FULL PAPER

Bone Marrow Niche-On-A-Chip www.small-journal.com

Deconstructed Microfluidic Bone Marrow On-A-Chip to

Study Normal and Malignant Hemopoietic Cell–Niche
Interactions
Julio Aleman, Sunil K. George, Samuel Herberg, Mahesh Devarasetty,
Christopher D. Porada,* Aleksander Skardal,* and Graça Almeida-Porada*

their respective BM niches have been pre-

Human hematopoietic niches are complex specialized microenvironments sumed to be similar to those of murine
that maintain and regulate hematopoietic stem and progenitor cells (HSPC). surrogates, there are few actual data
Thus far, most of the studies performed investigating alterations of HSPC- to support this assumption, and these
niche dynamic interactions are conducted in animal models. Herein, organ models have not always proven predic-
tive of results in humans.[2,3] Attaining a
microengineering with microfluidics is combined to develop a human bone
mechanistic understanding of the major
marrow (BM)-on-a-chip with an integrated recirculating perfusion system determinants of HSPC function in the
that consolidates a variety of important parameters such as 3D architecture, human BM has been difficult, due to
cell–cell/cell–matrix interactions, and circulation, allowing a better mimicry of challenges associated with visualizing
in vivo conditions. The complex BM environment is deconvoluted to 4 major and modeling the complex niche environ-
ment in humans. Given the important
distinct, but integrated, tissue-engineered 3D niche constructs housed within
biological differences that exist between
a single, closed, recirculating microfluidic device system, and equipped with human and mouse HSPC,[3] the develop-
cell tracking technology. It is shown that this technology successfully enables ment of tractable, physiologically relevant
the identification and quantification of preferential interactions—homing and models of normal and malignant human
retention—of circulating normal and malignant HSPC with distinct niches. BM niches would fill a critical gap, and is
essential to fully understand the human
hematopoietic system in health and
1. Introduction disease. Such a system would also enable the development of
therapies targeting deviant cells and/or signaling pathways.
Under normal conditions, hematopoietic stem/progenitor In addition to murine model systems, traditional in vitro 2D
cells (HSPC) reside within specific bone marrow (BM) niches. cultures have also been the foundation of countless important
These are comprised of different cell types located strategically scientific discoveries, but cells grown in 2D in traditional tissue
to provide a myriad of signals and physical interactions that culture dishes experience different surface topography, surface
maintain HSPC and orchestrate hematopoiesis. In myeloid stiffness, cell–cell/cell–matrix interactions, and a 2D versus
malignancies, the BM niche is remodeled by malignant cells, 3D architecture. As such, 2D culture systems fail to accurately
which displace HSPC, and create self-reinforcing malignant recapitulate the in vivo microenvironment, and growth under
niches that drive disease progression, drug-resistance, and 2D conditions can substantially alter the molecular and pheno-
relapse. Studies using the mouse model have led to a fairly typic properties of mammalian cells, producing experimental
detailed understanding of HSPC and their dynamic, special- outcomes that may not be indicative of what happens in vivo.[4]
ized microenvironment/niche.[1] While human HSPC and For example, we recently demonstrated that, when grown in 2D
tissue culture dishes, metastatic colon carcinoma cells exhib-
J. Aleman, Dr. S. K. George, Dr. M. Devarasetty, Dr. C. D. Porada, ited an epithelial morphology and expression profile, and it was
Dr. A. Skardal, Dr. G. Almeida-Porada only when they were transitioned into a 3D liver organoid envi-
Wake Forest Institute for Regenerative Medicine ronment that they adopted a mesenchymal and metastatic phe-
Wake Forest School of Medicine notype more reflective of their in vivo behavior.[5,6] Therefore,
391 Technology Way, Winston-Salem, NC 27101, USA
E-mail: cporada@wakehealth.edu; askardal@wakehealth.edu; 3D bioengineered platforms using human patient-derived cells
galmeida@wakehealth.edu can better mimic the cell–cell, cell-extracellular matrix (ECM),
Dr. S. Herberg and mechanical interactions of in vivo tissue, and are thus
Department of Opthamology more suitable for mechanistic research. Moreover, such plat-
State University of New York Upstate Medical University forms could one day even be deployed to improve personalized
4609 Institute for Human Performance, Syracuse, NY 13210, USA
medicine approaches in the clinic.[7] To date, only a handful
The ORCID identification number(s) for the author(s) of this article of such models have been developed that recapitulate some
can be found under https://doi.org/10.1002/smll.201902971.
aspects of the native bone and/or bone marrow microenviron-
DOI: 10.1002/smll.201902971 ment in vitro.[8–11]

Small 2019, 15, 1902971 1902971 (1 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

Herein, we combine organ microengineering[12] with micro- which to perform experiments to investigate the homing and
fluidics,[6,13,14] to develop a human bone marrow niche-on-a-chip engraftment/retention of normal and malignant human hemat-
(NOC) platform with an integrated recirculating perfusion system. opoietic stem/progenitor cells to the different cellular niche
This platform brings together a variety of important parameters constituents. To accomplish this, we sorted adult human bone
that allow a better mimicry of in vivo conditions, including 3D marrow mononuclear cells (Figure 1a) into 3 populations: Sinu-
architecture, cell–cell/cell–matrix interactions, circulation, and inte- soidal endothelial cells, arterial endothelial cells, and mesen-
gration of multiple niches/tissues. In the present studies, we have chymal stromal cells. We then differentiated a subset of MSCs
deconvoluted the complexity of the BM niche to 4 major distinct into osteoblasts to create a fourth niche population (Figure 1b).
bioengineered niche constructs that are housed within a single, Using a hyaluronic acid and gelatin-based hydrogel biofab-
closed, recirculating microfluidic device system, thus allowing rication technology (Figure 1c)–widely used in regenerative
direct manipulation and study of the impact of each individual medicine and tissue engineering[13,15,16]–the 4 niche cell popu-
niche on the homing and lodging/retention of circulating HSPC lations were then encapsulated in this ECM-derived hydrogel,
as well as the effects the distinct niches exert upon one another. forming distinct 3D niche constructs (Figure 1d), each in its
Using this human NOC platform, we showed in real-time, and for own chamber of a microfluidic device. This platform was then
5 days, the preferential interactions of infused HSPC, lymphoma, used to assess whether various healthy and malignant HSPC
and leukemic cells with periarterial (A), perisinusoidal (S), mesen- populations (Figure 1e), preferentially homed to specific niches
chymal (M), or osteoblastic (O) niches (N). Our studies establish following infusion into device (Figure 1f).
the feasibility of using microfluidic “on-a-chip” technology to rec-
reate deconstructed representations of the various niches within
the BM microenvironment, and they provide proof that this novel 2.2. Bone Marrow Niche Cell Characterization
system can be used to study the interactions of normal and malig-
nant HSPC with distinct cells of the niche. Following magnetic sorting of the cell populations used to
create each BM niche, and differentiation of MSCs into osteo-
blasts, the phenotype of each of the resulting 4 niche cell popu-
2. Results lations was characterized in 2D culture using well-established
markers for each niche/cell type.[1,2,17] Specifically, immuno-
2.1. Scientific Design fluorescence verified that the CD146+NG2+ arterial endothe-
lial cells expressed high levels of Ephrin B2 and were devoid
The overall goal of these studies was to create an ex vivo, of EphB4, while the CD146+NG2lo/- sinusoidal endothelial
deconstructed bone marrow niche-on-a-chip platform with cells expressed high levels of EphB4 but lacked expression of

Figure 1. Overall NOC experimental summary. a) Human bone marrow is b) separated into 3 niche populations (sinusoidal endothelial, arterial
endothelial, and mesenchymal) by magnetic sorting. A subset of the MSC population is differentiated to the osteoblastic lineage. c) Using an extracel-
lular matrix-mimicking hydrogel comprised of thiolated hyaluronic acid, thiolated gelatin, and a polyethylene glycol diacrylate (PEGDA) crosslinker, d)
individual niche populations are encapsulated in 3D niche constructs inside the f) NOC microfluidic device. Homing and lodging/retention studies e)
are initiated by infusing either healthy HSPC or malignant leukemia or lymphoma cells, after which homing and lodging/retention of infused cells that
have traveled through circulation to each of the niche constructs is quantified in an unbiased manner.

Small 2019, 15, 1902971 1902971 (2 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

Ephrin B2 (Figure S1a,b, Supporting Information). The MSC arterial, sinusoidal, MSC, and osteoblast niche constructs
niche population expressed both CD44 and Stro-1, two widely were formed and then maintained under circulating flow
accepted markers for MSC,[18] while the differentiated osteo- (10 µL min−1). Cells were incorporated into each niche con-
blastic niche cells showed calcium deposition, as evidenced by structs at 10 million cells mL−1 of hydrogel. At volumes of
positive staining with Alizarin Red, while MSC did not stain ≈6 µL, this would result in ≈60 000 cells per individual con-
with Alizarin Red (Figure S1c–e, Supporting Information).[19] struct. Figure 2c shows an enlarged top view of one of the niche
The distinct expression profile of each of these niche popula- constructs within the NOC and a cartoon schematic conceptu-
tions validated the successful isolation of the cell types required ally illustrating the tissue constructs under flow conditions, and
to recreate each of the four key bone marrow niches in vitro. suspended cells traveling with the recirculating media, ena-
bling them to pass by and/or interact with each of the niche
constructs.
2.3. Microfluidic Niche-On-A-Chip Device Fabrication and Niche
Construct Biofabrication In Situ
2.4. Platform Operation and Modeling
Devices were fabricated by soft lithography of a master mold
with PDMS.[20] The inlet and outlets of the negatively casted A 3D rendering of the fabricated NOC devices is shown in
hemicylindrical microfluidics PDMS layer were punched, fol- Figure 2d, and a photograph of a device under operation is
lowed by irreversible bonding to a glass slide; to form a sealed shown in Figure 2e. Devices were designed with a single inlet
fluidic device (Figure 2a). Importantly, the fabrication method port in the center of the PDMS part of the device. The con-
was designed to create hemicylindrical channels, to avoid nected inlet channel then bifurcates several times, with each
angular features that could potentially lead to trapping of cir- of the four resulting channels ending with a circular chamber,
culating cells. in which the niche construct resides. One important consid-
Following device fabrication, tissue constructs were bio- eration was achieving equivalent fluid flow in each of the four
fabricated by photopatterning-based cell encapsulation[5] arms of the device, to ensure that cells infused through the
(Figure 2b) using a hyaluronic acid (HA) and gelatin-based inlet and recirculating within the media were evenly distributed
hydrogel, commercially available as HyStem, that has been to each of the four niche constructs. This was first visualized by
employed extensively in tissue engineering and regenerative infusing dyed PBS, and confirming that the fluid was reaching
medicine applications including 3D culture,[15] tumor mod- each chamber at approximately the same time point (Movie S1,
eling,[21] bioprinting,[22] and biofabrication of organoids for Supporting Information). To more rigorously evaluate this
drug and toxicity screening.[13,21,23] We recently modified this important issue, we created a fluid dynamics computational
hydrogel system to provide faster gelation kinetics,[24] allowing model of the NOC using Flow3D software, in which a simula-
for tissue construct fabrication with significantly improved con- tion of infusing cells into the device was performed by infusing
trol over regions of hydrogel crosslinking, while maintaining 1000 roughly HSPC-sized particles. Figure 2f shows 4 images
established properties of the original non-photocrosslinkable that are screenshots of the video (Movie S2, Supporting Infor-
hydrogel. In contrast to many existing biomaterials, this system mation) of this simulation. Importantly, the simulation revealed
is comprised of naturally derived materials that are native to a heatmap of the fluid flow rates through the virtual device,
the body, namely, hyaluronic acid, which is present in high which indicated that fluid flow is approximately equal in each
concentrations in the ECM of the bone marrow.[25] Moreover, parallel region of the device. Moreover, arrows point to the loca-
this ECM hydrogel also supports the modular addition of addi- tion of the groups of infused particles, which are also present
tional factors, such as cytokines and growth factors, if desired, in approximately parallel locations within the chip at each time
via heparin-modulated binding, thus facilitating local presenta- point. These particle numbers were then quantified (Figure S2,
tion to nearby cells.[13,15,22,26] These capabilities and character- Supporting Information) using a custom-made MATLAB script
istics increase this hydrogel’s biomimetic properties compared (Code File S1 in the Supporting Information). This analysis
to other photocrosslinkable hydrogels such as those based on showed roughly equivalent numbers of infused particles were
PEGDA and methacrylated gelatin. Importantly, this hydrogel present in each bifurcation (Table S1, Supporting Information).
system supports on-demand, near-instantaneous photo- In addition, upon actual initiation of cell-based NOC device
crosslinking of discrete constructs in situ, where encapsulation studies, the physical locations of each niche type in the devices
of cells in 3D only occurs upon UV light exposure and within were rotated to avoid the possibility of introducing any position-
those regions that are exposed. This capability is not supported dependent bias to the observed homing and lodging/retention.
by more traditional gel materials that exhibit slow crosslinking
kinetics, such as collagen Type I and Matrigel, or fast, but more
difficult to control kinetics, such as alginate, making these 2.5. On-Chip Characterization of 3D Niche Constructs
materials less effective for biofabrication.[27]
Tissue construct photopatterning was performed in a step- Initially, NOC devices containing each of the 4 niche construct
wise fashion that we have previously described.[5] Figure 2b types were maintained under 10 µL min−1 flow to assess cell
shows a schematic of the NOC device and the biofabrication viability and the stability of the platform. The flow rate at which
procedure with the four patterned niche constructs, each circulating cells are transported through the device begins at
formed in an individually addressable fluidic chamber. Using 10 µL min−1. This was initially chosen to prevent a buildup of
this approach, in the sealed fluidic devices, the hydrogel-based pressure within the device, which could damage the device if

Small 2019, 15, 1902971 1902971 (3 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

Figure 2. Microfluidic chip device fabrication and niche construct integration. a) Microfluidic device fabrication is performed by bonding a molded
PDMS layer in which channels and chamber features are defined by soft lithography to a glass slide. b) In situ 3D microconstruct formation workflow:
Channels i) are filled with a mixture of photocurable hydrogel precursor, cells, and additional components (dark red, ii). A photomask (gray) is employed
to define construct shape and location iii), and the remaining solution washed away with fresh PBS iv). c) Side and top view depictions of the niche
constructs in the device chambers. d,e) Schematic and photograph, respectively, of operational recirculating multi-niche-on-a-chip systems. Scale bar:
1 cm. f) Four frames from a fluid dynamics computational model of the NOC using Flow3D software, in which a simulation of infusing HSPC into the
device was performed by infusing 1000 roughly HSPC-sized particles. Time (T) is in seconds. Heatmap represents fluid velocity (m s−1).

too high. However, when analyzing the flow rates throughout and the staining procedures described above. In parallel, via-
the device, as the channels bifurcate and flow reaches the con- bility was determined by LIVE/DEAD staining. To image the
struct chambers, the significantly increased cross-sectional area fluorophores on-chip, a macro-confocal microscope was uti-
means a significantly decreased velocity that mimics that in the lized to accommodate the 3D nature of the constructs. As can
bone marrow capillary bed.[28] Devices were maintained under be seen in Figure 3, all niche types continued to express appro-
these conditions for 8 days, after which staining was performed priate cell-specific markers within the 3D niche constructs,
directly on chip, using the antibodies to niche-specific markers with arterial constructs staining for Ephrin B2 (Figure 3a),

Small 2019, 15, 1902971 1902971 (4 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

Figure 3. Niche cell biomarkers in 3D niche constructs in NOC devices on Day 8 following construct biofabrication. a) Arterial niche constructs stained
for Ephrin B2 (B2); b) Osteoblastic niche constructs stained with Alizarin Red for calcium deposits; c) Sinusoidal niche constructs stained for EphB4
(B4); and d) mesenchymal (MSC) niche constructs stained for CD44. Panels are organized as: i) DAPI, ii) indicated stain, and iii) merged image;
e) LIVE/DEAD viability staining of each niche construct type at day 8. Green-calcein AM-stained viable cells; Red-ethidium homodimer-stained dead
cells. Scale bars: 100 µm.

Small 2019, 15, 1902971 1902971 (5 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

sinusoidal constructs staining for EphB4 (Figure 3c), mesen- microscopy, and then superimposing these images to precisely
chymal constructs staining for CD44 (Figure 3d), and osteo- visualize the localization of the labeled HSPC within the niche
blastic constructs staining positive for Alizarin Red, confirming constructs at any given point in time. An in-depth example of
calcium deposition (Figure 3b). Continued expression of these the comprehensive data that can be obtained with this approach
markers by each niche cell type was also confirmed on day 3, is shown in Figure 5, which provides a detailed overview of the
establishing the biological stability of the cells within this novel results of the homing studies. To obtain precise quantitation
platform (Figure S3, Supporting Information). Importantly, of labeled HSPC in each niche construct from these superim-
LIVE/DEAD staining over this same time period demonstrated posed images, we also developed a novel methodology utilizing
overwhelmingly high viability, with very few dead cells seen in ImageJ software, which is described in Figure S6 in the Sup-
any of the 4 niche construct types (Figure 3e). Interestingly, in porting Information. In brief, following biofabrication of each
many of the osteoblastic niche constructs, with time in culture, niche construct, a digital mask was created that corresponded
we observed fairly marked remodeling. Specifically, as the cells to the circumference of that particular construct, thereby
differentiated and deposited calcium, they also contracted to defining the region of the construct in which lodged/retained
form a tighter, more compact 3D structure, much like native HSPC could be detected (Figure S6a, Supporting Informa-
bone tissue (Figure S4, Supporting Information). tion). Next, the fluorescent and light microscopy superimposed
It should be noted here that 8 days for the total duration of image was created (Figure S6b,c, Supporting Information) at
the study (3 + 5 days after infusion of HSPC (described below), each subsequent time point during the study. The digital mask
was chosen for practical reasons. Due to the requirement of from the initial timepoint was then applied to these overlays,
closed loop circulation for the 5 days post HSPC infusion, after which the cells outside the mask were erased, and the red
spent media was not replaced with fresh media. After day 8 (or cells were transformed to white. Last, black/white thresholding
after 5 days in closed loop circulation), we began to see dete- was employed to turn the white regions into black, and a script
rioration of niche construct viabilities (Figure S5, Supporting was used to break large features into cell-sized objects for quan-
Information), which drove our time point choices. tification (Figure S6d, Supporting Information). As is detailed
in the following section, these analyses revealed that each of
the 3 populations of HSPC infused–healthy CD34+ cells, lym-
2.6. Healthy and Malignant Human HSPC Exhibit Distinct phoma cells (U937), and leukemia cells (MOLM13)–exhibited a
and Selective Homing to Specific Niches on NOC Devices marked preference for homing to and lodging within particular
niche constructs.
For all homing studies, the locations of the 4 distinct niche After each HSPC population had been infused and allowed
constructs were randomized to account for any potential bias to recirculate within the NOC device for 24 h, the first studies
of circulating cells based on the device geometry. Three days to quantify the co-localization of labeled HSPC with each of
after establishing the NOC devices, normal healthy adult the 4 niche constructs (homing) were performed. Figure 5
human bone marrow-derived CD34+ cells, human lymphoma depicts these initial homing trends. CD34+ HSPC homed with
cells (U937), or human acute myelogenous leukemia (AML) approximately equivalent proclivity to MSC, sinusoidal, and
cells (MOLM13) were fluorescently labeled with DiI fluores- osteoblastic niches, with very few cells homing to the arterial
cent membrane dye (red), and infused into the device via the niche (Figure 5a). This is in contrast to the U937 lymphoma
input port, and the cells were allowed to circulate for 5 days as cells which exhibited an increased trend in homing to the arte-
the microperistaltic pump recirculated the media. DiI-labeling rial and osteoblastic niches (Figure 5b), while the MOLM13
enabled facile tracking of the cells within the NOC devices via leukemia cells tended to home at higher numbers to the osteo-
direct visualization of red fluorescence using an on-chip micro- blastic niche (Figure 5c). It should be noted that, at this early
camera system we adapted from our previous organ-on-a-chip timepoint, these data exhibited trends, but did not achieve sta-
studies[6,21] to be compatible with the NOC platform (Figure 4a). tistical significance. However, as the studies progressed over
These camera systems consisted of an LED light source and a time and subsequent lodging/colonization of the niches took
lens that would sit on one side of the NOC, paired with a filter place, increased numbers of normal/ healthy CD34+ HSPC
and the camera on the other side of the NOC (Figure 4b). This were detected in the osteoblastic niche, while U937 and MOLM
platform could be employed to not only take individual snap- lodging/retention preferences paralleled that of early homing,
shots of the constructs, but also to capture videos in which one and these differences achieved statistical significance.
could observe, in real-time, labeled cells circulating through the Evaluation of lodging/retention at day 5 showed that CD34+
system and: 1) lodging within specific constructs (Figure 4c); HSPC lodged, to some degree, within the mesenchymal and
2) passing by the constructs without directly interacting sinusoidal niche constructs (Figure 6a,b), and while not statisti-
(Figure 4di–iii); or 3) transiently attaching to the construct, only cally significant, there was a trend to preferentially lodge in the
to then be released and re-enter the circulation (Figure 4div–vi). osteoblastic niche constructs (Figure 6c). Assessment over time
Movie S3 (Supporting Information) is a representative example showed that very few CD34+ HSCs homed to the arterial niche
of one such video that was captured. constructs, and even though small numbers of HSPC appeared
The most informative approach to quantitate the degree of to initially home within this niche, they were subsequently
homing (early migration/attachment to niche) and lodging/ released (Figure 6d). Relative quantification of the lodging/
retention (later, stable colonization of niche) of each HSPC retention of the normal/healthy HSPC within each niche con-
population within each of the niche constructs proved to be struct at day 5 is summarized in Figure 6e. U937 lymphoma
capturing images both in the fluorescent channel and by light cells lodged within the osteoblastic niche constructs (Figure 6h)

Small 2019, 15, 1902971 1902971 (6 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

Figure 4. Tracking of fluorescently labeled normal/healthy, leukemic, or lymphoma HSPC using an onboard fluorescent camera system to visualize
HSPC homing, lodging/retention, or passing by the 3D niche constructs. a) Working NOC device that is then monitored using b) a custom-built
fluorescent camera system comprised of the NOC sandwiched between an LED, lens, filter, and camera, shown in operation in the right panel. This
system allows (c,d) real-time visualization of infused labeled cells in the NOC system. c) A U937 cell is indicated (white arrows) moving toward an
arterial niche construct i–iv) and remaining in place after contact v–vi). d) A different U937 cell is indicated (white arrows) that passes around the
arterial construct i–iii), never making contact, while in panel iv), a second U937 cell is observed detaching from the construct and re-entering “circula-
tion” v–vi). Both sequences occur in ≈5–10 s time.

Small 2019, 15, 1902971 1902971 (7 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

Figure 5. Quantification of initial homing of HSPC to niches after 24 h. Cell counts at each niche for: a) normal/healthy CD34+ HSPC, b) U937 lym-
phoma cells, and c) MOLM13 leukemia cells.

and to some degree within the mesenchymal and sinusoidal device by implanting a scaffold into a mouse and allowing
niche constructs (Figure 6f,g). However, in marked contrast to it to be populated by murine cells. This construct was then
the normal HSPC, the lymphoma cells lodged/colonized the removed and placed in a microfluidic device that was then
arterial niche constructs with the highest frequency (Figure 6i). shown to support the maintenance of primitive hematopoietic
Relative quantification of the lodging/retention of the U937 cells.[11] Similarly, a study by Siebert and colleagues[9] described
cells within each niche construct at day 5 is summarized in a 3D microfluidic system to model the bone marrow, but the
Figure 6j. The last HSPC population examined, the leukemic biology of the system was significantly simplified by only
(MOLM13) cells lodged with greatest frequency within the oste- including a single niche cell type (MSC) to test the ability of
oblastic niche (Figure 6m), followed closely by the arterial niche the device to support primitive HSPC. These are all important
(Figure 6n), and lodged at lower levels within the mesenchymal studies that add to the field of microphysiological systems in
and sinusoidal niche constructs (Figure 6k,l). A summary of the the context of hematopoiesis and the bone marrow niche, yet
relative quantification of lodging/retention/colonization of each they all employed vastly different bioengineering approaches
niche type by the MOLM13 cells at day 5 appears in Figure 6o. and explored completely different biological questions than our
study.
Throughout the duration of our NOC studies, we showed
3. Discussion that this new system has the power and resolution to make
possible the direct visualization, tracking, and quantitation,
In the present studies, we developed a novel human bone in real-time, of the interactions that occur between fluores-
marrow niche-on-a-chip platform to model the in vivo com- cently-labeled normal human BM-derived HSPC, leukemic
plexity of the native human BM. Thus, we disentangled the cells (MOLM13), and lymphoma cells (U937), and each of the
BM microenvironment by breaking it down into 3D constructs 4 distinct NOCs. The microfluidic devices were designed and
representing the 4 major hematopoietic niches (N) that exist tested to provide an equal probability for interaction of these
within the BM, namely, the periarterial (A), perisinusoidal (S), three human hematopoietic cell populations with the four NOC
mesenchymal (M), and osteoblastic (O). Importantly, these 4 constructs, yet each population exhibited a marked predilection
distinct NOCs are housed within a single, closed, recirculating for homing to and subsequently lodging within specific NOC
microfluidic device, in which microchannels allow the con- constructs. Importantly, these preferences differed between
tinuous physiologic flow of human HSPC through each of the normal/healthy and malignant human HSPC, and also differed
specific niches, reproducing, in effect, a primitive circulatory depending upon the type of malignancy, such that HSPC lines
system. These NOCs contain 3D tissue constructs, sometimes from leukemia and lymphoma patients exhibited distinct pat-
referred to as organoids, fabricated at scales (200–400 µm) that terns of NOC homing and lodging/retention.
are well below the diffusion limit for nutrients, oxygen, and In addition, by employing optically-clear devices and an
small molecules,[29] allowing preservation of high viability for at onboard camera, we also showed the ability of this system to
least 8 days, and supporting the continued expression of appro- discriminate distinct phases of the interactive process between
priate phenotypic/functional markers by each distinct NOC the various niches and the infused human HSPC, as we were
throughout this time. able to clearly visualize the early stages of homing in which the
Despite the significant advancement and increased atten- flowing HSPC tethered and rolled along the specific niches, as
tion to organ-on-a-chip research in recent years, there have only well as their subsequent establishment of firm adhesive inter-
been a small number of published examples of attempts to pro- actions with the components of the niche to permanently lodge
duce an on-chip system that captures the biology of the bone for the duration of the 5-day study. We also observed brief inter-
marrow.[8–11] Importantly, these differ significantly from our actions in which the flowing HSPC transiently tethered and
platform and its intended objectives. For example, two recent rolled along a niche, only to then detach and re-enter the “circu-
studies have employed bone-based tissue chips to study cancer lation,” and ultimately lodge within/colonize a different niche.
cell migration from vasculature into bone tissue.[8] In another These observations were further supported by quantitation
study, Torisawa et al., engineered a bone marrow-on-a-chip at day 1 following cell infusion where cell–niche interactions

Small 2019, 15, 1902971 1902971 (8 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

Figure 6. Niche-on-a-chip lodging/retention experiments using CD34+ cells from normal/healthy adult donors (CD34+ HSPC), CD34- cell line derived
from a lymphoma patient (U937), CD34+ cells derived from an AML patient (MOLM13), which show distinct niche lodging/retention preferences
between cell types. a–e) Normal/healthy CD34+ HSPC located preferentially within the ON, exhibited moderate lodging/retention within the MN and
SN, and were only rarely found in the AN. f–j) The CD34- lymphoma cell line (U937) exhibited a marked predilection for the AN, followed by the ON.
k–o) CD34+ cells derived from an acute monocytic leukemia patient (MOLM13) engrafted/lodged primarily in the ON and AN, with some lodging/
retention in the MN and SN. a–n) show representative lodging/retention images with fluorescent images in the red channel highlighting the infused

Small 2019, 15, 1902971 1902971 (9 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

could be considered “homing”-based, versus quantitation at niches, and ultimately to dissect, in real-time, the molecular
day 5, by which time, cells could be considered to have truly mechanism(s) underlying the fate-dependent niche interactions
lodged within the niches. In some cases (CD34+ HSPC), early of normal and malignant human HSPC subpopulations, and
homing involved predilection to interact with both mesen- to delineate the HSPC-niche signaling pathways by which the
chymal and osteoblastic niches, but over time shifted to prefer- different niches regulate human HSPC function, under condi-
ential lodging/retention in the osteoblastic niches. In contrast, tions of both health and disease.
MOLM13 leukemia cells and U937 lymphoma cells had similar Looking specifically at myeloid malignancies, the BM niche
homing (early, day 1) and lodging/retention (late, day 5) niche is known to be profoundly remodeled by malignant cells, which
preferences, with these preferences becoming even more pro- displace resident HSPC, and create self-reinforcing malignant
nounced at later times. niches that drive disease progression and chemoresistance/
A critical aspect to validate the biological relevance of the relapse.[31,33] Therefore, the NOC platform we describe herein
NOC device/system described herein is its parity to the in can serve as a valuable tool for preclinical testing of drug effi-
vivo setting. Unfortunately, in vivo data of this type are not cacy/toxicity, and, since all cells in the NOC are human, should
available for the human system, making it impossible for us more accurately predict clinical response. We are currently
to compare the homing/lodgment patterns we observed on using this system to delineate the signaling pathways respon-
the NOC devices to what occurs in human patients receiving sible for the observed preferential HSPC/niche cell interac-
healthy HSC transplants, or human patients with leukemias/ tions, with the ultimate goal of using this knowledge to develop
lymphoma. However, multiple elegant experiments have been more effective treatments for hematological malignancies and
done in murine model systems in which healthy or leukemic enhance engraftment following HSPC transplant. Specifically,
human cells/cell lines have been infused and imaged using defining the differences within normal versus malignant
high resolution MRI and both real-time and “temporal snap- niches that alter the interactions with HSPC could lead to novel
shot” confocal microscopy to track the homing of these infused therapies targeting the deviant cells and/or signaling in the
human hematopoietic cells.[30,31] These studies have collec- malignant niche; a significant advance over current drugs to
tively shown that: 1) following infusion into mice, normal/ disease-specific driver mutations. Furthermore, repopulating the
healthy human HSC home to both the perivascular (sinusoidal NOCs with patient-derived cells could ultimately pave the way
endothelium and/or MSC/pericytes) and osteoblastic/endosteal for safer and more effective personalized medicine approaches
niches/regions; 2) HSC do not home to nor engraft the arte- in the clinic.
rial niches to any significant degree following infusion; and 3)
acute myelogenous leukemia cells (the same type of leukemia
as the MOLT-13 cells we employed herein as a model for leu- 4. Experimental Section
kemia) home almost exclusively to the endosteal (osteoblastic)
Study Design: The objective of these studies was to create an ex
niches/zones following infusion into mice. Since these patterns vivo, deconstructed bone marrow niche-on-a-chip platform with which
parallel those that we observed with our novel NOC platform, to perform experiments to investigate the homing and engraftment/
we are hopeful that the results we have obtained with this new retention of normal and malignant human hematopoietic stem/
in vitro system are likely to be predictive of aspects of the in progenitor cells to the different cellular niche constituents. All cells
vivo setting. employed in these studies were purchased commercially. All data
presented is based on studies in triplicate or higher.
An additional consideration that bears mention is the
Cell Sorting and Differentiation: Frozen human bone marrow
utility of this 3D in vitro system versus traditional 2D culture mononuclear cells (BMNC) were obtained from Stem Cell Technologies
approaches. One could argue that, in a more straightforward (Lot # 1507160920, harvested from a 28 year old male Caucasian, 77 kg,
experiment, a 2D control might be an appropriate control and 170 cm, nonsmoker). Thawed BMNC were divided into two aliquots for
a desired component of the study. However, we have published Stro-1 and CD146+ isolation using a MiniMACS Separator kit (Miltenyi
extensively in a variety of tissue types and pathologies (liver, Biotec, Auburn, CA) according to manufacturer’s instructions. Briefly,
cardiac, pancreas, cancer, etc.) showing that 3D in vitro models one aliquot of BMNC were labeled with IgM anti-human Stro-1 and
incubated with anti-rat IgM microbeads. The other fraction of BMNC was
functionally mimic in vivo tissue more accurately than their 2D labeled with a phycoerythrin (PE)-conjugated antibody to CD146 (Becton
counterparts.[6,13,15,23,32] Since the marrow microenvironment in Dickinson ImmunoSystems, San Jose, CA), followed by incubation
vivo exists in three dimensions, like all other tissues, we do not with anti-PE Multisort microbeads (Miltenyi Biotec), as previously
feel that including a 2D analog of our 3D system in our experi- described.[18] The Stro-1+ and CD146+ populations were then obtained
ments would be of significant value nor add to the validity of by magnetic sorting using an MS magnetic column (Miltenyi Biotec),
the data being presented. and the Stro-1+ mesenchymal stromal cells (MSC) were directly cultured
in mesenchymal cell growth medium (MSCGM; Lonza, Walkersville,
We thus feel that the data provided herein provide compel- MD) in fibronectin-coated flasks.
ling evidence of not only the feasibility, but also the validity, The CD146+ fraction of BMNC was then further sorted into NG2
of using microfluidics “niche-on-a-chip” technology as a tool positive (arterial endothelial cells; AEC) or negative (sinusoidal
to study the interactions of distinctive HSPC with different endothelial cells; SEC) populations by incubating the cells with

cell types overlaid on light microscopy images of each niche. Images were taken on day 1, day 3, and day 5 following infusion of the cells into each
NOC device. e,j,o) provide quantified average cell numbers of lodged/retained CD34+ HSPC, U937 lymphoma cells, and MOLM13 leukemia cells,
respectively, in each niche. Red highlighted borders indicate regions of common lodging/retention. One-way ANOVA (n = 10) **P < 0.005,***P < 0.001.
Scale bar: 250 µm.

Small 2019, 15, 1902971 1902971 (10 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

an allophycocyanin (APC)-conjugated anti-NG2 antibody (Novus buffer, pH 6.0. Nonspecific binding was blocked using Protein Block
Biologicals, Littleton, CO), followed by anti-APC microbeads and MS (Abcam, Cambridge, MA) for 1 h at room temperature. Slides were
magnetic column sorting. The CD146+NG2+ and CD146+NG2− cells incubated overnight at 4 °C in a humidified chamber with the following
were each cultured in MSCGM (Lonza) in fibronectin-coated flasks. primary antibodies, diluted in Antibody Diluent Reagent Solution (Life
To form the osteoblastic niche, an aliquot of the Stro-1+ MSC were Technologies; Thermo Fisher Scientific): rabbit anti-human Ephrin B2
induced to undergo osteogenic differentiation using the StemPro (NBP1-84830; Novus Biologicals, Littleton, CO), mouse anti-human
Osteogenesis Differentiation Kit, in accordance with the manufacturer’s EphB4 (37-1800; Thermo Fisher Scientific, Waltham, MA), sheep anti-
directions (Life Technologies, Grand Island, NY). In brief, at 50% mouse/rat/porcine/equine CD44 (AF6127; R&D Systems, Minneapolis,
confluence, MSCGM media was removed from the MSC and exchanged MN), and mouse anti-human Stro-1 (MAB1038; R&D Systems,
for freshly prepared osteocyte/chondrocyte differentiation basal media Minneapolis, MN). Following washing, sections were incubated with
with supplements. Partially differentiated osteoblasts were then Alexa Fluor 488- or 594- or 647- conjugated secondary antibodies (Life
harvested at day 3–4 and used for chip integration. Technologies; Thermo Fisher Scientific) for 1 h at room temperature in
3D Construct Biofabrication: The stromal constructs were fabricated a humidified chamber. Following washing, samples were counterstained
“off-chip” by encapsulating the Arterial, Sinusoidal, Osteoblast, or with 4′,6-diamidino-2-phenylindole (DAPI; Life Technologies; Thermo
mesenchymal (MSC) niche cells in thiol-modified hyaluronic acid-based Fisher Scientific) for 5 min, washed, and cover-slipped with ProLong
hydrogel (ESI-BIO, Alameda, USA), in separate wells. ECM-based HA/ Gold anti-fade mounting medium (Invitrogen; Thermo Fisher Scientific).
gelatin hydrogels were formed using HyStem-HP (ESI-BIO, Alameda, Controls included identical slides stained in parallel, in which the primary
CA). The thiolated HA component (Heprasil) and the thiolated gelatin antibody was either absent or was replaced by a nonspecific isotype-
component (Gelin-S) were dissolved in water containing 0.5% w/v of matched primary antibody conjugated to the respective Alexa-Fluor.
the photoinitiator 4-(2-hydroxyethoxy)phenyl-(2-propyl)ketone (Sigma St. To determine the presence of calcium deposition, a group of slides
Louis, MO) to make 1% w/v solutions. The polyethylene glycol diacrylate were stained with 40 × 10−3 m Alizarin Red S (ARS; Sigma-Aldrich,
crosslinker (Extralink, ESI-BIO) was dissolved in the photoinitiator St. Louis, MO), pH 4.1 for 5 min. Full osteoblastic commitment of
solution to make a 2% w/v solution. Heprasil, Gelin-S, and Extralink the differentiated MSC was assessed by treatment with tetracycline,
were then mixed in a 2:2:1 ratio by volume. The resulting solution used observing the resultant fluorescence staining of hydroxyapatite. In brief,
to resuspend cells at a cell density of 10 million cells mL−1. A 6 µL 40 µg mL−1 of tetracycline (Sigma-Aldrich, St. Louis, MO) was prepared
volume of the hydrogel precursor/cell mixture was placed on top of a 6 in MSCGM culture medium or StemPro Osteogenesis Differentiation
mm PDMS puck and exposed to UV light for 0.25 s (365 nm, 7 W cm−2) Kit medium, and the cells/constructs to be tested were then incubated
to crosslink, followed by washing with PBS, and addition of serum-free with this solution for 72 h, after which they were scored/imaged for
QBSF-60 Stem Cell Medium (Quality Biological, Gaithersburg, MD). At fluorescence.
volumes of ≈6 µL, this would result in ≈60 000 cells per individual niche All light and fluorescence microscopy images were captured using
construct. an Olympus BX63 microscope (Olympus, Center Valley, PA) equipped
Microfluidic Platform Fabrication: The microfluidic platform was with an X-Cite 120 LED Boost fluorescence lamp (Excelitas Technologies,
built using PDMS (10:1 Sylgard 184 silicone elastomer and curing Waltham, MA) and an Olympus DP80 dual camera (Olympus).
agent respectively, Dow Corning) and a glass slide (VWR, Radnor, Fluidic Dynamics and Cellular Infusion: To ensure homogeneous
PA); both treated and irreversibly bounded with O2 plasma. A convex distribution of the infused cells into the four downstream chambers,
hemicylindrical microfluidic channel (120 µm height) was fabricated transient 3D simulations were performed using computational fluid
using a modified previously reported standard photolithography dynamics software, (FLOW3D v11.2, Flow Sciences, Santa Fe, NM).
technique. Briefly, a primary SU-8 2050 (MicroChem, Westborough, MA) Mass-density particle species were assigned with random rate of
wafer was generated by standard lithography. A primary PDMS device generation with fully coupled particle-fluid interaction. The inlet
was generated by molding the square channel then the channels were source fluid flow rate was specified with four outlet sinks to simulate
sealed with a 40 µm PDMS membrane. The membrane in the channels a continuous loop system. Particle counts and lifespan were tracked
was deformed into the channels by a vacuum negative pressure system, using measuring flux planes and sample volumes within the software for
SU-8 was placed into them and also spin-coated evenly on a glass slide. quantification and visual analysis.
Both SU-8 covered surfaces were bind and UV-cured from the top. A After fabrication, niche constructs were allowed to equilibrate/
final PDMS channel mold was generated for the microfluidic layer. stabilize for three days, after which, a closed loop system was set up
Individual Niche Construct Biofabrication On-Chip: The in situ patterned in a four-channel precision micro peristaltic pump (Elemental Scientific,
stromal niche constructs were based on a modified reported technique. Omaha, NE). Previous to infusion, U937, MOLM13 (both from ATCC,
Chambers were filled with Arterial, Sinusoidal, mesenchymal (MSC), or Manassas, VA), and healthy CD34+ cells (StemCell Technologies,
Osteoblast cells encapsulated in thiol-modified hyaluronic acid based Vancouver, BC, Canada) were independently labeled with the Qtracker
hydrogel (ESI-BIO, Alameda, USA), in their respective chamber. A 605 Cell Labeling Kit (Life Technologies, Carlsbad, CA) following the
photomask with four equidistant 500 µm diameter circles was aligned in manufacturer’s instructions. In short, a 10 × 10−9 m labeling solution
the center of the chambers and exposed to UV light for 0.25 s (7 W cm−2) was prepared from component A and B. The solution was diluted in
to crosslink, washed with PBS to remove un-crosslinked solution, and 0.2 mL of fresh complete growth medium. The cells were incubated in
placed in serum-free QBSF-60 Stem Cell Medium (Quality Biological, the solution for 45 min and washed twice with media; 1 × 106 cells mL−1
Gaithersburg, MD) (Figure 2b). in QBSF-60 were put into a fresh reservoir and bundled into the system.
Assessment and Characterization of Niches: To confirm that the On the day of infusion, a close loop system was set up, and the labeled
individual niches exhibited the appropriate phenotype, both when U937, MOLM13, and healthy CD34+ cells were independently perfused
cultured in 2D and within the 3D constructs, immunofluorescent into the NOC device at a flow rate of 9.5 µL min−1 (Figure 2b). Upon
staining was performed at days 3, 7, and 14. Briefly, cells or tissue infusion, cells enter a main channel, and then evenly disperse into the
constructs were fixed in 4% paraformaldehyde at room temperature and four equidistant chambers, each of which houses a specific stromal
maintained at 4° in PBS until processed. Cells were cultured and stained niche (Figure 2b,d).
on coverslips, and off-chip constructs were processed by standard Unbiased Quantification of Engrafted Cells: Brightfield (BF)/Qtracker
paraffin-embedding; 5 µm sections were cut using a microtome (Leica stacked images in TIFF format were captured and processed in ImageJ
Microsystems Inc., Buffalo Grove, IL). Niche construct sections were in the following manner: 1) Using oval or elliptical tools, the stromal
deparaffinized and rehydrated with decreasing concentrations of construct area was selected. 2) Using Image ¦ Duplicate, a new image
ethanol. All samples were permeabilized in 0.1% Triton X100 for 5 min. was generated, and the surrounding area of the construct was cleared
Heat-induced epitope retrieval was performed using an automated (Clear outside). This generated “mask” was utilized on each remaining
Dako PT Link system (Agilent, Santa Clara, CA) with sodium citrate correspondent stromal construct. 3) Using Image ¦ Color ¦ Split

Small 2019, 15, 1902971 1902971 (11 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

Channels, a new image with the red labeled signals was separated from [6] A. Skardal, M. Devarasetty, S. Forsythe, A. Atala, S. Soker, Bio-
the composite BF/Qtracker image. 4) Using Image ¦ Adjust ¦ Threshold, technol. Bioeng. 2016, 113, 2020.
the red signal’s borders were defined and colored black with a white [7] a) A. R. Mazzocchi, S. A. P. Rajan, K. I. Votanopoulos, A. R. Hall,
background. 5) Using Analyze ¦ Analyze Particles, a constant size A. Skardal, Sci. Rep. 2018, 8, 2886; b) K. I. Votanopoulos, P. Shen,
inclusion filter (25–225 µm2) with a circularity of 0.0–1.0 was applied to A. Skardal, E. A. Levine, Surg. Oncol. Clin. North Am. 2018, 27, 551;
ensure that debris, noise, and bright stromal cells were excluded. c) K. I. Votanopoulos, A. Mazzocchi, H. Sivakumar, S. Forsythe,
Statistical Analysis: Statistical analyses were performed using J. Aleman, E. A. Levine, A. Skardal, Ann. Surg. Oncol. 2019, 26, 139.
statistical analysis software (GraphPad Prism 7, GraphPad Software Inc., [8] a) S. Bersini, J. S. Jeon, G. Dubini, C. Arrigoni, S. Chung,
USA). One-way ANOVA was employed for multiple comparisons. P <
J. L. Charest, M. Moretti, R. D. Kamm, Biomaterials 2014, 35, 2454;
0.01 or less was considered statistically significant. Data are presented
b) A. Marturano-Kruik, M. M. Nava, K. Yeager, A. Chramiec, L. Hao,
as means plus/minus standard deviation, and all experiments were
S. Robinson, E. Guo, M. T. Raimondi, G. Vunjak-Novakovic, Proc.
performed with n ≥ 4.
Natl. Acad. Sci. USA 2018, 115, 1256.
[9] S. Sieber, L. Wirth, N. Cavak, M. Koenigsmark, U. Marx, R. Lauster,
M. Rosowski, J. Tissue Eng. Regener. Med. 2018, 12, 479.
Supporting Information [10] Y. S. Torisawa, T. Mammoto, E. Jiang, A. Jiang, A. Mammoto,
A. L. Watters, A. Bahinski, D. E. Ingber, Tissue Eng., Part C 2016, 22,
Supporting Information is available from the Wiley Online Library or 509.
from the author. [11] Y. S. Torisawa, C. S. Spina, T. Mammoto, A. Mammoto, J. C. Weaver,
T. Tat, J. J. Collins, D. E. Ingber, Nat. Methods 2014, 11, 663.
[12] a) D. Huh, G. A. Hamilton, D. E. Ingber, Trends Cell Biol. 2011,
Acknowledgements 21, 745; b) A. M. Ghaemmaghami, M. J. Hancock, H. Harrington,
H. Kaji, A. Khademhosseini, Drug Discovery Today 2012, 17, 173.
J.A. and S.K.G. contributed equally to this work. A.S. acknowledges [13] A. Skardal, S. V. Murphy, M. Devarasetty, I. Mead, H. W. Kang,
funding from the Wake Forest Baptist Medical Center Clinical and Y. J. Seol, Z. Y. S. , S. R. Shin, L. Zhao, J. Aleman, A. R. Hall,
Translational Science Institute Open Pilot Program via NIH CTSA T. Hartung, A. Khademhosseini, S. Soker, C. E. Bishop, A. Atala, Sci.
UL1 TR001420. GAP acknowledges funding from NIH R21HL117704 Rep. 2017, 7, 8837.
and G.A.P and C.D.P are supported by NHLBI R01HL135853 and [14] J. Aleman, A. Skardal, Biotechnol. Bioeng. 2019, 116, 936.
R01HL130856. J.A. would like to thank Joel Jacob for providing advice [15] A. Skardal, L. Smith, S. Bharadwaj, A. Atala, S. Soker, Y. Zhang, Bio-
and assistance on the construction and implementation of the
materials 2012, 33, 4565.
microfluidics simulation.
[16] G. D. Prestwich, J. Control Release 2011, 155, 193.
[17] a) S. Gronthos, A. C. Zannettino, S. J. Hay, S. Shi, S. E. Graves,
A. Kortesidis, P. J. Simmons, J. Cell Sci. 2003, 116, 1827; b) S. Pinho,
Conflict of Interest T. Marchand, E. Yang, Q. Wei, C. Nerlov, P. S. Frenette, Dev. Cell
2018, 44, 634; c) T. M. Nguyen, A. Arthur, S. Gronthos, Int. J.
The authors declare no conflict of interest. Hematol. 2016, 103, 145.
[18] S. Mokhtari, E. Colletti, W. Yin, C. Sanada, Z. Lamar, P. J. Simmons,
S. Walker, C. Bishop, A. Atala, E. D. Zanjani, C. D. Porada,
G. Almeida-Porada, Leukemia 2018, 32, 1575.
Keywords [19] L. E. Sima, Curr. Stem Cell Res. Ther. 2017, 12, 139.
bone marrow niche, hematopoietic stem and progenitor cells, homing, [20] D. Qin, Y. Xia, G. M. Whitesides, Nat. Protoc. 2010, 5, 491.
microfluidics, tissue chip [21] M. Devarasetty, S. Forsythe, T. D. Shupe, S. Soker, C. E. Bishop,
A. Atala, A. Skardal, Biosensors (Basel) 2017, 7, e24.
Received: June 6, 2019 [22] A. Skardal, M. Devarasetty, H. W. Kang, I. Mead, C. Bishop,
Revised: July 31, 2019 T. Shupe, S. J. Lee, J. Jackson, J. Yoo, S. Soker, A. Atala, Acta Bio-
Published online: August 29, 2019 mater. 2015, 25, 24.
[23] S. D. Forsythe, M. Devarasetty, T. D. Shupe, C. E. Bishop, A. Atala,
S. Soker, A. Skardal, Front. Public Health 2018, unpublished.
[24] A. Skardal, M. Devarasetty, H. W. Kang, Y. J. Seol, S. D. Forsythe,
[1] a) M. Acar, K. S. Kocherlakota, M. M. Murphy, J. G. Peyer, H. Oguro, C. Bishop, T. Shupe, S. Soker, A. Atala, J. Visualized Exp. 2016, 110,
C. N. Inra, C. Jaiyeola, Z. Zhao, K. Luby-Phelps, S. J. Morrison, e53606.
Nature 2015, 526, 126; b) L. Ding, T. L. Saunders, G. Enikolopov, [25] D. N. Haylock, S. K. Nilsson, Regener. Med. 2006, 1, 437.
S. J. Morrison, Nature 2012, 481, 457; c) Y. Kunisaki, I. Bruns, [26] A. Skardal, S. V. Murphy, K. Crowell, D. Mack, A. Atala, S. Soker,
C. Scheiermann, J. Ahmed, S. Pinho, D. Zhang, T. Mizoguchi, J. Biomed. Mater. Res., Part B Appl Biomater. 2017, 105, 1986.
Q. Wei, D. Lucas, K. Ito, J. C. Mar, A. Bergman, P. S. Frenette, [27] S. R. Caliari, J. A. Burdick, Nat. Methods 2016, 13, 405.
Nature 2013, 502, 637; d) A. Mendelson, P. S. Frenette, Nat. Med. [28] K. P. Ivanov, M. K. Kalinina, I. Levkovich Yu, Microvasc. Res. 1981,
2014, 20, 833; e) S. J. Morrison, D. T. Scadden, Nature 2014, 505, 22, 143.
327. [29] M. Lovett, K. Lee, A. Edwards, D. L. Kaplan, Tissue Eng., Part B
[2] G. M. Crane, E. Jeffery, S. J. Morrison, Nat. Rev. Immunol. 2017, 17, 2009, 15, 353.
573. [30] a) I. Beerman, T. C. Luis, S. Singbrant, C. Lo Celso,
[3] P. A. Horn, B. M. Thomasson, B. L. Wood, R. G. Andrews, S. Mendez-Ferrer, Exp. Hematol. 2017, 50, 22; b) N. E. Bengtsson,
J. C. Morris, H. P. Kiem, Blood 2003, 102, 4329. S. Kim, L. Lin, G. A. Walter, E. W. Scott, Leukemia 2011, 25, 1223;
[4] W. J. Ho, E. A. Pham, J. W. Kim, C. W. Ng, J. H. Kim, D. T. Kamei, c) M. G. Bixel, A. P. Kusumbe, S. K. Ramasamy, K. K. Sivaraj,
B. M. Wu, Cancer Sci. 2010, 101, 2637. S. Butz, D. Vestweber, R. H. Adams, Cell Rep. 2017, 18, 1804;
[5] A. Skardal, M. Devarasetty, S. Soker, A. R. Hall, Biofabrication 2015, d) S. W. Lane, Y. J. Wang, C. Lo Celso, C. Ragu, L. Bullinger,
7, 031001. S. M. Sykes, F. Ferraro, S. Shterental, C. P. Lin, D. G. Gilliland,

Small 2019, 15, 1902971 1902971 (12 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.small-journal.com

D. T. Scadden, S. A. Armstrong, D. A. Williams, Blood 2011, 118, [31] S. W. Lane, D. T. Scadden, D. G. Gilliland, Blood 2009, 114, 1150.
2849; e) I. H. Oh, S. Y. Jeong, J. A. Kim, Curr. Opin. Hematol. 2019, [32] a) A. Mazzocchi, M. Devarasetty, S. Herberg, W. J. Petty, F. Marini,
26, 249; f) A. Sanchez-Aguilera, S. Mendez-Ferrer, Cell. Mol. Life Sci. L. D. Miller, G. L. Kucera, D. D. K., J. Ruiz, A. Skardal, S. Soker,
2017, 74, 579; g) K. Schepers, T. B. Campbell, E. Passegue, Cell Stem ACS Biomater. Sci. Eng. 2019, 5, 1937; b) A. Skardal, M. Devarasetty,
Cell 2015, 16, 254; h) D. A. Sipkins, X. Wei, J. W. Wu, J. M. Runnels, C. Rodman, A. Atala, S. Soker, Ann. Biomed. Eng. 2015, 43,
D. Cote, T. K. Means, A. D. Luster, D. T. Scadden, C. P. Lin, Nature 2361.
2005, 435, 969; i) K. H. Susek, E. Korpos, J. Huppert, C. Wu, [33] a) A. Colmone, M. Amorim, A. L. Pontier, S. Wang, E. Jablonski,
I. Savelyeva, F. Rosenbauer, C. Muller-Tidow, S. Koschmieder, D. A. Sipkins, Science 2008, 322, 1861; b) D. S. Krause,
L. Sorokin, Matrix Biol. 2018, 67, 47. D. T. Scadden, F. I. Preffer, Cytometry, Part B 2013, 84, 7.

Small 2019, 15, 1902971 1902971 (13 of 13) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim