Human respiratory syncytial virus (hRSV) is a leading cause of bronchiolitis and pneumonia in infants and is linked to the development of airway hyperreactivity in children who suffer severe disease early in life (Rosas-Salazar et al., 2023). hRSV infection in the elderly is also of concern, as the incidence of mortality may exceed that estimated for infants by a factor of 20 (Coultas et al., 2019).

hRSV is an enveloped virus belonging to the Pneumoviridae family, Orthopneumovirus genus. It possesses a single-stranded negative sense RNA genome and contains 10 genes that encode for 11 proteins (Battles and McLellan, 2019). Viral PAMPs (pathogen associated molecular patterns), like proteins and single-stranded or double-stranded RNA (ssRNA or dsRNA) trigger signaling pathways that activate transcription factors such as interferon regulatory factors (IRFs), NF-κB, ATF2 and c-Jun to establish an antiviral and proinflammatory state (Kawai and Akira, 2006; Carty et al., 2021). Particularly, hRSV nucleic acids are recognized through RLRs (ddx58/RIG-I and Ifih1/MDA5), TLR3, TLR7 and TLR8, while proteins are detected by TLR2, TLR6 and TLR4 (Ouyang et al., 2022).

Proinflammatory cytokines and chemokines that recruit leukocytes are produced during hRSV infection and contribute to viral clearance. However, in younger infants hRSV, severe disease is associated with a deregulated inflammatory response associated with the development of recurrent wheezing and asthma-like symptoms (Roe et al., 2011; Russell et al., 2017). The mechanisms linked to these respiratory complications are not clear. However, they appear to be multifactorial, involving factors such as age, host-genetic predisposition, immune responses, and the virus’s ability to establish persistent infections (Bertrand et al., 2015; Bont and Ramilo, 2011; Kato et al., 2012; Mejías et al., 2005; Piedimonte and Perez, 2014; Wu and Hartert, 2011).

hRSV persistence has been described in the respiratory tract of immunocompromised subjects and in mouse models (Tabatabai et al., 2018). hRSV genomic RNA and viral messenger RNA were detected for over 100 days after primary infection in mice. Recovery of infectious virus occurred only after depletion of CD4 and CD8 T lymphocytes, suggesting that the immune response has an essential role in viral elimination (Schwarze et al., 2004). Furthermore, hRSV persistence has been established in human epithelial cell lines and mouse macrophages (Martínez et al., 2009; Sarmiento et al., 2002; Valdovinos and Gómez, 2003). Cell lines are a valuable tool to characterize the virus-host cell interactions responsible for the establishment and maintenance of long-term infections. The study of the mechanisms, consequences and possible therapeutic targets of long-term viral infections is fundamental, given the large number of chronic diseases associated with viral persistence.

The role of macrophages in hRSV infection is complex and depends on the cell phenotype, time post infection and interaction with other cells. Macrophage polarization is influenced by different mechanisms that include cytokine microenvironment and lipid metabolism. Recently, M1 polarization has been characterized by a high rate of glycolysis, whereas oxidative phosphorylation and fatty acid oxidation are characteristic of M2-like macrophages (Batista-Gonzalez et al., 2020). Viruses often evolve to promote M2-like responses in order to survive in host cells (Sang et al., 2015). In coculture experiments with human bronchial epithelial cells and blood derived macrophages, M2-like macrophages were shown to increase hRSV infection, while M1-like macrophages protected against epithelial infection (Ronaghan et al., 2022).

We previously reported the ability of hRSV to persist for over 150 passages in the P388D1 mouse macrophage-like cell line (Sarmiento et al., 2002; Ruiz-Gómez et al., 2021) and showed that hRSV-persistently infected macrophages (piMϕ) have a predominantly proinflammatory profile or M1-like status. Indeed, conditioned medium from piMϕ can induce a proinflammatory/antiviral state in non-infected macrophages due to the presence of biologically active inflammatory cytokines and chemokines (Rivera-Toledo et al., 2017). Nevertheless, arginase 1 (Arg-1), a marker of M2 polarization in macrophages, is highly expressed and shows increased enzymatic activity in piMϕ, which interferes with the production of nitric oxide (NO), a potent antiviral effector molecule (Santiago-Olivares et al., 2019). This suggests that this M2 phenotype marker could contribute to viral persistence. It is not known whether hRSV persistence can induce the expression of other M2-associated genes.

Calreticulin (CRT) is a multifunctional and ubiquitous chaperone with canonical functions in the endoplasmic reticulum that include Ca++ homeostasis and correct glycoprotein folding (Gelebart et al., 2005; Michalak et al., 2009). However, CRT has also been reported outside the endoplasmic reticulum, in the cytoplasm, cell surface and extracellular space (Gold et al., 2010). In addition to its housekeeping functions, CRT has gained much attention due to its strong immunomodulatory activities that range from anti-inflammatory to pro-inflammatory effects. CRT has been identified in a wide range of parasites and influences cell responses and host-pathogen interactions. Mammalian CRT and its fragments have been shown to be potent stimulators of macrophages, stimulating the production of proinflammatory cytokines (Duo et al., 2014; Hong et al., 2010; Esperante et al., 2023; Huang et al., 2013). Many of the immunomodulatory functions of mammalian CRT are mirrored by parasite CRT.

Helminths are parasites that have coevolved with their human host for millennia. This long-term interaction has resulted in helminth survival strategies that regulate the host immune response through immunomodulatory molecules (Gazzinelli-Guimaraes and Nutman, 2018). A recent review by Ryan et al. (2020) suggests that helminth-derived product represent a “untapped pharmacopeia.” CRT from several parasites has been studied in different disease scenarios such as cancer, immune evasion, immune response, vaccination, and immunomodulation (Esperante et al., 2023; López et al., 2010; Rzepecka et al., 2009; Winter et al., 2005). Evidence suggests that CRT homologues exhibit comparable functions across different species. For example, CRT from different parasites can inhibit the complement system, thus facilitating their survival within the host (Esperante et al., 2023; Ramírez-Toloza and Ferreira, 2017). Furthermore, both the protozoan parasite Trypanosoma cruzi CRT (TcCRT) and the helminth Taenia solium CRT (TsCRT) recombinant forms exhibit antitumoral properties (López et al., 2010; Schcolnik-Cabrera et al., 2020).

However, TcCRT is more efficient in inhibiting the complement system and has a stronger antitumoral effect in comparison to human CRT (López et al., 2010; Ramírez-Toloza et al., 2020). The difference in efficiency can be attributed in part to conformational rearrangements based in species-specific structural features (Peña Álvarez et al., 2020).

We previously cloned and expressed CRT from the helminth Taenia solium as a recombinant protein (rTsCRT) (Mendlovic et al., 2004). Given the potential of mammalian CRT to activate macrophages, we hypothesized that rTsCRT can trigger a proinflammatory immune response in piMϕ, potentially leading to antiviral activity. In this study we analyzed the transcriptome profile of piMϕ with special emphasis on the antiviral and inflammatory responses, as well as on arginine and lipid metabolisms. We examined different time points after initiation of macrophage culture to identify early and late onset genes. In addition, we investigated the effects of rTsCRT treatment on the viral load and transcriptional signature induced by persistent hRSV infection.

In this study we investigated the transcriptional signature of piMϕ and found that persistent hRSV infection turns on predominantly genes involved in inflammation, antiviral response, as well as arginine and lipid metabolism, suggesting that piMϕ attain a mixed M1/M2 profile. Additionally, we analyzed the transcriptional changes induced by rTsCRT treatment and showed further stimulation of the inflammatory response. This stimulation was associated with a reduction in the number of hRSV copies in piMϕ.

In addition to epithelial cells, hRSV can infect both human and mouse alveolar macrophages (AMϕ) (de Souza et al., 2019; Johnson et al., 2007; Makris et al., 2016; Ravi et al., 2013). Evidence suggests that macrophage infection is temporarily productive as infective viruses are detected at very low titers (≤2 × 10¹ PFU/mL). Subsequently, it changes to a nonproductive state (absence of virions) with persistent expression of viral proteins, while recovery of viral RNA is sustained (Sarmiento et al., 2002; Ravi et al., 2013). The low-to-nonproductive hRSV infection in macrophages might not be important to propagate the virus. Nevertheless, viral proteins and nucleic acids persist as chronic stimuli for the immune system (Rivera-Toledo et al., 2017; de Souza et al., 2019; Ravi et al., 2013).

During acute hRSV infection viral dsRNA is detected by RIG-I and MDA5 to activate IRF3, which translocates to the nucleus and promotes transcription of IFN-β. IFN-β interacts in both autocrine and paracrine manners through the heterodimeric receptor composed of the IFNAR1 and IFNAR2 subunits to activate the Jak/STAT signaling pathway. Ultimately, the heterotrimeric interferon-stimulated gene factor 3 (ISGF3), composed of STAT-1, STAT-2 and IRF9 is assembled and translocates to the nucleus to initiate the transcription of multiple interferon-stimulated genes (David et al., 1993). Interestingly, in this study we observed that piMϕ downregulated genes involved in the recognition of viral RNA and genes transcribed in response to IFN-I, such as Ddx58/Rig1, Ifnar1, Ifnar2, Stat1, Stat2 and Irf9. Conversely, the expression levels of Ifna and Ifnb remained unaltered. Previously, we reported that IRF3 is active in piMϕ, as it is normally phosphorylated and located in the nucleus. However, piMϕ do not respond to either endogenous or exogenous IFN-β, as demonstrated by the lack of STAT-1 phosphorylation after a 45 min stimulation, thus limiting the expression of antiviral genes (Rivera-Toledo et al., 2015). A main function of the hRSV nonstructural proteins 1 and 2 (NS1 and NS1) is to evade the IFN-I-mediated antiviral response by targeting RIG-I, MAVS and STAT-2, as well as to induce expression of suppressor of cytokines signaling 1–3 (SOCS 1–3) that blocks the activation of the Jak/STAT pathway (Hashimoto et al., 2009; Sedeyn et al., 2019). It was recently described that NS1directly binds to regulatory elements of immune response genes during hRSV infection, suppressing their transcription (Pei et al., 2021). The antiviral transcriptomic profile reported herein agrees with these data. We observed either downregulation or transitory expression of many antiviral genes. However, genes such as Ifitm10, Ifi203, Ifi205, Ifit1 and Ifit2 showed an upregulated transcription at 6 and 24 h, suggesting that expression of some antiviral genes is regulated by alternative pathways to ensure host-cell survival during viral infections. Further underscoring the importance of antiviral pathways in controlling hRSV replication. The balance of an active immune system, capable of limiting virus replication, and the ability of the virus to dampen such antiviral responses are essential conditions to prevent cell death and facilitate long-term viral infections (Griffin, 2022).

In addition to IRF3, NF-kB can be activated by ligands of TLR3 in a MyD88-independent manner and by signaling via the IL-1/IL-1-Receptor through adaptor proteins expressed by Myd88, Tollip and Peli1. The latter two genes, along with Tlr3 were upregulated in piMϕ. Hence, IRF3 and NF-kB may be key transcription factors controlling the long-term interaction between the virus and host cells. IRF3 does not exclusively target type-I interferon genes, it also induces the transcription of other antiviral genes, as well as the chemokine Ccl5 that has been shown to have antiapoptotic properties in macrophages (Diamond and Farzan, 2013; Grandvaux et al., 2002; Jiang et al., 2004; Tyner et al., 2005). Accordingly, while we did not observe alterations in the expression of Ifn α/β, the ISGF3-related genes were downregulated in piMϕ. Overexpression of Ccl5 and some antiviral genes could be explained by an ISGF3-independent signaling pathway, potentially involving direct activity of IRF3 and/or NF-kB (Diamond and Farzan, 2013). In fact, NF-kB activation is induced by diverse stimuli making it an essential transcription factor that mediates responses to physiological stress (Pahl, 1999). Promoters of many interferon-stimulated genes (ISGs) have binding sites for NF-κB and cooperate with IRFs to establish the antiviral state (Antonczyk et al., 2019).

Other target genes for NF-kB include cytokines, chemokines, effectors of the oxidative stress and apoptosis, immunoreceptors and transcription factors (Pahl, 1999; Nilsson-Payant et al., 2021). In piMϕ we identified upregulation of many proinflammatory cytokine and chemokine genes, like Il1a, Il1b, Il6, Il12a, Ccl2, Ccl4, Ccl7, Ccl12, Cxcl2, Cxcl10 and Cxcl14. These proinflammatory cytokines are associated with a M1 phenotype in macrophages. In contrast, Tnf-α, Il18 and Il18 receptor accessory protein (Il18rap) that also have a central function as proinflammatory mediators, were downregulated (Rex et al., 2020). One pathway that can induce expression of TNF-α, IL-1α, IL-1β and IL-6 is through IL-18/IL-18 receptor and IL-18rap (Tsutsui et al., 1997). We found that the components of the IL-18 signaling pathway were downregulated in piMϕ. Accordingly, expression of Tnfa is negative regulated in piMϕ. However, we detected overexpression of Il1a, Il1b and Il6, suggesting that alternative pathways can compensate for the reduced activity of the IL-18 signaling in inducing these cytokines. Additionally, the apoptosis-associated speck-like protein containing a CARD domain (Asc or Pycard) and the caspase 1 genes, both related to processing of the pro-IL1β through the Nlrp3 inflammasome (Schroder and Tschopp, 2010) were downregulated, suggesting that Il1β might not be in its bioactive form. Indeed, we have observed very low levels of IL-1β in conditioned medium from piMϕ (Rivera-Toledo et al., 2017).

We compared 3 earlier transcriptional studies conducted in mouse macrophages and A549 epithelial cells acutely infected with hRSV for 20–24 h with our results in piMϕ (Supplementary Table S1). These studies were mainly focused on genes related to the immune response. We identified equivalent fold-change expression of Il1a, Il1b, Il6, and some antiviral genes in piMϕ and acute hRSV infection. Interestingly, we observed the downregulation of genes such as Tnf-α, Il18, Mx1, Rig1, Pkr, Stat1, Stat2 and Irf9, which appear to be hallmarks of hRSV persistence in contrast to the acute infection.

During early stages of hRSV infection TNF-α plays a protective role in the control of viral replication, while its chronic production can cause severe illness through immunopathology (Neuzil et al., 1996; Rutigliano and Graham, 2004). Therefore, TNF-α has a critical role in development and outcome of the hRSV infection and its expression is tightly regulated. The transcriptional regulator BCL3 that belongs to the IkB protein family, modulates Tnf-α expression by recruiting histone deacetylases to its promoter and induce a repressive chromatin state (Walker et al., 2013). We identified upregulation of Bcl3 gene at 6 and 24 h, accompanied by overexpression of the histone deacetylases (Hdac)-2, 8 and 10 in piMϕ. These observations suggest that these genes may participate in the pathway responsible for downregulating Tnf-α transcription. However, these findings require experimental validation.

Concerning proteins related to the transcription factor NF-kB, we identified upregulation of Nfkb1 (p50), Nfkbia (Ikba) and Nfkbiz (IκBζ). Previous studies have shown that inflammatory stimuli induce the expression of IκBζ in monocytes/macrophages (Sundaram et al., 2016). IκBζ can subsequently associate with p50 homodimers bound to the IL-6 promoter, facilitating its transcription (Kamata et al., 2010). These mechanisms may contribute to the upregulation of IL-6 mRNA in piMϕ. The role of IL-6 during viral infections may be protective, by inhibiting virus replication or deleterious, through a synergistic interaction with IL-17 that induces antiapoptotic molecules avoiding elimination of infected cells and promoting long-term infections (Velazquez-Salinas et al., 2019). Although IL-6 has been associated with hRSV severe disease in infants, its participation in virus persistence has not been studied (McNamara et al., 2004).

The transcriptional profile in piMϕ suggests the involvement of multiple mechanisms to maintain hRSV persistence, including inhibition of Tnf-α expression and the IFN-I response. The apparent paradoxical upregulation of some antiviral genes, such as the RNA editing enzyme Adar1, Isg20, and genes within the IFIT and IFITM families (Ifit1, Ifit2, and Ifitm10), which restrict viral entry by altering membrane fusion and virus genome replication, may indicate molecular strategies used by the virus to modulate its own replication and persistence. For instance, ADAR1 induces hypermutation (A-to-G substitutions) in the matrix gene of the measles virus, resulting in its defective expression. This leads to a lack of infectious viral particles and persistence of virus genome in the central nervous system, resulting in fatal subacute sclerosing panencephalitis (Samuel, 2012). These findings are consistent with our previous results, which show defective virus production and a lack of syncytia formation in piMϕ. Additionally, a high proportion of A-to-G transitions is a hallmark of the persistent hRSV genome (GenBank accession no. MT492011 and MT492012) (Sarmiento et al., 2002; Ruiz-Gómez et al., 2021).

Genes associated with arginine metabolism also showed profound alterations during hRSV persistence. Notably, Arg1 ranked third among the genes with the highest transcription fold change, in agreement with our earlier findings (Santiago-Olivares et al., 2019). In the arginine biosynthetic pathway, nitric oxide synthase 2 (NOS2) produces nitric oxide (NO) and citrulline from L-arginine. Then, argininosuccinate synthase 1 (ASS1) recycles citrulline to produce argininosuccinate, which is the substrate of arginosuccinate lyase (Rath et al., 2014). An increased expression of the Ass1 gene in piMϕ suggests that arginine metabolism is active to allow bioavailability of this amino acid as a NOS2 substrate, since NO is essential to inhibit viral infection. Nevertheless, overexpression of Arg1 that has higher affinity for arginine compared to that of NOS2, can importantly divert the synthesis of NO contributing to hRSV persistence (Santiago-Olivares et al., 2019).

Macrophages with an M2 phenotype are associated with expression of Arg1 and synthesis of ornithine, proline and polyamines (Martí et al., 2021) that modulate inflammation and in tissue repair (Kieler et al., 2021). By overexpressing Arg1, hRSV can shift the balance of arginine metabolism to the production of polyamines that may participate in genome packaging and activity of viral enzymes (Tomé, 2021), to ensure persistence and concomitantly evade elimination by NO. Indeed, ornithine aminotransferase (OAT) and ornithine decarboxylase (ODC) that produce proline and glutamate and polyamines, respectively (Caldwell et al., 2015) are upregulated in piMϕ.

The anti-inflammatory M2 phenotype is also favored by fatty acid oxidation (FAO) and oxidative phosphorylation (Batista-Gonzalez et al., 2020). Our transcriptome analysis showed upregulated expression of neutrophil gelatinase–associated lipocalin 2 (Lcn-2), peroxisome proliferator activated receptor-γ (Pparγ), as well as CD36, all of which are involved in FAO (Chawla, 2010; Jin et al., 2011; Lim et al., 2006). Moreover, piMϕ upregulated the cysteine proteases cathepsins (Cts) K, L and B, shown to promote a M2 macrophage phenotype in tumor-associated macrophages (Li et al., 2019). M2 polarization is induced by several viruses as an immune evasion mechanism for efficient replication (Sang et al., 2015; Yu et al., 2022). In alveolar macrophages, hRSV can polarize the initial M1 response to an M2 phenotype that can promote persistent infection and continuous activation of the immune response (Wang et al., 2022). Thus, the antiviral and inflammatory responses together with modulation of the arginine and lipid metabolism pathways contribute to viral persistence and host cell survival (Figure 7).

rTsCRT potentiated the proinflammatory state induced by the presence of the replicating hRSV in piMϕ. Ila, Il1b, Lcn2, Cxcl2 and Nos2 expression were stimulated in both niMϕ and piMϕ, while Il6, Tnf-α and Socs3 were upregulated only in piMϕ. The proinflammatory effects of rTsCRT treatment were accompanied by the expression of negative modulators such as Socs3, leucine rich repeat containing 25 (Lrrc25) in piMϕ, as well as Zc3h12a, a RNase that post-transcriptionally regulates Il6 mRNA, along Clec4a1, and IRAK3 in both niMϕ and piMϕ. Expression of zinc finger protein 36, C3H type-like 1 (Zfp36l1), critical for the decay of the mRNAs for TNF-α, was induced by rTsCRT treatment after 48 h (Matsushita et al., 2009). Thus, treatment with rTsCRT induced a proinflammatory phenotype accompanied by intrinsic regulation of the inflammatory response.

Treatment with rTsCRT resulted in a reduction in the viral load in piMϕ. This effect was accompanied by the induction of proinflammatory cytokine and chemokine genes, as well as by the production of TNF-α and IL-6 in rTsCRT-treated piMϕ. Accordingly, the transcriptomic analysis showed that rTsCRT reversed the inhibition of Tnf-α expression induced by viral persistence and upregulated the expression of Tnf-α in rTsCRT-treated piMϕ. TNF-α has been shown to have antiviral activity for several viruses including hRSV and can stimulate the NFκB and MAPK pathways, as well as genes such as Il1b, Il6, Csf2, Ptgs2, Socs3 and Nfkbia, all of which were upregulated in rTsCRT-treated piMϕ (Figure 7). Indeed, mouse recombinant CRT has been shown to stimulate the production of TNF-α in mouse peritoneal macrophages by activating IκBα-NFκB and the c-Jun N-terminal kinase (JNK), as well as the production of IL-6 by phosphorylating the extracellular signal-regulated kinase (ERK) (Duo et al., 2014).

In addition to the upregulation of IL-6 and TNF-α, rTsCRT stimulated Nos2 expression and we observed a marginally higher production of nitrites at 48 h after rTsCRT treatment (Supplementary Figure S3). Accordingly, mammalian recombinant CRT and its fragments can induce the accumulation of nitrites in peritoneal macrophages (Huang et al., 2013). NO has been shown to interfere with hRSV replication in both epithelial and macrophage cell lines (Santiago-Olivares et al., 2019; Kao et al., 2001). Thus, the effects on stimulation of cytokine and chemokine expression, as well as on Nos2 expression, could explain the observed rTsCRT antiviral activity. The mechanism of action and the receptors involved in mediating the antiviral effects of rTsCRT need to be elucidated. Several receptors have been implicated in binding CRT, including low-density lipoprotein receptor (LRP-1), scavenger receptor A and F, and TLR-4 (Berwin et al., 2004; Berwin et al., 2003; Gardai et al., 2003; Li et al., 2015; Ogden et al., 2001). Further analysis is required to identify the receptors for rTsCRT and define the signaling pathways involved in its antiviral effect.

hRSV persistent infection altered the expression of many genes. The hRSV model of persistence in macrophages showed downregulation of many genes associated with the antiviral response. Paradoxically, we also observed overexpression of a few antiviral genes, RNA editing enzymes, as well as proinflammatory cytokines and chemokines that induce an M1 phenotype and can control viral replication. Concomitantly, lipid metabolism-related genes, together with Arg1, which collectively promote an M2 status were overexpressed. This suggests that hRSV persistence induces a mixed M1/M2 phenotype characterized by a weakened antiviral response. Viral persistence depends on mutual adaptation between the virus and the host-cell to succeed in long-term infection. The seemingly contradictory effects on the macrophage transcriptome may represent virus-induced mechanisms aimed at modulating its own replication while simultaneously evading the immune response. Thus, hRSV persistence appears to involve a balance between pro-inflammatory and anti-inflammatory signals to control viral replication and ensure the survival of both the infected cell and the virus. Remarkably, rTsCRT showed that piMϕ are not refractory to external stimuli and that the apparent dampening of immune response signaling pathways is still reversible, resulting in a significant reduction in viral genome copies. Our observations have implications for the development of targeted therapies aimed at eliminating or controlling persistent viral infections often associated with immunopathology.