C-terminal truncation of CXCL10 attenuates inflammatory activity but retains angiostatic properties

  1. Luna Dillemans
  2. Karen Yu
  3. Alexandra De Zutter
  4. Sam Noppen
  5. Mieke Gouwy
  6. Nele Berghmans
  7. Mirre De Bondt
  8. Lotte Vanbrabant
  9. Stef Brusselmans
  10. Erik Martens
  11. Dominique Schols
  12. Pedro Elias Marques
  13. Sofie Struyf
  14. Paul Proost

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Interferon-γ-inducible protein of 10 kDa (IP-10/CXCL10) is a dual-function CXC chemokine that coordinates chemotaxis of activated T cells and natural killer (NK) cells via interaction with its G protein-coupled receptor (GPCR), CXC chemokine receptor 3 (CXCR3). As a consequence of natural posttranslational modifications, human CXCL10 exhibits a high degree of structural and functional heterogeneity. However, the biological effect of natural posttranslational processing of CXCL10 at the carboxy (C)-terminus has remained partially elusive. The truncated CXCL10 proteoform CXCL10 (1-73) , lacking the four endmost C-terminal amino acids, was previously identified in human cell culture supernatant. To further explore the functioning of CXCL10 (1-73) , we optimized its production in this study through Fmoc-based solid phase peptide synthesis (SPPS) and propose an SPPS strategy to efficiently generate human CXCL10 proteoforms. Compared to intact CXCL10 (1-77) , CXCL10 (1-73) had diminished affinity for glycosaminoglycans including heparin, heparan sulfate and chondroitin sulfate A. Moreover, CXCL10 (1-73) exhibited an attenuated capacity to induce CXCR3A-mediated signaling, as evidenced in calcium mobilization assays and through quantification of phosphorylated extracellular signal-regulated kinase-1/2 (ERK1/2) and protein kinase B/Akt. Furthermore, CXCL10 (1-73) incited reduced primary human T lymphocyte chemotaxis in vitro and evoked less peritoneal ingress of CXCR3 + T lymphocytes in mice receiving intraperitoneal chemokine injections. In contrast, loss of the four endmost C-terminal residues did not affect the inhibitory properties of CXCL10 on spontaneous and/or FGF-2-induced migration, proliferation, wound healing, phosphorylation of ERK1/2, and sprouting of human microvascular endothelial cells. Thus, C-terminally truncated CXCL10 has attenuated inflammatory properties, but preserved anti-angiogenic capacity.

GRAPHICAL ABSTRACT

Article activity feed

  1. Review Commons

    Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    The authors do not wish to provide a response at this time.

  2. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    The manuscript describes the strategy to efficiently synthesize a natural truncated version of the chemokine CXCL10 that lacks the last 4 amminoacids. In addition, it describes the biological activities of the CXCL10 truncated version (1-73) compared to the full length chemokine (1-77). By performing in vitro and in vivo experiments, authors have found that CXCl10 1-73 is not able to induce signalling and chemotaxis of CXCR3 expressing cells such as T lymphocytes. In addition, this C terminal truncated version does not bind GAGs while retains angiostatic activity, blocking migration and proliferation of endothelial cells.
    The paper is written very well, results are presented in a very logical sequence.

    Major comment

    The in vivo experiments shown in supplementary figures 7 and 8 are not significant and I suggest removing them from the manuscript.

    Minor comment

    In figure 9D authors showed the in vivo migration of CXCR3 positive T lymphocytes in the peritoneal cavity. However, the gating strategy showed in supplementary figure 6 is showing all the leukocytes CXCR3 positive. Please clarify.

    Significance

    The manuscript describes the biological activity of a truncated version of CXCL10 a very important chemokine that recruit Th1 lymphocytes and NK cells. The C terminal truncated version of CXCL10 is naturally occurring, but its functions were never described until now.

    The strength of the manuscript is the precise description of the synthesis and of the in vitro biological functions of the truncated CXCL10.

    For this reason, these results are of interest not only for a specialized audience working in the chemokine field, but also for a more broad audience for the development of an inhibitor of CXCR3 or for an angiostatic molecule.

  3. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    Summary:

    It has been reported that CXCL10 has several truncated forms (proteoforms) with different C-terminal truncation states, each with different functions. The authors have discovered and reported an efficient peptide synthesis method for CXCL10(a.a. 1-73) proteoform. The authors indicated that they could synthesize CXCL10(a.a. 1-73) proteoform consistent with the known functions of natural CXCL10(a.a. 1-73). The synthesized CXCL10(a.a. 1-73) successfully indicated the reduced effects on lymphocyte migration and similar effects on angiogenesis. These findings open the way for detailed functional analysis of CXCL10 (a.a. 1-73), which has been difficult to study in vivo and has potential for therapeutic use. However, as discussed below, the authors have made several statements that confuse their findings with the discoveries made by previous studies. The Introduction is a typical example of this. Also, there are several major issues noted in the next section.

    Major comments:

    1. Figure 4: Possibly an overinterpretation of results in CXCR3A overexpressing model cells
      The authors build their logic for the entire paper by drawing conclusions based on their assumption that CXCR3A overexpression model is equivalent to physiological lymphocytes and endothelial cells. However, CXCR3 has isoforms, including CXCR3A/B/alternative. They have different effects on cell proliferation and migration. Expression levels of CXCR3A/3B may vary among cell types and microenvironments. In addition, the downstream signals pAKT and pERK of CXCR3A/B are regulated by various regulatory factors. Therefore, it is important to perform the experiments shown in Figure 4 with primary lymphocytes and vascular endothelial cells, which are the subject of this paper. Based on the data presented by the authors, experiments with Primary lymphocytes and Endothelial cells would not be difficult.

    2. Figure 5: "In line with the observation of the signaling assays, COOH-terminal processing of CXCL10 also significantly diminishes its chemotactic properties on primary CXCR3+ T lymphocytes."
      The authors draw the conclusions described above from the results in Figure 4 and Figure 5. In other words, authors excluded other possibilities without data.
      In Figure 5, the Chemotaxis assay was performed in Transwells with 5 miro-meter pores pre-coated with fibronectin. CXCL10 is also known to interact with fibronectin. This suggests that, potentially, the interaction with fibronectin may be important for CXCL10 gradient formation on the transwell. However, interaction data between CXCL10(a.a. 1-73) proteoform and fibronectin is not shown. This information is essential in the interpretation of Figure 5 results.
      The authors should consider the possibility that readers unfamiliar with this experimental system may be given a false understanding that Chemotaxis shown here is determined solely via CXCR3A.
      Also, please indicate whether the conclusions here are supported in different Pre-coating (e.g. type I collagen, type 4 collagen, human fibronectin). How the activation changes with each Coating here is important information when considering how CXCL10(a.a. 1-73) behaves in the extracellular matrix in vivo. These add to the value of this study and provide important insights for readers to further work with CXCL10(a.a. 1-73).
      Furthermore, the Migration chamber here is pre-coated with bovine serum fibronectin. Please provide Lot and purity information for this Serum derived fibronectin. This is considered important both for the reader to reproduce the data and to interpret the results. Since Bovine serum fibronectin is a different species than human CXCL10 (a.a. 1-73), in order to correctly interpret its contribution to the Chemotaxis assay, it is interactions, respectively, should be evaluated.

    3. Figure 6: Over-interpretation of the results
      From Figure 6, the authors conclude that CXCL10 (a.a. 1-73) has no change in antiangiogenic action based on data on vascular endothelial cell migration and viability. Cell migration and endothelial cell viability are only one aspect of angiogenesis. It is problematic to conclude from these results that there is no change in "antiangiogenic action".
      Also, in Figure 6A, the authors cultured cells in the presence of FGF2 and in the presence of CXCL10 (a.a. 1-73) and CXCL10 (a.a. 1-77) for as long as 49 hours to evaluate Migration. Therefore, the results here include not only pure migration but also its effect on proliferation. However, Figures 6C/6D only show data on the viability of cells, not on the effects on cell proliferation. Therefore, in order to correctly interpret the results, the proliferation of vascular endothelial cells needs to be examined and presented.

    4. Figure 9 and supplemental Figure S6: Gating for T cells (gated as CD3+ NK1.1-) and activated CXCR3+ T cells (gated as CD3+ NK1.1- CXCR3+)
      Supplemental Fig. S6 raises a question as to whether the location of the Gating of CD3 and NK1.1 is correct. Please verify if this gating is proper by presenting Isotype control data as the basis.
      Gating for CXCR3 also seems to be gated in an unnatural position. Please present Isotype controls data and positive control data and explain the basis for this gating.

    5. Figure 9: Over-interpretation of the results
      It would be an oversimplified interpretation of the results here to explain them solely in terms of lymphocyte Migration. The authors should not rule out the possibility that the results obtained here could be due to effects quite different from those shown so far in vitro.
      Conclusions should be drawn after examining the following items

    6. Expression of lymphocyte adhesion-related molecules on the surface of vascular endothelial cells

    7. Effects on Tight junction of blood vessels

    8. Effect on vascular permeability

    If the above data are not presented, the authors should clearly describe that the author's conclusion is just one of the possibilities. The readers should be informed of the above possibilities, and the different potential mechanisms involved so that the readers do not misunderstand that the authors' conclusions are definitive conclusions.

    Minor comments:

    Figure 7C: Please provide higher-resolution images

    The quality and resolution of the images are low and very difficult to see. The image is of such low quality that it is barely possible to determine the presence or absence of cells. Here, providing higher-resolution images is important to give the reader a deeper understanding. The desired resolution is a resolution that allows determining what the Filopodia and Lamelipodia morphology of the cell looks like at the Edge of the Scratch, and how it differs or does not differ between CXCL10 (1-73) and 1-77, etc., desirable. Such an image could underpin the other data in this paper. Furthermore, such detailed forms can give the reader insights into more precise molecular mechanisms. In this sense, it is essential to provide high-quality images.

    Line 360-362, page 12 (Results)

    "Various naturally-occurring COOH-terminally truncated CXCL10 proteoforms were detected in human cell-culture supernatant of IFN-γ-stimulated human diploid skin/muscle-derived fibroblasts and primary human keratinocytes and potential processing enzymes were identified (Suppl. fig. 1). "
    This statement could be interpreted to mean that what is described in Supplemental figure 1 is identified in this paper. Although it is unlikely that most readers would make such a mistake, unnecessary misleading statements should be avoided.

    Line 508-509, page 16 (Discussion)

    "In the present study, we characterized the effects of a natural COOH-terminal truncation of CXCL10, which involves the shedding of the four endmost COOH-terminal amino-acids, on hallmark chemokine properties of CXCL10."

    The authors state that "we characterized the effects of a natural COOH-terminal truncation of CXCL10," which gives the reader the wrong impression.

    "Natural truncation of CXCL10" means physiological CXCL10, which is truncated form that normally occurs in vivo. These findings have been done in prior papers and were not first characterized in this paper. This should be described as a characterization of the synthesized peptide. This sounds like the authors have taken credit for prior studies.

    Figure 6A&6B: What is the "HRMVEs" on the Y-axis? Nowhere in the paper is there a description of this term.

    Figure 8A&8B: Some error bars are only on one side.

    Significance

    It has been reported that the functions of CXCL10 change dynamically in tissues depending on the C-terminal truncation state. However, this dynamic nature created a mixture of each Proteoforms (CXCL10 with different terminal truncation states), making the analysis of their functions difficult. CXCL10(a.a.1-73) is not commercially available like CXCL10(a.a.1-77) due to its difficult peptide synthesis; pure functional analysis of CXCL10(a.a.1-73) could not be performed in vivo. Therefore, the functions of CXCL10(a.a. 1-73) has been mainly reported as circumstantial evidence or in vitro studies using trace amounts of purified product purified using HPLC.

    In this study, the authors clarify the challenges of peptide synthesis and enable the synthesis of more CXCL10(a.a. 1-73). Thereby paving the way for implementing the function of pure CXCL10(a. 1-73) proteoform not only in vitro but also in vivo. It also potentially opens the door for the application of CXCL10(a.a. 1-73) in therapeutic interventions such as tissue repair.

    However, the paper has the problems mentioned above, and it would be desirable to verify and reinforce the reliability and logical development of the conclusions. Reinforcing additional experimental data such as that and validating the derivation of the conclusions would be a study of significance to basic medical researchers in vascular biology, immunology, and tissue repair, as well as to the clinical research community.

    General assessment:

    Strength:

    The authors have discovered and reported a stable method for synthesizing CXCL10 (a.a. 1-73), which has been difficult to synthesize in the past. This may provide researchers a way to solve the problem that it has been difficult to analyze clear molecular mechanisms due to the mixture of diverse CXCL10 proteoforms. The progress reported here may be expected to facilitate other researchers to investigate more detailed molecular mechanisms and explore unknown functions of CXCL10 (a.a. 1-73).

    It is also expected to solve the problem of in vivo analysis of CXCL10(a.a. 1-73) function, which has been impossible due to yield issues. In the future, this synthetic peptide may open the door to a variety of useful applications, such as therapeutic intervention for severe wound healing.

    Advance:

    The authors wrote the "Introduction" and "Abstract" focusing on the functional "discovery" of CXCL10 (a.a. 1-73). This may prevent the readers from understanding the true value of this study. The most significant finding of this study is the technological advance of increasing the yield of CXCL10 (a.a. 1-73), which has been difficult to synthesize to a level that allows in vivo experiments. Although there are many improvements to be made, I believe this is a significant study for the community described above if this synthesized peptide is widely available in the community.

    Audience:

    The current manuscript is suitable for a Specialized audience. If the issues raised here were solved, it might be suitable for broader audiences, including translational/clinical researchers.

    My field of expertise:

    Molecular biology, biochemistry, vascular biology, hematology and cancer

  4. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    The manuscript by Dillemans et al reports that synthesis, purification and functional characterisation of the truncated CXCL10 proteoform CXCL10(1-73). This lacks the four endmost COOH-terminal amino acids . The authors report that compared to the full length CXCL10(1- 77), CXCL10(1-73) had (i) diminished affinity for glycosaminoglycans, (ii) exhibited reduced capacity to induce signaling events (e.g. calcium mobilization as well as ERK and Akt phosphorylation) and (ii) reduced chemotactic T lymphocyte responses in vitro and in vivo. However, CXCL10(1-73) retained its anti-angiogenic properties, as assessed by inhibition of spontaneous and FGF-2-induced migration, wound healing and sprouting of human microvascular endothelial cells.

    The work is well performed though the pharmacological analysis is a little superficial and under-developed with incomplete/inconsistent concentration-dependent responses. The manuscript is rather verbose in places.

    Specific points:

    1. Fig 4: how do the authors know that the reduced calcium responses to full length CXCL10 following pr-treatment with the C-terminal truncated CXCL(1-73) is due to desensitisation rather than say partial agonism? They should compare internalisation of CXCR3 and/or loss of surface expression of CXCR3 following treatment with CXCL10 (1-73) versus CXCL13(1-77) to validate this.
    2. The choice of concentration ranges used for CXCL10(1.77) and CXCL10(1-73) across figs 4, 5 and 6 is inconsistent with no explanation given as to why.
    3. Figs 4: The dose response curves are rather limited narrow e.g.1, 3, 10 nM for CXCL10(1-77). The choice of concentrations for CXCL10(1-73) in fig 4 is a little unusual in Fig 4 (9, 45, 270nM). Has the maximum response to CXCL10(1-73) in figs 4-6 been achieved? It would be useful to know the EC50 values for both full length and truncated forms of CXCL10 in figs 4 and 5
    4. Fig 5: in contrast, to Fig 4, this figure has comparable concentration ranges at 5 points across (1-100 nM). What is the rationale for the inconsistent concentration ranges used across different assays?
    5. The bar graphs for pERK, pAkt responses would look better as line graphs and more complete concentration ranges (perhaps use 5 concentrations e.g. over 1-100 nM for CXCL10(1-77).
    6. Fig 6: the inhibitory effects of CXCL10(1-77) and CXCL10( 1-73) seem to occur at a single concentration (120 nM), Can the spontaneous HMVEC migration be further inhibited at higher doses of truncated and full length CXCL10? Both appear to have just reached 50% inhibition at 120 nM.
    7. Fig 7. What is the impact of both proteoforms on FGF-stimulated wound healing?
    8. Why is it necessary to provide Kd values in the main results text when these are already provided in Table 1. This is just one example of verbosity that that is often present in the manuscript
    9. The methods section is also very long.
    10. What phosphorylation sites are detected in the ERK1/2 and Akt ELISA assays? The authors should provide more details on this point. The ELISA assays alone does not really provide convincing analysis of phosphorylation and should be backed up with more robust assays to assess ERK and Akt phosphorylation e.g western blots and/or flow cytometry with phospho-specific Abs.

    Significance

    The study reveals that the COOH-terminal residues of CXCL10 Lys74-Pro77 are important for GAG binding, CXCR3A signaling, T lymphocyte chemotaxis, but dispensable for angiostasis .

    Study is of interest to basic researchers in areas of pharmacology, immunology and structural biology with relevance to drug discovery, inflammation and cancer biology.

  5. Version published to 10.1101/2023.07.10.548382v1 on bioRxiv