Rodney Porter and Gerald Edelman share the credit for “discovering” the chemical structure of the unseen and entirely theoretical antibodies said to provide some form of immunity from infectious disease. They were jointly awarded the Nobel Peace Prize in 1972 for their “discoveries.” The researchers both used techniques to break down immunoglobulins into smaller fragments by means of chemical and enzymatic treatments. They then claimed to analyze the physiochemical and biological characteristics of these fragments and proceeded to create hypotheses on how the fragments formed together. Keep in mind, neither researcher purified nor isolated any antibodies and they were still unable to see these entities. In fact, no one had ever seen an antibody as they were simply hypothetical molecules dreamt up in the late 1800’s to explain chemical reactions created in the lab. Many theories were proposed over the decades attempting to explain how these invisible “protectors” looked, formed, and functioned but no one had ever laid eyes on them nor had any direct evidence that an antibody existed as believed. Yet somehow, through the magic of chemistry experimentation, Porter and Edelman were said to be able to chemically define the structure of these imperceptible molecules.
A brief historical background of their work:
“The 1972 Nobel Prize in Physiology or Medicine was awarded jointly to Gerald M Edelman and Rodney R Porter “for their discoveries concerning the chemical structure of antibodies” [1,2]. Porter’s work used the protein-splitting enzyme papain to identify three fragments, two smaller very similar ones, both with capacity of combining with the antigen, and one larger piece lacking this capacity. Edelman’s contribution was the demonstration that immunoglobulin molecules, like most biologically active proteins, were composed of chain structures that were held together by sulfur bonds and could be separated. None of the resultant fragments retained the specific reactivity of the intact immunoglobulin molecule. Since then, progress has been made in understanding the finer characteristics of the various domains of the individual chains of the immunoglobulin molecule.
Rodney Porter 1959
According to the above source, Rodney Porter was able to use papain, a protein-splitting enzyme, to break down the antibody into its three distinct fragments. However, was this truly the case? Below are highlights from Rodney Porter’s contribution to this antibody puzzle.
The Hydrolysis of Rabbit y-Globulin and Antibodies
with Crystalline Papain
“The molecular size of rabbit y-globulin is such that any direct attempt to relate structure to biological activity is not feasible at present. An alternative approach is to degrade antibody molecules in such a way that activity will persist in smaller fragments and so to reduce the structural problems involved.
In an earlier investigation of this type (Porter, 1950b) it was found that, if rabbit y-globulin containing antiovalbumin was digested with crude papain, a fragment with a molecular weight of about 40,000 could be produced which retained the ability to combine specifically with ovalbumin though it would no longer form a precipitate. It seemed probable that this fragment contained an antibody-combining site.
In that work, the amounts of crude enzyme used relative to y-globulin were large, making subsequent investigation of the products of digestion rather difficult. As crystalline papain can now be prepared easily (Kimmel & Smith, 1954) and fractionation techniques have improved, these earlier experiments have been repeated. It has been found that y-globulin is split by papain into three large pieces with very little release of amino acids or small peptides. If the y-globulin contains antibody against any of the several antigens investigated, then two of these pieces retain combining though not precipitating power. The third piece, which may be readily crystallized, has no antibody activity but it has most of the antigenic sites of the original molecule. The isolation and properties of these three fractions will be described.
A preliminary account of this work has been given (Porter, 1958a).
Antisera. Rabbit antiovalbulmin, antibovine serum albumin and antihuman serum albumin were prepared by intravenous injection of the alum-precipitated protein (Porter, 1955).
Rabbit antipneumococci polysaccharide type 3 was prepared by injection of a suspension of the formalin-killed bacteria in 0-9 % NaCl solution.
Goat antirabbit y-globulin was prepared by intravenous injection of alum precipitate and subcutaneous injection of y-globulin with adjuvant (Freund & McDermott, 1942).
Rat antiserum was prepared according to the following schedule. The rats were given two intramuscular injections of protein and adjuvant into different thighs, with 1 week’s interval between injection. Five weeks later they were given three injections, with 3-day intervals between injections, of alum-precipitated protein into the tail vein. They were bled by heart puncture 6 days after the last injection.
Rabbit y-globulin. This was prepared either by chromatography, with diethylaminoethylcellulose (Sober, Gutter, Wyckoff & Peterson, 1956), or by Na2SO4 precipitation according to the method of Kekwick (1940).
Crystalline papain. This was prepared from crude enzyme powder purchased from Hopkin and Williams Ltd., London. The enzyme was crystallized once as the free enzyme and twice as the inactive Hg dimer (Kimmel & Smith, 1954). It was freeze-dried and stored as the dimer.
Quantitative estimation of precipitating antibody. This was carried out according to Kabat & Mayer (1948).
Estimations of inhibitory power. These were made by quantitative antibody assay in the presence of the inhibitor or by measure of the delay in precipitation caused by the inhibitor with the antibody and antigen mixed in optimum proportions.
Chromatography. Diethylaminoethylcellulose and carboxymethylcellulose were prepared according to Peterson & Sober (1956). Column size was 2-4 cm. (diam.) by 30-35 cm. (ht.); volume of mixing chamber was 1200 ml. Buffers used on carboxymethylcellulose columns were O OlM-sodium acetate, pH 5 5, with gradient to 0.9M-sodium acetate, pH 5.5. In the refractionation of fraction I on diethylaminoethylcellulose, the following system was used: 001m-sodium phosphate, pH 64, with gradient to 0 2 M-sodium phosphate, pH 6-2. All buffers were saturated with toluene. The pH was measured with an EIL Direct Reading pH meter (Electronic Instruments Ltd.).
Enzyme digestion. y-Globulin (150 mg.) and Hg papain (1.5 mg.) were dissolved in 10 ml. of buffer (0 1M-sodium phosphate, pH 7 0, 0*01 m-cysteine, 2 mM-ethylenediamine-
tetra-acetate). This solution was incubated at 370 for 16 hr. in the presence of toluene. It was then dialysed against water with vigorous stirring and several changes of the outer liquid over 48 hr. This procedure, which removed the cysteine and ethylenediaminetetra-acetate, and facilitated oxidation, inactivated the enzyme. N-Ethyl maleimide was also used to inactivate the enzyme but as there appeared to be no advantage this was not continued. The non-diffusible digestion products were either freeze-dried or fractionated directly by chromatography after dialysis against acetate buffer, pH 5-5.
Protein determinations. Protein concentrations were determined by reading the absorption at 280 and 260 mu in a 1 cm. cell in a Unicam SP. 500 spectrophotometer.
Radioactivity measurements. Injection of hydrolysed algal [14C]protein and counting of the y-globulin fraction were carried out as described previously (Askonas, Humphrey & Porter, 1956).
Analytical methods. Amino acid analysis was by the method of Moore & Stein (1951) as modified by McDermott & Pace (1957).
N-Terminal amino acids were estimated by the fluoro-dinitrobenzene technique (Porter, 1957a). Hexose as ‘glucose’ was estimated by the anthrone method of Mokrasch (1954). Hexosamine as ‘glucosamine’ was
estimated, after separation from amino acids on a cation-exchange resin (Boas, 1953), by a modification of the method of Elson & Morgan (1933).
When a digestion mixture, prepared as described above, was dialysed against water at 2° a precipitate formed which appeared to be crystalline. If, however, the dialysis was against 0 04N-acetic acid there was no precipitation and the recovery of the non-diffusible digestion products could be estimated either by measuring the absorption at 280 mp or by freeze-drying and weighing the dry powder. This has been done in a number of experiments, and by either method the recovery of y-globulin protein taken has fallen in the range 85-95 %. In view of the probable handling losses in dialysis and freeze-drying, it is considered that the higher figure is the most accurate. When such a digest was examined in the ultracentrifuge in 0O1M-phosphate, pH 6.7, some crystals again formed on dialysis but the supernatant showed only one peak (S20., 3.5). As y-globulin has S20,w 6-5, it was clear that the protein had been split into large fragments of similar size with very little production of diffusible peptides. Attempts to fractionate this mixture by zone electrophoresis were not successful, but resolution could be achieved by chromatography on carboxymethyl-cellulose. Acetate buffer, pH 5 5, was chosen because under these conditions most of the
carboxyl groups of the ion-exchanger are dissociated; if the digest was brought nearer to neutrality crystals formed, indicating a low solubility of at least one component. To help to keep all the material in solution chromatography was carried out at room temperature (20-23o). With a
gradient of increasing salt concentration at this pH, three components could be resolved which have been named fractions I, II and III in order of elution from the column (Fig. 1).
If the gradient on the column was reduced, fractions II and III were more spread and III ‘tailed’ badly, but there was no suggestion of any further resolution. Fraction I appeared very close to the solvent front and this was re-run on a diethylaminoethylcellulose column at pH 6-4. No fractionation was obtained but again with a slow
gradient there was considerable tailing. By these limited criteria the three fractions appeared to be single components and to be the only significant products of the digestion of y-globulin by papain.
Results with shorter times of hydrolysis under the same conditions suggest that the splitting is complete in very much less than 16 hr., the period which has always been used. It follows that these three fractions are exceptionally resistant to further digestion by papain.
Yields of the three fractions were measured by summing the absorption at 280 mp in each peak. The ratios of yield varied somewhat from experiment to experiment but averaged (I: II: III) 1: 0-8: 0-9, and total recovery from the column was 85-90%. When re-run, fractions I and III were recovered in about 95 % yield and fraction II in about 85% yield. The absorptions of peaks I, II and III at 280 m, at a concentration of 1 mg./ml. in water or 0-02N-acetic acid were 1-4, 1-4 and 1-0 respectively. If the relative yields are corrected for the lower recovery of fraction II, and the lower specific absorption at 280 m,p of fraction III, then the corrected relative yields (I:II:III) are 1:0-9:1-25 by wt. In some experiments the fractions were concentrated by pressure-dialysis in the cold against water or 0-02N-acetic acid; in-soluble material was discarded and the solution freeze-dried and weighed. The yields of I and III were similar to those above but II was somewhat lower. Considerable error can occur because of denaturation in dilute solution, and variable losses on chromatography, but it is considered that the average yields calculated from chromatography approximate to the yields of the fractions produced in the digestion.
Fraction III is readily identifiable as the material with a low solubility near neutrality. It may be crystallized and recrystallized by dialysing a solution in 0-02N-acetic acid against sodium phosphate buffer, pH 6-0-7-0, at 20. The crystals are diamond-shaped plates, often of considerable size but thin and easily broken (Fig. 2).
Molecular weights. The three fractions were studied in the ultracentrifuge (see Addendum) and the results for normal y-globulin are simmarized in Table 1. The sum of the molecular weights of the three fragments is very close to the molecular weight of the original y-globulin. This is in agreement with the high recovery of non-diffusible digestion products. The relative sizes of the three fragments were (I:II:III) 1:1-05:1-6, which is in
approximate though not exact agreement with the calculated relative yields of the fragments.”
“These results have been recalculated in terms of amino acid residues per mol. of fraction and per mol. of y-globulin. It has been assumed that the ash-free, moisture-free fractions have the same nitrogen content of 16 % as found for y-globulin by Smith, McFadden, Stockell & Buettner-Janesch (1955). In view of the difference in basic amino acid content the figure may be rather high for fractions I and II and low for fraction III. The results of this calculation are given in Table 3. The amino acid residues accounted for by the three fractions range from 83 to 116% of the figures for whole y-globulin. These discrepancies are probably explicable by errors in the different estimations, the assumption of 16 % of nitrogen in each fraction and by the peptide material lost in the dialysis. The overall recovery of amino acid residues in this calculation is 94 %.
“Carbohydrate estimations were made by Dr H. R. Perkins of this Institute, and the results with whole y-globulin and the fractions are given in Table 4. About two-thirds of the carbohydrate of the original molecule is found with fraction III, one third with fraction I and a small amount, which may not be significant, is with fraction II. It seems likely therefore that the carbohydrate moiety of y-globulin is in the two pieces, the larger with that part of the molecule which corresponds to fraction III and the smaller with the part corresponding to fraction I. If the total recovery of carbohydrate in the three fractions is compared with that present in the original, the yields are approximately 110 and 120 % for hexose and hexosamine respectively.
N-Terminal assay has been carried out on fractions I and II, and in both alanine was found to be the main terminal amino acid (about 0-9 mol./50 000), together with smaller amounts of aspartic acid (0.2 mol./50 000) and trace amounts of serine and threonine (together about 0- 1 mol./50 000).This is a very similar result to that obtained for whole rabbit y-globulin (Porter, 1950a; McFadden & Smith, 1955), except that the N-terminal amino acid content is more than three times as great as in the whole globulin. It is also in approximate agreement with the results of the digestion of rabbit antiovalbumin with papain powder, when an immunologically active fraction containing one N-terminal alanine per 40 000 was reported (Porter, 1950a). The significance of finding N-terminal alanine in both I and II is not certain, as there are about 110 alanine residues per molecule of y-globulin. An attempt to determine the N-terminal sequence was unsatisfactory owing to shortage of material, but it appeared that the sequence of I was alanylaspartyl and of II alanyl-leucyl. If this is correct, it suggests that II may derive from the N-terminal part of the molecule, as the sequence there is alanyl-leucylvalylaspartylglutamyl (Porter, 1950a; McFadden & Smith, 1955).
Immunological properties. The three fractions were prepared from digests of y-globulin which had been obtained from rabbit antisera against ovalbumin, bovine serum albumin, human serum albumin and antipneumococci polysaccharide type 3. None would precipitate with the corresponding antigen but fractions I and II prepared from y-globulin containing antiprotein antibodies inhibited the precipitation of the antigen by the homologous antiserum. This effect is specific; for example, I and II from antihuman serum albumin y-globulin had no effect when added to bovine serum albumin and its antiserum; fraction III, on the contrary, had no effect whatever its source or whatever the test system. Testing fraction III is difficult owing to its low solubility near neutrality, where antibody-antigen precipitation is carried out, but it is soluble to about 0 3 mg./ml. at 37* at pH 7-2 and it could be shown that it had less than one-thirtieth of the activity of the corresponding fractions I and II.
Quantitative estimates of the inhibitory power of I and II are shown in Fig. 3. In order to assess this on a molar basis it has been assumed that the fractions would have the same proportions of active molecules as the proportion of antibody in the y-globulin taken, and the weight of each fraction has been corrected accordingly when plotting the graph.”
“Antigenic activity. The power of the fractions to precipitate with goat antirabbit y-globulin serum was now tested and I and II showed neither precipitation nor inhibitory activity. Fraction III precipitated 70% of the antibody precipitated by y-globulin.
Rat antiserum was prepared against rabbit y-globulin and fractions I, II and III. Groups of three rats were used for each antigen and, after the
immunization course described above, the serum of each group was pooled, and gave antibody contents of 4-2, 7-8, 8-1 and 4-0 mg. of antibody/ml. for y-globulin, I, II and III respectively. All the fractions were at least as effective antigens as the original molecule, and I and II were the most effective, but as only three animals were used per group the difference may not be significant. When rat anti-y-globulin was tested with the fractions, rather different results from those with goat antiserum were obtained. Fraction III precipitated 50% of the antibody, and I and II 15% each. The antigenic specificities of the fractions were compared by measuring the precipitates formed by 1, II and III with rat anti-II serum. The precipitation curves are shown in Fig. 5. It can be seen that the curves for I and II are almost identical, but III gives almost no precipitate. This again emphasizes the great similarity of I and II and their sharp distinction from III.
Synthesis of the fraction in vivo. In view of the distinctive characters of the three fractions the possibility was considered that they might be synthesized separately and joined into the whole molecule in a separate step. If this were to happen the fractions would probably not all be synthesized from the same pool of amino acids at the same time, and hence if radioactive amino acids were injected into a rabbit the incorporation of radioactivity after a short time interval would vary from fraction to fraction. Hydrolysed algal [14C]protein was used as the source of labelled amino acids; 240uc was injected intravenously, and the rabbit
was bled after 1 and 4 hr. The plasma y-globuilin is not labelled until 30 min. after injection (Askonas et al. 1956). The y-globulin was prepared and hydrolysed with papain, and the fractions were isolated, precipitated with trichloroacetic acid, and counted, at infinite thickness on a 1 cm.2 disk. The specific activities are given in Table 5 and it can be seen that there is no significant difference between the different fractions. There is therefore no evidence to suggest that these fractions or any large parts of them are synthesized independently.
The results show that papain-digestion of rabbit y-globulin causes limited and highly selective splitting to give three large fragments and very few small peptides. As papain shows rather a wide specificity in the digestion of the B chain of insulin (Sanger, Thompson & Kitai, 1955), it is apparent that the structure of the native molecule must be such that many potentially hydrolysable bonds are protected by steric and other factors. Ultracentrifuge studies of the splitting of y-globuilins from different species, by a variety of enzymes (Petermann & Pappenheimer, 1941; Petermann, 1942, 1946), have all shown that a small number of fragments are the main products in each case. It is possible that in y-globulins only small parts of the peptide chain are accessible to proteolytic enzymes, so that even though different enzymes break different bonds in these vulnerable sections the principal large digestion products are similar.
Our results with the papain-digestion of rabbit y-globulin suggest that the single polypeptide chain (Porter, 1950a; McFadden & Smith, 1955) have been split into three distinct sections, which together comprise some 90% of the original molecule. However, fractions I and II are so similar that the question arises whether they could be derived very largely from the same section of the chain. They are almost indistinguishable in N-terminal amino acid, amino acid analysis, molecular weight, antigenic specificity and, if derived from antibody, in their antigen-binding capacity. They differ in chromatographic behaviour and carbohydrate and cystine content.
There are at least three possible ways of splitting a single chain to give results such as this, and they are illustrated in Fig. 6: A shows the obvious split into three distinct sections, B the production of two fractions from the same section of chain and C the result if y-globuilin consists of two types of molecule, such as euglobulin and pseudoglobulin, one of which gives rise to I + III and the other to II+ III. If either B or C were correct then the yield of III should exceed the sum of the yield of I and II. In fact the yield of each of the three fractions is very similar, as would be expected in A. In B and C the overall recovery, in view of the molecular weights found, would be 130,000/188,000, i.e. 70 %, considerably lower than the experimental figure of about 90%, which again is that which would be expected if A were correct. Similarly, with the individual amino acid residues, the recoveries would be only 70%, with the possibility of big variations in individual cases as much more material would be lost. In fact the recovery of residues is 90-95%, and the variation 83-120% in
the recoveries of individual amino acids is close to the range expected from the errors in the different measurements and the assumptions made in the calculation.
Further, in B it would be expected that if I and II were digested further by papain, either interconversion would occur or both would be reduced to a slightly smaller common product. In fact both appear to be stable to a further 16 hr. digestion with papain at 370, being unchanged in chromatographic behaviour and other properties. The simplest explanation of our results therefore seems to be that y-globulin has been split into three distinct fractions, as shown in A. It is probable that the inhibitor described in the earlier work (Porter, 1950b) was a mixture of fractions I and II, and that fraction III was prevented from crystallizing by the impurities introduced with the crude enzyme preparation.
It follows that two large sections of y-globulin (each about 30 % of the whole) are extremely similar in chemical, physical and biological properties. The finding of such close agreement between the amino acid analysis of these two fragments and also an almost identical antigenic specificity suggests that there may be almost a repeat of the amino acid sequence and configuration. This is in contrast with the properties of fraction III, which differs in every respect, so that there appears to be a large repeating unit (I or II) joined to a larger section (III) of entirely different character. This unusual make-up of the y-globulin molecule is presumably related to its antibody activity. The similarity of the pieces with mol.wt., 50 000, where the antibody-combining sites may be very much smaller (Kabat, 1956), raises the question whether large sections are required to maintain the structural integrity of a small combining site or whether antibody-combiing sites may occur anywhere in these pieces.
The big quantitative difference between the inhibitory power of fractions I and II, when derived from antiprotein or antipolysaceharide antibodies, may reflect important differences between the two types of combining site, but perhaps more probably arises from differences in the speed and mechanism of precipitation between the two systems.
The significance of the part of the y-globulin molecule represented by fraction III in antibody-antigen reactions is not known. The ease of crystallization could be taken as evidence of greater identity of structure among individual molecules in ImI than in the whole molecule, which appears to be complex by all available physical and biological data (see Porter, 1958b). Preliminary electrophoretic examination, however, suggests that it is as complex as the original molecule by this criterion. Most of the antigenic sites of y-globulin appear to be on m, but as these sites can only be defined in relation to a given antiserum (Porter, 1957b), variable results might be expected, and there is in fact a difference in this respect between the goat and rat anti-y-globulin serum. Fraction III is remarkably stable. After precipitation by 5% trichloroacetic acid at room temperature, washing with ethanol and ether and drying for 1 hr. at 1100, it will redissolve in 0-05N-acetic acid and it retains its power specifically to precipitate goat antirabbit y-globulin.
The finding that y-globulin appears to be built of three sections, one of exceptional stability and the other two containing the antibody-combining
sites, is reminiscent of Pauling’s (1940) theory of antibody formation, in which it is suggested that antibody molecules consist of a rigid centrepiece and two flexible ends capable of taking up configurations complementary to the antigen and hence forming antibody-combining sites on these flexible parts. Pauling further suggested that the flexibility might be due to a high content of proline residues. There is no evidence on the relative positions of fractions I, II and III in the whole molecule, except that II may be from the N-terminal end. Nor is there any evidence on the essential feature of Pauling’s theory, that the amino acid sequence of all antibodies is identical and that the different antibody-combining sites are formed only by refolding of the same polypeptide chain. The proline content of I and II is less than that of III, whereas the cystine content, often an important feature in determining the stability of a protein molecule, is lower in III than in I and II.
The equal rate of incorporation of radioactive amino acids into the different fractions is in agreement with the view that the whole molecule is synthesized simultaneously from the same pool of amino acids.
- Rabbit y-globulin, when digested by crystalline papain, gives three fragments which together form 90% or more of the original molecule.
- If the y-globulin contains antibodies, two fragments (I and II), of molecular weight 50,000-55,000, retain the power to combine with the antigen. The third fragment (III), molecular weight about 80,000, crystallizes easily and has much of the antigenic specificity of the original molecule.
- I and II are extremely similar in chemical and biological properties, and III differs very widely in all respects. This has led to the suggestion that rabbit y-globulin is formed of two pieces with very similar structure joined to a third piece of quite different character.
Rodney Porter is given credit for helping to determine the chemical structure and the Y-shape of the unseen antibodies. He did so by what appears to be reverse engineering the antibodies by way of breaking the protein substances down into the smallest pieces possible, analyzing them chemically, and then creating a theoretical framework and model for how they supposedly joined together. At no point were the assumed antibodies seen as a whole unit before being digested by papain. Unless you can convince yourself the Superman-like diamond trap image in the study is an antibody fragment, there were no credible images of any of the three fragments matching what an antibody is said to look like, fragmented or whole.
Even if there were accompanying images, it is apparent from his own conclusions that Rodney Porter, after supposedly breaking down the y-globulin into the three fragments, was uncertain if they were antibodies at all as he was apparently not convinced yet if the y-globulin contained antibodies. However, that did not stop Porter from creating a hypothesis based on the results from the chemical reactions of his experiments. He then applied his hypothesis to previously established antibody theories, such as that presented by Linus Paulimg with the Direct Template theory in 1940, which itself was eventually rejected by most immunologists in favor of Frank MacFarlane Burnett’s Clonal Selection Theory in 1957.
It is interesting to note that within the paper, the purification of the antibody fragments is never stated, even though different methods, such as ultracentrifugation and chromotograohy, were listed as having been used. Perhaps this is due to the fact that to achieve the most efficient purification levels required today, it is considered vital to have knowledge of the structure of the antibody first:
“Rodney Porter and Gerald M Edelman first elucidated the characteristic Y-shaped structure of antibodies. In 1972, they were awarded the Nobel Prize for Medicine and Physiology for their findings. These Y-shaped molecules were eventually identified as immunoglobulin G (IgG), the structure of which is composed of four polypeptide chains – two heavy (50 kDa) and two light chains (25 kDa) – linked by noncovalent bonds and disulfide bridges (Figure 1).”
“However, to perform the most efficient purification possible it is vital to have good knowledge of the structure of the antibody (or antibody derived structures) and its cognate antigen and ideally the affinity of their interaction.”
This appears to leave us with a chicken and the egg scenario. How would Rodney Porter have achieved the required purification needed in order to elucidate the chemical structure and the Y-shaped composition of an antibody if he lacked the vital knowledge of the structure of an antibody to begin with? Logic would dictate that he would not be able to do so. Also, Porter was working with what would be considered polyclonal antibodies as monoclonal antibodies were not created until the 1970’s. Polyclonal antibodies are said to be a complex mixture of antibodies contained within the serum of the host whereas monoclonal antibodies are supposedly a specific antibody cloned and cultured in the lab. Antigen-specific affinity purification is said to be required for polyclonal antibodies to prevent the inclusion of non-specific antibodies as more than one type is thought to be within the sample. This procedure was not utilized by Porter. In fact, depending on the chromatography equipment he would have utilized, these methods are said to be unable to purify antibodies from other proteins and macromolecules:
“By contrast, for polyclonal antibodies (serum samples), antigen-specific affinity purification is required to prevent co-purification of nonspecific immunoglobulins. For example, generally only 2–5% of total IgG in mouse serum is specific for the antigen used to immunize the animal.”
“Dialysis, desalting, and diafiltration can be used to exchange antibodies into particular buffers and remove undesired low-molecular weight (MW) components. Dialysis membranes, size-exclusion resins, and diafiltration devices that feature high molecular weight cut-offs (MWCO) can be used to separate immunoglobulins (>140 kDa) from small proteins and peptides. However, except with specialized columns and equipment, these techniques alone cannot purify antibodies from other proteins and macromolecules that are present in typical antibody samples. More commonly, gel filtration and dialysis are used following other purification steps, such as ammonium sulfate precipitation .”
Thus, Porter appears to be left with a pretty important knowledge and equipment gap that is required in order to properly purify antibodies away from other antibodies, proteins, and macromolecules that would also be contained within the sample. It is important to ask how would he have known that the fractions he was working with were even from an antibody at all if they were unable to be properly purified and isolated away from other components/impurities. It seems as long as one takes whatever results are achieved through experimentation and then fits them into a hypothetical model for what these invisible particles could possibly look like, a Nobel Prize awaits.
GERALD EDELMAN 1961
Gerald Edelman is the other half of the dynamic duo awarded for figuring out the chemical composition and structure of an antibody. It was said that he succeeded in splitting the IgG sulphide bonds, thus showing how these unseen fragments were supposedly held together. As was the case with Porter, Edelman tried to reverse engineer the theoretical molecules no one had ever seen before and fit his results into a model:
“Dr. Edelman spent the next several years working backward to recreate a model of the principal antibody molecule, which he achieved in 1969. During that time, he also hypothesized — and was later proven correct — that the vast diversification exhibited by antibodies is an example of the body turning a developmental flaw into an advantage. When cells divide, miniscule errors in transcription often occur, leading to the development of proteins with differences that in the immune system amount to a system of “strength through diversity.”
If it wasn’t clear that both researchers had no idea what an antibody looked like as they were conducting their experiments, it is stated in an article in Immunology Letters that after his experiments, Edelman proposed the Y-shape of an antibody which was subsequently “confirmed” by electron microscopy and x-ray diffraction:
“In the same year, Edelman showed that reduction of the disulfide bonds of antibodies in the presence of denaturizing agents led to dissociation of the molecule into smaller pieces, now known to be the light (L) and heavy (H) chains . Because the molecular weight of the original IgG molecule is 150 kDa, he concluded that the IgG molecule consisted of two heavy and two light chains linked by disulfide bonds and noncovalentinteractions. A Y-shaped configuration was proposed and then confirmed through electron microscopy and X-ray diffraction study. Thereafter, two antigenic types of light chains, denominated and chains were described (Fig. 6).”
Interestingly, no source was cited for this “confirmation” nor were any real images of an antibody supplied as evidence. Instead, we get a drawing and a photo of Edelman next to his model at the Rockefeller University:
Many of the methods Edelman utilized in his paper were similar to Porter. As the paper is 24 pages long, I am presenting just a few highlights from the beginning and the ending to give an overview of his work. You will see that Edelman tentatively (subject to further confirmation; not definitely) concluded that the fragments were held by disulfide bonds and he then created a hypothesis to explain his findings. As with Porter’s work, no images of purified and isolated antibodies nor antibody fragments are presented in the paper. Feel free to read the rest of Edelman’s work with the provided link if you desire the full breakdown of his chemistry experiments:
Studies on Structural Units of the Y-globulins
“In studying the molecular structure of the y-globulins, one is confronted at the outset with the problem of whether these proteins consist of one or of several polypeptide chains. The solution of this problem has significance in determining the chemical basis of antibody specificity, and in formulating a detailed theory of antibody production. In addition, it bears upon the relation between normal y-globulins and those of disease.
A previous report (1) showed that normal human 7S y-globulin and a pathological macroglobulin dissociated to components of lower molecular weight when treated with reagents that cleave disulfide bonds. The present study is concerned with an extension of this approach to a variety of normal and abnormal y-globulins. Partial separation of the dissociation products of y-globulin has been achieved by means of column chromatography and starch gel electrophoresis. The results suggest that y-globulin molecules are composed of several discrete subunits or polypeptide chains linked by disulfide bonds.”
“Porter (45) has recently demonstrated that rabbit y-globulin is cleaved by papain treatment into three fragments which together form over 90 percent of the original molecule. These fragments have molecular weights ranging from 50,000 to 80,000. Similar fragments have been found for papain-treated human y-globulins (46). The fragments obtained by treatment with papain do not seem to be identifiable with the subunits described above. It is likely, however, that the products of papain treatment are composed of portions of these subunits. This interpretation is strengthened by the finding that the two active antibody fragments obtained by treatment with proteolytic enzymes are linked by a single disulfide bond (47).
A unifying hypothesis may be formulated for the structure of proteins in the y-globulin family based on the findings presented above as well as on findings of other investigators. 7S y-globulin molecules appear to consist of several polypeptide chains linked by disulfide bonds. Bivalent antibodies may contain two chains that are similar or identical in structure. The 19S y-globulins would be composed of 5 or 6 multichain units of the size of 7S y-globulin. A provisional explanation for the wide molecular weight range of antigenically related globulins from Bence-Jones proteins to macroglobulins is suggested by this model. Heterogeneity and differences in isoantigenicity (48, 49) may arise from various combinations of different chains as well as from differences in the sequence of amino acids within each type of chain.
The finding that y-globulin contains dissociable subunits has a possible bearing upon the pathogenesis of diseases of y-globulin production. A primary defect in macroglobulinemia and multiple myeloma may be a failure of specificity and control in production and linkage of the various subunits to form larger molecules. Bence-Jones proteins may be polypepfide chains that have not been incorporated into the myeloma globulins because of a failure in the linkage process. Myeloma globulins may consist of combinations of subunits differing from those of the normal y-globulins, although both types of protein appear to contain subunits that are alike. This may explain in part the antigenic and chemical differences and similarities that have been found between the y-globulins of disease and normal y-globulin (36, 37). The hypothesis outlined above is capable of experimental test, since the products of various chemical treatments may now be separated and compared.
When human and rabbit 7S y-globulins were reduced in strong urea solutions by a number of procedures, their molecular weights fell to approximately of the original values. Partial separation of the reduction products was achieved using chromatography and starch gel electrophoresis in urea solutions. One of the components of reduced human 7S y-globulin was isolated by chromatography, identified by starch gel electrophoresis, and subjected to amino acid analyses. The amino acid composition of this component differed from that of the starting material and also from that of the remaining components.
A reduced pathological macroglobulin dissociated to components with an average molecular weight of 41,000. Several reduced human myeloma proteins, when subjected to starch gel electrophoresis, yielded individual patterns that nevertheless had features in common with those of reduced normal v-globulins. Reduction of normal and abnormal y-globulins was accompanied by the appearance of titratable sulfhydryl groups. Chemical treatments other than reduction were used to determine the type of bond holding the subunits together. It was tentatively concluded that they were linked by disulfide bonds. An hypothesis is presented to relate the structural features of the various y-globulins in terms of the multiplicity of polypeptide chains in these molecules.”
- Porter started off by admitting that the size of Rabbit y-globulin is such that any direct attempt to relate structure to biological activity was not feasible
- It seemed probable that rabbit y-globulin digested by crude papain contained fragments of antibody molecules
- Porter claimed that y-globulin split by papain formed three fractions
- His experiments were to see if y-globulin had antibody properties in two of the fractions and stated that there was none in the third fraction
- The amount of precipitating antibodies and the inhibiting power were estimated through various methods and calculations
- N-Terminal amino acids, glucose, and glucosamine content were also estimated
- As y-globulin has S20,w 6-5, Porter stated by this indirect measurement that it was clear that the protein had been split into large fragments of similar size with very little production of diffusible peptides
- Attempts to fractionate this mixture by zone electrophoresis were not successful
- Porter admitted that by his limited criteria, the 3 fractions seemed to be of a single molecule and were the only significant substances in the papain digestion
- Results with shorter times of hydrolysis under the same conditions suggested that the splitting is complete in very much less than 16 hr., the period which had always been used, and thus it was assumed that these three fractions were exceptionally resistant to further digestion by papain (i.e. he decided it was not necessary to attempt splitting the fragment for any longer than 16 hours as it could only be broken into three fragments and he assumed no further degradation would occur if a longer time was attempted)
- Porter admitted the yields are affected by considerable errors due to denaturation by dilute solution and variable losses on chromatography
- Fraction 3 formed crystals that were diamond-shaped plates, often of considerable size but thin and easily broken (that doesn’t sound like any image of an antibody fragment I’ve ever seen…🤔)
- The sizes of the 3 fractions were in approximate but not exact agreement with calculated yield
- It was assumed that the ash-free, moisture-free substance of y-globulin was the same 16% nitrogen concentration as found by previous researchers
- Porter admitted that there were discrepancies which were probably explicable errors in the different estimations, the assumption of 16% nitrogen in each fraction, and the peptide material lost in dialysis
- He believed it was likely that the carbohydrates were in fractions 1 and 2
- He admitted that the significance of finding the N-Terminal in fractions 1 and 2 was unknown
- Attempts to determine the N-Terminal sequence were unsatisfactory
- The three fractions were prepared from digests of y-globulin which had been obtained from:
- Rabbit antisera against ovalbumin (rabbit blood infused with chicken egg protein)
- Bovine serum albumin (cow blood)
- Human serum albumin (human blood)
- Antipneumococci polysaccharide type 3 (bacterial origin)
- There was no immunologic effect on fraction 3 no matter what source they used/tested
- It was assumed that the fractions would have the same proportion of active molecules as the proportion of antibody in the y-globulin taken
- The power of the fractions to precipitate with goat antirabbit y-globulin serum was tested and I and II showed neither precipitation nor inhibitory activity
- When rat anti-y-globulin was tested with the fractions, rather different results from those with goat antiserum were obtained as fraction III precipitated 50% of the antibody, and I and II 15% each
- As the precipitation curves for I and II were almost identical, but III gives almost no precipitate, this emphasized the great similarity of I and II and their sharp distinction from III
- Porter stated that in view of the distinctive characters of the three fractions the possibility was considered that they might be synthesized separately and joined into the whole molecule in a separate step
- He decided after injecting a rabbit, bleeding it, and then radioactively labelling the precipitates, that there was no evidence to suggest that these fractions or any large parts of them are synthesized independently
- Ultracentrifuge studies of the splitting of y-globuilins from different species, by a variety of enzymes, all showed that a small number of fragments are the main products in each case
- He claimed that it was possible that in y-globulins, only small parts of the peptide chain are accessible to proteolytic enzymes, so that even though different enzymes break different bonds in these vulnerable sections the principal large digestion products are similar
- Porter stated that his experiments suggested that the 3 fractions were from one molecule split into separate sections yet only made up 90%
- Fractions I and II were so similar that the question arose whether they could be derived very largely from the same section of the chain
- Porter stated that fractions I and II were almost indistinguishable in N-terminal amino acid, amino acid analysis, molecular weight, antigenic specificity and, if derived from antibody, in their antigen-binding capacity (thus, he was unsure if these fragments even belonged to an antibody)
- Porter stated that the range in recovery of residues and recovery of individual amino acids was close to the range of expected errors due to the different measurements and the assumptions made in the calculations
- He stated that the simplest explanation for his results was that y-globulin was split into 3 different fractions
- Porter stated that the discrepancies with his earlier work was probably due to impurities introduced in the crude enzyme preparation
- He also stated that the unusual make-up of the y-globulin is due to his presumption of it being an antibody
- The significance of the part of the molecule that fraction III represented (as they were trying to fit it into a predetermined model) was unknown
- Porter admitted that his findings of the 3 fractions of antibody molecule was similar to Pauling’s antibody formation theory in 1940
- However, he was not certain what position the 3 fractions were in the make-up of the molecule
- In the summary, Porter posed the question “If Y-globulin contains antibodies” thus showing the uncertainty in his own findings
- If they are antibodies, Porter surmised that the y-globulin is made of up two similar, nearly identical fractions and one very different one
- According to Edelman, in studying the molecular structure of the y-globulins, one is confronted at the outset with the problem of whether these proteins consist of one or of several polypeptide chains
- The solution of this problem has significance in determining the chemical basis of antibody specificity, and in formulating a detailed theory of antibody production
- The results suggested that y-globulin molecules are composed of several discrete subunits or polypeptide chains linked by disulfide bonds
- Edelman reiterated that Porter demonstrated that rabbit y-globulin is cleaved by
papain treatment into three fragments which together form over 90 percent of the original molecule (what happened to the remaining 10 percent?)
- Edelman stated that the fragments obtained by Porter were not the same as those which he discovered
- He did, however, make the interpretation that Porter’s fragments may be made up of the smaller subunits he himself obtained
- Edelman felt his interpretation was strengthened by the finding that the two active antibody fragments obtained by treatment with proteolytic enzymes were linked by a single disulfide bond
- Edelman believed that his findings along with those of other researchers could be combined into a unifying hypothesis
- He stated that antibodies may contain two chains that are similar or identical in structure
- His model suggested an explanation for the wide weight ranges between the various molecules
- The finding that y-globulin has dissociable subunits may have a bearing on pathogenesis of disease (and then again, it may not…🤷♂️)
- A primary defect in macroglobulinemia and multiple myeloma may be due to a failure of specificity and control in the linkage of smaller units into larger ones
- Myeloma globulins may consist of a combination of subunits differing from normal y-globulin
- This may explain the antigenic and chemical differences and similarities between y-globulins of disease and normal y-globulin (in other words, he stated that antibodies are found in both diseased and healthy)
- Edelman claimed that his hypothesis was able to be tested experimentally (i.e. it was a guess that had not been proven scientifically)
- He admitted that only partial separation of the reduction processes was achieved
- Edelman tentatively concluded through his experiments that the subunits were linked by disulfide bonds
- He then created a hypothesis to relate the many structural features of the various y-globulins in terms of the multiplicity of the polypeptide chains in these molecules
In order to break down and identify the chemical structure of an antibody, logic would dictate that an antibody, like “viruses,” would need to be properly purified and isolated from everything else and shown to actually exist first. However, from 1890 when antibodies were initially dreamt up in the mind of Emil Von Behring on up to this point in 1961 when it was decided that these entities were chemically defined structurally by two different researchers, antibodies had never been seen. They were (and still are) nothing more than hypothetical constructs based off of the results from grotesque animal and chemistry experiments which are used to explain the theoretical concept of immunity. As antibodies had never been observed in nature nor in the fluids taken from a host, they have only ever existed as a concept rather than as a fully functioning particle. This is why decades of different researchers have come up with their own guesses as to how these entities looked, formed, and functioned. This is why both Porter and Edelman had to try and reverse engineer an antibody and then fit their results into a conceptual model. Antibodies have never been scientifically proven to exist. Thus, any claims that Porter and Edelman were able to deduce the chemical structure and form of particles never shown to exist which they themselves never witnessed is ludicrous. All these two researchers did was add their own paragraphs to the immunology chapter in the long-form fiction known as germ theory.