Candidacidal activity prompted by N-terminus histatin-like domain of human salivary mucin (MUC7)1

Histidine-rich peptides (histatins, Hsn) in saliva are thought to provide a non-immune defense against Candida albicans. Sequence homology search of the human salivary mucin, MUC7, against histatins revealed a domain at the N-terminus (R3-Q17) having 53% identity to Hsn-5. To determine its candidacidal activity, this 15 residue basic histidine-rich domain of MUC7 (I) was prepared by solid-phase Fmoc chemistry. Various N- and C-terminal protected derivatives of I were also synthesized to correlate the effect of peptide overall charge in exhibiting cidal potency. Candidacidal activity measurement of I and its variants showed considerable ED50 values (effective dosage required to kill 50% of candida cells), albeit greater than Hsn-5 (ED50 approximately 4-6 microM). Of the various analogs tested, N-terminal free acid (I, ED50 approximately 40 microM) and amide (V, ED50 approximately 16 microM) exhibited appreciable candidacidal activities suggesting the possible role of peptide net charge in cidal action. Blocking of N-terminus with a bulky octanoyl group showed only marginal effect on the cidal activity of I or V, indicating that hydrophobicity of these synthetic constructs may not be important for exerting such activities. Membrane-induced conformational transition from random coil to helical structures of all the test peptides implied their tendency to adapt order structures at the lipid-membrane interface similar to that of Hsn-5. However, comparison of propensity for helical structure formation vs. ED50 indicated that cidal potency of MUC7 Hsn-like peptides depends largely on electrostatic interactions irrespective of secondary structural elements. Delineation of solution structure of the most active peptide (V) by 2D-NMR revealed essentially a non-structured conformation in aqueous medium, which further supported the fact that the peptide helical structure may not be a prerequisite for posing candidacidal activity. The formation of smaller truncated peptides and/or Hsn-like fragments on proteolytic degradation of intact MUC7 in the presence of oral flora provided indirect evidence that mucin could serve as a backup candidacidal agent to salivary Hsn.     

Role of alpha-helical conformation of histatin-5 in candidacidal activity examined by proline variants

Human salivary histatin-5 (Hsn-5) is a potent in vitro anticandidal agent. The aim of this study was to investigate the importance of alpha-helical structure of Hsn-5 for its candidacidal activity. The following three Hsn-5 variants, where one or more functionally nonessential residues were replaced with proline (potent alpha-helix breaker), were produced by Escherichia coli expression system: H21P (1P), H19P/H21P (2P), and E16P/H19P/H21P (3P). The activities of purified proteins were determined by candidacidal assays, and the secondary structures by circular dichroism (CD) spectroscopy in trifluoroethanol (TFE) that is considered the helix-promoting solvent, and lysophosphatidyl-glycerol (LPG) micelles, the environment that more closely resembles the biological membranes. Our results indicated that 3P variant displayed a candidacidal activity which was similar to that of unaltered Hsn-5 (0P), while 1P and 2P variants showed lower cidal activity. The CD spectra in TFE indicated that 3P variant has less helical characteristics than the 0P, 1P and 2P. These results suggested that the alpha-helical content of Hsn-5 proline variants does not correlate with the candidacidal activity. Further, the CD spectral analysis of peptides in LPG micelles indicated the formation of beta-turn structures in 0P and 3P variants. In conclusion, 3P variant which exhibited comparable candidacidal activity to 0P contains lower percentage of alpha-helical structure than 1P and 2P variants, which exhibited lower candidacidal activity. This suggests alpha-helix may not be important for anticandidal activity of Hsn-5.     

Structural and dynamic consequences of increasing repeats in a MUC1 peptide tumor antigen

MUC1 mucin is a large transmembrane glycoprotein whose extracelluler domain is composed of repeating units of a 20 amino acid sequence. In the cancer associated state, this protein expression becomes upregulated and underglycosylated. Previous studies, which show an enhanced binding of a 5-repeat over a 1-repeat MUC1 peptide to a panel of anti-MUC1 antibodies, have led us to investigate the structural and dynamic consequences of increasing repeat number. Two MUC1 peptides were studied: a 16mer corresponding to slightly less than one full repeat of the MUC1 tandem repeat sequence (GVTSAPDTRPAPGSTA) and a 40mer corresponding to two full repeats of the MUC1 sequence (VTSAPDTRPAPGSTAPPAHG)2. Isotopically labeled versions of these MUC1 peptides were cloned, expressed, purified, and evaluated structurally and dynamically using 15N- and 13C-edited NMR approaches. The data show that MUC1 structure, dynamics, and antibody binding affinity are invariant with increasing repeat number. In light of these results, we conclude that the enhanced antibody affinity of the 5-repeat over the 1-repeat MUC1 peptide is due to multivalency effects, and not due to the development of higher order structure in the longer length peptides. The implications of these results are discussed within the context of a multiple repeat MUC1 breast cancer vaccine design.     

Role of α-helical conformation of histatin-5 in candidacidal activity examined by proline variants

Human salivary histatin-5 (Hsn-5) is a potent in vitro anticandidal agent. The aim of this study was to investigate the importance of α-helical structure of Hsn-5 for its candidacidal activity. The following three Hsn-5 variants, where one or more functionally nonessential residues were replaced with proline (potent α-helix breaker), were produced by Escherichia coli expression system: H21P (1P), H19P/H21P (2P), and E16P/H19P/H21P (3P). The activities of purified proteins were determined by candidacidal assays, and the secondary structures by circular dichroism (CD) spectroscopy in trifluoroethanol (TFE) that is considered the helix-promoting solvent, and lysophosphatidyl-glycerol (LPG) micelles, the environment that more closely resembles the biological membranes. Our results indicated that 3P variant displayed a candidacidal activity which was similar to that of unaltered Hsn-5 (0P), while 1P and 2P variants showed lower cidal activity. The CD spectra in TFE indicated that 3P variant has less helical characteristics than the 0P, 1P and 2P. These results suggested that the α-helical content of Hsn-5 proline variants does not correlate with the candidacidal activity. Further, the CD spectral analysis of peptides in LPG micelles indicated the formation of β-turn structures in 0P and 3P variants. In conclusion, 3P variant which exhibited comparable candidacidal activity to 0P contains lower percentage of α-helical structure than 1P and 2P variants, which exhibited lower candidacidal activity. This suggests α-helix may not be important for anticandidal activity of Hsn-5.     

Degradation of MUC7 and MUC5B in Human Saliva

Background

Two types of mucins, MUC7 and MUC5B constitute the major salivary glycoproteins, however their metabolic turnover has not been elucidated in detail to date. This study was conducted to examine turnover of MUC7 and MUC5B in saliva, by focusing on the relationship between their deglycosylation and proteolysis.

Methodology/Principal Findings

Whole saliva samples were collected from healthy individuals and incubated at 37°C in the presence of various protease inhibitors, sialidase, or a sialidase inhibitor. General degradation patterns of salivary proteins and glycoproteins were examined by SDS-polyacrylamide-gel-electrophoresis. Furthermore, changes of molecular sizes of MUC7 and MUC5B were examined by Western blot analysis. A protein band was identified as MUC7 by Western blot analysis using an antibody recognizing an N-terminal epitope. The MUC7 signal disappeared rapidly after 20-minutes of incubation. In contrast, the band of MUC7 stained for its carbohydrate components remained visible near its original position for a longer time indicating that the rapid loss of Western blot signal was due to the specific removal of the N-termimal epitope. Pretreatment of saliva with sialidase facilitated MUC7 protein degradation when compared with samples without treatment. Furthermore, addition of sialidase inhibitor to saliva prevented proteolysis of N-terminus of MUC7, suggesting that the desialylation is a prerequisite for the degradation of the N-terminal region of MUC7. The protein band corresponding to MUC5B detected in both Western blotting and glycoprotein staining showed little sign of significant degradation upon incubation in saliva up to 9 hours.

Conclusions/Significance

MUC7 was highly susceptible to specific proteolysis in saliva, though major part of MUC5B was more resistant to degradation. The N-terminal region of MUC7, particularly sensitive to proteolytic degradation, has also been proposed to have distinct biological function such as antibacterial activities. Quick removal of this region may have biologically important implication.     

Structural features of the human salivary mucin, MUC7

Human salivary mucin (MUC7) is characterized by a single polypeptide chain of 357 aa. Detailed analysis of the derived MUC7 peptide sequence reveals five distinct regions or domains: (1) an N-terminal basic, histatin-like domain which has a leucine-zipper segment, (2) a moderately glycosylated domain, (3) six heavily glycosylated tandem repeats each consisting of 23 aa, (4) another heavily glycosylated MUC1- and MUC2-like domain, and (5) a C-terminal leucine-zipper segment. Chemical analysis and semi-empirical prediction algorithms for O-glycosylation suggested that 86/105 (83%) Ser/Thr residues were O-glycosylated with the majority located in the tandem repeats. The high (～25%) proline content of MUC7 including 19 diproline segments suggested the presence of polyproline type structures. CD studies of natural and synthetic diproline-rich peptides and glycopeptides indicated that polyproline type structures do play a significant role in the conformational dynamics of MUC7. In addition, crystal structure analysis of a synthetic diproline segment (Boc-Ala-Pro-OBzl) revealed a polyproline type II extended structure. Collectively, the data indicate that the polyproline type II structure, dispersed throughout the tandem repeats, may impart a stiffening of the backbone and could act in consort with the glycosylated segments to keep MUC7 in a semi-rigid, rod shaped conformation resembling a ‘bottle-brush’ model.     

Effect of glycosylation on MUC1 humoral immune recognition: NMR studies of MUC1 glycopeptide-antibody interactions

MUC1 mucin is a large transmembrane glycoprotein, of which the extracellular domain is formed by a repeating 20 amino acid sequence, GVTSAPDTRPAPGSTAPPAH. In normal breast epithelial cells, the extracellular domain is densely covered with highly branched complex carbohydrate structures. However, in neoplastic breast tissue, the extracellular domain is underglycosylated, resulting in the exposure of a highly immunogenic core peptide epitope (PDTRP in bold above) as well as the normally cryptic core Tn (GalNAc), STn (sialyl alpha2-6 GalNAc), and TF (Gal beta1-3 GalNAc) carbohydrates. In the present study, NMR methods were used to correlate the effects of cryptic glycosylation outside of the PDTRP core epitope region to the recognition and binding of a monoclonal antibody, Mab B27.29, raised against the intact tumor-associated MUC1 mucin. Four peptides were studied: a MUC1 16mer peptide of the sequence Gly1-Val2-Thr3-Ser4-Ala5-Pro6-Asp7-Thr8-Arg9-Pro10-Ala11-Pro12-Gly13-Ser14-Thr15-Ala16, two singly Tn-glycosylated versions of this peptide at either Thr3 or Ser4, and a doubly Tn-glycosylated version at both Thr3 and Ser4. The results of these studies showed that the B27.29 MUC1 B-cell epitope maps to two separate parts of the glycopeptide, the core peptide epitope spanning the PDTRP sequence and a second (carbohydrate) epitope comprised of the Tn moieties attached at Thr3 and Ser4. The implications of these results are discussed within the framework of developing a glycosylated second-generation MUC1 glycopeptide vaccine.     

Structure and Chromosomal Localization of the Human Salivary Mucin Gene, MUC7

We have isolated and characterized several MUC7 genomic clones encoding the human low-molecular-weight salivary mucin, MG2. The MUC7 gene spans ∼10.0 kb and comprises of three exons and two introns. Intron 1 is ∼1.7 kb long and is located in the 5′-untranslated region of the corresponding MUC7 cDNA. Intron 2 spans ∼6.0 kb and is located close to the boundary of the putative leader peptide and secreted protein. The entire region encoding the secreted peptide is located on exon 3, spanning ∼2.2 kb. The nucleotide sequence of sections of the MUC7 gene, including 1500 bp of the 5′-flanking region, was determined and analyzed for motifs identical or homologous to other known response elements. A modified RACE procedure was used to determine the 5′-end of the MUC7 mRNA. PCR, the human–hamster somatic cell hybrid panel PCRable DNAs kit, and anin situhybridization analysis on the complete metaphase chromosome spreads were used for the chromosomal localization of the MUC7 gene. It was mapped to chromosome 4q13–q21.     

Salivary histatin 5: dependence of sequence, chain length, and helical conformation for candidacidal activity

Histatin 5 (Asp1-Ser-His-Ala4-Lys-Arg-His-His8-Gly-Tyr-Lys-Arg12-Lys-Ph e-His-Glu16-Lys-His - His-Ser20-His-Arg-Gly-Tyr24), one of the basic histidine-rich peptides present in human parotid saliva and several of its fragments, 1-16 (N16), 9-24 (C16), 11-24 (C14), 13-24 (C12), 15-24 (C10), and 7-16 (M10), were synthesized by solid-phase procedures. Native histatin 5 from human parotid saliva was also purified. Their antifungal activities on two strains of Candida albicans have been studied and their conformational preferences both in aqueous and non-aqueous solutions examined by circular dichroism. The synthetic histatin 5, C16, and C14 peptides were highly active and inhibited the growth of C. albicans. The candidacidal activity data of synthetic histatin 5 were comparable to the values of the native histatin 5 isolated from parotid saliva and those reported previously, although the assay system used and the strains examined were different. The C16 fragment was as active as the whole peptide itself, whereas the N16 fragment was far less active than C14, suggesting that the sequence at the C-terminal is important for its fungicidal activity. An increase in the chain length of the C-terminal sequence from 12 to 16 residues increased the candidacidal activity, thereby indicating that a peptide chain length of at least 12 residues is necessary to elicit optimum biological activity. The CD spectra of these linear peptides showed that they are structurally more flexible, and they adopt different conformations depending on the solvent environment. CD studies provided evidence that histatin 5 and the longer fragments, C16, N16, and C14 preferred alpha-helical conformations in non-aqueous solvents such as trifluoroethanol and methanol, while in water and pH 7.4 phosphate buffers, they favored random coil structures. The shorter sequences seemed to adopt either turn structures or unordered structures both in aqueous and non-aqueous solutions. It appears that the sequence at the C-terminal of histatin 5 with a minimum chain length of 14 residues and alpha-helical conformation are the important structural requirements for appreciable candidacidal activity.     

Lack of evidence for an immunosuppressive role for MUC1

The in vitro anti-proliferative properties of various supernatants from MUC1-expressing cell lines and of purified preparations of MUC1 were evaluated. We have observed that supernatants from the MUC1- and MUC3-positive cell line T47D, but not from the MUC1- and MUC4-positive cell line MCF7, were able to inhibit proliferation of cells from various haematopoietic cell lines. Although the activity of T47D supernatants could be abrogated by immunodepletion of MUC1, immunopurified MUC1 from T47D was unable to inhibit cell proliferation. Significantly, supernatants from mouse 3T3 cells transfected with a secreted form of MUC1 or from BHK-21 cells infected with a recombinant vaccinia virus coding for the secreted form of MUC1, as well as preparations of purified MUC1 from bile or urine, were likewise unable to inhibit T cell proliferation. Surprisingly, a crude mixture of bile mucins had a suppressive effect on T cell growth. Our results suggest that other molecules, such as amino sugars or other mucins, which can associate with MUC1, are likely to be responsible for the observed anti-proliferative effects of T47D cells. Received: 20 August 1998 / Accepted: 3 December 1998     

Solution structure of O-glycosylated C-terminal leucine zipper domain of human salivary mucin (MUC7)

Solution structures of a 23 residue glycopeptide II (KIS* RFLLYMKNLLNRIIDDMVEQ, where * denotes the glycan Gal-beta-(1-3)-alpha-GalNAc) and its deglycosylated counterpart I derived from the C-terminal leucine zipper domain of low molecular weight human salivary mucin (MUC7) were studied using CD, NMR spectroscopy and molecular modeling. The peptide I was synthesized using the Fmoc chemistry following the conventional procedure and the glycopeptide II was synthesized incorporating the O-glycosylated building block (Nalpha-Fmoc-Ser-[Ac4-beta-D-Gal-(1,3)-Ac2-alpha-D-GalN3+ ++]-OPfp) at the appropriate position in stepwise assembly of peptide chain. Solution structures of these glycosylated and nonglycosylated peptides were studied in water and in the presence of 50% of an organic cosolvent, trifluoroethanol (TFE) using circular dichroism (CD), and in 50% TFE using two-dimensional proton nuclear magnetic resonance (2D 1H NMR) spectroscopy. CD spectra in aqueous medium indicate that the apopeptide I adapts, mostly, a beta-sheet conformation whereas the glycopeptide II assumes helical structure. This transition in the secondary structure, upon glycosylation, demonstrates that the carbohydrate moiety exerts significant effect on the peptide backbone conformation. However, in 50% TFE both the peptides show pronounced helical structure. Sequential and medium range NOEs, CalphaH chemical shift perturbations, 3JNH:CalphaH couplings and deuterium exchange rates of the amide proton resonances in water containing 50% TFE indicate that the peptide I adapts alpha-helical structure from Ile2-Val21 and the glycopeptide II adapts alpha-helical structure from Ser3-Glu22. The observation of continuous stretch of helix in both the peptides as observed by both NMR and CD spectroscopy strongly suggests that the C-terminal domain of MUC7 with heptad repeats of leucines or methionine residues may be stabilized by dimeric leucine zipper motif. The results reported herein may be invaluable in understanding the aggregation (or dimerization) of MUC7 glycoprotein which would eventually have implications in determining its structure-function relationship.     

Malignancy-induced autoimmunity to MUC1: initial antibody characterization.

Numerous reports document the existence of autoantibodies to MUC1 in the circulation of individuals with breast and other solid malignancies, with the majority of researchers utilizing MUC1 peptides in their detection. This report documents the purification, using peptide and whole molecule, and characterization of such autoantibodies from an individual with an unusual, highly MUC1-positive, myosarcoma. Purification of autoantibodies from serum was performed using affinity chromatography against either MUC1 peptide or whole molecule MUC1 [derived both from the patient (Pt-MUC1) and from a pool of sera from patients with advanced breast cancer (ABC-MUC1)]. Enzyme-linked immunosorbent assays (ELISAs) were used to compare specificity of purified autoantibodies. Peptide epitopes were determined by Ptifcan system against 7-mer peptides covering the 20 amino acid repeat of the MUC1 extracellular domain. Substantially higher amounts of autoantibodies were isolated when purifying against Pt-MUC1 rather than either ABC-MUC1 or peptide. Whole molecule purified autoantibodies demonstrated an increased specificity for tumour-derived MUC1. Pt-MUC1 autoantibodies were of both the immunoglobulin (Ig)G and IgM class, whilst autoantibodies purified against ABC-MUC1 and MUC1 peptide were IgG only. A greater range of peptide epitopes was defined by those autoantibodies purified against whole molecule. This report presents data indicating the presence of autoantibodies to MUC1 in an individual diagnosed with a MUC1 over-expressing myosarcoma. Confirmation of these autoantibodies as being specific for tumour-associated MUC1 is given. Further, it suggests that, although autoantibodies are present that recognize core protein determinants, the initial, and dominant, immunizing epitope is not purely pretentious in nature.     

Epitope mapping of antigenic MUC1 peptides to breast cancer antibody fragment B27.29: a heteronuclear NMR study

MUC1 mucin is a breast cancer-associated transmembrane glycoprotein, of which the extracellular domain is formed by the repeating 20-amino acid sequence GVTSAPDTRPAPGSTAPPAH. In neoplastic breast tissue, the highly immunogenic sequence PDTRPAP (in bold above) is exposed. Antibodies raised directly against MUC1-expressing tumors offer unique access to this neoplastic state, as they represent immunologically relevant "reverse templates" of the tumor-associated mucin. In a previous study [Grinstead, J. S., et al. (2002) Biochemistry 41, 9946-9961], (1)H NMR methods were used to correlate the effects of cryptic glycosylation outside of the PDTRPAP core epitope sequence on the recognition and binding of Mab B27.29, a monoclonal antibody raised against breast tumor cells. In the study presented here, isotope-edited NMR methods, including (15)N and (13)C relaxation measurements, were used to probe the recognition and binding of the PDTRPAP epitope sequence to Fab B27.29. Two peptides were studied: a one-repeat MUC1 16mer peptide of the sequence GVTSAPDTRPAPGSTA and a two-repeat MUC1 40mer peptide of the sequence (VTSAPDTRPAPGSTAPPAHG)(2). (15)N and (13)C NMR relaxation parameters were measured for both peptides free in solution and bound to Fab B27.29. The (13)C(alpha) T(1) values best represent changes in the local correlation time of the peptide epitope upon binding antibody, and demonstrate that the PDTRPAP sequence is immobilized in the antibody-combining site. This result is also reflected in the appearance of the (15)N- and (13)C-edited HSQC spectra, where line broadening of the same peptide epitope resonances is observed. The PDTRPAP peptide epitope expands upon the peptide epitope identified previously in our group as PDTRP by homonuclear NMR experiments [Grinstead, J. S., et al. (2002) Biochemistry 41, 9946-9961], and illustrates the usefulness of the heteronuclear NMR experiments. The implications of these results are discussed within the context of MUC1 breast cancer vaccine design.     

Proteomic and metabolic characterization of a Candida albicans mutant resistant to the antimicrobial peptide MUC7 12-mer

MUC7 12-mer is a cationic peptide derived from the N-terminal portion of human mucin MUC7, exhibiting potent antibacterial and antifungal properties. To advance our knowledge regarding the mechanisms of action of MUC7 peptide against an opportunistic fungal pathogen Candida albicans, we sought to develop and characterize mutant(s) resistant to this peptide. One of the selected mutants, designated #37, was much less susceptible to the MUC7 12-mer in a killing assay than the parental strain (ED(50)>40 vs. c. 6 microM, respectively). Difference gel electrophoresis (DIGE) analysis of the mutant revealed elevation of several glycolytic enzymes. The mutant also exhibited lowered ATP contents along with a relatively lower rate of oxygen consumption, as well as inability to grow on nonfermentable carbon sources. These properties are likely to be associated with changes in metabolic regulation, rather than lack of functional mitochondria, as determined by rhodamine 123 staining. Analysis of interaction between fluorescently labeled peptide and cells of both strains revealed that resistance of the mutant #37 is associated with changes in the process of transition between surface-bound state of the peptide to its internalization marking cell death.     

MUC5B Leads to Aggressive Behavior of Breast Cancer MCF7 Cells

The mucin MUC5B has a critical protective function in the normal lung, salivary glands, esophagus, and gallbladder, and has been reported to be aberrantly expressed in breast cancer, the second leading cause of cancer-related deaths among women worldwide. To understand better the implication of MUC5B in cancer pathogenesis, the luminal human breast cancer cell line MCF7 was transfected with a vector encoding a recombinant mini-mucin MUC5B and was then infected with a virus to deliver a short hairpin RNA to knock down the mini-mucin. The proliferative and invasive properties in Matrigel of MCF7 subclones and subpopulations were evaluated in vitro. A xenograft model was established by subcutaneous inoculation of MCF7 clones and subpopulations in SCID mice. Tumor growth was measured, and the tumors and metastases were assessed by histological and immunological analysis. The mini-mucin MUC5B promoted MCF7 cell proliferation and invasion in vitro. The xenograft experiments demonstrated that the mini-mucin promoted tumor growth and MCF7 cell dissemination. In conclusion, MUC5B expression is associated with aggressive behavior of MCF7 breast cancer cells. This study suggests that MUC5B may represent a good target for slowing tumor growth and metastasis.