IFN-λ Decreases Murid Herpesvirus-4 Infection of the Olfactory Epithelium but Fails to Prevent Virus Reactivation in the Vaginal Mucosa

Murid herpesvirus-4 (MuHV-4), a natural gammaherpesvirus of rodents, can infect the mouse through the nasal mucosa, where it targets sustentacular cells and olfactory neurons in the olfactory epithelium before it propagates to myeloid cells and then to B cells in lymphoid tissues. After establishment of latency in B cells, viral reactivation occurs in the genital tract in 80% of female mice, which can lead to spontaneous sexual transmission to co-housed males. Interferon-lambda (IFN-λ) is a key player of the innate immune response at mucosal surfaces and is believed to limit the transmission of numerous viruses by acting on epithelial cells. We used in vivo plasmid-mediated IFN-λ expression to assess whether IFN-λ could prophylactically limit MuHV-4 infection in the olfactory and vaginal mucosae. In vitro, IFN-λ decreased MuHV-4 infection in cells that overexpressed IFN-λ receptor 1 (IFNLR1). In vivo, prophylactic IFN-λ expression decreased infection of the olfactory epithelium but did not prevent virus propagation to downstream organs, such as the spleen where the virus establishes latency. In the olfactory epithelium, sustentacular cells readily responded to IFN-λ. In contrast, olfactory neurons did not respond to IFN-λ, thus, likely allowing viral entry. In the female genital tract, columnar epithelial cells strongly responded to IFN-λ, as did most vaginal epithelial cells, although with some variation from mouse to mouse. IFN-λ expression, however, failed to prevent virus reactivation in the vaginal mucosa. In conclusion, IFN-λ decreased MuHV-4 replication in the upper respiratory epithelium, likely by protecting the sustentacular epithelial cells, but it did not protect olfactory neurons and failed to block virus reactivation in the genital mucosa.


Introduction
Human gammaherpesviruses (γHVs) are highly seroprevalent and induce an important disease burden, including lymphoid and epithelial cancers. Around 90% of humans worldwide are infected with the Epstein-Barr virus (EBV) [1], and up to 30% with the Kaposi's sarcoma-associated herpesvirus (KSHV) [2]. Efficient control of those infections would be of public health interest, particularly in regions where co-infection with malaria [3] and human immunodeficiency virus-1 (HIV1) [4] can increase malignancies. Interruption of transmission would be key in the limitation of prevalence. EBV is mainly transmitted via saliva during childhood or adolescence [5]. It is more rarely transmitted via sexual contact [6]. Transmission of KSHV by the saliva is also predominant in populations with a

Animal Experiments
Ethics: Handling of mice and experimental procedures were conducted in accordance with the european community directive 86/609/CEE and the related Belgian law of 6 April 2010. The protocols used in this study were approved by the Committee on the Ethics of Animal Experiments of the University of Liège, under the permit number 1502 of 15 February 2017. BALB/c female mice were housed at the University of Liège, Department of Infectious Diseases, FARAH. B6.A2G-Mx1-IFNAR1 −/− female mice were kindly provided by Peter Stäheli (Freiburg University, Freiburg, Germany) [43].
Electroinjection: Mice were electroinjected in the tibialis anterior muscle with the empty vector (pcDNA3) or with IFN-λ2 expression vector (pSJ1) to induce the long-lasting expression of IFN-λ2. General anesthesia was induced with a mix of Medetomidin hydrochlorid 100 mg/mL (Domitor) and Ketamine 5 mg/mL (Anesketin) injected intraperitoneally (i.p.). Before DNA injection, mouse legs were shaved with a depilatory cream. 10 µg of endotoxin free plasmid DNA (pSJ1 or pcDNA3) was injected in 25 µL of phosphate-buffered (PBS) in the tibialis anterior muscle of each hind leg of the mice. Electric pulses (80 V per 4 mm, 8 pulses, 20 ms/pulse, pause: 480 ms) were administered using a Cliniporator system (Cliniporator, IGEA, Carpi, Italy) equipped with 4 mm electrode plates. To ensure electrical contact with the skin, conductive gel was used (Aquasonic 100, ultrasound transmission gel, Parker lab, Fairfield, NJ, USA). After electroinjection, mice were then woken up by an i.p. injection of 250 µL of Atipamezole 500 mg/mL (Antisedan).
Infection and sacrifice: Unless otherwise indicated, 8-week-old BALB/c female mice were infected intranasally with 10 4 PFU of Luc+ MuHV-4 diluted in 30 µL PBS, under general anesthesia with isoflurane (lower respiratory tract infection), or with 5 × 10 4 PFU diluted in 5 µL PBS without anesthesia (upper airway infections). They were sacrificed at day 11 or at day 28, and kidneys, spleens and serums were collected. For genital tract collection, the vagina, uterine cervix, and base of the uterine horns were taken jointly.
In vivo luminescence analysis: Mice were injected intraperitoneally (i.p.) with luciferin (60 mg/kg of body weight) and imaged 10 min later with an in vivo imaging system (IVIS) Spectrum instrument (Perkin Elmer, Waltham, MA, USA). For quantitative comparisons, we used Living Image software (Perkin Elmer) to obtain the average radiance (photons per second per square centimeter per steradian) over each region of interest. The background measured in the non-infected right abdominal region was removed from the measurements. The limit of detection of the assay is 10 2 photons/s/cm 2 /sr.

Viruses
KJ7 and TM967 are Theiler's murine encephalomyelitis virus (TMEV) (strain DA) derivatives carrying a capsid adapted to infect L929 cells [44]. Codons 5-to-67 of the leader protein coding region were replaced by the open reading frame (ORF) encoding the green fluorescent protein (GFP) (virus KJ7) or the mCherry (virus TM967) fluorescent protein [44]. Those viruses were produced by reverse genetics. The vesicular stomatitis virus (VSV) derivative expressing the GFP was a gift from Martin Schwemmle (University of Freiburg, Freiburg, Germany).
MuHV-4 virus (MHV-68) was derived from a MuHV-4 bacterial artificial chromosome (BAC) [45]. We used a mutant expressing firefly luciferase under the control of an additional M3 (lytic gene) promoter (Luc+ MuHV-4) [15]. The loxP-flanked BAC/enhanced GFP cassette present in the in vitro experiments was removed from the strain used for the in vivo experiments; detailed construction of those variants was previously described [46]. Virus stocks were grown in BHK-21 cells and infected cell supernatants were collected and cleared of cell debris by low-speed centrifugation (1000× g for 30 min). Viruses were then concentrated by high-speed centrifugation (58,000× g for 90 min). All viruses were titrated by a standard plaque assay on BHK-21 cells.

CRISPR Cas 9 Editing
The Ifnlr1 gene was inactivated in LKR10 cells with the pX461 plasmid (pSpCas9n-2A-GFP) coding for the Cas9 nickase and green fluorescent protein (GFP) [50]. The two single guide RNAs (sgRNAs) were designed using the MIT CRISPR design tool website [51]. Exon 1 of the mouse Ifnlr1 gene (GenBank: NM_174851) was targeted with two single guides sgRNAs, the combination of which produced no expected off-target cleavage sites during prediction: Annealed oligonucleotides that were cloned at the BpiI sites of pX461 were as follows (guide RNA sequences are underlined): sgRNA1: forward (Fw) 5 -CAC CGA GTA GGG GCG CCC ACC GGT-3 , reverse 5 -AAA CAC CGG TGG GCG CCC CTA CTC-3 ; sgRNA2: forward 5 -CAC CGT TCC TGT TGC AGA GCG CCC T-3 , reverse [50] 5 -AAA CAG GGC GCT CTG CAA CAG GAA C-3 . The obtained plasmids targeting Ifnlr1, pSJ3, and pSJ4, were co-transfected into LKR10 cells, using TransIT ® -LT1 transfection reagent (Mirus Bio LLC, Madison, WI, USA), according to the manufacturer's instructions. After 48 h, GFP-positive cells were sorted by fluorescence activated cell sorting (FACS Aria III, BD biosciences, Franklin Lakes, NJ, USA) and cloned in 96-well plates. Cell clones were screened for loss of type III IFN response with an anti-viral assay. Genome editing of the targeted exon in the clone was further confirmed by sequencing (Genewiz, Essex, United Kingdom).

ISGs and Viral Genome Quantification
ISG expression was measured by RT-qPCR. IFNLR1-KO and WT LKR10 cells were treated with 100 U/mL mouse IFN-αA, 700 pg/mL mouse IFN-λ3, or control supernatant (mock) for 24 h before RNA extraction. Total RNA was extracted from cells, kidney and genital tract using the technique of Chomczynski and Sacchi [52]. Reverse transcription and SybrGreen quantitative PCR (qPCR) for mRNA encoding mouse β-actin, Oasl2 and Usp18 were performed as previously described [53]. Spleen DNA was extracted using Wizard Genomic DNA Purification Kit (Promega, Madison, WI, USA). A Taqman probe was used for MuHV-4 genome quantification (ORF25) by qPCR, as described in [46]. ISG expression and viral genome copy amounts were normalized to those of β-actin, in the cDNA or DNA samples.

Flow Cytometry
For infection analysis, cells were dissociated with trypsin-EDTA and suspended in phosphate-buffered saline (PBS) containing 5% of filtered FCS and 0.5% of paraformaldehyde. Data acquisition was done with an LSR Fortessa flow cytometer (BD biosciences) using FACSDiva software. Data were analyzed using FlowJo 9.6.4. The rate of infection was defined as the percentage of mCherry-or-GFP-positive cells. Viral replication was estimated by the mean of mCherry or GFP fluorescence intensity. Infection efficiency was calculated by combining the rate of infection and viral replication (percentage of mCherry −/− GFP-positive cells multiplied by mean of mCherry/GFP fluorescence of these cells). For cell sorting, transfected (CRISPR-Cas experiments) or transduced (expression experiments) cells were suspended in PBS containing 1% FCS and 1 mM EDTA. GFP-positive cells were cloned at one cell per well in 96 wells plate using the FACS Aria III (BD biosciences, Franklin Lakes, NJ, USA).

Vectors
Plasmids pSJ1 and pSJ7 are pcDNA3 derivatives (Invitrogen, Thermo Fisher Scientific) carrying ORFs of mouse IFN-λ2 and IFN-λ3 [43]. The lentiviral vector pSJ12, used for expression of mouse IFNLR1, was obtained by cloning the Ifnlr1 ORF between the BamHI and XbaI sites of pTM945, a lentiviral vector allowing the coexpression of the cloned gene and of mCherry [54]. The Ifnlr1 ORF sequence was subcloned in this plasmid from pcr2.1-LICR2, kindly provided by Laure Dumoutier (de Duve institute, Brussels, Belgium).

IFN-λ Decreases MuHV-4 Infection in Epithelial Cells Overexpressing IFNLR1
The inhibitory potential of IFN-λ on MuHV-4 infection was characterized in vitro. In line with their epithelial origin, LKR10 cells were shown to express both type I and type III IFN receptors [43]. In these cells, mouse IFN-αA and IFN-λ3 treatments significantly up-regulated the expression of the IFN-stimulated genes (ISGs) Oasl2 ( Figure 1A) and Usp18 ( Figure 1B), and significantly decreased infection by TM967, a TMEV derivative expressing mCherry ( Figure 1C). As expected, IFN-λ-mediated ISG induction and anti-viral activity were abolished in an LKR10 cell clone in which the IFNLR1 receptor subunit had been knocked-out ( Figure 1A-C).

IFN-λ Decreases MuHV-4 Infection in Epithelial Cells Overexpressing IFNLR1
The inhibitory potential of IFN-λ on MuHV-4 infection was characterized in vitro. In line with their epithelial origin, LKR10 cells were shown to express both type I and type III IFN receptors [43]. In these cells, mouse IFN-αA and IFN-λ3 treatments significantly up-regulated the expression of the IFN-stimulated genes (ISGs) Oasl2 ( Figure 1A) and Usp18 ( Figure 1B), and significantly decreased infection by TM967, a TMEV derivative expressing mCherry ( Figure 1C). As expected, IFN-λ-mediated ISG induction and anti-viral activity were abolished in an LKR10 cell clone in which the IFNLR1 receptor subunit had been knocked-out ( Figure 1A-C). Next, we assessed the antiviral activity of IFN-λ against MuHV-4. After treatment with 1 ng/mL IFN-λ2, replication of vesicular stomatitis virus (VSV)-GFP and TMEV-GFP (KJ7) was significantly Next, we assessed the antiviral activity of IFN-λ against MuHV-4. After treatment with 1 ng/mL IFN-λ2, replication of vesicular stomatitis virus (VSV)-GFP and TMEV-GFP (KJ7) was significantly reduced but not that of MuHV-4-GFP ( Figure 1D-F, graphs on the left). Antiviral activity of IFN-λ against VSV and TMEV was, however, moderate as compared to that of IFN-αA ( Figure 1D,E), suggesting that LKR10 cells expressed only moderate amounts of the receptor. We thus tested the antiviral activity of IFN-λ in LKR10 cells that overexpress IFNLR1, in which the response to IFN-λ was shown to be enhanced [43]. In such LKR10-IFNLR1+ cells, IFN-λ2 treatment inhibited VSV and KJ7 ( Figure 1D,E, right graphs) infection by more than 95%, while MuHV-4 infection was decreased by 65% ( Figure 1F, right). Of note, cells overexpressing IFNLR1 likely exhibit increased basal STAT1 expression levels and behave as "IFN-primed" cells, as was observed previously in Fawa-λ-luc cells [43]. This may explain the lower infection efficiency observed in these cells with TMEV and MuHV-4. Why VSV infection efficiency was not affected is however unclear.
Taken together, these data show that MuHV-4 exhibits some resistance to IFN-λ, as previously shown for IFN-α in vivo [36]. MuHV-4 infection can, however, be reduced in cells that express high levels of the receptor.

IFN-λ Decreases Nasal but not Lung MuHV-4 Infection
We hypothesized that, in vivo, IFN-λ might protect some epithelial cells from MuHV-4 infection because these cells respond readily to type III IFNs. Using the Luc+ MuHV-4 strain that expresses the firefly luciferase gene, we analyzed the impact of IFN-λ on the primary infection of the respiratory tract using luminescence with an in vivo imaging system (IVIS).
Intranasal inoculation of mice under general anesthesia induces concomitant upper and lower airway infections. Two days before Luc+ MuHV-4 infection, mice were electroinjected with either an empty plasmid or a plasmid (pSJ1) allowing the long-lasting production of circulating mouse IFN-λ2. This technique of plasmid electroinjection has the advantage of leading to long-lasting circulating IFN production without the need for repeated injections, which might influence the inflammatory response. A control group was left untreated (Figure 2A). Four d.p.i., a significant decrease of signal was observed in the snout of IFN-λ2-treated mice as compared to untreated or empty vector-administered mice. In contrast, the signal was comparable in the lungs of all three groups, as in other infected organs up to 14 d.p.i ( Figure 2B,C). Virus replication (i.e., luciferase signal) in the nose and the lungs of all mice then dropped between 7 and 14 d.p.i. while it increased in lymph nodes, in agreement with previous reports showing that type I IFN gradually suppressed alveolar macrophage infection but failed to prevent B cell infection where MuHV4 would evade the IFN response [36,37].
Long term systemic IFN-λ production from the electroinjected plasmid was confirmed by a significant upregulation of the ISGs Oasl2 and Usp18 in the kidneys of the IFN-λ2-treated mice, 28 days post-treatment ( Figure 2D). Quantification of the viral genome by qPCR confirmed that the viral load was not modified in the spleen by the IFN-λ treatment ( Figure 2E). As the mock-treated group did not significantly differ from the controls, we concluded that the electroinjection process itself did not influence the infection profile.
In conclusion, IFN-λ expression decreased infection of the snout early after intranasal infection but did not prevent progression of the virus to the lungs, the lymph nodes, and the spleen.
We then analyzed the effect of IFN-λ expression on the upper airway infections, by infecting alert mice with a reduced inoculum volume. This virus inoculation method was reported to favor upper airway infections. As above, mice were electroinjected two days prior to infection, with the empty or the IFN-λ2-expressing plasmid ( Figure 3A). Luciferase activity was followed at 4, 7 and 11 d.p.i., the latter time point allowing clear detection of the signal in SCLNs after infection of the upper airways [15]. We then analyzed the effect of IFN-λ expression on the upper airway infections, by infecting alert mice with a reduced inoculum volume. This virus inoculation method was reported to favor upper airway infections. As above, mice were electroinjected two days prior to infection, with the empty or the IFN-λ2-expressing plasmid ( Figure 3A). Luciferase activity was followed at 4, 7 and 11 d.p.i., the latter time point allowing clear detection of the signal in SCLNs after infection of the upper airways [15].
A small but significant decrease of the luciferase signal was observed in the snout and superficial cervical lymph nodes of the IFN-λ plasmid-treated mice 4 d.p.i. (Figure 3B,C). As expected after infection of alert mice, no signal was observed in the lungs and the virus propagated directly from the nose to the lymphoid organs. Thirteen days post-treatment, ISGs were significantly more induced in mice that received the IFN-λ plasmid than in mice that received the empty vector A small but significant decrease of the luciferase signal was observed in the snout and superficial cervical lymph nodes of the IFN-λ plasmid-treated mice 4 d.p.i. (Figure 3B,C). As expected after infection of alert mice, no signal was observed in the lungs and the virus propagated directly from the nose to the lymphoid organs. Thirteen days post-treatment, ISGs were significantly more induced in mice that received the IFN-λ plasmid than in mice that received the empty vector ( Figure 3D). Given the lytic infection present 11 d.p.i. that induces IFN production, some extent of ISG induction was observed in the mock-treated group that had been infected but ISG induction was not as strong as in IFN-λ plasmid-treated mice. The viral load in the spleen 11 d.p.i did not significantly differ between groups ( Figure 3E).
Thus, after infection of both anesthetized and alert mice, prophylactic IFN-λ expression reduced early MuHV-4 infection in the upper respiratory tract but not in the lungs or spleen.
Viruses 2019, 11, x FOR PEER REVIEW 9 of 18 ( Figure 3D). Given the lytic infection present 11 d.p.i. that induces IFN production, some extent of ISG induction was observed in the mock-treated group that had been infected but ISG induction was not as strong as in IFN-λ plasmid-treated mice. The viral load in the spleen 11 d.p.i did not significantly differ between groups ( Figure 3E). Thus, after infection of both anesthetized and alert mice, prophylactic IFN-λ expression reduced early MuHV-4 infection in the upper respiratory tract but not in the lungs or spleen. Student's t-test was used to compare mock and IFN-λ-treated mice. In this analysis, one outlier defined by a Grubs' test (p < 0.01) was ignored, which strengthened the significance from * to **. (D,E) One-way ANOVA: * p < 0.05, ** p < 0.01 and *** p < 0.001; no mark-no significant difference.

Sustentacular Cells of the Olfactory Epithelium Respond to IFN-λ Whereas Olfactory Neurons Do Not
After i.n. inoculation of alert mice, Luc+ MuHV-4 was shown to infect olfactory neurons and sustentacular cells of the olfactory epithelium, but not respiratory epithelial cells [38,39]. While the respiratory epithelium had been shown to respond to IFN-λ [40], response to IFN-λ in the olfactory  = 3), mock, or IFN-λ2-treated mice at the time of sacrifice. (E) Viral genome copy numbers (ORF25) per 10 6 β-actin copies in spleen at the time of sacrifice showed no significant difference between groups. (C) Student's t-test was used to compare mock and IFN-λ-treated mice. In this analysis, one outlier defined by a Grubs' test (p < 0.01) was ignored, which strengthened the significance from * to **. (D,E) One-way ANOVA: * p < 0.05, ** p < 0.01 and *** p < 0.001; no mark-no significant difference.

Sustentacular Cells of the Olfactory Epithelium Respond to IFN-λ Whereas Olfactory Neurons Do Not
After i.n. inoculation of alert mice, Luc+ MuHV-4 was shown to infect olfactory neurons and sustentacular cells of the olfactory epithelium, but not respiratory epithelial cells [38,39]. While the respiratory epithelium had been shown to respond to IFN-λ [40], response to IFN-λ in the olfactory epithelium had not been characterized. We thus analyzed Mx1 expression as a marker of IFN response, in head sections of B6.A2G-Mx1-IFNAR1 −/− mice that were electroinjected with an empty vector or an IFN-λ2-expressing plasmid. Co-staining of neurons (olfactory marker protein-OMP), sustentacular cells (cytokeratine 18-CYK18) and Mx1 protein revealed that all luminal sustentacular cells (CYK18+/OMP−) were responsive to IFN-λ, while the underlying OMP+ neurons were not ( Figure 4A-H). The respiratory epithelium was strongly Mx1-positive in IFN-λ-expressing mice ( Figure 4A-C,E-G,I-J) whereas, as expected, only few sustentacular and respiratory epithelial cells were Mx1-positive in mock-treated mice ( Figure 4D,K-L).

IFN-λ Expresion Fails to Prevent Genital Reactivation from Latency in Female Mice
Local IFN-λ treatment has been shown to protect the mouse vaginal epithelium from herpes simplex virus (HSV)-2 infection [18]. We hypothesized that prophylactic vaginal epithelium exposure to IFN-λ might prevent genital reactivation of MuHV-4 and subsequent transmission to partners.
In a preliminary experiment, three groups of BALB/c female mice (n = 10) were infected intranasally with Luc+ MuHV-4 under general anesthesia. After establishment of latency, a control group was left untreated while the two others were electroinjected with either an empty or an IFN-λ2-expressing plasmid ( Figure 5A). IVIS measurements at 7 and 14 d.p.i. confirmed the expected propagation of the virus from the lungs to the lymphoid organs in all mice. IFN-λ expression in the group that received the IFN-λ2 plasmid was confirmed by a significant ISG upregulation in the kidney and in the genital tract at 28 d.p.i.. Infection of mice was confirmed by viral genomic DNA detection in the spleen at the time of sacrifice.

IFN-λ Expresion Fails to Prevent Genital Reactivation from Latency in Female Mice
Local IFN-λ treatment has been shown to protect the mouse vaginal epithelium from herpes simplex virus (HSV)-2 infection [18]. We hypothesized that prophylactic vaginal epithelium exposure to IFN-λ might prevent genital reactivation of MuHV-4 and subsequent transmission to partners.
In a preliminary experiment, three groups of BALB/c female mice (n = 10) were infected intranasally with Luc+ MuHV-4 under general anesthesia. After establishment of latency, a control group was left untreated while the two others were electroinjected with either an empty or an IFN-λ2-expressing plasmid ( Figure 5A). IVIS measurements at 7 and 14 d.p.i. confirmed the expected propagation of the virus from the lungs to the lymphoid organs in all mice. IFN-λ expression in the group that received the IFN-λ2 plasmid was confirmed by a significant ISG upregulation in the kidney and in the genital tract at 28 d.p.i.. Infection of mice was confirmed by viral genomic DNA detection in the spleen at the time of sacrifice. Genital signal occurrence (i.e., virus reactivation) and intensity was then recorded from day 17 to day 21 p.i. This five day follow-up coincides with the estrous cycle of the mouse and should include an estrogen peak accompanied by virus reactivation in the genital tract of most of the mice [16]. The mock-treated group (7/10 reactivations) did not significantly differ from the untreated controls (8/10 reactivations) ( Figure 5B). In the IFN-λ2 plasmid-treated group, a non-significant reduction of signal occurrence was observed (3/10 reactivations). Moreover, the genital signal in the IFN-λ2 expressing mice was only observed in the last days of follow-up, while in the other groups genital signal appeared randomly along the week ( Figure 5B). These data suggested that IFN-λ2 may reduce or delay virus reactivation in the genital mucosa, though non-significantly.
To confirm whether IFN-λ expression was able to decrease or to delay the genital infection, the experiment was reproduced with groups of 20 mock-and IFN-λ2-treated mice. IVIS measures were performed at 7 and 14 d.p.i. to confirm the usual evolution of infection. In this experiment, the reactivation follow-up period was extended up to 23 d.p.i. to detect a potentially delayed genital reactivation ( Figure 6A). No reduction in signal frequency and intensity was observed, however, after IFN-λ2 plasmid-treatment ( Figure 6B). Again, a non-significant delay of one day was observed in the appearance of the signal ( Figure 6C). Upregulated ISGs expression in the kidneys confirmed IFN-λ expression from the administered plasmid ( Figure 6D). Viral genome loads were detected by PCR in the spleen of all mice ( Figure 6E), and were surprisingly slightly higher in the spleen of the IFN-λ expressing mice. Genital signal occurrence (i.e., virus reactivation) and intensity was then recorded from day 17 to day 21 p.i. This five day follow-up coincides with the estrous cycle of the mouse and should include an estrogen peak accompanied by virus reactivation in the genital tract of most of the mice [16]. The mock-treated group (7/10 reactivations) did not significantly differ from the untreated controls (8/10 reactivations) ( Figure 5B). In the IFN-λ2 plasmid-treated group, a non-significant reduction of signal occurrence was observed (3/10 reactivations). Moreover, the genital signal in the IFN-λ2 expressing mice was only observed in the last days of follow-up, while in the other groups genital signal appeared randomly along the week ( Figure 5B). These data suggested that IFN-λ2 may reduce or delay virus reactivation in the genital mucosa, though non-significantly.
To confirm whether IFN-λ expression was able to decrease or to delay the genital infection, the experiment was reproduced with groups of 20 mock-and IFN-λ2-treated mice. IVIS measures were performed at 7 and 14 d.p.i. to confirm the usual evolution of infection. In this experiment, the reactivation follow-up period was extended up to 23 d.p.i. to detect a potentially delayed genital reactivation ( Figure 6A). No reduction in signal frequency and intensity was observed, however, after IFN-λ2 plasmid-treatment ( Figure 6B). Again, a non-significant delay of one day was observed in the appearance of the signal ( Figure 6C). Upregulated ISGs expression in the kidneys confirmed IFN-λ expression from the administered plasmid ( Figure 6D). Viral genome loads were detected by PCR in the spleen of all mice ( Figure 6E), and were surprisingly slightly higher in the spleen of the IFN-λ expressing mice.  In conclusion, IFN-λ non-significantly delayed genital reactivation of MuHV-4 infection despite extensive IFN-λ response in the epithelium of the uterine horns and a more variable response in the vaginal epithelium. In conclusion, IFN-λ non-significantly delayed genital reactivation of MuHV-4 infection despite extensive IFN-λ response in the epithelium of the uterine horns and a more variable response in the vaginal epithelium.

Discussion
Viral transmission is the condition for virus evolution and maintenance in the human population. Herpesviruses have a high prevalence that could be decreased by targeting their transmission. To infect new hosts, viruses need to disrupt or cross the epithelial barriers. IFN-λ is a

Discussion
Viral transmission is the condition for virus evolution and maintenance in the human population. Herpesviruses have a high prevalence that could be decreased by targeting their transmission. To infect new hosts, viruses need to disrupt or cross the epithelial barriers. IFN-λ is a key actor of the innate immune response that limits infection of viruses targeting epithelial cells. In the mouse digestive tract, IFN-λ limits norovirus shedding [19] and restricts host-to-host transmission [20] by acting on epithelial cells. Influenza transmission to naive contact was also restricted by IFN-λ, which protects the upper airways [40]. In the genital tract, prophylactic intra-vaginal IFN-λ treatment prevented HSV-2 infection in the vaginal epithelium of mice [18]. Although herpesviruses developed many strategies to escape the IFN response and persist, increasing the antiviral defense via IFN-λ treatment might tip the scale in favor of the host. γHVs escape the innate immune response and establish latency, mainly in B lymphocytes, a conserved pattern among the species [14]. Transmission between hosts occurs via re-excretion of the virus through epithelial cells. IFN-λ administration might thus prevent spreading of the virus by protecting the epithelial barrier, with limited side effects [41,42]. Here we analyzed the effect of IFN-λ on infection by MuHV-4, the mouse model for γHVs infection.
In vitro, IFN-λ treatment decreased MuHV-4 infection only when cells overexpressed the IFN-λ receptor. Antagonism of MuHV-4 infection was, however, less effective than that of VSV or TMEV. As was shown for type I IFNs [36], MuHV-4 partly resists IFN-λ signaling, but infection can be dampened in cells that express high levels of the receptor. As IFNLR1 is readily expressed by epithelial cells in vivo [23], we hypothesized that prophylactic IFN-λ treatment of mice might protect the respiratory and genital epitheliums from MuHV-4 infection and limit its transmission.
We first tested the antiviral effect of IFN-λ treatment on the primary infection of the respiratory tract and tracked the infection by bioluminescent analysis. Prophylactic IFN-λ treatment partially reduced the nasal luminescent signal, but no difference was seen in the lungs. The absence of lung protection in our experiments is in agreement with previous studies stating that alveolar macrophages, which are not expected to mount a strong IFN-λ response are the first cells infected in the lung [37].
After selective upper airway infections, the nasal and superficial lymph nodes' signal was reduced at 4 d.p.i. in IFN-λ-treated mice, but not at later time points. In the nasal cavity, infection of the olfactory epithelium by MuHV-4 has been well documented [38,39], and involvement of the respiratory epithelium was observed in IFNAR −/− mice [39]. Type I IFN has been shown to be instrumental in the protection of the olfactory bulb after VSV infection. Type I IFN produced by neuroectodermal cells and primarily by astrocytes activated microglial cells [57] and protected distal parts of the brain [58]. Regarding the type III IFN response, the influence of IFN-λ on respiratory epithelial cells to IFN-λ is well documented [40] but little is known about the response of olfactory neurons and sustentacular cells. The nasal luciferase signal (i.e., MuHV-4 replication) reduction observed at early time points after IFN-λ treatment indicates that some of the MuHV-4 target cells are responsive to type III IFNs. Mx1 immunostaining of B6.A2G-Mx1-IFNAR1 −/− IFN-λ-treated mice revealed that luminal sustentacular cells are strongly responsive to IFN-λ, while underlying olfactory neurons are not. It is thus likely that in IFN-λ-treated mice, the observed MuHV-4 infection delay reflected the resistance of sustentacular cells and that infection occurred through the unprotected olfactory neurons.
Sexual transmission of MuHV-4 was documented and follows transient virus reactivation in the vaginal mucosa of infected female mice [16]. Preventing viral shedding from the infected host would constitute an interesting therapeutic option to limit sexual transmission. Bioluminescent analysis of female mice treated with IFN-λ after the establishment of latency, revealed a non-significant delay in luciferase signal appearance, and no decrease in the intensity of the genital signal. In genital tract sections of IFN-λ-treated B6.A2G-Mx1-IFNAR1 −/− , columnar epithelial cells of the uterine horns strongly and consistently expressed Mx1. Vaginal epithelial cells were also responsive to IFN-λ but the response was less consistent between mice. This variability in the response of the vaginal epithelium was not due to ineffective IFN-λ expression given the strong Mx1 induction observed in other epithelial tissues, such as the uterine horns and olfactory epithelium of the same mice. Variability in Mx1 expression in the vaginal epithelium likely correlated with variation in the estrus cycle, as suggested by epithelial thickness differences. Vaginal epithelium differentiation and proliferation are known to be under hormonal control and to vary with the estrus cycle [59,60]. Some studies suggest that estrogens dampen the innate immune response in the female genital tract [58,61]. On the other hand, Caine et al. recently showed that in ovariectomized mice, IFN-λ administration was antiviral against Zika virus when mice were treated with estradiol and progesterone (pro-estrus stage) but not with progesterone only (diestrus) [62]. The hormonal status of the mouse thus influences the response to IFN-λ. Genital reactivation of MuHV-4 infection was reported to occur during the estrus phase of the cycle [16]. A decrease in IFNLR1 expression or signaling, at the time where the epithelial cells are permissive to MuHV-4 infection, might contribute to the viral escape to IFN-λ treatment.
In conclusion, MuHV-4 can escape to some extent, the control by IFN-λ. In vitro, MuHV-4 was less affected by IFN-λ treatment than VSV or TMEV. In vivo, MuHV-4 likely bypasses the IFN-λ response by infecting unresponsive cell types, such as neurons in the olfactory epithelium. In vaginal epithelial cells, MuHV-4 may take advantage of a transient decrease in the IFN-λ response during the estrus cycle.