There are many methods. Choose the one that suits you best.

Here is a paper entitiled "Comparison for four methods for extracting periplasmic proteins." It is not open-access, unfortunately.

http://www.sciencedirect.com/science/article/pii/0167701289900365

This paper looks useful: "Isolation of Bacteria Envelope Proteins:

http://sites.lsa.umich.edu/bardwell-lab/wp-content/uploads/sites/265/2015/05/Paper.pdf

Here is the original report of the chloroform method "Simple, Rapid and Quantitative Release of Periplasmic Proteins by Chloroform"

http://jb.asm.org/content/160/3/1181.full.pdf

And here is the cold osmotic shock method:

http://2014.igem.org/wiki/images/b/bb/Bielefeld-CeBiTec_2014-10-14_Cold_osmotic_Shock.pdf

Hi Shahrzad: I'd suggest to transform your plasmid in a few modified BL21 competent cells from the same vendor to see if which host has the best expression level.  Then analyze periplasmic and other fractions for your protein of interest.  Your own experiment will lead to a good protocol for extraction and purification.  Good luck.

There are many methods. Choose the one that suits you best.

Here is a paper entitiled "Comparison for four methods for extracting periplasmic proteins." It is not open-access, unfortunately.

http://www.sciencedirect.com/science/article/pii/0167701289900365

This paper looks useful: "Isolation of Bacteria Envelope Proteins:

http://sites.lsa.umich.edu/bardwell-lab/wp-content/uploads/sites/265/2015/05/Paper.pdf

Here is the original report of the chloroform method "Simple, Rapid and Quantitative Release of Periplasmic Proteins by Chloroform"

http://jb.asm.org/content/160/3/1181.full.pdf

And here is the cold osmotic shock method:

http://2014.igem.org/wiki/images/b/bb/Bielefeld-CeBiTec_2014-10-14_Cold_osmotic_Shock.pdf

Dear Shahrzad,

I would suggest using the Bugbuster Master Mix, and checking for your protein in successive insoluble fractions.

Abhishek

Dear Shahrzad,

although you are using the pelB leader sequence, consider that your protein may be directed to the periplasmic space in E. coli. However, what is important is to check whether your protein of interest is soluble or not. Therefore, try these few steps to test both expression and solubility:

1.) Transform competent cells with your plasmid, plate them and grow overnight

2.) inoculate 5ml LB+antibiotic with a single colony (do this with 5-10 colonies) and grow at 37°C

3.) when cells reach an OD600 of 0.4-0.6 induce with IPTG (here you can use different concentrations, e.g. 1mM, 0.5mM etc) and grow at 37°C for 3h (here you can use different temperatures and different incubation times, e.g. 30°C - 4h or room temperature - overnight).

Note: take 1 ml of the cell culture as non-induced control, spin down an equivalent of OD 1 of the cell cultures (i.e. volume in mL = 1/OD600), remove supernatant and store pellet at -20°C

4.) Meassure OD600. Spin down 2 tubes with an equivalent of OD 1 (see above) from your cell cultures as induced samples, one for expression test, the other for solubility test.

5.) run SDS-PAGE of your non-induced and induced samples by resuspending stored pellets in Laemmli buffer (depends on the concentration, e.g. for 2x Laemmli use 200ul to resuspend equivalent of OD 1), boiling for 10 min and load 10ul on gel.

6.) resuspend the remaining (induced) cell pellet in appropriate lysis buffer (here you have many different option) and add an protease inhibitor. Lyse cells, spin down and divide supernatant (soluble fraction) and pellet (insoluble fraction). Add 100 ul of 2x Laemmli buffer to each and run SDS-PAGE.

One simple method,treating with lysozyme on ice or 37℃,then  high speed centrifugation with optimum buffer ,that you may get half or even more

I know this is an old thread, but found it and wanted to add my recent experience. I have been expressing single-domain heavy chain (VHH) antibodies in WK6 cells with a pelB leader sequence. For purification, I have found that using the ice-cold TES (Tris, EDTA, sucrose) buffer followed by ice-cold water has worked well. The other "flavors" of this protocol involve adding a diluted TES buffer as the shock step and the fractionation buffer as listed in the protocol that Adam Shapiro posted above. I usually get around 3-5 mg/100ml of culture, so I don't know if it is worth trying other conditions, but realize that I could be leaving material behind (the lazy part of me doesn't want to do the bookkeeping :) ).