key: cord-103306-1wc3f1rl
authors: Sengupta, Sourodip; Addya, Sankar; Biswas, Diptomit; Sarma, Jayasri Das
title: Matrix metalloproteinases and tissue inhibitors of metalloproteinases in murine coronavirus-induced neuroinflammation
date: 2020-09-18
journal: bioRxiv
DOI: 10.1101/2020.09.17.302877
sha: 
doc_id: 103306
cord_uid: 1wc3f1rl

Mouse hepatitis virus (MHV) belongs to the same beta-coronavirus family as SARS-CoV-2, MERS-CoV, and SARS-CoV. Studies have shown the requirement of host cellular proteases for priming the surface spike protein during viral entry and transmission in coronaviruses. The metzincin family of metal-dependent endopeptidases called matrix metalloproteinases (MMPs) is involved in virus encephalitis, enhanced blood-brain barrier permeability, or cell-to-cell fusion upon viral infection. Here we show the role of MMPs as mediators of virus-induced host neuroinflammatory response in the MHV model. Infection of mice with wild-type MHV-A59 or its isogenic recombinant strains, RSA59 or RSMHV2 significantly upregulated MMP-3, MMP-8, and MMP-14 transcript levels. Functional network assessment with Ingenuity Pathway Analysis revealed a direct involvement of these MMPs in disrupting junctional assembly between endothelial cells via interaction with junctional adhesion molecules and thereby facilitating transmigration of peripheral lymphocytes. Our findings also suggest mRNA upregulation of Park7, which is involved in NADPH oxidase-dependent ROS production, following RSA59 infection. RSA59 infection resulted in elevated mRNA levels of RelA, a subunit of NF-κB. Infection with MHV-A59 is known to generate ROS, and oxidative stress can activate NF-κB. Thus, our findings indicate the existence of a possible nexus between ROS, NF-κB, and MMPs in RSA59-induced neuroinflammation. We also assessed the expression of endogenously produced regulators of MMP activities. Elevated mRNA and protein levels of tissue inhibitors of metalloproteinases 1 (TIMP-1) in MHV-A59 infection are suggestive of a TIMP-1 mediated host antiviral response. Importance The newly emergent coronavirus has brought the world to a near standstill. In the past, studies have focused on the function of host proteases in virus attachment and entry. Our research indicates the involvement of a group of metal-dependent host proteases in inflammation associated with coronavirus infection. Inflammation is the first response of the host to virus infection. While it helps in restricting the spread and clearance of viral particles, uncontrolled inflammation results in several inflammatory consequences. Therefore, it becomes vital to limit unchecked host immune response. The inhibition of specific metalloproteases represents a potential new therapeutic approach in coronavirus infection and disease outcome.

junctional adhesion molecules and thereby facilitating transmigration of peripheral 24 lymphocytes. Our findings also suggest mRNA upregulation of Park7, which is involved in 25 NADPH oxidase-dependent ROS production, following RSA59 infection. as the number of plaques times dilution factor (DF) per mL per gram of tissue per mL [PFU= 168 (no. of plaques*DF per ml)/ (tissue weight in gram per ml)]. 169

Gene expression analysis. Total RNA was extracted from brain tissues of MHV-A59, 170 RSA59 or RSMHV2 infected as well as mock-infected mice using TRIzol reagent 171 (Invitrogen) following the manufacturer's instructions. RNA concentration was measured 172 using a NanoDrop 2000/2000c Spectrophotometer (Thermo Fisher Scientific), and cDNA 173 was prepared with 1 µg of total RNA using a cDNA reverse transcription kit. Quantitative 174 real-time PCR (RT-qPCR) was performed using SYBR Green dye-based assay in a 175

QuantStudio 3 Real-Time PCR system (Thermo Fisher Scientific) with the following reaction 176 conditions: initial denaturation at 95°C for 7 min, 40 cycles of 95°C for 10 s and 60°C for 30 177 s, and melting curve analysis at 60°C for 30 s. Reactions were performed in triplicates (n=3). 178

Primer sequences are provided in Table 1 . The comparative threshold (ΔΔCT) method was 179 used for relative quantification. The mRNA levels of target genes were normalized with the 180 housekeeping GAPDH gene and represented as the relative fold change values compared to 181 their respective mock-infected controls. 182

Western blotting. Brain tissues (30 mg) were harvested from mice following 183 transcardial PBS perfusion and flash-frozen in liquid N2. Tissues were homogenized (using 9 Qiagen homogenizer) and lysed in 500 uL of RIPA buffer containing protease-cocktail 185 inhibitor and phosphatase inhibitors (1 mM NaVO4 and 10 mM NaF) for 1 hr 30 mins with 186 intermittent vortex every 15 mins. The samples were kept on ice during the entire process. 187

Samples were then centrifuged for 15 mins at 13,000 rpm at 4°C to separate the supernatant. 188

The total protein content in the supernatant was estimated with a BCA protein assay kit. For 189 immunoblotting, 60 µg of total protein per sample was resolved by SDS-PAGE on a 12% 190 polyacrylamide gel followed by transfer onto PVDF membranes using transfer buffer (25 191 mM Tris, 192 mM glycine, and 20% methanol) . Membranes were blocked for 1 hr at room 192 temperature in 5% v/v goat serum prepared in TBST (Tris-buffered saline containing 0.1% 193 v/v Tween 20) and subsequently incubated for overnight at 4°C in polyclonal anti-mouse 194 TIMP-1 antibody at 1:1000 dilution in blocking solution. The membranes were washed in 195 TBST and incubated for 1 hr at room temperature with HRP-conjugated donkey anti-goat 196 secondary IgG antibody. As an internal loading control, g-actin was used, and membranes 197 were blocked separately in 5% w/v non-fat skimmed milk in TBST. Polyclonal anti-mouse 198 g-actin antibody (1:5000 dilution) and HRP-conjugated goat anti-rabbit secondary IgG 199 antibody (1:10,000 dilution) were used. The blots were washed in TBST, and the 200 immunoreactive bands visualized using the chemiluminescent HRP substrate. Non-saturated 201 bands were visualized with Syngene G: box Chemidoc system using GENSys Software. to the median of all samples used as the baseline option. Data were filtered by percentile, and 218 a lower cut off was set at 25. A fold change of ≥ 1.5-fold was considered for differential 219 expression of a gene. Statistical analysis using unpaired student t-tests was performed to 220 compare two groups, with p-values ≤ 0.05 considered significant. The list of MMP and TIMP 221 genes from the microarray data was loaded into QIAGEN's Ingenuity Pathway Analysis 222 software (IPA®, QIAGEN, USA) to perform biological network and functional analyses. 223

Statistical analysis. Data shown are mean ± standard error mean (SEM) for all graphs. 224

Unpaired student t-test with Welch's correction, assuming unequal standard deviations, was 225 performed to examine significant differences between two groups. Multiple comparisons 226

were achieved using ordinary one-way ANOVA, followed by Dunnett's multiple comparison 227 test. A p-value < 0.05 was considered statistically significant. 228

Availability of data. All the data sets used and analyzed in the current study are 229 available from the corresponding author on request. 230

Four weeks old, male C57BL/6 mice inoculated with MHV-A59 (2000 PFU) or mock-233 infected were sacrificed at day 5-6 (acute), 10 (acute-chronic), and 15 (chronic) post-infection 234 (p.i), and brains were harvested. Routine plaque assay was performed with serially diluted 235 brain homogenates to estimate viral replication. MHV-A59 titer was significant between day 236 5-6 p.i (Fig. 1, A; p<0.001) and viral particles were below the detection limit at later time 237 points (data not shown). Total RNA was isolated from mock and virus-infected brain tissues 238 for expression analysis of viral nucleocapsid and Mmp genes through RT-qPCR. Primer 239 sequences are given in Table 1 . Levels of viral nucleocapsid mRNA ( Fig. 1, B ; p<0.0001) 240 coincided with viral replication reaching its peak between day 5-6 p.i, which also marks the 241 acute phase of inflammation. MHV-A59 infected mice exhibited neuroinflammation 242 reaching its peak by 5-7 days p.i, which is associated with meningitis, encephalitis, 243 perivascular cuffing, and macrophage/microglia nodule formation (31). We found similar 244 results in paraffin-embedded brain sections stained by hematoxylin-eosin and 245 immunohistochemistry (data not shown). Transcript levels of Mmp2, Mmp3, Mmp8, and 246

Mmp12 were significantly upregulated at day 5-6 p.i ( 

Tissue inhibitors of metalloproteinases or TIMPs are endogenous protein regulators of 253

MMPs. To understand the regulation of MMPs upon MHV-A59 infection, we also 254 considered the gene expression of TIMPs. As described above, total RNA from brain samples 255 of mock and MHV-A59 infected mice were subjected to RT-qPCR using specific primers 256 12 (Table 1) to determine the transcript levels of Timp1, Timp2, Timp3, and Timp4. MHV-A59  257 infection resulted in significant upregulation of Timp1 mRNA at day 5-6 p.i (Fig. 2, A;  258 p<0.001), while mRNA levels of Timp2, Timp3, and Timp4 remained significantly 259 downregulated (Fig. 2 , C-E; p-values varies as <0.05 to <0.0001). While Timp1 mRNA 260 followed a similar expression pattern as the Mmps following MHV-A59 infection-induced 261 inflammation, its protein levels remained high throughout post-infection, as shown in the 262 representative figure (Fig. 2, B) . Overall, MHV-A59 resulted in elevated TIMP-1 levels in 263 the brain of infected mice. 264

To determine whether the spike (S) protein has any role in inducing Mmp cord. To validate the findings from microarray data, we performed RT-qPCR from brain 277 13 samples of RSA59, RSMHV2, and mock-infected mice sacrificed at day 5-6, 10, and 15 p.i. 278

Brain samples were also harvested for titer assay to estimate viral replication. Like the 279 parental MHV-A59 strain, viral titer and nucleocapsid mRNA levels peaked by day 5-6 p.i 280 in both the recombinant strains as demonstrated in the representative graphs (Fig. 3, A-C) . 281

Although we detected no nucleocapsid mRNAs between 10-15 days p.i in RSMHV2, its 282 presence was observed at day 10 p.i in RSA59 (data not shown). This data corroborates with 283 previous findings that demyelinating RSA59 persists in the brain while RSMHV2 does not 284 persist or, if present, nucleocapsid level is significantly low compared with RSA59 (28). 

Biological and functional network analysis were performed for Mmp3, Mmp8, and 296

Mmp14 genes using QIAGEN's Ingenuity Pathway Analysis (IPA) software. IPA analysis 297 identified that these MMPs could influence several canonical pathways associated with an 298 14 immune response such as leukocyte extravasation signaling, granulocyte and agranulocyte 299 adhesion and diapedesis (Fig. 5, A) . Also, the top disease pathways involved both 300 inflammatory response and immune cell trafficking (Fig. 5, B) . Furthermore, IPA revealed 301 that MMPs facilitate the transmigration of firmly adhered granulocytes (Fig. 6, A) and 302 agranulocytes (Fig. 6, B) 

pathway genes 308

Brain samples from mice infected with RSA59 and sacrificed at day 5-6 and 10 p.i, were 309 harvested for total RNA isolation followed by cDNA synthesis. Mock-infected samples were 310 kept in parallel. We performed RT-qPCR using primers (Table 1) 

in the oxidative and anti-oxidative pathways. Transcript levels of Parkinson's disease 7 312 (Park7) gene were significantly upregulated following RSA59 infection and remained 313 elevated p.i compared to mock-infected samples (Fig. 7, A; p<0.05). RelA, a subunit of NF-314 B, also showed elevated mRNA levels during the acute infection, i.e., 5-6 days p.i (Fig. 7,  315 B; p<0.0001). On the contrary, mRNA levels of Nfb2, a negative regulatory subunit of NF-316 B, remained unchanged p.i (Fig. 7, C) . Similar to another study of our lab (unpublished 317 data), we detected significantly high mRNA levels of nuclear factor erythroid 2-related factor 318 2 (Nrf2) and heme oxygenase-1 (Hmox1) genes during the acute disease phase (Fig. 7 levels. In our current study, RSA59 infection increased transcript levels of Park7, which is 363 involved in oxidative stress. Park7 has a double-sword effect. In lower ROS concentration, 364 it can affect NADPH oxidase by phosphorylating its p47 phox subunit during NADPH oxidase 365 activation, which is crucial for NADPH oxidase-dependant ROS production (41). In one of 366 our studies (unpublished data), we show that Park7 also induces the anti-oxidative pathway 367 via Nrf2 and Hmox1activation during higher cellular ROS concentration. 368 17 Activation of metalloproteases via oxidative pathways has been demonstrated in the past (42-369 44). ROS-induced oxidative stress can also activate NF-B signaling (45). The nuclear factor 370 kappa-light-chain-enhancer of activated B cells (NF-B) acts as a transcription factor and is 371 known to induce inflammation-related genes. RelA, a subunit of NF-B, which gets activated 372 in the canonical pathway via toll-like receptors that recognize pathogenic patterns (46), 373 showed increased mRNA levels following RSA59 infection. On the other hand, Nfb2, 374 which acts as both a precursor and suppressor of NF-B (46), demonstrated unchanged 375 mRNA levels upon infection. Previously, it has been documented that NF-B can induce 376 MMP genes (47, 48). Therefore, our result indicates that Park7 mediated ROS generation 377 leads to the induction of MMP genes via NF-B signaling during MHV-induced acute 378 disease (Fig. 8) . 379

We also found that RSA59 infection-induced upregulation of Nrf2 and Hmox1 genes. The 380

anti-oxidative pathway mediated by nuclear factor erythroid 2-related factor 2 (Nrf2) and its 381 dependant heme oxygenase-1 (Hmox1) (49), could therefore play an essential role in 382 restoring homeostasis through inhibition of ROS overproduction. One limitation of this study 383 that will be addressed in our future experiments is that the interplay between ROS and MMPs 384 has not been validated using inhibitors of ROS as positive controls. 385

In previous studies (50-52) involving MHV-A59, it has been shown that virus infection 386 reduced expression of connexins (Cxs) that form intercellular gap junctional channels and 387 thereby disrupt functional communications between CNS glial cells and fibroblasts. 388

However, the mechanism through which Cx trafficking is altered is not well understood. The (0356) infected mice at different days post-infection (p.i) were harvested, and viral replication was 565 estimated by routine plaque assay. RNA isolated from brain tissues was subjected to cDNA 566

synthesis. An equal amount of cDNA template was used for RT-qPCR. Gene expression was 567 normalized to GAPDH and fold-change values obtained using ∆∆Ct method. A: MHV-A59 568 titer peaked between day 5-6 p.i. B: Viral nucleocapsid mRNA levels peaked at day 5-6 p.i 569 like viral replication. C-F: Mmp2, Mmp3, Mmp8, and Mmp12 mRNA levels elevated 570 between 5-6 days p.i. and coincided with viral replication peak. G: Membrane-associated 571 MMP-14 mRNA levels peaked only at later stages p.i. Data shown are mean ± SEM from 572 two independent biological experiments with nine technical replicates. A significant 573 difference between the two groups was compared with the student t-test. Multiple group 574 comparison was made with ordinary one-way ANOVA followed by Dunnet's test. A p-value 575 of < 0.05 was considered statistically significant (**, p<0.01; ***, p<0.001; ****, p<0.0001). 576

of RT-qPCR revealed a significant upregulation in Timp-1 mRNA levels at day 5-6 p.i in 578 MHV-A59 compared with mock-infected samples (A). Representative immunoblot assay 579 from two independent experiments showed elevated protein levels of TIMP-1 at all examined 580 days following MHV-A59 infection (B). In contrast, Timp-2, -3, and -4 mRNAs remained 581 downregulated throughout p.i (C-E). Graphs show mean ± SEM values from two independent 582 27 experiments with nine technical replicates. Ordinary one-way ANOVA followed by Dunnet's 583 test was performed for multiple group comparisons. Statistical significance was considered 584 for p values < 0.05 (*, p<0.05; ***, p<0.001; ****, p<0.0001). 585 strains of the wild-type MHV-A59 and differs only in the spike gene. Brain samples from 588 mice infected with RSA59 (25000 PFU) or RSMHV2 (100 PFU) were harvested at different 589 days p.i for routine plaque assay and total RNA extraction. A: Both RSA59 and RSMHV2 590 showed peak viral replication between day 5-6 p.i. No detectable viral particles were 591 observed at later time points in a routine plaque assay (data not shown). RT-qPCR data 592 showed significantly elevated mRNA levels of viral nucleocapsid gene in both RSA59 (B) 593

and RSMHV2 (C) at early days p.i, coinciding with peak levels of viral particles as detected 594 in a plaque assay. Data shown are mean ± SEM from three independent experiments having 595 three technical replicates each. Multiple group comparison was made with ordinary one-way 596 ANOVA followed by Dunnet's test. A p-value of < 0.05 was considered statistically 597 significant (***, p<0.001; ****, p<0.0001). 598 

China Novel Coronavirus I, Research T. 2020. A Novel 426

Coronavirus from Patients with Pneumonia in China

Coronavirus as a 429 possible cause of severe acute respiratory syndrome

Isolation of a novel 431 coronavirus from a man with pneumonia in Saudi Arabia

Structure, Function, and Evolution of Coronavirus Spike Proteins

SARS-CoV-2 Cell Entry Depends

Cathepsin L functionally cleaves the severe acute respiratory 458 syndrome coronavirus class I fusion protein upstream of rather than adjacent to the fusion peptide

Endosomal proteolysis 461 by cathepsins is necessary for murine coronavirus mouse hepatitis virus type 2 spike-mediated entry

Cleavage and activation of the severe 465 acute respiratory syndrome coronavirus spike protein by human airway trypsin-like protease

Protease-mediated enhancement 468 of severe acute respiratory syndrome coronavirus infection

Elastase-mediated activation of the severe acute respiratory 470 syndrome coronavirus spike protein at discrete sites within the S2 domain

Co-infection of respiratory bacterium with severe acute respiratory syndrome 474 coronavirus induces an exacerbated pneumonia in mice

Experimental demyelination produced 476 by the A59 strain of mouse hepatitis virus

Cleavage inhibition of the murine 478 coronavirus spike protein by a furin-like enzyme affects cell-cell but not virus-cell fusion

Zinc Metalloproteases for Virus Entry and Cell-Cell Fusion

Expression of 483 matrix metalloproteinases and their tissue inhibitor during viral encephalitis

Matrix metalloproteinase 485 expression correlates with virulence following neurotropic mouse hepatitis virus infection

Metalloproteinases: mediators of pathology and regeneration in the CNS

Intracellular substrate cleavage: a novel dimension in the 490 biochemistry, biology and pathology of matrix metalloproteinases

Chemokine and cytokine processing by matrix metalloproteinases and its 493 effect on leukocyte migration and inflammation

Biochemical and Biological Attributes of Matrix Metalloproteinases

Demyelination determinants map to the spike 497 glycoprotein gene of coronavirus mouse hepatitis virus

Mechanisms of primary axonal damage in 499 a viral model of multiple sclerosis

Enhanced green fluorescent protein 501 expression may be used to monitor murine coronavirus spread in vitro and in the mouse central nervous 502 system

Experimental optic neuritis induced 504 by a demyelinating strain of mouse hepatitis virus

Mouse hepatitis virus type-2 infection in mice: an 506 experimental model system of acute meningitis and hepatitis

Different 508 mechanisms of inflammation induced in virus and autoimmune-mediated models of multiple sclerosis 509 in C57BL6 mice

A proline insertion-deletion in the spike glycoprotein fusion peptide of 512 mouse hepatitis virus strongly alters neuropathology

Mouse hepatitis virus infection upregulates genes involved in innate immune responses

Demyelinating strain of mouse hepatitis virus infection bridging innate and adaptive immune 518 response in the induction of demyelination

Demyelinating and 520 nondemyelinating strains of mouse hepatitis virus differ in their neural cell tropism

Matrix 523 metalloproteinase 9 facilitates West Nile virus entry into the brain

Dengue-virus-infected dendritic cells trigger vascular leakage through metalloproteinase 526 overproduction

SIRT1 activating compounds reduce oxidative 528 stress mediated neuronal loss in viral induced CNS demyelinating disease

Park7 interacts with p47(phox) to direct 531 NADPH oxidase-dependent ROS production and protect against sepsis

Hypochlorous acid oxygenates the cysteine switch 533 domain of pro-matrilysin (MMP-7). A mechanism for matrix metalloproteinase activation and 534 atherosclerotic plaque rupture by myeloperoxidase

Activation of 536 matrix metalloproteinase-2 and -9 by 2-and 4-hydroxyestradiol

Matrix metalloproteinase-9 in pneumococcal meningitis: 539 activation via an oxidative pathway

NF-kappaB in Oxidative Stress

Transcriptional regulation via the NF-kappaB signaling 542 module

Hypoxia induces connexin 43 dysregulation by modulating 544 matrix metalloproteinases via MAPK signaling

Inhibition of transcription factor NF-kappaB reduces 546 matrix metalloproteinase-1, -3 and -9 production by vascular smooth muscle cells

Role of Nrf2/HO-1 system in 549 development, oxidative stress response and diseases: an evolutionarily conserved mechanism

Mouse Hepatitis Virus Infection Remodels 552

Gap Junction Intercellular Communication In Vitro and In Vivo

Microtubule-assisted altered trafficking of astrocytic 555 gap junction protein connexin 43 is associated with depletion of connexin 47 during mouse hepatitis 556 virus infection

Loss of Cx43-Mediated Functional Gap Junction 558

Communication in Meningeal Fibroblasts Following Mouse Hepatitis Virus Infection

We thank the Ministry of Education, India, and the Department of

mice infected with RSA59 (25000 PFU) or mock-infected was subjected to cDNA synthesis, 640 and subsequently, RT-qPCR was performed. A: Data analysis revealed elevated mRNA 641 levels of Park7 during acute (5-6 days p.i) and acute-chronic (day 10 p.i) disease phase. B: 642 mRNA upregulation was detected for RelA, a subunit of the NF-B transcription factor. C: 643In contrast, no change was observed in the mRNA level of Nfb2, a negative regulator of 644 30 NF-B. D & E: Moreover, RSA59 infection also induced increased transcription of Nrf2 and 645 Hmox1 genes. Data shown are mean ± SEM from two independent experiments. A 646 significant difference between multiple groups was compared with ordinary one-way 647 ANOVA, followed by Dunnet's test. A p-value of < 0.05 was considered statistically 648 significant (*, p<0.05; ***, p<0.001; ****, p<0.0001). 649