Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Airway hyperresponsiveness (AHR) is a fundamental feature of asthma, which is influenced by multiple genetic and environmental factors (1, 2). It consists of two components: (1) hyperexcitability, exemplified by a leftward shift of the dose-response curve; and (2) hyperreactivity, exemplified by an upward shift of the curve. Although several abnormalities can cause allergic AHR, recent advances have led to the view that two major factors contribute to its pathophysiology. The first is chronic inflammation of the allergic airway, which is initiated by increased allergen-induced IgE synthesis. Many cells, such as T cells, mast cells, eosinophils, and epithelial cells, play a role in this latter process (3, 4). This inflammation, along with associated edema of the bronchial wall and variable degrees of airway smooth-muscle (ASM) hypertrophy and hyperplesia, result, via a geometric factor, in airflow limitation. It must be noted that inflammation itself can alter airway smooth-muscle responsiveness. The second factor contributing to the pathophysiology of AHR is nonspecific hyperreactivity of the airway smooth-muscle cells (SMCs) (5). This causes difficulties in breathing that are usually associated with widespread but variable and reversible bronchospasm, and an increase in airway reactivity to a variety of stimuli in susceptible individuals.

Mechanical studies of strips of ASM in the canine model demonstrated that upon sensitization with antigen, maximal shortening capacity (L_max) and maximum velocity of shortening (V₀) were increased, whereas isometric force (P₀) showed no change (6). The change in L_max was of sufficient magnitude to increase airways resistance by 75%. However, because of the limitation in size of the mouse ASM, such studies have not been heretofore conducted in strips of this muscle, even though the sensitized mouse is a good model for human asthma (2, 7, 8).

Recent studies in rodents suggest that T cells play an important role in induction of the inflammatory cascade in the airways and lungs (9, 10). In mice, depletion of CD4 T cells prevented antigen-induced AHR and pulmonary eosinophilia (11). Studies with interleukin-4 (IL-4)-deficient mice demonstrated that absence of IL-4 not only blocked the Th2 cytokine response, but also prevented AHR (12, 13). Furthermore, inbred strains of mice differed in their ability to develop antigen-induced AHR as studied with electrical stimulation of tracheal smooth muscle (TSM) strips in vitro (8). Moreover, mouse strains differed in their acetylcholine- and methacholine-induced AHR in vivo, measured as air pressure-time index and pulmonary resistance (14). For the air pressure-time index and pulmonary resistance phenotypes, three putative loci controlling bronchial hyperreactivity (BHR) were said to have been identified (2, 14). In none of these studies, however, was hyperreactivity demonstrated, rather, only a change in excitability was reported, nor was ASM contractility directly measured. Parenthetically, it may be postulated that the latter may also be under genetic control.

Previously, to identify genes controlling IgE responsiveness, Mohapatra and associates established a mouse colony using ASW and SJL inbred strains of parent mice, which are high and low IgE responders, respectively (15). In the current study, we assessed whether increased ASM reactivity existed in mice, since it could provide an explanation for allergic bronchospasm. We report that smooth-muscle contractility can be validly measured in mice, and that two inbred strains of mice, ASW and SJL, differed in their mechanical response to electrical stimulation, suggesting a genetic basis for the airway hyperreactivity phenotype. Furthermore, the SJL mouse strain, which is a low IgE responder but expresses relatively high levels of the cytokines IL-4 and IL-5, demonstrated hyperreactivity of its ASM.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Mice

Inbred strains of ASW/SnJ and SJL/J mice were obtained from Jackson Laboratories, Bar Harbor, ME. In all, four groups of animals were studied: ASW control (ASW-C), ASW sensitized (ASW-S), SJL control (SJL-C), and SJL sensitized (SJL-S). Among the four groups of mice examined, only the weights of SJL-C and SJL-S differed from each other by 14% (P < 0.05), which is unlikely to be physiologically important. Also, no significant difference was found in muscle strip cross-sectional area (CSA) among the different groups.

Immunization of Mice

Groups of mice were immunized with ovalbumin (OVA) (Sigma Chemicals, Ltd., St. Louis, MO) coprecipitated with alum as adjuvant. Individual animals were injected intraperitoneally with 0.5 ml of inoculum comprising 5 µg OVA with 1 mg alum. Animals were challenged with similar doses of antigen after 3 wk, using the same preparation, and were bled 7 days thereafter to obtain samples for estimation of IgE levels. Total IgE was measured with an enzyme-linked immunosorbent assay and specific IgE with the passive cutaneous anaphylaxis (PCA) test.

Cytokine Measurement

IL-4 levels were determined in splenocyte cultures of mice, as described elsewhere (15). Amounts of IL-4 secreted were determined with an enzyme-linked immunosorbent assay (ELISA) in accord with the manufacturer's instructions. Briefly, 96-well ELISA plates were coated with 20 µg/ml purified rat antimouse IL-4 in coating buffer. After overnight incubation at 4°C, the plates were blocked for 2 h at room temperature and washed. Recombinant IL-4 standard and serial dilutions of culture supernatants at 100 µl per well were added to the plate. Plates were incubated at 4°C overnight, then washed four times. Biotinylated rat antimouse IL-4 (1 µg/ml) was added and incubated at room temperature for 1 h. The plates were then washed, and streptavidin-alkaline phosphatase was added for 1 h at room temperature. The plates were then washed again and p-nitrophenyl phosphate was added as the substrate. The plates were read with an automatic reader. The lower limit of detection for IL-4 was 2.5 pg/ml.

To measure cytokine expression at the messenger RNA (mRNA) level, total RNA from splenocyte culture was subjected to amplification with the reverse transcriptase- polymerase chain reaction (RT-PCR) according to procedures described previously (15).

Mechanical Studies

Mechanical studies were done with previously established methods (16). Briefly, mice were anesthetized with 0.5 ml of a 60 mg/ml solution of pentobarbital administered intraperitoneally. All tissues were placed immediately in ice-cold oxygenated (95% O₂/5% CO₂) Krebs-Henseleit solution of the following composition: 115 mM NaCl; 1.38 mM KH₂PO₄; 25 mM NaHCO₃; 2.5 mM KCl; 2.46 mM MgSO₄ · 7 H₂O; and 5.56 mM dextrose. Rings were cut transversely from the upper trachea and bisected with a dorsal midline section through the cartilage ring. Gentle eversion revealed the muscle strip. We have previously described this dissection method (6). The cartilage ends were pared down to provide attachment sites, and the epithelium was carefully removed. The cartilage at one end was fixed tightly in a clamp at the bottom of a jacketed glass bath at 37°C, which contained gassed Krebs-Henseleit solution including 1.91 mM CaCl₂. The cartilage at the upper end was attached via monofilament to the lever of a custom-built electromagnetic myograph that was originally developed by Brutsaert and colleagues (17). The output signals (force and shortening as functions of time) from the lever were digitized and transmitted to a computer that, with a customized software program, controlled the system and collected data for analysis and storage. The control system allowed abrupt (within 3 ms) changes in load on the muscle strip, leading to a new constant load (load clamp) at any desired time point, and also allowed critical damping to be applied. This permitted elicitation of force-velocity data at any time during the course of contraction. The Hill equation was fitted to the data, and a coefficient of determination (R²) was computed to demonstrate goodness of fit. The equation is that for a displaced rectangular hyperbola, and has been shown to apply to striated and smooth muscles. The equation is: (P + a)(v + b) = (P₀ + a)b, where P is the load applied to the muscle, v is the resulting maximum velocity for that load, and a and b are asymptote values for the hyperbola. a has units of force and b has units of velocity.

At the end of the equilibration period (30 min for mice), the ASM strip was stimulated every 4 min and allowed to contract isometrically at different muscle lengths in order to permit the determination of optimal muscle length (L₀), the length at which the muscle generates maximum active tension (P₀). Stress was expressed in mN/mm² of the total CSA of the strip, the latter being determined with a previously described method (16). After L₀ was determined, conventional force-velocity (16) and unloaded velocity-time data (16) were obtained through quick-release techniques. For these measurements, the muscle strip was always set at its predetermined optimal length (L₀).

We examined the mechanical properties of TSM in the inbred ASW and SJL strains of mice. Strips of TSM of both antigen-sensitized and littermate control mice, measuring 1 mm × 0.5 mm × 0.25 mm, were examined for development of maximum isometric force (P₀), maximum shortening capacity (L_max), and maximum velocity (V₀). Supramaximal electrical stimulation was employed, and the quick releases were made during the course of the ensuing isometric contraction. The activated muscle strips were released to a set of different, randomly chosen loads that were below the P₀ value. This resulted in different magnitudes and rates of shortening of the muscle for the different isotonic loads. The maximum shortening or displacement (Disp) developed by the muscle for any given load at 2 s after stimulus onset was measured from the curves in the lower panel of Figure 2A. The maximum shortening capacity was measured at a preload (1.25 mN load) equal to that required to stretch the muscle to its optimal length, L₀. The maximum slope of the slow transient following the rapid one in the same panel provided the V₀ for that particular load.

View larger version (31K): [in this window] [in a new window]   View larger version (21K): [in this window] [in a new window]   View larger version (10K): [in this window] [in a new window]   Figure 2.   Derivation of isotonic parameters. (A) Original recordings of isometric force (P0) and shortening (Lmax) versus time data obtained by quick release for various loads at 2 s of contraction. Top: Traces of force versus time elicited by electrical stimulation. Load clamps were applied at 2 s; the muscle strip was abruptly released to seven different loads (1.25 mN to 2.70 mN). Bottom: Shortening as a function of time at different loads: the magnitudes of different loads are shown by arrows. (B) Zero-load shortening of the muscle at different times during an isometric contraction obtained by quick-release and plotted with a customized computer program. Note that shortening is depicted as increasing in an upward direction. The maximal slope of the slow transients at the zero-load clamps at different times represents the maximal velocity of shortening at that time. (C) Velocities under zero-load shortening, plotted versus time. Force in both A and B is expressed in millinewtons (mN) and shortening as displacement (Disp) in millimeters (mm).

To confirm that mouse TSM resembles other mammalian TSMs in having normally cycling bridges and slowly cycling or latchbridges, zero-load velocity was determined by applying zero-load clamps to an isometrically contracting muscle at different points in time. It was expected that velocity at 2 s after stimulus onset would be maximum (normally cycling bridges), and would thereafter show a decrease of 20 to 30% at 8 to 10 s (demonstrating latchbridges).

Histology

Tracheas, including cervical and thoracic segments, were removed from individual ASW and SJL mice and equilibrated in Ca²⁺-deficient, aerated Krebs-Henseleit solution for 60 min. The tissues were then fixed in 10% formalin, and tracheal cross-sections (5 µm) were obtained, stained with hematoxylin and eosin (H&E), and studied with light microscopy.

Statistical Analysis

Mean values of measured variables were compared with a one way analysis of variance (ANOVA) with fixed constants; Duncan's new multiple range test was used to compare means on a post hoc basis, with P  0.05.  values were set at 0.80. Results were expressed as mean ± SE.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

IgE Antibody Levels in ASW and SJL Mice

We examined whether the differential reactivity of smooth muscles from the ASW and SJL strains of mice was related to any immunologic feature pertinent to the induction of inflammation. Comparison of the IgE levels in these two strains after allergen-sensitization indicated that SJL mice produced lower amounts of IgE antibodies than did ASW mice (Figure 1). Both total and OVA-specific IgE titers were significantly higher in ASW than in SJL mice. It is concluded that these strains differ in their IgE responsiveness. The IgG response of the two strains was similar (data not shown).

View larger version (11K): [in this window] [in a new window]   Figure 1.   Comparison of total and OVA-specific IgE levels in ASW and SJL mice. The total IgE was measured with an ELISA and specific IgE was measured with the passive cutaneous anaphylaxis (PCA) test. These results are shown in A and B.

Comparison of Mechanical Properties of SJL and ASW Inbred Strains

Figure 2A shows simultaneous records of force and Disp elicited by applying quick releases to different loads at 2 s after the onset of the electrical stimulus. The displacement traces in the lower panel of Figure 2A manifest a rapid transient (vertical deflection at the 3-s point) that originates from elastic recoil of the muscle's series elastic component. This is followed by a slow transient that is generated by the shortening of the contractile element. The maximum slope represents the maximum velocity (V₀) developed for the given load. These V₀ values are plotted against their isotonic loads. Velocity points are elicited for a set of loads and are used to plot the force-velocity curve of the muscle.

Smooth muscle differs from striated muscle in that the properties of its crossbridges do not remain the same throughout contraction. The properties of the bridges are determined by measuring their velocities at different points in time but always at the same load. Under these zero-load conditions, the velocity of the crossbridges at different time points is the same in skeletal muscle. To determine the time-dependent behavior of crossbridges in ASM, zero-load clamps were applied at different times. The results are shown in Figure 2B. The upper panel of the figure shows force records for zero-load applied at 1-s intervals during an isometric contraction. The corresponding shortening traces are shown in the lower panel. The maximum velocities for each slow transient are plotted against time and delineated in Figure 2C. The heterogeneous behavior of the crossbridges is clearly seen. Maximum V₀ develops at 1.25 s and decreases progressively thereafter. At 8 s, velocity has decreased by 20%. This confirms that ASM shows normally cycling bridges (2 s) and slowly cycling or latchbridges (8 s). This distribution is important for studies of asthmatic ASM. All four groups of muscles studied showed the same time-dependent behavior of velocity.

Using quick-release (to a range of different loads) techniques at 2 s after stimulus onset, we obtained force-velocity curves for ASM from sensitized SJL mice (SJL-S) and controls (SJL-C); these are shown in Figure 3. Although the maximum isometric force is the same for the two strains, the V₀ value for SJL-S mice is significantly (P < 0.05) greater than that for SJL-C mice. However, V₀ values obtained by quick releases at 8 s exhibited no significant difference and are therefore not shown.

View larger version (13K): [in this window] [in a new window]   Figure 3.   Experimental points obtained by quick release at 2 s for four SJL-C (open circles) and five SJL-S mice (closed circles). The two curves shown were computed with least-squares fitting techniques. P = force, P0 = maximum force, stress = P0/muscle cross-sectional area, L0 = optimal muscle length, a and b = asymptote values of the rectangular hyperbolic curves; a is in units of force and b in units of velocity.

Figure 4A shows mean maximum shortening capacities as percentages of L_max of muscles stretched to optimal length (L₀) by a suitable resting load, which was about 5% of P₀. The mean value for smooth muscle from ASW-S mice (30.0 ± 2.2 L₀) was very significantly greater (P < 0.01) than that for ASW-C mice (23 ± 3.6% L₀). The L_max for SJL-S mice (42 ± 3.6 L₀) was significantly greater than that for SJL-C mice (24 ± 6.5 L₀). A significant difference was found between control and sensitized muscles in the proportion of isotonic shortening occurring within the first 2 s. Seventy-five percent of the total shortening achieved by the muscle was complete within 2 s, which points up the importance of this measurement. The shortening seen at steady state, (i.e., at 8 s), is merely due to that already occurring at 2 s. Figure 4B shows mean ± SE values for V₀ at 2-s quick release. The differences found between control and sensitized muscle strips from ASW mice were not significant (P > 0.05), whereas those from SJL mice were significant (P < 0.05). Velocity is shown in L₀s per second. Values of maximal isometric force in percentages were not significantly different for control and sensitized ASM. The 100% value in SJL-S mouse ASM, shown in Figure 3, was 1.2 × 10⁵ ± 2.1 × 10⁴ N/m² (mean ± SE) in measured stress units.

View larger version (15K): [in this window] [in a new window]   Figure 4.   Comparison of ASW and SJL inbred strains with respect to the mechanical properties of their TSM. Groups of five to seven mice, consisting of both control and OVA-sensitized animals, were examined for (A) Lmax, (B) V0, and (C) P0.

Histology of Tracheal Smooth Muscle

Histologic analysis of tracheal smooth muscle demonstrated that SMCs were oriented in a parallel fashion along the long axis of the muscle strips used for mechanical studies (Figure 5). The orientation of the cells was assessed from an approximation that the cigar-shaped nuclei lie in the long axis of the cells. No differences were observed in the gross morphology of the tracheal muscle from ASW or SJL mice; the tracheal medial layer was about 0.25 mm thick and consisted of approximately nine to 14 layers of myocytes in both preparations (Figure 5). The photomicrographs shown in the figure demonstrate that murine smooth muscle inserts at the tips of the semicircular, tracheal cartilage rings; the tips of the latter were left in place on the smooth-muscle strip preparations used in mechanical studies, since they provided anchor points for attachment to the myograph. This method of attaching the strips to the myograph ensured that no distortion occurred in the orientation of the SMCs in the tracheal strips during in vitro mechanical studies, and that there was no injury to the muscle.

View larger version (121K): [in this window] [in a new window]   Figure 5.   Photomicrographs comparing the morphology of TSM from ASW and SJL mouse strains. Epithelium (E) was not removed from the samples processed for histology. All sections were stained with H&E. (A) Tracheal cross section (magnification: ×33) of an upper cervical segment of ASW trachea, demonstrating the site of muscle insertion on the tips of the cartilage rings. (B) Higher power (magnification: ×152) micrograph of A, showing the parallel arrangement of SMCs as evidenced by the orientation of nuclei in the medial layer. (C) Tracheal cross section (magnification: ×33) of an upper cervical segment of the trachea of an SJL mouse; muscle insertions, on the opposing tips of a cartilagenous ring, are shown by arrows. (D) Higher power (magnification: ×152) micrograph of C. C = cartilage, E = epithelium, SM = smooth muscle.

Cytokine Gene Eexpression in ASW and SJL Mice

We examined the expression of cytokines in splenocytes of ASW and SJL mice following immunization. For this purpose, groups of 13 ASW and 14 SJL mice were immunized with OVA-absorbed alum and stimulated in vitro with OVA or the polyclonal activator concanavalin A (Con A). Analysis of profiles of cytokines, including IL-2, IL-4, and IL-5, by RT-PCR, using RNA of splenocytes of ASW and SJL mice, showed that SJL mice expressed higher levels of mRNA than did ASW mice for all these cytokines (Figure 6A). Levels of IL-4 protein were determined by ELISA. Interestingly, under antigen stimulation, the high IgE responder strain ASW secreted lower levels of IL-2, IL-4, and IL-5 than did the low responder SJL strain (P < 0.01 for IL-2 and IL-4) (Figure 6B). Moreover, following Con A stimulation, significant differences between these two strains were also found in IL-4 production (P < 0.01). However, there was little difference in IL-2 secretion by polyclonal activation. These results suggest that the SJL mouse, though characterized by low IgE levels, can nevertheless induce significant levels of IL-4 and IL-5. Taken together, these results indicate that the SJL mouse, which was found to be a low IgE responder, produced higher levels of IL-4 and IL-5 than did the ASW mouse, and exhibited significant hyperreactivity of TSM.

View larger version (27K): [in this window] [in a new window]   Figure 6.   Cytokine production by ASW and SJL strains of mice. Splenocytes of OVA-sensitized mice were cultured in vitro with either Con A (nonspecific mitogen) or OVA. (A) RT-PCR analysis for IL-2, IL-4, and IL-5 with splenocyte mRNA after in vitro culture with OVA. C = control, A = antigen sensitized. (B) IL-4 protein measurements in supernatants of splenocyte cultures of mice.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Asthma may develop from the interaction of a number of factors, including allergic sensitization, airway inflammation, ASM hyperreactivity, and ASM hypertrophy and hyperplasia, each of which may be genetically controlled (1, 2, 14, 15). However, asthma does not necessarily involve a cascade of interactions, since asthmatic individuals without allergic sensitization can be found. Although definitive data are not available, owing to the difficulty in obtaining tissue from asthmatic individuals, the use of and response to bronchodilators suggests that the BHR of asthmatic patients results from abnormal bronchial smooth-muscle (BSM) contractile properties (5, 18). Similarly, sensitized canine TSM was shown to undergo nonspecific alterations in mechanical properties (19). These observations, along with the finding of Cockcroft and associates of nonspecific hyperreactivity of airway SMCs (5), suggest that the ultimate defect in allergic AHR may be at the muscle-cell level. We hypothesized that ASM hyperreactivity is a genetically controlled property, independent of IgE-mediated hypersensitivity. In view of the polygenetic control of asthma, development of a mouse model of asthma is considered important. In the present study we developed a murine model to study contractile properties of TSM. Observed differences between two IgE- and IL-4-disparate mouse strains in mechanical characteristics of their ASM suggest that the progenies of these strains will aid in clarifying the inheritance pattern and genetic analysis of BHR.

A major finding in the present study is that the contractile properties of tracheal SMCs, specifically in terms of shortening capacity (L_max) and velocity (V₀), can be measured in the mouse in a manner similar to that in which these properties are measured in other, larger animal models. Analysis of our results with a murine model suggest that upon allergen sensitization, these mice also develop allergic AHR, which can be quantitated ex vivo by examining the contractile properties of their TSMs, and that these results are similar to those previously reported for dogs (20). The majority of in vitro mechanical studies of ASM have been done with tracheas instead of bronchi, because obtaining a strip of pure BSM, especially in the mouse, is technically difficult. However, the mechanical properties of TSM and BSM of the dog were found to be similar. Further, TSM has been considered a mechanical model for airways down to the level of those generations of bronchi in which allergic AHR occurs (6). To date, all studies of airway muscle from mice, including our own, have been conducted on TSM, since mouse BSM is too small for in vivo studies. Nonetheless, even though we believe that these studies in mice trachea do reflect the mechanical properties of BSM, a caveat must be entered to the effect that this is not necessarily so, since we ourselves have reported quantitative differences in the mechanical properties of extrapulmonary and intrapulmonary ASM (24). We feel it highly unlikely that if the properties of control and sensitized TSM were different, those for BSM would not be. In support of this, we would like to point out that in tissues as dissimilar from the trachea as the saphenous vein, we have observed exactly the same changes in smooth muscle as we report here for TSM. Finally, we acknowledge the limitations of in vitro studies, and acknowledge that extrapolation from these to what occurs in vivo is perhaps hazardous. However, there are several reports of plethysmographic, in vivo measurements of AHR in antigen-sensitized rats (23) and direct-pressure measurements in vivo in sensitized mice (14). The changes in reactivity found in these studies were exactly in the same direction as we report here for the isolated ASM of the mouse.

Furthermore, our results demonstrate that control and sensitized SJL mice do not differ with respect to the development of isometric force (P₀). The SJL mouse therefore represents an excellent model for studying the mechanical properties of TSM, but only if shortening parameters are also studied in addition to isometric force.

Another major finding of this study was that the two strains of mice examined showed significant differences, in that only the SJL strain exhibited significant changes in L_max and V₀ of TSM. Since mice were housed in the same quarters and were under identical environmental conditions, the difference between these strains suggests that the contractile properties of smooth muscle are genetically controlled. Clearly, the nature of genetic elements and the pattern of inheritance remain to be elucidated. This will require examination of segregation of phenotypes in backcross progenies of mice derived from the inbred ASW and SJL strains of mouse. Because both L_max and V₀ significantly differed between control and sensitized mice, it is likely that both properties are coordinately regulated. It is worth pointing out that in both the canine and the murine models, increases in V₀ and L_max are only seen at 2 s. At 8 s no differences were seen in these parameters. This suggests that sensitization affects only the relatively rapidly cycling crossbridges that are active early in contraction. The slowly cycling or latchbridges appear to be unchanged. Thus, all of the mechanical changes accounting for allergic bronchospasm are essentially complete within 2 s. The changes computed at steady state (i.e., 8 s and later) are essentially those engendered at 2 s. The importance of conducting studies of smooth-muscle shortening early in contraction is clear.

It should be noted in passing that considerable differences exist in the contractile velocities of normally cycling and latchbridges. In canine and murine smooth muscle we found a decrease in velocity of only 30%, while others have found as much as a 400% decrease in hog carotid smooth muscle (21). The reason for this is unknown; it may relate to differences in smooth muscles (airway versus vascular) or in animal species (hog versus mouse), or to differing degrees of phosphorylation of the 20-kD myosin light chain (21).

Studies with ASW and SJL mice suggest that these changes in mechanical properties of TSM may be related to, and be driven by, the immunologic consequences of a sensitizing antigen. Our studies demonstrate that the SJL mouse, which produces significant amounts of IL-4 and IL-5 upon sensitization with antigen, also shows alterations in the mechanical properties of its TSM. Both immunologically specific and nonspecific mechanisms appear to control the inflammatory responses in asthma. Specifically, antiinflammatory cytokines such as IL-4 and IL-5 play important roles, since they initiate the inflammatory cascade, and their importance was evident in studies with IL-4-deficient (12, 13) and IL-5-transgenic mice (22). Although there is no evidence for this at present, it is likely that IL-4 and IL-5 affect the magnitude of airway inflammation, which in turn may influence TSM contractile properties.

The finding in the present study that the SJL mouse, despite its increased production of IL-4 and IL-5, produced low levels of IgE suggests that IgE is not a major player in the TSM hyperresponsiveness observed in this mouse. Increased levels of serum IgE have been positively correlated with AHR, but there is currently no proof that the genes governing IgE and AHR are the same. In human studies, polymorphic markers associated with IgE and AHR have been localized to chromosome 5q, but it has been suggested that the genes governing these phenotypes are different. In mice, however, the loci controlling methacholine-elicited AHR are located on chromosomes 2, 15, and 17 (2). We have recently found that a marker on a chromosome other than these shows significant association with the IgE response (S. Mohapatra, unpublished observation). Our results showing differences in segregation of IgE phenotype and TSM mechanical properties suggest that there exist separate genes for IgE responsiveness and AHR.