure
S3B). ORF1p/ORF2p-containing cytoplasmic foci also were
observed in U-2 OS cells transiently transfected with pAD2TE1-
NT, which lacks the mneoI retrotransposition indicator cassette
(Figure 5A) and with a pAD2TE1 derivative lacking the
heterologous cytomegalovirus immediate early (CMV) promoter,
although foci appeared 24–48 hours later as compared to cells
transfected with the wild type control, pAD2TE1 (data not shown).
ORF1p and ORF2p co-localization also was observed using an
anti-HA antibody to detect ORF2p (Figure 5A; pES2TE1) or
antibodies against endogenous ORF1p or ORF2p (Figure S3C;
pES2TE1). Qualitatively similar results were obtained when
pAD2TE1 was transiently transfected into HeLa or 143Btk cells,
which also support L1 retrotransposition [32,51] (data not shown).
To test whether L1 RNA co-localizes with ORF1p and ORF2p
to cytoplasmic foci, we transiently transfected pAD3TE1 into U-2
OS cells. In situ hybridization experiments using a fluorescently-
labeled probe complementary to the MS2 stem loop structures in
the L1 39UTR revealed the presence of L1 RNA in cytoplasmic
foci as well as in nuclei of transfected cells (Figure 5B and Figure
S3D). The co-localization of ORF1p, ORF2p, and L1 RNA was
pAD3TE1 and a plasmid expressing a fluorescently labeled MS2
protein (Figure S3D), and by staining with antibodies against
ORF1p and ORF2p (Figure S3C). As above, qualitatively similar
results were obtained upon transient transfection of pAD3TE1
into HeLa or HEK293 cells, which also support L1 retrotranspo-
sition [32,51] (data not shown).
To determine whether L1 foci are associated with specific
cytoplasmic substructures, we co-transfected U-2 OS cells with
pAD2TE1 and plasmids that express GFP fusion proteins that can
localize to processing bodies (i.e., P-bodies) and/or stress granules.
Consistent with previous analyses, ORF1p and ORF2p associated
with an Ago2-GFP fusion protein that localizes both to P-bodies
and stress granules (Figure 5C; panel 1) [31,52]. Refining this
analysis revealed that ORF1p and ORF2p co-localized with the
stress granule marker G3BP-GFP , but did not associate with
the P-body marker DCP1a-GFP  (Figure 5C; panel 2 and 3).
By comparison, experiments conducted with fluorescently labeled
antibodies specific for eIF3 and G3BP [53,55] revealed that stress
RNA when over-expressed
granules appear to closely associate with the L1 foci (Figure 5C,
panel 4 and 5).
Together, the above data demonstrate that ORF1p, ORF2p,
and L1 mRNA co-localize to cytoplasmic foci when over-
expressed from a variety of engineered L1 episomal expression
constructs and that many of these cytoplasmic foci associate with
stress granules. However, future experiments are needed to
determine whether cytoplasmic foci represent accumulation
depots for L1 RNPs or if they play an important role in L1
Mutations in ORF1p and ORF2p adversely affect the
formation of L1 cytoplasmic foci
We next examined if mutations in the L1-encoded proteins
affect L1 cytoplasmic foci formation. Transient transfection of
ORF1p mutant expression vectors into U-2 OS cells followed by
immunofluorescence staining with anti-T7 and anti-TAP antibod-
ies confirmed that ORF1p and ORF2p are expressed in these cells
(Figure 6A). Consistent with previous studies, mutations in the
ORF1p RRM domain (pAD113; NLR157–159ALA) and carboxyl-
terminal RNA binding domain (pAD105; RR261–262AA) led to a
reduction in the number of L1 cytoplasmic foci (Figure 6A and 6B)
. A reduction in the number of L1 cytoplasmic foci also was
observed for an RRM domain mutant (pAD102; REKG235–238
AAAA),an additional carboxyl-terminal
(pAD116; YPAKLS282–287AAAALA), and the putative leucine
zipper domain/carboxyl-terminal RNA binding domain double
mutant (pADL/R; L93,100,107,114V/RR261–262AA). By comparison,
mutations in the putative ORF1p LZ domain (pADLZC;
L93,100,107,114V) or mutations that affect the nucleic acid
chaperone activity of mouse ORF1p (pAD106; RR261–262KK
and pAD107; R261K) had little effect on L1 cytoplasmic foci
formation (Figure 6A and 6B), although we sometimes observed an
apparent nucleolar localization of ORF1p in pADLZC transfected
cells. None of the ORF2p mutations had a dramatic effect on L1
cytoplasmic foci formation (Figure 6A and 6B), although, we
observed a diffuse nuclear localization of TAP-tagged ORF2p in
cells transfected with either pAD162 or the putative leucine zipper
L93,100,107,114V/CWWDC1143–1147SWWDS) (Figure 6A).
The above data suggest that the ability of ORF1p to bind L1
RNA is critical for L1 cytoplasmic foci formation (Figure 6B).
Consistent with this idea, we were able to detect L1 cytoplasmic
foci, as well as diffuse ORF1p staining, in U-2 OS cells transiently
transfected with a T7-tagged ORF1p expression vector (Figure 6C;
pDK500). However, L1 cytoplasmic foci were not detected in U-2
OS cells transiently transfected with a TAP-tagged ORF2p
expression vector (Figure 5D; pAD500). Thus, these data, as well
as our previously published trans-complementation experiments
, suggest that ORF1p interacts with its encoding RNA in cis,
and that this association allows L1 cytoplasmic foci formation in
the absence of ORF2p.
ORF2p has been notoriously difficult to detect from engineered
human L1s in cultured cells. It has been hypothesized that human
ORF2p is translated at low levels when compared to ORF1p and/
or may be an unstable protein, which might help explain why it
has evaded detection [36–39,42]. Previous biochemical studies
have identified human ORF2p from vascular endothelial cells in
vivo  and have demonstrated that ORF2p RT activity co-
localizes with ORF1p and L1 RNA in cytoplasmic RNPs derived
from HeLa cells transfected with wild-type engineered human L1
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org9October 2010 | Volume 6 | Issue 10 | e1001150
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org10 October 2010 | Volume 6 | Issue 10 | e1001150
expression constructs [16,17]. Here, we have built on these studies
and have combined epitope and RNA tagging strategies to
physically detect L1 ORF1p, ORF2p and L1 mRNA in
Why our approach allows the ready detection of ORF2p
expressed from engineered human L1 constructs requires further
study. Experiments conducted with anti-TAP antibodies consis-
tently yielded more robust detection of ORF2p when compared to
anti-HA or anti-ORF2p antibodies. Thus, the inclusion of a large
carboxyl-terminal tag, such as the TAP-tag, might stabilize
ORF2p. However, since engineered L1 constructs containing
either the TAP or HA epitope tags on the carboxyl-terminus of
ORF2p remain retrotransposition-competent, the strategy de-
scribed here allows a way to both directly study the expression of
ORF1p and ORF2p from a bicistronic transcript and establishes
an experimental platform to determine how each protein interacts
with L1 RNA. Furthermore, this strategy now allows a
comprehensive means to assess how mutations in ORF1p and/
or ORF2p affect L1 RNP biogenesis and/or L1 retrotransposition.
Biochemical methods allowed us to assess how mutations in the
L1-encoded proteins affect RNP function. For example, in
agreement with previous studies, a mutation in the carboxyl-
terminal domain of ORF1p (pAD105, RR261–262AA) markedly
reduced ORF1p levels in RNPs, but did not noticeably affect
ORF2p accumulation or LEAP activity (Figure 6A; Figure S3)
[16,17]. Similarly, we could detect ORF2p and LEAP activity in
RNPs derived from cells transfected with a construct that lacks
ORF1 (Figure 6A and 6B). Thus, we conclude that ORF2p can
preferentially associate in cis with its encoding transcript to form an
RNP independently of ORF1p RNA binding (Figure 7).
Our studies further suggest that an interplay exists between
ORF1p, ORF2p and L1 RNA that is critical for proper L1 RNP
formation/function (Figure 7). For example, mutations in the
putative leucine zipper (pADLZC; L93,100,107,114V) or RRM
(pAD113; NLR157–159ALA) domains led to a reduced amount of
ORF2p in the RNP fraction, as well as a decrease in LEAP
activity. The L93,100,107,114V mutations reside in the N-terminal
coiled-coil domain of ORF1p and could potentially alter the
structure of the protein. Similarly, the NLR157–159ALA mutations
reside near coiled-coil domain/RRM junction and structural
studies indicate that a hydrogen bond between N157and D252is
important for correct folding of the RRM domain . Thus,
both of the above mutations may adversely affect the structural
integrity of ORF1p, leading to the destabilization of the resultant
L1 RNPs. Indeed, such a scenario could potentially account for
the reduced levels of ORF2p in RNPs and/or L1 RT activity in
these mutants (Figure 7). It is unlikely that the L93,100,107,114V,
L93,100,107,114V/RR261–262AA, and NLR157–159ALA mutants sig-
nificantly affect ORF2p translation, since our preliminary data
indicate that each mutant can serve as a ‘‘driver’’ in a genetic-
based trans-complementation assay (Doucet, Hulme et al., prelim-
As expected, a mutation in the endonuclease domain of ORF2p
(pAD136; H230A) had no discernable affect on the ability of
ORF1p to accumulate in RNPs when compared to a wild-type
control construct. However, this mutation consistently led to a
slightly reduced amount of ORF2p in RNPs, which correlated
with a lower LEAP activity . These findings could potentially
explain why the H230A mutant consistently exhibited lower levels
of endonuclease-independent L1 retrotransposition in Chinese
Hamster Ovary cells that are deficient in the non-homologous
end-joining pathway of DNA repair when compared to a D205A
endonuclease domain mutation .
SWWDS) did not have a major effect on the ability of ORF1p to
accumulate in the RNP fraction. However, these mutations led to a
reduced amount of ORF2p in the RNP fraction and a concomitant
decrease in LEAP activity when compared to a wild-type control
construct. How mutations in the C-domain affect ORF2p accumu-
lation in RNPs requires further study; however, it is possible that
these mutations alter the ability of ORF2p to interact with L1 RNA
and/or host factors that are important for the biogenesis of L1 RNPs
A second assay allowed us to determine how mutations in the
L1-encoded proteins affect L1 protein expression and cytoplasmic
foci formation shortly after transfection. First, we observed that
ORF1p and ORF2p can be detected when transiently expressed
from the wild-type and mutant L1 constructs used in the study. We
next measured the ability of these proteins to form cytoplasmic
foci. Consistentwith previous
defective L1s containing mutations in either the RRM (pAD113;
NLR157–159ALA, pAD102; REKG235–238AAAA) or carboxyl-
terminal domain of ORF1p (pAD105, RR261–262AA; pAD116,
YPAKLS282–287AAAALA) reduced L1 cytoplasmic foci formation
[31,42,59]. In contrast, mutations in the putative leucine zipper
domain (pADLZC; L93,100,107,114V) or mutations analogous to
those that adversely affect the nucleic acid chaperone activity of
mouse ORF1p (pAD106; RR261–262KK), which are not predicted
to inhibit L1 RNA binding, or mutations in ORF2p had little
effect on L1 cytoplasmic foci formation [16,35]. Thus, unlike our
biochemical assays, the L1 cytoplasmic foci formation assay does
not allow us to readily assess ORF2p function. Instead, it provides
a valuable tool to screen for ORF1p mutations that affect RNA
binding or perhaps protein stability (Figure 7).
Consistent with previous studies, we found that L1 cytoplasmic
foci are in close association with proteins that are components of
stress granules (Figure 7) [31,59]. Interestingly, recent studies have
shown an important role for another cytoplasmic structure (P-
Figure 5. Cellular identification of L1 cytoplasmic foci. A. Cellular localization of the L1-encoded proteins: Immunofluorescence was conducted
on pAD2TE1 transfected U-2 OS cells 48 hours post-transfection. T7-tagged ORF1p (green; left column) and TAP-tagged ORF2p (red; middle column)
staining are shown for representative transfected cells. A merged image is shown in the rightmost column; DAPI (grey) was used to stain nuclear
DNA. Cartoons of the constructs are indicated at the left of the micrographs. The blue rectangle in the constructs indicates the mneoI cassette. B.
Cellular localization of L1-encoded proteins and RNA: Immunofluorescence/RNA FISH was conducted on pAD3TE1 transfected U-2 OS cells 48 hours
post-transfection. T7-tagged ORF1p (green), TAP-tagged ORF2p (blue), L1 RNA (red), and DAPI (turquoise) staining are indicated in left four
micrographs. A merged image is shown in the rightmost panel. The cartoon of pAD3TE1 is shown above the micrographs. C. L1 cytoplasmic foci are
associated with stress granules: Immunofluorescence/fluorescence microscopy was performed on U-2 OS cells co-transfected with pAD2TE1 and one
of the following plasmids: 1) pAgo2-GFP (green staining, top row of images); 2) pDCP1a-GFP (green staining, second row of images); 3) pG3BP-GFP
(green staining, third row of images). T7-tagged ORF1p (red), and TAP-tagged ORF2p (blue) also are shown. A merged image is shown in the
rightmost panels; DAPI (grey) was used to stain nuclear DNA. Immunofluorescence also was performed on U-2 OS cells transfected with pAD3TE1.
Images using antibodies against the endogenous stress granule components eIF3 (aeIF3 (green)) and G3BP (aG3BP (green)) are shown. L1 RNA (red),
and ORF2p (blue) also are indicated. Arrows indicate the association of L1 cytoplasmic foci (white) and stress granules (yellow). A merged image is
shown in the rightmost columns; DAPI (grey) was used to stain nuclear DNA. All L1 constructs in panel B and C contain the mneoI retrotransposition
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org 11October 2010 | Volume 6 | Issue 10 | e1001150
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org12 October 2010 | Volume 6 | Issue 10 | e1001150
bodies) for Ty3 and Ty1 retrotransposition in yeast [60–62].
Whether L1 cytoplasmic foci play an important role in L1
retrotransposition awaits further experimentation.
In sum, we have developed a powerful system to physically
detect the proteins and RNA encoded by both retrotransposition-
competent and mutant L1 constructs in RNP complexes, which
now augments previous studies that were based on inferring the
presence of ORF2p from its enzymatic activity. It is noteworthy
that RNPs derived from the RC-L1s characterized in this study
exhibit the biochemical properties predicted of a ‘‘basal’’ L1
retrotransposition intermediate. Thus, we speculate that at least
some of the L1 cytoplasmic foci identified here could serve as bona
fide L1 retrotransposition intermediates. Finally, we predict that
the use of the L1 expression constructs developed here will allow a
powerful means to identify host factors that play a role in L1
retrotransposition and predict that adaptations of this system will
prove useful in identifying RNPs encoded by other non-LTR
Materials and Methods
Sequences of the oligonucleotides used in this study that have
been published previously or are available upon request.
39RACE adapter: 59- GCGAGCACAGAATTAATACGACT-
39RACE outer: 59-GCGAGCACAGAATTAATACGACT-39
GAPDH 39 end: 59-GACCCTCACTGCTGGGGAGTCC-39
Neo Promoter Sens (NPS): 59-GGTTGCTGACTAATTGA-
L1 39end: 59-GGGTTCGAAATCGATAAGCTTGGATCCA-
The following plasmids are based on the previously described
pJM101/L1.3 and pDK101 constructs [4,16]. The amino acid and
nucleotide numbers indicate the mutation position based on L1.3
accession number L19088 . The constructs were cloned into the
pCEP4 expression vector (Invitrogen) and contain the mneoI indicator
cassette [32,47] in the L1 39UTR unless otherwise indicated. PCR
sequences onto the 39 end of ORF2. As a result of this procedure, we
used in our cloning strategies are available upon request.
pADO2Tt contains a Tandem Affinity Purification epitope tag
(TAP tag)  on ORF2p and was cloned from the pZome-1-C
pAD2TE1 is derived from pDK101 (L1.3)  and contains
both the T7 gene 10 epitope tag on the carboxyl-terminus of
ORF1p and a TAP tag on the carboxyl-terminus of ORF2p.
pAD2TE1-D2 is derived from pAD2TE1, but lacks CMV
promoter and SV40 polyadenylation signal present in the original
pAD2TE1-NT is identical to pAD2TE1, but lacks the mneoI
pES2TE1 is identical to pAD2TE1, but contains a tandem
affinity FLAG-HA tag on the carboxyl-terminus ORF2p .
pAD500 is derived from L1.3DORF1NN , and contains a
TAP tag on the carboxyl-terminus of ORF2p.
pADL1MT is derived from pJM101/L1.3 and contains 24
repeats of the MS2 stem-loop (MS2 tag) upstream of the mneoI
indicator cassette in the L1 39UTR. The MS2 repeats were
subcloned from the pTRIP vector .
pAD3TE1 is identical to pAD2TE1, but contains the MS2 tag
in the 39UTR (at the same position as in pADL1MT).
pADO1S is identical to pAD2TE1, but contains three stop
codons in ORF1. The first two stop codons (R7Stop; K8Stop) were
generated by introducing a thymidine at nucleotide position 928 to
create a frameshift mutation and by mutating an A to a T at
nucleotide position 930. The third stop codon is from the construct
pJM108/L1.3 carrying the mutation S119Stop [19,32].
pADLZC is identical to pAD2TE1, but contains four leucine to
valine mutations (L93,100,107,114V) in the ORF1p putative leucine
pAD102 is identical to pAD2TE1, but contains the REKG235–238
AAAA mutations in the ORF1p RRM domain [16,32].
pAD105 is identical to pAD2TE1, but contains the RR261–262AA
mutations in the ORF1p C-terminal domain [16,19,32].
pAD106 is identical to pAD2TE1, but contains the RR261–262KK
mutations in the ORF1p C-terminal domain .
pAD107 is identical to pAD2TE1, but contains the RR261–262KR
mutation in the ORF1p C-terminal domain .
pAD113 is identical to pAD2TE1, but contains the NLR157–159
ALA mutations in the ORF1p RRM domain .
isidentical to pAD2TE1,
YPAKLS282–287AAAALA substitution in the ORF1p C-terminal
pAD135 is identical to pAD2TE1, but contains the D702A
mutation in the putative ORF2p RT active site .
pAD136 is identical to pAD2TE1, but contains the H230A
mutation in the ORF2p EN domain .
CWWDC1143–1147SWWDS mutations in the ORF2p C-domain
pADL/R is identical to pAD2TE1, but contains a putative
leucine zipper domain as well as a C-terminal domain mutant
(L93,100,107,114V; RR261–262AA) in ORF1p.
Figure 6. L1 cytoplasmic foci formation requires the nucleic acid binding domain of ORF1p. A. L1 cytoplasmic foci formation requires the
nucleic acid binding domain of ORF1p: Immunofluorescence was performed on U-2 OS cells transfected with the indicated pAD2TE1-derived mutant
plasmids (described in Figure 4A). T7-tagged ORF1p (green; top panels) and TAP-tagged ORF2p (red; middle panels) staining are shown for
representative transfected cells. A merged image is shown in the bottom panels; DAPI (grey) was used to stain nuclear DNA. B. Quantitative analyses
of L1 cytoplasmic foci formation: The number of U-2 OS transfected cells that contains L1 cytoplasmic foci were quantified. The name of the construct
used for each transfection is indicated on the X-axis (described in Figure 4A). The percentage of transfected cells displaying L1 cytoplasmic foci is
indicated on the Y-axis. pAD2TE1 serves as a positive control. The average of four independent experiments is indicated; error bars = standard
deviation of the mean. C. ORF1p is necessary and sufficient for L1 cytoplasmic foci formation: Immunofluorescence was performed on U-2 OS cells
transfected with pDK500 and pAD500. A cartoon of the constructs is shown at the left of the micrographs. T7-tagged ORF1p (green; left column) and
TAP-tagged ORF2p (red; middle column) staining are shown for representative transfected cells. A merged image is shown in the rightmost column;
DAPI (grey) was used to stain nuclear DNA. The graph indicates the percentage of cells exhibiting L1 cytoplasmic foci. The name of the construct is
indicated on the X-axis. The percentage of transfected cells displaying L1 cytoplasmic foci is indicated on the Y-axis. pAD2TE1 serves as a positive
control. Four independent analyses of 100 transfected cells were analyzed for each construct. Error bars = standard deviation of the mean. All L1
constructs contain the mneoI retrotransposition indicator cassette.
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org 13October 2010 | Volume 6 | Issue 10 | e1001150
Figure 7. A working model of L1 cytoplasmic RNP formation. A hypothetical model based on our data that builds on previous models of L1
retrotransposition by target-site primed reverse transcription (TPRT; recently reviewed in [6,8]). ORF1p (green oval), ORF2p (red oval), and L1 RNA
(waved blue line) associate with their encoding mRNA via cis-preference to form a ‘‘basal’’ retrotransposition complex (right side, pAD2TE1).
Mutations in ORF1p and/or ORF2p functional domains have different affects on L1 RNP formation and/or function (thin gray arrows). Mutations in the
ORF1p RNA recognition motif (pAD113) disrupt L1 cytoplasmic foci formation and lead to a severe reduction of ORF1p and ORF2p in cytoplasmic RNP
complexes (top left side). In some mutants (pAD105 and pAD500) ORF2p can still associate with L1 RNA in the absence of ORF1p RNA binding
(bottom left side). Mutations in the putative ORF1p leucine zipper domain (pADLZC) lead to a reduction in ORF1p and ORF2p in RNPs (top center; the
reduction in ORF2p is indicated by the striped red oval). Mutations in the ORF2p cysteine-rich domain (pAD162) still allow L1 cytoplasmic foci
formation, but adversely affect ORF2p accumulation in RNPs (bottom center). Mutations that disrupt ORF1p nucleic acid chaperone activity (pAD106)
or mutations in either the ORF2p endonuclease (pAD136) or reverse transcriptase (pAD135) domains form cytoplasmic RNPs containing ORF1p,
ORF2p, and L1 RNA (right side of figure). These mutations probably adversely affect L1 retrotransposition downstream of RNP formation and/or
during TPRT. Some RNP complexes localize to L1 cytoplasmic foci and frequently are found in association with stress granules (bold gray arrows).
However, whether these foci play a role in L1 retrotransposition remains unknown (indicated by the dotted line and question mark).
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org14 October 2010 | Volume 6 | Issue 10 | e1001150
pADL/C is identical to pAD2TE1, but contains a putative
leucine zipper domain mutation (L93,100,107,114V) in ORF1p as
well as a C-domain mutation (CWWDC1143–1147SWWDS) in
LZC is derived from pDK101 and contains four leucine to
valine mutations (L93,100,107,114V) in the ORF1p putative leucine
LZ1/2 is derived from pDK101 and contains two leucine to
valine mutations (L93,100V) in the ORF1p putative leucine zipper
LZ2/3 is derived from pDK101 and contains two leucine to
valine mutations (L100,107V) in the ORF1p putative leucine zipper
LZ3/4 is derived from pDK101 and contains two leucine to
valine mutations (L107,114V) in the ORF1p putative leucine zipper
pDK108, pDK116, pDK135, and pDK500 were described
pMS2-GFP-nls, pMS2-CFP, and pTRIP were generous
gifts from Edouard Bertrand [64–66].
pAgo2-GFP and pDCP1a-GFP were generous gifts from
Gregory Hannon .
pG3BP-GFP was a generous gift from Jamal Tazi .
Cell lines were maintained in a tissue culture incubator (37uC at
a 7% CO2 level) in high glucose Dulbecco’s modified Eagle
medium (DMEM) without pyruvate (GIBCO), supplemented with
10% fetal bovine calf serum and 1X Penicillin-Streptomycin-
Glutamine (GIBCO) as described previously .
The L1 retrotransposition assay
The cultured cell retrotransposition assay was conducted as
described previously [32,48]. Briefly, 26104HeLa cells/well were
plated in 6 well dishes. Within 24 hours, each well was transfected
with 1 mg of plasmid DNA (prepared with a Midiprep Plasmid
DNA Kit (QIAGEN)) using FuGene-6 transfection reagent
(Roche). Three days post-transfection, cells were grown in the
presence of G418 (400 mg/mL) to select for retrotransposition
events. The media was changed daily. After ,12 days of selection,
the resultant cells were washed with 1X Phosphate-Buffered Saline
(PBS), fixed, and stained with crystal violet to visualize colonies. In
parallel, HeLa cells were plated in 6 well dishes and transfected
with 0.5 mg of the same plasmids and hrGFP (Stratagene). Three
days post-transfection cells were subjected to flow cytometry and
the transfection efficiency was determined based on the number of
GFP positive cells by FACS. In some experiments, 26105HeLa
cells/well were transfected to monitor L1 retrotransposition.
Protein expression and western blot analysis
HeLa cells were transfected with a given L1 expression
construct in T-25, T-75, or T-175 tissue culture flasks. Whole
cell lysates then were prepared after 9 days of hygromycin
selection as described previously . The cells were washed in
1X PBS, scraped from plates in 1X PBS, and spun at 3,000 g for 5
minutes at 4uC. One volume of pelleted cells was lysed using two
volumes of the following buffer: 1.5 mM KCl, 2.5 mM MgCl2,
5 mM Tris-HCl, pH 7.5, 1% deoxycholic acid, 1% Triton X-100,
1X Complete Mini EDTA-free Protease Inhibitor Cocktail (Roche
Applied Science). The cells were resuspended by gentle pipetting
and incubated on ice for 10 minutes. The lysate was cleared by
centrifugation at 3,000 g for 5 minutes at 4uC. Untransfected
HeLa cell samples were obtained three days after plating. The
Bradford reagent (Bio-Rad) was used to determine the protein
concentrations . The same amount of total protein was
separated by SDS-PAGE. BenchMark Pre-Stained Protein Ladder
(Invitrogen) was used as a molecular weight marker. The proteins
were detected by western blot using the following primary
antibodies: mouse anti-T7-Tag (Novagen), rabbit anti-TAP (Open
Biosystems), rat anti-HA (3F10 clone, Roche), mouse anti-a-
tubulin (Sigma), rabbit anti-S6 (Cell Signaling Technology), rabbit
anti-ORF1p (a generous gift from Thomas Fanning ) and
rabbit a-ORF2p-N (a generous gift from John Goodier . Goat
anti-mouse, anti-rabbit and anti-rat HRP-conjugated secondary
antibodies were purchased from GE/Amersham. Western blots
were developed using either the pico or femto ECL substrate
(Pierce) according to manufacturer’s protocols.
RNA preparation and RT–PCR analysis
RNA isolation was performed with the RNeasy Kit (QIAGEN)
coupled to an on-column DNase treatment (QIAGEN). Whole cell
lysates (10–50 mL) were used as starting material. The isolated
RNAs were resuspended in Ultrapure distilled water (GIBCO) and
quantified using a Nanodrop spectrophotometer (Thermo Scien-
tific). For the LEAP assay controls, RNA was isolated from a
50 mL RNP sample (1.5 mg/mL). RT-PCR was performed on
0.5 mg total RNA, using the 39RACE adapter primer (0.4 mM)
and M-MLV reverse transcriptase (200U) (Promega). The
resultant cDNA products then were amplified by PCR using
HotStart Pfu Turbo polymerase (Stratagene) with one primer
specific to the transfected L1 constructs (L1 39 end) or GAPDH
(GAPDH 39 end) and the 39RACE outer primer, as described
previously . The PCR cycles were as follows: one cycle at 94uC
for 3 minutes, then thirty five cycles of 94uC for 30 seconds, 58uC
for 30 seconds and 72uC for 30 seconds. Then, a final extension
was performed at 72uC for 10 minutes.
The LEAP assay has been described previously . Briefly,
HeLa cells were plated at 66106cells/flask in T-175 flasks, and
transfected within 24 hours with 30 mg plasmid DNA (Midiprep
Plasmid DNA Kit (QIAGEN)) using FuGene-6 transfection
reagent (Roche). HeLa cells were grown in the presence of
hygromycin from days 3 to 9 post-transfection (200 mg/mL) to
select for episome-containing cells. HeLa cells grown for three
days in the absence of hygromycin served as an untransfected
(naı ¨ve) control. On day 9, transfected cells and naı ¨ve HeLa cells
were harvested, lysed, and the cleared whole cell lysates were
centrifuged through an 8.5%/17% (w/v) sucrose cushion at
178,000 g for 2 hours. The resultant pellet was resuspended with
100 mL dH2O +1X Complete EDTA-free protease inhibitor
cocktail (Roche). Bradford reagent (Bio-Rad) was used to
determine protein concentration and this RNP sample was diluted
to a final concentration of 1.5 mg/mL. An aliquot (1.5 mg) of the
RNP sample was added to 49 mL of LEAP assay master mix
(50 mM Tris-HCL (pH=7.5), 50 mM KCl, 5 mM MgCl2,
10 mM DTT, 0.4 mM 39RACE adapter primer, 20U RNasin
(Promega), 0.2 mM dNTPs, and 0.05% (v/v) Tween 20) and was
incubated at 37uC for 1 hour. LEAP cDNA products (1 mL) were
amplified in a standard 50 mL PCR reaction containing 0.4 mM of
the 39RACE outer primer and 0.4 mM of one of the following
forward primers: L1 39 end; Neo promoter sense; Neo8161S;
LEAP-86; LEAP-46, using HotStart Pfu Turbo polymerase
(Stratagene) according to the manufacturer’s protocol (see Figure
S2). The resultant products were visualized on 2% agarose gels.
PCR products were isolated, cloned into the pCR-Blunt vector
(Invitrogen), and sequenced to confirm their identity. The diffuse
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org15October 2010 | Volume 6 | Issue 10 | e1001150
profile of the amplification above 220 bp is explained by initiation
of reverse transcription at many places on L1 poly (A). Lower
bands, below 200 bp, are due to an internal initiation of reverse
transcription 59 of the poly (A) tail .
The affinity purification procedure described in Figure 3 was
adapted from a published protocol . To prepare the samples,
HeLa cells were plated in T-175 flasks and transfected as described
in the previous paragraph. Hygromycin selection on days 3 to 9
post-transfection was used to select for cells expressing the
respective constructs. Using these conditions, one T-175 flask
per plasmid was sufficient to yield enough cellular material (3 mg)
for an experiment. Cells were washed, scraped from the flasks in
1X PBS, and centrifuged at 3,000 g for 5 minutes at 4uC. Cells
were lysed by repeated pipetting with 3 volumes of IP FLAG
buffer (0.1% NP-40, 100 mM KCl, 20 mM Tris-HCl pH 8,
1 mM DTT, 10% Glycerol, 1X complete EDTA free Protease
inhibitor (Roche)) and incubated for 15 minutes on ice. The
cellular debris was removed by centrifugation at 3,000 g for 5
minutes at 4uC. The protein concentration of the supernatant was
quantified by a Bradford assay (Biorad Protein Assay).
For immunoprecipitation reactions, anti-FLAG beads (EZview
Red ANTI-FLAG M2 Affinity Gel, Sigma) were equilibrated in
0.1M Glycine (pH 2.2) (5 mL for 100 mL of beads) for 5 minutes at
room temperature. After addition of Tris-HCl (pH 8.0) (10 mL for
100 mL of beads), the beads were spun down for 3 minutes at 3000
rpm and then washed 3 times with IP FLAG buffer (mentioned
above). For each condition, 3 mg of protein extract (input) was
then incubated on a rotating wheel overnight at 4uC with 20 mL of
the pre-equilibrated anti-FLAG beads. The next day, the beads
were washed 5 times with 1 mL of IP FLAG Buffer for 10 minutes
at 4uC. The beads were incubated 1 hour (at 4uC on the wheel)
with 200 mL of IP FLAG buffer containing 200 mg/mL of 3X
FLAG peptide (Sigma). The elution fraction then was collected
and analyzed alongside the corresponding input fraction by
western blotting (as described above in the dedicated section). The
femto ECL substrate (Pierce) was used in the detection of both T7-
tagged ORF1p and FLAG-HA-tagged ORF2p in this experiment.
An aliquot (1 mL) of the input and elution samples then were used
to perform the LEAP assay (see previous section for detailed
Quantitative real-time PCR
Quantitative PCR was performed on LEAP cDNA samples or
M-MLV RT-PCR products using the 7300 Real Time PCR
system (Applied Biosystems). For analysis, 1 mL of LEAP or M-
MLV RT products was added to 19 mL of master mix (1X SYBR
Green PCR Master Mix (Applied Biosystems), 500 nM L1 39 end
primer, and 500 nM L1 Reverse primer), and amplified in a
standard Q-PCR run of 45 cycles. The average cycle threshold
(Ct) value for each experimental or control sample was calculated
from three independent reactions within a Q-PCR run. The
‘absolute quantitation by standard curve’ method was used to
determine the number of cDNA molecules in each LEAP RNP or
RNA sample. A standard curve was generated using dilutions of a
L1 LEAP product cloned into a plasmid, and a best fit line
(log(molecules) versus average Ct value) for these standards was
generated by linear regression. For each wild-type or mutant L1,
RNA levels from three independent RNP samples were examined
by at least one RT reaction and two Q-PCR runs. The level of
LEAP activity in each wild-type or mutant L1 was determined
from four independent RNP samples. These RNP samples were
characterized by at least one and up to three independent LEAP
RT reactions and one or two independent Q-PCR runs. For
LEAP activity, the negative control RT- (pDK135) gave a
background amplification level of ,15–30 molecules of cDNA
due to the presence of the transfected L1 plasmid in the RNP
sample. This RT- background control was included in each Q-
PCR run and the background amount of molecules was subtracted
from each experimental sample in Figure S2 and when calculating
Fluorescent In Situ Hybridization (FISH) and
U-2 OS cells were plated at 105onto sterile glass cover slips in 6
well tissue culture dishes. The following day, cells were transfected
using 1 mg of purified plasmid DNA (Midiprep Plasmid DNA Kit,
QIAGEN) and 3 mL of FuGene-6 Transfection Reagent (Roche
Applied Science). The FISH protocol was adapted from the
Robert Singer (Albert Einstein College of Medicine, New York)
lab protocol (available at http://www.singerlab.org/protocols) and
was modified to allow protein detection by immunofluorescence.
Briefly, 48 h post-transfection, cells were washed twice with 1X
PBS and fixed with 4% paraformaldehyde in 1X PBS for 10
minutes at room temperature. The fixed cells then were washed 2
additional times with 1X PBS. The fixed cells were permeabilized
by treatment with 70% ethanol overnight at 4uC. The following
day, cells were rehydrated with 1X saline-sodium citrate (SSC) and
10% formamide for 5 minutes at room temperature. To prepare
the hybridization solution, a first mix containing 40 mg of E.coli
tRNA (Sigma), 1X SSC, 10% formamide, and 7.5 ng of MS2-Cy3
probe (generous gift from Dr. Edouard Bertrand) was boiled for 1
minute at 100uC in order to denaturize the probe. The quantities
of probe and tRNA are indicated for hybridization of one slide. A
second mix was prepared with 10% dextran sulfate, 2 mM
vanadyl-ribonucleoside complex (Sigma), and 0.02% RNase free
BSA (Roche Applied Science). After probe denaturation, mixes 1
and 2 were combined to form the final hybridization solution. The
re-hydrated cells were hybridized overnight at 37uC in 30 mL of
this hybridization solution. Cells were then washed twice for 30
minutes at 37uC with 1X SSC, 10% formamide and 3% BSA and
then were incubated with primary antibodies for 1 hour at 37uC.
The cells were washed three times with 1X PBS and were
incubated with secondary antibodies and 0.2 mg/mL 49,69-
diamidino-2-phenylindole (DAPI, Molecular Probes) for 30
minutes at 37uC and washed three times with 1X PBS. The
primary and secondary antibodies were diluted in 1X PBS and 3%
BSA and are as follows: anti-T7 (Novagen), anti-TAP (Open
Biosystems), anti-HA (Roche), rabbit anti-ORF1p (a generous gift
from Thomas Fanning) ) and rabbit a-ORF2p-N (a generous
gift from John Goodier) , anti-eIF3 (Santa Cruz BioTechnol-
ogy), anti-G3BP (generous gift from Jamal Tazi), Alexa Fluor 488
anti-mouse and anti-rabbit (Invitrogen), Alexa Fluor 546 anti-
mouse and anti-rabbit (Invitrogen), Cy3-conjugated anti-rat
(Jackson Immuno Research) and Cy5-conjugated anti-mouse
and anti-rabbit (Jackson Immuno Research). Cells were rinsed
with water and mounted on slides with Vectashield (Vector
Laboratories). Samples were then analyzed with appropriate
fluorescent filters on DMRXA Leica microscope and images were
captured using a Zeiss LSM510 META confocal microscope.
The above protocol was used for both RNA and protein
detection analyses. In experiments where we only sought to detect
RNA, the protocol was stopped after hybridization with the MS2-
Cy3 probe and subsequent washes. Cover slides were stained with
DAPI and mounted on slides as described above. In experiments
where we only sought to detect protein, fixed cells were
permeabilized by treatment with anhydrous methanol for 1
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org 16October 2010 | Volume 6 | Issue 10 | e1001150
minute. After three washes with 1X PBS, the cells were incubated
with 3% BSA in 1X PBS for 30 minutes. Antibody incubation and
DAPI staining were performed as described above. We verified
that the protein A domain contained in the TAP tag of ORF2p did
not react with the secondary antibodies (data not shown).
Analysis of L1 cytoplasmic foci
In general, L1 cytoplasmic foci formation was measured
48 hours post-transfection. At least two independent series of
slides were analyzed. Each analysis corresponds to 100 transfected
cells that were quantified in a blinded manner. Cells in which we
were able to distinguish a concentrated cytoplasmic signal from a
diffuse cytoplasmic signal using a 63x or 100x objective (equivalent
of 1 micrometer of diameter) were considered as L1 cytoplasmic
A. Retrotransposition assays with mutant L1 constructs: 26104
HeLa cells were transfected with the indicated constructs.
pJM101/L1.3 and pAD2TE1 were used as positive controls. All
of the pAD-based constructs contain the ORF1p T7 epitope tag
and the ORF2p TAP-tag except for pAD500, which lacks ORF1.
pAD135 is an RT mutant (D702A), and serves as a negative
control. B. Retrotransposition assay with leucine zipper domain
mutants: 26105 HeLa cells were transfected with the indicated
pDK101-derived constructs. T7WT (pDK101) is a wild-type L1
(L1.3) that contains the T7 epitope tag on the carboxyl terminus of
ORF1p. pDK101 was modified to create LZ1/2 (L93V, L100V),
LZ2/3 (L100V, L107V), LZ3/4 (L107V, L114V), and LZC
(L93V, L100V, L107V, L114V). Each of the mutations abolished
Found at: doi:10.1371/journal.pgen.1001150.s001 (0.57 MB TIF)
Retrotransposition assays with mutant L1 constructs.
Results of western blot analyses: RNPs derived from wild-type
(pDK101) and the indicated mutant constructs were subjected to
western blot analyses with an anti-T7 antibody (aT7). An
,40 kDa band indicative of epitope-tagged ORF1p is shown.
Untransfected HeLa cells served as a negative control. The
ribosomal S6 protein was detected using an anti-S6 (aS6) antibody
(bottom panel; RNP prep control) and served as a loading control.
Molecular weight markers (Invitrogen) are indicated at the left of
the gel. B. Results of LEAP assays: Top panel: An aliquot of the
above RNPs was used to measure LEAP activity. RNPs derived
from wild-type (pDK101) generate strong LEAP products of
,220–400 bp and served as a positive control. Untransfected
HeLa cells and an RT mutant (pDK135; D702A) serve as negative
controls. LEAP products generated in ORF1p mutant RNPs are
shown. Reactions conducted without template (No Template) or
without RNPs (No RNP/No RNA) were used as negative controls.
Middle and bottom panels: RT-PCR with M-MLV RT and
primers specific to either the transfected L1 constructs or GAPDH
confirmed the presence of L1 RNA in RNPs and the integrity of
the RNA isolation procedure. DNA size markers (Invitrogen) are
indicated at the left of the gel. C. LEAP products derived from the
ORF1p RNA binding mutant (RR261-262AA) and putative
chaperone mutant (RR261-262KK): Representative LEAP prod-
ucts derived from wild-type (pDK101), an ORF1p RNA binding
mutant (pDK105; RR261-262AA), and a putative ORF1p nucleic
acid chaperone activity mutant (pDK106; RR261-262KK) are
depicted at the top gel. The middle and bottom gels are RT-PCR
reactions conducted with M-MLV RT and primers specific to L1
and GAPDH transcripts, respectively. The black arrow on the
The effect of ORF1p mutations on LEAP activity. A.
middle gel indicates the size of the specific L1 cDNA amplification
products. Untransfected HeLa cells and a RT mutant (pDK135)
served as negative controls. Additional negative controls include
reactions conducted without template (No Template) as well as
reactions conducted without RNPs or RNA (No RNP/No RNA).
DNA size markers (Invitrogen) are indicated at the left of the gel.
D. LZC RNPs have decreased reverse transcriptase activity: Top
panel: RNPs derived from T7WT (pDK101) generate strong
LEAP products of ,220–400 bp. By comparison, LZC (see Figure
S1B) had a less intense band at ,220–400 bp. LEAP products also
were seen in the RR261-262AA (pDK105) mutant and the LZC/
RR261-262AA mutants. No product was seen in untransfected
HeLa cells or for a L1 containing a RT active site mutation
(pDK135; RT-). Middle panels: RT-PCR with M-MLV RT and
primers specific to either the transfected L1 constructs or GAPDH
confirmed the presence of L1 RNA in RNPs and the integrity of
the RNA isolation procedure. No RNP/RT and dH2O served as
negative controls. DNA size markers (Invitrogen) are shown at the
left side of the gel. Bottom panels: Western blot against the T7
epitope tag detects ORF1p (aT7). Untransfected HeLa cells served
as a negative control. Western blot against ribosomal protein S6
was used as a loading control (aS6). Molecular weight markers
(Invitrogen) are indicated at the left of the blot. E. Quantitative
PCR of LEAP cDNAs from ORF1p LZC and carboxyl-terminal
nucleic acid binding domain mutants: Representative results of a
Q-PCR run are shown. Standard deviations are indicated on the
graph. Q-PCR was performed on four independent RNP preps for
T7WT (pDK101), LZC, RR261-262AA (pDK105), and the LZC/
RR261-262AA double mutant. Three of four preps showed a 5–7
fold decrease in LZC LEAP activity compared to wild-type. One
RNP prep showed a 16–23 fold decrease in LZC LEAP activity
compared to wild-type. F. LZC mutation does not appear to affect
reverse transcriptase elongation: Top panel: PCR was performed
on LEAP cDNAs using different primers pairs. The names of the
primers and the approximate size of each product are indicated in
the cartoon above the gel. LZC RNPs yield fewer products when
compared to T7WT (pDK101) RNPs with all tested primer sets.
Bottom panel: RT-PCR with M-MLV RT confirms the presence
of L1 RNA in the RNP samples. DNA size markers (Invitrogen)
are indicated at the left of the gel. The L1 constructs in all panels
contain the mneoI retrotransposition indicator cassette.
Found at: doi:10.1371/journal.pgen.1001150.s002 (1.42 MB TIF)
formation in U-2 OS cells. A. L1 retrotransposition assays:
26104 cells were transfected with the indicated L1 constructs.
pJM101/L1.3 and pAD2TE1 were used as positive controls.
Untransfected cells and pAD135 (RT mutant (D702A)) serve as
negative controls. A cartoon of each L1 is shown above the tissue
culture dishes. Green rectangle = ORF1; Red rectangle =
ORF2. The relative positions of the T7 and TAP tags also are
indicated. All constructs contain the mneoI retrotransposition
indicator cassette. B. Time course analyses of L1 cytoplasmic foci
formation: Cells were transfected with pAD2TE1. X-axis = time
after transfection. Y-axis= percentage of transfected cells
containing L1 cytoplasmic foci. For each time point, 100
transfected cells were analyzed for the presence of ORF1p and
ORF2p in L1 cytoplasmic foci. Error bars = standard deviation
(n=3). C. Cytoplasmic localization of L1 proteins and RNA: Top
panels: Cells were transfected with pAD3TE1. ORF2p was
visualized with an anti-ORF2 antibody (aORF2, green). ORF1p
was visualized with an anti-T7 antibody (aT7, blue). L1 RNA was
visualized with an MS2-Cy3 FISH probe (red). A merged image is
shown in the rightmost column; DAPI (grey) was used to stain
nuclear DNA. Middle panels: Cells were transfected with
L1 retrotransposition and L1 cytoplasmic foci
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org17 October 2010 | Volume 6 | Issue 10 | e1001150
pES2TE1. ORF2p was visualized with an anti-ORF2 antibody
(aORF2, green) and an anti-HA antibody (aHA, red). ORF1p was
visualized with an anti-T7 antibody (aT7, blue). A merged image
is shown in the rightmost column; DAPI (grey) was used to stain
nuclear DNA. Bottom panels: Cells were transfected with
pES2TE1. ORF1p was visualized with an anti-ORF1 antibody
(aORF1, green) and an anti-T7 antibody (aT7, blue). ORF2p was
visualized with an anti-HA antibody (red). A merged image is
shown in the rightmost column; DAPI (grey) was used to stain
nuclear DNA. D. Localization of L1 RNA: Top panels: Cells were
transfected with pADL1MT. L1 RNA was visualized with an
MS2-Cy3 FISH probe (MS2-Cy3, red). DAPI was used to stain
nuclear DNA (gray). A merged image is shown in the rightmost
column. Middle panels: Cells were co-transfected with pAD3TE1
and pMS2-GFP-nls. Fluorescence was used to visualize MS2-GFP-
nls (green). L1 RNA was visualized with an MS2-Cy3 FISH probe
(MS2-Cy3, red). ORF1p was visualized with an anti-T7 antibody
(aT7, blue). A merged image is shown in the rightmost column.
Bottom panels: Cells were co-transfected with pAD3TE1 and
pMS2-CFP. Fluorescence was used to visualize MS2-CFP (green).
L1 RNA was visualized with an MS2-Cy3 FISH probe (MS2-Cy3,
red). ORF2p was visualized with an anti-TAP antibody (aTAP,
blue). A merged image is shown in the rightmost column. The use
of MS2 binding protein system confirms the cytoplasmic
localization of the L1 RNA observed by FISH as well as the co-
localization of L1 RNA with ORF1p and ORF2p. The L1
constructs in all panels contain the mneoI retrotransposition
Found at: doi:10.1371/journal.pgen.1001150.s003 (2.01 MB TIF)
We thank Christine Beck, Jose ´ Luis Garcia Perez, Billy Giblin, Sandra
Richardson, Se ´verine Chambeyron, Alain Pe ´lisson, and Oliver Weichen-
rieder for helpful comments during the course of this study. We thank
Edouard Bertrand, Thomas Fanning, John Goodier, Greg Hannon, and
Jamal Tazi for generously providing reagents. We thank the IGH and
University of Michigan Sequencing Cores for help with DNA sequencing.
We are grateful to Eugenia Basyuk (FISH), Christelle Cayrou, Bijan
Sobhian (purification protocols), and Se ´verine Chambeyron (fluorescence
analysis) for assistance in designing/implementing experiments. We thank
Nicole Lautredou (Montpellier RIO Imaging) and Chris Edwards
(Microscopy and Image-analysis Laboratory at the University of Michigan,
Department of Cell and Developmental Biology) for assistance with
Conceived and designed the experiments: AJD AEH ES DAK JBM HCK
JNA JVM NG. Performed the experiments: AJD AEH ES DAK JBM
HCK JNA MH NG. Analyzed the data: AJD AEH ES DAK JBM HCK
JNA JVM NG. Contributed reagents/materials/analysis tools: AB JVM
NG. Wrote the paper: AJD AEH JVM NG.
1. Lander ES, Linton LM, Birren B, Nusbaum C, Zody MC, et al. (2001) Initial
sequencing and analysis of the human genome. Nature 409: 860–921.
2. Grimaldi G, Singer MF (1983) Members of the KpnI family of long interspersed
repeated sequences join and interrupt alpha-satellite in the monkey genome.
Nucleic Acids Res 11: 321–338.
3. Kazazian HH, Jr., Moran JV (1998) The impact of L1 retrotransposons on the
human genome. Nat Genet 19: 19–24.
4. Sassaman DM, Dombroski BA, Moran JV, Kimberland ML, Naas TP, et al.
(1997) Many human L1 elements are capable of retrotransposition. Nat Genet
5. Brouha B, Schustak J, Badge RM, Lutz-Prigge S, Farley AH, et al. (2003) Hot
L1s account for the bulk of retrotransposition in the human population. Proc
Natl Acad Sci U S A 100: 5280–5285.
6. Babushok DV, Kazazian HH, Jr. (2007) Progress in understanding the biology of
the human mutagen LINE-1. Hum Mutat 28: 527–539.
7. Goodier JL, Kazazian HH, Jr. (2008) Retrotransposons revisited: the restraint
and rehabilitation of parasites. Cell 135: 23–35.
8. Cordaux R, Batzer MA (2009) The impact of retrotransposons on human
genome evolution. Nat Rev Genet 10: 691–703.
9. Swergold GD (1990) Identification, characterization, and cell specificity of a
human LINE-1 promoter. Mol Cell Biol 10: 6718–6729.
10. Athanikar JN, Badge RM, Moran JV (2004) A YY1-binding site is required for
accurate human LINE-1 transcription initiation. Nucleic Acids Res 32:
11. Lavie L, Maldener E, Brouha B, Meese EU, Mayer J (2004) The human L1
promoter: variable transcription initiation sites and a major impact of upstream
flanking sequence on promoter activity. Genome Res 14: 2253–2260.
12. Scott AF, Schmeckpeper BJ, Abdelrazik M, Comey CT, O’Hara B, et al. (1987)
Origin of the human L1 elements: proposed progenitor genes deduced from a
consensus DNA sequence. Genomics 1: 113–125.
13. Dombroski BA, Mathias SL, Nanthakumar E, Scott AF, Kazazian HH, Jr. (1991)
Isolation of an active human transposable element. Science 254: 1805–1808.
14. Esnault C, Maestre J, Heidmann T (2000) Human LINE retrotransposons
generate processed pseudogenes. Nat Genet 24: 363–367.
15. Hohjoh H, Singer MF (1996) Cytoplasmic ribonucleoprotein complexes
containing human LINE-1 protein and RNA. Embo J 15: 630–639.
16. Kulpa DA, Moran JV (2005) Ribonucleoprotein particle formation is necessary
but not sufficient for LINE-1 retrotransposition. Hum Mol Genet 14:
17. Kulpa DA, Moran JV (2006) Cis-preferential LINE-1 reverse transcriptase
activity in ribonucleoprotein particles. Nat Struct Mol Biol 13: 655–660.
18. Martin SL (1991) Ribonucleoprotein particles with LINE-1 RNA in mouse
embryonal carcinoma cells. Mol Cell Biol 11: 4804–4807.
19. Wei W, Gilbert N, Ooi SL, Lawler JF, Ostertag EM, et al. (2001) Human L1
retrotransposition: cis preference versus trans complementation. Mol Cell Biol
20. Luan DD, Korman MH, Jakubczak JL, Eickbush TH (1993) Reverse
transcription of R2Bm RNA is primed by a nick at the chromosomal target
site: a mechanism for non-LTR retrotransposition. Cell 72: 595–605.
21. Kubo S, Seleme MC, Soifer HS, Perez JL, Moran JV, et al. (2006) L1
retrotransposition in nondividing and primary human somatic cells. Proc Natl
Acad Sci U S A 103: 8036–8041.
22. Feng Q, Moran JV, Kazazian HH, Jr., Boeke JD (1996) Human L1
retrotransposon encodes a conserved endonuclease required for retrotranspo-
sition. Cell 87: 905–916.
23. Cost GJ, Boeke JD (1998) Targeting of human retrotransposon integration is
directed by the specificity of the L1 endonuclease for regions of unusual DNA
structure. Biochemistry 37: 18081–18093.
24. Furano AV (2000) The biological properties and evolutionary dynamics of
mammalian LINE-1 retrotransposons. Prog Nucleic Acid Res Mol Biol 64:
25. Kolosha VO, Martin SL (1997) In vitro properties of the first ORF protein from
mouse LINE-1 support its role in ribonucleoprotein particle formation during
retrotransposition. Proc Natl Acad Sci U S A 94: 10155–10160.
26. Martin SL, Li J, Weisz JA (2000) Deletion analysis defines distinct functional
domains for protein-protein and nucleic acid interactions in the ORF1 protein of
mouse LINE-1. J Mol Biol 304: 11–20.
27. Moran JV, Gilbert N (2002) Mammalian LINE-1 retrotransposons and related
elements. In: Craig N, Craggie R, Gellert M, Lambowitz A, eds. Mobile DNA
II. Washington, DC: ASM Press. pp 836–869.
28. Duvernell DD, Turner BJ (1998) Swimmer 1, a new low-copy-number LINE
family in teleost genomes with sequence similarity to mammalian L1. Mol Biol
Evol 15: 1791–1793.
29. Furano AV, Duvernell DD, Boissinot S (2004) L1 (LINE-1) retrotransposon
diversity differs dramatically between mammals and fish. Trends Genet 20:
30. Khazina E, Weichenrieder O (2009) Non-LTR retrotransposons encode
noncanonical RRM domains in their first open reading frame. Proc Natl Acad
Sci U S A 106: 731–736.
31. Goodier JL, Zhang L, Vetter MR, Kazazian HH, Jr. (2007) LINE-1 ORF1
protein localizes in stress granules with other RNA-binding proteins, including
components of RNA interference RNA-induced silencing complex. Mol Cell
Biol 27: 6469–6483.
32. Moran JV, Holmes SE, Naas TP, DeBerardinis RJ, Boeke JD, et al. (1996) High
frequency retrotransposition in cultured mammalian cells. Cell 87: 917–
33. Basame S, Wai-Lun Li P, Howard G, Branciforte D, Keller D, et al. (2006)
Spatial Assembly and RNA Binding Stoichiometry of a LINE-1 Protein Essential
for Retrotransposition. J Mol Biol 357: 351–357.
34. Martin SL, Branciforte D, Keller D, Bain DL (2003) Trimeric structure for an
essential protein in L1 retrotransposition. Proc Natl Acad Sci U S A 100:
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org18October 2010 | Volume 6 | Issue 10 | e1001150
35. Martin SL, Cruceanu M, Branciforte D, Wai-Lun Li P, Kwok SC, et al. (2005) Download full-text
LINE-1 retrotransposition requires the nucleic acid chaperone activity of the
ORF1 protein. J Mol Biol 348: 549–561.
36. McMillan JP, Singer MF (1993) Translation of the human LINE-1 element,
L1Hs. Proc Natl Acad Sci U S A 90: 11533–11537.
37. Alisch RS, Garcia-Perez JL, Muotri AR, Gage FH, Moran JV (2006)
Unconventional translation of mammalian LINE-1 retrotransposons. Genes
Dev 20: 210–224.
38. Li PW, Li J, Timmerman SL, Krushel LA, Martin SL (2006) The dicistronic
RNA from the mouse LINE-1 retrotransposon contains an internal ribosome
entry site upstream of each ORF: implications for retrotransposition. Nucleic
Acids Res 34: 853–864.
39. Dmitriev SE, Andreev DE, Terenin IM, Olovnikov IA, Prassolov VS, et al.
(2007) Efficient translation initiation directed by the 900-nucleotide-long and
GC-rich 59 untranslated region of the human retrotransposon LINE-1 mRNA is
strictly cap dependent rather than internal ribosome entry site mediated. Mol
Cell Biol 27: 4685–4697.
40. Mathias SL, Scott AF, Kazazian HH, Jr., Boeke JD, Gabriel A (1991) Reverse
transcriptase encoded by a human transposable element. Science 254:
41. Fanning T, Singer M (1987) The LINE-1 DNA sequences in four mammalian
orders predict proteins that conserve homologies to retrovirus proteins. Nucleic
Acids Res 15: 2251–2260.
42. Goodier JL, Ostertag EM, Engleka KA, Seleme MC, Kazazian HH, Jr. (2004) A
potential role for the nucleolus in L1 retrotransposition. Hum Mol Genet 13:
43. Rigaut G, Shevchenko A, Rutz B, Wilm M, Mann M, et al. (1999) A generic
protein purification method for protein complex characterization and proteome
exploration. Nat Biotechnol 17: 1030–1032.
44. Nakatani Y, Ogryzko V (2003) Immunoaffinity purification of mammalian
protein complexes. Methods Enzymol 370: 430–444.
45. Fouts DE, True HL, Celander DW (1997) Functional recognition of fragmented
operator sites by R17/MS2 coat protein, a translational repressor. Nucleic Acids
Res 25: 4464–4473.
46. Valegard K, Murray JB, Stonehouse NJ, van den Worm S, Stockley PG, et al.
(1997) The three-dimensional structures of two complexes between recombinant
MS2 capsids and RNA operator fragments reveal sequence-specific protein-
RNA interactions. J Mol Biol 270: 724–738.
47. Freeman JD, Goodchild NL, Mager DL (1994) A modified indicator gene for
selection of retrotransposition events in mammalian cells. Biotechniques 17: 46,
48. Wei W, Morrish TA, Alisch RS, Moran JV (2000) A transient assay reveals that
cultured human cells can accommodate multiple LINE-1 retrotransposition
events. Anal Biochem 284: 435–438.
49. Holmes SE, Singer MF, Swergold GD (1992) Studies on p40, the leucine zipper
motif-containing protein encoded by the first open reading frame of an active
human LINE-1 transposable element. J Biol Chem 267: 19765–19768.
50. Leibold DM, Swergold GD, Singer MF, Thayer RE, Dombroski BA, et al.
(1990) Translation of LINE-1 DNA elements in vitro and in human cells. Proc
Natl Acad Sci U S A 87: 6990–6994.
51. Rangwala SH, Kazazian HH, Jr. (2009) The L1 retrotransposition assay: a
retrospective and toolkit. Methods 49: 219–226.
52. Kedersha N, Stoecklin G, Ayodele M, Yacono P, Lykke-Andersen J, et al. (2005)
Stress granules and processing bodies are dynamically linked sites of mRNP
remodeling. J Cell Biol 169: 871–884.
53. Tourriere H, Chebli K, Zekri L, Courselaud B, Blanchard JM, et al. (2003) The
RasGAP-associated endoribonuclease G3BP assembles stress granules. J Cell
Biol 160: 823–831.
54. Liu J, Valencia-Sanchez MA, Hannon GJ, Parker R (2005) MicroRNA-
dependent localization of targeted mRNAs to mammalian P-bodies. Nat Cell
Biol 7: 719–723.
55. Kedersha N, Chen S, Gilks N, Li W, Miller IJ, et al. (2002) Evidence that ternary
complex (eIF2-GTP-tRNA(i)(Met))-deficient preinitiation complexes are core
constituents of mammalian stress granules. Mol Biol Cell 13: 195–210.
56. Garcia-Perez JL, Doucet AJ, Bucheton A, Moran JV, Gilbert N (2007) Distinct
mechanisms for trans-mediated mobilization of cellular RNAs by the LINE-1
reverse transcriptase. Genome Res.
57. Ergun S, Buschmann C, Heukeshoven J, Dammann K, Schnieders F, et al.
(2004) Cell type-specific expression of LINE-1 open reading frames 1 and 2 in
fetal and adult human tissues. J Biol Chem 279: 27753–27763.
58. Morrish TA, Gilbert N, Myers JS, Vincent BJ, Stamato TD, et al. (2002) DNA
repair mediated by endonuclease-independent LINE-1 retrotransposition. Nat
Genet 31: 159–165.
59. Goodier JL, Mandal PK, Zhang L, Kazazian HH, Jr. (2010) Discrete subcellular
partitioning of human retrotransposon RNAs despite a common mechanism of
genome insertion. Hum Mol Genet 19: 1712–1725.
60. Beliakova-Bethell N, Beckham C, Giddings TH, Jr., Winey M, Parker R, et al.
(2006) Virus-like particles of the Ty3 retrotransposon assemble in association
with P-body components. RNA 12: 94–101.
61. Checkley MA, Nagashima K, Lockett SJ, Nyswaner KM, Garfinkel DJ (2010) P-
body components are required for Ty1 retrotransposition during assembly of
retrotransposition-competent virus-like particles. Mol Cell Biol 30: 382–398.
62. Dutko JA, Kenny AE, Gamache ER, Curcio MJ (2010) 59 to 39 mRNA decay
factors colocalize with Ty1 gag and human APOBEC3G and promote Ty1
retrotransposition. J Virol 84: 5052–5066.
63. Dombroski BA, Scott AF, Kazazian HH, Jr. (1993) Two additional potential
retrotransposons isolated from a human L1 subfamily that contains an active
retrotransposable element. Proc Natl Acad Sci U S A 90: 6513–6517.
64. Boireau S, Maiuri P, Basyuk E, de la Mata M, Knezevich A, et al. (2007) The
transcriptional cycle of HIV-1 in real-time and live cells. J Cell Biol 179:
65. Bertrand E, Chartrand P, Schaefer M, Shenoy SM, Singer RH, et al. (1998)
Localization of ASH1 mRNA particles in living yeast. Mol Cell 2: 437–445.
66. Fusco D, Accornero N, Lavoie B, Shenoy SM, Blanchard JM, et al. (2003) Single
mRNA molecules demonstrate probabilistic movement in living mammalian
cells. Curr Biol 13: 161–167.
67. Bradford MM (1976) A rapid and sensitive method for the quantitation of
microgram quantities of protein utilizing the principle of protein-dye binding.
Anal Biochem 72: 248–254.
68. Qi HH, Ongusaha PP, Myllyharju J, Cheng D, Pakkanen O, et al. (2008) Prolyl
4-hydroxylation regulates Argonaute 2 stability. Nature 455: 421–424.
Cytoplasmic LINE-1 RNPs
PLoS Genetics | www.plosgenetics.org 19October 2010 | Volume 6 | Issue 10 | e1001150