With respect to the fusion protein vector selection, there are many possible choices and the ideal choice may be quite dependent on your protein. The pATH fusion vectors (trpE fusion) have worked quite well for me, I have obtained a mg of pure protein from 50ml of culture with several different constructs using these vectors and the method described below. Other reports of similar yields with trpE fusions are present in the literature. The pUR vector series (lacZ fusion) has been successfully used by many people in the Rubin lab, the primary advantage of this and other lacZ fusion vectors is that they lead to a very large fusion protein (116kd + what you add) which is substantially larger than most of the other proteins which are present in a bacterial extract, facilitating purification by gel electrophoresis. lacZ fusion proteins; are also often insoluble or if soluble can be purified with a substrate analog affinity resin. Yields per ml of culture will generally be substantially lower than those obtainable with trpE fusions. Several lacZ fusion vector series other than the pUR series are availible at this writing, and they may provide superior yields, but I have not had any experience with these. The T7 vector system (pET vectors) deserves a mention ‹ this vector system is useful for producing protein initiated at the bona-fide initiation codon, or for making fusion proteins (to a portion of a T7 coat protein). Levels of protein production comparable to that obtained with trpE fusions can be achieved. In summary, with regard to selecting vectors my suggestion is that you choose two different vectors which have convenient restriction sites and which produce protein having the desired characteristics (i.e. intact or fusion, methods of purification). It is useful to have fusions to different proteins (e.g. lacZ and trpE) so that a sera raised against one fusion can be tested with the other, or one fusion protein can be used for affinity purification of sera; raised against another.
The choice of regions in your protein to use as antigen is somewhat arbitrary but there are several points worth mentioning. First, hydrophilic regions are often more strongly antigenic than are hydrophobic regions. Second, it is wise to try several different (preferably non-overlapping) regions if possible. Yields of fusion can be quite variable with different constructs, and one may also be a lot more antigenic than others. Also, if you have several sera directed against different portions of the protein which give identical results in histochemical, western experiments, etc. it is likely your sera are recognizing the protein you hope they are (in the absence of better controls).
The protocol below is based upon methods described for purifying insoluble trpE
and lacZ fusions. Therefore, while the conditions for the growth of cells and induction
are described for trpE fusions, it should in principle be applicable to purification of
any insoluble protein fusion. For general information about production and use of
antibodies, the laboratory manual by Harlow and Lane is highly recommended.
FUSION PROTEIN PURIFICATION-trpE fusions
1) Pick colony, grow overnight in L-broth with 100mg/ml AMP.
2) Inoculate M9-Casamino Acids media (100mg/ml AMP) 1:100; grow to O.D.600 of
approximately 0.2 and add a 1:1000 dilution of 10mg/ml -indoleacrylic acid
(Sigma). Let grow overnight (8-16hr).
3) Chill cells on ice, harvest 4krpm, 10'.
4) Store cell pellet at -80o.
(Volumes given in parentheses after instruction are for 25ml, 50ml, and 500ml cultures, respectively.)
5) Thaw cells on ice in 1/10 vol TEN (culture vol=1 vol), resuspend by vortexing.
(2.5ml, 5.0ml, 50ml)
6) Add lysozyme to 0.2mg/ml. (50µl, 100µl, 1ml)
7) Let sit on ice 20'.
8) Bring to 0.2% NP-40,10' on ice. (add 100µl, 200µl, 2ml)
9) Add .15 x culture vol 1.5M NaCl, 12mM MgCl2. (3.75ml, 7.5ml, 75ml)
10) Add DNase I to 4 µg/ml, leave on ice 1hr, stir regularly.
11) Layer onto sucrose cushion (at least 1/100 culture vol). Spin 30' 13krpm in
Sorvall HB-4 (swinging bucket) or equivalent.
12) Resuspend aggregate pellet in TEN and re-spin (to wash). This wash can be
repeated several times if desired.
13) The protein can be run over a prep gel; at the end of step 12 it should be at least
50% pure.
14) An extra step number, for luck.
Solutions:
TEN: 0.2 M NaCl, 10 mM Tris pH 8.0, 1 mM EDTA
sucrose cushion: 40% sucrose in TEN (filter sterilize and store at 4oC)
Assorted Notes:
1) Sonication can be substituted for DNase treatment (steps 9 and 10), both appeared
to work equally well in my hands. I found DNase treatment to be easier.
2) Protein can be electroeluted from the gel slice in protein gel running buffer and then dialyzed against 1xPBS. After quantitation, this material can be used for immunization. Generally, about 20mg/mouse is considered plenty for each immunization.
3) I found the use of protease inhibitors to be unnecessary. You may need them for your fusion, but since they are often nasty or expensive, I'd omit them if possible. It would seem likely that protein in inclusion bodies might be somewhat resistant to proteolysis, so any degradation products you see (you probably will see some) may be a result of degradation of protein during the period the cells are actively synthesizing it, and not during your purification.
REFERENCES
Science 214:1125 (1981)
J. Virol. 49:132(1984)
Cell 44:21 (1986)
Antibodies-A Laboratory Manual; Harlow, E., Lane, D. (1988) Cold Spring Harbor
Laboratory Press