Food products often contain multi-phase colloidal structures such as emulsions and foams due to their contributions to the textural and sensory properties. To maintain the stability, low-molecular and polymeric surfactants or emulsifiers are often exploited to either reduce the interfacial tension or provide steric hindrance against destabilization. Over the recent years, a global trend in developing surfactant-free and clean label products has been increasingly realized among food industries. Compared with low-molecular-weight-emulsifiers, Pickering particles of micron or nano-sizes can produce emulsions and foams with higher stability due to the irreversible adsorption of particles at interfaces, forming steric hindrance against droplet flocculation and coalescence. Naturally-derived fat globules are typical Pickering stabilizers in dairy products such as whipping cream and ice cream. However, the adverse effects of saturated fats on the environment and human health have also raised concerns. Lactic acid bacteria (LAB) as probiotic food ingredients possess micron sizes to be promising structural building blocks for colloidal food materials. Therefore, this PhD project focuses on the investigation and understanding of bacterial surface properties such as cell surface hydrophobicity (CSH) and aggregating ability, and how these properties can be tuned and utilized to create a series of colloidal structures based on the Pickering mechanism of LAB. As complex living entities, the chemical composition and spatial conformation of bacterial cell wall determined the major surface properties of LAB. To investigating this, chemical approaches including sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE), Fourier transform infrared (FTIR) spectroscopy and x-ray photoelectron spectroscopy (XPS) were used to get information on bacterial surface chemical composition. Concomitantly, physicochemical methods typical for bacteria including zeta potential measurement, microbial adhesion to solvents (MATS) and contact angle measurements (CAM) were employed to relate the obtained chemical information with specific adhesive forces such as electrostatic interactions, steric forces, CSH and Lewis acid-base (AB)/electron accepting-donating properties. Each of the aforementioned characterization techniques showed strengths and weaknesses as such, and therefore the combined use of multiple approaches was always preferable. According to the screening carried out for 31 naturally-occurring Lactobacillus strains, their surface were dominated by negative charges and a notable proportion of tested strains displayed considerable CSH with high adhesions to hexadecane. The high CSH of bacteria seemed to originate from the low Lewis AB characters due to a high protein to polysaccharide ratio on the surface. As a result, such bacteria tended to show higher degree of aggregation in aqueous solution due to the insufficient steric repulsions provided by hydrophilic polymers, primarily polysaccharides. For selected strains, chemical modifications were applied using acid anhydrides, acid chlorides and aldehydes to covalently graft hydrophobic moieties onto surface hydroxyl groups and amine groups. By probing bacterial surface after hexanoic anhydride (HA) and octanal modification, the CSH was drastically enhanced through effective grafting as confirmed by MATS and CAM, most likely by reducing Lewis AB interactions. Zeta potentials generally became slightly more negative due to the removal of surface amine groups. The bacterial aggregation was on the other hand differently affected by the dominance of either long-range steric repulsions or short-range Lewis AB attractions after each modification scheme. Selected LAB strains with intrinsic CSH or after chemical modifications were used to prepare a range of colloidal structures including emulsions, double emulsions, emulsion-templated and microbubble-templated colloidosomes and whipping creams. From the screening study, ~ 39% of the naturally-occurring LAB strains were able to generate Pickering emulsions remaining stable for three weeks, which in a way contradicted our recognitions that LAB as such are intrinsically hydrophilic. The hydrophilic strain, after lauroyl chloride (LC) modification, produced Pickering double emulsions with the inner water droplets stabilized by secondary molecular surfactants released from modification. Octenyl succinic anhydride (OSA) modification of the intrinsically- hydrophobic strain induced fast and rigid bacterial aggregation, enabling their strong adsorption and locking at interfaces, forming stable templates for colloidosome fabrication. The involvement of high proportion of naturally-hydrophobic and aggregating bacteria combined with proper hydrocolloids formulated whipping cream analogues. After whipping, bacteria demonstrated structural functions like fat globules by both adsorbing at air-water interfaces and forming solid network anchoring the air bubbles, transferring foam to a viscoelastic solid. All in all, this PhD project throws light on the understanding of bacterial surface properties and the practical utilization of these properties in facilitating their roles as not only health-promoting ingredients, but also active structural building blocks in colloidal food materials.