Spectrum of myelinated pulmonary afferents (II)

Abstract

Recently, it has been recognized that a single airway sensory unit may contain multiple receptive fields and that each field houses at least one encoder. Since some units respond to both lung inflation and deflation, we hypothesized that these units contain heterogeneous encoders for sensing inflation and deflation, respectively. Single unit activities were recorded from the cervical vagus nerve in anesthetized, open chest, and mechanically ventilated rabbits. Fifty-two airway sensory units with multiple receptive fields that responded to both lung inflation and deflation were identified. Among them, 13 units had separate receptive fields for inflation and deflation, where one of the fields could be blocked by local injection of 2% lidocaine (10 μl). In 8 of the 13 units, the deflation response was blocked without affecting the unit's response to inflation, whereas in the remaining five units, the inflation response was blocked without affecting the deflation response. Our results support the hypothesis that a single mechanosensory unit may contain heterogeneous encoders that can respond to either inflation or deflation.

two types of mechanoreceptors, which can respond to lung inflation, exist in the airways: rapidly adapting receptors (RARs) and slowly adapting receptors (SARs). They transmit mechanical information through myelinated afferents in the vagus nerve (5, 13, 15, 18). Although the behaviors of RARs and SARs are unique, they may demonstrate similarities (20). Thus, these myelinated pulmonary afferents were proposed being connected to a heterogeneous group of receptors, which behave as a spectrum with typical SARs, and RARs represent two extremes of the spectrum (20). Some afferents only respond to lung inflation, yet others respond to both inflation and deflation (2, 19). Conventionally, the responses to inflation and deflation are explained by a single receptor hypothesis, in which stimuli, such as lung inflation or deflation, are encoded by a single receptor (RAR or SAR) (1, 5, 9, 11, 13, 15, 16, 19, 20). Alternatively, this phenomenon can be explained by a multiple-receptor hypothesis (21), in which different stimuli are encoded by different receptors. Recently, a single pulmonary mechanosensory unit was found to possess multiple receptors (23). A sensory receptor is an encoder, which is the smallest device that can independently generate action potentials. A receptive field is a sensory region that houses receptors. A receptive field contains at least one receptor but often has multiple receptors. The discovery of multiple receptive fields in a single unit (23), along with multiple sensory structures connected to a single axon (22), leads to the hypothesis that an airway mechanosensory unit is a functional unit that contains multiple receptors with different characteristics. Each is capable of sensing one type of mechanical stimuli, such as lung inflation or deflation. Furthermore, each may detect a static (slowly adapting) vs. a dynamic (rapidly adapting) component for inflation or deflation. Thus, four different types of encoders exist: rapidly adapting inflation, slowly adapting inflation, rapidly adapting deflation, and slowly adapting deflation encoders. In the current studies, we tested this hypothesis by recording activity in a single unit that responded to both lung inflation and deflation. If we identified more than one receptive field, we injected local anesthetic into one of the fields to see whether inflation or deflation response could be eliminated. Indeed, the present results show that inflation and deflation responses could be blocked independently, supporting that there are heterogeneous encoders within a single sensory unit.

METHODS

General.

Experiments were carried out in male New Zealand white rabbits intravenously anesthetized with 20% urethane (1 g/kg). Additional doses (one-seventh of the initial dose) were given whenever necessary. Study procedures were in accordance with ethics codes set by the National Institutes of Health and approved by the Institutional Animal Care and Use Committee of the University of Louisville. A midline incision was made to expose the trachea and vagus nerve. The trachea was cannulated low in the neck close to the carina, and the lungs were mechanically ventilated with a Harvard ventilator (model no. 683; South Natick, MA). Positive-end-expiratory pressure (PEEP) was maintained by placing the expiratory outlet under 3–4 cm H₂O. Airway pressure was monitored at the tracheal tube inlet by a Statham pressure transducer (P23). The chest was opened widely in the midline to allow location of the receptive field during single-unit recording (see below). Unit responses to cyclic changes in airway pressure, PEEP removal, and lung inflation were examined before and after microinjection of local anesthetic (2% lidocaine). Lidocaine was injected into the receptive field with a needle (30 GD) in 10 μl. Airway pressure and afferent activities were recorded by a thermorecorder (Astro-Med Dash IV).

Recording of afferent activity.

Single-unit activities from vagal afferents were recorded according to conventional methods (20). The vagus nerve (either right or left) was separated from the carotid sheath, placed on a dissecting platform, and covered with mineral oil. A small afferent bundle was cut from the vagus nerve and was dissected into thin filaments with two pairs of fine forceps. The filaments were further divided and placed on recording electrodes to measure action potentials. The electrodes were connected to a high-impedance probe (Grass model no. HIP 511), from which the output was fed into an amplifier (Grass P 511). After suitable amplification, action potentials from a single unit of the vagal sensory receptors were displayed on an oscilloscope and monitored by a loudspeaker. In addition, a voltage analog of impulse frequency was produced by a rate meter (Frederick Haer, Brunswick, ME) at a bandwidth of 0.1 s. Adaptation rate was determined by an adaptation index.

Sensory unit identification.

The mechanosensory units that responded to both lung inflation and deflation were identified (20). Then, the receptive field was located by identifying the most sensitive point on the lung surface with a glass rod having a 0.5-mm round tip. At the center of this point, touching elicited a high discharge frequency, which can be easily identified by listening to the monitor and hearing a high-pitch burst of action potentials. This response subsided, as the distance away from the center increased. The sensory units containing more than one receptive field were further examined. That recorded activities were coming from a single unit was verified by inspection of the contour of action potentials with the oscilloscope under a high time resolution and by their discharge pitch from the loudspeaker. All receptive fields identified were in the lung parenchyma. It needs to be pointed out that lung collapse is a heterogeneous condition. The response of a receptor to collapse depends on its local conditions, as well as its inherent properties. For example, a deflation-activated receptor (DAR) may not respond to lung deflation if it is located in a lung region where air is trapped. In addition, RARs usually have a very low discharge frequency during normal tidal ventilation. An active search for RARs requires large inflations or deflations with negative pressure for each time a new strand of vagal afferents is put on the electrodes for recording. In the current studies, we did not perform such maneuvers; therefore, our studies would have undercounted the percentage of RARs in the myelinated afferent population.

Procedures.

When a sensory unit had been identified, the lung was deflated by removal of PEEP for four or five ventilatory cycles. If the units responded to PEEP removal, they were entered for a series of procedures. First, the lung was inflated to 30 cmH₂O constant pressure; then the lung was deflated to −7 cmH₂O. The unit responses to these pressures were used to calculate an adaptation index. The adaptation index is the difference between discharge frequencies of the peak and average of the second second, which is then divided by the peak frequency and expressed as a percentage. Peak discharge usually occurs within the first 0.3 s after pressure reached peak. Any units with a delayed discharge peak were excluded from the study. An adaptation index above 70% is defined as rapidly adapting. Then, the receptive fields would be identified again. Local anesthetic (2% lidocaine) was injected directly into one of the fields to determine whether the inflation or deflation response can be blocked. The deflation response was assessed by counting the number of action potentials in a complete deflation phase during the second ventilatory cycle after PEEP removal. We used PEEP removal for the quantitative analysis because this maneuver is simple, reproducible, and precise. The inflation response was assessed by counting the number of action potentials for one second after the first second following initiation of lung inflation.

RESULTS

In 42 rabbits, we encountered 52 mechanosensory units (1–2 units per rabbit) that responded to both lung inflation and deflation and had separate receptive fields. Three of the fifty-two units were rapidly adapting, and the remaining units were slowly adapting ones. Therefore, our study is mainly about slowly adapting units. Activities of these units increased during the lung inflation phase and decreased during the deflation phase. The activities increased again in response to lung deflation either by PEEP removal or negative pressure deflation. Upon constant pressure deflation, 10 of the units adapted rapidly. The adaptation rate was higher than that of lung inflation (Fig. 1). If rapidly adapting units (either to lung inflation or deflation) were removed, the adaptation rate to lung deflation was still higher than adaptation to lung inflation. There was no correlation between the two adaptation indexes. For any given unit, the adaptation index for lung inflation can be lower, higher, or the same as for lung deflation (Fig. 1).

Fig. 1.

Correlation between sensory units' inflation and deflation adaptation index (AI; r = 0.205, P = 0.1446; n = 52). Each symbol represents a sensory unit. Solid squares denote slowly adapting units, while open circles denote rapidly adapting units. Clearly, these sensory units were more rapidly adapting in response to lung deflation −7 cmH₂O) than inflation (−30 cmH₂O), as demonstrated by more units above the equal value line, with an average adaptation index (51 ± 19% for lung deflation and 34 ± 16% for lung inflation; P < 0.001; n = 52). Such a relationship remains after removing all of the rapidly adapting units (circular ones), when averaged AI was 46 ± 16% for lung deflation and 34 ± 14% for lung inflation (P < 0.001; n = 39).

Among the 52 units, activities in 10 units were blocked completely by local injection of lidocaine into one of the receptive fields; activities in 29 units decreased both in response to lung inflation and deflation. In these 29 units, however, the suppressive effects varied from unit to unit. The suppressive effects could be greater with inflation or deflation, or they could be equally suppressive. Figure 2 illustrates simultaneous suppressive effects. However, in the remaining 13 units, only one of the responses was blocked or significantly decreased by injection of lidocaine into one of the receptive fields. In 8 of the 13 units, the deflation response was significantly blocked without substantial effects on the inflation response (Fig. 3), whereas in the remaining five units, the inflation response was greatly attenuated, with little effect on the deflation response (Fig. 4). The grouped data for these 13 sensory units are illustrated in Fig. 5.

Fig. 2.

Blocking one receptive field may simultaneously suppress inflation and deflation receptors. A typical slowly adapting unit recorded from the left cervical vagus nerve in an anesthetized, open-chest, and mechanically ventilated rabbit. This unit has two separate receptive fields. The traces are IMP/s, sensory activity per second counted at each 0.1 s; IMP, impulses or action potentials; Paw, airway pressure measured in cmH₂O. A–C: intact, controls; D–F: block, after blocking one field. Positive-end expiratory pressure (PEEP) removal (A, D); Inf 20 cmH₂O (B, E) and Inf 30 cmH₂O (C, F), lung inflation to a constant pressure of 20 and 30 cmH₂O, respectively. Please note that though both inflation and deflation responses were attenuated, none was eliminated.

Fig. 3.

Blocking one receptive field blocked deflation response but not inflation response. A–D: intact (precontrols); E–H: block (after blocking one of the receptive fields); I–L: recovery (postcontrols). They were tested under PEEP removal (A, E, I), lung deflation at the pressure of −7 cmH₂O (B, F, J), lung inflation at pressure of 20 (C, G, K) and 30 (D, H, L) cmH₂O, respectively. Under the control conditions, after PEEP removal, the unit discharged during the expiratory phase (A), and was vigorously stimulated by deflation (B). Please note that the sensory unit responded to negative pressure deflation with two phases, an initial rapidly adapting phase followed by a slowly adapting phase. After blocking one of the receptive fields, the unit response to PEEP removal (E) and slowly adapting component in response to the negative pressure deflation (F) disappeared, leaving the rapidly adapting component intact. However, the unit response to lung inflation did not change (G, H). For figure legend abbreviations, please see Fig. 2.

Fig. 4.

Blocking one receptive field blocked inflation response but not deflation response. Sensory unit activities were recorded before (A–D) and after (E–H) blockade of one receptive field with 2% lidocaine (10 μl). The expiratory activity did not change, but the inspiratory activity was significantly decreased after the blockade. For figure legend abbreviations, please see Fig. 2.

Fig. 5.

Effects of lidocaine to block inflation or deflation responses in sensory units. The left four columns are grouped data for the blockade of inflation responses. The right four columns represent the data for the blockade of deflation responses (i.e., response to PEEP removal). The solid bars are deflation responses. The open columns are inflation responses. *P < 0.05; **P < 0.01. Histogram bars 1 and 3 are controls, whereas 2 and 4 are after blockade. Please note that when the inflation response was blocked, the deflation response did not change substantially (left bars). Similarly, when the deflation response was blocked, there were no significant effects on inflation responses (right bars).

Figure 6 illustrates a unit that clearly had both rapidly adapting and slowly adapting components in response to lung inflation, i.e., RAR and SAR shared an axon. Local injection of lidocaine blocked the RAR without affecting the SAR activity in the sensory unit. Figure 7 shows a unit that contained multiple heterogeneous encoders, including inflation- and deflation-activated encoders. Both inflation and deflation responses had rapidly adapting and slowly adapting components. This unit has an active DAR, which is blocked by local anesthetic without affecting the inflation response. Such a unit is not popular in the rabbit but is popular in the rat [Fig. 8 and (3, 17)]. Figure 8 illustrates that an active DAR was not affected by blocking the inflation encoder in a sensory unit from a rat.

Fig. 6.

SAR and RAR sharing an axon. This sensory unit has two heterogeneous encoders: SAR (low threshold and low frequency) and RAR (high threshold and high frequency). In A, during lung inflation, the unit starts with SAR activation, producing a low discharge frequency; as the airway pressure increases, the RAR is activated (indicated by two arrows), and the unit discharges with very high frequency. During constant lung inflation, the unit fires with a higher frequency (from the RAR encoder) initially, but rapidly adapts to a steady state (from the SAR encoder). In B, blocking the RAR encoder with local injection of 2% lidocaine, the rapidly adapting component disappeared and left only the slowly adapting component, which exhibited continuous lower discharge frequency. For figure legend abbreviations, please see Fig. 2.

Fig. 7.

A sensory unit with an active deflation activated encoder, i.e., it was active during lung deflation phase (indicated by arrows), which can be selectively blocked. A–E are controls with an intact sensory unit, showing PEEP removal, negative pressure deflation, and constant pressure inflation of the lung to 10, 20, and 30 cmH₂O, respectively. F–J were recorded after lidocaine blockade of one of the receptive fields. The active deflation-activated encoder was blocked. However, the unit was still activated during a strong deflation either by PEEP removal (F) or by negative pressure (G), although activity was significantly reduced. Clearly, the active deflation encoder has a low activating threshold, and this sensory unit contains more than one deflation encoder. For figure legend abbreviations, please see Fig. 2.

Fig. 8.

A sensory unit with an active deflation-activated encoder from a rat. This unit is active during both inflation phase and deflation phase under resting mechanical ventilation. This unit has two receptive fields, with one responding to inflation and the other to deflation. Please note that the inflation activity was blocked, but the deflation activity did not change after lidocaine injection. For figure legend abbreviations, please see Fig. 2.

DISCUSSION

Our current studies demonstrated that responses of a mechanosensory unit to both lung inflation and deflation could be blocked independently, indicating that they are coming from different receptors.

Sensory afferents that respond to lung deflation were first identified in a multiple-fiber preparation by Keller and Loeser in the rabbit (8) and then in a single fiber by Adrian in both the rabbit and the cat (1). These sensory units also responded to lung inflation, but often at a slowly adapting rate. Twenty (19) to thirty percent of the slowly adapting units were reported to respond to forced lung deflation (9). In our previous study, 11% of slowly adapting units responded to lung deflation with removal of PEEP (20). This type of unit has been studied extensively in the rabbit (1, 11, 14), as well as other animals such as cats (1, 9, 19), dogs (2), guinea pigs, (10) rats (3, 6, 12, 17), and mice (25). Airway sensory units can be slowly or rapidly adapting to either lung inflation or deflation. Adaptation indexes to lung inflation and deflation can be similar or different, but they behave independently. Currently, investigators categorize sensory units into rapidly or slowly adapting ones, according to their response to lung inflation. The differences in adaptation indexes to lung inflation and deflation justify further separation into groups, according to their stimulus.

Independent blockade of a sensory response to inflation or deflation suggests that the activities come from different encoders. We do not know yet whether one encoder can respond to both inflation and deflation. However, we provide compelling evidence that the inflation response is separate from the deflation response and is generated from two separate encoders. For example, as unit response to deflation was blocked, the response to inflation was unchanged (Figs. 3 and 5), and vice versa (Figs. 4 and 5). Such independent blockade was also demonstrated in the rat (Fig. 8). This supports our theory that a mechanosensory unit consists of many encoders and the mixed behavior is due to the heterogeneous composition of encoders. The sensory unit in Fig. 7 may possess all of the four receptor types. This is demonstrated when the lung was inflated to 30 cmH₂O or deflated to −7 cmH₂O, the unit discharged immensely, and the discharging frequency adapted rapidly to a steady-state level with no further significant adaptation. Because each encoder has its own characteristics in activation threshold, operation range, operating sensitivity, and deactivation threshold, within a type, individual encoders behave similarly, but not identically. When a sensory unit consists of homogenous encoders, it behaves as a typical SAR, RAR, or rapidly adapting or slowly adapting DAR. However, a unit may contain heterogeneous encoders, exhibiting mixed behavior. Thus, it may respond to both inflation and deflation, or behave as RAR-like SARs (Fig. 6) (20). In addition to Fig. 6, Figs. 2, 4, and 7 also demonstrate rapidly adapting and slowly adapting components during lung inflation to 30 cmH₂O, indicating coexistence of RAR and SAR in the sensory unit. Coexistence of RAR and SAR in a sensory unit has been found in mammalian muscle spindle (7). Similarly, many units had the rapidly and slowly adapting components during constant pressure deflation (Figs. 3 and 7). Interestingly, the slowly adapting deflation response could be selectively blocked without affecting rapidly adapting deflation response (Fig. 3).

An airway sensory afferent may connect to many receptive fields, which was found in about one-third of slowly adapting units (23). Multiple receptive fields were also found in sensory units in the other visceral organs, such as esophagus, stomach, and colon (4, 24). Morphologically, receptor structures may be as close as a few hundred micrometers (21). The techniques used in the current studies cannot identify two receptive fields less than 1 mm apart (23). Thus, a receptive field may contain multiple encoders. Multiple encoders within a receptive field may explain simultaneous suppression of inflation and deflation responses when one of the receptive fields was blocked. Alternatively, local injection is not precise, and the distribution of local anesthetic is complicated. Liquid may be blocked by the alveolar wall from infiltrating alveoli nearby, yet run through airways to more distant sites. Furthermore, the anesthetic may spread to adjacent areas. This may explain why blocking one receptive field may abolish the sensory activity.

In conclusion, our data support the contention that an airway mechanosensory unit consists of many encoders. Each encoder has a special function to detect a unique variable, such as inflation, deflation, static stimulation, or dynamic stimulation. When a sensory unit consists of homogenous encoders, it behaves as a typical SAR, RAR, or DAR, whereas when a sensory unit contains heterogeneous encoders, it exhibits mixed behavior. A significant amount of information is integrated at the sensory unit level. Thus, airway sensory units function not only as transducers but also as processors.

Pespectives and Significance

Airway mechanosensors provide the respiratory centers with crucial information about lung mechanics for controlling breathing. Therefore, a thorough understanding of mechanosensory behavior is necessary. Learning the pattern of the sensory discharges has attracted the attention of many investigators for more than a half-century; however, our understanding is still incomplete. The current results support a theory that an airway mechanosensory unit contains multiple receptors, including rapidly adapting receptors, slowly adapting receptors, and deflation-activated receptors. Each type of receptor is capable of sensing a particular variable of lung mechanics. Thus, the sensory unit is not only a transducer but also a processor that integrates a significant amount of information. This new insight will challenge our current view regarding how the central nervous system deciphers the incoming signals from the respiratory system.

GRANTS

This study was supported by a grant from VA Merit Review (PULM-029-10S).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

Author contributions: J.L. and J.Y. performed experiments; J.L. and J.Y. analyzed data; J.L. and J.Y. interpreted results of experiments; J.L. prepared figures; J.L. and J.Y. edited and revised manuscript; J.L. and J.Y. approved final version of manuscript; J.Y. conception and design of research; J.Y. drafted manuscript.