Original Contributions
Developmental Anatomy of the Eustachian Tube and Middle Ear in Mice KEEt -I‘YUN PARK, MD, KAZUYOSHI
UENO, MD, AND DAVID J. LIM, MD
Purpose: It is generally accepted that the development of the tubotympanum has significant bearing on the susceptibility to ear infection. A detailed study of the differentiation of ciliated cells in secretory elements will be useful in understanding both the normal physiology and the pathology of the tubotympanum. Method: Serially sectioned temporal bones of 76 mice ranging from gestational age day 11 to postnatal day 21 were examined microscopically. Results: During the period of gestation, the tubotympanic recess was formed at the 12th day and began to extend to form the middle ear between the 13th and 14th days. A rapid increase in the volume of the tubotympanic recess was observed between the 15th and 16th days when a definitive division of the tubotympanic recess into the eustachian tube and middle ear cavity was observed. Postnatally the tubotympanum attained an adult form around day 9, and the maximum change of middle ear volume was noted on day 11, when the mesenchymal tissue in the middle ear cavity disappeared completely. Development of the ciliated cells was observed concurrently in both the eustachian tube and middle ear on the 16th gestational day, one day earlier than the appearance of the epithelial secretory cells in both the eustachian tube and middle ear. The number of ciliated cells and secretory cells increased rapidly after birth. Tubal glands were well developed with evidence of secretory activity around the time of birth. Conclusions: Based on these findings, one can conclude that the mucociliary defense system starts to develop during the fetal stage and is well established immediately after birth. Copyright 0 1992 by W.B. Saunders Company
It is generally accepted that the development of the tubotympanum has significant bearing on the susceptibility to ear infection. The overwhelming concern of research focusing on the development of the tubotympanum has been the development of the tubal cartilage and muscles. In recent years, there has been good evidence to suggest that a dysfunctional muco-
From the Otological Research Laboratories, Department of Otolaryngology, The Ohio State University College of Medicine, Columbus, OH. Address reprint requests to David J. Lim, MD, Division of Intramural Research, NIDCDINIH, Bldg 31, Room 3CO6, 9000 Rockville Pike, Bethesda, MD 20692. Supported in part by Grant No. DC00090 from the National Institute of Deafness and Other Communication Disorders, National Institutes of Health. Copyright 0 1992 by W.B. Saunders Company 0196-0709/92/l 302-0004$5.00/O American
Journal
of Otolaryngology,
ciliary transport mechanism (such as in immotile cilia syndrome) in the tubotympanum and sinus leads to chronic or recurrent otitis media and/or sinusitis. Although the importance of the mucociliary defense system in the tubotympanum has been well established,‘*’ the development of the mucociliary system has been poorly documented. Although the developmental anatomy of the murine middle ear has been described earlier,3P4 a description of the development of the eustachian tube mucociliary system was lacking. A detailed study of the differentiation of ciliated cells and secretory elements will be useful in understanding both the normal physiology and pathology of the tubotympanum. In this report, the developmental anatomy of the murine eustachian tube and middle ear, with particular emphasis on the differentiation of epithelial cells (ciliated and secretory)
Vol 13, No 2 (March-April),
1992: pp 93-100
93
94
and luminal formation of the eustachian tube, including glands, was investigated using light microscopy (LM), computer-aided threedimensional (3D) reconstruction, and morphometry. MATERIALS AND METHODS The mouse was chosen for this study of the developmental morphology of the mammalian eustachian tube and middle ear because of its relatively short periods of gestational and postnatal development, and its rapid breeding ability. A total of 76 BALB/c mice, ranging in age from gestational day 11 through postnatal day 21,were used in this investigation. Gestational day was determined by the vaginal plug technique, establishing the first day of appearance of the vaginal plug as day one. Postnatal mice with evidence of middle ear infection were excluded. For LM observation, after decapitation under inhalation anesthesia with Halothane (Halocaroon Laboratories, Hackensack, NJ), specimens were fixed by immersion in 10% neutral buffered formalin for 7 days at 4°C. Temporal bones from postnatal mice were decalcified with 10% ethylene diamine tetraacetic acid (EDTA) in 0.1 m&L Trisbuffer (pH 6.95). The end point of EDTA decalcification was checked by Seilly’s chemical test5 Temporal bones were dehydrated with graded ethyl alcohol and embedded in glycol methacrylate (JB-4; Polysciences, Warrington, PA). Using glass knives, 5-km thick sections were made and stained with hematoxylin and eosin. Sections adjacent to the hematoxylin-eosin stained sections were stained with Alcian blue (pH 2.5) and periodic acid-Schiff (AB-PAS) to demonstrate mucosubstances of the secretory cells. For 3D reconstruction, three reference points were made on the JB-4 block near the specimen before serial sectionmg. For these reference points, three V-shaped cutting notches were made on different walls of the JB-4 block for orientation. The block was then re-embedded and serial sections were made at a 5-pm thickness. Because reference notches in the re-embedded block could not be identified after staining, unstained sections were used for tracing the outline of the tubotympanal cavity with the help of a projection microscope (Trisimplex, Bausch and Lomb, Rochester, NY) and a digitizing tablet attached to a desktop computer. At 30-pm intervals, the cross-section area of each eustachian tube and middle ear cavity was traced using a digitizing tablet, and the volume of the eustachian tube and middle ear cavity was computed using morphometric software (Jandel Scientific, Corte Madera, CA]. The traced images of the serial sections were aligned using the three reference points, and 3D reconstruction and rendering of the tubotympanal cavities were made using a 3D reconstruction program developed in this laboratory.
PARK, UENO, AND LIM
RESULTS During gestation, the eustachian tube and middle ear cavity are derived from the expanding terminal end of the endoderm lining the first pharyngeal pouch, which is an outpocketing of the foregut, and this primitive pharynx was observed on the 11th day. Between the 12th and 13th days, the first pharyngeal pouch elongated to form the tubotympanic recess (Fig 1A). At these ages, the lumen
of the tubotympanic recess showed a relatively smooth surface with nonciliated low columnar epithelium. Between the 13th and 14th days, the tubotympanic recess extended toward the middle ear near the stapedial primordium and Reichert’s cartilage (Fig 1B). At the 16th day, the eustachian tube and middle ear cavity became distinctly segregated (Fig 2A). As the future middle ear cavity enlarged after the segregation, its cavity was occupied by loose mesenchymal connective tissue,
Fig 1. (A) The pharyngeel pouch (P) elongates to form the tubotympanic recess (TRJ in e 12-day gestationel mouse. CO, cochlea. (Axial section, H&E, original magnification x33.1 (6) The tubotympenic recess (TN extends to the middle ear area from the pharynx(P) and surrounds the cochlea (CO1 in a U-day gestational mouse. (Axial section, H&E, original magnification x33.1
ANATOMY
OF EUSTACHIAN
TUBE/MIDDLE
EAR
95
tissue (Fig ZB). At around postnatal day 9, the eustachian tube attained an adult shape (Fig X). The middle ear cavity attained its adult shape near day 11 when the mesenchymal tissue in the middle ear disappeared (Fig 3A). At day 15, the epitympanic space attained its adult shape and size by an absorption of the mesenchymal tissue (Fig 3B). By day 21, the eustachian tube and middle ear cavity attained their adult size, about 0.007 mL in volume. A progression of the development of the eustachian tube lumen and middle ear cavity at different gestational and postnatal ages were reconstructed through a computer-aided 3D reconstruction program (Fig 4). The volume of
Fig 2. IA) The tubotympanic recess divides into the eustachian tube (ET) and middle eer cavity (ME) in a %-day gestational mouse. (Coronal section, H&E, original magnification x33.1 (Bl The future middle ear cavity is Clled with the mesenchymal tissue (MT). Tubal glands (G), muscle (MI, and cartilage (Cl are well identified in a l-day postnatal mouse. CO, cochlea. (Axial section, H&E, origins} magnification x33.1 (C) The eustachian tube (ET) attains an adult form in a g-day postnatal mouse. The mesenchymal tissue ChlTj is noted in the middle ear cavity. P, pharyngeel opening of the tube. (Coronal section, H&E, original magnification x33.)
which lay ture bony At birth, filled with
between the epithelium and the fustructure. most of the middle ear cavity was the loose mesenchymal connective
Fig 3. (A) The mesenchymal tissue (MT) has disappeared from the middle ear cavity in an 11-day postnatal mouse but is still around the ossicle (01. CO, cochlea; TM, tympanic membrane; ES, epitympanic space. (Coronal section, H&E, original magnification x55.1 lB) The mesenchymal tissue around the ossicle (01 has disappeared and the epitympanic space (ES) attains its adult size in a 15-day postnatal mouse. CO, cochlea; TM, tympanic membrane. (Coronal section. H&E, original magnification x55.1
96
PARK,
UENO,
AND
LIM
Fig 4. 3D reconstruction of the eustachian tube lumen and middle ear cavity at different gestational (013-18) and postnatal (Pl-21) ages. (Medial view of a right tubotympanal cavity from anterosuperior direction, original magnification x 15.)
the tubotympanic recess underwent a drastic change between the 15th and 16th gestation day (Fig 5A). During the postnatal period, the maximum volume change occurred on day 11, when the mesenchymal tissue in the middle ear cavity disappeared completely (Fig 5B). The luminal development was formed by the invagination of the tubotympanic recess and the resorption of the mesenchymal tissue in the tympanic cavity. The volume of the tubotympanic cavity increased in two distinct phases. The first phase occurred from gestation days 14 through 16, owing to the rapid development of the tympanic cavity. This cavity became prominent with the formation of bottleneck constriction of the tubotympanic
recess. The later phase occurred from postnatal days 9 through 15, owing to the rapid resorption of the tympanal mesenchyme. Cell division in the epithelium and mesenchymal regression were accelerated in these two phases. At the 16th day of gestation, the tubal lumen was lined by a tall ciliated pseudostratified columnar epithelium transformed from a low columnar epithelium (Fig 6A). Cuboidal ciliated cells were observed in the tympanic cavity. The muscle fibers of the tensor and levator veli palatini were first observed. By the 17th day, these became bundles (Fig 6B). At this time, the epithelial secretory cells were first observed in both the eustachian
ANATOMY
OF EUSTACHIAN
TUBE/MIDDLE
97
EAR
A
5
1-
,"
o-
,
,
17
16
., 13
12
14
15
16
.
19
Gestatlonal day
B
1
3
5
7
9
11
13
15
17
19
21
Postnatal day
Fig 5. (A) A graphic demonstration of the growth of the tubotympanic recess volume during gestation. The volume of the tubotympanic recess undergoes the maximum change bafore and after gestational day 15. So-caged bottleneck constriction is made at gestational day 16, at which the tubotympanic recess divides into the eustachian tube and middle ear cavity. WI A graphic demonstration of the growth of the eustachian tube and middle ear cavity volume after birth. The maximum changein volume is noted at postnatal day 11 when the mesenchymal tissue in the middle ear cavity disappeared.
tube and middle ear (Fig 7A). The tubal epithelial cells started to accumulate to granular AB-PAS-positive materials at day 16. There was an intense accumulation of fine granular AB-PAS-positive material in the adjacent mesenchymal tissue near the time of birth (Fig 7B). At the 18th day of gestation, a few glandlike structures were observed near the pharyngeal orifice of the eustachian tube. However, these glandular cells were not stained with AB-PAS. At postnatal day 1, tubal glands as well as tubal cartilages were clearly recognizable (Fig ZB). The tubal muscles were also well developed at this point. The number of ciliated and secretory cells in the eustachian tube epithelium was already near the adult level at birth. Table 1 summarizes the development of the eustachian tube and its related structures. The developing steps of the structures in the tubotympanic cavity of the mouse are compared with published data on the human (Table 2). The developmental process of the main parts of the murine eustachian tube and mid-
Fig 6. (A) The mucosal lining of the eustachian tube shows the development of cilia (arrowheads) in a 16-day gestational mouse (H&E, original magnification x640.) (B) The early development of the tubal muscle If41 is recognizable in a 17-day gestetional mouse. The bony cavity of the middle ear is occupied by the mesenchymal tissue (MT). ET, eustachian tube; ME, middle ear cavity; P, pharynx. (Coronal, HIE, original magnification x 33.)
dle ear is very similar to that of the human as described in several reports.6-12 The difference between the human and murine tubotympanal structures in development is that in the mouse they are fully developed around birth, whereas in the human, they are fully developed by the first half of the pregnancy.
DISCUSSION Our data suggest that ciliated cells appeared first at the 16th gestation day when the middle ear cleft was being formed, and were found concurrently in the tubal lumen as well as in the expanding tympanic cavity. This latter finding was unexpected, failing to support the commonly held view that the development of the tubal lumen precedes that of the tympanic cavity.
98
Fig 7. (A) A few positively stained epithelial sacretory cells (arrowheads) in both the eustachian tube (ET) and middle ear (ME) are concurrently noted in a 17-day gestational mouse. (ABPAS, original magnification x33.) (B) Intensively positive stainings (arrowheads) am noted pericellularly and in the messnchymal tissue underlying epithelium (arrows) in an W-day gestational mouse eustachian tube. (AB-PAS, original magnification x640.1
As in the case of ciliated cells, the secretory cells developed simultaneously in the middle ear and eustachian tube. The epithelial secretory cells had already synthesized secretory products at an early developmental age as evidenced by AB-PAS staining characteristics. Although the epithelial secretory cells appeared 1 day later than the ciliated cells, their distribution was parallel. This observation is consistent with our earlier report on humans and laboratory animals.13*14 The parallel distribution of ciliated and secretory cells is not surprising, considering the complementary roles they play in the physiology of the mucociliary system. Based on morphology and autoradiography conducted in this laboratory,“-17 Lim and associates have suggested that there are functionally different secretory cells containing
PARK, UENO, AND LIM
either dark, light, or mixed secretory granules.‘8s1g The dark granulated cells incorporated a greater amount of tritiated leucine than the light granulated cells (goblet cells], which incorporated a greater amount of tritiated glucose.18s1g The biochemical diversity of epithelial secretory cells may be needed to support the functioning mucociliary system that requires specialized components of a mucociliary system, such as a mucous blanket and serous periciliary fluid. A previous study in this laboratory using lectin labeling also showed that the secretory cells of the tubotympanal epithelium of mature animals secrete both serum type and mucoid type glycoconjugates.lg The serum type of secretory granules were “dark granule,” and the mucoid type of secretory granules were “light granule” as described earlier.13,15 Because the present study did not include lectin histochemistry, we were not able to determine whether this diversity of secretory cells existed during development. It has been commonly believed that the tympanal epithelium is formed by the invagination of the tubotympanic recess. Marovitz et al” suggested a dual origin: the tubal epithelium originates from the endoderm and the tympanal epithelium from the mesoderm (or mesenchyme). For mesenchymal cells to join the epithelium of an invaginating tubotympanic recess, a break in the basal lamina must occur. On the other hand, Hilding et a121 failed to demonstrate a break in the basal lamina through the whole tubotympanal epithelium. They concluded that the mesenchyme does not participate in tympanal epithelial formation in the tympanum. If formation of the middle ear cavity occurred by simple migration of the epithelium from the pharynx, ciliated or secretory cells should have been observed in the eustachian tube earlier than in the middle ear. However, our morphologic finding shows that the epithelial differentiation in the tympanic cavity appeared at the same time as in the eustachian tube, with the exception of poorly differentiated epithelium lining the mesenchymal tissue. This observation implies that the differentiation of the tubal epithelium and that of the middle ear may be independent events.
ANATOMY
OF EUSTACHIAN
TABLE1.
Presence
Postnatal
(P) Age
TUBE/MIDDLE
of Eustachian
99
EAR
Tube and Its Related
Structures
in Specimens
of Different
Gestational
(G) and
Epithelium
Age (d)
Lumen
Gil G12 G13 G14 G15 G16 G17 G18 Pl P3 P5 P7 P9 Pll P13 P15 P17 P19 P21
+ + + + + + + + + + + + + + + + + +
Abbreviations:
G, gestational;
Ciliated Cell
Secretory Cell
_ + + + + + + + + + + + + + +
+ + + + + + + + + + + + +
Gland
Muscle
_ _
-
+ + + + + + + + + + + +
Cartilage
_ + + + + + + + + + + + + + +
_ + + + + + + + + + + +
P, postnatal.
The mechanisms involved in cell differentiation during organogenesis of the eustachian tube and middle ear are not well understood. However, the differentiation of one type of epithelial cell into a specialized cell, such as a ciliated or secretory cell, is believed to be controlled by a genetic message encoded in the cell and by the influence of the mesenchymal cells underlying the epithelium.” It is well known that epithelial-mesenchymal interactions are required for epithelial differentiation in organogenesis. Our study showed an accumulation of granular AB-PAS-positive materials in the tubal epithelial cells and in the adjacent mesenchymal tissue during development, which may suggest a mesenchymal influence on epithelial differentiation. Developing organs are intimately associated with glycoconjugates that are produced by the epithelial cells themselves or mesenchymal cells underlying the epithelium. It is also known that cell-surface glycoproteins mediate signals for embryonic development, growth regulation, and the maintenance of normal function.23*24 It is interesting that the poorly differentiated epithelium in the tympanic cavity is intimately associated with large amounts of
the mesenchyme, and the disappearance of these mesenchymes coincides with the differentiation of the medial side of the tympanal epithelium. This finding can be interpreted to mean that the presence of mesenchyme is a prerequisite for epithelial differentiation. The role of the mesenchymal tissue in epithelial TABLE2. Differences Development
in Mouse and Human of Structures in Tubotympanal
Findings of Developing Structures in Tubotympanal Cavity Formation of tubal lumen begins Ciliated cell appears Secretory cell appears Tubal muscle appears Tubal gland appears Tubal cartilage appears Pneumatization of middle ear completed Pneumatization of epitympanum completed Abbreviations: G, gestational; ‘Swarts et al.g tTos.’ $ Bast et al.”
During Cavity
Age Mouse
(d)
Human
G12
(wk)
G8^
G16 G17 G16-17 G18-Pl Pl Pll
G15 G13T Gll,‘GlO-12t G13; G15T G12; G15t G30$
P15
G38*
P, postnatal.
100
migration and differentiation of the eustachian tube and middle ear requires further study. ACKNOWLEDGMENT The authors wish to thank Atha Ralston and Ilija Karanfilov for technical assistance, Asish Law for 3D computer graphics, Valerie Jones for photography, Teresa Black for manuscript preparation, and Sarah Bowers for word processing.
REFERENCES 1. Lim DJ: Functional morphology of the mucosa of the middle ear and eustachian tube. Ann Otol Rhino1 Laryngo1 2536-43, 1976 (suppl 85) 2. Shimada T, Lim DJ: Distribution of ciliated cells in the human middle ear: electron and light microscopic observations. Ann Otol Rhino1 Laryngol81:203-211,1972 3. Masuda Y, Honjo H, Naito M, et al: Normal development of the middle ear in the mouse: a light microscopic study of serial sections. Acta Med Okayama 4O:iOl-207, i986 4. Huanefu M. Saunders IC: Auditorv develoument in the mouse:“Structural maturation of the middle ear. J Morphol 176:249-259, 1983 5. Seilly DJ: A chemical test to determine the end point of EDTA decalcification. Med Lab Sci 39:71-73, 1982 6. Tos M: Growth of the fetal eustachian tube and its dimensions. Arch Klin Exp Ohren Nasen Kehlkopfheilkd 198:177-186, 1971 7. Tos M: Goblet cells in the human fetal eustachian tube. Arch Otolaryngol 93:365-373, 1971 8. Tos M: Development of mucous glands in the human eustachian tube. Acta Otolaryngol (Stockh) 70:340-350, 1970 9. Swarts JD, Rood SR, Doyle WJ: Fetal development of the auditory tube and paratubal musculature. Cleft Palate J 23:289-311, 1986 10. Kitajiri M, Sando I, Takahara T: Postnatal development of the eustachian tube and its surrounding structures. Ann Otol Rhino1 Laryngol 96:191-198, 1987 11. Bast T, Anson A: The Temporal Bone and the Ear:
PARK, UENO, AND LIM
The Origin and Development of the Middle Ear and Related Air Space. Springfield, IL, Thomas, 1949, pp 306336 12. Lim DJ: Functional morphology of the lining membrane of the middle ear and eustachian tube: An overview. Ann Otol Rhino1 Laryngol 11:5-18,1974 (suppl 83) 13. Lim DJ, Shimada T, Yoder M: Distribution of mucous-secreting cells in normal middle ear mucosa. Arch Otolaryngol 98:2-g, 1973 14. Lim DJ, Paparella MM, Kimura RS: Ultrastructure of the eustachian tube and middle ear mucosa in the guinea pig. Acta Otolaryngol 63:425-444, 1967 15. Hussl B, Lim DJ: Secretory cells in the middle ear mucosa of the guinea pig: Cytochemical and ultrastructural study. Arch Otolaryngol 89:691-699, 1969 16. Lim DJ, Shimada T: Secretory activity of normal middle ear epithelium: Scanning and transmission electron microscopic observations. Ann Otol Rhino1 Laryngol 80:319-329, 1971 17. Lim DJ, Vial1 J, Birck H, et al: The morphological basis for understanding middle ear effusion: An electron microscopic, cytochemical, and autoradiographic investigation. Laryngoscope 82:1625-1643, 1972 18. Lim DJ: Protein secreting cells in the normal middle ear mucosa of the guinea pig: An autoradiographic investigation. Ann Otol Rhino1 Laryngol 79:82-94, 1970 19. Ueno K, Lim DJ: Heterogeneity of glycoconjugates in the secretory cells of the chinchilla middle ear and eustachian tubal epithelia: A lectin-gold cytochemical study. J Histochem Cytochem 39:71-80, 1991 20. Marovitz WF, Porubsky ES: The embryological development of the middle ear space: a new concept. Ann Otol Rhino1 Laryngol 80:384-389, 1971 21. Hilding DA, Szachowicz E, Larsen SA: Development of the epithelium of the middle ear: Electron microscopic study of fine structures, including junctional complexes and basal lamina. Am J Otolaryngol 1:97-108,198O 22. Holtfreter J: Mesenchyme and epithelia in inductive and morphogenic processes, in Fleischmager R, Billingham RE (eds): Epithelial-Mesenchymal Interactions. Baltimore, MD, Williams & Wilkins, 1968, pp l-30 23. Wieland F: Are sulphated cell surface glycoproteins involved in the control of cell differentiation? Trends Biochem Sci 7:308-309, 1982 24. Yamada KH: Cell surface interactions with extracellular materials. Am Rev Biochem 52:761-799, 1983
Developmental Anatomy of the Eustachian Tube and Middle Ear in Mice KEEt -I‘YUN PARK, MD, KAZUYOSHI
UENO, MD, AND DAVID J. LIM, MD
Purpose: It is generally accepted that the development of the tubotympanum has significant bearing on the susceptibility to ear infection. A detailed study of the differentiation of ciliated cells in secretory elements will be useful in understanding both the normal physiology and the pathology of the tubotympanum. Method: Serially sectioned temporal bones of 76 mice ranging from gestational age day 11 to postnatal day 21 were examined microscopically. Results: During the period of gestation, the tubotympanic recess was formed at the 12th day and began to extend to form the middle ear between the 13th and 14th days. A rapid increase in the volume of the tubotympanic recess was observed between the 15th and 16th days when a definitive division of the tubotympanic recess into the eustachian tube and middle ear cavity was observed. Postnatally the tubotympanum attained an adult form around day 9, and the maximum change of middle ear volume was noted on day 11, when the mesenchymal tissue in the middle ear cavity disappeared completely. Development of the ciliated cells was observed concurrently in both the eustachian tube and middle ear on the 16th gestational day, one day earlier than the appearance of the epithelial secretory cells in both the eustachian tube and middle ear. The number of ciliated cells and secretory cells increased rapidly after birth. Tubal glands were well developed with evidence of secretory activity around the time of birth. Conclusions: Based on these findings, one can conclude that the mucociliary defense system starts to develop during the fetal stage and is well established immediately after birth. Copyright 0 1992 by W.B. Saunders Company
It is generally accepted that the development of the tubotympanum has significant bearing on the susceptibility to ear infection. The overwhelming concern of research focusing on the development of the tubotympanum has been the development of the tubal cartilage and muscles. In recent years, there has been good evidence to suggest that a dysfunctional muco-
From the Otological Research Laboratories, Department of Otolaryngology, The Ohio State University College of Medicine, Columbus, OH. Address reprint requests to David J. Lim, MD, Division of Intramural Research, NIDCDINIH, Bldg 31, Room 3CO6, 9000 Rockville Pike, Bethesda, MD 20692. Supported in part by Grant No. DC00090 from the National Institute of Deafness and Other Communication Disorders, National Institutes of Health. Copyright 0 1992 by W.B. Saunders Company 0196-0709/92/l 302-0004$5.00/O American
Journal
of Otolaryngology,
ciliary transport mechanism (such as in immotile cilia syndrome) in the tubotympanum and sinus leads to chronic or recurrent otitis media and/or sinusitis. Although the importance of the mucociliary defense system in the tubotympanum has been well established,‘*’ the development of the mucociliary system has been poorly documented. Although the developmental anatomy of the murine middle ear has been described earlier,3P4 a description of the development of the eustachian tube mucociliary system was lacking. A detailed study of the differentiation of ciliated cells and secretory elements will be useful in understanding both the normal physiology and pathology of the tubotympanum. In this report, the developmental anatomy of the murine eustachian tube and middle ear, with particular emphasis on the differentiation of epithelial cells (ciliated and secretory)
Vol 13, No 2 (March-April),
1992: pp 93-100
93
94
and luminal formation of the eustachian tube, including glands, was investigated using light microscopy (LM), computer-aided threedimensional (3D) reconstruction, and morphometry. MATERIALS AND METHODS The mouse was chosen for this study of the developmental morphology of the mammalian eustachian tube and middle ear because of its relatively short periods of gestational and postnatal development, and its rapid breeding ability. A total of 76 BALB/c mice, ranging in age from gestational day 11 through postnatal day 21,were used in this investigation. Gestational day was determined by the vaginal plug technique, establishing the first day of appearance of the vaginal plug as day one. Postnatal mice with evidence of middle ear infection were excluded. For LM observation, after decapitation under inhalation anesthesia with Halothane (Halocaroon Laboratories, Hackensack, NJ), specimens were fixed by immersion in 10% neutral buffered formalin for 7 days at 4°C. Temporal bones from postnatal mice were decalcified with 10% ethylene diamine tetraacetic acid (EDTA) in 0.1 m&L Trisbuffer (pH 6.95). The end point of EDTA decalcification was checked by Seilly’s chemical test5 Temporal bones were dehydrated with graded ethyl alcohol and embedded in glycol methacrylate (JB-4; Polysciences, Warrington, PA). Using glass knives, 5-km thick sections were made and stained with hematoxylin and eosin. Sections adjacent to the hematoxylin-eosin stained sections were stained with Alcian blue (pH 2.5) and periodic acid-Schiff (AB-PAS) to demonstrate mucosubstances of the secretory cells. For 3D reconstruction, three reference points were made on the JB-4 block near the specimen before serial sectionmg. For these reference points, three V-shaped cutting notches were made on different walls of the JB-4 block for orientation. The block was then re-embedded and serial sections were made at a 5-pm thickness. Because reference notches in the re-embedded block could not be identified after staining, unstained sections were used for tracing the outline of the tubotympanal cavity with the help of a projection microscope (Trisimplex, Bausch and Lomb, Rochester, NY) and a digitizing tablet attached to a desktop computer. At 30-pm intervals, the cross-section area of each eustachian tube and middle ear cavity was traced using a digitizing tablet, and the volume of the eustachian tube and middle ear cavity was computed using morphometric software (Jandel Scientific, Corte Madera, CA]. The traced images of the serial sections were aligned using the three reference points, and 3D reconstruction and rendering of the tubotympanal cavities were made using a 3D reconstruction program developed in this laboratory.
PARK, UENO, AND LIM
RESULTS During gestation, the eustachian tube and middle ear cavity are derived from the expanding terminal end of the endoderm lining the first pharyngeal pouch, which is an outpocketing of the foregut, and this primitive pharynx was observed on the 11th day. Between the 12th and 13th days, the first pharyngeal pouch elongated to form the tubotympanic recess (Fig 1A). At these ages, the lumen
of the tubotympanic recess showed a relatively smooth surface with nonciliated low columnar epithelium. Between the 13th and 14th days, the tubotympanic recess extended toward the middle ear near the stapedial primordium and Reichert’s cartilage (Fig 1B). At the 16th day, the eustachian tube and middle ear cavity became distinctly segregated (Fig 2A). As the future middle ear cavity enlarged after the segregation, its cavity was occupied by loose mesenchymal connective tissue,
Fig 1. (A) The pharyngeel pouch (P) elongates to form the tubotympanic recess (TRJ in e 12-day gestationel mouse. CO, cochlea. (Axial section, H&E, original magnification x33.1 (6) The tubotympenic recess (TN extends to the middle ear area from the pharynx(P) and surrounds the cochlea (CO1 in a U-day gestational mouse. (Axial section, H&E, original magnification x33.1
ANATOMY
OF EUSTACHIAN
TUBE/MIDDLE
EAR
95
tissue (Fig ZB). At around postnatal day 9, the eustachian tube attained an adult shape (Fig X). The middle ear cavity attained its adult shape near day 11 when the mesenchymal tissue in the middle ear disappeared (Fig 3A). At day 15, the epitympanic space attained its adult shape and size by an absorption of the mesenchymal tissue (Fig 3B). By day 21, the eustachian tube and middle ear cavity attained their adult size, about 0.007 mL in volume. A progression of the development of the eustachian tube lumen and middle ear cavity at different gestational and postnatal ages were reconstructed through a computer-aided 3D reconstruction program (Fig 4). The volume of
Fig 2. IA) The tubotympanic recess divides into the eustachian tube (ET) and middle eer cavity (ME) in a %-day gestational mouse. (Coronal section, H&E, original magnification x33.1 (Bl The future middle ear cavity is Clled with the mesenchymal tissue (MT). Tubal glands (G), muscle (MI, and cartilage (Cl are well identified in a l-day postnatal mouse. CO, cochlea. (Axial section, H&E, origins} magnification x33.1 (C) The eustachian tube (ET) attains an adult form in a g-day postnatal mouse. The mesenchymal tissue ChlTj is noted in the middle ear cavity. P, pharyngeel opening of the tube. (Coronal section, H&E, original magnification x33.)
which lay ture bony At birth, filled with
between the epithelium and the fustructure. most of the middle ear cavity was the loose mesenchymal connective
Fig 3. (A) The mesenchymal tissue (MT) has disappeared from the middle ear cavity in an 11-day postnatal mouse but is still around the ossicle (01. CO, cochlea; TM, tympanic membrane; ES, epitympanic space. (Coronal section, H&E, original magnification x55.1 lB) The mesenchymal tissue around the ossicle (01 has disappeared and the epitympanic space (ES) attains its adult size in a 15-day postnatal mouse. CO, cochlea; TM, tympanic membrane. (Coronal section. H&E, original magnification x55.1
96
PARK,
UENO,
AND
LIM
Fig 4. 3D reconstruction of the eustachian tube lumen and middle ear cavity at different gestational (013-18) and postnatal (Pl-21) ages. (Medial view of a right tubotympanal cavity from anterosuperior direction, original magnification x 15.)
the tubotympanic recess underwent a drastic change between the 15th and 16th gestation day (Fig 5A). During the postnatal period, the maximum volume change occurred on day 11, when the mesenchymal tissue in the middle ear cavity disappeared completely (Fig 5B). The luminal development was formed by the invagination of the tubotympanic recess and the resorption of the mesenchymal tissue in the tympanic cavity. The volume of the tubotympanic cavity increased in two distinct phases. The first phase occurred from gestation days 14 through 16, owing to the rapid development of the tympanic cavity. This cavity became prominent with the formation of bottleneck constriction of the tubotympanic
recess. The later phase occurred from postnatal days 9 through 15, owing to the rapid resorption of the tympanal mesenchyme. Cell division in the epithelium and mesenchymal regression were accelerated in these two phases. At the 16th day of gestation, the tubal lumen was lined by a tall ciliated pseudostratified columnar epithelium transformed from a low columnar epithelium (Fig 6A). Cuboidal ciliated cells were observed in the tympanic cavity. The muscle fibers of the tensor and levator veli palatini were first observed. By the 17th day, these became bundles (Fig 6B). At this time, the epithelial secretory cells were first observed in both the eustachian
ANATOMY
OF EUSTACHIAN
TUBE/MIDDLE
97
EAR
A
5
1-
,"
o-
,
,
17
16
., 13
12
14
15
16
.
19
Gestatlonal day
B
1
3
5
7
9
11
13
15
17
19
21
Postnatal day
Fig 5. (A) A graphic demonstration of the growth of the tubotympanic recess volume during gestation. The volume of the tubotympanic recess undergoes the maximum change bafore and after gestational day 15. So-caged bottleneck constriction is made at gestational day 16, at which the tubotympanic recess divides into the eustachian tube and middle ear cavity. WI A graphic demonstration of the growth of the eustachian tube and middle ear cavity volume after birth. The maximum changein volume is noted at postnatal day 11 when the mesenchymal tissue in the middle ear cavity disappeared.
tube and middle ear (Fig 7A). The tubal epithelial cells started to accumulate to granular AB-PAS-positive materials at day 16. There was an intense accumulation of fine granular AB-PAS-positive material in the adjacent mesenchymal tissue near the time of birth (Fig 7B). At the 18th day of gestation, a few glandlike structures were observed near the pharyngeal orifice of the eustachian tube. However, these glandular cells were not stained with AB-PAS. At postnatal day 1, tubal glands as well as tubal cartilages were clearly recognizable (Fig ZB). The tubal muscles were also well developed at this point. The number of ciliated and secretory cells in the eustachian tube epithelium was already near the adult level at birth. Table 1 summarizes the development of the eustachian tube and its related structures. The developing steps of the structures in the tubotympanic cavity of the mouse are compared with published data on the human (Table 2). The developmental process of the main parts of the murine eustachian tube and mid-
Fig 6. (A) The mucosal lining of the eustachian tube shows the development of cilia (arrowheads) in a 16-day gestational mouse (H&E, original magnification x640.) (B) The early development of the tubal muscle If41 is recognizable in a 17-day gestetional mouse. The bony cavity of the middle ear is occupied by the mesenchymal tissue (MT). ET, eustachian tube; ME, middle ear cavity; P, pharynx. (Coronal, HIE, original magnification x 33.)
dle ear is very similar to that of the human as described in several reports.6-12 The difference between the human and murine tubotympanal structures in development is that in the mouse they are fully developed around birth, whereas in the human, they are fully developed by the first half of the pregnancy.
DISCUSSION Our data suggest that ciliated cells appeared first at the 16th gestation day when the middle ear cleft was being formed, and were found concurrently in the tubal lumen as well as in the expanding tympanic cavity. This latter finding was unexpected, failing to support the commonly held view that the development of the tubal lumen precedes that of the tympanic cavity.
98
Fig 7. (A) A few positively stained epithelial sacretory cells (arrowheads) in both the eustachian tube (ET) and middle ear (ME) are concurrently noted in a 17-day gestational mouse. (ABPAS, original magnification x33.) (B) Intensively positive stainings (arrowheads) am noted pericellularly and in the messnchymal tissue underlying epithelium (arrows) in an W-day gestational mouse eustachian tube. (AB-PAS, original magnification x640.1
As in the case of ciliated cells, the secretory cells developed simultaneously in the middle ear and eustachian tube. The epithelial secretory cells had already synthesized secretory products at an early developmental age as evidenced by AB-PAS staining characteristics. Although the epithelial secretory cells appeared 1 day later than the ciliated cells, their distribution was parallel. This observation is consistent with our earlier report on humans and laboratory animals.13*14 The parallel distribution of ciliated and secretory cells is not surprising, considering the complementary roles they play in the physiology of the mucociliary system. Based on morphology and autoradiography conducted in this laboratory,“-17 Lim and associates have suggested that there are functionally different secretory cells containing
PARK, UENO, AND LIM
either dark, light, or mixed secretory granules.‘8s1g The dark granulated cells incorporated a greater amount of tritiated leucine than the light granulated cells (goblet cells], which incorporated a greater amount of tritiated glucose.18s1g The biochemical diversity of epithelial secretory cells may be needed to support the functioning mucociliary system that requires specialized components of a mucociliary system, such as a mucous blanket and serous periciliary fluid. A previous study in this laboratory using lectin labeling also showed that the secretory cells of the tubotympanal epithelium of mature animals secrete both serum type and mucoid type glycoconjugates.lg The serum type of secretory granules were “dark granule,” and the mucoid type of secretory granules were “light granule” as described earlier.13,15 Because the present study did not include lectin histochemistry, we were not able to determine whether this diversity of secretory cells existed during development. It has been commonly believed that the tympanal epithelium is formed by the invagination of the tubotympanic recess. Marovitz et al” suggested a dual origin: the tubal epithelium originates from the endoderm and the tympanal epithelium from the mesoderm (or mesenchyme). For mesenchymal cells to join the epithelium of an invaginating tubotympanic recess, a break in the basal lamina must occur. On the other hand, Hilding et a121 failed to demonstrate a break in the basal lamina through the whole tubotympanal epithelium. They concluded that the mesenchyme does not participate in tympanal epithelial formation in the tympanum. If formation of the middle ear cavity occurred by simple migration of the epithelium from the pharynx, ciliated or secretory cells should have been observed in the eustachian tube earlier than in the middle ear. However, our morphologic finding shows that the epithelial differentiation in the tympanic cavity appeared at the same time as in the eustachian tube, with the exception of poorly differentiated epithelium lining the mesenchymal tissue. This observation implies that the differentiation of the tubal epithelium and that of the middle ear may be independent events.
ANATOMY
OF EUSTACHIAN
TABLE1.
Presence
Postnatal
(P) Age
TUBE/MIDDLE
of Eustachian
99
EAR
Tube and Its Related
Structures
in Specimens
of Different
Gestational
(G) and
Epithelium
Age (d)
Lumen
Gil G12 G13 G14 G15 G16 G17 G18 Pl P3 P5 P7 P9 Pll P13 P15 P17 P19 P21
+ + + + + + + + + + + + + + + + + +
Abbreviations:
G, gestational;
Ciliated Cell
Secretory Cell
_ + + + + + + + + + + + + + +
+ + + + + + + + + + + + +
Gland
Muscle
_ _
-
+ + + + + + + + + + + +
Cartilage
_ + + + + + + + + + + + + + +
_ + + + + + + + + + + +
P, postnatal.
The mechanisms involved in cell differentiation during organogenesis of the eustachian tube and middle ear are not well understood. However, the differentiation of one type of epithelial cell into a specialized cell, such as a ciliated or secretory cell, is believed to be controlled by a genetic message encoded in the cell and by the influence of the mesenchymal cells underlying the epithelium.” It is well known that epithelial-mesenchymal interactions are required for epithelial differentiation in organogenesis. Our study showed an accumulation of granular AB-PAS-positive materials in the tubal epithelial cells and in the adjacent mesenchymal tissue during development, which may suggest a mesenchymal influence on epithelial differentiation. Developing organs are intimately associated with glycoconjugates that are produced by the epithelial cells themselves or mesenchymal cells underlying the epithelium. It is also known that cell-surface glycoproteins mediate signals for embryonic development, growth regulation, and the maintenance of normal function.23*24 It is interesting that the poorly differentiated epithelium in the tympanic cavity is intimately associated with large amounts of
the mesenchyme, and the disappearance of these mesenchymes coincides with the differentiation of the medial side of the tympanal epithelium. This finding can be interpreted to mean that the presence of mesenchyme is a prerequisite for epithelial differentiation. The role of the mesenchymal tissue in epithelial TABLE2. Differences Development
in Mouse and Human of Structures in Tubotympanal
Findings of Developing Structures in Tubotympanal Cavity Formation of tubal lumen begins Ciliated cell appears Secretory cell appears Tubal muscle appears Tubal gland appears Tubal cartilage appears Pneumatization of middle ear completed Pneumatization of epitympanum completed Abbreviations: G, gestational; ‘Swarts et al.g tTos.’ $ Bast et al.”
During Cavity
Age Mouse
(d)
Human
G12
(wk)
G8^
G16 G17 G16-17 G18-Pl Pl Pll
G15 G13T Gll,‘GlO-12t G13; G15T G12; G15t G30$
P15
G38*
P, postnatal.
100
migration and differentiation of the eustachian tube and middle ear requires further study. ACKNOWLEDGMENT The authors wish to thank Atha Ralston and Ilija Karanfilov for technical assistance, Asish Law for 3D computer graphics, Valerie Jones for photography, Teresa Black for manuscript preparation, and Sarah Bowers for word processing.
REFERENCES 1. Lim DJ: Functional morphology of the mucosa of the middle ear and eustachian tube. Ann Otol Rhino1 Laryngo1 2536-43, 1976 (suppl 85) 2. Shimada T, Lim DJ: Distribution of ciliated cells in the human middle ear: electron and light microscopic observations. Ann Otol Rhino1 Laryngol81:203-211,1972 3. Masuda Y, Honjo H, Naito M, et al: Normal development of the middle ear in the mouse: a light microscopic study of serial sections. Acta Med Okayama 4O:iOl-207, i986 4. Huanefu M. Saunders IC: Auditorv develoument in the mouse:“Structural maturation of the middle ear. J Morphol 176:249-259, 1983 5. Seilly DJ: A chemical test to determine the end point of EDTA decalcification. Med Lab Sci 39:71-73, 1982 6. Tos M: Growth of the fetal eustachian tube and its dimensions. Arch Klin Exp Ohren Nasen Kehlkopfheilkd 198:177-186, 1971 7. Tos M: Goblet cells in the human fetal eustachian tube. Arch Otolaryngol 93:365-373, 1971 8. Tos M: Development of mucous glands in the human eustachian tube. Acta Otolaryngol (Stockh) 70:340-350, 1970 9. Swarts JD, Rood SR, Doyle WJ: Fetal development of the auditory tube and paratubal musculature. Cleft Palate J 23:289-311, 1986 10. Kitajiri M, Sando I, Takahara T: Postnatal development of the eustachian tube and its surrounding structures. Ann Otol Rhino1 Laryngol 96:191-198, 1987 11. Bast T, Anson A: The Temporal Bone and the Ear:
PARK, UENO, AND LIM
The Origin and Development of the Middle Ear and Related Air Space. Springfield, IL, Thomas, 1949, pp 306336 12. Lim DJ: Functional morphology of the lining membrane of the middle ear and eustachian tube: An overview. Ann Otol Rhino1 Laryngol 11:5-18,1974 (suppl 83) 13. Lim DJ, Shimada T, Yoder M: Distribution of mucous-secreting cells in normal middle ear mucosa. Arch Otolaryngol 98:2-g, 1973 14. Lim DJ, Paparella MM, Kimura RS: Ultrastructure of the eustachian tube and middle ear mucosa in the guinea pig. Acta Otolaryngol 63:425-444, 1967 15. Hussl B, Lim DJ: Secretory cells in the middle ear mucosa of the guinea pig: Cytochemical and ultrastructural study. Arch Otolaryngol 89:691-699, 1969 16. Lim DJ, Shimada T: Secretory activity of normal middle ear epithelium: Scanning and transmission electron microscopic observations. Ann Otol Rhino1 Laryngol 80:319-329, 1971 17. Lim DJ, Vial1 J, Birck H, et al: The morphological basis for understanding middle ear effusion: An electron microscopic, cytochemical, and autoradiographic investigation. Laryngoscope 82:1625-1643, 1972 18. Lim DJ: Protein secreting cells in the normal middle ear mucosa of the guinea pig: An autoradiographic investigation. Ann Otol Rhino1 Laryngol 79:82-94, 1970 19. Ueno K, Lim DJ: Heterogeneity of glycoconjugates in the secretory cells of the chinchilla middle ear and eustachian tubal epithelia: A lectin-gold cytochemical study. J Histochem Cytochem 39:71-80, 1991 20. Marovitz WF, Porubsky ES: The embryological development of the middle ear space: a new concept. Ann Otol Rhino1 Laryngol 80:384-389, 1971 21. Hilding DA, Szachowicz E, Larsen SA: Development of the epithelium of the middle ear: Electron microscopic study of fine structures, including junctional complexes and basal lamina. Am J Otolaryngol 1:97-108,198O 22. Holtfreter J: Mesenchyme and epithelia in inductive and morphogenic processes, in Fleischmager R, Billingham RE (eds): Epithelial-Mesenchymal Interactions. Baltimore, MD, Williams & Wilkins, 1968, pp l-30 23. Wieland F: Are sulphated cell surface glycoproteins involved in the control of cell differentiation? Trends Biochem Sci 7:308-309, 1982 24. Yamada KH: Cell surface interactions with extracellular materials. Am Rev Biochem 52:761-799, 1983
