Differential innate immune response of endometrial cells to 1 porcine reproductive and respiratory syndrome virus type 1 2 versus type 2 3

Modification of cellular and immunological events due to porcine reproductive and 24 respiratory syndrome virus (PRRSV) infection is associated with pathogenesis in lungs. PRRSV 25 also causes female reproductive dysfunction and persistent infection which can spread to fetus, 26 stillbirth, and offspring. In this study, alterations in cellular and innate immune responses to 27 PRRSV type 1 or type 2 infection, including expression of PRRSV mediators, mRNA expression 28 of toll-like receptor (TLRs) and cytokine, and cytokine secretion, were examined in primary 29 porcine glandular endometrial cells (PGE). Cell infectivity as observed by cytopathic effect 30 (CPE), PRRSV nucleocapsid proteins, and viral nucleic acids was early detected at two days 31 post-infection (2 dpi) and persisted to 6 dpi. Higher percentage of CPE and PRRSV positive cells 32 were detected in type 2 infection. PRRSV mediator proteins, CD151, CD163, sialoadhesin (Sn), 33 integrin and vimentin, were upregulated following type 1 and type 2 infection. CD151 , CD163 34 and Sn were upregulated by type 2. Both PRRSV types upregulated TLR1 and TLR6. Only type 2 35 infection upregulated TLR3, but downregulated TLR4 and TLR8 . By contrast, both types 36 upregulated TLR4 and downregulated TLR6 protein expression. Interleukin ( IL )- 1  , IL - 6 and 37 tumor necrotic factor ( TNF ) -  were upregulated by type 2, but IL - 8 was upregulated by type 1. 38 Both PRRSV type 1 and 2 stimulated IL-6 but suppressed TNF-  secretion. In addition, IL-1  39 secretion was suppressed by type 2. These findings reveal one of the important mechanisms 40 underlying the strategy of PRRSV on innate immune evasion in endometrium which is 41 associated with the viral persistence. Widely prevalence of porcine reproductive and respiratory syndrome virus (PRRSV) remains the leading cause of huge economic losses to the global swine industry. Due to an infection of macrophages, PRRSV can persist in animals for extended periods of time associated with long- lasting reproductive disorders. Modification of cellular and immunological responses to PRRSV infection which may be related with the pathogenesis of reproductive disorders remains unclear. Herein, direct PRRSV infection of primary porcine glandular endometrial epithelial cell culture 52 (PGE) demonstrated that PRRSV type 1 and 2 upregulated the protein expression of PRRSV 53 mediators correlated with cell persistence of PRRSV. However, TLR and cytokines gene 54 expression, and cytokine secretion were differentially modulated in response to PRRSV type 1 55 vs. type 2. Our study provides new insights into the cellular mechanism associating with PRRSV 56 persistence in the endometrial cells and the underlying interaction of virus with the host.


46
Widely prevalence of porcine reproductive and respiratory syndrome virus (PRRSV) remains the 47 area at 2 dpi were significantly extended to 20-30% in both type 1 and type 2 inoculated PGE as 110 compared to mock and uninfected PGE (Fig. 1B). A greater extent of CPE area (40%) was found 111 in type 2 inoculated PGE at 4 dpi, but it was recovered at 6 dpi (Fig. 1B). By contrast, the CPE 112 area produced by type 1 were not significantly different from that by the mock and uninfected 113 groups at 6 dpi (Fig. 1B). Corresponding to CPE formation, immunoreactivity of PRRSV 114 envelop protein (GP5; Fig. 1C) was early observed at 2 dpi and remained up to 6 dpi. Type 2 115 inoculation increased the PGE immunoreactive cells about 3-4 times higher than type 1 116 inoculation ( Fig 1C, p < 0.05). 117 The PRRSV existing in type 2-inoculated PGE was 10 9 TCID50/mL which were a ten-118 fold higher than type 1 at 2 and 6 dpi ( Fig. 1D; p < 0.05). However, the PRRSV titers were 119 equivalent in both type 1 and type 2 infected PGE at 4 dpi ( Fig. 1D; p > 0.05). 120 PRRSV type 2 upregulated CD151, CD163 and Sn mRNA 121 expression 122 The mRNA expression of PRRSV mediators, CD151, CD163, Sn, integrin, and 123 vimentin, was determined at 4 dpi to examine whether they might be a target involved in PRRSV 124 infection and existent. The normal uninfected PGE expressed a relatively higher level of 125 vimentin (3-fold) and low level of CD151, CD163, Sn and integrin (0.5-fold) as normalized to 126 GAPDH (Fig 2A). At 4 dpi, only type 2 infection was found to upregulate the expression of 127 CD151, CD163 and Sn by 4-60-fold (p < 0.05, Fig 2B) with no effects on integrin or vimentin. 128 However, both mock and type 1 infection could not produce any effects on the mRNA 129 expression of PRRSV mediators tested (p > 0.05; Fig 2B).

166
Besides alveolar macrophages, the natural cell trophism of PRRSV, PGE has been 167 recently reported as an alternative target of PRRSV, which was preferentially infected via the 168 apical side of PGE monolayer [12]. In relation to the previous study, PGE infectivity as observed 169 by increases in CPE, PRRSV mediators and viral titers were found to be associated with changes 170 in cellular innate immune system, toll-like receptor expression, and cytokine expression and secretion following the apical PRRSV infection in the present study. Both PRRSV type 1 and 172 type 2 infection not only produced the cytotoxic effects but also persisted in PGE up to 6 dpi. 173 However, different cellular responses of PGE to PRRSV type 1 and type 2 infection were 174 observed. 175 In the current study, PRRSV type 2 infected PGE displayed more CPE and higher 176 percentage of PRRSV positive cells than type 1. Thus, more cellular destruction and PRRSV 177 existence caused by type 2 infection indicated the higher virulence to PGE than that by type 1.

Preparation of PRRSV inoculum
300 PRRSV was isolated from the lungs of pigs with respiratory and reproductive illness 301 and positive PRRSV sera at the Farm Animal Hospital (Faculty of Veterinary Science, 302 Chulalongkorn University, Nakorn Pathom, Thailand). To confirm and prepare PRRSV 303 inoculum, 2.3 g of the infected lung tissues was minced, homogenized in 15 mL of cold FBS-304 free DMEM, and centrifuged at 10,000 × g and 4°C for 10 min following the previous protocol 305 [37]. The supernatants were collected and filtered through a 0.2-µm syringe, diluted with FBS-306 free DMEM at a 1:1 ratio, and freshly proceeded to RT-qPCR using primers specific to ORF7 of 307 type 1/type 2, ORF 7 of type 1, and ORF 7 of type 2, as previously described [12,38]. 308 According to our previous study [12], 1 µg of cDNA template was mixed with qPCR 309 SYBR master mix in the presence of forward and reverse primers. All reactions were subjected 310 to CFX96™ Real-Time PCR Detection System (Bio-rad, Hercules, CA, USA) using the 311 following cycle: 95°C for 3 min to activate the reaction, followed by 40 cycles of amplification 312 steps, including denaturation at 95°C for 20 s, annealing at 60℃ for 30 s and extension at 72℃ 313 for 30 s, respectively. The specificity of amplified products was confirmed using 1.5% agarose 314 gel electrophoresis and melting curve analysis. The lungs of PRRSV-negative pigs were isolated 315 and used for mock infection. No amplicons were produced in the mock control. 316 supplemented with DMEM containing 10% FBS. To quantitate the virus concentration as described previously [12], PRRSV inoculum at 1 mL of 10-fold serial dilutions (10 -1 to 10 -6 ) 321

PRRSV inoculum quantification
were incubated with the confluent MARC-145 cells at 37°C in 5% CO2 for 1 h. After viral 322 inoculation, the cells were washed and replaced with the fresh media. CPEs in each MARC-145 323 cell culture well were observed microscopically at 2-, 4-and 6-days post-infection (dpi). 324 Following the Reed-Muench method [39], the dilution that produced CPEs by 50% was 325 considered the tissue culture infective dose 50% (TCID50)/mL PRRSV inoculum stock at the 326 endpoint dilution of 10 5 TCID50/mL to produce the pathological changes was used in this study 327 [12]. 328

PRRSV inoculation
329 PGE (1x10 6 cells/mL) were seeded into 24 mm membrane cell culture inserts 330 (Transwell, MA, USA) or 25 cm 2 flasks (Costar, MA, USA), and maintained in the culture 331 medium for 7 days to become confluence. PGE were then allocated to Mock-, PRRSV type-1 or 332 type-2 group (n = 5 pigs each group). According to our previous protocol [12], the PGE 333 monolayer in 24 mm culture inserts or 25 cm 2 flasks was respectively inoculated with 1 mL or 5 334 mL of PRRSV 10 5 TCID50/mL inoculum in 5% CO2 at 37C for 1 h. Inoculation with solution 335 extracted from PRRSV-negative lungs was used for the mock group. After 1 h of viral 336 adsorption, the cells were washed and replaced with fresh medium for 2-6 days. Each inoculation 337 was performed in duplicate. 338