Biofilms, the complex communities of microbiota that live in association with aquatic interfaces, are considered to be hotspots of microbial life in many aquatic ecosystems. Although the importance of attached algae and bacteria is widely recognized, the role of the highly abundant biofilm‐dwelling micrograzers (i.e., heterotrophic protists and small metazoans) is poorly understood. Studies often highlight the resistance of bacterial biofilms to grazing within the microbial food web and therefore argue that the micrograzers have a modulating role (i.e., have effects on biofilm phenotype) rather than a direct trophic role within biofilms. In the present review, we show that this view comes too short, and we establish a conceptual framework of biofilm food webs consisting of three major elements. (1) Energy pathways and subsidization from plankton. As inhabitants of interfaces, biofilm‐dwelling grazers potentially access both planktonic organisms and surface‐associated organisms. They can play an important role in importing planktonic production into the biofilm food web and thus in the coupling of the planktonic and benthic food webs. Nevertheless, specialized grazers are also able to utilize significant amounts of autochthonous biofilm production. (2) Horizontal complexity of the basal food web. While bacteria and algae within biofilms are edible in general, food quality and accessibility of both bacteria and algae can differ considerably between different prey phenotypes occurring during biofilm formation with respect to morphology, chemical defense, and nutrient stoichiometry. Instead of considering bacteria and algae within biofilms to be generally resistant to feeding by micrograzers, we suggest considering a horizontal food‐quality axis to be at the base of biofilm food webs. This food quality gradient is probably associated with increasing costs for the micrograzers. (3) Vertical food web complexity and food chain length. In addition to the consumption of bacteria and algae, many predatory micrograzers exist within biofilm food webs. With the help of video microscopy, we were able to demonstrate the existence of a complex food web with several trophic levels within biofilms. Our conceptual framework should assist in integrating food web concepts and processes into whole‐biofilm budgets and in understanding food‐web‐related interactions within biofilms.

Biofilms play an important role in the material flux of many aquatic ecosystems, but little is known about the mechanisms controlling their community structure under natural conditions. In the present study, we focused on the effects of ciliates on the quantity and taxonomic composition of heterotrophic flagellates (HF), and the effects of HF on the quantity and life forms (single cells vs. microcolonies) of bacteria in the early phase of biofilm colonization. For this purpose, we established semi-natural biofilms in flow cells connected to the river Rhine at Cologne, Germany. Using filter cartridges, we size-fractionated the potamoplankton, which is the source of the biofilm community, thus establishing biofilms containing (1) only bacteria (1.2 µm filter), (2) HF and bacteria (8 or 5 µm filter), or (3) ciliates, HF and bacteria (20 µm filter). The presence of ciliates negatively influenced the abundance of biofilm-dwelling HF and selectively altered the taxonomic composition of the HF community. The presence of HF resulted in a significant reduction in the abundance of single bacterial cells, but enhanced the abundance of bacterial microcolonies. Furthermore, the presence of ciliates stimulated the abundance of single-cell bacteria (probably due to an HF-mediated trophic cascade), but had no effect on bacterial microcolonies. Taken together, the results of this study show the importance of protozoan grazing in shaping the species composition and morphology of early river biofilms under semi-natural conditions.

The effects of protozoa (heterotrophic flagellates and ciliates) on the morphology and community composition of bacterial biofilms were tested under natural background conditions by applying size fractionation in a river bypass system. Confocal laser scanning microscopy (CLSM) was used to monitor the morphological structure of the biofilm, and fingerprinting methods (single-stranded conformation polymorphism [SSCP] and denaturing gradient gel electrophoresis [DGGE]) were utilized to assess changes in bacterial community composition. Season and internal population dynamics had a greater influence on the bacterial biofilm than the presence of protozoa. Within this general framework, bacterial area coverage and microcolony abundance were nevertheless enhanced by the presence of ciliates (but not by the presence of flagellates). We also found that the richness of bacterial operational taxonomic units was much higher in planktonic founder communities than in the ones establishing the biofilm. Within the first 2 h of colonization of an empty substrate by bacteria, the presence of flagellates additionally altered their biofilm community composition. As the biofilms matured, the number of bacterial operational taxonomic units increased when flagellates were present in high abundances. The additional presence of ciliates tended to at first reduce (days 2 to 7) and later increase (days 14 to 29) bacterial operational taxonomic unit richness. Altogether, the response of the bacterial community to protozoan grazing pressure was small compared to that reported in planktonic studies, but our findings contradict the assumption of a general grazing resistance of bacterial biofilms toward protozoa.

We tested the role of immigration from the plankton on the structure of local surface-associated communities of ciliates and heterotrophic flagellates (HF) in flow cells fed with fresh riverine water (Rhine, Germany). By applying size fractionation, we reduced the immigration potential of HF from the plankton in mature, pregrown biofilms by 93%, and observed no significant effects on the abundance and taxonomic composition of HF. Compared with a treatment with natural plankton density, a 100% reduction in the number of planktonic cells flowing over mature biofilms resulted in a reduced total abundance of ciliates, although only slight effects on the relative composition of present morphospecies were detectable. When starting with sterile flow cells, we found an initial linear increase in HF abundance proportional to their planktonic abundance; additionally supplemented planktonic bacteria showed that this initial colonization rate was resource independent. The initial phase was followed by an exponential increase, which was also independent from the resource level and strongest when initial colonization was low. In contrast to the early succession stages, the final abundances reached were independent of the plankton abundance but strongly dependent upon the local resource level. Immigration is an important factor controlling the initial substrate colonization and early growth, and ciliate and HF communities in biofilms become increasingly independent of immigration with maturation, being then controlled by local factors such as resources.