The discriminatory power of the T cell receptor
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
This paper will be of considerable interest to anybody focusing on highly sensitive T cell antigen recognition. It uses an extended experimental protocol and analytical methods to assess very low T cell receptor binding affinities, and to determine how T cells discriminate between self- and non-self antigens. The main conclusions are well supported by the presented analysis and provide a novel view on a previously considered concept.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Immunology and Inflammation (eLife)
Abstract
T cells use their T cell receptors (TCRs) to discriminate between lower-affinity self and higher-affinity non-self peptides presented on major histocompatibility complex (pMHC) antigens. Although the discriminatory power of the TCR is widely believed to be near-perfect, technical difficulties have hampered efforts to precisely quantify it. Here, we describe a method for measuring very low TCR/pMHC affinities and use it to measure the discriminatory power of the TCR and the factors affecting it. We find that TCR discrimination, although enhanced compared with conventional cell-surface receptors, is imperfect: primary human T cells can respond to pMHC with affinities as low as K D ∼ 1 mM. The kinetic proofreading mechanism fit our data, providing the first estimates of both the time delay (2.8 s) and number of biochemical steps (2.67) that are consistent with the extraordinary sensitivity of antigen recognition. Our findings explain why self pMHC frequently induce autoimmune diseases and anti-tumour responses, and suggest ways to modify TCR discrimination.
Article activity feed
-
-
-
Author Response:
Evaluation Summary:
This paper will be of considerable interest to anybody focusing on highly sensitive T cell antigen recognition. It uses an extended experimental protocol and analytical methods to assess very low T cell receptor binding affinities, and to determine how T cells discriminate between self- and non-self antigens. The main conclusions are well supported by the presented analysis and provide a novel view on a previously considered concept.
Reviewer #1 (Public Review):
The presented manuscript takes a comprehensive and elaborated look at how T cell receptors (TCR) discriminate between self and non-self antigens. By extending a previous experimental protocol for measuring T cell receptor binding affinities against peptide MHC complexes (pMHC), they are able to determine very low TCR-pMHC binding …
Author Response:
Evaluation Summary:
This paper will be of considerable interest to anybody focusing on highly sensitive T cell antigen recognition. It uses an extended experimental protocol and analytical methods to assess very low T cell receptor binding affinities, and to determine how T cells discriminate between self- and non-self antigens. The main conclusions are well supported by the presented analysis and provide a novel view on a previously considered concept.
Reviewer #1 (Public Review):
The presented manuscript takes a comprehensive and elaborated look at how T cell receptors (TCR) discriminate between self and non-self antigens. By extending a previous experimental protocol for measuring T cell receptor binding affinities against peptide MHC complexes (pMHC), they are able to determine very low TCR-pMHC binding affinities and, thereby, show that the discriminatory power of the TCR seems to be imperfect. Instead of a previously considered sharp threshold in discriminating between self and non-self antigen, the TCR can respond to very low binding affinities leading to a more transient affinity threshold. However, the analysis still indicates an improved discrimination ability for TCR compared to other cell surface receptors. These findings could impact the way how T cell mediated autoimmunity is studied.
The authors follow a comprehensive and elaborated approach, combining in vitro experiments with analytical methods to estimate binding affinities. They also show that the general concept of kinetic proofreading fits their data with providing estimates on the number of proofreading steps and the corresponding rates. The statistical and analytical methods are well explained and outlined in detail within the Supplemental Material. The source of all data, and especially how the data to analyze other cell surface receptor binding affinities was extracted, are given in detail as well. Besides being able to quantify TCR-pMHC interactions for very low binding affinities, their findings will improve the ability to assess how autoimmune reactions are potentially triggered, and how potent anti-tumour T cell therapies can be generated.
In summary, the study represents an elaborated and concise analysis of TCR-pMHC affinities and the ability of TCR to discriminate between self and non-self antigens. All conclusions are well supported by the presented data and analyses without major caveats.
Reviewer #2 (Public Review):
The paper revisits the question of ligand discrimination ability of TCRs of T cells. The authors find that the commonly held notion of very sharp discrimination between strongly and weakly binding peptides does not hold when the affinities of the weak peptides are re-measured more accurately, using their own new method of calibration of SPR measurements. They are able to phenomenologically fit their results with a ~2 step Kinetic Proofreading model.
It is a very carefully researched and thorough paper. The conclusions seem to be supported by the data and fundamental for our understanding of the T cell immune response with potentially very high impact in many scientific and applied fields. The calibration method could be of potential use in other cases where low affinities are an issue.
As a non-expert in the details of experimental technique, it is somewhat difficult to understand in detail the Ab calibration of the SPR curve - which is a central piece of the paper. The main question is - what are the grounds (theoretical and/or empirical) to expect that the B_max of the TCR dose response curve will continue to be proportional to the plateau level of the Ab. Figure 1D does suggest that, but it would be hard to predict what proportionality shape the curve will take for lower affinity peptides. Given that essentially all the paper claims rest on this assumption, this should explained/reasoned/supported more clearly.
We have revised the relevant Results and Methods sections to provide additional information. This information should clarify the expected relationship between Bmax and W6/32 binding. We emphasise that we have only interpolated within the curve and therefore, have not relied on any assumptions about the relationship between these two values outside of the empirical curve that we have generated.
On the theoretical side - I think the scaling alpha\simeq 2 in Figure 2 is indeed consistent with a two-step KPR amplification. However, there are some questions regarding the fitting of the full model to the P_15 of the CD69 response. As explained in the Supplementary Material the authors use 3 global and 2 local parameters resulting in 37 (or 27) parameters for 32 data points. To a naive reader this might look excessive and prone to overfitting. On the other hand, looking at Figure S8 shows the value ranges of lambda and k_p are quite tight. This is in contrast to gamma and dellta that look completely unconstrained.
We have revised the relevant Results section to explicitly indicate that the number of data points ex- ceeds the number of free parameters, which together with the ABC-SMC results, should provide additional confidence that we are not over-fitting.
Finally, one of the stated advantages of the adaptive proof-reading model is that it is capable of explaining antagonism. It is hard to see how a 'vanilla" KPR model is capable of explaining antagonism.
We have added a discussion paragraph to discuss antagonism, which cannot be explained by the basic KP model that we found is sufficient to explain our data on antigen discrimination in the presence of self pMHCs on autologous APCs. We describe how the methods we have employed can be used to study antagonism.
Reviewer #3 (Public Review):
Pettmann et al. aimed at significantly improving the accuracy of SPR-based measurements of low affinity TCR-pMHC interactions by including a 100% binding control (injecting of a conformation-specific HLA-antibody) in the surface plasmon resonance protocol. Interpolating with the information of saturated pMHC binding on the chip The authors arrive at KDs for low affinity binders that are significantly higher than the previously reported constants. If correct, this has considerable ramifications for the interpretations of the results obtained from functional assays measuring the T cell response towards pMHCs featured in a titrated fashion. Unlike what was put forward by earlier reports, the authors conclude that the discriminatory power of TCRs is far from perfect, as T cells still respond to low affinity pMHC-ligands without a sharp affinity threshold. This is also because they managed to detect T cells responding to even ultra-low affinity ligands if provided in sufficient numbers.
The body of work convinces in several regards:
(i) It is exceedingly well thought out and introduces a quality of analytical strength that is absent in most of the literature published thus far on this topic.
(ii) At the same time theoretical arguments are bolstered by a large body of experimental "wet" work, which combines a synthetic approach with cellular immunology and which appears overall well executed.
(iii) The data lead to hypotheses in the field of T cell antigen recognition in general and in the theatre of autoimmunity, cancer and infectious diseases.
There are a few aspects that may limit the impact of the study. I have listed them below:
(i) The study does not provide kinetic data for the low affinity ligand-TCR binding but rather argues from the position of affinities as determined via Bmax. This limits somewhat the robustness of the statements made with regard to kinetic proofreading.
We agree with this statement and are hoping to directly measure off-rates in the future. We note that in the published literature, including our own work, point mutations to the peptide generally modify the off-rate with only minor impact on the on-rate. An example of this can be found in Lever et al (2016) PNAS where point mutations led to 100,000-fold change in the off-rate but only a 10-fold change in the on-rate. This likely explains why antigen potency is often well-correlated with affinity when using point mutations to the peptide.
(ii) Thresholds for readouts were arbitrarily chosen (e.g. 15% activation). It appears such choices were based on system behavior (with the largest differences observed among the groups) but may have implications for the drawn conclusions.
We have chosen 15% in order to capture the ultra-low affinity pMHCs in our potency plots and have now added a sentence for why we have chosen this particular threshold. We did explore different thresholds but found that they produced similar values of α. The precise threshold could change the estimate of α if the shape of dose-response curves was dependent on antigen affinity but we did not find any evidence for this within our data.
In summary, the work presented contributes to demystifying the link between TCR-engagement and (membrane proximal) signaling. It also provides a fresh perspective on the potential of TCR-cossreactivity.
-
Reviewer #3 (Public Review):
Pettmann et al. aimed at significantly improving the accuracy of SPR-based measurements of low affinity TCR-pMHC interactions by including a 100% binding control (injecting of a conformation-specific HLA-antibody) in the surface plasmon resonance protocol. Interpolating with the information of saturated pMHC binding on the chip The authors arrive at KDs for low affinity binders that are significantly higher than the previously reported constants. If correct, this has considerable ramifications for the interpretations of the results obtained from functional assays measuring the T cell response towards pMHCs featured in a titrated fashion. Unlike what was put forward by earlier reports, the authors conclude that the discriminatory power of TCRs is far from perfect, as T cells still respond to low affinity …
Reviewer #3 (Public Review):
Pettmann et al. aimed at significantly improving the accuracy of SPR-based measurements of low affinity TCR-pMHC interactions by including a 100% binding control (injecting of a conformation-specific HLA-antibody) in the surface plasmon resonance protocol. Interpolating with the information of saturated pMHC binding on the chip The authors arrive at KDs for low affinity binders that are significantly higher than the previously reported constants. If correct, this has considerable ramifications for the interpretations of the results obtained from functional assays measuring the T cell response towards pMHCs featured in a titrated fashion. Unlike what was put forward by earlier reports, the authors conclude that the discriminatory power of TCRs is far from perfect, as T cells still respond to low affinity pMHC-ligands without a sharp affinity threshold. This is also because they managed to detect T cells responding to even ultra-low affinity ligands if provided in sufficient numbers.
The body of work convinces in several regards:
(i) It is exceedingly well thought out and introduces a quality of analytical strength that is absent in most of the literature published thus far on this topic.
(ii) At the same time theoretical arguments are bolstered by a large body of experimental "wet" work, which combines a synthetic approach with cellular immunology and which appears overall well executed.
(iii) The data lead to hypotheses in the field of T cell antigen recognition in general and in the theatre of autoimmunity, cancer and infectious diseases.
There are a few aspects that may limit the impact of the study. I have listed them below:
(i) The study does not provide kinetic data for the low affinity ligand-TCR binding but rather argues from the position of affinities as determined via Bmax. This limits somewhat the robustness of the statements made with regard to kinetic proofreading.
(ii) Thresholds for readouts were arbitrarily chosen (e.g. 15% activation). It appears such choices were based on system behavior (with the largest differences observed among the groups) but may have implications for the drawn conclusions.
In summary, the work presented contributes to demystifying the link between TCR-engagement and (membrane proximal) signaling. It also provides a fresh perspective on the potential of TCR-cossreactivity.
-
Reviewer #2 (Public Review):
The paper revisits the question of ligand discrimination ability of TCRs of T cells. The authors find that the commonly held notion of very sharp discrimination between strongly and weakly binding peptides does not hold when the affinities of the weak peptides are re-measured more accurately, using their own new method of calibration of SPR measurements. They are able to phenomenologically fit their results with a ~2 step Kinetic Proofreading model.
It is a very carefully researched and thorough paper. The conclusions seem to be supported by the data and fundamental for our understanding of the T cell immune response with potentially very high impact in many scientific and applied fields. The calibration method could be of potential use in other cases where low affinities are an issue.
As a non-expert in the …
Reviewer #2 (Public Review):
The paper revisits the question of ligand discrimination ability of TCRs of T cells. The authors find that the commonly held notion of very sharp discrimination between strongly and weakly binding peptides does not hold when the affinities of the weak peptides are re-measured more accurately, using their own new method of calibration of SPR measurements. They are able to phenomenologically fit their results with a ~2 step Kinetic Proofreading model.
It is a very carefully researched and thorough paper. The conclusions seem to be supported by the data and fundamental for our understanding of the T cell immune response with potentially very high impact in many scientific and applied fields. The calibration method could be of potential use in other cases where low affinities are an issue.
As a non-expert in the details of experimental technique, it is somewhat difficult to understand in detail the Ab calibration of the SPR curve - which is a central piece of the paper. The main question is - what are the grounds (theoretical and/or empirical) to expect that the B_max of the TCR dose response curve will continue to be proportional to the plateau level of the Ab. Figure 1D does suggest that, but it would be hard to predict what proportionality shape the curve will take for lower affinity peptides. Given that essentially all the paper claims rest on this assumption, this should explained/reasoned/supported more clearly.
On the theoretical side - I think the scaling alpha\simeq 2 in Figure 2 is indeed consistent with a two-step KPR amplification. However, there are some questions regarding the fitting of the full model to the P_15 of the CD69 response. As explained in the Supplementary Material the authors use 3 global and 2 local parameters resulting in 37 (or 27) parameters for 32 data points. To a naive reader this might look excessive and prone to overfitting. On the other hand, looking at Figure S8 shows the value ranges of lambda and k_p are quite tight. This is in contrast to gamma and dellta that look completely unconstrained.
Finally, one of the stated advantages of the adaptive proof-reading model is that it is capable of explaining antagonism. It is hard to see how a 'vanilla" KPR model is capable of explaining antagonism.
-
Reviewer #1 (Public Review):
The presented manuscript takes a comprehensive and elaborated look at how T cell receptors (TCR) discriminate between self and non-self antigens. By extending a previous experimental protocol for measuring T cell receptor binding affinities against peptide MHC complexes (pMHC), they are able to determine very low TCR-pMHC binding affinities and, thereby, show that the discriminatory power of the TCR seems to be imperfect. Instead of a previously considered sharp threshold in discriminating between self and non-self antigen, the TCR can respond to very low binding affinities leading to a more transient affinity threshold. However, the analysis still indicates an improved discrimination ability for TCR compared to other cell surface receptors. These findings could impact the way how T cell mediated …
Reviewer #1 (Public Review):
The presented manuscript takes a comprehensive and elaborated look at how T cell receptors (TCR) discriminate between self and non-self antigens. By extending a previous experimental protocol for measuring T cell receptor binding affinities against peptide MHC complexes (pMHC), they are able to determine very low TCR-pMHC binding affinities and, thereby, show that the discriminatory power of the TCR seems to be imperfect. Instead of a previously considered sharp threshold in discriminating between self and non-self antigen, the TCR can respond to very low binding affinities leading to a more transient affinity threshold. However, the analysis still indicates an improved discrimination ability for TCR compared to other cell surface receptors. These findings could impact the way how T cell mediated autoimmunity is studied.
The authors follow a comprehensive and elaborated approach, combining in vitro experiments with analytical methods to estimate binding affinities. They also show that the general concept of kinetic proofreading fits their data with providing estimates on the number of proofreading steps and the corresponding rates. The statistical and analytical methods are well explained and outlined in detail within the Supplemental Material. The source of all data, and especially how the data to analyze other cell surface receptor binding affinities was extracted, are given in detail as well. Besides being able to quantify TCR-pMHC interactions for very low binding affinities, their findings will improve the ability to assess how autoimmune reactions are potentially triggered, and how potent anti-tumour T cell therapies can be generated.
In summary, the study represents an elaborated and concise analysis of TCR-pMHC affinities and the ability of TCR to discriminate between self and non-self antigens. All conclusions are well supported by the presented data and analyses without major caveats.
-
Evaluation Summary:
This paper will be of considerable interest to anybody focusing on highly sensitive T cell antigen recognition. It uses an extended experimental protocol and analytical methods to assess very low T cell receptor binding affinities, and to determine how T cells discriminate between self- and non-self antigens. The main conclusions are well supported by the presented analysis and provide a novel view on a previously considered concept.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors)
-
-