Electrophoresis 1990,11,489-494
Kong So0 Khoo Josie A. Beeley Oral BiochemistryUnit, Dental School Oral Biology Group and Department of Biochemistry, University of Glasgow
Hybrid IEF of human salivary proteins
489
Isoelectric focusing of human parotid salivary proteins in hybrid carrier ampholyte-immobilizedpH gradient polyacrylamide gels Isoelectric focusing of human salivary proteins with carrier ampholyte-isoelectric focusing systems requires prior desalting and concentration of samples, a procedure which is time-consuming and requires relatively large volumes of samples. By contrast, immobilized pH gradient gels are more tolerant to salt loads. Thus pretreatment of samples consists only of centrifugation prior to isoelectric focusing. If larger loads (> 50 pg) are required, the samples may be concentrated by lyophilization and reconstitution in a smaller volume of water or by dialysis against 30 % w/v polyethylene glycol. Immobilized pH gradient polyacrylamide gels (incorporating a hybrid carrier ampholyte system) of two pH ranges (pH 4-9 and pH 3.5-5.0) have been used to separate the proteins in human parotid saliva. The effects of urea on focused patterns were studied; in pH 4-9 gels it gave improved resolution of protein bands, whereas in pH 3.5-5.0 gels it prevented protein precipitation. The salivary proteins were then visualized by staining with Coomassie Brilliant Blue G-250 or a silver procedure. Usl ing the latter, 25-30 well-resolved bands were formed on a p H 4-9 gel loaded with 20 pg of proteins. The method offers considerable advantages compared with carrier ampholyte-isoelectric focusing.
1 Introduction The analysis of human salivary proteins by isoelectric focusing (IEF) in carrier ampholyte (CA) polyacrylamide gels [ 1-41 has offered considerable advantages over other analytical procedures and has been invaluable in providing information regarding genetic variants in salivary proteins, such as variant a-amylase isozymes 151, or for studying salivary protein abnormalities in cystic fibrosis [ 61, Sjogren’s Syndrome [ 7 ] and rheumatoid arthritis [8]. While saliva is high in electrolyte levels, its protein content is low and variable (0.5-4.2 g/L) [9], making it necessary to desalt and concentrate before IEF. Failure to adequately reduce the salt content of samples will cause distortion ofthe pH gradient and give impaired resolution of protein bands [ lo]. Desalting saliva either by ultrafiltration [ l ] or dialysis [4] is time-consuming and requires relatively large volumes of saliva (preferably in excess of 2 mL) and can also result in the loss of some constituents; for example, proteins of low molecular weight ( M , < 10 000) may be lost by passage through the membranes. In view of this, immobilized pH gradients (IPGs) appear to offer considerable advantages over CA-based IEF for salivary protein analysis because their pH gradients are unaffected by electrolytes. Saliva may therefore be applied directly to an IPG. If a high protein load is required, the samples may be lyophilized and applied as concentrates. We have investigated the separation of human parotid salivary proteins on broad pH range (pH 4-9) and narrow pH range (pH 3.5-5.0) IPGs, followed by staining with either Coomassie Brilliant Blue G-250 or a silver procedure. IPGs of the latter pH range could be useful in studying acidic prolinerich proteins [ 11-141, considerable variation in these having been attributed to genetic polymorphisms [13, 151. It would Correspondence: Dr. Josie A. Beeley, Oral Biochemistry Unit, University of Glasgow Dental Hospital and School, 378 Sauchiehall Street, Glasgow G2 352,Scotland Abbreviations: BCA, bicinchoninic acid; CA, carrier ampholytes, IEF, isoelectric focusing; IPG, immobilized pH gradient polyacrylamide gel; pZ, isoelectric point 0VCH Verlagsgesellschaft mbH D-6940 Weinheim, 1990
also be of value in the study of the anionic salivary proteins associated with rheumatoid arthritis and Sjogren’s Syndrome 181.
2 Materials and methods 2.1 Chemicals Immobilines (pK’s 3.6, 4.6, 6.2, 7.0, 8.5 and 9.3) and Ampholines (pH 3.5-5 and pH 3.5-10) were purchased from Pharmacia-LKB Biotechnology (Milton Keynes, UK). GelBond P A G film was purchased from ICN Biochemicals (Bucks, UK). Glycerol, formaldehyde, ammonium persulfate, acrylamide, N,N’-methylenebisacrylamideand colored isoelectric point markers (pH 2.4-5.65 and pH 4.7-10.6) of “Analar” or “Electran” grade, polyethylene glycol 10 000 (“Organics” grade) and urea (“Aristar” grade) were from BDH (Poole, Dorset, UK). N,N,N’,N’-Tetramethylethylenediamine (TEMED), bovine a-chymotrypsinogen and Coomassie Brilliant Blue G-250 from Sigma Chemicals (Poole, UK), silver nitrate from Johnson Matthey Chemical Ltd. (Herts, UK) and bicinchoninic acid (BCA) Protein Assay Reagent from Pierce C hemical Company (Rockford, IL, USA).
2.2 Collection and preparation of saliva samples Parotid saliva from normal healthy volounteers was collected on ice with a modified Carlsson-Crittenden cup [161 and lemon juice used to stimulate flow. Samples were stored at -20 “C until use. After thawing, they were centrifuged at 20 000 g for 20 min at 4 OC to remove insoluble material. When higher protein loads (> 50 pg) are necessary, concentration may be achieved by either of the following methods: (i) lyophilization and reconstitution with water to a fifth of their original volumes, followed by centrifugation; or (ii) dialysis overnight, following centrifugation, against 30 % w/v polyethylene glycol using a BRL Model 1200 MA Microdialysis Unit (purchased from Bethesda Research Laboratories, Gaithersburg, MD, USA) fitted with a 6000-8000 molecular weight cut-off membrane. The latter method has the advan0 173-0835/90/0606-0489 $02.50/0
490
K. Soo Khoo and J. A. Beeley
tage of simultaneously reducing the salt content of the samples. The protein content of concentrated samples was determined by the BCA method [ 171 using bovine a-chymotrypsinogen as a standard [181.
2.3 Preparation of gels IPGs, pH 4-9 and p H 3.5-5.0, of dimensions 0.5 x 180 x 250 mm3 were cast according to the manufacturer's instructions (LKB ApplicationNote 324 and LKB Laboratory Manual for Multiphor I1 Electrophoresis System). After polymerization, washing and drying, IPGs were placed in an LKB reswelling cassette and rehydrated as follows: pH 4-9 IPGs were rehydrated with an aqueous solution containing 20 % w/v glycerol and 0.5 % w/v Ampholine, p H 3.5-10. The pH 3.5-5.0 IPGs were rehydrated in a solution containing 8 M urea and 0.5 % w/v pH 3.5-5 Ampholine in 20 % w/v glycerol. In order to prevent lateral dispersion of samples by electrolytes, 0.5 cm spaces were cut between 1.0 cm wide tracks. To investigate the effect ofurea om the focused patterns of proteins, an IPG of each pH range was rehydrated with a transverse urea gradient (0-8 M) perpendicular to the pH gradient by placing the dried gel in areswelling cassette and using a gradient former to superimpose the urea gradient. 2.4 Sample application Using a micropipette, sample volumes of up to 50 pL were applied to the surface of an IPG. Sample application was made at various positions on the gel to determine: the application point which causes the least distortion of the focused pattern.
Electrophoresis 1990,11,489-494
2.5 Running conditions The anolyte was 10 mM glutamic acid, and the catholyte was 10 mM sodium hydroxide. After sample application, IPGs were initially focused at 500 V for 1 h [ 191 to prevent the formation of strongly acidic or alkaline migrating boundaries, after which IEF was continued at 5000 V, 5 mA and 5 W overnight. An LKB 2297 Macro-Drive 5 constant power supply system was used, and the cooling temperature was 10 "C.
2.6 Fixing and staining Following IEF, gels were fixed by incubation in an aqueous solution of 10 % w/v trichloroacetic acid containing 3.5 % w/v perchloric acid I81 for 30 min followed by staining with an acid extract of Coomassie Brilliant Blue (3-250 [201. Gels were stained for 1 h, briefly washed in water, and destained in 10 % v/v acetic acid. After Coomassie-staining, gels could be silver stained when necessary [211. Gels which were silverstained without Coomassie-prestaining were fixed in the trichloroacetic acid-perchloric acid solution described above before staining. 2.7 Effect of the salt content on focused patterns The following experiment was designed to determine the salt concentration that would cause protein denaturation: 12 mL of parotid saliva was desalted by overnightdialysis against 1L of distilled water, and concentrated by a further overnight dialysis against 30 % w/v polyethylene glycol (a microdialysis unit, as describedin Section 2.2, was used) and theproteincon-
.Figure I . (A) IEF of human parotid salivary proteins from a single individual on a pH 4-9 IPG followed by Coomassie Brilliant Blue G-250 staining. The protein loads were: 10, 20,30, 50,60, 100, 125 and 250 kg. (B) The same gel as shown in (A) after silver staining.
Hybrid IEF of human salivary proteins
Electrophoresis 1990, 1 I , 489-494
centration determined. Proteins (25 pg) were loaded on a p H 4-9 I P G in increasing concentrations of NaCl(0-600 mM). This was achieved by adding measured volumes of stock NaCl solutions of concentrations 50 mM, 0.1 M, 0.5 M, and 1 M to fixed volumes of dialyzed saliva. In order to compare the focused patterns of undesalted and desalted lyophilizates, the following experiment was performed: seven I mL aliquots of parotid saliva from one individual were lyphilized and reconstituted with distilled water in order to prepare the following concentrated solutions: x 1 (i.e. unconcentrated), x 2, x 3, x 4, x 5, x 6 and x 7. Half of each sample was desalted by overnight dialysis (in a microdialysis unit) against distilled water. The IEF patterns of undialyzed and desalted samples were compared on adjacent tracks on a pH 4-9 I P G in order of increasing concentration. The sodium, potassium and calcium concentrations ofthe unconcentrated sample ofparotid saliva were determined, using a Corning flame photometer (Corning Ltd., Essex, UK), and were found to be 18 mM, 16 mM and 0.2 mM, respectively.
3 Results 3.1 Effect of application distances on focused pattern The results of analysis of human salivary proteins on hybrid CA-IPGs are shown in Fig. 1. U p to 25-30 well defined protein bands were clearly resolved, the degree of resolution and number depending on the protein load and the stain employed. Optimum results were obtained on a p H 4-9 I P G (data not shown) when the samples were applied 2-3 cm from the cathode, where the local pH is 6.5-7.0. Some degree of precipitation and streaking was observed if application was made at other distances. Application within 1 cm of the cathode resulted in the most severe streaking, probably due to precipitation caused by hydrophobic interactions between proteins and alkaline Immobilines 1221. In p H 3.5-5.0 IPGs, however, it was not possible to achieve good separation when I E F was performed under non-denaturing conditions. There was streaking and most bands failed to focus properly. This problem is not caused by salts in the samples, as even desalted samples failed to produce satisfactory bands, nor was separation improved even by mixing CAs with the samples prior to focusing. In order to obtain optimum results from an I P G of this pH range, it was necessary to perform IEF in 8 M urea(see Section 3.5).
49 1
3.2 Comparison of Coomassie and silver staining A range of protein loads (10-250 pg) was separated on a pH 4-9 I P G and subsequently stained by Coomassie Brilliant Blue G-250 followed by silver stain (Fig. 1). Using the silver stain technique, 20 pg of protein allowed visualization of approximately 25 bands in the gel, while Coomassie Blue staining required 250 pg in order to give similar results. The band patterns, however, were similar using both procedures, and the differences were only at the level of sensitivity rather than additional components being stained by the silver procedure.
3.3 Effect of urea
On a pH 4-9 I P G (Fig. 2) a secondary urea gradient (0-8 M urea) increases the resolution, especially above a urea concentration of 5 M. Most of the proteins focused as continuous bands throughout the length oftheIPG and somecomponents exhibited a slight PI shift towards the anode, a finding which has been reported with other proteins at urea concentrations greater than 2 M [23]. When a similar urea gradient is superimposed on a pH 3.5-5.0 IPG (Fig. 3) all the bands display a continuous shift to the anode as the concentration of urea increases. Below a concentration of 6 M urea the band patterns were poorly resolved, but above this value they were satisfactory.
3.4 Effect of salt levels Upon adding known levels of sodium chloride to samples, it was observed that denaturation occurred at concentrations of 200 mM and above (Fig. 4), the effect being more pronounced on the cathodic side of application points. The most abundant cations in stimulated parotid saliva [9] are sodium and potassium, which have mean concentrations of about 55 mM and I 6 mM, respectively, while the most abundant anions are chloride (33 mM) and phosphate (6 mM). Although there are other ionic species present in saliva, these will have the greatest effect in denaturing proteins as they will form strong acid and alkaline boundaries upon the application of a voltage across the gel [ 191. As such, NaCl has been chosen as the standard salt solution. Denaturation was observed only when the concentration of sodium was 200 mM and above, and this is considerably greater than the level normally found in stimulated parotid saliva. N o differences were observed between desalted and undialyzed lyophilizates which have been concentrated up to 7-fold (Fig. 4).
Figure 2. IEF of a continuous zone of normal human parotid salivary proteins applied 2 cm from the cathode throughout the length of the gel in a p H 4-9 IPG superimposed with a transverse 0-8 M urea gradient. 1 mg of protein was applied and the gel was silver stained.
492
Elenrophoresis 199O,II, 489-494
K . Soo Khoo and J . A. Beeley
Figure 3. IEF of a continuous zone of normal human parotid salivary proteins applied 2 cm from the cathode throughout the length of the gel in a pH 3.5-5.0 IPG superimposed with a transverse0-8 M urea gradient. The protein load was 10.5 mg and the gel was silver stained. Arrowheads indicate bands the resolution of which improved with increasing ureaconcentration.
Figure 4 . (A)IEF on pH 4-9 IPG of human parotid salivary protein samples containing increasing concentrations ofNaC1. The samples containea 25 pgof protein in 50 FL and were from the same individual. Arrowheads indicate precipitation and denaturation. (B) IEF on pH 4-9 IPG ofdesalted and undialyzed lyophilized concentrates of human parotid saliva. One- to 7-fold concentrates were applied to tracks (1) to (7). The tracks marked “U” were loaded with undialyzed samples while tracks marked“D” wereloaded with desalted samples; 25 pg ofprotein were applied on each track and the gel was silver stained.
3.5 Comparison of salivary protein patterns from different individuals Salivary proteins from 6 different subjects separated on a p H 4-9 IPG (Fig. 5) showed few major but several minor variations from individual to individual. The technique appears to give better resolution than that of Eckersall et al. [41 who performed IEF on CA gels and also reported that the band patterns of different individuals were essentially similar with only minor differences. Acidic range IPGs ((pH3.5-5.0) revealed few differences between normal individuals, and indeed very little material focusing in this area, but. numerous additional components were observed in patients with autoimmune diseases such as rheumatoid arthritis and Sjogren’s syndrome. Mairs and Beeley [Sl have previously separated the anionic salivary proteins associated with these disorders on CA-containing gels, but they were not well resolved as they precipitated out at or near their isoelectric points. Urea-containing pH 3.5-5.OIPGs (Fig. 6)overcame the problem ofproteinprecipitation and resolved these proteins into up to 20-25 components (Khoo, K. S., Beeley, J. A. andLaimey,P.-J., manuscript in preparation). By contrast, normal healthy individuals had about 6-10 minor bands focusing in this region.
4 Discussion
In order to further the understanding of the functions of salivary proteins and to develop the potentials of analysis of this fluid in clinical diagnosis, improved analytical electrophoretic procedures are needed. Whilst IEF of human parotid saliva using C A has given improved resolution as compared with other electrophoretic procedures and revealed further information about the nature of proteins present in both normal, healthy individuals [4,241 and in disorders such as connective tissue disease [81, use of the technique has been impaired by the need for prior desalting of samples because of the high electrolyte to protein levels. This procedure is time-consuming and leads to protein loss; furthermore, the sample sizes necessary for manipulation (> 2 mL) are frequently not available from diseased glands. The technique is also subject to problems such as cathodic drift together with the fact that the formation of “tailor-made” gradients necessary for the proteins under investigation can be difficult or impossible.
IPGs overcame most of these problems [251 but did sometimes give rise to streaking and sample precipitation at the application site. Addition of C A to the IPGs to form mixed-bed
Electrophoresis 1990. 1 1 . 489-494
Hybrid IEF 01‘ human salivary proteins
493
CA-IPGs [26, 271 resolved this problem; the procedure described in this paper is based on this methodology. Incorporation of ureain the gel gave improved resolution in pH 4-9 IPGs, and was found to be essential for the pH 3.5-5.0 IPGs; in its absence the components precipitated out and poor results were obtained. Rehydrating the gels with 20 % w/v glycerol was found to be advantageous as it prevented liquid exudation 1281. When liquid exudation or “sweating” occurs, the resulting droplets on the gel surface after prolonged focusing may adversely affect the focused pattern. As 8 M urea solutions have a weak inhibitory effect on liquid exudation, glycerol was also included in the rehydration solution for pH 3.5-5.0 gels. A limitation of IPGs is their inherently low conductivity, which causes difficulties such as slow sample entry, lateral zone spreading and the formation of salt fronts which halt protein migration. A hybrid CA-IPG system was employed to circumvent this problem [29,301.This was achieved by reswelling dried IPGs in CA-containing rehydration solutions.
Figure 5. IEF of parotid saliva from 6 different individuals (a)-(f) on a pH 4-9 IPG. The protein load was 25 Fg per sample and the gel was silver stained.
Figure 6 . IEF of human parotid saliva using pH 3.5-5.0 IPG rehydrated with 8 M urea in 20 % w/v glycerol and 0.5 % w/v pH 3.5-5 Ampholine. Tracks (v). (x) and (z) were loaded with 100 Fg of salivary proteins from healthy individuals, while track (w) was loaded with 100 pg salivary proteins from a patient with Sjogren’s syndrome and track (y) from a patient with rheumatoid arthritis; the gel was silver stained.
Salts present in samples are capable of causing denaturation because acidic and alkaline boundaries are produced at application points. Thus even low levels of salts (5 mM) have been reported to induce modification of haemoglobin patterns and higher levels to cause denaturation [ 191. In order to investigate whether the levels of electrolytes present in saliva would have any adverse effects, a study was performed to determine the salt level which would lead to protein precipitation and denaturation. We have found salivary proteins to be less affected by salts, and satisfactory patterns were obtained from samples containing up to 200 mM NaCI. This is well above the electrolyte levels in parotid saliva. No precipitation was observed when lyophilizates concentrated up to sevenfold without desalting were fractionated on IPGs. The pH 4-9 IPGs of parotid salivary proteins have produced highly resolved focused patterns. With samples containing as little as 20 pg of proteins, 25-30 components were evident on silver staining. Interestingly, IPGs, like CA-based IEF, although revealing some individual-to-individual differences in protein pattern in parotid saliva, they were less extensive than is the case with sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) [3 11. Urea is often used as a solubilizing agent in electrophoresis [321 and IEF in gels containing urea gradients perpendicular to the pH axis 133-351 may yield important clues about protein structure. Alterations in the pls ofproteins will occur if unfolding by urea results in the exposure of charged groups [361. In IPGs of both pH ranges, urea was found to cause shifts in the pls of most protein bands but results in better resolution. It was also necessary to incorporate 8 M urea into p H 3.5-5.0 IPGsinordertopreventprecipitation.UsingpH 3.5-5.OIPGs containing 8 M urea, larger sample loads (100 pg) were necessary to detect the lower levels of protein which focus in this range. Whilst some individual-to-individual variation in band pattern was observed in normal subjects, the abnormal proteins which are associated with connective tissue disorders [81 and focus in this region were resolved into 20 or more additional well-defined components. The I P G technique gives superior resolution to CA-IEF of human salivary proteins 141 in both broad and narrow pH range gels; this is especially the case in pH 3.5-5.0 gels, over which pH range carrier ampholytes produce virtually no
494
K. So0 Khoo and J. A. Beeley
separation. Analysis of salivary proteins by focusing on IPGs also obviates the need to desalt samples prior to analysis and can be used with sample volumes as small as 50 FL. Accordingly, it is now possible to analyze rapidly large numbers of parotid saliva samples (including those from patients with xerostomia) as a function of pl. This development should lead to considerable progress in the analysis ofhuman salivary proteins.
W e uregratefulfor~nunciulassistunce.frorn the Universityof Glusgow MucFeut Bequest. The valuubi'eadvice and technical assistance of M r . D. Sweeney is gratefully acknowledged. Received January 2, 1990
5 References Beeley, J. A., Archs. Oral. Biol. 1969, 14, 559-561. Arneberg, P., Scand. J . Dent. Res. 1972, 80, 134-138. Beeley, J. A., in: Catsimpoolas, N. and Drysdale, J. (Eds.), Biological and Biomedical Applications of Isoelectric Focusing, Plenum Press, London 1977, pp. 1-28. Eckersall,P.D.,Mairs,R. J. andBeeley,J. A.,Archs. Oral.Biol. 1981, 26,727-733. Pronk, J. C. and Frants, R. R., Hum. Hered. 1979,29, 181-186. Bustos, S. E. and Fung, L. in: Allen, R. C. and Arnaud, P. (Eds.), Electrophoresis '81, Walter de Gruyter, B,erlin 1981, pp. 3 17-328. Chisholm, D. M., Beeley, J. A. and Mason, D. K., Oral. Surg. 1973, 36,620-630. Mairs, R. J. and Beeley, J. A., Clin. Sci. 1985, 69, 727-735. Mason, D. K. and Chisholm, D. M., Saliiiary Glands in Health and Disease, W. B. Saunders, Philadelphia 1975, pp. 37-69. Jonsson, M., Electrophoresis 1980,1, 1411-149. Oppenheim, F. G., Hay, D. I. and Franzbdau, C.,J. Biochem. 1971, 10,4233-4238. Bennick, A., J. Biochem. 1977,163,229-239. Azen, E. A., Biochem. Genet. 1978,16, 7'3-99. Azen, E. A., Archs. Oral. Biol. 1978,23, 1173-1176.
Electrophoresis 1990,l I, 489-494
[ l 5 J Azen,E.A.andOppenheim,F.G.,Science 1973,180,1067-1069. [I61 Stephen, K. W. and Spiers, C. F., Br. J. Clin. Pharm. 1976, 3, 3 16-3 19. [ 171 Smith, P. K., Krohn, R. I., Hermanson, G .T., Mallia, A. K., Gartner, F. H., Provenzano, M. D., Fujimoto, E. K., Goeke, N. M., Olsen, B. J. and Klenk, D. C., Anal. Biochem. 1985,150,7645. [I81 Wilcox, P. E., Cohen, E. and Tan, W., J . Biol. Chem. 1947, 228, 999- 1019. I191 Righetti, P. G., Chiari, M. and Gelfi, C., Electrophoresis 1988, 9, 65-73. 1201 Mairs, R. J., Ph. D. Thesis, University of Glasgow, 1981, p. 65. 1211 Damerval, C., le Guilloux, M., Blaisonneau, J. and de Vienne, D., Electrophoresis 1987,8, 158-159. 1221 Rabilloud, T., Gelfi, C., Bossi, M. L. and Righetti, P. G., Electrophoresis 1987,8, 305-312. 1231 Gorg, A., Postel, W. and Johann, P., J. Biochem. Biophys. Methods 1985,10,341-350. 1241 Eckersall, P. D. and Beeley, J . A., Biochem. Genetics 1981, 19, 1055- 1062. 1251 Righetti, P. G., J . Chromatogr. 1984,300, 165-223. 1261 Rimpilainen, M. and Righetti, P. G., Electrophoresis 1985, 6, 4 19-422. L271 Righetti, P. G., J. Biochem. Biophys. Methods 1988, 16,99-108. 1281 Altland, K., Hackler, R. and Rossmann, U., Electrophoresis 1986,7, 25 1-259. I291 Altland, K., von Eckardstein, A., Banzhoff, A., Wagner, M., Rossmann, U., Hackler, R. and Becker, P., Electrophoresis 1987,8, 52-62. [30] Fawcett,J. S. andChrambach,A.,Electrophoresis 1986,7,266-272. [311 Lindsay, J. C. B. and Beeley, J. A. in: Dunn, M. J. (Ed.), Electrophoresis '86, VCH Publishers, Weinheim 1986,pp. 222-225. 1321 Andrews, A. T., Electrophoresis: Theory, Techniques and Biochemical and Clinical Applications,Oxford University Press, Oxford 1986, p. 252. 1331 Hobart,M. J.,in:Arbuthnott,J.P. andBeeley,J.A.(Eds.),Zsoelectric Focusing, Butterworths, London 1975, pp. 275-280. 1341 Altland, K., Roeder,T., Jakin, H. M., Zimmer, H. G. andNeuhoff,V., Clin. Chem. 1982,28,4, 1000-1010. 1351 Altland, K., Banzhoff, A., Hackler, R. and Rossmann, U., Electrophoresis 1984,5, 379-381. [361 Creighton, T. E., J. Mol. Biol. 1979,129,235-264.
Kong So0 Khoo Josie A. Beeley Oral BiochemistryUnit, Dental School Oral Biology Group and Department of Biochemistry, University of Glasgow
Hybrid IEF of human salivary proteins
489
Isoelectric focusing of human parotid salivary proteins in hybrid carrier ampholyte-immobilizedpH gradient polyacrylamide gels Isoelectric focusing of human salivary proteins with carrier ampholyte-isoelectric focusing systems requires prior desalting and concentration of samples, a procedure which is time-consuming and requires relatively large volumes of samples. By contrast, immobilized pH gradient gels are more tolerant to salt loads. Thus pretreatment of samples consists only of centrifugation prior to isoelectric focusing. If larger loads (> 50 pg) are required, the samples may be concentrated by lyophilization and reconstitution in a smaller volume of water or by dialysis against 30 % w/v polyethylene glycol. Immobilized pH gradient polyacrylamide gels (incorporating a hybrid carrier ampholyte system) of two pH ranges (pH 4-9 and pH 3.5-5.0) have been used to separate the proteins in human parotid saliva. The effects of urea on focused patterns were studied; in pH 4-9 gels it gave improved resolution of protein bands, whereas in pH 3.5-5.0 gels it prevented protein precipitation. The salivary proteins were then visualized by staining with Coomassie Brilliant Blue G-250 or a silver procedure. Usl ing the latter, 25-30 well-resolved bands were formed on a p H 4-9 gel loaded with 20 pg of proteins. The method offers considerable advantages compared with carrier ampholyte-isoelectric focusing.
1 Introduction The analysis of human salivary proteins by isoelectric focusing (IEF) in carrier ampholyte (CA) polyacrylamide gels [ 1-41 has offered considerable advantages over other analytical procedures and has been invaluable in providing information regarding genetic variants in salivary proteins, such as variant a-amylase isozymes 151, or for studying salivary protein abnormalities in cystic fibrosis [ 61, Sjogren’s Syndrome [ 7 ] and rheumatoid arthritis [8]. While saliva is high in electrolyte levels, its protein content is low and variable (0.5-4.2 g/L) [9], making it necessary to desalt and concentrate before IEF. Failure to adequately reduce the salt content of samples will cause distortion ofthe pH gradient and give impaired resolution of protein bands [ lo]. Desalting saliva either by ultrafiltration [ l ] or dialysis [4] is time-consuming and requires relatively large volumes of saliva (preferably in excess of 2 mL) and can also result in the loss of some constituents; for example, proteins of low molecular weight ( M , < 10 000) may be lost by passage through the membranes. In view of this, immobilized pH gradients (IPGs) appear to offer considerable advantages over CA-based IEF for salivary protein analysis because their pH gradients are unaffected by electrolytes. Saliva may therefore be applied directly to an IPG. If a high protein load is required, the samples may be lyophilized and applied as concentrates. We have investigated the separation of human parotid salivary proteins on broad pH range (pH 4-9) and narrow pH range (pH 3.5-5.0) IPGs, followed by staining with either Coomassie Brilliant Blue G-250 or a silver procedure. IPGs of the latter pH range could be useful in studying acidic prolinerich proteins [ 11-141, considerable variation in these having been attributed to genetic polymorphisms [13, 151. It would Correspondence: Dr. Josie A. Beeley, Oral Biochemistry Unit, University of Glasgow Dental Hospital and School, 378 Sauchiehall Street, Glasgow G2 352,Scotland Abbreviations: BCA, bicinchoninic acid; CA, carrier ampholytes, IEF, isoelectric focusing; IPG, immobilized pH gradient polyacrylamide gel; pZ, isoelectric point 0VCH Verlagsgesellschaft mbH D-6940 Weinheim, 1990
also be of value in the study of the anionic salivary proteins associated with rheumatoid arthritis and Sjogren’s Syndrome 181.
2 Materials and methods 2.1 Chemicals Immobilines (pK’s 3.6, 4.6, 6.2, 7.0, 8.5 and 9.3) and Ampholines (pH 3.5-5 and pH 3.5-10) were purchased from Pharmacia-LKB Biotechnology (Milton Keynes, UK). GelBond P A G film was purchased from ICN Biochemicals (Bucks, UK). Glycerol, formaldehyde, ammonium persulfate, acrylamide, N,N’-methylenebisacrylamideand colored isoelectric point markers (pH 2.4-5.65 and pH 4.7-10.6) of “Analar” or “Electran” grade, polyethylene glycol 10 000 (“Organics” grade) and urea (“Aristar” grade) were from BDH (Poole, Dorset, UK). N,N,N’,N’-Tetramethylethylenediamine (TEMED), bovine a-chymotrypsinogen and Coomassie Brilliant Blue G-250 from Sigma Chemicals (Poole, UK), silver nitrate from Johnson Matthey Chemical Ltd. (Herts, UK) and bicinchoninic acid (BCA) Protein Assay Reagent from Pierce C hemical Company (Rockford, IL, USA).
2.2 Collection and preparation of saliva samples Parotid saliva from normal healthy volounteers was collected on ice with a modified Carlsson-Crittenden cup [161 and lemon juice used to stimulate flow. Samples were stored at -20 “C until use. After thawing, they were centrifuged at 20 000 g for 20 min at 4 OC to remove insoluble material. When higher protein loads (> 50 pg) are necessary, concentration may be achieved by either of the following methods: (i) lyophilization and reconstitution with water to a fifth of their original volumes, followed by centrifugation; or (ii) dialysis overnight, following centrifugation, against 30 % w/v polyethylene glycol using a BRL Model 1200 MA Microdialysis Unit (purchased from Bethesda Research Laboratories, Gaithersburg, MD, USA) fitted with a 6000-8000 molecular weight cut-off membrane. The latter method has the advan0 173-0835/90/0606-0489 $02.50/0
490
K. Soo Khoo and J. A. Beeley
tage of simultaneously reducing the salt content of the samples. The protein content of concentrated samples was determined by the BCA method [ 171 using bovine a-chymotrypsinogen as a standard [181.
2.3 Preparation of gels IPGs, pH 4-9 and p H 3.5-5.0, of dimensions 0.5 x 180 x 250 mm3 were cast according to the manufacturer's instructions (LKB ApplicationNote 324 and LKB Laboratory Manual for Multiphor I1 Electrophoresis System). After polymerization, washing and drying, IPGs were placed in an LKB reswelling cassette and rehydrated as follows: pH 4-9 IPGs were rehydrated with an aqueous solution containing 20 % w/v glycerol and 0.5 % w/v Ampholine, p H 3.5-10. The pH 3.5-5.0 IPGs were rehydrated in a solution containing 8 M urea and 0.5 % w/v pH 3.5-5 Ampholine in 20 % w/v glycerol. In order to prevent lateral dispersion of samples by electrolytes, 0.5 cm spaces were cut between 1.0 cm wide tracks. To investigate the effect ofurea om the focused patterns of proteins, an IPG of each pH range was rehydrated with a transverse urea gradient (0-8 M) perpendicular to the pH gradient by placing the dried gel in areswelling cassette and using a gradient former to superimpose the urea gradient. 2.4 Sample application Using a micropipette, sample volumes of up to 50 pL were applied to the surface of an IPG. Sample application was made at various positions on the gel to determine: the application point which causes the least distortion of the focused pattern.
Electrophoresis 1990,11,489-494
2.5 Running conditions The anolyte was 10 mM glutamic acid, and the catholyte was 10 mM sodium hydroxide. After sample application, IPGs were initially focused at 500 V for 1 h [ 191 to prevent the formation of strongly acidic or alkaline migrating boundaries, after which IEF was continued at 5000 V, 5 mA and 5 W overnight. An LKB 2297 Macro-Drive 5 constant power supply system was used, and the cooling temperature was 10 "C.
2.6 Fixing and staining Following IEF, gels were fixed by incubation in an aqueous solution of 10 % w/v trichloroacetic acid containing 3.5 % w/v perchloric acid I81 for 30 min followed by staining with an acid extract of Coomassie Brilliant Blue (3-250 [201. Gels were stained for 1 h, briefly washed in water, and destained in 10 % v/v acetic acid. After Coomassie-staining, gels could be silver stained when necessary [211. Gels which were silverstained without Coomassie-prestaining were fixed in the trichloroacetic acid-perchloric acid solution described above before staining. 2.7 Effect of the salt content on focused patterns The following experiment was designed to determine the salt concentration that would cause protein denaturation: 12 mL of parotid saliva was desalted by overnightdialysis against 1L of distilled water, and concentrated by a further overnight dialysis against 30 % w/v polyethylene glycol (a microdialysis unit, as describedin Section 2.2, was used) and theproteincon-
.Figure I . (A) IEF of human parotid salivary proteins from a single individual on a pH 4-9 IPG followed by Coomassie Brilliant Blue G-250 staining. The protein loads were: 10, 20,30, 50,60, 100, 125 and 250 kg. (B) The same gel as shown in (A) after silver staining.
Hybrid IEF of human salivary proteins
Electrophoresis 1990, 1 I , 489-494
centration determined. Proteins (25 pg) were loaded on a p H 4-9 I P G in increasing concentrations of NaCl(0-600 mM). This was achieved by adding measured volumes of stock NaCl solutions of concentrations 50 mM, 0.1 M, 0.5 M, and 1 M to fixed volumes of dialyzed saliva. In order to compare the focused patterns of undesalted and desalted lyophilizates, the following experiment was performed: seven I mL aliquots of parotid saliva from one individual were lyphilized and reconstituted with distilled water in order to prepare the following concentrated solutions: x 1 (i.e. unconcentrated), x 2, x 3, x 4, x 5, x 6 and x 7. Half of each sample was desalted by overnight dialysis (in a microdialysis unit) against distilled water. The IEF patterns of undialyzed and desalted samples were compared on adjacent tracks on a pH 4-9 I P G in order of increasing concentration. The sodium, potassium and calcium concentrations ofthe unconcentrated sample ofparotid saliva were determined, using a Corning flame photometer (Corning Ltd., Essex, UK), and were found to be 18 mM, 16 mM and 0.2 mM, respectively.
3 Results 3.1 Effect of application distances on focused pattern The results of analysis of human salivary proteins on hybrid CA-IPGs are shown in Fig. 1. U p to 25-30 well defined protein bands were clearly resolved, the degree of resolution and number depending on the protein load and the stain employed. Optimum results were obtained on a p H 4-9 I P G (data not shown) when the samples were applied 2-3 cm from the cathode, where the local pH is 6.5-7.0. Some degree of precipitation and streaking was observed if application was made at other distances. Application within 1 cm of the cathode resulted in the most severe streaking, probably due to precipitation caused by hydrophobic interactions between proteins and alkaline Immobilines 1221. In p H 3.5-5.0 IPGs, however, it was not possible to achieve good separation when I E F was performed under non-denaturing conditions. There was streaking and most bands failed to focus properly. This problem is not caused by salts in the samples, as even desalted samples failed to produce satisfactory bands, nor was separation improved even by mixing CAs with the samples prior to focusing. In order to obtain optimum results from an I P G of this pH range, it was necessary to perform IEF in 8 M urea(see Section 3.5).
49 1
3.2 Comparison of Coomassie and silver staining A range of protein loads (10-250 pg) was separated on a pH 4-9 I P G and subsequently stained by Coomassie Brilliant Blue G-250 followed by silver stain (Fig. 1). Using the silver stain technique, 20 pg of protein allowed visualization of approximately 25 bands in the gel, while Coomassie Blue staining required 250 pg in order to give similar results. The band patterns, however, were similar using both procedures, and the differences were only at the level of sensitivity rather than additional components being stained by the silver procedure.
3.3 Effect of urea
On a pH 4-9 I P G (Fig. 2) a secondary urea gradient (0-8 M urea) increases the resolution, especially above a urea concentration of 5 M. Most of the proteins focused as continuous bands throughout the length oftheIPG and somecomponents exhibited a slight PI shift towards the anode, a finding which has been reported with other proteins at urea concentrations greater than 2 M [23]. When a similar urea gradient is superimposed on a pH 3.5-5.0 IPG (Fig. 3) all the bands display a continuous shift to the anode as the concentration of urea increases. Below a concentration of 6 M urea the band patterns were poorly resolved, but above this value they were satisfactory.
3.4 Effect of salt levels Upon adding known levels of sodium chloride to samples, it was observed that denaturation occurred at concentrations of 200 mM and above (Fig. 4), the effect being more pronounced on the cathodic side of application points. The most abundant cations in stimulated parotid saliva [9] are sodium and potassium, which have mean concentrations of about 55 mM and I 6 mM, respectively, while the most abundant anions are chloride (33 mM) and phosphate (6 mM). Although there are other ionic species present in saliva, these will have the greatest effect in denaturing proteins as they will form strong acid and alkaline boundaries upon the application of a voltage across the gel [ 191. As such, NaCl has been chosen as the standard salt solution. Denaturation was observed only when the concentration of sodium was 200 mM and above, and this is considerably greater than the level normally found in stimulated parotid saliva. N o differences were observed between desalted and undialyzed lyophilizates which have been concentrated up to 7-fold (Fig. 4).
Figure 2. IEF of a continuous zone of normal human parotid salivary proteins applied 2 cm from the cathode throughout the length of the gel in a p H 4-9 IPG superimposed with a transverse 0-8 M urea gradient. 1 mg of protein was applied and the gel was silver stained.
492
Elenrophoresis 199O,II, 489-494
K . Soo Khoo and J . A. Beeley
Figure 3. IEF of a continuous zone of normal human parotid salivary proteins applied 2 cm from the cathode throughout the length of the gel in a pH 3.5-5.0 IPG superimposed with a transverse0-8 M urea gradient. The protein load was 10.5 mg and the gel was silver stained. Arrowheads indicate bands the resolution of which improved with increasing ureaconcentration.
Figure 4 . (A)IEF on pH 4-9 IPG of human parotid salivary protein samples containing increasing concentrations ofNaC1. The samples containea 25 pgof protein in 50 FL and were from the same individual. Arrowheads indicate precipitation and denaturation. (B) IEF on pH 4-9 IPG ofdesalted and undialyzed lyophilized concentrates of human parotid saliva. One- to 7-fold concentrates were applied to tracks (1) to (7). The tracks marked “U” were loaded with undialyzed samples while tracks marked“D” wereloaded with desalted samples; 25 pg ofprotein were applied on each track and the gel was silver stained.
3.5 Comparison of salivary protein patterns from different individuals Salivary proteins from 6 different subjects separated on a p H 4-9 IPG (Fig. 5) showed few major but several minor variations from individual to individual. The technique appears to give better resolution than that of Eckersall et al. [41 who performed IEF on CA gels and also reported that the band patterns of different individuals were essentially similar with only minor differences. Acidic range IPGs ((pH3.5-5.0) revealed few differences between normal individuals, and indeed very little material focusing in this area, but. numerous additional components were observed in patients with autoimmune diseases such as rheumatoid arthritis and Sjogren’s syndrome. Mairs and Beeley [Sl have previously separated the anionic salivary proteins associated with these disorders on CA-containing gels, but they were not well resolved as they precipitated out at or near their isoelectric points. Urea-containing pH 3.5-5.OIPGs (Fig. 6)overcame the problem ofproteinprecipitation and resolved these proteins into up to 20-25 components (Khoo, K. S., Beeley, J. A. andLaimey,P.-J., manuscript in preparation). By contrast, normal healthy individuals had about 6-10 minor bands focusing in this region.
4 Discussion
In order to further the understanding of the functions of salivary proteins and to develop the potentials of analysis of this fluid in clinical diagnosis, improved analytical electrophoretic procedures are needed. Whilst IEF of human parotid saliva using C A has given improved resolution as compared with other electrophoretic procedures and revealed further information about the nature of proteins present in both normal, healthy individuals [4,241 and in disorders such as connective tissue disease [81, use of the technique has been impaired by the need for prior desalting of samples because of the high electrolyte to protein levels. This procedure is time-consuming and leads to protein loss; furthermore, the sample sizes necessary for manipulation (> 2 mL) are frequently not available from diseased glands. The technique is also subject to problems such as cathodic drift together with the fact that the formation of “tailor-made” gradients necessary for the proteins under investigation can be difficult or impossible.
IPGs overcame most of these problems [251 but did sometimes give rise to streaking and sample precipitation at the application site. Addition of C A to the IPGs to form mixed-bed
Electrophoresis 1990. 1 1 . 489-494
Hybrid IEF 01‘ human salivary proteins
493
CA-IPGs [26, 271 resolved this problem; the procedure described in this paper is based on this methodology. Incorporation of ureain the gel gave improved resolution in pH 4-9 IPGs, and was found to be essential for the pH 3.5-5.0 IPGs; in its absence the components precipitated out and poor results were obtained. Rehydrating the gels with 20 % w/v glycerol was found to be advantageous as it prevented liquid exudation 1281. When liquid exudation or “sweating” occurs, the resulting droplets on the gel surface after prolonged focusing may adversely affect the focused pattern. As 8 M urea solutions have a weak inhibitory effect on liquid exudation, glycerol was also included in the rehydration solution for pH 3.5-5.0 gels. A limitation of IPGs is their inherently low conductivity, which causes difficulties such as slow sample entry, lateral zone spreading and the formation of salt fronts which halt protein migration. A hybrid CA-IPG system was employed to circumvent this problem [29,301.This was achieved by reswelling dried IPGs in CA-containing rehydration solutions.
Figure 5. IEF of parotid saliva from 6 different individuals (a)-(f) on a pH 4-9 IPG. The protein load was 25 Fg per sample and the gel was silver stained.
Figure 6 . IEF of human parotid saliva using pH 3.5-5.0 IPG rehydrated with 8 M urea in 20 % w/v glycerol and 0.5 % w/v pH 3.5-5 Ampholine. Tracks (v). (x) and (z) were loaded with 100 Fg of salivary proteins from healthy individuals, while track (w) was loaded with 100 pg salivary proteins from a patient with Sjogren’s syndrome and track (y) from a patient with rheumatoid arthritis; the gel was silver stained.
Salts present in samples are capable of causing denaturation because acidic and alkaline boundaries are produced at application points. Thus even low levels of salts (5 mM) have been reported to induce modification of haemoglobin patterns and higher levels to cause denaturation [ 191. In order to investigate whether the levels of electrolytes present in saliva would have any adverse effects, a study was performed to determine the salt level which would lead to protein precipitation and denaturation. We have found salivary proteins to be less affected by salts, and satisfactory patterns were obtained from samples containing up to 200 mM NaCI. This is well above the electrolyte levels in parotid saliva. No precipitation was observed when lyophilizates concentrated up to sevenfold without desalting were fractionated on IPGs. The pH 4-9 IPGs of parotid salivary proteins have produced highly resolved focused patterns. With samples containing as little as 20 pg of proteins, 25-30 components were evident on silver staining. Interestingly, IPGs, like CA-based IEF, although revealing some individual-to-individual differences in protein pattern in parotid saliva, they were less extensive than is the case with sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) [3 11. Urea is often used as a solubilizing agent in electrophoresis [321 and IEF in gels containing urea gradients perpendicular to the pH axis 133-351 may yield important clues about protein structure. Alterations in the pls ofproteins will occur if unfolding by urea results in the exposure of charged groups [361. In IPGs of both pH ranges, urea was found to cause shifts in the pls of most protein bands but results in better resolution. It was also necessary to incorporate 8 M urea into p H 3.5-5.0 IPGsinordertopreventprecipitation.UsingpH 3.5-5.OIPGs containing 8 M urea, larger sample loads (100 pg) were necessary to detect the lower levels of protein which focus in this range. Whilst some individual-to-individual variation in band pattern was observed in normal subjects, the abnormal proteins which are associated with connective tissue disorders [81 and focus in this region were resolved into 20 or more additional well-defined components. The I P G technique gives superior resolution to CA-IEF of human salivary proteins 141 in both broad and narrow pH range gels; this is especially the case in pH 3.5-5.0 gels, over which pH range carrier ampholytes produce virtually no
494
K. So0 Khoo and J. A. Beeley
separation. Analysis of salivary proteins by focusing on IPGs also obviates the need to desalt samples prior to analysis and can be used with sample volumes as small as 50 FL. Accordingly, it is now possible to analyze rapidly large numbers of parotid saliva samples (including those from patients with xerostomia) as a function of pl. This development should lead to considerable progress in the analysis ofhuman salivary proteins.
W e uregratefulfor~nunciulassistunce.frorn the Universityof Glusgow MucFeut Bequest. The valuubi'eadvice and technical assistance of M r . D. Sweeney is gratefully acknowledged. Received January 2, 1990
5 References Beeley, J. A., Archs. Oral. Biol. 1969, 14, 559-561. Arneberg, P., Scand. J . Dent. Res. 1972, 80, 134-138. Beeley, J. A., in: Catsimpoolas, N. and Drysdale, J. (Eds.), Biological and Biomedical Applications of Isoelectric Focusing, Plenum Press, London 1977, pp. 1-28. Eckersall,P.D.,Mairs,R. J. andBeeley,J. A.,Archs. Oral.Biol. 1981, 26,727-733. Pronk, J. C. and Frants, R. R., Hum. Hered. 1979,29, 181-186. Bustos, S. E. and Fung, L. in: Allen, R. C. and Arnaud, P. (Eds.), Electrophoresis '81, Walter de Gruyter, B,erlin 1981, pp. 3 17-328. Chisholm, D. M., Beeley, J. A. and Mason, D. K., Oral. Surg. 1973, 36,620-630. Mairs, R. J. and Beeley, J. A., Clin. Sci. 1985, 69, 727-735. Mason, D. K. and Chisholm, D. M., Saliiiary Glands in Health and Disease, W. B. Saunders, Philadelphia 1975, pp. 37-69. Jonsson, M., Electrophoresis 1980,1, 1411-149. Oppenheim, F. G., Hay, D. I. and Franzbdau, C.,J. Biochem. 1971, 10,4233-4238. Bennick, A., J. Biochem. 1977,163,229-239. Azen, E. A., Biochem. Genet. 1978,16, 7'3-99. Azen, E. A., Archs. Oral. Biol. 1978,23, 1173-1176.
Electrophoresis 1990,l I, 489-494
[ l 5 J Azen,E.A.andOppenheim,F.G.,Science 1973,180,1067-1069. [I61 Stephen, K. W. and Spiers, C. F., Br. J. Clin. Pharm. 1976, 3, 3 16-3 19. [ 171 Smith, P. K., Krohn, R. I., Hermanson, G .T., Mallia, A. K., Gartner, F. H., Provenzano, M. D., Fujimoto, E. K., Goeke, N. M., Olsen, B. J. and Klenk, D. C., Anal. Biochem. 1985,150,7645. [I81 Wilcox, P. E., Cohen, E. and Tan, W., J . Biol. Chem. 1947, 228, 999- 1019. I191 Righetti, P. G., Chiari, M. and Gelfi, C., Electrophoresis 1988, 9, 65-73. 1201 Mairs, R. J., Ph. D. Thesis, University of Glasgow, 1981, p. 65. 1211 Damerval, C., le Guilloux, M., Blaisonneau, J. and de Vienne, D., Electrophoresis 1987,8, 158-159. 1221 Rabilloud, T., Gelfi, C., Bossi, M. L. and Righetti, P. G., Electrophoresis 1987,8, 305-312. 1231 Gorg, A., Postel, W. and Johann, P., J. Biochem. Biophys. Methods 1985,10,341-350. 1241 Eckersall, P. D. and Beeley, J . A., Biochem. Genetics 1981, 19, 1055- 1062. 1251 Righetti, P. G., J . Chromatogr. 1984,300, 165-223. 1261 Rimpilainen, M. and Righetti, P. G., Electrophoresis 1985, 6, 4 19-422. L271 Righetti, P. G., J. Biochem. Biophys. Methods 1988, 16,99-108. 1281 Altland, K., Hackler, R. and Rossmann, U., Electrophoresis 1986,7, 25 1-259. I291 Altland, K., von Eckardstein, A., Banzhoff, A., Wagner, M., Rossmann, U., Hackler, R. and Becker, P., Electrophoresis 1987,8, 52-62. [30] Fawcett,J. S. andChrambach,A.,Electrophoresis 1986,7,266-272. [311 Lindsay, J. C. B. and Beeley, J. A. in: Dunn, M. J. (Ed.), Electrophoresis '86, VCH Publishers, Weinheim 1986,pp. 222-225. 1321 Andrews, A. T., Electrophoresis: Theory, Techniques and Biochemical and Clinical Applications,Oxford University Press, Oxford 1986, p. 252. 1331 Hobart,M. J.,in:Arbuthnott,J.P. andBeeley,J.A.(Eds.),Zsoelectric Focusing, Butterworths, London 1975, pp. 275-280. 1341 Altland, K., Roeder,T., Jakin, H. M., Zimmer, H. G. andNeuhoff,V., Clin. Chem. 1982,28,4, 1000-1010. 1351 Altland, K., Banzhoff, A., Hackler, R. and Rossmann, U., Electrophoresis 1984,5, 379-381. [361 Creighton, T. E., J. Mol. Biol. 1979,129,235-264.
