Regulation of mucin secretion and inflammation in asthma; A role for MARCKS protein?

Abstract

A major characteristic of asthmatic airways is an increase in mucin (the glycoprotein component of mucus) producing and secreting cells, which leads to increased mucin release that further clogs constricted airways and contributes markedly to airway obstruction and, in the most severe cases, to status asthmaticus. Asthmatic airways show both a hyperplasia and metaplasia of goblet cells, mucin-producing cells in the epithelium; hyperplasia refers to enhanced numbers of goblet cells in larger airways, while metaplasia refers to the appearance of these cells in smaller airways where they normally are not seen. With the number of mucin-producing and secreting cells increased, there is a coincident hypersecretion of mucin which characterizes asthma. On a cellular level, a major regulator of airway mucin secretion in both in vitro and in vivo studies has been shown to be MARCKS (Myristoylated Alanine-Rich C Kinase substrate) protein, a ubiquitous substrate of protein kinase C (PKC). In this review, properties of MARCKS and how the protein may regulate mucin secretion at a cellular level will be discussed. In addition, the roles of MARCKS in airway inflammation related to both influx of inflammatory cells into the lung and release of granules containing inflammatory mediators by these cells will be explored.

Introduction

In asthma and many other chronic respiratory diseases, excessive mucus production and airway inflammation can ultimately contribute to morbidity and mortality in many patients. Subsequently, the development of drugs that inhibit overproduction of mucus and chronic inflammation is necessary. Although some conventional therapies, such as anticholinergics, β2-adrenoceptor agonists, and corticosteroids are available, they have variable effectiveness. In recent years, MARCKS (Myristoylated Alanine Rich C-Kinase Substrate) protein has emerged as a new target for inhibition of mucus hypersecretion and inflammation. This review will focus on MARCKS and its apparent roles in both secretion and inflammation, making it an attractive potential therapeutic target for asthma and other respiratory diseases characterized by mucus hypersecretion and inflammation.

The identification of MARCKS protein dates back to 1982 when it was found that an ’87 kDa’ substrate in rat brain nerve endings could be regulated by calcium and calmodulin (Ca2+/CaM) through the activation of PKC [1]. A similar substrate for PKC was purified from bovine brain and it was found to have widespread species, tissue, and subcellular distribution [2–3]. This protein captured the interest of several groups and subsequently the protein was officially named Myristolyated Alanine Rich C Kinase Substrate (MARCKS) [4]. The classical members of the MARCKS family are: MARCKS, an 87 kDa protein ubiquitously expressed in bovine, chicken, mouse, rat, cow and human, and MARCKS related protein (MRP, also known as MacMARCKS, F52, or MLP), a 20 kDa protein highly expressed in brain, reproductive tissues, and macrophages [5–6]. MRP, while smaller than MARCKS, contains the same three evolutionarily conserved domains as MARCKS (N-terminus, MH2, PSD) that are critical to its function (see below). However, MRP does appear to differ somewhat from MARCKS in that the extent of cooperation between the PSD and myristoylation that controls association of these proteins with membranes appears weaker in MRP, and MRP may not cross-link actin filaments as efficiently as MARCKS [7]. This diverse family of abundant proteins with unique structures has several functions that play important roles in a variety of cellular processes.

STRUCTURE & MEMBRANE BINDING

MARCKS has been identified as an 87 kDa substrate for Protein Kinase C (PKC) that is expressed ubiquitously in eukaryotic cells. This rod-shaped protein contains three distinct evolutionarily-conserved regions: the N-terminal myristoylated domain, the multiple homology 2 (MH2) domain, and the phosphorylation site domain (PSD) or effector domain. The N-terminus displays a consensus sequence for myristoylation, which is a co-translational lipid modification attaching a myristic acid. Myristic acid is a C₁₄ saturated fatty acid that is attached via an amide bond to the amino group of the N-terminal glycine residue, which aids in the anchorage of MARCKS to the plasma membrane. The amino terminus makes up the first 24 amino acids of MARCKS. The highly basic PSD domain consists of 25 amino acids containing several serine residues that can be phosphorylated by PKC. It also serves as the site for MARCKS binding to and cross-linking actin filaments, and binding to calcium/calmodulin. The MH2 domain, of unknown function, resembles the cytoplasmic tail of the cation-independent mannose-6-phosphate receptor. Overall, MARCKS is an acidic protein rich in alanine, glycine, proline, and glutamic acids which contribute to pI’s ranging from 4.12 to 4.42 in various species [8].

MARCKS binds to plasma membranes in various cell types including macrophages [9], neurons [10], fibroblasts [11] and epithelial cells[12]. It has been determined that this interaction is dependent upon two major features that are important characteristics of MARCKS; the protein has the ability to bind to membranes due to the myristate insertion hydrophobically into the phospholipid bilayer, as well as electrostatic interactions between the basic PSD and acidic head groups of membrane phospholipids [13–15]. Both interactions are required, since neither myristic acid insertion alone nor electrostatic interactions are sufficient to anchor the protein to the plasma membrane [16–17].

The association of MARCKS protein with the plasma membrane is critical for the underlying mechanisms which ultimately lead to regulation and function. MARCKS cycles from the plasma membrane to the cytoplasm in various cell types [18–19]. Phosphorylation of MARCKS by PKC leads to the attachment of negatively charged phosphate groups to serine residues within the effector region (PSD), which weakens the electrostatic interactions. Since myristoylation on its own is not sufficient to anchor the protein, MARCKS is released and moves into the cytosol [20–21]. Subsequently dephosphorylation leading to the removal of phosphate groups by protein phosphatase I, protein phosphatase 2A, or calcineurin allows MARCKS to return to the plasma membrane [17, 22–23]. Other factors, such as binding of MARCKS to calcium and calmodulin, may further influence these reactions and MARCKS translocation

As alluded to earlier, in addition to PKC, the MARCKS PSD is also a target for Ca²⁺ and calmodulin (CaM) binding. Upon activation due to increased Ca²⁺ concentrations, calcium-bound CaM is able to bind MARCKS, which aids in the release of MARCKS from the plasma membrane. This process is reversible, so once intracellular Ca²⁺ concentrations decrease, CaM will not bind to MARCKS, facilitating binding of MARCKS to the plasma membrane again [24–26]. It has been found that CaM binds with high affinity to the effector region of MARCKS and [27–28], yet the phosphorylation of MARCKS by PKC significantly decreases this affinity [29–30]. Therefore, Ca²⁺/CaM regulation in conjunction with PKC activation can influence MARCKSbinding characteristics.

Another influence on binding and subsequent translocation of MARCKS is the cytoskeletal protein actin. In fact, the discovery that MARCKS binds and cross-links actin was a profound contribution to the understanding of MARCKS function in cells. Additionally, both the phosphorylation of MARCKS by PKC and Ca2+/CaM were found to inhibit cross-linking, connecting two possible signaling pathways that control the downstream target, actin [31].

Although the general concept of actin binding and cross-linking is agreed upon, there are several explanations of the particular mechanism by which this takes place. Firstly, MARCKS may contain a single actin-binding site in the effector domain which allows for actin cross-linking via dimerization. Secondly, MARCKS may contain two actin binding sites within the effector (PSD) domain, and thirdly the effector domain of MARCKS might bundle actin filaments by lowering electrostatic repulsion between the filaments [32–35]. While there is still some controversy over the mechanisms involved in the regulation of actin via MARCKS interactions, it is clear that actin is essential to the overall functioning of MARCKS.

MUCUS, INFLAMMATION AND ASTHMA:A ROLE FOR MARCKS

MARCKS protein has a prominent role in regulation of secretion in vario us cell types. Previous experiments revealed that the phosphorylation of MARCKS mediated by PKC is important in neurotransmitter release [36]. Similarly, MARCKS is pivotal in glucose-induced secretion in isolated rat pancreatic islets [37]. It was also found that, upon stimulation, the phosphorylation of MARCKS causes rapid and early release of adrenocorticotropin (ATCH) in ovine anterior pituitary cells [38], and stimulation of platelets by thrombin inducesMARCKS phosphorylation and serot onin release [39].

In studies from this laboratory, MARCKS protein has been found to be a key molecule regulating mucin secretion in airway epithelial cells. Pivotal to these studies is utilization of a synthetic peptide corresponding to the first 24 amino acids of the N-terminal region of MARCKS, named the MANS (Myristoylated N-terminal sequence) peptide. This peptide was developed at a time when there were no known reagents to inhibit MARCKS function, and the MANS peptide has profound effects on a number of cell functions, while a control missense peptide consisting of the same amino acids but arranged randomly (the RNS peptide) is without effect. In the first published studies using well-differentiated normal human bronchial epithelial (NHBE) cells in vitro in an air/liquid interface system, cells that were shown to be filled in their cytoplasm with membrane-bound granules containing mucin, pretreatment of the cells with MANS, but not RNS, results in attenuation of mucin hypersecretion in these cells in response to PKC activation [12], implicating MARCKS protein in the secretory response. In addition, the cycles of MARCKS phosphorylation/dephosphorylation and binding to actin and myosin were shown to be critical to the secretory response.

We then turned our attention to mucin hyperproduction and secretion in asthma. Goblet cell hyperplasia and metaplasia are well-characterized features of the asthmatic airway, and we looked at the possibility of the MANS peptide affecting mucin secretion in a well-defined model of allergic inflammation, the ovalbumin (OVA) sensitized mouse. OVA sensitized and challenged mice develop a goblet cell metaplasia and, in response to methacholine challenge, secrete large amounts of mucin. However, pretreatment of these mice with the MANS, but not RNS peptide, for 15 min via intratracheal instillation attenuates, in a concentration-dependent manner, mucin secretion into the airway lumen in response to methacholine aerosol. This study implicated MARCKS in the process of mucin secretion in vivoin alle rgically inflamed mouse airways[40].

Additional studies with this model were performed in which inhibition of mucin secretion in vivo in OVA sensitized and challenged mice in response to intratracheal administration of the MANS peptide was again demonstrated. Of great interest, these studies also showed that inhibition of mucin secretion via treatment with the MANS peptide correlates with a substantial and significant lowering of airway obstruction and resistance when the mice are subjected to pulmonary function testing [41]. Similar effects of MANS on mucin secretion and pulmonary function were demonstrated in work using the elastase instillation model of airway inflammation and goblet cell hyperplasia/metaplasia in mice [42]. Finally, in a translational study in which human airway epithelial cells were derived from asthmatic patients, the MANS peptide was shown to inhibit expression of the mucin gene, MUC5AC, in epithelium from these patients (but not in epithelial cells from control non-asthmatic patients) after infection with Mycoplasma pneumonia [43].

Given the above, an obvious question to ask is whether or not levels of MARCKS and/or phosphorylated MARCKS are increased in airways or lavage fluid from human asthmatics or other patients with inflammatory lung disease. We have tried to address this question but, given the ubiquity of MARCKS and its high expression in other cell types, there is too much background to be able to show that MARCKS levels are increased in either tissue or BAL fluid from diseased individuals. Immunohistochemistry, for example, reveals MARCKS to be highly expressed in normal airways and also in asthmatic and bronchitic airways, with very high background. Studies looking at MARCKS phosphorylation status in these patients are presently underway. The fact that the MANS peptide, which inhibits MARCKS function both in vitro and in vivo, is only effective in knocking down expression of MUC5AC in cultured airway epithelial cells from asthmatics, whereas it has no effect on MUC5AC expression in cells cultured from non-asthmatic normal controls as described above certainly suggests that MARCKS is either more highly expressed or more activated in asthmatic epithelium.

The precise intracellular mechanisms by which MARCKS regulates secretion have not been fully elucidated, but additional studies from this laboratory have implicated the PKCδ isoform [44], chaperone proteins such as cysteine string protein (CSP) and heat shock protein 70 (HSP70 ) [45] and a novel non-muscle myosin isoform, myosin V [46] in the MARCKS-regulated mechanism.

MARCKS AND INFLAMMATION

Excessive inflammation is a key component of asthma and other airway diseases, including chronic bronchitis, bronchiectasis, and cystic fibrosis. Influx into the lung and airways of neutrophils, esoinophils, and other leukocytes can result in severe tissue damage. Since we had shown that mucin secretion was regulated by MARCKS and could be alleviated by treatment with the MANS peptide, we speculated that perhaps degranulation of inflammatory cells, a process similar to mucin secretion in that membrane-bound granules are released by cells, also could be linked to a MARCKS-dependent mechanism. In studies with isolated human neutrophils and other human leukocyte cell lines, release of myeloperoxidase from neutrophils, eosinophil peroxidase from the eosinophil-like cell line HL-60 clone 15, lysozyme from the monocytic leukemia cell line U937, and granzyme from the lymphocyte natural killer cell line NK-92 were each attenuated by pre-incubation of the cells with MANS but not with the missense control peptide. The results indicate that MARCKS protein also may play an important role in degranulation of leukocytes, and thus inhibition of MARCKS could be anti-inflammatory[47].

Another potential therapeutic target in asthma and other inflammatory airway diseases is actual migration of leukocytes, especially neutrophils and eosinophils, into the lung in response to chemoattractants released in the airway, such as IL-8. Since MARCKS is known to be an actin-binding protein, it was examined as a potential regulator of neutrophil migration towards a chemoattractant. The MANS peptide, but not the RNS peptide, was shown to attenuate migration of isolated human neutrophils in vitro towards the chemoattractants fmlf, IL-8 or LTD4 [48]. Furthermore, pretreatment of mice with the MANS, but not the RNS peptide, greatly attenuatesthe increase in levels of pro-inflammatory cytokines (IL-6, KC) and neutrophil migration in a mouse model of airway inflammation induced by exposure to ozone [49]. The same results were seen in the elastase model of airway inflammation in mice, where levels of inflammatory cytokines (KC, IL-6, IL-1β, TNFα, MCP-1) as well as influx of neutrophils, eosinophils and lymphocytes, are significantly decreased in BAL fluid of animals pretreated with the MANS peptide [42]. Thus, MARCKS protein appears to play an important role in both migration of inflammatory cells into the lung and release of granule-stored inflammatory mediators by these cells.

CONCLUSIONS

Asthma and other inflammatory airway diseases are characterized by excess mucin production and secretion as well as influx of inflammatory cells into the lungs and airways and increased levels of pro-inflammatory cytokines and mediators. In both of these processes, mucin production and secretion as well as inflammatory cell influx and release of inflammatory mediators, MARCKS protein appears to play a major role, as inhibition of MARCKS function with a peptide identical to the MARCKS N-terminus attenuates significantly, and in some cases dramatically, all of these processes in vitro and in vivo, and also enhances lung function in murine models of asthma and allergic inflammation. Thus, MARCKS represents a potentially important therapeutic target for treatment of airway diseases characterized by mucus hypersecretion and inflammation. The mechanisms of MARCKS protein regulation of both of these processes are still under investigation, and certainly several other proteins that may associate or interact with MARCKS no doubt are integrally involved in these responses; their exact contributions and interactive dynamics remain to be elucidated.

Figure 1. Hypothetical mechanism whereby MARCKS regulates exocytosis in mucin-secreting (goblet) cells in the airways.

A single goblet cell is represented in the figure, with the airway lumen at the top. A) Under normal, constitutive conditions, the cell is filled with “secretory modules”, membrane bound granules containing mucin in their interior and a number of proteins in the granule membrane, including cysteine string protein (CSP), vesicle-associated membrane proteins-2 and 8 (VAMP 2,8), protein phosphatase type II (PP2A), and GTPases like rab3. Under normal, non-stimulated conditions, MARCKS protein is bound to the inner face of the plasma membrane. B) Upon stimulation of the cell, for example binding of a secretagogue to a membrane receptor, Protein Kinase C (PKC) translocates to the membrane and phosphorylates MARCKS, causing it to move to the cytoplasm. C) In the cytoplasm, MARCKS binds to the chaperone Heat Shock Protein 70 (HSP70), which targets it to the secretory module via specific HSP70 – CSP affinity. D) After binding to the secretory module, MARCKS is dephosphorylated by PP2A, allowing it to bind to the actin/myosin contractile system at the PSD while still attached to the granule membrane at the N-terminus. E) Contraction then moves the MARCKS-granule complex to the cell surface where, in conjunction with a number of docking and scaffolding proteins (SNAPs, syntaxins, Muncs, etc.) the granule membrane fuses with the plasma membrane and the granule contents (mucin) are released into the airway lumen via an exocytotic process.