Immunohistochemical characterization of human olfactory tissue

Abstract

Objectives/hypothesis: The pathophysiology underlying human olfactory disorders is poorly understood because biopsying the olfactory epithelium (OE) can be unrepresentative and extensive immunohistochemical analysis is lacking. Autopsy tissue enriches our grasp of normal and abnormal olfactory immunohistology and guides the sampling of the OE by biopsy. Furthermore, a comparison of the molecular phenotype of olfactory epithelial cells between rodents and humans will improve our ability to correlate human histopathology with olfactory dysfunction.

Study design: An immunohistochemical analysis of human olfactory tissue using a comprehensive battery of proven antibodies.

Methods: Human olfactory mucosa obtained from 21 autopsy specimens was analyzed with immunohistochemistry. The position and extent of olfactory mucosa was assayed by staining whole mounts (WMs) with neuronal markers. Sections of the OE were analyzed with an extensive group of antibodies directed against cytoskeletal proteins and transcription factors, as were surgical specimens from an esthesioneuroblastoma.

Results: Neuron-rich epithelium is always found inferior to the cribriform plate, even at advanced age, despite the interruptions in the neuroepithelial sheet caused by patchy respiratory metaplasia. The pattern of immunostaining with our antibody panel identifies two distinct types of basal cell progenitors in human OE similar to rodents. The panel also clarifies the complex composition of esthesioneuroblastoma.

Conclusions: The extent of human olfactory mucosa at autopsy can easily be delineated as a function of age and neurologic disease. The similarities in human versus rodent OE will enable us to translate knowledge from experimental animals to humans and will extend our understanding of human olfactory pathophysiology.

Figure 1

Autopsy specimens provide tissue for both a comprehensive perspective of the extent of neuronal staining and a detailed characterization of olfactory mucosa. A. Whole mount staining of the left septum taken from a 75 year-old male with Alzheimer’s. The edges of the whole mount stripped from the underlying bone are delineated by dashed lines. The thin rectangular defect extending from the superior boarder corresponds to the region removed from the whole mount for histology. The pattern of PGP9.5 staining of the mucosal whole mount is represented in red and is superimposed on a photograph taken of the septum after removal of the mucosa. The inset is a higher power magnification of the actual PGP9.5 staining at the surface of the whole mount. Notice the circular patches of non-neuronal epithelium at the edges of the olfactory boarder and within the olfactory region seen in the whole mount representation and inset. Arrows indicate anterior and posterior extent of the cribriform plate, f = frontal sinus, a = anterior, p = posterior, scale bar = 50µ. B. Section of epithelium from specimen shown in (A) fluorescently double labeled with OMP and PGP9.5 antibodies. PGP9.5 labels mature and immature olfactory neurons and OMP labels only the mature neurons. In this specimen the double labeled mature neurons (arrows) are relatively few compared to the abundant PGP9.5(+)/OMP(−) immature neurons. arrowhead = basement membrane, scale bar = 25µ.

Figure 2

Non-neuronal components of the OE can be identified with various antibodies. A. DAB staining with the antibody to the transcription factor Hes1 labels nuclei of the supporting cells. B. The antibody to transcription factor Sox9 labels duct/gland cell nuclei only. C. Antibodies to K18 label both sustentacular cells as well duct/gland cells. D. Antibodies to E-Cadherin stain with a similar pattern to K18. E,F,G. Double immunofluorescent labeling of a sharp transitional zone between olfactory and respiratory epithelium with antibodies to Beta-tubulin IV and PGP9.5. Beta-tubulin IV densely labels the apical portion and cilia of the respiratory cells. Within the OE, PGP9.5 labels the cell body, axons and dendrites of the olfactory neurons. Several ciliated cells within the respiratory epithelium are labeled with both antibodies (arrows) and are clearly non-neuronal. Bg = bowman’s glands, arrowheads = basement membrane, scalebar = 25µ.

Figure 3

Horizontal basal cells (HBCs) can be identified using multiple antibodies. A. DAB staining with the Pankeratin antibody labels HBCs, supporting cells, and Bowman’s gland (bg) and duct cells. The K5 and K17 antibodies are specific for HBCs in the OE (B and C). D,E,F. An antibody to the transcription factor, p63, labels nuclei of the HBCs as confirmed with double labeling using the K5 antibody. G. DAB staining with antibodies to EGF receptor (EGFr) label HBCs as well as supporting cells to a lesser degree. H,I. Staining of adjacent sections with antibodies to transcription factors Pax6 and Sox2 label nuclei of supporting cells and HBCs, but Pax6 also labels duct/gland cells. J,K,L. Double immunofluorescent staining of the epithelium with p63 and Sox2 confirms co-labeling of HBCs, but occasional Sox2(+)/p63(−) cells are found just above the HBC layer (arrows) and likely correspond to globose basal cells. bg = Bowman’s glands, arrowhead = basement membrane, scale bar = 25µ.

Figure 4

Multiple antibodies can be used to identify subpopulations of olfactory neurons (ONs). A. DAB staining with OMP antibodies label mature ONs that tend to reside in the more apical portion of the epithelium. B. Antibodies to Tuj-1 label both mature and immature ONs with dense labeling of the dendrites and axons. C. The antibody to G_olf labels mature and immature ONs with at a higher density in the dendrites and cell bodies. Arrows = dendritic knobs. D,E,F. Double immunofluorscent labeling of the epithelium with antibodies to PGP9.5 and OMP demonstrates the relative numbers of mature vs. immature ONs. Normal olfactory epithelium contains abundant OMP(+)/PGP9.5(+) mature neurons and OMP(−)/PGP9.5(+) immature neurons (arrows) suggesting a dynamic neuroepithelium. G,H,I. With double immunofluorescent labeling of the epithelium using HBC marker K5 and ON marker PGP9.5, cells are identified just above the HBCs that are absent of staining for either antibody (arrows). These cells may be GBCs or early neuroprogenitor cells. Asterisk identifies olfactory nerve fasicles extending out from the basal layer of the epithelium. Arrowheads = basement membrane, scale bar = 25µ.

Figure 5

Transcription factor antibodies allow us to identify a subset of globose basal cells (GBCs). Immunofluorescent staining with Mash1 antibodies label clusters of cells in the OE (B,E,H,K). A,B,C. Mash1(+) cells (arrows) are not neurons as shown by lack of co-labeling with the anti-neuronal antibody Tuj-1. D,E,F. Mash1(+) cells are not HBCs as shown by absence of double staining with p63 antibodies. An arrow identifies a Mash1 stained cell sitting just above a p63 stained HBCs in a position typically occupied by GBCs. G,H,I. The absence of p27 double labeling with Mash1 (arrows) suggests that these cells are not quiescent. J,K,L. Many Mash1(+) cells (arrows) are mitotically active as shown by double labeling with Ki67 antibodies. M,N,O. Notch1 antibodies also label cells within the GBC compartment (arrows). Arrowheads = basement membrane, scale bar = 25µ.

Figure 6

Esthesioneuroblastoma can be characterized using antibody reagents similar to that used for OE. A. DAB stained section of a portion of an esthesioneuroblastoma taken from a 44 year-old female using the neuron-specific antibody Tuj-1 labels the majority of the tumor as expected. B. However, these cells are not mature ONs based on the absence of OMP staining. C. The sustentacular and gland/duct cell specific K18 antibody labels cords of cell separating the tumor into nests. (Scale bar = 250µ for A, B, and C.) D. The tumor does not stain with antibodies to K5, p63 (E), and Beta-Tubulin IV (F). Only cells of the epithelial surface overlying the tumor stain with antibodies to K5, p63, and Beta-Tubulin IV indicating a respiratory-like epithelium. (Scale bar = 50µ for D, E, and F.) G. DAB staining using the G_olf antibody also labels a large portion of the tumor in a pattern similar to Tuj-1 supporting an olfactory epithelial origin of the tumor. An arrow indicates extension of the tumor cells into the overlying epithelium (scale bar = 50µ). H. Immunofluorescent staining using antibodies to Mash1 also stains a large portion of the tumor in a pattern similar to Tuj-1 and G_olf and corroborates previous mRNA findings. Arrows indicate areas of tumor extension into the overlying epithelium (scale bar = 100µ). I. DAB staining with Pax6 antibodies labels a large portion of the tumor and the overlying epithelium (scale bar = 50µ). Arrowheads = basement membrane.

Figure 7

The esthesioneuroblastoma can be divided into two distinct populations using a combination of transcription factor and structural antibodies. A,B,C. Double immunofluorescent staining with Sox2 (arrows) and Sox9 antibodies show a mutually exclusive staining pattern within the tumor. Sox2(+) nuclei seem to occur at the periphery of the Sox9(+) cells and within the overlying epithelium (scale bar = 25µ). D,E,F. The staining of the tumor with Sox9 and K18 is also mutually exclusive and reveals a similar pattern to Sox9/Sox2 double staining (scale bar = 50µ). G,H,I. The Sox9 antibody labels Bowman’s duct/gland cell nuclei of the OE, but within the tumor it co-labels with Tuj-1(+) cells. The arrow indicates a neuronal process. Scale bar = 25µ.

Figure 8

Sox9(+)/Tuj-1(+) tumor cells do not follow the usual pattern of staining of olfactory epithelial gland cells. A,B,C. The separation of the tumor into two populations is confirmed using K18 and Tuj-1 antibodies. Similar to the two cell populations demonstrated with Sox2 and Sox9 antibodies, immunofluorescent labeling of cell cytoplasm and processes using K18 (arrows), does not co-label with the neuron-specific Tuj-1 antibody. D,E,F. E-Cad has a similar staining pattern as K18 in the epithelium labeling both gland and supporting cells, however within the tumor K18 is restricted to the Sox2(+)/Sox9(−)/Tuj-1(−) population while E-Cad labels the same NCAM(−) non-neuronal population (arrows) as well as the NCAM(+) neuronal population. Scale bar = 25µ.

Figure 9

A. The antibody to K17 labels only a portion of the tumor as shown with immunofluorescent staining and DAB at a lower magnification in the inset (inset scale bar = 100µ). B,C. Using double immunofluorescent staining with Tuj-1 a small number of cells are co-labeled with both antibodies (arrowheads), but most K17(+) cells seem to follow a pattern of staining similar to K18/Sox2 antibodies. White scale bar = 50µ.