Inhibition of Arginase I Activity by RNA Interference Attenuates IL-13-Induced Airways Hyperresponsiveness¹

Recently, we have characterized the transcriptome of two phenotypically similar models of experimental asthma induced by systemic sensitization and subsequent aeroallergen challenge with OVA or by direct inhalation of Aspergillus fumigatus (1). Notably, genes related to the metabolism of basic amino acids, specifically the cationic amino acid transporter 2 and arginase I and arginase II, were prominent among the asthma signature genes in both models (1). An important role for extrahepatic activity of arginase I is now recognized for the regulation of aspects of immunological processes, in particular the modulation of T cell function (e.g., T cell proliferation and expression of TCR ζ-chain) and participation in Th2 immunity (2, 3, 4). We have also shown that arginase I activity is up-regulated in the respiratory epithelium of allergic asthmatics, in mononuclear cells with macrophage morphology, and in some granulocytes isolated in the bronchoalveolar lavage fluid (BALF)³ (1). Although we have linked increased arginase I activity with allergic inflammation of the lungs of both mice and of human asthmatics, the contribution of this pathway to disease processes remains unknown.

l-arginine is the common substrate for inducible NO synthase (iNOS) and arginase I (4, 5) and is catabolized to promote the production of NO or a range of second messengers by both of these enzymes, respectively. Recently, iNOS and arginase I activities were shown to be differentially modulated by CD4⁺ Th type 1 lymphocytes (Th1 cells) and Th2 cells and their cytokines (2, 6). Th1 cytokines (e.g., IFN-γ) activate iNOS (2), whereas Th2 cytokines (e.g., IL-13) promote arginase I activity through substrate depletion in a STAT6-dependent manner (7). Differential regulation of iNOS and arginase I activity by Th type 1 and 2 cytokines has also been observed in mice infected with Schistosoma mansoni and is a critical determinant of the pathogenic outcome (2). Th2 cells and cytokines also play a central role in the development and expression of allergic inflammation of the airways disorder, asthma, in a STAT6-dependent manner (8). Thus, the regulation of the metabolic fate of l-arginine by IL-13 and during Th2-dominated inflammatory responses (hepatic infection with S. mansoni) suggests that IL-13 and Th2 cells may also operate the arginase I pathway through STAT6 to regulate disease processes induced by allergic inflammation.

Of the Th2 cytokines regulating allergic airways disease, IL-13 plays a critical role in the development of airways hyperresponsiveness (AHR) (9, 10). The mechanisms underlying the development of AHR and diminished airflow in asthma are considered to play central roles in disease pathogenesis as these underpin bronchoconstriction (8, 11, 12). The induction of AHR by IL-13 is dependent on signaling through the IL-4Rα-chain (IL-4Rα) and STAT6 (13, 14). There is also an important link between STAT6 expression in airway epithelial cells and the ability of IL-13 to induce AHR (15). STAT6 activation in resident pulmonary cells appears to be a critical molecular switch for the development of AHR (15). Airway epithelial cells and pulmonary macrophages are activated by IL-13 (16, 17), and we have previously shown that IL-13 can promote AHR independently of the infiltration of inflammatory cells (into the airways) in a STAT6-dependent manner (14). Collectively, these studies support the concept that IL-13 acts directly on resident pulmonary cells to promote AHR. Although IL-13 plays a critical role in the induction of AHR during allergic inflammation of the lung, the molecular processes used downstream of STAT6 to modulate airways reactivity remain unknown.

The development of AHR has been linked previously to the regulation of l-arginine catabolism by arginase I (18, 19). Diversion of l-arginine away from NO production may have important implications for the regulation of airways responsiveness to contractile agents and in the development of chronic disease. NO is a bronchodilator, which can attenuate the effect of spasmogens (20, 21, 22, 23). Furthermore, the downstream metabolic products of l-arginine catabolism by arginase I could potentially affect cellular function in the lung that predisposes to altered airways responsiveness to spasmogens (AHR) and airways obstruction. Arginase I hydrolyzes l-arginine to urea and l-ornithine, the latter metabolite being essential for the production of proline and polyamines, which regulate collagen production and cell proliferation and differentiation, respectively (4, 24). Polyamines also modulate ion channel function and thus cell excitability (25, 26, 27). l-Arginine can also promote glutamate (and subsequently the neurotransmitter, γ-aminobutyric acid) production (28, 29).

Although arginase I activity is induced by allergen provocation of the lung in experimental models of asthma and is enhanced in the airways of allergic asthmatics, the factors up-regulating its activity and role in disease pathogenesis remain unknown.

In this investigation, we show that arginase I expression and activity in the lung are regulated by IL-13 and Th2-induced allergic inflammation in a STAT6-dependent manner. Furthermore, we dissected the specific interaction between IL-13 and arginase I for the development of AHR by delivery of this cytokine directly to the airways. The induction of arginase I expression in the lung by IL-13 directly correlated with the development, persistence, and resolution of AHR. Pharmacological manipulation of the iNOS/arginase I axis by supplementation with a NO donor or l-arginine inhibited and amplified IL-13-induced AHR, respectively. Moreover, induction of loss of function of arginase I specifically in the lung by employing RNA interference (RNAi) strategies resulted in attenuation of IL-13-induced AHR. This investigation identifies arginase I as a potential target for the treatment of IL-13-regulated bronchoconstriction in asthma.

Wild-type (WT) or STAT6-deficient (STAT6^−/−) BALB/c mice (6–8 wk) were obtained from the Specific Pathogen Free Facility or the Gene Targeting Facility, The John Curtin School of Medical Research, Australian National University. STAT6^−/− mice (30) were backcrossed for 12 generations with the BALB/c strain. Mice were treated according to Australian National University Animal Welfare guidelines and were housed in an approved containment facility.

Mice were sensitized at 6–8 wk of age by i.p. injection of 50 μg of OVA/1 mg Alhydrogel (CSL) in 200 μl of 0.9% sterile saline. Nonsensitized mice received 1 mg of Alhydrogel in 200 μl of 0.9% saline. On days 12, 14, 16, and 18, all groups of mice were aeroallergen challenged with OVA, as described previously (31). Twenty-four hours after the last aeroallergen (OVA) challenge, airways responsiveness to methacholine was measured or tissues and cell samples collected for analyses.

Twenty-four hours after the last aeroallergen challenge, lungs were perfused with saline, removed, and fixed in 10% neutral-buffered formalin before sectioning and staining with Carbol’s Chromotrope-Hematoxylin or Alcian blue/periodic acid-Schiff to identify eosinophils or mucus-secreting cells, respectively.

Mice were anesthetized with an i.v. injection of 100 μl of Saffan solution (1/4 diluted with PBS) and then the trachea was intubated with a 22-gauge catheter needle, through which murine rIL-13 (provided by Dr. D. Donaldson, Wythe Genetics Institute, Cambridge, MA) (at maximally effective dose: 10 μg dissolved in 20 μl of vehicle: 0.1% BSA, BSA/PBS) or control vehicle (0.1% BSA/PBS) was instilled into the airways. Airways responsiveness to β-methacholine (methacholine) (doses of 3.12, 6.25, 12.5, 25, and 50 mg/ml) was measured 12 h after instillation (or as indicated in the text).

Airways responsiveness to methacholine was assessed in conscious, unrestrained mice by barometric plethysmography, using apparatus and software supplied by Buxco. This system yields a dimensionless parameter known as enhanced pause (Penh) that reflects changes in waveform of the pressure signal from the plethysmography chamber combined with a timing comparison of early and late expiration. Measurement was performed essentially as we have described previously (14, 32). Notably, we have confirmed in the BALB/c strain that changes in Penh in response to methacholine directly correlate with changes in airway resistance to this spasmogen (our unpublished data). Thus, measuring changes in Penh reflects alterations in resistance and is indicative of enhanced airways responsiveness.

Airway reactivity (resistance: Raw) was also measured using a modification of the low-frequency forced oscillation technique as described previously (33, 34).

Mice were anesthetized with 0.1 ml per 10 g of body weight with a mixture containing xylazine (2 mg/ml; Troy Laboratories) and ketamine (40 mg/ml; Parnell). Two-thirds of the dose was given to induce anesthesia, with the remainder given when the animal was attached to the ventilator (Scireq). Additional doses were given each 40 min as required. Mice were ventilated with a tidal volume of 8 ml/kg at a rate of 450 breaths/min, with a positive end-expiratory pressure of 2 cmH₂O using a custom-designed ventilator (flexivent; Scireq). Mice were then allowed to stabilize for 5 min.

Once stabilized, mice were challenged with a saline aerosol followed by increasing concentrations of β-methacholine (Sigma-Aldrich) (5, 10, 20, and 40 mg/ml). Aerosols were generated with an attached nebulizer (Scireq) and delivered to the inspiratory line between the piston and the animal. Each aerosol was delivered for 10 s during which time regular ventilation was maintained. Immediately following aerosol delivery, five measurements of airway reactivity were made at 1-min intervals and Raw determined. Changes in Raw following methacholine challenge were calculated as a percentage increase over saline control (baseline).

WT mice were injected i.p. with pharmacological concentrations of l-arginine (200 mg/kg in 100 μl/PBS) or the same volume of PBS every 12 h for 4 consecutive days. High concentrations of l-arginine are required to modulate extrahepatic arginase I activity as the liver rapidly sequesters this amino acid from the circulation. On day 3, IL-13 (1 μg/20 μl of 0.1% BSA/PBS) or control vehicle (0.1% BSA/PBS) was instilled into the trachea as described above. For these experiments, IL-13 was used at 1 μg/20 μl (50% of maximal response) instead of 10 μg (maximal effective dose). AHR was measured at 12 h to observe if l-arginase could potentiate the development of AHR.

DETA/NO was prepared in an identical fashion to that described by Hrabie et al. (35). Vestiges of solvent were removed in vacuo and the material stored under a nitrogen atmosphere at −20°C until use. Mice were treated with 10 μg of IL-13/20 μl of 0.1% BSA/PBS (maximal effective dose) or control vehicle (0.1% BSA/PBS) and airways responsiveness to methacholine measured 48 h later. In some experiments, IL-13-treated or saline control mice were also injected i.p. with DETA/NO (100 mg/kg in saline) 5 min before exposure to methacholine. The control mice for the DETA/NO-treated mice received acetate-buffered DETA (metabolite of DETA/NO) at a dose of 63 mg/kg. DETA/NO is degraded rapidly to liberate NO and DETA (35). These experiments were conducted 48 h after IL-13 treatment, where the effect of IL-13 on lung responsiveness to methacholine was maximal, to determine whether pharmacological supplementation with systemic NO could suppress AHR.

Arginase activity was measured as described previously (36). Briefly, lungs were excised and cells lysed with 0.5 ml of 0.1% Triton X-100. After 30 min, 0.5 ml of buffer (25 mM Tris-HCl and 5 mM MnCl₂ (pH 7.4)) was added to the lysate. Arginase was then activated by heating the cell suspension for 10 min at 56°C. l-Arginine hydrolysis was conducted by incubating 25 μl of the activated lysate with 25 μl of 0.5 M l-arginine (pH 9.7) at 37°C for 60 min. The reaction was stopped with 400 μl of an acidic mixture (H₂SO₄, H₃PO₄, and H₂O; 1:3:7 v/v). Urea was measured at 540 nm after addition of 25 μl of 9% α-isonitrosopropiophenone (dissolved in 100% ethanol) and heating at 100°C for 45 min to quantify arginase activity.

Levels of RNI from serum or BALF were determined by Griess reagent assay as described previously (37).

TOPO TA vector (Invitrogen Life Technologies) was used to construct the plasmid vector for the expression shRNA under the control of the murine U6 promoter (Fig. 1). The murine U6 promoter was cloned from genomic DNA, flanked with BamHI at the 5′ end, and cloned in front of the 19-nt sense sequence, which was separated by a short hairpin spacer (loop) from the reverse complement of the same sequence (antisense). Six thymidines (T6) were used as the termination signal. Two plasmids expressing shRNA were generated, arginase I short hairpin (AHP)1 and AHP2, to specifically target arginase I expression. The oligonucleotides for AHP1 and AHP2 were synthesized chemically (Proligo) and flanked with EcoRI at the 3′ end. The oligonucleotides were annealed and ligated downstream of the U6 promoter. All constructs were sequenced to confirm identity.

The oligonucleotides sequences are as follows: shRNA1 (AHP1), sense strand, 5′-CCAAAGTCCTTAGAGATTATTCAAGAGATAATCTCTAAGGACTTTGGTTTTTTGAATTC, and antisense strand, 5′-GAATTCAAAAAACCAAAGTCCTTAGAGATTATCTCTTGAATAATCTCTAAGGACTTTGG; and shRNA2 (AHP2), sense strand, 5′-CGATTCACCTGAGCTTTGATTCAAGAGATCAAAGCTCAGGTGAATCGTTTTTTGAATTC, and antisense strand, 5′-GAATTCAAAAAACGATTCACCTGAGCTTTGATCTCTTGAATCAAAGCTCAGGTGAATCG.

Brewer’s thioglycolate (3 ml of 4% solution) (Difco) was injected into the peritoneum of BALB/c mice, and peritoneal-derived macrophage (PDM) were then isolated 3 days later by washing the cavity with PBS (3 ml). Residual erythrocytes were eliminated with red cell lysis buffer as previously described (31) and PDM plated at 5 × 10⁵ cells/well in 6-well plates with complete RPMI 1640 medium. After 6 h, nonadherent cells were washed away with PBS, and adherent cells were allowed to recover in fresh complete RPMI 1640 medium. Over 99% of adherent cells were macrophages as determined by FCS scanning. PDM were cultured for 4 days before being used in transfection experiments. PDM were transfected with 500 ng of either plasmid containing the shRNA construct or a blank plasmid, with the transfection reagent ExGen500 (Fermentas). ExGen500/DNA complexes were administered in 5% w/v glucose in a total volume of 50 μl. IL-13 was added to the culture medium at a final concentration of 50 ng/ml 8 h after transfection. After IL-13 stimulation for 12 h, PDM were recovered and frozen at −70°C until RT-PCR and Western blot analysis was performed. We used ExGen500 in our studies because this compound belongs to an efficient new class of nonviral, nonliposomal gene delivery reagents (38, 39). ExGen500 has been shown to efficiently deliver DNA to various organs and tissues via i.v. and intratracheal administration and by direct injection into tissues (40, 41, 42). Importantly, ExGen500 has shown no or minimal toxicity in all in vivo systems tested to date (39, 40, 41, 42).

shRNA or control vectors (25 μg of plasmid vector) were complexed with ExGen500 and 5% w/v glucose, and a total volume of 50 μl was instilled into the trachea (and again 24 h later) through a 24-gauge catheter following the manufacturer’s instructions. IL-13 was delivered to the lungs (as described above) 3 h after the last delivery of shRNA or control vectors. Lungs were then collected and frozen at −70°C for analysis of loss of function of arginase I activity.

The method for quantitative PCR has been described in detail elsewhere (43, 44, 45). Simply, RNA was prepared from cells or tissue using the TRIzol RNA isolation buffer following the protocol suggested by the manufacturer (Invitrogen Life Technologies). cDNA was synthesized by an oligo(dT) primed reverse transcriptase reaction using 0.5 μg of RNA from each sample. Quantitative PCR was performed in an ABI PRISM 7000 Sequence Detection System (Applied Biosystems) using murine arginase I (forward, GGTCCACCCTGACCTATGTGT, and reverse, ACGATGTCTTTGGCAGATATGC), arginase II (forward, CTGCCATTCGAGAAGCTGG, and reverse, GGGATCATCTTGTGGGACATTAG), iNOS (forward, ACATCGACCCGTCCACAGTAT, and reverse, CAGAGGGGTAGGCTTGTCTC) and GAPDH (forward, CAGGTTGTCTCCTGCGACTT, and reverse, CCCTGTTGCTGTAGCCGTA). SYBR-green was used to detect changes in amplicon levels with each sequential amplification cycle. The fluorescence intensity was normalized to the rhodamine derivative ROX as a passive reference label, which was present in the buffer solution. The level of mRNA, relative to GAPDH, was calculated using the formula: relative mRNA expression = 2^{−(Ct of sample − Ct of GAPDH)} × 10⁶, where Ct is the threshold cycle value. Since the levels of target mRNA were considerably lower than the level of GAPDH, all mRNA values shown were arbitrarily multiplied by 10⁶ and normalized against the background level detected in naive lung tissue.

Cells and lungs were harvested at the indicated time point, washed twice with cold PBS, and lysed in TNT lysis buffer (20 mM Tris, 200 mM NaCl, and 1% Triton X-100) plus protease inhibitors (mixture solution; Roche). After 20 min on ice, lysates were centrifuged at 15,000 × g for 10 min to remove insoluble material. Protein concentrations were determined by using the bicinchoninic acid assay (Pierce). Twenty micrograms of lysate was separated on 12% SDS/PAGE gels and transferred to a nitrocellulose membrane (Pall) using a Multiphor Novablot semidry transfer system (Amersham Biosciences). The membrane was blocked with 2% BSA in TBST (Tris-buffered saline, 0.05% Tween 20) and probed with a 1/1000 dilution of goat anti-mouse arginase I (Santa Cruz Biotechnology), followed by an incubation with a detection Ab conjugated to HRP (DakoCytomation). Western blots were developed by using ECL (Amersham Biosciences).

The significance of differences between experimental groups was analyzed using Student’s unpaired t test. Values were reported as the mean ± SEM. Differences in mean values were considered significant if p < 0.05.

Delivery of OVA to the lung (aeroallergen challenge) of OVA-sensitized mice resulted in the induction allergic airways inflammation dominated by AHR, eosinophilia, and mucus hypersecretion (Fig. 2, A–C). Aeroallergen challenge and the development of allergic airways inflammation directly correlated with increased expression of arginase I and with a concomitant decrease in the expression of iNOS (Fig. 2, D and E). Increased expression of arginase I in response to aeroallergen challenge also correlated with increased activity (urea formed nmol/min/mg; saline (Sal) 10 ± 4; OVA 118 ± 20, mean ± SEM, n = 4 mice/group, p < 0.05) indicating active catabolism of l-arginine by arginase I in the inflamed lung. Arginase II expression was not enhanced by the induction of allergic airways diseases (Fig. 2 F). Although we observed a slight reduction in the relative expression of iNOS, we could not detect any measurable change in the level of RNI, nitrites and nitrates (by-products of NO production), in the BALF or serum after OVA treatment by comparison to saline-treated mice (data not shown). These data support the concept that Th2-dominated inflammatory responses shift the balance of l-arginine metabolism away from NO production by altering the expression and activity of arginase I (2).

Next, we determined if IL-13 alone was sufficient for the induction of AHR and arginase I activity by delivery of this cytokine directly to the lung. IL-13-induced AHR (Fig. 3,A), concomitant with promoting increased expression of arginase I activity (Fig. 3,B) and enhanced production of urea (urea formed nmol/min/mg; Sal 15 ± 2.0; OVA 103 ± 8.4, mean ± SEM, n = 4 mice/group, p < 0.05). IL-13 administration resulted in decreased iNOS activity as the level of RNI in the lung were significantly reduced (Fig. 3 C). Notably, alterations in the expression of transcripts encoding arginase I and iNOS induced in response to allergic airways inflammation or IL-13 were also critically dependent on STAT6 (results not shown). Thus, the development of AHR and activation of arginase I activity in response to allergen provocation or IL-13 are both intimately linked via STAT6.

To create an additional link between arginase I activity and the regulation of IL-13-induced AHR, we correlated expression with the development, persistence, and resolution of AHR over time. IL-13-induced AHR developed within 12 h and persisted for 96 h (data not shown) but was resolved by 8 days (Fig. 4,A). Enhanced expression of arginase I, but not arginase II, directly correlated with the presence of AHR (Fig. 4, B and E). By contrast, the recruitment of eosinophils and induction of mucus production persisted after the resolution of AHR (Fig. 4, C and D). These data show a direct temporal alignment between arginase I activity and the development and resolution of AHR.

As an additional link between arginase I and NO and the regulation of IL-13-induced AHR, we treated mice with l-arginine or the NO donor, DETA/NO, before exposure to methacholine. l-Arginine treatment amplified the development of IL-13-induced AHR (Fig. 5 A). However, supplementation did not affect baseline responsiveness to methacholine, specifically linking l-arginine metabolism with the mechanism of IL-13-induced AHR to this spasmogen.

The effect of IL-13 on methacholine responsiveness was maximal 48 h after treatment (Fig. 5,B). Exposure of mice to the NO donor DETA/NO significantly suppressed IL-13-induced AHR, and responses were similar to those observed in unexposed controls (PBS-treated mice) (Fig. 5 B). Thus, both l-arginine and NO can pharmacologically modulate IL-13-induced AHR to methacholine, strongly implicating the iNOS/arginase I axis in the mechanism underlying enhanced airways responsiveness.

We generated two constructs based on the arginase I gene sequence designed to express shRNA that may result in loss of function of arginase I (AHP1 and AHP2). To examine the potential effect of AHP1 and AHP2 on arginase I expression, we initially transfected PDM with the respective vector in the presence of IL-13 (Fig. 6). As we observed with lung (Fig. 4), treatment of PDM with IL-13 induced the expression of arginase I but not arginase II. Arginase II was constitutively expressed in contrast to arginase I. Notably, AHP2 but not AHP1 significantly inhibited the expression of transcripts for arginase I. The vectors expressing the shRNAs did not affect arginase II expression. Furthermore, controls for the transfection reagents or expression vectors did not affect IL-13-induced expression of arginase I (Fig. 6) or the constitutive expression of arginase II. Next, we determined if interference of arginase I expression by AHP2 resulted in attenuation of protein production in our IL-13-stimulated PDM. IL-13 stimulation of PDM resulted in increased protein production of arginase I, which was only inhibited by AHP2 treatment (Fig. 6 B). As a final control for specificity, we showed that none of the treatments affected the expression and protein production of GAPDH.

Next, we determined if AHP2 was effective in silencing arginase I function in the lung. Not all designed shRNA has the ability to induce loss of function, which may be due to target mRNA accessibility and individual three-dimensional structure (46, 47). We used AHP1 as a specific control for shRNAi by AHP2. AHP2 specifically inhibited IL-13-induced expression and protein synthesis of arginase I in the lung (Fig. 7, A and B, respectively), which directly correlated with the development of AHR to methacholine (Fig. 7,D). AHP2 (or AHP1) did not affect the expression or protein production (GAPDH) of nonrelated molecules (Fig. 7, A and B) or of arginase II (Fig. 7,C). Significantly, loss of function of arginase I in the lung by treatment with AHP2 resulted in a marked attenuation of IL-13-induced AHR to methacholine (Fig. 7 D). These data in the lung show a direct and specific role for arginase I in the lung in the mechanism underlying IL-13-induced AHR to methacholine.

The potency of IL-13 in amplifying pathways that control bronchoconstriction has identified this cytokine as a crucial regulator of airways obstruction and AHR in asthma. In this investigation, we show by specifically inducing loss of function in the lung by RNAi (AHP2) that arginase I plays a critical role in the mechanism whereby IL-13 regulates AHR. Our study also highlights the use of RNAi as a means to induce transient loss of function of target molecules in the airways for the treatment of respiratory disorders.

In our experimental model of asthma, Ag-specific Th2 cells and IL-13 are essential for the development of AHR (31, 48). In the current investigation, we extend previous observations and show that Th2 regulated allergic inflammation in the lung shifts the balance of l-arginine metabolism (iNOS/arginase I axis) away from NO production by enhancing the expression of arginase I. Increased expression of arginase I occurred concomitantly with a decrease in expression of iNOS (Fig. 2). Notably, changes in the expression of these enzymes correlated with the development of AHR (2). Increased expression of arginase I in the lung also correlated with increased activity and formation of urea, indicating active catabolism of l-arginine by arginase I in response to allergen inhalation.

To dissect the interaction between IL-13 and arginase I activation in the lung, we delivered rIL-13 protein directly to the airways (Fig. 3). IL-13-induced AHR and likewise promoted increased activity of arginase I and decreased production of RNI. Increased activation of arginase I in association with a reduction in the level of NO production in the lung suggests that iNOS activity was decreased by substrate depletion (Fig. 3). Although the in vitro kinetics of l-arginine catabolism by iNOS and arginase I appear to normally favor NO production, increased activity of arginase I in intact cells (macrophages, epithelial cells, and at sites of wound healing) can significantly limit the availability of l-arginine for NO synthesis (2, 7, 18, 49, 50). Thus, similar to observations made in the Th2-dominated granuloma models induced by S. mansoni egg deposition in the liver (2) and in vitro studies on isolated macrophages (7, 51, 52), IL-13 selectively modulates arginase I expression in the lung by altering the balance of l-arginine catabolism.

Although Th2 cell and IL-13-induced arginase I activity has been linked previously to the degree of pathology in granuloma models, this pathway has not been associated with other IL-13-regulated processes such as AHR. In this study, we confirm the importance of STAT6 for the development of AHR induced by allergic airways inflammation or IL-13 (data not shown). However, we extend previous observations to show that STAT6 also regulates the expression of arginase I in the lung in response to allergen inhalation or IL-13 instillation (data not shown). Furthermore, induction of arginase I expression by IL-13 temporally correlates with the development, persistence, and resolution of AHR. By contrast, other effects mediated by IL-13 such as mucus hypersecretion and eosinophil accumulation are temporally dissociated from the development and persistence of AHR (14). Bronchial hyperreactivity in asthmatics can also occur independently of the development of pulmonary inflammation (53, 54, 55, 56). This disassociation of airways inflammation and AHR has been confirmed in animal models of allergen-induced lung disease (57, 58, 59). Eosinophilia and mucus hypersecretion (albeit reduced) are also features of allergic airways inflammation in IL-13^−/− mice (60, 61); however, AHR is critically dependent on this cytokine. Thus, our data on the development of AHR with rIL-13 mimic that observed in animal models of allergen-induced allergic airway diseases. Furthermore, we have linked the three critical components (Th2 cells, IL-13, and STAT6) for the development of AHR with the activation of arginase I in the lung. However, we cannot exclude the possibility that other, as yet uncharacterized pathways, may also contribute to the mechanism whereby IL-13 through STAT6 mediates AHR in the allergic lung.

However, in models of allergic lung disease STAT6 activity and AHR are intimately linked and regulated by IL-13 (15, 62). To further implicate the regulation of the arginase I in the mechanism of IL-13-induced AHR, we treated mice pharmacologically with l-arginine or the NO donor DETA/NO. By this approach, we could directly determine the effect of l-arginine supplementation on airways reactivity during increased activation and expression of arginase I by IL-13. Furthermore, we could ascertain, through NO supplementation, the potential contribution of iNOS and subsequent NO production on modulating the mechanism whereby IL-13 specifically regulates airways responsiveness. Importantly, treatment of mice with l-arginine did not have an effect on baseline airway reactivity to methacholine (Fig. 5 A). However, supplementation of mice with l-arginine potently amplified IL-13 induced AHR. These data clearly demonstrate that increased substrate availability for arginase I, only at the time of IL-13 exposure, selectively promotes AHR to methacholine, further highlighting the link between these factors for the control of bronchoconstriction. By contrast, treatment of mice with a NO donor (DETA/NO) inhibited the development of IL-13-induced AHR. Thus, NO can play an important protective role against IL-13-induced processes that promote increased airways smooth muscle reactivity. Collectively, these data strongly suggest that the metabolic fate of l-arginine plays a central role in regulating IL-13-induced AHR.

To evaluate the direct contribution of arginase I activity in the lung to IL-13-induced AHR, we induced loss of enzyme function by shRNAi. AHP2 effectively blocked IL-13-induced expression and protein production of arginase I (but did not effect arginase II) in isolated macrophages and in the lung after transfection. Notably, inhibition of arginase I function in the lung resulted in marked attenuation of IL-13-induced AHR, and methacholine responses were reduced to near baseline. A number of products of l-arginine catabolism are cell-signaling molecules that can alter smooth muscle cell excitability (28, 29). Activation of arginase I in the lung may also reduce NO levels, which may promote increased airways reactivity to methacholine (5, 29, 63).

In summary, our investigation shows that Th2-regulated allergic airways inflammation and IL-13 induces alterations in l-arginine metabolism by activating arginase I in the lung in a STAT6-dependent manner. Moreover, we show that IL-13 induces arginase I expression and activity, which promote the development and persistence of AHR to methacholine, a hallmark feature of asthma. These observations may have important implications for the development of new approaches, downstream of Th2 cells and cytokines, to modulate IL-13-regulated bronchial hyperresponsiveness in asthma.

We gratefully acknowledge the generous support of Dr. D. Donaldson (Wyeth Genetics Institute, Cambridge, MA).

The authors have no financial conflict of interest.