Corrected 4 November 2019. See full text.

R ES E A RC H

SYNTHETIC BIOLOGY protein expression, we gently lysed the cells (~2 ×

106 cells per sample) and centrifugated the lysate
to enrich for buoyant particles, which would
Ultrasound imaging of gene include any gas vesicles. The top fraction of the
centrifugated lysate was then screened for gas

expression in mammalian cells vesicles using transmission electron microscopy

(TEM). These experiments took advantage of
the intrinsic stochasticity of transient cotrans-
Arash Farhadi1, Gabrielle H. Ho2*, Daniel P. Sawyer1, fection, in terms of the ratios of genes and the
Raymond W. Bourdeau2†, Mikhail G. Shapiro2‡ overall DNA quantity delivered to each cell, to
simultaneously sample a broad range of gene
The study of cellular processes occurring inside intact organisms requires methods to stoichiometries and expression levels without
visualize cellular functions such as gene expression in deep tissues. Ultrasound is a prior knowledge of parameters leading to gas
widely used biomedical technology enabling noninvasive imaging with high spatial and vesicle formation.
temporal resolution. However, no genetically encoded molecular reporters are available to The cotransfection of the gas vesicle genes
connect ultrasound contrast to gene expression in mammalian cells. To address this from Halobacterium salinarum and Anabaena
limitation, we introduce mammalian acoustic reporter genes. Starting with a gene cluster flos-aquae did not lead to the formation of
derived from bacteria, we engineered a eukaryotic genetic program whose introduction detectable gas vesicles. However, the cotrans-
into mammalian cells results in the expression of intracellular air-filled protein fection of nine gas vesicle–forming genes from
nanostructures called gas vesicles, which produce ultrasound contrast. Mammalian Bacillus megaterium (Fig. 1B) resulted in the
acoustic reporter genes allow cells to be visualized at volumetric densities below 0.5% and production of unmistakable gas vesicles, as evi-
permit high-resolution imaging of gene expression in living animals. denced by their appearance in TEM images

Downloaded from http://science.sciencemag.org/ on May 25, 2021

(Fig. 1C). The 9 genes originate from an eleven-
gene B. megaterium gene cluster previously used

T
he study of cellular function in the context isms, gas vesicles are encoded by clusters of 8 to express gas vesicles in Escherichia coli (13, 21),
of intact living organisms is a grand chal- to 14 genes, including one or two primary struc- with the exception of GvpR and GvpT, which
lenge in biological research and synthetic tural proteins, and several other essential genes were found to be unnecessary for gas vesicle
biology (1). Addressing this challenge re- encoding putative assembly factors or minor formation (fig. S1).
quires imaging tools to visualize specific shell constituents. Using the nine genes identified in our sto-
cells in tissues ranging from the developing brain The use of gas vesicles as reporter genes re- chastic screen, we set out to construct a poly-
to tumors, and to monitor gene- and cell-based quires the heterologous expression of their cog- cistronic mammalian operon for consistent gas
therapeutic agents in vivo (2). However, most nate multigene operon in a new cellular host, vesicle expression by joining these genes using
common methods for imaging cellular processes ensuring proper transcription and translation the viral cotranslational self-cleavage peptide P2A
such as gene expression rely on fluorescent or of each gene, functional folding of each corre- (22). Having determined that all genes except
luminescent proteins, which have limited per- sponding protein, and appropriate stoichiome- GvpB could tolerate P2A peptide additions (fig.
formance in intact animals because of the poor try and colocalization of the constituents for gas S2 and table S1), we constructed a polycistronic
penetration of light in biological tissue (3, 4). By vesicle assembly. Recently, a genetic engineer- plasmid containing the eight P2A-tolerant gas
contrast, ultrasound easily penetrates most tis- ing effort succeeded in expressing gas vesicles as vesicle genes connected by P2A sequences, and
sues, enabling deep, noninvasive imaging with acoustic reporter genes (ARGs) in commensal cotransfected it into HEK293T cells together
excellent spatial and temporal resolution (~100 mm bacteria, allowing their imaging in the mouse with a plasmid encoding GvpB. Unfortunately,
and ~1 ms, respectively) (2, 5). These capabilities, gastrointestinal tract (13). If ARGs could be this did not result in the production of gas ves-
along with its safety, portability, and low cost, developed for mammalian cells, then this would icles. We hypothesized that one or more of the
have made ultrasound a widely used technol- enable the study of how such cells develop, func- genes in our polycistronic plasmid was expressed
ogy in biomedicine. Despite these advantages, tion, and malfunction within the context of at an insufficient level, and used a comple-
to date, ultrasound has played a relatively small model organisms and enable the in vivo imaging mentation assay to identify GvpJ, GvpF, GvpG,
role in cellular imaging owing to the lack of ap- of mammalian cells engineered to perform di- GvpL, and GvpK as bottleneck genes (fig. S3).
propriate genetically encoded reporters. agnostic or therapeutic functions (14–16). How- This led us to construct a polycistronic “booster”
Recently, biomolecular contrast agents for ul- ever, developing ARGs for mammalian cells plasmid containing these five genes, ordered to
trasound were introduced based on gas vesicles, represents an even greater synthetic biology minimize P2A modifications to GvpJ and GvpK,
air-filled protein nanostructures that evolved in challenge because of the differences in tran- which were found to be most limiting. The co-
certain waterborne bacteria and archaea to pro- scription, translation, colocalization, and protein transfection of the booster plasmid, together with
vide cellular buoyancy (6, 7). Gas vesicles com- folding between prokaryotes and eukaryotes the two plasmids above (Fig. 1D), enabled robust
prise a 2-nm-thick protein shell enclosing a gas (17–19). To our knowledge, no genetic operon expression of gas vesicles in cells (Fig. 1E). We
compartment with dimensions on the order of larger than six genes has been moved between named this set of three genetic constructs “mam-
100 nm. The acoustic impedance mismatch be- these domains of life (20). malian acoustic reporter genes,” or mARGs.
tween their gas interior and surrounding aqueous Here, we describe the expression of ARGs in After establishing polycistronic constructs for
media allows gas vesicles to strongly scatter mammalian cells to enable ultrasound imaging mammalian gas vesicle assembly, we used an
sound waves and thereby serve as ultrasound of mammalian gene expression. To identify a integrase (23, 24) to incorporate them into the
contrast agents (8–12). In their native organ- set of genes capable of encoding gas vesicle cellular genome for stable expression under a
assembly in mammalian cells, we synthesized doxycycline-inducible TRE3G promoter, with
1
Division of Biology and Biological Engineering, California
individual gas vesicle genes from three differ- fluorescent proteins added to each construct
Institute of Technology, Pasadena, CA, USA. 2Division of ent microbial species using codons optimized as transfection indicators (Fig. 2A). We trans-
Chemistry and Chemical Engineering, California Institute of for human expression, cloned each gene into a fected these plasmids into HEK293-tetON cells
Technology, Pasadena, CA, USA. separate monocistronic plasmid, and transiently and used flow cytometry to sort cells according
*Present address: Department of Bioengineering, University of
Pennsylvania, Philadelphia, PA, USA. †Present address: Codiak
cotransfected mixtures of the genes from each to the expression level of each fluorescent re-
Biosciences, Cambridge, MA, USA. species into human embryonic kidney (HEK) porter. We found that the cell population com-
‡Corresponding author. Email: mikhail@caltech.edu 293T cells (Fig. 1A). After allowing 72 hours for bining the strongest expression of each construct

Farhadi et al., Science 365, 1469–1475 (2019) 27 September 2019 1 of 6

 Corrected 4 November 2019. See full text.
R ES E A RC H | R E PO R T

produced the largest number of gas vesicles in signal between the collapsing and postcol- HEK cells at varying ratios. We were able to de-
(Fig. 2B and fig. S4, A to D). To ensure that lapse frames reveal specifically the presence of tect the presence of mARG-expressing cells in
mARG expression was not limited to HEK293 gas vesicles. these mixtures down to 2.5% of total cells (Fig. 3H),
cells, we also transfected Chinese hamster ovary We implemented this collapse-based imaging corresponding to <0.5% volumetric density, or
cells (CHO-K1) and obtained similar results approach using an amplitude modulation pulse about three cells or 135 gas vesicles per voxel with
(fig. S4, E to G). sequence (10), which we found to provide the dimensions of 100 mm. A similar voxel-averaged
To generate a stable monoclonal cell line ex- best cancellation of non–gas vesicle signals. concentration of gas vesicles was detectable in
pressing mARGs for detailed analysis, we sorted When hydrogels containing mARG-HEK cells a monoculture of mARG-HEK cells induced to
individual high-expression HEK293-tetON cells were imaged with this technique at 18 MHz, express 1.4 ± 0.6 gas vesicles per cell (fig. S8).
for monoclonal growth (Fig. 2C), producing they were easily distinguishable from mCherry- In many imaging experiments, the output of a
30 cell lines, which we screened for viability, HEK controls on the basis of their contrast gene circuit is read out only once. However, in
fluorescence, and gas vesicle formation (Fig. 2D dynamics (Fig. 3C). Critically, although this im- some cases, it may be desirable to track gene
and table S2). The number of gas vesicles per aging paradigm requires the collapse of gas ves- expression over time. We therefore tested whether
cell was then estimated from TEM images, and icles inside cells, this does not affect cell viability mARG-expressing cells in which the gas vesicles
a cell line yielding the largest quantity of gas (Fig. 3D). have been collapsed during imaging could re-
vesicles was selected and named mARG-HEK. To determine whether mARGs can faithfully express these reporters to allow additional imaging.
When induced for 72 hours with doxycycline monitor circuit-driven gene expression (25, 26), mARG-HEK cells cultured in a nutrient-supported
(1 mg/ml) and 5 mM sodium butyrate (to reduce we measured the dynamic ultrasound response hydrogel produced clear ultrasound contrast 3 days
epigenetic silencing), this cell line produced on of mARG-HEK cells under the control of a after induction and were able to reexpress their
average 45 gas vesicles per cell (Fig. 2E). Using doxycycline-inducible promoter (Fig. 3E). After acoustic reporters over three additional days (Fig.
thin-section TEM, gas vesicles could clearly be induction with 1 mg/ml doxycycline, the cells 3, I and J).
seen in the cytosol of individual mARG-HEK showed a gradual buildup of ultrasound signal, Having engineered mammalian cells to stably

Downloaded from http://science.sciencemag.org/ on May 25, 2021

cells (Fig. 2F). From TEM images of cell lysates, with clear contrast appearing on day 2 and in- express gas vesicles and characterized their abil-
we measured the average dimensions of gas creasing over the next 4 days (Fig. 3F). These ity to produce ultrasound contrast in vitro, we
vesicles produced in this cell line to be 64 ± kinetics are similar to those observed with flu- next tested the ability of mARG expression to be
12-nm wide (standard deviation [SD], n = 1828 orescent indicators (fig. S7A). When the gene visualized in vivo with high spatial resolution.
gas vesicles) and 274 ± 212-nm long (SD, n = circuit was driven using a range of inducer We formed model tumor xenografts in immuno-
1828 gas vesicles), with some reaching lengths concentrations, the ultrasound contrast followed compromised mice by inoculating mARG-HEK
>1 mm (aspect ratios >30) (Fig. 2, G and H). This the expected transfer function of the promoter cells in Matrigel subcutaneously in their left
corresponds to an average gas vesicle volume (Fig. 3G and fig. S7B). flanks (Fig. 4A). In the same mice, the right
of 0.605 attoliters. Together, the 45 gas vesi- To determine how sensitively mARG-expressing flanks were inoculated with mCherry-HEK con-
cles expressed in an average mARG-HEK cell cells could be detected in a mixed cell population, trol cells. We induced reporter gene expression
are expected to occupy just 0.0027% of the cell’s we combined mARG-HEK cells with mCherry- in both tumors for 4 days with systemic injections
cytosolic volume.
The expression of gas vesicles did not change
the gross morphology of mARG-HEK cells (Fig. 2I)
and was nontoxic, as determined by three differ-
ent assays (Fig. 2J), compared with a similarly
prepared control cell line (mCherry-HEK) (fig. S5,
A and B). During a 6-day coculture, mARG-HEK
cells showed only a minor growth disadvantage
compared with mCherry-HEK cells (Fig. 2K). As
expected, both engineered cell lines grew more
slowly than wild-type HEK293T cells (fig. S6).
Having engineered mARG-HEK cells, we sought
to image their expression of acoustic reporters
with ultrasound. Gas vesicles encoded by the
B. megaterium gene cluster are expected to
produce linear ultrasound scattering (21). How-
ever, because mammalian cells themselves also
produce substantial linear contrast, detecting
gas vesicles expressed in such cells using linear
methods is challenging. To enable more selective
imaging of mARG expression, we took advan-
tage of the ability of gas vesicles to collapse
irreversibly above specific ultrasound pressure
thresholds (8, 9, 13, 21). A switch in the incident
ultrasound pressure from below to above such
a threshold results in a strong transient signal
from the gas vesicles, which decays to a lower
level in the next ultrasound frame owing to im- Fig. 1. Engineering of mammalian ARGs. (A) Schematic of the transient cotransfection assay used to
mediate dissolution of their gas contents and identify combinations of genes capable of producing gas vesicles in mammalian cells. (B) Schematic
the elimination of ultrasound scattering (Fig. 3, of nine genes from B. megaterium capable of encoding gas vesicle expression in mammalian cells.
A and B). Meanwhile, background tissue scat- Thin arrow denotes CMV promoter. polyA, SV40 polyadenylation element. (C) Representative TEM
tering rises with the increase in incident pres- image of purified gas vesicles expressed in HEK293T cells. (D) Gene cassettes comprising the
sure and remains constant at the new level. mammalian ARG construct, mARG. (E) Representative TEM image of gas vesicles purified from
Thus, images formed by taking the difference HEK293T cells transiently transfected with mARGs for 72 hours. All scale bars represent 500 nm.

Farhadi et al., Science 365, 1469–1475 (2019) 27 September 2019 2 of 6

 Corrected 4 November 2019. See full text.
R ES E A RC H | R E PO R T

Fig. 2. Formation, properties, and nontoxicity of gas vesicles in cells represents SEM (n = 3 biological replicates, each from two technical Downloaded from http://science.sciencemag.org/ on May 25, 2021
with genome-integrated mammalian ARGs. (A) Schematic of mARG replicates). (F) Representative TEM image of a 60-nm section through an
constructs used for genomic integration into cells with the piggyBac mARG-HEK cell showing an angled slice through two bundles of gas
transposase system. ITR, inverted terminal repeat; ChbGI, chicken b-globin vesicles in the cytosol. (G) Representative TEM image of gas vesicles
insulator; GFP, emerald green fluorescent protein; BFP, enhanced blue purified from mARG-HEK cells. (H) Size distribution of gas vesicles
fluorescent protein 2. (B) Representative TEM image of buoyancy-enriched expressed in mARG-HEK cells. The mean and SD of both distributions is
lysate from HEK293-tetON cells transfected with the constructs in (A) illustrated as a circle and with error bars (n = 1828 gas vesicles). (I) Phase-
and sorted for high expression of all three operons. (C) Fluorescence- contrast images of mARG-HEK and mCherry-HEK cells 72 hours after
activated cell sorting of HEK293-tetON cells transfected with the induction with 1 mg/ml doxycycline and 5 mM sodium butyrate. (J) Cell
constructs in (A). Red circles denote individual cells selected by sorting viability of mARG-HEK cells relative to mCherry-HEK cells after 72 hours of
to form monoclonal cell lines. (D) Selection process for monoclonal gene expression. Error bars indicate SEM. (K) Fraction of mARG-HEK cells
cell lines, including assays for viability, fluorescence intensity, and gas in coculture with mARG-mCherry cells seeded in equal numbers over
vesicle yield. (E) Number of gas vesicles expressed by monoclonal 6 days of gene expression (n = 3 biological replicates, each from four
HEK293-tetON cells after 72 hours of induced expression, as counted in technical replicates, with darker symbols showing the mean). Scale bars in
lysates using TEM. Bar represents the mean and the shaded area (B), (F), (G) represent 500 nm; scale bar in (I) represents 20 mm.

Farhadi et al., Science 365, 1469–1475 (2019) 27 September 2019 3 of 6

 Corrected 4 November 2019. See full text.
R ES E A RC H | R E PO R T

Fig. 3. Ultrasound
imaging of mammalian
gene expression in vitro.
(A) Illustration of the
collapse-based
ultrasound imaging
paradigm used to gener-
ate gas vesicle–specific
ultrasound contrast from
mARG-expressing cells.
(B) Representative non-
linear signal recorded
during a step change in
the incident acoustic
pressure, from 0.27 MPa
in the white-shaded
region to 1.57 MPa in the
gray-shaded region.
(C) Representative col-

Downloaded from http://science.sciencemag.org/ on May 25, 2021

lapse and postcollapse
ultrasound images of
mARG-HEK and mCherry-
HEK cells acquired during
this ultrasound imaging
paradigm and their
difference, indicating
gas vesicle–specific con-
trast. (D) Cellular viability
after being insonated
under 3.2-MPa acoustic
pressures, as measured
using the MTT assay.
(E) Schematic of a chem-
ically inducible gene
circuit with mARG
expression as its output.
All three mARG cassettes
in mARG-HEK cells are
under the control of the
doxycycline-inducible
TRE3G promoter (TRE),
with expression triggered
by incubation with doxy-
cycline. (F) Representa-
tive ultrasound images
and contrast measure-
ments in mARG-HEK cells
as a function of time after
induction with 1 mg/ml
doxycycline and 5 mM
sodium butyrate (n = 6
biological replicates, with
the darker dots showing
the mean). (G) Represen-
tative ultrasound images
and contrast measurements in mARG-HEK cells as a function of mean). (I) Schematic and representative ultrasound images from
doxycycline induction concentrations. Cells were allowed to express gas mARG-HEK cells in Matrigel reexpressing gas vesicles after acoustic
vesicles for 72 hours in the presence of 5 mM sodium butyrate (n = 6 collapse. Cells were induced with 1 mg/ml doxycycline and 5 mM sodium
biological replicates, with the darker dots showing the mean). A sigmoidal butyrate for 72 hours before and after 3.2-MPa acoustic insonation.
function is fitted as a visual guide. (H) Representative ultrasound Ultrasound images were acquired after an additional 72 hours in culture
images and contrast measurements in mARG-HEK cells mixed with after collapse. (J) Ultrasound contrast in mARG-HEK and mCherry-HEK
mCherry-HEK cells in varying proportions. Cells were induced with cells after initial expression, after collapse, after reexpression, and after
1 mg/ml doxycycline and 5 mM sodium butyrate for 72 hours before second collapse (n = 7 biological replicates, with the darker dots
imaging (n = 4 biological replicates, with the darker dots showing the showing the mean). GV, gas vesicles. All scale bars represent 1 mm.

Farhadi et al., Science 365, 1469–1475 (2019) 27 September 2019 4 of 6

 Corrected 4 November 2019. See full text.
R ES E A RC H | R E PO R T

of doxycycline and sodium butyrate (Fig. 4B). We

expected these nascent tumors to be mostly
vascularized at their perimeter, resulting in the
strongest inducible gene expression at the tumor
periphery (Fig. 4A). Ultrasound, with its sub–
100-mm spatial resolution (at 18 MHz), should
be able to discern this gene expression pattern,
whereas attaining such resolution would be chal-
lenging with optical techniques.
After 4 days of induction, we observed clear
ultrasound contrast in the flank inoculated with
mARG-HEK cells, and this was absent from the
contralateral side (Fig. 4, C and D). As expected,
the pattern observed with ultrasound revealed
mARG expression at the perimeter of the tumor,
whereas the core remained dark, and imaging
of adjacent ultrasound planes revealed that this
pattern of gene expression persisted across the
tumor mass (Fig. 4E and fig. S9).
The ultrasound-observed spatial distribution
of gene expression was consistent with the low
vascularity in the tumor core, as observed with

Downloaded from http://science.sciencemag.org/ on May 25, 2021

Doppler ultrasound (fig. S10). The peripheral
gene expression pattern was confirmed with
subsequent histological examination of the tis-
sue (Fig. 4F and fig. S11). In comparison, our
in vivo fluorescence images just showed the
presence of signal somewhere in the tissue and
not its precise distribution (Fig. 4G). These re-
sults, which were consistent across five animals
(fig. S12A), demonstrate that mARGs enable
gene expression imaging in vivo and highlight
the ability of ultrasound to visualize intricate
patterns of gene expression noninvasively. We
imaged three of the animals again after an ad-
ditional 4 days to look for reexpression of the
collapsed gas vesicles and observed ultrasound
contrast in each case (fig. S12B).
Our results establish the ability of an engi-
neered genetic construct encoding prokaryote-
derived gas vesicles to serve as a mammalian
reporter gene for ultrasound, providing the abil-
ity to monitor cellular location and function in-
side living organisms. mARGs provide many of
the capabilities associated with established ge-
netically encoded optical reporters, including
imaging cellular dynamics by promoter-driven
expression and mapping cellular populations in
complex samples. Whereas optical reporter genes
mainly provide these capabilities in culture and
surgically accessed tissues, mARGs enable gene
expression to be resolved noninvasively in vivo.
Although the genetic constructs described in
Fig. 4. Ultrasound imaging of mammalian gene expression in vivo. (A) Diagram of a mouse this work should be immediately useful in a
implanted with a subcutaneous tumor model, and the expected spatial pattern of vascularization and variety of contexts, considerable scope exists
doxycycline-induced reporter gene expression. (B) Experimental timeline. (C) Representative for further optimization to make ARGs as widely
ultrasound image of tumors containing mARG-HEK cells after 4 days of doxycycline administration. useful as green fluorescent protein (GFP) (5, 12).
mARG-specific contrast shown in the hot color map is overlaid on an anatomical B-mode image For example, accelerating mARG expression be-
showing the background anatomy. (D) Representative ultrasound image of tumors containing yond the day-scale kinetics shown in this study
mCherry-HEK cells after 4 days of doxycycline administration. (E) Ultrasound images of adjacent and developing sensitive imaging paradigms that
planes in the mARG-HEK tumor acquired at 1-mm intervals. The minimum and maximum values of do not require gas vesicle collapse would enable
color bars in (C) to (E) are 4000 and 40,000 arbitrary units, respectively. (F) Representative the imaging of more dynamic cellular processes.
fluorescence image of a histological tissue section of an mARG-HEK tumor. Blue color shows the In addition, although this study demonstrated
TO-PRO3 nucleus stain, green color shows GFP fluorescence, and red color shows mCherry essential mARG functionality with clonally se-
fluorescence. (G) Fluorescence image of a mouse implanted with mARG-HEK and mCherry-HEK lected cell lines, the expression of mARGs in
tumors on the left and right flanks, respectively, after 4 days of expression. Scale bars represent primary cells, their delivery to endogenous cells
1 mm for (C) to (F) and 1 cm for (G). by viral vectors, and their expression in transgenic

Farhadi et al., Science 365, 1469–1475 (2019) 27 September 2019 5 of 6

 Corrected 4 November 2019. See full text.
R ES E A RC H | R E PO R T

animals would greatly expand the utility of this 10. D. Maresca et al., Appl. Phys. Lett. 110, 073704 (2017). imaging of tissues was performed in the Biological Imaging
technology. To facilitate such uses, it would be 11. D. Maresca, D. P. Sawyer, G. Renaud, A. Lee-Gosselin, Facility of the Caltech Beckman Institute with support from the
M. G. Shapiro, Phys. Rev. X 8, 041002 (2018). Arnold and Mabel Beckman Foundation. We appreciate the
helpful to further condense the mARG constructs. 12. G. J. Lu, A. Farhadi, A. Mukherjee, M. G. Shapiro, Curr. Opin. help of A. Lee-Gosselin and Caltech’s Office of Laboratory
For example, genes could be consolidated into Chem. Biol. 45, 57–63 (2018). Animal Research with animal protocols and husbandry.
fewer clusters, and preliminary experiments show 13. R. W. Bourdeau et al., Nature 553, 86–90 (2018). Funding: A.F. was supported by an NSERC graduate fellowship.
that gvpB can be combined with the eight-gene 14. M. M. Davis, C. M. Tato, D. Furman, Nat. Immunol. 18, 725–732 D.P.S. was supported by an NSF graduate research fellowship
(2017). (grant no. 1745301). This research was supported by the
polycistron encoding gvpN-gvpU through an 15. A. H. Marblestone et al., Front. Comput. Neurosci. 7, 137 (2013). National Institutes of Health (grant nos. R01EB018975 and
internal ribosome entry sequence (IRES) (fig. 16. T. Schroeder, Nature 453, 345–351 (2008). U54CA199090 to M.G.S.), the Heritage Medical Research
S13). In addition, the total length of the coding 17. V. Gradinaru et al., Cell 141, 154–165 (2010). Institute (M.G.S.), the Packard Fellowship for Science and
sequence contained in mARG could be reduced 18. Y. W. Shieh et al., Science 350, 678–680 (2015). Engineering (M.G.S.), and the Burroughs Welcome Fund Career
19. E. Natan, J. N. Wells, S. A. Teichmann, J. A. Marsh, Curr. Opin. Award at the Scientific Interface (M.G.S.). Author
from 7.6 to 4.8 kb by eliminating the need for Struct. Biol. 42, 90–97 (2017). contributions: A.F. and M.G.S. conceived and planned the
redundant booster genes, relying instead on 20. D. M. Close et al., PLOS ONE 5, e12441 (2010). research. A.F. and G.H.H. performed the experiments. A.F. and
noncoding elements such as different-strength 21. A. Farhadi et al., AIChE J. 64, 2927–2933 (2018). R.W.B. designed the DNA sequences. A.F. and D.P.S. designed
promoters to tune expression stoichiometry. Fur- 22. A. L. Szymczak, D. A. A. Vignali, Expert Opin. Biol. Ther. 5, and optimized the ultrasound imaging sequences. A.F. analyzed
627–638 (2005). the data. A.F. and M.G.S. wrote the manuscript with input
ther optimization of mARG genetic constructs is from all authors. M.G.S. supervised the research. Competing
23. S. Ding et al., Cell 122, 473–483 (2005).
also needed to reduce epigenetic silencing and 24. M. H. Wilson, C. J. Coates, A. L. George Jr., Mol. Ther. 15, interests: A.F., G.H.H., D.P.S., and M.G.S. are inventors on
metabolic burden (27–29). Just as the engineer- 139–145 (2007). U.S. patent application 62/789,295 submitted by the California
ing of GFP over many years yielded brighter and 25. M. B. Elowitz, S. Leibler, Nature 403, 335–338 (2000). Institute of Technology that covers mammalian acoustic
26. T. S. Gardner, C. R. Cantor, J. J. Collins, Nature 403, 339–342 reporter genes. A.F., R.W.B., and M.G.S. are inventors on U.S.
more colorful reporters enabling new uses of flu- patent application 15/663,635 submitted by the California
(2000).
orescence microscopy, further engineering of the Institute of Technology that covers bacterial acoustic reporter
27. L. Gaidukov et al., Nucleic Acids Res. 46, 4072–4086
genetic constructs comprising mARGs would help (2018). genes. Data and materials availability: Plasmids encoding
cellular ultrasound penetrate and enable new GvpB, GvpNFGLKJU, and GvpJFGLK are available from M.G.S.

Downloaded from http://science.sciencemag.org/ on May 25, 2021

28. B. Jusiak et al., ACS Synth. Biol. 8, 16–24 (2019).
under a material agreement with the California Institute of
areas of mammalian biology and biomedicine. 29. J. J. Neville, J. Orlando, K. Mann, B. McCloskey, M. N. Antoniou,
Technology. The mARG genetic construct will be deposited with
Biotechnol. Adv. 35, 557–564 (2017).
Addgene at the time of manuscript publication. Raw data are
RE FE RENCES AND N OT ES
available from the authors upon reasonable request.
1. R. Y. Tsien, Nat. Rev. Mol. Cell Biol. 4, SS16–SS21 (2003). ACKN OWLED GMEN TS
2. D. I. Piraner et al., Biochemistry 56, 5202–5209 (2017). We thank D. Maresca, B. Ling, and A. Bar-Zion for help with SUPPLEMENTARY MATERIALS
3. J. Chu et al., Nat. Biotechnol. 34, 760–767 (2016). ultrasound imaging; N. Koulena for assistance with tissue
science.sciencemag.org/content/365/6460/1469/suppl/DC1
4. E. B. Santos et al., Nat. Med. 15, 338–344 (2009). histology; M. Abedi, J. Lee, and M. Y. You for assistance
Materials and Methods
5. D. Maresca et al., Annu. Rev. Chem. Biomol. Eng. 9, 229–252 with flow cytometry experiments; A. Collazo for confocal
Figs. S1 to S13
(2018). microscopy; C. Buser and the Oak Crest Institute of Science for
Tables S1 to S5
6. F. Pfeifer, Nat. Rev. Microbiol. 10, 705–715 (2012). cell sectioning and staining; and E. Jue and W. Chour for
References (30–31)
7. A. E. Walsby, Microbiol. Rev. 58, 94–144 (1994). assistance with initial experiments. Electron microscopy was
8. M. G. Shapiro et al., Nat. Nanotechnol. 9, 311–316 (2014). performed at the Beckman Institute Resource Center for 28 March 2019; accepted 30 August 2019
9. A. Lakshmanan et al., ACS Nano 10, 7314–7322 (2016). Transmission Electron Microscopy at Caltech. Fluorescence 10.1126/science.aax4804

Farhadi et al., Science 365, 1469–1475 (2019) 27 September 2019 6 of 6

Ultrasound imaging of gene expression in mammalian cells
Arash Farhadi, Gabrielle H. Ho, Daniel P. Sawyer, Raymond W. Bourdeau and Mikhail G. Shapiro

Science 365 (6460), 1469-1475.

DOI: 10.1126/science.aax4804

Sounding out mammalian cells

Live cell imaging allows us to observe cellular processes in real time. Most methods rely on light, and the poor
penetration of light into tissues limits their application. Ultrasound penetrates tissues, and cellular reporters that respond
to ultrasound have been developed recently. These reporters are air-filled protein structures that provide buoyancy in the
bacteria they are derived from, but when surrounded by a fluid medium, they reflect sound waves. Farhadi et al.
achieved expression from multiple genes to create these complex structures in mammalian cells. In addition to optimizing

Downloaded from http://science.sciencemag.org/ on May 25, 2021

reporter production and detection, they visualize cells in a proof-of-principle experiment in mouse tumor xenografts.
Science, this issue p. 1469

ARTICLE TOOLS http://science.sciencemag.org/content/365/6460/1469

SUPPLEMENTARY http://science.sciencemag.org/content/suppl/2019/09/25/365.6460.1469.DC1
MATERIALS

REFERENCES This article cites 31 articles, 2 of which you can access for free
http://science.sciencemag.org/content/365/6460/1469#BIBL

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Use of this article is subject to the Terms of Service

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement of
Science, 1200 New York Avenue NW, Washington, DC 20005. The title Science is a registered trademark of AAAS.
Copyright © 2019 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of
Science. No claim to original U.S. Government Works