CRISPR interference to model the Koolen-de Vries Syndrome in iNeurons derived form iPSCs

(1)

0

Bachelor Thesis

Tessa van der Heijden

June 2016

CRISPR Interference to model the Koolen-de

Vries Syndrome in iNeurons derived from iPSCs

(2)

1

CRISPR Interference to model the

Koolen-de Vries Syndrome in iNeurons Koolen-derived

from iPSCs

Student

Tessa van der Heijden ghm.vanderheijden@student.avans.nl

Student number 2064429

Supervisors

Dr. Nael Nadif Kasri n.nadif@donders.ru.nl

Katrin Linda katrin.linda1@radboudumc.nl

Teacher

Dr. Kees Rodenburg cw.rodenburg@avans.nl

Internship

Department of Cognitive Neuroscience Radboudumc

6500 HB Nijmegen

Donders Institute for Brain, Cognition and Behavior Centre for Neuroscience

6526 AJ Nijmegen

Education

Biology and Medical Laboratory Research Major Forensic Laboratory Research

Academy for the Technology of Health and Environment Avans University of Applied Sciences

Lovensdijkstraat 61-63 4818 AJ Breda

Date

Deadline Thesis: 13th_{of June 2016}

Presentation: 27th_{of June 2016}

(3)

2

Preface

This thesis is the result of my graduation research project at the Radboud University Medical Center in Nijmegen, for my Bachelor program Biology and Medical Laboratory Research, affiliated to the academy for the technology of health and environment at Avans University of Applied Sciences, Breda. This project is conducted from September 2015 to June 2016.

I would like to thank the principal investigator Dr. Nadif Kasri for offering me this research opportunity and for the supervision together with my supervisor Katrin Linda. I would also like to thank Dr. Kees Rodenburg for the supervision from school and my colleagues and fellow students for the great support and collaboration during my graduation internship.

(4)

3

Abstract

The Koolen-de Vries Syndrome (KdVS) is a neurodevelopmental disorder, also called the 17q21.31 microdeletion syndrome, which is characterized by intellectual disability, developmental delay, epilepsy, hypotonia and distinct facial features.6_{The main causative gene for this disease is the KANSL1}

gene, which is part of the non-specific lethal (NSL), is involved in the acetylation of histone H4 lysine 16 (H4K16), an epigenetic process that is mainly important during development and mitosis.3,7_{In order}

to study the pathogenesis of the KdVS, we used a cell line of induced pluripotent stem cells (iPSCs) wherein the KANSL1 gene was supposed to be knocked down by the sequence specific CRISPR interference (CRISPRi) technology. This technology is based on the co-expression of catalytically deficient endonuclease dead Cas9 protein (dCas9) and a sequence specific single guide RNA (sgRNA) that together form a complex, enabling sequence specific gene silencing of any gene complementary to the sgRNA.1_{Transformation of the KANSL1 deficient cells into induced neurons (iNeurons) would}

allow us to investigate the pathogenesis of the disease during development. In order to integrate the dCas9 gene in iPSCs, we used two different constructs with an EF1alpha promoter and a doxycycline inducible promoter. However, lentiviral delivery of these constructs did not result in dCas9 expression. Now we have a CRISPRi WTC cell line of iPSCs (obtained from Mandegar et al. 2016) that enables dCas9 expression, which was proven by means of immunostainings. This allows us to test the sgRNAs for performing CRISPRi in iPSCs.

(5)

4

Samenvatting

Het Koolen-de Vries Syndroom (KdVS) is een neuro-ontwikkelingsziekte, ook wel bekend als het 17q21.31 microdeletie syndroom, die wordt gekenmerkt door verstandelijke beperking, ontwikkelingsachterstanden, epilepsie, hypotonie en afwijkende gezichtskenmerken. Het belangrijkste gen dat verantwoordelijk is voor de ziekte is het KANSL1 gen dat deel uitmaakt van het non-specific lethal (NSL) complex en is betrokken bij de acetylatie in histoon H4 lysine 16 (H4K16). Dit is een belangrijk epigenetisch proces, vooral tijdens de ontwikkelingsfase en mitose van cellen.3,6_{Het doel}

van het project was om de pathogenese van het KdVS te kunnen bestuderen door middel van het maken van een cellijn van induced pluripotent stem cells (iPSC’s). Daarin wordt het KANSL1 uitgeschakeld aan de hand van CRISPR interference (CRISPRi). Dit systeem is gebaseerd op de co-expressie van katalytisch deficiënt endonuclease dead Cas9 (dCas9) wat een complex vormt met het target specifieke single guide RNA (sgRNA) en zorgt voor het inactiveren van het target gen.1_{Het idee}

is om deze KANSL1 deficiënte cellen te transformeren naar neuronen wat het analyseren van de pathogenese tijdens de ontwikkeling van neuronen mogelijk maakt. De iPSC’s werden getransduceerd met twee verschillende constructen om het dCas9 gen in het genoom van de cellen te integreren. Beide constructen bevatten een andere promotor, namelijk een EF1alfa promotor en een doxycycline induceerbare promotor. Echter heeft de lentivirale transductie van het dCas9-virus niet geleid tot de expressie van het dCas9 eiwit. Inmiddels maken we gebruik van een CRISPRi WTC cellijn van iPSC’s (afkomstig van Mandegar e.a., 2016) waarin het dCas9 eiwit wel tot expressie komt. Dit is bewezen middels immunocytochemie. Hierdoor kan het project voortgezet worden door verschillende sgRNA’s te testen om CRISPRi in iPSC’s te introduceren zodat het KANSL1 gen uitgeknokt wordt.

(6)

5

1. Introduction

In the department Cognitive Neuroscience at the University Medical Center Nijmegen (Radboudumc), there has been done research on cellular and molecular neurophysiology on several neurodevelopmental disorders. Cellular neurophysiological research has been done by means of electrophysiological recordings and micro-electrode arrays (MEAs) for analyzing cells at the single cell and network level. Furthermore, a nice in-vitro model of iNeurons derived from iPSCs is used to model e.g. the Kleefstra Syndrome for both molecular and electrophysiological investigations.

One of the other diseases that had been discovered in Nijmegen a few years ago is the Koolen-de Vries Syndrome, a neurodevelopmental disorder caused by haploinsufficiency of the KANSL1 gene. The role of KANSL1 in the brain is not yet well known, but seems to play an important role during development, which is not easy to study. Therefore the aim of this project is to differentiate KANSL1 deficient iPSCs into iNeurons, which enables us to investigate the pathogenesis of the disease during development of neurons. In order to generate KANSL1 knockdown in iPSCs, we perform CRISPRi for targeted gene silencing. After all, the use of a doxycycline inducible CRISPRi system allows the analysis on different timeframes later or earlier during development. We hypothesize that early KANSL1 knockdown in developing neurons will either lead to massive cell death or a low growth rate and KANSL1 knockdown in general will result in the outcome of autophagy.

The progress of introducing CRISPRi in iPSCs is described in this report, starting with the theoretical background of the KdVS and the methods in chapter 2. Then, the materials/methods and results are included in chapter 3 and 4, respectively. In chapter 5, the results are discussed, followed by the main conclusion in chapter 6. Based on the experiments performed, there are some future directions given in chapter 7. References and the supplementary data are found in chapters 8 and 9. For previous results and additional information, see the project plan and mid-term report.

(8)

7

2. Theoretical Background

In our body, we have many different types of cells. Despite the fact that the cells all contain the same genotype inside the nucleus, the cells show many differences in phenotype.8_{These differences are}

caused by epigenetic processes within the cell, which alter the gene activity without changing the DNA sequence itself.9_{DNA methylation and acetylation (figure 1) are examples of post translational}

modifications of histones that mainly occur during development.10,11_{During histone acetylation, an}

acetyl group is added to the lysine of the histone, which is performed by histone acetyltransferases (HATs). The addition of an acetyl group neutralizes the positive charge in the histone, leading to less affinity between the histone and DNA, therefore the chromatin structure becomes less condensed. The lower density of DNA structures increases the accessibility for transcription factors to specific genes, allowing genes to be transcribed. In contrast to histone acetylation, histone methylation increases the density of the chromatin structure so genes will be switched off again due to the inaccessibility for transcription factors and RNA polymerases.9,10

Figure 1, DNA (black) is wrapped around histones (purple). This structure can be altered due to epigenetic processes in the cells e.g. histone methylation (left) and histone acetylation (right). Due to the open structure when the DNA is acetylated, transcription factors have access to the DNA so genes can be described. However, when the DNA is methylated, DNA is inaccessible for transcription factors. Methyl (me) and acetyl (ac) groups are added to the DNA to alter the chromatin structure. Acetylation is performed by histone acetyltransferases (HATs) and the deacetylation by histone deacetylases (HDACs).12

Besides HATs and histone deacetylases (HDACs), there are many other epigenetic regulators involved in epigenetic alterations of chromatin inside cells. An example is the nonspecific lethal (NSL) complex, which is involved in global the regulation of transcription. Histone acetyltransferase 8 (KAT8), also called males absent on the first (MOF), is part of the NSL complex and is for a large fraction responsible for the acetylation of histone H4 lysine 16 (H4K16).7_{KAT8 regulatory NSL complex subunit 1 (KANSL1)}

is shown to be a scaffold protein in the NSL complex. Recently it was shown that mutations in the KANSL1 gene can lead to the complete phenotype of the Koolen-de Vries Syndrome (KdVS), also known as the 17q21.31 microdeletion syndrome. The

neurodevelopmental disorder is characterized by distinctive facial features (figure 2), mild-to moderate intellectual disability (ID), hypotinia, epilepsy and an amiable personality.6_{More than 50% of}

the patients suffers from structural defects of the brain, heart and the genitourinary system.13_{The syndrome can be caused by a}

microdeletion of varying sizes between 160 and 650 kb13,14_{in the}

q21.31 locus on chromosome 17 including at least the genes MAPT,

Figure 2, Facial characteristics of Koolen-de Vries patients.

(9)

8 STH and KANSL1. However, haploinsufficiency of the KANSL1 gene is sufficient to cause the complete phenotype of the KdVS.6,14,15

Not only KANSL1 is part of the NSL complex (figure 3), also MOF/MYST1/KAT8, KANSL1, KANSL1, MCRS1 and PHF20 are included in the multiprotein complex. MYST is a family of histone acetyltransferases and KAT is one type of HAT in this protein family (responsible for lysine acetyltransferase). The KANSL1 subunit within the multiprotein NSL complex consist of 1105 amino acid residues and is possible unstructured. As shown in figure 3, the PEHE domain of KANSL1 interacts with MOF/KAT8 and the KANSL1 subunit

interacts with three other proteins namely WDR5, MCRS1 and PHF20. The indirect connection between WDR5 and MOF/KAT8 is created by KANSL1 via conserved, short binding motifs (WIN in the figure 3). Four Zn-coordinating motifs are involved in the binding between KANSL2 and WDR5. According to figure 3, it can be concluded that KANSL1 acts as a scaffold protein in the NSL complex. Because of this, the loss of function of KANSL1 leads to improper bindings within the NSL complex, which results in less functional H4K16 acetylation. The function of the other sub proteins is not yet well known.

Also the function of KANSL1 within the brain is not really clear. However, it was recently shown that H4K16 acetylation plays an important role in the outcome of autophagy.16_{Since KANSL1 is part of the}

NSL complex7_{, which regulates the acetylation of H4K16, we hypothesize that loss of function of}

KANSL1 could affect autophagy in neurons. Autophagy is a lysosome-mediated process that results in the degradation of non-essential or damaged cell compounds and short-lived proteins.17,18_{This process}

is involved in starvation, development, cell death, tumor suppression and protein and organelle clearance.19_{Three main types of autophagy are known: micro-autophagy, chaperone-mediated}

autophagy and macro-autophagy. Both chaperone-mediated and micro-autophagy make directly use of lysosomes to degrade only small parts of the cytosol.20_{However, macro-autophagy is involved in}

the degradation of larger parts of the cytosol, whole organelles, intracellular pathogens and misfolded proteins using a double membraned vesicle called an autophagosome.18_{The pathway of autophagy}

(figure 4) can be divided in several phases: induction, nucleation/vesicle formation, fusion with lysosome and degradation. First, autophagy is induced by the activity of class I phospoinositide 3-kinase (PI3-K), which is usually inhibited by protein 3-kinase target of rapamycin (mTOR), for instance when cells were in nutrient supply is sufficient.21_{Also ULK1 is involved in the autophagy initiation step.}2

A lot of autophagy associated genes (ATGs) are involved in the autophagy pathway of whereas ATG8, ATG4, ATG7 and ATG3 are most important during autophagosome formation. Light chain protein 3 (LC3), an important marker for autophagy, is cleaved by ATG4 into LC3-I, which is present in the cellular cytosol. Then, ATG7 is responsible for the activation of LC3-II and the transfer of LC3-II towards ATG3, which attaches lipids and this results in the membrane bound form LC3-II. The lipidation of LC3 is very important during the autophagosome formation, but is also useful for the detection of autophagy since LC3-II is only present when autophagy is induced.20_{After the expansion and completion of the}

autophagosome formation, the autophagosome fuses with a lysosome containing hydrolytic enzymes

Figure 3, Composition of the human NSL complex with on the right side of the picture the multiprotein complex consisting of KANSL1, MOF/HAT, WDR5, PHF20, MCRS1, KANSL2 and KANSL3. KANSL1 is a scaffold protein because it interacts with MOF via its PEHE domain. Short binding motifs (WIN) generates the binding between WDR5 and KANSL1. Due to Zn-coordinating motifs, KANSL2 and WDR5 are connected.

(10)

9 resulting in an autolysosome (or autophagolysosome) that degrades the cellular components within the autolysosome.17_{The exact function of autophagy in neuronal cells remains unclear. Despite the}

fact that ATGs are highly expressed in rodent brains, the autophagy markers LC3-II and the number of autophagosomes are low.17_{Based on research of Collingridge et al. it can be concluded that}

degradation via endocytosis affects synaptic plasticity, in particularly in AMPA receptors. The role of autophagy in developmental biology is described in the project plan.

In order to get more insight into the role of H4K16 acetylation and the outcome of autophagy in developing neurons, we use induced pluripotent stem cells (iPSCs) derived from human fibroblasts. IPSCs can be differentiated into iNeurons, which enables to do research on the developing neurons that lack the KANSL1 gene due to CRISPRi. The use of iPSCs gives us a nice model to compare ‘Koolen-de Vries Syndrome cells’ with healthy ‘Koolen-developing neurons.

Figure 4, Autophagy pathway. Cellular autophagy is induced by the mTOR pathway. During the formation of the autophagosome, LC3-I becomes lapidated which results in the formation of LC3-II. The autophagosome fuses with lysosomes, which results in an autolysosome, wherein hydrolytic enzymes enter the cellular debris to degrade. From the figure can be deduced that many ATG protein are involved mainly during the nucleation phase and expansion of the autophagy pathway.2,3

(11)

10 IPSCs and embryonic stem cells (ESCs) have some characteristics in common of whereas the most important is that they maintain pluripotency, which means that cells have the capacity to differentiate into cells of all three germ layers namely endoderm, ectoderm and mesoderm. In addition, iPSCs and ESCs grow indefinitely and the Nanog gene is expressed in both iPSCs and ESCs, which is important for maintaining the pluripotent state of undifferentiated cells.22,23_{Because of ethical reasons, it is not}

easy to use ESCs for research and to treat several diseases. Besides, when modified cells are injected in patients to treat diseases, there is a big chance for rejection because of differences in genotype. Therefore, the use of iPSCs is a great solution since iPSCs can be derived from patient’s fibroblasts in order to make certain in-vitro disease models and transplantation therapies. Usually, fibroblasts are transformed into iPSCs by lentiviral delivery of transcription factors octamer3/4 (Oct3/4), SRY box-containing gene 2 (Sox2), kruppel-like factor 4 (Klf4) and c-Myc (figure 5). The expression of Oct4 is required for the initial specification of pluripotent stem cells.24_{Sox2 is able to form heterodimers with Oct4 transcription factors, because the}

binding site of Sox2 is adjacent to Oct4.25

Many protocols for differentiating iPSCs into iNeurons are possible to perform. However, most of the existing protocols are difficult to perform, lead to a low yield of differentiated neurons and the results are not reproducible. Using these protocols, the synapse formation is limited and the protocols only generates small amounts of functional neurons. A recently published paper describes26_{the protocol,}

which enables the transformation of iPSCs into neurons of nearly 100% within 2 weeks. This protocol is also used in our lab, with some adjustments (figure 6). The transformation of the iPSCs only requires the single transcription factor neurogenin-2 (Ngn-2), delivered via lentiviral infection, resulting in great synapse formation in iNeurons. The expression of transcription factor Ngn-2 is doxycycline inducible due to the doxycycline inducible promoter. The Ngn-2 construct is provided with a puromycin resistance gene, which allows the selection of Ngn-2 positive cells after the lentiviral delivery. In order to support the development of the iNeurons, primary astrocytes (glia cells) are added. Only a few days after the doxycycline inducible promoter is induced with doxycycline, the iPSCs already get a neuron-like shape and start to form axons and dendrites.26

Figure 5, Somatic cells, patient specific, can be reprogrammed back into their pluripotent state using at least transcription factors Oct4, Sox, Klf4, c-Myc. The patient specific iPSCs can be differentiated into all cell types to study disease models in vitro or for gene therapy.

(12)

11 In order to study the role of particular disease causing genes, such as KANSL1, RNAi has been the most-commonly used method to study gene functions, which is a sequence specific, RNA-mediated interference system that uses small interfering RNAs or short hairpin RNAs. The disadvantage of RNAi is that the results can be inefficient or even nonspecific.27_{In addition, zinc-finger proteins or TALEs}

were also used for both gene regulation and DNA targeting. However, this system is technically very challenging, because each DNA-binding protein is different for every different DNA-binding site, the proteins always need to be designed individually. Still, the protein binding is very robust, so the combination of the specificity in RNAi and the robustness of DNA-binding proteins could be a great technology.28

Recently, a new technique in molecular biology is discovered that is called Clustered Regularly Interspaced Palindromic Repeats Cas9 (CRISPR/Cas9). Originally, the CRISPR/Cas9 system is part of the adaptive immune system in 90% of archaea and 40% of bacteria and is a protection mechanism in those organisms against foreign DNA fragments.29_{The loci of CRISPR sequences consist of repeats that}

are separated by Cas genes. Research has already proven that these CRISPR sequences show homology to the genome of several bacteriophages, because a difference is observed in the CRISPR1 locus between Streptococcus thermophilus strains that were either resistant or non-resistant for a several bacteriophage. This difference means that extra spacers were included in the resistant strains. The sequences of these inserted spacers are similar to the genome of the bacteriophage where the strain was infected with. The Cas7 protein seems to be a very important gene in the CRISPR immune system as the Cas7 knock out strains were not able to change their immunity against provided bacteriophages. This could mean that especially Cas7 is important in the synthesis or insertion of the new spacers when the strains are exposed to the bacteriophages.29

Figure 6, Time schedule of the iNeuron differentiation derived from iPSCs. At day zero, the cells were plated on geltrex coated plates. After two days, the E8 medium containing doxycycline to activate the TetO promoter, is changed by DMEM medium provided with doxycycline, NT3, BDNF, NEAA and N2 (neuronal supplements, except doxycycline). On the second day, glia is added to prevent clustering of cells towards one place. On the fourth day, AraC is added to kill dividing cells (so not the neurons) to select only the post-mitotic neurons. In addition, on the fourth day, the DMEM medium is changed to neurobasal medium to support the development of neurons. After approximately 15 days, FBS is added to support the glia cells. After three weeks, the cells are ready to be analyzed.

(13)

12 The CRISPR/Cas9 technology used in molecular biology, is known as an RNA-guided DNA recognition platform, which is able to either alter or regulate the DNA transcription in a robust, scalable and precise manner.4_{The most commonly CRISPR technique for genome-engineering is Type II CRISPR system}

(figure 7a). This mechanism requires a few things, namely: (1) CRISPR-associated protein 9 (Cas9) that acts as a endonuclease, (2) an RNA complex that exists of CRISPR RNA (crRNA) and trans-acting RNA (tracrRNA) and (3) a short DNA sequence that is adjacent to the RNA complex binding site, which is called the protospacer-adjacent motif (PAM) sequence.1,4_{The Cas9 protein and the RNA-complex}

binding leads to the generation of a double stranded break in the target gene, however it is also possible to use an engineered single guide RNA (sgRNA) to create this double strand break.4_{Finally the}

target gene is edited by DNA repair of the homology-mediated repair machinery or by non-homologous end joining (NHEJ).30_{The loss of function in the target genes is then causes by in-frame}

insertions or deletions (INDELs).30_{In order to regulate the gene transcription instead of gene editing}

(figure 7b), CRISPRi is a nice CRISPR/Cas9 technology that makes use of the catalytically deficient endonuclease called dCas9. Due to two point mutations in nuclease domains RuvC1 and HNH, dCas9 lacks its endonuclease activity.1_{In figure 7, the differences between the functioning of Cas9 protein}

and dCas9 are presented.

Figure 7, the difference between CRISPR-Cas9 and CRISPR interference. a) Gene editing using nuclease Cas9, this is the main CRISPR-Cas9 technology. b) Gene regulation with CRISPR interference using nuclease deficient Cas9 because of two point mutations (RuvC1 and HNH) in the figure presented as two purple circles. Because the presence of the dCas9 protein, DNA polymerase is not able to either initiate or elongate the DNA transcription, depending on the target site. The upper part is called the nuclease (NUC) lobe and the other half is called the recognition (REC) lobe because of the sgRNA binding site. In both figures a and b, the NGG sequence is coding for the PAM sequence which always lies directly behind the target sequence.4

(14)

13 For CRISPRi, the binding of both a sequence-specific sgRNA and the dCas9 protein, which results in the dCas9-sgRNA complex, can lead to gene silencing up to 99.9% for the gene complementary to the sgRNA.1_{It depends on the target region of the dCas9-sgRNA complex whether the gene transcription}

is either blocked during the initiation step or the elongation by DNA polymerase II (figure 8).

CRISPRi systems including a doxycyline incucible promoter are a great opportunity to use CRISPRi for reversible and scalable gene silencing. In this project we used two constructs (construct 2 and 4, supplementary data 2 and 4) both containing a Tet-On 3G Tetracycline inducible gene expression system (TRE3G) promoter. This enables regulated and precisely controlling of gene expression and the systems are reproducible an reversible. An

example of the TRE3G inducible system is presented in figure 9, here is shown that the Tet-On 3G transactivator only binds to the TRE pomoter and performs DNA transcription in presence of doxycycline. This Tet-On 3G tranactivator is also called the reverse tetracycline controlled transactivator (rtTA) en therefore, Tet-On systems are dependent on rtTA expression, which codes for the transcription facor for the doxycycline inducible promoter in the gene of interest. Leaky expression of the gene of interest driven by the doxycyline inducible promoter is reduced significantly compared to older Tet-On systems due to this rtTA system.31_In

the next chaptor is described how the technologies explained here are applied.

Figure 8, overview of CRISPRi. The initiation of DNA transcription is blocked when dCas9 binds to the TSS or promoter in the template or non-template strand, because the transcription factor is unable to bind the TSS (in figure called TFBS: transcription factor binding site) or RNA polymerase is unable to bind to the promoter. The dCas9 binding in the 5’UTR region or protein coding region in the non-template strand blocks the elongation by RNA polymerase.1

Figure 9, Tet-On 3G transactivator is transcribed by an rtTA construct. This transactivator is only active when interacting with doxycycline, which results in a conformational change and therefore the transactivator is able to bind to the TRE3G promoter in the target gene.5

(15)

14

3. Materials and Methods

Maintaining iPSCs

IPSCs were cultured in Essential 8 (E8) basal medium (Gibco, cat. no A1517001) supplemented with 50X E8 supplement and 1% penicillin streptomycin (pen strep) (Sigma cat. no. P4333) and incubated at 37˚C/5%CO2. Plates were coated with 0.5 mg/ml vitronectin (VTN-N, Gibco, cat. no. A14700) in

Dulbecco’s PBS (DPBS, Gibco, cat. no 14190-094) and incubated for at least 1 hour at room temperature. The E8 medium was changed daily, except the day after splitting. For splitting the cells, cells were washed once with 2 ml DPBS (in case of 6-well plate) before EDTA (Invitrogen, 0.5 M, pH 8.0, cat. no. 1556487) was added and incubated for 3-8 minutes, depending on the cell density of the cells. After incubation, EDTA was changed by E8 medium to re suspend the cells gently so cells stay in colonies. The appropriate volume of cell suspension was added to the VTN-N coated plate and cells were distributed equally over the plate. Neomycin and puromycin resistant cells were kept in culture in E8 medium supplemented with 50 µg/ml puromycin (Sigma cat.no. P8833) and 0.5 µg/ml G418 (Gibco, cat.no. 10131-035).

Making lentiviral constructs using calcium phosphate transfection

For the production of lentiviruses, HEK293 cells were used, which were cultured in DMEM medium (Dulbecco’s Modified Eagle’s Medium – high glucose, Sigma, cat. no D0819) supplemented with 10% FBS (Fetal Bovine Serum, Sigma cat. no. F7524), 1% penicillin/Streptomycin (pen/strep, Sigma cat. no. P4333) and 1% sodium pyruvate (100mM, Sigma cat. no. S8636) and were incubated at 37˚C/5%CO2.

HEK293T cells were split when the confluency was higher than 80% by first washing two times with pre warmed PBS, followed by trypsinization (trypsin, 0.25%) and were then transferred to new plates containing fresh medium. For the formation of lentiviruses, packaging and envelope vectors PMDG.2 (1.2 µg, Addgene, plasmid no. 12259) and PSPAX (3.2 µg, Addgene, plasmid no. 12260) were added together in Eppendorf tubes to either construct 1 or construct 2 (10 µg) for dCas9 virus, or with construct 3 including the sgRNA (supplementary data 1, 2 and 3). Then the volume was adjusted to 250 µl with Tris-EDTA (1mM Tris, 0.1mM EDTA) and also 250 µl calcium chloride (500 mM CaCl2, 1 mM

Tris, 0.1 mM EDTA, pH between 7 and 8) was added and vortexed. 500µl HEPES buffered saline (HBS, 50 mM HEPES-NaOH, pH 7.3; 280 mM NaCl, 1.5 mM NaPO4) was pipetted in a snap-cap. Next, the calcium-DNA solution was added drop-wise to the HBS while vortexing (solutions for calcium phosphate transfection: Sigma, cat. no. CAPHOS). After the precipitate was formed, the solution was added to the cells in a 10cm dish. The medium was changed by fresh medium to remove the precipitate particles after 5-8 hours after incubation. Then, 48 hours after transfection, the medium containing viral particles was collected and centrifuged (Hettich Centrifugen, Rotina 380R) at 2000rpm for 5 min at 4˚C. The supernatant was filtered through a 0.45-umsyringe filter and stored in aliquots at -80˚C. For making concentrated virus, the filtered supernatant was transferred to an Ultra-Clear tube for an SW28 Rotor (Beckman Thin wall polyallomer tubes 326823). Tubes were centrifuged (Hettich Centrifugen, Rotina 380R) at 21000 rpm for 3 hours at 4˚C. Then the supernatant was removed and the pellet was re suspended in 100 µl Hanks’s balanced salt solution (Thermo Scientific, cat. no. 14185045).When the virus containing pellet was dissolved completely, this was added directly to the cells.

(16)

15

Infection with lentiviral constructs

Plate iPSCs as single cells

In order to plate iPSCs as single cells, plates were coated with diluted geltrex (1:100)(Gibco, cat. no. A1413201) in DMEM/F12 (Gibco Cat. no. 11320074) an hour before the cells were plated and incubated for at least 1 hour at 37˚C/5%CO2. After incubation at 37˚C, the plate was incubated at room

temperature for 1 hour before the E8 medium, consisting of 2 µM thiazovivin (Sigma cat. no. SML1045), was added with the appropriated volume. Accutase (accutase enzymes in DPBS containing 0.5 mM EDTA·4Na and 3 mg/L Phenol Red, Sigma cat. no. A6964) was added to loosen the cells and incubated for approximately 5 min at 37˚C/5%CO2. Cells were re suspended gently and washed in DMEM/F12 by

centrifuging 5 min at 1500 rpm to form a pellet. DMEM/F12 was changed by E8 medium provided with 2 µM thiazovivin when cells were re suspended to become single cells. Cells were counted under the microscope using 5 µl trypan blue and 5 µl cell suspension. The appropriate volume of cell suspension was added to the geltrex plate and incubated at 37˚C/5%CO2. In case of a doxycycline inducible

promoter in the construct, doxycycline (Sigma cat. no. D9891) was added to a final concentration of 4 µg/ml.

Lentiviral transduction

The day after plating single cells, lentiviral constructs were added in different volumes (constructs 1 and 2 for dCas9, construct 3 for sgRNAs). The spent medium was changed by fresh E8 medium provided with 8 µg/ml polybrene (Sigma cat.no. H9268) and 2 µM thiazovivin. Polybrene was also added to the virus to a final concentration of 8 µg/ml before the virus was added to the cells. Six hours after the infection, the virus containing medium was changed by fresh E8 medium. In case of a doxycycline inducible promoter in the construct, doxycycline was added to a final concentration of 4 µg/ml.

Antibiotic selection procedure

In order to make a stable cell line, the iPSCs were infected with both the rTTA construct containing a neomycin resistance gene and the Ngn-2 construct, which contains either a blasticidin or puromycin resistance gene. Selection with blasticidin was started 3 days after plating the cells with 6 µ/ml and stopped sixteen days after plating. The antibiotic selection for puromycin and G418 are described in table 1 and 2 respectively. After the selection period, neomycin and puromycin resistant cells were kept in culture in 50 µg/ml G418 and 0.5 µg/ml puromycin.

Day Final concentration G418 3 100 µg/ml

4-7 250 µg/ml

Day Final concentration puromycin

3 1 µg/ml

4 2 µg/ml

5 2 µg/ml

6 1 µg/ml

Table 1, Selection procedure for rtTA positive cells with G418

Table 2, Selection procedure for Ngn-2

(17)

16

iNeuron differentiation

Stable iPSC cell lines, containing the rTTA and Ngn-2 construct, were transformed into iNeurons by the addition of doxycycline. In order to generate the neurons, the stable cell line was plated as single cells on a plate coated with poly-L-ornithine (Sigma cat. no. MF001-118), which was diluted in MQ to a concentration of 50 µg/ml. The next day, the poly-L-ornithine was aspirated and the wells were washed twice with MQ. Laminin (Gibco cat. No. 23017) was diluted to a concentration of 10 µg/ml in cold DMEM/F12 and added to the poly-L-ornithine coated wells. The plate was incubated for at least two hours at 37˚C/5% CO2. The cells were plated as single cells according to the protocol in paragraph 3.3.

One day after splitting, the medium was changed with DMEM/F12 supplemented with N-2 (1:100, Sigma cat. No. 17502), NEAA (1:100, Sigma cat. no. M7145), NT3 (10 µg/ml, Promocell cat.no. C66212), BDNF (10 µg/ml, Promocell cat.no. C66425) and doxycycline (4 µg/ml). The medium was filtered before laminin was added to a concentration of 0.2 µg/ml. Two days after plating the cells as single cells, the medium was changed by neurobasal medium (always supplemented with B-27, glutamax and pen/strep) was provided with 10 µg/ml NT3 and BDNF, 2 µM AraC and 4 µg/ml doxycycline. Because cells were fixed approximately 5 days after they were plated, there was no need to add glia cells on the second day after plating.

Preparation and western blot

Making cell lysate

Cells were first washed twice with PBS before 100 µl lysis buffer was added. Lysis buffer was prepared by diluting complete mini (Roche cat. no. 11836153001) 7 times in RIPA buffer (10 mM Tris-HCl pH 8.0; 1 mM EDTA; 0.5 mM EGTA; 1% Triton X-100; 0.1% sodium doxycholate; 0.1% SDS; 140 mM NaCl). The cells were collected using a cell scraper and transferred to an Eppendorf tube. Then, the cells were incubated at 4˚C for 15 minutes in a rotating position and centrifuged 10 minutes at 15000 rpm at 4 ˚C. The supernatant was collected and snap frozen in liquid nitrogen before the samples were stored at -20˚C.

Determine protein concentration

Before the samples were analyzed for the presence of specific proteins, the protein concentration was measured to prevent exceedance of the maximum amount of 50 µg. The Pierce™ BCA Protein Assay Kit (Thermo Scientific cat. no. 23225) was used for making a dilution series with 2000, 1500, 1000, 750, 500, 250, 125, 25 and 0 µg/ml BSA protein diluted in lysis buffer. The working reagent was prepared by mixing 1 part of reagent A with 50 parts of reagent B. Each sample (10 times diluted) was mixed with 1 ml working reagent and directly incubated at 37˚C for 30 minutes. The absorbance was measured on 562 nm using a spectrophotometer.

Sample preparation for western blot

The samples for the western blot were commonly diluted to a concentration of 2.5 µg/µl (or less, depending on the lowest protein concentration measured) in a total volume of maximal 40 µl per sample in lysis buffer. One third of the total sample volume sample buffer was added to the samples. To prepare the sample buffer, DTT (1M/1ml) was diluted 10 times in 4x laemmli sample buffer (BioRad cat. no. 161-0747). After the sample buffer was added to the samples, samples were incubated at 95˚C for 10 minutes.

(18)

17

Western blot for KANSL1 detection

The prepared samples were loaded on a stain free gel (Mini-PROTEAN®_{TGX Stain-Free™ Protein Gels,}

BioRad, cat. no 456-8084) with a maximum of 50 µg per 40 ml per well. Empty wells were filled with 5 µl laemmli buffer and 5 µl marker was loaded on the gel. The gel was run for 20 minutes at 200V, then the gel was imaged in the ChemiDoc (Touch imaging system) for activating proteins which enables visualization on the blot. Proteins in the gel were transferred to a nitrocellulose membrane (Trans-Blot® _{Turbo™ Mini Nitrocellulose Transfer Packs, BioRad, cat. no. 1704158) using a blotter machine.}

After this, the proteins on the membrane were visualized to determine whether the proteins successfully were transferred from the gel to the membrane. The membrane was incubated in blocking buffer, consisting of 5% Blotting Grade Blocker (BioRad, cat. no. 1706404) in PBS-T 0.1% Tween (Merck Millipore, cat. no. 822184), for one hour at room temperature. After the blocking step, the, which was incubated overnight at 4˚C. After incubating the primary antibody, the membrane was washed three times for 5 minutes in PBS-T 0.1% before the blot was incubated with the secondary antibody in blocking buffer for 1 hour at room temperature. Then, again three washing steps were performed in PBS-T 0.1% and once in PBS. In the end, the membrane was transferred to the chemiluminescence tray and covered with ECL solution (SuperSignal™ West Pico Chemiluminescent Substrate, Thermo Scientific, cat. no. 34080) to visualize the proteins on the blot.

Primary antibodies for western blot and immunostaining: Anti CRISPR/Cas9 (mouse, Epigentec, cat. no. A-9000) and Anti KANSL1 (rabbit, Sigma, cat. no. HPA006874)

Immunofluorescence staining

Before the cells were fixed with 4% paraformaldehyde/4% sucrose (Sigma Aldrich, cat. no. MKBK8362V; 090M02112V) for 15 minutes at room temperature, the cells were washed once with ice cold PBS. After the cells were fixed they were washed 4 times for 5 minutes at room temperature with PBS. Then the cells were permeabilized with 0.2% triton(Merck Millipore, cat. no. 822184) in PBS for 10 minutes at room temperature followed by 3 wash steps of 5 minutes with PBS. The blocking buffer was added and incubated for at least 1 hour. The primary antibody mouse anti cas9 was diluted 500 times in blocking buffer, then added to the coverslips and incubated overnight. The next day, the coverslips were washed 3 times for 5 minutes with PBS before the secondary antibody was added. The secondary antibody used was goat anti mouse Alexa 568 and incubated 1 hour. Then the coverslips were washed 3 times for 5 minutes with PBS. The cells were stained with Höchst (Invitrogen, cat. no. 2249542) (1:10000 in PBS) 10 minutes at room temperature followed by a 5 minutes wash steps with PBS. Finally, the cells were mounted with DAKO (DAKO, cat. no. 10105463) and this was hardened for at least 1 hour before imaging.

(19)

18

4. Results

No KANSL1 knockdown observed in iPSCs transduced with construct 1

For gene silencing of KANSL1 using CRISPRi, we need a cell line that stably expresses and a sgRNA (inserted in construct 3, supplementary data) that is complementary to the target gene and therefore guides the dCas9 to the gene of interest. First the 29.3 control, a particular iPSC line, was transduced with a lentiviral construct (construct 1, supplementary data 1) containing a dCas9 sequence adjacent to an EF1alpha promoter and puromycin resistance gene. Then the puromycin selected cells were infected with different sgRNAs. For this, the untreated 29.3 control line was used as selection control, because both, the dCas9 vector and the vectors in which the sgRNAs (construct 3 supplementary data) were integrated contain the puromycin resistance gene. At the same time, the puromycin resistant cells were also transduced with the sgRNAs and incubated for at least 2 weeks to ensure total breakdown of the current KANSL1 proteins. The KANSL1 expression was detected by means of western blotting, this result is presented in figure 10. The samples in lane 1 (29.3) and 2 (dCas9) are both controls for KANSL1 (120 kD). KANSL1 expression is not affected in these lines, because they were not infected with the sgRNA-virus. The samples in lane 3, 4 and 5 (TSS, Prom1 and Prom2 respectively) contain bands with similar light intensities compared to the controls in lane 1 and 2. The western blot was performed twice and showed similar results, indicating that KANSL1 expression is not reduced in sgRNA treated cells.

Figure 10, Western blot for iPSCs (29.3 control) infected with construct 1 and three different sgRNAs targeting the transcription start site (TSS) and the promoter (Prom1 and Prom2) within the KANSL1 sequence. Cells were incubated 16 days with doxycycline. Samples 29.3 and dCas9 in the picture are untreated cells (29.3) and cells that were only infected with dCas9. The upper bands represent KANSL1 (120 kD), the weak lower band is KANSL1-like protein. Experiment performed twice.

(20)

19

No dCas9 expression observed in iPSCs and iNeurons infected with constructs 1 and 2

Reasons for the missing reduction in KANSL1 expression might be unspecific sgRNAs or lacking dCas9 expression. Perhaps, the EF1alpha promoter in construct 1 is not actively transcribed in iPSCs but rather in post mitotic neurons. In order to test dCas9 expression in either iPSCs or iNeurons we performed immunofluorescence on differentiated iPSCs using a specific antibody against Cas9 in cells transduced with construct 1. The immunostaining for dCas9 is presented in figure 11a is representative for all stained colonies in iPSCs and iNeurons transduced with dCas9 construct 1. According to the staining presented here, no clear dCas9 signal is detected. The observed background signal is different form signal that should be obtained from dCas9 immunostainings (see figure 11b, positive control).

Based on the western blot and immunostainings for dCas9, it can be concluded that neither dCas9 is expressed nor KANSL1 expression is reduced when construct 1 is used. Therefore, another construct is used that includes a doxycycline inducible promoter (construct 2, supplementary data). Because the transduced cell line already contains the Ngn-2 and rTTA construct, these cells are able to differentiate into iNeurons when the medium is provided with doxycycline. This allows us to determine whether dCas9 is expressed in iPSCs and iNeurons infected with construct 2 and if expression levels differ. Due to the neomycin resistance gene in both the rtTA construct and construct 2, there was no possibility for selecting cells that are positive for construct 2. Therefore we plated the cells transduced with construct 2 as single cells and picked colonies for immunostainings after doxycycline treatment to determine whether dCas9 positive colonies are present in the cell culture. We already did lentiviral delivery with construct 2 (see mid-term report) and since this did not succeeded we now used fresh made virus to prevent reduction in multiplicity of infections (MOIs) due to freeze-thaw cycles. Because this improves the delivery efficiency. Still the immunostaining for dCas9 in cells transduced with construct 2 gives a similar result as shown in figure 11a. Only very weak GFP expression is visible.

Figure 11, a) Staining for dCas9 in iPSCs (29.3 control) infected with the doxycycline inducible construct (construct 2) Doxycycline induction 72 hours. b) HEK293 cells transfected with Cas9 (data from product sheet anti CRISPR/Cas9 antibody, Epigentc cat. no. A9000)

Staining for dCas9 in CRISPRi cell line from San Francisco (CRISPRi WTC). This staining serves as a positive control for dCas9 expression. dCas9 is stained with GFP.

(21)

20

Protein expression of dCas9 is present in CRISPRi WTC cell line

A new CRISPRi cell line obtained from Gladstone Institue of Cardiovascular Disease, San Fransisco32

(Mandegar et al. 2016), was ordered as transfection with constructs 1 and 2 did not result in dCas9 expression in our cells. This CRISPRi cell line (from now called CRISPRi WTC) also contains a doxycycline inducible promoter as well as a mCherry fluorescent reporter (construct shown in supplementary data 4). After the cells were thawed and recovered, we cultured the CRISPRi WTC cells on two coverslips of which one well was treated with doxycycline to induce dCas9 expression. After 24 hours, mCherry expression was already visible in the doxycycline treated cells. The mCherry expression 72 hours after doxycycline induction is presented in figure 12a next to the non-induced cells in figure 12b. By comparing these signals, a clear difference was observed, indicating that the doxycycline inducible promoter in only active in the presence of doxycycline.

In order to ensure that dCas9 is expressed in the CRISPRi WTC cells and to test the antibody against dCas9, we performed immunocytochemistry. In figure 13 the dCas9 expression (GFP) is shown in a) non-induced and b) doxycycline treated iPSCs. A clear difference is visible between the two conditions, which means that the antibody seems to be specific for dCas9 and dCas9 is expressed in the cells where doxycycline is added.

Figure 12, a) Untreated CRISPRi WTC cells. b) CRISPRi WTC cells treated with doxycycline during 72 hour. The construct is stained with RFP. Nuclei are stained with DAPI.

Figure 13, a) Staining for dCas9 in the CRISPRi WTC cell line of cells that are non-induced and b) doxycycline treated cells during 72 hours. dCas9 is stained with GFP and nuclei with DAPI.

(22)

21

5. Discussion

The aim of this study was to generate a stable cell line enabling us to perform CRISPRi in order to knock down KANSL1. This was tried in HEK293 cells before (see mid-term report) using construct 2 (supplementary data 2). However, this did not succeed as the size of the construct seemed to be too big to be packaged efficiently into lentiviral particles. Therefore, we continued with another dCas9 vector (construct 1, supplementary data 1) with a smaller gene insert in the vector backbone. We hypothesized that this would increase the virus efficiency. According to the results, obtained from experiments performed with constructs 1 and 2, it can be concluded that both vectors from Kearns et al.33_{are not suitable for the use of CRISPRi since dCas9 is not transcribed in the transduced cells. This}

could be proven based on the immunostainings for dCas9, which resulted in a negative signal. Also the no reduction of KANSL1 expression (figure 10) was observed. The lack of dCas9 expression is also resulting in the unchanged levels of KANSL1 expression, indicated by the western blot that was performed. However, Kearns et al.33_{used the same constructs as construct 1 and 2 in the}

supplementary data. Yet, they have proven the effective performance of these constructs by generation either knockdown or transcription activation. We were not able to successfully use these constructs. Kearns et al. performed CRISPRi in human ESCs. Perhaps the promoter is rather active in ESCs than in iPSCs.

The iPSCs treated with puromycin after the transduction with construct 1 grew well, so it can be concluded that at least the puromycin is integrated into the genome. Based on this, we can ensure that construct 1 is integrated into the genome of the transduced cells, so the lack of dCas9 expression and KANSL1 reduction is not due to failure in the lentiviral delivery. Because we actually prefer a construct with a doxycycline inducible promoter as this enables gene silencing at different time points, we made fresh virus to prevent MOI due to the freeze-thaw cycle. The freeze-thaw cycle could cause breakdown of virus particles, so this decreases the amount of virus titers.

Because the constructs from Kearns at al.33_{were not working properly,}

we used the CRISPRi WTC cell line cloned by Mandegar et al. (construct shown in supplementary data 4) from the Gladstone Institute in San Francisco.32_{This construct (figure 14)}

also contains a doxycycline inducible promoter, which is very advantageous. First, the cells were cultured and exposed to doxycycline to ensure by means of immunostainings that the cells

express dCas9. In this construct, a mCherry reporter is present, which enables to check the presence and inducibility of the construct. In figure 14, you can see that the dCas9 insertion is located between the mCherry reporter and the doxycycline inducible promoter (TRE3G in the figure), so the mCherry fluorescence observed in the immunostaining due to the addition of doxycycline, suggests strongly that dCas9 is also expressed in the transduced cells. However, expression of the reporter does not necessarily mean that the gene of interest is stably expressed as well. The construct shown in figure 14 is inserted into the AAVS1 locus. This is because Hockemeyer et al.34_{figured out that transgenes}

integrated into the AAVS1 locus in the genome of iPSCs remains actively transcribed in iPSCs as well as

Figure 14, Overview of the TALEN targeted insertion of dCas9 and rtTA constructs into the AAVS1 locus. The CAG promoter drives the rtTA promoter. A neomycin resistance gene is included in the rtTA construct. dCas9-KRAB is driven by the third-generation doxycycline-response element. The localization of the rtTA construct in the opposite way prevents leaky expression of the constructs.

(23)

22 in differentiated cells. This ensures that the promoter will still be active during differentiation so dCas9 will still be actively transcribed and expressed in iNeurons. However, we cannot yet compare our results to this. Additionally, the doxycycline-controlled reverse transcriptional activator (rtTA) is driven by the CAG promoter, which is located in the opposite direction to the TRE3G promoter to prevent leaky expression of dCas9. The absence of leaky expression can be confirmed by the absence of mCherry expression in non-induced cells rather than in the doxycycline treated cells. Based on our result in figure 13, it can be concluded that dCas9 is indeed not leaky since no dCas9 was present in the absence of doxycycline.

In figure 15 is shown the immunostaining done by Mandegar et al. to detect dCas9 expression, which shows a similar result to our observations. This is what we hypothesized as it is the same cell line. Based on this information, it can be concluded that we now have an iPSC cell line that stably expresses dCas9 without leaky expression for doxycycline induction. However, based on our results, we cannot confirm the hypothesis that KANSL1 knockdown has an influence on the growth rate or will induce cell death and that KANSL1 knockdown will lead to the induction of autophagy in neuronal cells.

Figure 15, Immunostaining for CRISPRi colonies 48 hours after doxycycline treatment. dCas9-KRAB is stained with GFP and nuclei with DAPI.

(24)

23

6. Conclusion

Based on our results, it can be concluded that constructs 1 and 2 (supplementary data 1 and 2) are not suitable to express dCas9 in the 29.3 iPSCs and in iPSCs derived iNeurons as well. However, the CRISPRi WTC cell line does show dCas9 expression when induced by doxycycline. In addition, it can be concluded that expression of dCas9 in the CRISPRi WTC line is not leaky as neither mCherry, nor GFP signal upon dCas9 staining occurs.

7. Future perspective

Now that we have a cell line that expresses dCas9, we can continue the project with testing the sgRNAs. But first, western blot for dCas9 in the new CRISPRi WTC cell line could be done to ensure that the observed background signal in the dCas9 immunostainings is not due to low dCas9 expression levels. However, this is not wat we hypothesize because a green fluorescent calcium-modulated protein 6 fast type (GCaMP) calcium sensor is also included in the construct (supplementary data 4), which could lead to GFP expression without doxycycline treatement.32_{In case this is due to unspecific binding to}

another protein we would observe a band on the blot, which probably does not correspond to the protein size of dCas9. After the specificity of the antibody is confirmed, the CRISPRi WTC cells can be infected with the three different sgRNAs via lentiviral delivery to see if a dCas9-sgRNA complex formation is formed and leads to significant knockdown of KANSL1. For this experiment, it is not necessarily need to do a puromycin selection control for the sgRNAs as this was done before when we wanted to test the sgRNAs in cells infected with construct 1. There we saw that the sgRNA-virus is very efficient because almost no cell death was observed after puromycin treatment. It could be possible that the KANSL1 deficient cells are not able to grow well or to even survive because KANSL1 is involved in the regulation of H4K16 acetylation, which is really important during development. Additionally, it was recently shown that KANSL1 and KANSL3 are microtubule-associated proteins that serve for the localization of the mitotic spindles during mitosis.3_{When the KANSL1 knockdown results in too many}

dead cells, a possibility is to reduce the doxycycline concentration since this influences the level of dCas9 expression.32

When we have the stable cell line that performs CRISPRi for KANSL1 knockdown, the dCas9 positive cells can be differentiated into iNeurons by transducing them with the Ngn-2 transcription factor. The high sgRNA-virus efficiency enables us to infect dCas9 positive cells with the sgRNA in every desirable moment. Otherwise, the transformation of KANSL1 deficient (transduced with sgRNA) iPSCS into iNeurons would otherwise result in KANSL1 knockdown when the iNeuron differentiation is started. The development of the iNeurons can be studied and compared to a Koolen-de Vries Syndrome patient’s cell line. Additionally, the effect of KANSL1 on H4K16 acetylation and the outcome of autophagy can be studied. In the CRISPRi iPSCs and iNeurons as well, the role of Histone 4 Lysine 16 can be studied in order to investigate whether if the deficiency of H4K16ac plays a bigger role in iPSCs than in iNeurons as we hypothesized.

Using an LC3 construct designed by Zhou et al.35_{, we could monitor the autophagic flux in autophagy}

induced cells. In the research of Zhou et al., they use two constructs: mRFP-EGFP-LC construct and mTagRFP-mWasabi-LC3 construct. Both systems are based on the difference in pH stability of the fluorescent proteins within the constructs. RFP is less pH sensitive, so fluorescence signals remains active in acidic environments of a pH below 5, whereas the signal of GFP will be quenched in acidic

(25)

24 environments. Therefore, cells transfected with the LC3 construct that show red puncta indicate autolysosomes. However, little GPF expression is present in autolysosomes, which could result in yellow puncta when the RFP-GPF-LC3 construct is used. Yellow puncta, wherein both GFP and RFP are expressed represent autophagosomes. Based on this assay, it can also be concluded whether the autophagic flux is either induced or blocked. When only green and yellow puncta are present, this means that autophagic flux is blocked as no fusion of autophagosomes and lysosomes occurs. However, mWasabi in the mTagRFP-mWasabi-LC3 construct is more acid sensitive and GFP and mTagRFP results in brighter fluorescence than RFP which improves the assay when this construct is used instead of GFP-RFP-LC3. The use of the mTagRFP-mWasabi-LC3 construct enables a better discrimination between autophagosomes and lysosomes. This assay could be a nice comparison between the rapamycin induced autophagy in iNeurons and the KANSL1 deficient cells to see whether this shows the same or a different phenotype for autophagy. When we then do immunostainings, we can localize the autophagosomes and / or autolysosomes in for instance post-synaptic neurons.35_The

LC3 construct could also be used for patient’s line to investigate the autophagy induction based on the amount of autophagosomes and autolysosomes.

(26)

25

8. References

1. Larson, M.H. et al. CRISPR interference (CRISPRi) for sequence-specific control of gene expression. Nat Protoc 8, 2180-96 (2013).

2. Marzban, H. et al. Cellular commitment in the developing cerebellum. Front Cell Neurosci 8, 450 (2014).

3. Meunier, S. et al. An epigenetic regulator emerges as microtubule minus-end binding and stabilizing factor in mitosis. Nat Commun 6, 7889 (2015).

4. Dominguez, A.A., Lim, W.A. & Qi, L.S. Beyond editing: repurposing CRISPR-Cas9 for precision genome regulation and interrogation. Nat Rev Mol Cell Biol 17, 5-15 (2016).

5. Clontech Laboraties, I. Tet-On 3G Inducible Expression Systems (Clontech Loboraties, Inc). 6. Koolen, D.A. et al. Mutations in the chromatin modifier gene KANSL1 cause the 17q21.31

microdeletion syndrome. Nat Genet 44, 639-41 (2012).

7. Dias, J. et al. Structural analysis of the KANSL1/WDR5/KANSL2 complex reveals that WDR5 is required for efficient assembly and chromatin targeting of the NSL complex. Genes Dev 28, 929-42 (2014).

8. Williams, S.C. Epigenetics. Proc Natl Acad Sci U S A 110, 3209 (2013).

9. Weinhold, B. Epigenetics: the science of change. Environ Health Perspect 114, A160-7 (2006). 10. Simmons, D. Epigenetic Influences and Disease. (Nature Education, 2008).

11. Bannister, A.J. & Kouzarides, T. Regulation of chromatin by histone modifications. Cell Res 21, 381-95 (2011).

12. Johnstone, R.W. Histone-deacetylase inhibitors: novel drugs for the treatment of cancer. Nat Rev Drug Discov 1, 287-99 (2002).

13. Moreno-Igoa, M. et al. KANSL1 gene disruption associated with the full clinical spectrum of 17q21.31 microdeletion syndrome. BMC Med Genet 16, 68 (2015).

14. Koolen, D.A. et al. Clinical and molecular delineation of the 17q21.31 microdeletion syndrome. J Med Genet 45, 710-20 (2008).

15. Koolen, D.A. et al. The Koolen-de Vries syndrome: a phenotypic comparison of patients with a 17q21.31 microdeletion versus a KANSL1 sequence variant. Eur J Hum Genet (2015). 16. Fullgrabe, J. et al. The histone H4 lysine 16 acetyltransferase hMOF regulates the outcome of

autophagy. Nature 500, 468-71 (2013).

17. Shehata, M., Matsumura, H., Okubo-Suzuki, R., Ohkawa, N. & Inokuchi, K. Neuronal stimulation induces autophagy in hippocampal neurons that is involved in AMPA receptor degradation after chemical long-term depression. J Neurosci 32, 10413-22 (2012).

18. Banerjee, R., Beal, M.F. & Thomas, B. Autophagy in neurodegenerative disorders: pathogenic roles and therapeutic implications. Trends Neurosci 33, 541-9 (2010).

19. Mizushima, N. Autophagy: process and function. Genes Dev 21, 2861-73 (2007).

20. Cecconi, F. & Levine, B. The role of autophagy in mammalian development: cell makeover rather than cell death. Dev Cell 15, 344-57 (2008).

21. Klionsky, D.J. The molecular machinery of autophagy: unanswered questions. J Cell Sci 118, 7-18 (2005).

22. Zhao, R. & Daley, G.Q. From fibroblasts to iPS cells: induced pluripotency by defined factors. J Cell Biochem 105, 949-55 (2008).

23. Evans, M.J. & Kaufman, M.H. Establishment in culture of pluripotential cells from mouse embryos. Nature 292, 154-6 (1981).

24. Nichols, J. et al. Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell 95, 379-91 (1998).

25. Avilion, A.A. et al. Multipotent cell lineages in early mouse development depend on SOX2 function. Genes Dev 17, 126-40 (2003).

(27)

26 26. Zhang, Y. et al. Rapid single-step induction of functional neurons from human pluripotent

stem cells. Neuron 78, 785-98 (2013).

27. Jackson, A.L. et al. Expression profiling reveals off-target gene regulation by RNAi. Nat Biotechnol 21, 635-7 (2003).

28. Wolfe, S.A., Nekludova, L. & Pabo, C.O. DNA recognition by Cys2His2 zinc finger proteins. Annu Rev Biophys Biomol Struct 29, 183-212 (2000).

29. Barrangou, R. et al. CRISPR provides acquired resistance against viruses in prokaryotes. Science 315, 1709-12 (2007).

30. Hockemeyer, D. & Jaenisch, R. Induced Pluripotent Stem Cells Meet Genome Editing. Cell Stem Cell 18, 573-86 (2016).

31. Das, A.T., Tenenbaum, L. & Berkhout, B. Tet-On systems for doxycycline-inducible gene expression. Curr Gene Ther (2016).

32. Mandegar, M.A. et al. CRISPR Interference Efficiently Induces Specific and Reversible Gene Silencing in Human iPSCs. Cell Stem Cell 18, 541-53 (2016).

33. Kearns, N.A. et al. Cas9 effector-mediated regulation of transcription and differentiation in human pluripotent stem cells. Development 141, 219-23 (2014).

34. Hockemeyer, D. et al. Genetic engineering of human pluripotent cells using TALE nucleases. Nat Biotechnol 29, 731-4 (2011).

35. Zhou, C. et al. Monitoring autophagic flux by an improved tandem fluorescent-tagged LC3 (mTagRFP-mWasabi-LC3) reveals that high-dose rapamycin impairs autophagic flux in cancer cells. Autophagy 8, 1215-26 (2012).

(28)

27

Supplementary data

(29)

28

(30)

29

(31)

30