Introduction

Pneumonia is a medical condition characterized by the inflammation of terminal airways, alveoli, and interstitial lungs, often caused by a wide variety of microbial pathogens [1]. According to the World Health Organization (WHO), lower respiratory infections, including pneumonia, were ranked as the fourth leading cause of death globally [2]. Pneumonia can be categorized into hospital-acquired pneumonia (HAP) and community-acquired pneumonia (CAP), depending on the place where the infection is contracted [3]. CAP is a significant and life-threathening disease, causing about three million deaths worldwide annually [4]. A population-based cohort study in Germany in 2015 showed that the mortality rates for CAP cases in hospital, at 30-days, and at 1-year were 18.5%, 22.9%, and 44.5%, respectively [5].

Bacterial and viral infections are important causes of pneumonia [4, 6, 7]. The leading pathogenic bacteria in HAP include Acinetobacter baumannii, Pseudomonas aeruginosa, Klebsiella pneumoniae, and Staphylococcus aureus [8], while respiratory syncytial virus (RSV), parainfluenza virus, human rhinovirus (HRV), and influenza virus (IFV) are among the most commonly identified viral pathogens [9]. Numerous investigations have established the crucial role of viral infections, especially IFV, as a major risk factor for CAP. For instance, Jain et al. [10] found that HRV, IFV, and Streptococcus pneumoniae were the most common pathogens identified in adults with CAP, while Deng et al. [2a). In terms of species, the top five in average relative abundance in the IFVN group were Streptococcus pneumoniae, Burkholderia cenocepacia, Acinetobacter baumannii, Pseudomonas aeruginosa, and Streptococcus salivarius subsp. thermophilus, with Streptococcus pneumoniae being the most prevalent (4.51%). The top five species in the IFVP group were Acinetobacter baumannii, Streptococcus pneumoniae, Mycoplasma hyosynoviae, Streptococcus salivarius subsp. thermophilus, and Porphyromonas endodontalis, with Acinetobacter baumannii being the most abundant (11.70%) (Fig. 2b).

Fig. 2
figure 2

The top ten genera (A) and species (B) of bacteria in relative abundance, and the genera (C) and species (D) with significant difference between the IFVN group and the IFVP group. Note: IFVN, influenza virus negative; IFVP, influenza virus positive; *P < 0.05; **P < 0.01; ***P < 0.001. Wilcoxon rank sum test was used to compare the relative abundance of bacteria between the IFVN group and the IFVP group

Full size image

At the phylum level, only one bacterium exhibited a significant difference in relative abundance between the two groups, whereas at the class, order, family, genus, and species levels, four, seven, thirteen, thirty-four, and fourteen bacteria, respectively, were found to have significant differences. Among the genera, the top ten were Klebsiella, Clostridium sensu stricto-1, Gaiella, Rhodoplanes, Rikenella, Rodentibacter, GCA-900066575, 1174-901-12, Caulobacter, and Clade Ia. Meanwhile, the top six species with significant differences in relative abundance were Acinetobacter calcoaceticus, Bacteroides gallinaceum, Ruminococcus flavefaciens, Bacteroidia bacterium feline oral taxon 115, Bacillus funiculus, and Pantoea ananatis (Fig. 2c, d). The other bacteria with a statistically significant difference across the two groups at the genus or species level are presented in Additional file 1: Tables S3 and S4.

Bacterial diversity between the IFVN and the IFVP groups

The Chao1 index showed a slightly higher median in the IFVN group, while the Shannon and Simpson indexes displayed slightly lower medians in the IFVN group compared to the IFVP group. Compared with the IFVN group, the three indexes were more concentrated in the IFVP group. No statistical difference was observed in any of the three indexes across the two groups (Fig. 3a–c).

Fig. 3
figure 3

Bacterial alpha diversity between the IFVN and the IFVP groups (A, Chao1 index; B, Shannon index; C, Simpson index; D, rank abundance curve in the individual level; E, rank abundance curve in the average level). Note: IFVN, influenza virus negative; IFVP, influenza virus positive; OTU, operational taxonomic unit; NS, no significant difference

Full size image

Compared with the IFVN group, the rank abundance curve of the IFVP group appeared to be wider, suggesting a greater richness of bacteria in the IFVP group. In addition, the IFVP group exhibited a narrower vertical span in the rank abundance curve than the IFVN group, indicating a more even distribution of bacterial composition, although the difference between the two groups was not significant (Fig. 3d, e).

In the PCA analysis, samples from the IFVN group and the IFVP group were closely positioned to each other (PC1 6.02%; PC2 4.55%), indicating that the overall structure of the bacterial communities was similar between the two groups (Fig. 4a). Similarly, the PCoA analysis did not show any distinct clustering of the microbiomes of the IFVN and IFVP groups (PCoA1 30.09%; PCoA2 8.7%), and no significant difference was found in the Adonis analysis (R2 0.012; P 0.465) (Fig. 4b). Furthermore, the hierarchical clustering analysis based on the UPGMA method did not show obvious clustering pattern among the samples from the two groups (Fig. 4c). The LEfSe analysis identified several bacterial taxa with markable differences between the IFVN and IFVP groups. In the IFVN group, Klebsiella was the key contributor to the difference. In the IFVP group, Gaiellaceae, Gaiella, and Rhodoplanes were responsible for the difference across the two groups (Fig. 4d).

Fig. 4
figure 4

Bacterial beta diversity between the IFVN and the IFVP groups (A, PCA analysis; B, PCoA analysis; C, hierarchical cluster analysis; D, linear discriminant analysis effect size). Note: IFVN, influenza virus negative; IFVP, influenza virus positive; PCA, principal component analysis; PCoA, principal coordinate analysis; LDA, linear discriminant analysis

Full size image

Discussion

Next-generation sequencing technology in a high-throughput approach has emerged as a highly efficient tool for identifying multiple microbial pathogens, which has overcome the limitations of inaccuracy and instability in pathogen detection associated with traditional culture-based methods [34]. This study characterized the microbiome composition among severe pneumonia patients with or without IFV infection using 16S-rDNA sequencing technology. We found that severe pneumonia patients with IFV infection had a higher relative abundance of lung flora, with Acinetobacter baumannii being the most abundant. There was a higher abundance of Klebsiella in the IFVN group compared to that in the IFVP group. No statistically significant differences in alpha and beta diversity indexes were observed between the two groups.

A slightly higher diversity of bacteria was observed in the IFVP group compared to the IFVN group, though no statistically significant difference was observed. The exact mechanisms underlying the increased susceptibility to bacterial co-infection following IFV Infection remain elusive. One proposed mechanism involves alveolar macrophages, which play a crucial role in immune defense against bacterial infection by phagocytosing and eliminating foreign dust particles and pathogens. IFV infection may cause early depletion of alveolar macrophages, resulting in decreased immune function and increased susceptibility to bacterial co-infection [35]. In mouse models, IFV infection induced systemic glucocorticoids that promoted bacterial growth [36], and inhibited the expression of antimicrobial peptides in the lungs, rendering the host more susceptible to bacteria such as Staphylococcus aureus. However, these bacterial infections and inflammation reactions were alleviated and eliminated after injecting exogenous antimicrobial peptides [37]. In addition, viruses have been found to promote bacterial infections by disrupting the epithelial barrier and up-regulation of adhesion proteins [38]. Recent research by Bai et al. [39] revealed that IFV-A induced the expression of cyclophilin A, an intracellular receptor for cyclosporin A with immunosuppressive effects, to promote co-infection with Streptococcus. The mechanism involves cyclophilin A interacting with focal adhesion kinase (FAK) to inhibit the K48-linked FAK ubiquitination process, which positively regulates the expression of integrin α5 and actin rearrangement through the FAK/Akt signaling pathway, thereby promoting colonization and invasion of Streptococcus. Further research is required to fully elucidate the various mechanisms underlying the promotion of bacterial infections following IFV infection.

Previous studies have pointed out that the most common bacterial infections after IFV infection included Streptococcus pneumoniae [40], Staphylococcus aureus and Klebsiella pneumoniae [41]. In particular, there is a synergistic effect between Streptococcus pneumoniae and IFV [42]. Acinetobacter baumannii has also been identified as a common pathogen in adults with severe pneumonia and IFV infection [43]. In our study, we observed that Acinetobacter baumannii was the top bacterial species in the IFVP group, whereas Streptococcus pneumoniae was the top species in the IFVN group. Regardless of the IFV status, we found the most abundant bacteria among all the severe pneumonia patients were Acinetobacter baumannii, Streptococcus pneumoniae, and Escherichia-Shigella. Numerous studies have investigated the detection rate of bacteria and viruses among pneumonia patients in different regions. For example, an epidemiological study reported that the top three bacteria among Chinese adults with CAP were Streptococcus pneumoniae, Haemophilus influenzae, and Klebsiella pneumoniae [44]. A systematic review and meta-analysis suggested that Klebsiella pneumoniae, Streptococcus pneumoniae, and Escherichia coli were the most frequently detected bacterial agents among children under 5 years with CAP in China [45]. Another study carried out in **amen, China showed that Haemophilus influenzae, Streptococcus pneumoniae, Staphylococcus aureus, and Klebsiella pneumoniae were the most common bacteria among children with severe pneumonia [46]. In addition, Staphylococcus aureus, Pseudomonas aeruginosa, and Streptococcus pneumoniae were the most frequent pathogens among critically ill cancer patients with severe pneumonia. Therefore, Streptococcus pneumoniae is a consistent common bacterium among pneumonia patients across these studies, as well as in our study. It is recognized as the most important pathogen of CAP [47], and has the highest detection rate among adults with severe pneumonia after IFV infection [48]. However, different from the above studies, we found that Acinetobacter baumannii was the most frequent bacterium among adult patients with severe pneumonia. Acinetobacter baumannii is a multidrug-resistant pathogen, a major cause of nosocomial infections, with a higher occurrence rate in Asia [49, 50]. Wong et al. reported that Acinetobacter baumannii was a frequent cause of CAP in multiple countries and areas [51]. Therefore, the efforts for better surveillance and control of this bacterium should be strengthened, and targeted treatment plans need to be implemented for patients to overcome its bacterial resistance.

Klebsiella was observed to be more abundant in the IFVN group compared to the IFVP group in this study. Notably, a previous study showed that pre-infection with Klebsiella limited the excessive innate immune response induced by subsequent IFV infection and thereby protected mice from death [52]. However, another mouse model study reported Klebsiella pneumoniae infection following H9N2 IFV-A infection contributed to the development of pneumonia [53]. These conflicting findings highlight the need for further investigation to determine whether similar effects of Klebsiella exist in humans.

This study has some limitations that need to be acknowledged. Firstly, the relatively small sample size of our study might limit the generalizability of our findings. Secondly, the samples in our study were obtained from multiple hospitals, and the varying control measures for nosocomial infections in different hospitals might have affected the bacterial diversity and abundance to some extent. Thirdly, due to the cross-sectional nature of our study, the temporal relations between IFV infection and severe pneumonia occurrence could not be determined, and causality between them cannot be inferred. Therefore, a follow-up prospective trial is required to address this issue. Fourthly, although 16S-rDNA sequencing technology was used in this study, it provided limited taxonomic resolution at the species level, and it cannot provide absolute abundance of pathogens. Therefore, qPCR could be performed to investigate specific bacteria that interact with the IFV in the progression of pneumonia. Fifthly, due to the limited sample size of only 10 patients in both IFVN and IFVP groups who had used antibiotics, we did not stratify the data by antibiotic use status to investigate its influence on the microbiomes of pneumonia patients. However, the comparability of the two groups would not influence the results of differences in bacterial characteristics in the two groups. Additionally, although measures have been taken to control contamination, there is still a possibility of oral or environmental pollution from potential sources that cannot be entirely eliminated.

Conclusions

In summary, our study revealed differences in bacterial diversity and relative abundance between severe pneumonia patients with and without IFV infection. Severe pneumonia patients with IFV infection may be more susceptible to bacteria. Acinetobacter baumannii was the most abundant bacterium in the IFVP group and the overall samples, highlighting the urgency and necessity of bacterial surveillance and control in hospitals and communities. Our results shed new lights on the roles of IFV infection in the microbiome distribution among severe pneumonia patients. However, the mechanism underlying the interaction between IFV and Acinetobacter baumannii in the progression of pneumonia needs further investigation. These results provide valuable insights for the management and treatment of severe pneumonia patients, especially those with IFV infection.