Gapmers targeting conserved SARS-CoV-2 single-stranded regions reduce viral load and reshape inflammatory macrophage responses Greta Gallo, Lorena Donnici, Liam S Colley, Alessia Loffreda, Valeria Fumagalli, Matteo Iannacone, Davide Mazza, Danny Incarnato, Raffaele De Francesco, Emanuele Monteleone, Alessio De Magis, Marco E Bianchi Abstract The COVID-19 crisis is now over, but additional viral strains of SARS-CoV-2 that evade immunity are continuously emerging. We explored the design, evaluation, and optimization of anti-sense oligonucleotides (ASOs) targeting SARS-CoV-2 RNA, with the aim of providing a scalable strategy capable of responding to emergent outbreaks. Twenty-four ASOs were designed to target highly conserved, single-stranded regions of the SARS-CoV-2 genome. Extensive in vitro screening led to the identification of gapmers that exhibited high efficacy against four selected SARS-CoV-2 variants, significantly reducing intracellular viral RNA and virus production. Importantly, the antiviral efficacy correlates with viral RNA accessibility and is not due to activation of antiviral host responses. Finally, naked gapmers delivered intranasally can reach lung tissue, are preferentially internalized by alveolar macrophages, and modulate the inflammatory response to viral exposure. Taken together, our findings show that gapmers targeting the viral RNA are a promising approach to the development of antiviral therapies. Graphical abstract[ga1.webp][ga1.gif] Highlights•Gapmers targeting conserved SARS-CoV-2 ssRNA regions reduce viral load•Gapmer efficacy correlates with RNA structural accessibility (SHAPE-MaP)•Antiviral effect stems from RNA cleavage, not interferon activation•Intranasal gapmers reach lung macrophages and reshape inflammation Biological sciences; Microbiology; Virology Introduction Coronaviridae gained global recognition due to multiple serious outbreaks, including the Coronavirus disease 2019 (COVID-19) pandemic. Research on coronaviruses dates back to the early 1930s. The prominence of coronaviruses surged in 2003 with the emergence of the severe acute respiratory syndrome coronavirus (SARS-CoV) epidemic, and in 2012 when Middle East respiratory syndrome coronavirus (MERS-CoV) triggered a second epidemic. In the absence of specific antivirals or vaccines, both epidemics were ultimately controlled through conventional methods, such as isolating patients and restricting travel., SARS-CoV-2, the virus responsible for COVID-19, is the seventh known coronavirus to affect humans and the third to result in fatal infections. SARS-CoV-2 originated in bats and was first identified in Wuhan, Hubei province, China. In 2019, SARS-CoV-2 was declared a pandemic. In 2022, the World Health Organization (WHO) reported over 110 million confirmed COVID-19 cases and nearly 2.5 million deaths worldwide (World Health Organization website). SARS-CoV-2 spreads much more rapidly than previous coronaviruses (SARS-CoV and MERS-CoV) due to transmission during the early, asymptomatic stage of infection., The SARS-CoV-2 genome has ≈80% identity with SARS-like coronavirus. It is characterized by a 5′ capped and 3′ polyadenylated single stranded, linear, positive sense RNA (ssRNA+). The genome contains 5′ and 3′ untranslated regions and cis-acting secondary RNA structures,, and encodes 29 structural and non-structural proteins, including open reading frame 1a/b (ORF1a/b) polyproteins, spike (S), envelope (E), membrane (M), nucleocapsid (N1), and accessory proteins., SARS-CoV-2 infection requires the binding of the viral S protein with a host cell receptor, Angiotensin-converting enzyme 2 (ACE2). Then, the S protein undergoes conformational changes that facilitate membrane fusion, allowing the viral genome to enter the host. The cleavage of the S protein by host proteases, such as Transmembrane protease serine 2 (TMPRSS2), is crucial for this process as it activates the fusion machinery of the virus. Modulation of TMPRSS2 expression increased or decreased the sensibility to viral infections., However, beyond the productive infection of the respiratory epithelium through this canonical route, the virus is concurrently detected by the lung’s resident immune compartment, thereby initiating a localized immune surveillance response. Alveolar macrophages (AMs) serve as the primary sentinels in the lung, rapidly internalizing SARS-CoV-2 particles upon exposure. Clinical evidence from bronchoalveolar lavage fluid (BALF) of COVID-19 patients frequently reveals viral RNA and protein within these cells, a phenomenon indicative of restrictive uptake rather than productive replication., While this internalization is a fundamental component of the host’s innate defense, in the context of SARS-CoV-2 it often functions as a critical trigger for a dysregulated, hyper-inflammatory trajectory. This initial sensing event catalyzes a self-amplifying secretory program, called the ‘cytokine storm', which drives the severe tissue pathology and morbidity associated with the disease., This secretory program in the macrophage response to SARS-CoV-2 involves a coordinated set of pro- and anti-inflammatory cytokines that drive downstream type I interferon signaling,, antiviral interferons (IFN-β, IFN-γ), interferon-stimulated chemokines like C-X-C motif chemokine ligand 10 (CXCL10), core inflammatory mediators like Tumor Necrosis Factor (TNF-α) and interleukins (IL-1β, IL-6, IL-8, IL-12), and regulators of myeloid function such as Granulocyte-macrophage colony-stimulating factor (GM-CSF), alongside Interleukin-1 Receptor Antagonist (IL-1ra), which provides anti-inflammatory feedback.,,, Infected macrophages do not transition into classical lytic cell death but instead enter a state of ‘abortive pyroptosis’, characterized by a persistent inflammatory program that produces a robust yet pathological output of cytokines, such as IL-6 and CXCL10, which fuel the intense pulmonary inflammation associated with severe COVID-19., Further to its intrinsic pathogenicity, SARS-CoV-2 undergoes rapid evolution and diversification, leading to the emergence of more infectious variants. To date, the WHO has categorized as potential health risks the viral variants D614G, Delta, Omicron BA.1 and BA.5. The D614G mutation involves a single amino acid substitution in the S protein of SARS-CoV-2, where aspartic acid (D) at position 614 is replaced by a glycine (G); the D416G variant quickly supplanted the original Wuhan strain, becoming the predominant form of the virus by mid-2020. The Delta variant was first identified in October 2020 and soon became the predominant variant due to mutations that facilitate viral entry into host cells., The Omicron variant BA.1, first identified in late 2021, is characterized by a large number of mutations associated with altered transmissibility and pathogenicity compared to earlier variants. Notably, Omicron variant BA.5 has mutations, especially in the S protein, that increase its ability to escape immune responses and thus decrease the effectiveness of commercial vaccines. The higher mutation rate of the later variants and the reduced effectiveness of vaccines in immunocompromised patients (who are among those most at risk of death after infection) have pushed researchers to seek alternative solutions. An emerging class of therapeutic agents that have garnered significant attention in recent years for their potential to reduce viral load are Antisense Oligonucleotides (ASOs). ASOs can be designed to bind specific viral RNA sequences (e.g., S protein mRNA) and prevent their translation or trigger degradation. ASOs do not rely on immune activation, so they could be used in vulnerable patients such as immunocompromised patients. The therapeutic efficacy of ASOs is significantly enhanced through various chemical modifications that improve their stability, affinity, and overall pharmacological properties such as phosphorothioate (PS) and Locked Nucleic Acid (LNA) modifications. The former replaces one of the non-bridging oxygen atoms in the phosphodiester group with sulfur; the latter involves the incorporation of a methylene bridge that connects the 2′ oxygen and the 4′ carbon of the ribose sugar, effectively “locking” the nucleic acid in a rigid conformation. Both modifications lead to increased thermal stability and resistance to nucleases and improve the binding affinity of ASOs to their target RNA, facilitating effective gene silencing.,, ASOs that contain modifications only at their 5′ and 3′ ends are called gapmers. Gapmers have shown significant potential in various viral infections, including hepatitis C and SARS-CoV-2, by effectively reducing viral loads and inhibiting replication.,, Despite the promising results obtained by several gapmers against SARS-CoV-2 in reducing viral loads in a mouse model, a clear methodological framework for their design is still missing. Thus, in this work we developed an in-silico workflow incorporating RNA structural data and conservation metrics, providing a robust rationale for gapmer design. We then showed that a high proportion of these gapmers significantly reduce viral RNA levels across multiple SARS-CoV-2 variants. Furthermore, we showed that in infected macrophages gapmers can resolve the persistent dysregulated inflammatory state and promote the transition to a more regulated and physiological secretory profile., Ultimately, our study highlights the promise of RNA-based therapeutics against Coronaviridae or other ssRNA viruses, particularly for vulnerable populations such as immunocompromised individuals. Results Identification of SARS-CoV-2 genomic regions that are both single-stranded and conserved Our strategy for ASO design was based on the structural organization of the SARS-CoV-2 RNA genome, with the aim of identifying structurally accessible target regions that were also evolutionarily conserved. Overall, target prioritization was based on two criteria (A): (i) conservation of structural accessibility, and (ii) sequence conservation across SARS-CoV-2 VoCs and SARS-CoV (A). First, candidate regions were selected based on the genome-wide RNA structural analysis previously reported, where SARS-CoV-2 RNA secondary structure was experimentally mapped using the primer extension and mutational profiling (SHAPE-MaP) (A). Candidate sequences were then checked for complete sequence conservation among VoCs using Clustal Omega implemented in Jalview (). Almost all sequences passed this filter (column F, ), and were then stratified according to sequence conservation relative to SARS-CoV (Column G, ), including 16 regions displaying conservation scores >0.8 between SARS-CoV-2 and SARS-CoV and 8 additional regions with conservation scores >0.6 (). To further refine our target selection, dot-bracket secondary structure annotations were analyzed using custom R scripts. The first script “extractDot” () extracted the dot-bracket segment corresponding to each candidate region and identified consecutive runs of dots, representing unpaired nucleotides. For each candidate region, the longest continuous unpaired stretch was selected as the most structurally accessible target site. The corresponding nucleotide sequence was then processed using a second script (“revComp”), which generated the reverse-complement antisense sequence directed against the selected target (). Using this pipeline, we identified 24 ASO sequences ranging from 18 to 22 nucleotides in length (A, 1B, and ). Overall, the selected target regions have remained remarkably conserved throughout the course of the pandemic, suggesting that they are subject to strong evolutionary constraints and may therefore retain therapeutic relevance for future SARS-CoV-2 variants. Notably, many of these sequences correspond to protein coding regions, such as polymerase, protease, and structural proteins, as illustrated in C. This is expected, since protein-coding regions are generally subject to more severe selection pressure. Gapmers targeting highly conserved SARS-CoV-2 genomic regions effectively reduce viral load in vitro and in cells The reverse complement of the 24 sequences initially selected were chemically synthesized as unmodified antisense DNA oligonucleotides, generating what we call preliminary antisense oligonucleotides, or pre-ASOs (A). Then, we evaluated in vitro whether the pre-ASOs could facilitate RNase H cleavage of the viral RNA (A). RNase H specifically cleaves RNA/DNA hybrids; therefore, the amount of viral RNA remaining after RNAse H digestion is inversely proportional to the efficiency of RNase H-mediated cleavage. We reverse-transcribed (RT) the viral RNA incubated with RNAse H with 2 pre-ASO concentrations, 5 and 10 μM, and performed qPCR spanning the sequences matching the pre-ASOs. In this assay, ten pre-ASOs achieved at least 99% cleavage (A). We then introduced locked nucleic acid (LNA) modifications to these ten pre-ASOs to enhance their bioavailability and stability in a cellular system (). Since ASOs where all nucleotides are LNA do bind complementary RNA with high affinity, but are not good targets for RNAseH, we designed ASOs where LNA is present only at the 5′ and 3′ ends. Moreover, all phosphates in the ASO backbone were modified to phosphorothioates (PS); for the sake of simplicity, we will call these antisense oligonucleotides “gapmers”(A, 1B, and ). Gapmers were previously used by others, with excellent results. We screened our gapmers using A549-hACE2 cells, which are lung carcinoma epithelial cells engineered to overexpress human ACE2, and therefore are permissive to SARS-CoV-2 infection. Cells were transfected with gapmers and then infected with four different SARS-CoV-2 variants, D614G, Delta, Omicron BA.1 (Om BA.1) or Omicron 5 (Om BA.5), at 0.1 multiplicity of infection (MOI). We measured the amount of viral RNA in the cells by performing a qPCR targeting the S protein genomic region. In this assay, 8 out 10 gapmers significantly reduced viral RNA expression (A - left panel). We also measured the amount of virus in the supernatant by performing a qPCR targeting the N1 genomic region (A - right panel). Gapmers 33, 35, 40, and 41 reduced intracellular and supernatant viral RNA by more than 90% for all four variants (A). Similar results were obtained in another cell line, Huh-7.5, with increased permissivity to SARS-CoV-2 infection and replication (B). To visualize the cellular content of viral RNA in cells treated with the gapmers, we performed single molecule FISH (smFISH) with a probe that recognizes the RNA sequence corresponding to the S protein A549-hACE2 cells were treated with gapmers for 12 h (h), infected with the four SARS-CoV-2. variants and incubated for additional 24 h (A). The smFISH images show a reduced level of viral RNA in cells treated with all gapmers (A and A), corroborating the qPCR results. We next evaluated the cytotoxicity of our gapmers. A549-hACE2 cells were transfected with individual gapmers using Lipofectamine 3000 and cell viability was assessed by flow cytometry using Annexin V and Propidium iodide (PI) staining. Live cells (Annexin V and PI double-negative) were scored using a gating strategy based on positive (H2O2-treated cells) and negative controls (gapmer scramble, and Lipofectamine-only) (A). Flow cytometry demonstrated that Lipofectamine, either alone, with SCR gapmers or with most of our gapmers, induces mild cytotoxicity (A). Gapmers 33, 35, and 40 exhibited cytotoxicity in excess of the controls and of the other gapmers, with cell viability below 60% (B and A), suggesting that these gapmers have either additive or synergic cytotoxicity relative to Lipofectamine alone. Taken together, our results confirm that 8 of the designed gapmers effectively reduce SARS-CoV-2 RNA levels across multiple viral variants in a cellular system, highlighting their potential as broad-spectrum antiviral agents. The viral RNA accessibility to gapmers correlates with the decrease in viral replication As mentioned above, the design of the pre-ASOs and of the cognate gapmers was based on the SHAPE-MaP analysis of the RNA secondary structure of SARS-CoV-2/Leiden-0002 (GenBank accession: MT510999); however, no in-silico estimation of the RNA secondary structure of the VoCs was performed. Thus, we performed in vivo SHAPE-MaP of the VoCs, and identified the single nucleotide accessibility as reported in (Delta), (Omicron) and (Omicron 5) for the specific region of interest. As previously described we calculated the average SHAPE reactivity; the results () revealed that the structural accessibility of the selected gapmer target regions is highly conserved across SARS-CoV-2 variants. Pairwise comparisons of SHAPE reactivities measured in Delta, Omicron BA.1 and Omicron BA.5 showed strong correlation (R2 = 0.844–0.945), indicating that the local structural environments of the target regions are largely maintained (A). Furthermore, a statistically significant (p ≤ 0.05) correlation was observed between the SHAPE reactivity of each target region and the corresponding gapmer efficacy, in both intracellular and supernatant assays (B and 3C). Together with the strong conservation of SHAPE reactivities across variants (A), these results indicate that structurally accessible target sites can be identified reproducibly across SARS-CoV-2 variants and represent robust targets for antiviral gapmer design; moreover, gapmer efficacy is maintained across SARS-CoV-2 variants, as expected. These findings experimentally support the hypothesis that RNA structural accessibility is a major determinant of gapmer activity. In conclusion, our pipeline based on RNA secondary structure and sequence conservation has a high rate of success and yields gapmers with robust antiviral activity. Gapmers do not activate interferon responses in uninfected cells and decrease interferon responses in infected cells along with the decrease in viral replication Gapmers can be recognized as nucleic acids released by pathogens and detected by cytosolic DNA recognition pathways, e.g., Toll-like receptors (TLRs), activating an immune response, including the transcriptional program that results in the production of interferons and proteins encoded by interferon stimulated genes (ISG). Thus, the decrease in viral replication obtained with our gapmers might be due to the activation of the interferon pathway, which indeed affects SARS-CoV-2 replication in cells. To better understand the host’s response to SARS-CoV-2 variants and to gapmers, we first checked the expression of a panel of Interferon stimulated genes (ISGs) such as ISG54, ISG15, 2′-5′-oligoadenylate synthetase 1 (OAS1) and Interferon Induced Protein with Tetratricopeptide Repeats 1 (IFIT1), in cells infected for 24h with the four selected variants at 0.1 MOI. qPCR performed on cDNA reverse-transcribed from total cellular RNA showed a significant upregulation of the expression of all tested ISGs compared to the uninfected condition, for all viral variants (A). This shows that all variants elicit a robust antiviral response. Next, cells were transfected with gapmers using Lipofectamine but were not infected with viruses, and retrotranscribed total cellular RNA 36 h after transfection. Treatment with gapmers elicits a low ISGs response, indicating that our gapmers alone do not activate cytosolic DNA recognition and do not induce antiviral pathways (B). In the final experiment, we combined viral infection and treatment: cells were transfected with gapmers using lipofectamine and infected 12 h later with the four selected SARS-CoV-2 variants at 0.1 MOI. qPCR performed on cDNA from total RNA 24 h post-infection (h p.i.) revealed that treatment with gapmers strongly and significantly reduces the activation of ISGs in infected cells (C and A), in some cases to levels comparable to non-infected controls. In brief, our data indicate that gapmers do not activate ISGs, while they decrease the antiviral interferon response of infected cells. Taken together, these data suggest that gapmers cause a decrease in viral replication by directly promoting the destruction of viral RNA, and not indirectly by activating antiviral responses. Naked gapmers delivered in mouse lungs are effectively taken up in alveolar macrophages Since in vivo delivery of gapmers with lipofectamine was impractical due to intrinsic toxicity, we rather tested the delivery of naked gapmers to mouse lungs. To track gapmer uptake, Gapmer −35 was labeled with 5′-fluorescein (FAM-35). Anesthetized mice received an intranasal administration of 25 μL of FAM-35 or PBS (negative control) and were sacrificed 24 h later. FAM fluorescence was found in about 2.5% of lung cells in all 3 mice treated with naked FAM-35 (A). Thus, our gapmers can be taken up in at least in a small fraction of lung cells. It is well established that alveolar macrophages can uptake SARS-CoV-2 particles, and viral RNA/protein is detected in macrophages in BALF from COVID-19 patients; however, these findings indicate viral uptake rather than productive replication in macrophages. Accordingly, we aimed to verify that alveolar macrophages are infected by SARS-CoV-2. Lungs sections from the 3 gapmer–treated mice were then stained with a macrophage marker (F4/80) and imaged under the confocal microscope. Across lung sections from both PBS- and gapmer-FAM–treated mice, 341 out of 12,852 total cells (2.6%) were positive for F4/80, a fraction notably similar to the fraction of uninfected cells transfected by naked gapmers (B). In lungs from gapmer-FAM–treated mice (lower row), a clear FAM fluorescence signal frequently colocalized with F4/80 staining: 91 out of 97 FAM-positive cells (94%) were F4/80 positive (Venn diagram, B). As expected, lungs from PBS-treated mice (upper row) displayed no detectable FAM signal and, consequently, no FAM–macrophage colocalization. Although gapmer uptake was highly enriched in macrophages, a small fraction of FAM-positive cells lacked F4/80 expression, suggesting that a minor amount of gapmer is internalized by non-macrophage lung cells; however, the identity of these cells was not investigated in the present study. Collectively, these data demonstrate that naked gapmers are effectively taken up by alveolar macrophages, while their access to other cells is limited. Building on this observation, we took the opportunity to test whether gapmer treatment modulates the macrophage inflammatory programs associated with the cytokine storm characteristic of severe COVID-19. To define the baseline response of human macrophages, we analyzed cytokine production in human monocyte-derived macrophages (MDMs) differentiated from peripheral blood mononuclear cells (PBMCs) of healthy donors and infected with SARS-CoV-2 at MOI of 10. As shown in A and A, we assessed cytokine profiles in the supernatant at 24 and 48 h p.i. The results indicated that macrophages responded to viral exposure by releasing a specific set of inflammatory mediators: IL-6 levels were significantly elevated at both 24 and 48 h p.i., while the production of IL-12, IFN-γ, and GM-CSF increased significantly only at 48 h p.i. compared to uninfected controls. In contrast, IL-8, IFN-β, IL-1β, and TNF-α levels remained stable or showed only marginal increases. Notably, viral exposure also induced the secretion of the anti-inflammatory mediator IL-1Ra at 48 h p.i., alongside robust CXCL10 production at both time points. This profile is consistent with an abortive or restrictive infection model, where viral RNA triggers a subset of inflammatory pathways without achieving full productive replication. We next evaluated whether Gapmer-35 treatment could reprogram this inflammatory signature. As shown in B and B, control experiments using scramble (SCR)-treated macrophages showed no deviation from the untreated infection profile, confirming that non-specific oligonucleotide exposure does not inherently modulate macrophage activation in this model. In contrast, treatment with gapmer-35 significantly reconfigured the cytokine secretome: while unchallenged infection was characterized by a robust and potentially pathogenic IL-6 response, Gapmer-35 treatment led to a significant decrease in IL-6 levels. Concurrently, we observed a compensatory increase in both IL-1β and its regulatory counterpart IL-1Ra, alongside elevated TNF-α and IL-8 production. The decrease in IL-6, a primary driver of COVID-19-related lung injury, combined with the modulation of cytokines of the IL-1 family, suggests a shift toward a more regulated immune state. Taken together, these data indicate that a representative gapmer selectively reprograms the inflammatory responses of macrophages infected by SARS-CoV-2, rather than broadly suppress immune responses. Discussion The COVID-19 pandemic and the successive emergence of numerous mutated variant strains of SARS-CoV-2 have highlighted the need for effective therapies targeting multiple viral variants. RNA-based therapeutic strategies have spawned several promising candidate drug classes, including ASOs. Our study meticulously explored the design, evaluation, and optimization of gapmers targeting SARS-CoV-2 RNA to provide a methodological pipeline for potential in vivo translation. The identification of suitable gapmer targets within the SARS-CoV-2 genome required a dual focus on structural accessibility and sequence conservation. RNA secondary structure predictions based on SHAPE-MaP analyses served as the foundation for selecting single-stranded regions. These regions are essential for viral replication and protein translation and are therefore likely to be conserved across variants of concern. An initial list of 80 single-stranded regions was narrowed to 24 sequences with a high degree of sequence identity across VoCs including D614G, Delta, Omicron BA.1 and Omicron BA.5. Notably, many of the selected regions encode essential viral proteins, such as the RNA-dependent RNA polymerase (RdRp), N1, and S proteins, but do not match specific regions that have a high mutational rate, for example the receptor binding domain in subunit 1 of the S protein. This approach yielded sequences that are both single-stranded in the SARS-CoV-2 RNA genome, and therefore accessible to single-stranded complementary oligonucleotides, and highly conserved in different variants, which suggests that they cannot mutate easily. Initial in vitro screening revealed that 10 out of 24 pre-ASOs initially selected exhibited >99% efficiency in promoting viral RNA cleavage via RNase H. We then converted pre-ASOs into gapmers to improve their stability, binding affinity and resistance to nuclease degradation., Our design included the substitution of phosphorothioates for phosphates in the oligo chain, and the modification to LNA of the 5′ and 3′ ends of the pre-ASOs. Notably, four gapmers (33, 35, 40, 41) demonstrated consistent efficacy across multiple SARS-CoV-2 VoCs in two different cell lines (A549-hACE2 and Huh-7.5). These results highlight the robustness of our design strategy for designing gapmers with broad-spectrum antiviral activity. The main objective of this study was to experimentally validate the predicted structural accessibility of gapmer target regions in SARS-CoV-2 RNA. Our analysis adopted a targeted SHAPE-MaP approach to interrogate the structural state of the candidate gapmer target regions across multiple SARS-CoV-2 variants. Our data confirmed that these target regions are predominantly single-stranded. A significant correlation between SHAPE reactivity and gapmer efficacy highlights the role of structural accessibility of the viral genome, which can be predicted in silico but also includes RNA-protein interactions and sequence context effects that cannot be easily predicted. Our findings support prior evidence that gapmer performance is shaped by both target properties and cellular environment. Interestingly, the four gapmers with higher efficacy were all contiguous over the genomic region that encodes the structural proteins (C). This collinear configuration has a statistically significant probability of p = 7/210 = 0.03 (7 is the number of positions of 4 contiguous segments in a row of 10, and 210 is the number of arrangements containing 4 contiguous segments). Whether this is a statistical fluctuation or has a biological reason is impossible to determine from our data. Thus, the possibility that effective gapmers may consistently target the sequence coding for structural proteins in Coronaviruses should be further evaluated in future studies. The 6 gapmers with lower antiviral efficacy did not target poorly conserved sequences, but rather regions with less favorable RNA structural organization. In particular, these gapmers targeted regions partially constrained by nearby stem structures, which likely reduce the accessibility of the target sequence to gapmer binding and RNase H recruitment. Recent studies demonstrated that gapmer:RNA stability depends not only on sequence complementarity, but also on the dynamic accessibility and structural organization of the target RNA region. Favorable target architecture may combine high local accessibility with structural stability of the surrounding RNA context, whereas completely unstructured or highly dynamic regions may not maintain a sufficiently stable gapmer:RNA duplex for efficient RNase H activity. Importantly, the correlation between SHAPE reactivity and gapmer efficacy across multiple SARS-CoV-2 variants suggests that structural accessibility may represent a conserved functional feature of the viral genome. This indicates that conservation of RNA structure, rather than primary sequence alone, could serve as a robust predictor for broad-spectrum gapmer susceptibility against ssRNA virus. A possible confounding effect in the effectiveness of gapmer is the activation of cellular antiviral responses, especially because in our experiments gapmers were transfected prior to infection. However, gapmer transfection of uninfected cells showed no activation of ISGs, which is the hallmark of antiviral responses. Indeed, other gapmers like Mipomersen (Kynamro) and Volanesorsen (Waylivra) don’t show significant systemic IFN activation in humans. Infection of cells with SARS-CoV-2 variants significantly upregulated ISGs, indicating the expected robust innate immune activation in infected cells. In cells pretreated with gapmers before infection, ISG expression was much lower than in cells not transfected with gapmers, suggesting that the reduced viral replication and RNA expression leads to a lower host cell response. Mechanistically, this suggests that the antiviral activity of the gapmers depends on the destruction of the viral genome and not on the activation of the host antiviral response. In terms of the therapeutic effects of gapmers against SARS-CoV-2, reducing the activation of IFN pathways might be beneficial, as the patients most affected by SARS-CoV-2 infection show too much IFN and cytokine signaling, leading to a so-called cytokine storm. The cytokine profile induced by SARS-CoV-2 in human monocyte-derived macrophages (MDMs) reflects a dysregulated inflammatory program consistent with abortive or incomplete pyroptotic signaling., Although IL-6 and CXCL10 are robustly induced, indicating effective viral RNA sensing, the minimal modulation of IFN-β, IL-1β, and TNF-α suggests only partial engagement of antiviral and inflammasome pathways, even at high MOI. This pattern contrasts with early reports proposing productive macrophage infection and broad cytokine induction, including IFN-β., Subsequent studies, in fact, demonstrated that primary human MDMs are largely refractory to productive replication supporting a model of restricted or abortive inflammatory activation., The limited IL-1β maturation, together with delayed IL-1Ra induction, further indicates incomplete inflammasome activation, favoring DAMP release over a canonical IL-1–dominated response. Overall, SARS-CoV-2 appears to initiate early pyroptotic programming in macrophages while attenuating full inflammasome output, a phenotype potentially amplified under high viral exposure. Gapmer-35 treatment does not broadly suppress but selectively reshapes the inflammatory response of infected macrophages. The marked decrease in IL-6, a key mediator of COVID-19-associated hyperinflammation, indicates targeted attenuation of the pathological NF-κB axis. Concurrently, the increase in IL-1β and IL-1Ra suggests a more coordinated engagement of inflammasome pathways, shifting macrophages from abortive toward more regulated inflammatory signaling. Importantly, preserved levels of CXCL10, IFN-β, and GM-CSF confirm intact Retinoic acid-inducible gene I - Melanoma differentiation-associated protein 5 (RIG-I/MDA5)-dependent sensing. Together with our observation that gapmers reduce ISG expression by lowering viral burden rather than stimulating host antiviral pathways, the observation that gapmers rebalance macrophage inflammation suggests that gapmers may represent a promising strategy to mitigate macrophage-driven immunopathology in severe COVID-19. An important question for future studies concerns the emergence of viral resistance to gapmer treatment. The design strategy described here combines two features that are expected to limit the development of escape variants: structural accessibility and sequence conservation. By targeting RNA regions that are subject to strong evolutionary constraints, gapmers may induce less resistance-associated mutations than small antiviral molecules. However, direct testing of this hypothesis requires specific in vitro approaches and should then be confirmed in clinical trials. A critical hurdle in therapeutic development of gapmers does remain: their effective delivery into airway and lung cells. Our in vivo tests showed that naked gapmers can reach murine lung cells via intranasal delivery, but apart from macrophages the number of transfected cells was low. We confirmed significant cytotoxicity of lipofectamine, which is widely recognized as inadequate for in vivo delivery., We also tested polymers that have been developed for inhalation of small interfering RNA (siRNAs), albeit with unsatisfactory results (not shown). Thus, for therapeutic impact, gapmer delivery to the lung has to be developed. Overall, this study provides valuable insights into the design and optimization of gapmers as a therapeutic strategy against SARS-CoV-2 and by extension any other ssRNA virus especially for immunodeficient patients. Most notably, our study indicates that the accessibility of conserved single-stranded regions in genomic viral RNA is indeed the primary, albeit not sole, predictor of gapmer efficacy. Targeting those regions is a fast and effective approach that can be tested in cell cultures, with minimal delay between the identification of an emerging virus and the production of gapmers against it. Amid this promising landscape, there is still a prominent bottleneck: the lack of specific delivery systems that are specific or at least effective for inhaled gapmers. Indeed, our results provide a compelling motivation to invest in their development. Limitations of the study This study has several limitations. Gapmer efficacy was primarily assessed in two immortalized cell lines (A549-hACE2 and Huh-7.5) rather than in primary human airway epithelial cultures, which may not fully recapitulate the complexity of the respiratory epithelium in vivo. Our screen covered four SARS-CoV-2 variants of concern, and whether the selected target regions remain accessible and conserved in variants that emerge in the future remains to be tested. The in vivo experiments in mice demonstrated that naked gapmers can reach lung tissue and are preferentially taken up by alveolar macrophages following intranasal administration, but transfection efficiency in other lung cell types was low, and these studies assessed biodistribution rather than antiviral efficacy or therapeutic outcome in infected animals; an effective, non-toxic pulmonary delivery system for gapmers therefore remains to be developed. The apparent clustering of the most effective gapmers within the genomic region encoding structural proteins reached only borderline statistical significance, and its biological basis, if any, was not established. Finally, the potential for the virus to develop resistance to gapmer treatment through target-site mutation was not experimentally tested and will need to be addressed in future studies. Resource availability Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Marco E. Bianchi (bianchi.marco@hsr.it). Materials availability All unique/stable reagents generated in this study are available from the without restriction. Data and code availability - •Data: SHAPE-MaP raw sequencing data have been deposited at the National Center for Biotechnology Information (NCBI) Sequence Read Archive under accession number PRJNA1451745, as listed in the . - •Code: The original scripts for identification of accessible RNA regions and ASO design is reported in and described in the . - •Other items: Any additional information required to reanalyze the data reported in this paper is available from the upon request. Declaration of interests The authors declare no competing interests. Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this work, the authors used ChatGPT in order to improve language and readability. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication. STAR★Methods Key resources table Experimental model and study participant details Ethics statement Human PBMCs were obtained from the Biological Resource Center (Centro Risorse Biologiche, CRB) of the IRCCS Ospedale San Raffaele (Milan, Italy). PBMC were isolated from blood samples obtained from 3 healthy donors provided written informed consent for the collection and use of biological specimens for research purposes, in accordance with the Declaration of Helsinki, The biobanking protocol was approved by the Institutional Review Board/Ethics Committee of the IRCCS Ospedale San Raffaele under the protocol BBMRI-ERIC ID: IT_1383758011993577. A fully anonymized form to protect donor privacy, in compliance with institutional review board guidelines and GDPR regulations. Consequently, demographic characteristics such as age, sex/gender, and ancestry were unavailable to the investigators, and sex/gender-based stratification could not be performed. Experiments with mice were performed with hyACE2 knock-in mice. Six hyACE2 C57BL/6 female at 8–10 weeks of age were used. All experimental procedures were approved by Italian Ministry of Health (authorization 270/2022-PR). All other experiments were performed using established cell lines (A549-hACE2, HuH-7.5, VERO-E6-TMPRSS2) obtained from commercial or academic sources. Cell lines obtained from commercial suppliers were authenticated and routinely tested for mycoplasma contamination by the providers. Cell lines received from collaborating laboratories were authenticated and routinely tested for mycoplasma contamination by the respective laboratories. In addition, all cell lines were weekly tested for mycoplasma contamination in our laboratory. Cell culture The A549-hACE2 and HuH-7.5 cell lines were purchased from NCBI (ATCC number: CCL-185) and Japanese Collection of Research Bioresources (JCRB) Cell Bank, Osaka, Japan (Ref: JCRB0403), respectively. The former were cultured in RPMI 1640 Medium (Gibco, ref. 21875-034), supplemented with 10% fetal bovine serum (FBS, Gibco), 1% Penicillin-Streptomycin (Pen/Strep; Gibco, ref. 15140-122), 1% L-Glutamine 200 mM (Gibco, ref. 25030-024), 1% Non-Essential Amino Acids (MEM NEAA; Gibco, ref. 11140-050), 1% Sodium Pyruvate (Gibco, ref. 11360-070). The latter in Dulbecco’s modified Eagle’s medium (DMEM + GlutaMAX ; Gibco, ref. 61965-026) supplemented with 10% fetal bovine serum 1% Pen/Strep, 1% of MEM NEAA, and 1% Sodium Pyruvate. All cells were cultured at 37°C and 95–99% humidity, in an atmosphere containing 5% CO2. Mice Mice had ad libitum access to drinking water and chow (VRF1 standard diet, Safe, 801900). Mice were housed under specific pathogen-free conditions with a 12-h light/12-h dark cycle, a temperature ranging between 20 °C and 23 °C and 60% humidity. Method details Cell treatment HuH-7.5 or A549-hACE2 cells were plated in complete medium. After 24h cells were treated with 100 nM Gapmers and LIPOFECTAMINE 3000 according to manufacturer’s protocol. Infection: 12h after treatment the medium was replaced with DMEM supplemented with 2% FBS and the cells were infected with 0.1 MOI SARS-CoV-2 variants D614G, Delta, Omicron and Omicron 5 according. MOI was calculated as follows: (MOI x no. cell plated)/(viral concentration PFU/ML) x 1000. Virus preparation VERO-E6-TMPRSS2 cells were purchased from NCBI (ATCC number: CRL-1587). Cells were cultured in DMEM, supplemented with 2% FBS,1% Pen/Strep, 1% MEM NEAA, 1% Sodium Pyruvate. After 24 h, cells were inoculated with 0.001 or 0.01 MOI of SARS-CoV-2 variant D614G (GISAID ID: EPI_ISL_584051), SARS-CoV-2 variant Delta B.1.617.2 (GISAID ID: EPI_ISL_3073880), SARS-CoV-2 variant Omicron BA.1 (GISAID ID: EPI_ISL_10898045) and SARS-CoV-2 Omicron subvariant BA.5 (GISAID ID: EPI_ISL_14036188). The supernatant was collected after 48–72h; 200 μL were kept for titration and the rest centrifuged at 5000 g for 5 min. The pellet was resuspended in an appropriate volume, aliquoted and stored at −80°C. Virus stocks were titrated by plaque assay using Vero E6-TMPRSS as described in. Viral RNA extraction SARS-CoV-2 RNA (D614G variant) was extracted from the supernatant of VERO-E6 cells infected at MOI 0.1. The supernatant was harvested 48 h p.i. and centrifuged to remove debris. The supernatant was mixed with 5X PEG-it virus precipitation solution (System Bioscience, Cat. N. LV810A-1), vortexed and incubated overnight at 4°C. The supernatant/PEG-it mixture was centrifuged, transferred to a fresh tube and centrifuged again. The samples were lysed with TRIzol reagent (Invitrogen, Cat. N. 15596026) for RNA extraction, resuspend in RNAse-free water and stored at −20°C. RNA was quantified using a NanoDrop spectrophotometer, obtaining a concentration of 25 ng/μL. Samples were stored at −80°C prior to use. All experiments that involved active virus handling were carried out in the Bio Safety Level 3 (BLS-3) room in INGM. In vitro RNA folding 500 ng of viral RNA were denatured at 95°C for 2 min, followed by rapid cooling on ice for 1 min, followed by incubation with folding buffer containing 50 μM HEPES pH 7.5, 100 mM KCl at 37°C for 15 min 10 mM MgCl2 was added, and the sample was incubated for an additional 15 min at 37°C. Subsequently the folded RNA was hybridised with ASOs at varying final concentrations (0.1 μM, 1 μM, 5 μM, 10 μM) by incubating at 37°C for 15 min. The ASO-RNA solution was incubated with 2.5U of RNaseH (Thermo Fisher, Cat. N. EN0201) diluted in 20 mM Tris-HCl pH 7.5, 100 mM KCl, 10 mM MgCl2, 0.1 mM DTT, 4.7% sucrose. The digested RNA was reverse transcribed into cDNA using the SSIV Reverse Transcriptase (Invitrogen, ref. 18090050) according to the manufacturer’s protocol. Total RNA extraction 24 h p.i. the supernatants from each condition were mixed ViRNAExTM solution (Metacell s.r.o. Ref. 111560) according to the protocol and cells were lysed with TRIzol reagent for RNA extraction. The samples were stored at −20°C until RNA extraction, performed following the manufacturer’s protocol. After collection of the aqueous phase, an equal volume of 98% ethanol was added and the RNA cleanup was per-formed using the RNA Clean & Concentrator kit (Zymo Research; Ref: R1015) following the manufacturer’s protocol. RNA was eluted in 12 μL of DNase/RNAse-free water and quantified using a NanoDrop spectrophotometer. qPCR cDNA was diluted 1:150 in RNase-free water. qPCR was performed on RTcDNA using the primers reported in Table 5. The reaction mix was prepared according to LightCycler 480 SYBR Green I master (Roche Diagnostic, ref. 04887352001) or TaqPathTM Master Mix (Thermo Fisher Scientific; Ref: A15299) following the manufacturers’ protocols. The reactions were carried out using the Light-Cycler480 Instrument II (Roche) or QuantStudio 5 real-time PCR systems, respectively. Cycle thresholds (Cts) were normalised over the housekeeping gene (GAPDH) and ΔΔCt values were calculated relative to untreated controls. Cts obtained from extracellular RNA gapmers treatment were normalised over the non-treated sample. Macrophages isolation, infection and analysis Buffy coats were obtained from anonymous healthy donors. Human peripheral blood mononuclear cells (PBMCs) were isolated via Ficoll-Paque density gradient centrifugation, and CD14+ monocytes were subsequently purified through positive immunomagnetic selection using CD14 MicroBeads (Miltenyi Biotec, ref. 130–050-201) according to the manufacturer’s instructions. The isolated CD14+ monocytes were seeded in 24-well plates and cultured in RPMI 1640 medium (Gibco) supplemented with 10% fetal bovine serum (FBS; Gibco), 1% penicillin-streptomycin, 1% L-glutamine (200 mM), 1% non-essential amino acids, and 1% sodium pyruvate. Macrophage differentiation was induced by the addition of human Macrophage Colony Stimulating Factor (M-CSF; PeproTech, ref. 300-25) at a concentration of 20 ng/mL. After 6 days of differentiation, the M-CSF stimulus was removed, and the resulting macrophages were treated with 2.5 μM of Gapmer-35, a scrambled Gapmer control (SCR), or PBS. Following a 12-h incubation, the macrophages were infected with the SARS-CoV-2 Delta variant at a MOI of 10. Cell culture supernatants were harvested at 24 and 48 h p.i. Cytokine and chemokine profiles in the supernatants were characterized using a custom Human Magnetic Luminex Assay (R&D Systems, Ref: LXSAHM) according to the manufacturer’s protocol and analyzed on a Luminex 200 system. Imaging A549 cells (5x104) were plated onto 15 mm glass coverslips in a 24-well plate and treated/infected as reported above. After 36h medium was removed, and cells were washed twice with PBS. Subsequently, cells were fixed with 4% PFA for 15 min at room temperature (RT). For the mice lung section: 100 μM FAM-35 or PBS were intranasally administered into mice, under gaseous anesthesia (2–5% isoflurane). After 24h, the mice were sacrificed and lungs were collected. The lungs were left in 4% PFA overnight, washed with PBS, and incubated in 30% sucrose for at least 8h. After incubation, the lungs were embedded in OCT and stored at −80°C. The lungs were sliced into 20 μm sections and placed on glass slides. Lung slides were blocked 1 h at RT in 10% Goat Serum/PBS 1X and stained ON at 4°C with F4/80 primary antibody (Thermo Fisher Scientific, Ref: MA5-16630) diluted 1:100 in 2% Goat Serum/PBS 1X, followed by secondary fluorescent antibody. Three washes with PBS were performed after each antibody’s incubation. Finally, glass slides were incubated for 20 min at RT with 1 μg/mL DAPI in PBS. After rinsing with PBS, the glass slides were mounted with coverslips using mounting media (Merck; ref. 345789). SmFISH imaging A549 cells were plated in 24-well plates on glass coverslips and treatment/infection was performed as described above. SmFISH was performed as previously described. Briefly, cells were washed twice with PBS and fixed with 4% PFA for 15 min at RT. After two PBS washes, cells were permeabilised ON with 70% ice-cold ethanol at −20°C. Cells were then equilibrated at RT and washed twice with 2x SSC buffer for 5 min each. A single wash with 20% formamide in SSC 2x was performed for 5 min at room temperature. Coverslips were then incubated with Hybridization buffer (HB - 100 mg/mL Dextran Sulfate, 20% formamide in SSC 2x), containing the smFISH probe that targets the viral spike protein in a humidified chamber overnight at 37°C in the dark. The next day, coverslips were washed twice with 20% formamide in SSC 2x for 30 min each in the dark at RT. Nuclei were then stained 10 min with Hoechst solution at RT. Coverslips were washed twice with PBS and once with ultrapure water. Coverslips were mounted on microscope slides with mounting media and stored at 4°C at least overnight before imaging. Image detection and analysis Images were acquired, at least 2h after mounting, using an SP8 inverted confocal microscope (Leica Camera). Nuclei snaps were segmented in Fiji. Thresholding of the segmented nuclei masks was performed and watershed, then automatically counted in Fiji using the “Analyze Particles” function. The number of FAM-positive nuclei was expressed relative to the total number of nuclei within each field of view (FOV). Flow cytometer-based toxicity test Cells were seeded in 24-well plates and toxicity of Lipofectamine-transfected gapmers was determined by flow cytometry. 36 h after treatment, cells and supernatant were collected in FACS tubes, trypsinized and centrifuged at 300 g for 5 min, washed twice with PBS, and resuspended in 100 μL of Annexin-V buffer (10 mM HEPES, pH 7.4, 140 mM NaCl, 2.5 mM CaCl2 in PBS). Each tube was stained with 5 μL of APC Annexin-V (BioLegend, ref. 640920) and 5 μL of Propidium Iodide (Sigma-Aldrich; Ref: P4864). The controls included unstained and fluorescence minus one (FMO) performed on untreated and H2O2-treated samples (positive control: 20 mM H2O2 - Sigma-Aldrich; ref. 216763). The tubes were incubated for 15 min at room temperature in the dark. The samples were analyzed using a BD FACSCanto II (PI was read with a 488 nm laser and APC was read with a 633 nm laser) and normalised against the controls. SHAPE-MaP In vivo SHAPE-MaP was performed as previously described, with the exception that, unlike the genome-wide SHAPE-MaP analysis reported by Manfredonia et al., the present study focused exclusively on the genomic regions selected for gapmer design (column A, ). Accordingly, primer pairs were specifically designed to amplify only these target regions (). VERO-E6-TMPRSS2 cells, infected at 1.5 MOI, were treated with NAI from a 1 M stock solution (in DMSO) to a final concentration of 100 mM, while an equivalent volume of DMSO was added to control samples. For these experiments, NAI, synthetized as previously described in Spitale et al., was used instead of the more reactive 2A3 reagent to enable direct comparison with the Manfredonia SHAPE-MaP dataset. Cells were incubated at 37°C with constant shaking for 5 min, after which the reaction was quenched by the addition of DTT to a final concentration of 0.5 M. Samples were centrifuged at 10,000 × g for 1 min, the supernatant was discarded, and cell pellets were resuspended in 1 mL TRIzol. Total RNA was extracted according to the manufacturer’s instructions. Total RNA (3 μg in a final volume of 10 μL) was supplemented with 1 μL dNTP mix (10 mM each) and 1 μL of a 50 μM pooled forward primer mix (), containing the 8 forward library primers for all selected target regions. Samples were incubated at 70°C for 5 min and immediately transferred to ice for 1 min. Reverse transcription was performed in a final 20 μL reaction volume, using 200 U SuperScript II (Invitrogen; ref. 18064022), in the presence of 50 mM Tris-HCl pH 8, 75 mM KCl and 6 mM MnCl2, as previously described. Reactions were incubated at 42°C for 90 min, followed by 10 min at 50°C, 10 min at 55°C and 10 min at 60°C. Eight independent PCR reactions were performed for each sample, each using a specific forward and reverse primer pair corresponding to one target region (). Briefly, target regions were PCR amplified using OneTaq DNA Polymerase (NEB; Ref. M0480) and a touchdown approach, by first performing 10 cycles, starting at the optimal annealing temperature, and decreasing it by −0.5°C/cycle, followed by 20 cycles at the lowest temperature (ideal annealing temperature −5°C). The resulting amplicons were purified using NGS NucleoMag Beads (Macherey-Nagel; ref. 744970.50) at a 1:1 bead-to-sample ratio and eluted in 15 μL RNase-free water. Purified PCR products were verified on a 2% agarose gel, and each primer pair yielded a single amplicon of the expected size. The eight amplicons obtained from each sample were then pooled in equimolar amounts to a total of 50 ng DNA in a final volume of 20 μL. Libraries were prepared by blunting and dA-tailing the amplicons using the End Prep library kit (NEB; Ref: E7546) followed by NEBNext Ultra II Ligation Module kit (NEB, Ref: E7595) with the adapters obtained from NEBNext Multiplex Oligos for Illumina (NEB, Ref:E7600) following standard protocols. Reads obtained by SHAPE-MaP analysis were analyzed following the standard procedure. Analysis of SHAPE-MaP data SHAPE-MaP data were analyzed using the RNA Framework suite. Briefly, reads were mapped to the SARS-CoV-2 genome reference for each specific variant (Delta B.1.617.2 (GISAID ID: EPI_ISL_3073880), SARS-CoV-2 variant Omicron BA.1 (GISAID ID: EPI_ISL_10898045) and SARS-CoV-2 Omicron subvariant BA.5 (GISAID ID: EPI_ISL_14036188)) using the rf-map module and Bowtie2 v2.3.5.1. Mutations were then counted using the rf-count module (parameters: -m -rd), and raw reactivities normalized using the rf-norm module (parameters: -n 1000 -sm 3 -nm 3). Normalised SHAPE reactivities are reported in File 2 (Delta), File 3 (Omicron BA.1) and File 4 (Omicron BA.5). Each file contains the nucleotide sequence together with the corresponding normalized SHAPE reactivity value for each nucleotide. Reactivity values close to 0 indicate low structural accessibility, whereas higher values indicate increasing nucleotide flexibility and accessibility. In the present study, SHAPE reactivities were classified as low (<0.4), intermediate (0.4–0.7), or high (>0.7). Positions reported as NaN correspond to genomic regions that were not covered by the targeted SHAPE-MaP analysis and therefore were not experimentally analyzed. Software Immunofluorescence images were processed and analyzed using Fiji. Data analysis and graph generation were performed using Microsoft Excel and GraphPad Prism 6.2. RNA structure analysis and visualization were carried out using RStudio, VARNA, and Jalview. The viral genome architecture presented in C was created using BioRender. Quantification and statistical analysis Linear mixed effects models (LMEs) were used to evaluate differences in Spike, Nucleocapsid or ISGs expression. Following this, post-hoc pairwise comparisons were conducted by first estimating marginal means, essentially, the sum of coefficients derived from the LME model. Then, the differences in marginal means between groups were divided by the standard error to calculate the t statistic and P-values were adjusted using the Tukey’s HSD method. The analysis was performed in R using lme4, lmerTest and emmeans packages. p > 0.05 is considered not significant (ns); p ≤ 0.05 is considered significantly different. Statistical significance of SHAPE-MaP reactivities was calculated using a two-tailed Pearson’s correlation coefficient. Statistical significance for the correlation between SHAPE-MaP reactivity and efficacy of the gapmers was calculated using two-way ANOVA. Statistical significance of cytokine profiles, measured by Luminex assay, of supernatants from uninfected or infected macrophages was calculated using two-sided t test versus the relevant control. Footnotes Contributor Information Alessio De Magis, Email: demagis.alessio@hsr.it. Marco E. Bianchi, Email: bianchi.marco@hsr.it. Supplemental information