The present invention relates in a first aspect to a mutated adeno-associated virus (AAV) capsid protein or fragment thereof having an insert composed of an oligopep tide optionally having further amino acids representing linker sequences at both sites of said oligopeptide. These mutated AAV capsid proteins are more efficient in trans ferring genetic material into hepatic tissue, hepatocytes, hepatic cells and cell lines or hepatocellular carcinoma (HCC) partially with improved tropism for these cell types. Further, a mutated AAV particle (uniform or as hybrid) comprising the mutated AAV capsid protein according to the present invention is provided. In addition, a nucleic acid encoding the mutated AAV capsid protein according to the present invention is identified together with a corresponding nucleic acid vector, in particular, a plasmid.

In addition, a host cell containing the nucleic acid vector or the nucleic acid molecule according to the present invention. Further, a use of the AAV particle or the nucleic acid or the nucleic acid vector according to the present invention in the manufacture of a medicament for gene therapy. Moreover, the described protein, particle mole cules as well as nucleic acid vectors for use in targeting hepatocytes and/or FICC are described. In particular, said components for use in treating diseases involving hepatocytes or for treating FICC are disclosed, in particular, for use in gene therapy, e.g. for use in a transfer of a gene of interest into hepatocytes, hepatic tissue or HCC.

Background of the invention

The liver represents an essential organ with complex metabolic, immunological and hemostatic functions. The liver is affected by a broad range of monogenic disor ders and chronic metabolic conditions, see Marcellin, P. and Kutala, B. K., Liver In ternational, 2018, 38(S1 ), 2-6. Consequently, liver has become a key target for the development of gene therapy strategies in particular with AAV vectors as summa rized in Baruteau, J., Journal of Inherited Metabolic Disease, 2017, 497-517. As reviewed recently, promising results, for example, have been obtained for patients suffering from hemophilia, see Pierce, G. F. and lorio, A., Hamophilia, 1018, 24(Suppl 6), 60-67. However, still high vector doses are required to compensate for the low efficacy of AAV vectors in human patients. While efficacy is sufficient to ob tain the moderate physiological levels of blood clotting factors required to transform severe hemophilic phenotypes into mild forms, other liver diseases are in urgent need for more effective AAV vectors. In addition, patients establish neutralizing cap sid antibodies upon the first AAV vector dose, as described by Fitzpatrick et al. , Mo lecular Therapy: Methods & Clinical Development. Elsevier Ltd., 9 (June), 1 19-129. This impairs re-administration of AAV vectors with the same or cross-reacting AAV capsids for improving transgene expression levels. Therefore, AAV vectors with novel capsid properties for efficient liver-directed gene transfer are of utmost importance for improving current and future gene therapeutic approaches.

Beside the above-mentioned diseases, liver is prone to develop cancer, in par ticular hepatocellular carcinoma (HCC), or to being the“home” for metastasis origi nating from tumors other than HCC. With regard to HCC, it is reported to be the third most common cause of cancer-related deaths worldwide. It is a severe cancer dis ease with poor prognosis at advanced HCC stages. So far potentially curative treat ments are limited to surgical procedures like tumor resection, orthotopic liver trans plantation, or percutaneous radio frequency ablation, see Waghray, A. et al., WJH, 2015, 7(8), 1020-1029. The majority of patients are diagnosed at advanced stages of HCC for which current therapeutic options are very poor. Therefore, there is a strong need for novel therapeutic approaches including AAV vector-based cancer gene ther apy. AAV vectors may be used to deliver suicide genes, tumor-suppressor genes or immunotherapeutic transgenes to the tumor site, e.g. as discussed in Dhungel, B. et al., 2017, 7544. To provide efficient and safe delivery of transgene for cancer gene therapy, engineering of novel AAV vectors is required which are directed to the HCC target tissue and are capable of efficient therapeutic transgene expression.

AAV vectors are based on the adeno-associated virus (AAV), a non-pathogenic member of the parvoviridae and the genus Dependoparvovirus. AAV are composed of a single-stranded DNA genome of about 4.7 kB packaged into a non-enveloped icosahedral protein capsid. It is a small (25 nm in diameter) virus, consisting of 60 monomers forming the icosahedral capsid. These monomers are composed of the vi ral capsid proteins (VP) VP1 , VP2, and VP3 in a 1 : 1 : 10 ratio. The 4.5 kB single stranded coding DNA genome (plus inverted terminal repeats (ITRs)) is bearing two genes ( rep and cap) encoding for non-structural (Rep proteins and assembly activat ing protein) and structural proteins (VP1 , VP2, VP3). The AAV genome is flanked by ITRs, which serve as packaging and replication signals. AAV infection biology is mainly based on the interaction of the AAV capsid with the host cell. Specifically, the capsid is interacting with cell surface receptors of the host cell mediating cell entry and intracellular trafficking as well as processing of the virion. AAV has not been as sociated with any disease; rather, discussion arises about being tumor protective. It is unique in being dependent on the presence of a helper virus like adenovirus for progeny production. Further, AAV is able to transduce mitotic and postmitotic tissues and shows a broad tissue tropism. In fact, for in vivo application a major drawback of AAV vectors, in particular, AAV serotype 2-based vectors, is their broad tropism. Since AAV vectors are able to transduce a wide range of cell types, transduction effi cacy of a distinct target organ is lowered, and administration of higher vector doses are required to achieve therapeutic target cell transduction levels.

However, AAV, in particular, AAV2, has gained tremendous popularity as vector for gene therapy, but also as platform for vaccine development and as tool in preclini- cal research. AAV vectors are considered as gene delivery system of choice for in vivo gene therapy. Advantageous features of AAV include its apathogenicity, high stability, ability to transduce both dividing and non-dividing cells, long-term gene ex pression in post mitotic or slowly proliferating cells and low immunogenicity. In addi tion, AAV vectors lack an intrinsic genome integration activity and are therefore de fined as non-integrating vector system. This is a main advantage compared to retro- /lentiviral vectors as it significantly reduces the risk of insertional mutagenesis. Fur ther, vectors can be produced to high titer and purity owing to its stability.

To date, 13 different natural human and non-human primate AAV serotypes and over 100 natural variants have been isolated. Most of the serotypes have been vec torized as alternatives to AAV serotype 2, the prototype AAV vector, because they differ in tropism and in epitopes recognized by the immune system. For vectorization viral genomes are replaced by a transgene cassette. Usage of serotypes is simplified by the pseudopackaging technology allowing to package ITR2-flanked AAV vector genomes into non-AAV2 capsids. This enables the production of AAV vectors differ ing in capsid but delivering the same vector genome. The AAV vector capsid and ge nome can be modified and optimized for efficient and directed gene delivery in vitro and in vivo.

Pseudotyping (pseudopackaging) represents a strategy to change the tropism of the AAV vector system. However, it does not allow to re-direct the tropism towards a distinct cell type. In addition, its potency of increasing efficiency of AAV vector-me diated cell transduction is limited. Therefore, alternative strategies summarized as capsid-engineering technologies have been developed to tailor the host-AAV interac tion including the first step of cell transduction. Consequently, non-perm issive cell types became susceptible towards AAV transduction and transduction efficiencies for permissive cells could be improved. In addition, it became possible to redirect viral vector in vivo tropism towards a defined cell surface structure (BCining, H. et al. , Cur rent Opinion in Pharmacology, 2015, 24, 94-104).

In genetic capsid engineering approaches which focus on cell surface targeting, peptide ligands are inserted into a suitable position of the capsid. This position needs to be exposed on the capsid surface to influence target cell infection properties. To date, peptide insertions for tropism modification are mainly performed by the follow ing techniques: i) fusion to the N-terminus of VP2 or ii) insertion of peptides at the tip of variable region (VR)-VIII, for AAV2: insertion at amino acid position 587 and inser tion at position 588, or insertion at VR-IV, for AAV2: insertion at amino acid position 453 as described in BCining et al., see above; and BCining and Svrivastava, Mol Ther Methods Clin Dev, 2019, 12, 248-265. For example, EP 2 158 21 1 B1 identifies structure protein insertions having a length of 4 to 13 amino acids, which represent ligands mediating cell targeting accordingly. In addition, approaches have been con ducted for substituting specific amino acids present in the capsid proteins.

Since for a lot of target cells or tissues of therapeutic interest suitable peptide ligands that would mediate target cell transduction are unknown, high-throughput screenings of AAV peptide display libraries have been described. For example, librar ies have been disclosed in MCiller, O. J. et al., Nature Biotechnology, 2003, 21 (9), 1040-1046 and Perabo, L. et al., Molecular Therapy, 2003, 8(1 ), 151 -157. The AAV peptide display library established by Perabo et al. is composed of >4 E6 capsid vari ants. The library has been generated by insertion of a random 7-mer peptide into amino acid position 587 (VP1 numbering) of AAV2 capsid proteins. The random pep tide is flanked by linker sequences to help optimal presentation. This library is de scribed for example in EP 1 456 383 B1 and also in EP 2 363 487 A2. Selection of this library on target cells has been used to identify AAV capsid variants with novel and improved transduction properties.

In US 2002/0192823 A1 an AAV capsid variant and methods have been de scribed whereby said AAV capsid variant is derived from a shuffled AAV capsid li brary based on AAV2, AAV8, and AAV9.

Further strategies include attachment of foreign molecules to the outer surface of the vector that can mediate specific interaction with cellular receptors expressed on the membrane of target cell types. Chemical modifications of amino acids present on the surface or on the exterior of the capsid surface have been discussed. These modifications include binding of biotin for allowing coupling of further components, e.g. via streptavidin. Further, introduction of unnatural moieties have been disclosed. For example, WO 2012/149160 A2 describes the strategy for modification of virus capsid proteins based on introducing unnatural amino acid moieties which eventually allows to introduce further groups.

As noted, a main problem of the AAV vector system is the in vivo tropism as well as the transduction efficiency of specific tissue types, thus, requiring the use of high vector doses.

Summary of the present invention

Consequently, it is an object of the present invention to provide AAV vectors and capsid proteins overcoming the problems described in the art, namely, the prob lem of broad tropism and low transduction efficiency.

The present invention relates in a first aspect to a mutated adeno-associated vi rus (AAV) capsid protein or a fragment thereof wherein an insert is inserted after at least one of the amino acids having the amino acid numbers 139, 161 , 261 , 381 , 447, 453, 459, 534, 570, 573, 584, 585, 586, 587, 588, 589 of SEQ ID No. 2 correspond ing to the capsid protein of AAV2 or the homolog capsid protein of the other sero types of AAV, namely, AAV1 , AAV3b to AAV11 , and AAV bovine, the insert com prises an oligonucleotide of at least four amino acids and at most thirty amino acids, like consisting of five to twelve amino acids, in particular, seven amino acids, said in sert may optionally contain further amino acids representing a linker sequence at both sites of said oligopeptide, said linker sequence having independently from one another a size of 1 to 3 amino acids, if present; wherein the mutated AAV capsid pro tein or a fragment thereof or its homologs being more efficient in targeting hepatic tis sue, hepatocytes, hepatic cells and cell lines or hepatocellular carcinoma (HCC) with higher specificity.

In a second aspect, the present invention relates to a mutated AAV particle comprising the AAV capsid protein or fragment thereof according to the present in vention. Further, the present invention relates to a nucleic acid encoding the AAV vi rus capsid protein or fragment thereof according to the present invention. In addition, a nucleic acid vector, in particular, a plasmid comprising the nucleic acid molecules according to the present invention is disclosed as well as host cells containing said nucleic acid vector or said nucleic acid according to the present invention.

Moreover, the use of the AAV particle according to the present invention or the nucleic acid molecule according to the present invention or the nucleic acid vector according to the present invention in the manufacture of an AAV particle, or a manu facture of a medicament for gene therapy. Moreover, the components described herein including the mutated AAV capsid protein or a fragment thereof or a homolog as well as the mutated AAV particle or the nucleic acid molecule or the nucleic acid vector are suitable for use in targeting hepatocytes and/or HCC and/or improving transduction efficiency of AAV vectors in hepatocytes and/or HCC. Further, these components are suitable for use in transfer of a gene of interest into hepatocytes or HCC, in particular, for use in gene therapy.

That is, the present inventors identified new variants of mutated AAV capsid proteins or fragments thereof being more efficient in transducing hepatic tissue, hepatocytes, hepatic cells and cell lines, and HCC, respectively, with higher specific ity and/or efficiency.

Brief description of the drawings

Figure 1 : Mean transgene expression of rAAV vectors in Balb/c mice (A) over the course of 28 days and (B) at day 28. Animals received 4.8E1 1 particles of indi cated AAV vectors by tail vein injection. After 7, 15, and 28 days mice were analyzed for vector transduction. Specifically, animals received 1 .5 mg of D-Luciferin by tail vein injection and were positioned within IVIS ^® instrument. After 5 min images were taken with 5 min exposure. Radiance represents firefly luciferase-induced lumines cence signal intensity measured as photons/second/cm ² /steradiant [p/s/cm ² /sr]. Sta tistics: Error bars show SD. Ordinary Two-way ANOVA and Tukey’s multiple compari sons test were performed. ^* p<0.05; ^** p<0.01 , ^*** p<0.001 , ^**** p<0.00001 ;

Figure 2: Biodistribution of rAAV vector genomes determined by qPCR quanti fication. For determining vector genome content in liver compared to main-off target tissues (LIV=liver; SPL=spleen; LNG=lung; FIRT=heart), total DNA was isolated from animals of the above-described in vivo experiment (Figure 1 ). Indicated DNA sam ples were analyzed by relative qPCR quantification using firefly luciferase as target and mFIPRT as reference gene. The biodistribution is shown as relative transgene content for the different tissue samples. The target gene contents are defined as cTarget/Reference ratios and represent Target/Reference ratios calibrated by an in ter-plate calibrator. Error bars show SD. Mean of rAAV8 relative vector genome con tent in liver tissue was set as one. Statistics: Error bars show SD. Shown signifi cances relate to rAAV2. One-Way ANOVA und Tukeys Multiple Comparison Test:

^** p<0.01 ; ^**** p<0.0001 );

Figure 3: Transduction of primary human and murine hepatocytes with rMLIVI and rMLIV3 compared to parental rAAV2 and hepatotropic rAAV8 controls. Primary human (A) or murine (B) hepatocytes were seeded and subsequently transduced with a vector particle per cell ratio (GOI) of 10000. 72 hrs post transduction, cells were lysed and a luciferase assay was performed to determine level of transduction as means of luciferase activity. Luminescence signal shown in relative luminescence units (RLU). Mean RLUs were normalized to protein content measured by Bradford assay. Statistics: Error bars show SD. One-Way ANOVA und Tukeys Multiple Com parison Test: ^* p<0.05, ^** p<0.01 ; ^*** p<0.001 ; ^**** p<0.0001 );

Figure 4: HCC transduction efficiency of novel AAV2-based capsid variants (rHCCMI , rHCCM2, and rHCCM3) compared to parental rAAV2wt control. A) Repre sentative images of in vivo and in situ imaging of luciferase activity/luminescence sig nal 3 days post vector injection into the TGFa/c-myc HCC mice. B) Quantification of in situ luciferase activity/luminescence of tumor nodules; Animals received 1 .5E1 1 particles of indicated vectors by tail vein injection. Three days post injection, animals were prepared for in vivo imaging. 5 min after D-luciferin injection (1 mg per animal) a series of images was taken with 3 min of exposure time by IVIS ^® . One hour after the first measurement, animals received a second injection of D-luciferin. Animals were sacrificed after 5 min, livers were isolated and then imaged with 3 min of exposure time. Counts represent firefly luciferase-induced luminescence signal intensity meas ured as relative light units (RLU). Signal overlay images (A) generated with a color scale of 50 to 1200 counts. 100 RLU defined as back ground level. Dashed rectangle marks background (BKG) ROI on animal skin; RLU values represent background-cor rected signals. Circles represent ROIs marking tumor nodules. Dashed circle mark background (BKG) ROI on non-tissue area; ROI (BKG) tags show BKG-corrected to tal RLU values measured within ROIs; BKG tags show total RLU values of measured within BKG ROI;

Figure 5: Transduction of human hepatoma and human hepatocyte cell lines by indicated novel capsid variants compared to parental rAAV2wt control. A) HepG2, Huh7 B) and C) Pop10 were transduced with a particle per cell ratio (GOI) of 1000.

24 hrs post transduction, cells were lysed and luciferase assay was performed. Lucif- erase activity as luminescence signal is shown in relative luminescence units (RLU). Mean RLU of rAAV2 was set as one. Statistics: One-Way ANOVA und Tukeys Multi ple Comparison Test: ^* p<0.05, ^** p<0.01 ; ^*** p<0.001 ; ^**** p<0.0001 );

Figure 6: Transduction of human hepatoma cell line with indicated capsid vari ants compared to parental rAAV2wt control. HepG2 were transduced with GOI 1000 and harvested 24 hrs post transduction. Cells were lysed and luciferase assay was performed to detect luciferase activity as luminescence signal shown in relative lumi nescence units (RLU). Mean RLU of rAAV2 (rAAV2wt control) was set as one. Statis tics: One-Way ANOVA und Tukeys Multiple Comparison Test: ^* p<0.05, ^** p<0.01 ;

^*** p<0.001 ; ^**** p<0.0001 ).

Figure 7A-D: VP1 amino acid sequence alignment of AAV serotypes AAV1 , AAV2, AAV3b, AAV4, AAV5, AAV6, AAV7, AAV9, AAV10, AAV1 1 and AAVbovine. I- 587 and I-453 represent preferred homologous insertion sites after amino acid posi tions 587 and 453 (AAV2 VP1 numbering).

Detailed description of the present invention The present inventors identified new mutant AAV capsid protein or fragment thereof being more efficient and partially more specific in targeting hepatic tissue, hepatocytes, hepatic cells and cell lines or hepatocellular carcinoma (HCC).

The vector particles as well as the capsid proteins are characterized in contain ing insertions of an insert after at least one of the amino acids having the amino acid numbers 139, 161 , 261 , 381 , 447, 453, 459, 534, 570, 573, 584, 585, 586, 587, 588, 589 of SEQ ID No. 2 corresponding to the capsid protein of AAV, or the homolog capsid protein of the other serotypes of AAV, namely, AAV1 , AAV3 to AAV1 1 , and AAV bovine, the insert comprises an oligopeptide of at least four amino acids and at most 30 amino acids, like consisting of five to 12 amino acids, in particular, seven amino acids. The insert may optionally contain further amino acids representing a linker sequence at both sites of said oligopeptide, said linker sequence having inde pendently from one another a size of one to three amino acids, if present.

As used herein, the term“fragment thereof” refers to a protein or a polypeptide derived from the mutated adeno-associated virus capsid protein. Typically, the size of the fragment is at least 50 %, like at least 70 %, as 80 %, 90 %, like 95 % of the size or length of the sequence it refers to.

The term“mutated AAV capsid protein” means a capsid protein containing an insert composed of amino acids at the distinct positions identified herein referring to the sequence of SEQ ID No. 2 corresponding to the capsid protein VP1 of AAV2.

The term“homolog capsid protein” refers to the homolog capsid protein VP1 of the other serotypes of AAV, namely, AAV1 (SEQ ID No. 1 ), AAV3 (SEQ ID No. 3), AAV4 (SEQ ID No. 4), AAV5 (SEQ ID No. 5), AAV6 (SEQ ID No. 6), AAV7 (SEQ ID No. 7), AAV8 (SEQ ID No. 8, AAV9 (SEQ ID No. 9), AAV 10 (SEQ ID No. 10), AAV1 1 (SEQ ID No. 1 1 ) and AAV bovine (SEQ ID No. 12). That is, the homologs are serotype capsid proteins derived from the other AAV serotypes. An overview is given in Vance, M. A. et al., DOI: 10.5772/61988. The serotype alignment is shown in figure 7A-D. That is, the homologous insertion sites in other AAV serotypes corresponding to the AAV2 insertion sites can be easily determined by the skilled person,

The term“outer surface of the capsid” identifies the exterior of the capsid, thus, displaying the insert for allowing interaction with an interacting partner. The term “outer surface of the capsid” is used interchangeably with the term“exterior of the capsid”. The term“linker” refers to an amino acid sequence which is optionally present in the mutated AAV capsid protein according to the present invention. Said linker is a linking element between the oligopeptide present in the insert of the mutated AAV capsid protein and the wild type amino acid sequence of the VP1 protein.

The term“particle” or“AAV particle”, identify a capsid either without DNA or containing DNA. The term“vector particle”, or“vector” identify a capsid containing DNA. The term“AAV” refers to the adeno-associated virus or the derivatives thereof including recombinant AAV vector particles wherein the term“AAV wild type particle” designates the adeno-associated virus capsid as it occurs in nature without DNA or containing DNA. The term“recombinant AAV” or“recombinant vector particle” is an AAV wherein the genome of the virus is substituted with a vector genome, namely, a foreign DNA to be introduced into the cell and is also termed“rAAV”. The term“AAV” or the respective recombinant AAV vector particle and AAV wild type particle include the at least 13 different serotypes known in the art. The AAV of human serotype 2 is also mentioned as AAV2 or AAV2 particle or AAV2 vector particle.

The term“targeting” refers to the improvement of specificity for the target tis sues or target cells and/or the improvement of transduction efficiency in the target tis sues or target cells by genetic modification of the AAV capsid. The term“targeting molecule” refers to a molecule allowing targeting of AAV particle as described herein to target cells.

The term“target cells” or“target tissue” as used herein refers to specific cell types or tissues representing the target as defined by the targeting molecule present in the mutated AAV capsid proteins or fragments thereof according to the present in vention. Namely, the target cells or target tissue is hepatic tissue, hepatocytes, he patic cells and cell lines, or hepatocellular carcinoma.

The present inventors recognized that inserting the insert at the distinct posi tions of the capsid protein allows to provide capsid proteins and, eventually, mutated adeno-associated virus particles being more efficient in targeting hepatic tissue, hepatocytes, hepatic cells and cell lines or hepatocellular carcinoma with partially higher specificity.

In an embodiment, the mutated AAV capsid protein, the fragment thereof or a homolog thereof is a mutated AAV capsid protein being derived from AAV type 1 , 2, 3b, 4, 5, 6, 7, 8, 9, 10, 1 1 and bovine. In an embodiment, the mutated AAV capsid protein is derived from the VP1 of serotype AAV2.

In a further embodiment of the present invention, the mutated AAV capsid pro tein, the fragment thereof or a homolog thereof is a protein wherein the insert is at the position following at least one of the amino acids of 261 , 453, 534, 570, 573, 587 and 588. In a preferred embodiment, the position wherein the insert is introduced af ter positions 453 and 587, respectively. That is, the insertion is in between positions 453 and 454 of SEQ ID No. 2 and/or in between 587 and 588 of SEQ ID No. 2. In an embodiment, the insertion is present in between positions 587 and 588 of SEQ ID No. 2.

The sequence of SEQ ID No. 2 corresponds to the encoding sequence of the VP1 protein of AAV2. In another embodiment, the mutated AAV capsid protein, a fragment thereof, or a homolog thereof according to the present invention is a mu tated AAV capsid protein, a fragment thereof, or the homolog thereof comprising at least one further mutation selected from a point mutation, an internal or terminal dele tion, a second insertion and a mutation.

For example, the additional mutation may be a mutation described in the art. Further, the additional mutation may be a mutation as described in undisclosed EP 17193742.8.

The mutated AAV capsid protein according to the present invention were ob tained by screening the library in vivo in two different orthotopic FICC mouse models following a pre-clearing step for variants that bind heparin since this feature corre lates with unspecific binding to a wide variety of different cell types, which express heparan sulfate proteoglycane (FISPG), the AAV2 primary attachment cell surface re ceptor. One of the FICC mouse models was generated by transplanting a murine he patoma cell line (Flepa129) into the liver of C3FI mice (Schmitz et al, J. Flepatol.

2004. 787-797). A further mouse model represents a transgenic mouse model over expressing TGF-a and c-myc regulated by ZnCh-and albumin-inducible promoters, respectively (Flaupenthal J. et al. , Neoplasia, 2012, 14(5), 410-419).

The mutated AAV capsid proteins containing the insertion at the predetermined positions as disclosed herein are characterized in displaying the introduced insert from the capsid at the outer surface of the capsid when the capsid is assembled as an AAV particle. That is, the present inventors aim to improve the AAV vector system with regard to efficiency and selectivity of in vivo gene transfer into hepatocytes and its malignant counterpart, hepatocellular carcinoma cells, based on the mutated AAV capsid proteins according to the present invention. That is, superior transduction effi ciencies of hepatocytes and HCC have been achieved with higher cell entry rates and more efficient vector genome accessibility for transcription. The variants de scribed herein transduced hepatocytes and HCC more efficiently compared to wild type AAV2 or other known mutated capsid strains. In addition, the mutated AAV cap sid proteins according to the present invention are found to be superior regarding ex pression efficiency, i.e. transcripts per vector copy, arguing for a more efficient inter cellular processing again compared to the wild type AAV2.

That is, the displayed insert present in the outer surface of the capsid when the capsid is assembled as an AAV particle increases efficiency of transfection, thus, lower vector doses are required to achieve therapeutic target cell transduction levels in hepatocytes and HCC, respectively.

In an embodiment of the present invention, the mutated AAV capsid protein, the fragment thereof or a homolog thereof according to the present invention are a cap sid protein or a protein fragment or homolog thereof wherein the linker sequences in an amino acid sequence is containing or consisting of amino acids G, S and/or A. In a preferred embodiment, the linker consists of at least one of the amino acids A or S, in particular, having amino acids AAA N-terminally and AA C-terminally or ASA N-ter- minally and AA C-terminally, flanking the amino acid sequence of the oligopeptide in serted at the positions identified herein.

In an embodiment, the insert does not contain any stop codons. That is, it is clear that the mutated AAV capsid protein according to the present invention contains an insert continued with the wild type amino acid present after the amino acid wherein the insert is introduced. In addition, the nucleic acid molecule according to the present invention does not encode a stop codon, thus, stopping translation of the encoded peptide.

In an embodiment, the linker, in particular, the linker mentioned above, are flanking the oligopeptide of the at least five amino acids and at most 12 amino acids, like the seven amino acids described herein.

In an embodiment, the mutated AAV capsid protein, a fragment thereof, or a homolog thereof are a capsid protein, a protein fragment or homolog thereof wherein an oligonucleotide having a size of seven amino acids having a sequence according to any one of SEQ ID No. 13 to 275. In another embodiment, the mutated AAV cap sid protein, a fragment thereof, or a homolog thereof are containing the insert having a sequence of SEQ ID No. 276 to 538. The sequences according to any one of SEQ ID No. 13 to 275 are amino acid sequences showing the insert without any linker group while the insert of any one of SEQ ID No. 276 to 538 show the amino acid se quences of suitable insert to be present as inserts in the mutated AAV capsid pro teins according to the present invention containing linker sequences at both sites of the peptide sequences shown in SEQ ID Nos. 13 to 275.

That is, an embodiment of the present invention is a protein having an insertion in between 587 and 588 of SEQ ID No. 2 having an oligonucleotide of seven amino acids and a linker sequence at both sides of 1 to 3 amino acids. Preferably the se quence is any one of SEQ ID No. 13 to 21 and 276 to 284.

In a further aspect of the present invention, a mutated adeno-associated virus particle, also named vector particle, comprising the mutated adeno-associated virus capsid protein or a fragment thereof, or a homolog thereof according to the present invention is provided. This AAV particle is more efficient in targeting hepatic tissue, hepatocytes, hepatic cells and cell lines or hepatocellular carcinoma with partially higher specificity. That is, the efficiency of ex vivo or in vivo transduction of the target cell or target tissue as well as the applicability of AAV as a delivery tool is increased by increasing the capacity of the mutated AAV particle to transduce the cell with higher specificity and/or efficiency.

For example, this mutated AAV particle is useful in the manufacture of AAV par ticle in the manufacture of a medicament for gene therapy. For example, in case of a medicament for gene therapy, the AAV particle according to the present invention carries a functional nucleic acid fragment or a nucleic acid fragment of a molecule of interest. Said molecule of interest includes a nucleic acid fragment encoding a mole cule of interest. Being more efficient and having partially higher specificity of targeting the target cells or target tissue, here targeting hepatic tissue, hepatocytes, hepatic cells and cell lines, or FICC, allows to increase efficacy of gene therapy as well as vaccination strategies for preventing or treating malignancies of the mentioned tar geted cells and target tissue. In particular, the mutated AAV particle according to the present invention is suit able for use in targeting hepatocytes and/or HCC.

The mutated AAV particle according to the present invention is particularly use ful in the transfer of a gene of interest into hepatocytes or HCC, in particular, for use in gene therapy.

That is, the mutated AAV particles according to the present invention are useful in the expression of therapeutically effective transgenes. For example, metabolic fac tors, like enzymes may be expressed to overcome defects like genetic defects. In case of e.g. HCC the expression of tumor suppressor genes, suicide genes, pro-in flammatory factors, etc. may be envisaged.

Moreover, the mutated AAV particles are suitable to induce immunologic toler ance against transgenes or antigens introduced into the cells. This is possible partic ularly in the target cells and target tissue of the liver. Tolerance may help to treat au toimmune disease.

In a further aspect, the present invention relates to a nucleic acid molecule en coding the mutated AAV capsid protein or fragment thereof, or a homolog thereof ac cording to the present invention. That is, the nucleic acid molecule according to the present invention is a nucleic acid molecule encoding the AAV capsid protein or frag ment thereof as defined herein, for example, containing the oligopeptide of any one of SEQ ID No. 13 to 275 or the insert according to any one of SEQ ID Nos. 276 to 538. Of course, depending on the host cell, a codon optimization may be effected.

A further aspect relates to a nucleic acid vector, in particular, a plasmid com prising the nucleic acid molecule according to the present invention. The skilled per son is well-aware of suitable plasmid and vectors. In particular, based on the host cell to be transfected, a plasmid or vector is selected accordingly. Further the nucleic acid vector may contain further genetic AAV helper elements e.g. the rep open reading frame.

In a further aspect, a host cell containing the nucleic acid vector according to the present invention or the nucleic acid molecule according to the present invention is provided.

The host cell may contain further helper plasmids or nucleic acid molecules containing e.g. the adeno-viral helper genes E4, E2A and VA for the production of the recombinant AAV particles. In an embodiment of the present invention, the host cell is a producer for the production of the AAV particle according to the present in vention.

In a further aspect, the present invention relates to the mutated AAV particle ac cording to the present invention or the nucleic acid molecule according to the present invention or the nucleic acid vector according to the present invention or the compo sition as described herein with suitable transgene DNA components. Said transgene DNA components may be known transgene DNA or DNA encoding suitable peptides or proteins accordingly. In another embodiment, a composition is provided compris ing the mutated AAV particle according to the present invention, e.g. the composition is a pharmaceutical composition for usage in transducing hepatic tissue, hepato- cytes, hepatic cells and cell lines or hepatocellular carcinoma. Further, the present invention relates to the use of the AAV particle according to the present invention, the nucleic acid molecules according to the present invention or the nucleic acid vector according to the present invention in the manufacture of an AAV particle or the man ufacture of a medicament for gene therapy.

The mutated AAV capsid protein, a fragment thereof, or a homolog according to the present invention, the mutated AAV particle according to the present invention, the nucleic acid molecule according to the present invention or the nucleic acid vec tor according to the present invention are particularly useful in targeting hepatocytes and/or HCC as demonstrated herein.

Another aspect of the present invention relates to the mutated AAV capsid pro tein or a fragment thereof or a homolog thereof according to the present invention, the mutated AAV particle according to the present invention, the nucleic acid mole cule according to the present invention or the nucleic acid vector according to the present invention for use in treating diseases involving hepatocytes or for treating HCC: bleeding disorders including Hemophilia A, Hemophilia B, Von Willebrand dis ease, metabolic disorders including Alpha-1 -antitrypsin deficiency, familial hypercho lesterolemia, ornithine transcarbamylase deficiency, crigler-najjar syndrome, acute intermittent porphyria, glycogen storage disease type 1 a, liver cancer including hepa tocellular carcinoma, hepatoblastoma, collangiocarcinoma, autoimmune disorders in cluding multiple sclerosis, latent autoimmune diabetes, induction of tolerance towards allogenic transplants including liver, kidneys, heart, stem cells, surrogate liver-di rected gene therapy including lipoproteinlipase deficiency, diabetes and ornithine transcarbamylase deficiency.

That is, the components described herein including the mutated AAV capsid protein, fragment thereof or homolog thereof, the mutated AAV particle, the nucleic acid molecule or the nucleic acid vector are suitable for use in a method of treating an individual in need thereof. Hence, another aspect relates to a method of treating an individual in need thereof comprising the step of administering a mutated AAV particle according to the present invention or a nucleic acid molecule encoding the mutated AAV virus capsid protein, a fragment thereof, or a homolog thereof accord ing to the present invention, and/or a nucleic acid vector according to the present in vention to an individual in need thereof.

Moreover, administration can be conducted by known means. In particular, ad ministration can be conducted by intravenous route, or injection into the (hepatic) portal vein, or intrahepatically, or intratumorally, respectively.

In particular, the method of therapeutic treatment according to the present in vention is a method for therapeutic treatment of diseases involving hepatocytes or for treating HCC. Examples include bleeding disorders including Hemophilia A, Hemo philia B, Von Willebrand disease, metabolic disorders including Alpha-1 -antitrypsin deficiency, familial hypercholesterolemia, ornithine transcarbamylase deficiency, Crigler-Najjar syndrome, acute intermittent porphyria, glycogen storage disease type 1 a, liver cancer including hepatocellular carcinoma, hepatoblastoma, collangiocarci- noma, autoimmune disorders including multiple sclerosis, latent autoimmune diabe tes, induction of tolerance towards allogenic transplants including liver, kidneys, heart, stem cells, surrogate liver-directed gene therapy including lipoproteinlipase de ficiency, diabetes and ornithine transcarbamylase deficiency.

In an embodiment of the present invention, the method of prophylactic or thera peutic treatment is a gene therapy including the step of administering the AAV parti cle according to the present invention or comprising at least one of the mutated AAV capsid protein or fragment thereof, or homolog thereof according to the present in vention, a nucleic acid molecule according to the present invention and/or a nucleic acid vector according to the present invention , e.g. in form of a pharmaceutical com position. Said pharmaceutical composition or medicament may contain other suitable components including diluents, excipients or carriers.

Further, a method for ex vivo or in vivo transfer of a gene of interest into hepato- cytes or HCC, in particular, by gene therapy, namely, liver directed AAV gene ther apy, is disclosed.

The invention will be described further by way of examples without limiting the same thereto.

Examples

Characterization of MLIV capsid variants

Prior to AAV peptide display library selection the library was pre-cleared from HSPG binding capsid variants by heparin affinity column purification. In the library, the insert is located in between positions 587 and 588 of SEQ ID No 2. Capsid vari ants that accumulated in liver tissue after an in vivo AAV peptide display library se lections in the two different mouse models were named MLIV* library (Hepa129 grafted) and MLIV library (TGF-a/c-myc transgenic), respectively. 92 candidate vari ants for the grafted and 84 candidate variants for the transgenic mouse model have been identified. The most abundant variants of each of the two libraries were named MLIV1 (MLIV*) (SEQ ID No. 13) and MLIV3 (MLIV) (SEQ ID NO. 14) and character ized in detail. For in vivo evaluation, MLIV1 and MLIV3 were produced as recombi nant (r)AAV vectors expressing a luciferase transgene cassette. Transgene expres sion was controlled by the CMV promoter, a ubiquitously expressed promoter. In de tail, 4.8E11 rAAV vector particles were applied to healthy mice via tail vein injection. As control and for comparison, mice received rAAV2 and rAAV8 with natural sero type capsids delivering the same vector genome. 7, 14, and 28 days post injection, mice were injected with D-luciferin (1.5 mg per animal) and the luciferase activity in various body regions was monitored by in vivo imaging using a luminescence signal detecting camera. In all mouse cohorts a significant luciferase activity was detected in the upper abdomen, the liver region, of the animals. The novel capsid variants rMLIVI and rMLIV3 demonstrated a clear and significant improvement in murine liver transduction efficiency (up to 26-fold) compared to rAAV2 (Fig. 1 , A, B). Murine liver transduction efficiency of the new capsid variants is similar to the control vector (rAAV8), which is currently in use in liver-directed human clinical gene therapy trials. Regarding the biodistribution of rMLIVI and rMLIV3 as determined by qPCR anal yses of vector genomes, a clear tropism for murine liver was observed. Specifically, the amount of vector genomes detected in liver was up to 51 -fold higher for the novel capsid variants than in spleen, up to 18-fold higher than in the lung, and 3-fold higher than in the heart (Fig. 2). In particular, the observed de-targeting from murine spleen, one of the main off-target tissues of AAV2 (in the murine off-target tissues: rAAV2 ge nome content > rMLIV capsid variant genome content (182-fold higher in spleen; up to 24-fold higher in lung, 7-fold higher in heart)) (Fig. 2), is an important novel feature of the variants according to the present invention with regard to in vivo gene therapy.

It is well known that the non-human primate serotype rAAV8 shows high trans duction efficiency in liver tissue of mice. Flowever, this does not correlate with the transduction efficiency obtained for human liver. In comparison, human serotype rAAV2 vectors are taken up and processed more efficiently in human hepatocytes.

To investigate the efficiency of the newly developed capsid variants in the context of human hepatic tissue, primary human hepatocytes were transduced with rMLIVI and rMLIV3 (and the controls rAAV2 and rAAV8). In addition, the same vectors were as sayed on primary murine hepatocytes (GOI 1 E4 rAAV vector particles per cell, 72 hrs). For quantification of luciferase expression, luciferase activity was measured in cell lysates by adding luciferin followed by luminescent signal detection (Promega Lu ciferase Assay kit). Here, rAAV2 and the novel capsid variants rMLIVI and rMLIV3 were clearly superior to rAAV8, which is used in human clinical trials (Nathwani 2014 doi: 10.1056/NEJMoa 1407309; Pierce 2018, doi: 10.1 1 1 1 /hae.13489) (Fig. 3, A and B).

Due to the strong liver tropism across species and high hepatic transduction ef ficiency, MLIV1 and MLIV3 represent promising new developments for in vivo gene therapy. Furthermore, the significantly higher efficiency compared to rAAV8 in pri mary human hepatocytes which have been observed following ex vivo transduction and which is currently confirmed in humanized mice are an excellent basis for moving these novel vectors forward as tools for liver-directed human clinical gene therapy tri als. Characterization of HCCM capsid variants

Prior to AAV peptide display library selection the library was pre-cleared from HSPG binding capsid variants by heparin affinity column purification. Capsid variants that accumulated most abundantly in HCC tissue after in vivo AAV peptide display li brary selections in the two different mouse models were named HCCM* (selected in Hepa129 grafted mouse model) and HCCM (selected in TGF-a/c-myc transgenic mouse model). For the HCCM* selection route, before in vivo selection, the AAV pep tide display library was pe-selected in vitro on target cells Hepa129. We identified 103 candidate variants in the grafted and 89 candidate variants in the transgenic mouse model. These capsid variants were further scored with regard to enrichment in HCC (target) compared to liver (main off-target) tissue. The best scored capsid vari ants were ranked regarding enrichment in HCC compared to all analyzed off-target tissues (liver, heart, spleen, skeleton muscle, kidney, and pancreas). The best scored capsid variants were named HCCM1 (SEQ ID No.15), HCCM2 (SEQ ID No.16), and HCCM3 (SEQ ID No.17) (HCCM library) and HCCM*1 (SEQ ID No.18), HCCM*2 (SEQ ID No.19), HCCM*3 (SEQ ID No.20), and HCCM*4 (SEQ ID No.21 ) (HCCM* library). They were produced as rAAV vectors expressing a luciferase reporter transgene cassette for characterization. Expression of the transgene was controlled by the SFFV promoter.

HCCM1 , HCCM2, and HCCM3 were administered to TGF-a/c-myc mice via tail vein injection (1.5E11 rAAV vector particles per animal). For comparison the parental AAV (rAAV2 vector with natural occurring serotype delivering the same vector ge nome) was applied. 3 days post injection, mice were injected with D-luciferin (1 mg per animal) and after 5 min luciferase activity in the different body regions was meas ured by in vivo imaging analysis using a luminescence signal detecting camera (3 min exposure). In addition, from all animals tumor bearing livers were isolated 5 min after repeated administration of luciferin and the luminescence signal was deter mined in situ (3 min exposure). Exclusively, mouse cohorts treated with the novel rHCCM capsid variants demonstrated significant luciferase activity in the tumors, whereas in the rAAV2 cohort luminescence signals remained below background level (100 RLU) (Fig. 4, A, B). The novel capsid variants rHCCMI , rHCCM2, and rHCCM3 demonstrated a clear improvement in HCC transduction efficiency compared to con trol. rAAV2 is very efficient in transducing human hepatoma and hepatocytes cell lines (GOI of 1000 AAV vector particles per cell, 24 hrs). To estimate the efficiency of the newly developed HCCM and HCCM* capsid variants in the context of human he patic tissue, several human hepatoma/hepatocyte cell lines were transduced with the newly developed capsid variants rHCCMI , rHCCM2, rHCCM3, rHCCM*1 , rHCCM*2, rHCCM*3, and rHCCMM as well as the parental control (rAAV2). rAAV vectors ex pressed Renilla luciferase controlled by CMV promoter. Transgene expression level was quantified by luciferase assay (Promega Renilla Luciferase Assay kit). The novel capsid variants rHCCMI , rHCCM2, rHCCM3 (Fig. 5, A to B), rHCCM*2, rHCCM*3, and rHCCMM (Fig. 6) demonstrated transduction efficiencies comparable or even su perior to rAAV2 in these human hepatoma cell lines (Fig. 5, A and B).

To date, treatment of HCC per se and thus also AAV-based gene therapy tar geting HCC is limited by low efficiency. The novel HCCM capsid variants show clearly improved transduction of HCC nodules following intravenous injection and thus represent promising candidates for novel treatment strategies employing the AAV vector system.