(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



A human model of Buruli ulcer: The case for controlled human infection and considerations for selecting a Mycobacterium ulcerans challenge strain [1]

['Stephen Muhi', 'Peter Doherty Institute For Infection', 'Immunity', 'The University Of Melbourne', 'Melbourne', 'Victoria', 'Victorian Infectious Diseases Service', 'Royal Melbourne Hospital', 'Parkville', 'Walter']

Date: 2023-11

The ideal Mycobacterium ulcerans challenge strain

Is not associated with severe clinical disease, either in its individual provenance or in geographically associated cases. Causes a typical infection phenotype. Is amenable to a biologically plausible route of entry. Is susceptible to clinically relevant antibiotics. Can be cultured in a non-toxic, animal-free medium without genetic or chemical modification. Can be accurately enumerated to ensure consistent challenge dosing. Retains viability after cryopreservation. Produces the key virulence factor after in vitro culture (i.e., mycolactone). Maintains a conserved repertoire of genes encoding candidate vaccine antigens. Remains genetically stable during manufacture and challenge.

1. Is not epidemiologically associated with severe clinical disease (either in its individual provenance or in geographically associated cases)

A suitable M. ulcerans challenge strain should not be associated with a severe clinical phenotype. For example, the Australian isolate JKD8049 has been previously studied in murine models of M. ulcerans infection [69,70]. JKD8049 was obtained from an adult male in 2004, during an outbreak in Point Lonsdale, Victoria. The presentation was of a characteristic painless ulcer on the posterior calf, which was first noticed incidentally by an allied health practitioner.

Although Australian isolates of M. ulcerans belong to the classical lineage that also causes BU in Africa, infections acquired in Australia demonstrate notably different clinical characteristics in observational studies and serve as a useful comparison [71]. In Point Lonsdale, Victoria, Australia, an outbreak began abruptly in 2002, with BU affecting 79 people (48 residents and 31 visitors) [72]. A clinical description of 180 cases across the entire Bellarine Peninsula (an area including Point Lonsdale) during this period noted that 95% of patients had a single lesion and the majority of patients had a nodule or ulcer [73]. Only 5% of patients presented with oedema, the majority of whom were over 60 years of age, similar to observed disease severity across Victoria [74]. One patient had septic bursitis, and another had BU osteomyelitis [73]; this contrasts with osteomyelitis rates of approximately 6% in Benin [16]. In Benin, osteomyelitis has been observed to occur some distance from active or apparently healed BU lesions and even in some patients without any history of BU skin lesion [16]. This phenomenon has rarely been reported in Australia [75]. Given that M. ulcerans grows optimally at 30° to 33°C, the subacute haematogenous spread suggested by these cases is surprising, although isolates from Africa are more thermotolerant than those from temperate regions [76].

2. Reproduces a typical infection phenotype

A successful CHIM should be clinically relevant, safely and reliably reproducing a clinical infection endpoint that faithfully resembles what is seen after natural exposure [49]. In Australia, the most common site of infection in humans is the lower limb, between the knee and ankle [77]. In an Australian setting, using an Australian isolate such as JKD8049, the posterior calf appeals as a site for inoculation, equidistant from the knee and ankle joints, and separated from the underlying bone by the large gastrocnemius muscle, reducing the (already low) risk of septic arthritis and osteomyelitis by contiguous spread, and contractures due to scarring. A lesion at this site is also easily amenable to surgical excision, if required. Observational studies also suggest that it is not a site with an increased likelihood of oedematous lesions such as those involving the hand, elbow, or ankle [57]. M. bovis BCG vaccination provides a useful reference point for the expected local superficial scar in a BU CHIM.

Unlike other pathogens tested in CHIMs, M. ulcerans is not restricted to humans, and has been identified in numerous zoonotic hosts. The selected challenge strain may therefore be characterised phenotypically in a murine model, in order to demonstrate predictable patterns of virulence and immunological responses to infection. At least in the case of JKD8049, remarkably low doses are known to establish infection in mice. For this isolate, the infectious dose to infect 50% of mice (ID50) is known to be approximately 2.6 CFU [69], although 104 to 106 CFU have been used in most vaccine/challenge murine studies, using a variety of challenge strains [45]. Such unrealistically high doses are likely to be far greater than what is required to establish natural infection and may overwhelm immune responses and underestimate the true efficacy of candidate vaccines. The CHIM should therefore use a realistically low dose of M. ulcerans. The initial study to establish a human challenge model must also incorporate a dose-finding design, as the infectious dose in humans is unknown.

In order to understand the typical disease phenotype, it is anticipated that the study end point will be reached at the onset of ulceration, or alternatively, when any pre-ulcerative lesion has been present for a significant period of time. If a cellulitic or oedematous presentation is encountered, the study end point will be reached, and treatment will need prompt initiation.

3. Is amenable to a biologically plausible route of entry

Studies have previously demonstrated that M. ulcerans isolate JKD8049 can cause infection after inoculation procedures mimicking the hypothesised mosquito bite route of entry, and also after subcutaneous injection, using realistically low doses in mice [69]. Studies have also shown that abrasions are not amenable to initiating clinical infection, at least in a guinea pig model [78]. The ID50 curve reported by Wallace and colleagues [69] demonstrates that, according to this model, the number of organisms required to establish infection in ≥90% of mice is approximately 20 CFU. Needlestick puncture has shown that a low-dose mouse tail infection was achieved in 21 of 24 (88%) of mice (average approximately 30 CFU) [69], while approximately 14 to 20 CFU resulted in infection in a mouse tail model, with at least 80% of mice infected [79]. Therefore, an initial target range of approximately 20 to 30 CFU may be considered, although minor deviations above this range are unlikely to have a clinically significant impact.

Given the success of previous experiments at successfully initiating infection using a low-dose mechanical model of infection, skin-puncturing microtrauma is proposed as the most biologically plausible method of inoculation for a BU CHIM. Specifically, a 25-gauge (or smaller) hypodermic needle, used to inject up to 0.1 mL of culture material subcutaneously, approximately 2 to 3 mm under the skin, is proposed. This depth approximates the length of a mosquito proboscis [80], which is postulated to be a possible route of infection in Australia [69,72,81].

4. Is susceptible to clinically relevant antibiotics

Although clarithromycin is the preferred companion drug to rifampicin, fluoroquinolones (such as ciprofloxacin or moxifloxacin) may be used in combination with rifampicin when clarithromycin is unavailable, contraindicated, or poorly tolerated [82]. Australian guidelines [82] recommend the use of clarithromycin alongside a fluoroquinolone when rifampicin cannot be used, based on effectiveness in mouse models [83]. Antibiotic resistance in Australian isolates has not been reported, unlike African isolates, where rifampicin resistance is described [84]. An ideal challenge strain would therefore be susceptible to rifampicin, clarithromycin, and fluoroquinolones in vitro, and a known allergy to these antibiotics would be an exclusion criterion. Although there were previously no defined clinical breakpoints for in vitro susceptibility testing, the Clinical and Laboratory Standards Institute (CSLI) guidelines offer suggested susceptibility ranges to interpret minimum inhibitory concentrations (MICs) for slow-growing non-tuberculous mycobacteria such as M. ulcerans [85]. Authorities suggest that laboratories should establish their own in-house validation for fastidious non-tuberculous mycobacterial species [85]. If externally validated, such methodology may standardise M. ulcerans MIC testing for future work.

5. Can be cultured in a nontoxic and safe minimal media, with minimal chemical modification to enrich growth

One of the major limitations in M. ulcerans research is the organism’s slow growth, due to a reported doubling time of approximately 48 h. This results in the requirement for long incubation periods, generally up to 12 weeks, within a relatively narrow temperature window. The long incubation period also increases the opportunity for contamination, reinforcing the importance of performing all experimental work within strictly sterile conditions. Although there are numerous M. ulcerans isolates available for consideration, geographically diverse isolates are remarkably conserved, with minimal genetic diversity [86]. Nevertheless, the growth characteristics of various isolates should be investigated; a faster time to culture positivity would be an attractive option, minimising the opportunity for contamination. However, an isolate with rapid growth characteristics is not necessarily a requirement for a candidate CHIM strain; rather, an isolate which reliably reproduces a typical clinical phenotype should be prioritised. The standard for M. ulcerans liquid culture, Middlebrook 7H9, requires the addition of albumin and catalase to enrich growth; both are generally derived from bovine sources. This introduces the small but not insignificant risk of bovine spongiform encephalopathy (BSE), particularly if the origin of the product is from an endemic region.

Sauton’s medium is an alternative liquid culture medium, which is free of animal products, and has an established history of use for the culture of M. bovis BCG [87]. It contains nontoxic ingredients and is pH neutral [88]. Previous studies have demonstrated that M. bovis BCG retains virulence properties when cultured in this medium compared to research media [87], while M. ulcerans reportedly also retains the ability to produce mycolactone in this medium [89]. Although often routinely added to reduce clumping, surfactants (e.g., polysorbate/Tween) are ideally avoided, aiming to minimise chemical modification, particularly considering the presence of hydrophobic lipid-rich structures, including mycolactone, in M. ulcerans. Other alternatives to explore may include replacing animal containing supplements with synthetic proteins or proteins of non-animal origin. Such additional nutritional supplementation is anticipated to enhance M. ulcerans growth; vegetable protein alternatives, which are free of animal protein and nongenetically modified, also have a history of use in a CHIM [49].

6. Can be enumerated with accuracy to ensure precise challenge dosing

An issue that is familiar to researchers of slow-growing mycobacteria, and particularly M. ulcerans, is the propensity to form clumps and biofilm, potentially containing hundreds or thousands of bacilli (Fig 1). M. ulcerans has a known dose-dependent relationship with clinical phenotype; mice receiving larger doses of M. ulcerans demonstrate earlier onset of ulceration and more rapid tissue loss [90]. Therefore, a single clump may significantly overdose challenge participants, considering the low doses proposed. In the absence of detergent, mechanical de-clumping methodologies will need to be explored.

7. Remains viable after frozen storage

Challenge doses of M. ulcerans created by any manufacturing process will require storage, generally by cryopreservation in a nontoxic substance (such as glycerol). Quality control processes of this sample will need to quantify its viability, including pre- and post-cryopreservation, and its stability after thawing using viable CFU count and/or quantitative PCR analysis.

8. Produces the key virulence factor after in vitro culture (mycolactone)

A clinically relevant BU model of infection would require that the selected isolate produces the main virulence toxin, mycolactone. Mycolactone production should be further characterised because minor sequence modifications in the ML genes of the pMUM plasmid (that encode the enzymes that synthesise the toxin) have produced a variety of ML congeners with variable virulence (A/B, C, D, E, and F). ML A/B (interconverting stereoisomers) produced by African strains are the most cytotoxic, while the potency is thought to be more attenuated in ML C (produced by Australian strains) and the other structural variants [91,92]. In murine fibroblast L929 cells treated with a series of synthetic mycolactones, ML F has been shown to be about 2 times less active and ML C about 15 times less active than ML A/B, respectively [93]. In addition to mycolactone C, Australian M. ulcerans strains also produce a fraction of ML A/B; Scherr and colleagues posit that the ML A/B portion may be more important for the pathogenesis caused by Australian strains than ML C [93]. In summary, the ability to produce mycolactone in vitro should be confirmed with liquid chromatography-mass spectroscopy.

9. Maintains a conserved repertoire of genes encoding candidate vaccine antigens

In order to test candidate vaccines against cell-surface antigens, whole-genome sequencing should confirm the presence of these genes in the challenge strain; these have been reviewed elsewhere [45]. Due to niche adaptation, global M. ulcerans isolates are remarkably conserved, so it is anticipated that most isolates should display a conserved repertoire of relevant antigens [94].

10. Remains genetically stable during manufacture and challenge

Serial whole-genome sequencing of the isolate, at multiple time points along the manufacturing cycle and following challenge, should (1) evaluate for the presence of spontaneous nucleotide polymorphisms; and (2) confirm the presence of an intact pMUM virulence plasmid.

[END]
---
[1] Url: https://journals.plos.org/plosntds/article?id=10.1371/journal.pntd.0011394

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.