|
Conjugative transposons are genetic elements found
integrated into bacterial chromosomes that move from donor to recipient
cells by a mechanism that involves cell-cell contact. They carry
genes that confer antibiotic resistance. Tn916 the archetype conjugative
transposon encodes two proteins Int and Xis that are required for
excision and integration of the conjugative element from the donor
cell's DNA to the recipient cell's DNA. Purification of Xis involves
preparation of a lysate in 1M NaCl from bacteria carrying a plasmid
that expresses from an heterogeneous promoter a cloned Xis gene.
The salt concentration must then be adjusted so that it is low enough
for Xis in the lysate to bind to heparin agarose. Following elution
from the heparin agarose by an increasing concentration of NaCl
the protein is further purified by gel filtration. Variations in
the procedure used to produce the lysate and adjust the salt concentration
prior to heparin agarose chromatography have not been sufficiently
investigated to yield a robust purification protocol. In order to
optimize the protocol to give reproducible yield of Xis protein
I have investigated the conditions of cell lysis and adjustment
of salt concentration prior to chromatography. I have determined
the number of bursts of soncation to lyse over 90% of the bacteria
and the dialysis conditions necessary to adjust the salt concentration
of the lysate.
Conjugative transposons which typically confer antibiotic
resistance propagate from bacterium to bacterium by excising themselves
from the donor's DNA---Tn916 is the best studied conjugative transposon
(Churchward 2002; in Mobile DNA II Craig N et al. eds ASM Press
Washington D.C.). During transposition the element excises to form
a circular intermediate transfers from donor to recipient and integrates
into the recipient's DNA; the type of recombination involved in
integration and excision is extensively reviewed in the book Mobile
DNA II. Several transposon encoded proteins are responsible for
transfer of Tn916 from donor to recipient. These are Int Xis and
conjugal proteins (Churchward 2002). Int recombinase is a sequence
specific DNA binding protein with residues that catalyze DNA cleavage
and DNA ligation during excision and integration. Int cannot excise
the element without the aid of another DNA sequence specific binding
protein Xis. For the transposon to transfer conjugal proteins whose
individual functions are unknown act in similar ways as transfer
proteins of conjugal plasmids. Xis is a small architectural protein
that binds to specific DNA regions within Tn916; it is thought that
Xis establishes DNA-protein interactions which activate DNA recombination
(Mobile DNA II). Xis has two known functions--- transposon excision
and regulation of transposition (Hinerfeld and Churchward 2001;
Molec. Microbiol. 41: 1459-1467). If Xis binds to the left end of
Tn916 excision in vivo is more likely to occur; in fact if there
is a mutation in this binding site excision is reduced. Thus Xis
is necessary for excision. If Xis binds to the right end of the
element it takes on a regulatory role; excision is less likely to
occur. The published procedure for purifying Xis (Rudy et al. 1997;
Nucl. Acids Research 25:4061-4066) using a cloned gene inserted
into a plasmid involves lysing cells to produce lysate raising the
salt concentration to 1M NaCl to increase the amount of Xis in the
soluble fraction of the lysate lowering the salt concentration of
the lysate so that Xis will bind to heparin agarose eluting the
protein from the heparin with an increasing salt concentration and
gel filtration. This procedure has not been reproducible and so
we have investigated different steps of this protocol. We have focused
on cell lysis and on the dialysis step required to reduce the salt
concentration of the lysate prior to heparin agarose chromatography.
We have established conditions that lead to maximal cell lysis and
efficient binding of Xis to heparin agarose.
Bacterial cell lysis
A suspension of bacterial cells (30ml) was placed
in a 50ml plastic tube which was held in a beaker of crushed ice.
A 5mm sonication probe (Branson) was inserted as deeply as possible
into the suspension. Sonication was carried out in 30 sec bursts
at maximum power with 30 sec cooling periods in between. After each
burst an aliquot of the suspension was removed and centrifuged for
5 min in a microfuge at maximum speed. 10ml of the resulting supernatant
was diluted 100x in water and the OD260 of the diluted supernatant
was determined. The OD260 value of the last sample was set at 100%
to normalize the data.
Dilution of NaCl by dialysis
A standard curve relating NaCl concentration to conductivity
was made by creating a series of dilutions of buffer B (3.0M NaCl
50mM Tris-HCl 1mM EDTA) and determining their conductivity using
a conductivity meter equipped with a dip probe. 30ml of buffer C
(75mM NaCl 50mM Tris-HCl 1mM EDTA) and 15ml of buffer B (3.0M NaCl
50mM Tris-HCl 1mM EDTA) were mixed in a dialysis sac and dialyzed
against 4 L of buffer C at 4aC. Samples were removed at intervals
their conductivity determined with the conductivity meter and the
NaCl concentration in the sample was determined from the standard
curve. After 5 hours the buffer was changed.
Purification of Xis
E. coli carrying a plasmid expressing a cloned xis
gene were streaked out on an LA plate containing ampicillin (50mg/
ml) and chloramphenicol (34mg/ ml) and incubated overnight. Next
day a 100ml culture in LB containing the same concentrations of
antibiotics was inoculated with a single colony and incubated overnight.
Four 500ml cultures of LB containing antibiotics were inoculated
with 18ml of overnight culture and incubated at 37aC until the OD600
reached between 0.4 and 0.6. IPTG was added to each culture to a
final concentration of 0.8mM and the cultures were incubated for
4 hours at 37aC. Bacteria were harvested by centrifugation at 5.000g
for 7.5 minutes at 4aC the supernatant was discarded and the cell
pellets were stored at ¡V80aC overnight. The following day
the cell pellets were thawed on ice suspended in 30ml of modified
buffer C (75mM NaCl 50mM Tris-HCl 1mM EDTA) lysed by sonication
(on ice; 30 second bursts with a 30 second break between each burst).
15ml of buffer B (3.0M NaCl 50mM Tris-HCl 1mM EDTA) was added to
the lysate which was gently shaken at 4aC for 30 minutes. The lysate
was centrifuged at 47 000g for 20 minutes at 4aC. The supernatant
was placed in pre-soaked dialysis tubing (3 500 MWCO) and dialyzed
overnight against 4 liters of buffer C at 4aC. The buffer was changed
after 5 hours and dialysis continued for a further 16 hours. The
dialyzed lysate was centrifuged at 47 000g at 4aC for 20 minutes.
The supernatant was applied to a 5ml Hi-Trap heparin column. The
column was washed with 50ml of buffer C then eluted with a 150ml
gradient of 75mM to 3M NaCl. Fractions containing Xis were pooled
concentrated to a final volume of 1.0ml by centrifugation in a Centricon
YM-3 concentrator and applied to a Superdex 75 column. The column
was eluted with buffer E (0.5M NaCl 50mM Tris-HCl 1mM EDTA) and
the Xis-containing fractions were pooled.
Cell lysis. To determine the maximum efficiency of
sonication we measured the release of UV absorbing material into
the culture supernatant after repeated bursts of sonication. The
majority of UV absorbing material in the bacteria is RNA and DNA.
The results shown in Fig. 1 show that a minimum of 5 bursts was
required to release ƒ95% of the UV absorbing material from
the bacteria. 3 or 4 bursts which is the number commonly used resulted
in the release of ƒ¬75% of the material implying that 25%
of any cytoplasmic soluble protein will be lost in the pellet fraction
upon subsequent centrifugation of the lysate. Dialysis. To measure
the rate of which dialysis reduced NaCl concentration in a bacterial
lysate we set up a mock dialysis using conditions similar to those
employed in an actual protein purification. We constructed a standard
curve relating NaCl concentration to conductivity (Fig. 2). We then
used this curve to measure the concentration of NaCl within a dialysis
sac during dialysis. As shown in Fig. 3 the concentration of NaCl
falls during dialysis. The rate of reduction of NaCl concentration
was approximately 2x every 1.5 hours. After 5 hours the concentration
of NaCl within the sac remained significantly above the level of
the surrounding buffer. Purification of Xis. Based on the results
in the proceeding sections we set out to purify Xis from a bacterial
lysate. Fig. 4 shows that the majority of Xis present in the cell
lysate remained in the soluble fraction after dialysis to reduce
the salt concentration (compare lanes 1 and 3). Dialysis using the
procedure described for figure 3 using modified buffer C (75mM NaCl)
reduced the salt concentration so that no Xis was detectable in
the flow through of the heparin column (compare lanes 3 4 and 5)
Xis was eluted from the heparin column in the first four fractions
(lanes 6-9). Xis protein-containing fractions from the heparin column
were pooled concentrated and applied to a Superdex 75 column. Fractions
eluting from the column are shown in fig. 5. Comparison of lanes
3-13 shows that the Superdex column efficiently fractionated material
eluting from the heparin column o the basis of molecular weight
with the small Xis protein eluting in the last fraction. The entire
purification is summarized in fig. 6.
Experience gained in two unsuccessful attempts to purify Tn916
Xis protein led us to focus on the efficiency of cell lysis and
on the dialysis step performed to reduce the concentration of NaCl
in the cell lysate prior to heparin agarose chromatography. We found
that five or more bursts of sonication were required for maximal
cell lysis. The published procedure calls for four bursts which
according to our results leaves 25% of the potential soluble contents
of the cell in a state that can be pelleted by centrifugation. The
unsuccessful attempts to purify Xis followed the published procedure
for dialysis of the lysate. However dialysis was carried out for
different times and in each case all the Xis present in the lysate
passed through the heparin column in the column flow-through. The
conditions for dialysis established here coupled with a reduction
in the salt concentration in the buffer used for dialysis resulted
in all the detectable Xis present in the dialyzed lysate binding
to the column.
Despite the improvements in the purification protocol comparison
of our results with the published results show that the final yield
of Xis in the purification summarized in figure 6 is substantially
lower. In order to improve the yield we consider three possibilities
for future experimentation. The first is to determine if during
long-term storage the plasmid expressing Xis has undergone mutation
to reduce the level of expression. The second is to determine the
stability of the plasmid and to investigate culture conditions that
maximize the fraction of cells in the culture that carry the plasmid
at the time of induction. If either of these approached indicate
that Xis is deleterious to the host cells (presence of mutation;
loss of plasmid during growth) we would consider recloning the Xis
gene so that expression is more tightly regulated.
Thank you to Dr. Churchward, Churchward Lab, Moran Lab, and Igor
Lab. This work was supported by Howard Hughes Medical Institute
Grant No. 52003727, National Institutes for Health Grant No. T32
AI07470, and National Science Foundation Grant No. MBC - 0131471.
|
