Liu and Li Journal of Biomedical Science (2020) 27:3

https://doi.org/10.1186/s12929-019-0594-x

REVIEW Open Access

Photoacoustic imaging of cells in a three-

dimensional microenvironment
Wei-Wen Liu1 and Pai-Chi Li1,2*

Abstract
Imaging live cells in a three-dimensional (3D) culture system yields more accurate information and spatial
visualization of the interplay of cells and the surrounding matrix components compared to using a two-
dimensional (2D) cell culture system. However, the thickness of 3D cultures results in a high degree of scattering
that makes it difficult for the light to penetrate deeply to allow clear optical imaging. Photoacoustic (PA) imaging is
a powerful imaging modality that relies on a PA effect generated when light is absorbed by exogenous contrast
agents or endogenous molecules in a medium. It combines a high optical contrast with a high acoustic
spatiotemporal resolution, allowing the noninvasive visualization of 3D cellular scaffolds at considerable depths with
a high resolution and no image distortion. Moreover, advances in targeted contrast agents have also made PA
imaging capable of molecular and cellular characterization for use in preclinical personalized diagnostics or PA
imaging-guided therapeutics. Here we review the applications and challenges of PA imaging in a 3D cellular
microenvironment. Potential future developments of PA imaging in preclinical applications are also discussed.
Keywords: Photoacoustic imaging, Biomedical imaging, Three-dimensional cell culture, Tumor microenvironment

Introduction on protease in a dense collagen fibrillar meshwork with

In the past few decades, the conventional 2D cell cultures size-limiting pores [9, 10, 12–15]. A similar model for
have remarkably increased the knowledge in basic cell biol- tumor cell migration in the stromal ECM has also been de-
ogy and preclinical biomedical applications. However, cells scribed [13, 15]. Tumor cell intravasation and extravasation
cultured in a 2D monolayer lack a typical 3D architecture. through the vascular wall to other organs is a critical step
Moreover, cells inhabiting a rigid surface without a 3D elas- of tumor migration and metastasis [16]. 3D cell cultures
tic fibrous meshwork (i.e., the extracellular matrix [ECM]) have been utilized in biomimetic models of the tumor vas-
cannot normally response the physical or biochemical cues culature or angiogenesis for investigating the intrinsic or
from the surrounding physiological matrix substrate [1–5]. extrinsic modulation of the tumor vascular niche [17–20].
Recently, in efforts to target the tumor microenvironment Preclinical studies of the normalization of tumor vascula-
for improving both the effectiveness and efficiency of can- ture or drug screening for anti-angiogenesis have also been
cer therapeutics, several studies such as immunotherapy, performed [21, 22]. 3D culture models can recapitulate par-
tumor vasculature, or ECM remodeling has been led to a tial physiological diversity and thereby allow for dissecting
new era and yielded novel insights [6–8]. For example, ex- underlying regulatory mechanisms into separate units
travasated T lymphocytes infiltrated into the stromal ECM within a controllable microenvironment. The examinations
for migrating to the targeted tumor sites has been demon- performed in these studies may help to improve therapeutic
strated in 3D cell culture models [9–11]. An amoeboid interventions and inform clinical decisions.
shape change and contact guidance during T-cell locomo- The availability of appropriate imaging modalities for
tion in 3D collagen fibrils has been documented as a analyzing cell dynamics within 3D cell culture systems
protease-independent process, but it should be dependent can facilitate interpretations and precise quantification.
High-resolution imaging systems such as laser-scanning
* Correspondence: [email protected] confocal microscopy, electron microscopy, and multi-
1
Graduate Institute of Biomedical Electronics and Bioinformatics, National
Taiwan University, Taipei, Taiwan photon microscopy are usually employed by researchers
2
Department of Electrical Engineering, National Taiwan University, Taipei, to quantify and track cell behaviors. However, 3D cell
Taiwan

© The Author(s). 2020 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Liu and Li Journal of Biomedical Science (2020) 27:3 Page 2 of 9

cultures are usually thick and exhibit strong light scat- using a single- or multiwavelength laser. Conventional
tering, which results in the impinging light experiencing optical imaging using contrast agents with emitted fluor-
severe diffraction and diffusion. One method for acquir- escence or bioluminescence, which typically can be im-
ing images of high quality during live cell imaging is to aged with spatial resolution and the imaging depth in
culture cells on the surface of a thin layer of 3D ECM micrometer or sub-micrometer scale. By taking the ad-
matrix gel (also named a 2.5D culture) or a cell-laden vantage of the laser-based PA principles, photons can be
3D ECM matrix gel with a reduced thickness. Micro- converted into ultrasonic waves in biological samples.
scope objective lenses with numerical apertures (NA) Because of acoustic wave can travel through scattering
necessary for high-resolution imaging have very small tissue much far than photon does, PA imaging tech-
working distances, such as 100–200 μm for lenses with niques can surpass the depth limitation of optical im-
magnifications above 60×. When using an inverted aging systems. To provide a practical guide for choosing
microscope, the thickness of the holder (e.g., coverslip the appropriate technologies to examine the 3D struc-
or polydimethylsiloxane layer) supporting the 3D cell tural or functional information of biomaterials, cellular
culture will reduce the working distance to 0–50 μm, behaviors, and cell–biomaterial interactions, we compare
and so the deepest visible site of the 3D gel will be only the properties of the most widely used imaging modal-
50 μm from the bottom surface of the holder. However, ities to that of PA imaging modality (Table 1). As such,
when the thickness of a hydrogel is less than 50 μm, a we summarized the scalability, the chemical sensitivity,
finite-element model simulated that the hydrogel stress and the potential applications of acoustic imaging, op-
field around the indenter actually interacted with the tical imaging, PA imaging, and electron imaging. Among
rigid bottom support, leading to a stiffer response [23]. these techniques, PA imaging can achieve better spatial
Experimental evidence further proves that the cell aspect resolution than acoustic imaging, and its imaging depth
ratio, area, and migration speed are significantly in- can be larger than optical imaging and electron imaging.
creased in hydrogel with a height of less than 200 μm In this review, we first briefly outline the importance of
due to the mechanical gradient along its height [23]. using 3D cell cultures as novel physiological mimicry
Moreover, although yes-associated protein (YAP) has platforms, and then discuss the current challenges in
been documented as a key factor to mediate cancer pro- optic-based imaging of 3D cell cultures for the
gression through mechanotransduction, a recent report characterization of cell–biomaterial interactions. Since
challenges the established knowledge that the breast PA imaging can potentially obtain images at greater
cancer progression is regulated by YAP-dependent depths, we describe the physical background into how
mechanotransduction in 2.5D culture model, which is, PA imaging works and the principles of the two main
the independency of YAP in ECM stiffness-mediated PA imaging modalities. Combining PA imaging with the
breast cancer progression is found in 3D cultures and use of multiplex contrast agents makes it possible to
patient samples [24]. Therefore, data obtained from 2.5D monitor interactions between cells and 3D scaffolds.
cultures and thin cell-laden 3D cultures should be exam- Since most 3D cell cultures have no endogenous con-
ined carefully. Obtaining detailed information about the trast agents, the application of exogenous contrast
center region of 3D cell cultures usually requires the agents in 3D cell cultures will be more focused in this
biochemical processing of gel fixation following by thin review. Finally, we draw conclusions about the current
sections of embedded gels to produce samples whose bottlenecks and the future outlook on expanding the
structural, histological, or protein expression patterns capabilities of PA imaging through the use of multimod-
can be investigated using optical imaging systems. Re- ality and unconventional imaging toolkits.
grettably, these processing methods can cause gel de-
formation or damage, the loss of localized enzymes and Review
metabolite profiles, and alterations to cell dynamics and Fundamentals of PA imaging
chemical and nutrient gradients. PA imaging is based on the physical integration of optical ir-
To address the problem of deep imaging, PA imaging radiation and ultrasonic detection (Fig. 1) [25–27]. Irradiating
as a noninvasive and hybrid imaging modality that com- light-absorbing materials with a short-pulse laser induces an
bines optical excitation and ultrasonic detection to attain increase in pressure through thermoelastic expansion. The
better spatial resolution than traditional ultrasound (US) resulting pressure waves can be interpreted to US waves as
imaging and also achieve deeper penetration than purely the pressure wavefront propagates in the light-absorbing re-
optical imaging systems. PA imaging is a powerful im- gion. The US waves, also known as PA waves, can be de-
aging technique that can provide scalable and multicon- tected by US transducers to produce electrical signals. These
trast images of 3D cell culture scaffolds, ranging from signals are then amplified, digitized, decoded, and transferred
single cells to an organoid culture. Furthermore, both to a computer for image formation. The amplitude of the PA
structural and functional information can be obtained response is proportional to the concentration of the
Liu and Li Journal of Biomedical Science (2020) 27:3 Page 3 of 9

Table 1 Comparison of properties of the imaging modalitiesa

Imaging Spatial Imaging Chemical Potential applications
modality resolution depth sensitivity
EM 1 nm 0.1 μm N.A. Biomineralization, biomaterial surface structure, pore formation, cell–biomaterial
interaction
CM 0.4 μm 0.1 mm High Cell behavior, cell tracking, gene expression, cellular substructure, functional
activity
MPM 1 μm 1 mm High Biomaterial architecture, cell tracking, functional activity
OR-PAM 2.5 μm 1 mm High Cell tracking, functional activity, fluid dynamics, cell–biomaterial interaction
OCT 10 μm 2 mm Low Vascularization, biomaterial architecture, functional activity
AR-PAM 45 μm 3 mm High Vascularization, biomaterial architecture, cell metabolism, functional activity
US (20 MHz) 165 μm 3 cm Low Fluid dynamics, mechanics, spatial patterning of cells, functional activity
a
Abbreviations: EM electron microscopy, CM confocal microscopy, MPM multi-photon microscopy, OR-PAM optical resolution photoacoustic microscopy, OCT optical
coherence tomography, AR-PAM acoustic resolution photoacoustic microscopy, US ultrasound imaging, N.A. not available

absorbers, the optical absorption coefficient of the photoab- image with volumetric information to be generated.
sorber, and the thermal coefficient of volume expansion. The Because the degree of scattering is much lower for US
contrast of PA imaging when imaging in vivo or in vitro than for visible light in biological samples, PAM pro-
samples can be improved by utilizing the various available vides a better spatial resolution and a deeper penetra-
PA contrast agents as photoabsorbers such as hemoglobin tion depth [34, 35].
and gold nanoparticles [28–33]. The axial resolution and the achievable penetration
depth of PAM are determined by the central frequency
Photoacoustic microscopy of the US transducer. The axial resolution is equal to
Photoacoustic microscopy (PAM) is one type of PA half the spatial pulse width, and a higher operating fre-
imaging modality that aims to image at millimeter- quency has a smaller wavelength and hence generates
scale depths and micrometer-scale resolutions. Its shorter pulses, giving a better axial resolution. The lat-
microscopic essence is appropriate for visualizing eral resolution of PAM is determined by the combined
structural, functional, or molecular information such response of the point source from overlapping optical
as property alterations of a scaffold, cellular dynamics, excitation and acoustic detection by the PAM imaging
or engineered vasculature and angiogenesis inside 3D- system, known as the point spread function. Depending
scaffold-based samples. During PAM scanning, each on what directs the resolution of the imaging system,
laser-pulse-generated time-resolved PA signal re- PAM can be further categorized into optical-resolution
corded from the US transducer is converted into one- PAM (OR-PAM) and acoustic-resolution PAM (AR-
dimensional depth-resolved image (A-line) based on PAM) (Fig. 2). In OR-PAM, the optical focus is better
the sound velocity in the sample, and A-line scanning than the acoustic focus and a lateral resolution of a few
is used to form a 2D frame. Coupling this with 2D micrometers can be achieved, allowing for single-cell im-
raster scanning along the horizontal plane allows a 3D aging. Nonetheless, high optical scattering limits the

Fig. 1 Illustration of PA signal generation. Optical energy excited from a short-pulse laser is absorbed by optical absorbers, which causes an
increase in the local temperature. An US pressure wave, the so-called PA signal, is then generated by the thermal expansion of the absorber
Liu and Li Journal of Biomedical Science (2020) 27:3 Page 4 of 9

Fig. 2 Schematics of two types of PAM system: (a) OR-PAM and (b) AR-PAM. In this setup, 3D tumor spheres labeled with contrast agents are
cultured in a cuboidal matrix hydrogel for PA imaging. Note that the laser light is focused in OR-PAM but unfocused in AR-PAM, respectively.
Once the laser energy is delivered into the 3D cell culture and absorbed by endogenous or exogenous contrast agents, the absorbed energy is
converted into heat, leading to thermal expansion. Ultrasound signals are then generated and detected by the transducer located at the top of
the samples

penetration depth to around 1 mm in OR-PAM. In AR- are usually with optical absorption spectra in the near-
PAM, the acoustic focus is much better than the optical infrared (NIR) window (600–1100 nm) so as to ensure
focus, and a lateral resolution of a few tens of microme- their deeper penetration and hence the required imaging
ters can be achieved. The relatively weak acoustic scat- depth.
tering in AR-PAM allows a penetration depth of up to a For multiscale biological systems, several kinds of repre-
few centimeters, which enables investigations of pheno- sentative PAM systems are summarized based on the scal-
typic characteristics in a 3D configuration. In both OR- able imaging performance shown in Fig. 3 [44]. Generally,
PAM and AR-PAM, using objectives with low NA makes AR-PAM (i.e. using unfocused laser beam) can achieve im-
it possible to image a large field of view without sacri- aging depth beyond 1 mm, in contrast, OR-PAM can only
ficing the depth resolution. achieve imaging depth within 1 mm due to the limited pene-
tration of a focused laser beam. AR-PAM with a focused 50-
Monitoring a 3D microenvironment using PA contrast MHz ultrasound detector can provide lateral resolution of
agents and multiscale PAM 45 μm and axial resolution of 15 μm for detection of oxygen
PA contrast agents can be categorized into two types, saturation in a single blood vessel over 1 mm beneath the
endogenous and exogenous. Two well-known endogen- tissue surface [39]. The imaging depth can be extended to 4
ous contrast agents applied in in vivo label-free PA im- cm and the lateral resolution is enlarged to 100–560 μm
aging are melanin and hemoglobin. Except for cells when reducing the center frequency of the focused ultra-
containing melanin, PA contrast is usually undetectable sound detector to 5 MHz for macroscopic purpose [38].
in biomaterials and the cell-laden 3D culture, and so an Real-time imaging and the deeper penetration depth up to
exogenous contrast agent needs to be introduced for 7 cm can be achieved when using an ultrasound transducer
contrast enhancement. Exogenous contrast agents for array as the detector combined with a computed tomog-
use in PA molecular imaging have to possess certain raphy scanning system [37, 45]. OR-PAM for imaging cells
photophysical and biological properties, such as efficient has the lateral resolution of 1–5 μm, and the axial resolution
optical-to-PA conversion, long-lived excited-state life- can achieve to ~ 15 μm when combination with a 75-MHz
time, biocompatibility, distinct optical absorption spectra focused ultrasound detector [40] and it can be improved to
(where the endogenous contrast agents have a lower ab- 7.5 μm when using focused ultrasonic detector with a center
sorption), and the capability to pass through cellular and frequency of 125 MHz [41]. Combination with objectives
fibrillar barriers for successful labeling [29, 36]. Further- with a higher NA and sub-diffraction techniques, the lateral
more, both endogenous and exogenous contrast agents resolution of OR-PAM can be increased to 87 to 220 nm to
Liu and Li Journal of Biomedical Science (2020) 27:3 Page 5 of 9

Fig. 3 Scalability of PAM among multiscale biological systems. The blue circles denote lateral resolution, and green circles denote axial resolution.
Solid lines denote OR-PAMs, and dotted lines denote AR-PAMs. LA-PACT, linear-array PA computed tomography [37]; PAMac, PA macroscopy [38];
AR-PAM, acoustic resolution PAM [39]; OR-PAM, optical resolution PAM [40]; 125-MHz-PAM, PAM using a 125-MHz ultrasound detector [41]; SW-
PAM, subwavelength resolution PAM [42]; PI-PAM, photoimprint PAM [43]. Figure adapted from [44]

achieve the purpose for imaging organelle [42, 43]. Follow- or ex vivo functional imaging or to observe flow dynam-
ing sections will draw on the biomedical applications of PA ics in a fluid system, working with AR-PAM can achieve
imaging based on the properties of PA contrast agents in- deeper imaging up to few millimeters and spatial reso-
cluding probing functional biological processes, structural lution of 45–200 μm, but OR-PAM can provide cellular
imaging of biomaterial scaffolds and vasculature, cell track- level information such as intracellular calcium imaging
ing, and tumor detection in 3D microenvironments. Among in 3D cell culture systems.
these studies, to achieve PA imaging at the single-cell scale,
OR-PAM can be used, and AR-PAM can be used to achieve Structural imaging of 3D scaffolds/tissues
deeper penetration and tissue-scale imaging in in vivo ani- After implanting 3D engineered porous scaffolds into
mals/human studies. mice ears, neovascularization in the implanted scaffolds
could be noninvasively monitored and quantified using
Functional imaging of 3D cell cultures/tissues both AR-PAM and OR-PAM for up to 6 weeks [54]. Poly-
The change in the absorption spectra between oxyhemo- mer porous 3D scaffolds incorporating carbon nanotubes
globin and deoxyhemoglobin enabled the total concen- or 3D-printed alginate-polydopamine scaffolds can be
tration of hemoglobin and the oxygen saturation in the used for structural examinations of the 3D scaffolds with
rodent brain vasculature or tumor angiogenesis to be de- PA imaging [55, 56]. In these studies, to visualize the net-
tected in multiwavelength PAM [37, 45–47]. Gold nano- work of capillaries (5–10 μm) and 3D engineered porous
particles (AuNPs) are exemplar PA contrast agents that scaffolds, OR-PAM is used to achieve spatial resolution of
provide unique possibilities for both in vitro and in vivo 5 μm and axial resolution of 15 μm, but with a limited im-
molecular PA imaging. For example, AuNPs have been aging depth (1 mm). OR-PAM provides well-resolved im-
administered to blood vessels for blood flow velocity ages allowing quantification of the characteristics of the
measurements in chicken breast tissue [48, 49] and for 3D scaffolds such as pore size, porosity, or fiber formation,
monitoring the intravascular fluid pathway of the rat and AR-PAM provides a thicker image layer up to 2 mm.
brain [50]. The use of NIR-responsive PA dyes for func-
tional PA calcium imaging in in vitro 3D cell cultures Cell tracking and tumor cells detection in 3D cell cultures/
and in vivo animal models, including arsenazo III, chlor- tissues
ophosphonazo III, and genetically encoded calcium indi- Melanin, a naturally produced pigment in melanoma cells,
cators, has also been documented [51–53]. For in vivo provides good optical and PA contrast in melanoma relative
Liu and Li Journal of Biomedical Science (2020) 27:3 Page 6 of 9

to the surrounding tissue, and allowed for tracing the mel- guided tumor therapies for the purpose of theranostics.
anoma cells and monitoring the melanoma growth for 2 The use of both the PAM and US modalities provides
weeks [57]. This property means that melanoma cells are anatomical and functional information [32, 80–83]. Con-
traceable for monitoring cell proliferation in engineered 3D trast agents in multimodality imaging systems can en-
porous scaffolds [58]. Making use of cellular endocytosis hance the contrast in two or more modalities. For
processes, AuNPs can be loaded into stem cells or macro- example, the position of the sentinel lymph node can be
phages as a PA contrast agent, which opens the possibility displayed using US imaging, with PA imaging used to
of the long-term tracking and monitoring of stem cells or display the accumulation of methylene blue [83]. Com-
macrophages in a 3D fibrin or gelatin scaffold through bined PA and US imaging with PA contrast agents can
multimodal US and PA imaging for utilization in investiga- be further applied in image-guided photothermal ther-
tions of stem cell therapy [59–65]. Nanoparticles are gener- apy [52, 71, 72]. An US system could be used to monitor
ally more likely to accumulate in a tumor lesion due to the the targeting of AuNPs-encapsulated microbubbles, with
enhanced permeability and retention of the leaky tumor PA imaging used to monitor the US-assisted delivery of
blood vessels [66], which has been demonstrated by the AuNPs at the tumor lesion [66]. Moreover, phase-shifted
passive targeting and accumulation of AuNPs at a tumor droplets can be used as the contrast agent to enhance
site [67]. For tracking cells or delivering the contrast agent the contrast of combined US and PA imaging and also
to specific regions in order to reduce off-target effects, the therapeutic effects [28, 84, 85]. These previous stud-
strategies for conjugating the targeting ligands such as anti- ies have mainly relied on an optical droplet vaporization
bodies, peptides, and aptamers with contrast agents for ac- mechanism, and deep explorations of the underlying
tive targeting have been developed. AuNPs with molecular physics are now required to further optimize these tech-
targeting ability such as those conjugated with antibodies niques. The potential bioeffects should also be deter-
recognized to tumor protein biomarkers, and Arg-Gly-Asp mined to ensure safety. A very recent phantom study
(RGD) peptide are also commonly applied for tumor detec- employed the cancer drug doxorubicin as a PA contrast
tion in vivo in PA imaging [68–73]. Exploring the crosstalk agent has shed more light on tumor theranostics [86].
between stromal ECM and T cells is important for the cor- Further phantom and in vitro 3D cell culture validations
responding immunotherapy strategies. T cells that have should be performed to improve these methods with
taken up AuNPs or can be loaded in an in vitro 3D hydro- consideration of the tissue complexity before moving to
gel for tracking individual T cells when migrating to tumor- clinical applications.
spheres with OR-PAM [74, 75]. T cells labeled with NIR- Another aspect of PA imaging in a 3D cell microenvir-
797-isothiocyanate (an NIR PA and fluorescent dye) can be onment that needs further work is improving the im-
applied to imaging the dynamic change of T cells in lymph aging frame rate with the aim of achieving real-time
nodes in an in vivo mouse model by using AR-PAM [76]. functional applications, especially in thick 3D scaffolds.
One interesting application of using PA exogenous For example, acoustic-lens-based PA imaging [87, 88]
contrast agents is detecting the PA signals of matrix and optical US mapping [89] open up new possibilities
metalloproteinase-2 (MMP-2) in follicular thyroid can- to increase the imaging speed, spatial resolution, and
cer [77]. MMP-2 is abundant in several kinds of tumor field of view. Finally, quantitative studies for standardiz-
cells and is known to be closely associated with tumor ing preclinical applications are also important for trans-
progression and metastasis [78]. MMP-2 can be targeted lating the present results to the clinic.
by a modified activatable cell-penetrating peptide that is
labeled with two chromophores exhibiting different op- Conclusions
tical absorption wavelengths: BHQ-3 (675 nm) and Alexa PA imaging has been investigated in preclinical stud-
Fluor 750 (750 nm) [79]. Both chromophores can be de- ies over the past decade. This review has described
tected photoacoustically. Once MMP-2 is cleaved, only the current state of PA imaging, focusing on the ap-
the dye with the BHQ3-labeled cell-penetrating part of plication of PA imaging techniques to a 3D cellular
the probe accumulates in the cells, and the location of microenvironment. PA imaging provides a better
the cleaved probe is observable after background sub- penetration depth and can yield both structural and
traction. These synthesized contrast agents were used to functional information of 3D biological samples from
noninvasively detect the location of follicular thyroid the single-cell level to the organoid level. Combining
cancer in a mouse model by using AR-PAM [77] and a multiwavelength laser with the use of contrast
may be used in 3D tumor culture model as well. agents can produce multicontrast images. Hence, PA
imaging has been developed as a powerful tool to dis-
Bottlenecks and future prospects sect the mechanisms underlying spatiotemporal devel-
To expand the capabilities of multimodality imaging, opment in preclinical studies. However, it is difficult
PAM could be combined with US imaging in image- to compare the results obtained from different 3D cell
Liu and Li Journal of Biomedical Science (2020) 27:3 Page 7 of 9

culture systems and PA imaging systems due to the 10. Hartmann N, Giese NA, Giese T, Poschke I, Offringa R, Werner J, et al.
wide range of the in-house systems that are available. Prevailing role of contact guidance in intrastromal T-cell trapping in human
pancreatic cancer. Clin Cancer Res. 2014;20:3422–33.
Future works will focus on quantitative studies by 11. Salmon H, Franciszkiewicz K, Damotte D, Dieu-Nosjean MC, Validire P,
using various types of PA imaging systems to achieve Trautmann A, et al. Matrix architecture defines the preferential localization
standardization of each biological characteristic in dif- and migration of T cells into the stroma of human lung tumors. J Clin
Invest. 2012;122:899–910.
ferent 3D cell culture samples. 12. Wolf K, Muller R, Borgmann S, Brocker EB, Friedl P. Amoeboid shape change
and contact guidance: T-lymphocyte crawling through fibrillar collagen is
Abbreviations independent of matrix remodeling by MMPs and other proteases. Blood.
2D: Two-dimensional; 3D: Three-dimensional; AR-PAM: Acoustic-resolution 2003;102:3262–9.
microscopy; AuNPs: Gold nanoparticles; CM: confocal microscopy; 13. Wolf K, Te Lindert M, Krause M, Alexander S, Te Riet J, Willis AL, et al. Physical
ECM: Extracellular matrix; EM: electron microscopy; MMP-2: Matrix limits of cell migration: control by ECM space and nuclear deformation and
metalloproteinase-2; MPM: multi-photon microscopy; NIR: Near-infrared;
tuning by proteolysis and traction force. J Cell Biol. 2013;201:1069–84.
OCT: optical coherence tomography.; OR-PAM: Optical-resolution microscopy;
14. Kuczek DE, Larsen AMH, Thorseth ML, Carretta M, Kalvisa A, Siersbaek MS,
PA: Photoacoustic; RGD peptide: Arg-Gly-Asp peptidePAMPhotoacoustic
et al. Collagen density regulates the activity of tumor-infiltrating T cells. J
microscopy; US: Ultrasound; YAP: Yes-associated protein
Immunother Cancer. 2019;7:68.
15. Edsparr K, Basse PH, Goldfarb RH, Albertsson P. Matrix metalloproteinases in
Acknowledgements
cytotoxic lymphocytes impact on tumour infiltration and
Not applicable.
immunomodulation. Cancer Microenviron. 2011;4:351–60.
16. Folkman J. Role of angiogenesis in tumor growth and metastasis. Semin
Authors’ contributions
Oncol. 2002;29(6 Suppl 16):15–8.
WWL drafted this article. PCL supervised the work and gave final approval of
17. Oh S, Ryu H, Tahk D, Ko J, Chung Y, Lee HK, et al. “open-top” microfluidic
submitted/revised versions. Both authors have read and approved the final
device for in vitro three-dimensional capillary beds. Lab Chip. 2017;17:3405–14.
manuscript.
18. Paek J, Park SE, Lu Q, Park KT, Cho M, Oh JM, et al. Microphysiological
engineering of self-assembled and Perfusable microvascular beds for the
Funding
production of vascularized three-dimensional human microtissues. ACS
This work was supported by National Health Research Institutes (NHRI-EX107-
Nano. 2019;13:7627–43.
10624EI) and Ministry of Science and Technology (MOST 106–2221-E-002-001).
19. Chen MB, Lamar JM, Li R, Hynes RO, Kamm RD. Elucidation of the roles of
Availability of data and materials tumor integrin beta1 in the extravasation stage of the metastasis Cascade.
Not applicable. Cancer Res. 2016;76:2513–24.
20. Zervantonakis IK, Hughes-Alford SK, Charest JL, Condeelis JS, Gertler FB,
Ethics approval and consent to participate Kamm RD. Three-dimensional microfluidic model for tumor cell
Not applicable. intravasation and endothelial barrier function. Proc Natl Acad Sci U S A.
2012;109:13515–20.
Consent for publication 21. Amann A, Zwierzina M, Koeck S, Gamerith G, Pechriggl E, Huber JM, et al.
No applicable. Development of a 3D angiogenesis model to study tumour - endothelial cell
interactions and the effects of anti-angiogenic drugs. Sci Rep. 2017;7:2963.
Competing interests 22. Park JS, Kim IK, Han S, Park I, Kim C, Bae J, et al. Normalization of tumor
The author declares that he has no competing interests. vessels by Tie2 activation and Ang2 inhibition enhances drug delivery and
produces a favorable tumor microenvironment. Cancer Cell. 2016;30:953–67.
Received: 30 July 2019 Accepted: 18 November 2019 23. Rao SS, Bentil S, DeJesus J, Larison J, Hissong A, Dupaix R, et al. Inherent
interfacial mechanical gradients in 3D hydrogels influence tumor cell
behaviors. PLoS One. 2012;7:e35852.
References 24. Lee JY, Chang JK, Dominguez AA, Lee HP, Nam S, Chang J, et al. YAP-
1. Duval K, Grover H, Han LH, Mou Y, Pegoraro AF, Fredberg J, et al. Modeling independent mechanotransduction drives breast cancer progression. Nat
physiological events in 2D vs. 3D cell culture. Physiology (Bethesda). 2017; Commun. 2019;10:1848.
32:266–77. 25. Su JL, Wang B, Wilson KE, Bayer CL, Chen YS, Kim S, et al. Advances in
2. Gillet JP, Calcagno AM, Varma S, Marino M, Green LJ, Vora MI, et al. clinical and biomedical applications of Photoacoustic imaging. Expert Opin
Redefining the relevance of established cancer cell lines to the study of Med Diagn. 2010;4:497–510.
mechanisms of clinical anti-cancer drug resistance. Proc Natl Acad Sci U S 26. Wang LV, Hu S. Photoacoustic tomography: in vivo imaging from organelles
A. 2011;108:18708–13. to organs. Science. 2012;335:1458–62.
3. Katt ME, Placone AL, Wong AD, Xu ZS, Searson PC. In vitro tumor models: 27. Beard P. Biomedical photoacoustic imaging. Interface Focus. 2011;1:602–31.
advantages, disadvantages, variables, and selecting the right platform. Front 28. Fernandes DA, Kolios MC. Intrinsically absorbing photoacoustic and
Bioeng Biotechnol. 2016;4:12. ultrasound contrast agents for cancer therapy and imaging.
4. Liu Y, Chen YG. 2D- and 3D-based intestinal stem cell cultures for Nanotechnology. 2018;29:505103.
personalized medicine. Cells. 2018;7:225. 29. Weber J, Beard PC, Bohndiek SE. Contrast agents for molecular
5. Yamada KM, Cukierman E. Modeling tissue morphogenesis and cancer in photoacoustic imaging. Nat Methods. 2016;13:639–50.
3D. Cell. 2007;130:601–10. 30. Yao J, Wang L, Yang JM, Maslov KI, Wong TT, Li L, et al. High-speed label-
6. Chen Q, Liu G, Liu S, Su H, Wang Y, Li J, et al. Remodeling the tumor free functional photoacoustic microscopy of mouse brain in action. Nat
microenvironment with emerging Nanotherapeutics. Trends Pharmacol Sci. Methods. 2015;12:407–10.
2018;39:59–74. 31. Hai P, Imai T, Xu S, Zhang R, Aft RL, Zou J, et al. High-throughput, label-free,
7. Di Modugno F, Colosi C, Trono P, Antonacci G, Ruocco G, Nistico P. 3D single-cell photoacoustic microscopy of intratumoral metabolic
models in the new era of immune oncology: focus on T cells, CAF and heterogeneity. Nat Biomed Eng. 2019;3:381–91.
ECM. J Exp Clin Cancer Res. 2019;38:117. 32. Mallidi S, Luke GP, Emelianov S. Photoacoustic imaging in cancer detection,
8. Song HH, Park KM, Gerecht S. Hydrogels to model 3D in vitro microenvironment diagnosis, and treatment guidance. Trends Biotechnol. 2011;29:213–21.
of tumor vascularization. Adv Drug Deliv Rev. 2014;79–80:19–29. 33. Jathoul AP, Laufer J, Ogunlade O, Treeby B, Cox B, Zhang E, et al. Deep in vivo
9. Bougherara H, Mansuet-Lupo A, Alifano M, Ngo C, Damotte D, Le Frere-Belda photoacoustic imaging of mammalian tissues using a tyrosinase-based genetic
MA, et al. Real-time imaging of resident T cells in human lung and ovarian reporter. Nat Photonics. 2015;9:239.
carcinomas reveals how different tumor microenvironments control T 34. Zhou Y, Yao J, Wang LV. Tutorial on photoacoustic tomography. J Biomed
lymphocyte migration. Front Immunol. 2015;6:500. Opt. 2016;21:61007.
Liu and Li Journal of Biomedical Science (2020) 27:3 Page 8 of 9

35. Wang LV. Tutorial on Photoacoustic microscopy and computed 60. Li W, Chen X. Gold nanoparticles for photoacoustic imaging. Nanomedicine
tomography. IEEE J Sel Top Quantum Electron. 2008;14:171–9. (Lond). 2015;10:299–320.
36. Hatamimoslehabadi M, Bellinger S, La J, Ahmad E, Frenette M, Yelleswarapu 61. Jokerst JV, Thangaraj M, Kempen PJ, Sinclair R, Gambhir SS. Photoacoustic
C, et al. Correlation of Photophysical properties with the Photoacoustic imaging of mesenchymal stem cells in living mice via silica-coated gold
emission for a selection of established Chromophores. J Phys Chem C. 2017; nanorods. ACS Nano. 2012;6:5920–30.
121:24168–78. 62. Kubelick KP, Snider EJ, Ethier CR, Emelianov S. Development of a stem cell
37. Kim C, Erpelding TN, Jankovic L, Pashley MD, Wang LV. Deeply penetrating tracking platform for ophthalmic applications using ultrasound and
in vivo photoacoustic imaging using a clinical ultrasound array system. photoacoustic imaging. Theranostics. 2019;9:3812–24.
Biomed Opt Express. 2010;1:278–84. 63. Ricles LM, Hsieh PL, Dana N, Rybalko V, Kraynak C, Farrar RP, et al. Therapeutic
38. Song KH, Wang LV. Deep reflection-mode photoacoustic imaging of assessment of mesenchymal stem cells delivered within a PEGylated fibrin gel
biological tissue. J Biomed Opt. 2007;12:060503. following an ischemic injury. Biomaterials. 2016;102:9–19.
39. Zhang HF, Maslov K, Stoica G, Wang LV. Functional photoacoustic 64. Nam SY, Ricles LM, Suggs LJ, Emelianov SY. In vivo ultrasound and
microscopy for high-resolution and noninvasive in vivo imaging. Nat photoacoustic monitoring of mesenchymal stem cells labeled with gold
Biotechnol. 2006;24:848–51. nanotracers. PLoS One. 2012;7:e37267.
40. Maslov K, Zhang HF, Hu S, Wang LV. Optical-resolution photoacoustic 65. Dhada KS, Hernandez DS, Suggs LJ. In vivo Photoacoustic tracking of
microscopy for in vivo imaging of single capillaries. Opt Lett. 2008;33:929–31. Mesenchymal stem cell viability. ACS Nano. 2019;13:7791–9.
41. Zhang C, Maslov K, Yao J, Wang LV. In vivo photoacoustic microscopy with 66. Matsumura Y, Maeda H. A new concept for macromolecular therapeutics in
7.6-μm axial resolution using a commercial 125-MHz ultrasonic transducer. J cancer chemotherapy: mechanism of tumoritropic accumulation of proteins
Biomed Opt. 2012;17:116016. and the antitumor agent smancs. Cancer Res. 1986;46:6387–92.
42. Zhang C, Maslov K, Wang LV. Subwavelength-resolution label-free photoacoustic 67. Srivatsan A, Jenkins SV, Jeon M, Wu Z, Kim C, Chen J, et al. Gold nanocage-
microscopy of optical absorption in vivo. Opt Lett. 2010;35:3195–7. photosensitizer conjugates for dual-modal image-guided enhanced
43. Yao J, Wang L, Li C, Zhang C, Wang LV. Photoimprint photoacoustic photodynamic therapy. Theranostics. 2014;4:163–74.
microscopy for three-dimensional label-free subdiffraction imaging. Phys 68. Nie L, Wang S, Wang X, Rong P, Ma Y, Liu G, et al. In vivo volumetric
Rev Lett. 2014;112:014302. photoacoustic molecular angiography and therapeutic monitoring with
44. Yao J, Wang LV. Sensitivity of photoacoustic microscopy. Photoacoustics. targeted plasmonic nanostars. Small. 2014;10:1585–93.
2014;2:87–101. 69. Li PC, Wang CR, Shieh DB, Wei CW, Liao CK, Poe C, et al. In vivo
45. Gamelin J, Maurudis A, Aguirre A, Huang F, Guo P, Wang LV, et al. A real- photoacoustic molecular imaging with simultaneous multiple selective
time photoacoustic tomography system for small animals. Opt Express. targeting using antibody-conjugated gold nanorods. Opt Express. 2008;16:
2009;17:10489–98. 18605–15.
46. Chen SL, Burnett J, Sun D, Wei X, Xie Z, Wang X. Photoacoustic microscopy: 70. Cheng K, Kothapalli SR, Liu H, Koh AL, Jokerst JV, Jiang H, et al. Construction
a potential new tool for evaluation of angiogenesis inhibitor. Biomed Opt and validation of nano gold tripods for molecular imaging of living
Express. 2013;4:2657–66. subjects. J Am Chem Soc. 2014;136:3560–71.
47. Ku G, Wang X, Xie X, Stoica G, Wang LV. Imaging of tumor angiogenesis in 71. Wang YH, Chen SP, Liao AH, Yang YC, Lee CR, Wu CH, et al. Synergistic
rat brains in vivo by photoacoustic tomography. Appl Opt. 2005;44:770–5. delivery of gold nanorods using multifunctional microbubbles for enhanced
48. Liao CK, Huang SW, Wei CW, Li PC. Nanorod-based flow estimation plasmonic photothermal therapy. Sci Rep. 2014;4:5685.
using a high-frame-rate photoacoustic imaging system. J Biomed Opt. 72. Wang YH, Liao AH, Chen JH, Wang CR, Li PC. Photoacoustic/ultrasound
2007;12:064006. dual-modality contrast agent and its application to thermotherapy. J
49. Wei CW, Huang SW, Wang CR, Li PC. Photoacoustic flow measurements Biomed Opt. 2012;17:045001.
based on wash-in analysis of gold nanorods. IEEE Trans Ultrason Ferroelectr 73. Agarwal A, Huang SW, O’Donnell M, Day KC, Day M, Kotov N, et al. Targeted
Freq Control. 2007;54:1131–41. gold nanorod contrast agent for prostate cancer detection by
50. Wang Y, Xie X, Wang X, Ku G, Gill KL, O'Neal DP, et al. Photoacoustic photoacoustic imaging. J Appl Phys. 2007;102:064701.
tomography of a Nanoshell contrast agent in the in vivo rat brain. Nano 74. Lee PY, Liu WW, Chen SC, Li PC. Dual-wavelength optical-resolution
Lett. 2004;4:1689–92. photoacoustic microscopy for cells with gold nanoparticle bioconjugates in
51. Dean-Ben XL, Gottschalk S, Sela G, Shoham S, Razansky D. Functional three-dimensional cultures. Proc SPIE. 2016. https://doi.org/10.1117/12.2214037.
optoacoustic neuro-tomography of calcium fluxes in adult zebrafish brain 75. Huang RX, Fu Y, Liu W, Ma YT, Hsieh B-Y, Chen SC, et al. Dual-wavelength
in vivo. Opt Lett. 2017;42:959–62. OR-PAM with compressed sensing for cell tracking in a 3D cell culture
52. Liu W-W, Chen S-H, Li P-C. Functional calcium imaging using optical- system. Proc SPIE. 2018. https://doi.org/10.1117/12.2289975.
resolution photoacoustic microscopy in a 3D tumor cell culture. Proc SPIE. 76. Zheng S, Li H, Lai K, Chen M, Fu G, Liu WH, et al. Noninvasive photoacoustic
2019. https://doi.org/10.1117/12.2510936. and fluorescent tracking of optical dye labeled T cellular activities of
53. Dana N, Fowler RA, Allen A, Zoldan J, Suggs L, Emelianov S. In vitro diseased sites at new depth. J Biophotonics. 2018;11:e201800073.
photoacoustic sensing of calcium dynamics with arsenazo III. Laser Phys 77. Levi J, Kothapalli SR, Bohndiek S, Yoon JK, Dragulescu-Andrasi A, Nielsen C,
Lett. 2016;13:075603. et al. Molecular photoacoustic imaging of follicular thyroid carcinoma. Clin
54. Cai X, Zhang Y, Li L, Choi SW, MacEwan MR, Yao J, et al. Investigation of Cancer Res. 2013;19:1494–502.
neovascularization in three-dimensional porous scaffolds in vivo by a 78. Coussens LM, Fingleton B, Matrisian LM. Matrix metalloproteinase inhibitors
combination of multiscale photoacoustic microscopy and optical coherence and cancer: trials and tribulations. Science. 2002;295:2387–92.
tomography. Tissue Eng Part C Methods. 2013;19:196–204. 79. Levi J, Kothapalli SR, Ma TJ, Hartman K, Khuri-Yakub BT, Gambhir SS. Design,
55. Cai X, Paratala BS, Hu S, Sitharaman B, Wang LV. Multiscale photoacoustic synthesis, and imaging of an activatable photoacoustic probe. J Am Chem
microscopy of single-walled carbon nanotube-incorporated tissue Soc. 2010;132:11264–9.
engineering scaffolds. Tissue Eng Part C Methods. 2012;18:310–7. 80. Kolkman RG, Brands PJ, Steenbergen W, van Leeuwen TG. Real-time in vivo
56. Luo Y, Wei X, Wan Y, Lin X, Wang Z, Huang P. 3D printing of hydrogel photoacoustic and ultrasound imaging. J Biomed Opt. 2008;13:050510.
scaffolds for future application in photothermal therapy of breast cancer 81. Wilson K, Homan K, Emelianov S. Biomedical photoacoustics beyond
and tissue repair. Acta Biomater. 2019;92:37–47. thermal expansion using triggered nanodroplet vaporization for
57. Staley J, Grogan P, Samadi AK, Cui H, Cohen MS, Yang X. Growth of contrast-enhanced imaging. Nat Commun. 2012;3:618.
melanoma brain tumors monitored by photoacoustic microscopy. J Biomed 82. Harrison T, Ranasinghesagara JC, Lu H, Mathewson K, Walsh A, Zemp RJ.
Opt. 2010;15:040510. Combined photoacoustic and ultrasound biomicroscopy. Opt Express. 2009;
58. Zhang Y, Cai X, Choi SW, Kim C, Wang LV, Xia Y. Chronic label-free 17:22041–6.
volumetric photoacoustic microscopy of melanoma cells in three- 83. Erpelding TN, Kim C, Pramanik M, Jankovic L, Maslov K, Guo Z, et al. Sentinel
dimensional porous scaffolds. Biomaterials. 2010;31:8651–8. lymph nodes in the rat: noninvasive photoacoustic and US imaging with a
59. Chung E, Nam SY, Ricles LM, Emelianov SY, Suggs LJ. Evaluation of gold clinical US system. Radiology. 2010;256:102–10.
nanotracers to track adipose-derived stem cells in a PEGylated fibrin gel for 84. Chen Q, Yu J, Kim K. Review: optically-triggered phase-transition droplets for
dermal tissue engineering applications. Int J Nanomedicine. 2013;8:325–36. photoacoustic imaging. Biomed Eng Lett. 2018;8:223–9.
Liu and Li Journal of Biomedical Science (2020) 27:3 Page 9 of 9

85. Liu WW, Huang SH, Li PC. Synchronized optical and acoustic droplet
vaporization for effective Sonoporation. Pharmaceutics. 2019;11:279.
86. Kimm MA, Gross C, Dean-Ben XL, Ron A, Rummeny EJ, Lin HA, et al. Optoacoustic
properties of doxorubicin - a pilot study. PLoS One. 2019;14:e0217576.
87. He Y, Tang Z, Chen Z, Wan W, Li J. A novel photoacoustic tomography
based on a time-resolved technique and an acoustic lens imaging system.
Phys Med Biol. 2006;51:2671–80.
88. Dogra VS, Chinni BK, Valluru KS, Moalem J, Giampoli EJ, Evans K, et al.
Preliminary results of ex vivo multispectral photoacoustic imaging in the
management of thyroid cancer. AJR Am J Roentgenol. 2014;202:W552–8.
89. Zhang E, Laufer J, Beard P. Backward-mode multiwavelength
photoacoustic scanner using a planar Fabry-Perot polymer film
ultrasound sensor for high-resolution three-dimensional imaging of
biological tissues. Appl Opt. 2008;47:561–77.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.