| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
In their natural environment, trophozoites of E. histolytica live in the colonic region of the human intestine together with resident microbial flora, which under normal conditions is composed of a complex mixture of mostly anaerobic or microaerophilic bacteria.4 The predominant anaerobic species in this flora belong to Bacteroides, Bifidobacterium, Eubacterium, Clostridium, Peptococcus, Peptostreptococcus, and Ruminococcus, whereas facultative anaerobes, such as Escherichia coli, Enterobacter, Enterococcus, Klebsiella, Lactobacillus, and Proteus, are among the subdominant genera.5 The composition of the individual’s flora can fluctuate in conditions like acute diarrhea6 and antibiotic treatment. Although the microbiota in healthy adults have been studied extensively, little is known about the changes in flora that occur in protozoan-associated gastrointestinal infection.
It has been suggested that the bacterial flora provides anaerobic conditions or low redox potential beneficial for amebic growth.7 Trophozoites of E. histolytica grown in association with bacteria are active feeders that phagocytose numerous bacteria.8 E. histolytica appears to be selective with respect to association with different bacterial species, and only those bacteria that possess the appropriate recognition signals will become attached and ingested by the ameba. It has been speculated that certain bacterial species of the gut may trigger the virulent potential of the trophozoites while others may have no effect or may even cause avirulence.9 Metronidazole is known to be clinically more effective against ameba in tissue than lumenal ameba.10 This may be, in part, due to anaerobic bacteria that co-exist with the amebic trophozoites in vivo may bring about the reduction and elimination of metronidazole molecules needed for the killing of amebic trophozoites.10 The incidence of metronidazole-resistant anaerobic bacteria, especially some species of Bacteroides, appears to be on the rise.11 A molecular study demonstrated that the moderate resistance phenotype of these clinical strains was in all cases associated with the presence of the 5-NI (nitroimidazole) resistance gene (nim).12
Systematic studies are lacking to provide a clinical correlation between resident bacterial flora of the gut and the severity and spectrum of amebic disease. In the present study, we have looked into the profile of predominant gut flora of healthy individuals, asymptomatic E. histolytica carriers, and amebic liver abscess patients. We have also addressed the issue whether the changes observed in the gut flora may be attributed to the administration of metronidazole or the presence of the parasite. We have also looked for the presence of gut bacteria in the liver abscess aspirates. The prevalence of the metronidazole-resistance gene (nim) was also scored in these bacterial species. Our results indicate a significant loss of some predominant gut bacteria in amebiasis patients and the presence of some anaerobic species in amebic liver abscess aspirates.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Samples were also collected before and after metronidazole treatment (300 mg twice per day; 3-day course) from a group of 11 individuals: 8 healthy volunteers (residents of urban area, Jawaharlal Nehru University Campus, India) and 3 patients suffering from irritable bowel syndrome (IBS) reporting to AIIMS.
Informed consent was obtained from patients and healthy individuals prior to sample collection. Amebic liver abscess patients were from different economic backgrounds and came from different local areas in Delhi. Diagnosis of amebic liver abscess was based on four or more of the following criteria: (i) A space, occupying lesion in the liver diagnosed by ultrasonography and suggestive of abscess; (ii) clinical symptoms (fever, abdominal pain), (iii) enlarged and/or tender liver, usually without jaundice, (iv) in some cases diarrhea, and bloody diarrhea in a few cases, and (v) positive amebic serology. Of 35 patients, 33 were positive for amebic serology with ELISA Kit containing IgG antibodies against E. histolytica (R-Biopharm AG, Darmstadt, Germany). To avoid any secondary infection, amebic liver aspirates were collected only from those patients who were being aspirated for the first time. Out of 35 ALA patients, fecal samples were collected from 19 ALA patients.
Most of the ALA patients were males (33/35). Most of them had consumed metronidazole (mean duration of drug intake, 6 days).
Fecal and ALA pus specimens.
Fecal and pus samples were collected from various categories of individuals, as listed in the Tables 1
and 2. Aliquots of all fecal samples as well as pus samples were preserved at –20°C till DNA isolation was done. The aspiration procedure for ALA was carried out in a sterile environment in the Intensive Care Unit by a clinician with the required precautions.
|
|
PCR using genus-specific primers. Universal primer set; S-D-Bact-0008-a-S-20 (5' AGA GTT TGA TCC TGG CTC AG 3',13 which targets the domain Bacteria, and S-*-Univ-1492-b-A-21 (5' ACG GCT ACC TTG TTA CGA CTT 3',14 which targets all living organisms, were used to amplify bacterial 16S rDNAs from the collected samples to ensure the DNA quality. Later genus-specific primers were used to amplify the selected bacteria DNA from fecal and pus samples.
For all primer sets, 30-cycle PCR at different annealing temperatures was performed in a Tech gene thermal cycler (NuGEN Scientific, San Carlos, CA).15 Genus-specific primer sets were designed from 16S rRNA sequences of the following bacteria: Bifidobacterium (Bif), Clostridium (Clos), Ruminococcus (Rum), Campylobacter (Camp), Pseudomonas aeruginosa (Pseudo), Lactobacillus (Lacb), and Peptococcus (Pep). Specificity of these primers was checked by using target and non-targeted bacterial genomic DNA as described previously.15 Primers for Bacteroides (Bacto) and Peptostreptococcus productus (PSP) were obtained from the literature.16,17 All primers were commercially synthesized by Microsynth GmbH (Balgach, Switzerland).
Primer set used for the detection of Escherichia coli was based on the mal B gene17 while for Staphylococcus aureus, the primers mecA-1 and mecA-2 used were synthesized from methicillin-resistant gene.18
Genomic DNA isolated from human blood (Genomic DNA isolation kit, MBI Fermentas, St. Leon Rot, Germany) was used as a template DNA for ensuring primers’ specificity.
PCR products derived from healthy individuals’ fecal DNA were cloned and sequenced. These were used as controls in subsequent PCR experiments.
Detection of E. histolytica-positive samples. For the PCR-based detection of E. histolytica among all fecal and pus samples, E. histolytica-specific UEE primers were used.19
Detection of nim genes. PCR assays were carried out using the universal set of primers, Nim-3 and Nim-5, for all known nim genes.20 These primers amplify a PCR product of 458 bp. Amplification was done under conditions as previously described21 using an annealing temperature of 52°C. Wild-type strain Bacteroides fragilis (MTCC 1045) was used as a negative control.
Selected amplicons of nim genes (size = 458 bp) obtained for fecal and pus samples, were purified using a Qiagen Gel Extraction kit and cloned into a pGEMT-Easy vector (Promega Co., Madison, WI) as per the manufacturer’s instructions. Clones were further sequenced.
Analysis of sequenced data. Sequences obtained for representative cloned amplicons from each category were analyzed using the BLASTN program for a homology search within the existing database for each genus.
Statistical analysis. One-way ANOVA and Fisher’s exact tests were performed to check the statistical significance of the data. Statistical significance was accepted at the level of 0.05, and probability values were calculated for two-tailed possibilities.
|
RESULTS AND DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
In the present study, two major aspects have been dealt with: (i) changes in gastrointestinal flora as a result of E. histolytica infections and (ii) bacterial co-infection in amebic liver abscess.
Stool samples were analyzed from the following categories of individuals: (1) healthy E. histolytica negative (N = 19); (2) healthy, asymptomatic, E. histolytica positive (N = 11); and (3) ALA patients (N = 19). Pus samples were also analyzed from ALA patients (N = 35) for the presence of 11 prominent, mostly anaerobic bacteria found in the human gut. Bacterial detection was done by PCR amplification of total DNA from samples, using genus specific primers for 16S rRNA gene. The specificity of detection was ascertained by sequencing the PCR-amplified bands and confirming their identity by comparing with genus specific sequences available in the database.
Figure 1 represents typical pattern for the PCR amplification of various genera.
|
). The significance of the PCR-based percent incidence results was evaluated by statistical analysis. One-way ANOVA test and two-tailed Fisher’s exact test indicated a significant variation in fecal flora in the various categories tested (Table 3). To confirm the results of PCR analysis from stool and pus samples, randomly selected amplicons from each primer set were cloned and sequenced. BLASTN results revealed 94–100% identity with 16S rRNA sequences of the corresponding bacterial species (from which the primer was designed; data not shown). The sequences were submitted to the database, and accession numbers were obtained (Table 4). The high level of sequence identity showed that the assignment of bacterial genera based on 16S rRNA was reliable. Species names mentioned in Table 4 are based on the highest score values in BLAST N results for the query sequence.
|
|
). However, Lactobacillus was reduced in E. histolytica–asymptomatic carriers also (percent incidence = 36.3%; P = 0.0011). Asymptomatic carriers also showed a statistically significant decrease in the incidence of E. coli and increase in Pseudomonas aeruginosa compared with the other two categories of individuals. No significant changes were observed for Peptostreptococcus productus, Ruminococcus, Campylobacter, and Peptococcus among healthy controls and ALA patients (Table 1). Because ALA patients were on metronidazole treatment for ethical reasons, and E. histolytica–asymptomatic carriers had not taken metronidazole treatment, it was possible that, the observed reduction in Bacteroides, Bifidobacterium, and Clostridium in ALA patients alone was due to metronidazole treatment while reduction in Lactobacillus seen in both ALA patients and asymptomatic carriers was due to E. histolytica infection.
This was confirmed by directly testing the effect of metronidazole on the gut flora of healthy volunteers (residents of urban area) and IBS patients who were not suffering from amebiasis. Bacterial flora in fecal samples of urban area resident healthy volunteers (taken before metronidazole treatment) differed from the slum area resident healthy volunteers. This could be due to differences in their living conditions and dietary habits.23 These individuals were given metronidazole for a period of 3 days. None of these individuals was E. histolytica–positive, as tested by E. histolytica–specific UEE primers.21
There was no change in the incidence of Lactobacillus in these individuals as a result of metronidazole treatment, although there was a marked drop in the incidence of Bifidobacterium, Bacteroides, and Clostridium. Metronidazole is known to be very effective against obligate anaerobes, which explains the drop in Bifidobacterium, Bacteroides, and Clostridium.24 However, Lactobacillus, being facultative anaerobe, is not significantly affected by this drug. Thus, our results show that, of the 11 bacterial genera tested, at least Lactobacillus is significantly reduced in individuals infected with E. histolytica. Such a loss of anaerobic flora has been reported under conditions of severe diarrhea25 and Clostridium difficile–associated diarrhea.26
In keeping with the facts that metronidazole is effective only against obligate anaerobic bacteria and that P. aeruginosa is a strictly aerobic bacterium, no change was observed in the incidence of P. aeruginosa in healthy volunteers and IBS patients after metronidazole administration (Table 2). Our observations regarding the low prevalence of E. coli in the fecal samples of healthy adult individuals could be attributed to the fact that the level of this population may have been below detectable limits before the drug treatment. This could be due to the lower sensitivity of mal B gene–based PCR primers used for the detection of E. coli. The 16S rRNA gene could not be used for E. coli due to the lack of specificity.27 As a result of drug-treatment overgrowth of aerobic bacteria including E. coli may have occurred (Table 2), concomitant with the reduction in anaerobic bacteria as reported earlier.28 This was also supported by a study carried out with human fecal samples in which bifidobacteria were present in higher number than E. coli.29
No significant change was observed in P. productus or Ruminococcus titers among the various categories of individuals tested. Members of these genera are likely to be resistant both to drug action as well as E. histolytica infection.
To our knowledge, an extensive investigation of bacterial prevalence using sensitive DNA probes has not been carried out with E. histolytica infected individuals. This needs to be substantiated with more patients from a variety of geographical locations. If a general pattern emerges, then a probiotic therapy to replace the depleted bacterial species could be envisaged.
Bacteria in pus samples. DNA was extracted from ALA pus samples and tested for the presence of various bacterial genera as described for stool samples. No amplification was observed with primers specific for aerobic bacteria (E. coli, S. aureus, and P. aeruginosa).
A remarkable observation reported in this study is the highly significant occurrence of Peptostreptococcus (25/35 ALA cases, 71.4% occurrence) and, less frequently, Bacteroides (5/35 ALA cases, 14.2% occurrence) in the pus samples of ALA patients (Figure 2). Pus from these patients is generally considered to be sterile on the basis of culturing and microscopy.9 Because our detection method was based on PCR amplification, there could be a chance of primer cross-reaction or bacterial contamination.
|
Several species of Peptostreptococcus and Bacteroides including P. productus and B. fragilis are known to be involved in causing abscesses and other septicemia-related infections in humans. They are frequently associated with pyogenic liver abscess.30 However, their involvement with ALA as suggested here has not, so far, been reported. In a previous study, superinfection of an amebic abscess with Salmonella enteridis has been reported.31 However, in this case there was hemorrhage into the abscess due to physical trauma, which was followed by secondary bacterial superinfection. Reports are available about possible mechanisms of bacterial translocation to extraintestinal sites. These mechanisms, including intestinal bacterial overgrowth, are increased permeability of mucosal barrier (leaky gut syndrome) and deficiencies in host immune response.32 Whether these mechanisms continue to operate in mixed infection of these anaerobic bacteria along with the E. histolytica or whether bacteria passively gain entry into the liver abscess caused by E. histolytica trophozoites needs to be studied. It is important to know the extent of involvement of bacteria in initiating or enhancing the severity of ALA to effectively treat these patients.
Whether the bacteria detected by us were viable cannot be assessed because PCR amplification alone was used. In an attempt to cultivate bacteria from ALA pus samples from a population of Bangladesh, only 19% samples were found to be positive for a few aerobic gram- negative bacteria like E. coli, Proteus, and Pseudomonas. However, presence of any anaerobic bacteria was not reported.33 This could be due to the difficulty in cultivating anaerobes. The very high incidence of Peptostreptococcus reported here in pus samples (25 out of 35 ALA cases) is unprecedented. Bacteroides although found in 5 out of 35 ALA cases, is unreported so far from pus samples. Further confirmation of this data will require anaerobic cultivation of viable bacteria from ALA pus samples.
Presence of nim genes in ALA patients. In the last few years, the occurrence of nim genes associated with metronidazole resistance has been found to be widespread among anaerobes, including Bacteroides34 and Peptostreptococcus.35 Many of these nim genes are plasmid encoded, while some are also chromosomally located.36 Five nim genes, nimA to -E, have been identified that confer reduced susceptibility to 5-nitroimidazole antibiotics (e.g., metronidazole) on species of the B. fragilis group.34 The proposed resistance mechanism conferred by the nim genes is that they encode a 5-nitroimidazole reductase.37 Four of the nim genes have been shown to be associated with different mobile insertion sequence (IS) elements flanked by inverted repeats. There is strong evidence that these IS elements carry regulatory signals for expression of certain resistance genes, including the nim genes.38
The presence of nim genes was determined in all fecal and pus samples by PCR amplification with nim-specific primers.20 No amplicons were observed in healthy E. histolytica–negative individuals and asymptomatic carriers, while amplicons of the sizes expected for nim gene were present in both pus and fecal samples of ALA patients (Table 5). Selected clones of representative PCR products from both fecal and pus samples of ALA patients were sequenced. Sequence analysis using BLASTN revealed 99–100% identity with nim E gene from B. fragilis (database entry) for amplicons from both fecal and pus samples. Because ALA patients were receiving metronidazole, this result shows that nim genes may be rapidly amplified in bacterial populations after antibiotic challenge.
|
The study presented here initiates discussion of the need to rethink and reconsider the ameba-bacterium relationship inside the intestine and in extraintestinal tissues. The mechanism by which anaerobic bacteria reach the amebic liver abscess and their role, if any, in amebic pathogenesis need to be explored using more experimental evidence.
Received June 3, 2006. Accepted for publication July 14, 2006.
Acknowledgments: The authors acknowledge Miss Shweta Srivastava in screening E. histolytica–positive samples and Miss Rina Chakravorty (Indian Statistical Institute, New Delhi) for helping in statistical analysis of data.
Financial Support: This work was supported by grants from University Grants Commission and Department of Science and Technology, India. The American Society of Tropical Medicine and Hygiene (ASTMH) and the American Committee on Clinical Tropical Medicine and Travellers’ Health (ACCTMTH) assisted with publication expenses.
* Address correspondence to Jaishree Paul, School of Life Sciences, Jawaharlal Nehru University, New Delhi, 110067 India. E-mail: jpaul33{at}hotmail.com 
Authors’ addresses: Rekha Rani and Jaishree Paul, School of Life Sciences, Jawaharlal Nehru University, New Delhi, 110067 India, Fax: +91-011-26717580, E-mails: jpaul33{at}hotmail.com and rekhs2004garg{at}gmail.com. R.S. Murthy and Vineet Ahuja, Department of Gastroenteology, All India Institute of Medical Sciences, New Delhi, India. Sudha Bhattacharya, School of Environmental Sciences, Jawaharlal Nehru University, New Delhi, 110067, India. M.A. Rizvi, Department of Biosciences, Jamia Millia Islamia, New Delhi-110025, India.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
