key: cord-1012647-f536gs2w
authors: Hoffmann, Magnus A. G.; Kieffer, Collin; Bjorkman, Pamela J.
title: In vitro characterization of engineered red blood cells as potent viral traps against HIV-1 and SARS-CoV-2
date: 2020-12-21
journal: bioRxiv
DOI: 10.1101/2020.12.20.423607
sha: 173036971c4b335462b1c74aa66f3a18a56aba3a
doc_id: 1012647
cord_uid: f536gs2w

Engineered red blood cells (RBCs) expressing viral receptors could be used therapeutically as viral traps as RBCs lack nuclei and other organelles required for viral replication. Here we show that the combination of a powerful erythroid-specific expression system and transgene codon optimization yields high expression levels of the HIV-1 receptors CD4 and CCR5, as well as a CD4-glycophorin A (CD4-GpA) fusion protein on enucleated RBCs. Engineered RBCs expressing CD4 and CCR5 were efficiently infected by HIV-1, but CD4 or CD4-GpA expression in the absence of CCR5 was sufficient to potently neutralize HIV-1 in vitro. To facilitate continuous large-scale production of engineered RBCs, we generated erythroblast cell lines stably expressing CD4-GpA or ACE2-GpA fusion proteins, which produced potent RBC viral traps against HIV-1 and SARS-CoV-2. Our results suggest that this approach warrants further investigation as a potential treatment against viral infections.

Red blood cells (RBCs) exhibit unique properties that can be exploited for therapeutic 30 applications: they are the most abundant cell type, permeate all tissues, and have a lifespan of 31 120 days, making them attractive carriers for the delivery of therapeutic cargoes 1,2 . Moreover, 32

RBCs do not express class I major histocompatibility complex molecules 3 , thus therapeutic 33

RBCs from type O-negative blood could be universally administered to patients. 34

Engineered RBCs have been proposed as ideal candidates for the design of viral traps, as 35 they lack nuclei and other organelles required for viral replication 4-7 . Viruses could be lured 36 into attaching to and infecting RBCs that present viral receptors, thereby leading the virus to 37 a dead end and protecting viral target cells from infection. This approach has the potential to 38 prevent viral escape, as viruses must retain the ability to bind their receptors. However, 39 expression of viral receptors on RBCs is difficult to achieve since mature erythrocytes lack 40 the cellular machinery to synthesize proteins. Hence erythroid progenitor cells need to be 41 genetically-engineered to express the viral receptors and then be differentiated into 42 enucleated RBCs. During the erythroid differentiation process, transgene expression is 43 restricted through transcriptional silencing 8 , translational control mechanisms 9 and 44 degradation of proteins that are not normally present in RBCs 10 . 45

Here we show that the combination of a powerful erythroid-specific expression system and 46 transgene codon optimization yields high expression levels of the HIV-1 receptors CD4 and 47 CCR5 on enucleated RBCs to generate viral traps that potently inhibit HIV-1 infection in 48 vitro. We then applied these engineering strategies to generate erythroblast cell lines that can 49 continuously produce potent RBC viral traps against HIV-1 and SARS-CoV-2. 50

We used an in vitro differentiation protocol 11 to differentiate human CD34+ hematopoietic 53 stem cells (HSCs) into reticulocytes, an immature form of enucleated RBC that still contains 54 ribosomal RNA (Fig. 1a) . At the end of the proliferation phase, erythroid progenitor cells 55

were transduced using lentiviral vectors carrying CD4 or CCR5 transgenes by spinoculation 56 ( Fig. 1a; Supplementary Fig. 1a) . We also evaluated expression of a CD4-glycophorin A 57 (CD4-GpA) fusion protein that contained the extracellular CD4 D1D2 domains fused to the 58 N-terminus of GpA, an abundantly-expressed RBC protein. This strategy enabled expression 59 of single-domain antibodies in RBCs 11 , but is applicable only to CD4, a single-pass 60 transmembrane protein, but not to CCR5, a multi-pass transmembrane protein. Three days 61 post-transduction, transgene expression was evaluated by flow cytometry. Expression was 62 low for transgenes when the CMV promoter or alternative ubiquitous promoters were used 63 ( Fig. 1b; Supplementary Fig. 1b) . CD4-GpA expressed only marginally better than CD4, 64 suggesting that additional strategies are required to achieve robust expression of viral 65

To evaluate whether transcriptional silencing can be prevented by using an erythroid-specific 67 promoter, transgenes were subcloned into the CCL-βAS3-FB lentiviral vector 12 , which 68 contains regulatory elements that support the high expression levels of β-globin during 69 erythroid development (vectors β-CD4, β-CD4-GpA, and β-CCR5) ( Supplementary Fig. 1a) . 70

CD4 expression was greatly enhanced by this expression system, CCR5 expression increased 71 to a lesser extent, but CD4-GPA expression was not improved (Fig. 1b) . We hypothesized 72 that the limited availability of ribosomes and transfer RNAs restricted transgene expression 73 in differentiating erythroid cells and therefore codon-optimized transgene cDNA sequences 74 (β-CD4opt, β-CD4-GpAopt, and β-CCR5opt), resulting in greatly enhanced expression for all 75 transgenes (Fig. 1b) . 76

Genetically-engineered CD4+/CCR5+ erythroid progenitor cells differentiated efficiently 77 into enucleated RBCs (Fig. 1c) . After differentiation, almost all cells expressed GpA, of 78 which >80% did not stain for Hoechst nuclear dye, and May-Grunwald-Giemsa staining 79 confirmed that the majority of cells were enucleated RBCs (Fig. 1c,d) . Approximately 1/3 of shift of its emission spectrum from green (520 nm) to blue (447 nm) 13 . Whereas infectious 101 events were ≤0.3% in control RBCs and CD4-RBCs, 6.1% of CD4-CCR5-RBCs were 102 infected (Fig. 2b ). Since only 1/3 of these RBCs expressed both receptors (Fig. 1e) , this 103 corresponds to infection of almost 20% of CD4-CCR5-RBCs. Higher infection rates were 104 observed for RBCs that co-expressed CD4 and CXCR4 after incubation with a CXCR4-105 tropic HIV-1 HxBc2 pseudovirus ( Supplementary Fig. 3a,b) . However, RBCs that co-106 expressed the CD4-GpA fusion protein and CCR5 or CXCR4 were infected at lower 107 frequencies ( Fig. 2b; Supplementary Fig. 3b ). Addition of the CD4 D3D4 domains to CD4-108

GpA did not improve infection rates ( Supplementary Fig. 3c ). Unlike CD4, GpA does not 109 localize to lipid raft subdomains 14 , thus we speculate that low infection resulted from a lack 110 of co-localization between CD4-GpA and the CCR5 and CXCR4 co-receptors. 111

We assessed the therapeutic potential of RBC viral traps using a modified version of the 113

HIV-1 TZM-bl neutralization assay 15 (Fig. 2c) . After incubating RBCs with HIV-1YU2 114 pseudovirus, samples were centrifuged to remove RBCs and virions that attached to or 115 infected RBCs. Supernatants containing free virions that had not been captured by RBCs 116 were transferred to 96-well plates and TZM-bl cells were added to measure infectivity. CD4-117

GpA-RBCs neutralized HIV-1YU2 most potently at a half-maximal inhibitory concentration 118 (IC50) of 1.7x10 6 RBCs/mL (Fig. 2d ). This concentration is equivalent to 0.03% of the RBC 119 concentration of human blood (~5x10 9 RBCs/mL), suggesting that it is feasible to achieve 120 therapeutic concentrations in vivo. CD4-GpA-RBCs were ~4-fold more potent than CD4-121 RBCs, likely due to higher expression levels ( Fig. 2a; Supplementary Fig. 4) . Surprisingly, To generate a renewable source of RBC viral traps, we engineered the immortalized BEL-A 158 erythroblast cell line 17 to stably express high levels of CD4-GpA (Fig. 3a) . The CD4-GpA-159 BEL-A cells efficiently differentiated into enucleated RBCs, as >50% of CD71-expressing 160 cells did not stain for the nuclear marker DRAQ5 (Fig. 3b ). After differentiation, 161 CD71+/DRAQ5-RBCs were purified using fluorescence-activated cell sorting (FACS). The 162 majority of RBCs still expressed CD4-GpA (Fig. 3c) and potently neutralized HIV-1YU2 163 (IC50=2.1x10 7 RBCs/mL) (Fig. 3d) . 164

To evaluate if RBC viral traps could be effective against other viruses, we generated a BEL-165 A cell line that continuously produces RBC viral traps against SARS-CoV-2, the virus that 166 caused the ongoing COVID-19 pandemic 18 . BEL-A cells were transduced to stably express a 167 chimeric ACE2-GpA protein containing the extracellular domain of the SARS-CoV-2 168 receptor ACE2 18 fused to GpA (Fig. 3e) . Differentiation efficiency and transgene expression 169 on sorted CD71+/DRAQ5-RBCs was comparable to the CD4-GpA cell line (Fig. 3f,g) . 170

Importantly, lentivirus-based SARS-CoV-2 pseudovirus 19 was highly susceptible to ACE2-171

GpA-RBC neutralization (IC50=7x10 5 RBCs/mL) (Fig. 3h) suggesting that RBC viral traps 172 have the potential to become powerful anti-viral agents against a range of viruses. viruses. Since cell lines that generate RBC viral traps could be rapidly developed once a host 208 receptor has been identified, RBC viral traps could also become a rapid-response treatment 209 strategy for future viral outbreaks. Our results suggest that this approach warrants further 210 investigation as a potential treatment against viral infections. 211

Human cord blood or mobilized peripheral blood CD34+ HSCs (StemCell Technologies) 215

were differentiated into enucleated RBCs using a modified version of a previously-described 216 protocol 11 . Briefly, CD34+ HSCs were cultured in expansion medium (100 ng/mL rhFlt3, 217 100 ng/mL rhSCF, 20 ng/mL rhIL-6, 20 ng/mL rhIL-3, and 100 nM dexamethasone in 218 StemSpan II medium) at a density of 10 5 cells/mL for 4 days. Cells were then placed in 219 differentiation 1-2 medium (2% human AB plasma, 3% human AB serum, 3 U/mL heparin, 220 10 ng/mL rhSCF, 1 ng/mL rhIL-3, and 3 U/mL erythropoietin in StemSpan II medium) at a 221 density of 10 5 cells/mL for 3 days and at 2 x 10 5 cells/mL for an additional 3 days. The cells 222

were then passaged into differentiation 3 medium (2% human AB plasma, 3% human AB 223 serum, 3 U/mL heparin, 10 ng/mL rhSCF, and 1 U/mL erythropoietin in StemSpan II 224 medium) at a density of 2 x 10 5 cells/mL for 4 days. To induce RBC maturation, cells were 225 cultured in differentiation 4 medium (2% human AB plasma, 3% human AB serum, 3 U/mL 226 heparin, 0.1 U/mL erythropoietin, and 200 μg/mL holo-transferrin in StemSpan II medium) at 227 a density of 10 6 cells/mL for 4 days, and in differentiation 5 medium (2% human AB plasma, 228 3% human AB serum, 3 U/mL heparin, and 200 μg/mL holo-transferrin in StemSpan II 229 medium) at a density of 5 x 10 6 cells/mL for an additional 3 days. For morphological 230

Grünwald-Giemsa reagents (Sigma-Aldrich), and examined under an LSM800 laser scanning 232 confocal microscope (Zeiss). by transferring the cells into primary media (3% human AB serum, 2% FBS, 3 U/mL 332 heparin, 10 ng/mL rhSCF, 1 ng/mL rhIL-3, 3 U/mL erythropoietin, 200 µg/mL holo-transferrin, and 1 µg/mL doxycycline in StemSpan II medium) for 3-4 days at a density of 2 x 334 10 5 cells/mL. To induce RBC maturation, cells were moved into tertiary media (3% human 335 AB serum, 2% FBS, 3 U/mL heparin, 3 U/mL erythropoietin, 500 µg/mL holo-transferrin, 336 and 1 U/mL Pen-Strep in StemSpan II medium) for 4 days at a density of 1 x 10 6 cells/mL. 

Blood Cell Membrane 386 Processing for Biomedical Applications

Erythrocytes as Carriers 389 for Drug Delivery in Blood Transfusion and Beyond

Role of CD47 as a marker of self on red blood cells

Full-length CD4 electroinserted in the erythrocyte membrane as a 394 long-lived inhibitor of infection by human immunodeficiency virus

Functional expression of the CD4 protein after cross-397 linking to red blood cells with a bifunctional reagent

The erythrocyte viral trap: transgenic 400 expression of viral receptor on erythrocytes attenuates coxsackievirus B infection

Bone marrow stem cells to destroy circulating HIV: a hypothetical 403 therapeutic strategy

Extensive methylation of promoter sequences silences lentiviral 406 transgene expression during stem cell differentiation in vivo

Widespread and dynamic 409 translational control of red blood cell development

UBE2O remodels the proteome during terminal erythroid 412 differentiation

Genetically engineered red cells expressing single domain camelid 414 antibodies confer long-term protection against botulinum neurotoxin

beta-globin gene transfer to human bone marrow for sickle cell 417 disease

A sensitive and specific enzyme-based 419 assay detecting HIV-1 virion fusion in primary T lymphocytes

O-glycans direct selectin ligands to lipid rafts on leukocytes

Evaluating neutralizing antibodies against HIV, SIV, and SHIV in 424 luciferase reporter gene assays

Nanoparticles presenting clusters of CD4 expose a 427 universal vulnerability of HIV-1 by mimicking target cells

An immortalized adult human erythroid line facilitates 430 sustainable and scalable generation of functional red cells

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and 433 Is Blocked by a Clinically Proven Protease Inhibitor

Protocol and Reagents for Pseudotyping Lentiviral Particles 436 with SARS-CoV-2 Spike Protein for Neutralization Assays

Designed proteins induce the formation of nanocage-containing 440 extracellular vesicles

Foot-and-Mouth-Disease Virus 2a Oligopeptide Mediated 442

Cleavage of an Artificial Polyprotein