Pauling and Proteins: The Final Five Publications

Linus Pauling showing a molecular model to a young boy. 1950s.

[Part 3 of 3]

On March 31, 1951, Linus Pauling and numerous associates published seven revolutionary papers in a single issue of the Proceedings of the National Academy of Science. The research had been funded by the Rockefeller Foundation and carried out at the Gates and Crellin Laboratories of Chemistry, at Caltech. The first two articles: “The structure of proteins: Two hydrogen-bonded helical configurations of the polypeptide chain” and “Atomic coordinates and structure factors for two helical configurations of polypeptide chains,” have been discussed by us in the two weeks prior to this one. The remaining five will be described here in much shorter detail, as they are technical in the extreme.

The third article was titled “The structure of synthetic polypeptides,” and was written by Pauling and Robert B. Corey. The article claimed that the gamma helix and alpha-helix protein structures had forms that were also assumed by synthetic polypeptides. The authors discussed how the fibers of synthetic polypeptides had been analyzed using x-ray and infrared spectroscopy, which allowed them to determine the shape of the synthetic structures. Other scientists had also proposed the shapes of such structures, but Pauling and Corey rejected their hypotheses, as the structures the other scientists had proposed would have been “inherently unstable.” They concluded that their structure was the superior idea, and that while other structures potentially existed, they would be extremely difficult to measure due to their size.

The fourth article was more crucial to the narrative of protein structure that Pauling and his collaborators were weaving. The title was self-explanatory, and somewhat less technical than the others: “The pleated sheet, a new layer configuration of polypeptide chains.” In it, Pauling discussed how it had been long-believed that polypeptide chains are fully stretched and bound to adjacent, lateral chains of protein. He proposed instead a new idea, the so-called “pleated sheet.” In his suggested structure, the chains formed planes and certain bonds were arranged perpendicular to the planes of the chain, instead of coincidental with them. As a result, the chains are staggered and scrunched, instead of stretched in long, parallel lines. The rest of the article was devoted to the mathematics that Pauling had used to develop and explain the shape.

Feathers – specifically the atomic structure of feathers – was the topic of the fifth article, titled “The structure of feather rachis keratin.” The piece was written once again by Pauling and Corey and it analyzed rachis – a term with many meanings, but in this context referring to the central shaft of a feather – and keratins, which are structural proteins. The authors wrote that x-ray analysis of feather rachis keratin had shown the patterns of the polypeptide chains to be extremely complex, and notably shorter than expected. The rest of the article was spent explaining how the concept of the pleated sheet was mathematically relevant to feather rachis keratin.

Second to last was an article called “The structure of hair, muscle, and related proteins,” written by Pauling and Corey. In it, the authors pointed out that it had been many years since R.O. Herzog and Willie Jancke, in 1926, had made important x-ray photos of hair, muscle, nerve and sinew. Pauling and Corey felt that these photos, though revolutionary, were no longer adequate. Yet despite this deficiency, few modern attempts had been made to take better photographs. Two scientists named Lotmar and Picken had tried in 1942, but Pauling felt that their pictures were likewise not detailed enough. The Caltech researchers determined that their lab had found enough data though, and proposed structures for hair, muscle and “related proteins.”

This article differed from the other six in that it had an addition dated April 10, 1951. Written by Verner Schomaker, the addition revealed that subsequent research had shown that, while its basic premise was correct, the argument outlined on the piece’s first two pages was in fact wrong, and that the rest of the article hoped to amend that. Pauling and Corey argued that relaxed muscle was configured as a sheet, while contracted muscle formed an alpha-helix. The sheet configuration was inherently unstable relative to the alpha-helix, which made it easy for the hydrogen bonds holding the muscle in a sheet to break. This breakage allowed the polypeptide chains to coil and in turn made the muscle contract. The mechanism to prevent a chain reaction that might result in the sheet ripping itself apart during contraction was not understood, though Pauling had some ideas for that as well. The rest of the article was spent analyzing the amounts of energy released in frog muscle contractions to provide hypothetical amounts of energy expenditure and size for contractions in human muscle.

Representation of the collagen-gelatin molecule. April – May 1951.

The final article was “The structure of fibrous proteins of the collagen-gelatin group.” In it, Pauling wrote of his particular fascination with the protein in question:

Collagen is a very interesting protein. It has well-defined mechanical properties (great strength, reversible extensibility through only a small range) that make it suited to the special purposes to which it is put in the animal body, as in tendon, bone, tusk, skin, the cornea of the eye, intestinal tissue, and probably rather extensively in reticular structures of cells.

Another intriguing feature of collagen-gelatin was that it provided similar x-ray photos regardless of the source. In his article, Pauling noted that twenty-six samples, ranging from demineralized mammoth tusk to sheep gut lining, were all photographed by a scientist named Richard Bear and each resulted in remarkably similar images. Pauling compared them to a photograph of raw kangaroo tendon taken by Corey and Ralph W. G. Wyckoff, which also provided a view of what appeared to be the same structure. Pauling wrapped up the article discussing how three molecular chains wrapped into a distorted coil, and how the correlations between collagen-gelatin proteins and hydrogen could affect the structure.

The proteins work published by Linus Pauling and his Caltech colleagues in 1951 shook the scientific community and only added to Pauling’s growing fame. However, as time passed, evidence began to mount that his proposals regarding the gamma helix, muscle, and feather rachis were, in fact, wrong. Additionally, J.D. Bernal‘s lab found that the alpha helix, while fitting Pauling’s structural model, actually played a much smaller role in globular proteins than Pauling had suggested. However, Pauling’s media savvy and undeniable charisma won the day, at least in the short term. And so it was that, in the fall of 1951, (quoting Thomas Hager)

the 5 million readers of Life opened their new issues to find an enormous photo of Pauling, a big grin on his face, pointing to his space-filling model of the alpha helix. The headline read, ‘Chemists Solve a Great Mystery.’

Advertisements

Pauling and Proteins: Helices in the Air

Mounted models of the gamma helix and alpha helix, as housed in the Special Collections & Archives Research Center, Oregon State University Libraries.

Mounted models of the gamma helix and alpha helix, as housed in the Special Collections & Archives Research Center, Oregon State University Libraries.

[Part 2 of 3]

Linus Pauling sent shock waves through the scientific community when he published seven articles relating to the structure and function of proteins in the April-May 1951 issue of the Proceedings of the National Academy of Sciences. The first article of this volley was titled “The structure of proteins: Two Hydrogen-Bonded Helical Configurations of the Polypeptide Chain.” The second article was written by Pauling and Robert B. Corey, and was called “Atomic Coordinates and Structure Factors for Two Helical Configurations of Polypeptide Chains.” This paper was much more technical than was the first, and introduced two new important models developed by the Pauling group: the Gamma-helix and the Alpha-helix.

The article began by explaining in great detail Pauling’s idea for what he called the Gamma-helix (γ-helix). The γ-helix was the name assigned to the 5.1-residue helical configuration that Pauling, Corey, and Herman Branson had described in the first PNAS proteins article. The main difference between the γ-helix and the other configurations that they had proposed was that the bond angle between the C-N-C connection had been changed from 123˚ to 120˚. The authors explained that this small adjustment in the bond angle resulted in a miniscule change in the interatomic distances between various hydrogen bonds, but that these changes were significant enough to notably affect the number of residues per turn present within the structure.

51-residue

(It is worth noting that the unit used to measure the distance between molecules is the Angstrom (Å). One Å equals 10-10 m, or one ten-billionth of a meter. Considering the truly tiny sizes being measured, it is likewise worth noting that the changes between hydrogen bonds that Pauling was talking about were often measured in the thousandths of an Angstrom or ten-trillionths of a meter. One trillionth of a meter is known as a picometer.)

The article stressed that differences as small as 10 picometers could notably change bond angles, which would then change the number of residues per turn, thus dramatically affecting the shape of the helix. Next, the authors elaborated upon the likely arrangements of molecules within the helix due to symmetry or lack of symmetry in certain molecular bonds. Pauling and Corey further noted that they had used x-ray crystallography to validate their arguments and determine the crystal structures in question. From his earliest days as a scientist, Pauling had established himself as a major figure in x-ray crystallography, a technique by which an operator fires x-rays at a substance in question, then measures the way that the x-rays have deflected off of the substance. By analyzing these deflection patterns, researchers were then able to develop models of the shapes of molecular structures.

Once the γ-helix had been explained, the article moved on to discuss the Alpha-helix (α-helix). The group explained that the γ-helix and the α-helix were similar in terms of how the hydrogens bonded with other molecular groups, and how the residues fit under those configurations. They also detailed the exact interatomic distances between hydrogen and various other molecules in the structure, while pointing out that the distance between carbon and its other bonds determined the number of residues per turn. The number of turns, however, was variable; the smallest possible angle of 108.9˚ resulted in a residue of 3.6, while the largest possible angle of 110.8˚ resulted in a 3.67 residue.

37-residue

William Lawrence Bragg, an internationally famous scientist and proteins researcher of great import, was impressed by the α-helix, though he felt his rival Pauling to be overly excited about it. The α-helix was not a complete protein, except in a few cases including fibers, hair, and horn.  The structure also did not explain the functioning of proteins. As such, Bragg felt the paper to be an important first step – no more, no less. The rest of the scientific community was more enthusiastic than was Bragg and his team. Pauling received a Nobel Prize in 1954 in Chemistry “for his research into the nature of the chemical bond and its application to the elucidation of the structure of complex substances,” the α-helix being among the most famous of these “complex structures.” The National Science Foundation even named a research vessel The Alpha Helix in honor of the discovery.

Bragg was even less generous regarding the γ-helix. Though in his very carefully worded congratulatory letter to Pauling he did not say so, he felt the γ-helix to be far-fetched, perhaps existing only in Pauling’s imagination. While this was not the case, the γ-helix ultimately made less of an impact on the scientific community.

Regardless, the import of Pauling’s work was felt throughout the profession.  Though Francis Crick would write that the alpha helix did not give him and Jim Watson the idea that DNA was a double helix, he did suggest that “helices were in the air,” at the time “and you would have to be either obtuse or very obstinate not to think along helical lines.”

Pauling and Proteins: The Beginnings of a Revolution

Linus Pauling and Robert Corey examining models of protein structure molecules, ca. 1951.

[Part 1 of 3]

An article with the somewhat cumbersome title “The Structure of Proteins: Two Hydrogen-Bonded Helical Configurations of the Polypeptide Chain” appeared in the April-May, 1951 edition of the Proceedings of the National Academy of Sciences. The article was written by Linus Pauling, Robert B. Corey and Herman R. Branson, working collaboratively at the Gates and Crellin Laboratories of Chemistry at the California Institute of Technology and communicated to PNAS on Pauling’s fiftieth birthday. The article is immediately notable in that it is first of seven written by Pauling and his collaborators on the nature of protein and published in that single issue of PNAS.

And as it turns out, these seven articles were revolutionary. While the very act of mailing in all seven at once was audacious, their contents solved riddles that many researchers “believed would not be solved for decades.” Max F. Perutz, a competitor of Pauling’s in the field of biochemistry, read all seven papers in one morning, after which he immediately raced to his lab. Utilizing Pauling’s predictions, he was able to conduct experiments that validated the hypotheses proposed by the papers. He wrote to Pauling that

The fulfillment of this prediction and, finally, the discovery of this reflection in hemoglobin has been the most thrilling discovery of my life.

“The Structure of Proteins: Two Hydrogen-Bonded Helical Configurations of the Polypeptide Chain” was presented as the lead article in the April-May 1951 PNAS. The main question asked by the paper was: What is the structure of polypeptide chains – the backbones of proteins? Pauling, Corey, and Branson claimed that the best way to determine the answer to this question was to acquire an “accurate determination of the crystal structure of amino acids, peptides, and other simple substances related to proteins.” By determining the attributes of these components, which acted as the building blocks of polypeptide chains, the researchers could then make reasonable estimations of what the finished product would look like. Pauling’s group was specifically searching for the interatomic distances between molecules, the bond angles of the chemical bonds, and “other configurational parameters” fundamental to the structures.

Pauling and his colleagues felt that their work answered these questions and that they had determined the parameters satisfactorily. They then used their data to determine that their basic shape was a hydrogen-bonded, helical configuration. They wrote that “An amino acid residue (other than glycine) has no symmetry elements…” (In biochemistry, “residue refers to a specific monomer,” which is “a molecule that may bind chemically to other molecules…”) Because the residues lacked symmetry elements which would force the polymer chain into a symmetrical pattern, “…the only [possible] configurations for a chain…are helical configurations.”

Figures from “The Structure of Proteins: Two Hydrogen-Bonded Helical Configurations of the Polypeptide Chain,” 1951.

Furthermore, the group determined that, based on their calculations of bond angles, there were five possible angles for the helical turns; 165°, 120°, 108°, 97.2°, or 70.1°. Of these, only two, 97.2° and 70.1°, were “reasonable” potential configurations for the polypeptide chain, based on observed interatomic distances. Hydrogen, carbon, oxygen and nitrogen are the elements that make up the polypeptide chain. The chemical bond between C-O and C-N are both double bonds (they have four electrons instead of two). These tight bonds, along with the measured interatomic distances of other components, indicated that the interatomic distances for the chain would have to be small, otherwise the chain would very quickly become unstable. This is the reason that only the 97.2° and 70.1° configurations were acceptable. The other three angles were unstable and would have unraveled because the N-H bonds had too much space between them. Whether the helix turned at a 97.2° or a 70.1° rotation depended on the number of residues per turn in the chain. Pauling and his associates proposed either a 3.7-residue helix or a 5.1-residue helix.

The article ended by explaining why competing hypotheses on the shape of polypeptide chains were incorrect. The article specifically pointed out three hypotheses authored by William Astbury and Florence Bell, William Lawrence Bragg, John Kendrew, and Max Perutz, and Maurice Huggins as being inferior. The Caltech group asserted that each of their models assumed a set number of residues in the polypeptide chains instead of potential variables, and only gave rough estimates of interatomic distances and bond angles. While they all agreed that a helix was the correct shape, the specifics of all other helix models were incorrect because of these deficiencies.

This first paper was just a piece of the larger argument that Pauling was making. Each article was in itself useful, but only when considering the larger sum of the full publication bloc could the full import and implications of Pauling’s work be made visible. Pauling’s thinking proved to be revolutionary and controversial, as such ideas often are. William Lawrence Bragg, a key competitor of Pauling’s, was especially critical. He felt numerous of the Caltech group’s ideas to be outright false, and even the most solid of Pauling’s assertions to be just baby-steps rather than major breakthroughs. Pauling, naturally, disagreed.