The project set out to discover small molecules that bind the periplasmic adapter protein AcrA of the AcrAB‑TolC efflux system in Escherichia coli and thereby reduce antibiotic efflux. The hypothesis was that such binders would act as adjuvants, restoring the potency of existing antibiotics such as erythromycin. A series of 4(3‑aminocyclobutyl)pyrimidin‑2‑amine derivatives was synthesized and screened for binding to His‑tagged AcrA using microscale thermophoresis (MST). Binding curves were recorded with and without 0.08 % Pluronic F‑127, a non‑ionic surfactant that disrupts colloidal aggregates. In the presence of Pluronic, the curves for all tested compounds collapsed to baseline, indicating that the apparent affinity measured in the absence of surfactant was due to unspecific, aggregate‑mediated interactions. Consequently, all dissociation constants (KD) reported in the study are considered apparent values.

Dynamic light scattering (DLS) confirmed that the active compounds formed colloidal particles with hydrodynamic radii ranging from 1 000 to 10 000 nm, substantially larger than the protein alone. In contrast, a structurally related compound containing a diazacycloheptane ring did not form aggregates and behaved like free AcrA in DLS experiments. The aggregation propensity correlated with the observed synergy in checkerboard assays: only the aggregating derivatives showed enhanced activity with erythromycin, whereas non‑aggregating analogues were inactive.

The functional impact of aggregation on protein stability was assessed by nano‑differential scanning fluorimetry (nDSF). The melting temperature (TM) of AcrA alone was 50.4 °C. Addition of 2 mM of the three active compounds lowered TM to 49.5 °C (compound 30), 50.0 °C (24123034), and 49.8 °C (36287038). In contrast, the inactive analogues (11, 31, 57) produced no measurable shift, with TM values of 50.1 °C and 50.2 °C. These data support the conclusion that the compounds stabilize partially unfolded protein states only through nonspecific aggregation rather than specific binding.

The study was carried out within a joint effort between a German research institute and an Indian partner. The German team, led by Prof. Dr. Björn Windshügel, provided the protein expression, biophysical assays, and data analysis, while the Indian collaborator supplied the chemical library and performed initial screening. Two doctoral students were trained jointly, and regular video conferences were held to coordinate progress. The project ran from early 2021 until 31 March 2023, and was funded by the German Federal Ministry of Education and Research (BMBF) under grant number 3831/03.07_2. The findings were published in the peer‑reviewed journal Biomolecules (volume 13, issue 6, 2023, article 1000) and contribute to the growing understanding that many purported efflux pump inhibitors act as colloidal aggregators rather than specific ligands.