An integrated computational/experimental approach was used to predict and identify signal peptide cleavages among microbial proteins of environmental biofilm communities growing in acid mine drainage (AMD). SignalP-3.0 was employed to computationally query the AMD protein database of >16,000 proteins, which resulted in 1,480 predicted signal peptide cleaved proteins. LC-MS/MS analyses of extracellular (secretome) microbial preparations from different locations and developmental states empirically confirmed 531 of these signal peptide cleaved proteins. The majority of signal-cleavage proteins (58.4%) are annotated to have unknown functions; however, Pfam domain analysis revealed that many may be involved in extracellular functions expected within the AMD system. Examination of the abundances of signal-cleaved proteins across 28 proteomes from biofilms collected over a 4-year period demonstrated a strong correlation with the developmental state of the biofilm. For example, class I cytochromes are abundant in early growth states, whereas cytochrome oxidases from the same organism increase in abundance later in development. These results likely reflect shifts in metabolism that occur as biofilms thicken and communities diversify. In total, these results provide experimental confirmation of proteins that are designed to function in the extreme acidic extracellular environment and will serve as targets for future biochemical analysis.