History of the Washington University Inner Ear Protein Database

Isolde Thalmann PhD and Ruediger Thalmann MD

Drs. Ici and Ruedi Thalmann can be contacted by voice phone at 314-362-7505, by fax at 314-362-7568 or by E-Mail at thalmanni@ent.wustl.edu and

 
 
 
 
 

 thalmannr@ent.wustl.edu respectively




An all out effort has been initiated towards establishment of an Inner Ear Proteomics Project, i.e. towards the study of the 'total protein complement of the genome'. Unlike the genome of an organism, which is essentially fixed, the proteome varies with tissue, cell organelles, as well as developmental stage, disease state, or invironmental conditions. This endeavor, while gargantuan, is absolutely  worthwhile and  quite feasible.

The discussion that follows is not concerned with the Proteomics Project per se, but rather describes the approach that was chosen in the past to study inner ear proteins.

The Washington University Inner Ear Protein Database has been founded by Drs. Ici and Ruedi Thalmann and is the natural outgrowth of long-term ongoing  studies at the Department of Otolaryngology at Washington University in collaboration with numerous investigators from around the world. While the Database initially emphasized results obtained by the team at Washington University, results from other laboratories have been solicited and have been added.

For details consult the References of the respective topic (or protein). When available links to MEDLINE are provided.

The objective of the study carried out by the team at Washington Univeristy, and the strategy used:

The main objective of the studies is the comprehensive biochemical characterization of tissues and fluids of the inner ear, and correlation with functional parameters.

The mammalian inner ear  is an exquisitely sensitive instrument, unrivaled in its cellular specialization and differentiation. It consists of the older vestibular system, essential for the maintenance of balance and spatial orientation, and the newer auditory system, that converts sound waves to electrical signals which are sent to the brain. For a quick refresher course on this topic click on the various symbols on the painting on the index.

Understanding of cochlear and vestibular  macro-and micromechanics depends on the physical, bioelectrical, and biochemical properties of constituent tissues, and pervading and surrounding fluids of the inner ear. It is the macromolecules and their interactions that endow a tissue with its specific properties. Functional specialization is achieved through macromolecular specialization. Although cellular metabolism reflects the integrated activity of several thousand gene products, functional diversity is conferred by relatively few.

Pursuing this line of reasoning, we have focused our studies into auditory and vestibular function on:
 


We have chosen this approach rather than the more rapid identification of multiple tissue-specific genes through, for instance, subtractive screening of cDNA libraries. The latter suffers from the shortcomings that plague all fishing expeditions, i.e. low selectivity and variable efficiency. Having identified the putative candidate genes, their biological significance remains to be established. Furthermore, there is no guarantee that the most valuable genes made it into the net.

The core technology for large-scale screening of proteins (the term "Proteomics" has been coined for this) at present is  two-dimensional polyacrylamide gel electrophoresis (2D-PAGE). When combined with stringently controlled sample preparation, and image detection, thousands of proteins can be separated and hundreds identified through database searches. This technique was applied here. For details see  Method
 

Summary of results obtained:


Paradigm for identification of proteins satisfying the criteria described above, distribution and potential function:

1. Organ of Corti proteins OCP1 and OCP2:


Two-dimensional polyacrylamide gel electrophoresis
of guinea pig organ of Corti. Note the prominence
of OCP1 and OCP2

In 1980 we discovered two low molecular mass, strongly acidic proteins which are present at extremely high concentration in the organ of Corti. We termed them OCP1 and OCP2. They are among the most abundant proteins in the organ of Corti, each representing roughly 5% of the total protein. The two proteins were also detected at lower concentration in vestibular epithelia (more than one order of magnitude) concentrations. They are absent, or present in traces only in tissue other than ear.

In 1990 a  partial amino acid sequence was obtained for OCP2. It was used to raise an antibody (see figure below) and to clone the Ocp2 gene encoding OCP2 from a guinea pig organ of Corti cDNA library. OCP1 has resisted efforts at cloning until 1998. Extreme homology of OCP2 has been shown with Skp1, a cell cycle protein identified in S. cerevisiae. OCP1 shows homology with the neural F-box protein, NFB42 (except for amino acids 1-32), respectively.  The human genome includes at least five different genes with high homology to ocp2 and skp1. Even though the two proteins show extreme homology, ocp2 maps to 5q and skp1 to 7 and 12 q.


For 'historical' purposed the first immunocytochemical staining of the organ of Corti using
antibody raised to a synthetic oligopeptide derived from the partial amino acid sequence of OCP2
is shown. This picturewas featured on the cover of the June 10, 1995 issue of thejournalGenomics.
For anatomical details refer to the logo of the index page.


Most recently Richard Killick (unpublished) was able to obtain immunohistochemical double-labeling using antibodies agains OCP1 (left) and OCP2 (right) for all cochlear turns (guinea pig).

Both OCP1 and OCP2 are abundant in the supporting cells, and adjacent nonsensory epithelia including root cells and root processes, and interdental cells. They are not present in hair cells, spiral ligament and stria vascularis. Importantly they were not detected in the developing organ of Corti at birth.

The physiological roles of OCP2 and OCP1 are conjectural at present. Several lines of evidence suggest that OCP2, possibly in conjunction with OCP1 and other proteins, mediates gap junction gating, and provides a mechanism of K+ recirculation. However, a cell cycle role has not been ruled out. It could be speculated on the basis of emerging literature from other areas, it could be speculated that the OCPs may play a role in keeping the cells of the organ of Corti arrested thereby preventing regeneration.

2. Oncomodulin, a novel calcium-binding protein in the outer hair cells:

In 1995 we detected a novel calcium-binding protein (pI  3.1, Mr 15,000) in the organ of Corti, which was absent in other inner ear structures and other mammalian tissue. Determination of its partial amino acid sequence and subsequent cloning of the corresponding cDNA from a guinea pig organ of Corti cDNA library identified the protein as oncomodulin a parvalbumin of the beta lineage. Demonstration of the presence of oncomodulin in the organ of Corti is the first documentation of this substance in a postnatal mammalian tissue.


Light microscopic immunostaining of the upper basal turn of the gerbil cochlea.
Intense immunoreactivity is confined exclusively to outer hair cells.

Iimmunocytochemical data and quantitative biochemical studies indicate that oncomodulin is strongly expressed in the outer hair cells of the organ of Corti; this appears to be the only physiological site of expression in the mature animal. Recruitment of an unusual calcium-sensing protein specifically for the outer hair cells suggests an obligatory role in the physiology of this highly specialized cell.

3. Apo J and apo D, two proteins concentrated in endolymph and perilymph. For a review of the anatomy and physiology of inner ear fluids, the reader is referred to an excellent web page by Dr. Alec Salt.

a. Protein profiles of guinea pig inner ear fluids.

The primary purpose of this study at the outset was to determine whether the distribution of proteins in perilymph and endolymph is of guinea pig revealed  in 1992 that most proteins were found to be present in perilymph at levels in basic accord with the total protein gradient with respect to plasma; however, two proteins, were present in perilymph in almost one order of magnitude higher concentrations.

These findings indicated that ultrafiltration cannot be the sole mechanism of perilymph production. Endolymph proteins are uniformly five to eight times lower than perilymph levels, essentially following the total protein concentration gradient between the two compartments. This supports the view that endolymph is derived from perilymph rather than directly from blood.

 b. Protein profile of human perilymph.

More than 100 proteins were separated by 2D-PAGE, and some 30 identified and several quantitated in human perilymph.

The majority of proteins were found to be present in perilymph at levels reflecting the total protein gradient between perilymph and plasma. Two salient differences have been found: (1) beta-2 transferrin was detected in perilymph in concentrations similar to CSF, and (2) two high density lipoprotein-associated apolipoproteins, apo D and apo J were found to be present at concentrations 1-2 orders of magnitude higher than in plasma, in terms of total protein.

c. Application of human protein profile of inner ear fluids to the pathological ear (with emphasis on the diagnosis of perilymph fistula).

The technique has been evaluated to identify disease-specific protein patterns in human perilymph, in analogy to other body fluids, such as plasma and CSF. With the possible exception of presumed perilymphatic fistula, it is not envisioned that the analysis of perilymph will be used for diagnostic purposes but rather as a tool for the elucidation of the mechanisms underlying inner ear disease, whether localized or as part of systemic alteration.

The diagnosis of perilymph fistula remains a most controversial and challenging topic in otology. Perilymph fistula is an abnormal communication between the middle ear and the perilymphatic space. The major problem with the diagnosis and treatment of perilymph fistula that to date no reliable pre- and intraoperative diagnostic tests exist.  Diagnosis is based upon history, and physical examination. However, intraoperative criteria are based upon subjective observation of fluid accumlation in the oval or round window niche; this technique tends to lead to misdiagnosis. No consensus as to management exists as a restul of this.

A search for a specific marker to identify the fluid encountered in the middle ear upon exploratory surgery in patients with suspected perilymph fistula ensued. The establishment of protein maps of inner ear fluids and the revelation of a specific protein and of several  examples of proteins showing striking differences in concentration when compared with other body fluids seemed most promising.

2D-electrophoretic separation of proteins from perilymph of a 6.5 month-old infant,
collected 2.5 hrs post mortem. Proteins significantly different between plasma and perilymph,
and therefore relevant to perilymph fistula, are boxed.

The most promising candidates are:

aa. Beta-2 transferrin

We confirmed earlier findings that this protein was present in perilymph and CSF, but absent in normal plasma. In our experience, beta-2 transferrin was present in all control perilymph samples. While these samples are relatively large and uncontaminated, typical clinical samples of perilymph fistula fluid are small (usually 0.5 to 1 ul, or less) and likely contaminated to various degrees with red blood cells, plasma, middle ear transudate, or local anesthetic.  Even though beta-2 transferrin is not present in these contaminating fluids, with rising plasma contamination, beta-2 transferring becomes increasingly masked by transferrin. This, and other problems which arose  make this method meaningless.

bb.  Apolipoproteins D and J

While these two proteins are prominent on protein maps of perilymph and CSF and are deceivingly absent in maps of plasma when samples with similar total protein amounts are loaded onto the gel. We determined that these proteins are 70-100 fold higher in perilymph than plasma. However, the same problem discussed in the previous paragraph under beta-2 transferrin unfortunately applies here. Because the total protein content of perilymph versus plasma is 1:35, a relatively small volume of plasma contaminations increases the signal for the two proteins inordinately. Only judicious analysis of a number of reference proteins, in addition to apo J and apo J, will distinguish the type of fluids. Unfortunately, this is not feasible for the average clinical situation.

In summary, promising substances are available that can serve as reliable markers for perilymph fistula only if relatively clean samples are provided by the surgeon. Increasing the sensitivity of the chemical assay compound the problems rather than solve them.

4.  Otoconin-90, the major protein of otoconia

The gravity receptor system of the inner ear is one of the phylogenetically oldest sensory systems. The gravity receptor cells do not differ significantly from other mechanoreceptor hair cells of the acoustico-lateralis system which is conserved from the lateral line organ of the fish to the vestibular and auditory organs of higher vertebrates. The special adaptation to sense gravity is provided by the sensory superstructure, the otoconial complex that contains dense mineral particles embedded in a gelatinous membrane overlying the hair cells. Otoconia consist of proteins (otoconins) and a mosaic of calcium carbonate crystals which in mammals are present in the form of calcite.

Little is known about the mechanisms which regulate otoconial biosynthesis. In an effort to elucidate these mechanisms, we separated the mammalian otoconial proteins  as well as the underlying sensory epithelium by 2D-PAGE . We then partially sequenced and cloned the major protein component of the  otoconia, otoconin 90 (Figure C below - arrow). The derived amino acid sequence indicates two domains of homology to secretory phospholipase A2, which lack some residues considered important in calcium-binding and -catalytic action. High expression of otoconin 90 mRNA was found by RT-PCR in the  mouse otocyst, and localized by immunocytochemistry to the epithelial layer of the developing inner ear. Otoconin 90 is heavily glycosylated, and the extremely low pI suggest incorporation of sulfated GAGs.
 
 

We propose that a key event in the formation of otoconia requires an interaction between OC-90 and the globular substance. In this hypothetical model, the organized layer of acidic phospholipids of the globular substance leads to a high localized concentration of calcium. On binding of OC-90 to the lipid layer, the concomitant conformational change may facilitate stabilization of calcium at binding sites within the PLA domain of OC-90, with subsequent incorporation into the growing calcite lattice. The glycosaminoglycan  portion of OC-90 may act in the controlled inhibition of crystal growth. This model also explains the exclusive localization of otoconia to the macular superstructure, even though Oc90 is expressed throughout the entire developing inner ear.

5.  Proteins in endolymph of endolymphatic sac (preliminary data).

There is strong evidence that the endolymphatic sac is the primary site of volume regulation of endolymh [Salt AN: Fluid homeostasis in the inner ear. In: Meniere's Disease, Ed. JP Harris, Kugler, The Hague, 1999, pp. 93-101; see also Dr. Salt's web page], and that glycosylated macromolecules in the endolymhatic sac are critically involved in the process [Rask-Andersen H, Salt AN, DeMott JE, Bagger-Sjoback D: Morphological changes of the endolymphatic sac induced by microinjection of artificial endolymph into the cochlea. Hear Res, in press]. We therefore set out to identify and characterize the participating macromolecules under normal conditions and after experimental manipulations, as a key for the elucidation of the underlying biochemical mechanisms of Meniere's disease.

Two-dimensional electrophoretic pattern of contents of endolymphatic sac of guinea pig. (B LEFT) Untreated); (A RIGHT) Treated with deglycosylating enzymes (N-glycosidase F and neuraminidase). Small arrows indicate major proteins that have changed in molecular weight and/or pI.  The samples were obtained by dissection of  five freeze-dried endolymphatic sacs  (equivalent to approximately 450 nanoliter of sac endolymph).

Preliminary Results:

Return to Index Page