Paradoxical response of pulmonary slowly adapting units during constant pressure lung inflation

Abstract

Typically, unit discharge of slowly adapting receptors (SARs) declines slowly when lung inflation pressure is constant, although in some units it increases instead—a phenomenon hereinafter referred to as creeping. These studies characterize creeping behavior observed in 62 of 137 SAR units examined in anesthetized, open-chest, and mechanically ventilated rabbits. SAR units recorded from the cervical vagus nerve were studied during 4 s of constant lung inflation at 10, 20, and 30 cmH₂O. Affected SAR units creep more quickly as inflation pressure increases. SAR units also often deactivate after creeping, i.e., their activity decreases or stops completely. Creeping likely results from encoder switching from a low discharge to a high discharge SAR, because it disappears in SAR units with multiple receptive fields after blocking a high discharge encoder in one field leaves low discharge encoders intact. The results support that encoder switching is a common mechanism operating in lung mechanosensory units.

INTRODUCTION

Information from airway sensors is mainly carried in the vagus nerves for regulation of breathing and airway defense. Conventionally, airway sensors are divided into chemosensors and mechanosensors. The former are nonmyelinated bronchial and pulmonary C fibers, whereas the latter are myelinated mechanosensors (1–3). Regarding mechanosensors, investigators have adhered to a doctrine of one-sensor theory, i.e., one afferent fiber connects to a single sensor. Accordingly, lung inflation activates two types of mechanosensors: slowly adapting receptors (SARs) and rapidly adapting receptors (RARs). Recently, with advances in sensory physiology and morphology, it is recognized that a single-airway mechanosensory unit can contain numerous sensors with four different types, i.e., an afferent fiber connects to multiple heterogeneous or homogeneous sensors (multiple-sensor theory) (4). In addition to SARs and RARs, there are deflation-activated receptors (DARs), which can slowly adapt (dSARs) or rapidly adapt (dRARs). Each type senses a specific force and generates a unique response. Thus, sensor composition determines unit behavior. For example, RAR units may respond to deflation if they house DARs responsible for the Hering–Breuer deflation reflex. Multiple-sensor theory is a conceptual shift. It requires a description of how a sensory unit operates in response to stimuli by examining its discharge pattern. A mechanosensory unit model has been proposed to describe how the sensors interact within the unit (5). Airway mechanosensors are often examined during lung inflation. Here, we found some SAR units increased the activity, rather than the expected gradual decrease due to adaptation (please see the last 8 s of lung inflation in Fig. 5D of Ref. 6). Hereinafter this phenomenon is referred to as creeping. Although this puzzling phenomenon has been observed in several species: opossums, dogs, cats, guinea pigs, and turtles (7–10), only one paper discusses it as a side finding (7). Farber et al. inflated opossum lung to 10 cmH₂O constant pressure: 22.5% of the SAR units observed increased their activity. In view of its conserved behavior across species, SAR creeping appears to warrant further study. We retrospectively examined SAR unit behavior during constant pressure inflation at 10, 20, and 30 cmH₂O. Surprisingly, more than 45% demonstrated creeping. Closer examination of their behavior suggests creeping results from encoder switching or pace-maker switching. The mechanism of encoder switching has been reported in airway and other mechanosensors (11–13).

METHODS

This is a retrospective study reviewing vagal pulmonary afferent (SAR unit) behavior in anesthetized, open chest, mechanically ventilated rabbits (male, 1.8–2.4 kg body wt). SAR units recorded from June 2002 to September 2004 were included for data analysis. One hundred and thirty-seven pulmonary SAR units recorded in 52 rabbits were reviewed. The rabbits were intravenously anesthetized with 20% urethane (1 g/kg). The trachea was cannulated low in the neck and the chest was opened widely in the midline. The lungs were ventilated by a Harvard ventilator (model 683): positive end-expiratory pressure (PEEP) was maintained by placing the expiratory outlet under 3–4 cmH₂O. Electrical activities of SAR units were recorded with the conventional single unit recording technique. Details on recording the sensory afferent have been previously reported (6, 14). Briefly, the vagus nerve (either right or left) was sectioned and its peripheral end was separated from the carotid sheath, placed on a dissecting platform, and covered by mineral oil. A small slip was cut from the main vagus nerve and placed on recording electrodes. The electrodes were connected to a high-impedance probe (Grass HIP5) from which the output was led to an amplifier (Grass P511). After suitable amplification, action potentials from single fiber strands of the vagus nerve were displayed on an oscilloscope and monitored by a loudspeaker. In addition, a voltage analog of impulse frequency was produced by a ratemeter (Frederick Haer, Brunswick, ME) at a binwidth of 0.1 s. The single unit was ensured by a uniformity in amplitude and contour of action potentials displayed in the oscilloscope. Action potentials and analog frequency along with blood pressure and airway pressure were recorded by a thermorecorder (Astro-Med; Dash IV). The location of the receptors was determined by gently exploring the external surface of the lungs with a cotton tip to ensure the sensors are from the lung. No attempt was made to locate endings more precisely. Sensory activities were determined at three different levels of constant pressure of lung inflation, i.e., 10, 20, and 30 cmH₂O. The inflation pressures were applied randomly. All studies conformed to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Publication No. 85-53) and were approved by the Institutional Animal Care and Use Committee at University of Louisville and the Robley Rex Veterans Affairs Medical Center.

RESULTS

Among the 137 units examined, 62 units (45%) showed evidence of creeping (increased activity in response to constant pressure lung inflation). Sensory activity increases of more than 20 imp/s from the initial activity and lasting more than a second are considered as creeping. Overall SAR discharge frequency was directly related to lung inflation pressure, and was higher in creeping than in noncreeping units (Table 1). Three basic creeping patterns were observed: Gradual (Fig. 1, C–E); Abrupt (Figs. 2 and 3B); and In-combination, i.e., first gradual and then sudden (Fig. 3A). Creeping is often followed by deactivation during prolonged inflation (Fig. 1E). Creeping initiates more quickly as lung inflation pressure increases in a given SAR unit (Figs. 2 and 4). The same relationship appears among different sensory units, showing greater initial and peak discharge rates (Fig. 5). Among the three types of creeping, gradual is the most commonly observed creeping. For example, at 30 cmH₂O, 38 (69%), 10 (18%), and 7 (12%) of 55 units showed gradual, abrupt, and in-combination patterns.

Table 1.

Creeping versus noncreeping SAR inflation-related discharge frequencies

	Inflation Pressure, cmH2O	Discharge Frequency, Imp/s*
Creeping (n = 62)	10 20 30	72.8 (3.6) 124.9 (5.3) 165.6 (7.4)
Noncreeping (n = 75)	10 20 30	50.1 (3.4) 80.9 (4.6) 101.0 (0.3)

Unit discharge frequency at different levels of lung inflation. Analysis of covariance indicates the difference between creeping and noncreeping groups is significant, P < 0.0001. SAR, slowly adapting receptor.

*Values are mean (SE).

Figure 1.

A gradual pattern of creeping. Inflating the lung to 10 (A), 20 (B), 30 (C), and 40 cmH₂O (D and E), the slowly adapting receptor (SAR) unit starts to increase discharges indicated by the arrow. At high pressures 30 and 40 cmH₂O (C–E), the unit exhibits a gradual increase in activity (creeping). The pronounced increment is demonstrated by a steeper slope, which is more pronounced at 40 than 30 cmH₂O (comparing C with D and E). Prolonging the lung inflation, the unit deactivates after creeping (E), indicated by the second arrow.

Figure 2.

An abrupt pattern of creeping. In contrast to the gradual increase (Fig. 1), activity of this unit increases suddenly. Please also note that the creeping occurs earlier as the lung inflation pressure increases. The slowly adapting receptor (SAR) unit starts to increase its activity at 2.5 and 4 s at lung inflation pressures of 30 and 20 cmH₂O, respectively.

Figure 3.

A combination pattern of creeping (A). The unit creeps at two levels of dynamic lung compliance (Cdyn): (A) low Cdyn (indicated by a high airway pressure swing), this sensory unit has an initial gradual increase with a delayed abrupt increase in activity. This is probably either due to opening up the airway or gradual activation of the sensor, (B) high Cdyn, only a sudden increase in activity was observed with an initial activity being flat, indicating the sensor located airway is fully opened. It is interesting to note that the unit discharge frequency is about the same at the sudden creeping point and at the peak.

Figure 4.

Line plot showing creeping occurs earlier as the lung inflation pressure increases. Twenty-seven units show this effect at 20 and 30 cmH₂O lung inflation (2 units at 10 cmH₂O). This relationship exists in all except two units, one increases (arrow 1) and one does not change (arrow 2) in the time to the starting point.

Figure 5.

Interunit relationships between slowly adapting receptor (SAR) unit activity and creeping: Ordinate is time to creeping onset and abscissa peak discharge frequency (A) or initial discharge frequency (B). There is a negative correlation between the two variables, i.e. earlier creeping onset associates with greater peak and initial discharge frequency (P < 0.05 in both). Data are pooled from lung inflation at both 20 and 30 cmH₂O with a total of 94 units.

The following data support creeping results from encoder switching. The unit in Fig. 6 has multiple receptive fields and is activated with low discharge frequency with the lung inflated to 10 cmH₂O. At 20 cmH₂O, unit activity increases initially, and then shows double creeping. The second creep has a rapidly adapting component. At 30 cmH₂O, unit activity increases again without creeping. Interestingly, it also has a rapidly adapting component in the initial segment as the third encoder shown at 20 cmH₂O. Furthermore, we found multiple receptive fields of the unit in the lung. The double creeping can be blocked by local injection of lidocaine (2%, 20 µL) into the receptive fields. This phenomenon supports that three different sensors are responsible for the three levels of activity, respectively. Clearly, the higher activity at 30 cmH₂O overrides other encoder activity, that is, there is no creeping. Although Fig. 6 shows abrupt creeping effects can be blocked, Fig. 7 shows gradual creeping effects can also be blocked by local injection of lidocaine into one of the receptive fields.

Figure 6.

A double creeping (a unit shows 3 encoders with 2 stepwise increases in activity). A: when the lung is inflated to 10 cmH₂O, the unit discharges with a sustained low frequency. B: at 20 cmH₂O, there is a two-step increase in activity (i.e., two consecutive creeping, indicated by two arrows on top of the figure). C: there is no creeping at 30 cmH₂O. D: a repeat at 20 cmH₂O (please see the double creeping). E: after blocking 1 receptive field with 20 µL of 2% lidocaine (the highest discharging encoder was eliminated). F: after further blocking a second receptive field, the unit left with the lowest discharging encoder.

Figure 7.

Another example of how creeping effects can be eliminated by blocking a receptive field with 2% lidocaine (20 µL). A–C are 10, 30, and 30 cmH₂O lung inflation before; D–F are repeats after lidocaine blockade. Please note that at lung inflation to 30 cmH₂O (B and C), there was gradual creeping, which disappeared after lidocaine blockade. Instead of creeping, the unit deactivated (F). The deactivation can also be found in C, indicated by the arrow.

Creeping due to activation of a new encoder can also be supported by careful examination of unit behavior (Fig. 8). The creeping phenomena occur in both lung inflation-activated receptors and DARs (Fig. 9). This suggests the sensory properties of these different types of sensors are similar and creeping is common for airway mechanosensors.

Figure 8.

Creeping is due to activation of a new encoder. On the top of each figure, 1 denotes the low discharge encoder, 2 denotes the high discharge one. A: after a period of lung inflation at 30 cmH₂O, creeping occurs, indicated by arrow 2, which returns to its original level after inflation pressure was reduced to 15 cmH₂O due to deactivation (see arrow 1). B: with a short inflation period at 30 cmH₂O, the unit discharge did not increase (no creeping). C: repeating A. D: after cyclic increase in airway pressure, the unit activity increases gradually (creeping). Shifting the cyclic ventilation into a constant pressure inflation during the creeping, the high discharge encoder deactivates (activity switching from the encoder 2 to the encoder 1). As the cyclic ventilation resumes the low frequency encoder operates. Thus, it proves that activation of a new higher discharge encoder is responsible for creeping. E: short period manipulation with no creeping (only the encoder 1 operates). Please see question 4 for additional explanation.

Figure 9.

Creeping also occurs in the deflation-activated receptor (DAR). This unit has rapidly adapting and slowly adapting components during lung inflation to 30 cmH₂O. When the lung was deflated to −6 cmH₂O, a slowly adapting DAR was activated. Approximately 1 s after the activation, the unit creeps.

DISCUSSION

In the current study, the long-neglected, frequently occurring phenomenon of SAR unit creeping is characterized. Creeping is time- and stimulation strength-dependent. Our data suggest it occurs as the result of encoder switching (activation of a new encoder). Since creeping is common and occurs in different species, it must have an important physiological function.

Creeping frequently occurs in different species. In our previous studies in rabbits, some SAR units exhibited a gradual increase in activity after a period of constant lung inflation (6). The present study further demonstrates that 45% of the observed SAR units showed creeping. In other work, Farber et al. applied 10 cmH₂O constant airway pressure in opossums and found 9 out of 40 SARs (22.5%) increased activity (7). This occurrence rate is probably underestimated because lung inflation was limited to 10 cmH₂O (7). In the present study, the lung was only inflated for 4 s. With increased stimulation strength in Farber’s study or time in our study, the occurrence rate should be much higher. In Farber’s study, they mentioned that creeping exists in dogs, too (7). Indeed, creeping can be identified in the processed data in dog studies (Figs. 2 and 3A of Davenport et al.) (15). This can be demonstrated by increased SAR activity during maintained stimulation. Elsewhere, Karla et al. (8) demonstrated creeping in cats. It is worth noting that both the rabbit and cat were used in their study. Since RARs had been studied only in the cat, and Karla’s Fig. 4 included a RAR unit, the creeping observed is from the cat. Creeping appears in both of their figures as the lungs were inflated to 20 cmH₂O (8). Creeping is found also in the guinea pig [see Fig. 1 in Bergren’s review (9)]; and in turtles [judging from increased activities from processed group data, see Figs. 2 and 4 of Ref. (10)] during constant pressure inflation of the lung. Again, the single figure in Bergren’s (9) review article demonstrates creeping when the lungs were inflated to 40 cmH₂O. Thus, creeping occurs often and especially at high inflation pressure.

Creeping occurrence relates to sensory unit activity. The higher the unit activity, the higher the creeping occurrence rate and the earlier the unit starts to creep. For example, in a given unit, there is earlier onset as inflation pressure is increased (Fig. 4). This relationship also exists among different units, i.e., units with higher basal activity creep earlier (interunit relationship; Fig. 5). This relationship is also supported by the fact that the creeping units have higher basal activity than the noncreeping units (Table 1). Furthermore, this relationship may exist among different species (interspecies relationship). For example, Farber et al. (7) noticed that both basal activity in SARs and the occurrence rate of creeping are higher in opossums than in dogs.

Creeping develops due to encoder switching. Since SARs are stimulated by smooth muscle contraction (16), which can be induced by inspiratory drive output (17), Farber et al. (7) believed that creeping results from increased airway smooth muscle tone. Because bilateral vagotomy did not block creeping, they dismissed the possibility of a reflex contraction of smooth muscles and thought that lung inflation directly provoked smooth muscle contraction to activate the sensors. However, creeping can occur abruptly but smooth muscle tension cannot. Furthermore, this smooth muscle theory cannot explain that when inflation pressure increases, creeping may occur earlier, delay, or disappear. On the other hand, Figs. 6, 7, and 8 strongly support that encoder switching is the responsible mechanism for creeping, because it disappeared after blocking a high discharging encoder (Figs. 6 and 7), or without activating the high discharging encoder (Fig. 8). Thus, creeping is likely due to activation of an inactive encoder.

Creeping occurs in three different patterns. Gradual, abrupt, and combination patterns were found in the rabbit (Figs. 1, 2, and 3). They have also been observed in other species; for example, as combination and abrupt patterns [Figs. 4 and 5 in Ref. (8)] in the cat, and a gradual pattern in opossum (7) and guinea pig (9), which was followed by deactivation. At this point, we do not know why different patterns exist. However, the same mechanism, activation of a new encoder, is responsible for either abrupt (Fig. 6) or gradual (Figs. 7 and 8) creeping. The question remains why there are abrupt increases in some cases and gradual increases in others. It is likely that abrupt increase in activity represents full activation of an encoder, whereas gradual increase represents a transition process, such as a gradual decrease in threshold, or a gradual increase in sensitivity of the unit to the stimulus, or due to opening up the airway. As lung compliance decreases it takes longer to start creeping and to reach the peak. Although this remains unanswered, it is safe to say that a mechanical effect contributes to the creeping pattern. It is interesting to note that the unit discharge frequency is about the same at the sudden creeping point (Fig. 3).

Creeping often precedes sensory deactivation. We observed creeping followed by deactivation in many cases (Fig. 1E). This is also observed in the guinea pig [please see Fig. 1A, Bergren’s review (9)]. The reason is unclear. However, deactivation occurs as a result of overexcitation of the unit, either caused by high strength of mechanical stimulation (18) or by administration of Na⁺-K⁺-ATPase blocker ouabain (19, 20). Creeping causes the unit to discharge at a high frequency, which may lead to deactivation. We believe that both creeping and deactivation are caused by a same mechanism, that is, encoder switching. In creeping, the encoder switches to a higher discharging one, whereas in deactivation the encoder switches to a lower discharging one.

Creeping also occurs in DARs (Fig. 9). This important piece of information indicates that such properties are common not only to SARs but also to DARs and possibly other mechanosensors. Sensory activity is a dynamic process. A sensory unit possesses multiple heterogeneous sensors. Interaction among the sensors at any instant generates the sensory unit output. Multiple sensors provide a variety of behaviors of each unit. Our data support that encoder switching is a common mechanism for interaction of sensors in the unit. Using multiple-sensor theory, we are able to explain creeping phenomena, which cannot be explained by one-sensor theory. Revealing creeping behavior in turn lends strong support to the multiple-sensor theory (21). Clearly, this mechanism extends the sensory response range and guarantees the transmission of the information through a redundant mechanism.

Some important questions were raised during review that may be of interest to other investigators and need specific answers. These are discussed as follows in question-answer format.

Question 1. It seems that pressure vibrates when a high inflation pressure is maintained, which appears correlated to creeping (Fig. 1, C–E, and Fig. 2). In addition, individual mechanosensors might be stretched and/or pressed differently at different time, depending on the status of surround lung tissue’s physical location, posture, folding, closure, etc. Could these factors be a potential cause for creeping?

Answer 1. Yes, the pressure waveform is not absolutely flat, having some very subtle fluctuations. Personally, I don’t believe these fluctuations cause creeping, because they also occur in its absence (Figs. 3 and 8). Furthermore, fluctuation exists at low inflation pressures (Fig. 1, A and B) with no creeping. Our data indicate that creeping is pressure-related, but not to these subtle changes. Second, the fluctuations are not absolutely contemporaneous with creeping in these figures. Activities in SAR units follow the pressure waveform. A minute increase in pressure would induce a minute increase in neural activity only, but not a significant one. On the other hand, if creeping is due to activation of an inactive sensor, then this fluctuation is not prerequisite for creeping. However, it is possible minute alterations in the system may trigger creeping. Similarly, the other factors mentioned earlier are not prerequisite, but may trigger or modify creeping, i.e., they are potential contributing factors.

Question 2. Creeping seems accompanied with action potential amplitude and/or form changes (see IMP panels in Figs. 1, 2, 3, and 6), would you explain?

First, these action potentials are from a single unit recording, which can be verified from their regular discharges and smooth changes in the figures. This phenomenon occurs when the unit discharges at a very high frequency (21). They are probably due to reduced positive potential whereas negative potential increased, leading to a tendency of downward movement. The excursion of the action potential alters because of the redistribution of ions, which is electrogenic. Na⁺ enters and K⁺ exits from the axon. Such a phenomenon is not only observed during constant pressure inflation of the lung, but also during cyclic inflation (Fig. 8D, the four cycles before the second 2).

Question 3. It is implied that DARs are mutually exclusive from SARs and RARs? Is there a subpopulation of SARs or RARs that is sensitive to deflation?

Answer 3. This is a frequently encountered important question, which has been dealt with in relative detail (21). This possibility can neither been excluded nor proved by available techniques because many sensors are clustered together and cannot be differentially blocked. In conventional one-sensor theory, SARs or RARs may respond to both inflation and deflation, but in multiple-sensor theory, SARs and RARs operate separately from DARs. Lung inflation- and deflation-induced activities in a single sensory unit can be selectively abolished by injection of local anesthetics into their respective receptive fields (22), supporting multiple-sensor theory. Furthermore, deflation responsive SAR units can be stimulated by hypertonic saline, but not pure SAR units (i.e., the units do not respond to lung deflation) (23). This indicates DARs are physically separate from SARs, although they share the same axon. This may also explain that most RAR units can be stimulated by hypertonic saline, because most RAR units also respond to lung deflation (i.e., those containing DARs): others do not (24). Taken together, this information supports inflation and deflation sensors are mutually exclusive. The most difficult question for one-sensor theory is that how a sensor can respond to forces in opposite directions. Sant’Ambrogio believed only SAR units in the tracheal back membrane can respond to both inflation and deflation because the tracheal membrane can be mechanically stretched in both directions. For this very reason, he commented on mechanosensory units in the lung periphery responding to lung deflation: “The circumstances of activation of these receptors are obscure” (25). This is because it is mechanically unexplainable. Multiple-sensor theory is supported by many SARs with different subtypes [type I and type II (6); phasic high threshold and tonic low threshold (13)]; or even RARs and SARs (22) found in a single sensory unit and by morphological studies showing a axon may supply to many sensors (26). Furthermore, multiple-sensor theory explains many sensory unit behaviors that one-sensor theory cannot. For example, it can explain a RAR unit can respond to lung deflation with slowly adapting behavior and vice versa. It explains the existence of intermediate adapting receptors and the dynamic property of SARs (4). It also explains a sensory unit can have different discharge frequencies under the same mechanical condition, due to pace-maker switching. The unit switches from a high discharge sensor to a low discharge sensor [deactivation; (6, 18)], or from a low discharge sensor to a high discharge sensor (creeping; the present study). Putting these pieces of information together, I believe it is time for us to accept the multiple-sensor theory (4, 27).

Question 4. How the low discharge encoder “1” and high discharge encode “2” are designated in Fig. 8. For example, why “1, 2, 2, 1” instead of “1, 2, 1, 1” in Fig. 8C? The impression from Fig. 8, A and B is that any action potentials in responding to 15 or 30 cmH₂O without triggering creeping are supposedly assigned “1”? In addition, the rationale for “2”s and “1”s in Fig. 8D seems not obvious to general audience.

Answer 4. Yes, activities without creeping are designated 1. The unit in Fig. 8 adapts very slowly—almost not at all. Thus, its activity exactly follows the airway pressure waveform. In Fig. 8B, the unit did not creep, thus its activity serves as a perfect control for the low discharge encoder 1. Any activity above that level will be from the high discharge encoder 2. In Fig. 8C, when the lung inflates to 15 cmH₂O, the activity is from encoder 1. Then, at 30 cmH₂O, the unit starts to creep (i.e., encoder 2 takes over), indicated by the first 2. When the pressure returns to 15 cmH₂O, encoder 2 still serves as the pacemaker, because the activity is higher than encoder 1. Therefore, it is designated encoder 2. The last arrow 1 indicates the time point encoder 1 starts serving as the pacemaker, because the activity returns to encoder 1 level. Regarding Fig. 8D, the peak activity denoted by the first 2 is little bit bigger than the previous ones, while the peak airway pressure is the same. In the subsequent three breaths, while the pressure starts to drop slightly, the unit activity continues to rise. This is creeping (SAR activity should be dictated by airway pressure) because the activity is more than that of encoder 1. This can be verified by the high activities (paced by encoder 2) at the subsequent 15 and 30 cmH₂O. The unit activity drops to the low level (paced by encoder 1) as encoder 2 deactivates in the middle of lung inflation at 30 cmH₂O. Thus, encoder 1 was designated as the pacemaker (indicated by the 1).

Question 5. What is the physiological function of creeping?

Answer 5. This is an excellent important philosophical question. Creeping is newly identified. And while we are not quite sure about its function, it can extend the sensory domain. Since creeping can only be explained by multiple-sensor theory, hypothesizing within that framework, many sensors interact within a sensory unit. Each has its own threshold pressure, saturation pressure, and sensitivity—in other words, its own operating domain (afferent properties). Therefore, all sensors respond to the strength and profile of incoming stimuli. Overall, the unit responds in a unified manner, i.e., at any time point, only one pacemaker drives the unit (5). Such sensor interaction explains that a lower threshold SAR would be first activated, and then on inflation toward peak pressure, a high discharge frequency sensor may operate. It also explains RAR-SAR interaction (4). The creeping phenomenon tells us afferent properties are not fixed but dynamic; they may change in time domain and are influenced by stimulating profiles (please also see question 1 and the section of Creeping also occurs in DARs in discussion), which may alter the pacemaker. This highlights our knowledge of subcellular operating mechanisms in sensory neurons remains limited.

Question 6. In Fig. 7, why did the unit deactivate in F, but not in others?

Answer 6. In Fig. 7, the A–C is before and D–F is after local anesthetic blockade of the creeping sensor. Comparing E and F, the inflation pressure is slightly higher in F. Since the deactivation is directly related to the inflation pressure, a higher pressure may explain it. Regarding B and C, the sensor also deactivated in C. This can be discerned by a complete stop of activity before the airway pressure released, which is indicated by the arrow. There is no explanation why in C but not in B the sensor deactivated. It is very likely that the sensor in B is about to deactivate were inflation held a little bit longer. Although the response is very repeatable, activation and deactivation are not exactly the same during each stimulation.

In summary, creeping is a common behavior in different species. Based on our analysis, encoder switching—a common operating mechanism in sensory units—is responsible.

Perspectives and significance. Bronchial pulmonary mechanosensory units play important roles in breathing control and homeostasis. Conventionally, investigators adhere to one-sensor theory when interpreting the mechanosensory function. More recently, a multiple-sensor theory has been proposed. Heterogenous sensors may share an afferent axon to convey various types of information to the central nervous system. The current study demonstrates a common creeping phenomenon and provides appealing evidence it results from multiple sensor interaction. The fields of mechanosensation in general and the pulmonary mechanosensory system in particular may thereby be facilitated.

GRANTS

This study was supported by a VA Merit Review Award (PULM-024-17S).

DISCLAIMERS

The contents do not represent the views of the U.S. Department of Veterans Affairs or the United States Government.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author.

AUTHOR CONTRIBUTIONS

J.Y. conceived and designed research; performed experiments; analyzed data; interpreted results of experiments; prepared figures; drafted manuscript; edited and revised manuscript; and approved final version of manuscript.