| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The causative agent of trench fever is Bartonella quintana, which is a short, pleomorphic, Gram-negative bacterium thought to be transmitted through the feces of infected body lice to human beings by scratching of the skin.1 The human body louse, Pediculus humanus, is known to be the vector of B. quintana,1 but there are few studies that describe details of the growth kinetics and transmission of the bacteria. By using an animal model, Fournier and colleagues performed an experimental infection of rabbit-adapted body lice with B. quintana that expressed green fluorescent protein.9 They showed B. quintana being excreted in the feces and that life span and survival rate of body lice were not significantly influenced by B. quintana infection.9 However, their study was not sufficient to understand the proliferation dynamics in the body lice and persistent excretion of B. quintana in the feces.
Recently Kasai and colleagues10 developed an artificial blood feeding system using human whole blood, and the blood-fed adult body lice were able to be kept alive for at least 3 weeks in the laboratory.
In this study, a membrane feeding system was used to provide infectious blood meal and we report herein the proliferation dynamics of B. quintana in body lice and the massive excretion of B. quintana in louse feces.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bacterial strain and culture conditions. All procedures involving experimental infections of lice were conducted in a biosafety level 2 (BSL2) room at the National Institute of Infectious Diseases, Tokyo, Japan. The strain of B. quintana (Hassani strain), originally colonized from blood samples from homeless patients in Marseille, France, was used. The bacteria were cultured using selective media on Columbia sheep blood agar plate (5%, BioMerieux, Marcy l’Etoile, France) at 37°C under a 5% CO2 atmosphere, as described previously.8
On Day 10 post-plating, the bacteria were harvested by adding 4 mL of phosphate-buffered saline (PBS, Dulbecco, 0.1 M, pH 7.2), and 50 µL of the bacterial suspension was mixed with 1 mL of whole human blood and used for an experimental infection.
For the purpose of estimating the initial concentration of bacteria as colony forming units (CFU) per microliter, the bacterial suspension was diluted from 104 to 107 times and 10 µL of each dilution was cultured on the blood agar plates.
Cultures of B. quintana from feces of experimentally infected body lice on the 11 days after infection were also carried out by plating homogenized feces in PBS on the blood agar plates. After colonies formed on the plate, a second culture was performed and identification of bacterial species from the colonies was carried out by sequencing PCR-amplified product.
Membrane feeding system and louse infection with B. quintana.
The artificial membrane feeding method was performed according to Kasai.10 Human whole blood was prepared by adding an anti-coagulant, sodium citric acid (final concentration, 0.3%), followed by freezing at –80°C until use. Parafilm (American Can Co., Neemah, WI) was stretched to 4-fold (=16-fold of the original area) and put on the top of a glass Petri dish (90 mm in diameter). A circular cotton pad (30 mm in diameter and 2 mm in thickness) was put on the Para-film. One milliliter of blood was absorbed onto the cotton and covered with another sheet of stretched Parafilm. The circumference of the cotton pad was attached securely to ensure that the blood-containing cotton was sealed. The edge of the blood pad was clipped with a razor, placed on the bottom of another Petri dish, and incubated on a heating block at 37°C. A small piece of felt (round, 15 mm in diameter) infested with female lice was put on the center of the blood pad, and lice were released onto it to start feeding (Figure 1A
).
|
Isolation of genomic DNA. Genomic DNA was isolated using a SepaGene kit (Sanko Chemicals, Tokyo, Japan). Each louse was homogenized in 100 µL of the homogenization buffer (Reagent I; Tris-HCl buffer, pH 7.9) using a pestle for 30 sec in a 1.5-mL plastic tube. Feces in the 20-mL plastic tube were collected by centrifugation at 1500g for 10 min and then mixed with 200 µL of the homogenization buffer. The subsequent DNA extraction proceeded by the manufacturer’s instructions.
Quantitative real-time polymerase chain reaction (qRT-PCR). The qRT-PCR was performed with a SYBR-green detection system using iCycler (Bio-Rad, Tokyo, Japan). Specific primers were designed based on the gene sequence of the intergenic spacer (ITS) region of B. quintana: BQ-ITS-Forward, GCCGCCTTCGTTTCTCTTTC and BQ-ITS-Reverse: AGTGTCTTCCTTAAAGTCCCAAAG. Body louse sodium-channel gene-specific primers (L-SC-Forward, ACGATGATGAAGATGAAGATGAAGG and L-SC-Reverse, CAAGAGGAGTAGACGCCAATTC) were also designed to correct the efficacy of DNA extraction and standardize the number of B. quintana.12 The qRT-PCR conditions were as follows: the reaction mixtures were first kept at 95°C for 10 min, then put through 50 cycles of 95°C for 15 sec and 60°C for 60 sec. The number of B. quintana in each sample was calculated on the basis of DNA copy number of B. quintana. Standard curves were obtained using serial dilutions of purified PCR products obtained with the above primer sets, covering five template doses and yielding correlation coefficients of at least 0.98 in all experiments. Each standard value was determined in triplicate in every experiment. After 50 cycles of the PCR reaction, samples were run for the dissociation protocol, and we confirmed that only a single melting peak was obtained for all samples, showing no amplification of the non-target DNA. The number of amplified DNA copies were converted into the number of bacteria (1 attomoles = 3.01 x 105 cells). The average number of bacteria per louse was calculated from the data of 5 lice with standard error.
Preparation of feces of body lice for scanning electron microscopy. The feces from 4 body lice after 17-day infection were fixed with 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M PBS for 30 min and washed with 0.1 M PBS 3 times. After dehydration through an ascending alcohol series, the specimens were freeze-dried (model ES-2030, Hitachi, Tokyo, Japan) using tert-butyl ethanol, coated with osmium, and examined with a scanning electron microscope (model S-5200; Hitachi).
Preparation of feces of body lice for immunofluorescence assay. The feces from 10 body lice collected on Day 16 post-infection were fixed with 100% methanol for 10 min and dried at room temperature. The samples were then incubated with 10 µL of mouse anti-B. quintana monoclonal antibody (IgG subclass1) at a dilution of 1:10 in 0.1 M PBS for 30 min at 25°C. This antibody was established by Tsuguo Sasaki (not published) by immunizing whole bacterial lysate to BALB/c mice and fusing with myeloma cells (SP2/0-Ag14). Hybridomas specifically reacting with cell surface antigens of B. quintana were selected. Fecal samples were washed 3 times with 0.1 M PBS by flash centrifugation for 10 sec and incubated with 100 µL of fluorescein isothiocyanate (FITC)-conjugated goat Fc specific anti-mouse IgG (Funakoshi Inc., Tokyo, Japan) at a dilution of 1:150 in 0.1 M PBS for 30 min. After 3 final washes with 0.1 M PBS by flash centrifugation for 10 sec and being mounted on slides with glycerol solution half-diluted by PBS, they were observed by fluorescence microscopy (BX51, Olympus, Tokyo, Japan). Images were recorded by CCD camera (QICAM Fast cooled mono, 12 bit, QImaging, Burnaby, Canada). A negative control sample from the feces of non-infected body lice was prepared in the same manner as for the infected sample.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). From Day 1 to Day 21 after infection, a total number of 60 lice were sampled and their DNA was extracted to estimate numbers of bacteria. The average number of B. quintana in 1 body louse decreased until Day 3 post-infection to a minimum 2 x 103 bacteria, then increased logarithmically until Day 17 to a maximum of 1.3 x 108 bacteria (Figure 2). The number of B. quintana increased 
60,000 times when compared with the minimum number of bacteria. The number of B. quintana excreted in feces increased gradually until Day 15 (Figure 2). The maximum bacterial excretion in the feces was observed on Day 15 at 10 x 106 bacteria per louse. Logistic regressions were calculated from the data on Days 3–17 (Figure 2) to determine the excretion dynamics of B. quintana in the feces as it is associated with the level of bacteria in body lice. The doubling time of bacteria calculated from logistic regression was 21.3 hrs, which corresponds to the characteristically slow growth of B. quintana on blood agar plates. The average amount of blood sucked by one louse was estimated as 2.4 µL by the weight method; thus the estimated number of B. quintana in the blood of the experimental infection was 8.58 x 103 cell/µL.
|
4.21 cells formed 1 colony. From this number after conversion and the results of other papers, the initial dose of B. quintana taken by louse was similar to the bacteremia level in patients with urban trench fever.13 By plating of dissolved feces from infected body lice, translucent small colonies were successfully formed on the blood agar on Day 14 post-plating. Three colonies were chosen for the second culture, and 17 days afterward the bacteria were harvested from blood agars by adding 4 mL of PBS for DNA extraction. DNA of bacteria from 3 plates were amplified by PCR using B. quintana-specific primer, described previously, and sequencing analysis of the amplified DNA products showed that all of them contained the identical sequence with B. quintana.
Electron-microscopic observation of feces from infected lice.
Scanning electron microscopy identified a massive number of short-rod bacteria (1 µm) scattered on the inner and outer surfaces of fecal masses (Figure 1C). In the enlarged view (Figure 1D), excreted bacteria were observed to be in a mesh-like structure entwined with each other, which is called a "biofilm." There were great differences in the structure of biofilm of B. quintana. In some parts of feces, immature bio-film structures were observed (Figure 1C, in the box), while some parts showed more mature biofilm structures of B. quintana (Figure 1D).
Microscopic observation of immunostained feces from infected lice.
Immunofluorescence staining of feces from infected lice using monoclonal antibodies against bacterial ly-sate of B. quintana demonstrated a strong positive reaction on the outer and inner surface of feces (Figure 1B). No specific fluorescence was observed in the fecal sample from non-infected lice prepared in the same manner. Both the immunologic data and electron-microscopic examination of the feces of infected lice (Figure 1C) confirmed infection of B. quintana.
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In this study, we also provide the findings of the mesh-like structure of B. quintana in the fecal masses of infected body lice. In our detailed examination by scanning electron microscopy, B. quintana attached to the exopolysaccharide (EPS)-like matrix and fecal materials (Figure 1D). It is generally considered that biofilms are characterized by an EPS matrix created by the bacteria, and this EPS matrix plays a vital role in the survival of bacteria,14 providing ecological advantages to the organisms. This biofilm-like structure may contribute to long survival of B. quintana in a fecal environment.2
The culture of B. quintana in the feces from infected body lice suggests that its infectivity remains in an fecal environment for a long time. B. quintana infection has been reported to be related to poverty, alcoholism, and homelessness in developed countries7; however, some studies showed a high prevalence of anti-B. quintana antibodies in the healthy population.6,15,16 Massive excretion of infective B. quintana in the feces of heavily lice-infested people may be the cause of accidental infection in crowded public places in developed countries. Although cross-reactivity of human antibodies to B. quintana is reported,17 it would be necessary to carry out more-detailed epidemiologic studies, particularly focusing on healthy people.
Experimental infections of louse-borne diseases have been performed by preliminary artificial membrane feeding18 or rectal inoculation of the pathogen into body lice.19 Currently a rabbit model is commonly used and has been adapted for not only louse-borne diseases20,21 but also for other pathogens. However, animal models are not suitable for quantification of bacterial proliferation, because after intravenous injection of the pathogens the number of bacteria may be diminished by the influence of the immune system of the host animal. When the B. quintana is injected into a rabbit, the bacteria may be rapidly cleared by cellular defense systems and the initial concentration cannot be estimated accurately. Our artificial membrane feeding system has great advantages for quantitative studies of bacterial proliferation in body lice and for potential adaptation to experimental infection by other pathogens.
Sites of proliferation in body lice are not fully understood, although Kostrzewski describes extracellular proliferation of B. quintana in the midgut of body lice.2 Further detailed observations would be needed to determine whether B. quintana proliferates in the epithelial cells of the louse intestine or in the lumen of the gut.
In this paper, we have provided the first quantitative description of B. quintana proliferation in body lice and excretion in the feces. We hope that our results provide some explanation for trench fever outbreaks among not only homeless populations but also ordinary people in developed countries.
Our results suggest that control of body lice may be an essential step toward urban trench fever epidemics among indigent people, and proper management of louse feces containing huge numbers of B. quintana, which seems able to survive for a long time, should be considered by medical staff and health workers in health and welfare sections of local governments to avoid accidental infection themselves by trench fever.
Received January 19, 2007. Accepted for publication June 2, 2007.
Acknowledgments: The authors thank for Prof. Didier Raoult for providing the Hassani strain of B. quintana for this study. We are also indebted to Dr. Roger S. Nasci of Centers for Disease Control and Prevention for his careful review of the manuscript and useful suggestions.
Financial support: This study was partially supported by a grant-in-aid of Ministry of Health, Labor and Welfare of Japan (H15-Shinkou-ippan-014 and H18-Shinkou-ippan-009).
* Address correspondence to Mutsuo Kobayashi, Department of Medical Entomology, National Institute of Infectious Diseases, Toyama 1-23-1, Shinjuku-ku, Tokyo 162-8640, Japan. E-mail: mutsuo{at}nih.go.jp 
Authors’ addresses: N. Seki, S. Kasai, O. Komagata, M. Mihara, T. Sasaki, T. Tomita, and M. Kobayashi, Department of Medical Entomology, National Institute of Infectious Diseases, Toyama 1-23-1, Shinjuku-ku, Tokyo 162-8640, Japan, Telephone: +81 (0)3 5285-1111, ex. 2400, Fax: +81 (0)3 5285-1178, E-mail: mutsuo{at}nih.go.jp. N. Saito, Laboratory of Electron Microscopy, National Institute of Infectious Diseases, Toyama 1-23-1, Shinjuku-ku, Tokyo, 162-8640, Japan. T. Sasaki, Department of Bacterial Pathogenesis and Infection Control, National Institute of Infectious Diseases, Gakuen 4-7-1, Musashi-Murayama City, Tokyo, 208-0011, Japan.
Reprint requests: Mutsuo Kobayashi, Department of Medical Entomology, National Institute of Infectious Diseases, Toyama 1-23-1, Shinjuku-ku, Tokyo 162-8640, Japan. Telephone: +81 (0)3 5285-1111, ex. 2400, Fax: +81 (0)3 5285-1178, E-mail: mutsuo{at}nih.go.jp.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) Average number of B. quintana per louse, with standard error bar. Regression lines were drawn based on the multiplication rate of bacteria from 4 to 17 days after infection in louse (solid line) and feces (dotted line); regression formulas for each line were calculated as y = 13.91e0.7956x (R2 = 0.9454) and y = 101.23e0.7899x (R2 = 0.9577), respectively.