Bgdg

f BGAF, BGFL

benzyl)acetamide, potassium tert-butoxide, dmf, 88%; (c) l<2C03, methanol, 85%; (d) N-(+)-biotinyl-6-aminocaproic acid N-succinimidyl ester, triethylamine, DMF, 69%; (e) Digoxi-genin-3-0-methylcarbonyl-6-aminocaproic acid N-succinimidyl ester, DMF, triethylamine, 87%; (f) 5(6)-carboxyfluorescein diacetate N-succinimidyl ester (mixture of isomers), triethylamine, DMF, 8% (BGFL), 2% (BGAF).

Fig. 4.4.2. Labeling of W160hAGT fusion proteins in vitro and in vivo. (A) Analysis of the in vitro reaction ofW160hAGT (0.4 |im) with BGBT (15 |im). Aliquots were taken from the reaction mixture at the indicated time points and quenched with oligonucleotide containing BGBT (1 |im). Biotinylated W160hAGTwas detected by Western blotting using a strep-tavidin-peroxidase conjugate. The band Ycorre-sponds to a co-purified proteolytic degradation product of W160hAGT. (B) Fitting of the intensities obtained from the Western blot in (A) to a first-order reaction model, resulting in a

Fig. 4.4.2. Labeling of W160hAGT fusion proteins in vitro and in vivo. (A) Analysis of the in vitro reaction ofW160hAGT (0.4 |im) with BGBT (15 |im). Aliquots were taken from the reaction mixture at the indicated time points and quenched with oligonucleotide containing BGBT (1 |im). Biotinylated W160hAGTwas detected by Western blotting using a strep-tavidin-peroxidase conjugate. The band Ycorre-sponds to a co-purified proteolytic degradation product of W160hAGT. (B) Fitting of the intensities obtained from the Western blot in (A) to a first-order reaction model, resulting in a second-order rate constant of 600 M

for the reaction of W160hAGT with BGBT. (C) Analysis of total cell extract of E. coli BL21(DE3) expressing W160hAGTwith and without BGBT in the medium. Lane 1: SDS-PAGE of BL21(DE3) expressing W160hAGT and subsequent staining with Coomassie Blue; lanes

2-4: Western blot of total cell extracts of BL21(DE3) expressing or not expressing W160hAGT in the presence or absence of BGBT after 120 min. In lane 2, CCP was expressed instead ofW160hAGT. A streptavidin-peroxidase conjugate was used to detect biotin-labeled proteins. The band at 20 kDa corresponds to a protein that is biotinylated in E. coli in the absence of BGBT. (D) Analysis of cell lysates of yeast S. cerevisiae expressing or not expressing a W160hAGT-DHFR fusion protein in the presence or absence of BGBT (10 |im, 2.5 h) using an ELISA. Samples in columns B-D were also incubated with an oligonucleotide containing the nucleobase BGBTafter lysis. In column D a Ste14p-DHFR fusion protein was overexpressed instead of W160hAGT-DHFR. An anti-HA antibody was used as the primary antibody and an anti-mouse-HRP conjugate as a secondary antibody.

BGBT in E. coli was estimated to be 24% by comparing the signals resulting from the in-vivo labeling with those obtained from succeeding labeling of cells after lysis with oligonucleotides containing BGBT.

To demonstrate that the covalent labeling of hAGT fusion proteins in vivo can be applied to different hosts, we fused W160hAGT N-terminally to dihydrofolate re-

ductase (DHFR) from mouse to yield W160hAGT-DHFR and expressed this fusion protein in yeast S. cerevisiae. In this construct DHFR also carried a C-terminal HA epitope tag. Analysis of the biotinylation of W160hAGT-DHFR in yeast using Western blots proved to be impossible as the streptavidin-peroxidase conjugate strongly cross-reacted with several yeast proteins, one of them of the same size as the W160hAGT-DHFR fusion protein. To demonstrate that the W160hAGT-DHFR fusion protein can be biotinylated in yeast, cells expressing W160hAGT-DHFR were incubated with BGBT (10 mM), then lysed and the cell extracts transferred into streptavidin-coated microtiter plates. After washing of the wells the immobilized W160hAGT-DHFR fusion protein was detected by ELISA using an anti-HA antibody as the primary antibody (Figure 4.4.2D). The signals in the ELISA obtained from such samples were significantly above background, which was defined as the signal obtained from yeast not incubated with BGBT. To quantify the efficiency of the labeling in yeast, the signal from lysates that were treated with oligonucleotides containing BGBT after labeling in vivo were compared with those that were labeled solely in vivo (Figure 4.4.2D). These data enabled us to estimate that 10% of W160hAGT-DHFR is biotinylated in vivo under these conditions. No signal above background was observed in this assay when a different DHFR fusion protein, a fusion protein with the S. cerevisiae Ste14p (Ste14p-DHFR), was expressed using the same vector and incubated in vivo and in cell extracts with BGBT (Figure 4.4.2D). The relatively low efficiency of the in-vivo labeling with BGBT in E. coli and yeast is most probably because of the low intracellular concentration of the substrate, because biotin derivatives with similar linker structures as BGBT have been shown to have relatively low cell permeability [5].

To demonstrate the feasibility of our approach in standard mammalian cell cultures, we investigated the labeling of a W160hAGT fusion protein with fluorescein in Chinese hamster ovarian (CHO) cells using BGAF. In contrast to AGTs from yeast and E. coli all mammalian AGTs characterized so far accept BG as substrate [11]. To achieve specific labeling of the W160hAGT fusion protein, we used a previously described AGT-deficient CHO cell line [14]. To facilitate evaluation of the fluorescence labeling, a W160hAGT fusion protein was constructed and targeted to the nucleus of the CHO cells by fusing W160hAGT to three consecutive SV40 large T antigen nuclear localization sequences (NLS), yielding W160hAGT-NLS3 [15]. To demonstrate that W160hAGT-NLS3 fusion proteins are targeted to the nucleus of the CHO cell, we transfected CHO cells with a vector expressing a W160hAGT-ECFP-NLS3 fusion protein containing enhanced cyan fluorescent protein (ECFP). The location of W160hAGT-ECFP-NLS3 in the nucleus of the cell was verified by confocal fluorescence microscopy via fluorescence of ECFP. CHO cells transiently transfected with a vector expressing W160hAGT-NLS3 were incubated with BGAF (5 mm) for 5 min, washed, and monitored with confocal fluorescence microscopy (Figure 4.4.3A-C). During and after the washing step excess fluorophores leaked out of the cell and after 25 min only the nucleus of the cell had a strong fluorescence signal (Figure 4.4.3A-C). The background fluorescence in the cytosol was measured to be less than 10% under these conditions (Figure 4.4.3C). Two control experiments indicate that the observed fluorescence labeling of the nucleus is be-

0 0

Post a comment