For downstream analysis in gene expression, it is recommended to have at least 2-3 housekeeping genes. May I use gyrA and gyrB as well as a third gene, or as they encode for same enzyme but different subunits, it is not allowed?
Ideally, I think you should shoot for 2-3 housekeeping genes that are consitutively and highly expressed and perform different functions in the cell(s). Because treatments can affect whole cellular systems like the cytoskeleton or metabolism, only using housekeeping genes from one family can potentially skew your results. GAPDH and Actin make a good combination because one is related to metabolism and the other is related to the cytoskeleton. Actin and Tubulin make a bad combination because if your treatment affects the cytoskeleton, it is likely both will be affected and skew your results. Likewise, if your internal controls (housekeepers) are not only from the same functional family, but from the same enzyme, the potential for misleading or skewed results is increased.
I would try to find at least one other housekeeping gene and only use one of the gyrase subunits. I wouldn't say it isn't allowed to use two subunits of the same enzyme as a control, but there are better choices and those should be used unless there is a technical reason as to why these are the only internal controls you can use.