Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Jan 1.
Published in final edited form as: Pediatr Infect Dis J. 2013 Jan;32(1):72–77. doi: 10.1097/INF.0b013e318270d850

Density Interactions between Streptococcus pneumoniae, Haemophilus influenzae and Staphylococcus aureus in the Nasopharynx of Young Peruvian Children

Yu-Wen Chien 1, Jorge E Vidal 2, Carlos G Grijalva 3, Catherine Bozio 1, Kathryn M Edwards 4, John V Williams 4,5, Marie R Griffin 3, Hector Verastegui 6, Stella M Hartinger 7, Ana I Gil 6, Claudio F Lanata 6, Keith P Klugman 2,#
PMCID: PMC3525793  NIHMSID: NIHMS410385  PMID: 22935873

Abstract

Streptococcus pneumoniae, Haemophilus influenzae and Staphylococcus aureus are commonly carried in the nasopharynx (NP) of young children, and have been speculated to interact with each other. Although earlier studies used cultures alone to assess these interactions, the addition of real-time quantitative polymerase chain reaction (qPCR) provides further insight into these interactions. We compared results of culture and qPCR for the detection of these three bacteria in 446 NP samples collected from 360 healthy young children in a prospective cohort study in the Peruvian Andes. Patterns of concurrent bacterial colonization were studied using repeated measures logistic regression models with generalized estimating equations. Spearman correlation coefficients were employed to assess correlations among bacterial densities. At a bacterial density <105 colony forming units (CFU)/ml measured by qPCR, culture detected significantly less carriers (P<0.0001) for all three pathogens, than at a bacterial density >105 CFU/ml. In addition, there was a positive association between S. pneumoniae and H. influenzae colonization measured by both culture (OR 3.11 – 3.17, p < 0.001) and qPCR (OR 1.95 – 1.97, p < 0.01). The densities of S. pneumoniae and H. influenzae, measured by qPCR, were positively correlated (correlation coefficient 0.32, p < 0.001). A negative association was found between the presence of S. pneumoniae and S. aureus in carriage with both culture (OR 0.45, p = 0.024) and qPCR (OR 0.61, p < 0.05). The impact of density on detection by culture and the observed density-related interactions support use of qPCR in additional studies to examine vaccine effects on diverse bacterial species.

Keywords: Streptococcus pneumoniae, nasopharyngeal carriage, interaction

INTRODUCTION

Several common respiratory bacterial pathogens, including Streptococcus pneumoniae, Haemophilus influenzae and Staphylococcus aureus asymptomatically reside in the human nasopharynx. However, they can occasionally invade adjacent sites or the blood stream and cause disease such as otitis media, pneumonia, bacteremia and meningitis. The pneumococcus is commonly associated with these illnesses, (1) while H. influenzae is a common cause of acute otitis media (2) and S. aureus is a re-emerging cause of clinically important infections)(3), ranging from mild skin infections and sinusitis to severe diseases such as pneumonia, bacteremia, and endocarditis.

These different bacterial species may interact with each other by competing for resources and by producing chemicals or by inducing host immune responses that influence the growth of other bacteria in the nasopharynx.(4) Several observational studies using culture for bacterial detection have shown that S. aureus prevalence is negatively associated with S. pneumoniae prevalence,(5, 6) especially with vaccine-type pneumococci,(7-9) while some other studies found no association between them.(10, 11) Earlier studies have shown NP colonization by S. pneumoniae and H. influenzae to be positively associated.(5, 10-12)

The reported prevalence of nasopharyngeal (NP) colonization with these bacteria varies widely, depending on study population, age, exposure to antibiotics, socioeconomic status and variations in sampling and detection methods, and exposure to vaccines.(13) Previous studies using conventional culture have shown NP colonization of children ranged from 13 – 85% and 6 – 80% for S. pneumoniae and H. influenzae, respectively,(2, 8) and 10 – 35% for S. aureus .(7, 14, 15) Previous interaction studies used culture alone to detect and quantify bacteria. However, cultures have several drawbacks, such as low sensitivity compared with molecular methods, and it is time-consuming and laborious to perform quantification analysis. Real-time quantitative polymerase chain reaction (qPCR) is potentially a more sensitive and rapid alternative to culture.(13) Thus, the addition of qPCR to interaction assessments could provide valuable new insights.

Understanding factors that influence NP colonization by these bacteria is essential since colonization is the initial step for the development of disease. In addition, NP colonization of these pathogens in children is an important source of horizontal transmission to other individuals in the community.(16) Whether bacteria colonize or not is determined by a complex combination of factors including host characteristics that influence the exposure or susceptibility to specific bacterial species and direct interactions between different bacteria. Host factors that have been suggested to influence the colonization prevalence of S. pneumoniae, H. influenzae, or S. aureus include age, gender, ethnicity, immunity, crowding, number of siblings, daycare attendance, season, antibiotic therapy, acute respiratory infections, vaccine exposure, and environmental exposure to tobacco smoke.(2, 8, 12, 17, 18)

The goals of our study were to (1) compare culture and qPCR for detection of S. pneumoniae, H. influenzae, or S. aureus in NP samples; (2) describe the prevalence of NP colonization by S. pneumoniae, H. influenzae, and S. aureus in young children in rural communities of the Peruvian Andes; and (3) investigate the interaction between S. pneumoniae, H. influenzae, or S. aureus in two ways: first, evaluate whether colonization status (presence/absence) of one bacterium influences the colonization status of the other two bacteria, and second, evaluate whether the densities of the three bacteria are correlated.

MATERIALS AND METHODS

Study population and data collection

The population for this study was derived from a prospective cohort study of ~500 children 0-35 months of age in the District of San Marcos, Cajamarca, Peru. The study aimed to investigate whether indoor air pollution and acute respiratory infections (ARI) influenced nasopharyngeal colonization with S. pneumoniae in healthy rural children living between 2000 and 4000 meters above sea level. Children 0-35 months of age were enrolled and baseline demographic and socioeconomic information collected; the presence of ARI or pneumonia symptoms was collected at baseline and at weekly visits to the homes by trained field workers. Routine nasopharyngeal swabs for bacterial colonization were collected monthly on all children using Rayon swabs immediately placed in 1 ml of transport medium (skim-milk tryptone glucose glycerol, STGG), processed at a local laboratory according to WHO standards and stored at −70°C. Enrollment into this study started in May 2009. Here we present data for 446 consecutive NP samples collected among children 360 without ARI symptoms between August and September 2009.

Bacteriologic cultures

To increase the sensitivity of cultures, 200 μl of STGG sample were enriched in THY broth (Todd-Hewitt broth supplemented with 0.5% of yeast extract) containing 1% of rabbit serum and incubated for 6 h at 37°C with 5% CO2.(13) To identify S. pneumoniae strains, the enrichment broth was inoculated onto blood agar plates (BHI agar containing 5% sheep blood) and incubated overnight at 37°C in a 5% CO2 atmosphere.(13) Presumptive pneumococcus-like colonies were confirmed by the optochin susceptibility test and bile solubility. Equivocal results were confirmed by PCR using primers that target the cpsA gene(19) and DNA extracted from the S. pneumoniae isolates using the Chelex method (see below).(20)

For detection of S. aureus, the enriched THY broth was inoculated onto mannitol salt agar and incubated at 37°C in 5% CO2 for 24 hours; colonies morphologically suggestive of S. aureus were confirmed by performing PCR on Chelex-extracted DNA using published primers targeting the nuc gene.(21)

For detection of all serotypes and non-typable H. influenzae, 200 μl of STGG sample was enriched in brain heart infusion broth with 5% Fildes enrichment (BD Diagnostics, NJ, USA) for 6 h at 37°C in 5% CO2; the enriched broth was then plated onto chocolate agar with bacitracin and incubated overnight. The presence of H. influenzae was confirmed by performing PCR using the primers targeting the hpd gene which detects all H. influenzae strains(22) on DNA extracted from suspected colonies using the extraction method previously described by LaClaire et al.(23)

DNA extraction from cultured isolates using the Chelex method for PCR

A loopful of bacteria from the culture plate was placed into a 1.5 ml micro-centrifuge tube and mixed with 200 μl of 5% Chelex-100 resin (Bio-Rad) and 2 μl of Proteinase K (20mg/ml, QIAgen).(20) After incubation at 56°C for an hour then at 95°C for 10 minutes, the sample was mixed and then centrifuged at 13,000 rpm for 5 minutes to completely separate the layers. The DNA-containing supernatant was used as template in PCR reactions.

DNA extraction for qPCR from the original nasopharyngeal specimen and reference strains

Two-hundred μl of STGG sample were added with 100 μl of TE (10 mM Tris-HCl, 1 mM EDTA, pH 8.0) buffer containing 0.04 g/ml lysozyme and 75 U/ml of mutanolysin and then incubated for 1 h at 37°C in a water bath. Using the QIAamp DNA Mini protocol, DNA was eluted in 100 μl of elution buffer and kept at −70°C. Genomic DNA from the reference strains of S. pneumoniae (ATCC 33400 or TIGR4),(24) S. aureus (ATCC 25923) and H. influenzae type b (CDC reference strain M5216) was also extracted from overnight cultures using the QIAamp kit. DNA concentrations were measured by Nanodrop method (Nanodrop Technologies, Wilmington, DE) and serial dilutions in DNase-, RNase-free water were made to obtain the qPCR standards.

Real-time quantitative PCR (qPCR)

The total density of S. pneumoniae, measured in CFU/ml, was determined using pre-optimized concentrations of the forward primer (5′-ACGCAATCTAGCAGATGAAGCA-3′; 200 nM), reverse primer (5′-TCGTGCG TTTTAATTCCAGCT-3′; 200 nM), and probe (5′-FAM-TGCCGAAAACGCTTGATACAG GGAG-3-BHQ1; 200 nM) targeting the lytA gene as described previously.(24) To quantify the molecular bacterial load (CFU/ml) purified genomic DNA from S. pneumoniae reference strain TIGR4 (from our laboratory collection) was serially diluted to prepare standards representing 4, 4×101, 4×102, 4×103, 4×104, 4×105 or 4×106 CFU as previously described.(25) These standards were run along with our samples in a CFX96 real time PCR-detection system (Bio-Rad, Hercules CA) and final S. pneumoniae CFU/ml were calculated using the software Bio-Rad CFX manager.

The total density of S. aureus, measured in CFU/ml, was determined using the forward primer (5′-GTTGCTTAGTGTTAACTTTAGTTGTA-3′; 800 nM), reverse primer (5′-AATGTCGCAGGTTCTTTATGTAATTT-3′; 800 nM), and probe (5-FAM-AAGTCTAAGTAGCTCAGCAAATGCA-3-BHQ1; 400 nM) targeting the nuc gene.(21) To create standard curves, purified genomic DNA of the whole genome sequenced S. aureus reference strain N315 was used, assuming a genome size of 2.8 Mb. (26)

The total density of H. influenzae, measured in CFU/ml, was determined using recently published primers and probe: forward primer (5′-GGTTAAATATGCCGATGGTGTTG-3′; 100 nM), reverse primer (5′-TGCATCTTTACGCACGGTGTA-3′; 300 nM), and probe (5′-HEX-TTGTGTACACTCCGT “T-BHQ1” GGTAAAAGAACTTGCAC-3′; 100 nM) targeting the hpd gene.(22) To create standard curves, the purified genomic DNA of H. influenzae reference strain M5029 (a gift from Leonard Mayer, CDC, Atlanta GA, USA), was used, assuming a genome size of 1.8 Mb.

Real-time quantitative PCR was performed using the Bio-Rad CFX96TM Real-Time PCR Detection System (Hercules, CA, USA) in a reaction volume of 25 μl containing the EXPRESS qPCR Supermix Universal (Invitrogen by Life Technology, CA, USA), 2.5 μl of sample DNA, forward and reverse primers and fluorogenic probe with concentrations described above. For S. pneumoniae, the qPCR conditions were 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min; samples with cycle threshold (Ct) values ≤ 35 were considered positives. For H. influenzae, the cycling conditions included 50°C for 2 min, 95°C for 10 min, and 45 cycles of 15 s at 95°C followed by 1 minute at 60°C; and samples with Ct values ≤ 35 were considered positives. For S. aureus, the conditions were 50°C for 2 min, 95°C for 2 min, and 40 cycles of 95°C for 15 s and 60°C for 1 min. Positive samples were samples with Ct values ≤ 38.

Statistical analysis

All statistical analyses were performed using SAS version 9.1 (SAS institute, Inc, Cary, NC, USA). To assess whether colonization by one bacterial species was associated with colonization by the other two bacterial species, repeated measures logistic regression models with generalized estimating equations (GEE) were used because some children contributed more than one swab to the analyses. We modeled colonization by S. pneumoniae, H. influenzae and S. aureus separately, and each model included variables indicating the presence/absence of the other two bacterial species as the main exposures of interest. Covariates to be controlled for potential confounding included age in months, gender and antibiotic usage within the 7 days prior to sample collection. Separate analyses used bacterial culture and qPCR to define the colonization status of each bacterium.

To assess the degree of correlation between densities of each pair of bacteria in all samples, we used Spearman correlation coefficients. This test was preferred because bacterial densities determined by qPCR were not normally distributed because a large proportion was qPCR negative, and were assigned a density of 0. We then focused on those samples that were positive for two bacteria and re-calculated Spearman correlation coefficients. Positive samples were also categorized into high density and low density using median as a cutoff. Chi-square tests were performed to examine whether there was an association among bacterial densities (classified as high or low) of the three bacteria.

Ethical Approvals

This study was approved by the Ethical Review Boards (ERB) of the Instituto de Investigación Nutricional, Vanderbilt University and Emory University. An ERB-approved written informed consent form was obtained from one parent (usually the mother) of participating subjects at enrollment. The study was also approved by the local health authorities, and by community leaders.

RESULTS

Study population

A total of 446 consecutive nasopharyngeal samples from 360 children aged 0 – 35 months (mean 18.5 months, median 18.3 months), 53% male were included in this analysis. Among the enrolled children, 309 children had vaccination information available, and the percentages of children who received at least one dose of pneumococcal conjugate vaccines and H. influenzae type b vaccines before the swabs were collected were 9.1% and 67.3%, respectively. Of the 446 swabs, only 10 (2.2%) were collected from children who had received antibiotics within the 7 days prior to sample collection.

Comparison of culture and qPCR

All 446 samples were tested by both culture and qPCR for S. pneumoniae and S. aureus. For H. influenzae, 446 samples were tested using qPCR while only 351 samples were cultured because of inadequate remaining sample volume. A comparison of bacterial culture with qPCR showed that qPCR detected more pneumococcal carriage (in 77.4% of children compared to 60.1% of children by culture); and similarly, qPCR detected 37.3% carriage of H. influenzae compared to 23.6% by culture. For Staphylococcus aureus carriage qPCR detected carriage in 40.8% of children while only 11.9% were detected by culture (Table 1). For all three bacteria, all culture-positive swabs were also positive by qPCR and all qPCR-negative swabs were also negative by culture. As the bacterial density measured by qPCR increased, the detection by culture increased (Table 2). At a carriage density of <105 colony forming units (CFU)/ml as measured by qPCR, culture was significantly less sensitive than PCR at detecting pneumococcal carriage. Carriage was detected by culture in 196 of 273 (71.8%) of children carrying at low density, compared to detection by culture in 72/72 (100%) at a density >105 CFU/ml density (P<0.00001). Lesser sensitivity of culture at a density measured by PCR of <105 CFU/ml compared to a CFU >105 was also true for Haemophilus influenzae (detection of 1/19 (5.2%) versus 82/112 (73.2%); P< 0.0001) and staphylococcal carriage (detection of 13/132 (9.8%) versus 40/50 (80%); P<0.0001) (Table 2). The proportion of 351 samples indicating colonization by single or multiple species, by culture and qPCR, are shown in Table 3. Results of cultures showed that none, 1, or 2 bacteria species were concurrently present in 28.2%, 49.3% and 22.5% of NP swabs respectively, while qPCR detected none, 1, or 2 bacteria species in 8.0%, 40.2% and 51.8% of NP swabs, respectively.

Table 1.

Comparison of bacterial culture and qPCR for detection of Streptococcus pneumoniae, Staphylococcus aureus, and Haemophilus influenzae in nasopharyngeal swabs.

No. (%)
S. pneumoniae H. influenzae S. aureus
Culture
(−)		 Culture
(+)		 Total Culture
(−)		 Culture
(+)		 Total Culture
(−)		 Culture
(+)		 Total
qPCR(−) 101
(22.6%)		 0
(0.0%)		 101
(22.6%)		 220
(62.7%)		 0
(0.0%)		 220
(62.7%)		 264
(59.2%)		 0
(0.0%)		 264
(59.2%)
qPCR(+) 77
(17.3%)		 268
(60.1%)		 345
(77.4%)		 48
(13.7%)		 83
(23.6%)		 131
(37.3%)		 129
(28.9%)		 53
(11.9%)		 182
(40.8%)
Total 178
(39.9%)		 268
(60.1%)		 446
(100.0%)		 268
(76.4%)		 83
(23.6%)		 351
(100.0%)		 393
(88.1%)		 53
(11.9%)		 446
(100.0%)
Open in a new tab

Table 2.

Bacterial detection by culture, stratified by bacterial density (qPCR)

No. culture positive / No. in density category by qPCR (%)
Bacterial density
by qPCR (
CFU/ml)		 S. pneumoniae H. influenzae S. aureus
0
>0 - 104
>104 - 105
>105 -106
>106 		0/101 (0.0%)
65/131(49.6%)
131/142 (92.3%)
66/66 (100%)
6/6 (100%)		 0/220 (0.0%)
0/2 (0.0%)
1/17 (5.9%)
13/33(39.4%)
69/79 (87.3%)		 0/264 (0.0%)
1/43 (2.3%)
12/89 (13.5%)
18/25 (72.0%)
22/25 (88.0%)
Total 268/446 (60.1%) 83/351 (23.6%) 53/446 (11.9%)
Open in a new tab

Table 3.

identification of Streptococcus pneumoniae, Staphylococcus aureus, and Haemophilus influenzae in 351 nasopharyngeal swabs tested for all three bacteria using culture and qPCR

Colonizing bacteria Culture results
(n=351)		 qPCR results
(n=351)
none 99 (28.2%) 28 (8.0%)
S. pneumoniae alone 135 (38.5%) 97 (27.6%)
S. aureus alone 22 (6.3%) 33 (9.4%)
H. influenzae alone 16 (4.6%) 11 (3.1%)
One species total 173 (49.3%) 141(40.2%)
S. pneumoniae + S. aureus 17 (4.8%) 99 (28.2%)
S. pneumoniae + H. influenzae 61 (17.4%) 72 (20.5%)
S. aureus + H. influenzae 1 (0.3%) 11 (3.1%)
Two species total 79 (22.5%) 182 (51.8%)
S. pneumoniae + S. aureus + H. influenzae 0 (0%) 0 (0%)
Any S. pneumoniae 213 (60.7%) 268 (85%)
Any H. influenzae 78 (24.8%) 94 (29.8%)
Any S. aureus 40 (12.7%) 143 (45.4%)
Open in a new tab

Prevalence and incidence of colonization by S. pneumoniae, H. influenzae, and S. aureus as determined by qPCR

Among the first swabs collected from the 360 children, the prevalence of colonization by S. pneumoniae, H. influenzae, and S. aureus was 76.9%, 36.4% and 40.3%. For 86 children who contributed two swabs taken one month apart, 47.0% of 17 children who were originally not colonized by S. pneumoniae were colonized after one month, and 13.0% of 69 children who were originally colonized by S. pneumoniae cleared the bacteria. The percentages of acquisition and clearance after one month were 28.8% and 32.4% for H. influenzae, and 39.5% and 54.2% for S. aureus, respectively. All specimens were collected for this analysis between Aug and Sep 2009 to exclude any potential seasonality effect. Over the limited age range of the study (0 – 36 months) we did not detect a significant negative impact of increasing age on the density of these three pathogens. Overall density of pneumococcal carriage was not significantly different among the 9.1% of children previously in receipt of at least a single dose of PCV, nor was Haemophilus density influenced by previous receipt of Hib vaccine.

Assessing the associations between the colonization status of S. pneumoniae, H. influenzae, and S. aureus Bacterial cultures

Repeated measures logistic regression models of colonization (presence or absence) by S. pneumoniae and H. influenzae determined by cultures are shown in Table 4. The model of colonization by S. pneumoniae indicated that colonization by H. influenzae was positively associated with S. pneumoniae (OR 3.11, 95% CI 1.73 – 5.59, p < 0.001), while the presence of S. aureus was negatively associated with colonization by S. pneumoniae (OR 0.45, 95% CI 0.22 – 0.90, p = 0.024), controlling for age, gender, and antibiotic usage. A positive association between S. pneumoniae and H. influenzae was also observed in the model of colonization by H. influenzae (OR 3.17, 95% CI 1.76 – 5.72, p < 0.001).

Table 4.

Association between colonization with Streptococcus pneumoniae or Haemophilus influenzae (based on culture data) and selected variables

OR (95% CI)1 for colonization with
selected bacteria:
S. pneumoniae
(n=351)		 H. influenzae
(n=351)
S. pneumoniae
 Absent (ref)
 Present		 - 1
3.17 (1.76 – 5.72)
H. influenzae
 Absent (ref)
 Present		 1
3.11 (1.73 – 5.59) 		–
S. aureus
 Absent (ref)
 Present		 1
0.45 (0.22 – 0.90) 		1
0.71 (0.29 – 1.74)
Age (1-mo
 increase)		 1.00 (0.98 – 1.03) 1.02 (0.99 – 1.05)
Sex
 Female (ref)
 Male		 1
1.42 (0.90 – 2.25)		 1
0.61 (0.36 – 1.04)
Antibiotics
 No (ref)
 Yes		 1
0.97 (0.22 – 4.29)		 1
2.50 (0.53 – 11.9)
Open in a new tab
1

OR, odds ratio; CI, confidence interval. Significant ORs and 95% CI are shown in boldface. Model for colonization of each bacterium included variables representing presence/absence of the other two bacteria and potential confounders: age, sex and antibiotic usage within the past 7 days (Antibiotics).

Quantitative PCR

Repeated measures logistic regression models of colonization by S. pneumoniae, H. influenzae, and S. aureus determined by qPCR indicated a positive association between S. pneumoniae and H. influenzae (OR 1.97, 95% CI 1.18 – 3.29, p = 0.009) and a significant negative association with S. aureus (OR 0.61, 95% CI 0. 39 – 0.97, p = 0.031). The model of H. influenzae colonization also showed a positive association between H. influenzae and S. pneumoniae (OR 1.95, 95% CI 1.19 – 3.21, p = 0.009). The model of S. aureus colonization also showed that S. pneumoniae was negatively associated with S. aureus (OR 0.61, 95% CI 0.40 – 0.96, p = 0.039).

Assessing the correlation of bacterial densities of S. pneumoniae, H. influenzae and S. aureus

The Spearman correlation coefficients for the density between S. pneumoniae and H. influenzae, was 0.32 (p < 0.001). When only positive swabs (density > 0) were considered, the Spearman correlation coefficient was 0.38 (p < 0.001). When the density of the positive swabs was categorized into high and low for each bacterium, swabs with high density of S. pneumoniae were associated with high density of H. influenzae (OR 3.67, p<0.001). No association was found between S. pneumoniae and S. aureus or between H. influenzae and S. aureus densities.

DISCUSSION

In this study, we describe the colonization of young children in the rural Andes using molecular methods to complement information from traditional cultures. Our data indicate that molecular approaches increase the yield of detection of bacterial colonization by 17% to 29% for the three bacteria studied, suggesting that these molecular assays may become the assays of choice to detect bacterial colonization in the nasopharynx.

When comparing the results of culture and qPCR for detection of S. pneumoniae, H. influenzae and S. aureus for nasopharyngeal samples stored in the WHO standard medium for detection of pneumococci in the nasopharynx (STGG),(27) culture had low sensitivity when bacterial density was low, and its performance may be suboptimal for detecting low density carriage. The much higher detection rate using qPCR than using culture suggests that previous studies using culture alone underestimated the prevalence of bacterial colonization in the nasopharynx. However, qPCR could detect both viable and nonviable bacterial cells, which could lead to an overestimation of bacterial colonization prevalence. The use of molecular techniques with high sensitivity to detect pneumococcal colonization is important for evaluation and formulation of pneumococcal vaccines.(13)

To our knowledge, this is the first study to use qPCR to examine the prevalence of H. influenzae and S. aureus colonization in healthy Andean children. The clinical and epidemiologic relevance of using qPCR for detection of H. influenzae and S. aureus in nasopharyngeal samples is still to be determined. Several studies suggest that qPCR may provide a rapid quantification that can be used to distinguish between infection and colonization because bacterial density may be higher during infection than during colonization.(28-31) We found that age in months, gender and antibiotic usage were not associated with colonization by any of the three bacterial species we examined using either cultures or qPCR results. Our data suggest that if age related reductions in bacterial density occur, then they occur after three years of age which was the limit of age of children in this study. However, there were strong relationships among these three species. The negative association between S. pneumoniae and S. aureus colonization observed in this study is consistent with previous reports in Europe, Asia, America and Africa.(5-9) One possible mechanistic explanation for this bacterial interference phenomenon is that hydrogen peroxide produced by S. pneumoniae(32) may “protect” the host against S. aureus colonization. A recent randomized controlled trial showed that children who received heptavalent pneumococcal conjugative vaccine in The Netherlands were twice as likely to be colonized with S. aureus around twelve months of age compared to unvaccinated controls, but this effect might be transient. Whether this transient increase in S. aureus colonization influences the health of vaccinated children is currently unknown.(33)

Similar to several other epidemiologic studies,(5, 10-12) we found a positive association between colonization of S. pneumoniae and H. influenzae, and in addition, we found that their densities in the nasopharynx were positively correlated. This is consistent with a recent study showing that the presence of H. influenzae increases pneumococcal biofilm formation in vivo and in vitro.(34) However, in vitro experiments suggest that hydrogen peroxide and neuraminidase produced by S. pneumoniae can inhibit the growth of H. influenzae.(35, 36) Studies in mice suggest that the immune response primarily elicited by H. influenzae reduces the density of some pneumococcal strains,(4, 37) while the presence of S. pneumoniae facilitates the colonization by a new H. influenzae population.(4) Another epidemiologic study showed that these two bacteria were negatively associated in children with ARI, but the association shifted from negative to positive in the presence of M. catarrhalis.(38) Therefore, the mechanism by which S. pneumoniae and H. influenzae influence each other in the nasopharynx is complex and may be affected by which one initially colonized the nasopharynx, the host immune response, the presence of ARI symptoms, and other bacterial species present in the nasopharynx.

The observed association between bacteria in our study as well as in other epidemiologic studies may be a result of direct bacterial interaction or due to unmeasured host-specific confounders that simultaneously influence colonization status of different bacteria. Jacoby et al. recently used a hierarchical multivariate logistic model to analyze longitudinal data and to simultaneously model colonization status of different pathogens, trying to differentiate the host-level interaction and microbe-level interaction.(10) Interestingly, they found a positive association between S. pneumoniae and H. influenzae at host-level only among aboriginal children, not among non-aboriginal children, and they found neither host-level nor microbe-level associations between S. pneumoniae and S. aureus, which is contrary to many previous studies. We could not use their approach to analyze our data because only a small subset of children contributed more than one swab. Although we have controlled some host-level potential confounders, we cannot rule out that other factors may have influenced the observed bacterial associations.

Our study has several other limitations. Other pathogens potentially involved in interactions with the three bacteria we examined, such as M. catarrhalis (38) and coagulase-negative staphylococci,(18) were not studied. In addition, information on other potential confounders, such as number of siblings, family size and daycare attendance was only available on 65.7% of this sample. However, the association between the three bacterial species did not seem to be confounded by these factors in analyses confined to the subset of children from whom complete information was available (not shown). Bacterial density data on this population living at high altitude may differ from those living at lower altitudes but there are no data to date on the association of altitude with nasopharyngeal bacterial density. There are as yet limited data on serotype specific measures of nasopharyngeal density in children and these may add to the complexity of the density relationships in the nasopharynx.

The relationship between bacterial density in the nasopharynx and disease is attracting increasing attention. Increased nasopharyngeal density of respiratory bacteria has been linked to acute otitis media(39) in children as well as to pneumonia (30), but the relatively low specificity of the association has not allowed a critical level of density to be used diagnostically to date in children. Serotype specific density may be valuable as a diagnostic in the future. Nasopharyngeal pneumococcal density has however recently been shown to useful in the diagnosis of pneumococcal pneumonia in adults with a four log difference in nasopharyngeal density demonstrated between asymptomatic carriage and disease(40). The detection of low density carriage by molecular methods may thus be more important for understanding persistence of pneumococcal serotypes and transmission of the pneumococcus and other respiratory bacteria, than in making an association with disease in an individual patient.

Understanding how bacteria interact with each other in the nasopharynx, whether synergistically or competitively, is essential for designing preventive measures. This is especially true in this era of vaccines and antimicrobials which target specific bacteria and may unexpectedly influence the bacterial flora. Our study confirmed the previously observed negative association between S. pneumoniae and S. aureus and positive association between S. pneumoniae and H. influenzae using culture and qPCR to test nasopharyngeal samples collected from young children in rural communities of the Peruvian Andes. Our study demonstrates for the first time the impact of bacterial density on the detection of these bacteria by culture and the relationship between the density of pneumococcal and Haemophilus colonization. As colonization density may be an essential precursor to disease caused by these bacteria and may also be important for transmission, our study suggests that future colonization studies of these pathogens should also measure bacterial density.

Acknowledgments

Support and COI: This study was supported in part by Vanderbilt University, the Vanderbilt CTSA grant UL1 RR024975-01 from NCRR/NIH and Pfizer (IIR WS1898786(0887X1-4492)). In addition to an institutional grant from Pfizer, KPK has received consulted for Pfizer vaccines, GSK, Novartis, sanofi, and Merck, and he has received speaking fees from Pfizer and GSK. MRG has consulted for Novavax; CGG has consulted for GSK; JVW has consulted for Quidel and Medimmune.

Footnotes

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Lynch JP, 3rd, Zhanel GG. Streptococcus pneumoniae: epidemiology, risk factors, and strategies for prevention. Semin Respir Crit Care Med. 2009;30:189–209. doi: 10.1055/s-0029-1202938. [DOI] [PubMed] [Google Scholar]
  • 2.Garcia-Rodriguez JA, Fresnadillo Martinez MJ. Dynamics of nasopharyngeal colonization by potential respiratory pathogens. J Antimicrob Chemother. 2002;50(Suppl S2):59–73. doi: 10.1093/jac/dkf506. [DOI] [PubMed] [Google Scholar]
  • 3.Crum NF, Lee RU, Thornton SA, et al. Fifteen-year study of the changing epidemiology of methicillin-resistant Staphylococcus aureus. Am J Med. 2006;119:943–951. doi: 10.1016/j.amjmed.2006.01.004. [DOI] [PubMed] [Google Scholar]
  • 4.Margolis E, Yates A, Levin BR. The ecology of nasal colonization of Streptococcus pneumoniae, Haemophilus influenzae and Staphylococcus aureus: the role of competition and interactions with host’s immune response. BMC Microbiol. 2010;10:59. doi: 10.1186/1471-2180-10-59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Madhi SA, Adrian P, Kuwanda L, et al. Long-term effect of pneumococcal conjugate vaccine on nasopharyngeal colonization by Streptococcus pneumoniae - and associated interactions with Staphylococcus aureus and Haemophilus influenzae colonization--in HIV-Infected and HIV-uninfected children. J Infect Dis. 2007;196:1662–1666. doi: 10.1086/522164. [DOI] [PubMed] [Google Scholar]
  • 6.Zemlickova H, Melter O, Urbaskova P. Epidemiological relationships among penicillin non-susceptible Streptococcus pneumoniae strains recovered in the Czech Republic. Journal of Medical Microbiology. 2006;55:437–442. doi: 10.1099/jmm.0.46270-0. [DOI] [PubMed] [Google Scholar]
  • 7.Regev-Yochay G, Dagan R, Raz M, et al. Association between carriage of Streptococcus pneumoniae and Staphylococcus aureus in children. JAMA. 2004;292:716–720. doi: 10.1001/jama.292.6.716. [DOI] [PubMed] [Google Scholar]
  • 8.Bogaert D, van Belkum A, Sluijter M, et al. Colonisation by Streptococcus pneumoniae and Staphylococcus aureus in healthy children. Lancet. 2004;363:1871–1872. doi: 10.1016/S0140-6736(04)16357-5. [DOI] [PubMed] [Google Scholar]
  • 9.Quintero B, Araque M, van der Gaast-de Jongh C, et al. Epidemiology of Streptococcus pneumoniae and Staphylococcus aureus colonization in healthy Venezuelan children. European Journal of Clinical Microbiology and Infectious Diseases. 2011;30:7–19. doi: 10.1007/s10096-010-1044-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Jacoby P, Watson K, Bowman J, et al. Modelling the co-occurrence of Streptococcus pneumoniae with other bacterial and viral pathogens in the upper respiratory tract. Vaccine. 2007;25:2458–2464. doi: 10.1016/j.vaccine.2006.09.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Jourdain S, Smeesters PR, Denis O, et al. Differences in nasopharyngeal bacterial carriage in preschool children from different socio-economic origins. Clin Microbiol Infect. 2010;17:907–914. doi: 10.1111/j.1469-0691.2010.03410.x. [DOI] [PubMed] [Google Scholar]
  • 12.Abdullahi O, Nyiro J, Lewa P, Slack M, Scott JA. The descriptive epidemiology of Streptococcus pneumoniae and Haemophilus influenzae nasopharyngeal carriage in children and adults in Kilifi district, Kenya. Pediatr Infect Dis J. 2008;27:59–64. doi: 10.1097/INF.0b013e31814da70c. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.da Gloria Carvalho M, Pimenta FC, Jackson D, et al. Revisiting pneumococcal carriage by use of broth enrichment and PCR techniques for enhanced detection of carriage and serotypes. J Clin Microbiol. 48:1611–1618. doi: 10.1128/JCM.02243-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Shopsin B, Mathema B, Martinez J, et al. Prevalence of methicillin-resistant and methicillin-susceptible Staphylococcus aureus in the community. J Infect Dis. 2000;182:359–362. doi: 10.1086/315695. [DOI] [PubMed] [Google Scholar]
  • 15.Zetola N, Francis JS, Nuermberger EL, Bishai WR. Community-acquired methicillin-resistant Staphylococcus aureus: an emerging threat. Lancet Infect Dis. 2005;5:275–286. doi: 10.1016/S1473-3099(05)70112-2. [DOI] [PubMed] [Google Scholar]
  • 16.Murphy TF, Bakaletz LO, Smeesters PR. Microbial interactions in the respiratory tract. Pediatr Infect Dis J. 2009;28:S121–126. doi: 10.1097/INF.0b013e3181b6d7ec. [DOI] [PubMed] [Google Scholar]
  • 17.Howard AJ, Dunkin KT, Millar GW. Nasopharyngeal carriage and antibiotic resistance of Haemophilus influenzae in healthy children. Epidemiol Infect. 1988;100:193–203. doi: 10.1017/s0950268800067327. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Peacock SJ, Justice A, Griffiths D, et al. Determinants of acquisition and carriage of Staphylococcus aureus in infancy. J Clin Microbiol. 2003;41:5718–5725. doi: 10.1128/JCM.41.12.5718-5725.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Pai R, Gertz RE, Beall B. Sequential multiplex PCR approach for determining capsular serotypes of Streptococcus pneumoniae isolates. J Clin Microbiol. 2006;44:124–131. doi: 10.1128/JCM.44.1.124-131.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.de Lamballerie X, Zandotti C, Vignoli C, Bollet C, de Micco P. A one-step microbial DNA extraction method using “Chelex 100” suitable for gene amplification. Research in Microbiology. 1992;143:785–790. doi: 10.1016/0923-2508(92)90107-y. [DOI] [PubMed] [Google Scholar]
  • 21.Kilic A, Muldrew KL, Tang YW, Basustaoglu AC. Triplex real-time polymerase chain reaction assay for simultaneous detection of Staphylococcus aureus and coagulase-negative staphylococci and determination of methicillin resistance directly from positive blood culture bottles. Diagnostic Microbiology and Infectious Disease. 2010;66:349–355. doi: 10.1016/j.diagmicrobio.2009.11.010. [DOI] [PubMed] [Google Scholar]
  • 22.Wang X, Mair R, Hatcher C, et al. Detection of bacterial pathogens in Mongolia meningitis surveillance with a new real-time PCR assay to detect Haemophilus influenzae. International Journal of Medical Microbiology. 301:303–309. doi: 10.1016/j.ijmm.2010.11.004. [DOI] [PubMed] [Google Scholar]
  • 23.LaClaire LL, Tondella ML, Beall DS, et al. Identification of Haemophilus influenzae serotypes by standard slide agglutination serotyping and PCR-based capsule typing. J Clin Microbiol. 2003;41:393–396. doi: 10.1128/JCM.41.1.393-396.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Tettelin H, Nelson KE, Paulsen IT, et al. Complete genome sequence of a virulent isolate of Streptococcus pneumoniae. Science. 2001;293:498–506. doi: 10.1126/science.1061217. [DOI] [PubMed] [Google Scholar]
  • 25.Rodriguez-Lazaro D, Lewis DA, Ocampo-Sosa AA, et al. Internally controlled real-time PCR method for quantitative species-specific detection and vapA genotyping of Rhodococcus equi. Appl Environ Microbiol. 2006;72:4256–4263. doi: 10.1128/AEM.02706-05. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Kuroda M, Ohta T, Uchiyama I, et al. Whole genome sequencing of meticillin-resistant Staphylococcus aureus. Lancet. 2001;357:1225–1240. doi: 10.1016/s0140-6736(00)04403-2. [DOI] [PubMed] [Google Scholar]
  • 27.O’Brien KL, Nohynek H. Report from a WHO Working Group: standard method for detecting upper respiratory carriage of Streptococcus pneumoniae. Pediatr Infect Dis J. 2003;22:e1–11. doi: 10.1097/01.inf.0000049347.42983.77. [DOI] [PubMed] [Google Scholar]
  • 28.Smith CB, Golden CA, Kanner RE, Renzetti AD. Haemophilus influenzae and Haemophilus parainfluenzae in chronic obstructive pulmonary disease. Lancet. 1976;1:1253–1255. doi: 10.1016/s0140-6736(76)91733-5. [DOI] [PubMed] [Google Scholar]
  • 29.Abdeldaim GM, Stralin K, Kirsebom LA, et al. Detection of Haemophilus influenzae in respiratory secretions from pneumonia patients by quantitative real-time polymerase chain reaction. Diagnostic Microbiology and Infectious Disease. 2009;64:366–373. doi: 10.1016/j.diagmicrobio.2009.03.030. [DOI] [PubMed] [Google Scholar]
  • 30.Vu HT, Yoshida LM, Suzuki M, et al. Association between nasopharyngeal load of Streptococcus pneumoniae, viral coinfection, and radiologically confirmed pneumonia in Vietnamese children. Pediatr Infect Dis J. 2011;31:11–18. doi: 10.1097/INF.0b013e3181f111a2. [DOI] [PubMed] [Google Scholar]
  • 31.Smith-Vaughan H, Byun R, Nadkarni M, et al. Measuring nasal bacterial load and its association with otitis media. BMC Ear Nose Throat Disord. 2006;6:10. doi: 10.1186/1472-6815-6-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Regev-Yochay G, Trzcinski K, Thompson CM, Malley R, Lipsitch M. Interference between Streptococcus pneumoniae and Staphylococcus aureus: In vitro hydrogen peroxide-mediated killing by Streptococcus pneumoniae. Journal of Bacteriology. 2006;188:4996–5001. doi: 10.1128/JB.00317-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.van Gils EJ, Hak E, Veenhoven RH, et al. Effect of seven - valent pneumococcal conjugate vaccine on Staphylococcus aureus colonisation in a randomised controlled trial. PLoS ONE. 2011;6:e20229. doi: 10.1371/journal.pone.0020229. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Weimer KE, Armbruster CE, Juneau RA, et al. Coinfection with Haemophilus influenzae promotes pneumococcal biofilm formation during experimental otitis media and impedes the progression of pneumococcal disease. J Infect Dis. 202:1068–1075. doi: 10.1086/656046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Pericone CD, Overweg K, Hermans PW, Weiser JN. Inhibitory and bactericidal effects of hydrogen peroxide production by Streptococcus pneumoniae on other inhabitants of the upper respiratory tract. Infect Immun. 2000;68:3990–3997. doi: 10.1128/iai.68.7.3990-3997.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Shakhnovich EA, King SJ, Weiser JN. Neuraminidase expressed by Streptococcus pneumoniae desialylates the lipopolysaccharide of Neisseria meningitidis and Haemophilus influenzae: a paradigm for interbacterial competition among pathogens of the human respiratory tract. Infect Immun. 2002;70:7161–7164. doi: 10.1128/IAI.70.12.7161-7164.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Lysenko ES, Ratner AJ, Nelson AL, Weiser JN. The role of innate immune responses in the outcome of interspecies competition for colonization of mucosal surfaces. PLoS Pathog. 2005;1:e1. doi: 10.1371/journal.ppat.0010001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Pettigrew MM, Gent JF, Revai K, Patel JA, Chonmaitree T. Microbial interactions during upper respiratory tract infections. Emerg Infect Dis. 2008;14:1584–1591. doi: 10.3201/eid1410.080119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Binks MJ, Cheng AC, Smith-Vaughan H, et al. Viral-bacterial co-infection in Australian indigenous children with acute otitis media. BMC Infect Dis. 11:161. doi: 10.1186/1471-2334-11-161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Albrich WC, Madhi SA, Adrian PV, et al. Use of a rapid test of pneumococcal colonization density to diagnose pneumococcal pneumonia. Clin Infect Dis. 54:601–609. doi: 10.1093/cid/cir859. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES