J Appl Physiol 104: 262-268, 2008.
First published October 25, 2007; doi:10.1152/japplphysiol.00893.2007
8750-7587/08 $8.00
INNOVATIVE METHODOLOGY
Acoustic plethysmography measures breathing in unrestrained neonatal mice
J. Andrew Daubenspeck,1,2
Aihua Li,1 and
Eugene E. Nattie1
1Physiology Department, Dartmouth Medical School, Lebanon; and 2Thayer School of Engineering, Dartmouth College, Hanover, New Hampshire
Submitted 20 August 2007
; accepted in final form 24 October 2007
 |
ABSTRACT
|
|---|
Measurement of breathing volumes in neonatal mice is of growing importance in order to characterize the influence of development and genetic modifications on respiratory control to evaluate hypotheses concerned with human infant deficits that may affect sudden infant death syndrome, for example. Current techniques require undesirable physical constraints or incur possible artifacts specific to very small animals. We have examined the utility of a recently proposed approach using an acoustic resonance procedure that does not require undue physical constraint beyond placement in the acoustic plethysmograph. We show here that this approach can be applied to baby mice 5 days after birth and that it can be accurately calibrated. In addition, this approach should be useful to study unrestrained neonatal mice under conditions where body temperature approaches environmental temperature and barometric plethysmography cannot be used.
small animal plethysmography; Helmholtz resonator
MEASUREMENT of respiration in newborn mice is important in order to characterize the development of respiratory control mechanisms from birth through adulthood in normal mice and in mice that have been genetically modified to mimic defects that correlate to human diseases such as central hypoventilation syndrome (9, 11) and to defects in serotonin that may relate to sudden infant death syndrome (SIDS) (4).
This measurement presents many technological problems, particularly if the neonatal animal is conscious and unrestrained. Invasive instrumentation required to directly measure airflow is ruled out, and traditional head-out plethysmography applies the stress of a neck seal and immobilization. Barometric plethysmography has been used to measure breathing successfully in unrestrained, larger animals and depends on measuring the effects of heating and humidification of inspired gas, a process that causes the increase in alveolar volume to be greater than the inspiratory decrease in plethysmograph volume and measurably raises the chamber pressure.
Reynolds and Frazer (10) recently suggested a novel approach to unrestrained plethysmography using an acoustic resonance approach. In their method, a mouse is enclosed in a chamber similar in dimension to that used in barometric plethysmography. The chamber has a nozzle that is acoustically excited at a frequency close to, but not at the acoustic resonance frequency of the chamber and serves as a Helmholtz acoustic resonator. Reynolds and Frazer showed that such a device could be calibrated mechanically and used to measure respiratory volumes on the order of 200 µl in unrestrained adult mice.
We have adapted their methodology to permit measuring tidal volumes smaller than 15 µl in 5- to 7-day-old (P5–P7) mice by incorporating improved calibration and noise reduction techniques. Our approach provides a real-time output of volume vs. time that can be combined with other measurements to evaluate transient as well as steady-state changes in breathing. We show here that 10- to 30-µl calibration volumes are measured accurately in our device (the Dartmouth Acoustic Plethysmograph, DAP) and that we can measure tidal volumes and ventilation in such newborn mice that compare well with data from the literature. We intend to use this methodology to follow the development of respiratory responses to hypercapnia and hypoxia in order to characterize differences between normal and genetically modified mice from birth onward.
|
METHODS
|
|---|
Animals.
The mice used here were housed in a room with a light (resting) period from 7 AM to 7 PM and a dark (active) period from 7 PM to 7 AM. Food and water were available ad libitum. All experiments were performed between 8 AM and 3 PM. All procedures were within the guidelines of the National Institutes of Health for animal use and care and were approved by the Dartmouth College Institutional Animal Use and Care Committee.
Theory.
The theoretical basis for this measurement is that of Reynolds and Frazer (10) and will only be summarized briefly here except where we have selected a few options differently. The basic idea is that the acoustic pressure developed inside the chamber is a function of the excitation frequency of the driving speaker relative to the resonant frequency of the chamber. Equation 1 below is from equation 1 of Reynolds and Frazer (as are Eqs. 2 and 3) and shows how the physical characteristics of the system influence the measured acoustic pressure inside the chamber (Pc) relative to the excitation pressure applied by the speaker (Pe):
 | (1) |
where 
0 is chamber resonant frequency and e is excitation frequency in radians per unit time, R is airflow resistance of the nozzle, and C is chamber gas compliance. In turn, the resonance frequency 0 is a function of the speed of sound (c0), the area of the nozzle (S), the effective length of the nozzle (l'), and the volume of the chamber, V
 | (2) |
and the gas compliance C is a function of the chamber volume (V), and the density (
0) and c0 of the gas at whatever conditions pertain
 | (3) |
Essentially, the resonant frequency depends on the volume of gas within the chamber that is free to be compressed by the excitation wave (the gas in the chamber provides the elastic factor in a spring-mass oscillator) and the dimensions of the very low resistance nozzle that determine the nozzle gas mass factor for that oscillation. Equation 2 represents the derivation of Helmholtz and others as detailed by Alster (1), who noted that the true situation is considerably more complex for geometries and conditions commonly encountered.
The very insightful contribution of Reynolds and Frazer (10) was to see that as the mouse inside the plethysmograph inspires, gas is removed from the chamber and drawn into the lungs, which are acoustically isolated from the high-frequency excitation by the high-acoustic impedance of the mouse airway. The animal's chest wall, however, expands and forces some gas out of the low-resistance nozzle, thus reducing the amount of gas within the chamber free to resonate, the chamber volume V. This alters the relationship between the driving frequency, which is fixed, and the resonant frequency, which varies with breathing and thus alters the acoustic pressure (Pc) developed inside the chamber. Since V varies as breathing occurs, the chamber compliance C and resonant frequency 0 vary according to Eqs. 2 and 3, causing Pc to vary for the fixed excitation level Pe. The frequency of the acoustic pressure Pc in the plethysmograph will be the frequency of the excitation pressure () but the amplitude of Pc will vary with the changes in V due to breathing. The calibration of this change in acoustic pressure in terms of volume change is obviously dependent on the temperature and pressure conditions and the physical nature of the gas, as well as on the plethysmograph dimensions.
It should be noted that these equations describing the Helmholtz resonance apply when the wavelength of the excitation wave is an order of magnitude or more greater than any chamber dimension. Although this is the case here, the representation of Eq. 2 does not hold exactly and cannot be used for accurate prediction of the resonance frequency due to the end effects of the nozzle and geometric shape differences from the conditions under which the original Helmholtz relationship was derived (1). Therefore it is not possible to accurately predict the amplitude oscillations in Pc due to breathing from these equations. It is not important to make such predictions in order to measure breathing volumes, however, and Reynolds and Frazer took a more empirical approach, relying on calibrating the Pc amplitude change using known perturbations in chamber volume.
Apparatus.
Our plethysmograph resembles the design proposed by Reynolds and Frazer (10) but is somewhat larger (
120-ml chamber volume for the neonatal mice considered here vs. 75 ml) and has a few important differences. Figure 1 shows our implementation, which differs from their original design in the following ways: 1) we do not use a micrometer to mechanically calibrate the apparatus but rather use a built-in calibration piston as described later, 2) with very small neonatal mice we do not use a barrier within the chamber to keep the mouse from interfering with the nozzle, and 3) we mount the driving speaker on rubber shock mounts to help reduce background noise.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1. The acoustic plethysmograph is constructed from a polycarbonate cylinder 20 cm in length, inner diameter 5 cm and wall thickness 7 mm. The nozzle has a diameter of 19 mm and a length of 23 mm from the inside of the chamber. The moveable piston is machined from oil-filled, ultra-high-molecular-weight polyethylene to slide snugly within the chamber, and in these studies was positioned to form a 6-cm-long space from the removable end cap that holds the microphone and the slave calibration cylinder. This cylinder is a 5-ml glass syringe filled with water and connected via a water-filled catheter to the master cylinder, a 50-ml glass syringe, also water-filled, operated by the programmable syringe pump. The speaker is mounted to the frame holding the chamber using rubber vibration absorbing mounts. The actual orientation of the microphone and slave cylinder is in the horizontal plane, above the level occupied by the neonatal mouse. Pe, excitation pressure applied by the speaker; Pc, measured acoustic pressure inside the chamber.
|
|
We use the same acoustic measurement equipment that Reynolds and Frazer (10) described (Larsen Davis 2530 microphone, 910B preamplifier and 2200C power supply, PCB Piezotronics, Depew, NY) to achieve a low-noise, high-precision sound pressure acquisition system. The speaker is a 3-W mini-speaker (Panasonic EAS-3P133B6, Digi-Key, Thief River Falls, MN), 30 mm in diameter, which is driven by the digital (D)-to-analog (A) output of our data acquisition system (National Instruments PCI-6281 18-bit A-to-D inputs with dual 16-bit D-to-A outputs, National Instruments, Austin, TX). The selection of the 18-bit device A-to-D converter over 16-bit resolution as used by Reynolds and Frazer was necessary to permit adequate resolution to measure the much smaller volume changes in the neonatal mice. Other modifications to permit these measurements include application of noise reduction procedures and use of a specially constructed acoustic chamber, as will be described.
We built a virtual instrument using LabVIEW (National Instruments) to acquire the acoustic signal from the microphone through an interface with a 1-kHz anti-aliasing filter (National Instruments SCC-68 and SCC-LP04). This instrument includes the capabilities of exciting the chamber at a selectable frequency and amplitude and of acquiring and analyzing the acoustic signal in real time to give a voltage output proportional to the breathing movements of the mouse. A power amplifier is used to drive the speaker from the D-to-A output signal to provide a sound pressure level of 103 dB. This would be stressful for many species, but mice do not hear below 1.5 KHz (3) and the neonatal mice used here have not developed their sense of hearing at this age (5).
The data-acquisition system samples the acoustic signal at 10 kHz, and this is too fast for the software to be able to acquire and process each sample individually in real time on our laboratory computer (Dell E520, 2.8-GHz Pentium, 2-GB RAM with Windows Vista, Dell, Round Rock, TX). Therefore, to obtain a real-time output, we have designed a batch-processing acquisition and analysis technique optimized for speed that provides a sample of mouse breathing at 100 Hz, nearly real time (delayed by 0.01 s). The breathing signal appears as a very small modulation (mV) of the excitation signal (V) and is obtained by measuring the variation in the peak-to-peak amplitude of the acoustic signal in each 10-ms signal sample as acquired. This technique differs from that of Reynolds and Frazer, who applied a Hilbert transform to the acquired signal postacquisition to demodulate the breathing signal from the total acoustic waveform. We compared the results of our 10-ms, peak-to-peak demodulation technique with the Hilbert demodulation approach using the same data records in postacquisition analyses and obtained results that were virtually indistinguishable. We apply additional smoothing to our 100-Hz estimates to minimize the influence of background acoustic noise, and this reduces the effective maximum frequency content to about 30–50 Hz, depending on how much filtering is required to obtain a reasonably smooth volume signal. Since our neonatal mice breathe at 3–5 Hz, this is an acceptable effective frequency response.
Reynolds and Frazer (10) used a precision micrometer to vary the chamber volume by known increments to provide calibration. Calibration of our system is provided by a 5-ml glass slave syringe mounted through the end cap of the chamber next to the microphone (Fig. 1). This slave syringe is water filled and driven through a water-filled system by a glass master syringe mounted in a programmable syringe pump (NE-500, New Era Pump Systems, Wantagh, NY) that delivers a train of step changes in volumes of programmable size and duration and is activated by the same virtual instrument that acquires the data and drives the speaker. We generally use a four- or five-step calibration sequence of 0.1-ml increases followed by a single retraction to the original starting point. The initial one to two steps are not useful because of backlash in the pump driver, but direct volume calibration by diverting the master cylinder outflow to a 1-ml syringe without a plunger shows the succeeding steps to be accurate. The calibration sequence can be applied continuously or intermittently as desired, and the volume calibration is obtained by quantifying the baseline shifts in the acoustic signal as the animal breathes. Because the measurement depends on the speed of sound in the chamber, it is essential to perform the calibration whenever chamber gas composition changes. We have noted a reduction in sensitivity by over 50% when the chamber gas was changed from room air to 7% CO2-50%O2-balance N2, for example. It is likely that changes in chamber temperature will affect the calibration also.
The perturbation in the acoustic pressure for a 20-µl breath causes a change of <0.5 mV in the 600- to 800-mV sinusoidal microphone signal from the acoustic system, so noise minimization is critical to be able to measure the small tidal volume of a P5 mouse (20–30 µl). We operate the acoustic plethysmograph inside a box specially constructed of structural soundproofing material (McMaster-Carr Acoustical Polyurethane Foam, 35 mm thick). A sliding Plexiglas door allows physical access and permits visual monitoring of the mouse in the plethysmograph. Other noise-minimization procedures will be described in the next section.
Methodology.
There is flexibility in setting the operating frequency for the plethysmograph by adjusting the initial position of the movable piston in the chamber. The chamber acts like an acoustic amplifier close to its resonance frequency, and it will amplify room noise in this range. It is therefore necessary to determine the ambient acoustic noise profile in order to pick an operating point away from relative peaks in the noise, even though these peaks are small in absolute terms. We use a separate virtual instrument to measure the noise spectrum in the chamber with the mouse inside and the power amplifier turned on. We always see peaks in the spectrum at 60 Hz and harmonics thereof due in part to a very small amount of hum from the power amplifier (Fig. 2). There is always a relatively broad peak around the resonance frequency of the chamber that we determine more accurately with the mouse installed using another LabVIEW virtual instrument to excite the chamber over a range of frequencies that includes the resonance point. Once the chamber resonance frequency is determined and the room noise spectrum is available, we select an excitation frequency for the speaker (the driving frequency, fd) that is 10–40 Hz above or below the resonance frequency (f0), with fd positioned to be at a low point in the room noise spectrum. Most often, although not always, the lowest noise operating point is at a frequency above resonance where inspiratory volume shifts cause an increase in the acoustic pressure. (Operating the chamber below the resonance frequency causes this relationship to be inverted.) If a suitably low noise operating point cannot be determined, we adjust the resonance frequency to a new value in a lower noise portion of the spectrum by physically moving the piston to change the chamber volume.
Selection of this operating point is important to minimize the noise, but the sensitivity of the measurement is not very dependent on fd over a span of 20–40 Hz as long as there is at least about 10- to 15-Hz separation from f0. We find stable calibration values over a range of fd values and usually obtain the same values after replacing one test mouse with the next one of similar size.
Once a suitable excitation frequency is established, the amplitude of the excitation wave is set to give a microphone signal no greater than ±0.8 V since the analog input range is configured to be ±1 V to maximize resolution. This provides a root mean square sound pressure level in the chamber of 103 dB, satisfactory for our purposes.
The volume output from the computation is processed in the acquisition virtual instrument to remove the mean using a long-interval running average (here, 3,000 samples = 30 s) and to amplify the breathing signal before it is sent to a second D-to-A channel output for use by other data-acquisition equipment in the laboratory. Thus the acoustic plethysmograph looks like an independent measurement apparatus that provides a breathing signal in synchrony with other devices such as gas analyzers, etc. Once the mouse is installed in the chamber and the excitation parameters are set, no further interaction with the acoustic measurement is required.
|
RESULTS
|
|---|
Experimental validation.
Although the calibration technique described above is sufficient, we validated the response of the device by placing a small, water-filled balloon inside the chamber and altered its volume manually using a 50-µl glass syringe to provide volume changes of approximately 10, 20, and 30 µl. Manual positioning was done as quickly as possible using physical stops to achieve each desired volume change; the achieved volume change frequency was, however, much lower than the 150- to 300-breath/min rate of our neonatal mice. Figure 3 shows the results for two syringe pump calibration cycles, the first with manual volume changes from 0 to 25 s, and the second with only the calibration pump operating to change chamber volume. The drifting signal between calibration steps is due to the process that removes the mean, necessary to provide an output amplitude suitable for most standard data-acquisition systems. We have marked in Fig. 3 the points where we measured baseline shifts in the volume signal attributable to the calibration signal and have indicated the calculated sensitivity from each calibration maneuver. When the manual volume change was not applied, the baseline noise level can be seen in the variation about this baseline; here, the background noise is about 2–3 µl. Figures 4 and 5 show the calibration results for manual volume changes of 10 and 30 µl, respectively. Infrequently, the ambient acoustic noise is sufficiently high that measurements with this low level of background variation are difficult.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 3. Two pump calibration cycles with nominal 20-µl volume perturbations applied manually as described in the text during the first cycle only. The pump calibration steps are 0.1 ml each, and the horizontal lines are placed to indicate a best estimate to the baseline shift caused by the pump perturbation given the continuous baseline correction process. The numbers next to the steps are the calibration values obtained from the voltage signal for each step. The drift in the step levels is due to the running adjustment of the output value to subtract the mean value, required to provide an output usable by typical data-acquisition systems. The mean volume perturbation (delta V) estimated for the applied pulses was 23.5 ± 3.2 µl.
|
|
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 4. One pump calibration cycle similar to those in Fig. 3 but with nominal 10-µl volume perturbations applied manually as described in the text. The mean volume change for pulses was estimated to be 12.6 ± 1.9 µl.
|
|
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 5. One pump calibration cycle similar to those in Fig. 3 but with nominal 30-µl manual volume changes. The mean volume for these pulses was estimated at 31.5 ± 3.5 µl.
|
|
The volumes produced by these calibration maneuvers were acquired using standard procedures (DataPac 2K2 data acquisition and analysis software, Run Technologies, Mission Viejo, CA). The calibration factors determined from the 0.1-ml volume steps are shown in these figures next to each step and were used to quantify the acoustic pressure responses to the manual changes. Although it appears from calibration factors that there is a trend toward decreasing sensitivity as the calibration advances, this is not always seen and probably depends on the specific operating conditions, especially the separation between the resonance and driving frequencies. For our purposes here, we use the average of calibrations to evaluate the measured responses to the manual volume perturbations although more complex approaches could be used such as using the value for the closest calibration maneuver.
Figure 6 shows the measured calibration volumes vs. the nominal values applied. The volumes obtained from analysis of the manual changes were all quite close to the nominal changes but were all greater than the expected values as shown by comparison with the identity line. The manual maneuver used the bottom of the 50-µl syringe as a physical limit, and it was apparent that the end of the stroke went a small amount below the zero calibration mark; thus a volume change that started at the 20- µl mark would actually produce a somewhat larger volume change. The volume errors (difference between nominal and measured volumes) for each of the 10-, 20-, and 30- µl maneuvers are 2.6, 3.5 and 1.5 µl, respectively, and at least some of these errors are likely due to the true perturbations being somewhat greater that the nominal values.
View larger version (9K):
[in this window]
[in a new window]
|
Fig. 6. Comparison of the measured volume changes to the nominal values for the calibration maneuver. Plotted here are the measured means ± SD for each of the test volume changes applied to the calibration balloon inside the chamber. Since we could not measure the actual test volume changes, the plot here shows the values intended from setting the calibration syringe mechanical stop. As noted in the text, the actual applied volumes were likely to be somewhat greater than the nominal values.
|
|
An ideal calibration would involve simultaneous measures of mouse breathing using the acoustic technique and another standard approach. To our knowledge, this is not possible in these mice in which breathing is most commonly measured using barometric plethysmography. The small dimensions and principles of operation of the acoustic chamber do not permit other techniques: for example, we cannot incorporate a whole body plethysmograph inside the acoustic chamber since its rigid body enclosure would not permit breathing to alter the volume of gas in the acoustic chamber. We decided to measure breathing in small mice and to compare the results with expected values from the literature (8).
Figure 7 shows breathing in a 4.73-g, 6-day-old mouse. The calibration factors are shown, and analysis of the 115 breaths unaffected by the calibration steps gives a mean tidal volume of 0.0445 ml. Figure 8 shows 56 breaths measured with the calibration pump turned off. The ventilatory values measured here are well within those expected for this size mouse (8). Figure 8, inset, shows the first 5 s of the same data in order to show the effect of the instrument smoothing in detail. This figure shows how the acoustic measurement tracks thoracic volume and reflects variations in end-expiratory as well as end-inspiratory volume. Table 1 shows the mean tidal volumes measured in a set of six newborn mice ranging in age from P5 to P7. These values can be compared with those given by Mortola (Ref. 8, Fig. 2.15, p. 68), and this comparison indicates the values from the acoustic plethysmograph are quite comparable to expected values.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 7. Tracing from postnatal day 6 (P6) mouse 0710P6C showing breathing and calibration information. The numbers show the calibration factors obtained for each calibration step (0.1 ml) from the voltage data. As in the previous figures, the baseline trends reflect the auto-zero process required to keep the data in range for normal data-acquisition equipment (see text). VT, tidal volume.
|
|
View larger version (38K):
[in this window]
[in a new window]
|
Fig. 8. More data from the mouse of Fig. 7 with the calibration pump turned off. Note that there is no baseline drift since the auto-zero process stabilizes once the volume calibration shifts cease. Inset: an expanded version of the first 5 s of the data to demonstrate the temporal nature of the breathing signal. The values for mean VT, frequency (f), and minute ventilation (Vdot) for the 56 breaths in this 20-s run are shown. B/min, breaths/min.
|
|
View this table:
[in this window]
[in a new window]
|
Table 1. Acoustic plethysmograph results from 6 neonatal mice (C57BL6 with FOS-TUA-LacZ possible genetic modification)
|
|
|
DISCUSSION
|
|---|
Accurate measurement of breathing in small rodents is important for many reasons. Rats and mice show developmental changes in cardiorespiratory control processes that may reflect human infant development, and it is useful to examine how exogenous factors such as prenatal exposure to ethanol or cigarette smoke may affect this development. Mice can be genetically modified to interfere with this development of cardiorespiratory control in ways that may correlate to human diseases such as congenital hypoventilation (9, 11) or SIDS (4), for example. Adequate cardiorespiratory phenotyping requires measurements that are accurate, unaffected by the measurement procedures, and relatively efficient to perform to characterize population subsets.
Evaluation of the development of respiratory control obviously depends on accurate measurement of respiratory volume and timing, and this is true for estimation of respiratory mechanics measurements to determine the responses to environmental stresses. Problems with current whole body barometric plethysmography in small rodents have been described and examined in detail by others (e.g., Refs. 2, 6, 7) and will not be exhaustively reviewed here. One basic problem relates to the inadequate separation of volume-related from flow-related contributions to box pressure changes in animals with a high respiratory frequency. The acoustic plethysmographic approach does not avoid an artifact due to decompression of thoracic gas during airflow. A reviewer of this manuscript raised this issue, and we conclude that any technique that estimates the true inspired volume (the value that would have been measured by a displacement spirometer connected to the airway) using movement of the chest wall will be fooled when the chest wall does not accurately represent this volume change. Here, it is the chest wall movement with breathing that directly alters the resonating volume and is the signal measured. Therefore, if the thoracic gas is decompressed during inspiration, the chest wall signal will overestimate the volume that the spirometer would have measured anytime that flow is present. If we knew when the airflow was zero at the end of the breath, we could measure the volume change without artifact at that time, but this is not known.
An important advantage of the acoustic approach is that the signal does not change in character as the difference between the animal's body temperature and the environmental temperature decreases, so the acoustic technique is useful for measuring ventilatory responses in such conditions where the barometric signal disappears. The acoustic technique will, for example, measure accurately the chest wall volume changes when a mouse is cooled to room temperature, a condition where the barometric method can only measure the decompression signal correlated to airflow. The calibration will, of course, change as the physical characteristics of the chamber gas change with temperature (the speed of sound c0 in Eqs. 2 and 3 is a function of temperature), but the calibration approach proposed here will automatically account for this. Similarly, changing the gas composition to apply hypercapnia or hypoxia will alter the physical characteristics of the chamber gas, but the continuous calibration we use will similarly accommodate for this to give accurate volume measurements. The continuous, repeatable calibration scheme as implemented here removes any variation in the calibration due to changes in the rate at which it is applied.
In summary, we have extended the Reynolds and Frazer (10) approach to be useful in neonatal mice with tidal volumes <10% that of the adult mouse. Three advances we propose here to accomplish this are: 1) the continuous calibration scheme with the programmable syringe pump to facilitate using this approach in various conditions of temperature and gas composition, 2) the noise reduction techniques that include careful positioning of the acoustic operating point of the plethysmograph to minimize ambient noise amplification by the chamber, and 3) provision of a volume output in real time for use with other instrumentation. Although technical reasons prohibit validation of the technique by comparing simultaneous acoustic measurements with results obtained using another standard technique, we have shown that calibration by a balloon substitute is quite accurate and that measurements in newborn mice return values consistent with expectation. We anticipate that the acoustic approach will become an important tool in the measurement of breathing in unrestrained neonatal and developing mice to evaluate the influence of genetic modifications on respiratory control.
|
GRANTS
|
|---|
This research was supported in part by National Institutes of Health Grant 5-P01-HD-036379.
|
ACKNOWLEDGMENTS
|
|---|
The excellent technical support and efforts of Wilbur Clark and Chris Schumann are gratefully acknowledged. We thank Donald Bartlett for enlightening discussions on whole body plethysmography. We are also grateful for the very helpful advice and suggestions of Matt Sweetland of National Instruments.
|
FOOTNOTES
|
|---|
Address for reprint requests and other correspondence: J. A. Daubenspeck, Dept. of Physiology, Borwell Bldg., Rm. 756E, Dartmouth Medical School, Lebanon, NH 03756 (e-mail: andy.daubenspeck{at}dartmouth.edu)
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES
|
|---|
- Alster M. Improved calculation of resonant frequencies of Helmholtz resonators. J Sound Vibr 24: 63–85, 1972.[CrossRef]
- Enhorning G, van Schaik S, Lundgren C, Vargas I. Whole-body plethysmography, does it measure tidal volume of small animals? Can J Physiol Pharmacol 76: 945–951, 1989.[CrossRef]
- Fay RR. Comparative psychoacoustics. Hearing Res 34: 295–306, 1988.[CrossRef][Web of Science][Medline]
- Hodges MR, Chen Z, Deneris E, Johnson RL, Richerson G. Adult mice with 5-HT neuron-specific knockout of Lmx1b exhibit an attenuated hypercapnic ventilatory response (Abstract). FASEB J 20: A785, 2006.[Free Full Text]
- Kamiya K, Takahashi K, Kitamura K, Momoi T, Yoshikawa Y. Mitosis and apoptosis in postnatal auditory system of the C3H/He strain. Brain Res 901: 296–302, 2001.[CrossRef][Web of Science][Medline]
- Lai-Fook SJ, Lai Y. Airway resistance due to gas compression measured by barometric plethysmography in mice. J Appl Physiol 98: 2204–2218, 2005.[Abstract/Free Full Text]
- Lundblad LK, Irving CG, Adler A, Bates JH. A reevaluation of the validity of unrestrained plethysmography in mice. J Appl Physiol 93: 1198–1207, 2002.[Abstract/Free Full Text]
- Mortola JP. Respiratory Physiology of Newborn Mammals: A Comparative Perspective. Baltimore, MD: Johns Hopkins Univ. Press, 2001.
- Renolleau S, Dauger S, Vardon G, Levacher B, Simonneau M, Yanagisawa M, Gaultier C, Gallego J. Impaired ventilatory responses to hypoxia in mice deficient in endothelin-converting enzyme-1. Pediatr Res 49: 705–712, 2001.[Web of Science][Medline]
- Reynolds JS, Frazer DG. Unrestrained acoustic plethysmograph for measuring tidal volume in mice. Ann Biomed Eng 34: 1494–1499, 2006.[CrossRef][Web of Science][Medline]
- Shirasawa S, Arata A, Onimaru H, Roth KA, Brown GA, Horning S, Arata S, Okumura K, Sasazuki T, Korsmeyer SJ. Rnx deficiency results in congenital central hypoventilation. Nat Genet 24: 287–290, 2000.[CrossRef][Web of Science][Medline]
Copyright © 2008 by the American Physiological Society.
TOP
ABSTRACT
METHODS
