Biomaterials 124 (2017) 126e156

Contents lists available at ScienceDirect

Biomaterials
journal homepage: www.elsevier.com/locate/biomaterials

Three-dimensional growth of human endothelial cells in an

automated cell culture experiment container during the SpaceX CRS-8
ISS space mission e The SPHEROIDS project
Jessica Pietsch a, Samuel Gass b, Stefano Nebuloni b, David Echegoyen a, Stefan Riwaldt a, e,
Christin Baake a, Johann Bauer c, Thomas J. Corydon e, Marcel Egli d, Manfred Infanger a,
Daniela Grimm a, e, *
a
Plastic, Aesthetic and Hand Surgery, Otto-von-Guericke University Clinic, Magdeburg, Germany
b
RUAG Space, RUAG Schweiz AG, Nyon, Switzerland
c
Max-Planck Institute for Biochemistry, Martinsried, Germany
d
Luzerne University of Applied Sciences and Arts - School of Engineering & Architecture, Hergiswil, Switzerland
e
Department of Biomedicine, Aarhus University, Aarhus, Denmark

a r t i c l e i n f o a b s t r a c t

Article history: Human endothelial cells (ECs) were sent to the International Space Station (ISS) to determine the impact
Received 27 January 2017 of microgravity on the formation of three-dimensional structures. For this project, an automatic
Accepted 4 February 2017 experiment unit (EU) was designed allowing cell culture in space. In order to enable a safe cell culture,
Available online 7 February 2017
cell nourishment and fixation after a pre-programmed timeframe, the materials used for construction of
the EUs were tested in regard to their biocompatibility. These tests revealed a high biocompatibility for
Keywords:
all parts of the EUs, which were in contact with the cells or the medium used. Most importantly, we
Automated experiment container
found polyether ether ketones for surrounding the incubation chamber, which kept cellular viability
Biocompatibility test
Science validation test
above 80% and allowed the cells to adhere as long as they were exposed to normal gravity. After
Experiment sequence test assembling the EU the ECs were cultured therein, where they showed good cell viability at least for 14
International space station days. In addition, the functionality of the automatic medium exchange, and fixation procedures were
Endothelial cells confirmed. Two days before launch, the ECs were cultured in the EUs, which were afterwards mounted
on the SpaceX CRS-8 rocket. 5 and 12 days after launch the cells were fixed. Subsequent analyses
revealed a scaffold-free formation of spheroids in space.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction cytoskeletal proteins, or cell adhesion and extracellular matrix

Endothelial cells (ECs) form the inner layer of blood vessels
(intima) in mammalian organisms and are in direct contact with
the blood. They play a key role in functional blood vessel
morphogenesis [1] and in the regulation of blood pressure [2].
Experiments performed with a microgravity (mg)-simulating
device, the Random Positioning Machine (RPM), revealed that ECs
begin to form 3D cell aggregates after short-term RPM-exposure [3-
5]. In addition, a part of the ECs of each culture grew as tube-like
structures [6]. Proteins significantly altered in their content and
involved in the formation of 3D aggregates include b1-integrin,
* Corresponding author. Institute of Biomedicine, Aarhus University, Wilhelm
 4, DK-8000 Aarhus C, Denmark.
Meyers Alle
E-mail address: dgg@biomed.au.dk (D. Grimm).
proteins, such as laminin, fibronectin, collagen type I and III, and
osteopontin [6-8]. In short-term studies, an increase in apoptosis in
adherently growing RPM-exposed cells has been observed [8].
Supplementation with vascular endothelial growth factor (VEGF)
attenuated the apoptotic effect of microgravity, but had no obvious
influence on the rate of 3D spheroid formation [7,8].
Based on this information, a spaceflight to the International Space
Station (ISS) with acronym SPHEROIDS was carried out in 2016 in
order to study the EC behavior under spaceflight conditions and to
prove the gravity dependency of ECs' tube formation. The re-
quirements encompassed experiment durations of 7 days and 14 days
as well as fixation of ECs with RNAlater and paraformaldehyde (3.4%
PFA). A nutrient solution exchange after about 7 days was a prereq-
uisite for the 14-day-experiment. The storage of the fixed samples
was intended to take place at 4
C (PFA) and 80
C (RNAlater).

http://dx.doi.org/10.1016/j.biomaterials.2017.02.005
0142-9612/© 2017 Elsevier Ltd. All rights reserved.
 J. Pietsch et al. / Biomaterials 124 (2017) 126e156 127

Since experiments on Earth (i.e., with open liquid handling,
plastic consumables or manpower) cannot be replicated in space,
the design of an automated experimental unit (EU) was necessi-
tated. The materials intended for the construction of the EU were
tested for their biocompatibility before and after construction of
the EU (Breadboard Model, Science Model) (Figs. 1 and 2). To pre-
pare the spaceflight, two additional tests, the Science Validation
Test and the Experiment Sequence Test, were conducted practicing
the sequence of the individual steps with all persons required to
elucidate the reliability of the EUs. In addition, the assembly of the
EUs was performed considering all quality check marks required by
NASA (National Aeronautics and Space Administration), i.e. the as-
sembly of the EUs, time of the upload to the ISS, the powered
experiment durations (7 days and 14 days maximal time until fix-
ation) in microgravity with subsequent cool storage of the fixed
samples and the download time back to Earth. As the EUs were
stored on board the ISS inside the ESA-designed KUBIK experiment
container, which can harbor, empower and automatically control
the functionality of the EUs, two KUBIK containers available on
Earth were used for the Experiment Sequence Test. After this
careful preparation, for the first time we showed 3D growth of ECs
in space in samples fixed after 5d of ECs-exposure to real mg.
2. Materials & methods
2.1. Cell culture and experimental layout
The human endothelial cell line EA.hy926 is a permanent cell
line, which was established by hybridizing human umbilical vein
endothelial cells (HUVEC) with the lung carcinoma cell line A549.

Fig. 1. Flowchart of the biocompatibility tests performed according to six different scenarios. Test procedures and schedules of the scenarios I to VI are shown in AeF. When a
T25 cm2 cell culture flask was required for a test (A,B,C,F), 106 EA.hy926 cells suspended in 15 ml corresponding complete culture medium were seeded into a flask one day prior to
the start of the experiment. Pre-incubation of the various pieces of material was done in 50 ml tubes filled with about 20 ml of either medium, RNAlater or PFA solution (B,C,D). Cells
seeded into four-well cell culture slides (BD Bioscience) were used for immunofluorescence staining (scenario IV, D). Fixed samples were stored at 4
C prior RNA isolation and
measurement of the amount of RNA (C) or staining (D). For tests according to scenario V (E), the material sample was placed in the middle of a petri dish and the EA.hy926 cells were
seeded directly on the surface. After inoculation, the petri dish was carefully filled with medium trying to avoid the excessive stirring so that the cells were not washed of the
material. After 3, 24 and 72 h of incubation cells adherently growing to the materials were detected by HE staining.
128 J. Pietsch et al. / Biomaterials 124 (2017) 126e156

Fig. 2. Overview of the experiment plan and requirements (A) and the stages during the EU development presented in the last part of the manuscript (B).

Although a hybrid, EA.hy926 cells exhibit characteristic endothelial
cell features [9,10]. The cells were cultivated in RPMI 1640 medium
containing 100 mM sodium pyruvate and 2 mM l-glutamine (Life
Technologies) and 10% FCS (Biochrom), 100 U/mL penicillin and
100 mg/mL streptomycin (both Biochrom). In addition, human
microvascular endothelial cells (HMVEC) derived from dermis
(Provitro) were used for experiments on the Random Positioning
Machine (see 2.11). For culturing the HMVEC, cell growth medium
e advanced, supplemented with a supplement-mix containing FCS
and antibiotics (Provitro) was used.
One day prior to the biocompatibility experiments, the
EA.hy026 cells were suspended in RPMI 1640 medium so that cells
could be seeded in cell culture flasks at appropriate cell numbers
(e.g. 106 cells in T25 cm2 cell culture flasks) for a preliminary 24-h-
period of adherence. Subsequently, the cells were used in the
following experimental scenarios: 1) biocompatibility tests (BT), 2)
breadboard model tests (BBM), 3) Science Validation Tests (SVT),
and 4) Experiment Sequence Tests (EST) as well as for the space-
flight experiment on the ISS.
2.2. Tests of materials used for building the EUs
The suitability of the materials used for building the EUs, which
should harbor, automatically feed and fix the cells during a
spaceflight, was proved in six different test scenarios. The tests
were performed according to a modified scheme developed during
preparation of a former spaceflight experiment with thyroid cancer
cells [11]. By annexin V measurements (see section 2.4) the viability
of EA.hy926 cells was tested which had been in direct contact with
material samples at 37
C and 5% CO2 for 15 days, during which the
RPMI 1640 medium was replaced with fresh medium at day 7
(Scenario I, Fig. 1A). In addition, the viability of EA.hy926 cells was
tested after cultivation in complete RPMI 1640 medium, which had
been pre-incubated in direct contact with material samples for 15
 J. Pietsch et al. / Biomaterials 124 (2017) 126e156 129

days at 37
C (Fig. 1B, Scenario II). Furthermore, the amount and
integrity of RNA was determined for cells, which were incubated for
24 h after seeding and subsequently fixed with RNAlater (Ambion,
Life Technologies). The RNAlater had been in direct contact with
material samples for 15 days at 37
C, and incubated for further 7
days (Fig. 1C, Scenario III). Also, immunofluorescence histology was
performed on cells which incubated 24 h after seeding and were
subsequently fixed with 3.4% paraformaldehyde fixative (PFA, Carl
Roth), which had been in direct contact with material samples for
15 days at 37
C (Fig. 1D, Scenario IV).
The capability of the ECs to adhere to the different materials was
determined, after they had been seeded on the pieces of materials
indicated and subsequently incubated at 37
C and 5% CO2 for 3, 24,
and 72 h (Fig. 1E, Scenario V). Finally, the viability of EA.hy926 cells
was tested, which had been cultivated at 18
C or 37
C and 5% CO2
in normal culture flasks for 7 days (Fig. 1F, Scenario VI).
Material samples arrived at the laboratory at three different
time points (Table 3, biocompatibility test (BT) Batch number 1e3).
Direct after arrival each material batch (BT Batch #1e3) was
independently tested. For each material batch, a control without
additional material (no. 57e59, Table 3) as well as with a mix of all
material samples obtained by one batch (no. 60e62, Table 3) was
tested according to the scenarios I-V. All in all, 56 materials were
tested (Table 3). Each material investigated had been sterilized
before either by autoclaving at 120
C for 20 min or by submerging
the materials in 70% ethanol for 10 min prior to the tests. Table 3
indicates the kind of sterilization procedure applied for each ma-
terial, respectively. The cells used for viability tests, RNA isolation or
staining were taken from one T25 cm2 cell culture flasks each,
respectively.
2.3. Biocompatibility and functionality tests of assembled EUs at
various time points
The Breadboard Model test was carried out with a first proto-
type of the EU (Fig. 3) in which all single modules were placed side
by side on a metal plate. The tests consisted of three parts: 1) cell
seeding tests, 2) RNAlater freezing and long-term storage tests, and

Fig. 3. Schematic drawings of the EU. The dimensions of one EU hardware designed by RUAG are 10  10  10 cm. Each houses two separate experiment subunits, E1 and E2 (A).
Of each subunit the CC, MC, SC, FC and pump are interconnected via MINSTAC tubing. The flow direction of the liquids is controlled by the pumps as well as by valves (V1-V6), which
contain stainless steel amongst other things, attached to the bottom of the MCs and SCs. A valve consists of three holes, of which two are accessible by MINSTAC tubing. Inside the
valve is a mechanic switch, which allows the fluids to either enter the respective chamber or the second tubing attached to the valve. The valve switch (valve connectors), as well as
the pump operation (pump connectors) are controlled via the automatic electronic board (AEB) on which the experiments sequence with medium exchange and fixation is
programmed. B and C show cross-sectional drafts of an E2 subunit. Neither the CCs, nor the MCs and SC are actively tempered. However, the AEB as well as a temperature logger on
the outside record the temperature during the mission.
130 J. Pietsch et al. / Biomaterials 124 (2017) 126e156

3) Experiment Sequence Tests. The cell viability in the cell seeding Table 1
tests was determined by annexin V-measurements. The pass/fail Antibodies (all from Cell Signaling) used for immunofluorescence staining.

criteria for the RNAlater freezing and long-term storage test were Antibody Dilution Host
the total RNA yield and the RNA Integrity Number (RIN). The per- Actin-b 1:300 M
formance of the hardware was further proved by the experiment Cytokeratin 1:50 M
sequence tests, which comprised of two experiments: a) the Sci- Tubulin-a 1:300 M
ence Validation Test (SVT) and b) the Experiment Sequence Test Tubulin-b 1:200 R
Vimentin 1:100 R
(EST). For both, the final flight model was used (Fig. 3). The time
Secondary antibody anti-M or anti-R 1:1000 both G
frame of these tests was a simulated flight (i.e., upload of the
hardware to ISS, experiment and download of the hardware). The
temperature profiles for these tests were either a worst-case sce- stock solution, Sigma-Aldrich). Pictures of the stained cells were
nario (SVT) or the expected scenario (EST). Post-test analyses taken with a fluorescence microscope (Olympus).
included RNA isolation and RNA concentration determination for
RNAlater-fixed cells, flow cytometry analysis of PFA-fixed cells and 2.7. Hematoxylin and eosin staining
the Multi-Analyte Profiling (MAP) technology for cytokine
measurements. The hematoxylin/eosin (HE) staining method is a widely used
histochemical technique [14]. Briefly, the HE staining protocol used
2.4. Annexin V measurement in this study was as follows: cells were fixed with 2% PFA (Carl Roth)
for 10 min. Following the fixation, the cells were incubated for 5 min
The annexin V measurements were carried out as previously with distilled deionized water, then incubated for 10 min with an
described [12] using the Annexin V-FITC Apoptosis Detection Kit I acidic Mayer's hematoxylin solution (Carl Roth) to stain cell nuclei
(BD Bioscience). The EA.hy926 cells cultured in T25 cm2 cell culture blue. The dye was removed by washing the sample three times with
flasks were detached using 1 mL of trypsin (Gibco, Life Technolo- tap water. The cytoplasm and other cell structures were stained with
gies) and transferred with PBS (Gibco) into a 50 mL tube. After eosin G (Carl Roth) for 10 min and then washed three times with tap
centrifugation for 5 min at 2500 rpm, the cell pellet was re- water. The stained cells were visually identified and photographed
suspended in 1 binding buffer (BB) at a final concentration of and if possible observed with an inverted microscope.
106 cells/mL, and 100 mL were transferred into two round-bottom The paraffin-embedded 3D cell aggregates from the PFA-fixed
tubes. The cells in one tube were incubated with 5 mL of annexin cell samples from the ISS were sectioned into thin cuts and
V-FITC protein and 5 mL of propidium iodide. The cells in the other mounted on object slides, deparaffinized and after rehydration
tube were left untreated (negative control, gating of unspecific stained as described above.
fluorescence signals). After incubation for 15 min in the dark, the
cell suspensions were brought up to 500 mL with BB and analyzed 2.8. Multi-analyte profiling technology
using a FACSCanto II (Becton Dickinson). The analysis of the flow
cytometry data was carried out with Flowing Software (Version The multi-analyte profiling (MAP) measurements were per-
2.5.1). Six-times 5000 cells were counted for one measurement. formed by the company Myriad RBM Inc., using the Human Cytokine
MAP A and B. This technique enables the detection of several cyto-
2.5. Flow cytometry tests for caspase-3 antigen kines in the supernatants of cell cultures as previously described [3].
The MAP procedure utilizes antibody complexes of two different
The staining procedure for flow cytometry was modified from a antibodies, which generate a fluorescence signal upon binding to the
previous protocol [13]. Briefly, PFA-fixed cells were used instead of target cytokine. This signal is used for quantification of the cytokine.
ethanol-fixed cells. After carefully scraping the cells off the bottom
of the cultivation chamber (CC) of the flight model (Fig. 3), they
were collected in 50-mL tubes and centrifuged for 5 min at 1500  g 2.9. RNA isolation and quality check
at room temperature. The pelleted cells were resuspended in PBS,
and 105 cells each were transferred into two round-bottom tubes The RNA was isolated using the RNeasy Mini Kit from Qiagen as
and incubated with 0.1% Triton X-100 in PBS for 10 min. The cells in described previously [15]. Briefly, EA.hy926 cells were fixed with
one tube were incubated for 90 min with the primary antibody RNAlater, scraped off the bottom of the cell culture flasks and
anti-caspase-3 (Cell Signaling; dilution: 1:640), while the other was transferred into 50 mL tubes with PBS. After pelleting the cells, the
incubated with PBS. The cells in both round-bottom tubes were supernatant was discarded and the cells lysed and homogenized
incubated with the secondary antibody (Cell Signaling; 1:1000) for
Table 2
60 min. The measurement of the single cells was performed with a Primers (50 e30 ) used for quantitative real-time PCR.
FACSCanto II (Becton Dickinson). The analysis of the flow cytometry
Gene Primer Sequence
data was carried out with Flowing Software (Version 2.5.1).
18S rRNA 18S-F GGAGCCTGCGGCTTAATTT
2.6. Immunofluorescence staining 18S-R CAACTAAGAACGGCCATGCA
ACTB ACTB-F TGCCGACAGGATGCAGAAG
ACTB-R GCCGATCCACACGGAGTACT
As described previously [13], the fixed cells were washed once ITGA10 ITGA10-F CGAAGACTGGTAGGGAAACTGTTTA
with PBS (Gibco) and permeablized for 10 min with 0.1% Triton X- ITGA10-R GCCGACTGAGGTTCTTTGCT
100 (Carl Roth) in PBS. After removal of the Triton X-100 solution, ITGB1 ITGB1-F GAAAACAGCGCATATCTGGAAATT
ITGB1-R CAGCCAATCAGTGATCCACAA
different primary antibodies (Table 1) diluted with PBS were
KRT8 KRT8-F GATCTCTGAGATGAACCGGAACA
incubated with the cells overnight at room temperature. The KRT8-R GCTCGGCATCTGCAATGG
following day, the cells were washed thrice with PBS and incubated TUBB TUBB-F CTGGACCGCATCTCTGTGTACTAC
with the FITC-conjugated anti-mouse secondary antibody (1:1000; TUBB-R GACCTGAGCGAACAGAGTCCAT
Cell Signaling) overnight at room temperature. The nuclei were VIM VIM-F TTCAGAGAGAGGAAGCCGAAAAC
VIM-R AGATTCCACTTTGCGTTCAAGGT
counterstained with propidium iodide (dilution 1:100 of a 1 mg/mL
 J. Pietsch et al. / Biomaterials 124 (2017) 126e156 131

Table 3
List of materials required for the construction of the experiment unit (EU) and tested separately before EU assembling. Batch# e material batch number depending on
material arrival at laboratory, E e sterilization by incubating with 70% ethanol for 10 min, A e sterilization by autoclaving material for 20 min at 120
C, CC e cultivation
chamber, MC e medium chamber, SC e supernatant chamber, n.a. e not available.

Sample# Material Batch# Sterili- Photo Weight [g] Sample dimension Function Compartment Used
zation [mm] in EU

1 stainless steel 3 E 0.1476 7d  1.5 Spring Valve Yes

2 Dimethyl-oxosilane 2 A ~0.25 5d  10 Lubricant on external MC þ SC Yes

surface of membrane

3 Ceramic 3 E 0.2100 8d_o  5d_i  2 Pump Bearing Pump Yes

0.3160 5d_o  1d_i  4

4 Ceramic 3 E 2.1962 12d  12 Case Part Fitting Pump Yes

5 stainless steel 3 E 0.1133 6d  1.2 Plunger head assembly Valve Yes

6 Polyisoprene 1 E 0.0302 8d  1.5 No

7 Silicone 1 E 0.0377 8d  1.5 No

8 Polyepoxides 2 A 0.0267 6d  0.5 Epoxy Resin in Pump Pump Yes

9 Polyepoxides 3 E 0.0267 6d  0.5 Epoxy Resin in Pump Pump Yes

10 Fluoropolymer 3 E 1.4106 8d  2 O-rings, Gaskets Valve Yes

Elastomer

11 Polyphenylene 3 E 0.6276 12  10  3 Plunger head assembly Valve No

sulfide

12 Reinforced 3 E 0.1028 8d  3 Pump Shaft Seal Pump Yes

Polytetraflu-
oroethylene

(continued on next page)

132 J. Pietsch et al. / Biomaterials 124 (2017) 126e156

Table 3 (continued )

Sample# Material Batch# Sterili- Photo Weight [g] Sample dimension Function Compartment Used
zation [mm] in EU

13 Reinforced 2 A 0.0128 8d  3 Pump Shaft Seal Pump Yes

Polytetrafluoro-
ethylene

14 stainless steel 1 E 11.6178 14d  10 No

15 Polychloro- 1 E 0.0192 1d  5 Yes

trifluoroethylene

17 Transluscent 2 A 0.3123 10  10  2.5 Flexible Chamber CC No

Silicone Wall Membranes
Elastomer 1

19 Transluscent 3 A 0.1857 10  10  2.5 Flexible Chamber CC No

Silicone Wall Membranes
Elastomer 2

22 Polyamide 1 A 0.0414 40  10  0.08 Filter Material CC No

23 Polyamide 2 A 0.0035 9  5  0.08 Filter Material CC No

24 Polyamide 2 A 0.0035 9  5  0.1 Filter Material CC No

25 Polybutylene 3 E 0.5318 10  10  3 Housing Valve Yes

terephthalate

27 Polyether ether 3 A 0.3258 10  10  2.5 Chamber Cradle and CC þ MC þ SC Yes

ketone Extremities, Adherent
Surface MINSTAC

29 Polyether ether 3 A 1.9459 25  10  2.5 Chamber Cradle and CC þ MC þ SC n.a.

ketone Extremities, Adherent
Surface

30 Polyether ether 1 E 1.9459 25  10  2.5 Chamber Cradle and CC þ MC þ SC n.a.

ketone Extremities, Adherent
Surface
 J. Pietsch et al. / Biomaterials 124 (2017) 126e156 133

Table 3 (continued )

Sample# Material Batch# Sterili- Photo Weight [g] Sample dimension Function Compartment Used
zation [mm] in EU

33 Polyethylene 2 E 0.0077 10  10  0.08 Filter Material CC Yes

terephthalate

34 Polyethylene 2 E 0.0077 10  10  0.08 Filter Material CC Yes

terephthalate

35 Fluoropolymer 1 E 0.3262 25d_o  23d_i  2 Piston O-ring MC þ SC No

Elastomer

36 Polyphthalamide 3 E 0.5593 10  10  3 Plunger head Valve Yes

assembly

38 Transluscent 3 A 0.1964 10  10  2.5 Flexible Chamber CC No

Silicone Elastomer 3 Wall Membranes

40 Transparent 2 A 0.2733 10  10  2.5 Flexible Chamber CC þ MC þ SC Yes

Silicone Elastomer Wall Membranes

42 White Silicone 2 A 0.1552 10  10  2.5 Flexible Chamber CC No

Elastomer Wall Membranes

43 White Silicone 1 E 0.4646 65  10  0.08 Flexible Chamber CC No

Elastomer Wall Membranes

45 Transluscent 2 A 0.3349 10  10  2.5 Flexible Chamber CC No

Silicone Elastomer 4 Wall Membranes

46 stainless steel 2 A 2.9902 7d  10 Filter Flange CC þ MC þ SC No

47 stainless steel 2 A 8.5667 11d  20 Pump Case Pump Yes

48 stainless steel 3 E 8.5667 11d  20 Pump Case Pump Yes

49 stainless steel 1 A 1.7882 6d  8 Chamber Extremities MC þ SC No

(continued on next page)

134 J. Pietsch et al. / Biomaterials 124 (2017) 126e156

Table 3 (continued )

Sample# Material Batch# Sterili- Photo Weight [g] Sample dimension Function Compartment Used
zation [mm] in EU

50 Polytetrafluoro- 3 A 0.0351 1.6d  10 Connection of Yes

ethylene CC þ MC þ SC

51 Polytetrafluoro- 1 E 0.0622 1.6d  10 Connection of Yes

ethylene CC þ MC þ SC

52 Tungsten Carbide 3 E 0.1867 1.2d  12 Pump Shaft Pump Yes

53 Fluoropolymer 2 A 0.4947 10  10  2.5 Filter Gasket CC No

Elastomer

54 Fluoropolymer 3 A 0.4424 10  10  2.5 Filter Gasket CC No

Elastomer

55 Fluoropolymer 2 A 0.0150 10d_o  9.5d_i  0.5 Pump O-rings Pump Yes

Elastomer

56 Fluoropolymer 3 E 0.0150 10d_o  9.5d_i  0.5 Pump O-rings Pump Yes

Elastomer

57 Control Batch#1 1
58 Control Batch#2 2
59 Control Batch#3 3
60 Mix Batch#1 1
61 Mix Batch#2 2
62 Mix Batch#3 3

with RLT buffer. The lysate was loaded onto an RNeasy spin column,
which binds the RNA. The lysate was pushed through the spin
column by centrifugation at maximum speed for 15 s. After several
washing steps, the RNA was eluted from the column by adding
30e50 mL of RNase-free water. The RNA was quantified with the
microplate reader Spectra Max M2 (Molecular Device).
Assessment of the RNA quality was performed using the Agilent
2100 Bioanalyzer (an automated electrophoresis system) in
conjunction with the RNA 6000 Nano LabChip kit (both Agilent
Technologies) according to the manufacturer's instructions. In brief,
1 ml of isolated RNA of the samples, solved in RNase-free water was
heat-denatured (70
C for 2 min) and applied to each well of the 12-
sample-wells of the chip. Following sample analysis values for the
ribosomal ratio and the integrity number (RIN) for each RNA were
determined using the 2100 Expert software version B.02.08.SI648.
2.10. Quantitative real-time PCR
Quantitative real-time PCR was carried out to determine the
expression levels of b-actin (ACTB), integrin b-1 (ITGA10), integrin-a
1 (ITGB1), keratin type II cytoskeletal 8 (KRT8), b-tubulin (TUBB) and
vimentin (VIM). 18S rRNA was used as a housekeeping gene to
normalize expression data (see Table 2).
The primers were designed using the Primer Express software
(Applied Biosystems, Life Technologies) with a Tm of ~60
C. All
primers were synthesized by TIB MOLBIOL as previously described
[15].
2.11. Random positioning machine
The Random Positioning Machine (RPM) developed to simulate
microgravity [16], is manufactured by Airbus Defense and Space
and used routinely in our laboratory [4]. For the experiments with
EA.hy926 and human microvascular endothelial cells (HMVEC) the
cells were seeded in T25 cm2 cell culture flasks (n ¼ 15, each cell
line). After they had formed sub-confluent monolayers, the culture
flasks were completely filled with medium (air bubble-free), sub-
sequently positioned on the inner frame of the RPM and the rota-
tion was started at a speed of 60
/s using the real random mode
(random speed and random direction) for 7 days. For the experi-
ment, the RPM was positioned in a standard CO2 incubator at 37
C
and 5% CO2. In parallel, further T25 cm2 cell culture flasks were
 J. Pietsch et al. / Biomaterials 124 (2017) 126e156 135

placed next to the RPM in the incubator as 1g-controls (n ¼ 15, each
cell line).
2.12. SpaceX CRS-8 spaceflight
The SpaceX CRS-8 space mission was a cargo resupply space-
flight to the ISS. The Falcon 9 rocket was launched on April 8, 2016,
at 20:43 UTC from the launch site at Kennedy Space Center, Cape
Canaveral, FL, USA. The end of mission was in May 2016.
3. Results
3.1. Construction of the EUs and their role in the spaceflight
experiment
3.1.1. Intended use of various EUs
The experiments should last either 7 or 14 days, while the cells
are incubated in culture medium in the presence or absence of
VEGF (10,000 ng/ml) (Fig. 2A). Cell feeding was required after 7
days in cultures planned for 14 days. For technical reasons the
temperature of 37
C could be achieved two days after loading the
rocket, when the EUs had been installed in the KUBIK. Until then,
the temperature was kept above 18
C. Fixed samples should be
conserved either at 4
C (PFA-fixed) or at 80
C (RNAlater-fixed)
until analysis.
For this planned spaceflight experiment, EUs were constructed
in 5 steps (Fig. 2B). Each EU includes the two identical subunits E1
and E2. Each subunit is composed of one cultivation chamber (CC),
one medium chamber (MC), one supernatant chamber (SC) and a
pump (HNP Mikrosysteme) to move the liquids within the system
(Fig. 3A). Both subunits share one fixative chamber (FC) and are
assembled on a metallic main support. Their MCs are either filled
with medium (E1) or with medium containing VEGF (E2). The FCs
of different EUs were either filled with PFA solution (subsequent

Table 4
Results of the biocompatibility test (scenario I-III). The percentages given in the columns from Annexin V scenario I and Annexin V scenario II indicate the vital/living cells
determined by the Annexin V measurements. The amount given for RNA isolated according to scenario III is the total amount of the whole sample. Controls are cell samples
#57-#59 cultured in plain medium at 3 different time points (Batch #1-#3). The results shown for sample #60-#62 originate from samples incubated with the mix of all
materials of one test material batch. Batch# e material batch number, depending on material arrival at the laboratory.

Sample # Batch# Annexin V scenario I Annexin V scenario II RNA scenario III

1 3 84% 85% 2220 ng

2 2 80% 62% 7149 ng
3 3 81% 88% 2940 ng
4 3 81% 90% 2250 ng
5 3 84% 87% 2700 ng
6 1 88% 86% 8650 ng
7 1 82% 84% 5000 ng
8 2 66% 71% 5841 ng
9 3 83% 86% 3570 ng
10 3 83% 62% 2640 ng
11 3 82% 87% 1410 ng
12 3 85% 82% 2400 ng
13 2 66% 74% 6626 ng
14 1 14% 83% 6620 ng
15 1 78% 82% 6600 ng
17 2 80% 64% 467 ng
19 3 82% 94% 2160 ng
22 1 15% 18% 4150 ng
23 2 82% 71% 5059 ng
24 2 79% 71% 5461 ng
25 3 80% 89% 2970 ng
27 3 85% 88% 2580 ng
29 3 84% 83% 1620 ng
30 1 82% 84% 5820 ng
33 2 88% 84% 4542 ng
34 2 75% 84% 4939 ng
35 1 13% 83% 11,820 ng
36 3 81% 90% 2100 ng
38 3 56% 88% 2070 ng
40 2 85% 77% 3450 ng
42 2 78% 54% 1260 ng
43 1 18% 84% 1320 ng
45 2 86% 78% 4508 ng
46 2 77% 64% 5321 ng
47 2 72% 85% 3098 ng
48 3 83% 90% 2100 ng
49 1 26% 81% 8200 ng
50 3 85% 87% 2520 ng
51 1 58% 83% 3530 ng
52 3 85% 90% 1650 ng
53 2 77% 64% 3723 ng
54 3 87% 89% 2250 ng
55 2 80% 72% 5872 ng
56 3 83% 87% 2460 ng
57 1 88% 79% 6980 ng
58 2 88% 80% 5701 ng
59 3 86% 91% 3570 ng
60 1 70% 74% 5910 ng
61 2 72% 73% 7528 ng
62 3 84% 77% 1740 ng
136 J. Pietsch et al. / Biomaterials 124 (2017) 126e156

structural analyses) or with RNAlater solution (subsequent molec-
ular analysis).
The materials used for the construction had to be tested for their
biocompatibility. On the completion of these biocompatibility tests
(BT), the prototypes and the flight models were tested for their
functionality.
3.1.2. Design and parts of EUs
Due to the intricate design of the EU (Fig. 3), biomaterials with
different physical and chemical characteristics were required for its
construction.
3.2. Biocompatibility tests
3.2.1. Influence of separated materials on cell viability and RNAlater
stability
All the different materials finally used in the construction of the
EUs were tested for eligibility in regard to biocompatibility and
functionality before and after assembling the EUs. Numbers, names,
photos, weight, size and functions of the materials tested separately
before the construction are listed in Table 3, which also shows the
respective belongings to a batch (column Batch#) and the kinds of
pre-sterilization (column Sterilization). These 56 samples were
tested according to the test scenarios described above. The tests
were performed in three series as indicated by the Batch#. This
way, some types of material were tested twice, either sterilized by
autoclaving or by soaking in ethanol.
The results obtained in tests performed according to scenario I,
II, III (see 2.2) are shown in Table 4, respectively, where the samples
were numbered as in Table 3. Viability tests of cells investigated
according to scenario II (Table 4, second column from right) clearly
indicated that most of the materials were inert to culture medium,
i.e. they did not deliver substances to the medium during 15 days at
37
C, which would have prevented its use for cell feeding. In these
cases, the annexin V measurements revealed viabilities above 80%.
Especially, the polyether ether ketones, which form the chamber
walls surrounding a cell suspension, do not harm the cell growth
(see samples #27, #29, #30). Dimethyl-oxosilane, Polyepoxides,
Fluoropolymer Elastomer, Reinforced Polytetrafluoroethylene,
Transparent Silicone Elastomer polluted the culture medium. But
the parts of the EU, which included these substances, barely con-
tacted the culture medium (see Table 3, the last two columns on the
right). All the other substances, which reduced the RPMI 1640
medium's capability to support the life of the cells, were not used
for constructing the EU (compare second column from right of
Table 4 with the most right column of Table 3).
In addition, the cell viability investigated according to scenario I
(Table 4, second column from left) proves that the optimal material
was selected for the EU. Again, most of the materials tested did not
attenuate cell growth, when a direct contact between these mate-
rials and the cells was established. The polyether ether ketones,
which form the chamber walls to which the cells adhere as long as
they grow under 1g-condition, supported cell growth (see samples
#27, #29, #30). Polyepoxides, Reinforced Polytetrafluoroethylene,
Polychlorotri-fluoroethylene, Polyethylene terephthalate, stainless
steel, and Polytetrafluoro-ethylene were included in parts of the EU,
where cell growth did not occur. No substances, which reduced cell
growth after contact, were used for the final construction. A ma-
terial was deemed biocompatible for the scenarios I and II, when at
least 77% of living cells were detected by applying the apoptosis
detection kit.
To elucidate whether the materials emit substances, which
inactivate RNAlater, RNA was isolated from cells treated according
to scenario III (RNAlater) after the ECs had been fixed with pre-
treated RNAlater and stored for 7 days prior to isolation of RNA.
The amount of the isolated RNA is given in Table 4 (most right
column). RNA integrity numbers (RIN) determined for each RNA
sample with the help of an established algorithm following sample
assessment in an automated electrophoresis system (Agilent 2100
Bioanalyzer) are shown in Fig. 4 for various materials. The RIN
values demonstrate that no significant RNA degradation in the fixed
cells occurred, except in samples fixed with RNAlater after its pre-
treatment with one lot of the polyether ether ketone (sample #29).
From most samples, more than 2000 ng of RNA could be isolated.

Fig. 4. Assessments of the RNA integrity of total RNA isolated in test scenario III. Purified, heat-denaturated RNA from fixed cells was analyzed using the Agilent 2100 Bio-
analyzer and RNA integrity was assessed exploiting an established algorithm as described in 2.9. The results of the electrophoresis (representing data from three separate chips)
together with the RNA Integrity Number (RIN) are shown for some samples. Exemplary, the electropherograms of two RNAs show the low degradation of the lowest RIN values. A
RNA ladder (L) is shown to the left. Positions of 28S and 18S ribosomal bands are indicated to the right. The sample numbers refer to Table 3.
 J. Pietsch et al. / Biomaterials 124 (2017) 126e156 137

Table 5a
Immunofluorescence staining of the biocompatibility test scenario IV. The given pictures for each antibody are in the following order: left) primary antibody, middle)
propidium iodide, right) merge of 1 þ 2 (left þ middle). The density of cells tested after receiving the first batch of materials (Batch # 1) was extremely high. Therefore, a lower
amount of cells were seeded out for testing the materials of Batch #2 and #3. Nevertheless, the pictures were taken with the high density, but with a lower magnification. The
scale bars represent 100 mm. Batch # - material batch number depending on material arrival at the laboratory.
138 J. Pietsch et al. / Biomaterials 124 (2017) 126e156
J. Pietsch et al. / Biomaterials 124 (2017) 126e156 139
140 J. Pietsch et al. / Biomaterials 124 (2017) 126e156
 J. Pietsch et al. / Biomaterials 124 (2017) 126e156
When the RNAlater solution had been pre-treated either with
Polyphenylene sulfide (sample #11), or with Transluscent Silicone
Elastomer 1 (sample #17), or with White Silicone Elastomer
(sample #42 and #43), or with Tungsten Carbide (sample #52), less
than 2000 ng of RNA was obtained. From these materials, only the
Tungsten Carbide was used to form the pump shaft as the RNA
quality was very good (Fig. 4) and the contact of the material with
the RNAlater solution will last only for a few seconds.
For the Polyether ether ketone #29 (Table 3), the amount as well
as the quality of isolated RNA was poor and therefore, this material
was not chosen for the construction of the EU. The Fluoropolymer
Elastomers tested in experiments #54 and #56, both yielded a RNA
amount of more than 2000 ng and had a RNA quality of RIN of 8.4. Of
both, only the material of #56 was used for the construction of the EU.
3.2.2. Influence of separated materials on PFA stability
PFA was also used to fix the ECs for subsequent investigation of
the cell structures. Whether or not PFA was able to fix the cells was
determined by using immunofluorescence staining (Tables 5a and
b). When a staining could be performed and structures inside the
cytoskeleton of cells could be distinguished, the material passed
the biocompatibility test scenario IV.
141
3.2.3. Influence of separated elements on cell adherence
Some of the materials were not only tested in the biocompati-
bility tests scenario IeIV, but also in an adherence test (scenario V).
The materials for these tests were selected because they will be in
direct contact with the living cells either as the growth area inside
the cultivation chamber of the EU to which cells should adhere or as
filters inside to which cells should not adhere. All materials tested
in the adherence test are listed in Table 6. The last column in Table 6
states whether the material fulfilled the requirements and was
used for the construction of the EU and indicates its function as well
as the compartment in which the material was used.
3.2.4. Influence of temperature on cell viability
The upload of the EUs to the ISS, i.e. the time expected from
handover of the EU until installation into the KUBIK incubator on
board the ISS, was considered to take less than 7 days. For this time,
the cells must be exposed to a lowered temperature of at least 18
C
for technical reasons. In order to investigate, whether the
EA.hy926 cells would survive a transient exposure to 18
C, the
viability of ECs at around 18
C (Fig. 5A) and at 37
C (Fig. 5B) was
monitored according to scenario VI (Fig. 1A) prior to the start of the
Breadboard Model tests. For both temperatures, a cell viability
142 J. Pietsch et al. / Biomaterials 124 (2017) 126e156

Table 5b
Immunofluorescence staining of the biocompatibility test scenario IV. The given pictures for each antibody are in the following order: left) primary antibody, middle)
propidium iodide, right) merge of 1 þ 2. The density of cells tested after receiving the first batch of materials (Batch # 1) was extremely high. Therefore, a lower number of cells
were seeded out for testing the materials of Batch #2 and #3. Nevertheless, the pictures were taken with the high density, but with a lower magnification. The scale bars
represent 100 mm. Batch # - material batch number depending on material arrival at the laboratory.
J. Pietsch et al. / Biomaterials 124 (2017) 126e156 143
144 J. Pietsch et al. / Biomaterials 124 (2017) 126e156
J. Pietsch et al. / Biomaterials 124 (2017) 126e156 145
146 J. Pietsch et al. / Biomaterials 124 (2017) 126e156
above 77% could be assured for 6 days. On day 7, the EC viability at
18
C dropped below the 77% pass criteria. These results were
considered acceptable because an upload lasting 7 days was highly
unlikely, and the temperature normally does not drop below 20
C
inside the transport capsule. Performing the viability test shown in
Fig. 5, in parallel, we determined the optimal number of cells for the
7- and 14-day-experiments, which supported a cell viability above
the pass criteria of 77% in the annexin V apoptosis detection test.
Seeding 5 x 105 cells into the CC for the 7-day-experiment and 4 x
105 cells into the CC for the 14-day-experiment was optimal (data
not shown).
3.3. Breadboard model tests
3.3.1. Suitability of breadboard models to harbor cells
The Breadboard Model was the first prototype of the EU in
which all modules were arranged side by side on metal plates. The
Breadboard Model tests encompassed tests of the whole experi-
mental sequences including cell cultivation, programmed fixation
of the samples for both the 7- and the 14-day-experiments as well
as the automatic exchange of the nutrient solution for the 14-day-
experiment. The viability of the ECs was confirmed by harvesting
living and not fixed cells after 7 or 14 days and using these cells for
an apoptosis detection kit measurement (annexin V-measure-
ment). A cell viability of more than 80% was achieved for both
experimental durations (Fig. 6A). In case of a malfunction during
the medium exchange at the 7th day, the outcome of 14-day-ex-
periments was also studied. Fig. 6B shows viabilities of cells after a
14-day-incubation period in the T25 cm2 cell culture flasks during
which a complete, a half and no medium exchange occurred. Even
in the case of only a half or no exchange of the nutrient solution
during the 14-day-experiment, the cell vitality surpassed the
criteria of 77% of living cells (Fig. 6B).
3.3.2. Suitability of the breadboard model to store RNAlater-fixed
cells and fluids
A further aspect of the Breadboard Model test was the storage of
fixed EC samples. According to the experimental requirements,
cells fixed with RNAlater were destined for freezing on board the
ISS, preferably at 80
C. To determine the optimal temperature for
storing, freezing tests at 20
C (Freezing Test 1) as well as 80
C
(Freezing Test 2) were performed with cells fixed inside the culti-
vation chamber (Table 7). In these cases, the cultivation chambers
were stored for 7 days at the respective temperature, and the RNA
was subsequently isolated. Unexpectedly, the RNA yield was rather
low and not uniform across all of the experiments (if not non-
existent), even if one considers the small growth area of 9 cm2 of
the CC, which allowed cell adherence and growth prior to launch,
and the pass criteria of 1000 ng were not reached. Due to these
results, additional tests were conducted over a 7-day-period at
ambient temperature (24
C, Freezing Test 3) as well as 4
C
(Freezing Test 4) with satisfactory results. Following these short-
term tests, storing tests were conducted over a period of
6 weeks at either 24
C (Table 7, Freezing long-term storage 1) or
4
C (Table 7, Freezing long-term storage 2). Overall, storage at 4
C
produced the optimum RNA yield, and this temperature was chosen
for storage on board the ISS.
In addition, a storage test at ambient temperature (Freezing
long-term storage 3) was conducted over three weeks to validate if
the samples would be damaged by a power loss of the cooler. The
collection of RNA surpassing the criteria (1000 ng) suggested that
power loss would not be a problem. Since the completed experi-
ments would have to be transported by an astronaut from the
KUBIK incubator to the cooler at 4
C and an astronaut's time is
tightly scheduled, we determined how quickly after fixation the
relocation to 4
C was required by storing the fixed sample at 37
C
for several days. The resultant RNA yields clearly indicated that
Table 6
List of materials tested in the adherence tests (scenario V). Batch # - material batch number, depending on material arrival at the laboratory; E e sterilization by incubating
with 70% ethanol for 10 min; A e sterilization by autoclaving material for 20 min at 120
C; CC e cultivation chamber. All materials were incubated with cells and potentially
attached cells were stained with HE. Pictures of the materials were taken after the staining.

Sample # Material Batch # Sterili-zation Photo Cell adherence Weight (g) Sample dimension [mm] Function Compart- Used
ment in EU

16 Translucent Silicone 2 A No 1.2288 20  20  2 Flexible Chamber CC No

Elastomer 1 Wall Membranes

18 Translucent Silicone 3 A No 1.1377 25  25  2 Flexible Chamber CC No

Elastomer 2 Wall Membranes

20 Polyamide 2 A Yes 0.0397 25  25  0.08 Filter Material CC No

21 Polyamide 2 A Yes 0.0383 25  25  0.08 Filter Material CC No

26 Polyether ether 3 A Yes 1.7209 25  25  2 Chamber Cradle and CC Yes

keton Extremities, Adherent
Surface

28 Polyether ether 2 E Yes 1.8673 25  25  2 Chamber Cradle and CC No

ketone Extremities, Adherent
Surface

31 Polyethylene 2 A Yes 0.0384 25  25  0.08 Filter Material CC Yes

terephthalate

32 Polyethylene 2 A Yes 0.0410 25  25  0.08 Filter Material CC No

terephthalate

37 Translucent Silicone 3 A No 1.0599 25  25  2 Flexible Chamber CC No

Elastomer 3 Wall Membranes

39 Transparent Silicone 2 A No 1.0001 20  20  2 Flexible Chamber CC Yes

Elastomer Wall Membranes

41 White Silicone 2 E No 1.1517 20  20  3 Flexible Chamber CC No

Elastomer Wall Membranes

44 Translucent Silicone 2 A No 1.2745 20  20  2 Flexible Chamber CC No

Elastomer 4 Wall Membranes
148 J. Pietsch et al. / Biomaterials 124 (2017) 126e156

Fig. 5. Cell viability test. The ECs were cultivated at 18
C (A) and 37
C (B) for 7 days and the cell viability was determined each day. The required minimum temperature for the
experiment was 18
C during upload to the ISS. The diagrams summarize two experiments. The first was conducted encompassing in total 5 days. The second experiment extended
the time frame up to 7 days (the second and the last three days are given). The percentage of living cells is represented by a dark green, of early apoptotic cells by a light green, of
late apoptotic cells by a yellow and of dead cells by a red line/symbol (A and B). The right y-axis indicates the respective temperature by blue lines.

relocation of the fixed cell samples was required as soon as
possible.
3.3.3. Storage of PFA-fixed cells
In order to study whether a delayed immunostaining of PFA-
fixed cells is possible, the capacity of fixation with PFA over time
as well as different storage temperatures were also tested. Prior to
fixation, PFA solutions were stored for more than 10 days at 37
C.
For the fixation tests, ECs were seeded out one day prior to fixation
on glass slides and were fixed with the various stored PFA solutions.
For the first tests, ECs fixed with PFA solutions of concentrations of
4, 2 and 1.55% were stored at 4
C for 1 day and 1, 2 and 4 weeks
(Fig. 7A). The results showed that longer storage times in the
presence of higher PFA concentration improved the outcome of the
subsequent immunofluorescence detection (Fig. 7A). To determine
how the storage temperature influenced the fixation quality, ECs
were fixed in 3.4% PFA (stored at 37
C for more than 10 days be-
forehand) and stored for 7 days and 6 weeks at either 4, 24 or 37
C.
The concentration of 3.4% PFA was the highest concentration
allowed by NASA to bring on board the ISS. The results demonstrate
that 4
C is the optimal storage temperature (Fig. 7B). However,
even in case of a malfunction of the cooling capacities on board the
ISS, the analyses of the PFA-fixed samples stored at higher tem-
peratures appeared to be feasible if not promising.
 J. Pietsch et al. / Biomaterials 124 (2017) 126e156 149

Fig. 6. Viability test of cells cultured in a Breadboard Model. The first experiment sequence test (i.e., a complete experiment with medium exchange) in the Breadboard Model (A)
confirmed the high viability of the cells (seeded at a number of 5 x 105 for 7-day-experiment, 4 x 105 for 14-day-experiment) for the 7- and 14-day-experiments. The data shows
living cells by the black bar, and dead cells including early and late apoptotic cells by the white column (A). In B, the living cells are shown again in black, but the early apoptotic, late
apoptotic and dead cells are represented with their own column. Although the living cell decrease only slightly, there is a distinguished shift in cells migrating to late apoptotic or
even dead cells the less medium is exchanged.

Table 7
Storage tests of cells fixed inside the cultivation chamber of the EU with RNAlater. The Breadboard Model tests also comprised a storage test of fixed cell samples inside the
cultivation chamber. The table shows the amount of isolated RNA from the RNAlater-fixed samples, which were stored at different temperatures ranging from 80 up to 37
C
for several periods of time. The test was named a freezing test due to the intended storage temperature. Samples were stored for 7 days during the freezing tests 1e4 and for 3
and 6 weeks during the freezing long-term storage 1e3. The storage tests 1e4 at 37
C lasted 2e7 days. *: amount of only 276 ng RNA due to spillage of cell suspension during
sample retrieval (Freezing test 4, 2nd experiment). Each experiment of one given freezing temperature and duration was repeated three times. Their outcomes are stated in
columns 3, 5, and 7 from left. If after two experiments the results were not promising, the third experiment was omitted.

Temp. [
C] RNA [ng] Temp. [
C] RNA [ng] Temp. [
C] RNA [ng] Time of storage

Freezing test 1 20 450 20 n.a. 20 10,500 7 days

Freezing test 2 80 30 80 n.a. 7 days
Freezing test 3 ambient 26,720 24 6,990 24 1,740 7 days
Freezing test 4 4 2,400 4 276* 4 2,388 7 days
Freezing long-term storage 1 ambient 5,850 23-25 1,045 6 weeks
Freezing long-term storage 2 4 6,360 4 9,500 4 6,920 6 weeks
Freezing long-term storage 3 23-25 5,730 23-25 5,700 3 weeks
Storage test 1 37 429 2 days
Storage test 2 37 122 3 days
Storage test 3 37 420 4 days
Storage test 4 37 492 7 days

3.4. Science validation test and experiment sequence test
After the Breadboard Model tests, sequence tests were con-
ducted with the second prototypes, the Science Model (SM), which
had the configuration of the final EU (Fig. 3). The main objective of
these Science Model tests was the optimization of the functionality
of the EU. For the Science Validation Test, four EUs of the second
prototype, the Science Models 1e4, were used (Table 8). Different
scientists assembled each one of them and quality assurance offi-
cers recorded each step. The assembly of the EUs took place over a
3-day-course: day 1) sterilization of all modules; day 2) steriliza-
tion of pumps and valves, filling and assembly of all EUs except
integration of cultivation chambers and day 3) cell seeding and
integration of cultivation chambers.
The results of operation of these models during the Science
Validation Test are shown in Table 9 and Figs. 8 and 9. According to
the research strategy developed for the spaceflight, the RNAlater-
fixed samples should be used for RNA isolation, the PFA-fixed
samples should be analyzed for 3D cell aggregation (multicellular
spheroids, MCS) and the supernatants should be tested for cytokine
production. In accordance with these planned analyses, we isolated
the RNA (Table 9) and performed quantitative RT Polymerase Chain
Reaction (qPCR) analyses (Fig. 8), used the PFA-fixed samples to
stain and detect cells with a flow cytometer (data not shown),
measured soluble factors in EC supernatants (Fig. 9). These results
confirmed the feasibility for such analyses after the spaceflight.
For the actual spaceflight preparation, eight EUs had to be
assembled with an additional four as a backup. An Experiment
Sequence Test was conducted to train the steps of the assemblies of
12 EUs in parallel (Fig. 10A). All lessons learnt during the Science
Validation Test were used to optimize the procedure. The general
sequence was the same as that during the Science Validation Test
although 3 days were required for ‘day 2’. All in all, the assembly
took place over the course of 5 days. The post-experiment analyses
were kept at a minimum and were performed to confirm that
samples could be collected from each EU (Fig. 10B, C) as the protein
expression of caspase-3 indicating apoptosis was below a critical
threshold and reasonable amounts of RNA could be isolated. The
amount of the RNA was reasonable, but differed, depending on the
EU, from which it was isolated. Therefore, a strict quantitative
relationship between cell seeding and amount of isolated RNA
could not be recognized.
3.5. Assembly for spaceflight and preliminary results of samples
fixed on board the ISS
In preparation for the spaceflight, ECs were re-cultured 17 days
prior to launch of the SpaceX CRS-8 rocket to the ISS (launch April 8,
2016). 12 EUs were assembled and programmed according to the
timeline established during the Experiment Sequence Test. The
assembly of the EUs started on April 1, and finished 42 h before
launch. The EUs were installed inside KUBIK on-board the ISS on
150 J. Pietsch et al. / Biomaterials 124 (2017) 126e156

Fig. 7. Storage conditions of PFA-fixed cells. ECs were seeded out 1 day prior to fixation. (A) To determine the fixation capacity of different PFA concentrations over time, ECs were
fixed with PFA concentrations of 1.55, 2 and 4% and stored at 4
C for up to 4 weeks. The staining revealed a better visibility of the cell structures with the higher PFA concentration,
when the fixation lasted up to 4 weeks. (B) As the mission profile envisaged the storage of the PFA fixative at 37
C prior to fixation and the storage of fixed samples for some weeks,
the optimal storage temperature of the fixed samples was determined. The optimal storage temperature was 4
C. Since the maximum concentration decreed by NASA was PFA of a
concentration of 3.4%, this concentration was used for the test shown in B. As this test showed good fixation results, a PFA solution of 3.4% was prepared for the spaceflight. The scale
bar represents 100 mm.
 J. Pietsch et al. / Biomaterials 124 (2017) 126e156 151

Table 8
Survey on the science validation test. Each type of EU (SM1-SM4) included cells incubated without (E1) and with (E2) VEGF. Table 9 and graphs of Figs. 8 and 9 depict the
results of the follow-up analyses. An error occurred during the assembly of SM4 and, therefore, it was not possible to perform the E1 experiment (indicated with an X in Figs. 8
and 9 for SM4). During the medium exchange of the 14-day-experiments (SM1 and SM3) a malfunction of the MC was detected after the experiment.
The cells of two EUs were fixed after 7 days with either PFA (SM2) or RNAlater (SM4), respectively.
The cells of two additional EUs were fixed after 14 days with either PFA (SM1) or RNAlater (SM3). Supernatants were harvested prior to fixation. Their content of distinct
biological factors was determined.

Model Person Experiment in H/W Duration [days] VEGF Fixative Type RNA isolation Quantitative rt PCR Flow cytometry Multi-Analyte profiling

SM1 1 E1 14 N PFA Yes Yes

E2 14 Y PFA
SM2 2 E1 7 N PFA Yes Yes
E2 7 Y PFA
SM3 3 E1 14 N RNAlater Yes Yes Yes
E2 14 Y RNAlater
SM4 4 E1 7 N RNAlater Yes Yes Yes
E2 7 Y RNAlater

Table 9
Results of the Science Validation Test, RNA isolation. The amount of RNA, isolated from cells grown in the SM3 and the SM4 are shown. After fixation, two populations of cells
were harvested from each experiment subunit. One population comprised cells, which remained adherent after fixation (ad), the other one included cells floating in the fixative
fluid (floating cells, FC). The ad cells were the cells of the main interest during the SVT.

Model Experiment in H/W Cells RNA content [ng]

SM3 E1 Ad 5,000
E1 FC 5,400
E2 Ad 5,400
E2 FC 6,000
SM4 E1 Ad 5,000
E2 FC 5,800

April 10, 2016. After completion of the mission, the EUs returned to
Earth on May 5, 2016 and were transported to our laboratories.
All samples were successfully retrieved, including the RNAlater-
and PFA-fixed cells and their supernatants. Due to a 2 days gap
between loading of the cells and the launch of the rocket, the actual
times in real microgravity (r-mg) were approximately 5 days for the
7-day- experiment and approximately 12 days for the 14-day-
experiment. After the spaceflight, 3D cell aggregates were detected
floating in the cultivation chamber with cells fixed after 5 days of r-
mg (Fig. 11B). The cell aggregates were embedded in paraffin,
sectioned, mounted on slides and subsequently stained with HE
(Fig. 11B). The related ground reference test did not exhibit 3D cell
aggregates. Only adherent cells could be found. These were likewise
collected as described above and stained with HE (Fig. 11A).
These results are in compliance with the findings of RPM-
experiments performed on Earth (Fig. 11C). With this
microgravity-simulating device, similar 3D cell aggregates were
found after 7 days of culturing EA.hy926 EC as well as with the
human microvascular endothelial cells (HMVEC). Beside the 3D
round aggregates called spheroids, tube-like structures could be
distinguished.
4. Discussion
Many biological experiments with plants, small animals or
mammalian cells have been conducted in space on the ISS [17-22].
There are some articles describing the concept of EUs for use on
board the ISS [23-25], however only a few publications have

Fig. 8. Results of the Science Validation Test, usability of RNA. RNA isolated from cells grown in the SM3 and the SM4 were suitable for use in qPCR. As expected for 1g-cultures,
the results obtained for the ad cell population were good, whereas for the FC (floating cells) cell population results were rather arbitrary.
152 J. Pietsch et al. / Biomaterials 124 (2017) 126e156

Fig. 9. Results of the Science Validation Test, released biofactors in the supernatant. The results of the Multi-Analyte Profiling analyses are given. While B displays only the
amount of VEGF in the supernatants, in A several detectable soluble factors released by the EC are outlined. The X in A and B indicates that no samples were collected due to
investigations of the filled medium chambers (MC) by the company RUAG for SM1 and SM3. The experiment without VEGF of SM4 could not be performed due to an assembly error,
also indicated by an X. The supernatants of SM2 and SM4 were collected in the supernatant chamber (SC). The supernatants of SM1 and SM3 were collected in MC after 7-day-
feeding and in SC after 14 days.

reported the biocompatibility test results of the materials in use
[11,20].
During the Sino-German SimBox/Shenzhou-8 space mission in
2011, thyroid cancer cells were successfully grown in an automatic
culturing system [11]. This container has already several times
proven to endure large physical forces and support cell survival. It
was capable to change media with medium (cell feeding) or fixative
fluid (cell fixation) at programmed time points automatically [11].
The hardware was used again in 2014 for thyroid cancer cells in
space (NanoRacks-CellBox-Thyroid Cancer) and re-flown with
SpaceX CRS-3 to the ISS (http://www.nasa.gov/mission_pages/
station/research/experiments/1648.html) [22,26-28].
The cells returning from space were suitable for post-flight
proteomics investigations [26,28]. Furthermore, the supernatants
could be used to determine secreted cytokines by MAP technology
[22,27].
Fig. 10. Results of the experiment sequence test. The main goal of the Experiment Sequence Test was the optimization of the timing of the pre-flight preparations. (A) During the
Experiment Sequence Test, 12 EUs were assembled of which eight were chosen to participate in the experiment. Each experimental configuration, i.e. duration (7 or 14 days) and
fixative (PFA or RNAlater), was prepared thrice. 8 EUs were needed for the test. The other four EUs were potential backups in case one EU did not pass the “health check”. The “health
check” verifies the functionality of the valves and pumps in the assembled EU. Post-EST analyses comprised only of a flow cytometry measurement of PFA-fixed samples (B), which
showed the protein content of Caspase-3 indicating the degree of apoptosis at the time point of fixation and the isolation of RNA from the RNAlater-fixed samples (C).
154 J. Pietsch et al. / Biomaterials 124 (2017) 126e156

Fig. 11. Histochemical staining of EC cultured on ground (A), on board the ISS (B) and native pictures of endothelial cells cultivated on the RPM (C). Adherent cells of the
ground reference test are shown in A. These monolayer cultures were likewise embedded in paraffin and stained with HE. The 3D cell aggregates formed and fixed after about 5 days
in space in the EUs. They were collected floating in PFA inside the CC and were subsequently embedded in paraffin and stained with HE (B). B clearly shows the aggregation of
several single cells together in a spheroid. Similar structures like the aggregates obtained on board the ISS (B) could be observed for ECs cultivated on the RPM (C) after 7 days using
phase contrast microscopy. For the two cell types EA.hy926 and HMVEC, a second type of a 3D cell aggregate, which was tube-like (pictures in the middle) could be observed.

The material selected initially for the construction of the EUs
consisted of biomaterials normally used for medical applications
and brought in direct contact with tissues inside a mammalian
body regardless of the kind of material (e.g., metal, ceramic, poly-
mers, collagen, or hydrogels) [26]. However, the biocompatibility of
a material for one application does not necessarily extend to other
cases [26,27]. Therefore, all potential materials were tested for their
biocompatibility with ECs according to a sequence of tests per-
formed for the development of an EU [11]. A variety of 18 different
materials (Table 3) were finally chosen for the construction of the
EU. This primary selection was of fundamental relevance for the
successful completion of the experiment in space.
The tests performed with the prototypes of the EU were con-
ducted while keeping in mind the possible mission profile (timeline
and temperature conditions). These tests were based on the re-
quirements for adjusting the storage conditions of the RNAlater-
fixed samples from 80
C to 4
C as well as the time allowed
between fixation and storage at 4
C.
The Science Validation Test was performed in a first collabora-
tion of all scientists and companies involved in the preparation
leading to the handover of the assembled EUs to NASA. The parallel
assembly of several EUs took place over the course of several days.
Therefore, the time allotted for each step had to be measured
precisely to finalize and optimize the timeline for the Experiment
Sequence Test. As a result, both the Science Validation and Exper-
iment Sequence Tests were successfully performed and the time-
line for the assembly during the mission preparation determined.
In retrospect, because the tests have generated a well-trained team
working in a harmonic and constructive atmosphere during the
launch preparation this space experiment was possible.
The ESA-SPHEROIDS project investigated human endothelial
cells staying 5 days and 12 days in space (EA.hy926 cell line). The
EA.hy926 cell line was chosen for several reasons: Firstly, these EC
have proven to be robust enough to grow under spaceflight con-
ditions and for a while even at 18
C for the temperature tests.
Secondly, these ECs bear endothelial cell markers like factor 8 and
Weibel-Palade bodies as well as tissue-specific organelles. They
show the characteristics of differentiated endothelial cell functions
[29]. Thirdly, they can form tubes without any scaffolds under
simulated microgravity like the human microvascular ECs as
demonstrated in Fig. 11C [29]. One goal of this spaceflight was to
focus on 3D growth of endothelial cells in space. The so-called
vascular ECs have special functions and are involved in the for-
mation of new blood vessels (angiogenesis). They form the
 J. Pietsch et al. / Biomaterials 124 (2017) 126e156
endothelium in vivo, which is a thin layer of endothelial cells that
lines the inner surface of vessels in the organism. It forms the
interface between the circulating blood in the vessel lumen and the
vessel wall. Earlier research had shown that weightlessness
generated by an RPM induces 3D growth of these cells within a
short time. They grew in form of tubular structures on this device
[8]. Long-term investigations showed that they included a lumen.
The walls of these tubular structures in vitro consisted similar to the
in vivo situation mainly of single-layered ECs tissue-engineered on
the RPM [6]. After the ESA-SPHEROIDS spaceflight, we could
demonstrate for the first time with the 5d-samples (Fig. 11B) that
endothelial cells exhibit 3D growth in space. This corresponds to
the results obtained by simulated microgravity conditions with
different EC cell types (Fig. 11C) [3,6]. A recent space experiment
investigating the impact of microgravity on HUVECs did not report
scaffold-free 3D growth in space [29]. This was also observed when
HUVECs were exposed to the RPM (unpublished results). Interest-
ingly, the authors concluded that microgravity affects the same
molecular machinery responsible for sensing alterations of flow
and generates a pro-oxidative environment that activates inflam-
matory responses, alters endothelial behavior, and promotes
senescence [29].
Tissue engineering in simulated and real microgravity supple-
ments each other [30]. It contributes to Biomedicine and Space
Medicine and is a new methodology for tissue engineering of 3D
tissues, such as cartilage [31,32], intima constructs [3,6] or cancer
spheroids [33,34]. These 3D tissues can be applied to study path-
ways for example involved in angiogenesis or apoptosis [33,34]. In
addition, they are useful for drug testing (cancer chemotherapy) in
cancer research. Multicellular spheroids can be applied in the
research fields of Pharmacology, Cancer Research, Toxicology and
Radiation Biology.
Cell cultures in space are also very important because we learn
about the behavior of cells exposed to microgravity. In the near
future, many humans will travel in space and missions are planned
to the Moon and Mars. Therefore, it is necessary to study the
cellular changes of different cell types when they are exposed to
microgravity conditions in space [35,36]. We have to know how
cells change and how we can avoid negative alterations. A well-
functioning automatic cell culture system is a prerequisite for
such a project. In 2017, the next ISS-project with human thyroid
cancer cells on board will be launched with a SpaceX mission.
5. Conclusion
The results of all completed tests enabled the construction of a
reliable EU, which flew successfully on board the ISS in April/May
2016 with the ESA experiment SPHEROIDS. After the retrieval of the
PFA-fixed samples from the EUs, the 3D cell aggregates were
detected, which assembled after the launch of the experiment until
their fixation approximately 5 days later. These results were com-
parable to those obtained with the RPM. The developed automatic
cell culturing system has proved to be suitable for space experi-
ments with human adherent cells and can be used for future cell
biological space experiments on the ISS as well as on unmanned
missions by the space scientists.
Author contribution
J.P. conducted and analyzed the experiments for the biocom-
patibility tests; J.P. conducted experiments and analyzed the results
of the BBM tests; J.P., S.G., S.N., and D.E. participated in the Science
Validation Test; J.P., S.G., S.N., D.E., S.R. and C.B. performed the
Experiment Sequence Test; T.J.C performed the RNA quality as-
sessments and analyzed the results. S.G. and S.N. were responsible
155
for the EU design and construction; J.B., T.J.C., M.E., M.I. and D.G.
designed the experimental requirements; J.P., J.B., T.J.C. and D.G.
wrote the manuscript.
Acknowledgments
The research was supported by the German Space Agency (DLR,
grants 50BW1124 and 50BW1524) and the European Space Agency
(ESA, grant ESA-AO-2004-006). The authors would like to thank
Jason Hatton, Nadine Boersma and Andrea Koehler from ESA as well
as Raimondo Fortezza from Telespazio for their support of the
SPHEROIDS project. We like to thank Christa Baumann (University
of Regensburg) for performing the histochemical stainings of the
spheroids in the space samples. We also like to thank Margrethe
Kjeldsen (Research Unit for Molecular Medicine, Department of
Clinical Medicine, Aarhus University, Denmark) for assisting the
RNA quality assessments. We also like to thank Jayashree Sahana
(Department of Biomedicine, Aarhus University, Denmark) for the
performance of the RPM experiments.
References
[1] S.P. Herbert, D.Y. Stainier, Molecular control of endothelial cell behaviour
during blood vessel morphogenesis, Nat. Rev. Mol. Cell Biol. 12 (9) (2011)
551e564.
[2] D.B. Cines, E.S. Pollak, C.A. Buck, J. Loscalzo, G.A. Zimmerman, R.P. McEver,
J.S. Pober, T.M. Wick, B.A. Konkle, B.S. Schwartz, E.S. Barnathan, K.R. McCrae,
B.A. Hug, A.M. Schmidt, D.M. Stern, Endothelial cells in physiology and in the
pathophysiology of vascular disorders, Blood 91 (10) (1998) 3527e3561.
[3] D. Grimm, J. Bauer, C. Ulbrich, K. Westphal, M. Wehland, M. Infanger,
G. Aleshcheva, J. Pietsch, M. Ghardi, M. Beck, H. El-Saghire, L. de Saint-Georges,
S. Baatout, Different responsiveness of endothelial cells to vascular endothelial
growth factor and basic fibroblast growth factor added to culture media under
gravity and simulated microgravity, Tissue Eng. Part A 16 (5) (2010)
1559e1573.
[4] X. Ma, M. Wehland, H. Schulz, K. Saar, N. Hubner, M. Infanger, J. Bauer,
D. Grimm, Genomic approach to identify factors that drive the formation of
three-dimensional structures by EA.hy926 endothelial cells, PLoS One 8 (5)
(2013) e64402.
[5] X. Ma, A. Sickmann, J. Pietsch, R. Wildgruber, G. Weber, M. Infanger, J. Bauer,
D. Grimm, Proteomic differences between microvascular endothelial cells and
the EA.hy926 cell line forming three-dimensional structures, Proteomics 14
(6) (2014) 689e698.
[6] D. Grimm, M. Infanger, K. Westphal, C. Ulbrich, J. Pietsch, P. Kossmehl,
S. Vadrucci, S. Baatout, B. Flick, M. Paul, J. Bauer, A delayed type of three-
dimensional growth of human endothelial cells under simulated weightless-
ness, Tissue Eng. Part A 15 (8) (2009) 2267e2275.
[7] M. Infanger, C. Ulbrich, S. Baatout, M. Wehland, R. Kreutz, J. Bauer, J. Grosse,
S. Vadrucci, A. Cogoli, H. Derradji, M. Neefs, S. Kusters, M. Spain, M. Paul,
D. Grimm, Modeled gravitational unloading induced downregulation of
endothelin-1 in human endothelial cells, J. Cell Biochem 101 (6) (2007)
1439e1455.
[8] M. Infanger, P. Kossmehl, M. Shakibaei, S. Baatout, A. Witzing, J. Grosse,
J. Bauer, A. Cogoli, S. Faramarzi, H. Derradji, M. Neefs, M. Paul, D. Grimm, In-
duction of three-dimensional assembly and increase in apoptosis of human
endothelial cells by simulated microgravity: impact of vascular endothelial
growth factor, Apoptosis 11 (5) (2006) 749e764.
[9] C.J. Edgell, C.C. McDonald, J.B. Graham, Permanent cell line expressing human
factor VIII-related antigen established by hybridization, Proc. Natl. Acad. Sci. U
S A 80 (12) (1983) 3734e3737.
[10] C.J. Edgell, J.E. Haizlip, C.R. Bagnell, J.P. Packenham, P. Harrison, B. Wilbourn,
V.J. Madden, Endothelium specific Weibel-Palade bodies in a continuous hu-
man cell line, EA.hy926, In Vitro Cell Dev. Biol. 26 (12) (1990) 1167e1172.
[11] J. Pietsch, X. Ma, M. Wehland, G. Aleshcheva, A. Schwarzwalder, J. Segerer,
M. Birlem, A. Horn, J. Bauer, M. Infanger, D. Grimm, Spheroid formation of
human thyroid cancer cells in an automated culturing system during the
Shenzhou-8 Space mission, Biomaterials 34 (31) (2013) 7694e7705.
[12] J. Grosse, D. Grimm, K. Westphal, C. Ulbrich, J. Moosbauer, F. Pohl, O. Koelbl,
M. Infanger, C. Eilles, J. Schoenberger, Radiolabeled annexin V for imaging
apoptosis in radiated human follicular thyroid carcinomaseis an individual-
ized protocol necessary? Nucl. Med. Biol. 36 (1) (2009) 89e98.
[13] J. Pietsch, A. Sickmann, G. Weber, J. Bauer, M. Egli, R. Wildgruber, M. Infanger,
D. Grimm, Metabolic enzyme diversity in different human thyroid cell lines
and their sensitivity to gravitational forces, Proteomics 12 (15-16) (2012)
2539e2546.
[14] G. Avwioro, Histochemical uses of haematoxylin - a review, J. Pharm. Clin. Sci.
1 (2011) 24e34.
[15] G. Aleshcheva, J. Sahana, X. Ma, J. Hauslage, R. Hemmersbach, M. Egli,
156 J. Pietsch et al. / Biomaterials 124 (2017) 126e156
M. Infanger, J. Bauer, D. Grimm, Changes in morphology, gene expression and
protein content in chondrocytes cultured on a random positioning machine,
PLoS One 8 (11) (2013) e79057.
[16] T. Hoson, S. Kamisaka, Y. Masuda, M. Yamashita, Changes in plant growth
processes under microgravity conditions simulated by a three-dimensional
clinostat, Bot. Mag. ¼ Shokubutsu-gaku-zasshi 105 (1) (1992) 53e70.
[17] J.Z. Kiss, P. Kumar, R.N. Bowman, M.K. Steele, M.T. Eodice, M.J. Correll,
R.E. Edelmann, Biocompatibility studies in preparation for a spaceflight
experiment on plant tropisms (TROPI), Adv. Space Res. 39 (7) (2007)
1154e1160.
[18] R. Cancedda, Y. Liu, A. Ruggiu, S. Tavella, R. Biticchi, D. Santucci, S. Schwartz,
P. Ciparelli, G. Falcetti, C. Tenconi, V. Cotronei, S. Pignataro, The mice drawer
system (MDS) experiment and the space endurance record-breaking mice,
PLoS One 7 (5) (2012) e32243.
[19] E.M. Martinez, M.C. Yoshida, T.L.T. Candelario, M. Hughes-Fulford, Spaceflight
and simulated microgravity cause a significant reduction of key gene
expression in early T-cell activation, Am. J. of Physiol. Regulat. Integr. Comp.
Physiol. 308 (6) (2015) R480eR488.
[20] A. Guignandon, C. Genty, L. Vico, M.H. Lafage-Proust, S. Palle, C. Alexandre,
Demonstration of feasibility of automated osteoblastic line culture in space
flight, Bone 20 (2) (1997) 109e116.
[21] X. Ma, J. Pietsch, M. Wehland, H. Schulz, K. Saar, N. Hubner, J. Bauer, M. Braun,
A. Schwarzwalder, J. Segerer, M. Birlem, A. Horn, R. Hemmersbach, K. Wasser,
J. Grosse, M. Infanger, D. Grimm, Differential gene expression profile and
altered cytokine secretion of thyroid cancer cells in space, Faseb J. 28 (2)
(2014) 813e835.
[22] S. Riwaldt, J. Bauer, J. Pietsch, M. Braun, J. Segerer, A. Schwarzwalder,
T.J. Corydon, M. Infanger, D. Grimm, The importance of Caveolin-1 as key-
regulator of three-dimensional growth in thyroid cancer cells cultured un-
der real and simulated microgravity conditions, Int. J. Mol. Sci. 16 (12) (2015)
28296e28310.
[23] E. Brinckmann, ESA hardware for plant research on the International Space
Station, Adv. Space Res. 36 (7) (2005) 1162e1166.
[24] K.H. Hasenstein, A. Sato, A. Cogoli, K. Takahashi, E. Brinckmann, Life Sciences:
microgravity Research IISpaceflight opportunities on the ISS for plant research
d the ESA perspective, Adv. Space Res. 24 (6) (1999) 779e788.
[25] V.D. Kern, S. Bhattacharya, R.N. Bowman, F.M. Donovan, C. Elland, T.F. Fahlen,
B. Girten, M. Kirven-Brooks, K. Lagel, G.B. Meeker, O. Santos, Life sciences
flight hardware development for the International Space Station, Adv. Space
Res. 27 (5) (2001) 1023e1030.
[26] J.D.B. Joyce, Y. Wong, Donald R. Peterson, Biomaterials: Principles and
Practices, CRC Press Taylor & Francis Group, Boca Raton, FL, USA, 2012.
[27] D.A.-B. Gottfried Schmalz, Biocompatiblity of Dental Materials, Springer Ver-
lag, Berlin, Germany, 2009.
[28] S. Riwaldt, J. Pietsch, A. Sickmann, J. Bauer, M. Braun, J. Segerer,
A. Schwarzwalder, G. Aleshcheva, T.J. Corydon, M. Infanger, D. Grimm, Iden-
tification of proteins involved in inhibition of spheroid formation under
microgravity, Proteomics 15 (17) (2015) 2945e2952.
[29] S. Versari, G. Longinotti, L. Barenghi, J.A. Maier, S. Bradamante, The challenging
environment on board the International Space Station affects endothelial cell
function by triggering oxidative stress through thioredoxin interacting pro-
tein overexpression: the ESA-SPHINX experiment, Faseb J. 27 (11) (2013)
4466e4475.
[30] D. Grimm, M. Wehland, J. Pietsch, G. Aleshcheva, P. Wise, J. van Loon,
C. Ulbrich, N.E. Magnusson, M. Infanger, J. Bauer, Growing tissues in real and
simulated microgravity: new methods for tissue engineering, Tissue Eng. Part
B Rev. 20 (6) (2014) 555e566.
[31] L.E. Freed, R. Langer, I. Martin, N.R. Pellis, G. Vunjak-Novakovic, Tissue engi-
neering of cartilage in space, Proc. Natl. Acad. Sci. U S A 94 (25) (1997)
13885e13890.
[32] V. Stamenkovic, G. Keller, D. Nesic, A. Cogoli, S.P. Grogan, Neocartilage for-
mation in 1 g, simulated, and microgravity environments: implications for
tissue engineering, Tissue Eng. Part A 16 (5) (2010) 1729e1736.
[33] S. Kopp, L. Slumstrup, T.J. Corydon, J. Sahana, G. Aleshcheva, T. Islam,
N.E. Magnusson, M. Wehland, J. Bauer, M. Infanger, D. Grimm, Identifications
of novel mechanisms in breast cancer cells involving duct-like multicellular
spheroid formation after exposure to the Random Positioning Machine, Sci.
Rep 6 (2016) 26887.
[34] S. Kopp, E. Warnke, M. Wehland, G. Aleshcheva, N.E. Magnusson,
R. Hemmersbach, T. Juhl Corydon, J. Bauer, M. Infanger, D. Grimm, Mecha-
nisms of three-dimensional growth of thyroid cells during long-term simu-
lated microgravity, Sci. Rep 5 (2015) 16691.
[35] T.J. Corydon, S. Kopp, M. Wehland, M. Braun, A. Schutte, T. Mayer, T. Hulsing,
H. Oltmann, B. Schmitz, R. Hemmersbach, D. Grimm, Alterations of the cyto-
skeleton in human cells in space proved by life-cell imaging, Sci. Rep 6 (2016)
20043.
[36] T.J. Corydon, V. Mann, L. Slumstrup, S. Kopp, J. Sahana, A.L. Askou,
N.E. Magnusson, D. Echegoyen, T. Bek, A. Sundaresan, S. Riwaldt, J. Bauer,
M. Infanger, D. Grimm, Reduced expression of cytoskeletal and extracellular
matrix genes in human adult retinal pigment epithelium cells exposed to
simulated microgravity, Cell Physiol. Biochem 40 (1-2) (2016) 1e17.