scanning electron microscopy of the human organ of corti

Journal of the Royal Society of Medicine Volume 76 April 1983 269

Scanning electron microscopy of thehuman organ of Corti'

A Wright FRCSDepartment of Otorhinolaryngology, University of LiverpoolRoyal Liverpool Hospital, Liverpool L69 3BX

Summary: The human cochlea has been preserved from post-mortem autolysis by perfusionwith a fixative shortly after death. Subsequent staining with osmium permitsdissection of this structure from the temporal bone. (Temporal bones were

obtained from eight patients). When prepared for examination in the scanning electronmicroscope, the auditory sensory cells are found to be located in the band-like organ ofCorti which extends the length of the cochlea. The sensory cells have a cluster of stereociliaprojecting from their free upper surface and because of this are called hair cells. The haircells are divided into two separate groups: a single row of inner hair cells, which show littlevariation in their surface appearance along the length of the cochlea, and three or four rowsof outer hair cells whose cilia change in conformation and increase in length along thecochlea.

IntroductionStudying the structure -of the inner ear has always provided a challenge because of itscomplexity and its position deep in the petrous temporal bone. As new techniques havebecome available so different aspects of cochlear anatomy have been described. For most ofthis century a human temporal bone would have been decalcified, embedded and thensectioned - the sections being examined with the light or, more recently, the transmissionelectron microscope. Unfortunately, whole organ sections tend to restrict the appreciation ofthe three-dimensional nature of the cochlea, especially at the higher magnifications availablewith the electron microscope. A three-dimensional appreciation of the cochlea is very helpfulin understanding the physiology of hearing and in assessing the extent of pathologicalchange. During the last twenty years the technique of dissecting the cochlea out of its bonyshell, a method frequently used in the last century, has been revived. The surface of thesensory cell region (the organ of Corti) can now be studied with the light or phase contrastmicroscope and more recently with the scanning electron microscope. This development hasadded a new dimension by virtue of the greatly increased magnification and depth of focusthat is available. Unfortunately the increasing power of the electron microscope makesartefacts more obvious, and the changes due to post-mortem autolysis, which can beeliminated in animal material by rapid fixation, cause problems in the assessment of humantissue. Animal work using the transmission electron microscope has suggested that the delaybetween death and fixation should not exceed one hour since artefacts occur which cannotbe differentiated ,from pathological changes (Kimura et al. 1964).To overcome this problem, so that the scanning electron microscope could be used to

study the surface features of the human organ of Corti, the cochlea has been perfused with afixative within 40 minutes of death. This paper presents the surface features of the normalhuman organ of Corti.

Material and methodsThe ears for this study came from patients who died from head and neck cancer in theRoyal Liverpool Hospital. Before their deaths the patients had normal audiograms as1 Based on paper read to Section of Otology, 7 May 1982. Accepted 19 November 1982

0 1 41-0768/83/040269-10/$O 1.00/0 1983 The Royal Society of Medicine

270 Journal of the Royal Society of Medicine Volume 76 April 1983

judged by the criteria of Robinson & Sutton (1978). Temporal bones were obtained fromeight patients whose ages ranged from 29 to 70 years.

Within 40 minutes of the patient's death a tympanomeatal flap was lifted and the stapesdislocated. The cochlea was perfused with a fixative containing 2% glutaraldehyde and 2.5%formaldehyde in a 1.5 mol/l cacodylate buffer. The fixative was perfused using a right-angled cannula inserted in the round window so that it ascended the scala tympani, passedthrough the helicotreme and then descended via the scala vestibuli, the excess escapingthrough the oval window. The two windows were then sealed with plasticine. If permissionfor post-mortem was obtained, the temporal bones were removed and reperfused with thefixative and left for 24 hours. The tissues of the membranous labyrinth were further fixedand stained with 1% osmium tetroxide perfused through the cochlea. The osmium waswashed out and the temporal bone dissected until only a fine bony shell remained over themembranous labyrinth. This shell was picked away starting at the apex to reveal the spirallamina and scala media. Reissner's membrane, the stria vascularis and the tectorialmembrane were removed to reveal a half turn of the spiral lamina with the organ of Cortiupon it. This half turn was removed and the process repeated until all of the organ of Cortihad been dissected. Each segment was dehydrated, critical point dried, coated with gold andpalladium and examined in a Jeol 35C scanning electron microscope (SEM).

ResultsThe dissected cochlea is shown in Figure 1 and the helicotreme and apical turn of the spirallamina in Figure 2. The organ of Corti with the osmium-stained acoustic nerve fibres of thefully dissected apical segment is shown in Figure 3. When prepared for the SEM, onlysurface features can be seen. The ridge of the organ of Corti and the edge of the spirallimbus (from which the tectorial membrane has been removed) can now be seen (Figure 4).At a higher magnification these features are more clear (Figure 5). Tilting the specimen sothat the end of the organ of Corti comes into view gives an image resembling the moreconventional light microscope picture. The sensory cell bodies, the tunnel of Corti and thespaces of Nuel can be seen, as can the fibrous nature of the basilar membrane (Figure 6).Looking more closely at the surface of the organ of Corti, the sensory cells with clusters

of stereocilia projecting from their free surface can be divided into two distinct groups. Asingle row of inner hair cells (IHC) lies closer to the modiolus than the three or fourirregular rows of outer hair cells (OHC) (Figure 7). When the outer hair cells arephotographed from a different viewpoint the stereocilia projecting from the smoothcuticular plate can be seen to form ranks of increasing length arranged in a V-shaped cluster(Figure 8). These appearances are not, however, constant along the length of the cochlearduct. The longest stereocilia of the outer hair cells at the base of the cochlea are very muchshorter (2.75 ,um) than those at the apex (7 jm). There are fewer stereocilia arising from eachhair cell at the apex, and the configuration of the stereociliary cluster also changes, as can beseen in- Figure 9. The inner hair cells form a more homogeneous population. Theirstereocilia are arranged in straight lines parallel to the long axis of the cochlear duct, andthere is less variation in the length of the stereocilia along the cochlea (4,m base to 7.5 ,umapex).The number of hair cells present in each cochlea varies from subject to subject.

Unfortunately, hair cell counts could not be obtained from the cochleas of the youngest twopatients in this study, but in those whose ages ranged from 54 to 70 there were rarely morethan 3000 inner and 9000 outer hair cells.

DiscussionThe structure and workings of the inner ear have fascinated anatomists and physicians sinceearly times. Before the decline of Alexandrian and Greek science, Aristotle (c. 384-322 Bc)had probably noted the shape of the cochlea although modern translations of his 'HistoriaAnimalium' are ambiguous. Some 450 years later Galen (c. AD 131-201) developed his own

Journal of the Royal Society of Medicine Volwne 76 April 1983 271

Figure 1. The cochlea stained with osmium anddissected so that only a thin bony shell remains aroundthe membranous labyrinth. The oval (o) and roundwindows (r) are marked

Figure 2. Bony labyrinth removed from the apex of thecochlea to reveal the helicotreme and the spiral laminacurling up to its termination. Reissner's membrane isstill intact at this stage of the dissection

Figure 3. Apical segment of the spiral lamina; thefibres of the acoustic nerve are radiating out to theband-like organ of Corti which rests on the nearlytransparent basilar membrane. Length of the organof Corti is 4.75 mm

Figure 4. Segment of the spiral lamina prepared for SEM. The ridge of the organ of Corti can be seen in the upperportion of the micrograph with the clusters of cilia just visible. The tectorial membrane has been removed from thespiral limbus. Organ of Corti in this segment is 2.75 mm long. (Reproduced from Wright 1982, with kindpermission)


Figure 5. Cut end of a segment of the spiral lamina. The spiral limbus at the left of the picture has a ragged edgedue to removal of the tectorial membrane. The ridge-like shape of the organ of Corti is apparent and the cilia of thesensory cells are clearly seen. (Marker 100 microns)

Figure 6 Cut end of the specimen viewed from a more oblique angle. The sensory cell bodies are cylindrical withthe space of Nuel surrounding them. SL, spiral limbus. IHC, inner hair cell. OHC, outer hair cell. HC, Hensen'scells. CC, Claudius's cells. (Marker 100 microns) (Reproduced from Wright 1982. with kind permission)

Figure 7. Direct surface view of the organ of Corti. There is a single row of IHC's although some extra internalIHCs are present, and an irregular number of rows of OHCs, with occasional missing cells being replaced by'phalangeal scars'. (Picture width 200 microns)


... XX.. _: .... ............. .fi. >...

Figure 8. Outer hair cells. A: Three ranks of OHCs (picture width 35 microns). B: Single OHC. The tightly-packedcilia arise from a relatively smooth cuticular plate that is free of microvilli (picture width 7.5 microns)

Figure 9. OHCs from four different portions of thecochlea. The number in each micrograph represents thedistance in mm from the base of the cochlea. Themarker in each case is 1 micron. (Reproduced fromWright 1982, with kind permission)

Figure 10. Vesalius's diagram of the human temporal bone from the 1725 edition of 'De Fabrica'. The acousticnerve D ramifies within the petrous portion of the temporal bone but the labyrinth is not shown, possibly becausethis mastoid bone is rather cellular

... .. ....


ideas about the structure and function of the cochlea in 'De Usu Partium'. He thought thatthe acoustic nerve needed to be protected from the movements of the air which carried thequality of sound. Nature therefore had to strike a balance between protecting the nervewhilst allowing sensitivity in the perception of sound. 'Accordingly therefore she added adense hard bone and drilled through it with oblique coils like a labyrinth, being carefulgradually to dull by intricate deflections the direct force which the cold air would have if thepathway were straight.'Over the next thousand years virtually no new work concerning the ear was produced.

However, in the first half of the fifteenth century the study of anatomy flourished andvarious descriptions of the ear are to be found. Many are presumably copies of earlierworks, as at this time the ossicles had not been described. However, in 1521 Berengario daCarpi, in his massive work 'A Commentary on the Anatomy of Mondinus', produced thefollowing description: 'Within the substance of the aforesaid petrous bone there is in everyear a twisted cavity in which the beating air receives the form of hearing and offers them tothe auditory nerve which is spread out into a panniculus'. This 'twisted cavity' could easilybe interpreted as being a description of the cochlea were it not for a passage in hispublication 'A Short Introduction to Anatomy' that appeared a few years later (Berengarioda Carpi 1523). In this the following description is to be found: 'Within this bone (petrous)is a certain vacuity which is closed by a certain thin, solid panniculus. This panniculusaccording to some, arises from the auditory nerve. In the aforesaid vacuity which theaforesaid panniculus covers in front, the air is placed; this air receives the form of hearing.The air gives these forms to the auditory nerve dilated into the panninculus (which is calledthe meninge of the ear). Adjacent to the panniculus within the vacuity lie two little bones fitto be moved by the air with the nearest motion'. The 'twisted cavity' in the first descriptionseems unlikely to be the inner ear as its description is so similar to that of the 'vacuity' (themiddle ear) which contains the ossicles.Andrea Laguna (1535) paints a much more convincing picture, but does not describe the

ossicles, and it seems unlikely that he could have dissected the inner ear without findingthese. The following description from his work 'Anatomical Procedures' is not dissimilar tothat found in Greek writings: 'Each ear has a tortuous labyrinth made intricate with manyturns and twists'. Nicola Massa (1536) provides instruction on how to observe the ossiclesby removing bone from the floor of the middle cranial fossa, thereby opening up the middleand inner ears. He goes on to say: 'Examine carefully the windings of the inner bone for youwill see a nerve that passes through the substance of the bone to the eardrum'. Massatherefore appears to have been one of the first to have convincingly and clearly made thedistinction between the middle and inner ear.

In 1537, at the age of 23, Vesalius became professor of anatomy at Padua. He did notappear to be so well acquainted with anatomy of the ear as did some of his predecessors,describing only an inner circle, the outer circle being the tympanic membrane (Vesalius1550). In the illustration of the temporal bone taken from the 1725 edition of 'De CorporisHumani Fabrica' (Figure 10) the cochlea is not at all obvious. A few years after Vesalius,Eustachius (1563) depicted the cochlea and semicircular canals that had been described byFallopius in 1561. (Figure 11 is taken from the 1714 edition of Eustachius and clearly showsthese structures.) Duverney in 1683 was able to describe the bony labyrinth in more detail.He depicted the spiral lamina which he describes quite clearly as dividing the spiral duct intoan upper and lower compartment and of having a bony and a membranous portion(Figure 12; Duverney 1730). Valsalva (1717) was able to draw the labyrinth more clearly butadded nothing to the work of Duverney. He may even have caused some confusion since inone of his diagrams the labyrinth is upside down and has only one and a half turns(Figure 13). In the latter half of the eighteenth century many workers added pieces ofinformation, especially Comparetti, Cuvier, Scarpa and Sommering, and this body of workwas brought together by Breschet (1836) who added his own observation from 20 years ofdissection. A membranous labyrinth had been described but was restricted to thesemicircular canals and vestibule. The cochlea was still divided into only two compartments:


Figure 12. Duverney (1730) has provided a section of thetemporal bone and shows clearly the spiral lamina. Otherdiagrams in his work indicate clearly that themembranous spiral lamina extends to the outer wall ofthe cochlea

tIJ. I

Figure 11. The human temporal bone as depicted byEustachius (1714). The coils of the cochlea and thesemicircular canals are clearly shown in this section

Figure 13. Diagram of the bony labyrinth from Valsalva's1717 edition of 'De Aura Humana Tractatus'. It appearsto be printed upside down and there are only one and ahalf coils to the cochlea

Figure 14. Two illustrations from Breschet (1836). A: The entire bony labyrinth is shown, but the membranouslabyrinth depicted in the lower left of the illustration comprises only the semicircular canals, saccule and utricle. B:Mid-modiolar section of the cochlea shows the helicotreme and the spiral lamina but no scala media

a 'Rampe Vestibulaire' and a 'Rampe Tympanique'. The outstanding quality of Breschet'swork is shown in Figure 14.Although the gross anatomy of the cochlea had been carefully described by the first half

of the nineteenth century, further advances had to await the development of compoundmicroscopes free from the more serious lens aberrations. This happened in the half centuryfrom 1840-90. Alphonse Corti (1851) started the development of modern anatomy with hispainstaking description of the cochlea, and although it appears that the membranesubsequently discovered by Reissner had collapsed to cover the tectorial membrane andsensory cell area, Corti nevertheless was able to describe and measure the sensory cells whichsubsequently became called hair cells (Figure 15). Reissner (1854) described the membranethat formed the boundary of a separate compartment - the scala media, which contained the


Figure 15. Membranous spiral lamina as depicted by Corti in 1851. A: Conventional cross-section with the bonyspiral lamina and spiral limbus on the left. The organ of Corti has apparently collapsed beneath the tectorialmembrane, but the differentiation into inner and outer hair cells can be recognized. B: Surface view of the sameregion. The inner and outer hair cells are more clearly seen, as are the heads of the pillar cells

accurtelydescrbed he stuctues ofthennerars figr 16.yCrss-ectionofthcchudn ean.duThas

..... ................d

Duin theorganlf h wetehetryms mirs ofi Cotuiaes shown;tohlea striaperformed withtemporalbonesectvasuoar. one theen latra wallofnmisoecthenmirscala pedi

sTEnsor cellisofCreti'ergmagn,fctoandby1884 retisohadthon,rosuglye investigiatedondtaccuaratel descrnibeqhusrues ofthathcoleainner bear oftanyspeiesincludeting man.er the

Duringthe fairsto haiogher twentiicthicnturyamosto microscopienstudiesatofythuccleas werea

sections are studied and the whole reconstructed. This is a daunting task that few haveattempted (Spoendlin 1979). In the l950s various groups, especially that of Engstrom and

Journal of the Royal Society of Medicine Volwne 76 April 1983 277

colleagues (1966), redeveloped the techniques used in the last century, whereby the cochleawas dissected and the surface examined by light microscopy.The development of the scanning electron microscope (SEM), which enables surface

features to be examined at high magnification with a great depth of field, has extended thevalue of these surface preparations. But when the higher magnifications of the electronmicroscope are used, the artefacts resulting from post-mortem autolysis make the study ofhuman material particularly difficult. Corti (1851) commented in the notes to his article thathe was unfortunately unable to obtain human tissue in a state as fresh as animal material - aproblem that remains to this day. In the present study this difficulty has been overcome byperfusing the cochlea with a fixative shortly after death. The surface preparations of well-fixed tissue can provide information that is difficult to obtain from the more conventionaltemporal bone sections. The general anatomy of the cochlea and organ of Corti is easier tounderstand from scanning electron micrographs and, in addition, the hair cells can becounted directly and their distribution plotted in the form of a cytocochleogram (Wright1981). This avoids many of the problems involved in 'reconstructing' the cochlea from serialsections (Guild 1921). The number of hair cells counted by this method correlates closelywith the figures produced by Bredberg (1968) in his light-microscopic surface study of thehuman cochlea. This technique can therefore be used to plot the distribution of sensory cellloss caused by ageing, drugs, disease or injury.The higher magnification available also means that the stereocilia can be observed more

closely, and abnormalities, such as the formation of giant stereocilia, studied in detail(Wright 1982). The length of the stereocilia can be approximately measured directly frommicrographs. Accurate measurements, however, require the production of stereopairs and arigorous stereometric approach to remove errors caused by the stereocilia being tilted andtherefore appearing shorter than they really are.During preparation of the cochlea the stria vascularis is dissected, and this tissue can be

saved and prepared for both scanning and transmission electron microscopy (Wright 1980).This may be a useful method for investigating this structure, which is the site of action ofsome of the ototoxic drugs and dysfunction of which might be a cause of Meniere's disease.SEM provides -details of surface features, and although modifications of the basic

technique - freeze fracture and elemental X-ray analysis, for example - can be expected toprovide further information about the structure and function of the cochlea, transmission.microscopy is essential to provide details of intracellular structures and of the relationshipsbetween the various cell types found in the cochlea.

Acknowledgments: This work has been supported by a grant from the Merseyside RegionalHealth Authority. I am grateful to Dave Atkins of the Department of Medical Illustrationof the Royal Liverpool Hospital for taking the negatives of the old manuscripts.

ReferencesAristotle. Historia Animalium. Trans. d'Arcy Thompson. Clarendon, Oxford, 1910Berengario da Carpi (1521) A Commentary on the Anatomy of Mondinus. Trans. L R Lind, in 'PreVesalianAnatomy'. The American Philosophical Society, Philadelphia, 1975 (Harold Cohen Library, University ofLiverpool)

Berengario da Carpi (1523) A Short Introduction to Anatomy. Trans. L R Lind. University of Chicago Press, 1959Bredberg G (1968) Acta Otolaryngologica, Suppl. 236; pp 1-135Breschet G (1836) Recherches sur l'Organe de l'Ouie et sur l'Audition. Balliere, Paris (Liverpool Medical

Institution)Corti A (1851) Zeitschrfit fur Wissenschaftliche Zoologie 3, 109-140Duverney D (1730) Tractatus de Organo Auditus, Leyden (Liverpool Medical Institution)Engstrom H, Ades H W & Andersson A (1966) Structural Pattern of the Organ of Corti. Almqvist & Wiksell,

StockholmEustachius B (1714) Tabulae Anatomicae, Rome (Liverpool Medical Institution)Fallopius (1561) Observationes Anatomica, Venetis (Liverpool Medical Institution)Galen. On the Usefulness of the Parts of the Human Body. Trans. M T May. Cornell University Press, New York,

1968Guild S R (1921) Anatomical Record 22, 141-157

278 Journal of the Royal Society of Medicine Volune 76 April 1983

Kimura R S, Schuneckt H F & Sando I (1964) Acta Otolaryngologica 58, 390-408Laguna A (1535) Anatomical Procedure. Trans. L R Lind, in 'PreVesalian Anatomy'. The American Philosophical

Society, Philadelphia, 1975 (Harold Cohen Library, University of Liverpool)Massa N (1536) A Short Introduction to Anatomy. Trans. L R Lind, in 'PreVesalian Anatomy'. The American

Philosophical Society, Philadelphia, 1975Reissner E (1854) Zur Kenntniss der Schnecke im Gehororgan der Saugethiere und des Menschen. Muilers Archiv

F. Anat. Phys. u. Wiss. Med. pp 420-427Retzius G (1884) Das Gehororgan der Wirbelthiere, Vols I and II. Samson & Wallin, Stockholm (Harold Cohen

Library, University of Liverpool)Robinson D N & Sutton G J (1978) A Comparative Analysis of Data on the Relationship of Pure tone Audiometric

Threshold to Age. National Physical Laboratories Report AC84, EnglandSpoendlin H (1979) Journal of Laryngology and Otology 93, 853-877Valsalva A M (1717) De Aure Humana Tractatus, Utrecht (Liverpool Medical Institution)Vesalius A (1550) De Corporis Humani Fabrica. loannes Oporinus, Basilae (Sidney Jones Library, University of

LiverpoolVesalius A (1725) Opera Omnia Anatomica et Chirurgica. Boerhaave et Albini, LeydenWright A (1980) Archives of Otorhinolaryngology 228, 39-44Wright A (1981) Clinical Otolaryngology 6, 237-244Wright A (1982) Clinical Otolaryngology 7, 193-199

scanning electron microscopy of the human organ of corti

Documents