Requirement for IL-13 Independently of IL-4 in Experimental Asthma

See allHide authors and affiliations

Science  18 Dec 1998:
Vol. 282, Issue 5397, pp. 2261-2263
DOI: 10.1126/science.282.5397.2261


The pathogenesis of asthma reflects, in part, the activity of T cell cytokines. Murine models support participation of interleukin-4 (IL-4) and the IL-4 receptor in asthma. Selective neutralization of IL-13, a cytokine related to IL-4 that also binds to the α chain of the IL-4 receptor, ameliorated the asthma phenotype, including airway hyperresponsiveness, eosinophil recruitment, and mucus overproduction. Administration of either IL-13 or IL-4 conferred an asthma-like phenotype to nonimmunized T cell–deficient mice by an IL-4 receptor α chain–dependent pathway. This pathway may underlie the genetic associations of asthma with both the human 5q31 locus and the IL-4 receptor.

Allergic asthma is a complex disorder characterized by local and systemic allergic inflammation and reversible airway obstruction. Asthma symptoms, especially shortness of breath, are primarily related to airway obstruction, and death is almost invariably due to asphyxiation (1). Increased airway responsiveness to provocative stimuli, termed airway hyperresponsiveness (AHR), and mucus hypersecretion by goblet cells are two of the principal causes of airway obstruction observed in asthma patients (2). Data from animal models consistently reveal a critical role for TH2 (T helper 2) cells and potentially important roles for the cytokines IL-4 and IL-5 (3–7).

TH2 cells selectively develop and expand in the presence of IL-4 (8). To separate direct effects of IL-4 from developmental effects on TH2 cells in an asthma model, we compared the ability to establish the asthma phenotype in BALB/c mice deficient in either IL-4 or the IL-4 receptor α chain (IL-4Rα) (9). After intranasal challenge with the antigen ovalbumin (OVA), BALB/c mice developed a stereotyped asthma phenotype characterized by eosinophil influx of the airways, goblet cell metaplasia with mucus overproduction, and an increase in AHR as revealed by enhanced sensitivity to acetylcholine challenge (6,7). IL-4 and IL-4Rα–deficient mice showed incremental attenuation of each of these asthma indices (Fig. 1, C through E) (10). Thus, in agreement with prior studies (5–7), IL-4 contributes to the asthma phenotype, but these data suggest an independently greater contribution by IL-4Rα.

Figure 1

PAS-stained histologic sections of murine lungs. Arrowheads point to goblet cells within the respiratory epithelium. (A) Wild-type mice were primed with OVA and challenged with PBS intranasally. (B) Wild-type mice were administered IL-13 intranasally. (C) IL-4–deficient and (D) IL-4Rα–deficient mice were primed with OVA and challenged with OVA intranasally. Wild-type mice were primed with OVA and challenged intranasally with (E) OVA and human Fc control protein or with (F) OVA and IL-13R-Fc. Note the marked reduction in goblet cells in (D) and (F).

IL-13 is a cytokine closely related to IL-4 that binds to IL-4Rα and is also expressed by TH2 cells from asthma patients (11). To assess whether IL-13 might contribute to the asthma phenotype, we administered a soluble IL-13 recetor α2-human Fc fusion protein (IL-13R-Fc) to BALB/c mice sensitized to OVA and compared them to mice that received control protein (12). IL-13R-Fc selectively binds to and neutralizes murine IL-13 but not IL-4 (13). This treatment significantly attenuated the asthma phenotype, although little effect was seen on neutrophil influx into bronchoalveloar lavage (BAL) (Figs. 1, E and F, and 2). Thus, IL-13, like IL-4 (5–7), can contribute to the acute effector phase of experimental asthma.

Figure 2

Effect of neutralization of IL-13. Primed wild-type mice were administered intranasally human immunoglobulin (Ig control), Ig control and OVA, or IL-13R-Fc and OVA as indicated by (+). Data for (A) AHR, (B) goblet cell score, and numbers of (C) eosinophils and (D) neutrophils in the BAL fluid are plotted as means ±SEM. *P < 0.05 relative to PBS and Ig control–treated mice; †P < 0.05 relative to OVA and Ig control–treated mice. Data are representative of at least two comparable experiments with four to eight mice per group.

To assess the capacity of IL-13 and IL-4 to cause pathology independently of T and B cells, we administered each cytokine to nonimmunized BALB/c and RAG1 (recombinase activating gene 1)–deficient mice (14). Each cytokine alone induced the asthma phenotype (Figs. 1, A and B, and3). In contrast, administration of either cytokine to IL-4Rα–deficient mice resulted in no significant changes in any asthma parameter, demonstrating that their effects were dependent on signals mediated by IL-4Rα. Further, adoptive transfer of OVA-specific TH2 cells to IL-4Rα–deficient mice failed to elicit the asthma phenotype, whereas identical treatment of wild-type mice resulted in the full phenotype (15, 16). Thus, experimental asthma induced by antigen challenge, recombinant cytokine, or adoptive transfer of TH2 cells, is mediated through a final pathway dependent on IL-4Rα.

Figure 3

Effect of recombinant IL-4 and IL-13. Wild-type (WT), RAG1-deficient (RAG1−/−), and IL-4Rα–deficient (IL-4Rα–/–) mice were administered IL-4, IL-13, or control protein intranasally. Data for (A) AHR, (B) goblet cell score, and numbers of (C) eosinophils and (D) neutrophils in the BAL fluid are plotted as means ±SEM. *P < 0.05 relative to mice receiving control protein. Data are representative of at least two comparable experiments with four to eight mice per group.

Attenuated asthma phenotypes observed in IL-4–deficient mice may now be interpreted as representing the effects of residual IL-13 derived from IL-4–deficient TH2 cells (17). Parallel observations in experimental intestinal helminth infections demonstrate roles for both IL-4 and IL-13 in mediating critical final effector pathways via IL-4Rα (18). It is possible that human asthma represents a spectrum of disease also linked by a shared receptor effector pathway. The common embryological origin of tissues from the gut and lung (19) would support the presence of stereotyped responses in these organs.

The relevance of our data to human asthma remains an important issue that cannot be entirely addressed, given the complexity of the disease and the inadequacies of any animal model. Linkage analysis has mapped susceptibility to asthma to a region on human chromosome 5q25-31, which includes the genes for both IL-4 and IL-13 (20), and to mutations in two domains of the α chain of the IL-4 receptor (21). A number of additional regions in the genome have been linked to asthma in human studies, suggesting a complex multifactorial phenotype (22). As we suggest, however, diverse forms of asthma might follow a final common effector pathway mediated through signals transduced by IL-4Rα, thus creating a unified target for potential intervention.

  • * To whom correspondence should be addressed. E-mail: habari{at}


View Abstract

Navigate This Article