3-1 Design and expression of mutant ubiquitin constructs
To make thio-reactive fluorescence dye Alexa-488 conjugated ubiquitin, we need ubiquitin mutants with cysteine residue for dye labeling. We already had wild-type ubiquitin in pET-11b expression vector in our lab, and ubiquitin D39C, D39C/K63R M[C]Q-6His, and R72C-6His were constructed later based on the wild-type plasmid.
D39C and D39C/K63R mutant could be used directly, while M[C]Q-6His and R72C-6His could only be used after hexahistidine tag deletion. For this purpose, we mutated the first histidine codon (CAT) to stop codon (TGA) (figure 3.1). The resulting M[C]Q and R72C constructs were used to express cysteine-containing ubiquitins.
Furthermore, for better understandings of ubiquitination, we mutated lysine residues on ubiquitin M[C]Q to arginines. Seven mutants, M[C]Q/K6R, M[C]Q/K11R, M[C]Q/K27R, M[C]Q/K29R, M[C]Q/K33R, M[C]Q/K48R, and M[C]Q/K63R were therefore constructed and used in K-to-R ubiquitins expression later (figure 3.2). During making some constructs, the first DH5α colony picked on the plate didn’t contain the desired sequences in plasmids, and we had to pick and analyze several colonies to get the right ones.
28
Table 3.1 PCR protocol of site-directed mutagenesis
(A) The reagents added in PCR reactions. The polymerase used is PfuUltra II Fusion HS DNA Polymerase (Stratagene, Agilent, USA).
(B) PCR cycle condition. Determination of annealing temperature T* is according to the melting temperature (Tm) of the primer used.
Figure 3.1 PCR result of ubiquitin M[C]Q and R72C 6-His tag deletion
The protocol of PCR reaction is listed in Table 3.1 and the annealing temperature (T*) was set at 55°C. The PCR products were checked directly by DNA electrophoresis under 100 volts for 30 minutes. The gel was composed of 1% agarose in 1x TAE buffer with 1x SYBR® Safe DNA Gel Stain (Invitrogen, Life Technologies, USA). The result shows that the sizes of products were correct (about 6k bp). After plasmid 5’-phosphorylation and ligation, the samples were transformed to DH5α competent cells.
The sequences of plasmids were confirmed to be the expected ones by sequencing as chapter 2-2.1 described.
(A) (B)
29
Figure 3.2 PCR result of ubiquitin M[C]Q/K to R mutants.
The protocol of PCR reaction is listed in Table 3.1
(A) PCR result of ubiquitin M[C]Q/K6R. The annealing temperature was set at 52°C.
(B) PCR result of ubiquitin M[C]Q/K11R, K27R, K29R, K33R and K48R, respectively.
The annealing temperatures of all these 5 reactions were all set at 51°C.
(C) PCR result of ubiquitin M[C]Q/K63R. The annealing temperature was set at 55°C.
The DNA electrophoresis was done in gel composed of 1% agarose in 1x TAE buffer with 1x SYBR® Safe DNA Gel Stain (Invitrogen, Life Technologies, USA) under 100 volts for 30 minutes. The sizes of PCR products were all correct (about 6k bp). After plasmid 5’-phosphorylation and ligation, the samples were transformed to DH5α competent cells and the plasmids were amplified and extracted afterward. The sequences of plasmids were confirmed to be the expected ones by sequencing as chapter 2-2.1 described.
(A) (B) (C)
30
3-2 Small scale expression analysis
After transforming plasmid into E. coli BL21 StarTM (DE3) competent cells, we cultured the cells overnight on agar plates. The colonies on the plates were inoculated into LB/ampicillin medium. After overnight culture and the following subculture, IPTG was added for induction test. As shown in figure 3.3, the expression levels of ubiquitin mutants were all highly elevated. Those high-expression strains were prepared as glycerol stocks for later expression. 2 tubes of 1.2 mL glycerol stock were made for every mutant.
Figure 3.3 Small scale expression of ubiquitin M[C]Q and R72C (without 6-His tag).
Two colonies with ubiquitin M[C]Q plasmid (M[C]Q1 and M[C]Q2) and two colonies with ubiquitin R72C plasmid (R72C1 and R72C2) were picked and subcultured for induction test. The full procedures are described in chapter 2-2.2. The result shows that all cultures with IPTG induction overexpressed ubiquitin (arrow on the bottom). (“Ub”
stands for commercial wild-type ubiquitin as a positive control.)
31
3-3 Large scale expression and purification
3-3.1 Large scale protein expression
100 μL of ubiquitin M[C]Q, D39C, R72C and D39C/K63R glycerol stock was
added to 100 mL TB medium respectively and incubated overnight. Then the cultures were subcultured and ubiquitin expression was induced by IPTG. 4 hours later, the cultures were centrifuged and about 20 g of the cell pellet was collected for each culture.
Later the pellets underwent cell lysis, and about 200 mL of cell lysate was produced.
Then ammonium precipitation was performed. In every step, 20 uL of sample was collect and analyzed by electrophoresis (figure 3.4).
Figure 3.4 Examples of ubiquitin purification.
The purification method is listed in chapter 2-2.3.2 and 2-2.3.3. Ubiquitin D39C/K63R (D39 is replaced by cysteine, and K63 is replaced by arginine) construct was used to optimize purification procedures.
(A) Cell lysis and centrifugation after ubiquitin overexpression. As shown in the pictures, expressed protein appeared in supernatant, not pellet.
(B) Ammonium sulfate precipitation of collected supernatant. Most ubiquitin was precipitated within 40 - 80% saturation. (“SP” = supernatant; “Ub” = WT ubiquitin)
(A) (B)
32
3-3.2 Ion exchange chromatography & size exclusion chromatography
The supernatant was condensed to less than 50 mL and then filtered by 0.22 μm
filter for IEC. Ubiquitin was eluted at 30% to 40% B buffer, as it shown in figure 3.5.
The fractions which contained ubiquitin were pooled and concentrated for SEC. As shown in figure 3.6, ubiquitin appeared between fraction 23 and 32 (from 82 mL to 102 mL). However, since ubiquitin was not finely separated from the contaminants in this step, the collected fractions were concentrated and underwent a second round of SEC (figure 3.7). Figure 3.5 to 3.7 show the results of ubiquitin M[C]Q purification, and the
purification of ubiquitin D39C and R72C had similar results. The eluate with purified ubiquitin was dialyzed in 0.1% TFA and then lyophilized. Finally, about 5 μg of
ubiquitin dry powder was acquired.
33
Figure 3.5 IEC purification of Ub M[C]Q.
Ub M[C]Q mostly appeared in fraction 14 to 18, and these fractions were collected for SEC. (“Ub” represents “wild-type ubiquitin”, which was used as positive control.)
(A)
(B)
34
(B)
Figure 3.6 SEC purification of Ub M[C]Q
Ub M[C]Q appeared in fraction 22 to 28, and these fractions were pooled for second SEC. (“Ub” = WT Ub)
(A)
35
Figure 3.7 Second SEC of Ub M[C]Q
A single peak appeared in FPLC plot, and fraction 23 to 34 were collected, dialysis in 0.1% TFA, and then lyophilized.
36
3-4 Fluorescence dye labeling and purification
Ubiquitin M[C]Q, D39C, R72C and D39C/K63R were used to make Alexa-488 labeled ubiquitin, which called M[C]Q-A488, D39C-A488, R72C-A488 and D39C/K63R-A488. After reaction and dialysis, HPLC was applied to separate labeled ubiquitin from free ubiquitin. All samples eluted at 30% to 32% B buffer (figure 3.8 B- E). It should be noted that two peaks in M[C]Q-A488 sample with 488 nm absorption appeared, therefore we collected both peaks for further analysis. Purified and lyophilized samples were then sent to do ESI-TOF mass spectrometry analysis (figure 3.9 and 3.10). We confirmed that R72C-A488 was the correct product (figure 3.10), but it seemed that the Met1 of both M[C]Q-A488 samples were lost (figure 3.9), which agreed with previous result in our lab and researches from other groups [37, 38].
37
(A)
(B)
Sample volume = 220 μL
Sample volume = 50 μL
38
Figure 3.8 HPLC purification of Alexa-488 labeled ubiquitin
(A) HPLC purification of Ub M[C]Q-A488. Two peaks were eluted in M[C]Q-A488 sample, and both of them were collected.
(B) HPLC purification of Ub D39C-A488. A single peak was shown.
(C) HPLC purification of Ub R72C-A488. The right peak is dye-labeled R72C-A488, and the left peak is R72C free protein.
(D) HPLC purification of Ub D39C/K63R-A488. A single peak was shown.
Free Alexa-488 eluted at 11 - 13 minutes, at similar position as Ub-A488 samples.
(C)
(D)
Sample volume = 100 μL
Sample volume = 250 μL
39
Figure 3.9 ESI-TOF mass spectra of Ub M[C]Q and Ub M[C]Q-A488 (A) Theoretic mass of Met1 excised Ub M[C]Q is shown in the spectra.
(B) Theoretic mass of Met1 excised Ub M[C]Q-A488 (black arrow) is shown in the spectra of Ub M[C]Q sample peak 1.
(C) Theoretic mass of Met1 excised Ub M[C]Q-A488 (black arrow) is also shown in the spectra of Ub M[C]Q sample peak 2.
(A)
(B)
(C)
40
Figure 3.10 ESI-TOF mass spectra of Ub R72C and Ub R72C-A488 (A) No theoretic mass of Ub R72C (8511.8 Da) was obtained in this figure.
(B) However, two peaks with molecular weight of 8510.5 Da and 9209.0 Da (black arrows), which are related to Ub R72C (predicted mass = 8511.8 Da) and R72C-A488 (predicted mass = 9209.5 Da), appeared here.
(A)
(B)
41
3-5 Enzyme expression and purification
After ubiquitin-A488 samples are ready, Ubc5 (E2) and Rsp5 (E3) enzymes are expressed. Prepared glycerol stocks were used for large scale expression, and IMAC was performed to purify His-tagged enzymes. Our enzymes appeared in eluate (1st IMAC in figure 3.11A & B). In following SUMO-tag cleavage, Ubc5 and Rsp5 eluted at low imidazole concentration (2nd IMAC in figure 3.11 A & B).
Figure 3.11 Ubc5 (E2) purification and Rsp5 (E3) purification
The method of IMAC purification is described in chapter 2-2.5.2, and SUMO cleavage method is described in 2-2.5.3.
(A) Ubc5 purification. After cell lysis and centrifugation, Ubc5 appeared in supernatant.
In the first IMAC, Ubc5 was eluted when elution buffer with high imidazole was applied. After SUMO cleavage, the sample underwent the second IMAC. Ubc5 protein didn’t bind to IMAC column and appeared in flow through (imidazole concentration = 10 mM).
(B) Rsp5 purification. After cell lysis and centrifugation, Rsp5 appeared in supernatant.
In the first IMAC, Rsp5 was eluted when elution buffer with high imidazole was applied. After SUMO cleavage, Rsp5 protein bind loosely to IMAC column and was eluted when imidazole concentration > 20 mM
(A) (B)
42
3-6 Ubiquitination assay and its results
We performed in vitro ubiquitination assay after dye-labeled ubiquitin mutants and E2, E3 enzymes were ready to use. Every component was checked by silver staining prior to ubiquitination assay to confirm its position on gel (figure 3.12). We designed a series of ubiquitination assay, which specific reagents were added or not (table 3.2).
With this design, we showed that all reagents we added were required in ubiquitination (figure 3.13-3.16).
The fluorescent imaging, silver staining and Western blotting result of different dye-labeled ubiquitin (M[C]Q-A488, D39C-A488 and R72C-A488) are shown below.
We observed that ubiquitin M[C]Q-A488 and D39C-A488 conjugated to Pol II (figure 3.13-3.16), while R72C-A488 didn’t conjugate to substrate in the same condition (figure 3.17). To our surprise, no polyubiquitin signal was detected on fluorescent imaging, and it maybe indicates that dye-labeled ubiquitin is unable to form polyubiquitin chain. On the other hand, besides expected singnals, Western blot result revealed that non-specific bindings almost appeared in every plot (figure 3.13B, 3.15B and 3.17B).
43
Table 3.2 Experimental design of ubiquitination assay
A plus sign (+) means the reagent was added in this reaction, and a minus sign (-) means the reagent was absent in this reaction. An “F” means fluorescent-dye-labeled ubiquitin was added, while a “W” means wild-type ubiquitin was used instead. We expected that only reaction 1 and 7 would show Pol II ubiquitination.
Figure 3.12 Silver stain result of each reagent in ubiquitination
Pol II = yeast RNA polymerase II, E1 = human UBE1 (Sigma-Aldrich, USA), E2 = yeast Ubc5, E3 = yeast Rsp5, WT Ub = human wild-type ubiquitin (Sigma-Aldrich, USA). Ub M[C]Q, D39C and R72C were expressed and purified previously.
Two major bands in Pol II sample were Pol II subunit Rpb1 (upper) and Rpb2 (lower).
44
(A) (B)
Figure 3.13 Ubiquitination assay using Ub M[C]Q-A488 (E2, E3 with SUMO tag) Which reagents were added is listed above. “+” means this reagent was added, “-”
means this reagent was not added. At the row of Ub (ubiquitin), an “F” means fluorescent-dye-labeled Ub M[C]Q-A488, while a “W” means wild-type ubiquitin.
(A) Silver stain: in this figure, band shift after ubiquitin conjugation (especially E2-Ub) could be observed. The unmodified and Ub-conjugated E2 are marked by black arrow and red arrow, respectively.
(B) Western blot: polyubiquitination signal could be seen in ubiquitination assay using wild-type Ub (red arrow). Non-specific binding band is marked by black arrow.
(C) Fluorescent image: in this figure, we can distinguish ubiquitin monomer (marked by black arrwo), Ub conjugated E1 (E1-Ub, white arrow), E2-Ub (yellow arrow) and Rpb1-Ub (red arrow)
(C)
45
Figure 3.14 Ubiquitination assay using Ub M[C]Q-A488 (E2, E3 without SUMO tag)
Which reagents were added is listed above. The meanings of “+” and “-” are the same as figure 3.12. At the last row, an “F” means fluorescent-dye-labeled ubiquitin
M[C]Q-A488, and a “W” means wild-type ubiquitin.
(A) Silver stain: in this figure, band shift after ubiquitin conjugation could be observed.
The unmodified and Ub-conjugated E2 are marked by black arrow and red arrow, respectively. Ub-conjugated E3 could also be seen in this figure (blue arrow).
(B) Western blot (C) Fluorescent image: in this figure, we can distinguish ubiquitin monomer (marked by black arrwo), Ub conjugated E1 (E1-Ub, white arrow), E2-Ub (yellow arrow), E3-Ub (orange arrow) and Rpb1-Ub (red arrow)
(A) (B)
(C)
non-specific
46
(A) (B)
(C)
Figure 3.15 Ubiquitination assay using Ub D39C-A488 (E2, E3 with SUMO tag) Which reagents were added is listed above. The meanings of “+” and “-” are the same as figure 3.12. At the last row, an “F” means fluorescent-dye-labeled ubiquitin
D39C-A488, and a “W” means wild-type ubiquitin.
(A) Silver stain: in this figure, band shift after ubiquitin conjugation could be observed.
The unmodified and Ub-conjugated E2 are marked by black arrow and red arrow, respectively.
(B) Western blot: polyubiquitination signal could be seen in ubiquitination assay using wild-type Ub (red arrow). Non-specific binding band is marked by black arrow.
(C) Fluorescent image: in this figure, we can distinguish ubiquitin monomer (marked by black arrwo), Ub conjugated E1 (E1-Ub, white arrow), E2-Ub (yellow arrow) and Rpb1-Ub (red arrow)
47
Figure 3.16 Ubiquitination assay using Ub D39C-A488 (E2, E3 without SUMO tag) Which reagents were added is listed above. The meanings of “+” and “-” are the same as figure 3.12. At the last row, an “F” means fluorescent-dye-labeled ubiquitin
D39C-A488, and a “W” means wild-type ubiquitin.
(A) Silver stain: in this figure, band shift after ubiquitin conjugation (especially E2-Ub) could be observed. The unmodified and Ub-conjugated E2 are marked by black arrow and red arrow, respectively.
(B) Fluorescent image: in this figure, we can distinguish ubiquitin monomer (marked by black arrwo), Ub conjugated E1 (E1-Ub, white arrow), E2-Ub (yellow arrow), E3-Ub (orange arrow) and Rpb1-Ub (red arrow)
(A) (B)
48
(A) (B)
(C)
Figure 3.17 Ubiquitination assay using Ub R72C-A488 (E2, E3 with SUMO tag) Which reagents were added is listed above. The meanings of “+” and “-” are the same as figure 3.12. At the last row, an “F” means fluorescent-dye-labeled ubiquitin
R72C-A488, and a “W” means wild-type ubiquitin.
(A) Silver stain: in this figure, band shift after ubiquitin conjugation could be observed.
The unmodified and Ub-conjugated E2 are marked by black arrow and red arrow, respectively.
(B) Western blot: polyubiquitination signal could be seen in ubiquitination assay using wild-type Ub (red arrow). Non-specific binding band is marked by black arrow.
(C) Fluorescent image: only E1 (white arrow) and small amount of E2 (yellow arrow) are conjugated by R72C-A488. No fluorescent signal of Pol II could be seen on gel.
49