Continuous Beds for Microchromatography: Detection of Proteins by a Blotting Membrane Technique

Continuous Beds for Microchromatography: Detection of Proteins by a Blotting Membrane Technique

ANALYTICAL BIOCHEMISTRY ARTICLE NO. 241, 195–198 (1996) 0399 Continuous Beds for Microchromatography: Detection of Proteins by a Blotting Membrane ...

305KB Sizes 0 Downloads 29 Views

ANALYTICAL BIOCHEMISTRY ARTICLE NO.

241, 195–198 (1996)

0399

Continuous Beds for Microchromatography: Detection of Proteins by a Blotting Membrane Technique1 Jia-Li Liao, Cheng-Ming Zeng,2 Anders Palm, and Stellan Hjerte´n3 Department of Biochemistry, Uppsala University, Biomedical Center, P.O. Box 576, S-751 23 Uppsala, Sweden

Received May 16, 1996

Continuous beds have been used as matrices for cation- and anion-exchange chromatography of proteins on columns with an i.d. in the range of 0.005–0.015 mm. On-tube uv detection is not feasible at low protein concentrations with these narrow-bore columns. Therefore, a more sensitive detection system has been developed based on blotting technique: as the protein zones leave the microcolumn chromatographically they become adsorbed onto a rotating polyvinylidene difluoride blotting membrane. The protein spots can then be visualized by means of Coomassie brilliant blue, immunomethods, and other standard techniques. By using an immunomethod 0.015 ng of human transferrin can easily be detected. The blotting membrane can be washed with water without loss of adsorbed protein. This is an attractive feature because the presence of salts, etc., diminishes the accuracy in the determination of molecular weights of proteins by mass spectrometry. The microcolumns are easy to prepare. A solution of appropriate monomers is sucked into a piece of fused silica tubing. The rod formed upon polymerization contains channels through which the eluent can pass. No supporting frit is required because the polymer rod is anchored by covalent bonds to the tubing wall. q 1996 Academic Press, Inc.

In many biological, chemical, and physical disciplines there is steadily increasing interest in the systems and methods which permit the handling of minute amounts of material. This trend is very obvious in the 1 This study was supported by the Swedish Natural Science Research Council, the Swedish Research Council for Engineering Sciences, and the Carl Trygger Foundation. 2 Permanent address: Department of Bioanalytical Chemistry, Chongqing University of Medical Sciences, Chongqing 630 046, People’s Republic of China. 3 To whom correspondence should be addressed. Fax: /46 18 552139.

field of separation science, of which the rapid development of capillary electrophoresis is an example. The chromatographic counterpart, microchromatography, also has great potential, but grows relatively slowly, one reason being that it is difficult to pack extremely narrow column tubes uniformly with beads, which is a prerequisite for high resolution. This problem has been overcome because in the first paper (1) in a series of articles on microchromatography we have shown that continuous bed-based microcolumns have (1) a uniform packing even when the column diameter is as small as 10 mm, and (2) good separation power which is independent of the column diameter, at least for inner diameters in the range of 0.025–6 mm. However, the difficulty in detecting microamounts of solutes (proteins) by uv monitoring prompted us to develop an alternative, but more sensitive method. It is based on adsorption of proteins onto a moving blotting membrane followed by staining with standard methods. MATERIALS AND METHODS

Chemicals Piperazine diacrylamide (PDA),4 N,N,N*,N*-tetramethylethylenediamine (TEMED), ammonium persulfate (electrophoresis purity reagents), tris(hydroxymethyl)aminomethane (Tris), ammonium sulfate (HPLC grade), polyvinylidene difluoride (PVDF) membranes (pore size 0.2 mm), and an amplified alkaline phosphatase goat anti-rabbit immunoblot assay kit were obtained from Bio-Rad Laboratories (Hercules, CA); g-methacryloxypropyltrimethoxysilane (Bind-Silane) and transferrin were from Pharmacia LKB Biotechnology (Uppsala, Sweden); and cytochrome c, myoglobin, lysozyme, and lactoglobulin from Sigma (St. 4 Abbreviations used: PDA, piperazine diacrylamide; TEMED, N,N,N*,N*-tetramethylethylenediamine; PVDF, polyvinylidene difluoride; MA, methacrylamide; DMDA, dimethyl diallyl ammonium chloride; HEMA, 2-hydroxyethyl methacrylate.

195

0003-2697/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID

AB 9670

/

6m1e$$$621

09-19-96 19:37:18

aba

AP: Anal Bio

196

LIAO ET AL.

FIG. 1. A schematic drawing of the apparatus for transferring proteins from a microcolumn onto a moving blotting membrane. P, HPLC pump; U, union; T, Teflon tube; TU, tee union; SC, splitter column; MC, microcolumn; UV, detection window; W, wheel; F, filter paper; M, blotting membrane. As the proteins leave the column, MC, they are adsorbed onto the moving blotting membrane, M, and are made visible by staining. In a separate or the same experiment the solute zones can be detected optically as they pass the window, UV, provided that the concentration of the proteins is high enough (see Fig. 2).

Louis, MO). Human growth hormone was a gift from Professor Paul Roos, this department. Human serum was obtained from the University Hospital (Uppsala, Sweden). Methacrylamide (MA) was from Fluka Chemie AG (Buchs, Switzerland); acrylic acid was from Merck (Schuchardt, Germany); dimethyl diallyl ammonium chloride (DMDA) was from Polysciences (Warrington, PA); and 2-hydroxyethyl methacrylate (HEMA) was from Aldrich-Chemie (Steinheim, Germany). All other reagents used were of analytical grade. Fused silica capillaries with 0.015, 0.010, and 0.005 mm i.d. (o.d. 0.14 mm) were purchased from MicroQuartz (Munich, Germany). Apparatus A schematic drawing of the device for detection of proteins in the effluent from a microcolumn is shown in Fig. 1. A blotting membrane (M) and a moistened filter paper (F) are attached to a plastic wheel (W), the speed of which is adjusted by a gear and a step motor.

The microcolumn tube (MC), a piece of fused silica tubing, touches the membrane. As the proteins leave the column they become adsorbed onto the rotating membrane in the form of streaks (S), which are made visible by appropriate standard staining methods (2–6). The HPLC system consisted of a pump (Model 2150), an LC controler (Model 2152), and a recorder (Model 2210) from LKB (Bromma, Sweden) and a uv detector (Model 200) from Linear Instruments (Reno, NV). Preparation of Continuous Beds for Microchromatography Columns for cation-exchange chromatography. The columns (5, 10, and 15 mm i.d.) were prepared as described in Ref. (1). The column tube (a piece of fused silica tubing) was first washed sequentially with 0.1 M sodium hydroxide, 0.1 M hydrochloric acid, and distilled water, and then filled with a 10% (v/v) solution of gmethacryloxypropyltrimethoxysilane in acetone. After 1 h this solution was sucked out of the tube and nitro-

FIG. 2. Separation of proteins by cation-exchange microchromatography on a continuous bed and their detection by a uv monitor and a blotting technique. Sample: Myoglobin (M), cytochrome c (C), and lysozyme (L); 0.25 mg/ml of each protein. Buffer system: A, 0.02 M sodium phosphate, pH 5.8; B, 0.6 M sodium chloride in A. Linear gradient, 0–100% buffer B. The nonsplit gradient volume was 50 ml. (a) Column: 0.015 mm (i.d.) 1 80 mm (effective length). Optical detection at 220 nm (top chromatogram; AUF, 0.005) and detection by means of an amplified alkaline phosphatase goat anti-rabbit immunoblot assay (bottom chromatogram, 24 ng of protein in each spot). (b) Column: 0.010 mm (i.d.) 1 80 mm (effective length). Optical detection at 220 nm (top chromatogram; AUF, 0.002) and detection by means of an amplified alkaline phosphatase goat anti-rabbit immunoblot assay (bottom chromatogram, 7.5 ng of protein in each spot).

AID

AB 9670

/

6m1e$$$622

09-19-96 19:37:18

aba

AP: Anal Bio

MICROCHROMATOGRAPHY: DETECTION BY BLOTTING

FIG. 3. Detection of transferrin by the blotting technique described herein following cation-exchange microchromatography on a continuous bed. Column: 0.005 mm (i.d.) 1 70 mm (effective length). Buffer system, gradient, and detection: see the legend to Fig. 2. The concentration of transferrin was too low to permit uv detection. (a) Sample: 0.015 ng of transferrin (b) Sample: 0.050 nl of normal human serum.

gen was passed through to remove liquid adhering to the tube wall. By using this procedure the capillary wall was activated with methacryl groups, which then reacted with the monomers in the subsequent synthesis of the chromatographic bed to link the bed covalently to the wall. The continuous bed was prepared as follows (1). PDA (0.135 g) and MA (0.110 g) were dissolved in 1.0 ml of 0.05 M sodium phosphate (pH 7.0). Acrylic acid (14 ml), 5 M sodium hydroxide (28 ml), and ammonium sulfate (0.082 g) were then added. Air was expelled from this monomer solution for 2 min with a water-suction pump, followed by addition of 12 ml of a 10% (w/v) solution of

197

FIG. 5. Test of the sensitivity of the immunological dot-blotting technique described herein. Detection method: see the legend to Fig. 3. (a) Sample: human transferrin (4, 2, 1, 0.5, 0.2, 0.1, 0.075, and 0.050 ng of human transferrin). The blotting technique thus permits detection of 0.05 ng of transferrin. (b) Sample: human serum (1, 0.5, 0.25, 0.10, 0.075, 0.050, 0.025, and 0.010 nl).

ammonium persulfate and 12 ml of a 5% (v/v) solution of TEMED. The chromatographic tube was filled under microscopic inspection with this monomer solution by means of a syringe to a point 0.5 cm from the detection window (see below). The syringe and the fused silica tubing were connected by a reducing union. The polymerization was allowed to proceed at room temperature for at least 6 h. A 2-mm-wide section of the polyimide coating was then burned off by means of an electrically heated tungsten wire to create a uv-transparent window for on-tube detection. Columns for anion-exchange chromatography. The preparation procedure was similar to that described above for cation exchangers, although the composition of the monomer solution differed. Following dissolution of PDA (0.190 g) in 1.0 ml of 0.05 M sodium phosphate, pH 7.0, DMDA (0.40 ml), HEMA (0.24 ml), and ammonium sulfate (0.075 g) were added. Chromatographic Procedures

FIG. 4. Detection of proteins by Coomassie brilliant blue and the blotting technique described herein following anion-exchange microchromatography on a continuous bed. Buffer system: A, 0.02 M Tris–HCl, pH 8.6; B, 0.5 M sodium chloride in A. Linear gradient: 0–100% buffer B. The nonsplit gradient volume was 50 ml. (a) Column: 0.015 mm (i.d.) 1 100 mm (effective length). Sample: lactoglobulin (left; 24 ng) and human growth hormone (18 ng). Staining with Coomassie brilliant blue. (b) Column: 0.005 mm (i.d.) 1 70 mm (effective length). Sample: human transferrin (0.15 ng). Detection: see the legend to Fig. 3.

AID

AB 9670

/

6m1e$$$622

09-19-96 19:37:18

The apparatus used is outlined in the right-hand section of Fig. 1. Details, including how to create a concentration gradient for the elution, are described in Ref. (1). All chromatographic procedures were performed at room temperature. The buffers were filtered through a 0.25-mm cellulose nitrate filter (Schleicher & Schu¨ll, Dassel, Germany) and deaerated. The columns were equilibrated with buffer A (for composition, see the legends to the figures) for at least 40 min before each run. The linear salt gradients were created by sucking up into an 800-mm-long piece of Teflon tubing (i.d. 0.25 mm) 10 5-ml volumes of solutions of different concentrations of sodium chloride in buffer A (1). The 4-ml sample contained 3 proteins, which were detected in the effluent by a uv-monitor and/or a blotting technique (Fig. 1). The analysis times in the different experiments were 7–11 min.

aba

AP: Anal Bio

198

LIAO ET AL.

Blotting Method and Staining Techniques The device used is shown at the left in Fig. 1. A strip of filter paper (Munktell’s, No. 3, Grycksbo, Sweden) was attached to the wheel (W) and moistened with the same buffer as was employed for the equilibration of the column. A 1.0 1 8.0-cm piece was cut from the commercial PVDF blotting membrane and mounted on top of the filter paper and moistened with the same buffer (to wet the membrane it was first immersed in methanol for a few seconds). The filter paper keeps the blotting membrane wet during the chromatographic analysis. When the run was finished, the membrane was disconnected from the wheel (W) and washed three times with distilled water. The protein spots on the membrane were then stained with Coomassie brilliant blue (3, 4), except for the experiment on the 5-mm column, where the latter method was not sensitive enough to detect the small amount of transferrin applied. The transferrin was therefore detected by means of the amplified alkaline phosphatase goat anti-rabbit immunoblot assay kit. The staining was performed according to the supplier’s instructions with the difference that PVDF membranes were used instead of nitrocellulose membranes. The sensitivity of the immunoblot assay kit was studied by a dot-blot assay. For this purpose a sample of transferrin was diluted successively with buffer. From each solution 0.4 ml was withdrawn and put on a blotting membrane for detection, as described. The experiment was then repeated with human serum as sample. Before photographing, all membranes were dried on filter paper and sandwiched between polyester sheets. RESULTS AND DISCUSSION

The most widely used on-tube detection method for proteins in capillary electrophoresis and capillary (micro) chromatography is that based on the uv absorption of the solutes (top chromatograms in Fig. 2). However, when the inner diameter of the capillary is in the range of 0.005–0.015 mm, as in the experiments described herein, the sensitivity often is too low to permit satisfactory detection. Labeling of the proteins with appropriate groups for fluorescence measurements increases the sensitivity considerably, but requires further studies to become a general, cheap, simple, and reproducible detection method. We have earlier described a blot-

AID

AB 9670

/

6m1e$$$622

09-19-96 19:37:18

ting technique to detect small amounts of proteins in capillary electrophoresis ((7); see also Ref. (8)). With some modifications this technique can be used with advantage also in microchromatography, as shown herein. The proteins were stained by Coomassie brilliant blue (Fig. 4a) and an immunological method (Figs. 2 (bottom chromatograms), 3, and 4b). The sensitivity of the latter method is demonstrated in Fig. 5. However, many more techniques are available, for instance, those based on radioactive labeling (3) or staining with silver, colloidal gold (4, 5), mercurochrome, dimethylaminoazobenzene isothiocyanate (4), and india ink (6). The most narrow columns used in the experiments described in this paper had an inside diameter of only 0.005 mm. It is likely that these columns have a relatively uniform ‘‘packing’’ because columns of double the diameter exhibit a regular bed upon microscopy (1). Microcolumns are very useful for the concentration of minute volumes of proteins and other biopolymers, provided that the experimental conditions are such that the solutes are adsorbed onto the column. A desalting step must often precede a subsequent analysis when the desorption requires buffers of high concentration. For this purpose we have developed several methods, described in Refs. (9–11). A new technique based on the use of a hollow fiber is under development. All of these methods also permit concentration of the sample. REFERENCES 1. Li, Y.-M., Liao, J.-L., Nakazato, K., Mohammad, J., Terenius, L., and Hjerte´n, S. (1994) Anal. Biochem. 223, 153–158. 2. LeGendre, N. (1990) BioTechniques (Suppl.) 9, 788–805. 3. Baker, C. S., Dunn, M. J., and Yacoub, M. H. (1991) Electrophoresis 12, 342–348. 4. Christiansen, J., and Houen, G. (1992) Electrophoresis 13, 179– 183. 5. Nelson, T. J. (1993) Anal. Biochem. 214, 325–328. 6. Immobilon TM Tech Protocol TP 008, Millipore Corp., Milford, MA. 7. Eriksson, K.-O., Palm, A., and Hjerte´n, S. (1992) Anal. Biochem. 201, 211–215. 8. Beck, S. (1988) Anal. Biochem. 170, 361–366. 9. Hjerte´n, S., Valtcheva, L., and Li, Y.-M. (1994) J. Cap. Elec. 1, 83–89. 10. Hjerte´n, S., Liao, J.-L., and Zhang, R. (1994) J. Chromatogr. A 676, 409–420. 11. Liao, J.-L., Zhang, R., and Hjerte´n, S. (1994) J. Chromatogr. A 676, 421–430.

aba

AP: Anal Bio