Placenta (2005), 26, 353e357

doi:10.1016/j.placenta.2004.07.002

SHORT COMMUNICATION

Dissolved Oxygen Concentration in Culture Medium: Assumptions

and Pitfalls

D. Newby, L. Marks and F. Lyall*

Maternal and Fetal Medicine Section, Institute of Medical Genetics, Yorkhill, Glasgow G3 8SJ, UK
Paper accepted 8 July 2004

Oxygen is a key factor in the regulation of cytotrophoblast differentiation, proliferation and invasion in early pregnancy.
Abnormalities in oxygen concentration have also been linked to a number of pregnancy disorders. Cell culture models have been
used to study the effect of oxygen on cytotrophoblast behaviour in vitro, however, there is often little or no validation of oxygen
levels in these cell culture systems. In this study, dissolved oxygen levels in culture medium maintained in standard culture
conditions (18% O2) measured 18%. On transfer to a low oxygen environment (2% O2), oxygen levels decreased to 6e8% after
4 h and reached 2% only after 24 h in culture. Culture medium pre-gassed with nitrogen to remove dissolved oxygen quickly
absorbed oxygen when exposed to ambient air during dispensing and required further incubation in a 2% oxygen environment
before dissolved oxygen levels equilibrated to 2%. Thus, cultured cells placed in a low oxygen environment would be exposed to
varying levels of oxygen before the desired level of oxygen exposure is reached. This study highlights the importance of validation
of oxygen levels and potential problems associated with in vitro studies on the regulatory effects of oxygen.
Placenta (2005), 26, 353e357 Ó 2004 Elsevier Ltd. All rights reserved.

INTRODUCTION others make no mention of oxygen levels in their culture

A number of important clinical conditions such as pre-
eclampsia, fetal growth restriction and spontaneous mis-
carriage have been linked to the placenta being exposed to
abnormal oxygen concentrations [1]. Women living at high
altitude are exposed to chronic hypoxia and they too have
increased risk of complications of pregnancy [2], underscoring
the involvement of oxygen. Cytotrophoblast proliferation and
differentiation appears, at least in part, to be regulated by
oxygen tension. These observations have been the driving
force for investigators to develop in vitro models to determine
the effects of different oxygen concentrations on cell culture
models such as primary trophoblast cultures and placental
explant culture [3e7]. One of the key issues in studies of this
nature is the actual oxygen levels which cultured cells are
exposed to during incubation in different oxygen environ-
ments. Some authors report specific methods to control the
oxygen environments to which the cells are exposed and/or
the measurement of oxygen in culture medium [3,5,7,8], while
* Corresponding author. Tel.: C44 141 201 0657; fax: C44 141 357
4277.
E-mail address: f.lyall@udcf.gla.ac.uk (F. Lyall).
0143e4004/$esee front matter
systems other than in the incubator environment [6,9,10]. The
aim of this study was to define the oxygen environments
encountered in our cell culture systems and to address some of
the issues and potential problems associated with studies of
this nature.
METHODS
Cell culture incubators
Studies were carried out using two Forma Scientific water-
jacketed incubators (Forma Scientific, Inc. USA), one set at
5% CO2 in air for standard culture conditions with the sensor
on the incubator reading approximately 18.2% O2 and one set
at 5% CO2/93% N2 with the oxygen sensor reading
approximately 2% O2 for low oxygen conditions. Oxygen
levels in each incubator environment were measured using
a Fyrite Gas Analyser.
Measurement of dissolved oxygen levels
There are many commercially available oxygen probes and
meters, of varying specifications, for the measurement of
dissolved oxygen in a liquid environment but few are ideal for
the measurement of dissolved oxygen in cell culture systems
Ó 2004 Elsevier Ltd. All rights reserved.
354
such as those used in trophoblast and explants cultures. A
Jenway 970 Portable Dissolved Oxygen/(C Meter and
Electrode (Jencons Scientific Ltd, UK) was chosen for the
measurement of dissolved oxygen in this study due to the cost
and the suitability of the technical specifications, in particular
the integral temperature compensation. Issues relating to the
measurement of oxygen in cell culture systems and the factors
that influenced the choice of oxygen probe used in this study
are discussed in more detail later in this communication.
The oxygen meter was calibrated, according to the
manufacturer’s instructions, to 21% oxygen in water saturated
air and to 0% oxygen using the zero salts supplied with the
meter. In order to satisfy the requirements of the oxygen
electrode, experiments were carried out using approximately
125 ml of M199 (supplemented with 10% fetal bovine serum
and 1% antimycotic/antibiotic solution) (GibcoBRL Life
Technologies, UK) in a T75 vented-lid cell culture flask. This
was necessary so that the depth of the culture medium in the
upright flask was sufficient to cover the temperature compen-
sating element of the electrode and allowed stirring of the
culture medium during measurement. The flask was placed flat
in the incubators to allow maximum surface area for gaseous
exchange. The surface area/volume in the T75 flask was
75 cm2/125 ml and this compares to 8 cm2/4 ml in 35 mm
culture dishes, 1.9 cm2/0.4e1 ml in 24 well plates and 0.6 cm2/
200 ml in the Millicell inserts used by many investigators for
trophoblast and explant culture experiments. All experiments
were performed in triplicate and repeated three times.
Experiment 1
Culture medium was warmed to 37 (C in T75 flasks in
standard culture conditions (5% CO2 in air). Dissolved oxygen
levels were measured and the flasks were then transferred to
low oxygen conditions at 37 (C for up to 24 h. Dissolved
oxygen levels in the culture medium were measured at
intervals during the incubation period. Separate flasks were
used for each time point to prevent the medium being affected
by atmospheric oxygen.
Experiment 2
Nitrogen gas (Cryoservice, UK) was bubbled through culture
medium (2e3 psi) (125 ml) in a T75 flask for 30 min to
eliminate dissolved oxygen in the culture medium. The
opening to the flask was sealed with parafilm through which
there was an inlet for the pipette delivering the nitrogen gas
and an outlet to allow gas to escape. The flask was immediately
transferred to a 2% oxygen environment at 37 (C. Dissolved
oxygen levels in the culture medium were measured after 1 h
and following overnight incubation.
Experiment 3
Nitrogen gas was bubbled through culture medium (125 ml) in
a T75 flask for 30 min to eliminate dissolved oxygen in the
culture medium, as described above. The culture medium was
Placenta (2005), Vol. 26
20
Dissolved oxygen (%)
15
10
5
0 5 10 15 20 25
Time (h)
Figure 1. Dissolved oxygen concentration (% O2) in test culture medium
incubated at 37 (C in hypoxic conditions for up to 24 h. Oxygen levels were
measured using a Jenway Model 970 Dissolved Oxygen Meter and Electrode
(Jencons Scientific Ltd, UK).
then dispensed into smaller volumes (5 ! 25 ml) to mimic
culture medium changes (total dispensing time 4 min) and
then returned to the T75 flask. Dissolved oxygen in the culture
medium was measured immediately in the 125 ml of culture
medium and the flask placed in the low oxygen incubator at
37 (C. Dissolved oxygen levels were measured at intervals up
to 16 h.
RESULTS
Oxygen levels, measured using a Fyrite Gas Analyser, were
18e20% in the standard incubator and 2e3% in the low
oxygen incubator.
Experiment 1
The dissolved oxygen concentration in culture medium
maintained in standard culture conditions measured approx-
imately 18% (pO2 140 mmHg). When culture medium was
transferred to the low oxygen incubator, oxygen levels
decreased gradually to measure 6e8% (46e61 mmHg) after
4 h, 7% (53 mmHg) after 8 h, 3% (23 mmHg) after 16 h and
2e3% (15e23 mmHg) after 24 h of incubation (Figure 1).
The pH was unaffected by the changes in oxygen concentra-
tion (pH 7.2e7.4).
Experiment 2
When nitrogen gas was bubbled through culture medium,
dissolved oxygen levels fell from 18e20% to approximately
1.5% (11 mmHg) after 15 min and to 0% after 30 min. When
pre-gassed culture medium (0%) was placed directly into the
incubator with the 2% oxygen environment, oxygen levels in
the culture medium equilibrated to approximately 2% after 1 h
and remained at 2e3% (15e23 mmHg) following overnight
incubation.
Experiment 3
When culture medium pre-gassed with nitrogen to eliminate
oxygen (0%) had been dispensed in ambient air, dissolved
oxygen levels increased to 8.5% (65 mmHg). Further in-
cubation of the cell culture medium in a hypoxic environment
Newby et al.: Dissolved Oxygen Concentration in Culture Medium
of 2% oxygen for approximately 16 h was required for
dissolved oxygen levels in the culture medium to decrease to
2% (15 mmHg).
DISCUSSION
There are many factors that must be given careful consider-
ation when measuring dissolved oxygen levels in a cell culture
system. Although many oxygen electrodes and meters are
available, not all are suitable or ideal for use with cell cultures.
Many probes are simply too large for the direct measurement
of oxygen in cell cultures maintained in, for example, 35 mm
dishes, 24-well plates or Millicell inserts. Smaller probes, such
as the ISO2 electrode (World Precision Instruments, Inc,
USA) with a 2 mm diameter tip, are available but often at
much greater expense. Temperature is of vital importance as
temperature effects the concentration of dissolved oxygen in
a liquid. The MI-730 electrode (Microelectrodes, Inc, USA) is
quoted by the manufacturers as having a change in probe
response of 2.2% per degree change in temperature. Thus,
unless an electrode has an integral temperature compensation
function (e.g. Jenway 970, Jencons Scientific Ltd, UK), care
must be taken to ensure that all measurements, including
calibration, should be carried out at the same temperature (i.e.
37 (C in the case of cultured cells). This is not always possible
for a variety of reasons depending on the method used for
calibration. For example, many manufacturers recommend
calibration in ambient air, while pre-gassing liquids, either
for calibration or to establish defined oxygen levels, results in
changes in liquid temperature that are difficult to control.
Trophoblast cells and villous explants are most commonly
cultured in a static environment. However, dissolved oxygen
measurements cannot accurately be carried out in a static sys-
tem due to oxygen starvation at the membrane of the electrode.
This would result in false low oxygen readings, while agitation
of cultures may dislodge cells or explants. Thus it would be
difficult to obtain accurate measurements in many cell culture
systems. Some authors have reported using a blood gas analyser
to measure dissolved oxygen in culture medium sampled from
cell cultures [7]. We initially used a Bayer 288 blood gas
analyser (Bayer, Germany) in our experiments. However, this
was not suitable as the analyser could only accurately measure
oxygen levels in whole blood (correspondence with manufac-
turer) and gave false readings with culture medium. Kilani
et al. [11] have reported their experience using a blood gas
analyser and the requirement for addition of whole blood to
the medium to improve the accuracy of readings.
In order to characterise oxygen levels within the oxygen
controlled cell culture environments used in our laboratory
in a way that best compensated for the factors discussed
above, we chose an oxygen electrode with automatic temper-
ature compensation and measured dissolved oxygen levels in
test culture medium maintained within the same environ-
ments as our cultured cells, rather than directly in cell cul-
tures themselves. Dissolved oxygen levels in culture medium
incubated in a hypoxic environment of 2% oxygen following
355
prior exposure to atmospheric oxygen or standard culture
conditions decreased gradually to 2% over an incubation
period of 24 h. During this time cultured cells would be
exposed to varying oxygen concentrations and degrees of
hypoxia at the cell/culture medium interface depending on
the length of time in culture and before the desired oxygen
concentration, in this case 2%, is reached. These experi-
ments were carried out using a large volume of culture
medium due to the requirements of the oxygen electrode in
use. Presumably, with the much smaller volumes of culture
medium used in our actual cell culture experiments the rate
of gaseous exchange between the culture medium and the
environment would be greater and hypoxic conditions would
be established more rapidly. However, problems associated
with culturing cells in a static environment, which may limit
the rate of gaseous exchange, still remain. A rocking
platform placed inside the incubator may be useful in
instances where cells are firmly attached to substrate to allow
diffusion. However, the authors should ensure that effects
observed are not due to flow/shear stress per se.
One way of circumventing the delay in establishing the
appropriate culture conditions may be to pre-gas the culture
medium, either with nitrogen to eliminate oxygen from the
medium or with a gas of the desired mix and oxygen level. As
we have shown, after pre-gassing with nitrogen to 0% oxygen,
dissolved oxygen levels will equilibrate to 2% and remain at
that level when maintained within a 2% oxygen environment.
However, manipulation and microscopic monitoring of
cultured cells is often carried out in ambient air, thus
exposing the culture medium and cells to atmospheric oxygen.
We have demonstrated that culture medium that has been
pre-gassed with nitrogen to eliminate oxygen quickly absorbs
atmospheric oxygen on exposure to ambient air. Further
incubation in a 2% oxygen environment for 16 h was required
for dissolved oxygen levels in the culture medium to
equilibrate back to 2%, thus limiting the advantage of pre-
gassing the culture medium. Gassing cultures directly to
avoid uptake of atmospheric oxygen also has inherent
difficulties particularly when millicell dishes are used. Such
difficulties include dislodging the cells or explants during
gassing, the cooling effect of the gas on the culture medium
and the time involved in gassing individual cultures in an
experiment. Furthermore, cultures maintained in a low
oxygen environment would be exposed to increased oxygen
levels during handling and during the associated time delay
before the cell culture environment equilibrates to appropriate
levels on return to the incubator, thus disrupting the exposure
of the cells to the required oxygen concentration. The only
way to avoid uptake of atmospheric oxygen would be to use
a chamber flushed with the appropriate gas mix to allow
manipulation of cultures in the same oxygen environment as
the incubator atmosphere [3] and then to be able to transfer
the dishes to the incubator without exposure to any
atmospheric air.
This work highlights the importance of, and difficulties
associated with, the validation of oxygen levels in vitro cell
356
culture studies investigating the effects of oxygen on
placental development and function. Our studies focused
on plastic dishes since this is what most workers use.
However, studies with glass dishes may, because of differ-
ences in oxygen retention, reveal other issues. The issues
highlighted are particularly important with regard to
reproducibility and the comparison of data between different
research laboratories.
Based on our experiences we propose a number of
recommendations when investigators perform culture experi-
ments using varying oxygen conditions. Clearly, different
groups will be asking different questions in their model
systems and therefore the recommendations can only be used
as a guideline. These have been sub-divided into two
categories; Group 1: what this paper’s results have provided
evidence to support and Group 2: what the authors’ ‘‘expert
opinion’’ adds to the recommendations.
Group 1 recommendations
1. The oxygen concentration in the incubator or chamber
should be validated using a suitable device such as the
Fyrite Gas Analyser used in the present study.
2. Authors should state the volume of medium per area of
culture dish and how long it took to reach the required
oxygen concentration before the experiment in question
started.
3. Authors should specify if the culture medium was
changed during the course of the experiment and how
this aected the oxygen concentration. Ideally medium
should be changed in a chamber flushed with the desired
oxygen concentration and with medium that has been
pre-bubbled to the required oxygen concentration. We do
not recommend bubbling directly into culture dishes
where cells are growing. A gloved box cabinet would be
required with the pipette and pre-gassed medium placed
inside the box before the box was flushed to obtain the
desired oxygen concentration. Pipettes or tips should be
flushed once with medium to remove air trapped inside
the pipette. At this point the lid of the culture dish can be
removed. However, authors should bare in mind that
even with the lid on air exchange will still be taking place,
albeit more slowly.
ACKNOWLEDGEMENT
We are grateful to Action Research and the British Heart Foundation for funding.
REFERENCES
[1] Kingdom JCP, Kaufmann P. Oxygen and placental villous development:
origins of fetal hypoxia. Placenta 1997;18:623e4.
[2] Zamudio S, Palmer SK, Regensteiner JG, Moore LG. High altitude and
hypertension during pregnancy. Am J Human Biol 1995;7:182e93.
[3] Genbacev O, Joslin R, Damsky CH, Polliotti BM, Fisher SJ. Hypoxia
alters early gestation human cytotrophoblast differentiation/invasion in
Placenta (2005), Vol. 26
4. If the culture dishes are to be removed for examination
under the microscope, authors should state how this
aected the oxygen concentration.
Group 2 recommendations
5. Measurement of the oxygen concentration in the culture
medium should be made using an oxygen electrode with a
tip size which can be fully immersed in the culture medium.
A number of companies including Jencons (http://
www.jenway.com) and Diamond General (http://www.
diamondgeneral.com) sell a range of electrodes with dif-
fering tip sizes. The smaller tip electrodes are more ex-
pensive. These must be calibrated before use according to
the manufacturer’s instructions. The electrodes must have
an integral temperature compensation function as temper-
ature affects the dissolved oxygen in the culture medium.
6. We do not recommend shaking the culture dish contain-
ing cells to measure oxygen concentration when villous
explants are cultured on matrigel since this may dislodge
them slightly or completely thus altering their invasive
potential. However, gentle shaking of a control dish which
contains medium but no explant will give the researcher
an idea of how the oxygen concentration has changed as
a result of the electrode being in a static environment.
7. Authors should specify the oxygen concentration of the
culture medium that the placental villous tissue was
collected into after delivery of the placenta. The culture
medium should be gassed to the appropriate level in the
laboratory in a universal container and then the sample
could be quickly dropped into it at collection. They
should state how long it took from collection to
preparation and the conditions used to dissect the piece
of tissue. Ideally all procedures should be performed in
the starting oxygen concentration of the experiment. We
feel it would be unrealistic and technically very dicult to
isolate single trophoblast cells in a low oxygen environ-
ment because of the many steps involved.
8. After consideration of the above manipulations in their
experiment it would be helpful to other readers if the
authors then defined whether, on balance, their culture
system represents a model of hypoxia or a model or
re-perfusion.
vitro and models the placental defects that occur in preeclampsia. J Clin
Invest 1996;97:540e50.
[4] Genbacev O, Zhou Y, Ludlow JW, Fisher SJ. Regulation of human
placental development by oxygen tension. Science 1997;277:1669e72.
[5] Watson AL, Skepper JN, Janiaux E, Burton GJ. Susceptibility of human
placenta syncytiotrophoblastic mitochondria to oxygen-mediated damage
in relation to gestational age. J Clin Endocrinol Metab 1998;83:
1697e705.
Newby et al.: Dissolved Oxygen Concentration in Culture Medium
[6] Caniggia I, Homa M, Winter J, Gassmann M, Lye SJ, Kuliszewski M,
et al. Hypoxia-inducible factor-1 mediates the biological effects of
oxygen on human trophoblast differentiation through TGFb3. J Clin
Invest 2000;105:577e87.
[7] Huppertz B, Kingdom J, Caniggia I, Desoye G, Black S, Korr H, et al.
Hypoxia favours necrotic versus apoptotic shedding of placental
syncytiotrophoblast into the maternal circulation. Placenta 2003;24:
181e90.
[8] Fitzpatrick TE, Graham CH. Stimulation of plasminogen activator
inhibitor-1 expression in immortalized human trophoblast cells cultured
under low levels of oxygen. Exp Cell Res 1998;245:155e62.
357
[9] Kudo Y, Boyd CAR, Sargent IL, Redman CWG. Hypoxia alters expression
and function of syncytin and its receptor during trophoblast cell fusion
of human placental BeWo cells: implications for impaired tropho-
blast syncytialisation in pre-eclampsia. Biochim Biophys Acta 2003;1638:
63e71.
[10] Nelson DM, Smith SD, Furesz TC, Sadovsky Y, Ganapathy V,
Parvin CA, et al. Hypoxia reduces expression and function of system
amino acid transporters in cultured term human trophoblasts. Am J
Physiol Cell Physiol 2003;284:C310e5.
[11] Kilani RT, MacKova M, Davidge ST, Guilbert LJ. Placenta 2003;24:
826e34.