ll

Leading Edge

Primer
Bioprinting for the Biologist
Andrew C. Daly,1,3 Margaret E. Prendergast,1,3 Alex J. Hughes,1,2 and Jason A. Burdick1,*
1Department of Bioengineering, University of Pennsylvania, Philadelphia, PA 19104, USA
2Department of Cell and Developmental Biology, University of Pennsylvania, Philadelphia, PA 19104, USA
3These authors contributed equally

*Correspondence: burdick2@seas.upenn.edu
https://doi.org/10.1016/j.cell.2020.12.002

SUMMARY

Building tissues from scratch to explore entirely new cell configurations could revolutionize fundamental
understanding in biology. Bioprinting is an emerging technology to do this. Although typically applied to en-
gineer tissues for therapeutic tissue repair or drug screening, there are many opportunities for bioprinting
within biology, such as for exploring cellular crosstalk or cellular morphogenesis. The overall goals of this
Primer are to provide an overview of bioprinting with the biologist in mind, outline the steps in extrusion bio-
printing (the most widely used and accessible technology), and discuss alternative bioprinting technologies
and future opportunities for bioprinting in biology.

INTRODUCTION TO BIOPRINTING The range of bioprinting technologies available to biomedical re-

searchers is broad. The most common and accessible method is
Biological questions are traditionally investigated either with that of extrusion bioprinting, where the pressure-driven extrusion
cells seeded on two-dimensional (2D) hard surfaces (e.g., tis- of a bioink from a printer nozzle (sometimes referred to as printer
sue culture plates, glass) or with animal models. 2D cultures head) is used to print filaments with a user-defined design
can be explored with human cells but are non-physiological (Figure 1A) (Ozbolat and Hospodiuk, 2016). Inkjet printing falls
with regards to biophysical properties and cellular organiza- under the umbrella of extrusion bioprinting and involves the depo-
tion, whereas animal models may be challenging to implement sition of bioink droplets through a nozzle rather than as continuous
and monitor spatiotemporal cell behavior, particularly with hu- filaments (Li et al., 2020). In contrast, lithography bioprinting
man relevance (Benam et al., 2015; Ingber, 2020). Three- methods have also emerged where light is used to spatially pattern
dimensional (3D) cultures where cells are encapsulated and a cell-laden hydrogel resin (bioresin) into 3D constructs (Groll et al.,
cultured within soft materials are seeing increased use, as 2016; Sun et al., 2020). These techniques offer improved resolution
exemplified by Matrigel, which has particularly improved the when compared to extrusion bioprinting (Bertlein et al., 2017). Cell
culture and range of in vivo-like collective cell behaviors within spheroid-based bioprinting technologies (often termed bio-
organoids (Hughes et al., 2010). Motivated by limitations of assembly) have also emerged, where cell aggregates can be pre-
Matrigel such as batch variability and biochemical complexity, cisely assembled into cell-dense 3D constructs or structures con-
alternative 3D hydrogels from numerous biological and syn- taining organoids (Moldovan et al., 2017). The bioprinting method
thetic molecules have been developed and applied to the cul- used depends on the biological question and requisite consider-
ture of cells (Aisenbrey and Murphy, 2020; Cruz-Acuña et al., ations with respect to complexity, resolution, and cellularity.
2017; Gjorevski et al., 2016). Although these approaches are With respect to extrusion bioprinting, bioinks can generally be
advancing biology, they are still often limited to uniform struc- described as ‘‘a formulation of cells that is suitable to be pro-
tures and cultures of single cell types. cessed by an automated biofabrication technology’’ (Groll et al.,
To bring further organization to the culture of cells, the field of 2018). Common to the field is that the bioink is a hydrogel formu-
biofabrication has developed for the creation of cellular con- lation containing single-cell suspensions or cell aggregates. The
structs that are inspired by or mimic biological processes. These bioink may also be combined with cell-free biomaterial inks that
constructs incorporate living cells and extracellular or other are structural (to help support printed construct stability) or are
biochemical components and are configured into desired struc- sacrificial (meaning that they are only present temporarily during
tures, particularly for the engineering of tissue constructs for processing) (Kang et al., 2016). Further, although bioprinting
translational applications such as tissue repair and drug commonly involves the deposition of bioinks onto surfaces with
screening (Groll et al., 2016). Within the field of biofabrication, 3D structures built through the layering of printed filaments, an
there are a number of enabling technologies, one of which is emergent technique of great interest is the deposition of bioinks
bioprinting. Bioprinting is ‘‘the use of computer-aided transfer into suspension baths (also referred to as suspension media or
processes for patterning and assembly of living and non-living support hydrogels) that provide support during the printing pro-
materials with a prescribed 2D or 3D organization to produce cess (Figure 1B) (Bhattacharjee et al., 2015; Highley et al., 2015;
bio-engineered structures’’ (Moroni et al., 2018). Hinton et al., 2015; McCormack et al., 2020). This technique

18 Cell 184, January 7, 2021 ª 2020 Elsevier Inc.

 ll
Primer
Figure 1. Extrusion Bioprinting Process
In extrusion bioprinting, a bioink (formulation of
cells, often with a material) is deposited from a
printer nozzle either (A) onto a surface or (B) within
a suspension bath with a user-defined pattern.
There are numerous commercially available bio-
printers and biomaterials for use in bioinks that are
making bioprinting accessible to many users.

design of the overall print pattern and bio-

printing components (e.g., cells, bioinks,
biomaterial inks), (2) printing the desired
construct with the appropriate bioprinter,
and (3) processing the printed construct,
respectively (Figure 2). This all begins by
identifying the biological question of in-
terest, which will inherently guide the
other decisions during the planning and
printing stages. The biological question
may dictate the various cell populations
that are printed or seeded onto printed
constructs, the number and types of bio-
inks that are used, and the dimensions
and geometrical features that are
needed. In this section, we walk through
the steps in bioprinting in a general
manner, with an emphasis on the
commonly used extrusion bioprinting
enables the extrusion printing of bioinks that are otherwise chal- technique. Additional information and resources, such as
lenging to process using layer-by-layer methods. commercially available bioinks and bioprinters and links to
Although bioprinting has been widely explored for tissue fabri- user manuals for specific bioprinters, are included in Tables
cation in translational medicine (Moroni et al., 2018; Ozbolat and S1–S4. The following sections then provide numerous examples
Hospodiuk, 2016), there is much opportunity for the application where bioprinting has been implemented in biological questions.
of bioprinting to address fundamental biological questions.
Diverse cell-laden configurations are possible with extrusion bio- Plan
printing that span the cell matrix, cell-soluble factor, and cell-cell Print Design
interactions that drive biology. This is possible through the bioink The planning phase is a very important step in the bioprinting
selection and the use of multiple bioinks to create 3D constructs, process and includes two important aspects: creating the print
where the bioinks ultimately control the local cellular microenvi- design and selecting bioinks (Figure 2A). Printing designs are
ronment (i.e., biochemical and biophysical signals), and the often created through computer-aided-design (CAD) software,
placement of printed bioinks governs the macroscopic structure including commercially supplied or free software such as
and length scales across cell populations. It should be noted that FreeCAD, Solidworks, Blender, Onshape, and OpenSCAD
it is still challenging to replicate all of the structural, biochemical, (Junk and Kuen, 2016). Users can create a novel design from
and biophysical properties of tissues, and simplified versions are scratch or modify pre-existing designs, such as from patient/tis-
often bioprinted. sue scans or from other users. Additionally, many commercially
The goal of this Primer is to provide an overview of bioprinting available bioprinters come equipped with user-friendly software
for the biologist, which defines the steps and components to and support teams to help users with CAD models. For complex
extrusion bioprinting, reviews literature where bioprinting has prints, such as those mimicking tissue structures, there are
been used to address biological questions, highlights emerging numerous open source resources such as the NIH 3D print ex-
bioprinting technologies, and ends with an outlook of where bio- change, which provides medical and scientifically relevant
printing technology may be used in the future to address com- CAD models (Coakley et al., 2014). Once the CAD model is
plex biological questions. created, it can be saved and uploaded to printing software
to create G-code. Most commercial bioprinters accept
PRACTICAL STEPS TO BIOPRINTING STereoLithography (STL) file formats of CAD files, which save
3D structures as triangular tessellations, as is depicted in
There are several steps required to implement bioprinting, which Figure 2A. Open source software such as Repetier-Host or soft-
we define as (1) Plan, (2) Print, and (3) Process, referring to (1) the ware provided by bioprinting manufacturers is used to convert

Cell 184, January 7, 2021 19

 ll
Primer

(legend on next page)

20 Cell 184, January 7, 2021

 ll
Primer

Table 1. Considerations in the Selection of Bioinks for Extrusion Bioprinting

Specification Consideration
Rheological properties Rheological properties of a bioink will impact both cell response and printability. Shear-thinning hydrogels are often
considered ideal for bioprinting, as these materials can flow during extrusion and may protect cells from shear
stresses. Polymer concentration can be varied to control shear-thinning behavior with higher polymer concentrations
often possessing improved rheological properties (Liu et al., 2017a). Rheological additives such as gelatin or
methylcellulose can be used to induce shear-thinning behavior (Ahlfeld et al., 2020; Ouyang et al., 2020).
Method of gelation The method of gelation (e.g., photo-crosslinking, thermal) for a bioink should ideally be fast and nontoxic to cells.
The gelation method will determine compatibility with select bioprinters while length of gelation will determine
whether extra support, such as a suspension bath, is needed during bioprinting.
Biological properties Biological properties of a bioink will impact encapsulated cell response. Properties such as adhesion to cells and the
ability to degrade in culture will be important characteristics to understand in the context of an experiment.
Biophysical properties Biophysical properties, such as the elastic modulus of a bioink, can impact cellular responses such as growth
and differentiation.
Suspension bath If a bioink does not have ideal rheological properties or if the bioink has a long gelation time, the bioink can be
recommended? printed into a suspension bath or alongside a sacrificial biomaterial ink such as pluronic to offer temporary support
and improve resolution.

these STL files to G-code (Highley et al., 2015). G-code defines recapitulate complex kidney microarchitecture, the successful
the printing path for deposition of the bioink and can specify model focuses on a two-channel design that is easier to create
which bioinks are used throughout the print if more than one bio- and analyze but still effectively probes experimental study
ink is used. While STL file formats are acceptable for most bio- questions.
printing applications, methods to directly convert datasets into Bioink Selection
G-code are sometimes needed to avoid resolution loss (Bader Selection of bioinks is the other major step in the planning
et al., 2018). process of the bioprinting experimental workflow. While a brief
Some important considerations when creating or choosing a overview of this selection process is provided here, numerous
print design include considering the minimum complexity needed publications offer excellent in-depth reviews of commercially
in the print design and potential print settings such as needle available and state-of-the-art bioinks (Malda et al., 2013; Sun
used, extrusion flow rate, and print speed. The bioprinting plat- et al., 2020). The selection of the bioink is based on the
form is also important to consider as it can determine restrictions printability of the ink (e.g., compatibility with the printer, print res-
on bioink compatibility and achievable print resolution (e.g., ability olution) and the impact of the bioink on cell behavior. General
to heat or cool ink during extrusion, minimum extrusion pressure). considerations in bioink selection are described in Figure 2A
Key parameters in the printing process are interdependent on one and in more detail in Table 1. There are many commercially avail-
another and as a result the extrusion optimization process is often able bioinks that can be readily combined with desired cells (out-
iterative. For example, the needle diameter or filament flow rate lined in Table S1), as well as potentially useful biomaterial inks (to
chosen during the Print process inform the G-code design (print provide structure or that are sacrificial) and suspension baths
path, fill factor, and print speed). The needle diameter will influ- (outlined in Table S2).
ence the filament width and therefore the smallest feature size Printability is generally related to the rheological properties of
possible for the printed geometry (Blaeser et al., 2016). the bioink that permits extrusion during printing and the mech-
Most commercial platforms have excellent user manuals and anism that allows stabilization upon deposition onto a surface
training programs to guide new users through these parameters, or within a suspension bath. Traditional bioinks are viscous
and a selected list of these commercial platforms can be found in solutions, which may shear-thin during printing (meaning the
Tables S3 and S4. Ideally, it is best to simplify print designs as viscosity decreases as mechanical shear is applied during
much as possible to decrease unnecessary complexity in the extrusion from the nozzle) and then recover after deposition.
experimental workflow. For example, large, intricate designs In many cases, the bioink will undergo further stabilization
such as the kidney model depicted in Figure 2A may be possible and crosslinking (i.e., gelation), such as with light (photo-cross-
to fabricate but will take extended time to print and will likely be linking), chemical reaction (mixing, ionic, enzymatic), or tem-
difficult to culture and analyze. To address this, researchers have perature change (thermal). As bioprinting technologies and
simplified the kidney to an appropriate in vitro model (as dis- methods have developed, compatibility with various biomate-
cussed in more detail in Applications of Bioprinting in Biology) rial formulations has improved and techniques such as use of
to study crosstalk between renal kidney tubules and vasculature a suspension bath can aid in the processing of low-viscosity
(Lin et al., 2019). Instead of creating multiple channels or trying to bioinks (Figure 1B).

Figure 2. The Bioprinting Experimental Workflow

This workflow consists of three general parts: (A) planning (e.g., creating a print design and defining the bioink and biomaterial ink to be used), (B) printing the
construct, and (C) processing the printed construct over time with cell culture and identifying appropriate analytical outcomes.

Cell 184, January 7, 2021 21

 ll
Primer

With regards to cellular interactions of a bioink, the bioprinting of the bioink and printed structure, and media formulations
process may impact cell viability, and guidance should be taken that are dependent on the various cell populations included. As
from previous reports and commercial conditions to avoid with previous steps in the bioprinting process, users may have
exposing to excess shear stresses during extrusion (Blaeser to revisit the planning portions of the process to adjust bioink
et al., 2016). Each bioink will present different biochemical and formulation or print parameters based on results or updated pro-
biophysical features to cells, and the desired bioink may be tocols in the process phase. Constructs that are too large may
related to the specific cell type and biological question. For limit nutrient transport to incorporated cells. Custom bioreactors
instance, if the question relates to mechanobiology (the transla- may also be needed, such as to perfuse the channels within bio-
tion of local mechanics to biochemical signaling), a bioink where printed structures (Lin et al., 2019). While not the focus of this
the mechanical properties can be easily modulated should be article, more information on analysis of 3D cell-laden constructs
considered. In addition, the bioink must provide a suitable micro- and qualification of these models for industry or clinical use is
environment for the cell type being printed (primary, embryonic, detailed in previous publications (Caliari and Burdick, 2016;
or pluripotent derived); however, a detailed description of cell- Ekert et al., 2020).
hydrogel interactions is outside the scope of this paper, and
the reader is directed to published reviews (Caliari and Burdick, APPLICATIONS OF BIOPRINTING IN BIOLOGY
2016; Tibbitt and Anseth, 2009).
When possible, the use of commercially available and off-the- There are numerous examples where bioprinting has tackled
shelf materials is encouraged as these products come complete biological questions, particularly using extrusion bioprinting,
with rheological testing and suggested print settings, limiting and there are many opportunities still to explore. These studies
laborious troubleshooting and characterization needed from have been largely motivated by tissue development or tissue
the user. Some of these available bioinks are detailed in Table repair processes and have involved printing constructs with
S2. If developing a custom bioink for extrusion bioprinting or spatially patterned cell populations and/or biochemical factors.
other bioprinting platforms, the reader is directed to previous This section will provide the reader with various examples where
publications detailing the bioink development process (Gillispie bioprinting has been implemented in biological questions
et al., 2020). Other resources may also be found through manu- already and identify why bioprinting was useful over traditional
facturers of commercial printers, who often provide useful fabrication techniques.
guides for characterization of novel bioinks for specific platforms
(Table S4). Bioprinted Models to Study Tissue Development and
Repair
Print Biochemical Gradients
Once the planning process is complete, the user can move to Biochemical gradients provide spatiotemporal cues to direct cell
printing. There are a variety of bioprinting technologies that are differentiation in developing tissues (Rogers and Schier, 2011). It
well defined in previous reviews (Matai et al., 2020). Of these is challenging to recapitulate the spatiotemporal complexity of
technologies, extrusion-based systems tend to be the most developmental cascades where multiple signaling centers
versatile platforms for bioprinting. Extrusion bioprinting creates transiently arise to direct differentiation and morphogenesis;
3D constructs via the dispensing of bioink filaments through noz- however, bioprinting is a promising technology for such applica-
zles, which are controlled through pneumatic pressure or syringe tions as the spatial patterning of multiple biochemical species in
pumps (Matai et al., 2020). These systems are compatible with a 3D hydrogels is possible that can then diffuse throughout the hy-
wide variety of bioinks and include various features (heating, drogel to interact with cells. To study vascular angiogenesis in
cooling, light exposure) that allow processing of the aforemen- response to soluble factors, printing into suspension baths was
tioned bioinks (Figure 2B). Many systems also include multiple used to create vascular channels inside cell degradable hydro-
extruders to allow users to print multiple bioinks in a single print. gels, and a second channel was used to present a gradient of
There are a variety of commercial solutions that allow access to a cocktail of growth factors (Figure 3A) (Song et al., 2018). Inter-
extrusion bioprinting technology without requiring the ability or estingly, when endothelial cells (ECs) sprouted from the channel
time commitment to build custom systems, some of which are toward the biochemical gradient, increased sprouting was
outlined generally in Table S3, with further details provided in Ta- observed at curved locations. This study highlights how bio-
ble S4. Commercially available systems also come with signifi- printing technology is useful to probe how biochemical signaling
cant support, standardization, and communities of users, and is interpreted in complex geometric contexts to influence biolog-
their costs range from entry level to expert, based on features ical processes such as collective cell migration. Further, such
such as print resolution, number of print nozzles, and range of geometrically complex channels (e.g., introducing curvature,
printing technology included. creating interconnected channel networks) would be challenging
with other more traditional methods (e.g., sacrificial molding).
Process Combinatorial arrays of morphagen gradients are also possible
The final step in the bioprinting experimental plan is to process (Miller et al., 2009), where traditional techniques are limited to
bioprinted constructs (Figure 2C). This step involves both culture single gradients.
and analysis of printed constructs and is dependent on the spe- Biophysical Morphogenesis
cific biological question being asked. Important considerations During development as tissues grow and expand, internal and
include the length of the study, which may dictate the stability interfacial pressures and tensions are generated, which can

22 Cell 184, January 7, 2021

 ll
Primer

Figure 3. Bioprinted Models of Tissue Development and Disease

(A) (Left) Angiogenic sprouting. Schematic of 3D-printed microchannels within cell degradable hydrogels where the left channel is seeded with endothelial cells
and the right channel is perfused with angiogenic factors, and representative images of endothelial cell sprouting from the vascular channel toward the growth
factor gradient over 3 days of culture (Song et al., 2018). (Right) Tissue buckling. Schematic demonstrating embedded 3D printing of collagen filaments containing
fibroblasts into a granular yield stress media (i.e., suspension bath) and fibroblast contraction of the collagen matrix inside the printed filament. Contraction and
buckling of the collagen filament occurs over 24 h of culture as a function of the filament aspect ratio (length/diameter) (Morley et al., 2019).
(B) (Left) Glioblastoma model. Schematic and image of extrusion bioprinting of a mini-brain model containing compartmentalized regions of glioblastoma cells
and macrophages to study the role of macrophages in glioblastoma progression (Heinrich et al., 2019). (Right) Renal proximal tubule model. Schematic
demonstrating 3D printing of a sacrificial pluronic ink to generate convoluted perfusable channels inside a gelatin/fibrin matrix and seeding of the microchannels
with proximal tubule epithelial cells (PTECS) and glomerular microvascular endothelial cells (GMECs) to generate parallel vascular and renal epithelial channels to
study solute renal reabsorption (Lin et al., 2019).

lead to mechanical instabilities such as folding and buckling methods is challenging; however, cell-laden bioinks offer a
(Nelson, 2016). These shape-morphing events contribute to tis- promising approach where cell-generated forces and ECM me-
sue patterning through mechanotransduction-mediated cell chanics can be spatially controlled. Morley et al. (2019) used
specification and remodeling of the extracellular matrix (ECM) printing within a suspension bath to extrude a fibroblast-laden
(Mammoto and Ingber, 2010). Reconstructing biophysical collagen bioink into a granular support slurry and then measured
models of tissue morphogenesis using traditional in vitro culture the time-dependent changes in filament geometry as a result of

Cell 184, January 7, 2021 23

 ll
Primer

cell-generated traction on the collagen (Morley et al., 2019). By Tubular Disease Models
varying the filament length and diameter, and also the mechanics 3D bioprinting approaches using sacrificial inks offer an elegant
of the bioink and the suspension bath, a range of mechanical de- approach to generate perfusable microchannels inside 3D
formations including buckling, axial contraction, failure, and total hydrogels (Highley et al., 2015). While these approaches
static stability were observed (Figure 3A). This platform holds have been predominantly focused on engineering tissues for
tremendous potential for studying biophysical morphogenesis implantation, there is a significant opportunity to develop
in 3D across multiple cell types, and the design flexibility af- vascular and epithelial disease models. For example, sacrificial
forded by bioprinting technology allows investigation into how embedded printing has been used to engineer tissue models
geometrical features arise during morphogenesis. This approach to study reabsorption of solutes and crosstalk between renal kid-
provides advantages of the freedom of design and control over ney tubules and vasculature, which are in tight juxtaposition
geometrical features when compared to traditional methods of along non-linear paths, a difficult construction problem that war-
molded hydrogels (e.g., collagen, fibrin). ranted bioprinting innovation (Lin et al., 2019). Using a sacrificial
Paracrine Signaling and Co-culture pluronic ink, two parallel microchannels were printed inside a
During tissue development, cells communicate using paracrine fibrin matrix and epithelial and vascular monolayers were gener-
signals via several highly conserved receptors and pathways. ated by seeding one channel with proximal tubule epithelial cells
Bioprinting technologies offer a promising platform to study and a second channel with vascular endothelial cells (Figure 3B).
paracrine signaling in vitro as multiple cell populations can be Flow through the channels was controlled using a closed-loop
compartmentalized in 3D matrices with biomimetic patterning, perfusion system to study renal reabsorption of glucose from
which can be challenging to achieve using traditional cell culture the epithelial channels into the vascular channels, and the model
methods. In the context of liver development, hepatocytes and was able to recapitulate endothelial cell dysfunction and
endothelial cells have been printed into lobule-like geometries enhanced reabsorption in hyperglycemic disease conditions.
with biomimetic heterocellular localization, resulting in enhanced Fibrosis Disease Models
maturation compared to co-cultures that lacked geometric Following tissue injury in the heart, liver, and lung, adverse path-
structure (Kang et al., 2018). Bioprinting allows control over ological remodeling can lead to the development of non-func-
distinct cell populations to investigate paracrine signaling tional fibrotic tissue that disrupts surrounding healthy tissue
(Jeon et al., 2020), which is challenging or not possible with and leads to eventual organ failure. Engineered models of tissue
traditional methods (e.g., Transwell inserts, sequential micro- fibrosis could offer a significant opportunity to study disease pro-
molding). gression or tissue repair; however, it is challenging to reconstruct
the heterogenous cellular and extracellular patterning that arises
Bioprinted Tissues for Disease Modeling following scarring using traditional culture methods. To develop
Cancer Disease Models a model of liver fibrosis, extrusion bioprinting was used to create
Ex vivo cancer models are aiding in the design of personalized structured layers of hepatocytes, activated stellate cells, and
drug treatment regimes and in understanding the basic biology endothelial cells (Lee et al., 2020). The model exhibited charac-
that underlies disease. However, recreating the complexity of teristics of fibrotic remodeling including collagen accumulation,
the cancer environment in vitro—including stroma and immune cell apoptosis, and reduced liver function that was attributed
interactions, angiogenesis, and ECM remodeling—is chal- to the presence of the stellate cell population, and it was possible
lenging with traditional culture methods. In particular, the resis- to attenuate fibrosis using drugs targeting stellate cell activation.
tance of cancer cells to chemotherapy drugs is well known to
be modulated by interactions with surrounding stromal and im- ADVANCED BIOPRINTING TECHNOLOGIES
mune cells, and simplified 2D cell cultures do not capture this
complexity (Pauli et al., 2017). To develop a model of the glio- Although extrusion bioprinting is a common and accessible
blastoma microenvironment, extrusion bioprinting of decellular- bioprinting technology, there are other related technologies
ized ECM inks was used to create compartmentalized regions of that may be of use in the pursuit of biological questions. This
glioblastoma cells and endothelial cells, which better mimicked section highlights examples where these advanced bioprinting
the tumor-stroma interactions when compared to mixed co-cul- technologies have been implemented, and the advantages and
tures and reproduced clinically observed patient-specific resis- disadvantages of these techniques over extrusion technologies
tance to treatment (Yi et al., 2019). In another study, a mini-brain (Table 2).
model with compartmentalized regions of glioblastoma cells and
macrophages was developed through extrusion bioprinting Lithography Bioprinting
using a gelatin methacrylamide (GelMA) bioink (Figure 3B) (Hein- Lithography is achieved by concentrating light into a 2D plane to
rich et al., 2019). Glioblastoma cells were observed to actively re- locally solidify a hydrogel resin, and then a robotic stage verti-
cruit macrophages into the tumor region and polarized them into cally translates the crosslinked layer to allow sequential layer-
a glioblastoma-macrophage phenotype, which demonstrated by-layer crosslinking into a 3D solid. Several variations of lithog-
correlations with clinically generated transcriptome data. Future raphy exist depending on how light is delivered—stereolithogra-
studies will likely build on these techniques to integrate addi- phy (SLA) technologies utilize a scanning laser beam, whereas
tional vascular, immune, and stromal components to provide digital light processing (DLP) technologies utilize a digital mirror
predictive tissue models to dissect the multifactorial complexity device to rapidly mask light into 2D patterns (Lim et al., 2020).
of the cancer microenvironment. Lithography technologies can create physical features in the

24 Cell 184, January 7, 2021

 ll
Primer

Table 2. Comparison of Different Bioprinting Technologies

Bioprinting Technology Advantages Disadvantages
Extrusion bioprinting d Possible to create heterogenous constructs using d Moderate cell density (challenging to approach
multi-printhead bioprinters organ level density)
d Freedom of design, possible to create complex d Moderate feature size resolution (>100 mm)
geometrical features d Bioink must possess suitable rheological behavior for printing
d Relatively fast processing times
d Wide variety of commercial printers available

Lithography bioprinting d Possible to create high-resolution features down d Challenging to pattern heterogenous structures
to 5–10 mm d Low-cell-density constructs
d Freedom of design, possible to create highly d Bioink must be photocrosslinkable
complex geometrical features d Relatively slow processing times (note: emerging
volumetric technologies overcome this limitation)
d Limited availability of commercial printers

Spheroid bioprinting/ d High-cell-density constructs d Slow processing times

bioassembly d Possible to create heterogenous constructs with d Limited availability of commercial printers
high spatial precision d Challenging to create highly complex geometries
d Cells can be matured (i.e., the spheroid) prior to
bioprinting

10–100 mm range, which is a significant advantage compared to lithography bioprinting to engineer two entangled open-channel
extrusion technology, which has a minimum resolution of networks within a synthetic hydrogel (Grigoryan et al., 2019). The
100 mm (Bertlein et al., 2017; Lim et al., 2018). Further, the first network was perfused with deoxygenated red blood cells
development of biocompatible photo-crosslinking chemistries (RBCs) and the second network was perfused with humidified
has enabled the design of hydrogel resins that can support gaseous oxygen, resulting in reoxygenation of the RBCs during
high cell viability (i.e., bioresins) (Lim et al., 2020). flow. To further demonstrate the power of this technology, the
Deciding between extrusion or lithography bioprinting de- authors bioprinted a vascularized alveolar lung model containing
pends on the cell/tissue model being developed. For example, ventilated air sacs surrounded by perfused vascular beds to
extrusion bioprinters are relatively cheap compared to lithog- study oxygenation of RBCs in response to mechanical ventila-
raphy systems. In addition, extrusion bioprinting is compatible tion in the air sacs (Figure 4A).
with a wide variety of bioinks, whereas lithography bioprinting Due to the layer-by-layer nature of SLA and DLP lithography
is only compatible with photo-crosslinkable bioinks/bioresins. technologies, particularly long processing times are required to
It can also be challenging to fabricate heterogenous constructs create large volumes, which is a disadvantage of this technology.
using lithography methods as the bioresin cannot be easily To address this fundamental limitation, volumetric lithographic
switched out during fabrication, limiting applicability toward technologies have recently been developed in which light energy
co-culture models, although it should be noted that newer tech- is delivered to a material volume from a set of 2D image projec-
nologies have recently been developed to address this limitation tions delivered simultaneously from multiple angles (Figure 4A)
(Miri et al., 2018). As lithography technologies become more (Kelly et al., 2019; Loterie et al., 2020). The additive light dose
widespread and commercially available, they are increasingly exposure from multiple angles results in a 3D energy dose that
being used to engineer tissue and organ models; however, a rapidly solidifies a resin volume. In an important study, Bernal
detailed description of the steps to lithographic techniques and et al. (2019) demonstrated that this technology could be adopted
their components is outside the scope of this article. for bioprinting purposes, enabling rapid fabrication of anatomi-
There are numerous interesting examples where lithography cally shaped and human-scale cell-laden constructs (Bernal
techniques have been used to fabricate cell-laden structures et al., 2019).
or structures that are subsequently seeded with cells for biolog-
ical questions. For example, to recreate the microarchitectural Spheroid Bioprinting
complexity of a liver lobule, Ma et al. (2016) used DLP lithography The printing of high-cell-density constructs is an important
to pattern induced pluripotent stem cell (iPSC)-hepatocytes, consideration, as cells rarely exist in isolation and coordinated
endothelial cells, and adipose-derived stem cells in a biomimetic cellular collective processes mediated by cell-cell contact un-
hexagonal microarchitecture using a gelatin bioresin (Figure 4A) derlie developmental morphogenesis (Hall and Miyake, 1995).
(Ma et al., 2016). The bioprinted tri-culture model enhanced he- In addition, many disease states such as fibrosis or cancer are
patocyte functionality compared to 2D and 3D controls, with in- challenging to faithfully recapitulate when single cells are
creases in liver-specific gene expression, albumin secretion, and dispersed throughout gels. Cells self-organize into miniaturized
drug-metabolizing enzymes. spheroid or organoid structures and have been used by biolo-
Lithography bioprinting can also be used to create microchan- gists for years to study human development and disease
nels inside 3D hydrogels, to create hierarchical interconnected in vitro (Fatehullah et al., 2016). In particular, organoid models
networks (e.g., capillary beds). Grigoryan et al. (2019) utilized can display emergent levels of physiological structure and

Cell 184, January 7, 2021 25

 ll
Primer

(legend on next page)

26 Cell 184, January 7, 2021

 ll
Primer

function due to their high cell density and capacity to support ease (Cakir et al., 2019). To facilitate scaling up of organoids into
developmental-like cell sorting and differentiation (Rossi et al., vascularized 3D tissues, thousands of aggregates have been
2018). However, there is limited control over the self-organiza- jammed together in supporting molds to create self-healing
tion processes, and organoids possess non-polarized structur- granular tissue matrices that can support sacrificial embedded
ally immature microarchitectures compared to native organs 3D printing of perfusable vascular channels (Skylar-Scott et al.,
(Laurent et al., 2017). This has led to the development of hybrid 2019b). Embryoid bodies, cerebral organoids, and cardiac
bioprinting technologies capable of processing self-organized spheroids were compatible with the process, and the inclusion
tissues (often cell spheroids) into 3D constructs to scale and of vascular channels enhanced cell viability within core regions
direct self-organization. of the tissues. It should be noted that spheroid bioprinting has
Early work in this area demonstrated that pre-formed spher- some limitations of relatively long processing times and limita-
oids can fuse together through liquid-like coalescence to mini- tions in the complexity of printed structures; however, there
mize adhesive-free energy (Fleming et al., 2010). To scale this are many benefits related to the high cell densities produced
into a bioprinting process, multiple spheroids were fused into tis- that mimic tissue-like features.
sue strands, followed by automated extrusion of the strands into
larger 3D constructs (Norotte et al., 2009). The kenzan method OUTLOOK FOR BIOPRINTING IN BIOLOGY
has also been developed where cell spheroids are skewered
onto supporting metallic needles for fusion into 3D constructs, Bioprinting has great potential for the exploration of biological
followed by removal of the fused tissue from the needle supports questions in which traditional techniques are insufficient to build
(Figure 4B) (Ong et al., 2017). This system is commercially avail- in desired complexity and organization, and the technology is in
able and has been used to fabricate a range of different tissue its infancy with regards to its use in biological research. Below
models (Moldovan et al., 2017). More recently, hydrogel based we have outlined several biological contexts where the technol-
spheroid bioprinting technologies have been developed in which ogy may be particularly useful.
spheroids are printed into 3D constructs through sequential
layering of an uncrosslinked hydrogel precursor and spheroids, Biochemical Signaling at a Distance
followed by crosslinking of the hydrogel layer (Figure 4B) (Ayan The term ‘‘morphogen’’ was coined by the computer scientist
et al., 2020). These systems avoid mechanical disruption of Alan Turing to describe factors that form a spatially non-uniform
spheroids, enabling precise positioning in 3D with improved con- distribution spanning multiple cell lengths to instruct different cell
trol over geometry and heterogeneity (Ayan et al., 2020). To study fates at distinct levels (Green and Sharpe, 2015). Understanding
how far paracrine signals can travel in the ECM, Ayan et al. (2020) morphogen gradients in vivo is complex due to a limited ability to
used aspiration-assisted bioprinting to print endothelial cell change spatial features of developing tissues (Hiscock and Meg-
spheroids in a fibrin hydrogel at varying degrees of separation ason, 2015). Bioprinting could play a key role here by juxtaposing
(400, 800, and 1,200 mm) (Ayan et al., 2020). Limited interactions engineered or primary tissue-derived morphogen ‘‘sender’’ cells
were observed at high separation; however, at closer proximity, with morphogen ‘‘receiver’’ cells, perhaps in periodic arrays, ex-
enhanced EC sprouting and capillary network formation were tending emerging efforts in 2D to a 3D context (Li et al., 2018).
observed. Questions such as tissue size and composition, diffusion dis-
Although less widely used than extrusion bioprinting, spheroid tance, and diffusion/absorption rates are ideally suited to a com-
bioprinting technologies hold significant promise for developing bination of cell engineering and bioprinting methods, which can
organ and tissue models. As an example, several varieties of control these variables quantitatively (Ozbolat and Hospodiuk,
brain organoids have been developed to mimic different regions 2016; Song et al., 2018). For example, extrusion bioprinting
of the brain (Di Lullo and Kriegstein, 2017), and simple fusion be- could be used to create arrays of cell depots enabling combina-
tween two organoid phenotypes has been used to study regional tional screening of paracrine signaling between distinct cell pop-
interactions in vitro (Birey et al., 2017). Spheroid bioprinting ulations such as interactions between vascular and tumor cells
methods could also provide a powerful method to direct fusion (Figure 5A). Parameters such as the depot spacing could be
into more biomimetic organotypic assemblies. Finally, there varied across multiple cell types followed by mapping out cellular
has been increased interest in engineering vascularized organo- outcomes (proliferation, migration, ECM secretion, protein/gene
ids to enhance oxygen and nutrient delivery in core regions and expression) through live imaging (Figure 5A). Cells could also be
also to model vascular interactions during development and dis- engineered for signaling or biochemical secretion triggered by

Figure 4. Advanced Bioprinting Technologies

(A) (Left) DLP lithography bioprinting process where light is spatially projected onto a cell-laden bioresin using a digital micromirror device to create a liver lobule
construct (green regions contain iPSC-hepatocytes, and red regions contain endothelial cells and adipose-derived stem cells). Scale bars, 500mm (Ma et al.,
2016). (Right) DLP bioprinting of an alveolar lung model containing a central mechanically ventilated air sac surrounded by vascular channels perfused with red
blood cells, and demonstration of gaseous exchange through measurement of reoxygenation of the red blood cell population (green line) following oxygenation of
the air sac (blue line). (Grigoryan et al., 2019). (Bottom) Experimental setup for volumetric bioprinting including laser input followed by DLP projection modulation
of light onto a rotating platform containing the bioresin. Image of bioprinted human ear model created using a cell-laden GelMA bioresin, total printing time 22.7 s.
Scale bar, 2mm (Bernal et al., 2019; Loterie et al., 2020).
(B) (Left) Kenzan bioprinting method where cell spheroids are aspirated and then skewered onto metal needles for fusion into 3D constructs. (Ong et al., 2017)
(Right) Aspiration-assisted bioprinting of spheroids (labeled red and green) onto fibrin hydrogels at different separation distances to study paracrine signaling and
angiogenic sprouting. Scale bar, 400mm (Ayan et al., 2020).

Cell 184, January 7, 2021 27

 ll
Primer

Figure 5. Bioprinting Approaches to Biological Questions in Tissue Development and Homeostasis

(A) To study interactions between endothelial and tumor cells in a highly controlled manner, spatial combinatorial patterning of engineered ‘‘sender’’ and
‘‘receiver’’ cell arrays could test hypotheses around diffusible biochemical signaling and their influence on cell phenotype and function (e.g., protein/gene
expression, migration, proliferation). This could be achieved by patterning cell depots of distinct compositions at prescribed spacing within ECM hydrogels and
then monitoring cell behavior during culture.
(B) To study biophysical morphogenesis in neural crest development, strains at bioprinted tissue interfaces could be generated through internally generated cell-
cell or cell-ECM forces to create dynamic changes in tissue shape. Bioprinting could be used to interface two filaments where differences in cell behavior (e.g.,
contractility, proliferation) within filaments drive bending or buckling behaviors.
(C) Gradients in growth factor concentrations could stimulate formation of fluid-like and solid-like domains to guide dynamic remodeling of bioprinted tissues and
to study microenvironmental conditions that drive tissue fluidity (i.e., interplay between cell-cell interactions, cell-ECM interactions, and morphogen presenta-
tion), such as during head-to-tail elongation in the zebrafish embryo. Models could be produced by patterning morphogen depots adjacent to filaments
containing cell density gradients, allowing combinatorial screening of cell migration behavior across conditions.

28 Cell 184, January 7, 2021

 ll
Primer

light or small-molecules to study how specific morphogens influ- relieved by viscoelastic dissipation (Clément et al., 2017). Devel-
ence paracrine interactions in a highly controlled manner. oping tissues employ several strategies for inducing strain at in-
terfaces, including changes in apical dimension of cell sheets
Building Structure at Interfaces (apical constriction) (Martin et al., 2009) and differential growth,
Biological structure is often built at interfaces between cell pop- contractility, or mechanical constraint in one tissue layer at the
ulations with different properties and eventual fates. Examples interface relative to the other (Hughes et al., 2018; Spurlin
across developmental time include induction and spatial segre- et al., 2019). Other shape changes can also be achieved by
gation of the three germ layers during gastrulation (endoderm, appropriate spatial patterning of fluid-like and solid-like
mesoderm, ectoderm), and self-organization of distinct early (jammed) cell domains (Figure 5C) (Mongera et al., 2018), where
embryonic structures such as the neural tube and somites (early fluidity of a domain is associated with lower cell-cell adhesion,
body segments). Engineering efforts to control interfacial struc- higher cell motility, and/or lower cell density (Ihermann-Hella
ture formation will be crucial to forming new in vitro tissues for et al., 2014; Sadati et al., 2013). Bioprinting methods could aid
disease modeling and to explore biological questions. Indeed, in establishing the relationship between fluid-like states and geo-
cells show a range of dynamics at these interfaces that are metric, mechanical, and biochemical features of the tissue
open to engineering. These dynamics include cell sorting on microenvironment. Bioprinting could be used to create tissues
the basis of cell-cell adhesion or cell-ECM adhesion properties with cell density gradients to determine if cell density alone is
(Cerchiari et al., 2015), establishment of polarity-distinguishing sufficient to trigger formation of a tip-stalk phenotype, and if
apical (up) and basal (down) directionality to cells (Andrew and not, which additional microenvironmental features need to be
Ewald, 2010; Nissen et al., 2018), as well as in-plane direction- specified. For example, engineered tissue with intrinsic mechan-
ality (planar cell polarity) (Butler and Wallingford, 2017) and col- ical stress profiles that occur at high aspect ratio features due to
lective migration (Cetera et al., 2018). The positioning of inter- cell-matrix traction could be patterned within biochemical gradi-
faces and the behavior of cells at interfaces can also be ents thought to reinforce ‘‘tip’’ cell states (such as glial cell-
refined by cell-cell or cell-ECM repulsion/avoidance cues such derived neurotrophic factor in the developing kidney) (Gjorevski
as Eph-Ephrin and Versican signaling (Scheideler et al., 2020; et al., 2015; Menshykau et al., 2019). This could be achieved by
Szabó et al., 2016). In total, these collective cell decisions then patterning depots of morphogens adjacent to cell density gradi-
create a ‘‘blueprint’’ for future events that sculpt tissues. For ents for combinatorial screening of cell migration across a wide
example, planar cell polarity appears to have an important role range of microenvironmental conditions (Figure 5C). Such exper-
in driving epithelial tubule elongation and sculpting craniofacial iments could establish fundamental understanding for designing
cartilage (Kaucka et al., 2017). This is achieved by setting the di- bioprinted tissues that undergo, for example, programmed
rection of oriented cell divisions and cell ‘‘intercalation,’’ in which branching morphogenesis processes in vitro. They would also
groups of cells adjust their geometric relationship to each other lend quantitative understanding to shape-change phenomena
in such a way as to elongate in one direction while contracting in new embryo-like organoids (van den Brink et al., 2020).
along a radial or orthogonal direction (Kaucka et al., 2017).
Further, mechanical tension within tissues feeds back into planar SUMMARY
cell polarity and oriented cell division (Aw et al., 2016).
This reveals an opportunity to explore the effect of patterned This review provides an overview of bioprinting as a field,
tension fields on cell collective behaviors within bioprinted ob- describing the steps to implement the commonly used extrusion
jects over time. Indeed, bioprinting has a distinctive role to play bioprinting technology and reviewing examples where this tech-
here, because placing cells at synthetic interfaces in 3D would nology has been used to address biological questions. In some
begin to create a biochemical-to-morphological map that could instances, the information provided here can act as a guide for
be exploited to study the development of tissue interfaces such the bioprinting of simple structures, whereas in other cases it
as the neural crest (Figure 5B). 3D bioprinting could be used to may be useful to develop collaborations with the appropriate en-
interface two filaments within supportive ECM hydrogels and gineers or directly with bioprinting companies to help accelerate
several parameters could be varied such as the cell type (epithe- the adoption of bioprinting by biologists.
lial, mesenchymal), the ECM type (collagen/laminin rich), the Efforts in the use of bioprinting in biology will only grow as bio-
filament ECM mechanics (stiffness, viscoelasticity), and the printing technology advances—from new bioinks developed to
initial filament geometry (straight, curved) (Figure 5B). Such mimic the dynamic nature of biology to new bioprinters and bio-
approaches could be used to study how local differences in col- printing approaches that match the complexity of biology. To
lective cell behavior can generate internal tensions and forces further increase widespread adoption, engineers are continuing
that drive morphological changes at the interface, including local to streamline bioprinting technologies to improve automation in
bending and buckling (Figure 5B). order to limit the experience required by the operator. For
example, the development of automated ‘‘mid-print’’ feedback
Shape Change mechanisms between the machine and the bioprinted object
Coordinated shape change in cell sheets and tubules through could fully automate the bioprinting process (Sun et al., 2020).
intercalation and oriented cell division is complemented by a Bioprinters are also being developed with microfluidic extrusion
range of other shape-change phenomena at interfaces. In princi- printheads that enable fast and smooth switching across
ple, a shape change will accompany any local or global mechan- different bioink reservoirs during the print process making it
ical strain (change in length) parallel to an interface that is not easier to recapitulate the biological complexity of native tissue

Cell 184, January 7, 2021 29

 ll
Primer

and organs (Liu et al., 2017b). Lastly, extrusion printers with par- tor to Balance Printing Resolution and Stem Cell Integrity. Adv. Healthc. Mater.
allelized nozzles have also emerged to offer significantly 5, 326–333.
increased throughput (Skylar-Scott et al., 2019a). Butler, M.T., and Wallingford, J.B. (2017). Planar cell polarity in development
One particular area that will likely see advances with bio- and disease. Nat. Rev. Mol. Cell Biol. 18, 375–388.

printing is that of morphogenesis, which involves complex cell, Cakir, B., Xiang, Y., Tanaka, Y., Kural, M.H., Parent, M., Kang, Y.-J., Chapeton,
biochemical, and biophysical dynamics that sculpt the shape K., Patterson, B., Yuan, Y., He, C.-S., et al. (2019). Engineering of human brain
organoids with a functional vascular-like system. Nat. Methods 16,
and composition of living organisms and their constituent or-
1169–1175.
gans. These complexities can be recapitulated in some form
Caliari, S.R., and Burdick, J.A. (2016). A practical guide to hydrogels for cell
with bioprinted constructs, including merging with the rapidly ex- culture. Nat. Methods 13, 405–414.
panding area of organoid engineering. Thus, the future of bio-
Cerchiari, A.E., Garbe, J.C., Jee, N.Y., Todhunter, M.E., Broaders, K.E., Peehl,
printing provides great potential across wide-ranging biological D.M., Desai, T.A., LaBarge, M.A., Thomson, M., and Gartner, Z.J. (2015). A
questions. strategy for tissue self-organization that is robust to cellular heterogeneity
and plasticity. Proc. Natl. Acad. Sci. USA 112, 2287–2292.
SUPPLEMENTAL INFORMATION Cetera, M., Leybova, L., Joyce, B., and Devenport, D. (2018). Counter-rota-
tional cell flows drive morphological and cell fate asymmetries in mammalian
Supplemental Information can be found online at https://doi.org/10.1016/j. hair follicles. Nat. Cell Biol. 20, 541–552.
cell.2020.12.002.
Clément, R., Dehapiot, B., Collinet, C., Lecuit, T., and Lenne, P.-F. (2017).
Viscoelastic Dissipation Stabilizes Cell Shape Changes during Tissue Morpho-
ACKNOWLEDGMENTS genesis. Curr. Biol. 27, 3132–3142.e4.
Coakley, M.F., Hurt, D.E., Weber, N., Mtingwa, M., Fincher, E.C., Alekseyev,
The authors acknowledge support through the American Heart Association
V., Chen, D.T., Yun, A., Gizaw, M., Swan, J., et al. (2014). The NIH 3D Print Ex-
(postdoctoral fellowship to A.C.D.), the National Institutes of Health (MIRA
change: A Public Resource for Bioscientific and Biomedical 3D Prints. 3D Print
Award: R35GM133380), and the National Science Foundation (GRFP award
Addit. Manuf. 1, 137–140.
to M.E.P., STC Program [CMMI: 15-48571]).
Cruz-Acuña, R., Quirós, M., Farkas, A.E., Dedhia, P.H., Huang, S., Siuda, D.,
Garcı́a-Hernández, V., Miller, A.J., Spence, J.R., Nusrat, A., and Garcı́a, A.J.
REFERENCES
(2017). Synthetic hydrogels for human intestinal organoid generation and
colonic wound repair. Nat. Cell Biol. 19, 1326–1335.
Ahlfeld, T., Guduric, V., Duin, S., Akkineni, A.R., Schütz, K., Kilian, D., Emmer-
macher, J., Cubo-Mateo, N., Dani, S., Witzleben, M.V., et al. (2020). Methylcel- Di Lullo, E., and Kriegstein, A.R. (2017). The use of brain organoids to investi-
lulose - a versatile printing material that enables biofabrication of tissue gate neural development and disease. Nat. Rev. Neurosci. 18, 573–584.
equivalents with high shape fidelity. Biomater. Sci. 8, 2102–2110. Ekert, J.E., Deakyne, J., Pribul-Allen, P., Terry, R., Schofield, C., Jeong, C.G.,
Aisenbrey, E.A., and Murphy, W.L. (2020). Synthetic alternatives to Matrigel. Storey, J., Mohamet, L., Francis, J., Naidoo, A., et al. (2020). Recommended
Nat. Rev. Mater. 5, 539–551. Guidelines for Developing, Qualifying, and Implementing Complex In Vitro
Andrew, D.J., and Ewald, A.J. (2010). Morphogenesis of epithelial tubes: In- Models (CIVMs) for Drug Discovery. SLAS Discov. 10, 1174–1190.
sights into tube formation, elongation, and elaboration. Dev. Biol. 341, 34–55. Fatehullah, A., Tan, S.H., and Barker, N. (2016). Organoids as an in vitro model
Aw, W.Y., Heck, B.W., Joyce, B., and Devenport, D. (2016). Transient Tissue- of human development and disease. Nat. Cell Biol. 18, 246–254.
Scale Deformation Coordinates Alignment of Planar Cell Polarity Junctions in Fleming, P.A., Argraves, W.S., Gentile, C., Neagu, A., Forgacs, G., and Drake,
the Mammalian Skin. Curr. Biol. 26, 2090–2100. C.J. (2010). Fusion of uniluminal vascular spheroids: a model for assembly of
Ayan, B., Heo, D.N., Zhang, Z., Dey, M., Povilianskas, A., Drapaca, C., and Oz- blood vessels. Dev. Dyn. 239, 398–406.
bolat, I.T. (2020). Aspiration-assisted bioprinting for precise positioning of Gillispie, G., Prim, P., Copus, J., Fisher, J., Mikos, A.G., Yoo, J.J., Atala, A., and
biologics. Science Advances 6, eaaw5111. Lee, S.J. (2020). Assessment methodologies for extrusion-based bioink print-
Bader, C., Kolb, D., Weaver, J.C., Sharma, S., Hosny, A., Costa, J., and Ox- ability. Biofabrication 12, 022003.
man, N. (2018). Making data matter: Voxel printing for the digital fabrication Gjorevski, N., Piotrowski, A.S., Varner, V.D., and Nelson, C.M. (2015). Dynamic
of data across scales and domains. Science Advances 4, eaas8652. tensile forces drive collective cell migration through three-dimensional extra-
Benam, K.H., Dauth, S., Hassell, B., Herland, A., Jain, A., Jang, K.-J., Karalis, cellular matrices. Sci. Rep. 5, 11458.
K., Kim, H.J., MacQueen, L., Mahmoodian, R., et al. (2015). Engineered in vitro Gjorevski, N., Sachs, N., Manfrin, A., Giger, S., Bragina, M.E., Ordóñez-Morán,
disease models. Annu. Rev. Pathol. 10, 195–262. P., Clevers, H., and Lutolf, M.P. (2016). Designer matrices for intestinal stem
Bernal, P.N., Delrot, P., Loterie, D., Li, Y., Malda, J., Moser, C., and Levato, R. cell and organoid culture. Nature 539, 560–564.
(2019). Volumetric Bioprinting of Complex Living-Tissue Constructs within Green, J.B.A., and Sharpe, J. (2015). Positional information and reaction-diffu-
Seconds. Adv. Mater. 31, e1904209. sion: two big ideas in developmental biology combine. Development 142,
Bertlein, S., Brown, G., Lim, K.S., Jungst, T., Boeck, T., Blunk, T., Tessmar, J., 1203–1211.
Hooper, G.J., Woodfield, T.B.F., and Groll, J. (2017). Thiol-Ene Clickable Grigoryan, B., Paulsen, S.J., Corbett, D.C., Sazer, D.W., Fortin, C.L., Zaita,
Gelatin: A Platform Bioink for Multiple 3D Biofabrication Technologies. Adv. A.J., Greenfield, P.T., Calafat, N.J., Gounley, J.P., Ta, A.H., et al. (2019). Multi-
Mater. 29, 1703404. vascular networks and functional intravascular topologies within biocompat-
Bhattacharjee, T., Zehnder, S.M., Rowe, K.G., Jain, S., Nixon, R.M., Sawyer, ible hydrogels. Science 364, 458–464.
W.G., and Angelini, T.E. (2015). Writing in the granular gel medium. Sci. Adv. Groll, J., Boland, T., Blunk, T., Burdick, J.A., Cho, D.W., Dalton, P.D., Derby,
1, e1500655. B., Forgacs, G., Li, Q., Mironov, V.A., et al. (2016). Biofabrication: reappraising
Birey, F., Andersen, J., Makinson, C.D., Islam, S., Wei, W., Huber, N., Fan, the definition of an evolving field. Biofabrication 8, 013001.
H.C., Metzler, K.R.C., Panagiotakos, G., Thom, N., et al. (2017). Assembly of Groll, J., Burdick, J.A., Cho, D.W., Derby, B., Gelinsky, M., Heilshorn, S.C.,
functionally integrated human forebrain spheroids. Nature 545, 54–59. Jüngst, T., Malda, J., Mironov, V.A., Nakayama, K., et al. (2018). A definition
Blaeser, A., Duarte Campos, D.F., Puster, U., Richtering, W., Stevens, M.M., of bioinks and their distinction from biomaterial inks. Biofabrication 11,
and Fischer, H. (2016). Controlling Shear Stress in 3D Bioprinting is a Key Fac- 013001.

30 Cell 184, January 7, 2021

 ll
Primer
Hall, B.K., and Miyake, T. (1995). Divide, accumulate, differentiate: cell Lim, K.S., Galarraga, J.H., Cui, X., Lindberg, G.C.J., Burdick, J.A., and Wood-
condensation in skeletal development revisited. Int. J. Dev. Biol. 39, 881–893. field, T.B.F. (2020). Fundamentals and Applications of Photo-Cross-Linking in
Heinrich, M.A., Bansal, R., Lammers, T., Zhang, Y.S., Michel Schiffelers, R., Bioprinting. Chem. Rev. 120, 10662–10694.
and Prakash, J. (2019). 3D-Bioprinted Mini-Brain: A Glioblastoma Model to Lin, N.Y.C., Homan, K.A., Robinson, S.S., Kolesky, D.B., Duarte, N., Moisan,
Study Cellular Interactions and Therapeutics. Adv. Mater. 31, e1806590. A., and Lewis, J.A. (2019). Renal reabsorption in 3D vascularized proximal tu-
bule models. Proc. Natl. Acad. Sci. USA 116, 5399–5404.
Highley, C.B., Rodell, C.B., and Burdick, J.A. (2015). Direct 3D Printing of
Shear-Thinning Hydrogels into Self-Healing Hydrogels. Adv. Mater. 27, Liu, W., Heinrich, M.A., Zhou, Y., Akpek, A., Hu, N., Liu, X., Guan, X., Zhong, Z.,
5075–5079. Jin, X., Khademhosseini, A., and Zhang, Y.S. (2017a). Extrusion Bioprinting of
Shear-Thinning Gelatin Methacryloyl Bioinks. Adv. Healthc. Mater. 6, 1601451.
Hinton, T.J., Jallerat, Q., Palchesko, R.N., Park, J.H., Grodzicki, M.S., Shue,
H.-J., Ramadan, M.H., Hudson, A.R., and Feinberg, A.W. (2015). Three-dimen- Liu, W., Zhang, Y.S., Heinrich, M.A., De Ferrari, F., Jang, H.L., Bakht, S.M., Al-
sional printing of complex biological structures by freeform reversible embed- varez, M.M., Yang, J., Li, Y.-C., Trujillo-de Santiago, G., et al. (2017b). Rapid
ding of suspended hydrogels. Sci. Adv. 1, e1500758. Continuous Multimaterial Extrusion Bioprinting. Adv. Mater. 29, 1604630.

Hiscock, T.W., and Megason, S.G. (2015). Mathematically guided approaches Loterie, D., Delrot, P., and Moser, C. (2020). High-resolution tomographic volu-
to distinguish models of periodic patterning. Development 142, 409–419. metric additive manufacturing. Nat. Commun. 11, 852.

Hughes, C.S., Postovit, L.M., and Lajoie, G.A. (2010). Matrigel: a complex pro- Ma, X., Qu, X., Zhu, W., Li, Y.-S., Yuan, S., Zhang, H., Liu, J., Wang, P., Lai,
tein mixture required for optimal growth of cell culture. Proteomics 10, C.S.E., Zanella, F., et al. (2016). Deterministically patterned biomimetic human
1886–1890. iPSC-derived hepatic model via rapid 3D bioprinting. Proc. Natl. Acad. Sci.
USA 113, 2206–2211.
Hughes, A.J., Miyazaki, H., Coyle, M.C., Zhang, J., Laurie, M.T., Chu, D., Vav-
rusová, Z., Schneider, R.A., Klein, O.D., and Gartner, Z.J. (2018). Engineered Malda, J., Visser, J., Melchels, F.P., Jüngst, T., Hennink, W.E., Dhert, W.J.A.,
Tissue Folding by Mechanical Compaction of the Mesenchyme. Dev. Cell Groll, J., and Hutmacher, D.W. (2013). 25th anniversary article: Engineering hy-
44, 165–178.e6. drogels for biofabrication. Adv. Mater. 25, 5011–5028.
Mammoto, T., and Ingber, D.E. (2010). Mechanical control of tissue and organ
Ihermann-Hella, A., Lume, M., Miinalainen, I.J., Pirttiniemi, A., Gui, Y., Peränen,
development. Development 137, 1407–1420.
J., Charron, J., Saarma, M., Costantini, F., and Kuure, S. (2014). Mitogen-acti-
vated protein kinase (MAPK) pathway regulates branching by remodeling Martin, A.C., Kaschube, M., and Wieschaus, E.F. (2009). Pulsed contractions
epithelial cell adhesion. PLoS Genet. 10, e1004193. of an actin-myosin network drive apical constriction. Nature 457, 495–499.
Ingber, D.E. (2020). Is it Time for Reviewer 3 to Request Human Organ Chip Matai, I., Kaur, G., Seyedsalehi, A., McClinton, A., and Laurencin, C.T. (2020).
Experiments Instead of Animal Validation Studies? Advanced Science 7, Progress in 3D bioprinting technology for tissue/organ regenerative engineer-
2002030. ing. Biomaterials 226, 119536.

Jeon, S., Heo, J.-H., Kim, M.K., Jeong, W., and Kang, H.-W. (2020). High-Pre- McCormack, A., Highley, C.B., Leslie, N.R., and Melchels, F.P.W. (2020). 3D
cision 3D Biol.-Dot Printing to Improve Paracrine Interaction between Multiple Printing in Suspension Baths: Keeping the Promises of Bioprinting Afloat.
Types of Cell Spheroids. Adv. Funct. Mater., 2005324. Trends Biotechnol. 38, 584–593.

Junk, S., and Kuen, C. (2016). Review of Open Source and Freeware CAD Sys- Menshykau, D., Michos, O., Lang, C., Conrad, L., McMahon, A.P., and Iber, D.
tems for Use with 3D-Printing. Procedia CIRP 50, 430–435. (2019). Image-based modeling of kidney branching morphogenesis reveals
GDNF-RET based Turing-type mechanism and pattern-modulating WNT11
Kang, H.-W., Lee, S.J., Ko, I.K., Kengla, C., Yoo, J.J., and Atala, A. (2016). A 3D
feedback. Nat. Commun. 10, 239.
bioprinting system to produce human-scale tissue constructs with structural
integrity. Nat. Biotechnol. 34, 312–319. Miller, E.D., Phillippi, J.A., Fisher, G.W., Campbell, P.G., Walker, L.M., and
Weiss, L.E. (2009). Inkjet printing of growth factor concentration gradients
Kang, D., Ahn, G., Kim, D., Kang, H.-W., Yun, S., Yun, W.-S., Shim, J.-H., and and combinatorial arrays immobilized on biologically-relevant substrates.
Jin, S. (2018). Pre-set extrusion bioprinting for multiscale heterogeneous tis- Comb. Chem. High Throughput Screen. 12, 604–618.
sue structure fabrication. Biofabrication 10, 035008.
Miri, A.K., Nieto, D., Iglesias, L., Goodarzi Hosseinabadi, H., Maharjan, S.,
Kaucka, M., Zikmund, T., Tesarova, M., Gyllborg, D., Hellander, A., Jaros, J., Ruiz-Esparza, G.U., Khoshakhlagh, P., Manbachi, A., Dokmeci, M.R., Chen,
Kaiser, J., Petersen, J., Szarowska, B., Newton, P.T., et al. (2017). Oriented S., et al. (2018). Microfluidics-Enabled Multimaterial Maskless Stereolitho-
clonal cell dynamics enables accurate growth and shaping of vertebrate carti- graphic Bioprinting. Adv. Mater. 30, e1800242.
lage. eLife 6, e25902.
Moldovan, N.I., Hibino, N., and Nakayama, K. (2017). Principles of the Kenzan
Kelly, B.E., Bhattacharya, I., Heidari, H., Shusteff, M., Spadaccini, C.M., and Method for Robotic Cell Spheroid-Based Three-Dimensional Bioprinting. Tis-
Taylor, H.K. (2019). Volumetric additive manufacturing via tomographic recon- sue Eng. Part B Rev. 23, 237–244.
struction. Science 363, 1075–1079.
Mongera, A., Rowghanian, P., Gustafson, H.J., Shelton, E., Kealhofer, D.A.,
Laurent, J., Blin, G., Chatelain, F., Vanneaux, V., Fuchs, A., Larghero, J., and Carn, E.K., Serwane, F., Lucio, A.A., Giammona, J., and Campàs, O. (2018).
Théry, M. (2017). Convergence of microengineering and cellular self-organiza- A fluid-to-solid jamming transition underlies vertebrate body axis elongation.
tion towards functional tissue manufacturing. Nat. Biomed. Eng. 1, 939–956. Nature 561, 401–405.
Lee, H., Kim, J., Choi, Y., and Cho, D.-W. (2020). Application of Gelatin Bioinks Morley, C.D., Ellison, S.T., Bhattacharjee, T., O’Bryan, C.S., Zhang, Y., Smith,
and Cell-Printing Technology to Enhance Cell Delivery Capability for 3D Liver K.F., Kabb, C.P., Sebastian, M., Moore, G.L., Schulze, K.D., et al. (2019).
Fibrosis-on-a-Chip Development. ACS Biomater. Sci. Eng. 6, 2469–2477. Quantitative characterization of 3D bioprinted structural elements under cell
Li, P., Markson, J.S., Wang, S., Chen, S., Vachharajani, V., and Elowitz, M.B. generated forces. Nat. Commun. 10, 3029.
(2018). Morphogen gradient reconstitution reveals Hedgehog pathway design Moroni, L., Burdick, J.A., Highley, C., Lee, S.J., Morimoto, Y., Takeuchi, S.,
principles. Science 360, 543–548. and Yoo, J.J. (2018). Biofabrication strategies for 3D in vitro models and regen-
Li, X., Liu, B., Pei, B., Chen, J., Zhou, D., Peng, J., Zhang, X., Jia, W., and Xu, T. erative medicine. Nat. Rev. Mater. 3, 21–37.
(2020). Inkjet Bioprinting of Biomaterials. Chem. Rev. 120, 10793–10833. Nelson, C.M. (2016). On Buckling Morphogenesis. J. Biomech. Eng. 138,
Lim, K.S., Levato, R., Costa, P.F., Castilho, M.D., Alcala-Orozco, C.R., van 021005.
Dorenmalen, K.M.A., Melchels, F.P.W., Gawlitta, D., Hooper, G.J., Malda, J., Nissen, S.B., Rønhild, S., Trusina, A., and Sneppen, K. (2018). Theoretical tool
and Woodfield, T.B.F. (2018). Bio-resin for high resolution lithography-based bridging cell polarities with development of robust morphologies. eLife 7,
biofabrication of complex cell-laden constructs. Biofabrication 10, 034101. e38407.

Cell 184, January 7, 2021 31

 ll
Primer
Norotte, C., Marga, F.S., Niklason, L.E., and Forgacs, G. (2009). Scaffold-free Skylar-Scott, M.A., Uzel, S.G.M., Nam, L.L., Ahrens, J.H., Truby, R.L., Damar-
vascular tissue engineering using bioprinting. Biomaterials 30, 5910–5917. aju, S., and Lewis, J.A. (2019b). Biomanufacturing of organ-specific tissues
Ong, C.S., Fukunishi, T., Zhang, H., Huang, C.Y., Nashed, A., Blazeski, A., DiS- with high cellular density and embedded vascular channels. Science Ad-
ilvestre, D., Vricella, L., Conte, J., Tung, L., et al. (2017). Biomaterial-free three- vances 5, eaaw2459.
dimensional bioprinting of cardiac tissue using human induced pluripotent Song, K.H., Highley, C.B., Rouff, A., and Burdick, J.A. (2018). Complex 3D-
stem cell derived cardiomyocytes. Sci. Rep. 7, 4566. Printed Microchannels within Cell-Degradable Hydrogels. Adv. Funct. Mater.
Ouyang, L., Armstrong, J.P.K., Lin, Y., Wojciechowski, J.P., Lee-Reeves, C., 28, 1801331.
Hachim, D., Zhou, K., Burdick, J.A., and Stevens, M.M. (2020). Expanding
and optimizing 3D bioprinting capabilities using complementary network bio- Spurlin, J.W., Siedlik, M.J., Nerger, B.A., Pang, M.-F., Jayaraman, S., Zhang,
inks. Science Advances 6, eabc5529. R., and Nelson, C.M. (2019). Mesenchymal proteases and tissue fluidity
remodel the extracellular matrix during airway epithelial branching in the em-
Ozbolat, I.T., and Hospodiuk, M. (2016). Current advances and future perspec-
bryonic avian lung. Development 146, dev175257.
tives in extrusion-based bioprinting. Biomaterials 76, 321–343.
Pauli, C., Hopkins, B.D., Prandi, D., Shaw, R., Fedrizzi, T., Sboner, A., Sailer, Sun, W., Starly, B., Daly, A.C., Burdick, J.A., Groll, J., Skeldon, G., Shu, W., Sa-
V., Augello, M., Puca, L., Rosati, R., et al. (2017). Personalized In Vitro and kai, Y., Shinohara, M., Nishikawa, M., et al. (2020). The bioprinting roadmap.
In Vivo Cancer Models to Guide Precision Medicine. Cancer Discov. 7, Biofabrication 12, 022002.
462–477.
Szabó, A., Melchionda, M., Nastasi, G., Woods, M.L., Campo, S., Perris, R.,
Rogers, K.W., and Schier, A.F. (2011). Morphogen gradients: from generation and Mayor, R. (2016). In vivo confinement promotes collective migration of
to interpretation. Annu. Rev. Cell Dev. Biol. 27, 377–407. neural crest cells. J. Cell Biol. 213, 543–555.
Rossi, G., Manfrin, A., and Lutolf, M.P. (2018). Progress and potential in orga-
Tibbitt, M.W., and Anseth, K.S. (2009). Hydrogels as extracellular matrix
noid research. Nat. Rev. Genet. 19, 671–687.
mimics for 3D cell culture. Biotechnol. Bioeng. 103, 655–663.
Sadati, M., Taheri Qazvini, N., Krishnan, R., Park, C.Y., and Fredberg, J.J.
(2013). Collective migration and cell jamming. Differentiation 86, 121–125. van den Brink, S.C., Alemany, A., van Batenburg, V., Moris, N., Blotenburg, M.,
Scheideler, O.J., Yang, C., Kozminsky, M., Mosher, K.I., Falcón-Banchs, R., Vivié, J., Baillie-Johnson, P., Nichols, J., Sonnen, K.F., Martinez Arias, A., et al.
Ciminelli, E.C., Bremer, A.W., Chern, S.A., Schaffer, D.V., and Sohn, L.L. (2020). Single-cell and spatial transcriptomics reveal somitogenesis in gastru-
(2020). Recapitulating complex biological signaling environments using a mul- loids. Nature 582, 405–409.
tiplexed, DNA-patterning approach. Science Advances 6, eaay5696. Yi, H.-G., Jeong, Y.H., Kim, Y., Choi, Y.-J., Moon, H.E., Park, S.H., Kang, K.S.,
Skylar-Scott, M.A., Mueller, J., Visser, C.W., and Lewis, J.A. (2019a). Voxe- Bae, M., Jang, J., Youn, H., et al. (2019). A bioprinted human-glioblastoma-on-
lated soft matter via multimaterial multinozzle 3D printing. Nature 575, a-chip for the identification of patient-specific responses to chemoradiother-
330–335. apy. Nat. Biomed. Eng. 3, 509–519.

32 Cell 184, January 7, 2021