Co-existence and co-infection of influenza A viruses and coronaviruses: Public health challenges

Phylogenetic dynamics for IAVs and SARS-CoV-2

IAVs (eg, H1, H3, H5, and H7) have experienced rapid evolution and established divergent lineages and genotypes. Phylogenies of hemagglutinin (HA) genes of seasonal influenza H1N1 and H3N2 viruses exhibit ladder-like tree topologies and constant genetic diversity within lineages across time, and some lineages persist from one influenza season through to the next years.58, 59, 60 Given the genetic divergence and phylogeny relationship of HA genes, global H1N1 and H3N2 viruses can be divided into different clades (Figure 2).^61,62 Of multiple co-existent clades, H1N1 clade A5a and H3N2 clade A1b are the currently dominant clades circulating the world (Figure 2; https://nextstrain.org/flu/seasonal/).^62,63 Regarding H5, H7, and H9 AIVs of public health concern, H5 influenza was designated into multiple clades by the WHO/OIE/FAO H5N1 Evolution Working Group according to the phylogenetic topology and genetic divergence of HA genes.⁶⁴ Two main lineages of novel H7N9 have been established based on the HA phylogeny relationship, the Yangtze River Delta lineage and the Pearl River Delta lineage, since its emergence in 2013.^65,66 At least three HA clades of H9 AIVs are co-circulating among poultry.³⁹ Moreover, new subclades or lineages are gradually emerging as the evolution of H5, H7, and H9 AIVs.^38,39

A dynamic nomenclature system has been proposed for the expanding phylogenetically divergent SARS-CoV-2 viruses (https://cov-lineages.org/lineage_list.html).⁶⁷ The phylogeny of global SARS-CoV-2 can be grouped into two main lineages: lineage A and lineage B. Lineage A is a minor group and shares two nucleotides, at 8782 nt of ORF1ab and 28144 nt of ORF8, with the genetically close SARS-CoV-2-related bat CoVs RaTG13 and RmYN02.^13,47 Lineage B includes the currently persistent and dominant SARS-CoV-2 variants. Based on the increased risk of SARS-CoV-2 variants to public health and corresponding biological and clinical features, WHO designated the variants of concern (VOC) and variants of interest (VOI) using the Greek alphabet (https://www.who.int/en/activities/tracking-SARS-CoV-2-variants/). On July 14, 2022, the VOC group includes Omicron (B.1.1.529) SARS-CoV-2 variants, and previously circulating VOCs, Alpha (B.1.1.7), Beta (B.1.351), Gamma (P.1), and Delta (B.1.617.2), have been removed from this group. The VOCs and VOIs can be reclassified given the circulation, epidemiological situation, and biological properties of corresponding variants. In addition, the within host diversity of SARS-CoV-2 is relatively low, with narrow transmission bottlenecks at high viral loads (at least at the early infection), but the transmitted SARS-CoV-2 variant could spread rapidly.⁶⁸

The IAV and SARS-CoV-2 underwent genetic diversity and relatively rapid evolution dynamics. The partial cross-immunity and the acute and short infectious period, to a large extent, facilitate the evolutionary dynamics and seasonal epidemics of human H1N1 and H3N2 IAVs.⁶⁹ Of note, the vaccine breakthrough infection and reinfection with SARS-CoV-2^70,71 mean that vaccine-induced and natural immunity are not enough to protect humans from virus infection, especially for the current variants, suggesting probably long-term co-circulation of SARS-CoV-2 with seasonal influenza viruses. In addition, the rapid evolution with antigenic changes and global persistence of seasonal influenza require the intensive surveillance of influenza activity and formulation of well-matched vaccines before each annual influenza season.⁷² These valuable lessons and experiences should be learned to control the current pandemic and possibly annual SARS-CoV-2 epidemics in the future.

Genetic reassortment and recombination in IAVs and HCoVs

Virus ecology affects the genetic evolution of IAVs and CoVs. When co-infection of multiple IAVs or CoVs happens in the same host, genetic reassortment among IAVs and recombination among CoVs potentially occur and further facilitate virus evolution and novel variant emergence. To our knowledge, the genetic interactions between IAVs and HCoVs have not yet been documented, while the potential recombination between the two types of viruses might also occur during their co-infections in one host.

IAV is an enveloped virus with a negative-sense, RNA-segmented genome that can encode for more than 17 proteins.⁷³ The segmented genome drives the exchange of gene segments between IAVs (genetic reassortment) when they simultaneously infect the same host or cell.⁷⁴ Reassortment facilitates the formation of novel influenza variants with new genomic constellations. Moreover, the reassortant virus could obtain fitness advantage, cross-species transmission, evasion from host immune responses, and even cause pandemics/epidemics in humans.¹⁶ Since the 20^th century, at least three of four influenza pandemics were caused by reassortants: the 1957/H2N2 virus (HA, NA, and PB1 from AIV, the other five genes from human IAV), the 1968/H3N2 virus (HA and PB1 from AIV, the other six genes from human IAV), and the 2009/H1N1 virus (PB2 and PA from AIV, PB1 from human IAV, and others from swine IAV). Notably, the novel H7N9 AIVs also emerged by reassortment in 2013 and have caused five infection waves in humans.^65,75 The internal genes of H7N9 originate from H9N2 AIVs that adapted well in chickens and H7 and N9 genes from viruses found in aquatic birds (Figure 3). Later, the novel H7N9 strain evolved into more genotypes by further reassortments with diverse H9N2 variants and other AIV subtypes.^66,76

CoV is also an enveloped virus but carries a large, positive-sense, single-stranded RNA genome.⁷⁷ The common mutations and recombination in the positive-strand RNA viruses with the largest genome contribute to genetic divergence and novel CoV variant emergence.^9,78,79 Following co-infection with more than one CoVs, recombination may occur during virus replication when multiple subgenomic RNAs are generated, and genetically related genes are readily recombinant among different CoVs by template switching.⁸⁰ Genetic recombination has been reported in human and animal CoVs, but the recombination breakpoints are commonly random.^9,12,81,82 In addition, the novel recombinant CoVs could lead to cross-species transmission and outbreaks in humans and/or animals.^12,81,82 The MERS-CoV lineage 5 is a recombinant between lineage 3 and 4 and caused human outbreaks in Saudi Arabia in 2015 (Figure 3).¹² The SARS-CoV TOR2 strain might originate from recombination among SARS-CoV, alpha CoV (HCoV-229E), and gamma CoV (avian infectious bronchitis virus).⁸¹ Further, genetic evidence^13,14,47, 48, 49 suggests that SARS-CoV-2 is of natural origin by genetic recombination and is related to bat CoVs.

Cleavage sites with polybasic amino residues in the surface proteins of AIVs and HCoVs

The HA protein of IAVs and spike (S) protein of HCoVs can be cleaved into two subunits at their respective cleavage sites (CS) by host cell proteases. The seasonal IAVs H1N1 and H3N2 do not contain multibasic amino acids at their CSs. In the case of AIVs, HPAIVs always possess a motif with polybasic amino acids at the HA cleavage site (HA1/HA2 junction) (Figure 4), which is a genetic marker for H5 and H7 HPAIVs.^83,84 Also, a polybasic motif was identified at the S cleavage site (S1/S2 junction) in four HCoVs, including SARS-CoV-2, MERS-CoV, HCoV-OC43, and HCoV-HKU1 (Figure 5). Furthermore, the analogous polybasic CS on HA of HPAIV and S of SARS-CoV-2 could be correlated with increased virus virulence in poultry and/or mammals.^13,85, 86, 87

Regarding AIVs, the CS is flanked by PQ/L for H5 and PE for H7 at the N terminus and by GLF for both H5 and H7 AIVs at the C terminus (Figure 4). H5 and H7 low pathogenic AIVs (LPAIVs) can evolve into HPAIVs by the insertion of a polybasic motif at the HA CS (Figure 4).⁸⁸ In addition, varied lengths and polymorphisms of the CS were observed in naturally occurring viruses.⁸⁹ There are several potential mechanisms for the insertion of a polybasic CS: (1) substitution of non-basic amino acids with the basic R or K residues, (2) insertion related to duplication of purine triplets due to the polymerase slippage, and (3) recombination with other gene segments of IAVs or host 28S ribosomal RNA.^83,88,89

The polybasic CS is a crucial determinant for the infectivity and virulence of AIVs.⁸⁹ Generally, multibasic CSs of HPAIV HA protein can be recognized and cleaved by the ubiquitous expressed cellular proteases such as furin and PC6, causing systemic infection and even a fatal outcome in poultry.⁸⁸ However, the monobasic CS of LPAIV HA can merely be cleaved by trypsin and trypsin-like proteases.^88,89 The distribution of proteases for LPAIV HA cleavage restricts virus replication in respiratory and/or intestinal tracts and causes asymptomatic or mild symptoms in birds.^88,89 Further, the multibasic motif at CS also affects AIV virulence in mammals.⁹⁰

The SARS-CoV-2 also has a polybasic CS (PRRAR) and forms a furin site at the S1/S2 junction of the S protein; this insertion has not been previously observed in other beta-CoVs of clade 2b (lineage B) (Figure 5).^47,85,91 The mutations at the CS have been found in the Alpha (HRRAR), Delta (RRRAR), and Omicron (HRRAR) SARS-CoV-2 variants. In addition, SARS-CoV-2 variants with deletions and mutations at the CS were readily found during virus passage in Vero-E6 cells (Figure 5).^92,93 Furthermore, a three-residue (PAA) insertion at the S1/S2 junction was identified in a bat SARS-CoV-2-like virus (RmYN02).⁴⁷ Currently, how the insertion of polybasic residues into the SARS-CoV-2 CS occurred is still elusive. However, these results support the natural diversity of the polybasic CS in SARS-CoV-2 or SARS-CoV-2-related viruses.

The correlation between the furin CS and the infectivity and pathogenicity of SARS-CoV-2 was explored in in vivo and in vitro experiments. Loss of the furin CS (PRRA) decreased SARS-CoV-2 replication in human respiratory cells and attenuated virus pathogenesis in both hamster and mouse models compared with its parental virus.^85,86,93 Further, the cell-passaged SARS-CoV-2 variant with eight-residue deletions at CS was deficient in transmission among co-housed ferrets.⁹⁴ These results about AIVs and SARS-CoV-2 suggest that the polybasic CS is a crucial determinant for infection and replication ability, virulence, and transmissibility of HPAIVs and SARS-CoV-2 in poultry and/or mammals.

Sialic acid receptor and cell and organ tropisms for IAVs

IAVs utilize sialic acid receptors linked to glycoproteins and gangliosides for cell entry, especially N-acetyl-neuraminic acids, which are crucial in influenza viruses infecting hosts.²¹ The tissue distribution and expression of the sialic acid receptors are different in humans and avian hosts.⁹⁵ In humans, α-2,6 sialic acid (α2-6-SA) is predominantly expressed in the upper respiratory tract, while α2-3-SA is mainly expressed in the lower respiratory tract.⁹⁶ However, α2-6-SA and α2-3-SA were also occasionally detected in the lower respiratory tract and nasal mucosa, respectively.^97,98 However, in avian species, α2-3-SA predominates in both upper and lower respiratory tracts, and it is also extensively expressed in the intestinal epithelial cells of birds.⁹⁹ Typically, human IAVs (eg, H3N2) preferentially bind to human-type receptors (α2-6-SA), while AIVs preferentially bind to avian-type receptors (α2-3-SA). Hence, receptor specificity is a major determinant for the host range of influenza viruses. The preference of AIVs for human-type receptors was considered a crucial warning for cross-species infection of AIVs to humans.

The tissue distribution of sialic acid receptors corresponds to cell and organ tropisms for IAVs. The respiratory system is the primary target for IAV infections. Human IAVs have been documented to infect nasal mucosa, the epithelial cells of trachea, bronchi, and bronchioles, and type I pneumocytes (AT1) in the alveoli (Figure 6).^99,100 In contrast, the antigens of AIVs were primarily detected in the lower respiratory tract of humans;⁹⁹ AIVs tend to target cells in bronchioles and also infect epithelial cells, AT2, and macrophages in the alveoli (Figure 6).^99,100

Sialic acids also act as receptors for some alpha-, beta-, and gamma-CoVs.²¹ Specifically, O-acetylated sialic acids serve as receptors for beta-CoVs, eg, HCoV-OC43 and bovine CoV.²¹ N-acetyl- and N-glycolylneuraminic acids were recognized as receptor determinants of cell infection for transmissible gastroenteritis virus (alpha-CoV) and infectious bronchitis virus (gamma-CoV).²¹

ACE2 receptor and cell and organ tropisms for SARS-CoV-2

ACE2 serves as a receptor for SARS-CoV-2, SARS-CoV, and HCoV-NL63 to enter cells.^20,101 SARS-CoV-2 can efficiently use ACE2 as a receptor for cell entry, with up to 10- to 20-fold higher affinity than for SARS-CoV.¹⁰² ACE2 is expressed in the lungs, hearts, kidneys, livers, testes, and intestines of humans.¹⁰³ The ratio of ACE2-expressing cells of human lungs is relatively low.^104,105 About 0.64% cells that can express ACE2 in normal human lungs were profiled by single-cell RNA sequencing.¹⁰⁴ Specifically, in human lungs, ACE2 expresses in AT2, AT1, airway epithelial cells (ciliated and club cells), fibroblasts, endothelial cells, and immune cells (macrophages and T cells).^104,105 In the trachea, ACE2-expressing cells include club, goblet, ciliated, and proliferative cells.¹⁰⁶ The expression patterns of ACE2 are variable in current reports, so more samples and studies on ACE2 expression and distribution are needed to obtain solid results and understanding.

The tissue expression and distribution of ACE2 receptors correlate with the cell and organ tropisms for SARS-CoV-2 infection. SARS-CoV-2 can infect multiple organs, including the lung, trachea, pharynx, intestine, kidneys, pancreas, brain, and heart of humans.^22,107 The respiratory system is also the primary target for SARS-CoV-2 infection. The co-location of viral antigens with cell markers by immunofluorescence staining was used to uncover cell tropism details for SARS-CoV-2.²² In lungs of postmortem specimens, viral antigens were found in basal cells, ciliated cells, club cells, AT2, AT1, proliferative cells, and vascular endothelial cells (Figure 6).^22,108 SARS-CoV-2 infections in alveolar cells and airway cells have also been reported in human distal lung organoids¹⁰⁹ and ex vivo cultures of human bronchus and lung.¹¹⁰ In AT2-AT1 cell cultures, AT2 cells are preferentially infected by SARS-CoV-2 over AT1 cells.¹⁰⁸ In the trachea of autopsied humans, viral infection was reported in ciliated and goblet cells of the mucosa and epithelial cells of conduits and glands.²²

Target cells and receptors responsible for co-infection of IAV and SARS-CoV-2

The IAV and SARS-CoV-2 are airborne pathogens, and they primarily target the respiratory system of humans, including the nasal mucosa, trachea, bronchi, and alveoli.¹¹¹ The co-location and expression of sialic acid and ACE2 receptors may contribute to the co-infection of IAVs and SARS-CoV-2 in the same organs and cells. For example, both IAV and SARS-CoV-2 could target and infect AT2 cells in alveoli.^108,112 Further, co-infections in the same organs and cells could facilitate pathogenic and immunological interactions between the two types of viruses. Moreover, IAVs could promote the infectivity of SARS-CoV-2, probably due to IAV pre-infection elevating ACE2 expression in a human cell line.¹¹¹

Disease burden of co-infection with seasonal IAVs and SARS-CoV-2

During the COVID-19 pandemic, co-infections of SARS-CoV-2 with other respiratory pathogens have been reported, and IAVs were one of the most common co-infections.¹¹³ Patients co-infected with IAV and SARS-CoV-2 have been sporadically identified in multiple countries, such as China, Japan, Iran, the United States, Turkey, and Germany.^17,18 However, patients co-infected with IAV and SARS-CoV-2 or only infected with either virus presented similar clinical respiratory symptoms and radiological images (Table S2). The most common symptoms of co-infections are fever, cough, dyspnea, and myalgia, and the typical changes in chest radiology images include ground-glass opacities.^{18,19,114,115} However, patients with IAV and SARS-CoV-2 co-infection usually presented more severe clinical outcomes compared with those with SARS-CoV-2 infection alone.¹⁸ The concentrations of serum cytokines/chemokines in co-infections should be further studied to understand the potential immunological mechanism for the observed clinical manifestations.¹¹⁶ Some contradictory disease severity of co-infection could result from the small sample size or insufficient consideration of the confounding factors, including the order of infection, virus subtypes, and underlying comorbidity.^117,118

The clinical outcomes of co-infection with IAV and SARS-CoV-2 have been evaluated in laboratory-based studies.^111,113 Co-infection of SARS-CoV-2 with IAVs caused more severe disease in vivo than mono-infection with either virus. Further, the IAV pre-infection could interfere with the SARS-CoV-2 replication but increase disease severity.¹¹³ However, another study indicated that IAV pre-infection promoted the infectivity and viral load of SARS-CoV-2 and caused more severe lung damage in mice, while several other respiratory viruses did not enhance SARS-CoV-2 infectivity in cells.¹¹¹

During the COVID-19 pandemic, the prevalence of patients co-infected with IAV and SARS-CoV-2 varied substantially across regions and studies.^{18,19,119,120} Some studies documented a co-infection rate of influenza and COVID-19 as high as 57.3% in Wuhan,^119,120 while a relatively low ratio of influenza infections, 0.8%, was also reported in patients with COVID-19 based on a systematic review and meta-analysis.¹⁹ Non-pharmaceutical interventions for the pandemic, including mask wearing, social distancing, and travel restrictions, could reduce the transmission of IAVs, resulting in a low co-infection rate.¹²¹

Disease dynamics caused by the co-existence of seasonal IAVs and SARS-CoV-2

The COVID-19 pandemic suppressed the number of influenza infections in the 2020–2021 season.^122,123 In the Southern and Northern Hemispheres, seasonal influenza cases reduced drastically in the 2020 and 2020–2021 influenza seasons, respectively.¹²² In the United States and Europe, the positive influenza rate of samples in the 2020–2021 influenza season is about 0.15%–0.2%, which is estimated to be 18%–23% in the 2019–2020 influenza season.¹²² The decreased influenza burdens may have resulted from several factors. One explanation could be the interference and competition between distinct viruses in shared niches.¹²² At the onset of the A(H1N1)2009 pandemic, rhinovirus infections delayed the circulation of the A(H1N1)2009 in France.¹²⁴ Also, the negative transmission dynamics between the seasonal IAV and rhinovirus were documented in the UK.¹²⁵ Influenza infections could destroy cells and/or the surface receptors of the cells and thus decrease the number of susceptible cells for subsequent infection by other viruses.¹²⁵ Additionally, non-specific innate immune responses, such as interferon secretion, could be activated by prior virus infection and then suppress other virus infections.^124,125 Third, the non-pharmaceutical mitigation measures for the current pandemic could limit the influenza transmission routes analogous to SARS-CoV-2.¹²⁶ Also, the changed physician visiting behaviors and limited surveillance capacity could affect the documented transmission patterns of influenza during the pandemic.¹²³ Fourth, many countries (eg, the United States, Australia, and France) increased and accelerated the influenza vaccine uptakes to reduce the burden of overwhelming health systems in the COVID-19 pandemic.¹²² Given the above factors, the ecological and immunological relationship between SARS-CoV-2 and IAVs is a key determinant for their future infection dynamics,125, 126, 127 which should be evaluated in a unified model incorporating the interaction of multiple pathogens.

The atypically low influenza circulation in the seasons overlapping with the COVID-19 pandemic indicated a reduction in natural population-level immunity for influenza, which could shift the immune landscape, build up a susceptible pool, and further alter the timing, trajectories, and severity of influenza in future seasonal epidemics.^122,123 Moreover, the low influenza circulation could challenge the choice of vaccine strains and the vaccine efficacy and therefore increase the influenza disease burden, including overwhelmed healthcare capacities and severe morbidity and mortality, in the next season. Further, rapid virus evolution and immunity dynamics in humans could complicate the forecast modeling of influenza.¹²³ Anticipating the potentially atypical transmission patterns of influenza epidemics, especially after the COVID-19 pandemic, alongside well-prepared vaccines and antivirals, could guide targeted interventions and reduce the burden on healthcare systems.

Mitigation strategies of co-infection with seasonal IAVs and SARS-CoV-2

As the pandemic wanes and the countermeasures against COVID-19 become relaxed worldwide, circulation of IAVs and HCoVs may resume at pre-pandemic levels, together with SARS-CoV-2 continuing to circulate in humans. Co-infections with IAVs and SARS-CoV-2 may increase, complicating the diagnosis, treatment, and prognosis for patients and aggravating the disease burden for the medicine and healthcare system.¹²¹ Given this, the early detection of co-infection by multiplex molecular diagnostics should be implemented to timely initiate antiviral therapy and improve the prognosis of patients.^113,121 Practical preventive actions (eg, wearing masks, washing hands, and social distancing) could protect against infection with these airborne pathogens.¹²¹ Further, vaccinations could reduce co-infections, clinical severity, and disease burden, especially in high-risk individuals and the elderly.¹²⁸ Moreover, the live attenuated influenza virus-vectored COVID-19 vaccine has been designed to induce co-immunity against influenza virus and SARS-CoV-2,¹²⁹ and vaccine deployments will simplify the vaccination strategy against these two types of viruses.