The immunological ‘Big Bang’ that gave rise to RAG-based antigen receptor gene rearrangement in jawed vertebrates produced an adaptive immune system in which each naive B and T cell expresses a clonotypic B or T cell receptor (BCR or TCR) with unique antigen specificity (Bernstein et al., 1996; Flajnik, 2014). The B and T cell repertoires can therefore be thought of as combinatorial libraries from which individual clonotypes, expressing receptors specific to antigen, expand to mount a tailored response. This strategy provides jawed vertebrates with long-lived protection against microbial infection, neoplastic transformation, and is the basis for vaccines; yet it also presents the risk of reactivity against self. As a result, mechanisms have evolved to ensure that the adaptive immune system of jawed vertebrates is on high alert to respond to foreign antigens while maintaining tolerance to self.

For the CD4⁺ T cell repertoire, discriminating self from foreign begins in the thymus where developmental checkpoints test the strength with which a thymocyte’s clonotypic TCR interacts with composite surfaces of self-peptides embedded within class II MHC (pMHCII) (Huseby et al., 2005). TCRs that interact weakly with self-pMHCII direct positive selection toward the CD4⁺ lineage, while those that strongly recognize self-pMHCII induce apoptosis to establish central tolerance by removing autoreactive TCR clonotypes from the repertoire by negative selection. For CD4⁺ T cells that emerge from the thymus, the nature of TCR–pMHCII engagement determines their homeostasis, activation, and differentiation to one of a variety of helper (Th) or regulatory (Treg) phenotypes (Gottschalk et al., 2010; Tubo and Jenkins, 2014). These Th and Treg cells then influence the responses of a variety of other immune cell types.

The conversion of pMHCII-specific information into intracellular signals is an emergent property of five distinct modules: TCR, CD3γε, CD3δε, CD3ζζ, and CD4 (Kobayashi et al., 2020; Kuhns and Davis, 2012). The TCR is the receptor module. It deciphers information encoded within the composite pMHCII surface and relays the information to the immunoreceptor tyrosine-based activation motifs (ITAMs) of three associated signaling modules (CD3γε, CD3δε, and CD3ζζ) (Chen et al., 2022; Gil et al., 2002; Lee et al., 2015). CD4 is the coreceptor module. It binds MHCII on the outside of the CD4⁺ T cell and interacts with the Src kinase, Lck, via an intracellular CQC clasp motif (Kim et al., 2003; Turner et al., 1990). According to the TCR signaling paradigm, CD4 recruits Lck to phosphorylate the ITAMs of TCR–CD3 complexes when both TCR–CD3 and CD4 coincidently engage pMHCII (Rudd, 2021). Phosphorylation of the ITAMs initiates pMHCII-specific signaling, connecting TCR–CD3 complexes to the broader intracellular signaling machinery (Courtney et al., 2018; Gaud et al., 2018). In this model, the biophysical properties that govern TCR–pMHCII interactions are the key determinants for T cell fate decisions while CD4 plays a supporting role.

Recent evidence suggests however that the role of CD4 within the TCR signaling paradigm requires refinement. CD4’s extracellular domain (ECD) can increase TCR dwell time on pMHCII and position its intracellular domain (ICD) in a defined relationship with the TCR–CD3 complex through coordinated rather than coincident interactions (Glassman et al., 2016; Glassman et al., 2018; Guy and Vignali, 2009). Furthermore, CD4 molecules that are not associated with Lck have been proposed to compete with those that are to limit the number of TCR–CD3 complexes phosphorylated by Lck, thus setting a threshold for the duration of TCR–pMHCII interactions required to initiate signaling (Stepanek et al., 2014). In addition, CD4 molecules that are associated with Lck are reported to play a vital role in pMHCII restriction by sequestering Lck away from TCR–CD3 to prevent off-target signaling by TCR interactions with non-pMHCII molecules (Van Laethem et al., 2013). These models help explain how the stability and composition of TCR–CD3–pMHCII–CD4(+/−Lck) assemblies influence and regulate ITAM phosphorylation. They also suggest that CD4, which has been less-well studied than the TCR, warrants more attention.

Accordingly, we reconstructed the evolutionary history of extant CD4 homologs from boney fish, reptiles, birds, and mammals to evaluate the results of ~435 million years of CD4 evolution in jawed vertebrates. Our analyses identified five putative motifs within the CD4 transmembrane domain (TMD) and ICD that are unique to mammals, or found only in eutherians (placental mammals), and contain residues under purifying selection. Further analyses identified residues within motifs in the ECD, TMD, and ICD that have covaried over evolutionary time. Follow-on structure–function analyses revealed a paradox that cannot be explained by the current TCR signaling paradigm. Specifically, mutating the transmembrane and intracellular motifs increased CD4–Lck association and impaired CD4-driven responses. Conversely, mutating the ICD helix or a motif therein reduced CD4–Lck interactions and enhanced responses. These findings have broad implications for how multisubunit transmembrane receptors relay ligand-specific information across the cell membrane, for our understanding of CD4⁺ T cell biology, and for biomimetic engineering of synthetic receptors.

The goal of this study was to gain novel insights into how ~435 million years of natural selection shaped eutherian CD4 function. We therefore used evolutionary and covariation analyses to guide structure–function studies of mouse CD4 in 58α⁻β⁻ T cell hybridomas. We note that while these cells may lack some elements of signal transduction found in real T cells, their IL-2 responses to pMHCII are CD4 dependent so relevant signaling pathways are intact (Glassman et al., 2018). Furthermore, protein–protein interactions between the CD4 motifs studied here with their interacting partners should be the same at a biochemical level as in real T cells provided the interacting partners are expressed. Indeed, the link between the CQC clasp motif, CD4–Lck interactions, and IL-2 production were established in seminal work using 58α⁻β⁻ cells (Glaichenhaus et al., 1991). Finally, because expression levels of CD4, Lck, adaptor proteins, or other molecules that might interact with CD4 vary between phenotypically different T cell populations (http://immpres.co.uk/), the motifs studied here may affect different outcomes in different T cell subsets. Our data are therefore likely to reflect general principles concerning the function of the motifs we identified, even if they may not reflect exactly how the motifs uniquely influence thymocyte, naive, Th, Treg, or Tm cell behavior.

A central tenet of the TCR signaling paradigm, and related models, is that CD4−Lck interactions via the CQC clasp allow CD4 to recruit Lck to phosphorylate the CD3 ITAM upon pMHCII engagement (Glaichenhaus et al., 1991; Rudd, 2021; Stepanek et al., 2014). However, when we tested this model directly with our Clasp mutant, we did not observe consistent reductions in CD3ζ phosphorylation as expected in response to agonist or weak agonist pMHCII. We did observe reduced Zap70 and Plcγ1 phosphorylation in response to agonist pMHCII, which helps explain the reduced IL-2 responses reported here and elsewhere for CQC clasp mutants (Glaichenhaus et al., 1991). We interpret these data as evidence that, in our system, Lck can efficiently phosphorylate CD3ζ ITAMs, but not other substrates (e.g., Zap70), even if CD4−Lck abundance is reduced by mutating the CQC clasp. We cannot, however, rule out that mutating the CQC clasp would impact CD3ζ phosphorylation in response to low densities of agonist pMHCII where CD4 is known to be critical for downstream signaling responses such as calcium mobilization or IL-2 production (Glassman et al., 2018; Irvine et al., 2002); however, on this point it is worth noting that our previous work points to an essential role for the CD4 ectodomain in mediating sensitivity to weak agonist pMHCII, as well as low doses of agonist pMHCII (Glassman et al., 2018). Overall, a key takeaway from the results with our Clasp mutant is that focusing solely on CD4−Lck interactions via the CQC clasp fails to convey a full understanding of the functional significance of CD4−Lck pairs.

Indeed, the Clasp mutant is better understood when considered with our TP mutant, which was nearly identical to the Clasp mutant regarding IL-2 responses even though it had ~123% more total CD4−Lck pairs than WT CD4. Enrichment of CD4−Lck pairs in the DSMs of TP mutants corresponded with lower CD3ζ, Zap70, and Plcγ1 phosphorylation, providing evidence that the Clasp and TP mutant IL-2 phenotypes are similar for different reasons. We favor the following interpretation: (1) Lck freed by our Clasp mutant can phosphorylate CD3ζ ITAMs but cannot as efficiently phosphorylate other substrates; (2) CD4−Lck pairs sequestered to the wrong membrane compartment by our CD4 TP mutation does not efficiently phosphorylate CD3ζ ITAMs and other substrates. Taken together, we think the simplest interpretation of these data is that CD4 association with Lck regulates the phosphorylation of Lck targets, including ITAMs, because membrane domain localization of CD4−Lck pairs is regulated. This interpretation supports a variant of the TCR signaling paradigm in which CD4 sequesters Lck away from TCR–CD3 to prevent spurious signaling until reciprocal engagement of pMHCII by TCR–CD3 and CD4 allows Lck recruitment to the CD3 ITAMs (Glassman et al., 2018; Van Laethem et al., 2007). Varying CD4 palmitoylation would then allow for tuning of CD4 function.

Comparing the Clasp mutant with our intracellular helix mutants provides additional insights into the relationship between CD4−Lck interactions and CD4 function. First, although our LL mutation reduced CD4−Lck interactions to half of WT levels, we again did not observe clear evidence of reduced CD3ζ phosphorylation. Second, if we consider that only a small fraction of WT CD4 molecules are naturally paired with Lck (~6% in 58α⁻β⁻ cells) then we can intuit that the intracellular helices of those CD4 molecules that are not paired with Lck are free to mediate other function (Parrish et al., 2016; Stepanek et al., 2014). If we further consider that our Clasp mutant has an intact IKRLL motif within the intracellular helix that is generally not occupied by Lck, and is thus free to mediate those other functions, then we can infer from the increased IL-2 production driven by our H, HC, and LL mutants (which have ~25%, ~14%, and ~49% of WT levels of CD4−Lck association, respectively, but disrupted IKRLL motifs) that a key function of the IKRLL motif is to prevent CD4 molecules that are not paired with Lck from driving pMHCII-specific signaling. Indeed, comparing our Clasp and LL mutant signaling phenotypes implies that one potential function of the IKRLL motif is to inhibit free Lck from phosphorylating substrates other than ITAMs (e.g., Zap70). This would explain the conundrum of why our H, HC, and LL mutants relieve CD4−Lck interactions and yet increase IL-2 responses. Together, we consider the simplest interpretation of our data to be that the IKRLL motif of the intracellular helix is under purifying selection because it mediates CD4−Lck interactions, CD4 endocytosis, and performs a previously unreported inhibitory function. Our SS, pSS, and LL + pSS mutants all suggest that the phosphorylation states of the helix serines regulates these functions. We take these data as evidence that the multifunctional intracellular helix is a key regulator of pMHCII responses.

This conclusion is further supported by our finding of evolutionary covariation between proximal as well as distant residues and motifs. One example is the covariation of proximal residues within the intracellular helix, such as the serines and IKRLL motif, that regulate CD4−Lck interactions and CD4 inhibitory activity. For more distant residues, the GGXXG and CV +C co-arose in the predicted most recent common ancestor of eutherians, have been maintained under purifying selection, and together regulate CD4−Lck membrane domain localization. They also counterbalance the inhibitory activity of the ICD helix, suggesting a functional link that is supported by residue covariation between the GGXXG motif of the TMD and ICD helix. Moreover, our analyses provide evidence that interactions mediated by the ectodomain nonpolar patch, which enhances signaling by stabilizing TCR–CD3–pMHCII–CD4 assemblies on the outside of the cell, are coupled with the intracellular helix that regulates signaling (Glassman et al., 2018). Key residues in these motifs covaried over evolutionary time, the motifs co-arose in the predicted most recent common ancestors of mammals, and they have been conserved in marsupials and eutherians under purifying selection. Additionally, our data show that the intracellular helix and nonpolar patch counterbalance each other with regard to IL-2 production. The broader conclusion from these results is that the emergence of the CD4 intracellular helix, with its ability to counterbalance the signal-enhancing activity of other motifs, was a key moment in the regulation of mammalian CD4 function.

Understanding how these motifs work individually to influence eutherian CD4 function and CD4⁺ T cell biology in fine molecular detail represents fertile ground for future directions. Because GXXXG and GG motifs can interact with cholesterol, and palmitate can interact with protein-bound cholesterol (PDB 4IB4), coupling of rapidly reversible palmitoylation of the CV +C motif with a cholesterol-bound state of the GGXXG motif may allow finer regulation of CD4 membrane domain localization and function than can be achieved with a palmitoylation motif alone (Fessler, 2016; Song et al., 2014; Teese and Langosch, 2015; Wacker et al., 2013). Structural analysis will help test this hypothesis. For the intracellular helix, the IKRLL motif clearly has multiple interacting partners that dictate its activity. Some are known, such as Lck and AP2 (Kelly et al., 2008; Kim et al., 2003; Sleckman et al., 1992); yet our data lead us to hypothesize that the inhibitory activity of the helix is the result of one or more additional partners. The data here and elsewhere suggest that the preference of partners is likely to be regulated both by their relative abundance, membrane domain localization, and the phosphorylation state of the helix serines. In addition, given that T443 just outside the helix has covaried over evolution with L438 in the IKRLL motif as well as Q445 of the CQC clasp motif, it is reasonable to speculate that the phosphorylation state of this residue might further regulate interactions between the helix and/or clasp and their binding partners. Moreover, given that the CQC clasp links CD4 to Lck by coordinating a zinc, it is reasonable to consider if and how changes in zinc concentration regulate CD4−Lck interacts for their own function. Importantly, zinc-regulated changes in Lck association would impact occupation of the IKRLL motif by Lck, which would impact the ability of the intracellular helix to interact with other partners to exert other activity. In sum, CD4 function is likely to be regulated by the switch-like activity of multiple motifs in its ICD.

Future studies aimed at understanding how these motifs work together will allow us to better understand why some of them have covaried over evolutionary time. Two plausible, nonmutually exclusive modes can be envisioned for how individual motifs may work together within the network of counterbalancing activities described here. First, given the switch-like activity of the intracellular helix serines, palmitoylation sites in the CV +C motif, and zinc coordination of the CQC clasp, CD4 activity might be finetuned by the sum of the switch states of each motif in the way that the sum of digital states determines a computational output. Second, allosteric effects may also be at play. For example, extracellular interaction mediated by the D3 nonpolar patch might induce an allosteric change in the transmembrane GGXXG motif or intracellular helix that might impact binding partner preferences. These possibilities, which are not mutually exclusive, suggest just how complex the regulation of CD4 may be and give insights into the work that will be required to describe exactly how CD4 functions.

In closing, our multidisciplinary results highlight a network of function-regulating motifs within eutherian CD4 that would not otherwise be obvious. In so doing we extend a theme of counterposing activities that regulate eutherian pMHCII responses at the population level (e.g., helper and regulatory CD4⁺ T cells) and cellular level (e.g., costimulatory and coinhibitory molecules) to that of a single molecule (e.g., CD4 nonpolar patch and ICD helix). Furthermore, our data concerning the residue covariation and functional coordination of the ectodomain nonpolar patch and intracellular helix provide evidence that a multidomain transmembrane protein, which serves as one component of a multimodule receptor complex, can coevolve binding activity on the outside of a cell with regulatory activity on the inside of the cell to dictate the molecule’s function within the multimodule receptor complex. We expect this to emerge as a common theme for other complex transmembrane receptors tasked with relaying information across the membrane to the intracellular signaling machinery. Collectively, our results advance our understanding of T cell biology and have translational value given efforts to engineer synthetic receptors for therapeutic purposes, such as chimeric antigen receptors (CARs). While CAR-T cell therapy has shown considerable promise, problems with sensitivity and side effects now suggest that the absence of mechanisms to mediate or regulate the relay of information across the membrane have a fitness cost for this form of CAR (Labanieh et al., 2018). We therefore think that biomimetic designs, incorporating strategies refined over ~435 million years of iterative testing in a variety of vertebrates, will ultimately lead to more sensitive and reliable synthetic receptors (Kobayashi et al., 2020). Such biomimetic engineering will require a doubling down on basic research efforts to elucidate the evolutionary blueprint for key immune receptors.

Raw data, including alignments and phylogenetic trees, associated with figures 1 and S1 as well as source data and statistics for remaining figures are available on Dryad (https://doi.org/10.5061/dryad.59zw3r26z).

1. 1. Lee M
   2. Tuohy P
   3. Kim C
   4. Lichauco K
   5. Parrish H
   6. Van Doorslaer K
   7. Kuhns M

   (2022) Dryad Digital Repository

   Data associated with "Enhancing and inhibitory motifs have coevolved to regulate CD4 activity".

   https://doi.org/10.5061/dryad.59zw3r26z

1. 1. Ashkenazy H
   2. Erez E
   3. Martz E
   4. Pupko T
   5. Ben-Tal N

   (2010) ConSurf 2010: calculating evolutionary conservation in sequence and structure of proteins and nucleic acids

   Nucleic Acids Research 38:W529–W533.

   https://doi.org/10.1093/nar/gkq399

   • PubMed
   • Google Scholar
2. 1. Ashkenazy H
   2. Abadi S
   3. Martz E
   4. Chay O
   5. Mayrose I
   6. Pupko T
   7. Ben-Tal N

   (2016) ConSurf 2016: an improved methodology to estimate and visualize evolutionary conservation in macromolecules

   Nucleic Acids Research 44:W344–W350.

   https://doi.org/10.1093/nar/gkw408

   • PubMed
   • Google Scholar
3. 1. Babu MM
   2. van der Lee R
   3. de Groot NS
   4. Gsponer J

   (2011) Intrinsically disordered proteins: regulation and disease

   Current Opinion in Structural Biology 21:432–440.

   https://doi.org/10.1016/j.sbi.2011.03.011

   • PubMed
   • Google Scholar
4. 1. Balagopalan L
   2. Barr VA
   3. Samelson LE

   (2009) Endocytic events in TCR signaling: focus on adapters in microclusters

   Immunological Reviews 232:84–98.

   https://doi.org/10.1111/j.1600-065X.2009.00840.x

   • PubMed
   • Google Scholar
5. 1. Bernstein RM
   2. Schluter SF
   3. Bernstein H
   4. Marchalonis JJ

   (1996) Primordial emergence of the recombination activating gene 1 (RAG1): sequence of the complete shark gene indicates homology to microbial integrases

   PNAS 93:9454–9459.

   https://doi.org/10.1073/pnas.93.18.9454

   • PubMed
   • Google Scholar
6. 1. Brown CA
   2. Brown KS

   (2010) Validation of coevolving residue algorithms via pipeline sensitivity analysis: ELSC and OMES and ZNMI, oh my!

   PLOS ONE 5:e10779.

   https://doi.org/10.1371/journal.pone.0010779

   • PubMed
   • Google Scholar
7. 1. Buslje CM
   2. Santos J
   3. Delfino JM
   4. Nielsen M

   (2009) Correction for phylogeny, small number of observations and data redundancy improves the identification of coevolving amino acid pairs using mutual information

   Bioinformatics 25:1125–1131.

   https://doi.org/10.1093/bioinformatics/btp135

   • PubMed
   • Google Scholar
8. 1. Capra JA
   2. Singh M

   (2007) Predicting functionally important residues from sequence conservation

   Bioinformatics 23:1875–1882.

   https://doi.org/10.1093/bioinformatics/btm270

   • PubMed
   • Google Scholar
9. 1. Capra EJ
   2. Perchuk BS
   3. Lubin EA
   4. Ashenberg O
   5. Skerker JM
   6. Laub MT

   (2010) Systematic dissection and trajectory-scanning mutagenesis of the molecular interface that ensures specificity of two-component signaling pathways

   PLOS Genetics 6:e1001220.

   https://doi.org/10.1371/journal.pgen.1001220

   • PubMed
   • Google Scholar
10. 1. Chen Y
    2. Zhu Y
    3. Li X
    4. Gao W
    5. Zhen Z
    6. Dong D
    7. Huang B
    8. Ma Z
    9. Zhang A
    10. Song X
    11. Ma Y
    12. Guo C
    13. Zhang F
    14. Huang Z

    (2022) Cholesterol inhibits TCR signaling by directly restricting TCR-CD3 core tunnel motility

    Molecular Cell 82:1278–1287.

    https://doi.org/10.1016/j.molcel.2022.02.017

    • PubMed
    • Google Scholar
11. 1. Colell EA
    2. Iserte JA
    3. Simonetti FL
    4. Marino-Buslje C

    (2018) MISTIC2: comprehensive server to study coevolution in protein families

    Nucleic Acids Research 46:W323–W328.

    https://doi.org/10.1093/nar/gky419

    • PubMed
    • Google Scholar
12. 1. Courtney AH
    2. Lo WL
    3. Weiss A

    (2018) TCR signaling: mechanisms of initiation and propagation

    Trends in Biochemical Sciences 43:108–123.

    https://doi.org/10.1016/j.tibs.2017.11.008

    • PubMed
    • Google Scholar
13. 1. Crise B
    2. Rose JK

    (1992) Identification of palmitoylation sites on CD4, the human immunodeficiency virus receptor

    The Journal of Biological Chemistry 267:13593–13597.

    https://doi.org/10.1016/S0021-9258(18)42253-3

    • PubMed
    • Google Scholar
14. 1. Dunn SD
    2. Wahl LM
    3. Gloor GB

    (2008) Mutual information without the influence of phylogeny or entropy dramatically improves residue contact prediction

    Bioinformatics 24:333–340.

    https://doi.org/10.1093/bioinformatics/btm604

    • PubMed
    • Google Scholar
15. 1. Dyson HJ
    2. Wright PE

    (2005) Intrinsically unstructured proteins and their functions

    Nature Reviews. Molecular Cell Biology 6:197–208.

    https://doi.org/10.1038/nrm1589

    • PubMed
    • Google Scholar
16. 1. Fessler MB

    (2016) The intracellular cholesterol landscape: Dynamic integrator of the immune response

    Trends in Immunology 37:819–830.

    https://doi.org/10.1016/j.it.2016.09.001

    • PubMed
    • Google Scholar
17. 1. Flajnik MF

    (2014) Re-evaluation of the immunological Big Bang

    Current Biology 24:R1060–R1065.

    https://doi.org/10.1016/j.cub.2014.09.070

    • PubMed
    • Google Scholar
18. Preprint

    1. Foley G
    2. Mora A
    3. Ross CM
    4. Bottoms S
    5. Sützl L
    6. Lamprecht ML
    7. Zaugg J
    8. Essebier A
    9. Balderson B
    10. Newell R
    11. Thomson RES
    12. Kobe B
    13. Barnard RT
    14. Guddat L
    15. Schenk G
    16. Carsten J
    17. Gumulya Y
    18. Rost B
    19. Haltrich D
    20. Sieber V
    21. Gillam EMJ
    22. Bodén M

    (2020) Engineering Indel and Substitution Variants of Diverse and Ancient Enzymes Using Graphical Representation of Ancestral Sequence Predictions (GRASP)

    bioRxiv.

    https://doi.org/10.1101/2019.12.30.891457

    • Google Scholar
19. 1. Fragoso R
    2. Ren D
    3. Zhang X
    4. Su MWC
    5. Burakoff SJ
    6. Jin YJ

    (2003) Lipid raft distribution of CD4 depends on its palmitoylation and association with Lck, and evidence for CD4-induced lipid raft aggregation as an additional mechanism to enhance CD3 signaling

    Journal of Immunology 170:913–921.

    https://doi.org/10.4049/jimmunol.170.2.913

    • PubMed
    • Google Scholar
20. 1. Gaud G
    2. Lesourne R
    3. Love PE

    (2018) Regulatory mechanisms in T cell receptor signalling

    Nature Reviews. Immunology 18:485–497.

    https://doi.org/10.1038/s41577-018-0020-8

    • PubMed
    • Google Scholar
21. 1. Gibson TJ

    (2009) Cell regulation: determined to signal discrete cooperation

    Trends in Biochemical Sciences 34:471–482.

    https://doi.org/10.1016/j.tibs.2009.06.007

    • PubMed
    • Google Scholar
22. 1. Gil D
    2. Schamel WWA
    3. Montoya M
    4. Sánchez-Madrid F
    5. Alarcón B

    (2002) Recruitment of Nck by CD3 epsilon reveals a ligand-induced conformational change essential for T cell receptor signaling and synapse formation

    Cell 109:901–912.

    https://doi.org/10.1016/s0092-8674(02)00799-7

    • PubMed
    • Google Scholar
23. 1. Glaichenhaus N
    2. Shastri N
    3. Littman DR
    4. Turner JM

    (1991) Requirement for association of p56lck with CD4 in antigen-specific signal transduction in T cells

    Cell 64:511–520.

    https://doi.org/10.1016/0092-8674(91)90235-q

    • PubMed
    • Google Scholar
24. 1. Glassman CR
    2. Parrish HL
    3. Deshpande NR
    4. Kuhns MS

    (2016) The CD4 and CD3δε cytosolic juxtamembrane regions are proximal within a compact TCR-CD3-pMHC-CD4 macrocomplex

    Journal of Immunology 196:4713–4722.

    https://doi.org/10.4049/jimmunol.1502110

    • PubMed
    • Google Scholar
25. 1. Glassman CR
    2. Parrish HL
    3. Lee MS
    4. Kuhns MS

    (2018) Reciprocal TCR-CD3 and CD4 engagement of a nucleating pMHCII stabilizes a functional receptor macrocomplex

    Cell Reports 22:1263–1275.

    https://doi.org/10.1016/j.celrep.2017.12.104

    • PubMed
    • Google Scholar
26. 1. Gottschalk RA
    2. Corse E
    3. Allison JP

    (2010) TCR ligand density and affinity determine peripheral induction of Foxp3 in vivo

    The Journal of Experimental Medicine 207:1701–1711.

    https://doi.org/10.1084/jem.20091999

    • PubMed
    • Google Scholar
27. 1. Guy CS
    2. Vignali DAA

    (2009) Organization of proximal signal initiation at the TCR:CD3 complex

    Immunological Reviews 232:7–21.

    https://doi.org/10.1111/j.1600-065X.2009.00843.x

    • PubMed
    • Google Scholar
28. 1. Hochberg GKA
    2. Thornton JW

    (2017) Reconstructing ancient proteins to understand the causes of structure and function

    Annual Review of Biophysics 46:247–269.

    https://doi.org/10.1146/annurev-biophys-070816-033631

    • PubMed
    • Google Scholar
29. 1. Hopf TA
    2. Morinaga S
    3. Ihara S
    4. Touhara K
    5. Marks DS
    6. Benton R

    (2015) Amino acid coevolution reveals three-dimensional structure and functional domains of insect odorant receptors

    Nature Communications 6:6077.

    https://doi.org/10.1038/ncomms7077

    • PubMed
    • Google Scholar
30. 1. Hur EM
    2. Son M
    3. Lee OH
    4. Choi YB
    5. Park C
    6. Lee H
    7. Yun Y

    (2003) LIME, a novel transmembrane adaptor protein, associates with p56lck and mediates T cell activation

    The Journal of Experimental Medicine 198:1463–1473.

    https://doi.org/10.1084/jem.20030232

    • PubMed
    • Google Scholar
31. 1. Huseby ES
    2. White J
    3. Crawford F
    4. Vass T
    5. Becker D
    6. Pinilla C
    7. Marrack P
    8. Kappler JW

    (2005) How the T cell repertoire becomes peptide and MHC specific

    Cell 122:247–260.

    https://doi.org/10.1016/j.cell.2005.05.013

    • PubMed
    • Google Scholar
32. 1. Irvine DJ
    2. Purbhoo MA
    3. Krogsgaard M
    4. Davis MM

    (2002) Direct observation of ligand recognition by T cells

    Nature 419:845–849.

    https://doi.org/10.1038/nature01076

    • PubMed
    • Google Scholar
33. 1. Jones DT
    2. Taylor WR
    3. Thornton JM

    (1992) The rapid generation of mutation data matrices from protein sequences

    Computer Applications in the Biosciences 8:275–282.

    https://doi.org/10.1093/bioinformatics/8.3.275

    • PubMed
    • Google Scholar
34. 1. Katoh K
    2. Misawa K
    3. Kuma K
    4. Miyata T

    (2002) MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform

    Nucleic Acids Research 30:3059–3066.

    https://doi.org/10.1093/nar/gkf436

    • PubMed
    • Google Scholar
35. 1. Kelly BT
    2. McCoy AJ
    3. Späte K
    4. Miller SE
    5. Evans PR
    6. Höning S
    7. Owen DJ

    (2008) A structural explanation for the binding of endocytic dileucine motifs by the AP2 complex

    Nature 456:976–979.

    https://doi.org/10.1038/nature07422

    • PubMed
    • Google Scholar
36. 1. Kim KJ
    2. Kanellopoulos-Langevin C
    3. Merwin RM
    4. Sachs DH
    5. Asofsky R

    (1979)

    Establishment and characterization of BALB/c lymphoma lines with B cell properties

    Journal of Immunology 122:549–554.

    • PubMed
    • Google Scholar
37. 1. Kim PW
    2. Sun Z-YJ
    3. Blacklow SC
    4. Wagner G
    5. Eck MJ

    (2003) A zinc clasp structure tethers Lck to T cell coreceptors CD4 and CD8

    Science 301:1725–1728.

    https://doi.org/10.1126/science.1085643

    • PubMed
    • Google Scholar
38. 1. Kobayashi S
    2. Thelin MA
    3. Parrish HL
    4. Deshpande NR
    5. Lee MS
    6. Karimzadeh A
    7. Niewczas MA
    8. Serwold T
    9. Kuhns MS

    (2020) A biomimetic five-module chimeric antigen receptor (^5MCAR) designed to target and eliminate antigen-specific T cells

    PNAS 117:28950–28959.

    https://doi.org/10.1073/pnas.2012495117

    • PubMed
    • Google Scholar
39. 1. Kosakovsky Pond SL
    2. Frost SDW

    (2005) Not so different after all: a comparison of methods for detecting amino acid sites under selection

    Molecular Biology and Evolution 22:1208–1222.

    https://doi.org/10.1093/molbev/msi105

    • PubMed
    • Google Scholar
40. 1. Kosakovsky Pond SL
    2. Poon AFY
    3. Velazquez R
    4. Weaver S
    5. Hepler NL
    6. Murrell B
    7. Shank SD
    8. Magalis BR
    9. Bouvier D
    10. Nekrutenko A
    11. Wisotsky S
    12. Spielman SJ
    13. Frost SDW
    14. Muse SV

    (2020) HyPhy 2.5-A customizable platform for evolutionary hypothesis testing using phylogenies

    Molecular Biology and Evolution 37:295–299.

    https://doi.org/10.1093/molbev/msz197

    • PubMed
    • Google Scholar
41. 1. Kowarsch A
    2. Fuchs A
    3. Frishman D
    4. Pagel P

    (2010) Correlated mutations: a hallmark of phenotypic amino acid substitutions

    PLOS Computational Biology 6:e1000923.

    https://doi.org/10.1371/journal.pcbi.1000923

    • PubMed
    • Google Scholar
42. 1. Kuhns MS
    2. Davis MM

    (2012) TCR signaling emerges from the sum of many parts

    Frontiers in Immunology 3:159.

    https://doi.org/10.3389/fimmu.2012.00159

    • PubMed
    • Google Scholar
43. 1. Labanieh L
    2. Majzner RG
    3. Mackall CL

    (2018) Programming CAR-T cells to kill cancer

    Nature Biomedical Engineering 2:377–391.

    https://doi.org/10.1038/s41551-018-0235-9

    • PubMed
    • Google Scholar
44. 1. Ladygina N
    2. Martin BR
    3. Altman A

    (2011) Dynamic palmitoylation and the role of DHHC proteins in T cell activation and anergy

    Advances in Immunology 109:1–44.

    https://doi.org/10.1016/B978-0-12-387664-5.00001-7

    • PubMed
    • Google Scholar
45. 1. Larson SM
    2. Di Nardo AA
    3. Davidson AR

    (2000) Analysis of covariation in an SH3 domain sequence alignment: applications in tertiary contact prediction and the design of compensating hydrophobic core substitutions

    Journal of Molecular Biology 303:433–446.

    https://doi.org/10.1006/jmbi.2000.4146

    • PubMed
    • Google Scholar
46. 1. Lee MS
    2. Glassman CR
    3. Deshpande NR
    4. Badgandi HB
    5. Parrish HL
    6. Uttamapinant C
    7. Stawski PS
    8. Ting AY
    9. Kuhns MS

    (2015) A mechanical switch couples t cell receptor triggering to the cytoplasmic juxtamembrane regions of CD3ζζ

    Immunity 43:227–239.

    https://doi.org/10.1016/j.immuni.2015.06.018

    • PubMed
    • Google Scholar
47. Preprint

    1. Lee MS
    2. Tuohy PJ
    3. Kim C
    4. Lichauco K
    5. Parrish HL
    6. Doorslaer KV
    7. Kuhns MS

    (2021) Enhancing and Inhibitory Motifs Have Coevolved to Regulate CD4 Activity

    bioRxiv.

    https://doi.org/10.1101/2021.04.29.441928

    • Google Scholar
48. 1. Letourneur F
    2. Malissen B

    (1989) Derivation of a T cell hybridoma variant deprived of functional T cell receptor alpha and beta chain transcripts reveals a nonfunctional alpha-mRNA of BW5147 origin

    European Journal of Immunology 19:2269–2274.

    https://doi.org/10.1002/eji.1830191214

    • PubMed
    • Google Scholar
49. 1. Lockless SW
    2. Ranganathan R

    (1999) Evolutionarily conserved pathways of energetic connectivity in protein families

    Science 286:295–299.

    https://doi.org/10.1126/science.286.5438.295

    • PubMed
    • Google Scholar
50. 1. Marks DS
    2. Colwell LJ
    3. Sheridan R
    4. Hopf TA
    5. Pagnani A
    6. Zecchina R
    7. Sander C

    (2011) Protein 3D structure computed from evolutionary sequence variation

    PLOS ONE 6:e28766.

    https://doi.org/10.1371/journal.pone.0028766

    • PubMed
    • Google Scholar
51. 1. Martin LC
    2. Gloor GB
    3. Dunn SD
    4. Wahl LM

    (2005) Using information theory to search for co-evolving residues in proteins

    Bioinformatics 21:4116–4124.

    https://doi.org/10.1093/bioinformatics/bti671

    • PubMed
    • Google Scholar
52. 1. Parrish HL
    2. Glassman CR
    3. Keenen MM
    4. Deshpande NR
    5. Bronnimann MP
    6. Kuhns MS

    (2015) A transmembrane domain ggxxg motif in CD4 Contributes to its Lck-independent function but does not mediate CD4 dimerization

    PLOS ONE 10:e0132333.

    https://doi.org/10.1371/journal.pone.0132333

    • PubMed
    • Google Scholar
53. 1. Parrish HL
    2. Deshpande NR
    3. Vasic J
    4. Kuhns MS

    (2016) Functional evidence for TCR-intrinsic specificity for MHCII

    PNAS 113:3000–3005.

    https://doi.org/10.1073/pnas.1518499113

    • PubMed
    • Google Scholar
54. 1. Pike LJ

    (2006) Rafts defined: a report on the Keystone Symposium on Lipid Rafts and Cell Function

    Journal of Lipid Research 47:1597–1598.

    https://doi.org/10.1194/jlr.E600002-JLR200

    • PubMed
    • Google Scholar
55. 1. Pond SLK
    2. Frost SDW
    3. Muse SV

    (2005) HyPhy: hypothesis testing using phylogenies

    Bioinformatics 21:676–679.

    https://doi.org/10.1093/bioinformatics/bti079

    • PubMed
    • Google Scholar
56. 1. Popik W
    2. Alce TM

    (2004) CD4 receptor localized to non-raft membrane microdomains supports HIV-1 entry. Identification of a novel raft localization marker in CD4

    The Journal of Biological Chemistry 279:704–712.

    https://doi.org/10.1074/jbc.M306380200

    • PubMed
    • Google Scholar
57. 1. Price MN
    2. Dehal PS
    3. Arkin AP

    (2010) FastTree 2--approximately maximum-likelihood trees for large alignments

    PLOS ONE 5:e9490.

    https://doi.org/10.1371/journal.pone.0009490

    • PubMed
    • Google Scholar
58. 1. Reynolds KA
    2. McLaughlin RN
    3. Ranganathan R

    (2011) Hot spots for allosteric regulation on protein surfaces

    Cell 147:1564–1575.

    https://doi.org/10.1016/j.cell.2011.10.049

    • PubMed
    • Google Scholar
59. 1. Rudd CE

    (2021) How the discovery of the CD4/CD8-p56^lck complexes changed immunology and immunotherapy

    Frontiers in Cell and Developmental Biology 9:626095.

    https://doi.org/10.3389/fcell.2021.626095

    • PubMed
    • Google Scholar
60. 1. Sezgin E
    2. Levental I
    3. Mayor S
    4. Eggeling C

    (2017) The mystery of membrane organization: composition, regulation and roles of lipid rafts

    Nature Reviews. Molecular Cell Biology 18:361–374.

    https://doi.org/10.1038/nrm.2017.16

    • PubMed
    • Google Scholar
61. 1. Sleckman BP
    2. Shin J
    3. Igras VE
    4. Collins TL
    5. Strominger JL
    6. Burakoff SJ

    (1992) Disruption of the CD4-p56lck complex is required for rapid internalization of CD4

    PNAS 89:7566–7570.

    https://doi.org/10.1073/pnas.89.16.7566

    • PubMed
    • Google Scholar
62. 1. Song Y
    2. Kenworthy AK
    3. Sanders CR

    (2014) Cholesterol as a co-solvent and a ligand for membrane proteins

    Protein Science 23:1–22.

    https://doi.org/10.1002/pro.2385

    • PubMed
    • Google Scholar
63. 1. Stepanek O
    2. Prabhakar AS
    3. Osswald C
    4. King CG
    5. Bulek A
    6. Naeher D
    7. Beaufils-Hugot M
    8. Abanto ML
    9. Galati V
    10. Hausmann B
    11. Lang R
    12. Cole DK
    13. Huseby ES
    14. Sewell AK
    15. Chakraborty AK
    16. Palmer E

    (2014) Coreceptor scanning by the T cell receptor provides a mechanism for T cell tolerance

    Cell 159:333–345.

    https://doi.org/10.1016/j.cell.2014.08.042

    • PubMed
    • Google Scholar
64. 1. Teese MG
    2. Langosch D

    (2015) Role of GxxxG motifs in transmembrane domain interactions

    Biochemistry 54:5125–5135.

    https://doi.org/10.1021/acs.biochem.5b00495

    • PubMed
    • Google Scholar
65. 1. Tompa P

    (2011) Unstructural biology coming of age

    Current Opinion in Structural Biology 21:419–425.

    https://doi.org/10.1016/j.sbi.2011.03.012

    • PubMed
    • Google Scholar
66. 1. Tubo NJ
    2. Jenkins MK

    (2014) TCR signal quantity and quality in CD4+ T cell differentiation

    Trends in Immunology 35:591–596.

    https://doi.org/10.1016/j.it.2014.09.008

    • Google Scholar
67. 1. Turner JM
    2. Brodsky MH
    3. Irving BA
    4. Levin SD
    5. Perlmutter RM
    6. Littman DR

    (1990) Interaction of the unique N-terminal region of tyrosine kinase p56lck with cytoplasmic domains of CD4 and CD8 is mediated by cysteine motifs

    Cell 60:755–765.

    https://doi.org/10.1016/0092-8674(90)90090-2

    • PubMed
    • Google Scholar
68. 1. Van Laethem F
    2. Sarafova SD
    3. Park J-H
    4. Tai X
    5. Pobezinsky L
    6. Guinter TI
    7. Adoro S
    8. Adams A
    9. Sharrow SO
    10. Feigenbaum L
    11. Singer A

    (2007) Deletion of CD4 and CD8 coreceptors permits generation of alphabetaT cells that recognize antigens independently of the MHC

    Immunity 27:735–750.

    https://doi.org/10.1016/j.immuni.2007.10.007

    • PubMed
    • Google Scholar
69. 1. Van Laethem F
    2. Tikhonova AN
    3. Pobezinsky LA
    4. Tai X
    5. Kimura MY
    6. Le Saout C
    7. Guinter TI
    8. Adams A
    9. Sharrow SO
    10. Bernhardt G
    11. Feigenbaum L
    12. Singer A

    (2013) Lck availability during thymic selection determines the recognition specificity of the T cell repertoire

    Cell 154:1326–1341.

    https://doi.org/10.1016/j.cell.2013.08.009

    • PubMed
    • Google Scholar
70. 1. Venkatesh B
    2. Lee AP
    3. Ravi V
    4. Maurya AK
    5. Lian MM
    6. Swann JB
    7. Ohta Y
    8. Flajnik MF
    9. Sutoh Y
    10. Kasahara M
    11. Hoon S
    12. Gangu V
    13. Roy SW
    14. Irimia M
    15. Korzh V
    16. Kondrychyn I
    17. Lim ZW
    18. Tay B-H
    19. Tohari S
    20. Kong KW
    21. Ho S
    22. Lorente-Galdos B
    23. Quilez J
    24. Marques-Bonet T
    25. Raney BJ
    26. Ingham PW
    27. Tay A
    28. Hillier LW
    29. Minx P
    30. Boehm T
    31. Wilson RK
    32. Brenner S
    33. Warren WC

    (2014) Elephant shark genome provides unique insights into gnathostome evolution

    Nature 505:174–179.

    https://doi.org/10.1038/nature12826

    • PubMed
    • Google Scholar
71. 1. Wacker D
    2. Wang C
    3. Katritch V
    4. Han GW
    5. Huang X-P
    6. Vardy E
    7. McCorvy JD
    8. Jiang Y
    9. Chu M
    10. Siu FY
    11. Liu W
    12. Xu HE
    13. Cherezov V
    14. Roth BL
    15. Stevens RC

    (2013) Structural features for functional selectivity at serotonin receptors

    Science 340:615–619.

    https://doi.org/10.1126/science.1232808

    • PubMed
    • Google Scholar
72. 1. Waddell PJ
    2. Steel MA

    (1997) General time-reversible distances with unequal rates across sites: mixing gamma and inverse Gaussian distributions with invariant sites

    Molecular Phylogenetics and Evolution 8:398–414.

    https://doi.org/10.1006/mpev.1997.0452

    • PubMed
    • Google Scholar
73. 1. Weaver S
    2. Shank SD
    3. Spielman SJ
    4. Li M
    5. Muse SV
    6. Kosakovsky Pond SL

    (2018) Datamonkey 2.0: A modern web application for characterizing selective and other evolutionary processes

    Molecular Biology and Evolution 35:773–777.

    https://doi.org/10.1093/molbev/msx335

    • PubMed
    • Google Scholar
74. 1. Wei Q
    2. Brzostek J
    3. Sankaran S
    4. Casas J
    5. Hew LSQ
    6. Yap J
    7. Zhao X
    8. Wojciech L
    9. Gascoigne NRJ

    (2020) Lck bound to coreceptor is less active than free Lck

    PNAS 117:15809–15817.

    https://doi.org/10.1073/pnas.1913334117

    • PubMed
    • Google Scholar
75. 1. Willbold D
    2. Rösch P

    (1996) Solution structure of the human CD4 (403-419) receptor peptide

    Journal of Biomedical Science 3:435–441.

    https://doi.org/10.1007/BF02258047

    • PubMed
    • Google Scholar
76. 1. Wu H
    2. Kwong PD
    3. Hendrickson WA

    (1997) Dimeric association and segmental variability in the structure of human CD4

    Nature 387:527–530.

    https://doi.org/10.1038/387527a0

    • PubMed
    • Google Scholar

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting the paper "Enhancing and inhibitory motifs have coevolved to regulate CD4 activity" for consideration at eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Senior Editor. Although the work is of interest, we regret to inform you that the findings at this stage are too preliminary for further consideration at eLife.

All three reviewers agreed that the comparative genomics/evolution-motivated approach to identify new CD4 functions is intriguing and likely successfully identified 'novel' regions of CD4 that contribute to its function. However, there was also agreement that the major findings of the manuscript, while intriguing, fail to meet the significance bar for eLife due largely to a lack of mechanistic details of how distal regions of CD4 interact (thermodynamically or functionally) and control downstream TCR activation. In addition, the reviewers commented that the manuscript is written in a way that is largely inaccessible to non-experts that comprise much of the eLife readership and, as such, the broad and generalizable impact of the work was not apparent.

Reviewer #1 (Recommendations for the authors):

Lee and colleagues use evolutionary genomics to identify regions of CD4 that appear to have evolved under functional constraint and which were not previously appreciated to contribute to CD4 function. They proceed to test the function of these regions using in vitro assays to measure how mutations in these regions affect several aspects of CD4 biology including changes in CD4 localization on the membrane, interaction of CD4 with the downstream kinase Lck, and ultimately IL-2 production. The authors find effects on IL-2 production from mutations in all of identified motifs. Of interest, mutations in three of the regions tend to decrease IL-2 production, but mutations in one motif of the intracellular domain (termed "helix" or "IKRLL") increase IL-2 production. From this data, the authors also find evidence that interaction with the Lck kinase may not be necessary to stimulate IL-2 production, counter to the established model of CD4 signaling. Next, an analysis of statistical covariation in CD4 protein sequences showed significant covariation between the intracellular helix and most of the newly identified functional motifs in the intracellular, transmembrane, and extracellular domains. Finally, the authors combine mutations in the intracellular helix (which increase IL-2; "LL") with mutations in evolutionarily covarying regions (which decrease IL-2; "R426A, Clasp, TP") to measure the effects of the double mutants relative to each single mutant. All double mutants appear to behave as the additive effects of the individual mutations, without clear non-additive or epistatic effects.

Overall, the authors use a powerful combination of evolutionary analyses and biochemistry/molecular biology to identify novel functional regions of CD4. However, the author's conclusion that these motifs have coevolved to "finetune" the MHCII response are overstated and not supported by the functional data. Specifically, it is unclear what evidence the authors provide to support regulation of the ECD by intracellular or transmembrane regions. In figure 7, R426A, clasp, and TM mutants all decrease IL-2 on their own. The effects of these mutations in the background of the LL mutation seem to be simply additive, providing no evidence of interaction between these motifs (a la a network of coevolving residues). The covariation could reflect a 'rheostat' wherein mutations in a stimulatory domain are compensated by mutations in an enhancing domain and vice versa, but there need be no regulation in the sense of a perturbation/input at one motif affects the function of another/allostery.

To be suitable for a broad readership journal like eLife, the manuscript would benefit from a figure which shows the biological context (all the relevant molecules and the regions of interest in CD4). In addition, the diagrams included in the supplementary showing the mutations and their location within CD4 could be included in this or should be moved to the main figures. Similarly, some summary figure showing the mutations and their effects in a simple manner (up/down/wt) would be immensely useful to process the proposed interaction of mutations.

The relevance of the functions measured is not clear – what do we know about the function of raft vs non-raft CD4? Most CD4 is in DSM/non-raft fraction. Why use CTxB and fractionation? Do they tell different things? What about when they don't agree? A clear setup of the why these assays have been chosen is crucial to understanding the relevance of the data.

The language should be more precise regarding evolutionary inferences, with a clear delineation of what is a statistical test and what is an inference of those tests. For example:

– The pairwise conservation/mutual information analysis detects statistical covariation which suggests coevolution. This analysis does not show coevolution.

– They find statistical signatures of selection, not identify selection.

– (ln 104-105) "how ancient and ongoing environmental challenges have influenced CD4" should read "find motifs that have been preserved".

The various monikers used for each motif are hard to keep track of. It would greatly help the reader to have a single term to apply to each motif in the text and figures (e.g. Palm and CV+C, ECD and Domain 3).

Reviewer #2 (Recommendations for the authors):

"Enhancing and inhibitory motifs have coevolved to regulate CD4 activity" takes an evolutionary approach to investigate the mechanism(s) by which CD4 regulates TCR activation and downstream functional responses. The authors identify conserved motifs in the extracellular, transmembrane, and intracellular domains of CD4 that appear to regulate multiple aspects of its function, including its localization to lipid RAFTs, its capacity to interact with Lck, and its ability to promote IL-2 production. Notably, one of these motifs, comprising a helix just N-terminal to the cysteine clasp responsible for interaction with Lck, has an inhibitory effect on TCR signaling, and it seems to have coevolved in eutherians together with a motif in extracellular domain 3 that promotes T cell activation. Collectively, these results suggest that the regulation of CD4 activity has been finely tuned during evolution such that the acquisition of activity-promoting mutations is balanced by the emergence of inhibitory regions. Although this idea is interesting, it should be accompanied by more in-depth mechanistic work to demonstrate exactly how specific parts of CD4 control TCR activation. The notion that motifs within CD4 have coevolved is not particularly unexpected. These evolutionary relationships should be connected to mechanistic and functional insights into how the molecule works.

1) Just because motifs have coevolved and their mutations have additive effects doesn't necessarily mean that they are functionally coupled. Mechanistic insight will require a more in-depth analysis of membrane proximal signaling events and the activation status of downstream pathways (e.g. Erk, Ca, NFKB).

2) The idea that the role of CD4 is to control the colocalization of the TCR and Lck to RAFTs is quite interesting, and it should be tested. Could RAFT localization be modulated independently of CD4, and would this reverse the effects of the relevant CD4 mutations?

3) It is difficult to get a sense of what the fractionation and lipid association results really mean in terms of membrane organization (e.g. Figure 2). CTxB binding in the context of a bead-based IP is particularly unphysiological. A positive control would be helpful here. What would an established disruption of RAFT localization look like in this assay (e.g. a mutation in the Lck N-terminal domain).

4) Based on the presented data, one might conclude that mutation of the ICD helix enhances T cell responses simply by inhibiting CD4 internalization, thereby maintaining the density of CD4 on the cell surface. Indeed, all of the mutations that enhance IL-2 responses appear to result in higher surface expression. Do the authors favor this hypothesis, or can it be ruled out?

5) Given the evolutionary focus of this study, it would have been more interesting if the authors had actually used sequences from non-eutherian CD4s, as opposed to Ala and Gly substitutions. Was this attempted?

6) It is hypothesized that the importance of the GGXXG motif may depend on the local cholesterol concentration of the membrane domain in which it resides. Does changing membrane cholesterol modulate the effects of mutations in the GGXXG motif?

7) The authors should confirm that the CV+C mutation actually has the intended consequence of altering CD4 palmitoylation.

Reviewer #3 (Recommendations for the authors):

In this manuscript, Lee et al., the authors interrogate the function of CD4 from an evolutionary perspective. Adaptive immunity within the jawed vertebrates is believed to have arisen as a consequence of the "Immunological Big Bang". This event marks the evolutionary origin of both T and B cell receptors, as well as many of the molecules that mediate their signaling (e.g., the TCR co-receptors CD4 and CD8, as well as the MHC I and II proteins). Although orthologs of TCRs, CD4, and MHCII are broadly conserved throughout jawed vertebrates, these proteins have not yet been identified outside of this lineage. Notably, all three molecules first appear within the cartilaginous fishes, suggesting that their functions may be tightly linked. However, most studies on TCR function focus specifically on TCRs interacting directly with either CD4 or MHCII. In this study, the authors draw on evolutionary analyses to identify broadly conserved sequence motifs that may be involved in direct interactions between CD4 and MHCII to fine-tune the strength of TCR signaling. They complement these computational analyses with biochemical studies to support their hypothesis.

The primary strength of this manuscript lies in the author’s integrative approach; their evolutionary hypotheses are considerably strengthened by the findings from the experimental techniques. However, in its current form, the manuscript is very difficult to follow for a non-specialist in TCR signaling. The authors use a considerable amount of jargon and abbreviations. Consequently, the logic underlying their experiments is not always clear. Thus, while their biological findings may be of interest to researchers interested in the specifics of TCR signaling (particularly from the standpoints of engineering biomimetics and clinically manipulating TCR signaling strength), it is difficult to extend the significance of these findings more broadly.

As mentioned above, the manuscript contains a nearly impenetrable amount of abbreviations and jargon. I suggest that the authors have the manuscript revised by someone outside the field who is familiar with the work to address this problem. This will make the manuscript more considerably more accessible to a broad range of biologists.

Two additional points that should be addressed:

1. On line 55, the authors state that "[in sharks]… an orthologous gene encoding CD4

appears to be absent, as are genes for proteins associated with CD4⁺ 56 T cell helper (Th) or regulatory (Treg) functions (e.g. FoxP3 and Rorc) (Venkatesh et al., 2014)" Although this was the primary finding described in the original elephant shark genome paper, subsequent analysis of this genome (JM Dijkstra, Nature 511, 2014) revealed that the CD4⁺ T cell lineage is likely present within cartilaginous fish.

2. The manuscript relies heavily on evolutionary analysis of CD4 molecules collected from jawed vertebrates. However, the authors do not adequately describe if/how they verified the homology of these molecules. The authors state that "The criteria used for including orthologous CD4 sequences in our analyses were that they have a domain structure consisting of four extracellular Ig domains followed by a TMD and a C-terminal ICD". Many proteins have convergently evolved this domain architecture, which makes this criteria insufficient to assign orthology. A phylogenetic approach is required to confidently determine whether or not the sequences are truly orthologous to CD4. Additionally, all accession numbers and/or aligned sequences should be made freely available in the supplemental material.

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "Enhancing and inhibitory motifs regulate CD4 activity." for further consideration by eLife. Your revised article has been evaluated by Carla Rothlin (Senior Editor) and a Reviewing Editor.

The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

1) The reviewers feel that the new signaling data are conceptually the right experiments but may be technically limited by overstimulating T cells and sparsely sampling antigen concentrations in some assays. As discussed by Reviewer 3, despite the sparse dose response applied in the signaling experiments, the manuscript presents the conclusions as if there can be only one interpretation of the results.

For example, whether or not the Clasp plays a positive or negative role in TCR signaling is not settled science (which is why your manuscript is interesting). The reviewers don't feel like the added experiments completely resolve these questions, so we request that your either add additional experiments or appropriately temper the presentation and discussion of your conclusions to acknowledge the remaining gaps in our understanding of this important topic.

2) We understand the difficulty of crafting a narrative that is accessible to all readers but still respects the conventions and standards of both immunology and molecular evolution. While we appreciate the revisions made to improve the readability of the manuscript, we respectfully request that you have a colleague outside the evolution and immunology read the manuscript and highlight sections or terminology or acronyms that could be modified with the goal of increasing the accessibility of the narrative. The extent and substance of these revisions are left to your discretion.

Reviewer #1 (Recommendations for the authors):

Lee and colleagues use phylogenetics and amino acid conservation/covariation to identify previously unstudied motifs within CD4 that emerged in the last common ancestor of eutherians, statistically coevolved, and have contrasting effects on downstream signaling when mutated. Historically, the activity of CD4 on TCRs has been shown to be indirect – mediated by CD4 binding Lck and Lck phosphorylating CD3. The data in this manuscript (based on mutations in positions of CD4 shown by the authors to coevolve) suggest that CD4 also directly regulates or uses an unknown cofactor (not Lck) to regulate TCR signaling.

The revisions by the authors have significantly improved the manuscript by trimming potentially unsupported assertions about the interaction of distal patches on CD4 and addressing previous concerns regarding interpretation of the evolutionary analyses.. The added experiments to test the effects of various mutants on downstream signaling steps provide a limited glimpse of how the different regions of CD4 may be differentially regulating TCR signaling.

My main concern is still whether the very detailed and rigorous analysis of the interactions among CD4-Lck-CD3-TCR and the resulting findings are accessible enough for a broad readership to understand the potential impact of the finding. The evolutionary analyses are interesting, but not especially novel, so I think the impact of this paper is best judged in the context of its contribution to the CD4/TCR literature.

Reviewer #2 (Recommendations for the authors):In their manuscript, "Enhancing and inhibitory motifs regulate CD4 activity", Lee et al., employ an evolutionary approach to identify specific residues within CD4 that impact TCR signaling. CD4 is a transmembrane protein composed of four immunoglobulin domains that mediates T cell signaling in two ways: first, through direct interactions with MHC II on antigen presenting cells; and second, by facilitating intracellular signaling by interacting with the Src kinase, Lck. Traditionally, the researchers have focused on the TCR-MHC II interactions as the key determinants of T cell signaling. However, recent evidence suggests that CD4 may also regulate the outcome of T cell activation. In this study, the authors use phylogenetic analysis as a relatively unbiased means of defining residues within CD4 that are important for signaling. This is predicated on the assumption that protein evolution is constrained by function; residues that directly mediate signaling are less likely to mutate over time. CD4 is a central component of the adaptive immune system and its function is limited to this context. Accordingly, orthologs of CD4 have only been identified within vertebrates.In this study, the authors use computational analyses to characterize the evolution of CD4 within the jawed vertebrates as a foundation for functional experiments. Overall, the manuscript is well-written and the conclusions are supported by the data shown. This work could potentially be of interest to a broad group of researchers, including immunologists interested in adaptive immunity and comparative immunologists as well as cell biologists who are focused on transmembrane cell signaling proteins.

My primary concern with the manuscript is that, in its current form, it is difficult to interpret by a non-specialist in TCR signaling. The manuscript contains a considerable amount of jargon, abbreviations and specific details without the necessary context. I would suggest having the manuscript read by someone who is not familiar with the work and can highlight the more challenging sections.

In a similar vein, the figures are quite complex. The authors might consider simplifying to emphasize the main points.

Specific points to address are outlined below.

1. In the Introduction, the authors refer to the "immunological 'Big Bang' that gave rise to RAG-based antigen receptor gene rearrangement in jawed vertebrates…". This discounts several important advances in the evolution of adaptive immunity, including the characterization of the VLR-based adaptive immune system that evolved in parallel within the jawless vertebrates as well as the discovery of orthologs of the RAG1/2 and CDA enzymes within invertebrate deuterostomes. These findings, which are described and synthesized in Flajnik, 2014 (doi:10.1016/j.cub.2014.09.070), will provide important evolutionary context for this manuscript.

2. Line 69. The authors describe MHC-TCR signaling as mediated by "5 distinct modules", but it is not clear from the subsequent text what these modules are. This should be clarified.

3. Although the authors describe MHC-TCR signaling in great detail, the structure of CD4 is not described. This would be particularly useful to readers who are not specialists in T cell biology and might include a figure showing the four Ig domains, the transmembrane region, and the intracellular domain. The abbreviations ECD, TMD, and ICD should be clearly defined early in the text.

4. Line 115. The authors should briefly mention in the text which CD4 orthologs were used. For example, "X CD4 orthologs were analyzed from Y species, which included the vertebrate lineages …". This should be further elaborated in the methods section. The authors state (line 682) that "… available CD4 orthologs were downloaded from Genbank." More details are necessary. There are many transmembrane proteins that consist of four Ig domains; only a subset of these have been correctly annotated as CD4. The authors should specifically describe their inclusion/exclusion criteria (i.e., define the algorithms and parameters used and the "etc" on line 685). Accession numbers for all these sequences should be included in the manuscript.

5. Although most mammals have a single ortholog of CD4, due to whole genome duplications, many teleost genomes contain two paralogs (e.g., doi:10.4049/jimmunol.1600222). Were these included in the analysis? Conversely, the authors cite Venkatesh et al., 2014 as evidence of the absence of CD4 in sharks. This manuscript was followed by a comment (doi:10.1038/nature13446) that identified many of the "missing" components of adaptive immunity from the elephant shark genome sequence. It would be wise to double-check these findings.

6. Line 131. It is not clear how the distribution of residues subject to purifying selection suggests that "mutation putative linear motifs within the ICR is selected against". This should be clarified or justified.

Reviewer #3 (Recommendations for the authors):

This paper takes an evolutionary approach to investigate the mechanism(s) by which CD4 regulates TCR activation and downstream functional responses. The authors identify conserved motifs in the extracellular, transmembrane, and intracellular domains of CD4 that appear to regulate multiple aspects of its function, including its localization to lipid RAFTs, its capacity to interact with Lck, and its ability to elicit membrane proximal signaling and promote IL-2 production. Notably, one of these motifs, comprising a helix just N-terminal to the cysteine clasp responsible for interaction with Lck, has an inhibitory effect on TCR signaling, and it seems to have coevolved in eutherians together with a motif in extracellular domain 3 that promotes T cell activation. Collectively, these results suggest that the regulation of CD4 activity has been finely tuned during evolution such that the acquisition of activity-promoting mutations is balanced by the emergence of inhibitory regions.

The revisions and new data have improved this manuscript, although I still have some concerns.

1) The manuscript remains a difficult read. I'm not sure how best to deal with this. The acronyms get a bit overwhelming. Perhaps it would be better to remove some of the less used acronyms (e.g. MRCA) and just write the words out in full?

2) The new signaling data (Figure 6) are a welcome addition to the manuscript, but they are rather sparse. In particular, I am worried that the authors may be missing important phenotypes by overstimulating the TCR expressing cells. A dose response should be performed for each signaling readout. It's possible, for instance, that at a lower antigen dose they would observe more substantial differences in CD3z phosphorylation in Figure 6A. This would be consistent with the idea that CD4-Lck interaction matters more when antigenic peptide is limiting.

[Editors’ note: the authors resubmitted a revised version of the paper for consideration. What follows is the authors’ response to the first round of review.]

Reviewer #1 (Recommendations for the authors):

Lee and colleagues use evolutionary genomics to identify regions of CD4 that appear to have evolved under functional constraint and which were not previously appreciated to contribute to CD4 function. They proceed to test the function of these regions using in vitro assays to measure how mutations in these regions affect several aspects of CD4 biology including changes in CD4 localization on the membrane, interaction of CD4 with the downstream kinase Lck, and ultimately IL-2 production. The authors find effects on IL-2 production from mutations in all of identified motifs. Of interest, mutations in three of the regions tend to decrease IL-2 production, but mutations in one motif of the intracellular domain (termed "helix" or "IKRLL") increase IL-2 production. From this data, the authors also find evidence that interaction with the Lck kinase may not be necessary to stimulate IL-2 production, counter to the established model of CD4 signaling. Next, an analysis of statistical covariation in CD4 protein sequences showed significant covariation between the intracellular helix and most of the newly identified functional motifs in the intracellular, transmembrane, and extracellular domains. Finally, the authors combine mutations in the intracellular helix (which increase IL-2; "LL") with mutations in evolutionarily covarying regions (which decrease IL-2; "R426A, Clasp, TP") to measure the effects of the double mutants relative to each single mutant. All double mutants appear to behave as the additive effects of the individual mutations, without clear non-additive or epistatic effects.

Overall, the authors use a powerful combination of evolutionary analyses and biochemistry/molecular biology to identify novel functional regions of CD4. However, the author's conclusion that these motifs have coevolved to "finetune" the MHCII response are overstated and not supported by the functional data. Specifically, it is unclear what evidence the authors provide to support regulation of the ECD by intracellular or transmembrane regions. In figure 7, R426A, clasp, and TM mutants all decrease IL-2 on their own. The effects of these mutations in the background of the LL mutation seem to be simply additive, providing no evidence of interaction between these motifs (a la a network of coevolving residues).

Please note that we have removed the term “epistatic effects” from the manuscript. We previously used the term in a manner that we view as consistent with ‘epistatic’ interactions as described by Starr and Thornton (Starr and Thornton, 2016). They identify “permissive epistatic interactions [that] made the residue tolerable in its native background, that restrictive epistatic mutations made it intolerable in the other, or both.” In our system, the LL mutation leads to increased IL2 production. It is likely that the increased signaling that resulted in increased IL-2 would lower the fitness of an organism that acquired this mutation, although admittedly our current experimental setup does not allow us to test this directly. Individually, mutating the other motifs lowers the IL2 response. However, when combined with the LL mutation, the effects of mutating other motifs is indistinguishable from WT, indicating that these mutations counterbalance each other. To us, this suggests that the LL mutation, likely in combination with the non-polar patch with which it co-arose, restricted the evolutionary path of CD4 in mammals leading to the acquisition of the other regulatory motifs. Therefore, in our view, the LL and non-polar patch are examples of permissive epistatic interactions that allowed (and likely necessitated) the evolution of the other motifs identified in this study. We will explore this idea at some point in a future review where this speculative idea may be more appropriate.

The covariation could reflect a 'rheostat' wherein mutations in a stimulatory domain are compensated by mutations in an enhancing domain and vice versa, but there need be no regulation in the sense of a perturbation/input at one motif affects the function of another/allostery.

The comment that motifs with covarying residues could reflect a rheostat mediated by motifs with compensatory activity rather than allosteric interactions between the motifs is in line with our thinking. Indeed, we think it is more likely that residues in the nonpolar patch that stabilize TCR-CD3-pMHCII-CD4 assemblies are not allosterically interacting with the ICD helix but rather that interactions on the outside of the cell serve to dictate the spatial position of the CD4 ICD relative to the TCR-CD3 ICDs, and that the IKRLL motif of the ICD helix determines what the CD4 ICD interacts with and recruits to TCRCD3; however, we cannot rule out an allosteric effect. We did not intend to favor one or the other in the prior submission. Based on this helpful comment, we have tried in the revised Discussion section to be clear that experimentally determining the exact mechanistic basis for the function of the motifs we are studying individually, or in concert with each other, goes beyond the scope of the present study. In short, we’ve made observations that provide evidence of activity that has not been reported before, or even considered in play for CD4. These insights would have been hard to gain from less multi-disciplinary approaches and it is our hope that this study will serve as a hypothesis generator that opens up fertile ground for new discoveries.

To be suitable for a broad readership journal like eLife, the manuscript would benefit from a figure which shows the biological context (all the relevant molecules and the regions of interest in CD4). In addition, the diagrams included in the supplementary showing the mutations and their location within CD4 could be included in this or should be moved to the main figures. Similarly, some summary figure showing the mutations and their effects in a simple manner (up/down/wt) would be immensely useful to process the proposed interaction of mutations.

Thank you for the suggestions. We have now moved the structural model of CD4 to the main Figure 1 (Figure 1C), added Table 1 to describe the mutants, and added Table 2 to summarize the biochemical and functional impacts of the CD4 mutants relative to the WT. We were unclear on what exactly is being requested with regards to a figure showing the biological context, or where to call out such a figure in the text, as we read the suggestion as requesting the types of figures typically found in review articles. We have therefore refrained from making such a figure at this time. We can add one if needed. At issue is that we are unclear as to what would satisfy the requirement of “all relevant molecules” since there could be many. We did take care to overview many of the relevant molecules in the introduction (specifically, the third paragraph). Furthermore, the citations in that paragraph refer to several reviews with figures depicting molecules that are relevant to the current study. In particular, the reviews from the Weiss and Love Labs have what we think are nice illustrations that might satisfy the request (Courtney et al., 2018; Gaud et al., 2018). It is our hope that it is sufficient to reference these reviews for those within the broader community who are looking for more information in order to put our work in a broader context. Please let us know if you would prefer that we add a similar figure, and what level of detail is suitable.

The relevance of the functions measured is not clear – what do we know about the function of raft vs non-raft CD4? Most CD4 is in DSM/non-raft fraction. Why use CTxB and fractionation? Do they tell different things? What about when they don't agree? A clear setup of the why these assays have been chosen is crucial to understanding the relevance of the data.

First, we note that rafts are and have been extensively debated. We have adopted the definition outlined by Linda Pike as reported from the Keystone symposium on lipid rafts and cell function (Pike, 2006): “Membrane rafts are small (10–200 nm) heterogeneous, highly dynamic, sterol-and sphingolipid-enriched domains that compartmentalize cellular processes. Small rafts can sometimes be stabilized to form larger platforms through protein-protein and protein-lipid interactions.” Second, we note that it is now clear that cellular membranes are complex, with a variety of membrane rafts or islands with distinct lipid and protein compositions and that these play critical roles in cellular function (Lillemeier et al., 2010; Lillemeier et al., 2006).

What is known about the function of CD4 in membrane rafts is limited. One study reported a role in immunological synapse formation and another implicated an impact on intracellular signaling after antibody crosslinking (Balamuth et al., 2004; Fragoso et al., 2003). We have not seen studies that have related CD4 membrane raft localization to a downstream function, such as IL-2 production, in response to natural agonist pMHCII ligand as we have done here.

We used CTxB as a staining agent for GM1 (a ganglioside that is reported to concentrate in membrane rafts). Given that rafts are heterogenous, and sucrose gradient fractionation only segregates membranes and associated proteins into DRMs and DSMs, we used CTxB in combination with sucrose gradient fraction to ask if the motifs we are studying influences CD4 localization to distinct membrane domains within DRMs or DSMs. In other words, we interpret CTxB staining as providing additional information about the subdomains in which CD4 resides. Although this widely used approach is limited in the information it can provide, we think it is suitable for asking basic question about protein association with membrane subdomains. The results provide clear, albeit low resolution evidence that some of the motifs we are studying do influence membrane domain fractionation and thus justify future experimentation.

The language should be more precise regarding evolutionary inferences, with a clear delineation of what is a statistical test and what is an inference of those tests. For example:

– The pairwise conservation/mutual information analysis detects statistical covariation which suggests coevolution. This analysis does not show coevolution.

– They find statistical signatures of selection, not identify selection.

– (ln 104-105) "how ancient and ongoing environmental challenges have influenced CD4" should read "find motifs that have been preserved".

To make the manuscript accessible to a broad audience we lost precision in how we described the evolutionary assays and their potential interpretation. We appreciate the reviewer pointing this out and have carefully reworded the text to clarify the distinction between test and interpretation.

The various monikers used for each motif are hard to keep track of. It would greatly help the reader to have a single term to apply to each motif in the text and figures (e.g. Palm and CV+C, ECD and Domain 3).

We appreciate that it can be hard to keep track of what mutations have been made and where they reside within CD4. To aid the reader, we have moved our model CD4 structure to the main Figure 1 and added Table 1, which lists the mutant name, location, the motif mutated, and the specific residues that are mutated for easy reference.

Reviewer #2 (Recommendations for the authors):

"Enhancing and inhibitory motifs have coevolved to regulate CD4 activity" takes an evolutionary approach to investigate the mechanism(s) by which CD4 regulates TCR activation and downstream functional responses. The authors identify conserved motifs in the extracellular, transmembrane, and intracellular domains of CD4 that appear to regulate multiple aspects of its function, including its localization to lipid RAFTs, its capacity to interact with Lck, and its ability to promote IL-2 production. Notably, one of these motifs, comprising a helix just N-terminal to the cysteine clasp responsible for interaction with Lck, has an inhibitory effect on TCR signaling, and it seems to have coevolved in eutherians together with a motif in extracellular domain 3 that promotes T cell activation. Collectively, these results suggest that the regulation of CD4 activity has been finely tuned during evolution such that the acquisition of activity-promoting mutations is balanced by the emergence of inhibitory regions. Although this idea is interesting, it should be accompanied by more in-depth mechanistic work to demonstrate exactly how specific parts of CD4 control TCR activation. The notion that motifs within CD4 have coevolved is not particularly unexpected. These evolutionary relationships should be connected to mechanistic and functional insights into how the molecule works.

Regarding: “Although this idea is interesting, it should be accompanied by more indepth mechanistic work to demonstrate exactly how specific parts of CD4 control TCR activation.”

In consideration of this comment we have worked to rule in or out the most obvious mechanism, based on the TCR signaling paradigm and related models, for how the CD4 mutants with the most dramatic Lck and IL-2 phenotypes (Clasp, TP, and LL) influence TCR-CD3 signal initiation (Glaichenhaus et al., 1991; Rudd, 2021; Stepanek et al., 2014). The TCR signaling paradigm posits that CD4 contributes to TCR-CD3 signal initiation by recruiting Lck, via interactions at the CQC clasp, to phosphorylate the CD3 ITAMs when TCR-CD3 and CD4 both engage pMHCII. This model predicts that our Clasp and LL mutants, both of which reduce CD4-Lck association, should lead to reduced phosphorylation of the CD3z ITAMs. In contrast, because our TP mutant increases CD4-Lck association, it should lead to increased CD3z ITAM phosphorylation. We also analyzed phosphorylation of another substrate of Lck (Zap70), as well as an indirect downstream target (Plcg1), to gain deeper insights into how the Clasp, TP, and LL mutants impact TCR-CD3 proximal signaling events upstream of IL-2 production.

The results are presented in Figure 6. To summarize: 1) our Clasp mutant does not impact CD3z ITAM phosphorylation but does reduce Zap70 and Plcg1 phosphorylation compared to WT CD4; 2) our TP mutant reduces phosphorylation of all three molecules compared to WT CD4; 3) our LL mutant does not have a discernable impact on the phosphorylation of any of these signaling intermediates compared to WT CD4. Our interpretation of these results, and their implications, are detailed in the Discussion (for quick reference: the Clasp mutant (CQC motif) reduces CD4-Lck interactions and IL-2 production; the TP mutant (GGXXG plus CV+C motifs) increases CD4-Lck interactions, localizes CD4-Lck pairs to DSMs, and reduces IL-2; the LL mutant (IKRLL motif) decreases CD4-Lck interactions and increases IL-2 production.)

We will note here that the signaling events that originate at TCR-CD3-pMHCII-CD4 assemblies split into multiple signaling cascades, involving multiple intermediates and second messengers, to direct nuclear localization of multiple transcription factors that together drive IL-2 expression (our endpoint readout for pMHCII-specific signaling) (Courtney et al., 2018; Gaud et al., 2018; Malissen and Bongrand, 2015). We could form several hypothesize to explain how, for example, the IKRLL motif (mutated with our LL mutant) regulates TCR-CD3 signaling but we cannot test them all in one study to conclusively demonstrate how this one mutant functions. We think the most likely explanation for the inhibitory activity of the IKRLL motif is that it is mediated by an unidentified interacting partner. Finding this putative partner will require a screen and extensive characterization. Therefore, while we agree that further mechanistic analysis of the motifs described here is warranted, we ask the reviewer to please consider that demonstrating exactly how specific parts of CD4 control pMHCII-specific signaling could take years and numerous additional datasets. For this reason, we hope the reviewer will agree that doing so goes beyond the scope of the current study but that our body of data make a compelling case that further study is warranted.

Regarding: “The notion that motifs within CD4 have coevolved is not particularly unexpected.”

To our knowledge, the question has not been asked or investigated for CD4 or other components of multi-module activating immune receptors.

Regarding: “These evolutionary relationships should be connected to mechanistic and functional insights into how the molecule works.”

Our study was designed to describe evolutionary, mechanistic, and functional relationships by relating how the distinct motifs identified by our evolutionary analysis influence membrane domain localization, CD4 association with Lck, and IL-2 production in response to pMHCII. We chose membrane domain localization and Lck association because both have been proposed to regulate CD4 function, and we chose IL-2 production as an endpoint readout for differences in pMHCII-specific signaling (Fragoso et al., 2003; Glaichenhaus et al., 1991; Stepanek et al., 2014). We acknowledge that differences in IL-2 production can tell us that there is a difference in signaling between a particular mutant of CD4 and the WT molecule, but not where and in what pathway that defect lies. As detailed above, we have added analysis of TCR-CD3 proximal signaling events. We think that our results support the conclusion that CD4 function is more complex than can be explained by the TCR signaling paradigm and related models. We appreciate Reviewer #2’s view that more mechanistic work is needed and trust that, once our study is available to the field, the current study will help fuel interest in this topic.

1) Just because motifs have coevolved and their mutations have additive effects doesn't necessarily mean that they are functionally coupled. Mechanistic insight will require a more in-depth analysis of membrane proximal signaling events and the activation status of downstream pathways (e.g. Erk, Ca, NFKB).

We appreciate this comment. As with any study, we have put forth an interpretation of our data that we think is most consistent with all of the results, and other published work. We acknowledge that we cannot know with certainty why particular residues or motifs have coevolved. That being written, there is a growing body of evidence that points to a functional relationship between residues that coevolve, even if they are located at distant sites.

We have removed the term “functionally coupled” from the revision in consideration that there may be a difference in perspective regarding how it should be used or interpreted. We now use the term “counterbalance” to describe the neutral phenotype (equal to WT) we observe when we combine a motif that reduces IL-2 with one that enhances IL-2.

As discussed above, we have added additional mechanistic insights about our Clasp and TP mutants. We agree that more is needed and think that there is a long road ahead in understanding the function of the ICD helix, and IKRLL motif therein, as it is clearly a multifunctional hub with which other residues have covaried over evolutionary time. For the reasons outlined above, we think that providing the exact mechanistic details of how different motifs interact functionally requires that we first understand the function of the ICD helix, and IKRLL motif therein. We think that doing so goes beyond the scope of this study.

2) The idea that the role of CD4 is to control the colocalization of the TCR and Lck to RAFTs is quite interesting, and it should be tested. Could RAFT localization be modulated independently of CD4, and would this reverse the effects of the relevant CD4 mutations?

We agree that this is an interesting question and think that interrogating it properly would require a separate study. There are real technical challenges to doing this well.

3) It is difficult to get a sense of what the fractionation and lipid association results really mean in terms of membrane organization (e.g. Figure 2). CTxB binding in the context of a bead-based IP is particularly unphysiological. A positive control would be helpful here. What would an established disruption of RAFT localization look like in this assay (e.g. a mutation in the Lck N-terminal domain).

While membrane domains and their composition are clearly important for several biological processes, they are challenging to study. We used a classic, well-accepted biochemical technique (sucrose gradient fractionation) and a quantitative flow-based IP readout (basically a bead-based ELISA) to evaluate how the motifs we are studying impact segregation of CD4 molecules or CD4-Lck pairs into DRMs (membrane rafts) or DSMs (non-rafts)(Fragoso et al., 2003; Glassman et al., 2016; Parrish et al., 2016; Schrum et al., 2007). One reason for using this approach is that the impact of mutating some of the motifs studied here on sucrose gradient fractionation has already been published and thus those studies provide us with a benchmark for comparison [i.e. the CQC clasp motif, the CV+C palmitoylation motif, and the HXXR motif (shown in original submission but removed from the revision) (Fragoso et al., 2003; Popik and Alce, 2004)]. Those studies showed in single representative Western blots that mutating the CQC, CV+C, and HXXR motifs reduced CD4 localization to DRMs. Mutating these motifs yields similar results in our analysis. We hope that this comparison addresses the positive control comment.

We stained with CTxB because it binds the ganglioside GM1 that is reported to be enriched in membrane rafts. We agree that CTxB binding of bead-based IP is not physiological, although Western blot analysis of sucrose fractionated cell lysates is also not physiological. We looked at CTxB staining because we thought it might provide further insights into CD4 localization in membrane subdomains within DRMs and DSMs. We think the finding that our Clasp and Palm mutants have very different CTxB staining within the DRMs suggests that these two motifs control localization to distinct subdomains within DRMs for those CD4 molecules still localized in DRMs. We also think it is interesting that our TMD and Palm mutants have the same CTxB staining in DRMs as WT CD4 and that the TP mutant (TMD+Palm) has less CTxB staining that the Palm mutant. These data suggest that together, the GGXXG and CV+C motifs impact membrane domain, or subdomain, localization differently than either one individually. We do not know exactly what this means about membrane organization as the approach is admittedly low-resolution and our interpretation is limited. Nevertheless, we included the data in the manuscript because we think that some people in the field might find it interesting, useful, or otherwise hypothesis generating. We can remove the CTxB analysis from the manuscript if that is preferred.

4) Based on the presented data, one might conclude that mutation of the ICD helix enhances T cell responses simply by inhibiting CD4 internalization, thereby maintaining the density of CD4 on the cell surface. Indeed, all of the mutations that enhance IL-2 responses appear to result in higher surface expression. Do the authors favor this hypothesis, or can it be ruled out?

We have added additional data to address this point. Specifically, we have analyzed additional ICD helix mutants in Figure 4 to understand how serines flanking the IKRLL motif influence the inhibitory function of the IKRLL motif. Endocytosis of all of the ICD helix mutants is impaired, yet they have distinct IL-2 phenotypes suggesting that IL-2

production is more heavily influenced by the ICDs than expression levels. In addition, we included data with a C-terminally truncated CD4 molecule, termed CD4-T1 in accordance with previous labeling, for a mutant that is truncated immediately after the CV+C motif and thus lacks the remaining ICD, including the helix (see Table1 and Figure 4 —figure supplement 5) (Glaichenhaus et al., 1991). CD4-T1 expresses at equivalent levels to the LL mutant, and fails to endocytose, but does not direct increased IL-2 production relative to the WT the way the LL mutant does. Instead, it directs slightly reduced IL-2 production. These data clearly indicate that the differences in IL-2 production between the CD4-T1 and LL mutants are due to differences in their ICDs and not their expression levels at the initiation or throughout the co-culture with APCs.

Of additional note, we have moved data concerning the H and HC mutants to Figure 3. In revising the manuscript, we realized that the cell lines used for the flow cytometry and sucrose data were not the same as those used in the IL-2 (not generated on the same day). We have now adjusted the data such that all data is from a single set of cell lines generated at the same time. Specifically, we updated the flow cytometry and sucrose data because the cell lines shown throughout the revised figure was more closely matched in CD4 expression, which is why they had originally been chosen for the IL-2 data. Now, for all figures in the manuscript, the flow data, sucrose gradient data, and IL2 data are all from the same set of lines wherein the WT and mutants being compared were generated at the same time.

5) Given the evolutionary focus of this study, it would have been more interesting if the authors had actually used sequences from non-eutherian CD4s, as opposed to Ala and Gly substitutions. Was this attempted?

The goal of this study was to better understand CD4 in general, and eutherian CD4 in particular. We focused specifically on mouse and human CD4 since the former is an extensively-used model for the latter, which has obvious implications for immunotherapeutic engineering and human health. This is in accordance with our NIH funded project. Given our goal, we decided to take a more conventional structure function approach and use mutations that would disrupt the chemical nature of the motifs under investigation.

We think that replacing eutherian sequences with those from non-eutherian CD4 would also be incredibly interesting; however, such experiments would be performed in response to a different question with a different end-goal in mind. We also think this approach may be harder to interpret because the non-eutherian CD4 homologous evolved along a different trajectory, in potentially different cellular environments (e.g. temperature, membrane composition). We do not know how to predict how non eutherian motifs would behave in the background of mouse CD4 in mouse cells. We also think there may be many caveats associated with interpreting such data. Overall, we think this is a different question that would be worthy of a separate study.

6) It is hypothesized that the importance of the GGXXG motif may depend on the local cholesterol concentration of the membrane domain in which it resides. Does changing membrane cholesterol modulate the effects of mutations in the GGXXG motif?

Changing membrane cholesterol levels or interfering in other ways is expected to have effects beyond CD4, making any results challenging to interpret (Chen et al., 2022; Swamy et al., 2016; Wang et al., 2016). Therefore, we thought that speculating about the interplay between cholesterol concentrations and the GGXXG motif was more appropriate for the Discussion.

7) The authors should confirm that the CV+C mutation actually has the intended consequence of altering CD4 palmitoylation.

Because previous publications have established that the cysteines of the CD4 CV+C motif are plamitoylated, and other studies have shown that similar motifs are also palmitoylated in other proteins, we consider this to be well established in the field (Arcaro et al., 2000; Crise and Rose, 1992; Fragoso et al., 2003; Ladygina et al., 2011).

Reviewer #3 (Recommendations for the authors):

In this manuscript, Lee et al., the authors interrogate the function of CD4 from an evolutionary perspective. Adaptive immunity within the jawed vertebrates is believed to have arisen as a consequence of the "Immunological Big Bang". This event marks the evolutionary origin of both T and B cell receptors, as well as many of the molecules that mediate their signaling (e.g., the TCR co-receptors CD4 and CD8, as well as the MHC I and II proteins). Although orthologs of TCRs, CD4, and MHCII are broadly conserved throughout jawed vertebrates, these proteins have not yet been identified outside of this lineage. Notably, all three molecules first appear within the cartilaginous fishes, suggesting that their functions may be tightly linked. However, most studies on TCR function focus specifically on TCRs interacting directly with either CD4 or MHCII. In this study, the authors draw on evolutionary analyses to identify broadly conserved sequence motifs that may be involved in direct interactions between CD4 and MHCII to fine-tune the strength of TCR signaling. They complement these computational analyses with biochemical studies to support their hypothesis.

The primary strength of this manuscript lies in the author’s integrative approach; their evolutionary hypotheses are considerably strengthened by the findings from the experimental techniques. However, in its current form, the manuscript is very difficult to follow for a non-specialist in TCR signaling. The authors use a considerable amount of jargon and abbreviations. Consequently, the logic underlying their experiments is not always clear. Thus, while their biological findings may be of interest to researchers interested in the specifics of TCR signaling (particularly from the standpoints of engineering biomimetics and clinically manipulating TCR signaling strength), it is difficult to extend the significance of these findings more broadly.

See specific comments below.

As mentioned above, the manuscript contains a nearly impenetrable amount of abbreviations and jargon. I suggest that the authors have the manuscript revised by someone outside the field who is familiar with the work to address this problem. This will make the manuscript more considerably more accessible to a broad range of biologists.

We understand and agree that field-specific nomenclature can, at times, feel like a foreign language to those outside of the field. Indeed, as an interdisciplinary collaboration, our team has had to spend time and effort learning each other’s language. This has been challenging, but also fun. We appreciate the suggestion and will note that we did have colleagues who understand the evolutionary analysis but not the immunology, and vice versa, read the manuscript and provide comments. Their suggestions were incorporated into the initial submission as we anticipated that we were likely to get one or more comments such as this and can assure the reviewer we have tried to make the manuscript accessible. We will note that our attempts backfired somewhat as Reviewer #1 took issue with the precision of our language regarding our evolutionary analysis which was intended to make the work more accessible.

We ask the reviewer to consider that we are constrained in how we relate our work to the reader. The immunological terms, including names of the cell types and proteins, must be kept consistent with the field in order for people inside and outside the field to be able to relate our work to that of others in the field. And, as called out by Reviewer #1, we must also use precision when discussing the evolutionary aspects of the work. In short, we think that we are generally obliged to stick to convention when discussing key aspects of this work. With regards to other nomenclature, or jargon, we have tried to use standard terms. For example, we used ECD, TMD, and ICD to refer to the extracellular, transmembrane, and intracellular domains of a protein as is common in biochemistry and molecular and cellular biology because these abbreviations should be broadly accessible. Similarly, referring to the detergent resistant and detergent soluble membrane fractions of sucrose gradients as DRMs and DSMs is conventional. We would be happy to use the long form instead of the abbreviations if Reviewer #3 and the editor think it would be more appropriate and we are not constrained by space.

Finally, we acknowledge that the mutant names may be a problem as we sometimes have a hard time keeping track of all the mutants. For this reason, we chose to name our mutants in a way that relates to their location or particular motif. For example, we found that calling the CQC clasp motif mutant “Clasp” and our CV+C palmitoylation motif mutant “Palm” was easier for us to keep track of than using a residue-based naming system such as C444S+C446S and C418S+C421S, respectively. We have now added Table #1 as a quick reference for the mutant names, locations, residue number being changed, and the nature of the mutation. We hope that this helps.

Two additional points that should be addressed:

1. On line 55, the authors state that "[in sharks]… an orthologous gene encoding CD4

appears to be absent, as are genes for proteins associated with CD4⁺ 56 T cell helper (Th) or regulatory (Treg) functions (e.g. FoxP3 and Rorc) (Venkatesh et al., 2014)" Although this was the primary finding described in the original elephant shark genome paper, subsequent analysis of this genome (JM Dijkstra, Nature 511, 2014) revealed that the CD4⁺ T cell lineage is likely present within cartilaginous fish.

We appreciate this comment and have removed this mention of the Venkatesh paper from our manuscript due to the controversial nature of their finding.

2. The manuscript relies heavily on evolutionary analysis of CD4 molecules collected from jawed vertebrates. However, the authors do not adequately describe if/how they verified the homology of these molecules. The authors state that "The criteria used for including orthologous CD4 sequences in our analyses were that they have a domain structure consisting of four extracellular Ig domains followed by a TMD and a C-terminal ICD". Many proteins have convergently evolved this domain architecture, which makes this criteria insufficient to assign orthology. A phylogenetic approach is required to confidently determine whether or not the sequences are truly orthologous to CD4. Additionally, all accession numbers and/or aligned sequences should be made freely available in the supplemental material.

As the reviewer points out, it is hard to demonstrate that proteins are orthologous. Our dataset is based on reciprocal blast-based searches of the NCBI database. Sequences were aligned and used to construct phylogenetic trees. Sequences that either did not have a canonical CD4 structure or contained large indels were removed from the analysis. Likewise, we assumed that the evolution of CD4 should reflect that of the host species. Sequences that did not cluster as expected were removed from the dataset. The final dataset, the aligned sequences, and phylogenetic trees are available from Dryad (https://doi.org/10.5061/dryad.59zw3r26z).

References cited in our response to reviewers:

Arcaro, A., Gregoire, C., Boucheron, N., Stotz, S., Palmer, E., Malissen, B., and Luescher, I.F. (2000). Essential role of CD8 palmitoylation in CD8 coreceptor function. J Immunol 165, 2068-2076.

Balamuth, F., Brogdon, J.L., and Bottomly, K. (2004). CD4 raft association and signaling regulate molecular clustering at the immunological synapse site. J Immunol 172, 58875892.

Chen, Y., Zhu, Y., Li, X., Gao, W., Zhen, Z., Dong, Huang, B., Ma, Z., Zhang, A., Song, X., et al. (2022). Cholesterol inhibits TCR signaling by directly restricting TCR-CD3 core tunnel motility. Mol Cell.

Courtney, A.H., Lo, W.L., and Weiss, A. (2018). TCR Signaling: Mechanisms of Initiation and Propagation. Trends in biochemical sciences 43, 108-123.

Crise, B., and Rose, J.K. (1992). Identification of palmitoylation sites on CD4, the human immunodeficiency virus receptor. J Biol Chem 267, 13593-13597.

Davey, N.E., Cyert, M.S., and Moses, A.M. (2015). Short linear motifs – ex nihilo evolution of protein regulation. Cell Commun Signal 13, 43.

Fragoso, R., Ren, D., Zhang, X., Su, M.W., Burakoff, S.J., and Jin, Y.J. (2003). Lipid raft distribution of CD4 depends on its palmitoylation and association with Lck, and

evidence for CD4-induced lipid raft aggregation as an additional mechanism to enhance CD3 signaling. J Immunol 170, 913-921.

Gaud, G., Lesourne, R., and Love, P.E. (2018). Regulatory mechanisms in T cell receptor signalling. Nat Rev Immunol 18, 485-497.

Glaichenhaus, N., Shastri, N., Littman, D.R., and Turner, J.M. (1991). Requirement for association of p56lck with CD4 in antigen-specific signal transduction in T cells. Cell 64, 511-520.

Glassman, C.R., Parrish, H.L., Deshpande, N.R., and Kuhns, M.S. (2016). The CD4 and CD3deltaepsilon Cytosolic Juxtamembrane Regions Are Proximal within a Compact TCR-CD3-pMHC-CD4 Macrocomplex. J Immunol 196, 4713-4722.

Kim, P.W., Sun, Z.Y., Blacklow, S.C., Wagner, G., and Eck, M.J. (2003). A zinc clasp structure tethers Lck to T cell coreceptors CD4 and CD8. Science 301, 1725-1728.

Kobayashi, S., Thelin, M.A., Parrish, H.L., Deshpande, N.R., Lee, M.S., Karimzadeh, A., Niewczas, M.A., Serwold, T., and Kuhns, M.S. (2020). A biomimetic five-module chimeric antigen receptor ((5M)CAR) designed to target and eliminate antigen-specific T cells. Proc Natl Acad Sci U S A 117, 28950-28959.

Ladygina, N., Martin, B.R., and Altman, A. (2011). Dynamic palmitoylation and the role of DHHC proteins in T cell activation and anergy. Adv Immunol 109, 1-44.

Lillemeier, B.F., Mortelmaier, M.A., Forstner, M.B., Huppa, J.B., Groves, J.T., and Davis, M.M. (2010). TCR and Lat are expressed on separate protein islands on T cell membranes and concatenate during activation. Nature immunology 11, 90-96.

Lillemeier, B.F., Pfeiffer, J.R., Surviladze, Z., Wilson, B.S., and Davis, M.M. (2006). Plasma membrane-associated proteins are clustered into islands attached to the cytoskeleton. Proc Natl Acad Sci U S A 103, 18992-18997.

Malissen, B., and Bongrand, P. (2015). Early T cell activation: integrating biochemical, structural, and biophysical cues. Annu Rev Immunol 33, 539-561.

Parrish, H.L., Deshpande, N.R., Vasic, J., and Kuhns, M.S. (2016). Functional evidence for TCR-intrinsic specificity for MHCII. Proc Natl Acad Sci U S A.

Parrish, H.L., Glassman, C.R., Keenen, M.M., Deshpande, N.R., Bronnimann, M.P., and

Kuhns, M.S. (2015). A Transmembrane Domain GGxxG Motif in CD4 Contributes to Its Lck-Independent Function but Does Not Mediate CD4 Dimerization. PLoS One 10, e0132333.

Pike, L.J. (2006). Rafts defined: a report on the Keystone Symposium on Lipid Rafts and Cell Function. J Lipid Res 47, 1597-1598.

Popik, W., and Alce, T.M. (2004). CD4 receptor localized to non-raft membrane microdomains supports HIV-1 entry. Identification of a novel raft localization marker in CD4. J Biol Chem 279, 704-712.

Rudd, C.E. (2021). How the Discovery of the CD4/CD8-p56(lck) Complexes Changed Immunology and Immunotherapy. Front Cell Dev Biol 9, 626095.

Schrum, A.G., Gil, D., Dopfer, E.P., Wiest, D.L., Turka, L.A., Schamel, W.W., and Palmer, E. (2007). High-sensitivity detection and quantitative analysis of native proteinprotein interactions and multiprotein complexes by flow cytometry. Sci STKE 2007, pl2.

Starr, T.N., and Thornton, J.W. (2016). Epistasis in protein evolution. Protein Sci 25, 1204-1218.

Stepanek, O., Prabhakar, A.S., Osswald, C., King, C.G., Bulek, A., Naeher, D., BeaufilsHugot, M., Abanto, M.L., Galati, V., Hausmann, B., et al. (2014). Coreceptor scanning by the T cell receptor provides a mechanism for T cell tolerance. Cell 159, 333-345.

Swamy, M., Beck-Garcia, K., Beck-Garcia, E., Hartl, F.A., Morath, A., Yousefi, O.S., Dopfer, E.P., Molnar, E., Schulze, A.K., Blanco, R., et al. (2016). A Cholesterol-Based Allostery Model of T Cell Receptor Phosphorylation. Immunity 44, 1091-1101.

Wang, F., Beck-Garcia, K., Zorzin, C., Schamel, W.W., and Davis, M.M. (2016). Inhibition of T cell receptor signaling by cholesterol sulfate, a naturally occurring derivative of membrane cholesterol. Nat Immunol 17, 844-850.

[Editors’ note: what follows is the authors’ response to the second round of review.]

The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

1) The reviewers feel that the new signaling data are conceptually the right experiments but may be technically limited by overstimulating T cells and sparsely sampling antigen concentrations in some assays. As discussed by Reviewer 3, despite the sparse dose response applied in the signaling experiments, the manuscript presents the conclusions as if there can be only one interpretation of the results.

For example, whether or not the Clasp plays a positive or negative role in TCR signaling is not settled science (which is why your manuscript is interesting). The reviewers don't feel like the added experiments completely resolve these questions, so we request that your either add additional experiments or appropriately temper the presentation and discussion of your conclusions to acknowledge the remaining gaps in our understanding of this important topic.

We appreciate this comment and have taken your suggestions to modify the manuscript. In brief:

Please note that we have now also added analysis of two additional independently generated WT and Clasp mutant cell line pairs to our proximal signaling analysis in Figure 6A for a total of four independent generated lines. While we found no difference in the pCD3z MFI compared to the WT for the two independently generated cell lines we originally tested, one of the two additional cell line showed pCD3z MFI that was ~83% of WT. The other line showed a very small reduction in pCD3z MFI (~92% of WT), which is nevertheless statistically significant. The key point remains that we do not see an absence of pCD3z MFI with the Clasp mutant, nor do we see a consistent reduction in pCD3z MFI as is expected based on the prevailing paradigm.

Thank you. We followed this suggestion and asked a colleague, Dr. Benjamin Renquist who works on obesity in mice, to read the manuscript and provide suggestions on how to make the manuscript more accessible. We greatly appreciate the gift of his time and have worked to incorporate his suggestions into the manuscript.

We understand the difficulty of crafting a narrative that is accessible to all readers but still respects the conventions and standards of both immunology and molecular evolution. While we appreciate the revisions made to improve the readability of the manuscript, we respectfully request that you have a colleague outside the evolution and immunology read the manuscript and highlight sections or terminology or acronyms that could be modified with the goal of increasing the accessibility of the narrative. The extent and substance of these revisions are left to your discretion.

Reviewer #2 (Recommendations for the authors):

In their manuscript, "Enhancing and inhibitory motifs regulate CD4 activity", Lee et al., employ an evolutionary approach to identify specific residues within CD4 that impact TCR signaling. CD4 is a transmembrane protein composed of four immunoglobulin domains that mediates T cell signaling in two ways: first, through direct interactions with MHC II on antigen presenting cells; and second, by facilitating intracellular signaling by interacting with the Src kinase, Lck. Traditionally, the researchers have focused on the TCR-MHC II interactions as the key determinants of T cell signaling. However, recent evidence suggests that CD4 may also regulate the outcome of T cell activation. In this study, the authors use phylogenetic analysis as a relatively unbiased means of defining residues within CD4 that are important for signaling. This is predicated on the assumption that protein evolution is constrained by function; residues that directly mediate signaling are less likely to mutate over time. CD4 is a central component of the adaptive immune system and its function is limited to this context. Accordingly, orthologs of CD4 have only been identified within vertebrates.

In this study, the authors use computational analyses to characterize the evolution of CD4 within the jawed vertebrates as a foundation for functional experiments. Overall, the manuscript is well-written and the conclusions are supported by the data shown. This work could potentially be of interest to a broad group of researchers, including immunologists interested in adaptive immunity and comparative immunologists as well as cell biologists who are focused on transmembrane cell signaling proteins.

My primary concern with the manuscript is that, in its current form, it is difficult to interpret by a non-specialist in TCR signaling. The manuscript contains a considerable amount of jargon, abbreviations and specific details without the necessary context. I would suggest having the manuscript read by someone who is not familiar with the work and can highlight the more challenging sections.

We understand the comment and thank you for the suggestion. We did have another colleague, Dr. Benjamin Renquist who studies obesity in mouse models, read the paper to provide additional critical feedback. He also indicated that the numerous acronyms and sheer number of mutants made for a tough read. He found that he could navigate the manuscript with the aid of Figure 1C and Table 1, suggested by the reviewers (thank you), and he suggested as with Reviewer #3 that we should take the space to spell out key acronyms. We have now done this. He also suggested adding an acronym key to Table 2, which we have now done. In addition, following this line of thinking, we have added acronym keys to the figure legends.

In a similar vein, the figures are quite complex. The authors might consider simplifying to emphasize the main points.

We also appreciate this comment but are unsure how to further modify them beyond what we have already done. As noted above, we did add acronym keys to the figure legends.

Specific points to address are outlined below.

1. In the Introduction, the authors refer to the "immunological 'Big Bang' that gave rise to RAG-based antigen receptor gene rearrangement in jawed vertebrates…". This discounts several important advances in the evolution of adaptive immunity, including the characterization of the VLR-based adaptive immune system that evolved in parallel within the jawless vertebrates as well as the discovery of orthologs of the RAG1/2 and CDA enzymes within invertebrate deuterostomes. These findings, which are described and synthesized in Flajnik, 2014 (doi:10.1016/j.cub.2014.09.070), will provide important evolutionary context for this manuscript.

We appreciate the comments and suggestion. We think these are all important evolutionary developments in adaptive immunity that serve as a nice introduction to the paper from “30,000ft” before zooming in specifically on the evolution of CD4 and its role in T cell biology. We have now cited the Flajnik paper to point interested readers to a key resource for a broader sense of these important issues. We now take more care to be clear to state that this paper is focused on adaptive immunity in jawed vertebrates.

2. Line 69. The authors describe MHC-TCR signaling as mediated by "5 distinct modules", but it is not clear from the subsequent text what these modules are. This should be clarified.

We appreciate this suggestion and have now more carefully described each module.

3. Although the authors describe MHC-TCR signaling in great detail, the structure of CD4 is not described. This would be particularly useful to readers who are not specialists in T cell biology and might include a figure showing the four Ig domains, the transmembrane region, and the intracellular domain. The abbreviations ECD, TMD, and ICD should be clearly defined early in the text.

Figure 1C shows a model of CD4 with the elements highlighted as suggested. We have also added an acronym key to the figure legend to aid the reader in understanding the figure.

4. Line 115. The authors should briefly mention in the text which CD4 orthologs were used. For example, "X CD4 orthologs were analyzed from Y species, which included the vertebrate lineages …". This should be further elaborated in the methods section. The authors state (line 682) that "… available CD4 orthologs were downloaded from Genbank."

We have updated to text to read:

We performed multiple analyses of available vertebrate CD4 ortholog sequences (n=99 distinct sequences), representing ~435 million years of evolution, to understand how ancient and ongoing environmental challenges have influenced CD4. The analyzed sequences represent fish, reptiles (including birds), marsupials, and placental mammals. Details related to ortholog selection are outlined in Materials and methods. All sequences and files are available through the DataDryad repository associated with this manuscript.

More details are necessary. There are many transmembrane proteins that consist of four Ig domains; only a subset of these have been correctly annotated as CD4. The authors should specifically describe their inclusion/exclusion criteria (i.e., define the algorithms and parameters used and the "etc" on line 685).

Available CD4 orthologs were identified through reciprocal blast-based searches and downloaded from GenBank. BLAST may not only identify orthologs so additional criteria were used for including putative orthologous CD4 sequences in our analyses: presence of a domain structure consisting of four extracellular Ig domains followed by a transmembrane domain and a C-terminal intracellular domain, including the presence of the Lck binding clasp (CxC). Sequences that were shorter, contained frameshift mutations, or displayed high sequence variability were excluded from the analysis.

Accession numbers for all these sequences should be included in the manuscript.

Accession numbers are available through the DataDryad repository associated with this manuscript.

5. Although most mammals have a single ortholog of CD4, due to whole genome duplications, many teleost genomes contain two paralogs (e.g., doi:10.4049/jimmunol.1600222). Were these included in the analysis?

Based on the criteria described above we filtered out potential duplicates.

Conversely, the authors cite Venkatesh et al., 2014 as evidence of the absence of CD4 in sharks. This manuscript was followed by a comment (doi:10.1038/nature13446) that identified many of the "missing" components of adaptive immunity from the elephant shark genome sequence. It would be wise to double-check these findings.

We removed the statement in question and reference to the Venkatesh paper.

6. Line 131. It is not clear how the distribution of residues subject to purifying selection suggests that "mutation putative linear motifs within the ICR is selected against". This should be clarified or justified.

The data demonstrates that 45% of the residues in the intracellular domain of CD4 are under purifying selection. We interpret this to demonstrate that mutating these residues that are primarily located in unstructured protein regions negatively affect fitness and are selected against.

Reviewer #3 (Recommendations for the authors):

This paper takes an evolutionary approach to investigate the mechanism(s) by which CD4 regulates TCR activation and downstream functional responses. The authors identify conserved motifs in the extracellular, transmembrane, and intracellular domains of CD4 that appear to regulate multiple aspects of its function, including its localization to lipid RAFTs, its capacity to interact with Lck, and its ability to elicit membrane proximal signaling and promote IL-2 production. Notably, one of these motifs, comprising a helix just N-terminal to the cysteine clasp responsible for interaction with Lck, has an inhibitory effect on TCR signaling, and it seems to have coevolved in eutherians together with a motif in extracellular domain 3 that promotes T cell activation. Collectively, these results suggest that the regulation of CD4 activity has been finely tuned during evolution such that the acquisition of activity-promoting mutations is balanced by the emergence of inhibitory regions.

The revisions and new data have improved this manuscript, although I still have some concerns.

1) The manuscript remains a difficult read. I'm not sure how best to deal with this. The acronyms get a bit overwhelming. Perhaps it would be better to remove some of the less used acronyms (e.g. MRCA) and just write the words out in full?

We understand and appreciate both the feedback and suggestion. We have written out the acronyms in the manuscript.

2) The new signaling data (Figure 6) are a welcome addition to the manuscript, but they are rather sparse. In particular, I am worried that the authors may be missing important phenotypes by overstimulating the TCR expressing cells. A dose response should be performed for each signaling readout. It's possible, for instance, that at a lower antigen dose they would observe more substantial differences in CD3z phosphorylation in Figure 6A. This would be consistent with the idea that CD4-Lck interaction matters more when antigenic peptide is limiting.

We appreciate the comment and agree that our Clasp mutation might impair CD3z phosphorylation at low levels of agonist MCC, particularly at the levels where CD4 has been shown to be critical for calcium mobilization as a readout of proximal signaling events (<20). However, formally testing this in a convincing manner would be very challenging and would take a good deal of time. We have therefore addressed the concern as follows:

Please know that we considered the potential pitfall of overstimulation, and other technical issues, during internal deliberations about how best to study the influence of CD4 on proximal signaling events. After several pilot experiments, we settled on an approach that we consider to be robust, well-controlled, and interpretable within the parameters tested.

Key considerations in designing our experiments were that we wanted to: 1) synchronize TCR engagement on our WT and mutant cell lines; 2) use agonist pMHCII to engage TCR-CD3 and CD4 with their physiological ligands to avoid non-physiological signaling; 3) generate detectable levels of signal to be able to confidently distinguish signal from noise for data interpretation; and, 4) distinguish between differences in the frequency of responding cells and differences in the intensity of responses among responding cells, which is something that cannot be achieved with bulk cell signaling assays such as Western blots.

Antibody crosslinking (e.g. anti-CD3) is typically used in T cell signaling studies to activate cells. This approach provides reasonable control over the timing of signaling initiation, such that many cells within a sample can be activated with relative synchrony due to the high levels of antibody used to ensure robust signal. We were concerned that using anti-CD3/anti-CD4 crosslinking to synchronize activation of our sample populations would stimulate our cells in a non-physiological, and potentially overstimulating manner due to many factors including drastically slowed kinetics of engagement relative to TCR-CD3 and CD4 engagement of pMHCII on APCs (Glassman et al., 2018). We therefore reasoned that, with such an approach, any differences we might observe between the WT and mutant cells may not faithfully mirror any differences that result from engagement of agonist pMHCII, making it hard to confidently interpret the results. Using APCs expressing high densities of agonist pMHCII ensures that when the T cells and APCs are spun together and shifted to 37ºC to initiate signaling, that TCRs and CD4 should rapidly find their ligands to initiate signaling in sufficient numbers to allow for detectable signals with current approaches.

We are concerned that experiments using APCs with low doses of agonist pMHCII would be challenging for the following reasons:

Author details

1. Mark S Lee

   Department of Immunobiology, The University of Arizona College of Medicine, Tucson, United States

   Contribution

   Resources, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

2. Peter J Tuohy

   Department of Immunobiology, The University of Arizona College of Medicine, Tucson, United States

   Contribution

   Data curation, Formal analysis, Validation, Investigation, Writing – review and editing

   Competing interests

   No competing interests declared

3. Caleb Y Kim

   Department of Immunobiology, The University of Arizona College of Medicine, Tucson, United States

   Contribution

   Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

4. Katrina Lichauco

   Department of Immunobiology, The University of Arizona College of Medicine, Tucson, United States

   Contribution

   Investigation, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-9480-2893

5. Heather L Parrish

   Department of Immunobiology, The University of Arizona College of Medicine, Tucson, United States

   Contribution

   Conceptualization, Resources, Validation, Investigation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

6. Koenraad Van Doorslaer

   1. Department of Immunobiology, The University of Arizona College of Medicine, Tucson, United States
   2. School of Animal and Comparative Biomedical Sciences, University of Arizona, Tucson, United States
   3. Cancer Biology Graduate Interdisciplinary Program and Genetics Graduate Interdisciplinary Program, The University of Arizona, Tucson, United States
   4. The BIO-5 Institute, The University of Arizona, Tucson, United States
   5. The University of Arizona Cancer Center, Tucson, United States

   Contribution

   Conceptualization, Resources, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing – original draft

   For correspondence

   vandoorslaer@arizona.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2985-0733

7. Michael S Kuhns

   1. Department of Immunobiology, The University of Arizona College of Medicine, Tucson, United States
   2. Cancer Biology Graduate Interdisciplinary Program and Genetics Graduate Interdisciplinary Program, The University of Arizona, Tucson, United States
   3. The BIO-5 Institute, The University of Arizona, Tucson, United States
   4. The University of Arizona Cancer Center, Tucson, United States
   5. The Arizona Center on Aging, The University of Arizona College of Medicine, Tucson, United States

   Contribution

   Conceptualization, Resources, Data curation, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing – original draft, Project administration, Writing – review and editing

   For correspondence

   mkuhns@email.arizona.edu

   Competing interests

   has disclosed an outside interest in Module Therapeutics to the University of Arizona. Conflicts of interest resulting from this interest are being managed by the University of Arizona in accordance with their policies

   "This ORCID iD identifies the author of this article:" 0000-0002-0403-6313

• 1,075

  Page views

• 259

  Downloads

• 3

  Citations

Article citation count generated by polling the highest count across the following sources: PubMed Central, Crossref, Scopus.