The main barrier that will hinder the penetration of antibiotics into the biofilm is the extracellular matrix [7, 26]. The extracellular matrix is the backbone of the biofilm and it is very complex in its composition, wide ranging between polysaccharides, proteins, nucleic acids, and lipids. The extracellular polymeric substances (EPS) provide not only physical and adhesive stability to the biofilm, but they also form the scaffold for the three-dimensional architecture that interconnects and organizes cells in biofilms [26].

Critical to matrix function is the distribution of the varied molecular-complex components that influences the developmental, homeostatic, and defensive processes in biofilms. Because of the marked diversity of EPS – inclusive of glycoproteins, proteoglycans, and insoluble hydrophobic polymers, among other components depending on the species involved – it is not surprising that this slimy substance delays considerably the diffusion of antimicrobials [81]. For example, it has been directly observed a profound retardation in the delivery of a penicillin antibiotic from penetrating a biofilm formed by a betalactamase-positive bacterium [3].

Due to the physical protection provided by the biofilm matrix, intense research is ongoing that aim to target the identification of novel matrix components. This novel research on matrix components will provide evidence for the identification and application of matrix-degrading enzymes that may prevent formation and/or activate dispersal of biofilms [45]. Some examples of novel biofilm matrix components that are currently studied are listed in Table 1.1.

It has been observed that throughout the various sections of the biofilm, cells are in different physiological states. Cells at the base of the film, for example, may be dead or lysing, while those near the surface may be actively growing [19, 80]. However, the majority of time cells in biofilms are in a dormant state that is equivalent to cells in the stationary phase of growth [64, 65]. In particular this dormant state is hypothesized to be common in biofilms that are formed in microenvironments where nutrients are scarce, such as treated root canals of teeth [14]. This dormant physiological state related to the general stress response and associated survival responses may offer an explanation for the resistance of biofilm cells to antimicrobials.

Bacteria under the stress of nutrient deprivation have developed efficient adaptive regulatory mechanisms to modify their metabolic balance away from biosynthesis and reproduction [40, 73]. One such mechanism involves the stringent response, a global bacterial response to nutritional stress that is mediated by the accumulation of the alarmones guanosine tetraphosphate and guanosine pentaphosphate, collectively known as (p)ppGpp [25, 68, 85]. For example, (p)ppGpp plays an important role for low-nutrient survival of E. faecalis, an organism that is known to withstand prolonged periods of starvation and remain viable in root-filled teeth for at least 12 months [58, 67]. Furthermore, the alarmone system (p)ppGpp has also a profound effect on the ability of E. faecalis to form, develop, and maintain stable biofilms [15]. These improved understanding of the alarmone mechanisms underlying biofilm formation and survival by E. faecalis may facilitate the identification of pathways that could be targeted to control persistent infections by this organism.

From the perspective of the persisting root canal flora, it is reasonable to assume that such dormant cells might “wake up” at some point in time and resume their metabolic activity to provoke periapical inflammation. Thus, from the metabolic perspective, the reactivation of dormant cells will render biofilm bacteria able to contribute to the persistence of inflammation. For example, a recent case report of a tooth that was adequately treated and showed no signs of disease revealed recurrent disease after 12 years. Histopathologic and histobacteriologic analyses showed a heavy dentinal tubule infection surrounding the area of a lateral canal providing evidence on the persistence of an intraradicular infection caused by bacteria possibly located in dentinal tubules [90].

The above hypothesis on the reactivation of biofilm cells was tested in a recent study [14]. Biofilm cultures of oral isolates of Streptococcus anginosus and Lactobacillus salivarius were forced to enter a state of dormancy by exposing them to nutrient deprivation for 24 h in buffer. After the starvation period the number of metabolically active cells decreased dramatically to zero and their cell membrane integrity was kept intact. Biofilm cells were then exposed to a “reactivation period” with fresh nutrients, but even after 96 h, the cultures were dominated by undamaged cells that were metabolically inactive. This phenomenon was not observed for cells in a planktonic state that were rapidly reactivated after 2 h. The data produced by this study showed that biofilm cells exhibit a slow physiological response and, unlike cells in planktonic culture, do not reactivate in short time periods even under optimal conditions. This observation highlights the difference in physiology between the biofilm and planktonic cultures and also confirms the slower physiological response of biofilm cells [53, 54], a mechanism that may account as a strategy of biofilm bacteria to resist stressful conditions.

Bacteria within biofilms differ in their phenotype, depending on the spatial location of the cells within the community [81, 96]. There is now consistent evidence that has proven the presence of subpopulations of cells within biofilms that significantly differ in their antibiotic susceptibility [32, 41]. This phenomenon is correlated with differences in chemical concentration gradients that create unique microenvironments within biofilm communities. Simultaneously, adaptive variability allows the cells to respond to their local environmental conditions [69, 97]. Numerous studies have investigated the creation of these phenotypically different subpopulations and their mechanisms including genetic alterations, mutations, genetic recombination, and stochastic gene expression. For example, Weiser et al. described two distinct phenotypic variants in S. pneumoniae that switched between a phenotype with the ability to adhere and coexist among eukaryotic cells and a phenotype that was less capable to adhere but was better adapted to evade the host immune response during inflammation or invasive infection [94]. Of interest is the fact that both phenotypes of S. pneumoniae differed in their production of capsular polysaccharide having the inflammation-resistant phenotype an increased production of up to two to six times more capsular polysaccharide. These differences were accentuated by changes in the environmental concentration of oxygen; decreased oxygen levels correlated with an increase in capsular polysaccharide expression.

Interestingly, the formation of subpopulation in biofilms, where physiological differences are in play, has been demonstrated to occur in multispecies biofilms by root canal bacteria [11]. This was shown using four root canal bacterial isolates that, when cocultured, reacted concurrently to the absence of glucose in the culture medium. Although the overall cell viability of the four-species community was not affected by the lack of glucose, there was a significant variation in the 3D structure of the biofilms. In addition, patterns of physiologic adaptation by members of the community to the glucose-deprived medium were observed. The metabolic activity was concentrated in the upper levels of the biofilms, while at lower levels the metabolism of cells was considerably decreased. Subpopulations of species with high glycolytic demands, streptococcus, and lactobacilli were found predominating in the upper levels of the biofilms. This distinct spatial organization in biofilms grown in the lack of glucose shows a clear reorganization of the community in order to satisfy their members’ metabolic pathways in order to enable the long-term persistence of the community. This result lends support to the hypothesis that the reorganization of subpopulations of cells in multispecies biofilms is also important for survival to stress factors from the environment [76].

Groups of cells have been found to persist following exposure to lethal doses of antibiotics and new growing populations appear in the culture [48, 49]. These persister cells (a) may represent cells in some protected part of their cell cycle, (b) are capable of rapid adaptation, (c) are in a dormant state, or (d) are unable to initiate programmed cell death in response to the stimulus [49]. Thus, such persister cells represent a recalcitrant subpopulation that will not die and are capable of initiating a new population with normal susceptibility once the antibacterial effect has been dissipated. To date, these cells have only been reported to occur after the exposure of a bacterial population to high doses of a single antimicrobial agent, which triggered the appearance of persister cells exhibiting multiple drug resistance [51]. The frequency of persister occurrence and the mechanism(s) involved in their appearance are unclear, although one hypothesis with Escherichia coli suggests that persister cells are regulated by the expression of chromosomal toxin–antitoxin genes [42]. In this case, the operon HipA seems to be responsible for tolerance to ciprofloxacin and mitomycin C in stationary-phase planktonic cells and E. coli biofilms [42]. It has also been proposed that the expression of toxins drives bacteria reversibly into the slow-growing, multiple drug-tolerant phenotypes by “shutting down” antibiotic targets [50]. In the context of root canal bacteria, the formation of such persisting populations that are capable of surviving imposed endodontic treatment measures, as rise of the alkaline levels due to application of calcium hydroxide [12], would explain how organisms are able to survive and remain in the environment until the effects of noxious stimuli have dissipated.