(Received for publication, November 11, 1996, and in revised form, May 22, 1997)
From the Division of Laboratory Medicine, Department
of Pathology and Medicine, Washington University Medical School,
St. Louis, Missouri 63110 and the ¶ Molecular Cytogenetic
Section, Laboratory of Experimental Carcinogenesis, NCI, National
Institutes of Health, Bethesda, Maryland 20892
Ninjurin is a novel protein that is up-regulated after nerve injury both in dorsal root ganglion (DRG) neurons and in Schwann cells. We previously reported that ninjurin demonstrates properties of a homophilic adhesion molecule and promotes neurite outgrowth from primary cultured DRG neurons. We have now found that ninjurin is widely expressed in both adult and embryonic tissues, primarily in those of epithelial origin. Aggregation assays were used to demonstrate that ninjurin-mediated adhesion requires divalent cations and is an energy-dependent process. The critical domain for ninjurin-mediated homophilic adhesion was localized to an 11-residue region (between Pro26 and Asn37) by mutagenesis and by employing synthetic oligopeptides as competitive inhibitors of ninjurin-mediated adhesion. Of particular importance are the Trp residue at position 29 and the 3 arginines in the region. Furthermore, we show that the peptide which inhibits aggregation of Jurkat cells expressing ninjurin is also capable of blocking the ability of ninjurin to promote neurite extension from DRG neurons. Using FISH analysis, the ninjurin gene was localized to human chromosome 9q22. Several genetic diseases of unknown etiology have been mapped to this region, including hereditary sensory neuropathy type 1, self-healing squamous epithelioma, split-hand/foot deformity type 1, and familial dilated cardiomyopathy.
Cell surface adhesion proteins play an important role in embryonic development, in organogenesis, and in tissue regeneration after injury (1-3). In the nervous system, adhesion molecules function as receptors for axonal guidance cues and are important for the development of neuronal connections (4). In the immune system, cell surface molecules mediate signaling mechanisms that are critical for specificity in lymphocyte homing (5). In addition, normal cell-cell interactions are often aberrant in tumors, and there is increasing evidence that impaired cellular interaction due to loss of adhesion proteins plays a role in tumorigenesis, as well as tumor invasion and metastasis (6, 7). Indeed, a gene encoding an adhesion molecule similar to neural cell adhesion molecule (NCAM) and N-cadherin has been identified as a tumor suppressor gene (8, 9).
In vertebrates, a number of cell surface glycoproteins have been identified as adhesion molecules, including integrins, cadherins, and those containing an immunoglobulin (Ig)1-like motif (3, 10, 11). The sequence motifs that mediate the adhesive interactions of these molecules have been identified for only a few of these proteins. One of the most well characterized sequence motifs of this type is the tripeptide Arg-Gly-Asp (RGD), which was identified as the sequence within fibronectin that mediates cell attachment (12). Many integrins recognize this RGD motif within their respective binding partners, and these interactions then mediate either cell-substratum or cell-cell interactions. Members of the cadherin family contain multiple copies of the Asp-Arg-Glu (DRE) and Asp-Xaa-Asn-Asp-Asn (DXNDN) sequences (13). Structural analysis of cadherin indicates that these motifs may be situated such that they can form a zipper-like structure that may be critical for cell adhesion (3). Shared sequence motifs for members of the Ig-superfamily of adhesion molecules are less well understood although it has been proposed that a decapeptide sequence (KYSFNYDGSE) in the third Ig-like domain of NCAM is responsible for its homophilic binding interactions (14).
We have identified a cell membrane-associated protein, ninjurin, that possesses homophilic adhesion properties but has no significant homology with other proteins (15). Ninjurin is widely expressed in embryonic and adult rats, predominantly in epithelial cells. To define the mechanism of ninjurin-mediated adhesion, we focused on identifying sequences necessary for the binding interaction. Using a cell aggregation inhibition assay and synthetic peptides corresponding to the predicted ninjurin extracellular domain, we localized the binding region to residues 26 through 37. Based on these results, we tested ninjurin molecules with mutations in this region for their adhesive properties and showed that these ninjurin mutants were incapable of mediating cell aggregation. Furthermore, to demonstrate that ninjurin function is associated with its adhesive properties, we demonstrate that ninjurin-stimulated neurite outgrowth from dorsal root ganglia neurons is reversed by administration of peptides that block ninjurin-mediated cell aggregation.
Total RNA was prepared from adult rat tissues and
samples (10 µg) were electrophoresed on 1% agarose-formaldehyde gels
and blotted onto nylon membranes as described previously (16).
Membranes were probed with a 32P-labeled fragment of
ninjurin cDNA. For in situ hybridization analysis of
embryos, timed pregnant rats of the indicated gestational age were
sacrificed, and the embryos were immediately frozen. In situ
hybridization was performed using 33P-labeled antisense or
sense RNA probes transcribed from the ninjurin cDNA (nt 518-1026)
on fresh frozen tissue samples as described previously (17). For
immunohistochemical analysis of adult rat tissues, male Sprague-Dawley
rats (200-300 g) were anesthetized and perfused with 4%
paraformaldehyde in phosphate-buffered saline (PBS). The analysis was
performed on 15 µm cryostat sections using standard methods. The
purified anti-ninjurin antiserum was used at a 1:2500 dilution, and
specific staining was detected with indocarbocyanine (Cy3)-conjugated
anti-rabbit IgG (Jackson Immunological Laboratories). All the tissue
samples were stored at 70 °C. For immunofluorescent staining of
live cultured cells, Chinese hamster ovary (CHO) cells expressing
ninjurin were washed 3 times with PBS containing 0.5% Triton X-100.
The cells were then incubated with a mixture of rabbit anti-ninjurin
antiserum and a mouse anti-
-actin monoclonal antibody (1:500
dilution, Sigma) for 30 min at 4 °C and then fixed with 4%
paraformaldehyde in PBS. Immunoreactivity was visualized with
either Cy3-conjugated anti-rabbit IgG (ninjurin) or
fluorescein-conjugated anti-mouse IgG (
-actin). Immunoblotting using
the anti-ninjurin antibodies was performed as described previously
(18).
A ninjurin genomic probe was labeled with digoxigenin (Random Primed DNA Labeling Kit, Boehringer-Mannheim). The labeled probe was used for in situ hybridization of human chromosomes derived from methotrexate-synchronized normal peripheral lymphocyte cultures. The conditions of hybridization, the detection of hybridization signals, digital-image acquisition, processing and analysis as well as the procedure for direct visualization of fluorescent signals to banded chromosomes were performed as described previously (19, 20).
Peptide PreparationPeptides were synthesized using an
Applied Biosystems peptide synthesizer. Lyophilized crude peptides were
purified by reverse phase-high performance liquid chromatography using
an elution gradient of 0-60% acetonitrile with 0.1% trifluoroacetic
acid in water. The purity and composition of the purified peptides were
verified by mass spectrometry. Purified peptides were dissolved in
distilled water and stored at 70 °C.
The ninjurin expression vector
(pCMV-ninjurin) has been previously described (15). Expression vectors
containing ninjurin mutants pCMV-ninjurin(W29A), which replaces the Trp
with Ala at position 29, and pCMV-ninjurin(R32,34N), which replaces the
Args at positions 32 and 34 with Asn, were prepared by introducing the
indicated mutations into synthetic oligonucleotides
(5-GCTCTCCGCGGCGCTCTTCTTGTTGGCGTAATGGTTTACATTGATGGGCCGGTTCCTCAAACCCGCGCGGGGTGG for W29A and 5
-GCTCTCCGCGGCGCTCTTCTTG
TTGGCGTAATGGTTTACATTGATGGGGTTGTTGTTCAAACCCCAGCGGGGTGG for R32,34N).
PCR reactions using the pCMV-ninjurin plasmid as template were
performed in which the mutagenic primer was paired with a primer
corresponding to nt 840-860 of the pCMV vector. The amplified products
were digested with BglII and SacII, and the
fragments containing the mutations were ligated into pCMV-ninjurin digested with BglII and SacII so as to replace
the corresponding wild-type fragment and create pCMV-ninjurin(W29A) and
pCMV-ninjurin(R32,34N).
CHO and Jurkat cells were grown as described previously (15). The expression vectors pCMV-ninjurin, pCMV-ninjurin(W29A) and pCMV-ninjurin(R32,34N) were transfected into CHO cells via calcium phosphate precipitation or into Jurkat cells via electroporation. Stable transfectants were selected by growing in medium containing G418 (0.4 mg/ml).
Primary cultures of neurons from dorsal root ganglia (DRG) were prepared from E17 rat embryos and dissociated neurons were grown on confluent monolayers of either native CHO cells or CHO cells expressing ninjurin in 24-well plates. Cultures were grown in a medium that consisted of 90% Eagle's minimal essential medium (Life Technologies, Inc.), 10% FCS, 20 µg/ml each fluorodeoxyuridine and uridine, and 50 ng/ml mouse nerve growth factor (gift from E. Johnson, Washington University). Synthetic peptides were added at a concentration of 0.16 mM to wells where indicated. Six h after plating, the cells were fixed in 4% paraformaldehyde in PBS and immunohistochemistry was performed using anti-neurofilament H antibody. The length of the neurites extended by the DRG neurons was measured as described previously (21).
Cell Aggregation AssaysAggregation assays were performed in 96-well microtiter plates as described previously (22) using Jurkat cells stably transfected with ninjurin expression vectors (either wild type or mutant). Results of the assays were quantitated after 1 h unless otherwise indicated. Reagents and peptides were added to the cell suspension prior to performing the assays at the indicated concentrations. The requirement for divalent cations was tested in Hank's balanced salt solution without calcium and magnesium containing 10% dialyzed fetal bovine serum. Cell quantitation was performed as described previously (22). Briefly, photographs of the cells were taken at the indicated time points, and the total number of cells as well as the number of cells incorporated into aggregates were counted and a ratio was determined. At least 4 assays were performed for each set of experimental conditions. Statistical analysis was performed by SigmaPlot (version 3.0).
For aggregation experiments performed using a mixture of cells, cells expressing wild-type ninjurin were stained red with 1 mM CMTMR, and cells expressing mutant ninjurin were stained green with 1 mM CMFTA as described in the manufacturer protocol (Molecular Probes). The cells were resuspended to 1 × 106 cells/ml, and 3 ml of the mixed cell suspension was allowed to form aggregates in 6-well culture plates. The composition of the aggregates was monitored by fluorescence microscopy.
In previous work, we showed that ninjurin is
expressed in the nervous system, particularly after nerve injury (15).
To further examine the expression pattern of ninjurin in the adult rat,
we performed RNA blot analysis on samples isolated from a variety of
tissues. The highest levels of ninjurin mRNA were found in the
liver. In addition, thymus, heart, adrenal gland and spleen also had
significant levels of ninjurin transcripts, whereas brain and DRG had
low levels of expression (Fig. 1). To
examine ninjurin expression during development, in situ
hybridization was performed on rat embryos sacrificed at embryonic days
17 and 19. Ninjurin expression is observed in a variety of embryonic
tissues, with the most abundant expression observed in tissues where it
is highly expressed in the adult (e.g. liver, adrenals, and
spleen) (Fig. 1). Ninjurin was also detected in the vertebra and limbs,
where its expression increased with increasing embryonic age. The
signal was detected primarily in regions of active ossification, such as the terminal areas of the limb and vertebral bones. The most intense
signal was observed over layers of dividing chondrocytes, rather than
over regions that were already ossified. Ninjurin was also highly
expressed in the skin, where it was primarily detected in the
epithelium. In the central nervous system, ninjurin expression was very
low throughout the embryonic period examined. Tissues hybridized with a
ninjurin sense probe or tissues treated with ribonuclease prior to
hybridization showed no appreciable signal (data not shown).
Ninjurin Is Expressed Predominantly in Epithelial Cells
To
further characterize the ninjurin protein, we prepared a polyclonal
antiserum against a peptide derived from the N terminus (amino acids
1-15). Anti-ninjurin antibodies were then purified by immunoaffinity
chromatography. These antibodies recognized a 22-kDa protein that was
present in rat liver and in CHO cells stably transfected with a
ninjurin expression vector, but not in native CHO cells (Fig.
2).
After verifying their specificity for ninjurin, these antibodies were
used to examine cell type-specific ninjurin expression via
immunohistochemistry. This analysis confirmed that ninjurin is
expressed in a wide variety of tissues in the adult rat, including the
liver, kidney, thymus, uterus, adrenal gland, retina, and dorsal root
ganglia (Fig. 3). In the dorsal root
ganglia, ninjurin was not detected in neurons but rather was expressed
in the satellite cells that ensheath the neuronal cell bodies. In the
adrenal gland, ninjurin is expressed throughout the cortex and appears
to be on the surface of the cortical cells. Ninjurin immunoreactivity was also observed on the surface of the hepatocytes of the liver. In
the kidney, ninjurin was detected in the podocytes and/or mesangial cells of the glomerulus, but other renal cell types were negative. In
the thymus, scattered cells stained intensely. These
ninjurin-expressing cells could represent a specific subset of
thymocytes or the thymic epithelial cells, which provide the proper
microenvironment for lymphocyte maturation. Flat cells located on the
surface of the thymic cortex and in the area adjacent to the blood
vessels also expressed ninjurin. Staining for ninjurin was also
observed in the uterus, where ninjurin-positive cells were found in the
myometrium. These cells were distinct from smooth muscle cells, and
their distribution and morphology was consistent with fibroblasts of the associated connective tissues. In the retina, ninjurin was detected
in the Muller cells near the apical ends, which project into the
interphotoreceptor matrix. The brush-like staining pattern strongly
indicates that the intense immunoreactivity is restricted to the apical
microvilli of the Muller cells. Taken together, these results
demonstrate that ninjurin immunoreactive cells are found in a wide
variety of tissues.
The pattern of staining we observed suggests that ninjurin is
associated with the cell surface, consistent with its role in mediating
cell adhesion and with the cell surface labeling experiments previously
reported (15). To provide further confirmation of its surface
localization, immunostaining of live cells was performed. CHO cells
stably expressing ninjurin were incubated with anti-ninjurin antibodies
at 4 °C prior to fixation, conditions during which antibodies are
excluded from penetrating the cell. The ninjurin immunoreactivity was
intense and appeared to be restricted to the cell surface (Fig.
4). When antibodies to the intracellular protein -actin were used to stain the cell under similar conditions, no immunoreactivity was observed. These results confirm that the antibodies were unable to enter the cell under these staining conditions and clearly indicate that the N-terminal region of the
ninjurin molecule (to which the antibodies were raised) is located
extracellularly.
Chromosomal Localization of the Human Ninjurin Gene
To obtain
further clues regarding the function of ninjurin and its possible
association with human disease, the chromosomal localization of the
human ninjurin gene was determined. In two experiments with lymphocytes
from different individuals, 80% of the chromosome spreads had specific
fluorescent signals at identical sites on both chromatids of the long
arm of chromosome 9. From a total of 190 metaphases examined, 175 had
fluorescent signals on both chromosomes 9 (Fig.
5A). The location of the
fluorescent signal was determined directly in 50 metaphases with
4,6-diamidino-2-phenylindole-enhanced G-like banding at region
9q.22.1-q22.3 where we assign the locus of the ninjurin gene (Fig.
5B). Several diseases of unknown pathogenesis have been
linked to this region, including hereditary sensory neuropathy type 1 (23), self-healing squamous epithelioma (24), split-hand/foot deformity
type 1 (25), and familial dilated cardiomyopathy (26).
Properties of Ninjurin-mediated Adhesion
We had previously
used a cell aggregation assay to demonstrate that Jurkat T-cell
leukemia cells expressing ninjurin showed increased adhesiveness
compared with native Jurkat cells (15). Additional studies demonstrated
that ninjurin adhesion was mediated via homophilic interactions.
Inspection of the ninjurin sequence did not reveal any motifs present
in other adhesion molecules that might mediate homophilic adhesion. To
characterize the adhesive properties of ninjurin, cell aggregation
assays using ninjurin-expressing Jurkat cells were performed under
various conditions (Fig. 6). Ninjurin-mediated adhesion was completely inhibited by prior addition of 20 µM cytochalasin B which depolymerizes actin
filaments, as well as by 0.02% sodium azide, an inhibitor of oxidative
phosphorylation. Aggregation was also inhibited when the assays were
conducted at 4 °C. These results demonstrate that ninjurin-mediated
adhesion requires a functional cytoskeleton and is an active,
energy-dependent process. We also examined the pH and
divalent cation requirements of the interaction. Minimal aggregation of
ninjurin-expressing Jurkat cells was observed when assays were
performed in Ca2+- and Mg2+-free medium, or in
normal medium containing greater than 5 mM EDTA.
Aggregation could be restored by the addition of either CaCl2 or MgCl2, indicating that
ninjurin-mediated adhesion requires divalent cations.
Ninjurin-expressing Jurkat cells showed the same degree of aggregation
between pH 7 and 11, whereas aggregation was inhibited at a pH less
than 6.
Characterization of Ninjurin Adhesion Domain
Analysis of the
ninjurin sequence predicted that it contains two transmembrane domains
(between residues 72 and 100 and between 118 and 139) (15). This
analysis further suggested that the amino terminus of the molecule is
likely to be located extracellularly. To directly investigate the
hypotheses that the N-terminal hydrophilic region is located
extracellularly and is responsible for homophilic adhesion, we tested
whether peptides from this region could inhibit ninjurin-mediated
aggregation (27) (Fig. 7). Partially
overlapping synthetic peptides whose sequences are derived from the
predicted ninjurin extracellular domain were tested for their ability
to inhibit ninjurin-mediated adhesion (Fig.
8A). Peptides 1 and 2 inhibited aggregation in a dose-dependent manner.
Aggregation was completely abolished by peptide 2 at concentrations
above 0.33 mM, whereas peptide 1 inhibited aggregation less
efficiently. Peptide 4 had no effect on aggregation, and we were
unable to test peptide 3 due to its very poor solubility. These results indicated that the ninjurin N terminus was indeed located
extracellularly, and that residues corresponding to peptide 2 were
critical for binding.
To further delimit the site of interaction, we tested four additional peptides (peptides P5-P8; Fig. 7) for their ability to inhibit ninjurin-mediated aggregation. These results demonstrated that a peptide (P6), corresponding to ninjurin residues 26-37, showed inhibitory activity comparable with peptide 2 (Fig. 8B, data not shown for P5, P7, and P8). Aggregation was inhibited at P6 concentrations greater than 0.02 mM and was completely abolished at concentrations above 0.4 mM. To identify the amino acids most critical for ninjurin-mediated adhesion, we synthesized a series of peptides (M1-M7) extending from residue 26 to 37, each containing a single amino acid mutation (Fig. 7). The ability of each of these peptides to inhibit ninjurin-mediated adhesion in a dose-dependent manner was examined (Fig. 8B). We found that mutation of the Trp or any of the Arg residues resulted in a loss of the ability to inhibit aggregation, implying that these 4 residues play an important role in ninjurin-ninjurin interactions. Mutations made to the non-charged residues including Gly30 and Leu31 did not alter the ability to inhibit aggregation, indicating that these residues are not critical to the interaction. One of the mutations (Asn33 to Leu) resulted in a peptide with greater inhibitory activity than wild type, suggesting that it interacts with ninjurin very strongly.
Ninjurin Mutants Fail to Support AggregationAggregation
inhibition experiments using synthetic peptides revealed that residues
26-37 (the ninjurin adhesion motif) are responsible for
ninjurin-mediated homophilic cellular adhesion. To confirm the
importance of this motif, two Jurkat cell lines expressing ninjurin
mutants were generated. One was designed to express ninjurin(W29A), in
which Trp29 was mutated to Ala; and the other expressed
ninjurin(R32,34N), in which Arg32 and Arg34
were both mutated to Asn. Protein blot and immunohistochemical analyses
of these cell lines confirmed that the expression levels for wild
type and mutant ninjurin molecules were comparable and that the
ninjurin mutants were expressed on the cell surface (Fig. 9). Aggregation assays with these
ninjurin transfected cell lines revealed that Jurkat cells stably
expressing ninjurin(W29A) showed no increase in aggregation over native
Jurkat cells, whereas cells expressing ninjurin(R32,34N)
showed greater aggregation than native Jurkat cells but significantly
less than Jurkat cells expressing wild-type ninjurin (Fig.
10), a result consistent with that of the peptide competition assay. To determine whether the mutated ninjurin molecules could interact with wild-type ninjurin, we performed
aggregation assays on mixtures of wild type and mutant ninjurin
molecules, which were differentially stained by fluorescent dyes. When
equal numbers of cells expressing wild-type ninjurin (colored
red) and ninjurin(W29A) (colored green) were
mixed together, aggregates contained predominantly wild-type
ninjurin-expressing cells. Aggregates containing both types of cells
were not observed, indicating that the mutant molecules are incapable
of interacting with wild-type ninjurin. A similar experiment conducted
with cells expressing wild-type ninjurin (colored red) and
ninjurin(R32,34N) (colored green) resulted in an increased
number of aggregates. Some of the aggregates showed a mosaic pattern,
indicating that this ninjurin mutant could interact with wild-type
ninjurin. Taken together, these results are consistent with those
obtained using the peptide competition assays and indicate that
residues 26 to 37 are critical for ninjurin-mediated adhesion.
Ninjurin-stimulated Neurite Outgrowth Is Mediated through the Adhesion Motif
Using ninjurin-expressing CHO cells, we have
previously shown that ninjurin promotes the extension of neurites from
primary cultured DRG neurons. To determine whether this phenomenon
occurs via adhesive properties of ninjurin, we tested the ability of peptides that blocked ninjurin aggregation to inhibit neurite outgrowth. DRG neurons from E17 rat embryos were dissociated and seeded
onto confluent monolayers of either native or ninjurin-expressing CHO
cells. The neuronal cultures were treated with either P6 (wild type) or
M2 (W29A) peptide at a concentration of 0.16 mM, and neurite extension was monitored (Fig.
11). As previously documented, neurites
from neurons grown on CHO cells expressing ninjurin showed increased
neurite outgrowth (569 ± 134 µm) compared with those grown on
native CHO cells (297 ± 104 µm) (15). However, the promotion of
neurite outgrowth by ninjurin was inhibited (317 ± 83 µm) when
the cultures were treated with the peptide of native sequence (P6),
whereas the peptide (M2) containing the W29A mutation had no
significant inhibitory effect on neurite extension by DRG neurons
(638 ± 209 µm) (Fig. 11). Neither of the peptides affected the
neurite extension when neurons were cultured on native CHO cells (data
not shown). These results indicate that the residues identified as
important for adhesion using the Jurkat cell aggregation assay are also
required for ninjurin-stimulated neurite outgrowth, suggesting that the
adhesive properties of ninjurin play an important functional role in
this process.
Ninjurin was first identified as a molecule that is up-regulated in Schwann cells and neurons after peripheral nerve injury. Subsequent analysis of ninjurin function revealed that it is a cell surface molecule that promotes cell aggregation and stimulates neurite outgrowth, suggesting that it may play an important role in nerve regeneration (15). The mechanism by which ninjurin promotes aggregation has been unclear, since it does not contain sequence motifs found in other adhesion molecules. Mutation analysis has enabled us to identify a region in ninjurin that functions as a homophilic adhesion motif that bears no resemblance to previously identified adhesion motifs. Using aggregation inhibition assays and synthetic peptides containing mutated residues within this region, we found that a Trp and three Arg residues within this 10 amino acid region are important for ninjurin-mediated recognition. To verify the critical role of this adhesion motif in ninjurin-mediated functions, peptides that inhibited ninjurin-mediated cell aggregation were used to show that ninjurin-stimulated neurite outgrowth is dependent on its adhesive properties.
The ninjurin adhesion motif (residues 26-37) contains a tryptophan and a cluster of arginine residues. None of the previously reported homophilic adhesion molecules contain this peptide motif or a combination of tryptophan and arginines as functionally relevant residues (28). Competition experiments with peptides derived from this region demonstrated that replacement of tryptophan by alanine dramatically decreased the ability to inhibit aggregation, suggesting that this tryptophan residue plays a major role in ninjurin-ninjurin interactions. The importance of the tryptophan residue was also confirmed by mixed aggregation assays using Jurkat cells expressing ninjurin(W29A), which demonstrated that these cells do not adhere to themselves or to cells expressing wild-type ninjurin. This tryptophan residue may be directly involved in the physical interaction between ninjurin molecules, or it may be necessary for the overall structure of the domain. Replacement of the arginines showed significant, but less dramatic effects on ninjurin adhesion. Interestingly, even though ninjurin binding is dependent on divalent cations, the ninjurin adhesion motif does not contain acidic residues commonly associated with cation binding motifs. Overall, ninjurin contains 13 Glu and Asp residues, with 6 of these residues located within the amino-terminal 23 residues. Perhaps this acidic region is related to the cation dependence of ninjurin and plays a role in the formation of the functional ninjurin adhesion domain.
The demonstration that the inhibitory peptide we identified via cell aggregation assays was able to reverse the ninjurin-stimulated neurite outgrowth from neurons indicates that the biological function of ninjurin is related to its adhesive properties. We also observed that the basal adhesion observed for wild-type Jurkat cells was abolished in the presence of peptides containing the adhesion motif of ninjurin (data not shown). This suggests that other molecule(s) expressed by Jurkat cells possibly share this motif. If other molecules with the ninjurin-like adhesion motif exist, then the possibility that ninjurin may participate in heterophilic as well as homophilic interactions must be considered. Clearly, the ability of ninjurin to participate in heterophilic binding would greatly extend the number of interactions, and potentially the number of functions, in which ninjurin is involved.
In addition to its expression by Schwann cells and neurons after nerve injury (15), we found that ninjurin is present in a wide variety of tissues (e.g. thymus, kidney, liver, adrenal gland), both during development and in adulthood. Ninjurin is predominantly expressed in epithelial cells, suggesting that it is important in the normal development and/or function of a number of tissues, in addition to its role in promoting nerve regeneration. For instance, its presence in the thymus may indicate a role in thymocyte development as adhesion molecules on thymic epithelial cells are thought to be important in this process (29). Ninjurin may also be important in the formation of cellular polarity as it is located in a restricted region in Muller cells, the highly polarized cells of the retina. Within this region, ninjurin may be present in a specific intercellular adhesion apparatus, or it may be interacting with specific cytoskeletal components.
The mapping of the ninjurin gene to human chromosome 9q22 is intriguing as several diseases of unknown pathogenesis have been linked to this region and because mutations in another neural cell adhesion molecule, L1, have been associated with a number of severe genetic disorders, including X-linked hydrocephalus, MASA syndrome (Mental retardation, Aphasia, Shuffling gait, and Adducted thumbs), and spastic paraplegia type 1 (30, 31). The role of ninjurin in promoting nerve regeneration and its expression in sensory neurons of the DRG makes it a reasonable candidate gene for hereditary sensory neuropathy type 1, a disease characterized by progressive degeneration of DRG and motor neurons followed by sensory loss, muscle weakness, and neural deafness (23). The linkage to familial dilated cardiomyopathy is also interesting as there is evidence suggesting that an autoimmune mechanism directed against cell surface adhesion molecules could be involved in this disease (26, 32). The expression pattern of ninjurin, its identification as a new adhesion protein, and its potential linkage to a number of hereditary disorders suggest that it plays an important role in the proper development and function of a variety of tissues.
We thank members of the Milbrandt laboratory for comments on the manuscript and Paul Allen for peptide synthesis and purification.