**4. Conclusions**

In this work, we showed that the genes responsible for the anaerobic degradation of benzoate in all aromatic-degrader *Azoarcus*/*Aromatoleum* strains whose genome sequences are currently available are organized in a well-conserved *bzd* cluster. Thus, a *bzdR* regulatory gene and a *P N* promoter could be identified upstream of the catabolic *bzd* operon in all *bzd* clusters analyzed. Interestingly, all *P N* promoters show a conserved architecture that includes three BzdR operator regions (OR1–3). Whereas OR1 is su fficient for the BzdR-mediated repression of the *P N* promoter, the presence of OR2 and OR3 is required for the de-repression promoted by the benzoyl-CoA inducer molecule. Our results sugges<sup>t</sup> that the dimeric protein BzdR binds to the *P N* promoter in the form of four dimers; two of the dimers are initially bound to OR1, and the complex formed may favor the interaction of two additional dimers at OR2 and OR3. The BzdR/*P N* complex formed induces a DNA loop that wraps around the BzdR dimers and generates a superstructure that was observed by AFM. This three-dimensional (3D) configuration may keep the −10/−35 sequences of *P N* inaccessible to the RNAP for achieving an e fficient repression and, at the same time, could facilitate the *P N* de-repression when benzoyl-CoA is generated. However, the elucidation of the molecular mechanism underlying the participation of OR2 and OR3 in the de-repression of *P N* requires further research. In this sense, it should be taken into account that regulatory proteins other than BzdR, e.g., AcpR and AccR, are also involved in the control of the *P N* promoter [24,25] and, hence, more complex structure–function relationships at the *P N* promoter should be expected. In any case, the work presented here strongly suggests the existence of a common BzdR-dependent mechanism to control the expression of the *bzd* genes in *Azoarcus*/*Aromatoleum* strains.

**Author Contributions:** G.D.-R. carried out the genetic manipulations, physiology experiments, and in vitro enzyme assays. P.G.-d.-A. and M.V. designed and performed the AFM experiments and interpreted all the data obtained from this technique. G.D.-R., E.D., and M.C. conceived the whole study, designed the experiments, and contributed to the discussion of the research and interpretation of the data. G.D.-R., E.D., and M.C. wrote the article.

**Funding:** This work was supported by gran<sup>t</sup> BIO2016-79736-R from the Ministry of Economy and Competitiveness of Spain, by gran<sup>t</sup> CSIC 2016 20E093 from the CSIC, and by European Union H2020 gran<sup>t</sup> 760994.

**Acknowledgments:** The author thanks A. Valencia for the technical assistance and S.L. Secugen for DNA sequencing. We are indebted to C. Alfonso and G. Rivas for their help in the analytical ultracentrifugation experiments.

**Conflicts of Interest:** The authors declare that there are no competing interests.
