2.2. General aspects of Chlamydiae

2.2.1. Taxonomy

Chlamydiae are obligate intracellular gram-negative bacteria. They were at first considered as viruses because of their unique biphasic intracellular life cycle (Grayston & Wang 1975). Thus far, four species of the genus Chlamydiae have been identified: C. trachomatis, C. pneumoniae, C. psittaci, and C. pecorum. The characteristics of the three main chlamydia species are presented in Table 1. Due to defects in energy metabolism, such as an inability to synthesise many major metabolites, including ATP (McClarty 1999), Chlamydiae are dependent on the host cell for their replication. The chlamydial organism is surrounded by an outer membrane which is mainly composed of a major outer membrane protein (MOMP) (Caldwell et al. 1981). The outer membrane is very similar to that of many other gram-negative organisms, but differs especially with regard to the amount of peptidoglycan, which Chlamydiae synthesise only in small amounts (Stephens et al. 1998). Therefore, beta-lactam antibiotics, such as penicillin, have no bactericidal activity against Chlamydiae. However, it seems that Chlamydiae contain some penicillin-binding proteins, because penicillin can interrupt the normal chlamydial development cycle (McClarty 1999).

C. trachomatis species have been divided into 18 serotypes or serovars (Wang & Grayston 1970, Grayston & Wang 1975, Wang et al. 1985), according to the Omp 1 gene that encodes MOMP. At the moment, typing of the Omp 1 gene is used to classify C. trachomatis strains. Further, C. trachomatis serotypes can be divided into three biovars: trachoma, lymphogranuloma venereum (LGV) and mouse pneumonitis (MoPn). The trachoma biovar includes the serotypes A, B, Ba, C, D, Da, E, F, G, H, I, Ia, J and K, the LGV biovar the serotypes L1, L2, L2a and L3, but MoPn is a single serovar type (Wang & Grayston 1970, Grayston & Wang 1975, Wang et al. 1985). The number of serotypes within the C. psittaci species is unknown, but it is genetically more heterogeneous than C. trachomatis (Grillner 1991). C. pecorum, which causes infections in ruminants, was established as a species distinct from C. psittaci (Fukushi & Hirai 1992). Recently, studies on the possible specific relationships with the Chlamydiae have highlighted the need for a assessment of the taxonomy of chlamydia-related organisms (Everett et al. 1999). This is currently under discussion.

So far, it seems that C. pneumoniae has only one serovar, with almost 100% DNA homology between the strains but less than 10 % homology with the other Chlamydiae (Kuo et al. 1995).

2.2.2. Biology of chlamydia

2.2.3. Pathogenetic mechanisms of chlamydial infections

Chlamydiae are responsible for a wide variety of human and animal infections and have a tendency to cause recurrent, persistent or chronic infections (Schachter 1989, Kuo et al. 1995, Ward 1995). However, there are no typical clinical symptoms or features which could differentiate chlamydial from other infections.

Chlamydial species and biovars infect different hosts and different cell types. C. trachomatis and C. pneumoniae strains are considered strictly human pathogens with no known animal reservoir (Schachter 1989, Kuo et al. 1995). However, it seems that C. pneumoniae of other mammals will be discovered and C. pneumoniae may be more diverse and widespread species than the current literature suggests (Girjes et al. 1994). C. psittaci infects a wide variety of avian and mammalian species, including humans. On the other hand, C. pecorum seems to be exclusively an animal pathogen (Schachter 1989, Fukushi & Hirai 1992). The trachoma biovars A-K of C. trachomatis infect primarily columnar but not squamous epithelium, causing mucosal infections, whereas, LGV strains of C. trachomatis and C. pneumoniae and C. psittaci are invasive (Schachter 1989). In vitro studies, C. pneumoniae has been shown to multiply in macrophages, vascular endothelium and smooth muscle cells, thus making its systemic dissemination through the circulation possible (Kaukoranta-Tolvanen et al. 1994, Godzik et al. 1995, Gaydos et al. 1996).

No long lasting protective immunity for chlamydia develops during acute infections. Thus, Chlamydiae may cause repeat infections (Grayston et al. 1985, Saikku 1992). Several studies have indicated that chlamydial infections may persist for a long time in humans. However, there is only little evidence about the clinical significance of the persistent chlamydia infections (Beatty et al. 1994b, Ward 1995). The persistence of the bacterium has also been established in vitro by interrupting the normal development cycle of the bacteria with antibiotics, nutrient-deficient conditions and immune-system regulating factors (Beatty et al. 1994b). It has been suggested that the immune response may be ineffective in eradicating chlamydia and protecting the host from reinfections. In humans and monkeys, for example, recurrent infection may cause an intense inflammatory reaction, while primary eye infections resolve with little or no residual tissue damage (Patton & Taylor 1986, Schachter 1989). Repeated or persistent infections, which provide an opportunity for long-term stimulation of the host with chlamydial antigens, result in tissue damage (Beatty et al. 1994a, Ward 1995).

The sequel of unsuspected silent infections associated with the genetically restricted host response are possibly due to Hsps. Repeated or prolonged exposure to Hsp antigens may cause a strong host response against bacterial Hsps and self-Hsps homologous to the bacterial ones (Peeling & Mabey 1999). Therefore, the immune response against these conserved sequences of the Hsps shared by the microbe and the host might potentially lead to an autoimmune reaction (Zugel & Kaufman 1999). Hsp60 seems to be the key antigen in chronic chlamydial infections. It is produced in chronic infections and has been associated with the hypersensitivity phenomenon and the immunopathology seen in these infections (Morrison et al. 1989, Peeling & Mabey 1999). Molecular mimicry between bacterial and viral proteins and endogenous molecules, such as the chlamydial Hsp60 antigen and its human homologue (Brunham & Peeling 1994, Paavonen et al. 1994), has been implicated to have a role in exacerbation ongoing autoimmune process (Peeling & Mabey 1999).

C. trachomatis has been shown to persist in an unculturable intracellular state, in which the synthesis of structural proteins is greatly reduced but the Hsp60 production actually increases (Beatty et al. 1993). Several studies have indicated that an enhanced immune reaction against C. trachomatis Hsp60 is more typically associated with chronic upper genital tract conditions, including ectopic pregnancy, chronic pelvic pain, perihepatitis, tubal factor infertility and fallopian tube damage than with acute infections of the lower genital tract (Peeling & Mabey 1999). The scarring tissue damage in trachoma is also connected to Hsp60 targeted immune responses (Peeling & Mabey 1999). Hsp10 reactivity may further contribute to the immunopathologic manifestations of severe upper genital tract complications of chlamydial disease in women (La Verda et al. 1999).

Chlamydial Hsp60 has been associated with the severity of the chronic sequelae of not only C. trachomatis infections but also other chlamydial infections. Hsp60 has been localized in human atheromatous tissue (Kol et al. 1998), and associated with the development of atherosclerosis (Xu et al. 1993). Recently, chlamydial infections and heart diseases were shown to be linked by antigenic mimicry (Bachmaier et al. 1999). Interestingly, an antibody against the chlamydial Hsp70 protein has been shown to neutralise chlamydia infectivity in vitro (Danilition et al. 1990) suggesting that the antibody is associated with protective immunity.

2.2.4. Clinical spectrum of C. trachomatis infection

C. trachomatis causes trachoma, which continues to be an important cause of blindness in some parts of the developing world. In these areas, poor economic conditions predominate and young children are frequently exposed to C. trachomatis, being the main reservoir of the organism. Primary C. trachomatis infection is generally regarded as benign and self limiting, though healing may take months (Beatty et al.1994b). Persistent and recurrent chlamydial infections leads to scarring of the conjuctiva, and disease severity seem to be associated with repeated C. trachomatis infections (Grayston et al. 1985, Bobo et al. 1997), which may ultimately cause blindness many years later (Beatty 1994b, Ward 1995). Certain highly virulent C. trachomatis serotypes (Bobo et al. 1997) may also be responsible for the more severe ocular forms of trachoma. Therefore, trachoma is considered a prototype of chronic chlamydial infection. It is the leading cause of preventable blindness in the world. C. trachomatis is also the most common cause of neonatal conjuctivitis and one of the most common causes of pneumonia in early infancy (Black 1997).

Although C. trachomatis still is an important ocular pathogen in the developing countries, C. trachomatis studies have mostly focused on sexually transmitted infections, since the same organism that causes trachoma is considered the world’s most common sexually transmitted bacterial pathogen. The World Health Organisation estimates that about 90 million of all new STD infections are caused by C. trachomatis. The highest rates are found in young, sexually active populations (WHO 1996). Compared to older females, young women often have cervical ectopy, where the squamocolumnar junction, a primary host target for C. trachomatis, is everted and thus more exposed to Chlamydiae.

Most infections caused by C. trachomatis in women are asymptomatic. However, clinical manifestations include cervicitis, urethritis, endometritis, pelvic inflammatory disease (PID), and abscess of the Bartholin´s glands (Stamm et al. 1999). The initial site of infection is usually the cervix, but the urethra and the rectum may also be infected (Stamm et al. 1980). Culture studies have shown that approximately half of the women with C. trachomatis are infected at both the cervix and the urethra, while one third have only cervical infections, and 5 to 30% have only urethral infections (Paavonen 1979, Paavonen et al. 1982, Phillips et al. 1987). Lower genital tract infections, urethritis and cervicitis, are completely asymptomatic or carry a wide range of symtoms (Paavonen 1979, Stamm et al. 1980, Cates & Wasserheit 1991, Horner et al. 1995). Most women with chlamydial cervicitis have minimal symptoms (Grayston & Wang 1975, Paavonen 1979, Cates & Wasserheit 1991).

The predominant C. trachomatis serotypes in urogenital tract infections are the serotypes D, E, and F (Wang et al. 1985, Saikku & Wang 1987, van Duynhoven et al. 1998). In women, serotype G has been associated with symptomatic infection (Lan et al. 1995) and the serotypes D and F with asymptomatic infection (Workowski et al. 1994, Lan et al. 1995). Serotype E has been found both symptomatic and asymptomatic women (Dean et al. 1995, Lan et al. 1995). Furthermore, Dean et al. (1998) have shown that almost all patients with repeat C. trachomatis infection are infected with uncommon C complex serotypes, suggesting that the C complex is associated with chronic or recurrent infections.

C. trachomatis infection may persist subclinically in the endometrium for a long time (Paavonen et al. 1985a,b) and produce chronic subclinical infection analogous to trachoma. The presence of plasma cells in the endometrial stroma [i.e. plasma cell endometritis (PCE)]; is characteristic to chronic endometritis (Greenwood & Moran 1981, Kiviat et al. 1990). C. trachomatis has been reported as a causative agent of PCE cases (Kiviat et al. 1986, Paavonen et al. 1987, Paukku et al. 1999) and also associated more often in severe PCE with lymphoid follicles (Paavonen et al. 1985b) than non-chlamydial endometritis.

Women with chlamydia isolated from the cervix often show no signs or symptoms of infection. On examination, however, at least a third generally have local signs of infection, such as endocervical bleeding, mucopurulent endocervical discharge, and edema within the area of ectopy (Paavonen et al. 1988). It has been reported that colposcopic features of immature squamous cell metaplasia of the cervix are associated with chlamydial infection (Paavonen et al. 1988). The number of polymorphonuclear leukocytes in cervical mucus also correlates with chlamydial infection of the cervix (Kiviat et al. 1985). Finally, patients with cervicitis caused by C. trachomatis are at risk for further development of PID (Paavonen et al. 1985a, Paavonen et al. 1987, Hillier et al. 1996).

PID has been defined as a syndrome associated with spreading of micro-organisms from the vagina and cervix to the endometrium, salpingeal tubes and adjacent structures (Weström 1980). Today, the majority of PID episodes of known aetiology are caused by C. trachomatis (Paavonen et al. 1987, Heinonen & Miettinen 1994, Paavonen & Lehtinen 1996, Paavonen 1998). A large proportion of C. trachomatis infections in the salpingeal tubes are asymptomatic, subclinical or atypical, and difficult to recognise as PID. With repeated infections, the risk of PID increases (Hillis et al. 1997).

In tubal infection, the fibrosis and scarring (Weström 1994, Paavonen & Lehtinen 1996) lead to permanent tubal damage, which increases the risk of ectopic pregnancy (Hillis et al. 1997) and tubal factor infertility (Weström 1980). The more episodes of PID one has had, the higher is the risk for infertility (Weström 1980, Paavonen & Lehtinen 1996).

Pregnant women with chlamydial infections are at an increased risk for adverse outcomes of pregnancy and postpartum endometritis (Smith & Taylor-Robinson 1993, Claman et al. 1995, Paavonen & Lehtinen 1996). C. trachomatis may persist in the upper genital tract for months or even years (Shepard & Jones 1989), and the persistent infection in the endometrium may cause repeated adverse pregnancy outcomes. C. psittaci infection is also suggested to cause abortion by inducing acute inflammatory response in the placenta (Roberts et al. 1967, Johnson et al. 1985, Flanagan et al. 1996).

LGV is a sexually transmitted systemic infection caused by C. trachomatis strains L1, L2 and L3. It is uncommon in industrialised countries but frequent in parts of Africa, Asia and South America. It predominantly infects lymphatic tissue (Schachter & Osoba 1983), but may also occur as an acute symptomatic infection without apparent lymph node involvement or tissue reaction at the point of infection (Perine & Stamm 1999). Acute LGV is reported over five times more frequent in men than in women (Schachter 1977). LGV has various acute and late manifestations. Most of the patients recover from LGV without late sequels. In some patients, however, the persistence of chlamydia in anogenital tissue may induce a chronic inflammatory response and may cause an anogenital syndrome with subacute manifestations, such as proctocolitis and hyperplasia of intestinal and perirectal lymphatic tissue. Perirectal abscesses, ichiorectal and rectovaginal fistulas, anal fistulas and rectal stricture or stenosis are chronic or late manifestations of LGV. Antibiotic treatment during the second stage of LGV, i.e. inguinal or anogenital syndrome, prevents the late complications of the disease (Perine & Stamm 1999).

2.2.5. Clinical spectrum of C. pneumoniae infection

C. pneumoniae infections occur world-wide both endemically and epidemically, and the prevalence varies from one region to another. C. pneumoniae is primarily a human respiratory pathogen (Saikku et al. 1985, Grayston et al. 1986, Ekman et al. 1993a, Kuo et al. 1995), and it is probably transmitted from person to person by respiratory secretions (Grayston et al. 1986, Grayston et al. 1990, Mordhorst et al. 1994). Transmission usually takes place outside home. Closed communities, such as military garrisons, schools and large families, have an important role in pneumonia outbreaks. The incubation time is around 3 weeks (Mordhorst et al. 1994). The infection spreads inefficiently, and perhaps only a few infected persons transmit the organisms. However, transmission may also occur through asymptomatic carriers (Kleemola et al. 1988).

C. pneumoniae infections occur yearly, however, cyclic variations has been shown in the incidence: two to three year periods of high incidence is followed by 3- to 10- year periods of low incidence (Grayston et al. 1990, Karvonen et al. 1993). In children in tropical countries, C. pneumoniae infection is more common and more severe (Saikku et al. 1988a) than in the developed countries, where very few patients under five years of age have serological evidence of past infection. The prevalence increases clearly after the age of 5 years, and approximately half of the population aged 20 years have antibodies against the organism. Unlike C. trachomatis, C. pneumoniae antibody prevalence is higher in males than in females (Saikku 1992). Seropositivity with C. pneumoniae antibodies continues to rise steadily in the population along with age, while in the case of C. trachomatis, antibody prevalence clearly falls after 40 to 50 years of age. This indicates that most people have two or three C. pneumoniae infections during their lifetime or alternatively a possible persistent C. pneumoniae infection (Grayston et al. 1990, Saikku 1992, Kuo et al. 1995). The C. pneumoniae antibody prevalence is also higher in smokers than among non-smokers (Karvonen et al. 1994).

The clinical features of C. pneumoniae infections are not typical. Apart from pneumonia, the most frequent illness associated with C. pneumoniae is bronchitis (Grayston & Wang 1975, Kuo et al. 1995). Most infections are subclinical (Kleemola et al. 1988), and clinical symptoms and pneumonia are more frequent in patients older than 20 years of age (Ekman et al. 1993a). Among teenagers and young adults, pneumonia or prolonged bronchitis is caused by a primary infection, and patients are often febrile and hoarse (Grayston et al. 1990, Kuo et al. 1995).

Respiratory illnesses caused by C. pneumoniae seldom require hospitalisation, as the infection is mostly relatively mild and patients usually respond to antimicrobial treatment. However, in patients with COPD (Kauppinen & Saikku 1995) and particularly in older people (Kauppinen et al. 1995, Peeling et al. 1997), C. pneumoniae pneumonia may be severe and complete recovery may be slow regardless of antibiotic therapy (Kuo et al. 1995, Peeling et al. 1997). Pharyngitis, laryngitis, sinusitis and otitis media are also caused by C. pneumoniae (Kuo et al. 1995).

Several chronic inflammatory diseases involving both respiratory and non-respiratory organs have been associated with C. pneumoniae infections. Chronic inflammatory conditions of the respiratory tract are logical consequences of C. pneumoniae infections. High prevalence of IgG antibodies and local sputum IgA antibodies to C. pneumoniae have been observed in patients with chronic bronchitis, suggesting a chronic respiratory C. pneumonie infection (von Hertzen et al. 1996). The onset of asthma and asthma exacerbations have also been proposed to occur in association with C. pneumoniae infection (Hahn & Allegra 1999). Smoking has been identified as the main risk factor for the development of chronic bronchitis and COPD; most of the patients are elderly male smokers. Serological studies have shown that C. pneumonie is involved in 4-5% of acute exacerbations of COPD (Blasi et al. 1993). Stable elevated IgA antibodies in sputum and frequent presence of circulating ICs, the markers of chronic C. pneumoniae infection, may reflect a defence mechanism mitigating airway inflammation (von Hertzen et al. 1997).

Even before the association of C. pneumoniae infection with chronic respiratory diseases was recognised, a connection between C. pneumonie infection and atherosclerosis was discovered (Saikku et al. 1988b). Saikku et al. showed (1988) that patients with acute myocardial infarction (AMI) and coronary heart disease (CHD) had more often elevated C. pneumoniae IgG and IgA antibody levels than healthy controls. Later, the same investigators showed that AMI and CHD patients had circulating ICs containing chlamydial LPS or C. pneumoniae protein-specific ICs present in their sera (Leinonen et al. 1990, Saikku 1992, Linnanmäki et al. 1993). So far, the serological association between C. pneumoniae and atherosclerosis has been confirmed in approximately thirty studies (reviewed by Saikku 1997, Campbell et al. 1998). In addition, the presence of C. pneumonie antigens and nucleic acid has been demonstrated in atherosclerotic lesions (Kuo et al. 1993, Kuo et al. 1995).

Sarcoidosis, hilar lympadenopathy and reactive arthritis (Kuo et al. 1995) have also been associated with C. pneumoniae infection.

2.2.6. Diagnosis of chlamydial infection