Tetramer Immunization and Selection Followed by CELLISA Screening to Generate Monoclonal Antibodies against the Mouse Cytomegalovirus m12 Immunoevasin

The use of mAb has been fundamental in advancing our understanding of basic immunology and has revolutionized the fields of applied biotechnology, diagnostic medicine, and clinical therapeutics. mAb have proven to be effective agents to treat cancer, in which they elicit direct cytotoxic effects by guiding immune cells to more effectively recognize cells expressing tumor Ags or block inhibitory signals on immune cells, rendering them more active at eliciting effector functions via checkpoint inhibition (1–3). Broadly neutralizing Ab can also protect against pathogens and have been described for HIV, Ebola, dengue, and Zika virus (4).

The generation of mAb can often be a challenge attributable not only to immunization protocols that require large amounts of Ag but also due to labor-intensive procedures often involving screening of thousands of culture wells to identify suitable hybridomas. These approaches are often not only laborious but are also associated with high screening costs. We have previously described an efficient method to generate mAb using a reporter cell–based system (5, 6). Briefly, by constructing a chimeric Ag fused to the intracellular portion of the CD3ζ signaling domain, reporter cells can be generated that are capable of detecting and reporting Ab-mediated cross-linking; this eases the screening procedure to a low-cost, high-throughput assay whereby the reporter cells generate their own enzyme (5, 6). Initially, immunization protocols used the reporter cells as prime-boost immunogens themselves; however, in this study, we demonstrate in a proof-of-principle approach that it is also possible to effectively 1) immunize host animals and 2) selectively enrich Ag-specific B cells prior to hybridoma fusion using tetrameric forms of the Ag as both the immunogen and an antigenic bait for enrichment, followed by subsequent validation screening.

We recently reported that mouse CMV (MCMV) encodes a viral immunoevasin that targets and antagonizes the NKR-P1 family of NK cell receptors (7, 8). Considering this, we sought to generate mAb specific for this viral decoy ligand, an NKR-P1B immunoevasin, to further characterize functional interactions. In this report, we describe an efficient approach for the immunization, isolation, and screening of hybridomas producing mAb of interest using tetramers of the viral m12 protein. Specifically, we used the MCMV-encoded m12 protein in tetrameric form as the prime-boost immunogen, leveraged the ability of m12 tetramers to enrich for Ag-specific B cells prior to hybridoma fusion, and subsequently validated and characterized several novel mAb specific for m12. Importantly, we last show that these mAb may have passive immunotherapeutic value against CMV infection.

C57BL/6 (B6) mice were purchased from The Jackson Laboratory. Animals were housed and maintained according to approved animal protocols at the University of Toronto and Sunnybrook Research Institute.

YAC-1 cells were obtained from the American Type and Culture Collection. BWZ.36 cells were provided by Dr. N. Shastri (University of California, Berkeley). HEK293T and P3XAg8.653.1 (P3) cells were obtained from Dr. D. H. Raulet (University of California, Berkeley). Cells were cultured in complete high-glucose DMEM (DMEM-HG) or RPMI 1640 supplemented with 2 mM glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, 50 μg/ml gentamicin, 110 μg/ml sodium pyruvate, 50 μM 2-ME, 10 mM HEPES, and 10–20% FCS. Splenic lymphokine-activated killer (LAK) cells were cultured in complete RPMI 1640 supplemented with 1000 U/ml recombinant human IL-2 (rhIL-2; Teceleukin).

For transfections, HEK293T cells were plated 1 d prior to transfection in six-well plates (4 × 10⁵ per well). Transfections were performed using Lipofectamine 2000 according to the manufacturer’s protocol (Thermo Fisher Scientific). Cells were analyzed 48 h posttransfection using flow cytometry. All mammalian expression vectors used in this study have been previously described (7, 9).

Tetrameric forms of m12^Smith were generated as previously described (7, 10, 11). Briefly, DNA encoding m12 from the MCMV^Smith strain (residues 29–208) was ligated into the pFastbac vector (Thermo Fisher Scientific) upstream of 6× histidine and BirA tags and used to generate recombinant baculovirus according to the manufacturer’s instructions. Recombinant baculovirus was produced by transfection into Sf9 cells, and the supernatant was used to infect Hi5 insect cells. Protein was harvested 48–72 h later from cell media and was then purified using nickel affinity and size-exclusion chromatography using HisTrap and Superdex 200 columns 16/60 (GE Healthcare), respectively. For biotinylation, purified m12 was buffer exchanged into 10 mM Tris and enzymatically biotinylated overnight at 20°C prior to removal of free biotin via size-exclusion chromatography. Tetramers were generated by incubating biotinylated m12^Smith monomeric protein with streptavidin (SA)–PE or SA–allophycocyanin in a 4:1 M ratio, aliquoted in 1/10 volumes over a period of 3 h at 4°C.

On day 0, 6–8-wk-old B6 mice were immunized with m12^Smith–allophycocyanin tetramer (25 μg) emulsified in CFA by i.p. injection. MCMV m12^Smith tetramers have previously been described (7). Twenty-eight days later, the mice were boosted using m12^Smith–allophycocyanin tetramer (50 μg) emulsified in IFA. On day 42, the mice received one final boost of purified m12^Smith–biotin protein (25 μg, no SA–allophycocyanin) in IFA, then sacrificed 3 d later. On the day of sacrifice, cardiac puncture blood was collected for purification of polyclonal antisera. Splenocytes were harvested, exposed to 1× ammonium–chloride–potassium lysis buffer for 5 min on ice, thoroughly washed with ice-cold PBS, then stained with m12^Smith–PE tetramers (no SA–allophycocyanin) before enrichment using anti-PE magnetic selection (STEMCELL Technologies), as previously described (12). These enriched m12^Smith–PE⁺ splenocytes were fused with P3X63.Ag8.653.1 (P3) hybridoma fusion partners at a 2:1 ratio using polyethylene glycol 1500 (MilliporeSigma) according to an established protocol (13). Hybridomas were resuspended in 20% DMEM-HG and aliquoted into a 96-well plate (one plate per fusion). The following day, selection using 20% DMEM-HG supplemented with 1× hypoxanthine–aminopterin–thymidine (HAT) was initiated. After 7 d, the medium was switched to 20% DMEM-HG with 1× hypoxanthine–thymidine (HT), and 3–5 d later, the cells were screened for mAb of interest.

Screening of hybridomas was initially accomplished by using Ig-capture plate-bound stimulations of CELLISA arrays using BWZ.CD3ζ–m12^Smith reporter cells, as described below. Once candidate hybridomas were identified and subcloned, mAb were further validated and characterized using several approaches described below.

Cells were stained with primary mAb supernatants (100 μl of hybridoma supernatant) in FACS buffer (HBSS, 0.5% BSA, and 0.03% NaN₃) on ice for 30 min, washed, incubated with secondary Ab for another 30 min, then analyzed using an FACSCanto II or LSR II (BD Biosciences). Cells were gated by forward and side light scatter properties and propidium iodide or DAPI exclusion for viability. Data were analyzed using FlowJo software (FlowJo). Alexa Fluor 647–conjugated goat anti-mouse IgG, IgG₁, IgG_2a, IgG_2b, IgG_2c, IgG₃, IgM, and IgG + IgM secondary reagents were purchased from Jackson ImmunoResearch Laboratories. The anti-mouse NKR-P1B^B6 mAb (2D12) was a kind gift from Drs. K. Iizuka and W. M. Yokoyama (14). Anti-mouse NK1.1/NKR-P1C^B6 (PK136), NKp46 (29A1.4), and CD3ε (145-2C11) mAb were purchased from BioLegend.

BWZ.CD3ζ–NKR-P1B^B6, BWZ.CD3ζ–NKR-P1B¹²⁹, BWZ.CD3ζ–NKR-P1B^FVB, BWZ.CD3ζ–NKR-P1C^B6, and BWZ.CD3ζ–m12^Smith reporter cells have previously been described (7, 15). Briefly, the ectodomains of the NKR-P1B receptor or m12^Smith were fused with intracellular mouse CD3ζ signaling domains in the murine stem cell virus vector expressing type II and type I fusion cassettes, respectively, and used to transduce BWZ.36 cells, as previously described (6). Transduced cells were then sorted for GFP⁺ expression using FACS.

For plate-bound stimulations (CELLISA assays), anti-mouse IgG + IgM capture plates were prepared by incubating high-binding chemistry, flat-bottom, 96-well plates (Corning) with polyclonal goat anti-mouse IgG + IgM Ab (Jackson ImmunoResearch Laboratories) at 10 μg/ml in 75 μl PBS per well, then incubated overnight at 37°C in a tissue culture incubator. Plates were washed three times with 200 μl of PBS, followed by Ab capture via addition of 100 μl of hybridoma supernatant directly to the wells in 96-well arrays and incubated overnight. Plates were triple washed with 200 μl of PBS, then reporter cells (BWZ.CD3ζ–m12^Smith, 5 × 10⁴ in 200 μl medium/well) were added and incubated overnight. The following morning, cells were pelleted, washed with PBS, and then resuspended in 150 μl of 1× chlorophenol-red-β-d-galactopyranoside buffer (90 mg/l CPRG [MilliporeSigma], 9 mM MgCl₂, and 0.1% NP-40 in PBS). Plates were incubated at room temperature, then analyzed using a Varioskan microplate reader (Thermo Fisher Scientific) with a signal–background subtraction set at OD signal 595–655-nm background (∆OD_595–655).

For cell-based stimulation assays, 5 × 10⁴ BWZ reporter cells were cocultured with 5 × 10⁴ stimulator cells overnight in a tissue culture incubator. Hybridoma supernatants (100 μl) were preincubated with stimulator cells for 1 h at 37°C to test blocking potential. Plates were developed as described above. As positive controls, reporter cells were stimulated with 10 ng/ml PMA and 0.5 μM ionomycin in complete DMEM-HG.

mAb were isotyped using a mouse Ig isotyping ELISA kit following the manufacturer’s protocol (BD Biosciences). Because the ELISA kit detects BALB/c strain Ab isotypes, we confirmed Ab cross-reactivity using secondary anti-mouse IgM and IgG reagents by flow cytometry.

Monoclonal IgG Ab were purified from hybridoma supernatants using Capturem Protein G Maxiprep Columns according to manufacturer’s protocols (Takara Bio). mAb were concentrated using Amicon Ultra columns (MilliporeSigma).

NK-mediated cytotoxicity was measured using a flow cytometry–based assay. Briefly, NK-LAK cells were generated by processing spleens into single-cell suspensions then culturing them in complete RPMI 1640 supplemented with 1000 U/ml rhIL-2 (Teceleukin; National Cancer Institute). On day 7, cells were harvested and used as effectors in cytotoxicity assays. YAC-1 targets expressing m12^Smith, m12^MW97, or control vector were labeled with CellTrace Violet (CTV; Thermo Fisher Scientific), and then incubated with NK-LAK effectors at different E:T ratios in the presence or absence of purified Ab. Cells were cocultured for 4 h at 37°C and then resuspended in FACS buffer containing propidium iodide. For control wells, bleach or media were used to determine maximal and spontaneous cellular death in the assay, respectively. Percentage of specific lysis was calculated by determining the percentage of live CTV⁺ targets in each well relative to control cells as follows:

Data were analyzed using Prism 7 (GraphPad) employing one-way ANOVA analysis (see figure legends). All graphs show mean ± SEM (*p < 0.05, **p < 0.01, and ***p < 0.001). All data are representative of at least three independent experiments.

We have recently described how MCMV targets the NKR-P1B/Clr-b missing-self NK cell recognition axis, including the evolution of a virally encoded decoy immunoevasin, m12, which inhibits NK cell responses via NKR-P1B (7–9, 16, 17). Importantly, we also observed that the activating NKR-P1A and NKR-P1C paralogs could also bind this ligand (7). To further characterize m12 function, we set out to generate m12-specific mAb. Using our previously generated m12^Smith tetramers (Fig. 1A) (7), we initiated an immunization protocol using these tetramers as the prime-boost immunogen emulsified in CFA adjuvant (Fig. 1B). Briefly, B6 mice were immunized with 25 μg of m12^Smith–allophycocyanin tetramer in CFA, boosted on day 28 using 50 μg of m12^Smith–allophycocyanin tetramer emulsified in IFA, then finally boosted on day 42 using 25 μg of purified m12^Smith–biotin monomers (to avoid stimulating SA–allophycocyanin carrier-specific B cells) in IFA. Mice were sacrificed 3 d later, spleens were harvested, and m12^Smith–PE–tetramer⁺ splenocytes (to avoid allophycocyanin-reactive B cells) were isolated using an anti-PE⁺ enrichment kit, akin to a previously described method (12). As seen in Fig. 1C, the anti-PE enrichment using m12^Smith-PE tetramers resulted in 20–30% of the splenocytes being NKp46^–m12^Smith–tetramer⁺, corresponding to roughly a 10-fold enrichment of m12-reactive B cells. Interestingly, this approach also allowed coenrichment of m12^Smith-tetramer⁺ NK cells, which largely represents the NKR-P1B⁺ subset (NKp46⁺m12–tetramer⁺; Fig. 1C). These cells were then fused with P3 fusion partners, selected using HAT followed by HT media and then tested for production of mAb of interest.

First, to confirm the successful generation of anti-m12 Ab, we tested the polyclonal antisera from immunized mice. Importantly, only the antisera from animals immunized with m12^Smith reagents were capable of staining BWZ.CD3ζ–m12^Smith reporter cells, but not parental BWZ^– cells, using flow cytometry (Fig. 2A), thus validating our immunization approach. We then proceeded to isolate hybridomas for m12-specific mAb.

To test the efficiency of generating Ab, we initially screened hybridomas using our CELLISA approach (5, 6). This high-throughput screen was accomplished using BWZ.CD3ζ–m12^Smith reporter cells (BWZ.ζ–m12^Smith) in Ab-capture, plate-bound stimulation arrays. These reporter cells display the m12^Smith extracellular domain fused with the intracellular signaling domain of CD3ζ and express the LacZ gene (β-galactosidase enzyme) upon downstream NFAT activation, thus facilitating measurements of receptor ligation using an inexpensive colorimetric assay whereby the responding cells make their own enzyme (6). Briefly, high-binding chemistry plates were initially coated with anti-mouse IgG + IgM capture Ab, thoroughly washed, then incubated with target hybridoma supernatants. Wells were then washed, BWZ.ζ–m12^Smith reporter cells were incubated on plates overnight, and assays were developed by adding CPRG solution containing substrate and lysis buffer following the removal of media and washing using PBS. As shown in Fig. 2B, this approach yielded a significant frequency and number of hybridomas capable of stimulating BWZ.ζ–m12^Smith reporter cells. In this initial screen, the first and second host mice gave rise to 5 and 46 anti-m12 mAb⁺ clones, respectively. We then froze down replicates of these initial plates while expanding 50 of the strongest responding wells for further validation and characterization.

We next validated and characterized the expanded hybridoma clones. Of the 50 chosen candidates, only 40 grew as viable long-term stable cultures, and of these, only 18 consistently produced m12-reactive mAb once stably established. These hybridoma clones were then characterized in several biological assays. First, we confirmed that their mAb were capable of stimulating BWZ.ζ–m12^Smith reporters in plate-bound assays (Fig. 3A). Second, we tested their ability to stain BWZ.ζ–m12^Smith cells. As shown in Fig. 3B, all candidate m12 mAb stained m12^Smith-bearing reporter cells, but not parental BWZ^– cells. Using anti-mouse IgG– and anti-mouse IgM–specific secondary reagents, we observed that most of the mAb were IgG isotype, whereas two clones produced IgM isotypes (data not shown). This was subsequently confirmed using a mouse mAb isotyping kit (Table I).

We investigated which mAb were capable of blocking the m12^Smith/NKR-P1B interaction in a cellular context. To this end, we used our previously generated BWZ target cells transduced with constructs expressing two different m12 alleles (BW.m12^Smith and BW.m12^MW97), which differ in a single amino acid (E91K). To confirm that these cells were surface m12⁺, we used hybridoma supernatants and observed that ∼5–10% of the BW.m12^Smith and ∼0.5–1% of the BW.m12^MW97 were m12⁺, respectively. Therefore, we sorted these cells for m12 expression using a selected 2C12 hybridoma (Supplemental Fig. 1). Interestingly, the BW.m12^Smith allele consistently had higher 2C12 anti-m12 expression versus the BW.m12^MW97 allele (observed with all hybridoma stains; data not shown). Next, we performed reporter cell assays using CD3ζ–NKR-P1B fusion receptor–bearing reporter cells (BWZ.ζ–NKR-P1B) with BW.m12 targets as stimulator cells. We preincubated hybridoma supernatants with these stimulator cells for 1 h prior to coculture with BWZ.ζ–NKR-P1B^B6, BWZ.ζ–NKR-P1B¹²⁹, and BWZ.ζ–NKR-P1B^FVB reporter cells. These alleles cover a range from low to medium to high affinities for m12^Smith (B6 < 129 < FVB) (7). In this study, we observed that 2C12 was capable of blocking interactions between m12^Smith and NKR-P1B^B6, NKR-P1B¹²⁹, and NKR-P1B^FVB; notably, 2D1 partially blocked B6 and 129 alleles, whereas 1E3 and 1F7 were only capable of partially blocking NKR-P1B (Fig. 3C).

We next determined if anti-m12 mAb could recognize other allelic variants of m12. In this study, we transfected HEK293T cells with constructs expressing m12^Smith, m12^MW97, m12^C4A, or m12^G4 and tested the ability of our mAb to detect these alleles at the cell surface. Interestingly, we observed that all mAb were capable of recognizing m12^Smith, m12^MW97, and m12^C4A, but not m12^G4. This was not surprising given that m12^G4 is highly polymorphic relative to m12^Smith (26-aa differences) (7). Worth noting, m12^G4 has 3-aa differences in residues that contact NKR-P1B^B6, whereas m12^MW97 and m12^C4A do not have any differences in contact residues (7). Also, consistent with our observations using BW.m12 targets, the m12^MW97 transfectants displayed lower staining levels compared with their m12^Smith counterparts (Fig. 4A; Supplemental Fig. 1). We next determined if any mAb could cross-react with other m02 or m145 family MCMV immunoevasins by transfecting the various family members into HEK293T cells then staining them using m12 mAb supernatants. In this study, we confirmed that m12 mAb were uniquely specific for the m12^Smith among all other m02 and m145 family members (Fig. 4B). Because some of these initial parental hybridomas were very likely to be oligoclonal, we further selected for monoclonal hybridoma subclones by limiting dilution. In this study, we obtained 19 mAb that retained their anti-m12 specificity using CELLISA assays (Table II; data not shown). We then reassessed these mAb in their ability to block NKR-P1B/m12 interactions in BWZ reporter cell assays. To further evaluate these mAb, we also performed assays with either m12^Smith- or m12^MW97-expressing stimulator cells. Importantly, m12^MW97 only interacts with the 129 and FVB, but not B6 NKR-P1B alleles, whereas m12^Smith interacts with all with varying affinities. Consistent with previous findings (Fig. 3C), 1E3, 1F7, and 2D1 partially blocked NKR-P1B^B6 interactions, whereas these mAb were less effective at blocking binding to NKR-P1B^FVB (Figs. 3C and 5A). All 2C12 subclones were very effective in blocking engagement of m12^Smith to all NKR-P1B alleles, whereas 2D1 only completed blocked NKR-P1B¹²⁹ (Fig. 5A). Importantly, 2C12 also blocked BW.m12^MW97 interactions with BWZ.ζ–NKR-P1B¹²⁹ and BWZ.ζ–NKR-P1B^FVB (Fig. 5B). Interestingly, in BW.m12^MW97:BWZ.ζ–NKR-P1B¹²⁹ cocultures, we observed some mAb actually increased reporter cell stimulation, which may be the result of mAb stabilizing this m12 allele at the cell surface (Fig. 5B). Consistent with our previous findings, cocultures of the immunogen variant BW.m12^Smith:BWZ.ζ–NKR-P1C^B6,129,FVB alleles did not yield significant signals (7) (data not shown). Collectively, these data demonstrate that m12 reactive mAb hybridoma supernatants could recognize the original m12^Smith prime-boost immunogen, all subclones cross-reacted with the m12^MW97, and m12^C4A, but not m12^G4 alleles, and some subclones were capable of blocking m12 interactions with multiple NKR-P1B receptor alleles.

To determine if some of these Ab were capable of influencing target recognition by NK cells, cytotoxicity assays were performed. For these experiments, we selected one mAb for each IgG isotype (2A3 = IgG_2b, 2C12 = IgG₁, and 2H4 = IgG_2c) and purified the mAb from hybridoma supernatants. Splenic NK-LAK cells (Fig. 6A) from B6 mice were cultured and used as effectors against NK cell–sensitive YAC-1 targets expressing m12 alleles (Fig. 6B). As expected, there was no difference in killing of YAC-1 controls (YAC.Vector) when incubated with any of the mAb. Similar results were observed with m12^MW97-expressing targets because m12^MW97 is poorly engaged by the NKR-P1B^B6 allele (Fig. 6B) (7). In contrast, in cocultures with YAC-1–expressing m12^Smith, there was a slight increase in killing with 2A3 and 2H4 (possibly because of FcγRIII-mediated, Ab-dependent cellular cytotoxicity [ADCC]); however, 2C12 yielded efficient killing of these targets most likely because of its superior ability to block m12^Smith engagement with NKR-P1B. Collectively, these results demonstrate that the 2C12 is strikingly effective at blocking m12-mediated NKR-P1B–dependent decoy immunoevasion (in multiple cellular contexts), thereby restoring NK cell function and perhaps additionally enhancing cytotoxicity via ADCC.

In this study, we describe a novel sensitive high-throughput protocol for the efficient generation of Ag-specific hybridomas and mAb. We used tetrameric forms of the candidate protein of interest for both prime-boost immunization and selective enrichment of candidate-responding B cells to generate hybridomas and Ag-specific mAb along with a cost-effective, high-throughput assay for the screening and validation of mAb specificities (CELLISA). We demonstrate that this method enhances immunization and eases screening while greatly diminishing the effort required to fuse and screen hybridomas; CELLISA also facilitates a rapid and sensitive high-throughput assay for identification of positive mAb-secreting clones to cell surface Ags.

The utility of this assay is supported by the following theoretical and real advancements: 1) tetramers of the protein of interest facilitate the activation of both low-affinity and high-affinity B cells during prime-boost immunization and should also permit greatly reduced amounts of protein immunogen upon further experimentation; 2) the final boost using purified tetramers free of SA–allophycocyanin carrier facilitates the selective reactivation of Ag-specific B cells prior to fusion, minimizing cross-reactive, carrier-specific B cells; 3) magnetic (or perhaps enhanced FACS based) SA–PE⁺ tetramer enrichment of only Ag-reactive B cells prior to hybridoma fusion greatly minimizes the number of fusion partner cells required (logarithmically lower numbers), as well as the plating and screening efforts required to generate Ag-specific hybridomas while at the same time minimizing or eliminating undesired or allophycocyanin-reactive clones; 4) the BWZ.CD3ζ–fusion reporter cells bearing the Ag of interest facilitate rapid, efficient, and high-throughput screening of Ag-specific hybridomas in a single day using an inexpensive colorimetric assay because they produce their own enzyme. Using this approach, our fusions yielded one plate per spleen rather than the 30 plates per spleen, as is commonly used in the absence of enrichment, resulting in an economy of labor, materials, and time. The use of reporter cells, which drive potent β-galactosidase enzyme production in response to Ag–CD3ζ chimeric receptor stimulation, in plate-bound stimulation colorimetric assays greatly enhances the sensitivity and ease of screening (5, 6). In addition, our technique facilitates the detection of Ab that engage an Ag in its native state. Importantly, although in this study we used anti-mouse IgG and anti-mouse IgM Ab for capturing hybridoma-produced mAb, this could easily be adapted for targeting specific Ab isotypes, such as IgA or IgE. Therefore, the combined use of reporter cells, tetrameric proteins for both immunization and enrichment and hybridoma supernatants allow for a rapid, efficient, and specific validation of several Ab parameters in a short time period, including Ag expression levels, affinity and avidity measurements (6), comparative allelic reactivities, functional blocking assays, cross-reactivities with antigenic variants, mAb isotype variants, and theoretically complement activation, Fc receptor binding, ADCC function, etc. Although prior studies have used FACS to enrich Ag-specific B cells for hybridoma production (18), our method combines this approach with tetrameric Ag immunization and a sensitive functional reporter screening technique. We are now testing whether this protocol will allow for similar generation and screening of mAb specific for intracellular protein targets in addition to cell surface proteins.

Importantly, in light of the global COVID-19 pandemic, this approach may be particularly useful as a basis for the rapid generation of novel neutralizing or diagnostic mAb to emerging pathogens, such as the SARS-CoV2 coronavirus, the causative agent behind COVID-19. We are currently pursuing this endeavor.

The authors have no financial conflicts of interest.