Whooping cough in Pakistan: Bordetella pertussis vs Bordetella parapertussis in 2005-2009

Pertussis, or whooping cough, is an acute respiratory disease mainly affecting infants and children and is caused by Bordetella pertussis and Bordetella parapertussis. The aim of this study was to investigate the share of Bordetella species from potential whooping cough cases during 2005-2009. Eight hundred and two samples from suspected pertussis cases were collected, mainly from 2 provinces of Pakistan. Bacterial culture, identification, DNA extraction and routinely used polymerase chain reaction (PCR) methods using IS1001, IS1002 and IS481 were used to identify the Bordetella species. The results were unexpected, because all of the isolates collected from the different cities were identified as B. parapertussis (7.4%); B. pertussis was not isolated from any sample. However, PCR results indicated the presence of a small percentage (0.6%) of B. pertussis among the total cases studied. This study suggests that vaccines to protect against both B. pertussis and B. parapertussis should be considered.     

Expression of pertussis toxin correlates with pathogenesis in Bordetella species

Pertussis toxin is a principal determinant of virulence produced by Bordetella pertussis in the disease whooping cough and is the primary toxinogenic component of the pertussis vaccine. This toxin is not produced by the closely related species Bordetella parapertussis. Toxinogenic strains of B. parapertussis were constructed by the conjugative introduction of cloned pertussis toxin genes. These and analogous toxinogenic and nontoxinogenic strains of B. pertussis were assayed for their toxicity and reactogenic activities. Expression of active toxin by either Bordetella sp. correlated with the induction of leukocytosis, anaphylaxis, and histamine sensitivity.     

Evidence of efficacy of the Lederle/Takeda acellular pertussis component diphtheria and tetanus toxoids and pertussis vaccine but not the Lederle whole-cell component diphtheria and tetanus toxoids and pertussis vaccine against Bordetella parapertussis infection.

A subanalysis of a recent cohort efficacy trial of a pertussis vaccine was performed to determine its efficacy against cough illnesses due to Bordetella parapertussis infections. Infants received four doses of either the Lederle/Takeda acellular pertussis component diphtheria and tetanus toxoids and pertussis (DTaP) vaccine or the Lederle whole-cell component diphtheria and tetanus toxoids and pertussis (DTP) vaccine at 3, 4.5, 6, and 15-18 months of age; controls received three doses of diphtheria and tetanus toxoids (DT) vaccine only. All subjects were prospectively followed for cough illnesses of > or = 7 days' duration; cases of B. parapertussis infection were confirmed by positive culture, household contact, or serology. Seventy-six cough illnesses due to B. parapertussis were identified; 24 occurred in 929 DTaP recipients, 37 in 937 DTP recipients, and 15 in 321 DT recipients, resulting in an efficacy of 50% for DTaP vaccine (95% CI [confidence interval], 5% to 74%) and 21% for DTP vaccine (95% CI, -45% to 56%). The data in the present analysis suggest that the Lederle/Takeda DTaP vaccine but not the Lederle whole-cell component DTP vaccine has efficacy against B. parapertussis infection.     

Patterns of Bordetella parapertussis respiratory illnesses: 2008-2010

Clinical specimens from 9 states during 2008-2010 were tested by PCR for Bordetella pertussis and Bordetella parapertussis. Of the positive samples, 13.99% were identified as B. parapertussis. It was concluded that B. parapertussis infections are more common than previously realized and contribute to cases thought to be vaccine failures.     

Parapertussis and pertussis: differences and similarities in incidence, clinical course, and antibody responses.

OBJECTIVES: To compare the incidence, clinical course, and serologic response to Bordetella antigens in patients with parapertussis and pertussis. DESIGN: Two studies were performed in Sweden during the 1990s, when pertussis vaccines were used only in clinical trials. Study I was a retrospective study of patients with positive Bordetella cultures obtained in clinical routine, and study II involved an active search for patients with Bordetella infections during a placebo-controlled trial of a pertussis toxoid vaccine. RESULTS: Study I includes 58, and study II 23 patients with parapertussis. In study I, the incidence of parapertussis was 0.016 cases per 100 person years in children 0 to 6 years old and 0 in older children and adults. In study II, the incidence rates of parapertussis and pertussis were 0.2 and 16.2 per 100 person years, respectively, in children followed from 3 months to 3 years of age. The median number of days with cough was 21 in parapertussis and 59 in pertussis. The proportions of children with whooping and vomiting were lower in parapertussis than in pertussis. Geometric mean serum filamentous hemagglutinin IgG increased from 6 to 63, and pertactin IgG from 4 to 12 units/mL in parapertussis patients, which was similar to increases in children with pertussis. CONCLUSIONS: Disease caused by Bordetella parapertussis is diagnosed less commonly and is milder and of shorter duration than disease caused by Bordetella pertussis. Parapertussis induced serum IgG against filamentous hemagglutinin and pertactin of similar magnitude as does pertussis, and did not induce serum IgG against pertussis toxin.     

Domestic animals as carriers of Bordetella species in Senegal

Despite intense efforts to maintain a high level of vaccine coverage against human whooping cough, rural senegalese areas are still endemic for Bordetella pertussis. One explanation being the potential existence of animal reservoirs, the objective of this work was to precise the carriage by domestic animals of bacteria belonging to the genus Bordetella in Senegal. Bacteriological samples (swabs and aspirates) were obtained from various domestic animals living in different parts of the country. No B. pertussis nor B. parapertussis were isolated. However, for the first time to our knowledge, B. bronchiseptica was identified from small ruminants located in Africa. The positive animals were two goats and two sheep from Dakar slaughterhouse together with a goat living in a rural compound. The fact that it was identified in goats and sheep underlines the potential zoonotic of that bacterial species in countries where small ruminants are of economical and cultural relevance.     

A multiplex PCR assay for the detection of respiratory bacteriae in nasopharyngeal smears from children with acute respiratory disease

To elucidate the frequency of infections with pathogenic respiratory bacteriae during an inter-epidemic period a multiplex PCR assay was used to screen nasopharyngeal smears for the presence of DNA specific for Bordetella pertussis, Bordetella parapertussis, Chlamydophila pneumoniae and Mycoplasma pneumoniae. 187 samples from children aged 2-14 y were analysed with this method in addition to classical bacteriology and compared to results obtained with commercially available PCR kits for each single parameter. From 82 samples positive by bacteriology, 8 (4.3%) were also positive by PCR, whereas from 105 negative samples, 12 (6.4%) were positive only by PCR. From the total of 20 samples positive by PCR, 4 were found to be positive for M. pneumoniae, 6 for B. pertussis, 3 for B. parapertussis and 7 for both B. pertussis and B. parapertussis. Multiplex PCR is a very useful approach for the diagnosis of bacterial infections not detectable by classical bacteriology. In some patients, PCR was the only method giving a positive result, and in others double infections were diagnosed only because of the PCR contribution. Combination of classical bacteriology with multiplex PCR allows a precise diagnosis of infections in the upper respiratory tract, resulting in a more effective therapy.     

Infection with Bordetella parapertussis but not Bordetella pertussis causes pertussis-like disease in older pigs

The 3 major Bordetella species--namely, B. pertussis, B. parapertussis, and B. bronchiseptica--can be distinguished by their different host ranges. B. bronchiseptica infects a wide range of mammals (including humans), whereas B. pertussis infects only humans and, under experimental conditions, mice and pigs. In contrast, B. parapertussis, also a causative agent of pertussis, displays a unique host specificity with 2 subgroups, one infecting only humans and the other infecting only sheep. Here, we show that both strains of B. parapertussis also infect older piglets when delivered intrapulmonarily. Infected piglets displayed mild fever and respiratory symptoms, such as coughing and breathing difficulties. Importantly, transmission was observed between infected and noninfected piglets. In tracheal organ cultures, adherence to ciliated epithelial cells was observed. Furthermore, both strains of B. parapertussis displayed higher resistance than B. pertussis to neutralization by porcine beta-defensin 1 in the respiratory tract, which has been demonstrated to be associated with protection against B. pertussis disease in older pigs. The development of this new model will assist us in better understanding the pathogenesis of this disease and in the development of more-effective vaccines against pertussis.     

Frequency of serological evidence of Bordetella infections and mixed infections with other respiratory pathogens in university students with cough illnesses.

Banked acute-phase and convalescent-phase serum samples from a previous study of respiratory illness in university students were examined for significant (>/=2-fold) increases in ELISA titers of IgA and IgG antibody to Bordetella pertussis filamentous hemagglutinin, pertactin, and fimbriae-2 and >/=4-fold titer increases to agglutinogens by agglutination. ELISA titers of antibody to pertussis toxin could not be determined because of technical problems. Chlamydia pneumoniae infections were diagnosed by culture or by a >/=4-fold increase in immunofluorescence assay titer or a single high titer (>/=512). Mycoplasma pneumoniae, influenza A and B, adenovirus, and respiratory syncytial virus infections were diagnosed by >/=4-fold increases in complement fixation titer or a single high titer (>/=64). There were 319 subjects with cough of >/=5 days' duration, and of these, 47 (15%) had significant increases in antibody to B. pertussis antigens; 26 (8%) had significant increases to fimbriae-2 or agglutinogens, indicative of B. pertussis infection, and 2 (1%) had evidence of non-B. pertussis bordetella infections. Seventeen (36%) had evidence of mixed infections or cross-reacting antibodies (influenza B infections, 5; adenovirus infections, 4; influenza A infections, 3; C. pneumoniae infections, 3; and M. pneumoniae infections, 2). Our findings suggest that bordetella infections are common in young adults with cough illnesses (incidence, 9%), and a surprising number of these are mixed infections with other respiratory pathogens.     

Rapid detection of Bordetella pertussis and Bordetella parapertussis in clinical and molecular proficiency panel specimens with a novel intercalating dye-based PCR assay

A novel rapid internally-controlled duplex polymerase chain reaction (PCR) assay for Bordetella pertussis and Bordetella parapertussis that uses a new intercalating dye, LCGreen, for amplicon detection was designed and evaluated using clinical specimens and the 2009 Quality Control for Molecular Diagnostics (QCMD) molecular proficiency testing panel. The sensitivity and specificity of the new PCR assay were equal to that of reference standard uniplex probe-based assays. The assay can be performed in 1 h including DNA extraction. The new assay thus offers a simple and rapid alternative for the detection of B. pertussis and B. parapertussis.     

Serologic diagnosis of whooping cough by an enzyme-linked immunosorbent assay using fimbrial hemagglutinin as antigen

Diagnosis of whooping cough by an enzyme-linked immunosorbent assay (ELISA) that measures serum IgG, IgM, and IgA antibody to the fimbrial hemagglutinin of Bordetella pertussis was compared to isolation from nasopharyngeal swabs in a prospective study. Of 77 patients with upper respiratory tract infections of unknown etiology in which B. pertussis infection could not be excluded on clinical grounds, 26 were culture-positive, including one for Bordetella parapertussis. All 26 patients were positive by ELISA except one asymptomatic erythromycin-treated patient (ELISA sensitivity, 96%). Among culture-negative patients, 24 additional patients were positive by ELISA. Thus, only about one half of the patients with whooping cough were identified by culture under optimal conditions. Positive titers of IgM antibody and/or high titers of IgG antibody in the first serum sample, allowing for a rapid diagnosis, were found only in 26 (53%) of 49 serologically positive patients. By combining culture testing with serology, rapid diagnosis was obtained in 41 (82%) of 50 patients.     

Neutralizing antibodies to pertussis toxin in whooping cough

The development and duration of neutralizing antibodies (antitoxin) to pertussis toxin were studied in 38 patients with culture-verified infections due to Bordetella pertussis and one patient with infection due to Bordetella parapertussis. An in vitro neutralization test in microplate culture of Chinese hamster ovary cells was used. An antitoxin response was recorded in 36 patients, the exceptions being two patients treated early with erythromycin (one of whom developed clinical pertussis two years later) and the patient with infection due to B. parapertussis. A long-term follow-up for several months to several years after disease showed maintenance of high antitoxin levels. These results are in accordance with the hypothesis that antibodies to pertussis toxin mediate long-term immunity to whooping cough.     

Bordetella pertussis and chronic cough in adults.

To evaluate Bordetella pertussis as a cause of persistent cough in adults, we examined 201 patients who had a cough for 2-12 weeks and no pulmonary disease. We obtained the following at presentation: medical history, chest radiograph, respiratory function measurement, nasopharyngeal aspirate for polymerase chain reaction (PCR), nasopharyngeal swab specimen for culture, and a blood sample (acute serum). Four weeks later a second blood sample (convalescent serum) was obtained. Control sera were obtained from 164 age-matched healthy blood donors with no history of cough during the previous 12 weeks. Four patients were B. pertussis culture-positive; 11 (including the culture-positive patients) were B. pertussis PCR-positive; and 33, including 10 of the 11 PCR-positive patients, had serological evidence of recent B. pertussis infection. Pertussis-positive and -negative patients could not be discriminated by a history of cough. We conclude that B. pertussis infection is a common cause of persistent cough in adults. This is of concern, because these patients may be B. pertussis reservoirs from which transmission may occur to infants, in whom the disease can be devastating.     

Serological diagnosis of pertussis: IgM, IgA and IgG antibodies against Bordetella pertussis measured by enzyme-linked immunosorbent assay (ELISA)

Igm, IgA and IgG antibodies against Bordetella pertussis were measured by enzyme-linked immunosorbent assay (ELISA) with an ultrasonicate of formalin-killed bacteria (a mixture of strains 1, 2 and 1, 2, 3) as antigen and disposable polystyrene 9-cuvette blocks as the solid phase. The specificity properties of the assay were assessed by an inhibition technique. Of the microbes tested, only B. parapertussis was able to cause a significant inhibition. In addition, IgM and IgA antibodies against B. pertussis were only found in some sporadic cases of respiratory infections caused by other microbes. Sera, nasal swabs and cough plates were received from 198 patients with suspected whooping-cough. ELISA determinations were mostly made from only one serum sample of each patient. Paired sera were studied only from the culture-positive infants under 3 months of age. The number of positive cultures was highest in group under 3 months of age (41%), where the frequency of positive ELISA was lowest (20%). The use of paired sera strikingly increased the number of ELISA-positive individuals in this youngest patient group. In later life, the relationship between these tests changed: isolation was positive in only about 10% of the patients, whereas 29-64% yielded positive titres in ELISA. This study shows that pertussis ELISA is a valuable aid in the rapid diagnosis of pertussis, particularly of the atypical forms of the disease which mostly are culture-negative.     

Evaluation of serology and nasopharyngeal cultures for diagnosis of pertussis in a vaccine efficacy trial

Nasopharyngeal cultures and titer rises in paired sera were evaluated in a placebo-controlled pertussis vaccine efficacy trial. IgG ELISA for filamentous hemagglutinin (FHA) identified 30 (88%) of 34 placebo recipients and 33 (89%) of 37 vaccine recipients with culture-verified Bordetella pertussis infections, whereas IgG ELISA for pertussis toxin (PT) showed higher diagnostic sensitivity in the placebo group than in the vaccine groups. The CHO cell assay did not improve sensitivity. Children with Bordetella parapertussis infections had rises of titers of antibody to FHA of the same magnitude as children with B. pertussis infections. Sensitive serologic criteria, based on the intraassay variations, identified 105 additional culture-negative cases with significant titer rises in paired sera. IgG ELISA for FHA and PT and IgA ELISA for FHA were reliable assays, and bacterial isolation rates were lower in vaccine recipients than in placebo recipients with serologically defined pertussis.     

A search for Bordetella pertussis infection in university students.

University students with persistent cough of greater than or equal to 6 days' duration were evaluated for evidence of infection with Bordetella pertussis. Of 130 students studied during a 30-month period, 34 (26%) were found to have evidence of recent infections with B. pertussis. Infection was identified by direct fluorescent antibody assay of a nasopharyngeal specimen in one student and serologically in 33 additional subjects. B. pertussis was not recovered on culture of nasopharyngeal specimens from any subjects. Students with B. pertussis infection were identified in seven of the eight 3-month periods in which students were enrolled during the 30-month investigation, suggesting an endemic rather than epidemic pattern of infection in this university population. Illnesses of students with pertussis were similar to the illnesses of students without pertussis. The findings in this study suggest that adult populations in which endemic illness occurs at a relatively constant rate may be the reservoirs for pertussis outbreaks in susceptible children. Immunization programs in the future will need to employ booster doses for adults if complete control of B. pertussis infection is our goal.     

Bordetella pertussis in a four‐time kidney transplant recipient: A call for immunization programs at transplant centers

Pertussis, or whooping cough, is a highly contagious respiratory illness caused most frequently by Bordetella pertussis. Clinical presentation ranges in severity, but life‐threatening illness disproportionately affects children and immunocompromised individuals. Acellular vaccines for pertussis have been available for decades, and they are recommended throughout the lifespan. A patient who had received a kidney transplant presented with respiratory distress and dry cough as manifestations of co‐infection with B pertussis and Bordetella parapertussis/bronchiseptica. The goal of this case report was to highlight the importance of immunization programs at transplant centers, which are in the unique position to care for patients both with end‐stage organ disease and in the post‐transplant setting.     

Structural variability and originality of the Bordetella endotoxins

Structural studies of Bordetella endotoxins (LPSs) have revealed remarkable differences: (i) between their LPSs and those of other bacterial pathogens; (ii) among the LPSs of the seven identified Bordetella species; and (iii) among the LPSs of some Bordetella strains. The lipid As have the "classical" bisphosphorylated diglucosamine backbone but tend to have fewer and species-specific fatty acid components compared to those of other genera. Nevertheless, three strains of B. bronchiseptica have at least three different fatty acid distributions; however, the recently identified B. hinzii and B. trematum LPSs had identical lipid A structures. The B. pertussis core is a dodecasaccharide multi-branched structure bearing amino and carboxylic groups. Another unusual feature is the presence of free amino sugars in the central core region and a complex distal trisaccharide unit containing five amino groups of which four are acetylated and one is methylated. The B. pertussis LPS does not have O-chains and that of B. trematum had only a single O-unit, unlike the LPSs of all the other species of the smooth-type. The O-chain-free cores of non-B. pertussis LPSs were always built on the B. pertussis core model but most were species-specifically incomplete. The LPS structures of three B. bronchiseptica strains were found to be different from each other. The O-chains of B. bronchiseptica and B. parapertussis were almost identical and had some features in common with B. hinzii O-chain. Serological analyses are consistent with the determined LPS structures.     

Frequency of pertussis in children with prolongued cough

To determine the frequency of pertussis in children < or = 16 y who had prolonged cough (> or = 14 d), a prospective study was conducted at an outpatient clinic of a paediatric hospital. Nasopharyngeal swabs were taken for culture and nucleic acid testing by polymerase chain reaction (PCR) for Bordetella pertussis. Immunoglobulin A and immunoglobulin G antibodies against pertussis toxin (PT) were tested by ELISA in paired serum samples. A total of 148 patients were recruited during 1 y. Pertussis was detected in 25 (16.9%) patients with at least 1 of the tests. PCR was positive in 12 patients, and 9 cases was diagnosed serologically. Both PCR and serology were positive in 4 children. Duration of cough was longer in the patients with pertussis (median 33 vs 20, p = 0.03). Seropositivity of pertussis toxin was higher in pertussis negative patients during enrollment (24% vs 65%, p = 0.005). From the results of this study, B. pertussis seems to be common in our population despite high immunization rates with whole cell vaccine. Although the duration of cough is defined as longer than 21 d in some studies for pertussis case definition criteria, it was shorter than this in 3 of our cases.     

Bordetella pertussis binds human C1 esterase inhibitor during the virulent phase, to evade complement-mediated killing

C1 esterase inhibitor (C1inh) is a major inhibitor of several pathways of inflammation in humans. In this study, we show that virulent-phase cultures of Bordetella pertussis, the etiological agent for whooping cough, but not other Bordetella species specifically recruit C1inh from human serum. Using a spontaneous mutant of B. pertussis that was deficient in C1inh binding, we demonstrate that the ability of B. pertussis to acquire high levels of human C1inh and wild-type levels of serum resistance are well correlated, suggesting that, in addition to and independent of BrkA expression, acquisition of C1inh is vital to B. pertussis resistance to complement-mediated killing.