The Chemotactic Behavior of Eosinophils in Patients With Chronic Rhinosinusitis

Julie L. Wei, MD; Hirohito Kita, MD; David A. Sherris, MD; Eugene B. Kern, MD; Amy Weaver, MS; Jens U. Ponikau, MD

Objective: To characterize peripheral eosinophil migration in patients with chronic rhinosinusitis in the presence of nasal mucin and nasal tissue extracts. Study Design: Prospective, controlled, ex-vivo. Meth­ods: Peripheral blood eosinophils, nasal mucin, and nasal tissue were harvested at the time of sinus sur­gery in 10 patients, as well as obtained in 10 healthy control subjects. Extracts were prepared from nasal mucin and nasal tissue. A modified Boyden chamber was used to study eosinophil migration from both patients and healthy control subjects in the presence of both extracts. Results: Patients with chronic rhino-sinusitis and all healthy control subjects demon­strated a concentration-dependent increased migra­tion of eosinophils in the presence of both nasal mucin and nasal tissue extracts. The percentage of migration was consistently higher for eosinophils from patients with chronic rhinosinusitis compared with control subjects. The difference attained statis­tical significance in the presence of 50% tissue extract (median percentage of migration, 23.3% vs. 7.8% [P = .0331). Conclusions: Nasal mucin and nasal tissue in chronic rhinosinusitis contains chemoattractants, which can induce active eosinophil migration. The eosinophil migration from patients with chronic rhi­nosinusitis was consistently higher compared with eosinophils from healthy control subjects. Because the eosinophils were obtained from the peripheral blood, this finding suggests activation of eosinophils in the systemic circulation in chronic rhinosinusitis. Key Words: Chronic rhinosinusitis, eosinophils, migration.

INTRODUCTION

Chronic rhinosinusitis (CRS) is one of the most com­mon chronic diseases in the United States .' The patient's quality of life is altered because of recurrent infections, nasal obstruction, and anosmia. There seems to be no "cure" because CRS frequently recurs, requiring subse­quent treatment including surgery. Although the patho­physiology of CRS is unknown, the histopathology clearly features tissue eosinophilia in the nose and paranasal sinuses. Toxic major basic protein released by the eosino­phils was found to be associated with the epithelial dam-age in patients with CRS.2

It was recently found that the eosinophils in CRS, whenever present in the tissue, actually migrate, ulti­mately, into the airway mucin, where they typically form typical clusters and eventually become necrotic.' This phenomenon was observed in 96% of consecutive surgical patients (n = 101) diagnosed with CRS. This significantly higher incidence of eosinophilic mucin than previously reported was probably a result of improved mucus collec­tion techniques during surgery. In addition, the observa­tion was made that in that eosinophilic mucin, eosinophils degranulate morphologically differently than they do in allergic rhinitis (Ponikau JU, et al., unpublished data, April 1999). The purpose of our study was to further elucidate the migration patterns of the eosinophil in this common chronic disease.

MATERIALS AND METHODS

After approval through the Institutional Review Board (IRB) of the Mayo Clinic, 10 patients with CRS and 10 healthy control subjects were enrolled into the study. The eosinophil migration experiments were carried out using a modified Boyden chamber technique.' Eosinophils migrate from one chamber to another because of active chemotaxis through a porous membra­nous filter. The pores of the filter are significantly smaller than the diameter of the eosinophil, thus requiring active migration of the cells to traverse and "squeeze" through the narrow pores (Fig. 1).

Purification of Eosinophils From Peripheral Blood Eosinophils could not be isolated in nasal mucin because the vast majority of these eosinophils had degranulated and were rendered necrotic and nonviable. Therefore, all experiments were performed using eosinophils isolated from the peripheral blood of 10 patients with CRS undergoing functional endoscopic sinus surgery (FESS) and 10 healthy control subjects without any evi­dence of CRS, asthma, or allergy.

Fig. 1. Schematic drawing of modified Boyden chamber. Eosinophils are intro­duced into the chamber on the left and migrate through the porous membrane toward the patient mucin or patient tis­sue extract located in the chamber on the right. The diameter of the membrane pores is less than the diameter of the eosinophil, thus requiring active eosino­philic migration.

Eosinophils were isolated from the peripheral blood using a magnetic cell separation system (MACS) (Becton Dickinson, San Jose , CA ), as described by Hansel et al.,5 with minor modifica­tions. In brief, 40 mL venous blood anticoagulated with 50 U/mL heparin was diluted with phosphate-buffered solution (PBS) at a 1:1 ratio. Diluted blood was overlayered on an isotonic Percoll solution (density of 1.085 g/mL, Sigma Chemical Co., St. Louis , MO ) and centrifuged at 200g at 4°C using a Beckman TJ-6 cen­trifuge. The supernatant and mononuclear cells at the interface were carefully removed. The inside wall of the centrifuge tube was wiped twice with sterile gauze to eliminate mononuclear cell adherence to the inside wall of the tube. Erythrocytes in the sediment were lysed by exposure to two cycles of sterile water. Isolated granulocytes were washed twice with piperazine-N, N'-bis[2-ethanesulforic acid] (PIPES) buffer (25 mmol/L PIPES, 50 mmol/L NaCl, 5 mmol/L KC1, 25 mmol/L NaOH, and 5.4 mmol/L glucose, pH 7.4) containing 1% defined calf serum (DCS, HyClone Laboratories, Logan UI), and an approximately equal volume of anti-CD 16 antibody conjugated with magnetic particles (Miltenyi Biotic, Bergisch-Gladbach, Bergisch-Gladbach, Germany) was added to the cell pellet. After 60 minutes of incubation on ice, cells were loaded into the separation column positioned on the MACS magnetic field. Cells were washed three times with 5 mL PIPES buffer with 1% DCS. The purity of eosinophils counted by Ran­dolph stain was greater than 98%. The contaminating cells were neutrophils. No mononuclear cells or basophils were present.

Preparation of the Nasal Mucin and Tissue Extracts

Surgical specimens of nasal mucin were collected into a conical tube lined with mesh and weighed. Five times the mucin weight was added in the form of RPMI 1640 medium. The tube was vortexed for 30 seconds, then placed on ice. After 3 hours, the tube was again vortexed for 30 seconds. The mesh was gently removed from the conical tube. Sterile forceps were used to gently squeeze the mesh-entrapped nasal mucin to collect the extract solution into the tube. Mesh and nasal mucin were discarded. The extract was centrifuged at 3000 rpm for 10 minutes at 4°C with low brake. The sediment was discarded. The supernatant was placed on ice until ready for use. Nasal tissue extracts were prepared in an identical fashion. Commercially available eotaxin, a powerful eosinophil chemotactic factor, was used as a positive control.

Measurement of Eosinophil Migration in Patients and Healthy Control Subjects

Costar Transwell ( Corning , NY ) kits, each containing 12 wells, were used. Each well has an upper and lower chamber separated by a membranous filter of 3-p.m pores. The lower cham­bers contained no extract (negative control), 10%, 25%, or 50% concentration of either nasal mucin or nasal tissue extract, or eotaxin (positive control). The eosinophilic suspension (100 gL of 1 x 106 cells/mL) from patients with CRS or healthy control subjects was gently added to the upper chamber. The wells were incubated at 37°C for 1 hour. The upper chambers and the filters were removed from each well and 600 µL of 10 mmol/L ethyl­enediamine tetra-acetic acid (EDTA) in PBS was added to each lower chamber, which then was placed on ice for 20 minutes. The cell suspensions were centrifuged at 3000 rpm for 10 minutes in a microcentrifuge at 4°C. Next, 100 uL of 10 mmol/L EDTA in PBS was added to the sediment to suspend the eosinophils. Ten microliters of cell suspension was loaded onto the hemocytometer to count the number of eosinophils. The percentage of cells that migrated was calculated after subtracting the background migra­tion observed for each specimen in the negative control well.

Statistical Analysis

Nasal mucin and nasal tissue extracts were evaluated sep­arately. The values for percentage of migration were analyzed after applying the square root transformation because the origi­nal distributions were positively skewed. For each group of sub­jects (patients with CRS and healthy control subjects), a separate one-way, repeated-measures analysis of variance (ANOVA) model was fit (with extract concentration level [10%, 25%, or 50%] handled as a quantitative factor) to assess whether the eosinophil migration followed a concentration-dependent pattern. In addi­tion, a two-way repeated measures ANOVA model (with effect terms for subject group and extract concentration level) was fit to evaluate differences in the percentage of migration. Within this model, contrast statements were specified to compare the per­centage of migration between the subject groups at each concen­tration level. All calculated P values were two-sided, and P values less than .05 were considered statistically significant. The ANOVA models were fit using the SAS MIXED procedure.

RESULTS

Information regarding patient demographics is pre­sented in Table I. Nine of 10 patients had a clinical history of asthma and were on maintenance regimens of inhaled steroid therapy for asthma control. The average number of previous sinus surgeries was 3.8 per patient. All patients had significant sinus disease as documented by computed tomography (CT) scan using the Lund-MacKay CT staging system.'

Four patients with CRS had positive results on skin testing to common aeroallergens; the remaining six pa­tients, as well as the healthy control subjects, were non-allergic. Four patients were on a regimen of systemic steroids for asthma control, three patients were on a reg­imen of oral prednisone therapy, and one patient received intramuscular Kenalog injections every 3 to 6 months. None of the patients was using intranasal steroids. The median percentage of background migration was 3% (mean value, 4.4%) for patients with CRS and 4% (mean value, 4.5%) for healthy control subjects.

Eosinophil Migration Toward Tissue Extract

After subtracting the background migration observed from each specimen, the eosinophils from all 10 healthy control subjects demonstrated increased eosinophil migra­tion with a concentration-dependent pattern in the pres­ence of nasal tissue extract (P = .003). For nasal tissue extract concentrations of 10%, 25%, and 50%, the median percentages of eosinophils migrating were 0.5%, 2.5%, and 7.8%, respectively (Fig. 2).

TABLE II.
Summary of the Median Percent Migration for CRS Patients and
Control Subjects Across the Extract Concentration Levels.

Healthy Control Patients With CRS Subjects

Extract Median (IQR) Median (IQR)

Nasal tissue extract

10% 1.5 (-0.5, 6.8) 0.5 (-0.9, 5.5)

25% 5.9 (1.2, 17.5) 2.5 (0.7, 8.8)

50% 23.3 (16.5, 29.3) 7.8 (2.9, 14.5)

Nasal mucin extract

10% 0.2 (-1.3, 0.8) -0.7 (-2.8, 0)

25% 4.9 (0.5, 5.8) 0.8 (-2.1, 5.3)

50% 10.0 (2.8, 12.1) 3.1 (0.5, 8.5)

Eotaxin 9.9 (7.2, 15.8) 17.0 (6.4, 25.0)

CRS = chronic rhinosinusitis; IQR = interquantile range, 25th and 75th percentiles.

The eosinophils from patients with CRS also demon­strated increased, concentration-dependant eosinophilic migration toward the tissue extract (P <.001). At the 10%, 25%, and 50% concentration levels, the median percent-ages of migration were 1.5%, 5.9%, and 23.3%, respec­tively (Fig. 2). Although the percentage of migration was consistently higher for eosinophils from patients with CRS than from control subjects, this difference only attained significance in the presence of 50% tissue extract (P = .033).

Eosinophil Migration Toward Mucin Extract

A similar finding as with the tissue extract was ob­served for the migration toward the mucin extract. Pe­ripheral blood eosinophils from both healthy control sub­jects and patients with CRS demonstrated migration toward mucin extract, which was, again, concentration dependent (P = .004 and P <.001, respectively). For healthy control subjects, the median percentages of migration were -0.7% (background migration was higher), 0.8%, and 3.1% for 10%, 25%, and 50% mucin concentra­tion, respectively. For eosinophils from patients with CRS, the median percentages of migration were 0.2%, 4.9%, and 10.0% for 10%, 25%, and 50% mucin concentration, respec­tively (Fig. 3).

TABLE I.
Demographic and Clinical Data on 10 Patients With CRS.

	Systemic	No. of	CT Staging		Number of Eosinophils in
Age	Steroid	Previous			Peripheral Blood
(y)	Sex Asthma Use	Surgeries	Left	Right	(x106)/cc
78	M +	0	12	12	5.3
52	M + -	3	12	12	3.6
64	F + +	4	9	10	2.9
19	M + +	6	11	10	1.1
35	M -	0	NA	NA	8.0
35	F +	3	10	9	4.1
42	F +	5	12	12	27.5
47	M + +	4	12	10	12.3
39	F +	12	12	12	8.5
54	M + +	1	11	12	24.5

*The sinus CT scan was staged using the Lund-McKay CT staging. CRS = chronic rhinosinusitis.

Control Eos n Patient Eos

10% 25% 50% Eotaxin Extract Concentration

Fig. 2. Eosinophilic migration toward tissue extract. The graft demonstates the migration of eosinophils toward the tissue extract, dependent on its concentration. Note the increased migration of patients' eosinophils compared with healthy control subjects.

Impact of Steroids and Allergy

No trend was observed when comparing degree of eosinophil migration in patients receiving steroid therapy with patients without steroid therapy. Greater eosinophil migration was observed from the four allergic patients in response to both the nasal mucin and nasal tissue extracts when compared with eosinophils from nonallergic pa­tients. However, the migration difference between allergic and nonallergic patients was not statistically significant.

Eotaxin as Positive Control

When eotaxin was used as a positive control, the pattern was reversed. The eosinophils from healthy con­trol subjects were migrating better toward the eotaxin compared with the eosinophils from the patients with CRS (median percentage of migration, 17.0% vs. 9.9%, respec­tively [P = .41] ).

Control Eos n Patient Eos

10% 25% 50% Eotaxin Extract Concentration

Fig. 3. Eosinophilic migration toward mucin extract. The graph demonstrates the migration of eosinophils toward the mucin ex-tract, dependent on its concentration. Note again the increased migration of patients' eosinophils compared with healthy control subjects.

DISCUSSION

The eosinophil, not the neutrophil, is the character­istic and predominant cell in CRS. To further investi­gate the role of these eosinophils in CRS, it is important to understand the migration pattern they follow in the dis­ease state.

In the majority of the CRS cases, the eosinophils observed in the tissue were found to be merely in transit and, ultimately, to migrate into the mucin of the airway lumen. There, these eosinophils formed clusters, typifying the characteristic eosinophilic mucin.3

Our findings clearly demonstrate the ability of these eosinophils to migrate toward the nasal tissue and, ulti­mately, into the nasal mucin of patients with CRS. A concentration-dependent relationship was observed be­tween the concentration of both the nasal mucin extracts and the nasal tissue extracts and the degree of eosino­philic migration. It appears that the higher concentrations of extract may contain higher concentrations of chemo­kines or other still unknown responsible factors, which explain this linear relationship.

Eosinophils from healthy subjects without history of allergy, asthma, or CRS demonstrated similar concentration-dependent migration patterns in the presence of nasal mucin extracts and nasal tissue extracts in patients with CRS, but to a significantly lesser degree.

The unique and major finding of the current study is that eosinophils harvested from the peripheral blood of patients with CRS consistently demonstrated signifi­cantly higher median percentage of migration toward tis­sue extract than eosinophils from healthy control subjects. This suggests that the eosinophilic inflammatory cells mediating the inflammation in CRS are already distinctly activated in the systemic circulation. Thus, these periph­eral eosinophils are induced to migrate from the vascula­ture into the nasal tissue and subsequently migrate through the epithelium into the mucin of the nasal lumen and paranasal sinuses.

The findings in the current study should stimulate further research focusing on the elucidation of the eosin­ophil activation and the chemotaxis pathway so that in­terruption of these behavior patterns could possibly pro-vide novel targets to treat patients with CRS.

BIBLIOGRAPHY

1. Lanza DC, Kennedy DW. Adult rhinosinusitis defined. Otol Head Neck Surg 1997;117:51-S7.
2. Harlin SL, Ansel DG, Lane SR, et al. A clinical and pathologic study of chronic sinusitis: the role of the eosinophil. J Allergy Clin Immunol 1988;81:867-875.
3. Ponikau JU, Sherris DA, Kern EB, et al. The diagnosis and incidence of allergic fungal sinusitis (AFS). Mayo Clin Proc 1999;74:877-884.
4. Hakansson L, Westerlund D, Venge P. New method for the measurement of eosinophil migration. J Leukoc Biol 1987; 42:689-696.