This paper describes several salivary components and their distribution in other mucosal secretions. Histatins are polypeptides which possess exceptional anti-fungal and anti-bacterial activities, but are nevertheless present only in saliva. Proline-rich proteins (PRPs) are members of a closely related family, of which the acidic PRPs are found solely in saliva, whereas the basic PRPs are also found in other secretions. Mucins are a group of glycoproteins that contribute to the visco-elastic character of the mucosal secretions. Despite the similarities in their structure and behavior, mucins have distinct tissue distributions and amino acid sequences. Other salivary proteins are present in one or more mucosal secretions. Lysozyme is an example of a component belonging to an ancient self-defense system, whereas secretory immunoglobulin A (sIgA) is the secreted part of a sophisticated adaptive immune system. Cystatins are closely related proteins which belong to a multigene family. Alpha-Amylase is a component that is believed to play a specific role in digestion, but is nevertheless present in several body fluids. Kallikrein and albumin are components of blood plasma. But whereas albumin diffuses into the different mucosal secretions, kallikrein is secreted specifically by the mucosal glands. The presence of these proteins specifically in saliva, or their distribution in other mucosal secretions as well, may provide important clues with respect to the physiology of those proteins in the oral cavity.

Extra Parotid Glycoprotein (EP-GP) is a glycoprotein isolated from human saliva, having homologues in several other body fluids. The biological role of EP-GP and its homologues is unknown. Recently, EP-GP was shown to bind in vitro to the bacterium Streptococcus salivarius HB. In contrast, no binding to a number of other oral microorganisms could be demonstrated. In the present study we have determined whether binding of EP-GP to bacteria occurs in vivo in saliva and in other EP-GP containing body fluids. Therefore the presence of EP-GP on bacteria in vivo was determined by analyzing oral, skin and ear floras by confocal fluoresence microscopy using specific antibodies. About 12% of the in vivo oral flora had EP-GP present on their surface, while approximately 5% of the bacteria from ear canal or skin was positive for EP-GP. IgA was detected on approximately 65% of the salivary bacteria, whereas the high-molecular weight mucin (MG1) and cystatin C were not detectable on any oral bacterium. Using a replica-plate assay, a number of EP-GP binding strains in saliva were isolated and identified as Gemella haemolysans, Gemella morbillorium, Streptococcus acidominimus, Streptococcus oralis, Streptococcus salivarius and Streptococcus parasanguis. Bacteria from the ear canal and skin bacteria were identified as Staphylococcus hominis. It is concluded that EP-GP is selectively bound in vivo to several oral and non-oral bacterial species.

In human saliva, two different mucin populations can be distinguished, viz., high-molecular-weight mucins (MG1, mol. wt > 1 x 10(6)) and low-molecular-weight mucins (MG2, mol. wt approximately 125 kD). The carbohydrate moiety of MG1 displays a wide spectrum of oligosaccharide structures, varying in composition, length, branching, and acidity. The biological significance of the heterogeneity in carbohydrate structures of mucins is unclear. The present investigation focused on the question whether MG1, because of its diverse carbohydrate side-chain population, can bind to a large variety of oral micro-organisms. A replica plate technique, in combination with immunochemical detection with monoclonal antibodies against MG1, was used to screen in vivo human oral microflora for the presence of micro-organisms which could bind the high-molecular-weight salivary mucin MG1. Binding to purified MG1 was established for Hemophilus (para)influenzae species, whereas other species, including Streptococcus and Staphylococcus, were negative. MG1 binding to Hemophilus parainfluenzae could be abolished by protease treatment of MG1. In contrast, periodate acid treatment, partial deglycosylation, or addition of monosaccharides did not affect MG1 binding to H. parainfluenzae, indicating that MG1 carbohydrate side-chains were not directly involved in the binding. The binding was pH-dependent, showing an increase when the pH was lowered from 8.0 to 4.0. These data indicate that MG1 can be bound in a selective manner by Hemophilus spp. and suggest that the 'naked' unglycosylated polypeptide moiety of MG1 is involved in its binding to Hemophilus parainfluenzae.

The interaction of the human salivary glycoprotein EP-GP with a number of oral bacterial species, following incubation with human whole saliva, has been investigated. EP-GP could be detected with a specific monoclonal antibody, by means of ELISA or by electrophoresis in combination with Western Transfer. The results indicated that EP-GP is bound only by Streptococcus salivarius, and not by the other tested strains of bacteria, Actinomyces viscosus, A. naeslundii, Actinobacillus actinomycetemcomitans, Bacteroides fragilis, S. gordonii, S. oralis, S. sanguis, S. mitis, S. mutans, S. sobrinus, S. rattus, S. constellatus, and S. anginosus. Binding of EP-GP to S. salivarius is mediated by a protein-protein interaction, which was found to be pH-dependent with a maximum binding between pH 5 and 6. For further characterization of the binding of EP-GP to S. salivarius, four mutants were tested, each of them lacking different cell wall antigens. EP-GP was bound to all mutants in amounts comparable with the wildtype, in spite of the different surface antigen compositions. We were able to identify a 27-kD EP-GP binding protein, by extraction of S. salivarius-cell wall antigens and electrophoretic techniques. In addition to EP-GP, S. salivarius also bound two other salivary proteins, namely, secretory IgA and low-molecular-weight mucin (MG-2).

Two salivary 20-kDa proteins [the human lipocalin Von Ebner's gland protein (VEGh) and extraparotid glycoprotein (EP-GP)] show several remarkable similarities and differences. The latter is identical to secretory actin-binding protein (SABP), gross cystic disease fluid protein-15 (GCDFP-15), prolactin-induced protein (PIP), and 17-kDA CD4-binding glycoprotein (gp17). Much is known about the distribution, localization, biochemical characteristics, and molecular biology of these two proteins, yet there are only few clues about their functions.