Muscular Dystrophies in Human Urine

CLAIM OF PRIORITY

This application claims the benefit of U.S. Provisional Application Serial No. 62/365, 139, filed on July 21, 2016. The entire contents of the foregoing are incorporated herein by reference.

FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

This invention was made with Government support under Grant No.

CA069246 awarded by the National Institutes of Health. The Government has certain rights in the invention.

TECHNICAL FIELD

Described herein are methods for diagnosing and monitoring subjects with diseases associated with aberrant splicing, based upon detecting properly spliced isoforms and mis-spliced isoforms in a urine sample from the subject.

BACKGROUND

Pre-mRNA splicing occurs when introns are removed to generate a protein-coding message, while alternative splicing involves inclusion or exclusion of certain exons to code for different protein isoforms from the same gene. These splice variants are a fundamental process of nature designed to increase biodiversity, mainly in eukaryotes. Mis-regulation of pre-mRNA alternative splicing is found in a number of neurologic and neuromuscular diseases \ For example, in myotonic dystrophy type 1 (DM1) an expanded trinucleotide repeat in the 3' UTR of the DMPK transcript disrupts splicing regulator proteins in the muscleblind-like (MBNL) family, causing abnormal splicing of a number of pre-mRNAs ² · ³ .

SUMMARY

Urine contains extracellular RNA (exRNA) markers of urogenital cancers.

However, the capacity of genetic material in urine to identify systemic diseases outside the urinary tract is unknown. In clinical trials for myotonic dystrophy type 1

(DM1) and Duchenne muscular dystrophy (DMD), non-invasive detection of mRNA splicing outcomes is needed to monitor therapeutic antisense oligonucleotide (ASO) drug effects. The present inventors examined whether ex-mRNA splice variants in human urine could meet sensitivity and specificity as robust markers of muscular dystrophies and other conditions associated with aberrant splicing, e.g., conditions associated with muscle weakness or dystrophy.

Ten transcripts were identified that are spliced differently in urine from DM1 patients as compared to unaffected individuals and disease controls. The predictive model was 100% accurate in our independent validation set. Urine also contains mutation-specific dystrophin deletion mRNAs amenable to therapeutic exon skipping ASO strategies in DMD patients, and a dystrophin cryptic splice site in a patient with Becker muscular dystrophy.

These results show that urine provides a renewable source of ex-mRNA splice variants that can serve as a powerful composite biomarker of DM1 or personalized genetic markers of DMD, suggesting its potential to monitor therapeutic response.

Thus, provided herein are methods that include obtaining a sample comprising urine from a subject who has, or is suspected to have, a disease associated with aberrant mRNA splicing; isolating extracellular mRNA in the sample; determining one or more selected mRNA in the sample, wherein the one or more selected mRNA is aberrantly spliced in the subject, and is suspected to be present in a plurality of spliced isoforms in the sample, wherein the spliced isoforms comprise properly spliced isoforms and mis-spliced isoforms; quantitating levels of the properly spliced isoforms and mis-spliced isoforms of the selected mRNA in the sample; and determining a ratio of the properly spliced isoforms to the mis-spliced isoforms in the sample.

Also provided are methods for diagnosing a disease associated with aberrant mRNA splicing. The methods include obtaining a sample comprising urine from a subject who has, or is suspected to have, a disease associated with aberrant mRNA splicing; isolating extracellular mRNA in the sample; determining one or more selected mRNA in the sample, wherein the one or more selected mRNA is aberrantly spliced in the subject, and is suspected to be present in a plurality of spliced isoforms in the sample, wherein the spliced isoforms comprise properly spliced isoforms and mis-spliced isoforms; quantitating levels of the properly spliced isoforms and mis- spliced isoforms of the one or more selected mRNAs in the sample; determining a ratio of the properly spliced isoforms to the mis-spliced isoforms of the one or more selected mRNAs in the sample; and comparing the ratio of properly spliced to mis- spliced in a subject to a reference ration, wherein a ratio in the subject that is less than the reference ratio indicates the presence of a disease associated with aberrant mRNA splicing.

In addition, provided herein are methods for monitoring the efficacy of a treatment for a disease associated with aberrant mRNA splicing in a subject. The methods include determining a first ratio of properly spliced isoforms to mis-spliced isoforms in a sample from the subject using a method described herein; administering a treatment for the disease to the subject; determining a subsequent ratio of properly spliced isoforms to mis-spliced isoforms in a sample from the subject using a method described herein; and comparing the first and subsequent ratios, wherein a ratio in the second sample that is higher than the ratio in the subsequent sample indicates that the treatment is effective.

In some embodiments, the treatment that is intended to correct splicing; to inhibit or reduce levels of mis-spliced transcripts; or to alter splicing to produce a functional protein. In some embodiments, the treatment is an antisense

oligonucleotide.

In some embodiments, the disease is myotonic dystrophy type 1 (DM1); Duchenne muscular dystrophy (DMD); Becker muscular dystrophy (BMD); limb girdle muscular dystrophy type IB (LGMD1B); LMNA-linked dilated

cardiomyopathy (DCM); Hutchinson-Gilford progeria syndrome (HGPS); Familial partial lipodystrophy type 2 (FPLD2); spinal muscular atrophy (SMA); or amyotrophic lateral sclerosis (ALS).

In some embodiments, the disease is myotonic dystrophy type 1 (DM1), and wherein the one or more selected mRNAs is selected from the group consisting of the transcript for insulin receptor (INSR); muscleblind like splicing regulator 2

(MBNL2); SOS Ras/Rac guanine nucleotide exchange factor 1 (SOS1); cytoplasmic linker associated protein 1 (CLASP 1); muscleblind like splicing regulator 1

(MBNL1); mitogen-activated protein kinase kinase kinase 4 (MAP3K4); nuclear factor I X (NFIX); nuclear receptor corepressor 2 (NCOR2); VPS39, HOPS complex subunit (VPS39); and microtubule associated protein tau (MAPT).

In some embodiments, the selected mRNAs comprise MBNL2, MBNL1, SOS 1, CLASP 1, MAP3K4, and optionally INSR. In some embodiments, the disease is associated with aberrant splicing of dystrophin (DMD); lamin A/C (LMNA); survival of motor neuron 2, centromeric (SMN2); solute carrier family 1 member 2 (SLC1A2); TAR DNA-binding protein (TARDP); or FUS RNA binding protein (FUS).

In some embodiments, one or more selected mRNAs is selected from the group consisting of the transcript for insulin receptor (INSR); muscleblind like splicing regulator 2 (MBNL2); SOS Ras/Rac guanine nucleotide exchange factor 1 (SOS l); cytoplasmic linker associated protein 1 (CLASPl); muscleblind like splicing regulator 1 (MBNL1); mitogen-activated protein kinase kinase kinase 4 (MAP3K4); nuclear factor I X (NFIX); nuclear receptor corepressor 2 (NCOR2); VPS39, HOPS complex subunit (VPS39); microtubule associated protein tau (MAPT); dystrophin (DMD); lamin A/C (LMNA); survival of motor neuron 2, centromeric (SMN2); solute carrier family 1 member 2 (SLC1A2); TAR DNA-binding protein (TARDP); and FUS RNA binding protein (FUS).

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Methods and materials are described herein for use in the present invention; other, suitable methods and materials known in the art can also be used. The materials, methods, and examples are illustrative only and not intended to be limiting. All publications, patent applications, patents, sequences, database entries, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control.

Other features and advantages of the invention will be apparent from the following detailed description and figures, and from the claims.

DESCRIPTION OF DRAWINGS

Figures 1A-B. Extracellular mRNA in human urine and serum. We isolated and examined extracellular RNA (exRNA) from urine (N = 15 DM1 patients and 13 UA control subjects without muscular dystrophy) and serum (N = 12 DM1 and 8 UA controls). (A) Expression of DMPK mRNA (upper) and GTF2B mRNA (middle) by qPCR as measured by cycle threshold (Ct) values, and DMPK expression normalized to GTF2B (lower) in human urine, and (B) in human serum. Individual data points represent the mean of duplicate assays for each sample. Error bars = mean ± s.e.m. **** P < 0.0001; *** P = 0.0005 (/-test). Figures 2A-C. mR A alternative splicing in exR A from human urine. We isolated urine exRNA from 27 DM1, 14 DMD/BMD controls (MDC), and 26 unaffected (UA) subjects, and examined alternative splicing by RT-PCR and gel electrophoresis. ⁴ (A) Representative gel images showing alterative splicing of human insulin receptor (INSR) exon 11, MBNL2 exon 6, SOS1 exon 25, MBNLl exon 7, CLASP1 exon 20, MAP3K4 exon 17, NFIX exon 7, NCOR2 exon 45a, VPS39 exon 3, and MAPT exons 2 and 3. PCR cycle number was 36 (INSR, MBNL2, SOS1, CLASP1, MAP3K4, NFIX, NCOR2, VPS39) or 37 (MBNLl, MAPT). Control muscle cDNA was diluted 1 :50 (MAPT) or 1: 100 (INSR, MBNL2, SOS 1, MBNLl, CLASP 1, MAP3K4, NFIX, NCOR2, VPS39) and amplified in the same PCR reaction as urine samples. (B, C) Individual data points represent quantitation of splicing of all individual urine samples examined. Error bars = mean ± s.e.m. **** P < 0.0001 (1-way ANOVA); ** = mean difference 9.4, 95% CI of difference 2.9 to 15.9.

Figures 3A-C. Principle component analysis and predictive modeling of urine splicing outcomes. Using principal component regression, a linear combination of 10 urine transcripts that show differential splicing in DM1 subjects (INSR, MBNL2, SOS1, MBNLl, CLASP1, MAP3K4, NFIX, NCOR2, VPS39, and MAPT) was used to develop a predictive model of DM1. (A) Principle component (PC) score for each subject (N = 23 DM1, 8 MDC, 22 UA). (B) DM1 (N = 23) and UA (N = 22) subjects were combined (N = 45 total), then 34 randomly assigned, irrespective of genotype, to a training set that was used to generate a predictive model of urine splicing outcomes. Using a singular value decomposition algorithm for the fitting and a threshold of 0.5 (see Methods), the model produced zero false positives and false negatives in a 5-fold cross-validation test. The Receiver Operating Characteristic (ROC) curve is shown. (C) The remaining 11 subjects, plus an additional 8 subjects examined while blinded to genotype (N = 19 total), formed an independent validation set that produced zero false positives and false negatives.

Figures 4A-C. Reliability of exRNA alternative splicing outcomes in human urine. We collected separate urine specimens from 12 DM1, 4 MDC, and 10 UA subjects several months apart and examined exRNA splicing outcomes by RT-PCR. Note that due to low collection volume in one of the UA specimens, 8 of the 10 transcripts have replicates from only 9 UA specimens.

Figures 5A-B. Alternative splicing of exRNA in human serum. (A) We used RT- PCR to screen splicing in DM1 (N = 5) and UA control (N = 4) serum exRNA samples of several transcripts mis-spliced in DM1 muscle biopsies. ²⁸ Splicing in normal human muscle tissue served as a control. Transcript name is shown on the left and target exon/intron on the right. PCR cycle number was 36 (VPS39, CAPZB) or 40 (MBNLl, MAPT, CAMK2B,

ARFGAP2, ALPK3). Examination of VPS39 in an additional 9 DM1 and 7 UA samples and ALPK3 in an additional 4 DM1 and 3 UA samples also showed no difference in splicing patterns by genotype. Alternative splicing patterns in serum appeared similar in DM1 and UA for all transcripts tested. (B) Alternative splicing by RT-PCR and gel electrophoresis of INSR exon 11 (N = 14 DM1, 10 UA), MBNL2 exon 6 (N = 9 DM1, 9 UA), SOS1 exon 25 (N = 14 DM1, 12 UA), and NFIX exon 7 (N = 14 DM1, 12 UA) in human serum exRNA. PCR cycle number was 36 (MBNL2, SOS1, NFIX), and 40 (INSR). Control muscle cDNA was diluted 1 : 100 and amplified in the same PCR reaction as serum samples. Individual data points represent quantitation of splicing of all serum samples examined. Error bars = mean ± s.e.m.

Figures 6A-E. DMD genetic markers in urine exRNA. We used RT-PCR and

DNA sequencing to examine urine exRNA from five DMD subjects with frame-shifting deletions of exons 18 - 22 (Subject 1; SI), exons 51 - 53 (S2), exons 45 - 52 (S3, S4), and exons 24 - 43 (S5). Deletions in S2, S3, and S4 are candidates for exon skipping antisense oligonucleotide drugs to restore the DMD reading frame. (A) Using RT-PCR, we detected DMD deletion mRNAs corresponding to the genetic mutation for all five subjects (SI - S5). The boxes show the exons amplified, "bp" = base pairs; "L" = DNA ladder. (B) DNA sequencing of extracted bands in a) identified the deletions for S 1 and S2. (C) RT-PCR analysis of urine and serum exRNA from an unaffected subject appears similar to muscle tissue, while those from a subject with Becker muscular dystrophy (BMD) due to a t-to-g substitution in intron 67 identifies a second larger band. (D) DNA sequencing of the lower BMD bands showed normal splicing of exon 67/68, while the upper BMD band is a heteroduplex containing both the normally spliced DMD transcript and a 2 ^nd transcript that includes the 1 ^st five nucleotides of intron 67. (E) Diagram of the cryptic splice site in intron 67 that shifts the reading frame and produces a poorly functional dystrophin protein.

Figures 7A-H. Characterization of exRNA in human urine and serum from DM1 and unaffected (UA) control subjects. (A) Nanoparticle analysis in human urine and serum samples. We determined nanoparticle size and concentration in urine and serum samples from DM1 (N = 12 urine, 10 serum) and UA control (N = 9 urine, 8 serum) subjects. For accurate measurements, we diluted serum samples 1 : 1000 and urine either 1 : 10 or 1 :20 in saline to stay in the target concentration range of 1.0 x 10 ⁸ and 2.5 x 10 ⁹ particles/milliliter (see Methods). Representative traces of nanoparticle size from DM1 and UA subjects in urine (left) and serum (right) are shown. (B) Mean nanoparticle size from each individual sample in urine (left) and serum (right) ± s.e.m. Note that nanoparticle concentration in urine was lower than serum, while mean particle size was higher. (C) Using ultracentrifugation and Trizol (see Methods), we isolated exR A from urine (N = 14 DM1 and 12 UA) and serum (N = 5 DM1 and 7 UA) samples and examined optical density using a microvolume spectrophotometer (Nanodrop). Representative optical density spectra of exRNA urine (left) and serum (right) are shown. Vehicle (water) served as reference. The peak of 268 nm reflects residual phenol that was used to purify the RNA (Krebs et al., Anal Biochem 387, 136-138 (2009)), which seemed to have no effect on

electropherogram analysis (E) or cDNA synthesis (Figures 2A-C, 9A-B, 10, 11A-B). (D) We used capillary gel electrophoresis (Agilent Bioanalyzer) to analyze the concentration, size distribution, and quality of exRNA in urine (N = 22 DM1; N = 21 UA controls) and serum (N = 5 DM1 and 7 UA controls). Using the RNA

concentration, we determined the quantity of RNA recovered per volume of urine (left) or serum (right) starting material (range 38 ml - 120 ml urine; 5.5 - 8.5 ml serum). Individual data points from each sample are shown. Error bars represent the mean ± s.e.m. (E) Representative electropherogram traces of exRNA size in nucleotides from urine (left) and serum (right). Most of the species are < 200 nt. (F) RNA integrity number (RIN) in urine (left) and serum (right), as calculated from electropherogram traces using a software algorithm (Agilent). RIN results for three DM1 and two UA serum samples were read as undetermined by the software algorithm. Error bars represent mean ± s.e.m. (G) qPCR analysis of GAPDH and DMPK gene expression in human urine samples (N = 14 DM1 and 13 UA controls). Individual data points indicate the mean of duplicate assays from each individual sample examined. Error bars represent mean ± S.E.M. ** P = 0.003, *** P = 0.0002; /-test with Welch's correction. (H) Serum mRNA expression of reference gene GAPDH (left) and DMPK relative to GAPDH (right) by qPCR. Error bars represent mean ± s.e.m.

Figures 8A-B. Regulation of Map3k4, Claspl, and Ncor2 splicing by MBNL1 protein and response to ASO treatment. (A) We used RT-PCR to analyze alternative splicing of Map3k4, Claspl, and Ncor2 in gastrocnemius muscles from 2 mouse models of DM1, the Mbnll knockout (Kanadia et al, Science 302, 1978-1980 (2003)) (Mbnll ^ΔΕ3/ΔΕ3 ; N = 4) and HSA ^LR transgenic (Mankodi et al., Science 289, 1769-1773 (2000)) (N = 3), and FVB wild-type (N = 3). The graph shows quantitation of alternative exon 17 splicing in each individual replicate. * *** P < 0.0001 ( 1-way ANOVA). (B) We treated HSA ^LR with either saline or ASO 445236 (Wheeler et al, Nature 488, 111-115 (2012)) (N = 4 each) using a dose of 25 mg/kg twice weekly for 4 weeks, and analyzed alternative splicing in quadriceps muscles from these mice by RT-PCR. Gastrocnemius muscles from untreated FVB wild type (N = 2) served as controls. The graph shows quantitation of alternative exon 17 splicing in each individual replicate. **** P < 0.0001 (1-way ANOVA). These data demonstrate that alternative splicing of Map3k4 is regulated by Mbnll protein and that mis-splicing of Map3k4 in the HSA ^LR mouse model of DM1 is rescued by ASO treatment, similar to ASO rescue of other alternatively spliced exons that are regulated by MBNLl protein.

Figures 9A-C. Alternative splicing in urine exRNA isolated from DM1 and unaffected subjects. (A,B) We used RT-PCR to screen 23 candidate DM mis- regulated splice events (Nakamori et al., Ann Neurol 74, 862-872 (2013)) in urine exRNA from DM1 (N = 5) and UA control (N = 4) subjects. Splicing in normal muscle tissue served as a control. Transcript name is shown on the left and target exon/intron on the right. PCR cycle number was 34 (ARFGAP2), 36 (CACNA1S, ALPK3, COPZ2, ANK2, CAPZB, GFPT1, IMPDH2, MAPT, BIN1, FN1, NRAP, OPA1, PHKA1, UBE2D3), and 40 (CAMK2B, ATP2A1, CLCN1, KIF13A, DMD, LMNA). ALPK3 band intensity was variable, requiring 40 cycles to identify bands in an additional 7 DM1 and 4 UA samples (not shown). Splicing of CAMK2B, ARFGAP2, and CAPZB in an additional 5 DM1 and 3 UA control samples was identical to samples shown above. We detected the ATP2A1 exon 22 exclusion band in 6/8 DM1 samples and 2/7 UA controls. (C) Quantitation of alternative splicing by RT-PCR of KIF13A and DMD in DM1 (N = 14), MDC (N = 6 or 5), and UA (N = 14 or 12) subjects. Note that KIF13A bands were absent in 1 DM1 and 4 UA subjects. Error bars = s.e.m.

Figures 10A-B. Alternative splicing in urine cell pellet RNA isolated from DM1 and unaffected subjects. We isolated total RNA from urine cell pellets obtained from DM1 (N = 9), MDC (N = 4), and UA (N = 9) subjects and analyzed alternative splicing by RT-PCR and gel electrophoresis. (A) Representative gel images showing alterative splicing of human insulin receptor (INSR) exon 11, MBNL2 exon 6, SOSl exon 25, MBNLl exon 7, CLASPl exon 20, MAP3K4 exon 17, NFIX exon 7, NCOR2 exon 45a, VPS39 exon 3, and MAPT exons 2 and 3. PCR cycle number was 36 (INSR, MBNL2, SOSl, CLASPl, MAP3K4, NFIX, NCOR2, VPS39) or 37 (MBNLl, MAPT). Control muscle cDNA was diluted 1:50 (MAPT) or 1: 100 (INSR, MBNL2, SOSl, MBNLl, CLASPl, MAP3K4, NFIX, NCOR2, VPS39) and amplified in the same PCR reaction as urine samples. (B) Individual data points represent quantitation of splicing of all individual urine samples examined. ** = mean difference 28.1, 95% CI of difference 11.08 to 45.14; * = mean difference 15.33, 95% CI of difference 1.973 to 28.69 (MBNL2) and mean difference 21.61, 95% CI of difference 3.807 to 39.41 (NCOR2).

Figures 11A-B. RT-PCR analysis of alternative splicing in human and mouse kidney and skeletal muscle. (A) Urine EVs are derived from the kidney and urinary tract (Erdbrugger and Le, J Am Soc Nephrol 27, 12-26 (2016)), and splicing of transcripts derived from these tissues may explain the different splicing pattern of several transcripts in urine exRNA than in skeletal muscle (Figs 2A-C, 9A-B). To determine if the urinary tract may be the predominant source of urine exRNA, we screened alternative splicing of 12 splice events in commercially available mRNA from normal human bladder (B), urothelial (Ut; transitional epithelial) cells, kidney (K) and muscle (M) tissue by RT-PCR using random primers and 26 cycles for each transcript. (B) Percent exon inclusion of each transcript shown in (A) and mean percent exon inclusion ± standard error of the mean (SEM) of urine ex-RNA and urine cell pellets (Figures 2A-C, 9A-B, 10A-B). "n/d" = not done.

Figures 12A-D. We examined splicing of 6 alternatively spliced transcripts by RT-PCR in bladder (N = 3), kidney (N = 2), and muscle tissue (N = 2) from FVB wild-type mice and muscle tissue from the HSA ^LR transgenic mouse model of DM1

(N = 2) using oligo dT and 26 cycles for each transcript. Transcript name is shown on the left and target exon on the right. The percent exon inclusion is indicated below each lane. (A) Urine EVs are derived from the kidney and urinary tract (6), and splicing of transcripts derived from these tissues may explain the different splicing pattern of several transcripts in urine exRNA than in skeletal muscle (Figs 2, S7). To determine if the kidney may be the predominant source of urine exRNA, we screened alternative splicing of 10 transcripts in commercially available mRNA from normal human kidney (k) and muscle (m) tissue (Ambion) by RT-PCR using random primers and 26 cycles for each transcript. (B) We examined splicing of 6 alternatively spliced transcripts by RT-PCR in bladder (N = 3), kidney (N = 2), and muscle tissue (N = 2) from FVB wild-type mice and muscle tissue from the HSA ^LR transgenic mouse model of DM1 (N = 2) using oligo dT and 26 cycles for each transcript. Transcript name is shown on the left and target exon on the right. The percent exon inclusion is indicated below each lane. (C) We used qPCR to analyze expression of genes DMPK, GTF2B, GAPDH, and CKM in mRNA derived from human kidney and muscle tissue. (Upper) the mean qPCR cycle threshold (Ct) of duplicate assays for each gene is shown. (Lower) DMPK expression mRNA expression level normalized to reference genes GTF2B (left) and GAPDH (right). Expression of DMPK in kidney tissue indicates kidney is a potential source of these transcripts in urine exRNA. (D) DMD transcripts expressed in human kidney (k) and muscle (m) tissue using RT-PCR. "E" = empty lane; "Lad" = DNA ladder; "NT" = nucleotides.

DETAILED DESCRIPTION

In DM1 patients, pre-mRNA splicing outcomes in muscle biopsies are biomarkers of disease severity ⁴ , while in DM1 mice they serve as sensitive indicators of therapeutic drug effects ⁵ · ⁶ . Less invasive biomarkers to assess disease state and response to therapy in DM are currently unavailable, and optimal outcome measures of therapeutic success remain undefined. As a result, a recent clinical trial of an antisense oligonucleotide (ASO) drug for DM1 required participants' consent to multiple muscle biopsies to monitor splicing outcomes in response to therapy and was restricted to adult patients ¹ . This experimental drug for DM1 is designed to induce knockdown through the RNase H pathway of mutant transcripts, thereby rescuing muscle cells from the pathogenic effects of splicing mis-regulation ⁶ · ⁸ .

Extracellular vesicles (EVs) include exosomes, microvesicles, and other membrane- encased nanoparticles released and taken up by cells as a form of extracellular

communication ⁹ . EVs in serum and urine contain mRNA and non-coding RNAs, including microRNA (miRNA), termed exRNAs, released from different tissues and can serve as genetic biomarkers of cancers and other disease states ^10"12 . Mutations, deletions, translocations, and transcriptome variations also have been shown extensively in EVs, especially for cancers ^13"15 . Differentiated skeletal muscle cells in culture release EVs ^{16, 17} and a handful of miRNA biomarkers and several protein signatures have been identified in serum of muscular dystrophy patients ¹⁸ . However, the capacity of muscle-derived exRNA in urine to serve as biomarkers for muscular dystrophies seems unlikely given that they would be released into the blood circulation and would be unable to pass through the glomerular filtration of serum in the kidney ¹⁹ . The present results demonstrate that RNA splice products in human urine have sufficient sensitivity and specificity to be robust biomarkers of muscular dystrophies.

As shown herein, mRNA splicing patterns in "liquid biopsies" present a rich source of personalized biomarkers with applications to a number of genetic diseases. For DM1, we found 10 alternative splice variants in urine that serve as a robust composite biomarker of DM1 disease activity. Mis-regulated alternative splicing outcomes in muscle tissue were sensitive indicators of therapeutic response in DM1 mice ^{5, 6} and disease activity in DM1 patients ⁴ . Indeed, splice products in muscle biopsies were used in a recent clinical trial as measures of ASO activity in DM1 patients ¹ . The present methods using splicing outcomes provides powerful biomarkers of DM1, in part because the disease mechanism involves mis- regulated alternative splicing; in addition, the ratiometric measurements of exon inclusion/exclusion described herein are inherently more sensitive than the unidirectional changes that are typical of most biomarkers.

The detection of differential splicing in urine and not in serum was surprising, and suggests that the source of exRNA in these biofluids may be different and that the primary source in serum is unlikely to be muscle tissue. Because DM1 is primarily a disease of skeletal muscle, heart, and the central nervous system (CNS), it is counter-intuitive that exRNA reflecting the characteristic mis-regulated splicing events appears in urine rather than in blood, as exRNA has not been shown to pass from the blood through the proximal tubules of the kidney ²⁹ . In earlier pre-clinical studies, therapeutic ASOs induced target knockdown and exon skipping in kidney tissue of mice and non-human primates ⁸ · ³⁰ , suggesting ASOs could have similar effects in human kidney and other tissues lining the urinary tract that release exRNA into the urine. The potential to evaluate exRNA splicing outcomes as pharmacodynamic biomarkers in urine has the advantage of being non-invasive and can be repeated routinely over the course of treatment to evaluate efficacy. For example, due to the need for general anesthesia and the absence of a therapeutic benefit, muscle biopsies generally are avoided in children with DM1. Consequently, detailed study of splicing outcomes in children with DM1 remains an unmet medical need. Urine exRNA should enable comprehensive non-invasive investigation of splicing outcomes in children with DM for the first time, facilitate clinical trials to these patients earlier, and enable convenient titration of dose. The shared pathogenic mechanism of alternative splicing misregulation in DM1 and DM2 ⁴ suggests urine exRNA also may be useful for monitoring disease activity in DM2 patients.

For DMD, the urine splice products are more than traditional biomarkers: they are personalized genetic markers that are designed specifically for each individual patient and enable the possibility to monitor splice-shifting ASO drug effects ¹¹ ' ²⁸ . Dystrophin protein measurement in biopsy tissue is presently used as a surrogate marker of drug effect that led to the accelerated approval of eteplirsen by the U.S. Food and Drug Administration ³¹ .

However, monitoring the ratio of skipped/unskipped DMD splice products in urine during the course of treatment may be used to complement RT-PCR analysis of muscle biopsies and/or in place of dystrophin protein measurement as a surrogate marker of therapeutic effect as newer and better splice-shifting drugs are developed. The finding of a DMD cryptic splice site responsible for Becker MD phenotype in an individual with dystrophinopathy further suggests the value of liquid biopsies as a means to identify novel splice variants that may help correlate genotype with phenotype for a number of diseases for which non-invasive biomarkers are unavailable. For example, in patients with Hutchinson-Gilford progeria syndrome (HGPS), point mutations in the LMNA gene activate a weak splice site in exon 11 that shortens the transcript and produces a truncated progerin protein ³² . ASOs that reduce use of this weak splice site are being evaluated as strategy to treat HGPS ³³ . The presence of LMNA exon 11 in urine (Figure 9B) shows that the present methods including use of exRNA can be used to monitor drug effects in these patients as well. The present findings also support development of exRNA from urine, serum, or CSF as a biomarker replacement for tissue biopsies in other diseases with altered mRNA splicing, including limb girdle muscular dystrophy type IB, spinal muscular atrophy and amyotrophic lateral sclerosis ^34"36 .

Table A provides a list of exemplary conditions that can be diagnosed, treated, or monitored using the present methods, along with the mutated genes (though note that the mutation may or may not result in altered splicing of that specific gene, or not only that specific gene).

Table A - Diseases associated with aberrant mRNA splicing

Additional diseases are known in the art, including limb-girdle muscular dystrophy type 2B, Miyoshi myopathy, distal myopathy with anterior tibial onset and Fukuyama congenital muscular dystrophy, see, e.g., Scotti and Swanson, Nature Reviews Genetics 17:19-32 (2016), and Touznik et al., Expert Opin Biol Ther. 2014 Jun; 14(6): 809- 19.

Myotonic dystrophy type 1 (DM1)

DM1 is caused by a heterozygous trinucleotide repeat expansion (CTG)n in the 3- prime untranslated region of the dystrophia myotonica protein kinase gene (DMPK); a repeat length exceeding 50 CTG repeats is pathogenic (Musova et al., Am. J. Med. Genet. 149A: 1365-1374, 2009). The CUG repeats form RNA hairpins that bind proteins including muscleblind-like 1 (MBNLl), a splicing regulatory factor; nuclear sequestration of MBNLl prevents its activity and results in aberrant splicing of several genes. As shown herein, a number of pre-mRNAs are aberrantly spliced in urine (see Tables 5 and 6); of those, the following showed differential urine exRNA splicing in DM1 vs MDC and UA controls:

The methods can include determining ratios of properly spliced mRNA to aberrantly spliced mRNA for all or a subset, e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9 or all 10 of the above. In some embodiments, the subset includes 1, 2, 3, 4, or all of MBNL2, MBNLl, SOS1, CLASP 1, MAP3K4, and optionally also INSR.

DM1 affects skeletal and smooth muscle as well as the eye, heart, endocrine system, and central nervous system. Symptoms can include muscle weakness, e.g., in the leg, hand, neck, and/or face; myotonia, e.g., grip myotonia or percussion myotonia; and posterior subcapsular cataracts (which are detectable as red and green iridescent opacities on slit lamp examination).

The methods described herein can also include administering a treatment for DM1, e.g., a treatment that is expected or intended to affect splicing, e.g., to correct splicing or to inhibit or reduce levels of aberrantly spliced transcripts, e.g., as described herein. The present methods can be used to monitor efficacy, e.g., to determine whether the treatment affects splicing, e.g., by detecting a change in the ratio of properly spliced mRNA to aberrantly spliced mRNA. An increase in the properly spliced mRNA, and/or a decrease in aberrantly spliced mRNA, would result in an increase in the ratio and indicates that the treatment is effective. Duchenne muscular dystrophy (DMD) and Becker muscular dystrophy (BMD)

DMD and BMD are inherited progressive muscle disorders that are noninflammatory and not associated with a central or peripheral nerve abnormality. The disease affects the muscles with definite fiber degeneration but without evidence of morphologic aberrations, resulting in progressive muscle wasting, and are caused by defects in the dystrophin gene DMD. See, e.g., Aartsma-Rus et al., J Med Genet. 2016 Mar;53(3):145-51; Flanigan et al., Hum Mutat. 2011 Mar; 32(3): 299-308. In some cases, DMD or BMD are caused by mutations that affect splicing of the transcript, e.g., acceptor or donor splice site mutations. The present methods can be used to detect these alternative mRNA splice variants or mRNA of different lengths. The methods can include determining ratios of properly spliced DMD mRNA to aberrantly spliced DMD mRNA.

The methods described herein can also include administering a treatment for DMD or BMD, e.g., a treatment that is expected or intended to affect splicing, e.g., to correct splicing of the dystrophin transcripts, reduce levels of aberrant transcripts, or to produce transcripts that encode functional dystrophin protein. In general, ASOs for DMD are used to induce new splicing changes that serve to restore the open reading frame rather than correct aberrant splicing. It may be possible that some DMD mutations that lead to a Duchenne phenotype and that ASOs could be designed to treat this. Alternative splicing of DMD transcripts typically includes exon 71, 78, and perhaps exon 68. The remainder of the DMD exons are spliced constitutively. Urine RNA can also be used to identify novel aberrant splicing, as in our Becker patient with a cryptic splice site.

The present methods can be used to monitor efficacy, e.g., to determine whether the treatment affects splicing, e.g., by detecting a change in the ratio of properly spliced mRNA (or mRNA of a desired size or sequence) to aberrantly spliced mRNA (or mRNA of a non- desired size or sequence). An increase in the properly spliced mRNA or mRNA of a desired size or sequence, and/or a decrease in aberrantly spliced mRNA or mRNA of a non-desired size or sequence, would result in an increase in the ratio and indicates that the treatment is effective.

Limb girdle muscular dystrophy type IB/ Hutchinson-Gilford progeria syndrome (HGPS)/ LMNA-linked dilated cardiomyopathy (DCM)/Familial partial lipodystrophy type 2 (FPLD2)

Mutations in the lamin A (LMNA) gene that result in aberrant splicing are associated with a number of hereditary disorders. See Scotti and Swanson, Nature Reviews Genetics 17: 19-32 (2016).

Hutchinson-Gilford progeria syndrome (HGPS) is caused by mutations within the LMNA gene that lead to increased usage of an internal splice site, resulting in alternative lamin A transcript with internal deletions of 150 nucleotides (LMNA G608G (GGOGGT) mutation); see Eriksson et al., Nature 423, 293-298 (2003); Rodriguez et al., Eur J Hum Genet. 2009 Jul; 17(7): 928-937.

Limb-girdle muscular dystrophies (LGMD) are a heterogeneous group of muscle disorders; symptoms begin in the voluntary muscles of the hips and shoulders. See Nigro and Saverese, Acta Myol. 2014 May;33(l): l-12.

In familial partial lipodystrophy type 2 (FPLD2), a G>C mutation leads to aberrant intron 8 retention, nonsense-mediated decay and may lead to translation of a truncated lamin A/C. see Scotti and Swanson, Nature Reviews Genetics 17: 19-32 (2016)).

LMNA-linked dilated cardiomyopathy (DCM) is associated with an alternative 3' splice site generated by an A>G mutation (c. 640-10A>G); see Scotti and Swanson, Nature Reviews Genetics 17: 19-32 (2016)).

The methods described herein can also include administering a treatment for a conditions associated with mis-splicing of LMNA, e.g., a treatment that is expected or intended to affect splicing, e.g., to correct splicing of the dystrophin transcripts, reduce levels of aberrant transcripts, or to produce transcripts that encode functional dystrophin protein. The present methods can be used to monitor efficacy, e.g., to determine whether the treatment affects splicing, e.g., by detecting a change in the ratio of properly spliced mRNA (or mRNA of a desired size or sequence) to aberrantly spliced mRNA (or mRNA of a non-desired size or sequence). An increase in the properly spliced mRNA or mRNA of a desired size or sequence, and/or a decrease in aberrantly spliced mRNA or mRNA of a non-desired size or sequence, would result in an increase in the ratio and indicates that the treatment is effective.

Spinal muscular atrophy

SMA is associated with mutations in the SMN1 gene (including c.922 + 6 T/G deletion) and loss of SMN full-length protein; see Lorson et al., Proc. Natl Acad. Sci. USA 96, 6307-6311 (1999); Lefebvre et al., Cell 80, 155-165 (1995); Scotti and Swanson, Nature Reviews Genetics 17: 19-32 (2016)). Approximately 98% of spinal muscular atrophy (SMA) patients have a survival of motor neurons (SMN) gene that has been deleted or mutated. SMN is part of a large multi-protein complex (with additional proteins, including Gemeins 2- 7) that is necessary for biogenesis of small nuclear RNA ribonucleoproteins (snRNPs), which are major components of pre-mRNA splicing machinery. Genetic alterations in the SMN gene result in the reduced capacity for snRNP assembly, and defects in RNA splicing. See Dreyfuss, Eukaryon, 6:75-79, 2010, herein incorporated by reference in its entirety.

The methods described herein can also include administering a treatment for SMA, e.g., a treatment that is expected or intended to affect splicing, e.g., to correct splicing of the dystrophin transcripts, reduce levels of aberrant transcripts, or to produce transcripts that encode functional dystrophin protein. Splicing of SMN2 normally involves skipping of exon 7 in the majority of SMN2 transcripts. ASO treatment of SMA involves increasing inclusion of SMN2 exon 7, which results in an increase of SMN protein levels, the same protein that is lost by mutations in SMN1. The present methods can be used to monitor efficacy, e.g., to determine whether the treatment affects splicing, e.g., by detecting a change in the ratio of properly spliced mRNA (or mRNA of a desired size or sequence) to aberrantly spliced mRNA (or mRNA of a non-desired size or sequence). An increase in the properly spliced mRNA or mRNA of a desired size or sequence, and/or a decrease in aberrantly spliced mRNA or mRNA of a non-desired size or sequence, would result in an increase in the ratio and indicates that the treatment is effective.

Amyotrophic lateral sclerosis

Approximately 60%-70% of patients with sporadic Amyotrophic lateral sclerosis (ALS) display a loss of the astrocytic glutamate transporter protein EAAT2 (also known as SLC1A2) in motor cortex and spinal cord. See Rothstein et al., Ann. Neurol. 38:73-84, 1995, herein incorporated by reference in its entirety. Defective pre-mRNA splicing in the motor cortex and spinal cord is responsible for the loss of EAAT2 protein. This defective splicing is caused by a defect in a splicing regulatory factor, rather than a mutation in the EAAT2 gene that causes alternative aberrant splicing or a defect in a general splicing apparatus, such as the spliceosome. The defective pre-mRNA splicing process for EAAT2 can skip normal 5' and 3 ' splice sites (donor and acceptor splicing sites), or use inappropriate 5' and 3 ' splice sites (i.e., other than the normal GU or AU for the donor site, and AG or AC for the acceptor site), resulting in multiple abnormal RNAs in ALS patients. The aberrant splicing results in transcripts that partially retain introns or skip exons, as well as transcripts that have exonic sequences at random sites. Two aberrantly spliced EAAT2 mRNAs are found predominantly in sporadic ALS patients. These include an mRNA transcript that partially retains intron 7 and an mRNA transcript that skips exon 9 of the gene. The intro 7-retaining RNA causes a dominant-negative effect on normal EAAT2 that has been shown to result in a loss of protein and activity. See, e.g., Honig et al., Neurology. 2000 Oct 24;55(8): 1082-8; Lin et al., Neuron. 1998 Mar;20(3):589-602; Meyer et al., J Neurol Sci. 1999 Nov 15;170(l):45-50. Certain forms of ALS are associated with the presence of mutations in the TARDP (c. 991C>A), (c.l009A>G) and FUS (c. 15660T), (c. 1561T>G) genes; see Scotti and Swanson, Nature Reviews Genetics 17: 19-32 (2016)); Bai and Lipton, 20(3):363-366 (1998); Zhou et al., PLoS Genet. 2013 Oct;9(10):el003895; Orozco and Edbauer, J Mol Med (Berl). 2013

Dec;91(12): 1343-54; Belzil et al., J Mol Med (Berl). 2013 Dec;91(12): 1343-54.

The methods described herein can also include administering a treatment for conditions associated with mis-splicing of SLC1 A2, TARDP, or FUS, e.g., a treatment that is expected or intended to affect splicing, e.g., to correct splicing of the dystrophin transcripts, reduce levels of aberrant transcripts, or to produce transcripts that encode functional dystrophin protein. The present methods can be used to monitor efficacy, e.g., to determine whether the treatment affects splicing, e.g., by detecting a change in the ratio of properly spliced mRNA (or mRNA of a desired size or sequence) to aberrantly spliced mRNA (or mRNA of a non-desired size or sequence). An increase in the properly spliced mRNA or mRNA of a desired size or sequence, and/or a decrease in aberrantly spliced mRNA or mRNA of a non-desired size or sequence, would result in an increase in the ratio and indicates that the treatment is effective.

Methods of Diagnosis and Monitoring

Included herein are methods for diagnosing and monitoring subjects with a disease associated with a genetic mutation that results in aberrant splicing, e.g., myotonic dystrophy type 1 (DM1), Duchenne muscular dystrophy (DMD), Becker muscular dystrophy (BMD), limb girdle muscular dystrophy type IB (LGMD1B), LMNA-linked dilated cardiomyopathy (DCM); Hutchinson-Gilford progeria syndrome (HGPS); Familial partial lipodystrophy type 2 (FPLD2), spinal muscular atrophy (SMA) and amyotrophic lateral sclerosis (ALS). The methods can also be used to diagnose and monitor subjects with other splicing diseases, e.g., progeria.

The methods rely on detection of ratios of properly spliced (which can include transcripts that are spliced by an ASO) to mis-spliced isoforms of affected transcripts in urine samples. As used herein, "properly spliced" means that the transcript has a desired splice pattern, e.g., has wild-type splicing, or is spliced in a way that is desired, e.g., to produce a functional protein. For example, in some embodiments, the DMD exons targeted by ASOs are constitutively spliced, meaning they are always included in DMD patients and UA individuals. Treatment with ASOs in DMD is designed to induce a new unique splice event absent in DMD patients or UA individuals, and the ratio of the inclusion of the unique splice site (properly spliced) to inclusion of the constitutive splice (mis-spliced) provides a convenient estimation of ASO drug effects in urine a new unique splice event absent in DMD patients or UA individuals; in this case, a transcript that was "properly spliced" in a subject with DMD being treated with these ASOs would include the unique splice event. The methods can include determining that the mR A is of a desired size or sequence (i.e., "properly spliced"), or is of a non-desired size or sequence ("mis-spliced").

The methods include obtaining a urine sample from a subject determining levels of properly spliced and aberrantly spliced extracellular transcripts, and determining a ratio of properly spliced to mis-spliced transcript in the sample. The methods can include comparing the ratio with one or more reference ratios, e.g., a control reference that represents a normal ratio of properly spliced:mis-spliced transcript, e.g., a level in an unaffected subject, and/or a disease reference that represents a ratio associated with the disease. For example, in some embodiments a reference ratio of properly spliced:mis-spliced transcripts in an unaffected subject may approach 1 :0, since there would not be expected to be a large number of mis- spliced transcripts in such individuals.

Various methods are well known within the art for determining levels of properly spliced and aberrantly spliced extracellular transcripts. These methods can include identification and/or isolation and/or purification of a transcript from a sample. An "isolated" or "purified" biological marker is substantially free of cellular material or other contaminants from the cell or tissue source from which the biological marker is derived i.e. partially or completely altered or removed from the natural state through human intervention. For example, nucleic acids contained in the sample can be isolated according to standard methods, for example using filtration, centrifugation, or other methods of purification to obtain a sample that contains extracellular transcripts but does not contain cells or cellular transcripts. The methods can include using chemical solutions nucleic acid-binding resins following the manufacturer's instructions. In one example, the entire volume of urine is centrifuged, e.g., at 2,000-3,000 x g, e.g., at 2,450 x g for 5-15 minutes, e.g., 10 minutes at room temperature, and then the supernatant is passed through a filter, e.g., a 0.8 μηι filter, before being ultracentrifuged, e.g., at 100,000 x g 2 hours at 4° C, to pellet the RNA. Then the supernatant is removed and, RNA is extracted from the translucent ribonucleoprotein pellet, e.g., using Trizol (Life Technologies) according to manufacturer instructions. To enhance RNA pellet visibility, linear acrylamide (Ambion) or other reagents can be added.

The transcripts can be evaluated using methods known in the art, e.g., using polymerase chain reaction (PCR), reverse transcriptase polymerase chain reaction (RT-PCR), quantitative or semi -quantitative real-time RT-PCR, digital PCR i.e. BEAMing ((Beads, Emulsion, Amplification, Magnetics) Diehl (2006) Nat Methods 3:551-559) ; RNAse protection assay; Northern blot; various types of nucleic acid sequencing (Sanger, pyrosequencing, NextGeneration Sequencing); fluorescent in-situ hybridization (FISH); or gene array/chips) (Lehninger Biochemistry (Worth

Publishers, Inc., current addition; Sambrook, et al, Molecular Cloning: A Laboratory Manual (3. Sup.rd Edition, 2001); Bernard (2002) Clin Chem 48(8): 1178-1185; Miranda (2010) Kidney International 78: 191-199; Bianchi (2011) EMBO Mol Med 3:495-503; Taylor (2013) Front. Genet. 4: 142; Yang (2014) PLOS One

9(1 l):el 10641); Nordstrom (2000) Biotechnol. Appl. Biochem. 31(2): 107-112; Ahmadian (2000) Anal Biochem 280: 103-110. In some embodiments, high throughput methods, e.g., protein or gene chips as are known in the art (see, e.g., Ch. 12, Genomics, in Griffiths et al., Eds. Modern genetic Analysis, 1999,W. H. Freeman and Company; Ekins and Chu, Trends in Biotechnology, 1999, 17:217-218;

MacBeath and Schreiber, Science 2000, 289(5485): 1760-1763; Simpson, Proteins and Proteomics: A Laboratory Manual, Cold Spring Harbor Laboratory Press; 2002; Hardiman, Microarrays Methods and Applications: Nuts & Bolts, DNA Press, 2003), can be used to detect the presence and/or level of different splice isoforms. multiple- exon-skipping detection assay (MESDA) can also be used (see Singh et al., 2012, PLoS One. 2012;7(l l):e49595). Measurement of the level of different splice isoforms can be direct or indirect. For example, the abundance levels of various differently spliced isoforms can be directly quantitated, e.g., based on size or the presence or absence of a selected sequence. In some embodiments a technique suitable for the detection of alterations in the structure or sequence of nucleic acids, such as the presence of deletions, amplifications, or substitutions, can be used for the detection of different splice isoforms. Gene arrays are prepared by selecting probes which comprise a polynucleotide sequence, and then immobilizing such probes to a solid support or surface. For example, the probes may comprise DNA sequences, RNA sequences, co-polymer sequences of DNA and RNA, DNA and/or RNA analogues, or combinations thereof, which detect various spliced isoforms. The probe sequences can be synthesized either enzymatically in vivo, enzymatically in vitro (e.g. by PCR), or non-enzymatically in vitro.

In some embodiments, the methods can be used to diagnose a condition described herein; for example, when the ratio of properly spliced:mis-spliced transcripts in a subject (e.g., a subject who has one or more symptoms associated with the disease) is comparable to a reference ratio in a representative subject with the disease, then the subject can be diagnosed with the disease. In some embodiments, once it has been determined that a person has a disease described herein, e.g., DM1, DMD, BMD, LGMD IB, SMA, or ALS, then a treatment, e.g., as known in the art or as described herein, can be administered.

Suitable reference values can be determined using methods known in the art, e.g., using standard clinical trial methodology and statistical analysis. The reference values can have any relevant form. In some cases, the reference comprises a predetermined value for a meaningful ratio, e.g., a control reference level that represents a normal level ratio, e.g., a level in an unaffected subject, and/or a disease reference that represents a ratio associated with the disease, e.g., a level in a subject having a disease as described herein, e.g., DMD, DM1, BMD, LGMD, HGPS, DCM, HGPS, FPLD2, SMA, or ALS.

The predetermined ratio can be a single cut-off (threshold) value, such as a median or mean, or a ratio that defines the boundaries of an upper or lower quartile, tertile, or other segment of a clinical trial population that is determined to be statistically different from the other segments. It can be a range of cut-off (or threshold) ratios, such as a confidence interval. It can be established based upon comparative groups, such as where association with presence of disease in one defined group is a fold higher, or lower, (e.g., approximately 2-fold, 4-fold, 8-fold, 16-fold or more) than the presence of disease in another defined group. It can be a range, for example, where a population of subjects (e.g., control subjects) is divided equally (or unequally) into groups, such as a low-risk group, a medium-risk group and a high-risk group, or into quartiles, the lowest quartile being subjects with the lowest risk and the highest quartile being subjects with the highest risk, or into n-quantiles (i.e., n regularly spaced intervals) the lowest of the n-quantiles being subjects with the lowest risk and the highest of the n-quantiles being subjects with the highest risk.

In some embodiments, the predetermined ratio is a ratio in the same subject, e.g., at a different time point, e.g., an earlier time point.

Subjects associated with predetermined values are typically referred to as reference subjects. For example, in some embodiments, a control reference subject does not have a disorder described herein (e.g., DMD, DM1, BMD, LGMD, HGPS, DCM, HGPS, FPLD2, SMA, and ALS). In some cases it may be desirable that the control subject is a first or second degree relative of the subject to be tested.

A disease reference subject is one who has (has been diagnosed with) a disease as described herein, e.g., DMD, DM1, BMD, LGMD, HGPS, DCM, HGPS, FPLD2, SMA, or ALS.

Thus, in some cases the ratio of properly spliced:mis-spliced in a subject being less than a reference ratio is indicative of a clinical status (e.g., indicative of presence of a disorder as described herein, e.g., DMD, DM1, BMD, LGMD, HGPS, DCM, HGPS, FPLD2, SMA, or ALS), or indicative of an ineffective therapy. In other cases the ratio in a subject being greater than or equal to the reference ratio is indicative of the absence of disease, or an effective therapy. In some embodiments, the amount by which the ratio in the subject is the less than the reference ratio is sufficient to distinguish a subject from a control subject, and optionally is a statistically significantly less than the ratio in a control subject. In cases where the ratio in a subject being equal to the reference ratio, the "being equal" refers to being approximately equal (e.g., not statistically different).

The predetermined ratio can depend upon the particular population of subjects (e.g., human subjects) selected. For example, an apparently healthy population may have a different 'normal' range of ratios than will a population of subjects which have, are likely to have, or are at greater risk to have, a disorder described herein. Accordingly, the predetermined values selected may take into account the category (e.g., sex, age, health, risk, presence of other diseases) in which a subject (e.g., human subject) falls. Appropriate ranges and categories can be selected with no more than routine experimentation by those of ordinary skill in the art. In characterizing likelihood, or risk, numerous predetermined values can be established.

Methods of Treatment

The methods described herein can include administering a treatment of disorders associated with aberrant splicing. In some embodiments, the disorder is

DMD, DM1, BMD, LGMD, HGPS, DCM, HGPS, FPLD2, SMA, or ALS. Generally, the methods include administering a treatment to a subject identified using a method described herein.

As used in this context, to "treat" means to ameliorate at least one symptom of the disorder associated with aberrant splicing. For example, where the disease is a muscular dystrophy, a treatment can result in a reduction in muscle weakness or a reduction in rate of muscle loss or weakening.

Exon-skipping antisense oligonucleotides (ASOs) that correct missplicing can be used, e.g., as described in Siva et al., Nucleic Acid Ther. 2014 Feb 1; 24(1): 69-86; Scotti and Swanson, Nature Reviews Genetics 17: 19-32 (2016). For example, bicyclic-locked nucleic acids (LNAs), ethylene-bridged nucleic acids (ENAs), 2'-0- methyl phosphorothioate AO (20ME-PSs), peptide nucleic acids (PNAs), or phosphorodiamidate morpholino oligomers (PMOs) have been described that correct missplicing in clinical trials and animal models; see, e.g., Brolin and Shiraishi, Artif DNA PNA XNA. 2011 Jan-Mar; 2(1): 6-15; Scotti and Swanson, Nature Reviews Genetics 17: 19-32 (2016); Touznik et al., Expert Opin Biol Ther. 2014

Jun;14(6):809-19. The ASOs can be delivered, e.g., parenterally in liposomal complexes, e.g., cationic lipoplexes, or using a viral vector, e.g., a lentivirus, adenovirus, or adeno-associated virus. See e.g., Jarver et al., Nucleic Acid Ther. 2014;24(l):37-47; Aartsma-Rus et al., Hum Gene Ther. 2014;25(10):885-892,

McNally and Wyatt, J Clin Invest. 2016 Apr 1; 126(4): 1236-8; Imbert et al, Genes 2017, 8(2), 51; doi: 10.3390/genes8020051.

Exon skipping uses antisense oligonucleotides (ASOs) to alter transcript splicing; the present methods can be used to detect these transcripts with desired splicing. These treatments can include antisense oligonucleotide-targeted exon skipping to induce near normal, e.g., for dystrophin, e.g., as described in Aartsma- Rus, Methods Mol Biol. 2012;867:97-116. Clinical trials of ASOs in DMD have been conducted, see, e.g., Koo and Wood, Hum Gene Ther. 2013 May;24(5):479-88; Voit et al, Lancet Neurol. 2014; 13(10):987-996.

An exon 11 antisense oligonucleotide (ASO) that increased lamin C production has been shown to shift the output of LMNA more toward lamin C and reduce levels of the mutant protein in fibroblasts derived from patients with HGPS (Lee et al., J Clin Invest. 2016 Apr 1; 126(4): 1592-602).

Exon skipping ASOs directed against an intron splice silencer in SMN2 increase the amount of full-length SMN transcript in the CNS, restoring SMN to treat subjects with spinal muscular atrophy (SMA) (see Burghes and McGovern, Genes Dev. 2010 Aug 1; 24(15): 1574-1579). ASO drug nusinersen enhanced exon 7 inclusion in a clinical trial, see Chiriboga et al, Neurology. 2016 Mar 8; 86(10): 890- 897.

Specific ASOs for use in exon 51 skipping therapy, e.g., in DMD, include PRO051 (20ME-PS, Netherlands) and AVI-4658 (PMO, UK). A plurality of ASOs can also be used, e.g., to induce exon skipping in multiple exons; see, e.g., Wood et al, Brain. 2010 Apr;133(Pt 4):957-72 See also Fletcher et al, Mol Ther Nucleic Acids. 2012 Oct; 1(10): e48; McClorey et al, Curr Opin Pharmacol. 2005

Oct;5(5):529-34.

Similar methods can be used in DM1, as described in Chamberlain and Chamberlain, Nature Medicine 16: 170-171 (2010). For example, an ASO inhibiting mutant DMPK transcripts can be used, e.g., a 149-bp antisense RNA complementary to the (CUG)13 repeats and to the 110-bp region following the repeats sequence has been described, see Furling et al., Gene Ther. 2003 May; 10(9):795-802. See also Magana and Cisneros, J Neurosci Res. 2011 Mar;89(3):275-85; Thornton et al., Curr Opin Genet Dev. 2017 Jun;44: 135-140; Gao and Cooper, Hum Gene Ther. 2013 May; 24(5): 499-507.

See also Gao et al., Hum Gene Ther. 2013 May; 24(5): 499-507; Wheeler et al, Science 2009, 325, 336-339; Wheeler et al., Nature 2012, 488, 111-115;

Wojtkowiak-Szlachcic et al., Nucleic Acids Res. 2015, 43, 3318-3331; Mulders et al, Proc. Natl. Acad. Sci. USA 2009, 106, 13915-13920; Francois et al., Nat. Struct. Mol. Biol. 2011, 18, 85-87; Cavazzana-Calvo et al., Science 2000, 288, 669-672; Cornetta et al., Mol. Ther. J. Am. Soc. Gene Ther. 2011, 19, 557-566. Small molecule therapeutics can also be used, e.g., PTC124, a 284.24-Da, achiral, 1,2,4-oxadiazole linked to fluorobenzene and benzoic acid rings, which selectively induces ribosomal read-through of premature but not normal termination codons, see Welch et al., Nature 447: 87-91, 2007, and has been used in clinical trials for DMD.

EXAMPLES

The invention is further described in the following examples, which do not limit the scope of the invention described in the claims.

Materials and Methods

The following materials and methods were used in the Examples set forth below.

Human subjects. The Partners Health Service/MGH IRB approved all studies involving human subjects described here. We recruited study participants from the MGH Neuromuscular Diagnostic Center. Three groups were studied: (1) individuals with DM1 (N = 23), (2) individuals with a muscular dystrophy besides DM (N = 8 total; 6 DMD, 1 BMD, 1 FSHD), and (3) individuals with no known muscular dystrophy (unaffected; N = 22) that were either a parent, spouse, or cousin of a study participant with muscular dystrophy. Inclusion criteria for DM1 subjects were age 13 years or older, a diagnosis of DM1 based on genetic testing that identified a DMPK-CTG repeat expansion of > 50, or clinical diagnosis of DM1 and a 1 ^st degree relative with DM1 due to a DMPK-CTG repeat expansion of > 50, and ability to provide informed consent or assent for participation. Inclusion criteria for MDC subjects included known diagnosis of DMD, BMD, or FSHD, ages 13 years or older, and ability to provide informed consent. Inclusion criteria for unaffected individuals were age 18 years or older, no known history of any muscular dystrophy, and ability to provide informed consent. The training cohort consisted of a combined 34 DM1 or UA participants chosen randomly, and the remaining 11 combined DM1 and UA subjects comprised the validation cohort. Prior to participation in the study, informed consent was obtained for blood and/or urine collection from all subjects; due to severe autism, informed consent for the individual with Becker muscular dystrophy was obtained from his mother/legal guardian, according to IRB protocol. Subject information is shown in Tables 1A-1B.

Tables 1A-1B. Clinical data from DM1 subjects.

Table 1A Table IB

(IA) Gender, age, and age range of participating DM1, MDC, and UA subjects.

(IB) We examined urine from a subset of 18 DM1 and 16 UA subjects using urinalysis reagents strips (ACON Laboratories) as a general screening tool for urinary tract disorders, endocrine disorders, or metabolic or systemic diseases that affect kidney function. Specific gravity values (detection range 1.000 - 1.030) tended to be higher in DM1 than in UA subjects (P = 0.0568; Mann-Whitney test). Specific gravity is measure of the kidney's ability to concentrate the urine. Urobilinogen detection range is 0.2 - 1.0 mg/dL (3.5 - 200 μπιοΙ/L). Collection and processing of human urine. Subjects donated urine (range from 20

- 120 milliliters) in a standard specimen container. To remove cells, we centrifuged the entire volume at 2,450 x g for 10 minutes at room temperature, passed the supernatant through a 0.8 μπι filter into sterile 50 ml tubes, and placed on wet ice within 2 hours of collection. We proceeded with exRNA isolation from specimens either immediately or after storage at 4° C overnight. To analyze total RNA in urine cell pellets, we used Trizol (Life Technologies) according to manufacturer recommendations.

Collection and processing of human serum. Blood was collected in two standard red top serum separator tubes (Becton Dickinson), incubated at room temperature for 30 - 45 minutes, and centrifuged at 2,450 x g for 10 minutes at room temperature. To remove any remaining cells, we passed the serum through a 0.8 μπι filter into a sterile 15 ml tube, placed on wet ice within 2 hours of collection, and stored at -80° C. The volume of serum recovered ranged from 5.5 - 8.5 ml. The blood sample from one individual with DM1 was unusable due to hemolysis and total volume of less than 3 ml.

Experimental mice. The MGH IACUC approved all experiments involving mice. HSA ^LR transgenic andMbnll knockout (Mbnll^ ³ ™ ³ ) models of DM1 (both FVB background) have been described ² · ²⁶ . FVB wild-type mice served as controls. HSA ^LR mice that were treated with antisense oligonucleotide (ASO) 445236 received subcutaneous injections of 25 mg/kg twice weekly for 4 weeks, as previously described ⁶ . ASO 445236 was a gift of Dr. Frank Bennett at Ionis Pharmaceuticals (Carlsbad, CA).

Nanoparticle tracking. To determine nanoparticle size and concentration, we used the Nanosight LM10 system and Nanoparticle Tracking Analysis 2.0 analytical software according to manufacturer instructions (Malvern). The system uses a laser beam, light microscope, and CCD camera to visualize and video record particles in liquid suspension moving under Brownian motion. For accurate measurements, we diluted serum samples 1 : 1000 and urine either 1 : 10 or 1 :20 in saline to stay in the target concentration range of 1.0 x 10 ⁸ and 2.5 x 10 ⁹ particles/milliliter. We recorded 60-second videos and analyzed data in auto mode.

Isolation of exR A from hiot ^' lu ids We ultracentrifuged urine and serum samples at 100,000 x g 2 hours at 4° C, removed the supernatant, extracted RNA from the translucent ribonucleoprotein pellet using 700 μΐ Trizol (Life Technologies) according to manufacturer instructions. To enhance RNA pellet visibility, we added 1.4 μΐ linear acrylamide (Ambion) to each sample and mixed well prior to isopropanol precipitation. Pellets were re-suspended in molecular grade water. exRNA analysis. We measured optical density spectra using a microvolume spectrophotometer (Nanodrop). To measure exRNA size, quality, and total mass of recovered, we used chip-based capillary gel electrophoresis according to manufacturer instructions (2100 Bioanalyzer, Agilent Technologies). Using electropherogram traces, a software algorithm (Agilent) automatically determined the RNA integrity number (RIN) based on using a numbering system of 1 (most degraded) to 10 (fully intact) ³⁷ . Quantitative real-time RT-PCR (qPCR). To quantitate DMPK gene expression, we used Taqman qPCR (Applied Biosciences 7500) and standard assays for GAPDH, and GTF2B (Applied Biosciences, FAM-MGB; assay IDs Hs00976255 mi and Hs0275899i gl) as normalization controls. The primer probe set for DMPK was published previously ⁶ . To determine expression levels, we used the mean of duplicate assays from individual samples.

RT-PCR analysis of splicing outcomes. We generated cDNA using Superscript III reverse transcriptase (Life Technologies) and random primers, and performed PCR using Amplitaq Gold DNA polymerase (Life Technologies) and gene specific primers (Tables 2 and 3). We used previously published primers for INSR and ΑΡΤ2Α1 ³ · ³⁸ and designed all other primers using Primer3 software ^{39, 40} . Due to the small size of the exRNA species, we targeted the product size for exon exclusion isoforms to be ~ 100 - 200 nucleotides whenever possible. Total RNA from normal human skeletal muscle and kidney (Ambion AM7982 and AM 7976) served as tissue controls. We separated PCR products using agarose gels, stained with SYBR I green nucleic acid gel stain, and quantitated band intensities using a transilluminator, CCD camera, XcitaBlue™ conversion screen, and Image Lab image acquisition and analysis software (Bio-Rad).

Table 2. PCR primers for human transcripts.

NFIX 7 250 127

NRAP 12 235 130

OPA1 4b 250 196

PHKA1 19 318 141

PHKA1 28 212 173

SOS1 25 208 163

UBE2D3 10 456 406

VPS39 3 274 241

#: SEQ ID NO:

*, progerin

Table 3. PCR primers for mouse transcripts.

#: SEQ ID NO:

Sample size. Splicing patterns in human urine and serum, or even whether alternative splice isoforms are present or detectable in these biofluids, were unknown.

Therefore, we were unable to choose a sample size ahead of time to ensure adequate power to detect disease-specific differences. Instead, we chose a sample size based on splicing outcomes in muscle biopsies ⁴ and a goal of enrolling a similar number of DM1 and UA controls. In mice, we chose sample sizes for splicing analysis in muscle based on previously reported differences in muscle tissue of these models ² · ³ · ⁵ · ⁶ . Mice ranged from 2 to 4 months of age and were chosen randomly by genotype, stratified for sex to allow an approximately equal number of females and males, and examined without blinding.

Statistics. Group data are presented as mean ± s.e.m. We compared groups using an unpaired two-tailed /-test or analysis of variance (ANOVA) as indicated. We used the F test to compare variances between DM1 and UA control samples analyzed by qPCR and RT-PCR (Table 4). In groups with statistically significant difference in variance, we used t-test with Welch's correction to determine differences between groups. A P value < 0.05 was considered significant.

Principle component analysis was performed using R statistical software. The principle component score for each subject was calculated using a linear combination of the 10 splicing outcomes shown in Fig. 2 (INSR, MBNL2, SOS1, MBNLl, CLASP1, MAP3K4, NFIX, NCOR2, VPS39, and MAPT). Table 4. Variance between groups.

We used the F test to compare variances between DM1 and UA control groups for gene expression in urine and serum samples analyzed by qPCR and splicing in serum samples by RT-PCR. In groups with statistically significant difference in variance, we used t-test with Welch's correction to determine differences between groups. The difference in variances of some groups may represent true differences in the two populations, and may be as important as the finding of different means (Figure 1, Figures 7G and7 H, Figures 11A and 1 IB). Predictive model. We used principal component regression to develop a predictive model of DM1 using the splicing quantification of the 10 genes, shown in Fig. 2. The pis package in R, which uses singular value decomposition algorithm for the fitting, was used to implement the model ⁴¹ . For the model, DM1 (N = 23) and UA (N = 22) subjects were randomly assigned to a training cohort that consisted of 17 subjects with DM1 and 17 unaffected controls. For subjects that provided 2 ^nd samples, the mean value of the splicing quantitation measurements was used. Only the first principal component was used for prediction.

Example 1. Characterization of exRNA in biofluids from DM1 and UA subjects

To examine the possibility of detecting biomarkers of muscular dystrophies (MDs) in human biofluids, we analyzed exRNA microarray and raw sequencing data from two previous studies and found that more than 30 transcripts previously reported as "splicing biomarkers" in DM1 muscle biopsy tissue could be detected in control human serum and urine (Tables 5 and 6) ²⁰ - ²¹ . To determine whether splice variants of these transcripts are also detectable in human biofluids, we collected blood and/or urine from 23 subjects with DM1, 22 unaffected (UA) individuals, and 8 MD controls (MDC) (Table 1). First we screened biofluids for the presence of exRNA and found that nanoparticle content was greater in serum than urine, and showed no difference in number or size between DM1 and controls (Figure 7A). Conversely, nanoparticle size spanned a larger range in urine than serum in both DM1 and UA controls (Figure 7B). Optical density curves at 260 nm appeared similar in DM1 and UA controls in both urine and serum (Figure 7C). Using capillary gel electrophoresis, we calculated RNA mass recovered per milliliter of biofluid, exRNA size distribution, and RNA quality (Figure 7D to F). Although the concentration of exRNA was 2.5 - 3 fold greater in serum than urine, the total recovery of exRNA was greater in urine due to larger specimen volume (20 - 120 ml urine vs 5.5 - 8.5 ml serum).

Table 5. Transcripts mis-spliced in DM1 muscle tissue and other muscle transcripts

Gene Systematic

Sequence # Probe Name Avg Name name

TTTTGGATGCACTGAGACCCCGACATTCCT

DMPK NM 004409 89. A 23 P50535 5.66

CGGTATTTATTGTCTGTCCCCACCTAGGAC

GTTCAGAGATCGTTCCTATACATTTCTGTT

INSR* NM 000208 90. A 23 P4764 4.87

CATCTTAAGGTGGACTCGTTTGGTTACCAA

ATCTTTCTGTAACACTTAAAGAATTCCCTC

MBNL2* NM 144778 91. A 24 P56317 4.60

ATTCATTACCTTACAGTGTAAACAGGAGTC

TTATTACCACCACGAGAACCTGTGAGGACA

SOS1* NM 005633 92. A 23 P343808 4.77

CCTGATGTTTTCTCAAGCTCACCACTACAT

TTATCAAGCGTAATGTTACACTTTAAAGGA

CLASP1* NM 015282 93. A 23 P311232 4.57

CAGCAAATAAGAACTTTGTAGAATCCCACC

ATCCTTTCAAACCCTCATGACTGACAAAAA

MBNL1* NM 021038 94. A 23 P357811 6.26

CTCCATGGGGCCAAATCTGCCTGAAGATCA

AAAGATTAAGCCCTGAAGGAAAGGACTTCC

MAP3K4* NM 005922 95. A 23 P42096 5.07

TTTCTCACTGCCTTGAGAGTGACCCAAAGA

ACCTGGTCATGGTGATTTTGTTTAAGGGGA

NFIX* NM 002501 96. A 23 P165295 4.85

TCCCCCTGGAAAGTACTGATGGGGAGCGGC

TTCGATGCGTATTCTGTGGCCGCCATTTGC

NCOR2* NM 006312 97. A 23 P203891 5.84

GCAGGGTGGTGGTATTCTGTCATTTACACA

GGCAAGAACAGCAGGACGCTGGTTTAAAAA

VPS39* NM 015289 98. A 24 P167825 5.08

TAACTCACCGCCAAACCTGTGGAGCAGTGT

ACCAGTTCTCTTTGTAAGGACTTGTGCCTC

MAPT* NM 016835 99. A 23 P207699 5.19

TTGGGAGACGTCCACCCGTTTCCAAGCCTG

TCATATTCATTCCCTGGGATGTTTAGTTAC

KIF13A NM 022113 100. A 23 P214111 4.69

CAGTTTTCCCAAAGTGTTCTGGTAGCATCT

AGAAAATATAGTCACAGGAAACTACTCACG

DMD NM 004019 101. A 23 P321860 4.44

TAAGTAGTAATGATTCTCAAGATCAAAGGG

TCCATCTTCCAGTCCCTGCTTCACTGCTTG

CLCN1 NM 000083 102. A 23 P59772 6.90

CTGGGCAGAGCTCGCCCCACAAAGAAGAAA

GTCCTCAAGATCTCACTGCCAGTCATTGGG

ATP2A1 NM 173201 103. A 23 P72462 5.17

CTCGACGAAATCCTCAAGTTCGTTGCTCGG

GCTGAACGAGATCTACTTTGGAAAAACAAA

CAPZB NM 004930 104. A 23 P126752 5.22

GGATATCGTCAATGGGCTGAGGTCTGTGCA

TGGAGTCCTCCATGCCTGAGGACAGAAAGA

CACNA1S NM 000069 105. A 23 P85765 5.80

GCTCCACACCAGGGTCTCTTCATGAGGAGA

GATCATTAAGACCACGGAGCAGCTCATCGA

CAMK2B NM 172082 106. A 23 P42882 4.79

GGCCGTCAACAACGGTGACTTTGAGGCCTA

COPZ2 NM 016429 GGTTCTTCAGTCTGCCAAGGAACAAATTAA 107. A 23 P101093 4.76 Gene Systematic

Sequence # Probe Name Avg Name name

ATGGTCGTTATTGAAATGAAGGCTGTGGAT

TGAAGGCATCCTTGCTGGTGAATTGAAACA

GFPT1 NM 002056 108. A 23 P44083 6.03

TGGCCCTCTGGCTTTGGTGGATAAATTGAT

TTGGACTCTTCCCAGGGAAATTCCATCTTC

IMPDH2 NM 000884 109. A 24 P166042 5.00

CAGATCAATATGATCAAGTACATCAAAGAC

TGTGTACCCCTTAGCAGGGTGTCTGGGGAC

ALPK3 NM 020778 110. A 23 P348728 5.95

TTACGCCTTTGGAATTGCTCTTCATTCAGA

CCCCATCCTCTTTAACTATAAAGCTAATTT

ANK2 NM 001148 Ill . A 23 P133068 5.50

GTGACCAAAGATGGCATCCTTCATACTGGA

AGCAAAGGGAAATCAAGAGGAGACCCCCAG

BIN1 NM 139346 112. A 23 P165333 5.63

GCAGAGGGGCGTTCTCCCAAAGATTAGGTC

GGCCAGAGAGGAAGTTTGTTCACCAGAGAC RAP NM 198060 113. A 23 P402765 4.58

AGGCTTCAGATGGCTTTGATTTCGGCAAGC

GTGCTTCCCAGCCTCACAATGTGGGAATTT

OPA1 NM 130837 114. A 23 P211797 6.54

GACATAGGATGAGAGTCAGAGTATAGGTTT

GCTATGTTCAGAAAGATGCTTGGGTCCGAG

PHKAl NM 002637 115. A 32 P186121 5.26

ATAATGTGTACAGCATCTTGGCTGTGTGGG

ATATAGCACTGAATAAATGATGCAAGTTGT

UBE2D3 NM 181886 116. A 24 P363005 6.05

CAATGGATGAGTGATCAACTAATAGCTCTG

AAACATACACTTAGCTATGTTTTGCAACTC

PDLIM3 NM 014476 117. A 23 P110403 4.51

TTTTTGGGGCTAGCAATAATGATATTTAAA

TTTTTTGCCTGTGTGAATTCTACTTTTTAG

LDB3 NM 007078 118. A 32 P98227 5.11

CAAAAATAAAGCCCCCCAAAGGATGTGCAA

CTGACAACCCTGATCATCATGGACGTACAG

TTN NM 133378 119. A 23 P85269 4.66

AAACAAGATGGTGGACTTTATACCCTGAGT

ATCATGGACCTTCTGGTGCAGTCAGTGACC

FHOD1 NM 013241 120. A 23 P37778 5.46

AAGAGCAGTCCTCGTGCCTTAGCTGCTAGG

CAACACAGATACCAGTGTCCTCAGATGTCT

TBC1D15 NM 022771 121. A 23 P139558 5.10

GCAGATTAACACCTGCATGATCACTGTTCT

GAGTCTTATGTCTGGAAGATGTACCAAGAG

RYR1 NM 000540 122. A 23 P78867 4.76

AGATGTTGGGATTTCTTCCCAGCTGGTGAT

TCGGACGGTGCTTTTGGTGGATGCGTCTAG

DTNA NM 001392 123. A 23 P208158 4.82

ATGGATAACATGACTTCTTCTACCCTAAAA

AACAGGAGGAAGGCTGAGGATGAGGCCCGG

TNNT2 NM 000364 124. A 23 P34700 5.28

AAGAAGAAGGCTTTGTCCAACATGATGCAT

NM 001013 AGTCTTTACATCGCACTTTCAGTTCCTCCA

FXR1 125. A 23 P132784 5.47

439 TTTGGAATTCATAAAGGGGAGGGATCCTGA

ACACTCGGAGCTTGTGCTTTGTCTCCACGC

CKM NM 001824 126. A 23 P50250 4.75

AAAGCGATAAATAAAAGCATTGGTGGCCTT

CCGCAGTCACTTTCTTTGTAACAACTTCCG

ACTA1 NM 001100 127. A 23 P1102 4.98

TTGCTGCCATCGTAAACTGACACAGTGTTT

CTAAGACTCGAGACTTCACCTCCAGCAGGA

MYH3 NM 002470 128. A 23 P26865 4.86

TGGTGGTCCACGAGAGTGAAGAGTGAGCCA

Serum samples from healthy control subjects (Ctrl; N=7) were filtered through an 0.8 μπι filter to remove cells, ultracentrifuged at 100,000 x g for 90 minutes to collect extracellular RNA in EVs and particles, and the EV/ribonucleoprotein (RNP) pellet was lysed. RNA was extracted using the Qiagen miRNeasy kit and examined by mRNA microarray analysis. The data are represented as quartile normalized with background subtraction and values indicate expression levels of each gene (Noerholm et al., BMC Cancer 12, 22 (2012)). More than two-dozen transcripts of the mis-spliced mRNAs found in tibialis anterior (TA) muscle tissue of DM1 patients and reported as biomarkers of muscle weakness (Nakamori et al., Ann Neurol 74, 862-872 (2013)) are detected in the serum EV/RNP mRNA fraction and are candidate serum biomarkers. Note the similar levels of expression of muscle transcripts between healthy controls. The presence of muscle transcripts in EVs in this study is consistent with previous reports of EVs released by muscle cells (Romancino et al., FEBS Lett 587, 1379-1384 (2013)) and measurement of muscle-enriched miRs in human serum (Aoi et al., Front Physiol 4, 80 (2013)). (*: transcripts we report here that show differential urine exRNA splicing in DM1 vs MDC and UA controls; #, SEQ ID NO:). Table 6. Screening of human urine exosomes/microvesicles for

A previous study collected 3300 ml of urine from a human male, centrifuged the entire volume at low speed to pellet cells, passed the supernatant through a 0.8 μπι filter to remove remaining debris, ultracentrifuged the filtered supernatant, isolated RNA from the exosome/microvesicle pellet using a commercially available kit (Qiagen), and identified genes present using massively parallel RNA sequencing (Miranda et al., PLoS One 9, e96094 (2014)). Shown are raw and normalized counts of genes from that study that also were reported as biomarkers of DM1 in muscle biopsies 5, as well as other genes associated with muscular dystrophies. (*: transcripts we report here that show differential urine exRNA splicing in DM1 vs MDC and UA controls). Example 2. Quantitative gene expression in hiot ^' lu ids from DM1 and UA subjects

Based on qPCR cycle threshold (Ct) values of reference genes GTF2B and GAPDH, urine ex-mRNA content tended to be higher in DM1 as compared to UA subjects (Figure 1 A and Figure 7G). The lower expression of DMPK, the gene causing DM1, in urine from DM1 vs. UA subjects may be due to retention of mutant transcripts in the nucleus preventing their release into the cytoplasm and incorporation into EVs ²² . In serum, these transcripts were expressed at similar levels in both DM1 and UA subjects, and were present at lower levels than in urine (Figure IB and Figure 7H).

Example 3. Alternative mR A splice variants in exRNA and urinary tract tissues

Next we examined splice products in exRNA, focusing on transcripts previously reported as biomarkers of DM1 disease severity in muscle biopsies ⁴ . In urine exRNA, we identified 10 candidate DMl-specific splice products of 32 examined (Figure 2), 9 of which are MBNLl protein-dependent (Figures 8A-B) ⁴ · ²³ . Principle component analysis of these 10 splicing outcomes confirmed separation of DM1 from MDC and UA individuals (Figure 3A). We randomly assigned 76% of urine specimens from DM1 and UA subjects regardless of genotype to a training cohort and generated a predictive model that was 100% accurate in identifying the outcome of the remaining 24% of individuals in the validation cohort (Figure 3B and C). Splice products appeared similar in consecutive samples collected several months apart from the same individual (Figure 4), suggesting reliability of the assay. Interestingly, splicing of several other candidate DM mis-regulated splice events ⁴ was similar in urine from DM1 and UA individuals (Figures 9A-9C). Surprisingly, examination of these splice products in serum exRNA failed to show a significant difference between DM1 and UA controls for any of the transcripts examined (Figure 5A and B).

In previous studies, RNAs in urine have been used as biomarkers of prostate cancer, bladder cancer, and kidney transplant rejection, suggesting that cells lining the urinary tract are the primary contributors to the urine ex-RNA pool ^{12, 24, 25} . Our finding of slightly different exon inclusion/exclusion percentages of some transcripts in urine as compared to muscle tissue suggests the exRNA found in urine may represent a pool from multiple different cell types along this urinary route. To determine whether the urinary tract is the primary contributor of the ex-mRNA alternative splice variants in urine, we examined splice isoforms in human and mouse kidney and mouse bladder (Figures 10 and 12). Splicing of all transcripts examined showed different exon inclusion/exclusion percentages in normal human kidney compared to skeletal muscle (Figures 10 and 12A). In wild-type mouse bladder and kidney, splicing showed no consistent pattern relative to muscle tissue from wild-type or the HSA ^LR mouse model of DM1 (Figure 12B) ²⁶ . qPCR analysis also revealed that kidney expresses DMPK, although at a lower level than muscle (Figure 12C).

Example 4. Personalized DMD deletion transcripts in human urine

ASOs also are being evaluated therapeutically for another form of muscle disease, Duchenne muscular dystrophy (DMD), to modify dystrophin pre-mRNA splicing directly by inducing skipping of a target exon to restore the open reading frame and produce a truncated, partially functional protein ²¹ · ²⁸ . Detection of therapeutic drug effects in DMD patients involves multiple muscle biopsies to examine splicing outcomes and dystrophin protein production. To test whether biofluid exRNA contains DMD deletion transcripts, we examined urine from several subjects with DMD and found patient-specific DMD deletion transcripts (Figures 6A and B), suggesting this biofluid exRNA is a viable approach to monitor therapeutic exon-skipping ASO drug effects in DMD patients as personalized genetic markers ^{27, 28} .

Example 5. Identification of a novel DMD cryptic splice site

We also examined exRNA from a BMD patient with a normal DMD coding sequence, but a point mutation in intron 67 (c9807+ 6 T>G substitution). The normal coding sequence presumably produces a full-length dystrophin protein, suggesting the mutation in this patient causes dystrophinopathy by an overall reduction of dystrophin protein expression. RT-PCR analysis identified a splice product corresponding to the normal DMD exon 67 - 68 sequence in urine and serum from this patient and a UA subject, identical to muscle tissue (Figure 6C). In addition, a second larger product unique to the BMD samples was evident. DNA sequencing confirmed the larger band was a heteroduplex containing the normal product identical to that in the lower band, as well as one with inclusion of the 1 ^st five nucleotides of intron 67, indicating a cryptic splice site (Figure 6D) created by the mutation. The result is a frame shift and premature termination codon in exon 68, reducing functional dystrophin protein expression (Figure 6E). Thus, urine exRNA also can be used to identify this molecular disease mechanism. The expression in the kidney of DMD transcripts spanning the deletions and point mutation (Figure 12D) is consistent with the urinary tract as the primary source of exRNA in urine.

References

1. Scotti, M.M. & Swanson, M.S. RNA mis-splicing in disease. Nat Rev Genet 17, 19- 32 (2016).

2. Kanadia, R.N. et al. A muscleblind knockout model for myotonic dystrophy. Science

302, 1978-1980 (2003). Lin, X. et al. Failure of MBNLl -dependent post-natal splicing transitions in myotonic dystrophy. Hum Mol Genet 15, 2087-2097 (2006).

Nakamori, M. et al. Splicing biomarkers of disease severity in myotonic dystrophy. Ann Neurol 74, 862-872 (2013).

Wheeler, T.M. et al. Reversal of RNA dominance by displacement of protein sequestered on triplet repeat RNA. Science 325, 336-339 (2009).

Wheeler, T.M. et al. Targeting nuclear RNA for in vivo correction of myotonic dystrophy. Nature 488, 111-115 (2012).

Clinicaltrials.gov https://clinicaltrials.gov/ct2/show/NCT02312011. (2016).

Pandey, S.K. et al. Identification and characterization of modified antisense oligonucleotides targeting DMPK in mice and nonhuman primates for the treatment of myotonic dystrophy type 1. J Pharmacol Exp Ther 355, 329-340 (2015).

Tkach, M. & Thery, C. Communication by Extracellular Vesicles: Where We Are and Where We Need to Go. Cell 164, 1226-1232 (2016).

Skog, J. et al. Glioblastoma microvesicles transport RNA and proteins that promote tumour growth and provide diagnostic biomarkers. Nat Cell Biol 10, 1470-1476 (2008).

Chen, W.W. et al. BEAMing and Droplet Digital PCR Analysis of Mutant IDH1 mRNA in Glioma Patient Serum and Cerebrospinal Fluid Extracellular Vesicles. Mol Ther Nucleic Acids 2, el09 (2013).

Nilsson, J. et al. Prostate cancer-derived urine exosomes: a novel approach to biomarkers for prostate cancer. Br J Cancer 100, 1603-1607 (2009).

San Lucas, F. A. et al. Minimally invasive genomic and transcriptomic profiling of visceral cancers by next-generation sequencing of circulating exosomes. Ann Oncol 27, 635-641 (2016).

Khan, S. et al. Early diagnostic value of survivin and its alternative splice variants in breast cancer. BMC Cancer 14, 176 (2014).

Neeb, A. et al. Splice variant transcripts of the anterior gradient 2 gene as a marker of prostate cancer. Oncotarget 5, 8681-8689 (2014).

Romancino, D.P. et al. Identification and characterization of the nano-sized vesicles released by muscle cells. FEBS Lett 587, 1379-1384 (2013).

Forterre, A. et al. Myotube-derived exosomal miRNAs downregulate Sirtuinl in myoblasts during muscle cell differentiation. Cell Cycle 13, 78-89 (2014).

Hathout, Y. et al. Clinical utility of serum biomarkers in Duchenne muscular dystrophy. Clin Proteomics 13, 9 (2016). Moeller, M.J. & Tenten, V. Renal albumin filtration: alternative models to the standard physical barriers. Nat Rev Nephrol 9, 266-277 (2013).

Noerholm, M. et al. RNA expression patterns in serum microvesicles from patients with glioblastoma multiforme and controls. BMC Cancer 12, 22 (2012).

Miranda, K.C. et al. Massively parallel sequencing of human urinary

exosome/microvesicle RNA reveals a predominance of non-coding RNA. PLoS One 9, e96094 (2014).

Davis, B.M., McCurrach, M.E., Taneja, K.L., Singer, R.H. & Housman, D.E.

Expansion of a CUG trinucleotide repeat in the 3' untranslated region of myotonic dystrophy protein kinase transcripts results in nuclear retention of transcripts. Proc Natl Acad Sci USA 94, 7388-7393 (1997).

Wagner, S.D. et al. Dose-Dependent Regulation of Alternative Splicing by MBNL Proteins Reveals Biomarkers for Myotonic Dystrophy. PLoS Genet 12, el006316

(2016) .

Motamedinia, P. et al. Urine Exosomes for Non-Invasive Assessment of Gene Expression and Mutations of Prostate Cancer. PLoS One 11, e0154507 (2016).

Urquidi, V. et al. Urinary mRNA biomarker panel for the detection of urothelial carcinoma. Oncotarget 7, 38731-38740 (2016).

Mankodi, A. et al. Myotonic dystrophy in transgenic mice expressing an expanded CUG repeat. Science 289, 1769-1773 (2000).

Mendell, J.R. et al. Eteplirsen for the treatment of Duchenne muscular dystrophy. Ann Neurol 74, 637-647 (2013).

Kinali, M. et al. Local restoration of dystrophin expression with the morpholino oligomer AVI-4658 in Duchenne muscular dystrophy: a single-blind, placebo- controlled, dose-escalation, proof-of-concept study. Lancet Neurol 8, 918-928 (2009). Erdbragger, U. & Le, T.H. Extracellular Vesicles in Renal Diseases: More than Novel Biomarkers? J Am Soc Nephrol 27, 12-26 (2016).

Morcos, P.A., Li, Y. & Jiang, S. Vivo-Morpholinos: a non-peptide transporter delivers Morpholinos into a wide array of mouse tissues. Biotechniques 45, 613-614, 616, 618 passim (2008).

Aartsma-Rus, A. & Krieg, A.M. FDA Approves Eteplirsen for Duchenne Muscular Dystrophy: The Next Chapter in the Eteplirsen Saga. Nucleic Acid Ther 27, 1-3

(2017) .

Eriksson, M. et al. Recurrent de novo point mutations in lamin A cause Hutchinson- Gilford progeria syndrome. Nature 423, 293-298 (2003). 33. Lee, J.M. et al. Modulation of LMNA splicing as a strategy to treat prelamin A diseases. J Clin Invest 126, 1592-1602 (2016).

34. Hua, Y. et al. Peripheral SMN restoration is essential for long-term rescue of a severe spinal muscular atrophy mouse model. Nature 478, 123-126 (2011).

35. Honda, D. et al. The ALS/FTLD -related RNA-binding proteins TDP-43 and FUS have common downstream RNA targets in cortical neurons. FEBS Open Bio 4, 1-10 (2013).

36. Genschel, J. & Schmidt, H.H. Mutations in the LMNA gene encoding lamin A/C.

Hum Mutat 16, 451-459 (2000).

37. Imbeaud, S. et al. Towards standardization of RNA quality assessment using user- independent classifiers of microcapillary electrophoresis traces. Nucleic Acids Res 33, e56 (2005).

38. Savkur, R.S., Philips, A.V. & Cooper, T.A. Aberrant regulation of insulin receptor alternative splicing is associated with insulin resistance in myotonic dystrophy. Nat Genet 29, 40-47 (2001).

39. Koressaar, T. & Remm, M. Enhancements and modifications of primer design

program P ner3. Bioinformatics 23, 1289-1291 (2007).

40. Untergasser, A. et al. Primer3~new capabilities and interfaces. Nucleic Acids Res 40, el l5 (2012).

41. Mevik, B.-H. & Wehrens, R. The pis Package: Principal Component and Partial Least Squares Regression in R. J Statistical Software 18, 1-24 (2007).

OTHER EMBODIMENTS

It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.