Advances in Bioscience and Biotechnology, 2010, 1, 398-408 ABB
doi:10.4236/abb.2010.15053 Published Online December 2010 (
Published Online December 2010 in SciRes.
Direct measurement of oxygen consumption rates from
attached and unattached cells in a reversibly sealed,
diffusionally isolated sample chamber
Timothy J. Strovas1,7,*, Sarah C. McQuaide1,7, Judy B. Anderson2,7, V ivek Nandakumar3,7,
Marina G. Kalyuzhnaya4,7, Lloyd W. Burgess5,7, Mark R. Holl3,7, Deirdre R. Meldrum3,7,
Mary E. Lidstrom4,6,7
1*Department of Electrical Engineering, University of Washington, Seattle, USA;
2Department of Pathology, University of Washington, Seattle, USA;
3Center for Ecogenomics, The Biodesign Institute, Arizona State University, Tempe, USA;
4Department of Microbiology, University of Washington, Seattle, USA;
5Department of Chemistry, University of Washington, Seattle, USA;
6Department of Chemical Engineering, University of Washington, Seattle, USA;
7Microscale Life Sciences Center (MLSC), a National Institute of Health (NIH) National Human Genome Research Institute
(NHGRI) Center of Excellence in Genomic Science (CEGS), Los Angeles, USA.
Received 4 August 2010; revised 25 August 2010; accepted 3 September 2010.
Oxygen consumption is a fundamental component of
metabolic networks, mitochondrial function, and
global carbon cycling. To date there is no method
available that allows for replicate measurements on
attached and unattached biological samples without
compensation for extraneous oxygen leaking into the
system. Here we present the Respiratory Detection
System, which is compatible with virtually any bio-
logical sample. The RDS can be used to measure
oxygen uptake in microliter-scale volumes with a re-
versibly sealed sample chamber, which contains a
porphyrin-based oxygen sensor. With the RDS, one
can maintain a diffusional seal for up to three hours,
allowing for the direct measurement of respiratory
function of samples with fast or slow metabolic
rates. The ability to easily measure oxygen uptake in
small volumes with small populations or dilute sam-
ples has implications in cell biology, environmental
biology, and clinical diagnostics.
Keywords: Respirometry; Oxygen Consumption Rate;
Reversible Diffusional Seal; Pt-Porphyrin
The measurement of respiration rates can be a versatile
tool for the diagnosis of cellular and metabolic state. The
ability to directly determine the oxygen uptake rate of
cellular samples and monitor how the rates change in
response to stimuli has broad implications in furthering
the understanding of a wide array of biological systems
from single cells to complex ecosystems. For example,
respiration rates can be indicators of subtle phenotypes
and metabolic states in bacterial systems [1].
Environmentally, oxygen-linked respiration is the main
sink of organic matter in nature and can be considered a
fundamental component of global element cycling [2].
Differences in oxygen uptake within complex natural
communities could lend insight into the utilization of
energy sources in the environment as well as into pri-
mary production [2]. The availability of a simple tech-
nique that could provide direct and precise oxygen
measurements in marine or freshwater samples with low
cell counts would significantly improve our knowledge
of aquatic ecosystem functioning at local or global
In addition to applications for environmental studies,
oxygen uptake rates are important for eukaryotic cell
biology. Oxygen uptake rates can be directly related to
mitochondrial function and have been implicated in the
activation of eukaryotic cells and the progression of dis-
ease states [3-5]. The ability to predict the progression of
cell death pathways would lend insight into the diagnosis,
treatment, and prevention of inflammatory diseases such
as infection, stroke, heart disease, diabetes, and cancer
[6-8]. Potential applications for the direct measurement
of respiration rates would be as a gauge of cellular health
and metabolic state for stress response, toxicity studies,
T. J. Strovas et al. / Advances in Bioscience and Biotechnology 1 (2010) 398-408
Copyright © 2010 SciRes. ABB
and the assessment of the effects on tissues from extra-
neous perturbation.
To date the most common method for measuring res-
piration rates has been electrochemical measurements
via traditional Clark electrodes [1,9-11]. However, the
drawbacks of electrodes include low sensitivities, signal
drift, and consumption of oxygen by the electrode itself
[10,11]. Other recent methods have entailed scanning
electrochemical microscopy (SECM) and utilization of
nanobead sensors attached to the outer membrane of
cells [12,13]. However, these techniques have been em-
ployed in openly diffusible environments and only indi-
cate oxygen flux near the cell membrane or the oxygen
Optical methods for the measurement of oxygen con-
sumption rates (OCRs) have become both widely used
and accessible in the last several years [14,15]. While
palladium and ruthenium based phosphores are used for
oxygen sensing, Pt-phosphores have become the most
relied upon oxygen sensor dye and have been used rou-
tinely in research and industry for over twenty years
[1,5,15-21]. In addition, calibration, signal acquisition,
and signal processing have become routine in the utiliza-
tion of phosphorescent oxygen sensors. Many of the
approaches employing optical sensors are based on fab-
ricated well arrays or on standard well plate formats
commercially available [5,18,20,22-26]. In general with
well and plate based systems, the sensor is deposited into
the bottom of a well or positioned above the well for
oxygen concentration measurements. Plate-based sys-
tems have the advantage of low costs associated with the
production of mold injected plastic plates compared to
the fabrication of well systems. However, the main
drawback of plate-based sensing platforms is that the
materials used in these techniques are themselves per-
meable to oxygen and require extensive characterization
to separate the actual respiration rates from the leakage
or diffusion of extraneous oxygen into the sample
chamber [20,22,25,27-28]. In addition, difficulties with
reproducible sealing of these devices present additional
variability. The result is that the signal acquired for op-
tical probes is that of flux of the oxygen near the sensor
and not of a direct measurement of the OCR. Thus far,
the only known method for directly measuring the OCR
of cells has been an optical technique employed in a dif-
fusionally sealed borosilicate chamber [1,18].
Herein, we describe the development of the Respira-
tory Detection System (RDS), which is capable of di-
rectly measuring the respiration rate of biological sam-
ples in a simple and reproducible manner. The RDS is
unique in its ability to measure OCR’s of dilute biologi-
cal samples in a reversibly sealed sample chamber that
can maintain a diffusional seal for over two hours.
Commercially-available slides as well as fabricated glass
well systems were tested as components of the RDS, and
the latter was shown to provide rapid and reproducible
measurement of OCRs for attached mammalian cell
lines, low optical density bacterial cultures, and envi-
ronmental samples.
2.1. Cellular Samples and Growth Conditions
Methylobacterium extorquens AM1 was grown in batch
culture at 28 in minimal salts medium using either
0.3% (vol/vol) methanol or 0.4% succinate as a carbon
source [29].
2.2. Lake Washington Sediment
Lake Washington sediment core samples were collected
as described previously [30]. Sediment cores were
transported to the laboratory on ice and stored at 9.
The samples of the top layer of the sediment were ana-
lyzed within 24 hours after collection.
2.3. Mammalian Cell Cultures
Human lung carcinoma cells (A549) and human colon
carcinoma cells (HCT-116) were gifts from the Rabino-
vitch Lab (Department of Pathology, University of
Washington). The mouse macrophage-like cell line, RAW
267.4 was a gift from the Cookson Lab (Department of
Laboratory Medicine, University of Washington).
A549 and HCT cells were cultured under identical
conditions as previously described with the exception
that 5% Fetal Cone III (HyClone, Logan, UT) was used
[18]. RAW 267.4 cells were cultured as previously de-
scribed [18].
Normal human bronchial epithelial (NHBE) cells
were purchased from Lonza (Allendale, NJ). Cell cul-
tures were grown in bronchial cell growth media
(BEGM) with supplemental factors for NHBE cells
(Lonza, Allendale, NJ).
2.4. Prototype System
Glass slides containing three concave depressions (Erie
Scientific, Portsmouth NH) were employed as macro-
wells. The slides were cut into thirds (with a diamond
tipped pen) to separate the three depressions. In order to
normalize the data as oxygen consumption rate per cell,
the volumes of the depressions were measured by accu-
rately weighing the macrowell and cover slip dry and
then again with the depression filled with water and
sealed with the cover slip. The calculated volume and
the estimated number of cells were used to convert the
oxygen concentration depletion rates in the macrowell to
average oxygen consumption rate per cell.
T. J. Strovas et al. / Advances in Bioscience and Biotechnology 1 (2010) 398-408
Copyright © 2010 SciRes. ABB
2.5. Respiratory Detection System Description
The glass chips containing the RDS-wells were fabri-
cated in borosilicate glass as previously described re-
sulting in square RDS-wells with dimensions of 5, 8, and
10 mm2 and a well depth of 45 μm resulting in well
volumes of 1.2, 3.2 and 4.5 μl respectively (Figure 1),
[18]. In addition, RDS-wells with a volume of 1.2, 3.2
and 4.5 μl had a corresponding lip area of 100, 49, and
25 mm2. The fabrication process yielded 24, 15 x 15 x
0.5 mm chips per wafer, each containing one etched
square well. Lids for sealing the glass RDS-wells for
experiments on biological samples were fabricated from
borosilicate glass by dicing 4-inch wafers into 20 x 20
mm squares.
During experiments, RDS-wells were housed within a
miniature cell incubator (MCI), which consisted of a
microscope platen, a quartz viewing window, and a del-
rin structure to form a 4 ml MCI-well (Figure 1), [18].
The MCI-well containing 2 ml growth medium was
mounted directly onto the microscope stage. RDS-wells
were held in place within the MCI-well with a 0.4 mm
thick stainless steel chip holder. For routine experiments
a quartz coverslip (VWR, West Chester, PA) was used as
a lid.
2.6. Oxygen Sensor
The oxygen sensors used were 1 μm Pt phosphor car-
boxylate-modified microspheres (Invitrogen, Eugene,
OR) rinsed in deionized water (dH2O; 1 μm beads dis-
continued, 40 nm beads available as product #F-20886).
Approximately 0.1 μl of beads were pipetted into the
RDS-wells after a 1 minute plasma etch to ensure that
the surface of the well was hydrophilic. Sensor patches
were cured for 10 minutes at 170 to ensure surface
adhesion by melting the microspheres. Following sensor
deposition, the RDS-wells were ethanol sterilized and
stored in sterile 12 well plates.
2.7. Sensor Calibration
Sensor patches were calibrated by saturating the liquid in
the RDS wells with gas containing 0, 10, and 20% oxy-
gen (5% CO2, N2 balance). The gasses were passed
through 5 ml of water to humidify them and bathed over
2 ml of dH2O or the mammalian cell culture medium
DMEM (Dulbecco’s Modified Eagle’s Medium) in the
MCI at 24 and 37.
2.8. Seal Tests
The lid used for seal testing the glass RDS-wells (seal
test lid) was made of 5 layers of 0.5 mm thick, 20 x 20
mm borosilicate glass affixed to a piston comprised of
0.25” stainless steel drillstock with a tapered end, fabri-
cated in-house. A thin layer of polydimethylsiloxane
(PDMS) the size of the tapered tip was glued to the tip of
the piston between the stainless steel and the glass as a
conformal layer. The thick layers of glass ensured that
no significant flexing of the piston tip would occur under
pressure. All layers were glued together using medi-
cal-grade biocompatible UV-curable adhesive (Dymax
Corp, Torrington, CT) and all parts were cleaned in an
air plasma (Harrick Plasma Corp, Ithaca, NY) before
gluing to strengthen the adhesive bond.
The diffusional seal tests were conducted in water and
DMEM at 24 by equilibrating 1ml of fluid to 0% dis-
solved oxygen with nitrogen, placing the seal test lid
over the 4.5, 3.2, and 1.2 μl RDS-well sizes with corre-
sponding lip areas of 100, 49, and 25 mm2 at 0 and 15
pounds psi, and turning off the gas line to allow the sur-
rounding liquid outside of the sealed RDS-well to
re-equilibrate to ambient oxygen concentrations. Sen-
sor signal was acquired before the lid was added to en-
sure that the fluid was properly equilibrated and for up to
18 hours after the gas line was turned off.
Figure 1. Schematic of a sealed RDS-well housed in the miniature cell incubator (Molter et al. 2009).
(a) Aluminum platen; (b) Delrin structure that forms the MCI-well to hold cell media; (c) Glass lid;
(d) RDS-well with prophyrin sensor; (e) Chip holder; (f) Quartz viewing window.
T. J. Strovas et al. / Advances in Bioscience and Biotechnology 1 (2010) 398-408
Copyright © 2010 SciRes. ABB
2.9. Bacterial Culture and Environmental
Sample Manipulation
Prior to the measurement of respiration rates, the RDS-
wells were seeded with the cells of interest. For cul-
tures of M. extorquens AM1 and Lake Washington
sediment, 2 ml samples were pipetted directly into the
MCI-well. Aliquots of M. extorquens AM1 cultures in
logarithmic growth were diluted to OD600nm of ~0.10.
The number of M. extorquens AM1 cells in the sample
was calculated via the equation y = 4 x 108x – 4 x 106
such that y = colony forming units and x = OD600nm [31].
For the mammalian cell lines, A549, HCT, and RAW
267.4 cell suspensions were centrifuged for 10 minutes
at 24 at 1200 RPM. The pellet was resuspended in
fresh medium and counted using a hemacytometer with
0.1% Trypan Blue (Invitrogen, Carlsbad, CA).
2.10. Mammalian Culture Spot Seeding Method
A spot seeding method was used for all experiments with
RAW 267.4 cells with the prototype system wells, initial
baseline experiments with A549 cells in RDS-wells, and
hypoxic experiments with A549 cells in RDS wells. Spot
seeding entailed pipetting cells directly into a well con-
taining a sensor patch, ensuring that the cell solution did
not spill over onto the lip of the well. Cell suspensions of
50-150 μl RAW 267.4 cells (~200,000 cells) and 15-30
μl A549 cells (~20,000 cells) were used. The seeded
prototype macrowells and RDS-wells were incubated
under standard culture conditions for one hour to allow
cells to attach to the glass surface. The prototype ma-
crowells and RDS-wells, housed in a sterile 100 mm
petri dish, were then immersed in 20 ml DMEM prior to
the measurement of oxygen uptake rates.
Prior to the measurement of OCRs using the prototype
macrowells, 350 µL of medium was added to the de-
pression and the macrowell was sealed by the addition of
a glass cover slip with the excess medium removed by
pushing down on the cover slip with light force. The
sealed macrowell was placed directly on a glass slide for
imaging and maintained at 37 [24]. RDS-wells spot
seeded with A549 cells were manipulated for experi-
mentation as described above.
2.11 . Mammalian Culture Masked-Well Seeding
Prior to RDS-well seeding, Blue Tack (Semiconductor
Equipment Corporation, Moorpark, CA) was applied to
the surface of the sterilized chip and tape covering the
well was removed leaving the tape only on the lip of the
chip around the well, which will be referred to as a
masked RDS-well. The masked RDS-wells were subse-
quently placed into one well of a 12-well plate. 100,000
A549, HCT, or NHBE cells were seeded in a 2ml vol-
ume and incubated overnight to allow cells to settle and
fully attach. NHBE seeding was conducted with the ad-
dition of 0.01% fibronectin (Invitrogen, Carlsbad, CA)
Prior to experiments the tape covering the lip of the
RDS-well was removed and the seeded RDS-wells were
manipulated as described above.
2.12. Experimental Design
Hypoxic recovery experiments were carried out by in-
cubating A549 cultures at 1% oxygen (5% CO2, N2 bal-
ance) in a triple gas incubator (Thermo, Asheville, NC)
24 hours prior to well seeding. Once spot seeded, the
wells were maintained at 1% oxygen until measurements
were made. For NHBE cells, RDS-wells were masked-
well seeded as described above. 24 hours prior to ex-
periments, the 12 well plates containing seeded RDS-
wells were placed within a 1% oxygen atmosphere.
For respiratory inhibition experiments, RDS-wells
masked-well seeded with A549 cells were incubated in
the presence of 1, 10, or 100 nM rotenone (Sigma Al-
drich, St. Louis, MO) for 15 minutes. Treated RDS-wells
were immediately placed into a petri dish containing 10
ml growth medium without rotenone and transferred into
a MCI-well for the detection of oxygen uptake rates.
To seal an RDS-well seeded with cells, a glass lid was
placed directly on top of the well and moderate pressure
was applied to achieve a good surface tension seal.
Measurements were conducted on a Zeiss LSM 510
META (Thornwood, NY) using a 10 × 0.3 N.A. objec-
tive at 24 for M. extorquens AM1 cultures and Lake
Washington sediment, and 37 for mammalian cell
lines. Temperature control was maintained as previously
described [24]. Sensor signal acquisition was carried out
from 10-60 minutes depending on conditions and cell
type. All data collection and analysis was conducted as
previously described [18]. For RDS-wells seeded with
cells 24-48 hrs prior to experiments, the cells were
stained with Calcien AM (Invitrogen, Carlsbad, CA) for
30 minutes and the entire well was imaged using the tile
feature in the Zeiss LSM software (version 4.0 SP2)
resulting in a 13.5 × 13.5mm image (15 × 15 tiles). Ex-
citation was conducted with a 488nm argon laser at 1%
power and detected in transmission or fluorescence de-
tection mode.
In order to count the number cells in the RDS-wells,
the tile images obtained were first converted to TIF files
using the Zeiss LSM software. The constituent tiles were
separated using Matlab (Mathworks, Novi, MI)
resulting in 225 tiles of 512 × 512 pixels. Each tile im-
age was converted to its grayscale equivalent using Im-
ageJ ( /), [32], processed to re-
move image noise, and segmented using a modified wa-
tershed algorithm in Cell-Profiler (,
[33], which accounted for segmentation errors due to
T. J. Strovas et al. / Advances in Bioscience and Biotechnology 1 (2010) 398-408
Copyright © 2010 SciRes. ABB
clustered cells and over-exposure during image acquisi-
tion. The accuracy of this automated count procedure
was estimated by visually inspecting the segmented cells
overlaid on the original image. The maximum error thus
obtained was computed to be 9.6%.
The OCRs in ppm O2 min-1 cell-1 were calculated
based on the number of estimated or counted cells within
the prototype macrowells or RDS-wells, the volume of
the well used, and the Stern-Volmer equation [1,18]. All
statistical values were calculated using standard two-
sample t significance tests.
3.1. Respiratory Detection System Prototype
In order to test the concept of a small volume respiration
detection system using optical sensors, a prototype sys-
tem was developed. This initial prototype system con-
sisted of phosphorescent sensor deposited into depres-
sions of commercially available glass slides that were
sealed by the application of a coverslip. Once the cover-
slip was applied, the cell sample in the depression con-
sumed oxygen, and the sensor measured the rate of that
decrease (the OCR). Since the oxygen concentration is
monitored in real-time, the experiment can be ended by
the removal of the glass lid before the dissolved oxygen
concentration reaches a predetermined threshold, thereby
preventing the samples from experiencing hypoxic shock.
RAW267.4 cells were used to characterize the ability of
the prototype system to measure OCR. The average rate
per cell was found to be 1.62 +/– 0.65 fmol/min, which
correlated well with previously published data (Table 1),
[34]. The results verified the applicability of the proto-
type system to measure respiration rates. However, the
depression volumes of the commercial slides were
highly variable (32.6-84.2 l), requiring measurement of
each volume for OCR calculation. In order to test the
concept further, a system involving inhouse fabricated
wells (the RDS) was developed.
3.2. RDS
The RDS consists of a low volume square well etched in
borosilicate glass and containing a phosphorescent por-
phyrin sensor used for the measurement of dissolved
oxygen concentrations, (the RDS-well). As in the proto-
type system, a lid can be applied for a period of time to
diffusionally isolate the sample in the well, and the sen-
sor measures the OCR. Once a rate is obtained, the lid is
removed. It should be noted that the microsphere immo-
Table 1. Compiled respiration rates measured. Data presented as mean +/- s.d.
Cell Type Respiration Rate
(mol min-1 cell-1)1 Range
(mol min-1 cell-1)1 Growth Condition Seeding
Method N
Methylobacterium extorquens
AM1 9.00 +/– 7.24 × 10-17 0.43 – 2.80 × 10-16 Succinate - 9
Methylobacterium extorquens
AM1 1.32 +/– 0.95 × 10-16 0.46 – 3.71 × 10-16 Methanol - 9
Lake Wa shington Sedime nt 6.20 +/– 1.83 × 10-12 4.11 – 7.50 × 10-12 - - 3
(prototype system) 1.62 +/– 0.65 m × 10-15 1.01 – 3.69 × 10-15 SCC SS 22
Human Lung Carcinoma
(A549) 4.01 +/– 1.53 × 10-15 2.28 – 7.24 × 10-15 SCC SS 11
Human Lung Carcinoma
(A549) 2.16 +/– 0.52 × 10-15 1.61 – 2.86 × 10-15 1% Atmospheric
Oxygen SS 8
Normal Human Lung
Epithelium (NHBE) 1.12 +/– 0.89 × 10-15 0.52 – 2.85 × 10-15 SCC MW 6
Normal Human Lung
Epithelium (NHBE) 2.67 +/– 0.47 × 10-15 1.60 – 3.2 × 10-15 1% Atmospheric
Oxygen MW 8
Human Lung Carcinoma
(A549) 2.06 +/– 0.36 × 10-15 1.39 – 3.60 × 10-15 SCC MW 20
Human Lung Carcinoma
(A549) 2.29 +/– 0.59 × 10-15 1.64 – 3.03 × 10-15 10 nM dose of
Rotenone MW 7
Human Lung Carcinoma
(A549) 0.75 +/– 0.55 × 10-15 0.25 – 1.74 × 10-15 100 nM dose of
Rotenone MW 12
Colon Cancer
(HCT-116) 4.16 +/– 2.14 ×x 10-15 1.29 – 7.73 × 10-15 SCC MW 10
1. Lake Washington Sediment is shown as mol min-1 Total-Sample-Volume-1; SS = Spot seeded; MW = Masked-well seeded; SCC = Standard Culture Conditions
T. J. Strovas et al. / Advances in Bioscience and Biotechnology 1 (2010) 398-408
Copyright © 2010 SciRes. ABB
bilization protocol used in this study had no effect on
sensor response. Preliminary experiments with RAW
267.4 cells demonstrated the feasibility of using the RDS
to measure OCRs. Therefore, the RDS was assessed
3.3. Sensor Calibration and Seal Verification for
the RDS
Sensor calibration was carried out as described previ-
ously, and the calibration data generated were consistent
with the previous study (data not shown), [18]. Next, the
RDS was tested to determine whether pressure on the lid
was required to obtain an oxygen barrier for the time
period expected for experimental protocols. Seal tests
were conducted in triplicate on RDS-wells with lip areas
of 100, 49, and 25 mm2 using 0 and 15 lbs pressure for
up to 18 hours after the seal test lid was applied (Figure
2). The time for which the diffusional seal was main-
tained, or the seal time, ranged from 38.35 to 66.95 min
for 0 lbs (force applied onto seal lid) and 94.84 to 178.88
min for 15 lbs for all RDS-well sizes combined. Figure
2(b)-(c) demonstrates that the RDS-well lip area had a
negligible effect on seal times and the corresponding
diffusional rate after the seal expired. However, it was
found that diffusional rates of oxygen into the wells after
the seal expired modestly correlated with the length of
the seal time with 0 and 15 lbs force (Figure 2(d)).
3.4. Detection of Respiration Rates from
Bacterial Cultures and
Environmental Samples
To test the use of the RDS with bacterial cultures, loga-
rithmically growing cultures of the gram-negative bacte-
rium Methylobacterium extorquens AM1 were diluted
and tested at an OD600nm of ~0.10 during growth on suc-
cinate or methanol (Figure 3(a)). Respiration rates/cell
were calculated to be 90 +/- 73 amol min-1 cell-1 and 132
+/- 95 amol min-1 cell-1 during growth on succinate and
methanol respectively. The calculated rates had a range
of 43 – 280 amol min-1 cell-1 for succinate growth and
46-371 amol min-1 cell-1 for methanol growth.
To test the use of the RDS with environmental sam-
0 lbs15 lbs
Seal T ime ( min )
1.2 l Well
3.2 l Well
4.5 l Well
0200 400 600 8001000
ppm Oxyg en
Time (min)
(a) (b)
0 lbs15 lbs
1.2 l Well
3.2 l Well
4.5 l Well
y = -3E-06x + 0.000
R² = 0.498
y = -2E -07x + 4E -05
R² = 0.3
050100 150 200
Seal Ti me ( mi n)
0 lb s
15 lbs
Linear (0 lbs)
Line ar ( 15 lbs)
(c) (d)
Figure 2. Summary of seal tests for RDS-wells with volumes of 1.2, 3.2, and 4.5 l using 0 and 15 lbs of force.
RDS-wells with volumes of 1.2, 3.2, and 4.5 l correspond to lip areas of 100, 49, 25 mm2 respectively. (a) Represen-
tative data from a 4.5 l RDS-well after a seal was initiated with 15 lbs and the diffusion rate of oxygen into the well
after the seal expired (8.8 × 10-6 ppm O2 min-1 mm-2). The arrow marks the expiration of the diffusional seal after
154.7 minutes. (b) Average seal time for all well volumes using 0 and 15 lbs. (c) Average diffusional rate of oxygen
into the RDS-wells after the diffusional seal expired. (d) Diffusion rates of oxygen after the seal expired versus seal
times. Data displayed as mean +/– s.d.
T. J. Strovas et al. / Advances in Bioscience and Biotechnology 1 (2010) 398-408
Copyright © 2010 SciRes. ABB
y = - 0 .112x + 6.608
R² = 0.954
y = -0.072x + 4.440
R² = 0.947
0 10203040
ppm Oxygen
Time (min)
y = - 0 .0 69x + 4.537
R² = 0.990
y = -0.041x + 6.648
R² = 0.73 3
y = - 0 .0 75x + 4.961
R² = 0.985
0 1020304050
ppm Oxygen
Time (min)
Figure 3. Oxygen uptake data acquired from bacterial cultures and Lake Washington sediment. (a) Example oxygen
consumption rates obtained from ~ 1.7 × 105 M. extorquens AM1 cells grown on succinate or methanol with an optical
density (600 nm) of ~0.10. (b) Summary of respiration rates obtained from 3 replicates for the top aerobic layer of
Lake Washington sediment.
ples, the oxygen consumption rate from the top aerobic
layer (1 cm) of Lake Washington sediment cores was
measured. The minimum incubation time required for a
rate calculation was 30-40 minutes, which was more
than 40 times shorter compared to routine incubation
approaches (Figure 3(b)), [2]. The rate of oxygen con-
sumption in the RDS system for the sediment samples
was 6.20 1.83 × 10-12 mol min-1 Total-Sample-Volume-1
or 116.25 +/– 34.27 mmol O2 m
-3 hr-1. The values ob-
tained fall within the range previously reported for fresh-
water lakes [2].
3.5. Oxygen Consumption Rates of Mammalian
Cell Lines and the Effects of Extraneous
Experiments were carried out to demonstrate efficacy of
the RDS for assessing response to perturbations with
cultured mammalian cells. The respiration rates from
A549, HCT-116, and NHBE cells were tested under
standard tissue culture conditions and were found to be
similar to literature values for the cell lines A549 [5,18]
and HCT-116 [35], (Table 1).
For one set of perturbation experiments, the effects of
anoxic conditions on respiration rates were tested for
A549 and NHBE cells grown at 20% or 1% oxygen for
24 hours prior to the measurement of respiration rates.
For these experiments, A549 cells were spot seeded and
NHBE cells were masked-well seeded. A549 cells grown
in a 1% oxygen environment displayed an oxygen up-
take rate that was half of the rate observed from control
cells, matching trends in the literature (p-value < 0.005),
[36], while NHBE cells displayed a two-fold increase in
respiration rates compared to control cells (p-value <
0.01; Figure 4). NHBE cells were attached to the glass
surface of the RDS-wells via the addition of 0.01% fi-
bronectin to the cell seeding solution. Precoating the
RDS-wells with fibronectin or adding the fibronectin to
the cell solution were both found to be effective in fa-
cilitating cell attachment. However, if the lip of an
RDS-well became coated with fibronectin, a diffusional
seal was not possible. This problem was alleviated by
applying Blue Tack Tape to the lip of an RDS-well and
then removing it after seeding, which was effective in
allowing both cell attachment in the wells and sealing of
the wells. This process is termed masked-well seeding.
As a second perturbation test, masked-well seeded
A549 cells were treated with 0, 1, 10, and 100 nM rote-
none, a respiratory inhibitor. Concentrations greater than
100 nM were observed to completely arrest oxygen up-
take rates, while a concentration of 1 nM rotenone had
no effect (data not shown). Cells treated with 100 nM
rotenone showed a three-fold decrease in respiration
rates compared to controls (p-value < 0.0005) and cells
treated with 10 nM exhibited no effect (p-value > 0.15;
Figure 5).
After the respiration rates from cells seeded on masked
RDS-wells were measured, the cells were stained with
Calcien AM and imaged (Figure 6). Since cells were
seeded into the wells 24-48 hours prior to measurements,
it was necessary to count the number of live cells in the
wells for an accurate calculation of respiration rates per
cell (see methods section). Two approaches were used
for counting, hemacytometer and image analysis. Com-
parisons of baseline data of A549 cells spot seeded ver-
sus masked-well seeded indicate that spot seeding cells
based on cell counts generated by a hemacytometer
greatly underestimated the number of cells within the
sample chamber when compared to directly counting the
number of seeded cells via image analysis.
In this report we demonstrate the applicability of the
Respiratory Detection System for directly measuring
OCRs from different types of cell samples. Previous
efforts have focused on the measurement of single cell
respiration rates, and here we show that the RDS is a
T. J. Strovas et al. / Advances in Bioscience and Biotechnology 1 (2010) 398-408
Copyright © 2010 SciRes. ABB
y = -0.501x + 5.147
R² = 0.990
y = - 0 .2 32x + 3.792
R² = 0.995
0510 15
ppm Oxygen
Time (min)
20% Oxygen
1% Oxyge n
Linear (20%
y = -0.084x + 7.13
R² = 0.903
y = -0.193x + 3.349
R² = 0.996
0 10203040
ppm Oxygen
Time (min)
20% Oxygen
1% Oxygen
Linear (20%
Figure 4. Oxygen uptake data obtained from mammalian cell lines. (a) Example data from A549 cells grown 24 hours
under standard tissue culture conditions and in a 1% oxygen environment. (b) Example data from NHBE cells main-
tained 24 hours under a 20% and 1% oxygen environment.
Control10nM Rotenone100nM Rotenone
Respiration Rate (mol O2 min-1 cell-1)
Figure 5. Summary of respiration rates obtained from A549
cultures treated with 10 and 100 nM rotenone. Data displayed
as mean +/– s.d.
viable alternative for use with bulk populations. The
RDS was demonstrated to be compatible with attached
mammalian cell lines, low optical density bacterial cul-
tures, and environmental samples. In addition, the RDS
was demonstrated to maintain a diffusional seal for up to
three hours to allow for OCR measurements on highly
diluted or slowly respiring samples. These features and
the ability to directly measure the OCR are a significant
improvement compared to oxygen flux-based measure-
ments in sample chambers made from plastics. The de-
sign parameters of the RDS integrate large glass-glass
interfaces which allows for long diffusional seals. Due to
the fabrication method to make RDS-wells, well dimen-
sions can be easily changed to accommodate different
sample types with the potential to measure OCR’s from
larger multi-cellular organisms (e.g. nematodes, zebraf-
ish embryos). In addition to these novel characteristics,
the simple design, inexpensive fabrication, and straight-
forward implementation of the RDS would readily allow
for its integration into a preexisting plate reader format,
on an optical microscope system, or as a portable device
for use in the field.
Our results indicate that repeatable oxygen uptake
rates could be measured with experimental times as short
as ten minutes in the largest RDS-well, depending on the
cellular sample. The versatility of the RDS is demon-
strated by the broad spectrum of biological samples that
can be tested, including mammalian cell strains that re-
quire special coatings to assist in cell attachment and
The RDS utilizes a commercially available phospho-
rescent oxygen sensor to measure oxygen uptake rates.
However, any optically based oxygen sensor could be
employed with the RDS. Since phosphorescence-based
oxygen reporters are detected by the lifetime decay of
the excited phosphor [15], the RDS can be used to per-
form both fluorescence-based and phosphorescence-based
measurements due to the difference in decay timescales,
nanoseconds vs. microseconds respectively. The ability
to multiplex oxygen sensing with additional fluorescence
reporters allows for a wide array of fluorescence-based
reporters to be used simultaneously, which could be im-
aged on a microscope, detected with a plate reader, or
with a basic filter-based light emitting diode/photo mul-
tiplier tube optical train [5,9].
The RDS is a device that can be used to directly
measure the respiration rate of bulk samples in a reversi-
ble diffusionally sealed sample chamber. The simplicity
of the device and the ability to use microliter-scale cell
samples makes it a logical alternative or enhancement to
other established methods for measuring oxygen. By
integrating the RDS design with existing systems, OCR
measurements would directly reflect oxygen consump-
tion rather than the flux of oxygen near the sensor. These
characteristics of the RDS would allow the system to be
employed as a portable device, a stand-alone bench top
system, a microscope mounted device, or it could be
integrated into a standard plate reader format. The latter
approach would allow taking advantage of the absorb-
ance, fluorescence, and luminescence detection capabili-
ties of microplate readers to multiplex a wide array of
assays with the detection of oxygen uptake rates [37].
The respiration rate of a cell can be used as a gauge of
T. J. Strovas et al. / Advances in Bioscience and Biotechnology 1 (2010) 398-408
Copyright © 2010 SciRes. ABB
(a) (b)
(c) (d)
Figure 6. Images of an RDS-well seeded with NHBE cells. (a) A transmission tile image in channel D. (b) Tile image of NHBE cells
stained with Calcien AM. (c) Single image from a transmission tile scan showing the edge of the sensor ring. (d) Single image from
the fluorescence tile scan of NHBE cells stained with Calcien AM.
metabolic state and cellular health. Thus, the RDS has a
broad range of possible applications, especially in cases
in which the sample size is limited. The ability to main-
tain a diffusional seal allows for greater sensitivity when
measuring subtle changes in OCRs and when the sample
exhibits a slow respiration rate such as dilute bacterial
cultures or environmental samples. This sensitivity could
be beneficial in the study of the effects of experimental
drugs on target cells as an assay in a commercial envi-
ronment or as a clinical screen of fertilized embryos to
be used for in vitro fertilization [16]. This same concept
is relevant in gauging the effects of stimuli on cells in
the investigation of stress response and toxicity studies.
Lastly, the compatibility of the RDS with attached cells
T. J. Strovas et al. / Advances in Bioscience and Biotechnology 1 (2010) 398-408
Copyright © 2010 SciRes. ABB
and an optically based oxygen sensor allows for fast
replicate measurements without the need for stirring, as
compared to detection methods utilizing openly diffus-
ible sample chambers. These characteristics impart sig-
nificant improvements to existing methods for the meas-
urement of respiration rates from biological samples.
This work was supported by an NIH National Human
Genome Research Institute (NHGRI) Centers of Excel-
lence in Genomic Sciences grant (P50 HG 002360) and
the U. S. Department of Energy (DE-PS02-07ER07-14).
[1] Strovas, T.J., Dragavon, J.M., Hankins, T.J., Callis, J.B.,
Burgess, L.W. and Lidstrom, M.E. (2006) Measurement
of respiration rates of Methylobacterium extorquens
AM1 cultures by use of a phosphorescence-based sensor.
Applied Environmental Microbiology, 72, 1692-1695.
[2] Bidanda, B. (2005) Respiration in aquatic ecosystems. In:
Del Giorgio, P.A. and Williams, P.J. le B., Eds., Journal
of Plankton Research, Oxford University Press, New
[3] Fink, S.L. and Cookson, B.T. (2005) Apoptosis, pyrop-
tosis, and necrosis: mechanistic description of dead and
dying eukaryotic cells. Infection and Immunity, 73,
[4] Weissig, V., Boddapati, S.V., Jabr L., and D'Souza, G.G.
(2007) Mitochondria-specific nanotechnology. Nano-
medicine, 2, 275-285.
[5] Wu, M., Neilson, A., Swift, A.L., Moran, R., Tamagnine,
J., Parslow, D., Armistead, S., Lemire, K., Orrell, J.,
Teich, J., Chomicz, S. and Ferrick, D.A. (2007) Multi-
parameter metabolic analysis reveals a close link be-
tween attenuated mitochondrial bioenergetic function and
enhanced glycolysis dependency in human tumor cells.
American Journal of Physiology: Cellular Physiology,
292, 125-136.
[6] Bell, E.L. andChandel, N.S. (2007) Mitochondrial oxy-
gen sensing: Regulation of hypoxia-inducible factor by
mitochondrial generated reactive oxygen species. Essays
in Biochemistry, 43, 17-27.
[7] Fink, S.L. and Cookson, B.T. (2007) Pyroptosis and host
cell death responses during Salmonella infection. Cellu-
lar Microbiology, 9, 2562-2570.
[8] Maley, C.C. (2007) Multistage carcinogenesis in Bar-
rett’s esophagus. Cancer Letters, 245, 22-32.
[9] Eklund, S.E., Thompson, R.G., Snider, R.M., Carney, C.
K., Wright, D.W., Wikswo, J. and Cliffel, D.E. (2009)
Metabolic discrimination of select list agent by monitor-
ing cellular responses in a multianalyte microphysiome-
ter. Sensors, 9, 2117-2133.
[10] Ramamoorthy, R., Dutta, P.K. and Akbar, S.A. (2003)
Oxygen sensors: Materials, methods, designs, and appli-
cations. Journal of Material Science, 38, 4271-4282.
[11] Renger, G. and Hanssum, B. (2009) Oxygen detection in
biological systems. Photosynthesis Research, 102, 487-
[12] Kuang, Y. and Walt, D.R. (2007) Detecting oxygen con-
sumption in the proximity of Saccharomyces cerevisiae
cells using self-assembled fluorescent nanosensors. Bio-
technology Bio engineering, 96, 318-325.
[13] Saito, T., Wu, C.C., Shiku, H., Yasukawa, T., Yokoo, M.,
Ito-Sasaki, T., Abe, H., Hoshi, H. and Matsue, T. (2006)
Oxygen consumption of cell suspension in a poly (di-
methylsiloxane) (PDMS) microchannel estimated by
scanning electrochemical microscop. Analyst, 131, 1006-
[14] Warkentin, M., Freese, H.M., Karsten, U. and Schumann,
R. (2007) New and fast method to quantify respiration
rates of bacterial and plankton communities in freshwater
ecosystems by using optical oxygen sensor spots. Ap-
plied Environmental Microbiology, 73, 6722-6729.
[15] Papkovsky, D.B. and O’Riordan, T.C. (2007) Emerging
applications of phosphorescent metalloporphyrins. Jour-
nal of Fluorescence, 15, 569-584.
[16] Abe, H. (2007) A non-invasive and sensitive method for
measuring cellular respiration with a scanning electro-
chemical microscopy to evaluate embryo quality. Journal
of Mammalian Ova Research, 24, 70-78.
[17] Borisov, S.M., Nuss, G. and Klimant, I. (2008) Red
light-ex- citable oxygen sensing materials based on
platinum(II) and palladium(II) benzoporphyrins. Ana-
lytical Chemistry, 80, 9435-9442.
[18] Molter, T.W., McQuaide, S.C., Suchorolski, M.T.,
Strovas, T.J., Burgess, L.W., Meldrum, D.R. and Lid-
strom, M.E. (2009) A microwell array device capable of
measuring single-cell-oxygen consumption rates. IEEE
Sensors and Actuators B, 135, 678-686.
[19] O’Donovan, C., Twomey, E., Alderman, J., Moore, T.
and Papkovsky, D. (2006) Development of a respiromet-
ric biochip for embryo assessment. Lab on a Chip, 6,
[20] O'Mahony, F.C., O’Donovan, C., Hynes, J., Moore, T.,
Davenport, J. and Papkovsky, D.B. (2005) Optical oxy-
gen microrespirometry as a platform for environmental
toxicology and animal model studies. Environmental
Science and Technology, 39, 5010-5014.
[21] Wilson, D.F., Vanderkooi, J.M., Green, T.J., Maniara, G.,
Defeo, S.P. and Bloomgarden, D.C. (1987) A versatile
and sensitive method for measuring oxygen. Advances in
Experimental Medical Biology, 215, 71-77.
[22] Alderman, J., Hynes, J., Floyd, S.M., Krüger, J.,
O’Conner, R. and Papkovsky, D.B. (2004) A low-volume
platform for cell-respirometric screening based on
quenched- luminescence oxygen sensing. Biosensors and
Bioelectronics, 19, 1529-1535.
[23] Birmele, M., Roberts, M. and Garland, J. (2006) Charac-
terization of methods for determining sterilization effi-
cacy and reuse efficiency of oxygen biosensor multiwell
plates. Journal of Microbiological Methods, 67, 619-623.
[24] Dragavon, J., Molter, T., Young, C., Strovas, T.,
McQuaide, S., Holl, M., Zhang, M., Cookson, B., Jen, A.,
Lidstrom, M. and Burgess, L. (2008) A cellular isolation
system for real-time single-cell oxygen consumption
monitoring. Journal of the Royal Society Interface, 5(S2),
[25] Will, Y., Hynes, J., Ogurtsov, V.I. and Papkovsky, D.B.
(2007) Analysis of mitochondrial function using phos-
phorescent oxygen-sensitive probes. Nature Protocols, 1,
[26] Wodnicka, M., Guarino, R.D., Hemperly, J.J., Timmins,
M.R., Stitt, D. and Pitner, J.B. (2000) Novel fluorescent
T. J. Strovas et al. / Advances in Bioscience and Biotechnology 1 (2010) 398-408
Copyright © 2010 SciRes. ABB
technology platform for high throughput cytotoxicity and
proliferation assays. Journal of Biomolecular Screening,
5, 141-152.
[27] Arain, S., Weiss, S., Heinzle, E., John, G.T., Krause, C.
and Klimant, I. (2005) Gas sensing in microplates with
optodes: Influence of oxygen exchange between sample,
air, and plate material. Biotechnology and Bioengineer-
ing, 90, 271-280.
[28] Gerencser, A.A., Neilson, A., Choi, S.W., Edman, U.,
Yadava, N., Oh, R.J., Ferrick, D.A., Nicholls, D.G. and
Brand, M.D. (2009) Quantitative microplate-based res-
pirometry with correction for oxygen diffusion. Analyti-
cal Chemistry, 81, 6868-6878.
[29] Strovas, T.J., Sauter, L.M., Guo, X. and Lidstrom, M.E.
(2007) Cell-to-cell heterogeneity in growth rate and gene
expression in Methylobacterium extorquens AM1. Jour-
nal of Bacteriology, 189, 7127-7133.
[30] Kalyuzhnaya, M.G., Bowerman, S., Nercessian, O., Lid-
strom, M.E. and Chistoserdova, L. (2005) Highly diver-
gent genes for methanopterin-linked C1 transfer reac-
tions in Lake Washington, assessed via metagenomic
analysis and mRNA detection. Applied Environmental
Microbiology, 71, 8846-8854.
[31] Guo, X. (2006) Intergrative study of metabolism of Me-
thylobacterium extorquens AM1 from a metabolomic
perspective. Ph.D. dissertation, University of Washington,
Department of Chemical Engineering.
[32] Abramoff, M.D., Magelhaes, P.J. and Ram, S.J. (2004)
Image processing with ImageJ. Biophotonics Interna-
tional, 11, 36-42.
[33] Carpenter, A.E., Jones, T.R., Lamprecht, M.R., Clarke, C.
Kang, I.H., Friman, O., Guertin, D.A., Chang, J.H.,
Lindquist, R.A., Moffat, J., Golland, P. and Sabatini, D.
M. (2006) CellProfiler: image analysis software for iden-
tifying and quantifying cell phenotypes. Genome Biology,
7, R100.
[34] Nalwaya, N. and Deen, W.M. (2005) Nitric oxide, oxy-
gen, and superoxide formation and consumption in
macrophage cultures. Chemical Research in Toxicology,
18, 486-93.
[35] Leschelle, X., Goubern, M., Andriemihaja, M., Blottiere,
H.M., Coupland, E., Gonzalez-Barroso, M.D., Petite, C.,
Pagniez, A., Chaumentet, C., Mignotte, B., Bouillaud, F.
and Blachier, F. (2005) Adaptative metabolic response of
human colonic epithelial cells to the adverse effects of
the luminal compound sulfide. Biochimica et Biophysica
Acta, 1725, 201-212.
[36] Heerlein, K., Schulze, A., Hotz, L., Bartsch, P. and
Mairbaurl, H. (2005) Hypoxia decreases cellular ATP
demand and inhibits mitochondrial respiration of A549
cells. American Journal of Respiratory Cell and Mo-
lecular Biology, 32, 44-51.
[37] Lidstrom, M.E. and Meldrum, D.R. (2003) Life-on-a-
chip. Nature Reviews Microbiolo gy, 1, 158-164.