Dissecting Glycosyltransferase Activity in Human Mesenchymal Stem Cells to Enforce Cellular Migration

UNIVERSIT Y O F COPENHAGEN F A C U L T Y O F H E A L T H A N D M E D I C A L S C I E N C E S Dissecting Glycosyltransferase Activity in Human Mesenchymal Stem Cells to Enforce Cellular Migration Master’s Thesis, Medicine (12,5 ECTS) Amalie Dahl Haue Supervisor: Hans H. Wandall Submission Date: September 12th 2016 Name of department: Department of Cellular and Molecular Medicine Primary site of activity: Copenhagen Center for Glycomics Author: Amalie Dahl Haue, BSc, BA KU Username: fhq750 Title: Dissecting Glycosyltransferase Activity in Human Mesenchymal Stem Cells to Enforce Cellular Migration Supervisor: Professor Hans H. Wandall, MD, PhD Submission date: September 12th 2016 Word count: 5.825 The provided data concludes my recent research training as a summer student. Research training period: June 8th 2016 – August 19th 2016 Research training place: Laboratory of Dr. Robert Sackstein 77 Avenue Louis Pasteur, Boston, MA 02115, USA ___________________________________________________________________ Title page illustration has been cropped from EMBL Symposium Poster. Symposium held 5 December 2011 in Heidelberg, Germany. i Preamble The study of glycans dates back to Emil Fischer’s work systematizing the description of hexose sugars by making actual 3D-structures accessible in a 2D-format. Today, roughly a century later, the field of glycobiology is promising as a diagnostic and therapeutic tool. Years of intensive and systematic study have even given rise to the term glycomedicine, demonstrating the great progress and promising future of the field. Thanks to combined insight, effort and hospitality of Drs. Robert Sackstein, Director of Program of Excellence in Glycosciences and Hans H. Wandall, Professor of Glycomedicine, I was very privileged to spend The Summer of 2016 as a research trainee in The Laboratory of Dr. Robert Sackstein. The purpose of this report is to present and evaluate my research training experience. Formally, the report will constitute The Master’s Thesis of my Medical Degree. The experiments that I conducted during my recent research training were in accordance with Dr. Sackstein’s therapeutical mission, including weekly consultations and written consent agreeing to his intellectual property. Post Doc Nandini Mondal performed routinely supervision and laboratory manager Kyle C. Martin provided technical as well as theoretical assistance. Stud.med. Amalie Dahl Haue Panuminstituttet, København Den 11. september 2016 ii Resumé Formålet med nærværende opgave er at præsentere min erfaring som medicinstuderende og sommergæst i et forskningslaboratorium til enhver med medicinske interesser, intentioner og visioner i et format der opfylder “Recommendations for the Conduct, Reporting, Editing and Publication of Scholarly Work in Medical Journals”. Læringsmæssigt havde opholdet to komponenter, hvoraf den ene var at undersøge en specifik kulhydratstruktur og den anden var at udvide mit molekylærbiologiske metodekendskab. Den specifikke kulhydratstruktur kaldes the sialyalted lewis x binding determinant, og betegner en struktur der indgår i den klassiske beskrivelse af cellulær migration under såvel normalfysiologiske forhold som sygdomsmæssige processer. Histologisk er den udtrykt i adskillige humane væv. Dog er den ikke udtrykt i humane mesenkymale stamceller. Rent topologisk er the sialyalted lewis x binding determinant udtrykt på mange proteiner herunder CD44, hvilket er et protein der grundet dets gennemgribende kompleksitet er interessant at studere i biologiske og fysiologiske systemer. Udvidelse af mit molekylærbiologiske metodekendskab involverede indsigt i to in vitro metoder. Disse metoder er henholdsvis praktisk håndtering af humane mesenkymale stamceller samt enzymatisk behandling af intakte celler, herunder også humane mesenkymale stamceller. Sidstnævnte foregik ved at inducere glykosyleringen af membranbundne proteiner ved anvendelse en metode, der betegnes exofucosylation. Opholdets væsentligste eksperimentelle fund var at ved forsøg på at slukke for fire individuelle glycosyltransferaser i humane mesenkymale stamceller var i stand til at detektere diskrete fænotyper. Opgaven er inddelt i fem sektioner. Den første del indeholder en generel introduktion til glykobiologi. Den anden sektion forbinder den generelle glykobiologi med det eksperimentelle arbejde, som jeg udførte under mit ophold. Den tredje sektion præsenterer og diskuterer det datasæt, jeg indsamlede under opholdet. I den fjerde sektion fremstilles en syntese af hovedkonklusionerne. Den femte og sidste sektion præsenterer en detaljerede beskrivelse af henholdsvis materialer og metoder, som jeg konstruerede og arbejdede med under opholdet. iii Abstract The aim of the present thesis is to communicate my summer research training experience as a Medical Student to anyone with an intention to treat and a vision to prevent in a format the strives to fulfill the “Recommendations for the Conduct, Reporting, Editing and Publication of Scholarly Work in Medical Journals”. The two main components of the research training experience were (i) to study one particular classic glycan modification and (ii) to expand my technical experimental repertoire. The glycan structure in questions is the sialylated lewis x binding determinant, which is a structure known to facilitate cellular migration in health and disease. Histologically, it is expressed in many tissues of man, but it is characteristically absent from human mesenchymal stem cells. Topologically, the sialylated lewis x binding determinant is expressed on many proteins, including CD44, a protein of great biological interest owing to its complexity. Expansion of my technical experimental repertoire counted two, namely working with human mesenchymal stem cells and in vitro enzymatic preparation of intact cells, by enforced cell surface glycosylation benefitting from method known as exofucosylation. Scientifically, I was able to detect only modest phenotypes by knockout of four individual glycosyltransferase genes expressed in human mesenchymal stem cells. The thesis has five sections. The first section is a general introduction to glycobiology with both a structural and a functional focus. The second section links general glycobiology with the experimental work I conducted as a research trainee. The third section presents and evaluates the data that I collected during my training. A synthesis of the main conclusions will be presented in section four. The fifth and final section presents a detailed description of the materials and methods that I prepared and worked with during my stay. 1 CONTENTS THEORETICAL INTRODUCTION 2 GLYCOBIOLOGY AT A GLANCE 2 GLYCANS AT WORK: THE MULTI-STEP PARADIGM OF CELLULAR MIGRATION 3 BRIDGE 5 EXPERIMENTAL MOTIVATION 5 THE SIALOFUCOSYLATED LACTOSAMINE DETERMINANT 5 EXPERIMENTAL OBJECT 7 RESULTS 8 DISCUSSION 11 CONCLUSION 13 MATERIALS AND METHODS 15 HUMAN MESENCHYMAL STEM CELLS (MSCS) 15 CHARACTERIZATION OF GLYCOSYLTRANSFERASES EXPRESSED IN HUMAN MSCS 15 CONSTRUCTION OF VECTORS FOR TRANSFECTION 16 TRANSFECTION OF MSCS 16 ENZYMATIC PREPARATION OF HUMAN MSCS 17 CHARACTERIZATION OF TRANSFECTED MSCS BY PCR ON GENOMIC DNA 17 CHARACTERIZATION OF TRANSFECTED MSCS BY FLOW CYTOMETRY 17 LITERATURE 19 2 Theoretical Introduction Glycobiology at a Glance Glycosylation is a post-translational modification present on membrane bound proteins and lipids of both proand eukaryotes and are categorized depending on their relation to the protein backbone or lipid structure. The glycoprotein family counts Nglycosylation, O-glycosylation and glycosaminoglycans characteristically displaying repeating disaccharide units. Glycosaminoglycans are also qualitatively different from the former two, as they exist both individually and protein bound. Also, glycosaminoglycans are linear and synthesized by a distinct class of enzymes. As with glycoproteins, glycolipids can be modified by the terminal glycan sialic acid. If a glycolipid is sialylated it is called a ganglioside (Ohtsubo & Marth, 2006). The enzymatic machinery responsible for the creation of Nand O-glycosylation resides in the Golgiapparatus. Phylogenetically, they all belong to the glycosyltransferase family which occupies approximately 1% of the human genome (Rini J, 2009). The glycosyltransferases function in a step-wise fashion, where individual sugar moieties or polysaccharides are added to the non-reducing end of the developing glycoprotein or -lipid. At a molecular level glycosylation is unique in that the structures are not directly translated from the genome. In other words, it is neither possible to predict a glycan structure from an amino acid sequence, nor possible to derive the enzymatic cascade that precedes a specific glycan structure. This interdependency between the molecules present in the micro milieu and the evolving glycan structure is termed microheterogeneity. Functionally, post-translational glycan modifications have a substantial impact as actors in biological systems, as they may be mediators of nearly all biological processes including cell adhesion, molecular trafficking and clearance, receptor activation, signal transduction and endocytosis (Ohtsubo & Marth, 2006). Figure 1. Common classes of glycans: Modified from Varki A. 1997. FASEB J. Downloaded from: http://www.ncbi.nlm.nih.gov/books/NBK1931/figure/ch1.f6/?report=objectonly 3 Classical examples of glycoproteins that mediate cell adhesion are the integrins defined by their RGD-binding motif (Arnaout, 1990). Functionally, it has been stipulated and evidentially supported that glycans are implicated in molecular trafficking and clearance by a variant of endocytosis, that is characteristically clathrin-independent (Stanley, 2014). A classic example of glycans involved in receptor activation stems from studies of the Ashwell-Morell receptor, which belongs to the family of asialoglycoprotein receptors (Grewal et al., 2008). Other examples that manifest the biological significance of glycosylation in signal transduction has evolved from studies of the Notch receptor, which is a transmembrane protein with inherent catalytic activity facilitating homophilic receptor interactions (Takeuchi & Haltiwanger, 2014). Glycans at Work: The Multi-Step Paradigm of Cellular Migration In a clinical-physiological context, the significance of glycans dates all the way back to Karl Landsteiners discovery of red blood cells ability to agglutinate and subsequently mapping of the AB0-blood group system (Bayne-Jones, 1931). To date, the field of glycobiology is rapidly evolving as an empirical platform with diagnostic and therapeutical perspectives benefitting from sophisticated and continuously evolving genetic engineering technologies (Galluzzi et al., 2014; Hudak & Bertozzi, 2014). The multi-step paradigm of cellular migration is an example of a physiological model that is strictly dependent on receptor signaling involving glycan structures. Specifically, it involves the interaction of selectins expressed on the endothelium and the sialofucosylated lactosamine binding determinant (introduced in greater detail below) expressed on circulating cells, mainly leukocytes. Indeed, the interaction of circulating cells with the endothelial bed is a ubiquitously physiological phenomenon. Examples include cellular homing, a concept originally introduced to explain the observed flux of lymphocytes between blood and lymph (Sackstein, 2012) and conversely recruitment of neutrophils from the marrow to sites of infection in systemic circulation (Borregaard, 2010). Originally, the multi-step model of cellular migration was developed to describe the extravasation of leukocytes to the vasculature as a physiological component of stress (Springer, 1994). During embryogenesis, the receptor-ligand interactions of selectins and the sialofucosylated lactosamine expressed on circulating cells operate to form cell-cell adhesion responsible for the colonization of lymphoid organs by lymphocytes. Additionally, the sialofucosylated lactosamine is an important mediator of the inflammatory response as well as a structural actor of a malignant hallmark, namely metastasis. In fact, the selectins have been ascribed the primary function of glycan recognition in mammalian immune function (Marth & Grewal, 2008). 4 Originally, three consecutive steps defined the multi-step paradigm of cellular migration. It is in the first step that the binding of the sialofucosyalted lactosamine and the selectins is prominent. Other important mediators of cellular migration the chemokines (e.g. TNFand IL-1) and integrins (e.g. LFA-1 and VLA-4), operating chemoattractively through G-protein-coupled receptors and binding between opposing transmembrane type I proteins, respectively (Springer, 1994). Figure 2. The multi-step model of cellular migration: Sackstein R, Immun. Review, 2009. 11 MFI was reduced to 38%, 26% and 48% of baseline level in knockout of ST3GalTIII, -IV and –VI, respectively. Thus, the data suggests that ST3GalTIV is the major sialyltransferase in generation of the sLex determinant in human MSCs (figures 1.3 and 1.4). Explicitly knockout of ST3GalTIV yields the largest HECA-452 negative population (figure 1.4, left panel). Figure 1.4. Detection of cell surface sialylated and unsialylated lactosamines in ST3GalT knockouts. Left panel: Exofucosylation with FT6, then staining with HECA-452 mAb, specific for sLex Right panel: Exofucosylation with FT6, then staining with anti-CD15 mAb, specific for Lex MFI: Mean fluorescence intensity Discussion The results provide evidence that multiple enzymes are involved in the synthesis of sLex binding determinant in human MSCs. The data suggests, that there is not a one to one relation between a particular structure (i.e. the disaccharide bond) and a particular enzyme (i.e. a glycosylor sialyltransferase). Therefore, a number of biological regulatory mechanisms may explain the discrete phenotypes. In addition, the fact that a number of concerning technical challenges remain unsettled, complicate a rigorous analysis. The biological as well as technical considerations will be covered in greater detail below. The technical challenges are partly due to the rigorous dependency on appropriate controls. We did not, for instance have an appropriate control in detection of the type 2 lactosamines by ECL (right panel, figure 1.2). Consequently, the data does not assess if staining with ECL of human MSCs correlates with the expected ECL specificity. However, the data suggests that native human MSCs present with type 2 lactosamine binding epitopes, as they are ECL positive (right panel, figure 1.2). Correspondingly, native MSCs are CD15 negative (right panel, figure 1.3). In effect, the majority of type 2 lactosamine epitopes on human MSCs are unfucosylated. Thus, 12 this piece of data correlates with the fact that native human MSCs express no (1,3) fucosyltransferases (addendum from Dr. Sackstein). Moreover, as anti-CD15 mAb recognizes the unsialylated type 2 lactosamine (Lex), the data underscores that native human MSCs are highly sialylated (discussed in greater detail below). Another technical challenge is that it is notoriously difficult to assess specifically how enzymatic preparation of cells affects the cellular phenotype including viability. Explicitly, prior to detection of the sLex binding determinant, cells were exofucosylated and prior to detection of the Lex binding determinant, cells were neuraminidase treated. Therefore, the three reporters (i.e. HECA-452, ECL and antiCD15 mAb) are not readily comparable. The current strategy in dissection of the glycsosyltransferase redundancy is strictly dependent on continuous in vitro availability of human MSCs. In effect of being a primary human cell line, the MSCs present are a major experimental opportunity, as they closely resemble the theoretical basis of the enzymatic dissection. However, practically it is not the most efficient way to assay cell surface glycosylation as human MSCs are rarely passed for more than a month in vitro (ex vivo). Thus most likely, establishment of a stable knockout cell line is restricted by the limited replication potential of MSCs in culture. In order to extend the in vitro replication potential of human MSCs, we attempted to optimize the growth conditions by substituting the media component of fetal bovine serum (FBS) for platelet lysate. In casual evaluation by direct light microscopy over a period between 24 and 48 hours, we observed altered morphology manifested as smaller cells and less spindle formation accompanied by accelerated growth. Besides, a considerable limit to the quality of the data is the quantity of cells that were included for analysis. All individual histograms represent counts of less than 5.000 cells. For comparison, it is reported that a cell population of 1 x 106 individual cells is analyzed for publication (Dykstra et al., 2016). The modest phenotype in the 4GalT1 knockout population is a conceivable consequence of the technical challenges outlined above or a manifestation of glycosyltransferase redundancy. It may even be a combination of both. Taken together, the data suggests that enzymatic redundancy account for lack phenotype in the 4GalT1 knockouts as an isoenzyme, namely 4GalT2 is expressed in human MSCs (figures 1.1 and 1.2). A supporting observation is that the present data set identifies that the expression level of 4GalT2 is similar to the expression levels of ST3GalTIII and –IV (figure 1.2) one of which, presumably, accounts for creation of most sialylated type 2 lactosamines in human MSCs (figure 1.4). In sum, enzymatic dissection of sialylated cell surface glycans is internally coherent as MFI in non-transfected human MSCs is equal in the two unrelated measurements (left panels of figures 1.1 and 1.2, respectively). Further, the data suggests that human MSCs are highly sialylated, as staining of native MSCs with HECA-452 is positive (left panel, figure 1.4). The current model system does not discriminate the specific 13 backbone of the adjacent glycan structures. Therefore, the sialic acid may be predominantly expressed on either N-linked, O-linked glycans or glycolipids. For reference (data not included), the abundance of sialic acid on human MSCs was assessed specifically by staining with indirectly FITC conjugated Maackia Amurensis Lectin II (MAL II), a lectin that specifically recognizes (2,3) sialic acids. Consistently, reporting with MAL II suggests that human MSCs are highly sialylated. Knockout of the individual sialyltransferase genes reduces the MFI to 85%, 43% and 69% (ST3GalTIII, -IV and -VI, respectively) of baseline level corresponding to MAL II staining of non-transfected human MSCs. Based on the width of the histograms, the data presents an unexpected variation in assessment of the sLex and Lex binding determinants, respectively. Possibly, the variation reflects cellular senescence, as the two sets of experiments were conducted two weeks apart (figure 1.3). Furthermore, literature reports that MSCs express a βgalactosidase activity related to senescence (see Materials & Methods and figure 0.3 for details). Thus, the presence of lactosamine determinants decreases as cell age and at such may account for an altered phenotype unrelated to genetic engineering. The key step in establishment of a sensitive evaluation of glycosyltransferase activity in human MSCs consists in optimization of the transfection procedure. For enrichment strategies, introduction of a green fluorescence protein (GFP) tag on the cas9 endonuclease enables enrichment based on fluorescent activated cells sorting (FACS) methodologies. Although the technique has been successfully reported (Yang et al., 2015), it does not enable sustained monitoring of target gene expression level as the cas9 endonuclease is only transiently expressed. In contrast, continuous monitoring of the transfection efficiency can be obtained by introduction of a reporter gene at the cleavage site of the cas9 endonuclease. Potentially an even more sophisticated strategy is to generate a human mesenchymal stem cell line deficient of sialylated cell surface epitopes. The perspective can either be obtained by cotransfection of multiple plasmids each targeting the individual genes, construct one plasmid that harbors more than one gRNA or knockout of selected enzymes upstream of the enzymatic pathways involving catalysis of glycosyland sialyltransferases, respectively (Steentoft et al., 2014). Conclusion In conclusion, the report presents evidence that synthesis of the type 2 lactosamine determinant in human MSCs is a complex and highly regulated process; consistent with conventional evaluation of glycosylatransferase activity and specificity (Wandall et al., 1997). Additionally, the report emphasizes some technical challenges to the experiments (see Discussion for details). 14 Although only a minor effect was observed in knockout of β4GalT1, the transfection resulted in an expected decreased presence of type 2 lactosamines epitopes on the cell surface of human MSCs. In addition, it cannot be stressed enough that lack of positive controls to assess β4GalT1 knockout phenotype muddles the conclusion even further. Regardless, it appears that β4GalT1 is dispensable for creation of type 2 lactosamines in human MSCs. At such a “rescue phenotype” dominated by expression of β4GalT2 may explain the minor effect of knocking out β4GalT1. Knockout of the individual sialyltransferases expressed in human MSCs, resulted expectedly in reduced presence of sialylated type 2 lactosamines epitopes on the cell surface. Further, the data suggests that although redundantly expressed, the sialyltransferases do not have equal enzymatic capacity. That is, the sialyltransferase knockout data suggests that sialylation of type 2 lactosamine in human MSCs is dominated by activity of ST3GalIV. In comparison, knockout of the sialyltransferases in human MSCs has a greater phenotypic impact than knockout of β4GalT1. In sum, we identified at least four enzymes involved in synthesis of the type 2 sialylated lactosamine in human MSCs. In addition to the rigorous dependency on proper controls, the immediate technical challenges include two. These are optimizing management of human MSCs in vitro and optimizing the transfection procedure per se. The former can be obtained by adjusting the specific media formulation. Although not evidentially supported, a considerable effect was observed by substituting FBS for platelet lysate. Perspectives on optimization of the transfection procedure include application of FACS preceded by introduction of a GFP-tag and inducing the expression of a reporter gene. According to the biological perspective (see Experimental Motivation for details), the establishment of a robust assay based on the current strategy, will generate a descriptive dataset. Moreover, the therapeutical perspectives demands generation of an operational platform, where a particular physiological effect (i.e. enforced cellular migration) is obtained by genetic engineering of human MSCs. Explicitly, the latter readout is functional. The former is structural. Despite the qualitative difference between the two readouts (i.e. functional and structural), they are dependent on each other as genetic dissection of enzymatic regulation is necessary in order to benefit therapeutically from physiological models assessed biochemically. Thus, genetic dissection of the glycosyltransferase activity in human MSCs demands careful and persistent attention in order to obtain enforced cellular migration and, subsequently differentiation to tissues were they are needed for immunity and tissue repair. 15 Materials and Methods Human Mesenchymal Stem Cells (MSCs) All reported cellular assays throughout were conducted in human mesenchymal stem cells (MSCs). Consequently, their cellular characteristics will be introduced briefly. Experimental work with human MSCs has been reported for half a century (Costero, Chevez, Barroso-Moguel, & Pomerat, 1955). They are defined by their ability to give rise to a broad set of differentiated mesenchymal cell types including osteoblasts, adipocytes, myocytes and chondrocytes. At such, they have a robust and characteristic differentiation potential. By definition, MSCs are categorized as non-hematopoietic stem cells. Phenotypically, they are defined by the expression of CD105, CD73 and CD90 and absence of CD11b, CD14, CD45 and CD34. Additionally, there is consensus that human MSCs express CD44 (Uccelli, Moretta, & Pistoia, 2008). MSCs are particularly interesting as a research platform in the attempt to advance therapeutical options for regenerative medicine, as MSCs consequently to their infinite cell cycle have potential to foster tissue repair. In connection to their microenvironment, MSCs also have potent anti-proliferative and -inflammatory effects (Uccelli et al., 2008). As stem cells age, they alter phenotypically at a molecular level as well as metabolically. MSCs are no exception. For instance expression of the tumor suppressor p53 is part of the mesenchymal stem cell senescence response. Also, MSCs undergo irreversible changes and typically exhibit –galactosidase activity related to senescence. Physiologically, MSCs have a remarkable capacity of inhibiting the immune response, making them – seemingly – even more promising in facilitating tissue regeneration in man (Dominici et al., 2006). In culture, human MSCs are adherent, which is an additional defining characteristic of MSCs (Dominici et al., 2006). However, in comparison with other adherent cell lines (e.g. HaCaT cell line (Boukamp et al., 1988)) they are lifted easily. Morphologically, they are characteristically star-shaped and exhibit some similarities with fibroblast. Cells from the average donor may survive ex vivo for about 8 passages, corresponding to the order of 40 days (personal observations). Characterization of Glycosyltransferases Expressed in Human MSCs Expression levels of the enzymes in question were assessed by quantitative polymerase chain reation (qPCR) and measured relative to the expression of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Schmittgen & Livak, 2008). Amplification of the genes in question was set up for PCR in 12 L using the SYBR® Select Master Mix (appliedbiosystems, by Life Technologies). Primers were prepaired at a stock concentration of 10 M (5 M of forward and reverse primer). For reactions, primers were diluted to a final concentration of 1.25 M. The PCR templates consisted in cDNA transcribed in vitro from mRNA of two individual donors. The relative expression level of each gene was based on the geometric mean derived from replicates of three and was expressed relative to GAPDH, serving the 16 function of a housekeeping gene. The amplification was performed with The StepOnePlus™ Real Time System (ThermoFisher Scientific) and analyzed using GraphPad Prism. Construction of Vectors for Transfection Plasmid vectors for transfection of MSCs were prepared in accordance with the CRISPR/cas9-technology (Cong et al., 2013). Primers targeted at the genes of interest were designed using the UCSC Genome Browser. Search was limited to exon one and primers were ordered from Applied Biosystems™. By evaluating each set of primers at the database available at crispr.mit.edu, optimal on target and minimal off target effects were ensured. The plasmid constructs were analyzed by restriction digestion and sequencing prior to transfection. The plasmid of choice was pX330 (Addgene), which has been designed by the manufacturer such that it harbors sequences encoding nucleation location signals and a promotor upstreams of the modified CRISPR gene locus. For selection purposes, the manufacturer has also designed the plasmid to express an ampicillin resistance gene. The DNA inserts were designed with overhangs compatible with cleavage motive of the restriction enzyme minimizing risk of selfligation. The restriction enzyme was BbsI (New England BioLabs®). Restriction cleavage was performed in a buffer composed of 50 mM NaCl, 10 mM Tris-HCl, 10 mM MgCl2 and 100 g/mL BSA. Reaction was performed at pH 7.9 at 37°C for 1 hour. For ligation, 50 ng of linearized plasmid was incubated with 37.5 ng DNA insert corresponding to a stoichiometric ratio 3:1 between insert and linearized vector backbone. As a means to introduce Gibbs energy into the system and initial challenges producing the plasmid targeting the genetic locus of interest, DNA inserts were phosphorylated prior to annealing. Running buffer was 50 mM Tris-HCl, 10 mM MgCl2, 1 mM ATP and 10 mM DTT. Reaction was performed at pH 7.5. For amplification of the ligated plasmid, it was transduced into One Shot® Stbl3™ Chemically Competent E. coli (invitrogen™, by Life Technologies) and cultured on agar plates with ampicillin at 100 g/mL. DNA was isolated upon binding to a silica membrane and eluted by an increasing salt concentration gradient in accordance with the manufactures protocol (©Qiagen). In order to verify insertion of relevant guide RNA (gRNA) in to the vector, samples were sent for sequencing prior to transfection (Biopolymers Facility @ Harvard Medical School, https://genome.med.harvard.edu/). Transfection of MSCs The MSCs were harvested from human donors. Donors were included according to the Human Experimentation and Ethics Committees of Partners Cancer Care Institutions (Dykstra et al., 2016). The mononuclear fraction was isolated by FicollHypaque density separation, and plated at 2-5 106 cells/mL in T-175 culture flasks. Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM), 10% platelet lysate and 1 % Antibiotic-Antimycotic (Gibco®, by Life Technologies) diluted as prescribed to at final concentration of 100 U/mL of penicillin and streptomycin, respectively. Cells were incubated at 37°C for 18 hours and washed in PBS, whereby 17 isolation of the adherent cell population, that is the MSCs, was obtained. Cells were grown confluent in T-175 flasks, corresponding to 500.000 cells per flask. Prior to transfection, cells were lifted with 0.05% trypsin and 0.5 mM EDTA diluted in phosphate-buffered saline (GIBCO® PBS). Transfection was performed using Lipofectamine® 3000 Transfection Reagent (invitrogen™, by Life Technologies) using 3 g of pre-paired plasmid construct (as described above). Enzymatic Preparation of Human MSCs For exofucosylation, cells were treated with recombinant fucosyltransferase 6 expressed in Pichia pastoris expression system (internal development). Reaction volume was 30 L. Reaction buffer was diluted in Hanks Balanced Salt Solution (HBSS) from 25X stock (internal formulation). Additionally, GDP-fucose was added to a working concentration of 1 mM GDP-fucose. Cells were treated for 1 hour at 37°C and reactions were stopped by dilution in HBSS. For neuraminidase treatment, cells were treated with Neuraminidase (Sialidase) from Clostridium perfringens (Roche). Enzyme was diluted 1:10 corresponding to a working activity of 10 U/mg. Reaction was performed in PBS for 1 hour at 37°C and stopped by dilution with PBS. Characterization of Transfected MSCs by PCR on Genomic DNA Media was removed from transfected MSCs and cells were washed. Subsequently, cells were lifted with trypsin as described above. Then, genomic DNA was isolated with PureLink® Gennomic DNA Kit (Invitrogen, by lifetechnologies) according to manufactures protocol and sample type categorized as mammalian cell lysate. The isolated DNA served as a template in the PCR. Reaction was prepared with volume picomoles of forward and reverse primer, a total of 10 mM dNTPs and 1 L of DNA polymerase. Reaction volume was 50 L and PCR was performed in MJ Research PTC-200 Peltier Thermal Cycler DNA Engine Dual Alpha Blocks. Program was initial denaturation, one cycle at 94 for 2 minutes, then 30 cycles of amplification composed of 94°C for 30 seconds, 56°C for 30 seconds and 72°C for 30 seconds. Final extension was ensured by one cycle at 72°C for 5 minutes. Amplified PCR products were analyzed on a 2% agarose gel prepared with 3% ethidium bromide. Gel ran at 100 V for 30 minutes. Running buffer was Tris-acetateEDTA. Characterization of Transfected MSCs by Flow Cytometry Upon transfection, cells were re-grown in T-175 flask till a confluency of >80%, correlating with an incubation time of approximately 4 weeks. Of note is that, cellular doubling time was slowed significantly after transfection. Cellular development was monitored on a regular basis by direct light microscopy. When cells were confluent, they were washed and lifted as described above. Cells were divided according to the mode of assessment. For ECL-staining, cells were incubated with Fluorescein labeled Erytrhina Cristagalli Lectin (ECL, ECA) (Vector Laboratories) diluted 1:4000 in PBS, incubated at 4°C for 30 min. For staining against the HECA-452 epitope, cells were incubated with FITC anti-human/mouse Cutaneous Lymphocyte Antigen Antibody (Biolegend®) diluted 1:10 and incubated at 4°C for 30 min. For assessment 18 of CD15 epitopes, cells were incubated with FITC anti-human CD15 (SSEA-1) Antibody (Biolegend®). As a staining control, FITC Mouse IgM, Isotype Ctrl Antibody (Biolegend®) was used upon dilution 1:50 in PBS. Data was acquired using FC 500 Series Flow Cytometer (Beckman Coulter) and analyzed by FlowJo®. 19 Literature Alberts B, J. A., Lewix J, Raff M, Roberts K and Walter P. (2007). Intracellular Vesicular Traffic Molecular Biology of The Cell (pp. 776-777): Garland Science. Arnaout, M. A. (1990). Structure and function of the leukocyte adhesion molecules CD11/CD18. Blood, 75(5), 1037-1050. Bayne-Jones, S. (1931). DR. KARL LANDSTEINER NOBEL PRIZE LAUREATE IN MEDICINE, 1930. Science, 73(1901), 599-604. doi:10.1126/science.73.1901.599 Borregaard, N. (2010). Neutrophils, from marrow to microbes. Immunity, 33(5), 657670. doi:10.1016/j.immuni.2010.11.011 Boukamp, P., Petrussevska, R. T., Breitkreutz, D., Hornung, J., Markham, A., & Fusenig, N. E. (1988). Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line. J Cell Biol, 106(3), 761-771. Cong, L., Ran, F. A., Cox, D., Lin, S., Barretto, R., Habib, N., . . . Zhang, F. (2013). Multiplex genome engineering using CRISPR/Cas systems. Science, 339(6121), 819-823. doi:10.1126/science.1231143 Costero, I., Chevez, A., Barroso-Moguel, R., & Pomerat, C. M. (1955). [Fibroblastic system. VI. Morphology of mesenchymal cells in vitro]. Arch Inst Cardiol Mex, 25(2), 125-170. Cummings RD, E. J. (2009). Principles of Glycan Recognition. In C. R. Varki A, Esko JD, et al. (Ed.), Essentials of Glycobiology. Cold Spring Harbor (NY): Cold Spring Harbor Laboraroty Press. Dominici, M., Le Blanc, K., Mueller, I., Slaper-Cortenbach, I., Marini, F., Krause, D., . . . Horwitz, E. (2006). Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy, 8(4), 315-317. doi:10.1080/14653240600855905 Dykstra, B., Lee, J., Mortensen, L. J., Yu, H., Wu, Z. L., Lin, C. P., . . . Sackstein, R. (2016). Glycoengineering of E-Selectin Ligands by Intracellular versus Extracellular Fucosylation Differentially Affects Osteotropism of Human Mesenchymal Stem Cells. Stem Cells. doi:10.1002/stem.2435 Galluzzi, L., Vacchelli, E., Bravo-San Pedro, J. M., Buque, A., Senovilla, L., Baracco, E. E., . . . Kroemer, G. (2014). Classification of current anticancer immunotherapies. Oncotarget, 5(24), 12472-12508. doi:10.18632/oncotarget.2998 Grewal, P. K., Uchiyama, S., Ditto, D., Varki, N., Le, D. T., Nizet, V., & Marth, J. D. (2008). The Ashwell receptor mitigates the lethal coagulopathy of sepsis. Nat Med, 14(6), 648-655. doi:10.1038/nm1760 Hudak, J. E., & Bertozzi, C. R. (2014). Glycotherapy: New Advances Inspire a Reemergence of Glycans in Medicine. Chem Biol, 21(1), 16-37. doi:10.1016/j.chembiol.2013.09.010 Lofling, J., & Holgersson, J. (2009). Core saccharide dependence of sialyl Lewis X biosynthesis. Glycoconj J, 26(1), 33-40. doi:10.1007/s10719-008-9159-z Manzella, S. M., Hooper, L. V., & Baenziger, J. U. (1996). Oligosaccharides containing beta 1,4-linked N-acetylgalactosamine, a paradigm for proteinspecific glycosylation. J Biol Chem, 271(21), 12117-12120. 20 Marth, J. D., & Grewal, P. K. (2008). Mammalian glycosylation in immunity. Nat Rev Immunol, 8(11), 874-887. doi:10.1038/nri2417 Mourant, A. E. (1946). A new human blood group antigen of frequent occurrence. Nature, 158, 237. Ohtsubo, K., & Marth, J. D. (2006). Glycosylation in cellular mechanisms of health and disease. Cell, 126(5), 855-867. doi:10.1016/j.cell.2006.08.019 Rini J, E. J., Varki A. (2009). Glycosyltransferases and Glycan-processing Enzymes. In C. R. Varki A, Esko JD, et al. (Ed.), Essentials of Glycobiology: Cold Spring Harbor Laboratory. Sackstein, R. (2009). Glycosyltransferase-programmed stereosubstitution (GPS) to create HCELL: engineering a roadmap for cell migration. Immunol Rev, 230(1), 51-74. doi:10.1111/j.1600-065X.2009.00792.x Sackstein, R. (2012). Glycoengineering of HCELL, the human bone marrow homing receptor: sweetly programming cell migration. Ann Biomed Eng, 40(4), 766776. doi:10.1007/s10439-011-0461-8 Schmittgen, T. D., & Livak, K. J. (2008). Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc, 3(6), 1101-1108. Springer, T. A. (1994). Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell, 76(2), 301-314. Stanley, P. (2014). Galectins CLIC cargo inside. Nat Cell Biol, 16(6), 506-507. doi:10.1038/ncb2983 Steentoft, C., Bennett, E. P., Schjoldager, K. T., Vakhrushev, S. Y., Wandall, H. H., & Clausen, H. (2014). Precision genome editing: a small revolution for glycobiology. Glycobiology, 24(8), 663-680. doi:10.1093/glycob/cwu046 Takeuchi, H., & Haltiwanger, R. S. (2014). Significance of glycosylation in Notch signaling. Biochem Biophys Res Commun, 453(2), 235-242. doi:10.1016/j.bbrc.2014.05.115 Uccelli, A., Moretta, L., & Pistoia, V. (2008). Mesenchymal stem cells in health and disease. Nat Rev Immunol, 8(9), 726-736. doi:10.1038/nri2395 Varki A, K. R., Toole BP. (2009). Glycosylation Changes in Cancer. In C. R. Varki A, Esko JD, et al. (Ed.), Essentials of Glycobiology. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press. Wandall, H. H., Hassan, H., Mirgorodskaya, E., Kristensen, A. K., Roepstorff, P., Bennett, E. P., . . . Clausen, H. (1997). Substrate specificities of three members of the human UDP-N-acetyl-alpha-D-galactosamine:Polypeptide Nacetylgalactosaminyltransferase family, GalNAc-T1, -T2, and -T3. J Biol Chem, 272(38), 23503-23514. Yang, Z., Wang, S., Halim, A., Schulz, M. A., Frodin, M., Rahman, S. H., . . . Clausen, H. (2015). Engineered CHO cells for production of diverse, homogeneous glycoproteins. Nat Biotechnol, 33(8), 842-844. doi:10.1038/nbt.3280