Respiratory viruses cause the most common infectious illnesses in humans—acute RTIs that can be divided into upper RTIs (URTI), e.g., the common cold, and lower RTIs (LRTI), e.g., bronchitis and pneumonia. These illnesses affect all age groups annually and cause a high burden on health care systems and global economics due to absenteeism from daycares, school, and work. Over 200 virus types have been identified as causative agents for respiratory illnesses [1
]. In most cases, especially illnesses of the upper respiratory tract are mild to moderate and often self-limiting. On the other hand, LRTIs leading to pneumonia can be especially fatal among children and the elderly, or immunocompromised subjects [2
In the past decade, studies linking the microbiota and immune system function have laid the foundation for the opportunities in microbiota modulation and bacterial therapeutics in health and disease. Further, studies have associated the gut and airway microbiota with upper and lower respiratory tract health and immunity [3
]. Thus, modulation of the gut microbiota and immunity by dietary supplements or pharmaceuticals is of increasing interest in finding novel solutions to manage RTIs. Among the promising candidates are probiotics that have been studied for immune function modulation and viral infections. Meta-analyses suggest that probiotics could be beneficial in the context of acute URTI typically caused by viruses [5
We conducted a literature search with keywords “probiotics”, “Lactobacillus”, Bifidobacterium” “respiratory tract infections”, “viral respiratory tract infections”, and “respiratory virus” in common databases such as PubMed, Google Scholar, and Web of Knowledge up to March 2020. In this narrative review, we focus on evaluating the current knowledge in the scientific literature on the immunomodulatory effects of probiotics in the context of viral RTIs by reviewing key pre-clinical and clinical evidence.
2. Etiology and Epidemiology of Viral Respiratory Tract Infections
The most common viruses causing respiratory infections are picornaviruses (rhinoviruses and enteroviruses), coronaviruses, adenoviruses, respiratory syncytial virus (RSV), parainfluenza, and influenza viruses [1
]. In recent decades, advancements in molecular diagnostics have also enabled the identification of several novel respiratory viruses, including human bocaviruses and metapneumovirus as well as highly pathogenic coronaviruses (severe acute respiratory syndrome: SARS-CoV, SARS CoV-2, and MERS-CoV) [8
]. Estimations show that rhinoviruses that circulate in the community throughout the year cause approximately half of all the common cold cases, in contrast to seasonal viruses such as influenza and coronaviruses that cause approximately one-third of the common colds [1
]. Individuals can also be infected simultaneously with multiple viruses.
Children tend to have more frequent infections than adults and the elderly due to their still maturing immune system [10
]. The elderly, on the other hand, are more susceptible to severe complications due to aging-associated decline of the immune system function. The duration of the infection and presentation of symptoms, i.e., illness, varies between individuals [11
]. Several respiratory viruses are detectable in asymptomatic subjects and, for instance, one-fifth of rhinovirus-infected individuals do not experience symptoms and do not feel ill despite carrying the rhinovirus [12
]. On the other extreme, respiratory illnesses can last for weeks and be fatal, as in the case of emerging viruses, i.e., new influenza strains or coronaviruses capable of causing SARS. The underlying reason between these differences is due to genetic and epigenetic differences in immune responses and in the characteristics of the infecting virus, but potentially also the differences in the respiratory microbiota [13
3. Immune Responses against Respiratory Viruses
Viruses that cause RTIs are found in various virus families, which differ in virulence and utilize variable strategies to infect the host cells and to evade the host immune system [14
]. Respiratory viruses spread via nasal secretions that can be transmitted through the air or by hand-to hand and surface-to-hand contact [16
]. Infection requires penetration of the virus through the host mucus layer, including the microbiota, and antiviral molecules in the mucus, such as antibodies and collectins. Once on the mucosal epithelial cells, respiratory viruses attach to specific receptors, such as Intercellular Adhesion Molecule (ICAM)-1 (rhinoviruses), peptidases (coronaviruses), or sialic acids (influenza viruses), that mediate the internalization of the virus by endocytosis [17
]. The viral receptors are differentially expressed on host cells resulting in virus-specific host cell tropism that is one key factor in viral pathogenesis. For example, influenza viruses typically infect bronchial cells, whereas rhinoviruses infect the epithelial cells of the upper airways, resulting in differences in illness presentation.
Viral genomic structure can be, for example, positive- or negative-sense single-stranded (ss) or double-stranded (ds) RNA or DNA [19
]. Most RTI causing viruses, picornaviruses, influenza viruses, and coronaviruses are ssRNA viruses, whereas adenoviruses are dsDNA viruses. The genomic structures are recognized by different receptors in the host and activate different types of immune responses. Some respiratory viruses, e.g., coronaviruses, are surrounded by a viral envelope which confers additional protection from the host immune system. Respiratory viruses have further developed molecules that help in evading the immune response, for example, by disrupting the interferon (IFN) response and hijacking the host’s cellular machinery for the production of virus copies [15
Once the viruses have penetrated into the host cells, the epithelial and immune cells detect the viral structures by pattern recognition receptors (PRR) of which Toll-like receptors (TLRs) and retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs) play a central role. TLR3, TLR7, TLR8, and TLR9 are located in the endosomes and can identify viral ss (TLR7 and 8) and ds (TLR3) RNA structures, and DNA (TLR9) [18
]. The recognition by TLRs leads to activation of transcription factor nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) and IFN regulatory factors (IRF) 3, 5, and 7 [12
], resulting in expression of pro-inflammatory cytokines and type I IFNs, IFN-α and IFN-β. Type I IFNs are broadly secreted by cells, but epithelial cells further secrete type III lambda IFNs in response to viral infections. Cytoplasmic RNAs, on the other hand, are recognized by RLRs, of which RIG-I recognizes ssRNA and melanoma differentiation associated 5 (MDA-5) dsRNA. The activation of RLRs leads to type I (and type III in epithelial cells) IFN production via mitochondrial antiviral signaling protein (MAVS). Type I and III IFNs induce an antiviral state in the surrounding cells which is not, however, necessarily sufficient to resist the infection, but delays the spreading of the infection [12
]. The activation of RLRs, and TLRs by viral infection and cellular stress, leads to formation of nucleotide-binding oligomerization domain (NOD)-, leucine-rich repeat (LRR)-, and pyrin domain-containing protein 3 (NLRP3) inflammasome [23
]. Although the role of NLRP3 is still somewhat unclear in viral RTIs, it seems to play a role at least in rhinovirus, influenza, adenovirus, and RSV infections. NLRP3 inflammasome activation drives caspase 1-dependent IL-1β and IL-18 cytokine response and inflammatory programmed cell death (pyroptosis) [24
Epithelium-derived pro-inflammatory cytokines tumor necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6, chemokine (C-C motif) ligand (CCL) 2, CCL5, chemokine (C-X-C motif) ligand (CXCL)8, and CXCL10 induce innate cellular responses by attracting and activating Natural Killer (NK) cells, macrophages, and neutrophils that further amplify the innate cytokine and chemokine response [17
]. The role of other innate cells like, for example, intraepithelial lymphocytes, γδT cells, mucosa associated invariant T cells (MAIT), and innate lymphoid cells (ILC) is less well described in viral infections, however, they are likely to contribute to innate and adaptive responses against viral infections, as exemplified by the NK cells [25
] and by the role of ILC2 cells in overcoming an influenza infection [27
]. If the innate immune or memory responses cannot clear the pathogen effectively and the adaptive immune system is unexperienced with the virus, an adaptive immune response is initiated and required. Key are dendritic cells (DCs) that present the viral antigens and induce B and T cell responses against the pathogen in the secondary lymph nodes. B and T cell responses initiate within four–six days post-infection and peak later at days 7–14 depending on the respiratory virus [28
]. Typically, common respiratory viruses, such as rhinovirus and influenza virus, are cleared before adaptive immune responses are activated [22
] indicating that memory responses and innate immunity are essential in viral eradication. However, the induction of cytotoxic CD8 T cells, CD4 T cells, and antibody responses is key for virus eradication by adaptive immunity and for establishing protective immunity for secondary infections.
The activation of the epithelium, innate immune cells, and adaptive responses is important for defense against respiratory viruses, but on the other hand, the host inflammatory response is the major cause of symptoms and more severe pathologies [12
]. Chronic activation of CD8 T cell responses and adaptive immunity may lead to pulmonary damage and acute respiratory distress syndrome, like in severe cases of coronavirus infections (e.g., SARS-CoV or SARS CoV-2) and pandemic influenza virus infections [18
]. In milder colds, rhinoviruses are not cytolytic and do not actually cause considerable damage to host cells and may pass asymptomatically. Presentation of cold symptom severity seems to correlate with host inflammatory response. Specifically, the early expression of pro-inflammatory IL-8 [33
] and high levels of neutrophils in nasal aspirates [34
] have been shown to correlate with symptom severity of rhinovirus and influenza infection [35
]. Production of anti-inflammatory IL-10, resolvins, and regulatory T cell responses acts as a natural mechanism to control lung inflammation during acute influenza virus (and others) infection [37
]. Virus–host immune interactions are key to viral pathogenesis and to ultimately determine the outcome of the infection.
5. Clinical Evidence
Accumulating clinical evidence suggests that probiotics in general may have favorable effects against RTIs. For instance, several systematic reviews and/or meta-analyses have evaluated the effects of prophylactic ingestion of probiotics on the RTI-associated outcomes, e.g., either only in children [79
], or both in children and adults [5
] (Table 1
). Of note, the majority of the outcomes in these analyses are related to URTI, and data on LRTI outcomes are either not available or are very limited. Therefore, in the below chapter, we primarily focus on clinical trials on probiotics’ effects on URTI symptoms/episodes/duration.
In children (below 18 years), the meta-analysis by Wang et al., 2016, reported that probiotic use compared with placebo significantly decreased the number of subjects having at least one RTI episode, had fewer numbers of days of RTIs per person, and had fewer numbers of days absent from daycare or school [80
]. However, the meta-analysis did not find a statistically significant difference on the illness episode duration between the probiotic and the placebo. Laursen and Hojsak [79
] limited the analysis to children up to 7 years old and reported that probiotic use was associated with reduced risk of at least one URTI and reduced the risk of antibiotic use, but the use was not associated with a reduction in RTI duration or missed days of daycare due to RTI [79
]. This meta-analysis also discussed the effects of the individual probiotic strains on RTI outcomes. The results of the analysis showed that the most effective probiotic strains on RTI-related outcomes were L. rhamnosus
GG (RTI duration) and L. acidophilus
NCFM as a single supplement and in combination with B. lactis
Bi-07 (RTI duration and antibiotic use). Interestingly, these strains have shown in vitro the ability to induce antiviral IFN signaling pathways (see Section 4.2
) which may potentially explain their beneficial effects observed in RTIs. However, as multiple studies with probiotic strains other than L. rhamnosus
GG are limited or lacking, comparison and interpretation of the strain specific results should be made carefully.
Meta-analyses that pool data from clinical trials conducted with children, adults, and the elderly show that probiotic use is more beneficial over placebo in reducing the number of participants experiencing episodes of acute URTI [5
], reducing antibiotic prescription rates for acute URTIs [5
], and reducing the mean duration of an episode of an acute URTI as well as cold-related school absences [5
]. When the literature search was conducted, meta-analyses were not found in the databases searched on probiotic effects on respiratory infections restricted to the elderly population, potentially due to the fact that data are fairly limited regarding this age group.
While there is consensus that probiotics could have potential in reducing the risk for RTIs, it should be noted that clinical trials in the meta-analyses have been conducted in populations of different ages and genetic backgrounds, with various strains and/or their combinations, supplementation matrices, and doses. Moreover, the measured outcomes and data collection procedures between the trials (i.e., infection episode definition) are not harmonized and therefore may vary considerably. Consequently, pooling all the data creates a bias, as the probiotic effect is generally dependent on the dose, population, and strain. Moreover, as discussed above, the probiotics effects on the immune system are strain-specific which affects the interpretation of the results.
With regard to probiotics effects to specific respiratory viruses in clinical settings, several trials have characterized the respiratory infection etiology in infants [81
], in children [82
], in adults [85
], and in the elderly [86
]. In addition, two clinical trials have investigated the efficacy of probiotics in an experimental rhinovirus challenge model [87
] (Table 2
In the clinical trials conducted in free-living subjects in the community, no consistent data exist that show that specific probiotics would reduce the incidence of laboratory-confirmed respiratory virus infections as such. In preterm infants, the use of L. rhamnosus
GG for 60 days was associated with lower incidence of rhinovirus-induced episodes (comprising 80% of all RTI episodes) compared with the placebo. However, L. rhamnosus
GG had no effect on rhinovirus RNA load during infections, duration of rhinovirus RNA shedding, duration or severity of rhinovirus infection, or the occurrence of rhinovirus RNA in asymptomatic infants. In children attending daycare, L. rhamnosus
] consumption for 28 weeks did not reduce the occurrence of any of the common respiratory viruses either. In otitis-prone children, supplementation of a combination of L. rhamnosus
GG, L. rhamnosus
Lc705, B. breve
99, and Propionibacterium jensenii
JS, for six months, reduced the number of human bocavirus-positive nasopharyngeal samples when compared with placebo, but not the number of rhino/enterovirus-positive samples [84
]. Furthermore, in schoolchildren, the consumption of Levilactobacillus brevis
KB290 during influenza season was associated with lower incidence of physician-diagnosed influenza virus cases [82
]. In adults attending military service, the use of a combination of L. rhamnosus
GG and B. lactis
BB-12 for either 90 or 150 days was not overall associated with lower occurrence of common respiratory viruses upon presentation of cold symptoms [85
]. However, in a subgroup, there was a lower occurrence of rhino/enteroviruses after three months in the probiotic group when compared with the placebo. In nursing home residents, Wang et al., 2018, reported that the use of L. rhamnosus
GG for six months was not associated with the reduction in occurrence of confirmed viral respiratory infections [86
]. The differences between the findings in these trials may be explained by the fact that these studies were conducted in various age groups with different immune system statuses (infants vs. children vs. healthy adults vs. the elderly), different seasons, as well as variable probiotic strains, strain combinations, doses, and variable lengths of intervention. Furthermore, most of the studies were not designed for analyzing the viral infection etiology as the primary outcome and the diagnosis for the identification of the viral agent was not applied.
Since over 200 respiratory virus types can cause respiratory infections and, in many cases, the infections and symptoms overlap, or the etiology is undiagnosed, the potential antiviral effects of probiotics against specific viruses can be difficult to determine in clinical trials targeting free-living subjects within the community. To overcome this caveat, two probiotics have been investigated in an experimental rhinovirus challenge model that allows investigation of the effect of a probiotic strain to a specific viral pathogen. In a rhinovirus (type 39) challenge model, B. lactis
Bl-04 was administered for 28 days prior and during five days of experimental rhinovirus infection to healthy volunteers [89
]. B. lactis
Bl-04 supplementation resulted in significantly lower rhinovirus titers in nasal washes during the infection as well as in a lower number of infected participants shedding the virus compared with the placebo. Moreover, B. lactis
Bl-04 induced a significantly higher concentration of IL-8 in nasal washes after 28 days of supplementation and prior to infection. Given the reduced viral titer, an increase in IL-8 could indicate priming of the mucosal immune system prior to infection. This hypothesis is in line with a clinical study conducted in healthy active adults, where supplementation of B. lactis
Bl-04 reduced the risk of URTI episodes compared with placebo [90
]. In another similar experimental rhinovirus type 39 challenge pilot trial, no significant antiviral effect was seen with live or inactivated L. rhamnosus
GG supplementation compared with placebo [87
], suggesting potential strain-specific differences on the efficacy of probiotics in respiratory virus infections. Nevertheless, further adequately powered trials with harmonized study designs are necessary to draw conclusions on the efficacy of probiotics against specific respiratory viruses.
6. Discussion and Conclusions
Viral RTIs are the most common infections of mankind and the health and financial impact of seasonal epidemics and global pandemics on society is high. Due to the large number of various respiratory viruses, the development of efficient therapies, such as vaccines, is challenging. When preventative measures are scarce or lacking, the role of a well-functioning immune system becomes crucial for providing resistance to an infection. Within the past decade, research highlighting the importance of the microbiota on immune system function has raised interest in understanding the role of microbiota modulation and bacterial therapeutics by dietary and pharmaceutical solutions in health and disease. Of the available solutions, probiotic bacteria have been studied for immune function modulation in the context of respiratory viral infections. In this review, we have summarized the current evidence on the effects of probiotics on antiviral immune function in vitro and in vivo, and clinical evidence on the effect of probiotics on viral RTIs and on the course of RTI (Figure 1
In vitro data indicate that probiotics have strain-specific immunomodulatory effects on the host and immune cells by engaging TLRs that stimulate IFN pathways. The upregulation of IFN response seems to prime cells for better resistance against virus infection as probiotics were shown effective in inhibiting the replication of various respiratory viruses, including influenza viruses and RSV. Similar effects have been demonstrated in mice with the ability of the probiotics to reduce virus titers in lung tissues and to modulate antiviral and pro-inflammatory gene expression before and after viral infection. Interestingly, some studies in mice show an increase in IL-10 response, suggesting control of the pro-inflammatory response that typically drives lung pathology in severe infections. Most likely probiotics’ effects in the gut are transferred into the respiratory tract via the gut–respiratory tract axis, however, this mechanism of action remains to be studied in more detail. The pre-clinical studies further show improvement in the symptom scores of mice, suggesting potential clinical benefits. Indeed, some evidence exists for specific probiotic strains, e.g., from the species of L. rhamnosus, L. acidophilus, and B. lactis for their ability to induce antiviral immune responses in pre-clinical models, which is in agreement with their effects observed in clinical trials in reducing the risk of RTI-associated outcomes. However, translation of probiotic effects from cell culture and animal studies to humans can be challenging and variable confounding factors, e.g., age, diet, microbiome, genetic and epigenetic immune status of an individual, study season, and variable viral epidemiology, all have an impact on the study outcome and are difficult to standardize. The clinical studies that have diagnosed and characterized viral etiology are limited, nevertheless, the meta-analyses investigating probiotic clinical interventions on RTIs show that probiotic use is associated with lower incidence and duration of mild RTIs, both in children and in adults. Further studies aiming at discovering the mechanism of action of probiotics and establishing the association of immune system function stimulation and clinical efficacy are warranted.