I have used constant cell disruptor such as French press and sonicator for protein purification. French press is useful to lysis bacteria at high cell density and in large volumes. So it may useful to purify membrane proteins in large scale by using convensional protein purification methods (such as gel filtration, ion chromatograhy etc). In contract, sonication can be used to lyse bacteria in 1ml to 500ml volume. If you want to affinity purify protein in a small scale first and then want to scale up the process, then I would recommend sonication for lysing bacteria.
otherwise while using sonication protect the denaturation of enzyme by heat produced by sonicator. keep the sample carefully in ice filled box and prevent the enzyme from excessive heating.
Best homogenization system I have used, for bacteria or mammalian cells is the Stansted Pump (a French press system. See http://www.frenchpressurecell.com). It is scalable. Benchtop sonication is not scalable. Even small changes in volume, vessel, transducer etc. will give you different results. Also, cavitation effects of sonication can cause significant protein denaturation. Stansted Pump is scalable from 1 ml batch to 1000 litre/h continuous (on their line of instruments). If you are buying, and have a serious need for scalability, consider this. If you are on a limited budget, or have limited access to equipment, success is going to depend more on your skill in process development than on the choice of disruption equipment.
I would suggest to use E.coli BL21 or similar for protein expression and DH5alpha for plasmid production.
Second:
In my hands, sonication works okay (but not perfect). However, if you have access to a microfluidizer system, definetley use that instead of sionication, because there is three advantages:
1) better and more complete cell disruption , means higher output (~90%)
2) more gentle cell disruption, e.g. less heat produced = better for intact protein recovery