In vitro intermittent hypoxia: challenges for creating hypoxia in cell culture

doi:10.1016/S1569-9048(03)00077-6

Respiratory Physiology & Neurobiology

Volume 136, Issues 2–3, 16 July 2003, Pages 131-139

https://doi.org/10.1016/S1569-9048(03)00077-6 Get rights and content

Abstract

Intermittent hypoxia has been implicated in morbidities associated with sleep apnea, and may be a novel cellular signal for inflammation [J. Appl. Physiol. 90 (2001) 1986]. Standard cell culture has two major limitations for studying the effects of steady-state P_O₂ and intermittent hypoxia. First, convective mixing in the culture media can be variable, making precise control of cellular P_O₂ difficult. Second, diffusion of oxygen through the culture media slows changes in cellular P_O₂ after rapid changes in the gas phase P_O₂. Our estimates of diffusional transients for standard cell culture suggest significant restrictions in the ability to cycle P_O₂ at frequencies relevant to intermittent hypoxia. We present a novel system for forced convection cell culture with adherent cells inside capillary tubing. Steady state cellular P_O₂ is regulated to an accuracy of approximately 1 Torr. The response time for cycling of P_O₂ is less than 1.6 sec. This system is ideally suited for studies of intermittent hypoxia in adherent cells.

Introduction

Isolated culture of adherent cells is an important reductionist approach for investigating the role of specific cells in inflammatory responses. Despite a longstanding interest in the effects of hypoxia and re-oxygenation on the expression of inflammatory mediators, there have been few methods available for precise regulation of cellular P_O₂ in adherent cell culture (Lewis et al., 1995, Wei et al., 1999, Manevich et al., 2001). Recently, interests in the effects of oxygen on inflammation have extended to the effects of intermittent hypoxia (Prabhakar, 2001). Several studies have identified an association between obstructive sleep apnea, a disease characterized by episodic de-saturation, and increased expression of several mediators of inflammation (Chin et al., 2000, Vgontzas et al., 2000). No cell culture methods have been reported, however, for rapid and precise cycling of cellular oxygen tension.

A major obstacle to precise regulation of cellular P_O₂ in culture is resistance to the transport of oxygen from headspace gas to the cells. Oxygen is transported through the media surrounding cells by molecular diffusion and by convective mixing of the media (Bird et al., 1960). Characterization of the relationship between headspace P_O₂ and cellular P_O₂ requires an estimate of the convective contributions to oxygen transport. In standard cell culture, with the cells adherent to the bottom of culture wells and covered by several millimeters of media, mixing of the media is governed by free convection. These free convective flows typically are driven by thermally generated buoyancy differences, and by mechanical vibrations transmitted from the surrounding environment (Bird et al., 1960, Vivian and King, 1964). Convective transport has been treated by some investigators as zero (i.e. pure diffusional transport in an unstirred layer) (Metzen et al., 1995, Mamchaoui and Saumon, 2000, Chen and Deen, 2002) and by other investigators as infinite (i.e. complete mixing such that headspace P_O₂ equals cellular P_O₂) (Albina et al., 1995, McCormick et al., 2000). Direct measurements of the P_O₂ gradient in media overlying cells, however, suggest that the amount of convective mixing in some culture systems lies between these two extremes (Otto and Baumgardner, 2001). Additionally, measurements of the P_O₂ gradient suggest that convective contributions to oxygen transport can vary substantially over time (Otto and Baumgardner, 2001).

When the focus of investigation is not just precise steady-state control of cell P_O₂, but also precise and rapid switching of P_O₂ to simulate intermittent hypoxia, transport of oxygen to the cells presents an additional problem. Diffusion of oxygen through the media is not instantaneous, and rapid step changes in P_O₂ in the headspace gas are not translated directly into step changes in P_O₂ at the cell. The time constant that characterizes the delay due to diffusion transients is a function of the media thickness squared, divided by the effective diffusivity (Bird et al., 1960, Crank, 1975). Thus for any system that has less than perfect mixing, and therefore, a finite effective diffusivity, the thickness of media surrounding the cells is a dominant factor in determining how rapidly the cellular P_O₂ can be changed.

In this report, we use a range of reported values for the effective diffusivity of oxygen through culture media, in standard cell culture apparatus, to estimate a range of time constants for changes in cellular P_O₂. Then we present a novel system for precise control of cellular P_O₂ in cultured cells that uses forced convective flow over cells adherent to the inside of impermeable capillary tubing. Because the cylindrical geometry is well characterized, steady-state cellular P_O₂ can be precisely determined with an error of less than 1 Torr. The dimensions of the capillary, low dead volume at the entrance of the capillary, and rapid transit times through the capillary are ideally suited for rapid and precise switching of cellular P_O₂.

Section snippets

Cycling cellular P_O₂ in standard cell culture

An estimate of diffusion transients after a step change in the headspace (i.e. gas phase) P_O₂ in a culture chamber can be calculated by considering solutions to a similar problem where the media is initially held at uniform P_O₂, the media surface at time zero undergoes a step change to a different P_O₂, and the bottom of the media well is impermeable to oxygen. Analytic solutions for the change in P_O₂ at the bottom of the media versus time, for this simpler problem (which neglects oxygen

Results

For standard cell culture apparatus, calculated oxygen tension at the bottom of a culture well after a step change in P_O₂ at the media surface is shown in Fig. 3. Even for an effective diffusivity of 50 times the molecular diffusivity, which would reflect vigorous mixing of the media, the P_O₂ at the bottom of the well takes 56 sec to reach 90% of the steady-state value. For more realistic effective diffusivities, such as directly measured in a previous study (Otto and Baumgardner, 2001), the

Discussion

Although standard cell culture has been adapted for studies of the effects of oxygen tension on cultured cells (Albina et al., 1995, McCormick et al., 2000, Otto and Baumgardner, 2001, Prabhakar, 2001, Chen and Deen, 2002), for quantitative studies of the effects of cellular P_O₂ the standard apparatus has two shortcomings. First, the amount of mixing in culture media due to free convection is difficult to predict in any given environment (Bird et al., 1960), and in some environments may vary

Acknowledgements

The authors gratefully acknowledge the excellent technical assistance of Julia Fox. The authors are also thankful for helpful discussions, unfailing support, and inspiration from Dr J.A. Quinn. Supported in part by NIH GM59274, NIH HL59052, AHA 0151528U and NIH GM64486.

References (24)

K. Chin et al.
Effects of nasal continuous positive airway pressure on soluble cell adhesion molecules in patients with obstructive sleep apnea syndrome
Am. J. Med.
(2000)
P.E. James et al.
The effects of endotoxin on oxygen consumption of various cell types in vitro: an EPR oximetry study
Free Radic. Biol. Med.
(1995)
R.S. Lewis et al.
Kinetic analysis of the fate of nitric oxide synthesized by macrophages in vitro
J. Biol. Chem.
(1995)
E. Metzen et al.
Pericellular P_O₂ and O₂ consumption in monolayer cell cultures
Respir. Physiol.
(1995)
J.E. Albina et al.
Macrophage activation by culture in an anoxic environment
J. Immunol.
(1995)
J.E. Baumgardner et al.
Effect of ventilator management on arterial P_O₂ oscillations (abstract)
Shock
(2002)
R.B. Bird et al.
Transport Phenomena
(1960)
B. Chen et al.
Effect of liquid depth on the synthesis and oxidation of nitric oxide in macrophage cultures
Chem. Res. Toxicol.
(2002)
J. Crank
The Mathematics of Diffusion
(1975)
J.M. Davis et al.
Hollow-fiber cell culture
Methods Mol. Biol.
(1997)

A.B. Fisher et al.

Endothelial cellular response to altered shear stress

Am. J. Physiol. Lung Cell. Mol. Physiol.

(2001)

J.A. Frangos et al.

Flow effects on prostacyclin production by cultured human endothelial cells

Science

(1985)

Cited by (36)

Limitations of oxygen delivery to cells in culture: An underappreciated problem in basic and translational research
2017, Free Radical Biology and Medicine
Citation Excerpt :
However, to estimate the amount of time needed to equilibrate cell-free medium to a novel oxygen concentration, a derivation of Fick's law must be used. There are several publications describing the use of Fick's law in this manner, however, the mathematics behind this calculation are complex and are best reviewed elsewhere [29,31]. Diffusion times in cell-free medium have been experimentally determined by several researchers.
Molecular oxygen is one of the most important variables in modern cell culture systems. Fluctuations in its concentration can affect cell growth, differentiation, signaling, and free radical production. In order to maintain culture viability, experimental validity, and reproducibility, it is imperative that oxygen levels be consistently maintained within physiological “normoxic” limits. Use of the term normoxia, however, is not consistent among scientists who experiment in cell culture. It is typically used to describe the atmospheric conditions of a standard incubator, not the true microenvironment to which the cells are exposed. This error may lead to the situation where cells grown in a standard “normoxic” oxygen concentration may actually be experiencing a wide range of conditions ranging from hyperoxia to near-anoxic conditions at the cellular level. This apparent paradox is created by oxygen's sluggish rate of diffusion through aqueous medium, and the generally underappreciated effects that cell density, media volume, and barometric pressure can have on pericellular oxygen concentration in a cell culture system. This review aims to provide an overview of this phenomenon we have termed “consumptive oxygen depletion” (COD), and includes a basic review of the physics, potential consequences, and alternative culture methods currently available to help circumvent this largely unrecognized problem.
Vessel network formation in response to intermittent hypoxia is frequency dependent
2015, Journal of Bioscience and Bioengineering
A bioreactor for subjecting cultured cells to fast-rate intermittent hypoxia
2012, Respiratory Physiology and Neurobiology
Citation Excerpt :
Nevertheless, the equipment used in these experimental approaches is not readily available to most labs because of the need for special technologies to construct the high-precision miniaturized cell culture settings. Moreover, it is worth noting that our approach, in contrast to bioreactors based on liquid convection through cell cultures inside capillary tubing (Baumgardner and Otto, 2003), presents the advantage that the cells are cultured on standard glass coverslips and are thus readily available for inspection with any microscopy technique or bio-analytical assay. In contrast with previously proposed methods, the novel bioreactor described in this work is easily assembled since it requires simple commercially available components used in various kinds of bioreactors (Burdick and Vunjak-Novakovic, 2009).
High frequency intermittent hypoxia is one of the most relevant injurious stimuli experienced by patients with obstructive sleep apnea (OSA). Given that the conventional setting for culturing cells under intermittent hypoxia conditions is limited by long equilibration times, we designed a simple bioreactor capable of effectively subjecting cultured cells to controlled high-frequency hypoxic/normoxic stimuli. The bioreactor's operation is based on exposing cells to a medium that is bubbled with the appropriate mixture of gases into two separate containers, and from there it is directed to the cell culture dish with the aid of two bidirectional peristaltic pumps. The device was tested on human alveolar epithelial cells (A549) and mouse melanoma cells (B16-F10), subjecting them to patterns of intermittent hypoxia (20 s at 5% O₂ and 50 s at 20% O₂), which realistically mimic OSA of up to severe intensity as defined by the apnea hypopnea index. The proposed bioreactor can be easily and inexpensively assembled and is of practical use for investigating the effects of high-rate changes in oxygen concentration in the cell culture medium.
Oxygen-dependent regulation of nitric oxide production by inducible nitric oxide synthase
2011, Free Radical Biology and Medicine
Citation Excerpt :
As a consequence, there are many opportunities for physiological and pathophysiological cellular oxygen tensions to influence NO production. Despite the potential impact of PO2 on many factors influencing NO production, surprisingly little attention has been paid, in cell culture studies, to specifying cellular PO2 or specifying any of the experimental factors that could create a gradient between headspace PO2 and cellular PO2, for example, medium layer thickness, cell plating density, or specific cellular O2 metabolism [91, 204, 205]. As this field moves forward, increased attention to estimating or measuring cellular PO2 may facilitate comparison of studies with differing results.
Inducible nitric oxide synthase (iNOS) catalyzes the reaction that converts the substrates O₂ and l-arginine to the products nitric oxide (NO) and l-citrulline. Macrophages, and many other cell types, upregulate and express iNOS primarily in response to inflammatory stimuli. Physiological and pathophysiological oxygen tension can regulate NO production by iNOS at multiple levels, including transcriptional, translational, posttranslational, enzyme dimerization, cofactor availability, and substrate dependence. Cell culture techniques that emphasize control of cellular PO₂, and measurement of NO or its stable products, have been used by several investigators for in vitro study of the O₂ dependence of NO production at one or more of these levels. In most cell types, prior or concurrent exposure to cytokines or other inflammatory stimuli is required for the upregulation of iNOS mRNA and protein by hypoxia. Important transcription factors that target the iNOS promoter in hypoxia include hypoxia-inducible factor 1 and/or nuclear factor κB. In contrast to the upregulation of iNOS by hypoxia, in most cell types NO production is reduced by hypoxia. Recent work suggests a prominent role for O₂ substrate dependence in the short-term regulation of iNOS-mediated NO production.
Experimental Models to Study End-Organ Morbidity in Sleep Apnea: Lessons Learned and Future Directions
2022, International Journal of Molecular Sciences
Modelling human CNS injury with human neural stem cells in 2- and 3-Dimensional cultures
2020, Scientific Reports

View all citing articles on Scopus

View full text

In vitro intermittent hypoxia: challenges for creating hypoxia in cell culture

Abstract

Introduction

Section snippets

Cycling cellular PO2 in standard cell culture

Results

Discussion

Acknowledgements

Am. J. Med.

Free Radic. Biol. Med.

J. Biol. Chem.

Respir. Physiol.

Macrophage activation by culture in an anoxic environment

J. Immunol.

Effect of ventilator management on arterial PO2 oscillations (abstract)

Shock

Transport Phenomena

Effect of liquid depth on the synthesis and oxidation of nitric oxide in macrophage cultures

Chem. Res. Toxicol.

The Mathematics of Diffusion

Hollow-fiber cell culture

Methods Mol. Biol.

Endothelial cellular response to altered shear stress

Am. J. Physiol. Lung Cell. Mol. Physiol.

Flow effects on prostacyclin production by cultured human endothelial cells

Science

Cycling cellular P_O₂ in standard cell culture

Effect of ventilator management on arterial P_O₂ oscillations (abstract)