Abbreviations:

URT — upper respiratory tract

MS — maxillary sinus

NA — nucleic acids

PNS — paranasal sinuses

PRS — polyposis rhinosinusitis

PCR — polymerase chain reaction

SNM — sinonasal microbiome

MNM — middle nasal meatus

CRS — chronic rhinosinusitis

Introduction

An extensive and detailed study of the human microbiome began at the end of the last century. Every year more and more scientific papers appear that reveal new aspects of this problem. In 2020 about 71,000 publications can be found in international scientific databases using the keywords «microbiome and «microbiota». The presence of a physiological microbiome (i.e. microbial colonization of the skin surface and mucous membranes) is vital for human growth and development, the correct formation of metabolic and immune homeostasis. The development of many diseases such as asthma, diabetes mellitus, obesity, some malignant tumors, rheumatoid arthritis, and even such as autism, multiple sclerosis, Alzheimer's and Parkinson's diseases are associated with microbiome disorders including sinonasal ones. Autoimmune diseases inherited in families are explained now not by the inheritance of the person genetic information but the inheritance of microbiome pathological features [1—4].

Bacteria inhabiting the intestines are involved in the process of food digestion, in the regulation of the immune system, protect a person from pathogenic microorganisms and participate in the synthesis of vitamins (thiamine, riboflavin, B₁₂, K). Almost all components of the human immune system are formed under the influence of the microbiome of the upper respiratory tract (URT). It is known that the microorganisms themselves growing in the nasal cavity and pharynx or the products of their vital activity directly or through epithelial cells affect dendritic cells and macrophages. The functions of CD4- and CD8-T-cells and the production of antibodies against the respiratory infections, particularly in influenza A depend on the composition of the intestinal microbiome [2, 5, 6].

The numbers of microorganisms inhabit the surface of our skin and hollow organs significantly exceed the number of the body's own cells. According to rough estimation the total number of cells in the human body is 10¹³ while the number of bacteria reaches 10¹⁴, viruses — 10¹⁵, fungi — 10¹². The number of microorganisms growing in the human body is 100 times higher than the population of the Earth; their total weight reaches 2 kg while the main part populates the large intestine [1, 4].

Terminology

Microbiota is the composition of microorganisms (not only bacteria, but also fungi, viruses, archaea, protozoa, bacteriophages) that live in a particular ecological niche. Microbiome is understood as the totality of the genetic material of all microorganisms that make up the microbiota. As for the human microbiome — this is the entire set of nucleic acids (NA) of all representatives of the micro-world located on the surface of the skin, mucous membranes or in the tissues of the human body. There are terms are used to characterize the bacterial microbiome in the English literature; some of them have not yet full-fledged Russian analogues:

— richness or abundance — quantitative composition of the microbiome;

— (bio)diversity — total number of microorganisms taxa;

— evenness — the ratio of different types / genus in the total composition of the microbiome.

The growing interest to the problem is associated primarily with the development of microbiological methods for whole genome sequencing, bio-informatics and meta-genomic analysis. Traditional strains methods still remain the basis of bacteriological research: in the clinical aspect they are quite informative, relatively cheap and make it possible to study the sensitivity of the revealed microorganisms to antibiotics and bacteriophages in vitro. However, in the «strains» era we were able to identify and classify only about 20% of the bacteria that exist on Earth. Conventional nutrient platform did not ensure the conditions for the all microbiota types’ growth. So, in the treatment of chronic rhinosinusitis (CRS), for too long we had to focus only on the results of strains studies which did not always create correct ideas about the role of certain microorganisms in the disease pathogenesis. L. Hauser et al. [7] who studied the biopsy specimens of the ethmoid labyrinth mucosa in 54 patients with CRS using standard strains methods and 16S-ribosomal RNA (rRNA) sequencing identified on average 3 types of bacteria in standard methods and 21.5 ± 12.5 in genome-wide analysis. Only 47.7% of microorganisms with NA detected by sequencing could be grown on nutrient platform.

Secondly, from the microbiology point of view the ideas about the normal state of hollow organs, such as the lungs’ alveoli, the bladder, and now the paranasal sinuses (PNS) which were previously considered sterile, have changed. Studies carried out in the last two decades have shown that in healthy condition these organs are colonized by communities of microbes consisting of commensals and potential pathogens while pathogens are present in quantities that are not capable of causing disease. Commensals making up the microbiome of hollow organs are in symbiosis with the host organism and form a kind of barrier that prevents the invasion of pathogens. An increase in the relative number of opportunistic pathogens causes imbalance / dysbiosis a diversity decrease and imbalance in the microbiota which leads to a decrease in the representation of key commensals and the response of the immune system to these changes causes acute or chronic inflammation [3, 4].

Methodological problems

Attempts to systematize data on the human synonasal microbiome (SNM) face methodological problems associated primarily with the lack of a systematic approach to the collection and analysis of material. So, works describing the composition of SNM in CRS and, even more so, in the normal state, as a rule refer to the results of the study of swabs from the middle nasal meatus (MNM), although the extrapolation of these data to the PNS microbiome is not entirely legitimate even in the case when the material taken under endoscopic control. It is well known that a significant part of inhaled air contaminated with microorganisms enters the MNM while air exchange in the PNS with the normal size of the natural anastomosis is minimal and should exist in them, also due to the differences in the gas composition of the air the unique composition of the microbiome that differs from the nasal cavity. Material sampling by sinus puncture is practically not used in the study of SNM; studies of mucosal biopsies obtained during surgical interventions on PNS are few [7—9]. For example, in our study of the synonasal microbiota in CRS significant differences were found between the composition of the MNM microflora and the contents of the maxillary sinus (MS) obtained during its puncture. Full identity of the microflora in MS and MNM was noted only in 8.9% of studies; the ratio of aerobic and anaerobic microorganisms was also different: anaerobes were significantly more often present in the MS. Staphylococcus aureus (4.5% of cases) and Streptococcus constellatus (also 4.5% of cases) were cultivated more often than others in both localizations. [10].

The study results by E. Copeland et al. [8] showed that the microbial composition of smears taken from PNS during endoscopic surgery in patients with CRS and those in the control group who do not suffer from this disease differs significantly due to the bigger representation of the Escherichia in CRS while Corynebacterium and Dolosigranulum dominate in individuals without CRS, i.e. operated on for other indications. The microbial composition of samples from MNM in patients with CRS and patients in the control group did not have such significant differences and the representation of Escherichia was not so excessive here as in samples from the sinuses. Thus, the microbiocenosis of MNM and PNS is significantly different especially in chronic inflammation, and the collection of smears from MNM even under endoscopic control is not an ideal method for studying the PNS microbiome.

The normal nasal microbiome

The nasal cavity along with the oral cavity is the main gateway of our body; and the nasal mucosa is forced to continuously react and respond to life-long «bombing» with a huge amount of pollutants, allergens and microorganisms. A healthy person inhales more than 7000 liters of air a day while each cubic meter of inhaled air contains 10⁴—10⁶ bacteria and their number and composition vary and depend on the state of the environment and the conditions of the inhaled air (humidity, gas composition, etc.) [3, 9]. It is important to note that more than 92% of a person's life time is spent indoors; therefore, the set of microorganisms entering the URT with inhaled air depends on the sanitary and hygienic conditions in the home and at the workplace [1]. Together with the peculiarities of the anatomical structure of the intranasal structures the composition and degree of contamination of the inhaled air determine the formation of specific local microbiocenoses of the respiratory tract where both permanent and transient representatives of the micro-world are present [4, 9].

The nature of the microbial «lining» of the URT consistently changes from the nose cavity entrance to the oropharynx depending on the environment (skin, multi-row ciliated or squamous cell epithelium), the state of the immune system and other parameters. In this case the composition of individual microbial «niches» (in our case the nasal cavity) may depend on the state of «neighbors», and the existing boundaries of microbiota can also shift [11]. Information on the composition of the microbiome of the nasal cavity and nasopharynx in children is rather contradictory. In a newborn it is poor; in fact it copies the composition of the microflora of the vagina and mother's skin and it’s usually represented by firmicuts (Lactobacillus, Staphylococcus, Streptococcus, Dolosigranulum), proteobacteria (Moraxella, Haemophilus) and actinobacteria (Corynebacterium) among which one or two of the mentioned taxa may dominate [3, 12]. However, immediately after the birth the new microorganisms coming from the external environment and begin to colonize the URT of the newborn; the evolution of the microbiota depends on a number of factors: not only the conditions of the child grows up but also to a large extent on the circumstances of childbirth (by the usual route or by cesarean section) and the nature of feeding. It is considered that the Dolosigranulum and Corynebacterium dominate in the microbiocenosis of the URT of the child at the age of 1.5 months with breastfeeding; the representation of S. aureus increases with artificial feeding [13—15]. Furthermore, the microbial community of the child's nasopharynx gradually undergoes serious changes becoming less compact but more diverse, and the evolution of the microbiota is in close interaction with the formation of the immune system. For example, it is known that a decrease in the representation and diversity of Bacteroidetes in the respiratory tract microbiome in small children predisposes to the development of allergic diseases. There is evidence that children with moraxella-dominated nasopharyngeal microbiome are less likely to suffer from the cold & flu. The only exception is Moraxella catarrhalis which presence together with Haemophilus influenza and Streptococcus pneumonia is associated with the wheezing development in children and some researchers consider the dominance of streptococci to be a clear predictor of the bronchial asthma development [15—17]. In general, observations of children under 2 years old show that the dominance of the Dolosigranulum and Moraxella in combination with corynebacteria ensures the formation of a more resistant microbiome in the nasopharynx in the future, more resistant to respiratory infections. On the contrary, the prevalence of H. influenza and S. pneumoniae contributes to increased susceptibility to respiratory viruses and the development of bronchiolitis and pneumonia in early childhood. [12, 16—18].

Even the first studies of the microbiome of the nasal cavity of a healthy adult (in swabs from the middle and upper nasal meatus) have shown that it is a unique and variable microbial community which is fundamentally different from the same communities inhabiting the nasal cavity entrance, nasopharynx, pharynx and mouth cavities despite its close anatomical neighborhood. Comparison of samples taken from different parts of the URT of an adult shows that the composition of the microbiome of the nasal cavity, pharynx and oral cavity is represented by the same types of bacteria but at the level of families and genera it is radically different. The uniqueness of the nasal microbiome is determined by the predominant representation of the types Actinobacteria (dominated by the Corynebacteriaceae and Propionibacteriaceae) and Firmicutes, and to a lesser extent — Proteobacteria [19]. In a study by C. Bassis et al. [20] the representation of Corynebacteriaceae copies in swabs taken from the nasal cavity of 10 patients ranged from 1.5 to 62.8%, and Propionibacteriaceae — from 0.4 to 42.4%. For comparison, in smears from the mucous membrane of the cheek and the tongue root, Corynebacteriaceae accounted for slightly more than 1% of the community while Propionibacteriaceae were practically not found. Another critical difference between the nasal and oral microbiome is the representation of different firmicuts. NA of Staphylococcaceae in the samples from the nasal cavity accounted for 2.2—55.0% and in the oropharynx they practically did not occur but NA of Streptococcaceae were found much more often among the representatives of firmicutes.

The microbial communities of the nasal cavity itself and its entrance also differ significantly; the composition of the nasal cavity entrance is much less diverse, although firmicuts and actinobacteria also prevail in it and the representation of proteobacteria is much less. The groups of researchers involved in the study of MNM agree that the main factor determining the balance of synonasal microbiocenosis and, accordingly, the health of the mucous membrane is the ratio of the representation of staphylococci (type Firmicutes) and corynebacteria (type Actinobacteria). Normally, corynebacteria prevail accounting for 36% while staphylococci (S. aureus, Staphylococcus epidermidis) — 26%. The microbial composition of the nasal cavity of healthy people who are not permanent carriers of staphylococcus is represented mainly by the type Actinobacteria (Propionibacterium spp. and Corynebacterium spp.) with a relatively low representation of the types Firmicutes (Staphylococcus spp.) and Proteobacteria (Enterobacter spp.). The percentage of staphylococci (S. aureus) in the nasal cavity and in MNM in its carriers increases precisely due to a decrease in the total number of Actinobacteria, in particular, Propionibacterium acnes [21, 22].

The normal nasopharyngeal microbiome also differs significantly from the nasal microbiome. According to an extensive monocentric study that included 100 healthy adults the types of microbial communities where Streptococcus and Fusobacterium prevail are specific only to the nasopharynx and practically are not found in the nasal cavity. In general, there are more potential pathogens in the normal nasopharynx than in the nasal cavity. [11].

It is well known that the composition of the microbiome changes with age. Thus, the majority of MNM bacteria in healthy individuals under 50 years old are represented by actinobacteria but at an older age their number decreases in favor of firmicuts; the percentage of proteobacteria practically does not change [19, 23].

According to other data the representatives of the Cutibacterium / Propionibacterium, Corynebacterium and Staphylococcus in the microbiome of the nasal cavity increase in people aged 40–65 years. After 65 years SNM gradually becomes poorer, loses variety and changes even more approaching the composition of the microbial community of the pharynx and oral cavity that can be explained by a decrease in the nasal mucosa defenses and a slowdown in mucociliary clearance [23—25].

The results of a large-scale international study ISMS (International Sinonasal Microbiome Study) with 410 samples of SNP microflora collected from all continents except Africa and Antarctica were analyzed by 16S rRNA sequencing suggest that the unified SNM characteristic of all healthy people does not exist. Three main microbiotypes were identified: in the first type the Corynebacterium dominated (54.14% of the relative average representation), in the second — Staphylococcus (28.54%), in the third (17.32%) there was no predominant dominance but there were representatives of other «key» genera that make up the microbiocenosis of the nasal cavity and PNS: Streptococcus, Haemophilus, Moraxella and Pseudomonas. In percentage terms, these three variants were presented in different ways in different regions: in Australia and Asia the first microbiotype was found more often, it was detected a little less often in America, in Europe the second type was found much more often than the first one. The third type of microbiome turned out to be the least widespread on all continents but geographical differences in general did not affect the basic composition of key participants in the synonasal microbiocenosis [26]. Thus, the nasal cavity and PNS look like two specific biological niches and depending on the state of the immune system, age and other factors the microbial communities inherent can inhabit in each of them but different ones in their diversity. However, close neighboring to the surrounding territories — the nasal cavity entrance and nasopharynx where the composition of the microbiome is more aggressive under the certain conditions it threatens to shift the established boundaries. As a result, the spread of opportunistic pathogens in those areas of the nasal cavity where commensals normally prevail; it can lead to the development of various diseases [19-21, 24]. It would be logical to assume that these pathogens can be Haemophilus influenzae and pneumococci whose expansion from the nasopharynx at viral infection can lead to the development of acute rhinosinusitis. Aggression from the opposite side, from the nasal cavity entrance, contributes to excessive contamination of the middle and upper nasal meatus with other potential pathogens; in particular Staphylococcus aureus can presumably cause the development of CRS. However, this is only theoretical premises that require scientific proof so far.

Basic concepts on the bacterial microbiome composition of the nasal cavity and PNS of a healthy adult are presented in the table. It is formed by 3 main types: actinobacteria, firmicutes and proteobacteria. A number of researchers also assign a certain role to the type Bacteroidetes.

Table. The main types and genus of bacteria in the microbiome of the nasal cavity and PNS of a healthy adult

Microbial competition within the microbiome

Most of the microorganisms that make up the SNM are in a state of constant direct or indirect syntropy / competition for ecological niches. This competition can take place both between commensals and potential pathogens and, to some extent, between microbes and the host itself. In the process of competing for the food and life support bacteria have different paths: they can absorb the waste products of the host itself or produce special molecules, for example, the so-called siderophores which are able to utilize the atomic iron from the environment. There are other mechanisms of direct and indirect influence on the ecology of habitat [9, 27]. Understanding the nature of inter-microbial competition in the future is important for the development of pathogenesis methods of chronic diseases treatment of the nasal cavity and PNS, in particular, associated with the persistence of Staphylococcus aureus. This opportunistic pathogen transiently or permanently but more often asymptomatically colonizes the nasal mucosa, however when it multiplies excessively it can cause a number of pathological conditions. The thiazolidine—containing cyclic peptide lugdenin which inhibits the growth of S. aureus in vitro may become one of the theoretically effective substances for the treatment of such conditions. Another candidate also the antimicrobial peptide nukacin IVK45 is produced by epidermal staphylococcus under the conditions of oxidative stress and a lack of iron ions [27-29]. Perhaps the most interesting and most studied in this regard are interactions between staphylococci and corynebacteria. Thus, some strains of C. pseudodiphteriticum and C. accolens can inhibit the growth of S. aureus. In general, the corynebacteria themselves and even the cell-free platform prepared on their basis were able to reduce the virulence and hemolytic activity of S. aureus as well as its capabilities in terms of excessive colonization of the mucous membrane, i.e. actually put them in a commensal position [27, 30, 31].

In the competition for such components necessary for the vital activity of bacteria as methionine and iron ions, coagulase—negative staphylococci which produce fewer siderophores appear to be weaker players [32]. Recently, it has been shown that the species C. accolens which is normally a commensal living on the skin of the nasal cavity entrance is able to inhibit the growth of S. pneumoniae due to the production of free fatty acids which, in turn, increase the expression of the human antibacterial β-defensin—2 [ 33]. Competitive interactions between some representatives of the URT microbiome are shown in the figure.

Fig. Some of the known interspecies competitive ratios of bacteria that compose the synonasal microbiome.

Some types of corynebacteria are capable of inhibiting or, conversely, promoting the growth of staphylococci and S. pneumoniae in vitro. Other microorganisms are capable reduce the virulence of S. aureus or support the formation of its bio-membranes (adapted from [9]).

Fungal microbiome (mycobiome) of the paranasal sinuses

At the end of the twentieth century the hypothesis about the leading role of fungi in the pathogenesis of CRS attracted much attention. Then, using methods of molecular analysis the presence of fungi mainly Alternaria in PNS was detected in more than 90% of patients with CRS, and then in the same number of healthy individuals [34, 35]. However, the initial optimism associated with the preliminary results of the topical antifungal drugs’ usage (Amphotericin B) in the treatment of all forms of CRS was subsequently refuted by the negative results of multicenter controlled trials [36, 37].

Subsequent studies using the 18S rDNA sequencing method gave more specifics although largely contradictory ideas about the MNM and PNS mycobiome. In one of these studies the Malassezia became the prevailing fungal species and the mycobiome itself in patients with CRS was richer than in healthy individuals (on average 12.14 versus 8.18 species in the sample) while the total number of fungal copies decreased after the surgery on the PNS [38]. In another study, on the contrary, no qualitative differences were found between the mycobiomes of healthy individuals and patients with CRS but in patients the composition of the PNS fungal association was richer and more diverse. Both in patients and healthy subjects the Cryptococcus neoformans prevailed but in CRS they were detected more often (90% versus 61%). Representatives of Malassezia were in third place here and more common in healthy individuals (4.7% versus 1.4%) [39].

In general, modern studies show that fungal DNA is presented in MNM and PNS much poorer than bacterial; potentially pathogenic fungi were detected only in some patients with polyposis rhinosinusitis (PRS) but in none of the patients with CRS without polyps and in none from healthy persons [40].

Viral microbiome (virome) of the paranasal sinuses

Most exacerbations of CRS occur during the viral infections and as for bronchial asthma where the role of rhinoviruses in the pathogenesis of exacerbations is well proven, one can assume their important role in exacerbations of CRS. A study using multiplex polymerase chain reaction (PCR) for respiratory viruses confirmed this assumption initially. The authors found the viral NAs (mainly rhinovirus) in 64% of scrapings and in 50% of lavage fluid samples from patients with CRS; in the control group these figures were significantly lower — 30 and 14%, respectively [41]. In a Chinese study the respiratory viruses were detected in the epithelial cells of MNM in 68.66% of patients with PRS, in 73.77% of patients with CRS without polyps and in 75.47% of healthy individuals; the difference between the groups was insignificant [42].

In contrast to these results, Y. Jang et al. [43] also by PCR revealed the NA of rhinovirus only in 21% of the epithelial cells of the middle nasal meatus in patients with CRS, and A. Wood et al. [44] found no traces of typical pathogens of respiratory viral infections in this category of patients at all. The authors of these works do not exclude that not persistence but a transient viral infection may be the impetus for the development of CRS or its exacerbations. Experimental studies on the cells sample of nasal polyps inoculated with rhinovirus also gave contradictory results [44]. Thus, the role of changes in the synonasal virome and the respiratory viruses themselves in the pathogenesis of CRS and its exacerbations remains unclear and requires further study. Taking into account the exclusively parasitic variant of the symbiotic relationship of viruses with the human body it is obvious that human viruses can hardly be permanent participants in microbiocenoses and we can only talk about transient participation in the regulation of microbiomes of the URT mucous membrane. However, if we consider the viruses of bacteria (bacteriophages) this can be very interesting both from the standpoint of the possibility of regulating microbiocenosis (including due to prophages) [45] and from the standpoint of using phages / prophages as markers / regulators of the functioning of microbiocenoses [46 ]. The latter mechanism is clearly underestimated from the standpoint of understanding the functioning of microbial communities.

Authors’ contribution:

Study concept and design — A.S. Lopatin

Collection and processing of material — A.S. Lopatin, I.S. Azizov

Text writing — A.S. Lopatin

Editing — I.S. Azizov, R.S. Kozlov

The authors declare no conflicts of interest.