Following the production phase of our project, we had 10 StarCore constructs to test. We set out to characterize their antimicrobial properties and mechanism of action. We had many questions in the following categories.

We evaluated the impact of StarCores on bacterial growth by treating bacterial cultures with the fusion proteins and monitoring the OD 600 through time. Representative results are shown in Fig. 1.

Fig. 1. Growth curves of E. coli (A) and B. subtilis (B) in the presence of StarCores.

We also quantified their antimicrobial activity by determining the minimum inhibitory concentrations (MIC). Ovispirin, a commonly used antimicrobial peptide (AMP) with no star-shaped geometry, was used as a control in every experiment. The results of the MIC determination are summarized in Table 1.

StarCores displayed a range of MICs, generally similar to control values. The top performing StarCore was the Ferritin-Alyteserin fusion, with an activity almost 10 times higher than that of the control.

We performed MIC determinations for both E. coli, a Gram-negative bacterium and B. subtilis, a Gram-positive strain. In general, StarCores displayed higher antimicrobial activities than the control Ovispirin in E. coli. While some StarCores exhibited a higher activity towards one bacterial class, others were largely nonspecific (Fig. 2).

Differences in StarCore activity may be attributed to differences in membrane lipid charge and electrostatic potential, which vary among species and are believed to mediate AMP-membrane interactions. This idea is explored in more detail in the modelling and optimization sections.

Fig. 2. MIC of E. coli and B. subtilis in the presence of (A) Ovispirin, (B) Ferritin-Ovispirin and (C) Pyruvate Dehydrogenase-Ovispirin.

In order to investigate the influence of the architecture and the composition of the StarCores on their antimicrobial efficiency, we compared the MIC of constructs containing the same core but different AMPs (Fig. 3A) and that of constructs containing the same AMPs but different cores (Fig. 3B).

In general, StarCores of varying geometry produced similar MIC values. This suggests to us that StarCores may act via a relatively nonspecific mechanism. For example, simply bringing positively charged AMPs to the bacterial membrane at a high local concentration may be sufficient to cause disruption.

Fig. 3. MIC comparison between constructs with the same core but different AMPs (A) and constructs with the same AMPs but different cores (B). The results for the control AMP ovispirin are also shown as reference.

We used time-lapse microscopy to observe the effect of StarCores on growing bacteria. StarCores were able to disrupt log-phase growth in B. subtilis. We observed both bacteriostatic and bacteriolytic activities.

These results are consistent with described mechanisms for AMP activity: depolarization of the membrane potential followed by lysis.

Video 1: Untreated B. subtilis culture

Video 2: B. subtilis culture treated with OV-1

Video 3: B. subtilis culture treated with p81

Video 4: B. subtilis culture treated with p89

As assessed by MIC determination and time-lapse microscopy, the StarCores we have obtained show antimicrobial activities. Specifically, 8 out of the 10 constructs showed antimicrobial activity towards E. coli while 9 out of the 10 constructs displayed antimicrobial potential towards B. subtilis. The antimicrobial effect of the StarCores was exerted at concentrations similar to those of the reference AMP ovispirin. Furthermore, StarCores exhibited some species-specific killing, being particularly active against E. coli. Finally, a clear correlation between the geometry of the StarCores and their antimicrobial activity was not observed, indicating a non-specific mechanism of action.

Overnight cultures of bacteria (E. coli or B. subtilis) were diluted 1:40 into fresh media and grown to log phase (~2 hours) at 37°C with agitation. Cells were washed 3x with PBS, then resuspended in Mueller-Hinton Broth at a final OD of 0.01. The appropriate quantities of AMP or StarCore proteins were then added, and the cultures incubated for 24 hours after which the OD600 was measure with a spectrophotometer. The MIC was defined as the lowest concentration of antimicrobial at which there was no significant growth compared to the background (Fig. 5).


Bacterial seed cultures were prepared as described for the MIC assay in 96 well plates. Growth was measured by monitoring the temporal progression of the OD at 600 nm every 15 minutes for 12 hours with agitation (Fig. 6).

Agar pads consisted of Mueller-Hinton medium with 1.5% agarose and antimicrobial compounds at the appropriate concentrations. Overnight cultures of B. subtilis were back-diluted and grown to log phase (OD 600 = 0.1), thoroughly washed and resuspended in PBS and then spotted onto agar pads. Glass coverslips were placed on top to seal the specimens (Fig. 7), following the method of Graham et al., (2011). Five field positions were imaged overnight for each treatment, then assembled into a time-lapse movie with ImageJ.

Graham, J., Robertson, B.D., and Williams, K.J. (2011). A modified agar pad method for mycobacterial live-cell imaging. BMC Research Notes 4:73.