Mechanisms of permanent loss of olfactory receptor neurons induced by the herbicide 2,6-dichlorobenzonitrile: effects on stem cells and noninvolvement of acute induction of the inflammatory cytokine IL-6

Abstract

We explored the mechanisms underlying the differential effects of two olfactory toxicants, the herbicide 2,6-dichlorobenzonitrile (DCBN) and the anti-thyroid drug methimazole (MMZ), on olfactory receptor neuron (ORN) regeneration in mouse olfactory epithelium (OE). DCBN, but not MMZ, induced inflammation-like pathological changes in OE, and DCBN increased interleukin IL-6 levels in nasal-wash fluid to much greater magnitude and duration than did MMZ. At 24h after DCBN injection, the population of horizontal basal cells (HBCs; reserve, normally quiescent OE stem cells) lining the DMM became severely depleted as some of them detached from the basal lamina, and sloughed into the nasal cavity along with the globose basal cells (GBCs; heterogeneous population of stem and progenitor cells), neurons, and sustentacular cells of the neuroepithelium. In contrast, the layer of HBCs remained intact in MMZ-treated mice, as only the mature elements of the neuroepithelium were shed. Despite the respiratory metaplasia accompanying the greater severity of the DCBN lesion, residual HBCs that survived intoxication were activated by the injury and contributed to the metaplastic respiratory epithelium, as shown by tracing their descendants in a K5CreEr(T2)::fl(stop)TdTomato strain of mice in which recombination causes HBCs to express TdTomato in advance of the lesion. But, contrary to published observations with MMZ, the HBCs failed to form ORNs. A role for IL-6 in suppressing ORN regeneration in DCBN-treated mice was rejected by the failure of the anti-inflammatory drug dexamethasone to prevent the subsequent respiratory metaplasia in the DMM, suggesting that other factors lead to HBC neuro-incompetence.

Keywords: 2,6-dichlorobenzonitrile (or dichlobenil); DCBN; DEX; DMM; Dichlobenil; HBCs; IFN; IL; Inflammatory cytokines; K5; MMZ; Methimazole; Neurodegeneration; OE; OMP; ORNs; Olfactory stem cells; PBS; Regeneration; TNF; dexamethasone; dorsal medial meatus; horizontal basal cells; interferon; interleukin; keratin 5; methimazole; olfactory epithelium; olfactory marker protein; olfactory receptor neurons; phosphate-buffered saline; tumor necrosis factor.

Fig. 1. Regeneration of the OE after DCBN or MMZ treatment

Two- to three-month old male and female C57BL/6 mice were treated with a single i.p. injection of DCBN or MMZ, or with the vehicle alone. Representative cross sections of the nasal cavity, obtained at the level 3 of Young (1981), are shown. A, H&E staining. Eight weeks after DCBN injection (at 50 mg/kg), the DMM of the nasal cavity was covered by a single layer of non-neuronal epithelium, in contrast to the pseudostratified neuroepithelium seen in vehicle-treated mice. B, Immunohistochemical staining of the ORN marker, OMP. Eight weeks after DCBN injection, there were no OMP-positive cells in the DMM of the nasal cavity, in contrast to the robust staining of ORNs in the neuroepithelium, and the nerve bundles in the submucosa, in vehicle-treated mice. C, H&E staining. Four weeks after DCBN injection at a lower dose of 12.5 mg/kg, the DMM was still covered by a single layer of non-neuronal epithelium. In contrast, four weeks after MMZ treatment, at 75 mg/kg, the DMM was already covered by normal pseudostratified neuroepithelium. Scale bar: 20 μm. BG, Bowman's gland; OE, olfactory epithelium; NB, neural bundles. Magnification: 400×.

Fig. 2. Histological analysis of the acute nasal mucosal damages caused by DCBN or MMZ treatment

Two- to three-month old male C57BL/6 mice were treated with a single i.p. injection of DCBN (at 50 mg/kg), or MMZ (at 50 mg/kg), or with the vehicle alone. Representative H&E stained cross sections of the nasal cavity, obtained at the level 3 of Young (1981), are shown. Twenty four hours after DCBN injection, the nasal cavity was congested by detached epithelium and exudations. In the submucosa, there was edema and degenerated Bowman's glands. In contrast, 24 h after MMZ injection, the detached epithelium was no longer visible, and the submucosa was denuded. Arrow indicates the position of the basement membrane. Boxed areas in the top panels are shown at higher magnification in the bottom panels.

Fig. 3. Immunohistochemical staining for K5 at 24 h after DCBN or MMZ injection

Two-to three-month old male C57BL/6 mice were treated with a single i.p. injection of DCBN (at 50 mg/kg) (A,D) or MMZ (at 50 mg/kg) (B,E), or with the vehicle alone (C,F). Representative cross sections of the nasal cavity (DMM, A-C; septum, D-F), obtained at the level 3 of Young (1981), are shown. Nuclei were identified by DAPI staining (blue). K5 was stained as red. *, opening of a Bowman's gland duct; SP, nasal septum; ET, ethmoid turbinate. Magnification: 200×.

Fig. 4. Time course for changes in K5-positive cells in OE following DCBN treatment

Two- to three-month old male mice were treated with a single i.p. injection of DCBN (50 mg/kg) or MMZ (50 mg/kg), or with the vehicle alone. Tissues were obtained for immunohistochemical staining for K5, at 24 h, or 2, 4, and 7 d, or 8 weeks after DCBN injection. Representative cross sections of the nasal cavity, in the DMM region, obtained at the level 3 of Young (1981), are shown. A negative control (omitting primary antibody) was also included. Arrow indicates position of the basement membrane. Magnification: 400×.

Fig. 5. Time course for changes in Pax6-positive cells in OE following DCBN treatment

Two- to three-month old male mice were treated with a single i.p. injection of DCBN (50 mg/kg) or MMZ (50 mg/kg), or with the vehicle alone. Tissues were obtained for immunohistochemical staining for Pax6, at 24 h, or 2, 4, and 7 d, or 8 weeks after DCBN injection. Representative cross sections of the nasal cavity, in the DMM region, obtained at the level 3 of Young (1981), are shown. A negative control (omitting primary antibody) was also included. Arrow indicates position of the basement membrane.

Fig. 6. Spared HBCs give rise to metaplastic respiratory epithelium

Bigenic, K5-CreER^T2:fl(stop)CAG-TdTomato mice, were administered tamoxifen 3 weeks before DBCN intoxication and then euthanized 8 weeks later. A) The progeny of spared HBCs, marked by the expression of TdTomato are found in the metaplastic respiratory epithelium lining the dorsomedial meatus (dorsal is toward the top of the figure). The boundary between respiratory and olfactory epithelium is marked by the fall-off in CD133 staining as indicated by the arrows. RM – metaplastic respiratory epithelium, OE – olfactory epithelium. B) The HBC-derived cells in the metaplastic respiratory epithelium lining the dorsomedial meatus are both columnar and basal (the three arrows demarcate a row of basal cells. The columnar cells are capped by cilia (inset). C) The basal cell-derived cells in normal, unlesioned respiratory epithelium located anterior and ventral to the olfactory epithelium closely resemble those in the metaplastic respiratory epithelium (inset). D) The HBC-derived columnar cells in the metaplastic respiratory epithelium lining the dorsomedial meatus do not retain CD54 staining, which is characteristic of HBCs, while the basal cells do (arrow). E) The metaplastic respiratory epithelium lining the dorsomedial meatus is sharply bounded, and that boundary is marked by the rapid transition from neuron-containing epithelium in which ORNs are marked by the expression of PGP9.5 to the metaplastic respiratory epithelium characterized by columnar TdTomato expressing epithelial cells. Scale bars: A – 100 μm; D – 50 μm (applies to B, C, as well); E - 25 μm.

Fig. 7. IFN-y and IL-6 levels in mouse nasal wash fluid after DCBN or MMZ treatment

Two- to three-month old male mice were treated with a single i.p. injection of DCBN (50 mg/kg) or MMZ (50 mg/kg), or with a vehicle. Nasal wash fluid was collected for analysis at 2, 6, or 24 h after the injection, as indicated. For the DEX+DCBN group, mice were treated twice with DEX (s.c, 5 mg/kg, at 0 and 2 h after DCBN injection), and nasal wash fluid was collected for cytokine determination at 24 h after the DCBN injection. IFN-γ (A) and IL-6 (B) levels are shown by box plots, which show the minimum/maximum, the 25^th/75^th percentiles, and the median. *, P<0.05; **, P<0.01; compared to the vehicle-treated group (n=4; one way ANOVA on ranks, followed by Dunn's post-hoc test).

Fig. 8. Effects of DEX treatment on nasal mucosal damages caused by DCBN treatment

Two- to three-month old male mice were treated with a single i.p. injection of DCBN (at 50 mg/kg) and two subcutaneous injections of saline or DEX (5 mg/kg) at 0 and 2 h after DCBN injection. A, Tissues were obtained for histopathological examination at 4 weeks after DCBN injection. Representative H&E-stained cross sections of the nasal cavity, obtained at the level 3 of Young (1981), are shown. In DCBN-treated groups, metaplastic bone formation was observed in the submucosa. B, Immunohistochemical staining for K5 at 24 h after DCBN. Arrows indicate position of the basement membrane.