This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Mast cells are not only involved in immediate-type allergic reaction, but also in innate immunity, inflammation, angiogenesis, and tissue remodelling, which are responsible for tissue-dwelling results. Increasing evidence suggests that mast cells play an important role in asthma pathogenesis, such as mast cell accumulation in the airways, or the infiltration of airway smooth muscle (ASM) of asthma patients by mast cells where the adhesion is a fundamental mechanism facilitating cellular cross-talk. Current research into mast cell biology has revealed an adhesion molecule, called cell adhesion molecule 1 (CADM1) that not only acts as glue but also seems to promote functional communication between nerve and mast cells and between smooth muscle and mast cells. CADM1, also called Necl-2, SgIGSF, TSLC1or SynCAM consists of three immunoglobulin-like motifs in the extracellular domain that mediate cell-cell adhesion, a transmembrane domain and a 47 amino acid C-terminal domain which interacts with other proteins via the protein 4.1-binding and PDZ-binding motifs. If either the FERM-binding or PDZ-binding domains are deleted, CADM1 is no longer able to mediate cell-cell adhesion or act as tumour suppressor in epithelial cells. CADM1 forms homodimers on the cell membrane and binds to adhesion receptors on fibroblasts. In this project, the biological roles of CADM1 present in human mast cells are studied using established techniques by overexpressing mutant CADM1. To generate CADM1 mutants, the entire cytoplasmic domain (aa 398-442), the FERM-binding domain (aa 398-410), the PDZ-binding domain (aa 432-442) are deleted. Another mutant is generated by single point mutation to replace the tyrosine by phenylalanine (aa 440). This project is looking at the expression changes in the apoptosis markers like Mcl-1 or Bim and the tyrosine kinase receptors such as c-Kit by overepression, downregulation and mutation of CADM1 in human mast cells. On the other hand, the extracellular domains of difference CADM1 isoforms presented in human mast cells are cloned and expressed to study the role they play in the molecular interactions.
I would like to thank my supervisor Dr. Mark Leyland, and research associates Professor Peter Bradding and Dr. Elena Moiseeva for their guidance and expertise. Many thanks to Dr.Xiaowen Yang, Dr. Sharad Mistry, Samarein Ahamed and Eva Jordan for their technical assistance.
A special thanks to my parents for their support.
List of figures
1.1 Cellular interactions between human lung mast cells and airway smooth muscle cells involved in bronchial asthma pathophysiology.
1.2 The molecular structure of CADM1 and Nectin.
1.3 Alternatively spliced isoforms of CADM1.
2.1 The structural differences of wild-type CADM1 and its mutants.
3.1 The effects of modulation of CADM1 in HMC-1 cells and human lung mast cells on expression of c-Kit and Mcl-1.
3.2 The expression of c-Kit and Mcl-1 in HMC-1 cells overexpressing either wild-type (SP4) or C-terminus mutants of CADM1.
3.3 Modulation of CADM1 affects HMC-1 cell viability.
3.4 Homotypic mast cell-cell adhesion affected by the C-terminus mutations of CADM1.
3.5 CADM1-mediated homophilic cell adhesion is affected by its overexpression and mutation in the C-terminus.
3.6 CADM1-mediated homotypic adhesions over 48 hours in IMDM.
3.7 Production of CADM1-SP4-ECD fusion proteins in E. coli at 37°C.
3.8 Production of fusion SP1-ECD fragment in E. coli at 37°C.
3.9 Production of fusion SP6-ECD fragment in E. coli at 37°C.
3.10 Production of fusion SP4-ECD in E. coli under lower temperatures.
3.11 Production of fusion SP1/ SP6-ECD in E. coli at 18°C.
3.12 Affinity purification of GST/His-CADM1-ECD.
3.13 Screening for the production of soluble fusion SP4 with a disulfide oxidase Erv1p pre-expression in E. coli.
3.14 Refolding screening for GST-/His-SP4-ECD fusion proteins.
Chapter 1 Introduction
1.1 Mast cells
The name of mast cells originally came from the Greek word "mastos", which means breast. In 1878, Paul Ehrlich thought of this name because he believed that the intracellular granules contained nutrients. Mast cells are 6-12 Âµm in diameter, and known for their prominent granules that are responsible of allergic diseases for decades. Mast cells, resident in all normal tissues, are believed to play significant roles in tissue homeostasis, wound healing, and host defence especially bacterial infection (BraddingHYPERLINK "#_ENREF_10" et al.HYPERLINK "#_ENREF_10", 2006).
Mast cells originate from pluripotent CD34+ hematopoietic progenitor cells in the bone marrow, circulate as undifferentiated mononuclear cells in the peripheral circulation, and eventually mature under local influences following migration into tissue. The maturation and differentiation of mast cells occur locally; following the migration of mast-cell precursors to the vascularised tissues or serosal cavities in which mast cells will eventually reside (Kitamura, 1989). The proliferation and differentiation of human mast cells are regulated by a variety of physiologically active molecules. Based on published results, stem cell factor (SCF) stimulates both the proliferation and maturation, while Thrombopoietin (TPO) and IL-9 only promote mast cell proliferation. IL-4 and IL-6 both inhibit the proliferation but stimulate the maturation of human mast cells. Factors like retinoids have negative effects on both the proliferation and maturation of mast cells (BraddingHYPERLINK "#_ENREF_8" et al.HYPERLINK "#_ENREF_8", 1992, BraddingHYPERLINK "#_ENREF_9" et al.HYPERLINK "#_ENREF_9", 1994). In 1986, two types of human mast cells were identified based on their content of tryptase together with skin chymotrptic proteinase (TC mast cells) or of tryptase alone (T mast cells) (IraniHYPERLINK "#_ENREF_38" et al.HYPERLINK "#_ENREF_38", 1986). These two types represent human mast cells distributed in connective tissues or at the mucosal surfaces, respectively.
Human mast cells play diverse roles in the human body. First of all, they can function as effecter cells. For example, mast cells are capable of killing pathogens and degrading potentially toxic endogenous peptides or compounds of venoms. They are also capable of regulating the number, viability, distribution and phenotype of structural cells including fibroblasts and vascular endothelial cells. Such functions can be observed during innate and adaptive immune responses. Secondly, mast cells also affect many aspects of the immune cells in terms of their recruitment, survival, development and functions. This so-called "immunomodulatory" function can be either beneficial or detrimental. On the positive side, mast cells can promote the migration, maturation, differentiation and function of immune cells by the secretion of tumour-necrosis factors (TNF), chemokines, histamines, leukotriene B4 (LTB4) and protoeases. Mast cells are able to present antigens to T cells (on MHC class I or II molecules) or increase antigen presentation by capturing Immunoglobulin-E (IgE) bounded antigens through FcÆRI and then undergoing apoptosis. With the secretion of TNF, interleukin-4 (IL-4) and IL-13 mast cells can promote the airway smooth muscle cells to produce cytokines and chemokines (BrightlingHYPERLINK "#_ENREF_12" et al.HYPERLINK "#_ENREF_12", 2003b). These are only a few examples of course. The tactical positioning of mast cells that is at the interface between the host and the external environment near blood vessels, lymphatic vessels, nerve fibres, and a set of immune cells allows mast cells to act quickly against any changes in the environment by communicating with different cells (WellerHYPERLINK "#_ENREF_93" et al.HYPERLINK "#_ENREF_93", 2011). On the other hand, mast cells are known as a promoter of inflammation. For instance, the chemokine receptor CXCR4 on mast cells mediates the migration of mast cells from skin to the draining lymph nodes induced by the ultraviolet irradiation, leading to immunosuppression (ByrneHYPERLINK "#_ENREF_15" et al.HYPERLINK "#_ENREF_15", 2008). Mast cells also suppress sensitization for contact hypersensitivity through ultra-violet-B-light-induced histamine production (GrimbaldestonHYPERLINK "#_ENREF_29" et al.HYPERLINK "#_ENREF_29", 2007). By secreting IL-10, mast cells suppress cytokine production by T cells and monocytes, pro-inflammatory cytokines and chemokines productions by keratinocytes, and increase dentritics cells' ability to decrease T-cell proliferation and cytokine production. Mast cells and mast cell-derived IL-10 are the key players in influencing many responses including inflammation, epidermal hyperplasia and skin ulceration, yet the pathways involved in these complicated processes remain to be elucidated.
Regarding the kinetics and function of mast cells in allergic diseases, an increasing number of circulating mast cell colony-forming cells was found in this allergic disorder (MwamtemiHYPERLINK "#_ENREF_62" et al.HYPERLINK "#_ENREF_62", 2001). Actually, mast cells are believed to be related closely to the onset and progression of various human diseases such as artherosclerosis, cardiomegaly, hypertension (HaraHYPERLINK "#_ENREF_31" et al.HYPERLINK "#_ENREF_31", 1999, HamadaHYPERLINK "#_ENREF_30" et al.HYPERLINK "#_ENREF_30", 1999), multiple sclerosis (SecorHYPERLINK "#_ENREF_79" et al.HYPERLINK "#_ENREF_79", 2000), and sudden infantile death syndrome (PlattHYPERLINK "#_ENREF_69" et al.HYPERLINK "#_ENREF_69", 1994, HolgateHYPERLINK "#_ENREF_34" et al.HYPERLINK "#_ENREF_34", 1994).
This hypothesis is supported by numerous experimental observations, indicating that mast cells not only are involved in allergic diseases, but also many other physiological, pathological and immunological processes such as tissue remodelling, wound healing, pathological fibrosis, arthritis, angiogenesis, and host reactions against neoplasia (Holgate, 2000). In the review by Kalesnikoff and Galli (2008), new evidences of the contribution made by mast cells to the pathology of some cardiovascular disorders and certain cancers at least in rodents were presented and they also brought up the importance of understanding the mechanisms behind those mast cell responses and how to exploit this understanding clinically (KalesnikoffHYPERLINK "#_ENREF_45" et al.HYPERLINK "#_ENREF_45", 2008).
In conclusion, it is necessary and extremely urgent to gain a better understanding of the development and function of mast cells in vivo as well as in vitro to analyze the pathophysiology of these mast cell-related diseases.
1.2 The role of mast cells in the pathology of asthma
Mast cells play a crucial role in the initiation of inflammation as a primary defence mechanism against pathogens. However, their chronic activation also leads to the pathophysiology of numerous diseases such as autoimmune disorders, allergy and asthma (GalliHYPERLINK "#_ENREF_26" et al.HYPERLINK "#_ENREF_26", 2010). Weller et al stated that the inappropriate responses by the components of the immune system, especially mast cells, to innocuous antigens could lead to allergy. The fact that mast cells produce IL-4 and IL-13 suggested that they might have the potential to influence the IgE production by B cells (WellerHYPERLINK "#_ENREF_93" et al.HYPERLINK "#_ENREF_93", 2011).
The discovery of IgE and its association with mast cell histamine release in the Gel and Coombs type I hypersensitivity reaction formed the initial understanding of the role of mast cells in asthma and acute allergic reactions. In the 1970s, the cross-linking of IgE on high-affinity receptors on mast cells was regarded as the mechanism for the variable airflow obstruction that happens in asthma, and was the major target for therapeutic development of drugs (HolgateHYPERLINK "#_ENREF_33" et al.HYPERLINK "#_ENREF_33", 1985). Nevertheless, treatment with antihistamines was disappointing. After the 1980s, allergic inflammation featured by eosinophil recruitment into tissues was shown to be essential in asthma pathology; thus inhaled corticosteroids that could inhibit eosinophilic inflammation became the most recognised therapy for asthma.
Recent research by Bradding et al has found out two major differences between asthma and eosinophilic bronchitis lacking airway hyperresponsiveness and airflow obstruction was found to be infiltrated of airway smooth muscle (ASM) by mast cells (BraddingHYPERLINK "#_ENREF_10" et al.HYPERLINK "#_ENREF_10", 2006). Once present in the ASM bundle, adhesion of mast cells to ASM cells is likely to be important for the retention of mast cells and the functional interaction between the two cell types (HollinsHYPERLINK "#_ENREF_35" et al.HYPERLINK "#_ENREF_35", 2008). Recent study on the interactions between mast cells and airway smooth muscle implied that mast cell infiltration of the airways in asthma depends on T cells, and TH2 cytokines from T cells and other sources act in mast cell expansion from circulating and tissue precursors, which could contribute to airway hyperresponsiveness in asthma (Robinson, 2004). However, the mechanism of mast cells activation in the asthmatic bronchial mucosa is extremely complicated because a diversity of stimuli including monomeric IgE alone, proteases (such as tryptase), cytokines (such as stem cell factor, TNF-Î± and IFN-Î³) complement adenosine, Toll-like receptor (TLR) ligands, neuropeptides and hyperosmolality can activate mast cells (BraddingHYPERLINK "#_ENREF_10" et al.HYPERLINK "#_ENREF_10", 2006). There is increasing evidence that the production of Th2 cytokines such as IL-4, IL-13 and IL-5 could lead to the development and maintenance of airway inflammation in asthma. Brightling et al (2003) have found out that airway smooth muscle mast cells expressed IL-4 and IL-13 but not IL-5, and airway smooth muscle cells expressed the IL-4 high affinity Î± chain of the IL-4 receptor and the IL-13RÎ±I and IL-13RÎ±II chains in asthma (BrightlingHYPERLINK "#_ENREF_11" et al.HYPERLINK "#_ENREF_11", 2003a). This is mediated in part through a molecule known as tumour suppressor in lung cancer 1 (TSLC1), also known as cell adhesion molecule 1 (CADM1) which is highly expressed by human lung mast cells (HLMCs). However, the CADM1-independent adhesion mechanism remains a mystery (Figure 1.1). CADM1 is believed to participate the adhesion of mouse mast cells to mouse fibroblasts which is dependent on an interaction between membrane-bound stem cell factor (SCF) and c-Kit/CD117 (ItoHYPERLINK "#_ENREF_42" et al.HYPERLINK "#_ENREF_42", 2003, AdachiHYPERLINK "#_ENREF_1" et al.HYPERLINK "#_ENREF_1", 1992). Based on the research done by Yang et al (2006), the effect of Ca2+ chelation and TSLC1 blockade on HMC-1 and HLMC adhesion to ASM implicated a novel mechanism which is CADM1 dependent and Ca2+-independent. And the Ca2+-dependent HLMC-ASM adhesion still needs further work (YangHYPERLINK "#_ENREF_98" et al.HYPERLINK "#_ENREF_98", 2006).
According to Kaur et al (2010), the mast cells resided with the ASM bundle express fibroblast markers and their number in the ASM bundle is closely related to airway hyperrespensiveness (AHR). In 2011, human lung mast cells were found to be able to modulate their own recruitment and smooth muscle functions locally in asthma (AlkhouriHYPERLINK "#_ENREF_2" et al.HYPERLINK "#_ENREF_2", 2011). All these studies provide new perspectives in the pathogenesis of asthma and might offer more effective approach for modern asthma treatment.
1.3 Cell adhesion molecule 1 (CADM1) and its roles in allergic diseases
Adhesive interactions between cells are a fundamental mechanism of cell communication, contributing to the accurate delivery of cell-cell signals. Cell adhesion molecules consist of four categories: the integrins, the selectins, the cadherins and the immunoglobulin superfamily (IGSF) (Hynes, 1999). IGSF is the largest and comprises of cell surface receptors such as major histocompatility (MHC), neural cell adhesion molecules (NCAM), intercellular adhesion molecules (ICAM), nectins and nectin-like molecules (AdachiHYPERLINK "#_ENREF_1" et al.HYPERLINK "#_ENREF_1", 1992). This superfamily not only acts as active adhesion molecules but also as cell surface recognition molecules involved in different cellular processes. In terms of the structural characteristics, IGSF contains highly-conserved immunoglobulin-like domains. The members of IGSF are Ca2+-independent and prefer homophilic and/or heterophillic interactions (BensonHYPERLINK "#_ENREF_4" et al.HYPERLINK "#_ENREF_4", 2000).
Cells of haematopoietic origin including mast cells can survive and function in both adherent and non-adherent status. CADM1 is an immunoglobulin protein first discovered in 1999 as a tumour suppressor in non-small cell lung cancers (TSLC1) with an extracellular domain homologous to immunoglobulin superfamily proteins (YagetaHYPERLINK "#_ENREF_96" et al.HYPERLINK "#_ENREF_96", 2002). It is also called Necl-2, SgIGSF, TSLC1, SynCAM1, or immunoglobulin superfamily 4A due to its diverse roles in different cell types discovered by independent research groups. It consists of three immunoglobulin-like motifs in the extracellular domain that mediate cell-cell adhesion, a transmembrane domain and a 47 amino acid C-terminal domain which interacts with other proteins via the FERM/protein 4.1-binding and PDZ-binding motifs (Figure 1.2). According to previous research, TSLC1 mediates intracellular adhesion through homophilic interactions. In 2003, Ito et al also identified CADM1 as a novel mast cell adhesion molecule, realising that CADM1 is an adhesion molecule common to nerves and mast cells. In the same year and the next, researchers of testis and brain found that this molecule also played an important role in spermatogenic cell attachment and synapse formation (HamadaHYPERLINK "#_ENREF_30" et al.HYPERLINK "#_ENREF_30", 1999, KomaHYPERLINK "#_ENREF_50" et al.HYPERLINK "#_ENREF_50", 2004).
1.3.1 Nectin-like molecule 2 (Necl-2) and cell-cell adhesion
The name of Necl-2 is closely connected with cell-cell adhesions and novel regulations of cell activities, such as cell polarization, movement and survival. The nectin and nectin-like (necl) families are consisted of four and five members, respectively. The word "nectin" comes from Latin word necto which means "to connect" (TakahashiHYPERLINK "#_ENREF_84" et al.HYPERLINK "#_ENREF_84", 1999). Necls are ubiquitously expressed in human epithelial tissues and involved in various physiological avtivities, but seldom detectable in fibroblasts such as NIH3T3, Swiss3T3 and L cells (ShingaiHYPERLINK "#_ENREF_80" et al.HYPERLINK "#_ENREF_80", 2003).
Nectins and nectin-like molecules contain one membrane distal Ig-V-type domain and two membrane-proximal Ig-C-type domains in their extracellular domains (SakisakaHYPERLINK "#_ENREF_74" et al.HYPERLINK "#_ENREF_74", 2004, TakaiHYPERLINK "#_ENREF_85" et al.HYPERLINK "#_ENREF_85", 2003). In addition, two types of binding interfaces in the C-C'-C''-D b-strands of the Ig-V-type domains are believed to be involved in both homotypic and heterotypic interactions according to previous work (Du Pasquier, 2004, FabreHYPERLINK "#_ENREF_22" et al.HYPERLINK "#_ENREF_22", 2002). Necl-2 interacts with Necl-2, Necl-1, nectin-3 and a class-I-restricted T-cell-associated molecule through cis and trans homotypic contacts or heterotypic adhesions in a Ca2+-independent manner (ShingaiHYPERLINK "#_ENREF_80" et al.HYPERLINK "#_ENREF_80", 2003, MasudaHYPERLINK "#_ENREF_56" et al.HYPERLINK "#_ENREF_56", 2002, GalibertHYPERLINK "#_ENREF_25" et al.HYPERLINK "#_ENREF_25", 2005). The study by Boles et al demonstrated that nature killer cells and CD8+ T cells could recognize Necl-2 through a class-I-restricted T-cell-associated molecule (CRTAM), and this CRTAM-Necl-2 interaction improves the cytotoxicity of certain lymphocytes (BolesHYPERLINK "#_ENREF_6" et al.HYPERLINK "#_ENREF_6", 2005), assuming that Necl-2 is a molecular target that allows the immunosurveillance network to distinguish tumour cells from normal cells. In their research, Boles et al (2005) argued that Necl-5, nectin 2 and their counter receptors CD226 and CD96 had the similar binding interfaces, however, necl-2 binded specifically and tightly to the cell surface receptor CRTAM under static and flow conditions through its first domain.
In 2008, Takai et al described Necl-2 as a cell-cell adhesion molecule which was localized on the basolateral membranes in epithelial cells, and also functioned in spermatogenesis and synapse formation. In addition it is a tumour suppressor in lung carcinoma. As a matter of fact, Necl-2 is down-regulated because of hypermethylation of its gene promoter or/and loss of heterozygosity at chromosome 11q23.2 in many types of cancer cells (FukamiHYPERLINK "#_ENREF_23" et al.HYPERLINK "#_ENREF_23", 2002). Recently, research has showed that Necl-2 interacted with Intergrin a6b4 in cis through its extracellular region, resulting in the stabilization of hemidesmosome structure in human colorectal adenocarcinoma Caco-2 cells (MizutaniHYPERLINK "#_ENREF_59" et al.HYPERLINK "#_ENREF_59", 2011). The function of Necl-5 is opposite to that of Necl-2 as it is found overexpressed in many tumour cells, yet Necl-5 also binds in cis with integrins and growth factor receptors contributing to the formation of a leading edge structure like lamellipodia, focal complexes and focal adhesions (MinamiHYPERLINK "#_ENREF_58" et al.HYPERLINK "#_ENREF_58", 2007, NagamatsuHYPERLINK "#_ENREF_63" et al.HYPERLINK "#_ENREF_63", 2008). Necl-5 is expressed in some fibroblasts and believed to enhance cell motility and survival (AmanoHYPERLINK "#_ENREF_3" et al.HYPERLINK "#_ENREF_3", 2008). In terms of its adhesion properties, Necl-2 interacts with scaffolding proteins like membrane-associated guantylate kinase (MAGuK) and Band 4.1 protein family members, whereas nectins are able to bind with the filamentous(F)-actin-binding protein afadin and mediate epithelial cell-cell adhesions through adherens junctions together with cadherins (TakaiHYPERLINK "#_ENREF_86" et al.HYPERLINK "#_ENREF_86", 2008, TogashiHYPERLINK "#_ENREF_90" et al.HYPERLINK "#_ENREF_90", 2011, OgitaHYPERLINK "#_ENREF_67" et al.HYPERLINK "#_ENREF_67", 2010).
1.3.2 SgIGSF in spermatogenic cells
During the discovery process of CADM1, some researchers cloned a new IGSF molecule from the mouse testis according to its homologous nucleotide sequence to neural cell adhesion molecule 1 (NCAM-1) and NCAM-2 cDNA, and they found that the mRNA of this molecule was expressed preferentially in spermatogenic cells. In 2001, SgIGSF (Spermatogenic immunoglobulin superfamily) was cloned from a cDNA library of adult mouse testes by Wakayama et al. Based on the finding that Sertoli cells did not express CADM1, Wakayama et al (2003) predicted that CADM1 on spermatogenic cells bound to another molecule on the Sertoli cells in a heterophillic manner, which was confirmed with a primary culture of Sertoli cells and a fusion protein containing the extracellular domain of CADM1 (sCADM1) (TakahashiHYPERLINK "#_ENREF_84" et al.HYPERLINK "#_ENREF_84", 1999). Thus the name spermatogenic immunoglobulin superfamily was given. Immunoglobulin superfamily number 4 (IGSF4) is known as a human homolog of SgIGSF because there is 98% identity between these two molecules at amino acid level (GomyoHYPERLINK "#_ENREF_28" et al.HYPERLINK "#_ENREF_28", 1999). Later on, IGSF4 was proved to be identical to tumour suppressor in lung cancer 1 (TSLC1) (PletcherHYPERLINK "#_ENREF_70" et al.HYPERLINK "#_ENREF_70", 2001).
CADM1 mRNA exists in two forms in the mouse testis: 2.1 and 4.5 kb (BrightlingHYPERLINK "#_ENREF_12" et al.HYPERLINK "#_ENREF_12", 2003b, WellerHYPERLINK "#_ENREF_93" et al.HYPERLINK "#_ENREF_93", 2011). The knockout mouse model for CADM1 was recently generated. Male knockout mice develop infertility due to defective spermatogenesis (ButterfieldHYPERLINK "#_ENREF_14" et al.HYPERLINK "#_ENREF_14", 1988, CastelloneHYPERLINK "#_ENREF_16" et al.HYPERLINK "#_ENREF_16", 2004, CoussensHYPERLINK "#_ENREF_17" et al.HYPERLINK "#_ENREF_17", 1999, BorrelloHYPERLINK "#_ENREF_7" et al.HYPERLINK "#_ENREF_7", 2008). One of the heterophillic binding partners of CADM1 has been proven to be a poliovirus receptor, another member of the immunoglobulin superfamily which is expressed in Sertoli cells (TakaiHYPERLINK "#_ENREF_86" et al.HYPERLINK "#_ENREF_86", 2008).
Based on the research by Ito et al, transcription of the SgIGSF gene was significantly regulated by microphthalmia transcription factor (MITF) in murine mast cells by examining the cell-to-cell adhesion phenotypes of cultured mast cells derived from 3 types of MITF mutant mice. Their report also showed the specific role SgIGSF played in mast cell adhesions (ItoHYPERLINK "#_ENREF_42" et al.HYPERLINK "#_ENREF_42", 2003), survival (ItoHYPERLINK "#_ENREF_43" et al.HYPERLINK "#_ENREF_43", 2004) and the attachment of mast cells to murine superior cervical ganglia (SCG) neurites in vitro (FurunoHYPERLINK "#_ENREF_24" et al.HYPERLINK "#_ENREF_24", 2005) and in the murine mesentery which subjected mast cells more susceptible to neural activation (ItoHYPERLINK "#_ENREF_40" et al.HYPERLINK "#_ENREF_40", 2007).
1.3.3 TSLC1 in non-small cell lung cancer
The tumour suppressor gene, TSLC1, which was identified by functional complementation through the suppression of tumour formation by the human lung adenocarcinoma cell line A549 (KuramochiHYPERLINK "#_ENREF_52" et al.HYPERLINK "#_ENREF_52", 2001), was reported to directly associate with DAL-1, a gene product of another lung tumour suppressor from the protein 4.1 family (YagetaHYPERLINK "#_ENREF_96" et al.HYPERLINK "#_ENREF_96", 2002). Based on their experiments, the cytoplasmic domain of TSLC1 harbours a sequence of 10 amino acids which matches the protein 4.1 binding motif perfectly. In addition, redistribution of both TSLC1 and DAL-1 to the newly generated membrane ruffling areas implicated that these proteins are also involved in cell motility through the actin rearrangement. Biochemical analysis has revealed that TSLC1 is an N-linked glycoprotein with a molecular mass of 75 kDa, and it can form homodimers by cis interaction within the plane of the cell membranes (MasudaHYPERLINK "#_ENREF_56" et al.HYPERLINK "#_ENREF_56", 2002). In 2003, the critical role of the protein 4.1 (FERM)-binding and PDZ-interacting domains of TSLC1 in tumour suppressor in non-small lung cell cancer was reported by Mao et al. Their results showed that the deletion of these two motifs abrogates the tumour suppressor activity of TSLC1 in murine cells, and the cytoplasmic domain of TSLC1 plays an important role in cell adhesion activity and proliferation in an anchorage-dependent manner (MaoHYPERLINK "#_ENREF_54" et al.HYPERLINK "#_ENREF_54", 2003). Not until recently, the crystal structure of a complex between the DAL-1 FERM domain and a portion of the TSLC1 cytoplasmic domain was revealed, which indicates that DAL-1 FERM domain binds TSLC1 in the C-lobe instead of the Î±-lobe (BusamHYPERLINK "#_ENREF_13" et al.HYPERLINK "#_ENREF_13", 2011).
1.3.4 SynCAM1 and its synaptic function
Biederer et al (2002) have shown that the SynCAM (Synaptic cell adhesion molecule) is a brain-specific, immunoglobulin domain-containing protein that binds to intracellular PDZ-domain proteins and functions as a homophilic cell adhesion molecule at the synapse. They stated that the expression of the isolated cytoplasmic tail of SynCAM in neurons inhibited synapse assembly, while expression of full-length SynCAM in nonneural cells induced synapse formation by cocultured hippocampal neurons with normal release properties (YangHYPERLINK "#_ENREF_98" et al.HYPERLINK "#_ENREF_98", 2006). In 2006, Biederer demonstrated that a family of genes homologous to SynCAM1 comprising of four genes found only in vertebrates by bioinformation characterisation. Their structure contains three Ig-like domains, a single transmembrane region, and a short cytosolic tail with a protein 4.1 interaction sequence and a PDZ type II motif. Alternative splicing generates isoforms of SynCAM proteins, yet their interaction motifs in the cytosolic sequence are highly conserved among all four members indicating its unique roles in function. Last but not least, SynCAM1 is the family member with the highest degree of alternative splicing, which leads to its functional variability.
1.4 CADM1 isoforms
According to Bierderer (2006), human and mouse genes that encode SynCAM1 consist of 12 and 11 exons, respectively. This difference in exon number is resulting from human exon 10, which is missing in transcripts of mouse SynCAM1. In this case, the cell adhesion molecules are actually a family of transmembrane proteins that are connected with several pathological and physiological processes, which has been discussed by many researchers. Until 2007, researchers have identified four CADM/SynCAM genes in tetrapod vertebrates which encode proteins with a strict conservation of their structural organization(Biederer, 2006). PietriHYPERLINK "#_ENREF_29" et al.HYPERLINK "#_ENREF_29" (2008) have managed to identify six orthologs of CADM in zebrafish by using mammalian family members as templates, and they compared those six isoforms with murine CADMs. Not surprisingly, the C-terminal portions of the CADMs, including the intracellular protein-protein interaction domains are highly conservative, which indicates again that both the FERM/protein 4.1B and the PDZ binding domains play extremely significant roles in function (MayrHYPERLINK "#_ENREF_57" et al.HYPERLINK "#_ENREF_57", 2003). However, genes encoding CADM family members are not found in invertebrates up to date.
Among the four family members in human, CADM1 is particularly open to alternative splicing which could lead to functional variability. As CADM1 is widely expressed in various epithelial tissues, different tissues are detected with different CADM1 isoforms. For instance, in the brain sample deglycosylated CADM1 had a molecular mass of about 45kDa, while in the mouse lens epithelium, the deglycosylated CADM1 had a mass of more than 50kDa (De MariaHYPERLINK "#_ENREF_18" et al.HYPERLINK "#_ENREF_18", 2011). According to their report, lenses from CADM1-null animals seemed normal in overall observation. However, the changes in cell shape might suggest that the organization of the lens fiber cell cytoskeleton was altered in the CADM1-knockout lens.
The basic molecular structures of alternatively spliced isoforms of CADM1 are shown in Figure 1.3. Based on the paper by Biederer, the splice product 4 (SP4) and SP1 both contain a highly threonine-rich sequence which is almost full of potential O-glycosylation sites only that SP1 is slightly extended by the potential O-glycosylation sites. SP2 shares a partial sequence of SP1 extended by a few potential O-glycosylation sites, while SP3 is short of these O-glycosylation sites. SP5 exists as a soluble form of the CADM1 extracellular domain. Recently, a unique splicing variant of CADM1 containing additional extracellular fragments corresponding to exon 9 in addition to SP4 which is common in epithelia was detected in small cell lung cancer (SCLC) tumours and testis by (KikuchiHYPERLINK "#_ENREF_46" et al.HYPERLINK "#_ENREF_46", 2012), suggesting that CADM1 enhances the malignant features of SCLC and possibly provides a promising target for both the diagnosis and treatment of SCLC.
On the other hand, the three functional and non-functional isoforms of CADM1 were successfully cloned from human lung mast cells and human mast cell lines (MoiseevaHYPERLINK "#_ENREF_61" et al.HYPERLINK "#_ENREF_61", 2012). A new isoform of CADM1 namely SP6 was identified by Moiseeva et al, and CADM1 has an undeniable impact on human mast cell survival and aggregation. Much less is known about human CADM1 compared to murine CADM1, yet CADM1 seems to be an attractive target for various mast-cell dependent diseases due to the specific CADM1-induced signalling in human mast cells.
1.5 Aims of prospect
The aims of this project involved two main aspects. One is identification of the effects of CADM1 overexpression, knockdown and mutation on human mast cell homotypic adhesion, survival and apoptosis, as well as on c-Kit tyrosine kinase receptor (CD117) expression. Secondly, different extracellular domains (ECD) of CADM1 isoforms presented in human mast cells were cloned and purified as inclusion bodies in E. coli, followed by optimising the strategies in order to produce soluble and functional target proteins for future examinations on how the structural differences in ECD affect their molecular interactions.
Chapter 2 Materials and Methods
2.1 Cell culture
2.1.1 HLMC isolation and culture
Human lung mast cells were purified from macroscopically normal lung using the method described by Hollins et al (2008). Final mast cell purity was >99%.
Freshly isolated HLMCs were used for cloning and the study of CADM1 expression. Stabilised 1- to 2-week-old HLMCs, adapted to cell culture, were used for adenoviral transductions.
2.1.2 HMC-1 culture
The human mast cell line HMC-1.1 (V560G) was maintained in Iscove's Modified Dulbecco's Medium (IMDM) with 10% iron-supplemented foetal calf serum (FCS) as described previously (DuffyHYPERLINK "#_ENREF_21" et al.HYPERLINK "#_ENREF_21", 2001, SanmugalingamHYPERLINK "#_ENREF_76" et al.HYPERLINK "#_ENREF_76", 2000).
2.1.3 E coli. Strains and media
E. coli strain BL21 (DE3) was used for expression studies in all experiments unless stated otherwise. Cells were grown in Lysogeny Broth (LB) media supplemented with 50 mg/ml Kanamycin, 50 mg/ml Amplicillin or 35 mg/ml Chloramphenicol when necessary.
2.2 Primary antibodies
For immunoblotting, anti-CADM1 3E1 IgY mAb from Medical and Biological Laboratories, Japan was employed, an antibody that reacts against the extracellular domain of SynCAM on Western Blotting.
Anti-c-Kit (E1) is a mouse monoclonal antibody raised against aa 23-322 mapping near the N-terminus of c-Kit of human origin (Santa Cruz Biotechnology). HRP-conjugated anti-b-actin was also purchased from Santa Cruz Biotechnology. Antibodies against Bim (mAb C34C5) and Mcl-1 (polyclonal #4572) were purchased from New England Biolabs.
2.3 Western blot analysis
Mast cell lysates were prepared using a lysis buffer containing 1% non-ionic detergent Nonidet P-40 (NP-40), 50mM Tris (pH=7.4), 250mM NaCl, 5mM EDTA, 50mM NaF, 0.02% NaN3, 1mM Na3VO4 and a protease inhibitor cocktail supplemented with 1mM PMSF prior to use. Cell pellets were lysed in cell lysis buffer for 15 minuters on ice with vortexing 15s to 30s intervals. The cell extracts were then centrifuged at 13,000 rpm for 10 min at 4 °C to remove insoluble materials and the supernatants were collected. The protein concentration of each sample was determined by the DC protein assay. For western blotting, cell lysates were suspended in a sample buffer containing 50 mM dithiothreitol (DTT) and incubated at 70°C for 10 min. The samples were fractionated using 4-12% gradient SDS-PAGE, transferred to a nitrocellulose membrane, and blotted with appropriate primary antibodies. The binding of the primary antibody was detected with Western blotting Substrate using a peroxidise-conjugated secondary antibody.
SDS-PAGE and immunoblotting were performed using the NuPAGE electrophoresis system (Invitrogen, USA). Blots were cut into horizontal strips with proteins of a close range of molecular weights, probed with appropriate antibodies. Alternatively, whole blots were sequentially probed with antibodies from different species. Protein bands on exposed films were quantified using a Syngene system (Syngene, UK).
2.4 Cloning and analysis of CADM1 isoforms
CADM1 in bacterial clones p24 (SP4), p33 (SP1) and p489 (SP6) were from previous research (MoiseevaHYPERLINK "#_ENREF_61" et al.HYPERLINK "#_ENREF_61", 2012). Total DNA was purified using the QIAprep Spin Miniprep Kit (QIAGEN) and quantified by Absorbance at 260nm. The extracellular domains (ECM) of these isoforms were subcloned into E. coli DE3 expression vectors pLEICS-04 (GST-tagged) and pLEICS-03 (His-tagged), or mammalian expression vectors both with histidine tags: pLEICS-12 and pLEICS-49 by the PROTEX at the University of Leicester (http://www2.le.ac.uk/departments/biochemistry/research-groups/protex). Then the subclones were sequenced for verification. All DNA sequencing was done by the PNACL at the University of Leicester (http://www.le.ac.uk/mrctox/pnacl/).
2.5 Homophilic cell aggregation assay
Human mast cells, dispersed as a single cell suspension, were allowed to aggregate in a 96-well plate for 3h, overnight and 48h in the incubator at 37°C. The images were captured on EVOS digital inverted microscope (www.amgmicro.com), and were viewed and analysed by CellF imaging software (Olympus). The aggregates were outlined and cross sectional areas of the largest 5 aggregates per photo of a well were expressed in or pixels2 or arbitrary units. The size of cell aggregates in each condition was estimated as mean of the 5 largest aggregates in 4 wells, respectively.
2.6 Cell viability
Cell viability was examined by measuring cell number and caspase-3 activity as described previously (MoiseevaHYPERLINK "#_ENREF_60" et al.HYPERLINK "#_ENREF_60", 2007). Cell number was estimated using the ATPlite 1step kit (Perkin-Elmer, USA) using a calibration curve with a known number of HMC-1 cells, with luminescence detected in a 96-plate reader. Caspase-3 activity was measured as accumulation of a specific fluorogenic substrate using the Apoptosis Homogenous Caspase-3/7 assay (Promega, USA), detected in a plate reader and expressed in fluorescent units/cell (FU/cell).
2.7 Manipulation of CADM1 expression
To generate dominant-negative CADM1 mutants, the entire cytoplasmic domain (aa 398-442), the FERM-binding domain (aa 398-410) or the PDZ- binding domain (aa 432-442) were deleted, and they were named C397, PDZ, and FERM respectively in this dissertation. Deletion of the cytoplasmic and PDZ-binding domains was produced by PCR amplification while specific deletion of FERM-binding domains was generated using splicing by overlap extension PCR. A single point mutation was designed on aa 440, allowing tyrosine (Y) replaced by phenylalanine (F), thus named Y440F. Their sequence differences are shown in Figure 2.1.
Cloned wild-type CADM1 (SP4) and mutants were inserted into a replication-incompetent Ad5 adenovirus (Biofocus DPI, Leiden, Holland) which was shown to allow reliable overexpression in these cells (WykesHYPERLINK "#_ENREF_95" et al.HYPERLINK "#_ENREF_95", 2007). Adenoviruses containing SP4, GFP, PDZ, FERM and Y440F were adopted to transduce human mast cells. CADM1 sh RNAs (sh3, sh4, sh5) were designed by BioFocus outside areas of alternative splicing as stated previously (MoiseevaHYPERLINK "#_ENREF_61" et al.HYPERLINK "#_ENREF_61", 2012). Mutant CADM1 expression in HMC-1 cells 6 days after transduction resulted in visible expression at variable levels, similar to the previously reported (WykesHYPERLINK "#_ENREF_95" et al.HYPERLINK "#_ENREF_95", 2007). Transductions of HMC-1 cells and HLMCs with a multiplicity of infection of 50 IU/cell (17Ã-3 infection unit/cell for mixed sh RNAs) for 6 days were used in all experiments, unless stated otherwise.
For each transformation, 1 ml plasmid DNA was added to 100 ml competent E. coli BL21 (DE3) cells and mixed gently with the pipette tip, and incubated on ice for 30 minutes. Then the cells were heat shocked at 42°C for 90 seconds, and incubated on ice for another 2 minutes. Cells were then grown in LB broth at 37°C with shaking. Finally, 200 ml of transformed cells were plated onto a LB agar plate containing appropriate antibiotics with antiseptic techniques, and left in the incubator overnight at 37 °C. The next day, one positive colony of each transformation was picked and grown overnight in LB media with antibiotics at 37 °C for further storage and analysis.
All the proteins expressed in E. coli were analysed by 10% SDS-PAGE. After SDS-PAGE, gels were rinse once with distilled water, and then stained by Coomassie blue for 30 minutes. Gels were then de-stained with destain solution containing 25% methanol, 10% acetic acid and distilled water till clear background was obtained. Gels were scanned and analyzed later.
2.10 Optimization of fusion protein expression and solubility test
After transformation, 1 positive colony of each transformation was picked for overnight incubation in 5 ml LB medium with appropriate antibiotics. The next day, 0.4 ml of the overnight cultures was inoculated into 20 ml fresh LB with antibiotics. When the absorbance at 600nm reached 0.5-0.8, 1 ml sample was taken out into an eppendorf tube, miscrofuged at 13,000 rpm for 1 minute to remove the media and frozen at -80°C as a zero time-point negative control. IPTG was added to the remaining culture to a final concentration of 1mM, 0.1 mM or 0.05 mM followed by incubating with shaking. 1 ml samples were taken out at 1 hour, 2 hour, 4 hour and overnight after adding IPTG. Cells were harvested as before and kept at -80°C for further SDS-PAGE analysis.
In order to test the solubility of the fusion proteins, cell pellets from overnight induction with IPTG were resuspended in 0.5 ml Tris-NaCl buffer (pH=8.0) and sonicated on ice by Soniprep 150 (4Ã-10s bursts with 30-60s cooling in between). 55ml of 10% Triton X-100 was added to lysed cells and incubated for 15 minutes on ice to improve the solubility. Soluble proteins were separated by microfugation at 13,000 rpm for 10 minutes at 4°C. Samples of the supernatants (soluble proteins) and the pellets (insoluble proteins) were analysed by 10% SDS-PAGE.
2.11 Affinity purification of fusion proteins
2.11.1 GST pull-down assay
GST-CADM1 fusion proteins expressed in E. coli were purified using glutathione Sepharose 4B. Subcultures from overnight incubation were induced by 0.1 mM IPTG overnight, and then the cell pellets were resuspended in 1 ml TN buffer compsed of 20mM Tris-HCl and 100mM NaCl (pH=8.0). Cells were lysed by sonication, and subjected to centrifugation to separate soluble and insoluble fractions. The glutathione-Sepharose beads were resuspended in TN buffer after washed three times. Cell lysates were added onto glutathione-Sepharose beads and incubated with mixing for 60 minutes in the cold room. After washing three times with TN buffer, the beads were heated at 95°C and subjected to SDS-PAGE analysis.
2.11.2 Immobilised cobalt resin affinity purification
For His-tagged fusion proteins, cells were induced and lysed as before. The insoluble pellets were solubilised in 8M urea (pH=8.0). Both the cell lysates and solubilised pellets were mixed with HIS-select cobalt affinity resin beads (Sigma-Aldrich, USA) for 60 minutes in the cold room followed by SDS-PAGE analysis.
2.12 Soluble expression of recombinant proteins in E. coli
To help the disulfide bond formation in E. coli, a helper plasmid (pFH255) expressing a sulfhydryl oxidase Erv1p from the inter-membrane mitochondrial space of S. cerevisiae, a natural catalyst of de novo disulfide bond formation was transformed into the cytoplasm of BL21 (DE3) together with our cloned genes of interest (NguyenHYPERLINK "#_ENREF_66" et al.HYPERLINK "#_ENREF_66", 2011).
For expression in LB media, E. coli strains containing expression vectors were streaked out from glycerol stocks at -70°C onto LB agar plates containing suitable antibiotics to allow selection. The next day one colony from these plates were used to inoculate 5 ml of LB media, containing suitable antibiotics, and grown overnight at 30 °C, 200 rpm. This overnight culture was used to seed a 50 ml culture of LB containing suitable antibiotics in a 250 ml conical flask to an optional density of 0.05 at 600 nm (OD 600). This culture was grown at 30 °C, 200 rpm until the OD600 reached 0.4 at which point the sulfhydryl oxidase Erv1p production was induced by the addition of 0.5% w/v arabinose followed 30 minutes, later by 1mM IPTG to induce affinity tagged CADM1 expression. The cells were then grown for a total of 4 hours post induction at 30 °C, 200 rpm and the final OD600 measured (GilfillanHYPERLINK "#_ENREF_27" et al.HYPERLINK "#_ENREF_27", 2006). The cells were collected by centrifugation and lysed by sonication as mentioned before, followed by SDS-PAGE analysis and solubility tests.
2.13 Refolding screening kit
The QuickFold refolding screening kit (Athena Enzyme Systems, USA) was used to optimise the refolding conditions of the target proteins after denaturation by 8M urea (pH=8.0). All the procedures were done according to the manufacture's manual, and successful refolding was assessed by 10% SDS-PAGE (JoHYPERLINK "#_ENREF_44" et al.HYPERLINK "#_ENREF_44", 2006, HuangHYPERLINK "#_ENREF_36" et al.HYPERLINK "#_ENREF_36", 2008).
2.14 Data analysis
Analysis was performed using GraphPad Prism 5 software. All data are presented as the mean+/- SE. Differences among the groups were analyzed using an one-way ANOVA, followed by Dunnett's test to determine whether the groups were different from a control group, or Bonferroni's test to compare multiple groups.
Chapter 3 Results
3.1 Characterisation of the roles that SP4 and its Carboxyl-terminal domains play in human mast cell biology.
The cytoplasmic domain of CADM1 contains a juxtamembrane sequence with protein 4.1/FERM binding motif and a class II PDZ binding motif (RyderHYPERLINK "#_ENREF_73" et al.HYPERLINK "#_ENREF_73", 2008). There is strong evidence suggesting that the C-terminal motifs of CADM1 have significant functions in the suppression of tumour invasion and metastasis, but their precise roles remain elusive. The PDZ binding domain interacts with the scaffolding molecules such as MAGuKs (SalvatoreHYPERLINK "#_ENREF_75" et al.HYPERLINK "#_ENREF_75", 2006, SoucekHYPERLINK "#_ENREF_82" et al.HYPERLINK "#_ENREF_82", 2007); while FERM binding domain interacts with the protein 4.1 family such as protein 4.1B/DAL-1 that connects CADM1 to the actin cytoskeleton during actin reorganization (YagetaHYPERLINK "#_ENREF_96" et al.HYPERLINK "#_ENREF_96", 2002). In a three-dimensional collagen culture, both the FERM motif and the PDZ binding motif were found to be responsible for the lateral localization of CADM1, yet only the deletion of the PDZ binding domain shifted the subcellular localization of CADM1 from the lateral membrane to the basal surface of the Madin-Darby canine kidney (MDCK) cysts (RyderHYPERLINK "#_ENREF_73" et al.HYPERLINK "#_ENREF_73", 2008). In addition, the cytoplasmic domain of CADM1 is demonstrated to directly associate with the PDZ domain of Tiam1, T-lymphoma invasion and metastasis 1 (SchneiderHYPERLINK "#_ENREF_78" et al.HYPERLINK "#_ENREF_78", 1977), and induce formation of lamellipodia through Rac activation in Adult T-cell Leukaemia cell lines (ScarpinoHYPERLINK "#_ENREF_77" et al.HYPERLINK "#_ENREF_77", 2000). Although CADM1 does not bind to Afadin as nectins do, it can bind a membrane guanylate kinase Pals2 through the C-terminal consensus motif of four amino acids and the PDZ domain of Pals2 intracellularly, mediating the localizations of transmembrane proteins in Caenorhabditis elegans (SoucekHYPERLINK "#_ENREF_82" et al.HYPERLINK "#_ENREF_82", 2007).
The roles of CADM1 in mast cell survival, cell adhesion and localization of transmembrane proteins in mammalian cells are greatly connected with the C-terminal motifs, which was investigated in the first half of this project.
3.1.1 SP4 overexpression and CADM1 downregulation affect HLMC and HMC-1 viability and the expression levels of c-Kit and Mcl-1.
The first aim is to re-examine the earlier work on SP4 trasductions into HLMCs, which demonstrates that SP4 is a significant survival factor in human mast cells (MoiseevaHYPERLINK "#_ENREF_61" et al.HYPERLINK "#_ENREF_61", 2012). Similarly apoptosis-related protein expression in previously transduced HMC-1 cells which were stored at -80°C was analysed. After thawing and lysing the cells, protein amounts were normalised and then analysed by Western blot.
CADM1-SP4 was present as a single band of approximately 105 kDa in HLMCs and HMC-1 cells because it is a highly-glycosylated protein with a protein core of about 50 kDa and dual N- or O-glycosylation among different isoforms.
Mcl-1 band intensities in combined downregulation of CADM1 became slightly weaker when compared to control groups (GFP) and SP4-overexpressing groups in pooled HLMCs (Figure 3.1). No doublet bands of Mcl-1 were observed from the blots as cells were not cultured in deprived of serum indicating no modifications of Mcl-1 at this point. On the other hand, there seems to be no significant differences in Mcl-1 levels in HMC-1 and HLMCs between SP4-overexpressing and control groups (Figure 3.1). In terms of c-Kit expression, SP4 downregulation resulted in decreased c-Kit levels compared to SP4 overexpression in HLMCs (Figure 3.1). No changes were observed between HMC-1 cells transduced with GFP viruses and with SP4 viruses (Figure 3.1). In conclusion, the normal expression and function of CADM1 is essential to mast cell survival as previous described.
Figure 3.1 The effects of modulation of CADM1 in HMC-1 cells and human lung mast cells on expression of c-Kit and Mcl-1.
From left to right, lane 1 and lane 3 were untransduced HMC-1 and pooled HLMCs from Donor 616, respectively. HMC-1 cells were transduced with adenoviruses carrying SP4 (lane 2). Pooled HLMCs from donors (D616) and (D673+D674) were transduced either with CADM1 sh mixed RNAs (lane 4), GFP (lane 5), SP4 (lane 6) or CADM1 sh mixed RNAs (lane 7) viruses. Cell lysates were then subjected to 4-12% Gradient SDS-PAGE (25 mg/lane) followed by western blotting.
3.1.2 The effects of various C-terminal protein-interaction domains on c-Kit and Mcl-1 expression in HMC-1 cells.
Various constructs were made with CADM1 C-terminus mutations; either the FERM binding domain or the PDZ binding domain was deleted from the wild-type SP4 (referred to as FERM or PDZ respectively), and a single point mutation was introduced in the PDZ binding domain at aa 440 (referred to as Y440F). HMC-1 cells transduced with adenoviruses containing GFP, SP4, FERM, PDZ and Y440F were incubated in DMEM at 37 °C for 6 days followed by Western blot analysis. Three different SDS-PAGEs were used to optimise the Western blotting results. HMC-1 cells transduced with GFP viruses were used as a negative control in this experiment.
The difference in the molecular weight of CADM1 mutants is not significant with SP4 being the heaviest, therefore the CADM1 bands appeared at the similar positions with SP4 bands being slightly higher in all blots. However, the intensity of CADM1 bands was much stronger with overexpression compared to GFP, indicating successful transductions. The expression levels of Mcl-1 in transduced cells were quite low thus difficult to tell the differences among different mutants. c-Kit showed up as a doublet band because of alternative splicing at its gene exon/intron junction of exon 9, which leads to the presence and absence of a four amino acid sequence GNNK (glycine-asparagine-asparagine-lysine, codons 510-513) in its juxtamembrane domain (SumimotoHYPERLINK "#_ENREF_83" et al.HYPERLINK "#_ENREF_83", 2006, TanHYPERLINK "#_ENREF_87" et al.HYPERLINK "#_ENREF_87", 2006). And the GNNK- isoform exists dominantly in both human and mouse (TheoharidesHYPERLINK "#_ENREF_89" et al.HYPERLINK "#_ENREF_89", 2004). The Kit levels expressed in the transduced HMC-1 cells seemed similar among all different groups compared to the control group (GFP) (Figure 3.2). The band intensities of Mcl-1 appeared almost undetectable in this case, maybe due to the choice of HMC-1 cells. The amount of b-actin indicated equal loading amount of proteins in each lane.
The Western blot results indicate that neither deletion of the FERM binding domain nor deletion of the PDZ binding domain exerts significant effects on c-Kit and Mcl-1 expression in HMC-1 cells. Also the replacement of tyrosine (Y) at aa 440 by phenylalanine (F) in the PDZ binding domain seems to make no significant difference either.
Figure 3.2 The expression of c-Kit and Mcl-1 in HMC-1 cells overexpressing either wild-type (SP4) or C-terminus mutants of CADM1. (a) HMC-1 cells, previously transduced with adenoviruses containing GFP (control), SP4 and C-teminus mutants of CADM1 (FERM, PDZ, Y440F), were thaw on ice and lysed by NP40 cell lysis buffer. Protein expression levels were detected by 7% SDS-PAGE (left-hand side blots) and 4-12% Gradient SDS-PAGE (right-hand side blots) followed by western blotting(30mg/lane). (b) HMC-1 cells were transduced with adenoviruses carrying GFP, SP4, and different C-terminus mutants of CADM1 (FERM, PDZ, Y440F) and incubated at 37° C for 6 days. Cell lysates were then analysed by 7% SDS-PAGE (30mg/lane) and western blotted with relevant antibodies to visualize protein bands. The results shown are representative of three independent experiments.
3.1.3 The role of CADM1 C-terminal domains in HMC-1 cell viability
HMC-1 cells transduced with SP4, PDZ, FERM or Y440F survived to a similar extent (no statistically significant difference) over 48 hours in the absence of survival factors (Figure 3.3a). The cell number of each well was detected but not normalised to an equal number (Figure 3.3b). The caspase-3/7 activity of the PDZ mutant after 48 hours in IMDM alone was not significantly higher than of the SP4-overexpressing or the other two mutants (Figure 3.3a).
The calcium ionophore A23187 is an inhibitor of the membrane events associated with immunologic stimulation by transferring the divalent cations across the cell membrane (FabreHYPERLINK "#_ENREF_22" et al.HYPERLINK "#_ENREF_22", 2002); it can also induce histamine release by human mast cells dependent on its concentration (Du Pasquier, 2004). The effect of the calcium ionophore A23187 on the viability of HMC-1 cells with SP4, PDZ, FERM and Y440F overexpression or control viruses was studied. Surprisingly, cells overespressing SP4 demonstrated worse survival compared to other groups (Figure 3.3c, 3.3d).
Figure 3.3 Modulation of CADM1 affects HMC-1 cell viability. Transduced cells were washed and then incubated in IMDM alone in a 96-well plate for 45 hours at 37 °C. The caspase 3/7 activity (a) and cell number per well was measured (b) in two experiments in quintuplicate. Caspase 3 activity was expressed in fluorescent units/cell. Transduced HMC-1 cells were washed and incubated in IMDM alone (c) or IMDM with 2 mM A23187 (d) in a 96-well plate for 24 hours at 37 °C. Cell number per well was measured as mentioned in (d). GFP-transduced cells were treated as control. Two experiments were performed in quintuplicate, respectively.
3.1.4 CADM1 mediates homotypic human mast cell adhesion.
To examine the role of CADM1 in homotypic mast cell adhesion, transduced HMC-1 cells were washed and then cultured for 3 hours in IMDM with growth factors (10% FCS) for analysis of cell aggregation. HMC-1 cells transduced with SP4 and control viruses (GFP) formed cell aggregates from single cell suspension similarly, whereas cells with Y440F mutant were present mostly as single cells (Figure 3.4a). In order to estimate the sizes of the aggregates, cross-sectional areas of the largest aggregates were compared. SP4-overexpressing cells displayed larger aggregates compared to cells transduced with PDZ, FERM and Y440F mutants (Figure 3.4b).
Repeat experiment was done from the transductions of HMC-1 cells to cell adhesion assay after incubating the cells in IMDM + 10% FCS for 3 hours. However, this time additional groups with the cultivation conditions changed to the media lack of growth factors (IMDM alone) and longer time period (24 hours) were also investigated. In the end, HMC-1 cells with SP4 overexpression formed larger aggregates than with PDZ, FERM overexpression or control group, but not cells transduced with Y440F overexpression (Figure 3.5a, 3.5b). We also examined cell aggregation over a longer time period (24 hours) in IMDM in the absence of serum with the same transduced cells. SP4-transduced cells still showed largest aggregates among the groups, while PDZ-transduced cells were present mostly as single cell suspension (Figure 3.5c). Photographs were analysed the same way as mentioned before with CellF software and two-way Anova (Figure 3.5d). Cell adhesion was influenced by C-terminal structure of CADM1 as well as culture conditions.
Last but not least, the cell aggregation across even longer incubation time (48 hours) in IMDM alone was studied. The fact that HMC-1 cells proliferate fast in the presence of growth factors thus the cell number of each well will become dramatically different after 48 hours to allow any data analysis.
To summarize, SP4-overexpressing HMC-1 cells formed larger aggregates than the control group (Figure 3.6a), and this observation is in support with previous results (MoiseevaHYPERLINK "#_ENREF_61" et al.HYPERLINK "#_ENREF_61", 2012). The mutations on the C-terminal motifs did not change HMC-1 cell aggregation significantly over 48 hours in the absence of serum (Figure 3.6a). Later on the cross-sectional data was analysed by a two-way ANOVA, and the mutation on aa 440 (Y440F) seemed make a difference (p<0.05) on cell adhesion (Figure 3.6b). Cell number of each well in the 96-well plate was tested to make sure there were equal numbers of HMC-1 cells in the single cell suspension (Figure 3.6c).
Figure 3.4 Homotypic mast cell-cell adhesion affected by the C-terminus mutations of CADM1. (a) Transduced HMC-1 cells were resuspended as single cell suspension and incubated in growth medium for 3h at 37°C. The cross-sectional areas of the largest aggregates were estimated on photographs of five wells for each condition. (b) Data analysis by GraphPad Prism 5 software. Differences among the groups were analysed by a one-way ANOVA, followed by Dunnett's test to determine whether the groups were different from a control group, or Bonferroni's test to compare multiple groups. n=48, ***p<0.001.
Figure 3.5 CADM1-mediated homophillic cell adhesion is affected by its overexpression and mutation in the C-terminus. (a) HMC-1 cells were transduced with adenovirus carrying GFP, SP4, or CADM1 mutants for 6 days. GFP adenoviruses were used as controls. Transduced cells were resuspended as single cell suspension and incubated in growth medium for 3 hours at 37 °C. Original magnification Ã-100; each experiment was performed in quintuplicate (five times). The cross-sectional areas of the largest aggregates were estimated on photographs of five wells for each variant. (b) (n=48, **p<0.01, ***p<0.001) (c) Same transduced HMC-1 cells as in (a) were resuspended in single cell suspension and incubated in IMDM for 24 hours at 37 °C and analysed in the same way. (d) (n=48, ***p<0.001)
Figure 3.6 CADM1-mediated homotypic adhesions over 48 hours in IMDM. (a) HMC-1 cells were transduced with adenovirus carrying GFP, SP4, or CADM1 mutants for 6 days. GFP adenoviruses were used as controls. Transduced cells were resuspended as single cell suspension and incubated in deprived of growth factors for 48 hours at 37 °C. Original magnification Ã-100; each experiment was performed in quintuplicate (five times). The cross-sectional areas of the largest aggregates were estimated on photographs of five wells for each variant. (b) (n=48, *p<0.05) (c) Same amount of the transduced HMC-1 cells as in (a) were resuspended in single cell suspension and incubated in IMDM for 48 hours at 37 °C and the cell number in each well was detected by ATPlite kit.
3.2 Production of CADM1 extracellular domains (ECD) in E. coli
During the second half of this project, the ECD of different CADM1 isoforms were studied. Their structural differences lie in the transmembrane domains, and the transmembrane motifs could influence the cell-cell binding through controlling the localizations of the proteins at the cell membrane and the signalling pathways (MinamiHYPERLINK "#_ENREF_58" et al.HYPERLINK "#_ENREF_58", 2007, SongHYPERLINK "#_ENREF_81" et al.HYPERLINK "#_ENREF_81", 2012). Besides, the communications between the extracellular activities and intracellular interactions are possibly through the transmembrane domains.
In order to produce reasonable amount of target proteins in the lab, several constructs were first designed with the overexpression in E. coli. As to simplify the protein purification, expression vectors able to express the protein of interest with affinity tags and protease cleavage were chosen.
3.2.1 Time course of CADM1-ECD fusion protein expression and solubility tests
To begin with, the extracellular domains of SP4 were cloned into expression vectors under the control of a strong T7 promoter to produce fusion proteins with N-terminal glutathione S-transferase (GST) or poly-Histidine (His) tag, and transformed into E. coli strain BL21 (DE3) (able to express T7 RNA polymerase). In this paper, GST-SP4 and His-SP4 stand for GST or His-tagged SP4-ECD fusion proteins.
In order to optimise the expression conditions, different concentrations of IPTG, induction time and temperatures were tested. Figure 3.7 showed that although GST-SP4 and HIS-SP4 can be detected by SDS-PAGE after induction, most of the fusion proteins were found in inclusion bodies (insoluble pellets). The molecular weight of GST-SP4 is around 68 kDa, and His-SP4 being around 40 kDa. In summary, the expression of GST-SP4 and His-SP4 under 37 °C with 4 hour induction by 1mM IPTG or overnight induction by 0.1mM IPTG can produce the highest level of proteins.
Figure 3.7 Production of CADM1-SP4-ECD fusion proteins in E. coli at 37°C. Coomassie blue stained SDS-PAGE analysis of GST-SP4 and HIS-SP4 fusion