Search in book...
Toggle Font Controls
Create new playlist

Name your new playlist

Playlist description (optional)
Sign In

Email address

Password

Forgot Password?

or

Continue with Facebook

Continue with Google
Sign Up

Full Name

Email address

Confirm Email Address

Password

or

Continue with Facebook

Continue with Google

Droplet-Based Bioprinting^∗

Abstract

Droplet-based bioprinting (DBB) offers greater advantages due to its simplicity and agility with precise control on deposition of biologics including cells, growth factors, genes, drugs, and biomaterials, and has been a prominent technology in the bioprinting community. Due to its immense versatility, DBB technology has been adopted by various application areas, including, but not limited to, tissue engineering and regenerative medicine, transplantation and clinics, pharmaceutics and high-throughput screening, and cancer research. Despite the great benefits, the technology currently faces several challenges such as a narrow range of available bioink materials, bioprinting-induced cell damage at substantial levels, limited mechanical and structural integrity of bioprinted constructs, and restrictions on the size of constructs due to lack of vascularization and porosity. This chapter presents DBB technology and its modalities including inkjet, electrohydrodynamic, acoustic, and microvalve bioprinting. The recent notable studies are highlighted and the relevant bioink and substrate biomaterials are expounded. In addition, current limitations are discussed and future prospects are presented to the reader.

Keywords

Acoustic bioprinting; Bioink; Droplet-based bioprinting; Electrohydrodynamic bioprinting; Inkjet bioprinting; Microvalve bioprinting

The only merit of which I personally am conscious was that of having pleased myself by my studies, and any results that may be due to my researches were owing to the fact that it has been a pleasure for me to become a physicist

John William Strutt, 3rd Baron Rayleigh

5.1. Introduction

Droplet-based bioprinting (DBB) has its roots in inkjet printing technology, which has its beginning in the 1950s when Elmqvist of Siemens patented the first practical inkjet device in 1951 (Le, 1998). Later, Sweet from Stanford University spearheaded the development of continuous inkjet (CIJ) printing system in 1960s. In 1970s, Zoltan, Kyser, and Sears pioneered the development of drop-on-demand (DOD) inkjet printing system. Their invention was licensed in the first commercial DOD inkjet printer, the Siemens PT-80, in 1977.

The idea of printing biologics was first introduced by Klebe in 1987 when he used a commercially available Hewlett–Packard (HP) thermal DOD inkjet printer to deposit a bioink solution comprising collagen and fibronectin (Klebe, 1988). Afterward, the first inkjet-based 3D printer was developed by Objet Geometries in 2000 (Wohlers and Gornet, 2014). In 2003, Boland demonstrated the feasibility of using a modified thermal DOD inkjet printer to deposit living cells in a viable form (Wilson and Boland, 2003) and introduced the concept of inkjet bioprinting (Boland et al., 2007). Subsequently, Nakamura's group successfully fabricated viable 3D tubular tissue constructs using a commercially available electrostatic DOD inkjet printer (Nishiyama et al., 2008). Later, several research groups have successfully adopted DBB technologies for bioprinting of a wide array of cells for various purposes, including, but not limited to, bioprinting for stem cell research (Faulkner-Jones et al., 2015; Gurkan et al., 2014; Xu et al., 2013a; Lee et al., 2010), tissue engineering (Xu et al., 2013a,b; Fang et al., 2012), controlled release (Cooper et al., 2010), transplantation (Xu et al., 2013b, 2010), drug screening (Rodríguez-Dévora et al., 2012), high-throughput arrays (Suntivich et al., 2014), and cancer research (Xu et al., 2011; Fang et al., 2012).

Despite the commonly used extrusion-based bioprinting (EBB) (Ozbolat and Hospodiuk, 2016) and the high-precision laser-based bioprinting (LBB) (Odde and Renn, 1999; Barron et al., 2004; Koch et al., 2010), DBB (Xu et al., 2005, 2010; Eagles et al., 2006; Demirci and Montesano, 2007b) offers several advantages due to its simplicity, agility, versatility, and the great control over the deposition pattern. It enables bioprinting with controlled volumes of bioink deposition at predefined locations (Murphy and Atala, 2014) facilitating spatially heterocellular constructs with well-defined positioning of cells (Xu et al., 2012). DBB, as shown in Fig. 5.1, comprises inkjet (Klebe, 1988; Xu et al., 2005, 2012; Murphy and Atala, 2014), electrohydrodynamic (EHD) jet (Jayasinghe et al., 2006; Onses et al., 2015; Gasperini et al., 2013; Eagles et al., 2006), acoustic-droplet-ejection (or simply acoustic) (Tasoglu and Demirci, 2013), and microvalve bioprinting (Faulkner-Jones et al., 2013, 2015; Lee et al., 2009a; Xu et al., 2011; Gurkan et al., 2014). Inkjet bioprinting is classified into two: (1) CIJ and (2) DOD bioprinting. CIJ bioprinting leverages Rayleigh–Plateau instability to break bioink jets into droplets. DOD bioprinting, on the other hand, uses thermal or piezoelectric actuators or electrostatic forces to generate droplets. In contrast, EHD jet bioprinting uses high ranges of electric voltage to eject droplets. Whereas acoustic bioprinting uses acoustic waves to produce droplets, microvalve bioprinting uses a solenoid pump to eject droplets.

Figure 5.1 Classification of droplet-based bioprinting modalities.

5.2. Inkjet Bioprinting

Inkjet bioprinting physically manipulates a bioink solution to generate droplets. It leverages gravity, atmospheric pressure, and the fluid mechanics of the bioink solution to eject droplets onto a receiving substrate. To dispense the bioink, the following relations should be satisfied for a set of physical properties (Saunders and Derby, 2014):

$Z = \frac{\sqrt{σ ρ l}}{η} = \frac{\sqrt{We}}{Re} = \frac{1}{Oh}$ $Z = \frac{\sqrt{σ ρ l}}{η} = \frac{\sqrt{We}}{Re} = \frac{1}{Oh}$

(5.1)

where σ, ρ, and η are the surface tension, density, and dynamic viscosity of the bioink, respectively. In Eq. (5.1), l is a characteristic length (the diameter of the nozzle tip), Re is the Reynolds number, We is the Weber number, and Oh is the Ohnesorge number. Reis et al. suggested a limited range for inkjet printing (1 < Z < 10). If Z < 1, the bioink is too viscous to eject, and if Z > 10, inkjet printing generates satellite formation accompanying the main droplet (Reis et al., 2005). Other experimental work reported slightly larger range of limits (1 < Z < 14) (Derby, 2011; Jang et al., 2009).

5.2.1. Continuous Inkjet Bioprinting

In CIJ bioprinting, the bioink solution is forced under pressure through a nozzle, which subsequently breaks up into a stream of droplets owing to Rayleigh–Plateau instability (Derby, 2008) as illustrated in Fig. 5.2. The Rayleigh–Plateau instability is a physical phenomenon, where a thread of jet breaks up into droplets to minimize its surface tension. It occurs when the length of a cylindrical volume of liquid jet surrounded by a gas exceeds its radius by a certain limit (Rayleigh, 1878; Cardoso and Dias, 2006; Atkins and Escudier, 2013). During equilibrium state, the jet flows with a radius R₀ and pressure P₀ (where

P_{0} = \frac{σ}{R_{0}}

$P_{0} = \frac{σ}{R_{0}}$

) assuming no external pressure and the influence of gravity is neglected (see Fig. 5.2C). Under perturbed conditions, the jet radius (R) can be represented as a wave function:

Figure 5.2 (A) Inkjet bioprinter with (B) the droplet formation. (C) An illustration of the intermediate stage of jet breaking up into droplets.

$R = R_{0} + ε e^{ω t + i k z}$ $R = R_{0} + ε e^{ω t + i k z}$

(5.2)

where ε is the perturbation amplitude (ε << R), ω is the growth rate and k is the wave number of the disturbance in z-direction. The wavelength (λ) can be represented as λ = 2π/k. The radial (u_r) and the axial (u_z) component of the perturbation velocity, and the perturbation pressure (P) can be represented as:

$u_{r} = R (r) e^{ω t + i k z}$ $u_{r} = R (r) e^{ω t + i k z}$

(5.3)

$u_{z} = Z (r) e^{ω t + i k z}$ $u_{z} = Z (r) e^{ω t + i k z}$

$P = P (r) e^{ω t + i k z}$ $P = P (r) e^{ω t + i k z}$

Substituting the previously mentioned perturbation fields into the cylindrical Navier–Stokes equations, one can obtain the following momentum (r,z) and continuity (m) equations:

$r : ω R = - \frac{1}{ρ} \frac{d P}{d r}$ $r : ω R = - \frac{1}{ρ} \frac{d P}{d r}$

(5.4)

$z : ω Z = - \frac{i k}{ρ} P$ $z : ω Z = - \frac{i k}{ρ} P$

$m : \frac{\partial R}{\partial r} + \frac{R}{r} + i k z = 0$ $m : \frac{\partial R}{\partial r} + \frac{R}{r} + i k z = 0$

By eliminating P and Z to get a differential equation for R, the following equation can be obtained:

$r^{2} \frac{d^{2} R}{d r^{2}} + r \frac{d R}{d r} - (1 + k^{2} r^{2}) R = 0$ $r^{2} \frac{d^{2} R}{d r^{2}} + r \frac{d R}{d r} - (1 + k^{2} r^{2}) R = 0$

(5.5)

The solution of Eq. (5.5) is a Bessel function of the first kind, which yields the following equation:

$R (r) = C I_{1} (k r)$ $R (r) = C I_{1} (k r)$

(5.6)

In Eq. (5.6), C is a constant. Using the relation between P and R given in Eq. (5.4) and the Bessel function identity

I_{0}^{'} (ζ) = I_{0} (ζ)

$I_{0}^{'} (ζ) = I_{0} (ζ)$

, the perturbation pressure can be obtained as the following:

$P (r) = - \frac{ω ρ C}{k} I_{0} (k r)$ $P (r) = - \frac{ω ρ C}{k} I_{0} (k r)$

(5.7)

To find C, equations for boundary conditions should be further developed. The first boundary condition is the normal stress balance on the free surface:

$P_{0} + P = σ \nabla \cdot n = σ [\frac{1}{R_{1}} + \frac{1}{R_{2}}]$ $P_{0} + P = σ \nabla \cdot n = σ [\frac{1}{R_{1}} + \frac{1}{R_{2}}]$

(5.8)

In Eq. (5.8), R₁ and R₂ are the principle radii of curvature of the jet and perturbation, respectively (see Fig. 5.2C). R₁ can be approximated as:

\frac{1}{R_{1}} = \frac{1}{R_{0} + ε e^{ω t + i k z}} ≅ \frac{1 - \frac{(ε e^{ω t + i k z})}{R_{0}}}{R_{0}}

$\frac{1}{R_{1}} = \frac{1}{R_{0} + ε e^{ω t + i k z}} ≅ \frac{1 - \frac{(ε e^{ω t + i k z})}{R_{0}}}{R_{0}}$

. The second radius of curvature can be obtained using the second derivative definition as

R_{2} = \frac{1}{ε k^{2} e^{ω t + i k z}}

$R_{2} = \frac{1}{ε k^{2} e^{ω t + i k z}}$

. The second boundary condition is the kinematic condition on the free surface energy:

$\frac{\partial R}{\partial t} = u \cdot n ≅ u_{r}$ $\frac{\partial R}{\partial t} = u \cdot n ≅ u_{r}$

(5.9)

By substituting Eqs. (5.3) and (5.6) into Eq. (5.9), the constant C can be obtained as

C = \frac{ε ω}{I_{1} (k R_{0})}

$C = \frac{ε ω}{I_{1} (k R_{0})}$

. Then, by combining Eqs. (5.3) and (5.7) and inserting into Eq. (5.8), the growth rate can be reformulated as:

$ω^{2} = (1 - k^{2} R_{0}^{2}) \frac{k σ}{ρ R_{0}^{2}} \frac{I_{1} (k R_{0})}{I_{0} (k R_{0})}$ $ω^{2} = (1 - k^{2} R_{0}^{2}) \frac{k σ}{ρ R_{0}^{2}} \frac{I_{1} (k R_{0})}{I_{0} (k R_{0})}$

(5.10)

From Eq. (5.10), it can be concluded that unstable modes can be obtained when kR_o < 1, and the perturbation grows exponentially and eventually the jet distorts itself to minimize its potential energy and breaks up into a stream of droplets. The condition kR_o < 1 can be reformulated as 2πR_o < λ, which draws the conclusion that unstable modes can be achieved when the perimeter of the perturbation is less than its wavelength.

5.2.2. Drop-on-Demand Inkjet Bioprinting

DOD inkjet bioprinting is preferred over CIJ bioprinting for tissue bioprinting purposes. DOD inkjet bioprinters generate droplets when required, which makes them more economical, handy to control, and easy to pattern biologics (Derby, 2010). DOD inkjet bioprinters rely on three different mechanisms to generate droplets including (1) thermal inkjet (TIJ), (2) piezoelectric inkjet (PIJ), and (3) electrostatic bioprinting. TIJ bioprinters generate droplets using a thermal actuator whereas PIJ bioprinters generate droplets using a piezoelectric actuator (Wijshoff, 2010). The two techniques are identical in other aspects. Electrostatic bioprinting, in contrast, ejects droplets using an electrostatic force (Kamisuki et al., 1998; Nishiyama et al., 2008).

DOD bioprinters consist of a single or multiple printheads. Each printhead contains a fluid chamber and a single or multiple nozzles. Fig. 5.3 demonstrates a sample DOD bioprinter for high-throughput fabrication of mini–tissue models. The bioink stored in the fluid chamber is held in place by the surface tension at the nozzle orifice (Derby, 2010). Pressure pulses are introduced in the fluid chamber through means of a thermal or a piezoelectric actuator such that a droplet is ejected when the bioink overcomes the surface tension. Some printhead assemblies may require pneumatic pressure (static pressure through means of pressurized air) commonly referred to as the back pressure to supplement the pressure pulses to overcome the surface tension.

5.2.2.1. Thermal Inkjet Bioprinting

The thermal actuator in TIJ bioprinting locally heats the bioink solution when a voltage pulse is applied. The localized heating forms a vapor bubble as shown in Fig. 5.4A. Subsequently, the bubble expands rapidly and collapses (explodes), which generates a pressure pulse inside the fluid chamber (Cui and Boland, 2009; Mohebi and Evans, 2002). Consequently, the bioink overcomes the surface tension at the nozzle orifice and a droplet is ejected.

TIJ bioprinters are capable of dispensing various biologics such as proteins (Roth et al., 2004) and mammalian cells. Bioprinted biologics include Chinese hamster ovary cells (Xu et al., 2005), rat aortic smooth muscle cells (Roth et al., 2004), canine smooth muscle cells (Xu et al., 2013a), rat neural cells (Xu et al., 2005, 2006), bovine aortal endothelial cells (bECs) (Xu et al., 2013a), beta-TC6 cells (Xu et al., 2008), feline cardiomyocytes (Xu et al., 2009), H1 cardiomyocytes (Xu et al., 2009), human microvascular endothelial cells (HMVECs) (Yanez et al., 2014; Cui and Boland, 2009), neonatal human dermal fibroblasts (NHDFs) (Yanez et al., 2014), neonatal human epidermal keratinocytes (Yanez et al., 2014), human articular chondrocytes (Cui et al., 2012a,b), human mesenchymal stem cells (hMSCs) (Gao et al., 2014, 2015), and human amniotic fluid-derived stem cells (hAFSCs) (Xu et al., 2013a). The bioink temperature during bubble formation reaches 200–300ᵒC for a few milliseconds (ms) (Roth et al., 2004). Moreover, the orifice diameter, in general, is about 50 μm (Xu et al., 2006) and the diameter of the ejected droplets is around 30–60 μm (Xu et al., 2008). Postbioprinting cell viability ranges from 75% to 90% at cell concentrations from 10⁵ to 10⁷ cells/mL (Xu et al., 2013a, 2005; Cui et al., 2012a,b; Gao et al., 2014, 2015). Furthermore, bioprinted cells retain their functional phenotype and genotype, and proliferation capacity (Xu et al., 2006, 2008, 2013a; Cui et al., 2012a). In summary, bioprinting studies pertaining TIJ bioprinting technology to date have been to assess its impact on the functionality of bioprinted cells. For example, Fig. 5.5A1 and A2 shows TIJ-bioprinted neurons tagged with neural markers 15 days after bioprinting. As seen in the figure, TIJ bioprinting did not affect the neuronal phenotype and electrophysiological functions of the bioprinted neurons. Furthermore, a myriad of studies have utilized modified commercial 2D TIJ printers and explored their role in tissue regeneration such as cardiac tissue (see Fig. 5.5B; Xu et al., 2009), vascular tissue (Cui and Boland, 2009), and in situ cartilage repair (Cui et al., 2012a).

Figure 5.3 A drop-on-demand (DOD) bioprinter with multiple printhead is used for high-throughput fabrication of mini–tissue models in an agarose mold with an array of microwells.

Figure 5.4 The mechanism of (A) thermal and (B) piezoelectric drop-on-demand bioprinting.

5.2.2.2. Piezoelectric Inkjet Bioprinting

The piezoelectric actuator in PIJ bioprinting changes its shape when a voltage pulse is applied, as depicted in Fig. 5.4B. This deforms the fluid chamber (Wijshoff, 2010) in which a sudden change in the volume of the fluid chamber causes a pressure wave. As a result, the surface tension at the nozzle orifice is overcome and a droplet of the bioink is ejected (Singh et al., 2010).

Piezoelectric bioprinters, similar to TIJ bioprinters, are suitable for bioprinting various biologics including proteins (Pataky et al., 2012; Wilson and Boland, 2003), and bacterial (Choi et al., 2011) and mammalian cells [i.e., bECs (Wilson and Boland, 2003), HeLa cells (Arai et al., 2011; Yusof et al., 2011), National Institutes of Health 3T3 mouse fibroblasts (NIH 3T3 cells) (Pataky et al., 2012; Xu et al., 2012; Christensen et al., 2015), C2C12 mouse myoblasts (Matsusaki et al., 2013), MCF-7 breast cancer cells (Cheng et al., 2014), human umbilical vein endothelial cells, and NHDFs (Matsusaki et al., 2013)]. Generally, the nozzle orifice diameters range from 21.5 μm (for bioprinting bacteria) (Choi et al., 2011) to 120 μm (Xu et al., 2012; Cheng et al., 2014; Christensen et al., 2015) whereas the ejected droplet diameters range from 50 (Choi et al., 2011; Cheng et al., 2014) to 100 μm (Xu et al., 2012). Meanwhile, a postbioprinting cell viability of 70–95% was observed for cell concentrations of 10⁵–10⁷ cells/mL (Wilson and Boland, 2003; Xu et al., 2012; Matsusaki et al., 2013; Christensen et al., 2015). Overall, studies on PIJ bioprinting technology have thus far investigated the feasibility and possibility of fabricating 3D tissue constructs (Arai et al., 2011; Xu et al., 2012; Christensen et al., 2015; Matsusaki et al., 2013). For example, Fig. 5.5C depicts a PIJ-bioprinted vascularlike tissue construct with horizontal and vertical bifurcations. The tissue construct comprising alginate and NIH 3T3 cells was fabricated by a single nozzle PIJ bioprinter. Moreover, some studies optimized process parameters, including the bioink constituents (Pataky et al., 2012), piezoelectric element actuation modes (pull–push and push–pull) (Yamaguchi et al., 2012), and voltage pulse characteristics (amplitude, rise and fall times, dwell time, echo time, and frequency) (Xu et al., 2012; Christensen et al., 2015), to improve process efficiency. Furthermore, few other studies attempted to improve process reliability (Cheng et al., 2014) and control the number of encapsulated cells in ejected droplets for potential applications in the areas of diagnostics, therapeutics, and cell biology (Yusof et al., 2011; Yamaguchi et al., 2012).

Figure 5.5 Droplet-based bioprinting of cells and biologics. Thermal drop-on-demand bioprinting of neurons tagged with neuronal markers after 15 days of culture, (A1) immunostaining of dendrites of rat embryonic cortical neurons with MAP2 monoclonal antibodies (green), (A2) immunostaining of dendrites of rat embryonic hippocampal with anti-MAP2 monoclonal antibodies (green) and axons of the neurons with antineurofilament monoclonal antibodies (red) (Reproduced/adapted with permission from Xu et al. (2006)); (B) thermal inkjet bioprinting of 3D cardiac tissue (Reproduced/adapted with permission from Xu et al. (2009)); (C) piezoelectric inkjet bioprinting of tissue constructs (NIH 3T3 cells with sodium alginate) with bifurcations (Reproduced/adapted with permission from Christensen et al. (2015)); (D) electrostatic bioprinting of alginate tubular construct (Reproduced/adapted with permission from Nishiyama et al. (2008)); (E) Electrohydrodynamic jet bioprinting of porcine vascular smooth muscle cells after 40 days postbioprinting (Reproduced/adapted with permission from Jayasinghe (2007)); (F) acoustic bioprinting of AML-12 cells after 12 days postbioprinting (Reproduced/adapted with permission from Demirci and Montesano, 2007b)); microvalve bioprinting of fibroblasts and keratinocytes, (G1) immunostained 3D image of the cells and its side views, (G2) keratin layer of KC, (G3) β-tubulin of keratinocytes and fibroblasts (Reproduced/adapted with permission from Lee et al. (2009a)).

5.2.2.3. Electrostatic Bioprinting

Electrostatic bioprinters, identical to PIJ bioprinters, generate droplets by temporarily increasing the volume of the fluid chamber, without heating the bioink unlike TIJ bioprinters (Kamisuki et al., 1998). The brief increase in the volume of the fluid chamber as well as the bioink solution is achieved through the means of a pressure plate as shown in Fig. 5.6. The pressure plate deflects when a voltage pulse is applied between the plate itself and an electrode. The pressure plate regains its original shape in the absence of the voltage pulse and subsequently ejects droplets. Electrostatic bioprinters have been primarily used by Nakamura's group (Nishiyama et al., 2008) to fabricate 3D acellular (see Fig. 5.5D) as well as cellular constructs comprising of HeLa cells (6 × 10⁶ cells/mL with 70% postbioprinting viability).

5.3. Electrohydrodynamic Jet Bioprinting

DOD bioprinters generate droplets by propelling bioink solutions through a nozzle. Hence, a very high level of pressure is required when ejecting cell-laden droplets through a nozzle with an extremely small orifice diameter, which is at times harmful to cells. In contrast, EHD bioprinters use an electric field to pull the bioink droplets through the orifice obviating the need for a substantially high pressure (Onses et al., 2015). As a result, EHD bioprinters are ideal for bioprinting applications requiring nozzles with very small orifice diameters (≤100 μm) and highly concentrated bioink solutions (up to 20% weight by volume) (Jayasinghe et al., 2006).

The working mechanism of EHD bioprinting is illustrated in Fig. 5.7 (Sutanto et al., 2012; Gasperini et al., 2015; Onses et al., 2015), where a bioink solution is fed through a metallic nozzle using a certain back pressure (i.e., pneumatic or mechanical pressure) such that the bioink forms a spherical meniscus at the tip of the nozzle owing to the surface tension (Poellmann et al., 2011). A high voltage range (0.5–20 kV) (Hayati et al., 1986; Gasperini et al., 2015) is applied between the nozzle and the substrate, which generates an electric field (as a function of the applied voltage and the distance between the nozzle and the substrate) between them. As a result, the electric field leads to the accumulation of mobile ions in the bioink near the surface of the suspended bioink meniscus. Subsequently, the Columbic or the electrostatic repulsions between the ions deform the meniscus into a conical shape called a “Taylor cone.” Consequently, bioink droplets are ejected when the electrostatic stresses overcome the surface tension at the orifice under a sufficiently high voltage depending on the process parameters. The strength of the electric field (applied voltage), the bioink flow rate, and bioink properties (including cell type and concentration (Jayasinghe and Townsend-Nicholson, 2006)) determine the jetting mode (Kim et al., 2007; Onses et al., 2015) and cell viability (Workman et al., 2014). Dripping mode is observed at low voltages and flow rates, whereas streams of distinct droplets are seen at intermediate voltages and/or flow rates. At the same time, a continuous stream of the bioink, known as cone-jet-mode (continuous presence of a Taylor cone), is observed at high voltages.

Figure 5.6 The mechanism of electrostatic bioprinting.

Figure 5.7 The mechanism of electrohydrodynamic jet bioprinting.

EHD jet bioprinters are suitable for bioprinting proteins, such as collagen (Kim et al., 2007) and antibody immunoglobulin G (IgG) (Poellmann et al., 2011), without any loss of their functionality. In addition, EHD bioprinters are capable of bioprinting mammalian cells including Jurkat cells (Jayasinghe et al., 2006), human astrocytoma cells (Jayasinghe and Townsend-Nicholson, 2006), mouse CAD (Cath.a-differentiated) cells (Eagles et al., 2006), hepatocytes G2 cells (Xie and Wang, 2007), THP-1 cells (Workman et al., 2014), white blood cells (Mongkoldhumrongkul et al., 2009), erythrocytes (red blood cells) (Mongkoldhumrongkul et al., 2009), B50 rat neuronal cells (Gasperini et al., 2013), and 3T3 cells (Gasperini et al., 2015). At times, the applied voltage ranges from 3 to 20 kV for a flow rate ranging from 10⁻¹⁰ m³/s (0.36 mL/h) to 10⁻⁸ m³/s (36 mL/h). Although the nozzle orifice diameter ranges from 100 to 1000 μm, the ejected droplet diameters range from 10 to 2000 μm depending on the bioink constituents (i.e., media, hydrogels, etc.) and the process parameters (i.e., applied voltage). At the same time, a postbioprinting cell viability over 90% was reported for cell concentrations of 10⁶–10⁷ cells/mL, where bioprinted cells retained their functionality, phenotype and genotype, and proliferation capacities (Jayasinghe et al., 2006; Mongkoldhumrongkul et al., 2009). EHD jet bioprinting studies to date have investigated the practicality of bioprinting living cells and for understanding the effect of process parameters on characteristics of ejected droplets. For example, Fig. 5.5E shows EHD-bioprinted porcine vascular smooth muscle cells (PVSMCs) after 40 days postbioprinting. Subsequently, flow cytometric analysis indicated that the bioprinted cells retained their viability (Jayasinghe, 2007). In summary, EHD bioprinters generate a continuous jet or a stream of droplets at a time rather than a single droplet. Hence, they are not suitable for bioprinting applications requiring very accurate placement of cells.

5.4. Acoustic Bioprinting

Acoustic bioprinting, as illustrated in Fig. 5.8, employs a gentle acoustic field to eject droplets from an open pool unlike inkjet or EHD bioprinting, which ejects droplets through a nozzle (Demirci and Montesano, 2007b). As a result, the bioink and the constituent living cells are not exposed to detrimental stressors such as heat, high pressure, large voltage, and significant shear stress during droplet ejection. Typically, an acoustic bioprinter consists of a single or an array of 2D microfluidic channels, in which the bioink is held in place owing to the surface tension at the small channel exit. Further, the bioprinter setup generally consists of a piezoelectric substrate and interdigitated gold rings placed on the substrate to generate surface acoustic waves on demand. The waves are circular in geometry and form an acoustic focal point at the interface between the air and the bioink near the channel exit. The droplets are ejected when the force, exerted by the acoustic radiation at the focal point, exceeds the surface tension at the exit of the channel (Demirci, 2006; Demirci and Montesano, 2007b).

Bioprinting often involves printhead and/or substrate movement; however, a moving printhead and/or a substrate can introduce undesirable disturbances in acoustic-based bioprinting when compared to nozzle-based systems including inkjet and EHD bioprinters. Consequently, the disturbances can bring loss of control over the droplet ejection. In addition, gentle acoustic fields may not be capable of ejecting droplets of viscous bioinks such as hydrogels with high cell concentrations. Sadly, studies investigating acoustic-based bioprinting are very few to date. Acoustic-based bioprinting has enabled bioprinting of a myriad of cell types including mouse embryonic stem cells, 3T3 fibroblasts, AML-12 hepatocytes, Raji cells (a lymphocyte-like B-cell line), HL-1 cardiomyocytes, C2C12 myofibroblasts, MDA MB 231 breast cancer cells, and HEK 293 cells (Demirci and Montesano, 2007b; Fang et al., 2012). Further, the droplet diameter range is specific to the bioink composition and the bioprinter, and is usually between 10 and 500 μm. At the same time, the observed postbioprinting cell viability was above 90% for cell concentrations of 10⁵–10⁷ cells/mL. For example, Fig. 5.5F shows acoustically bioprinted AML-12 cells after 12 days postbioprinting (Demirci and Montesano, 2007b). The bioprinted cells were viable and became confluent in 12 days.

Figure 5.8 The mechanism of acoustic-droplet-ejection bioprinting.

5.5. Microvalve Bioprinting

Microvalve bioprinting uses an electromechanical valve to generate droplets as shown in Fig. 5.9 (Moon et al., 2010; Lee et al., 2009a; Faulkner-Jones et al., 2013). The bioink in the fluid chamber is pressurized (back pressure) and the nozzle orifice is gated by a microvalve. A typical microvalve consists of a solenoid coil and a plunger, which blocks the orifice. The valve coil generates a magnetic field when a voltage pulse is applied and the magnetic field pulls the plunger upward. As a result, the nozzle is unblocked and the bioink is ejected out when the back pressure is sufficiently large enough to overcome the surface tension at the orifice. The back pressure and the valve-gating time determine the mode of the droplet generation, either CIJ or DOD.

Figure 5.9 The mechanism of microvalve (solenoid) bioprinting.

Microvalve bioprinters are favorable for bioprinting various proteins such as collagen (Lee et al., 2009a, 2009b; Moon et al., 2010; Xu et al., 2010), bone morphogenetic protein-2 (BMP-2), and transforming growth factor-β1 (TGF-β1) (Gurkan et al., 2014). Additionally, they are suitable for bioprinting various cell types, including, but not limited to, AML-12 cells (Demirci and Montesano, 2007a), rat embryonic astrocytes, rat embryonic neurons (Lee et al., 2009b), human fibroblasts (Lee et al., 2009a; Xu et al., 2011), human keratinocytes (Lee et al., 2009a), rat primary bladder SMCs (Moon et al., 2010; Xu et al., 2010), epithelial human ovarian cancer cells (OVCAR-5) (Xu et al., 2011), C2C12 cells (Ferris et al., 2013), human embryonic stem cells (hESCs) (Faulkner-Jones et al., 2013, 2015), hMSCs (Gurkan et al., 2014), and human-induced pluripotent stem cells (hiPSCs) (Faulkner-Jones et al., 2015). Process parameters, such as pneumatic pressure, nozzle geometry, cell concentration, and the bioink constituents, determine the droplet volume as well as the cell viability (Faulkner-Jones et al., 2015).

Process parameters, including a nozzle orifice diameter of 150 μm, a pneumatic pressure ranging 6.89–20.7 kPa, and a valve open/close duration of 200 μs, generate droplets of 250 μm in diameter (Lee et al., 2009a, 2009b). Whereas process parameters, including an orifice diameter of 101.6 μm, a pneumatic pressure of 13.79 kPa, and a valve open/close duration of 100 μs, generate droplets of 300 μm in diameter (Faulkner-Jones et al., 2013). At the same time, a postbioprinting cell viability of 90% or greater was reported for cell concentrations of 10⁵–10⁷ cells/mL (Demirci and Montesano, 2007a; Lee et al., 2009a,b; Moon et al., 2010; Xu et al., 2010, 2011; Ferris et al., 2013; Faulkner-Jones et al., 2013, 2015; Gurkan et al., 2014). Also the bioprinted cells retained their functionality, phenotype and genotype, and proliferation capacity (see Fig. 5.5G1 and G3) (Lee et al., 2009a). In addition, the differentiation capacity of human stem cells was not affected by the bioprinting process (Gurkan et al., 2014; Faulkner-Jones et al., 2015).

In summary, studies pertaining microvalve bioprinting technology to date have examined the impact of bioprinting process on cell viability and functionality. Indeed, studies demonstrating the fabrication of complex 3D tissue constructs, similar to PIJ studies (Christensen et al., 2015; Xu et al., 2012), are lacking. Overall, microvalve bioprinters require low range of pneumatic pressure compared to that of PIJ bioprinters and hence they are less prone cell injury and damage (Lee et al., 2009a). However, microvalve bioprinters dispense larger droplets than other DBB modalities (including TIJ, PIJ, and EHD bioprinters) when identical nozzles are used (Tasoglu and Demirci, 2013). Thus the resolution of microvalve bioprinters is lower than that of TIJ and PIJ bioprinters.

Each DBB modality has its own strengths and limitations, and its use should be carefully considered according to the desired application area and tissue type to be bioprinted. Table 5.1 compares different DBB modalities according to their bioprinting parameters (i.e., nozzle size, cell viability, droplet size, operating conditions, etc.) and present their advantages and disadvantages.

Table 5.1

Comparison of Droplet-Based Bioprinting Modalities

DBB Modality	Nozzle Size	Droplet Diameter	Cell Viability	Bioink Printability			Operating Conditions	Advantages	Disadvantages
DBB Modality	Nozzle Size	Droplet Diameter	Cell Viability	Material	Maximum Concentration (w/v)	Maximum Viscosity (mPa.s)	Operating Conditions	Advantages	Disadvantages
Thermal DOD	50 μm (Xu et al., 2006)	30–60 μm (Xu et al., 2008, 2006)	75–90% (Xu et al., 2013a,b, 2005; Cui et al., 2012a,b; Gao et al., 2014, 2015)	Alginate (NaAlg) (Xu et al., 2009, 2008; Boland et al., 2007)	2.3% (Xu et al., 2009, 2008; Boland et al., 2007)	N/A	10⁵−10⁷ cells/mL (Xu et al., 2013a,b, 2005; Cui et al., 2012a,b; Gao et al., 2014, 2015) exposed to 200–300^°C for few milliseconds (ms) (Roth et al., 2004)	Low cost, ideal for feasibility studies	Limited cell types and clogging issues owing to smaller nozzle diameter, thermal and mechanical stress on cells during droplet ejection, difficult to clean as cartridges are designed for 2D paper printing, limited range of orifice diameters, and not available in single nozzle configuration
				PEG (Gao et al., 2015)	10% (Gao et al., 2015)	1.85 (Gao et al., 2015)
				PEG-GelMA (Gao et al., 2015)	10% PEG and 5% GelMA (Gao et al., 2015)	4 (Gao et al., 2015)
				Thrombin (Cui and Boland, 2009)	200 unit/mL (Cui and Boland, 2009)	N/A
Piezoelectric DOD	21.5–120 μm (Xu et al., 2012; Cheng et al., 2014; Christensen et al., 2015; Choi et al., 2011)	50–100 μm (Herran and Huang, 2012; Xu et al., 2012; Choi et al., 2011; Cheng et al., 2014)	70–95% (Wilson and Boland, 2003; Matsusaki et al., 2013; Xu et al., 2012; Christensen et al., 2015)	Alginate (NaAlg from Sigma–Aldrich) (Herran and Coutris, 2013; Xu et al., 2012)	2% (Herran and Coutris, 2013; Xu et al., 2012)	140 (ƞ₀) (Herran and Coutris, 2013; Xu et al., 2012)	10⁵–10⁷ cells/mL (Wilson and Boland, 2003; Xu et al., 2012; Matsusaki et al., 2013; Christensen et al., 2015)	Control over droplet generation and placement, wide range of nozzle diameters, available in single nozzle configuration, cleanable as long as bioink materials are not dried out	Nozzle clogging, satellite droplets, mechanical stress on cells during droplet ejection, made of glass that is not suitable for certain bioink materials such as fibrinogen
Table Continued

DBB Modality	Nozzle Size	Droplet Diameter	Cell Viability	Bioink Printability			Operating Conditions	Advantages	Disadvantages
DBB Modality	Nozzle Size	Droplet Diameter	Cell Viability	Material	Maximum Concentration (w/v)	Maximum Viscosity (mPa.s)	Operating Conditions	Advantages	Disadvantages
Electrostatic DOD	N/A	10–60 μm (Nishiyama et al., 2008)	70% (Nishiyama et al., 2008)	Alginate (Nishiyama et al., 2008)	1% (Nishiyama et al., 2008)	10 (Nishiyama et al., 2008)	10⁶–10⁷ cells/mL (Nishiyama et al., 2008)	Low cost, ideal for feasibility studies	Limited cell types and clogging issues owing to smaller nozzle diameter, mechanical stress on cells during droplet ejection, difficult to clean as cartridges are designed for 2D paper printing, limited range of orifice diameters, and not available in single nozzle configuration
Table Continued

DBB Modality	Nozzle Size	Droplet Diameter	Cell Viability	Bioink Printability			Operating Conditions	Advantages	Disadvantages
DBB Modality	Nozzle Size	Droplet Diameter	Cell Viability	Material	Maximum Concentration (w/v)	Maximum Viscosity (mPa.s)	Operating Conditions	Advantages	Disadvantages
Electrohydrodynamic jetting	2–1000 μm (Gasperini et al., 2013; Workman et al., 2014; Jayasinghe and Townsend-Nicholson, 2006; Xie and Wang, 2007; Poellmann et al., 2011; Gasperini et al., 2015; Kim et al., 2007)	5–2000 μm (Gasperini et al., 2013; Workman et al., 2014; Jayasinghe and Townsend-Nicholson, 2006; Xie and Wang, 2007; Poellmann et al., 2011; Gasperini et al., 2015; Kim et al., 2007)	>90% (Workman et al., 2014; Xie and Wang, 2007; Gasperini et al., 2015)	Alginate (NaAlg from Sigma–Aldrich) (Workman et al., 2014)	2% (Workman et al., 2014)	>2000 (Workman et al., 2014)	Applied voltage of 0.250–20 kV (Gasperini et al., 2013; Workman et al., 2014; Jayasinghe and Townsend-Nicholson, 2006; Xie and Wang, 2007; Poellmann et al., 2011; Gasperini et al., 2015; Kim et al., 2007), flow rate from 10⁻¹⁰ (0.36 mL/h) to 10⁻⁸ m³/s (36 mL/h) (Gasperini et al., 2013; Workman et al., 2014; Jayasinghe and Townsend-Nicholson, 2006; Xie and Wang, 2007; Poellmann et al., 2011; Gasperini et al., 2015; Kim et al., 2007), 10⁶–10⁷ cells/mL (Gasperini et al., 2013; Workman et al., 2014; Jayasinghe and Townsend-Nicholson, 2006; Xie and Wang, 2007; Poellmann et al., 2011; Gasperini et al., 2015; Kim et al., 2007)	Droplets smaller than nozzle orifice diameter, low mechanical stress on cells during droplets ejection, viscous bioink materials are dispensable	Expensive, unavailability of commercial complete systems, unsafe for operators, not capable of ejecting single droplet at a time
Electrohydrodynamic jetting				Collagen (Kim et al., 2007)	3% in 3% acetic acid (Kim et al., 2007)	N/A
Table Continued

DBB Modality	Nozzle Size	Droplet Diameter	Cell Viability	Bioink Printability			Operating Conditions	Advantages	Disadvantages
DBB Modality	Nozzle Size	Droplet Diameter	Cell Viability	Material	Maximum Concentration (w/v)	Maximum Viscosity (mPa.s)	Operating Conditions	Advantages	Disadvantages
Acoustic bioprinting	n/a	5–300 μm (Demirci and Montesano, 2007b; Fang et al., 2012)	>90% (Demirci and Montesano, 2007b; Fang et al., 2012)	Ethylene glycol (Demirci and Montesano, 2007b)	N/A	18 (Demirci and Montesano, 2007b)	10⁵–10⁷ cells/mL (Demirci and Montesano, 2007b; Fang et al., 2012)	No mechanical stress on cells during droplet ejection, easy to fabricate	Viscous bioinks are not dispensable, unavailability of commercial complete systems
Microvalve bioprinting	100–300 μm (Lee et al., 2009b; Faulkner-Jones et al., 2013; Horváth et al., 2015)	100–600 (Lee et al., 2009b; Faulkner-Jones et al., 2013; Horváth et al., 2015)	>90% (Demirci and Montesano, 2007a; Lee et al., 2009a,b; Moon et al., 2010; Xu et al., 2010, 2011; Ferris et al., 2013; Gurkan et al., 2014; Faulkner-Jones et al., 2015, 2013; Faulkner-Jones et al., 2013)	Collagen type I (Zhao et al., 2012)	0.9% (Zhao et al., 2012)	N/A	10⁵–10⁷ cells/mL (Demirci and Montesano, 2007a; Lee et al., 2009a,b; Moon et al., 2010; Xu et al., 2010, 2011; Ferris et al., 2013; Faulkner-Jones et al., 2013, 2015; Gurkan et al., 2014)	Low cost, viscous bioink materials are dispensable, cleanable as long as bioink materials are not dried out, interchangeable nozzles	Significantly larger droplet diameters than nozzle orifice diameter, greater shear stress on cells during droplet ejection as nozzles are not available in tapered configuration
Table Continued

DBB Modality	Nozzle Size	Droplet Diameter	Cell Viability	Bioink Printability			Operating Conditions	Advantages	Disadvantages
DBB Modality	Nozzle Size	Droplet Diameter	Cell Viability	Material	Maximum Concentration (w/v)	Maximum Viscosity (mPa.s)	Operating Conditions	Advantages	Disadvantages
				Fibrinogen (Lee et al., 2010)	6.2% (Lee et al., 2010)	N/A
				Thrombin (Lee et al., 2010)	133 unit/mL (Lee et al., 2010)	N/A

5.6. Droplet-Substrate Interactions

In addition to jet formation, impingement of a droplet onto the substrate is also crucial during DBB as it affects the fidelity and spreading of the droplet. While the ultimate goal of bioprinting is to pattern the cells and fabricate 3D tissue constructs, the ideal placement of droplets should accommodate the preservation of the droplet integrity as splashing or spreading of the droplet results in displacement of the deposited cells from their desired position or structural failure in 3D bioprinting. For viscoelastic hydrogels used in DBB, there exist two major impingement characteristics including splashing and spreading of the droplet. In splashing, the droplet disintegrates into secondary droplets after colliding with a substrate. In spreading, on the other hand, the droplet spreads over the surface and expands its surface area. According to Worthington, the droplet shape highly depends on the velocity of the droplet, where higher velocity generates splashing and lower velocity leads to spreading (Worthington, 1876). According to Weber, the critical number (We) of droplet splashing is formulated as:

$We = \frac{ρ d U^{2}}{σ}$ $We = \frac{ρ d U^{2}}{σ}$

(5.11)

where d is the characteristics length that corresponds to the diameter of the droplet and U is the velocity. In general, larger Weber numbers generate splashes and lower ones lead to spreading. As these properties are important for the droplet formation (Son et al., 2008; Stringer and Derby, 2009; Rioboo et al., 2002), properties of the substrate are also important as the factors including but not limited to wettability of the surface, surface roughness as well as viscous forces (Roisman et al., 2002). Furthermore, the spreading of the droplet occurs faster than its polymerization in a typical bioprinting setup (Pataky et al., 2012) and the parameters associated with the crosslinking mechanism (i.e., ionic crosslinker solution concentration) influence the spreading behavior of the droplet (Xu et al., 2008). Droplet–substrate interactions broadly consist of two regimes from a fluid mechanics perspective (Saunders and Derby, 2014; Stringer and Derby, 2009). The first is the dynamic regime during which the kinetic energy of the droplet is dissipated. The second is the viscous dissipation regime during which surface energy interactions between the droplet and the substrate determine the spreading of the droplet to an equilibrium shape. In addition, gravity has minimal influence on landing and spreading of the droplet (Schiaffino and Sonin, 1997).

Transitioning from 2D (on a substrate) to 3D bioprinting necessitates a 3D-printing mechanism, which enables fabrication of 3D constructs through deposition of droplets in a layer-by-layer fashion. Depending on the utilized hydrogel and its crosslinking mechanisms, four types of 3D-printing schemes have been utilized in fabrication of tissue constructs including (1) alternating printing of the bioink and the crosslinker solutions (Fig. 5.10A), (2) bioprinting of the bioink solution into a reservoir filled with the crosslinker solution (Fig. 5.10B), (3) bioprinting of the bioink solution followed by spraying the crosslinker solution on top (Fig. 5.10C), and (4) bioprinting of the bioink solution followed by exposing it to a UV source (Fig. 5.10D). As fabricating scale-up tissue constructs necessitate the integration of vascular networks, such vascularization can be generated by printing thermally crosslinking sacrificial materials (i.e., gelatin) and liquefying them thereafter.

Figure 5.10 Droplet–substrate interactions and fabrication of 3D constructs through layer-by-layer deposition. (A) Droplets of a cell-laden hydrogel precursor solution are deposited at specific locations on the lateral plane by controlling the movement of the printhead and/or the substrate. The deposited droplets spread and coalesce to form lines on the substrate. The coalesced lines in turn assemble to form the first layer of the desired 3D pattern, which is subsequently polymerized by depositing the droplets of crosslinking (ionic or enzymatic) solution. This cycle is repeated until the fabrication of the entire construct is completed. (B) Alternatively, each layer of the bioprinted hydrogel precursor solution is polymerized by lowering the substrate into a reservoir of the crosslinker and raising it again before a new layer is bioprinted. (C) Each layer of the bioprinted hydrogel precursor solution is polymerized by spraying the crosslinking solution. (D) Each layer of the bioprinted hydrogel precursor solution (photocurable bioink) is polymerized using ultraviolet (UV) light.

The resolution at which the 3D construct is fabricated depends on various factors including the volume and velocity of ejected droplets, droplet–substrate and droplet–droplet interactions, and the applied crosslinking mechanism. The volume of ejected droplets is mediated by several operating parameters such as the bioink material characteristics, the printhead geometry (orifice diameter), and its actuation voltage pulse characteristics (Herran and Huang, 2012; Lee et al., 2009a,b; Faulkner-Jones et al., 2015; Demirci and Montesano, 2007a; Fang et al., 2012; Workman et al., 2014; Gasperini et al., 2013; Nishiyama et al., 2008). For example, the width of each coalesced line in Fig. 5.10A depends on several fabrication parameters including the volume of droplets, the spacing between droplets, and the printing speed, which consequently control the bioprinting resolution (Xu et al., 2012; Nishiyama et al., 2008; Soltman and Subramanian, 2008). In addition, how droplets interact with the substrate affects the spreading behavior of the droplet, which consequently affects the bioprinting resolution (Son et al., 2008; Schiaffino and Sonin, 1997; Stringer and Derby, 2009).

5.7. Biomaterials Used in Droplet-Based Bioprinting

A bioprintable material comprising various biologics (i.e., cells, growth factors, deoxyribonucleic acid (DNA), or drugs loaded in a delivery medium such as media or hydrogels) that is employed to fabricate 3D constructs with or without the use of external stimulations is rightfully referred to as “bioink,” as introduced earlier in Chapter 3. Essential characteristics of bioink include low viscosity, suitable biodegradability and biocompatibility, enhanced cell adhesive properties, bioprintability, and high mechanical strength. However, such characteristics limit the range of exploitable biomaterials for DBB. Therefore, a limited range of hydrogels are available as bioink in DBB. Alternatively, hydrogels are also used as a substrate material when their viscosity is higher and/or the nozzle orifice diameter is extremely small. Hence, cells and other biologics are directly bioprinted into them.

5.7.1. The Bioink Consideration

A limited range of hydrogels including alginate, collagen, fibrin, methacrylated gelatin (GelMA), and polyethylene glycol (PEG) have been used in DBB due to their own or their crosslinkers' ease of ejectability and the compatibility of their crosslinking mechanism with different DBB modalities. This section discusses the bioink within the context of DBB and provides the reader with examples of bioprinted tissue constructs using the previously mentioned bioink materials. The reader is referred to Chapter 3 for a comprehensive discussion on a myriad of bioink materials used in bioprinting technologies.

Alginate undergoes ionic crosslinking, through the negatively charged carboxylate (COO⁻) group that is present in its polymeric backbone. When the negatively charged COO⁻ group is exposed to positively charged ions such as divalent calcium cations (Ca²⁺), it yields a crosslinked hydrogel network. Exploiting this crosslinking mechanism of alginate with CaCl₂, Atala's group fabricated heterogeneous tissue constructs comprising AFSCs, dSMCs, and bECs (Xu et al., 2013a). Alginate was also combined with several other materials to enhance the mechanical and functional properties of the fabricated constructs. Blaeser's group, for instance, fabricated a bifurcated vascular tissue construct using a composite blend of 3% (w/v) low-melting agarose and 3% (w/v) low-viscosity alginate (Blaeser et al., 2013). The bioprinting process was performed under nontoxic fluorocarbon, which provided the necessary buoyancy forces needed to support the soft tissue structure. Additionally, Boland and his coworkers bioprinted the crosslinking solution (CaCl₂) using a modified HP DeskJet inkjet printer to fabricate a heart-like tissue construct with connected ventricles (Xu et al., 2009). In another study, Huang's group mixed 3T3 fibroblast cells in 1% sodium alginate solution and bioprinted it into the crosslinker pool to fabricate zigzag cellular constructs (Xu et al., 2012). Similarly, alginate comprising hepatocyte-like cells, which were obtained through directed differentiation of hiPSCs, was bioprinted to fabricate a 3D mini–liver tissue model (Faulkner-Jones et al., 2015).

Collagen type I has been extensively exploited in tissue engineering for fabrication of tissue constructs due to its biocompatibility and cell adhesive properties (Roth et al., 2004); however, it has a limited use in DBB due to its fibrous nature as fibers can likely to clog the nozzle. In one study Boland's group used it as a bioink constituent to investigate adhesion and proliferation of cells on collagen-coated cell repellant substrates (Roth et al., 2004). Also, Boland's group fabricated a bilayer skin graft, which generated neoskin identical to native skin with microvessels (Yanez et al., 2014). In another study, fibrin–collagen bioink comprising one of the two cell types, AFSCs or MSCs, was bioprinted into wound sites for treating skin burns (Skardal et al., 2012).

Methacrylated gelatin forms a biomimetic (Nichol et al., 2010) as well as an enzymatically degradable (Hutson et al., 2011) hydrogel that is mechanically strong when photo-crosslinked with UV in presence of a photoinitiator providing a suitable material for DBB. For example, Demirci's group incorporated GelMA and growth factors (BMP-2 and TGF-β1) as bioink constituents to imitate native fibrocartilage microenvironments that differentiated bioprinted hMSCs toward osteogenic and chondrogenic lineages spatially (Gurkan et al., 2014).

Fibrin, a hydrogel formed by the reaction of thrombin with fibrinogen, supports extensive cell growth and proliferation (Cui and Boland, 2009). Boland's group used fibrin to engineer microcapillaries by bioprinting HMVECs-laden thrombin and Ca²⁺ solution on a fibrinogen substrate (Cui and Boland, 2009). Employing TIJ bioprinting, HMVECs were precisely bioprinted on crosslinked fibrin. Bioprinted HMVECs aligned well in fibrin and formed into an extensive capillary network after 21 days of culture. In another study, the same group bioprinted alternating layers of neural cells and fibrin gel to fabricate viable neural constructs for potential neural engineering applications (Xu et al., 2006). In another study, Atala's group used fibrin to engineer cartilage tissue with enhanced mechanical and functional properties employing a hybrid method involving electrospinning and microvalve bioprinting (Xu et al., 2013b). In general, it is more convenient to bioprint thrombin instead of fibrinogen due to the fibrous nature of fibrinogen leading to clogging issues.

Polyethylene glycol has greater mechanical properties compared to naturally derived polymers, such as alginate, fibrin, agarose, and collagen type I, making it an appealing bioink material for DBB. Altering the composition of PEG-based hydrogels in tandem with photo-crosslinking them in presence of a photoinitiator enables the control of structural, functional, and mechanical properties of fabricated tissues. Using TIJ bioprinting, Cui et al. 2012a,b bioprinted human articular chondrocytes in PEGDMA in a layer-by-layer fashion concurrent UV crosslinking of each layer, which yielded 3D construct with homogenous distribution of cells and supported neocartilage formation. In a similar study, acrylated PEG was bioprinted with acrylated peptide containing the Arg-Gly-Asp (RGD) sequence (essential for cell adhesion), followed by photopolymerization of the bioprinted layer (Gao et al., 2015). Bone marrow–derived hMSCs were suspended in PEGDMA in conjunction with bioactive glass and hydroxyapatite nanoparticles, and bioprinted using a TIJ bioprinter (Gao et al., 2014). This approach allowed control over the spatial delivery of hMSCs and bioactive ceramic materials in the fabricated bone tissue constructs.

There are a limited number of hydrogels that can be employed in DBB owing to very fine nozzle diameters. Therefore, it is essential to use bioink materials with low viscosity to obviate nozzle clogging issues. Bioprintability, cost, crosslinking mechanisms, and viscosity are some of the factors that are essential to consider while selecting a bioink for DBB.

5.7.2. The Substrate Consideration

In addition to bioprinting hydrogels, a myriad of endeavors has been made to bioprint macromolecules (i.e., growth factors, proteins, or even cells in media) directly onto hydrogel substrates. By controlling the spatial distribution of growth factors on hydrogel substrate, differentiation of stem cell into specific lineages has been widely attempted. Phillippi et al. (2008) used cyanin-3–labeled BMP-2 in media as a bioink solution and bioprinted it on fibrin-coated glass slides using a piezoelectric DOD bioprinter (Phillippi et al., 2008). They spatially immobilized BMP-2 with a varying concentration according to a predesigned pattern and cultured primary muscle-derived stem cells (MDSCs) on the BMP-2 patterned surface. This led to the differentiation of MDSCs into multiple lineages, namely osteogenic and myogenic in that study. Another study conducted by Ilkhanizadeh et al. (2007) showed that the fate and differentiation ability of neural stem cells could be effectively controlled by inkjet bioprinting of various macromolecules, such as fibroblast growth factor-2, ciliary neurotrophic factor, and fetal bovine serum, on neural stem cell–seeded polyacrylamide-based hydrogel substrates.

Demirci's group fabricated co-culture cancer models by bioprinting human ovarian cancer cells and fibroblasts in a controlled manner on Matrigel-coated glass culture dish (Xu et al., 2011). Collagen type I and fibrinogen has also been used as gel substrates for spatially controlled bioprinting of HMVECs mixed in thrombin for fabricating skin grafts (Yanez et al., 2014). This bilayered skin graft comprising of keratinocytes and fibroblasts in collagen (as the top layer and bottom layer, respectively), and HMVECs encapsulated in fibrin network as the middle layer resulted in a portable construct, which was eventually placed on full thickness wounds in mice model for in vivo studies. Polyacrylamide gel substrates have also been employed for micropatterning proteins and cells by peeling off the protein from functionalized glass slides coated with PA substrates (Tang et al., 2012). Poly(allylamine hydrochloride)/poly(styrene sulfonate) (PAH/PSS) based films were patterned on alginate gels for vascular tissue engineering by peeling off from the substrate as well (Kerdjoudj et al., 2011). These studies reveal the feasibility of bioprinting cells on gel substrates, and depending on the application, these substrates can be engineered to be peeled off for further studies.

5.8. Comparison of Droplet-Based Bioprinting With Other Bioprinting Techniques

DBB has several advantages and disadvantages with respect to other bioprinting techniques, including EBB (mechanical (Skardal et al., 2010; Jakab et al., 2008; Gaetani et al., 2012; Hockaday et al., 2012; Owens et al., 2013), pneumatic (Khalil et al., 2005; Fedorovich et al., 2008; Ozbolat et al., 2014; Marchioli et al., 2015; Yu et al., 2014), or valve-based (Snyder et al., 2011; Markstedt et al., 2015; Chang et al., 2008, 2010)) or LBB (stereolithography (Salonitis, 2014) and its modifications (Lin et al., 2013), laser-guidance direct writing (Odde and Renn 1999, 2000), and laser-induced forward transfer (Mézel et al., 2010; Xiong et al., 2015; Barron et al., 2004)).

DBB technology is a multifaceted technology. Highly complex heterocellular tissue constructs with different compositions of biologics (i.e., biomaterials, cells, growth factors, drugs, and genes) can be easily patterned when compared to EBB and LBB techniques as it is highly challenging to incorporate multiple types of biologics in LBB and generating heterogeneity in a delicate manner is difficult using EBB. While DBB has a process resolution higher than that of EBB and possesses a greater versatility in incorporating multiple biologics, DBB has attracted several researchers in the bioprinting community as well as researchers adopting bioprinting technology in other field of studies such as regenerative medicine and pharmaceutics (Horváth et al., 2015; Rodríguez-Dévora et al., 2012). Moreover, it has a reasonable resolution comparable to LBB, which allows better control on the geometry and size of bioprinted constructs by mediating the gelation process precisely as crosslinker and precursor hydrogel solutions can be selectively deposited. This enables mighty control on swelling and shrinkage properties of the bioprinted constructs.

Droplet-based bioprinters are highly versatile and affordable, where a simple HP printer can be easily modified and used as a bioprinter (Cui et al., 2014). A wide range of droplet-based bioprinters are commercially available within affordable price band (Mironov et al., 2009). If the reproducibility and flexibility becomes a concern, extra capabilities can be easily implemented with a reasonable additional cost. For example, printheads in DBB may not be suitable for certain bioink materials such as fibrous bioink (i.e., fibrinogen and collagen) and generate inconsistent results due to nozzle clogging or accumulation of cell debris or fibers anywhere in the line from reservoir to the nozzle in the tubing system. A commercial droplet-based bioprinter can be modified to overcome such issues by replacing the original dispenser with dispensers having larger nozzles or dispensers with different droplet generation mechanisms as the rest of bioprinter subcomponents are common for all DBB modalities.

DBB technology is also user-friendly and easy to implement. It can be readily used by operators, who have limited exposure to the technology while generated computer-aided design (CAD) models can be easily transferred to print out by simply pressing “bioprint.” In other words, it has a non-steep learning curve circumventing the need for extensive experimentation, which is a major impediment in EBB as the operator needs to understand the shear-thinning behavior of hydrogels as well as their bioprintability on the printing stage.

The other advantage of DBB is that it facilitates rapid bioprinting through an array of nozzles in a highly reproducible manner. This capability enables rapid fabrication of an array of samples, which is highly desirable in high-throughput screening applications such as drug testing and cancer screening (Fang et al., 2012; Xu et al., 2011; Yusof et al., 2011). Producing high-throughput arrays using EBB or LBB, on the other hand, is highly challenging and not practical. In addition to its appealing features, DBB has a great translational potential in clinical use for tissue bioprinting. It is highly convenient for in situ bioprinting purposes as defects (i.e., cranio- or maxilla-facial defects, skin burns or deep wounds) on human body can be easily reconstructed using DBB as DBB operates in a noncontact manner. The defects can be easily filled by jetting droplets from a distance into the defect. This feature also enables bioprinting of growth factors or other biologics on existing tissue constructs as biologics can be selectively sprayed over the tissue constructs (Cooper et al., 2010). The noncontact nature of DBB processes alleviates other major issues observed in EBB such as collision between the printhead and the bioprinted constructs, or unexpected increase in clearance between the orifice and the receiving substrate.

5.9. Recent Achievements in Droplet-Based Bioprinting

Recent achievements in DBB are in the areas of stem cell research, organs-on-chip models and regenerative medicine including tissue regeneration using in situ bioprinting. DBB does not affect stem cells functionality and differentiation capacity (Faulkner-Jones et al., 2015; Gurkan et al., 2014; Xu et al., 2013a; Gao et al., 2015). For example, Shu's group engineered 3D liver tissue models with hepatocytes derived from hiPSCs and hESCs (Faulkner-Jones et al., 2015). The group bioprinted a bioink solution (comprising hepatocytes and sodium alginate) using a microvalve bioprinter to fabricate 3D tissue model (see Fig. 5.11A1 and A2). During the bioprinting process, alternating layers of the bioink and crosslinker solutions [calcium chloride (CaCl₂)] were bioprinted to enable the crosslinking of alginate. After 17 days postbioprinting (after 23 days postdifferentiation), bioprinted cells maintained their differentiated phenotype, which was confirmed through the presence of hepatic markers such as hepatocyte nuclear factor 4 alpha (HNF4α) and albumin. In another study, Atala's group demonstrated that bioprinted stem cells retained their functionality and the differentiation capacity both in vitro and in vivo (Xu et al., 2013a). The group bioprinted three different cell types, hAFSCs, dSMCs, and bECs, along with CaCl₂ solution using a TIJ bioprinter. Consecutive layers of cell-laden CaCl₂ crosslinker solution were bioprinted into a sodium alginate-collagen solution to fabricate 3D pie-shaped heterogeneous tissue constructs. The pie-shaped constructs were comprised of three distinct sections each with a particular cell type as shown in Fig. 5.11B1. Similarly, 3D cuboidal homogenous tissue constructs of each cell type were also fabricated. Later, bioprinted constructs were cultured for 1 week and subcutaneously implanted into outbred athymic nude mice. Afterward, the constructs were surgically retrieved either after 4 or 8 weeks (see Fig. 5.11B2). Subsequent analysis showed that biological functions (i.e., viability, proliferation, phenotypic expression, and physiological properties) of the bioprinted cells of each type were not affected significantly both in vitro and in vivo. Another notable observation was the vascularization of implanted bEC constructs with substantial blood vessels compared to that of control groups.

Figure 5.11 Recent achievements in droplet-based bioprinting. (A1) Three-dimensional (3D) liver tissue model (top view) comprising of 40 layers of bioprinted alginate and HLCs acquired through differentiation of hiPSCs and hESCs and the side view (A2) (Reproduced/adapted with permission from Faulkner-Jones et al. (2015)); (B1) 3D heterogeneous tissue model consisting of bioprinted dSMCs (red) labeled with PKH 67 dye, hAFSCs (blue) labeled with CMHC dye, and bECs (green) labeled with PKH 26 dye that retained its functionality and the differentiation capacity both in vitro and in vivo, (B2) vascularization of the bECs constructs 8 weeks after implantation (Reproduced/adapted with permission from Xu et al. (2013a)); Bioprinted 3D cartilage tissue transplants (C1) maintained their biological functions both in vitro and in vivo, (C2) cartilage tissue construct fabrication by layer-by-layer deposition of chondrocytes-fibrinogen-collagen into a previously electrospun layer of PCL fibers (Reproduced/adapted with permission from Xu et al. (2013a)).

Achievements in regenerative medicine include tissue transplants (Xu et al., 2013b; Yanez et al., 2014) and in situ bioprinting (Cui et al., 2012a) for improved wound healing. In a recent study (Xu et al., 2013b), Atala's group engineered hybrid cartilage tissue by employing microvalve bioprinting and electrospinning. In that study, a bioink solution comprising chondrocytes, fibrinogen, and collagen was bioprinted into previously electrospun layers of poly-ε-caprolactone (PCL) fibers as illustrated in Fig. 5.11C1 and C2. In addition, thrombin was bioprinted on each bioprinted layer of the bioink solution to facilitate crosslinking. Afterward, the constructs were cultured in vitro to evaluate cell proliferation and organization. Additionally, some constructs were cultured in vitro for 2 weeks and implanted subcutaneously in immunodeficient mice. Subsequently, the constructs were surgically retrieved after 2, 4, and 8 weeks for characterization. The characterization study indicated that cartilage constructs maintained their biological functions both in vitro and in vivo. At the same time, the cartilage constructs possessed enhanced biological and mechanical characteristics than the cartilage constructs fabricated without incorporating the electrospun PCL fibers. Further, the constructs supported formation of a new cartilage-like tissue.

In situ bioprinting is an alternative approach to two-step bioprinting (bioprinting followed by implantation), where cells and other biologics are directly bioprinted into lesion sites (Skardal et al., 2012; Cui et al., 2012a). In one study (Cui et al., 2012a), Lima's group engineered a cartilage tissue with comparable characteristics of the native cartilage. Using a TIJ bioprinter, human articular chondrocytes were bioprinted within photopolymerizable PEGDMA into 2–5 mm deep defects in osteochondral explants. The explants were previously harvested from bovine femoral condyles using an 8-mm-diameter stainless steel punch. The defects in explants were repaired by bioprinting layers of chondrocytes and PEGDMA while crosslinking each bioprinted layer through photopolymerization. Afterward, the repaired explants were cultured in vitro for 2, 4, and 6 weeks postbioprinting. After 4 weeks, more chondrocytes with higher glycosaminoglycan (GAG) and collagen type II production were observed in tissue constructs bioprinted into the osteochondral explants than the constructs bioprinted in vitro (control group). Thus in situ bioprinting leveraged the presence of native cartilage tissue to accelerate chondrogenesis and extracellular matrix (ECM) production. Similarly, Skardal et al. (2012) employed in situ bioprinting to regenerate skin tissue with AFSCs and MSCs. To compare their healing properties, the two cell types were separately bioprinted over surgical skin wounds (2.0 × 2.0 cm) on the back (middorsal region) of nude mice using a microvalve bioprinter. Overall, three layers of thrombin solution and two layers of cell-laden collagen-fibrinogen solution were alternatively bioprinted over the wounds resulting in 5 × 10⁶ AFSCs or SMCs in each wound. Photographic images were taken immediately as well as 7 and 14 days postbioprinting to monitor wound healing. Furthermore, regenerated skin was harvested from animals at 7 and 14 days for histological analysis. Results indicated that AFSCs were comparable to MSCs in skin regeneration and in situ bioprinting of cells overall accelerated the wound healing process.

5.10. Limitations

Despite the great advantages of DBB, the technology possesses considerable limitations and drawbacks. One of the major limitations of DBB is the clogging of orifice during bioprinting process as highly small fragments in the bioink can accumulate within the orifice and obstruct the flow. This can sometimes necessitate the replacement of the entire orifice if the clogged material does not dissolve or cannot be removed. Because of the small orifice diameter ranging from 10 to 150 μm, as discussed earlier, a limited number of biomaterials are available for DBB such as low-viscosity hydrogels or their components. Thus a wide majority of the bioink materials used in EBB, including cell aggregates, microcarriers, and highly viscous hydrogels (Ozbolat and Hospodiuk, 2016), cannot be used in DBB. Due to this issue, researchers have preferred to create substrates of hydrogels and bioprint cells or other biologics on them using cell media as a delivery medium (Gurkan et al., 2014). This approach has been widely employed for bioprinting arrays of droplets for high-throughput screening (Suntivich et al., 2014). Although the noncontact nature of DBB provides great advantages as discussed earlier, gelation characteristics of printed droplets should be well experimented as ejected droplets can quickly gel in the air and do not assemble to the bioprinted substrate easily.

The other limitation of DBB is the inability to fabricate mechanically strong and structurally well-integrated constructs due to the limited range of available bioink materials, particularly in high concentrations (Dababneh and Ozbolat, 2014). This can be however alleviated to some extent by infiltrating the bioprinted constructs within another biomaterial as a postprocess. Alternatively, a reinforcement approach can also be employed, where nanofibers of stronger polymers can be reinforced into the inkjet bioprinted hydrogels using a hybrid fabrication setup (Xu et al., 2013b). While DBB facilitates fabrication of constructs in a discontinuous manner, where discrete droplets are assembled in 3D, it is highly challenging to fabricate porous tissue constructs. Porous tissue constructs are favorable for perfusion purposes to facilitate sufficient media exchange and can be easily bioprinted using EBB or LBB (only stereolithography and its modifications) techniques. Using DBB, porous architecture can be created either using a plotting medium as a support material with the help of buoyancy or utilizing a two-step approach such as applying porogens or fugitive hydrogels. Lastly, as the resolution of DBB is higher than that of EBB, it takes a longer time to fabricate scalable tissue constructs, where some minor issues can be experienced such as change in the size of the construct due to swelling, contraction, or dehydration (when not bioprinted into the medium).

5.11. Future Directions

Different bioprinting modalities, including DBB, are envisioned to fabricate functional replacement human organs in the future (Ozbolat and Yu, 2013); however, several challenges have yet to be overcome to make it a reality. The first challenge is the printhead design for DBB. The physical characteristics of currently available printheads limit the control over several parameters including droplet volume, the number of cells to be encapsulated in each droplet, the precise placement of droplets, cell concentration, bioink material properties (viscosity), and the long-term reliability of the entire system. Printhead design constraints arise because of the current microfabrication processes, which impose several restrictions including the nozzle geometry. Hence, new nano- or microfabrication techniques are required to overcome the physical limitations with novel nozzle and printhead designs.

The second challenge is associated with materials that constitute the bioink. Each human organ is comprised of several billions of cells of various types (Bianconi et al., 2013). Hence, acquiring cells such as stem cells in such quantities for autologous transplantation applications is constrained by cell cycle times, which may take several weeks to months (Cooper, 2000; Bruce et al., 2002). Hence, new strategies for accelerating the cell cycle time are required. Another bioink-related challenge is the availability of biomimetic materials with controlled degradability and signaling cues to stimulate the proliferation and differentiation of cells (Murphy and Atala, 2014). For example, matrix material properties, such as elasticity, impact the differentiation of bioprinted cells into specific phenotypes (Engler et al., 2006). Thus development of new materials or mechanisms that instill the bioink material with specific biomimetic characteristics, particularly after bioprinting, is desired as the nozzle geometry imposes material constraints. Growth factors and other signaling molecules could partially reduce the necessity of biomimetic materials; however, transporting them to specific bioprinted cells in a sustained manner over time imposes its own set of challenges. Perhaps, the transport of targeted growth factors and signaling molecules is possible by controlling the micropore geometry of bioprinted hydrogels such that of molecules of particular conformation (shape) are selectively transported.

The third challenge is the fabrication of 3D tissue constructs of complex conformations at submicrometer to micrometer resolution. For example, fabrication of complete vascular network at the single–cell level is challenging because DBB at high resolution is limited by the nozzle geometry, which constraints the droplet volume (size) and the bioink viscosity. Moreover, available bioprintable materials or hydrogels are characterized by weak mechanical strength and hence simultaneous codeposition of degradable and biocompatible support materials is essential for counteracting gravity. Although, strategies such as support by means of liquid buoyancy have been proposed (Christensen et al., 2015), leveraging natural mechanisms of cells is potentially most effective of all as it addresses several challenges at once. For instance, angiogenesis of microcapillaries (Lee et al., 2014) obviates the codeposition of support materials and the development of novel printheads with extremely small orifice diameter.

Currently, DBB is not capable of fabricating functional replacement human organs at clinically relevant dimensions; however, it can improve drug discovery and disease modeling as it enables fabrication of spatially patterned multicellular microenvironments in a high-throughput and reproducible manner. Further, only a tiny tissue model can be sufficient for drug screening. At present, many drugs are not effective. For example, 97% of the patients see no benefits from antihypertensives (McCormack, 2014) given for high blood pressure whereas 98% of the patients see no benefits from statins given for high cholesterol (Newman, 2015). The low efficacy of prescription drugs can be attributed to the low numbers of human test subjects (Friedman et al., 2010), which may not account for the genetic diversity among millions of patients. Bioprinted organ-on-a-chip models based on hiPSCs derived from diverse groups can account for the genetic variations and improve the drug discovery. Further, genomic analysis tools and individual genetic tests are becoming inexpensive and they can be used to personalize treatment plans or drug doses. Moreover, 3D tissue models are better at mimicking human physiology and pathology than currently used 2D cell culture models (Wüst et al., 2011; Billiet et al., 2012) as well as the animal models (Shanks et al., 2009). Hence, DBB in combination with targeted genome editing tools such as CRISPR (clustered regularly interspaced short palindromic repeats)/CAS9 (Barrangou and Marraffini, 2014; Hsu et al., 2014) can improve disease modeling (Hinson et al., 2015; Liu et al., 2016).

5.12. Summary

DBB offers great advantages due to its simplicity, agility, and versatility with great control on the deposition pattern. Although the technology currently enables fabrication of heterocellular tissue constructs in a high-throughput and reproducible manner, and has been widely used in several application areas such as tissue engineering and regenerative medicine, transplantation, drug testing and high-throughput screening, and cancer research, the technology currently faces several limitations such as weak structural and mechanical properties of bioprinted tissue and organ constructs as well as their lack of vascularization and perfusability, and the limited translation of the technology into clinics. Despite these limitations, novel breakthroughs such as angiogenesis and in situ bioprinting, which leverage nature-driven mechanisms, make the eventual clinical translation of DBB technology inevitable.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.

Table of Contents for 5. Droplet-Based Bioprinting

Create new playlist

Sign In

Sign Up