Free light chain (FLC) concentrations are demonstrated to be increased in different inflammatory disorders and are proposed to mediate mast cell–dependent immune responses. A role for mast cells is suggested in chronic rhinosinusitis with nasal polyposis (CRSwNP), which is characterized by a local Th2 inflammatory response. However, clear mast cell–activating factors are not always apparent. In this study, the presence of FLCs in CRS patients with or without nasal polyps (CRSw/sNP) was investigated and the effect of different treatments on FLC expression was analyzed.
Nasal tissue, nasal secretion, and serum of control patients, patients with CRSwNP, and CRSsNP were analyzed for the presence of kappa and lambda FLC. The expression of FLCs in nasal polyp tissue was investigated using immunohistochemistry. In addition, FLC was measured in serum and nasal secretion of nasal polyp patients treated with methylprednisolone, doxycycline, anti-IL-5, or placebo.
Free light chain concentrations were increased in nasal secretion and mucosal tissue homogenates in patients with chronic rhinosinusitis, and this effect was most prominent in CRSwNP patients. Immunohistochemical analysis confirmed the increased FLC concentrations in nasal polyp tissue. In CRSwNP patients, treatment with methylprednisolone or anti-IL-5 resulted in the reduction in systemic or local FLC concentrations, respectively.
The presence of FLC in CRSwNP and CRSsNP suggests a possible role in mediating the local immune reaction in the paranasal cavities. Furthermore, the decrease in local FLCs after treatment with anti-IL-5 presumes that IL-5 creates an environment that favors FLC production.
Chronic rhinosinusitis (CRS) is a chronic inflammatory disorder of the nose and paranasal cavities, which is currently divided in CRS without (CRSsNP) and with nasal polyps (CRSwNP) based on the histology and inflammatory patterns . CRSsNP is considered as a fibrotic disease characterized by a Th1 cell-mediated inflammation with high numbers of neutrophils and high concentrations of transforming growth factor-β1 (TGF-β1) and interferon-γ (IFN-γ) [2, 3]. Nasal polyps observed in CRSwNP are benign edematous masses protruding from the nasal and paranasal mucosa. Most Caucasian CRSwNP patients have a massive polyp infiltration of inflammatory cells, mainly eosinophils [2, 3]. The local Th2 polarization observed in these polyps is characterized by high levels of eosinophilic markers [eosinophil cationic protein (ECP) and eotaxin], IL-5, and local production of polyclonal IgE [4, 5]. However, the mechanisms underlying CRS remain poorly understood.
Activation of mast cells can result in an immediate release of granule-stored mediators, and a large array of cytokines, chemokines, and growth factors several hours after activation . A functional role of mast cells is suggested in ex vivo experiments showing nasal polyp mast cell activation upon IgE-crosslinking [7, 8] and by the finding that continuously released mast cell mediators enhanced the recruitment of eosinophils in nasal polyp tissue . Recently it was shown that mast cells are localized in the glandular epithelium of NPs in a unique pattern and that they have a distinct phenotype compared with controls . The involvement of IgE in nasal mast cell activation in CRSwNP patients is supported by local polyclonal IgE production in the stroma and the epithelium of nasal polyps [11-13]. In contrast to CRSwNP, the number of mast cells in CRSsNP patients is not significantly different from healthy controls. Moreover, putative functional differences between these diseases are unknown [14, 15].
Next to IgE, immunoglobulin free light chains (FLCs) can mediate antigen-specific mast cell activation . B cells produce light chains [either of kappa (κ) or lambda (λ) isotype] that can be secreted as unbound FLCs. Free light chain concentrations are increased in many inflammatory disorders, including asthma and rhinitis [17-19], and high FLC concentrations are shown to correlate with an increased disease severity in rheumatoid arthritis and multiple sclerosis [19-21].
Interestingly, preclinical models showed that FLCs might be involved in models for contact hypersensitivity and asthma by mediating mast cell activation via a yet unknown receptor in an antigen-specific manner [16, 18]. Because FLCs are able to specifically bind antigen [16, 22], investigating their possible involvement in (allergic) diseases associated with allergen-specific non-IgE-mediated mast cell activation is of great interest. In this study, the presence of FLCs in CRSwNP and CRSsNP patients was investigated and the effect of different treatments on FLC expression was analyzed.
The analysis of FLC concentrations in tissue homogenates, nasal secretion, and serum was performed on 25 control subjects, 46 patients with CRSsNP, and 41 patients with CRSwNP, included at the Department of Otorhinolaryngology of the University Hospital Ghent, Belgium, between May 2007 and December 2009. The inclusion and exclusion criteria are described in the Supporting Information.
The effects on serum and nasal secretion FLC concentrations were analyzed in CRSwNP patients included in two randomized controlled trials investigating three alternative treatment options as previously published [23, 24]. In the first study, 14 patients received methylprednisolone (32 mg/day on days 1–5; 16 mg/day on days 6–10; 8 mg/day on days 11–20), 14 received doxycycline (200 mg on day 1; 100 mg/day on days 2–20), and 19 received placebo (lactose for 20 days). Nasal secretion and serum were analyzed at baseline and 2 weeks after treatment . The second study included 20 patients, who received mepolizumab [anti-IL-5; two single IV injections (4 weeks interval) of 750 mg], and ten patients, who received placebo. The primary endpoint was 8 weeks after the first administration of medication .
The ethical committee of the Ghent University Hospital, Ghent, Belgium, approved the studies, and written informed consents were obtained from all subjects.
Methods of sample collection (serum, nasal secretion, and tissue homogenates), immunohistochemical staining, and statistical methods are described in the Supporting Information.
Measurement of FLC, IgE, and other inflammatory parameters
Total κ- and λ-FLC concentrations were determined in all supernatants and sera using an ELISA adapted from Abe et al. [24, 25]. In brief, plates were coated (o/n; 4°C) with goat–anti-mouse IgG (M4280; Sigma, Zwijndrecht, The Netherlands), blocked (1 h; RT), and incubated with mouse–anti-human κ- or λ-FLC, which bind free light chains exclusively [24, 25] (obtained from Dr. A. Solomon, Tennessee, US). After incubation with different dilutions of samples and standards (The BindingSite, Birmingham, UK), plates were incubated with HRP-labeled goat F(ab′)2-anti-human κ- or λ-light chain antibodies (AHI1804 and AHI1904, respectively; Biosource, Life Technologies Europe, Bleiswijk, the Netherlands). TMB was used as substrate. Per sample, at least three data points within the range of the standard curve were used to estimate the FLC concentration.
Different inflammatory parameters were analyzed by ELISA (IL-5, IFN-γ, IL-8, IL-6 and MPO; R&D Systems, Minneapolis, MN, USA) and Uni-CAP system (ECP, total and specific SAE IgE; Pharmacia, Uppsala, Sweden) according to the manufacturer's instructions.
Demographic and clinical features
Age and asthma incidence differed significantly between patient groups analyzed for FLC in tissue homogenates, nasal secretion, and serum (Table S1 in Data S1). Because serum FLC concentrations are hardly influenced by age in adults (<80 years) , this observed difference in age is not likely to bias current results. Patients with CRSwNP have a higher risk for developing asthma, which is demonstrated by more asthmatics in the CRSwNP group compared with CRSsNP. Concentrations of the analyzed inflammatory parameters in the different disease groups are shown in Table 1. No significant differences in baseline characteristics between different treatment groups were observed, as published previously [23, 24].
Table 1. Different inflammatory parameters in tissue, nasal secretion, and serum in control subjects, CRSsNP and CRSwNP patients
Control (n = 25)
Control vs CRSsNP
CRSsNP (n = 46)
CRSsNP vs CRSwNP
CRSwNP (n = 41)
CRSwNP vs Control
Differences between groups were analyzed for each parameter by a Kruskal–Wallis test followed by a post hoc Dunn's multiple comparison test on all patient groups. Corresponding P-values are indicated. All numbers represent mean (SEM).
CRSwNP, chronic rhinosinusitis with nasal polyposis; CRSsNP, chronic rhinosinusitis without nasal polyposis; ECP, eosinophil cationic protein; FLC, Free light chain.
Total kappa FLC (μg/ml)
Total lambda FLC (μg/ml)
Total IgE (kU/l)
SAE IgE (kU/l)
Total kappa FLC (μg/ml)
Total lambda FLC (μg/ml)
Total IgE (kU/l)
Total kappa FLC (mg/L)
Total lambda FLC (mg/L)
Total IgE (kU/l)
SAE IgE (kU/l)
FLC concentrations are increased locally in mucosal tissue and nasal secretion in patients with CRS
Free light chain concentrations in control, CRSsNP, and CRSwNP subjects are shown in Table 1 and Fig. 1A–C.
Correlations between κ-FLC and λ-FLC concentrations measured in nasal secretion (control: P < 0.001, r = 0.94; CRSsNP: P < 0.001, r = 0.71; CRSwNP: P < 0.001, r = 0.81), mucosal tissue homogenates (control: P < 0.001, r = 0.90; CRSsNP: P < 0.001, r = 0.72; CRSwNP: P < 0.001, r = 0.76), and serum (control: P < 0.001, r = 0.90; CRSsNP: P < 0.001, r = 0.82; CRSwNP: P < 0.001, r = 0.86) were highly significant in all groups, indicating a polyclonal response.
Analysis of correlations between FLC and different inflammatory parameters in tissue indicated a significant correlation between FLC and total IgE (κ: P = 0.002; λ: P = 0.003), ECP (κ: P = 0.050; λ: P = 0.016), IL-5 (κ: P = 0.013; λ: P = 0.004), and IL-6 (κ: P = 0.006; λ: P = 0.007) in CRSwNP. No correlations were found with IFN-γ in CRSwNP. The amount of FLC in CRSwNP was independent of the presence of SAE IgE. In CRSsNP, a significant correlation was found for FLC and MPO (κ: P = 0.006; λ: P = 0.023) and IL-8 (κ: P = 0.020; λ: P = 0.014). In contrast, no significant correlations were found in control patients.
FLCs are diffusely distributed in nasal polyp tissue
Numbers of tryptase-positive mast cells, B cells, and plasma cells were all significantly higher in polyp tissue as compared to control tissue (mast cells: 6.7 (2.9) vs 20.7 (4.2); P = 0.050; B cells: 4.8 (4.4) vs 33.4 (9.3); P = 0.02; plasma cells: 0.1 (0.1) vs 45.8 (17.7) P = 0.008, control vs NP, mean ± SEM). Detection of κ- and λ-FLCs was analyzed using antibodies detecting only free light chains and not those attached to complete immunoglobulins . Control nasal mucosa showed little FLC-positive cells (Fig. 2A), whereas nasal polyp tissue showed more intense FLC staining (Fig. 2B). The diffuse staining in polyp tissue likely results in an underestimation of the number of FLC-positive cells. Isotype control antibodies did not show any positive staining in polyp tissue (Fig. 2C). We did not detect clear colocalization of FLC with either tryptase or CD138-positive cells.
Analysis of FLC staining intensities showed a significant difference for λ-FLC between both groups (κ-FLCs: control: 11.0 ± 2.8; NP: 21.2 ± 3.3; P = 0.1, λ-FLC: control: 10.1 ± 1.5; NP: 20.1 ± 2.9; P = 0.03; Fig. 3A, left panel). Correlations between the number of FLC-positive cells and staining intensity were highly significant for both κ-FLCs (P = 0.003; r = 0.67, Fig. 3A, middle panel) and λ-FLCs (P = 0.002; r = 0.67, Fig. 3A, right panel). Moreover, both κ-and λ-FLC staining intensities highly correlated with FLC concentrations in NP tissue homogenates from the same tissue analyzed with the FLC ELISA (κ-FLCs: P < 0.001; r = 0.87, λ-FLC: P = 0.001; r = 0.68, Fig. 3B). A significant correlation with the number of FLC-positive cells was only observed for κ-FLCs (P = 0.003; r = 0.80).
Serum and nasal FLC concentrations after treatment with methylprednisolone and doxycycline in CRSwNP
Baseline FLC concentrations in serum and nasal secretion were not significantly different between the two treatment groups. Serum FLC concentrations were significantly decreased in CRSwNP patients after treatment with the systemic corticosteroid methylprednisolone (κ-FLCs: before: 35.4 ± 2.76; after: 28.6 ± 2.62, P < 0.001; λ-FLC: before: 33.8 ± 2.43; after: 29.6 ± 2.27, P = 0.005). Placebo treatment did not affect FLC serum concentrations (κ-FLCs: before: 34.5 ± 3.31; after: 36.3 ± 2.78, P = 0.59; λ-FLC: before: 30.3 ± 3.25; after: 32.6 ± 2.95, P = 0.19) (Fig. 4A–B). Free light chain concentrations in nasal secretion were unaffected. Doxycycline treatment did not affect serum or nasal FLC concentrations (data not shown).
Mepolizumab treatment significantly reduces local FLC concentrations in patients with severe nasal polyposis
Baseline FLC concentrations in serum and nasal secretion were not significantly different between the two treatment groups. Anti-IL-5 treatment (mepolizumab) significantly reduced both κ- and λ-FLC concentrations in nasal secretion (κ-FLCs: before: 18.3 ± 3.02; after: 12.2 ± 4.17, P = 0.047; λ-FLC: before: 15.8 ± 3.76; after: 9.11 ± 3.59, P = 0.03). In this study, placebo treatment did not affect nasal FLC concentrations (κ-FLCs: before: 28.6 ± 9.33; after: 47.3 ± 17.76, P = 0.65; λ-FLC: before: 23.0 ± 6.2; after: 44.0 ± 18.09, P = 0.91) (Fig. 5A–B). Two placebo-treated patients showed highly increased FLC concentrations after treatment resulting in high mean FLC concentrations and increased standard deviations. Serum FLC concentrations were unaffected by mepolizumab. Serum and nasal total IgE did not significantly change after treatment . Although 12 of 20 responded to treatment, correlations between changes in FLC concentrations and treatment effect were not found. Moreover, baseline FLC (and IL-5) concentrations in polyp tissue homogenates did not predict the treatment outcome.
In this study, the presence of FLCs in CRSwNP and CRSsNP patients was investigated and the effect of different treatments on FLC expression was analyzed. CRS patients had highly increased FLC concentrations in mucosal tissue and nasal secretion, and this effect was most prominent in CRSwNP patients. These increased FLC concentrations were confirmed by immunohistochemical analysis of polyp tissue. Interestingly, local FLC concentrations are reduced in CRSwNP patients after anti-IL-5 treatment. Together, these observations further support an ongoing local immune reaction in the paranasal cavities of CRSwNP and CRSsNP patients and indicate a role for IL-5 in creating an environment that favors FLC production.
Chronic rhinosinusitis with nasal polyposis in Caucasian patients is characterized by a Th2-mediated eosinophilic immune response with high IL-5, ECP and IgE concentrations, whereas CRSsNP is mainly a Th1 immune response with high levels of IFN-γ and TGF-β . Highest FLC concentrations were found in tissue homogenates and nasal secretion from CRSwNP patients, which is in line with high local IgE, IgA, and IgG concentrations and increased CD19 + B cells and plasma cell numbers in NP tissue [2, 27]. Moreover, FLC concentrations in CRSwNP tissue are correlated with IL-5, IL-6, and local IgE. This polyclonal IgE production is independent of an atopic status, and Staphylococcus aureus is shown an important activator of this IgE production [4, 5]. Although less pronounced, FLC concentrations were also increased in mucosal tissue from CRSsNP patients and correlated with MPO and IL-8 concentrations. Further research should delineate whether these correlations are causally related.
On the basis of these findings it appears that local FLC production is not specifically associated with either a Th1- or Th2-mediated inflammatory response. This is in line with previous findings showing increased FLC production in both IgE- and non-IgE-mediated disorders, including asthma and rhinitis [17, 18]. Compared to most CRSwNP cases, eosinophilic infiltration of the nasal mucosa is characteristic for nonallergic rhinitis, especially in nonallergic rhinitis with eosinophilia syndrome (NARES) patients . In this latter group, we previously showed that FLCs are highly increased in nasal secretion . Moreover, the number of FLC-positive cells in nasal mucosa including mast cells and plasma cells, as detected by laser microdissection and subsequent RT-PCR analysis, were increased in both IgE- and non-IgE-mediated rhinitis patients as compared to controls. Both CRSwNP and NARES are characterized by a tissue eosinophilia. Therefore, FLCs might be complementary actors in both disorders next to tissue eosinophilia.
Previous results from our group showed that FLCs are able to mediate antigen-specific mast cell activation [16, 18, 22]. At present, the data in this study do not provide proof for a functional role for FLCs in mediating NP pathology. Chronic rhinosinusitis with nasal polyposis pathology is highly associated with Staphylococcus aureus infection because Staphylococcus aureus enterotoxin (SAE) specific IgE is present in nasal polyp tissue and its ex vivo cross-linking leads to mast cell–mediated early phase-like responses [5, 29]. Because FLCs are also able to specifically bind antigen, it would be therefore interesting to analyze the antigen binding of nasal FLCs to SAE. Furthermore, as combined crosslinking of IgE and FLC results in synergistic mast cell activation in in vitro assays , a possible functional role for FLCs in local mast cell activation could be further underscored. Unfortunately, in contrast to the cell-localized FLC staining pattern observed in nasal mucosa from rhinitis patients , the diffuse FLC staining in NP tissues hampered the identification of possible FLC-positive cells to further substantiate this hypothesis.
Recently, the therapeutic effect of oral glucocorticosteroids and antibiotics (methylprednisolone and doxycycline, respectively) was investigated in a double blind randomized manner, demonstrating a reducing nasal polyp size compared with placebo . In line with the lack of a clear decrease in local IgE concentrations after these treatments, nasal FLC concentrations were not affected. Factors that influence FLC production remain largely undetermined. The decrease in serum FLC concentrations after methylprednisolone treatment can be related to its potent systemic anti-inflammatory action. This effect was absent in doxycycline-treated patients, despite the ability of doxycycline to inhibit the immunoglobulin production .
IL-5 is highly expressed in nasal polyp tissue, and this cytokine likely plays a critical role in chemotaxis, activation, and survival of eosinophils [2, 32]. It is proposed that eosinophils highly contribute to nasal polyp growth by the release of toxic mediators. In accordance with these findings, anti-IL-5 treatment with reslizumab induced a decrease in eosinophil count in serum and of ECP levels in nasal secretion . Moreover, mepolizumab, a humanized monoclonal anti-IL-5 antibody, showed to be effective in reducing severe nasal polyposis in 12 of the 20 patients analyzed in this study . We observed a clear decrease in nasal FLC concentrations after anti-IL-5 treatment in these 20 patients, whereas nasal IgE and serum IgE and FLC concentrations were unaffected . Even though clear effects of IL-5 on B-cell function are described in mice , it is unclear how blockage of IL-5 leads to a specific decrease in FLC production in human. The discrepancy between the effects of anti-IL-5 treatment on IgE and FLC concentrations suggests that different mechanisms are responsible for their local expression. Alternatively, this finding could be explained by the existence of cells that appear to selectively produce FLC instead of complete immunoglobulins, which is also observed in patients with NARES  and in idiopathic pulmonary fibrosis patients who showed little IgE and IgG but highly increased FLC in their BAL fluid . Clear regulatory mechanisms involved in FLC production by B cells are lacking; though, it has been demonstrated that unstimulated mature B cells are mainly responsible for FLC production [36, 37].
In summary, FLCs are highly present in nasal mucosa and nasal polyp tissue of CRSsNP and CRSwNP patients, respectively, emphasizing the local production. The exact functional role of the increased local FLC expression, including their potential role in mediating local mast cell activation, needs to be further investigated. Methylprednisolone and doxycycline treatment of CRSwNP patients did not reduce local FLC concentrations, although serum concentrations significantly decreased after corticosteroid treatment. In contrast, anti-IL-5 treatment of severe CRSwNP patients dramatically decreased nasal FLC concentrations. Future studies should elucidate whether IL-5 directly influences FLC production and whether interference in FLC production/effector function may contribute to the therapy of nasal polyposis.
Tom Groot Kormelink and Lien Calus were involved in data collection, statistical analysis, and interpretation of data. Both were involved in writing the manuscript. Natalie De Ruyck was involved in the sample collection. Gabriele Holtappels assisted during the experimental analyses. Claus Bachert and Philippe Gevaert supervised the clinical studies, contributed to the sample collection interpretation of the data, and controlling of the manuscript. Frank Redegeld contributed to the interpretation of the data and controlling of the manuscript. All authors critically revised the manuscript and approved the final version. The opinions, results, and conclusions reported in this article are those of the authors and are independent of the funding sources. The corresponding author had full access to all data in the study and had final responsibility for the decision to submit for publication.
Conflict of interest
All authors declare that they have no conflict of interest.