Electrophoresis 1990, I I , 957-962
Recycling isoelectric focusing of multiple sclerosis plasma
focusing protein in the RF3 were Fredicted by initial analysis of the protein preparation in ana ytical isoelectric focusing gels. We are continuing to investig Ite the utility of the instrument for preparation of enzymes from a variety of source materials. Received June 5 , 1990
5. References I 11 Maniatis, T., Fritsch, E. F. and Sar it rook, J., Molecular Cloning: A 121 I31 [41 I51
Laboratory Manual Cold Spring 1 I u b o r Laboratory, Cold Spring Harbor, NY 1982, pp. 237-238. McKeown, M. and Firtel, R. A., C ?I, 1981,24, 799-807. Henikoff,S..Methods Enzymol. 1'87,155, 156-165. Vogt, V. M., Methods Enzymol. 15 81465,248-255. Olesun,A.E. andSasakuma,M.,Ai crr.Biochem.Biophys. 1980,204, 36 1-3 70.
Cherry H. Tamblyn' Geoffrey V. F. SeamanZ Ned B. Egen3 Milan Bier3 'University of Washington, Seattle, WA 'P. 0.Box 1977, Rancho Mirage, CA 3Universityof Arizona, Tucson, AZ
957
161 Shishido, K. and Habuka, N., Biochim. Biophys. Acta 1986, 884, 2 15-2 18. 171 Bier, M., Twitty, G. E. and Sloan, J . E., J . Chromatogr. 1989, 470, 369-376. 181 Vogt, V. M., Eur.J. Biochem. 1973,33, 192-200. 191 Bradford, M. M., Anal. Biochem. 1976, 72,248-254. 101 Laemmli, U. K., Nature 1970,227, 680-685. 11 11 Ando, T., Biochim.Biophys. Acta. 1966,114, 158-168. [121 Rushizky, G. W., Gene Amplif. Anal. 1981,2,205-215. I131 Righetti, P. G. and Gianazza, E., Biochim. Biophys. Acta 1978,532, 137-146. [ 141 Gianazza E. and Righetti, P. G., Biochim. Biophys. Acta 1978,540, 357-364. 1151 Righetti, P. G., Brown, R. P. and Stone, A. L., Biochim. Biophys. Acta 1978,542,232-244. 1161, Sutton, W. D., Biochim. Biophys. Acta 1971,240, 522-531. 1171 Hahn, W. E. and Van Ness, J., Nucleic Acids Res. 1976, 3, 14 19- 1423. 181 Rushizky,G. W.,Shaternikov,V. A.,Mozejko, J. H. and Sober, H. A,, Biochemistry I975,14,4221-4226.
Membrane active plasma factor in multiple sclerosis: Characterization and isolation by recycling isoelectric focusing Recycling isoelectric focusing is a rapid, high resolution technique that has the capability of fractionating complex mixtures of proteins on a preparative scale on the basis of their isoelectric points (pfs). For this reason, it appeared to be an ideal tool to further characterize and isolate the surface active plasma component(s) which is abnormal in multiple sclerosis (MS). The normal control and the abnormal MS plasma components, or factors, proved to be stable under the conditions used in this technique, including deionization by electrodialysis, dialysis against distilled water, lyophilization and the presence of 3~ urea and carrier distilled water, lyophilization and the presence of 3~ urea and carrier ampholytes. The presence or absence of plasma factor activity was determined by incubating red blood cells in a test plasma, or plasma fraction, followed by the determination of the red blood cell electrophoretic mobility in the presence and absence oflinoleic acid. Both the normal and MS factor had a p l of 4.0 & 0.1 under the conditions used. A high degree of purification was achieved and albumin was eliminated as a possible candidate for the factor(s).
1 Introduction Abnormalities in the electrophoretic mobilities ofplatelets I1I, lymphocytes [21 and red blood cells (RBCs) [3-61 from patients with multiple sclerosis (MS) suggest the presence of either abnormal blood plasma components or intrinsic membrane defects. Bolton et al. [ 11 showed that the MS ~
Correspondence: Dr. G. V. F. Seaman, P. 0. Box 1977, Rancho Mirage, CA 92270, USA Abbreviations: ED, electrodialysis; IEF, isoelectric focusing; MS;multiple sclerosis; PBS, phosphate buffered saline; pZ, isoelectric point; PAG, polyacrylamide gel; RBC, red blood cell; RIEF, recycling isoelectric focusing; UEA, unsaturated fatty acids
0VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
platelet abnormalities originated from the plasma in which they were suspended because, after incubation in normal plasma, the platelet electrophoretic mobilities normalized. Similarly, Seaman's group found that erythrocytes from MS patients had normal electrophoretic mobilities in the presence of linoleic acid after incubation in normal plasma. The reverse was also true, since normal RBC acquired abnormal mobilities after incubation in MS plasma [ 7 , S l . As a further demonstration that the electrophoretic abnormalities were conferred by plasma components, Seaman et al. I91 used biologically inert, polystyrene latex particles (PSL) as carrier particles for plasma components. Incubation of PSL in MS plasm a resulted in electrophoretic mobilities in the presence of linoleic acid that were slower than PSL incubated in normal plasma. 0173-0835/90/1111-0957 $3.50+.25/0
958
Electrophoresis 1990, I I . 957-962
Ch. H. Tarnblyn ef al.
Preliminary studies (C. H. Tamblyn el al., in preparation) with the surface active plasma components or factors which conferred these abnormal electrophoretic properties on carrier particles showed that they are removed or structurally altered by the clotting process 18, 101 since they are not detectable in serum. The factors are heat labile, stable at low temperature and have a molecular weight range of 50 000- 100 000 D a by ultrafiltration. The factors are stable in 3 M urea, are water soluble, and can be lyophilized without loss of activity. The characterization, purification and eventual identification of the MS and normal plasma factors could provide valuable information about MS at the molecular level. A method which appeared to achieve these goals, recycling isoelectric focusing (RIEF), was developed by Bier et al. [ 1 11 for the fractionation and characterization of complex mixtures of proteins and peptides on a preparative scale. The samples in the present study are plasmas from MS patients or control subjects that were separated into 10 or 20 fractions.
2 Materials and methods 2.1 Blood sampling and processing Blood samples were drawn from patients (under the care of Dr. Sibley, Department of Neurology, University of Arizona. Tucson, AZ) with clinically definite MS [ 131 and from healthy control subjects, matching in age and sex. The blood was anticoagulated with 3.8 % trisodium citrate and was processed as described previously IS, 61 to obtain the plasma and the saline-washed, fresh RBCs. The electrophoretic mobility of washed RBCs in Medium 199 (GIBCO, Grand Island, NY) was determined according to the method described by Seaman [ 141 by means of an analytical particle electrophoresis apparatus (Rank Bros., Bottisham, England) equipped with reversible silver/silver chloride electrodes and maintained at 25 "C. The RBC-unsaturated fatty acids(UFA)electrophoretic mobility tests for MS entailed measuring the in dividual mobilities of 20 RBCs in each sample in the absence of and then in the presence of 80 pg/mL linoleic acid. The percentage of change in RBC mobility upon the addition of linoleic acid was then determined, as described previously 15, 61. These studies were carried out using a double blind technique.
2.3 RIEF of blood plasma The RIEF apparatus is a modular, preparative scale IEF instrument that separates a pH gradient into discrete fractions. For the present work, it was configured with both 10 or 20 fractions, called flow channels. To operate, the fluid is mixed with a carrier ampholyte and introduced into the apparatus which consists of three basic components: the focusing chamber, which processes the sample; the heat exchange reservoir, which dissipates the Joule heat generated by the focusing chamber and stores the bulk ofthe process fluid; and a multichannel pump which recirculates the process fluid between the two components. A more detailed description of the instrument has been presented I 1 1I. Plasma samples to be fractionated by R I E F were desalted by electrodialysis (ED) since the presence of salts in the RIEF would distort the pH gradient. The E D apparatus consists of a three-compartment electrolytic cell: a sample compartment separated from the electrode compartments by ion exchange membranes. The bulk ofthe sample and the electrolytes (0.05 Msodium sulfate) were maintained in flasks chilled in ice water, while a threechannel pump recirculated the electrolytes and sample between these reservoirs and their respective ED cell subcompartments. Each plasma sample was either diluted to 35 mL with water or diluted with urea and water (final volume 35 mL, 3 M urea) prior to ED. Desalting occurred at 60 V to a final resistance of 400 K ohm/cm, within 5 min. Desalted samples were then centrifuged at 2000 x g for 15 min to remove the precipitate. Less precipitation occurred in the presence of urea. The supernatant fluid was then filtered through Whatman No. I paper and 0.45 pm Millex H A filters (Millipore Corporation, Bedford, MA), mixed with 1 m L o f p H 3.5- 10.0 Ampholine carrier ampholytes (LKB Produkter, Bromma, Sweden), and added to the RIEF primed with 200 mL of the same prefocused carrier ampholyte solution, either with or witout urea, consistent with the prior treatment.
Eight RIEF runs were performed: In runs 1 and 2 the sample consisted of a I 3 mL pool from two MS patients whose RBCs tested positive for MS in the RBC-UFA test. Half of this sample was processed in a 20-channel RIEF with 3 M urea and the other half was processed in the same apparatus without urea. The pooled fractions 1-6 from both the urea and nonurea runs contained the plasma factor activity and were then combined into one fraction and rerun (run 3) with pH 3.5-5.0 Ampholine carrier ampholytes in RIEF. Runs 4-6 were identical to runs 1-3 using normal, instead of MS, plasma. Run 7 was performed on a 29 mL plasma pool from five MS patients 2.2 Assay for plasma factor activity and was processed in a 10-channel RIEF in 3 M urea. Run 8 was identical to run 7, using plasma from normal subjects. The This technique is described elsewhere (C.H. Tamblyn et al., in plasma samples were focused in the RIEF apparatus for 4 h at preparation); briefly, it entails incubating saline-washed pack- 400 V. The fractions were then collected, their pH measured, ed blood, group 0 RBC, in an equal volume of whole blood and then dialyzed against phosphate buffered saline (PBS, plasmaor atestfractionobtainedfrom wholeplasma, at 37 "C 0.15 M NaCl and 0.01 M sodium phosphate) before being for 1 h. The RBCs were then washed three times in saline and assayed for the plasma factor activity. Aliquots of each RIEF once in Medium 199, and were submitted to the RBC-UFA fraction and the original pooled plasma samples were examintest. When test results for normal RBCs shifted to the MS ed by isoelectric focusing (IEF) in polyacrylamide gels range (greater than 3 %I slowing of electrophoretic mobility (PAGs, 5.3 %I acrylamide and0.2 % bisacrylamide(Bi0-Rad, with the addition of linoleic acid), the plasma or plasma frac- Richmond, CA), 5 % w/v Ampholine carrier ampholytes, pH tion was said to have MS plasma factor activity. Conversely, 3.5- 10.0. The samples were visualized by staining with solutions were said to have normal plasma factor activity Coomassie Brilliant Blue G-250 1.51. The protein concentrawhen, after incubation in the solution, the MS RBC test scores tion in each dialyzed RIEF fraction and the pooled plasma changed from the MS range to the normal range (less than 3 96 samples was determined by the spectrophotometric method of slowing). Waddell [ 161.
Recycling isoelectric focusing of multiple sclerosis plasma
Eleclrophoresis 1990.11, 957-962
959
Table 1. Assays for MS and normal plasma factor activities after R E F in pH 3.5 -10.0 Ampholines in the presence and absence of 3 w urea Types of RBC used for assay
Incubation medium to be assayed
Presence of 3 w urea
(+I-)
Change in RBC mobility with linoleic acid
Plasma factor assay result
%
Normal
MS
EDSN 1 2
EDSN
2
No plasma MS plasma-whole MS plasma RIEF Fractions: 1- 6 7-12 13-20 1- 6 7-12 1 3-20
-
+ 2.1
Normal
-
-5.3
MS
-t
- 8.6
-t
+ 0.4
MS Normal Normal
No plasma Normal plasma-whole Normal plasma RIEF 1- 6 Fractions: 7-12 13-20 1- 6 7-12 13-20
-
3
3
4 5
4
6
5 6
7
7
8
8
+
-
-
-
+ 1.1 - 5.4
-
MS Normal
0.0
-t
-
MS Normal Normal
-5.5
+ -t -
9
-0.4 -8.8 +0.7 -t 3.3
Normal MS
-
-
+ 0.9
Normal MS
-5.0 -
10 11 12 13 14 15 16 1 7 18 19 20 EDSN
9 10
ll 12 13 1 4 15 16 1 7 23
19
20
EDSN
Figure I . Analytical IEF in PAG, pH 3.5-10, 3 M urea of RIEF fractions. Aliquots of the 20 RIEF fractions obtained with 3.5-10 Ampholine and 3 M urea and the desalted plasma (two outer lanes, EDSN) were applied to the gel. (A) MS plasma. (B) Normal plasma. Identical gels, not presented, were obtained from the corresponding RIEF runs performed without urea.
960
Ch. H. Tamblyn ef al.
Electrophoresis 1990, 11,957-962
3 Results
Channels 5 , 6 and 7, corresponding to pH 3.9-4.1, showed activity in both the MS and normal plasma fractionations. Aliquots from the 20 channels of these two RlEF fractionations were then analyzed by IEF on PAGs. Figures 2a and 2b show the distribution of MS and normal plasma proteins, respectively. Coomassie Brilliant Blue staining showed no discernible bands in the active fractions, numbers 5 , 6 and 7; however, 200-3 10 pg/mL protein were found in these fractions (Table 3). The results of somewhat larger scale 10-channel RIEF fractionations using pH 3.5- 10.0Ampholine carrier ampholytes are shown in Table 4. Assays showed that the MS and normal plasma factors remained active with a p l of 3.9 to 4.1.
In order to ascertain the stability of the MS plasma factor sujected to the conditions of RIEF fractionation, two preliminary RIEF runs were performed, using plasma pooled from two MS patients. Both runs were in the presence of pH 3.5- 10.0 Ampholine carrier ampholytes, one in the presence and the other in the absence of 3 M urea. Two identical experiments were also performed using plasma from control subjects. The results (Table 1) showed that both the MS and normal plasma factor activities remained intact and were found in the pooled fractions 1-6 in the lowest pH range (approximately pH 3.5-4.5). Figures l a and 1b show the distribution of M S and normal plasma proteins, respectively, separated in the RIEF apparatus and then analyzed by 1EF on PAGs containing 3 M urea and pH 3.5-10.0 Ampholinecarrier ampholytes. Pooled channels 1-6 contained the plasma factors. Similar gels were obtained when 3 M urea was omitted from the R l E F runs and the gels.
4 Discussion RIEF is a powerful technique developed by Bier et a/. I 1 11 for the fractionation of complex mixtures ofproteins or peptides, on a preparative scale. This would appear to be an ideal tool for the characterization and purification of the abnormal surface active plasma component demonstrated in MS plasma 11. 7- I01 as well as the equivalent normal plasma factor, or component. Two initial runs on the R l E F apparatus employed a pool of plasma from two MS patients and a pool of plasma from two control subjects. The separations were carried out by a two-step process, beginning with a separation of the plasmas into 20 fractions by means of carrier ampholytes with a broad pH range of 3.5-10.0. The initial R l E F test runs were carried out, both in the presence and absence of 3 M urea, to determine optimum conditions for maintaining plasma factor
Since the factors remained active both in the presence and absence of 3 M urea, the low pH fractions 1-6 from both MS plasma RIEF runs were pooled and the normal plasma fractions were treated identically. These two pools were then fractionated separately in the RIEF apparatus using narrower range pH 3.5-5.0 Ampholine carrier ampholytes to achieve further purification. The results of preliminary assays on pooled fractions indicate that MS plasma fractions 1-5 and 6-9 contained the plasma factor activity (Table 2). In order to more precisely identify the presence ofthe plasma factors, aliquots from individual RIEF channels were assayed (Table 2).
Table 2. Assays for MS and normal plasma factor activities after RIEF in pH 3.5 -5.0 Ampholinesa) Type of RBC used for assay
Normal
MS
Incubation medium to be assayed
Change in RBC mobility with linoleic acid %
Plasma factor assay result
pH of RlEF fraction
No plasma
+0.5
Normal
-
MS plasma RIEF Fractions: 1- 5 6- 9 10-14 15-19 20 4 5 6 7 8 9
-3.3 -5.1 + 1.1 + 1.0 + 0.2 -2.0 -7.9b) -5.1b) - 7.2b) 0.0 +2.1
MS MS Normal Normal Normal Normal MS MS
-1.0 -5.7 -5.1 -4.3 -5.9 + 2.8c) +2SC) -0.9') -5.6
I 8 ~~
~
7.91
MS
3.85 3.93 4.02 4.06 4.14 4.22 -
MS MS MS MS MS Normal Normal Normal MS
3.46-3.81 4.28 -4.8 1 4.91-5.84 7.18 3.86 3.96 4.02 4.09 4.20
blS Normal Normal
- 9.4
No plasma Normal plasma RIEF Fractions: 1- 3 9-14 15-19 20 4 5 6
3.51-3.93 4.02-4.22 4.30-4.64 4.81-5.58
~
~~~
~
a) Fractions 1-6 of the previous RIEF separations in pH 3.5-10.0 Ampholines were the samples run in this experiment b) MS plasma factor activity in this fraction c) Normal plasma factor activity in this fraction
Recycling isoelectric focusing of multiple sclerosis plasma
Elecrruphuresis 1990.11, 951-962
96 1
Figure 2. Analytical IEF in PAG, pH 3.5-10, 3 M urea of RIEF fractions. Fractions 1-6 from the RIEF runs both with and without urea were pooled and rerun in the RIEF, containing a more narrow pH range Ampholine, 3.5-5, and applied to theabove gels. (A) Pools from the M S plasma RIEF 0x9
1
2 3
4
5
6 7
8
9 10
11 1 2 1 3 1 4 1 5 16 17 18 19 20 Org
Table 3. Assay of MS and normal plasma protcins present in RIEF fractions using pH 3.5 --5.0 Ampholinc carrier ampholytcs Plasma sample
MS plasma
1:raction No.
Original plasma")
Normal plasma
Reagent blank Standard human albumin (0.1 mg/mL)
Protein concentration mg/mL
5 6 7
10.906 0.200 0.220 0.290
Original plasma") 3 4 5 6 I 9
9.097 0.354 0.297 0.260 0.310 0.240 0.354
~
0.000 0.095
a ) Fractions 1-6 of previous R E F using pH 3.5-10.0 Ampholine
~ ~ ~ ~ j B u ) n ~from o o lthe s
'Iasma
activities. The results in Table 1 indicate that both MS and normal plasma factor activities were stable under the conditions used and were present in pooled fractions 1-6 whether or not 3 M urea was present. Fractions 1-6 from both MS plasma runs (+ 3 M urea) were then combined and refocused over a narrower pH range by means of p H 3.5-5.0 carrier ampholytes. The normal plasma fractions were treated in the same manner. Sharp focusing of the proteins of interest was shown by assay to occur at p H 4.0 & 0.1 (Table 2). PAC-IEF was carried out on the original sample applied to and the fractions resulting from - the RIEF experiments. The initial gel shows substantial protein bands in fractions 1-6 of the RIEF where plasma factor activity was present (Fig. la, b). However, the second RIEF runs produced fractions that resulted in the gels shown in Fig. 2a and 2b. Only fractions 5 , 6 and 7, from both MS and normal plasma runs, gave positive results in the plasma factor activity assays, but no protein bands were detectable. A small amount of protein (2003 10 pg/mL) was shown to be present (Table 3). These results indicate considerable purification and high specific activities
962
Ch. H. Tamblyn et at.
Electrophoresis 1990,I1,951-962
Table 4. Assays for the presence of MS and normal plasma factor activities in fractions from two RIEF fractionations of whole MS or normal plasma Type of RBC used for assay
Incubation medium assayed
Change in RBC mobility with linoleic acid %
Plasma factor activity result
PH
-
1 MS plasma fractionated in pH 3.5 -10.0 LKB Ampholines in the presence of 3M urea Normal No plasma -0.7 Normal Whole MS plasma -9.4 MS RIEF fraction: 1 -7.2 MS 2 - 10.1 MS Normal 3 + 1.8 Normal 4 -2.9 5 Normal + 0.4 + 0.5 Normal 6 Normal 7 -0.8 Normal 8 + 1.0 Normal 9 -0.5 Normal 10 -1.3 2 Normal plasma fractionated in pH 3.5-10.0
MS
LKB Ampholines in the presence of 3
Plasma Whole normal plasma RIEF fractions: 1 2 3
-11.5
-
3.93a) 4.09 4.26 4.37 4.63 4.96 5.30 5.63 5.99 7.47 urea
MS
+ 0.6 + 0.6
Normal Normal Normal
-5.7
MS
- 1.0
M
-
-
3.99) 4.1 2a) 4.32
a) pH of fractions showing plasma factor activity
of the factors in these channels. Significantly, the RIEF channels containing albumin (the protein with the highest concentration in plasma), fractions 12-20 (Fig. 2a, b) showed no plasma factor activity (Table 2). With a pI of 4.8, albumin should have focused in channel 14 for the MS and 15 for the normal plasma fractionation, but its high concentration and other molecules (such as fatty acids), absorbed on albumin, resulted in relatively low resolution for this protein. Thus, the RIEF technique successfully achieved the rapid, high resolution separation of the abnormal M S plasma factor from the bulk of the remaining plasma proteins and also provided estimates of its PI.Similar results were obtained for the normal plasma factor. Under the experimental conditions used, a p l of 4.0 -t 0. I was obtained for the factor. It was also shown that the factors were stable during deionization by ED, dialysis against distilled water and lyophilization. They were also stable in 3 M urea and carrier ampholyte systems. One possible application for the purified plasma factors would be the production of antibodies against the factors, which could then be used to pave the way to the future identification and separation of the MS plasma factor and, furthermore, increase our knowledge of MS at the molecular level.
We thank Dr. Sibley for providing samples of bloodfrom M S patients, Lou Zawadski for his technical expertise and enthusiasm, and f a n Wellsfor his timely and incisive comments. This research work was made possible by supportfroin The Margaret W. and Herbert Hoover, Jr., Foundation and by N A S A grant NAGW693. Received June 26, 1990
5 References 1 I I Bolton, C. H., Hampton. J. R. and Phillipson, 0. T., Lancet 1968, i. 99-104. 121 Field, E. J., Shenton, B. K . and Joyce, G., Brit. Med. J. 1974, 1,
412-4 14. 131 Field. E. J.. Joyce. G. and Smith, B. M.. J . Neurol. 1977, 214, 113-127. 141 Field, E. J. and Joyce, G. Lance/ 1976, ii, 367-368. 151 Seaman, G. V. F., Swank, R. L., Tamblyn, C. H. and Zukoski IV. C. F., Lancet 1979, i 1138-1 139. [61 Tamblyn, C. H., Swank, R. L., Seaman, G. V. F. and Zukoski IV, C. F., Neurol. Res. 1980.2, 69-83. 171 Seaman, G. V. F.,Swank, R. L. andZukoski IV.C.F.,Lancer 1979.i. 1139. 181 Seaman, G. V. F., Swank, R. L. andTamblyn, C. H., Lance/ 1980.;. 938. 191 Scaman,G. V . F.,Swank, R. L. andTamblyn,C. H., Newol. 1984.34. 547-549. I101 Field, E. J. and Joyce, G., Neurol. Res. 1986.8,57-60. [ I I I Bier, M., Egen, N. B., Allgyer, T. T.. Twitty, G. E. and Mosher, R . A,. in: Gross. E. and Meienhofer. J. (Eds.), Pepfides: S/cucri,re and Biological Function, Proceedings ofthe Sixth American Peptide Symposium, Pierce Chemical Company, 1979, pp. 79-89. 1 121 Bini0n.S. B., Rodkey, L. S., Egcn, N. B. and Bier, M., Electrophoresis 1982,j. 284-288. 1 131 Hallpike, J. F.. in: Hallpike, J. F.. Adams, C. W. M. andTourtellotte. W. W. (Eds.), Mirlfiple Sclerosis: Pu/hologj,Diagnosis und Manage rnent, Iliarns & Wilkins, Baltimore 1983. pp. 129-162. 1141 Seaman. G. V . F.. in: Mac Suigmor. D. (Ed.), 7 h e Red Blood Cell, Academic Press, New York 1975, Vol. 2 pp. 1135-1229. I151 Blakes1ey.R. W.andBoezi,J.A.,Anal.Biochem. 1977,82,580-585. 1161 Waddell.W.J.,J.Lub.Clin.Med. 1956.48,311-314.
Recycling isoelectric focusing of multiple sclerosis plasma
focusing protein in the RF3 were Fredicted by initial analysis of the protein preparation in ana ytical isoelectric focusing gels. We are continuing to investig Ite the utility of the instrument for preparation of enzymes from a variety of source materials. Received June 5 , 1990
5. References I 11 Maniatis, T., Fritsch, E. F. and Sar it rook, J., Molecular Cloning: A 121 I31 [41 I51
Laboratory Manual Cold Spring 1 I u b o r Laboratory, Cold Spring Harbor, NY 1982, pp. 237-238. McKeown, M. and Firtel, R. A., C ?I, 1981,24, 799-807. Henikoff,S..Methods Enzymol. 1'87,155, 156-165. Vogt, V. M., Methods Enzymol. 15 81465,248-255. Olesun,A.E. andSasakuma,M.,Ai crr.Biochem.Biophys. 1980,204, 36 1-3 70.
Cherry H. Tamblyn' Geoffrey V. F. SeamanZ Ned B. Egen3 Milan Bier3 'University of Washington, Seattle, WA 'P. 0.Box 1977, Rancho Mirage, CA 3Universityof Arizona, Tucson, AZ
957
161 Shishido, K. and Habuka, N., Biochim. Biophys. Acta 1986, 884, 2 15-2 18. 171 Bier, M., Twitty, G. E. and Sloan, J . E., J . Chromatogr. 1989, 470, 369-376. 181 Vogt, V. M., Eur.J. Biochem. 1973,33, 192-200. 191 Bradford, M. M., Anal. Biochem. 1976, 72,248-254. 101 Laemmli, U. K., Nature 1970,227, 680-685. 11 11 Ando, T., Biochim.Biophys. Acta. 1966,114, 158-168. [121 Rushizky, G. W., Gene Amplif. Anal. 1981,2,205-215. I131 Righetti, P. G. and Gianazza, E., Biochim. Biophys. Acta 1978,532, 137-146. [ 141 Gianazza E. and Righetti, P. G., Biochim. Biophys. Acta 1978,540, 357-364. 1151 Righetti, P. G., Brown, R. P. and Stone, A. L., Biochim. Biophys. Acta 1978,542,232-244. 1161, Sutton, W. D., Biochim. Biophys. Acta 1971,240, 522-531. 1171 Hahn, W. E. and Van Ness, J., Nucleic Acids Res. 1976, 3, 14 19- 1423. 181 Rushizky,G. W.,Shaternikov,V. A.,Mozejko, J. H. and Sober, H. A,, Biochemistry I975,14,4221-4226.
Membrane active plasma factor in multiple sclerosis: Characterization and isolation by recycling isoelectric focusing Recycling isoelectric focusing is a rapid, high resolution technique that has the capability of fractionating complex mixtures of proteins on a preparative scale on the basis of their isoelectric points (pfs). For this reason, it appeared to be an ideal tool to further characterize and isolate the surface active plasma component(s) which is abnormal in multiple sclerosis (MS). The normal control and the abnormal MS plasma components, or factors, proved to be stable under the conditions used in this technique, including deionization by electrodialysis, dialysis against distilled water, lyophilization and the presence of 3~ urea and carrier distilled water, lyophilization and the presence of 3~ urea and carrier ampholytes. The presence or absence of plasma factor activity was determined by incubating red blood cells in a test plasma, or plasma fraction, followed by the determination of the red blood cell electrophoretic mobility in the presence and absence oflinoleic acid. Both the normal and MS factor had a p l of 4.0 & 0.1 under the conditions used. A high degree of purification was achieved and albumin was eliminated as a possible candidate for the factor(s).
1 Introduction Abnormalities in the electrophoretic mobilities ofplatelets I1I, lymphocytes [21 and red blood cells (RBCs) [3-61 from patients with multiple sclerosis (MS) suggest the presence of either abnormal blood plasma components or intrinsic membrane defects. Bolton et al. [ 11 showed that the MS ~
Correspondence: Dr. G. V. F. Seaman, P. 0. Box 1977, Rancho Mirage, CA 92270, USA Abbreviations: ED, electrodialysis; IEF, isoelectric focusing; MS;multiple sclerosis; PBS, phosphate buffered saline; pZ, isoelectric point; PAG, polyacrylamide gel; RBC, red blood cell; RIEF, recycling isoelectric focusing; UEA, unsaturated fatty acids
0VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
platelet abnormalities originated from the plasma in which they were suspended because, after incubation in normal plasma, the platelet electrophoretic mobilities normalized. Similarly, Seaman's group found that erythrocytes from MS patients had normal electrophoretic mobilities in the presence of linoleic acid after incubation in normal plasma. The reverse was also true, since normal RBC acquired abnormal mobilities after incubation in MS plasma [ 7 , S l . As a further demonstration that the electrophoretic abnormalities were conferred by plasma components, Seaman et al. I91 used biologically inert, polystyrene latex particles (PSL) as carrier particles for plasma components. Incubation of PSL in MS plasm a resulted in electrophoretic mobilities in the presence of linoleic acid that were slower than PSL incubated in normal plasma. 0173-0835/90/1111-0957 $3.50+.25/0
958
Electrophoresis 1990, I I . 957-962
Ch. H. Tarnblyn ef al.
Preliminary studies (C. H. Tamblyn el al., in preparation) with the surface active plasma components or factors which conferred these abnormal electrophoretic properties on carrier particles showed that they are removed or structurally altered by the clotting process 18, 101 since they are not detectable in serum. The factors are heat labile, stable at low temperature and have a molecular weight range of 50 000- 100 000 D a by ultrafiltration. The factors are stable in 3 M urea, are water soluble, and can be lyophilized without loss of activity. The characterization, purification and eventual identification of the MS and normal plasma factors could provide valuable information about MS at the molecular level. A method which appeared to achieve these goals, recycling isoelectric focusing (RIEF), was developed by Bier et al. [ 1 11 for the fractionation and characterization of complex mixtures of proteins and peptides on a preparative scale. The samples in the present study are plasmas from MS patients or control subjects that were separated into 10 or 20 fractions.
2 Materials and methods 2.1 Blood sampling and processing Blood samples were drawn from patients (under the care of Dr. Sibley, Department of Neurology, University of Arizona. Tucson, AZ) with clinically definite MS [ 131 and from healthy control subjects, matching in age and sex. The blood was anticoagulated with 3.8 % trisodium citrate and was processed as described previously IS, 61 to obtain the plasma and the saline-washed, fresh RBCs. The electrophoretic mobility of washed RBCs in Medium 199 (GIBCO, Grand Island, NY) was determined according to the method described by Seaman [ 141 by means of an analytical particle electrophoresis apparatus (Rank Bros., Bottisham, England) equipped with reversible silver/silver chloride electrodes and maintained at 25 "C. The RBC-unsaturated fatty acids(UFA)electrophoretic mobility tests for MS entailed measuring the in dividual mobilities of 20 RBCs in each sample in the absence of and then in the presence of 80 pg/mL linoleic acid. The percentage of change in RBC mobility upon the addition of linoleic acid was then determined, as described previously 15, 61. These studies were carried out using a double blind technique.
2.3 RIEF of blood plasma The RIEF apparatus is a modular, preparative scale IEF instrument that separates a pH gradient into discrete fractions. For the present work, it was configured with both 10 or 20 fractions, called flow channels. To operate, the fluid is mixed with a carrier ampholyte and introduced into the apparatus which consists of three basic components: the focusing chamber, which processes the sample; the heat exchange reservoir, which dissipates the Joule heat generated by the focusing chamber and stores the bulk ofthe process fluid; and a multichannel pump which recirculates the process fluid between the two components. A more detailed description of the instrument has been presented I 1 1I. Plasma samples to be fractionated by R I E F were desalted by electrodialysis (ED) since the presence of salts in the RIEF would distort the pH gradient. The E D apparatus consists of a three-compartment electrolytic cell: a sample compartment separated from the electrode compartments by ion exchange membranes. The bulk ofthe sample and the electrolytes (0.05 Msodium sulfate) were maintained in flasks chilled in ice water, while a threechannel pump recirculated the electrolytes and sample between these reservoirs and their respective ED cell subcompartments. Each plasma sample was either diluted to 35 mL with water or diluted with urea and water (final volume 35 mL, 3 M urea) prior to ED. Desalting occurred at 60 V to a final resistance of 400 K ohm/cm, within 5 min. Desalted samples were then centrifuged at 2000 x g for 15 min to remove the precipitate. Less precipitation occurred in the presence of urea. The supernatant fluid was then filtered through Whatman No. I paper and 0.45 pm Millex H A filters (Millipore Corporation, Bedford, MA), mixed with 1 m L o f p H 3.5- 10.0 Ampholine carrier ampholytes (LKB Produkter, Bromma, Sweden), and added to the RIEF primed with 200 mL of the same prefocused carrier ampholyte solution, either with or witout urea, consistent with the prior treatment.
Eight RIEF runs were performed: In runs 1 and 2 the sample consisted of a I 3 mL pool from two MS patients whose RBCs tested positive for MS in the RBC-UFA test. Half of this sample was processed in a 20-channel RIEF with 3 M urea and the other half was processed in the same apparatus without urea. The pooled fractions 1-6 from both the urea and nonurea runs contained the plasma factor activity and were then combined into one fraction and rerun (run 3) with pH 3.5-5.0 Ampholine carrier ampholytes in RIEF. Runs 4-6 were identical to runs 1-3 using normal, instead of MS, plasma. Run 7 was performed on a 29 mL plasma pool from five MS patients 2.2 Assay for plasma factor activity and was processed in a 10-channel RIEF in 3 M urea. Run 8 was identical to run 7, using plasma from normal subjects. The This technique is described elsewhere (C.H. Tamblyn et al., in plasma samples were focused in the RIEF apparatus for 4 h at preparation); briefly, it entails incubating saline-washed pack- 400 V. The fractions were then collected, their pH measured, ed blood, group 0 RBC, in an equal volume of whole blood and then dialyzed against phosphate buffered saline (PBS, plasmaor atestfractionobtainedfrom wholeplasma, at 37 "C 0.15 M NaCl and 0.01 M sodium phosphate) before being for 1 h. The RBCs were then washed three times in saline and assayed for the plasma factor activity. Aliquots of each RIEF once in Medium 199, and were submitted to the RBC-UFA fraction and the original pooled plasma samples were examintest. When test results for normal RBCs shifted to the MS ed by isoelectric focusing (IEF) in polyacrylamide gels range (greater than 3 %I slowing of electrophoretic mobility (PAGs, 5.3 %I acrylamide and0.2 % bisacrylamide(Bi0-Rad, with the addition of linoleic acid), the plasma or plasma frac- Richmond, CA), 5 % w/v Ampholine carrier ampholytes, pH tion was said to have MS plasma factor activity. Conversely, 3.5- 10.0. The samples were visualized by staining with solutions were said to have normal plasma factor activity Coomassie Brilliant Blue G-250 1.51. The protein concentrawhen, after incubation in the solution, the MS RBC test scores tion in each dialyzed RIEF fraction and the pooled plasma changed from the MS range to the normal range (less than 3 96 samples was determined by the spectrophotometric method of slowing). Waddell [ 161.
Recycling isoelectric focusing of multiple sclerosis plasma
Eleclrophoresis 1990.11, 957-962
959
Table 1. Assays for MS and normal plasma factor activities after R E F in pH 3.5 -10.0 Ampholines in the presence and absence of 3 w urea Types of RBC used for assay
Incubation medium to be assayed
Presence of 3 w urea
(+I-)
Change in RBC mobility with linoleic acid
Plasma factor assay result
%
Normal
MS
EDSN 1 2
EDSN
2
No plasma MS plasma-whole MS plasma RIEF Fractions: 1- 6 7-12 13-20 1- 6 7-12 1 3-20
-
+ 2.1
Normal
-
-5.3
MS
-t
- 8.6
-t
+ 0.4
MS Normal Normal
No plasma Normal plasma-whole Normal plasma RIEF 1- 6 Fractions: 7-12 13-20 1- 6 7-12 13-20
-
3
3
4 5
4
6
5 6
7
7
8
8
+
-
-
-
+ 1.1 - 5.4
-
MS Normal
0.0
-t
-
MS Normal Normal
-5.5
+ -t -
9
-0.4 -8.8 +0.7 -t 3.3
Normal MS
-
-
+ 0.9
Normal MS
-5.0 -
10 11 12 13 14 15 16 1 7 18 19 20 EDSN
9 10
ll 12 13 1 4 15 16 1 7 23
19
20
EDSN
Figure I . Analytical IEF in PAG, pH 3.5-10, 3 M urea of RIEF fractions. Aliquots of the 20 RIEF fractions obtained with 3.5-10 Ampholine and 3 M urea and the desalted plasma (two outer lanes, EDSN) were applied to the gel. (A) MS plasma. (B) Normal plasma. Identical gels, not presented, were obtained from the corresponding RIEF runs performed without urea.
960
Ch. H. Tamblyn ef al.
Electrophoresis 1990, 11,957-962
3 Results
Channels 5 , 6 and 7, corresponding to pH 3.9-4.1, showed activity in both the MS and normal plasma fractionations. Aliquots from the 20 channels of these two RlEF fractionations were then analyzed by IEF on PAGs. Figures 2a and 2b show the distribution of MS and normal plasma proteins, respectively. Coomassie Brilliant Blue staining showed no discernible bands in the active fractions, numbers 5 , 6 and 7; however, 200-3 10 pg/mL protein were found in these fractions (Table 3). The results of somewhat larger scale 10-channel RIEF fractionations using pH 3.5- 10.0Ampholine carrier ampholytes are shown in Table 4. Assays showed that the MS and normal plasma factors remained active with a p l of 3.9 to 4.1.
In order to ascertain the stability of the MS plasma factor sujected to the conditions of RIEF fractionation, two preliminary RIEF runs were performed, using plasma pooled from two MS patients. Both runs were in the presence of pH 3.5- 10.0 Ampholine carrier ampholytes, one in the presence and the other in the absence of 3 M urea. Two identical experiments were also performed using plasma from control subjects. The results (Table 1) showed that both the MS and normal plasma factor activities remained intact and were found in the pooled fractions 1-6 in the lowest pH range (approximately pH 3.5-4.5). Figures l a and 1b show the distribution of M S and normal plasma proteins, respectively, separated in the RIEF apparatus and then analyzed by 1EF on PAGs containing 3 M urea and pH 3.5-10.0 Ampholinecarrier ampholytes. Pooled channels 1-6 contained the plasma factors. Similar gels were obtained when 3 M urea was omitted from the R l E F runs and the gels.
4 Discussion RIEF is a powerful technique developed by Bier et a/. I 1 11 for the fractionation of complex mixtures ofproteins or peptides, on a preparative scale. This would appear to be an ideal tool for the characterization and purification of the abnormal surface active plasma component demonstrated in MS plasma 11. 7- I01 as well as the equivalent normal plasma factor, or component. Two initial runs on the R l E F apparatus employed a pool of plasma from two MS patients and a pool of plasma from two control subjects. The separations were carried out by a two-step process, beginning with a separation of the plasmas into 20 fractions by means of carrier ampholytes with a broad pH range of 3.5-10.0. The initial R l E F test runs were carried out, both in the presence and absence of 3 M urea, to determine optimum conditions for maintaining plasma factor
Since the factors remained active both in the presence and absence of 3 M urea, the low pH fractions 1-6 from both MS plasma RIEF runs were pooled and the normal plasma fractions were treated identically. These two pools were then fractionated separately in the RIEF apparatus using narrower range pH 3.5-5.0 Ampholine carrier ampholytes to achieve further purification. The results of preliminary assays on pooled fractions indicate that MS plasma fractions 1-5 and 6-9 contained the plasma factor activity (Table 2). In order to more precisely identify the presence ofthe plasma factors, aliquots from individual RIEF channels were assayed (Table 2).
Table 2. Assays for MS and normal plasma factor activities after RIEF in pH 3.5 -5.0 Ampholinesa) Type of RBC used for assay
Normal
MS
Incubation medium to be assayed
Change in RBC mobility with linoleic acid %
Plasma factor assay result
pH of RlEF fraction
No plasma
+0.5
Normal
-
MS plasma RIEF Fractions: 1- 5 6- 9 10-14 15-19 20 4 5 6 7 8 9
-3.3 -5.1 + 1.1 + 1.0 + 0.2 -2.0 -7.9b) -5.1b) - 7.2b) 0.0 +2.1
MS MS Normal Normal Normal Normal MS MS
-1.0 -5.7 -5.1 -4.3 -5.9 + 2.8c) +2SC) -0.9') -5.6
I 8 ~~
~
7.91
MS
3.85 3.93 4.02 4.06 4.14 4.22 -
MS MS MS MS MS Normal Normal Normal MS
3.46-3.81 4.28 -4.8 1 4.91-5.84 7.18 3.86 3.96 4.02 4.09 4.20
blS Normal Normal
- 9.4
No plasma Normal plasma RIEF Fractions: 1- 3 9-14 15-19 20 4 5 6
3.51-3.93 4.02-4.22 4.30-4.64 4.81-5.58
~
~~~
~
a) Fractions 1-6 of the previous RIEF separations in pH 3.5-10.0 Ampholines were the samples run in this experiment b) MS plasma factor activity in this fraction c) Normal plasma factor activity in this fraction
Recycling isoelectric focusing of multiple sclerosis plasma
Elecrruphuresis 1990.11, 951-962
96 1
Figure 2. Analytical IEF in PAG, pH 3.5-10, 3 M urea of RIEF fractions. Fractions 1-6 from the RIEF runs both with and without urea were pooled and rerun in the RIEF, containing a more narrow pH range Ampholine, 3.5-5, and applied to theabove gels. (A) Pools from the M S plasma RIEF 0x9
1
2 3
4
5
6 7
8
9 10
11 1 2 1 3 1 4 1 5 16 17 18 19 20 Org
Table 3. Assay of MS and normal plasma protcins present in RIEF fractions using pH 3.5 --5.0 Ampholinc carrier ampholytcs Plasma sample
MS plasma
1:raction No.
Original plasma")
Normal plasma
Reagent blank Standard human albumin (0.1 mg/mL)
Protein concentration mg/mL
5 6 7
10.906 0.200 0.220 0.290
Original plasma") 3 4 5 6 I 9
9.097 0.354 0.297 0.260 0.310 0.240 0.354
~
0.000 0.095
a ) Fractions 1-6 of previous R E F using pH 3.5-10.0 Ampholine
~ ~ ~ ~ j B u ) n ~from o o lthe s
'Iasma
activities. The results in Table 1 indicate that both MS and normal plasma factor activities were stable under the conditions used and were present in pooled fractions 1-6 whether or not 3 M urea was present. Fractions 1-6 from both MS plasma runs (+ 3 M urea) were then combined and refocused over a narrower pH range by means of p H 3.5-5.0 carrier ampholytes. The normal plasma fractions were treated in the same manner. Sharp focusing of the proteins of interest was shown by assay to occur at p H 4.0 & 0.1 (Table 2). PAC-IEF was carried out on the original sample applied to and the fractions resulting from - the RIEF experiments. The initial gel shows substantial protein bands in fractions 1-6 of the RIEF where plasma factor activity was present (Fig. la, b). However, the second RIEF runs produced fractions that resulted in the gels shown in Fig. 2a and 2b. Only fractions 5 , 6 and 7, from both MS and normal plasma runs, gave positive results in the plasma factor activity assays, but no protein bands were detectable. A small amount of protein (2003 10 pg/mL) was shown to be present (Table 3). These results indicate considerable purification and high specific activities
962
Ch. H. Tamblyn et at.
Electrophoresis 1990,I1,951-962
Table 4. Assays for the presence of MS and normal plasma factor activities in fractions from two RIEF fractionations of whole MS or normal plasma Type of RBC used for assay
Incubation medium assayed
Change in RBC mobility with linoleic acid %
Plasma factor activity result
PH
-
1 MS plasma fractionated in pH 3.5 -10.0 LKB Ampholines in the presence of 3M urea Normal No plasma -0.7 Normal Whole MS plasma -9.4 MS RIEF fraction: 1 -7.2 MS 2 - 10.1 MS Normal 3 + 1.8 Normal 4 -2.9 5 Normal + 0.4 + 0.5 Normal 6 Normal 7 -0.8 Normal 8 + 1.0 Normal 9 -0.5 Normal 10 -1.3 2 Normal plasma fractionated in pH 3.5-10.0
MS
LKB Ampholines in the presence of 3
Plasma Whole normal plasma RIEF fractions: 1 2 3
-11.5
-
3.93a) 4.09 4.26 4.37 4.63 4.96 5.30 5.63 5.99 7.47 urea
MS
+ 0.6 + 0.6
Normal Normal Normal
-5.7
MS
- 1.0
M
-
-
3.99) 4.1 2a) 4.32
a) pH of fractions showing plasma factor activity
of the factors in these channels. Significantly, the RIEF channels containing albumin (the protein with the highest concentration in plasma), fractions 12-20 (Fig. 2a, b) showed no plasma factor activity (Table 2). With a pI of 4.8, albumin should have focused in channel 14 for the MS and 15 for the normal plasma fractionation, but its high concentration and other molecules (such as fatty acids), absorbed on albumin, resulted in relatively low resolution for this protein. Thus, the RIEF technique successfully achieved the rapid, high resolution separation of the abnormal M S plasma factor from the bulk of the remaining plasma proteins and also provided estimates of its PI.Similar results were obtained for the normal plasma factor. Under the experimental conditions used, a p l of 4.0 -t 0. I was obtained for the factor. It was also shown that the factors were stable during deionization by ED, dialysis against distilled water and lyophilization. They were also stable in 3 M urea and carrier ampholyte systems. One possible application for the purified plasma factors would be the production of antibodies against the factors, which could then be used to pave the way to the future identification and separation of the MS plasma factor and, furthermore, increase our knowledge of MS at the molecular level.
We thank Dr. Sibley for providing samples of bloodfrom M S patients, Lou Zawadski for his technical expertise and enthusiasm, and f a n Wellsfor his timely and incisive comments. This research work was made possible by supportfroin The Margaret W. and Herbert Hoover, Jr., Foundation and by N A S A grant NAGW693. Received June 26, 1990
5 References 1 I I Bolton, C. H., Hampton. J. R. and Phillipson, 0. T., Lancet 1968, i. 99-104. 121 Field, E. J., Shenton, B. K . and Joyce, G., Brit. Med. J. 1974, 1,
412-4 14. 131 Field. E. J.. Joyce. G. and Smith, B. M.. J . Neurol. 1977, 214, 113-127. 141 Field, E. J. and Joyce, G. Lance/ 1976, ii, 367-368. 151 Seaman, G. V. F., Swank, R. L., Tamblyn, C. H. and Zukoski IV. C. F., Lancet 1979, i 1138-1 139. [61 Tamblyn, C. H., Swank, R. L., Seaman, G. V. F. and Zukoski IV, C. F., Neurol. Res. 1980.2, 69-83. 171 Seaman, G. V. F.,Swank, R. L. andZukoski IV.C.F.,Lancer 1979.i. 1139. 181 Seaman, G. V. F., Swank, R. L. andTamblyn, C. H., Lance/ 1980.;. 938. 191 Scaman,G. V . F.,Swank, R. L. andTamblyn,C. H., Newol. 1984.34. 547-549. I101 Field, E. J. and Joyce, G., Neurol. Res. 1986.8,57-60. [ I I I Bier, M., Egen, N. B., Allgyer, T. T.. Twitty, G. E. and Mosher, R . A,. in: Gross. E. and Meienhofer. J. (Eds.), Pepfides: S/cucri,re and Biological Function, Proceedings ofthe Sixth American Peptide Symposium, Pierce Chemical Company, 1979, pp. 79-89. 1 121 Bini0n.S. B., Rodkey, L. S., Egcn, N. B. and Bier, M., Electrophoresis 1982,j. 284-288. 1 131 Hallpike, J. F.. in: Hallpike, J. F.. Adams, C. W. M. andTourtellotte. W. W. (Eds.), Mirlfiple Sclerosis: Pu/hologj,Diagnosis und Manage rnent, Iliarns & Wilkins, Baltimore 1983. pp. 129-162. 1141 Seaman. G. V . F.. in: Mac Suigmor. D. (Ed.), 7 h e Red Blood Cell, Academic Press, New York 1975, Vol. 2 pp. 1135-1229. I151 Blakes1ey.R. W.andBoezi,J.A.,Anal.Biochem. 1977,82,580-585. 1161 Waddell.W.J.,J.Lub.Clin.Med. 1956.48,311-314.
