Prevalance of genes involved in pili and biofilm formation and in vitro adherence properties of medical and fecal strains of Enterococcus faecalis isolated in Gdańsk

Abstract

Adherence properties of 78 Enterococcus faecalis strains from clinical specimens and healthy volunteers were measured by two methods: microculture and nephelometric assays. In addition selected genes involved in adherence and pili formation were detected by PCR. They were more prevalent in clinical strains compared with fecal isolates from healthy volunteers. Microculture assay showed good correlation between enterococcal adherence and the presence of ebpA/ebpB pili genes. In contrast to the previous method, the results of nephelometric assay were not influenced by the presence of ebpA/ebpB genes. Both methods together could provide meaningful information on adherence and differentiate the group of genes responsible for it.

Introduction

Enterococci may be commensals and a part of the normal human flora or they may be pathogens if isolated from a normally sterile site. One of their virulence factors is adherence and the ability to produce biofilm on various intravenous and prosthetic devices [1–5]. Protein adhesins, e.g. enterococcal surface protein (Esp), are thought to mediate enterococcal attachment to host tissues [3], [6–8] and are necessary for biofilm formation [9–12], while aggregation substance (which is a product of agg gene) promotes aggregation of replicating microbes [1], [2], [13]. Among Gram-negative bacteria it has long been known that adhesion is a result of the presence of pili or fimbriae [14], [15], but only recently limited information on Gram-positive pili has been published [13–16]. The authors demonstrated the presence of pili composed of EbpA, EbpB, and EbpC on the surface of E. faecalis and suggested that pili are required for the establishment of enterococcal infections. It is still unclear as to which of the above-mentioned adherence factors are most important.

The aim of the present study was: 1) to study the presence of pili-encoding genes ebpA and ebpB in medical and commensal isolates of E. faecalis in the Gdańsk region and 2) to analyze the in vitro adherence ability of the studied strains.

Materials and methods

A total of 78 strains of E. faecalis were examined, including 38 strains from healthy volunteers (stool samples) and 40 clinical isolates (6 from ulcers, 6 from vagina, 21 from urine, 7 from respiratory specimens). All strains were collected from 2005 to 2007 in various hospitals in the Gdańsk region. Strains OG1R(pCF10) and OG1RF (pAD1) were used as reference strains ( kindly provided by Gary Dunny from the University of Michigan).

Detection of adherence genes

Bacterial DNA was isolated using a commercially available kit (A&A Biotechnology, Poland). The presence of agg, esp, ebpA, and ebpB genes was detected by the PCR method, as described earlier [14], [17] (agg forward AAGAAAAAGAAGTAGACCAAC, reverse AAACGGCAAGACAAGTAAATA, esp forward TTGCTAATGCTAGTCCACGACC, reverse GCGTCAACACTTGCATTGCCGAA, ebpA forward CCGCTCGAGAACTAACAAAAATGATTCGGCTCCAG, reverse CCGCTCGAGCCAT CTCACGCATTTTATCTTCAACT, ebpB forward CCGCTCGAGCTGAAGGAAAAACGGTCCAA, reverse CCGCTCGAGCTTTTGCGTCGTCAGTGTGT). PCR was performed in a 50 µl reaction mixture using 5 µg of DNA, 20 pmol of each primer, and 1 U Taq DNA polymerase (MBI Fermentas, Lithuania) in a Biometra thermocycler (Biometra, Germany). Sample without DNA was used as a negative control. Initial denaturation lasted for 2 min at 94°C, annealing at 40°C (detection of agg and esp genes), 50°C or 60°C (detection of ebpA and epbB), and elongation at 72°C for 2 min followed by 29 cycles of denaturation at 94°C for 2 min, annealing at the appropriate temp for 15 s, and final elongation at 72°C for 2 min. Results were visualized on 2% agarose (Prona Marine Research Institute, Spain) stained with ethidium bromide.

The biofilm assay was performed using flat-bottomed highly adherent polystyrene plates (Sigma, USA) as described earlier [8]. All bacterial strains were grown at 37°C for 18 h in brain heart infusion (BHI). Optical densities of bacterial suspension were measured at 600 nm using a Spekol spectrophotometer and normalized (OD 0.1). Bacterial cultures were diluted 1:280 in BHI supplemented with 0.5% glucose; 150 µl of diluted culture was dispersed in nine wells in a single row of a sterile plate. As a control, an additional nine wells were filled with 150 µl sterile TSB. After incubation for 18 h at 37°C the medium was aspirated and wells were washed with phosphate-buffered saline (PBS). The plates were dried at room temperature for 1 h and 200 µl of 0.25% aqueous crystal violet solution was added to each well and incubated for 1 h. After incubation, the wells were washed three times with PBS. For biofilm quantification, crystal violet bound to biofilm cells was extracted with 200 µl of 4:1 ethanol and acetone mixture. Absorbance was measured at 595 nm with the Spekol spectrophotometer. Each experiment was done three times.

Adherence was determined as described previously by nephelometric assay [3]. Briefly, cells from all overnight cultures were centrifuged, washed, and suspended in 0.9% NaCl to reach a density of 5×10⁴ cfu/ml (T_at550nm=50%). Cells were incubated with 10% hydroxyapatite solution for 30 min at room temperature with gentle mixing and left for sedimentation for another 30 min. Transmission (reduction of light intensity in suspension) was measured and inoculum value was estimated following the experimentally designated relation between transmission (T) and density of inoculum (cfu/ml) equation [3]. Adherence was estimated as the difference in inoculum before and after the addition of hydroxyapatite. Each experiment was done three times.

Statistical analysis

Statistical significance was set a priori at p <0.05. The influence of particular genes on adherence was compared with ANOVA tests (Statistica software, Statsoft, USA). We also used the χ² test.

Results and discussion

Pathogenesis of E. faecalis relies on fewer virulence factors than in many well known pathogens and differentiation of commensal and pathogenic strains is difficult. As a role of pili in virulence of this bacterium has been described only recently, we decided to compare the prevalence of two pili-related genes as a marker of pili-mediated virulence in our strains

The enterococcal pili genes detected are presented in Table I. We confirmed a significantly higher prevalence of both ebp genes in clinical compared with fecal isolates (p <0.05). and the difference in ebp composition between fecal and clinical strains (Table I). Comparison of prevalence of esp and agg genes studied previously [17] showed no such significant difference (e.g agg gene was detected in 64.1% of commensal strains while in medical isolates it varied from 75.0% to 88.18% [17]), so the result shows the importance of the studied ebp genes in pathogenesis. However, the role of genes encoding surface protein is still under discussion. Some reports showed high prevalence of esp genes in medical isolates and lack of this gene in strains isolated from healthy volunteers [9], [17] but another study did not support such findings [8].

Table I.  Prevalence of enterococcal pili genes in clinical and fecal isolates.

Bacteria	No ebp genes	ebpA alone	ebpB alone	ebpA + ebpB
Fecal isolates	10/38	9/38 (24%)	7/38 (18%)	12/38 (31%)
Clinical isolates	0/40*	0/40*	0/40 *	40/40 (100%)*
Reference strains	0/2	0/2	0/2	2/2

*Differences between fecal and clinical isolates at p<0.05.

Results obtained from phenotypic assay are shown in Tables II and III. Biofilm assay showed increased adherence in strains when only one of the studied genes was present and even more increased adherence in the case of two genes detected. It was statistically significant for ebpA and ebpB genes and for agg and esp (p <0.05). The role of enterococcal pili in biofilm formation has recently been proved by some authors [14]. Both surface protein genes (agg and esp) and pili genes (ebpA and ebpB) demonstrated some influence on adhesion properties determined by biofilm assay. Adherence was shown to be associated with the product of esp gene by many authors [10–12]. One should also notice a difference in biofilm properties determined in vitro in strains within the same type of clinical specimens [6]. These divergences make any assumptions difficult, especially since factors influencing biofilm formation are not sufficiently known [11], [12].

Table II.  Adherence determined by biofilm assay (ANOVA comparison between columns 2–4 and 5–7).

Presence of genes	No adherence genes	esp or agg gene	agg + esp	No pili-encoding genes	ebpA or ebpB	ebpA + ebpB
Mean optical density	0.175	0.4	1.97	0.175	1.32	1.96
SE	0.638	0.45	0.23	0.69	0.49	0.25
ANOVA comparison	F = 4.65, p =0.02			F = 7.58, p=0.003

Table III.  Adherence determined by nephelometric method (ANOVA comparison between columns 2–4 and 5–7).

Presence of genes	No adherence genes	esp or agg gene	agg + esp	No pili-encoding genes	ebpA or ebpB	ebpA + ebpB
Percentage of adherent cells	62.5%	70.5%	84%	62.5%	87.2%	83%
SE	0.02	0.01	0.07	0.038	0.022	0.02
ANOVA comparison	F = 136.31, p<0.01			F = 1.5849, p=0.2248

In contrast to results obtained with the biofilm assay, adherence measured by the nephelometric assay was not influenced by the presence of pili genes. In this case the mechanism of adhesion seems to be different. The adsorption of bacteria, which have a negative charge, is due to the electrostatic attraction between the negatively charged cell and the positively charged calcium atoms of the hydroxyapatite [18]. So comparison of results obtained by biofilm assay and nephelometric assay can be used to differentiate pili-mediated and non-pili-mediated adherence.

Our study shows an interesting influence of pili-encoding genes and other adherence-related enterococcal genes on adherence.

Acknowledgements

We thank Prof G. Dunny for supplying reference strains. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.