Serum sphingolipid profile in asthma

Abstract

Sphingolipids metabolism is an important cell process and plays critical roles in asthma. However, the involvement of sphingolipids in the pathogenesis of asthma and its subtypes is unknown. The present study aimed to determine the role of sphingolipids in asthma and its subtypes. Clinical data from 51 asthma patients and 9 healthy individuals were collected and serum samples were performed to analyze the levels of serum sphingolipids by liquid chromatography-mass spectrometry-based targeted metabolomics. Results showed that the levels of sphingomyelin (SM) including SM34:2, SM38:1, and SM40:1 were significantly decreased in asthmatic patients compared to healthy controls. Moreover, serum SM levels were obviously decreased in the blood noneosinophilic asthma (bNEA) group compared with blood eosinophilic asthma group. Similar tendencies of serum SM level changes were observed in the early-onset group compared with late-onset group. Correlation analysis revealed that SM 40:1 was negatively related to sputum IL-17A (r = −0.621, P = 0.042). The present study presented that the SM may be a protective factor of asthma and contributes to the mechanism of asthma, especially bNEA. SM may be a potential biomarker and therapeutic target in asthma.

ceramide, eosinophilic asthma, noneosinophilic asthma, sphingomyelin

INTRODUCTION

Asthma, characterized by chronic airway inflammation, tissue remodeling, and airway hyperresponsiveness, is one of the most common chronic diseases worldwide.¹ More than 300 million people suffer from asthma, and the prevalence continues to increase globally.² A combination of genetic predisposition, allergens, and viral triggers are strong risk factors of asthma, but the pathogenesis of asthma is poorly understood.

Sphingolipids are diverse and complex lipids accompanied by abundant variations in their sphingoid bases, fatty acids, and head groups,³ and involved in several disorders.⁴ Sphingomyelin (SM) species and glycolipids found in bronchoalveolar lavage fluid from patients with asthma was the first evidence for a role of sphingolipids in asthma.⁵ Accumulating evidence also revealed that sphingolipids, especially ceramide (Cer), glycosphingolipid, and sphingosine-1-phosphate (S1P), are important factors involved in asthma.⁶ Inhibition of Cer synthesis aggravated allergen-induced airway inflammation.⁷ Conversely, reducing glycosphingolipid levels in mast cells partially reversed asthma manifestations in mice.⁸ S1P is an important proinflammatory and anti-apoptotic factor that played an essential role in the development of asthmatic phenotype in different experimental models in mice.⁹ SMs are key regulatory steps of sphingolipids synthesis and degradation. SM levels are maintained by catabolic action of neutral or acid sphingomyelinase (ASMase), releasing Cer and the corresponding head group, phosphorylcholine, or by anabolic action of sphingomyelin synthase (SMS), basing on Cers as substrate.¹⁰ In addition, ASMase was found to be involved in the regulation of the typical T helper 2 phenotype¹¹ and was implicated in inflammatory diseases, such as acute lung injury¹² and rheumatoid disease.¹³ However, the role of SM in asthma remains elusive.

According to previous studies, we speculated that sphingolipid metabolism may be implicated in the mechanisms of asthma and in different subtypes of asthma, especially the different inflammation subtypes. However, most previous studies on sphingolipid metabolism have rarely focused on asthma. In this study, we tested and explored the differences in sphingolipid levels among asthma patients and between different asthma subgroups. These results provide new insight into the underlying mechanisms involved in this heterogeneous disease.

MATERIALS AND METHODS

Clinical data collection

Fifty-one patients with asthma recruited from the outpatient clinic at Peking University Third Hospital between January 2015 and December 2015 were diagnosed based on the 2014 Global Initiative for Asthma guidelines. Patients with other lung diseases, obstructive sleep apnea-hypopnea syndrome, malignant disease, acute or chronic respiratory failure, or severe cardiovascular disease were excluded. Nine healthy volunteers who underwent a physical examination at our hospital were enrolled as the control group. All patients were required to undergo pulmonary function tests using spirometry (Elite series, MGC Diagnostics, St. Paul, MN, USA), and the percentage predicted forced expiratory volume in 1 s (FEV₁%pred) and FEV₁/forced vital capacity (FVC) were recorded. The study was approved by the Ethics Committee of Peking University Third Hospital (approval ID 2014071). Written informed consent was provided by all subjects.

Liquid chromatography-mass spectrometry, cytokines analysis, and ASMase activity assay

Serum sphingolipid levels were measured by ACQUITY ultra performance liquid chromatography (UPLC) using a UPLC BEH C18 column (1.7 μm, 100 × 2.1 mm i.d.; Waters, Milford, MA, USA) as previously described.¹⁴ The scanning strategy used for multiple-reaction monitoring of sphingolipids is shown in Table 1. The concentrations of cytokines, including IL-4, IL-5, IL-10, IL-13, IL-17A, and IFN-γ, in the peripheral blood and induced sputum were measured with the LEGENDplex Multi-Analyte Flow Assay Kit (BioLegend, San Diego, California, USA). TGF-β was detected with the Human TGF-beta 1 Quantikine ELISA Kit (R&D Systems, Inc., Minneapolis, Minnesota, USA). The concentration of matrix metalloprotease 9 (MMP9) was measured with the Human MMP-9 Quantikine ELISA Kit (R&D Systems, Inc., Minneapolis, Minnesota, USA). ASMase activity was measured with the Acid Sphingomyelinase Assay Kit II (Colorimetric) (BioVision, San Francisco, California, USA).

TABLE 1

Sphingolipid scanning strategy using liquid chromatography-mass spectrometry

LC-MS/MS run	Lipid class	Internal standard	Parent ion	Daughter ion	CE
1	Ceramide (Cer)	Cer d18:1/17:0	[M+H]+	264.4	42
1	Sphingomyelin (SM)	SM d18:1/17:0	[M+H]+	184.1	52
2	Ceramide-1-phosphate	Cer d18:1/17:0	[M-H]-	78.9	−65

Grouping criteria for asthma subgroups

A body mass index (BMI) < 18.5 kg/m² was considered underweight, a BMI ≥ 25 kg/m² was considered overweight, a BMI ≥ 28 kg/m² was considered obese, and other values were considered normal. Onset ages ≤12 yr old were assigned to the early-onset group, and the others were assigned to the late-onset group. Asthma control was determined according to the asthma control test (ACT).¹⁵ A score of 20 to 25 points was considered good control, 15 to 20 points was considered general control, and 5 to 15 points was considered poor control. The inflammatory subtypes of asthma were grouped according to the quantity of inflammatory cells in induced sputum and blood, respectively.^16,17 Based on sputum eosinophil and neutrophil proportions, airway inflammation in asthma can be categorized into the sputum eosinophilic asthma (sEA) group (sputum eosinophil proportion > 3% and sputum neutrophil proportion ≤61%), sputum neutrophilic asthma (sNA) group (sputum neutrophil proportion > 61% and eosinophil proportion ≤3%), sputum mixed asthma (sMA) group (sputum neutrophil proportion > 61% and eosinophil proportion > 3%), and sputum oligogranulocytic asthma group (sputum neutrophil proportion ≤61% and eosinophil proportion ≤3%). According to the absolute eosinophil count in the blood, blood eosinophilic asthma (bEA) group (≥300 cells/μl) and blood noneosinophilic asthma (bNEA) group (< 300 cells/μl) were determined.

Statistical analysis

Multivariate analysis of sphingolipid profiles was performed using an online statistical tool (http://www.metaboanalyst.ca). T-tests and fold change analyses were used to identify sphingolipids that differed significantly between two groups. ASMase activity between two groups was analyzed using t-tests. Analyses were performed with GraphPad Prism (GraphPad Software, La Jolla, CA, USA).

RESULTS AND DISCUSSION

Comparison of serum sphingolipid profiles between asthma patients and healthy controls

A total of 51 asthma subjects, among of which 46 cases had valid data, and 9 age and gender-matched healthy individuals were included in this study. Serum SM levels in asthma patients were significantly lower than in healthy controls, whereas serum levels of Cer and Ceramide-1-phosphate were not changed (Fig. 1A). Previous studies have described relationships between SM in the bronchoalveolar lavage fluid or erythrocyte membrane and asthma,^18,19 whereas few focused on the serum SM levels of asthma patients and healthy controls. The present study found that the serum SM levels including SM34:2, SM38:1, and SM40:1 were significantly decreased in the asthma patients, indicating that the serum SM may be thought of as a protective factor in asthma. Another study also showed that serum SM levels were decreased in patients with aspirin-exacerbated asthma during the lysine-aspirin bronchoprovocation test,²² which is consistent with our result. In addition, Zheng et al. reported that SM increased intracellular magnesium concentrations in cultured vascular smooth muscle cells and induced bronchial relaxation,^20,21 which may explain, at least in part, why the reduction in serum SM levels is involved in the pathogenesis of asthma.

FIGURE 1

Sphingolipids profiles of patients and subgroups. Serum levels of sphingomyelin (SM) in asthma patients were lower than that in the normal healthy (A). No significant differences of SM level were detected in patients with normal body mass index (BMI) compared to overweight and obesity (B), in patients of different asthma control test (ACT) control grades (C) and in patients of different airway inflammation phenotypes (E). But serum levels of SM were significantly lower in early-onset patients than that in late-onset patients (D), and also lower in bNEA than bEA (F). sEA: sputum eosinophilic asthma; sNA: sputum neutrophilic asthma; sMA: sputum mixed asthma; bEA: blood eosinophil asthma; and bNEA: blood noneosinophil asthma. Mean ± sem. ^*P < 0.05

Comparison of serum sphingolipids between different subgroups of asthma patients

To further explore the potential role of SM in asthma subtypes, we analyzed the levels of SM in different subtypes of asthma. We found that serum SM levels were obviously decreased in the bNEA group compared with the bEA group. Similar tendencies of serum SM level changes were observed in the early-onset group compared with the late-onset group as shown in Figure 1B–F.

The difference in serum SM levels between bEA and bNEA imply that the serum SM was possibly involved in the pathogenesis of asthma. Due to the interdependent nature of the SM cycle, the influence of a certain lipid may be due to conversion or breakdown into another lipid mediator.²³ SM can be hydrolyzed into Cer by sphingomyelinase, and Cer has already been identified as a crucial factor in inflammation.²⁴ Cer was negatively correlated with eosinophilic asthma, and this negative correlation was related to the orosomucoid-like (ORMDL) gene, especially ORMDL3 expressed in airway epithelial cells, which can regulate the degranulation, transport, and recruitment of eosinophils.^25,26 Extracellular Cer can suppress mast cell-dependent allergic responses, including eosinophil migration to the airway epithelium.²⁷ Although our results showed no significant change in Cer levels in the eosinophil subgroup, the predicted trend was present (Fig. 1F). This imbalance between SM and Cer may lead to asthma. However, to date, the association of SM with eosinophils is not clear.

The difference in serum SM levels between the early-onset and late-onset groups might be associated with the different main inflammation types of these two groups. Previous studies have suggested that lung (tissue and sputum) eosinophils, such as a type 2 (IL-5) biomarker, are more prominent in adult-onset patients, whereas early-onset patients have less eosinophilia.²⁸ The allergic disease can affect people of all ages, and the prevalence in childhood is up to 39%.²⁹ However, the diagnosis of asthma in children is usually difficult because information is mostly given by the parents who are not always with their child,³⁰ and the exercise challenge, the most commonly used provocation test, is difficult to carry out correctly.³¹ Although there are some methods to test airflow reversibility in children, the criteria are under debate. In this study, we found that SM levels in early-onset asthma (≤12 yr old) were significantly lower, indicating that SM may be a diagnostic indicator in children with asthma.

Correlation analysis of SM with inflammatory factors and lung function indicators

Asthma is a chronic inflammatory airway disease associated with the type 2 cytokines IL-4, IL-5, and IL-13, and the lung function of asthmatic patients might get worse as the disease progresses. To further explore whether inflammatory factors and lung function indicators were related to serum SM, we applied Pearson correlation analysis. And the results showed SM 40:1 negatively correlated to sputum IL-17A (r = −0.621, P = 0.042) (Table 2). Several studies have shown that SM has a role in the anti-inflammation process. SM can bind to CD1d, a major histocompatibility complex class I-like glycoprotein serving as an antigen presenting biomacromolecule, and then suppress the activation of invariant natural killer T.³² SM is also a physiologic ligand for human CD300f, inhibiting high-affinity IgE receptor mediated mast cell activation.³³ IL-17A is a cytokine with strong proinflammatory properties that is thought to contribute to neutrophilic airway inflammation³⁴ and is an independent risk factor for severe asthma that impacts airway smooth muscle remodeling.³⁵ At the same time, the decreased SM may induce smooth muscle contraction independent on eosinophils,³⁶ and we speculated that the contribution of SM to the pathogenesis of asthma is more prominent in patients with bNEA. However, there are few studies on the relationship between SM and IL-17A, which is worthy of further study to explore the mechanism involving SM in asthma and in its various inflammatory subtypes.

TABLE 2

Correlations between sphingomyelin (SM) 34:2 and lung function indicators and inflammatory factors

	SM 34:2		SM 38:1		SM 40:1
	Pearson correlation coefficient	P	Pearson correlation coefficient	P	Pearson correlation coefficient	P
FEV1 (%predicted)	−0.046	0.765	−0.075	0.625	−0.032	0.834
FEV1/FVC (%)	0.033	0.830	0.036	0.815	0.103	0.502
Leptin (pg/ml)	0.313	0.236	0.211	0.159	0.185	0.218
Periostin (ng/ml)	0.190	0.205	0.211	0.159	0.185	0.218
Eosinophil count (/μl)	0.056	0.764	0.086	0.647	0.080	0.668
Blood TGF (ng/ml)	−0.113	0.453	−0.117	0.440	−0.026	0.864
Induced sputum MMP9 (pg/ml)	0.196	0.244	0.135	0.425	0.055	0.746
Induced sputum IL-5 (pg/ml)	−0.296	0.285	−0.254	0.360	−0.277	0.317
Induced sputum IL-13 (pg/ml)	−0.106	0.676	−0.086	0.735	−0.001	0.996
Induced sputum IL-10 (pg/ml)	−0.289	0.363	−0.269	0.399	−0.281	0.377
Induced sputum IFN-γ (pg/ml)	0.129	0.648	0.090	0.749	0.010	0.971
Induced sputum IL-17A (pg/ml)	−0.571	0.067	−0.577	0.063	−0.621	0.042a
Induced sputum IL-4 (pg/ml)	−0.088	0.836	−0.081	0.849	−0.084	0.843
Blood IL-5 (pg/ml)	−0.036	0.899	−0.002	0.995	−0.013	0.962
Blood IL-13 (pg/ml)	−0.119	0.727	−0.142	0.677	−0.105	0.760
Blood IL-10 (pg/ml)	−0.582	0.100	−0.646	0.060	−0.633	0.067
Blood IFN-γ (ng/ml)	0.080	0.785	0.141	0.631	0.156	0.594
Blood IL-17A (pg/ml)	−0.273	0.601	−0.306	0.556	−0.267	0.609
Blood IL-4 (pg/ml)	−0.183	0.513	−0.143	0.611	−0.132	0.639

a

P < 0.05.

FEV₁%pred: percentage predicted forced expiratory volume in 1 s; FVC: forced vital capacity; MMP: matrix metalloprotease.

Differences in ASMase activity between asthma patients and healthy controls

ASMase, a key enzyme that hydrolyzes SM to produce Cer, is the type most widely distributed in lung tissue.^11,37 In contrast to that in the healthy control group, the serum ASMase activity in the asthma group was significantly reduced, but there was no significant difference in induced sputum supernatant ASMase activity between the two groups (Table 3). ASMase is thought to convert SM into Cer in the pathogenesis of asthma.^11,37,38 In this study, the SM level was found to be decreased in asthma patients, and thus, we speculated that ASMase activity may be up-regulated in asthma patients. However, the results showed that the serum ASMase activity in asthma patients was lower than that in healthy controls. The same trends for SM and ASMase were also found by Gupta et al.¹⁸ These results indicate that the level of serum SM in asthma patients may not be mainly regulated by ASMase. SMS is a key enzyme responsible for the production of SM from Cer.³⁹ Based on the result of relatively low SM levels with concurrent relatively low ASMase activity in the asthma group, we speculated that SMS may participate in this pathogenesis. Whether the level of SMS varies should be further detected to explore the effect of SMS on sphingolipid metabolism in asthma patients.

TABLE 3

Differences in acid sphingomyelinase (ASMase) activity between asthma patients and healthy controls

	Asthma	Healthy control	P
Serum ASMase activity (mU/ml)	21.58 ± 7.83	33.25 ± 8.56	<0.001
Induced sputum supernatant ASMase activity (mU/ml)	22.51 ± 14.91	25.75 ± 12.63	0.439

Data were analyzed by a t-test. Mean ± sd.

Our study's strength is that it represents a rare exploration of sphingolipid metabolism in asthma patients; in particular, it is important that we carried out subgroup analyses and found some clues to the mechanism of asthma. The limitations of our study include the relatively small size of the study population. There is a lack of information on corticosteroid treatment, which may change eosinophil counts and sphingolipid metabolism. We need to perform further large-scale research, including animal model and cell model studies, to further explore and confirm the roles of these sphingolipids in the pathogenesis of asthma.

In summary, the decreased level of serums SM in patients with asthma showed the SM may be a protective factor of asthma. The change in SM level in the different subtypes of asthma and the association between the SM level and cytokines presented that the SMs contribute to the mechanism of asthma, especially bNEA. As a result, SM may be a potential biomarker and therapeutic target in asthma, particularly bNEA.

ACKNOWLEDGMENTS

The authors thank and acknowledge the study investigator, coordinator team, and participants for their support of this study. This study was supported by the National Natural Science Foundation of China (No. 81970028), National Natural Science Foundation of China (No. 82070305) and Natural Science Foundation of Tianjin City (No. 19JCQNJC10100).

AUTHORSHIP

C.G. analyzed data, created figures, and wrote the original manuscript. L.S. acquired clinical data and reviewed and edited the article. L.Z. wrote the original manuscript and performed the experiments. F.D. performed the experiments. X.Z. performed the experiments and interpreted the data. C.C. and L.Y. conceptualized the project, designed the research, and reviewed the article; C.C. was the lead author.

C.G., L.S., and L.Z. contributed equally to this work.

DISCLOSURES

The authors declare no conflicts of interest.