Porous tissue strands: Avascular building blocks for scalable tissue fabrication

Abstract

The scalability of cell aggregates such as spheroids, strands, and rings has been restricted by diffusion of nutrient and oxygen into their core. In this study, we introduce a novel concept in generating tissue building blocks with micropores, which represents an alternative solution for vascularization. Sodium alginate porogens were mixed with human adipose-derived stem cells, and loaded into tubular alginate capsules, followed by de-crosslinking of the capsules. The resultant cellular structure exhibited a porous morphology and formed cell aggregates in the form of strands, called 'porous tissue strands (pTSs).' Three-dimensional reconstructions show that pTSs were able to maintain ∼25% porosity with a high pore interconnectivity (∼85%) for 3 weeks. Owing to the porous structure, pTSs showed up-regulated cell viability and proliferation rate as compared to solid counterparts throughout the culture period. pTSs also demonstrated self-assembly capability through tissue fusion yielding larger-scale patches. In this paper, chondrogenesis and osteogenesis of pTSs were also demonstrated, where the porous microstructure up-regulated both chondrogenic and osteogenic functionalities indicated by cartilage- and bone-specific immunostaining, quantitative biochemical assessment and gene expression. These findings indicated the functionality of pTSs, which possessed controllable porosity and self-assembly capability, and had great potential to be utilized as tissue building blocks in distinct applications such as cartilage and bone regeneration.

Full text

Biofabrication
ACCEPTED MANUSCRIPT
Porous tissue strands: Avascular building blocks for scalable tissue
fabrication
To cite this article before publication: Yang Wu et al 2018 Biofabrication in press https://doi.org/10.1088/1758-5090/aaec22
Manuscript version: Accepted Manuscript
Accepted Manuscript is “the version of the article accepted for publication including all changes made as a result of the peer review process,
and which may also include the addition to the article by IOP Publishing of a header, an article ID, a cover sheet and/or an ‘Accepted
Manuscript’ watermark, but excluding any other editing, typesetting or other changes made by IOP Publishing and/or its licensors”
This Accepted Manuscript is © 2018 IOP Publishing Ltd.
During the embargo period (the 12 month period from the publication of the Version of Record of this article), the Accepted Manuscript is fully
protected by copyright and cannot be reused or reposted elsewhere.
As the Version of Record of this article is going to be / has been published on a subscription basis, this Accepted Manuscript is available for reuse
under a CC BY-NC-ND 3.0 licence after the 12 month embargo period.
After the embargo period, everyone is permitted to use copy and redistribute this article for non-commercial purposes only, provided that they
adhere to all the terms of the licence https://creativecommons.org/licences/by-nc-nd/3.0
Although reasonable endeavours have been taken to obtain all necessary permissions from third parties to include their copyrighted content
within this article, their full citation and copyright line may not be present in this Accepted Manuscript version. Before using any content from this
article, please refer to the Version of Record on IOPscience once published for full citation and copyright details, as permissions will likely be
required. All third party content is fully copyright protected, unless specifically stated otherwise in the figure caption in the Version of Record.
View the article online for updates and enhancements.
This content was downloaded from IP address 150.231.246.1 on 29/10/2018 at 23:39

1
Porous Tissue Strands: Avascular Building Blocks
for Scalable Tissue Fabrication
Yang Wu1,2, Monika Hospodiuk1,3, Weijie Peng4, Hemanth Gudapati1,2, Thomas Neuberger5,6, Srinivas Koduru7,
Dino J. Ravnic7, Ibrahim T. Ozbolat1,2,6,8
1 Engineering Science and Mechanics Department, The Pennsylvania State University, State College, PA, USA
2 The Huck Institutes of the Life Sciences, The Pennsylvania State University, State College, PA, USA
3 Department of Agriculture and Biological Engineering, Penn State University, State College, PA, USA
4 Department of Pharmacology, Nanchang University, Nanchang, Jiangxi, China
5 High Field MRI Facility, Huck Institutes of the Life Sciences, Pennsylvania State University, University Park, PA, USA,
6 Department of Biomedical Engineering, Penn State University, University Park, PA, USA
7 Department of Surgery, Penn State Health Milton S. Hershey Medical Center, Hershey, PA, USA
8 Materials Research Institute, Penn State University, University Park, PA, USA
Corresponding author: Ibrahim Tarik Ozbolat, PhD
E-mail: ito1@psu.edu
Address: W313 Millennium Science Complex,
Penn State University,
University Park, PA 16802
Tel: +1(814)863-5819
Fax: +1(814)863-4358
Page 1 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

2
Abstract
The scalability of cell aggregates such as spheroids, strands, and rings has been restricted by diffusion of
nutrient and oxygen into their core. In this study, we introduce a novel concept in generating tissue
building blocks with micropores, which represents an alternative solution for vascularization. Sodium
alginate porogens were mixed with human adipose-derived stem cells (ADSCs), and loaded into tubular
alginate capsules, followed by de-crosslinking of the capsules. The resultant cellular structure exhibited
a porous morphology and formed cell aggregates in the form of strands, called ‘porous tissue strands
(pTSs).’ Three-dimensional (3D) reconstructions show that pTSs were able to maintain ~25% porosity
with a high pore interconnectivity (~85%) for three weeks. Owing to the porous structure, pTSs showed
up-regulated cell viability and proliferation rate as compared to solid counterparts throughout the culture
period. pTSs also demonstrated self-assembly capability through tissue fusion yielding larger-scale
patches. In this paper, chondrogenesis and osteogenesis of pTSs were also demonstrated, where the
porous microstructure up-regulated both chondrogenic and osteogenic functionalities indicated by
cartilage- and bone-specific immunostaining, quantitative biochemical assessment and gene expression.
These findings indicated the functionality of pTSs, which possessed controllable porosity and self-
assembly capability, and had great potential to be utilized as tissue building blocks in distinct applications
such as cartilage and bone regeneration.
Keywords: tissue strands, porosity, cell aggregates, micro-fabrication, adipose-derived stem cells
Page 2 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

3
1. Introduction
The conventional scaffold-based tissue engineering (TE) involves seeding cells onto a scaffold, which
has several shortcomings such as low cell density, limited cell-cell interactions, and biomaterial
degradation [1, 2]. These drawbacks have induced the development of new bottom-up approaches, in
which cells are able to self-assemble into tissue constructs [3, 4]. As compared to monolayer cultures,
cells that self-assemble into cell aggregates, imitate the fundamental mechanism in the origin of life,
achieve elevated gene expression, and maintain their phenotypes [5]. The emerging field of
biofabrication strives to get large tissue constructs using cell aggregates as building blocks to fabricate
tissues and organs in vitro [6]. The outermost region of cell aggregates, which exposed to cell culture
media in vitro or surrounding tissue in vivo, always has rich supply of nutrient, oxygen and other
metabolites. However, cellular viability decreases inside the core due to the constrained diffusion and
lack of vascularization, leading to the hypoxic core [7]. Hence, hypoxia usually occurs in cell aggregates
with diameter over 500 µm, which is detrimental for survival of highly metabolic cells [8]. Micro-
vascularized tissue constructs would significantly improve clinical outcome in terms of innervations and
complete functional recovery [9, 10]. In highly vascularized tissues, such as bone, liver and kidney,
formation of new blood vessels becomes essential for a tissue to grow beyond the diffusion limit (> 200
µm) [11, 12]. Vascularization is also strived to be improved for blood circulation, in order to encourage
vessel in-growth into implanted constructs from the host, or to generate new circulation system through
stem cell therapy [13, 14].
Prior to vascularization in free-standing cell aggregates, involvement of porous internal microstructures
can be the first step for building larger tissue constructs [15, 16], which provides space for vessel
ingrowth in the next phase. In scaffold-based tissue engineering, researchers have strived to incorporate
pores into tissue engineered scaffolds with various approaches, such as freeze-drying [17], gas foaming
[18-20], solvent casting and particulate leaching [21-23], phase separation [24], photolithography [25-
Page 3 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

4
27] and three-dimensional (3D) printing technologies [28-30]. In the case of cell-laden hydrogels,
techniques for pore incorporation has been extensively explored in the recent years [31-33]. For example,
it has been reported that a cell-laden hydrogel was prepared by combining human osteoblast-like cells,
alginate suspended with collagen microcarriers, and collagenase to create porous hydrogel constructs
[32]. The cell-laden collagen microcarriers provided a cell-friendly environment, while the collagenase
promoted the disintegration of collagen, leading to pore generation for cell spreading and migration.
As compared to the abovementioned approaches, cell aggregate-based (or scaffold-free) constructs could
result in higher cell density and more space for extracellular matrix (ECM) deposition, facilitate better
cell-cell interactions, generate tissues with close biomimicry, and eliminate biodegradation related issues
[8, 34]. In our previous study, we demonstrated scaffold-free tissue strands as building blocks for scale-
up tissue biofabrication [6]. This study aims to biofabricate porous cell aggregates in form of strands,
henceforth referred as ‘porous tissue strands (pTS),’ which is the first time to demonstrate porous
architecture in scaffold-free constructs. The schematic presented in figure 1 elucidates the fabrication of
scaffold-free pTSs. Harvested using human adipose-derived stem cells (ADSCs), mixed with
micrometer-sized alginate beads as porogens for pore formation, were microinjected into coaxially
extruded tubular semi-permeable capsules. Microfabricated pTSs revealed up-regulated cell viability,
and facilitated rapid fusion and maturation through self-assembly, and also better supported
differentiation of ADSCs towards chondrogenic and osteogenic lineages compared to solid tissue strands
(sTSs).
Page 4 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

5
Figure 1: Steps taken in fabrication of scaffold-free pTSs: First, tubular alginate capsules were extruded
using a coaxial nozzle. Then, alginate porogens which were inkjet printed, and ADSCs that were
harvested in large numbers to obtain highly dense cell pellets, were mixed to obtain homogeneous
cell/porogen mixture. Cell/porogen mixture was then microinjected into alginate capsules to enable the
aggregation of cells, followed by de-crosslinking of the alginate capsule and porogens to obtain pTSs.
To investigate their functionality, pTSs were placed close to each other for self-assembly and also
differentiated towards chondrogenic and osteogenic lineages.
2. Materials and methods
2.1 Fabrication of tubular alginate capsules
Tubular alginate capsules were fabricated using a coaxial nozzle apparatus as described in our previous
studies [6, 35, 36]. Briefly, sodium alginate and calcium chloride (CaCl2) powder (Sigma Aldrich, MO)
Page 5 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

6
was sterilized with ultraviolet (UV) light for 30 min. Sterilized sodium alginate powder was dissolved in
sterile deionized (DI) water and subjected to magnetic stirring overnight to obtain 4% (w/v) alginate
solution. 4% CaCl2 solution in sterile DI water was used as a crosslinker during and after the coaxial
extrusion process. For capsule preparation, a coaxial nozzle unit was connected to a pneumatic air
dispenser and a syringe pump for alginate and CaCl2 extrusion, respectively. The coaxial nozzle
comprised a 22G inner nozzle for CaCl2 deposition and a 14G outer nozzle for alginate extrusion. The
dispensing pressure of alginate was 82.7 kPa, while the CaCl2 dispensing rate was 16 ml/min
[6]. Crosslinked alginate capsules were collected into a CaCl2 pool, and left in the pool overnight for
further crosslinking. The morphology of alginate capsules was observed using an EVOS FL Auto
inverted microscope (ThermoFisher, Pittsburgh, PA), and the inner and outer diameters, and wall
thickness were measured using ImageJ software (National Institutes of Health, MD).
2.2 Fabrication of alginate porogens
Piezoelectric drop-on-demand inkjet (PIJ) printer (MicroFab Technologies Inc., TX) was used for
fabrication of porogens, which comprised a XYZ motion stage, bioink reservoir, PIJ dispenser within a
laminar flow cabinet (Air Science, FL), and a control unit (figures 2a-b). 0.5% sodium alginate solution
was prepared as explained in Section 2.1, and transferred from the reservoir to the dispenser by using a
positive back pressure. Immediately afterward, the back pressure was changed to negative (-1.87 kPa),
such that the resulting meniscus of the alginate solution at the dispenser nozzle orifice formed stable
droplets. The porogens were fabricated by dispensing sodium alginate droplets into a crosslinker pool
(4% CaCl2) using the PIJ dispenser with an orifice diameter of 120 µm (figure 2c, MicroFab
Technologies). The diameter of porogens was controlled by the voltage pulse characteristics, sodium
alginate and CaCl2 concentrations, and the nozzle orifice diameter. The voltage pulse was applied at a
dwell time of 30 µs duration, an echo time of 30 µs duration and an idle time of 20 µs duration. In
addition, the rise and fall time was set at 2 µs for the voltage pulse, and an excitation frequency of 100
Page 6 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

7
Hz was used. The printed porogens were left in the crosslinker pool overnight for complete crosslinking.
The morphology of alginate porogens was observed using the EVOS microscope, and the diameters were
measured using ImageJ software.
Figure 2: Fabrication of sodium alginate porogens. (a) Set-up of the PIJ printer in a laminar flow cabinet.
(b) Piezoelectric dispenser with the orifice diameter of 120 µm. (c) The image of a sodium alginate
porogen captured by a high-speed camera during ejection. (d) The morphology of the printed alginate
porogens after collecting them in the crosslinker pool showing uniform sizes of the beads. (e) The
histogram of the diameter distribution of porogens (n=80).
2.3 Cell preparation
To obtain human ADSCs, surgically discarded adipose tissues were obtained from patients who
underwent an elective adipose tissue removal process (e.g., panniculectomy) at the Pennsylvania State
University (Hershey, PA) with patients’ consent and approval from the Institutional Review Board (IRB
protocol #4972). The tissue samples were minced and rinsed vigorously multiple times with Hanks
Balanced Salt Solution (HBSS) until excess blood was removed. Minced adipose tissue was digested
with 0.2% of collagenase (Gibco, MD) and incubated at 37 °C on a shaker for 2 h. The collagenase-
digested tissue was centrifuged for 10 min at 2,000 rpm and the pellet was obtained. The pellet was then
suspended in red blood cell (RBC) lysis buffer and filtered through 100 and 40 µm filters to remove large
tissue particles and then centrifuged. The pellet was re-suspended in HBSS buffer and layered on Ficoll
Page 7 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

8
gradient centrifuge for 30 min at 2,000 rpm, and the white layer (middle) band was collected, which was
called stromal vascular fraction (SVF). SVF was re-suspended in magnetic activated cell sorting (MACS)
buffer and subjected to AutoMACs Pro cell sorter (Milteyni Biotec, CA) to isolate ADSCs using CD90
microbeads, followed by verification of flow cytometry against CD73 and CD90 (supplementary figure
S1), which were specific for stem cells. The sorted ADSCs were cultured in DMEM/F12 supplement
with 20% FBS, 100 U/mL penicillin and 100 µg/mL streptomycin at 37 °C with 5% CO2. Cell medium
was changed every third day.
2.4 Micro-fabrication of pTSs
ADSCs were expanded, suspended in culture medium, and centrifuged at 1,600 rpm for 5 min to form
cell pellet at the bottom of a 15 mL conical tube. After removal of the supernatant, ADSCs were then
transferred into 1.5 mL tubes, and further centrifuged at 2,000 rpm for 5 min to remove the culture
medium and obtain concentrated cell pellets. Porogens in CaCl2 solution were then transferred into 1.5
mL tubes, and centrifuged at 5,000 rpm for 8 min, followed by removal of the supernatant to obtain
porogens.
Next, cells and porogens, in 5:3 ratio, were transferred into another 1.5 mL tube using a syringe
(Hamilton Company, NV). The cell/porogen mixture was stirred, aspirated and released for two times
using the syringe, followed by injecting it into alginate capsules with approximately 3 mm per 1 µL
cell/porogen mixture. Both ends of the capsules were closed using vascular clamps (Thomas Scientific,
NJ). The injected cell/porogen mixture was cultured for seven days, and culture medium was changed
every other day. Subsequently, alginate capsules were de-crosslinked using sodium citrate solution (4%
(w/v) in sterile DI water) for 5 min leaving pTSs behind. The pTSs were rinsed with phosphate buffer
saline (PBS) twice for 10 min after removing the sodium citrate solution. The sTSs (solid counterparts
as a control group) were also fabricated without inclusion of porogens. The pTSs and sTSs were cultured
Page 8 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

9
for two more weeks under the same culture conditions described in Section 2.3, where the media was
changed every other day. Cell densities for pTSs and sTSs were ~1 and ~1.6×105 cells/mm, respectively.
The morphology of cell/porogen mixture and pTSs were observed by the EVOS microscope, and their
diameter was measured at different time points (Days 0, 2, 4, 7, 10, 14 and 21) using ImageJ software.
2.5 3D Reconstruction of pTSs
The pTSs cultured for different time periods (Days 7, 14 and 21) were fixed with 4% paraformaldehyde
overnight. They were then cut into smaller samples with 3-4 mm in length and transferred into a
transparent glass tube (1.6 mm inner diameter) with PBS, followed by sealing the ends with Parafilm
(Bemis Company, Inc., WI). The glass tube was inserted into another glass tube with a larger diameter
(4.5 mm outer diameter), sealed on both ends with Parafilm, and inserted into a home-built solenoid coil
(4.6 mm inner diameter) for magnetic resonance microscopy (MRM). MRM was conducted on a 14.1
Tesla Agilent (Palo Alto, CA) micro-imaging system using a 1,000 mT/m gradient set. Standard 3D
gradient echo imaging was conducted. With a repetition time of 180 ms (echo time: 15 ms) and 16
averages, the acquisition time was about 6.5 h. The achieved isotropic resolution was 20 µm (field of
view: 4 × 1.8 × 1.8 mm; matrix size: 200 × 90 × 90). By zero filling the dataset in all three directions by
a factor of two (Matlab, MathWorks, MA), the final resulting pixel resolution in images was 10 µm
isotropic. All MRM images were imported into Avizo software (Thermo Fisher Scientific, OR), and 3D
segmentation was performed to identify pores and tissue in the images. Upon creation of the 3D model,
the volume of pores (Vp) and tissue (Vt) were calculated in Avizo, and the porosity (P) was calculated
using Equation (1):
𝑃 = 𝑉
𝑝
𝑉
𝑝 + 𝑉
𝑡
×100%
(Equation 1)
Page 9 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

10
The volume of connected components of pores (Vcp) was measured using a built-in module in Avizo and
the pore interconnectivity (IC) was calculated using Equation (2):
𝐼𝐶 =𝑉
𝑐𝑝
𝑉
𝑝
× 100%
2.6 Scanning electron microscope (SEM) imaging
The surface topography of pTSs was accessed using field emission scanning electron microscopy
(FESEM, Zeiss SIGMA VP, Zeiss, NY). Samples cultured for different time periods (Days 7 and 21)
were fixed in 4% paraformaldehyde overnight. Samples were then washed with PBS and dehydrated
using graded ethanol solutions (25, 50, 75, 95 and 100%, sequentially). To ensure complete removal of
water, dehydrated samples were then further dried in a critical point dryer (CPD300, Leica, Germany).
Upon complete dehydration, samples were sputter coated with gold using a sputter coater (Bal-Tec SCD-
050, Leica, Germany), and observed at an accelerating voltage of 3kV.
2.7 Histological analysis
The pTSs and sTSs cultured for different time periods (Days 7 and 21) were fixed with 4%
paraformaldehyde and embedded in paraffin using an automatic tissue processor (TP 1020, Leica,
Germany). Next, strands were gradually dehydrated in alcohol, cut into 8 µm sections and placed onto
charged slides. Sections were then stained with hematoxylin and eosin (H&E) using Leica Autostainer
XL (Leica, Germany). Coverslips were mounted on the slides with Xylene Substitute Mountant (Thermo
Fisher Scientific, PA). Samples were imaged using the EVOS fluorescence microscope, and the
diameters of the internal pores and the porous area were measured using ImageJ software.
2.8 Cytoskeleton staining
Organization of ADSCs in pTSs and sTSs was assessed by F-actin/nucleus staining. At the designated
time points (Days 7 and 21), samples were histologically sectioned in 8 µm thickness as explained
(Equation 2)
Page 10 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

11
previously. Samples were treated in 0.1 % Triton-X 100 for 10 min, followed by being blocked in 1 %
bovine serum albumin (BSA) for 30 min. Subsequently, samples were incubated in Alexa Fluor 568-
conjugated phalloidin (1:100, 1 h) for F-actin cytoskeleton visualization, and Hoechst 33258 (1:200, 5
min) for cell nucleus visualization, sequentially. Samples were then imaged using the EVOS fluorescence
microscope.
2.9 Cell viability and proliferation assays
Alginate capsules were de-crosslinked after 7 days of culture, and cell viability of pTSs and sTSs was
assessed using LIVE/DEAD staining, in which calcein acetoxymethyl ester (calcein AM, Life
Technologies, CA) labeled living cells in green, and ethidium homodimer-1 (EthD-1, Invitrogen, CA)
stained dead cells in red. Samples were rinsed with PBS before staining. After 30 min of incubation with
calcein AM and EthD-1 at a concentration of 1.0 µM each, samples were rinsed with PBS twice for 10
min, and imaged using the EVOS fluorescence microscope. ImageJ software was used for quantitative
analysis for red- and green-fluorescent cells.
Proliferation of cells in pTSs and sTSs was further compared quantitatively using a Cell Counting Kit-8
(CCK-8; DOJINDO molecular technologies, Inc., Japan). To ensure that the initial cell number for each
sample was identical, 5 µL of cell pellet was used in order to prepare both pTSs and sTSs. Upon culturing
seven days, alginate capsules were de-crosslinked, and the proliferation rate of cells were measured at
Days 7, 10, 14 and 21. Samples were transferred to a 24-well plate and incubated with 10% CCK-8
reagent for 4 h. After incubation with CCK-8 reagent, the formazan dye generated by the activity of
dehydrogenases in cells was proportional to the number of living cells and the absorbance at 450 nm was
measured using a microplate reader (PowerWaveX, BioTek, Winooski, VT).
Page 11 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

12
2.10 Self-assembly of pTSs
After seven days of culture, alginate capsules were dissolved and then samples were washed with PBS.
Next, samples were fluorescently labeled using cell trackers of 5-chloromethyl fluorescein diacetate
(CMFDA) and 5-(and-6)-(((4-chloromethyl) benzoyl) amino) tetramethylrhodamine (CMTMR,
Invitrogen, MA) in green and red, separately. Samples were stained with cell trackers with a
concentration of 1 µM for 30 min in serum-free medium according to the manufacturer’s protocol. The
stained pTSs were then closely placed to each other with ensuring contact. A minimum amount of culture
medium was applied to facilitate cell survival without losing physical contact for one week. Fluorescence
images were taken every day using the EVOS fluorescence microscope, to monitor the intermixing of
pTSs. The total area of two pTSs and intermixing region were measured separately using ImageJ software,
and the intermixing ratio was calculated by dividing the intermixing region by the projectional area of
fusing pTSs [37].
2.11 Differentiation of pTSs into tissue-specific lineages
The functionality of pTSs was evaluated by differentiating them into chondrogenic and osteogenic
lineages. pTSs were cultured in alginate capsules for one week to facilitate cell aggregation prior to
differentiation. Next, strands were cultured in human chondrocyte differentiation media and human
osteoblast differentiation medium (Cell Applications, CA) for chondrogenesis and osteogenesis in
alginate capsules for another week, respectively, followed by de-crosslinking of the alginate capsules.
The pTSs under chondrogenic and osteogenic differentiation were further cultured for two and three
more weeks, respectively, according to manufacturer’s instruction. Differentiated pTSs were then fixed,
sectioned, and stained with H&E for imaging as described previously. The sTSs were used as a negative
control group.
Page 12 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

13
2.12 Immunohistochemistry assay
All primary monoclonal antibodies were purchased from Abcam (AM) and fluorescence-conjugated
secondary antibodies were purchased from Life Technologies (CA). Sections of pTSs (differentiated into
the chondrogenic linage) were then treated using Triton-X 100 (0.1 % in PBS, 10 min), followed by
being blocked with normal goat serum (NGS, 10 % in PBS, 1 h). The samples were then incubated with
the monoclonal rabbit anti-human collagen type II (COL-II, 1:200), mouse anti-human aggrecan (1:50)
and NGS (negative control) for 1 h, respectively. Samples were washed twice with PBS and incubated
with the secondary antibodies (goat anti-rabbit IgG (H+L)-Alexa Fluor 647 for COL-II, and goat anti-
mouse IgG (H+L)-Alexa Fluor 488 for aggrecan, both at 1:200 dilution) for another hour. After rinsing
thrice with PBS, samples were finally incubated with Hoechst 33258 (1:200, 5 min) for nucleus
visualization, and mounted with Fluoromount-GTM (Invitrogen, MA). Using the same protocol, sections
of pTSs undergoing osteogenic differentiation were stained for bone-specific primary antibodies,
including rabbit anti-human bone sialoprotein (BSP, 1:200) and runt-related transcription factor 2
(RUNX2, 1:200), followed by incubation with the secondary antibodies (goat anti-rabbit IgG (H+L)-
Alexa Fluor 647 for BSP, and goat anti-mouse IgG (H+L)-Alexa Fluor 488 for RUNX2, both at 1:200
dilution). Images for each marker were taken by a Zeiss Axiozoom microscope (Carl Zeiss Microscopy
LLC, NY). The sTSs were used as a negative control group.
2.13 Biochemical assay
To ensure that the initial cell numbers for each sample were identical, 10 µL of cell pellet was used to
prepare both pTSs and sTSs. For samples undergoing chondrogenic differentiation, sulfated
glycosaminoglycan (sGAG) content was determined by 1,9-dimethylmethylene blue (DMMB) dye-
binding assay [38]. Samples were de-crosslinked, followed by being rinsed with PBS and digested in 500
μL solution of 0.1 mg/mL papain extraction reagent at 65 °C for 18 h. 20 µL of the digested samples
Page 13 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

14
were mixed with 200 µL DMMB solution, and the absorbance was measured at 525 nm using a
microplate reader. Serially diluted solution of solution of chondroitin 4 sulfate was prepared as standard,
and sGAG contents of samples were calculated according to the standard curve. For samples undergoing
osteogenic differentiation, alkaline phosphatase (ALP) activity assay was conducted using an assay kit
(K412-500; BioVision, Inc., Milpitas, CA) according to the manufacturer's instructions. The
differentiated samples were re-suspended in assay buffer and subsequently centrifuged at 13,000 g for 3
min at 4˚C to remove insoluble material. Supernatant was mixed with p-nitrophenyl phosphate (pNPP)
substrate and incubated at 25˚C for 60 min. The optical density of the resultant pNPP at 405 nm was
determined. A Quant-iTTM PicoGreen dsDNA Assay Kit (Molecular Probes Inc., Eugene, OR) was used
to determine the DNA amount of pTSs and sTSs, according to the manufacturer’s instructions.
Fluorescence intensity was determined by a SpectraMax multi-detection microplate reader (Molecular
Devices, Inc., Sunnyvale, CA), using the wavelength of 480 nm (excitation) and 520 nm
(emission). sGAG content and ALP activity was normalized to dsDNA content.
2.14 Gene expression using quantitative real-time polymerase chain reaction (qRT-PCR)
In order to evaluate the cartilage- and bone-specific gene expression levels, differentiated pTSs were
homogenized in TRIzol reagent (Life Technologies, Carlsbad, CA), followed by adding 0.2 mL
chloroform per 1 mL TRIzol reagent and centrifuging the mixture at 12,000 g for 15 min at 4˚C. The
upper aqueous phase with RNA was transferred, and the RNA was then precipitated by adding 0.5 mL
isopropyl alcohol per 1 mL TRIzol reagent, followed by centrifuging at 12,000 g for 10 min at 4˚C.
Subsequently, the precipitated RNA was rinsed twice by 75% ethanol, air-dried for 10 min and dissolved
in 50 µL diethyl pyrocarbonate (DEPC)-treated water. RNA concentration was measured using a
Nanodrop (Thermo Fisher Scientific, PA). Reverse transcription was performed using AccuPower®
CycleScript RT PreMix (BIONEER, Korea) following the manufacturer's instructions. Gene expression
was analyzed quantitatively with SYBR Green (Thermo Fisher Scientific, PA) using a 7500 Real-Time
Page 14 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

15
PCR system (Applied Biosystems®, Life Technologies, USA). Cartilage-specific genes tested in
chondrogenic pTSs included collagen type II (COL2A1), Aggrecan, collagen type I (COL1) and
chondrogenic transcription factor SOX9. Bone-specific genes tested in osteogenic pTSs included BSP,
RUNX2, COL1 and collagen type VI (COL6A1). The gene sequences are listed in table 1. Expression
levels for each gene were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH).
Table 1: Primer information for qRT-PCR
Gene
Primer
Aggrecan
Forward
5’-TCCCCTGCTATTTCATCGAC-3'
Reverse
5’-CCAGCAGCACTACCTCCTTC-3'
SOX9
Forward
5’-AGCGAACGCACATCAAGAC-3'
Reverse
5’-CTGTAGGCGATCTGTTGGGG-3'
COL2A1
Forward
5’-CCAGATGACCTTCCTACGCC-3'
Reverse
5’-TTCAGGGCAGTGTACGTGAAC-3'
COL1
Forward
5’-CAGAACGGCCTCAGGTACCA-3'
Reverse
5’-CAGATCACGTCATCGCACAAC-3'
BSP
Forward
5’-AACGAAGAAAGCGAAGCAGAA-3'
Reverse
5’-TCTGCCTCTGTGCTGTTGGT-3'
RUNX2
Forward
5’-GGTTAATCTCCGCAGGTCACT-3'
Reverse
5’-CACTGTGCTGAAGAGGCTGTT-3'
COL6A1
Forward
5’-CCTGGAGGGCTACAAGGAA-3'
Reverse
5’-GTGCTTGGCCTCGTTCAC-3'
2.15 Statistical analysis
All data were presented as the mean ± standard deviation unless stated otherwise, and were analyzed
using Student’s t-test and one-way analysis of variance (ANOVA) to test for significance when
comparing the data. Post-hoc Tukey’s multiple comparison test was used to determine the individual
differences among the groups. Differences were considered significant at p < 0.05 (*), p < 0.01 (**), and
p < 0.001 (***). All statistical analysis was performed by Statistical Product and Service Solutions
software (SPSS, IBM, USA).
Page 15 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

16
3. Results
3.1 Formation of pTSs
In this study, PIJ printing was applied to generate alginate porogens (figure 2). The porogens exhibited
a spherical morphology after jetting out of the nozzle (figure 2c), and well maintained their shape after
submerging into the crosslinker pool (figure 2d). The average diameter of porogens was 85 ± 11 µm,
where most of the porogens distributed within 80 to 90 µm range (figure 2e). Tubular alginate capsules
were obtained with permeability and uniform structure [36], and they preserved their shape during the
entire culture period. The average inner and outer diameter, and wall thickness of alginate capsules were
896 ± 27, 1,254 ± 26 and 182 ± 50 μm, respectively (figure 3a).
In order to fabricate pTSs, pellets of ADSCs were mixed with porogens in 5:3 ratio homogeneously
(figure 3b), and injected into the alginate capsules (figure 3c1). After two days of culture in capsules,
aggregation started and some large pores were observed between cell aggregates which were generated
by alginate porogens (figure 3c2). The diameter of the pTSs showed no significant difference when
cultured in alginate capsules from Day 0 to 7 (figure 3d). After seven days of culture, alginate capsules
were dissolved and pTSs did not segregate (figure 3c4), demonstrating sufficient cell-cell adhesion of
ADSCs. After the removal of alginate capsules, the diameter of pTSs decreased from 879 ± 42 μm at
Day 7 to 541 ± 35 μm at Day 14, and kept constant thereafter (figures 3c5-c7 and 3d).
Page 16 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

17
Figure 3: Formation and morphology change of pTSs. (a) Dimensions of alginate capsules (n=3). (b)
Homogenous distribution of cells and porogens in the mixture, with red arrows indicating the porogens
and an image with higher magnification showing well-defined porogens surrounded by cells. (c)
Morphology change of microinjected cells in the alginate capsule before (c1-c3) and after (c4-c7) de-
crosslinking. (d) Diameter change of pTSs overtime (n=3).
3.2 Pore morphology change
The porosity and pore interconnectivity of pTSs were visualized and quantified by 3D reconstruction
using magnetic resonance imaging (MRI) at a high resolution (10 µm). 3D reconstructed pTSs exhibited
diminished number of pores and denser structure on the surface (figures 4a1-c1). The cross-sectional
view of the reconstructed samples demonstrated the porous internal structure with interconnected pores
(figures 4a2-c2). Quantitative analysis revealed a porosity-level of ~34% at Day 7, which significantly
diminished to ~29% at Day 14 and ~25% at Day 21 (figure 4d). Additionally, pore interconnectivity
maintained a higher value during 21 days of culture, which decreased from ~96% at Day 7 to ~88% at
Day 14 and ~84% at Day 21 (figure 4e). As expected, the generated pTSs were well interconnected (with
~100% connectivity) during the entire culture period (figure 4f).
Page 17 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

18
Figure 4: 3D reconstruction of pTSs samples. (a-c) Orthogonal and cross-sectional view of the
reconstructed samples at different time points. (d-f) Quantification of porosity, pore interconnectivity
and tissue interconnectivity of pTSs (n=3; *p < 0.05, **p < 0.01, and ***p < 0.001).
The ultra-morphology of pTSs that were cultured for two weeks after the removal of alginate capsules
was also characterized by SEM. The pTSs at Day 7 exhibited a porous morphology with a rough surface
and visible pores (figure 5a1). Their surface became denser due to diminishing pores at the end of an
extended culture period (21 days) (figure 5a2). Histological characterization was performed to evaluate
the change of pore morphology inside pTSs. At Day 7, pores showed irregular shapes and diameters,
with a total porous area of ~33% (figure 5b1). In contrast, sTSs showed a uniform and densely packed
cell arrangement without apparent pores (figure 5b2). After two more weeks of culture, pTSs revealed
Page 18 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

19
decreased external diameter and internal pore size (spherical morphology and smaller dimensions (~50
µm) with respect to porogens) due to the contraction of the tissue (figure 5c3). Meanwhile, the total
porous area at Day 21 decreased to ~23% in the sagittal cross-section. In comparison to the cell
morphology at Day 7, cells at Day 21 were more densely organized around the pores. To further access
the effect of pores on cellular organization, cytoskeletal staining was performed. In two weeks of culture,
F-actin fibers were evenly distributed and spread in sagittal sections during the prolonged culture time.
The nuclei tended to organize around the pores in pTSs (figure 5c1 and c3), while those in the sTSs were
randomly oriented (Figure 5c2).
Figure 5: (a) SEM images demonstrating the surface morphology of pTSs at Days 7 and 21. (b) Images
of H&E stained sagittal sections at Days 7 and 21 demonstrating the maintenance of the porous
microstructure as compared to the control group. (c) Fluorescence images of cytoskeletal staining of
ADSCs in pTSs and sTSs (yellow dashed lines indicate pores).
3.3 Up-regulated cell viability and proliferation in porous strands
Page 19 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

20
Cell viability of pTSs was compared to that of sTSs (control). LIVE/DEAD staining of pTSs at Day 7
showed a small number of dead cells, while a relatively large amount of dead cells were observed in the
control group (figures 6a-b). The quantitative analysis demonstrated that cell viability in pTSs (> 90%)
was higher than that of sTSs (< 70%, p < 0.05) at Day 7 (figure 6c). Since cell density was too high in
samples during the prolonged culture time, particularly in their core, LIVE/DEAD imaging beyond Day
7 was not sufficient for quantification of cellular viability. Hense, CCK-8 assay was performed and the
absorbance was normalized to the value of sTSs obtained at Day 7 (figure 6d). At Day 7, the proliferation
rate of pTSs were ~20% higher than the control group, which was consistent with the results obtained
from LIVE/DEAD straining. The proliferation rate of the control group decreased to 74% at Day 10 and
62% at Day 14 compared to Day 7, and maintained at 62% at Day 21. On the contrary, proliferation rate
of pTSs increased by ~13% at Day 10 compared to that at Day 7, demonstrating that cell number
increased after de-crosslinking the alginate capsule. After Day 10, the proliferation rate of pTSs kept
constant, which was significantly higher than that of the control group (p < 0.05), demonstrating superior
cell survival within pTSs.
Page 20 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

21
Figure 6: (a-b) Z-stack fluorescence images of LIVE/DEAD staining of ADSCs in pTSs and sTSs at
Day 7. (c) Cell viability in pTSs and sTSs at Day 7 (n=3; *p < 0.05 vs. control group). (d) Proliferation
rates in pTSs and sTSs during a 21-day culture (n=3; *p < 0.05, and **p < 0.01 vs. control group).
3.4 Self-assembly of pTSs
In order to investigate the self-assembly properties of pTSs, cells in two pTSs were fluorescently labeled
with red and green separately, and the pTSs were closely placed next to each other with an initial
overlapping region of 9 ± 2% (figure 7a). Their fusion started rapidly in the first two days, indicated by
the sharp increase in the overlapping area of two fluorescent signals (figure 7b-d). The area of the
overlapped region increased from 36% at Day 1 to 70% at Day 2. The speed of fusion slowed down from
Day 3, and the overlapped area slightly increased from 81% at Day 3 to 89% at Day 7 (figure 7e). After
a week of culture, pTSs fused into a single patch of tissue, evidenced by the invisible boundary in between.
Page 21 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

22
Figure 7: (a-d) Cellular intermixing of pTSs at different time points with increased area of the
overlapped fluorescent signals. Inset: optical images indicating the absence of the gap between the two
pTSs over culture time. (e) Quantitative measurement of the ratio of the intermixed region over a week
culture (n=3).
3.5 pTSs promote chondrogenic and osteogenic differentiation of ADSCs
To further investigate the potential of pTSs for functional tissue engineering applications, pTSs were
differentiated into chondrogenic and osteogenic lineages. The chondrogenic tissue was densely packed
both in solid and porous forms, indicating the chondrogenic differentiation of cells and deposition of
cartilage-specific ECM proteins (figures 8a-b). Particularly, pTSs were robust against chondrogenic
induction with well-maintained integrity and internal porosity (figure 8a). Immunohistochemistry
staining was performed to evaluate the expression of cartilage-specific markers including COL-II and
Aggrecan (figures 8c-f). Images showed a significant expression of COL-II and Aggrecan distributed
throughout pTSs, which demonstrated that ADSCs were able to be chondrogenically differentiated in the
form of strands, presented chondrogenic phenotype, and formed cartilage-like tissue in vitro. The
chondrogenic induction was further confirmed by quantitative sGAG deposition and cartilage-specific
gene expression. With the identical initial cell numbers, differentiated pTSs supported significantly
Page 22 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

23
greater sGAG matrix deposition (~2-fold increase compared to sTSs, figure 8g). When sGAG contents
were normalized against the dsDNA content, pTSs exhibited an ~1.6-fold increase compared to sTSs
(figure 8h). In terms of gene expression, pTSs had an up-regulated expression of Aggrecan and COL2A1
(with ~3.2 and ~4.1-fold increase, respectively) compared to the control group (figure 8i). In addition,
the gene expression in pTSs revealed an ~3.1-fold increase with regard to SOX9, a transcription factor
required for maintaining the chondrocyte phenotype during chondrogenesis [6], and ~1.9-fold increase
for COL1 against the control group.
Figure 8: Chondrogenesis of pTSs. (a-b) H&E staining showing tissue and pore morphology (insert:
high magnification images demonstrating cellular morphology). (c-f) Expression of the cartilage-specific
markers (COL-II and Aggrecan) after three weeks of differentiation in pTSs and sTSs. (g) Quantitative
analysis of sGAG content in differentiated samples (n=3, *<0.05). (h) sGAG content normalized to DNA
amount. (i) Cartilage-related gene expression (n=3, *p < 0.05 vs. the control group).
Page 23 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

24
The pTSs were also successfully differentiated under osteogenic conditions. As compared to the solid
counterparts, cell density throughout pTSs was relatively uniform, indicated by dense tissue formation
near the core and around the internal pores (figure 9a). As compared to the uniform cell density
throughout the bodies of chondrogenically differentiated sTSs, higher cell density was found at the
superficial area of osteogenically induced ones, which was indicated by the dark purple area in figure 9b.
Immunohistochemistry of bone-specific markers including BSP and RUNX2 revealed positive staining
throughout both pTSs and sTSs (figures 9c-f). BSP, a highly glycosylated and sulfated phosphoprotein,
is found almost exclusively in mineralized connective tissues such as bone [39], and RUNX2 is a key
transcription factor associated with osteoblast differentiation [40]. Stronger staining was
apparently observed inside the deep regions and around the pores for both markers (figures 9c-e). In the
biochemical assessment of ALP activity, the pTSs showed superior overall (~4.2-fold increase, p < 0.05)
and DNA-normalized (~3.1-fold increase, p < 0.05) ALP levels (figures 9g-h). Effectively, pTSs
exhibited significantly higher gene expression levels of BSP and RUNX2 (~5.8 and ~2.1-fold increase,
respectively, figure 9i) and at the same time, demonstrated greater expression of COL1 and COL6A1
(~4.4 and ~2.3-fold increase, respectively) as compared to sTSs. These results demonstrated that ADSCs
differentiated into an osteogenic lineage, in which pTSs were advanced over their solid counterparts.
Page 24 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

25
Figure 9: Osteogenesis of pTSs. (a-b) H&E staining showing tissue and pore morphology (insert: high
magnification images demonstrating cellular morphology). (c-f) Expression of the bone-specific markers
(BSP and RUNX2) after four weeks of differentiation in pTSs and sTSs. (g) Quantitative analysis of
ALP activity in differentiated samples (n=3; *p < 0.05). (h) ALP activity normalized to DNA amount.
(n=3; *p < 0.05) (i) Bone-related gene expression in pTSs and sTSs (n=3; *p < 0.05 and **p < 0.01 vs.
the control group).
4. Discussions
In this study, we presented the biofabrication of porous cell aggregates using the mixture of ADSCs and
sodium alginate beads as porogens in the form of strands. During the fabrication of pTSs, de-crosslinking
was performed using sodium citrate solution to remove the alginate capsules and porogens near the
surface of the strands after the strand formation. Since the sodium citrate solution was difficult to reach
the depth of the strand which had high cell density, and alginate was also hard to diffuse out of the strand,
Page 25 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

26
some porogens or alginate residuals might remain inside pTSs without effecting their structure integrity.
It was shown that the internal porous architecture of pTSs were preserved during the 3-week culture.
There are reasons for choosing alginate to fabricate capsules and porogens. As unmodified alginate
inhibits cell adhesion due to lack of cell receptors [41], cells did not adhere to the capsule and porogens,
which promoted cell-cell interactions and hence facilitated their aggregation. In addition, the integrity of
pTSs was not impacted during the de-crosslinking process. The detachment of cells from the capsule
wall also supported cell-cell adhesion. More importantly, alginate is semi-permeable, which allows cell
culture media and oxygen to permeate through the capsule wall, and at the same time prevents cells from
migrating out during the aggregation process [6]. Coaxial system has been utilized to print cell-
encapsulated alginate conduits in our previous and some other studies [36, 42, 43]. In theory, the coaxial
system can support one-step formation of porous strands by feeding the cell/porogen mixture through the
inner nozzle of the coaxial apparatus; however, the preparation of cell/porogen mixture was time-
consuming and costly compared to the preparation of alginate capsules. In this study, therefore,
subsequent injection of cell/porogen mixture into alginate capsules was performed instead of direct
coaxial bioprinting, which significantly reduced the waste of biologics.
In terms of pore generation, cell aggregation started after the cell/porogen mixture was injected into
alginate capsules, resulting in dark (cells) and bright (porogens) regions from Day 2 (figure 3c). The
densely packed cells interacted with each other biologically and enclosed the porogens to keep the
integrity of the entire strand. In the literature, a three-step model that involves initial cell-cell contacts,
cadherin accumulation, and aggregate compaction has been proposed for cell aggregate formation [44-
46]. First, loose cells rapidly aggregate via the binding of cell surface integrin to arginylglycylaspartic
acid (RGD) motifs in the ECM. A delay phase follows this aggregation and exhibits up-regulated
cadherin expression and accumulation. Finally, homophilic cadherin-cadherin binding between two cells
confers strong cell adhesion, forming compact cellular aggregates [44]. Particularly, it has also been
Page 26 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

27
reported that ADSCs cultured in the form of 3D aggregates produced much more ECM proteins
(e.g., tenascin C, collagen VI α3, and fibronectin) and cytokines compared to monolayer 2D culture [47,
48].
After the de-crosslinking of alginate capsules and porogens near the surface of the strand, a porous
morphology was observed on the surface of pTSs at Day 7 (figure 5a1). During the extended culture time,
pores on the external surface diminished gradually due to the proliferation of cells (figure 5a2). On the
other hand, internal porous structure was able to maintain for a longer term. This might be due to the fact
that porogens inside pTSs were entrapped, which slowed down their de-crosslinking process. At Day 7,
although cell aggregates formed and were sufficient to keep the integrity of pTSs, cell density inside
pTSs was still not as dense as that in solid counterparts (figure 5b2), and the cells organized around the
porogen clusters (figure 5b1). Conversely, individual pores were able to be apparently distinguished from
each other at Day 21, which could be attributed to the cellular invasion into the gaps between porogens
(figure 5b3). The cell density in non-porous regions was similar to that in solid counterparts.
Cell aggregates in the form of spheroids [49], strands [6], and rings [50] have the advantages of enhanced
cell-cell contacts and tight junctions, which imitate the cellular arrangement of native tissues. However,
cell aggregates are usually short of vascularization, leading to mass transfer to the cells only relying on
diffusion, which is detrimental to cells [51]. For example, in spheroids with a diameter larger than 200
µm, diffusional gradients limit the cells in the core to obtain adequate nutrients and remove the waste,
resulting in cell death [51]. The oxygen diffusion distance is typically within the scale of hundred
micrometers in metabolically highly active tissues such as liver and kidney [52, 53]. In terms of TE
constructs, it is expected that most cells reside within a couple of hundred microns from the convective
flow. In this study, sTSs showed a relatively low cell viability during the entire culture period as indicated
by the quantitative CCK-8 assay, in which pTSs exhibited ~2.1-fold increase of proliferation rate
compared to sTSs at Day 21 (figure 6d). It has been considered that alginate is capable of media and
Page 27 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

28
oxygen transport and supply, and hence, it is commonly used as a structural matrix for cell-laden
constructs. Here, the alginate porogens, remained inside pTSs, provided a connective pathway for
nutrient and oxygen perfusion, which maintained the distances of perfused medium to the surrounding
cells within hundreds of microns and facilitated high cell viability.
From the standpoint of scaffold-based TE, high porosity (60-70%) is desired in scaffold design [28].
Since polymers and crosslinked hydrogels are usually mechanically stronger than immatured cell
aggregates and function as support for tissue growth, high porosity can be achieved without sacrificing
the integrity of the fabricated constructs. In addition, as scaffolds are expected to induce tissue ingrowth,
degrade and finally be replaced by regenerated tissue in the long term, the amount of supporting
biomaterials is minimized through structure optimization [54, 55], which also results in high porosity.
However, pTSs are nearly free of exogenous biomaterials and most space is occupied by cells. Thus, the
degradation of exogenous biomaterials and cell ingrowth are no longer issues. Additionally, our
preliminary data (not presented in this study) revealed that high ratio of porogens (e.g., >50%) hindered
cell-cell contacts, resulting in local cell aggregations without formation of intact pTSs. In this study, the
results showed that porosity of ~30% throughout the culture period supported the pTS integrity and
facilitated high cell viability as well. Another advantage of the fabricated pTSs was their high pore
interconnectivity (~90%), which kept most cells within a diffusive distance of nutrient supply.
Tissue fusion refers to a process in which multiple tissue pieces make contact and self-assemble to form
a single patch of tissue [56, 57]. Tissue fusion is regulated by several factors including cell migration,
cell-cell interactions and cell-matrix interactions [58, 59]. To initiate the fusion process, tissue building
blocks (e.g., pTSs in this study) must be in direct contact [57]. In this presented work, upon close
placement of pTSs, the fusion started at Day 1. Further fusion was observed with larger intermixing over
time resulting in gap closure between pTSs. At Day 7, pTSs were almost completely fused into a single
patch of tissue, without visible gap in between (figure 7). The intermixed pTSs demonstrated the potential
Page 28 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

29
of building large-scale tissues as building blocks for TE, which can be further coupled with bioprinting
[6].
In addition, ADSCs were used as a cell source for the fabrication of pTSs, which were multipotent to
differentiate into multiple lineages, and offer the capability of pTSs to be utilized in various TE
applications. In this study, chondrogenic and osteogenic differentiation were performed. Although both
cartilage- and bone-like tissues were obtained after differentiation, difference in terms of cell density and
organization was observed. The uniformly distributed cell density in chondrogenic culture was probably
attributed to the anaerobic nature of chondrocytes. It has been predicted that oxygen concentration was
1-5% within adult human articular cartilage tissue [60, 61]. The detection of a tonic activation of hypoxia
inducible factor 1 alpha (HIF-1α) within the chondrocytes validated the hypothesis on a hypoxic
environment [62, 63]. The improved chondrogenic differentiation of immature precursor cells under
hypoxic conditions also verified the importance of a low oxygen concentration in chondrogenesis [64,
65]. In the case of chondrogenesis of tissue strands in this study, sTSs were expected to have better
hypoxic environment as compared to pTSs. However, since the strands were differentiated in the form
of ADSC strands, the non-porous structure might hinder the differentiation of ADSCs in the depth of
strands. In comparison, although more oxygen was delivered in the case of pTSs, the porous pathway
might assist the chondrogenic media to reach deeper zones in pTSs, promoted chondrogensis, and offset
the effect of oxygen supply. Such claim was supported by the greater deposition of sGAG matrix and
stronger cartilage-related gene expression than sTSs (figure 8).
With osteogenic induction, more cells migrated to the periphery of sTSs, which could be induced by
insufficient oxygen and nutrient supply. Consequently, limited fluorescent signal of bone-related gene
expression was detected only at the edge of sTSs. In contrast, although cells were also inclined to move
to the edges of pTSs, dense tissue was presented around the pores in deeper zones, which could be the
indication of oxygen supply through pores, resulting in more uniform tissue formation compared to the
Page 29 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

30
solid ones. Such assumption was further confirmed by the expression of BSP and RUNX2 in
immunostaining, ALP activity in biochemical test and bone-related gene expression, which were all
strongly expressed in pTSs (figure 9). Since bone is vascularized and cells are aerobic, insufficient
oxygen and nutrient diffusion inhibit osteoblast viability in the depth of tissue, leading to bone formation
primarily on the periphery [66, 67]. Microporosity is a critical osteoconductive property of tissue-
engineered constructs, which facilitates bone tissue regeneration and vascularization. In conventional
scaffold-based TE, structural parameters including pore size, porosity, and interconnectivity are critical
for proper cell growth and migration, nutrient flow, vascularization, gene expression and ECM
production [52, 68]. Since cell migration is no longer a concern in scaffold-free tissue strands, nutrient
flow and vascularization became vital challenges. This study offers a promising solution to include pores
in dense, packed cell aggregates to enhance mass transfer, particularly for metabolically highly active
tissue types.
The mass transport distance of the nutrients and oxygen, which is known as the intercapillary distance
(ICD), varies with the type of tissue and different sections of the same type of tissue. For example, ICD
of bone has been reported to be ~300 µm [69]. In addition, capillaries are numerous in the depth of 35
µm of the synovium (with an ICD of 17 µm), and the number gradually decreases when the depth
increases [70]. In another study, it has also been reported that the ICD can reach as large as 570 µm in
depth of the synovium [71]. In this study, the diameter of strands was ~600 µm, which was larger than
the ICD in most of native tissues [72, 73], and made the inclusion of porogens important for nutrient
supply. The necessity of porogens in strands with different diameters should be investigated in future
studies. The impact of porogens might be shaded in the case when the strand diameter is closed to ICD
of the targeted tissue, since the nutrient and oxygen would be able to diffuse from the surrounding culture
media. Additionally, the porogen effect may have a limitation when the metabolic activity exceeds
porogen-based mass transport in large-size strands. In such a case, larger porogens and higher porogen
Page 30 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

31
ratio might be helpful to enhance the capability to facilitate sufficient supplement diffusion. Thus, the
porogen ratio and capsule diameter need to accommodate to specific applications depending on the
required ICD and strand size.
Functional neovascularization is another challenge in repair of metabolic tissues [52]. Since engineered
tissue constructs lack established vasculature as compared to transplants [74], survival of the implanted
tissues relies on diffusive nutrient supply and waste removal processes until neovascularization takes
place. In addition, it was recently reported that robust intra-vascularization mainly occurs when cell
aggregates are embedded in hydrogels instead of in a free-standing form [75], which implies that dense
cell aggregates may have limited capability of supporting vascularization. The fabrication of pTSs may
open a new avenue for vascularization of scaffold-free constructs. The porous internal microstructure
may provide space for potential vascularization, which can further enhance the mass exchange and
promote cell viability in the center of the tissue, leading to the prospect for large-scale tissue formation.
5. Conclusion
Herein, we reported a novel concept in inducing micropores in scaffold-free micro-tissues, which can be
used as building blocks for biofabrication of scalable human tissues. ADSCs cultured in a semi-
permeable capsule were able to aggregate and form a structurally integrated pTS after the de-crosslinking
of the capsule and porogens, which enabled nutrient perfusion to the deeper region, preserved cellular
viability and promoted tissue maturation for a long term. In terms of their functionality, pTSs were also
able to self-assemble to larger scale tissue patches, which can be further scaled-up using automated
processes such as bioprinting. In addition, chondrogenesis and osteogenesis of pTSs were successfully
performed, which showed that pTSs better supported regeneration of metabolically highly active tissues
due to the enhanced nutrient supply. The capability of differentiating pTSs of human stem cells to distinct
Page 31 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

32
lineages may open a new avenue in tissue engineering applications, especially patient-specific scalable
tissues on demand.
Acknowledgment
This work has been supported by National Science Foundation Award # 1624515 and the Hartz
Professorship awarded to I.T.O. W.P. was supported through the China Scholarship Council
201308360128 and the Oversea Sailing Project from Jiangxi Association for Science and Technology
(2013). The authors also acknowledge the support from Engineering Science and Mechanics Department,
the Huck Institutes of Life Sciences and Materials Research Institutes.
Competing interests
The authors confirm that there are no known conflicts of interest associated with this publication and
there has been no significant financial support for this work that could have influenced its outcome.
References
[1] Yang S, Leong K F, Du Z and Chua C K 2001 The design of scaffolds for use in tissue engineering.
Part I. Traditional factors Tissue Eng. 7 679-89
[2] Wu Y, Wong Y S and Fuh J Y H 2017 Degradation behaviors of geometric cues and mechanical
properties in a 3D scaffold for tendon repair J. Biomed. Mater. Res. A 105 1138-49
[3] Jakab K, Norotte C, Damon B, Marga F, Neagu A, Besch-Williford C L, Kachurin A, Church K H,
Park H, Mironov V, Markwald R, Vunjak-Novakovic G and Forgacs G 2008 Tissue engineering
by self-assembly of cells printed into topologically defined structures Tissue Eng. Part A 14 413-
21
[4] Norotte C, Marga F S, Niklason L E and Forgacs G 2009 Scaffold-free vascular tissue engineering
using bioprinting Biomaterials 30 5910-7
[5] Napolitano A P, Chai P, Dean D M and Morgan J R 2007 Dynamics of the self-assembly of
complex cellular aggregates on micromolded nonadhesive hydrogels Tissue Eng. 13 2087-94
[6] Yu Y, Moncal K K, Li J, Peng W, Rivero I, Martin J A and Ozbolat I T 2016 Three-dimensional
bioprinting using self-assembling scalable scaffold-free “tissue strands” as a new bioink Sci. Rep.
6
[7] Edmondson R, Broglie J J, Adcock A F and Yang L 2014 Three-dimensional cell culture systems
and their applications in drug discovery and cell-based biosensors Assay Drug Dev. Technol. 12
207-18
[8] Ozbolat I T 2015 Scaffold-based or scaffold-free bioprinting: competing or complementing
approaches? J. Nanotechnol. Eng. Med. 6 024701
Page 32 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

33
[9] Hatch J and Mukouyama Y S 2015 Spatiotemporal mapping of vascularization and innervation in
the fetal murine intestine Dev. Dyn. 244 56-68
[10] Criswell T L, Corona B T, Wang Z, Zhou Y, Niu G, Xu Y, Christ G J and Soker S 2013 The role
of endothelial cells in myofiber differentiation and the vascularization and innervation of
bioengineered muscle tissue in vivo Biomaterials 34 140-9
[11] Richards D, Jia J, Yost M, Markwald R and Mei Y 2017 3D bioprinting for vascularized tissue
fabrication Ann. Biomed. Eng. 45 132-47
[12] Santos M I and Reis R L 2010 Vascularization in bone tissue engineering: physiology, current
strategies, major hurdles and future challenges Macromol. Biosci. 10 12-27
[13] Lovett M, Lee K, Edwards A and Kaplan D L 2009 Vascularization strategies for tissue engineering
Tissue Eng. Part B Rev. 15 353-70
[14] Datta P, Ayan B and Ozbolat I T 2017 Bioprinting for vascular and vascularized tissue
biofabrication Acta Biomater. 51 1-20
[15] Carrier R L, Rupnick M, Langer R, Schoen F J, Freed L E and Vunjak-Novakovic G 2002 Perfusion
improves tissue architecture of engineered cardiac muscle Tissue Eng. 8 175-88
[16] Radisic M, Yang L, Boublik J, Cohen R J, Langer R, Freed L E and Vunjak-Novakovic G 2004
Medium perfusion enables engineering of compact and contractile cardiac tissue Am. J. Physiol.
Heart Circ. Physiol. 286 H507-H16
[17] Ho M H, Kuo P Y, Hsieh H J, Hsien T Y, Hou L T, Lai J Y and Wang D M 2004 Preparation of
porous scaffolds by using freeze-extraction and freeze-gelation methods Biomaterials 25 129-38
[18] Chung EJ, Sugimoto M, Koh JL and Ameer GA 2011 Low-pressure foaming: a novel method for
the fabrication of porous scaffolds for tissue engineering Tissue Eng. Part C Methods 18 113-21
[19] Harris LD, Kim BS and Mooney DJ 1998 Open pore biodegradable matrices formed with gas
foaming J. Biomed. Mater. Res. 42 396-402
[20] Wang X, Li W and Kumar V 2006 A method for solvent-free fabrication of porous polymer using
solid-state foaming and ultrasound for tissue engineering applications Biomaterials 27 1924-9
[21] Katoh K, Tanabe T and Yamauchi K 2004 Novel approach to fabricate keratin sponge scaffolds
with controlled pore size and porosity Biomaterials 25 4255-62
[22] Nam Y S and Park T G 1999 Biodegradable polymeric microcellular foams by modified thermally
induced phase separation method Biomaterials 20 1783-90
[23] Park J S, Woo D G, Sun B K, Chung H M, Im S J, Choi Y M, Park K, Huh K M and Park K H
2007 In vitro and in vivo test of PEG/PCL-based hydrogel scaffold for cell delivery application J.
Control. Release. 124 51-9
[24] Khutoryanskaya O V, Mayeva Z A, Mun G A and Khutoryanskiy V V 2008 Designing temperature-
responsive biocompatible copolymers and hydrogels based on 2-hydroxyethyl (meth) acrylates
Biomacromolecules 9 3353-61
[25] Tsang V L and Bhatia S N 2004 Three-dimensional tissue fabrication Adv. Drug Deliv. Rev. 56
1635-47
[26] Bryant S J, Cuy J L, Hauch K D and Ratner B D 2007 Photo-patterning of porous hydrogels for
tissue engineering Biomaterials 28 2978-86
[27] Hahn M S, Taite L J, Moon J J, Rowland M C, Ruffino K A and West J L 2006 Photolithographic
patterning of polyethylene glycol hydrogels Biomaterials 27 2519-24
[28] Hollister S J 2005 Porous scaffold design for tissue engineering Nat. Mater. 4 518-24
[29] Wu Y, Wang Z, Fuh J Y H, San Wong Y, Wang W and San Thian E 2016 Mechanically-enhanced
three-dimensional scaffold with anisotropic morphology for tendon regeneration J. Mater. Sci.
Mater. Med. 27 115
Page 33 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

34
[30] Wu Y, Sriram G, Fawzy A S, Fuh J Y H, Rosa V, Cao T and Wong Y S 2016 Fabrication and
evaluation of electrohydrodynamic jet 3D printed polycaprolactone/chitosan cell carriers using
human embryonic stem cell-derived fibroblasts J. Biomater. Appl. 31 181-92
[31] Hwang C M, Sant S, Masaeli M, Kachouie N N, Zamanian B, Lee S H and Khademhosseini A
2010 Fabrication of three-dimensional porous cell-laden hydrogel for tissue engineering
Biofabrication 2 035003
[32] Zhong M, Sun J, Wei D, Zhu Y, Guo L, Wei Q, Fan H and Zhang X 2014 Establishing a cell-
affinitive interface and spreading space in a 3D hydrogel by introduction of microcarriers and an
enzyme J. Mater. Chem. B 2 6601-10
[33] Liu B, Liu Y, Lewis A K and Shen W 2010 Modularly assembled porous cell-laden hydrogels
Biomaterials 31 4918-25
[34] Gao G, Schilling A F, Hubbell K, Yonezawa T, Truong D, Hong Y, Dai G and Cui X 2015
Improved properties of bone and cartilage tissue from 3D inkjet-bioprinted human mesenchymal
stem cells by simultaneous deposition and photocrosslinking in PEG-GelMA Biotechnol. Lett. 37
2349-55
[35] Akkouch A, Yu Y and Ozbolat I T 2015 Microfabrication of scaffold-free tissue strands for three-
dimensional tissue engineering Biofabrication 7 031002
[36] Zhang Y, Yu Y, Akkouch A, Dababneh A, Dolati F and Ozbolat I T 2015 In vitro study of directly
bioprinted perfusable vasculature conduits Biomater. Sci. 3 134-43
[37] Olsen T R, Mattix B, Casco M, Herbst A, Williams C, Tarasidis A, Simionescu D, Visconti R P
and Alexis F 2015 Manipulation of cellular spheroid composition and the effects on vascular tissue
fusion Acta Biomater. 13 188-98
[38] Farndale R W, Sayers C A and Barrett A J 1982 A direct spectrophotometric microassay for sulfated
glycosaminoglycans in cartilage cultures Connective tissue research 9 247-8
[39] Ganss B, Kim R H and Sodek J 1999 Bone sialoprotein Crit Rev Oral Biol Med 10 79-98
[40] Brubaker K D, Vessella R L, Brown L G and Corey E 2003 Prostate cancer expression of runt-
domain transcription factor Runx2, a key regulator of osteoblast differentiation and function
Prostate 56 13-22
[41] Lee K Y and Mooney D J 2012 Alginate: properties and biomedical applications Prog. Polym. Sci.
37 106-26
[42] Zhang Y, Yu Y, Dolati F and Ozbolat I T 2014 Effect of multiwall carbon nanotube reinforcement
on coaxially extruded cellular vascular conduits Mater Sci Eng C 39 126-33
[43] Dai X, Liu L, Ouyang J, Li X, Zhang X, Lan Q and Xu T 2017 Coaxial 3D bioprinting of self-
assembled multicellular heterogeneous tumor fibers Sci Rep 7 1457
[44] Lin R Z and Chang H Y 2008 Recent advances in three-dimensional multicellular spheroid culture
for biomedical research Biotechnol. J. 3 1172-84
[45] Lin R Z, Chou L F, Chien C C and Chang H Y 2006 Dynamic analysis of hepatoma spheroid
formation: roles of E-cadherin and beta1-integrin Cell Tissue Res. 324 411-22
[46] Sart S, Tsai A C, Li Y and Ma T 2014 Three-dimensional aggregates of mesenchymal stem cells:
cellular mechanisms, biological properties, and applications Tissue Eng. Part B Rev. 20 365-80
[47] Amos P J, Kapur S K, Stapor P C, Shang H, Bekiranov S, Khurgel M, Rodeheaver G T, Peirce S
M and Katz A J 2010 Human adipose-derived stromal cells accelerate diabetic wound healing:
impact of cell formulation and delivery Tissue Eng Part A 16 1595-606
[48] Gimble J M, Bunnell B A and Guilak F 2012 Human adipose-derived cells: an update on the
transition to clinical translation Regen Med 7 225-35
[49] Ozbolat I T and Yu Y 2013 Bioprinting toward organ fabrication: challenges and future trends
IEEE Trans. Biomed. Eng. 60 691-9
Page 34 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

35
[50] Lin R Z, Chu W C, Chiang C C, Lai C H and Chang H Y 2008 Magnetic reconstruction of three-
dimensional tissues from multicellular spheroids Tissue Eng. Part C Methods 14 197-205
[51] Glicklis R, Merchuk J C and Cohen S 2004 Modeling mass transfer in hepatocyte spheroids via
cell viability, spheroid size, and hepatocellular functions Biotechnology and bioengineering 86
672-80
[52] Logeart-Avramoglou D, Anagnostou F, Bizios R and Petite H 2005 Engineering bone: challenges
and obstacles J. Cell Mol. Med. 9 72-84
[53] Wang Z, Ying Z, Bosy-Westphal A, Zhang J, Schautz B, Later W, Heymsfield S B and Müller M
J 2010 Specific metabolic rates of major organs and tissues across adulthood: evaluation by
mechanistic model of resting energy expenditure– Am. J. Clin. Nutr. 92 1369-77
[54] Lin C Y, Kikuchi N and Hollister S J 2004 A novel method for biomaterial scaffold internal
architecture design to match bone elastic properties with desired porosity J. Biomech. 37 623-36
[55] Lin C Y, Hsiao C C, Chen P Q and Hollister S J 2004 Interbody fusion cage design using integrated
global layout and local microstructure topology optimization Spine 29 1747-54
[56] Jakab K, Norotte C, Marga F, Murphy K, Vunjak-Novakovic G and Forgacs G 2010 Tissue
engineering by self-assembly and bio-printing of living cells Biofabrication 2 022001
[57] Mironov V, Visconti R P, Kasyanov V, Forgacs G, Drake C J and Markwald R R 2009 Organ
printing: tissue spheroids as building blocks Biomaterials 30 2164-74
[58] Perez-Pomares J M and Foty R A 2006 Tissue fusion and cell sorting in embryonic development
and disease: biomedical implications Bioessays 28 809-21
[59] Jakab K, Neagu A, Mironov V, Markwald R R and Forgacs G 2004 Engineering biological
structures of prescribed shape using self-assembling multicellular systems Proc. Natl. Acad. Sci.
U.S.A. 101 2864-9
[60] Zhou S, Cui Z and Urban J P 2004 Factors influencing the oxygen concentration gradient from the
synovial surface of articular cartilage to the cartilage–bone interface: a modeling study Arthritis.
Rheum. 50 3915-24
[61] Egli R J, Bastian J D, Ganz R, Hofstetter W and Leunig M 2008 Hypoxic expansion promotes the
chondrogenic potential of articular chondrocytes J Orthop Res 26 977-85
[62] Coimbra I B, Jimenez S A, Hawkins D F, Piera-Velazquez S and Stokes D G 2004 Hypoxia
inducible factor-1 alpha expression in human normal and osteoarthritic chondrocytes1 Osteoarthr
Cartil 12 336-45
[63] Brucker P U, Izzo N J and Chu C R 2005 Tonic activation of hypoxia-inducible factor 1α in
avascular articular cartilage and implications for metabolic homeostasis Arthritis Rheum 52 3181-
91
[64] Robins J C, Akeno N, Mukherjee A, Dalal R R, Aronow B J, Koopman P and Clemens T L 2005
Hypoxia induces chondrocyte-specific gene expression in mesenchymal cells in association with
transcriptional activation of Sox9 Bone 37 313-22
[65] Hirao M, Tamai N, Tsumaki N, Yoshikawa H and Myoui A 2006 Oxygen tension regulates
chondrocyte differentiation and function during endochondral ossification J. Biol. Chem. 281
31079-92
[66] Karageorgiou V and Kaplan D 2005 Porosity of 3D biomaterial scaffolds and osteogenesis
Biomaterials 26 5474-91
[67] Amini A R, Laurencin C T and Nukavarapu S P 2012 Bone tissue engineering: recent advances
and challenges Crit. Rev. Biomed. Eng. 40
[68] Zeltinger J, Sherwood J K, Graham D A, Müeller R and Griffith L G 2001 Effect of pore size and
void fraction on cellular adhesion, proliferation, and matrix deposition Tissue Eng. 7 557-72
Page 35 of 36 AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript

36
[69] Keramaris N C, Kaptanis S, Lucy Moss H, Loppini M, Pneumaticos S and Maffulli N 2012
Endothelial progenitor cells (EPCs) and mesenchymal stem cells (MSCs) in bone healing Stem Cell
Res Ther 7 293-301
[70] Stevens C, Blake D, Merry P, Revell P and Levick J 1991 A comparative study by morphometry
of the microvasculature in normal and rheumatoid synovium Arthritis Rheum 34 1508-13
[71] Wilkinson L and Edwards J 1989 Microvascular distribution in normal human synovium J Anat
167 129
[72] Stoker M E, Gerdes A M and May J F 1982 Regional differences in capillary density and myocyte
size in the normal human heart Anat Rec 202 187-91
[73] Yoshii Y and Sugiyama K 1988 Intercapillary distance in the proliferating area of human glioma
Cancer Res 48 2938-41
[74] Kruyt M, de Bruijn J D, Wilson C, Oner F, Van Blitterswijk C, Verbout A and Dhert W 2003
Viable osteogenic cells are obligatory for tissue-engineered ectopic bone formation in goats Tissue
Eng. 9 327-36
[75] Hospodiuk M, Dey M, Ayan B, Sosnoski D, Moncal K K, Wu Y and Ozbolat I T 2018 Sprouting
angiogenesis in engineered pseudo islets Biofabrication 10 035003
Page 36 of 36AUTHOR SUBMITTED MANUSCRIPT - BF-101688.R1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript