The nonsense mutation stop+4 model correlates with motor changes in
Duchenne muscular dystrophy

Claudia Brogna a,b,¶, Giorgia Coratti a,b,¶, Rachele Rossi c,¶, Marcella Neri c, Sonia Messina d,e, Adele D’ Amico f, Claudio Bruno g, Simona Lucibello a,b, Gianluca Vita e, Angela Berardinelli h,
Francesca Magri i, Federica Ricci l, Marina Pedemonte g, Tiziana Mongini l, Roberta Battini m,n,
Luca Bello o, Elena Pegoraro o, Giovanni Baranello p, Luisa Politano q, Giacomo P. Comi i, Valeria A Sansone r, Emilio Albamonte r, Alice Donati s, Enrico Bertini f, Nathalie Goemans t,
Stefano Previtali u, Francesca Bovis v, Marika Pane a,b,&, Alessandra Ferlini c,&,
Eugenio Mercuri a,b,&,∗, on behalf on the International DMD group
a Pediatric Neurology, Università Cattolica del Sacro Cuore, Rome, Italy b Centro Clinico Nemo, Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Largo Agostino Gemelli 8, Rome 00152, Italy c Unit of Medical Genetics, Department of Medical Sciences, University of Ferrara, Ferrara, Italy
d Department of Clinical and Experimental Medicine, University of Messina, Messina, Italy e Nemo SUD Clinical Center, University Hospital “G. Martino”, Messina, Italy f Department of Neurosciences, Unit of Neuromuscular and Neurodegenerative Disorders, Bambino Gesù Children’s Hospital, IRCCS, Rome, Italy g Center of Translational and Experimental Myology, IRCCS Istituto Giannina Gaslini, Genoa, Italy h Child Neurology and Psychiatry Unit, ‘‘Casimiro
Mondino’’ Foundation, Pavia, Italy i Department of Pathophysiology and Transplantation, Fondazione IRCCS Ca’ Grande Ospedale Maggiore Policlinico,
Dino Ferrari Center, , University of Milan, Milan, Italy l Neuromuscular Center, AOU Città della Salute e della Scienza, University of Torino, Italy
m Department of Developmental Neuroscience, Stella Maris Institute, Pisa, Italy n Department of Clinical and Experimental Medicine, University of Pisa,
Pisa, Italy o Department of Neurosciences, University of Padua, Padua, Italy p Fondazione IRCCS Istituto Neurologico Carlo Besta, Milan, Italy
q Cardiomiologia e Genetica Medica, Dipartimento di Medicina Sperimentale, Università della Campania Luigi Vanvitelli, Napoli, Italy r The NEMO Center in Milan, Neurorehabilitation Unit, University of Milan, ASST Niguarda Hospital, Milan, Italy s Metabolic Unit, A. Meyer Children’s Hospital, Florence, Italy t Department of Child Neurology, University Hospitals Leuven, Leuven, Belgium u Neuromuscular Repair Unit, Inspe and Division of Neuroscience,
IRCSS San Raffaele Scientific Institute, Milan, Italy v Department of Health Sciences (DISSAL), University of Genova, Genoa, Italy
Received 12 November 2020; received in revised form 1 February 2021; accepted 17 February 2021

The aim was to assess 3-year longitudinal data using 6MWT in 26 ambulant boys affected by DMD carrying nonsense mutations and to compare their results to other small mutations. We also wished to establish, within the nonsense mutations group, patterns of change according to several variables. Patients with nonsense mutations were categorized according to the stop codon type newly created by the mutation and also including the adjacent 5′ (upstream) and 3′ (downstream) nucleotides. No significant difference was found between nonsense mutations and other small mutations (p > 0.05) on the 6MWT. Within the nonsense mutations group, there was no difference in 6MWT when the patients were subdivided according to: Type of stop codon, frame status of exons involved, protein domain affected. In contrast, there was a difference when the stop codon together with the 3′ adjacent nucleotide (“stop+4 model”) was considered (p < 0.05) with patients with stop codon TGA and 3′ adjacent nucleotide G (TGAG) having a more rapid decline. Our finding suggest that the stop+4 model may help in predicting functional changes. This data will be useful at the time of interpreting the long term follow up of patients treated with Ataluren that are becoming increasingly available. © 2021 Published by Elsevier B.V. Keywords: Duchenne; Nonsense mutation; Stop+4 model. ∗ Corresponding author at: Centro Clinico Nemo, Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Largo Agostino Gemelli 8, Rome 00152, Italy. E-mail address: [email protected] (E. Mercuri). ¶ CB, GC, RR are all First Authors. & AF, MP, EM are all Senior Authors. 0960-8966/© 2021 Published by Elsevier B.V. 1.Introduction Duchenne muscular dystrophy (DMD) is a progressive, X- linked neuromuscular disorder caused by mutations in the dystrophin gene [1]. Deletions are the most frequent type of mutation in DMD patients (65%) followed by small mutations (25%) and duplications (9%) [2–4]. A number of studies have suggested possible differences in motor functional outcomes according to the type and site of mutations in the dystrophin gene [5–9]. Little has been reported about motor outcome within the group of small mutations: this appears to be challenging as this is a heterogeneous group. Nonsense mutations occur in approximately 15% of DMD boys [4] and are due to single nucleotide variations, creating a single premature stop codon in the dystrophin mRNA which prevents the production of full-length, functional dystrophin protein [10,11]. Some variability in outcome has been reported, with a wide range of age at LOA [5], but there has been no attempt to establish whether this could be related to the site or the type of mutation involved. In addition frameshift mutations are caused by deletion, insertion or deletion-insertion events (generally involving 1– 5 nucleotides). Splicing mutations include mutations within introns leading to dystrophin mRNA splicing changes. Missense mutations are small mutations in which a single nucleotide change results in a codon that codes for a different amino acid. In this paper we assessed 3-year longitudinal motor function data using 6-minute walk test, assessing the spectrum of functional changes in patients carrying small mutations and comparing them to patients with deletions or duplications. We were particularly interested in establishing (i) the functional variability within the subgroup with nonsense mutations; (ii) possible differences with other small mutations; (iii) the possible correlation with the type of stop codon and their adjacent nucleotides, the protein domain which is affected by the nonsense mutations and/or the exon in which the nonsense mutation has occurred. 2.Materials and methods The study is a longitudinal multicentric cohort study involving 14 tertiary neuromuscular centers in Italy and one center in Belgium. Patients were recruited between January 2008 and March 2015 and followed for at least three years. Patients were included if they had genetically proven DMD diagnosis, were still able to walk independently and had no severe or moderate intellectual disability or behavioral problems which could interfere with the assessments. Genetic and treatment information were collected and classified following the criteria used in our previous study [7,9,12,13]. The LOVD database DMD reading frame-checker tool (https: // was used to determine whether the exon in which the nonsense mutation occurs would produce, if omitted from the transcript by spontaneous alternative splicing events, an in frame or out of frame mRNA. DYS protein domains where nonsense mutations were located were classified as actin binding domain (ABD) (exon 1 to 8), Rod domain (RD) (exons 9–62), CRD (exon 63–69) and C terminal domain (CTD) exon 70–79). Patients with nonsense mutations were further categorized according to the type of the stop codon newly created by the mutation (TAG, TAA, TGA) and, separately, also including the adjacent 5′ (upstream) and 3′ (downstream) nucleotides. We suggest that this new way to consider a nonsense mutation together with its adjacent nucleotide could be defined “stop+4 model”. 2.1.6MWT The six-minute walking test (6MWT) is routinely used for the evaluation of endurance in DMD. All DMD ambulant boys performed 6MWT according to previously reported guidelines [14]. Data were collected from the first assessment after recruitment (baseline) and from the 12, 24, 36- month assessments. Details of the training and interobserver reliability have been previously reported [9,12]. 2.2.Statistical analysis To decribed the analyzed population Summary statistics (N, mean, SD, Range) were used. The statistical analysis was performed with SPSS software v26. A one-way ANOVA model was used to assess heterogeneity of 6MWD changes at 12, 24 and 36-months assessments. The one- way ANOVA model was performed on the whole cohort, according to the type of mutation (deletion, duplications and small mutation) and age < or > 7 years and the Tuckey correction was used for multiple testing.
In the analysis of small mutation and nonsense groups, disease progression was analyzed using the Kruskal Wallis test to compare the distribution of 6MWT meters among groupsand Dunn correction for multiple testing was performed as post-hoc test when the Kruskal Wallis test indicated a significant difference. Mann–Whitney U test was used to further analyze disease progression of nonsense mutation subgroup, comparing 12, 24 and 36 months 6MWT changes according to the frame transcript (in-frame, out of frame). P value was set at < 0.05. 2.3.Standard protocol approvals, registrations, and patient consents The study was approved by the Ethical Committee of each center. Parents of participants (all our patients were minor/children) provided written consent. 2.4.Data availability statement The data that support the findings of this study are available from the corresponding author, EM, upon reasonable request. Table 1 6MWT values and changes at baseline, 12-, 24- and 36-month subdivided by type of mutation. BASELINE 12-MONTH 24-MONTH 36-MONTH 12-MONTH CHANGES 24-MONTH CHANGES 36-MONTH CHANGES COHORT (N:294) Mean SD 369.46 77.71 362.86 106.34 340.15 147.12 310.42 174.51 -8.13 75.75 -27.87 121.8 -59.04 154.94 Min Max 140 620 0 658 0 688 0 642 -351 193 -425 272 -425 285 Duplications (N:33) Mean SD 370.36 63.84 373.6 76.53 357.7 122.28 336.89 164.89 4.34 54.31 -6.71 110.57 -33.46 153.19 Min Max 248 518 237 521 0 529 0 625 -88 130.5 -305 173 -375 200 Small Mutations (N:72) Mean SD Min Max 362.25 86.66 174 567 363.02 109.75 0 534 344.95 147.35 0 545 335.2 154.18 0 511 0.19 71.32 -204 193 -15.87 122.39 -330 272 -27.05∗ 132.06 -330 285 Deletions (N:189) Mean SD 372.06 76.52 360.92 109.86 335.56 151.11 296.36 182.48 -13.35 80.2 -35.66 123.19 -75.69∗ 161.39 Min Max 140 620 0 658 0 688 0 642 -351 150 -425 200 -425 242 Key to table: ∗ Highlights statistical significance. Detail of post hoc test are reported in the results section. 3.Results 3.1.Whole DMD cohort The DMD cohort included 294 patients (mean age 7.39 ± 2.6 SD): 189 (64.29%) patients had deletions (mean age 7.49±2.69 SD), 33 (11.22%) had duplications (mean age 7.19±2.48 SD,) and 72 (24.49%) had small mutations (mean age 7.15±2.44 SD). Of the 294 boys, 87 were not on steroids at baseline, 102 were on intermittent and 105 on daily deflazacort. At baseline, the 6MWD ranged between 140 and 620 m (mean 369.46 ± 77 SD) in the whole cohort, between 248 and 518 m (mean 370 ± 63.84 SD) in the duplication group, between 140 and 620 (mean 372.06 ± 76.52 SD) in the deletion group and between 174 and 567 m (mean 358.32 ± 86.66 SD) in the small mutation group (Table 1). The difference in 6MWT changes between small mutations, duplications and deletions was not significant at 12 months (p = 0.281) or at 24 months (p = 0.328) and showed a trend of significance at 36 months (p = 0.046). Tuckey test was performed in order to determine means that are significantly different from each group. Results determined a trend of significant difference between deletions and small mutation (p = 0.06), but not between duplications and small mutation (p = 0.979) nor between deletions and duplications (p = 0.314). When comparing the groups according to the steroid treatment (no treatment, intermittent, daily) no significant difference was found between small mutations, duplications and deletions. 3.2.Small mutations group In the group with small mutations, 26 were nonsense (mean age 7.73 ± 2.03 SD), 27 frameshifting (mean age 7.05 ± 2.97 SD), 14 splicing (mean age 6.00 ± 1.92 SD), 3 atypical (mean age 6.59 ± 0.56 SD) and 2 missense (mean age 9.79 ± 0.40 SD) mutations. Table 2 and Fig. 1 provide details (mean, range and SD) of the 6MWD in different subgroups with small mutation at baseline, 12, 24 and 36-month assessments. No significant difference was found between the different types of small mutations (atypical, frameshifting, splicing, nonsense mutations) at 12 (p = 0.800), 24 (p = 0.854) and 36 months (p = 0.868) 6MWT changes (Table 2, Fig. 1), or when comparing the groups according to the age above and under 7 years or when comparing steroid treatment at baseline (no treatment, intermittent, daily). 3.3.Nonsense mutations subgroup The 6MWT at baseline ranged between 174 and 481 m (mean: 352.60±80.12 SD). Molecular details and transcript annotation of the nonsense mutation subgroup, codon type, adjacent nucleotides, exon, frame status, and domain are described in Table 3. 3.4.In and out of frame exons Eight patients with nonsense mutations had mutations in exons that, if omitted, would produce an in-frame DMD mRNA and 18 in exons that, if omitted, would lead to an out of frame transcript. There was no difference at baseline or at 12-, 24- and 36-months between the 8 patients who had mutations predicted to be in-frame and the 18 who had mutations predicted to be out of frame (p = 0.849 at 12 months; p = 0.978 at 24 months; p = 0.644 at 36 months). 3.5.Dystrophin domains The nonsense mutations are distributed along all DMD coding sequence, from the most 5′ exon 6 to the very 3′ exon 481 Fig. 1. 6MWT trajectories and changes in small mutations in patients below or above 7 years of age. (A) Patients <7 years of age. (B) Patients ≥ 7 years of age. Color coding: Red = atypical, Yellow = missense, Green = splicing, Blue = Nonsense, Gray = frameshift, Black = all. Table 2 Age and 6MWT at baseline for small mutations, 12-, 24- and 36-month by type of mutation and age groups. AGE BASELINE 12- MONTH 24- MONTH 36- MONTH 12-MONTH CHANGES 24-MONTH CHANGES 36-MONTH CHANGES Nonsense ALL (n:26) Mean 7.73 352.60 344.60 333.37 317.02 -3.30 -9.84 -35.58 SD 2.03 80.12 144.45 182.21 183.19 82.23 128.09 132.98 Min 5.00 174.00 0.00 0.00 0.00 -204.00 -284.00 -284.00 Max 12.11 481.00 508.90 545.00 503.00 127.20 173.00 199.00 <7 years (n:11) Mean 5.83 339.65 365.92 391.23 375.91 25.20 51.57 36.25 SD 0.63 62.04 135.48 149.92 157.45 89.79 110.68 125.40 Min 5.00 204.00 0.00 0.00 0.00 -204.00 -204.00 -204.00 Max 6.74 443.00 476.00 545.00 503.00 127.20 173.00 199.00 ≥7 years (n:15) Mean 9.13 362.10 SD 1.46 92.11 328.20 154.33 280.33 198.88 273.83 193.63 -25.22 71.77 -66.13 120.21 -88.27 115.39 Min 7.00 174.00 0.00 0.00 0.00 -174.00 -284.00 -284.00 Max 12.11 481.00 508.90 494.00 481.50 45.60 52.00 40.00 Frameshift ALL (n:27) Mean 7.05 355.10 361.31 339.80 323.56 2.05 -22.95 -31.54 SD 2.97 88.46 90.85 157.78 157.49 76.66 139.46 142.33 Min 3.40 193.00 0.00 0.00 0.00 -193.00 -330.00 -330.00 Max 17.70 507.00 489.20 493.10 485.00 193.00 272.00 285.00 <7 years (n:16) Mean 5.23 354.36 382.88 405.35 379.00 21.19 37.44 24.64 SD 0.98 94.56 51.09 47.84 91.74 78.65 95.40 111.12 Min 3.40 200.00 300.00 325.00 174.00 -125.00 -86.00 -126.00 Max 6.80 489.70 489.20 493.10 485.00 193.00 272.00 285.00 ≥7 years (n:11) Mean 9.70 356.18 SD 2.90 83.26 333.86 122.20 256.36 207.60 242.91 199.48 -22.32 69.97 -99.82 152.45 -113.27 147.48 Min 7.50 193.00 0.00 0.00 0.00 -193.00 -330.00 -330.00 Max 17.70 507.00 440.00 470.00 461.00 64.40 70.00 86.00 Splicing ALL (n:14) Mean 6.00 364.79 374.21 352.01 376.36 9.43 -3.90 11.57 SD 1.92 88.43 76.44 39.29 75.86 53.15 95.12 101.34 Min 3.00 212.00 279.00 321.00 250.00 -121.00 -246.00 -132.00 Max 8.70 567.00 534.00 440.00 511.00 104.00 163.00 211.00 <7 years (n:9) Mean 4.85 333.78 355.78 362.35 383.89 22.00 28.57 50.11 SD 1.26 74.26 49.78 40.52 85.79 36.70 60.12 97.18 Min 3.00 212.00 287.00 322.15 250.00 -22.00 -35.00 -100.00 Max 6.73 475.00 453.00 440.00 511.00 104.00 163.00 211.00 ≥7 years (n:5) Mean 8.09 420.60 SD 0.60 91.26 407.40 109.05 321.00 0.00 362.80 60.22 -13.20 74.11 -101.33 127.75 -57.80 71.66 Min 7.38 325.00 279.00 321.00 280.00 -121.00 -246.00 -132.00 Max 8.70 567.00 534.00 321.00 435.00 72.00 -4.00 45.00 Atypical ALL (n:3) Mean 6.59 452.60 428.33 416.67 354.67 -24.27 -35.93 -97.93 SD 0.56 105.06 75.22 58.16 191.69 32.03 110.86 220.78 Min 5.98 351.00 350.00 350.00 134.00 -60.80 -103.80 -312.00 Max 7.08 560.80 500.00 457.00 480.00 -1.00 92.00 129.00 <7 years (n:2) Mean 6.34 398.50 392.50 396.50 307.00 -6.00 -2.00 -91.50 SD 0.51 67.18 60.10 65.76 244.66 7.07 132.94 311.83 Min 5.98 351.00 350.00 350.00 134.00 -11.00 -96.00 -312.00 Max 6.70 446.00 435.00 443.00 480.00 -1.00 92.00 129.00 ≥7 years (n:1) Mean 7.08 560.80 SD N/A N/A 500.00 N/A 457.00 N/A 450.00 N/A -60.80 N/A -103.80 N/A -110.80 N/A Min 7.08 560.80 500.00 457.00 450.00 -60.80 -103.80 -110.80 Max 7.08 560.80 500.00 457.00 450.00 -60.80 -103.80 -110.80 Missense ≥ 7 years (n:2) Mean 9.79 431.00 SD 0.40 87.68 420.00 147.08 392.50 116.67 411.50 82.73 -11.00 59.40 -38.50 28.99 -19.50 4.95 Min 9.50 369.00 316.00 310.00 353.00 -53.00 -59.00 -23.00 Max 10.07 493.00 524.00 475.00 470.00 31.00 -18.00 -16.00 75. In 19/26 the mutations occurred in the DYS CR domain, in 4 in the CT, in 2 in the AB domain, and in one at hinge 4.motif (Supplemental Figure-1). 3.6. Stop codon types and adjacent nucleotide The stop codon generated by the nonsense mutation was TAA in 2/26, TAG in 11, and TGA in 13. There was no difference in 6MWT progression over 36 months among the three subgroups (p = 0.260 at 12 months; p = 0.281 at 24 months; p = 0.326 at 36 months) also when the nucleotide change (C>T, G>T, T>A, G>A) (p = 0.158 at 12 months;
p = 0.247 at 24 months; p = 0.385 at 36 months) were considered.
Fig. 2 shows individual trajectories according to the nucleotide change and the stop codon type.
Nonsense mutations and stop codon adjacent nucleotides (tetranucleotides) 5′ and 3′ adjacent nucleotides are detailed according to the DMD mRNA Refseq (NM_004006.2).
The 6 patients with TGA and adjacent +4 nucleotide G (TGA G) had a more rapid decline than the 5 TGA with adjacent +4 nucleotide C (TGA C) who had a more stable course of progression (Fig. 3).

Table 3
Molecular details of the nonsense mutation subgroup.

NM_004006.2 transcript reference sequence (LOVD database)
Generated StopCodon
Codons and adjacent nucleotides (5′ -3′ )
Exon Exonframing DYS domain

1c.433C>T p.Arg145∗ TGA GTCTGACAA 6 out-of-frame ACTIN BINDING(CH1/CH2)
2c.583C>T p.Arg195∗ TGA CAATGACTG 7 out-of-frame ACTIN BINDING (CH2)
3c.2302C>T p.Arg768∗ ° TGA GAGTGAGAA 19 out-of-frame ROD-LIKE (R4)
4c.2302C>T p.Arg768∗ ° TGA GAGTGAGAA 19 out-of-frame ROD-LIKE (R4)
5c.2474G>A Trp825∗ TAG AACTAGCTG 20 out-of-frame ROD-LIKE(R4/R5)
6c.2791 G>T Glu931∗ TAG AAATAGCTA 21 out-of-frame ROD-LIKE(R5)
7c.3742C>T p.Gln1248∗ TAG TACTAGTGG 27 in-frame ROD-LIKE(R8)
8c.5089C>T p.Gln1697∗ TAG ATTTAGGCT 36 in-frame ROD-LIKE(R13)
9c.5563 C>T p.Gln 1855∗ ° TAG TAATAGACA 39 in-frame ROD-LIKE(R14)
10c.5563 C>T p.Gln1855∗ ° TAG TAATAGACA 39 in-frame ROD-LIKE(R14)
11c.6283C>T p.Arg2095∗ TGA GACTGACAA 43 out-of-frame ROD-LIKE(R16)
12c.6283C>T p.Arg2095∗ TGA GACTGACAA 43 out-of-frame ROD-LIKE(R16)
13c.6292C>T p.Arg2098∗ TGA GGGTGATTT 44 out-of-frame ROD-LIKE(R16/R17)
14c.6310G>T p.Glu2104∗ TAG GTTTAGAAA 44 out-of-frame ROD-LIKE(R16/R17)
15c.6805C>T p.Gln2269∗ TAA AAATAATTA 47 in-frame ROD-LIKE (R18)
16c.6889C>T p.Gln2297∗ TAG AAGTAGACA 47 in-frame ROD-LIKE (R18)
17c.7736T>A p.Leu2579∗ TAA ATGTAAAAG 53 out-of-frame ROD-LIKE(R20/R21)
18c.8880G>A p.Trp2960∗ TGA TCCTGACAG 59 out-of-frame ROD-LIKE(R23/R24)
19c.8791 G>T, p.Glu2931∗ ° TAG GATTAGACC 59 out-of-frame ROD-LIKE(R23/R24)
20c.8791 G>T, p.Glu2931∗ ° TAG GATTAGACC 59 out-of-frame ROD-LIKE(R23/R24)
21c.8944C>T p.Arg2982∗ TGA CTTTGAGGA 60 in-frame ROD-LIKE(R24)
22c.9100C>T p.Arg3034∗ TGA GACTGAGTC 61 out-of-frame HINGE 4/WW
23c.10108C>T p.Arg3370∗ TGA GTTTGAGAC 70 out-of-frame C TERMINAL
24c.10108 C>T p.Arg3370∗ TGA GTTTGAGAC 70 out-of-frame C TERMINAL
25c.10141C>T p.Arg3381∗ TGA TTTTGAACC 70 out-of-frame C TERMINAL
26c.10651C>T p.Gln3551∗ TAG CCCTAGAGT 75 in-frame C TERMINAL
Key to table = bold: Newly generated stop codon; underlined: +4 adjacent (3′ ) nucleotide. °: brothers DYS dystrophin protein.

The 7 patients with TAG and adjacent +4 nucleotide A (TAG A) also had a more stable course of progression. All the other combinations of stop codons and adjacent nucleotides were limited to one or, in a few cases, two patients as shown in Fig. 3.
When comparing groups, a difference in 6MWT progression was not found at 12 months (p = 0.192), but was
found at 24 months (p = 0.008) and at 36 months (p = 0.032). At 24 months, there was a difference in the 6MWT
progression between TGAG and TGAC (p = 0.0093), between TGAG and TGAGA (p = 0.0091) but not between TGAC and TAGA (p = 0.4672).
At 36 months, there was a difference in the 6MWT progression between TGAG and TGAC (p = 0.0264), between
TGAG and TGAGA (p = 0.0305) but not between TGAC and TAGA (p = 0.4646).
As a negative control group, we also stratified patients’ 6MWT trajectories according to the adjacent nucleotide -1 but no pattern of 6MWT changes could be identified with any of the new combinations using the -1 adjacent nucleotide. The difference among the different subgroups was not significant (p = 0.472).
4. Discussion

The advent of a therapeutic approach, Ataluren, targeting nonsense mutations, has resulted in an increased interest in these mutations [15–20]. A recent paper [18], reporting the results of an international drug registry highlights the need
to have longitudinal natural history reference data for the interpretation of the long term findings in the treated cohort.
So far, the literature about natural history in this group of mutations has been very scanty. The results of the placebo group in the two placebo controlled studies [17,20] show a wide variability in the 6MWT after 12 months. Nothing has been reported about longer term follow up with the exception of a recent study reporting a wide variability in LOA, with a number of patients with nonsense mutations still walking at age of 20 years and others having an early LOA [5].
In the present study, we report 36-month longitudinal natural history data using 6MWT in DMD patients with small mutations, also focusing on those with nonsense mutations. Our results confirm previous studies reporting a better overall outcome in DMD boys with small mutations and duplications than in those with deletions, especially in the group of DMD boys above the age of 7 years [7]. The difference between the different types of small mutations (atypical, frameshifting, splicing, nonsense, and missense mutations) was not significant. Although the different groups included boys with different age range and were not easily comparable, some differences could be noted. Early LOA (before the age of 9 years) occurred more frequently in boys with nonsense mutations. Patients with splicing mutations appeared to have overall less decline on 6MWT and, although the results should be interpreted with caution as most of them were still younger than 12 years at the time of the last assessment, unlike those with nonsense mutations, none lost ambulation before the age

Fig. 2. Details of the trajectories according to age, flanking nucleotide and stop codon type.
– – – -: TAA, -. -: TAG, —: TGA. Red line: nucleotide change C > T; Light blue line: nucleotide change G > T, Purple line: nucleotide change T > A, Green line: nucleotide change G > A.

of 12 years. A better disease course in this group would not be unexpected, since splicing mutations are susceptible to be modulated by alternative splicing events, which may favorably reframe the DMD transcript, therefore leading to a milder phenotype [21].
In this paper we were particularly interested in the group of 26 boys with nonsense mutations. We confirmed previous findings that although early LOA can be observed in this group of mutations [5], there is an overall wide variability in the progression. After the age of 10 years, while some patients showed a rapid decline and lost ambulation, others remained relatively stable. The variability could not be only explained by the steroid regime as both daily and intermittent deflazacort could be found in patients with different patterns of progression.
Following the hypothesis that nonsense variations occurring in-frame exons may cause milder phenotypes via spontaneous, mRNA re-framing, exon skipping of the mutated exon, which may lead to some DYS protein production, we explored the possibility that the different progression could be related to the “re-framing” ability of exon in which the nonsense mutations occurred. We were however unable to find any difference as nonsense mutations located in both “in-frame” and “out of frame” exons were associated with
both early decline and relative stability. A previous study has analyzed correlations of clinical phenotype (DMD vs. IMD or BMD) with the strength of exon recognition motifs as exon splicing enhancers/silencers (ESE, ESS) located in- frame exons, showing a trend towards milder phenotypes with weaker recognition motifs [22]. In our DMD population there were no mutations occurring in previously identified exons predicted with the “weakest” recognition motifs, as exons 25, 29, 31, and 37, therefore we could not confirm this observation.
Similarly, no differences could be found concerning the domain of the DYS protein where the nonsense mutation lies, suggesting that the occurrence of the premature truncation in a specific DYS domain does not play a role in the clinical outcome. It has been previously reported that mutations occurring toward near the 3′ end of the DMD mRNA may result in a less premature protein truncation which may favor a downstream protein translation rescue or may reduce the extent of mRNA nonsense-mediated decay, therefore causing a milder phenotype [23,24]. When we looked at the 5′ to 3′ mRNA position of the nonsense mutations, the few patients with mutations clustered in the C terminal appeared to have a relatively stable course of progression, but the numbers were too small to draw definite conclusions. Most of the patients

Fig. 3. Details of the trajectories according to age, 3′ adjacent +4 nucleotide (tetranucleotide) and stop codon type.
Color coding: Violet = TAAA, Yellow = TAGG, Blue = TGAC, Orange = TAGA, Light blue = TAAT, Green = TGAG, Brown = TAGC, Pink = TGAA, Gray = TAGT, Red = TGAT.

had mutations in the rod domain and these were associated with a variable decline. Patients with mutations at the 3′ downstream rod domains (R21–23) were more often more stable than those with mutations at the 5′ upstream repeats (R 4–8) but this was not always true for individual cases. Therefore, our data do not support the 5′-3′ severity trend in all cases with nonsense mutations, though of course this should be further verified in a larger number of patients.
The variability could also not be explained by the different stop codon type created by the mutations. Based on the literature, the translation termination efficiency varies among the three codons being TAA>TGA>TAG [11,25] and we would therefore expect that boys carrying a TAA nonsense mutation might be more severely affected since TAA cannot be spontaneously read through [25]. Conversely, TGA and TAG are less robustly recognized by the termination complex and releasing factors and can be more frequently and physiologically read through, possibly leading to some protein translation [26]. In our cohort both subgroups carrying TGA and TAG showed some variability, but the difference was not significant and no specific pattern of progression could be identified.
It is also known that the adjacent 3′ nucleotide of the stop codon (called +4) plays a crucial role in stop codon recognition and its consequent efficiency [26,27], with an impact on disease severity, as described in Lysosomal
storage diseases and HERG gene mutation [28,29]. This suggests that the recognition of a nonsense mutation by the ribosome translational factors, as RF1 and RF2, might be
influenced by the +4 nucleotide. The remark that the +4C nucleotide significantly increases the stop codon read-through by gentamicin and aminoglycosides, being therefore more “permissive” in terms of translation rescue was hypothesized in 2003 by Politano et al. [30] and further supported recently [31]. Similarly, in a pilot study exploring the nonsense codon context in culture HEK293 cells harboring TAA, TAG or TGA nonsense allele treated with increasing concentration of Ataluren it was shown that the termination efficiency of Ataluren was greater when a cytosine was located +4 position and that the TGA C termination context was the only
efficiently suppressed [32]. According to this+4 model, the RFs recognize a tetranucleotide (instead of the trinucleotide classical codon) which contains very limited redundancy, and confers higher specificity to ribosomal recognition site [23].
The role of the +4 nucleotide has never been explored in the DMD gene in relation to long term functional outcome. When we looked at the combination of the stop codons and the adjacent nucleotides in our cohort, some patterns of 6MWT changes could be observed. Patients with TGA had a different progression depending on the adjacent nucleotide; those with adjacent nucleotide +4 G (TGA G) had a more rapid decline than those with +4 C (TGA C), who had a

more stable course of progression. Interestingly, also patients with TAG and adjacent nucleotide +4 A (TAG A) had a more stable course of progression. Our results, although obtained in a small cohort of patients, support the already reported +4 C nucleotide role in the TAG C tetranucleotide, as weakening the stop codon ribosomal recognition, and the novel +4 A nucleotide effect in the TAG A tetranucleotide, having the very same effect. This novel stop+4 model allows predicting a better motor outcome in DMD boys.
This observation, that should be further confirmed, improves our understanding of genotype-phenotype relationship in nonsense mutation. These findings may provide new clues for the prognosis of individual cases of small mutations in both untreated patients and in those currently treated with nonsense mutation correction strategies.
Although this was a multicenter international effort, involving 13 tertiary care centers and recruiting all the patients consecutively seen in clinic, the number of patients was relatively small. This is due to the fact that in the last few years most patients with nonsense mutations had been treated with Ataluren and could therefore not be part of a natural history study. Also, the study design which excluded BMD and most IMD patients may have reduced the power to establish genotype-phenotype correlation especially for nonsense mutations.
In each subgroup there was some variability that could not always be justified by an individual clinical characteristic, such as steroids regime, or genetic variable (codons, domain, etc.). Other variables, such as genetic modifiers [33–35], that have not been assessed in this study, may account for the observed variability. Including other variables, such as additional functional measures, height or body mass index, and a larger panel of measures may also be of help, as recently shown in other DMD cohorts, to establish more accurate trajectories of progression using advanced statistical analysis [36,37].
Despite of these limitations, our findings provide, for the first time, longitudinal data using the 6MWT in a relatively large cohort of patients with nonsense mutations and a completely new model for nonsense mutation readiness, the stop+4 model. This data will be of help at the time of interpreting the long term follow up of patients treated with Ataluren that are becoming increasingly available [18,38].

Declaration of Competing Interest



We are grateful to the Italian Telethon (GUP 15011) for the financial support to this study and the International MDEX Consortium (Drs Volker Straub, Newcastle; Laurent Servais, Paris; Dr Erik Nick and Jan Verschuuren, Leiden, and Dr Imelda DeGroot, Nijmegen) for the participation to the AFM funded natural history study (Grant N. 17100). The Italian Duchenne Parent Project is acknowledged for funding the DMD diagnostic project and DMD cell biobank.

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.nmd.2021.02. 015.


[1]Mercuri E, Muntoni F. Muscular dystrophies. Lancet 2013;381(9869):845–60.
[2]Bladen CL, Salgado D, Monges S, Foncuberta ME, Kekou K, Kosma K, et al. The TREAT-NMD DMD global database: analysis of more than 7,000 Duchenne muscular dystrophy mutations. Hum Mutat 2015;36(4):395–402.
[3]Tuffery-Giraud S, Beroud C, Leturcq F, Yaou RB, Hamroun D, Michel-Calemard L, et al. Genotype-phenotype analysis in 2,405 patients with a dystrophinopathy using the UMD-DMD database: a model of nationwide knowledgebase. Hum Mutat 2009;30(6):934–45.
[4]Neri M, Rossi R, Trabanelli C, Mauro A, Selvatici R, Falzarano MS, et al. The genetic landscape of dystrophin mutations in Italy: a nationwide study. Front Genet 2020;11:131.
[5]Bello L, Morgenroth LP, Gordish-Dressman H, Hoffman EP, McDonald CM, Cirak S, et al. DMD genotypes and loss of ambulation in the CINRG Duchenne natural history study. Neurology 2016;87(4):401–9.
[6]Ricotti V, Ridout DA, Pane M, Main M, Mayhew A, Mercuri E, et al. The northstar ambulatory assessment in Duchenne muscular dystrophy: considerations for the design of clinical trials. J Neurol Neurosurg Psychiatry 2016;87(2):149–55.
[7]Pane M, Mazzone ES, Sormani MP, Messina S, Vita GL, Fanelli L, et al. 6 Min walk test in Duchenne MD patients with different mutations: 12 month changes. PLOS ONE 2014;9(1):83400.
[8]Servais L, Montus M, Guiner CL, Ben Yaou R, Annoussamy M, Moraux A, et al. Non-ambulant Duchenne patients theoretically treatable by exon 53 skipping have severe phenotype. J Neuromuscul Dis 2015;2(3):269–79.
[9]Brogna C, Coratti G, Pane M, Ricotti V, Messina S, D’Amico A, et al. Long-term natural history data in Duchenne muscular dystrophy ambulant patients with mutations amenable to skip exons 44, 45, 51 and 53. PLOS ONE 2019;14(6):0218683.
[10]Wang RT, Barthelemy F, Martin AS, Douine ED, Eskin A, Lucas A, et al. DMD genotype correlations from the Duchenne registry: endogenous exon skipping is a factor in prolonged ambulation for individuals with a defined mutation subtype. Hum Mutat 2018;39(9):1193–202.
[11]Wang L, Xu M, Li H, He R, Lin J, Zhang C, et al. Genotypes and phenotypes of DMD small mutations in Chinese patients with dystrophinopathies. Front Genet 2019;10:114.


We acknowledge the role of the Association Francaise contre les Myopathies (AFM), the UCL MRC Neuromuscular Biobank and the Muscular Dystrophy UK for the support given to the Dubowitz Neuromuscular Center.
[12]Pane M, Mazzone ES, Sivo S, Sormani MP, Messina S, DA A, et al. Long term natural history data in ambulant boys with Duchenne muscular dystrophy: 36-month changes. PLOS ONE 2014;9(10): 108205.
[13]Pane M, Mazzone ES, Sivo S, Sormani MP, Messina S, D’Amico A, et al. Correction: long term natural history data in ambulant boys

with Duchenne muscular dystrophy: 36-month changes. PLOS ONE 2015;10(12):0144079.
[14]Henricson E, Abresch R, Han JJ, Nicorici A, Goude Keller E, de Bie E, et al. The 6-minute walk test and person-reported outcomes in boys with duchenne muscular dystrophy and typically developing controls: longitudinal comparisons and clinically-meaningful changes over one year. PLOS Curr 2013;5.
[15]Finkel RS, Flanigan KM, Wong B, Bonnemann C, Sampson J, Sweeney HL, et al. Phase 2a study of ataluren-mediated dystrophin production in patients with nonsense mutation Duchenne muscular dystrophy. PLOS ONE 2013;8(12):81302.
[16]Landfeldt E, Lindberg C, Sejersen T. Improvements in health status and utility associated with Ataluren for the treatment of nonsense mutation Duchenne muscular dystrophy. Muscle Nerve 2020;61(3):363–8.
[17]McDonald CM, Campbell C, Torricelli RE, Finkel RS, Flanigan KM, Goemans N, et al. Ataluren in patients with nonsense mutation Duchenne muscular dystrophy (ACT DMD): a multicentre, randomised, double-blind, placebo-controlled, phase 3 trial. Lancet 2017.
[18]Mercuri E, Muntoni F, Osorio AN, Tulinius M, Buccella F, Morgenroth LP, et al. Safety and effectiveness of Ataluren: comparison of results from the STRIDE registry and CINRG DMD natural history study. J Comp Eff Res 2020;9(5):341–60.
[19]Muntoni F, Domingos J, Manzur AY, Mayhew A, Guglieri M, Network UKN, et al. Categorising trajectories and individual item changes of the North Star Ambulatory Assessment in patients with Duchenne muscular dystrophy. PLOS ONE 2019;14(9):0221097.
[20]Bushby K, Finkel R, Wong B, Barohn R, Campbell C, Comi GP, et al. Ataluren treatment of patients with nonsense mutation dystrophinopathy. Muscle Nerve 2014;50(4):477–87.
[21]Tuffery-Giraud S, Miro J, Koenig M, Claustres M. Normal and altered pre-mRNA processing in the DMD gene. Hum Genet 2017;136(9):1155–72.
[22]Flanigan KM, Dunn DM, von Niederhausern A, Soltanzadeh P, Howard MT, Sampson JB, et al. Nonsense mutation-associated Becker muscular dystrophy: interplay between exon definition and splicing regulatory elements within the DMD gene. Hum Mutat 2011;32(3):299–308.
[23]Kerr TP, Sewry CA, Robb SA, Roberts RG. Long mutant dystrophins and variable phenotypes: evasion of nonsense-mediated decay? Hum Genet 2001;109(4):402–7.
[24]Tuffery-Giraud S, Saquet C, Thorel D, Disset A, Rivier F, Malcolm S, et al. Mutation spectrum leading to an attenuated phenotype in dystrophinopathies. Eur J Hum Genet 2005;13(12):1254–60.
[25]Shi M, Zhang H, Wang L, Zhu C, Sheng K, Du Y, et al. Premature termination codons are recognized in the nucleus in a reading-frame dependent manner. Cell Discov 2015;1:15001. doi:10.1038/celldisc. 2015.1. Epub 2015 May 5.

[26]Brown A, Shao S, Murray J, Hegde RS, Ramakrishnan V. Structural basis for stop codon recognition in eukaryotes. Nature 2015;524(7566):493–6.
[27]Tate WP, Poole ES, Horsfield JA, Mannering SA, Brown CM, Moffat JG, et al. Translational termination efficiency in both bacteria and mammals is regulated by the base following the stop codon. Biochem Cell Biol 1995;73(11–12):1095–103.
[28]Yu H, Meng Y, Zhang S, Tian C, Wu F, Li N, et al. Stop codons and the +4 nucleotide may influence the efficiency of G418 in
rescuing nonsense mutations of the HERG gene. Int J Mol Med 2019;44(6):2037–46.
[29]Keeling KM. Nonsense suppression as an approach to treat lysosomal storage diseases. Diseases 2016;4(4):32.
[30]Politano L, Nigro G, Nigro V, Piluso G, Papparella S, Paciello O, et al. Gentamicin administration in Duchenne patients with premature stop codon. Preliminary results. Acta Myol 2003;22(1):15–21.
[31]Wangen JR, Green R. Stop codon context influences genome-wide stimulation of termination codon readthrough by aminoglycosides. Elife 2020;9.
[32]Welch EM, Barton ER, Zhuo J, Tomizawa Y, Friesen WJ, Trifillis P, et al. PTC124 targets genetic disorders caused by nonsense mutations. Nature 2007;447(7140):87–91.
[33]Bello L, Flanigan KM, Weiss RB, United Dystrophinopathy P, Spitali P, Aartsma-Rus A, et al. Association study of exon variants in the NF-kappaB and TGFbeta pathways identifies CD40 as a modifier of Duchenne muscular dystrophy. Am J Hum Genet 2016;99(5):1163–71.
[34]Bello L, Piva L, Barp A, Taglia A, Picillo E, Vasco G, et al. Importance of SPP1 genotype as a covariate in clinical trials in Duchenne muscular dystrophy. Neurology 2012;79(2):159–62.
[35]Flanigan KM, Ceco E, Lamar KM, Kaminoh Y, Dunn DM, Mendell JR, et al. LTBP4 genotype predicts age of ambulatory loss in Duchenne muscular dystrophy. Ann Neurol 2013;73(4):481–8.
[36]Goemans N, Vanden Hauwe M, Signorovitch J, Swallow E, Song J. Collaborative trajectory analysis P. Individualized prediction of changes in 6-minute walk distance for patients with Duchenne muscular dystrophy. PLOS ONE 2016;11(10):0164684.
[37]Mercuri E, Signorovitch JE, Swallow E, Song J, Ward SJ, Group DMDI, et al. Categorizing natural history trajectories of ambulatory function measured by the 6-minute walk distance in patients with Duchenne muscular dystrophy. Neuromuscul Disord 2016;26(9):576–83.
[38]Muntoni F, Desguerre I, Guglieri M, Osorio AN, Kirschner J, Tulinius M, et al. Ataluren use in patients with nonsense mutation Duchenne muscular dystrophy: patient demographics and characteristics from the STRIDE registry. J Comp Eff Res 2019;8(14):1187–200.