INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

trkA is the prototype of a family of genes which also includes trkB and trkC, encoding tyrosine kinase receptors for the neurotrophins of the nerve growth factor (NGF) family. Thus, NGF is the preferred ligand for TrkA, brain-derived neurotrophic factor and NT4/NT5 are ligands for TrkB, and NT3 is the only known ligand for TrkC (2). Neurotrophins are responsible for the survival, differentiation, and maintenance of specific populations of neurons in the developing and adult nervous system (7). NGF-triggered TrkA signaling is required for the survival of sensory and sympathetic neurons. In human neuroblastomas, expression of trkA is a good prognostic marker, suggesting that lack of trkA expression contributes to malignancy; perhaps because it results in the loss of signaling pathways important for growth arrest and/or differentiation of the neural crest-derived cells from which these tumors originate (4). On the other hand, in some tumors an autocrine loop, NGF-TrkA, is responsible for tumor progression, as is the case in prostatic carcinoma, in which tumor growth can be blocked by TrkA kinase inhibitors (8). Consequently, TrkA gain-of-function mutations can result in oncogenesis (9, 10).

The extracellular region of TrkA is characterized by a number of distinct structural motifs (22). The amino-terminal sequence consists of three tandem leucine repeats (LRM) flanked by two cysteine clusters. Following the cysteine-rich region there are two immunoglobulin (Ig)-like C2 type domains which contribute significantly to NGF binding (14, 24, 33). As for other receptor tyrosine kinases, ligand-induced homodimerization and conformational changes of TrkA have been proposed as mechanisms for activation of its intrinsic tyrosine kinase activity, followed by transphosphorylation of the two receptor molecules present in the dimer (16).

trkA was originally isolated from a human colon carcinoma as a transforming oncogene activated by a somatic rearrangement that fused a nonmuscle tropomyosin gene with the receptor tyrosine kinase-encoding trkA gene (19, 20). Similar mechanisms are responsible for the malignant activation of trkA in a significant fraction of papillary thyroid carcinomas (9, 10). Different oncogenic forms of trkA have been also identified by transformation of NIH 3T3 cells in culture (6). Among them, one had a partial deletion of the sequences encoding the second Ig-like domain (trk-5 oncogene), and another was mutated in a conserved cysteine residue within the second Ig-like domain. The rest of the oncogenic forms lacked the sequences encoding the extracellular and transmembrane domains (6). The mechanisms by which these receptors are activated by ligand-independent modes are unknown. The existence of TrkA oncoproteins lacking the transmembrane domain indicated that this region is not required for the activation of the tyrosine kinase domain (2). However, exchanging the TrkA transmembrane domain with the corresponding region of other receptors, such as that for tumor necrosis factor receptor 2, yields nonfunctional receptors (5), suggesting that the transmembrane region of TrkA might be required to properly transduce the signals leading to receptor dimerization and autophosphorylation.

In an attempt to determine the mechanisms that regulate TrkA activation and to investigate how the Ig-like domains are involved in TrkA dimerization, we have studied the differentiating and oncogenic potential of several deletion and chimeric TrkA mutants, as well as those of some single-amino-acid mutations in conserved TrkA residues considered crucial to maintain the structure of the Ig-like domains. Our results indicate that the Ig-like domains of TrkA serve not only to bind NGF but also, in the absence of ligand, to inhibit dimerization.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Generation of trkA mutants. DNA manipulations were carried out by established protocols (1, 27). Escherichia coli DH5 was used as the host for propagation of plasmids. Strains CJ236 and MV1190 were used for in vitro mutagenesis. Base substitutions were introduced according to the Muta-gene in vitro mutagenesis kit protocol (Bio-Rad).

Cell culture and transfections. The human embryonic epithelial kidney HEK293 and Rat-1 fibroblast cell lines were plated onto plastic tissue culture dishes in Dulbecco's modified Eagle's medium (DMEM) (Bio-Whittaker, Verviers, Belgium) supplemented with 10% heat-inactivated fetal calf serum, 2 mM L-glutamine, penicillin (50 U/ml), and streptomycin (50 µg/ml) (Bio-Whittaker). The rat adrenal pheochromocytoma PC12nnr5 cell line (11) was plated in DMEM supplemented with 6% heat-inactivated horse serum, 6% heat-inactivated fetal calf serum, penicillin (50 U/ml), streptomycin (50 µg/ml), and 2 mM L-glutamine. Cells were incubated at 37°C in a 95% air-5% CO₂ atmosphere.

Immunoprecipitation and immunoblot analysis. Cells were lysed in a buffer containing 137 mM NaCl, 20 mM Tris-HCl (pH 8), 10% glycerol, 1% NP-40, 2 mM EDTA, and protease inhibitors (0.15 U of aprotinin per ml, 20 µM leupeptin, and 1 mM phenylmethylsulfonyl fluoride) at 4°C for 20 min. Immunoprecipitation was performed for 3 h at 4°C using 2 mg of total protein extract and the anti-203 pan-Trk antiserum (20) (1:1,000 dilution). After several washes, immunoprecipitates were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by Western blot with the pan-Trk antiserum (1:5,000) and horseradish peroxidase (HRP)-conjugated anti-rabbit Ig. Reactive protein bands were visualized by enhanced chemiluminescence detection (Amersham Corp.).

Dimerization assay. Spontaneous or NGF-induced dimerization of TrkA receptors was assayed by introducing two different epitopes in some of the receptor variants. Two hemagglutinin (HA) epitopes were introduced between amino acids 43 and 70, just after the signal peptide of the molecule. The sequence encoding the Myc epitope was introduced at the MluI site previously engineered into the rat trkA cDNA, after the second Ig-like domain (24).

[³H]thymidine incorporation. Cells were seeded in 24-well plates at a density of 2 × 10⁴ cells/well and incubated for 2 days in DMEM with 10% calf serum. Cells were starved for 22 h in medium with 0.2% serum and then treated with either NGF (100 ng/ml) or 10% calf serum or left in 0.2% serum for 12 h. At that time [³H]thymidine was added (0.5 µCi/ml), and incubation was continued for another 4 h at 37°C. Cells were washed with phosphate-buffered saline (PBS) containing 1 mM CaCl₂ and 0.5 mM MgCl₂, precipitated with 10% trichloroacetic acid, and resuspended in 0.1 ml of 0.2 N NaOH. The solution was neutralized by adding 0.1 ml of 0.2 N HCl. The amount of [³H]thymidine incorporated was quantitated by liquid scintillation counting.

Soft agar colony formation. Rat-1 cell lines expressing mutant receptors were resuspended at a concentration of 10³ cells in 2 ml of DMEM with 10% fetal calf serum and 0.5% agar at 45°C and overlaid onto plates containing 4 ml of solidified DMEM supplemented with 10% fetal calf serum and 1% agar. Plates were kept at 4°C for 5 min and incubated at 37°C for 21 days. Twice per week the cells were fed with 0.5 ml of DMEM plus 10% fetal calf serum.

Tumorigenesis in athymic nude mice. Different Rat-1 cell lines expressing the mutant receptors were injected subcutaneously at 10⁴ cells/mouse into 6-week-old female athymic Swiss nu/nu mice (two sites per mouse). Animals were housed under sterile conditions in a germ-free protected unit and fed ad libitum. Tumor growth was assessed weekly by caliper measurement of the tumor in three dimensions. When xenografts reached a mean diameter of 6 mm, the cell lines were scored as tumorigenic. Three mice were used for each cell line.

NGF binding studies. Mouse submaxillary NGF (2.5S) was obtained from Harlan Scientific and radioiodinated by lactoperoxidase treatment as described previously (13). The specific activity ranged from 3,000 to 3,500 cpm/fmol. The [¹²⁵I]NGF was used within 2 weeks of labeling. No proteolysis was detected by SDS-PAGE. [¹²⁵I]NGF binding assays were performed in HEK293-derived cell lines expressing TrkA mutant receptors, and the dissociation constants (K_d) were determined by Scatchard plot analysis, as described (18, 24).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Generation of mutant and chimeric TrkA receptors. Like other receptor tyrosine kinases, TrkA undergoes dimerization and activation upon ligand binding. Thus, in the absence of NGF, some domains of the receptor, perhaps the same ones responsible for ligand binding, must be impeding its spontaneous dimerization at the cell surface. To investigate the contribution of the different domains within the extracellular region of TrkA to receptor function, each of the two Ig-like domains was individually deleted. Deletions of both Ig domains and of the entire extracellular domain (ECD) were also made (Fig. 1A).

View larger version (30K): [in this window] [in a new window]   FIG. 1.   Schematic representation of TrkA mutant receptors. (A) Mutants generated by deleting the Ig-like domains or the entire extracellular region. (B) TrkA/c-Kit chimeric receptors. TM, transmembrane domain; TK, tyrosine-kinase domain.

Ligand-independent neurite formation by TrkA mutants in PC12nnr5 cells. Ligand-activated TrkA promotes differentiation of the PC12 rat pheochromocytoma cell line. Addition of NGF to these cells causes them to stop proliferating and to acquire neurites in 2 to 3 days. We have used the PC12nnr5 mutant cell line, which does not express endogenous TrkA, to study the ability of TrkA mutants to induce differentiation in the absence of NGF.

View larger version (21K): [in this window] [in a new window]   FIG. 2.   Ligand-independent neurite formation in PC12nnr5 cells transfected with TrkA mutant receptors. Neurite outgrowth was quantified 3 days after transfection by assessing the percentage of -galactosidase-positive cells bearing neurites at least twice the length of their cell bodies. Results were normalized to the trk-5 oncogene response (set at 100%). Values were calculated from at least five independent experiments. Means and standard deviations (SD) are shown.

Dimerization of TrkA mutant receptors in the absence of NGF. Dimerization is considered the mechanism responsible for initiating the activation of TrkA (16, 36). To determine whether the mutant receptors were capable of spontaneous dimerization, we tagged some of these mutants with either HA or Myc epitopes (Fig. 3A) and analyzed if they coimmunoprecipitated after transient cotransfection into HEK293 cells. Receptors were immunoprecipitated with anti-HA antibodies and detected by Western blotting using the 9E10 anti-Myc antibody. As a positive control, we cotransfected TrkA-HA and TrkA-Myc into HEK293 cells and treated them with NGF (100 ng/ml) for 5 min before lysis and immunoprecipitation. Neither the HA nor the Myc epitopes altered the affinity of wild-type TrkA for NGF (data not shown). The results obtained with wild-type TrkA and the deletion mutants TrkA-Ig1, TrkA-Ig2, and TrkA-Ig1,2 are shown in Fig. 3B. Anti-Myc immunoblot analysis of the anti-HA immunoprecipitates showed the presence of Myc-tagged TrkA proteins in the HA-TrkA immunoprecipitates of NGF-treated cells and in the HA-TrkA-Ig1, HA-TrkA-Ig2, and HA-TrkA-Ig1,2 immunoprecipitates in the absence of NGF. Therefore, significant spontaneous receptor dimerization occurs in those cells.

View larger version (37K): [in this window] [in a new window]   FIG. 3.   Ligand-independent dimerization of TrkA deletion mutants. (A) Schematic representation of HA- and Myc-tagged wild-type TrkA receptors. These epitopes were introduced in the same position within the different TrkA mutants. (B) Dimerization analysis in HEK293 cells transiently transfected with the following pairs of expression vectors: HA-TrkA and Myc-TrkA; HA-TrkA-Ig1 and Myc-TrkA-Ig1; HA-TrkA-Ig2 and Myc-TrkA-Ig2; and HA-TrkA-Ig1,2 and Myc-TrkA-Ig1,2. As a positive control, we used HA-TrkA- and Myc-TrkA-transfected cells treated with NGF (100 ng/ml) for 5 min. Two days after transfection, cells were lysed and immunoprecipitations (IP) were performed using the 12CA5 anti-HA antibody. Western blot was done with either 9E10 anti-Myc antibody (upper panel) or anti-203 pan-Trk antiserum (bottom panel). (C) Western blot of whole-cell extracts (40 µg of total protein) done with either 203 antiserum, 9E10, or 12CA5. Immunoreactive protein bands were detected by chemiluminescence. Sizes are shown in kilodaltons.

Ligand-independent phosphorylation of TrkA mutant. To confirm that the activated receptors were capable of spontaneous autophosphorylation, expression vectors for the wild-type and mutant receptors were transiently transfected into HEK293 cells. Two days after transfection, cells were lysed and extracts were analyzed by Western blot using the antiphosphotyrosine monoclonal antibody 4G10. As shown in Fig. 4 (upper panel), constitutive phosphorylation was observed for the TrkA deletion mutants except TrkA-Ig1 and for the chimeric mutants except T-Kit 3.2, which behaved like wild-type TrkA. These results were consistent with ligand-independent differentiating activity of the receptors, suggesting that this activity was due to their constitutive tyrosine autophosphorylation.

View larger version (67K): [in this window] [in a new window]   FIG. 4.   Tyrosine phosphorylation of TrkA mutant receptors in the absence of NGF. Expression vectors for the different mutants were transiently transfected into HEK293 cells. Two days after transfection, cells were lysed, and 50 µg of total protein extract was analyzed by Western blot using either 4G10 antiphosphotyrosine antibody (upper panel) or anti-203 pan-Trk antiserum (bottom panel). Sizes are shown in kilodaltons. Arrows indicate positions of the TrkA mutant receptors.

Proliferative potential of Rat-1 cell lines expressing TrkA mutants. Wild-type and mutant trkA cDNAs were stably transfected in Rat-1 cells. Selected G418-resistant cell lines were isolated and analyzed for receptor expression by immunoprecipitation and Western blot using the 203 pan-Trk antiserum. A series of cell lines were used for further studies (Fig. 5A and B). To analyze the proliferative and oncogenic potentials of the TrkA mutants, we measured the rate of DNA synthesis as [³H]thymidine incorporation after serum starvation and the ability to grow in soft agar of Rat-1-derived cell lines expressing those mutant receptors. These experiments were also performed in the absence of NGF, since none of the deletion or chimeric receptors was capable of binding the ligand; therefore, we were analyzing the constitutive (i.e., ligand independent) activity of these receptor variants.

View larger version (36K): [in this window] [in a new window]   FIG. 5.   (A and B) Expression of mutant TrkA receptors in stably transfected Rat-1 cell lines. Immunoprecipitation and Western blot were performed on total cell extracts (2 mg of protein) using the 203 pan-Trk antiserum and HRP-labeled secondary anti-rabbit Ig antibody. Sizes are shown in kilodaltons. (C) [3H]thymidine incorporation by serum-starved Rat-1 cell lines stably expressing mutant TrkA receptors in the absence of NGF. Values are the means from at least five experiments performed with two independent clones of each mutant. Data are normalized to the [3H]thymidine incorporated by the cells in the presence of serum and expressed as a percentage of the incorporation measured in cells expressing the trk-5 oncogene (100%). Means and SD are shown.

View larger version (91K): [in this window] [in a new window]   FIG. 6.   Soft agar colony formation of Rat-1 cell lines expressing mutant TrkA receptors. Colonies were counted after 21 days in culture. Values are the means from at least five experiments performed with two independent clones of each mutant. The percentage of the foci formed by the trk-5-expressing clones (set at 100%) is shown in parentheses for each receptor.

In vivo tumorigenesis of TrkA mutant receptors. We also assessed the ability of the different Rat-1 cell lines expressing TrkA mutant receptors to form tumors when injected into nude mice. Cells were injected subcutaneously, and animals were examined twice per week for tumor formation. A summary of the results is presented in Table 1. Neither normal Rat-1 cells nor cells expressing wild-type TrkA formed tumors. In contrast, all cells expressing deletion mutants formed progressively growing tumors, although there were differences in their aggressiveness. Thus, TrkA-Ig1 tumors developed very late (latency, 22 days), whereas TrkA-Ig2 and TrkA-Ig1,2 were extremely aggressive. Cells expressing the chimeric T-Kit3.2 mutant did not form tumors, whereas cells expressing either the T-Kit4.1 or T-Kit4.2 chimeric receptor caused tumors that were considerably more aggressive in the case of cells expressing T-Kit4.2. Overall, the tumorigenicity data correlated with those of the proliferation and soft agar growth assays and definitively demonstrated that altering the structure of the Ig-like domains leads to the constitutive, oncogenic activation of the TrkA receptor.

Specific point mutations in the second Ig-like domain can cause spontaneous activation of TrkA. From the results described above, it appears that the second Ig-like domain of TrkA plays a critical role in preventing spontaneous activation of the receptor. This is consistent with an important role of this domain in NGF binding. To gain insights on the role of specific residues within this domain, we generated several single-amino-acid mutations, aiming at altering its Ig-like structure. These mutations affected amino acids that are conserved in all members of the Trk family of receptors and that are considered important for maintaining the structure of the Ig-C2 type domains (29). Specifically, the cysteine residues that form the disulfide bridge (C302 and C348) were changed to serines, and the conserved P313 and W317 residues were changed to alanines (Fig. 7A). A tryptophan residue, separated approximately 15 amino acids from the first cysteine, is conserved in 80% of the Ig-C2 type domains, whereas P313 is conserved in only 30% of the Ig-C2 domains (29) but is present in all Trk receptors.

View larger version (26K): [in this window] [in a new window]   FIG. 7.   (A) Schematic representation of point mutations affecting individual amino acid residues within the extracellular region of TrkA. (B) Expression and ligand-independent phosphorylation of mutant TrkA receptors in stably transfected Rat-1 cell lines. Immunoprecipitation (IP) was performed on total cell extracts (2 mg of protein) using the anti-203 pan-Trk antiserum, followed by Western blot with either anti-203 antiserum (upper panel) or 4G10 antiphosphotyrosine monoclonal antibody (lower panel). Expression of TrkA in PC-12 cells (2 mg of protein) is shown for comparison. Sizes are shown in kilodaltons. (C) [3H]thymidine incorporation by Rat-1 cell lines stably expressing mutant TrkA receptors in the absence of serum. Values are the means from at least five experiments performed with two independent clones of each mutant. Data are normalized relative to the [3H]thymidine incorporated by the cells in the presence of serum and expressed as a percentage of the incorporation measured in cells expressing the trk-5 oncogene (100%). Means and SD are shown.

9

9

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Uncontrolled cell proliferation has commonly been associated with deregulated activity of oncogenes. Receptor tyrosine kinases, like many other signaling proteins, can function as oncoproteins. Upon ligand binding, wild-type receptor tyrosine kinases undergo dimerization or conformational changes or both (12, 15, 30), and those changes lead to the activation of the intrinsic protein tyrosine kinase activity that causes the trans-autophosphorylation of the receptor and the phosphorylation of other substrates.

Constitutive, ligand-independent activation of a tyrosine kinase receptor can occur by different mechanisms. In the present study, we have used a series of mutations in the extracellular region of TrkA in order to gain insights into the molecular basis of the activation of this receptor. In particular we have evaluated the hypothesis that the two Ig-like ligand-binding domains might be involved in blocking receptor homodimerization in the absence of ligand. To achieve this we have engineered several types of mutations and analyzed these TrkA mutant receptors in different functional assays. PC12nnr5 neurite outgrowth allowed us to analyze the ability of mutant receptors to induce differentiation in a neural environment, whereas their transforming potential was assessed by analyzing [³H]thymidine incorporation, soft agar growth, and in vivo tumorigenesis of Rat-1-derived cell lines expressing TrkA mutants. The results obtained from these different assays gave a consistent picture: elimination of the first or second Ig-like domain leads to oncogenic activation of the receptor. This suggests a role for both Ig-like domains in the stabilization of the monomeric form of the receptor, perhaps through repulsion that can be negated by ligand binding. The activation caused by deletion of the first Ig-like domain was weak in all the assays compared to that caused by deleting the second domain. The second Ig-like domain appears to be more critical in preventing spontaneous receptor dimerization, and this correlates with the more important role played by this domain in NGF binding (24, 33, 34, 36). In platelet-derived growth factor receptors, Ig-like domains 1 to 3 participate in ligand binding; however, only deletion of the Ig-like domain 3 caused oncogenic activation (32).

Chimeric receptors between TrkA and c-Kit demonstrated that dimerization could also be achieved by retaining the Ig-like structure but exchanging the TrkA Ig-like domains for a c-Kit domain responsible for dimerization. It has been postulated that in c-Kit (stem cell factor receptor), the fourth Ig-like domain favors receptor dimerization once the ligand is bound to Ig-like domains 1 to 3 (3, 23). Substitution of the second Ig domain of TrkA with the third Ig domain of c-Kit (T-Kit3.2) did not cause spontaneous activation, whereas its substitution with the fourth c-Kit domain (T-Kit4.2) renders the receptor highly oncogenic in the absence of NGF. The weaker activation of T-Kit4.1 than of T-Kit4.2 indicated again that the second Ig-like domain of TrkA provides a stronger barrier to spontaneous dimerization.

Our results showed similar but not identical requirements for cell transformation in fibroblasts and for differentiation in PC12 cells. Mutants lacking the entire ECD of TrkA were less active in neuritogenesis than mutants lacking only the Ig-like domains. This suggests that there are sequences in the ECD, perhaps the LRM, which play a positive role in differentiation. Consistent with this model, deleting the two Ig-like domains eliminated a dimerization block and stimulated the differentiation activity of the receptor, whereas further deletions of the ECD reduced this activity. This effect was not observed in proliferation assays, where deletion of the entire ECD had the same activating effect as the deletion of Ig domains 1 and 2. It appears that there are sequences in the ECD, outside the Ig domains, that play a positive role in differentiation activity but are not required to stimulate cell proliferation or transformation. Some observations suggest that the LRM leucine repeats may be responsible for this effect. Thus, mutant L92V L95V exhibits reduced ligand-induced neuritogenesis compared to the wild-type receptor in spite of having the same affinity for NGF (not shown). This is also consistent with the results reported previously (17).

The biological activities of the single-amino-acid mutants within the second Ig domain of TrkA corroborated the results obtained with the deletion and chimeric receptors and added information on the role of specific residues within this domain. The importance of W317 has been unveiled. The different activities of mutants C302S and C348S are difficult to interpret, since these residues form a disulfide bridge in the external part of the domain that is different from that found in canonical Ig domains (31, 36). Some activating mutations in the fibroblast growth factor receptor (FGFR) family result in the loss of a cysteine residue that disrupts disulfide bond formation between the two highly conserved cysteines of the third Ig-like domain (22). These mutations give rise to an unpaired cysteine residue that may form an intermolecular disulfide bond (26). The fact that mutation C302S in TrkA is considerably more activating than C348S could be explained by invoking a similar mechanism by which the first cysteine would be capable of forming intermolecular disulfide bonds more easily than C348. However, the double mutation C302S C348S is also activating (data not shown). This fact, together with the crystal structure of the NGF-bound domain, suggests that the unusual disulfide bridge formed by these two cysteine residues, in the external face of the domain, is not necessary to maintain the structure of the domain and that residue C302 is required to bind NGF and to keep the monomeric form of the receptor in the absence of ligand, whereas C348S is not.

It is intriguing that the point mutation P287A, affecting a proline conserved only in some Ig-like domains that allows binding of NGF with an affinity similar to that of the wild type receptor, can also induce ligand-independent TrkA-mediated transformation in Rat-1 fibroblasts. A mechanism other than alteration of the Ig-like structure must be responsible for the spontaneous dimerization of this mutant receptor. Some of the activating mutations in the colony stimulating factor-1 (CSF-1) receptor fall near the fourth Ig-like domain (35), and these mutant receptors are able to bind and respond to macrophage CSF M-CSF. Similarly, the only mutation identified in the human FGFR1 is a P252R substitution in the linker region between Ig-II and Ig-III domains that is associated with the Pfeiffer syndrome (21). Interestingly, this mutant receptor seems to bind slightly more radiolabeled FGF than the wild-type receptor (22). An analogous mutation has been identified in FGFR2 in individuals with Apert syndrome (37). Recent crystallization of the FGFR2 Ig-II and Ig-III domains bound to FGF2 has shown that the linker region between domains is important for the structure of the receptor (25). Only the second Ig-like domain of TrkA, bound to NGF, has been crystalized (36); therefore, the effect of mutations within this region cannot be interpreted.

In addition to ligand-receptor contacts, direct receptor- receptor interactions might contribute to dimerization. Our results indicate that the equilibrium between attractive and repulsive forces that is necessary to block TrkA receptor dimerization in the absence of ligand is achieved through the two Ig-like ligand-binding domains. As we have discussed above, indications exist pointing to a role of the extracellular domain in positively regulating TrkA receptor activity, since truncation of the entire ectodomain resulted in an activated receptor that was, however, less biologically active than other mutant receptors in its ability to stimulate neurite outgrowth.

In summary, the data presented here indicate that the ECD of TrkA exerts multiple regulatory effects on the intracellular catalytic domain by mechanisms involving ligand-induced dimerization and also by inhibiting spontaneous receptor dimerization and phosphorylation.