Correlation of Utrophin Levels with the Dystrophin Protein Complex and Muscle Fibre Regeneration in Duchenne and Becker Muscular Dystrophy Muscle Biopsies

Duchenne muscular dystrophy is a severe and currently incurable progressive neuromuscular condition, caused by mutations in the DMD gene that result in the inability to produce dystrophin. Lack of dystrophin leads to loss of muscle fibres and a reduction in muscle mass and function. There is evidence from dystrophin-deficient mouse models that increasing levels of utrophin at the muscle fibre sarcolemma by genetic or pharmacological means significantly reduces the muscular dystrophy pathology. In order to determine the efficacy of utrophin modulators in clinical trials, it is necessary to accurately measure utrophin levels and other biomarkers on a fibre by fibre basis within a biopsy section. Our aim was to develop robust and reproducible staining and imaging protocols to quantify sarcolemmal utrophin levels, sarcolemmal dystrophin complex members and numbers of regenerating fibres within a biopsy section. We quantified sarcolemmal utrophin in mature and regenerating fibres and the percentage of regenerating muscle fibres, in muscle biopsies from Duchenne, the milder Becker muscular dystrophy and controls. Fluorescent immunostaining followed by image analysis was performed to quantify utrophin intensity and β-dystrogylcan and ɣ –sarcoglycan intensity at the sarcolemma. Antibodies to fetal and developmental myosins were used to identify regenerating muscle fibres allowing the accurate calculation of percentage regeneration fibres in the biopsy. Our results indicate that muscle biopsies from Becker muscular dystrophy patients have fewer numbers of regenerating fibres and reduced utrophin intensity compared to muscle biopsies from Duchenne muscular dystrophy patients. Of particular interest, we show for the first time that the percentage of regenerating muscle fibres within the muscle biopsy correlate with the clinical severity of Becker and Duchenne muscular dystrophy patients from whom the biopsy was taken. The ongoing development of these tools to quantify sarcolemmal utrophin and muscle regeneration in muscle biopsies will be invaluable for assessing utrophin modulator activity in future clinical trials.


Introduction
Duchenne muscular dystrophy (DMD) is a lethal inherited muscle wasting disease caused by mutations in the dystrophin gene that disrupt the open reading frame, preventing production of a functional dystrophin protein [1]. The absence of dystrophin protein from the muscle fibre membrane results in progressive fibre degeneration. A milder allelic form, Becker muscular dystrophy (BMD) is usually caused by in-frame mutations in the dystrophin gene, resulting in the synthesis of reduced levels of a possibly partially functional dystrophin [2,3]. Possible treatment strategies in late stage clinical development include the use of antisense oligonucleotides (AONs) to skip mutated dystrophin exons, allowing the production of a BMD-like dystrophin [4][5][6]. One drawback of this approach is that the AONs are exon-specific, so one AON is applicable only to a subset of patients carrying specific mutations. Another genetic strategy utilises small molecules to prevent premature protein termination. This approach will allow for a full length dystrophin to be synthesised and will be appropriate for around 10-15% of all DMD patients. Another promising approach in pre-clinical development that has been tested in dystrophin-deficient mice and dogs [7,8], uses adeno-associated virus (AVV) to insert a gene coding for a truncated dystrophin into muscle fibres. However, the major challenges facing AAV gene delivery is the titer necessary to achieve bodywide transfection in the human, the immune reaction against the AAV vector precluding re-administration and how functional the highly truncated dystrophin will be long term. Utrophin, an autosomally-encoded homologue of dystrophin, is first expressed in skeletal muscle during fetal development [9]. Dystrophin and utrophin are both expressed at early gestational stages but by birth in humans utrophin in normal human muscle is confined to the neuromuscular and myotendinous junctions and blood vessels [10]. In dystrophic muscle utrophin is present on regions of fibre repair. [11]. Consequently in some muscular dystrophies, including DMD and BMD, utrophin is present at the muscle fibre sarcolemma due to the significant regeneration taking place, in addition to the up-regulation in DMD and BMD that is present in the absence of dystrophin [9,11]. Utrophin expression is driven by two different promoters, A and B, which regulate the expression of two transcripts utrophin A and B, which have unique expression patterns. The A isoform is expressed in skeletal muscle and present at the sarcolemma of normal fetal muscle fibres, and on both regenerating and mature fibres in DMD and BMD muscle. As with dystrophin, utrophin specifically binds to the costamere in order to stabilize the sarcolemma from contraction-induced damage. Utrophin B is more ubiquitous, being found in most other tissues and cell types [10,12]. Knockdown of utrophin in dystrophin-deficient mdx mice causes a considerable worsening of the muscle pathology and function [13,14], suggesting that the presence of utrophin may ameliorate the phenotype of dystrophin-deficient muscles. It also has to be noted that there are some functional differences between dystrophin and utrophin where dystrophin localizes nNOS to the membrane and plays a role in microtubule organization. However there are a number of BMD patients lacking the nNOS binding site in dystrophin, who remain mildly affected and ambulant, suggesting that nNOS tethering at the sarcolemma is not an absolute requirement in order to have a mild phenotype [15,16]. There is therefore considerable interest in utrophin up-regulation as a possible therapeutic option for DMD, as this strategy may have potential for all DMD patients, regardless of their dystrophin mutation. Studies in the mdx mouse showed promise with marked reduction in myofibre degeneration and improved phenotype following transgenic up-regulation of utrophin [17]. Further work has demonstrated that a small molecule utrophin modulator, SMT C1100, increased utrophin in mdx mice skeletal muscle and diaphragm, leading to a significant decrease in pathology and increased functional benefit [18]. SMT C1100 is currently in clinical development having been shown to be safe and well tolerated in a Phase 1 healthy volunteer study [19]. To determine the efficacy of any therapeutic intervention to up-regulate utrophin, reliable and reproducible quantification of utrophin levels is essential. We have developed a method to quantify sarcolemmal utrophin in frozen muscle biopsies, using methods used previously in our laboratory to quantify intensity of sarcolemmal associated proteins [20]. We quantified utrophin in both mature and regenerating fibres in muscle biopsies from patients with BMD, DMD, carriers of DMD or BMD and non-pathological controls. The dystrophin-associated protein complex (DAPC) complex plays a major role with dystrophin or utrophin to link the extracellular matrix to the actin cytoskeleton. All members of the DAPC are reduced in DMD [21][22][23] and induction of dystrophin by antisense oligonucleotides leads to increased sarcolemmal expression of DAPC proteins [24,25]. We therefore also investigated whether utrophin plays a role in the stabilisation of members of the DAPC, by correlating the levels of utrophin with β-dystroglycan and ɣ-sarcoglycan levels. If utrophin levels can be maintained at the sarcolemma of a dystrophin-deficient myofibre with a utrophin modulator drug, this should protect the fibre from undergoing necrosis and as a result, reduce muscle regeneration [26]. Therefore we developed a second complementary assay to quantify the percentage number of regenerating fibres in a biopsy, and to help validate the assay we compared the amount of regeneration in BMD and DMD muscle biopsies. Regenerating fibres were identified by staining with antibodies to developmental and fetal myosin as markers of fibre regeneration. Both antibodies were used as a cocktail in order to unequivocally identify all recently regenerating fibres. Serial sections were stained with a monoclonal antibody to beta-spectrin (Novocastra NCL-SPEC1-CE 1:100) for 1 hour followed by biotinylated anti-mouse antibody and streptavidin conjugated to Alexa Flour A594 as described above.
All incubations were carried out at room temperature. Controls, in which the primary antibody was omitted, were performed (S2 Fig). A second control, in which the second primary antibody to β -dystroglycan was omitted, shows that the secondary Alexa Fluor A594 goat anti mouse IgG2A antibody did not cross-react with the mouse IgG1 antibody used to detect utrophin (S3 Fig).

Intensity measurements
Immunolabelled sections were evaluated using a Leica DMR microscope interfaced to Meta-Morph (Molecular Devices, Downingtown, PA, USA). Intensity measurements were obtained using Metamorph software following capture of 4 randomly selected fields, selected out of focus at x20 magnification. An area of stained muscle section taken from an 8 year old DMD patient (P15; Table 1) expressing high levels of utrophin was used to set the exposure times and images of all sections in one experiment were captured at the same time. A control muscle section (C2; Table 1) was used to set ɣ-sarcoglycan and β-dystroglycan exposure times. The intensity was measured by randomly placing 10 regions of interest (ROI) in each image using Metamorph software capturing 40 fibres. If the ROI was in the centre of the fibre, on an area of fibrosis, on a neuromuscular junction, or if more than one measurement per fibre was selected, the region was moved slightly to the nearest fibre membrane. The measured regions included a portion of the cytoplasm and the sarcolemma. In this way, 40 ROI on a total of 40 fibres was captured. No two ROIs were placed on the same muscle fibre. Very small fibres, with a mean diameter of 6μM or less, which were similar in size to capillaries, were excluded from the analysis. Fetal and developmental myosin intensity measurements were taken from the sarcoplasm of fibres as indicators of fibre regeneration using the same method and all intensity measurements were recorded as described previously [20]. Fibre diameter measurements were taken from sections that had been stained with utrophin to delineate fibre membranes, by measuring the diameter of the lesser side of the fibre using Metamorph software [27]. For each region, the minimum intensity value recorded (representative of the cytoplasm or background intensity) was subtracted from the maximum intensity value (which corresponded to the sarcolemma) to correct each measurement for background intensity. To correct for variation of sarcolemmal integrity between samples, we performed the same measurements on serial sections stained with the monoclonal antibody to beta-spectrin. The spectrin intensity values obtained for the control samples were set as the standard to calculate normalization factors.
For each of the antibodies, the minimum intensity value was subtracted from the maximum, then these values (one for each of the 40 fibres analysed) were normalized with the beta-spectrin measurements and plotted on a graph.
There was only background sarcolemmal intensity in sections in which the primary antibodies were omitted (S1 Table;

Data analysis
The mean ± standard error of mean (SEM) is represented for normally-distributed data, whilst the median and interquartile range (IQR) is shown for data not normally distributed. A t-test or one-way ANOVA was used to determine differences between groups. Post hoc comparisons were performed with the Tukey test and if data did not meet the assumptions of ANOVA a Kruskal-Wallis one-way ANOVA test was used. Correlations were performed using Pearson's or Spearman's test. Statistical analysis was done using Graph Pad Prism and significance level was set at p 0.05.

DMD patient biopsies have significantly higher utrophin levels compared to BMD patients
Intensity quantification methods were used to quantify utrophin on mature muscle fibre sarcolemma in sections of quadriceps muscles (Table 1). Utrophin levels were normalised to spectrin stained on serial sections.
To determine reproducibility of the method, utrophin quantification on mature muscle fibres was repeated on serial sections from the same control, BMD and DMD muscles (Fig 2b;  Table 2). There was little variability in the measurements from BMD samples between the two experiments and no significant differences between utrophin intensity in each experiment for 4 of the 7 DMD samples. However a Wilcoxon matched-pairs signed rank test revealed significant variability between replicate experiments for the remaining three DMD patients P9 (p = 0.0025), P10 (p = 0.0219) and P14 (p = 0.0417) (Fig 2b; Table 2).
Regenerating fibres have higher utrophin levels than mature fibres in both BMD and DMD skeletal muscle biopsies Next, we compared sarcolemmal utrophin levels between mature and regenerating muscle fibres (determined by the presence of developmental and fetal myosin) in sections from 3 BMD patients from 4 to10 years of age (P5, P6 and P7) and 6 DMD patients from 4 to 8 years of age (P9, P10, P11, P12, P14 and P15) by co-immunostaining utrophin with a cocktail of fetal and developmental myosin antibodies. Myosin and utrophin intensity readings and fibre diameter measurements were recorded in fibres that were positive and fibres negative for fetal and developmental myosin in randomly-encountered areas.
Analysis of fibre diameter showed that, as expected, the regenerating fibres were significantly smaller than mature muscle fibres in both BMD and DMD muscle (Fig 3b). However there was no difference in size between regenerating fibres in BMD (median: 23.2; IQR: 19-29 μM) and DMD muscle (median: 24.6; IQR: 17.4-31.5 μM) (Kruskal-Wallis test). Stabilisation of ɣ-sarcoglycan and β-dystrogylcan by utrophin in DMD muscle biopsies The correlation of utrophin levels with the dystrophin associated protein complex (DAPC) proteins β-dystroglycan (β-DG) and ɣ-sarcoglycan (ɣ-SG) levels was investigated in muscle sections from 1 young (P11; 4 years of age) and 1 older DMD patient (P12; 8 years of age) by co-labelling utrophin with either β-DG or ɣ-SG antibody (S1 Fig). Intensity analysis revealed a significant positive correlation between utrophin levels and ɣ-SG subunit levels in individual fibres in the muscle of the 4 year old DMD patient P11 (Pearson's, p = 0.0008) but not in the  Fig 4a and 4b). There was also a positive correlation between utrophin and β-DG levels in muscle fibres of DMD P11 (Spearman's, r 2 = 0.75) and P12 (Pearson's, r 2 = 0.45), both of which were highly significant (p<0.0001 ; Fig 4c and 4d). We quantified the levels of β-DG by normalising to spectrin intensity (Fig 4e). There was a 44% and 66% reduction in β-DG intensity in the sarcolemma of muscle fibres from the young DMD  There are significantly more regenerating muscle fibres in DMD than in BMD muscle biopsies Muscle regeneration was assessed in the same BMD (P5, 6 and 7) and DMD (P9, 10, 11, 12,14 and 15) sections used to show that regenerating fibres have higher utrophin levels, by quantifying the number of fibres containing both fetal and developmental myosin and expressing this as a percentage of the total number of fibres. The amount of regeneration varied between the 3 BMD muscle biopsies, with P7 having significantly more regenerating fibres (median: 15.4; IQR: 5-39) than BMD P5 (median: 3.7; IQR: 0-6.9), which was significant (Mann-Whitney ttest, p = 0.0194; Fig 5a). Five of the six DMD muscle biopsies (P9, 10, 11, 12 and 15, aged 4.8, 4.3, 4.8, 7.8 and 8.2 years) analysed had high numbers of regenerating fibres, ranging from 24 to 33% of total muscle fibres in the 10 areas analysed. There were no significant differences in percentages of regenerating fibres between these 5 DMD cases (One way ANOVA with Tukey's test; Fig 5a). However, DMD muscle biopsy P14 had a lower number of regenerating fibres (median: 6; IQR: 3.1-13.7) when compared to P9 (mean 32.8 ± 3%; p = 0.0036), P10 (mean 28 ± 3.8%;  Whitney t test, Fig 5b).
To determine whether the fetal and developmental myosins were differentially expressed in fibres, serial sections of these BMD and DMD muscles were immuno-labelled with either fetal or developmental myosin antibodies. Although both myosins were expressed at similar mean intensity levels in BMD and DMD muscles, there were a number of fibres with increased intensity of both myosins in DMD (developmental; median: 428; IQR: 353-588, neonatal, median: 441; IQR: 356-564) compared to BMD muscles (developmental; median 453; IQR: 356-495, fetal: median: 399; IQR: 342-544). This difference was not statistically significant (Kruskal-Wallis followed by Dunn's test, Fig 5c). However, it should be noted that only 2 of the 3 BMD muscle sections (P6 and 7) used for this experiment stained with developmental myosin contained regenerating/positive fibres. Correlation between amount of fibre regeneration and functional motor assessment in BMD and DMD The percentage of regenerating fibres in BMD (P5 and P6) and DMD (P9, 10, 11, 12, 14 and P15) muscle biopsies were correlated with the patient's functional motor score (Fig 6). BMD P7 was omitted from the analysis, as no clinical report was available. A Pearson's test revealed a negative correlation between the level of regeneration and the functional motor score, with a lower motor score evident with an increasing percentage of regenerating fibres within the muscle biopsy, which was significant (p = 0.0435; Fig 6).

Discussion
A therapeutic strategy that is being pursued in DMD is the modulation of utrophin in dystrophin-deficient muscle fibres, i.e. the re-programming of utrophin transcription such that utrophin RNA and protein is continually expressed in mature fibres. The replacement of dystrophin with utrophin has been shown to be effective following transgenic and pharmacological approaches to increase utrophin in mdx mice [18,28].
For therapeutic efficacy, sufficient utrophin linked to the DAPC must be present at the sarcolemma of muscle fibres to prevent or slow the fibres from undergoing necrosis. To determine this, accurate quantification of utrophin and components of the DAPC at the muscle fibre sarcolemma needs to be demonstrated. If utrophin modulation results in more utrophin localised to the sarcolemma in mature dystrophin deficient fibres for longer, then a reduction in muscle fibre necrosis would be predicted.
To this end, we have developed protocols to quantify sarcolemmal utrophin and members of the DAPC using a semi-quantitative immunocytochemical technique originally developed to quantify dystrophin [20]. A novel protocol has been developed immunostaining fetal and developmental myosins in the sarcoplasm of muscle fibres to quantify the percentage of regenerating fibres in a muscle biopsy (Fig 2). Utrophin quantification was reproducible between experiments (Fig 2b).
We found, as expected, increased numbers of mature muscle fibres with utrophin in DMD compared to BMD [29] and control muscle biopsies [30] (Fig 1a). There are lower levels of utrophin in BMD compared to DMD mature muscle fibres, as the former express some dystrophin that would be expected to influence the level of utrophin. There is a significant difference between utrophin levels in regenerating and non-regenerating DMD muscle fibres (Fig 3). The difference between the situation in fibre repair and the utrophin therapeutic approach is where the utrophin is localised. In DMD, the transverse section identifies the fibres that are regenerating by virtue of the utrophin staining. However this is a very local event where the damaged membrane is being degraded and repaired via myoblast fusion. If one were to travel further along that fibre, one would find regions that have not started the regenerative process. This intermittent expression of utrophin along the fibre will have limited effect on protecting the fibre and the muscle. The therapeutic approach is to modulate normal utrophin transcription such that utrophin is present along the whole length of the fibre with no regions without utrophin at the membrane. Indeed this was achieved in the original utrophin transgenic studies where utrophin was ubiquitously expressed in muscle and associated with functional benefit [18]. Compounds developed by Summit to modulate utrophin in the mdx mouse have been shown to produce uniform continual staining along the length of the fibres and this is associated with clinical benefit. In contrast, in untreated mdx muscle there is punctate staining where it is presumed focussed repair of the fibre is taking place [31].
There was a significant positive correlation of utrophin intensity with β-dystroglycan (β-DG) intensity at the sarcolemma of muscle fibres in a 4 and an 8 year old DMD patient (P11 and P12), but ɣ-sarcoglycan (ɣ-SG) and utrophin intensities had a significantly positive correlation only in fibres from P11, not P12 (Fig 4c and 4d). We also quantified the levels of β-DG by normalising to spectrin intensity in sections of the same muscles. β-DG intensity in fibres in the older P12 was significantly lower than in the younger P11, but whether this is due to the patient age, or other factors, is not known. In both these 2 DMD patients that lack dystrophin (but have utrophin), β-DG intensity was less than in control muscles that have normal dystrophin (but no utrophin). This is not surprising, as there are functional differences between dystrophin and utrophin. Dystrophin binds actin via two separate contact sites whereas utrophin interacts with actin through a single continuous unit (for a review, see [32]). Furthermore, utrophin shows a twofold lower affinity for β-dystroglycan [33]. Nevertheless, both proteins are fully capable of interacting with F-actin and β-dystroglycan, and each alone is sufficient to preserve the sarcolemmal integrity. But dystrophin can recruit nNOS to the sarcolemma, whereas utrophin cannot [34]. In addition, dystrophin binds microtubules with high affinity and pauses microtubule polymerization, whereas utrophin has no activity in either assay [35]. Interestingly, some β-DG remains at the sarcolemma of muscle fibres in the absence of utrophin and dystrophin [36].
The amount of ɣ-SG and β-DG correlated with the level of utrophin in muscle fibres of some DMD patients, indicating that utrophin is able to recruit components of the DAPC at the sarcolemma of dystrophin-deficient muscle fibres (Fig 4). This provides further support of the advantageous effects of utrophin modulation in DMD patients in that pharmacologically maintaining utrophin levels at the costameric binding sites in fibres should also allow for conservation or localisation of the DAPC. This is the first time that the intensity of utrophin and members of the DAPC complex have been quantified simultaneously and significant correlations demonstrated.
The identification of regenerating muscle fibres is not straightforward. Although they can be distinguished from mature muscle fibres on the basis of their size and expression of developmental myosins, there is no single marker that unequivocally identifies a regenerating fibre. Some developmental myosins may be present in muscle fibres that are not regenerating, for example in denervated muscle fibres, and the time after the onset of regeneration affects the expression pattern [37,38]. Embryonic and fetal myosin heavy chains are expressed sequentially in muscle development [39]. They are also expressed in regenerating muscle fibres from approximately 2-3 days after injury and persist for 2-3 weeks [39]. The antibodies that we employed recognise a myosin heavy chain present during fetal development but down regulated by birth (MHCd: developmental myosin antibody) and a myosin present during the first few months after birth in humans (MHCn: neonatal myosin antibody) (unpublished observations). Regenerating fibres express variable levels of both proteins, depending on the stage of regeneration. In dystrophic muscle, the MHCn antibody often detects more fibres than MHCd, which recognises fibres at an earlier stage of maturation. Hence, we chose to use a combination of fetal and embryonic myosin antibodies to identify regenerating fibres at all stages of maturation. Using this antibody combination, we found, as expected [27] a significantly higher percentage of regenerating muscle fibres in DMD biopsy sections compared to BMD biopsies. We also confirmed that regenerating or recently regenerated fibres have increased sarcolemmal utrophin levels [11,[40][41][42].
Interestingly, the percentage of regenerating fibres in BMD and DMD muscle biopsies correlated inversely with the patient's functional motor score [43], suggesting that muscles undergoing less degeneration and regeneration are more functional. However, the interval between the age at which the biopsy was done and the age at which muscle functional tests were performed (Table 1) will have an effect. However, our sample numbers are small; in addition, it is known that the patient age and DMD genotype, neither of which we have taken into account in our analysis, have an effect on the motor function [44]. Other factors, including the number of surviving myofibres and extent of fibrosis, will also affect muscle function. Similarly, a significant relationship between muscle regeneration and ankle dorsiflexion force in limb girdle muscular dystrophy type 2I (LGMD2I) patients has been demonstrated, but in that study only neonatal myosin was used to distinguish the regenerating fibres [38]. The controls we used had morphologically normal muscle, although the patients had been referred to our neuromuscular service for a clinical reason.
Kleopa et al. [45] quantified utrophin on Western blot and compared the ratio of utrophin: myosin in DMD muscle samples to the ratio of utrophin:myosin in a normal human fetal muscle sample, in which utrophin expression is high. Using this measure, the mean utrophin in DMD was 55% of that in normal fetal muscle, but there was a very large standard deviation. The significance of utrophin levels and severity of disease has been highlighted recently [46] utilizing semi-quantitative western blots to demonstrate similar levels of utrophin protein in severe and less severe DMD samples. The downside to this approach is that any difference in levels of regeneration across the whole biopsy are missed and it is this subtlety that may correlate with disease severity. The development of the quantitative imaging methods described in our manuscript is to remove this variability of total protein quantification approaches with absolute quantification of utrophin levels at the membrane of each fibre. This gives novel insights into the biology of utrophin in regenerating and non-regenerating fibres and the interaction of utrophin with components of the dystroglycan complex in human muscle. This is a proof of principle study to show that it is possible to quantify differences between DMD and BMD patient biopsies. Future clinical trials would require high throughput methods, as the method we describe is labour-intensive. Consequently the next steps in the further development of these protocols is to develop solutions where the whole biopsy section is scanned and data generated using automated imaging technologies such as those recently used for the quantification of dystrophin [47].

Conclusions
The tools that we have developed to quantify sarcolemmal utrophin, DAPC proteins and muscle regeneration in muscle biopsies have led to interesting observations correlating levels of utrophin and members of the DAPC and percentage of regenerating fibres to muscle function. These tools will be invaluable for confirming utrophin modulator activity in future clinical trials. A. Serial sections of P9 and P11 were labelled with either the secondary antibody used for utrophin immunostaining only, or with a combination of both secondary antibodies used for fetal myosins and utrophin immunostaining. Images were captured for sarcolemmal intensity measurement in 20 random fibres, using the same exposure times used for utrophin intensity measurements. There was no difference in mean sarcolemmal intensity between the sections stained with one or both secondary antibodies for both P9 (unpaired t-test; p = 0.58) and P11 (unpaired t-test; p = 0.32). B. Serial sections of P11 were stained with either (1) Secondary antibody for utrophin only (2) secondary antibodies for utrophin and fetal myosins (3) secondary antibodies for utrophin and gamma sarcoglycan (g-SG) (4) secondary antibodies for utrophin and beta-dystroglycan (b-DG) and images captured for sarcolemmal intensity measurement. 1 and 2: 20 fibres analysed for each; 3 and 4: 10 fibres analysed for each. There were no differences in sarcolemmal intensity between the 4 groups (1 way ANOVA with Tukey's test). C. Serial sections of P12 were stained with secondary antibodies for (1) utrophin and g-SG, or (2) utrophin and b-DG and images captured for sarcolemmal intensity analysis. 10 fibres were quantified for each combination. There was no significant difference between sarcolemmal intensity between the two groups (unpaired t-test; p = 0.32). The sarcolemmal intensities when only secondary antibodies had been applied were similar to the mean utrophin intensity levels in control muscles (38 ± 1.7; Fig 2). (TIFF) S3 Fig. The secondary antibody used to detect beta-dystroglycan does not identify utrophin. Section of P12 was stained with mouse monoclonal antibody to utrophin, (NCL-DRP2 IgG1), followed by Alexa Fluor 488 donkey anti mouse antibody (Molecular Probes A21202, anti IgG) and the secondary antibody that was used to detect beta-dystroglycan (Alexa Fluor A594 goat anti mouse IgG2A (Molecular Probes A21135). A. Fibres expressing utrophin, detected by Alexa Flour 488 donkey anti mouse IgG antibody. B. Alexa Fluor A594 goat anti mouse IgG2A antibody does not cross-react with utrophin. C. Merged image of A and B. Scale bar = 50μM. (TIFF) S1 Table. Sarcolemmal intensity measurements after application of secondary and tertiary antibodies.