FLP recombinase Expression

Escherichia coli Strain HBlOl  was used for constcuction and propagation of
plasmids.

The construction of the substrate plasmids pBAl12, pLB112 and pCS38
has been described previously (Andrew6 et al.. 1985; Beatty L SedOW6ki. 1988;
described previously Isabineau. et al, 19851. SChWartz 6 Sadowski. 19901. Plasmids pov56. pDv64, cvI9 and pBAlO4 were
Plasmid pLO3 contains the FLP codlng sequence joined to the
promoter. This plasmid was constructed in three steps [see Figure 81. (a1
ribosome binding site of the T7 gene 10 protein downstream from the Tl(10
The entire FLP coding sequence contained on a 1.4 kb fragment was exclsed
pRR3039 (from F.W. Studieri. This plasmid contains the T7t10 promoter. the
from plasmid pDv56 with SrI and ligated into the =HI site of plarnld
tranblation initiation site for the T7 gene 10 protein, a %HI cloning Site
followed by the transcrlptian terminator for T7 RNA polymerase. Cloning of
the FLP gene fragment from pOV56 gives a coding sequence consisting of the
first 11 codons from the gene 10 protein, four extraneous codons followed by
the entire FLP coding sequence. This plasmid was called pDC-1 (Figure EA!.
Although this Construct produced a larqe amount of a protein that was the
correct size to be the FLP fusion product inor shown!. the protein was
apparently insoluble and inactive. Plasmid pDC-1 was therefore modified to
remove the extraneous amino acid codons in front of the FLP coding region.
lb) Plasmid pOY56 was linearized with *I, which cuts in the ATG initiation
codon of the FLP gene. The 3' protruding ends were made blunt with T4 DNA
polymerase and deoxytriphosphates and a double nonphosphorylated +I linker
precipltation to remove excess linkers and heating to 80' for 5 minutes. the
15'-CATATGCRTATG-3'1 was ligated to the blunted termini. After spermine
plasmid termini were allowed to anneal and the resultant plasmids were
transformed into HB101. This pcocedure resulted in the replacement Of the
*I site in the ATG of the FLP gene with an -1 site. The resulting
plasmid was called pLB4 (Figure 801. (c1 A 1.1 kb fragment from pLB4 was
excised with *I and transferred to plasmid pDC-1 Chat had been Cut with
+I. The resulting plasmid. pLD3. has therefore lost the 15 extra codons
that were originally present in pDC-1 (Figure 8Cl. The DNA sequence of the
junction ceglon was verified by DNA sequencing. The plarrold was then
Brookhaven ~ational Laboratory!. This Strain contains a lambda prophage with
transferred into strain BL2 (*DE31 pLy6 5 [obtained fcom F.W. Studier,
the gene for Tl RNA polymerase under control of the & VV 5 pcomoter as well
as a plasmid (p~yssl that constitutively expresses the qene encoding T7
polymerase activity that ib made from the & UV5 promoter (Studier et al..
lysozyme. This enzyme binds to and inhibits any basal amount of T7
19901. 

Assays of PLP Activity
which Contains the A form of the 2 micron plasmid. One unit of activity is
ILP activity in crude extracts was assayed using the inversion plasmid cv19.
B form of the 2 micron plasmid (Bablneau et al., 19851. The activity of FLP
the amount Of protein required to invert 50 per cent Of the Substcate to the
was followed during the pur~flcdtion procedure using the llnear excision
Substrate, plasmid pBRlO4. A unlt Of activity is the amount of FLP required
to excise 50 per cent of the substrate (Babineau et al., 19851. The
composition of vac~ous fractions was assessed by SDS polyacrylamide gels 115
per cent separatlng el. cast in Bio-Rad mini-spparaturl, followed by
staining with coomassie blue. Protein concentrations were detemlned by the
Bradford colorlnetric method [Bradford, 19761.
Pucification of FLP Protein
Induction of PLP expression from Tl promoter We previously induced FLP
expression from the promoter by the addition of IPTG to the medium
pcarnoter using the same protocol. we were unable to detect any activity at
IBebineau et al., 19851. when we attempted to produce FLP from the T7
levels Of protein where FLP activity was readily detectable fcm the
construct. It was possible that the FLP protein was being highly expressed
from the Tl promoter, but that the protein vas inactive in the 5. cell
amount of a band that coaigrated with authentic FLP protein when cells
[Marston, 19861. This hypothesis was confirmed by the finding of a large
containing the plasmid pL03 were lnduced with IPTG, and analyzed by SDS-PAGE
fallwed by coomasrie blue staining Inat shown). This band was present one
hour after induction and its intensity increased markedly after three hour6
of induction. It therefore seemed that a large amount of inactive FLP
protein was being expressed from the Tl promoter. We therefore sought
methods which might facilitate the synthehls of enzymatically competent
protein and found that inductlan of FLP expression at low tenrperature led to
the synthesis of substantla1 amounts of active protein. When we induced FLP
expression at 2S0c we obtalned about forty-fold more FLP activity in crude
cell extracts than we previously detected in cell extracts where FLP was
expcessed from the % promoter at 31'.
Growth of cells A 200 1 volume of LB Brath 110 9 Tryptone. 5 g yeast
extract. 5 g NaCl, 0.1 ml. of 10 N NaUH per I! was inoculated with a 2 liter
overnight culture of strain PSl31 IBL2IACE31, pLysS, pLO31 that had been
grown at 31.. Both cultures canrained 20114 per ml of ampicillin and 30ug per
ml of chloramphenicol. The ZOO 1 culture was grown at 31' with an air flow
of 4 cubLC feet per minute, and an agltation speed of 200 cpm in a New
cooled to 22' and solid lPTG was added to a concentration of 5OppI. Growth
Brunswick fermentor, After the reached 0.73. the temperature was
was continued far six hours at 22O: the cells ".ere collected by
Centrlfugation and stored at -70'. 
Preparation of crude extras 1\11 the procedures were carried out at 0.4'~
weight1 were thawed with gentle stirring on ice in 450 m1 of sonication
unle55 otherwise indicated. One hundred and fifty grams Of frozen cells (wet
mixture was sonicated for 3 minutes to suspend the cells and then for twenty
buffer I50mM Tris-C1, pH 1.4, 50mM INH412SD4, 1nM EUTR, 0.5mM OTTI. The
periods of one minute. interspersed wrth 45 seconds of coaling in an ice-salt
bath. The sonicated extract I520 ml 18 ag/ml Of protein] was centrifuged in a
Beckman Type 70 Ti rotor at 40,000 rpm for 40 nin. The supernatant fraction
1500ml. 14 mq/ml Of protein) was loaded directly onto a Bio-Rex 10 column.
- Bio-Rei 10 chromatography Bia-Rex 70 cesln 1100-200 mesh1 vas packed into d
radial flow column (8.6 cm diameter. 11.5 cm high, superflo-500 column.
Sepragen Corpocat~onl. The column was equlllbrsted with BiO-Rex buffer 150mn
NaP04 buffer, pH 6.5. laH EDTA. 0.5~" DTT, 158 glycerol). The crude extract
was applied dlrectly to the column and the pa%s-throUgh fraction 1500 nll was
Cole-Parmer pump. The column was then washed with 1000 m1 of Bio-Rex buffer.
collected. The flow rate was maintained at 25-35 ml/nrin with the aid of a
with I000 a1 of Bio-lex buffer containing 700 an NaCl. The two 700 mM NaC1
and eluted with 1500 ml of Bio-Rex buffer containing 300 nM NaCl and finally
fractions (500 ml each1 contained >90% of the FLP activity. They were pooled
and the protein was concentrated with ammonium Sulfate.
Ammonium sulfate precipitation The pooled 700 all NaCl fractions from the
010-Rex column (1000 mlI were brought to 60% Saturation by the addition of
390 grams of sol~d ammaniun sulfate Iultrapure, Schwarz/Mann Biotech.] which
was dissolved by stirring at 4'c for 30 mln.. The protein was pelleted by
precipitate vas dlssalved in 25 a1 of Blo-Rex buffer IRS fraction, pcotein
centrifugation at 16.000 rpm for 20 m~n. in the Sarvall SS34 rotor. The
concentration - 19 ng/mli and the fraction was frozen immediately at - 10'C

http://www.jbc.org/content/266/17/11347.long
""til application to a gel filtration column.
s?@hcryl 5-300 gel filtrat_lo-n chromatography A Sephacryl 5-100 IPharmacial
column was packed at room temperature 15 Crn x 90 cm: 1,640 ml bed voluneI and
equilibrated I" the cold wlth 5-100 elutlon buffer I50aN NIP04 buffer, pH
6.7. ImH EDTA. 1mM DTT, 20% glycerol and 150aM NaC11. Ten a1 of the AS
fraction were loaded onto the column and elution was continued with 5-100
buffer at a flowrate of 2.5 al/min. The first 300 ml were collected as a
single void fractlon and then 12 ml fractlons Yece collected with a fraction
collector. Prote~n cancentrat~an was monitored by Bradford assays. The FLP
actlvity eluted after about 0.6 column volume and about 15% of the starting
activity was present in 6 fractions Of specific activity greater than IO5
units per mg of protexn [Fig. 91\ lanes 2-11. These fractions were pooled and
stored at -7O'C.
FPLC ion exchange chromatography &n FPLC cation exchange column Iflono S HR
5/5. 5x50 mnl was equllibrated in buffer A 150aM NaP04 buffer, pH 6.1, 0.2mN
EDTA. ImH UTT. 25% glycerol) on a FPLC system (Phacmacia). Nine nl 14.1 mg
protein) of the pooled fractions from Sephacryl s-300 column was directly
applied to the column at a fiawrate of 0.5 al/ain at room temperature. The
A. protein concentration was monitored by uv absorbance at 280 m and 0.5 ml
 ample vas eluted with a 32.6 ml salt gradient 10.00-700 ml NaCll in buffer
fractions wece collected. FPLC fcactions were analyzed for purity by SDS-PAGE
and staining with Coomassie Blue and for FLP activity. The pure fractions of
specific actlvlty greater than 4.8~10~ units per mg Of protein were pooled
and stared at -loot.
CoYente on the purification The purification scheme is surnmarired in Table
a fast flowrate. FurtheCmOre, the resin can be regenerated several tines with
1. The initial Bio-Rex chromatography step used a radial flow column that has
alkali and the column can be repacked. This step yields a fraction in which
FLP in about 108 pure and the major contaminants are Of low molecular weight
(Fig. 9 C and 0. lane 21. The Sephacryl column also flows rapidly, remove6
meny Of these contaminants and results in a few fold purification with 61
percent recovery [Fig. 9 C and D, lane 3. Table 11. In the purification we
described here. we used 40 percent Of the annonivln Sulfate fraction. but
double the volun~e can be applied to the Sephacryl column with little
deterioration in purification. We have chromatographed less than 5 mg of the
5-300 fraction to the Nono s column and recover 1.5-2 mg of homogeneous FLP
per run. TO amass milligcamb of homogeneous FLP fcom 150 gcamb of frozen
cells would require up to twenty FPLC runs, although several 1""s can be done
in one day. This procedure is capable Of generating 30 mg of homogenous FLP
Stability Of the purified ILP The aanonlum Sulfate fraction has been stable
from 150 grams of cell6 with excellent yields.
for one year at -70°C. The Sephacryl fractlons are stable far up to 6 months
at -1O'C and will withstand three cycles of freezing and thawing. The FPLC
several months.
fraction can withstand freezing and thawing several times and is stable for
when stained with Comna~sie Blue (Fig.90; 9C lane 41. Since the gel lane
Purity Of the purified FLP FLP in FPLC fraction6 YEIS apparently homogenous
contained 8 pg of FLP protein (Fig. 4C lane 41 and since 100 ng of protein is
readily visible as a Coornambie slue-stained band in such a gel. we ebtinarc
estimate of the degree of contamination, we stained the same gel with sllver
that the pceparation is greater than 98% puce. To obtain a nore sensitive
and found scarcely any contaminating bands. The Oo280/oo260 ratio was 1.99,
indicating that the preparation 16 free of nucleic acid. The was
11.61 using the Bradford method and bovine plesma albumin as the standard. 
The FLP protein war purified from an E.m transfarmant carry~ng plasmid
pLD3 as described above. R Sephacryl S-300 fraction of greater than 85%
purlty was used. ~estriction enzymes were from Boehringer-mnnhein, Pharaacia
and BRL. AMY reverse transcriptase Yao fr