Basic A1 Protein of the Myelin Membrane

Twenty-seven tryptic peptides were isolated from the Al protein from bovine spinal cord. These, together with 16 peptic peptides, were utilized to establish the complete amino acid sequence of the 170 residues of the Al protein. Peptide T, derived from the Al protein by cleavage of the carboxyl of the single tryptophan residue, was useful in positioning the peptides which comprise the COOH-terminal end. Seven chymotryptic and nine tryptic peptides were isolated from Peptide T. An unusual feature of the sequence is the methylated arginine residue at position 107 which is present as both the dimethyl and monomethyl derivatives. The methylated derivatives appear to be relatively resistant to tryptic hydrolysis. Located close to the methylated arginine residue is a Pro-Arg-Thr-Pro-Pro-Pro sequence, a structure which may induce a sharp bend in the molecule, suggesting a conformation more compatible with an open double chain structure than a random coil. The over-all sequence reveals no obvious periodicity but rather a general distribution of basic residues over the polypeptide chain, making the interaction with phospholipids within the myelin matrix highly probable. Several peptide segments of 9 residues exist, in which basic residues are missing; the somewhat nonpolar character of these regions suggests a possiblity for participation in nonpolar interactions.


SUMMARY
Twenty-seven tryptic peptides were isolated from the Al protein from bovine spinal cord.
These, together with 16 peptic peptides, were utilized to establish the complete amino acid sequence of the 170 residues of the Al protein.
Peptide T, derived from the Al protein by cleavage of the carboxyl of the single tryptophan residue, was useful in positioning the peptides which comprise the COOH-terminal end.
Seven chymotryptic and nine tryptic peptides were isolated from Peptide T.
An unusual feature of the sequence is the methylated arginine residue at position 107 which is present as both the dimethyl and monomethyl derivatives.
The methylated derivatives appear to be relatively resistant to tryptic hydrolysis. Located close to the methylated arginine residue is a Pro-Arg-Thr-Pro-Pro-Pro sequence, a structure which may induce a sharp bend in the molecule, suggesting a conformation more compatible with an open double chain structure than a random coil.
The over-all sequence reveals no obvious periodicity but rather a general distribution of basic residues over the polypeptide chain, making the interaction with phospholipids within the myelin matrix highly probable.
Several peptide segments of 9 residues exist, in which basic residues are missing; the somewhat nonpolar character of these regions suggests a possiblity for participation in nonpolar interactions.
The basic Al protein' is a major structural proteinof thecentral nervous system myelin in which it constitutes 30% of the total protein (see Reference 1 for a review).
The Al protein has been isolated (24) from human and bovine brain and bovine spinal cord in homogeneous form as shown by gel electrophoresis, ultracentrifugation, and immunoelectrophoresis. Physicochemical *This work was supported by United States Public Health Service Grant lRO1 NB 08268-02. Send renrint reauests to The Merck Institute, Rahway, New Jersey 07665. A $ Present address, School of Dental Medicine, Harvard Medical School, Boston, Massachusetts 02115.
1 During the past 3 years, we have referred to the basic protein from centrai nervous system myelin as the Al protein in order to distinguish it from other basic proteins of the nervous system. The general terminology of "basic protein" or "basic myelin protein" appears too general and has led to confusion. studies revealed it to be a highly basic protein, behaving as a random coil in solution (5-7).
We have sought (S-11) to correlate the chemical structure of the Al protein with its biological properties, most notably the induction of the autoimmune disease, experimental allergic encephalomyelitis (see Reference 12 for a review).
The disease-inducing site of the protein (13, 14) appears to be a function of the primary structure defined by a unique short peptide segment of 9 residues, which, alone or as part of the Al protein, induces the identical chemical and histological signs associated with EAE2 in guinea pigs. We report here the complete amino acid sequence of the 170 residues of the Al protein from bovine spinal cord as established primarily with tryptic and peptic peptides. ' Preliminary reports on the sequence of both the bovine (15) and human ii1 proteins have appeared (15,16).
Although proteins are known which are either specific to the nervous system (17) or possible structural proteins of membranes (18), the Al protein is the first protein in these categories to have its sequence defined. Such information should aid our understanding of the normal role of this protein in the myelin substructure and possibly in the process of myelinogenesis.
In addition to its immunological relevance, knowledge of the sequence will be useful for comparative studies with Al proteins from other species, from peripheral nerve, and with other basic proteins from nervous tissue such as histones.
The sequence has already proven invaluable in elucidating the regions of the Al protein responsible for inducing disease in guinea pigs (8,13) and rabbits (19,20), and thereby providing the model for peptide synthesis (14) which has led to definition of the essential residues: the three amino acids tryptophan, glutamine, and lysine. The sequence has also proven helpful in elucidating the site of glycosylation (21) by the polypeptide N-acetylgalactosaminyl transferase, thus defining the receptor sequence recognized by this enzyme (22), and suggesting a model for the sequence in mucins.
Issue of September 25, 1971 E. H. Eylar, S. Brostof, G. Hashim, J. Caccam, and P. Burnett phoresis at pH 10.5 was used, minor microheterogeneity was observed; 85 to 90% of the protein migrated in a single leading band toward the cathode followed by two faint bands.
In all cases hydrolysis with carboxypeptidase was carried out at 37" in 0.2 1\f triethylamine-bicarbonate buffer, pH 8.1, using a 5O:l ratio of peptide to enzyme (carboxypeptidases A and B together) unless otherwise noted (11). Hydrolysis with aminopeptidase M was usually carried out under the same conditions.
The hydrazinolysis procedure was used as described earlier (8, 10) to identify the COOH-terminal residue of some peptides. High voltage electrophoresis was performed at pH 4.7, 3500 volts, in buffer containing 2.5% pyridine and 2.5% acetic acid, unless otherwise noted.
Ascending paper chromatography was also used for preparation and assessing purity of peptides. The system contained butanol-acetic acid-pyridine-water (122 : 38: 189: 151). The monomethyl-and dimethylarginine derivatives were identified as described previously (23).
The direct Edman procedure used was that in which the phenylthiohydantoin-amino acids were identified by thin layer chromatography as described by Blomback et al. (24). For the subtractive Edman procedure, the amino acid analysis of the peptide residue was performed after removal of the phenylthiohydantoin-amino acid. In some cases, the dansylation procedure was used to identify the NHz-terminal residue (25). All amino acid analyses were performed as previously described (3)(4)(5).
The peptide mapping of the tryptic peptides from the bovine Al prot,ein was performed as described previously (4,9). For Peptide T, derived from the Al protein by cleavage of the COOHtryptophanyl linkage with N-bromosuccinimide (26) or BNPSskatole (27), the tryptic and chymotryptic peptides were similarly mapped.
Nomenclature-Tryptic peptides are designated by T, and peptic peptides by P. Tryptic and chymotryptic peptides derived from Peptide T are designated TT and TC, respectively. All pept,ides are numbered beginning with 1 for the NH*-terminal peptide and progressing to the COOH-terminal peptide. Since the complete amino acid sequence has been determined, each residue has been numbered in accord with the sequence of 170 residues.
Reference to Fig. 6 will aid in following the description of each peptide.
Isolation of Peptides-Peptides were isolated from the peptic and tryptic digests of the bovine Al protein by (a) resolution into fractions by ion exchange chromatography on Cellex-P or Dowex 50 resin or, for a few tryptic peptides, directly from a peptide map of the Al protein or Peptide T; (b) resolution of the fractions into subfractions by gel filtration on Sephadex G-25, G-50, or G-75; and (c) isolation of peptides from the subfractions by preparative paper chromatography or high voltage electrophoresis or both. With these procedures, we were able to isolate most of the peptides, accounted for by the peptide mapping, in sufficient quantities for defining the sequence of the Al protein.
In no case were peptides found which did not fit the proposed sequence. Each peptide was judged pure when it gave a single band ou both high voltage electrophoresis at pH 4.6 and paper chromatography, and gave amino acid residues in near integral ratios on amino acid analysis.
If contaminating peptides were found, then preparative peptide mapping was used with 1 to 3 mg of material.
This technique was adequate to purify all of the tryptic and peptic peptides, although, in some cases, electrophoresis for 4 to 8 hours was needed to resolve peptides of similar charge. The yield of each peptide was estimated either from dry weight measurement or amino acid analysis.
Recovery varied from 50 to 85% from paper electrophoresis and 30 to 70% from paper chromatography.
It was observed that the cation exchange columns adsorbed a considerable portion of the peptides, particularly the very basic peptic peptides.
Tryptic Hydrolysis of Bovine Al Protein and PuriZcation of Tryptic Peptides-The digestion of 3.0 g of bovine Al protein (170 Bmoles) was carried out for 4 hours at 37" with 90 mg of trypsin (treated with L-(tosylamido 2-phenyl)ethyl chloromethyl ketone) in 250 ml of 0.2 M triethylamine buffer, pH 8.1. After the solution was lyophilized, the peptide mixture was dissolved in 300 ml of Hz0 and applied to a column of Dowex AG-50X2 which had been equilibrated with 0.03 M ammonium acetate, pH 3.7. The peptide elution pattern is shown in Fig. 1. A linear gradient in pH and ionic strength was established at tube 147; at tube 900 the pH was 7.75, at which point 0.3 M NH40H was applied directly to the column.
Just prior to elution of Peak I, the pH was 8.1; after elution, the pH was 10.4. The elution position of each tryptic peptide is designated in Fig. 1 Each fraction was desalted and further fractionated by gel filtration on a Sephadex G-10-G-25 column (92 x 2.5 cm) in 0.1 N acetic acid (Fig. 2). Resolution into several subfractions, based generally on peptide size, was achieved in every case. Each subfraction was then subjected to paper chromatography or electrophoresis or both.
For chromatography, approximately 5772 Basic Al Protein of Myelin Vol. 246,No. 18 FIG. 2. The gel filtration elution pattern is shown for tryptic peptide fractions C, E, and G (see Fig. 1) applied to Sephadex G-10 upper layer (45 cm)-G-25 lower layer (47 cm), 92 X 2.5 cm in 0.05 M acetic acid. The absorbance at 235 rnp was used as a measure of peptide content; 8 ml per tube were collected.
I, Fraction C eluate was divided into three regions, Cl, C2, and C3; ZZ, Fraction E was divided into two regions, El and E2; and ZZZ. Fraction G was divided into two regions, Gl and G2. ORIGIN+ TII n 0 T6 T22 FIQ. 3. The peptide map of the bovine Al protein was obtained after digestion with trypsin (5O:l ratio of protein to enzyme) for 4 hours at 37" in 0.2 M triethylamine bicarbonate buffer, pH 8.1. Electrophoresis was carried out for 2.5 hours at 3500 volts, pH 4.7, using 2.5y0 pyridine, 2.5% acetic acid, and 5% butanol; ascending chromatography was carried out for 18 hours in butanolacetic acid-pyridine-water (122:38:189:151).
0.5 mg of peptide per inch was applied; up to 2 mg per inch were applied for electrophoresis. Preparative peptide mapping of the tryptic peptides from the bovine Al protein was carried out following a 2-hour incubation with trypsin as described above. The peptide mixture (5 to 10 mg) was applied to Whatman No. 3MM paper for electrophoresis at 3500 volts, 150 ma, in pyridine-acetate buffer, pH 4.7, for 3 hours. The paper was dried at 25"; chromatography (ascending) was then performed for 12 to 16 hours. After drying, the paper was dipped in ninhydrin solution (0.25% in acetone) and developed at 25" overnight.
Prior to cutting and elution of the spots from the paper with 5'% ammonia, the paper was washed in acetone to remove excess ninhydrin.
The peptide map of the bovine Al protein is shown in Fig. 3. The above procedure was very reproducible; 15 to 70% of the theoretical yield for each peptide could be recovered.
Most of the peptides were cleanly separated by this procedure as shown in Fig. 3 where each peptide has been identified.
Peptide Tl was obtained from Peptide CBl which was derived from the Al protein after cleavage of the COOH-methionyl bond with CNBr, as described previously (11).
Peptide T9 was also obtained from Peptide CBl (11). Although Peptide T2 generally overlapped with Peptide T3, in the usual peptide map they were cleanly resolved with a chromatographic system lacking pyridine (11).
Peptide TS was the major component of Fraction D, and was easily purified by electrophoresis (75% yield). Peptide T4 was prepared from Fraction I. One of the largest tryptic peptides, a dodecapeptide, it was one of the last to elute from the Dowex column in tubes 942 to 952. As shown in Fig. 3, it could be separated from Peptides T8, T13, and T24 by electrophoresis.
It was subsequently prepared in homogeneous form by paper chromatography which clearly separated it from traces of Peptide T13 and a small amount of faster moving peptide referred to as Peptide T6A.
On peptide mapping, it sometimes partially overlapped Peptide T19.
Peptide T6 was not found in the eluate from the Dowex 50 column.
It was prepared from the peptide map where it migrates rapidly because of charge +2, but exhibits a very high RF on chromatography (Fig. 3). Peptide T6 was not found in the elute from Dowex 50. It was easy to prepare, however, from the peptide map because it migrated faster on electrophoresis than any other peptide. Peptide T6A was found in Fraction I; it was separated from Peptide T4 by paper chromatography as described under "Peptide T4." This peptide was not detected on the peptide map. Peptide T7 was one of the first peptides to elute from the Dowex 50 column (Fraction B). It was also one of the most acidic of the peptides with charge -1 and was separated cleanly by peptide mapping as well. This peptide was purified from traces of Peptide T15 by paper electrophoresis.
Peptide T8 was prepared from Fraction I by paper electrophoresis which separated it from Peptides T12, T13, T17, and T24 (Fig. 3). It is the least basic of this group of peptides with charge 0, and as such migrates much more slowly than Peptides T17, T12, T13, and T24, the latter having the highest mobility of this group.
Peptide T9 was located in Fraction C from the Dowex 50 column, and was separated from other peptides on gel filtration (Fig. 21). It eluted mainly in the middle region (C2) which contained 62% Peptide T9, 30% T20, and 8% T15. On electrophoresis it migrated much faster than either Peptide T15 or T20 (the slowest).
It is of interest that Peptide T9, a tetrapeptide, should elute earlier from the gel filtration than the Pentapeptide T20, which constituted 88% of the third region (C3). Peptide T15, an octapeptide, clearly eluted first and constituted 950/, of the first region (Cl).
Peptide TlO was derived from Fraction E of the Dowex 50 eluate. Resolution on gel filtration (Fig. 2111) showed Peptide TlO primarily in Region E2. When Fraction E2 was subjected to electrophoresis, this peptide separated cleanly from Peptide T23 (much slower) and Peptide T3 (slower).
Peptide Tll was obtained from Fraction G following gel filtration, as shown in Fig. 2111. Only Region Gl showed a significant yield of peptide (94% of material applied).
Peptide Tll constituted 43% of the Gl region, and was purified by both electrophoresis and chromatography. On electrophoresis it migrated together with Peptide T22, just slower than Peptide T21 from which it was separated.
Peptide Tll was subsequently separated by paper chromatography from Peptide T22. By this procedure, Peptide Tll appeared homogeneous when examined by peptide mapping. 25,1971 E. H. Eylar,S. Brostof,G. Hashim,J. Caccam,and,P. Burnett 5773 Peptide Tld was found in Fraction I (tubes 953 to 962) in a group which included Peptides T8, T13, T18, and T24 as described under "Peptide T8."

Issue of September
When this mixture was subjected to electrophoresis, Peptides T12 and T13 migrated together slightly ahead of Peptide T18 but slower than Peptide T24 (see Fig. 3). Using paper chromatography, Peptides T12 and T13 were separated; a yield of 70% was obtained.
Peptide T12 appeared homogeneous by peptide mapping.
Peptide TiS was found in Fraction I (tubes 953 to 962), and was isolated as described under "Peptide T12." It was also found in Fraction I (tubes 942 to 952) and separated by electrophoresis from Peptide T4 as shown in Fig. 3.
Peptide Ti4 was located in Fraction F. It appeared homogeneous by peptide mapping and thus required no further purification.
Peptide T15 was located in Fraction C. On gel filtration, Fig.  21, it was the dominant peptide in the Region Cl.
It constituted 94% of the Cl fraction and was resolved well from Peptide T9 on electrophoresis.
Peptide T16, a dipeptide, was one of the last peptides to elute from the Dowex 50 column in Fraction I (tubes 963 to 972). It was separated from other peptides by 'paper chromatography and appeared homogeneous.
It was not found on peptide mapping (Fig. 3).
Peptide T16A was not found in the Dowex 50 eluate, presumably because of the presence of 2 arginine residues, but was obtained in homogeneous form by preparative peptide mapping (Fig. 3).
Peptide T17 was eluted in low yield in Fraction I (tubes 963 to 972) and purified on paper chromatography.
It also resolved well on peptide mapping, as shown in Fig. 3.
Peptide T18 was located in Fraction I (tubes 952 to 962). On electrophoresis it migrated just slightly faster than Peptide T8. Although it has zero net charge at pH 7, apparently at pH 4.7 the glutamic acid residue is partially titrated, giving the peptide a slightly basic character, barely more than Peptide T8 which contains aspartic acid and arginine.
Peptide TlQ was not isolated from the Dowex 50 column. It is a large peptide (18 residues) containing 2 basic residues.
On peptide mapping, where it was isolated for structural studies, it separated well from most peptides, but it was necessary to chromatograph for over 24 hours in order to resolve it from Peptide T4 which moves slightly slower.
Peptide TZO was located in Fraction C. On gel filtration it partially adsorbed to the Sephadex material and thus eluted in t.he last region, C3, where it constituted 92 To of this fraction.
It was prepared in homogeneous form from both Regions C2 and C3 by electrophoresis (Fig. 21).
Peptide T9i was found in Fraction G of the eluate from the Dower; 50 column.
This peptide comprised 36% of Fraction G (see Fig. 2111) and together with Peptides Tll and T22 constituted the bulk of this fraction.
It was easily purified from Region Gl by electrophoresis because it migrated more rapidly than the other component peptides by virtue of a charge of +2.
Peptide T%f?, which constituted 21% of Fraction Gl (Fig.  2ZZZ), was isolated following both electrophoresis and chromatography as described under "Peptide Tll." Peptide T.93, a decapeptide, was one of the main components of Fraction E (Fig. 2ZZ), and was located mainly in Region El on gel filtration.
It was isolated following paper chromatography. Peptide Tg4 was spread over the whole of Fraction I and was easily purified by electrophoresis because it migrated more rapidly than any other peptide of this fraction.
Peptide TZ% was not found in the Dowex 50 eluate but was resolved by peptide mapping (Figs. 3 and 4).
Peptide T96 was likewise not found in the Dowex 50 eluate, but was separated clearly from all other peptides by peptide mapping (Fig. 3). Since this peptide has charge 0, it should have eluted from the Dowex 50 column, but was possibly missed prior to or in Fraction A.
Peptide T.97 was located by peptide mapping (Fig. 3) but was not found in the Dowex 50 eluate.
If the Arg-Arg linkage, which is relatively resistant to trypsin (28), had not been cleaved, then this peptide may have contained both arginine residues and would be strongly adsorbed to the Dowex 50. However, the material from the peptide map shows only 1 arginine residue.
The amino acid compositions of the tryptic peptides, other than those previously reported (11,20), are given in Table I. Although some peptides overlap, because of the partial resistance of some linkages to trypsin attack, i.e. methylated arginine 107 (23), the combined compositions account for all of the amino acid residues in the Al protein except for arginine residues 53 and 170 which are liberated as free arginine.
Peptide Tl, a neutral peptide because of the acetylated NH2 terminus (ll), was probably not detected with the ninhydrin procedure; other peptides, however, such as Peptide T2, are relatively basic, having a charge +2, and may not have eluted.
It appeared generally that arginine-containing peptides were bound more strongly than lysine-containing peptides. Peptide T19 is relatively large (18 residues) and hydrophobic and was probably strongly bound to the Dowex beads. It is fortunate, however, that under the conditions used the Lys-Pro linkage was partially split, giving Peptide TM, which was obtained in a 9% yield.
In the usual peptide mapping procedure, therefore, Peptide T18 is not observed; only Peptide T19.
For most of t.he tryptic peptides, a yield of 15 to 70% of theoretical was obtained, a value high considering the losses on paper electrophoresis or chromatography (Table I). In some cases, however, the yield was lower, either due to adsorption on the Dowex 50 column or incomplete hydrolysis as in the case of Peptide T18 and Peptides TlO and Tll where the Lys-Asp linkage is partially resistant to hydrolysis by trypsin (29).

SEQUENCE STUDIES ON TRYPTIC PEPTIDES
Peptide Ti (Residues 1 to 5): Acetyl-Ala-Ser-Ala-Gln-Lys-The sequence of this ninhydrin-negative peptide was established as described earlier (11) using the peptide fragment CBl cleaved at methionine residue 20 with CNBr. Peptide T2 (Residues 6 to 10): Arg-PreSer-Gln-Arg-The sequence of this peptide was determined previously (11) as part of Peptide CBl.
Peptide TS (Residues 11 and id): Ser-Lys-The sequence is obvious from specificity of trypsin.
Peptide T4 (Residues IS to 24): Tyr-Leu-Ala-Ser-Ala-Ser-Thr-Met-AspHis-Ala-Arg-The sequence of the first 8 residues was determined previously (11) as part of Peptide CBl. The remainder of the sequence was determined in three ways. Direct Edman degradation of Peptide T4 confirmed the sequence from tyrosine to methionine as well as aspartic acid as the 9th residue. Direct Edman degradation of Peptide CB2 (from CNBr treatment) gave Asp( )-Ala, thus showing aspartic acid and alanine to occupy the first and third positions from the methionine residue.
It had been established earlier (11) that Peptide CB2 was linked to Peptide CBl through the methionyl residues. Finally, Peptide T4, when treated with carboxypeptidase for 30 min, gave arginine (1.08), alanine (0.55), and histidine (0.17). This information thus establishes the sequence of the COOH-terminal region as Met-Asp-His-Ala-Arg.
Peptide T5 (Residues 25 to SO): His-Gly-Phe-Leu-Pro-Arg-The sequence of this peptide was determined with the subtractive Edman method.
Peptide T6 (Residues Si fo $2): His-Arg-The sequence is established by the specificity of trypsin.
It is apparent from these data that the Arg-Asp linkage between Peptides T6 and T7 was not split.
Such linkages are hydrolyzed more slowly by trypsin because of the car-boxy1 group (29).
Use of the subtractive Edman technique gave the Asp-Thr-Gly-Ile-sequence for the NH2 terminal region and thus confirms isoleucine at the fourth position.
Step 1: Asp, 427,; Step 2: Thr, 78%; Step 3: Gly, 37%; Step 4: Ile, 57%. Carboxypeptidase action revealed the sequence of the COOH-terminal region as -Leu-Gly-Arg. These data are sufficient to establish the sequence. The nonamidated form of the aspartic residues was also suggested from the electrophoretic mobility which placed it with those peptides of charge -1 (Fig. 3).
Peptide T8 (Residues 48 to 48): Phe-Phe-Gly-Ser-Asp-Arg-Four steps of the direct Edman procedure established the sequence of this peptide.
The nonamidated form of the aspartic acid was inferred from the mobility on electrophoresis (Fig. 3). Peptide T9 (Residues 4.9 to 52): Gly-Ala-Pro-Lys-The sequence was reported previously as part of peptic Peptide R (20.) Peptide TIO (Residues 54 to 57): Gly-Ser-Gly-Lys-The sequence of this peptide was also previously reported as part of peptic Peptide R (20).
Peptide Tll (Residues 58 to 64): Asp-Gly-His-His-Ala-Ala-Arg-The sequence of this peptide was also previously reported as part of peptic Peptide R (20).
Peptide TiS (Residues 75 to 91): Ala-Gln-Gly-His-Arg-Pro-Gln-Asp-Glu-Asn-Pro-Val-Val-His-Phe-Phe-Lys-The determination of the sequence of this peptide was reported previously as part of Peptide R (20) except for the COOH-terminal Phe-Lys segment.
Carboxypeptidase treatment for 5 min gave relative values of Lys, 1.0; Phe, 1.9; His, 0.4; and Val, 0.3. Thus the COOH-terminal region appears to be Phe-Phe-Lys. Peptide T14 (Residues 92 to 97): Asn-Ile-Val-Thr-Pro-Arg-The direct Edman procedure gave Asn-Ile-Val-Thr-for the first 4 residues, thus establishing the sequence of this peptide. The sequence was confirmed by the subtractive Edman procedure.
It was not determined whether the basic residue was arginine of the monomethylarginine derivative.
It has been established (23) that position 2 is occupied by methylated arginine residues (mono-and dimethylarginine derivatives). It appears likely that the resistance of the methylated arginine site to trypsin digestion is due to the methylation, since, under the conditions used, even a highly resistant Lys-Pro linkage was partially cleaved producing Peptide T18.
Peptide Ti8 (Residues ii4 to ids): Phe-Ser-Trp-Gly-Ala-Glu-Gly-G&&s-The direct Edman procedure gave Phe-Ser-Trp-Gly-Ala-Glu-Gly-for the first 7 residues and clearly establishes the sequence. Carboxypeptidase treatment for 3 hours released only 2 residues, lysine, 0.7, and glutamine, 0.65. Aminopeptidase M treatment for 10 hours gave the precise analysis as found by acid hydrolysis (Table I) except that 1 glutamic residue was amidated.
The remainder of the sequence was determined by using peptides obtained after 2-hour digestion with chymotrypsin.
The sequence of Peptide T19-3 was shown by carboxypeptidase which released relatively 1.0 mole of Arg and 0.2 mole of Gly in 30 min.
Peptide T23 (Residues 143 to 152): Gly-Nis-Asp-Ala-Gln-Gly-Thr-Leu-Ser-Lys-This decapeptide was subjected to eight steps of the direct Edman procedure which established the sequence of the first 8 residues.
Aminopeptidase M was not helpful because it released equal amounts of isoleucine and phenylalanine.
Peptide T25 (Residues 156 to 159): Leu-Gly-Gly-Arg-The direct Edman technique gave Leu-Gly at the NH2 terminus which establishes the sequence shown.
The indirect Edman confirmed the NHz-terminal Leu with a 60% reduction in leucine after one step.
Peptide T26 (Residues 160 to 162): Aspfier-Arg-The NH2 terminus of this tripeptide is aspartic acid as shown by the direct Edman procedure.
The indirect Edman also established Asp at the NH*-terminal end: Step 1: Asp, 74%. Peptide T27 (Residues I63 to 169): Ser-Gly-Ser-Pro-Met-Ala-Arg--Five steps of the direct Edman procedure established the Ser-Gly-Ser-Pro-Met sequence and thus the sequence of the peptide.
Hydrolysis with Pepsin and Pzcri$cation of Peptic Peptides-The bovine Al protein (3.6 g) was treated with pepsin at 37" for 75 min as described elsewhere (10) and lyophilized.
The digest was taken up in 0.001 M ammonium acetate, pH 7.0, and applied to a column of Cellex-P, as shown in Fig. 1 of Reference 20. The elution profile of the peptides was monitored by absorbance measurement at 235 nm. High voltage electrophoresis of selected tubes from each peak was also performed.
The elution was carried out at 4" with 0.1 M acetic acid; 8 to 10 ml per tube were collected.
The elution pattern was followed by measuring the absorbance at 235 and 280 nm. The purity and identity of the peptides were evaluated by high voltage electrophoresis at pH 4.7 and by paper chromatography.
Appropriate tubes containing subfractions of peptides were combined and lyophilized.
Desalted mixtures containing two or more peptides were further purified either by preparativepaperchromatographyorhighvoltageelectrophoresis.
Peptide PI was derived from Fraction PM, one of the last peaks to elute from the Cellex-P column.
On gel filtration, most of the material of this fraction eluted in a single peak. On preparative paper chromatography, one major peptide band was found with traces of three other peptides: 39 mg were eluted. This material gave one band on high voltage electrophoresis. Thus Fraction PM is composed primarily of one peptide component.
Peptide P2 was also found in one of the last fractions to elute from the Cellex-P column, Fraction PN, and thus is one of the most basic peptides.
On gel filtration, one major peak was observed, which on paper electrophoresis resolved into one main peptide, which migrated more rapidly than three minor bands. The major peptide was purified by preparative electrophoresis: 7 mg were obtained.
This material appeared homogeneous. Peptide PS was found in Fraction PC, one of the first peaks eluted from Cellex-P.
One major peak was observed on gel filtration which resolved into six peptides on paper electrophoresis. This neutral pept,ide (9.3 mg) appeared homogeneous on paper chromatography.
Peptide PSA was present in the first Cellex-P fraction PA; on gel filtration, it was found in the second of two main peaks. On paper electrophoresisit separated clearly from six other peptides, moving as an acidic peptide toward the anode. The peptide (27 mg) was eluted from the paper in 41% theoretical yield.
Peptide P.4 was derived from Cellex-P Fraction PG. After gel filtration the material eluting just prior to the single main peak was subjected to paper chromatography; six bands were Issue of September 25, 1971 E. H. Eylar, S. Brostoff, G. Hashim, J. Caccam, and P. Burnett found. The fastest moving peptide was eluted in low yield (3.3 w).
Peptide P5 was located in Cellex-P Fraction PO, the last fraction to elute from the column.
Peptide P5 (referred to elsewhere at Peptide R) was isolated in relatively high yield (45%) from this fraction as described elsewhere (20).
Peptide P6 was found in Cellex-P Fraction PH. On gel filtration, this peptide eluted in the first of two peaks. From paper electrophoresis, 39 mg of Peptide P6 were obtained; only a trace of another peptide was seen. This peptide, therefore, comprises 95% of the first peak found on gel filtration.
Peptide P7 was found in Cellex-P Fraction PJ. On gel filtration, this peptide (8 mg) was found in the shoulder prior to the single main peak. It migrated slowly as a homogeneous band on electrophoresis.
Peptide P8 was obtained by Cellex-P chromatography in phosphate-saline buffer as described earlier (10). It appeared homogeneous from the Cellex-P column.
Peptide P9 was derived from Cellex-P Fraction PE. On gel filtration, this peptide eluted in the major front peak in homogeneous form with yield of 37 mg, 30% theoretical.
A minor second peak was also observed.
Peptide PlO was obtained primarily from Fraction PD of the Cellex-P eluate.
Gel filtration of Fraction PD revealed one main peak which contained several peptides including traces of Peptides P8 and P9. Peptide PlO was obtained in homogeneous form by paper chromatography.
The yield was 37 mg. Peptide PiOA was obtained from Cellex-P Fraction PK following gel filtration.
A single major peak II-as obtained, which on paper electrophoresis separated into two peptides which were recovered in equal amounts (8.5 mg each). Peptide PlOA migrated slightly faster than the second band, later identified as a combination of Peptides P9 and PlO. Peptide Pi1 was found in Cellex-P Fraction PG and was puri-fied by gel filtration followed by paper elcctrophoresis. This peptide was identical with Pept.ide E, the encephalitogenic peptide described previously (8, 10).
Peptide PllA was found in Cellex-P Fraction PH. On gel filtration, this peptide was found in the second peak as the major component along with three trace contaminants whi.ch were removed by paper electrophoresis; 14 mg were obtained. Peptide P19 was found in Cellex-P Fraction PD along with Peptides P9 and PlO. On gel filtration, Peptide P12 was the last peptide to elute from the column.
All peptides containing tryptophan, Peptides Pll, PllA, P12, P12A, appeared to be retarded on gel filtration, and thus eluted later than smaller peptides. Peptide P12 was purified by paper electrophoresis; 19 mg were obtained.
Peptide PldA was located primarily in Cellex-P Fraction PF. On gel filtration this peptide eluted as a small trailing band after the large single peak. On paper chromatography, five bands were observed but only the slowest, Peptide P12A, was recovered in adequate quantity (4.8 mg) for study.
Peptide PlS was found in Cellex-P Fraction PC, and eluted on gel filtration in the single main peak. It was resolved on paper electrophoresis into five peptides, Peptide P13 migrating slightly toward the cathode and Peptide P3 migrating slightly toward the anode. Approximately 5.2 mg were recovered.
Peptide PI4 was found in Cellex-P Fraction PO along with the large Peptide P5. This peptide was easily isolated following paper chromatography, as shown in Fig. 2 of Reference 20, where it migrated more rapidly than the other two peptide components. Based on the weight of the peptide eluted from the paper, the yield was 18% of theoretical.
The amino acid compositions of the peptic peptides are shown in Table II. It should be noted that Peptides PI0 and PlOA are identical except for the replacement in the latter peptide of 1 arginine residue by the dimethylarginine derivative. It is evi-    Vol. 246,No. 18 dent that the methyl&ion of the single arginine residue accounts for t,he differences in chromatographic properties of these peptides. Because of the broad specificity of pepsin activity, other peptides were observed but were not studied because of low yield or the presence of impurities.
The blocked NH2 terminus and the sequence of the COOH-terminal region identify this peptide as the NH2 terminus of the Al protein.
It is also found as part of Peptide CBl, the sequence of which has been reported (11).
The sequence of Peptide P2 was established as shown by the data in Table III. Peptide P2-CN2 was derived from the NHzterminal region since it contained homoserine, and was shown by Edman degradation to have the Leu-Ala-Ser-Ala-Ser-Thr-Met sequence previously demonstrated (11). Direct Edman degradation of Peptide P2-CN2 positions the tryptic peptides derived from this peptide.
Hydrolysis with carboxypeptidase A for 1 hour released phenylalanine (1 mole) only. When carboxypeptidase B was then added, arginine (1.0 mole) and glycine (0.5 mole) were released in 10 min.
The COOH-terminal leucine was found on hydrazinolysis.
Peptide P6  Peptide PI0 (Residues 96 to III): Pro-Arg-Thr-Pro-Pro-Pro-Ser-Gln-Gly-Lys-Gly-Arg-Gly-Leu-Ser-Leu-The sequence of this peptide was determined from the dat,a shown in Table IV. Hydrolysis with carboxypeptidase released nearly 1 mole of leutine, and some serine. The Edman degradation gave the sequence of the first 11 residues.
The additional data needed to to determine the entire sequence were obtained from tryptic peptides (5 mg of peptide, 5 hours, 37"). The tryptic digest was resolved by paper electrophoresis into five peptides.
The Pro-Arg peptide is derived from the NHz-terminus.
Peptide PlO-T2 is positioned from the Edman degradation results, which also positions the Gly-Arg dipeptide.
The tetrapeptide, PlO-T4, occupies the COOH terminus since it is the only peptide containing leucine.
These data were sufficient to establish the complete sequence of Peptide PlO.
Peptide PlO-T5 is of interest because it contains the uncleaved Arg-Gly linkage which, under normal conditions, would be hydrolyzed by trypsin to give Peptides PlO-T3 and 1'10.T4.
Based on the recoveries of these peptides, it was found that only 60% of the Arg-Gly linkage had been hydrolyzed, thus showing a relatively high resistance to trypsin.
Since it is this site, arginine residue 107, at which methylation occurs (23) it is likely that this residue exists as the monomethyl derivative in Peptide PlO, a modification which could render it partially resistant to trypsin. Direct evidence for monomethylarginine in this peptide was not obtained since it cannot be distinguished from arginine by the short basic column of the Beckman AutoAnalyzer; no dimethylarginine was observed.
Peptide PlOA has the sequence identical with Peptide PlO, except that 1 arginine residue is present entirely as the dimethyl derivative.
When treated wit,h trypsin, as described for Peptide PlO, the peptide analogous to Peptide PIO-T5 was found, but Peptides PlO-T3 and PlO-T4 were not found.
These data show that the arginine residue 107 is present in this peptide as the dimethyl derivative, and is highly resistant to hydrolysis by trypsin.
Its sequence was originally reported to contain 2 additional residues (8), and since been corrected (13). The data shown in Table V establish the sequence. Ten steps of the direct Edman procedure give the sequence through glutamine.
At the 11th step, however, glycine predominated over lysine and an unequivocal choice could not be made. The COOH-terminal residue was identified as Phe by hydrazinolysis.
The COOH-terminal portion of Peptide Pll was established by the isolation of Peptide Pll-Tl, a neutral peptide in 30% yield following treatment with trypsin (2 mg of peptide, 30 hours, 37", 10: 1 ratio of peptide to enzyme).
It is apparent that, under these conditions, trypsin slowly hydrolyzes a Lys-Pro linkage, giving a Pro-Gly-Phe tripeptide. This finding parallels that of Carnegie (16), who found that in the analogous peptide from the human Al protein an Arg-Pro linkage is slowly hydrolyzed.
It has an NH*-terminal region identical with Peptide 1'11 as shown by the 11 steps of the direct Edman procedure (Table V). This peptide differs from Peptide 1'11 only by an extension of the COOHterminal region.
As shown in Table V, the sequence of this region was determined by using chymotryptic peptides. Thus the COOKterminal Phe of Peptide I'll is further extended in Peptide I'llA as follows: -Phe-Gly-Tyr-Gly-Gly-Arg-Ala-Ser-Asp.
Peptide Pi.2 (Residues 115 to 127): Xer-Trp-Gly-(Ala,2 Glu,2 Gly , Lys, Pro)-Phe-Gly-Tyr-This la-residue peptide is from the same tryptophan region as Peptides Pll and PllA and appears to have the identical sequence based on amino acid analysis and the partial sequence determination.
Peptide 12d. (Residues 115 to 1SS)-This peptide lacks the KHzterminal tripeptide Ser-Arg-Phe present in l'eptide 1lA but otherwise is identical as shown by amino acid composition. Three steps of the Edman degradation gave Ser-Trp-Gly, thus showing an identity with Peptide 1'12 over 13 residues.
The first 12 residues of the NH2 terminus were determined by the Edman degradation. The Gly-His-Asp-Ala sequence of I'eptide P14-T4 was also obtained from Edman degradation.
To obtain further data, I'eptide P14 (5 mg) was treated with with trypsin (120 pg) for 6 hours at 37". The tryptic peptides were resolved by high voltage electrophoresis.
All of the expected peptides were found.
Peptide P14-T9 moved identically with free arginine, however, as shown by amino acid analysis before and after acid hydrolysis.
The Edman degradation of Peptide P14-T8 gave the Ser-Gly-Ser-Pro sequence. These data, combined with the Ala-Arg sequence obtained from the COOH terminus with carboxypeptidase, complete the sequence as shown. The positions and sequences of Peptides P14-T5, T6, and T7 were not determined.

~0 170)
Peptide T was used in this study to position the peptides located in the carboxyl-terminal region. When the Al protein was treated with N-bromosuccinimide, the COOH-tryptophanyl linkage was cleaved, and a large peptide, referred to as Peptide T, was subsequently isolated (26). This peptide occupies the 54 residues of the COOH-terminal region of the Al protein, from tryptophan to arginine, since it begins at the NH2 terminus with Gly-Ala-Glu-Gly-Gln-and terminates at the other end with Ala-Arg-Arg (26)) the same COOH-terminal region found for the Al protein.
The same peptide was isolated (27) in higher yield following treatment of the Al protein with BNPS-skatole, a highly specific reagent which cleaves solely at the COOH-tryptophanyl linkage giving two peptide fragments, Peptides L and T.
The NHz-terminal region of Peptide T, derived from BNPSskatole degradation, was evaluated with the direct Edman procedure; seven steps were successful, giving Gly-Ala-Glu-Gly-Gln-Lys-Pro.
This result confirms the sequence found in tryptic Peptide T18 and peptic Peptide P12, both of which are derived from the t.ryptophan region.
Tryptic Peptides-Peptide T was treated with trypsin under the same conditions as the Al protein.
Nine peptides and arginine were found on peptide mapping as shown in Fig. 4. Although 10 to 50% of the NH*-terminal residue of each peptide was destroyed by the ninhydrin treatment, the composition of each peptide was determined and, in some cases, subtractive Edman degradation was performed.
The tryptic peptides were easily identified by comparisons of their composition with the analogous tryptic peptides obtained directly from the Al protein, and are designated as such on the peptide map (Fig. 4). The only peptide not obtained from the tryptic digest of the Al protein, Peptide TTl, must originate from the NH2 terminus of Peptide T; the composition is shown in Table VII, along with the chymotryptic peptides.
Chymotryptic Peptides-Peptide T was digested with chymotrypsin under the same conditions as the tryptic digestion. Seven peptides were produced and clearly resolved as shown by by peptide mapping (Fig. 4). The composition of these peptides is given in Table VII.
Sequence of Peptide T-From the tryptic and chymotryptic peptides, the sequence of Peptide T can be directly determined, utilizing in addition information from certain peptic peptides. The sequence of Peptide T is shown in Fig. 5. It is evident by comparison with tryptic Peptide T18 and T19 and peptic Peptides Pll, PllA, P12, and P12A that Peptides TCl and TTl, which contain a tyrosine and several glycine residues, are derived from the NHz-terminal region. Peptides TCl and TTl differ only by the Gly-Gly-Arg sequence present in the COOH-terminal region of Peptide TTl.
Peptide TC7 is identical with the COOH-terminal residues of the Al protein, Ala-Arg-Arg, a tripeptide which is released upon splitting the COOH-methionyl linkage with CNBr (11). Thus Peptide TC6, which contains methionine, must be joined to Peptide TC7.
Peptide TC5, therefore, bridges Peptides TC4 and TC6 and must have the sequence Ser-Lys-Ile-Phe which agrees with the terminal end of Peptide TT5 (T23) and Peptide TT6 (T24).
With four steps of the Edman procedure, the NH&erminal region of Peptide TC6 was found to be Lys-Leu-Gly-Gly-, thus establishing the union between Peptides TT6 (T24) and TT7 (T25).
Peptides T27 and P14-T8 are in fact identical as shown by Edman degradation.
These data are sufficient to establish the sequence of Peptide T as shown in Fig. 5.
The strategy used for positioning the various tryptic and peptic peptides t,o give the complete amino acid sequence of the 170 residues of the Al protein can best be followed by reference to Fig.  6. The sequence of residues 1 to 20 was determined previously (11) by using Peptide CBl, derived from CNBr cleavage at the carboxyl end of the methionine at position 20. Tryptic peptide T4 overlaps Peptide CBl and extends the sequence to residue 24. Residues 25 to 37 were established with peptic Peptide P2 which, after CNBr treatment, showed the Arg-His bridge between tryp- 5. The amino acid sequence of Peptide T is shown with the tryptic peptides denoted by TT and the chymotryptic peptides denoted by TC. The peptides and amino acids are numbered in order beginning at the NHz-terminal end.
tic Peptides T4 and T5. The His-Arg region (residues 31 to 32) must fall between Peptides T5 and T7 because of the results from peptic Peptide P2 and tryptic Peptide T6A where the Arg-Asp linkage was not fully hydrolyzed by trypsin. The sequence from residues 38 to 43 was determined with peptic Peptides P3 and P4, and tryptic Peptide T7 and T8. The sequence of residues 44 to 89 (Peptide P5 or Peptide R) was elucidated previously (20). Peptide R is joined at both terminal positions by Phe-Phe linkages as shown by tryptic Peptide T8 at the NH, terminus and T13 at the COOH terminus.
The bridge from Peptide R to the trytophan residue, 1.16, was established with Peptides T14-T18 as positioned by peptic Peptides P9, PlO, Pll, and PllA. This is one of the most interesting regions since it contains methylated arginine residue 107 and the triproline sequence. From the tryptophan residue to the COOH-terminal arginine, residue 170, Peptide T, the sequence of which is shown in Fig. 5, was used to complete the P1-lo-' p*-20-N-Ac-Ala-Ser-Ala-Gln-Lys-Arg-Pro-Ser-Gln-Arg-Ser-Lys-Tyr-Leu-Ala-Ser-Ala-Ser-Thr-Met --TT1-vT2-I  bT3'i---TT4 p2-30 I -4o&-A~p-His-Ala-Arg-H~~~Phe-Leu-Pro-Arg-His-Arg-Asp-Thr-Gly-lle-~~-Asp-Ser-teu-Gly-Arg-lrT6--I T7 dence of trace peptides other than those from incomplete or secondary cleavage was discovered. All of the espected tryptic peptides were found on peptide mapping, a factor which has proven useful in subsequent studies to determine the phylogenetic variation in the Al protein sequence (15). The question arises, therefore, concerning the reported microheterogeneity of the Al protein at pH 10.5 (20). This phenomenon cannot be attributed to the state of methylation of the single arginine residue 107 since the protein would still have a high positive charge at that pH, and it is unlikely that the difference in pK between arginine and monomethylarginine, the major derivative, would be significant.
A more reasonable interpretation is that deamidation occurs either during acid extraction or at alkaline pH during assay. Deamidation of glutamic or aspartic residues may occur more readily if adjacent to a basic residue (31). In the Al molecule, there are several such potential sit,es of deamidation: Gln-Lys or Gln-Arg (residues 4 to 5, 9 to 10, 73 to 74, 121 to 122) and Lys-Asn (91 to 92). Small quantities of the deamidated forms of Peptides T12, T14, T18 and T19 were observed in some preparations. ordering of the peptides.
These data combine to give t,he complete sequence of the bovine Al protein.

DIscusslox
The strategy used in this study for determining the sequence of the bovine Al protein was determined in part by the highly basic character of this protein.
It was often necessary, as in the case of histone IV (28), to utilize preparative paper elect.rophoresis and chromatography to obtain homogeneous peptides. Pepsin was particularly helpful because it provided large peptides, useful not only for positioning the tryptic peptides, but also for localizing disease-inducing sit,es. Although only 19 of the 27 possible tryptic peptides were recovered from the Dowex 50 column, the remaining tryptic peptides were easily identified and isolated from peptide maps of the Al protein or Peptide T. The high solubility and open conformation of the Al molecule no doubt facilitate tryptic hydrolysis and lead to unusually well defined peptide maps. Although 80 to 90% of the available linkages are cleaved by trypsin in 30 min, it is of interest that the Lys-Asp linkage (Nos. 57 to 58) and Arg-Asp linkage (Nos. 33 to 34) are more slowly hydrolyzed by trypsin than are conventional linkages as shown by recovery of some of the uncleaved peptides. The COOH-terminal Arg-Arg sequence appeared to be cleaved by trypsin under the conditions used, unlike the situation in histone IV (28). The hydrolysis of the Lys-Pro linkage (Nos. 122 to 123) and the Arg-Pro linkage @OS. 79 to 80) by trypsin was particularly helpful in determining t,he sequences in those regions. Peptide T18, resulting from cleavage of the Lys-Pro linkage, was isolated from the Dowex 50 column whereas the larger unhydrolyzed fragment, Peptide T19, was apparently retained. The sequence of residues 80 to 84 was difficult to determine; hydrolysis of the Arg-Pro linkage in Peptide T13 gives two peptides which helped to confirm this sequence (20).
Depending on the degree of methylation of arginine residue 107, it is apparent that the sequence reported for the bovine Al protri!l is unique (Fig. 6) ; no variants were found and no evi- Vol. 246,No. 18 An alternate explanation of the microheterogeneity comes from possible degradation in situ because of the open conformation of the Al protein and its high susceptibility to proteolysis (9). Traces of lysosomal enzymes might cause limited hydrolysis at the most vulnerable linkages; degradation of the Al protein occurs in situ (2) in bovine spinal cord. As a consequence of the significant turnover of the Al protein, (half-life 21 days (32)), the additional bands seen on gel electrophoresis could reflect the presence of large peptide fragments.
What properties of the Al protein, associated with its role as a structural protein of the myelin membrane, can be inferred from the amino acid sequence? A rare feature of the sequence is the methylated arginine residue (No. 107), the only methylated residue detected in the molecule and the first localization of methylated arginine in a protein sequence (23). Both the Al protein (33) and histones (34) serve as acceptor proteins for methylation by enzymes from brain and other tissues. The phylogenetic importance of the methylated arginine is illustrated by its presence at the analogcus position in the ,41 proteins of many species including turtle, chicken, rat, rabbit, guinea pig, cow, monkey, and human (23). In accord with the open conformation of this molecule, it is likely that the methylation of this highly specific arginine residue involves the recogrution of a unique segment of the polypeptide chain by the appropriate methylase.
Near the middle of the polypeptide chain there is a Pro-Arg-Thr-Pro-Pro-Pro sequence. Although other conformations are possible, the restrictions imposed by the 4 proline residues could induce a sharp U-shaped bend in the molecule (23) and thereby induce the polypeptide chain to fold back on itself. Such an interpretation would explain the relatively low axial ratio of 10: 1 found for the Al protein by viscosity studies (5), a lower ratio than would be predicted for a completely open conformation lacking significant secondary or tertiary structure.
The proposed open double chain structure (23) could be stabilized within the myelin membrane by interaction of the Al protein with lipids or other proteins, such as proteolipid, or perhaps by cross-chain interaction within the Al molecule itself. The close proximity of methylated arginine residue to t,he triproline region suggests a cooperative function; i.e. the methylated arginine is more nonpolar than arginine and might participate in stabilization of the proposed double chain structure by either cross-chain interaction with t,he two proximal phenylalanine side chains (see Fig. 2 of Reference 23) or by interaction with lipids. It is of interest that the triproline sequence occurs infrequently in proteins, and has been reported (35) in rabbit IgG in which it constitutes the hinge region.
Adjacent to the triproline sequence is found threonine residue 98. This threonine residue is the sole focus for glycosylation in the Al protein by the polypeptide N-acetylgalactosaminyl transferase from the submaxillary glands (36) ; a GalNAc-0-Thr linkage is formed.
Many other proteins were tested for acceptor activity (21) ; in addition to the Al protein, only the denuded polypeptide chain of the submaxillary glycoprotein, the natural acceptor for this enzyme, was functional.
The question arises whether the Al protein, which contains no carbohydrate, may be glycosylated and deglycosylated during its synthesis and secretion as part of the myelin membrane.
The sequence of the Al protein allows relevant interpretation of the structural role of the Al protein within the myelin membrane.
If it can be a.ssumed that the unfolded form of the Al protein in solution prevails in situ, then the Al molecule appears ideally designed to promote maximum interaction with other components such as phospholipids.
It is likely that this interaction is primarily electrostatic; the basic groups are distributed quite randomly without any obvious periodicity and could interact with the phosphate groups of lipids.
The basic polypeptide chain may in fact serve as a type of template which directs phospholipids to the positive regions. It is of interest that, in the Al protein, regions of 8 to 10 residues exist in which basic residues are absent; nonpolar and even negative charges are found in these gaps and could provide a separate focus for attract.ion of lipids or possibly proteins such as proteolipids.
The Al protein is the agent in the central nervous system which is responsible for induction of experimental allergic encephalomyelitis.
In defining the immunopathological role of the Al protein, the amino acid sequence again assumes major significance because of the unfolded conformation of the polypeptide chain. Unlike most proteins, therefore, peptide fragments derived from the Al protein are themselves immunogenic; i.e. they induce a delayed hypersensitive response, a process involviug the specific sensitization of lymphocytes that ultimately leads to the disease stat,e (2, 12-14).
The g-residue segment (Peptide T18) surrounding the single tryptophan residue is the major disease-inducing site in guinea pigs (13). In rabbits, both Peptide I'5 (referred t.o as Peptide R (20)) and Peptide T18 are highly encephalitogenic (19,20). Although Peptide P5 contains 45 residues, it is of interest that a small region exists that has a close similarity to the tryptophan region: (In the Peptide P5 region) $tr--Gly-Ser-Leu-Pro-Gln-Lz (Peptide T18) Trp-Gly-Ala-Glu-Gly-Gln-Lys 116 122 It has been shown (14) that the tryptophan, glutamine, and lysine residues are essential for encephalitogenic activity in guinea pigs. The sequence similarities in these regions suggest that similar requirements may exist for disease induction in rabbits with the exception that the requirements for the aromatic residue are less specific; i.e. tyrosine can substitute for tryptophan. In this regard, preliminary evidence for encephalitogenic activity of this region (Peptide T12) in rabbits has been reported (37). In the guinea pig, the tryptophan requirement appears absolute (13, 14) ; both Peptides P5 and T12 are inactive.
Moreover, it is possible that other peptide sequences quite distinct from these regions may be capable of disease induction in other species such as monkey, rat, dog, and even human.
This subject is currently being investigated in our laboratory.