Page 121 - Science

RESEARCH

MOLECULAR BIOLOGY (FISH) signals (Fig. 1C), and efficient, as 60% to
70% of cells across different lines were consistently
Transcription-coupled changes in labeled with one or two puncta with a sufficiently
high signal to background ratio (Fig. 1, D and E).
We next investigated both the precision of
nuclear mobility of mammalian the dCas9 targeting and its potential interfer-
ence with enhancer activation during differen-
cis-regulatory elements tiation. We performed anti-GFP (green fluorescent
protein) and anti-H3K27ac (acetylation of histone
H3 at lysine 27) chromatin immunoprecipitation
1
1
1
Bo Gu, Tomek Swigut, Andrew Spencley, 1,2 Matthew R. Bauer, Mingyu Chung, 1 (ChIP)–quantitative polymerase chain reaction
1
Tobias Meyer, Joanna Wysocka 1,3,4,5 * (qPCR) analyses from dCas9-eGFP mESCs with
or without a CARGO array, and from mEpiLCs
derived from them through 48 hours of differ-
To achieve guide RNA (gRNA) multiplexing and an efficient delivery of tens of distinct gRNAs
into single cells, we developed a molecular assembly strategy termed chimeric array of entiation. Using qPCR amplicons spanning either
gRNA oligonucleotides (CARGO). We coupled CARGO with dCas9 (catalytically dead Cas9) theCARGO-arraytargeted region(a toc)oren-
hancer E1 itself (d and e) (fig. S2B), we observed
imaging to quantitatively measure the movement of enhancers and promoters that undergo
that: (i) dCas9 recruitment is dependent on the
differentiation-associated activity changes in live embryonic stem cells. Whereas all
array and localized to the targeted region, and
examined functional elements exhibited subdiffusive behavior, their relative mobility
(ii) enhancer activation, as indirectly measured
increased concurrently with transcriptional activation. Furthermore, acute perturbation of
by H3K27ac, retains its developmental dynamics
RNA polymerase II activity can reverse these activity-linked increases in loci mobility.
and is not considerably affected by dCas9-eGFP
Through quantitative CARGO-dCas9 imaging, we provide direct measurements of cis-
recruitment in the vicinity (fig. S2, C and D). In
regulatory element dynamics in living cells and distinct cellular and activity states and agreement, Fgf5 mRNA single-molecule FISH
uncover an intrinsic connection between cis-regulatory element mobility and transcription.
(smFISH) analysis of mEpiLC dCas9-eGFP cells Downloaded from
with or without a CARGO array showed no signif-
is-regulatory DNA elementssuch as promot- into single cells (diagram in Fig. 1A). With this icant differences in transcript numbers between
ers and long-range enhancers mediate pre- strategy, gRNA 12- and 18-nucleotide monomers the two populations (fig. S4A). Taken together,
cisespatiotemporalcontrolof geneexpression can be readily assembled in a single step with these results demonstrate that CARGO-dCas9
(1–4). Understanding regulatory elements’ 70% and 60% efficiencies, respectively, and an imaging provides a specific and noninvasive strat-
C dynamics in living cells and their changes even higher degree of multiplexing can be readily egy to label functional cis-regulatory sequences in
in relation to transcriptional status and cellular accomplished (fig. S1; see materials and methods their native chromosomal context.
states is important for comprehending gene ex- and other supplementary materials for details and To follow the movement of the Fgf5 enhancer
pression control. To explore these questions, we protocol). We then asked whether robust labeling in its active and inactive state, we performed live http://science.sciencemag.org/
first had to address technical limitations asso- of nonrepetitive cis-regulatory elements can be CARGO-dCas9 imaging in mESCs or in mEpiLCs
ciated with the labeling of small, nonrepetitive achievedinmammaliancellsbycombiningCARGO after 48 hours of differentiation and tracked
genomic elements. Previous strategies typically with dCas9 imaging. To this end, we chose mouse centers of labeled loci with high temporal reso-
relied on inserting heterologous arrays of bacte- embryonic stem cells (mESCs) and their in vitro lution (Fig. 2A and movie S1). Visual inspection
rial operator sequences that are much larger than differentiation to epiblast-like cells (mEpiLCs) as of the recorded time-lapse images revealed that
a typical individual enhancer or promoter and, a cell fate transition model (16–18) (fig. S2A). We a substantial fraction of Fgf5 enhancers displayed
owing to their highly repetitive nature, may be have previously shown that during this transition increased mobility in mEpiLCs compared with
subject to regulation specific for repetitive se- a cluster of enhancers located ~30 to 54 kb down- mESCs (compare movies S2 and S3), and this was on March 1, 2018
quences (5–10). Alternatively, dCas9 (catalytically stream from the Fgf5 promoter is activated de also evident when we examined individual trajec-
dead Cas9) imaging can facilitate RNA-guided novo (17) (fig. S2B) and that these changes are tories over a fixed time interval (Fig. 2A, compare
labeling of native genomic regions (11), but accompanied by the transcriptional induction c and d). To characterize the Fgf5 enhancer move-
similarly to other methods, it requires a large of the Fgf5 gene (17). To label this developmen- ment quantitatively, we computed the mean square
number of fluorescent molecules bound to the tally regulated enhancer cluster, we designed displacement (MSD) for these trajectories (Fig.
target locusatanygiventimetoenablemicroscopic three CARGO arrays, each harboring 12 different 2B) and extracted two parameters: the scaling
visualization. Consequently, in the absence of ap- gRNAs, spanning a 2-kb window immediately exponent a and the apparent diffusion coefficient
proaches allowing for highly multiplexed and uni- upstream of the first enhancer (E1) within the D app . Time-averaged MSD (tMSD) plots show an
form delivery of guide RNAs (gRNAs) to target cells, cluster (fig. S2B). We generated clonal mESC lines anomalous scaling exponent of a =0.53±0.21 in
dCas9 imaging has been practically limited to with a stably integrated, inducible dCas9-eGFP mESCs and a =0.51 ±0.26 in mEpiLCs (Fig. 2B),
the labeling of repetitive sequences (12–15). transgene expressed at relatively low (albeit suggesting subdiffusive behavior of the enhancer,
To overcome this bottleneck, we developed a clone-to-clone–variable) level upon induction similar to chromosome movement in bacte-
molecular assembly strategy, termed chimeric with doxycycline (Dox) (fig. S3, A and B). As ex- ria, yeast, and B cells (8, 19, 20). Additionally, a
array of gRNA oligonucleotides (CARGO), that pected (11), dCas9-eGFP displayed previously comparable scaling exponent a (0.54 in mESCs
can achieve highly multiplexed gRNA delivery observed nucleolar retention in the absence of and 0.51 in mEpiLCs) was obtained from the
gRNA and robustly labeled telomeric repeats upon time- and ensemble-averaged MSD (eMSD) plots
1 Department of Chemical and Systems Biology, Stanford transfection with telomere gRNA (Fig. 1B). Upon (Fig. 2B, shaded area denotes SEM), indicating
University School of Medicine, Stanford, CA 94305, USA. transfection of the three Fgf5 enhancer CARGO that the loci movements are ergodic.
2 Cancer Biology Program, Stanford University School of
3
Medicine, Stanford, CA 94305, USA. Department of arrays, one or two puncta were clearly visible in Whereas the scaling exponent extracted from
Developmental Biology, Stanford University School of live cells to which arrays were successfully in- our MSD measurements was similar for the two
4
Medicine, Stanford, CA 94305, USA. Institute for Stem troduced, as measured by a fluorescent marker examined cell states, the apparent anomalous
Cell Biology and Regenerative Medicine, Stanford encoded by the CARGO plasmid (Fig. 1B). The diffusion coefficient of the Fgf5 enhancer exhib-
University School of Medicine, Stanford, CA 94305, USA.
5 Howard Hughes Medical Institute, Stanford University labeling was specific, as confirmed by the colocal- ited a significant increase in mEpiLCs compared
School of Medicine, Stanford, CA 94305, USA. ization of the dCas9-eGFP immunofluorescence with ESCs (Fig. 2C and table S2). Moreover, dis-
*Corresponding author. Email: wysocka@stanford.edu and Fgf5 DNA fluorescence in situ hybridization tribution of D app was unimodal in the mESC state,
Gu et al., Science 359, 1050–1055 (2018) 2 March 2018 1of6