Immunogenicity of Theileria parva p67C Antigen Delivered via Adjuvanted CoPoP Liposomes in Cattle and Mice

Round 1

Reviewer 1 Report (Previous Reviewer 3)

Comments and Suggestions for Authors

The authors have adequately addressed all the reviewers’ comments and queries. Few final comments;

Line 492: "BoLa" should be "BoLA".

Table S2: Please replace PHAD and QS21 in 'liposome name' column.

Author Response

Reviewer 1

The authors have adequately addressed all the reviewers’ comments and queries. Few final comments;

Comment 1: Line 492: "BoLa" should be "BoLA".

Response 1: The section where the word appeared has been eliminated from the manuscript.

Comment 2: Table S2: Please replace PHAD and QS21 in 'liposome name' column.

Response 2: Thank for the correction, this is now changed in Table S2.

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

Manuscript ID: vaccines-4180496

Title: Immunogenicity of Theileria parva p67C antigen delivered via adjuvanted CoPoP liposomes in cattle and mice

The article needs extensive improvements. Currently, the article clearly lacks integrated approach in almost every part starting from simple summary to the end. Authors are strongly suggested to do following improvements:

- Why did the study focus on cattle and mice?

- The abstract lacks specificity regarding the data presented in the study. It is recommended to highlight key findings or notable data to give readers a clearer understanding of the study's contributions.

- The aim of the study is not clearly stated in either the abstract or the Introduction. I recommend stating this clear.

- Introduction lacks the necessary information and includes unnecessary one.

- In all text: Long sentences should be avoided.

- Go through all manuscript and improve structures of the sentences and English grammar.

- Discussion section needs extensive improvement regarding linking findings of the current study with previous work. There is not sufficient information which can explain the rationale behind changes observed in the current study. Authors are encouraged to discuss the mechanism of action with caution based on available data.

- The conclusion section is missing; please add it at the end of the manuscript.

- Line 126: Change " lower immune response" to "a lower immune response"

- Line 142: Change "ethanol injection" to "the ethanol injection"

- Line 195: Change "equivalent of" to "equivalent to"

- Line 196: Change "sequence), with" to "sequence) with"

- Line 197: Change "incubated" to "incubated for"

- Line 215: Change "ratio" to "ratio of"

- Line 216: Change " prior of " to " prior to"

- Line 221: Change " day 0 and 42" to " days 0 and 42"

- Line 237: Change " Results expressed " to " Results are expressed "

- Line 215: Change " Ficoll gradient;" to " a Ficoll gradient,"

- Line 215: Change " magnetics beads " to " magnetic beads "

- Line 263: Change " respectively), or " to " respectively) or "

- Line 286: Change " negative control" to "a negative control"

- Line 336: Change " previous" to " previously"

- Lines 359 & 366: Change " in average " to " on average "

- Lines 358-363: A long sentence of six lines. Long sentences should be avoided.

- Line 398: Change " poor " to " poorly"

- Line 416: Change " 15,2" to " 15.2"

- Line 460: Change " both human, companion animal" to " human, companion animals"

- Line 538: Change " alone, or" to " alone or"

- Line 543: Change " [15], in" to " [15] in"

- Lines 561-564: Rewrite this sentence

Author Response

Reviewer 2

The article needs extensive improvements. Currently, the article clearly lacks integrated approach in almost every part starting from simple summary to the end. Authors are strongly suggested to do following improvements:

Comment 1: Why did the study focus on cattle and mice?

Response 1: Theileria parva specifically infects cattle, and previous studies evaluating the p67C sporozoite antigen have been conducted in this species with promising outcomes. This subsequent study was therefore designed to build on and improve those earlier results, making the use of the same animal model both logical and scientifically appropriate. In parallel, several prior studies using CoPoP liposome formulations have been conducted in murine models, where they elicited strong antibody responses, particularly against Plasmodium spp. Consequently, the inclusion of mice in this study provides a necessary baseline for comparison with earlier CoPoP liposome studies, enabling cross-validation of immunogenicity across established experimental systems.

The rationale of using cattle and mice in the experiments is explained throuout the manuscript with pre-existing statements and new ones. Some examples:

• Line 110: “However, the potential of this technology in large livestock animals remains to be evaluated.
• Lines 145-147: “Here we present the evaluation of four different preparations of cobalt-porphyrin–phospholipid (CoPoP) combined with immunomodulators to enhance the immune responses of the Theileria parva model antigen p67C in bovine and mice models”
• Lines 378-380: “Four groups of five cattle were immunised with the p67C antigen bound to different CoPoP-liposomes (Table 1) to assess the capacity of SNAP technology to enhance the immunogenicity of p67C.”
• Lines 426-427: “In the interest of ascertaining the appropriateness of the CoPoP technology for the delivery of p67C, a poorly immunogenic antigen, a group of six mice was immunised with…”
• Line 505-507: “In this sense, p67C, with limited immunogenic capacity, is an ideal model antigen to test new technologies for vaccine delivery in cattle, since it allows the measurement of subtle variations in the immune response.”
• Line 508-510: “In cattle, and in comparison with our historical data, CoPoP liposomes did not perform better than soluble p67C (s-p67C) or previously evaluated nanoparticle technologies administered with commercial emulsion adjuvant (Montanide ISA206 VG, Seppic) [23,24].”
• Lines 562-563: “ The use of saponins for full-length p67 and p67C immunisation in cattle was previously reported with very successful antibody responses (reviewed in [22])…”

Comment 2: The abstract lacks specificity regarding the data presented in the study. It is recommended to highlight key findings or notable data to give readers a clearer understanding of the study's contributions.

Response 2: The abstract has been extensively modified to accommodate the journal guidelines and following the reviewers’ recommendations.

Lines 25-54:

“Background: Effective vaccines are essential to overcome the limitations of livestock immunisation, particularly in low- and middle-income countries (LMICs), where scalable, thermostable, and easy-to-administer solutions are needed. Nanoparticle-based delivery systems, such as the Spontaneous Nanoliposome Antigen Particleization (SNAP) technology, using CoPoP liposomes, offer a promising alternative for subunit vaccine development, although their performance in large animal species remains poorly characterised. CoPoP enables the rapid non-covalent multimeric display of His-tagged protein antigens combined with immunomodulators on liposomes incorporating cobalt-porphyrin-phospholipid (CoPoP).
Objective: To evaluate the immunogenicity of CoPoP-based liposomes delivering the Theileria parva p67C antigen in cattle, and compare their performance in murine models.
Methods: Cattle and mice were immunised with p67C formulated in CoPoP liposomes incorporating QS-21 and/or PHAD immunomodulators. Humoral and cellular responses were assessed. Parallel in vitro stimulation of bovine PBMC with Quil-A was used to investigate the mechanistic effects of saponins on bovine cells.
Results:  CoPoP liposome formulations did not improve p67C immunogenicity in cattle, with antibody responses at least two-fold lower than previously reported results, and no detectable cellular responses. In contrast, the same platform induced up to 2,000-fold higher antibody titres in mice. This disparity is likely driven by differences in antigen dose relative to body mass, tissue architecture, lymphatic accessibility, and innate immune signalling differences. PHAD-mediated TLR4 activation appeared less effective in cattle, whereas QS-21 induced a broader immune activation, likely through conserved inflammasome pathways. Despite limited immunogenicity, antigen presentation by CoPoP liposomes was preserved.
Conclusions: SNAP-based CoPoP liposomes showed strong immunogenicity in mice but limited efficacy in cattle, highlighting the challenges of cross-species translation. Optimisation of antigen dose and adjuvant selection for the targeted species is required, with QS-21 representing a more promising candidate than the TLR4-agonist. The scalability and versatility of SNAP technology support its continued development for multivalent livestock vaccines.”

Comment 3: The aim of the study is not clearly stated in either the abstract or the Introduction. I recommend stating this clear.

Response 3: The objective of the study has been included in the abstract, lines34-35 “Objective: To evaluate the immunogenicity of CoPoP-based liposomes delivering the Theileria parva p67C antigen in cattle, and compare their performance in murine models.”; and in the Introduction, lines 145-147 “Here we present the evaluation of four different preparations of cobalt-porphyrin–phospholipid (CoPoP) combined with immunomodulators to enhance the immune responses of the Theileria parva model antigen p67C in bovine and mice models”

Comment 4: Introduction lacks the necessary information and includes unnecessary one.

Response 4: The introduction has been modified to include all relevant background information necessary to follow the manuscript. A new paragraph has been included to increase clarity and improve the background and rationale.

Lines 74-87: “Nanotechnology provides innovative approaches with great potential to enhance vaccine efficacy. The use of nanoparticles in vaccine delivery is an effective strategy owing to their small size, usually below 1 mm in diameter, making them efficiently phagocytosed by the antigen-presenting cells (APCs) [10]. The increase in antigen uptake results in increased antigen presentation to the immune system [11,12]. In addition, nanoparticles promote preservation of antigen structure and controlled antigen release, which prolongs the immune system stimulation [10]. Taking all these characteristics together, the use of nanoparticles enhances the immunogenicity of antigens.

Nanoparticles explored for use in veterinary vaccines include liposomes, polymers, and virus-like particles (VLPs), all of which can be engineered to facilitate efficient antigen presentation to the immune cells [13,14]. Some examples of improvements are the co-delivery of the nanoparticle-antigen with adjuvant molecules, enhancing even further the immune response by targeting specific cells and pathways; and the incorporation of metals such as cobalt and nickel, which enhance antigen binding and presentation [15].”

 

 

 

Comment 5: In all text: Long sentences should be avoided.

Response 5: We have corrected the English grammar and avoided long sentences throughout the manuscript.

Comment 6: Go through all manuscript and improve structures of the sentences and English grammar.

Response 6: We have corrected the English grammar and avoided long sentences throughout the manuscript.

Comment 7: Discussion section needs extensive improvement regarding linking findings of the current study with previous work. There is not sufficient information which can explain the rationale behind changes observed in the current study. Authors are encouraged to discuss the mechanism of action with caution based on available data.

Response 7: The discussion section has been extensively modified according to the reviewers’ recommendations. Some information has been removed, and a comprehensive section explaining the disparity of results between the experiments in cattle and mice has been included (lines 517-561).

Comment 8: The conclusion section is missing; please add it at the end of the manuscript.

Response 8: The conclusion section has been included in lines 616-637.

Minor corrections:

- Line 126: Change " lower immune response" to "a lower immune response"

This has been corrected in line 148

- Line 142: Change "ethanol injection" to "the ethanol injection"

This has been corrected in line 163

- Line 195: Change "equivalent of" to "equivalent to"

This has been corrected in line 217

- Line 196: Change "sequence), with" to "sequence) with"

This has been corrected in line 218

- Line 197: Change "incubated" to "incubated for"

This has been corrected in line 219

- Line 215: Change "ratio" to "ratio of"

This has been corrected in line 237

- Line 216: Change " prior of " to " prior to"

This has been corrected in line 238

- Line 221: Change " day 0 and 42" to " days 0 and 42"

This has been corrected in line 243

- Line 237: Change " Results expressed " to " Results are expressed "

This has been corrected in line 259

- Line 215: Change " Ficoll gradient;" to " a Ficoll gradient,"

This has been corrected in line 283

- Line 215: Change " magnetics beads " to " magnetic beads "

This has been corrected in line 284

- Line 263: Change " respectively), or " to " respectively) or "

This has been corrected in line 285

- Line 286: Change " negative control" to "a negative control"

This has been corrected in line 308

- Line 336: Change " previous" to " previously"

This has been corrected in lines 360

- Lines 359 & 366: Change " in average " to " on average "

This has been corrected in lines 388 and 394

- Lines 358-363: A long sentence of six lines. Long sentences should be avoided.

We corrected long sentences in the manuscript. However, in this case the sentence might look long because it includes data in brackets.

In the manuscript (lines 387-392):

In line with these findings, and despite the large variability between the outbreed cattle in the same group, on average Group 4 (p67C-CPQ) and Group 3 (p67C-CQ) cattle developed higher antibody titres (p < 0.05) from day 42 to the end of the experiment (day 42 titres: 78.8 ± 73.5 mg/ml and 84.2 ± 67.9 mg/ml, respectively; Figure 2C, D, and E) than animals in Group 1 (p67C-CA at day 42 titres: 11.42 ± 16.1 mg/ml) or Group 2 (p67C-CP at day 42: 6.7 ± 9.2 mg/ml) (Figure 2 A, B, and E).

Removing the data in brackets:

In line with these findings, and despite the large variability between the outbreed cattle in the same group, on average Group 4 (p67C-CPQ) and Group 3 (p67C-CQ) cattle developed higher antibody titres from day 42 to the end of the experiment than animals in Group 1 or Group 2.

- Line 398: Change " poor " to " poorly"

This has been corrected in line 427                   

- Line 416: Change " 15,2" to " 15.2"

This has been corrected in line 446

- Line 460: Change " both human, companion animal" to " human, companion animals"

This has been corrected in line 490

- Line 538: Change " alone, or" to " alone or"

This section has been removed from the discussion.

- Line 543: Change " [15], in" to " [15] in"

This section has been removed from the discussion.

- Lines 561-564: Rewrite this sentence

The sentence has been improved. Lines 585-587: “and interestingly, Group 3 exhibited a substantial portion of the antibody response directed towards a sequence also recognised by sera from cattle immunised with the protective p67C-I53-50, which has neutralising capacity (pin 75, EEEVKKILDEIVKDP)”

Reviewer 3 Report (New Reviewer)

Comments and Suggestions for Authors

Major comments:

1. Physicochemical characterisation of the formulations

The manuscript reports the particle size and polydispersity index of the liposomal formulations. However, surface zeta potential measurements are not included. Zeta potential is a critical parameter for nanoparticle-based vaccine systems as it affects colloidal stability, antigen adsorption and interactions with biological membranes and immune cells. Therefore, the authors should determine and report the surface zeta potential of the different formulations (CA, CP, CQ and CPQ). Including this information would provide a more comprehensive physicochemical characterisation of the vaccine platform and facilitate the interpretation of immunological results.

2. Visualisation of vaccine components

 

The introduction describes several vaccine components and adjuvants, including CoPoP liposomes, QS21 and PHAD/MPLA. However, including schematic diagrams or chemical structures of these components would improve the manuscript. Including a figure that illustrates the chemical structures of the main adjuvant molecules and the SNAP antigen display mechanism would greatly improve the readability of the manuscript and help readers to better understand the formulation strategy.

3. Discussion of species-specific immune responses

The study reports strong antibody responses in mice, but comparatively weak and variable responses in cattle. While the Discussion briefly addresses this observation, the explanation could be expanded. The authors are encouraged to discuss potential species-dependent immunological differences in more detail, including variations in innate immune receptor signalling, adjuvant responsiveness and antigen presentation mechanisms between rodents and bovine species.

Minor comments:

The introduction could be streamlined slightly to focus more specifically on nanoparticle vaccine platforms and SNAP technology.

The authors should clarify whether the efficiency of antigen binding to CoPoP liposomes was comparable across all formulations tested.

Figure legends should include more detailed descriptions of the experimental conditions.

Showing individual animal data points in antibody response graphs would help to visualise variability within the cattle groups.

Further minor language editing would improve the readability of the manuscript.

Comments on the Quality of English Language

Minor language editing would improve the readability of the manuscript

Author Response

Reviewer 3

 

Major comments:

Comment 1: Physicochemical characterisation of the formulations. The manuscript reports the particle size and polydispersity index of the liposomal formulations. However, surface zeta potential measurements are not included. Zeta potential is a critical parameter for nanoparticle-based vaccine systems as it affects colloidal stability, antigen adsorption and interactions with biological membranes and immune cells. Therefore, the authors should determine and report the surface zeta potential of the different formulations (CA, CP, CQ and CPQ). Including this information would provide a more comprehensive physicochemical characterisation of the vaccine platform and facilitate the interpretation of immunological results.

Response 1: We have now measured and included the surface zeta potential for all p67C antigen formulated liposomes (CA, CP, CQ, and CPQ). These data have been added to Figure 1E, line 170 (“zeta potential”) and lines 361-362 “and the zeta potential of the p67C-liposomes was slightly negatively charged (Figure 1E)”.

Comment 2: The introduction describes several vaccine components and adjuvants, including CoPoP liposomes, QS21 and PHAD/MPLA. However, including schematic diagrams or chemical structures of these components would improve the manuscript. Including a figure that illustrates the chemical structures of the main adjuvant molecules and the SNAP antigen display mechanism would greatly improve the readability of the manuscript and help readers to better understand the formulation strategy.

Response 2: We have revised Figure 1A to include schematic illustrations of chemical structures of the key adjuvant components, including CoPoP, QS-21, and PHAD.

Comment 3: Discussion of species-specific immune responses. The study reports strong antibody responses in mice, but comparatively weak and variable responses in cattle. While the Discussion briefly addresses this observation, the explanation could be expanded. The authors are encouraged to discuss potential species-dependent immunological differences in more detail, including variations in innate immune receptor signalling, adjuvant responsiveness and antigen presentation mechanisms between rodents and bovine species.

Response 3: The discussion has been improved by providing different possible explanations for the disparity of results between cattle and mice.

Lines 517-561:

“The significant disparity in antibody titres, despite a 62.5-fold increase in the dose of adjuvanting molecules and 35-fold increase in antigen dose in cattle, is likely to be multifactorial. First, it is important to consider the large disproportion in the immunomodulator-to-body-volume ratio. The dose of immunomodulators used in mice can easily reach immune saturation, while in cattle, the dose used is diluted in a larger volume (approximately 128 mg/kg in mice and 0.8 mg/kg in cattle of PHAD or QS-21) [32,33]. In line with this, the nanoparticles were injected intramuscularly in both species, and while the distance to the nearest draining lymph node is negligible in mice, it is considerable in cattle. The nanoparticles can become trapped in the dense muscle tissue, exacerbating the dilution factor of both the immunomodulators and the antigen [34,35]. These findings underscore the capital importance of optimizing the dose of antigen and immunomodulators for use in large livestock animals.

At the molecular level, the suboptimal performance of p67C-CP (PHAD-only formulation) in cattle might indicate species-specific limitations in the TLR4 ligand recognition and signalling pathway. Structural divergence in the TLR4/MD-2 (myeloid differentiation protein 2) complex across species can significantly alter ligand-binding affinity. Specifically, sequence variations in the MD-2 may diminish its capacity to recognise PHAD in cattle compared to mice, hence decreasing the intensity of the signal [36–38]. Furthermore, while the TLR4/MD-2/CD14 complex is essential for optimal activation in the bovine model [37,39], murine studies suggest that CD14 might be dispensable for efficient signalling in certain contexts, highlighting a less restrictive TLR4 activation in mice than in cattle [40]. Importantly, both species might differ in the downstream pathway activation. Although both MyD88-dependent and -independent cascades are present in both species, the MyD88-dependent cascade appears to play a more dominant role in bovine immunity [36,38,41], and the activation of a specific pathway is ligand specific [42]. Collectively, these differences highlight the need for further investigation to identify the most effective TLR4 agonist for maximising immunogenicity in cattle.

On the contrary, QS-21 might have a more universal adjuvant capacity by directly stimulating the NLRP3 inflammasome upon the loss of cellular homeostasis, without the need for receptor-ligand interaction [43–45]. Thus, QS-21 might be able to generate a broader and more potent signal, which is translated into higher antibody titres than PHAD. However, the formulation is lacking the synergistic effect that QS-21 and PHAD often have when administered as an immunostimulatory combo, as it happens after administration of antigens combined with AS01 adjuvant [43,46].

Although saponin mode of action has been extensively studied in mice and human models, the cellular targets and signalling pathways triggered by saponins in cattle remain poorly defined. Our results suggest that a similar signal cascade might be stimulated. Bovine PBMCs stimulated in vitro for 24 h with increasing concentrations of Quil-A exhibited a transcriptional profile indicative of NLRP3 inflammasome activation [43,44,46]. Key signature pro-inflammatory cytokines, including IL-1α, IL-1β, and TNF-α, were markedly upregulated. Notably, the anti-inflammatory cytokine IL-10 was also slightly induced, likely as a regulatory mechanism to counterbalance the heightened inflammatory response. Together, these results indicate that saponins can activate conserved innate immune pathways across species, overcoming the differences in the genetic background in a way that TLR4 agonist, PHAD, could not.”   

Minor comments:

Comment 4: The introduction could be streamlined slightly to focus more specifically on nanoparticle vaccine platforms and SNAP technology.

Response 4: An extra section has been included in the introduction in lines 74-87:

“Nanotechnology provides innovative approaches with great potential to enhance vaccine efficacy. The use of nanoparticles in vaccine delivery is an effective strategy owing to their small size, usually below 1 mm in diameter, making them efficiently phagocytosed by the antigen-presenting cells (APCs) [10]. The increase in antigen uptake results in increased antigen presentation to the immune system [11,12]. In addition, nanoparticles promote preservation of antigen structure and controlled antigen release, which prolongs the immune system stimulation [10]. Taking all these characteristics together, the use of nanoparticles enhances the immunogenicity of antigens.

Nanoparticles explored for use in veterinary vaccines include liposomes, polymers, and virus-like particles (VLPs), all of which can be engineered to facilitate efficient antigen presentation to the immune cells [13,14]. Some examples of improvements are the co-delivery of the nanoparticle-antigen with adjuvant molecules, enhancing even further the immune response by targeting specific cells and pathways; and the incorporation of metals such as cobalt and nickel, which enhance antigen binding and presentation [15].”

Comment 5: The authors should clarify whether the efficiency of antigen binding to CoPoP liposomes was comparable across all formulations tested.

Response 5: Two complementary figures are included in the manuscript, Figure 1B and 1C showing the capacity of p67C to bind to the CoPoP liposomes in an equivalent and efficient manner.

Extra text has been included in line 349 to further clarify this aspect: “to all the CoPoP liposomes in an equivalent capacity…”

Comment 6: Figure legends should include more detailed descriptions of the experimental conditions.

Response 6: We have included more details about the experimental conditions in the legend of all figures.

Comment 7: Showing individual animal data points in antibody response graphs would help to visualise variability within the cattle groups.

Response 7: The individual animal data points in antibody titres have been shown in Figure 2A-D. In Figure1F all individual animal data points are represented by individual dots.

Comment 8: Further minor language editing would improve the readability of the manuscript.

Response 8: This has been addressed in the whole manuscript.

Reviewer 4 Report (New Reviewer)

Comments and Suggestions for Authors

The presented manuscript may be a useful contribution to the field after correcting the following issues.

1) The abstract and discussion present SNAP/CoPoP as a promising platform for large livestock, but the core findings tell a more cautious story: no detectable cell-mediated immunity in cattle, and humoral responses below the ISA206VG benchmark. The manuscript should be reframed around what was shown: QS21-containing formulations improve antibody responses over CoPoP alone, but the platform in its current form does not yet meet the immunological standard set by adjuvanted soluble p67C in cattle. This is the main result.

2) Cattle received 70 µg p67C / 500 µg CoPoP / 200 µg adjuvant per dose; mice received 2 µg / 8 µg / 3.2 µg. No mg/kg or allometric rationale is provided for either species. The authors acknowledge the need to optimize dosing for large animals, but this should be stated as a limitation of the current design, not just a future direction. The difference in immunogenicity between species may reflect suboptimal dosing in cattle rather than a species biology effect, and this alternative interpretation should be discussed.

3) The comparison to soluble p67C + ISA206VG is based on data from a separate experiment, not a concurrent control group. Given the high inter-animal variability in outbred cattle, statements like "2-fold lower" or "approaching the level of" should either be replaced with informal qualitative language or accompanied by an explicit caveat about the limitations of cross-experiment comparisons.

4) The PBMC stimulation experiment uses Quil-A (not QS21), a single 24-hour timepoint, and three naive donors. The pro-inflammatory cytokine signature observed is interesting but does not explain the absence of cellular responses in vivo. This section should be clearly framed as hypothesis-generating only.

Minor comments

1) Supplementary Table S2. The row labeled "CoPoP + QS21 (CP)" lists PHAD as a component, and "CoPoP + PHAD (CQ)" lists QS21 — the reverse of the nomenclature used throughout the manuscript (CP = CoPoP+PHAD; CQ = CoPoP+QS21). This should be corrected.

2) It is stated that Kruskal–Wallis was used for cattle group comparisons, but no post-hoc test is specified. Additionally, the results section describes differences "from day 42 to the end of the experiment," while the statistical methods section only addresses day 42.

Author Response

Reviewer 4

The presented manuscript may be a useful contribution to the field after correcting the following issues.

Comment 1: The abstract and discussion present SNAP/CoPoP as a promising platform for large livestock, but the core findings tell a more cautious story: no detectable cell-mediated immunity in cattle, and humoral responses below the ISA206VG benchmark. The manuscript should be reframed around what was shown: QS21-containing formulations improve antibody responses over CoPoP alone, but the platform in its current form does not yet meet the immunological standard set by adjuvanted soluble p67C in cattle. This is the main result.

Response 1: Three sections of the manuscript have been reframed and improved following the reviewer’s suggestion: the abstract, the introduction, and the discussion.

Comment 2:  Cattle received 70 µg p67C / 500 µg CoPoP / 200 µg adjuvant per dose; mice received 2 µg / 8 µg / 3.2 µg. No mg/kg or allometric rationale is provided for either species. The authors acknowledge the need to optimize dosing for large animals, but this should be stated as a limitation of the current design, not just a future direction. The difference in immunogenicity between species may reflect suboptimal dosing in cattle rather than a species biology effect, and this alternative interpretation should be discussed.

Response 2: The interpretation of the disparity of results between mice and cattle has been comprehensively discussed in the new version of the manuscript.

Lines 517- 561:

“The significant disparity in antibody titres, despite a 62.5-fold increase in the dose of adjuvanting molecules and 35-fold increase in antigen dose in cattle, is likely to be multifactorial. First, it is important to consider the large disproportion in the immunomodulator-to-body-volume ratio. The dose of immunomodulators used in mice can easily reach immune saturation, while in cattle, the dose used is diluted in a larger volume (approximately 128 mg/kg in mice and 0.8 mg/kg in cattle of PHAD or QS-21) [32,33]. In line with this, the nanoparticles were injected intramuscularly in both species, and while the distance to the nearest draining lymph node is negligible in mice, it is considerable in cattle. The nanoparticles can become trapped in the dense muscle tissue, exacerbating the dilution factor of both the immunomodulators and the antigen [34,35]. These findings underscore the capital importance of optimizing the dose of antigen and immunomodulators for use in large livestock animals.

At the molecular level, the suboptimal performance of p67C-CP (PHAD-only formulation) in cattle might indicate species-specific limitations in the TLR4 ligand recognition and signalling pathway. Structural divergence in the TLR4/MD-2 (myeloid differentiation protein 2) complex across species can significantly alter ligand-binding affinity. Specifically, sequence variations in the MD-2 may diminish its capacity to recognise PHAD in cattle compared to mice, hence decreasing the intensity of the signal [36–38]. Furthermore, while the TLR4/MD-2/CD14 complex is essential for optimal activation in the bovine model [37,39], murine studies suggest that CD14 might be dispensable for efficient signalling in certain contexts, highlighting a less restrictive TLR4 activation in mice than in cattle [40]. Importantly, both species might differ in the downstream pathway activation. Although both MyD88-dependent and -independent cascades are present in both species, the MyD88-dependent cascade appears to play a more dominant role in bovine immunity [36,38,41], and the activation of a specific pathway is ligand specific [42]. Collectively, these differences highlight the need for further investigation to identify the most effective TLR4 agonist for maximising immunogenicity in cattle.

On the contrary, QS-21 might have a more universal adjuvant capacity by directly stimulating the NLRP3 inflammasome upon the loss of cellular homeostasis, without the need for receptor-ligand interaction [43–45]. Thus, QS-21 might be able to generate a broader and more potent signal, which is translated into higher antibody titres than PHAD. However, the formulation is lacking the synergistic effect that QS-21 and PHAD often have when administered as an immunostimulatory combo, as it happens after administration of antigens combined with AS01 adjuvant [43,46].

Although saponin mode of action has been extensively studied in mice and human models, the cellular targets and signalling pathways triggered by saponins in cattle remain poorly defined. Our results suggest that a similar signal cascade might be stimulated. Bovine PBMCs stimulated in vitro for 24 h with increasing concentrations of Quil-A exhibited a transcriptional profile indicative of NLRP3 inflammasome activation [43,44,46]. Key signature pro-inflammatory cytokines, including IL-1α, IL-1β, and TNF-α, were markedly upregulated. Notably, the anti-inflammatory cytokine IL-10 was also slightly induced, likely as a regulatory mechanism to counterbalance the heightened inflammatory response. Together, these results indicate that saponins can activate conserved innate immune pathways across species, overcoming the differences in the genetic background in a way that TLR4 agonist, PHAD, could not.  

Comment 3:  The comparison to soluble p67C + ISA206VG is based on data from a separate experiment, not a concurrent control group. Given the high inter-animal variability in outbred cattle, statements like "2-fold lower" or "approaching the level of" should either be replaced with informal qualitative language or accompanied by an explicit caveat about the limitations of cross-experiment comparisons.

Response 3: We agree that cross-experiment comparisons in outbred cattle should be interpreted with caution due to inherent inter-animal variability. Our intention was not to make a strict quantitative comparison, but rather to provide context regarding the relatively modest magnitude of the immune response observed with the PHAD-only formulation. For this reason, we used cautious wording such as “at least 2-fold lower” or “approaching the level” in reference to previously reported data with soluble p67C (s-p67C) and other nanoparticles, which represents one of the less potent formulations described to date. We consider the language to be more qualitative than quantitative.

Comment 4:  The PBMC stimulation experiment uses Quil-A (not QS21), a single 24-hour timepoint, and three naive donors. The pro-inflammatory cytokine signature observed is interesting but does not explain the absence of cellular responses in vivo. This section should be clearly framed as hypothesis-generating only.

Response 4: The aim of the in vitro experiment was not to shed light on the absence of cellular response in vivo. The aim was to evaluate the type of innate response triggered by saponins on bovine PBMCs to support the hypothesis that saponins might stimulate the immune response through the inflammasome pathway in this animal model, as suspected in the in vivo experiment.

A new section in the discussion was added to clarify this point:

Lines 544-561:

“On the contrary, QS-21 might have a more universal adjuvant capacity by directly stimulating the NLRP3 inflammasome upon the loss of cellular homeostasis, without the need for receptor-ligand interaction [43–45]. Thus, QS-21 might be able to generate a broader and more potent signal, which is translated into higher antibody titres than PHAD. However, the formulation is lacking the synergistic effect that QS-21 and PHAD often have when administered as an immunostimulatory combo, as it happens after administration of antigens combined with AS01 adjuvant [43,46].

Although saponin mode of action has been extensively studied in mice and human models, the cellular targets and signalling pathways triggered by saponins in cattle remain poorly defined. Our results suggest that a similar signal cascade might be stimulated. Bovine PBMCs stimulated in vitro for 24 h with increasing concentrations of Quil-A exhibited a transcriptional profile indicative of NLRP3 inflammasome activation [43,44,46]. Key signature pro-inflammatory cytokines, including IL-1α, IL-1β, and TNF-α, were markedly upregulated. Notably, the anti-inflammatory cytokine IL-10 was also slightly induced, likely as a regulatory mechanism to counterbalance the heightened inflammatory response. Together, these results indicate that saponins can activate conserved innate immune pathways across species, overcoming the differences in the genetic background in a way that TLR4 agonist, PHAD, could not.”   

Minor comments

Comment 5:  Supplementary Table S2. The row labeled "CoPoP + QS21 (CP)" lists PHAD as a component, and "CoPoP + PHAD (CQ)" lists QS21 — the reverse of the nomenclature used throughout the manuscript (CP = CoPoP+PHAD; CQ = CoPoP+QS21). This should be corrected.

Response 5: This has been corrected.

Comment 6: It is stated that Kruskal–Wallis was used for cattle group comparisons, but no post-hoc test is specified. Additionally, the results section describes differences "from day 42 to the end of the experiment," while the statistical methods section only addresses day 42.

Response 6: Thank you for pointing out this mistake, this has now been corrected in lines 338-339: “, followed by a pairwise comparison using Dunn’s multiple comparison test with Bonferroni correction.”

We have removed the sentence “from day 42 to the end of the experiment”.

Round 2

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

The authors have addressed all of my comments.

Reviewer 3 Report (New Reviewer)

Comments and Suggestions for Authors

The authors have satisfactorily addressed all the comments raised by this reviewer.

Comments on the Quality of English Language

Minor language editing would improve the readability of the manuscript

Reviewer 4 Report (New Reviewer)

Comments and Suggestions for Authors

Accept in a current form

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Overall the paper was well written and there are very few comments to make   A - Typographical errors   Introduction - sentence on lines 65-70; there is something here that needs editing as the sentence as it currently reads doesn't make complete sense. - sentence on lines 74-76; there is something here that needs editing as the sentence as it currently reads doesn't make complete sense ('have' in front of 'extended shelf life'??) - sentence on lines 84-88; there is something here that needs editing as the sentence as it currently reads doesn't make complete sense. - line 132; change 'large' to 'target'   Materials and Methods Section 2.4 - line166; I assume this was his-tagged 67C protein? Section 2.4 - line 176; change ', occurring when p67C-DY490 binds to the liposomes,' to '(quenching occurs when p67C-DY490 binds to the liposomes)'   B - Minor scientific questions - section 2.5; ratio of p67C and CoPoP liposomes appears to be 1:5 here rather than 1:4 that is cited in section 2.3 and section 3.3 - can the authors comment on the significance (if any) of this? - section 3.2; what are the author's thoughts in the large variation in antibody titres observed between individuals in the same experimental groups  - section 4 (discussion); last paragraph - can the authors really justify the statement that the SNAP platform is 'an exceptionally appealing solution for the delivery of subunit vaccines that can be administered repeatedly throughout the life span of high-value animals, such as cattle' - a critical evaluation of the data would suggest that the platform i) didn't induce T-cell responses and ii) induced low and highly variable levels of antibody responses. I am not sure that aligns with the output you would want from a vaccine delivery platform!   Summary  - well written and performed study, minimal corrections required. - not sure I agree with the final statement!

Author Response

General comments about the manuscript
Overall the paper was well written and there are very few comments to make.

Comment 1: sentence on lines 65-70; there is something here that needs editing as the
sentence as it currently reads doesn't make complete sense.
Response 1: Corrected: now lines 67-73: “The Spontaneous Nanoliposome Antigen
Particleization (SNAP) vaccine platform utilises liposomes incorporating metal-chelating
lipids to rapidly capture and display His-tagged proteins in a multimeric form. The
technology offers a rapid plug-and-play system for the delivery of multimerized antigens,
an alternative to less plastic systems, like virus-like particle (VLPs), ferritin or Outer
Membrane Vesicles (OMVs) [10]. The technology enables particle formation without the
need for chemical conjugation or deep knowledge of the protein/antigen structure [11].”

Comment 2: sentence on lines 74-76; there is something here that needs editing, as the
sentence as it currently reads doesn't make complete sense ('have' in front of 'extended shelf
life'??)
Response 2: This has been corrected, now lines 76-77: “Remarkably CoPoP-liposomes have
been demonstrated to be very stable after injection [12] and have extended shelf life after
lyophilization [13].

Comment 3: sentence on lines 84-88; there is something here that needs editing, as the
sentence as it currently reads doesn't make complete sense. - line 132; change 'large' to
'target'.
Response 3: We have modified the sentence to adjust, now lines 89-90: “However, the
potential of the technology in large livestock animals remains to be evaluated.”

Comment 4: Section 2.4 - line166; I assume this was his-tagged 67C protein?
Response 4: Yes, this has been corrected, now lines 172-173.

Comment 5: Section 2.4 - line 176; change ', occurring when p67C-DY490 binds to the
liposomes,' to '(quenching occurs when p67C-DY490 binds to the liposomes)'.
Response 5: We have corrected this section to make it more understandable. Now lines 181-
186: “Samples were diluted 10-fold before measuring. Fluorescence quenching occurs when
p67C-DY490 binds to the liposomes, which was measured using a Tecan Safire microplate
fluorescence reader (excitation wavelength at 490 nm, emission wavelength at 515 nm). The
results are presented as the fluorescence intensity (FI) of the liposome-bound p67C-DY490
compared with the fluorescence of free antigen in PBS, using the formula: [1-FIp67CDY490_
liposomes/FI free p67C-DY490 in PBS]X100”

Comment 6: section 2.5; ratio of p67C and CoPoP liposomes appears to be 1:5 here rather
than 1:4 that is cited in section 2.3 and section 3.3 - can the authors comment on the
significance (if any) of this?
Response 6: Thank you for pointing this out. We used a 1:4 ratio of p67C to CoPoP in mice
and 1:5 ratio of p67C to CoPoP in cattle, as this higher ratio also resulted in higher adjuvant
doses.

Comment 7: section 3.2; what are the author's thoughts in the large variation in antibody
titres observed between individuals in the same experimental groups.
Response 7: We have been observing these variations when using p67C in both cattle breeds
we have been historically working with (Bos taurus and Bos indicus). It was not discussed in
detail in this manuscript but we hypothesise it might have a genetic component on the
MHC-II (BoLa-II for cattle). We modified lines 490-493 to capture this hypothesis: “Although
the antibody response in cattle was low and a large variation between animals was
observed, most probably due to genetic differences in the BoLa type II (bovine major
histocompatibility complex II), it is undeniable that the use of the refined saponin QS21 had
a positive effect on antigen-specific antibody production.”

Comment 8: section 4 (discussion); last paragraph - can the authors really justify the
statement that the SNAP platform is 'an exceptionally appealing solution for the delivery of
subunit vaccines that can be administered repeatedly throughout the life span of high-value
animals, such as cattle' - a critical evaluation of the data would suggest that the platform i)
didn't induce T-cell responses and ii) induced low and highly variable levels of antibody
responses. I am not sure that aligns with the output you would want from a vaccine delivery
platform!
Response 8: This is a good point, however the mouse data shows the potential for the
approach, underscoring challenges to transition vaccine approaches from small to large
animals. Claims and statements have been adapted to the presented data.
- Lines 42-45 in the Abstract: “Our findings confirm the potential of SNAP as a
versatile vaccine delivery system, but improving the dosage and the type of
immunomodulatory adjuvants included in the CoPoP particle would be key for
advancement in the application of this technology to large livestock animals.”
- Lines 135-139 Introduction: “Overall, the SNAP platform shows potential as a
versatile and easy-to-formulate vaccine delivery system. However, improving the
dosage and the type of immunomodulatory adjuvants included in the CoPoP
particle would be key for advancement in the application of this technology to large
livestock animals, particularly for complex pathogens such as Theileria parva, the
causative agent of East Coast fever (ECF).” And
- Lines 566-579 in the discussion: “Even though p67C remains a promising antigen,
there are other regions within p67 with protective potential (reviewed in [17]), and
we have further identified several additional antigens with neutralising capacity [43].
In line with this, and despite the fact that the formulation of SNAP for large livestock
species implementation needs further research, it would be of special interest to
assess the performance of the technology on the delivery of multivalent vaccines
against East Coast fever. Improving the dosage and the type of immunomodulators
included in the CoPoP would be key for advancement in the application of this
technology to large livestock animals. A series of comprehensive in vitro and in vivo
studies is imperative to identify the most effective immunostimulants in camle. The
versatility of SNAP technology, functioning as a plug-and-play system for Histagged
proteins, combined with the lack of reactivity against the liposomal carrier,
still positions it as a practical approach for delivering subunit vaccines that require
repeated administration in high-value livestock such as camle, including for
neglected diseases like ECF.”

Comment 9: Summary - well written and performed study, minimal corrections required. -
not sure I agree with the final statement!
Response 9: We have toned down the final sentence as follows (lines 575-579): “The
versatility of SNAP technology, functioning as a plug-and-play system for His-tagged
proteins, combined with the lack of reactivity against the liposomal carrier, still requires
investigation as a practical approach for delivering subunit vaccines that require repeated
administration in high-value livestock such as cattle, including for neglected diseases like
ECF.”

Reviewer 2 Report

Comments and Suggestions for Authors

I appreciate the authors’ effort to explore the SNAP/CoPoP platform for delivering the p67C antigen in cattle and mice, addressing an important challenge in East Coast fever vaccine development. However, after a thorough evaluation, I find that the manuscript in its current form does not meet the standards for publication in Vaccines. While the concept is innovative, the experimental design, data interpretation, and overall strength of evidence are insufficient to support the conclusions drawn. The lack of critical controls limited statistical rigor, and absence of functional validation significantly undermine the impact and reliability of the findings. For these reasons, I recommend rejection rather than revision.

Major Comments

1. Absence of an Appropriate Positive Control

   • The study compares CoPoP formulations to historical data on s-p67C with Montanide ISA206VG rather than including this as an in-study control (Section 3.2, Discussion lines 466–472). This omission makes it impossible to fairly assess whether the CoPoP platform offers any advantage over established approaches.

2. Insufficient Evidence of Efficacy in the Target Species

   • The cattle data show modest antibody responses only when QS21 is included (Groups 3 and 4, Figure 2C–E), and these titers remain at least two-fold lower than those achieved with conventional adjuvants in previous studies. No functional assays (e.g., sporozoite neutralization) were performed to demonstrate biological relevance, which is critical for a vaccine study.

3. Failure to Deliver on Key Claims

   • The abstract and introduction emphasize the potential of SNAP to enhance both humoral and cellular immunity (lines 30–31), yet no detectable T-cell response was observed in cattle (Figure 3). This discrepancy is not adequately addressed in the discussion, and the conclusions remain overstated.

4. Overreliance on Murine Data

   • The strong antibody response in mice (Figure 2F) is highlighted as evidence of platform potential, but this is not translatable to cattle. The discussion does not sufficiently temper these claims or acknowledge species-specific differences in adjuvant responsiveness.

5. Limited Statistical Power and Transparency

   • With only five animals per group (Table 1) and high variability in responses, the study is underpowered. While non-parametric tests are mentioned (Section 2.11), no effect sizes or confidence intervals are reported, and multiple comparison corrections are absent. This weakens confidence in the reported significance.

6. Critical Gaps in Platform Characterization

   • Despite emphasizing thermostability and scalability (Introduction, lines 50–52), the manuscript provides no data on liposome stability post-antigen binding, after lyophilization, or under field-relevant conditions. These are essential to substantiate the practical value of the platform.

7. Mechanistic Insights Are Superficial

   • The in vitro PBMC stimulation with Quil-A (Section 3.5, Figure 5) is disconnected from the in vivo findings and does not clarify how QS21 within CoPoP liposomes influences immune responses in cattle.

8. Economic and Practical Feasibility Not Addressed

   • The introduction frames the work as relevant for LMIC settings, yet the discussion omits any consideration of cost, scalability, or regulatory feasibility for QS21/PHAD-based formulations in livestock vaccines.

Minor Comments

• Abstract: The phrase “needs optimization” (line 42) is vague; specify which parameters require optimization.
• Figures: Figure 2 is cluttered; consider moving individual animal kinetics to supplementary material.
• Terminology: Ensure consistent naming of formulations (e.g., p67C-CPQ vs. CPQ).
• Grammar: Minor typos (e.g., “ration” instead of “ratio” in Section 2.5).
• References: Heavy reliance on a few prior studies; broaden the literature base.
• Supplementary Data: Table S1 is referenced but not summarized in the main text.

Comments on the Quality of English Language

The quality of English in the manuscript is generally good, with clear sentence structure and appropriate scientific terminology. However, there are several areas that need attention:

• Minor grammatical issues: For example, “ration” instead of “ratio” in Section 2.5, and occasional missing articles.
• Consistency in terminology: Formulation names (e.g., p67C-CPQ vs. CPQ) should be standardized throughout.
• Stylistic clarity: Some sentences in the Introduction and Discussion are overly long and could be broken down for better readability.
• Abstract and Discussion: Phrases like “needs optimization” are vague and should be replaced with more precise language.

Author Response

General comments about the manuscript
I appreciate the authors’ effort to explore the SNAP/CoPoP platform for delivering the p67C
antigen in cattle and mice, addressing an important challenge in East Coast fever vaccine
development. However, after a thorough evaluation, I find that the manuscript in its current
form does not meet the standards for publication in Vaccines. While the concept is
innovative, the experimental design, data interpretation, and overall strength of evidence
are insufficient to support the conclusions drawn. The lack of critical controls limited
statistical rigor, and absence of functional validation significantly undermine the impact and
reliability of the findings. For these reasons, I recommend rejection rather than revision.

Comment 1: Absence of an Appropriate Positive Control. The study compares CoPoP
formulations to historical data on s-p67C with Montanide ISA206VG rather than including
this as an in-study control (Section 3.2, Discussion lines 466–472). This omission makes it
impossible to fairly assess whether the CoPoP platform offers any advantage over
established approaches.
Response 1: While this would be a useful control, our research group has generated
abundant data using p67C as a candidate antigen in two cattle breeds, Bos taurus and Bos
indicus, from several experiments. All results reported in peer-reviewed publications,
accessible by the research community in an open-access fashion. Instead, we assessed 4
groups of cattle with various adjuvant components, thereby establishing the importance of
saponin and the relative inefficiency of MPLA in cattle, which we believe will be useful to
the research community.

Comment 2: Insufficient Evidence of Efficacy in the Target Species. The cattle data show
modest antibody responses only when QS21 is included (Groups 3 and 4, Figure 2C–E), and
these titers remain at least two-fold lower than those achieved with conventional adjuvants
in previous studies. No functional assays (e.g., sporozoite neutralization) were performed to
demonstrate biological relevance, which is critical for a vaccine study.
Response 2: We have adjusted our statements to align with the data generated in this study
throughout the whole manuscript. Although the best-performing SNAP–CoPoP particles
induced antibody titres, these remained approximately two-fold lower than those obtained
with soluble p67C formulated with Montanide ISA206VG. Given this and considering that
the Theileria parva sporozoite seroneutralization assay is both costly and highly labourintensive,
we did not include it in the current work. In addition, we are not implying that
the immune responses observed here will directly correlate with protection against East
Coast fever. While functional assays are valuable, they are not essential when p67C is used
primarily as a model antigen to evaluate the ability of the SNAP platform to elicit immune
responses in large livestock species.

Comment 3: Failure to Deliver on Key Claims. The abstract and introduction emphasize the
potential of SNAP to enhance both humoral and cellular immunity (lines 30–31), yet no
detectable T-cell response was observed in cattle (Figure 3). This discrepancy is not
adequately addressed in the discussion, and the conclusions remain overstated.
Response 3: The statements have been adapted to the presented data throughout the whole
manuscript. Please see examples below in comment 4.

Comment 4: Overreliance on Murine Data. The strong antibody response in mice (Figure 2F)
is highlighted as evidence of platform potential, but this is not translatable to cattle. The
discussion does not sufficiently temper these claims or acknowledge species-specific
differences in adjuvant responsiveness. =
Response 4: We have stated in several paragraphs the importance of not translating the
results from one species to another one and also the need to optimise the type of
immunostimulants and the doses to the target species. Moreover, we have toned down the
claims and statements related to the potency of the SNAP technology in cattle.
- Lines 473-478: “In ca&le, compared to our historical data, CoPoP liposomes did
not perform be&er than soluble protein (s-p67C) or previously evaluated
nanoparticle technologies that were delivered in the presence of commercial
emulsion adjuvant (Montanide ISA206 VG, Seppic) [19,20]. Antibody responses
were a minimum of 2-fold lower when using the best CoPoP-liposome formulation
that included QS21 saponin (p67C-CQ and p67C-CPQ), and no cellular response
could be detected with any of the evaluated CoPoP-liposomes.”
- Lines 486-489: “These findings underscore the capital importance of optimising the
dose and composition of the adjuvants and/or immunostimulants used for the
delivery of antigens to the target species, and the limitations of extrapolating
successes from one species to another.”
- Lines 466-579: “Even though p67C remains a promising antigen, there are other
regions within p67 with protective potential [17] reviewed in (Nene et al., 2016)), and
we have further identified several additional antigens with neutralising capacity [43].
In line with this, and despite the fact the performance of SNAP in large livestock
species needs optimization, it would be of special interest to assess the performance
of the technology on the delivery of multivalent vaccines against East Coast fever.
Improving the dosage and the type of immunomodulators included in the CoPoP
would be key for advancement in the application of this technology to large livestock
animals. A series of comprehensive in vitro and in vivo studies is imperative to
identify the most effective immunostimulants in camle. The versatility of SNAP
technology, functioning as a plug-and-play system for His-tagged proteins,
combined with the lack of reactivity against the liposomal carrier, still positions it as
a practical approach for delivering subunit vaccines that require repeated
administration in high-value livestock such as camle, including for neglected diseases
like ECF.”

Comment 5: Limited Statistical Power and Transparency. With only five animals per group
(Table 1) and high variability in responses, the study is underpowered. While nonparametric
tests are mentioned (Section 2.11), no effect sizes or confidence intervals are
reported, and multiple comparison corrections are absent. This weakens confidence in the
reported significance.
Response 5: We are aware the statistical power will be a minor component of the presented
data. The main objective of the experimental design was to evaluate the capacity of different
formulations of SNAP to trigger an immune response in large livestock animals with
minimal animal suffering. Increasing the number of animals would not have made the
immune response better and would only have increased animal stress and suffering, and
eventually reach the same conclusions. The statistical analysis included in the manuscript is
enough for the conclusions reached.

Comment 6: Critical Gaps in Platform Characterization. Despite emphasizing
thermostability and scalability (Introduction, lines 50–52), the manuscript provides no data
on liposome stability post-antigen binding, after lyophilization, or under field-relevant
conditions. These are essential to substantiate the practical value of the platform.
Response 6: We acknowledge that we have not discussed in great detail the costeffectiveness
of SNAP for livestock vaccines. SNAP is still a technology applied in research
areas where cost of goods is not a limiting factor. To address this challenge we have
included a paragraph in the discussion in lines 548-553: “Particular emphasis should be
placed on identifying formulations that are not only effective in enhancing immune
responses but also economically viable and scalable for use in livestock species. Such efforts
would help ensure that promising vaccine platforms, like SNAP, can be translated into
practical tools for livestock vaccinology, where cost, accessibility and ease of deployment are
critical factors for successful implementation [41,42].” The stability of the CoPoP
formulations with the same immunmodulators has been reported previously in Huang et al.
(2018, Nature Nanotechnology, doi:10.1038/s41565-018-0271-3), and we included
information and references in the introduction lines 76-77 “Remarkably CoPoP-liposomes
have been demonstrated to be very stable after injection [12] and have an extended shelf life
after lyophilization [13].”

Comment 7: Mechanistic Insights Are Superficial. The in vitro PBMC stimulation with Quil-
A (Section 3.5, Figure 5) is disconnected from the in vivo findings and does not clarify how
QS21 within CoPoP liposomes influences immune responses in cattle.
Response 7: Thanks to the in vitro stimulation of bovine PBMCs with Quil-A, we gained
insight into the nature of the induced response, which appears to be predominantly
inflammatory, a mechanism commonly exploited by adjuvants to enhance antibody
production. This section of the manuscript helps clarify how saponins may support the
induction of p67C-specific immune responses and shows that the magnitude of the response
increased in a dose-dependent manner with the amount of saponin used for stimulation.

Comment 8: Economic and Practical Feasibility Not Addressed. The introduction frames the
work as relevant for LMIC settings, yet the discussion omits any consideration of cost,
scalability, or regulatory feasibility for QS21/PHAD-based formulations in livestock
vaccines.
Response 8: We appreciate this point and agree that economic, scalability, and
practical/regulatory feasibility are critical for translation of livestock vaccines. We have
revised the Discussion to explicitly address manufacturability and deployment
considerations for the SNAP platform and the QS-21/PHAD-based adjuvant package. Lines
553-557: “In this context, SNAP is positioned for practical translation, with CoPoP
manufacturing demonstrated at the 100 g scale, supporting cost-conscious scale-up for
livestock use. QS-21 can be addressed through established sourcing and dose-optimization
strategies to support cost-conscious livestock deployment.”

Comment 9: Abstract: The phrase “needs optimization” (line 42) is vague; specify which
parameters require optimization.
Response 9: Corrected, lines 43-45: “…but improving the dosage and the type of
immunomodulators included in the CoPoP particle would be key for the advancement in
the application of this technology to large livestock animals.”

Comment 10: Figures: Figure 2 is cluttered; consider moving individual animal kinetics to
supplementary material.
Response 10: For clarity we think it’s better to keep the individual animal panels and a
summary of results in panel E.

Comment 11: Terminology: Ensure consistent naming of formulations (e.g., p67C-CPQ vs.
CPQ).
Response 11: Corrected.

Comment 12: Grammar: Minor typos (e.g., “ration” instead of “ratio” in Section 2.5).
Response 12: Corrected.

Comment 13: References: Heavy reliance on a few prior studies; broaden the literature base.
Response 13: We have made considerable efforts to broaden the references considering we
are few groups working with SNAP and East Coast fever.

Comment 14: Supplementary Data: Table S1 is referenced but not summarized in the main
text.
Response 14: Table S1 now Table S2 is references in line 202.

Comment 15: The quality of English in the manuscript is generally good, with clear sentence
structure and appropriate scientific terminology. However, there are several areas that need
attention:
• Minor grammatical issues: For example, “ration” instead of “ratio” in Section 2.5,
and occasional missing articles. Corrected.
• Consistency in terminology: Formulation names (e.g., p67C-CPQ vs. CPQ) should be
standardized throughout. Corrected.
• Stylistic clarity: Some sentences in the Introduction and Discussion are overly long
and could be broken down for better readability. Corrected.
• Abstract and Discussion: Phrases like “needs optimization” are vague and should be
replaced with more precise language. Corrected.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors evaluated four different preparations of SNAP platform with p67C model antigen with or without adjuvants in cattle and mice. After two IM injections, the authors found p67-CQ and CPQ induced significant antibody responses but no cellular immune responses. The manuscript is well-written and easy to follow but requires some clarifications before publication. Specific comments follow.

Major points:

1. Line 136, Material and Methods: Please indicate details of reagents (Source and catalogue number) so that the readers can follow your methods.
2. Do the authors have stability data of the SNAP vaccines?
3. Can the authors determine IgG sub class ELISA?
4. Figure 2E: Can the authors draw graphs to show individual animal like cattle graphs? I’d like to see how many mice responded in each group.

Minor points:

1. Line 333, Figure 1A: “CoPOP” should be “CoPoP”.
2. Line 333, Figure 1C: “As01” should be “AS01”.

Author Response

General comments about the manuscript
The authors evaluated four different preparations of SNAP platform with p67C model
antigen with or without adjuvants in cattle and mice. After two IM injections, the authors
found p67-CQ and CPQ induced significant antibody responses but no cellular immune
responses. The manuscript is well-written and easy to follow but requires some clarifications
before publication. Specific comments follow.

Comment 1: Line 136, Material and Methods: Please indicate details of reagents (Source and
catalogue number) so that the readers can follow your methods.
Response 1: In material and methods we are providing enough information for the
experiments to be repeated. The manufacturer name is listed in all the reagents and the
clone name is identified for the antibodies used. Moreover, the names of the reagents match
the name of the manufacturer’s. If more information is required by a reader a more detailed
protocol can be shared by the authors.

Comment 2: Do the authors have stability data of the SNAP vaccines?
Response 2: No, we did not perform in vitro stability experiments with the p67C-CoPoP
formulations. However, previous studies have demonstrated stability of the SNAP platform
at 4 °C. In particular, Huang et al. (2018, Nature Nanotechnology, doi:10.1038/s41565-018-
0271-3) showed that CoPoP/PHAD liposomes with pre-bound antigen remain stable during
storage.

Comment 3: Can the authors determine IgG sub class ELISA?
Response 3: The results using CoPoP with p67C are not as increased compared to protein
with adjuvant, as previously used. Thus, expanding on the ex vivo analysis of samples won’t
be of interest.

Comment 4: Figure 2E: Can the authors draw graphs to show individual animal like cattle
graphs? I’d like to see how many mice responded in each group.
Response 4: In the mice experiment only two time points were evaluated, before
immunization when the antibody titers were undetectable and at day 42, which is the data
presented in the graph in Figure 2 panel F as individual animals (dots) and the mean and SD
for the group at this time point.

Comment 5: Line 333, Figure 1A: “CoPOP” should be “CoPoP”.
Response 5: Corrected.

Comment 6: Line 333, Figure 1C: “As01” should be “AS01”.
Response 6: Corrected.