Amplification with molecular beacon primers and reverse line blotting for the detection and typing of human papillomaviruses.

A novel method for the detection and typing of human papillomavirus (HPV) was developed using molecular beacon primers. The method is based on the use of HPV-specific primers containing a hairpin loop structure in which fluorescent donor and quencher groups are held in close proximity such that fluorescence is quenched. Amplification of the target sequence results in the opening of the loop and the resulting fluorescence can be detected on a sequence detector system (SDS) 7700 (Applied Biosystems), as used for TaqMan assays. Fluorescent amplicons were identified on the SDS 7700 and then typed by a single hybridisation with specific probes immobilised in lines on a nylon membrane and detected on a fluorescent scanner. This novel beacon primer method compared well with conventional PCR for cervical scrape specimens. The combination of the beacon primer method and reverse line blotting should enable large-scale population studies of HPV infection.

Single-tube real-time nested polymerase chain reaction for detecting human papillomavirus DNA.

A single-tube real-time nested polymerase chain reaction (PCR) was developed to detect human Papillomavirus (HPV) DNA in a closed tube system. The oligonucleotide primers MY09/MY11 and GP5+/GP6+ were included in contiguous reactions, thus eliminating the need to transfer first round PCR product into a second tube. The sensitivity and specificity of the optimized single-tube nested PCR were comparable with that achieved by two separate reactions on a conventional thermal block system using serial dilutions derived from plasmids containing DNA of 20 HPV types. A minimum of 10 copies of HPV types 11 and 16 DNA could be detected by both systems. In clinical samples, HPV types 1A, 2, 3, 5, 6-8, 10, 11, 14, 16, 17, 18, 20, 31, 33, 35, 39, 45, 49, 50, 52-54, 57, 62, 66, 70, CP8304 and LVX82/MM7 could be detected by both PCR methods. A total of 145 samples collected from patients were tested for the presence of HPV DNA with the two PCR systems; 124 (86.1%) of 144 samples gave concordant results in both assays. The HPV DNA positive PCR amplicons were typed and concordant results were obtained in 47 of 67 positive samples tested in both amplicons. In samples containing multiple HPV types at least one type was common to both amplicons.

Detection and typing of human papillomavirus DNA in paired urine and cervical scrapes.

The prevalence of human papillomavirus (HPV) in paired cervical scrape and urine specimens from 144 women attending a clinic for genitourinary medicine was determined by polymerase chain reaction (PCR) and nested PCR, using degenerate and general primer pairs localized within the L1 region. HPV typing was by restriction fragment length polymorphism (RFLP), type-specific PCR (HPV 6, 11, 16, 18, 33), and partial DNA sequencing of PCR products. HPV DNA was detected in 114 (84%) women. HPV DNA was detected in the specimens of 58 patients after amplification with MY09/MY11 primers and in a further 54 patients after nested PCR with the GP5+/GP6+ primers. A total of 106/136 (78%) of women had HPV DNA positive cervical scrapes and 89 (65%) had HPV DNA positive urine specimens. Both the urine and cervical specimens of 81 women were positive. In 25 women HPV DNA was detected in the cervical specimen only, and in 8 women HPV DNA was detected in the urine specimens only. A total of 108 specimens from 75 patients were typed. For 33 patients HPV typing was achieved in both the cervical and the urine specimens and 19 women had identical types in paired specimens. Multiple HPV infections could be detected in 15 (20%) of 75 women where either the cervical and urine specimen or both of the specimens could be typed. More then one HPV type was found in 8 specimens and from multiple sites (cervix and urinary tract) in the same patients on 7 occasions. The results of this study indicate that the detection of HPVs in the urogenital tract can be maximised through the testing of both cervical scrapes and urine specimens in conjunction with the use of a nested PCR to increase the sensitivity of HPV DNA detection. Also, urine cannot be a direct substitute for a cervical scrape as different HPV types are often detected in the urine compared with those detected in the cervix.

Characterisation of non-capsulate Haemophilus influenzae by repetitive extragenic palindromic (REP)-PCR.

The use of repetitive extragenic palindromic (REP)-PCR to characterise non-capsulate Haemophilus influenzae (NCHI) for epidemiological studies was validated by application to four outbreak-associated and three epidemiologically unrelated NCHI strains of the same phenotype which had been well characterised by other methods. The REP-PCR patterns were reproducible and showed the unrelated isolates to be distinguishable from each other, whereas the outbreak-associated isolates were indistinguishable. The results were concordant with those from outer-membrane protein enriched profiles, ribotyping and randomly amplified polymorphic DNA analysis. When applied to six further isolates from two different suspected outbreaks, rapid results were obtained from boiled supernates prepared from one colony and indicated that the isolates in question were not related. REP-PCR provides a rapid method of strain characterisation suitable for NCHI, which is ideal for use in conjunction with other methods.

Genomic DNA digestion and ribotyping.

Bacteria may need to be characterized for a number of reasons. Newly discovered organisms are characterized to determine their taxonomic position; clinical isolates are characterized to provide an indication of pathogenic potential and likely antibiotic susceptibility. Such studies generally mvolve characterization to the genus and species level and provide an identification for the organism. Highly discrimmatory methods of intraspecies characterization are required for epidemiologic purposes in order to establish sources and routes of mfection to which control measures may be directed. For an mdlvidual patient, differentiation between relapsmg infection or reinfection may ard treatment choice and clinical management. The pattern of DNA fragments produced by digestion of genomic DNA with a restriction endonuclease, directly or after hybridization, can be highly discrimmatory and enable strain characterization, or typing, suitable for epidemiologic studies. The practice of obtaining these patterns (or profiles) for epidemiologic purposes is the subject of the present chapter, although the methods can also be applied to questions regarding identification at higher taxonomic levels.

Source of variation detected in ribotyping patterns of Haemophilus influenzae: comparison of traditional ribotyping, PCR-ribotyping and rDNA restriction analysis.

The pattern of EcoRI restriction fragments of chromosomal DNA that hybridize with a probe for genes encoding 16S and 23S rRNA is highly discriminatory for non-capsulate Haemophilus influenzae (NCHI). The source of variation detected by these probe-based ribotyping patterns was investigated by restriction analysis of rRNA operon (rrn) amplification products from nine representative strains. Digestion of rrn amplification products with EcoRI indicated one conserved EcoRI site within 16S rDNA and no EcoRI sites within the 16S-23S intergenic spacer region of the nine strains, and an EcoRI site at the 5' end of 23S rDNA from seven of the nine strains. Comparison of the EcoRI ribotyping patterns obtained with separate probes for 16S and 23S rDNA showed more variation with the 23S probe indicating variation in EcoRI sites downstream from the operon. Restriction analyses of 16S and 23S rDNA amplification products with AluI, HhaI, HaeIII and TaqI divided the nine 'traditional' ribotypes into a maximum of three and eight groups, respectively. Similar analyses of the 16S-23S intergenic regions (PCR-ribotyping) failed to distinguish any of the nine representative strains. Therefore, there is probably insufficient variation within the operon for it to form a good target for PCR-based typing methods. In contrast, 'traditional' ribotyping with cDNA from 16S plus 23S rRNA detects restriction site differences in the sequences flanking the operon, which show considerably more variation between strains. 'Traditional' ribotyping should therefore remain the standard for characterising NCHI in epidemiological investigations.

Pyrolysis mass spectrometry in epidemiological and population genetic studies of Haemophilus influenzae.

Haemophilus influenzae serotype b (Hib) vaccines have reduced the amount of invasive Hib disease in immunised infants. However, Hib disease remains in unvaccinated infants and adults and non-capsulate H. influenzae (NCHi) still causes infections, including outbreaks of respiratory disease. Characterisation of strains and the bacterial population as a whole is therefore necessary to detect outbreaks of infection with NCHi or changes in the population, for example, to vaccine-resistant clones of Hib. The rapid, simple and objective technique of pyrolysis mass spectrometry (PMS) was investigated as an alternative to current complex, subjective methods. PMS was compared with ribotyping and multilocus enzyme electrophoresis (MLEE) for population genetic analyses of Hib and with ribotyping and protein profiling for epidemiological analyses of NCHi. PMS clustered all the isolates of Hib together whereas MLEE and ribotyping distinguished certain clones - this is probably because the three methods examine different (and unrelated) characteristics of the organisms. The PMS results were essentially similar to those from ribotyping and protein profiling for the epidemiological analyses of outbreaks of NCHi disease. Therefore, PMS is probably unsuitable for comparisons of Hib populations but it is a useful addition to the arsenal of techniques for the characterisation of NCHi.

Epidemiological typing of coagulase-negative staphylococci from nosocomial infections.

Biotyping, antibiograms, bacteriophage typing, plasmid profile analysis and SDS-PAGE protein profiles were used to determine the relatedness of 44 Staphylococcus epidermidis and four S. haemolyticus isolates from 14 patients. A selection of these were further characterised by ribotyping. Biotyping classified the isolates into three major groups but was considered a poor strain marker. Although antibiograms classified the S. epidermidis isolates into 20 groups, some changes in the susceptibility patterns of related isolates from a single patient were demonstrated. Bacteriophage typing was the least discriminatory of the methods used. SDS-PAGE gave highly related patterns for the majority of S. epidermidis isolates. Plasmid profile analysis and ribotyping, with a minimum of two restriction endonucleases, were the most discriminatory methods for typing S. epidermidis. Nonetheless, some isolates from the same patient - probably representing a single strain - varied in plasmid profile indicating plasmid instability. One of six related isolates from a single patient lacked two bands from the ribotyping pattern of the other isolates. Although no single method proved entirely satisfactory on all occasions, the combination of typing methods was sufficient to provide evidence of the relatedness of S. epidermidis isolates from individual patients.

A molecular analysis of Greek and UK Haemophilus influenzae conjugative resistance plasmids.

Antibiotic resistance in Haemophilus influenzae has been associated with the presence of large, chromosomally integrated, conjugative plasmids. The plasmids of 10 beta-lactamase-positive, ampicillin-resistant strains, two from the UK and eight from Greece, were investigated. Plasmids were detected and isolated after transfer to a rec-deficient recipient. Purified whole plasmid was used as probe. In addition a 12 kb PstI fragment containing the putative point of recircularization in one plasmid, p1056, was cloned and used as a probe. All plasmids shared a high degree of sequence homology suggesting that plasmids of diverse geographical origin are highly related. All plasmids also shared sequence homology with the 12 kb PstI fragment containing the point of recircularization, suggesting that the sequences involved in excision and recircularization are conserved.

When should healthcare workers be screened for methicillin-resistant Staphylococcus aureus?

The role of screening of healthcare workers (HCWs) in the control of methicillin-resistant Staphylococcus aureus (MRSA) is controversial. It is recommended in guidelines by expert groups in both North America and the United Kingdom, although the role of MRSA carriage by HCWs in outbreaks is not clearly defined. The present report describes the spread of a distinct strain of MRSA to patients by a single HCW on three separate occasions over 27 months. The isolates from this HCW and patient contacts were shown to be indistinguishable by antibiogram and repetitive extragenic palindromic polymerase chain reaction (REP/PCR); none were typeable by lytic phage-typing. Throat carriage of the MRSA probably recurred in this HCW, despite attempts to eradicate it on three occasions. Over the same period, nine other small clusters were seen in the Oxford Hospital Group, involving 66 patients and 22 HCWs colonized, or occasionally infected, with a variety of MRSA strains. In none of these instances could HCWs be implicated in the initiation of an outbreak. The advantages of a screening policy include the determination of the full extent of MRSA-colonization and work exclusion; the disadvantages include detection of transient nasal carriage, disruption of staff routine and stigmatization. Screening of HCWs can be a valuable tool in the control of MRSA outbreaks but it should be used selectively. This strategy remains an important part of a control programme.

Analysis of the prevaccine population of noncapsulate Haemophilus influenzae and identification of a putative epidemic clone.

For six months prior to the introduction of Haemophilus influenzae serotype b vaccines, all noncapsulate Haemophilus influenzae received by our laboratory were characterised by biotyping, antibiogram, outer-membrane protein profiling, and ribotyping. Simpson's index of diversity (SID) showed the population was heterogeneous with multiple clones. The study identified a clone within noncapsulate Haemophilus influenzae biotype II that caused more disease than other strains. This clone was shown to have previously caused two outbreaks of respiratory disease and to possess a small extrachromosomal plasmid encoding ampicillin resistance. The study shows that describing the diversity within a bacterial population with SID may negate the need for retrospective subtyping comparisons.

The incidence and epidemiology of beta-lactam resistance in Haemophilus influenzae.

Consecutive isolates of Haemophilus influenzae were collected by the Public Health Laboratory, Bath between 1 June 1992 and 31 May 1993. Of 379 apparently distinct isolates, 216 originated from the respiratory tract, 102 from eyes and 61 from other sites. The minimum inhibitory concentrations of amoxycillin, amoxycillin/clavulanate, cefaclor, cefuroxime, cefotaxime and cefpodoxime were determined for each isolate. Forty strains (10.6%) were beta-lactamase producers. MIC50 and MIC90 values and the range of MICs were determined for all isolates. The overall resistance rates were: amoxycillin (MIC > 1.0 mg/L), 22.7%; amoxycillin/clavulanate (MIC > 1.0 mg/L), 14.8%; cefuroxime (MIC > 1.0 mg/L), 18.5%, (MIC > 4.0 mg/L), 5.5%; cefaclor (MIC > 8 mg/L), 15.6%; cefpodoxime (MIC > 1.0 mg/L), 0.3%; cefotaxime (MIC > 1.0 mg/L), 0%. Twenty non-beta-lactamase producing but beta-lactam resistant strains (cefuroxime MIC > 4.0 mg/L) were matched with 20 susceptible strains on the basis of patient age, sex, and specimen type. The strains were characterised by outer-membrane protein (OMP), random amplified polymorphic DNA (RAPD) and ribotyping patterns. Eleven of the 20 resistant strains were indistinguishable by the methods used, suggesting spread of a single beta-lactam resistant, non-beta-lactamase producing clone. The distribution of resistant strains within the local community was plotted geographically.

Molecular epidemiology of a multiple strain outbreak of methicillin-resistant Staphylococcus aureus amongst patients and staff.

Since methicillin-resistant Staphylococcus aureus (MRSA) isolates are not endemic in our hospital, which is a tertiary referral centre, the finding of 13 MRSA isolates from 12 patients associated with an acute vascular surgery ward between October 1993 and December 1993 prompted further epidemiological and laboratory investigations. Two strains were distinguished by antibiogram and phage-typing. One strain, resembling EMRSA-16, colonized six patients and was probably introduced from another hospital in the Oxford Region. Five other patients were colonized by a second strain, gentamicin-resistant and non-typable by phage-typing, probably introduced into the hospital 12 months previously by a patient from Nairobi, Kenya. A 12th patient was colonized by both strains simultaneously. Of 46 staff members screened three were colonized--one by an EMRSA-16 strain, a second by the gentamicin-resistant 'Nairobi'-strain and a third member carried yet a further distinct MRSA strain. The healthcare worker colonized by the 'Nairobi'-strain had been carrying the isolate 12 months previously and was the likely source of this strain. These isolates were also characterized by the repetitive extragenic palindromic-polymerase chain reaction (REP-PCR), a novel PCR-based methodology which has not been previously used in characterizing Staphylococcus aureus in an outbreak. This method corroborated the strain classifications provided by the traditional methods, confirming that there had been spread of two strains simultaneously. Our study demonstrates that multiple strains of MRSA may circulate amongst patients and staff during an outbreak, patients may be colonized by more than one strain simultaneously and long-term staff carriage (> 12 months) may be an important source of colonization in patients. REP-PCR is a rapid and effective molecular typing method for MRSA.

Haemophilus influenzae: then and now.

Haemophilus influenzae has long been recognised as a major cause of serious infection and mortality in children less than 5 years old. Prior to the introduction of Haemophilus influenzae type b (Hib) immunisation, the incidence of a child suffering an invasive Haemophilus infection was 20-50/100,000 in industrialised countries and up to ten times higher in developing regions. The introduction of a Hib vaccine programme results in a rapid and dramatic decline in the incidence of Hib infection in the susceptible childhood population. For example, within two years of the introduction of routine Hib vaccination of infants in the UK, the risk of serious Hib infection had fallen from 1:600 to 1:30,000 by 5 years of age. Many other European countries have introduced, or are in the process of introducing, a routine Hib immunisation programme. Because the epidemiology of Haemophilus influenzae infection is changing so dramatically, it is opportune to review Haemophilus influenzae as it was perceived in the pre-vaccine era (the past) and during vaccine implementation (the present), and how its role may change in the post-vaccination era (the future). This review will summarise the historical landmarks that have led to our present-day understanding of Haemophilus influenzae pathogenicity, the concerns about antibiotic resistance, the features of the host immune response to Haemophilus influenzae, and the introduction of the Hib vaccine. Furthermore, the possible importance of this organism in the future will be discussed.

Characterization of Streptococcus pneumoniae from human immunodeficiency virus--seropositive patients with acute and recurrent pneumonia.

Thirty-two isolates of clinically significant Streptococcus pneumoniae from 11 human immunodeficiency virus (HIV)-seropositive patients with single or multiple episodes of pneumonia were characterized by antibiotic susceptibility testing, serotyping, ribotyping, and repetitive extragenic palindromic polymerase chain reaction (REP-PCR). The isolates comprised 10 serotypes, 12 ribotyping patterns, and 12 REP-PCR patterns. There was close but not absolute correlation between techniques. By combining these characterization methods, 14 strains were identified. Five strains were found in > 1 patient, suggesting their frequent occurrence in this population. Two isolates of different serotype from 1 patient were highly related by ribotyping and REP-PCR, suggesting possible in vivo serotype change. Acute infection was associated with single strains or coinfection by distinct strains. Recurrent pneumonia was identified as relapse with the same strain or reinfection with new strains. The molecular characterization of pneumonococci from HIV-seropositive persons refines our understanding of pneumonococcal infection in these patients.

Development of a ribotyping scheme for Haemophilus influenzae type b.

Ribotyping and outer-membrane protein subtyping were used to characterise 283 consecutive isolates of Haemophilus influenzae type b. These isolates were obtained primarily from patients with invasive disease in the UK and were received by the Public Health Laboratory Service Haemophilus Reference Laboratory prior to the implementation of Haemophilus influenzae serotype b vaccine in the UK. A subtyping scheme using the ribotyping method is suggested. Twenty-two ribotypes are described, 14 of which were found amongst the 283 clinical isolates characterised in this study. In contrast, only four outer-membrane protein subtypes were found amongst the 283 isolates. The ribotyping profiles were further used to estimate the relatedness of isolates. The resulting dendrogram suggested a population genetic structure different from that previously described for Haemophilus influenzae type b using multi-locus enzyme electrophoresis. This study shows the value of ribotyping as a subtyping method for epidemiological studies of Haemophilus influenzae type b. However, the further use of ribotyping for population genetic structure analysis of Haemophilus influenzae type b may be misleading and therefore inappropriate.

Farm animals as a putative reservoir for vancomycin-resistant enterococcal infection in man.

Using a highly selective enrichment broth, 62 isolates of vancomycin-resistant Enterococcus faecium were obtained from non-human sources; 35 isolates from raw sewage, 22 from farm animals and 5 from uncooked chickens. All strains possessed the Van A gene, conferring high-level resistance to vancomycin (MIC > or = 256 mg/L). Ribotyping of 42 of these isolates resulted in 14 distinguishable patterns. Two ribotyping patterns were found among isolates from animals and sewage and those from clinical sources. A blood and a urine isolate from separate hospital patients and porcine isolates shared the same ribotyping pattern number 6 and a stool isolate from a patient in the community and sewage isolates shared another pattern, number 10. This finding suggests that animals may serve as a reservoir of vancomycin-resistant enterococci (VRE), which may enter the human food chain. The emergence of VRE in hospital patients may reflect selection of these organisms in the hospital environment by antibiotic usage from which nosocomial spread might occur.

Faecal carriage and nosocomial spread of vancomycin-resistant Enterococcus faecium.

Eight clinical isolates of vancomycin-resistant Enterococcus faecium (VRE) were obtained from four renal and four other in-patients within an 11 week period during 1992. Characterisation of the isolates by restriction enzyme analysis with Sal I and rRNA gene restriction patterns (ribotyping) showed them to be clonally related. During the next 3 months an additional 14 VRE were isolated from hospital patients, nine of which were indistinguishable by ribotyping from the strain associated with the outbreak. An epidemiological survey was instigated in order to determine the level of carriage of this VRE. A total of 354 stool specimens was screened using a highly selective enrichment broth. VRE were detected in the stools of 11/73 (15%) of renal patients, 5/97 (5%) of other hospital patients and 3/184 (2%) of patients based in the community. Of the 25 stool isolates that were further characterised by ribotyping, 17 were indistinguishable from the outbreak strain. The remaining eight isolates gave seven different patterns. Patients harbouring the outbreak strain stayed in hospital significantly longer and had received more antibiotic treatment, for longer, than those patients from whom other VRE had been isolated. There was no significant difference in vancomycin or cephalosporin usage between the two groups of patients. Ribotyping showed there to be a number of clones of VRE carried by patients and that one of these clones was especially prevalent and has been responsible for the outbreak of infection in the renal unit. The technique also showed the presence of different VRE in general practice patients suggesting they are not just a hospital phenomenon.

Polymerase chain reaction-based strain characterization of noncapsulate Haemophilus influenzae.

A polymerase chain reaction-based typing method for noncapsulate Haemophilus influenzae was developed. Randomly amplified polymorphic DNA fingerprints were generated from boiled supernatants prepared directly from bacterial colonies without the need for DNA extraction. The technique was applied to isolates obtained during putative outbreaks of chest infection and validated by comparison with sodium dodecyl sulfatepolyacrylamide gel electrophoresis analysis of outer membrane protein-enriched preparations and rRNA gene restriction analysis. There was complete concordance between the three techniques. The results show that randomly amplified polymorphic DNA analysis provides a highly discriminatory method of characterizing strains of noncapsulate H. influenzae which is eminently suitable as an epidemiological tool for the rapid investigation of outbreaks of infection.