Binding of native bacterial
protein SlyD to metal affinity matrices remains
a major problem in affinity purification of His-tagged recombinant proteins
from Escherichia coli cells. In this
study, four novel E. coli strains that
lack the expression of SlyD as well as SlyX, were engineered using l-red
mediated chromosomal deletion. The resultant SlyD/SlyX-deficient E. coli
strains allow us to obtain contaminant-free proteins immediately after metal
affinity chromatography, and eliminate additional purification processes. As a model protein, bispecific antibodies composed of
anti-F4/80 VHH module and
anti-TNF VHH module (MYSTI-2) were used. Using this protein we
have shown that the SlyD/SlyX-deficient E. coli strains allow to obtain
a fully functional protein.
metal affinity chromatography (IMAC) is the most important, and an extremely
technique for protein purification in low and high throughput
environments (1–3). IMAC involves the fusion of
4–10 histidine tags to the target protein. These tags interact with a
chromatography matrix charged with metal ions (mainly nickel and cobalt) because
of the affinity of histidine for these ions (4). The advantages of IMAC
include high protein loading capacity, ligand stability, and the possibility of
multiple column regeneration at low cost. Moreover, his-tag is a unique tag that
allows protein purification at native as well as denaturing conditions. However,
new challenges arise along with the progress in recombinant protein expression
and purification methods. Expression of a heterologous protein in a cell causes
the initiation of cellular response to foreign molecules and over-production of
stress responsive enzymes (5). Some of these native
bacterial proteins show high affinity to divalent cations that allow co-purification
of these proteins along with the target protein. Metal
binding proteins are present in E. coli
strains of various genetic backgrounds, such as BL21(DE3), C41(DE3), C43(DE3),
Rosetta 2(DE3), Origami(DE3), B834(DE3), and XL1-blue
etc. The co-purification of contaminant and recombinant proteins is
particularly problematic when one or more E.
coli native proteins are expressed at high levels and/or when their size is
similar to that of the recombinant protein. This situation might ultimately
result in the purification of an undesired bacterial protein rather than the
recombinant protein of interest (6).
The most common contaminant of
IMAC purification is SlyD, a 196-amino acid protein (Mw=20.85
kDa), known to be present in E. coli during cell lysis that
occurs due to bacteriophage fX174 infection (7). SlyD consists of a peptidyl-prolyl isomerase (PPIase) domain,
molecular chaperone domain, and a C-terminal tail rich in metal-binding
residues responsible for the high affinity of this protein for divalent
cations. It was initially discovered as
a major contaminant of non-denaturing IMAC-based procedures. SlyD has
the highest percentage of histidine residues (10.2%) among almost 20 proteins that
get co-purified along with the target protein during IMAC. The high affinity of
SlyD for metal ions is retained even under strong denaturing conditions such as under 8 M urea or 6 M guanidine hydrochloride. Consequently,
the complete removal of SlyD is a very difficult task, which leads to
significant loss of recombinant protein during additional purification
procedures. Presence of SlyD interferes with analysis and the subsequent use of
recombinant proteins for proteomic analysis, studies of protein-protein and
protein-nucleic acid interactions, crystallization, etc. (8). Proteins used in
therapy must be absolutely pure to ensure that the pharmaceutical effect is solely
because of the target protein and not because of one of the contaminants.
of the current approaches to isolate SlyD and other contaminants from a target
protein is the use of a two-step chromatographic method: IMAC is the capture
step and size exclusion chromatography (SEC) is the polishing step (9). This approach allows to successfully resolve a target protein from
SlyD, and to achieve >95% purity of the desired protein. However, this approach can be considered only when the molecular
weight of the desired protein sufficiently differs from that of SlyD for
separation by SEC.
method to remove contaminants involves the use of double tags, i.e., the combination
of polyhistidine with glutathione-S-transferase, maltose-binding protein, or
other tags (10–12). However, it requires excessive
sample manipulation during cleavage of additional tags, which may lead to proteolysis,
aggregation, and insolubility of the recombinant protein (13).
study, we aimed to obtain a SlyD-free target protein, immediately after IMAC,
without any additional purification steps. On the E.coli genome upstream of slyD located slyX (Figure 1). The SlyX function is unknown, but the C-terminal
domain contains a conserved sequence (PPHY) responsible for the secretion of
Insulin-degrading enzyme (14). It can be assumed that bacterial SlyX is involved in
proteolytic degradation of proteins. Therefore, we decided to knock out both slyD and slyX.
To achieve this goal, the chromosomal sequence of SlyD/SlyX
in the E. coli genome was replaced by
an antibiotic resistance gene using the l-red chromosomal deletion strategy (15). A set of E. coli
strains that are commonly used for heterologous expression were modified to
obtain the new strains for SlyD/SlyX-free protein expression.
of bacterial strains extends the range for
purification of potential recombinant proteins.
The strains of E. coli included in
this study were BL21(DE3), C41(DE3), C43(DE3), and B834(DE3).
assess the functional activity of proteins derived from SlyD/SlyX-deficient E.
coli strains we chose a therapeutic protein MYSTI-2. It is a bispecific
antibody composed of anti-F4/80 VHH and anti-TNF VHH
module and has a 6-histidine tag at the C-terminus. The structure of this
protein was described in detail previously (16).
Materials and Methods
Strains and Plasmids
In this study, all the
expression and purification experiments were performed using E. coli “DE3” strains, namely, BL21,
B834, C43, and C41. BL21(DE3) strain is deficient in the OmpT and Lon
proteases, which may interfere with the isolation of intact recombinant
proteins (17). E. coli DE3 strains contain lambda DE3 lysogen, and are intended
to be used for strong expression of proteins encoded by plasmids with the T7 RNA
polymerase promoter. The C41(DE3) and C43(DE3) strains were derived
from BL21(DE3), as described in a previous study (18). These strains have at least
one uncharacterized mutation that prevents the occurrence of cell death
associated with expression of highly toxic membrane proteins. The B834(DE3)
strain is a Met auxotroph. This strain has been widely used for highly specific
35Smethionine-labeling of target proteins, and recently, for
selenomethionyl derivatization of proteins for X-ray crystallography (19). Detailed
information about the bacterial strains used in this study is provided in Table 1.
MYSTI-2 construct, which has a 6-histidine tag at the C-terminal, was
cloned into plasmid pET22b+ as previously described (16). The template plasmid pKD4 is
a derivative of pANTSg that contains an
FRT-flanked kanamycin resistance (KmR)
gene from pCP15 (15). The Red helper plasmid pKD46
is a derivative of pINT-ts that contains araC-ParaB and g b exo with the tL3 terminator
downstream of exo. This plasmid
expresses the Red system under the control of a well-regulated ParaB promoter
to avoid unwanted recombination events from taking place under non-inducing
conditions. pCP20 is an ampicillin (ApR) and chloramphenicol (CmR)
resistance plasmid that exhibits temperature-sensitive replication and thermal
induction of FLP synthesis (20).