Bicistronic AAV vector for RNA interference in ALS

Abstract

The present invention relates to a bicistronic expression vector for silencing a gene specifically in astrocytes and neurons, comprising two expression cassettes comprising a first and a second silencer sequence, respectively, wherein the expression of said first silencer sequence within astrocytes is regulated by an astrocyte-specific promoter and the expression of said second silencer sequence within neurons is regulated by a neuron-specific promoter. In a preferred embodiment, said first and second silencer sequences are SOD1 silencer sequences. Pharmaceutical composition comprising said bicistronic vector and the use of the same in the treatment of motoneuron diseases are further described.

Claims

1. A bicistronic expression vector for silencing a gene in astrocytes and neurons comprising two expression cassettes, wherein a first cassette comprises astrocyte specific promoter-posttranscriptional regulatory element-the first silencer sequence-polyA tail and a second cassette comprises neuron specific promoter-posttranscriptional regulatory element-the second silencer sequence-polyA tail; wherein, when injected into a subject, the bicistronic expression vector is able to target both the astrocytes and neurons brain cell populations with one injection.

2. The bicistronic vector according to claim 1, wherein said first and second silencer sequences are both SOD1 silencer sequences.

3. The bicistronic vector according to claim 2, wherein said SOD1 silencer sequences are independently selected from the group comprising RNA sequences that can be transcribed from DNA to interfere with SOD1 expression, including small hairpin RNA (shRNA) against SOD1, micro RNA (miRNA) against SOD1, antisense RNA sequences against SOD1 and guide RNA sequences for CRISPR/Cas9 mediated targeting of the SOD1 gene.

4. The bicistronic vector according to claim 1, wherein said first and second silencer sequences are both a miRNA designed to suppress SOD1 expression (miR SOD1).

5. A bicistronic expression vector for silencing a gene in astrocytes and neurons comprising two expression cassettes, wherein a first cassette comprises astrocyte specific promoter-posttranscriptional regulatory element-the first silencer sequence-polyA tail and a second cassette comprises neuron specific promoter-posttranscriptional regulatory element-the second silencer sequence-polyA tail; wherein said first and second silencer sequences are both a miRNA designed to suppress SOD1 expression (miR SOD1); and, wherein said miR SOD1 targets the coding sequence of human SOD1 (NM 000454), nt 209-229 (miR SOD1) (SEQ ID No. 2).

6. A bicistronic expression vector for silencing a gene in astrocytes and neurons comprising two expression cassettes, wherein a first cassette comprises astrocyte specific promoter-posttranscriptional regulatory element-the first silencer sequence-polyA tail and a second cassette comprises neuron specific promoter-posttranscriptional regulatory element-the second silencer sequence-polyA tail; wherein said first and second silencer sequences are both a miRNA designed to suppress SOD1 expression (miR SOD1); and, wherein said miRNA sequence comprises a 6 or 7 nucleotides-long seed sequence as shown in SEQ ID No. 2 which is fully homologous to the human SOD1 mRNA transcript (NM_000454).

7. A bicistronic expression vector for silencing a gene in astrocytes and neurons comprising two expression cassettes, wherein a first cassette comprises astrocyte specific promoter-posttranscriptional regulatory element-the first silencer sequence-polyA tail and a second cassette comprises neuron specific promoter-posttranscriptional regulatory element-the second silencer sequence-polyA tail; wherein said first and second silencer sequences are both a miRNA designed to suppress SOD1 expression (miR SOD1); and, wherein said miR SOD1 mature sequence has at least 50%, 60%, 80%, still more preferably 90% sequence identity with the corresponding coding sequence set forth in SEQ ID No 1: 5-ATT ACT TTC CTT CTG CTC GAA-3 (SEQ ID, No 1).

8. The bicistronic vector according to claim 7, wherein said miR SOD1 mature sequence corresponds to the coding sequence: 5-ATT ACT TTC CTT CTG CTC GAA-3 (SEQ ID. No. 1).

9. The bicistronic vector according to claim 1, wherein said astrocyte specific promoter is selected from the group comprising the GFAP promoter, the glutamine synthase promoter, and minimal GFAP promoter gfaABC.sub.1D.

10. The bicistronic vector according to claim 1, wherein said neuron specific promoter is selected from the group comprising synapsin, cmv, platelet-derived growth factor B-chain (PDGF-), the methyl-CpG binding protein 2 (MeCP2), and human synapsin I promoter hSyn1.

11. The bicistronic vector according to claim 1, wherein said vector is an Adeno-Associated Viral (AAV) vector.

12. The bicistronic vector according to claim 11, wherein said AAV vector is of a serotype compatible with widespread transgene delivery to astrocytes and motoneurons, including serotype 9 or of serotype 6.

13. A method of treating of motoneuron diseases, including amyotrophic lateral sclerosis (ALS), comprising injecting a bicistronic expression vector according to claim 1 into a subject in need thereof.

14. A pharmaceutical composition comprising a bicistronic AAV vector comprising two expression cassettes comprising a first and a second gene silencer sequence, wherein the expression of said first silencer sequence within astrocytes is regulated by an astrocyte-specific promoter and the expression of said second silencer sequence within neurons is regulated by a neuron-specific promoter and a pharmaceutically acceptable carrier and, wherein, when the pharmaceutical composition is injected into a subject, the bicistronic expression vector is able to target both the astrocytes and neurons brain cell populations with one injection.

15. The pharmaceutical composition according to claim 14 for use in the treatment of motoneuron diseases, including amyotrophic lateral sclerosis (ALS).

16. A method of using an AAV vector-based biological delivery and expression system according to claim 1 for expressing a first and a second silencer sequence in astrocytes and motoneurons.

Description

DESCRIPTION OF THE DRAWINGS

(1) FIG. 1: cartoon diagram of vectors used in the present invention. A), B), C) comparative examples. D) an embodiment of the bicistronic vector of the present invention.

(2) FIG. 2: SEQ ID NO.2 showing SOD1 silencer sequence (miR SOD1), miRNA stem-loops, with the guide strand shown in grey.

(3) FIG. 3: compound muscle action potential measured in triceps surae muscles during time after infection of the mice. Study on the therapeutic efficiency, comparing the bicistronic vector of the present invention to an astrocyte-specific vector (GFAP promoter, panel A) and neuron-specific vectors (cmv and hsyn1 promoters, panel B).

(4) FIG. 4: the swimming performance of the mice is evaluated over time after intrathecal injection of the different vectors. Study on motor ability comparing the therapeutic efficiency of the bicistronic vector of the present invention to an astrocyte-specific vector (GFAP promoter, panel A) or neuron-specific vectors (cmv and hysn1 promoters, panel B).

(5) FIG. 5: Motoneuron counts per lumbar spinal cord section. Motoneuron numbers are significantly increased for the bicistronic group (white column) when compared to transgenic control animals.

(6) FIG. 6: Neuromuscular junction integrity expressed as percentage of bungarotoxin-positive motor end plates that are occupied by a SV2-positive motoneuron terminal. Innervation is fully preserved in the bicistronic group (white column), which is significantly different from the partial preservation in the other treated groups.

DETAILED DESCRIPTION

(7) The present invention describes a bicistronic expression vector, allowing continuous expression of a first and a second silencer sequence specifically in motoneurons and astrocytes.

(8) The bicistronic expression vector of the present invention comprises a first expression cassette comprising a first nucleotide sequence encoding a first silencer sequence operably linked to a promoter functional in astrocytes and a second expression cassette comprising a second nucleotide sequence encoding a second silencer sequence operably linked to a promoter functional in neurons.

(9) Said bicistronic vector, in a preferred embodiment, comprises, in both said first and second cassettes, a posttranscriptional regulatory element upstream of said sequences encoding said silencer sequences and a polyA tail downstream from the same.

(10) In a preferred embodiment, said first and second silencer sequences are SOD1 silencer sequences.

(11) In a further preferred embodiment, said vector is an AAV vector and the regulatory sequences are selected so that the total size of the vector is below 5 kb, which is compatible with packaging within the AAV capsid. Said AAV vector is of any serotype compatible with widespread transgene delivery to astrocytes and motoneurons; preferably, said AAV vector is of serotype 9 or of serotype 6.

(12) In a preferred embodiment, the astrocyte specific promoter is selected from the group comprising the GFAP promoter, the glutamine synthase promoter, preferably is the minimal GFAP promoter gfaABC.sub.1D; the neuron specific promoter is selected from the group comprising synapsin, cmv, platelet-derived growth factor B-chain (PDGF-), the methyl-CpG binding protein 2 (MeCP2), preferably it is the human synapsin I promoter hSyn1.

(13) The above indicated mammalian promoters are compatible with long-term transgene expression and they are known to restrict the expression of the controlled gene to subsets of cells of interest in the CNS.

(14) Said SOD1 silencer sequence is selected from the group comprising RNA sequences that can be transcribed from naturally occurring or artificial DNA to interfere with SOD1 expression, including small hairpin RNA (shRNA) against SOD1, micro RNA (miRNA) against SOD1, antisense RNA sequences against SOD1, or guide RNA sequences for CRISPR/Cas9 targeting of the SOD1 gene. In a preferred embodiment, the SOD1 silencer is a miRNA targeting SOD1 mRNA, preferably targeting the nt 209-229 (SEQ ID No. 1) (miR SOD1) of the coding sequence of human SOD1. For the purpose of the present invention, miR SOD1 is related to a miRNA targeting SOD1, where said SOD1 targeting sequence is referred to the nt 209-229 of the coding sequence of human SOD1 (SEQ ID No. 1).

(15) In a further preferred embodiment, said miRNA sequence comprises a 6 or 7 nucleotides-long seed sequence which is fully homologous to the human SOD1 mRNA transcript (NM_000454).

(16) In a further preferred embodiment miRNA mature sequence has at least 50%, 60%, 80%, 90% sequence homology, wherein for the scope of the present invention sequence homology is referred to sequence identity, with the corresponding coding sequence set forth in SEQ ID No. 1: 5-ATT ACT TTC CTT CTG CTC GAA-3 (SEQ ID. No. 1). More preferably, said miRNA mature sequence corresponds to the coding sequence: 5-ATT ACT TTC CTT CTG CTC GAA-3 (SEQ ID No. 1).

(17) In a preferred embodiment, said mature miRNA is carried by the pre-miRNA backbone of murine miR-155. The corresponding DNA sequence was synthesized and subcloned into a pAAV shuttle plasmid according to standard procedures.

(18) The inclusion of posttranscriptional regulatory element in the expression cassette, said regulatory element being, as an example, an intron or a Woodchuck hepatitis Posttrascriptional Regulatory Element (WPRE), results in a significantly increased expression of the SOD1 silencer sequence encoded by the said vector. Said intron is preferably selected from the group comprising MVM (minute virus of mice), F.IX truncated intron 1 (human factor IX), -globin splice donor SD/immunoglobin heavy chain splice acceptor SA, adenovirus SD/immunoglobin SA. However, other natural or artificial introns known to the person skilled in the art can be used. Alternatively, a posttranscriptional enhancer element such as the WPRE can be used.

(19) Compared to other SOD1 silencing systems used so far in the field of ALS, this bicistronic expression vector surprisingly compiles the following advantages that are relevant to therapeutic application: the use of cell-type specific promoters helps limiting expression of the SOD1 silencer sequence to neurons and astrocytes. This minimizes chances of deleterious effects due to off-target silencing of SOD1 or activation of the miRNA pathways in other non-therapeutically relevant cell types. This is particularly true given that SOD1 is a ubiquitous enzyme. Only one allele is mutated in familial ALS. It is unclear whether SOD1 activity produced by the normal allele is critical to the function and survival of certain cell types. Therefore, it is preferred to limit SOD1 silencing to diseased cells, such as motoneurons and astrocytes, in order to limit the risk of unwanted side effects. the here obtained successful incorporation of two expression cassettes into a single vector allows to express SOD1 silencer sequence simultaneously and specifically in motoneurons and astrocytes following a single injection of a viral vector that has a tropism for the two cell types, thus avoiding the limitations of using vector cocktails. These limitations include decrease in transduction efficiency due to the lower injectable dose of each individual vector, as well as the need to develop two separate therapeutic products to be ultimately combined. Also in regulatory terms, this solution brings a considerable advantage: in the event that two separate vectors are used, two procedures are required to get the regulatory approval for both of them. In this case, the burden of a single procedure has to be afforded.

(20) Compared to vectors with single expression cassettes, the here claimed bicistronic expression vector drastically improves the therapeutic benefits obtained by gene therapy aiming at silencing SOD1 in ALS patients.

(21) The here claimed bicistronic expression vector allows to target two different cell types by administration of a single vector, where the combined injection of two viral vectors would not only be a problem concerning the maximal vector dose tolerated by the organism but would also significantly raise the regulatory hurdles towards clinical application.

(22) Moreover, as it will become evident from the examples that follow, a maximization of the therapeutic efficiency is obtained. Since motoneurons and astrocytes play distinctive roles in the pathology and SOD1 silencing in each of these cell types has complementary effects, targeting both motoneurons and astrocytes is a critical factor.

(23) It is also claimed a bicistronic expression vector according to the present invention for use in the treatment of ALS.

(24) It is further described a pharmaceutical composition, comprising a bicistronic AAV vector comprising two expression cassettes each comprising a first and a second silencer sequence, wherein the expression of said first silencer sequence within astrocytes is regulated by an astrocyte-specific promoter and the expression of said second silencer sequence within neurons is regulated by a neuron-specific promoter and a pharmaceutically acceptable carrier for use in the treatment of motoneuron disease, preferably of ALS.

(25) In a preferred embodiment, in said pharmaceutical composition said first and second silencer sequences are SOD1 silencer sequences.

(26) It is further described a method of treatment of a subject in need thereof, said method comprising administering to a subject in need thereof the described vector in a range comprised between 1E12-5E14 Vector Genomes VG/kg body weight.

(27) Said administration, in a preferred embodiment, is via the cerebrospinal fluid. More preferably, it is Intrathecal (IT), Intracisternal (IC) or Intraventricular (ICV). In a further preferred embodiment, said administration is Intravenous (IV). The recombinant virions are preferably introduced to the subject in combination with an adjunctive pharmacological therapy.

Example 1

(28) SOD1.sup.G93A ALS mice, carrying a transgene which is the mutant human SOD1 containing the Gly.sup.93-->Ala (G93A) substitution, are used in the experimental setting. SOD1.sup.G93A ALS mice are widely used as a model for ALS and their limbs become progressively paralyzed beginning around six to seven months of age. Life expectancy is normally four to six weeks beyond onset of symptoms. SOD1.sup.G93A ALS mice have a highly predictable course of denervation that takes place in two successive episodes. There is an initial pruning of fast-twitch fatigable (FF) motoneurons, resulting in the denervation of type IIb muscle fibers. Vacant neuromuscular junctions (NMJ) are then partially re-innervated by the sprouting of fast-twitch fatigue-resistant (FR) and slow-twitch fatigue-resistant (S) motoneurons until FR motoneurons also start pruning their intramuscular nerve branches leading to a sustained loss of NMJ (Pun, Santos et al., 2006; Kanning, Kaplan et al., 2010). SOD1.sup.G93A ALS mice were injected intrathecally at the age of 4.5 weeks with 2.1E+12 VG of either one of the following AAV9 viral vectors: A) AAV9-cmv-RFP-miR SOD1 (control, to silence SOD1 in motoneurons) B) AAV9-hsyn1-RFP-miR SOD1 (control, to silence SOD1 in motoneurons) C) AAV9-gfaABC.sub.1D-GFP-miR SOD1 (control, to silence SOD1 in astrocytes) D) AAV9-bicistronic miR SOD1 (vector of the present invention, to silence SOD1 in motoneurons and astrocytes).

(29) In FIG. 1 a schematic representation of the above-indicated vectors is reported. In this experimental setting, the SOD1 silencer sequence was engineered based on the murine miRNA-155 flanking sequences. Its guiding strand targets the nt 209-229 of the coding sequence of human SOD1 (FIG. 2). miR SOD1 is an engineered miRNA based on the flanking sequences and loop from the murine miRNA-155 sequence (represented in black in FIG. 2). It has been designed for the specific silencing of human SOD1. The SOD1 targeting sequence (bold font) is composed of a guide strand (in grey) that binds to SOD1 mRNA and contains a 7-nucleotide-long seed sequence (in italic and underlined), and the partially complementary passenger strand. Said guide strand gives rise to a miR SOD1 mature sequence, which correspond to the coding sequence SEQ ID No. 1.

(30) To evaluate the therapeutic efficiency of these viral vectors, the four groups of mice (7 animals per group except for bicistronic vector group, which had 8 animals) were compared to PBS-injected wild-type and SOD1.sup.G93A ALS control animals (16 and 19 animals per group, respectively). In each group, males and females were equally distributed. Littermates were split among groups.

(31) Electromyography (EMG) was performed on a weekly basis on the triceps surae muscles to assess neuromuscular function. EMG consists in electrically stimulating the motoneurons of the spinal cord and recording the muscular response, called Compound Muscle Action Potential (CMAP). EMG is one of the most reliable tests to predict clinical disease onset and disease progression in ALS mice (Mancuso, Osta et al., 2014) and ALS patients (Simon, Lomen-Hoerth et al., 2014).

(32) Results are reported in the graph of FIG. 3.

(33) In PBS-injected SOD1.sup.G93A mice, CMAP amplitude declined between postnatal days 50 and 60, and the amplitude then stabilized at around 40% of the starting value. At 12 weeks of age, the CMAP measurements of SOD1.sup.G93A control mice (33.410.2 mV) did not statistically differ from mice injected with AAV9 vectors silencing SOD1 in either motoneurons or astrocytes. There was, however, a trend towards improved muscle response in treated animals (AAV9-cmv-RFP-miR SOD1, 46.816.2 mV; AAV9-hsyn-RFP-miR SOD1, 44.713.4 mV; gfaABC.sub.1D-GFP-miR SOD1, 50.814.3 mV).

(34) Remarkably, a significant difference was already observed at 8 weeks between the AAV9-bicistronic group and control ALS mice (p<0.05). At 12 weeks of age the difference between the two groups was even more pronounced (p<0.001). Indeed, at this time point the CMAP values of animals injected with the bicistronic expression vector (75.613.4 mV) did not statistically differ from the values obtained for wild-type control mice (84.818.3 mV).

(35) By further evaluating the CMAP at 20 weeks of age, where the symptoms are evident in the used animal model, the values obtained in the animals injected with the bicistronic expression vector are unchanged with respect to the value obtained at 12 weeks of age (76.57.2 mV), while the values further decreased in PBS injected SOD1 mice (17.51.4 mV), which is consistent with the second wave of denervation typically observed in the SOD1.sup.G93A mice at this age (FIG. 3). Overall, a significant difference was observed between the group injected with the bicistronic vector and the group injected with the vector containing the astrocyte-specific GFAP-promoter (p<0.001). Significant difference was as well observed between the bicistronic vector group and the groups injected with the vectors containing the promoters for preferential expression in neurons (for cmv and hsyn1, p<0.001) (FIG. 3, panel B). This surprising result confirms the validity of the here proposed approach.

(36) This is further confirmed when assessing motor performance of the mice, by measuring the time taken by the animals to swim to a 1 m distant platform. Results are represented in FIG. 4. Overall, a significance difference in swimming performance was observed between the group injected with the bicistronic vector and the group injected with the vector containing the astrocyte-specific GFAP promoter (p<0.05). A significant difference was as well observed between the bicistronic vector group and the group injected with vectors containing the promoters for preferential expression in neurons (for cmv p<0.01 and for hysn1 p<0.05), while almost no significative difference was observed between the motor performances of the bicistronic group and the group of the wild-type control animals.

(37) These results clearly demonstrate the superior therapeutic potential of the here claimed bicistronic expression vector when compared to expression vectors carrying a single expression cassette that expresses the miR SOD1 in motoneurons or astrocytes only.

(38) The FIGS. 5 and 6 confirm at a histological level that silencing SOD1 in both neurons and astrocytes using the bicistronic vector has a greater therapeutic outcome than SOD1 silencing in only one of the two cell types. Spinal cord sections were obtained from treated animals at end stage. Choline acetyltransferase (ChAT)-positive motoneurons were counted in the lumbar region of the spinal cord. Indeed, at the level of motoneuron survival (FIG. 5) the bicistronic vector shows a statistically significant increase in motoneuron survival when compared to vectors expressing miR SOD1 under the control of either the GFAP or the hsyn promoter alone.

(39) Neuromuscular junction (NMJ) integrity (FIG. 6) has then been assessed. NMJ occupancy was evaluated on 20 m muscle sections stained with an anti-SV-2 (Developmental Studies Hybridoma Bank, University of Iowa, Iowa City, Iowa) or VAChT antibody and tetramethylrhodamine -bungarotoxin (Invitrogen). About 100 bungarotoxin-positive motor end plates were identified and checked for overlapping SV-2 or VAChT-positive motoneuron terminal using an Olympus AX70 microscope or Leica DM5500 microscope. The bicistronic vector (white column) shows a higher percentage of bungarotoxin-positive motor end plates that are occupied by a SV2-positive motoneuron terminal when compared to the vector with the vectors expressing miR SOD1 under the control of either the GFAP or the hsyn promoter alone.