Efficacy of a synthetic peptide Chlamydia pecorum major outer membrane protein vaccine in a wild koala (Phascolarctos cinereus) population

Chlamydiosis is a significant disease affecting Eastern Australian koala (Phascolarctos cinereus) populations, impacting individual animal welfare and fecundity and therefore influencing population dynamics. The aim of this study was to investigate the effect of a synthetic peptide vaccine based on 4 components of the Chlamydia pecorum major outer membrane protein (MOMP), over an 18-month period in a koala population severely impacted by chlamydiosis. Wild koalas were recruited into a vaccination or a placebo treatment group on a random allocation, then followed through a period of 18 months, with recapture at 6 monthly intervals. Vaccination did not alter clinical disease expression or chlamydial shedding from the ocular or urogenital sites. Vaccination did not stimulate a significant plasma anti-MOMP IgG response, when compared to the placebo group. There was no significant effect of vaccination on IFN-γ and IL-17A mRNA expression of peripheral blood lymphocytes when stimulated with rMOMP. We have demonstrated that a synthetic peptide vaccination against chlamydiosis is not an effective management tool in a koala population with a high prevalence of C. pecorum infection and related disease. The lack of antigenic response found in this study suggests that further research utilising a larger, full-length antigen is an avenue worth investigation if we are to consider vaccination as a part of a management strategy in diseased koala populations.

Cohort recruitment.Koalas recruited to this study were captured within a 15 km radius from − 31.159855,150.118456, south-west of the town of Gunnedah, New South Wales, using the 'noose and flag' technique 33 .Alfaxalone (Jurox), was used to sedate the koalas at a dose rate of 1.8 mg/kg by intramuscular injection and then oxygen and isoflurane was administered to effect through a fitted mask to anaesthetise the koala.The same veterinarian (SJS) assessed and sampled all koalas.At initial assessment, koalas were classified as either diseased or non-diseased, based on external signs of chlamydiosis, (presence of a "wet bottom" or ocular disease) 34 .These criteria were chosen as these clinical signs can improve with veterinary treatment 35 and infection status of individuals was unknown at initial capture.Following this classification, the koala was randomly assigned to one of two treatment groups, vaccination or placebo.The veterinarian who examined the koalas and administered the treatment was blinded to treatment group allocation.VHF collars (transmitter model M3420, ATS Australia)  were fitted to the koalas so they could be radio-tracked and visually monitored from a distance at 2-month intervals throughout the study, and recaptured and clinically evaluated at 6, 12 and 18 months following vaccination.

Koala clinical examination.
The following procedures were performed at timepoint 0 (recruitment), 6, 12 and 18 months post vaccination.Digital images were taken of the koala's identification tag, eyes, teeth (for age estimation) 36 and perineal region.Koalas were weighed and a body condition score (BCS) ranging from 1 to 5 was recorded 37 .Both eyes were assessed for ocular discharge, conjunctival proliferation, and conjunctival chemosis, each criterion with a score between 0 and 3 allocated to each eye 34 .The perineal region was examined for urine leakage, erythema, or the presence of purulent discharge.A wet bottom score was allocated, based on a modification of the Griffith 34 scoring system, whereby only scores 0-6 were used because scores 7-9 required prolonged observation, not amenable to field examinations.Ultrasonography was performed with the koala in dorsal recumbency [38][39][40] .For ultrasonographic examination of the kidneys, a small patch of fur was clipped on the abdomen ventral to the approximate site of each kidney.Both kidneys were identified and the presence of any structural abnormalities such as dilated renal pelvis, dilated ureters or presence of renal calculi were recorded.For females, the transducer was placed inside the pouch and female reproductive structures visualised by scanning the region cranio-lateral to the bladder.If paraovarian cysts were identified, it was noted if the cyst/s were unilateral, bilateral, multiloculated or singular.A paraovarian cyst score was allocated with 0 = no abnormalities, 1 = unilateral paraovarian cysts identified and 2 = bilateral paraovarian cysts identified.If paraovarian cysts were not detected, a thorough scan of the caudal abdomen was performed to confirm their absence.In males, ultrasound was performed to identify the prostate by directing the transducer caudo-ventral from the bladder.Ultrasound was also performed on both testes.
Swabs for detection of C. pecorum DNA were collected from the conjunctiva bilaterally and the female urogenital sinus or male urethra (Copan, Interpath, 160C).For conjunctival sampling, swabs were placed into the conjunctival fornix and rotated five times.In females, swabs were inserted 2-3 cm into the urogenital sinus and rotated five times.In males, the penis was everted, and the swab placed 2 cm into the urethra and rotated five times.All swabs were stored at − 20 °C until processing.P1 H-EGMSGDPCDPCATW-OH, P2 H-INYHEWQVGAALSYRLNMLIP-OH, utilised previously 18 , P3 H-VLQIVSLQINKLKSRKACG-OH, P4 H-KKLLKSAFLSAAFFAG-OH, representative of the fifth conserved region and the leader sequence of MOMP, respectively.All peptides were synthetised by Mimotopes (Melbourne, Australia), at a purity of > 70% determined by high pressure liquid chromatography.Two of these peptides were chosen as they have been shown to stimulate humoral and cell mediated immune responses in healthy koalas following vaccination 18 .The additional two peptides were selected as they represent identified B-cell vaccine specific epitopes of conserved regions within C. pecorum MOMP that have been recognised by koalas, following MOMP antigen vaccination 25 .Kyte-Doolittle plots were constructed to determine hydrophobic and hydrophilic regions of each four peptides.The antigen was combined with a three-component adjuvant, containing Poly I:C (250 μg), Host Defence Peptide-Innate Defence Regulator IDR-1002 (500 μg), and Polyphosphazene EP3 (250 μg) (VIDO-Intervac, University of Saskatchewan, Canada).Vaccine preparation involved mixing the Poly I:C and IDR-1002 and incubating for 15 min at room temperature, with gentle rocking.Next, EP3 and the four peptides were added, with the final mixture incubated for a further 15 min at room temperature, with gentle rocking, then stored at 4 °C until use 25 .Vaccination and placebo injections (Hartmann's solution) were administered subcutaneously (0.5 ml) into the interscapular region at first capture and then again at 6 months.
Urogenital and ocular swab DNA extraction and qPCR.DNA was extracted from ocular and urogenital swabs using the MagMAX™ CORE Nucleic Acid Purification Kit #A32702.A sterile swab was included as a control to monitor for contamination during the extraction process.A multiplex qPCR assay was used to quantify the following genes from each extracted swab sample, P. cinereus beta actin (HEX), 23S Chlamydia (ROX) and C. pecorum ompB (FAM); primers adapted from Hulse et al. 41  Plasma anti-major outer membrane protein (MOMP) IgG ELISA.All concentrations of reagents, blocking and incubation conditions were determined by optimisation experiments using chequerboard titrations.In the optimised assay, 96 well flat bottom plates (Grenier #0030125150) were coated with 2 µg of recombinant MOMP serovar G (produced utilising methods previously described) 14 in carbonate-bicarbonate buffer (Sigma-Aldrich #C3041) at 100 µl/well at 4 °C overnight.Wells were emptied and then blocked with 300 µl of 5% skim milk powder in PBS with Tween (0.05%) (PBST) at 37 °C for 1 h.Wells were emptied and koala plasma was applied to the wells at a concentration of 1:400 at 100 µl/well and incubated at 37 °C for 1 h.Wells were then emptied and washed five times with PBST with a microplate washer (Biorad model 1575).An in-house sheep anti-koala-IgG antibody 18 100 µl was added to each well at a dilution of 1:8000 and incubated at 37 °C for 1 h.Following incubation, wells were washed as described above.100 µl of rabbit anti-sheep IgG HRP conjugated (Abcam #ab6747) was added to each well (1:20,000 dilution) and incubated at 37 °C for 1 h.Following incubation, wells were washed as described above, using PBS only.Finally, 100 µl of 3,3′,5,5′Tetramethylbenzidine (Sigma-Aldrich #T5525) was added to each well and allowed to develop at room temperature, in the dark, for 20 min before 100 µl of 1 M H 2 SO 4 was added to each well.Optical density (OD) was read at 450 nm.Each plate contained a blank well (no antigen or plasma), the same negative control from a captive, C. pecorum negative koala and a standard curve based on doubling dilutions of a known strong positive sample, selected during optimisation, from a C. pecorum positive koala.The inter-assay coefficient of variation was < 15% and the intra-assay coefficient of variation was accepted if it was < 10%.Each sample was run in two wells coated with antigen and two without antigen (four wells total).All samples were run in duplicate individual wells, with one well coated with antigen and one without antigen, and the mean was calculated from these replicates.All OD values were blank adjusted; the OD value of the no-antigen well was subtracted from the OD value of the antigen and plasma well for each sample.The highest concentration standard was given a nominal value of 32.The other values were calculated relative to that standard, based on dilution.The OD values of the samples were compared to the standard curve using a 4-parameter logistic curve.

Lymphocyte stimulations.
Three millilitres RPMI medium (Sigma-Aldrich #R7388), was incubated for 30 min at 37 °C.The same volume of heparanised blood from each koala was centrifuged at 3000 g for 5 min and the leukocyte fraction (buffy coat) was aspirated and suspended in the pre-incubated media.In duplicate, 5 µg of rMOMP genotype G in 25 µl of foetal calf serum (FCS) or, in the case of negative controls FCS 25 µl, was added to 220 µl of the buffy coat solution.Suspensions were then incubated for 12 h at 37 °C then 750 µl of RNA later (Sigma-Aldrich #R0901) was added to each sample and the sample stored at room temperature for 24 h, then frozen at − 20 °C until processing.
RNA extraction and cDNA synthesis.RNA was extracted from the leukocyte suspensions using the RiboPure-Blood kit (Invitrogen #AM1928) according to the manufacturer's protocol.Extracts were treated with RNAse-free DNAse, (Thermofisher #EN0521) and cDNA synthesis was then performed using the Rever-tAid First Strand cDNA Synthesis Kit (K1622) according to the manufacturer's protocol.

IFNγ and IL-17A expression.
A qPCR assay was used to estimate expression of the following genes: GAPDH, IFNγ and IL-17A 42,43 .Reactions were made to a final volume of 20 µl consisting of 10 µl of SsoAd-vanced™ Universal SYBR Green Supermix (BioRad), 0.5 µM of each primer with 6 µl of H 2 O (GAPDH) and 0.3 µM of primer with 6.8 µl of H 2 O (IFNγ and IL-17A) and 2 µl of DNA.Cycling conditions for the different genes were applied as previous 42,43 .All samples were run in duplicate with negative controls at each step.IFNγ and IL-17A expression for each koala for each time point were normalised to GAPDH by the 2 −ΔΔCT method, where ΔΔCT = (CT of target − CT of GAPDH) − (CT of target-CT of GAPDH) and presented as a fold change in relation to the unstimulated sample.

Statistical analysis.
All data analyses were performed using the R Statistical Environment (Version 3.6.1) 44.
Data were initially assessed using the 'ggplot' and 'ggdensity' functions from the 'ggplot2' 45 and 'ggpubr' 46 packages respectively.Shapiro-Wilk tests were used to check for normality.Data that did not conform to normality were log transformed 47 .
Spearman's Rank Correlations were used to test for relationships between immunological markers, chlamydial shedding, and disease variables.A variable was removed if a pair of variables had a correlation coefficient > 0.5 (or < -0.5) 48.To normalise C. pecorum CT against swab yield (koala beta-actin DNA) from the same swab sample, ΔCT values were generated (ΔCT = CT of C. pecorum CT of beta-actin).Chlamydial shedding is expressed as the inverse of C. pecorum ΔCT values because, as chlamydial shedding decreases, C. pecorum ΔCT values increase.
The effect of treatment (vaccination or placebo), time since vaccination, their interaction and sex, was tested with different linear mixed effects models (LMM), constructed using the 'lme4' package 49 , with individual koala as the random factor.Dependant variables examined included clinical disease variables, specifically ocular disease scores, wet bottom score and paraovarian cyst score (females only), chlamydial shedding (ocular and urogenital ΔCT values), anti-MOMP plasma IgG values (log transformed) and cytokine expression (log transformed fold change).

Results
Cohort recruitment.Fifty-two koalas were recruited into the study (Table 1).Post vaccination data were measured for all koalas at 6 months, 41 koalas at 12 months and 28 koalas at 18 months.The decrease in sample size was due to koalas dying from natural causes.The average tooth wear class at the beginning of the study was pre-molar four being "flat" with molar one having some wear which signifies an estimated average age of 9 years and above 36 .C. pecorum infection prevalence of the koalas sampled at the beginning of the study was 79% (Table 1).
Ultrasonography.Repeated ultrasonography showed that the presence of paraovarian cysts can change over time.Unilateral or bilateral paraovarian cysts of five female koalas were not detected subsequently.The maximum size of any paraovarian cystic structure that was not identified on subsequent ultrasound was 55 mm in diameter.Four additional female koalas showed fluctuations over 18 months, whereby paraovarian cysts would be detected, not detected, and then detected again, sometimes on opposing sides.

Vaccination response.
No correlations were sufficiently strong to warrant exclusion of variables from the model (Supplementary Table S1), in supplementary material.

Discussion
Immunological markers, along with the degree of chlamydial shedding and clinical response parameters in this blinded randomised placebo-controlled vaccination trial, indicate that this synthetic peptide vaccination regime is not an effective management tool to manage chlamydiosis within a high disease prevalence koala population.At 6, 12 and 18 months post vaccination, there was no significant difference in chlamydial shedding, disease expression, plasma anti-MOMP IgG levels or antigen stimulated IFNγ or IL-17A expression between vaccinated and placebo koalas.Additional findings include apparent vaccine-independent resolution of paraovarian cysts and the low prevalence of chlamydial shedding among koalas with clinical ocular disease.This synthetic peptide vaccine provided no identifiable reduction in chlamydial shedding or disease as measured by ocular, urogenital and paraovarian cyst scoring.The inability to demonstrate a reduction in chlamydial shedding or disease is likely a result of no identifiable vaccine-induced cell-mediated or humoral immune stimulation.The lack of response found in this study contrasts studies utilizing full length MOMP as a vaccine antigen in mice 50 , humans 51 and koalas 21,22 .The antigen used in this study comprised of four synthetic peptides of the MOMP protein that ranged from 16 to 21 amino acids in length.There is evidence that synthetic peptide vaccines utilising MOMP peptide components can stimulate immune responses and reduced chlamydial shedding in mice 31,32,52 .Our contrasting results may be the result of the size of peptides; the peptides examined in murine studies were larger and thus likely to be a more immunogenic antigen 31 .An alternate possibility is that the antigens chosen did not represent immunogenic epitopes or that the peptide was not conformationally consistent with that of the full MOMP protein.We hypothesise that either of these two scenarios would result in a lack of immune recognition of the C. pecorum elementary body (EB) when the koala is infected.It is possible that a response occurred to the vaccine peptides but was not detected in our assays as the full length recombinant MOMP was used as an antigen in the ELISA and lymphocyte stimulations as it is this response that we hypothesised is needed for vaccine efficacy.Another alternative is that there was a transient response prior to first sampling at 6 months; previous synthetic peptide vaccine trials examined vaccine response over weeks rather than months 31 .Tifrea et al. 31 detected an elevation in plasma anti-MOMP IgG and an increase in plasma IFN-γ at 4 weeks following vaccination, however, did not measure immune responses beyond this time.Therefore, it remains possible that the synthetic peptide vaccination provided a positive effect over the short term (weeks) but, as our study aimed to evaluate longer-term responses to build evidence on the use of widespread vaccination programs, short term vaccine response was out of scope of this study.While the vaccine protocol within this study included a single booster, it is plausible that the observed lack of response could stem from the need for  additional booster vaccines.In studies involving other marsupials, multiple booster vaccinations have shown a tenfold rise in antibody levels 53 .However, Nyari et al. 18 demonstrated that a single dose of a vaccine comprised of two synthetic peptides of the MOMP antigen stimulated a plasma anti-MOMP IgG response in healthy koalas.Therefore, it is likely that the characteristics of our population impacted our ability to demonstrate a vaccine induced immune response.
The demographics of this population may have contributed to the lack of development of vaccine induced immune responses.There is evidence of reduced fecundity within our study population, with only one breeding female observed within this cohort across two breeding seasons.This decline in fertility has coincided with an increase in C. pecorum prevalence within the population 5 .As a consequence of the reduced fecundity over a period of years, the population is geriatric, with the average estimated age from tooth class wear being 9 years and above 36 .Immunosenescence is well documented in humans and refers to an age-related dysfunction of immune cells 54 .One example of immunosenescence is the reduced antibody production and cell mediated responses (IFNγ) in geriatric patients following influenza vaccination 54 .In koalas, increased age, specifically 9 years and older, has been associated with a decrease in CD4 and CD8β responses 55 .The lack of immune response development in this study may be in part, therefore, a result of our study population demographics.
Koala populations with a high degree of pathology attributable to chronic inflammation may not be a suitable population to assess vaccine responses and vaccination may not be a suitable tool to address individual animal health effects however may provide a suitable tool to reduce chlamydial shedding and therefore reduce transmission events.Urogenital fibrotic inflammation that results from C. pecorum infection is likely somewhat irreversible 7 .In humans, severe persistent conjunctival inflammation following C. trachomatis infection, is associated with scarring and fibrosis 56 and can persist despite medical intervention 57 .Histopathological assessment of koalas with severe ocular disease demonstrates conjunctival fibrosis 6 , which is unlikely to resolve with medical treatment.At the beginning of this study, 5% of the males and females showed evidence of severe conjunctival proliferation and 73% of the females had evidence of genital tract pathology (paraovarian cysts) attributable in some part to chronic inflammatory process such as fibrosis of the salpingeal ducts.This chronic pathology was previously thought to be a permanent structural change and therefore unlikely to be altered by medical treatment alone, including vaccination 7 .In populations severely impacted by chlamydiosis, vaccination may therefore be ineffective at altering disease outcomes.The results of this study suggest that reducing transmission events by sterilisation or culling of infected individuals may offer better potential as management tools in these severely affected populations 58 .
The sonographic resolution or change in paraovarian cysts in nine koalas, independent of vaccination, and monitored by a single operator, further supports the notion that, in at least some koalas, cyst formation is less likely to be primarily due to fibrotic occlusions and may have a more dynamic pathogenesis than previously thought 27 .Historically, due to strong associations between the two, studies investigating the pathogenesis of paraovarian cysts in koalas have hypothesised that paraovarian cysts develop due to fibrosis and occlusion of the salpingeal lumen and were presumed to be irreversible 7,59 .More recently, Pagliarani et al. 27 demonstrated paraovarian cyst development in the absence of fibrotic inflammatory occlusions and hypothesised disruptions in epithelial homeostasis of the ovarian bursa due to C. pecorum infection was the main factor resulting in the accumulation of fluid in cystic ovarian bursae.Our sonographic findings further support the evidence that paraovarian cysts have a dynamic pathogenesis that may not be solely associated with fibrosis of the reproductive tract, and yet were still not influenced by the vaccine treatment in this study.
Within this population there was a disparity between clinical ocular disease and infection status, measured by qPCR.At the commencement of this study 3% of this population tested positive for C. pecorum at the ocular site despite 46% demonstrating clinical ocular disease.Within this population, higher disease scores correlated to severe proliferation of the conjunctiva, which likely represents the chronic stages of inflammation.The low sensitivity is likely a result of the fact that chronic inflammation and clinical signs can persist despite clearance of infection.This finding is consistent with humans with clinical trachomatis 60 and the demonstration of persistent clinical signs despite clearance of infection following antibiotic therapy 61 .Another possibility for this low sensitivity is the incorrect classification of signs such as mild ocular discharge classified to be solely as a result of C. pecorum infection, rather than other pathogenesis.
We have demonstrated that a synthetic peptide vaccination against chlamydiosis is not an effective management tool in an aging koala population with a high prevalence of C. pecorum infection and related disease.Vaccination failed to achieve a reduction in chlamydial shedding and therefore is unlikely to reduce transmission of infection in this setting.The lack of antigenic response found in this study suggests that further research utilising a larger, full-length antigen warrants further investigation if we are to consider vaccination as a part of a management strategy in diseased populations.Although the demographics and chronic pathology exhibited by our population may have impacted the assessment of vaccine efficacy, the scenario is, nonetheless, typical of many koala populations for which vaccination might be considered desirable as a management option.Evaluation of future full length antigen vaccines should include investigation of longer-term efficacy, surpassing 6 months of vaccination, in wild koala populations less severely affected by chlamydiosis.Once a viable vaccine is developed, disease and demographic status of the population should be considered, and context-specific efficacy evaluated, prior to incorporating vaccine into population wide disease management plans.
Figure 1. C. pecorum ∆CT values at the ocular site of vaccinated and placebo koalas, 0-18 months following vaccination (n = 52 at 0 and 6 months, n = 41 at 12 months and n = 28 at 18 months).Note: ∆CT reflects C. pecorum CT-β-Actin CT (normalising gene) whereby an increase in ∆CT reflects a decrease in chlamydial shedding.

Figure 2 .
Figure 2. C. pecorum ∆CT values at the urogenital site of vaccinated and placebo koalas, 0-18 months following vaccination (n = 52 at 0 and 6 months, n = 41 at 12 months and n = 28 at 18 months).∆CT reflects C. pecorum CT-β-Actin CT (normalising gene) whereby an increase in ∆CT reflects a decrease in chlamydial shedding.

Figure 3 .
Figure 3. Plasma anti-MOMP IgG values of vaccinated and placebo koalas, 0-12 months following vaccination (n = 52 at 0 and 6 months, n = 41 at 12 months).Plasma anti-MOMP IgG values on the Y axis represent values derived from a 4-parameter logistic curve.

Figure 4 .
Figure 4. Fold change in IFNγ mRNA in response to rMOMP stimulation of peripheral blood lymphocytes of vaccinated and placebo koalas, 0-12 months following vaccination (n = 52 at 0 months and n = 41 at 12 months).

Figure 5 .
Figure 5. Fold change in IL-17A mRNA in response to rMOMP stimulation of MOMP stimulated peripheral blood lymphocytes of vaccinated and placebo koalas, 0-12 months following vaccination (n = 52 at 0 months and n = 41 at 12 months).
. PCR reactions were made to a final volume of 20 µl consisting of 10 µl of SensiFAST™ Probe, 400 nM of each primer, 200 nM of each probe, 4.4 µl of dH 2 O and 2 µl of DNA.The cycling conditions included initial denaturation for 2 min at 98 °C, followed by 40 cycles of denaturation for 15 s at 98 °C and a combined annealing and extension step for 30 s at 58 °C.Samples were determined to be C. pecorum positive if the CT value of C. pecorum ompB was ≤ 34.Samples were determined to be suspect C. pecorum positive if the CT value of C. pecorum ompB was 34-40 and the 23S Chlamydia was 34-40.qPCR assay was used to quantify P. cinereus beta actin and C. pecorum ompB genes.

Table 1 .
Cohort characteristics of enrolled study group at recruitment.Numbers indicate number of koalas.