Declining population sizes of koalas (Phascolarctos cinereus) in SE Queensland (QLD), Australia can partially be attributed to chlamydiosis, with the majority of epidemiological studies focusing on the prevalence of infection and associated pathology in female koalas, with lesser attention given to males. We aimed to explore the epidemiology of Chlamydia pecorum infection in the male urogenital tract from wild (hospitalized and free-ranging) koalas in SE QLD. Although 67% of male koalas were infected with C. pecorum in their urogenital tract and 55% were shedding the organism in their semen, only a third of the males sampled presented with overt signs of urogenital disease. Infection with C. pecorum was lower in populations from rural locations, compared with periurban locations, with a corresponding low association between urogenital infection and clinical disease. The presence of C. pecorum in penile urethral swabs was a good predictor of the presence of C. pecorum in semen, with a significant correlation (P=0.006) in 58% of males. In contrast, the C. pecorum load in penile urethral swabs was not a good predictor of the C. pecorum load in semen, with no significant correlation. In addition, 57% of male koalas had large numbers of bacterial copy numbers in the penile urethra (upper quartile) and 40% shedding into semen with no overt signs of disease. Investigation of the association of C. pecorum infection, body condition score, and age revealed that the highest incidence of urogenital infection occurred in males with the lowest body score (1 out of 10). Furthermore, 63% of sexually mature male koalas (>2 yr old) had urethral infections and 50% had C. pecorum in their semen. Our study suggested that the role of chlamydia in male koala infertility has been previously underestimated.
The serious decline of free-ranging koala (Phascolarctos cinereus) populations located along the eastern coast of Australia over the past few decades has been attributed to a combination of natural and anthropogenic factors, but of major concern to the survival of this iconic species is the pathology associated with chlamydiosis. Chlamydial infection, predominantly Chlamydia pecorum, is considered to be widespread in most koala populations, with prevalence rates varying with the population studied and the method of detection used. Chlamydial infection is the main etiology of ocular, urogenital, and reproductive disease in koalas and commonly manifests clinically as keratoconjunctivitis leading to blindness, urinary tract inflammation resulting in urinary incontinence and cystitis (identified by wetness and brown staining of fur around the rump area), and reproductive tract lesions (Blanshard and Bodley 2008; Wan et al. 2011; Polkinghorne et al. 2013; Nyari et al. 2017). Various studies have revealed that approximately 50% of Chlamydia-infected koalas in SE Queensland (QLD) are subclinical, with no overt clinical signs of ocular or urogenital disease (Wan et al. 2011; Polkinghorne et al. 2013; Nyari et al. 2017). In the case of reproductive disease, there are often no overt clinical signs denoting the presence of disease, but Chlamydiarelated pathology may render the koala infertile (McColl et al. 1984; Blanshard and Bodley 2008; Johnston et al. 2015).
Although it is now well established that chlamydial disease renders females infertile (Jackson et al. 1999; Blanshard and Bodley 2008; Loader 2010; Robbins et al. 2018), by contrast there has been limited research dedicated to the effects of Chlamydia on male koala fertility. Deif (2011) was the first to provide a comprehensive review of studies that examined the effect of Chlamydia on the reproductive tract in female and male koalas and clearly demonstrated that there were three times the number of studies on females compared with males. Investigation into pathogenesis of chlamydial genital disease in the female and its impact on fertility has been well documented in the literature, despite the male being instrumental in venereal transmission of the disease. Brown and Grice (1986) were the first to report the adverse effects of Chlamydia on male koalas when they demonstrated that penile urethral inoculation of Chlamydia pscittaci into a male koala resulted in a purulent urethritis and later, the death of the animal that exhibited clinical signs of cystitis. Further, Canfield (1989) demonstrated that chlamydial infection present in the urinary tract of male koalas was associated with cystitis and occasionally prostatitis; Hemsley and Canfield (1997) later confirmed these findings.
The primary method of diagnosis of urogenital chlamydial disease is via veterinary clinical assessment of koalas presented to wildlife hospitals. Veterinary assessment includes a combination of sonographic examination of the urogenital tract; cytological assessment of a urinary sediment sample collected via cystocentesis; visual assessment of overt clinical signs to provide a score (0–3) of cystitis (Wan et al. 2011); and collection of conjunctival, urogenital sinus, and cloacal or rectal swabs for PCR analysis. Swab collection and veterinary examination are only indicators for detection of infection but do not always assist in detection of lesions linked to chronic sequelae within the lower and upper reproductive tract and resultant reproductive disease.
Detection of chlamydia by quantitative real-time PCR (qPCR) not only facilitates sensitive species-specific detection of chlamydia, but also allows determination of the copy number of the bacterium. Establishment of this technique has given researchers a more informative means of investigating animals that do not exhibit clinical signs of chlamydiosis. This is important in epidemiological studies to establish the prevalence of chlamydia infection within koala populations to potentially correlate these findings to the physical and health status of the population. In an effort to better understand chlamydial disease in wild koala populations in eastern Australia, Markey et al. (2007) and Wan et al. (2011) investigated the association between chlamydial load using real-time qPCR and disease severity. Despite the relatively low number of samples analyzed, both studies concluded that the range of infectious load varied considerably between animals and reflected the severity of clinical disease in koalas with clinical signs.
Previous epidemiological studies of chlamydia have focused on rural free-ranging wild and periurban koala populations (White and Timms 1994; Jackson et al. 1999; Nyari et al. 2017), periurban being defined as areas on the urban periphery into which cities expand or influence (Buxton et al. 2011). These studies have revealed a significant incidence of chlamydial infection in koala populations of SE QLD, with prevalence rates ranging between 0% and 85% within subpopulations (Devereaux et al. 2003; Polkinghorne et al. 2013; Nyari et al. 2017), as well as prevalence of chlamydial disease of up to 88% in southern Australian populations (Patterson et al. 2015; Legione et al. 2016; Speight et al. 2016). In the majority of such studies, samples were obtained by swabbing of the urogenital tract or conjunctiva of live koalas. However, penile urethral swabs may not be a reliable indicator of infection prevalence within the upper reproductive tract and reproductive lesions in males are more challenging to sonographically identify than those in females (Loader 2010). Although the use of ultrasound is noninvasive and provides an index of urogenital and reproductive tract pathology (Larkin et al. 2018), it is typically not helpful for assessing whether the infection is active. Semen collection, on the other hand, not only enables detection of chlamydial DNA in the upper reproductive tract but also allows an assessment of the effect of inflammation and pathology of the reproductive tract on semen quality and therefore, potentially, male fertility.
To better assess prevalence and understand the epidemiology of chlamydia infection on male reproductive disease in wild koala populations in SE QLD, we investigated whether detection of chlamydia in semen provided further information on the prevalence of disease that was not detected by swabbing of the urogenital tract alone. In addition, we investigated whether the presence of C. pecorum in penile urethral swabs was a reliable indicator of the presence of the bacterium in the semen, whether bacterial load in semen was predictive of bacterial load in urethral swabs, and whether either measure was associated with clinical severity grade. Furthermore, we examined possible relationships between C. pecorum infection prevalence as detected by penile urethral swabs and semen with respect to age class and body condition score (BCS).
MATERIALS AND METHODS
Animals and sample collection
Swabs of the penile urethra were collected from 250 koalas over a 4-yr period from both free-ranging hospitalized (n=211) and nonhospitalized (n=39) wild koalas using cotton-tipped swabs (COPAN Diagnostics, Murrieta, California, USA) and stored at –20 C. The hospitalized koalas were presented to Moggill Koala Hospital (Brisbane, QLD), Endeavour Veterinary Ecology (Brisbane, QLD), and Currumbin Wildlife Hospital (Gold Coast, QLD) for veterinary treatment because of trauma, ill thrift, poor body condition, or exhibiting clinical signs of disease. Samples were collected under the Department of Environment Scientific Purposes permit WISP16751015 and University of Queensland Animal Ethics permit AE02305. The nonhospitalized koalas were from wild populations located at St Bees Island (20°55′0″S, 149°25′59″E), Mackay (21°48′54″S, 150°22′32″E), and Mount Byron (27°6′55″S, 152°40′23″E) and were swabbed as part of the University of Queensland research project KRG024—Pathways of Pathogen Transmission in the Koala (University of Queensland Ethics approval CMLR/304/13/QLD GOVT and CMLR/091/12/ARC/RIO TINTO, Central Queensland University Animal Ethics approval A72/04-282), and Scientific Purposes permits (WISP16162915 and WISP15517315).
Koala semen was collected by means of electroejaculation (Johnston et al. 1994) from 109 sexually mature (>2 yr old) free-ranging hospitalized koalas as part of the University of Queensland research project KRG005—The Pathology, Incidence, Treatment and Management of Chlamydiosis in the Male Koala (University of Queensland Animal Ethics approval AE02305) and Scientific Purposes permit WISP16751015. Semen samples for PCR analysis were immediately frozen in liquid nitrogen (–196 C) postcollection. All animals were maintained under inhalant anaesthesia (Vogelnest 2008) while penile urethral swabs and semen samples were collected.
Koala age was estimated using the tooth-wear class system described by Gordon (1991). Body condition score was assessed by palpation of the musculature over the scapulae using a scoring system of 1 to 10 (Ellis and Carrick 1992), with 1 representing poor and 10 representing excellent body condition. Clinical signs of urogenital chlamydial disease were scored on the basis of intensity of urine scalding of fur around the rump (Fig. 1) and described as grade 0: normal fur; grade 1: slight discoloration of fur around rump, evidence of mild fresh urine leakage; grade 2: stained greasy fur covering the rump; and grade 3: stained, greasy, wet matted fur around a large area of the rump (Griffith 2010).
Isolation of DNA and real-time qPCR
Swab samples were removed from storage and 200 µL of 1× Tris–ethylenediaminetetraacetic acid buffer was added to each swab and vortexed vigorously for 30 min. We extracted DNA using MagJET genomic DNA kit (Thermo Scientific, Melbourne, Victoria, Australia) followed by purification of DNA with KingFisher Flex (Thermo Scientific) as per the manufacturer's instructions. Semen samples were removed from storage and DNA was similarly extracted from 100 µL of sample and purified. The DNA template was assayed using two multiplex real-time PCR panels (Hulse et al. 2018). Chlamydia pecorum bacterial load was quantified using a standard curve derived from DNA obtained from quantified reference C. pecorum strain culture concentrations, and the assay limit of detection was determined at 74 infectious forming units per milliliter (Hulse et al. 2018).
Chlamydia incidence within populations was analyzed by means of odds ratio (OR; 95% confidence interval) and uncorrected chi-square statistic (P<0.05) to quantitatively compare the strength of association of the presence or absence of overt signs of chlamydial disease with qPCR detection of C. pecorum in the male koala using EpiTools epidemiological calculators (Sergeant 2018). Interpretation of the OR with respect to association between C. pecorum detection and correlation to disease was interpreted using the following criteria: 1) if OR=1 then C. pecorum presence was not associated with disease; 2) if OR>1 then C. pecorum presence was strongly associated with disease; and 3) if OR<1 there was a minimal association between C. pecorum presence and disease (Sergeant and Perkins 2015). Analysis of the association between detection of C. pecorum in koala semen via sampling with a penile urethral swab, in addition to comparing chlamydial load, with respect to clinical sign grades was done using a Pearson's chi-square test of association (P<0.05) in IBM SPSS Statistics, version 24 (IBM Corp., Armonk, New York, USA). The interquartile range (Q1–Q3) of C. pecorum copy number, determined via qPCR, was calculated separately for penile urethral swabs and semen samples. Each quartile was then further partitioned into clinical severity grade. Data were only used from koalas where there was matching penile urethral swabs and semen samples with corresponding clinical sign disease grade recorded (n=79). The relationship between C. pecorum infection, via detection with penile urethral swabs and semen, with age class and body score was compared using a paired t-test (P<0.05).
Prevalence of infection and disease
Chlamydia pecorum DNA was detected in 67% of penile urethral swabs and 55% of semen samples (Tables 1, 2). For both penile urethral swabs and semen samples, the highest prevalence was identified in periurban locations such as Redlands and Moreton Bay, whereas the lowest prevalence was in rural areas. For urethral swabs, there was a strong association between C. pecorum infection and clinical signs of chlamydial disease in all locations except the Lockyer Valley, Scenic Rim, and Somerset locations (Table 1). For semen samples, there was a strong association between C. pecorum infection and the presence of clinical signs of chlamydial disease for all shires except Redlands and Toowoomba (Table 2).
Relationship to clinical signs of cystitis
Figure 2 illustrates the relationship between the clinical grade of urogenital disease and the interquartile ranges of C. pecorum copy number in penile urethral swabs and semen, respectively. Table 3 shows the C. pecorum copy number (measured by qPCR) interquartile range thresholds for penile urethral swabs and semen. Of the 250 penile urethral swabs, 62% (155/250) were PCR positive and of the 109 semen samples, 59% (65/109) were PCR positive. Of male koalas exhibiting no clinical signs of urogenital disease (grade 0), 57% (69/122) had detectable C. pecorum DNA in their penile urethra and 40% (21/53) had detectable C. pecorum in semen.
Relationship between the presence of C. pecorum in the penile urethral swab and semen
Investigation into whether a C. pecorum PCR-positive penile urethral swab sample was a predictor of the presence of Chlamydia in the semen revealed a significant association within each clinical grade; Chlamydia was detected in the semen by sampling with a urogenital swab (χ2=10.1, P=0.006). In 58% of the male koalas sampled, Chlamydia was present in both the penile urethra and semen (Table 4). In contrast, there was no significant association indicating that chlamydial load will be equivalent between koala semen and urogenital swabs (χ2=8.08, P=0.089), suggesting that quantification of bacterial load in penile urethral swabs is not a reliable indicator of Chlamydia load in the corresponding semen sample. With the exception of samples collected from koalas with a clinical sign grade of 2 or 3, the C. pecorum load was consistently greater in the penile urethral swab when compared with the bacterial load in the semen (Table 5).
Prevalence of urogenital C. pecorum infections as a function of age and BCS
There was a significant difference between the prevalence of C. pecorum infection via detection with penile urethral swabs and semen with respect to age class (n=207) as determined by a paired t-test (t9=2.924, P=0.017; Fig. 3A), and there was higher prevalence of infection in males estimated to be >4 yr old. Although there was no significant difference in the prevalence of C. pecorum infection between swabs and semen with respect to BCS (t10=1.801, P=0.102; n=242; Fig. 3B), there was a higher prevalence of infection detected by both penile urethral swabs and semen in koalas with BCSs less than 3. The prevalence decreased as BCS increased, with the exception of male koalas with a score of 10, which had a high prevalence of infection in penile swabs but no detection of C. pecorum in their semen samples.
Koalas are known to be infected with two chlamydial species, C. pecorum and Chlamydia pneumoniae, with C. pecorum being the more pathogenic of the two species (Jackson et al. 1999; Devereaux et al. 2003). Although the focus of our study was C. pecorum, we did not detect C. pneumoniae in penile urethral swabs or semen of koalas in this study. Chlamydial disease has been suggested as a key agent of population decline of koalas in SE QLD (Rhodes et al. 2011), although the extent to which the impact of disease on koala mortality and fecundity has exacerbated declines at a population level remains controversial (McCallum et al. 2018). Gonzalez-Astudillo et al. (2017) investigated causes of morbidity and mortality of koalas in SE QLD and found that, over a 16-yr time frame, 52% (n=21,619) of koalas were admitted to wildlife hospitals with signs of chlamydiosis. Cystitis was the most common sign (26.8%), followed by conjunctivitis (17.2%), bursitis (13.5%), pneumonia (12.3%), nephritis (12%), and metritis (5%). Interestingly, male-specific chlamydialike signs such as prostatitis and orchitis were diagnosed very minimally, at only 0.04% and 0.01% respectively, potentially indicating underdiagnosis of male reproductive disease (Gonzalez-Astudillo et al. 2017).
Of the male koalas that we examined, 67% had urethral infections with C. pecorum and 55% had evidence of shedding the bacterium in their semen. This supported the findings of Jackson et al. (1999) and Bodetti et al. (2002) that Chlamydia is highly likely to be a sexually-transmitted disease. Interestingly, only a third of the males that we sampled exhibited clinical signs of disease despite infection being present in both the urogenital tract and semen. This strongly suggested that male koalas are likely to act as subclinical carriers of infection and highlighted the likelihood of a significant underestimation of the prevalence of infection within wild populations (Nyari et al. 2017).
Our study also revealed that C. pecorum infection prevalence was lower in populations from rural areas compared with periurban populations, and the association between urogenital infection and clinical disease was weaker in rural populations. However, this may have been due to sampling bias of hospitalized koalas where the prevalence of C. pecorum was overrepresented compared with wild healthy koalas located in rural areas. Significant association between infection and clinical disease in male koalas of periurban populations supported the proposition that urbanization and habitat loss have imposed chronic stress factors on koala populations that may influence expression of disease (Weigler et al. 1988; McCallum et al. 2018). McCallum et al. (2018) suggest that stress increases susceptibility to and reduces tolerance of infection, although there are currently no conclusive studies directly comparing stress and susceptibility to chlamydial infection in koala populations located in areas where there has been human encroachment on habitat. Further investigation into whether there is an indirect or direct correlation between stress and either resistance to chlamydial infection or tolerance of infection without clinical disease is warranted.
One of our most significant findings was that 57% of male koalas had large numbers of bacterial copy numbers in the upper quartile of the penile urethra and 40% were shedding chlamydial bodies into the semen with no overt signs of disease. This may have been due to sampling of males while they were in the early stages of infection, or individual tolerance of periodic or recurrent high levels of infection without developing pathology. Others also found high levels of urogenital shedding, as detected by swabs, in the absence of clinical disease (Wan et al. 2011; Polkinghorne et al. 2013). Male koalas shedding high loads of chlamydia into their penile urethra and semen with no clinical signs of disease are likely to be carriers and thereby capable of transmitting the organism into the female reproductive tract. This has disease management implications and highlights the importance of appropriate diagnostic protocols, such as qPCR, to identify infectious animals, especially those that might be carriers of the infectious organism but that present as subclinical.
We presented evidence that if C. pecorum is detected in the penile urethra by means of a swab, then it is likely that C. pecorum will also be present in the semen. Semen traversed the penile urethra before sample collection and there was a chance of contamination of the semen by organisms present in the penile urethral cells. The major sites of C. pecorum infection in male koalas are the prostate and bulbourethral glands, with 63.5% of male koalas that were admitted into a SE QLD wildlife hospital having prostatitis and 16.6% bulbourethral gland pathology (Palmieri et al. 2019). Prostatic secretions represent the bulk of seminal plasma and are essential for sperm survival and motility in the female reproductive tract (Elzanaty et al. 2002), whereas the bulbourethral gland secretions assist in forming the seminal plug (Johnston and Holt 2014). We showed the presence of chlamydia in koala semen and the lower reproductive tract, and consideration needs to be given to the suggestion that chlamydiosis is an ascending infection of the male reproductive tract (Deif 2011; Mackern-Oberti et al. 2013). The source of chlamydia in the semen may be the testis or epididymis, although chlamydia were detected in the majority of regions of the lower and upper reproductive tracts (Palmieri et al. 2019), which indicates that there are multiple sites within the tract that could be the source of chlamydia.
When the relationship between age (estimated from tooth wear) and C. pecorum infection was investigated, 40% of sexually immature males (<1 yr old) were infected, as detected from the urethral swab. Transmission of infection in this age cohort is likely to occur during birth, through pap feeding, or during residency in the mother's pouch (Jackson et al. 1999; Polkinghorne et al. 2013; Nyari et al. 2017; Russell et al. 2018). In addition, 63% of sexually mature male koalas (>2 yr old) had urethral infections and 50% had C. pecorum in their semen, results potentially with major implications for fertility at a population level. Regardless of whether a male has reached sexual maturity by 2 yr of age, mating may not begin until 4 yr of age (Martin and Handasyde 1990), by which time C. pecorum urogenital infection would be well established within the reproductive tract. Our results showed an increase of infection (in both urethral swabs and semen) in males estimated to be 4 to 5 yr old and suggested that, as males become more sexually active with age, there is an increased level of infection within the urogenital tract and in semen.
We found no significant differences between the prevalences of infection of urogenital swabs, semen, and body score. The highest incidence of urogenital infection occurred in males with the lowest body score of 1; these are koalas that are considered emaciated, which is in line with the observations from Obendorf (1983) that progressive wasting or emaciation in a koala can be due to chlamydiosis; this might also suggest that either chlamydia infection results in low body condition or that it could be due to stress causing poor condition, in turn predisposing the koala to chlamydial disease. Although the prevalence of C. pecorum urogenital infection decreased as BCS increased, male koalas with a score of >9 showed a distinct difference in C. pecorum infection between swabs and semen. Chlamydia pecorum was not detected in semen but there was a high prevalence in swabs, which may have indicated that male koalas in excellent condition could be carriers of C. pecorum or possibly that the infection was present only in a persistent form and was not being actively shed. Further studies are required for confirmation.
The high incidence of chlamydia detected in the penile urethra and semen in our study highlighted the need to further investigate the effect of chlamydia on sperm quality and its possible contribution to male infertility. In vivo and in vitro (inoculation) studies of chlamydia infection of human spermatozoa show an adverse effect on fertility by causing premature apoptosis of sperm due to coincubation with chlamydia lipopolysaccharide, increase in tyrosine phosphorylation causing premature capacitation of sperm, and increased incidence of sperm DNA fragmentation (Satta et al. 2006; Gallegos et al. 2008; Eley and Pacey 2011). That we found chlamydial DNA in the penile urethra and semen of male koalas with no clinical signs of disease suggested that the role of chlamydia in infertility has been underestimated. Further studies are required to establish whether chlamydia shed in semen from infected males (with and without clinical signs of the disease) are capable of transmitting the infection to females.
We acknowledge core project funding from the Queensland State Government Koala Research Grant: The Pathology, Incidence, Treatment and Management of Chlamydiosis in the Male Koala (project KRG005, Department of Environment and Heritage Protection).