|
Expression of mammalian proteins in bacteria is a
useful method to create the amounts of these proteins needed for
biochemical analyses. A commonly used method includes the use of
IPTG to induce protein expression in BL21cells carrying pET-based
vectors. However, plasmids are unstable in BL21 cells grown in LB
medium and cell density must be monitored regularly for optimal
induction in log phase growth. Studier and colleagues have developed
new media and methods that eliminate these limitations upon the
use of BL21 cells for protein expression. The new system uses auto-induction
of the desired protein at high cell density to also obtain substantially
higher levels of protein expression. This method was tested with
cultures of BL21 cells that were transformed with plasmids directing
expression of five Arf proteins and 16 Arf-like (Arl) or Sar proteins.
We found that the use of P-0.5G minimal medium allowed BL21 cells
to much more stably maintain the pET vectors than did LB medium.
In addition, with only a few exceptions (Arf4 and Arf5) this newer
system of protein expression allowed much higher levels of protein
to be expressed in bacteria. Fully 15 of the 21 proteins tested
expressed to high enough levels that they were readily observed
in Coomassie blue stained samples of bacterial lysates. As most
of the Arl proteins have never been purified or analysed, we began
the process of purifying as many of these as possible, to allow
biochemical analyses of these regulatory proteins for the first
time. We adopted methods previously determined for Arf purification
(Randazzo et al., 19952) from bacteria to generate preparations
of Arl2, Arl3, Arl5, Arl5B, Arl8, and Arl11. From one liter of bacterial
culture we typically obtained in excess of 50 mg purified proteins
in each case. Thus, these new media and system for bacterial protein
production appear to significantly enhance our ability to obtain
large amounts of purified Arfs and Arls.
The Arf family is a group of proteins that are about
21 kDaltons in size. They are members of the Ras superfamily of
GTP-binding proteins used in cell signaling and regulation. There
are two divisions of the Arf family, based on function, Arf and
Arf-like (Arl) proteins, with a total of 21 Arf family members in
human cells. The most common method of generating large amounts
(mgs) of protein for biochemical and structural studies is the pET
vector and BL21 cell growth in LB medium, as first developed by
Studier, et al (Studier and Moffatt, 19863; Studier et al., 19904).
While very useful for many years, this system suffers from variability
in protein expression, likely caused by the need to use freshly
transformed bacteria in LB medium as plasmids are unstable, and
the need to monitor cell density carefully to get optimal protein
expression. Studier and colleagues have now developed a newer method
of protein expression in bacteria involving media that create more
stable transformants as well as producing higher protein expression,
without the need for induction. E. coli are transformed with pET-based
plasmids as in the earlier method, but cultures are then grown in
a medium (P-0.5G) in which expression is suppressed and plasmids
are very stable. A small sample of this culture can be used to inoculate
medium (ZYP-5052) that promotes auto-induction at high cell density.
Expression is much higher and proteins are anecdotal evidence suggests
proteins may remain more soluble.
I. Media Preparation
LB medium was prepared according to standard practice. The new media
included P-0.5G, in which BL21 cells were grown for long- and short-term
storage and used to inoculate large scale cultures for protein expression,
and ZYP-5052, used for large scale protein expressions. P-0.5G consisted
of 1 mM MgSO4, .5% glucose, .025 M (NH4)2SO4, .05 M KH2PO4, .05
M Na2HPO4, 100 ??g/mL methionine, 50 ??g/mL ampicillin, 5 ??M Fe,
2 ??M Ca, 1 ??M Mn, 1 ??M Zn, .2 ??M Co, .2 ??M Cu, .2 ??M Ni, .2
??M Mo, .2 ??M Se, and .2 ??M H3BO3. The final concentrations of
reagents in ZYP-5052 were 1 mM MgSO4, .5% glycerol, .05% glucose,
.2% ? -lactose, .025 M (NH4)2SO4, .05 M KH2PO4, .05 M Na2HPO4,
50 ??g/mL ampicillin, 50 ??M Fe, 20 ??M Ca, 10 ??M Mn, 10 ??M Zn,
2 ??M Co, 2 ??M Cu, 2 ??M Ni, 2 ??M Mo, 2 ??M Se, and 2 ??M H3BO3.
II. Transformation
BL21(DE3) cells were transformed with pET-based plasmids made by
former Kahn lab member Shana Kerr, that direct expression of each
of the 21 human Arf family members. Transformants were plated onto
LB+ampicillin plates and grown overnight for selection
III. Culture in P-0.5G
Colonies were picked the next day and used to inoculate liquid growth
in P-0.5G medium. After overnight growth, an aliquot of this culture
was mixed with 80% glycerol and stored at -80 degrees C. Although
not done in our studies these frozen cells can be used to inoculate
ZYP-5052 cells directly in the future.
IV. Culture in ZYP-5052 ZYP-5052
(1 L) cultures were inoculated with a ~0.1- 1 mL of a P-0.5G culture
and grown overnight at room temperature (RT) and at 37 degrees C.
These cultures were tested for expression by resolving bacterial
lysates in SDS-PAGE gels and staining with Coomassie blue to visualize
proteins.
V. Cell Lysis and Protein Purification
Pellets were resuspended in TM buffer (20 mM Tris-Cl pH 7.5 1 mM
MgSO4) and lysed by three passes through a French pressure cell.
Soluble proteins were obtained by centrifugation at 100 000xg for
one hour and the supernatant was run on a Macro Q column for ion
exchange chromatography on an FPLC machine using a linear gradient
of 0-1000 mM NaCl. An SDS-PAGE gel was run on fractions chosen from
UV readout. Most of the GTPases eluted very early from this column
which effectively resolves them from the majority of bacterial proteins.
VI. Concentration and Further Protein Purification
Fractions from the Macro Q column containing the desired protein
were pooled and put in a stirred cell concentrator with a PM10 filter.
When the volume was ~2-5 ml it was applied onto a Sephacryl S75
column for gel filtration chromatography. Fractions were again analyzed
by SDS-PAGE gel to determine which fractions contained the desired
protein. These fractions were pooled concentrated and frozen at
-80 degrees C for later analyses.
The methods and media developed by Studier, et al are a much faster,
easier, more reproducible, and cheaper way of producing proteins
on a large scale than the current widely-used method. IPTG is unnecessary,
re-transformation of bacteria is unnecessary, and higher expression
is achieved. Altogether this is a far superior method of obtaining
proteins for study. In the future, nucleotide binding to the new
Arls will be studied by members of the Kahn lab. This will include
defining under what conditions Arls bind GTP and GDP, if lipids
are necessary in binding activity, if there is GTPase activity,
whether Arl proteins bind directly to cell membranes as Arf proteins
do, and whether Arls are post-translationally modified as Arfs are
(N-myristoylated) or Arfrp is (N-acetylated). Mass spectroscopy
of these proteins purified from mammalian cells will also be done
to learn if there is a group covalently attached to the Arl proteins.
Eventually, specificity for binding to Arf GEFs and GAPs will be
tested as well. Such studies will teach us a lot about how cells
regulate a large number of essential cellular processes, including
membrane traffic and the cytoskeleton.
This work was supported by NIH funding provided to the Kahn lab
of Emory School of Medicine, Department of Biochemistry, Atlanta,
GA. Acknowledgements go to the members of the Kahn lab for their
assistance and expertise.
References
1 Kahn, R. A. (2004). ARF Family GTPases, Vol 1 (Dordrecht, Kluwer
Academic Publishers).
2 Randazzo, P. A., Weiss, O., and Kahn, R. A. (1995). Preparation
of recombinant ADP-ribosylation factor. Methods Enzymol 257, 128-135.
3 Studier, F. W., and Moffatt, B. A. (1986). Use of bacteriophage
T7 RNA polymerase to direct selective high-level expression of cloned
genes. J Mol Biol 189, 113-130. 4Studier, F. W., Rosenberg, A. H.,
Dunn, J. J., and Dubendorff, J. W. (1990). Use of T7 RNA polymerase
to direct expression of cloned genes. Methods Enzymol 185, 60-89.
The ability to make a large amount of proteins is important to
understanding how they work. This involves transforming bacteria
with plasmids, or making them take up a small, circular piece of
DNA that contains a sequence that codes for the protein you want
to express. A very common yet cumbersome method of making transformants
uses a chemical called IPTG, which you can only add at a precise
point in bacterial growth. The medium that the bacteria is grown
in is a simple mixture called LB. The plasmids are not stable in
bacteria grown in this medium, so if you want more bacteria with
plasmids in them, you have to transform those bacteria again. Studier
labs have developed a new set of media in which plasmids are stable
in the bacteria, and in which you don't have to add IPTG. We transformed
bacteria with plasmids that contained sequences for expression of
proteins called Arf proteins, to discern if this new method worked
better than the old one.
Bacterial transformation, Plasmid DNA isolation, gel electrophoresis,
SDS-PAGE, culturing, centrifugation, protein determination, French
press, FPLC (ion exchange chromatography, gel filtration chromatography),
spectrophotometry, and liquid scintillation.
|
