Figure 4

Germicidal properties of the proteinaceous apocrine secretion isolated from the lumen of late prepupal SGs. Apocrine secretion from the equivalent of 5, 20 or 100 wild-type SGs was applied in a 10 μl volume of sterile PBS onto 6-mm diameter Rotilabo filters placed on 10-cm LB-agar (a,b) or YPD-agar (c,d) plates previously spread with E. coli (a), S. aureus (b), S. cerevisiae (c), or C. albicans (d). Sectors 1, 2, and 3 had no filter, an untreated filter, or a filter with only PBS, respectively. Sectors 4, 5, and 6 had filters with the apocrine secretion from the equivalent of 5, 20, or 100 pairs of SGs, respectively. Greater amounts of the apocrine secretion were associated with increased growth inhibition for each microbe tested. Images show data from one representative experiment. The graph shows the mean diameter of growth inhibition, measured using Image J, observed in three replicate experiments (Error bars: 95% CI). There were statistically significant differences in the degree of growth inhibition of each microbe by different amounts of SG extract as determined by one-way ANOVA (E. coli:F(2,6) = 2788, p = 1.2e−09; S. aureus: F(2,6) = 192.8, p = 3.6e−06; S. cerevisiae: F(2,6) = 162.4, p = 6.0e−06; C. albicans: F(2,6) = 80.6, p = 4.6e−05). For each microbe, post-hoc TukeyHSD tests revealed significant differences (adjusted p < 0.05) in the mean growth inhibition by different amounts of SG extract in all pairwise comparisons. Differences in growth inhibition of Gram-negative (a) and Gram-positive (b) bacteria may partly reflect differences in culture density.