CHRONIC HYPERPLASTIC SINUSITIS - ASSOCIATION OF TISSUE EOSINOPHILIA WITH MESSENGER-RNA EXPRESSION OF GRANULOCYTE-MACROPHAGE COLONY-STIMULATING FACTOR AND INTERLEUKIN-3
Dl. Hamilos et al., CHRONIC HYPERPLASTIC SINUSITIS - ASSOCIATION OF TISSUE EOSINOPHILIA WITH MESSENGER-RNA EXPRESSION OF GRANULOCYTE-MACROPHAGE COLONY-STIMULATING FACTOR AND INTERLEUKIN-3, Journal of allergy and clinical immunology, 92(1), 1993, pp. 39-48
Background: We investigated the association among tissue eosinophilia,
cellular infiltration, and cytokine mRNA expression in chronic hyperp
lastic sinusitis (CHS). Methods: Percutaneous biopsies of the maxillar
y sinuses and nasal polyps were performed in 12 adult patients (six me
n and six women) of whom seven were nonallergic and 11 were asthmatic.
Tissues were compared with biopsy specimens from the inferior and mid
dle turbinates of normal control subjects. Results: Histologically, an
eosinophil-predominant inflammatory infiltrate was seen in 10 of 12 p
atients, whereas a mild to moderate neutrophilic infiltrate was seen i
n 4 of 12 patients. As determined by immunocytochemistry, diseased tis
sues and normal control tissues differed significantly in terms of the
number of activated (EG2+) eosinophils (p = 0.005) but not in terms o
f CD3+ or CD4+ T lymphocytes, elastase-positive neutrophils or CD68+ m
acrophages. The number of eosinophils did not correlate with that of a
ny other cell type. By in situ hybridization, CHS tissues showed signi
ficantly higher numbers of granulocyte-macrophage colony-stimulating f
actor (GM-CSF) and interleukin (IL)-3 mRNA-positive cells than normal
control tissues (p = 0.002 and 0.0005, respectively) per high-powered
field. There was a significant correlation between the number of infil
trating EG2+ eosinophils and cells that expressed mRNA for GM-CSF (r =
0.60, p = 0.041) or IL-3 (r = 0.69, p = 0.013). Furthermore, epitheli
al cells did not show detectable mRNA expression for GM-CSF or IL-3. N
o significant correlation was found between IL-5 mRNA expression and i
nfiltrating EG2+ eosinophils in diseased tissues. However, the IL-5 de
nsity was significantly higher in the five patients with CHS who had p
ositive allergy skin test results than in the seven patients with nega
tive skin test results (p = 0.017) or in normal control subjects. Conc
lusions: Our data support a role for GM-CSF and IL-3 in the eosinophil
ia characteristic of CHS and show that IL-5 mRNA expression is not a p
rominent feature of nonallergic inflammation. The cellular sources of
GM-CSF and IL-3 in CHS remain to be definitely determined.