Biochimica et Biophysica Acta, 379 (1975) 627-637
© Elsevier Scientific Publishing Company, Amsterdam - - Printed in The Netherlands BBA36906 P A R T I A L A M I N O ACID SEQUENCE OF A N IgA2 H U M A N I M M U N O G L O B U L I N HEAVY CHAIN*
CARLOTA WOLFENSTEIN-TODEL, BLAS FRANGIONE and EDWARD C. FRANKLIN Irvington House Institute, Rheumatic Diseases Study Group, New York University Medical Center, 550 First Avenue, New York, N.Y. 10016 (U.S.A.)
(Received May 24th, 1974) (Revised manuscript received September 3rd, i974)
SUMMARY The amino acid sequence of the heavy chain of an IgA2, A2m(l) polymeric myeloma protein (Avi) was studied. Altogether, sequence data were obtained for some 130 residues. Including the amino acids placed by homology with IgAx, this accounts for some 170 residues, thus representing more than one-third of the a 2 chain. The sequence includes 26 amino acids from the amino-terminal end (VH III), and 25 residues at the "hinge" region. Of a total of 17 cysteine residues, 14 were located in regions of the molecule which were identical or homologous in the a~ and az chains. These striking homologies, together with the results obtained by diagonal maps of the hinge region, suggest a similar arrangement for the disulfide bridges in both subclasses of IgA. Study of the cysteine-containing peptides of the J chain are consistent with the conclusion that the J chains associated with different classes of immunoglobulins are identical.
INTRODUCTION Human immunoglobulin A consists of two antigenically distinguishable subclasses (IgA1 and IgA2), which differ in the structure of their heavy chains [1-3]. A genetic polymorphism was found in the A2 subclass [4-6]. Proteins belonging to the more common A2m(1)genetic variant lack the disulfide bond linking the heavy and light chains. In the minor allotype A2m(2), this bridge is present [7], and the a chain peptide involved in the heavy-light bridge has an amino acid sequence identical to that of IgA1 molecules [8-9]. Studies of secretory IgA, polymeric myeloma IgA and IgM have demonstrated the presence of a third chain, the J chain which appears to be immunologically and chemically identical in all of them [10-11]. Recently it has been shown that J chain is bridged to the carboxy-terminal end of the a chain [ 12-13]. Sequences around cysteine residues of human a~ chains [9, 14-16], as well as those of the J chain obtained from IgA1 [17], have been reported and a model for the * This work was presented in part at the International Symposium on the IgA System, Birmingham, Alabama, October 23, 1973.
628 hinge region has been proposed [18]. In the case of IgA2, only labile disulfide bonds have been characterized [19]. In this report, we present sequences around the remaining cysteine containing peptides obtained from the heavy chain of a human IgA2, myeloma protein and its J chain, and compare them to those of human IgA~. Furthermore, the sequence of 26 residues from the amino-terminal end is presented. MATERIALS AND METHODS Protein Avi is an lgA2, k, A2m(l) myeloma protein. It was immunologically typed and purified as previously described [19]. For partial reduction and alkylation, protein Avi (250 mg) was dissolved at a protein concentration of 20 mg/ml in 0.27 M Tris-HC1 buffer (pH 8.2) and exposed to 10 mM dithiothreitol at room temperature under N2 for ! h. Reduction was terminated by addition of iodo-[14C]acetic acid (22 mM, specific activity to 0.7 Ci/mole). After incubation for 1 h, the mixture was dialyzed against water and subsequently 1 M acetic acid. Heavy, light and J chains were then separated by chromatography on Sephadex G-100 in 1 M acetic acid (column 3 cm × 130 cm). The column effluent was monitored both by measuring the absorbance of the fractions at 280 nm and the radioactivity in 50-/~1 aliquots. The heavy, light and J chains were pooled, freeze-dried and tested for purity by immunoelectrophoretic analysis and polyacrylamide gel electrophoresis [17].
Complete reduction and radioactive alkylation This was performed on heavy chains that had been prepared as described above but using unlabelled iodoacetic acid as the alkylating agent. The protein was dissolved at a concentration of 20 mg/ml in 7 M guanidine-0.27 M Tris-HCl buffer (pH 8.2) and was reduced with 10 mM dithiothreitol at room temperature under N2 for 1 h. Radioalkylation was achieved by added iodo-[~aC]acetic acid (22 mM, specific activity 0.7 Ci/mole) and allowing the mixture to stand for 1 h in the absence of light. Excess reagents were removed by exhaustive dialysis against water followed by freezedrying.
Enzyme digestions and separation of peptides The labelled J chain was digested with L-(1-tosylamido-2-phenyl)-ethylchloromethyl ketone trypsin (Worthington) in 0.2 M NH4HCO3 (pH 8.3) for 15 h at 37 °C, enzyme substrate ratio 1:50 (w/w). The digest was freeze-dried and subjected to pepsin digestion (enzyme: substrate, I : 50 (w/w)) in 5 % formic acid at a concentration of 20 mg/ml for 15 h at 37 °C. The radioactive peptides were purified as described [9, 16]. The completely reduced and alkylated heavy chain (200 rag) was digested with L-(l-tosylamido-2-phenyl)-ethylchloromethyl ketone trypsin in 0.2 M NH4HCO3 (pH 8.3) for 15 h at 37 °C, enzyme :substrate ratio, 1:50 (w/w)). The digest was freezedried, dissolved in 5 ml of 1 M acetic acid and fractionated by chromatography on a column of Sephadex G-50 (3.5 cm × 180 cm) equilibrated in 1 M acetic acid at room temperature. Fractions of 6 ml were collected at a flow rate of 25 ml/h. The eluate was monitored both by measuring the absorbance of the fractions at 280 nm and the radioactivity on 100-#1 aliquots dissolved in a toluene-based counting solution containing 10% (v/v) Biosolv 3 (Beckman) and counted in a Beckman LS-150 liquid
629 scintillation counter. The insoluble material was further digested with pepsin, enzyme substrate ratio 1:50 (w/w) in 5% formic acid at 37 °C for 15 h. The procedures employed in the purification of radioactive peptides, amino acid analyses and determinations of amino acid sequences have been described previously [9]. All mobilities are given relative to the distance between e-(2,4-dinitrophenyl)-lysine and aspartic acid at pH 6.5 [20]. Automatic amino acid sequence determination was performed with a Beckman model 890 Sequencer by the method of Edman and Begg [21] on 20 mg of completely reduced and alkylated heavy chain. Phenylthiohydantoin amino acids were identified by gas chromatography with a Hewlett-Packard gas chromatograph (HewlettPackard Co., Palo Alto, Calif.) [22] and by amino acid analysis of the hydrolyzed derivatives. The N-terminal peptide from the J chain was obtained with the technique used by Press et al. [23] after pronase digestion of the completely reduced protein (enzyme :substrate ratio, 1:20) for 2 h at pH 8. The complete amino acid sequence of all the peptides could not be determined due to the limited amount of protein available. RESULTS a chain
The amino acid sequence of the N-terminal end of the heavy chain as determined by automatic sequence analysis and manual Edman degradation of the peptictryptic peptide containing the first cysteine residue, is shown in Table I, and identifies TABLE I AMINO ACID SEQUENCE OF THE AMINO-TERMINALEND OF THE HEAVY CHAIN OF PROTEIN Avi Glu-Val-Gln-Leu-Glu-Glu-Ser-Gly-Gly-Gly-Leu-Val-Lys-Pro-Gly-
!
Sequencer Gly-Ser-Leu-Arg-Leu-Ser-Cys-Ala-Ala-Ser-Gly
I
I
Peptide PT 11
I
it as belonging to the VH III subgroup, which is the predominant subgroup among a chains [24]. This sequence is in agreement with the one previously determined by Dr A. C. Wang (personal communication). The isolated a chain was completely reduced and alkylated with iodo-[~4C]acetic acid, and subjected to tryptic digestion. After digestion the insoluble material was separated by centrifugation. Fig. 1 shows the elution pattern of the soluble tryptic peptides fractionated on a Sephadex G-50 column in 1 M acetic acid. The eluted peptides were pooled into five fractions. The insoluble material was further digested with pepsin. The labelled peptides were purified by high-voltage paper electrophoresis at pH 6.5, 3.5, and 2.1 using radioautography to localize them. The amino acid compositions and mobilities at pH 6.5 of tryptic and tryptic-peptic peptides are shown in Table II.
630
CPM
0.5
0,000
0.4
8,000
0.3
6,000
o
0.1
.,,,"
2,000
:..o-..t 70
90
110
130
150
t70
Tab{' N~
Fig. 1. Gel filtration of a tryptic digest of completely reduced and 14C-labeled carboxymethylated heavy chain (protein Avi, IgA2, A2m(1)) on Sephadex G-50. Peptides from 200 mg of protein were dissolved in 5 ml of 1 M acetic acid and added to the column (3.5 cm × 180 cm) equilibrated in the same solvent. Fractions of 6 ml were collected at a flow rate of 25 ml/h. The eluate was monitored both by measuring the absorbance of the fractions at 280 nm (-) and the radioactivity on 100#1 aliquots ( - - - ) . The pooled fractions are indicated on the figure.
TABLE II A M I N O ACID COMPOSITION OF CARBOXYMETHYLCYSTEINE PEPTIDES OBTAINED FROM THE COMPLETELY R E D U C E D A N D ALKYLATED HEAVY CHAIN OF PROTEIN Avi (IgA2, A2m(1)) Peak:
I
Peptide:
T1
II
IV
IIl
Precipitate
Amino acid Lysine Histidine Arginine CM cysteine Aspartic acid Threonine Serine Glutamic acid Proline Glycine Alanine Valine Methionine Leucine Isoleucine Tyrosine Phenylalanine CHO
1.0 1.1 1.3 0.6 2.3 3.0 1.9 2.1 1.0 3.3 2.7 2.0 0.5 5.8
Mobilities at p H 6.5
0.19
T2
T3
0.8 1.8 0.6 1.8 2.6 2.2 2.0 0.9 2.2 2.2
1.6 1.0 1.7 2.0 1.8 1.1 1.1 9.2
2.8 7.5
T4
T5
T6
T7
0.8 1.0
1.0
1.9 1.0
0.7 1.0
1.5 1.4 1.9 2.5 2.1 2.2 2.1 2.0 1.5
0.6 4.0 3.8 3.9 3.2 4.2 2.1 2.1
0.6
1.9
2.8
0.5 1.1 1.8 1.2 2.0 1.0 2.1 2.0 0.9 0.7 2.1
0.7 0.9
1.5
T8
1.0 0.7 1.8 1.7 0.6 2.0 2.0 2.0 1.1
TP9
TP10
TPll
TP12
0.6
0.6
0.6
1.1 0.6
0.9
1.7
0.6
1.0
1.0 2.0
1.1
0.9
1.0
1.1
1.0
0.55
0.55
0.34
0
0.9 0.5 ?
? + 0.20
0.7 0.8
0.7 1.0
1.8 +
0.19
0.41
0.34
0
--0.3
0.4
631
Peptide TI The exact amount of phenylalanine is difficult to ascertain due to the presence of carbohydrate, which gives a peak eluting with phenylalanine in the amino acid analysis. Amino-terminal sequence by the dansyl-Edman procedure was Leu-AlaGly-Asp-Pro. In view of its large size, relatively low yield, and the presence of lysine and arginine this peptide may represent the result of incomplete tryptic cleavage and may contain some of the other peptides.
Peptide T2 This large glycopeptide has an amino acid composition that closely resembles that of a peptide reported for IgA1 by Moore and Putnam [15] and Mendez et al. [16] and differs only by having 7.5 leucyl residues instead of 9. This discrepancy may be due to technical problems. In this peptide too, the amount of phenylalanine could not be determined because of the presence of large amounts of carbohydrate. The dansylEdman procedure gave a sequence of the first amino acids as Leu-Ser-Leu. Previously [19] a tryptic-peptic peptide Thr-Cys-Thr-Leu was isolated from the same protein, and we assume that it is part of peptide T2, since a homologous peptide from protein Ha (IgA1) [15] (see Table IV) contains the same sequence Thr-Cys-Thr-Leu.
Peptide T3 This 25 residues peptide has the amino acid composition characteristic of the "hinge" peptide. Since carboxymethylcysteine is destroyed during hydrolysis, the value of 1.7 residues of carboxymethylcysteine actually represents 3 residues after correction. This tryptic peptide has 8 residues more than the previously reported peptic-tryptic peptide [19]. The dansyl-Edman procedure was used to identify the first 14 amino acids, and gave the sequence: His-Tyr-Thr-Asx-Pro-Ser-Glx-AsxVal-Thr-Val-Pro-Cys-Pro. This provides the overlap with the previously established sequence of the peptic-tryptic "hinge" peptide, and gave the complete sequence: His-Tyr-Thr-Asx-Pro-Ser-Glx-Asx-Val-Thr-Val-Pro-Cys-Pro-Val-Pro-Pro-ProPro-Pro-Cys-Cys-His-Pro-Arg.
Peptide T4 This peptide contains 25 residues. The corrected value for carboxymethylcysteine is 3 residues. There are 3 residues of serine and 2 of valine, since the peptide bonds involving valine are not completely hydrolyzed in 20 h. The sequence of the first 16 steps was established by the dansyl-Edman methods as: Asp-Leu-Cys-GlyCys-Tyr-Ser-Val-Ser-Ser-Val-Leu-Pro-Gly-Cys-Ala which is identical with the one reported for IgA1, proteins Pat [9] and Oso [16]. However, the differences in the amino acid composition point to the existence of several differences at the carboxy terminal end compared to the homologous peptide from the al chain (see Table IV).
Peptide 7"5 Amino acid analysis of this peptide showed it to contain 30 residues. The dansyl-Edman procedure gave its amino terminal sequence: Asx-Phe-Pro-Pro-SerGlx-Asx-Ala-Ser-Gly-Asx-Leu-Tyr. This peptide appears to contain the tryptic-
632 peptic subtilisin peptide that was previously shown to form a disulfide bridge with the "hinge" peptide: G l x - C y s - P r o - A s p - G l y - L y s [19].
Peptide 7"6 The amino acid composition of this peptide is identical to one of the peptides reported by Mendez et al. [16] for IgAl and it has the same mobility. The dansylEdman procedure showed an identical amino terminal sequence: L y s - G l y - A s x - T h r Phe. It is located near the C-terminal end because it overlaps with the sequence reported by Chuang et al. [25] for the carboxy terminus of the a chain.
Peptide T7 This peptide had the same amino acid composition as a peptide reported for IgA1, protein Hal [15] and Oso [16]. The sequence of this basic peptide was established by the dansyl-Edman methods as: Ser-Val-Thr-Cys-His-Val-Lys and preliminary observations suggest that it is a constant feature of the Fd fragment.
Peptide T8 Amino acid analysis of this glycopeptide showed it to have 15 residues. DansylEdman degradation gave the amino terminal sequence Gly-Glx-Asx.
Peptide TP 9 This acidic tripeptide was obtained after pepsin digestion of the tryptic insoluble material. The dansyl-Edman procedure yielded the sequence Ala-Cys-Leu.
TABLE IIi AMINO ACID COMPOSITION OF THE CARBOXYMETHYLCYSTEINE PEPTIDES ISOLATED AFTER ENZYME DIGESTION OF PARTIALLY REDUCED AND ALKYLATED J CHAIN OBTAINED FROM IgA3, A2m(l) Compositions are moles of amino acid per mole of peptide. Hydrolysis time 20 h. PT = pepsintrypsin, Pro ~ pronase. Peptides
PT1
PT2
PT3
PT4
PT5
PT6
PT7
Pro 1
Lysine CM Cys SO2" Aspartic acid Threonine Serine Glutamic acid Proline Glycine Alanine Valine Leucine Isoleucine
1.0 0.6
1.0 0.6 1.0
0.5
0.5
1.0 1.9
1.3
0.5 2.0
0.5 1.0 0.9
0.6 1.0
0.6 1.0
Mobility pH 6.5**
0
1.3 0.9 0.9 2.0 1.6
0.9 0.7 1.1 1.0
1.0 1.0
2.0
1.0
0.6 1.4
0.6
0.55
0.6
0.5 0
0.95
1.0
1.0
1.25
* CM Cys SO2 carboxymethylcysteinesulfone. The low yield of CM Cys SO2 was due to destruction during hydrolysis [31]. ** Mobilities are given relative to the distance between e-dinitrophenyl-lysineand aspartic acid.
633 Peptide TP 10 and TP 11 These peptides were derived by peptic digestion from the insoluble residue remaining after trypsin digestion. They are related and both had leucine as their Nterminal amino acid. The sequence obtained for peptide TP 11 by dansyl-Edman degradation was: Leu-Ser-Cys-Ala-Ala-Ser-Gly, which places peptide TP 10: Leu (Ser, Cys) at the amino terminal end of TP 11. Peptide TP 12 The sequence of this neutral pentapeptide was established by the dansylEdman procedure as: Thr-Cys-Leu-Ala-Arg. J Chain The amino acid compositions and mobilities at pH 6.5 of the major peptictryptic carboxymethylcysteine peptides obtained from J chain are shown in Table III. The small amount of J chain available did not allow further characterization. The N-terminal peptide of the J chain was isolated after pronase digestion according to the procedure followed by Meinke and Spiegelberg [26] and Mendez et al. [17] for the J chain of IgA~. By comparison with peptides obtained from the J chain and of an IgAa protein [17], the peptides obtained appeared to be identical. DISCUSSION Studies of amino acid sequence of cysteine-containing peptides account for more than one-third of the a chain. Table IV shows a comparison of all the available sequences of the cysteine-containing peptides of several IgA1 and protein Avi IgA2, A2m(1). Several of the peptides obtained appear to be the same for both subclasses. These include the large glycopeptide Avi T2, the previously reported C-terminal peptide [19, 27], the 21 residue peptide Avi T6, the basic heptapeptide Avi T7, the neutral pentapeptide Avi TPI2 and the tripeptide Ala-Cys-Leu (Avi TP 9) which was part of a larger peptide containing the heavy-light bridge in IgA 1 [16]. Other peptides show a high degree of homology. Such is the case of the "hinge" peptide Avi T3, which, as previously reported [19], has a gap of 13 residues, including the carbohydrate moiety, just in the place where duplication of a small fragment was identified in A1 molecules. It has instead a remarkable pentaproline sequence. The extended sequences of the tryptic peptides from the hinge of the ai and a2 chains reveal that their amino- and carboxy-terminal sequences are virtually identical. Residue 2 of this peptide from protein Avi was found to be tyrosine, as in protein Ha (IgA1) [28], while in another IgA~, protein Oso [16] it was reported to be proline. The significance of this difference remains to be evaluated. Another of the homologous peptides is Avi T~, which includes 3 cysteines. Its sequence at the N-terminal end is identical to the corresponding peptide of IgA1, but it appears to differ at the carboxy terminal end. Peptide Avi T5 includes the peptide Glx-Cys-Pro-Asp-Gly-Lys, previously described [18] as being involved in a disulfide bridge with the first cysteine of that "hinge" peptide. The homologous peptide of IgA1 (see Table IV) includes the sequence Cys-Leu-Ala which is also bound to the "hinge" [18], and has an almost identical sequence, except for the amino acids located next to the cysteine. Thus, the position of this disulfide bridge is the same in both
634
$o
m
I
o
u_
o
~
?
o
o
~
,!
i
I
×
~
o
L
¢ o
o
C 0)
"
d
d
2
C
i
C ~
d
~"
V
O'
S
~
°
m
~
m
~
o
o
?
m
m
~
5
5
M
,4
'~
1
.'
L~
o o o .~
m
~
r-i
z :Z
,
m
g
g
e
o
~
o
o
co
m
r~
>
~
~
~
(5
0
0
o
$
o
M
~4
M
M
M
M
>