WO2024208989A1 - Host cells engineered for lentivirus production - Google Patents

Abstract

1. A method of identifying a lentiviral modulation factor in the genome of a host cell which results in the production of replication-incompetent lentiviral particles at altered yield or infectivity, comprising: a) a genomic perturbation and screening of host cells for modulated expression of one or more genes within a host cell's genome screen, thereby obtaining a pooled repertoire of perturbed host cells to cover perturbation of a diversity of genes; b) utilizing the repertoire of perturbed host cells to produce a lentiviral pool of replication-incompetent lentiviral particles in a perturbed host cell pool culture, where host cells bearing perturbations of lentiviral modulation factors produce lentiviruses with altered yield or infectivity; c) infecting a target cell line with said lentiviral pool and culturing the infected target cells in a target cell pool culture; and d) determining in the target cell pool culture the production of lentiviruses at altered yield or infectivity against the target cells, and identifying by simultaneous perturbation analysis which is a lentiviral modulation factor that results in the production of lentiviruses at altered yield or infectivity against the target cells in a host cell that comprises a respective perturbation for the lentiviral modulation factor, wherein the lentiviruses produced by the perturbed host cells comprise a respective nucleic acid barcode, and the lentiviral modulation factor is identified by determining the relative amounts of the barcodes in the target cell pool culture.

Description

HOST CELLS ENGINEERED FOR LENTIVIRUS PRODUCTION

FIELD OF THE INVENTION

The invention relates to methods of identifying a lentiviral modulation factor in the genome of a host cell and engineering host cells to improve lentivirus production, in particular the production of replication-incompetent lentiviral particles that are used as packaging vectors to deliver a transgene to a target cell genome.

BACKGROUND OF THE INVENTION

The delivery of a therapeutical gene via viruses has proven effective in treating medical disorders. Several viruses can be, due to their specific characteristics, applied for gene therapy: Adenoviruses, adeno-associated viruses (AAV), gamma retroviruses, herpes viruses, retroviruses and lentiviruses. Recently, clinical trials and basic research focused on the usage of AAV and especially lentiviruses (LV). The advantages of lentiviruses are: 1 ) Devoid of viral proteins, 2) replication incompetent, 3) able to transduce non-dividing or slowly growing cells, 4) stable integration of a transgene 5) a broad cell tropism, 6) a low immunogenicity and 7) T cells as native hosts (Ghassemi et al., 2018; Escors et al., 2010). These properties are advantageous in gene therapy as most applications target post-mitotic or immune cells (T-cells). Lentiviruses are currently used in studies treating several diseases such as p-thalassemia, sickle cell anemia, X- linked adrenoleukodystrophy or Wiskott-Aldrich syndrome (Ghassemi et al., 2018).

For decades, the primary options to treat leukaemia or lymphoma was long-term chemotherapy or radiotherapy which burden patients with severe side effects. A novel approach termed chimeric antigen receptor (CAR) - T therapy has the potential to significantly improve the treatment outcomes for affected patients (Ivica and Young et al. 2021 , Albinger et al., 2021 ). Numerous CAR - T therapies are under clinical evaluation and products against B cell malignancies have been approved. CAR-T therapy utilizes patient’s lymphocytes, in particular T-cells which have the ability to detect and eliminate cancerous cells. These T-cells can get genetically modified to express chimeric antigen receptors (CARs), representing synthetic receptors which can re-direct T-cells to recognize and target tumour cells expressing certain target antigens (Sterner et al., 2021 ; Albinger et al., 2021 ). The entire process of CAR-T therapy includes the extraction of blood from a patient, isolation and activation of T-cells, the introduction and integration of the CAR gene into the T-cell genome, expression of the CAR on the T-cells and the expansion and injection of the CAR-T cells into the patient (Hucks & Rheingold et aL, 2019; Albinger et aL, 2021 ; Levine et aL, 2017; Labbe et aL, 2021 ). All marketed CAR-T therapies and most clinical trials utilize lentiviruses as delivery vehicle for the CAR. Whilst CAR-T therapies have had significant positive impact on patients’ health, there are still several shortcomings within this approach such as the generation of large amounts of recombinant lentiviruses at higher titres, CAR-T cell-associated toxicities, antigen escape, poor tumour infiltration or the immunosuppressive microenvironment (Sterner et al., 2021 ).

Currently, lentiviruses are the main choice within CAR - T trials due to their safe integration profiles, broad cell tropism and ability to infect lymphocytes (T-cells) (Levine et aL, 2017; Labbe et aL, 2021 ). Since an enormous number of patient-derived isolated T-cells are required to be genetically modified with a CAR-gene, high titres of lentiviruses need to be the manufactured.

There are published studies that address the lentiviral entry, uncoating and transcription in a pooled fashion (Hein et aL, 2022, Li et aL, 2020, Zhu et aL, 2021 ).

Lentiviruses used for the delivery of genes to a target host cell usually comprise lentiviral packaging safety features, and are typically replication-incompetent.

To obtain recombinant lentiviral particles to deliver a cargo to a target cell, one typically uses a “reverse genetics system” in which several lentiviral components are utilized: A transfer vector comprising a gene of interest, and a series of one or more packaging plasmids containing genes encoding other elements necessary for packaging the lentiviral particles. These plasmids are typically introduced into a production host cell by co-transfection or by electroporation. One area of great concern with lentiviral products is the inadvertent generation of recombinant viruses that are capable of autonomous replication; this is mitigated by alterations to the transfer vector and by distributing the packaging components on one or more distinct plasmids, thus significantly decreasing the chance of viral self-replication and resulting in significantly increased biological safety.

Although the packaging plasmids deliver viral helper proteins (typically Gag, Pol, Env, Rev, in certain cases also Tat) via delivery of the respective genes on separate plasmids, there is a risk that recombination occurs as all the genomic features of the virus are present in the producer cell, albeit it on separate DNA strands. Consequently, the transfer vector is equipped with additional safety features. When choosing lentiviral packaging components, for optimal safety, a most advanced generation packaging mix is preferably used that is compatible with a transfer vector. The standard definitions are: • 2^nd generation systems contain an all-in-one packaging vector expressing the Tat gene.

• 3^rd generation systems do not express Tat, and the packaging is split in two or more plasmids.

• 4^th generation systems express Tat, but are split across many more plasmids to further decrease the risk of recombination inherent to original 2^nd generation mixes.

Wild-type lentiviruses can undergo multiple cycles of infection. However, lentiviral vectors used for CAR-T therapy have been engineered to be “self-inactivating” i.e., they can only undergo a single round of infection. As outlined above, this can be achieved by providing genes as necessary for the virus to be active, such as one or more or all of gag, pol, env, and rev (the viral protein-coding genes) on separate plasmids. Upon infection of the target cell, the virus will lack the respective gag, pol, env and rev, and can thus not assemble for a second round of infection.

In addition, self-inactivation is achieved by truncating the viral 3’ long terminal repeat (3’ LTR). This 3’ LTR is located downstream of a transgene. After the rolling circle amplification of the virus, this 3’ LTR gets copied upstream of the transgene and becomes the new 5’LTR. In the wild-type virus, the 5’LTR functions as a promoter, driving the transcription of the sequence downstream and thus giving rise to the viral mRNA that gets packaged into the virus. In the truncated 3’LTR, this promoter function has been eliminated, thus preventing the expression of a functional viral mRNA upon integration of the virus in the target cell (Zufferey et al., 1998).

Viruses are obligate intracellular parasites and it is well known that the genetic make-up of the host cell has an influence on the viral life cycle. Some host genes are utilized by the virus and the inactivation of these genes will typically abolish the virus infection. Other host genes dampen specific sections of the viral life cycle. Most notably, this includes genes involved in innate immunity towards viral infection and those genes are known as “viral restriction factors”. Such factors were described by Sauter et al., 2021 ; Pagani et al., 2022; Chemudupati et al., 2019; Hatziioannou et al., 2016; Boso et al., 2020).

Han et al. (2021 ) disclose a multi-gene knockout packaging line. Two host restriction factors were identified in HEK293T packaging cells that impeded LV production, 2’-5’-oligoadenylate synthetase 1 (OAS1 ) and the low-density lipoprotein receptor (LDLR). Knocking out these two genes separately led to about 2-fold increases in viral titer. A monoclonal cell line, CRISPRed HEK293T to Disrupt Antiviral Response (CHEDAR), was produced by successively knocking out OAS1 , LDLR, and PKR, another factor impeding LV titers. In addition, transcription elongation factors, SPT4 and SPT5, were overexpressed during packaging.

Dobson et al. (2022) disclose receptor-antigen pairing by targeted retroviruses, which combines viral pseudotyping and molecular engineering approaches to enable one-pot library-on-library interaction screens by displaying antigens on the surface of lentiviruses and encoding their identity in the viral genome. Antigen-specific viral infection of cell lines expressing human T or B cell receptors allows readout of both antigen and receptor identities via single-cell sequencing.

Hain et al. (2021 ), Adamson et al. (2018), and Piccioni et al. (2018) disclose CRISPR-based genetic screens and single-cell RNA sequencing (Perturb-seq). The phenotype of interest is determined on a single cell basis, not in a pool of infected target cells.

OhAinle et al. (2018) disclose a CRISPR/Cas9-mediated knockout screen in which the level of HIV virus itself serves to report levels of infection to identify genes important for HIV infection.

There is a need for improved LV production, in particular where replicationincompetent LV was to be produced at high yield.

SUMMARY OF THE INVENTION

It is the objective to provide engineered host cells for improved LV production, in particular replication-incompetent LV which can be used as a transfer vector to introduce a gene of interest into a target cell. It is a further object to improve a screening method for lentiviral modulation or restriction factors in the genome of host cells, which would be the target for modifying the host cell’s genome, aiming to alter LV production capacities by the host cell.

The object is solved by the subject matter as claimed and as further described herein.

The invention is based on a novel approach to uncover host cell factors that govern production and secretion of lentiviruses. An exemplary CRISPR-Cas9 screen has been developed that is based on two rounds of lentiviral infection: In the first round, a replication-incompetent lentivirus was used which harbors a variety of sgRNAs to infect a host cell harboring Cas9, thus establishing perturbations for the variety of genes that the individual sgRNAs targets. In a second round, the viral mRNA is reactivated from the genome, such that it produces a virus harboring the sgRNA that was present in the producer cell and being capable of infecting an indicator cell or target cell in the second round. If a gene knockout enhances lentivirus production or secretion, more of the respective sgRNA is found in the second infection round. For example, the amount of respective sgRNAs can be assessed by “counting sgRNAs” i.e., by PCR amplification of the sgRNAs and next generation sequencing.

This method described herein particularly enables pooled screens for genes regulating the translation, protein processing, packaging, assembly and budding of lentiviruses. Specifically, the pool screening is by simultaneous perturbation analysis. Specifically, the pool screening is a functional screen for infectious LV.

It turned out that the pooled screening could be conveniently used with replication-incompetent lentivirus. These typically have truncated 3’LTRs as a safety feature which are incompatible with a second round of infection. To alleviate this, a modified lentivirus bearing a wild-type 3’LTR (WT-LTR) was used, which following the first round of infection becomes the new 5’ WT-LTR.

Such a modified lentivirus can be reactivated from the host cell genome by providing packaging plasmids comprising genes as necessary to produce active virus, such as one or more or all of the genes selected from gag, pel, env, rev, and tat, thus providing for the second round of infection by the activated LV.

Specifically, tat is provided alongside in trans. Of note, tat can drive the transcription of the 5’WT-LTR to activate the replication-incompetent LV.

According to specific examples, knockout/s (KO) of host cell genes have been successfully produced which modulate LV production. Specifically, the LV produced by such KO cells have proven to be infective and capable of infecting target cells. Such infection of target cells was indicative of the LV which is functional as a transfer vector when replication-incompetent.

The invention provides for a method of identifying a lentiviral modulation factor in the genome of a host cell which, upon its perturbation in the genome of the host cell, results in the production of replication-incompetent lentiviral particles at altered yield or infectivity, comprising: a) a genomic perturbation screen, thereby obtaining a pooled repertoire of perturbed host cells to cover perturbation of a diversity of genes; b) utilizing the repertoire of perturbed host cells to produce a lentiviral pool of replication-incompetent lentiviral particles in a perturbed host cell pool culture, where host cells bearing perturbations of lentiviral modulation factors produce lentiviruses with altered yield or infectivity; c) infecting a target cell line with said lentiviral pool and culturing the infected target cells in a target cell pool culture; and d) identifying a lentiviral modulation factor which, upon the respective perturbation of the host cell, has resulted in the production of lentiviruses at altered yield or infectivity against the target cells, wherein the lentiviruses produced by the perturbed host cells comprise a respective barcode, and the lentiviral modulation factor is identified by determining the relative amount of the barcodes in the target cell pool culture.

The invention particularly provides for a method of identifying a lentiviral modulation factor in the genome of a host cell which results in the production of replication-incompetent lentiviral particles at altered yield or infectivity, comprising: a) a genomic perturbation and screening of host cells for modulated expression of one or more genes within a host cell’s genome screen, thereby obtaining a pooled repertoire of perturbed host cells to cover perturbation of a diversity of genes; b) utilizing the repertoire of perturbed host cells to produce a lentiviral pool of replication-incompetent lentiviral particles in a perturbed host cell pool culture, where host cells bearing perturbations of lentiviral modulation factors produce lentiviruses with altered yield or infectivity; c) infecting a target cell line with said lentiviral pool and culturing the infected target cells in a target cell pool culture; and d) determining in the target cell pool culture the production of lentiviruses at altered yield or infectivity against the target cells, and identifying by simultaneous perturbation analysis which is a lentiviral modulation factor that results in the production of lentiviruses at altered yield or infectivity against the target cells in a host cell that comprises a respective perturbation for the lentiviral modulation factor, wherein the lentiviruses produced by the perturbed host cells comprise a respective nucleic acid barcode, and the lentiviral modulation factor is identified by determining the relative amounts of the barcodes in the target cell pool culture.

Specifically, production of lentiviruses at an altered yield or infectivity against the target cells is determined as compared to a host cell without the respective perturbation for the lentiviral modulation factor. Specifically, where a gene is perturbed in the host cell and causes a higher yield or infectivity of lentivirus, such gene qualifies as a lentiviral modulation factor.

Specifically, a perturbed host cell is understood as a host cell that is engineered for modulated expression of one or more genes within the host cell genome, also understood as genomic modulation. Such genomic modulation may comprise transcriptional or translational upregulation or downregulation of one or more genes. It may also include the deletion of a gene or its knockout e.g., by introducing a small insertion or deletion into the coding sequence of the gene.

Specifically, the perturbation is directed to the host cell’s genome, in particular the genome of a wild-type host cell.

Specifically, genomic perturbation screens examine the cellular changes following an intended genetic perturbation. Such screens have become a widely-utilized method for investigating gene functions and molecular mechanisms. The present screen is performed with cells in a pooled culture which has advantages over arrayed screens where physically separated (single) cells are individually perturbed with distinct guide RNAs (delivered synthetically or by lentiviral transduction) which are laborious and non- scalable. The pooled method allows for the simultaneous perturbation analysis of thousands of genes.

Specifically, the repertoire of perturbed host cells is produced by the genomic perturbation screen, which is provided in a first pool, in particular in a first cell culture containment.

Specifically, in a first step, the repertoire of perturbed host cells is created by infecting a host cell line with a lentiviral pool, thus giving rise to a first pool of cells in which each cell contains a different perturbation of a host cell gene. Perturbations can be marked with molecular barcodes.

Specifically, each of the different perturbated genes and/or each of the genes that are differently perturbed, produces lentivirus comprising a different molecular or nucleic acid barcode. A high amount of a barcode (e.g., an amount that is higher than the amount of another barcode) may indicate which is the perturbation that results in a high amount of infectious lentiviral particles. Specifically, by determining the amounts of different barcodes (such as relative amounts e.g., in relation to each other) in a pool of target cells that have been infected with differently barcoded infectious LV, each indicating a perturbation of different gene and/or a different perturbation of a gene, will indicate which of the perturbed genes is a lentiviral modulation factor. In some embodiments, the sgRNA introducing the perturbation can serve as the molecular barcode.

Specifically, the repertoire of perturbed host cells is cultured in vitro in a pool culture, under conditions that lead to the differential expression the respective gene of interest (GOI) that was targeted by the genomic perturbation. Following introduction of one or more genes as necessary to activate the LV, such as one or more or all of gag, pol, env, rev, and tate.g., gag, pol, env, and rev, alongside with tat, a pool of recombinant lentiviruses is produced from the perturbed host cells.

Specifically, the target cell line is infected with the lentiviral pool and cultured in a second pool, which is the target cell pool culture, in particular in a second cell culture containment that is preferably different from the first cell culture containment.

Specifically, the lentiviral pool is harvested from the perturbed host cell pool culture, in particular by separating the cell culture supernatant from the host cells. The harvested supernatant contains the lentiviral pool which can be prepared for infecting the target cell line. A method of preparing the lentiviral pool may be used which provides for reactivation of the replication-incompetent LV, thereby providing a reactivated lentiviral pool.

The lentiviral pool may then be used for infecting the target cell line, thereby producing a number of lentiviral infection products in the target cell pool culture. Specific lentiviral infection products may comprise the barcode that was used with a lentivirus.

According to a specific aspect, differently engineered host cells of said repertoire of perturbed host cells are cultured to produce LV in the first pool, and the lentiviral modulation factor is identified by identifying which genomic modulation in the first pool has resulted in differential amounts of barcodes in the second pool.

According to a specific aspect, the lentiviral modulation factor is identified by determining the relative amounts of the barcodes in the target cell pool culture e.g., upon genomic DNA extraction of the target cell pool culture. Alternatively, the relative amounts of the barcodes can be determined in a cellular fraction or the supernatant of the target cell pool culture. Specifically, the relative amounts of the barcodes can be determined by PCR amplification and next generation sequencing.

According to a specific aspect, a relative increase of a barcode is indicative of the respective lentiviral modulation factor, whose perturbation in the production host cell has caused a higher yield or infectivity of lentivirus. Specifically, the yield or infectivity of lentivirus is higher than a reference level, which is produced by such production host cell without such perturbation.

Specifically, the yield or infectivity of lentivirus is higher than a reference level, which is produced by such production host cell without the respective perturbation of the host cell for the identified lentiviral modulation factor. Specifically, where a gene is perturbed in the host cell and causes a higher yield or infectivity of lentivirus, such gene qualifies as a lentiviral modulation factor.

Specifically, the host cell bearing the perturbation of the identified lentiviral modulation factor produces lentiviruses with an increased yield and/or infectivity, such as e.g., at least any one of 1 .2, 1 .3, 1 .4, 1 .5, 1 .6, 1 .7, 1 .8, 1 .9, or 2-fold increased yield or infectivity compared to the host cell without such perturbation.

Specifically, the yield of lentivirus production is increased when producing the lentivirus from a production host cell bearing one or more e.g., only one (a “single”) knockout of a lentiviral modulation factor, as compared to the host cell without such knockout.

Specifically, the barcode is a molecular or nucleic acid barcode.

The barcode amount can be determined in the target cell pool culture e.g., directly in the target pool cell culture, or in a fraction of the target pool cell culture such as in the cell culture supernatant, or a cell culture fraction comprising the barcode. For example, it can be determined in infected cells. It can likewise be determined in the cell culture supernatant of the first pool of perturbed cells, in which case, the infection of the target cells is not required.

Lentiviruses which comprise a particular barcode may or may not be produced by a host cell with altered yield and/or infectivity, depending on the genomic modulation of the perturbed host cell. Depending on the yield and/or infectivity of the LV, the amount of the respective barcode of the LV may differ. Therefore, the amount of the respective barcode is indicative of the modulated LV production in the perturbed host cell.

The barcode as used herein may indicate which LV originates from which host cell. Therefore, the amount of a barcode in the target cell or target cell culture is a measure of the effect of the genomic perturbation in the host cell on producing the LV, in particular infectious LV. In the respective genomic perturbation screen, the amount of a number of different barcodes can be determined in the target cell or target cell culture, and relative amounts of the barcodes (relative to each other, or relative to a reference) indicate which genomic modulation was more or less effective in altering the LV yield and/or infectivity.

Specifically, the determination of barcodes in the target cell pool culture may result in different amounts of a variety of barcodes, and the relative amounts of barcodes to each other determines which perturbed host cell has produced the respective LV with altered yield and/or infectivity.

Specifically, a barcode reader is used to identify which genomic perturbation in the repertoire of perturbed host cells results in the production of the respective LV with altered yield and/or infectivity.

Specifically, the barcode can be incorporated within a larger entity, such as fused, conjugated or otherwise integrated within a larger entity. Specifically, the barcode is a nucleotide sequence that is comprised in or part of an oligonucleotide or polynucleotide.

The amounts of the barcodes in the target cell pool culture can be determined in a semiquantitative or quantitative manner. For example, a nucleic acid barcode such as a barcode consisting of an oligonucleotide or polynucleotide, can be determined employing any one or more of the following methods: polymerase chain reaction amplification (PCR), sanger sequencing, next-generation sequencing (NGS), quantitative PCR readout (qPCR) or hybridization of a probe with the barcode. Specifically, quantitative real-time PCR, hybridization approaches, microarrays, NanoString or PCR and next generation sequencing may be used.

The relative amounts of barcodes can e.g., be determined by comparing the amounts of individual barcodes and ranking according to their relative abundance (either relative to one another, or relative to a reference). A reference could be a barcode that is produced from unperturbed cells. Specifically, a reference barcode could be a guide RNA that is non-targeting (e.g., which has no complementary sequence in the host cell genome), or that targets a gene desert whose perturbation will not have any impact on the lentiviral life cycle.

Specifically, the relative amounts of barcodes can be determined by comparing to a reference level. For example, a barcode amount is compared to the respective control barcode in a cell that is not engineered by said genomic modulation, or that is engineered for targeting a region of the genome used as a negative control reference.

For example, to provide a control, wild-type host cells can be used in the genomic perturbation screen, and a LV that comprises a control barcode is produced by the wildtype host cells. The amount of the control barcode can then serve as a control amount to determine whether an LV produced by a perturbed host cell and the respective barcode amount has been increased or reduced in the target cell pool culture, as compared to the control amount.

According to a specific aspect, the genomic perturbation screen comprises an RNA-guided programmable nuclease (RPN)-mediated perturbation screen using a lentivirus harboring a library of single guide RNAs (sgRNAs), which are designed to target the diversity of genes in the host cell genome, wherein the repertoire of perturbed host cells is created using host cells containing the RPN, and wherein the lentiviral modulation factor is identified by determining a relative alteration of the respective sgRNA or its coding sequence, which is used as the barcode, compared to other sgRNAs in the target cell pool culture.

Specifically, the host cells are expressing the RPN, thereby containing the RPN.

Specifically, the host cells can be transduced with the RPN delivered as a protein, thereby containing the RPN.

Specifically, a repertoire of host cells expressing RPN is produced in which each host cell comprises a different genomic modulation such as targeting different genes that are endogenous to the host cell. According to specific examples, the repertoire of host cells is engineered to express an RPN and each host cell bears a knock-out for a different human gene.

According to a specific aspect, the genomic modulation in the perturbation screen is performed by manipulating the cells using CRISPR or other techniques.

Specifically, the RPN is a CRISPR enzyme.

A variety of different Cas nucleases, such as Cas9 (from Streptococcus pyogenes), Cas14, CasX, CasY, Cas12a, Cas13a, Cas13b, Cas13d, Cas14a, etc. can be used. Variant forms of such Cas nucleases are also contemplated, e.g., High-Fidelity Cas9, eSpCas9, SpCas9-HF1 , HypaCas9, Fokl-Fused dCas9, xCas9, dCas9, etc.

Specifically, the RPN is a CRISPR-Cas complex, preferably selected from the group consisting of a Cas9, Cas12a, Cas12f.

Specifically, the RPN is a zinc finger nuclease (ZFN), or a TAL-effector nuclease (TALEN).

Specifically, the programmable nuclease is catalytically inactive.

According to a specific aspect, the RPN-mediated perturbation screen is a CRISPR-Cas9 screen, preferably a CRISPR knockout, CRISPR activation, or CRISPR interference screen. Specifically, CRISPRi or CRISPRa methods are used.

CRISPR interference (CRISPRi) and activation (CRISPRa) are widely used to modulate gene expression at the transcriptional level. For the purpose of CRISPRi or CRISPRa, guide RNA designs in proximity to the gene’s promoter region or the transcriptional start site (TSS) are used, to result in transcriptional gene silencing or activation, respectively. (Gilbert et al. 2013).

Specifically, the perturbation screen employs CRISPR perturbation by e.g., the CRISPRi or CRISPRa modality.

Targeting the transcriptional start site (TSS) or the promoter region of a gene with a dCas9 repressor fusion leads to the downregulation of the cognate gene and is referred to as CRISPR interference (CRISPRi). Conversely, targeting of the transcriptional start site (TSS) or the promoter region of a gene with a dCas9 activator fusion leads to the upregulation of the cognate gene and is referred to as CRISPR activation (CRISPRa).

According to specific examples, the method described herein comprises a gene perturbation screen (e.g., CRISPRi screen, or CRISPRa screens) with a readout which comprises determining LV infection products, in particular an LV-specific barcode in the target cell pool culture. CRISPR screening can be combined with sgRNA sequencing, directly linking sgRNA expression to LV production by individual cells in the pool.

In some embodiments, a CRISPR-Cas system is used employing a targeted repressor or activator. In some embodiments, a targeted repressor and/or activator system is used comprising a CRISPR enzyme such as a Cas9 polypeptide, which is engineered to reduce (or inactivate) nuclease activity, and one or more guide RNAs (gRNA) that bind to a target region.

A Cas9 protein can be mutated so that the nuclease activity is reduced or inactivated. An inactivated Cas9 protein (“iCas9”, also referred to as “dCas9”) with no endonuclease activity has been targeted to genes in bacteria, yeast, and human cells by gRNAs to silence gene expression through steric hindrance. Exemplary mutations with reference to the S. pyogenes Cas9 sequence (UniProtKB - Q99ZW2) to inactivate the nuclease activity include: D10A, E762A, H840A, N854A, N-863A, and/or D986A. Exemplary mutations with reference to the S. aureus Cas9 sequence (UniProtKB - J7RUA5) to inactivate the nuclease activity include D10A and N580A. In certain embodiments, the Cas9 protein is a mutant S. aureus Cas9 protein. In some embodiments, CRISPRi can be based on Acidaminococcus sp. (strain BV3L6) Cas12a/ Cpf1 (UniProtKB - U2UMQ6). Exemplary mutations with reference to the Acidaminococcus sp. (strain BV3L6) Cas12a sequence include D908A.

According to a specific aspect, a transcriptional effector can be used as a modulator of gene expression, also referred to as a transcriptional modulator. Specifically, transcriptional modulators are chosen based on their ability to further repress, or alternatively, to activate the expression of a GOI.

In some embodiments, the present disclosure teaches tethering or translationally fusing a transcriptional modulator with the RPN, e.g., through the use of a fusion construct.

In some embodiments, a fusion protein can be used which comprises an RPN fused to a transcription modulating domain e.g., a fusion which is any one of a dCas9- KRAB (repressor), dCas9-VP64 (activator), dCas9-VPR, VP64-dCas9-VP64, dCas9- p300, or dCas9-Tet1 c, dCas9-EZH2 or dCas9-DNMT3A (as reviewed in Xie et al., 2018). Regulatory factors appended to dCas9 may affect histone methylation or acetylation, DNA methylation or heterochromatin status. They may also aid in the recruitment of active transcription factors. In a specific embodiment, CRISPR-mediated gene silencing may occur using the CRISPRoff approach (Nunez et al., 2021 ).

Specifically, the target of modulation can be a non-coding region which is a regulatory region which controls the transcription of a gene of interest (GOI). Exemplary target sites are regulatory regions such as promoter sequences, terminators, translational regulatory sequences such as ribosome binding sites and internal ribosome entry sites, enhancers, silencers, insulators, boundary elements, replication origins, matrix attachment sites and locus control regions. Specifically, the target site is in proximity to the GOI’s promoter region or the transcriptional start site (TSS) of the promoter.

According to a specific aspect, a transcription modulating domain can be used, which is: a) a transcriptional repressor peptide for gene silencing, preferably a Kriippel- associated box (KRAB) domain, such as selected from the group consisting of ZIM3, or the KRAB domain of KOX1 ; or b) a transcriptional effector for gene activation, preferably transcriptional activation domain, such as selected from the group consisting of VP64, VPR (VP64, p65 and Rta), the SAM domain (consisting of p65 and HS1 ), the SunTag, the p300/CBP histone acetyl transferase, or SPH. A non-limiting list of the transcriptional repressors compatible with the presently disclosed methods include KRAB domains (Margolin et al., 1994; Wolf et al., 2015).

Fusion constructs may generally be prepared using standard techniques. For example, DNA sequences encoding the peptide components may be assembled separately, and ligated into an appropriate construct. The ligated DNA sequences are operably linked to suitable transcriptional or translational regulatory elements. The 5' or 3' end of the DNA sequence encoding one peptide component is ligated, with or without a peptide linker, to the 3' or 5' end, respectively, of a DNA sequence encoding the second peptide component so that the reading frames of the sequences are in frame. This permits translation into a single fusion protein that retains the biological activity of both component peptides.

In some embodiments, a CRISPR enzyme or fusion enzyme may be expressed recombinantly and purified and introduced into the cells by electroporation. Specifically, the sgRNA may be delivered by a lentivirus. In some embodiments, the CRISPR enzyme may be encoded by a lentiviral vector and may be introduced into the target cell using lentiviral infection.

Specifically, the repertoire of perturbed host cells is cultured in vitro in a pool culture, under conditions to differentially express the respective GOI that was targeted by the genomic perturbation, and to produce lentiviruses from the perturbed host cells.

In certain aspects, nucleic acids are introduced into cells, which can be used to modify the genetic material of a cell, for example, its genome. Techniques such as CRISPR or other related techniques may be used to modify the genetic material of the cell, e.g., as guided by the nucleic acids. This may allow, in some embodiments, for the accurate identification of genetic manipulations of the cells, and their corresponding phenotypes, using barcodes as identification portions to identify the genotypes that lead to the observed phenotypes. For example, a nucleic acid that is delivered to a cell may include a guide portion and a barcode. The guide portion may contain sgRNA or another recognition sequence that can be used to recognize a target site, e.g., within the genome of a host cell. The barcode may include one or more sequences that can be used to distinguish various nucleic acids containing different guide portions from each other. For example, the identification portion may include one or more barcode sequences that can be read using a corresponding nucleic acid probe (e.g., a “barcode probe”).

Specifically, a guide portion may include a sequence, such as an RNA sequence, that recognizes a target region of interest, e.g., on genomic DNA of the host cell. In some cases, the guide portion may also include a binding sequence, such as a Cas binding sequence, that Cas or another nuclease is able to recognize. For instance, in certain cases, the guide portion may be suitable for allowing CRISPR editing of the genome to occur. For example, the guide portion may include an sgRNA. In some embodiments, the sgRNA may include a crispr RNA portion (crRNA), which is a sequence complementary to a target sequence (e.g., to a target DNA), and a tracrRNA portion, which the Cas nuclease, or another nuclease, can recognize. In some cases, the crRNA portion may have 17, 18, 19, or 20 nucleotides.

CRISPR and related techniques, and kits useful for conducting CRISPR experiments are readily available commercially.

Specifically, a library of sgRNAs is delivered to the host cells, such that individual cells receive one or more single sgRNAs. According to a specific aspect, each cell of a repertoire receives a different sgRNA, or a different set of sgRNAs.

A library of sgRNA nucleic acids may be prepared, e.g., having different crRNA portions, e.g., for binding to different target sequences in a genome. In certain cases, there may be at least 10, at least 10², at least 10³, at least 10⁴, at least 10⁵, etc. possibilities for the guide portion, e.g., able to bind to different target sites within a genome, and/or able to cause different changes or manipulations of the genome, etc. Thus, a plurality of distinguishable nucleic acids may be prepared using one or more identification portions (such as barcodes described herein) and one or more guide portions.

Specifically, the cells are annotated with sgRNA-specific barcodes or directly detectable sgRNAs.

Specifically, the sgRNA comprises a barcode which is a unique identifiable nucleotide sequence that is indicative for the respective perturbation of the host cell induced by the sgRNA. In such case, the sgRNA-barcode can be conveniently determined by sequencing the sgRNA sequence.

Specifically, the sgRNA sequence itself is a unique identifiable nucleotide sequence and can thus be used as a sgRNA-barcode.

The amount of the sgRNA-barcode within the target cell pool culture can be conveniently determined by any method of determining RNA or a corresponding nucleic acid sequence, such as e.g., described herein.

For example, a sequencing analysis of the sgRNA-barcode library can be performed, which allows for a quantitative assessment of the sgRNA-barcode library in the target cell pool culture. By comparing the read counts of the sgRNA-barcode library to a known standard, such as a reference library or a control sample, the absolute amount of the sgRNA-barcode in the target cell pool culture can be estimated.

According to a specific aspect, the lentiviral modulation factor is a wild-type gene that is endogenous to the host cell.

Specifically, the lentiviral modulation factor is modulating attachment, entry, uncoating, fusion, reverse transcription, integration, transcription, translation, packaging, trafficking, assembly, secretion or budding off of the lentivirus.

Specifically, such lentiviral modulation factor is targeted by a genomic modulation of a host cell which results in improved LV production e.g., increased yield and/or increased infectivity of the LV. Specifically, said lentiviral modulation factor is directly or indirectly targeted by said genomic modulation.

According to a specific aspect, the method comprises at least two or at least three cycles of lentiviral infection, wherein a) in a first cycle, said repertoire of perturbed host cells is generated by infection with replication-incompetent lentiviral particles in pooled culture, thereby producing a pool of host cells harboring lentiviral donor templates; b) said pool of host cells harboring lentiviral donor templates produced by the first cycle is activated, thereby obtaining said lentiviral pool of activated lentivirus particles; and c) in a further cycle, the target cells are infected with the lentiviral pool of activated lentivirus particles in the target cell pool culture.

Specifically, a library of replication-incompetent lentivirus is used for genomic modulation of the host cells and producing the repertoire of perturbed host cells. Upon infection with the lentivirus, the host cells are cultivated in a pool culture and produce LV with altered yield and/or infectivity.

Specifically, the pool of host cells harbors lentiviral donor templates.

Specifically, a lentiviral donor template is typically confined by long terminal repeat sequences which determine the transcriptional start (at the 5’LTR) and the transcriptional termination (at the 3’LTR). In addition, conveniently, a lentiviral donor template harbors key elements, including

(i) a cargo, in particular a cargo that is meant to be delivered by the lentivirus e.g., a barcode or an sgRNA, (ii) a central polypurine tract (cPPT) sequence, in particular a cPPT sequence which is required for proviral DNA synthesis, such as e.g., a cPPT sequence of or derived from HIV-1 ,

(iii) a Psi sequence, in particular a Psi sequence which is required for packaging of the template via the nucleocapsid, such as e.g., a Psi sequence of or derived from HIV-1 , and

(iv) a rev response element (RRE), in particular an RRE which can be bound by the rev protein to initiate nuclear export, such as e.g., an RRE of or derived from HIV-1 .

In a specific embodiment, the lentiviral donor template includes a molecular barcode that gets packaged during lentiviral production and allows the identification of the perturbed host cell based on the barcode. In a specific embodiment, the 3’LTR is not truncated and corresponds to the wild-type LTR (LTR), thus enabling a second round of infection following the provision of one or more genes as necessary for lentiviral activation, such as any one or more or all viral genes selected from gag, pel, env, rev and tat, in particular one or more of gag, pel, env, and rev, alongside with tat in trans.

According to a specific aspect, said repertoire of perturbed host cells harboring lentiviral donor templates is bearing a variety of perturbations introduced by an RPN and a pooled library of sgRNAs, which are designed to target the diversity of genes in the host cell genome.

According to a specific aspect, the replication-incompetent lentiviral particles are assembled by utilizing a lentiviral donor template harboring a wild-type intact 3’ long terminal repeat (LTR) to exchange for a mutated or truncated 3’ LTR as previously used in replication-incompetent lentiviral particles.

Specifically, a replication-incompetent lentivirus of the second or third generation is used in which a lentiviral donor plasmid is used which harbors an intact wild-type LTR at the 3’end and a guide RNA expression cassette flanked by the 5’LTR and the 3’LTR.

Specifically, the lentiviral donor template is activated, thereby activating the replication-incompetent lentivirus.

Specifically, the replication-incompetent lentivirus is activated to render it active (i.e., replication-competent).

Specifically, the replication-incompetent lentivirus is activated by adding or delivering all elements required for activation, such as e.g., viral genes as necessary to produce the virus. For example, one or more or all viral genes selected from gag, pel, env, rev, and tat can be provided to the replication-incompetent lentivirus, and thereby activate the virus to be replication-competent.

Said one or more viral genes can be provided by one or more separate helper plasmids or packaging plasmids.

Said one or more viral genes may originate from a lentivirus such as HIV-1 , or from any other corresponding or compatible gene of another virus, including wild-type genes, homologs or functional mutants thereof, such as selected from genes as used in pseudotype retroviral or lentiviral vectors e.g., originating from vesicular stomatitis virus (VSV). For example, the Env protein may be VSV-G, a glycoprotein of VSV, and env is a nucleotide sequence encoding Env.

Specifically, tat can drive the transcription of the 5’WT-LTR to activate the replication-incompetent LV.

Specifically, the replication-incompetent lentivirus is activated, thereby providing for the second round of infection.

Specifically, activation is by the exogenous addition, expressing or otherwise delivering of one or more of the respective proteins encoded by said viral genes, in particular one or more proteins selected from Gag, Pol, Env, Rev, and Tat.

According to a specific example, for said activation of the lentiviral donor templates, gag, pel, and env are delivered to the pool of host cells in the presence of tat, which host cells are harboring the lentiviral donor templates, preferably wherein said gag, pel, and env are delivered on two or more separate plasmids, and said tat is provided in trans.

The transactivator Tat interacts with its cognate transactivation response RNA structure (TAR) to activate transcription of the lentiviral donor template from the upstream LTR promoter if said LTR is a wild type LTR.

Another possibility to activate the lentiviral donor templates, if the LTR is truncated, is via the implementation of a short promoter into the LTR (Dobson et al., 2022).

A promoter exchange for a heterologous promoter can be made within the viral donor cassette or a lentiviral donor plasmid, to improve the reactivation process.

According to a specific aspect, the lentiviral donor template comprises a viral donor sequence which harbors a weak heterologous promoter.

According to a specific aspect, a weak promoter is included in the viral donor sequence to drive expression of a reporter gene such as GFP or Puromycin resistance gene, in addition to the wild-type LTR promoter. It turned out that a weak promoter used instead of a strong promoter (e.g., EF1 A promoter) displays increased viral transcription from the 5’WT-LTR and, consequently, increases re-activation efficiency. The likely explanation for this observation is promoter competition: The promoter which resides within the viral donor sequence may compete with the 5’WT-LTR for the cellular machinery initiating and maintaining transcription. Use of a weaker internal promoter may free some of the transcriptional machinery so that it can be recruited to the 5’WT- LTR.

Specifically, a heterologous promoter is introduced in the viral donor sequence, preferably between the 5’LTR and the 3’LTR.

Specifically, the heterologous promoter is a weak promoter.

The strength or weakness of a promoter is specifically understood as its ability to recruit transcription factors, the RNA polymerase II complex, and/or relevant cofactors, with the goal to drive strong or weak transcription of the transgene that lies downstream. For example, weakness of a promoter can be determined by fusing the promoter to a reporter gene, such as luciferase or GFP, and assessing its strength based on the expression of the reporter (Qin et al., 2010). The latter can be assessed by using a chemiluminescence-based reporter assay, flow cytometry or quantitative RT-PCR. Where the strength is less than a 5000 FITC-MFI signals determined in a quantitated GFP intensity readout via flow cytometry, the promoter is understood to be a weak promoter (Qin et al., 2010).

Specifically, the heterologous promoter is a weaker promoter than the native wildtype LTR promoter, which is under the control of the TAT transactivator.

Specifically, the heterologous promoter is an artificial promoter such as a composite or hybrid promoter, or a promoter that is heterologous to the expression construct, such as a promoter that originates from a different expression construct, or a promoter that is not naturally-occurring. In particular, a truncated promoter, such as the EFS promoter which is a truncated variant of the EF1 A promoter, can be used.

Specifically, the heterologous promoter is weaker than the EF1 a promoter, as determined by fusing the promoter to a reporter gene, such as luciferase or GFP, and assessing its strength based on the expression of the reporter. The latter can be assessed by using a chemiluminescence-based reporter assay, flow cytometry or quantitative RT-PCR. Specifically, the heterologous promoter is a short promoter i.e., a promoter that is derived from a naturally-occurring promoter, but has a shorter length, such as a promoter fragment.

Specifically, the heterologous promoter is any one of an EF1 S promoter (also referred to as EFS promoter, SEQ ID NO:16), PGK promoter, TK promoter, GAPDH promoter, RSV promoter or UBC promoter.

It turned out that by using a weak promoter, such as an EFS promoter, significantly higher reactivation rates were obtained e.g., yielding at least 50%, 60%, 70%, 80%, or 990% of marker-positive cells. Specifically, the amount of re-activated lentiviral particles could be increased, such as e.g., 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, or 10-fold increased.

According to a specific aspect, the lentiviral pool is harvested from the perturbed host cell pool culture and transferred to the target cell pool culture.

According to a specific aspect, the repertoire of perturbed host cells is of packaging host cell lines suitable for producing lentiviral particles for delivery of a transgene to a target cell. Specifically, said transgene encodes a chimeric antigen receptor (CAR) for T-cell immunotherapy, a T cell receptor (TCR) for T-cell immunotherapy or a transgene for gene replacement therapy.

Specifically, the host cells are human cells such as selected from the group consisting of HEK cells, such as HEK293T, 293FT, Lenti-X, or 293SF-3F6.

Specifically, the target cell is provided as an isolated clone of a cell, or an in vitro cell culture of such clone.

According to a specific aspect, the target cells are human cells, preferably primary cells or stem cells, preferably selected from the group consisting of T cells, B cells, Macrophages, PBMCs and iPS cells.

Specifically, the target cells are selected from cancer cells, induced pluripotent stem cells (iPSCs), cells obtained from human iPSCs by differentiation, or primary immune cells.

Specifically, the target cell is a cancer cell or respective cell line, such as originating from a solid (e.g., epithelial tumor) or hematologic tumor. Exemplary cancer cells are human cell lines, such as e.g., selected from the following cell lines: K562 (ATCC, CCL-243), U937 (ATCC, CRL-1593, HEK293 (ATCC, CRL-1573), HELA (ATCC, CCL-2), RKO (ATCC, CRL-2577), HCT1 16 (ATCC, CCL-247), MCF7 (ATCC, HTB-22), MCF10A (ATCC, CRL-10317), A549 (ATCC, CCL-185), NCI-H358 (ATCC, CRL-5807), BEAS2B (ATCC, CRL-9609), NCI-H1437 (ATCC, CRL-5872), NCI-H1975 (ATCC, CRL-5908), JURKAT (ATCC, TIB-152), SW837 (ATCC, CCL-235), GM12878 (Coriell), SF126 (EMD/ Millipore), THP1 (ATCC, TIB-202).

Specifically, the target cell is a primary cell or an immune cell, such as of a cell type selected from the group consisting of a Natural Killer (NK) cell, a microglia cell, a macrophage, or a T cell, such as a cytotoxic T lymphocyte (CTL), a regulatory T cell or a T helper cell.

According to a specific aspect, the diversity of genes comprises at least 10 different genes, preferably wherein the genes are endogenous to the wild-type host cell genome. Specifically, the diversity is at least any one of 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 150, 200, 250, 300, 350, 400, 450, or 500 different genes, in specific cases up to 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or all of the endogenous genes comprised the host cell genome, in particular the wild-type host cell genome.

According to a specific aspect, a library of perturbed host cells is obtained from the perturbation screen, wherein the library covers perturbation of a diversity of at least 10 different genes, preferably wherein the diversity of genes is as further described herein.

Specifically, the host cells are transduced with replication-incompetent lentiviral delivery particles.

The invention further provides for a library comprising a repertoire of perturbed host cells which covers perturbation of a diversity of at least 10 different genes, wherein the host cells are transduced with replication-incompetent lentiviral delivery particles.

Specifically, the perturbation is directed to the host cell’s genome, in particular the genome of a wild-type host cell.

Specifically, the diversity is at least any one of 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 150, 200, 250, 300, 350, 400, 450, or 500 different genes, in specific cases up to 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or all of the endogenous genes comprised the host cell genome, in particular the wild-type host cell genome.

The library may be contained in one or more containments e.g., in a pool or in an arrayed device.

The invention further provides for the use of the library described herein, for identifying a lentiviral modulation factor which upon its perturbation in a producer host cell results in altered yield production of replication-incompetent lentiviral delivery particles. The invention further provides for a method of engineering a producer host cell line producing lentiviral delivery particles at high titer or high infectivity, comprising: a) identifying a lentiviral modulation factor in the genome of a host cell according to the method described herein, wherein the respective perturbation of the host cell for the identified lentiviral modulation factor has resulted in the production of lentiviruses at higher yield or infectivity against the target cells; and b) engineering a producer host cell line comprising said perturbation of the identified lentiviral modulation factor.

Specifically, the yield or infectivity of lentivirus is higher than a reference level, which is produced by such production host cell without the respective perturbation of the host cell for the identified lentiviral modulation factor. Specifically, where a gene is perturbed in the host cell and causes a higher yield or infectivity of lentivirus, such gene qualifies as a lentiviral modulation factor.

According to a specific aspect, the producer host cell line produces lentiviruses with an increased yield and/or infectivity, such as e.g., at least any one of 1 .2, 1 .3, 1 .4, 1 .5, 1 .6, 1 .7, 1 .8, 1 .9, or 2-fold increased yield or infectivity compared to the host cell line without such perturbation.

According to a specific aspect, the producer host cell line may contain two or more perturbations e.g., two or more lentiviral modulation factors may be knocked out, preferably wherein CRISPR/Cas9 is used for the genomic perturbation screen.

The invention further provides for a producer host cell line obtainable by a method described herein, which comprises a knockout of the identified lentiviral modulation factor, wherein the identified lentiviral modulation factor is a gene selected from the group consisting of a gene selected from the group consisting of ZC3HAV1 , ADGRG1 , MEN1 , EPHB4, BARHL1 , C11 orf71 , TNFRSF6B, OR10G9, C17orf58, CDCA7, TRIM25, C1 1 orf68, TMEM125, FUT2, RCE1 , ZNF398, TLDC2, FCAMR, TNFRSF14, CYBC1 , PDXK, CASS4, MMP23B, HRG, GSN, TP53INP2, WDR81 , RBM4, NFKBIB, LYNX1 - SLURP2, SLC7A2, SYNM, DEPTOR, SIGIRR, FOSL1 , FGF4, GDF5, PLAAT3, TNPO2, HLA-DPA1 , MAGEB1 , PNMA6E, IRF2BP2 and TRPT1 .

Specifically, the producer host cell line originates from a human cell, preferably a human cell that is used as a target cell as described herein, preferably a HEK cell such as described herein. The invention further provides for a method of producing lentiviral delivery particles in a producer host cell line produced as described herein, preferably wherein said lentiviral delivery particles are replication-incompetent.

According to a specific aspect, said lentiviral delivery particles are packaged with a transgene for delivery to a target cell genome, preferably wherein said transgene encodes a chimeric antigen receptor (CAR) for T-cell immunotherapy, a T cell receptor (TCR) for T-cell immunotherapy, or a transgene for gene replacement therapy.

FIGURES

Figure 1 : Generation of HEK293T-SpCas9-Blasticidin clones. HEK293T-WT cells were transduced with a lentivirus encoding “SpCas9-Blasticidin” followed by Blasticidin selection and single clone generation (shown: clone “16“ termed from now on “clone 1 ” and “23”, termed from now on “clone 2”). Subsequent, single clones were tested via FACS for gene knockout efficiency. Knockout was determined via an anti-B2M-APC antibody (B2M-01 ; Fisher Scientific GmbH). As controls, HEK293T WT cells, receiving no lentivirus, and polyclonal HEK293T-SpCas9 cells transduced with a lentivirus encoding a non-targeting control sgRNA (NTC) and GFP have been used. HEK293T WT cells showed no GFP and high B2M - APC signal (Plot 1 ); Polyclonal HEK293T- SpCas9 showed GFP and APC signal (Plot 2); HEK293T-SpCas9 single clones “16” and “23” transduced with a lentivirus encoding a sgRNA against B2M and GFP as a marker for transduction efficiency showed similar knockout efficiencies of ~80% against the B2M gene (Plot 3 and 4).

Figure 2: Growth curve of HEK293T-SpCas9-Blasticidin clones / RNA sequencing. a) Viability and cell growth was monitored for 4 days for HEK293T WT, HEK293- SpCas9-clone 1 and HEK293-SpCas9-clone 2. No aberrant growth defect compared to the HEK293T-WT cells has been observed. b) Total RNA was isolated from HEK293T WT, HEK293T-SpCas9-polyclonal, HEK293-SpCas9-clone 1 and HEK293-SpCas9-clone 2 (3 biological replicates each) and deep sequenced. Differential gene expression analysis showed a Spearman correlation > 0.98 within all cell lines.

Figure 3: Comparison standard lentivirus and improved system. Direct comparison of knockout efficiency between the “truncated LTRs system” (standard lentivirus generation system plot 2 and 3 - top) and “WT LTRs system” (improved system plot 4 and 5 - bottom). The transduction of HEK293T-SpCas9 polyclonal cells with a NTC sgRNA (truncated system or WT-LTRs system) or a sgRNA against B2M (truncated system or WT-LTRs system) led to similar transduction and knockout efficiencies of about 45% within both systems.

Figure 4: Lentivirus production and re-activation. HEK293T-SpCas9-clone 1 was transduced with a “NTC sgRNA - GFP” or “B2M sgRNA - GFP” cassette, encoded within the WT-LTR donor template. B2M - APC staining revealed ~95% B2M-APC levels within the NTC (Plot 2- top 1^st virus infection) and 25% B2M-APC levels within the B2M condition (Plot 3 - bottom top 1 ^st virus infection), translating to a knockout efficiency of 75% (Plot 3). The integrated viruses (WT LTRs NTC and WT LTR B2M) were reactivated and a 2^nd cell line (Jurkats Clone E6-1 TIB-152) was transduced. Anti B2M-APC staining showed no knockout within the NTC sgRNA (Plot 4 - top 2^nd virus infection) and a 66% knockout efficiency within the B2M sgRNA condition (Plot 5- bottom top 2^nd virus infection).

Figure 5: Proof of concept screen; SpCas9 HEK293T cells enriched with WT-LTR sgRNA library. HEK293T-SpCas9-clone 1 cell line was transduced with a knockout sgRNA library cloned into the WT-LTR system at MOI=0.1 . After Blasticidin selection the three transduced replicates showed about 83 to 93% GFP signal (Plot 2,3,4).

Figure 6: Transduction efficiency of 2^nd cell line with the WT-LTR sgRNA library. The WT-LTR sgRNA library was reactivated and a 2^nd cell line (HEK293T-WT) was transduced. After Blasticidin selection transduced HEK293T-WT showed 71 -80% GFP positive signal.

Figure 7: sgRNA library recovery within 1^st and 2^nd cell line. gDNA was harvested from three replicates of the HEK293T-SpCas9-clone 1 cell line after transduction of the WT-LTR sgRNA library and selection via Blasticidin. Similarly, gDNA was harvested from the 3 replicates of the 2^nd cell line (HEK293T-WT) after transduction with the reactivated virus and selection via Blasticidin. Data analysis of the amplified sgRNAs identified all sgRNAs within the three replicates of the HEK293T-SpCas9-clone 1 cell line (top three plots; “Input samples”). Data analysis of the 2^nd cell line (bottom three plots; “Output samples”) revealed a non-uniform distribution of sgRNAs due to the dropout of sgRNAs targeting essential gene and enrichment of lentiviral repressing sgRNAs.

Figure 8: Analysis WT-LTR re-activation screen from total of about 400 genes. Analysis of output vs. input samples, shown are top and bottom 30 genes from total of 400 screened genes. The top two genes, TRIM25 and ZC3HAV1 are significant enriched (FDR <=0.05) while most essential genes dropped out and were found within the bottom 30 genes.

Figure 9: Knockout of TRIM25 or ZAP improve lentivirus production.

HeLa-WT and HEK293T-WT cells were transduced with lentivirus produced from the SpCas9-HEK293T, SpCas9-HEK293T - TRIM25-knockout sgRNA 1 , SpCas9- HEK293T- TRIM25-knockout sgRNA 2 and SpCas9-HEK293T - ZC3HAV1 knockout cell lines. The knockout cell lines showed higher GFP levels in a FACS readout compared to the parental SpCas9-HEK293T cell line.

Fig. 10: Test of lentiviral setups as used in the Examples.

Fig. 10a outlines a self-inactivated vector (SIN) comprising a truncated 3’LTR, unable to be re-activated after the addition of helper plasmids. Fig. 10b describes a lentivirus re-activation system which comprises a promoter within the 3’LTR, namely LeAPsPro. Fig. 10c and d outline a re-activation system utilizing WT-LTRs.

Fig. 1 1 : Lentiviral systems; 1^st round of transduction & selection

HEK WT cells received no lentivirus (Fig. 1 1 a control) or were transduced with the lentiviral systems (3’LTR U3, EF1 a, EFs or LeAPs Pro) described in Fig. 10. The transduction efficiency for all 4 systems was similar with -30-55%. Subsequent selection of the transduced cells led to a lentivirus library containing cell population of 98-99.5% (Fig. 1 1 b).

Fig. 12: Lentiviral system; 2^nd round of re-activation

After re-activation of the lentivirus and transduction of HEK WT cells, the WT- LTR-EF1 a as well as the LeAPs Pro system led to -20% transduced cells while the WT- LTREFs system infected -75% of cells.

Fig. 13: Genome-wide screen; Transduction & selection of cells

3 replicates of SpCas9-HEK293T cells were transduced with a genome-wide lentivirus library. The transduction efficiency for all 3 replicates was -30-40% (Fig. 13a) with subsequent selection leading to 91 -95% SpCas9-HEK293T cells with a viral integration (Fig. 13b).

Fig. 14: Genome-wide screen; Correlation of samples

Genome-wide data outline a high correlation between the plasmid library and the input samples (Spearman >0.8) as well as a correlation within the output samples (Spearman >0.65). Input samples were collected before the viral re-activation while output samples were harvested after the transduction of a 2^nd cell line. The difference between the detected sgRNAs within the input samples and the output samples determined the identified hits.

Fig. 15: Validation screen workflow

A lentiviral oligo pool library was generated based on the data from the previous genome-wide screen. SpCas9-HEK293T cells were transduced with a low MOI to ensure one infection event per cell. After integration and knockout of a target gene, cells were selected for several days to enrich for a population comprising the lentivirus library. Next, cells received helper plasmids leading to the re-activation of the lentivirus within the selected cells. The virus was harvested and another batch of SpCas9-HEK293T cells was transduced. This transduction-re-activation process was repeated three times leading to 3x input samples and 9x output samples.

Fig. 16: Validation screen; Correlation of samples

Analysis of the deep sequencing data showed a very high correlation of the 3x input samples (Spearman correlation >0.9) as well as a high correlation within the output samples (Spearman correlation >0.7).

Fig. 17: Validation screen; Correlation of genome-wide & validation screen

A high correlation of the log-fold changes (LFC) for the cumulative LFC over the 3 rounds of the validations screen vs. genome-wide screen was observed. In addition, a significant increase in viral infection rates upon knockout of anti-viral genes, no effect within non-essential control genes and a decrease in viral infection rates due to the knockout of pro-viral genes was observed.

Fig. 18: Top significant hits of validation screen

44 genes were identified with a false discovery rate (FDR) <0.05 that significantly affected lentiviral production, assembly or secretion.

Fig. 19: Sequences referred to herein.

DETAILED DESCRIPTION OF THE INVENTION

Unless indicated or defined otherwise, all terms used herein have their usual meaning in the art, which will be clear to the skilled person. Reference is for example made to the standard handbooks. Genetic modifications described herein may employ tools, methods and techniques known in the art, such as described by J. Sambrook et al., Molecular Cloning: A Laboratory Manual ^(3rd edition), Cold Spring Harbor Laboratory, Cold Spring Harbor Laboratory Press, New York (2001 ), Lewin “Genes ”V”, Oxford University Press, New York, (1990), and Janeway et al“ “Immunobiology” (5^th Ed., or more recent editions), Garland Science, New York, 2001. The terms “comprise”, “contain”, “have” and “include” as used herein can be used synonymously and shall be understood as an open definition, allowing further members or parts or elements. “Consisting” is considered as a closest definition without further elements of the consisting definition feature. Thus “comprising” is broader and contains the “consisting” definition.

The term “about” as used herein refers to the same value or a value differing by +/-10% or +/-5% of the given value.

The subject matter of the claims specifically refers to artificial products or methods employing or producing such artificial products, which may be variants of native (wildtype) products. Though there can be a certain degree of sequence identity to the native structure, it is well understood that the materials, methods and uses of the invention, e.g., specifically referring to isolated nucleic acid sequences, amino acid sequences, fusion constructs, expression constructs, transformed host cells and modified proteins, are “man-made” or synthetic, and are therefore not considered as a result of “laws of nature”.

Specific terms as used throughout the specification have the following meaning:

The term “barcode,” as used herein, generally refers to a label, or identifier (in particular a unique identifier), that conveys or is capable of conveying information about an analyte. A barcode can be part of an analyte. A barcode can be independent of an analyte. A barcode can be a tag attached to an analyte (e.g., nucleic acid molecule) or a combination of the tag in addition to an endogenous characteristic of the analyte (e.g., size of the analyte or end sequence(s)). A barcode may be unique. Barcodes can have a variety of different formats. For example, barcodes can include: polynucleotide barcodes; random nucleic acid and/or amino acid sequences; and synthetic nucleic acid and/or amino acid sequences. A barcode can be attached to an analyte in a reversible or irreversible manner. A barcode can be added to, for example, a fragment of a deoxyribonucleic acid (DNA) or ribonucleic acid (RNA) sample before, during, and/or after sequencing of the sample. Barcodes can allow for identification and/or quantification of individual sequencing-reads.

Nucleic acid molecules (e.g., RNA or DNA, or chimeras) such as polynucleotides or oligonucleotides can be barcoded by a molecular barcode incorporated in the nucleotide sequence. The length of barcoded nucleotide sequences as used for the purpose described herein typically ranges from 4 to 60 nucleotides, preferably between 20 and 30 nucleotides. The barcoded nucleotide sequences can be single-stranded or double-stranded. A guide RNA or a guide RNA-coding sequence can represent a barcode for a sample containing this guide RNA or guide RNA-coding sequence.

Specifically, a barcoded nucleotide sequence array can be used for a repertoire of host cells comprising a diversity of genomic modulations of a, each with a unique barcode. A nucleotide sequence array can be constructed so that different members of the array contain different sequences (a unique sequence is used as a barcode). Such arrays are commercially available with user-defined nucleotide sequences. For example, custom barcoded guide RNA arrays are commercially available, such as from TWIST Bioscience, Merck, Cellecta, ThermoFisher, IDT, Synthego, Horizon Discovery or GenScript.

A barcoded nucleotide sequence array typically comprises: (i) a sequencing primer at the attached end; (ii) unique barcodes for each member; and (iii) short sticky- end adapter sequences at their free ends to allow easy incorporation into a target vector by cloning.

In certain embodiments, unique barcodes are used to perform Perturb-seq, such as to link guide RNAs with single cells in conventional single-cell RNA sequencing workflows. In certain embodiments, a guide RNA is detected by single-cell RNA-seq using an RNA Polymerase II transcript expressed from a vector encoding the guide RNA. The transcript may include a unique barcode specific to the guide RNA or the guide RNA itself may serve as unique barcode. The transcript may include the guide RNA sequence (see, e.g., Datlinger, et al., 2017).

In certain embodiments, a guide RNA and guide RNA barcode is expressed from the same vector, and the barcode may be detected by RNA-seq. Thus, a perturbation may be assigned to a single cell by detection of a guide RNA barcode in the cell.

In certain embodiments, a cell barcode is added to the RNA in single cells, such that the RNA may be assigned to a single cell. Generating cell barcodes is specifically provided for single-cell sequencing methods, such as employing arrayed or pooled methods of sequencing.

In certain embodiments, a Unique Molecular Identifier (UMI) is comprised in each individual LV.

In specific embodiments, Perturb-seq is performed using a guide RNA barcode and a UMI.

In certain embodiments, CRISPR-based perturbations are readily compatible with Perturb-seq, including specific DNA-targeting proteins as editors. For example, barcoded LV can be used. The LV may comprise an sgRNA which incorporates a barcode that is unique for the target of genomic modulation in the producer host cell. Upon infecting a host cell, such LV can mediate targeted modulation of a gene in the host cell’s genome that expresses or otherwise comprises a CRISPR enzyme. As described herein, the host cell produces barcoded LV, which can then be harvested from the host cell culture.

As described herein, to determine functionality of the barcoded LV produced by the host cell, a target cell line is infected with the LV. Where a replication-incompetent LV has been used in the first round of infection (“first cycle”), the harvested LV can be reactivated by suitable engineering and means, to allow infection of the target cell and subsequent identification of the barcodes, thus providing a link to the genomic perturbation introduced into the producer host cell.

The term “cell” with respect to an engineered cell comprising an operation system or a “host cell” as used herein shall refer to a single cell, a single cell clone, or a cell line of a host cell.

The term “host cell”, particularly in the context of a “producer host cell” as used herein, shall particularly apply to any cell, which is suitably used for lentiviral infection and/or replication and/or production. It is well understood that the term “host cell”, in particular “producer host cell”, refers to recombinant host cells and does not include human beings. Specifically, recombinant host cells as described herein are artificial organisms and derivatives of native (wild-type) host cells. Such recombinant host cells are provided as isolated host cell and respective host cell culture, which is understood as an ex vivo host cell or host cell culture. Recombinant host cells can be used for in vivo therapies, or cultured ex vivo to produce host cell products.

It is well understood that the host cells, methods and uses described herein, e.g., specifically referring to those comprising one or more genetic modifications, heterologous expression cassettes or artificial expression constructs, said transfected or transformed host cells and recombinant proteins, are non-naturally occurring, are “man-made” or synthetic, and are therefore not considered as a result of “law of nature”.

The term “cell line” as used herein refers to an established clone of a particular cell type that has acquired the ability to proliferate over a prolonged period of time. A cell line is typically used for producing host cell products, such as a virus (e.g., a lentivirus), expression products such as a protein of interest, or cell metabolites. A “production host cell line” or “producer host cell line” is commonly understood to be a cell line ready-to-use for cell culture in a bioreactor to obtain the product of a production process, such as a virus (e.g., a lentivirus), expression products such as a protein of interest, or cell metabolites. Typically, the production host cell line is cultured in an ex vivo cell culture, and the products of the production process are obtained in the cell culture or a fraction thereof such as the cell culture supernatant.

A “packaging host cell line” as referred to herein is understood as a production host cell line that produces lentiviral particles upon the introduction of a lentiviral donor template. Production of the lentiviral particle may require the exogenous addition of viral helper genes (one or more of gag, pol, env, rev, and tat), e.g., on a one or more viral helper plasmids. Alternatively, any one or more of such viral genes may already be present and stably or inducibly expressed in a “packaging host cell line”.

As described herein, a host cell may be engineered by genomic perturbation. For example, expression of a gene may be upregulated or downregulated, a gene may be overexpressed, such as expressed at higher levels, or underexpressed, such as expressed at lower levels, as compared to a host cell without such engineering. The yield of host cell products may be increased or reduced. A gene editing method may be used for host cell engineering. Specifically, the host cell may express a CRISPR enzyme to support engineering the host cell’s genome by CRISPR-mediated perturbation.

Genomic perturbation of a host cell line targeting different host cell genes or respective regulatory elements, can result in a diversity of host cells with differences in the respective host cell’s genome.

As described herein, a repertoire of host cells is produced by genomic perturbation of the host cell. Specifically, host cells originating from the same host cell line are subject to genomic perturbation, which results in the same type of host cells, but with a difference in their genome that corresponds to the target and type of genomic modulation caused by the perturbation.

A “target host cell line” as referred to herein is understood as the cell line that is the target of infection by a virus such as a lentivirus.

As described herein, lentiviral particles produced by a repertoire of host cells are used to infect a target cell line. This allows for quantifying target cell infection as a measure of the amount of LV produced by the host cells. Where the LV comprises a barcode, the barcode can typically be read or otherwise determined upon infection of a target cell with the LV e.g., in the target cell culture or a fraction thereof. The term “cell culture” or “culturing” as used herein with respect to a host cell refers to the maintenance of cells in an artificial, e.g., an in vitro environment, under conditions favoring growth, differentiation or continued viability, in an active or quiescent state, of the cells, specifically in a controlled bioreactor according to methods known in the industry. When culturing a cell culture using appropriate culture media, the cells are brought into contact with the media in a culture vessel or with substrate under conditions suitable to support culturing the cells in the cell culture. Standard cell culture media and techniques are well-known in the art.

A culture of different cells, such as a repertoire of diverse host cells which differ in their host cell genome, can be performed in a pool. Such pool culture is typically carried out with a mixture of diverse host cells, also referred to as being a “co-culture” of different host cell clones, or a “polyclonal” cell culture. The pool culture is conveniently performed in only one cell culture containment (such as a bioreactor or vessel).

Alternatively, a repertoire of diverse host cells can be cultured in an array e.g., to culture a reduced diversity or even a single cell or clone in one cell culture containment (a “monoclonal” cell culture), such as in an array of cell culture bioreactors or vessels.

As described herein, a LV preparation can be produced in a host cell culture, by culturing in an appropriate medium and harvesting LV e.g., by isolating a fraction comprising LV, and producing a LV preparation by a suitable method.

Host cells described herein can be tested for their capability to produce infectious LV, such as by determining the viral load e.g., the amount of virus or of nucleic acid molecules (of a fragment thereof) comprised in the LV. The viral load may be structurally determined, directly or indirectly such as by functionally determining the viral load.

According to a specific aspect described herein, the functionality of LV produced by the host cells is determined. Functionality of viruses can be determined in multiple biological assays including the determination of the viral titer by qPCR, by determining the viral titer using an ELISA assay for one of the viral component proteins (e.g. the HiV- 1 p24 antigen), or by infecting target cells and determining the infection rate using a virally encoded reporter (e.g., GFP). Functionality of a lentivirus particularly understood as infectivity of the virus, and can e.g., be determined by a respective infectivity test well- known in the prior art.

For example, the viral load or functionality can be determined by a phenotypic measurement of the amount of infectivity of a LV preparation. To this end, target cells may be infected with the LV preparation, and the number of infected cells or the viral load in the target cell or target cell culture indicates the amount of LV produced by the host cells. The viral load may be determined by any of the following tests: sequencing, qPCR (in particular to detect one or more genes encoded by the LV), ELISA via the detection of a viral protein, e.g. p24, or FACS via the infection of a 2^nd cell line and the readout via a fluorophore which is encoded within the virus.

As used herein, the term “gene” e.g., as used in the context of “target gene” or “gene of interest” or “GOI”, includes a DNA region encoding a gene product, and optionally one or more (or all) DNA regions which regulate the production of a host cell product, whether or not such regulatory sequences are adjacent to coding and/or transcribed sequences. Accordingly, a gene may be understood to include promoter sequences, exons and introns, terminators, translational regulatory sequences such as ribosome binding sites and internal ribosome entry sites, enhancers, silencers, insulators, boundary elements, replication origins, matrix attachment sites and locus control regions.

The term “expression” as used herein refers to transcription of a polynucleotide or gene from an expression cassette, or to the translation of the respective polypeptide or protein. As used herein, the term “expression” also refers to host cell products expressed to produce a virus or viral components, in particular the lentiviral particle.

The devices, facilities and methods used for the purpose described herein are specifically suitable for use in and with culturing any desired cell line, and are particularly suitable for production operations configured for production of expression products, such as pharmaceutical and biopharmaceutical products e.g., polypeptide products, nucleic acid products (for example DNA or RNA), cells and/or viruses such as those used in cellular and/or viral therapies, in particular LV such as used as viral vectors or in human therapies. Expression products can be produced in vitro such as in an isolated ex vivo cell culture, or in vivo, such as by administration of the cells or the respective operating system to a subject (in particular a mammal, or a human being, such as a patient or healthy subject) e.g., for gene therapy, in particular cell-based gene therapy.

Specific expression products are viruses, in particular LV.

Specific expression products are CRISPR enzymes as used in genomic modulation of the cell that expresses a CRISPR enzyme in conjunction with a guiding element to target a gene or genomic site of interest.

The term “expression cassette” is herein understood to refer to nucleic acid molecules (herein also referred to as polynucleotides), which contain a desired coding sequence (herein referred to as a gene), and control sequences in operable linkage, so that hosts transformed or transfected with these molecules incorporate the respective sequences and are capable of producing the encoded proteins or host cell metabolites.

An expression cassette may comprise one or more nucleic acid molecules (e.g., a gene) that are endogenous or heterologous to a host cell.

An expression cassette can be engineered ex vivo or in vivo. For example, an expression cassette can be engineered as part of a vector or an artificial expression construct that can be provided as isolated expression construct that is engineered using in vitro techniques, and optionally incorporated in a cell culture. An artificial expression construct can also be engineered in vivo, by incorporating one or more heterologous elements of an expression cassette, such as e.g., a promoter or a gene switch that is not operably linked to a target gene in a wild-type host cell, into a wild-type expression cassette that is endogenous to a host cell, thereby modifying the wild-type expression cassette to produce a recombinant, artificial expression cassette.

An element of an expression cassette that is not operably linked to a certain gene in a wild-type host cell, which gene is endogenous to the wild-type host cell, is herein understood as “not natively associated with” such gene. A wild-type host cell is herein understood to be a naturally-occurring host cell that is not recombined by any artificial means. Such expression cassettes comprising one or more elements that are not natively associated with a gene, are herein understood as artificial or recombinant expression cassettes.

One or more expression cassettes are herein also understood as “expression system”. The expression system may be included in an expression construct, such as an artificial heterologous expression cassette, a vector and in particular a plasmid. The relevant DNA of an expression cassette or construct may also be integrated into a host cell chromosome. Expression may refer to secreted or non-secreted expression products, including polypeptides or metabolites or viruses.

Expression cassettes are conveniently provided as expression constructs e.g., in the form of “vectors” or “plasmids”, which are typically DNA sequences that are required for the transcription of cloned recombinant nucleotide sequences i.e., of recombinant genes and the translation of their mRNA in a suitable host organism. Expression vectors or plasmids usually comprise an origin for autonomous replication or a locus for genome integration in the host cells, selectable markers (e.g., an amino acid synthesis gene or a gene conferring resistance to antibiotics such as zeocin, kanamycin, G418 or hygromycin, nourseothricin), a number of restriction enzyme cleavage sites, a suitable promoter sequence and a transcription terminator, which components are operably linked together. The terms “plasmid” and “vector” as used herein include autonomously replicating nucleotide sequences as well as genome integrating nucleotide sequences, such as artificial chromosomes e.g., a bacterial artificial chromosome (BAC) or yeast artificial chromosome (YAC).

Expression vectors may include but are not limited to cloning vectors, modified cloning vectors and specifically designed plasmids. Preferred expression vectors described herein are expression vectors suitable for expressing a recombinant gene in a eukaryotic host cell and are selected depending on the host organism. Appropriate expression vectors typically comprise regulatory sequences suitable for expressing DNA encoding polypeptide or protein of interest in a eukaryotic host cell. Examples of regulatory sequences include promoter, operators, enhancers, ribosomal binding sites, and sequences that control transcription and translation initiation and termination. The regulatory sequences are typically operably linked to the DNA sequence to be expressed.

To allow expression of a recombinant nucleotide sequence in a host cell, a promoter sequence is typically regulating and initiating transcription of the downstream nucleotide sequence, with which it is operably linked. An expression cassette or vector typically comprises a promoter nucleotide sequence which is adjacent to the 5’ end of a coding sequence, e.g., upstream from and adjacent to the coding sequence (e.g., gene of interest) or if a signal or leader sequence is used, upstream from and adjacent to said signal and leader sequence, respectively, to facilitate translation initiation and expression of coding sequences to obtain the expression product.

The term “gene expression”, or “expressing a polynucleotide” or “expressing a nucleic acid molecule” as used herein, is meant to encompass at least one step selected from the group consisting of DNA transcription into mRNA, mRNA translation and processing, mRNA maturation, mRNA export, protein folding and/or protein transport.

The term “genomic perturbation” and “genomic perturbation screen” as used herein is specifically understood in the following way.

A cell may be engineered for genomic perturbation of one or more genes (i.e., targets(s) of genomic perturbation), thereby modulating the respective gene(s) expression, such as upregulating, downregulating, overexpressing or underexpressing (understood as expressing at higher and lower levels than normal, respectively), or in particular by completely knocking out the target(s) of the perturbation. A target of perturbation is also referred to as a “perturbed gene”. A cell that is undergoing targeted perturbation is also referred to as “perturbed cell”.

A repertoire of diverse perturbed cells can be produced by genomic perturbation of host cells to target different genes in individual cells. For example, a repertoire comprises a diversity of cells, wherein each cell is perturbed to target a different gene.

For perturbation of a diversity of genes, a library of targeting molecules can be used, such as e.g., a library of different guide RNAs (e.g., sgRNAs), each targeting a different gene. The targeting molecules such as guide RNAs (e.g., sgRNAs) conveniently comprise a barcode with a unique identifier of the target gene.

In an RPN-mediated perturbation screen, the library of targeting molecules can be used in the presence of an RPN that mediates the genomic perturbation in cells, thereby obtaining a repertoire of diverse cells, wherein the diversity originates from the respective targeting molecules (such as guide RNAs e.g., sgRNAs) that cause different perturbations in the cells.

Perturbation methods and tools are well-known in the art. In one embodiment, the method comprises (1 ) introducing single-order or combinatorial perturbations to a population of cells, (2) measuring genomic, genetic, proteomic, epigenetic and/or phenotypic differences in single cells and (3) assigning a perturbation(s) to the single cells. A perturbation may be linked to a phenotypic change, preferably changes in gene or protein expression. In specific embodiments, measured differences that are relevant to the perturbations are determined by applying a model accounting for co-variates to the measured differences. The model may include the capture rate of measured signals, whether the perturbation actually affected the cell (phenotypic impact), the presence of subpopulations of either different cells or cell states, and/or analysis of matched cells without any perturbation. In certain embodiments, the measuring of phenotypic differences and assigning a perturbation to a single host cell is determined by determining a respective barcode in the host cell’s product (such as a LV) and/or the effect of the host cell’s product (such as the amount of LV infecting a target cell or the respective barcode in the target cell culture).

In one embodiment, CRISPR systems using a CRISPR enzyme, such as CRISPR/Cas9, may be used to perturb protein-coding genes or non-protein-coding DNA. In the context of a Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) system, a “target”, refers to a genomic sequence to which a guide sequence is designed to have complementarity, where hybridization between a target sequence and a guide sequence promotes the formation of a CRISPR complex. Where a transcription-modulating domain is employed, upon such hybridization of the guide sequence, there target GOI is typically upregulated or downregulated in terms of its expression level. Typically, the target sequence is a regulatory sequence, such as a promoter, which controls the expression of said GOI.

There are currently four main types of programmable nucleases (sometimes also referred to as “site specific nucleases”, “RNA-guided nuclease” or “targetable nuclease”) in use: zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), RNA-guided nucleases (RGNs) such as the Cas proteins of the CRISPR/Cas Type II system, and engineered meganucleases. ZFNs and TALENs comprise the nuclease domain of the restriction enzyme Fokl (or an engineered variant thereof) fused to a site-specific DNA binding domain (DBD) that is appropriately designed to target the protein to a selected DNA sequence. In the case of ZFNs, the DNA binding domain comprises a zinc finger DBD. In the case of TALENs, the site-specific DBD is designed based on the DNA recognition code employed by transcription activator-like effectors (TALEs), a family of site-specific DNA binding proteins found in plant-pathogenic bacteria such as Xanthomonas species. The CRISPR Type II system is a bacterial adaptive immune system that has been modified for use as an RNA-guided endonuclease technology for genome engineering. The bacterial system comprises two endogenous bacterial RNAs called crRNA and tracrRNA and a CRISPR-associated (Cas) nuclease, e.g., Cas9. The tracrRNA has partial complementarity to the crRNA and forms a complex with it. The Cas protein is guided to the target sequence by the crRNA/tracrRNA complex, which forms an RNA/DNA hybrid between the crRNA sequence and the complementary sequence in the target. For use in genome modification, the crRNA and tracrRNA components are often combined into a single chimeric guide RNA (sgRNA or gRNA) in which the targeting specificity of the crRNA and the properties of the tracrRNA are combined into a single transcript that localizes the Cas protein to the target sequence so that the Cas protein can cleave the DNA. The sgRNA often comprises an approximately 20 nucleotide guide sequence complementary or homologous to the desired target sequence followed by about 80 nt of hybrid crRNA/tracrRNA. One of ordinary skill in the art appreciates that the guide RNA need not be perfectly complementary or homologous to the target sequence. For example, in some embodiments it may have one or two mismatches. The genomic sequence which the gRNA hybridizes is typically flanked on one side by a Protospacer Adjacent Motif (PAM) sequence although one of ordinary skill in the art appreciates that certain Cas proteins may have a relaxed requirement for a PAM sequence. The PAM sequence is present in the genomic DNA but not in the sgRNA sequence. The Cas protein will be directed to any DNA sequence with the correct target sequence and PAM sequence. The PAM sequence varies depending on the species of bacteria from which the Cas protein was derived. Specific examples of Cas proteins include Cas1 , Cas2, Cas3, Cas4, Cas5, Cas6, Cas7, Cas8, Cas9 and Casi o or Cas12a or Cas12g. In some embodiments, the site-specific nuclease comprises a Cas9 protein. For example, Cas9 from Streptococcus pyogenes (Sp), Neisseria meningitides, Staphylococcus aureus, Streptococcus thermophiles, or Treponema denticola may be used. A number of engineered variants of the site-specific nucleases have been developed and may be used in certain embodiments. For example, engineered variants of Cas9 and Fok1 are known in the art. Furthermore, it will be understood that a biologically active fragment or variant can be used. Other variations include the use of hybrid site specific nucleases. For example, in CRISPR RNA-guided Fokl nucleases (RFNs) the Fokl nuclease domain is fused to the amino-terminal end of a catalytically inactive Cas9 protein (dCas9) protein. Site-specific nucleases that produce a single-stranded DNA break are also of use for genome editing. Such nucleases, sometimes termed “nickases” can be generated by introducing a mutation (e.g., an alanine substitution) at key catalytic residues in one of the two nuclease domains of a site-specific nuclease that comprises two nuclease domains (such as ZFNs, TALENs, and Cas proteins). Examples of such mutations include D10A, N863A, and H840A in SpCas9 or at homologous positions in other Cas9 proteins. A nick can stimulate HDR at low efficiency in some cell types. Two nickases, targeted to a pair of sequences that are near each other and on opposite strands can create a single-stranded break on each strand (“double nicking”), effectively generating a DSB, which can optionally be repaired by HDR using a donor DNA template (Ran et al., 2013). In some embodiments, the Cas protein is a SpCas9 variant. In some embodiments, the SpCas9 variant is a R661 A/Q695A/Q926A triple variant or a N497A/R661A/Q695A/Q926A quadruple variant. In some embodiments, the targetable nuclease (e.g., site-specific nuclease) has at least 90%, 95% or 99% polypeptide sequence identity to a naturally occurring targetable nuclease. In one embodiment, perturbation is by deletion of regulatory elements. Noncoding elements may be targeted by using pairs of guide RNAs to delete regions of a defined size and by tiling deletions covering sets of regions in pools.

In one embodiment, perturbation of genes is by RNAi. The RNAi may be obtained using shRNAs, siRNAs or micro RNAs targeting genes. The shRNAs may be delivered by any methods known in the art. In one embodiment, the shRNAs may be delivered by a viral vector. The viral vector may be a lentivirus, adenovirus, or adeno associated virus (AAV).

A specific approach comprises target mRNA degradation by using small interfering RNA (siRNA) to transfect the cell and targeting a mRNA encoding the target antigen expressed by said cell.

Gene silencing, gene knock-down and gene knockout refers to techniques by which the expression of a gene is reduced, either through genetic modification or by treatment with an oligonucleotide with a sequence complementary to either an mRNA transcript or a gene. If the change in gene expression is caused by an oligonucleotide binding to an mRNA or temporarily binding to a gene, this results in a temporary change in gene expression without modification of the chromosomal DNA and is referred to as a transient knock-down.

In a transient knock-down, which is also encompassed by the above term, the binding of this oligonucleotide to the active gene or its transcripts causes decreased expression through blocking of transcription (in the case of gene-binding), degradation of the mRNA transcript (e.g., by small interfering RNA (siRNA) or antisense RNA) or blocking mRNA translation.

Other approaches to carry out gene silencing, knock-down or knockout are known to the skilled person from the respective literature, and their application in the context of the present invention is considered as routine. Gene knockout refers to techniques by which the expression of a gene is fully blocked i.e., the respective gene is inoperative, or even removed. Methodological approaches to achieve this goal are manifold and known to the skilled person. Examples are the production of a mutant which is dominantly negative for the given gene. Such mutant can be produced by site directed mutagenesis (e.g., deletion, partial deletion, insertion or nucleic acid substitution), by use of suitable transposons, or by other approaches which are known to the skilled person from the respective literature, the application of which in the context of the present invention is thus considered as routine. One example is knockout by use of targeted Zinc Finger Nucleases. A respective Kit is provided by Sigma Aldrich as “CompoZR knockout ZFN”. Another approach encompasses the use of Transcription activator-like effector nucleases (TALENs). Inactive mutants of ZFNs or TALENs can be combined with transcriptional repressors or activators to enable targeted gene modulation.

The delivery of a dominant negative construct involves the introduction of a sequence coding for a dysfunctional gene expression product, e.g., by transfection. Said coding sequence is functionally coupled to a strong promoter, in such way that the gene expression of the dysfunctional enzyme overrules the natural expression of the gene expression product, which, in turn, leads to an effective physiological defect of the respective activity of said gene expression product.

A conditional gene knockout allows blocking gene expression in a tissue- or timespecific manner. This is done, for example, by introducing short sequences called loxP sites around the gene of interest. Again, other approaches are known to the skilled person from the respective literature, and their application in the context of the present invention is considered as routine.

One other approach is gene alteration which may lead to a dysfunctional gene product or to a gene product with reduced activity. This approach involves the introduction of frame shift mutations, nonsense mutations (/.e., introduction of a premature stop codon) or mutations which lead to an amino acid substitution which renders the whole gene product dysfunctional, or causing a reduced activity. Such gene alteration can for example be produced by mutagenesis (e.g., deletion, partial deletion, insertion or nucleic acid substitution), either unspecific (random) mutagenesis or site directed mutagenesis or by using programmable nucleases targeting endogenous genes. Protocols describing the practical application of gene silencing, gene knockdown, gene knockout, delivery of a dominant negative construct, conditional gene knockout, and/or gene alteration are commonly available to the skilled artisan, and are within his routine. The technical teaching provided herein is thus entirely enabled with respect to all conceivable methods leading to an inhibition or reduction of gene expression of a gene product, or to the expression of a dysfunctional, or inactive gene product, or with reduced activity.

Genetic modifications described herein may employ tools, methods and techniques known in the art, such as described by J. Sambrook etal., Molecular Cloning: A Laboratory Manual (3^rd edition), Cold Spring Harbor Laboratory, Cold Spring Harbor Laboratory Press, New York (2001 ).

Differential expression of a target gene may result from its perturbation, such as by interfering (or downregulating) or activating (or upregulating) of the target gene expression.

A target gene is understood to be “perturbed”, if there is a perturbation of the gene transcripts. Specifically, the gene expression can be downregulated (thereby reducing expression) or upregulated (thereby increasing expression). Downregulation or upregulation of gene expression is herein also referred to as “perturbation of gene expression”. Specifically, the perturbation can occur by introducing a small insertion or deletion into a coding exon, thus partially or completely eliminating the functional gene product from the target cell.

The term “reduce expression” generally refers to any amount less than an expression level exhibited by a reference standard, which is a cell prior to the engineering to reduce expression of a certain gene, or which is otherwise expressed in a cell of the same type or species which is not engineered to downregulate expression of said gene. Typically, a cell is engineered to downregulate expression of a GOI by genetic modification to reduce expression of said gene, thereby obtaining an expression level of a gene product which is less than the expression of the same gene product prior to said genetic modification or in a comparable cell which does not comprise said genetic modification. “Less than” includes, e.g., at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80, 90% difference, or more. No expression of the gene product is also encompassed by the term “reduction of expression”.

The term “reduction” in the context of gene expression as used herein refers to an experimental approach leading to reduced expression of a given gene compared to expression in a control cell. Downregulation of expression of a gene can be achieved by various experimental means such as introducing nucleic acid molecules into the cell which hybridize with parts of the gene’s mRNA leading to its degradation (e.g., shRNAs, RNAi, miRNAs) or altering the sequence of the gene in a way that leads to reduced transcription, reduced mRNA stability or diminished mRNA translation. A specific approach uses CRISPRi.

The term “increase expression” generally refers to any amount higher than an expression level exhibited by a reference standard, which is a cell prior to the engineering to reduce expression of a certain gene, or which is otherwise expressed in a cell of the same type or species which is not engineered to upregulate expression of said gene. Typically, a cell is engineered to upregulate expression of a GOI by genetic modification to increase expression of said gene, thereby obtaining an expression level of a gene product which is higher than the expression of the same gene product prior to said genetic modification or in a comparable cell which does not comprise said genetic modification. “Higher than” includes, e.g., at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80, 90% difference, or more.

A cell that comprises a perturbed gene expression, such as described herein, is herein also referred to as “modulated”.

The term “modulate” as used herein shall mean to cause or facilitate a qualitative or quantitative change, alteration, or modification in a process, pathway, or phenomenon of interest. Without limitation, such change may be an increase, decrease, or change in relative strength or activity of different components or branches of the process, pathway, or phenomenon. A “modulator” is typically an agent that causes or facilitates a qualitative or quantitative change, alteration, or modification in a process, pathway, or phenomenon of interest. In certain embodiments, the modulator is a transcription modulator, such as a transcription repressor or activator. Modulating expression of a gene may be accomplished or facilitated, for example, by any suitable agent (e.g., a nucleic acid molecule or compound) that causes or facilitates a qualitative or quantitative change, alteration, or modification in the expression of the gene in a subject.

Transcription modulators may be repressors or activators. Transcription modulators typically contain DNA binding domains that recognize and bind recognition sites or sequences in the promoters of transcriptionally active or inactive genes, and also contain activation or repression domains that activate or suppress gene transcription when the transcription modulator binds to the recognition site or sequence. Transcription modulator binding motifs are known in the art. In some embodiments, the transcription modulator is a zinc finger protein. In some embodiments, the transcription modulator belongs to the GLI-Kruppel class of zinc finger proteins. In some embodiments, the transcription modulator is a widely or ubiquitously distributed transcription factor belonging to the GLI-Kruppel class of zinc finger proteins and is involved in repressing and activating a diverse number of promoters. The transcription modulator as disclosed herein is not limited and may be any transcription factor that associates with an enhancer-promoter DNA loop. In some embodiments, the transcription modulator binds to an enhancer and a promoter region of the genome of the cell. In some embodiments, the enhancer and promoter regions are both located in the same insulated neighborhood of the genome of the cell.

The term “lentivirus” (LV) as used herein shall refer to a retrovirus of the genus lentivirus that express reverse transcriptase and optionally integrase. These two enzymes are used to integrate viral RNA into host DNA, allowing for the exploitation of host machinery to express viral genes. Lentiviruses can convert viral RNA into complementary DNA (cDNA) and may integrate a significant amount of viral cDNA into the DNA of a host cell. Lentiviruses can efficiently infect nondividing cells, so they are one of the most efficient methods of gene delivery and are broadly applicable across proliferating and non-proliferating cell types. They can become endogenous, integrating their genome into the host germline genome, so that the virus is henceforth inherited by the host’s descendants.

As used herein, the term “lentivirus”, abbreviated LV, shall also encompass lentiviral particles, and in particular replication-incompetent lentiviral particles.

LV are commonly used as a vector for delivery of a nucleic acid molecule (or “gene”) into the host cell’s genome. Such nucleic acid molecule is herein referred to as a “transgene”. LV can be packaged with a transgene for transgene delivery.

LV can be used as a vector (a “transfer vector” or plasmid) for delivery of a transgene to a host cell genome. According to the classification of the International Committee on Taxonomy of Viruses (ICTV), the genus lentivirus belongs to the family Retroviridae and currently comprise of nine species: seven animal lentiviruses and two human lentiviruses. Animal lentiviruses are bovine immunodeficiency virus (BIV), caprine arthritis encephalitis virus (CAEV), equine infectious anemia virus (EIAV), feline immunodeficiency virus (FIV), puma lentivirus (PLV), simian immunodeficiency virus (SIV) and visna/maedi virus (VMV). Human species are well-known human immunodeficiency virus 1 (HIV-1 ) and human immunodeficiency virus 2 (HIV-2).

To infect human cells, typically, LV of the family Retroviridae and species immunodeficiency virus 1 (HIV) are used.

Lentivirus used for the purpose described herein is typically based on HIV-1 . For safety reasons, the coding regions of the viral proteins are deleted, and the cis-acting regulatory elements are retained. Specifically, the required elements within viruses used for the purpose described herein, are: 5' long terminal repeat (LTR), the psi (ip) packaging signal, the central polypurine tract/chain termination sequence (cPPT/CTS), Rev responsive element (RRE), and 3' LTR, including a polyadenylation signal.

Typically, lentiviral vectors are modified from a wild-type lentivirus such as e.g., HIV-1 , with many of the viral genes removed. Such lentiviral vectors are referred to as “lentiviral particles” or “lentivirus particles”.

Using lentiviral vectors to deliver genes into targeted cells has proven to be a dependable, efficient, and safe method for research.

Specifically, a lentiviral vector only contains the LTRs and the packaging signal, MT In addition, lentiviral packaging genes are provided on separate plasmids (one or more “packaging plasmids”), so the pseudo lentiviral particles are replicationincompetent, with a deficiency to replicate.

Lentiviruses use the gag, pol, and env genes for packaging, revfor exporting viral RNA from the cell core into the cytosol through binding to an RRE, and Tat to activate expression. Infectious transgenic lentiviruses that are replication-incompetent lentiviral particles are specifically produced by a host cell that is transfected with the transfer vector (comprising the LTRs, the packaging signal and optionally a transgene), one or more packaging vector(s) (comprising gag, pol, env, rev) and the presence of Tat. Specific helper plasmids (packaging plasmids) encode for Gag, Pol, Tat and Rev. Rev is included in the packaging plasmids (2^nd and 3^rd generation) and is a core-essential gene for the virus. Rev is required for the post-transcriptional transport of the unspliced viral mRNAs from the nuclei to cytoplasm. The Rev protein binds to the Rev responsive element (RRE).

In particular, the packaging plasmid contains the structural (gag), and replication (pol) genes which code for some of the proteins required to produce the lentivirus. It also encodes the viral env gene, which encodes the envelope protein that defines the tropism (i.e., the range of infectable cells). The envelope plasmids and packaging plasmids provide all of the proteins essential for transcription and packaging of an RNA copy of the expression construct into recombinant pseudoviral particles.

Second generation packaging systems express the HIV gag, pol, env, rev, and tat genes all from a single packaging plasmid.

Third generation packaging systems express gag and pol from one packaging plasmid and env from another. Third generation packaging systems do not express tat. Tat is eliminated from the third generation system through the addition of a chimeric 5’ LTR fused to a heterologous promoter on the transfer plasmid. Expression of the transgene from this promoter is no longer dependent on Tat transactivation. Third generation lentiviral systems are considered safer than second generation systems.

Exemplary second or third-generation packaging systems comprise the following plasmids (referred to as Vectors):

2^nd generation system:

Vector 1 : encodes for gag, pol, rev and tat (e.g., Addgene psPAX2 #12260);

Vector 2: encodes for env (e.g., Addgene pMD2.G #12259);

3^rd generation system:

Vector 1 , encodes for gag and pol (e.g., Addgene pMDLg/pRRE #12251 );

Vector 2: encodes for rev, (e.g., Addgene pRSV-REV #12253);

Vector 3: which encodes for env (e.g., Addgene pMD2.G #12259).

LV preparations can be stored or centrifuged to concentrate virus. Crude or concentrated virus can be used to transduce a target cell. LV preparations produced as described herein can be purified e.g., by ultracentrifugation, precipitation using polyethylene glycol or polyethylene imine, or other methods well-known for preparing purified virus preparations.

Upon infection of a cell, an LV may or may not replicate, depending on whether the virus is replication-incompetent or replication-competent.

The yield of an LV, such as an infectious LV particle, in particular a replicationincompetent infectious particle, can be determined by any of the methods described as follows; qPCR, FACS, ELISA, NGS or hybridization.

A functional and complete LV particle is called a transduction unit, and can be quantified by a qPCR method, in Integrated Genome units (IG). An LV titer can be determined in IG/ml units e.g., as a count of functional particles.

The infectivity of a LV, such as an infectious LV particle, in particular a replicationincompetent infectious particle, can be determined by any of the methods described as follows: qPCR, FACS, ELISA, NGS or hybridization

The infection rate for an LV may depend on the type of target cell. The infection rate is

A lentiviral particle produced as described herein has a certain transduction efficiency, preferably a transduction efficiency of at least 5% or at least 10% e.g., as determined by a FACS readout

The terms “polynucleotide”, “nucleic acid molecule(s)” or “nucleic acid sequence(s)” are interchangeably used herein and refer to nucleotides, either ribonucleotides or deoxyribonucleotides or a combination of both, in a polymeric unbranched form of any length. Preferably, a polynucleotide refers to deoxyribonucleotides in a polymeric unbranched form of any length. Here, nucleotides consist of a pentose sugar (deoxyribose), a nitrogenous base (adenine, guanine, cytosine or thymine) and a phosphate group.

As described herein the genomic perturbation screen is specifically targeting genes that are endogenous to the host cell’s genome, in particular of a wild-type host cell.

The term “endogenous” as used herein is meant to include those molecules and sequences, in particular genes or proteins, which are present in a wild-type (native, not recombinant) host cell or expressed by such wild-type host cell, thereby “endogenous” to said wild-type host cell. In particular, an endogenous nucleic acid molecule (e.g., a gene) or protein that does occur in (and originates from, or can be obtained from) a particular host cell as it is found in nature, is understood to be “host cell endogenous” or “endogenous to the host cell”. Moreover, a cell “endogenously expressing” a nucleic acid or protein expresses that nucleic acid or protein as does a host of the same particular type as it is found in nature. Moreover, a host cell “endogenously producing” or that “endogenously produces” a nucleic acid, protein, or other compound produces that nucleic acid, protein, or compound as does a host cell of the same particular type as it is found in nature.

Upon perturbation and/or genomic modulation of one or more target gene(s), an endogenous nucleic acid or encoded protein can be upregulated or overexpressed, such as to be expressed at a higher level, or downregulated such as to be expressed at a lower level, as compared to the level expressed by a respective wild-type host cell without such engineering. Even if a gene is no more expressed by a host cell, such as e.g., a knockout mutant of the host cell, where the gene is inactivated or deleted, the gene is herein still referred to as “endogenous”.

The term “heterologous” as used herein with respect to a nucleotide sequence, an expression cassette or construct, or any element or part of any one of the foregoing, an amino acid sequence or protein, refers to a compound which is either foreign to a given host cell, i.e. “exogenous”, such as not found in nature in said host cell; or that is naturally found in a given host cell, e.g., is “endogenous”, however, in the context of a heterologous construct or integrated in such heterologous construct, e.g., employing a heterologous nucleic acid fused or in conjunction with an endogenous nucleic acid, thereby rendering the construct heterologous. The heterologous nucleotide sequence as found endogenously may be produced in an unnatural, e.g., greater than expected or greater than naturally found, amount in the cell. The heterologous nucleotide sequence, or a nucleic acid comprising the heterologous nucleotide sequence, possibly differs in sequence from the endogenous nucleotide sequence but may still encode the same protein as found endogenously. Specifically, heterologous nucleotide sequences are those not found in the same relationship to a host cell in nature. Any recombinant or artificial nucleotide sequence is understood to be heterologous. An example of a heterologous polynucleotide is a nucleotide sequence not natively associated with a promoter, e.g., to obtain a truncated or short promoter, or a hybrid promoter, or a hybrid nucleic acid molecule comprising a promoter and a gene switch, or operably linked to a coding sequence, as described herein. As a result, a hybrid or chimeric polynucleotide may be obtained. A further example of a heterologous compound is a protein encoding polynucleotide operably linked to a transcriptional control element, e.g., a promoter, a hybrid promoter, or a hybrid nucleic acid molecule comprising a promoter and a gene switch, to which an endogenous, naturally-occurring protein coding sequence is not normally operably linked.

In some aspects described herein, a diversity of barcodes can be determined in a pool by next generation sequencing.

The term “next generation sequencing” is herein understood as follows.

Next generation sequencing (NGS) is a massively parallel sequencing technology that offers ultra-high throughput, scalability, and speed. The technology is used to determine the order of nucleotides in entire genomes or targeted regions of DNA or RNA.

In some embodiments, the sequencing can be done using an Illumina platform. However, the sequencing and related methods can be adapted to other sequencing platforms that use long single reads or shorter paired-end reads or short single-end reads as well-known to one of ordinary skill in the art.

In some embodiments, the primers used for amplification can be compatible with use in any next generation sequencing platform in which primer extension is used, e.g., Illumina’ s reversible terminator method, Roche’s pyrosequencing method, Life Technologies’ sequencing by ligation (the SOLiD platform), Life Technologies’ Ion Torrent platform or Pacific Biosciences’ fluorescent base-cleavage method.

The sequencing step can be done using any convenient next generation sequencing method and can result in at least 10,000, at least 50,000, at least 100,000, at least 500,000, at least 1 M at least 10M at least 10OM or at least 11 B sequence reads. In some cases, the reads are paired-end reads.

As described herein, targeted DNA-based next generation sequencing techniques can be used to determine a diversity of barcodes in a pool such as in a target cell pool culture, or a fraction thereof.

The term “recombinant” as used herein shall mean “being prepared by or the result of genetic engineering”. A “recombinant cell” or “recombinant host cell” is herein understood as a cell or host cell that has been genetically engineered or modified to comprise a nucleic acid sequence which was not native to said cell. A recombinant host may be engineered to delete and/or inactivate one or more nucleotides or nucleotide sequences, and may specifically comprise an expression vector or cloning vector containing a recombinant nucleic acid sequence, in particular employing nucleotide sequence foreign to the host. A recombinant protein is produced by expressing a respective recombinant nucleic acid in a host. The term “recombinant” as used herein with respect to expression products, includes those compounds that are prepared, expressed, created or isolated by recombinant means, such as isolated from a host cell transformed or transfected to express the expression products.

Certain recombinant host cells are “engineered” host cells which are understood as host cells which have been manipulated using genetic engineering, i.e. by human intervention. Specific examples of an engineered host cell are perturbed host cells as described herein.

When a host cell is engineered for perturbation of a certain gene, the host cell is specifically manipulated such that the host cell has the capability to express such gene and respective protein, respectively, to a different extent compared to the host cell under the same condition prior to manipulation, or compared to the host cells that are not engineered for such gene perturbation.

Therefore, the present invention provides for altering LV production in host cells by engineering a host cell for improved LV production. It turns out that large amounts of lentiviruses can be produced from such engineered production host cell line.

The subject of the invention is particularly based on the use of a reactivated replication-incompetent LV bearing a WT-LTR which is compatible with multiple rounds of infections. Thereby, pooled screens were possible to determine genes regulating the translation, protein processing, packaging, assembly and budding of LV. Specifically, the pooled screen is a functional screen for infectious LV. Specifically, functionality of the LV produced by the pool is determined by a functional screen.

According to a specific example, a constitutive active human host cell line expressing an RPN was used to enable the knockout of different target genes in individual host cells, using a lentivirus preparation containing an sgRNA library thereby obtaining a repertoire of diverse host cells which comprise different knockouts of target genes. In a proof-of-concept experiment, a small-scale genomic perturbation screen was targeting about 400 potential lentiviral restriction factors with about 2.000 lentiviral sgRNAs. The sgRNAs were cloned into the lentiviral plasmid backbone bearing a WT- LTR 3’LTR and served as a barcode identifying the respective perturbed gene. Upon reactivation of the LV comprising the sgRNA library with helper plasmids expressing gag, pel, env, and rev, alongside with tat, the produced LV was harvested and used to infect another cell line, the target cell line. The helper plasmids encode the proteins used for the packaging of the virus wherefore they are also called packaging (or helper) plasmids

Transduced target cells containing the re-activated virus were enriched, the genomic DNA was harvested and the sgRNAs were amplified by PCR and quantified by deep sequencing. As a result, a series of factors were found to increase LV yield when being the target of the KO in the host cell, among them known viral restriction factors (which validate the screen), but also several factors that were previously not found relevant regarding viral production.

The foregoing description will be more fully understood with reference to the following examples. Such examples are, however, merely representative of methods of practicing one or more embodiments of the present invention and should not be read as limiting the scope of invention.

The invention is further described by one or more of the following items.

1 . A method of identifying a lentiviral modulation factor in the genome of a host cell which, upon its perturbation in the genome of the host cell, results in the production of replication-incompetent lentiviral particles at altered yield or infectivity, comprising: a) a genomic perturbation screen, thereby obtaining a pooled repertoire of perturbed host cells to cover perturbation of a diversity of genes; b) utilizing the repertoire of perturbed host cells to produce a lentiviral pool of replication-incompetent lentiviral particles in a perturbed host cell pool culture, where host cells bearing perturbations of lentiviral modulation factors produce lentiviruses with altered yield or infectivity; c) infecting a target cell line with said lentiviral pool and culturing the infected target cells in a target cell pool culture; and d) identifying a lentiviral modulation factor which, upon the respective perturbation of the host cell, has resulted in the production of lentiviruses at altered yield or infectivity against the target cells, wherein the lentiviruses produced by the perturbed host cells comprise a respective barcode, and the lentiviral modulation factor is identified by determining the relative amount of the barcodes in the target cell pool culture.

2. The method of item 1 , wherein said genomic perturbation screen comprises an RNA-guided programmable nuclease (RPN)-mediated perturbation screen using a lentivirus harboring a library of single guide RNAs (sgRNAs), which are designed to target the diversity of genes in the host cell genome, wherein the repertoire of perturbed host cells is created using host cells containing the RPN, and wherein the lentiviral modulation factor is identified by determining a relative alteration of the respective sgRNA or its coding sequence, which is used as the barcode, compared to other sgRNAs in the target cell pool culture.

3. The method of item 2, wherein the RPN mediated perturbation screen is a CRISPR-Cas9 screen, preferably a CRISPR knockout, CRISPR activation, or CRISPR interference screen.

4. The method of any one of items 1 to 3, wherein the lentiviral modulation factor is identified by determining the relative amount of the barcode in the target cell pool culture, preferably by PCR amplification and next generation sequencing.

5. The method of any one of items 1 to 4, wherein a relative increase of a barcode is indicative of the respective lentiviral modulation factor that has induced a higher yield or infectivity of lentivirus produced by the respective perturbed host cell, preferably wherein the host cell bearing the perturbation of the identified lentiviral modulation factor produces lentiviruses with at least 1 .2-fold increased yield or infectivity compared to the host cell without such perturbation.

6. The method of any one of items 1 to 5, wherein the lentiviral modulation factor is a wild-type gene that is endogenous to the host cell, preferably modulating attachment, entry, uncoating, fusion, reverse transcription, integration, transcription, translation, packaging, trafficking, assembly, secretion or budding off of the lentivirus 7. The method of any one of items 1 to 6, which comprises at least two cycles of lentiviral infection, wherein a) in a first cycle, said repertoire of perturbed host cells is generated by infection with replication-incompetent lentiviral particles in pooled culture, thereby producing a pool of host cells harboring lentiviral donor templates; b) said pool of host cells harboring lentiviral donor templates produced by the first cycle is activated, thereby obtaining said lentiviral pool of activated lentivirus particles; and c) in a further cycle, the target cells are infected with the lentiviral pool of activated lentivirus particles in the target cell pool culture.

8. The method of item 7, wherein said said repertoire of perturbed host cells harboring lentiviral donor templates is bearing a variety of perturbations introduced by an RPN and a pooled library of sgRNAs, which are designed to target the diversity of genes in the host cell genome.

9. The method of item 7 or 8, wherein the replication-incompetent lentiviral particles are assembled by utilizing a lentiviral donor template harboring a wild-type intact 3’ long terminal repeat (LTR) to exchange for a mutated or truncated 3’ LTR as previously used in replication-incompetent lentiviral particles.

10. The method of item 9, wherein for said activation of the lentiviral donor templates, gag, pol, env, rev and tat are delivered to the pool of host cells harboring lentiviral donor templates, preferably wherein said gag, pol, env, rev and tat, are delivered on two or more separate plasmids.

Specifically, said gag, pol, env, and rev are delivered in the presence of tat.

Specifically, said gag, pol, env, and rev are delivered, alongside with tat in trans.

11 . The method of any one of items 1 to 10, wherein the lentiviral pool is harvested from the perturbed host cell pool culture and transferred to the target cell pool culture.

12. The method of any one of items 1 to 1 1 , wherein the repertoire of perturbed host cells is of packaging host cell lines suitable for producing lentiviral particles for delivery of a transgene to a target cell, preferably wherein the host cells are human cells such as selected from the group consisting of HEK cells, such as HEK293T 293FT, Lenti-X, or 293SF-3F6

13. The method of any one of items 1 to 12, wherein the target cells are human cells, preferably primary cells or stem cells, preferably selected from the group consisting of T cells, B cells, Macrophages, PBMCs and iPS cells. 14. A library comprising a repertoire of perturbed host cells which covers perturbation of a diversity of at least 10 different genes, wherein the host cells are transduced with replication-incompetent lentiviral delivery particles.

15. Use of the library of item 14, for identifying a lentiviral modulation factor which upon its perturbation in a producer host cell results in altered yield production of replication-incompetent lentiviral delivery particles.

16. A method of engineering a producer host cell line producing lentiviral delivery particles at high titer or high infectivity, comprising: a) identifying a lentiviral modulation factor in the genome of a host cell according to the method of any one of items 1 to 12, wherein the respective perturbation of the host cell has resulted in the production of lentiviruses at higher yield or infectivity against the target cells; and b) engineering a producer host cell line comprising said perturbation of the identified lentiviral modulation factor.

17. The method of item 16, wherein the producer host cell line produces lentiviruses with at least 1 .2-fold increased yield or infectivity compared to the host cell without said perturbation of the identified lentiviral modulation factor.

18. A method of producing lentiviral delivery particles in a producer host cell line produced according to item 16 or 17, preferably wherein said lentiviral delivery particles are replication-incompetent.

19. The method of item 18, wherein said lentiviral delivery particles are packaged with a transgene for delivery to a target cell genome, preferably wherein said transgene encodes a chimeric antigen receptor (CAR) for T-cell immunotherapy, a T cell receptor (TCR) for T-cell immunotherapy, or a transgene for gene replacement therapy.

EXAMPLES

EXAMPLE 1 : SpCas9-HEK293T cell line

A constitutive active SpCas9-HEK293T cell line was generated to enable the knockout of single genes. Therefore, a SpCas9-Blasticidin cassette (SEQ ID NOU ) was stably integrated into HEK293T cell (ATCC-CRL-3216) via a lentivirus generated with the 2^nd generation lentivirus production system comprising the helper plasmids psPAX2 (Addgene #12260) and pMD2.G (Addgene #12259). The HEK293T cells were selected with Blasticidin via concentration of 8pg/m I . As a next step, single clones were tested for SpCas9 activity via the knockout of the surface marker beta-2-microglobulin (B2M). To this end, SpCas9-HEK293T clones got transduced with a lentivirus containing an sgRNA (GGCCGAGATGTCTCGCTCCG, SEQ ID NO:5) against the surface marker B2M to observe a KO efficiency. After transduction, SpCas9-HEK293T were kept for 3 days on Puromycin (1 ,5ug/ml) to select for the presence of the guide RNA expression vector and tested via fluorescence-activated cell sorting (FACS) to observe a reduction of the surface receptor B2M inflicted by SpCas9 and the B2M sgRNA. We detected a knockout efficiency of about 80% for the best performing clones (clone 1 & 2) (Fig. 1 ). In addition, we performed a growth assay (Fig. 2a) and bulk RNA sequencing (Fig. 2b) to assess whether the introduction of Cas9 had altered the cell. Two clones were selected for future work. Both of these had high Cas9 activity and were highly similar (Spearman correlation of >98%) to the parental HEK293T cell population.

EXAMPLE 2: Test of the Wild-type long terminal repeat system

A lentivirus backbone plasmid comprising cPPT, RRE and Psi sequence was used to exchange the 3’ truncated LTR (SEQ ID NO:1 1 ) with a wild type LTR (WT-LTR) (SEQ ID NO:12). In addition, the truncated 5’ LTR, coupled to a promoter placed upstream of the 5’ LTR (SEQ ID NO:2) was exchanged with a WT-LTR (SEQ ID NO:3). The WT-LTR template vector (SEQ ID NO:4) was integrated with the 2^nd generation lentivirus system into the previous generated SpCas9-HEK293T via 2 helper plasmids (Addgene #12259, #12260).

For a first test, the lentiviral donor template harboring WT-LTRs (both at the 5’ and the 3’end) was compared to the conventional lentiviral donor template harboring truncated LTRs. To this end, the lentiviral donor template harbored an expression cassette for a guide RNA targeting B2M (see example 1 ). Infection of a HEK293T cell harboring SpCas9 with a lentivirus bearing this guide RNA should lead to a B2M gene knockout in the recipient cell.

To test this, the lentiviral donor templates (bearing either WT-LTRs or truncated LTRs) were combined with suitable packaging plasmids in HEK293T. Briefly, 10 million cells within a 10cm dish were transfected with plasmids (Addgene #12259, #12260 and vector SEQ ID NO:4 flanked with WT-LTRs or vector SEQ ID NO:4 flanked with conventional lentiviral donor template) using polyethylenimine (PEI) according to manufacturer’s conditions. Following a media change 24 hours post transfection, two batches of virus were harvested 48h and 72h post transfection and filtered the combined batches through 0,45 pm PVDF filter (Sartorius). Active lentiviral particles were enriched by precipitation using polyethylene glycol (Sigma-Aldrich) and 4M NaCL Then, active virus was transferred onto HEK293T cells bearing SpCas9, thus infecting the cells and delivering the B2M guide RNA. Cells were selected with 1 ,5 g/ml Puromycin to enrich for cells bearing the guide RNA and were assessed by flow cytometry 72h post infection.

As expected, an efficient gene knockout was observed using either the WT-LTR or the truncated LTR (Figure 3). This suggested that, during the first round of lentivirus generation, an active viral species was produced which was capable of infecting the target cell (here HEK293T harboring SpCas9).

Next, it was tested if the WT-LTR lentiviral system allows the re-activation of an integrated lentivirus following the addition of the helper plasmids. To this end, cells described above harboring a guide RNA against B2M in the context of WT-LTRs or truncated LTRs were combined with the viral helper plasmids Addgene #12259 and #12260 using polyethylenimine (PEI) according to manufacturer’s conditions. Following a media change 24 hours post transfection, two batches of virus were harvested 48h and 72h post transfection and filtered the combined batches through 0,45pm PVDF filter (Sartorius). Active lentiviral particles were enriched by precipitation using polyethylene glycol (Sigma-Aldrich) and 4M NaCL If the second round of infection were successful, it was expected to see a B2M gene knockout in the recipient cells bearing SpCas9. Note that, here, during the second round of infection, SpCas9 is not formally needed. It is merely used as an indicator of efficiency.

A lentivirus with a 3’ LTR truncation was included as a reference point. In this condition, all cells succumbed to cell death following selection with 1 ,5pg/ml Puromycin. This is to be expected as a lentiviral donor template harboring truncated LTRs cannot produce any active virus upon provision of viral helper genes. In contrast, the activation of the lentiviral donor template harboring WT-LTRs gave rise to active virus that could be used to infect the target cells in the second of infection. Infection efficiency was judged by flow cytometry, using the knockout of B2M which was caused by the guide RNA delivered by the LV as an indirect measure of virus production. Specifically, in this condition, it was observed that, upon the delivery of a B2M guide RNA using the WT- LTR architecture, 66% of the target cells were B2M negative, in contrast to 0% of the target cells in a condition in which a non-targeting guide RNA had been delivered using the same approach.

In summary, this experiment showed that a virus with WT-LTRs can produce an efficient knockout, can be re-activated after stable integration and accurately assembles to generate again a knockout in a second round of infection (Fig. 4; 2^nd cell line, stable Cas9-Jurkat cells (ATCC Clone E6-1 - TIB-152). This was a key finding to establish the screening paradigm outlined above.

EXAMPLE 3: WT-LTR proof of concept screen:

Having established the feasibility of the approach, a small-scale proof-of-concept (POC) screen was performed, in which about 400 potential lentiviral restriction factors were targeted with about 2.000 lentiviral sgRNAs. The sgRNAs were cloned into the WT- LTR lentiviral plasmid backbone and sgRNA representation was validated via deep sequencing, ensuring that the library displayed a uniform representation of all sgRNAs within the WT-LTR vector (data not shown). Furthermore, starting from the cloned plasmid sgRNA library (TWIST Bioscience), a lentiviral sgRNA pool was produced, utilizing LentiX (Takara Bio) as packaging cells. LentiX cells were transfected via PEI with the WT-LTR vector comprising the library of about 2000 sgRNAs and the two viral helper plasmids Addgene #12259 and #12260. Following a media change 24 hours post transfection, two batches of virus were harvested 48h and 72h post transfection, combined and filtered through a 0,45 pm PVDF filter (Sartorius). Active lentiviral particles were enriched by precipitation using polyethylene glycol (Sigma-Aldrich) and 4M NaCL As a next step the SpCas9-HEK293T cell line was transduced with the viral sgRNA library at an MOI of 0.1 (delivering one sgRNA per single cell on average) and selected via Puromycin at a concentration of 1 .5ug/ml to enrich for cells harboring a sgRNA, and a pool of cells comprising the sgRNA library was established (Fig. 5; 3 replicates). Cells were selected and expanded for a total of 12 days in order to establish a stable knockout of the respective genes targeted by the sgRNA library. To monitor which sgRNA were present within this pool of cells, prior to the re-activation phase, a fraction of the transduced and selected SpCas9-HEK293T cells were collected at day 12 and genomic DNA was extracted. Subsequent, on the genomic DNA, nested PCRs were performed to enriched via specific primers (SEQ ID NO:6) the sgRNAs located within the lentivirus. For all 3 replicates (“Input samples”), almost all sgRNAs from the initial library were recovered outlining a high coverage of the sgRNA library within the cells (Fig.7 “Input samples”). As a next step, the SpCas9-HEK293T cells bearing lentiviral donor templates (WT-LTR) were re-seeded and transfected with 2 helper plasmids (Addgene #12259 and #12260) which enables the re-activation of the integrated lentivirus cassette comprising the sgRNA library. The produced lentivirus was harvested from the cellular supernatant 48h and 72h post transfection, combined and filtered through a 0,45 pm PVDF filter (Sartorius). Active lentiviral particles were enriched by precipitation via polyethylene glycol (Sigma-Aldrich) and 4M NaCI and used to infect a 2^nd cell line (HEK293T ATCC-CRL-3216). This 2^nd cell line was again selected via Puromycin to enrich for transduced cells containing the re-activated virus. Subsequently, the genomic DNA (“Output samples”) of the 2^nd cell line was harvested after a total of 9 days, and the sgRNAs was again enriched for deep sequencing (Fig. 7 “Output samples”).

After analysis of the input vs. output samples, a significant enrichment for the genes TRIM25 and ZC3HAV1 (FDR <=0.05) was identified within the 2^nd cell line (Fig 8). This suggests that knockouts for TRIM25 and ZC3HAV1 , introduced into the producer cell (HEK293T bearing SpCas9), produced higher titres of infectious virus, leading to an enrichment of these guide RNAs upon infection of target cells. This is in agreement with published literature on the TRIM25 and ZC3HAV1 (also named: ZAP) which suggests that these genes are involved in antiviral defence and therefore represent potential roadblocks for viral production (TRIM25: Sanchez et al., 2018; Choudhury et al., 2022; ZC3HAV1 : Kozaki et al., 2015). It was noted that 1 ) an sgRNA targeting eGFP dropped out of the experiment (Fig 8 “GFP”). This can be explained as the lentiviral vectors harbors an eGFP-2A-PuroR transgene. Targeting eGFP with a sgRNA will lead to the destruction of the eGFP-2A-PuroR open reading frame and, following puromycin selection, will lead to the elimination of these cells from the pool; 2) essential genes dropped out; and 3) lentiviral restriction genes and non-essential genes were enriched.

EXAMPLE 4: Knock-out of lentiviral modulation factors that were identified in the proof of concept screen

SpCas9-HEK293T clone 1 was transduced, utilizing the 2^nd generation lentivirus system with a lentivirus donor template encoding for the fluorophore marker GFP, the resistance gene Puromycin and either sgRNAs against TRIM25 (SEQ ID NO:13; SEQ ID:14) or a sgRNA against ZC3HAV1/ZAP (SEQ ID NO:15). The transduced cells were selected with Puromycin via a concentration of 1 .5pg/ml and screened after 10 days via FACS, observing 100% GFP signal within the selected cell population (datapoint not shown). As a next step, the SpCas9-HEK293T clone 1 - TRIM25-knockout cell lines, SpCas9-HEK293T clone 1 - ZC3HAV1/ZAP-knockout cell line and SpCas9-HEK293T clone 1 control cell line were transfected with 2 helper plasmids (Addgene #12259 and #12260) and a 2^nd generation lentivirus donor template encoding for GFP and a NTC sgRNA. The produced lentiviruses were harvested from the cellular supernatant 48h and 72h post transfection, combined and filtered through a 0,45 pm PVDF filter. Afterwards, HeLa-WT (Fig.9a) or HEK-WT cells (Fig.9b) were transduced with the produced lentiviruses and GFP levels were measured within HeLa-WT (Fig.9a) or HEK-WT cells, indicating the amount of lentivirus generated in the producer cell lines. A clear difference was observed for the SpCas9-HEK293T clone 1 (20% GFP signal) and the individually generated knockout cells lines SpCas9-HEK293T - TRIM25 KO sgRNA 1 and 2 and ZC3HAV1/ZAP KO sgRNA 1 (~35-40% GFP signal), which showed up to 20% more GFP signal within the transduced HeLa-WT or HEK-WT cells. These data suggest that the knockout of the lentiviral repressing factors TRIM25 or ZC3HAV1 can lead to more and reproducible (sgRNA 1 and 2 against TRIM25 showed similar improvements) lentivirus production in producer cell lines.

Example 5: Boosting the re-activation process of a lentivirus

Here, a process was established that allows CRISPR screens for lentiviral infection phenotypes. Importantly, this process does not involve a replication-competent virus, which would be prohibitive for safety reasons. Instead, it involves multiple cycles of infection and “reactivation”: During the infection cycle, a replication-deficient virus is made in producer cells and utilized to infect suitable target cells, where it integrates. During the reactivation cycle, the cells bearing the integrated lentivirus are combined with suitable helper plasmids to assemble and secrete another round of virus.

During the establishment of the infection-reactivation workflow, a low reactivation efficiency of the lentivirus was observed after the addition of the helper plasmids, i.e. viral yields from the reactivation cycle were considerably lower than those obtained during the infection cycle. This represents a bottleneck for the screening approach presented here as low-efficiency reactivation requires very large numbers of cells to ensure an accurate sgRNA representation.

In general, lentiviral vectors are replication-incompetent after integration due to two main features: (i) the helper genes (gag, pol and env) are provided on separate plasmids and are thus not present in the infected cell and (ii) a truncation within the 3’LTR prevents the LTR from initiating viral transcription (Fig. 10a). A viral re-activation can be enabled in multiple ways:

1 ) the placement of a promoter within the 3’LTR (Fig. 10b; “LeAPs Pro”). After a rolling circle amplification and integration of the lentivirus into the genome, the 3’LTR promoter lands upstream of the viral cassette and enables the transcription of the viral cassette.

2) The exchange of the truncated 3’ LTR to a wild-type LTR (Fig. 10c, Fig 10d). For this experiment, two vectors were generated utilizing this wild-type LTRs strategy which contain either a strong internal promoter (Fig. 10c EF1 a) or weak internal promoter (Fig.10d; EF1 s) also referred to as EFS; SEQ ID NO:16; see also SEQ ID NO:1.

To this end, the four configurations outlined in Figure 10 (a-d) were compared. As control cells were taken along which didn’t receive a virus or cells which received a replication deficient 3’ truncated virus (Fig. 10a).

During the first round (infection round), HEK293 WT cells were transduced and a high transduction efficiency was observed for all approaches (Fig. 11 a). This suggests that all architectures are in principle functional and can give rise to active viruses. Following selection via puromycin (at a concentration of 1.5 pg/ml) suitable helper plasmids (psPAXI and pMD2.G) were transfected into the HEK293 WT cells to enable the reactivation of the virus. Importantly, reactivation requires the tat transactivator which was also included as described before. Next, a fresh batch of HEK293T WT cells was transduced with the re-activated virus using a cationic polymer “polybrene” (at a concentration of 4 pg/ml) and cells were selected via puromycin at a concentration of 1.5 pg/ml. Then, surviving cells were assessed using crystal violet staining. Cells surviving puromycin selection must have been infected with a lentivirus, hence the number of cells on the dish should be proportional to the viral titre produced during the reactivation step.

The control cells died throughout the selection process, outlining that no viral reactivation occurred in the configuration where the truncated LTR had been used. In contrast, all the three re-activation approaches gave rise to surviving cells, suggesting that all three approaches had worked to some extent.

To enable a detailed quantitative comparison, surviving cells from these three batches were analyzed by flow cytometry (Figure 12). The original [configuration (WT- LTR + EF1 a) lead to -20% marker-positive cells and so did the LeAPs Pro approach. Strikingly, the approach in which the WT-LTR had been combined with the weaker EFS promoter led to significantly higher reactivation rates, yielding around 74% of markerpositive cells. This 5.6-fold increase of re-activated particles outlines a successful improvement.

Example 6: Genome-wide screen

It was aimed to conduct an unbiased genome-wide screen to identify novel targets involved in lentiviral production, assembly or secretion. Therefore, a genome-wide knockout library comprising 76.500 sgRNAs targeting 19.027 genes with 4 sgRNAs/gene on average was designed. The sgRNAs were cloned into the WT-LTR lentiviral plasmid backbone and validated via deep sequencing, ensuring that the library displayed a uniform representation of all sgRNAs within the WT-LTR vector (data not shown). Furthermore, starting from the cloned plasmid sgRNA library a lentiviral sgRNA pool was produced utilizing LentiX as packaging cells. As a next step 130 million HEK293T-SpCas9 cells were transduced for each of the 3 replicates (Fig.13a) with the viral sgRNA library at an MOI of 0.4 (represents 1 -2 sgRNA per single cell) and selected via puromycin (cone.: 1.5ug/ml) to enrich for transduced cells and establish a pool of cells comprising the sgRNA library (Fig.13b). Cells were selected and expanded for a total of 12 days, which allows the stably integrated SpCas9 protein together with the sgRNA to induce gene knockouts. To monitor which sgRNAs are present within the pool of cells before the re-activation phase, a fraction of the transduced and selected HEK293T-SpCas9 cells were collected, and genomic DNA was extracted (“Input samples”). Subsequent, on the genomic DNA, nested PCRs were performed to enriched via specific primers (SEQ ID NO:17-20) the sgRNA-encoding element located within lentiviral donor sequence.

As a next step, SpCas9-HEK293T cells were re-seeded and transfected with 2 helper plasmids (psPAX2 and pMD2.G) which enables the re-activation of the integrated lentivirus cassette comprising the sgRNA library. The produced lentivirus was harvested from the cellular supernatant and used to infect HEK293 WT cells. This 2nd cell line was again selected via puromycin to enrich for transduced cells containing the re-activated virus. Subsequently the genomic DNA (“Output samples”) of the 2nd cell line was harvested after a total of 4 days and the sgRNA sequences were amplified by PCR. Next, PCR products from the “Input samples” and the “Output samples” (triplicates each) were analyzed by NGS. Analysis of the data showed a high coverage of the sgRNA library with an overall coverage of 500-1000 reads/sgRNA (data not shown) as well as a high correlation of the 3x input (Spearman correlation >0.8) and 3x output samples (Spearman correlation >0.65) (Fig.14).

When analyzing the data in more detailed and comparing it to the POC screen (Example 3), the top 3 hits from the POC screen amongst the 200 most significant hits were identified from the genome-wide screen. More specifically, ZAP (ZC3HAV1 ) was found on rank 13, POU5F1 was found on rank 22 and TRIM25 was found on rank 126. This suggests that the genome-wide screen was robust and confirmed some of the key factors identified in the POC screen.

Example 7: Validation screen

Due to the large number of hits identified in the genome-wide screen, a validation screen was set up based on the results of the genome-wide screen. Specifically, sgRNAs targeting 1 ) the top 400 genome-wide identified “anti-viral” genes were included. These restriction factors should inhibit lentivirus production, assembly or secretion and the cognate gene knockout is thus expected to produce more virus. 2) 100 neutral/non-essential controls and 3) 100” pro-viral" genes which are required for lentiviral production, assembly or secretion. If these genes are knocked out, less virus should be made. The 2.500 sgRNAs targeting these genes were cloned into the WT- LTR lentiviral plasmid backbone comprising the EFS promoter and were validated via deep sequencing, ensuring that the library displayed a uniform representation of all sgRNAs within the WT-LTR + EFS vector (data not shown). Since the re-activation workflow is compatible with multiple rounds of infection and reactivation, it was set out to perform three rounds of transduction and re-activation of the lentiviral sgRNA library to potentially increase the significance of the identified hits. 8.5 million SpCas9- HEK293T cells were transduced for each of the 3 replicates with the viral sgRNA library at an MOI of 0.4 (represents 1 -2 sgRNA per single cell) and selected via puromycin (cone.: 1.5ug/ml) to enrich for transduced cells and establish a pool of cells comprising the sgRNA library (data not shown). The selection and expansion of cells was done for a total of 12 days, which allows the stably integrated SpCas9 protein together with the sgRNA to establish a knockout of a target gene. To monitor which sgRNA are present within this pool of cells a fraction of the transduced and selected SpCas9-HEK293T cells got collected and genomic DNA was extracted (“Input samples”). Subsequently, on the genomic DNA, nested PCRs were performed to enriched via specific primers (SEQ ID NO:17-20) the sgRNA-encoding region located within the lentivirus.

For all 3 replicates (“Input samples”), almost all sgRNAs from the initial library were recovered outlining a high coverage of the sgRNA library within the cells (data not shown). As a next step, HEK293T cells were re-seeded and transfected with two helper plasmids (psPAX2 and pMD2.G) which enables the re-activation of the integrated lentivirus cassette comprising the sgRNA library. The produced lentivirus was harvested and used for a second round of infection of naive HEK293T cells bearing SpCas9. These cells were again selected via puromycin to enrich for transduced cells containing the reactivated virus. Subsequently the genomic DNA (“Output samples”) was harvested from a fraction of the cells after a total of 12 days and amplified the sgRNAs by PCR for deep sequencing.

Importantly, 3 rounds of transduction and re-activation were performed and 3x Input and 9x Output samples (3 rounds x 3 replicates were obtained; Fig. 15). Analysis of the deep sequencing data showed a high correlation of the 3x input (Spearman correlation >0.9) as well as a correlation within the output samples (Spearman correlation >0.7) (Fig. 16). Moreover, a high correlation of the log-fold changes (LFC) was seen upon comparing the cumulative LFC over the 3 rounds of the validations screen vs. genome-wide screen (Fig.17). There was a significant increase in viral infection rates upon knockout of anti-viral genes (restriction factors), no effect within non- essential control genes and a decrease in viral infection rates due to the knockout of pro-viral genes (host factors) (Fig.18). In total, 44 genes were identified with a false discovery rate (FDR) <0.05 that significantly affected lentiviral production, assembly or secretion. The top hit was (ZAP) ZC3HAV1 , outlining the importance of this gene as well as the reproducibility of the screen. These 44 genes appear to represent the core machinery of host genes that is present and active in HEK293T cells and whose modulation will affect the production, assembly, and secretion of lentiviruses. REFERENCES

Ghassemi S, Nunez-Cruz S, O'Connor RS, Fraietta JA, Patel PR, Scholler J, Barrett DM, Lundh SM, Davis MM, Bedoya F, Zhang C, Leferovich J, Lacey SF, Levine BL, Grupp SA, June CH, Melenhorst JJ, Milone MC. Reducing Ex Vivo Culture Improves the Antileukemic Activity of Chimeric Antigen Receptor (CAR) T Cells. Cancer Immunol Res. 2018 Sep;6(9):1 100-1109. doi: 10.1 158/2326-6066.CIR-17-0405. Epub 2018 Jul 20. PMID: 30030295; PMCID: PMC8274631.

Escors D, Breckpot K. Lentiviral vectors in gene therapy: their current status and future potential. Arch Immunol Ther Exp (Warsz). 2010 Apr;58(2):107-19. doi: 10.1007/S00005-010-0063-4. Epub 2010 Feb 9. PMID: 20143172; PMCID: PMC2837622.

Ivica NA, Young CM. Tracking the CAR-T Revolution: Analysis of Clinical Trials of CAR-T and TCR-T Therapies for the Treatment of Cancer (1997-2020). Healthcare (Basel). 2021 Aug 19;9(8):1062. doi: 10.3390/healthcare9081062. PMID: 34442199; PMCID: PMC8392279.

Albinger, N., Hartmann, J. & Ullrich, E. Current status and perspective of CAR-T and CAR-NK cell therapy trials in Germany. Gene Ther 28, 513-527 (2021 ). https://doi.Org/10.1038/S41434-021 -00246-w

Sterner, R.C., Sterner, R.M. CAR-T cell therapy: current limitations and potential strategies. Blood Cancer J. 1 1 , 69 (2021 ). https://doi.org/10.1038/s41408-021 -00459-7

Hucks, G., Rheingold, S.R. The journey to CAR T cell therapy: the pediatric and young adult experience with relapsed or refractory B-ALL. Blood Cancer Journal 9, 10 (2019). https://doi.Org/10.1038/S41408-018-0164-6

Levine BL, Miskin J, Wonnacott K, Keir C. Global Manufacturing of CAR T Cell Therapy. Mol Ther Methods Clin Dev. 2016 Dec 31 ;4:92-101. doi: 10.1016/j.omtm.2016.12.006. PMID: 28344995; PMCID: PMC5363291.

Labbe RP, Vessillier S, Rafiq QA. Lentiviral Vectors for T Cell Engineering: Clinical Applications, Bioprocessing and Future Perspectives. Viruses. 2021 Aug 2;13(8):1528. doi: 10.3390/v13081528. PMID: 34452392; PMCID: PMC8402758.

Hein, M.Y., Weissman, J.S. Functional single-cell genomics of human cytomegalovirus infection. Nat Biotechnol 40, 391-401 (2022). https://doi.Org/10.1038/S41587-021 -01059-3

Li, B., Clohisey, S.M., Chia, B.S. et al. Genome-wide CRISPR screen identifies host dependency factors for influenza A virus infection. Nat Commun 11 , 164 (2020). https://doi.Org/10.1038/S41467-019-13965-x

Zhu, Y., Feng, F., Hu, G. et al. A genome-wide CRISPR screen identifies host factors that regulate SARS-CoV-2 entry. Nat Commun 12, 961 (2021 ). https://doi.Org/10.1038/S41467-021 -21213-4

Zufferey R, Dull T, Mandel RJ, Bukovsky A, Quiroz D, Naldini L, Trono D. Selfinactivating lentivirus vector for safe and efficient in vivo gene delivery. J Virol. 1998 Dec;72(12):9873-80. doi: 10.1 128/JVI.72.12.9873-9880.1998. PMID: 9811723; PMCID: PMC1 10499.

Sauter D, Kirchhoff F. (2021 ) Evolutionary conflicts and adverse effects of antiviral factors eLife 10:e65243.

Pagani I, Demela P, Ghezzi S, Vicenzi E, Pizzato M, Poli G. Host Restriction Factors Modulating HIV Latency and Replication in Macrophages. Int J Mol Sci. 2022 Mar 11 ;23(6):3021 . doi: 10.3390/ijms23063021 . PMID: 35328442; PMCID: PMC8951319.

Chemudupati M, Kenney AD, Bonifati S, Zani A, McMichael TM, Wu L, Yount JS. From APOBEC to ZAP: Diverse mechanisms used by cellular restriction factors to inhibit virus infections. Biochim Biophys Acta Mol Cell Res. 2019 Mar;1866(3):382-394. doi: 10.1016/j.bbamcr.2O18.09.012. Epub 2018 Oct 2. PMID: 30290238; PMCID: PMC6334645.

Boso, G.; Kozak, C.A. Retroviral Restriction Factors and Their Viral Targets: Restriction Strategies and Evolutionary Adaptations. Microorganisms 2020, 8, 1965. https://doi.Org/10.3390/microorganisms8121965

Jiaying Han, Kevin Tam, Curtis Tam, Roger P. Hollis, Donald B. Kohn, Improved lentiviral vector titers from a multi-gene knockout packaging line, Molecular Therapy - Oncolytics, Volume23, 2021 ,Pages582-592,ISSN2372-7705, https://doi.Org/10.1016/i.omto.2021 .1 1 .012.

Dobson, C.S., Reich, A.N., Gaglione, S. et al. Antigen identification and high- throughput interaction mapping by reprogramming viral entry. Nat Methods 19, 449-460 (2022). https://doi.Org/10.1038/S41592-022-01436-z

Gilbert LA, Larson MH, Morsut L, Liu Z, Brar GA, Torres SE, Stern-Ginossar N, Brandman O, Whitehead EH, Doudna JA, Lim WA, Weissman JS, Qi LS. CRISPR- mediated modular RNA-guided regulation of transcription in eukaryotes. Cell. 2013 Jul 18;154(2):442-51 . doi: 10.1016/j.cell.2O13.06.044. Epub 2013 Jul 11. PMID: 23849981 ; PMCID: PMC3770145.

Nunez JK, Chen J, Pommier GC, Cogan JZ, Replogle JM, Adriaens C, Ramadoss GN, Shi Q, Hung KL, Samelson AJ, Pogson AN, Kim JYS, Chung A, Leonetti MD, Chang HY, Kampmann M, Bernstein BE, Hovestadt V, Gilbert LA, Weissman JS. Genome-wide programmable transcriptional memory by CRISPR-based epigenome editing. Cell. 2021 Apr 29;184(9):2503-2519.e17. doi: 10.1016/j.cell.2O21 .03.025. Epub 2021 Apr 9. PMID: 338381 1 1 ; PMCID: PMC8376083.

Margolin JF, Friedman JR, Meyer WK, Vissing H, Thiesen HJ, Rauscher FJ 3rd. Kruppel-associated boxes are potent transcriptional repression domains. Proc Natl Acad Sci U S A. 1994 May 10;91 (10):4509-13. doi: 10.1073/pnas.91 .10.4509. PMID: 8183939; PMCID: PMC43815.

Wolf G, Greenberg D, Macfarlan TS. Spotting the enemy within: Targeted silencing of foreign DNA in mammalian genomes by the Kruppel-associated box zinc finger protein family. Mob DNA. 2015 Oct 2;6:17. doi: 10.1 186/s13100-015-0050-8. PMID: 26435754; PMCID: PMC4592553.

Datlinger P, Rendeiro AF, Schmidl C, Krausgruber T, Traxler P, Klughammer J, Schuster LC, Kuchler A, Alpar D, Bock C. Pooled CRISPR screening with single-cell transcriptome readout. Nat Methods. 2017 Mar;14(3):297-301 . doi:

10.1038/nmeth.4177. Epub 2017 Jan 18. PMID: 28099430; PMCID: PMC5334791.

Ran FA, Hsu PD, Lin CY, Gootenberg JS, Konermann S, Trevino AE, Scott DA, Inoue A, Matoba S, Zhang Y, Zhang F. Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity. Cell. 2013 Sep 12;154(6):1380-9. doi: 10.1016/j.cell.2O13.08.021. Epub 2013 Aug 29. Erratum in: Cell. 2013 Oct 10;155(2):479-80. PMID: 23992846; PMCID: PMC3856256.

Sanchez JG, Sparrer KMJ, Chiang C, Reis RA, Chiang JJ, Zurenski MA, Wan Y, Gack MU, Pornillos O. TRIM25 Binds RNA to Modulate Cellular Anti-viral Defense. J Mol Biol. 2018 Dec 7;430(24):5280-5293. doi: 10.1016/j.jmb.2O18.10.003. Epub 2018 Oct 17. PMID: 30342007; PMCID: PMC6289755.

Choudhury NR, Trus I, Heikel G, Wolczyk M, Szymanski J, Bolembach A, Dos Santos Pinto RM, Smith N, Trubitsyna M, Gaunt E, Digard P, Michlewski G. TRIM25 inhibits influenza A virus infection, destabilizes viral mRNA, but is redundant for activating the RIG-I pathway. Nucleic Acids Res. 2022 Jul 8;50(12):7097-71 14. doi: 10.1093/nar/gkac512. PMID: 35736141 ; PMCID: PMC9262604. Kozaki T, Takahama M, Misawa T, Matsuura Y, Akira S, Saitoh T. Role of zinc- finger anti-viral protein in host defense against Sindbis virus. Int Immunol. 2015 Jul;27(7):357-64. doi: 10.1093/intimm/dxv010. Epub 2015 Mar 10. PMID: 25758257; PMCID: PMC4565983. Xie et al., Novel Epigenetic Techniques Provided by the CRISPR/Cas9 System. 2018. Stem Cells Int.; 2018: 7834175.

Hain Marco Y. et al., Nature Biotechnology, 2021 , vol. 40(3), pp. 391 -401

Adamson B. et al., bioRxiv, 2018, pp. 1 -14

Piccioni F. et al., Current Protocols in Molecular Biology, 2018, vol. 121 (1 ) OhAinle et al. (2018) A virus-packageable CRISPR screen identifies host factors mediating interferon inhibition of HIV eLife 7:e39823.

Qin JY, Zhang L, Clift KL, Hulur I, Xiang AP, Ren BZ, Lahn BT. Systematic comparison of constitutive promoters and the doxycycline-inducible promoter. PLoS One. 2010 May 12;5(5):e10611 . doi: 10.1371/journal.pone.0010611 . PMID: 20485554; PMCID: PMC2868906.

Claims

1 . A method of identifying a lentiviral modulation factor in the genome of a host cell which results in the production of replication-incompetent lentiviral particles at altered yield or infectivity, comprising: a) a genomic perturbation and screening of host cells for modulated expression of one or more genes within a host cell’s genome screen, thereby obtaining a pooled repertoire of perturbed host cells to cover perturbation of a diversity of genes; b) utilizing the repertoire of perturbed host cells to produce a lentiviral pool of replication-incompetent lentiviral particles in a perturbed host cell pool culture, where host cells bearing perturbations of lentiviral modulation factors produce lentiviruses with altered yield or infectivity; c) infecting a target cell line with said lentiviral pool and culturing the infected target cells in a target cell pool culture; and d) determining in the target cell pool culture the production of lentiviruses at altered yield or infectivity against the target cells, and identifying by simultaneous perturbation analysis which is a lentiviral modulation factor that results in the production of lentiviruses at altered yield or infectivity against the target cells in a host cell that comprises a respective perturbation for the lentiviral modulation factor, wherein the lentiviruses produced by the perturbed host cells comprise a respective nucleic acid barcode, and the lentiviral modulation factor is identified by determining the relative amounts of the barcodes in the target cell pool culture.

2. The method of claim 1 , wherein said genomic perturbation screen comprises an RNA-guided programmable nuclease (RPN)-mediated perturbation screen using a lentivirus harboring a library of single guide RNAs (sgRNAs), which are designed to target the diversity of genes in the host cell genome, wherein the repertoire of perturbed host cells is created using host cells containing the RPN, and wherein the lentiviral modulation factor is identified by determining a relative alteration of the respective sgRNA or its coding sequence, which is used as the barcode, compared to other sgRNAs in the target cell pool culture, preferably wherein the RPN mediated perturbation screen is a CRISPR-Cas9 screen, preferably a CRISPR knockout, CRISPR activation, or CRISPR interference screen.

3. The method of claim 1 or 2, wherein the lentiviral modulation factor is identified by determining the relative amounts of the barcodes in the target cell pool culture, preferably by PCR amplification and next generation sequencing, preferably wherein a relative increase of a barcode is indicative of the respective lentiviral modulation factor that has induced a higher yield or infectivity of lentivirus produced by the respective perturbed host cell, preferably wherein the host cell bearing the perturbation of the identified lentiviral modulation factor produces lentiviruses with at least 1.2-fold increased yield or infectivity compared to the host cell without such perturbation.

4. The method of any one of claims 1 to 3, which comprises at least two cycles of lentiviral infection, wherein a) in a first cycle, said repertoire of perturbed host cells is generated by infection with replication-incompetent lentiviral particles in pooled culture, thereby producing a pool of host cells harboring lentiviral donor templates; b) said pool of host cells harboring lentiviral donor templates produced by the first cycle is activated, thereby obtaining said lentiviral pool of activated lentivirus particles; and c) in a further cycle, the target cells are infected with the lentiviral pool of activated lentivirus particles in the target cell pool culture.

5. The method of claim 4, wherein said repertoire of perturbed host cells harboring lentiviral donor templates is bearing a variety of perturbations introduced by an RPN and a pooled library of sgRNAs, which are designed to target the diversity of genes in the host cell genome.

6. The method of claim 4 or 5, wherein the replication-incompetent lentiviral particles are assembled by utilizing a lentiviral donor template harboring a wild-type intact 3’ long terminal repeat (LTR) to exchange for a mutated or truncated 3’ LTR as previously used in replication-incompetent lentiviral particles, preferably wherein for said activation of the lentiviral donor templates, gag, pol, env, rev and tat are delivered to the pool of host cells harboring lentiviral donor templates, preferably wherein said gag, pol, env, rev and tat are delivered on two or more separate plasmids.

7. The method of claim 6, wherein the lentiviral donor template comprises a viral donor sequence which harbors a weak heterologous promoter.

8. The method of any one of claims 1 to 7, wherein the lentiviral pool is harvested from the perturbed host cell pool culture and transferred to the target cell pool culture.

9. The method of any one of claims 1 to 8, wherein the repertoire of perturbed host cells is of packaging host cell lines suitable for producing lentiviral particles for delivery of a transgene to a target cell, preferably wherein the host cells are human cells such as selected from the group consisting of HEK cells, such as HEK293T 293FT, Lenti-X, or 293SF-3F6

10. The method of any one of claims 1 to 9 wherein the target cells are human cells, preferably primary cells or stem cells, preferably selected from the group consisting of T cells, B cells, Macrophages, PBMCs and iPS cells.

1 1. The method of any one of claims 1 to 10, wherein the diversity of genes comprises at least 10 different genes.

12. The method of any one of claims 1 to 1 1 , wherein a library of perturbed host cells is obtained from the perturbation screen, wherein the library covers perturbation of a diversity of at least 10 different genes.

13. The method of claim 12, wherein the host cells are transduced with replication-incompetent lentiviral delivery particles.

14. A library comprising a repertoire of perturbed host cells which covers perturbation of a diversity of at least 10 different genes, wherein the host cells are transduced with replication-incompetent lentiviral delivery particles.

15. Use of the library of claim 14, for identifying a lentiviral modulation factor which upon its perturbation in a producer host cell results in altered yield production of replication-incompetent lentiviral delivery particles.

16. A method of engineering a producer host cell line producing lentiviral delivery particles at high titer or high infectivity, comprising: a) identifying a lentiviral modulation factor in the genome of a host cell according to the method of any one of claims 1 to 13, wherein the respective perturbation of the host cell for the identified lentiviral modulation factor has resulted in the production of lentiviruses at higher yield or infectivity against the target cells; and b) engineering a producer host cell line comprising said perturbation of the identified lentiviral modulation factor.

17. The method of claim 16, wherein the producer host cell line produces lentiviruses with at least 1 .2-fold increased yield or infectivity compared to the host cell without said perturbation of the identified lentiviral modulation factor.

18. A method of producing lentiviral delivery particles in a producer host cell line produced according to claim 16 or 17, preferably wherein said lentiviral delivery particles are replication-incompetent.

19. The method of claim 18, wherein said lentiviral delivery particles are packaged with a transgene for delivery to a target cell genome, preferably wherein said transgene encodes a chimeric antigen receptor (CAR) for T-cell immunotherapy, a T cell receptor (TCR) for T-cell immunotherapy, or a transgene for gene replacement therapy.

20. A producer host cell line obtainable by a method of any of claims 16 or 17, which comprises a knockout of the identified lentiviral modulation factor, wherein the identified lentiviral modulation factor is a gene selected from the group consisting of ZC3HAV1 , ADGRG1 , MEN1 , EPHB4, BARHL1 , C1 1 orf71 , TNFRSF6B, OR10G9, C17orf58, CDCA7, TRIM25, C1 1 orf68, TMEM125, FUT2, RCE1 , ZNF398, TLDC2, FCAMR, TNFRSF14, CYBC1 , PDXK, CASS4, MMP23B, HRG, GSN, TP53INP2, WDR81 , RBM4, NFKBIB, LYNX1 -SLURP2, SLC7A2, SYNM, DEPTOR, SIGIRR, FOSL1 , FGF4, GDF5, PLAAT3, TNPO2, HLA-DPA1 , MAGEB1 , PNMA6E, IRF2BP2 and TRPT1.

21. The producer host cell line of claim 20, which originates from a human cell, preferably a human cell that is used as a target cell recited in any one of claims 1 to 13.