, 1360 (2014);
et al. Je Hyuk Lee
Highly Multiplexed Subcellular RNA Sequencing in Situ
This copy is for your personal, non-commercial use only.
clicking here.colleagues, clients, or customers by
, you can order high-quality copies for your
If you wish to distribute this article to others
The following resources related to this article are available online at
here. following the guidelines
can be obtained by
Permission to republish or repurpose articles or portions of articles
Updated information and services,
): October 24, 2014 www.sciencemag.org (this information is current as of
version of this article at:
including high-resolution figures, can be found in the online
can be found at:
Supporting Online Material
, 8 of which can be accessed free:
cites 19 articles
4 articles hosted by HighWire Press; see:
This article has been
This article appears in the following
registered trademark of AAAS.
is a Science 2014 by the American Association for the Advancement of Science; all rights reserved. The title
Copyright American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005.
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience
on October 24, 2014
on October 24, 2014
on October 24, 2014
on October 24, 2014
on October 24, 2014
or apoptosis. As such, small-molecule inhibitors
of RIPK3 may have limited therapeutic benefit
because of their potential to promote apoptotic
Nec-1 blocked necroptosis induced by TNF/
zVAD.fmk (fig. S5C), was protective in an in-
flammatory disease model (15), and does not
bition of RIPK1 rather than RIPK3 may have
therapeutic benefit. To mimic RIPK1 inhibition
inthe wholeanimal,wegenerated Ripk1 knock-
for kinase activity mutated to Asn (fig.S6, Aand
B). Whereas Ripk1−/−mice died soon after birth,
Ripk1D138N/D138Nmice were viable. Consistent
in TNF-induced necroptosis, Ripk1D138N/D138N
bone marrow–derived macrophages (BMDMs)
and E1A-immortalized MEFs were as resistant
as Ripk3−/−cells to killing by TNF/zVAD.fmk
(Fig. 4A and fig. S6C). Ripk1D138N/D138Ncells ex-
pressed normal amounts of RIPK3 and MLKL
Ripk3−/−mice in their systemic response to TNF,
exhibiting less hypothermia than did their wild-
type counterparts (Fig. 4B). Unlike RIPK1 loss,
caused by RIPK3 D161N (fig. S4). This result is
consistent with nec-1 not protecting MEFs ex-
pressing RIPK3 D161N (fig. S5B) and indicates
that the kinase activity of RIPK1 is not required
for activation of caspase-8 by RIPK3 D161N.
RIPK1 is required for TNF-induced nuclear
factor kB (NF-kB) and mitogen-activated pro-
tein kinase signaling (16, 17). Wild-type RIPK1
and RIPK1 D138N restored these signals in
Ripk1−/−cells (18), suggesting that RIPK1 has an
essential scaffold function in this setting, where-
as its kinase activity is dispensable. Indeed,
Ripk1D138N/D138Nand wild-type BMDMs were
indistinguishable in their phosphorylation of
inhibitor of NF-kB a (IkBa), c-Jun N-terminal
kinase (JNK), and extracellular signal-regulated
kinase (ERK) in response to TNF (Fig. 4C).
These results, together with the viability of
Ripk1D138N/D138Nmice, are encouraging because
they suggest that inhibiting the kinase activity
of RIPK1 has no deleterious effects, at least in
the short term. These Ripk1D138N/D138Nmice can
be used to explore the contribution of RIPK1
and necroptosis to various mouse models of
References and Notes
1. K. Moriwaki, F. K. Chan, Genes Dev. 27, 1640–1649
2. Y. S. Cho et al., Cell 137, 1112–1123 (2009).
3. S. He et al., Cell 137, 1100–1111 (2009).
4. D. W. Zhang et al., Science 325, 332–336 (2009).
5. N. Takahashi et al., Cell Death Dis. 3, e437 (2012).
6. K. Newton, X. Sun, V. M. Dixit, Mol. Cell. Biol. 24,
7. E. E. Varfolomeev et al., Immunity 9, 267–276
8. A. Oberst et al., Nature 471, 363–367 (2011).
9. W. J. Kaiser et al., Nature 471, 368–372 (2011).
10. L. Salmena et al., Genes Dev. 17, 883–895 (2003).
11. X. Sun, J. Yin, M. A. Starovasnik, W. J. Fairbrother,
V. M. Dixit, J. Biol. Chem. 277, 9505–9511 (2002).
12. Y. H. Park, M. S. Jeong, H. H. Park, S. B. Jang, Biochim.
Biophys. Acta 1834, 292–300 (2013).
13. M. Muzio et al., Cell 85, 817–827 (1996).
14. J. Seibler et al., Nucleic Acids Res. 31, e12 (2003).
15. L. Duprez et al., Immunity 35, 908–918 (2011).
16. M. A. Kelliher et al., Immunity 8, 297–303 (1998).
17. T. H. Lee et al., Mol. Cell. Biol. 23, 8377–8385
18. T. H. Lee, J. Shank, N. Cusson, M. A. Kelliher, J. Biol.
Chem. 279, 33185–33191 (2004).
Acknowledgments: We thank J. Starks, K. O’Rourke,
A. Maltzman, M. Schlatter, T. Soukup, L. Tam, R. Pattni,
R. Newman, H. Ngu, and D. Siler for technical assistance.
S. Sun, W. Alexander, S. Akira, and Y. He kindly provided
knockout mice. All authors were employees of Genentech.
Materials and Methods
Figs. S1 to S6
5 December 2013; accepted 6 February 2014
Published online 20 February 2014;
Highly Multiplexed Subcellular RNA
Sequencing in Situ
Je Hyuk Lee,1,2*† Evan R. Daugharthy,1,2,4* Jonathan Scheiman,1,2Reza Kalhor,2
Joyce L. Yang,2Thomas C. Ferrante,1Richard Terry,1Sauveur S. F. Jeanty,1Chao Li,1
Ryoji Amamoto,3Derek T. Peters,3Brian M. Turczyk,1Adam H. Marblestone,1,2
Samuel A. Inverso,1Amy Bernard,5Prashant Mali,2Xavier Rios,2John Aach,2George M. Church1,2†
Understanding the spatial organization of gene expression with single-nucleotide resolution
requires localizing the sequences of expressed RNA transcripts within a cell in situ. Here, we
describe fluorescent in situ RNA sequencing (FISSEQ), in which stably cross-linked complementary
DNA (cDNA) amplicons are sequenced within a biological sample. Using 30-base reads from
8102 genes in situ, we examined RNA expression and localization in human primary fibroblasts
with a simulated wound-healing assay. FISSEQ is compatible with tissue sections and whole-mount
embryos and reduces the limitations of optical resolution and noisy signals on single-molecule
detection. Our platform enables massively parallel detection of genetic elements, including gene
transcripts and molecular barcodes, and can be used to investigate cellular phenotype, gene regulation,
and environment in situ.
localization methods are limited to a handful of
to localize RNA transcriptome-wide (1–3). We
(FISSEQ) in 2003 and subsequently developed
methods to sequence DNA amplicons on a solid
substrate for genome and transcriptome se-
RNA in situ for gene expression profiling re-
he spatial organization of gene expression
can be observed within a single cell, tis-
sue, and organism, but the existing RNA
quires a spatially structured sequencing library
and an imaging method capable of resolving the
S1), RNA was reverse-transcribed in fixed cells
with tagged random hexamers (fig. S2A). We in-
corporated aminoallyl deoxyuridine 5′-triphosphate
and refixed the cells using BS(PEG)9, an amine-
reactive linker with a 4-nm spacer. The cDNA
fragments were then circularized before rolling
circle amplification (RCA) (fig. S2C), and BS
containing aminoallyl dUTP (fig. S2, D and E).
We found that random hexamer-primed RT was
inefficient (fig. S3A), but cDNA circularization
was complete within hours (fig. S3, B to D). The
resultwas single-strandedDNA nanoballs200 to
400 nm in diameter (fig. S4A), consisting of nu-
merous tandem repeats of the cDNA sequence.
BS(PEG)9 reduced nonspecific probe binding
after probe hybridization (fig. S4C). As a result,
the amplicons could be rehybridized many times,
with minimal changes in their signal-to-noise ra-
tio or position (fig. S4, D and E). Using SOLiD
sequencing by ligation (fig. S5), the signal over-
potent stem (iPS) cells, the amplicons counter-
stained subcellular structures, such as the plasma
membrane,the nuclear membrane,thenucleolus,
and the chromatin (Fig. 1A, fig. S6, and movies
S1 to S3). We were able to generate RNA se-
quencing libraries in different cell types, tissue
1Wyss Institute, Harvard Medical School, Boston, MA 02115,
MA 02115, USA.3Department of Stem Cell and Regenerative
Biology, Harvard University, Boston, MA 02138, USA.4Depart-
ment of Systems Biology, Harvard Medical School, Boston, MA
02115, USA.5Allen Institute for Brain Science, Seattle, WA
*These authors contributed equally to this work.
†Corresponding author. E-mail: firstname.lastname@example.org (J.H.L.);
21 MARCH 2014VOL 343
sections, and whole-mount embryos for three-
tiple resolution scales (Fig. 1, B and C).
High numerical aperture and magnification
cells (8–10), but many gene expression patterns
and wide-field mode, where it typically becomes
difficult to distinguish single molecules because
of the optical diffraction limit and low sensitivity
(11). To obtain a spot density that is high enough
and yet sufficiently low for discerning individual
which preextended sequencing primers are used
to reduce the number of molecular sequencing
tion site (Fig. 2A). Progressively longer sequenc-
ing primers result in exponential reduction of the
observed density, and the sequencing primer can
be changed during imaging to detect amplicon
pools of different density.
by tissue-specific artifacts and autofluorescence,
which impede accurate identification of objects.
If objects are nucleic acids, however, discrete
sequences,rather than the analogsignalintensity,
can be used to analyze the image. For FISSEQ,
putative nucleic acid sequences are determined
pared with reference sequences, and a null value
is assigned to unaligned pixels. With a suitably
long read length (L), a large number of unique
sequences (n) can be used to identify transcripts
or any other objects with a false-positive rate of
approximately n/4Lper pixel. Because the inten-
sity threshold is not used, even faint objects are
registeredon the basisof their sequence,whereas
background noise, autofluorescence, and debris
are eliminated (Fig. 2B).
We applied these concepts to sequence the
ends–polymerase chain reaction (RACE-PCR)
(12).AfterRTand molecular amplificationof the
5′ end followed by fluorescent probe hybridiza-
tion (fig. S7A), we quantified the concentration-
and time-dependent mCherry gene expression
in situ (fig. S7B). Using sequencing-by-ligation,
we then determined the identity of 15 contiguous
bases from each amplicon in situ, corresponding
to the transcription start site (fig. S7C). When the
sequencing reads were mapped to the vector
sequence, 7472 (98.7% ) amplicons aligned to
amplicons mapped within two bases of the pre-
dicted transcription start site (fig. S7D).
We then sequenced the transcriptome in hu-
man primary fibroblasts in situ (Fig. 3A) and gen-
per-base error rate of 0.64% (fig. S8). Using an
automated analysis pipeline (fig. S9), we identified
4171 genes, of which 13,558 (90.6%) amplicons
mapped to the correct annotated strand (Fig. 3B,
fig. S10, and table S1). We found that mRNA
(43.6%) was relatively abundant even though
random hexamers were used for RT (Fig. 3C).
Ninety genes with the highest expression counts
included fibroblast markers (13), such as fibro-
nectin (FN1); collagens (COL1A1, COL1A2,
COL3A1); matrix metallopeptidases and inhib-
itors (MMP14, MMP2, TIMP1); osteonectin
(SPARC); stanniocalcin (STC1); and the bone
morphogenesis–associated transforming growth
factor (TGF)–induced protein (TGFBI), repre-
senting extracellular matrix, bone development,
and skin development [Benjamini-Hochberg false
discovery rate (FDR) <10−19, 10−5, and 10−3,
respectively] (Fig. 3D) (14). We made Illumina
seq.Pearson’sr correlation coefficientbetween
RNA-seq and FISSEQ ranged from 0.52to0.69
(P< 10−16),excluding one outlier (FN1). For 854
genes with more than one observation, Pearson’s
(P < 10−3) between FISSEQ and RNA-seq from
fibroblasts, lymphocytes, and iPS cells, respec-
tively (Fig. 3E). When FISSEQ was compared
with gene expression arrays, Pearson’s r was as
high as 0.73 (P < 10−16) among moderately ex-
pressed genes,whereasgenes with low or high ex-
Fig. 1. Construction of 3D RNA-seq li-
braries in situ. After RT using random hex-
amers with an adapter sequence in fixed cells,
the cDNA is amplified and cross-linked in situ.
(A) A fluorescent probe is hybridized to the
adapter sequence and imaged by confocal
microscopy in human iPS cells (hiPSC) (scale
bar: 1 mm) and whole-mount Drosophila em-
bryos (scale bar: 5 mm), although we have not
sequenced these samples. (C) 3D rendering of
dized to cDNA amplicons. FISH, fluorescence
in situ hybridization.
Fig. 2. Overcoming resolution limitations and enhancing the signal-to-noise ratio. (A) Ligation
of fluorescent oligonucleotides occurs when the sequencing primer ends are perfectly complementary to the
than using an arbitrary intensity threshold, color sequences at each pixel are used to identify objects. For
removing unaligned pixels, the nuclear background noise is reduced in fibroblasts (scale bar: 20 mm).
VOL 343 21 MARCH 2014
Highly abundant genes in RNA-seq and gene ex-
pression arrays were involved in translation and
splicing (figs. S11 and S12), whereas such genes
12,427 (83.1%) and 2533 (16.9%) amplicons in
that nuclear RNA was 2.1 [95% confidence in-
terval (CI) 1.9 to 2.3] times more likely to be non-
coding (P < 10−16), and antisense mRNA was
1.8 [95% CI 1.7 to 2.0] times more likely to be
nuclear (P < 10−16). We confirmed nuclear enrich-
mitochondrial 16S ribosomal RNA (rRNA) (table
S2),whereasmRNA,such as COL1A1, COL1A2,
and THBS1, localized to the cytoplasm (table S3).
Fig. 3. Whole-transcriptome in situ RNA-seq inprimary fibroblasts.
(A) From deconvolved confocal images, 27-base reads are aligned to the
reference, and alignments are spatially clustered into objects. (B) Of the
amplicons, 90.6% align to the annotated (+) strand. (C) mRNA and non-
coding RNA make up 43.6% and 6.9% of the amplicons, respectively.(D) GO
term clustering for the top 90 ranked genes. (E) FISSEQ of 2710 genes
fromfibroblasts comparedwithRNA-seq forfibroblast, Bcell,and iPScells.
Pearson’s correlation is plotted as a function of the gene expression level.
(F) Subcellular localization enrichment compared to the whole transcriptome
distribution. (G) Of the amplicons, 481 map to the FN1 mRNA, showing an
alternatively spliced transcript variant and a single-nucleotide polymorphism
Fig. 4. Functionalanalysisoffibroblastsduring
simulated wound healing. (A) In EGF medium,
rRNA makes up 82.7% of the amplicons. (B) EGF
from FBS medium (different colors denote genes).
(C) The top 100 ranked genes from FBS versus EGF
FISSEQ clustered for functional annotation. (D) An
The image segments are based on the cell morphol-
ogy. (E) Comparison of 4533 genes from migrating
and contact-inhibited cells. (F) Twelve genes are dif-
ferentially expressed (Fisher’s exact test P < 0.05
and>fivefold; 180genes). (SeetableS4.)(G)The
top 100 genes in fibroblasts are enriched for terms
associated with ECM-receptor interaction and focal
adhesion kinase complex (bold letters). During cell
migration, genes involved in ECM-receptor-cytoskeleton
signaling and remodeling are differentially expressed
(red letters). THBS, thrombospodin; COMP, cartilage
IBSP, integrin-binding sialoprotein; PKC, protein
kinase C; FAK, focal adhesion kinase; PI3K, phos-
phatidylinositol 3-kinase; MLC, myosin light chain;
PAK, p21-activated protein kinase; WASP, Wiskott-
Aldrich syndrome protein.
21 MARCH 2014VOL 343
FN1 has three variable domains referred to as Download full-text
EDA, EDB, and IIICS, which are alternatively
spliced (15). We did not observe development-
associated EDB, but observed adult tissue–
associated EDA and IIICS (Fig. 3G).
We also sequenced primary fibroblasts in situ
after simulating a response to injury, obtaining
156,762 reads (>5 pixels), representing 8102
annotated genes (Fig. 4A and fig S13, A to D).
Pearson’s r was 0.99 and 0.91 between different
wound sites and growth conditions, respectively
(Fig. 4B and fig. S13, E and F). In medium with
epidermal growth factor (EGF), 82.7% of the
amplicons were rRNA compared to 42.7% in
medium were enriched for fibroblast-associated
GO terms, whereas rapidly dividing cells in EGF
medium were less fibroblast-like (Fig. 4C) with
alternative splicing of FN1 (fig. S14). In regions
expression (Fisher’s exact test P < 0.05 and
>fivefold change) (Fig. 4, D to F, and table S4),
eight of which were associated with the extra-
cellular matrix (ECM)–receptor–cytoskeleton in-
teraction, including GID4, FHDC1, PRPF40A,
LMO7, and WNK1 (Fig. 4G and table S4).
In summary, we present a platform for
transcriptome-wide RNA sequencing in situ and
demonstrate imaging and analytic approaches
FISSEQ correlates well with RNA-seq, except
for genes involved in RNA and protein process-
ing, possibly because some cellular structures or
classes of RNA are less accessible to FISSEQ. It
is notable that FISSEQ generates far fewer reads
than RNA-seq but predominantly detects genes
characterizing cell type and function. If this find-
ing can be generalized, FISSEQ may be used to
identify cell types based on gene expression pro-
bine transcriptome profiling and in situ mutation
detection in a high-throughput manner (16–18).
Using RNA barcodes from expression vectors,
one can label up to 4N(N = barcode length) cells
bination of fluorescent proteins (19). Similar to
next-generation sequencing, we expect advances
in read length, sequencing depth and coverage,
and library preparation (i.e., fragmentation, rRNA
depletion, targeted sequencing). Such advances
may lead to improved stratification of diseased tis-
sues in clinical medicine. Although more work re-
mains, our present demonstration is an important
References and Notes
1. G. Diez-Roux et al., PLOS Biol. 9, e1000582 (2011).
2. H. Zeng et al., Cell 149, 483–496 (2012).
3. E. Lécuyer et al., Cell 131, 174–187 (2007).
4. J. Shendure et al., Science 309, 1728–1732 (2005).
5. J. B. Kim et al., Science 316, 1481–1484 (2007).
6. R. D. Mitra et al., Proc. Natl. Acad. Sci. U.S.A. 100,
7. R. Drmanac et al., Science 327, 78–81 (2010).
8. A. M. Femino, F. S. Fay, K. Fogarty, R. H. Singer, Science
280, 585–590 (1998).
9. J. M. Levsky, S. M. Shenoy, R. C. Pezo, R. H. Singer,
Science 297, 836–840 (2002).
10. A. Raj, P. van den Bogaard, S. A. Rifkin, A. van Oudenaarden,
S. Tyagi, Nat. Methods 5, 877–879 (2008).
11. S. Itzkovitz, A. van Oudenaarden, Nat. Methods 8
(suppl.), S12–S19 (2011).
12. M. A. Frohman, M. K. Dush, G. R. Martin, Proc. Natl.
Acad. Sci. U.S.A. 85, 8998–9002 (1988).
13. R. J. McAnulty, Int. J. Biochem. Cell Biol. 39, 666–671
14. W. Huang, B. T. Sherman, R. A. Lempicki, Nat. Protoc. 4,
15. A. R. Kornblihtt et al., FASEB J. 10, 248–257 (1996).
16. R. Ke et al., Nat. Methods 10, 857–860 (2013).
17. K. Zhang et al., Nat. Methods 6, 613–618 (2009).
18. J. H. Lee et al., PLOS Genet. 5, e1000718 (2009).
19. J. Livet et al., Nature 450, 56–62 (2007).
Acknowledgments: Data can be downloaded from
http://arep.med.harvard.edu/FISSEQ_Science_2014/ and Gene
Expression Omnibus (gene expression arrays: GSM313643,
GSM313646, and GSM313657; RNA-seq: GSE54733). We
thank S. Kosuri, K. Zhang, and M. Nilsson for discussions;
A. DePace for Drosophila embryos; and I. Bachelet for
antibody conjugation. Funded by NIH Centers of Excellence
in Genomic Sciences grant P50 HG005550. J.H.L. and
co-workers were funded by the National Heart, Lung, and
Blood Institute, NIH, grant RC2HL102815; the Allen Institute
for Brain Science, and the National Institute of Mental
Health, NIH, grant MH098977. E.R.D. was funded by NIH
grant GM080177 and NSF Graduate Research Fellowship
Program grant DGE1144152. A.H.M. was funded by the Hertz
Foundation. Potential conflicts of interests for G.M.C. are
listed on http://arep.med.harvard.edu/gmc/tech.html. J.H.L.,
E.R.D., R.T., and G.M.C. are authors on a patent application
from the Wyss Institute that covers the method of generating
three-dimensional nucleic acid–containing matrix.
Materials and Methods
Figs. S1 to S14
Tables S1 to S4
Movies S1 to S6
26 December 2013; accepted 13 February 2014
Published online 27 February 2014;
Structure of the Mitochondrial
Translocator Protein in Complex
with a Diagnostic Ligand
Łukasz Jaremko,1,2* Mariusz Jaremko,1* Karin Giller,1Stefan Becker,1† Markus Zweckstetter1,2,3†
The 18-kilodalton translocator protein TSPO is found in mitochondrial membranes and mediates the
import of cholesterol and porphyrins into mitochondria. In line with the role of TSPO in mitochondrial
function, TSPO ligands are used for a variety of diagnostic and therapeutic applications in animals
and humans. We present the three-dimensional high-resolution structure of mammalian TSPO reconstituted
in detergent micelles in complex with its high-affinity ligand PK11195. The TSPO-PK11195 structure is
described by a tight bundle of five transmembrane a helices that form a hydrophobic pocket accepting
PK11195. Ligand-induced stabilization of the structure of TSPO suggests a molecular mechanism for
the stimulation of cholesterol transport into mitochondria.
its gene family is present in almost all organisms
(1–3). TSPO was first described as a peripheral
benzodiazepine receptor, a secondary receptor
for diazepam (1, 4). TSPO was subsequently
found to be responsible for the transport of cho-
he 18-kD translocator protein TSPO is
preferentially expressed in mitochondrial
membranes of steroidogenic tissues, and
lesterol into mitochondria, thereby influencing
adaptation (7). Expression of TSPO is strongly
up-regulated in areasof brain injury andin neuro-
inflammatory conditions including Alzheimer’s
and Parkinson’s diseases (2). TSPO is located at
drial membrane, and was suggested to be part of
anion channel, and the adenine nucleotide trans-
locator) the mitochondrial permeability transition
TSPO ligands have potential diagnostic and
(2). TSPO ligands such as XBD-173 might also
be useful in treating anxiety with reduced side
effects relative to traditional benzodiazepine-
related drugs (9). The best-characterized ligand
of TSPO is 1-(2-chlorophenyl)-N-methyl-N-
(PK11195), which binds to TSPO with nanomo-
lar affinity in many species (10–13). PK11195 is
used as a biomarker in positron emission tomog-
raphy to visualize brain inflammation in pa-
tients with neuronal damage (2, 10). Moreover,
1Max-Planck-Institut für Biophysikalische Chemie, 37077
Göttingen, Germany.2Deutsches Zentrum für Neurodegenera-
tive Erkrankungen, 37077 Göttingen, Germany.3Center for
Nanoscale Microscopy and Molecular Physiology of the Brain,
University Medical Center, 37073 Göttingen, Germany.
*These authors contributed equally to this work.
(M.Z.); email@example.com (S.B.)
VOL 343 21 MARCH 2014