PD-L1 Expression in Mesenchymal Stem/Stromal Cells: Impacts on Innate and Adaptive Immunity, Therapeutic Potential, and Biomarker Utility

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. The authors have provided a relatively well-written review on the role of PD-L1 expression in MSCs with special emphasis onthe PD-1/PD-L1 axis as a key checkpoint pathway in MSC-mediated immune regulation, both on the innate as well as on the adaptive immune systems. The synergistic interactions with other immune checkpoints and the dual function of PD-L1, both as a therapeutic effector and a dynamic biomarkerare deliberated in detail. The relevance of PD-L1 in clinical contexts such as autoimmune diseases, graft-versus-host disease, sepsis, and transplantation as well as the utility of PD-L1 expression as a biomarker for MSC potency are discussed. An extensive mechanistic evaluation of PD-L1 in sepsis, in hyperinflammatory and immunosuppressive diseases was provided. The figures and one Table have summarized nicely the essential issues raised. These are the strengths of the manuscript.
2. It is, on the other hand, a very comprehensive and convoluted article summarizing the multiple cellular and molecular aspects of PD-L1 expression in MSCs and the potentials of acquiring this knowledge for eventual clinical application. The authors went into extensive details, for example, in paragraph: “8. Challenges and Future Directions” – deliberating on Safety Concerns and Tumor Immunity, Safety Concerns and Viral Immune Evasion, Donor Variability and MSC Source, Regulatory and Manufacturing Challenges, Personalized MSC Therapies. Some preliminary clinical data, either their own or from the literature, that provide some evidence for the perturbation of the proportionate PD-L1 high versus low populations in MSCs in the respective disease entities mentioned might have strengthened the manuscript. Do we have any preliminary assessment on the proportion of high versus low PD-L1 expressing MSCs in normal human subjects at steady state as a reference. How does this change with autoimmune disease, GVHD, or after transplantation.
3. In addition, the authors deliberated in extenso on “Engineering MSCs for Enhanced Immunomodulation”, “Exosome-Based, Cell-Free Therapeutic Platforms”, “Synergistic Checkpoint Modulations”, “Longitudinal and Mechanistic Studies”. Before we come to such technical details, some data or evidence on the changes in relative distribution of high versus low PD-L1 expressing in MSCs in inflammation, in specific autoimmune diseases, or in patients with GVHD, and if corrections such MSC populations could induce any improvement.

Author Response

We sincerely thank the reviewer for the thorough and thoughtful evaluation of our manuscript and for highlighting its strengths regarding mechanistic depth, clinical relevance, and clarity of the figures and table. We greatly appreciate the constructive suggestions aimed at further strengthening the translational value of the review.

Comment 1

The authors have provided a relatively well-written review on the role of PD-L1 expression in MSCs with special emphasis on the PD-1/PD-L1 axis as a key checkpoint pathway in MSC-mediated immune regulation, both on the innate as well as on the adaptive immune systems. The synergistic interactions with other immune checkpoints and the dual function of PD-L1, both as a therapeutic effector and a dynamic biomarker are deliberated in detail. The relevance of PD-L1 in clinical contexts such as autoimmune diseases, graft-versus-host disease, sepsis, and transplantation as well as the utility of PD-L1 expression as a biomarker for MSC potency are discussed. An extensive mechanistic evaluation of PD-L1 in sepsis, in hyperinflammatory and immunosuppressive diseases was provided. The figures and one Table have summarized nicely the essential issues raised. These are the strengths of the manuscript.

Response:

We thank the reviewer for this encouraging assessment. We are pleased that the emphasis on the PD‑1/PD‑L1 axis, its mechanistic breadth across innate and adaptive immunity, and its relevance as both a therapeutic effector and biomarker were found to be clear and well supported. We also appreciate the positive comments on the figures and table, which were designed specifically to synthesize complex information for translational readers.

Comment 2

It is, on the other hand, a very comprehensive and convoluted article summarizing the multiple cellular and molecular aspects of PD-L1 expression in MSCs and the potentials of acquiring this knowledge for eventual clinical application. The authors went into extensive details, for example, in paragraph: “8. Challenges and Future Directions” – deliberating on Safety Concerns and Tumor Immunity, Safety Concerns and Viral Immune Evasion, Donor Variability and MSC Source, Regulatory and Manufacturing Challenges, Personalized MSC Therapies. Some preliminary clinical data, either their own or from the literature, that provide some evidence for the perturbation of the proportionate PD-L1 high versus low populations in MSCs in the respective disease entities mentioned might have strengthened the manuscript. Do we have any preliminary assessment on the proportion of high versus low PD-L1 expressing MSCs in normal human subjects at steady state as a reference. How does this change with autoimmune disease, GVHD, or after transplantation.

Response:

We agree with the reviewer that quantitative or semi-quantitative data on PD‑L1^high versus PD‑L1^low MSC subpopulations would substantially strengthen the field and provide an important biological anchor for translational interpretation. At present, however, we are not aware of robust human datasets that systematically define baseline distributions of PD‑L1 expression in MSCs from healthy individuals, nor longitudinal or cross-sectional data across autoimmune disease, GvHD, or post-transplant settings.

Most available human studies focus on relative changes induced by inflammatory licensing (e.g., IFN‑γ exposure) or compare PD‑L1 expression between donors or tissue sources, rather than defining stable population-level reference ranges or disease-specific shifts. Similarly, clinical MSC studies rarely report single-cell–resolved PD‑L1 stratification, instead relying on bulk expression or functional suppression assays.

To address this important point, we revised the manuscript to recognize this knowledge gap in Sections 6, to emphasize the absence of validated reference distributions for PD‑L1^high/low MSCs in humans as a key limitation and priority for future clinical-translational studies, to clarify that current evidence supports PD‑L1 primarily as a dynamic, inducible marker, rather than a fixed phenotypic classifier at steady state.

We have added the revisions to section 6.3 and 9 and merged the relevant paragraphs to improve readability and logical flow.

The revised text now includes the following additions:

• Updated section hading (page 12, lines 447-448): 6.3 Standardization, Regulatory Consideration and (Lack of) Reference Distributions of PD-L1 in MSCs
• Addidtion of the sentence (page 12, lines 453-454): … and define biologically meaningful expression categories such as PD-L1^high and PD-L1^low
• Addidtion of the paragraphs (page 12-13, lines 460-485): A major limitation is the absence of systematic human data defining baseline PD-L1 ex-pression distributions in MSCs under steady-state conditions. The relative proportions of PD-L1high versus PD-L1low MSC subpopulations in healthy individuals have not been es-tablished in a standardized or comparable manner, restricting the ability to contextualize disease-associated changes. Consequently, reference ranges needed to interpret shifts in PD-L1 expression in pathological contexts, such as autoimmune disease, GvHD, sepsis, or post-transplant settings, remain lacking.

 

Preclinical research has strengthened the rationale for focusing on PD-L1, consistently demonstrating that elevated PD-L1 expression enhances MSC immunosuppressive function and therapeutic efficacy. Studies such as Bai et al. (2023) and Lou et al. (2025) show that PD-L1high MSCs outperform PD-L1low counterparts, supporting PD-L1 as a functional biomarker of MSC potency [34, 43]. Additionally, a substantial body of preclinical work has also explored mechanisms to induce PD-L1 expression in MSCs and MSC-derived ex-tracellular vesicles across both in vitro and in vivo systems [36]. However, clinical transla-tion of these findings remains limited, as existing human studies primarily assess induc-ible PD-L1 expression following inflammatory licensing (e.g., IFN-γ, TNF-α), rather than resolving stable or disease-specific subpopulation structures at single-cell resolution. No-tably, no clinical studies to date have prospectively stratified or engineered MSC products based on PD-L1 subpopulation composition, nor correlated these distributions with ther-apeutic outcomes in patients.

Addressing these combined challenges will require the development of standardized, quantitative, and single-cell–resolved frameworks to define PD-L1 expression in human MSCs. Establishing robust assays, reproducible reference distributions, and clearly de-fined disease-associated deviations will be essential for transitioning PD-L1 from a con-text-dependent marker to a validated and regulatory-acceptable potency attribute in clini-cal MSC manufacturing.

This section includes a new reference [43] (page 23, lines 876-877): Luo, X., et al., Single-cell RNA sequencing identifies PD-L1 + mesenchymal stem cells with enhanced immunomodulatory capacity and alleviated the degree of ectopic new bone formation in ankylosing spondylitis. Stem Cell Res Ther, 2025. 16(1): p. 684.

Addidtion of the sentence in section 9 (page 18, lines 728-730): the distribution of PD‑L1high and PD‑L1low MSC subpopulations in healthy individuals and across disease contexts remains poorly defined;

 

Comment 3

Before discussing engineering strategies and advanced platforms, the reviewer asks for clearer evidence regarding changes in PD‑L1^high versus PD‑L1^low MSCs during inflammation or disease, and whether correcting such distributions improves outcomes.

Response:

We fully agree with the reviewer’s prioritization. Indeed, much of the enthusiasm for engineering MSCs or developing exosome-based approaches rests on preclinical evidence demonstrating that higher PD‑L1 expression correlates with enhanced immunosuppressive function, rather than on direct demonstrations of correcting naturally occurring PD‑L1^low populations in vivo.

To better reflect this hierarchy of evidence, we revised Section 8 to more clearly distinguish what has been demonstrated (preclinical augmentation of PD‑L1 enhancing efficacy) from what remains hypothetical (therapeutic correction of endogenous PD‑L1^low MSC distributions in patients), reframed engineering and exosome-based strategies as proof‑of‑concept approaches that highlight biological sufficiency of PD‑L1, rather than as clinically validated corrective interventions, and stated that no clinical trials to date have prospectively stratified or corrected MSC products based on PD‑L1^high/low population structure.

Addidtion of the paragraphs (page 14-15, lines 571-582):

Several of the strategies discussed in this section - including genetic or epigenetic engineering of MSCs, exosome-based cell-free platforms, and combinatorial checkpoint modulation, are supported predominantly by mechanistic and preclinical evidence. While these approaches compellingly demonstrate that enhanced PD L1 expression is biologically sufficient to augment MSC mediated immunoregulation, they cannot be interpreted as absolute evidence that endogenous PD L1low MSC populations may presently be identified, corrected, or therapeutically rebalanced in human disease.

Accordingly, these strategies are best viewed as proof-of-concept explorations that reinforce the central role of the PD 1/PD L1 axis in MSC biology, rather than as clinically validated interventions. Clarifying how naturally occurring PD L1 expression heterogeneity arises in vivo, and whether it can be meaningfully modified in patients, remains an unmet prerequisite for translation.

Reviewer 2 Report

Comments and Suggestions for Authors

This review provides a broad and timely overview of PD-L1 expression in mesenchymal stem/stromal cells, with a strong focus on immune regulation, therapeutic relevance, and biomarker potential. The manuscript is well organized and translationally relevant, particularly in its integration of mechanistic pathways with applications in autoimmune disease, GvHD, transplantation, and sepsis. The following additions may further strengthen the review:

• Please provide a clearer comparison of PD-L1 with other MSC potency markers. Since the review proposes PD-L1 as a potential biomarker, it would be useful to discuss how it compares with markers such as IDO1, PGE2, TGF-β, and others, and whether PD-L1 is sufficient as a standalone marker or is better interpreted within a multi-parameter potency panel.
• Consider adding a brief section on hypoimmunogenic engineered cells. As the review highlights PD-L1 as an important immune-regulatory mechanism, it would be valuable to discuss how PD-L1 fits within broader strategies to generate immune-evasive or hypoimmunogenic cell products, particularly in the context of allogeneic MSC or stem cell-based therapies.
• Please add a short section distinguishing preclinical promise from clinical evidence. The manuscript would benefit from a clearer summary of what has been demonstrated in animal models versus what is currently supported by human studies, especially in the areas of GvHD, transplantation, autoimmune disease, and sepsis.

 

Author Response

We thank the reviewer for the positive assessment of the manuscript’s organization, translational relevance, and disease-focused integration. We are grateful for the concrete suggestions, which we believe will clearly strengthen the review.

Comment 1

Please provide a clearer comparison of PD-L1 with other MSC potency markers. Since the review proposes PD-L1 as a potential biomarker, it would be useful to discuss how it compares with markers such as IDO1, PGE2, TGF-β, and others, and whether PD-L1 is sufficient as a standalone marker or is better interpreted within a multi-parameter potency panel.

Response:

We fully agree with this point. PD‑L1 does not operate in isolation, and MSC immunomodulation is inherently multifactorial. In the revised manuscript, we expanded Section 6 to provide a more explicit comparison between PD‑L1 and established potency markers such as IDO1, PGE2, and TGF‑β, highlighting:

• PD‑L1 as a cell-surface, inducible, and mechanistically well-defined checkpoint marker.
• IDO1 and PGE2 as enzymatic and metabolic mediators with complementary but distinct kinetics and regulatory properties.
• The advantage of PD‑L1 in enabling flow-cytometry–based, single-cell–resolved assessment, compared with soluble mediators.
• We also more clearly stated that PD‑L1 is unlikely to be sufficient as a standalone potency marker, and that its greatest value lies within multi-parameter potency panels, an approach already hinted at but now made more explicit.

Addidtion of the paragraph (page 13, lines 487-498):

MSC immunomodulatory potency is mediated by a complex network of surface molecules, soluble factors, and metabolic enzymes, including indoleamine 2,3-dioxygenase (IDO1), prostaglandin E2 (PGE2), transforming growth factor-β (TGF-β), nitric oxide, and additional checkpoint ligands. Within this framework, PD-L1 represents a mechanistically distinct class of potency marker, functioning as a cell-surface immune checkpoint with direct relevance to T-cell exhaustion, anergy, and regulatory T-cell induction. Compared with soluble mediators such as PGE2 or TGF-β, PD-L1 offers practical and analytical advantages, including direct quantification by flow cytometry and single-cell resolution, enabling the assessment of cellular heterogeneity within MSC products. In comparison, markers such as IDO1 and PGE2 primarily reflect downstream metabolic activity and are often influenced by culture conditions and the duration of inflammatory licensing.

 

Comment 2

Consider adding a brief section on hypoimmunogenic engineered cells. As the review highlights PD-L1 as an important immune-regulatory mechanism, it would be valuable to discuss how PD-L1 fits within broader strategies to generate immune-evasive or hypoimmunogenic cell products, particularly in the context of allogeneic MSC or stem cell-based therapies.

Response:

We appreciate this suggestion. In the revised version, we added a short, dedicated subsection linking PD‑L1 biology in MSCs to broader strategies for generating hypoimmunogenic or immune-evasive cell products, including: Comparisons with approaches used in allogeneic stem cell and iPSC-derived therapies (e.g., checkpoint modulation, MHC attenuation).

Addidtion of a new section (page 17, lines 675-689):

8.6.2 PD-L1 and Hypoimmunogenic Cell Engineering Strategies

The increasing focus on PD‑L1 in MSC immunobiology aligns with broader efforts in regenerative medicine to develop hypoimmunogenic or immune‑evasive cell products, particularly for allogeneic applications. In stem cell–derived therapies, strategies such as immune checkpoint augmentation, attenuation of antigen presentation pathways, and modulation of NK‑cell recognition have been explored to prolong graft persistence and reduce host immune rejection [52].

Within this broader landscape, PD‑L1 overexpression in MSCs represents a targeted and biologically grounded approach that exploits an endogenous immunoregulatory pathway rather than wholesale immune invisibility. This may confer advantages in safety and controllability, particularly when compared with more extensive genetic modifications aimed at eliminating immune recognition altogether.
These observations position PD‑L1 not merely as an MSC‑specific feature, but as part of a generalizable framework for designing next‑generation cell therapies with enhanced immune compatibility.

This section includes a new reference [52] (page 24, lines 895-896): Asadi-Sarabi, P., et al., Hypoimmunogenic pluripotent stem cells: A game-changer in cell-based regenerative medicine. Int Immunopharmacol, 2025. 162: p. 115134.

As a result of the addition of a new section 8.6.2, the following 3 sections have been renamed to sections 8.6.3, 8.6.4, and 8.6.5 respectively.  

Comment 3

Please add a short section distinguishing preclinical promise from clinical evidence. The manuscript would benefit from a clearer summary of what has been demonstrated in animal models versus what is currently supported by human studies, especially in the areas of GvHD, transplantation, autoimmune disease, and sepsis.

Response:

We thank the reviewer for this important and constructive suggestion. We agree that a clearer distinction between preclinical and clinical evidence strengthens the translational perspective of the manuscript. Accordingly, we have added a new paragraph to explicitly address this point.

In this paragraph, we now distinguish between the robust preclinical evidence—derived from animal models and in vitro human systems—supporting PD-L1–associated MSC immunomodulatory mechanisms, and the more limited and largely indirect clinical evidence currently available across indications such as GvHD, transplantation, autoimmune diseases, and sepsis. We further emphasize the need for prospective clinical studies integrating PD-L1 profiling with clinical outcomes to bridge this gap and support its potential role in translational applications.

Addidtion of a new paragraph (page 18, lines 738-747):

It is important to distinguish between the strong preclinical evidence supporting PD‑L1–dependent MSC immunomodulation and the limited clinical data currently available. While animal models and in vitro human systems consistently demonstrate that PD‑L1 expression correlates with enhanced suppression of pathological immune responses, clinical studies in GvHD, autoimmune disease, transplantation, and sepsis have largely inferred PD‑L1 involvement indirectly, without systematic stratification or outcome correlation.
Bridging this gap of evidence will require prospective human studies that integrate PD‑L1 expression profiling with clinical endpoints, ideally within standardized manufacturing and potency assessment frameworks.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have responded appropriately and adequately the issues raised in my initial review.