Human Hepatocyte Morphology and Functions in a Multibore Fiber Bioreactor

Article (PDF Available)inMacromolecular Bioscience 7(5):671-80 · May 2007with42 Reads
DOI: 10.1002/mabi.200600281 · Source: PubMed
Abstract
The viability and liver specific functions of human hepatocytes in a multibore fiber bioreactor are reported. Human hepatocytes were cultured in the intraluminal compartment of the bioreactor. Human hepatocytes on the membranes maintained their round shape and showed focal adhesions as sites of interaction with the membrane surface. Cells in the bioreactor expressed liver specific functions, including synthetic and detoxification activity up to 14 d of culture. The results demonstrate that human hepatocytes cultured in the multibore fiber bioreactor are able to sustain the same in vivo liver functions in vitro.
Human Hepatocyte Morphology and Functions
in a Multibore Fiber Bioreactor
Loredana De Bartolo,
*
Sabrina Morelli, Maria Rende, Carla Campana,
Simona Salerno, Nino Quintiero, Enrico Drioli
Introduction
The engineering of human tissue analogues gives a
significant advantage to applied research as differentiated
human tissue is made accessible for e.g. drug testing, study
of disease and pathology and therapeutics molecules. On
the contrary to cell lines and conventional culture systems
that frequently induce qualitative changes in the cells,
miniaturized human in vitro tissue analogues provide the
chance to improve human predictability. The development
of both biomaterial and culture technique able to maintain
liver specific functions could help to realize a system which
comprehensively reproduces human liver functions. Iso-
lated hepatocytes are able to continue the full range of
known in vivo liver specific functions for only a short time
when they are maintained under standard in vitro cell
culture conditions.
[1]
Various culture methods have been developed to foster
retention of hepatocyte functions including coculture with
nonparenchymal cells,
[2]
culture in a sandwich collagen
gel,
[3]
in three-dimensional systems in spheroids
[4]
or a
variety of dynamic systems such as bioreactors.
[5–12]
Full Paper
The viability and liver specific functions of human hepatocytes in a multibore fiber bioreactor
are reported. Human hepatocytes were cultured in the intraluminal compartment of the
bioreactor. Human hepatocytes on the membranes maintained their round shape and showed
focal adhesions as sites of interaction with the
membrane surface. Cells in the bioreactor expressed
liver specific functions, including synthetic and
detoxification activity up to 14 d of culture. The
results demonstrate that human hepatocytes cul-
tured in the multibore fiber bioreacto r are able to
sustain the same in vivo liver functions in vitro.
L. De Bartolo, S. Morelli, M. Rende, C. Campana, S. Salerno,
N. Quintiero, E. Drioli
Institute on Membrane Technology, National Research Council of
Italy, ITM-CNR, c/o University of Calabria, Via P. Bucci, cubo 17/C,
Rende (CS), Italy
Fax: þ39 09 8440 2103; E-mail: l.debartolo@itm.cnr.it
M. Rende, E. Drioli
Department of Chemical Engineering and Materials, University of
Calabria, Via P. Bucci, cubo 45, Rende (CS), Italy
Macromol. Biosci. 2007, 7, 671–680
ß 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim DOI: 10.1002/mabi.200600281 671
Differently from static culture methods, the bioreactors
allow the culture of cells under tissue specific mechanical
forces such as pressure, shear stress and interstitial flow.
Furthermore, in this system a constant turnover of tissue
culture medium augments the gas and nutrient exchange,
which together with the complete fluid dynamics control
ensures the long-term maintenance of cell viability and
functions.
[13]
One of the most-used bioreactors for mammalian tissue
growth is the hollow fiber membrane bioreactor. This
bioreactor meets the main requirements for cell culture:
wide area for cell adhesion, oxygen and nutrient transfer,
removal of catabolites and protection from shear
stress.
[14,15]
Hollow fiber membranes may serve as scaff-
olding material in several ways: 1) initiate tissue
formation, 2) guide the microarchitecture of the develop-
ing tissue by providing histo-appropriate mechanical
forces, 3) confer mechanical strength. The multibore
membrane of modified poly(ethersulfone) (PESM) is
completely new, combining the advantage of having
7 compartments represented by 7 capillaries arranged in
one single fiber so the risk of membrane rupture can be
minimized by using this multibore membrane. Moreover
the 7-hole arrangement enables a high stability to the
foamy support structure that is located in-between the
capillaries and shows a permeability that is higher than
those of other membranes with similar pore size. An
additional advantage of the multibore membrane is the
high chemical and mechanical resistance. This type of fiber
could be useful in the culture and co-culture of cells in
several intraluminal compart-
ments and therefore in the
reconstruction of tissue envir-
onment in vitro. We used this
membrane in a bioreactor for
the first time for the culture of
human hepatocytes. Since an
efficient transport of meta-
bolites and nutrients is
required for in vitro mainte-
nance of hepatocyte viability
and functions, we evaluated
the hydraulic permeance and
the mass transfer of metabo-
lites through the membranes.
We investigated the morpho-
logical behavior of cells in the
multibore fiber bioreactor by
confocal laser scanning micro-
scopy (CLSM) and scanning
electron microscopy (SEM).
The liver specific functions of
human hepatocytes in the
membrane bioreactor have
been explored with time. In order to investigate the
biotransformation functions of human hepatocytes in
the multibore fiber bioreactor Hypericum perforatum
(St. John’s wort) extract and paracetamol were used.
H. perforatum is a herbal remedy widely used for the
treatment of depression.
[16]
Among its constituents
hyperforin is one of the major constituents responsible
for antidepressant activity and is biotransformed through
the cytochrome P450 enzyme system. Acetaminophen,
paracetamol or N-acetyl-p-aminophenol (APAP), is an
over-the-counter drug commonly used for its analgesic
and antipyretic properties that is metabolized by hepato-
cytes through glucuronidation and sulfonation reac-
tions.
[17]
Both components were used to assess the
performance of the human hepatocyte multibore fiber
bioreactor.
Experimental Part
Multibore Fiber Bioreactor
The multibore fiber membrane bioreactor consists of 3 multibore
membranes
1
of PESM (Inge AG, Greifenberg, Germany) in order to
increase hydrophilicity and strength of membrane. One single
multibore fiber contains 7 compartments in which cells are seeded.
The fibers are located in parallel within a glass housing and potted at
each end in order to establish two separate compartments: an
intraluminal compartment within the fiber, and an extraluminal
compartment or shell outside of the fibers. The two compartments
communicate through the pores in the fiber wall. The total inner
surface area of the multibore membranes used is 1.2 cm
2
.
L. De Bartolo et al.
Figure 1. Scheme of the multibore fiber bioreactor and the perfusion system.
672
Macromol. Biosci. 2007, 7, 671–680
ß 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim DOI: 10.1002/mab i.200600281
The bioreactor (volume: 18 mL) was connected to the perfusion
system consisting of a glass medium reservoir, tubing, oxy-
genator, a micro-peristaltic pump and a glass medium waste
(Figure 1).
The medium enters from the reservoir before the oxygenator
where it is heated and oxygenated, then the medium flows into
the membrane bioreactor. The oxygenated medium was con-
tinuously fed to the cells adhered on the membrane in the
bioreactor with a flow rate Q of 0.64 mL min
1
that was set on
the basis of average retention time. The stream Q
out
leaving the
bioreactor was monitored and recycled to ensure the continuous
mixing of the medium. Media samples were collected daily from
the outlet stream to evaluate the product formation and the
substrate clearance. Fresh medium was perfused in single-pass
and the stream leaving the bioreactor Q
out
was collected as waste
until approaching the steady state. When the system reached the
steady state, the stream leaving the bioreactor was recycled ( Q
r
)in
order to obtain the accumulation of products. The fluid dynamics
were characterized without the cells by tracer experiments using
William’s medium. The bioreactor was challenged by changing
the tracer concentration stepwise in the feed stream (C
in
) and
the outlet concentration (C
out
) was continuously monitored by
on-line spectrophotometer (UV Cord Pharmacia, Uppsala,
Sweden). The bioreactor fluid dynamics were characterized in
terms of the cumulative residence time distribution (RTD) to a step
inputs:
Fðt Þ¼C
out
=C
in
(1)
where t is the actual time.
[18]
The experimental mean residence time was calculated as a
mean value of the E curve:
F ¼
Z
t
0
E dt (2)
The theoretical mean residence time was calculated as:
t ¼ V=Q (3)
where Q is the perfusion flow rate and V is the volume of the
bioreactor.
[18]
Characterization of Morphological and
Physico-Chemical Properties of the Membrane
Dried multibore membranes were cut in cross-sections, mounted
with double-faced conductive adhesive tape, and sputter-coated
with gold. Treated membranes were analyzed by scanning
electron microscope (SEM, ESEM FEG QUANTA 200, FEI Company,
Oregon, USA). Representative images were acquired to obtain
information about cross-sectional structure and thickness, intra-
and extra-lumen morphology and diameters, and the shape and
size of membrane pores.
The hydrophobic/hydrophilic character of the investigated
membranes was estimated by contact angle technique. Water
contact angles were measured using the sessile drop method
at ambient temperature by CAM 200 contact angle meter
(KSV Instruments LTD, Helsinki, Finland), depositing the liquid
on the membrane surface using an automatic microsyringe.
Transport Characterization
The mass transport of metabolites through the membranes was
evaluated by a characterization apparatus. The multibore fiber
membranes were potted with an epoxy compound inside glass
modules (length: 20 cm, inner diameter: 1.5 cm) that allowed
access to both the intraluminal and extraluminal compartments.
A peristaltic pump (ISMATEC, General Control, Milan, Italy) circu-
lated the metabolite solution by pumping the fluid (feed) from the
reservoir into the inlet port of the module at flow rates of 0.48–
0.64 mL min
1
. Pressures were monitored at inlet and outlet
of the module by online manometers (Allemano, accuracy:
0.98 mbar). Inlet pressures were varied in order to obtain
transmembrane pressures (DP
TM
) from 4 to 70 mbar. The extralu-
minal flow (permeate) was measured continuously, and the
concentration of the metabolite permeating through the mem-
branes was monitored by an online UV-spectrophotometer (LKB
Uvicord SII, Pharmacia, Italy) and/or by assay with phenol in the
case of dextran. Metabolites include albumin (molecular weight
66 kD), immunoglobulin (IgG) (molecular weight 155 kD) and
dextran (molecular weight 473 kD), all purchased from Sigma,
Milan, Italy. Test solutions were prepared by dissolving separately
0.5 mg mL
1
for albumin, 0.05 mg mL
1
for IgG and 0.1 mg
mL
1
for dextran in phosphate buffer at pH ¼ 7.4.
The multibore fiber module was initially fed with water to
evaluate hydraulic permeance, then it was filled with metabolite
solutions for solute permeation measurements. For each type of
metabolite a new module was used. In order to ensure a high
reproducibility of permeance data, tests were carried out on four
modules and the average hydraulic permeance was reported.
Human Hepatocytes Culture
Primary human hepatocytes (Cambrex Bio Science, Milan, Italy)
isolated from the non-transplantable tissue of young single
donors were used for cell culture experiments. The purity of
isolated hepatocytes is 95% and non-parenchimal cells are present
in a very low percentage (5%). Cryopreserved human hepatocytes
were quickly thawed in a 37 8C water bath with gentle shaking.
Then, the cell suspension was transferred slowly into a tube
containing 25 mL of cold hepatocyte culture medium (HCM
TM
,
Cambrex Bio Science, Milan, Italy), and centrifuged at 50 g at 4 8C
for 3 min. The HCM
TM
is constituted of hepatocyte basal medium
(HBM
TM
) together with all the components provided in HCM
TM
bulletkit
1
(Cambrex Bio Science, Milan, Italy): epidermal growth
factors, insulin, ascorbic acid, transferrin, hydrocortisone 21-
hemosuccinate, bovine serum albumin-fat acid free 2% (BSA-FAF)
and gentamicin sulfate 50 mg mL
1
, amphotericin B 50 ng mL
1
.
The cell pellet was suspended in HCM
TM
and tested for the cell
viability by trypan blue exclusion, as reported elsewhere.
[19]
Human hepatocytes were seeded in the lumen of the multibore
fiber bioreactor, previously conditioned with HCM
TM
containing
BSA, to give a concentration of 1.3 10
5
cells cm
2
and incubated
for the first 24 h at 37 8C in a 5 vol.-% CO
2
/20 vol.-% O
2
atmosphere
Human Hepatocyte Morphology and Functions in a Multibore Fiber Bioreactor
Macromol. Biosci. 2007, 7, 671–680
ß 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mbs-journal.de 673
with 95% relative humidity in HCM
TM
containing 2% BSA.
Thereafter, the culture was continued under serum-free condi-
tions for the whole culture time up to 14 d.
The functions of the human hepatocytes cultured in the
bioreactor were investigated in terms of urea synthesis, secretion
of total proteins and biotransformation functions. In particular,
the biotransformation functions were investigated by adding
to culture medium paracetamol (0.3 10
3
M) and hyperforin
(1.8 mg mL
1
).
Biochemical Assays
Samples of the culture medium were collected from the bioreactor
in pre-chilled tubes and stored at 20 8C until assayed. The urea
concentration was assayed by the enzymatic urease method
(Sentinel, Milan, Italy).
The protein content in the samples was determined by protein
assay using bicinchoninic acid solution (Sigma, Milan, Italy) by
spectrophotometer analysis.
The statistical significance of the experimental results was
established according to the Unpaired Statistical Student’s t-test
and ANOVA test (p < 0.05).
HPLC Analysis of Paracetamol and Metabolites
The samples were HPLC analyzed using LiChrosorb RP-18 column
(5 mm), 250 4 mm (Merck, Darmstadt, Germany). The mobile
phase consisted of a potassium phosphate buffer (0.5
M)atpH¼
6.5 and methanol (95:5 v/v). The flow rate of eluent was
1mL min
1
with an analysis time of 30 min at a UV wavelength
of 242 nm. A polymeric column (Strata X, Chemtek Analytica,
Bologna, Italy) was used to treat the samples before injection
into the HPLC system. After conditioning with ethanol and water
(1:1, v/v), the samples were acidified with
2% H
3
PO
4
and diluted with potassium
phosphate buffer at pH ¼ 6 (1:1, v/v).
[20]
The recovery of paracetamol and its meta-
bolites was performed by using two dif-
ferent conditions: the first with 1 mL of
water with 5% of methanol and the second
with 1 mL of methanol. The recovered
fraction was dried under a gentle stream
of nitrogen and it was re-dissolved in 0.2 ml
of water before injecting into the HPLC
system.
HPLC Analysis of Hyperforin
The hyperforin was analyzed using a
LiChrosorb RP-18 column (5 mm), 250
4 mm at a UV wavelength of 270 nm.
Mobile phase A (acetonitrile) and mobile
phase B (2% H
3
PO
4
in water) were used in a
step gradient program at 1 mL min
1
: from
0 min to 10 min a linear change from 15% A
and 85% B to 20% A and 80% B; from 10 min
to 20 min , a linear change from 20% A and
80% B to 75% A and 25% B, and from 20 min
to 25 min a linear change from 75% A and
25% B to 80% A and 20% B. The samples were diluted with
acetonitrile (1:1, v/v) and after acidification with 0.1 mL of H
3
PO
4
,
vortexing for 1 min and then centrifuged at 6 000g for 10 min.
[21]
Cell Morphology
Sample Preparation for SEM
Specimens of cell cultures were prepared for SEM by fixation in
2.5% glutaraldehyde, pH ¼ 7.4 phosphate buffer, followed by
post-fixation in 1% osmium tetroxide and by progressive
dehydration in ethanol. The specimens were examined by SEM
after plating with gold under vacuum.
Hepatocyte Staining for LCSM
The morphological behavior of human hepatocytes cultured in the
membrane bioreactor was investigated after 48 h of culture by
LCSM. Hepatocytes cultured in the multibore PESM membranes
were washed with phosphate-buffered saline (PBS), fixed for
15 min in 3% para-formaldehyde in PBS at room temperature,
permeabilized for 5 min with 0.5 % Triton-X100 and saturated for
15 min with 2% normal goat serum (NGS). To visualize vinculin, a
specific monoclonal anti-vinculin antibody (Sigma, Milan, Italy),
diluted 1:50 in 1% NGS, was incubated for 30 min at room
temperature.
[22]
Then samples were washed twice in PBS and
incubated for 30 min with goat anti-mouse IgG tetramethylrho-
damine isothiocyanate (TRICT) conjugated (Sigma, Milan, Italy),
diluted 1:100 in PBS. To visualize the actin, the samples were
incubated in phalloidin conjugated with fluorescein isothiocya-
nate (FITC) (50 mg mL
1
) (Fluka, Milan, Italy) in PBS. Then samples
were washed twice in PBS and incubated for 20 min with
diamidino-2-phenylindole (DAPI) (Sigma, Milan, Italy) to visualize
L. De Bartolo et al.
Figure 2. Representative scanning electron micrographs of cross-section of multibore
PESM membrane.
674
Macromol. Biosci. 2007, 7, 671–680
ß 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim DOI: 10.1002/mab i.200600281
nucleic acid. Finally, the samples were washed, mounted and
viewed with Laser Confocal Scanning Biological Microscope
(Fluoview FV300, Olympus, Milan, Italy).
Results and Discussion
Properties of the PESM Multibore Membranes
The morphology of multibore membrane fibers is shown in
Figure 2. One fiber (diameter: 4.1 mm) consists of 7
capillaries with diameters of 960 mm and an inner layer
that is a very thin selective surface. The capillaries are
surrounded by a porous support structure that gives good
stability and mechanical resistance to the fiber. The PESM
membrane has a good surface wettability as the water
contact angle and water sorption measurements demon-
strated (Figure 3). The capillaries have a wettable surface,
in fact the water contact angle measured on this
membrane was 69.5 2.48 at t ¼ 0 s, and the increase of
water sorption was about 14% after 10 s. It has been shown
that the surface wettability of the membrane favors
interactions with the cells.
[23]
These properties together with hydraulic permeance are
not only important for the attachment of cells but also as
routes for the delivery of metabolites towards the cell mass
and the selective removal of secreted products in the
membrane bioreactor.
In this work, experiments aimed at evaluating the
hydraulic permeance of the membranes and the transport
of metabolites like BSA, IgG and dextran, under pressure
gradients, through PESM membranes were carried out. The
observed steady-state hydraulic permeance of multibore
membranes, calculated as the slope of the flux (J)
versus transmembrane pressure (DP
TM
) straight line,
was 1.067 L m
2
h mbar, with an R
2
value of 0.95
(Figure 4a). The structural characteristics of PESM mem-
branes were responsible for the high water permeance.
The foamy support structure in-between capillaries shows
a permeability that is 1 000 times higher than the filtration
surface so that an equal distribution throughout the whole
cross section of the fiber is ensured.
Permeation measurements of metabolite solutions
through multibore membranes are reported in Figure 4b.
The convective permeance calculated were: 0.845
L m
2
h mbar (R
2
¼ 0.997) for albumin solutions, 0.724
L m
2
h mbar (R
2
¼ 0.993) for IgG solutions and 0.382
L m
2
h mbar (R
2
¼ 0.93) for dextran solutions. The
average value of the transmembrane fluxes for the
investigated metabolites decreased with increasing
the MW of metabolites. As a result, the flux for BSA
solutions was significantly greater than that for IgG and
dextran.
Concentration profiles of metabolites permeating
through multibore PESM membrane, as monitored in
the extrafiber space and normalized on the feed solution
Human Hepatocyte Morphology and Functions in a Multibore Fiber Bioreactor
Figure 3. Time-related water contact angle (
) and water sorp-
tion (&) on multibore PESM membrane surface. The reported
values are the mean of 30 measurements of different droplets on
different surface regions of each sample standard deviation.
Figure 4. a) Hydraulic permeation measurements of multibore
PESM membranes. Experimental values (symbols) were averaged
on 10 measurements. The interpolation of experimental data is
reported as solid line. Slope: 1.067 L m
2
h mbar. b) Convective
permeation measurements of metabolite solutions through mul-
tibore PESM membranes: (~) BSA, (&) IgG, (*) dextran at
different transmembrane pressures (DP). Experimental values
(symbols) were averaged on 10 measurements. The interpolations
of experimental data are reported as solid line. Slopes: 0.845
L m
2
h mbar for albumin, 0.724 L m
2
h mbar for IgG and
0.382 L m
2
h mbar for dextran.
Macromol. Biosci. 2007, 7, 671–680
ß 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mbs-journal.de 675
concentration, increased with time and reached a plateau;
this behavior is exemplified in Figure 5. The same trend
was observed at different transmembrane pressure
differences. This characterization provided metabolite
and transport information specific to the membranes,
which should guarantee a sufficient level of mass transfer
for the survival of hepatocytes.
Fluid Dynamics of the Bioreactor
An important characteristic of the membrane bioreactor is
the fluid dynamics characterization under operating
conditions. The bioreactor was connected to a peristaltic
pump that allowed the feeding of nutrients and metabo-
lites and the continuous mixing of the medium inside the
bioreactor. Tracer experiments were performed at an
increasing perfusion flow rate Q
in
in order to investigate
the bioreactor fluid dynamics. As is shown in Figure 6, for
medium challenge, after a relatively short transience, the
medium concentration in the stream leaving the biore-
actor attained a steady state, then remained constant
throughout the duration of the experiment and in a time
closely depending on the flow rate Q
in
. Specifically,
the time needed to attain steady state decreased with
the increasing of the flow rate in the inlet stream. Diffusive
resistances in the intraluminal compartment of fiber
were observed at low flow rate (0.48 mL min
1
), in which
the fractional exit concentration values evidenced a
different concentration between the shell and intralum-
inal compartments. Increasing the flow rate to a value of
0.64 mL min
1
allowed having the same tracer concen-
tration in both compartments. In order to obtain the same
molecule distribution in the shell and lumen compartment
and an appreciable cell metabolic conversion, a perfusion
flow rate Q
in
of 0.64 mL min
1
was chosen. Under these
operating conditions the metabolite concentration in the
intraluminal compartment is uniform and equal to that in
the shell compartment of the bioreactor. By plotting the
RTD according to Equation (2) the mean residence time
was calculated corresponding to a value of 27.5 min. This
value is in agreement with that theoretically predicted
according Equation (3).
Cell Culture in the Multibore Fiber Bioreactor
In the membrane bioreactor we cultured human hepato-
cytes inside the multibore membranes, which is an
experimental system by which human-specific metabolic
information can be studied.
After 48 h of culture the cells on the membranes were
observed by confocal laser microscopy and appeared to be
reorganized in small aggregates. Actin (green staining) was
organized in the cytoskeleton ensuring a round shape to
the cells (Figure 7a–c). A peripheral distribution of vinculin
(red staining) of the hepatocytes evidences the localization
of focal adhesion (white arrows) and a central position of
nucleic acid (blue). The actin is more condensed and
exhibited a circumferential organization which reflects
balanced cell-membrane interactions and cell-cell con-
tacts.
[24,25]
No stress fibers were developed by the cells. As
the vinculin staining shows, cells developed focal adhesion
contacts, which are the sites of cell attachment to the
substratum that occurs through the link between the
extracellular matrix (ECM) and the molecules of actin
cytoskeleton. The focal adhesions consist of clustered
integrins that are employed, during the steps of cell
adhesion, in physical anchoring processes as well as in
signal transduction through the cell membrane that
influence the maintenance of liver specific functions.
[24,25]
A sufficient number of focal adhesions are visible
indicating the sites of interactions with the membrane
surface in different bores of the fiber (Figure 7a–c).
Hepatocytes maintained their morphology with time.
After 14 d of culture they appeared at the SEM analysis
L. De Bartolo et al.
Figure 5. Concentration profile appearance of IgG (~) and albu-
min (&) in the extrafiber side of multibore membrane module at
constant transmembrane pressure (15 mbar).
Figure 6. Cumulative RTD of the bioreactor following a step
change of William’s medium concentration in the perfusion
medium. Approach to a steady state of the fractional exit con-
centration for Q ¼ 0.48 mL min
1
(&), Q ¼ 0.64 mL min
1
(^);
(- - - -) tracer concentration stepwise.
676
Macromol. Biosci. 2007, 7, 671–680
ß 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim DOI: 10.1002/mab i.200600281
(Figure 8a, b) to be mostly round in shape and surrounded
by an ECM-like structure. The cells attached onto the
membrane established cell-cell contacts (Figure 8b). The
maintenance of this morphology is important for
the expression of the hepatocyte differentiated functions.
As demonstrated by other studies on hepatocytes cultured
in other hollow fiber membrane bioreactors, limitation
to mass transport of oxygen and nutrients affect
the distribution and viability of cells.
[13]
Interesting
approaches undertaken by some authors developing
mixing hollow fiber bioreactor with different fibers or
coaxial hollow fiber membrane reactors confirmed the
Human Hepatocyte Morphology and Functions in a Multibore Fiber Bioreactor
Figure 8. SEM images of human hepatocytes after 14 d of culture
in the lumen of multibore PESM membrane at different magni-
fications: (a) distribution of cells inside the fiber lumen; (b) tight
junctions.
Figure 7. Confocal images of human hepatocytes in multibore
fiber bioreactor after 48 h of culture by actin staining with
FITC-phalloidin (green) and vinculin staining with primary anti-
vinculin and secondary TRICT antibody conjugated (red) and by
nucleic acid staining with DAPI (blue). White arrows indicate focal
adhesions. (a), (b) and (c) Human hepatocytes in different bores of
the fibers.
Macromol. Biosci. 2007, 7, 671–680
ß 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mbs-journal.de 677
importance to enhance mass transfer.
[6,26]
In the multibore
fiber bioreactor, the high permeability of the fiber support
structure and the optimized fluid dynamics conditions
ensure an adequate mass transport throughout the whole
cross section of the fiber with the maintenance of cell
viability.
Functions of Human Hepatocytes in the Multibore
Fiber Bioreactor
One of the primary goals of this work was to sustain the
cell-specific functions of the human hepatocytes cultured
in the multibore fiber bioreactor. Specific metabolic
functions in terms of urea synthesis as well as secretion
of proteins were sustained for the investigated culture
time, demonstrating the maintenance of viability of the
hepatocytes cultured in the membrane bioreactor. Figure 9
shows the concentration profile of urea in the shell of the
bioreactor. The urea was produced by cells in the
bioreactor at high levels maintaining their specific activity
for 14 d of culture. The protein secretion by hepatocytes
was monitored by culturing cells in serum-
free medium (Figure 10). The concentration
of proteins in the shell was maintained for
most of the time at a value of around
150 mg mL
1
. These data take into account
all proteins secreted and released by cells in
the culture medium including the extra-
cellular matrix proteins.
The liver is the most important organ
concerning biotransformation of xenobio-
tics that occurs through the cytochrome
P450s, which is a class of constitutive and
inducible hemoprotein enzymes that meta-
bolize many endogenous substrates as well
as numerous xenobiotics and therapeutic
agents. CYP3A4 is the isoform present in
the highest quantity and is considered to
catalyze the transformation of a large
number of xenobiotics. The biotransformation
functions of human hepatocytes in the bioreactor
were investigated by using the H. perforatum
extract and paracetamol.
In particular the elimination of hyperforin that is
one of the most abundant lipophilic compounds
and an important active component of St. John’s
wort was evaluated. Human hepatocytes cultured
in the membrane bioreactor were able to biotrans-
form the hyperforin contained in the H. perforatum
present in the medium maintaining a good
metabolic activity for most of the culture time
(Figure 11). The hyperforin biotransformation
occurs via a hydroxylation pathway catalyzed by
CYP3A isoforms. This component contributes also
to the CYP3A4 inducing effect of the H. perforatum
extract. The hyperforin is a potent ligand for the pregnane
X receptor, an orphan nuclear receptor that regulates the
expression of the cytochrome P450 3A4 monooxy-
genase.
[27]
Treatment of human hepatocytes with hyper-
icum extract results in an induction of CYP3A4 expression,
which is involved in the oxidative metabolism of more
than 50% of all drugs including paracetamol.
Hepatocytes are also able to biotransform acetamino-
phen through the formation of the main metabolite
paracetamidophenyl-b-glucuronide, which is the product
of a glucuronidation reaction (Figure 12). The APAP is
primarily metabolized in the liver by the phase II routes,
sulfation and glucuronidation. In vivo the majority of the
paracetamol is metabolized by glucuronidation which
yields a relatively non-toxic metabolite. A small amount
(approximately 10% of paracetamol dose) of the drug
is metabolized by isoenzyme CYP1A2 and CYP2E1 into
N-acetylbenzoquinoneimine, which is extremely toxic
to liver tissue and it is immediately inactivated by
conjugation with glutathione.
[28]
In the human hepatocyte
L. De Bartolo et al.
Figure 9. Urea synthesis of human hepatocytes cultured in the multibore
fiber bioreactor for 14 d. The values are the mean of six experiments
standard deviation.
Figure 10. Protein secretion of human hepatocytes cultured in the multibore fiber
membrane bioreactor for 14 d. The values are the mean of six experiments
standard deviation.
678
Macromol. Biosci. 2007, 7, 671–680
ß 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim DOI: 10.1002/mab i.200600281
bioreactor we detected paracetamidophenyl-b-glucuronide
as a product of the glucuronidation of the drug. These results
demonstrated that hepatocytes maintain their biotransfor-
mation functions, which involve the activity of enzymes
catalyzing the conjugation reactions and the CYP1A2,
CYP2E1 and CYP3A, the most active isoforms involved in
the metabolism of hyperforin and acetaminophen.
Conclusion
In summary we report a new membrane bioreactor confi-
guration for the maintenance of human hepatocyte
specific functions. In contrast to the conventional fiber
the PESM multibore membranes have the arrangement
of seven capillaries in one fiber. The multibore fiber bio-
reactor ensures a sufficient oxygenation process, nutrient
feeding, end-product removal and distribution of fluid
molecules inside the cell environment. Human hepato-
cytes maintained their round shape and showed focal
adhesions as sites of interaction with the membrane
surface. Cells expressed liver specific functions, including
synthetic and detoxification activity for the investigated
culture time as demonstrated by the urea synthesis,
secretion of proteins, the elimination of hyperforin and the
formation of a paracetamol metabolite. These results
demonstrated the feasibility of the multibore fiber
membrane bioreactor that can be used as in vitro liver
tissue model for the evaluation of hepatic metabolic
transformation of compounds and therapeutic molecules
in a well controlled microenvironment.
Acknowledgements: The authors acknowledge the financial
support of the Italian Ministry of University and Research, MIUR,
for the grant to the FIRB research project RBNE012B2K and the
financial support of the European Commission for the grant to the
Livebiomat project STREP NMP3-CT-2005-013653.
Received: December 10, 2006; Revised: January 26, 2007;
Accepted: January 31, 2007; DOI: 10.1002/mabi.200600281
Keywords: drug biotransformation; human hepatocytes; mem-
brane bioreactor; morphology; synthetic functions
[1] M. N. Berry, A. M. Edwards, ‘‘The Hepatocyte Review’’, Kluwer
Academic Publishers, Dordrecht, Boston, London 2000, p. 365.
[2] M. T. Donato, J. V. Castell, M. J. Gomez-Lechon, Cell Biol.
Toxicol. 1991, 7, 1.
[3] J. C. Dunn, R. G. Tompkins, M. L. Yarmush, Biotechnol. Prog.
1991, 7, 237.
[4] F. J. Wu, J. R. Friend, C. C. Hsiao, M. J. Zilliox, W. J. Ko, F. B. Cerra,
W. S. Hu, Biotechnol. Bioeng. 1996, 50, 404.
[5] J. Rozga, F. Williams, M. S. Ro, D. F. Neuzil, T. D. Giorgio,
G. Backfisch, A. D. Moscioni, R. Hakim, A. A. Demetriou,
Hepatology 1993, 17, 258.
Human Hepatocyte Morphology and Functions in a Multibore Fiber Bioreactor
Figure 11. Hyperforin concentration in the inlet medium (&) and in outlet medium from bioreactor loaded with cells (&) in presence of
H. perforatum (1.8 mg mL
1
) added to the culture medium.
Figure 12. Time course of paracetamidophenyl-b-glucuronide for-
mation by human hepatocytes cultured in multibore fiber bio-
reactor for 14 d in the presence of paracetamol (0.3 10
3
M).
Macromol. Biosci. 2007, 7, 671–680
ß 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mbs-journal.de 679
[6] J. Gerlach, N. Schnoy, M. D. Smith, P. Neuhaus, Artif. Organs
1994, 18, 226.
[7] H. O. Jauregui, C. J. P. Mullon, D. Trenkler, S. Naik,
H. Santangini, P. Press, T. E. Muller, B. A. Solomon, Hepatology
1995, 21, 460.
[8] S. L. Nyberg, R. A. Shatford, M. W. Peshwa, J. G. White,
F. B. Cerra, W. S. Hu, Biotechnol. Bioeng. 1992, 41, 194.
[9] J. F. Patzer, Ann. N. Y. Acad. Sci. 2001, 944, 320.
[10] L. De Bartolo, G. Jarosch-Von Schweder, A. Haverich, A. Bader,
Biotechnol. Prog. 2000, 16, 102.
[11] M. J. Powers, D. M. Janigian, K. E. Wack, C. S. Baker, D. Beer
Stolz, L. G. Griffith, Tissue Eng. 2002, 8, 499.
[12] L. M. Flendrig, J. W. La Soe, G. G. A. Joerning, A. Steenbeck,
O. T. Karlsen, W. M. M. Bove´e, N. C. J. J. Ladiges, A. A. te Velde,
A. F. M. Chamuleau, J. Hepatol. 1997, 26, 1379.
[13] Y. Martin, P. Vermette, Biomaterials 2005, 26, 7481.
[14] K. E. Dionne, B. M. Cain, R. H. Li, W. J. Bell, E. J. Doherty,
D. H. Rein, M. J. Lysaght, F. T. Gentile, Biomaterials 1996, 17,
257.
[15] E. Curcio, L. De Bartolo, G. Barbieri, M. Rende, L. Giorno,
S. Morelli, E. Drioli, J. Biotechnol. 2005, 117, 309.
[16] Y. Cui, Y. W. A. Catharina, R. D. Beger, T. M. Heinze, L. Hu,
J. Leakey, Drug Metab. Dispos. 2004, 32, 28.
[17] D. A. Stoyanovsky, A. I. Cederbaum, Toxicol. Lett. 1999,
106, 23.
[18] O. Levenspiel, ‘‘Chemical Reaction Engineering’’, J. Wiley &
Sons, New York 1972.
[19] L. De Bartolo, S. Morelli, M. C. Gallo, C. Campana, G. Statti,
M. Rende, S. Salerno, E. Drioli, Biomaterials 2005, 26, 6625.
[20] M. Brolis, B. Gabetta, N. Fuzzati, R. Pace, F. Panzeri,
F. Peterlongo, J. Chromatogr. A 1998, 825, 9.
[21] M. V. Vertzoni, H. A. Archontaki, P. Galanopoulou, J. Pharmac.
Biomed. Anal. 2003, 32, 487.
[22] N. Krasteva, T. H. Groth, F. Fey-Lamprecht, G. Altankov,
J. Biomater. Sci., Polym. Ed. 2001, 12, 613.
[23] R. M. Ezzell, W. H. Goldmann, N. Wang, N. Parasharama,
D. E. Ingber, Exp. Cell Res. 1997, 231, 14.
[24] U. Hersel, C. Dahmen, H. Kessler, Biomaterials 2003, 24, 4385.
[25] L. B. Moore, B. Goodwuin, S. A. Jones, G. B. Wisley,
C. J. Serabjit-Singh, T. M. Willson, J. L. Collins, S. A. Kliewer,
PNAS 2000, 97, 7500.
[26] J. Macdonald, M. Grillo, O. Schmidlin, D. Tajiri, T. James, NMR
Biomed. 1998, 11, 55.
[27] A. Hollownia, J. J. Braszko, Biochem. Pharmacol. 2004, 68, 513.
[28] L. De Bartolo, S. Morelli, A. Bader, E. Drioli, Biomaterials 2002,
23, 2485.
L. De Bartolo et al.
680
Macromol. Biosci. 2007, 7, 671–680
ß 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim DOI: 10.1002/mab i.200600281
    • "The hyperforin was analysed using LiChrosorb RP-18 column (5 m), 250 mm × 4 mm (Merck, Milan, Italy) and UV wavelength of 270 nm. Mobile phase A (100% acetonitrile) and mobile phase B (H 3 PO 4 at the 2% in water) were used in a step gradient program at 1 mL/min: from 0 to 10 min a linear change from 15% A and 85% B to 20% A and 80% B; from 10 to 20 min, a linear change from 20% A and 80% B to 75% A and 25% B, and from 20 to 25 min a linear change from 75% A and 25% B to 80% A and 20% B. The samples were diluted with acetonitrile (1:1, v/v) and after acidified with 0.1 mL of H 3 PO 4 , vortexing for 1 min and then centrifuged at 6000 × g for 10 min (De Bartolo et al., 2007). Paracetamol and metabolites were analysed using LiChrosorb RP-18 column (5 m), 250 mm × 4 mm (Merck, Milan, Italy). "
    [Show abstract] [Hide abstract] ABSTRACT: In this study we developed a PEEK-WC hollow fiber (HF) membrane bioreactor for the maintenance of human peripheral lymphocytes as model system for the in vitro investigation of disease pathogenesis, chemical effects and individual drug sensitivity. Peripheral lymphocytes isolated from donor's human buffy coat were cultured in the shell compartment of the PEEK-WC-HF bioreactor and stimulated with PHA 5microg/mL for the first 48h of culture to enhance cytokine production and cell proliferation. Thereafter, cells were cultured in the presence of Hypericum perforatum (St. John's wort) in order to induce cytochrome P450s enzymes, CYP2E, involved in the biotransformation of endogenous molecules and exogenous compounds. The metabolic activity of cells with respect to glucose consumption and oxygen uptake was maintained for all the culture time without the addition of mitogen. Two cytokines IL-2 and IL-10, which are specific pattern of lymphocytes T helper 1 and T helper 2, respectively, were produced in the bioreactor up to 14 days of culture. Lymphocytes were also able to biotransform acetaminophen through the formation of the main metabolite paracetamidofenil-beta-glucuronide, which is the product of glucuronidation reaction, as a result of the Hypericum perforatum administration that induced the catalytic activity of the CYP2E1. These results demonstrated the usefulness of the bioreactor as the support system that reproduces physiological parameters such as a constant perfusion of medium, nutrients and oxygen maintaining the in vitro integrity of lymphocyte viability and functions.
    Full-text · Article · Nov 2007
  • [Show abstract] [Hide abstract] ABSTRACT: Primary mouse hepatocytes (between 10-250 cells per capsule) were immobilized within 1.0% w/v alginate microbeads. The textural properties of the alginate matrix were characterized and a full protocol based upon the measurement of the initial rate of Resazurin reduction was studied and standardized. Using this method, the decay rate constant (K(d) = 0.45 +/- 0.01 days(-1)) and the time in which the cell viability decreases in half (VI(50) = 37 +/- 0.7 h) have been measured. The method was compared with the analysis of cell vitality using Calcein A/M and Ethidium Homodimer I. Differences between the two methods were found in the viability profile due to the significant presence of double stained cells along the culture time. According to the author's knowledge, this is the first report of a systematic study and determination of the K(d) value for immobilized hepatocytes, incorporating a wide range of cell concentrations within the alginate matrix.
    Article · Jul 2009
  • [Show abstract] [Hide abstract] ABSTRACT: IntroductionMembranes for Human Liver ReconstructionHuman Lymphocyte Membrane BioreactorMembranes for Neuronal-Tissue ReconstructionConcluding RemarksReferences
    Full-text · Chapter · Jul 2009 · Journal of Microencapsulation
Show more

  • undefined · undefined
  • undefined · undefined
  • undefined · undefined