Endo-lysosomal assembly variations among human leukocyte antigen class I (HLA class I) allotypes
Curation statements for this article:-
Curated by eLife
eLife assessment
In this manuscript, the authors provide mechanisms by which HLA-I polymorphism affects the capacity in the endo-lysosomal assembly of HLA-I molecules for constitutive expression and during cross-presentation. The findings may have implications for allotype-dependent variation in T cell responses to antigens localized in different subcellular compartments. However, additional biochemical and quantitative data is essential to bolster the central claims of the paper.
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Immunology and Inflammation (eLife)
Abstract
The extreme polymorphisms of human leukocyte antigen class I (HLA class I) proteins enable the presentation of diverse peptides to cytotoxic T lymphocytes. The canonical endoplasmic reticulum (ER) HLA class I assembly pathway enables presentation of cytosolic peptides, but effective intracellular surveillance requires multi-compartmental antigen sampling. Endo-lysosomes are generally sites of HLA class II assembly, but human monocytes and monocyte-derived dendritic cells (moDCs) also contain significant reserves of endo-lysosomal HLA class I molecules. We hypothesized variable influences of HLA class I polymorphisms upon outcomes of endo-lysosomal trafficking, as the stabilities and peptide occupancies of cell surface HLA class I molecules are variable. Consistent with this model, when the endo-lysosomal pH of moDCs is disrupted, HLA-B allotypes display varying propensities for reductions in surface expression, with HLA-B*08:01 or HLA-B*35:01 being among the most resistant or sensitive, respectively, among eight tested HLA-B allotypes. Perturbations of moDC endo-lysosomal pH result in accumulation of HLA-B*35:01 in LAMP1 + compartments and increase HLA-B*35:01 peptide receptivity. These findings reveal the intersection of the vacuolar cross-presentation pathway with a constitutive assembly pathway for some HLA-B allotypes. Notably, cross-presentation of epitopes derived from two soluble antigens was also more efficient for B*35:01 compared to B*08:01, even when matched for T cell response sensitivity, and more affected by cathepsin inhibition. Thus, HLA class I polymorphisms dictate the degree of endo-lysosomal assembly, which can supplement ER assembly for constitutive HLA class I expression and increase the efficiency of cross-presentation.
Article activity feed
-
-
eLife assessment
In this manuscript, the authors provide mechanisms by which HLA-I polymorphism affects the capacity in the endo-lysosomal assembly of HLA-I molecules for constitutive expression and during cross-presentation. The findings may have implications for allotype-dependent variation in T cell responses to antigens localized in different subcellular compartments. However, additional biochemical and quantitative data is essential to bolster the central claims of the paper.
-
Reviewer #1 (Public Review):
The manuscript entitled "Endo-lysosomal assembly variations among Human Leukocyte 1 Antigen class I (HLA-I) allotypes" by Eli Olson and co-workers reports an interesting observation that HLA-I alleles known not to require the standard peptide-loading complex for assembly and egress from the endoplasmic reticulum, may assemble with peptide ligands within the endo/lysosomes and are more adept at antigen cross-presentation to the CD8+ subset of T cells.
The strengths of the work are (a) a novel hypothesis that HLA-I allotype variations caused by HLA-I gene polymorphisms control endo/lysosomal HLA-I assembly and antigen cross-presentation even though there is ample evidence for cross-presentation using the endo/lysosomal pathway; and (b) new evidence to support this hypothesis.
Weaknesses are (a) the use of …
Reviewer #1 (Public Review):
The manuscript entitled "Endo-lysosomal assembly variations among Human Leukocyte 1 Antigen class I (HLA-I) allotypes" by Eli Olson and co-workers reports an interesting observation that HLA-I alleles known not to require the standard peptide-loading complex for assembly and egress from the endoplasmic reticulum, may assemble with peptide ligands within the endo/lysosomes and are more adept at antigen cross-presentation to the CD8+ subset of T cells.
The strengths of the work are (a) a novel hypothesis that HLA-I allotype variations caused by HLA-I gene polymorphisms control endo/lysosomal HLA-I assembly and antigen cross-presentation even though there is ample evidence for cross-presentation using the endo/lysosomal pathway; and (b) new evidence to support this hypothesis.
Weaknesses are (a) the use of qualitative serologic assays in which specificity of broadly reactive antibodies such as the anti-Bw6 antibody cannot be easily controlled; (b) poor resolution of co-localization micrographs and quantification based on such data; (c) evidence that endo/lysosomal pathway dominates in cross-presentation by B35.1 allotype is weak as the data suggest a significant role for the standard cytoplasmic pathway itself in this immunologic process; and (d) narrow focus on a single member of the B7 supertype that is prevalent at low frequency in the African American (AA: 0.05) and White American (WA: 0.07) populations. These weaknesses can be addressed by using (a) more quantitative biochemical assays; (b) high resolution microscopy; & (c) extending cellular biochemical studies to one or more additional allotypes in the B7 superfamily-e.g., B7.2 (AA: 0.08; WA: 0.155) itself &/or B35.3 (AA: 0.005; WA: 0.027) or B53.1 (AA: 0.133; WA: 0.004).
-
Reviewer #2 (Public Review):
The manuscript reports on the complex variability of expression, trafficking, assembly/stability, and peptide loading among different MHC I haplotypes. In particular by analyzing two distinct MHC I molecules as representative members of groups of allotypes, that favor canonical or non-canonical assembly modes, the PI reports on preferential cytosolic or endo-lysosomal MHC I loading. Overall, the data shed light on the intersection between MHC I conformation and subcellular sites of peptide loading and help explain MHC I immunosurveillance at a different subcellular location.
In the first series of experiments the authors report an uneven surface expression of HLA-B vs HLA-A, and C on circulating monocytes, with HLA-B being expressed 4 times higher, also they report that as compared to the TAP-dependent …
Reviewer #2 (Public Review):
The manuscript reports on the complex variability of expression, trafficking, assembly/stability, and peptide loading among different MHC I haplotypes. In particular by analyzing two distinct MHC I molecules as representative members of groups of allotypes, that favor canonical or non-canonical assembly modes, the PI reports on preferential cytosolic or endo-lysosomal MHC I loading. Overall, the data shed light on the intersection between MHC I conformation and subcellular sites of peptide loading and help explain MHC I immunosurveillance at a different subcellular location.
In the first series of experiments the authors report an uneven surface expression of HLA-B vs HLA-A, and C on circulating monocytes, with HLA-B being expressed 4 times higher, also they report that as compared to the TAP-dependent allotype B*08:01 the TAP-independent allotype B*35:01 has a lower surface half-life and if often present as an empty molecule. These data set the basis for the author's hypothesis that B*35:01 could traffic in Rab11+ compartment and be involved in cross-presentation, which indeed is demonstrated in a series of pulse-chase peptide experiments and using cathepsin inhibitors.
Overall, the experiments could be improved by performing subcellular fractionation and organelle purification to conclusively demonstrate the differential trafficking of B*08:01 vs B*35:01, as well as quantitative mass spectrometry to determine cytosolic vs endosomal processing for one selected epitope presented by the different haplotypes.
-
Reviewer #3 (Public Review):
The work by Olson and colleagues provides novel, fundamental insights into the role of HLA polymorphisms in the processing of exogenous antigens via the non-canonical vacuolar and cytosolic pathways. The choice of the two exemplar HLA-B allotypes leverages a significant amount of background work done both by the Raghavan lab and others, together with a series of novel and very elegant in vitro assays to elucidate a trend where differences in peptide binding preferences and other molecular features can have a drastic effect on non-canonical processing of exogenous antigens. Finally, using two related cell types (monocytes and monocyte-derived DCs) it highlights important differences in endo-lysosomal assemble within different cell types, an aspect of the non-canonical antigen processing that has not been …
Reviewer #3 (Public Review):
The work by Olson and colleagues provides novel, fundamental insights into the role of HLA polymorphisms in the processing of exogenous antigens via the non-canonical vacuolar and cytosolic pathways. The choice of the two exemplar HLA-B allotypes leverages a significant amount of background work done both by the Raghavan lab and others, together with a series of novel and very elegant in vitro assays to elucidate a trend where differences in peptide binding preferences and other molecular features can have a drastic effect on non-canonical processing of exogenous antigens. Finally, using two related cell types (monocytes and monocyte-derived DCs) it highlights important differences in endo-lysosomal assemble within different cell types, an aspect of the non-canonical antigen processing that has not been sufficiently addressed in previous studies. While the number of allotypes and cell types utilized in this study is small (n=2 in each case), it provides an elaborate view into the vacuolar processing pathway and motivates further studies on a more expanded set of alleles in future studies. Finally, it underscores the importance of defining the expression of HLA expression levels in the context of specific cell types, setting a standard for future studies in the field.
Moreover, the work outlined in this study is technically sound, with sufficient attention to detail, adequate control experiments, and a rigorous statistical analysis of the resulting data when needed. Overall, the conclusions are well supported by the data. The manuscript is written in a clear, succinct manner to comprehend by a wide audience of readers.
One shortcoming of the paper is a lack of molecular characterization of the peptide-receptive MHC-I species at different stages of their assembly and trafficking process. For instance, while the authors utilize a monoclonal antibody (HC10) to probe empty MHC-I conformers and their dynamics, they don't provide further analysis of interactions with the light chain, a component of the complex that is known to be critical for regulating the internalization and peptide-loading process, both on the cell surface and at different intracellular compartments. Finally, while the overall effects on the cross-presentation of specific EBV antigens by the two allotypes are well described, what is lacking is a more quantitative analysis of the number of molecules, their densities, and distribution on the cell surface, all of which are known to have important consequences for T cell stimulation.
-
Author Response:
Reviewer #2 (Public Review):
The manuscript reports on the complex variability of expression, trafficking, assembly/stability, and peptide loading among different MHC I haplotypes. In particular by analyzing two distinct MHC I molecules as representative members of groups of allotypes, that favor canonical or non-canonical assembly modes, the PI reports on preferential cytosolic or endo-lysosomal MHC I loading. Overall, the data shed light on the intersection between MHC I conformation and subcellular sites of peptide loading and help explain MHC I immunosurveillance at a different subcellular location.
In the first series of experiments the authors report an uneven surface expression of HLA-B vs HLA-A, and C on circulating monocytes, with HLA-B being expressed 4 times higher, also they report that as compared to the …
Author Response:
Reviewer #2 (Public Review):
The manuscript reports on the complex variability of expression, trafficking, assembly/stability, and peptide loading among different MHC I haplotypes. In particular by analyzing two distinct MHC I molecules as representative members of groups of allotypes, that favor canonical or non-canonical assembly modes, the PI reports on preferential cytosolic or endo-lysosomal MHC I loading. Overall, the data shed light on the intersection between MHC I conformation and subcellular sites of peptide loading and help explain MHC I immunosurveillance at a different subcellular location.
In the first series of experiments the authors report an uneven surface expression of HLA-B vs HLA-A, and C on circulating monocytes, with HLA-B being expressed 4 times higher, also they report that as compared to the TAP-dependent allotype B*08:01 the TAP-independent allotype B*35:01 has a lower surface half-life and if often present as an empty molecule. These data set the basis for the author's hypothesis that B*35:01 could traffic in Rab11+ compartment and be involved in cross-presentation, which indeed is demonstrated in a series of pulse-chase peptide experiments and using cathepsin inhibitors.
Overall, the experiments could be improved by performing subcellular fractionation and organelle purification to conclusively demonstrate the differential trafficking of B*08:01 vs B*35:01, as well as quantitative mass spectrometry to determine cytosolic vs endosomal processing for one selected epitope presented by the different haplotypes.
We thank the reviewer for this suggestion, and agree that this would be a powerful method for further validating differential HLA-B trafficking and antigen processing. Unfortunately, we were unable to perform subcellular fractionation experiments for mass spec, as protocols for fractionation require upwards of 10 million cells to obtain endosomal fractions. For our donor samples, we typically obtain 1- 2 million moDCs after isolation and differentiation, greatly limiting the types of experiments we can perform with primary cells from specific donors. We considered performing these experiments in a cell line but were concerned that ER as well as endosomal trafficking and processing pathways might differ between cell lines and primary cells, which would necessitate a number of additional studies to validate use of the cell lines.
-
