An observational study to test the potency of GLP-grade phenol-free lyophilised leishmanin antigen formulation

Summary

Background

Leishmaniasis is a vector-borne disease transmitted through sand fly vectors carrying protozoan parasites from the genus Leishmania. The leishmanin skin test (LST), used to detect exposure to Leishmania parasites, is no longer available due to a lack of well-characterised antigens. Towards reintroducing the LST, we previously developed a preparation of the leishmanin antigens using a freeze-thaw (FT) process, which induced a delayed-type hypersensitivity (DTH) response in hosts exposed to Leishmania parasites. The storage of FT preparation antigens requires refrigeration, which is not practical for use in field settings.

Methods

We developed a safe, simple and scalable antigen extraction process and a phenol-free lyophilised-LST antigen formulation that obviates the need for a cold-chain under Good Laboratory Practice conditions (GLP-LST^Lyo). The immunogenicity of GLP-LST^Lyo was evaluated in murine models of Leishmania infections in support of two distinct use cases: (i) to detect latent infections of Leishmania in endemic areas and (ii) as a surrogate of vaccine-induced immunogenicity. Towards this goal, leishmanisation with wild-type Leishmania major and immunisation of mice with the live-attenuated L. major vaccine strain lacking the centrin gene (LmCen^−/−) were used as animal models.

Findings

The GLP-LST^Lyo antigen was administered, and its ability to induce a DTH response was monitored for 48 h. Additionally, the potency of the GLP-LST^Lyo was assessed by immune phenotyping of the cells isolated from the DTH tissue via flow cytometry. Only the DTH sites from pre-exposed animals, but not the excipient controls, showed enrichment for predominantly CD8⁺ T cells, along with CD4⁺ T cells and macrophages, consistent with previous studies. Additionally, in an interferon gamma release assay (IGRA), stimulation of whole blood collected from healed, active cutaneous leishmaniasis (CL) cases in a Leishmania endemic area in Brazil with GLP-LST^Lyo induced IFN-γ and IL-10.

Interpretation

These results demonstrated the potency of our formulation in IGRA studies, which could aid immunogenicity studies in vaccine trials, and surveillance studies in both endemic and emerging areas of Leishmania infection.

Funding

These studies are supported by funding from GHIT Fund, Japan, CIHR, Canada, intramural funding from the FDA, and FIOCRUZ and INCT-DT Brazil.

Research in context.

Evidence before this study

The need for the development of the lyophilised formulation of leishmanin antigens for the reintroduction of Leishmanin Skin Test (LST) and the Interferon Gamma Release Assay (IGRA) in the field was ascertained by a literature search in PubMed, Google Scholar, and informal consultations with the experts at WHO. We consulted with colleagues in disease-endemic areas, including East Africa and South America, to assess the need for leishmanin antigens to detect pan-Leishmania species. We undertook the development and testing of lyophilised formulations that can be used in field conditions, partly at the urging of WHO, to fulfil the need for a reagent in support of disease surveillance studies and vaccine development efforts. The search terms included LST, leishmanin, Montenegro Skin Test, and Interferon Gamma Release Assay (IGRA).

Added value of this study

Our study demonstrates the utility of a lyophilised formulation of the leishmanin antigens derived from an Indian strain of Leishmania donovani in inducing cytokines in whole blood or Peripheral Blood Mononuclear Cells (PBMCs) isolated from patients with active or healed cutaneous leishmaniasis. Availability of a cGMP-grade lyophilised formulation will enable disease surveillance studies in resource-limited settings either by measuring DTH response or by IGRA.

Implications of all the available evidence

We are currently investigating the utility of lyophilised formulation in cohorts of healed visceral leishmaniasis (VL) cases in Brazil, Ethiopia and Spain. Due to the potency of the selected lyophilised leishmanin formulation demonstrated in ex vivo studies, we will develop the cGMP-grade antigens and proceed with the further steps necessary for regulatory approval, including toxicology studies and testing in human subjects. The availability of cGMP-grade leishmanin antigens will benefit disease surveillance studies across all endemic areas.

Introduction

Human leishmaniases caused by the protozoan parasite Leishmania are transmitted by infected sand flies and are endemic in over 90 countries.¹ Leishmaniasis can occur as clinically distinct cutaneous, visceral, and mucocutaneous forms. The majority of the visceral leishmaniasis (VL) cases that may result in death without timely treatment occur in the Americas, East Africa, North Africa, and West and South-East Asia.² Cases of human leishmaniasis are emerging in non-endemic or low-endemic areas due to human migration,3, 4, 5 and resurging in regions where various control programs are already in place.^6,7 Travel-acquired and autochthonous cases of human leishmaniasis have also been reported in the United States.8, 9, 10 Outbreaks of VL in resource-constrained areas such as Kenya also remain a public health challenge.¹¹ Currently, no approved prophylactic vaccine against human leishmaniasis exists.

Timely diagnosis and surveillance of leishmaniasis cases are important tools in the control and elimination of human leishmaniasis. Effective surveillance requires accurate measurement of the true infection rate in a community. Identifying both symptomatic and asymptomatic cases is necessary to understand the epidemiology of the disease, assess the real burden of infection, and evaluate the effectiveness of control measures.¹² Diagnosis of active cases of leishmaniasis is accomplished by a wide variety of methods, including microscopy, serology, and molecular analysis.¹³ In comparison, methods for the detection of asymptomatic infection, and tests of cure to monitor treatment outcomes are less well optimised.^14,15 The Montenegro Skin Test (MST), a variant of the Leishmanin Skin Test (LST), typically consisting of whole Leishmania parasites treated with phenol is no longer produced by CPPI [Centro de Produção e Pesquisa de Imunobiológicos, Brazil], and reintroduction of the LST and the IFN-γ release assay (IGRA)- using a standardised GMP grade Leishmanin antigen are therefore sought to address this unmet need.^16,17 LST has been used to detect prior exposure to Leishmania parasites by a delayed-type hypersensitivity (DTH) response that relies on the recall of the cell-mediated immunity. This property is beneficial not only for detecting latent infections of leishmaniasis but also as a correlate of vaccine-induced immunity in assessing experimental vaccines. Prevailing low disease transmission rates render typical Phase-3 efficacy trials costly, and DTH positivity as a surrogate biomarker of acquired immunity will be an invaluable tool for the development of prophylactic vaccines. However, LST was discontinued in the field mainly due to the lack of cGMP-grade antigen preparations. The need to reintroduce LST and IGRA cannot be overemphasised due to the lack of reliable methods firstly to detect asymptomatic infections and secondly to assess vaccine efficacy. Studies with recombinant or crude preparations of leishmanin antigens, in some instances freeze-dried antigens (Leish-kit), have been used in whole blood assays or agglutination tests to detect infections18, 19, 20, 21, 22 but have not resulted in a commercial scale production or broader availability. To support the development of an optimised leishmanin antigen preparation, we previously reported on the selection of a source strain, manufacturing of GLP-grade leishmanin antigens, and evaluation of its potency in appropriate animal models.²³ Production of leishmanin antigens based on the freeze-thaw (FT) process showed promising results in eliciting a DTH response in both leishmanised mouse models and in hamsters that were immunised with a live attenuated vaccine strain of Leishmania major lacking the centrin gene (LmCen^−/−).²³ However, significant challenges remain in the development of consistent and reproducible leishmanin antigens that have typically been used in the LST. For instance, (1) the freeze-thaw cycles-based legacy process is not amenable to large-scale production of leishmanin antigens (2) production of LST antigens to be used in human subjects have to be manufactured under validated and controlled conditions so as to maintain lot-to-lot consistency as required by regulatory authorities, and (3) a lyophilised formulation of leishmanin antigens with long shelf-life is desirable to meet the goal of reintroducing LST in the field to facilitate testing in those disease-endemic areas where the necessity of a cold-chain is a serious limitation.

Unlike tuberculin tests,²⁴ the immunological mediators of the DTH response in Leishmania infections are not well studied. Histochemical analysis of DTH sites showed that CD3⁺ T lymphocytes and macrophages are associated with a positive DTH response.^25,26 Accordingly, our data with GLP-grade FT leishmanin antigen showed that depletion of either CD4⁺ or CD8⁺ T cells abrogates the DTH response in mice inoculated with wild-type L. major (LmWT) or genetically attenuated LmCen^−/− parasites.²³ To develop leishmanin antigens that meet regulatory product standards, standardised tests of purity, potency, safety and identity must be established. Further, phenol was used in the preparation of leishmanin antigens as an anti-microbial agent in previous studies. However, phenol is an irritant and is known to promote a DTH response by itself resulting in false–positive reactions.^27,28 Therefore, in the present study, the use of phenol has been replaced with lyophilisation for the final leishmanin antigen preparation. Finally, it was necessary to compare and optimise the extraction buffers to generate the most antigenic preparation of leishmanin prior to lyophilisation as detailed within.

In support of two distinct use cases for LST (1) detection of asymptomatic infections, and (2) use as a surrogate of vaccine-induced immunity, suitable preclinical models and assays must be developed. Further, since LST cannot be performed repeatedly due to sensitisation of the immune cells, ex vivo assays such as IGRA with a standardised antigen preparation may be more appropriate. In this report, we describe the development of a GLP-grade phenol-free lyophilised formulation of soluble Leishmania antigens and characterisation of its potency in preclinical animal models. We show that leishmanin antigens recruit both CD4⁺ and CD8⁺ T cells along with macrophages to the DTH site, similar to FT and GLP-grade leishmanin preparations, demonstrating that GLP-LST^Lyo antigens are suitable for further development through GMP manufacturing and for eventual field deployment.

In addition to the LST, the IGRA is also valuable for detecting asymptomatic infections and assessing protective immunity and can be performed ex vivo.¹⁶ It has been demonstrated that IGRA can detect cell-mediated immunity with high sensitivity and specificity in patients following cure.^18,19 We therefore further established that the GLP-LST^Lyo antigen can be used in the IGRA to identify individuals in Brazil with prior exposure to Leishmania.

Methods

Source strain of the parasite and formulation of the antigens

GLP-grade L. donovani-Indian strain (Ld-Ind) characterised in a previous study²³ was used as the source strain [MHOM/IN/93/BI2301/LRC-751] to produce leishmanin antigens by osmotic lysis (OL), followed by preparation of lyophilised formulations. Four lyophilised formulations (formulations A–D) were prepared that primarily differ in the excipients such as glycerol, mannitol, sucrose and trehalose. The composition of the four lyophilised antigen formulations including excipients is provided (Supplementary Table S1).

Antigen preparation

The leishmanin antigen products of FT and OL manufacturing processes were thawed on ice prior to inoculation into animals. The phenol-free lyophilised antigen preparations were then reconstituted in 0.5 ml sterile water and allowed to equilibrate on ice for 30 min with mixing by occasional swirling of the glass vial.

Mouse infection and immunisation

Female 6 to 8 week-old C57BL/6 mice were purchased from The Jackson Laboratory. In all Leishmania literature, it common to use female mice for logistical convenience as they can be housed in groups unlike male mice. Therefore, we followed the same precedent. The mice were immunised in the right ear with either virulent LmWT (L. major Friedlin FV9) parasites (1 × 10⁴) or LmCen^−/− parasites (1 × 10⁶) by intradermal needle injection in the ear in 10 μl PBS using a 29 or 30-gauge needle (BD Ultra Fine). After 8–10 weeks of immunisation, mice were inoculated intradermally in the contralateral ear with the freeze thaw (GLP-LST^FT), osmotic lysis (GLP-LST^OL) or reconstituted lyophilised (GLP-LST^Lyo) antigen formulations by needle injection in the ear in 10 μl volume using a 30-gauge needle. Corresponding excipient solutions were used as buffer controls. The sample size for the animal experiments was determined through an a priori power analysis conducted using preliminary estimates of the mean and variance. The calculation assumed a two-way ANOVA design (treatment × mouse group), with the interaction term as the primary outcome. Power analysis using G∗Power (version 3.1.9.x) indicated that a minimum of four animals per group is required to achieve ≥80% power (α = 0.05). Therefore, each experiment was designed to include ≥4 animals per group, with the exact number fluctuated slightly depending on animal availability and ethical considerations guided by 3Rs principles.

LST by intradermal inoculation

Contralateral ears of immunised or leishmanised mice were intradermally injected with lyophilised leishmanin antigens or matching excipient control using 28 or 29 G needle. Leishmanin antigen (5 μg) in 10 μl volume was used for inoculation. Ear thickness of the inoculated mice was measured prior to inoculation (0 h), and at 24 h and 48 h post-injection. To enable comparison of results between both collaborating laboratories, ear thickness measurements were normalised to 100% relative to the baseline initial thickness (0 h) values from the naïve, no LST group. Changes in ear thickness were determined for all other groups and time points.

Tissue collection and single cell suspension

48 h after leishmanin antigen inoculation, the injected ears were dissected from the euthanised mice and briefly rinsed in 70% ethanol, followed by PBS. Then, the DTH response site was excised in a Petri dish, and the ear sheets were peeled and immersed in 1.5 ml RPMI at room temperature (RT) containing Collagenase P 2 mg/ml (11249002001) and 0.44 μl of DNAse (79254) in 6 well plates. The plates were incubated at 37 °C in 5% CO₂ for 120 min to allow enzymatic digestion. The tissue and enzymatic cocktail was transferred to a gentleMACS C Tube (Miltenyi Biotech Cat. No. 130-093-237) for cell dissociation with an additional 8.5 ml complete RPMI at RT. Cell dissociation was performed using a Gentle-MACS device and 3 consecutive cycles of the Mouse Intestine protocol available as pre-set options in the device, for 1 min each. The cell suspension was passed through a 70 μm Filcon filter. Single cell suspensions were pelleted by centrifugation at 1600 rpm for 6 min at 4 °C, followed by one wash with 10 ml PBS. The cells were resuspended in 1 ml PBS and counted in a Cellometer prior to staining.

Flow cytometric analysis

The single-cell suspensions were placed in PBS in flat bottom 96-well plates and incubated with Fc-blocking (CD16/32) antibody and Live/dead Aqua fixable viability dye for 15 min at 4 °C prior to surface staining with fluorochrome-conjugated antibodies in a final 100 μl of PBS for 30 min at 4 °C (See Supplementary Table S2). Later, cells were washed with PBS and fixed using BD Cytofix™ Fixation Buffer (Cat. No. 554655) for 30 min at 4 °C, protected from the light. The fixed cells were washed twice with PBS.

Stained cells were analysed by flow cytometry using the five-laser spectral flow cytometer Aurora (Cytek Biosciences, Fremont, CA) containing 16 violet, 14 blue, 10 yellow-green, and 8 red channels (4L-16V-14B-10YG-8R) for the evaluation of T cells in GLP-LST^FT, and FACSymphony A1 (BD bioscience) for the rest of the flow cytometry experiments in formulations A–D. During the acquisition, compensation was performed using reference controls integrated into the SpectroFlo software (Cytek Biosciences). Each fluorochrome peaked in individual channels, and the spectral pattern was validated using the BD Flow references provided (Supplementary Table S2). Acquired samples were analysed as FCS files using FlowJo software version 10.10.0 (BD Biosciences, Ashland, OR).

Immunohistochemistry

At 48 h post-injection with lyophilised (GLP-LST^Lyo) antigens of each formulation, mice were euthanised, and the respective ears were collected. Ears were stored in 4% paraformaldehyde-phosphate buffered solution for at least 48 h, followed by 70% ethanol until further processing for paraffin embedding and tissue sections. Some of the sections were stained with Haematoxylin and Eosin (H&E) and other sections were processed for immunostaining. Anti-CD4 (Clone: D7D2Z, Cell Signaling Technology) and anti-CD8α (Clone D4W2Z, Cell Signaling Technology) antibodies were used to stain for CD4⁺ or CD8⁺ T cells. All histochemical staining was performed at the Central Institute for Experimental Medicine and Science (CIEM), Japan and the Biomedical Research Support Center (BRSC), Nagasaki University School of Medicine, Japan.

Ethics

All animals were housed in pathogen-free conditions in the vivarium at the Food and Drug Administration (FDA) animal facility, Silver Spring (MD) and the Animal Research Center, Nagasaki University. All animal experiments in this study were reviewed and approved by the Animal Care and Use Committee of the Center for Biologics Evaluation and Research, U.S. Food and Drug Administration (ASP 1995#26) and the animal care committee of Nagasaki University for ethics on animal experiments (approval number 2004271624, 2004271625, 2104011711), and align with ARRIVE guidelines.

The study using clinical material was conducted in the Health Post of Corte de Pedra, municipality of Tancredo Neves, an ATL-endemic area in the southeastern region of Bahia, Brazil. This study was conducted according to the principles specified in the Declaration of Helsinki and according to the Ethics Committee at Fiocruz Bahia (CEP FIOCRUZ/Bahia, CAAE: 86870725.0.0000.0040). All participants gave written informed consent to participate. Participants were patients with active Cutaneous Leishmaniasis (CL) (n = 15) and patients with a previous history of CL (n = 26). Diagnosis of active CL was confirmed by the presence of a typical ulcer, documentation of DNA of Leishmania braziliensis by PCR and one of the following: a positive MST or identification of amastigotes in the skin biopsies by immunohistochemistry.²⁹ Patients were treated with meglumine antimoniate (MA) in doses of 20 mg/kg per day for 20 days. Patients with a previous history of CL were recruited at the same site. Inclusion criteria included clinical records confirming diagnosis of CL in the past, presence of a scar at the site of the former lesion, indicating complete healing and re-epithelisation. Exclusion criteria included individuals <18 years old, pregnancy, presence of other comorbidities, or a positive serologic test for human immunodeficiency virus (HIV). Healthy individuals, residents of areas non endemic for leishmaniasis (Salvador, Bahia) and without history of travelling to endemic areas were enrolled as controls (n = 7).

Statistics

Comparisons were made between DTH responses in various GLP-LST formulations with GraphPad Prism V10 (GraphPad Software, San Diego, CA). The DTH response using FT and OL products were analysed using two-way ANOVA (excipient antigen) followed by Bonferroni test. DTH response measurements with lyophilised products were compared between excipient control and antigen inoculated groups at equivalent time points (LmCen^−/− and LmWT) using an unpaired t-test. The pre-clinical datasets were analysed using two-way ANOVA (treatment × group) followed by Tukey's post-hoc test. In studies performed with human cells or blood, comparisons between control cultures (medium) and antigen-stimulated cultures (LbSLA and GLP-LST^Lyo-C) were performed by Kruskal–Wallis, followed by Dunn's multiple comparisons test. Accordingly, ∗ represents p < 0.05, ∗∗ represents p < 0.005, ∗∗∗ represents p < 0.001, and ∗∗∗∗ represents p < 0.0001.

Role of funders

The funders of the current study had no role in the design, data collection, data analysis, data interpretation, or writing of this manuscript. Our (TPF, LK, HM, NA, MB, CH, IJ, HLN, SG) contributions are considered an ‘informal communication’ and represent our own best judgement. These comments do not bind or obligate FDA.

Results

Manufacturing of leishmanin antigen through osmotic lysis

Leishmanin antigen was prepared from a well characterised L. donovani strain (Ld-Ind) under GLP conditions. A frozen vial of L. donovani promastigotes was thawed and the seed culture was inoculated in M199 based growth medium using T-flask and shake flask (Fig. 1A). The seed was further inoculated into a bioreactor with 5L of M199 based growth medium. The promastigotes were grown in the bioreactor for four days under controlled conditions. Parasite density and viability were measured using an Acridine orange/Propidium iodide double stain method and automated cell counter Cellometer. The bioreactor was inoculated at 2.0 × 10⁶ parasites/ml of density and reached ∼56.0 × 10⁶ parasites/ml after four days of cultivation (Fig. 1B, Supplementary Table S3). Parasites viability greater than 95% was observed during bioreactor cultivation (Fig. 1B, Supplementary Table S3). Microscopic observations showed a well-defined healthy parasite morphology (Fig. 1C). After four days of cultivation when culture reached in stationary phase, parasites were harvested for LST antigen extraction. The harvested parasites were washed three times by centrifugation at 1300 g for 10 min using sterile 0.8% (w/v) phosphate buffered saline (PBS), pH 7.2 with 1% (w/v) glucose. After the final wash, the pellet was re-suspended in sterile water at a parasite concentration of 500–1000 × 10⁶ parasites/ml. The solution was incubated for 30–45 min at room temperature of 20–25 °C with continuous stirring for osmotic lysis. The antigen suspension was centrifuged at 10,000 × g, 4 °C for 30 min. The supernatant (GLP-LST^OL) was collected, and phosphate buffer was added before filtration with 0.2 μm membrane filter. From this GLP-LST^OL, a final formulation was prepared for lyophilisation after addition of excipients. From this final formulation, lyophilised GLP-LST antigen (GLP-LST^Lyo) vials were produced (Fig. 1A). The composition of the four lyophilised antigen formulations including excipients is provided (Supplementary Table S1). The critical quality attributes (CQA) of lyophilised vials are described in Fig. 1A. The products of osmotic lysis were evaluated as in-process controls that showed the complete array of antigens as was observed in our previous freeze thaw cycles-based manufacturing process²³ (GLP-LST^FT) on SDS-PAGE analysis (Fig. 1D). To establish the potency of the osmotic lysis intermediate product, mice were intradermally immunised with LmCen^−/− vaccine strain for 8–10 weeks and were inoculated in the contralateral ear with either GLP-LST^OL antigen or GLP-LST^FT antigen or the excipient buffer control. The DTH response measured at 0, 24, and 48 h following leishmanin antigen inoculation showed that LmCen^−/− vaccine strain-inoculated mice exhibited a detectable DTH response while the naïve mice, and excipient control-inoculated mice showed a baseline response (Fig. 1 E). The DTH response in LmCen^−/− inoculated mice was significantly higher in GLP-LST^OL group 48 h post-inoculation (Fig. 1E). A well-defined DTH response was detected at the site of inoculation with GLP-LST^OL and GLP-LST^FT whereas the excipient controls showed no identifiable response (Fig. 1F). The DTH response in GLP-LST^OL across the excipient, naïve and immunised groups was comparable to the response observed in GLP-LST^FT antigen inoculated mice (Fig. 1E).

Fig. 1.

Manufacturing steps in GLP-LST antigen production and their validation by DTH response. Manufacturing of GLP grade leishmanin antigen by L. donovani (Ld-Ind) promastigotes. A) A brief process flow of LST antigen manufacturing from frozen parasites vial to Lyophilised LST antigens vials. B) Parasite density and viability during bioreactor cultivation. A bioreactor containing 5L of growth medium was cultivated for 4 days. Parasite density and viability was measured daily using automated Nexcelom automated cell counter. C) Morphology of L. donovani promastigotes was observed during bioreactor cultivation by using inverted microscope at 400-X magnification. D) SDS-PAGE analysis of leishmanin antigens extracted from osmotic lysis. Lane 1, molecular markers, lane 6 reduced and lane 7, non-reduced form of antigens. kDA-kilodalton. E) DTH site measurement comparisons of GLP-LST^FT and GLP-^OL show the increase in ear thickness in vaccinated animals at 24- and 48-h post inoculation in response to antigens prepared by freeze-thaw (FT) and Osmotic lysis (OL) methods. F) Representative ear pictures are shown where the DTH site is observed (red circle) 24 h after LST performance. Data are presented as mean ± SD. Statistical analysis was performed using unpaired t-test. ∗p < 0.05.

Manufacturing of lyophilised leishmanin antigens

Towards developing a stable lyophilised leishmanin antigen, the osmotic lysis intermediate was used as the starting material. While the freeze thaw process yielded higher protein content compared to the osmotic lysis-based process, the parasite DNA concentration in the resulting material was well above the regulatory product standards.³⁰ The antigen material from the osmotic lysis process was used in combination with suitable buffer and a “Generally Recognised as Safe” (GRAS) by the U.S. Food and Drug Administration (FDA) approved list of excipients to produce four different lyophilised formulations (Fig. 2A). The antigen composition in the four different formulations and the input osmotic lysis intermediate was assessed on a silver stained SDS gel that showed comparable banding patterns (Fig. 2B). Two different analytical chromatographic methods were used to further assess the antigen composition of the four different formulations as in-process controls. Results from both methods showed comparable antigen compositions in all four formulations and matching with the pattern observed through the osmotic lysis intermediate (Fig. 2C and D). All four lyophilised formulations contained less than 20 ng of residual parasite DNA per 5 ug dose, and <1 EU/ml of endotoxin levels (Supplementary Table S4–5). Immunogenicity of the lyophilised antigen material was tested in mice immunised with LmCen^−/− parasites for 8 weeks. A clearly distinguishable and significant DTH response was observed 48 h after inoculation with GLP-LST^Lyo antigens in mice immunised with LmCen^−/− parasites compared to excipient control and LmCen^−/− inoculated groups (Fig. 2E). The antigen inoculated ears showed a comparable DTH response as was observed in the FT product (Fig. 2F).

Fig. 2.

Development of a detectable DTH response in after inoculation of the lyophilised Leishmanin antigen (GLP-LST^Lyo), after inoculation (LST) in 10-week immunised mice (LmCen^−/−). A) GLP lyophilised leishmanin antigen was generated following good manufacturing conditions (GMP). B) Four different formulations of excipient were generated and compared via SDS gel against a GMP-Freeze thaw leishmanin antigen. C) Reverse phase HPLC elution profiles indicate all four LST antigens possess similar hydrophobicity. D) All four LST antigen formulations eluted from the SEC column. E) DTH site measurement comparisons of GLP-LST^Lyo show the increase in ear thickness in vaccinated animals at 24 and 48 h post inoculation in response to lyophilised antigen after 12 weeks of vaccination. F) Representative ear pictures are shown where the DTH site is observed (red circle) 24 h after LST performance. Data are presented as mean ± SD. Statistical analysis was performed using unpaired t-test, ∗∗∗p < 0.001.

Selection of the formulation for product development

While developing a field-deployable LST, four different formulations of phenol-free lyophilised leishmanin antigen were configured containing different excipients. Compositions of the excipients differed in sugars and PS-80. Formulations A (GLP-LST^Lyo-A) and B (GLP-LST^Lyo-B) are similar in their formulation, whereas formulations C (GLP-LST^Lyo-C) and D (GLP-LST^Lyo-D) are similar in their formulation.

We tested all four formulations in animals that were either vaccinated with 1 × 10⁶ LmCen^−/− or leishmanised with 1 × 10⁴ virulent strain LmWT (Fig. 3). After 10 weeks of either immunisation or leishmanisation (when the lesions were resolved in the latter), mice were injected intradermally in contralateral ears with 5 μg of one of the four freshly reconstituted preparations. GLP-LST^Lyo-A showed to be capable of eliciting a significantly higher DTH response after both vaccination and leishmanisation 48 h post inoculation, marked by the increase in the ear thickness (Fig. 3A). However, the DTH response was not significantly higher than some of the naïve animals exposed to GLP-LST^Lyo-A, or the immunised animals exposed to the excipient (Fig. 3A, 24 h LmWT group). Similarly, inoculation with GLP-LST^Lyo-B in both vaccinated and leishmanised animals showed a DTH response that was not clearly distinguishable from the naïve and excipient control animals, and excipient and LmWT groups (Fig. 3B). Formulation B was therefore deemed not appropriate for further evaluation. With formulations GLP-LST^Lyo-C (Fig. 3C) and GLP-LST^Lyo-D (Fig. 3D), we observed that both induced significantly better DTH responses at 24 and 48 h post antigen inoculation, in both vaccinated and leishmanised animals compared to naïve and excipient controls (Fig. 3C and D). Together these data showed that formulations GLP-LST^lyo-C and GLP-LST^Lyo-D are superior to formulations GLP-LST^lyo-A and GLP-LST^Lyo-B in eliciting a DTH response in both vaccinated and leishmanised animals.

Fig. 3.

Comparison of four Lyophilised LST formulations. Four lyophilised antigens were tested to evaluate the DTH response induced by with different excipient composition in 12 weeks vaccinated mice. Formulations A) GLP-LST^LyoA, B) GLP-LST^LyoB, C) GLP-LST^LyoC, and D) GLP-LST^LyoD. Changes in ear thickness show the development of a DTH response at 24 and 48 h after LST. Data are presented as mean ± SD. Statistical analysis was performed using unpaired t-test, ∗∗p < 0.05, ∗∗∗p < 0.01, ∗∗∗p < 0.001 and ∗∗∗∗p < 0.0001.

DTH response in GLP-LST^FT inoculated mice is associated with CD4⁺ and CD8⁺ T cells

Data on the immunological correlates of DTH reaction following inoculation with leishmanin antigens are relatively sparse compared to those available for the tuberculin tests.³¹ To fully reveal the correlates of DTH reaction, we developed a flow cytometry protocol to characterise the immune cell populations in the DTH site (Fig. 4). Following spontaneous resolution of the cutaneous lesion in C57BL/6 mice caused by wild-type L. major parasites, the mice were used as leishmanisation models. As a correlate of vaccine-induced immunogenicity, DTH response in mice immunised with the LmCen^−/− vaccine strain was used. To first establish a baseline for the immune cell populations in the DTH site, we characterised the response using GLP-LST^FT antigen. The DTH response tissue was harvested, and single cell suspensions were produced by enzymatic digestion (Fig. 4A, Supplementary Table S6). The cell suspensions were stained with a cocktail of fluorescently tagged antibodies for analysis by flow cytometry. The gating scheme to identify various immune cell populations in the DTH site is shown in Supplementary Fig. S2. Since T cells are necessary for the development of a DTH response,²³ we analysed the percentages of CD4⁺ and CD8⁺ T cells in the DTH site, 48 h after inoculation of GLP-LST^FT antigens (Fig. 4B). Results showed that CD4⁺ T cells were increased in the LmWT group after inoculation with GLP-LST^FT (Fig. 4C), while CD8⁺ T cells were enriched in both LmCen^−/− and LmWT groups upon inoculation with GLP-LST^FT compared to naïve animal receiving the GLP-LST^FT (Fig. 4D). Haematoxylin and eosin (H&E) stained sections of the DTH site showed clear recruitment of cells in LmCen^−/− vaccinated and leishmanised mice compared to naïve controls (Fig. 4E). Immunohistochemistry staining with anti-CD4 and anti-CD8 antibodies showed the presence of CD4⁺ and CD8⁺T cells in the DTH response site in mice injected with LmCen^−/− and LmWT parasites but only a basal response in naïve controls (Fig. 4E).

Fig. 4.

T cell responses against freeze-thaw Leishmanin antigen (GLP-LST^FT). A) Experimental procedure timeline consisting of an initial leishmanisation (LmWT) or vaccination (LmCen^−/−) of C57Bl/6 mice 12 weeks before performing the LST. For the LST, whether GLP-LST^FT, GLP-LST^OL, or GLP-LST^Lyo leishmanin antigens were injected intradermally. After 24 and 48 h the DTH site was measured and subsequently collected for flow cytometry immunophenotyping. B) Representative dot plots of the flow cytometry analysis of CD4⁺ CD8⁺ T cells from ear tissues of vaccinated mice after injecting with GLP-LST^FT Leishmanin antigen. Changes in the percentage of C) CD4⁺ and D) CD8⁺ T cells in the site of inoculation are shown. E) H&E-stained ear tissue sections at 1 × (scale bar 500 μm) and at × 20 (scale bar 50 μm) magnification showing the cellular infiltration at the site of injection in the vaccinated (LmCen^−/−) and the leishmanised (LmWT) mice. Immuno-histochemical staining of the ear sections shows the infiltration of CD4⁺ and CD8⁺ T cells after injection with the GLP-LST^FT. Scale bar 50 μm. n = 3–6 mice per group. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA followed by Tukey's post-hoc test. ∗p < 0.05 and ∗∗∗p < 0.001.

CD4⁺ and CD8⁺ T cells mediate the DTH response in phenol-free lyophilised leishmanin antigens similar to GLP-LST^FT antigens

To test which of the lyophilised antigen formulations (GLP-LST^Lyo) upon reconstitution show a level of potency substantially equivalent to the freeze-thaw antigens (GLP-LST^FT), immunophenotyping of the DTH site was performed following inoculation with GLP-LST^Lyo-A or GLP-LST^Lyo-C. Frequencies of CD4⁺ and CD8⁺ T cells in the DTH site were analysed, 48 h after inoculation of either GLP-LST^Lyo-A (Fig. 5A–D) or GLP-LST^Lyo-C (Fig. 5E–H) in mice infected with LmWT or vaccinated with LmCen^−/− parasites or non-immunised naïve mice. Immunophenotyping of the DTH site via flow cytometry (Fig. 5A) showed a significantly increased frequency of CD8⁺ (Fig. 5C) but not CD4⁺ (Fig. 5B) T cells in both the LmWT or LmCen^−/− groups 48 h after the inoculation with GLP-LST^Lyo-A (Fig. 5B and C), compared to inoculation of the excipient buffer and compared to the naïve group inoculated with GLP-LST^Lyo-A. In contrast, a significantly increased frequency of CD4⁺ (Fig. 5F) and CD8⁺ (Fig. 5G) T cells in LmCen^−/− but not LmWT group 48 h after the inoculation with GLP-LST^Lyo-C, compared to inoculation of the excipient buffer and to naïve group inoculated with GLP-LST^Lyo-C. Similar to the data from GLP-LST^Lyo-A, CD8⁺ T cells showed a significant enrichment in both LmCen^−/− and LmWT groups in response to GLP-^Lyo-C inoculation (Fig. 5G). The immunophenotyping results from GLP-LST^Lyo-C were similar to those observed in GLP-LST^FT (Fig. 4C and D). H&E-stained sections of the DTH site showed clear recruitment of cells in LmCen^−/− vaccinated and LmWT infected mice compared to naïve controls in GLP-LST^Lyo-A and GLP-LST^Lyo-C (Fig. 5D and H). Further, immunohistochemistry staining with anti-CD4 and anti-CD8 antibodies showed the presence of CD4⁺ and CD8⁺T cells in the DTH response site in mice injected with LmCen^−/− and LmWT parasites but only a basal response in naïve controls upon inoculation with in GLP-LST^Lyo-A and GLP-LST^Lyo-C (Fig. 5D and H). Immunophenotyping of the GLP-LST^Lyo-D groups showed a similar pattern of recruitment of CD4+ and CD8+ T cells by flow cytometric method to the GLP-LST^Lyo-C groups in the LmCen^−/− vaccinated mice (Supplementary Fig. S2A–C). H&E staining of the ear dermis inoculation site revealed cellular infiltration in the skin tissue (Supplementary Fig. S2D). Immunohistochemical staining showed that an infiltration of both CD4⁺ and CD8⁺ T cells was observed in the LmCen^−/− vaccinated mice and the leishmanised mice, while the naïve mice showed only background levels (Supplementary Fig. S2D), therefore, revealing that the LST responses by GLP-LST^Lyo-C and GLP-LST^Lyo-D were associated with an infiltration of both CD4⁺ and CD8⁺ T cells in the LmCen^−/− vaccinated mice and the leishmanised mice. Definitive quantitative estimates are hard to draw from immunohistochemistry results thus demonstrating the utility of data from our flow cytometry analysis in selecting the right formulation suitable for further development.

Fig. 5.

Evaluation of T cell responses against lyophilised leishmanin antigens in formulations A and C. T cell analysis of the DTH site 48 h after LST with GLP-LST^Lyo-A(A–D) and GLP-LST^Lyo-C(E–H). A) Representative dot plots display the presence of T cell after GLP-LST^Lyo-A. Percentages of B) CD4⁺ T cells and C) CD8⁺ T cells were estimated via flow cytometry. D) H&E-stained ear tissue sections at 1 × (scale bar 500 μm) and at × 20 (scale bar 50 μm) magnification showing the cellular infiltration at the site of injection in the vaccinated (LmCen^−/−) and the leishmanised (LmWT) mice. Immuno-histochemical staining of the ear sections showing the infiltration of CD4⁺ and CD8⁺ T cells after injection with the GLP-LST^Lyo-A. Scale bar 50 μm. E) Representative dot plots display the presence of T cell after GLP-LST^Lyo-C. Percentages of F) CD4⁺ T cells and G) CD8⁺ T cells were estimated via flow cytometry. H) H&E-stained ear tissue sections at 1 × (scale bar 500 μm) and at × 20 (scale bar 50 μm) magnification showing the cellular infiltration at the site of injection in the vaccinated (LmCen^−/−) and the leishmanised (LmWT) mice. Immuno-histochemical staining of the ear sections showing the infiltration of CD4⁺ and CD8⁺ T cells after injection with the GLP-LST^Lyo-C. Scale bar 50 μm. n = 4–7 mice per group. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA followed by Tukey's post-hoc test. ∗p < 0.05, ∗∗p < 0.005 and ∗∗∗p < 0.001.

Recruitment of innate cells to the DTH site

In addition to CD4⁺ and CD8⁺ T cells, we also evaluated the composition of innate cells (neutrophils, macrophages, dendritic cells, and Langerhans cells) at the DTH site of immunised mice, as well of at the site of injection of naïve mice. Representative flow plots for neutrophils and macrophages are shown in Supplementary Fig. S3, while dendritic cells and Langerhans cells are observed in Supplementary Fig. S4. We observed a significant increase of non-T cells from the live cells in the DTH sites in LmWT or LmCen^−/− groups in GLP-LST^FT (Fig. 6A) compared to naïve controls. Similar recruitment of non-T cells was also observed in LmWT or LmCen^−/− groups upon inoculation with either GLP-LST^Lyo-A or GLP-LST^Lyo-C antigens (Fig. 6B and C). Deconvolution of the non-T cell populations revealed that there were significant changes in the Ly6G⁺ CD11b⁺ neutrophils in GLP-LST^Lyo-A (Fig. 6E) but not in either GLP-LST^FT (Fig. 6D) and GLP-LST^Lyo-C (Fig. 6F) compared to excipient controls inoculations in both LmWT or LmCen^−/− groups (Fig. 6D–F). Significant increase of the percentages of F4/80⁺ CD11b⁺ macrophages in the DTH site was observed in LmWT group in the GLP-LST^FT (Fig. 6G) as well as both GLP-LST^Lyo-A (Fig. 6H) or GLP-LST^Lyo-C (Fig. 6I) inoculations groups compared to naïve controls (Fig. 6G–I). However, in case of LmCen^−/− groups, the significant increase of the percentage of macrophages were observed in both GLP-LST^Lyo-A (Fig. 6H) or GLP-LST^Lyo-C (Fig. 6I) but not in GLP-LST^FT (Fig. 6G) compared to naïve controls (Fig. 6G–I). The percentages of CD11c⁺ MHCII⁺ dendritic cells in the DTH sites from the LmCen^−/− vaccinated mice and the leishmanised mice showed no significant change in the GLP-LST^FT as well as both GLP-LST^Lyo-A or GLP-LST^Lyo-C inoculations groups compared to naïve controls (Fig. 6J–L). The percentage of CD207⁺ CD11c⁺ Langerhans cells in the DTH site from the leishmanised mice was significantly reduced in the GLP-LST^FT group (Fig. 6M). However, no significant differences were observed in GLP-LST^Lyo-A or GLP-LST^Lyo-C inoculations (Fig. 6N and O). Conversely, the percentage of CD207⁺ CD11c⁺ Langerhans cells in the DTH site from the LmCen^−/− vaccinated mice showed no significant change in the GLP-LST^FT as well as both GLP-LST^Lyo-A or GLP-LST^Lyo-C inoculations groups compared to naïve controls (Fig. 6M−O). Similar to GLP-LST^Lyo-c, GLP-LST^Lyo-D (Supplementary Fig. S5) showed an overall increase of non-T cells (Supplementary Fig. S5A) and macrophages (Supplementary Fig. S5C) in leishmanised and vaccinated animals, neutrophils and dendritic cells showed no change (Supplementary Fig. S5B and D), while Langerhans cells percentage was reduced only in the vaccinated group (Supplementary Fig. S5E). Collectively, these results suggest that GLP-LST^Lyo-C is the best candidate for GMP leishmanin antigen production due to its ability to recruit immune cells comparable to the standard GLP-LST^FT.

Fig. 6.

Evaluation of innate cell responses against GLP-grade leishmanin antigens. A–C) Percentage of Leukocytes (CD45+ CD3−) from live cells as well as D–F) Ly6G⁺ CD11b⁺ neutrophils, G–I) F4/80⁺ CD11b⁺ macrophages, J–L) CD11c⁺ MCH-II⁺ dendritic cells and M–O) CD207⁺ CD11c⁺ Langerhans cells from live cells at the DTH site. The graphs show the innate immune cell environment after 48 h of inoculation with excipient (grey bars) or leishmanin antigens (blue bars): GLP-LST^FT (top panel), formulation A (GLP-LST^Lyo-A, middle panel), and formulation C (GLP-LST^Lyo-C; bottom panel), in naïve mice, vaccinated (LmCen^−/−) mice, and leishmanised (LmWT) mice, respectively. n = 3–7 mice per group. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA followed by Tukey's post-hoc test. ∗p < 0.05, ∗∗p < 0.005 and ∗∗∗p < 0.001.

Ex vivo potency of GLP-LST^Lyo-C in individuals naturally exposed to Leishmania

Following the demonstration that GLP-LST^Lyo-C antigen induces a DTH response in mice, with predominant recruitment of CD8⁺ T-cells, along with CD4⁺ T cells and macrophages, we set to determine whether this antigen preparation would be able to induce recall responses in individuals naturally exposed to Leishmania. Studies were conducted in Bahia, Brazil, in an area endemic for CL caused by L. braziliensis (Fig. 7A). We did not observe significant differences regarding baseline characteristics of participants with active CL compared to healed individuals (Supplementary Table S7). Healed individuals had the initial lesion as early as 2021 and time to heal was 85 days following standard treatment with pentavalent antimonials. Whole blood samples collected from individuals with active CL or with history of CL (healed) were stimulated with GLP-LST^Lyo-c antigen. Whole blood from naïve controls, residing in a non-endemic area, were used as controls (Fig. 7B). Levels of IFN-γ were unchanged in samples from healthy subjects (Fig. 7C) whereas stimulation with GLP-LST^Lyo-c antigen induced a significantly higher production of IFN-γ in individuals with active CL and those with history of CL (healed) (Fig. 7D and E, respectively). Similar results were observed for IL-10 (Fig. 7 F and H). Of note, no significant differences in cytokine levels were detected comparing LbSLA and GLP-LST^Lyo-c antigen, indicating that this source recapitulates the antigen repertoire present in LbSLA. Importantly, GLP-LST^Lyo-c antigen also induced IFN-γ and IL-10 production by PBMCs of individuals with active disease or history of CL (Supplementary Fig. S6). Collectively, these data show the potential of GLP-LST^Lyo-c antigen for the purposes of identifying immunised individuals as well as for surveillance studies, advancing towards the possibility of re-introducing LST through the production of the antigen in regulatory agency-approved conditions.

Fig. 7.

Cytokine production by whole blood from Leishmania-exposed individuals. A) Enrolment of subjects into the observational study with inclusion criteria is shown. B) Whole blood from Leishmania-exposed individuals (actice CL, n = 15 and healed CL, n = 26) or from naïve individuals residing in non-endemic areas (n = 7) were stimulated in vitro with soluble leishmania antigen (SLA) from L. braziliensis or with GLP-LST^LyoC for 24 h. Whole blood was collected, and IFN-γ C–E) and IL-10 F–H) were performed by ELISA. Data are shown individually. Comparisons between control cultures (medium) and antigen-stimulated cultures (LbSLA and GLP-LST^Lyo-C) were performed by Kruskal–Wallis, followed by Dunn's multiple comparisons test. ∗p < 0.05, ∗∗p < 0.005, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001.

Discussion

The need for well-characterised and consistent production of leishmanin antigen for the LST and the IGRA for the control of leishmaniasis is widely recognised. Availability of well-characterised leishmanin antigens will enable widespread adoption of LST and IGRA in the field for better disease surveillance and developing control strategies. Secondly, the LST and IGRA will allow assessment of the efficacy of therapeutic regimens following treatment in test of cure formats. Additionally, the potency and durability of the immune response induced by prophylactic vaccines currently being developed can be tested using the LST and IGRA as a surrogate of protective immunity. Towards these goals, we developed GLP-grade leishmanin antigens using a well-characterised L. donovani as a source strain (Ld-Ind).²³ Cross-reactivity of the antigens derived from a visceralising species of Leishmania (L. donovani) has been demonstrated in murine models of leishmanisation (virulent L. major Friedlin strain), and in mice and hamsters inoculated with a live attenuated vaccine strain of L. major (LmCen^−/−) by the DTH response elicited upon intradermal administration of the GLP-LST antigens.²³ Similar utility of LST in detecting various clinical presentations of cutaneous leishmaniasis has been previously demonstrated.³²

Despite the need for reintroduction of the LST in the field, no systematic studies have been undertaken for large-scale manufacture of leishmanin antigens from a well characterised strain of the Leishmania parasite. Previous versions of leishmanin antigens made at the Pasteur Institute of Iran were formulations of killed L. major parasites suspended in a phosphate buffer and anti-microbial agents.³³ Similar studies in Brazil with LST, leishmanin antigen preparations using Leishmania amazonensis or Leishmania panamensis showed that cold-chain management for storage is necessary to maintain the potency of LST.³⁴ Variable results in different leishmanin preparations could be controlled if standardised antigens are manufactured at commercial scale. One of the common steps in the typical manufacturing process of a biological product is the control of host DNA. However, the parasite DNA content in LST preparations has never been reported in previous studies. The safety risk of residual DNA in biological products has been under investigation since the advent of recombinant DNA technology.^35,36 Residual DNA in the preparations in certain biologics has been shown to cause unintended immune response through activation of Toll-like receptors (TLRs).³⁷ Accordingly, a standard of 10 ng DNA per dose has been established,³⁰ and methods to assess the residual DNA in biologics have been extensively studied to meet the applicable standard.³⁸ Previous preparations manufactured using a freeze-thaw based process were never rigorously tested for the residual DNA and were not developed to ensure critical quality attributes are maintained in the leishmanin antigens. During GLP-LST^FT production reported previously,²³ our freeze-thaw cycles-based process typically yielded antigen preparations that contained >1 μg/ml quantities of parasite DNA (Supplementary Table S4). To overcome this limitation, we developed an osmotic lysis-based manufacturing process that allowed better control over the residual DNA content in the antigen preparations. The intermediate product resulting from the osmotic lysis process was also tested in animal models to ensure elicitation of a DTH response, and no effect on the potency was found as observed in our immunised animal models.

Previous versions of LST from the Pasteur Institute Iran contained 6 × 10⁶ killed L. major parasites suspended in 1 ml Phosphate buffered saline and 0.1% Thimerosal,³² or from Wellcome that contained 1 × 10⁷ L. major promastigotes/ml treated with phenol.³³ Similarly, leishmanin antigens extracted from Leishmania parasites used in several studies contained phenol as an anti-microbial agent.³⁹ Leishmanin antigens have been used in epidemiological studies in Africa and South America with varying degrees of specificity and sensitivity.^25,26,33,40, 41, 42, 43, 44 Phenol is an irritant and is known to promote a DTH response by itself resulting in false–positive reactions while using LST.²⁸ Thus, developing a leishmanin antigen formulation without Phenol would enhance the specificity of LST. Accordingly, we prepared all four formulations without Phenol and tested the potency. More importantly, a lyophilised formulation will maintain sterility, would eliminate the need for cold-chain management and improve the ease of use in remote and resource constrained settings.

Potency testing and selecting the best performing lyophilised formulation necessitated development of assays that meet the regulatory product standards beyond the demonstration of a strong DTH response in an appropriate animal model. Therefore, we developed a flow cytometry-based method to characterise the cellular composition of the DTH site. Studies of LST describing the cellular composition and the mediators of cell recruitment at the DTH site are limited. CD3⁺ T lymphocytes and large mononuclear cells were found to be prevalent in canine DTH tissues.⁴⁵ CD4⁺ and CD8⁺ T cells were found in DTH site in asymptomatic and chronic cutaneous leishmaniasis cases when tested with heat killed and thimerosal-fixed L. panamensis and L. amazonensis promastigotes.²⁵ Similarly, immunohistochemical analysis revealed that CD4⁺ T, CD8⁺ T cells and macrophages were predominantly found in the DTH sites.²⁶ A recent study in cutaneous leishmaniasis cases in Brazil identified Granzyme⁺ CD8⁺ T cells to be associated with the DTH response.⁴⁶ Similar studies on the tuberculin skin test on which LST was modelled also reported CD3⁺ T lymphocytes, macrophages, and Langerhans cells are recruited to the DTH site.²⁴

Accordingly, our flow cytometry-based protocol catalogued CD4⁺ and CD8⁺ T cells in addition to neutrophils, macrophages, dendritic cells, and Langerhans cells. Data showed that various lyophilised formulations differed in their capacity to recruit CD4⁺ and CD8⁺ T cells to the DTH site. A comparison of the frequencies of CD4⁺ and CD8⁺ T cells showed that formulations GLP-LST^Lyo-C and GLP-LST^LyoD were superior to GLP-LST^Lyo-A in recruiting these cells to the DTH site. The frequencies of CD4⁺ T cells were notably indistinguishable from the GLP-LST^Lyo-A and its corresponding excipient control, even though significantly more CD8⁺ T cells were found in the same DTH site compared to excipient control in leishmanised and immunised models. Since previous reports showed the presence of both CD4⁺ and CD8⁺ T cells in the DTH site of LST, the formulation GLP-LST^Lyo-C was selected for further development towards cGMP production. Notably, our data is corroborated by our previous observations with a freeze thaw-based GLP-LST where we showed depletion of either CD4⁺ or CD8⁺ T cells prior to inoculation with the leishmanin antigen abrogated the DTH response.²³ Data from the immune histochemistry staining of the DTH sites from GLP-LST^Lyo-C and GLP-LST^LyoD inoculations validated the recruitment of CD4⁺ and CD8⁺ T cells.

In addition to the well-established role of CD4⁺ and CD8⁺ T cells in mediating the DTH response to leishmanin antigens, our study sheds further light on the recruitment and dynamics of innate immune cells at the DTH site. Flow cytometric analysis revealed significant infiltration of macrophages and, to a lesser extent, neutrophils, following antigen challenge in both LmCen^−/− vaccinated and leishmanised mice. These findings are consistent with previous reports that innate cells contribute to the early phases of the DTH response and help shape the local inflammatory milieu necessary for effective T cell recruitment and activation. Notably, our results indicate that the phenol-free lyophilised formulations GLP-LST^Lyo-C and GLP-LST^Lyo-D not only recruit adaptive immune CD4⁺ and CD8+ T cells but also induce a pattern of innate cell infiltration comparable to the standard GLP-LST^FT. The significant increase in macrophages at the DTH site suggests that these cells may play a supportive role in antigen presentation and the amplification of the local immune response. The recruitment of neutrophils, observed particularly with GLP-LST^Lyo-A, may reflect formulation-dependent differences in the early inflammatory response, which could influence the specificity and intensity of the DTH reaction. Importantly, the lack of excessive neutrophil recruitment with GLP-LST^Lyo-C and GLP-LST^Lyo-D may contribute to improved specificity by minimising background inflammatory responses. The observation that Langerhans cell percentages were reduced only in certain groups further suggests a nuanced regulation of skin-resident dendritic cell subsets during the DTH response. While the functional significance of this reduction remains to be fully elucidated, it may reflect the migration of activated Langerhans cells to draining lymph nodes for antigen presentation, as has been described in other skin test models. Collectively, these findings highlight the importance of evaluating both adaptive and innate immune cell recruitment when optimising leishmanin antigen formulations. The ability of GLP-LST^Lyo-C to induce a balanced recruitment of macrophages and T cells, without excessive innate cell-mediated inflammation, may underlie its superior performance as a diagnostic and immunological tool. Future studies should further investigate the interplay between innate and adaptive immune responses in the context of LST, as this may inform the development of even more specific and sensitive assays for leishmaniasis surveillance and vaccine evaluation.

In the context of vaccination, a test that can measure IFN-γ production in immunised individuals is essential. To this end, the development of an IFN-γ release assay (IGRA) would be an important tool to evaluate immunogenicity of vaccine candidates as well as for disease surveillance. Stimulation of whole blood with Leishmania antigen peptides induced IFN-γ release in a sensitive and specific manner and use of soluble leishmanin instead of Leishmania peptide antigens improved sensitivity, indicating that an IFN-γ release assay (IGRA) is feasible and providing compelling evidence that a standardised leishmanin product is necessary to advance the vaccine trials.⁴⁷ Our results build on this proposition by showing that GLP-LST^Lyo induces IFN-γ release in whole blood assays in individuals with active disease or with history of CL at comparable levels, suggesting that L. donovani (Indian Strain) derived soluble antigens cross-react with cutaneous strains circulating in South America. Similar studies in other endemic areas with the GLP-LST formulation will test the reactivity against pan-Leishmania species. Of note, IFN-γ release was similar employing cognate antigen (LbSLA) and GLP-LST^Lyo antigen, indicating the potency of GLP-LST^Lyo antigen to induce IFN-γ release across exposure to diverse Leishmania species. IGRA assays with lyophilised GLP-LST may also help in situations where repeated LSTs are not desirable. Moreover, availability of lyophilised LST antigen will further help resolve PCR negative cases and ambiguous results in complex asymptomatic leishmaniasis cases with comorbidities, such as HIV⁺ cases found in disease endemic areas.41, 42, 43^,48,49 Further, reintroducing LST will empower testing of vaccine efficacy.^50,51 Availability of well-characterised leishmanin antigens allows detection of asymptomatic infections in ex-vivo studies when repeated LSTs are not possible or desirable due to cellular sensitisation.

Availability of well characterised and GMP grade lyophilised leishmanin antigens may eliminate the need for cold chain storage and variability in the results from test of cure due to inconsistencies in antigen preparations. In summary, a phenol free lyophilised formulation of LST antigen was manufactured using an osmotic lysis-based process that meets the quality standards for biological products and the potency was tested based on an immunophenotyping assay besides solely relying on the subjective DTH response. This reinforces the idea that DTH response does represent correlate of immunity. A fully characterised product like GLP-LST^Lyo-C will advance the reintroduction of a highly standardised diagnostic test into the field for consistent and reliable detection of latent infections, while also serving as a surrogate for vaccine immunogenicity in future vaccine trials.

The current study has been undertaken at a single centre with a small sample size of healed/active CL cases. Future studies with larger, more diverse cohorts and targeted recruitment to include asymptomatic cases, low/high endemicity, unusual clinical presentations, diverse Leishmania species will be critical to more precisely assess potency characteristics of the lyophilised antigen formulation used in the current study. More importantly, utility of the GLP-LSTLyo-C will have to be tested against healed VL and CL cases in other disease endemic areas including East Africa where different species of Leishmania may be prevalent.

Contributors

TF, LK, SI, KS, HM, NA, MB, SB, FV, CH, IJ, PT, SS, PK, LS, PB, MS, TC performed the experiments. TF, LK, SI, KS, SK, CO, EC, GM, ARS, SS, SH, SG, and HLN analysed the data and edited the manuscript. ARS, GM, SH, SS, SG, and HLN were responsible for funding acquisition, supervising, writing, and editing the manuscript. All authors read and approved the final version of the manuscript. SG, HLN, CO, SH, SI, TPF accessed and verified the underlying data.

Data sharing statement

Requests for materials and raw data should be addressed to Sanjay Singh (sanjay.singh@gennova.co.in), Shinjiro Hamano (shinjiro@nagasaki-u.ac.jp), Camila Oliveira (camila.indiani@fiocruz.br), Sreenivas Gannavaram (sreenivas.gannavaram@fda.hhs.gov), or Hira L. Nakhasi (hira.nakhasi@fda.hhs.gov) and will be made available upon request following publication.

Declaration of interests

The authors declare the following competing interests: The FDA is currently a co-owner of two US patents that claim attenuated Leishmania species with the Centrin gene deletion (US7,887,812 and US 8,877,213). Kamaleshwar P. Singh, Swarnendu Kaviraj and Sanjay Singh are employees of Gennova Biopharmaceuticals. Gennova Biopharmaceuticals filed a patent on the lyophilised formulation of the leishmanin antigens (Indian Patent Provisional Application No. 202521049141). GM is an elected member of Royal Society of Canada; HLN and ARS are elected fellows of the American Society of Tropical Medicine and Hygiene, and ARS is an elected member of American Society of Microbiology. The remaining authors declare no competing interests.