From genetic fingerprinting to the genetics of fingerprints! Kellie's project

It was very early on during my first year at the Liverpool UTC, that I recognised the power of inviting students to design their own research projects. Now, as I approach the end of my original two year secondment to the UTC, with the REAL (Research Enhanced Active Learning) programme in place, I wanted to highlight another project (the first one, Jack's project was covered in an earlier post, and there will be more to follow). Before I do, let me say a few words about the concept of research itself, as defined by the OED it is: 

"The systematic investigation into, and study of, materials and sources in order to establish facts and reach new conclusions"

What an unsatisfactory definition. Let's try somewhere else! If not Oxford, then Cambridge:

"A detailed study of a subject, especially in order to discover (new) information or reach a (new) understanding"

Better...or even better, from Merriam-Webster:

"studious inquiry or examination; especially :  investigation or experimentation aimed at the discovery and interpretation of facts, revision of accepted theories or laws in the light of new facts, or practical application of such new or revised theories or laws"

The latter is also given by Wikipedia, in its discussion of research, and I think it covers the concept pretty well. So with a "formal definition" in hand, let's look at some examples of research and how the word is used. You will have heard someone say: "I am going to research my family history", or maybe: "I fancy a career in research"; or perhaps "use your iPads to research the topic and then write a proposal". So when Kellie and I discussed possible research projects what did she think I meant and what did I expect her to do!

After a short discussion in which I explored Kellie's interests, we settled on the the chemical and genetic origins of fingerprints, or FPs for short (I wont list the rejected topics!). Fingerprinting was first used to successfully convict Harry Jackson for the unlawful removal of a set of billiard balls from a house in Denmark Hill, London (an area where I spent many enjoyable weekends as a young student!) in 1902. Interesting as the subject may be, at that point in our discussion, I realised I had a very superficial appreciation of FP genetics (as opposed to genetic fingerprinting!). So this was going to be a journey for both of us. And, on reflection, I realised that this was how I prepared for my PhD project over the summer months of 1980, as I earned some cash, by sweeping the changing rooms at a well known ICI plant. Getting back to Kellie and I deciding on the first steps: as the cartoon depicts, this was a case of "the blind leading the blind".

A quick search on the internet led me to the work of Alan Turing and his ideas, published in 1952 on chemically induced morphogenesis, in the period just after his great achievements at Bletchley Hall (and if you haven't seen Benedict Cumberbatch's portrayal in the "Imitation Game": you must!). As an aficionado of the molecular mechanisms underpinning epigenetics, I thought this was an interesting topic and, surprisingly, one that has yet to be resolved in molecular terms. So what was the research question? After some discussions and reading around, we settled on "How is the observation that genetically identical twins have different fingerprints (see top LHS), best explained at a molecular level".

This is how the project has evolved over the last school year, as Kellie juggled time devoted to class lessons with work on her project. The first suggestion I made was that she see for herself how easy/difficult it is not only to "take" a fingerprint, but to take one that gives a clear pattern, that might potentially be used to identify or eliminate someone in a criminal situation. After some time the finger prints were pretty good and it was easy to see that they were unique, but not after quite a number of inky smudges had been collected.

Unfortunately, we were unable to find two identical twins to provide fingerprint samples, and so we had to rely on the published evidence. At this point I should make it clear that this is standard practice in Scientific Research, however there is often some overlap between the observations that you make for yourself, with those made by others. It may be that you want to assess the magnitude of a particular measurement, or you may want to alter the pH, temperature etc. (for a legitimate reason). Given the historically accepted view on the differences between FPs from identical twins, together with contemporary data, obtained  with more sensitive instrumentation, we both felt it was appropriate to rely on the body of literature that exists today, since there were many, independent observations drawing the same conclusion.

The second phase of investigation concerned the Turing hypothesis. Turing had suggested from a perspective drawn from his world of physics, chemistry and mathematics, that the differential diffusion of a soluble molecule in the amniotic sac, would reach the boundary of each of the two developing embryos at two different times. If this molecule could trigger expression of one or more genes (determinants of FP patterning) in a time dependent manner, then it would be possible for a given local concentration of such a molecule, to "fix" the levels of  expression (for example) of the FP genes, in a unique way. Each embryo would then produce a distinctive fingerprint at the end of the developmentally regulated patterning process. 

Mathematically, the probability of producing a unique finger print has been discussed since Victorian times: for example, in the early, elegant work of Francis Galton FRS he discusses the metrics of human characteristics (take a look at the Galton Institute web site and the series of articles that appeared in Nature papers and Royal Society publications obtainable here). More recently (and I do not claim to have carried out an exhaustive literature search) I noticed a paper by the mathematician, Jim Morrow, from The University of Washington (Seattle) which you can read here. In it, Morrow derives the probability of finding any individual sharing identical FPs (alive or even throughout human history) to be one in 10 raised to the power 36, ie so small as to be impossible: hence the metrics, providing a good FP sample has been collected, appear to be unassailable.

Going back to the consideration of diffusion, by adding a dye into a simple aqueous vessel, and monitoring the time taken for the die to equilibrate throughout the vessel, Kellie demonstrated that a 50ul sample of dye, dispensed into 500ml, took around 15-20 minutes. Clearly the amniotic sac would not be completely still and the shape of the glass vessel, was hardly biomimetic! Nevertheless it allows an approximation to made. And, using this information Kellie asked the final question: how long does it take to express a gene in vivo? Again simplifying the system, if a gene is activated when a specific molecule exceeds a threshold level, and if following transcription, it is spliced and translated, how long does this typically take in vivo? Kellie is now drawing her experimental data together with her literature search and it is emerging that there is a good match between diffusion rates and the time scale of gene expression (even when under some form of epigenetic modulation), to support the original Turing hypothesis.

I think this kind of project, which is perfectly simple to carry out with even the most basic level of facilities, is given a strong intellectual focus by the historical and contemporary literature relating to the core question of the mechanisms underpinning the uniqueness of fingerprints. Equally, I can't think of many other projects that draw on so many aspects of Science: Maths, Physics, Chemistry and last but not least, Biology! As we return next week for the last term before summer, Y12 students will be developing their new projects. I can't wait to see what they will come up with. Oh yes, and why are we not doing this at a much earlier stage at undergraduate level?

Bar code of a killer!

Last night I watched the concluding part of the ITV drama "Code of a Killer", in which the Leicestershire police (led by DCS David Baker, played by David Threlfall) sought the help of (now Sir) Alec Jeffreys (played by John Simm) to identify the rapist and murderer of two young girls in the early 1980s. I discussed the background to the science in an earlier post, but here is a quick (and personal) summary of the Scientific "climate of technology and discovery" surrounding molecular genetics in the late '70s-early 1980s.

 

Whilst methods were developed as log ago as 1974 for gene sequencing (thanks to Fred Sanger, Walter Gilbert and Allan Maxam), by 1982, molecular biologists had access to the genomes of just two bacteriophages, along with the (rather modest, by today's standards!) human mitochondrial genome. And just in case you are too young to remember, finding a computer in a Biology lab in the early 1980s, to analyse the sequences, was truly exceptional. Most high profile molecular biology labs at this time had their sights firmly trained on the common features of gene sequences (which is of course of fundamental importance), however, when Alec Jeffreys arrived at the University of Leicester, he set out to explore the differences between the genes of closely related species. Around this time, Nobel Prizes were awarded for the discovery of the tools that Alec Jeffreys would employ in his work (Arber, Nathans and Smith: Restriction Enzymes) and work on the immune system had first produced monoclonal antibodies (Jerne, Kohler and Milstein) and then Tonegawa was recognised for his work on the genetic origins of antibody diversity. It was some time later that Sharp and Roberts were independently rewarded for their work in the late 1970s on the discovery of split genes (exons and introns), but such events were timely for Jeffreys to lay the foundations for DNA fingerprinting, or profiling. I should also mention that David Botstein had used restriction enzymes to expose subtle variations in genes, referred to as RFLPs (restriction fragment length polymorphisms), arising through the loss or gain of a restriction site through mutation. Finally, I should probably point out that whilst gene sequencing was now becoming popular in many research labs, it was largely a manual process and it would be some years before it would become automated to a level that large genome sequencing projects could be considered feasible. In fact, the method of choice for molecular analysis of DNA (and RNA) was Southern (or Northern) Blotting, the former technique being developed by Ed Southern, even though Kary Mullis had discovered the technique for PCR in 1980. At this point, I hope (presumably in vain) that I haven't left out any important discoveries! 

The challenge that Alec Jeffreys faced, is one of the most fundamental aspects of genome science: on the one hand Darwinian evolution leads us to look for the similarities between genes in say mice and men. However, what is important in forensics (and paternity testing) are the elements in our genomes that make us different. These are the "stutters" that are mentioned in the drama. The image on the left shows Alec holding an autoradiograph revealing a series of DNA fragments which have been "highlighted" through the use of a radioactive "probe" designed to "pick out"  DNA sequences complementary to the probe itself. In this way a scene of crime sample can be unequivocally identified as belonging to the suspect, or (importantly) NOT. Recall that the phenomenon of base pairing is a key component of the structure of DNA in which the bases G and C and A and T form "complementary" pairs. So, if a fragment of DNA contains the sequence 5'GATTCCGGATTCA3' (for example), then the probe sequence 5'TGAATCCCCGGAATC3' would "hybridise" to it. If the probe was radioactively (or fluorescently) labelled, then the complementary sequence separated on a gel (agarose for long fragments and polyacrylamide for short fragments of DNA) can be "visualised" using a suitable film or detector. When Alec Jeffreys is shown in episode one, trying various probes to explore similarities and differences in genes, his focus on the seal myoglobin gene is shown because this is where he obtained the first high quality data, from a highly variable sequence, linked to a common core sequence within an intron in the myoglobin gene. 

The second part of the drama begins with a press conference in which the police announce that their prime suspect has been excluded by genetic fingerprinting! There are few applications of science that have been so closely intertwined with the police and the legal profession. Ten years later, the US sports celebrity O.J. Simpson would be famously acquitted of the murder of his wife, following a high profile trial in which the evidence from genetic fingerprinting was dismissed, not because the science was suspect, but rather the "audit trail" of sample collection and analysis was shown to be unreliable. What I particularly liked about the second episode of Code of a Killer, is the recognition (and conviction) shown by DC Baker, that Scientific evidence will provide the truth. Moreover, we mustn't underestimate the importance of Alec Jeffreys' success in communicating this so effectively to the police officer. The other parallel I like in the drama, is that between the work of a principal scientist (PI) and that of a chief investigating officer (CIO): both involve the systematic collection, evaluation and rigorous testing of data. DC Baker's leadership of his team will be a familiar tale to many PIs, as he navigates  through misleading results (false leads) and the challenges of funding and competing priorities (budgets and politics!) to finally apprehend the killer (or more usually clone the gene!).

I am not sure that this production will receive the accolades reserved for great TV drama, but it does join a handful of productions in which the profound value of science to society is showcased effectively. I also managed to get over the somewhat hackneyed references to the "committed scientist burning the midnight, oil whilst neglecting his family", in episode one. In fact, the scene in the second part, in which Alec Jeffreys persuades the community to submit to testing, by holding aloft his own genetic fingerprint alongside that of the killer, is a lovely moment that in my view, fully vindicates the pursuit and public support for Blue Skies Science. If you have been fortunate enough to spend an evening chatting with Alec, or have been present at one of his seminars, you will undoubtedly share my view (and that of the people of Leicester, who awarded him the freemanship of the city in 1992!) that he is not only an outstanding scientist but an exceptional human being. I will leave you with the thought that occurred to me when Alec Jeffreys aligned the autorads of the sample taken from Colin Pitchfork and that recovered from a victim: Alec Jeffreys' greatest legacy may be to Human Justice and not Science!

The science behind DNA fingerprinting

This week's TV drama "Code of a Killer" is perhaps best described (by one critic) as "workman-like". I can't comment on the accuracy of the portrayal of the policeman, but the role of Alec Jeffreys, or rather the depiction of a Molecular Biologist at that time, is more  clichéd than I would have  a liked. However, I think the Science has been explained pretty well. Although it can't compete with Daniel Craig's explanation of "Uncertainty" in Michael Frayn's "Copenhagen" when he (as  Werner Heisenberg) meets Niels Bohr; or Tim Piggott-Smith's masterful explanation (as Francis Crick) of the significance of Rosalind Franklin's diffraction data (above) in "Life Story" (the drama behind the determination of the double helical structure of DNA). 

The challenge that Alec Jeffreys faced, is one of the most fundamental aspects in genome science: on the one hand Darwinian evolution leads us to look for the similarities between genes in say mice and men. However, what is important in forensics (and paternity testing) are the elements in our genomes that make us different. These are the "stutters" that are mentioned in the drama. The image on the right shows a series of DNA fragments which have been "highlighted" through the use of a radioactive "probe" designed to "pick out"  DNA sequences complementary to the probe itself. In this way a scene of crime sample can be identified as that of suspect 2. Recall that the phenomenon of base pairing is a key component of the structure of DNA in which the bases G and C and A and T form "complementary" pairs. So, if a fragment of DNA contains the sequence 5'GATTCCGGATTCA3' (for example), then the probe sequence 5'TGAATCCCCGGAATC3' would "hybridise" to it. If the probe was radioactively (or fluorescently) labelled, then the complementary sequence separated on a gel (agarose for long fragments and polyacrylamide for short fragments of DNA) can be "visualised" using a suitable film or detector. When Alec Jeffreys is shown trying various probes to explore similarities and differences in genes. His focus on the seal myoglobin gene is shown because this is where he obtained the first high quality data, from a highly variable sequence, linked to a common core sequence within an intron in the myoglobin gene. 

Let me try to simplify. While many labs in the late 1970s were studying the generic properties of genes, Alec Jeffreys (left) was interested in human genetic variation. For example, in the USA, Richard Roberts and Phil Sharp were about to announce their Nobel Prize winning discovery of introns; that is non-coding sequences of DNA that are interwoven between the exons (coding sequences) in the human (and most eukaryotic) genomes. Back in Leicester, Alec Jeffreys was more concerned with obtaining an experimental handle on those regions in our chromosomes, where there had been (over evolutionary time) less selective pressure to maintain integrity, i.e. where variation in sequence was and is tolerated. These regions, often called mini-satellites are typically flanked by common sequences. Hence, if you can design a "probe" to reveal the number of  copies of a repeated sequence, or one that will pick out the common element and it bring with it the number of variable sequence, then using restriction enzymes (recall these are the sequence-specific DNA cutting enzymes) or more recently PCR-based strategies, it is possible to compare individuals in such a way that the differences are highlighted. Importantly, as Jeffreys proved, when you get this right, these differences are sufficiently distinctive and statistically robust, making identification (or importantly in the case highlighted in the drama) elimination of a suspect both possible and definitive. If you recall the scene in the drama where the investigating officer is in his office with his second in command and Alec Jeffreys explains the experimental data: the scientist must be completely confident in advising the police that their prime suspect is not the culprit. Today this genetic "bar coding" forms a major element in any identification process, ranging from crime scene investigations, through parental disputes to the tragedy of the identification of victims in warfare or, as we have recently witnessed, the passengers on Germanwings flight 4U9525. 

For me Alec Jeffrey's work and its application, represents one of the finest examples of how "Blue Skies" science can improve our world. Not surprisingly, Professor Sir Alec Jeffreys has since then, been given many accolades. In my view, there couldn't (in my view) be a more deserving scientist.

Molecule of the Month for April 2015: Alchemase

Structure This month I have chosen the enzyme Alchemase as the subject of my post. The general structure of the enzyme is somewhat different than most enzymes, and moreover, its mechanism of action is unique. Through a series of redox steps, alchemase catalyses the only triple-cofactor dependent transmutation reaction so far identified in living organisms. The other striking feature of alchemase that is novel, is that it is a nucleozyme. Following the demonstration by Szostak and Joyce (see open access publication here), several years ago, that not only RNA, but DNA can provide the structure and reactivity, normally thought of as the preserve of proteins, alchemase was the first example of a naturally occurring DNA enzyme to be isolated. There are many RNA enzymes known, but so far only one deoxyribozyme has been structurally and mechanistically, characterised. The 3 cofactors, which are all variants of Pyridoxal Phosphate, associate with the exposed bases, seen projecting from the central core, which itself comprises the deoxyribose phosphates. Cofactor 1 is purine specific, cofactor 2 is pyrimidine specific, while cofactor 3 seems to show a dual specificity. Each cofactor has a distinctively different redox potential when measured in standard biological buffer at pH7, this is thought to be pivotal to the energetically unfavourable reaction. 

Source The organism from which alchemase was first isolated is Clostridium seaborgii, an obligate anaerobe, found (typically) in the oral cavities of adults over the age of 60, particularly those with a history of high carbohydrate diets. When the carbohydrate diet is supplemented with Bismuth (usually derived from over the counter medication such as Pepto Bismol, well known in the USA, but available now in the UK in a chew formulation at many high street pharmacies), the alchemase is induced. Induction does not follow the usual lac operon model, but occurs through a mechanism in which a site specific nuclease (similar to Type II restriction endonucleases, but with a dependency on Bismuth instead of Magnesium), releases multiple copies of the DNA sequence from the C.seaborgii genome. The alc gene (encoding alchemase), is present in multiple copies, similar to ribosomal RNA genes. 

Chemistry It is during cleavage reaction that Bismuth ions are first activated through cofactor 1 (and subsequently by cofactors 2 and 3), as part of a sequential conversion of the Bismuth nucleus to Gold. The reaction was formally demonstrated by Glenn Seaborg (from where the organism gets its name) over 30 years ago.  Through systematic proton and neutron transfer reactions, the Bismuth (tightly complexed via the phosphate backbone of alchemase), is converted to Gold. Recall from your chemistry that Bismuth (83) is in the post-Transition Metal cluster, whilst Gold (79) is just inside the Transition Metal group. It might not seem like a big jump from Bismuth to Gold, but the energy required is probably equivalent to a day's worth of ATP, confined to the volume of a typical unicellular microbe. So whilst the reaction is therefore pretty slow, compared with many enzymes (and anaerobes are generally known to have much slower doubling times than aerobes), the energy per cubic metre (assume a sphere of radius 1 micron for the anaerobe) is equivalent to that in the  core of a nuclear reactor. The key role of the cofactor triad in this reaction is currently the subject of a major investigation at the Aprilscherztag Institute fur Radiochemie in Berlin through funding by the German Green Party, who are keen to establish an alternative organic programme to solve Germany's energy crisis in the wake of their decision to axe conventional nuclear power. I for one hope that this process can be harnessed soon, since renewable energy is such an important problem to solve globally. 

Applications of alchemase in Synthetic Biology. One of the most promising ideas emerging from the discovery of the alc gene and its enzymatic properties, is the possibility of inducing in situ dental repairs. As you will all no doubt be aware, gold has been shown over many years to be one of the best metals for dental repair. Whist it is still popular in some cultures, the woman on the right is from Tajikistan, where gold teeth are a status symbol, in Europe and the USA, gold has been replaced by materials that match the "ivory" colour of most teeth. The use of metals in the form of amalgams, which combine lead, mercury or increasingly "composites", remains commonplace for dental repairs at the rear of the mouth. Here is where the use of the alc genes comes in. By transplanting the genes (multiple copies, see earlier) for alchemase (based on similar concepts related to the recent CRISPR phenomenon), together with that encoding the alc specific restriction enzyme (CseI), scientists at the Goldgraber Zentrum in Augsburg, Germany have shown evidence of nutrient induced gold deposition by their synthetic microbe. I thought "Synthia" (the name given to Craig Venter's first synthetic microbe) was a bit cheesy, but I quite like the name "Midas"! Whatever your opinion of the name, with the discovery of this novel nucleic acid enzyme, I am certain some interesting fundamental science and subsequent applications will be developed over the coming years that will prove transformative! Finally it would be remiss of me not to acknowledge the help of several colleagues in writing this Blog, in particular Professors Geber, Magnus, Paracelsus and not forgetting many helpful discussions with the late Professor Newton, who is was always a great inspiration to me and whose ideas I have always felt, were not only worth their weight in gold, but bring a certain gravitas to any scientific discussion: he is fondly remembered and affectionately missed, but did live to a ripe old age.