Comparison Of The Protein Expression Profiles With Program Damapep

After calculating signal (SigS(ij)) and background (BKS(ij)) values for all spots, the program DAMAPEP identifies those proteins differentially expressed between normal and cancer cell lines. For this purpose, three different criteria were utilized to identify those candidate proteins from the DAMA staining images (Fig. 2b).

First a decision was made according to the average intensity change of the protein at spot(ij) between all cancer cells and all normal cells. SigCan(ij), defined as the average SigS(ij) for all seven cancer cells (T47D, MCF7, Zr-75-1, MDA-MB-231, BT549, HS578T, and MDA-MB-435s), was calculated for every spot(ij) (Fig. 2, b and cEquation 3). This value represents the average intensity change of the protein at spot(ij) in all cancer cell lines relative to the reference cell line MCF10A. Similarly SigNor(ij), the average SigS(ij) for two normal cells (MCF12A and HS578Bst), was also calculated for every spot(ij) (Fig. 2cEquation 4). SigNor(ij) corresponds to the average intensity change of the protein at spot(i,j) in other normal cell lines relative to the intensity in the reference cell MCF10A. The Ratio(ij) calculated (Fig. 2cEquation 5) thus represents the average intensity change of the protein at spot(ij) between cancer cells and normal cells relative to that of the normal cells (Fig. 2b). The higher the absolute value of Ratio(ij), the larger the intensity difference. Therefore, DAMAPEP can predict a list of proteins with different expression levels between normal and cancer cells based on the selected cutoff value of Ratio(ij).

When 2.0 was used as a cutoff value of Ratio(ij), 54 proteins were predicted to be differentially expressed proteins between normal and cancer cells. However, 22 of the 54 proteins had lower spot intensity in at least one of the 10 cell lines. The contribution of those weak intensity spots to the corresponding value of Ratio(ij) could be enlarged. To test the possibility, the expression levels of eight proteins, randomly selected from the 22 proteins, were compared among 10 cell lines by Western blotting analysis. All eight proteins showed low or undetectable expression levels (data not shown). Therefore, to decrease false positives caused by lowered spot intensity, DAMAPEP automatically compares the intensity of each spot with their background from the ScanAlyze exported files (CH1I versus CH1B and CH1AB; CH2I versus CH2B and CH2AB). The spots with intensities lower than their corresponding backgrounds in any one of the cell lines were excluded from the final prediction list (Fig. 2b).

The third criteria, Sigcan(ij) (the averaged SigS(ij) values for all seven cancer cell lines), could easily be biased by large numbers in one or two cell lines. To eliminate this bias, the SigS(ij) value was compared with the corresponding background BKS(ij) for all seven cancer cell lines. Only those proteins at spot(ij) having more than four cell lines with higher SigS(ij) value than the corresponding background value BKS(ij) were included in the final prediction list.

Cancer Cells By Dissociable Antibody Microarray Dama Staining

Dissociable antibody microarray (DAMA) staining is a technology that combines protein microarrays with traditional immunostaining techniques. It can simultaneously determine the expression and subcellular location of hundreds of proteins in cultured cells and tissue samples. We developed this technology and demonstrated its application in identifying potential biomarkers for breast cancer. We compared the expression profiles of 312 proteins among three normal breast cell lines and seven breast cancer cell lines and identified 10 differentially expressed proteins by the data analysis program DAMAPEP (DAMA protein expression profiling). Among those proteins, RAIDD, Rb p107, Rb p130, SRF, and Tyk2 were confirmed by Western blot and statistical analysis to have higher expression levels in breast cancer cells than in normal breast cells. These proteins could be potential biomarkers for the diagnosis of breast cancer.

Protein microarrays have recently been attracting great attention for their potential use in high throughput studies of protein function (16). The ultimate goal of developing this technology is to construct ordered arrays of individual proteins for biochemical study at the molecular level. A number of different sources of peptides and proteins have been used for protein microarray manufacture, including synthetic peptides (78), recombinant proteins (911), and monoclonal and polyclonal antibodies (1214). Microarrays have been utilized to study protein expression profiles (1517), protein-protein interactions (1819), drug analysis (1120), and the diagnosis of diseases such as cancer, food allergies, and infection by pathogenic viruses and bacteria (2124). Most of the microarrays use the capture microarray platform (35). In a standard capture microarray procedure, an array of proteins is immobilized on a membrane or a glass slide to capture protein ligands from a protein mixture; the captured ligands are detected either with a different set of labeled antibodies (1517) or a detectable tag attached to the ligands (25).

We have developed a different protein microarray platform, dissociable antibodymicroarray (DAMA)1 staining (26). This technology combines the power of immunohistochemical staining and the parallel analysis of antibody microarrays. DAMA staining provides a new approach in the global analysis of protein expression and subcellular localization. In DAMA staining, targeted cells are grown on a coverslip in a culture dish or mounted on a coverslip and fixed and permeabilized with the standard protocol for traditional immunostaining. In the next step, instead of adding a primary antibody, an array of antibodies immobilized on a membrane is placed on top of the specimen. Pressure is applied to maintain close contact between the antibody array and cells. During incubation, antibodies dissociate from the array support and bind to their respective antigens in the cells without significant lateral diffusion. In this way, hundreds of antibodies are delivered to the targeted cells or tissues in a position-dependent manner. The bound antibodies are then detected either with enzyme- or fluorophore-conjugated secondary antibodies. The expression profiles of hundreds of proteins can thus be determined simultaneously. Furthermore when stained with fluorophore-conjugated secondary antibodies, subcellular localization profiles of hundreds of proteins can be obtained from a single staining by a fluorescence microscope equipped with a computer-controlled motorized stage.

Here we report the development of DAMA staining technology and its application in identifying potential biomarkers for breast cancer. Breast cancer is the most common form of cancer in women with ∼210,000 new cases diagnosed annually in the United States alone (27). This disease causes significant mortality, accounting for ∼40,000 deaths in the United States per year (28) and many more fatalities worldwide. Earlier detection and better treatment will improve prognosis and survival of the disease (27). Detection of new molecular biomarkers will not only be very useful for breast cancer diagnosis but also may identify novel and key targets for therapy. In this report, we compared the expression profiles of 312 proteins among 10 different normal breast and breast cancer cell lines using DAMA staining. We also developed a data analysis program to identify differentially expressed proteins and validated the prediction by Western analysis in the same cell lines.

EXPERIMENTAL PROCEDURES

Preparation of Antibody Microarray Array-320—

Array-320 contains 312 antibodies in a 16 × 20 format. The antibodies were spotted on a membrane by a robotic arrayer from Gesim (Dresden, Germany) using piezoelectrically driven microdosage heads. This array has an overall size of 44 (width) × 42 mm (height) with 2 mm between different antibody spots. Each spot contains 50 ng of antibody and is ∼500 μm in diameter. The antibody list for Array-320 is shown in the supplemental materials. Antibodies were selected from Hypromatrix's collection. All antibodies have been characterized and demonstrated to bind their targets in various assays.

Cell Culture—

Ten different breast cell lines, MCF10A, MCF12A, Hs578Bst, MCF7, T-47D, ZR-75-1, MDA-MB-231, BT549, Hs578T, and MDA-MB-435S, were purchased from American Type Culture Collection (ATCC). Cells were maintained and propagated as recommended by the ATCC and were grown on 10-cm cell culture dishes until 90–95% confluence for DAMA staining.

The DAMA Staining Procedure—

Cells were fixed with formaldehyde solution, permeabilized with Triton X-100, and blocked with goat serum. A 320-antibody microarray (Array-320) was placed over the cells and was incubated at room temperature under the optimal pressure of ∼100g/cm2 as measured by a pressure detection sensor in an Economical Load and Force (ELF) single handle system (Tekscan, Inc.). Bound antibodies were detected by alkaline phosphatase-conjugated secondary antibodies (both goat anti-rabbit and goat anti-mouse) with 1-step™ NBT/BCIP (nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate) substrate (Pierce). Images of protein expression profiles were scanned by an HP Scanjet 4890 and processed by Photoshop for intensity integration.

Intensity Integration and Initial Data Analysis—

The ScanAlyze program developed for DNA microarray analysis was used for intensity integration of the DAMA staining images. One image (e.g. the reference cell line, such as MCF10A) was used as the channel 1 data, and the other image (e.g. other sample cell lines, such as MCF7) was used as the channel 2 data. The scanned DAMA staining images were first compressed into eight-bit grayscale images with a resolution of 600 d.p.i. and color-inverted by Photoshop. The preprocessed images were then combined into an RGB overlay image with one in red channel 1 and another in green channel 2. A grid was created with a circular mask defining the boundary for each spot. For the purpose of intensity integration, the location of the mask was refined for every spot. MRAT(ij) values, the median ratio of intensity between channel 1 and channel 2 within the mask, were determined and exported to an Excel file. The subsequent analysis was done with a customized program (DAMAPEP). Briefly MRAT(ij) was converted to log2(MRAT(ij)), normalized for all 312 proteins by subtracting the median log2(MRAT(ij)) value, and scaled by dividing the root mean square of all normalized log2(MRAT(ij)) values. The median intensity of every spot and their corresponding median and mean background values, CH1I, CH1B, and CH1AB for the reference cell line R and CH2I, CH2B, and CH2AB for the sample cell line S, were also exported for the analysis by the DAMAPEP program.

Western Blotting Analysis—

Ten breast cell lines were grown on 10-cm cell culture dishes in corresponding medium to around 90–95% confluence. The whole cell lysates were extracted, concentrated, then quantified with the Bradford assay, and adjusted to the same values. Equal amounts of total protein were loaded in each lane (29).

Gel Quantification and Statistical Analysis—

For every protein, relative band densities of 10 cell lines were scanned and digitized by UN-SCAN-IT software (Silk Scientific, Orem, UT). These values were normalized against the highest value and averaged over all experiments. Relative expression levels of five proteins between normal and cancer cells were analyzed by using the Student's t test. For every protein, expression levels in three normal cells and in seven cancer cells were analyzed by a two-sample t test assuming equal variances. The degree of freedom is 9 (n1 = 3, n2 = 7). The one-tailed pvalues were used for statistical inference.

RESULTS

Determining Protein Expression Profiles in Different Breast Cell Lines by DAMA Staining—

DAMA staining is a high throughput technology that can simultaneously determine the expression profiles and subcellular locations of hundreds of proteins (26). The goal of this work was to develop the DAMA staining technology for protein expression profiling and to demonstrate its application in identifying potential biomarkers for breast cancer. The technology includes the following steps: determination of protein expression profiles by DAMA staining, data extraction by the ScanAlyze program, data analysis and prediction by the DAMAPEP program, and evaluation by Western blot analysis.

For this purpose, protein expression profiles of 10 different cell lines from human mammary glands were obtained from DAMA staining (Fig. 1). The 10 breast cell lines include three normal cell lines (MCF12A, MCF10A, and Hs578Bst), three estrogen receptor-positive carcinomas (T-47D, MCF7, and ZR-75-1), and four estrogen receptor-negative carcinomas (MDA-MB-231, BT549, Hs578T, and MDA-MB-435S). The cells were grown on a 10-cm cell culture dish to 90–95% confluence and fixed and permeabilized by a standard protocol. Array-320 was then used to deliver an array of 312 primary antibodies to those fixed cells. Bound antibodies were detected by using alkaline phosphatase-conjugated secondary antibodies. The resulting array of gray dots was imaged and subjected to intensity integration and data analysis to generate expression profiles for the 312 proteins. The experiments were repeated at least twice for each cell line. The representative images and the Pearson's correlation coefficients (Rr) of the replicated images for all 10 breast cell lines are summarized in Fig. 1. Most cell lines had reproducible DAMA staining images with the Pearson's correlation coefficients higher than 0.5.

FIG. 1.

Summary of DAMA staining images in 10 different breast cell lines. The images were obtained by using the 312-antibody microarray, Array-320. The distance between the spots is 2 mm, and the dimensions of the images within four corner spots is 38 mm in width and 30 mm in height. The cell line names are labeled either aboveor below the corresponding images. The spot positions are labeled from A to P for rows and from 1 to 20 for columns. Experiments were repeated at least twice, and one set of results is shown here. The Pearson's correlation coefficients (Rr) of two replicated images for all 10 cell lines are included.

Intensity Integration of the DAMA Staining Images by ScanAlyze—

The images obtained from DAMA staining were analyzed using standard methods for intensity integration, normalization, and scaling. Among the 10 breast cell lines, one normal cell line, MCF10A, was used as a common reference for intensity integration. The other nine cell lines were used as sample cell lines. The DAMA staining images of the nine cell lines were individually compared with the image of MCF10A by using ScanAlyze, an intensity integration program for DNA microarrays. Protein expression profiles between different samples were quantitatively compared by ScanAlyze with one image (e.g. the reference cell line R) as the channel 1 data and the other image (e.g. the sample cell line S) as the channel 2 data. MRAT(ij), the median intensity ratio between channel 1 and channel 2 at the spot of the ith row and jth column (spot(ij)), was exported. The logarithm of the MRATs of each pair was normalized and scaled by using a procedure similar to that used for data analysis of DNA microarrays.

Initial Data Analysis for the DAMA Staining Images with Program DAMAPEP—

A program, DAMAPEP (DAMA protein expression profiling), was developed to retrieve, normalize, and scale the data from the exported ScanAlyze files. The method for this data analysis is shown in Fig. 2. As the protein expression profiles for each cell line were determined at least twice, there are two independent DAMA staining images for each spot: two for the sample cell line (S1 and S2) and two for the reference cell line (R1 and R2). Therefore, four different sets of MRAT values, MRAT(1, 1), MRAT(1, 2), MRAT(2, 1) and MRAT(2, 2), corresponding to the intensity ratios of S1 to R1, S1 to R2, S2 to R1, and S2 to R2, respectively, were obtained for every spot(ij) (Fig. 2a). The logarithms of those MRAT values were normalized and scaled. The average of those scaled log2(MRAT(ij)) values, defined as SigS(ij), represents the intensity change between the sample cell line (S) and the reference cell line (R) for the protein at spot(ij) (Fig. 2cEquation 1).

Fig. 2.

View larger version:

Desalting Of Dagk Prep Of Samples For Faruv Cd Of Dagk Protocol

The key to far-UV CD is to have a clear solution which also does not have anything in it (besides protein) which absorbs in the 180-250nm range. Even if any such absorptive substance produces no CD signal, it can strongly absorb both components of the polarized light making acquisition of a CD spectrum impossible (the equivalent of trying to take standard UV absorbance measurements on a solution with A > 3.0). For DAGK, one would have to worry especially about imidazole. For far UV CD, Olga made samples in the range of 0.2 mg/ml. Special quartz cuvettes are used having 1 mm (instead of the usual 1 cm) path lengths. Olga‘s Biophysical Journal paper has additional info on how she made measurements and also has representative spectra. One way to produce DAGK samples for near-UV CD would be to purify the enzyme using something other than imidazole as the eluting agent. This can be done. However, for our studies we may more often be interested in taking a CD spectrum of a sample eluted the normal way and subjected to other experiments (activity measurement, cross-linking, disulfide mapping, etc.)… so that we have a CD spectrum of EXACTLY the same batch of DAGK as used in these other experiments. For this, a better strategy would be to de-salt the DAGK.

Desalting Buffer: 1% decyl maltoside

100 mM NaCl

20 mM phosphate

pH 7.0

1.Prepare 3 ml of Biogel P-4 resin by incubating with DM-containing buffer for 4 hours at room temperature.

2.Pour off the excess solution (with any non-sedimented suspension).

3.Load it on to a 0.7 cm diameter column. Make sure the column has been well-equilibrated in buffer before applying the protein (flow at least 3 column volumes of DM buffer through it if it has not already been equilibrated).

4.Apply 0.5 mg of DAGK to the column in as small a volume as possible (1 ml or less). The starting DAGK should be concentrated enough so that even when it is diluted by the desalting column it will be concentrated enough upon elution to use directly in CD.

5.Once the sample flowed into the column, chase. the sample in with about 0.25 ml of DM/buffer.

6.Apply a small (0.1 ml) volume of 3 mM tyrosine in buffer. If you can cleanly separate the tyrosine from the DAGK (elution judged by the UV-detector/chart) then you can be sure DAGK is completely separated from imidazole or other small molecules.

7.Collect only the first . of the DAGK peak (ideally the top . of the total peak, not saving the wings.).

8.Once all of the tyrosine is eluted from the desalting column, let another 3 bed volumes of buffer continue to flow through (to make sure all of the imidazole is completely eluted).

9.Now the column is ready to use again: you donet have to pour a new column. Only use a column 2 times before making a new one! We have observed that resolution really degrades with repetitive use of the same column (this was true whether BioGel or Sephadex resins were used).

Glutaraldehyde Crosslinking Of Dagk Mutants Protocol

Purpose: Normal wild type DAGK will get covalently linked by glutaraldehyde to form primarily trimeric DAGK which can be observed in SDS-PAGE. This reflects the fact that DAGK functions as a homotrimer. Some DAGK mutants may not fold correctly and assemble into homotrimers. In this case, covalent homotrimers would not be expected to form in the presence glutaraldehyde. Thus, we use glutaraldehyde cross-linking followed by SDS-PAGE as one way of testing for correct folding of DAGK.

Glutaraldehyde cross-linking buffer: 2% decyl maltoside

20 mM phosphate

50 mM NaCl

1 mM EDTA

1 mM DTT (add fresh on day of use, optional)

pH 7.5

1. When cross-linking a mutant, always run a parallel reaction (identical conditions)

on WT, C46, or Cysless as a standard.

2. Add DAGK to cross-linking buffer to 25 ¥ìM (roughly 0.35 mg/mL), for a final volume of 200 ¥ìL.

3. Add glutaraldehyde to 16 mM from 25% aqueous stock solution (Sigma, store at -20¢ªC). {25% solution is 2.5 M, add 3.2 microliters per 500 microliters of solution).

4. Incubate for fixed period of time (2-24 hours, but always do the same amount of time.little additional reaction occurs after 2 hours) at room temp with shaking, ~300 RPM.

5. Run cross-linked samples on SDS-PAGE. Remember to use C46, Cysless or WT as control. 

Western Blot Detection Of Histagged Proteins Protocol

Use flat-bladed tweezers to handle the membrane. A small yellow tip box works well for this procedure. Gently agitate on platform shaker during each wash/incubation step. Leave for 10 minutes during the wash steps. Use about 30 mL of solution for each step.

1. Run SDS-PAGE and transfer proteins from gel to nitrocellulose membrane according to the instructions that came with the gel apparatus.

2. Wash membrane twice in TBS.

3. Incubate for 1 hr in blocking buffer.

If it is desired to stop for the day, leave blot in blocking buffer overnight at 4 °C.

4. Wash membrane twice in TBSTT.

5. Wash membrane in TBS.

6. Incubate membrane in blocking buffer with 1/2000 dilution of anti-pentaHis antibody (add 15 ìL stock solution to 30 mL blocking buffer), for 1-2 hours.

7. Wash membrane twice in TBSTT.

8. Wash membrane in TBS.

9. Incubate membrane in blocking buffer containing 1/10000 dilution of anti-mouse/AP conjugate (add 3 ìL antibody stock solution to 30 mL blocking buffer) for 1-2 hours.

10. Wash 4 times with TBSTT.

11. Stain with AP staining solution until the signal is visible. Do not shake blots during color development.

12. Stop the chromogenic reaction by rinsing twice with water.

13. Photograph/scan the membrane as soon as possible. It will be OK if covered in water and kept away from light. If dried and exposed to light, the bands will fade quickly. 

1X NuPAGE Transfer Buffer (1L):

Transfer Buffer (20X)

50 mL

Methanol

100 mL

Deionized H2O

850 mL

To prepare 20X Transfer Buffer, dissolve the following in 100 mL deionized H2O:

Bicine

10.2 g

Bis-Tris

13.1 g

EDTA

0.75 g

Mix well and adjust the volume to 125 mL. Adjust pH to 7.2, if needed.

Western Blot Solutions (all amounts given per L):

Tris Buffered Saline (TBS)

(20 mM Tris, 140 mM NaCl, pH 7.5)

2.42 g Tris base

8.18 g NaCl 

TBS + Tween/Triton (TBSTT)

(0.1 % Tween 20, 0.2 % Triton X-100 in TBS)

for 1 L:

1 mL Tween 20

2 mL Triton X-100 

Blocking Buffer (make fresh – you will need at least 90 mL for one blot)

3% Bovine Serum Albumin (fraction V, Sigma A-7906) in TBS 

Staining solutions (for step 11.)

The complete AP Staining solution is Buffer A with 0.33 mg/mL NBT and 0.166 mg/mL BCIP.

To prepare the final staining solution,

add 1 mL of NBT stock and 100 μL 5% BCIP stock to 30 mL Buffer A.

Individual components:

1. Buffer A (AP Staining Solution):

100 mM Tris-Cl (12.1 g/L)

100 mM NaCl (5.84 g/L)

5 mM MgCl2 (0.48 g/L)

Start with Tris base and pH to 9.5 with HCl.

2. NBT stock is 1% NBT in water

10 mg NBT (1 tablet)

1 mL of dH2O (make fresh every time)

3. BCIP stock is 5% BCIP in 100% DMF

25 mg BCIP (1 tablet)

500 μL DMF

can be stored in aliquots at -20 °C 

Bowie Procedure For Purifying Dagk Protocol

1. Lyse cells as described for the usual Sanders procedure.
2. To extract, add -octyl glucoside to 5% (5g/100 mL).
3. Tumble 30 min. (only) at 4C (do NOT use a stir bar, instead gently rotate a sealed container).
The lysate cannot be frozen at this point!!!
4. Remove insoluble material by spinning 20 min., 4C, at 15000 RPM in 50 mL tube Beckman rotor.
5. Discard pellet.
6. Equilibrate nickel resin with 4 x 1 bed volumes Buffer A solution. Use 1 mL packed resin per gram wet cells.
7. Add resin to supernatant.
8. Tumble 30 min., 4C.
9. Spin 15 min. in a tabletop centrifuge.
10. Decant sup.
11. Disperse resin in BOG/A buffer (see below). Pour resin into column. (Alternately, resin can be frozen in a Falcon tube at this point.)
12. Wash resin with Bowie Wash until A280 goes down and stays at a minimum.
13. Rinse resin with 8 X 1 bed volumes of Bowie Rinse to re-equlibrate with decylmaltoside as the detergent.
14. Elute DAGK with Bowie Elute.
After making all detergent-containing buffers, carefully bubble Argon through the solutions for half an hour.
DTT should not be used with regenerated resin; use new resin.
-Buffer A:
40 mM HEPES
300 mM NaCl
10 uM BHT
0.5 mM DTT. Add to buffer A only on day of purification, and only if will not interfere with
further experiments.
pH = 7.5
-BOG/A: Buffer A + 1.5% BOG
-Bowie Wash:
Buffer A + 1.5% BOG + 0.04 M imidazole, pH 7.8
-Bowie Rinse: 1% decyl maltoside in 25 mM Na-Phosphate, pH 7.2: on day of use add DTT to 0.5 mM unless you are going to be doing experiments with DAGK (such as SH-modification) that require that no reducing agent be present)… only to volume you will actually use
-Bowie Elute: 1% decyl maltoside + 0.25 M imidazole, pH 7.8 (adjust the pH then add the decyl maltoside) (on day of use add DTT to 0.5 mM unless you are going to be doing experiments with DAGK—such as SH-modification or -S-S- formation—that require that no reducing agent be present… only to volume you will actually use): .
B-octyl glucoside and B-decyl maltoside (DM) are both detergents. BOG has a critical micellar concentration (CMC) near 25 mM, while for DM the CMC is 2 mM. DAGK is more stable in DM which is why it is used in the final steps of the purification and to store DAGK.
The DAGK containing pool can now be stored. If you now freeze the DAGK solution in liquid nitrogen you can store at -80 and later thaw with only an about 10% loss of activity. Alternately, you can freeze-dry (lyophilize) the solution, although lyophilizing DM/formic acid solutions of DAGK does not work well- the enzyme denatures. Also, you cannot lyophilize DAGK solutions containing high salt (e.g., 300 mM NaCl or 25 mM MgAcetate).

Regenerating Used Qiagen Ninta Resin Protocol

1. Pour used resin into filter funnel. May connect to vacuum line, but only if necessary.
25
2. Wash twice with 6M guanidine HCl + 2% formic acid. Stir well after EACH wash, and allow to drain completely between washes.
3. Wash twice with 100mM EDTA + 3% Empigen at neutral or basic pH.
4. Wash twice with water.
5. Incubate resin twice with batches of 200 mM NiSO4 (hazardous – must collect all NiSO4 as hazardous waste).
6. Wash 3-4 times with water.
7. Store in 20% ethanol in water, at 4 °C.

Ni2 Ninta Resin Incubation And Elution Protocol

1. Superflow. Nickel-Agarose resin should be equilibrated by rinsing with Buffer A. Use about 1.2 ml of packed resin for every gram of wet cells in the lysate. Pack resin into a column and rinse twice with 2 bed volumes of Buffer A.
2. Transfer the resin into a tube containing the Empigen-extracted lysed cells (on ice).
3. Tightly close the lid and rotate the tube for . hour (no longer) in the cold room. During this time the detergent-solubilized protein will bind to the nickel resin.
4. Following incubation, isolate the resin by centrifugation of the solution. 70-100% speed for any tabletop centrifuge for 15 minutes should be fine.
5. Pour off the supernatant (try not to lose any of the resin) and either freeze the resin in liquid N2 and store until later use, or transfer to an appropriately-sized column. The height of the packed bed should be more than 4 times the bed diameter, and the total column volume should be about 3-5 times the bed volume. Do not keep the DAGK on resin sitting around at room temperature any longer than you have to; keep on ice and purify immediately following incubation or thawing.
FROZEN DAGK-on-resin can be safely stored at -80 ‹C for about 3 months. However, even at that very low temperature, Ni(II)-catalyzed protein oxidation should be a concern and very old DAGK/resin should be suspect as a source of high quality protein.
6. Wash the resin with about 5 X 1 bed volume of ice-cold Emp/A.
7. Turn on chart recorder and start monitoring A280.
8. Wash column with cold Sanders wash buffer until the junk peak has finished eluting (as monitored by the chart recorder). The wash buffer contains enough imidazole to knock proteins off the column which have a weak affinity for the nickel ions but which do not have the His6 tail. After this step, the target protein will be just about the only protein left sticking to the resin.
9. Rinse column with 12 X 1 bed volume portions of cold rinse buffer.
This does not mean a single rinse of 12 bed volumes. It means to do 12 portions
(“pulses”), 1 at a time, allowing rinse to enter top of column before adding next portion.
The purpose of this buffer is to switch from empigen to DPC or DM.
10. Elute the protein with elution buffer. Target protein will elute as a sharp band which
can be monitored using the chart recorder. Only this band need be collected. If you
are using an eluting buffer containing deuterated imidazole, use only the amount
actually needed to elute. Remember to tare your collection vial while it is empty so
you can measure the volume of protein solution eluted.
11. The protein-containing pool can now be stored if the elution buffer contained DM.
In this case, you can now freeze the solution in liquid nitrogen you can store at -80 °C
and later thaw (for DAGK, loss of activity from freeze-thawing is typically <10%).
If the elution buffer contained DPC, do NOT freeze the solution. Go straight to reconstitutive refolding or store at 4 °C for up to a few days. (Super DAGK in DPC can be stored for weeks at 4 °C with no loss of activity). When subjected to freeze-thaw DAGK in DPC often loses considerable activity.
DAGK Elution Notes: DAGK has an extinction coefficient at 280 of 1.8 O.D. units per mg/ml. Use elution buffer to zero the spectrophotometer. The real absorbance of the DAGK solution at this point may be 15 O.D. units. However, the spectrophotometer only can accurately measure O.D.s of solution with real absorbencies of 2.5 or less. Thus, normally you will add 50-100 microliters of your solution to 1 ml of elution buffer and measure the absorption of that and then multiply by the dilution factor in order to determine the actual O.D. of the DAGK pool. If solution is cloudy it can normally be clarified by adding formic acid to 0.5 % (5 microliters of acid per ml solution).

Lysis Procedure Protocol

1. Take E. coli cell pellet and dilute 20X with lysis buffer in a sealable bottle (i.e. use 20 mL lysis buffer per gram of wet cells). Disperse cells in the solution (mild mixing).
2. Add PMSF (phenylmethylsulfonyl fluoride- a poison!) from a 20 mg/ml stock solution in isopropanol (this can be stored indefinitely in the freezer) to a concentration of 20 mgs per 100 ml of sup (1.1 mM). PMSF is a protease inhibitor which will help keep protein from getting chewed up.
3. Add the following:
lysozyme powder 0.2 mg/ml
powdered DNase and RNase 0.02 mg/ml each
MgAcetate to 5 mM from a 500 mM stock (stock: 11 g/100 ml)
The RNA and DNA in E. coli tend to form thick suspensions.
RNase and DNase will break up this goop.
Seal the container and incubate for ½ hour at room temperature with tumbling
(do not mix with stir bar).
 
4. Tip sonicate at 50% power, 50% duty cycle for 5 minutes (5 sec on, 5 sec off). Place your sample in an ice water bath during sonication.
5. If indicated for your future analyses of pure protein, add dithiothreitol (DTT) to a concentration of 0.5 mM (10 mgs per 100 ml). DTT is a reducing agent which will help keep the Cys thiol groups from getting oxidized. If more than 0.5 mM DTT is added, new nickel resin, not regenerated, must be used.
6. If you wish to stop at this point, lysate can now be divided up into „T40 ml portions in 50 ml Falcon tubes. These can then be frozen in liquid nitrogen and stored in the -80 ¢XC freezer. Note: it can be frozen and stored at this point (before adding detergent), but not after adding detergent.

Protein Expression Protocol

GROWING CULTURES FOR PROTEIN EXPRESSION
 
The goal is to grow large cultures. However, you cannot simply add a colony from a plate to 1 L of media. Instead, we first grow a small liquid culture (5 ml or more in LB media) from one plate colony. This small liquid colony is then used to inoculate 1L of media. The 1L culture is grown until cells reach a certain optical density (determined by A600 ). At this point, IPTG is added to the culture to turn on the lac operon and induce protein synthesis. Cells will normally stop growing at this point and put all their energy into making protein, which is allowed to continue for at least a few hours, after which cultures are harvested.
 
GROWTH PROCEDURE A: 37 °C GROWTH in LB
 
Day 1
Plate cells from a glycerol stock, or transform competent cells. Incubate LB plate at 37 °C overnight.
 
Day 2, evening:
1. Using sterile serological pipet, add 5 mL LB plus appropriate antibiotics to a loose-capped culture tubes (we currently use 17 X 100 mm sterile culture tube)
2. Inoculate the tubes with a single colony using a disposable loop, sterile toothpick, or pipette tip.
3. Incubate with 250 rpm shaking overnight at 37 °C. Do not tighten the lid of the tube; bacteria need oxygen to grow.
 
Day 3, morning:
1. The tubes should be very cloudy as a result of the growth of the microbial culture. Carefully transfer the contents of each tube into 1L (each) of freshly autoclaved LB medium with antibiotics. Put foil covering back on top of flask.
2. Shake at 250 rpm, at 37 °C and monitor cell growth by measuring A600.
3. When A600 reaches 0.7-1 (typically about 5-6 hours), add 0.2 g/ml stock IPTG to a concentration of 0.2 g/liter. IPTG is a substance which activates the promoter which controls transcription on the plasmid. By adding IPTG, you are telling the E. coli culture to start making your protein.
4. Shake culture at 37 °C for another 3 hours to overnight.
5. Harvest cells by centrifugation (see following section for details).
 
GROWTH PROCEDURE B:
Growth at 37 °C, induction at 12 °C
 
This procedure is used for proteins that do not express well at 37 °C. We have found this to be effective for a number of difficult-to-express membrane proteins. The detailed protocol is tucked away in the literature as on-line supporting infomation to a Correction in JACS: Tian, C., Breyer, R. M., Kim, H. K., Karra, M.D., Friedman, D. B., and Sanders, C. R. (2005) Correction to: Solution NMR spectroscopy of the human vasopressin V2 receptor, a G protein-coupled receptor. J. American Chemical Society 128, 5300 (2006). Contact CS if you have trouble tracking this down.
 
GROWTH PROCEDURE C:
Minimal medium for uniformly 15N labeled proteins using H2O
Reference: Oxenoid, K., Kim, H.-K., Jacob, J., Sonnichsen, F. D., and Sanders, C. R. (2004) NMR Assignments for a Helical 40 kDa Membrane Protein in 100 kDa Micelles. J. American Chemical Society 126, 5048-5049.
As with growing cells in LB medium, we first grow a small liquid culture (3-6 ml of LB) from a single colony. This small starter culture is then used to inoculate a 0.5-1L of minimal medium.
 
1. Make minimal medium
„« Add the following amounts per liter of medium:
Na2HPO4 6 g (12.8 g if the 7H2O hydrate form is used)
KH2PO4 3 g
NaCl 0.5 g
15NH4Cl 1.0 g
1) B. Adjust pH to 7.0.
„« C. Autoclave at 121 ¢XC, 15-20 minutes.
„« D. Let medium cool, then add:
1 mL 0.1 M CaCl2
1 mL 1 M MgSO4 ¡P 7 H2O
10 mL 40% glucose (or 20 mL 20% glucose)
+ vitamin solution (see below)
+ antibiotics
 
„X 0.1 M CaCl2
(MW=111 g/mol, 0.111 g CaCl2 in 10 mL MQH2O, filter sterilize)
„X 1 M MgSO4 ¡P 7 H2O
(MW=246 g/mol, 2.46 g MgSO4 ¡P 7 H2O in 10 mL MQH2O, filter sterilize)
„X 40% Glucose (or 20% glucose ¡V can be autoclaved)
„X Vitamins:
Option 1 (CVS vitamins)
Vitamin prep: smash 1 vitamin and mix with 20 ml of H2O. Mix, bath sonicate, and remove insoluble junk by low speed centrifugation. Filter with steriflip. Store at -20 ¢XC. Use 2 mL per L minimal media.
Option 2.(MEM vitamins)
Use commercially available MEM vitamin solution, 10 mL to 1L minimal medium.
 
2. Grow 5 ml LB/amp cultures overnight with shaking at 37 °C.
3. Use one 5 ml tube of LB to inoculate each large flask of minimal medium.
4. Grow with shaking at 37 °C until A600 reaches 0.7-1.0 (about 6-10 hours).
5. Induce with 0.2 g/L IPTG.
6. Continue shaking at 37 °C overnight.
7. Harvest cells the following morning.
 
GROWTH PROCEDURE D:
Cells to use when expressing DAGK: BL21 with pSD005 plasmid containing an inducible synthetic gene for the I53C/I70L/V107D triple mutant of E. coli diacylglycerol kinase. Be careful not to get confused and grow this same mutant in WH1061 (which is Leu auxotroph). BL21 is amp resistant but not Kan resistant.
Note: if WH1061 is used, then you also need to include leucine in the medium. Use 0.25 g/L. Leucine is autoclavable.
 
Small tube (3-6ml) culture:
LB medium with ampicillin 100 ug/ml and NO kanamycin.
Medium scale (50 ml in 250 ml flask) cultures with 100 ug/ml amp:
(1) unlabeled,
(2) 70% D2O as only label,
(3) 99% D2O plus labeled glucose and 15NH4Cl.
Large scale (0.5 L per 2 L flask) culture:
Minimal medium with 100 ug/ml amp in 99% D2O plus labeled glucose and 15NH4Cl.
1. Make the appropriate amounts of the 3 different types of minimal medium (above) and place into correct flasks for autoclaving.
A. Add the following amounts per liter of medium:
Use the appropriate amounts of H2O and/or 99% D2O
Na2HPO4 6 g (12.8 g if the 7-H2O hydrate form is used)
KH2PO4 3 g
NaCl 0.5 g
15NH4Cl 1.0 g
B. Adjust pH=7.0
C. Autoclave
D. Let medium cool, then add:
CaCl2 1 ml/L
MgSO4/7H2O 1 ml/L
40% Glucose (possibly 13C and 2H-labeled) 10 ml/l
Ampicillin 100 mg/l
Vitamins (see below). 2 ml/liter
 
Stock solution of CaCl2 is prepared by dissolving 1.47g of CaCl2 in 100 ml of dd H2O. Filter-sterilize..
Stock solution of MgSO4/7H2O is prepared by dissolving 24.65 g of MgSO4/7H2O in 100 ml of dd H2O. Filter-sterilize.
40% sterilized Glucose is ordered from molecular biology prep lab or can be prepared by large scale sterile filtration.
Ampicillin solutions are prepared and filter sterilized as described above in LB medium section.
Vitamin prep: smash 1 vitamin and mix with 20 ml of H2O. Mix, bath sonicate, and remove insoluble junk by low speed centrifugation. Filter with steriflip.
 
2. Grow 2-5 ml LB/amp cultures overnight with shaking at 37 °C (NO Kanamycin).
3. Use 1 tube of LB to inoculate a 50 ml culture of unlabeled minimal medium.
4. Grow with shaking at 37 °C to OD600=0.5 (about 5 hours). Only a single 50 ml culture is required.
5. Take 0.5 ml aliquot of culture and inoculate 50 ml of minimal medium which is 70% D2O (no other labels need to be included). Only a single 50 ml culture is required.
6. Grow with shaking at 37 °C to OD600=0.5 (about 16 hours).
7. Take 0.5 ml aliquot and inoculate 50 ml of 99% D2O minimal medium which also includes 15N-NH4Cl and sometimes 13C6-glucose (which may also be perdeuterated). Only a single 50 ml culture is required.
8. Grow with shaking at 37 °C to OD600=0.5 (about 20 hours).
9. Take 5 ml aliquot and inoculate 500 ml of 99% D2O minimal medium which also includes 15N-NH4Cl and sometimes 13C6-glucose (which may also be perdeuterated). The same 50 ml culture can be used to inoculate multiple 500 ml cultures.
10. Grow with shaking at 37 °C to OD600=1.0 (about 20 hours using either deuterated or non-deuterated glucose).
11. Induce with 0.2 g/L IPTG and continue shaking at 37 °C for 8 hours to overnight.
12. Harvest cells.