Skip to main content
Download PDF

The Staphylococcus aureus-antagonizing human nasal commensal Staphylococcus lugdunensis depends on siderophore piracy

Microbiome volume 12, Article number: 213 (2024) Cite this article

Abstract

Background

Bacterial pathogens such as Staphylococcus aureus colonize body surfaces of part of the human population, which represents a critical risk factor for skin disorders and invasive infections. However, such pathogens do not belong to the human core microbiomes. Beneficial commensal bacteria can often prevent the invasion and persistence of such pathogens by using molecular strategies that are only superficially understood. We recently reported that the commensal bacterium Staphylococcus lugdunensis produces the novel antibiotic lugdunin, which eradicates S. aureus from the nasal microbiomes of hospitalized patients. However, it has remained unclear if S. lugdunensis may affect S. aureus carriage in the general population and which external factors might promote S. lugdunensis carriage to enhance its S. aureus-eliminating capacity.

Results

We could cultivate S. lugdunensis from the noses of 6.3% of healthy human volunteers. In addition, S. lugdunensis DNA could be identified in metagenomes of many culture-negative nasal samples indicating that cultivation success depends on a specific bacterial threshold density. Healthy S. lugdunensis carriers had a 5.2-fold lower propensity to be colonized by S. aureus indicating that lugdunin can eliminate S. aureus also in healthy humans. S. lugdunensis-positive microbiomes were dominated by either Staphylococcus epidermidis, Corynebacterium species, or Dolosigranulum pigrum. These and further bacterial commensals, whose abundance was positively associated with S. lugdunensis, promoted S. lugdunensis growth in co-culture. Such mutualistic interactions depended on the production of iron-scavenging siderophores by supportive commensals and on the capacity of S. lugdunensis to import siderophores.

Video Abstract

Conclusions

These findings underscore the importance of microbiome homeostasis for eliminating pathogen colonization. Elucidating mechanisms that drive microbiome interactions will become crucial for microbiome-precision editing approaches.

Introduction

Host-associated microbiomes are shaped by mutualistic or antagonistic interactions among microbiome members. Some of these members depend on each other because they collaborate in the degradation and utilization of complex nutrient sources or the exchange of essential cofactors [1]. These processes can be of mutual benefit for two partners or can support only one of them, which may exploit the other bacterium’s metabolic capacities. Interacting microorganisms can also antagonize each other directly by the release of antimicrobial bacteriocins, which inhibit major competitors to enhance the producer’s fitness [2]. Such mutualistic and antagonistic mechanisms govern complex, often multi-dimensional interaction networks, which have remained largely unexplored.

Elucidating mechanisms that can promote the persistence of beneficial and impair that of harmful bacteria in microbiomes are of particular relevance for the prevention of infections caused by bacterial pathogens that use human microbiomes as their major reservoirs [3]. All of the notoriously antibiotic-resistant pathogens, vancomycin-resistant Enterococcus faecalis and Enterococcus faecium, methicillin-resistant Staphylococcus aureus (MRSA), and carbapenemase- or extended-spectrum beta-lactamase-producing Klebsiella pneumoniae, Acinetobacter baumannii, and Escherichia coli, can be found in the microbiomes of healthy humans or of at-risk patients [4]. Carriage of such antibiotic-resistant pathogens strongly increases the risk of invasive, difficult-to-treat infections [5]. Microbiome composition has a dominant role in the capacity of pathogens to colonize, most probably as a consequence of antagonistic effects exerted by beneficial commensals [6]. However, current options for the decolonization of pathogens are very limited, which demands the development of effective and specific pathogen eradication regimes, which should maintain microbiome integrity [7].

S. aureus colonizes the nares of 30% to 40% of the human population, which represents a major risk factor for severe S. aureus infections, in particular when caused by MRSA strains [8, 9]. The abundance of IgA in the nose has been found to shape bacterial densities in the nose but not the presence or absence of S. aureus or other nasal species [10]. The nasal microbiome is highly diverse, and its composition is crucial for the capacity of S. aureus to colonize [11]. Based on the most prevalent species or genera within the nasal microbiome, Liu et al. defined seven community state types (CSTs). Whereas CST1 is dominated by S. aureus, its abundance in other CSTs is strongly reduced, most likely due to the presence of specific microbiome members, or even absent as in CST2 which is Enterobacteriaceae-dominated [11]. Nevertheless, we are only beginning to understand the mechanisms used by nasal commensals to exclude S. aureus.

We recently reported that the production of bacteriocins is very frequent among commensal bacterial species of the human nasal microbiomes suggesting that such mechanisms should have a strong impact on microbiome composition and, potentially, the capacity to exclude specific pathogens [12, 13]. Notably, many of these compounds did not or not only act against closely related species, which contradicts previous definitions of bacteriocins’ roles and indicates that bacteriocins are often produced to combat major competitors, irrespective of the relatedness of bacteriocin producer and target strain [2]. One of the antimicrobial molecules, the novel fibupeptide lugdunin, produced by the coagulase-negative Staphylococcus (CoNS) species Staphylococcus lugdunensis, was explored in detail [14]. Almost all of the nasal S. lugdunensis isolates contained the lugdunin-biosynthetic gene cluster in the chromosome. S. lugdunensis eradicated S. aureus in a lugdunin-dependent fashion, during cultivation in laboratory media and in animal models. Moreover, hospitalized patients carrying S. lugdunensis in their noses had a 6-fold lower risk of being colonized by S. aureus. However, only 9.1% of the hospitalized patients were S. lugdunensis carriers, raising the question of why only certain nasal microbiomes may permit the persistence of S. lugdunensis with its S. aureus-eradicating activity [14].

Here we describe that S. lugdunensis is equally rare in healthy humans as in hospitalized patients and promotes the elimination of S. aureus in both groups of humans. A long-run analysis showed the persistence of S. lugdunensis colonization in about 50% of initial carriers. S. lugdunensis was positively associated with several other commensals, which were required for S. lugdunensis to thrive by providing essential iron-scavenging siderophores.

Methods

Bacterial strains and media

Bacterial strains used in this study are summarized in Supplementary Table S3. All bacteria were grown on basic medium (BM; 1% soy peptone, 0.5% yeast extract, 0.5% NaCl, 0.1% glucose, and 0.1% K2HPO4, pH 7.2), which was supplemented with 5% sheep blood (Oxoid) and 1.5% agar (BD European Agar) if needed. Cutibacterium acnes and Lawsonella clevelandensis were incubated under anaerobic conditions using an anaerobic jar and AnaeroGen™ (Thermo). For the phenotypic identification of S. lugdunensis, S. aureus, and other staphylococcal species in nasal swabs, basic medium (BM), blood agar, and selective agar SSL [15] were used. For the siderophore experiments, bacteria were grown in E-BHI, composed of brain heart infusion medium (BHI; Roth) supplemented with 10 µM of the iron chelator ethylenediamine-di(o-hydroxyphenylacetic acid) (EDDHA; LGC Standards GmbH), for 72 h at 37 °C and constant shaking at 160 rpm. For transformation experiments with E. coli DC10B and S. lugdunensis, tryptic soy broth (TSB; Oxoid) or tryptic soy agar (TSA) were used and, when necessary, supplemented with antibiotics at concentrations of 10 µg/ml chloramphenicol (Sigma-Aldrich), 100 µg/ml ampicillin (Roth) and 1 µg/ml anhydrotetracycline (Fluka).

Human volunteer selection, nasal swabbing, and detection of S. lugdunensis and S. aureus

The sample collection procedures were approved by the clinical ethics committee of the University of Tübingen (No. 109/2009 BO2) and informed written consent was obtained from all volunteers. Nasal swabs were taken exclusively from healthy adults not showing obvious symptoms of any disease. Samples from 270 such volunteers, mostly students from practical courses as well as staff members from different departments of the University of Tübingen were collected by swabbing both nares with cotton swabs and suspending them in 1 ml phosphate buffered saline (PBS). Various dilutions of each sample were plated on BM, blood agar, and selective S. lugdunensis medium agar (SSL) [15] for phenotypic identification of S. lugdunensis, S. aureus, and other staphylococcal species. The plates were incubated for 24 to 48 h at 37 °C under aerobic and anaerobic conditions, respectively. The bacterial identity was evaluated by matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry (mass spectrometer: AXIMA Assurance, Shimadzu Europa GmbH, Duisburg, database: SARAMIS with 23.980 spectra and 3.380 superspectra, BioMérieux, Nürtingen).

Metagenome sequencing of nasal microbiome samples

Nasal microbiome samples were taken from individuals by swabbing both nares successively with one nylon-flocked E-Swab (ThermoFisher Scientific) and suspending them in 1 ml Amies transport medium. Two replicate swabs were consecutively taken per volunteer. For degradation of contaminating host DNA and subsequent preparation of bacterial DNA, the suspended samples were mixed with AHL buffer and treated with the QIAamp DNA Microbiome Kit (Qiagen) according to the manufacturer’s instructions. Finally, DNA was eluted from the spin columns in 50 µl AVE buffer. DNA concentration was determined with Qubit 3.0 (Thermo Fisher Scientific) using High Sensitivity (HS) reagents for low DNA concentrations. 2.5 µl of the preparation was used for whole-genome amplification with REPLI-g Single Cell Kit (Qiagen). Subsequently, the amplified DNA was purified with Genomic DNA Clean & Concentrator (Zymo Research) and the final DNA concentration was determined with Qubit 3.0.

About 10 µg of the purified DNA was used for next-generation sequencing by GATC (Konstanz, Germany). In brief, libraries of fragments of ca. 400 bases were prepared and submitted to Illumina paired-end sequencing with read lengths of 150 bases. After quality filtering, an average of 30 million read pairs was obtained per sample and provided as a compressed fastq.gz file (separate files for each end of the sequenced inserts).

In the first metagenome analysis (after 18 months) we obtained a median of 70.13 million quality-filtered, paired-end reads per sample. In the second metagenome analysis (after 23 months), a median of 52.71 million reads per sample was obtained. Despite taking measures to reduce the proportion of human DNA in the samples we found varying percentages of metagenomic reads mapping to the human DNA reference database: at the first time point (18 months), human-derived reads ranged from 14.4 to 94.3% (median 74.75%) per sample while at time point 2 (23 months) between 0.5% and 90.9% (median 28.15%) of the reads mapped to sequences from the hg19 database. At time point 1, a median of 11.45 million bacterial reads per sample (sample D had exceptionally low read numbers due to antibiotic treatment and was left out as an outlier) was obtained, while at the second time point the median of bacterial reads was 21.81 million.

Bioinformatics

The sequence reads from each file was mapped against the NCBI non-redundant (nr) protein database by using the optimized BLASTX algorithm of DIAMOND [16]. For the subsequent functional and taxonomic analysis with MEGAN6 [17], the DIAMOND alignment files were “meganized” by combining the paired-end sequences for each sample and providing them with functional annotations. Taxonomic binning, functional analysis, and statistical analyses were performed by using the corresponding functions of MEGAN6. For the comparison of the metagenomes on the genus level, pairwise distances were calculated based on ecological distance according to Hellinger [18].

To exploit the genetic information obtained with the metagenome sequence reads we assembled the species-specific reads into larger contiguous sequences. To this end, we used the MetaSpades tool, which is implemented in the SPADES assembler [19]. For species-specific read assemblies, we first used MEGAN6 to extract the reads assigned to the species of interest as a Fasta file. Then, we used MetaSpades to assemble these reads into contigs.

In order to determine the S. lugdunensis strains that reside in the anterior nares of the carriers we selectively extracted the S. lugdunensis reads where possible and assembled them into contigs. The contigs were compared with a database composed of S. lugdunensis genome sequences by BlastN and the genomes getting the highest number of hits were regarded as the closest relative(s) to the strains present in the S. lugdunensis carriers. Thus, we obtained strain profiles based on the total alignment lengths of hits mapped to a phylogenetic tree of S. lugdunensis strains (Fig. 2).

Correlation assays were performed by network analysis with the web-based tool MetagenoNets, (https://web.rniapps.net/metanets/; [20]). To this end, an abundance table with read numbers of the bacterial species in the samples was uploaded to MetagenoNets. The abundance data were transformed to centered log ratios, filtered for a prevalence (minimum proportion of reads in samples) of 0.1%, and an occurrence (minimum percentage of samples showing the species) of 20%. Network inference was calculated by NAMAP (a modified ReBoot method; [21, 22]) based on Pearson correlation coefficients. The results were downloaded as a correlation matrix, imported into R [23] using the package corrr [24], and visualized with the R package ggplot2 [25].

Siderophore production and detection

Bacterial strains were incubated on BM blood agar plates from 24 to 48 h at 37 °C. Subsequently, bacterial material was scratched from the plates, washed once with E-BHI medium, adjusted to an optical density at 600 nm (OD600) of 0.05, and incubated in 1 ml E-BHI for 72 h at 37 °C and constant shaking at 160 rpm in 24 well plates. In the case of Corynebacterium accolens, Corynebacterium tuberculostearicum, Corynebacterium aurimucosum, and Corynebacterium kroppenstedtii, the medium was additionally supplemented with 0.4% Tween80 (Sigma-Aldrich); C. acnes and L. clevelandensis were incubated in 1.5-ml reaction tubes to minimize gas exchange with atmospheric oxygen; Dolosigranulum pigrum was incubated with 10% sterile-filtered spent medium of C. accolens to enhance growth [26]. After 72 h of incubation, the OD600 was determined, the bacterial cultures were centrifuged, and the supernatants were sterile-filtered using a 0.22-µm filter (Millex) to obtain the spent media. In order to quantify siderophore concentrations in the spent media, a siderophore detection kit (SideroTec Assay™ from Emergen Bio) was used according to the manufacturer’s instructions with small alterations; instead of 100 µl samples, 50 µl samples were used and mixed with 50 µl Chelex treated MilliQ-purified H2O in order to reduce background signals of the medium.

Analysis of growth support for S. lugdunensis under iron-restriction by xenosiderophores produced by nasal bacteria

In order to generate iron-restricted BHI agar plates, 2 × BHI was treated with Chelex 100 resin (Sigma-Aldrich) for 24 h at 4 °C and supplemented with 40 µM EDDHA. After sterile filtration, the medium was heated to 50 °C and mixed 50:50 with 2 × agar (50 °C) containing 20% complement-inactivated horse serum (10% final concentration) (Sigma-Aldrich) to provide an iron source with complexed iron. Freshly grown S. lugdunensis was picked from blood agar plates, washed once with PBS, adjusted to an OD600 of 0.5, and streaked evenly onto the BHI agar plates. To evaluate the growth support for S. lugdunensis by the siderophore-containing spent media, 10 µl aliquots were spotted on previously plated S. lugdunensis cells as follows. Depending on the endpoint OD600 of each culture, different volumes of spent medium were taken and mixed, if necessary, with fresh E-BHI to a final volume of 10 µl (for instance, from a culture with an endpoint OD600 of 0.2, 10 µl spent medium was taken whereas, from a culture with an endpoint OD600 of 2.0, 1 µl spent medium was taken and mixed with 9 µl E-BHI medium). After spotting 10 µl of each spent medium, the plates were incubated for 20 h at 37 °C. If the spent media contained siderophores that could promote the growth of S. lugdunensis, zones of enhanced growth appeared that were characterized according to their diameters.

Generation and complementation of S. lugdunensis and Staphylococcus epidermidis mutants

For the construction of S. lugdunensis and S. epidermidis mutants, the thermosensitive plasmid pIMAY [27] and the primers listed in Supplementary Table S4 were used. Targeted mutagenesis was performed as described in [27]. In brief, 500 bp DNA fragments up- and downstream of the ferric hydroxamate uptake (fhu) gene locus for S. lugdunensis and of the staphyloferrin A biosynthetic genes (sfaDAB) for S. epidermidis were amplified by PCR using the corresponding primers ((1)/(2) and (3)/(4) for fhu knockout, (9)/(10) and (11)/(12) for sfaDAB knockout). The resulting PCR products contained overlapping sequences, which facilitated hybridization in order to construct the gene deletion sequence. This DNA sequence was inserted into pIMAY and digested via SacI/KpnI (Thermo) for fhu knockout and via SalI/SacI (Thermo) for sfaDAB knockout, using Gibson assembly (New England BioLabs) according to the manufacturer’s instructions. After transformation of E. coli DC10B and isolation of the plasmid construct, S. lugdunensis or S. epidermidis were transformed via electroporation and incubated at 30 °C in TSB with 10 µg/ml chloramphenicol. Selection for plasmid integration into the chromosome was performed at 37 °C in TSB with 10 µg/ml chloramphenicol, and positive integration clones were subsequently grown at 30 °C without chloramphenicol to promote excision of the plasmid. The loss of plasmid was selected on TSA containing 1 µg/ml anhydrotetracycline and correct S. lugdunensis ∆fhu or S. epidermidis ∆sfaDAB knock-out clones were confirmed by chloramphenicol susceptibility testing and PCR (primers (5)/(6) for ∆fhu, (13)/(14) for ∆sfaDAB).

For complementation in S. lugdunensis, primers (7)/(8) were used to amplify the wild-type fhu gene. The DNA was inserted into SalI/XbaI digested pRB474 using Gibson assembly and transformed into E. coli DC10B. Subsequently, the complementation plasmid pRB474_fhu was used to transform S. lugdunensis ∆fhu. Chloramphenicol-resistant transformants were sub-cultured and correct complementation constructs were confirmed via PCR using primers (15)/(16).

Lugdunin susceptibility assay

Bacterial strains were incubated on BM blood agar plates for 24 h to 48 h at 37 °C. Subsequently, bacterial material was scratched from the plates, washed once with 1 × PBS, adjusted to an optical density at 600 nm (OD600) of 0.5, and streaked evenly onto BM sheep blood agar plates. Lugdunin dilutions in DMSO were prepared as follows: 0 µg/ml, 10 µg/ml, 20 µg/ml, 30 µg/ml, 40 µg/ml, and 50 µg/ml. Of each concentration, 2 µl were spotted onto the lawn of bacteria (0 to 100 ng lugdunin per spot). After drying, bacteria were incubated for 24 h to 48 h at 37 °C (anaerobic bacteria were incubated under anaerobic conditions as described in “bacterial strains and media”). After incubation, the spot with the lowest amount of lugdunin showing a clear inhibition was used to determine lugdunin susceptibility.

Results

S. lugdunensis colonization reduces the risk of S. aureus carriage in healthy volunteers 5-fold

We wondered if our recent report on a strong negative correlation between nasal colonization by S. lugdunensis and S. aureus [14] might have been confounded by underlying health problems or prior antibiotic treatment in the hospitalized patients included in this group. To assess the relation between S. lugdunensis and S. aureus in non-healthcare-associated humans a cohort of 270 healthy human volunteers was analyzed for nasal carriage of one or both bacterial species. The average age was 26.7 years and 63.7% of the participants were women. Nasal swabs were plated on selective media that allowed the specific detection of either S. aureus or S. lugdunensis. The identity of representative colonies from each donor was also verified by matrix-assisted laser desorption/ionization time-of-flight mass spectroscopy (MALDI-TOF).

The S. aureus carriage rate was 29.3%, which is very close to that of the hospitalized patient cohort and other previous studies [8, 14]. 6.3% of the participants were colonized by S. lugdunensis (Table 1), which is slightly lower than for the hospitalized patient cohort (9.1%) [14]. Nasal carriage of S. lugdunensis was much higher in males (11.9%) compared to female participants (2.9%). Notably, only one of the 270 participants was colonized by both, S. aureus and S. lugdunensis, which corresponds to a 5.2-fold lower propensity of S. lugdunensis-positive humans to carrying S. aureus compared to S. lugdunensis-negative humans. This ratio corresponds well to the 5.9-fold reduced risk of S. lugdunensis carriers being colonized by S. aureus, reported for the hospitalized patient’s cohort [14]. All S. lugdunensis isolates from healthy volunteers contained the lugdunin gene cluster and all S. aureus isolates were highly susceptible to lugdunin. Thus, the negative correlation between S. lugdunensis and S. aureus is the same in healthy and hospitalized human populations.

Table 1 S. aureus and S. lugdunensis distribution in study participants
Full size table

S. lugdunensis colonization is more frequent than estimated and remains largely stable over time

To analyze the dynamics of nasal S. lugdunensis persistence or loss, the colonization status of eight initial carriers was monitored at four-time points for a period of 23 months (probands A-H; Fig. 1A). In addition, four initial S. lugdunensis non-carriers were included as random controls (probands I-L). One of the latter was an S. aureus carrier (proband I). In four participants (A-C, F) S. lugdunensis was detected at all four time points by cultivation on agar plates. In one carrier (G) S. lugdunensis was not detected after 7 months but it reemerged after 18 and 23 months. In two (D, E) or one (H) carrier(s), S. lugdunensis could not be cultivated at two or three subsequent time points, respectively. Interestingly, S. lugdunensis could also be cultivated from the noses of all four initially negative participants after 7 months and, additionally, in the nose of one initially negative participant (K) after 23 months (Fig. 1A). These data suggested that S. lugdunensis may have a pronounced capacity to persist in the noses of some humans (here at least 50% of initial carriers) while it temporarily emerges and disappears in others.

Fig. 1
figure 1

Detection of S. lugdunensis and S. aureus in the nose of study participants by cultivation or metagenome analysis. A Culture outcome for S. lugdunensis and S. aureus in the nasal samples and abundance of metagenome reads for the two species. For each time point and each proband, the culture results for S. lugdunensis (circles filled in grey) and S. aureus (circles filled in yellow) are shown (the initially culture-negative group of probands I–J, which was negative for S. lugdunensis at the first time point, is separated by two dashed lines). A negative culture outcome is indicated by empty circles. For time points 18 and 23 months, also the relative DNA abundances (= rel. proportions of reads specifically for S. lugdunensis and S. aureus) in the metagenomes are indicated by color-coded squares. The color intensity indicates the relative DNA abundance compared to the corresponding median value of DNA abundance (see legend on the right). B Relative proportions (in percent) of S. lugdunensis or S. aureus metagenome reads in microbiomes with positive or negative culture outcomes. Left: relative proportions of S. lugdunensis reads versus culture outcome. The median percentage for culture-positive samples: 0.33%; the median percentage for culture-negative samples: 0.09% (Mann-Whitney test, p = 0.0159). Right: relative proportions of S. aureus reads versus culture outcome. Median percentage for culture-positive samples: 47.27%; median percentage for culture-negative samples: 1.34% (Mann-Whitney test, p = 0.0072). Relative read proportions are shown on the y-axes in the log scale. Medians are indicated by horizontal lines with 95% confidence intervals

Full size image

The nasal swabs were also analyzed for viable S. aureus cells. One of the S. lugdunensis carriers (F) became an S. lugdunensis-S. aureus co-carrier at months 18 and 23 (Fig. 1A). The initial S. aureus carrier from the control group (I) lost its S. aureus carrier status at months 18 and 23 while one of the initial non-carriers (K) became and remained positive after 7 months. Co-carriage of S. aureus and S. lugdunensis was observed at individual time points in two participants (I, K) of the random control group.

At two time points (18 and 23 months), the colonization status was also assessed by metagenome analysis of nasal swabs. To this end, bacterial isolates from nasal swabs were disintegrated, contaminating host DNA was degraded, and the remaining DNA was amplified and shotgun-sequenced on an Illumina system. An average of 30 million read pairs was obtained per sample, which was mapped to the NCBI nr protein database by DIAMOND and analyzed with the MEGAN6 algorithm. S. lugdunensis genome DNA was indeed found in most of the culture-positive and some additional samples (Fig. 1A) with a median relative abundance of S. lugdunensis-specific reads of 0.33% ranging from 0.03% to 3.37% (Supplementary Table S1). In only one of eleven culture-positive samples no S. lugdunensis DNA was found (Fig. 1A). Surprisingly, S. lugdunensis DNA was also detected in several of the culture-negative samples, at a low median relative abundance of 0.09% (range between 0.04% and 0.20%) indicating that S. lugdunensis may have remained present in most of the original carriers but, occasionally, at very low numbers, which did not reach the cultivation-based detection limit. Of the four original S. lugdunensis carriers who were culture-negative at one or more of the subsequent time points, three contained S. lugdunensis DNA even at culture-negative time points. Conversely, S. lugdunensis DNA was absent in the noses of two of the S. lugdunensis culture-negative control participants while two others contained some S. lugdunensis DNA with a relative abundance between 0.04% and 0.27% (the latter revealing an S. lugdunensis/S. aureus co-culture-positive state) (Fig. 1, Supplementary Table S1). Thus, S. lugdunensis might be more frequently colonizing human noses than estimated by swab cultivation. Samples with a positive culture outcome for S. lugdunensis revealed a significantly higher relative abundance of S. lugdunensis reads than those that were culture-negative (0.33% versus 0.09% median relative abundance, Mann-Whitney test, p = 0.0159, Fig. 1B). Thus, the capacity to cultivate S. lugdunensis from carriers appears to depend on a threshold relative abundance in metagenomes, which probably lies somewhere between the median abundances observed for the cultivation-positive and cultivation-negative groups.

The metagenome analysis revealed similar findings for S. aureus as for S. lugdunensis-cultivation-positive samples had also high percentages of nasal S. aureus DNA with a median relative abundance of 47.3% (ranging from 2.0 to 91.4%) but even all culture-negative samples contained traces of S. aureus DNA (median relative abundance of 1.3%, ranging from 0.06 to 6.2%, Fig. 1A, Supplementary Table S1). These data suggest that S. aureus carriage is much higher in the human population than estimated via subcultivation of nasal swabs and that culture positivity depends on a threshold DNA abundance in metagenomes of above 2.0%. It is interesting to note that the threshold abundance for successful cultivation is obviously much higher for S. aureus than for S. lugdunensis (Fig. 1B).

The S. lugdunensis contigs from each consistently colonized proband (A, B, C, F, and K) matched with specific S. lugdunensis genomes from databases of only one or two clonal complexes (CC1, CC3, CC6, CC7, or a currently undefined CC) per proband (Fig. 2) suggesting that a given human is usually colonized by one or only a few S. lugdunensis clones. This association remained largely the same at the two sampling times, pointing to a stable clonal distribution (samples G-2, K; see Fig. 2).

Fig. 2
figure 2

S. lugdunensis-containing metagenomes include only one or a few S. lugdunensis clones. S. lugdunensis strain profiles based on BLASTN comparisons of contigs from assembled metagenome reads with S. lugdunensis genome sequences. Sample numbers (1 or 2) indicate the corresponding metagenome analysis after 18 or 23 months, respectively. Blue bars represent the percentages (top 10%) of total hit alignment lengths between sample contigs and the corresponding S. lugdunensis strain (thresholds: 99.9% identity, 99.9% contig length in each hit alignment, only those contigs with a length of at least 250 nucleotides were considered). Red bars indicate the highest percentage of total alignment length in the corresponding sample (see the legend on the top of the right side). The phylogenetic tree of S. lugdunensis strains was downloaded from the NCBI database as a Nexus file and edited with Dendroscope [28]. Please note that genome data for S. lugdunensis NCTC12217 are presented twice in the NCBI tree as complete and as a draft genome sequence. The strain profiling presented here is based on the complete genome sequence (NCTC12217). Volunteers who tested positive for S. lugdunensis in the accompanying culture assays are indicated by grey rectangles, co-carriers by green rectangles, S. aureus carriers by yellow rectangles, and non-carriers by non-filled rectangles (see the legend on the right side). Microbiomes of the CST3/5, CST1, and CST7 types are labeled accordingly. Samples labeled with areas shaded in light grey yielded no analyzable assemblies of the S. lugdunensis reads. Sequence types (ST) and corresponding clonal complexes (CC) of the profiled S. lugdunensis strains were obtained from https://bigsdb.web.pasteur.fr/staphlugdunensis. nd, no CST or ST/CC assignment is possible

Full size image

S. lugdunensis is associated with CSTs three, five, and seven

Based on the metagenome data the nasal microbiome composition of the two samples from twelve participants was elucidated and compared with the previously reported CST classification [11]. Notably, most of the metagenomes could be allocated to one of the seven CSTs (Fig. 3A, Supplementary Table S1). Most S. lugdunensis carriers could be assigned to CST3 or CST5, which are dominated by S. epidermidis or Corynebacterium sp., respectively, at both (participants A, B, E) or at least one-time point (F, G, H). Several of these samples changed over time from CST5 to CST3 or vice versa. The two CSTs had very similar composition and clustered in principal component analysis (Fig. 3B). The dominance by either S. epidermidis or Corynebacterium sp. appears to vary over time, suggesting that CST3 and CST5 represent dynamic states of the same CST, which we refer to as CST3/5 from now on. Participant D had undergone systemic antibiotic therapy before the first microbiome sampling time point, which was probably the reason for the unusual metagenome composition, which changed from a non-CST-classifiable microbiome composition dominated by Streptococcus and Staphylococcus to an unprecedented nasal community dominated by Haemophilus influenzeae. The nasal microbiome of participant F was dominated by Moraxella, Streptococcus, and Peptoniphilus and could also not be assigned to one of the established CSTs (Fig. 3A). S. lugdunensis associated most consistently with CST3/5, and all CST3/5 samples (except H-1) were found to contain S. lugdunensis genomic DNA at above-average abundances between 0.1% and 3.4% (median 0.3%) (Supplementary Table S1). The four initially culture-negative participants had CST7, CST1, or CST3/5, which were largely the same at the two different time points. Genomic DNA amounts in CST1, which is defined by S. aureus dominance, revealed very high S. aureus DNA abundances of 91.4 and 67.5%.

Fig. 3
figure 3

Most of the S. lugdunensis-colonized study participants can be assigned to CST3/5 and CST7. A Nasal microbiome profiles at the genus level. Relative proportions of bacterial genera present in the nasal microbiomes at time points 18 months and 23 months are presented by stacked bar charts. The different bacterial genera are indicated by colours as shown in the legend on the right. The presence of S. lugdunensis reads in the metagenomes is indicated (Slu). The microbiome profiles that could be assigned to the community state types (CST) published by Liu et al. [11] are labeled accordingly. Profiles not fitting to one of these types are designated “nd”. B Clustering of the microbiome genus profiles by principal coordinates analysis. Comparison of the microbiome profiles at genus level by principal coordinates analysis based on evolutionary distances according to Hellinger [18]. The data showed the highest variation along principal component 1 (PC1, 52.6%). Principal component 2 (PC2) represents a variation of 15.0%. The microbiomes are represented by colored circles and numbers correlating with the corresponding metagenome analysis time point (−1, 18 months or −2, 23 months). Clusters representing the two main microbiome types CST3/5 and CST7 are labeled correspondingly. Vectors drawn and labeled in light grey point towards the prevalent genera in the clusters/subclusters. Note that sample K-2 clusters with the CST3/5 microbiomes but is de facto regarded as CST1 because the Staphylococcus proportion is mainly composed of S. aureus

Full size image

In addition to S. epidermidis, S. lugdunensis, and S. aureus, more than 50% of the metagenomes contained Mammaliicoccus sciuri (formerly Staphylococcus sciuri), albeit at low numbers, and at least one third also traces of other CoNS such as Staphylococcus capitis and Staphylococcus caprae (Supplementary Figure S1). The microbiomes contained a remarkable diversity of Corynebacterium species, but the abundance of specific species was similar for a specific CST (Supplementary Figure S2). CST3/5 was dominated by Corynebacterium accolens with lower representation by Corynebacterium tuberculosteraricum, Corynebacterium diphteriae, and others (Table 2, Supplementary Figure S2). In contrast, CST7 contained major amounts of genomic DNA from Corynebacterium propinquum and Corynebacterium pseudodiphtheriticum (Table 2, Supplementary Figure S2). In conclusion, nasal S. lugdunensis carriage was associated with a high abundance of S. epidermidis, C. accolens, C. tuberculostearicum, and in one study participant, D. pigrum, C. propinquum, and C. pseudodiphtheriticum.

Table 2 Bacterial species dominating the S. lugdunensis-associated CST3/5- and CST7-type microbiomes. Species that occur in all microbiomes of the indicated CST type are listed by decreasing median relative abundances (%) of species-specific metagenome reads. A threshold of 0.06% relative abundance was applied to the co-occurrence analysis with MEGAN6
Full size table

Quantitative metagenomic analysis permitted the assignment of bacterial species, which were negatively or positively correlated with the presence of S. lugdunensis. Based on a centered log-ratio-transformed abundance table of species reads we performed correlation network analyses with the Namap/Pearson correlation inference option of MetagenoNets [20] with thresholds for prevalence (0.1%) and occurrence (20%) (see “Methods” section). Based on these parameters, we obtained correlation measures for 28 species throughout all 24 samples (Fig. 4, Supplementary Table S2). We observed correlation coefficients for S. lugdunensis ranging from −0.34 corresponding to the highest negative association with Streptococcus pneumoniae to +0.50, the strongest positive correlation with the Clostridium-related Bacillota (formerly Firmicutes) Finegoldia magna (Fig. 4). Most of the species dominating ST3/5 (Table 2) were also positively associated with S. lugdunensis. The data also revealed a strong negative association of S. lugdunensis with S. aureus (correlation coefficient −0.22) confirming our previous studies on the antagonistic interplay between these species [14] (Fig. 4). Moreover, several species, which are not typical nasal colonizers, found only at trace amounts were negatively associated with S. lugdunensis. In addition to the positive association with Corynebacterium tuberculostearicum, several other Actinomycetota including Cutibacterium acnes, Corynebacterium striatum, Corynebacterium efficiens, and the CoNS S. epidermidis and S. capitis, showed positive correlations with S. lugdunensis (Fig. 4. Supplementary Table S2).

Fig. 4
figure 4

Specific bacterial species are positively or negatively associated with the abundance of S. lugdunensis reads in nasal metagenomes. Correlation coefficients were inferred based on a centered log-ratio transformed abundance table of species reads by correlation network analyses using the Namap/Pearson algorithm of MetagenoNets [20] with thresholds for prevalence and occurrence of 0.1% and 20%, respectively. Based on these parameters, we obtained correlation measures for 28 species throughout all 24 samples. Shown are the associations of S. lugdunensis with other microbiome species (shown at the x-axis). Positive associations are indicated by green bars and negative associations by red bars. The corresponding Pearson correlation coefficients are indicated at the x-axis

Full size image

The presence and abundance of S. aureus metagenome reads correlated negatively with S. lugdunensis, and other species, which have been shown to interfere with S. aureus colonization, including D. pigrum, C. propinquum, S. pneumoniae, and F. magna (Supplementary Figure S3) [11, 29, 30]. In contrast, positive correlations for S. aureus were found for the Actinomycetota Lawsonella clevelandensis, Corynebacterium aurimucosum, the CoNS S. epidermidis and S. capitis, and the Clostridium-related Bacillota Anaerococcus sp. (Supplementary Figure S3, Supplementary Table S2).

Commensals dominating the S. lugdunensis-positive CST3/5 or CST7 support S. lugdunensis growth

The positive or negative association of S. lugdunensis with specific microbiome members raised the question, of which mechanisms might be the basis for the antagonistic or mutualistic interactions. Some of the species among the typical nasal microbiome members, which were strongly negatively associated with S. lugdunensis, S. pneumoniae, and S. aureus (Fig. 4), were highly susceptible to lugdunin (Fig. 5), which underscores the potential capacity of lugdunin to exclude specific microbiome members from the human nasal microbiome.

Fig. 5
figure 5

Lugdunin susceptibility of typical nasal bacterial species. Different amounts of lugdunin (0 to 100 ng in 2 μl) were spotted on lawns of nasal bacteria, resulting in zones of inhibition after incubation for 24–48 h. Data points represent average values of three independent experiments, and vertical red lines show the medians of each group. The dashed line indicates the threshold of positive (above) and negative (below) correlations of nasal bacteria with S. lugdunensis

Full size image

S. lugdunensis was positively associated with a particularly high number of nasal bacterial species (Fig. 4) raising the question of how it can interact with so many different bacteria in a mutualistic manner. S. lugdunensis has been found to be strongly impaired in growth on iron-deprived media [31], and experimental data indicated that S. lugdunensis is probably unable to synthesize iron-sequestering siderophores and may therefore depend on the uptake of siderophores produced by other nasal bacteria [32, 33]. This possibility was assessed by monitoring the growth of S. lugdunensis on iron-deficient agar plates supplemented with culture filtrates from other nasal bacteria. Culture filtrates of C. tuberculostearicum, S. epidermidis, and S. capitis, which were positively associated with S. lugdunensis (Fig. 4), had indeed a strong growth-promoting impact on S. lugdunensis (Fig. 6A). However, the tested isolates of a given bacterial species differed profoundly in their impact on S. lugdunensis growth indicating that the strength of mutualistic interaction with S. lugdunensis is probably strain specific.

Fig. 6
figure 6

Nasal bacterial species producing siderophores have the capacity to support growth of S. lugdunensis. A Occurrence and sizes (diameter in mm) of growth zones of S. lugdunensis promoted by siderophore-containing spent media from nasal bacteria. Staphylococcal species, C. propinquum, and C. tuberculostearicum support S. lugdunensis growth under iron-restricted conditions. B Siderophore concentration in spent media from nasal bacteria after three days of incubation was detected via SideroTec AssayTM kit. High concentrations of siderophores could be detected in spent media from staphylococcal species and C. propinquum; siderophore detection was not assessable for C. accolens, C. aurimucosum, and C. tuberculostearicum due to essential Tween 80 in growth medium that interacts with the reaction solutions provided by the SideroTec AssayTM kit. Data points represent the average values of three independent biological replicates, and vertical red lines show the means of each species

Full size image

To assess if the growth-promoting bacterial microbiome members may release siderophores that S. lugdunensis can utilize, supernatants of cultures of the test bacteria grown under iron-restricted conditions were analyzed for the presence of siderophores by a colorimetric assay. Siderophore production was found, again, to be strongly strain-dependent, with some isolates of a given species showing no or very strong siderophore release (Fig. 6B). Strongly siderophore-producing isolates were found for C. propinquum and S. epidermidis but not for most of the nasal commensals that did not support growth of S. lugdunensis. C. tuberculostearicum could not be tested in this assay because essential components in its growth medium interfered with the colorimetric siderophore detection assay.

While S. lugdunensis lacks genes for siderophore synthesis, it encodes the siderophore uptake systems Hts and Sir allowing the acquisition of staphyloferrin A and B [32]. Additionally, it encodes the Fhu system recognizing hydroxamate-type and the Sst-system for the acquisition of catechol/catecholamine-type siderophores [34]. The fhu operon encoding also FhuC, the ATPase probably required for all of these uptake systems to function (except the SSt-system) [35] was deleted in S. lugdunensis HKU09-01 and IVK28 and the resulting mutants were compared to the wild type strains for their capacity to grow in the presence of siderophore-containing culture filtrates. The fhu mutants were unable to grow in the presence of culture filtrates from C. propinquum and S. epidermidis while the HKU09-01 mutant complemented with a plasmid-encoded copy of fhu grew equally well as the wild type (Fig. 7). A S. epidermidis mutant lacking the sfaDAB genes for staphyloferrin A biosynthesis was also included in the experiment. Culture supernatants of this mutant were not able to support the growth of any of the S. lugdunensis strains, which confirms the critical role of siderophore provision by other nasal bacteria for the capacity of S. lugdunensis to grow under conditions resembling the human nasal habitat.

Fig. 7
figure 7

S. lugdunensis growth promotion requires the fhu siderophore uptake locus and depends on siderophore production by S. epidermidis. S. lugdunensis HKU09-01 and IVK28 wild-type (WT), isogenic Δfhu mutants, and complemented strains were grown on iron-restricted agar. Growth promotion was observed only for the wild-type and the complemented strains when spent media were spotted/added that contained staphyloferrin A produced by S. epidermidis WT. Bars represent the mean of three independent experiments and black dots values of single experiments

Full size image

Discussion

The human nasal microbiome is less complex in composition than the intestinal microbiome [11]. Why the various human CSTs are dominated by specific bacterial species and why other species are found only in certain CSTs but absent from others has remained largely unclear. Nevertheless, a recent twin-cohort study has revealed that host genetics are of minor importance while the interaction among established microbiome residents appears to play key roles [11]. Our study demonstrates that even bacterial species that can be cultivated only from a minority of human nasal microbiomes are more prevalent than previously thought when analyses are based on metagenome data. Both, S. lugdunensis and S. aureus were identified in several nasal metagenomes from noses, which were culture-negative for these species. These data corroborate previous reports on the frequent presence of S. aureus DNA in nasal samples that did not yield positive S. aureus cultures [11, 36]. They also shed new light on the human S. aureus carrier status, which might be higher than the 30–40% previously estimated in culture-based studies. We found that the probability to cultivate S. aureus increased with the numbers of S. aureus reads in metagenomes indicating that a certain threshold number of S. aureus cells is necessary to achieve a positive culture. Bacterial numbers may change over time, which might explain the volatile S. aureus intermittent carrier state, characterized by a repeated change between culture-negative and culture-positive conditions with only low S. aureus counts per nare [37]. Very low numbers of S. aureus may not increase the risk of an at-risk patient developing an invasive S. aureus infection. However, even very low S. aureus numbers may be of high relevance if it is an MRSA strain that might spread to other patients on a hospital ward and may elicit MRSA outbreaks. Future pathogen monitoring approaches should therefore not only rely on bacterial cultures but also on the analysis of metagenome data to identify even low numbers of notorious pathogens and their antibiotic resistance genes, for which bioinformatic tools have already been developed [38]. A different explanation for the presence of S. aureus DNA in culture-negative samples is the non-viability of these cells, either by adoption of a viable but not cultivable (VBNC) state or the DNA originates from lately introduced cells that were not able to colonize but were killed. Many different bacteria can turn into VBNC cells under nutrient-poor environments or stressful conditions [39], and S. aureus was shown to adapt into a VBNC state under certain conditions [40]. Currently, we can only speculate about the origin of S. aureus and S. lugdunensis DNA in culture-negative samples, but clear discrepancies between culture and sequencing outcomes have been reported before [41].

Many bacterial species, including S. aureus and S. pneumoniae were negatively associated with S. lugdunensis and were also highly susceptible to lugdunin. However, several species, that were positively associated with S. lugdunensis, including several coynebacterial and CoNS species, were also found to be lugdunin-susceptible, indicating that further criteria beyond the mere susceptibility to lugdunin need to be considered to explain the antagonistic interactions of S. lugdunensis. These might include the proximity or distance of S. lugdunensis to other microbiome members in a specific nasal sub-niche. In addition, lugdunin is also able to induce human-derived antimicrobial peptides which synergistically act especially against S. aureus [42]. It should be noted that the bacterial strains tested for susceptibility to lugdunin were representative isolates obtained from available strain collections, which might differ from strains in the nasal microbiomes of the study participants in their properties.

Several studies have been published with a focus on bacterial competition in the human anterior nares [43,44,45], which found in part similar positive or negative associations as in our study. The results of our correlation analyses confirm previously described associations between Corynebacterium, Staphylococcus, Cutibacterium, Dolosigranulum, and Streptococcus species in the anterior nares [45]. We observed a negative association of C. accolens with S. pneumoniae (correlation coefficient -0.28), which corresponds to the reported antagonism of these two species [43]. Furthermore, we confirm the positive and negative associations of D. pigrum with other nasal microbiome species [44]. D. pigrum exhibited a strong positive association with C. propinquum (correlation coefficient +0.66) and a negative association with S. aureus (correlation coefficient −S0.41) (Supplementary Table S2).

With microbiome precision editing approaches, it could become possible to eliminate S. aureus from the noses of at-risk patients or at least to reduce its numbers strongly enough to keep infection risks to a minimum [8]. Finding a way to establish robust nasal S. lugdunensis colonization in at-risk patients could become a sustainable way of preventing S. aureus carriage and infections. We provide strong evidence that S. lugdunensis is auxotrophic for iron-scavenging siderophores and may need other microbiome members to release siderophores. Such compounds do not only support the producer but act as “public goods” for other bacteria with cognate siderophore uptake systems. Of note, the nutritional surrounding of the nasal cavity is iron-restricted and iron acquisition systems of S. aureus are expressed during the colonization of humans and cotton rats [46, 47]. Additionally, we show in a parallel submitted manuscript that siderophore acquisition of S. aureus is important during nasal colonization of cotton rats and that a plethora of siderophore-mediated interactions exist between nasal commensals and S. aureus (Zhao Y. et al.; mBio, in revision).

S. lugdunensis does not produce siderophores but encodes ABC importers for carboxylate, hydroxamate, and catechol/catecholamine siderophores [32, 34]. Interestingly, C. propinquum, which supported S. lugdunensis growth under iron-limiting conditions, has been reported to bear genes for biosynthesis of the siderophore dehydroxynocardamine [48]. The capacities of different isolates of S. epidermidis, S. hominis, S. capitis, and S. aureus to support S. lugdunensis growth was highly strain-dependent, which may point to substantial differences in the presence of such genes or their expression levels.

Several of the nasal bacterial species that correlated positively with S. lugdunensis did not support growth of S. lugdunensis under iron-restricted conditions. It is possible that the nasal strains differed in the production of siderophores from the test strains from our culture collection, used for in-vitro co-cultivation. On the other hand, some of the positive associations may result from higher-order interactions in larger bacterial networks with community members supporting other bacteria that were able to secrete siderophores.

Conclusion

We demonstrate that the nasal cavity of only a minority of the human population is colonized by S. lugdunensis, which eliminates S. aureus by production of antimicrobial lugdunin. S. lugdunensis cannot produce siderophores, and it prefers nasal microbiomes dominated by bacterial species that support iron acquisition by S. lugdunensis by production of suitable siderophores. Understanding and harnessing such complex mechanisms will allow the development of innovative strategies for pathogen exclusion from human microbiomes. Engineering S. lugdunensis to produce its own siderophores or co-administration of other commensals that produce siderophores for S. lugdunensis could become valuable strategies for the sustained prevention of S. aureus infections.

Data availability

The datasets generated and analyzed during the current study are deposited at NCBI under the BioProject number PRJNA1078731 and are available via the following link (https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA1078731). All additional data generated or analyzed during this study are included in this published article and its supplementary information files.

References

  1. Kost C, Patil KR, Friedman J, Garcia SL, Ralser M. Metabolic exchanges are ubiquitous in natural microbial communities. Nat Microbiol. 2023;8(12):2244–52. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41564-023-01511-x.

    Article  PubMed  CAS  Google Scholar 

  2. Heilbronner S, Krismer B, Brötz-Oesterhelt H, Peschel A. The microbiome-shaping roles of bacteriocins. Nat Rev Microbiol. 2021;19(11):726–39. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41579-021-00569-w.

    Article  PubMed  CAS  Google Scholar 

  3. Keith JW, Pamer EG. Enlisting commensal microbes to resist antibiotic-resistant pathogens. J Exp Med. 2019;216(1):10–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1084/jem.20180399.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  4. Tacconelli E, Sifakis F, Harbarth S, Schrijver R, van Mourik M, Voss A, et al. Surveillance for control of antimicrobial resistance. Lancet Infect Dis. 2018;18(3):e99–106. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/s1473-3099(17)30485-1.

    Article  PubMed  Google Scholar 

  5. Tacconelli E, Autenrieth IB, Peschel A. Fighting the enemy within. Science. 2017;355(6326):689–90. https://doiorg.publicaciones.saludcastillayleon.es/10.1126/science.aam6372.

    Article  PubMed  CAS  Google Scholar 

  6. Tsolis RM, Bäumler AJ. Gastrointestinal host-pathogen interaction in the age of microbiome research. Curr Opin Microbiol. 2020;53:78–89. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.mib.2020.03.002.

    Article  PubMed  CAS  Google Scholar 

  7. Zhang ZJ, Lehmann CJ, Cole CG, Pamer EG. Translating microbiome research from and to the clinic. Annu Rev Microbiol. 2022. https://doiorg.publicaciones.saludcastillayleon.es/10.1146/annurev-micro-041020-022206.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Krismer B, Weidenmaier C, Zipperer A, Peschel A. The commensal lifestyle of Staphylococcus aureus and its interactions with the nasal microbiota. Nat Rev Microbiol. 2017;15(11):675–87. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nrmicro.2017.104.

    Article  PubMed  CAS  Google Scholar 

  9. Lee AS, de Lencastre H, Garau J, Kluytmans J, Malhotra-Kumar S, Peschel A, Harbarth S. Methicillin-resistant Staphylococcus aureus. Nat Rev Dis Primers. 2018;4:18033. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nrdp.2018.33.

    Article  PubMed  Google Scholar 

  10. van Dalen R, Elsherbini AMA, Harms M, Alber S, Stemmler R, Peschel A. Secretory IgA impacts the microbiota density in the human nose. Microbiome. 2023;11(1):233. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40168-023-01675-y.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  11. Liu CM, Price LB, Hungate BA, Abraham AG, Larsen LA, Christensen K, et al. Staphylococcus aureus and the ecology of the nasal microbiome. Sci Adv. 2015;1(5):e1400216. https://doiorg.publicaciones.saludcastillayleon.es/10.1126/sciadv.1400216.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Janek D, Zipperer A, Kulik A, Krismer B, Peschel A. High frequency and diversity of antimicrobial activities produced by nasal staphylococcus strains against bacterial competitors. PLoS Pathog. 2016;12(8):e1005812. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.ppat.1005812.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  13. Torres Salazar BO, Heilbronner S, Peschel A, Krismer B. Secondary metabolites governing microbiome interaction of staphylococcal pathogens and commensals. Microb Physiol. 2021;31(3):198–216. https://doiorg.publicaciones.saludcastillayleon.es/10.1159/000517082.

    Article  PubMed  CAS  Google Scholar 

  14. Zipperer A, Konnerth MC, Laux C, Berscheid A, Janek D, Weidenmaier C, et al. Human commensals producing a novel antibiotic impair pathogen colonization. Nature. 2016;535(7613):511–6. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nature18634.

    Article  PubMed  CAS  Google Scholar 

  15. Ho PL, Leung SM, Tse H, Chow KH, Cheng VC, Que TL. Novel selective medium for isolation of Staphylococcus lugdunensis from wound specimens. J Clin Microbiol. 2014;52(7):2633–6. https://doiorg.publicaciones.saludcastillayleon.es/10.1128/JCM.00706-14.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Buchfink B, Xie C, Huson DH. Fast and sensitive protein alignment using DIAMOND. Nat Methods. 2015;12(1):59–60. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nmeth.3176.

    Article  PubMed  CAS  Google Scholar 

  17. Huson DH, Beier S, Flade I, Gorska A, El-Hadidi M, Mitra S, et al. MEGAN community edition - interactive exploration and analysis of large-scale microbiome sequencing data. PLoS Comput Biol. 2016;12(6):e1004957. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pcbi.1004957.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  18. Mitra S, Gilbert JA, Field D, Huson DH. Comparison of multiple metagenomes using phylogenetic networks based on ecological indices. ISME J. 2010;4(10):1236–42. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/ismej.2010.51.

    Article  PubMed  Google Scholar 

  19. Nurk S, Meleshko D, Korobeynikov A, Pevzner PA. metaSPAdes: a new versatile metagenomic assembler. Genome Res. 2017;27(5):824–34. https://doiorg.publicaciones.saludcastillayleon.es/10.1101/gr.213959.116.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  20. Nagpal S, Singh R, Yadav D, Mande SS. MetagenoNets: comprehensive inference and meta-insights for microbial correlation networks. Nucleic Acids Res. 2020;48(W1):W572–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/nar/gkaa254.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  21. Faust K, Sathirapongsasuti JF, Izard J, Segata N, Gevers D, Raes J, Huttenhower C. Microbial co-occurrence relationships in the human microbiome. PLoS Comput Biol. 2012;8(7):e1002606. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pcbi.1002606.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  22. Yadav D, Ghosh TS, Mande SS. Global investigation of composition and interaction networks in gut microbiomes of individuals belonging to diverse geographies and age-groups. Gut Pathog. 2016;8:17. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13099-016-0099-z.

    Article  PubMed  PubMed Central  Google Scholar 

  23. R Core Team. R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing; 2020. https://www.r-project.org/.

  24. Cimentada MKaSJaJ. corrr: correlations in R. R package version 0.4.2 edn. 2020.

  25. Wickham H. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York; 2016. ISBN 978-3-319-24277-4. https://ggplot2.tidyverse.org.

  26. Brugger SD, Eslami SM, Pettigrew MM, Escapa IF, Henke MT, Kong Y, Lemon KP. Dolosigranulum pigrum cooperation and competition in human nasal microbiota. bioRxiv. 2020:678698. https://doiorg.publicaciones.saludcastillayleon.es/10.1101/678698.

  27. Monk IR, Shah IM, Xu M, Tan MW, Foster TJ. Transforming the untransformable: application of direct transformation to manipulate genetically Staphylococcus aureus and Staphylococcus epidermidis. mBio. 2012;3(2):e00277-11. https://doiorg.publicaciones.saludcastillayleon.es/10.1128/mBio.00277-11.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  28. Huson DH, Scornavacca C. Dendroscope 3: an interactive tool for rooted phylogenetic trees and networks. Syst Biol. 2012;61(6):1061–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/sysbio/sys062.

    Article  PubMed  Google Scholar 

  29. Lemon KP, Klepac-Ceraj V, Schiffer HK, Brodie EL, Lynch SV, Kolter R. Comparative analyses of the bacterial microbiota of the human nostril and oropharynx. mBio. 2010;1(3):e00129-10. https://doiorg.publicaciones.saludcastillayleon.es/10.1128/mBio.00129-10.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  30. Wos-Oxley ML, Plumeier I, von Eiff C, Taudien S, Platzer M, Vilchez-Vargas R, et al. A poke into the diversity and associations within human anterior nare microbial communities. ISME J. 2010;4(7):839–51. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/ismej.2010.15.

    Article  PubMed  Google Scholar 

  31. Flannagan RS, Brozyna JR, Kumar B, Adolf LA, Power JJ, Heilbronner S, Heinrichs DE. In vivo growth of Staphylococcus lugdunensis is facilitated by the concerted function of heme and non-heme iron acquisition mechanisms. J Biol Chem. 2022;298(5):101823. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jbc.2022.101823.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  32. Brozyna JR, Sheldon JR, Heinrichs DE. Growth promotion of the opportunistic human pathogen, Staphylococcus lugdunensis, by heme, hemoglobin, and coculture with Staphylococcus aureus. Microbiologyopen. 2014;3(2):182–95. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/mbo3.162.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  33. Sheldon JR, Heinrichs DE. The iron-regulated staphylococcal lipoproteins. Front Cell Infect Microbiol. 2012;2:41. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fcimb.2012.00041.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  34. Lebeurre J, Dahyot S, Diene S, Paulay A, Aubourg M, Argemi X, et al. Comparative genome analysis of Staphylococcus lugdunensis shows clonal complex-dependent diversity of the putative virulence factor, ess/Type VII Locus. Front Microbiol. 2019;10:2479. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fmicb.2019.02479.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Sheldon JR, Heinrichs DE. Recent developments in understanding the iron acquisition strategies of gram positive pathogens. FEMS Microbiol Rev. 2015;39(4):592–630. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/femsre/fuv009.

    Article  PubMed  Google Scholar 

  36. Cole AL, Sundar M, Lopez A, Forsman A, Yooseph S, Cole AM. Identification of nasal gammaproteobacteria with potent activity against Staphylococcus aureus: novel insights into the “Noncarrier” State. mSphere. 2021;6(1):e01015-20. https://doiorg.publicaciones.saludcastillayleon.es/10.1128/mSphere.01015-20.

    Article  PubMed  PubMed Central  Google Scholar 

  37. van Belkum A, Verkaik NJ, de Vogel CP, Boelens HA, Verveer J, Nouwen JL, et al. Reclassification of Staphylococcus aureus nasal carriage types. J Infect Dis. 2009;199(12):1820–6. https://doiorg.publicaciones.saludcastillayleon.es/10.1086/599119.

    Article  PubMed  Google Scholar 

  38. Lanza VF, Baquero F, Martínez JL, Ramos-Ruíz R, González-Zorn B, Andremont A, et al. In-depth resistome analysis by targeted metagenomics. Microbiome. 2018;6:11. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40168-017-0387-y.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Oliver JD. The viable but nonculturable state in bacteria. J Microbiol. 2005;43:93–100.

    PubMed  Google Scholar 

  40. Gonçalves FDA, de Carvalho CCCR. Phenotypic modifications in cells exposed to high concentrations of vancomycin and teicoplanin. Front Microbiol. 2016;7:13. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fmicb.2076.00013.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Kaspar U, Kriegeskorte A, Schubert T, Peters G, Rudack C, Pieper DH, et al. The culturome of the human nose habitats reveals individual bacterial fingerprint patterns. Environ Microbiol. 2016;18(7):2130–42. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/1462-2920.12891.

    Article  PubMed  CAS  Google Scholar 

  42. Bitschar K, Sauer B, Focken J, Dehmer H, Moos S, Konnerth M, et al. Lugdunin amplifies innate immune responses in the skin in synergy with host- and microbiota-derived factors. Nat Commun. 2019;10(1):2730. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41467-019-10646-7.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  43. Bomar L, Brugger SD, Yost BH, Davies SS, Lemon KP. Corynebacterium accolens releases antipneumococcal free fatty acids from human nostril and skin surface triacylglycerols. mBio. 2016;7(1):e01725-15. https://doiorg.publicaciones.saludcastillayleon.es/10.1128/mBio.01725-15.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  44. Brugger SD, Eslami SM, Pettigrew MM, Escapa IF, Henke MT, Kong Y, Lemon KP. Dolosigranulum pigrum cooperation and competition in human nasal microbiota. mSphere. 2020;5(5):e00852-20. https://doiorg.publicaciones.saludcastillayleon.es/10.1128/mSphere.00852-20.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  45. Hardy BL, Merrell DS. Friend or foe: interbacterial competition in the nasal cavity. J Bacteriol. 2021;203(5):e00480-20. https://doiorg.publicaciones.saludcastillayleon.es/10.1128/JB.00480-20.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  46. Burian M, Wolz C, Goerke C. Regulatory adaptation of Staphylococcus aureus during nasal colonization of humans. PLoS One. 2010;5(4):e10040. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pone.0010040.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  47. Burian M, Rautenberg M, Kohler T, Fritz M, Krismer B, Unger C, et al. Temporal expression of adhesion factors and activity of global regulators during establishment of Staphylococcus aureus nasal colonization. J Infect Dis. 2010;201(9):1414–21. https://doiorg.publicaciones.saludcastillayleon.es/10.1086/651619.

    Article  PubMed  CAS  Google Scholar 

  48. Stubbendieck RM, May DS, Chevrette MG, Temkin MI, Wendt-Pienkowski E, Cagnazzo J, et al. Competition among nasal bacteria suggests a role for siderophore-mediated interactions in shaping the human nasal microbiota. Appl Environ Microbiol. 2019;85(10):e02406-18. https://doiorg.publicaciones.saludcastillayleon.es/10.1128/AEM.02406-18.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

Download references

Acknowledgements

The authors thank Vera Augsburger, Cosima Hirt, and Darya Belikova for excellent technical support and Stephanie Grond for synthetic lugdunin. We acknowledge support from the Open Access Publication Fund of the University of Tübingen.

Funding

Open Access funding enabled and organized by Projekt DEAL. This work was financed by grants from the German Research Foundation (DFG) GRK1708 (S.H., A.P.), TRR261 (S.H., A.P.), TRR156 (A.P.), and Cluster of Excellence EXC2124 Controlling Microbes to Fight Infection (CMFI) (S.H., A.P., B.K.); from the German Center of Infection Research (DZIF) to S.H., A.P., B.K..; and the European Innovative Medicines Initiate IMI (COMBACTE) (A.P.).

Author information

Author notes

  1. Ralf Rosenstein and Benjamin O. Torres Salazar contributed equally to this work.

Authors and Affiliations

  1. Cluster of Excellence EXC 2124 “Controlling Microbes to Fight Infections (CMFI)”, University of Tübingen, Tübingen, Germany

    Ralf Rosenstein, Benjamin O. Torres Salazar, Simon Heilbronner, Bernhard Krismer & Andreas Peschel

  2. Interfaculty Institute of Microbiology and Infection Medicine Tübingen, Infection Biology, University of Tübingen, Auf der Morgenstelle 28, Tübingen, 72076, Germany

    Ralf Rosenstein, Benjamin O. Torres Salazar, Claudia Sauer, Simon Heilbronner, Bernhard Krismer & Andreas Peschel

  3. German Centre for Infection Research (DZIF), Partner Site Tübingen, Tübingen, Germany

    Ralf Rosenstein, Benjamin O. Torres Salazar, Claudia Sauer, Simon Heilbronner, Bernhard Krismer & Andreas Peschel

  4. Ludwig-Maximilians-Universität München, Faculty of Biology, Microbiology, Munich, Germany

    Simon Heilbronner

  5. Interfaculty Institute of Microbiology and Infection Medicine, Institute for Medical Microbiology and Hygiene, UKT Tübingen, Tübingen, Germany

    Simon Heilbronner

  6. Present Address: Faculty of Biology, Microbiology, Ludwig Maximilian University of Munich, Munich, Germany

    Simon Heilbronner

Authors
  1. Ralf Rosenstein
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Benjamin O. Torres Salazar
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Claudia Sauer
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Simon Heilbronner
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Bernhard Krismer
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Andreas Peschel
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

R.R., B.O.T.S., and C.S. performed and evaluated the majority of experiments; R.R. assembled metagenomes and performed bioinformatic analysis. B.O.T.S. constructed the ΔfhuC and Δsfa strain for siderophore uptake experiments. C.S. collected and cultured nasal swabs and subsequently identified bacterial species. R.R., S.H., B.K., and A.P. designed the study; A.P. acquired funding and wrote the first draft of the manuscript. B.K., and A.P. finalized the manuscript. All authors read, reviewed, and approved the final manuscript.

Corresponding author

Correspondence to Bernhard Krismer.

Ethics declarations

Ethics approval and consent to participate

This study was approved by the Institutional Review Board for Human Subjects at the University of Tübingen (project number 577/2015A), and informed written consent was obtained from all study participants before sample collection.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

40168_2024_1913_MOESM1_ESM.pdf

Supplementary Material 1: Supplementary Table S1. Comparison of metagenome data with community state types (CST). Relative proportions (in percent) of selected taxa that had been described as signature for community state types (CST) as described in [11]. Most abundant taxa are highlighted in bold. Round 1 or 2 indicates the corresponding time of metagenome analysis after 18 months or 23 months, respectively. CST: classification of designated CST types according to the prevalence of indicator taxa in the microbiomes or none if not determinable (nd). CST6 for sample H-2 was based on a relative proportion of 54.4% Moraxella (not listed). For S. lugdunensis, the corresponding relative DNA abundances are also shown (in percent). The culture-based carrier states determined in parallel to the metagenome sequencing are indicated under “Culture”: Slu, S. lugdunensis carrier; Sau, S. aureus carrier; co, co-carrier; non, non-carrier. Supplementary Table S2. Correlation coefficients of species present in the nasal microbiomes. Correlation coefficients were inferred based on a centered log-ratio-transformed abundance table of species reads by correlation network analyses using the Namap/Pearson algorithm of MetagenoNets [20] with thresholds for prevalence and occurrence of 0.1% and 20%, respectively. The table shows the resulting correlation measures for 28 species throughout all 24 samples. Positive correlations are emphasized by green shades (with intensity corresponding to the hight of the values), while negative correlations are shaded in red. Supplementary Table S3. Bacterial strains used in this study. Supplementary Table S4. Plasmids and oligonucleotides used in this study.

40168_2024_1913_MOESM2_ESM.pdf

Supplementary Material 2: Supplementary Figure S1. Microbiome profiles within the genus Staphylococcus at the species level. Relative proportions of staphylococcal species present in the nasal microbiomes at time points 18 months and 23 months are presented by stacked bar charts. The different bacterial species are indicated by colours as shown in the legend on the right. The presence of S. lugdunensis reads in the metagenomes is indicated (Slu). The microbiome profiles that could be assigned to the CSTs published by Liu et al. [11] are labelled accordingly. Profiles not fitting to one of these types are designated “nd”. Supplementary Figure S2. Microbiome profiles within the genus Corynebacterium at the species level. Corynebacterium species profiles determined in metagenome analyses at time points 18 and 23 months. Relative proportions of Corynebacterium species present in the nasal microbiomes are presented by stacked bar charts. The different Corynebacterium species are indicated by colours as listed in the legend on the right. For C. accolens, C. pseudodiphteriticum, and C. tuberculostearicum, different strains (“KPL strains”) were determined but labeled with the same colours according to a strain assignment by the Lemon lab [45]. The presence of S. lugdunensis reads in the metagenomes is indicated (Slu). The microbiome profiles that could be assigned to the CSTs published by [11] are labelled accordingly. Profiles not fitting to one of these types are designated “nd”. Supplementary Figure S3. Correlation of S. aureus with other microbiome species. Correlation coefficients were inferred based on a centered log-ratio-transformed abundance table of species reads by correlation network analyses using the Namap/Pearson algorithm of MetagenoNets [20] with thresholds for prevalence and occurrence of 0.1% and 20%, respectively. Based on these parameters, we obtained correlation measures for 28 species throughout all 24 samples. Shown are the associations of S. aureus with other microbiome species (shown at the x-axis). Positive associations are indicated by green bars, negative associations by red bars. The corresponding Pearson correlation coefficients are indicated at the x-axis. Supplementary Figure S4. Nasal bacterial species producing siderophores have the capacity to support growth of S. lugdunensis. (A) Siderophore concentration in spent media from nasal bacteria after three days of incubation detected via SideroTec AssayTM kit. Siderophores could be detected in spent media from staphylococcal species and E. coli; siderophore detection was not assessable for C. kroppenstedtii due to the essential presence of Tween 80 in growth medium that interacts with the reaction solutions provided by the SideroTec AssayTM kit. (B) Occurrence and sizes (diameter in mm) of growth zones of S. lugdunensis promoted by siderophore-containing spent media from nasal bacteria. Staphylococcal species and E. coli support S. lugdunensis growth under iron-restricted conditions. Data points represent average values of three independent biological replicates, and vertical red lines show means of each species. Supplementary Figure S5. Representative pictures of lugdunin susceptibility determination of nasal bacteria and S. lugdunensis growth promotion by siderophore-containing spent media from nasal bacteria. (A) Lugdunin susceptibility determination by the example of S. aureus USA300. 2 μl containing different amounts of lugdunin (0 to 100 ng) were spotted onto lawns of bacteria on sheep blood agar plates, resulting in zones of inhibition for lugdunin-susceptible bacteria. The spot with the lowest amount of lugdunin showing a clear inhibition was used to determine lugdunin susceptibility. (B) Growth promotion of S. lugdunensis HKU09-01 by siderophore-containing spent media of nasal bacteria. Iron-restricted BHI agar supplemented with 10% horse serum (as transferrin source) were inoculated with a lawn of S. lugdunensis. Spent media of different bacteria were spotted onto S. lugdunensis lawns resulting in zones of enhanced growth when siderophores were present. Diameters were measured and noted. Supplementary Figure S6. Lugdunin susceptibility of nasal bacterial species. Different amounts of lugdunin (0 to 100 ng in 2 μl) were spotted onto lawns of nasal bacteria, resulting in zones of inhibition. Data points represent average values of three independent experiments, and vertical red lines show medians of each group.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Rosenstein, R., Torres Salazar, B.O., Sauer, C. et al. The Staphylococcus aureus-antagonizing human nasal commensal Staphylococcus lugdunensis depends on siderophore piracy. Microbiome 12, 213 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40168-024-01913-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40168-024-01913-x

Keywords

Microbiome

ISSN: 2049-2618

Contact us