In this study, we present a novel utrophin promoter activation assay, which we have used to screen a library of approved drugs and natural compounds. After initial screening, hit confirmation and independent validation, we have identified a lead compound, nabumetone, that is a potential therapeutic candidate for DMD.
The utrophin promoter activation assay performed well in tests of robustness, with a Z-factor of 0.6 and % CV under 10%. The number of hits as a percentage of the library (1.8%) was comparable to other published screens 
, suggesting that the assay was specific with a low number of false positives. Nevertheless, we had set the initial screen threshold of 20% upregulation low enough to avoid false negatives, in the expectation that some of the initial hits would be false positives due to ‘statistical noise’. Consistent with this, only 14 out of 20 initial hits (70%) were confirmed upon dose-response testing.
In screening assays that use luciferase as a reporter, false positives can arise from compounds that act as luciferase inhibitors. These compounds bind to and stabilise luciferase in cells, increasing its levels, and are then competed off by the substrate in the luciferase assay reagent, such that an artifactually high luciferase activity is produced 
. Thus, it is important to independently confirm the effects of the hit compounds on the endogenous utrophin A promoter, mRNA and protein. We did this for one candidate, nabumetone, using a TaqMan qRT-PCR assay for utrophin A mRNA and Western blotting for utrophin protein. Validation of the remaining compounds is ongoing. However, definitive in vitro
validation experiments are challenging, in part due to the differences in utrophin protein expression compared to the in vivo
situation, where utrophin is enriched at specific locations, such as neuromuscular junctions 
, that do not exist in cultured cells. To move our findings closer to the clinic, it will be essential to determine the efficacy of the compounds in vivo
, using animal models of DMD.
Previous studies suggest that an increase of approximately 2-fold in utrophin protein in muscle is sufficient for correction of the dystrophic phenotype in mice 
. In our study, we identified several compounds that could upregulate the utrophin A promoter up to 3.5-fold. Independent validation of nabumetone showed that it could increase endogenous utrophin A mRNA levels approximately 2-fold, and increase utrophin protein by 1.2-fold. This is extremely promising given that even a very small upregulation of utrophin appears to have a beneficial effect in dystrophin deficient mice 
A variety of potential therapies for DMD are being investigated, and some have reached clinical trials (http://www.clinicaltrials.gov/
). While this is greatly encouraging, there are still many obstacles to be overcome before all patients with DMD can be treated successfully. In many cases, problems of delivery, safety and large-scale, cost-effective manufacture have not yet been resolved. Some approaches, such as antisense oligonucleotide-mediated exon-skipping and nonsense codon suppression, are only applicable to subsets of patients with particular types of dystrophin mutations 
. Regulatory body approval may also be more complicated for new kinds of drug molecules such as proteins and oligonucleotides. For example, under current regulation, each of the potentially hundreds 
of mutation-specific exon-skipping oligonucleotides would be treated as separate drugs 
. It is also important to consider that, initially at least, combinations of treatments may be needed in order to achieve therapeutic efficacy. Therefore, the continuation of research along multiple therapeutic avenues, including utrophin upregulation, is of great importance to ensure the development of effective therapies for all patients with DMD.
There are a number of advantages to small molecule-mediated utrophin upregulation that make it both a strong candidate for DMD therapy and a complimentary approach to those discussed above. The introduction of dystrophin protein into the body of a DMD patient where it has never been expressed could provoke an immune reaction against the protein, which might be recognised as ‘non-self’. Utrophin is expressed in muscle in the foetus and at high levels in other tissues such as liver, lung and kidney throughout life in DMD patients (as well as healthy people) 
, so increasing its production therapeutically in muscle would not risk inciting an immune response. Additionally, the utilisation of the endogenous utrophin gene provides an elegant solution to problems arising from the large size of the dystrophin coding sequence (14 kilobases) 
, which makes it difficult to incorporate into viral vectors, except in truncated form. Finally, the use of a traditional ‘drug-like’ small molecule, with favourable absorption, distribution, metabolism and excretion properties 
, to upregulate utrophin offers obvious advantages in terms of delivery, stability and bioavailability.
Drug repositioning, the exploitation of existing drugs for new applications, is becoming an increasingly important part of research and development for the pharmaceutical industry 
. Our approach of screening a library of regulatory body-approved drugs and natural compounds offers a distinct advantage in terms of the speed and efficiency of future therapy development. All the hits identified in our screen are compounds that have been shown to be safe in humans, and for which pharmacokinetic data is available. This eliminates a significant proportion of the time and expense required when developing a novel compound as a drug, and gives the potential for a rapid progression from the lab to the clinic.
To date, we have validated one drug, nabumetone, at the mRNA and protein level, in C2C12 cells. Nabumetone is a COX-1/COX-2 inhibitor that shows a preference for COX-2 inhibition in vitro 
. It is used for the management of pain and inflammation in osteoarthritis and rheumatoid arthritis 
. Nabumetone is generally well-tolerated by patients 
, and its anti-inflammatory activity might be beneficial in DMD, since inflammation is a component of the disease 
. There is some evidence that the use of selective COX-2 inhibitors may increase the risk of adverse cardiovascular events, especially in patients who already have an increased risk 
. This is less of a concern with COX-1/COX-2 inhibitors, possibly because COX-1 inhibition has an antiplatelet effect, which may protect against thrombotic events 
. Nonetheless, because of the involvement of the heart in DMD pathology, the safety of nabumetone use in this group would need to be carefully evaluated.
In developing our assay, we used as positive controls a number of substances already known to upregulate utrophin: heregulin, TSA, okadaic acid and L-arginine. Of these, only L-arginine has been used in human beings, as a supplement. There are some safety concerns about its use, particularly at high dosages 
. Its use in DMD patients has not been investigated.
In terms of doses, it is not possible to directly compare in vitro and in vivo doses without considering pharmacokinetics; however, using a crude calculation based on an average total human body fluid volume of 42 l, the optimum dose for nabumetone determined in cell culture (25 µM) lies far below that used in human beings (equivalent to approximately 100–200 µM). As a comparison, L-arginine was effective in activating our utrophin upregulation assay at 2 mM, whereas doses used in humans would correspond roughly to 0.2–5 µM.
In our experiments, we observed a smaller increase in utrophin at the protein level than at the mRNA level. Although these experiments were done at different time points to allow for the expected slow synthesis of the large utrophin protein (approximately 400 kDa), this difference may also reflect the regulation of utrophin at the translational level. Indeed, it is known that utrophin expression is influenced by post-transcriptional mechanisms, acting via the 5′- and 3′-untranslated regions (UTRs) of the utrophin mRNA 
. It may be that by combining drugs that activate the utrophin promoter with therapies targeting points of post-transcriptional expression control, or therapeutic substances such as biglycan that promote localisation and stabilisation of utrophin at the sarcolemma 
, a far greater upregulation of utrophin can be achieved.
In conclusion, we have taken a novel approach to the problem of DMD therapy by screening existing drugs for utrophin promoter activation. Following the successful screening project and independent validation presented here, the lead compound nabumetone will be tested in preclinical trials for its ability to upregulate utrophin in vivo and improve the phenotype of dystrophic mdx mice. This venture offers great promise for the rapid development of an effective drug therapy for DMD. However, we caution that although nabumetone is an FDA-approved drug that is used safely in human beings, it will still be important to conduct thorough preclinical studies in animals, as well as clinical trials, to determine the safety and efficacy of its long-term use in DMD.