ADAMTS4 cDNA induces neurite outgrowth in neurons grown on astrocyte monolayers. Primary cortical neurons transfected with empty pCMS vector or pCMS inserted with ADAMTS4 cDNA were grown on a previously prepared astrocyte monolayer for 7 days. Representative photomicrographs depict neurons transfected with empty pCMS (A) or with the ADAMTS cDNA insert (C). Photographs were taken of GFP-expressing neurons and neurites were traced as a measure of neurite outgrowth (B, pCMS; D, pCMS-ADAMTS4). Means ± SEM of traced neurons reveal ADAMTS4 increases several parameters of neurite outgrowth including the mean longest primary and secondary neurites, overall mean primary and secondary neurites and number of secondary neurites (* signifies p ≤ 0.05 compared to control pCMS vector alone). No significant difference was observed in the mean number of primary neurites in response to ADAMTS4 cDNA transfection (p > 0.05). A single independent neuronal culture was utilized to measure 117 primary and 84 secondary neurites from 50 empty pCMS-transfected neurons and 127 primary and 124 secondary neurites from 37 pCMS-ADAMTS4-transfected neurons. Similar data was obtained when the experiment was repeated. Scale bars, 50μm.

An inactive mutant ADAMTS4 cDNA enhances neurite outgrowth. Primary cortical rat neurons transfected with empty pCMS vector, pCMS inserted with ADAMTS4 cDNA, or pCMS vector inserted with an inactive mutant form of ADAMTS4 cDNA were grown on poly-L-lysine/laminin coated coverslips in the absence of an astrocyte monolayer for 7 days. Due to a point mutation in the catalytic domain (362 E→Q), this ADAMTS4 protein possesses the same structural domains as native ADAMTS4, but lacks any proteolytic capacity. Representative photomicrographs depict pCMS (A), pCMS-ADAMTS4 (C) and pCMS inactive ADAMTS4 mutant (E) expressing neurons. Photographs were taken of the fluorescing neurons and neurites were traced as a measure of neurite outgrowth (B, pCMS; D, pCMS-ADAMTS4, F, pCMS inactive ADAMTS4 mutant). Means ± SEM of traced neurons reveal ADAMTS4 and inactive ADAMTS4 mutant induced statistically significant (* signifies p ≤ 0.05 compared to pCMS vector alone) increases in several parameters of neurite outgrowth including the mean longest primary and secondary neurites, overall mean primary and secondary neurites and the number of secondary neurites. There was an additional increase in the mean length of primary neurites of neurons transfected with pCMS inactive ADAMTS4 mutant cDNA compared to neurons transfected with pCMS-ADAMTS4 cDNA, (** signifies p ≤ 0.05 compared to pCMS-ADAMTS4). No significant differences (p > 0.05) were observed in the mean number of primary neurites in response to ADAMTS4 cDNA transfection. A single independent neuronal culture was utilized and 288 primary and 181 secondary neurites from 111 empty pCMS-tranfected neurons were examined. 207 primary and 171 secondary neurites from 70 pCMS-ADAMTS4-transfected neurons were measured, and 325 primary and 316 secondary neurites from 129 pCMS-inactive ADAMTS4-transfected neurons were quantified. Similar data was obtained when the experiment was repeated. Scale bars, 50μm.

Tyrosine residue phosphorylation in total protein extract from neurons after treatment with human recombinant ADAMTS4. Representative immunoblot showing protein bands immunoreactive with anti-phophotyrosine antibody in extracts from neurons, at 5 days in culture and treated with culture medium or 50 nM ADAMTS4 for 20, 40 and 60 min. Proteins immunoreactive for phospho-tyrosine were observed at several molecular weights, 42 (A), 44 (B), 50 (C), 115 (D), >120 (E). Protein bands from this single experiment were quantified by densitomety and values are shown in panel (F) indicate increases in band density for proteins at several different molecular weights, especially prominent is a band at about 40 kD. This blot is from a representative experiment and results were similar in three replicate experiments.

MAP kinase inhibitors attenuate neurite outgrowth induced by human recombinant ADAMTS4 treatment. PD98059, an inhibitor of MAP kinase (A) U0126, a potent inhibitor of MEK-1 and -2 (B) and U0124, an inactive analog of U0126 (C) were added to the culture medium of neurons at 2 and 5 days in culture to determine whether inhibition of the MAP kinase could influence the growth of neurites induced by ADAMTS4. Cells were treated with vehicle (DMSO) or inhibitor alone or with inhibitor thirty minutes before the addition of 50nM ADAMTS4, followed by fixation, immunostaining and quantification of neurite outgrowth by tracing neurites on day 6 of the culture. Area occupied by neurites (μm²) from neurite tracings was divided by number of neurons in the field to generate a mean ± SEM (μm²/neuron). Each MAP kinase inhibitor significantly and dose-dependently impaired the neurite outgrowth promoting effects of ADAMTS4 with little effect on basal neurite outgrowth with the exception of high doses of U0126 which appeared to be toxic toward the neurons. * signifies p ≤ 0.05 compared to “control” treatment; ** signifies p ≤ 0.05 compared to treatment with 50 nM ADAMTS4 alone. This experiment was performed using an independent culture and at least 10 fields were examined for each treatment group representing about 75 neurons per treatment group. Similar data was obtained when the experiment was repeated.

Proteoglycan (lectican)-tenascin-hyaluronan matrix complex in the CNS and lectican cleavage by ADAMTSs. (A), Intact complexes of extracellular matrix form an inhibitory boundary toward neurite outgrowth by hyaluronic acid binding to the N-terminus, tandem repeats of the lecticans and tenascin binding to the C-terminus of the lecticans. (B), Cleavage of lectican by an ADAMTS loosens the stable, matrix network, potentially increasing fluidity and allowing for increased plasticity. (C, D), Cleavage of the lectican, brevican, by human recombinant ADAMTSs. Proteoglycans were purified from rat brain extracts contained in fractions eluted from a DEAE column with 1 M NaCl. Lectican-containing fractions were incubated with active ADAMTSs, or heat-inactivated protease (Inactive), subjected to SDS-PAGE, and probed with anti-brevican antibody (C) to reveal holoproteins and the N-terminal cleavage product or with anti-EAVESE antibody to selectively reveal the N-terminal fragment that contains the EAVEAE neoepitope exposed by ADAMTS-cleavage of brevican (D). “Control” lane shows brevican immunoreactivity as a smear at a molecular weight of >145 kD. Samples containing inactive ADAMTS show little cleavage, whereas brevican substrate incubated with human recombinant ADAMTSs 1, 4 and 5 (25 nM) for 3 hours reveal an abundance of the 55 kD, N-terminal cleavage fragment (C) that is observed when probed with antibody selective for the ADAMTS-cleaved fragment (D).

ADAMTS4 cDNA induces neurite outgrowth in neurons grown on a poly-L-lysine/laminin substrate. Primary cortical neurons transfected with empty pCMS vector or pCMS inserted with ADAMTS4 cDNA were grown on poly-L-lysine/laminin coated coverslips in the absence of an astrocyte monolayer for 7 days. Representative photomicrographs depict neurons transfected with empty pCMS (A) or with the ADAMTS cDNA insert (C). Photographs were taken of the GFP-expressing neurons and neurites were traced as a measure of neurite outgrowth (B, pCMS; D, pCMS-ADAMTS4). Means ± SEM of traced neurons reveal ADAMTS4 induced statistically significant (* signifies p ≤ 0.05 compared to pCMS vector alone) increases in several parameters of neurite outgrowth including the mean longest primary and secondary neurites, overall mean primary and secondary neurites and the number of primary and secondary neurites. 112 primary and 48 secondary neurites from 45 empty pCMS-transfected neurons and 136 primary and 107 secondary neurites from 44 pCMS-ADAMTS4 transfected neurons derived from a single independent neuronal culture were analyzed. Similar data was obtained when the experiment was repeated. Scale bars, 50μm.

Human recombinant ADAMTS4 induces neurite outgrowth. Representative photomicrographs and tracings are shown of primary cortical rat neurons grown on poly-L-lysine/laminin and treated with ADAMTS4 at 1, 10 (D, E) and 50nM (F, G) and ADAMTS5 at 10nM (H, I) or culture medium containing BSA (B, C) at days 2 and 5 of culture and neurite outgrowth measured at day 6. At the end of culture, neurons were visualized by using anti-microtubule associated protein-2 immunocytochemistry and the Alexafluor 488-conjugated actin binding protein, phalloidin. The photomicrographs (B, D, F, H) were converted to grayscale and neurites traced (C, E, G, I). Data was calculated by dividing the area of the photograph occupied by neurite signal (μm²) divided by number of neurons in the field to generate mean ± SEM (μm²/neuron) that are shown in panel A. Treatment with ADAMTS4 resulted in a dose-dependent increase in neurite outgrowth (see panel A) and 10nM ADAMTS5 also significantly increased neurite outgrowth (* p ≤ 0.05 compared to untreated or medium + BSA treated wells). Three separate cultures were utilized in these experiments, with at least 10 fields in total for each treatment group, representing over 1,000 neurons for each group. Scale bars, 100μm.

ERK1/2 is activated in neurons in response to treatment with human recombinant ADAMTS4. Representative blots (A and B) and densitometric values of immunoreactive bands (C and D) of ERK1/2 immunoreactivity in extracts from neurons at 5 days in culture and treated with ADAMTS4. Culture medium vehicle or human recombinant ADAMTS4 at 50 nM was added to neuronal culture medium and cell lysates were collected 20, 40 and 60 minutes after treatment, the lysates subjected to SDS-PAGE and Western blotting using anti-42/44kD phospho-ERK1/2 (A) and anti-pan-ERK1/2 (B) antibodies. Densitometry of the immunoreactivity bands was quantified and the ratio of phospho-ERK1/2 to pan-ERK1/2 was calculated for each sample and for each isoform of ERK 1/2. (C) ratio of 42 kD activated phospho-ERK to pan- 42 kD ERK, (D), ratio of 44 kD activated phospho-ERK to pan- 44 kD ERK. Marked and apparently sustained increases in activated ERK1/2 were observed in response to treatment with ADAMTS4 This experiment was repeated four times, with increases seen in each experiment, however, the time of the peak magnitude of the increase was variable among experiments, Thus, results from a single, representative experiment is shown.