Human lung vascular endothelium may limit infection with HRV16 via IFN‐β‐dependent mechanisms

Human rhinovirus 16 (HRV16) may induce inflammatory and antiviral responses in the human lung vascular endothelium (ECs) and impair its barrier functions after infection. However, ECs may regain barrier and metabolic functions. Mechanisms of limitation of HRV16 infection in the lung vascular endothelium are unknown. Human lung vascular endothelium (HMVEC‐L) was infected with HRV16. IFN‐β, OAS‐1, and PKR expression was assessed by real‐time PCR, flow cytometry, and confocal microscope. To prove the significance of IFN‐β in the limitation of HRV16 replication, HMVEC‐Ls were preincubated with anti‐IFN‐β Abs. To prove the involvement of OAS‐1 and PKR in the IFN‐dependent limitation of HRV16 replication, HMVEC‐Ls were transfected with respective siRNA. HRV16 stimulated IFN‐β production and activated intracellular mechanisms of antiviral immunity based on OAS‐1 and PKR activation. Blocking of IFN‐β contributed to the inhibition of intracellular mechanisms of antiviral immunity (OAS‐1, PKR) and boosted replication of HRV16. Effective OAS‐1 silencing by siRNA caused the increase of HRV16 copy numbers after HRV16 infection. siRNA upregulated the other genes related to the antiviral response. The infected lung vascular endothelium may limit the HRV16 infection. This limitation may be associated with the induction of IFN‐β‐dependent intracellular mechanisms based on OAS‐1 and PKR activity.

Human rhinoviruses (HRV), the non-enveloped viruses with a single-stranded, positive-sense RNA, are responsible for respiratory infections, such as common cold and asthma exacerbations.They are supposed to infect mainly epithelial cells [1].However, according to our several late studies, the lung vascular endothelium may be infected in in vitro conditions by respiratory viruses [2][3][4] as well as during airway viral infections in vivo [5].The vascular endothelium that acts as a barrier between the blood and tissues regulates the transport of nutrients, ions, and immune cells from the blood to tissues, and conversely, therefore it plays an important role in maintaining homeostasis [6].It was also shown to be the source of a plethora of inflammatory mediators and Type I and III interferons (IFNs) [2,3].
Interferons are a key component of innate immune responses and the "first line" of defense against viral infection.Type I IFNs activate the signaling cascade that leads to the induction of more than 300 IFN-stimulated genes (ISGs) [7], but a few of these are directly involved in the antiviral defense.Indeed, agents, including 2 0 -5 0 oligoadenylate synthase 1 (OAS-1) or ribonuclease L (RNase L), protein kinase R (PKR), and the GTPase MX-1 (myxovirus resistance 1) play a crucial role in antiviral immunity.
In our previous studies, we have shown that rhinovirus HRV16 infection of the human lung vascular endothelium induced a strong inflammatory cytokine and Type I and III IFN synthesis accompanied by an upregulation of OAS-1 and PKR mRNA expression [2,4].We also proved the ability of HRV16-infected endothelium to restore and gain its barrier capacities [8].OAS-1 and PKR are IFN-dependent agents that may be directly involved in the inhibition of viral replication by degrading viral RNA and suppressing the translation of major structural viral proteins, such as capsid proteins.
In the present study, we show that the human lung vascular endothelium may eradicate infection with HRV16 via IFN-b-dependent mechanisms.To our knowledge, this is the first study showing that lung vascular endothelium is able to use universal molecular antiviral mechanisms to fight the rhinoviral infection.

Isolation of mRNA and real-time polymerase chain reaction (RT-PCR)
mRNA was extracted from HMVEC-Ls using the RNeasy Mini Kit (74106; Qiagen, Hilden, Germany) following the manufacturer's instructions.Subsequently, cDNA was synthesized by using the High Capacity cDNA Reverse Transcription Kit (4268813; Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's instructions.Primers were used for IFN-b, RANTES (CCL5), OAS-1, PKR, MX-1, TLR-3, RIG-I, and ICAM-1 (Thermo Fisher Scientific).mRNA expression was made using Step One Plus TM real-time PCR system (Applied Biosystems) and results were normalized to GAPDH (Thermo Fisher Scientific) as the housekeeping gene.Quantitative reverse transcriptase real-time PCR (qRT-PCR), using the Genesig Kit (PrimerDesign, Chandler's Ford, UK) for detection and quantification of HRV16.

Analysis of OAS-1 and PKR concentration by flow cytometry
Intracellular OAS-1 and PKR concentration in HMVEC-L was assessed by flow cytometry (BD LSR Fortessa flow cytometer) at 0, 5, 24, and 72 h after infection.HMVEC-L were fixed and permeabilized using Cytofix/Cytoperm Solution (554722, Franklin Lakes, NJ, USA) for 20 min at 4 °C.Following, the cells were washed with Perm/Wash buffer (554723; BD Biosciences, Franklin Lakes, NJ, USA) and then were incubated with antibodies for OAS-1 and PKR (sc-374656, sc-6282; Santa Cruz Biotechnology, Dallas, TX, USA) or isotype controls for 30 min at 4 °C.After incubation cells were analyzed by flow cytometry.Data were analyzed using BD FACS Diva Software.

Confocal microscopy
HMVEC-L monolayers were fixed in 4% paraformaldehyde (PFA) for 15 min at room temperature and then were permeabilized using 0.05% Triton X-100 for 10 min.After permeabilization HMVEC-Ls were incubated with anti-OAS-1 and anti-PKR antibodies at a concentration 4 lg/mL overnight at 4 °C.Subsequently, the cells were washed and the cell nuclei was stained with DAPI (4 0 ,6diamidino-2-phenylindole) (1:2000, 15 min at room temperature) (D1306; Invitrogen, ThermoFisher Scientific, Waltham, MA, USA).Mouse IgG2a kappa Isotype Control, Alexa Fluor 594 (IC003T; R&D Systems, Minneapolis, MN, USA) was used as an isotype control.Leica TCS SP8 confocal microscope (Leica-Microsystems, Wetzlar, Germany) featuring 639/1.40lens (HC PL APO CS2 Oil Immersion) was used to evaluate the state of intercellular production of proteins.For visualization of cells supercontinuum, there were used: laser with 594 nm excitation and emission at 567-594 nm for OAS-1 and PKR.For nuclei stained with DAPI, excitation and emission parameters were 405 nm and 460-480 nm, respectively.Leica software (Wetzlar, Germany) was used to analyze the data.

IFN-b blocking
IFN-b blocking was performed using neutralizing anti-IFN-b antibodies (ab24309; Abcam, Cambridge, UK).HMVEC-Ls were infected with HRV16 as described above and after HRV16 incubation time in fresh culture medium was added anti-IFN-b antibodies in a concentration of 3 lg/mL (the most effective concentration completely neutralizing IFN-b over 24 h).Medium with anti-IFN-b antibodies was incubated throughout the experiment.
OAS-1 and PKR silencing using siRNA OAS-1 and PKR silencing in HMVEC-Ls was performed using small interference RNA (siRNA).Endothelial cells were transfected using OAS-1, PKR, and Negative Control siRNA (sc-61241, sc-36263, sc-37007, Santa Cruz Biotechnology) following the manufacturer's instruction.Cells were incubated with siRNA for 6 h.Subsequently, 24 h after siRNA transfection endothelial cells were infected with HRV16 as described above.To confirm the transfection of siRNA to HMVEC-Ls was performed fluorescent staining using control siRNA Fluorescent Conjugate (sc-36869; Santa Cruz Biotechnology) and subsequently visualized and analyzed in confocal microscopy (percentage of positively transfected cells were analyzed).A minimum of 300 cells were analyzed.

Infectivity of supernatants from the lung vascular endothelium infected with HRV16
HMVEC-Ls were incubated with HRV-16 for 3 h, washed, and cultured for subsequent 24 or 96 h.Additionally, half of the infected HMVEC-L were washed out in 72 h to further collect supernatants in 96 h.Next, HMVEC-L were cultured for 24 h with supernatants from previously infected HMVEC-Ls and analyzed viral copy number or RANTES mRNA expression and release.

Statistical analysis
The results are presented as mean AE SEM for variables with a normal distribution of values.The Mann-Whitney U-test was used to analyze differences between the two groups.Statistical evaluations were performed using GraphPad Prism (GraphPad Software, Inc., La Jolla, CA, USA) software.Differences less than p < 0.05 were regarded as significant.

HRV16 infects human lung endothelial cells and stimulates IFN production
To better illustrate the dynamics of virus replication, HRV16 copy numbers were detected in cell lysates at the early time points: 0, 2, 5, 8, 10, 12 and the persistence of a copy of the virus in the subsequent hours: 20, 24, 48, and 72 after infection.A peak occurring at 5 h upon the exposition to the virus (14 450 AE 7062 copies/lL) was followed by its rapid decrease that ended up with a value lower than 10% of a peak value at 72 h (Fig. 1A).HRV16 induced the release of IFN-b (261 AE 154.8 pg/mL vs. 12.6 AE 6.2 pg/mL in mock cells, p < 0.05) and IFN-k (79.9 AE 44.4 pg/mL vs. 5.9 AE 5.4 pg/mL in mock cells, p < 0.05) 24 h upon incubation (Fig. 1B).

Effective OAS-1 and PKR silencing caused the increase in HRV16 copy numbers
To analyze the relevance of OAS-1 and PKR in the limitation of HRV16 replication in HMVEC-Ls, siRNA silencing OAS-1 and PKR mRNA were used.
siRNA silences target mRNAs, but they upregulate genes related to antiviral response We observed that siRNA may inhibit specific targets of IFN-b-dependent antiviral protein expression, but it may simultaneously induce the expression of another.Therefore, we wanted to check the effect of anti-OAS-1 and anti-PKR siRNA, and siRNA without any specificity (negative control, neg siRNA) on the expression of IFN-b-dependent antiviral agents (OAS-1, PKR), RNA sensors (TLR3, RIG-I), and inflammatory markers (RANTES, ICAM-1) mRNA expression and proteins in 6 and 24 h.In general, each siRNA induced mRNA expression of studied agents.Moreover, neg siRNA caused a 78% increase in OAS-1 (p < 0.01), anti-PKR siRNA caused 89% enhancement in OAS-1 (p < 0.01), and anti-OAS-1 siRNA led to the 5% up-regulation of PKR (p < 0.01) protein expression as compared to not transfected cells (Fig. 4).

Infectivity of supernatants from the HRV16-infected lung vascular endothelium
To check the production and release of HRV16 particles from HRV16-infected HMVEC-Ls, healthy endothelium was incubated with supernatants from primarily HRV16-infected ECs (primary infection).We observed that supernatants from HRV16infected cells taken in 24 h infected the secondary endothelium (2136 AE 2037 copy/lL) and induced RANTES mRNA expression (490-fold, p < 0.05) and protein release (2694 AE 1352 vs. 80 AE 51 pg/ mL, p = 0.05) as compared to supernatants from the uninfected cells.Supernatants from HRV16infected cells taken 96 h post-infection caused a similar, but less intense effect (HRV16 copy numbers 514 AE 2459 copy/lL; RANTES: mRNA 84-fold increase and protein 1426 AE 1049 pg/mL).However, supernatants from HRV16-infected endothelium that was washed in 72 h post-infection were not able to infect secondary endothelium (HRV16 copy numbers 2 AE 1 copy/lL; RANTES: mRNA 4fold increase and protein 168 AE 86 pg/mL) (Fig. 5).That means that HMVEC-Ls do not produce new viral particles from 72 to 96 h upon the infection.

DISCUSSION
In our previous papers, we have demonstrated that the human lung vascular endothelium may be infected with rhinovirus HRV16, leading not only to strong inflammatory activation but also to the induction of antiviral response [2,4].In a short-term manner, HRV16 impairs barrier functions and regenerative properties of ECs due to the cytopathic effect [4].However, in the long-term period, the lung vascular endothelium infected with HRV16 was shown to limit the infection, recover in time, and regain barrier properties and metabolic functions, thus leading to the functional restoration of the tissue.
Additionally, we noticed that the rhinoviral infection of EC is always accompanied by Type I IFN-b and Type III IFN-k release that is associated with activation of several pattern recognition receptors such as TLR3, TLR7, MDA-5, and RIG-I as well as interferon regulatory factors, including IRF-3 and 7 [2][3][4].Having in mind the observations mentioned above, we put the hypothesis that the activation of Type I IFNdependent responses, especially IFN-b which is strongly released by the endothelium after infection, may serve as a key mechanism determining the eradication of HRV16 infection and subsequent restoration of endothelial barrier and functional properties.
First, we confirmed the typical dynamics of HRV16 copy numbers change in infected ECs.A peak occurring at 5 h upon the exposition to the virus was followed by its rapid decrease that ended up with a value lower than 10% of a peak value at 72 h.HRV16-infected EC released IFN-b and k.What we further noticed was that the pattern of viral copy changes in infected ECs resembled the dynamics of mRNA expression of 2 0 -5 0 -oligoadenylate synthetase 1 (OAS-1), protein kinase R (PKR), and interferon-induced GTP-binding protein Mx-1 (MX-1).OAS-1, PKR, and MX-1 are intracellular enzymes activated by the IFN-b mostly in an autocrine and paracrine manner.They are responsible for the degradation of viral RNA and the inhibition of viral protein production (i.e., capsid proteins) [7].They have been also shown to act as receptors sensing the viral RNA.Furthermore, the increase of OAS-1 and PKR mRNA expression was   accompanied by the enhancement of protein expression.We believe that the inhibition of HRV16 replication at 5 h after infection may be related to the rapid activation of the antiviral response based on interferon production, as well as the activation of intracellular mechanisms-OAS-1, PKR, and MX-1.Type I IFNs are secreted when cells are infected with viruses.Their main function is to induce intracellular antiviral mechanisms in both infected and neighboring, yet uninfected cells in an autocrine and paracrine manner to limit viral replication and the spread of viral infection [9].Indeed, we observed that the neutralization of IFN-b by anti-IFN-b monoclonal antibodies led to a considerable increase of HRV16 copy numbers in HRV16infected ECs.This means that IFN-b deprivation of ECs may lead to the enhancement of viral replication and the increase in viral particles.Blocking of IFN-b in HRV16-infected ECs resulted in an elegant inhibition of OAS-1, PKR, and MX-1 mRNA expression and subsequent protein production.These data indicate that OAS-1, PKR, and MX-1 depending on IFN-b might be key intracellular factors determining the eradication of HRV16 infection of ECs.Concomitantly, IFN-b blockade led to the partial decrease of RANTES and total inhibition of IL-6 release by HRV16-infected ECs (but not IL-8).It suggests that the synthesis of these inflammatory cytokines during rhinoviral infections depends to some extent on IFN-b.Indeed, IFN-b is known to promote the activation and production of plenty of inflammatory cytokines and chemokines [10].
IFN-b-dependent intracellular mechanisms of antiviral immunity based on OAS-1 and PKR have been shown to have a crucial role in the suppression of HRV16 replication and production of new viral particles in airway epithelium [11].Therefore, in the next step, we wanted to analyze the potential contribution of OAS-1 and PKR in the IFN-b-dependent limitation of HRV16 load in ECs.For this reason, we used an in vitro model of silencing the OAS-1 and PKR mRNA in EC cells with small interfering RNA (siRNA OAS-1, siRNA PKR).Basically, we confirmed an effective transfection of ECs with siRNA.We observed, first, that when anti-OAS-1 siRNA effectively silenced the expression of OAS-1, the viral copy numbers increased as compared to HRV16-infected ECs without OAS-1 silencing.However, in cases in which OAS-1 was not effectively silenced, HRV16 numbers were even lesser as compared to cells infected with HRV16.Surprisingly, we noticed that OAS-1 silencing with siRNA in each case led to the increase of PKR expression.Second, the effective siRNA silencing of PKR by anti-PKR siRNA was not accompanied by the increase in viral copies.However, this might be related to the concomitant increase of OAS-1 expression that we observed in ECs transfected with siRNA anti-PKR.Unexpectedly, anti-OAS-1 siRNA and anti-PKR siRNA caused an intense increase in MX-1 mRNA expression.
Analyzing these discrepant results, we put the hypothesis that the specific siRNA used in the study may silence the target antiviral agent, but at the same time, it may cause nonspecific enhance the expression of the other, whose increased activity in limiting the viral replication diminishes the silencing the former.Moreover, this might have more relevant consequences, as in this model of mRNA silencing, ECs are exposed to HRV16 24 h after the transfection with siRNA.Indeed, siRNA gene silencing has been shown to induce IFN-dependent agents' expression (MX-1, OAS, etc.) [12,13].In further experiments, we observed that the negative siRNA (neg siRNA) having no specificity to any mRNA was able to increase RNA receptors (TLR3, RIG-I) and inflammatory agents (RANTES, ICAM-1) expression 24 h after the transfection.However, what is most importantly, negative siRNA significantly increased IFN-b mRNA expression accompanied by the enhancement of IFN-b-dependent OAS-1, PKR, and MX-1 mRNA and protein expression in ECs.Similarly, the siRNA specific to its target mRNA induced an inflammatory activation and IFN-b increase together with IFN-bdependent intracellular antiviral agents.In light of these findings, we may presume that HRV16 entering the cytoplasm of ECs transfected with siRNA 24 h earlier encounters activated OAS-1 and/or PKR, and MX-1 proteins that immediately prevent and limit viral replication.In conclusion, siRNA used in this model might sense by PRRs as if it was of a viral origin, which leads to the induction of IFN-b-dependent antiviral mechanisms interfering with RNA coming from the real HRV16.Paradoxically, siRNA specific for the target may silence the target RNA right after transfection, but soon it may increase it via the induction of IFN-b-dependent mechanisms.Therefore, our study shows that when siRNA anti-OAS-1 effectively silenced the expression of OAS-1, the viral copy numbers increased.However, in cases in which OAS-1 was not effectively silenced, HRV16 numbers were decreased.Finally, there is no doubt that IFN-b neutralization leads to a nonspecific decrease of OAS-1 and PKR protein expression, and MX-1 mRNA expression increases HRV16 replication, which confirms that OAS-1, PKR, and MX-1 are crucial for eradication of HRV16 infection in ECs.
Finally, considering the rapid decrease of viral copy numbers along the infection of ECs, we assessed the infectivity of supernatants taken at different time points of infection.We confirmed that at 96 h HRV16infected ECs which were flushed at 72 h do not produce active viral particles capable of a secondary infection of healthy endothelium.In contrast, supernatants taken 24 h after infection and 96 h (but not flushed at 72 h) effectively infected healthy ECs.
To sum up, in the present study, we show that the lung vascular endothelium infected with HRV16 may eradicate the infection and restore its barrier and functional properties.Furthermore, we prove that this eradication may be associated with the induction of IFN-b-dependent intracellular mechanisms based on OAS-1, PKR, and MX-1 activity.Our data support the hypothesis that as a sensor of viral RNA and a strong source of IFN-b, the lung vascular endothelium may support the innate antiviral responses during rhinoviral infections of airways.Finally, to our knowledge, this is the first study showing that the lung vascular endothelium is able to use universal molecular antiviral mechanisms to fight the rhinoviral infection.Its relevance in airway antiviral immunity requires confirmation in studies on animal models.

Ó
2023 The Authors.APMIS published by John Wiley & Sons Ltd on behalf of Scandinavian Societies for Pathology, Medical Microbiology and Immunology.113 IFN-Β LIMITS HRV16 IN THE LUNG ENDOTHELIUM

Fig. 1 .
Fig. 1.Induction of antiviral mechanisms in the lung vascular endothelium infected with HRV16.HRV16 viral copy number (A); IFN-b and IFN-k release (B); OAS-1 and PKR mRNA expression and intracellular production in FACS and confocal microscopy (red -OAS-1 or PKR, blue -DAPI) (C, D; respectively).The Mann-Whitney U test was used to analyze differences between the two groups.Data are representatives of at least three independent experiments as means AE SEM; n = 3-7; *p < 0.05.

Fig. 2 .
Fig. 2. The effect of IFN-b inhibition on the activity of IFN-dependant mechanisms in the lung vascular endothelium infected by HRV16.The data shows the results after IFN-b blocking for: HRV16 copy number (A); IFN-b release (B); RANTES, IL-8, IL-6 release (C); OAS-1 mRNA expression and production (D, E); PKR mRNA expression and production (F, G).Data are representatives of at least three independent experiments as means AE SEM; n = 3-4.

Fig. 3 .
Fig. 3.The effect of silencing of OAS-1 and PKR on the HRV16 copy numbers in the lung vascular endothelium infected with HRV16 (5 h after HRV16 infection).These data show the results of mRNA expression of OAS-1 and PKR or HRV16 copy number (left panel) and intracellular production of OAS-1 and PKR in FACS (right panel).The Mann-Whitney U test was used to analyze differences between the two groups.Data are representatives of at least three independent experiments as means AE SEM; n = 3-7; *p < 0.05; **p < 0.01.

Fig. 5 .
Fig. 5. Infectivity of supernatants from the lung vascular endothelium infected with HRV16 in different time points after the exposure to HRV16.HRV16 copy number (A); RANTES mRNA expression (B); RANTES production (C).Data are representatives of at least two independent experiments as means AE SEM; n = 4; *p < 0.05.

Ó
2023 The Authors.APMIS published by John Wiley & Sons Ltd on behalf of Scandinavian Societies for Pathology, Medical Microbiology and Immunology.119 IFN-Β LIMITS HRV16 IN THE LUNG ENDOTHELIUM