Short and Simple (ish) Guide to X-ray Diffraction

  X-ray diffraction (or X-ray crystallography) was the chief physical method used to determine the structure of DNA. In this post, I will briefly and as simply as I can (which with my non-scientific background should not be a problem!) explain what x-ray diffraction technique is and its relative importance to the overall discovery.

What does X-ray diffraction actually mean?

 X-ray diffraction is the method of projecting a beam of X-ray radiation at a target object and through to a photographic film on the far side. A series of spots appear on the photographic film following this exposure, which is formed by the x-ray radiation diffracting off the structure that they passed through. These diffraction patterns give an indication of the general structure of the object (such as an inorganic crystal or macro- molecule such as DNA) which can then be delineated using complex mathematical formulas.

Why use X-rays in the first place?

The reason why X-ray beam is required in the first place is that atoms are too small (0.1nm between them, bearing in mind that 1 millimetre = 1000000 nanometres) to be revealed using visible light and therefore could not be viewed by a light microscope (even an electron microscope does not possess the required magnification). X-ray radiation fits the appropriate wavelength to be diffracted by the object and produce visible results.   

What causes the diffraction of the X-ray beams?

What the X-ray beam are diffracting is not the entire atom but the orbiting electrons (one of the component parts of an atom) that are close enough to the core (nucleus) of the atom to give a good indication of the structure of the unit cell (the term used for the repeating unit found in crystals and macromolecules). The end image is known as an electron density map of that unit cell. However due to the incredibly weak image a single molecule would produce, a crystalline structure is used instead, for example common salt (NaCl), since a crystalline structure provides a huge number of molecules arranged in the same orientation and therefore produces the same scattering effect on the X-ray beams. 

In this diagram, the diffraction of the X-ray beam causes an image with a helical arrangement to form as all the DNA molecules in a fibre are aligned in the same direction.

X-ray diffraction of nucleic acids at King’s College London from 1950 to 1953 

 X-ray diffraction studies on DNA began in June 1950 when Maurice Wilkins asked PhD student Raymond Gosling to assist him in diffracting the DNA fibre samples prepared by the Swiss biochemist, Rudolf Signer. Fibre diffraction did not usually provide good quality images because of the thinness of the fibres and therefore a very small mass to scatter the radiation. Nevertheless, the fibres’ remarkable uniformity when wetted allowed Wilkins to manipulate them into a bundle and mount them on a wire frame to obtain x-ray diffraction images. The initial images showed promise but Wilkins and Gosling were greatly assisted by J T Randall’s own experience with X-ray diffraction.  He advised how the surrounding air could affect the x-ray scattering. The solution was to pass hydrogen through the camera and control the relative humidity of the sample.  With this in place, the resulting images were much sharper and showed a clear crystalline diffraction pattern.

X-ray diffraction pattern obtained by M H F Wilkins and R Gosling in late 1950 showing a clear crystalline arrangement.

It was in late 1950 that the theoretical physicist Alec Stokes first noticed an interesting observation from the images. He realised that there was no diffraction at all along the length of the molecules: a sign that DNA might be helical.  However, the King’s College team needed far sharper images to confirm this hypothesis. This required a new X-ray camera that could work on single fibres.  Through a fortunate coincidence, Werner Ehrenberg and W E Spears had just developed one at Birkbeck: this was generously loaned to the King’s College team.

Before the new camera was set up, it was decided that Rosalind Franklin, who was joining the laboratory from Paris, would replace Wilkins in producing the x-ray diffraction images with the continued assistance of Raymond Gosling. Both Stokes and Wilkins continued working on the problem with the latter embarking on some rough tests with the old X-ray diffraction camera on various DNA specimens that produced an observed “X” crossed pattern. The X pattern of diffraction was created by the x-ray radiation scattering at right angles off the "zigzag" structure of the DNA chain.  This interpretation was further supported when Franklin and Gosling produced the first “B” structure X-ray patterns in the late summer of 1951. This was a crucial development as it showed two observed states of DNA: crystalline “A” and semi-crystalline “B” (the best B structure diffraction photograph became known as “Photo 51”). The photos also supported the predicted observed readings of a helix that Alec Stokes had developed using the mathematical technique known as Bessel functions.

Plot of Bessel Functions for a smooth helix, named "Waves at Bessel-on-sea" by Alec Stokes who completed the calculations for the diagram over a single train journey.

It was now Maurice Wilkins and Rosalind Franklin disagreed over the direction of the research in finding the overall structure. Wilkins was keen on hypothetical model building while Franklin favoured a more systematic study of the structure. This parting of ways can be partially explained as stemming from the limitations of the x-ray diffraction process itself. For example, the evidence from the photos clearly pointed to a helical structure but this begged question: what type of helix? Helices in nature could occur in single, double and even triple strands and there was no clear indication, which was the right number. This is why the King’s College London attempt at model building proved to be a failure when the model made by Bruce Fraser showed a triple helix based on the chemical readings but was unable to fit with the rest of the x-ray data. A crucial piece of the puzzle was missing and related closely to DNA’s function of providing the genetic material for life: it was only when Jim Watson and Francis Crick came up with the base pair hypothesis that the double helix seemed the ideal form. 

In this diagram, we can see the general similarity between a single and a double helix.

X-ray diffraction studies undertaken at King's College London provided part of the experimental structural data needed to solve the general structure of the DNA double helix. Yet, as important as these observations were other methods and disciplines were of equal importance in unravelling the overall structure, in particular the biochemical work of Erwin Chargaff and the biological insight of Jim Watson. X-ray diffraction work on DNA at King’s did not finish with the unveiling of the structure in March 1953 but continued for another decade as Wilkins and his team worked to test to the correctness of the "Watson-Crick" model. 

A potted account of the research on DNA at King's College London

In this post, there will be a brief overview of the research carried out here at King's on the structure of DNA. This of course is not a definitive history of the events and interactions which led to the discovery of a double helix. For a more comprehensive scientific history of DNA, I would recommend Robert Olby's book "The Path to the Double Helix: The Discovery of DNA" (1974). My purpose is to introduce some of the techniques and findings that occurred here at King's and relate it to the overall contribution to identifying the structure of DNA.

DNA enters the 'Circus'

The new Biophysics Laboratory created by J T Randall at King's College London was a lively and unconventional institution. Randall's ambition to create a laboratory where physicists could work on biological problems and vice versa were being realised at King's and an assortment of young research workers from diverse scientific backgrounds found their feet in these new disciplines. Maurice Wilkins was one of these workers. His first years at King's had proved fruitless in the field of ultrasonics causing mutations in fruit flies and he swiftly moved on to construct, with his colleagues Bill Seeds and K P Norris reflecting achromatic microscopes which he began to use on ultraviolet and dichroism studies on the Tobacco Mosaic Virus (TMV), nucleic acids and nucleoproteins. During these initial experiments with DNA Wilkins found a significant finding he wrote in 1962:

"While examining oriented films of DNA prepared for ultraviolet dichroism studies, I saw in the polarizing microscope extremely uniform fibres getting clear extinction between crossed nicols...each time that I touched the gel with a glass rod and removed the rod, a thin and almost invisible fibre of DNA was drawn out like a filament of spider's web. The perfection and uniformity of the fibres suggested that the molecules in them were regularly arranged"

Polarising microscope view of DNA fibre stretched at room humidity

The excellent quality of this extracted DNA was not through chance. The sample was supplied by Rudolph Signer, a Swiss biochemist who had been since the 1930s endeavouring to produce high quality extracted DNA. The sample made in 1949 with his student H Schwander made that grade. On the 12 May 1950, Signer was invited to the Faraday Society in Cambridge to discuss his work on preparing DNA samples. At the end of the talk he distributed bottles of his best DNA and Maurice Wilkins was one of the lucky recipients. Wilkins later reflected that this was "a generous act in the best tradition of science!".

Having witnessed the remarkable uniformity of the fibres, Wilkins took the DNA fibres to Raymond Gosling for X-ray diffraction. Gosling was the only person using X-ray diffraction techniques at the time to complement Randall's interest in the X-ray study of ram's sperm heads. Initially the two were unsuccessful  the specimen to the film distance too large and the X-ray tube too weak to yield a pattern but they improvised as Raymond Gosling explains:

"I wound these fibres around a wire frame, forming a dense bundle which on the conventional Raymax tube produced a diffraction pattern recordable in a few hours. If that sounds rather scientific, I must tell you that the 'wire frame' was simply a bent paper clip and the 'dense bundle' was formed by applying Lepages quick setting cement, purchased from Woolworth's in the Strand!"

First multifibre specimen taken on the Raymax tube Unicam Camera, filled with hydrogen (1950)

The above photograph is the one that Maurice Wilkins showed at the Naples conference that so captivated Jim Watson. This was obtained thanks to a suggestion by Randall , by passing hydrogen thought the camera and sealing it where possible to prevent air scattering, which caused a fog on the film. From these images it was possible to demonstrate that the molecule was packed together like cylinders 2.0 nm in diameter and  that the structure was very crystalline. The final study in this initial period was experimenting with the water content of the molecule. By drying and heating the specimen they obtained an amorphous scatter pattern .This contrasted when they repeated the process but wetted the hydrogen at 90% humidity for 12 hours prior to exposure and obtained a crystalline pattern similar tp the example above. This demonstrated that water played a vital role in maintaining an ordered crystalline structure.

X-ray Diffraction Years

In June 1950, the old wartime Siemens X-ray tube broke down leaving the department without a working X-ray diffraction camera. They soon obtained designs for new fine-focus X-ray equipment produced by Ehrenberg and Spear at Birkbeck College. Whilst the camera was being constructed Randall came to the decision that they needed a professional crystallographer to keep the work progressing. Rosalind Franklin was already on her way to the department as a research fellow to work on proteins but Randall expressed in a letter the change in orientation. The letter stated that "as far as the experimental X-ray effort is concerned there will be at the moment only yourself and Gosling..." and  gave no indication of Wilkins' continued involvement on the project. This may account for the grievance that Franklin held from what she viewed as Wilkins interpreting her problem but this issue is in no way definitive and has been heatedly debated along with the wider recognition of Franklin's role ever since. Yet, it is best to leave the issue of Franklin and Wilkins relationship to one side for a moment and recount for what was actually achieved when she joined the department.

One of the first achievements of this collaborations was a vital one in solving the structure of DNA. Franklin brought her expertise to the job by fixing the humidity and the water content of the exposures by passing the hydrogen through saturated aqueous solutions of appropriate compounds through which the hydrogen could bubble into the camera at any given temperature. They soon found that the sodium salt of DNA supplied by Signer could transform into two forms, Structure A and Structure B.

Soon after this discovery, the division between the DNA workers at King's was cemented with Franklin and Gosling continuing to work with the fine focus X-ray tube using Signer DNA to outline the Structure A pattern of DNA whilst Wilkins and Alex Stokes used the old Raymax camera and work on the Structure B pattern using Erwin Chargaff's DNA samples. This situation did not change until Franklin left King's College London in February 1953 with virtually no communication between the two groups.

In October 1951 Wilkins, who had been reading Linus Pauling's famous paper on the protein alpha-helix , wondered why Pauling had not calculated the X-ray diffraction of the structure. After discussing the matter with Stokes he came back the next day with a Bessel function calculation of diffraction of a helix. The remarkable aspect of the 'Waves at Bessel on Sea' diagram was how much it corresponded to the new B diffraction patterns that Franklin was achieving. Franklin reacted furiously to her results being interpreted and the matter was set aside.

The following month November 1951, saw the unveiling of two DNA models: one by Bruce Fraser at King's and the other by Jim Watson and Francis Crick in Cambridge. Both these models were three chained helixes and lacked the key base-pair element. Fraser's model (as described in a previous post) was a closer approximation of the correct version as a fundamental flaw in the first Cambridge model was that the helix was inside out with the bases on the outside due to Watson misjudging the water content. The failure of the Cambridge model put a temporary injunction on the pairs DNA interest, whilst model building was not pursued at King's College London after Fraser left the department shortly after this. 

At the start of 1952, Franklin, taking the advice of a Paris mentor, Vittorio Luzzati , decided to elucidate the structure of the A pattern using the crystallographic method of cylindrical Patterson function. This laborious method was a way of calculating the Fourier transform of the intensities of the spots on the X-ray films ,and involved measuring different reflexions of the specimen which required a new tilting microcamera to be designed for this process. In order to calculate the Fourier transform, Franklin and Gosling had to consult Beavers and Lipson strips (pictured below). Ray Gosling recalled that:

"These assembled the values of the periodic functions all set out at appropriate intervals and arranged sequentially in a handsomely polished mahogany box. I used to have nightmares...that I had dropped a box of 'strips' on the floor and had to sort them all out in the correct order!"

By the end of 1952, Franklin and Gosling had the preliminary results back for the cylindrical Patterson function of Structure A. Although in hindsight the data from the cylindrical Patterson and then the 3-dimensional Patterson analysis supported a double helix in the A form such a conclusion was not reached by Franklin who before leaving for Birkbeck College had begun to investigate the B form of DNA (with which, as shown in her notebooks, she would come close to solving the structure with). 

The solution to the structure:

On the 7th March 1953, James Watson and Francis Crick finished the model of the double helix. The Cambridge pair started model building again after Watson was inspired on account of being shown 'Photo 51' by Wilkins when he visited King's on the 30th January. Watson deduced that a double helix rather than a triple helix fitted with genetic transference and was supported by the biochemical work of Erwin Chargaff who had discovered that the quantities of the base pairs were equal. On the 12th March 1953 the King's team were invited up to view the model. Wilkins wrote of seeing the model: 

"...a feeling came through to me that the model, though only bits of wire on a lab bench, had a special life of its own. It seemed like an incredible new-born baby that spoke for itself, saying 'I don't care what you think - I know I am right' "

                                                              

 The diagram above on the left hand side shows some of the essential features of the double helix from the original paper by Watson and Crick such as the two sugar-phosphate chains running in opposite directions linked together by hydrogen bonded base-pairs stacked on top of each other. The diagram on the right shows a wire-model of the double helix used by Watson and Crick in their representation of the original double helical model of DNA.                     

Conclusion:

The contribution of the the Biophysics department at King's to the discovery of the structure of DNA was vital. The x-ray diffraction studies and other experimental methods provided the essential properties for Watson and Crick to elucidate the structure.Yet these achievements were not in isolation and needed to be combined with the knowledge acquired from Pauling and Chargaff along with many others to lead to the structure of DNA. The cracking of the structure should not be seen in terms of a race but the culmination of advances in chemistry, biology and physics spanning back to the nineteenth century when Fritz Miescher extracted DNA for the first time. 

Fortieth Anniversary of the Discovery of the structure of DNA. Pictured from left to right are four of the five named workers featured on the commemorative plaque (exception being Rosalind Franklin) they are: Ray Gosling, Herbert Wilson, Maurice Wilkins and Alec Stokes.

                                    

DNA Story at King's: The Hidden DNA workers

Fortieth Anniversary at King's College London of the discovery of the structure of DNA in 1993. Pictured are four of the five names commemorated in  the grey plaque on the background wall ( from left-right they are: Raymond Gosling, Herbert Wilson, Maurice Wilkins and Alexander Stokes)

On the unveiling of the grey plaque to commemorate the fortieth anniversary of the discovery of the Double Helix at King's College London, Maurice Wilkins said these words:

"I'd like to emphasize that my presence in front of this plaque is to emphasize all five names there, including that of Rosalind Franklin who is not able to be present".

This magnanimous gesture was not just based on politeness and modesty but reflected the key collaborative effort required to provide the experimental data needed to crack the structure of the double helix. In this blog, I will provide a little background on some of the key collaborators at King's during the early years of the DNA work at King's who have been somewhat overshadowed in DNA history due to the grand narratives of Jim Watson and the biographers of Rosalind Franklin.

Raymond Gosling (born 1926):

Raymond Gosling is relatively well known in the DNA story because of his collaboration with Rosalind Franklin on the X-ray crystallography of DNA. However, Gosling's role in the DNA story pre-dated Franklin's arrival at the lab and it is this work in collaboration with Maurice Wilkins which was also of great importance to the discovery of the double helix. He first joined the lab as a PhD student under the supervision of John Randall in 1949 and began working on the cell nucleus. This approach soon led to Gosling working on studies involving DNA and by 1950 Randall asked Gosling to gather information on ram sperm using X-ray diffraction. These fuzzy pictures were in rough accordance with the X-ray diffraction pictures of DNA by Astbury. The introduction of Signer DNA and the expert manipulation by Wilkins to obtain DNA threads allowed Gosling to improve these initial results and produce X-ray pictures that showed DNA's crystalline structure. With help from Randall, Gosling and Wilkins were able to produce the "Structure A" form of DNA through the bubbling of hydrogen through the camera to prevent air scattering. In early 1951, Rosalind Franklin joined the lab and Gosling was transferred to work under her in a crystallographic analysis of DNA. The crystallographic work of the two provided a key component to obtaining the structure of DNA by vastly improving the crystallographic images and distinguishing structures A and B of DNA. There collaboration continued until Rosalind Franklin left of Birkbeck College in early 1953 and is thankfully well documented  due to the survival of Rosalind Franklin's experimental notebooks (found at the Churchill Archives Centre, Cambridge) and also several articles the two published in Nature and Acta Crystallographica.

Alexander Stokes (1919-2003):
 

  Whilst Gosling was the diligent lab worker, Alex Stokes was the theoretician who provided the crucial mathematical interpretations of the x-ray diffraction studies in order to guide the King's team in the right direction. Stokes was one of the initial appointments into the unit by Randall and came with valuable experience of X-ray crystallography from his time at the Cavendish Lab in Cambridge during the war. What distinguished Stokes from his other colleagues was his consummate ease in translating the patterns created on an X-ray diffraction film into a description of the atomic arrangement using his skills in mathematics. For example, it was Stokes who first noticed in 1950 that the X-ray diffraction photographs of DNA gave an indication of a helical structure by the absence of diffraction along the length of the molecule. Stokes later used complex mathematics in the form of Bessel functions to underline this, by famously working it out on a single train journey from his home in Welwyn Gardens City to London before christening the diagram, " Waves at Bessell-on-sea".

"Waves at Bessell-on-sea" by Alexander Stokes

Herbert Wilson (1929-2008):

Though Herbert Wilson arrived six months before Watson and Crick unveiled the double helix to the world, his work with Maurice Wilkins uncovered essential information regarding the structure of DNA. Wilson arrived at the biophysics unit in September 1952 under tenure of the University of Wales. He soon began X-ray diffraction studies of DNA, nucleoproteins and cell nuclei under the guidance of Maurice Wilkins. The two collaborated on a number of investigations beginning in the autumn of 1952 comparing, under different humidities, different samples of DNA (such as pig thymus, squid sperm, and wheatgerm DNA). Their observations confirmed what Franklin and Gosling had concluded that the phosphate groups were found on the outside of DNA. They then extended the study to look at the effect of undried preparation of live trout sperm to support the hypothesis that the drying process had no affect on the in vivo structure which it subsequently confirmed. The importance of these comparative studies was affirmed by the growing number of samples that were collected that not only showed para-crystalline patterns but also the A-type of DNA. This indicated that the crystalline appearance of DNA was not laboratory induced but occurred in biologically active samples and that the work of Franklin and Gosling would have universal applications.

The Molecular Configuration of Nucleic Acids", twenty six people along with organisations that contributed. Along with those mentioned already they also include from King's Sir John Randall, Bill Seeds, Bruce Fraser, Geoffrey Brown, Gerald Oster, Watson Fuller and Struther Arnott.