INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Heightened levels of airway responsiveness are produced by various insults to the airway. One of the most clinically relevant insults in humans is viral infection of the respiratory tract (1). The mechanisms that interact to produce altered airway function are multiple and complex (2), but they include disruption of neural pathways that normally mediate contraction or relaxation of airways (3). In this respect, we previously reported that the consequences of infection of cotton rats with the human respiratory syncytial virus (HRSV) include a significant increase in the airways neurally mediated cholinergic contractile response as well as a decrease in the nonadrenergic noncholinergic inhibitory (NANCi) response shortly after the infection (4).

The consequences of an insult to an airway should be considered not only in terms of the acute event, but also as a function of the age at which the insult is delivered to the host. This may be especially important at a time of rapid somatic growth when the normal mechanisms of airway control are developing over time. In this respect, our laboratory has shown that rabbits do not normally have a functional NANCi response at birth, and this pathway subsequently undergoes significant postnatal development (5). Furthermore, neonatal sensitization to an allergen appears to disrupt the normal functional development of this pathway (5).

In the present study, we addressed the more chronic effects of HRSV infection on airway control in young ferrets during a period of rapid somatic growth. For this study, ferrets were employed because the pattern and age-dependency of viral replication within the respiratory track of this species closely resembles the clinical course of the disease that is found in humans (6).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Pregnant female ferrets (Mustela furo) were obtained from Marshall Farms (North Rose, NY) Animals were shipped approximately 3 wk prepartum and were housed in large plastic guinea pig cages in which nesting boxes were placed. The ferrets were maintained on a diet specific for ferrets (Marshall Farms Premium Ferret Diet). Room lighting was regulated to provide 16 h of light each day. Litters ranged in size from 6 to 10 infant ferrets, with an average per litter of eight animals. All the procedures employed in this study were approved by the Animal Care and Use Committee of the National Jewish Medical and Research Center and conformed to NIH guidelines.

Virus

The HRSV, strain A_Long, was used in all experiments. As previously described (7), stocks of this virus were prepared in Hep-2 cells, which were obtained from the American Type Culture Collection (Rockville, MD).

Mode of Infection

Infant ferrets were infected at 2 to 10 d of age (average age, 5 d). Animals were infected intranasally under light anesthesia (Methoxyflurane; Pitman-Moore, Mundelein, IL) with 10⁵ plaque-forming units (PFU) of virus. The volume of the inoculum given to the anesthetized ferrets was approximately 0.1 ml. Control animals were given uninfected cell culture medium in a similar manner. Previous studies have shown that under these conditions the inoculum uniformly reached the lungs, with viral replication documented in both the upper and the lower respiratory tracts of ferrets given HRSV (6).

Viral Assays

To document the effectiveness of our method of RSV delivery into the lower respiratory track, two ferrets that received RSV and one ferret that received medium alone (control) were killed 3 d after challenge of their airways. The top half of each left lung, including the mainstem bronchus, was placed in 10 volumes of cold Hanks' balanced salt solution supplemented with 0.218 M sucrose, 4.4 mM glutamate, 3.8 mM KH₂PO₄, and 7.2 mM K₂HPO₄. The tissue was homogenized on ice with a teflon/glass Wheaton homogenizer (Millville, NJ). The supernatant of the homogenate was subsequently assessed for RSV using the USU-104 cell line. This continuous monkey kidney cell line was seeded into a 150-cm² tissue culture flask and maintained on Medium 199 with Earle's balanced salt solution and 5% fetal bovine serum (FBS) in a 37° C humidified 5% CO₂ incubator until confluency. Flasks were confluent in 48 to 72 h and contained approximately 10 × 10⁶ cells. They were then refed with the Medium 199 and 1% FBS and inoculated with the supernatant. The appearance of the cell sheet was subsequently monitored for cytopathic effects.

Preparation of Airways for In Vitro Studies

Tracheal smooth muscle (TSM) segments approximately 0.75 cm in length were obtained from groups of control and HRSV-infected animals when they were 4, 8, or 24 wk of age. After anesthesia with 5 mg/ kg xylazine (Mobay Co., Shawnee, KS) and 35 mg/kg ketamine (Aveco, Fort Dodge, IA), animals were killed with an additional 5 mg xylazine and 25 mg ketamine injected intracardially. The airway segments were cleaned of loose connective tissue and placed in 3-ml organ baths (Harvard Apparatus Co., Millis, MA) and supported longitudinally by triangular stainless steel wire supports. The lower support was attached to a stainless steel hook at the base of the organ bath, and the upper support was attached with a gold chain to an isometric force transducer (Model FT.03C; Grass Instrument Co., Quincy, MA). The latter was mounted on a rack and pinion clamp permitting adjustments of the resting length of each TSM segment.

The tissues were bathed in modified Krebs-Henseleit solution (in mM): 118 NaCl, 25 NaHCO₃, 2.8 CaCl₂ H₂O, 1.17 MgSO₄, 4.7 KCl, 1.2 KH₂PO₄, and 2 g/L dextrose. The baths were aerated with a 95% O₂-5% CO₂ gas mixture, and a pH of 7.43 ± 0.03 was established for the duration of each experiment. The temperature of the baths was maintained at 37° C. Each TSM segment was equilibrated in the bath for 60 min at an optimal resting tension of 1 g. This optimal resting tension was established by assessing its maximal contractile response to the following standard electrical field stimulation (EFS): 2 V; 1 ms pulse duration; 20 Hz stimulus frequency. During this equilibration time, the tissue was challenged one time with KCl at a final concentration in the bath of 120 mM as in previous studies (5, 8, 9). Tissues that would not develop a contractile response were excluded from further studies. Tissues were rinsed with fresh K-H solution periodically and allowed to relax to their initial tension after reaching maximal contraction. At the end of each experiment the TSM segments were blotted on a gauze pad and weighed.

Recordings of resting tensions and TSM contractile responses were made on a SensorMedics Dynagraph Recorder (R612) equipped with type 9853A couplers, 461D preamplifiers, and 412 amplifiers (SensorMedics Corp., Anaheim, CA). EFS was delivered by a Grass S44 stimulator connected to a stimulus isolation unit (Grass Instruments, Quincy, MA). The stimulus was applied transmurally across the tissues by means of parallel platinum electrodes (each 0.3 cm²), and tension was recorded.

Assessment of Contractile Responses to EFS

Neurally mediated contractile responses of TSM were assessed in vitro on TSM segments from control and infected animals bathed in Krebs-Henseleit solution. Full frequency-response curves (0.5 to 30 Hz; 2 V; 1 ms pulse duration) were performed using the stimulator and recording apparatus described above. Each frequency of stimulation was applied to the tissue until a plateau in the contractile response was observed (approximately 20 to 30 s). A 1-min period of time elapsed before application of the next stimulation frequency. Results were expressed in terms of maximal contractile response (Tmax) and frequency of stimulation causing 50% of the maximal contractile response (ES₅₀). With the parameters of stimulation used, these responses were both neurally and cholinergically mediated in that either tetrodotoxin (TTX) (Sigma Chemical, St. Louis, MO), an inhibitor of neural transmission, or atropine, a muscarinic receptor antagonist, completely abolished EFS-induced contractions of TSM segments.

Assessment of Contractile Responses to Methacholine

To assess postsynaptic cholinergic sensitivity of the tissues, cumulative dose-response curves to methacholine were performed in half-log increments employing concentrations ranging from 10⁸ to 10⁴ M. Results in the control and experimental groups at each time point were expressed in terms of the maximal contractile response as well as the concentration of methacholine causing 50% of the maximal contraction (EC₅₀).

Assessment of NANCi Responses

To evaluate NANCi responses, experiments were performed in the presence of atropine (Aldrich, Milkwaukee, WI), at a final concentration of 1 × 10⁶ M, propranolol (Sigma), at a final concentration of 5 × 10⁶ M, and indomethacin (Sigma), at a final concentration of 1 × 10⁶ M. These concentrations of atropine and propranolol, chosen after preliminary dose-response curves, have been previously used by us (5, 8) and other investigators (10) in the study of NANCi responses in vitro. Neurokinin A (NKA) (Sigma), at a final concentration of 5 × 10⁷ M, was used to induce tone in TSM. This concentration of NKA, selected after preliminary dose-response curves, produced approximately 50% of the maximal contractile response to methacholine. After obtaining a plateau response with NKA, EFS was applied at stimulation frequencies ranging from 1 to 30 Hz (2 V; 1 ms pulse duration). Each frequency of stimulation was applied to the tissue until a plateau in the relaxant response was observed (approximately 15 s). Tissues were allowed to return to their baselines before application of the next stimulation frequency. Changes in tension from baseline NKA contractile response were recorded and expressed as percent relaxation (mean ± SEM). To further define the neural contribution to the relaxant response, the ability of TTX (final concentration of 5 × 10⁶ M) to diminish the relaxant response was assessed in selected experiments. The stimulation parameters were unchanged, and each tissue served as its own control. For some tissues, TTX was not added to the bath before the second stimulation so that the reproducibility of the relaxant responses could be defined.

Histology

To evaluate the histopathologic changes induced by HRSV infection, tracheas from both the control and the experimental groups were removed and fixed in formalin at the 4- and 8-wk time points. Histologic sections were stained with hematoxylin-eosin and examined by light microscopy.

Statistical Analysis

Data were expressed as mean ± SEM when "n" equaled the number of experimental observations (i.e., the number of TSM segments studied). At each time point, TSM was obtained from a minimum of three animals. The results were analyzed with a two-tailed unpaired t test. A p value of < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Viral Assays

The presence or absence of virus within the lung of ferrets after intranasal inoculation was monitored as outlined above. When supernatants from the lung homogenates from the two ferrets that received HRSV were applied to the USU-104 cell line, cytopathic effects were detected, whereas in the ferret given medium without HRSV, no cytopathic effects were seen.

Contractile Response to EFS

To evaluate the neurally mediated contractile response of TSM segments, full frequency response curves to EFS (0.5 to 30 Hz) were assessed in terms of ES₅₀ and Tmax values. When matched for age, there was no significant differences between the control and HRSV-infected groups in terms of maximal contractile responses (data not shown). Conversely, we found increased contractile responses as reflected by ES₅₀ values in TSM from infected animals compared with the control group at both 4 and 8 wk. In this respect, the ES₅₀ (mean ± SEM) in TSM from control animals was 1.3 ± 0.1 (n = 6), whereas the value in the infected group (n = 7) was 0.7 ± 0.1 at the first time point (p = 0.001). At 8 wk of age, these values were 2.0 ± 0.2 in the control group (n = 15) and 0.9 ± 0.2 in the infected group (n = 8; p = 0.008). As shown in Figure 1, this difference had resolved by 24 wk of age in that the frequency response curves were now superimposable and the ES₅₀ values were not significantly different (2.0 ± 0.2 in control ferrets [n = 15] versus 2.5 ± 0.4 in TSM from infected ferrets [n = 19]).

View larger version (15K): [in this window] [in a new window]   Figure 1.   Frequency response curves to EFS (0.5 to 30 Hz) of TSM segments obtained from control (closed circles) and RSV-infected ferrets (open squares) at 4, 8, and 24 wk of age. Tensions generated from TSM segments are expressed as a percent of the maximal response (mean ± SEM) with n = 6 (control) and 7 (RSV) at the first time point, n = 15 (control) and 8 (RSV) at the second time point, and 15 (control) and 19 (RSV) at the last time point. Tracheal segments from infected animals were more responsive to EFS as demonstrated by significantly lower ES50 values at the 4 and 8 wk time points, but not in 24-wk-old ferrets.

Contractile Response to Methacholine

There were no differences in TSM responses to MCh in terms of either Tmax or EC₅₀ values between the two experimental groups when matched for age. This is shown in Table 1. At the same time, Tmax did decrease significantly in magnitude from the first to the second and third time points in TSM from both the control and the HRSV-infected ferrets.

NANCi Responses in Normal and in Infected Ferrets

A representative tracing demonstrating NANCi responses in tracheal segments from a control ferret is shown in Figure 2. This response was not found in normal 4-wk-old ferrets, but it was present at both 8 and 24 wk of age. When ferrets were infected with HRSV during the first 10 d of life, NANCi responses were significantly decreased in 8-wk-old ferrets (Figure 3). For example, at a stimulation frequency of 20 Hz, EFS-induced relaxation was 32.0 ± 4.2% in TSM from control ferrets (n = 12), whereas the value was 2.5 ± 2.5% in the HRSV-infected group (n = 8; p = 0.0001). A significant difference persisted in control versus infected ferrets at 24 wk of age, albeit to a lesser degree (NANCi response at 20 Hz: 32.0 ± 4.2% [n = 12] versus 16.0 ± 2.3%, respectively; p = 0.008). This change in NANCi response as a function of the age of the ferrets and HRSV status is shown in Figure 3.

View larger version (7K): [in this window] [in a new window]   Figure 2.   Representative tracing demonstrating NANCi responses at varying frequencies of stimulation in ferret airways in vitro. The dark vertical bar represents 1 g of tension. The responses were obtained in the presence of indomethacin, atropine, and propranolol in TSM contracted with NKA.

View larger version (11K): [in this window] [in a new window]   Figure 3.   NANCi responses of TSM segments obtained from control (closed bars) and RSV-infected (open bars) ferrets at stimulation frequencies of 10 and 20 Hz at both 8 and 24 wk of age. Values are expressed as percent relaxation (mean ± SEM) with n = 12 (control) and 8 (RSV) at both time points. The tracheal segments obtained from infected animals showed significantly decreased NANCi relaxations when compared with control animals at both 10 and 20 Hz (p = 0.02 and 0.0001, respectively) at the earlier time point. A significant difference (p = 0.008) was still present at a stimulation frequency of 20 Hz in TSM from older ferrets.

The ability of TTX to inhibit the NANCi response was assessed. When TTX was not introduced into the baths, the relaxant responses were reproducible from the first to the second series of stimulations. However, when TTX was added to TSM from normal ferrets, it completely inhibited relaxation at 10 Hz and inhibited the NANCi response by 32.0 ± 4.1% at 20 Hz (n = 8). In TSM from HRSV-infected ferrets, the TTX inhibition at 20 Hz was not significantly different (30.0 ± 3.9%; n = 6). The tissues for the latter experiment were from older ferrets (24 wk) in which the NANCi response was easily measurable but not normal in magnitude, as displayed in Figure 3.

Histology

The histology of the trachea was assessed in both control and HRSV-inoculated ferrets at all three time points. No differences were apparent in this analysis. Specifically, there was no patchy loss of the airways' epithelial cells or submucosal edema in the groups with prior HRSV infection. These abnormalities were seen within the central airways in a previous study of HRSV infection performed in cotton rats (4) when the histology was assessed 4 d after administration of HRSV to the airways. Conversely, at least 3 wk elapsed between viral administration and obtaining tissue for analysis in this study.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The ability of the respiratory syncytial virus (RSV) to cause disease in humans has been appreciated for four decades (11). Prominent among the clinical syndromes produced by this agent are viral-induced wheezing in infants (bronchiolitis) as well as exacerbations of asthma in people of all ages, but especially in children (12). In several studies, bronchiolitis requiring hospitalization early in life has been identified as one risk factor associated with the subsequent development of asthma and airway hyperresponsiveness (reviewed in 1). Several hypotheses have been advanced regarding mechanisms via which RSV (as well as other viral pathogens) might alter airway function both acutely and on a more chronic basis. These theories include epithelial damage, immunologic and inflammatory mechanisms, and dysfunction of the airways' neural control (1, 2, 13). In terms of the latter possibility, our laboratory recently reported that the acute consequences of infection of cotton rats with HRSV include a significant increase in the airway's neurally mediated cholinergic contractile response as well as a decrease in the NANCi response (4). These observations were made 4 d postinfection in 5-wk-old cotton rats at a time when viral replication should reach a peak in tracheal tissues. Neither the time course of these abnormalities in neural control nor the effects of infection at an early postnatal age were assessed in this earlier study (4).

The possibility that events early in life might alter the normal development of neural control within an airway has recently received attention in terms of experimental studies. For example, when investigating the functional development of NANCi innervation in rabbits, we found that early sensitization to an allergen (begun on the first day of life) alters the normal maturation of this neural pathway as demonstrated by significantly reduced NANCi responses in allergen-sensitized animals as they matured (5). In another study, recurrent aspiration of milk in rabbits during the first month of life led to marked changes in the contractile response to EFS at a time after aspiration had ceased (17). These observations led to the current study in which the primary pathogenic viral agent encountered in early life was assessed in terms of its ability to produce long-term alterations in neutral control.

In the present study, we evaluated the effect of HRSV infection on both the airways' contractile and relaxant responses to EFS. We found significantly decreased NANCi responses in infected ferrets when compared with control animals at both of the latter time points (8 and 24 wk), whereas the abnormal contractile responses were limited to the assessments at 4 and 8 wk. Although a dysfunction of the NANCi system might lead to increased contractile responses to EFS, the dissociation of the responses at the first and last time points suggests that separate mechanisms may be responsible for these changes. In terms of the EFS-induced increase in contractile responses, one or more of several mechanisms might explain this alteration. An increase in the trachea's smooth muscle response to acetylcholine (ACh) released by neural stimulation could take place because of postjunctional alterations in muscarinic receptors (increased number, affinity for ACh). However, the observation that the in vitro response to methacholine was not different between tissues from infected and control ferrets (Table 1) makes this less likely. In addition, in other experimental models of viral infection, other laboratories have also reported that the contractile response of airway smooth muscle to muscarinic agonists in vitro was unaltered by the infection (18). Prejunctionally, an increase in the neural release of ACh from parasympathetic nerve endings could also lead to the enhanced contraction. This mechanism was invoked to explain the increased response of airway smooth muscle to EFS in the absence of an increased response to exogenous ACh in an in vivo model of herpes virus I infection (21). Within airways, release of endogenous ACh is under the local control of muscarinic autoreceptors (M₂) on postganglionic parasympathetic nerves (3, 22). A dysfunction of M₂ autoreceptors with loss of this inhibitory control has been described as an acute consequence of parainfluenza infections in guinea pig (23) and rat models (24). In addition, parainfluenza infection of guinea pigs has also been noted to lead to a more chronic change in this receptor's function in that it becomes dependent upon an inducible form of cyclooxygenase (COX II) for normal activity (25). Both the acute and more chronic effects of this viral infection on M₂ autoreceptors could lead to facilitation of cholinergic neurotransmission and explain in part this virus-evoked airway hyperresponsiveness. Although direct measurement of release of this endogenous neurotransmitter from nerves was not performed as part of this study, the results suggest that RSV-induced alterations of the airways' neural control are more likely to be explained by abnormal presynaptic than postsynaptic cholinergic mechanisms.

The mechanisms through which a HRSV infection contributes to a dysfunction of the NANCi system are undefined. Possible mechanisms that could decrease NANCi responses have been proposed (3) and include alterations in the messenger(s) that mediate the NANCi response by the products of inflammation. For example, proteolytic enzymes released from inflammatory cells drawn into an airway could destroy neuropeptides that mediate relaxation. This might be especially important during the acute infectious process. Conversely, persistence of respiratory syncytial virus genome and protein after acute bronchiolitis in guinea pigs has been described (26). The persistence of viral genome within airway cells (mononuclear cells and/or airway epithelial cells) might alter the ability of the airway to relax by altering the function/ products of these cells. These potential mechanisms as well as others remain speculative at this point in part because of the controversial nature of the neurotransmitter(s) mediating EFS-induced relaxation in this and in other species.

The nature of the NANCi response in ferrets deserves additional comment. As noted in RESULTS, this EFS-induced relaxant response was only partially inhibited by TTX at the higher stimulation frequency of 20 Hz. This is in contrast to work within our laboratory with other species, including rabbits (5, 8) and cotton rats (4), in which the relaxant responses at 20 Hz and less were almost completely abolished by addition of TTX to the tissue bath. The significance and possible reasons for TTX-insensitive relaxations in various species, including humans have been reviewed (27). Most pertinent to the results presented here, the findings in ferrets are similar to what is described in humans where TTX-resistant relaxations appear to be the major component of the relaxant response produced by field stimulation. This work suggests that HRSV affects both the TTX-sensitive (neural) as well as the TTX- resistant component of relaxation.

The consequences of any insult to an airway should be considered not only in terms of the acute effects on airway structure and function but also on the longer-term effects. Both the acute and more persistent alterations may vary in terms of severity as a function of the age at which the insult is delivered to the host. As noted previously, the insult may be especially harmful at a time of rapid somatic growth when the normal mechanisms of airway control are developing over time. For example, a series of studies with parainfluenza type 1 virus (Sendai) in rats has shown that neonatal infection of genetically susceptible rats leads to chronic abnormalities in airway morphology, histology, and physiology (28), whereas infections in weanling and older rats resolve more quickly and do not cause permanent airway structural abnormalities (31). Previous work from our laboratory has demonstrated that mechanisms of neural control develop postnatally in some species. Rabbits do not normally have a functional NANCi response at birth, and this pathway subsequently undergoes significant postnatal development (5). This study has demonstrated that another mammalian species (ferrets) normally develops this response only after a period of postnatal growth. We have demonstrated that both neonatal sensitization to an allergen in rabbits (5) and infection with HRSV in ferrets disrupt the normal functional development of this pathway. In both species, the abnormalities in NANCi responses are present at a time when the contractile responses to EFS are either normal (rabbit) or have normalized with time (ferret). Although these observations demonstrate the potential vulnerability of the NANCi system to differing insults, it is pertinent to note that recurrent aspiration early in life leads to a different pattern of dysfunction. In this latter model in developing rabbits, the NANCi responses normalized after aspiration ceased, whereas the EFS-induced contractile responses became more abnormal (17). Thus, the normal balance of an airway's contractile and relaxant responses are likely disrupted via different mechanisms depending in part on the nature of the insult and the stage of development of the airway at the time of the insult.

In summary, the present study was designed to determine if an RSV infection early in life in a mammalian model would lead to long-lasting alterations of neural control within the airways. We found that significantly reduced NANCi responses in infected animals persisted to at least 24 wk of age after the acute infection. This study further demonstrates the ability of RSV to produce dysfunction of this neural pathway. We also demonstrated abnormal cholinergic mechanisms, as demonstrated by increased contractile response to EFS in infected tissues. Although the abnormal contractile response resolved by the last time point, significant alterations were still present in terms of the NANCi response. Partial loss of the modulatory activity of the NANCi system might contribute to altered airway control in that factors that mediate relaxation are not functionally normal. Although the pathways by which respiratory viruses, including RSV, produce airway hyperresponsiveness are multifactorial, an alteration in the normal balance of neural responses within an airway may prove to be one of the more important consequences of infection with this virus when the infection is severe and/or occurs very early in life.