WO2014077693A1

WO2014077693A1 - Means and methods for reducing an effect of aging in a mammalian cell

Info

Publication number: WO2014077693A1
Application number: PCT/NL2013/050827
Authority: WO
Inventors: Vered RAZ; Silvère Maria VAN DER MAAREL
Original assignee: Leids Universitair Medisch Centrum LUMC
Current assignee: Leids Universitair Medisch Centrum LUMC
Priority date: 2012-11-16
Filing date: 2013-11-18
Publication date: 2014-05-22
Anticipated expiration: 2015-05-16

Abstract

Aging is associated with genome-wide changes in expression profiles. Poly(A) Binding Protein Nuclear 1 (PABPN1) function is associated with aging and cell functionality. Mutation in PABPN1 is the genetic cause for the late-onset oculopharyngeal muscular dystrophy (OPMD). We identified the E3-ligase, ARIH2, as a regulator of PABPN1 accumulation. The wide-spread effects of PABPN1 can be manipulated by altering the level of ARIH2 m RNA and/or protein in a cell.

Description

Title: Means and methods for reducing an effect of aging in a mammalian cell.

The invention relates to the field of aging. The invention in particular relates to means and methods for preventing or delaying the reduced vitality or increased degeneration of cells during aging.

Proper protein turnover and protein homeostasis are crucial for maintaining cellular integrity. Imbalance of protein homeostasis leading to an increase in protein aggregation is among the cellular signatures of aging (Ben-Zvi et al. 2009)(Lindner & Demarez 2009; David et al. 2010). Protein aggregation characterizes a large spectrum of late-onset neuromuscular degenerative disorders, such as Alzheimer's disease, Huntington's and Parkinson's disease. In these disorders misfolded proteins accumulate in insoluble inclusions or nuclear inclusions (reviewed in (Bingol & Sheng 2011)). It is unclear whether nuclear inclusions or oligomers that precede inclusion formation are pathogenic and harmful to the cell (Shao & Diamond 2007). An alternative explanation may be that irreversible protein entrapment in insoluble bodies depletes the cell from soluble and functional protein (Junghans 2009; de Mezer et al. 2011; Raz et al. 2011b).

Adult neurodegenerative disorders can be caused by single amino acid repeat expansions. Most expansion mutations generate expanded polyglutamine tracts, but a subset is caused by polyalanine expansions (Brais 2009). A

polyalanine tract expansion in the poly(A)-b iding protein nuclear 1 (PABPN1) causes oculopharyngeal muscular dystrophy (OPMD) (Brais et al. 1998). In OPMD, disease symptoms appear only after midlife and initially are limited to a subset of skeletal muscles. As disease progresses, most skeletal muscles become affected. Accumulation of expanded (exp)PABPNl into insoluble intranuclear nuclear inclusions in skeletal muscles is the pathological hallmark of OPMD (Tome & Fardeau 1980; Calado et al. 2000). Naturally occurring inclusions of wild- type (WT) PABPN1 have also been reported (Berciano et al. 2004). In contrast to the nuclear inclusions in OPMD, inclusions of WT-PABPN1 are not disease-associated but are aging-associated (Hipkiss 2006; David et al. 2010).

In the present invention, WT and exp PABPNl are used to discriminate between pathogenic and non-pathogenic age-associated protein-aggregation. In vitro, both WT- and exp PABPNl form indistinguishable insoluble inclusions, but the aggregation process of WT- and exp PABPNl significantly differs (Raz et al. 2011a). Compared with the WT protein, a slower protein turnover of expPABPNl is caused by reduced poly-ubiquitination (Raz et al. 2011b). Expression of

expPABPNl is associated with down-regulation of proteasome encoding genes (Anvar et al. 2011a; Anvar et al. 2011b; Raz et al. 2011b). In OPMD the ubiquitin proteasome system (UPS) is down-regulated, and components of the UPS are entrapped in PABPNl nuclear inclusions (Corbeil-Girard et al. 2005; Tavanez et al. 2005; Anvar et al. 2011a). Being the substrate-specific component of the UPS, the expression of E3-ligases is also significantly deregulated in OPMD, and E3-ligases are also entrapped in PABPNl nuclear inclusions (Anvar et al. 2011a). Protein entrapment in PABPNl aggregates is gradual (Raz et al., 2011a).

Among the proteins that co-localize with PABPNl at an early step of the aggregation process are ubiquitin and poly(A) polymerase (Raz et al., 2011a), which bind to soluble PABPNl (Abu-Baker et al., 2003; Kerwitz et al., 2003).

In the present invention it was found that the E3-ligase TRIAD 1, ariadne homolog 2 (Drosophila), ARIH2 (named here ARIH2) (human

Erisernbi:ENSG00000177479) co-localizes with PABPNl at an early step of the aggregation process, and regulates PABPNl turnover with a preference for the WT protein. Down-regulation of ARIH2 induces a decrease in PABPNl expression. In turn, a decrease in PABPNl, for instance by mRNA down-regulation or by entrapment of soluble PABPNl in aggregates, causes a change in ARIH2 3'- untranslated region (UTR) length through a proximal instead of a distal

polyadenylation site usage. We suggest that ARIH2 is a regulator of PABPNl expression, and that PABPNl regulates ARIH2 3'-UTR length and expression levels in a feed forward loop. SUMMARY OF THE INVENTION

The invention provides a method for increasing the level of PABPN1 mRNA and/or protein in a cell, said method comprising increasing the level of ARIH2 protein in said cell.

The invention also provides a method for modifying poly-adenylation site usage in a cell comprising modifying the level of ARIH2 protein in said cell.

The invention further provides a method for inhibiting a molecular effect of aging in an adult cell, said method comprising increasing the level of ARIH2 protein in said cell.

Also provided is a method for modifying a molecular effect of aging in an adult cell said method comprising providing said cell with an

- antisense oligonucleotide that is complementary to and capable of hybridizing to a poly-adenylation site; or

- antisense oligonucleotide that is complementary to and capable of hybridizing to a miRNA target sequence,

- antisense oligonucleotide that is complementary to and capable of hybridizing to a regulatory sequence in the 3'UTR,

of a pre-mRNA encoded by the gene HILFA; EGFR; SUMO l;

PASMD 14; SLC1A4; E2F1; BMI1; RBI; TP63; SLC2A4; ING1; MIB1; PTEN ;

SUB1 ; IL10 ; VEGFA ; MORF4L1; DNM1L; SOD2; RAC1; ITGB1; HIF1A; SRF;

MCL1; RAD17; GSK3B; RYR1; MEF2A; GRB2; ZMYND11; DNMT3A; CDK7;

DIABLO; JARID2; EIF4E; UBE3A; BECN1; KHDRBS3; NOLC1; LPL; NF2;

ROCK2; STAT5A; ITSNl; STAT5B; HUSl; HSPA9; LPL; ABI3; CAMK2D; HSPDl; ITPR1; CSNK2A1; PAK1; BIRC2; RSL1D1; GCLM; CDS1; HDAC3; YY1; RPA1;

RRM2B; CD55; HDAC4; KSR1; CD59; HMGB1; GCLC; PSMD14; CADM1; TFRC;

RBX-1; PDK1; AGFG1; PSMA2 ; PLCB4; LAMP1; RHOA ; RAP-1; UGCG; H2AFZ;

CANX; UBE2I; TPP2; REV1; EMB; COL5A1; RAB1A; IMPACT; VAMP2; TANK;

HSPE1; MYL1; PRKAR1A; WASL; HNRNPK; SGCB; GABPA; MLF1; SMEK2; M6PR; PERP; PPPICC; HSF2; TDP-43; C90RF72; or MAP3K7. The gene symbols as used herein are the official gene symbols for the genes at the time of filing of the application. The genes exhibit significant changes in expression upon aging and upon disease progression in the OPMD model. The changes are correlated to alternate poly-adenylation site (PAS) usage. With the method it is possible reduce at least the effect of aging on the altered gene expression of the targeted gene.

The invention further provides a method for modifying PAS usage in pre-mRNAs expressed by a cell, comprising providing said cell with an anti-sense oligonucleotide that is complementary to and capable of hybridizing to a PAS of a pre-mRNA encoded by the gene ARIH2; HILFA; EGFR; SUMOl; PASMD14;

SLC1A4; E2F1; BMI1; RBI; TP63; SLC2A4; ING1; MIB1; PTEN ; SUB1 ; IL10 ;

VEGFA ; MORF4L1; DNM1L; SOD2; RAC1; ITGB1; HIF1A; SRF; MCL1; RAD17;

GSK3B; RYR1; MEF2A; GRB2; ZMYND11; DNMT3A; CDK7; DIABLO; JARID2; EIF4E; UBE3A; BECNl; KHDRBS3; NOLCl; LPL; NF2; ROCK2; STAT5A; ITSNl;

STAT5B; HUSl; HSPA9; LPL; ABI3; CAMK2D; HSPDl; ITPRl; CSNK2A1; PAKl;

BIRC2; RSL1D1; GCLM; CDS1; HDAC3; YY1; RPA1; RRM2B; CD55; HDAC4;

KSR1; CD59; HMGB1; GCLC; PSMD 14; CADM1; TFRC; RBX-1; PDK1; AGFG1;

PSMA2 ; PLCB4; LAMP1; RHOA ; RAP- 1; UGCG; H2AFZ; CANX; UBE2I; TPP2; REV1; EMB; COL5A1; RAB1A; IMPACT; VAMP2; TANK; HSPE 1; MYLl;

PRKAR1A; WASL; HNRNPK; SGCB; GABPA; MLF1; SMEK2; M6PR; PERP;

PPP1CC; HSF2; TDP-43; C90RF72; or MAP3K7,

said method characterized in that said cell is an adult cell, preferably a muscle cell, preferably a skeletal muscle cell, a senescent cell, neuronal cell, a satellite cell, an adult stem cell, preferably a mesenchymal stem cell.

The invention also provides an isolated oligonucleotide having 12-40 bases, wherein said oligonucleotide comprises a continuous stretch of at least 7 bases that is complementary to and capable of hybridizing to a poly-adenylation site or comprises a continuous stretch of at least 7 bases that is complementary to and capable of hybridizing to a miRNA target sequence, or comprises a continuous stretch of at least 7 bases that is complementary to and capable of hybridizing to a regulatory 3' UTR target sequence, of an ARIH2 pre-mRNA or of a (pre-)mRNA encoded by the gene HILFA; EGFR; SUMOl; PASMD14; SLC1A4; E2F1; BMI1;

RBI; TP63; SLC2A4; ING1; MIB1; PTEN ; SUB1 ; IL10 ; VEGFA ; MORF4L1; DNM1L; SOD2; RAC1; ITGB1; HIF1A; SRF; MCL1; RAD17; GSK3B; RYR1;

MEF2A; GRB2; ZMYND11; DNMT3A; CDK7; DIABLO; JARID2; EIF4E; UBE3A;

BECNl; KHDRBS3; NOLCl; LPL; NF2; ROCK2; STAT5A; ITSNl; STAT5B;

HUS l; HSPA9; LPL; ABI3; CAMK2D; HSPDl; ITPRl; CSNK2A1; PAKl; BIRC2; RSL1D1; GCLM; CDS 1; HDAC3; YY1; RPA1; RRM2B; CD55; HDAC4; KSR1; CD59; HMGBl; GCLC; PSMD14; CADMl; TFRC; RBX- 1; PDKl; AGFGl; PSMA2 ; PLCB4; LAMP1; RHOA ; RAP- 1; UGCG; H2AFZ; CANX; UBE2I; TPP2; REV1; EMB; COL5A1; RAB1A; IMPACT; VAMP2; TANK; HSPE1; MYL1; PRKAR1A; WASL; HNRNPK; SGCB; GABPA; MLFl; SMEK2; M6PR; PERP; PPPICC; HSF2; TDP-43; C90RF72; or MAP3K7.

Also provided is an isolated oligonucleotide having 12-40 bases, comprising a continuous stretch of at least 7 bases of sequence:

ARIH2* 5' - GT ATAATT GTAC AAC CTTT GAAAG- 3'

PSMD14 5'-GAGCGCCACTGACAGCTCTCTTA-3'

GRB2 5'-GACAAGAAACCAAGTGGGC-3' STAT5B 5'-GAAGTGTTAATACTAGTTGT-3'.

The invention further provides a compound for increasing the level of ARIH2 protein in a cell for use in the treatment of an individual suffering from aging. In a preferred embodiment said individual is suffering from a

neurodegenerative disease. In one embodiment said disease is Sarcopenia,

Alzheimer or Parkinson. In a preferred embodiment said disease is Sarcopenia. Said compound is preferably selected from an antisense oligonucleotide is that is complementary to and capable of hybridizing to ARIH2 (pre-)mRNA produced by said cell; a nucleic acid that encodes ARIH2 protein; an MDM2 inhibitor that inhibits the level or activity of MDM2; a compound that increases the level or activity of or of HoxAlO in said cell; and/or all-trans retinoic acid. The invention further provides a method for the treatment of an individual suffering from an age related degenerative disease comprising administering to the individual in need thereof a compound for increasing the level of ARIH2 protein in a cell. In a preferred embodiment said individual is suffering from a neurodegenerative disease. In one embodiment said disease is Sarcopenia, Alzheimer or Parkinson. In a preferred embodiment said disease is Sarcopenia. Said compound is preferably selected from an antisense oligonucleotide is that is complementary to and capable of hybridizing to ARIH2 (pre-)mRNA produced by said cell; a nucleic acid that encodes ARIH2 protein; an MDM2 inhibitor that inhibits the level or activity of MDM2; a compound that increases the level or activity of or of HoxAlO in said cell; and/or all-trans retinoic acid.

DETAILED DESCRIPTION OF THE INVENTION

PABPNl is a regulator of mRNA processing: it regulates

polyadenylation tail length (Kuhn et al. 2009; Apponi et al. 2010) and PAS usage (de Klerk et al. 2012; Jenal et al. 2012). Changes in the PABPNl protein level have genome-wide consequences for mRNA stability.. We observed that ARIH2 and PABPNl expression levels significantly decline during aging and in the late-onset muscle disorder OPMD. A decline in expression is initiated around midlife both in healthy humans as in OPMD patients. The decline in expression correlates with disease onset. PABPNl down-regulation causes changes in PAS usage that is similar to over-expression of expPABPNl (de Klerk et al. 2012; Jenal et al. 2012). Expression of expPABPNl causes reduced levels of soluble PABPNl (Raz et al. 2011b; E. de Klerk 2012). A decline in soluble PABPNl leads to alternative PAS usage, including the ARIH2 PAS usage. We found that PABPNl mRNA levels significantly reduce in vivo during muscle aging, aged human mesenchymal stem cells, OPMD muscles and in vitro senescent cells.

Accumulation of soluble PABPNl is regulated by the UPS (Abu-Baker et al. 2003; Raz et al. 2011b), and in the present invention we identified ARIH2 as a regulator of PABPNl protein turnover. ARIH2 contains a ring between ring fingers (RBR) domain, and is part of the largest family of E3-ligases. RBR E3- ligases have attracted interest because of their involvement in late onset protein aggregation disorders such as Parkinson' disease, Lewy body dementia, and Alzheimer's disease (Eisenhaber et al. 2007). Here we found that a decline in ARIH2 expression during aging in, among others, skeletal muscles that

demonstrate aging-associated muscle weakness. Corresponding with severe muscle weakness in OPMD, reduced ARIH2 and PABPNl expression levels were greater in OPMD. The expression trend of ARIH2 and PABPNl in aging overlaps with an age-dependent decline in muscle strength (Beenakker et al. 2010). In vivo, the correlation between expression levels of both genes is highly significant.

We identified that in muscle cells manipulation of expression levels of

ARIH2 or PABPNl affects the levels of the other gene. The means and methods of the invention can be applied to a variety of different cells and cell types. A preferred example of such a cell or cell type is selected from the group of a muscle cell, preferably a skeletal muscle cell, a senescent cell, a neuronal cell, a satellite cell, an adult stem cell, preferably a mesenchymal stem cell. In a particularly preferred embodiment the cell is a skeletal muscle cell.

The cell is preferably a cell of an animal, preferably of a mammal or a bird. In a particularly preferred embodiment the cell is a primate cell, preferably a human cell. In primates and particularly in humans the PABPNl level declines with age and is associated with muscle weakness.

An adult human cell of the present invention is preferably a cell derived from or of an individual that is at least 40 years old, preferably at least 50 years old. In a particularly preferred embodiment the cell is derived from or of a 60 old individual. An adult cell of a different organism is preferably derived from an individual of a comparable age is indicated herein above for the human.

The effect that the concerted decline of the levels of ARIH2 and PABPNl has in a cell, for instance, as a result of the aging of the cell, can be decreased and/or reversed by increasing the level of ARIH2 mRNA and/or protein in the cell. Increasing the level of ARIH2 mRNA and/or protein in a cell has the effect of elevating the level of PABPNl mRNA and/or protein in the cell. With the terms elevating or increasing the level of PABPNl is meant a higher level when compared to the same circumstances in the absence of ARIH2 manipulation. The term also encompasses a stabilization or slower decrease of the level of PABPNl over time, when in the absence of ARIH2 mRNA and/or protein increase in otherwise similar circumstances the level of PABPNl decreases or decreases more, respectively. The level of PABPNl preferably refers to the level of soluble PABPNl. Maintaining of PABPNl in turn rejuvenates the cell. Without being bound by theory, it is believed that due to a genome-wide change in mRNA stability, the expression of many different genes, including HUB molecules shifts the cell from functional/healthy to less functional/unhealthy. Re-adjusting mRNA stability of one or more of those HUB molecules pushes a cell into a normally functional state. PABPNl is a HUB in the spliceosome and ARIH2 regulates PABPNl. Increasing the level of ARIH2 mRNA and/or protein in a cell can be used in a treatment of an individual suffering from aging. In a preferred

embodiment said individual is suffering from a neurodegenerative disease. In one embodiment said disease is Sarcopenia, Alzheimer or Parkinson. In a preferred embodiment said disease is Sarcopenia.

Alternative poly-adenylation site usage is regulated by PABPNl, and as levels of PABPNl are regulated by ARIH2, poly-adenylation site usage in a cell can be modified by modifying the level of ARIH2 protein in a said cell. Lower PABPNl levels results in proximal poly-adenylation site usage. Increasing the level of ARIH2 protein in a cell with lower levels of PABPNl results in an increase in the level of PABPNl in the cell. This, in turn, reverses the utility of a more distal PAS in a gene is as indicated herein. Particularly in the gene HILFA; EGFR; SUMOl; PASMD 14; SLC1A4; E2F1; BMI1; RBI; TP63; SLC2A4; ING1; MIB1; PTEN ;

SUB1 ; IL10 ; VEGFA ; MORF4L1; DNM1L; SOD2; RAC1; ITGB1; HIF1A; SRF; MCL1; RAD17; GSK3B; RYR1; MEF2A; GRB2; ZMYND11; DNMT3A; CDK7;

DIABLO; JARID2; EIF4E; UBE3A; BECN1; KHDRBS3; NOLC1; LPL; NF2;

ROCK2; STAT5A; ITSNl; STAT5B; HUSl; HSPA9; LPL; ABI3; CAMK2D; HSPDl; ITPR1; CSNK2A1; PAK1; BIRC2; RSL1D1; GCLM; CDS1; HDAC3; YY1; RPA1; RRM2B; CD55; HDAC4; KSR1; CD59; HMGB1; GCLC; PSMD14; CADM1; TFRC; RBX-1; PDK1; AGFG1; PSMA2 ; PLCB4; LAMP1; RHOA ; RAP-1; UGCG; H2AFZ; CANX; UBE2I; TPP2; REV1; EMB; COL5A1; RAB1A; IMPACT; VAMP2; TANK; HSPE1; MYLl; PRKAR1A; WASL; HNRNPK; SGCB; GABPA; MLF1; SMEK2; M6PR; PERP; PPP1CC; HSF2; TDP-43; C90RF72; or MAP3K7 or a combination thereof. The level of ARIH2 protein in a cell is preferably increased by providing the cell with

- an antisense oligonucleotide is that is complementary to and capable of hybridizing to ARIH2 (pre-)mRNA produced by said cell;

- providing the cell with a nucleic acid that encodes ARIH2 protein;

- by decreasing the level and/or activity of MDM2 in said cell; - by increasing the level and/or activity of HoxAlO in said cell; and/or

- by contacting the cell with an effective amount of all-trans retinoic acid, or a combination thereof. The level of ARIH2 protein in a cell is preferably increased by providing the cell with an antisense oligonucleotide is that is complementary to and capable of hybridizing to ARIH2 (pre-)mRNA produced by said cell. The antisense oligonucleotide is preferably complementary to and capable of hybridizing to the proximal poly-adenylation signal sequence located at position 314-337 in the ARIH2 sequence of figure 8.

In one embodiment the level of ARIH2 mRNA and/or protein is increased by providing the cell with an antisense oligonucleotide (AON) that is complementary to and capable of hybridizing to an ARIH2 pre-mRNA, preferably to ARIH2 3' UTR produced by said cell. In a preferred embodiment the antisense oligonucleotide is complementary to and capable of hybridizing to a PAS of the ARIH2 pre-mRNA. Preferably said PAS is the proximal PAS. Preferably said antisense oligonucleotide is complementary to and capable of hybridizing to the proximal poly-adenylation signal sequence located at position 314-337 in the ARIH2 sequence of figure 8 (highlighted in BOLD).

Alternatively the oligonucleotide is complementary to and capable of hybridizing to a miRNA binding site, located in the ARIH2 (pre-)mRNA. The binding site is preferably located in the 3' UTR of the ARIH2 (pre-)mRNA. In a preferred embodiment the miRNA binding site is a miR19 binding site of the ARIH2 (pre-)mRNA. Preferably said miR19 binding site is located downstream of the proximal PAS of the ARIH2 pre-mRNA. In a particularly preferred

embodiment the AON directed towards the miR19 binding site in the ARIH2 pre- mRNA is directed towards the sequence located at position 1055- 1074 in the ARIH2 sequence of figure 8 (underlined). The oligonucleotide can also be directed towards a regulatory sequence in the 3' UTR of ARIH2.

An antisense oligonucleotide directed towards the proximal PAS in the

ARIH2 pre-mRNA, preferably comprises 12-40 bases. The AON preferably comprises a continuous stretch of at least 7 bases of sequence 5'-GTA TAA TTG TAC AAC CTT TGA AAG-'3. Preferably, the AON comprises a continuous stretch of at least 15 bases of sequence 5'-GTA TAA TTG TAC AAC CTT TGA AAG-'3. In a particularly preferred embodiment the AON comprises a continuous stretch of at least 20 bases of sequence 5'-GTA TAA TTG TAC AAC CTT TGA AAG-'3. In a particularly preferred embodiment the AON comprises the sequence:

5' GTA TAA TTG TAC AAC CTT TGA A; 5' GTA TAA TTG TAC AAC CTT TGA;

5' GTA TAA TTG TAC AAC CTT TG; or

5' GTA TAA TTG TAC AAC CTT TGA AAG.

In a preferred embodiment the AON comprises the sequence 5' GTA TAA TTG TAC AAC CTT TGA AAG.

In another embodiment the AON comprises the sequence

5'-TA TAA TTG TAC AAC CTT TGA AAG T-'3;

5'-A TAA TTG TAC AAC CTT TGA AAG TT-'3 ;

5'-T GTA TAA TTG TAC AAC CTT TGA AA-'3 ; or

5'-TT GTA TAA TTG TAC AAC CTT TGA A-'3.

An antisense oligonucleotide directed towards the miR19 binding site downstream of the proximal PAS of the ARIH2 pre-mRNA, preferably comprises 12-40 bases. The AON preferably comprises a continuous stretch of at least 7 bases of sequence 5'-TAA CTT GTG CAA ACA CAG CC-3'.

Preferably, the AON comprises a continuous stretch of at least 15 bases of sequence 5'- TAA CTT GTG CAA ACA CAG CC -'3. In a particularly preferred embodiment the AON comprises a continuous stretch of at least 20 bases of sequence 5'- TAA CTT GTG CAA ACA CAG CC -'3. In a particularly preferred embodiment the AON comprises the sequence:

5' TAA CTT GTG CAA ACA CAG CC;

5' ATA ACT TGT GCA AAC ACA GC; or

5' AAT AAC TTG TGC AAA CAC AG.

In a preferred embodiment the AON comprises the sequence 5'TAA CTT GTG CAA ACA CAG CC-3'.

In another embodiment the AON comprises the sequence

5'-AA CTT GTG CAA ACA CAG CCC-3';

5'- A CTT GTG CAA ACA CAG CCC C-3';

5'- A TAA CTT GTG CAA ACA CAG C-3'; or

5'- AA TAA CTT GTG CAA ACA CAG -3'.

Another way of increasing the level of ARIH2 mRNA and/or protein in a cell is by providing the cell with a nucleic acid that code for a functional part, derivative and/or analogue thereof. The nucleic acid preferably comprises an ARIH2 coding region together with the appropriate in cis required expression signal sequences for transcription and/or translation of the coding region in the target cell. Such sequences can encompass for instance, a suitable promoter, a shine delgano sequence, poly-adenylation sequences and the like. The ARIH2 coding region is preferably a coding region encoding a mammalian ARIH2, preferably a primate ARIH2 and more preferably a human ARIH2.

It is preferred that the species of that ARIH2 protein is derived from is the same as the species the cell is derived from.

ARIH2 protein levels in a cell are also influenced by other factors. For instance the level of ARIH2 protein is increased when the level or activity of

MDM2 in the cell is decreased. Thus in a preferred embodiment of the invention the level of ARIH2 mRNA and/or protein in a cell is increased by decreasing the level and/or activity of MDM2 in said cell.

ARIH2 protein levels in a cell are also influenced by the level or activity of HoxAlO in the cell. The level of ARIH2 protein is increased when the level or activity of HoxAlO in the cell is increased. Thus in a preferred embodiment of the invention the level of ARIH2 mRNA and/or protein in a cell is be increased by increasing the level and/or activity of HoxAlO in said cell.

ARIH2 protein levels in a cell are also influenced by contacting said cell with resveratrol. The level of ARIH2 protein is increased by contacting the cell with an effective amount of resveratrol. Thus in a preferred embodiment of the invention the level of ARIH2 mRNA and/or protein in a cell is increased by contacting said cell with an effective amount of resveratrol.

ARIH2 protein levels in a cell are also influenced by contacting said cell with a SIRT drug. The level of ARIH2 protein is increased by contacting the cell with a SIRT drug. Thus in a preferred embodiment of the invention the level of ARIH2 mRNA and/or protein in a cell is increased by contacting said cell with a SIRT drug.

ARIH2 RNA and/or protein levels in a cell are also influenced by contacting said cell with all-trans retinoic acid (Pietschmann et al., The

International Journal of Biochemistry & Cell Biology 44 (2012) 132- 138; Marteijn et al., 2005). The level of ARIH2 protein is increased by contacting the cell with an effective amount of a\\-trans retinoic acid. Thus in a preferred embodiment of the invention the level of ARIH2 mRNA and/or protein in a cell is increased by contacting said cell with all- trans retinoic acid.

On the other hand, PABPNl expression in a cell can be down-regulated by decreasing the level of ARIH2 mRNA and/or protein in that cell. This feature is for instance useful in artificially aging cells. This property can be used, for instance, to the treatment of cancerous cells. The invention further provides a method for the treatment of an individual suffering from cancer said method comprising administering a PABPNl, and/or an ARIH2 inhibitor to the individual in need thereof. A preferred inhibitor is an antibody that binds to PABPNl or ARIH2, or a derivative or analogue of said antibody. The antibody is preferably an antibody that neutralizes the activity of PABPNl or ARIH2. In a preferred embodiment said antibody is an intracellular antibody; in a particularly preferred embodiment said intracellular antibody is an ScFv antibody. The antibody may be used as such or be provided to the cell by means of a nucleic acid delivery vehicle comprising one or more nucleic acids encoding said antibody, or derivative or analogue thereof. For intracellular antibodies or derivatives and/or analogues thereof it is preferred that it is provided to the cell by means of said gene delivery vehicle. In another preferred embodiment said PABPNl or ARIH2 inhibitor is an RNA inhibitor. Presently there are many different RNA molecules that can inhibit translation of an mRNA and/or decrease the stability of the RNA. The RNA inhibitor is preferably an RNAi molecule specific for PABPNl or ARIH2 mRNA; shRNA molecule specific for PABPNl or ARIH2 mRNA; an AON that induces exon skipping of a PABPNl or ARIH2 exon, The exon to be skipped preferably introduces a frameshift resulting in premature termination of the protein. Thus the invention further provides a method for the treatment of an individual suffering from cancer comprising administering to the individual in need thereof an effective amount of a PABPNl inhibitor and/or an ARIH2 inhibitor.

We found that down-regulation of ARIH2 induces a decline in PABPNl expression. Recently, others and we showed that PABPNl regulates the PAS usage (E. de Klerk 2012; Jenal et al. 2012). Reduced levels of soluble PABPNl, either by down-regulation PABPNl or by aggregation of expPABPNl, enhance the usage of proximal PAS. In concurrence, here we found a decrease in the ratio between long and short ARIH2 3' UTR in A17.1 OPMD mice, in PABPNl down-regulated cells, and in ARIH2 mRNA fraction that binds to PABPNl. Thus the reduced expression of ARIH2 is caused by alternative PAS usage and this is regulated by PABPNl (Figure 7C). In turn, we show that ARIH2 regulates PABPNl protein

accumulation. Thus the expression of both proteins is maintained via interplay between both mRNA stability and protein turnover regulatory machineries. This feed-forward regulatory loop is noticeable during aging. An aging-associated decline in soluble and therefore functional PABPNl induces genome-wide changes in mRNA stability and turnover (Figure 7C). In OPMD, aggregation of expPABPNl leads to a higher decrease in soluble PABPNl, and therefore the levels of ARIH2 are lower compared with controls (Figure 7C).

As mentioned herein above the invention also provides a method for modifying a molecular effect of aging in an adult cell said method comprising providing said cell with an

- antisense oligonucleotide that is complementary to and capable of hybridizing to a poly-adenylation site;

- antisense oligonucleotide that is complementary to and capable of hybridizing to a miRNA target sequence; or

of a pre-mRNA encoded by the gene HILFA; EGFR; SUMOl; PASMD14;

SLC1A4; E2F1; BMI1; RBI; TP63; SLC2A4; ING1; MIB1; PTEN ; SUB1 ; IL10 ;

VEGFA ; MORF4L1; DNM1L; SOD2; RAC1; ITGB1; HIF1A; SRF; MCL1; RAD17; GSK3B; RYR1; MEF2A; GRB2; ZMYNDll; DNMT3A; CDK7; DIABLO; JARID2;

EIF4E; UBE3A; BECNl; KHDRBS3; NOLCl; LPL; NF2; ROCK2; STAT5A; ITSNl;

STAT5B; HUSl; HSPA9; LPL; ABI3; CAMK2D; HSPDl; ITPRl; CSNK2A1; PAKl;

BIRC2; RSL1D1; GCLM; CDS1; HDAC3; YY1; RPA1; RRM2B; CD55; HDAC4;

KSR1; CD59; HMGB1; GCLC; PSMD 14; CADM1; TFRC; RBX-1; PDK1; AGFG1; PSMA2 ; PLCB4; LAMP1; RHOA ; RAP- 1; UGCG; H2AFZ; CANX; UBE2I; TPP2;

REV1; EMB; COL5A1; RAB1A; IMPACT; VAMP2; TANK; HSPE 1; MYL1;

PRKAR1A; WASL; HNRNPK; SGCB; GABPA; MLF1; SMEK2; M6PR; PERP;

PPP1CC; HSF2; TDP-43; C90RF72; or MAP3K7. The invention further provides a method for modifying PAS usage in a (pre-)mRNA expressed by a cell, comprising providing said cell with an antisense oligonucleotide that is complementary to and capable of hybridizing to a poly- adenylation site of a pre-mRNA encoded by the gene ARIH2; HILFA; EGFR;

SUMOl; PASMD14; SLC1A4; E2F1; BMI1; RBI; TP63; SLC2A4; ING1; MIB1;

PTEN ; SUB1 ; IL10 ; VEGFA ; MORF4L1; DNM1L; SOD2; RAC1; ITGB1; HIF1A; SRF; MCLl; RAD 17; GSK3B; RYRl; MEF2A; GRB2; ZMYNDl l; DNMT3A; CDK7; DIABLO; JARID2; EIF4E; UBE3A; BECN1; KHDRBS3; NOLC1; LPL; NF2;

ROCK2; STAT5A; ITSNl; STAT5B; HUSl; HSPA9; LPL; ABI3; CAMK2D; HSPDl; ITPR1; CSNK2A1; PAK1; BIRC2; RSL1D1; GCLM; CDS1; HDAC3; YY1; RPA1; RRM2B; CD55; HDAC4; KSR1; CD59; HMGB1; GCLC; PSMD14; CADM1; TFRC; RBX-1; PDK1; AGFG1; PSMA2 ; PLCB4; LAMP1; RHOA ; RAP-1; UGCG; H2AFZ; CANX; UBE2I; TPP2; REV1; EMB; COL5A1; RAB1A; IMPACT; VAMP2; TANK; HSPE1; MYL1; PRKAR1A; WASL; HNRNPK; SGCB; GABPA; MLF1; SMEK2; M6PR; PERP; PPP1CC; HSF2; TDP-43; C90RF72; or MAP3K7 said method characterized in that said cell is an adult cell, preferably a muscle cell, preferably a skeletal muscle cell, a senescent cell, neuronal cell, a satellite cell, an adult stem cell, preferably a mesenchymal stem cell. The modification of the PAS usage of one or more of the mentioned gene affects aging of the cell.

The gene is preferably selected from ARIH2; PSMD14; STAT5B;

HILFA; PTEN; VEGFA; GSK3B; ITGB1; EGFR; YY1; SUMOl; PASMD14;

SLC1A4; RBI; HDAC3; RHOA; GRB2; E2F1; TDP-43; C90RF72; or RAC1. In a preferred embodiment the gene is selected from ARIH2; PSMD14; STAT5B; TDP- 43; C90RF72; or GRB2. In a particularly preferred embodiment the gene is ARIH2.

ARIH2 is also referred to as ARI2 or TRIAD 1.

External IDs OMIM: 605615 MGI: 1344361 HomoloGene: 48424

Orthologs

Species Human Mouse

Entrez 10425 23807

Ensembl ENSG00000177479 ENSMUSG00000064145

UniProt 095376 Q9Z1K6

RefSeq (mRNA) NM_006321.2 NM_011790.4 RefSeq (protein) NP_006312.1 NP_035920.1

Location (UCSC) Chr 3 Chr 9

48.96 - 49.02 Mb 108.6 - 108.65 Mb An antisense oligonucleotide that is complementary to and capable of hybridizing to a PAS (directed or targeted to a PAS) decreases usage of the thus targeted PAS in the cell. As a result other PAS will be used. The ratio of the amount of mRNA of the specific gene with poly-adenylation at the targeted PAS over the total amount of mRNA of the specific gene decreases. An antisense oligonucleotide is preferably directed towards the proximal PAS. In that way usage of the more proximal PAS is at least in part inhibited and a more downstream PAS is utilized. The proximal PAS in an mRNA of a gene is the PAS that is downstream of the translation stop codon of the coding region in the mRNA and it is the PAS that is closest to that translation stop codon.

An antisense oligonucleotide that is complementary to and capable of hybridizing to a miRNA target sequence (directed or targeted to a miRNA target sequence) inhibits the effect of the miRNA that it competes with in the cell. The miRNA target sequence is preferably located in the 3' untranslated region of the (pre-)mRNA. The miRNA target sequence is preferably located downstream or distal to the proximal PAS.

A molecular effect of aging in an adult cell is preferably modified by providing the cell with an AON directed towards a pre-mRNA encoded by the gene ARIH2; Psmdl4; Stat5b; HILFA; PTEN; VEGFA; GSK3B; ITGB1; EGFR; YY1; SUMOl; PASMD14; SLC1A4; RBI; HDAC3; RHOA; GRB2; E2F1; RAC1 TDP-43; or C90RF72. Similarly the PAS usage in a (pre-)mRNA expressed by a cell is preferably modified by providing the cell with an AON directed towards a pre- mRNA encoded by the gene ARIH2; Psmdl4; Stat5b; HILFA; PTEN; VEGFA; GSK3B; ITGB1; EGFR; YY1; SUMOl; PASMD14; SLC1A4; RBI; HDAC3; RHOA; GRB2; E2F1; RAC1.

In a preferred embodiment the AON is directed towards a pre-mRNA encoded by the gene ARIH2; Psmdl4; Stat5b; or Grb2. An AON of the invention is preferably directed towards the proximal PAS in a pre-mRNA. Often the default (and normal) mRNA processing is via the distal poly-adenylation site, the proximal poly-adenylation site usage alters mRNA metabolism in a cell. Masking the proximal PAS by specific AON revert mRNA processing to a distal poly-adenylation site usage, and can restore cellular defects (Figure 18).

An antisense oligonucleotide or AON of the invention preferably comprises a sequence that is complementary to the target site on the (pre-)mRNA. With the term complementary is meant capable of hybridizing to the sense strand comprising the target RNA. Such a complementary AON is typically the reverse complement of the sense strand. The AON preferably contains a continuous stretch of between 8-50 nucleotides that is complementary to the target site on the pre- mRNA. An AON of the invention preferably comprises a continuous stretch of at least 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27 , 28 , 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49 or 50 nucleotides that is complementary to the target site. In a preferred embodiment said AON contains a continuous stretch of between 12-45 nucleotides that is complementary to the target site on the pre-mRNA. More preferably a stretch of between 15-41 nucleotides. Depending on the chemical modification introduced into the AON the complementary stretch may be at the smaller side of the range or at the larger side. A preferred antisense oligonucleotide according to the invention comprises a T-0 alkyl phosphorothioate antisense oligonucleotide, such as 2'-0- methyl modified ribose (RNA), 2'-0-ethyl modified ribose, 2'-0-propyl modified ribose, and/or substituted derivatives of these modifications such as halogenated derivatives.

A most preferred AON according to the invention comprises of 2'-0- methyl phosphorothioate ribose. Such AON typically do not need to have a very large complementary stretch. Such AON typically contain a stretch of between 15- 25 complementary nucleotides. Such AON also render the target (pre-mRNA) resistant to the action of RNAse H. As described herein below, another preferred AON of the invention comprises a morpholino backbone. AON comprising such backbones typically contain somewhat larger stretches of complementarity. Such AON typically contain a stretch of between 25-40 complementary nucleotides. When in this invention reference is made to range of nucleotides, this range includes the number(s) mentioned. Thus, by way of example, when reference is made to a stretch of between 8-50, this includes 8 and 50.

An AON of the invention that is complementary to a target RNA is capable of hybridizing to the target RNA under stringent conditions. Typically this means that the reverse complement of the AON is at least 90% and preferably at least 95% and more preferably at least 98% and most preferably at least 100% identical to the nucleotide sequence of the target at the targeted sited. An AON of the invention thus preferably has two or less mismatches with the reverse complement of the target RNA, preferably it has one or less mismatches with the reverse complement of the target RNA. Preferably it has no mismatch with the reverse complement of the target RNA. A mismatch is defined herein as a nucleotide or nucleotide analogue that does not have the same base pairing capacity in kind, not necessarily in amount, as the nucleotide it replaces. For instance the base of uracil that replaces a thymine and vice versa is not a mismatch. A preferred mismatch comprises an inosine. An inosine nucleotide is capable of pairing with any natural base in an RNA, i.e. capable of pairing with an A, C, G or U in the target RNA.

Different types of nucleic acid may be used to generate the oligonucleotide. Preferably, the oligonucleotide comprises RNA, as RNA/RNA hybrids are very stable. Since one of the aims of the present invention is to alter poly-adenylation in subjects it is preferred that the oligonucleotide comprises a modification providing the AON with an additional property. Some types of anti- sense oligonucleotides render the target (pre-mRNA) sensitive to nucleases such as RNAseH. Oligonucleotides that render the duplex formed by the antisense oligonucleotide and the target pre-mRNA sensitive to RNAseH degradation are not preferred for the present invention as these can reduce the amount of target RNA in the cell via RNAi type mechanisms. When only PAS-usage alteration is intended, it is preferred that the anti-sense oligonucleotide does not have this property and thus comprises a modification that renders the renders the duplex formed by the antisense oligonucleotide and the target pre-mRNA resistant to RNAseH. Various oligonucleotides and modifications and variants thereof do not promote the nuclease degradation of the target (pre-)mRNA. Examples of such types of oligonucleotides are 2'0-methyl phosphorothioate oligonucleotides and morpholinos. Other features that may be added to or enhanced in said anti-sense oligonucleotide are increased stability (for instance in a bodily fluid), increased or decreased flexibility, reduced toxicity, increased intracellular transport, tissue- specificity, etc.

In a preferred embodiment, the AON comprises a modified backbone. Examples of such backbones are provided by morpholino backbones, carbamate backbones, siloxane backbones, sulfide, sulfoxide and sulfone backbones, formacetyl and thioform acetyl backbones, methyleneformacetyl backbones, riboacetyl backbones, alkene containing backbones, sulfamate, sulfonate and sulfonamide backbones, methyleneimino and methylenehydrazino backbones, and amide backbones. Phosphorodiamidate morpholino oligomers are modified backbone oligonucleotides that have previously been investigated as antisense agents. Morpholino oligonucleotides have an uncharged backbone in which the deoxyribose sugar of DNA is replaced by a six membered ring and the phosphodiester linkage is replaced by a phosphorodiamidate linkage. Morpholino oligonucleotides are resistant to enzymatic degradation and appear to function as antisense agents by arresting translation or interfering with pre-mRNA splicing rather than by activating RNase H. Morpholino oligonucleotides have been successfully delivered to tissue culture cells by methods that physically disrupt the cell membrane, and one study comparing several of these methods found that scrape loading was the most efficient method of delivery; however, because the morpholino backbone is uncharged, cationic lipids are not effective mediators of morpholino oligonucleotide uptake in cells. A recent report demonstrated triplex formation by a morpholino oligonucleotide, because of the non-ionic backbone; these studies showed that the morpholino oligonucleotide was capable of triplex formation in the absence of magnesium. A modified backbone is typically preferred to increase nuclease resistance of the AON, the target RNA or the AON/target RNA hybrid or a combination thereof. A modified backbone can also be preferred because of its altered affinity for the target sequence compared to an unmodified backbone. An unmodified backbone can be RNA or DNA, preferably it is an RNA backbone.

It is further preferred that the linkage between the residues in a backbone does not include a phosphorus atom, such as a linkage that is formed by short chain alkyl or cycloalkyl internucleoside linkages, mixed heteroatom and alkyl or cycloalkyl internucleoside linkages, or one or more short chain

heteroatomic or heterocyclic internucleoside linkages.

A preferred nucleotide analogue or equivalent comprises a Peptide Nucleic Acid (PNA), having a modified polyamide backbone (Nielsen, et al. (1991) Science 254, 1497-1500). PNA-based molecules are true mimics of DNA molecules in terms of base-pair recognition. The backbone of the PNA is composed of 7V-(2- aminoethyl)- glycine units linked by peptide bonds, wherein the nucleobases are linked to the backbone by methylene carbonyl bonds. An alternative backbone comprises a one- carbon extended pyrrolidine PNA monomer (Govindaraju and Kumar (2005) Chem. Commun, 495-497). Since the backbone of a PNA molecule contains no charged phosphate groups, PNA-RNA hybrids are usually more stable than RNA- RNA or RNA-DNA hybrids, respectively (Egholm et al (1993) Nature 365, 566-568).

A further preferred backbone comprises a morpholino nucleotide analog or equivalent, in which the ribose or deoxyribose sugar is replaced by a 6- membered morpholino ring. A most preferred nucleotide analog or equivalent comprises a phosphorodiamidate morpholino oligomer (PMO), in which the ribose or deoxyribose sugar is replaced by a 6-membered morpholino ring, and the anionic phosphodiester linkage between adjacent morpholino rings is replaced by a non- ionic phosphorodiamidate linkage.

In yet a further embodiment, a nucleotide analogue or equivalent of the invention comprises a substitution of one of the non-bridging oxygens in the phosphodiester linkage. This modification slightly destabilizes base-pairing but adds significant resistance to nuclease degradation. A preferred nucleotide analogue or equivalent comprises phosphorothioate, chiral phosphorothioate, phosphorodithioate, phosphotriester, aminoalkylphosphotriester, H-phosphonate, methyl and other alkyl phosphonate including 3'-alkylene phosphonate, 5'-alkylene phosphonate and chiral phosphonate, phosphinate, phosphoramidate including 3'- amino phosphoramidate and aminoalkylphosphoramidate, thionophosphoramidate, thionoalkylphosphonate, thionoalkylphosphotriester, selenophosphate or boranophosphate. A further preferred nucleotide analogue or equivalent of the invention comprises one or more sugar moieties that are mono- or disubstituted at the 2', 3' and/or 5' position such as a -OH; -F; substituted or unsubstituted, linear or branched lower (C1-C10) alkyl, alkenyl, alkynyl, alkaryl, allyl, aryl, or aralkyl, that may be interrupted by one or more heteroatoms; 0-, S-, or N-alkyl; 0-, S-, or N-alkenyl; 0-, S-or N-alkynyl; 0-, S-, or N-allyl; O-alkyl-0- alkyl, -methoxy, - aminopropoxy; -amino xy; methoxyethoxy; -dimethylaminooxyethoxy; and - dimethylaminoethoxyethoxy. The sugar moiety can be a pyranose or derivative thereof, or a deoxypyranose or derivative thereof, preferably a ribose or a derivative thereof, or a deoxyribose or a derivative thereof. Such preferred derivatized sugar moieties comprise Locked Nucleic Acid (LNA), in which the 2'-carbon atom is linked to the 3' or 4' carbon atom of the sugar ring thereby forming a bicyclic sugar moiety. A preferred LNA comprises 2'-0,4'-C-ethylene-bridged nucleic acid (Morita et al. 2001. Nucleic Acid Res Supplement No. 1: 241-242). These substitutions render the nucleotide analogue or equivalent RNase H and nuclease resistant and increase the affinity for the target RNA.

It is understood by a skilled person that it is not necessary for all positions in an antisense oligonucleotide to be modified uniformly. In addition, more than one of the aforementioned analogues or equivalents may be incorporated in a single antisense oligonucleotide or even at a single position within an antisense oligonucleotide. In certain embodiments, an antisense oligonucleotide of the invention has at least two different types of analogues or equivalents. As mentioned herein above a preferred AON according to the invention comprises a T-0 alkyl phosphorothioate antisense oligonucleotide, such as 2'-0-methyl modified ribose (RNA), 2'-0-ethyl modified ribose, 2'-0-propyl modified ribose, and/or substituted derivatives of these modifications such as halogenated derivatives. A most preferred AON according to the invention comprises of 2'-0-methyl phosphorothioate ribose.

An AON of the invention can be linked to a moiety that enhances uptake of the antisense oligonucleotide in cells. Examples of such moieties are cholesterols, carbohydrates, vitamins, biotin, lipids, phospholipids, cell- penetrating peptides including but not limited to antennapedia, TAT, transportan and positively charged amino acids such as oligoarginine, poly-arginine, oligolysine or polylysine, antigen-binding domains such as provided by an antibody, a Fab fragment of an antibody, or a single chain antigen binding domain such as a cameloid single domain antigen-binding domain, or a recombinant affinity binder such as a affibody.

Preferably, additional flanking sequences are used to modify the binding of a protein to said AON, or to modify a thermodynamic property of the AON, more preferably to modify target RNA binding affinity.

AON administration in humans is typically well tolerated. Clinical manifestations of the administration of AON in human clinical trials have been limited to the local side effects following subcutaneous (SC) injection (on the whole intravenous (i.v.) administration seems to be better tolerated) and generalized side effects such as fever and chills that similar to the response to interferon

administration, respond well to paracetamol. More than 4000 patients with different disorders have been treated so far using systemic delivery of first generation AON (phosphorothioate backbone), and approximately 500 following local administration. The typical dosage used ranged from 0.5 mg/kg every other day for 1 month in Crohn's disease, to 200 mg twice weekly for 3 months in rheumatoid arthritis, to higher dosages of 2 mg/kg day in other protocols dealing with malignancies. Fewer patients (approx. 300) have been treated so far using new generation AON (uniform phosphorothioated backbone with flanking 2' methoxyethoxy wing) delivered systemically at doses comprised between 0.5 and 9 mg/kg per week for up to 3 weeks. Delivery of AON to cells of the brain can be achieved by various means. For instance, they can be delivered systemically through (coated) liposomes that are targeted to the brain or, directly to the brain via intracerebral inoculation (Schneider et al, journal of Neuroimmunology 195 (2008) 21-27). Alternatively, the AON can be coupled to a single domain antibody or the variable domain thereof (VHH) that has the capacity to pass the Blood Brain barrier.

Preferably the complementary part is at least 50% of the length of the oligonucleotide of the invention, more preferably at least 60%, even more preferably at least 70%, even more preferably at least 80%, even more preferably at least 90% or even more preferably at least 95%, or even more preferably 98% and most preferably up to 100% of the length of the oligonucleotide of the invention.

An AON of the invention preferably comprises a sequence that is complementary to part of said pre-mRNA as defined herein. In a more preferred embodiment, the length of said complementary part of said oligonucleotide is of at least 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27 , 28 , 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50 nucleotides. Preferably, additional flanking sequences are used to modify the binding of a protein to said molecule or oligonucleotide, or to modify a

thermodynamic property of the oligonucleotide, more preferably to modify target RNA binding affinity. An AON of the invention may further comprise additional nucleotides that are not complementary to the target site on the target pre-mRNA. In a preferred embodiment an AON contains between 8-50 nucleotides. An AON of the invention preferably comprises a stretch of at least 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27 , 28 , 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49 or 50 nucleotides. In a preferred embodiment said AON contains a continuous stretch of between 12-45 nucleotides, more preferably a stretch of between 15-41 nucleotides. Depending on the chemistry of the backbone as indicated herein above an AON of the invention contains between 15-25 nucleotides. An AON of the invention with a morpholino backbone typically contains a stretch of between 25-40 nucleotides. In a preferred embodiment the indicated amounts for the number of nucleotides in the AON refers to the length of the complementarity to the target pre-mRNA. In another preferred embodiment the indicated amounts refer to the total number of nucleotides in the AON.

With respect to AON that also contain additional nucleotides, the total number of nucleotides typically does not exceed 50, and the additional nucleotides preferably range in number from between 5-25, preferably from 10-25, more preferably from 15-25. The additional nucleotides typically are not complementary to the target site on the pre-mRNA but may be complementary to another site on said pre-mRNA or may serve a different purpose and not be complementary to the target pre-mRNA, i.e., less then 95% sequence identity of the additional nucleotides to the reverse complement of the target pre-mRNA.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1. Arih2 co-immunoprecipitates and co-localize with soluble PABPNl.

A. Immunoprecipitation of PABPNl in soluble and insoluble protein fractions isolated from myotubes that stably express wild-type (AlalO) or exp PABPNl (Ala 17). Upper panel shows blots of protein extract inputs of

PABPNl -FLAG (53 kDa) and MSA loading control. Lower panel shows

immunoprecipitation of PABPNl -FLAG with rabbit-anti-FLAG. VHH-3F5 immunoprecipitates both endogenous (48 kDa) and PABPN1-FLAG fused protein. Immunoblots were carried out with mouse-anti-FLAG (PABPN1- FLAG), 3F5-VHH, anti-Arih2. Muscle actin (MSA) shows equal loading in input. Protein molecular weights are indicated in kDa.

B. Arih2 co-IP with wild type PABPNl (AlalO) and expPABPNl (Ala 17) in the presence of iodoacetamide. Immunoprecipitation of FLAG-tagged PABPNl was carried out with rabbit-anti-FLAG. Immunoblots of IP were carried out with mouse-anti-FLAG and anti-ARIH2 antibodies. Muscle actin (MSA) shows equal loading in input. Protein molecular weights are indicated in kDa.

Figure 2. Arih2 co-localizes with PABPNl in nuclear inclusions.

Hela cells were transfected with YFP-Alal6-PABPN1, and immunofluorescence was performed with anti-Arih2 (red) and anti-ubiquitin (Ubi; blue) antibodies. Co- localization is shown in the merged image and in the fluorescence intensity plots. Small PABPNl foci are indicated with arrows in both images and intensity plots.

Figure 3. PABPNl aggregation is increased in Arih2 down-regulated cells.

A. Hela cells were transduced with shRNA specific to human ARIH2 (272, 273). Cells transduced with shRNA to mouse Arih2 (261) or with HI empty vector were used as controls. Histogram shows RT-qPCR oiARIH2 and PABPNl mRNA expression. Fold change was normalized to GUSB housekeeping gene and to non- transfected cells. Averages are of 3 biological replicates. Asterisks indicate significant down-regulation (pvalue<0.05). Lower panel shows protein levels of ARIH2 in the transduced cell cultures. Actin was used as a loading control.

B. and C. Stably HI- or sh273- transduced cells were transfected with YFP- AlalO- or YFP-Alal6-PABPN1 constructs. B. Cells were immunolabeled with anti-

Arih2 antibody and visualized with Alexa-594 conjugated secondary antibody. Co- localization between PABPNl and ARIH2 is shown in the merged image and in the intensity distribution plots (lower panel) of two representative nuclei. The cross- section lines of intensity distribution are shown in the merged image. Scale bar is 10 pm. C. Histogram shows the fraction of transfected cells with nuclear PABPNl nuclear inclusions. Number of nuclei is indicated in the bars. Fold enrichment of PABPNl nuclear inclusions in ARIH2-DR cells compared to HI control is indicates (±SD). Averages are of two independent experiments.

Figure 4. ARIH2 regulates PABPN1 protein turnover and affects muscle cell fusion.

A. Quantification oiArih2 in C2C12 myoblast cultures transduced with shRNA specific to mouse Arih2 (sh023, sh073). Non-transduced (NT) and cells transduced with shRNA specific to human ARIH2 (sh273) were used as controls. Histogram shows RT-qPCR oiArih2 and Pabpnl mRNA expression. Fold change was normalized to Hprt housekeeping gene and to NT. Averages are of 4 biological replicates. Lower panel shows western blot analysis of Arih2 protein levels in the transduced cultures. Tubulin is used as a loading control. Asterisks indicate significant down-regulation (p-value<0.05).

B. Controls (NT), and Arih2-DR (sh073, sh023) C2C12 cells were treated with 15 μΜ cyclohexamide (CHX) for 6 hours. Immunoblots are of soluble protein extracts. Tubulin shows equal loading. MyoD is used as a control for CHX treatment.

C. Control (NT) and Arih2-DR sh073 cell cultures were transduced with AlalO- or Alal6-PABPN1 fused to YFP. YFP-PABPN1 fused proteins were

immunoprecipitated with anti-GFP antibodies. Protein aliquots that were applied for IP are shown in input. Endogenous Pabpnl was detected with VHH-3F5 and Arih2 with goat-anti-ARIH2 antibody.

D. Immunoprecipitation of endogenous Pabpnl with Arih2 in protein extracts from control and Arih2-DR cells was performed with the VHH-3F5 (IP: VHH-3F5). Pabpnl and Arih2 co-IP was detected by western blotting. Input shows levels of Pabpnl or Arih2 protein in extracts. Equal loading is shown with tubulin and VHH for the input and IP fractions, respectively.

E. Robust quantification of cell fusion. Images show segmentation of myosin heavy chain (Myol) (red) and nuclei (blue) in 6 days fused mock or sh073 Arih2-DR C2C12 cultures. Myonuclei in fused cells are depicted in green. Histogram shows the percentage of nuclei in Myol expressing cells in mock, sh023 or sh027 C2C1 cultures. N indicates that number of quantified nuclei. Averages and SD are of 4 replications.

Figure 5. Arih2 expression decreases in OPMD and during muscle aging and correlates with PABPNl expression. A. RT-qPCR analysis οΐΑΡιΙΗ2 in quadriceps from expPABPNl carriers at pre- symptomatic (pre-symp; N=6) or symptomatic (OPMD; N=9) stages, and age- matched control groups. B-box plots show fold changes, which are normalized to the average of GUSB housekeeping gene. Student's T-Test statistical significance (p<0.05) is indicated with an asterisk.

B. RT-qPCR analysis oiARIH2 in blood from OPMD patients (N=10) and age- matched controls (N=13). B-box plots show fold changes, which are normalized to the average of GUSB housekeeping gene.

C. RT-qPCR analysis oiARIH2 and PABPN1 in Vastus lateralis in healthy controls age 17-83 years (N=73). Log2 expression level oiARIH2 is depicted in black diamonds and PABPNl in white circles. Quadratic fit curves are indicated. Statistical significance of quadratic fit (p<0.005; adjusted for gender) is indicated with asterisks.

D. Scatter plot of PABPNl vs. ARIH2 expression levels (N=73). Line indicates a linear correlation. Beta of linear regression between ARIH2 and PABPNl expression is indicated. Statistical significance (p<0.005; adjusted for gender) is indicated with two asterisks.

E. The correlation between ARIH2 and PABPNl expression levels in expPABPNl carriers at pre-symptomatic (pre-symp; N=6) and symptomatic (OPMD; N=9) patients. A linear correlation is depicted with hashed or continuous lines for OPMD and pre-symptomatic, respectively. Betas of linear regression are indicated.

Statistical significance (p<0.005; adjusted for gender) is indicated with two asterisks, and pO.01 with one asterisk.

Figure 6. A switch in poly(A) site usage oiArih2 is regulated by PABPN1.

A. Schematic presentation of the Arih2 3'-UTR. Proximal and distal

polyadenylation sites (PAS) are indicated, as well as the relative PAS usage in wild type (black), A17.1 mice, or Pabpnl-OR (red). The location of primers that were used to quantify long or short Arih2 transcripts is indicated with double-head arrows.

B. Histogram shows the ratio between long and short transcripts of Arih2 in A17.1 (N=5) and FVB wild type (WT) control (N=5). RT-qPCR of RNA from quadriceps of mice at 6 and 16 weeks. Statistical significance between WT and A17.1 (p<0.0005) is indicated with asterisks.

C. Histogram shows the ratio between Arih2 transcripts with long or short 3'UTRs in RNA-Pabpnl complex in Pabpnl-OR or HI and untransduced control C2C12 myoblasts. A control RNA-IP incubated sepharose-A beads within VHH-3F5 (no ab) is included. Averages and SD are from three biological replicates. Statistical significance between PABPN1-DR and controls (p<0.005) is indicated with asterisks

D. A schematic model describing the regulation of PABPN1 aggregation in OPMD and aging. PABPN1 protein accumulation is regulated by ARIH2 and depends on the proteasome. During aging ARIH2 expression level declines and induces PABPN1 accumulation and a subsequent aggregation. Reduced levels of soluble PABPN1 cause alternative PAS usage in ARIH2 transcript, and a decline in ARIH2 expression. A reduction in soluble PABPNl results in genome-wide changes in mRNA stability, leading to muscle dysfunctionality. In OPMD, pre- aggregated structures of expPABPNl cause faster aggregation of PABPNl and reduced levels of soluble PABPNl, causing a decrease in ARIH2 expression. The levels of ARIH2 are also reduced by entrapment in PABPNl nuclear inclusions. The combination of protein aggregation and a decline in expression levels of both PABPNl and ARIH2 cause premature muscle weakness in OPMD. Figure 7. Anti-sense oligonucleotides against the proximal PAS of ARIH2 or the mirl9 binding site in the 3'UTR of ARIH2 result in increased levels of ARIH2 RNA in the cell. Muscle cell cultures were transfected with AON lOOOmicroMolar using

polyethylene glycol. The control AON 47/2 is fluorescence-conjugated and was use to evaluate transfection efficiency. &2 hours post-transfection RNA was extracted from cells and cDNA was made with oligo d(T). Expression levels were determined with RT-qPCR using primers specific to ARIH2 last exon. Fold change was calculated by normalization to GUSB house keeping gene and control transfection. levels of GapDH are used as control for specific effect of AON. ARIH2 AON3 has the sequence 5'-GTATAATTGTACAACCTTTGAAAG-'3. Mir 19 AON has the sequence 5'-TAACTTGTGCAAACACAGCC-3';

Figure 8. Sequence of the untranslated region of human ARIH2.

The AON binding site for the AON directed towards the proximal PAS is indicated in bold. The AON directed towards the miR19 binding site is underlined. Figure 9. PABPNl protein accumulation significantly decreases in aging in senescent cells. A) Chart bar shows PABPNl accumulation in human mesenchymal stem cells. PABPNl accumulation was determined from western blot analysis of total protein extracts from 5 young (19-29 years) and 6 old (79-87 years) donors. PABPNl expression was normalized to actin. Averages from all samples represent two independent experiments. Cultures were propagated in vitro for a single passage. B) Chart bar shows PABPNl accumulation senescent human fibroblasts. PABPNl accumulation was determined from western blot analysis of soluble protein extracts from cultures at passage 13 and passage 27. PABPNl expression was normalized to actin. Averages are from four independent experiments. C) Chart bar shows nuclear PABPNl accumulation in primary muscle cell cultures that were fused for 6 days. PABPNl was visualized with the immuno-fluorescence procedure and nuclear PABPNl was measured from segmented nuclei. PABPNl integrated fluorescence was normalized to DAPI staining. Averages are from 100 nuclei and three experiments.

Figure 10. Pabpnl and Arih2 expression in aging control mice. Expression levels were normalized to Hrpt house keeping gene and mice at 4 month. The age- associated increase in expression is highly significant. P<0.0005. Figure 11: ARIH2 expression decreases in OPMD and during muscle aging and correlates with PABPNl expression.

A. RT-qPCR analysis of ARIH2 in quadriceps from expPABPNl carriers at pre- symptomatic (pre-symp; N=6) or symptomatic (OPMD; N=9) stages, and age- matched control groups. B-box plots show fold changes, which are normalized to GUSB housekeeping gene. Student's T-Test statistical significance (p<0.05) is indicated with an asterisk.

B. RT-qPCR analysis of ARIH2 in blood from OPMD patients (N=10) and age- matched controls (N=13). B-box plots show fold changes, which are normalized to

GUSB housekeeping gene.

C. RT-qPCR analysis of ARIH2 and PABPNl in Vastus lateralis muscles in healthy controls age 17-83 years (N=73). Log2 expression level of ARIH2 is depicted in black diamonds and PABPNl in white circles. Quadratic fit curves are indicated and for both genes the quadratic fit is significant (p<0.005; adjusted for gender).

D. Scatter plot of PABPNl vs. ARIH2 expression levels (N=73) in healthy controls. Line indicates a linear correlation. Beta of linear regression between ARIH2 and PABPNl expression is indicated. Statistical significance (p<0.005; adjusted for gender) is indicated with two asterisks.

Statistical significance (p<0.005; adjusted for gender) is indicated with two asterisks.

Figure 12: ARIH2 co-localizes with PABPNl aggregates.

HeLa cells were transfected with YFP-Alal6-PABPN1, and immunofluorescence was performed with anti-ARIH2 (red) and anti-ubiquitin (Ubi; blue) antibodies. Co- localization is shown in the merged image and in the fluorescence intensity plots. Small PABPNl foci are indicated with arrows in both images and intensity plots. Figure 13: PABPN1 aggregation is increased in ARIH2 down-regulated cells.

A. HeLa cells were transduced with shRNA specific to human ARIH2 (272, 273). Cells transduced with shRNA to mouse ARIH2 (261) or with HI empty vector were used as controls. Histogram shows RT-qPCR of ARIH2 and PABPN1 mRNA expression. Fold change was normalized to GUSB housekeeping gene and to non- transfected cells. Averages are of 3 biological replicates. Asterisks indicate significant down-regulation (p-value<0.05). Lower panel shows protein levels of ARIH2 in the transduced cell cultures. Actin was used as a loading control.

B. Co-immunoprecipitation of ARIH2 or PABPN1 in HI- or ARIH2-DR (sh273) cultures. Proteins were extracted in the presence of iodoacetamide and

immunoprecipitation (IP) for ARIH2 was carried out with R_anti-ARIH2 and for PABPN1 with VHH_3F5. Immunoblots were carried out with goat_anti-ARIH2, VHH-3F5 or mouse anti-Ubiquitin (Ubi). Loading controls are shown in the input fraction. Arrow indicates ubiquitinated PABPN1.

C. and D. HI- or ARIH2-DR (sh273) cultures were transfected with YFP-AlalO- or YFP-Alal6-PABPN1 constructs. All experiments were performed two days after transfection. C. ARIH2 co-IP with PABPN1. (i) Immunoblot was carried out with goat-anti-ARIH2 and rabbit- anti-GFP antibodies. Loading controls are shown in the input fraction, (ii) Chart bar shows the ratio of PABPN1 to ARIH2 in the IP fraction after normalization to PABPN1 input. PABPN1 in input is normalized to Actin. Results are from a representative experiment, the exact values varied between experiments according to transfection efficiency. D. ARIH2 is visualized with anti-ARIH2 and Alexa-594 conjugated secondary antibody, (i) Co-localization between PABPN1 and ARIH2 is shown in the merged image and in the intensity distribution plots (lower panel) of two representative nuclei. The cross-section lines of intensity distribution are shown in the merged image. Scale bar is 10 pm. (ii) Bar chat shows the fraction of transfected cells with nuclear PABPN1 nuclear inclusions. Number of nuclei is indicated in the bars. Fold enrichment shows PABPN1 nuclear inclusions in ARIH2-DR cells compared to HI control. Averages are from three independent experiments. Figure 14: ARIH2 co-IP with soluble wild type PABPN1 in myotube cultures.

Soluble or insoluble proteins were extracted from 4 days myotube cultures that stably express wild type (AlalO) or expPABPNl (Alal7) fused to FLAG. Parental culture (IM2) was used as control. PABPN1 IP was carried out with VHH-3F5.

Immunoblot was carried out with goat-anti-ARIH2, mouse -anti- FLAG antibodies or antibodies to muscle actin (MSA). Loading controls are shown in the input fractions. Molecular weights are indicated in kDa. Figure 15: ARIH2 regulates PABPN1 protein turnover and affects muscle cell fusion.

A. Results are from a representative experiments down-regulated stable cultures in C2C12 myoblasts were generated with shRNA specific to mouse ARIH2 (sh023, sh073) lentiviruses. Non-transduced (NT) and cells transduced with shRNA specific to human ARIH2 (sh273) were used as controls. Histogram shows RT- qPCR of ARIH2 and PABPN1 mRNA expression. Fold change was normalized to HPRT housekeeping gene and to NT culture. Averages are of 4 biological replicates. Lower panel shows western blot analysis of ARIH2 protein levels in the transduced cultures. Tubulin is used as a loading control. Significant down- regulation (p-value<0.05) is indicated with asterisks.

B. Co-immunoprecipitation of ARIH2 with PABPN1. Control (NT) and ARIH2- DR (sh073) cultures were transduced with AlalO- or Alal6-PABPN1 fused to YFP. YFP-PABPN1 fused proteins were immunoprecipitated with anti-GFP antibodies. Protein aliquots are shown in input. PABPN1 was detected with VHH-3F5 and ARIH2 with goat-anti-ARIH2 antibody.

C. PABPN1 protein turnover in C2C12 cultures. Controls (NT), and ARIH2-DR (sh073, sh023) C2C12 cells were treated with 15μΜ cycloheximide (CHX) for 6 hours. Soluble proteins were used to western blot analysis. PABPN1 was detected with VHH-3F5. Tubulin shows equal loading and MyoD was used as a control for CHX treatment.

D. Quantification of cell fusion. Images show segmentation of myosin heavy chain (MHC1) (red) and nuclei (blue) in 6 days fused mock or ARIH2-DR (sh073) C2C12 cultures. Myonuclei in fused cells are depicted in green. Histogram shows the percentage of nuclei in MHC1 expressing cells in mock or ARIH2-DR cultures. The number of nuclei is indicated (N). Averages and SD are of 6 experiments.

Statistical significance (p<0.05) is indicated with an asterisk. Figure 16: PABPN1 and ARIH2 transcripts binds to PABPN1 proteins.

PABPN1 and ARIH2 mRNA abundance in input and PABPN1-RIP fractions from PABPN1-DR or HI control C2C12 cultures were determined with RT-qPCR. Fold change in the input fraction was determined after normalization to HRPT house keeping gene, and fold enrichment in the RIP fraction was calculated after normalization to the input fraction. RIP was carried out with VHH-3F5. Averages and SD are from four biological replicates. Statistical significance (p<0.05) between PABPN1-DR and control cultures is indicated with an asterisk.

Figure 17: A switch in poly(A) site usage of ARIH2 3'-UTR is regulated by PABPN1.

A. Schematic presentation of the ARIH2 3'-UTR. Proximal and distal polyadenylation sites (PAS) are indicated, as well as the relative PAS usage in wild type (black), A17.1 mice (N=6) or PABPN1-DR (red) (N=5). The location of primers that were used to quantify long or short ARIH2 transcripts is indicated with double-head arrows.

B. A snapshot UCSC genomic browser of RNA-seq from the 3-UTR in ARIH2 mRNA. A change in PAS preference is found in the A17.1 OPMD mouse compared with FVB control mouse.

C. Bar chart shows the ratio between long and short transcripts of ARIH2 (left) and expression fold change (right) in A17.1 (N=5) and FVB control (N=5). RT-qPCR of RNA from quadriceps of mice at 6 weeks. Statistical significance (p<0.005) between FVB and A17.1 is indicated with an asterisk.

D. Bar chart shows the ratio between ARIH2 transcripts with long or short 3'- UTRs in input and RIP fractions from PABPN1-DR or HI control C2C12 cultures. RIP was carried out with VHH-3F5 and incubation with sepharose-A beads without VHH-3F5 was used as a control (mock). Averages and SD are from three biological replicates. Statistical significance (p<0.05; p<0.005) between PABPN1- DR and control cultures is indicated with one or two asterisks, respectively. Figure 18: Blocking proximal ARIH2 PAS restores ARIH2 and PABPN1 expression levels and restores cell fusion defects in PABPN1-DR cells.

A. Histogram shows expression levels of ARIH2 in PABPN1-DR and in control cultures, non-transformed and HI empty vector) 7304 myoblast cultures. Fold change was calculated after normalization to GUSB housekeeping gene and to NT. Averages and SD are from three biological replicates. Statistical significance (p<0.005) in PABPN1-DR cells is indicated with an asterisk.

B. Histogram shows expression levels of ARIH2 and PABPN1 in 7304 control cultures (left) or in PABPN1-DR myoblast cultures (right), after transfection with control AON or AONs to ARIH2 proximal PAS (AON1 and AON2). Fold change was calculated after normalization to GUSB housekeeping gene and to NT control. Averages and SD are from six biological replicates. Statistical significance (p<0.05; p<0.005) is indicated with one or two asterisks, respectively.

C. Quantification of cell fusion after AON transfection. Cell fusion was analyzed in fused cultures, 7 days after AON transfection. Images show

segmentation of myosin heavy chain (MHCl) (red) and nuclei (blue) in mock or PABPN1-DR 7304 cultures. Myonuclei in fused cells are depicted in yellow.

Histogram shows the percentage of nuclei in MHCl expressing cells in mock, 7304 cultures after transfection with control AON, and AONs to ARIH2 proximal PAS, AON1 and AON2. Per condition, >10,000 nuclei were analyzed. Averages and SD are of three experiments. Statistical significance (p<0.05) is indicated with one asterisk. Figure 19: A feed-forward graphical model for age-regulated expression of ARIH2 and PABPN1 in muscles.

PABPN1 regulates ARIH2 mRNA stability (depicted in black) via PAS usage and PABPN1 protein (depicted in blue and expPABPNl is depicted in red)

accumulation is regulated by ARIH2 protein (depicted in orange). With age expression levels of both PABPN1 and ARIH2 mRNAs declines due to two regulatory loops: reduced ARIH2 induces increase in PABPN1 aggregation and thus reduced educed levels of soluble and functional PABPN1. Reduced PABPN1 causes proximal PAS usage in ARIH2 transcript (depicted in gray), and thus a decline in ARIH2 expression. In addition, PABPNl self-regulates its mRNA levels. A decline in both ARIH2 and PABPNl levels is aging-regulated and is not affected in carriers of mutant PABPNl at a pre-symptomatic stage. However, in muscles of OPMD patients ARIH2 and PABPNl levels further reduced due aggregation of mutant PABPNl and entrapment of ARIH2 in PABPNl aggregates (aggregates are marked with a gray circle). Reduced soluble PABPNl causes genome-wide changes in mRNA expression profiles, and in muscle tissues leads to muscle degeneration.

EXAMPLES

Example 1 RESULTS E3 ligase Arih2 preferentially binds to wild type PABPNl

Protein turnover of nuclear proteins is predominantly regulated by the ubiquitin proteasome system. Poly-ubiquitination levels of PABPNl differ between wild type (WT) and alanine -expanded PABPNl resulting in differences in protein

accumulation and aggregation (Raz et al. 2011b). We searched for an E3 ligase regulating PABPNl protein accumulation, the pathological hallmark of OPMD muscle. The expression of E3 ligases is significantly affected in OPMD (Anvar et al., 2011a). Among those, the expression of E3-ligase ARIH2 is consistently down- regulated in OPMD and model systems to OPMD (Anvar et al., 2011a). To investigate whether ARIH2 binds PABPNl, co-immunoprecipitation (IP) experiments were performed in muscle cells that stably express FLAG-tagged wild type- (WT) or Alanine expanded PABPNl (expPABPNl) at similar expression levels (Raz et al. 2011b). IP with antibodies to PABPNl revealed Arih2 co-IP (Figure 1A). PABPNl IP was performed with either anti-FLAG, which binds only the transgene product, or with VHH-3F5, which binds both the transgene product and endogenous PABPNl, and Arih2 was co-IP in both conditions, indicating that binding of Arih2 to PABPNl is not an artifact of the IP. Interestingly, Arih2 co-IP was enriched in WT-PABPN1 expressing myotubes compared with expPABPNl (Figure 1A). Since functional PABPNl is soluble (Kuhn et al., 2009), we next compared Arih2 co-IP with soluble or insoluble PABPNl. PABPNl was IP from both soluble and insoluble protein fractions, while Arih2 preferentially co-IP with soluble PABPNl (Figure 1A). Together this suggests a preference binding of Arih2 to WT and soluble PABPNl.

Since poly-ubiquitination of WT-PABPN1 is higher than that of exp PABPNl (Raz et al. 2011b), we next investigated whether Arih2 binds to ubiquitinated PABPNl. In the presence of iodoacetamide, an inhibitor of deubiquitinating enzymes, the co- IP of WT-PABPN1 with Arih2 was more abundant compared with expPABPNl (Figure IB). This suggests that Arih2 binds to ubiquitinated PABPNl.

E3 ligase Arih2 regulates PABPNl aggregation

PABPNl aggregation is considered as the cause for muscle weakness in OPMD. PABPNl aggregation in mitotic cells, like HeLa cells is fast and is caused by expression of both WT- and expPABPNl (Abu-Baker et al. 2003; Raz et al. 2011a). High-resolution fluorescence microscopy reveals that ARIH2 co-localizes with PABPNl fluorescence foci in HeLa cells expressing expPABPNl fused to yellow fluorescent protein (YFP) (Figure 2). Importantly, co-localization of both PABPNl and ARIH2 was also with ubiquitin, suggesting a complex of the three (Figure 2). Therefore, to investigate a role for ARIH2 in PABPNl aggregation, ARIH2 expression was stably down-regulated (ARIH2-DR) in HeLa cells using different shRNA vectors. A significant reduction in ARIH2 RNA and protein levels was obtained with shRNAs 272 and 273 (Figure 3A). Empty vector (HI) and an unspecific shRNA (261), were used as controls (Figure 2A). Down-regulation of ARIH2 resulted in reduced co-localization between foci of PABPNl fused to yellow fluorescent protein (YFP) and ARIH2 (Figure 3B). Noticeably, ARIH2 co- localization with expanded (Alal6) YFP-PABPN1 was less affected in ARIH2-DR cells (Figure 3B).

Next, we investigated whether ARIH2-DR also affects PABPNl aggregation.

Overexpression of both WT or expPABPNl in Hela cells leads to PABPNl aggregation, and in ARIH2 DR-cells the proportion of cells with aggregated

PABPNl was significantly higher compared with control cells (Figure 3C). This suggests that ARIH2 regulates accumulation of PABPNl. The increase in nuclei with PABPNl nuclear inclusions was about 2-fold higher in WT-PABPN1 transfected cells compared with expPABPNl-transfected cells (Figure 3C), further suggesting a preference of ARIH2 for WT-PABPN1, which is consistent with preferential binding of Arih2 to WT-PABPN1 (Figure 1). Arih2 regulated PABPNl protein turnover and affects muscle cell fusion

Since in OPMD exp PABPNl leads to muscle symptoms, we next investigated whether ARIH2 affects PABPNl protein accumulation in muscle cells. Same as in HeLa cells, Arih2 expression was down-regulated in C2C12 mouse muscle cells using shRNAs 023 and 073, which are specific to mouse Arih2. Both shRNAs led to Arih2 mRNA down-regulation by 50% and 40%, respectively, which was confirmed on the protein level (Figure 4A). The shRNA 273, which is specific to human ARIH2, did not affect Arih2 levels (Figure 4A). To determine whether Arih2 regulates PABPNl turnover cells were treated with cyclohexamide (CHX), an inhibitor of protein biosynthesis. After 6 hours CHX treatment Pabpnl protein accumulation was reduced in control cells but not in Arih2-DR cells (Figure 4B). This suggests that Arih2 regulates Pabpnl protein accumulation.

To validate a preferential binding of Arih2 to WT-PABPN1, control and Arih2- shRNA C2C12 cells were transduced with lentivirus vectors expressing WT (AlalO)- or expPAPBNl (Ala 16) fused YFP. Reduced co-IP of Arih2 with PABPNl was found in Arih2-DR cells (Figure 4C). This result further demonstrates that

Arih2 preferentially binds to WT PABPNl. Moreover, in this cell model Arih2 co-IP with endogenous PABPNl (Figure 4D). Interestingly, western blot analysis in input reveals that in the Arih2-DR cells PABPNl expression levels decreases (Figure 4D). RT-qPCR of PABPNl in those cells shows that in the Arih2-DR cells expression levels of PABPNl mRNA also decreases.

PABPNl knockdown in muscle cells leads to reduced cell fusion (Apponi et al. 2010). Also in our cell model, down-regulation of PABPNl leads to reduced cell fusion, which is associated with reduced expression of myogenic genes

(Supplementary Figure 1). Therefore, we investigated whether Arih2-DR cells also has myogenic defects. A robust semi-automatic image quantification protocol was applied to determine the fusion index. Fusion index is calculated from the proportion of nuclei in cells expressing MYH1. A significant decrease in C2C12 cell fusion was found in both sh023 and sh073 Arih2-DR cells compared with control cells (Figure 4E). As both down-regulation of both PAPBN1 and Arih2 induce reduced cell fusion and both proteins co-IP in muscle cells it suggests a casual relationship between the two proteins. ARIH2 expression levels decline in OPMD and in muscle aging and are strongly associated with PABPNl expression levels in muscles

Microarray study of quadriceps from OPMD patients suggests a decline in

ARIH2 expression in OPMD (Anvar et al. 2011a). To validate this observation, RT-qPCR was performed on extended Vastus lateralis muscles (VL) samples. A significant decrease in ARIH2 expression was found in OPMD patients (age range 49-69) compared with controls (Figure 5A). This decline in ARIH2 expression was associated with symptoms and was not found in exp PABPNl carriers at a pre-symptomatic stage (age range 33-41) (Figure 5A). Since we observed that a decline in ARIH2 causes a decline in PABPNl expression in muscle cell culture (Figure 4A), the expression of PABPNl was next determined in the same muscle samples. Similar to ARIH2 trend in OPMD, PABPNl expression was also significantly decreased (Figure 5A). This suggests that changes in ARIH2 and PABPNl expression in OPMD are associated with symptoms. To verify this, RT-qPCR was performed in blood samples from OPMD patients and controls. However, no changes in ARIH2 or PABPNl expression were found between OPMD and controls (Figure 5B). This suggests that a decrease in the expression oiARIH2 and PABPNl in OPMD is associated with symptoms.

Since OPMD is a late onset disorder, we next investigated whether the expression oiARIH2 and PABPNl is also age -regulated. A significant decline in ARIH2 and PABPNl expression was found in a cross-sectional cohort of 74 in VL muscles from healthy controls (Figure 5C). A quadratic regression model suggests a decrease in the expression of both genes from mid-life (Figure 5C). When applying a linear regression model on age groups, a significant decrease in the expression oiARIH2 was found from 36 years, and from 43 years for PABPNl (Table 1 and Anvar et al., 2013, respectively). Age-regulated expression of both ARIH2 and PABPNl was not found in the full dataset from blood, kidney medulla, brain cortex and kidney cortex (Table 1, and Anvar 2013, respectively). Moreover, an age-regulated expression of both ARIH2 and PABPN 1 was also not found in Rectus abdominis skeletal muscles, a muscle that shows minor aging-associated changes (Marzani et al. 2005). In brain cortex we identified a small reduction in PABPN 1 expression in elderly (>70 years) (Anvar 2013). Compared with PABPN1 expression in VL muscles, the decrease in the brain cortex was smaller and delayed (Anvar 2013). Interestingly, the trend oiARIH2 expression was highly similar to that oi PABPN 1 (Figure 5C), and a significant positive correlation (p<0.005) was found between ARIH2 and PABPN 1 expression during aging (Figure 5D). A positive correlation between ARIH2 and PABPN 1 was also found in RT-qPCR from expPABPNl carriers at a pre-symptomatic stage but not in symptomatic patients (Figure 5E). Together, this analysis suggests that ARIH2 and PABPN 1 expression levels are tightly correlated in healthy skeletal muscles. This conclusion is supported by the in vitro experiment showing that ARIH-DR in C2C12 cells but not in HeLA cells causes a decline in PABPN 1 expression (Figure 4A and Figure 3A, respectively).

PAPBN1 levels regulate PAS usage ΪΑΚΙΗ2

PABPN1 regulates mRNA stability by regulating PAS usage (E. de Klerk 2012; Jenal et al. 2012), we therefore also studied whether changes in Arih2 mRNA levels are associated with alternative PAS usage. Distal and proximal PAS generate long or short 3' UTR, respectively. A change in the ratio between long and short 3' UTR can be a measure for a change in PAS usage. RT-qPCR analysis was performed with primers that specifically amplify long or short Arih2 3' mRNA ends (figure 6A). In the OPMD mouse model, A17.1 the ratio between long and short Arih2 transcripts significantly decreased compared with the wild-type control mice (Figure 6A). An increase in the short transcript was consistent in both 6 weeks and 26 weeks-old mice. This suggests that reduced expression oiArih2 in A17.1 is associated with an alternative PAS usage. Next we investigated whether a change in PAS is associated with Arih2 mRNA binding to PABPN1 using the RNA- immunoprecipitation (RIP) procedure. PABPN 1-RNA complexes were

immunoprecipitated with VHH-3F5 and Arih2 transcripts were amplified with the same primer sets as in Figure 6A. The ratio between PCR products representing the long or short 3' UTRs significantly decreased in sh536, Pabpnl-DR cells, compared with controls (Figure 6B). Western blot analysis of input indicates equal aliquots in IP, and a specific pull-down of PABPNl in the immune complex (Supplementary Figure 2). The specificity of protein-RNA pull-down was demonstrated in an IP reaction without an antibody (Figure 6B). A decrease in Pabpnl level in sh536 cells was confirmed by western blot and RT-qPCR analysis (Supplementary Figure 2). Consistent with the RIP result, the ratio between long and short Arih2 3' UTR in sh536-transduced cells indicates a higher expression of the shorter 3' UTR PCR product in cells with reduced levels of endogenous Pabpnl (Supplementary Figure 2). Together, these experiments reveal a post- transcriptional regulation oiArih2 expression by PABPNl levels, where either a decline in Pabpnl expression or aggregation of expPABPNl result in a decrease ratio between Arih2 transcripts containing long or short 3' UTR.

Materials and Methods

Human materials, RNA extraction and RT-qPCR

A summary of human samples is listed in Supplementary Table 1. All human muscle biopsies presented in this study were collected at Radboud Hospital, Nijmegen, Canisius-Wilhelmina Hospital, Nijmegen, The Netherlands, and

Rigshospitalet, Denmark, after an approval of the medical ethical committee Arnhem-Nijmegen (CMO nr. 2005/189) and of the local ethical committee, from The NL and Denmark, respectively. OPMD patients and pre-symptomatic were genetically confirmed and underwent clinical investigation including MRC score prior to sampling of muscle biopsy. All quadriceps biopsies were collected using the Bergstrom needle procedure. The biopsies froze immediately in liquid nitrogen and stored at -80 before RNA extraction.

Cell culture and lentivirus transduction

HeLa cells were grown in Dulbecco's modified Eagle's medium (DMEM) +

Glutamax™-I (GIBCO^® Invitrogen) with 4,5 g/L glucose, supplemented with 10% fetal calf serum (FCS) (GIBCO^® Invitrogen) and 100 U/ml antibiotics

(streptomycin, penicillin) in a humidified 5% C02-air atmosphere at 37 °C.

IM2 cells and the PABPN1 clones were previously described (Raz 2011). In brief, cells were grown in Dulbecco's modified Eagle's medium (DMEM) + Glutamax™-I (GIBCO^® Invitrogen) with 4,5 g/L glucose, supplemented with 20% fetal calf serum (FCS) (GIBCO^® Invitrogen), 0,5% chicken embryo extract (CEE) (PAA

Laboratories, Somerset, UK), 20 U/ml INF-γ (Hycult^® Biotech) and 100 U/ml antibiotics (streptomycin, penicillin) (GIBCO^® Invitrogen) at 33 °C in humidified 10% C02-air atmosphere. For fusion, the myoblasts were seeded onto collagen coated dishes and grown to a confluence of -80%. Differentiation was induced by complete removal of the growth medium and substitution with a fusion medium (DMEM + Glutamax™-I with 4,5 g/L glucose, complemented with 5% horse serum (HS) and 100 U/ml antibiotics (streptomycin/penicillin) (all GIBCO^® Invitrogen) and following incubation at 37 °C in humidified 5% C02-air atmosphere. The cells were harvested after 3 to 5 days, when the successful fusion to myotubes became visible under the microscope.

C2C12 cells were grown in DMEM supplemented with 20% FCS and antibiotics. Treatments with 5 μΜ cyclohexamide were conducted for 10 hours in culture at 90% confluence.

Cells at about 60% confluence were transduced with 1-2 MOI lentivirus particles in a medium supplemented with 0.1% polybrene. The virus was washed after 18 - 24 hours and was replaced by a normal growth medium. Puromycin selection (4 pg/ml) was applied to cells transduced with shRNA. The shRNA constructs were obtained from Sigma-Aldrich. shRNA for human ARIH2 (NM_006321) are:

TRCN0000034269 (269); TRCN0000034272 (272); TRCN0000034273 (273). shRNA for mouse Arih2 (NM_011790) are: TRCN0000041023 (023); TRCN0000041027 (027). ShRNA for mouse PABPN1 is: TRCN0000102536 (536), and for human PABPN1: TRCN0000000121 (121), TRCN0000000122 (122) and TRCN0000000123 (123). Between brackets are the symbol for each sh-vector that are used in the document. Lentiviruses for YFP-Alal6-PABPN1 or YFP-Alal0-PABPN1 were generated from the expression vectors described in (Raz et al. 2011a). Lentivirus particles were produced as described in (Raz et al. 2006). Four independent transduction experiments were performed for every virus.

Microarray and RT-qPCR analysis

ARIH2 and PABPN1 expression levels were determined from the previously described microarray studies in mouse (GEO GSE26604, (Trollet et al. 2010), in human quadriceps (GEO GSE26605) (Anvar et al. 2011a), frontal cortex: (GEO- GD707, GEO-GSE1572) (Lu et al. 2004), Rectus abdominis (GEO-GSE5086) (Zahn et al. 2006), blood (GEO-GSE16717) (Passtoors et al. 2012), kidney cortex and medulla (Rodwell et al. 2004). Statistical analyses of linear and quadratic models, in microarray studies and RT-QPCR were carried out with PASWStatistics 18.0 for Mac (IBM). Total RNA extraction from cell cultures and muscle biopsies for RT- qPCR was performed as described in (Anvar et al. 2011b). 3ng of purified cDNA served as template per RT-qPCR reactions. Detection of the PCR product was provided by iQ-SYBR Green (Bio-Rad). For calculations of fold change

normalization with house keeping genes GUS in HeLa cells, in human biopsies expression was normalized to both GUSB and GapDH, in C2C12 HPRT was used as a house keeping gene, followed by normalization with un-manipulated cells or with age-group 17-25 years. The primers used are listed in table 2. Protein assays

Total proteins were extracted in a RIPA-buffer (20mM Tris pH 7.5, 150mM NaCl, 5mM EDTA, 1% NP40, 0.05% SDS), and soluble proteins were extracted with a buffer composed of 20mM Tris pH 7.5, 10% glycerol, 150mM NaCl and 5mM EDTA. Protease inhibitor cocktail (SigmaFAST™ protease inhibitor tablets, Sigma- Aldrich^®) was freshly added. Soluble fraction was separated from the insoluble fraction by centrifugation. Insoluble proteins were recovered with RIPA buffer supplemented with 1% triton and 0.05% SDS, and ensuing sonification. When indicated, inhibition of deubiquitinating enzymes was performed with lOmM iodoacetamide (Sigma- Aldrich) in the extraction buffer. Immunoprecipitations of PABPNl with Rabbit anti-FLAG (Sigma-Aldrich) or VHH-3F5 were performed as described in (Raz et al. 2011b). Proteins were fractionated on SDS-PAGE and transferred onto polyvinylidene fluoride (PVDF) membrane. First antibodies used are: Goat anti-Arih2 (Everest Biotech), mouse -anti -muscle actin (Santa Cruz Biotechnology^®), mouse-anti-FLAG (Sigma-Aldrich^®); mouse anti-Ubiquitin (FK2, Tebu-bio); mouse anti-tub ulin (Sigma-Aldrich^®); MyoD (M-318 Santa Cruz). The VHH-3F5 (Verheesen et al. 2006) intrabody was detected with rabbit anti-VHH. Antibodies were detected with the Odyssey® Infrared Imaging System (LI-COR® Biosciences) and applicable IRDye Secondary antibodies.

Immunohistochemistry

Cells were seeded on glass cover-slides and transfected with either YFP-AlalO or YPF-Alal6 PABPNl as described in (Raz et al. 2011a). 48h after transduction cells were fixed with 2% formaldehyde, and permeabilized with 1% triton for 15 minutes. Immunohistochemistry was performed with mouse anti-Ubiquitin (FK2) TeBu-Bio; goat anti-Arih2, Everest Biotech; anti-Myosin MF20 (Sigma-Aldrich, MO, USA); anti-Desmin (1:500; Cell Signalling Technology, MS, USA) and the anti- PABPN1, 3F5 llama single chain antibody (1: 1000), recognised with rabbit-anti- VHH (1:2000). Secondary antibodies were donkey-anti mouse Cy5; donkey anti- goat- alexa594. Slides were imbedded with Citifluor (Agar Scientific) after ethanol dehydration. Microscopic images were captured using the Leica DMRA

fluorescence microscope with the 40x and 100X lenses NA 1.4 plan Apo objective, equipped with a photometries Quantix B/W camera. Image quantification was conducted with Image J 1.44 (http://rsb.info.nih.gov/ij/).

Quantification of cell fusion

Cells were seeded onto 48 plastic cell-well plate in triplicates, and were induced to fuse at 90% confluence with 5% horse-serum in DMEM. 6 days after fusion, cultures were subjected to immunofluorescence with anti-MHCl MF20 (Sigma- Aldrich). Cells were embedded in Citofluor (Agar Scientific) supplemented with DAPI. Semi-automatic quantitative analysis was carried out with the ArrayScan VTI HCS Reader (Cellomics, Thermo Scientific), using a co- localization protocol. Fusion index shows the ratio of nuclei in cells expressing MHCl to total number of nuclei in culture.

RNA immunoprecipitation (RIP)

RIP was performed using C2C12 myoblasts of shl21- or Hl-transduced and untransduced cultures. Proteins were extracted with lysis buffer (100 mM KC1, 5 mM MgC , 10 mM HEPES (pH 7.0), 0.5% NP40, 1 mM DTT, 80 U RNAse Inhibitor (Roche), Protease Inhibitor Cocktail (Roche)), and lysates were passed 5 times through a 29G needle and incubated for 10 minutes on ice. The lysates were then clarified by centrifugation at 16.000 rcf for 5 minutes at 4 °C and the supernatants were recovered. DNA was removed from the protein-RNA extracts by a DNAse I (Fermentas) treatment. Aliquots of protein extracts were used for

immunoprecipitation of PABPN1 using the VHH-3F5 antibody. Immunecomplexes were isolated with Protein A Sepharose beads (GE Healthcare) pre-coated with sperm-DNA. Following extensive washing with lysis buffer, RNA was isolated from the immunecomplexes using Trizol, and RT-qPCR was performed on mRNA aliquots. Equal protein loading and the expression levels of PABPN1 were determined with western blot analysis of protein extracts. Consistent and efficient IP between experiments was confirmed by western blot of IP protein yields. Example 2

Materials and Methods

Human and mouse materials

All human samples were reported in ⁿ. OPMD patients and pre-symptomatic individuals were genetically confirmed and underwent clinical investigation including MRC score prior to sampling of muscle biopsy and are reported in ¹⁹. All human muscle biopsies presented in this study were collected at Radboud Hospital, Nijmegen, Canisius-Wilhelmina Hospital, Nijmegen, The Netherlands, and

Rigshospitalet, Denmark, after an approval of the medical ethical committee

Arnhem-Nijmegen (CMO nr. 2005/189) and of the local ethical committee, from The NL and Denmark, respectively. All quadriceps biopsies were collected using the Bergstrom needle procedure. The biopsies froze immediately in liquid nitrogen and stored at -80 °C before RNA extraction. Mouse samples were previously reported ²⁰.

Cell culture lentivirus transduction and AON transfection

HeLa cells were grown in Dulbecco's modified Eagle's medium (DMEM) +

GlutamaxTM-I (GIBCO® Invitrogen) with 4,5 g/L glucose, supplemented with 10% fetal calf serum (FCS) (GIBCO® Invitrogen) and 100 U/ml antibiotics

(streptomycin, penicillin) in a humidified 5% C02-air atmosphere at 37 °C.

C2C12 immortalized mouse myoblasts were cultured in DMEM supplemented with 20% FCS and antibiotics. Treatments with 5μΜ cycloheximide were conducted for 10 hours in culture at 90% confluence. IM2 cultures are detailed in ¹¹. Cell growth and fusion conditions were performed as described in ¹⁷

The human 7304 immortalized myoblasts, generated by expressing telomerase (hTERT) and cyclin- dependent kinase 4²¹, were propagated in a medium

containing F-10 + 20% Fetal Calf Serum supplemented with an equal volume of Skeletal Muscle Cell Basal Medium supplemented with 5% FBS, 5 ng hEGF, and 200 ng Dexamethasone (PromoCell GmbH, Germany), at 37°C under 5% C02. 80- 90% confluent culture was induced to differentiate into myotubes in DMEM supplemented with 5% HS, 1% P/S, 1% Glutamax.

For lentivirus transduction, cells at about 60% confluence were transduced with 1- 2 MOI lentivirus particles in a medium supplemented with 0.1% polybrene. The virus was washed after 18 - 24 hours and was replaced by a normal growth medium. Puromycin selection (4 pg/ml) was applied to cells transduced with shRNA. The shRNA constructs were obtained from Sigma- Aldrich. shRNA for human ARIH2 (NM_006321) are: TRCN0000034269 (269); TRCN0000034272 (272); TRCN0000034273 (273). The shRNA for mouse ARIH2 (NM_011790) are: TRCN0000041023 (023); TRCN0000041073 (073). The shRNA for mouse PABPNl is: TRCN0000102536, and for human PABPNl: TRCN0000000121,

TRCN0000000122 and TRCN0000000123. Between brackets are the names for each sh-vector that are described here. The shRNA for PABPNl are described in ¹⁵ and in ⁿ. Lentiviruses for YFP-Alal6-PABPN1 or YFP-Alal0-PABPN1 were generated from the expression vectors described in ²². Lentivirus particles were produced as described in ²³. Four independent transduction experiments were performed for every virus. AON transfection

AON design was carried out as described in ²⁴, AON sequences are shown in Table 3. AON transfection was conducted as described in ¾ 26_ p_or RNA expression analysis 143 ng AON were transfected, and for fusion experiments 200 ng AON. AON in 0.15M NaCl were transfected into 7304 cells using 2.5 μΐ polyethylenimine (ExGen 500, MBI Fermentas) or TurboFect Transfection Reagent (Thermo

Scientific) per pg of AON. Fluorescent labeled control AON was used to monitor transfection efficiency. RNA was collected 48 hours after AON transfection, and cell fusion analysis was carried out in cultures 7 days post-transfection. Microarray and RT-qPCR analysis

ARIH2 and PABPNl expression levels were determined from the previously described ¹⁸. The microarray studies in used in this paper are publically available: in human quadriceps (GEO GSE26605), frontal cortex: GEO-GD707, GEO- GSE1572 Rectus abdominis GEO-GSE5086 ²⁸, blood GEO-GSE16717 ²⁹, and kidney cortex and medulla ³⁰. Statistical analyses of linear and quadratic models, in microarray studies and RT-qPCR were carried out with PASWStatistics 18.0 for Mac (IBM). Total RNA extraction from cell cultures and muscle biopsies for RT- qPCR was performed as described in ¹⁸. 3ng of purified cDNA served as template per RT-qPCR reactions. Detection of the PCR product was provided by iQ-SYBR Green (Bio-Rad). For calculations of fold change, normalization with housekeeping genes was followed by normalization with un-manipulated cells, in human with age-matched control groups, and in the aging study the age-group 17-25 years. The house keeping genes that were used here: GUSB and HRPT in human samples and biopsies and HRPT in C2C12 and mouse muscles are according to previous studies ¹¹ and ¹⁹. Fold change was determined with a primer set that covers that last axon and the 3'-UTR. The primer sets for each gene are listed in Table 2. Protein assays

Immunoprecipitation and western blot analyzes

Total proteins for direct western blots were extracted with a RIPA-buffer (20mM Tris pH 7.5, 150mM NaCl, 5mM EDTA, 1% NP40, 0.05% SDS), and for

immunoprecipitation experiments soluble proteins were extracted with a buffer composed of 20mM Tris pH 7.5, 10% glycerol, 150mM NaCl and 5mM EDTA. Protease inhibitor cocktail (SigmaFAST™ protease inhibitor tablets, Sigma- Aldrich®) was freshly added. Where indicated, 10 mM iodoacetamide was added to the extraction buffer. Soluble proteins were collected in the supernatant centrifugation. Insoluble proteins were recovered with RIPA buffer supplemented with 1% triton and 0.05% SDS, and ensuing sonification. Immunoprecipitation of PABPN1 was performed with VHH-3F5 was performed as described in ¹¹. Proteins were fractionated on SDS-PAGE and transferred onto polyvinylidene fluoride (PVDF) membrane. First antibodies used are: Goat anti-ARIH2 (Everest Biotech), mouse- anti-muscle actin (Santa Cruz Biotechnology®), mouse- anti-FLAG (Sigma- Aldrich®); mouse anti-Ubiquitin (FK2, Tebu-bio); mouse anti-tubulin (Sigma- Aldrich®); MyoD (M-318 Santa Cruz). The VHH-3F5 ³¹ intrabody was detected with rabbit anti-VHH. Antibodies were detected with the Odyssey® Infrared Imaging System (LI- COR® Biosciences) and applicable IRDye Secondary antibodies. Immunohistochemistry

Cells were seeded on glass cover-slides and transfected with either YFP-AlalO or YPF-Alal6 PABPN1 as described in ²². 48h after transduction cells were fixed with 2% formaldehyde, and permeabilized with 1% triton for 15 minutes.

Immunohistochemistry was performed with mouse anti-Ubiquitin (FK2) TeBu-Bio; goat anti-ARIH2, Everest Biotech; anti-Myosin MF20 (Sigma- Aldrich, MO, USA); anti-Desmin (1:500; Cell Signaling Technology, MS, USA) and the anti-PABPNl, 3F5 llama single chain antibody (1: 1000), recognized with rabbit-anti-VHH

(1:2000). Secondary antibodies were donkey-anti mouse Cy5; donkey anti-goat - alexa594. Slides were imbedded with Citifluor (Agar Scientific) after ethanol dehydration. Microscopic images were captured using the Leica DMRA

Quantification of cell fusion

Semi-automated quantification of cell fusion was carried out as in ⁿ. In brief:

aliquot of cells for 100% confluence were seeded onto 48 plastic cell-well plate in triplicates, and were induced to fuse with 5% horse-serum in DMEM few hours after seeding. 6 days after fusion, cultures were fixed and immunolabelled with anti-MHCl MF20 (Sigma- Aldrich). After incubation with secondary antibody, cells were embedded in Citofluor (Agar Scientific) supplemented with DAPI. Semiautomatic quantitative analysis was carried out with the ArrayScan VTI HCS Reader (Cellomics, Thermo Scientific), using the co-localization BioApplication V4. Fusion index is calculated as the percentage of nuclei in cells expressing MHC1 to total number of nuclei in culture.

RNA immunoprecipitation (RIP)

RIP was performed using C2C12 myoblasts of PABPN1-DR or Hl-transduced and un-transduced cultures as described in ¹⁵. Immunoprecipitation of PABPN1 was carried out with VHH-3F5. Immunecomplexes were isolated with Protein A

Sepharose beads (GE Healthcare) pre-coated with sperm-DNA. RNA isolated from RIP was subjected for RT-qPCR. Equal protein loading and the expression levels of PABPN1 were determined with western blot analysis of protein extracts.

Consistent and efficient IP between experiments was confirmed by western blot of IP protein yields. Results

ARIH2 mRNA levels decline in OPMD and in muscle aging and are highly associated with PABPN1 mRNA levels in muscles

Dysregulation of the UPS is among the most significant in models expressing expPABPNl ¹⁵> ¹⁸. Within the UPS E3-ligases, which determine substrate specificity for protein degradation, were highly dysregulated ¹⁸.

First we validated changes in expression levels of ARIH2 mRNA compared with age-matched control group in human Vastus lateralis muscles (VL) using RT- qPCR. A significant decrease in ARIH2 level was found in OPMD patients (age range 49-69) compared with an age-matched control group (Figure 11A). To assess whether a decline in ARIH2 expression is associated with symptoms or with expPABPNl expression, RT-qPCR was also performed on VL from expPABPNl carriers at a pre-symptomatic stage (age range 33-41). After normalization to a younger control age group (age 17-25) a significant decline in ARIH2 was found only in symptomatic but not in pre-symptomatic carriers. Symptoms in OPMD patients are predominantly restricted to skeletal muscles. In blood samples from OPMD and controls this decrease in ARIH2 levels could not be detected (Figure

IIB) . This suggests that a decline in ARIH2 is associated with symptoms and/or with age but not per se with expPABPNl expression.

Recently we reported that the expression of PABPN1 significantly decreases in

OPMD VL muscles and are significantly associated with symptoms and with age¹¹. We then investigated whether ARIH2 levels are also age-regulated and correlated with PABPN1 expression. RT-qPCR was performed on VL muscles from healthy controls age 17-89 and a significant decline in ARIH2 levels was found with a multiple-regression model with gender as a covariance (Figure 11C). The age- dependent trend in ARIH2 levels was highly similar to that of PABPN1 (Figure

IIC) . To evaluate the significance of age -associated decline in ARIH2, linear regression model was applied on two age groups. The earlier age where a significant decline in the expression of ARIH2 mRNA was found at 36 years, whereas below 36 years the levels of ARIH2 mRNA did not significantly change (Table 1). In this dataset a decline in PABPNl mRNA also started only from mid-life (Table 1). This suggests that the expression of both genes can mark age-regulated expression in skeletal muscles. The expression of both is highly correlated (p<0.005; Figure 11D), and confirmed in expPABPNl carriers at a pre-symptomatic stage but not in

symptomatic patients (Figure HE). This suggests that the tight correlation in expression of both genes is disrupted in OPMD.

Since decease in ARIH2 mRNA levels were found in in OPMD in affected muscles but not in blood, when then assessed whether a decline in ARIH2 levels is specific for aging- affected muscles or is also found in other tissues. An age-regulated expression of ARIH2 and PABPNl was examined in microarray studies from blood, kidney medulla, brain cortex, kidney cortex and Rectus abdominis skeletal muscles. Age-regulated expression of both ARIH2 and PABPNl could not be found in these tissues (Table 1). Of these, Rectus abdominis skeletal muscles have previously been reported to show minor aging-associated changes ³². Together, this suggests that expression of PABPNl and ARIH2 and associated and an age- regulated decline in both ARIH2 and PABPNl mRNA levels is specific to skeletal muscles.

ARIH2 E3-ligase binds to PABPNl and regulates PABPNl aggregation

To assess ARIH2 as a candidate PABPNl regulator co-localization

immunofluorescence studies were conducted in HeLa cell expressing PABPN1- YFP. ARIH2 protein was found to be nuclear localized and co-localized with PABPNl aggregates (Figure 12). Importantly, together with ARIH2, ubiquitin co- localization was also found (Figure 12).

We then investigated whether ARIH2 binds to PABPNl and regulates PABPNl aggregation. ARIH2 expression was stably down-regulated (ARIH2-DR) in HeLa cells by RNA interference using two independent shRNA lentiviruses. We chose HeLa cells as a cell model since PABPNl aggregation is highly reproducible and quantifiable ²²> ³³. A significant reduction in ARIH2 RNA and protein levels was obtained with two shRNA to ARIH2 (Figure 13A). An empty vector (HI) and an unspecific shRNA were used as negative controls (Figure 13A).

Co-immunoprecipitation (IP) experiments suggest that ARIH2 bind to PABPNl (Figure 13B). Reduced ARIH2 IP and PABPNl co-IP were found in the ARIH2-DR cell cultures (Figure 13B), suggesting that ARIH2 binding to PABPNl is specific. Previously we reported that PABPNl protein accumulation with regulated by the UPS ¹⁷. Therefore, we next investigated whether PABPNl poly-ubiquitination is regulated by ARIH2. Proteins were extracted in the presence of iodoacetamide, which inhibits deubiqutination, and the ubiquitinated PABPNl was visualized with anti-ubiquitin antibodies in the IP fraction. While ubiquitinated PABPNl was found in control cells (Figure 13B, indicated with a narrow), in the ARIH2-DR cell cultures ubiquitinated PABPNl was undetectable. This indicates that ARIH2 regulates PABPNl ubiquitination. As the wild type (WT) PABPNl ubiquitination is higher compared with expAPBPNl, and is associated with a reduced protein accumulation ¹¹ , we then investigated whether ARIH2 co-IP with PABPNl differs between WT and expPABPNl. Control (HI) and ARIH2-DR cell cultures were transiently transduced with WT (AlalO) or alanine-expanded (Alal6) PABPNl fused to yellow-fluorescent protein (YFP). PABPNl IP was carried out two days after transduction and Co-IP with ARIH2 was determined in immunoblots.

Loading controls and PABPNl transgenes are shown in the input fraction (Figure 13C). Quantification of the ratio of PABPN1-YFP to ARIH2 in the IP fraction reveals that ARIH2 preferentially co-IP with WT PABPNl whereas co-IP with expPABPNl was less efficient (Figure 13C). ARIH2 co-IP with PABPNl is severely affected in ARIH2-DR cultures (Figure 13C). A preferential association of ARIH2 with WT PABPNl was also found in co-localization experiments. Focusing on the small PABPNl fluorescence foci, co-localization was stronger in YFP-AlalO transfected cells, compared with cells expressing YFP-Alal6, and this co- localization reduced in ARIH2-DR cells (Figure 13D). Consistently, ARIH2-DR had a clearer effect on ARIH2 co-localization with WT-PABPN1 compared with expPABPNl (Figure 13Di). Importantly, down-regulation of ARIH2 causes a significant increase in the proportion of nuclei that contain PABPNl aggregates (Figure 13Dii). Consistent with the co-IP and co-localization results, ARIH2-DR has smaller effect on expPABPNl aggregation compared with the effect on WT- PABPNl aggregation. Together, these experiments indicate that ARIH2 binds to PABPNl and regulates its ubiquitination and aggregation, with a preference for WT-PABPN1 over expPABPNl. ARIH2 regulates PABPNl protein turnover and affects muscle cell fusion

Next we also examined binding of ARIH2 to PABPNl in muscle myotube cultures. In this cell model cell model AlalO-PABPNl or Alal7-PABPN1 are fused to the FLAG tag and are stably expressed in fused cultures ¹⁷. PABPNl was

immunoprecipitated with the VHH-3F5 and FLAG-fused protein product was detected with anti-FLAG antibodies. The specificity of IP is demonstrated in the control IM2 cells, which do not express PABPN1-FLAG (Figure 14). ARIH2 co-IP was found in AlalO-PABPNl IP whereas its co-IP was Alal7-PABPN1 was noticeably reduced (Figure 14). ARIH2 was found in co-IP with soluble PABPNl but not with insoluble PABPNl (Figure 14). This suggests that ARIH2 binds to the functional form of PABPN 1.

To investigate whether ARIH2 regulates PABPNl protein accumulation, stable ARIH2-DR C2C12 cultures were generated using two shRNAs to mouse ARIH2. Both shRNAs led to 50% and 40% down-regulation, respectively, which was confirmed on RNA and protein level (Figure 15A). The shRNA 273, whose sequence is specific to human ARIH2, was used as a negative control (Figure 15A). A preferential binding of ARIH2 with WT-PABPN1 in this muscle cell model was found in ARIH2-DR cultures that were transduced with YFP-AlalO or YFP-Alal6 (Figure 15B). This conclusions is consistent with that found in HeLa cells.

Therefore, we next investigated whether PABPNl protein accumulation is regulated by ARIH2. We examined the accumulation of endogenous PABPNl since overexpression of PABPNl causes aggregation. After 6 hours cyclohexamide (CHX) treatment PABPNl protein accumulation significantly reduced in control cells but not in ARIH2-DR muscle cells (Figure 15C). This indicates that indeed ARIH2 regulates levels of PABPNl protein accumulation.

To determine a cellular impact of ARIH2-DR in muscle cells we determined muscle cell fusion using a semi-automatic image quantification protocol that calculates the proportion of nuclei in cells expressing myosin heavy chain 1 (MHC1) (Figure 15D). With this robust procedure (N >20,000 nuclei) we observed that ARIH2-DR also causes a significant decrease in fusion index (Figure 15D).

In these ARIH2-DR muscle cell cultures PABPNl expression was significantly reduced (Figure 15A). PABPNl down— regulation in muscle cell culture causes reduced fusion index as well as a decrease in cell growth, expression of muscle fusion genes, mitochondrial metabolic rate and an accumulation of fat bodies ⁿ. This suggests that ARIH2 and PABPNl regulation of cell myogenesis are interconnected. Therefore, we then investigated whether PABPNl directly regulate mRNA levels of ARIH2.

PAPBN1 levels regulate the PAS choice in ARIH2 transcripts

First we determined whether ARIH2 mRNA specifically co-IP with PABPNl protein using PABPN1-DR cultures. In these cultures ARIH2 mRNA levels were also significantly reduced (Figure 16, input). RNA-immunoprecipitation (RIP) with antibodies to PABPNl was performed in C2C12 cell cultures, and mRNA bound to PABPNl was determined with RT-qPCR. RIP reveals that ARIH2 mRNA was IP with PABPNl protein, and binding was enriched in PABPN1-DR cells (Figure 16). Interestingly, PABPNl mRNA was IP with PABPNl protein but was decreased in IP from PABPNl -DR cells. Importantly, HRPT mRNA, which was used as a housekeeping gene in the input fraction, was not found in RIP fraction. This suggests that not all transcripts bind to PABPNl. This experiment suggests that PABPNl directly regulates ARiH2 mRNA levels. Moreover, PABPNl could regulate its expression levels, but in a mechanism that differ from ARIH2 regulation.

Others and we demonstrated that PABPNl regulates PAS usage ¹⁴> ¹⁵. In humans and mice, the 3'-UTR in ARIH2 contains proximal and distal PAS, leading to short or long ARIH2 transcripts, respectively (Figure 17A). A change in PAS usage was determined from the ratio between long and short 3'-UTR (Figure 17A).

Overexpression of expPAPBNl in the A17.1 mouse model causes a preference for the proximal over the distal PAS (Figure 17B and ¹⁵). A change in the choice of PAS was validated with RT-qPCR using primers in ARIH2 3'-UTR that detect long or short transcripts. In the A17.1 mice reduced ARIH2 mRNA level was associated with a decrease in the ratio between long and short ARIH2 transcripts (Figure 17C). Also in PABPN1-DR C2C12 cells the ratio long to short ARIH2 3'-UTR transcripts significantly decreased (Figure 17D, input). Together these results indicate that a preference in proximal PAS usage in ARIH2 mRNA causes reduced expression levels.

Next we investigated whether a change in ARIH2 PAS usage is a direct result of ARIH2 mRNA binding to PABPNl using the RIP procedure. The ratio long:short ARIH2 mRNA significantly decreased in PABPNl -DR cells (Figure 17D, RIP). Background protein-RNA IP was determined by incubating the extract to the beads (Figure 17D; mock). Together, these experiments indicate that reduced levels of ARIH2 are regulated by PABPNl via a PAS usage.

Masking ARIH2 proximal PAS elevates expression levels of ARIH2 and PABPNl and restores cell fusion in PABPNl-DR cultures

To further demonstrate that PAS choice causes a change in ARIH2 expression we blocked proximal PAS usage by expressing antisense oligonucleotides (AON) specific to the ARIH2 proximal PAS (Table 3). An unspecific FITC-conjugated AON was used as a control for transfection efficiency and a negative control for evaluation of the efficiency of the AONs to the proximal PAS of ARIH2. AON transfection experiments were performed in 7304.1 (named here 7304)

immortalized myoblasts ³⁴, as transfection efficiency and cell fusion are both higher in those cells compared with C2C12. AON transfection was studied in both control and PABPNl-DR myoblast cell cultures. 7304 PABPNl-DR cells were generated after stable expression of the shl21 lentivirus ⁿ. As in C2C12 cells, also in 7304 cultures PABPNl-DR caused reduced ARIH2 expression (Figure 18A). A

significant increase in ARIH2 mRNA expression in PABPNl-DR cells was found 3 days after transfection of AON to proximal ARIH2, compared with control AON (Figure 18B). As we found in vivo that ARIH2 and PABPNl expression are highly correlated we also determined PABPNl expression after AON transfection. Similar to ARIH2, levels of PABPNl were also elevated after transfection with AONs to the proximal PAS of ARIH2 (Figure 18B). Importantly, the increase in both ARIH2 and PABPNl mRNA levels was more pronounced in PABPNl-DR cell cultures (Figure 18B). Based on higher binding energy for AON1 compared with AON2 (Table 3), we predicted a better effect for AON1 ²⁴. Indeed, the effect of AON1 on ARIH2 or PABPNl expression levels was more pronounced compared with AON2 (Figure 18B).

Next, a functional impact for masking of ARIH2 proximal PAS on muscle cell fusion was determined. Reduced cell fusion, which is induced by PABPNl-DR, was maintained in cultures after transfection with control AON (Figure 18C). A small but significant increase in fusion index was found after transfection with AON1 or 2 to the proximal PAS of ARIH2 into PABPNl-DR cells (Figure 18C). After ARIH2 AON transfection fusion index did not differ between control and PABPNl-DR cultures. This suggests that AON to proximal ARIH2 PAS restores the PABPN1- induced fusion defects. Consistent with an increased expression of PABPNl after AON1 transfection, the effect on cell fusion was also higher. This suggests that AON1 is more potent in masking ARIH2 proximal PAS compared with AON2.

Discussion

Aging is a multi-factorial complex process, where the control of multiple cellular processes progressively loosens as age raises. Among those, genome-wide changes in RNA expression are found in aging and degenerated muscles ³⁵> ³⁶. This age- regulated mRNA expression profiles are regulated, in part, by PABPNl levels ⁿ. PABPNl is a multi-functional orchestrator of mRNA processing regulating different steps in mRNA processing 12. Reduced PABPNl levels causes proximal over distal PAS usage in the 3'-UTR, affecting mRNA abandnece ¹⁴> ¹⁵. Levels of PABPNl decline during muscle aging, and in vitro it induces myogenic defects ⁿ. Here we identified ARIH2-E3 ligase as a regulator of PABPNl protein

accumulation and aggregation. In addition, we demonstrate that ARIH2 mRNA levels are regulated by PABPNl. Our studies here suggest that together ARIH2 and PABPNl coordinate muscle cell function via a feed-forward loop between protein turnover and mRNA stability (Figure 19). Therefore the decline in PABPNl during muscle aging is likely to be regulated by an age-regulated decline in ARIH2. In turn a decrease in ARIH2 mRNA levels is regulated by PABPNl (Figure 19). A cytosolic positive feedback pathway regulating the aggregation of oc-synuclein in Parkinson disease ³⁷ , supports recent concepts for how progression is derived in protein aggregation disorders ³⁸. In addition, we found that PABPNl protein binds to its own mRNA and therefore could self regulate its expression. In

polyadenylation RNA-seq studies, PABPNl levels were not found to be affected by PAS usage ¹⁵. In OPMD muscles, a decrease in expression of both ARIH2 and PABPNl is exaggerated compared with age-matched control group. In vitro, reduced ARIH2 induces an increase in PABPNl aggregation. In models with expPAPBNl expression, levels of soluble PABPNl decrease ¹⁷ and proximal PAS usage increases ¹⁴> ¹⁵. Aggregated PABPNl is associated with reduced levels of soluble PABPNl ¹¹ , and likewise levels of proteins that are entrapped in PABPNl aggregates would deplete ²²> ³⁹. We found that ARIH2 co-localizes with PABPNl aggregates. Therefore the exaggerated decrease in PABPNl and ARIH2 expression in OPMD, could be explained by a combination of age-regulated decline in expression and depletion of functional proteins due to entrapment in insoluble nuclear aggregates (Figure 19). Consequently, this would cause in OPMD greater mRNA expression dysregulation and more severe muscle degeneration (Figure 19). The UPS, as a cellular regulator of protein homeostasis has also been implicated as a major contributor to a decline in cellular homeostasis during aging and in late- onset protein aggregation neurodegenerative disorders ⁴⁰. Aging-regulated dysregulation is suggested to accelerate protein aggregation in a feed-forward loop ⁴¹> ⁴². In aging muscles proteasome activity declines ⁴³, and several E3-ligases, like Murfl and Atrogin- 1 have been identified as key regulators of muscle atrophy in animal disuse models ⁴⁴. Expression dysregulation those genes are found in atrophic and disuse muscles ¹⁸> ²⁰> ⁴⁵. ARIH2 E3-ligase contains a ring between ring fingers (RBR) domain, and is part of the largest family of E3-ligases. RBR E3- ligases have attracted interest because of their involvement in late-onset protein aggregation disorders such as Parkinson' disease, Lewy body dementia, and Alzheimer's disease ⁴⁶. Here we show that ARIH2 regulates PABPNl

ubiquitination, protein accumulation and aggregation in myogenic cells. These RBR E3-ligases, as regulators of aggregation-prone protein accumulation, could be a collective target for therapy of protein aggregation disorders.

The significance of the proximal PAS usage in ARIH2 on its expression level was demonstrated in this study by masking the proximal PAS with specific AONs, which elevated ARIH2 and PABPN1 levels and restored myogenesis in PABPN1- DR cells. Antisense oligonucleotides have been widely demonstrated to be an efficient molecular tool to modulate RNA processing due to their small size, stability and high efficiency delivering into the nucleus ⁴¹. In Duchenne patients AONs application for exon-skipping is progressed as therapeutic treatment ⁴⁸. In addition, AON treatment can also redirect PAS selection ⁴⁹. Our results here demonstrate that manipulation of expression level by AONs to proximal PAS also cause a functional impact on cell fusion. This opens a therapeutic opportunity to restore expression levels of regulatory genes whose levels significantly change by PAS usage, and possibly could be a clinical strategy for the treatment of diseases of RNA metabolism.

References for example 2 1. Lecker SH, Goldberg AL, Mitch WE: Protein degradation by the ubiquitin- proteasome pathway in normal and disease states, J Am Soc Nephrol 2006, 17: 1807- 1819

2. Sandri M: Signaling in muscle atrophy and hypertrophy, Physiology

(Bethesda) 2008, 23: 160-170

3. Lindner AB, Demarez A: Protein aggregation as a paradigm of aging,

Biochimica et Biophysica Acta (BBA) - General Subjects 2009, 1790:980-996

4. David DC, Ollikainen N, Trinidad JC, Cary MP, Burlingame AL, Kenyon C:

Widespread protein aggregation as an inherent part of aging in C. elegans, PLoS biology 2010, 8:e 1000450

5. Bingol B, Sheng M: Deconstruction for reconstruction: the role of proteolysis in neural plasticity and disease, Neuron 2011, 69:22-32

6. Brais B: Oculopharyngeal muscular dystrophy: a polyalanine myopathy, Current neurology and neuroscience reports 2009, 9:76-82

7. Brais B, Bouchard JP, Xie YG, Rochefort DL, Chretien N, Tome FM, Lafreniere RG, Rommens JM, Uyama E, Nohira O, Blumen S, Korczyn AD,

Heutink P, Mathieu J, Duranceau A, Codere F, Fardeau M, Rouleau GA: Short GCG expansions in the PABP2 gene cause oculopharyngeal muscular dystrophy, Nature genetics 1998, 18: 164-167 8. Tome FM, Fardeau M: Nuclear inclusions in oculopharyngeal dystrophy, Acta neuropathologica 1980, 49:85-87

9. Raz V, Butler-Browne G, van Engelen B, Brais B: 191st ENMC

International Workshop: Recent advances in oculopharyngeal muscular dystrophy research: From bench to bedside 8-10 June 2012, Naarden, The Netherlands, Neuromuscular disorders : NMD 2013, 23:516-523

10. Berciano MT, Villagra NT, Ojeda JL, Navascues J, Gomes A, Lafarga M, Carmo-Fonseca M: Oculopharyngeal muscular dystrophy-like nuclear inclusions are present in normal magnocellular neurosecretory neurons of the hypothalamus, Human molecular genetics 2004, 13:829-838

11. Anvar SY, Raz Y, Verway N, van der Sluijs B, Venema A, Goeman JJ, Vissing J, van der Maarel SM, t Hoen PA, van Engelen BG, Raz V: A decline in PABPN1 induces progressive muscle weakness in Oculopharyngeal muscle dystrophy and in muscle aging, Aging (Albany NY) 2013, 5:412-426

12. Simonelig M: PABPN1 shuts down alternative poly(A) sites, Cell Res 2012, 22: 1419- 1421

13. Benoit B, Mitou G, Chartier A, Temme C, Zaessinger S, Wahle E, Busseau I, Simonelig M: An essential cytoplasmic function for the nuclear poly(A) binding protein, PABP2, in poly(A) tail length control and early development in Drosophila, Dev Cell 2005, 9:511-522

14. Jenal M, Elkon R, Loayza-Puch F, van †Haaften G, KV°hn U, Menzies FM, Vrielink JAFO, Bos AJ, Drost J, Rooijers K, Rubinsztein DC, Agami R: The

Poly(A) -Binding Protein Nuclear 1 Suppresses Alternative Cleavage and

Polyadenylation Sites, Cell 2012, 149:538-553

15. de Klerk E, Venema A, Anvar SY, Goeman JJ, Hu O, den Dunnen JT, van der Maarel SrM, Raz V, ,Aot Hoen PAC: Poly(A) binding protein nuclear 1 levels affect alternative polyadenylation, Nucleic acids research 2012, 40:9089-9101

16. Beaulieu YB, Kleinman CL, Landry-Voyer AM, Majewski J, Bachand F: Polyadenylation-dependent control of long noncoding RNA expression by the poly(a)-binding protein nuclear 1, PLoS Genet 2012, 8: 15

17. Raz V, Routledge S, Venema A, Buijze H, van der Wal E, Anvar S,

Straasheijm KR, Klooster R, Antoniou M, van der Maarel SM: Modeling

oculopharyngeal muscular dystrophy in myotube cultures reveals reduced accumulation of soluble mutant PABPN1 protein, Am J Pathol 2011, 179: 1988- 2000

18. Anvar SY, t Hoen PA, Venema A, van der Sluijs B, van Engelen B, Snoeck M, Vissing J, Trollet C, Dickson G, Chartier A, Simonelig M, van Ommen GJ, van der Maarel SM, Raz V: Deregulation of the ubiquitin-proteasome system is the predominant molecular pathology in OPMD animal models and patients, Skeletal muscle 2011, 1: 15

19. Raz V, Sterrenburg E, Routledge S, Venema A, van der Sluijs BM, Trollet C, Dickson G, van Engelen BG, van der Maarel SM, Antoniou MN: Nuclear

entrapment and extracellular depletion of PCOLCE is associated with muscle degeneration in oculopharyngeal muscular dystrophy, BMC Neurol 2013, 13: 1471- 2377

20. Trollet C, Anvar SY, Venema A, Hargreaves IP, Foster K, Vignaud A, Ferry A, Negroni E, Hourde C, Baraibar MA, t Hoen PA, Davies JE, Rubinsztein DC, Heales SJ, Mouly V, van der Maarel SM, Butler-Browne G, Raz V, Dickson G: Molecular and phenotypic characterization of a mouse model of oculopharyngeal muscular dystrophy reveals severe muscular atrophy restricted to fast glycolytic fibres, Human molecular genetics 2010, 19:2191-2207

21. Zhu CH, Mouly V, Cooper RN, Mamchaoui K, Bigot A, Shay JW, Di Santo JP, Butler-Browne GS, Wright WE: Cellular senescence in human myoblasts is overcome by human telomerase reverse transcriptase and cyclin-dependent kinase 4: consequences in aging muscle and therapeutic strategies for muscular

dystrophies, Aging cell 2007, 6:515-523

22. Raz V, Abraham T, van Zwet EW, Dirks RW, Tanke HJ, van der Maarel SM: Reversible aggregation of PABPN1 pre-inclusion structures, Nucleus 2011,

2:208-218

23. Raz V, Carlotti F, Vermolen BJ, van der Poel E, Sloos WC, Knaan-Shanzer S, de Vries AA, Hoeben RC, Young IT, Tanke HJ, Garini Y, Dirks RW: Changes in lamina structure are followed by spatial reorganization of heterochromatic regions in caspase-8-activated human mesenchymal stem cells, Journal of cell science 2006, 119:4247-4256

24. Aartsma-Rus A, van Vliet L, Hirschi M, Janson AA, Heemskerk H, de Winter CL, de Kimpe S, van Deutekom JC, t Hoen PA, van Ommen GJ: Guidelines for antisense oligonucleotide design and insight into splice-modulating

mechanisms, Mol Ther 2009, 17:548-553

25. Aartsma-Rus A: Overview on DMD exon skipping, Methods Mol Biol 2012, 867:97-116

26. Aartsma-Rus A, Kaman WE, Bremmer-Bout M, Janson AA, den Dunnen JT, van Ommen GJ, van Deutekom JC: Comparative analysis of antisense

oligonucleotide analogs for targeted DMD exon 46 skipping in muscle cells, Gene Ther 2004, 11: 1391- 1398

27. Lu T, Pan Y, Kao SY, Li C, Kohane I, Chan J, Yankner BA: Gene regulation and DNA damage in the ageing human brain, Nature 2004, 429:883-891

28. Zahn JM, Sonu R, Vogel H, Crane E, Mazan-Mamczarz K, Rabkin R, Davis RW, Becker KG, Owen AB, Kim SK: Transcriptional profiling of aging in human muscle reveals a common aging signature, PLoS Genet 2006, 2:ell5

29. Passtoors WM, Boer JM, Goeman JJ, Akker EB, Deelen J, Zwaan BJ, Scarborough A, Breggen R, Vossen RH, Houwing-Duistermaat JJ, Ommen GJ,

Westendorp RG, Heemst D, Craen AJ, White AJ, Gunn DA, Beekman M, Slagboom PE: Transcriptional profiling of human familial longevity indicates a role for ASF1A and IL7R, PloS one 2012, 7:e27759

30. Rodwell GE, Sonu R, Zahn JM, Lund J, Wilhelmy J, Wang L, Xiao W, Mindrinos M, Crane E, Segal E, Myers BD, Brooks JD, Davis RW, Higgins J, Owen AB, Kim SK: A transcriptional profile of aging in the human kidney, PLoS biology 2004, 2:e427

31. Verheesen P, de Kluijver A, van Koningsbruggen S, de Brij M, de Haard HJ, van Ommen G-JB, van der Maarel SrM, Verrips CT: Prevention of oculopharyngeal muscular dystrophy- associated aggregation of nuclear poly (A) -binding protein with a single-domain intracellular antibody, Human molecular genetics 2006, 15: 105- 111

32. Marzani B, Felzani G, Bellomo RG, Vecchiet J, Marzatico F: Human muscle aging: ROS-mediated alterations in rectus abdominis and vastus lateralis muscles, Exp Gerontol 2005, 40:959-965

33. Abu-Baker A, Messaed C, Laganiere J, Caspar C, Brais B, Rouleau GA: Involvement of the ubiquitin-proteasome pathway and molecular chaperones in oculopharyngeal muscular dystrophy, Human molecular genetics 2003, 12:2609- 2623

34. Perie S, Mamchaoui K, Mouly V, Blot S, Bouazza B, Thornell L-E, St Guily JL, Butler-Browne G: Premature proliferative arrest of cricopharyngeal myoblasts in oculo-pharyngeal muscular dystrophy: Therapeutic perspectives of autologous myoblast transplantation, Neuromuscular disorders : NMD 2006, 16:770-781

35. Zahn JM, Sonu R, Vogel H, Crane E, Mazan-Mamczarz K, Rabkin R, Davis RW, Becker KG, Owen AB, Kim SK: Transcriptional profiling of aging in human muscle reveals a common aging signature, PLoS Genet 2006, 2:ell5

36. Zhan M, Yamaza H, Sun Y, Sinclair J, Li H, Zou S: Temporal and spatial transcriptional profiles of aging in Drosophila melanogaster, Genome research 2007, 17: 1236-1243

37. Mazzulli Joseph R, Xu Y-H, Sun Y, Knight Adam L, McLean Pamela J, Caldwell Guy A, Sidransky E, Grabowski Gregory A, Krainc D: Gaucher Disease Glucocerebrosidase and a-Synuclein Form a Bidirectional Pathogenic Loop in Synucleinopathies, Cell 2011, 146:37-52

38. George JL, Mok S, Moses D, Wilkins S, Bush AI, Cherny RA, Finkelstein DI: Targeting the progression of Parkinson's disease, Current neuropharmacology 2009, 7:9-36

39. Calado A, Tome FM, Brais B, Rouleau GA, Kuhn U, Wahle E, Carmo-

Fonseca M: Nuclear inclusions in oculopharyngeal muscular dystrophy consist of poly(A) binding protein 2 aggregates which sequester poly(A) RNA, Human molecular genetics 2000, 9:2321-2328

40. Ciechanover A, Brundin P: The ubiquitin proteasome system in

neurodegenerative diseases: sometimes the chicken, sometimes the egg, Neuron 2003, 40:427-446

41. Bennett EJ, Bence NF, Jayakumar R, Kopito RR: Global impairment of the ubiquitin-proteasome system by nuclear or cytoplasmic protein aggregates precedes inclusion body formation, Molecular cell 2005, 17:351-365

42. Bence NF, Sampat RM, Kopito RR: Impairment of the ubiquitin-proteasome system by protein aggregation, Science 2001, 292: 1552- 1555 43. Ferrington DA, Husom AD, Thompson LV: Altered proteasome structure, function, and oxidation in aged muscle, FASEB journal : official publication of the Federation of American Societies for Experimental Biology 2005, 19:644-646

44. Sandri M, Sandri C, Gilbert A, Skurk C, Calabria E, Picard A, Walsh K, Schiaffino S, Lecker SH, Goldberg AL: Foxo Transcription Factors Induce the

Atrophy-Related Ubiquitin Ligase Atrogin- 1 and Cause Skeletal Muscle Atrophy, Cell 2004, 117:399-412

45. Edstrom E, Altun M, Hagglund M, Ulfhake B: Atrogin-l/MAFbx and MuRFl Are Downregulated in Aging-Related Loss of Skeletal Muscle, The

Journals of Gerontology Series A: Biological Sciences and Medical Sciences 2006, 61:663-674

46. Eisenhaber B, Chumak N, Eisenhaber F, Hauser M-T: The ring between ring fingers (RBR) protein family, Genome biology 2007, 8:209

47. Bennett CF, Swayze EE: RNA targeting therapeutics: molecular

mechanisms of antisense oligonucleotides as a therapeutic platform, Annu Rev Pharmacol Toxicol 2010, 50:259-293

48. Goemans NM, Tulinius M, van den Akker JT, Burm BE, Ekhart PF, Heuvelmans N, Holling T, Janson AA, Platenburg GJ, Sipkens JA, Sitsen JMA, Aartsma-Rus A, van Ommen G-JB, Buyse G, Darin N, Verschuuren JJ, Campion GV, de Kimpe SJ, van Deutekom JC: Systemic Administration of PRO051 in Duchenne's Muscular Dystrophy, New England Journal of Medicine 2011,

364: 1513-1522

49. Vickers TA, Wyatt JR, Burckin T, Bennett CF, Freier SM: Fully modified 2' MOE oligonucleotides redirect polyadenylation, Nucleic acids research 2001, 29: 1293- 1299

Table 1

ARIH2 PABPN1

Age-range Beta ±SE P-value Beta ±SE P-value

-0.017 ±0.005 -0.029 ±0.006

0.002 <0.0001

Vastus lateralis muscles 74 17-89 (36-89; N=47) (43-89; N=40)

-0.002 ±0.002 -0.006 ±0.009

0.25 0.37

(17-35; N=27) (17-42; N=34)

0.001 ±0.002 0.72 -0.000 ±0.003 0.94

Rectus abdominis muscles 62 24-83 0.002 ±0.003 0.39 0.010 ±0.007 0.13

-0.001 ±0.008 0.89 0.001 ±0.003 0.64

Frontal brain cortex 30 26-95 0.000 ±0.002 0.90 0.002 ±0.007 0.73

-0.001 ±0.001 0.68 -0.001 ±0.002 0.76

Kidney cortex 72 27-92 0.000 ±0.001 0.76 0.001 ±0.002

0.000 ±0.001 0.64 0.003 ±0.002

-0.001 ±0.001 0.40 -0.003 ±0.002 0.11

Kidney medulla 61 29-92 -0.003 ±0.002 0.10 0.001 ±0.002 0.76

0.001 ±0.001 0.48 -0.004 ±0.002 0.06

-0.001 ±0.005 0.88 0.001 ±0.003 0.69

Blood 150 42- 102 0.001 ±0.001 0.15 -0.003 ±0.003 0.39

-0.001 ±0.002 0.45 0.003 ±0.002 0.11

Betas ±standard errors of the mean are per probe and show an age- association linear model. Values for three independent probes shown for datasets from Kidney cortex, Kidney medulla, Rectus Abdominis. P-values are adjusted for gender, except in kidney datasets and are corrected for false discovery rate. Significant changes are highlighted in bold. N indicates number of samples. Ag range is indicates in years. In Vastus lateralis muscles a linear regression model was applied on two age groups; age-range and number of samples are indicated between brackets.

Table 2

Gene Primer sequence

Human/mouse PABPN1 FW 5' CGTTGGCAATGTGGACTATG 3'

RV 5' ACACGGTTGACTGAACCACA 3' human ARIH2 FW 5' GAATAGCCAGGGGTCTGACA 3'

RV 5' CTGGTACTCCTCGGGATCAA 3'

GUSB FW 5' CTCATTTGGAATTTTGCCGATT 3'

RV 5' CCGAGTGAAGATCCCCTTTTTA 3'

GapDH FW 5' CAAC GAATTT G G CTAC AGC A 3'

RV 5' AGGGGTCTACATGGCAACTG 3'

ARIH2 3'-UTR FW 5' TTCTGCAACTTTGCTGGATG 3'

RV 5' CTGCTCCACATCACTGGCTA 3'

Mouse ARIH2 exon 11/12 FW 5' TGGAGGCTGAGATCGAAAAC 3'

RV 5' GCTGCTCTGCTATGTGCATC 3'

Hprt FW 5' CGTCGTGATTAGCGATGATG 3'

RV 5' TTTTCCAAATCCTCGGCATA 3'

Table 3

Sequence Binding energy Melting temperature Control AON 5' 2'0Me(CUU GAG CUU AUU UUC AAG UUU) 3' - 50

ARIH2 AON 1 5' 2'0Me(GUA UAA UUG UAC AAC CUU UGA AAG) 3' 27 49

ARIH2 AON 2 5' 2'0Me(GUA UAA UUG UAC AAC CUU UGA A) 3' 23 46

CITED ART (except for example 2)

Abu-Baker A, Messaed C, Laganiere J, Caspar C, Brais B , Rouleau GA (2003). Involvement of the ubiquitin-proteasome pathway and molecular chaperones in oculopharyngeal muscular dystrophy. Hum Mol Genet. 12, 2609-2623.

Anvar SY, t Hoen PA, Venema A, van der Sluijs B, van Engelen B, Snoeck M, Vissing J, Trollet C, Dickson G, Chartier A, Simonelig M, van Ommen GJ, van der Maarel SM , Raz V (2011a). Deregulation of the ubiquitin-proteasome system is the predominant molecular pathology in OPMD animal models and patients. Skelet Muscle. 1, 15.

Anvar SY, Tucker A, Vinciotti V, Venema A, van Ommen GJ, van der Maarel SM, Raz V , t Hoen PA (2011b). Interspecies translation of disease networks increases robustness and predictive accuracy. PLoS Comput Biol. 7, e 1002258.

Anvar et al (2013). A decline in PABPN1 induces progressive muscle weakness in Oculopharyngeal muscle aging. Aging Vol 5, pp 412. Apponi LH, Leung SW, Williams KR, Valentini SR, Corbett AH , Pavlath GK

(2010). Loss of nuclear poly(A)-binding protein 1 causes defects in myogenesis and mRNA biogenesis. Hum Mol Genet. 19, 1058- 1065.

Beenakker KG, Ling CH, Meskers CG, de Craen AJ, Stijnen T, Westendorp RG , Maier AB (2010). Patterns of muscle strength loss with age in the general population and patients with a chronic inflammatory state. Ageing Res Rev. 9, 431- 436.

Ben-Zvi A, Miller EA , Morimoto RI (2009). Collapse of proteostasis represents an early molecular event in Caenorhabditis elegans aging. Proc Natl Acad Sci U S A. 106, 14914-14919. Berciano MT, Villagra NT, Ojeda JL, Navascues J, Gomes A, Lafarga M , Carmo- Fonseca M (2004). Oculopharyngeal muscular dystrophy- like nuclear inclusions are present in normal magnocellular neurosecretory neurons of the hypothalamus. Hum Mol Genet. 13, 829-838.

Bingol B , Sheng M (2011). Deconstruction for reconstruction: the role of proteolysis in neural plasticity and disease. Neuron. 69, 22-32.

Brais B (2009). Oculopharyngeal muscular dystrophy: a polyalanine myopathy. Curr Neurol Neurosci Rep. 9, 76-82.

Brais B, Bouchard JP, Xie YG, Rochefort DL, Chretien N, Tome FM, Lafreniere RG, Rommens JM, Uyama E, Nohira O, Blumen S, Korczyn AD, Heutink P, Mathieu J, Duranceau A, Codere F, Fardeau M , Rouleau GA (1998). Short GCG expansions in the PABP2 gene cause oculopharyngeal muscular dystrophy. Nat Genet. 18, 164- 167.

Calado A, Tome FM, Brais B, Rouleau GA, Kuhn U, Wahle E , Carmo-Fonseca M (2000). Nuclear inclusions in oculopharyngeal muscular dystrophy consist of poly(A) binding protein 2 aggregates which sequester poly(A) RNA. Hum Mol Genet. 9, 2321-2328.

Corbeil-Girard LP, Klein AF, Sasseville AM, Lavoie H, Dicaire MJ, Saint-Denis A, Page M, Duranceau A, Codere F, Bouchard JP, Karpati G, Rouleau GA, Massie B, Langelier Y , Brais B (2005). PABPN1 overexpression leads to up regulation of genes encoding nuclear proteins that are sequestered in oculopharyngeal muscular dystrophy nuclear inclusions. Neurobiol Dis. 18, 551-567.

David DC, Ollikainen N, Trinidad JC, Cary MP, Burlingame AL , Kenyon C (2010). Widespread protein aggregation as an inherent part of aging in C. elegans. PLoS Biol. 8, e 1000450. de Klerk E, Venema A, Anvar SY, Goeman JJ, Hu O, den Dunnen JT, van der Maarel SrM, Raz V , ,Aot Hoen PAC (2012). Poly(A) binding protein nuclear 1 levels affect alternative polyadenylation. Nucleic Acids Res. de Mezer M, Wojciechowska M, Napierala M, Sobczak K , Krzyzosiak WJ (2011). Mutant CAG repeats of Huntingtin transcript fold into hairpins, form nuclear foci and are targets for RNA interference. Nucleic Acids Res. 39, 3852-3863.

E. de Klerk AV, S.Y. Anvar, J.J. Goeman, O. Hu, J.T. den Dunnen, S.M. van der Maarel, V. Raz, P.A.C. 't Hoen (2012). Poly(A) binding protein nuclear 1 (PABPNl) levels affect alternative polyadenylation. Nucleic Acids Res. In press.

Edstrom E, Altun M, Hagglund M , Ulfhake B (2006). Atrogin- 1/MAFbx and MuRFl Are Downregulated in Aging-Related Loss of Skeletal Muscle. The

Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 61, 663-674.

Eisenhaber B, Chumak N, Eisenhaber F , Hauser M-T (2007). The ring between ring fingers (RBR) protein family. Genome Biology. 8, 209.

Hipkiss AR (2006). Accumulation of altered proteins and ageing: causes and effects. Exp Gerontol. 41, 464-473.

Jenal M, Elkon R, Loayza-Puch F, van †Haaften G, KV°hn U, Menzies FM, Vrielink JAFO, Bos AJ, Drost J, Rooijers K, Rubinsztein DC , Agami R (2012). The Poly(A) -Binding Protein Nuclear 1 Suppresses Alternative Cleavage and

Polyadenylation Sites. Cell. 149, 538-553.

Junghans RP (2009). Dystrophia myotonia: why focus on foci? Eur J Hum Genet. 17, 543-553.

Kuhn U, Gundel M, Knoth A, Kerwitz Y, Rudel S , Wahle E (2009). Poly(A) tail length is controlled by the nuclear poly (A) -binding protein regulating the interaction between poly(A) polymerase and the cleavage and polyadenylation specificity factor. J Biol Chem. 284, 22803-22814.

Lindner AB , Demarez A (2009). Protein aggregation as a paradigm of aging.

Biochimica et Biophysica Acta (BBA) - General Subjects. 1790, 980-996.

Lu T, Pan Y, Kao SY, Li C, Kohane I, Chan J , Yankner BA (2004). Gene regulation and DNA damage in the ageing human brain. Nature. 429, 883-891. Marzani B, Felzani G, Bellomo RG, Vecchiet J , Marzatico F (2005). Human muscle aging: ROS-mediated alterations in rectus abdominis and vastus lateralis muscles. Exp Gerontol. 40, 959-965.

Passtoors WM, Boer JM, Goeman JJ, Akker EB, Deelen J, Zwaan BJ, Scarborough A, Breggen R, Vossen RH, Houwing-Duistermaat JJ, Ommen GJ, Westendorp RG, Heemst D, Craen AJ, White AJ, Gunn DA, Beekman M , Slagboom PE (2012). Transcriptional profiling of human familial longevity indicates a role for ASF1A and IL7R. PLoS One. 7, e27759. Raz V, Abraham T, van Zwet EW, Dirks RW, Tanke HJ , van der Maarel SM

(2011a). Reversible aggregation of PABPN1 pre-inclusion structures. Nucleus. 2, 208-218.

Raz V, Carlotti F, Vermolen BJ, van der Poel E, Sloos WC, Knaan-Shanzer S, de Vries AA, Hoeben RC, Young IT, Tanke HJ, Garini Y , Dirks RW (2006). Changes in lamina structure are followed by spatial reorganization of heterochromatic regions in caspase-8-activated human mesenchymal stem cells. J Cell Sci. 119, 4247-4256. Raz V, Routledge S, Venema A, Buijze H, van der Wal E, Anvar S, Straasheijm KR, Klooster R, Antoniou M , van der Maarel SM (2011b). Modeling oculopharyngeal muscular dystrophy in myotube cultures reveals reduced accumulation of soluble mutant PABPN1 protein. Am J Pathol. 179, 1988-2000. Rodwell GE, Sonu R, Zahn JM, Lund J, Wilhelmy J, Wang L, Xiao W, Mindrinos M, Crane E, Segal E, Myers BD, Brooks JD, Davis RW, Higgins J, Owen AB , Kim SK (2004). A transcriptional profile of aging in the human kidney. PLoS Biol. 2, e427.

Sandri M, Sandri C, Gilbert A, Skurk C, Calabria E, Picard A, Walsh K, Schiaffino S, Lecker SH , Goldberg AL (2004). Foxo Transcription Factors Induce the

Atrophy-Related Ubiquitin Ligase Atrogin- 1 and Cause Skeletal Muscle Atrophy. Cell. 117, 399-412.

Shao J , Diamond MI (2007). Polyglutamine diseases: emerging concepts in pathogenesis and therapy. Hum Mol Genet. 16, R115-R123. Tavanez JP, Calado P, Braga J, Lafarga M , Carmo-Fonseca M (2005). In vivo aggregation properties of the nuclear poly(A) -binding protein PABPN1. RNA. 11, 752-762.

Tome FM , Fardeau M (1980). Nuclear inclusions in oculopharyngeal dystrophy. Acta Neuropathol. 49, 85-87.

Trollet C, Anvar SY, Venema A, Hargreaves IP, Foster K, Vignaud A, Ferry A, Negroni E, Hourde C, Baraibar MA, t Hoen PA, Davies JE, Rubinsztein DC, Heales SJ, Mouly V, van der Maarel SM, Butler-Browne G, Raz V , Dickson G (2010). Molecular and phenotypic characterization of a mouse model of oculopharyngeal muscular dystrophy reveals severe muscular atrophy restricted to fast glycolytic fibres. Hum Mol Genet. 19, 2191-2207.

Verheesen P, de Kluijver A, van Koningsbruggen S, de Brij M, de Haard HJ, van Ommen G-JB, van der Maarel SrM , Verrips CT (2006). Prevention of

oculopharyngeal muscular dystrophy-associated aggregation of nuclear poly(A)- binding protein with a single-domain intracellular antibody. Hum Mol Genet. 15, 105-111. Zahn JM, Sonu R, Vogel H, Crane E, Mazan-Mamczarz K, Rabkin R, Davis RW, Becker KG, Owen AB , Kim SK (2006). Transcriptional profiling of aging in human muscle reveals a common aging signature. PLoS Genet. 2, el 15.

Claims

A method for increasing the level of PABPNl mRNA and/or protein in a cell, said method comprising increasing the level of ARIH2 protein in said cell. A method according to claim 1, wherein said cell is a cell of an adult.

A method according to claim 1 or claim 2, wherein the level of ARIH2 protein is increased by:

- providing the cell with an antisense oligonucleotide is that is complementary to and capable of hybridizing to ARIH2 (pre-)mRNA produced by said cell;

- providing the cell with a nucleic acid that encodes ARIH2 protein;

- by decreasing the level and/or activity of MDM2 in said cell;

- by increasing the level and/or activity of HoxAlO in said cell; and/or

- by contacting the cell with an effective amount of all- trans retinoic acid.

A method according to claim 3, wherein said antisense oligonucleotide is complementary to and capable of hybridizing to the poly-adenylation signal sequence located at position 314-337 in the ARIH2 sequence of figure 8. A method according to claim 3, wherein said antisense oligonucleotide is complementary to and capable of hybridizing to a miRNA target sequence located in the 3' untranslated region of the ARIH2 sequence of figure 8, wherein said miRNA target sequence preferably is a mir9 target sequence. A method according to any one of claims 1-5, wherein said cell is a muscle cell, preferably a skeletal muscle cell, a senescent cell, neuronal cell, a satellite cell, an adult stem cell, preferably a mesenchymal stem cell.

A method for modifying poly-adenylation site usage in a cell comprising modifying the level of ARIH2 protein in said cell.

A method for inhibiting a molecular effect of aging in an adult cell said method comprising increasing the level of ARIH2 protein in said cell.

A method for modifying a molecular effect of aging in an adult cell said method comprising providing said cell with an - antisense oligonucleotide that is complementary to and capable of hybridizing to a poly-adenylation site;

of a pre-mRNA encoded by the gene HILFA; EGFR; SUMOl; PASMD14; SLC1A4; E2F1; BMI1; RBI; TP63; SLC2A4; ING1; MIB1; PTEN ; SUB1 ; IL10 ; VEGFA ; MORF4L1; DNM1L; SOD2; RAC1; ITGB1; HIF1A; SRF; MCL1; RAD17; GSK3B; RYR1; MEF2A; GRB2; ZMYNDll; DNMT3A; CDK7; DIABLO; JARID2; EIF4E; UBE3A; BECN1; KHDRBS3; NOLC1; LPL; NF2; ROCK2; STAT5A; ITSN1; STAT5B; HUS1; HSPA9; LPL; ABI3; CAMK2D; HSPD 1; ITPR1; CSNK2A1; PAK1; BIRC2; RSL1D1; GCLM; CDS1; HDAC3; YYl; RPAl; RRM2B; CD55; HDAC4; KSRl; CD59; HMGBl; GCLC; PSMD 14; CADM1; TFRC; RBX-1; PDK1; AGFG1; PSMA2 ; PLCB4; LAMP1; RHOA ; RAP-1; UGCG; H2AFZ; CANX; UBE2I; TPP2; REVl; EMB; COL5A1; RABIA; IMPACT; VAMP2; TANK; HSPE1; MYL1; PRKAR1A; WASL; HNRNPK; SGCB; GABPA; MLF1; SMEK2; M6PR; PERP; PPP1CC; HSF2; TDP-43; C90RF72; or MAP3K7.

A method for modifying PAS usage in a (pre-)mRNA expressed by a cell, comprising providing said cell with an antisense oligonucleotide that is complementary to and capable of hybridizing to a poly-adenylation site of a pre-mRNA encoded by the gene ARIH2; HILFA; EGFR; SUMOl; PASMD14; SLC1A4; E2F1; BMI1; RBI; TP63; SLC2A4; ING1; MIB1; PTEN ; SUB1 ; IL10 ; VEGFA ; MORF4L1; DNM1L; SOD2; RAC1; ITGB1; HIF1A; SRF; MCL1; RAD17; GSK3B; RYR1; MEF2A; GRB2; ZMYNDll; DNMT3A; CDK7; DIABLO; JARID2; EIF4E; UBE3A; BECN1; KHDRBS3; NOLC1; LPL; NF2; ROCK2; STAT5A; ITSN1; STAT5B; HUS1; HSPA9; LPL; ABI3; CAMK2D; HSPD 1; ITPR1; CSNK2A1; PAK1; BIRC2; RSL1D1; GCLM; CDS1; HDAC3; YYl; RPAl; RRM2B; CD55; HDAC4; KSRl; CD59; HMGBl; GCLC; PSMD 14; CADM1; TFRC; RBX-1; PDK1; AGFG1; PSMA2 ; PLCB4; LAMP1; RHOA ; RAP-1; UGCG; H2AFZ; CANX; UBE2I; TPP2; REVl; EMB; COL5A1; RABIA; IMPACT; VAMP2; TANK; HSPE1; MYL1; PRKAR1A; WASL; HNRNPK; SGCB; GABPA; MLF1; SMEK2; M6PR; PERP; PPP1CC; HSF2; TDP-43; C90RF72; or MAP3K7,

11. A method according to claim 9 or claim 10, wherein said poly-adenylation site is the proximal poly-adenylation site.

12. A method according to claim 9 or claim 10, wherein said miRNA target

sequence is in the 3' untranslated region of the (pre-)mRNA.

13. A method according to claim 12, wherein said miRNA target sequence is

distal to said proximal poly-adenylation site.

14. An isolated oligonucleotide having 12-40 bases, wherein said oligonucleotide: comprises a continuous stretch of at least 7 bases that is complementary to and capable of hybridizing to a poly-adenylation site; comprises a continuous stretch of at least 7 bases that is complementary to and capable of

hybridizing to a miRNA target sequence; or comprises a continuous stretch of at least 7 bases that is complementary to and capable of hybridizing to a regulatory sequence in the 3'UTR, of an ARIH2 pre-mRNA or of a (pre- )mRNA encoded by the gene HILFA; EGFR; SUMOl; PASMD14; SLC1A4; E2F1; BMI1; RBI; TP63; SLC2A4; ING1; MIB1; PTEN ; SUB1 ; IL10 ;

VEGFA ; MORF4L1; DNM1L; SOD2; RAC1; ITGB1; HIF1A; SRF; MCL1; RAD17; GSK3B; RYR1; MEF2A; GRB2; ZMYND11; DNMT3A; CDK7;

DIABLO; JARID2; EIF4E; UBE3A; BECN1; KHDRBS3; NOLC1; LPL; NF2; ROCK2; STAT5A; ITSN1; STAT5B; HUS1; HSPA9; LPL; ABI3; CAMK2D; HSPD 1; ITPR1; CSNK2A1; PAK1; BIRC2; RSL1D1; GCLM; CDS1; HDAC3; YYl; RPAl; RRM2B; CD55; HDAC4; KSRl; CD59; HMGBl; GCLC; PSMD 14; CADM1; TFRC; RBX-1; PDK1; AGFG1; PSMA2 ; PLCB4; LAMP1; RHOA ; RAP-1; UGCG; H2AFZ; CANX; UBE2I; TPP2; REVl; EMB; COL5A1; RABIA; IMPACT; VAMP2; TANK; HSPE1; MYL1; PRKAR1A; WASL; HNRNPK; SGCB; GABPA; MLF1; SMEK2; M6PR; PERP; PPP1CC; HSF2; TDP-43; C90RF72; or MAP3K7.

15. An oligonucleotide according to claim 14, comprising a modification that

renders the oligonucleotide nuclease insensitive.

16. An oligonucleotide according to claim 15, wherein said oligonucleotide comprises a phosphorothioate backbone.

17. An isolated oligonucleotide having 12-40 bases, comprising a continuous stretch of at least 7 bases of sequence:

ARIH2* 5'-GTA TAA TTG TAC AAC CTT TGA AAG-3'

Psmdl4 5'-GAGCGCCACTGACAGCTCTCTTA-3'

Grb2 5'-GACAAGAAACCAAGTGGGC-3'

Stat5b 5'-GAAGTGTTAATACTAGTTGT-3'

18. A method for the treatment of an individual suffering from cancer said method comprising administering a PABPNl, and/or an ARIH2 inhibitor to the individual in need thereof.