Journal of an Orthopaedic Researcher's Diary for Action Arthritis
Profiling gene expression of proteinases in Dupuytren's Disease
Phil Johnston, Specialist Registrar

Over the last year I have been studying Dupuytren's Disease of the hand, and its relationship with a family of chemicals named matrix metalloproteinases (MMPs). This work has been partly funded by Action Arthritis

Background

Dupuytren's Disease (DD) is named after the Frenchman who described it. DD is a common condition which presents as clawing or contractures of the fingers. It is due to tissue under the skin becoming like scar tissue being thickened and knotted. As the contracture worsens, the affected finger cannot be straightened. This means that it gets in the way and catches during some activities such as putting the hand in a pocket. When this severe it is operated to remove this thickened scar-like tissue and straighten the affected digits.

The MMPs are a family of chemicals whose primary action is digestion although not in the gut. They are responsible for the breakdown of extracellular matrix. These structural and supportive proteins which normally keep cells together and limit their movement. There has been a huge amount of research into MMPs as there is strong evidence for their involvement in cancer. Since cancers involve the abnormal movement of cells, MMPs help this movement by removing the extracellular matrix. Therefore giving drugs that stop MMPs is one way that has been tried to stop cancers spreading. In fact, in several trials, patients treated in this way developed a condition in the hand closely resembling DD.

However, the link between MMPs and DD has not been properly explored so far. Several groups have investigated some of the MMP, but not all. Professor Ian Clark's laboratory, within the BioMedical Research Centre at the University of East Anglia (UEA), is at the cutting edge of research into MMPs and related chemicals. They have made important discoveries for MMPs in cancer and also arthritis. Combining this expertise with the help of the orthopaedic surgeons at the Norfolk & Norwich University NHS Trust (NNUH) was an obvious choice.

During this year, my salary was obtained in part from the Institute of Orthopaedics in Norwich and also from work as Joint Replacement Fellow at the NNUH. However, the funding enabling me to perform the research (paying for all chemicals used, etc.) was by way of a generous grant from Action Arthritis, without which the project could not have been undertaken.

August 2005

The project starts! In fact, even though this represents the beginning of my year's sabbatical from Specialist Registrar training, we have been up and running for 6 months. During the first few months of the year, we produced our experimental protocol, obtained local Research and Development and national Research Ethics Committee approval (both essential to allow research on human tissue samples to be undertaken), have written numerous grant applications (and even attended one interview) all unsuccessful and collected over half of our tissue samples for the disease group (patients with Dupuytren's Disease of the palm).
The first month sees my induction into Professor Ian Clark's laboratory in the School of Biological Sciences at the UEA. I learn basic lab techniques; How to collect tissue samples in theatre and preserve them for lab analysis and how to keep cells alive in culture without introducing infection (an ever-present problem). I begin collecting tissue samples from our control group patients without Dupuytren's, who have agreed to donate a piece of tissue during their carpal tunnel release (a condition that causes hand pain), to act as a benchmark for comparison. Carpal tunnel release is a procedure to remove pressure from a nerve at the wrist: during the operation, there is the opportunity to remove a piece of the tissue being incised. This collection process sees me walking to and from UEA and the NNUH across the sports fields with a small flask of liquid nitrogen (to store samples) and the various chemicals and test tubes required for storage.

September 2005

During the first month, but more so during September and October, I am learning how to extract RNA (the genetic message contained within all cells, closely related to DNA) from the tissues collected. I am acting under the guidance of Rose Davidson, a PhD student of Ian Clark, who is running a project on similar lines to mine in cartilage and synovium from patients with arthritis (also partially funded by Action Arthritis). Certainly the RNA extraction technique ought to be transferable from one human tissue-type to another? Try as I might, the methods, with minor adjustments, are depleting my stocks of normal tissue without producing "pure" RNA. The frustration is compounded, because only at the end a long day's work am I able to quantify the amounts and purity of the result both of which are far below that which is acceptable for downstream experiments. The lab are very supportive that's science for you! and this is why scientists are generally unflappable. You need the patience of a saint and a good grasp of the bigger picture, not to be put off by minor set-backs in obtaining results.

October 2005

Every member of the lab presents their data regularly in turn, for discussion amongst the others. With only early data (mostly negative), I give a largely clinical talk about Dupuytren's Disease and Carpal Tunnel syndrome, and we discuss my efforts to produce pure RNA. As a result, I put the tissue RNA extraction on the back-burner and work with an expendable source (cancer cells in plentiful supply) to see whether my failure so far can be attributable to properties of the tissue, details of the technique, or just operator factors (i.e.me!). During the process I am learning more lab techniques, including how to "run a gel" (electrophoresis) and how to grind tissue to a powder by placing it in liquid nitrogen and shaking it violently. Finally, with the aid of a "spin column" (a highly sophisticated filter), I am able to obtain reasonable amounts of highly pure RNA. We are back in business.

November 2005

The tissue is too gristly to break up simply by pounding it with ball bearings (a common technique). We turn to a machine called the Ultra Turrax, which is effectively a miniature blender with an extremely sharp blade. The extraction reagent contains phenol, a highly caustic chemical with the ability to burn. All work is therefore done very carefully in a fume hood to minimise risk. As I progress with this technique - success: all of the RNA produced remains pure and in large amounts. Once all 60 samples have been processed (each one takes about 3 - 4 hours), I can use the RNA in a chemical reaction to produce complementary DNA (this process is called the Reverse Transcriptase reaction). This is the first step towards quantifying the genes within the samples and produces a much more stable compound to work with, while the RNA has to be stored at -80oC to prevent it from decaying, the cDNA is stable at -20oC and can even be left in the fridge for days at a time without problem. This month sees my first attendance at a British Society for Matrix Biology meeting (BSMB) at the Kennedy Institute for Rheumatology in London. I hear talks by various experts from around the world, all related to matrix metalloproteinases. By now, I have obtained a working knowledge of the subject. The talks are not all Greek to me (as they would have been in August)! I also attend the British Society for Surgery of the Hand's annual meeting (BSSH) at the Royal College of Surgeons in London, with several talks on Dupuytren's Disease, but none investigating our field, or using our approach. It seems that we remain the leaders in this area.

December 2005

I am introduced to a new machine, called TaqMan. This puts the samples through another chemical reaction (called Polymerase Chain Reaction, or PCR) during which the relative amounts of the gene of interest can be measured. The reaction uses a digital camera-style optical receiver to detect when the fluorescence of a "probe attached to the specific gene is detectable. PCR doubles the amount of gene per cycle of the reaction. After 40 cycles, I calculate therefore that the gene has been amplified about 1,000,000,000,000 times from the starting amounts! If the machine cannot detect fluorescence after this amount of amplification, then presumably there is none of that gene within that sample. Also, a gene detected after only 15 cycles is clearly much more highly expressed than one which takes 30 cycles to detect. Each run of the PCR takes 2 hours. As December draws to a close, the weather gets colder. I even spend a night or two in Norwich, so I can get the first reactions going at 6am and leave the last ones to start when I leave at midnight. Having got so far, I want to complete this stage by the end of the year!

January 2006

The TaqMan machine has produced an enormous amount of data. I have looked at the expression of about 50 genes in 60 samples, which thus forms an "expression profile" for this tissue. This will obviously need analysis and interpretation to put some sense to it all. At about this time, I begin my studies of research methodology and application of statistical methods to clinical data, enrolling on several courses at the UEA (modules from their MSc programme), and attending lectures at the NNUH. Fortunately, with the gene profiling now finished, I can concentrate on the huge amounts of homework that these courses generate. The experimentation is put on hold for a month.

February 2006

The expression profile has thrown up many genes of interest, some of which were never previously known to be important in Dupuytren's disease. The TaqMan system works using a series of fluorescent "probes" which are specific to identifying individual genes (the power of this approach). I learn how to create such a probe, specific for a protein called "alpha smooth-muscle actin", a chemical produced preferentially by the cells involved in the disease process. This is tested in the TaqMan machine, although in this case, there appears to be little difference between disease and normal tissues in the amounts of this protein produced.

March / April 2006

My year will finish at the end of July and I will return to Specialist Registrar training in Addenbrooke's Hospital, Cambridge. Ian Clark has strong connections with a senior scientist in the Rheumatology Research Unit in Cambridge, Dr Graham Riley. Graham and his team have developed a technique to study the contraction behaviour of cells from patients with ruptured Achilles tendons. We think this would be a useful technique to emulate in our Dupuytren's cells as well. As the experiments will almost certainly run over into the autumn (or longer!), we decide to move the second phase of the project to Cambridge. I transport frozen stocks of cells, and a large crate of tissue culture paraphernalia, to Addenbrookes, and look around my new home for the next few months. I present the data for the first time at the East Anglian Registrars' Prize presentations in Cambridge: well received and with positive comments to improve the talk for future presentations. My second presentation is for the Medico-Chirurgical Society in Norwich, and a poster of the work is presented at the annual congress of the BSMB in Cambridge, again to a favourable reception.

May 2006

Lectures progress at the UEA, with the associated coursework. I continue my operating at the NNUH and keep up with Ian Clark's group. At present in Cambridge I am expanding the stocks of cells by allowing them to fill their culture flask (achieve "confluence") and then dividing them 1 in 3 to repeat the process and treble the yield. Once I have frozen some stocks down for future use, I can begin the experiments, looking at contraction behaviour of the cells, when they are grown in a 3-dimensional collagen lattice (collagen is one of the proteins of the extracellular matrix, the skeleton offering support to cells in organs). The premise is that the most active cells (hopefully those from the patients with DD) will display most contractility in this artificial collagen lattice, and that this model will be representative of the cells' behaviour in patients' hands. Once their behaviour has been studied by adding stimulants to contraction, the plan is to add inhibitors to some of the MMP genes identified in the expression profile and to observe whether this alters their behaviour (especially if this reduces the cells' tendency to contract). This model could theoretically then be applied to patients, perhaps as an adjunct to surgery, to reduce the rate of disease progression. The first few gels do not form correctly, and the members of the Riley lab collectively scratch heads to work out why.

June 2006

Despite a fungal infection wiping out half of my cell stocks in the incubator (thankfully I had frozen some spares to replace these), I have now got the contraction gels up and running, by tweaking the amounts of chemicals in the mix. Initial results appear to show a difference between the Dupuytren's cells and those from the patients without disease, with the Dupuytren's cells contracting faster than the "normals". This will need to be corroborated by greater numbers to demonstrate a real effect rather than a one-off. The final few statistics lectures have opened up new avenues in analysis of the data from the expression profile, including some multi-variate analysis techniques that are worth pursuing. I present the data to the entire BioMedical Research Centre staff at the UEA (probably the most nerve-racking talk as these guys really know what I'm talking about). My presentation in the Norfolk & Norwich Benjamin Gooch lecture for research performed by doctors in training wins first prize.

July 2006

One further presentation, this time at the British Orthopaedic Research Society annual meeting in Southampton. I must have presented a water-tight case, as no one as able to ask a question to pole-axe me. As I write, with three weeks left before my return to a more structured timetable, I am a changed man. I have gone from relative ignorance in many areas of research (although, more worryingly, I was often unaware of this lack of knowledge) to a position of firm foundation, having trained in the methodology of clinical research, learned some techniques of molecular and cell biology and become more confident in presentation. Also the expression profile we have produced is unique and original, representing an important first for our research collaboration between UEA and NNUH.

The next year's work will progress at a slower rate (needing to be dove-tailed in with routine clinical duties) and the thesis for my MD will need to be completed. The data will continue to be presented, certainly at the British Orthopaedic Association annual congress in Glasgow and hopefully also at the BSSH in London this autumn and possibly further afield. The first paper, detailing the expression profile, is almost complete to be submitted for publication. This autumn will also see me returning to Norwich for the final few 1-year follow-up assessments of the Dupuytren's patients. We hope to be able to correlate their clinical outcome following surgery with the expression profile, which would of course lead to the use of future tissue samples in prognosis and directing treatments, perhaps leading to more aggressive therapy in the cases more likely to recur.

There follow some pictures for interest.

Me in action in the cell culture laboratory gloves and lab coat - and within a laminar flow hood (similar to operating theatres) to reduce infections.

Graph showing the different amounts of gene expression for the three sample groups (yellow is Dupuytren's 'cord' (less active); red is Dupuytren's nodule (active); and blue is normal samples. Obvious differences can be seen between groups.

Microscope photograph of Dupuytren's cells in culture - forming long whorls which resemble fingerprints, and typical of such cells, called 'fibroblasts'.

An example of the readout from the contraction gel experiments. The vertical columns represent different concentrations of cells per gel; the top two horizontal rows have had the stimulant added and the bottom two rows have not. The differences between contraction of the discs can clearly be seen within the pink culture fluid.