key: cord-0284192-wjgg174y
authors: White, Timothy M.; Stanfield, Brent A.; Bonavita, Cassandra M.; Rudd, Jared S.; Cardin, Rhonda D.
title: Development of a mouse salivary gland-derived mesenchymal cell line for immunological studies of murine cytomegalovirus
date: 2022-03-03
journal: bioRxiv
DOI: 10.1101/2022.03.03.482882
sha: a6ce2b43791cef701e6d0076d1d57450bdb5f4e4
doc_id: 284192
cord_uid: wjgg174y

The salivary glands are a crucial site of replication for human cytomegalovirus (HCMV) and its murine counterpart, murine cytomegalovirus (MCMV). Studies of MCMV often involve the use of BALB/c strain mice, but most in vitro assays are carried out in the NIH 3T3 cell line, which is derived from Swiss Albino mice. This report describes a BALB/c-derived mouse salivary gland cell line immortalized using the SV40 large T antigen. Cells stained positive for PDGFR1 and negative for E-cadherin and PECAM-1, indicating mesenchymal origin. This cell line, which has been named murine salivary gland mesenchymal (mSGM), shows promise as a tool for ex vivo immunological assays due to its MHC haplotype match with the BALB/c mouse strain. In addition, plaque assays using mSGM rather than NIH 3T3 cells are significantly more sensitive for detecting low concentrations of MCMV particles. Finally, it is demonstrated that mSGM cells express all 3 BALB/c MHC class I isotypes and are susceptible to T cell-mediated ex vivo cytotoxicity assays, leading to many possible uses in immunological studies of MCMV.

217 The m164-and IE1-derived immunodominant peptides are presented on H-2D d and H-2L d MHC 218 isotypes, respectively [23] . Our studies did not confirm precisely which epitopes were presented, 219 only that antigen-specific T cell-mediated cytotoxicity could be detected in response to both m164 220 and IE1 proteins. Our flow cytometry results indicate that all 3 isotypes are present on the cell 221 surface. Future investigation of MHC peptide complexes on transgene-expressing or virus-infected 222 mSGM cells could be used to confirm epitopes matching those seen in MCMV-infected BALB/c 223 mice. Transgene-expressing mSGM cells were readily produced using simple lentivirus 224 transduction, extending their possible uses. In particular, the m164 protein which was efficiently 225 expressed after lentivirus transduction is an important T cell target in MCMV-infected BALB/c 226 mice. A similarly transduced line may therefore be useful for CTL assays in the future.

In addition to the BALB/c strain-specific proteins that mSGM cells express, the mSGM 228 cell line was also found to be significantly more sensitive during plaque assays. Detection of low 229 numbers of viral particles in sera, tissue homogenates, or culture supernatant is subject to the limit 230 of detection of the assay. Quantitative PCR is a common substitute for plaque assays, with very 231 high sensitivity, but the presence of viral genome copies does not always correlate with infectious 232 viral particles. Using mSGM cells, it was found that infectious MCMV could be detected with 3-233 to 4-fold greater sensitivity compared to the same sample titrated on NIH 3T3 fibroblasts. 279 Cells were incubated for 3 hours at 37° C. Then, transfection solution was removed and replaced 280 with 3 ml fresh culture medium. After 24 hours, roughly 50% cell death had occurred and medium 281 was changed to remove debris. At 48 hours post-transfection, G418 selection agent was added to 282 the medium at a concentration of 100µg/ml. An un-transfected control well was used to ensure 283 that the G418 concentration was enough to induce cell death. After the initial selection phase, the 305 Medium was then removed and replaced with fresh complete medium and cultures were incubated 306 for 6 days. At 1-day intervals, wells were scraped using a cell scraper and cells and supernatant

PDGFR1, a mesenchymal marker (d)

single-and multi-step growth curves were conducted using wild-type MCMV 95 (KP strain). MCMV replication in mSGM cells was found to be nearly identical to growth in NIH 96 3T3 fibroblasts at both low and high multiplicity of infection (MOI) (Figs 2a-b). No significant 97 differences were found at any time point

After 40 passages, morphology had not changed, and plaque efficiency 103 remained comparable to low-passage mSGM cells. After 60 passages, however, the mSGM cells 104 were found to divide too rapidly for efficient plaque formation. Nevertheless, this represents a 105 substantial improvement over NIH 3T3 cells

Figure 2. mSGM cells are permissive to MCMV infection and replication. mSGM cells and 110 NIH 3T3 cells were infected with KP strain MCMV. (a) cells were infected at MOI of 2 for 111 single-step growth. (b) cells were infected at MOI of 0.01 for multi-step growth

To determine whether mSGM cells could be used for plaque assays, studies using identical 115 virus dilutions on NIH 3T3 cells or mSGM cells were performed, overlaid with identical medium, 116 and stained at 5 dpi. Both cell types grew well under overlay medium and plaques could be 117 visualized in both cell lines, although mSGM cells tended to produce larger plaques with more 118 clearly defined borders (Fig 3a-b). Efficiency of plaque formation was compared between the two 119 cell lines and mSGM cells were found to produce significantly greater numbers of plaques than 120 NIH 3T3 cells when the same MOI of virus was used (Fig 3c)

Centrifugal 122 enhancement, in which the 1-hour incubation at 37° is replaced with a 1-hour centrifuge step prior 123 to overlay

Figure 3. mSGM cells are suitable for highly sensitive MCMV plaque assay. (a-b) NIH 3T3 125 cells (a) and mSGM cells (b) were infected with 30 plaque forming units (pfu) (initially 126 determined on NIH 3T3 cells) per well of a 12-well plate. After 4 (left column) or 5 days (right 127 column) under overlay, plaques were stained with Giemsa

Images are at 40x magnification on an inverted microscope

Significance is indicated as *, p<0

One of the key drawbacks to using NIH 3T3 cells for studies of MCMV is the difference 135 in MHC haplotypes between Swiss Webster albino mice and other mouse strains, making NIH 136 3T3 cells unsuitable as antigen presenting cells for immunological studies using BALB/c mice

BALB/c 138 mice express 3 MHC class I isotypes: H-2D d , H-2K d , and H-2L d [18]. In order to determine the 139 expression levels of each isotype, mSGM cells were stained with antibodies against each of these 140 MHC isotypes and found to express all 3 isotypes at high levels on the cell surface (Fig 4a-c). This against MHC class I H-2D d (a)

Besides MHC class I molecule expression on the cell surface, processing and presentation 150 of viral protein-derived peptide epitopes is essential for T cell-mediated cytotoxicity. Therefore, 151 mSGM cells were tested as antigen presenting cells for MCMV-specific T cell cytotoxicity. T cells 152 were enriched from latently MCMV-infected or mock-infected BALB/c mice. An adenovirus 153 vector expressing the MCMV immediate-early protein 1 (IE1), a major CD8+ T cell epitope (Ad5-154 MCMV_IE1) [19], or a matched adenovirus control (Ad5-control) which expressed only an 155 irrelevant sequence, were used to introduce MCMV antigens. Uninfected mSGM cells were used 156 as a secondary control. Enriched T cells were added at an effector:target (E:T) ratio of 6:1 and 157 cytotoxicity was monitored using an Incucyte live-cell imaging system

Apoptosis was quantified using NucView 488, a non-toxic substrate 159 which fluoresces brightly after cleavage by caspase-3

T cell-mediated cytotoxicity was most

As expected, uninfected mSGM cells and Ad5-control infected mSGM cells, neither of 163 which produces MCMV-associated epitopes, did not induce T cell-mediated cytotoxicity (Fig 5a)

Some residual cytotoxicity was detected in the MCMV-naïve group and 167 may be explained by the expression of stress ligands by the mSGM cells, which activates T cell-168 mediated cytotoxicity, or by secretion of cytokines from T cells incidentally activated by the 169 purification process [22]. To control for differential cell growth rates, apoptosis is shown as a ratio 170 of individual fluorescent cells to percent confluence of the monolayer

Fluorescent events were gated by size to remove background fluorescence and any T cells which 174 underwent apoptosis (detailed in Materials and Methods section)

Figure 5. mSGM cells are vulnerable to ex vivo MCMV antigen-specific T cell mediated 176 cytotoxicity. mSGM cells were infected with Ad5-control (a) or Ad5-MCMV_IE1 (b)

splenic T cells from uninfected (white bar) or MCMV-infected (lined bar) mice at 28 days post 179 infection were added at E:T ratio of 6:1. Data represent 2 animals with 6 technical replicates 180 each

Retroviral vector transduction is a valuable tool for constructing stable transgeneduring MCMV latency

These results further demonstrate the broad potential applicability of before (a) and after transduction (b) with MCMV m164-FLAG lentivirus. (c) After 198 selection, cells were stained for FLAG epitope (upper panels) or isotype control (lower panels) to HCMV, and the tissue and cell tropism and replication

BALB/c mice are also inexpensive and easily 206 maintained in comparison with larger CMV host species, such as guinea pigs or rhesus macaques

commercially-available cell lines used 208 to produce MCMV virus stocks and for plaque assays are typically NIH 3T3 cells, a mouse 209 embryonic fibroblast cell line isolated from a Swiss albino mouse. NIH 3T3 cells therefore express 210 different cellular markers, including different MHC haplotypes. This precludes their use in some 211 common ex vivo assays such as T cell-or NK cell-mediated cytotoxicity assays

High expression of all 3 MHC class I isotypes on mSGM cells allows presentation of 307 were collected and frozen. Collected samples were thawed and sonicated, and titer was determined to 6-week-old female BALB/c ByJ mice were obtained from Jackson Laboratory

USA) and maintained under specific pathogen free conditions at the Louisiana

Cells 319 were then washed twice with phosphate buffered saline containing 2% FBS, passed through a 70-320 µm cell strainer, and fixed with 2% formaldehyde

Results were analyzed with FlowJo software

Tullagreen, Carrigtwohill, Ireland), washed, fixed with 2% 328 formaldehyde, and washed again in PBS. Primary antibodies were added at 1:100 dilutions and 329 slides were incubated at 4° on a rocker overnight. Cells were then washed thoroughly with PBS 330 and secondary antibodies were added at a 1:100 dilution for 1 hour. After washing with PBS, slides 331 were cover-slipped using ProLong Diamond Antifade Mountant with DAPI (Invitrogen, Eugene, 332 OR) and glass coverslips

The secondary antibody used was chicken anti-rat IgG AF488

Whole mouse spleens were homogenized in 3 ml complete medium and filtered through 337 70-micron cell strainers. Nucleated cells were separated from red blood cells and cell debris using

Cedarlane Laboratories Ltd., Canada) according to manufacturer's instructions

Germany) according to manufacturer's 342 instructions, using an input of 10 7 cells. Enrichment was confirmed to be >95% efficient by flow 343 cytometry. T cells were resuspended in RPMI 1640 (Gibco) with 10% FBS, 1x 344 penicillin/streptomycin solution (Gibco), and beta-mercaptoethanol

After allowing 12-24 hours for cells to adhere, culture medium was 348 replaced with RPMI containing 10% FBS and 1x penicillin/streptomycin solution (Gibco). MACS-349 enriched T cells were then added at the indicated effector:target ratio

Incucyte system set to scan 2 fields from each well of a 96-well cell culture plate at 1-hour 353 intervals. In an initial study, effector:target ratios were tested from 1:1 to 10:1. The ratio 6:1 was 354 found to provide optimal T cell-mediated killing with minimal background fluorescence during 355 the 48-hour test period

Cloning was performed using the FastCloning method described by Li et al

MCMV IE1 gene was cloned into the plasmid pAd5-Blue [15] to make pAd5-MCMV_IE1. First, 360 PCR amplification of the IE1 gene was conducted using 10ng of purified MCMV (KP) DNA

Primers ClaI_MCMV_IE1_F 5'-AGAGAGATCGATCACCATGGAGCCCGCCGCAC-3' and 362 SpeI_FLAG-MCMV_IE1_R

1 μg pAd5-Blue was digested 369 with ClaI and XbaI and dephosphorylated using Quick CIP

Enzymes were heat inactivated by heating to 80 °C for 5 minutes and DNA precipitated 371 using 100% ethanol. Digested pAd5-Blue and IE1 PCR products were then ligated using T4 372 DNA ligase

Plasmids were then transformed into XL-10 Gold chemically competent Escherichia coli cells 374 (Agilent Technologies

For lentivirus plasmid generation, first the MCMV m164 gene was PCR amplified from

For control lentivirus, insert was amplified using primers 382 pLV_CDS_FC_F 5'-ACCCAGCTTTCTTGTACAAAGTGGTG-3' and pLV_CDS_StartFC_R 383 5'-GGTGGCAGCCTGCTTTTTTGTAC-3'. The ligated products were then

Adenovirus containing the MCMV IE1 gene was generated by transfecting purified

Once plaques were visualized, the plate 390 was freeze/thawed 3 times and cellular debris pelleted by centrifugation at 1000g for 5 minutes. 391 100μL of supernatant was then used to infect a 70% confluent T-75 flask of 293A cells and 392 infection allowed to progress for 5 days. Following infection, cells were freeze thawed 3 times and 393 cellular debris pelleted by centrifugation at 1000 x g for 5 minutes. The resulting supernatant was 394 then titrated on 293A cells by limiting dilution and frozen at -80 °C until use

Lentivirus containing the MCMV m164 gene was produced by co-transfection of pLV-396 m164-FLAG with psPAX2 (a gift from Didier Trono

Addgene plasmid #8454) [16] mixed at a 1:1:1 molar ratio using

Supernatant was collected 399 after 4 days and frozen at -80° C until use. For transduction of mSGM cells, lentivirus was added 400 directly to culture medium of sub-confluent monolayers of mSGM cells in 12-well tissue culture 401 plates. After 48 hours, cells were selected under 1 ng/mL puromycin

For most figures, data 406 were tested for normality and ANOVA with Tukey's post-hoc test was used. Graphs were 407 produced using Prism 9

Live cell images were analyzed with Incucyte software with parameters set to exclude 409 fluorescent cells smaller than 300 µm 2 , which allowed us to exclude any apoptotic leukocytes 410 (which are typically 9-11 µm in diameter, or under 100 µm 2 ) while including mSGM cells

TW designed and carried out experiments and wrote manuscript. BS designed and assisted 415 with experiments, produced lentivirus and adenovirus, and edited manuscript. CB assisted with 416 experiments and edited manuscript. JR provided SV40 plasmid and carried out transfection

Acknowledgments This work is supported by NSTP fellowship grant number AP19VSNL00C037 from the

Structure and Function of the 425

Murine Cytomegalovirus sggl Gene: a Determinant of Viral Growth in Salivary Gland 426 Acinar Cells

Salivary glands are a target for SARS-CoV-2: a source for saliva contamination

Absence of 431 Cross-Presenting Cells in the Salivary Gland and Viral Immune Evasion Confine 432

Cytomegalovirus Immune Control to Effector CD4 T Cells

Latent murine cytomegalovirus DNA in splenic 435 stromal cells of mice

Stromal Cells as a Potential Reservoir of Human Cytomegalovirus

Murine Sarcoma and Leukemia Viruses: Assay 440

Using Clonal Lines of Contact-Inhibited Mouse Cells

The M33 Chemokine

Receptor Homolog of Murine Cytomegalovirus Exhibits a Differential Tissue-Specific Role 444 during In Vivo Replication and Latency

Peripheral 447 blood mononuclear phagocytes mediate dissemination of murine cytomegalovirus

Latent Cytomegalovirus 450 Infection of BALB/c Mouse Spleens Detected by an Explant Culture Technique

Heart Dysfunction Following Long-Term 453

Murine Cytomegalovirus Infection: Fibrosis, Hypertrophy, and Tachycardia. Viral 454 Immunology

Murine cytomegalovirus 456 binds reversibly to mouse embryo fibroblasts: implications for quantitation and explanation 457 of centrifugal enhancement

Characterization of murine 459 cytomegalovirus infection and induction of calcification in Murine Aortic Vascular Smooth Muscle Cells (MOVAS)

Murine Cytomegalovirus IE2, an 463 Activator of Gene Expression, Is Dispensable for Growth and Latency in Mice

FastCloning: a highly simplified, 466 purification-free, sequence-and ligation-independent PCR cloning method

Direct Ligation System for Engineering

Lentivirus-472 delivered stable gene silencing by RNAi in primary cells

Platelet-475 derived Growth Factor Receptor Regulates Salivary Gland Morphogenesis via Fibroblast 476

Growth Factor Expression

Structural relation of murine "third locus" (H-2L) major 479 histocompatibility antigens to the products of H-2K and H-2D loci

A pentapeptide as minimal antigenic 481 determinant for MHC class I-restricted T lymphocytes

Src/lck 484 inhibitor dasatinib reversibly switches off cytokine release and T cell cytotoxicity following 485 stimulation with T cell bispecific antibodies

In vivo imaging of 488 early stage apoptosis by measuring real-time caspase-3/7 activation

Activation of NK Cells and 491 T Cells by NKG2D, a Receptor for Stress-Inducible MICA

Two Antigenic Peptides from Genes 493 m123 and m164 of Murine Cytomegalovirus Quantitatively Dominate CD8 T-Cell Memory 494 in the H-2d Haplotype

Fields Virology