THE JOURNAL OF BIOLOGICAL CHE~WGTRY Vol. 246, No. 1, Issue of January 10, pp. 9-15, 1971

Printed in U.S.A.

Phylogeny of Immunoglobulin Structure and Function

IV. IMWUNOGLOBULINS OF THE GIANT GROUPER, EPINEPHELUS ITAIRA*

(Received for publication, May 25, 1970)

L. IV. CLEM

From the Department of Microbiology, College of Medicine, University of Florida, Gainesville, Florida 32601 and Marineland Research Laboratories, Marinelancl, Florida 3208~

SUMMARY

Giant grouper serum was shown to contain 16 S (~3 mg per ml of serum) and 6.4 S (-6 mg per ml of serum) immu- noglobulins. The 16 S immunoglobulin had a molecular weight of -700,000, a relatively high hexose content, and was composed of approximately equimolar amounts of H and L polypeptide chains. It is suggested that this molecule resembles immunoglobulin M on the basis of polypeptide chain properties but that it is most likely a tetramer instead of the “typical” 19 S pentamer of immunoglobulin M. The 6.4 S immunoglobulin appeared structurally to represent a fragment of the 16 S molecule; the major difference being that the 6.4 S H chain was missing -30,000 daltons.

In attempting to fit the class Osteichthyes into a scheme of vertebrate immunoglobulin evolution, it was suggested that the “primitive” form of polymeric immunoglobulin M may be a tetramer of subunits held together by both disulfide and noncovalent bonds.

The ability to synthesize humoral antibody in response to stimulation with complex antigens appears to be possessed by representatives of all true vertebrates (reviewed in References 2 and 3). Numerous studies have also been reported regarding the structural relationships between the immunoglobulins of Yower” vertebrates and those of “higher” animals. Certain cyclostomes (4) and elasmobranchs (5-8) apparently have only one class of immunoglobulin; this immunoglobulin resembles immunoglobulin M as defined for rabbits (9) and man (10). Certain amphibians (11) have two immunoglobulin classes; one resembles IgM1 and the other IgG of higher animals. Very little physicochemical data is available for the immunoglobulins of the numerically very large “intermediate” group, namely the class Osteichthyes, and thus the phylogenetic level at which multiple classes of immunoglobulin appeared is obscure. There-

* This investigation was supported by National Science Foun- dation Grant GB 8632. The previous paper in this series is Reference 17.

1 The abbreviations used are: IgA, IgM, IgG, and F(ab’p)t nomenclature recommended by a conference on human immuno- globulins sponsored by the World Health Organization (1)) DNP-, 2,4-dinitrophenyl-.

fore, in light of the paucity of data on bony fishes in general, and the finding that the giant grouper, Epinephelus itaira, yielded appreciable amounts of high and low molecular weight antibodies to the 2,4-dinitrophenyl determinant (12, 13), studies on the physicochemical properties of the immunoglobulins of this species were undertaken. This report presents the results of these studies.

MATERIALS AND METHODS

Source of Grouper Serum-The giant groupers used in this study were the same as those used in the study described in the following report (13). Normal serum was obtained from each animal prior to immunization. Supernatants from which anti- bodies to DNPl had been removed by immune precipitation (13) were also employed and gave results similar to those obtained with normal serum.

Fractionation Procedures-DEAE-cellulose chromatography and Sephadex G-200 gel filtration were performed as described in detail previously (6). The exceptions were that the buffers for DEAF-cellulose chromatography contained 1 M urea and the NaCl gradients were linear.

Analytical Procedures-Antigenic analysis of grouper proteins, preparation of rabbit antisera, preparation of polypeptide chains by partial and extensive reduction, analytic ultracentrifugation, determination of extinction coefficients and hexose contents, and fingerprints of tryptic digests were all conducted as described in detail for similar studies on lemon shark immunoglobulins (6). Radioiodination of grouper proteins was performed by the chlor- amine-T method (14). Amino acid compositions were deter- mined using an automatic amino acid analyzer (JLC-5AH Japan Optical Electronics Laboratories, Ltd., Tokyo) on acid hydroly- sates, (6 N HCl, 105”, 20 hours), according to the method of Spackman, Stein, and Moore (15). Partial specific volumes (s) were calculated from the amino acid composition (16). No corrections were made for losses during hydrolysis and a hexose to total carbohydrate (assumed carbohydrate @ = 0.61) ratio of 1:1.6 as for human IgM (10) was assumed for these calcula- tions.

RESULTS

Purifiation of Grouper Antibodies and Immunoglobulins- Grouper 16 S and 6.4 S antibodies specific for the DNP- determi- nant were purified by the dissolution of immune precipitates as described in the following report (13). In order to establish the

9

by guest on February 10, 2020http://w

ww

.jbc.org/D

ownloaded from

Inanmnoglobulins of Giant Grouper Vol. 246, No. 1

FRACTION NUMBER

FIG. 1. DEAE-Cellulose chromatography of grouper serum “soiked” with 12SI-labeled grouuer antibodies to DNP-. Condi- tions of fractionation: 150 ml se&m was applied to a column (5 X 40 cm), fraction volumes, 12 ml. Recoveries: ODzgo = 90%, lZsI = 92%.

B 16slq H

FRACTION NUMBER

FIG. 2. Sephadex G-200 gel filtration of antibody-containing pools (d and B) obtained by DEAE-cellulose chromatography of whole grouper serum spiked with 1261-labeled grouper antibodies to DNP-.

elution position of grouper immunoglobulins, normal grouper serum was “spiked” prior to fractionation with Y-labeled 16 S and 6.4 S antibodies (~0.1 fig of antibody per ml of serum). Initial efforts at chromatographing grouper serum on DEAE- cellulose were hampered by the insolubility of several proteins, especially the macroglobulins, in the low ionic strength buffers employed. This insolubility could be readily circumvented by the addition of 1 M urea to the buffers. Fig. 1 presents the pro- tein, radioactivity and NaCl gradient profiles of a DEAE-cellu- lose chromatographic separation of whole grouper serum and shows several points for discussion. First, no grouper protein is eluted until the conductance reaches about 4000 mhos (-0.05 M NaCl); this lack of cationic proteins is in contrast to the results obtained with the sera from other species, such as rabbits and

sharks (5, 6). Secondly, the radioactivity (16 S and 6.4 S anti- bodies) elutes in two distinct peaks (designated A and B). Each of these radioactive peaks was then subjected to Sephadex G-200 gel filtration; Fig. 2 gives the elution profiles of two such columns. Most of the grouper protein (and radioactivity) in Peak A eluted in an included column volume and was designated 6.4 S im- munoglobulin (see below). Fractionation on Sephadex G-200 of Peak B from the DEAE-cellulose column yielded both in- cluded and excluded proteins. Most of the radioactivity was excluded and, on the basis of evidence presented below, this fraction was designated as 16 S immunoglobulin. The second peak from this latter column was found to contain several pro- teins in addition to some 6.4 S immunoglobulin by immuno- electrophoretic analysis. Although approximately 40% of the protein in this peak was 6.4 S immunoglobulin, it was not studied further. Based upon the recoveries of purified immunoglobulins and specific radiolabeled antibodies from such experiments, grouper serum contains -3 mg of 16 S immunoglobin per ml and ~6 mg of 6.4 S immunoglobulin per mI.

Physicochemical Properties-Grouper immunoglobulins pre- pared in the manner described above were assayed for purity by immunoelectrophoresis. Fig. 3 shows that each immunoglobulin was pure and exhibited a slow /3 mobility as seen with grouper antibodies (13). Also evident in this figure is the absence of cathodic moving proteins in grouper serum. The purified pro- teins were considered to be immunoglobulins by virtue of being antigenically identical to their specific antibody counterparts as shown in Fig. 4; also evident here is the antigenic deficiency of the 6.4 S immunoglobulin relative to the 16 S immunoglobulin as noted for specific antibodies (13).

Each of the grouper immunoglobulins was analyzed by sedi- mentation velocity and representative Schlieren patterns of these are shown in Fig. 5. Each immunoglobulin was judged to be homogeneous by this method. The s&,~ of the 6.4 S im- munoglobulin from a slight concentration dependence was taken to be 6.35 S (7.2 mg per ml = 5.7 S; 3.6 mg per ml = 6.05 S; 1.8 mg per ml = 6.2 S). The sZo,u, of the 16 S immunoglobulin exhibited considerable concentration dependence (10 mg per ml = 13.0 S; 5 mg per ml = 15.0 S; 2.5 mg per ml = 15.6 S). Therefore the .s&,~ of this latter protein was taken to be 16.1 S.

Grouper antibodies to DNP- and immunoglobulins were sub- jected to sedimentation equilibrium analysis. LnC uersus X2 plots of representative data (Fig. 6) indicate that the heteroge- neity is minimal. Weight average molecular weights calculated from these data are presented in Table I. The partial specific volumes used in these caIculations were determined from the amino acid compositions, i.e. 0.708 for the 16 S immunoglobulin and 0.716 for the 6.4 S immunogJobulin. There appeared to be no concentration dependence and thus the weights for each speed and protein concentration were averaged. The moIecular weights of the grouper 16 S immunoglobulin and antibodies to DNP- each appear to be -700,000 and of the 6.4 S immuno- globulin and anti-DNP- molecules to be -120,000. The value of -700,000 for the 16 S immunoglobulin was quite unexpected (see “Discussion”). Thus an experiment was performed wherein lemon shark 19 S immunoglobulin (mol wt -900,000 (6)) and grouper 16 S immunoglobulin labeled with 1251 were mixed and gel filtered on a column of Sepharose 4B. The elution profile of this column is presented in Fig. 7 and shows that the grouper 16 S immunoglobulin eluted after the lemon shark protein. This experiment supports the molecular weight data indicating the

by guest on February 10, 2020http://w

ww

.jbc.org/D

ownloaded from

Issue of January 10, 1971 L. w. Clem 11

FIG. 3. (left). Immunoelectrophoresis of grouper serum, im- munoglobulins, and polypeptide chains. Rabbit antisera and grouper serum were used undiluted. The grouper protein concen- trations were 2 to 3 mg per ml.

FIG. 4. (middle). Ouchterlony-type studies with purified grouper antibodies @lb), immunoglobulins (Zg) and polypeptide chains (ZZ). Rabbit antisera were used undiluted. Grouper pro- teins were used at 2 to 3 mg per ml.

-’

- IO -Ab

36 36.5 37 RADIU5’km’)

FIG. 6. Plots of the logarithm of the protein concentration (fringe displacement, in centimeters) versus distance from the center of rotation squared for grouper 16 S and 6.4 S antibodies and immunoglobulins. Protein concentrations and centrifugal speeds employed were 0.25 mg per ml and 16,260 rpm for the 6.4 S proteins and 0.25 mg per ml and 8,225 rpm for the 16 S proteins.

grouper 16 S immunoglobulin to be smaller than the shark 19 S immunoglobulin.

Extinction coefficients and hexose contents of the two grouper immunoglobulins were measured and the results are given in Table II. The 16 S immunoglobulm has about three times more hexose than the 6.4 S molecule. It should be mentioned that these determinations were performed simultaneously with

FIG. 5. (right), Schlieren patterns of purified grouper immuno- globulins: A, partially reduced grouper 6.4 S immunoglobulin (3.5 mg per ml), S20ea = 6.1 S; B, grouper 6.4 S immunoglobulin (8 mg per ml), s200a = 5.6 S; C, partially reduced grouper 16 S im- munoglobulin (5 mg per ml), ~20,~ = 14.9 S; D, grouper 16 S im- munoglobulin (5 mg per ml), ~20,~ = 15.0 S. The solvent employed was 0.14 Y NaCl, 0.015 M, Tris-HCl, pH 7.4.

TABLE I Weight average molecular weights (XIPa) of grouper immunoglobu-

lins and antibodies

The solvent was 0.14 M NaCl, 0.015 M Tris-HCl, pH 7.4.

Protein

16 S Ig

16 S Anti-DNP-

6.4 S Ig

6.4 S Anti-DNP-

-

--

-

rPm \ 0.25 I I 0.50 0.75

6,166 720 8,225 694

w/ml 759 640

703 650

6,166 760 799 700 8,225 647 690 618

13,410 123 106 136 16,200 120 116 110

13,410 16,200 I -

122 128 126 121 108 110

Concentration

-

--

-

724 = 661 693

750 = 652 701

122 = 115 119

125 113 = 119

several control proteins (see Reference 17). The amino acid compositions of the grouper 16 S and 6.4 S immunoglobulins are given in Table III. The two proteins showed a remarkable degree of similarity with only slight differences in serine, glycme, and isoleucine content being evident.

Polypepticle Chain Structure--Each of the grouper immuno- globulins was extensively reduced and alkylated and then sub- jected to Sephadex G-200 gel filtration in the presence of 5.0 M guanidme hydrochloride. The presence of 1261-labeled antibodies to DNP- in these immunoglobulin preparations permitted an evaluation of the elution volumes of antibody polypeptide chains. The elution profiles from two such columns are pre-

by guest on February 10, 2020http://w

ww

.jbc.org/D

ownloaded from

12 Immunoglobulins of Giant Grouper Vol. 246, No. 1

LEMON SHARK

? 19s lg +

E 04-

8

- 6000

CJ f I

:03- 5

- 6000 2

6 2 n

lg 02- F3

GROUPER - 4000 o 0

k g=Ip H

I Ol- p '9, - 2000

"@' 0 \

1 I 50 60 70 60 90

FRACTION NUMBER

FIG. 7. Sepharose 4B gel filtration of a mixture of lemon shark 19 S immunoglobuiin (mol wt ~900,000) and grouper 16 S immuno- globulin and 6.4 S immunoglobulin labeled with “61.

TABLE II

Extinction coeficients in various solvents and hexose contents of puri$ed grouper immunoglobulins

Proteins Solvent E;;oorl an h S.D. Hexose

% Grouper 0.3 M KC1 16.57 i 0.19 1.1 6.4 S Ig 0.1 N NaOH 17.82 f 0.18

5.0 M Guanidine hydro- 16.50 f 0.31 chloride

Grouper 0.3 M KC1 13.78 i 0.44 y3.5 16 S Ig 0.1 N NaOH 15.03 i 0.41

5.0 M Guanidine hydro- 13.53 f 0.28 chloride

TABLE III Amino acid compositions of grouper 16 S and 6..4 S immunoglobulins

Amiio acid 16 S Immunoglobulin

g per cent carbohydde-free )rOlei?Z

6.4 S hmunoglobulii

Lysine. . . . . . . . . . . . . 5.10 4.7 Histidine. . . . . . . . . 1.6 1.5 Arginine . . . . . . . . . . . 3.6 4.1 Aspartic acid. . . . . . . 14.2 13.6 Threonine . . . . . . . . . 11.0 11.1 Serine.............. 7.3 8.4 Glutamic acid. . . . . . 12.1 12.0 Proline. . . . . . . . . . . . . 5.8 5.4 Glycine. . . . . . . . 5.4 6.4 Alanine. . . . . . . . . . . 3.7 4.1 Cysteineb.. . . . . . . . . 0.9 0.7 Valine. . . . . . . 7.2 6.8 Methionine. . . . . . 1.2 0.7 Isoleucine . . . . . , . . . 3.8 3.0 Leucine . . . . . . . . 7.0 7.1 Tyrosine . . . . . . . . . . 5.6 5.8 Phenylalanine . . . . . . 4.6 4.4

a Based upon average of duplicate determinations. b Measured as cystine.

I 1 64s lg I

xx) 25a MO $50 VOLUME(ml)

FIG. 8. Sephadex G-200 gel filtration in 5 M guanidine hydro- chloride of extensively reduced and alkylated grouper immuno- globulins (O-O) and antibodies to DNP- (C- - -0).

FIG. 9. Fingerprints of tryptic digests of grouper 16 S and 6.4 S H polypeptide chains. The left panels are photographs of the original chromatograms. The right panels are traced photographs showing spots visible to the eye on the originals. Designations employed are Lys, lysine marker for electrophoresis: red, phenol red marker for chromatography; y, yellow spots; B! brown spots: @ = spotsunique to 6.4 S H chain; @ = spots unique to 16 S H chain. The origin is in the lower right-hand portion of each pattern.

sented in Fig. 8; in each experiment -95% of the applied ma- terial was recovered in the profiles presented here. The optical density (280 rnp) elution profile of the 16 S immunoglobulin indicated that -70% of the recovered material eluted in a vol- ume corresponding to a molecular weight of ~70,000 (16 S H

by guest on February 10, 2020http://w

ww

.jbc.org/D

ownloaded from

Issue of January 10, 1971 L. W. Clem 13

chains). The remainder eluted in a volume indicative of mo- lecular weight -22,000 (L chains). The optical density profile of the 6.4 S immunoglobulin showed 58% of the recovered ma- terial to elute in a volume corresponding to a molecular weight of -40,000 (6.4 S H chains) with the remainder appearing as L chains. The radioactivity profiles of these columns indicated the antibody chains to have molecular weights indistinguishable from those of their immunoglobulin counterparts.

In view of the 2 H-2 L chain subunit structure that is probably ubiquitous among all immunoglobulins, efforts were made to study the subunit structure of the grouper immunoglobulins. Partial reduction (0.1 M dithioerythritol or cysteine) of the grouper 16 S and 6.4 S immunoglobulins followed by dialysis against neutral aqueous buffer did not alter their sedimentation characteristics as shown in Fig. 5. Subsequent gel filtration of such partially reduced proteins in 5 M guanidine hydrochloride (or 1 M propionic acid) resulted in the liberation of H and L chains. The failure of unreduced grouper immunoglobulins to dissociate in 5 M guanidine hydrochloride indicates that the par- tial reduction procedures employed had in fact broken some inter- chain disulfide bonds and therefore, the polypeptide chains were held together by both noncovalent and covalent interactions.

Antigenic Xtudies-As stated previously, partial reduction of grouper immunoglobulins followed by gel filtration in the pres- ence of 5 M guanidine hydrochloride resulted in the liberation of H and L chains. Preliminary studies indicated these chains to be immunologically reactive although it was necessary to recycle the 6.4 S immunoglobulin chains to obtain pure preparations. Employing the technique of immunodiffusion in agar (Ouchter- lony), rabbit antisera to the 16 S L chains and to whole grouper serum indicated the L chains from the 16 S and 6.4 S immuno- globulins to be identical (not shown). Rabbit antisera to the 6.4 S H chains showed these chains to be identical to the 16 S H chains, whereas antisera to the 16 S H chains showed them to have determinants not present on the 6.4 S H chains, as depicted in Fig. 4. Another representation of the antigenic deficiency of the 6.4 S H chains relative to the 16 S H chains can be seen in Fig. 3. The double precipitin arcs observed between the anti- sera to the 16 S H chains and whole grouper serum are presuma- bly due to reactions with the slow diffusing 16 S molecule and the faster diffusing but antigenically deficient 6.4 S molecule.

Peptide Map Comparisons-Further evidence relating to the nature of the apparent mass and antigenic deficiency of the 6.4 S H chain relative to the 16 S H chain is obtained from peptide analysis of these chains. Fig. 9 shows the peptide maps of tryptic digests of grouper 16 S and 6.4 S H chains. These maps indicate that these polypeptide chains contain numerous common spots including 4 readily identifiable ones, which stain yellow with cadmium-ninhydrin. The maps of the 6.4 S H chain digest appear to contain only 1 unique spot not present in the 16 S H digest, whereas at least 13 unique spots are evident with the 16 S H chain digest.

DISCUSSION

The data presented here indicate that the grouper 16 S im- munoglobulin has a molecular weight of -700,000, has a rela- tively high hexose content, and is composed of disulfide linked H (mol wt -70,000) and L (mol wt ~22,000) polypeptide chains. The polypeptide chain yields from extensively reduced material (based upon optical density readings) indicate there are probably equimolar amounts of H and L chains in the intact 16 S mole-

cule, i.e. 8 H and 8 L chains. The one disconcerting aspect of the studies on this molecule has been the unexpected failure to obtain reductive subunits; if, however, one assumes that the 2 H-2 L chain subunit is ubiquitous among all immunoglobulins, then the grouper 16 S molecule may be a tetramer of subunits held together by both covalent and noncovalent bonds.

The grouper 6.4 S immunoglobulin was found to have a mo- lecular weight of -120,000, a low carbohydrate content, and a composition of disulfide-linked H (mol wt -40,000) and L (mol wt -22,000) chains. Here again, assuming comparable extinc- tion coefficients as for the 16 S immunoglobulin polypeptide chains, there appears to be equimolar H and L chains, i.e. 2 H-2 L. The relationship of the grouper 6.4 S immunoglobulin to the grouper 16 S immunoglobulin appears to be that of a frag- ment wherein the H chain of the former is lacking a segment present on the latter. There are four experimental observations consistent with this hypothesis. (a) The antigenie data indicate that a portion of the 16 S H chain is not present on the 6.4 S H chain and (6) the molecular weight data indicate that this miss- ing portion accounts for -30,000 daltons. (c) The peptide maps indicate most of the 6.4 S H chain tryptic peptides to be present in the 16 S H chain; the one exception may represent the C-terminal peptide of the 6.4 S H chain. On the other hand, the 16 S H chain contains numerous unique peptides; these may account for the “missing” -30,000 daltons. (d) The low hexose content of the 6.4 S immunoglobulin would be expected if this molecule represents an F(ab’&like fragment of the 16 S mole- cule. Additional studies are required to establish the possible F(ab’,&like nature of the 6.4 S immunoglobulin and to establish any metabolic (anabolic versus catabolic) interrelationships be- tween the grouper 16 S and 6.4 S immunoglobulins. In light of amino acid sequence studies which suggest that immuno- globulin polypeptide chains of other species are the products of several cistrons (for example see Reference 18), the finding of an animal synthesizing both intact and “partial” chains would certainly be exciting.

It would now seem appropriate to attempt to relate the grouper immunoglobulins to those of other species. First of all it should be stated that the most valid method to make such comparisons would entail the use of amino acid sequence data. This approach would be more apt to show that a phylogenetically conserved cistron (or cistrons) provided the genetic information for the synthesis of the polypeptide chains in question. In the absence of sequence data, any comparisons must be made on the basis of “grosser” molecular parameters, such as molecular weights of the whole molecule and its component polypeptide chains, carbohydrate content, and electrophoretic mobilities. Initial efforts in this direction with lower vertebrates focused on the 19 S and 7 S immunoglobulins of sharks (5-8). Such studies indicated that these two shark immunoglobuhns belonged to the same immunoglobulin class and that this class closely resembled IgM as defined for higher vertebrates. On the other hand, amphibians, such as the bullfrog (II), had 19 S and 7 S immuno- globulins which were considered to be phylogenetic counterparts of IgM and IgG respectively. Somewhat “intermediate” be- tween these two classes of vertebrates is the lungfish, a Dipnoid fish related to the Crossopterygii, i.e. an “uncle” of the tetrapods (19). The lungfish has recently been shown (20) to possess a 19 S immunoglobulin which, on the basis of its H chains, resem- bles IgM; molecular weights of the intact molecule were not ob- tained. The lungfish 5.9 S immunoglobulin appears to belong

by guest on February 10, 2020http://w

ww

.jbc.org/D

ownloaded from

14 Immunoglobulins of Giant Grouper Vol. 246, No. 1

to a different immunoglobulin class than the 19 S molecule, but because it has an H chain with a molecular weight of only -40,000, does not seem to have a readily identifiable mammalian counterpart (20); its relationship, if any, to a similar sized im- munoglobulin in the duck (21) remains obscure.

When attempting to relate the immunoglobulins of a member of the class Osteichthyes to these other lower vertebrates, the one “certainty” would have been based on the expected presence of a 19 S IgM (mol wt -900,000) ; this was not found in the study reported here. The grouper 16 S immunoglobulin had a mo- lecular weight of only -700,000. Thus, with this protein, the only comparative statements must rely solely upon polypeptide chain properties. The 16 S H chain molecular weight (-70,000) and the relatively high hexose content are compatible with an IgM designation for the grouper 16 S immunoglobulin with the implication, as mentioned above, that based upon the molecular weight of the whole molecule it may be a tetramer of subunits instead of the “typical” pentameric 19 S form of IgM. In de- fense of this tenuous immunoglobulin class designation, it should be stated that molecular mass cannot be the only criterion for making such comparisons as evidenced by the presence of 7 S IgM in newborn (22, 23) and pathological (24) human sera and what is presumably 19 S IgG in piglet serum (25). Since, as discussed above, the grouper 6.4 S immunoglobulin appears to resemble a fragment of the 16 S molecule, it seems unprofitable here to discuss it as a significant aspect in the evolution of multi- ple immunoglobulin classes. Its structural relationship to the grouper 16 S molecule suggests it to be quite different from the low molecular immunoglobulins of the lungfish (20) and duck

(21). The question thus arises as to the evolutionary significance of

the tetrameric grouper IgM since the general scheme of im- munoglobulin evolution that appeared to be emerging suggested the shark pentameric IgM to be the “starting point.” How- ever, in light of statements in the literature that sharks do not reflect ancestral properties of higher vertebrate evolution (19, 26), perhaps the emerging scheme may have been misleading. For example, is the data sufficient to say that the “primitive” macroglobulin antibody was not a tetramer? Obviously it is dangerous to generalize from one species, in this case the grouper, and therefore detailed studies on other species of Osteichthyes are necessary. However, fragmentary data in the literature do support the hypothesis that the macroglobulin antibodies of numerous Osteichthyes are smaller than 900,000 daltons. Su- crose gradient ultracentrifugation studies with immune sera from margates and snappers, two species of marine teleosts, indicated that the macroglobulin antibodies did not sediment as rapidly as did rabbit IgM antibodies (27). Recent studies aimed at assessing the molecular weight of the margate macroglobulin have been hampered by the presence of both 16 S and 21 S populations in purified immunoglobulin preparations. Prelimi- nary sedimentation equilibrium studies of such mixtures suggest molecular weights of -700,000 and -1,400,OOO for these two species of protein2 The s!&,~ of purified goldfish macroglobulin was found to be 16.4 S (28), and therefore these proteins may have a molecular weight comparable to that of the grouper. The paddlefish, a chondrostean considered to be a primitive member of the class Osteichthyes, yielded a macroglobulin with

2 L. W. Clem, unpublished data.

an s& of -14 S (29) ; in the absence of s 2rersu.s c data it is im- possible to comment on the molecular weight of this protein.

Even if the high molecular weight immunoglobulin of all mem- bers of the class Osteichthyes were found to have a molecular weight of -700,000, the question could still remain as to whether this really represents the primitive form of polymeric IgM. The only reply to this suggestion must involve a discussion of the incomplete data available on the immunoglobulins of representa- tives of the cyclostomes, an even more ancient group in terms of phylogenetic origins. The lamprey has both high (-14 S) and low (-7 S) molecular weight immunoglobulins which, on the basis of antigenic comparisons, appear to belong to the same im- munoglobulin class. The H chain of the 7 S protein was felt to resemble the p chain and thus the 14 S and 7 S molecules were considered to be IgM (4). Preliminary reports suggest that the molecular weight of the lamprey macroglobulin is considerably less than 900,000 (30), and thereby strongly indicates that the phylogenetically ancient version of what is probably IgM could not be a pentamer of 2 H-2 L chain of subunits.

One additional interesting aspect of the lamprey immuno- globulin is the relative lack of interchain disulfide bond in favor of noncovalent interactions. The observed failure of partially reduced grouper (reported here) and paddlefish (29) immuno- globulins to dissociate in neutral aqueous buffers, suggests these intersubunit noncovalent interactions are still important at the level of the Osteichthyes. Recent data showing that intra- cellular mouse IgM exists only in the 7 S form and that polym- erization occurs during release from the synthesizing cell (31), is consistent with the hypothesis that initial efforts (phylogeneti- tally speaking) at polymerization may have resulted in the as- sembly of other than pentameric forms. The presence of strong noncovalent intersubunit bonds would be expected to help stabilize this hypothetical primitive molecule.

Aclcnowledgments-The author wishes to thank Mr. Edsal McLean for excellent technical assistance and Drs. P. A. Small, Jr., and Gerrie Leslie for numerous helpful discussions. Dr. John Cebra suggested the usage of urea in the buffers for DEAE- cellulose chromatography.

REFERENCES

1.

2.

3.

4.

5.

6.

7.

8.

9.

10. 11.

12.

Nomenclature for Human Immunoglobulins, Bull. World Health Organ., 30, 447 (1964).

CLEM, L. W., AND LESLIE, G. A., in M. ADINOLFI (Editor), Immunology and development, Spastics International Medi- cal Publications, London, 1970, p. 62.

GOOD, R. A., AND PAPERMASTER, B. W., Advan. Immunol., 4, 1 (1964).

M.~RCH.YLONIS, J., AND EDELMAN, G. M., J. Exp. Med., 137, 891 (1968).

MARCHi\LONIS, J., ANDEDELMAN,G. M.,J. Ezp,Med.,122,601 (1965).

CLEM, L. W., AND SMALL, P. A., JR., J. Exp. Med., 125, 893 (1967).

CLEM, L.W., DEBOUTIUD, F., AND SIGEL, M.M.,J.Immunol., 99, 1226 (1967).

SURAN, A. S., T~R.uL, M. H., .~ND PAPERMASTER, B. W., J. Immunol., $9, 679 (1967).

LAMM, M. E., AND SMALL, P. A., JR., Biochemistry, 6, 267 (1966).

M;LL&, F., AND METZGER, H., J. Biol. Chem., 240,3325 (1965). MARCH~LONIS, J., ANDEDELM~N,G. M.,J. Exp.Med., 124,901

(1966). LESLIE, G. A., AND CLEM, L. W., J. Immunol., 103, 613 (1969).

by guest on February 10, 2020http://w

ww

.jbc.org/D

ownloaded from

Issue of January 10, 1971 L. w.

13.

14.

15.

16.

17. 18.

19. 20.

21. 22.

CLEM, L. W., AND SMALL, P. A., JR., J. Exp. Med., 132, 385 (1970).

MCCONAHEY, P., AND DIXON, F., Int. Arch. Allergy, 29, 185 (1966).

SPACEMAN, D. H., STEIN, W. H., AND MOORE, S., Anal. Chem., 30, 1190 (1958).

COHN, E. J., AND EDSALL, J. T., Proteins, amino acids, and peptides as ions and dipolar ions, Reinhold Publishing Corporation. New York. 1943.

LESLIE, G. A., XND CLEM, i. W., J. Exp. Med., 130,1337 (1969). HILL, R. L., FELLOWS, R. E., AND LEBOVITZ, H., Proc. Nat.

Acad. Sci. U. S. A., 66, 1762 (1966). ROMER, A. S., Science, 168, 1629 (1967). MARCHALONIS, J. J., Aust. J. Exp. Biol. Med. Xci., 47, 405

(1969). GREY, H. M., J. Immunol., 98, 811 (1967). PERCHALSKI, J. E., CLEM, L. W., AND SMALL, P. A., JR., Amer.

J. Med. Sci., 266, 107 (1968).

Clem 15

23. HUNTER, A., FEINSTEIN, A., AND COOMBS, R. A., Immunology, 16. 381 (1968).

24. SOL&AN,‘A., END MCLAUGHLIN. C. L.. J. C.&n. Invest.. 49. 150 (1970).

, I

25. KIM, Y. B., BRADLEY, S. G., AND WATSON, D. W., J. Immunol., 97. 189 (1966).

26. ROMER, A‘. S.,.Notes and comments on vertebrate paleontology, University of Chicago Press, Chicago, 1968, p. 41.

27. CLEM, L. W., AND SIGEL, M. M., in R. T. SMITH, P. A. MEI- SCHER, AND R. A. GOOD (Editors), Phylogeny of immunity, University of Florida Press, Gainesville, 1966, p. 209.

28. TRUMP, G. N., J. Zmmunol., 104, 1267 (1970). 29. FISH, L., POLLARA, B., AND GOOD, R. A., inR. T. SMITH, P. A.

MEISCHER, AND R. A. GOOD (Editors), Phylogeny of im- munity, University of Florida Press, Gainesville, 1966, p. 99.

30. LITMAN, G., HOWELL, J., FINSTAD, J., GOOD, R. A., AND POL- LARA, B., Fed. Proc., 28, 819 (1969).

31. PARKHOUSE, R. M. E., AND ASI~ONAS, B. A., Biochem. J., 116, 163 (1969).

by guest on February 10, 2020http://w

ww

.jbc.org/D

ownloaded from

L. W. ClemIMMUNOGLOBULINS OF THE GIANT GROUPER, EPINEPHELUS ITAIRA

Phylogeny of Immunoglobulin Structure and Function: IV.

1971, 246:9-15.J. Biol. Chem. 

  http://www.jbc.org/content/246/1/9Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/246/1/9.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on February 10, 2020http://w

ww

.jbc.org/D

ownloaded from