If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
* This study was supported in part by a University Research grant (to J. L. R. and K.-Y. L.) and an operating grant from the Canadian Institutes of Health Research (to K.-Y. L.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ** Alberta Heritage Foundation for Medical Research Senior Scholar.
We have previously shown evidence for the existence of a calcium-independent, GTP-regulated mechanism of secretion from neutrophils, but this secretory mechanism remains to be fully elucidated. Cyclin-dependent kinase 5 (Cdk5), the various substrates of which include Munc18 and synapsin 1, has been implicated in neuronal secretion. Although the Cdk5 activator, p35, and Cdk5-p35 activity are primarily associated with neurons, we report here that p35 also exists in neutrophils and that an active Cdk5-p35 complex is present in these cells. Cdk5-p35 activity in human neutrophils is mostly localized in secretory granules, which show an increase in Cdk5-p35 level and activity upon GTP stimulation. The potent Cdk5 inhibitor, roscovitine, completely blocks GTP-stimulated granule Cdk5 activity, which accompanies lactoferrin secretion from neutrophil-specific granules. Roscovitine also inhibits GTP-induced lactoferrin secretion and surface localization of the secretion markers, CD63 and CD66b, to a certain extent. Furthermore, neutrophils from wild-type mice treated with roscovitine and neutrophils from p35–/– mice exhibit comparable surface expression levels of both CD63 and CD66b upon GTP stimulation. Although our data suggest that other molecules control GTP-induced secretion from neutrophils, it is clear that Cdk5-p35 is required to elicit the maximum GTP-induced secretory response. Our observation that multiple proteins in neutrophil granules serve as specific substrates of Cdk5 further supports the premise that the kinase is a key component of the GTP-regulated secretory apparatus in neutrophils.
To effectively function in host defense, neutrophils secrete bactericidal proteins and proteolytic enzymes that destroy invading pathogens. Nucleotides such as ATP, ATPγS,
). Such mechanism requires GTP hydrolysis to stimulate secretion in the presence or absence of calcium. Several proteins have been implicated in regulating secretion from neutrophils, including the small ras-related GTPases (
). Unlike other Cdks, however, the cellular functions of Cdk5 are seemingly unrelated to the cell cycle. In addition, activation of Cdk5 requires association with its non-cyclin regulatory partner, p35 (or its truncated form, p25) (
). Cdk5 is most abundant in neurons and its activators, p35 and p39, have been presumed to be neuron-specific. Although Cdk5 activity has generally been associated with brain, recent studies have implicated a role for Cdk5 in other tissues such as muscle (
), and believed to be an essential component of the secretion machinery, has implicated Cdk5 in neuronal secretion. An increasing line of evidence also demonstrates Cdk5 as a modulator of neurocytoskeletal dynamics (
), which play an essential role in regulated secretion.
Previous studies have shown that neutrophils share some secretory mechanisms with neurons. In this study, the possibility that Cdk5 participates in the regulation of the calcium-independent GTP-mediated secretion from neutrophils was investigated using SO-permeabilized cells. We show for the first time that Cdk5 and its known neuronal activator, p35, are present in neutrophils and that p35-associated Cdk5 activity is present in these cells as well. Cdk5 activity, which is most abundant in neutrophil granules, increases with GTP stimulation and secretion from specific granules. The potent Cdk5 inhibitor, roscovitine, inhibits such effects. In accord with these findings, we found two neutrophil granule proteins that serve as effective substrates of Cdk5. Together, our observations suggest that Cdk5 plays an important role in regulating GTP-mediated secretion from neutrophils.
MATERIALS AND METHODS
Neutrophil Isolation, Permeabilization, and Stimulation—Neutrophils obtained from normal human volunteers and mice (p35–/– and corresponding wt; ref.
). In brief, permeabilization using SO (Murex Diagnostics Inc., Norcross, GA) was performed in two steps: 1) SO binding at 4 °C for 10 min followed by washing to remove unbound SO, and 2) pore formation by resuspending cells in prewarmed (37 °C) permeabilization buffer and incubation at 37 °C for 10 min. Stimulation with 300 μm GTP (Roche Molecular Biochemicals, Indianapolis, IN) in the presence or absence of 20 μm roscovitine (Calbiochem-Novabiochem) was performed simultaneously with the second step of permeabilization. Incubation at 37 °C was continued for an additional 15 min after the pore formation step. Cells treated with roscovitine during GTP stimulation were also pre-incubated with 20 μm roscovitine for 30 min at RT before permeabilization.
PCR/RT-PCR—To detect the specific PCR products of Cdk5 (650 bp) and p35 (450 bp), cDNAs generated from RNAs (isolated using the RNeasy® mini kit; Qiagen) by RT reaction (using SuperScript III reverse transcriptase; Invitrogen) were used as templates for PCR reactions. PCR was carried out using Taq from Invitrogen. The following primers were used: Cdk5 forward primer (TGAGGGTGTGCCAAGTTCAGC) and reverse primer (GGCATTGAGTTTGGGCACGAC); p35 forward primer (ACCTCTGCAGGGACACCCAAACG) and reverse primer (GTGGGTCGGCATTGATCTGCAGC); and β-actin forward primer (GAACCCTAAGGCCAACCGTGA) and reverse primer (AGGAAAGGATGCGGCAGTGG).
Preparation of Cell and Granule Lysates and Subcellular Fractionation—Cells were sonicated in HEPES buffer A (25 mm HEPES, pH 7.2) containing 2 mm MgCl2, 150 mm NaCl, 2.5 mm dithiothreitol, 10 mm NaF, 1 μg/ml sodium orthovanadate, and 1 mm diisopropyl fluorophosphate. Samples were then processed to obtain the designated cell lysate or subcellular fractions. To prepare the cell lysates, sonicated samples were incubated with Triton X-100 (final concentration, 1%) and additional NaCl (final concentration, 800 mm) at 4 °C for 1 h on a rocker. The supernatant obtained after centrifugation at 14,000 rpm for 10 min was designated as cell lysate. To prepare the subcellular fractions, sonicated samples were centrifuged at 420 × g for 10 min to separate unbroken cells and nuclei. The supernatant was centrifuged at 30,000 × g for 15 min to obtain the granule fraction. The supernatant was further centrifuged at 100,000 × g for 60 min to obtain the membrane (pellet) and cytosol (supernatant) fractions. Granule lysates were prepared by sonication of the granule fraction in HEPES buffer A followed by incubation with 1% Triton X-100 and 800 mm NaCl for 1 h at 4 °C ona rocker. After centrifugation at 30,000 × g for 15 min, the supernatant was designated as the granule lysate.
Immunoprecipitation—To immunoprecipitate Cdk5 and p35, cell or granule lysates precleared with protein A/G plus agarose beads were incubated with fresh beads pre-adsorbed with an antibody to either Cdk5 (C-8 or DC-17) or p35 (C-19) (Santa Cruz Biotechnology, Santa Cruz, CA) for 2 h on a rocker at 4 °C. The immunocomplexes were washed extensively with Tris-buffered saline (10 mm Tris-HCl, pH 7.4, 150 mm NaCl, and 2.6 mm KCl) containing 0.1% Tween 20.
SDS-PAGE and Western Blotting—SDS-PAGE was performed according to the method of Laemmli (
). For Western blot analysis, proteins were transferred onto a nitrocellulose membrane (Millipore, Bedford, MA), probed with a primary antibody (C-8 or DC-17 anti-Cdk5, or C-19 anti-p35; Santa Cruz Biotechnology) then incubated with a horseradish peroxidase-conjugated secondary antibody (Pierce Chemical). Immunoreactive protein bands were visualized using the ECL detection system (Amersham Life Sciences, Piscataway, NJ).
Immunofluorescence Microscopy—Cells suspended in Hanks' balanced salt solution containing 0.15% bovine serum albumin were allowed to adhere to coverslips incubated at 37 °C with 5% CO2 for 30 min. The media was replaced gently with fresh Hanks' balanced salt solution/bovine serum albumin ± 20 μm roscovitine and incubation at 37 °C was continued for another 30 min. After rinsing with permeabilization buffer (Ref.
), cells were incubated with prewarmed (37 °C) permeabilization buffer ± 300 μm GTP ± 20 μm roscovitine. After treatment, cells were rinsed with phosphate buffered saline, pH 7.2, and fixed in 3.7% paraformaldehyde for 10 min at RT. After washing, cells were incubated with either the CD63 or CD66b antibody (monoclonal antibodies; BD Pharmingen, San Diego, CA), washed, and incubated with a Cy3-conjugated secondary antibody (Jackson ImmunoResearch Labs, West Grove, PA). The coverslips were mounted on glass slides using the SlowFade mounting medium (Molecular Probes). Immunofluorescence staining was visualized using an Olympus IX71 inverted microscope, and images were acquired using a cooled charge-coupled device camera.
). In brief, samples were incubated with 20 units of calf intestinal alkaline phosphatase (Pfizer, New York, NY) for 2 h at 37 °C. Dephosphorylation was stopped by incubating the samples at 70 °C for 15 min, which also inactivates endogenous kinases and phosphatases. Phosphorylation of granule proteins by Cdk5 was then performed by incubating the dephosphorylated granule samples with a Cdk5 immunoprecipitate (from a neutrophil lysate) ± 20 μm roscovitine in the presence of 30 μm [γ-32P]ATP in kinase assay buffer (
) at 30 °C for 40 min. The reaction was stopped by adding SDS-PAGE sample buffer and heating for 5 min at 95 °C. Samples were then resolved in a 12.5% SDS-polyacrylamide gel, transferred to a nitrocellulose membrane, and phosphorylation of granule proteins detected by autoradiography.
Miscellaneous—Quantitation of the extent of permeabilization and secretion of lactoferrin into the extracellular medium was performed as we described previously (
). Control mouse brain samples were homogenized in 25 mm HEPES buffer, pH 7.0, containing 250 mm NaCl, 1% Triton X-100, 1 mm dithiothreitol, 1 mm EDTA, 0.6 mm phenylmethylsulfonyl fluoride, 1 μg/ml each of leupeptin, antipain and aprotinin, 0.3 mg/ml benzamidine, and 0.1 mg/ml soybean trypsin inhibitor) using a tissue homogenizer (CH-6010; Kinematica, Kriens-Luzern, Switzerland). Samples were clarified by centrifugation at 16,000 × g for 40 min at 4 °C. Protein concentration was determined using the Bradford microassay (
) where the kinase is found to have remarkable activity. To investigate the possibility that Cdk5 participates in the regulation of secretion from neutrophils, we initially examined human neutrophils for the expression of transcripts of Cdk5 and its known neuronal activator, p35. As shown in Fig. 1A, RT-PCR revealed the presence of transcripts of the ubiquitous Cdk5 (left) and the more tissue-specific p35 (right) in human neutrophils. The presence of p35 is supported by parallel analysis of brain cDNA from p35–/– (negative control) and corresponding wt (positive control) mice. Because p35 is an intronless single exon gene, genomic DNA contamination will result in a false positive PCR product. To prove that the PCR product was, indeed, derived from p35 cDNA from the RT reaction and not from potential genomic DNA contamination, isolated RNA was used as template for the p35 PCR negative control reaction (lane 3). RT-PCR amplification with rat β-actin gene primers spanning intron II was used to confirm the success of RT reactions using cDNA as template (bottom). Western blot analysis revealed that both Cdk5 (Fig. 1B, lane 2, left) and p35 (Fig. 1B, lane 2, right) proteins are also present in neutrophils. Cdk5 immunoprecipitates from neutrophil lysates also showed a considerable amount of histone H1 kinase activity (Fig. 1C, bar 2), indicating that Cdk5 exists as an active kinase in neutrophils as well.
Cdk5-p35 Activity in Neutrophils Is Mostly Associated with the Secretory Granules—Secretion from neutrophil granules results from the fusion of granule membranes with the plasma membrane, implying important roles for certain molecules in these subcellular compartments during the secretory process. To examine whether Cdk5-p35 localizes in the granule and/or membrane compartments of neutrophils, the kinase was immunoprecipitated from neutrophil subcellular fractions using a p35 antibody (C-19). Western blotting showed that although Cdk5 levels are comparable in the membrane and granule fractions (Fig. 2A), a much higher level of Cdk5 from the granule fraction coimmunoprecipitates with p35 (Fig. 2B), indicating the formation of a Cdk5-p35 complex in neutrophils that is most abundant in granules. The level of Cdk5-p35 in the subcellular compartments correlates with the level of Cdk5 kinase activity (Fig. 2C). Therefore, it seems that Cdk5-p35 in neutrophils has an important function in granules and, potentially, in the secretion of granule contents.
GTP Stimulates Granule Cdk5-p35 Activity That Accompanies Lactoferrin Secretion—Because Cdk5-p35 activity in neutrophils is mostly associated with the secretory granules, we sought to examine a potential role for the kinase in GTP-dependent secretion from SO-permeabilized neutrophils. As shown in Fig. 3A, GTP stimulation elicited an increase in Cdk5-p35 kinase activity in neutrophil granules. It seems that such an increase in kinase activity is caused by an increase in Cdk5-p35 level (Fig. 3A, inset). Although roscovitine, a potent Cdk5 inhibitor (
), did not inhibit the association of Cdk5 with p35 (Fig. 3A, inset, lane 3), it completely inhibited the kinase activity of the p35 immunoprecipitate, suggesting that the increase in histone kinase activity in granules was in fact caused by Cdk5-p35. We then examined whether GTP-stimulated lactoferrin secretion from specific granules can also be inhibited by roscovitine. As shown in Fig. 3B, roscovitine caused inhibition of GTP-stimulated lactoferrin secretion from specific granules. The incomplete effect of roscovitine suggests that other molecules regulate the secretion of lactoferrin. Nonetheless, it seems that participation of Cdk5-p35 is essential for maximum GTP-stimulated secretion from specific granules.
Neutrophils from p35–/–Mice Display Reduced GTP-stimulated Surface Localization of CD63 and CD66b—Because we also detected both Cdk5 (Fig. 4B, left) and p35 (Fig. 4B, middle) transcripts in mouse neutrophils, the relevance of p35-associated Cdk5 activity in secretion by GTP-stimulated neutrophils was further examined using cells from p35–/– mice and their corresponding wt. Secretion was assessed by analyzing the surface expression of CD63 and CD66b, which have been shown as effective markers of secretion from azurophil and specific granules, respectively (
). As shown in Fig. 4, GTP stimulation resulted in an obvious increase in surface expression of both CD63 and CD66b in wt neutrophils. Such effect of GTP in wt neutrophils is partially inhibited by roscovitine treatment. In p35–/– mouse neutrophils, GTP stimulation resulted in CD63 and CD66b surface expression that is comparable with that of roscovitine-treated GTP-stimulated wt neutrophils. The partial induction of surface expression of the CD63 and CD66b secretion markers in GTP plus roscovitine-treated neutrophils from wt mice and in GTP-treated neutrophils from p35–/– mice corroborates our finding that roscovitine partially inhibits lactoferrin secretion from human neutrophils (Fig. 3B). Together, our observations suggest that although GTP-stimulated secretion from neutrophils may occur in the absence of Cdk5-p35 activity, the maximum GTP-induced secretory response from both the azurophil and specific granules could only be achieved in the presence of Cdk5-p35 activity.
Cdk5 Has Two Major Substrates in Neutrophil Granules— Because our findings demonstrate the regulation of GTP-dependent secretion from neutrophil granules by Cdk5, it is likely that at least one granule protein that also plays a role in GTP-mediated secretion serves as a substrate for Cdk5. Indeed, by phosphorylation assay, we found multiple granule proteins that were phosphorylated by Cdk5 (Fig. 5, lane 3). Two of these granule proteins, which have apparent molecular masses of ∼49 and ∼54 kDa, were intensely phosphorylated by Cdk5, suggesting that both proteins act as major substrates of Cdk5. The presence of roscovitine inhibited the phosphorylation of these granule proteins (Fig. 5, lane 4), indicating the specificity of phosphorylation by Cdk5. Phosphorylation of at least one of these proteins by Cdk5 may have a crucial role in GTP-mediated secretion from neutrophil granules. Further characterization of these 49- and 54-kDa proteins is in progress.
Although Cdk5 is ubiquitously expressed in mammalian tissues, Cdk5 activity has mostly been associated with the brain tissue, where the Cdk5 activators, p35 and p39, were initially found. In this study, we demonstrate for the first time that p35-associated Cdk5 activity is present in neutrophils as well. Neutrophils offer the first line of defense against infection and are potent effectors of inflammation. To achieve these functions, neutrophils generate reactive oxygen intermediates and secrete proteolytic enzymes. However, the components of the neutrophil secretory apparatus and their specific functions are not yet well characterized. We have previously shown that a calcium-independent GTP-mediated mechanism of secretion exists in neutrophils (
). In the current study, we demonstrate that Cdk5-p35 is an important component of the GTP-induced secretory machinery in neutrophils.
Our observation that Cdk5-p35 expression and activity in neutrophils is mostly associated with secretory granules suggests that the kinase has a relevant function in this subcellular compartment. Increased Cdk5-p35 level and activity in neutrophil granules upon GTP stimulation supports this notion. These findings are consistent with the observation in neuroendocrine cells whereby stimulation of secretion corresponds to a rapid translocation of cytosolic Cdk5 to the particulate fraction as well as an increase in Cdk5 kinase activity (
). The complete inhibition of GTP-stimulated kinase activity in granules by roscovitine substantiates the premise that the increase in kinase activity is specific to Cdk5. Inhibition of GTP-stimulated lactoferrin secretion by roscovitine further points to a role for Cdk5 in GTP-mediated secretion from neutrophils. This concurs with the inhibitory effect of roscovitine on the GTP-stimulated increase in surface expression of CD63 and CD66b in neutrophils. This increase in plasma membrane expression of CD63 and CD66b has been shown in the past to closely correlate with conventional assays of secretion from neutrophil azurophil and specific granules, respectively (
). Although roscovitine may cause some inhibitory effect on the cell cycle Cdk1 and Cdk2, such effect may be disregarded in secretion by terminally differentiated neutrophils.
The partial effect of roscovitine on GTP-induced secretion from human and mouse neutrophils is consistent with the observations in GTP-stimulated p35–/– mouse neutrophils. It would seem that other molecules are involved in GTP-mediated secretion, but it also seems that participation of Cdk5-p35 is crucial to reach the maximum GTP-triggered secretory response. Our observations parallel a previous report that Cdk5-p35 acts as a positive regulator of insulin secretion from pancreatic β cells (
) is also regulated by Cdk5. It is possible that the calcium-independent secretory mechanism in pancreatic β cells is analogous to that of the GTP-induced, calcium-independent secretory mechanism that exists in neutrophils. Perhaps Cdk5-p35 is one of the kinases responsible for the phosphorylation events that may be critical to achieve the maximum GTP-induced release of granule contents from neutrophils.
Obvious phosphorylation of multiple granule proteins by Cdk5, particularly of the ∼49- and ∼54-kDa proteins, which seem to act as effective substrates of the kinase, may point to the possible targets of Cdk5 during GTP-induced secretion. At least one of the Cdk5 phosphorylated proteins may serve as an effector molecule for GTP-mediated secretion from neutrophil granules. This theory concurs with the finding that phosphorylation of the mammalian neuronal homologue of Sec1p, Munc18, by Cdk5 reduces its affinity to syntaxin1 (
), allowing the interaction between syntaxin1 and the other effectors of secretion. It seems that in pituitary cells, Cdk5 phosphorylation of the RhoGEF, Trio, is also involved in the actin dynamics required for endocrine cell secretion (
). It is possible that in neutrophils, phosphorylation of the 49- and/or 54-kDa granule proteins by Cdk5 results in a positive interaction among the protein effectors of GTP-mediated secretion as well. Further studies are required to substantiate this premise.
We thank Dr. Inez Vincent (University of Washington) for providing the p35–/– mice breeding pair.