A Reproducible Oral Microcosm Biofilm Model for Testing Dental Materials

Objectives: There is increasing interest in biofilm effects on dental materials. Most studies have used single-species biofilms, although some have used up to nine species. None of those models replicate oral diversity. Microcosm biofilms grown directly from plaque are much more diverse, but more difficult to characterize. Here we used the Human Oral Microbial Identification Microarray (HOMIM) to validate a reproducible oral microcosm model.

Methods: Whole saliva and plaque were collected four times from four donors. Hydroxyapatite and dental composite disks were first coated with filter-sterilized saliva supernatant, and then treated with plaque from the same donor. Disks were placed into a biofilm reactor containing basal mucin medium (BMM), and incubated at 37° C under constant shear (125 rpm) for 24 h. BMM was then flowed through the reactor for 48 h at 20 ml/min (125 rpm; 37° C). In later experiments, the reactor was pulsed with 20% sucrose eight times during flow. We also made stocks from microcosms, and grew biofilms from stock multiple times. DNA from plaque and microcosms were analyzed by HOMIM for 272 species. Plate counts were determined, and disks were examined by scanning electron microscopy (SEM).

Results: Disk plate counts ranged from 10⁸ – 10¹⁰ CFU/ml. SEM showed complex multi-layered structures. Disk biofilms typically included about 60% of plaque species, including the species that were most abundant in the original plaque samples. Cluster analysis indicated that biofilms from the same experiment were extremely similar, regardless of material. Sucrose-pulsing increased the abundance of Streptococcus and Veilonella. Biofilms from the same donor, grown at different times, clustered together, as did biofilms grown from stocks at different times.

Conclusions: This model produces reproducible microcosm biofilms that are representative of the oral flora. Ongoing studies are using it to evaluate biofilm effects on composite materials.

Supported by NIH DE021366