The CD4 and CD3 Εδ Cytosolic Juxtamembrane Regions Are Proximal Within a Compact TCR−CD3−Pmhc−CD4 Macrocomplex This Information Is Current As of September 29, 2021

The CD4 and CD3 εδ Cytosolic Juxtamembrane Regions Are Proximal within a Compact TCR−CD3−pMHC−CD4 Macrocomplex This information is current as of September 29, 2021. Caleb R. Glassman, Heather L. Parrish, Neha R. Deshpande and Michael S. Kuhns J Immunol 2016; 196:4713-4722; Prepublished online 2 May 2016; doi: 10.4049/jimmunol.1502110 Downloaded from http://www.jimmunol.org/content/196/11/4713

Supplementary http://www.jimmunol.org/content/suppl/2016/04/30/jimmunol.150211 http://www.jimmunol.org/ Material 0.DCSupplemental References This article cites 55 articles, 19 of which you can access for free at: http://www.jimmunol.org/content/196/11/4713.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 29, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2016 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

The CD4 and CD3d« Cytosolic Juxtamembrane Regions Are Proximal within a Compact TCR–CD3–pMHC–CD4 Macrocomplex

Caleb R. Glassman,* Heather L. Parrish,* Neha R. Deshpande,*,† and Michael S. Kuhns*,†,‡

TCRs relay information about peptides embedded within MHC molecules (pMHC) to the ITAMs of the associated CD3g«, CD3d«, and CD3zz signaling modules. CD4 then recruits the Src kinase p56Lck (Lck) to the TCR–CD3 complex to phosphorylate the ITAMs, initiate intracellular signaling, and drive CD4+ T cell fate decisions. Whereas the six ITAMs of CD3zz are key determi- nants of T cell development, activation, and the execution of effector functions, multiple models predict that CD4 recruits Lck proximal to the four ITAMs of the CD3 heterodimers. We tested these models by placing FRET probes at the cytosolic juxtamembrane regions of CD4 and the CD3 subunits to evaluate their relationship upon pMHC engagement in mouse cell lines. Downloaded from The data are consistent with a compact assembly in which CD4 is proximal to CD3d«, CD3zz resides behind the TCR, and CD3g« is offset from CD3d«. These results advance our understanding of the architecture of the TCR–CD3–pMHC–CD4 macrocomplex and point to regions of high CD4–Lck + ITAM concentrations therein. The findings thus have implications for TCR signaling, as phosphorylation of the CD3 ITAMs by CD4-associated Lck is important for CD4+ T cell fate decisions. The Journal of Immunology, 2016, 196: 4713–4722. http://www.jimmunol.org/ ndividual peptides embedded within class II MHC molecules ITAM phosphorylation determine cell fate decisions, with fewer (pMHC) create composite surfaces that encode essential in- than seven functional ITAMs per complex leading to a breakdown formation for T cell–mediated immunity. Thymocytes and in central tolerance (8, 9). Thus, assembly of a TCR–CD3–pMHC– I + + CD4 T cells survey the libraries of pMHC expressed on thymic CD4 macrocomplex is key to CD4 T cell development and effector epithelial cells and APCs, respectively, with clonotypic TCRs functions. that relay mechanical information about TCR–pMHC interactions The sensitivity and specificity of this macromolecular machin- across the membrane to the ITAMs of the associated CD3 sig- ery have inspired considerable interest in how it works. Although naling modules (CD3dε, CD3gε, and CD3zz) (1–5). This infor- TCR–CD3 cross-linking with mAbs can activate T cells, not all mation is then converted to chemical signals when CD4, which mAbs can induce this outcome (10); furthermore, not all TCR–pMHC by guest on September 29, 2021 also binds class II MHC, recruits the Src kinase p56Lck (Lck) to interactions of seemingly sufficient affinity activate T cells if the TCR–CD3 complex to phosphorylate the ITAMs (6, 7). In this they dock in a noncanonical topology (11). Thus, an ordered way, pMHC engagement by both TCR and CD4 acts as a coin- relationship of signaling molecules appears to be important for cidence detector to drive the generation, maintenance, and func- T cell activation, particularly when receptors are confronted tion of a pMHC-specific TCR repertoire. There are 10 ITAMs with their natural ligands (1, 2, 12, 13). within a TCR–CD3 complex, with CD3d,CD3g, and CD3ε having In support of this idea, two crystallography studies indicate that one each and CD3z having three. The quantity and quality of aV-likearchisformedwhenCD4andtheTCRsimultaneously bind a pMHC. One structure included the D1 and D2 domains of CD4 in complex with MHC but without a TCR, whereas the other *Department of Immunobiology, The University of Arizona College of Medicine, Tucson, AZ 85724; †The Arizona Center on Aging, The University of Arizona constituted a ternary complex of a TCR, pMHC, and the four College of Medicine, Tucson, AZ 85724; and ‡The BIO5 Institute, The University extracellular domains of a CD4 molecule that was affinity mat- of Arizona College of Medicine, Tucson, AZ 85724 urated for binding to MHC (14, 15). Both structures suggest that Received for publication September 28, 2015. Accepted for publication March 29, the TCR–pMHC axis constitutes one arm of the arch, whereas 2016. CD4 constitutes the other and the CD4–MHC binding site forms This work was supported by The University of Arizona College of Medicine (M.S.K.), the BIO5 Institute (M.S.K.), National Institutes of Health/National Institute of Allergy the apex. and Infectious Diseases Grant R01AI101053 (M.S.K.), and Cancer Center Support Because TCR–pMHC interactions occur via a canonical dock- Grant CCSG-CA 023074 for flow cytometry. M.S.K is a Pew Scholar in the Biomedical ing orientation, the implication is that a geometrically constrained Sciences, supported by The Pew Charitable Trusts. TCR–CD3–CD4–pMHC architecture is functionally mandated to Address correspondence and reprint requests to Dr. Michael S. Kuhns, The University of Arizona College of Medicine, Medical Research Building, Room 224, 1656 East Mabel achieve a specific relationship between Lck and the ITAMs for Street, P.O. Box 245221, Tucson, AZ 85724. E-mail address: [email protected] proper signaling. Mutagenesis and nuclear magnetic resonance The online version of this article contains supplemental material. data place the ectodomains of CD3dε and CD3gε on one side Abbreviations used in this article: CD28T, C-terminally truncated CD28; CD4T, C-terminally of the TCR, with CD3dε and CD3gε positioned inside the arch truncated version of CD4; CD3dTG,CD3dT fused to mEGFP; CD4TmCh,CD4Tfusedatthe formed with CD4, whereas electron microscopy data indicate that JM region to mCherry; FRET, Fo¨rster resonance energy transfer; FRETE, FRET efficiency value; JM, juxtamembrane; Lck, Src kinase p56Lck; MCC, moth cytochrome c peptide; these heterodimers are nested near the base of the TCR (1, 16–20). mEGFP, monomeric enhanced GFP; PD-1T, C-terminally truncated PD-1; pMHC, pep- Owing to the turn of the TCRa transmembrane helix and ∼140˚ tide embedded within class II MHC molecule; ROI, region of interest; TIRFM, total offset of the positively charged residues that interact with CD3zz internal reflection fluorescence microscopy. and CD3dε, CD3zz should reside on the opposite side of the Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 TCRa subunit from CD3dε (21). In addition, because the CD3 www.jimmunol.org/cgi/doi/10.4049/jimmunol.1502110 4714 ARCHITECTURE OF THE TCR–CD3–pMHC–CD4 MACROCOMPLEX heterodimers have short, rigid connecting peptides and the existing single CD3–mEGFP species. For cell lines without a CD3z–mEGFP fu- structural data for activating immune receptor transmembrane sion, CD3z was fused to the biotin acceptor peptide, AP-3. This is a short domains indicate a minimal crossing angle, the CD3 transmem- tag that was considered irrelevant for these studies but is connected by the same linker as mEGFP. M12 control lines consisted of C-terminally brane domains are likely to emerge into the cytoplasm in a spatial truncated CD28 [(CD28T) aa: 1–179] or PD-1 [(PD-1T) aa: 1–199] fused orientation that roughly mimics their ectodomain (22, 23). Con- to mEGFP and mCherry via an AAAG linker and were cotransduced with sequently, Lck is predicted to be closest to the four ITAMs of 2.5 equivalents of full-length CD3 subunits lacking CD3z to account for viral load. the CD3 heterodimers upon pMHC engagement (1, 2, 12, 16). A 2 2 To generate CD3εTG::CD4TmCh or CD3zTG::CD4TmCh 58a b cell priori, such a model is surprising because the six ITAMs within lines for FRET comparisons, parental cells were transduced with vectors the CD3zz signaling module (three per subunit) clearly play an encoding CD3εTG-T2A-5c.c7a and CD3zT-T2A-5c.c7b, or CD3εT-T2A- important role in T cell development and activation (6, 8, 9). 5c.c7a and CD3zTG-T2A-5c.c7b along with a CD4 construct and a construct encoding all full-length versions of the CD3 subunits (8, 16). To generate 58a2b2 Testing this model is thus necessary to gain better insights into the G mCh architecture and function of the TCR–CD3–pMHC–CD4 macro- cell lines comparing CD3dT fused to mEGFP (CD3dT )::CD4T and CD3gTG::CD4TmCh FRET, cells were transduced with individual constructs complex. encoding 5c.c7a, 5c.c7b,CD3xT-mEGFP,CD4TmCherry,andallfull- In this study, we used Fo¨rster resonance energy transfer length CD3 subunits. 2 2 (FRET) to experimentally assess the spatial relationship between The 58a b control cell lines were transduced with individual con- the cytoplasmic juxtamembrane (JM) regions of CD4 and the structs encoding 5c.c7a, 5c.c7b, PD-1T (aa: 1–199) fused to mEGFP and ε mCherry via an AAAG linker, and all full-length CD3 subunits. Further CD3g,CD3d,CD3,andCD3z subunits upon concurrent CD4 information is available upon request. and TCR engagement of agonist pMHC. Replacing the intracellular domains with FRET probes ensured that any proximity Soluble proteins for bilayers and immobile surfaces Downloaded from measurements were dependent upon binding of CD4 and the Production of soluble pMHC and ICAM-1 was performed with a baculovirus TCR to pMHC, but independent of signaling or postsignaling expression as described elsewhere (3, 26). interactions that may tether CD4 to the TCR–CD3 complex (7). Such a system allowed us to make predictions about the native Lipid bilayers spatial relationships between kinase and substrate in the context Bilayers were prepared with a lipid mixture consisting of 97.5 mol % POPC,

of a living cell. The data indicate that CD4 adopts an ordered 1 mol % DGS NiNTA, 1 mol % biotin-CAP PE, and 0.5 mol % DOPE- http://www.jimmunol.org/ spatial arrangement with respect to the TCR–CD3 complex only PEG5000 and were extruded to generate unilaminar vesicles (Avanti Polar Lipids) (26, 38). Liposomes were injected onto a cleaned glass when both the TCR and CD4 bind an agonist pMHC. Therein, coverslip, and bilayer mobility was assessed by photoablation recovery of the JM region of CD4 is positioned closer to that of CD3d than streptavidin conjugated to APC, as previously described (26, 38). For TCR CD3g and closer to the CD3ε subunits than the CD3zz subunits. engagement experiments, each well received 0.05 mgMCC:I-EK and 0.08 mg 2 The implications of these data for the architecture of the TCR–CD3– ICAM-1 to produce an agonist pMHC density of ∼60 mol/mm (17, 26, 38). pMHC–CD4 macrocomplex, and its function in early signaling Peptide–MHC surfaces events, are discussed. Biotinylated poly-L-lysine–coated coverslips were incubated with 5 mg/ml

streptavidin, washed, and incubated with 400 ml PBS + 2% BSA + 0.01% by guest on September 29, 2021 Materials and Methods sodium azide containing a total pMHC–biotin concentration of 5 mg/ml (Hb:I-Ek and MCC:I-Ek) and 0.5 mg/ml biotinylated anti-H2-Dd (3, 39). Cell lines and constructs M12 and 58a2b2 T cell hybridoma lines were generated by retroviral Microscopy transduction and drug selection, as previously described (3, 24–26). The Total internal reflection fluorescence microscopy (TIRFM) was performed 5c.c7 TCR, which is specific for the moth cytochrome c peptide (MCC) at 37˚C, 5% CO , and 50% relative humidity. Cells were adhered to the (88–103) in I-EK (27, 28), was used in this study in complex with 2 ε glass coverslip, lipid bilayers, or immobile surface for 20 min and then C-terminally truncated CD3 (aa: 1–139), CD3d (aa: 1–132), CD3g (aa: imaged for 20–30 min following adhesion. TIRF images were acquired 1–143), and CD3z (aa: 1–57) subunits (CD3xT) or full-length CD3 using a Marianas workstation built on a Zeiss Axio Observer Z1 [Intelli- subunits where indicated in the text (3, 17). When a CD3T subunit was gent Imaging Innovations (3I)] with a Zeiss fluorescent microscope using expressed as a fusion with monomeric enhanced GFP (mEGFP), we a 363 Zeiss TIRF objective coupled to a Zeiss motorized TIRF slider used a GGGSAAAG linker. C-terminally truncated versions of CD4 (numerical aperture 1.46). TIRFM was performed with a Laser Stack (3I) (CD4T aa: 1–421), with or without the Dbind mutation (aa: 68–73 containing 50-mW 488-nm and 561-nm solid-state lasers set at 20% power KGVLIR to DGDSDS), fused to mCherry via an AAAG linker were output. Photoablation of mCherry was performed with a Vector high-speed used in this study (26, 29). Glycine-based linkers were used for our point scanner (3I) at 100% 561-nm laser output within a 6.45-mm2 region fusion proteins because flexibility is required to allow equivalent of interest (ROI). Images were collected at 500-ms intervals [Photometrics probability of dipole alignment between FRET pairs, and even a single Evolve EMCCD; 1 pixel = 0.25 mm (H) 3 0.25 mm (V) at 363]. glycine in a dipeptide linker can allow a strand to double back upon itself (30–34). The AAA segment of each linker represents the reading frame of Image analysis a NotI restriction enzyme cut site used for cloning. This was chosen because Ala has been used with Gly in flexible linkers (35). Furthermore, Median mEGFP intensity and mCherry intensity for the ROI targeted poly-Ala peptides have been reported to assume similar backbone con- for mCherry ablation were background subtracted using SlideBook6 (3I) formation diversity to that of poly-Gly (36), provided the AAA stretch is and exported. Data were processed in MATLAB (MathWorks), and FRET not flanked by Glu and Lys residues, which it was not in our study (37). was calculated as FRET efficiency value (FRETE)=12 (Q/DQ), where Cell lines were assessed for TCR–CD3 and CD4 surface expression by Q (quenched) is the mEGFP intensity prior to mCherry ablation and flow cytometry. DQ (dequenched) is the mEGFP intensity following ablation, as previously Murine stem cell virus–based retroviral expression vectors pP2 (IRES- reported (3, 26). Efficiency of mCherry ablation was calculated as the ratio puromycin resistance) and pZ4 (IRES-zeocin resistance) were used for the of postbleach to prebleach mCherry intensity, Abl = mCh (postbleach)/ generation of mouse cell lines (16). We took advantage of the 2A cleavage mCh (prebleach). Subsets designated All cells in the figures had system to generate polycistronic constructs that reduced the number of mCherry ablation below 12.5% prebleach intensity, as described previ- constructs needed per cell line (8). ously (3, 26). This resulted in the analysis of populations with an average Experimental M12 FRET lines were generated using constructs encoding ablation below 10% prebleach intensity (not shown). Subsets designated CD3εT(G)-T2A-5c.c7a,CD3zT(G)-T2A-5c.c7b,andCD3dT(G)-T2A-CD3gT(G) Matched subset included events with ablation below 12.5% of prebleach along with a construct encoding CD4T fused at the JM region to mCherry mCherry intensity in which analyzed populations were matched for (CD4TmCh) or CD4TDbind.mCh. The CD3xT(G) indicates that constructs mCherry intensity prior to photobleaching and GFP intensity following were built encoding only the truncated CD3 subunit, or a fusion of a photobleaching to account for differential donor quenching in experi- truncated CD3 subunit to mEGFP (G), to generate cell lines expressing a mental conditions as well as GFP/mCherry ratio. The Journal of Immunology 4715

Flow-based fluorophore-linked immunosorbent assay a CD4 molecule in a distinct membrane domain appears to be G physically limited to those that are located at the boundary of each DDM (1% n-dodecyl-b-D-maltoside) lysates from CD3xT cell lines were subject to immunoprecipitation with 6.0-mm polystyrene beads membrane domain (42). In contrast, disulfide-bonded CD28 homo- (Polysciences) coated with anti-TCRb mAbs (H57-597; BioLegend). After dimers will speciate into CD28TG::CD28TmCh,CD28TG::CD28TG, washing, the beads were then probed with anti-TCRa (a-Va11 RR8-1 and CD28TmCh::CD28TmCh pairs within the same membrane do- ε allophycocyanin; eBioscience) and anti-CD3 (145-2C11 PE; BD) with main based on the relative expression of FRET constructs. These fluorescently labeled mAbs. Analysis was then performed by flow cytometry to assess coprecipitation of TCRa,CD3ε, and CD3xTG (mEGFP) with distinctions are important because FRETE measurements are TCRb, similarly to previously described protocols (29, 40, 41). based on changes in the fluorescent intensity of all FRET do- nors within an ROI after acceptor ablation of all donor mole- Statistical analysis cules, whether or not they are paired with an acceptor. Owing to All statistical analyses were performed with Prism 6 (GraphPad Software). the complications outlined above, and differences in expression As the data presented in this article are nonparametric, the Mann–Whitney profiles, comparisons between the CD3dTG::CD4TmCh FRET U test and Kruskal–Wallis test with Dunn’s multiple comparisons were E used where appropriate. and CD28 provide limited quantitative information. Given the challenges associated with comparing FRET between distinct molecular species, we also approached the problem by Results asking if mutating the D1 domain region of CD4, which binds d Proximal association of CD4 and CD3 upon TCR and CD4 class II MHC in crystal structures (14, 15), would impair the engagement of agonist pMHC G mCh CD3dT ::CD4T FRETE signal observed on pMHC-coated

The existing experimental data suggest that the formation of a coverslips. To this end, we used a CD4TDbind mutant that im- Downloaded from TCR–CD3–pMHC–CD4 macrocomplex positions Lck closest to pairs T cell hybridoma activation (26, 29). In this situation, G Dbind.mCh the ITAMs of CD3dε (1). Consequently, we transduced B cell lym- CD3dT ::CD4T FRETE was significantly lower than G mCh phoma M12 cells, which do not express endogenous TCR–CD3 CD3dT ::CD4T FRETE at the bulk population level (Fig. 1C). subunits, to express the 5c.c7 TCR, C-terminally truncated CD3g, One caveat to this approach is that CD4–MHC interactions influ- CD3ε, and CD3z subunits (CD3gT, CD3εT, and CD3zT) that enced accumulation of CD4 at the contact interface (Fig. 1D–F). G lack ITAMs and cannot signal (3, 24), and CD3dT as the FRET However, this difference in donor and acceptor concentrations is http://www.jimmunol.org/ donor. We also expressed CD4T fused to mCherry as a FRET unlikely to be responsible for the difference in FRET efficiency, as acceptor. FRET was measured by donor recovery after acceptor intensity- and ratio-matched subsets showed a similar reduction in G Dbind.mCh photobleaching via TIRFM to determine a relative FRETE for FRETE for the CD3dT ::CD4T compared with the FRETE molecules on the cell membrane, as previously described (3, 26). measured in CD3dTG::CD4TmCh cells (Fig. 1D–G). No difference This experimental approach allowed us to reduce the question to was observed between the two cell lines on glass coverslips (Fig. 1H), one of spatial proximity driven purely by pMHC engagement via again demonstrating pMHC dependence for the observed FRET. the TCR and CD4 in the absence of competition with endogenous Altogether, these data strongly suggest that the CD3dTG::CD4TmCh subunits, signaling, and signaling feedback. FRET results from concurrent TCR and CD4 engagement.

G mCh k by guest on September 29, 2021 CD3dT ::CD4T FRETE was first measured for M12 cells We next used mobile lipid bilayers presenting MCC:I-E adhered to glass coverslips to establish a baseline signal in the and ICAM-1 to assess whether these differences in FRET absence of pMHC engagement. Coverslips coated with immobile would be observed between the CD3dTG::CD4TDbind.mCh and G mCh agonist pMHC were then used to determine if concurrent TCR CD3dT ::CD4T FRETE cells under more physiological con- and CD4 engagement of agonist pMHC increased FRET be- ditions (Fig. 2A). Again, we saw a decrease in FRET efficiency tween CD3d and CD4 (Fig. 1A) (26). Because we have previously with the Dbind mutation, consistent with the results obtained on established that disulfide-bonded CD28 homodimers and PD-1 immobile surfaces (Fig. 2B). monomers serve as positive and negative controls, respectively, Taken together, the data suggest that the CD4 and CD3d JM G mCh for FRET in M12 cells, we analyzed CD3dT ::CD4T FRETE regions assume a close spatial proximity that depends upon the D1 along with these controls (3, 26). No difference in FRETE was domain of CD4 binding class II MHC, but does not require sig- observed between the PD-1TG::PD-1TmCh negative control and naling. However, because TCR–CD3 complexes and CD4 accumu- our experimental CD3dTG::CD4TmCh cells imaged on glass cov- late at the binding interface, these data alone do not demonstrate an erslips. This finding is consistent with data suggesting that CD4 ordered spatial relationship. The observed CDdTG::CD4TmCh and TCR do not preassociate in an unengaged state (42). Adherence of FRET could be explained by random FRET interactions oc- CD3dTG::CD4TmCh cells to glass coverslips functionalized with ago- curring in trans owing to increased molecular concentrations nist pMHC resulted in a significant increase in CD3dTG::CD4TmCh (i.e., clustering) rather than a defined orientation enforced by FRETE, compared with cells on uncoated coverslips or the negative specific interactions. control cells, indicating that concurrent TCR and CD4 engagement of agonist pMHC positions CD3d and CD4 in a close spatial relationship Coordinate TCR–pMHC and CD4–MHC interactions create an (Fig. 1B). ordered CD4–CD3 spatial relationship G mCh G mCh Although the CD3dT ::CD4T FRETE signal was lower than If CD3dT ::CD4T FRET results from random clustering, then that of the positive control cells, it is unclear if this is due to at any given point CD4 should, on average, be equidistant to all differences in distance between FRET donor and acceptor, the CD3 subunits. Alternatively, if FRET is occurring as a conse- frequency of productive FRET pairs, or both. In the case of quence of the formation of an ordered structure, then differences CD3d::CD4 FRET, coordinate binding of CD4 and TCR to an in FRETE should be observed between CD4 and the CD3 subunits agonist pMHC creates a transient mEGFP and mCherry pairing based on the relative distances of their defined spatial relation- for a subset of TCR and CD4 molecules that engage pMHC on an ships. To test these possibilities, we generated M12 cells expressing immobile surface. This happens because CD4 and the TCR have CD4TmCh and TCR–CD3 complexes containing CD3dTG,CD3gTG, been shown by superresolution imaging to inhabit distinct mem- CD3εTG,orCD3zTG. Each CD3 subunit was fused to mEGFP via brane domains (42). Consequently, the number of TCRs within a a common flexible linker (GGGSAAAG) that should allow the membrane domain that are available to engage the same pMHC as probe to extend beyond the short intracellular domains of the TCR 4716 ARCHITECTURE OF THE TCR–CD3–pMHC–CD4 MACROCOMPLEX Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 1. FRET between CD4 and CD3d is dependent on concurrent TCR and CD4 engagement of pMHC on immobile surfaces. (A)Repre- sentative TIRF images and intensity traces showing donor recovery after acceptor photobleaching. Experimental M12 cells expressing the 5c.c7 TCR, truncated CD3 subunits and the ﬂuorescently tagged proteins are shown adhered to glass coverslips (left: unengaged) or coverslips functionalized with MCC:I-Ek (right: TCR + CD4 engagement), as labeled and described in the text. The white box represents the region targeted for photoablation and analysis. Scale bar, 5 mm. (B) pMHC engagement increases FRETE between CD4 and CD3d at a population level. FRETE measurements of experimental M12 cells were compared for cells on glass coverslips (unengaged) or MCC:I-Ek functionalized coverslips, as labeled and described in the text.

****p , 0.0001, Kruskal–Wallis with Dunn’s multiple comparisons posttest. (C)FRETE between CD3d and CD4 is dependent on CD4 D1 domain k interactions with MHC. FRETE was measured for the indicated cell lines adhered to MCC:I-E functionalized coverslips, as labeled. ****p , 0.0001, Dbind Mann–Whitney. (D–G) The reduced FRETE seen with CD4T is not due to decreased accumulation of CD4 at the contact interface. Boxed region represents matched subsets from (C)for(D) mEGFP/mCherry ratio, (E) mEGFP intensity, and (F) mCherry intensity. (G)AnalysisofFRETE for matched subsets as shown in (D) through (F) and described in Materials and Methods. ****p , 0.0001, Mann–Whitney. (H) The decreased FRETE observed with CD4TDbind is dependent on MHC engagement (ns p . 0.05, Mann–Whitney). Experiments were performed (Figure legend continues) The Journal of Immunology 4717

FIGURE 2. FRET between CD4 and CD3d is dependent on concurrent TCR and CD4 engagement of pMHC on lipid bilayers. (A) Representative images showing donor recovery after acceptor photobleaching of experimental M12 cells adhered to supported lipid bilayers presenting MCC:I-Ek and ICAM-1 Downloaded from (LBL), as labeled and described in the text. Images are as described in Fig. 1. (B) CD4 D1 domain interactions with MHC are critical for observed FRET. Population analysis of matched subsets, as described in Materials and Methods. Circles represent individual cells, N = number of cells analyzed, and black bars represent median values. Data are representative of at least two experiments. ****p , 0.0001, Mann–Whitney.

(∼5–9 aa) and other truncated CD3 subunits (∼5 aa) and allow for (Supplemental Fig. 1B). The slightly lower ratio of CD3zTG an equivalent probability of dipole alignment. Because the CD3 compared with CD3εTG is expected owing to the presence of intracellular domains, which do not play a role in complex as- partially assembled complexes, because it is well established that http://www.jimmunol.org/ sembly (3, 16, 17, 21), are largely absent in this system, they CD3z is the last subunit to join the complex and partially as- should not inadvertently have an impact on the rotation or position sembled complexes lacking CD3z can be found in whole-cell of the CD3-associated mEGFP. This design strategy should make lysates (21, 43). These data establish that the mEGFP probes do FRET a function of the spatial position of the donor within the not differentially affect complex assembly. G mCh TCR–CD3 complex relative to the position of a CD4-associated Signiﬁcantly higher FRETE was measured for CD3dT ::CD4T acceptor when both the TCR and the CD4 engage a pMHC. than CD3gTG::CD4TmCh for ratio- and intensity-matched subsets G mCh Consequently, any differences in FRETE between compared lines on bilayers (Fig. 3A). Likewise, CD3εT ::CD4T was signiﬁ- G mCh

should reflect the distance of the particular CD3 JM regions relative cantly higher than CD3zT ::CD4T for ratio- and intensity- by guest on September 29, 2021 to CD4 if an ordered macrocomplex is formed. In our experiments, matched subsets on bilayers (Fig. 3B). These data are best explained CD3dTG and CD3gTG were compared directly because each is by the formation of an ordered macrocomplex upon coordinate G present only once in a single TCR–CD3 complex, and CD3εT TCR and CD4 binding to pMHC. Of note, no difference in FRETE and CD3zTG were compared directly because there are two copies was observed on glass coverslips for the relevant comparisons per complex. (Fig. 3C, 3D). Cell surface expression of TCR–CD3 complexes was confirmed We next asked if FRET between the distal CD3g and CD3z by flow cytometry (not shown) as well as by TIRFM (Fig. 3). subunits was dependent on CD4 engagement of MHC. Analysis Because the TCR trafficks to the cell surface only in association of matched subsets revealed that the FRET efficiencies of with the CD3 subunits, these data strongly suggest that the FRET CD3gTG::CD4TmCh and CD3zTG::CD4TmCh were greater than probes do not negatively impact complex assembly. To further their CD4TDbind.mCh counterparts (Fig. 3E, 3F). These data indi- confirm that fusing mEGFP to the individual CD3 subunits did cate that the observed FRET was dependent on CD4 engagement. not differentially impact complex assembly, we assessed the As a second approach to resolving random versus ordered composition of the TCR–CD3 complexes using a flow-based FRET between CD4 and the TCR–CD3 complex, we measured fluorophore-linked immunosorbent assay. Appropriate complex FRET on immobile surfaces coated with agonist pMHC diluted assembly involves the incorporation of two CD3ε per TCR–CD3 into nonstimulatory pMHC. These surfaces prevent the cells from complex, independent of the mEGFP-tagged CD3 subunit, gathering agonist pMHC into a high local concentration, so at low whereas mEGFP will be present according to the copy number agonist concentrations any measured FRET should be as a con- of the tagged subunit: one for CD3d and CD3g, or two for CD3ε sequence of concurrent TCR–CD3 and CD4 association with an and CD3z. Comparison of anti-CD3ε intensities with mEGFP agonist ligand surrounded by null ligands. Indeed, our streptavidin levels revealed proportional signal (i.e., diagonal) for the CD3dΤG capture system allows for the possibility of pMHC molecules versus CD3gΤG and CD3εΤG versus CD3zΤG cell lines, with being presented in orientations that are not accessible to TCR, CD3εTG and CD3zTG shifted toward brighter mEGFP signal CD4, or both, further reducing the actual availability of agonist (Supplemental Fig. 1A). This observation was confirmed by pMHC for coordinate TCR and CD4 engagement. Because our analysis of beads with matched TCRa intensities (i.e., equivalent cells express the 5c.c7 TCR that recognizes the MCC (88–103) abTCR load). In matched samples, the ratio of mEGFP signal presented in I-Ek (MCC:I-Ek) as an agonist pMHC, we diluted relative to CD3ε staining was similar between the CD3dΤG and MCC:I-Ek into a null pMHC complex comprising the mouse CD3gΤG lines and between the CD3εTG and CD3zTG lines hemoglobin d allele peptide (64–76) presented in I-Ek (HB:I-Ek). and analyzed as described in Materials and Methods. Images were acquired at 500-ms intervals with a 25-ms exposure, 50-camera intensification. Circles represent individual cells, N = number of cells analyzed, and black bars represent median values. Data are representative of at least two experiments. 4718 ARCHITECTURE OF THE TCR–CD3–pMHC–CD4 MACROCOMPLEX Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 3. CD4 adopts an ordered arrangement with respect to the CD3 subunits. (A and B) Agonist pMHC engagement results in differential FRETE between CD4 and the CD3 subunits. Experimental M12 cells were adhered to supported lipid bilayers presenting MCC:I-Ek and ICAM-I (LBL), as labeled and described in the text. (C and D) Differences in FRETE depend on ligand engagement. Experimental M12 cells were adhered to glass coverslips (unengaged). (E and F) FRET between CD4 and (E) CD3g or (F) CD3z subunits is dependent on CD4 D1 domain–MHC interactions. Experimental M12 cells were adhered to supported lipid bilayers presenting MCC:I-Ek and ICAM-I (LBL), as labeled and described in the text. Circles represent individual cells, N = number of cells analyzed, and black bars represent median values. Data are representative of at least two experiments per condition. Subset analysis and subset criteria are as detailed in Materials and Methods. ****p , 0.0001, Mann–Whitney.

This null ligand has previously been shown to lack co-agonist control 58a2b2 (PD-1TG::PD-1TmCh) cells that expressed TCR– activity for the 5c.c7 TCR (44). CD3 complexes and were able to adhere to the bilayers (Fig. 5A). G mCh Analysis of FRETE for a titration of agonist into null pMHC CD3dT ::CD4T FRETE was observed to be higher than the G mCh revealed dose-dependent FRET. CD3dT ::CD4T FRETE was negative control (Fig. 5B). This signal was dependent on CD4 G mCh Dbind observed to be higher than CD3gT ::CD4T FRETE,whereas binding to MHC, as the CD4T mutant reduced this signal G mCh G mCh G mCh CD3εT ::CD4T FRETE was higher than CD3zT ::CD4T (Fig. 5C, 5D). Importantly, CD3dT ::CD4T FRETE was G mCh G mCh FRETE across the titration range (Fig. 4A, 4B), including at higher than CD3gT ::CD4T FRETE, and CD3εT ::CD4T G mCh low ligand densities (Fig. 4C, 4D). Collectively, these data are FRETE was higher than CD3zT ::CD4T FRETE (Fig. 5E, 5F). best explained by nonrandom FRET between CD4 and the The magnitude of these differences is smaller than that observed TCR–CD3 complex upon coordinate TCR and CD4 binding of in the M12 system. This is likely due to the presence of untagged agonist pMHC. subunits or the potential for signaling and signaling feedback. Altogether, these data indicate that the spatial relationships ob- The same spatial relationship occurs in T cell membranes served in M12 cells in the absence of CD3 ITAMs holds in T cell Finally, we measured FRETE between CD4 and each CD3 subunit hybridomas expressing a full complement of ITAMs. in 58a2b2 T cell hybridomas to verify that the spatial proximity of these molecules demonstrated the same hierarchy when mea- Discussion sured in a T cell membrane environment in the presence of full- In this article, we show that CD4 adopts an ordered spatial rela- length, signaling-competent CD3 subunits (26). First, we com- tionship with the TCR–CD3 complex when the TCR and CD4 bind G mCh pared CD3dT ::CD4T FRETE on lipid bilayers with negative to agonist pMHC. The results are important for our understanding The Journal of Immunology 4719 Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 4. FRET between CD4 and CD3 subunits depends on agonist pMHC concentration. Experimental M12 cells were adhered to functionalized k K coverslips presenting agonist (MCC:I-E ) diluted into null (HB:I-E ) pMHC, as labeled. The total amount of pMHC on the surface was constant. FRETE was measured, and analysis was performed as described in Materials and Methods.(A and B) Dose-dependent FRET between CD4 and the CD3 subunits persists over a range of agonist pMHC concentrations. CD3dTG::CD4TmCh and CD3gTG::CD4TmCh did not differ on completely null surfaces [N(d) = 69, N(g) = 40, ns p . 0.05] but did for all other ligand concentrations tested [4% N(d) = 113, N(g) = 101, *p , 0.05; 12.5% N(d) = 96, N(g) = 75, **p , 0.01; 100% N(d) = 86, N(g) = 56, ****p , 0.0001]. Similarly, CD3εTG::CD4TmCh and CD3zTG::CD4TmCh did not differ on completely null surfaces [N(ε) = 44, N(z) = 54, ns p . 0.05] but did for all other ligand concentrations tested [4% N(ε) = 71, N(z) = 79, *p , 0.05; 12.5% N(ε) = 52, N(z) = 86, **p , 0.01; 100% N(ε) = 55, N(z) = 66, ***p , 0.001]. Cells were intensity and ratio matched (matched subset as described in Materials and Methods) at each titration point, and medians were compared using a Mann–Whitney U test. Each data point in the graph represents the mean 6SEM. (C and D) Differences between subunits persist at low ligand densities. Comparisons made on 4% agonist surfaces for matched subsets, as described in Materials and Methods. Circles represent individual cells, N = number of cells analyzed, and black bars represent median values. Data are representative of at least two experiments. *p , 0.05, Mann–Whitney. of the architecture of this macrocomplex and provide a conceptual (23). Within this context, the simplest interpretation of our data is framework for further interrogating its function. that the JM regions of CD3dε are between the JM regions of CD4 Overall, the data shed light on the subunit arrangement within and TCRa, with the CD3zz JM regions sitting behind TCRa. the TCR–CD3–pMHC–CD4 macrocomplex. Speciﬁcally, they Because CD3gε interacts via transmembrane charge interactions indicate that the CD4 JM region resides closer to the JM region of with TCRb, its JM regions would be offset at a distance from CD3d than to CD3g and closer to the JM region of the CD3 CD3dε (21). heterodimers than to CD3zz. TCRa is known to interact with Three distinct arrangements of the TCR–CD3 complex subunits CD3dε and CD3zz via positively charged residues that are offset have been proposed and are considered in this article within the by ∼140˚ in its transmembrane domain (21). This ﬁnding suggests TCR–CD3–pMHC–CD4 macrocomplex. Our FRET data are in- that the CD3zz JM regions are similarly displaced (i.e., ∼140˚) consistent with a model in which CD3gε is positioned in line with relative to the JM regions of CD3dε when associated with TCRa CD4 such that CD3dε is situated adjacent to CD3gε, but offset at a 4720 ARCHITECTURE OF THE TCR–CD3–pMHC–CD4 MACROCOMPLEX Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 5. CD3 subunit–dependent FRET differences are maintained in 58a2b2 T cell hybridomas. (A) Representative TIRF images of donor recovery after acceptor photobleaching for PD-1TG:: PD-1TmCh (left)andCD3dTG::CD4mCh (right)58a2b2 T cell hybridomas adhered to supported lipid bilayers (LBL) presenting MCC:I-Ek and ICAM-1, as described in the text. Images were acquired as described in Fig. 1 and Materials and Methods.(B)CD3dTG::CD4TmCh G mCh FRETE is greater than PD-1T :: PD-1T FRETE on MCC–ICAM lipid bilayers in the presence of full-length CD3 subunits. (C) FRET between CD4T and CD3dT depends on the CD4 D1 domain interacting with class II MHC. (D) The CD4TDbind mutation does not affect FRET efﬁciency in unengaged cells. (E and F)

CD4 adopts a specific orientation with respect to the TCR–CD3 complex in a T cell membrane. FRETE measurements were performed and analyzed as described in Materials and Methods. Circles represent individual cells, n = number of cells analyzed, and black bars represent median values. Data are representative of at least two experiments. *p , 0.05, ****p , 0.0001, Mann–Whitney. greater distance from CD4 (12). Slight adjustments in the place- striking is that, by the general approximations allowed by such ment of the CD3 heterodimers might reconcile these differences modeling, the JM regions of CD3g and either CD3zz subunit are for this model; however, previous proximity analysis of the CD3 .100A˚ away from CD4 (Supplemental Fig. 2D). The same is true JM regions does not support a juxtaposed placement of CD3d and if the CD3 heterodimers are positioned on opposite sides of the TCR CD3g (17). Furthermore, because of the 140˚ offset of the TCRa (Supplemental Fig. 2E). This observation is important because such transmembrane domain charge residues that interact with CD3dε estimated distances between CD4 and CD3g or CD3z place them at and CD3zz, too much adjustment would place CD4 closer to or beyond the outer limits of detection by FRET with these probes CD3zz than either CD3ε. The data presented in this article are (34, 46). That our CD3gTG::CD4TmCh and CD3zTG::CD4TmCh compatible with models in which CD3dε and CD3gε are either FRET are dependent upon the CD4 D1 domain suggests that the situated on opposite sides of the TCR (22, 45) or reside on one V-like arch identified in the crystal structures might be an initial, side of the TCR such that the subunits are ordered d:ε:ε:g (1, 2, but not the final, arrangement of the macrocomplex. 16, 17). This last model is most consistent with other experimental Such a conclusion is consistent with multiple studies that suggest data, as well as recent nuclear magnetic resonance data, so we CD4 makes contacts with class II MHC beyond those observed in currently consider it the best approximation of the subunit orga- the crystal structures, and may even contact the TCR–CD3 com- nization of the TCR–CD3 complex (1, 19). plex. For example, mutagenesis of class II MHC or the D1 and D2 When integrating this model with the ternary arched crystal domains of CD4 suggest more extensive CD4–MHC interactions structure of TCR–pMHC–CD4 (Supplemental Fig. 2), what is most than are revealed by the existing structures (47–49). Furthermore, The Journal of Immunology 4721 domain swap and mutagenesis analysis of the membrane proximal TCR–CD3–pMHC–CD4 macrocomplex in which the cytosolic D3 and D4 domains of CD4 suggest additional contacts with the JM regions of CD4 are most closely associated with the JM re- TCR–CD3 complex (50, 51). gions of the CD3 heterodimers, and in particular the CD3dε In consideration of these data, it has been proposed that CD4 subunits. Although it would be technically challenging to tem- might preassociate with the TCR or dock loosely along a composite porally resolve the consequences of this apposition by ITAM surface formed by the TCR–CD3 complex to form more avid phosphorylation analysis, these data provide a conceptual frame- interactions than the 200 mM–2 mM affinity estimates for CD4 work from which to devise experiments aimed at testing whether binding to MHC in isolation (1, 44). Although preassociation is the CD3dε and CD3gε ITAMs represent the ignition point for not supported by our data, or by superresolution imaging, such signaling under low pMHC density conditions that initiate im- interactions could occur at a low frequency (42). Docking of CD4 munological synapse formation and signal potentiation. along the composite surface created by TCR–CD3–pMHC would be consistent with the data presented in this article and could help Acknowledgments explain the evolutionary advantage of having weak CD4–MHC We thank Mark S. Lee and David I. Duron for thoughtful discussions, interactions in the absence of the TCR–CD3 complex. For ex- critical feedback, and technical assistance; Dominik Schenten for criti- ample, CD4 binding to MHC via the D1 domain could provide cally reading the manuscript; members of the Kuhns, Nikolich-Zugich, sufficient interactions to facilitate preliminary nucleation of CD4 Frelinger, Wu, and Schenten laboratories for critical feedback; Hemant B. and TCR, yet be weak enough to allow for CD4 to pivot toward a Badgandi for contributing novel constructs; and the University of Arizona more avid interaction along the TCR–CD3 composite interface. In Cancer Center/Arizona Research Labs Cytometry Core Facility for flow this way, CD4 binding would be coordinated with TCR affinity, cytometry. Downloaded from consistent with a kinetic proofreading model for TCR signaling (52). It could also help explain CD4’s Lck-independent function Disclosures by providing a structural contribution to TCR–pMHC interactions The authors have no financial conflicts of interest. that facilitates signaling (1, 53). Also of note, interactions between the SH2 domain of Lck and phosphorylated ITAMs would not References contribute to the close association observed in this study because 1. Kuhns, M. S., and H. B. Badgandi. 2012. Piecing together the family portrait of http://www.jimmunol.org/ the clasp domain of CD4 that interacts with Lck was replaced by a TCR-CD3 complexes. Immunol. Rev. 250: 120–143. FRET probe (7, 40). Such interactions, however, could normally 2. Kuhns, M. S., and M. M. Davis. 2012. TCR signaling emerges from the sum of many parts. Front. Immunol. 3: 159. extend a continuous contact interface of a compact macrocomplex 3. Lee, M. S., C. R. Glassman, N. R. Deshpande, H. B. Badgandi, H. L. Parrish, inside the cell that would be relevant for a recently proposed Lck C. Uttamapinant, P. S. Stawski, A. Y. Ting, and M. S. Kuhns. 2015. A me- come and stay/signal duration model (40). Additional work is chanical switch couples T cell receptor triggering to the cytoplasmic juxtamembrane regions of CD3zz. Immunity 43: 227–239. needed on this long-standing problem, but the data presented in 4. Minguet, S., M. Swamy, B. Alarcoń, I. F. Luescher, and W. W. Schamel. 2007. this study are suggestive of a compact TCR–CD3–pMHC–CD4 Full activation of the T cell receptor requires both clustering and conformational changes at CD3. Immunity 26: 43–54. macrocomplex that is consistent with the existing mutagenesis data. 5. Gil, D., W. W. Schamel, M. Montoya, F. Sańchez-Madrid, and B. Alarcoń. 2002. Regardless of how CD4 and the TCR–CD3 complex achieve a Recruitment of Nck by CD3 epsilon reveals a ligand-induced conformational by guest on September 29, 2021 close spatial relationship upon binding pMHC, the data presented change essential for T cell receptor signaling and synapse formation. Cell 109: 901–912. in this study provide evidence for models in which CD4 brings 6. Guy, C. S., and D. A. Vignali. 2009. Organization of proximal signal initiation at Lck in closer proximity to the CD3 heterodimers than CD3zz. the TCR:CD3 complex. Immunol. Rev. 232: 7–21. Indeed, the data indicate that the cytosolic JM regions of 7. Chakraborty, A. K., and A. Weiss. 2014. Insights into the initiation of TCR signaling. Nat. Immunol. 15: 798–807. CD4 most closely approach the CD3dε ITAMs, suggesting that 8. Holst, J., H. Wang, K. D. Eder, C. J. Workman, K. L. Boyd, Z. Baquet, H. Singh, CD4-associated Lck and these ITAMs are likely to reside in the K. Forbes, A. Chruscinski, R. Smeyne, et al. 2008. Scalable signaling mediated by T cell antigen receptor-CD3 ITAMs ensures effective negative selection and highest local concentrations upon TCR and CD4 binding of ago- prevents autoimmunity. Nat. Immunol. 9: 658–666. nist pMHC. Given that experimental evidence places the cytosolic 9. Guy, C. S., K. M. Vignali, J. Temirov, M. L. Bettini, A. E. Overacre, JM regions of the two CD3ε subunits within ∼40 A˚ upon assembly M. Smeltzer, H. Zhang, J. B. Huppa, Y. H. Tsai, C. Lobry, et al. 2013. Distinct TCR signaling pathways drive proliferation and cytokine production in T cells. into the TCR–CD3 complex, and unobstructed from each other Nat. Immunol. 14: 262–270. (17), the data suggest that CD4 will also bring Lck proximal to the 10. Yoon, S. T., U. Dianzani, K. Bottomly, and C. A. Janeway, Jr. 1994. Both high ITAMs of the CD3gε heterodimer, particularly if the kinase do- and low avidity antibodies to the T cell receptor can have agonist or antagonist activity. Immunity 1: 563–569. main can reach beyond its immediate locale when extended in an 11. Adams, J. J., S. Narayanan, B. Liu, M. E. Birnbaum, A. C. Kruse, open and active conformation. How Lck could access the CD3zz N. A. Bowerman, W. Chen, A. M. Levin, J. M. Connolly, C. Zhu, et al. 2011. ITAMs is less obvious, particularly within the context of the V-like T cell receptor signaling is limited by docking geometry to peptide-major his- tocompatibility complex. Immunity 35: 681–693. arch stabilized by ∼200 mM–2 mM interactions. However, if CD4 12. Rangarajan, S., and R. A. Mariuzza. 2014. T cell receptor bias for MHC: co- does dock along a composite TCR–CD3–pMHC surface, then evolution or co-receptors? Cell. Mol. Life Sci. 71: 3059–3068. 13. Garcia, K. C. 2012. Reconciling views on T cell receptor germline bias for interactions between CD4-associated Lck and the CD3zz ITAMs MHC. Trends Immunol. 33: 429–436. are more conceivable, given that each CD3z subunit has long 14. Wang, J. H., R. Meijers, Y. Xiong, J. H. Liu, T. Sakihama, R. Zhang, intracellular domains with three ITAMs that might somehow reach A. Joachimiak, and E. L. Reinherz. 2001. Crystal structure of the human CD4 N-terminal two-domain fragment complexed to a class II MHC molecule. Proc. toward Lck. Such a docking model would also be consistent with Natl. Acad. Sci. USA 98: 10799–10804. how CD8 has been proposed to function (54, 55). 15. Yin, Y., X. X. Wang, and R. A. Mariuzza. 2012. Crystal structure of a complete In closing, the canonical docking of TCR on pMHC is thought to ternary complex of T-cell receptor, peptide-MHC, and CD4. Proc. Natl. Acad. Sci. USA 109: 5405–5410. be functionally mandated to achieve a specific relationship between 16. Kuhns, M. S., and M. M. Davis. 2007. Disruption of extracellular interactions Lck and the ITAMs for positive selection and T cell activation (1, 2, impairs T cell receptor-CD3 complex stability and signaling. Immunity 26: 357–369. 12, 13). A variety of data have predicted that CD4-associated Lck 17. Kuhns, M. S., A. T. Girvin, L. O. Klein, R. Chen, K. D. Jensen, E. W. Newell, and the ITAMs of the CD3 heterodimers are most closely asso- J. B. Huppa, B. F. Lillemeier, M. Huse, Y. H. Chien, et al. 2010. Evidence for a ciated upon coordinate CD4 and TCR binding to pMHC, but this functional sidedness to the alphabetaTCR. Proc. Natl. Acad. Sci. USA 107: 5094–5099. prediction has not been experimentally tested (16–19). The data 18. Fernandes, R. A., D. A. Shore, M. T. Vuong, C. Yu, X. Zhu, S. Pereira-Lopes, in this article provide evidence for a highly ordered, compact H. Brouwer, J. A. Fennelly, C. M. Jessup, E. J. Evans, et al. 2012. T cell receptors 4722 ARCHITECTURE OF THE TCR–CD3–pMHC–CD4 MACROCOMPLEX

are structures capable of initiating signaling in the absence of large conformational 37. Amet, N., H. F. Lee, and W. C. Shen. 2009. Insertion of the designed helical rearrangements. J. Biol. Chem. 287: 13324–13335. linker led to increased expression of tf-based fusion proteins. Pharm. Res. 26: 19. He, Y., S. Rangarajan, M. Kerzic, M. Luo, Y. Chen, Q. Wang, Y. Yin, 523–528. C. J. Workman, K. M. Vignali, D. A. Vignali, et al. 2015. Identification of the 38. Beemiller, P., J. Jacobelli, and M. F. Krummel. 2012. Integration of the move- docking site for CD3 on the T cell receptor b chain by solution NMR. J. Biol. ment of signaling microclusters with cellular motility in immunological syn- Chem. 290: 19796–19805. apses. Nat. Immunol. 13: 787–795. 20. Birnbaum, M. E., R. Berry, Y. S. Hsiao, Z. Chen, M. A. Shingu-Vazquez, X. Yu, 39. Huse, M., L. O. Klein, A. T. Girvin, J. M. Faraj, Q. J. Li, M. S. Kuhns, and D. Waghray, S. Fischer, J. McCluskey, J. Rossjohn, et al. 2014. Molecular ar- M. M. Davis. 2007. Spatial and temporal dynamics of T cell receptor signaling chitecture of the ab T cell receptor-CD3 complex. Proc. Natl. Acad. Sci. USA with a photoactivatable agonist. Immunity 27: 76–88. 111: 17576–17581. 40. Stepanek, O., A. S. Prabhakar, C. Osswald, C. G. King, A. Bulek, D. Naeher, 21. Call, M. E., J. Pyrdol, M. Wiedmann, and K. W. Wucherpfennig. 2002. The M. Beaufils-Hugot, M. L. Abanto, V. Galati, B. Hausmann, et al. 2014. Coreceptor organizing principle in the formation of the T cell receptor-CD3 complex. Cell scanning by the T cell receptor provides a mechanism for T cell tolerance. Cell 111: 967–979. 159: 333–345. 22. Sun, Z. Y., S. T. Kim, I. C. Kim, A. Fahmy, E. L. Reinherz, and G. Wagner. 2004. 41. Schrum, A. G., D. Gil, E. P. Dopfer, D. L. Wiest, L. A. Turka, W. W. Schamel, Solution structure of the CD3epsilondelta ectodomain and comparison with and E. Palmer. 2007. High-sensitivity detection and quantitative analysis of CD3epsilongamma as a basis for modeling T cell receptor topology and sig- native protein-protein interactions and multiprotein complexes by flow cytom- naling. Proc. Natl. Acad. Sci. USA 101: 16867–16872. etry. Sci. STKE 2007: pl2. 23. Call, M. E., K. W. Wucherpfennig, and J. J. Chou. 2010. The structural basis for 42. Roh, K. H., B. F. Lillemeier, F. Wang, and M. M. Davis. 2015. The coreceptor intramembrane assembly of an activating immunoreceptor complex. Nat. CD4 is expressed in distinct nanoclusters and does not colocalize with T-cell Immunol. 11: 1023–1029. receptor and active protein tyrosine kinase p56lck. Proc. Natl. Acad. Sci. USA 24. Kim, K. J., C. Kanellopoulos-Langevin, R. M. Merwin, D. H. Sachs, and 112: E1604–E1613. R. Asofsky. 1979. Establishment and characterization of BALB/c lymphoma 43. Minami, Y., A. M. Weissman, L. E. Samelson, and R. D. Klausner. 1987. lines with B cell properties. J. Immunol. 122: 549–554. Building a multichain receptor: synthesis, degradation, and assembly of the 25. Letourneur, F., and B. Malissen. 1989. Derivation of a T cell hybridoma variant T-cell antigen receptor. Proc. Natl. Acad. Sci. USA 84: 2688–2692. deprived of functional T cell receptor alpha and beta chain transcripts reveals a 44. Krogsgaard, M., Q. J. Li, C. Sumen, J. B. Huppa, M. Huse, and M. M. Davis. nonfunctional alpha-mRNA of BW5147 origin. Eur. J. Immunol. 19: 2269–2274. 2005. Agonist/endogenous peptide-MHC heterodimers drive T cell activation Downloaded from 26. Parrish, H. L., C. R. Glassman, M. M. Keenen, N. R. Deshpande, and sensitivity. Nature 434: 238–243. M. P. Bronnimann, and M. S. Kuhns. 2015. A transmembrane domain GGxxG 45. Brazin, K. N., R. J. Mallis, D. K. Das, Y. Feng, W. Hwang, J. H. Wang, motif in CD4 contributes to its Lck-independent function but does not mediate G. Wagner, M. J. Lang, and E. L. Reinherz. 2015. Structural features of the CD4 dimerization. PLoS One 10: e0132333. abTCR mechanotransduction apparatus that promote pMHC discrimination. 27. Hedrick, S. M., L. A. Matis, T. T. Hecht, L. E. Samelson, D. L. Longo, E. Heber- Front. Immunol. 6: 441. Katz, and R. H. Schwartz. 1982. The fine specificity of antigen and Ia deter- 46. Akrap, N., T. Seidel, and B. G. Barisas. 2010. Fo¨rster distances for fluorescence minant recognition by T cell hybridoma clones specific for pigeon cytochrome c. resonant energy transfer between mCherry and other visible fluorescent proteins.

Cell 30: 141–152. Anal. Biochem. 402: 105–106. http://www.jimmunol.org/ 28. Newell, E. W., L. K. Ely, A. C. Kruse, P. A. Reay, S. N. Rodriguez, A. E. Lin, 47. Ko¨nig, R., L. Y. Huang, and R. N. Germain. 1992. MHC class II interaction with M. S. Kuhns, K. C. Garcia, and M. M. Davis. 2011. Structural basis of specificity CD4 mediated by a region analogous to the MHC class I binding site for CD8. and cross-reactivity in T cell receptors specific for cytochrome c-I-E(k). J. Nature 356: 796–798. Immunol. 186: 5823–5832. 48. Ko¨nig, R., X. Shen, and R. N. Germain. 1995. Involvement of both major his- 29. Parrish, H. L., N. R. Deshpande, J. Vasic, and M. S. Kuhns. 2016. Functional tocompatibility complex class II alpha and beta chains in CD4 function indicates evidence for TCR-intrinsic specificity for MHCII. Proc. Natl. Acad. Sci. USA a role for ordered oligomerization in T cell activation. J. Exp. Med. 182: 779–787. 113: 3000–3005. 49. Xu, H., and D. R. Littman. 1993. A kinase-independent function of Lck in po- 30. Mayer, M. L. 2005. Crystal structures of the GluR5 and GluR6 ligand binding tentiating antigen-specific T cell activation. Cell 74: 633–643. cores: molecular mechanisms underlying kainate receptor selectivity. Neuron 45: 50. Vignali, D. A., R. T. Carson, B. Chang, R. S. Mittler, and J. L. Strominger. 1996. 539–552. The two membrane proximal domains of CD4 interact with the T cell receptor. 31. Mayer, M. L., R. Olson, and E. Gouaux. 2001. Mechanisms for ligand binding to J. Exp. Med. 183: 2097–2107.

GluR0 ion channels: crystal structures of the glutamate and serine complexes 51. Vignali, D. A., and K. M. Vignali. 1999. Profound enhancement of T cell acti- by guest on September 29, 2021 and a closed apo state. J. Mol. Biol. 311: 815–836. vation mediated by the interaction between the TCR and the D3 domain of CD4. 32. Armstrong, N., and E. Gouaux. 2000. Mechanisms for activation and antagonism J. Immunol. 162: 1431–1439. of an AMPA-sensitive glutamate receptor: crystal structures of the GluR2 ligand 52. McKeithan, T. W. 1995. Kinetic proofreading in T-cell receptor signal trans- binding core. Neuron 28: 165–181. duction. Proc. Natl. Acad. Sci. USA 92: 5042–5046. 33. Reddy Chichili, V. P., V. Kumar, and J. Sivaraman. 2013. Linkers in the structural 53. Killeen, N., and D. R. Littman. 1993. Helper T-cell development in the absence biology of protein-protein interactions. Protein Sci. 22: 153–167. of CD4-p56lck association. Nature 364: 729–732. 34. Zal, T., and N. R. Gascoigne. 2004. Using live FRET imaging to reveal early 54. Mallaun, M., D. Naeher, M. A. Daniels, P. P. Yachi, B. Hausmann, I. F. Luescher, protein-protein interactions during T cell activation. Curr. Opin. Immunol. 16: N. R. Gascoigne, and E. Palmer. 2008. The T cell receptor’s alpha-chain con- 418–427. necting peptide motif promotes close approximation of the CD8 coreceptor 35. Antczak, A. J., T. Tsubota, P. D. Kaufman, and J. M. Berger. 2006. Structure of allowing efficient signal initiation. J. Immunol. 180: 8211–8221. the yeast histone H3-ASF1 interaction: implications for chaperone mechanism, 55. Doucey, M. A., L. Goffin, D. Naeher, O. Michielin, P. Baumga¨rtner, species-specific interactions, and epigenetics. BMC Struct. Biol. 6: 26. P. Guillaume, E. Palmer, and I. F. Luescher. 2003. CD3 delta establishes a 36. Brant, D. A., W. G. Miller, and P. J. Flory. 1967. Conformational energy esti- functional link between the T cell receptor and CD8. J. Biol. Chem. 278: mates for statistically coiling polypeptide chains. J. Mol. Biol. 23: 47. 3257–3264. A

5 10 105 CD3δTG CD4TmCh CD3γTG CD4TmCh Parental M12 104 104

PE-A 3 PE-A 103 10

102 102

0 0

0 102 103 104 105 0 102 103 104 105 GFP-A GFP-A CD3 subunit (GFP)

5 10 105 CD3εTG CD4TmCh CD3ζTG CD4TmCh Parental M12 4 10 104 GFP/PE

PE-A 3 10 PE-A 103

2 2 (2C11 PE) (2C11 10 10 ε 0 0 CD3

2 3 4 5 0 10 10 10 10 0 102 103 104 105 GFP-A GFP-A CD3 subunit (GFP)

B 0.10

0.08

0.06 )/(2C11 PE) )/(2C11 0.04 mEGFP ( 0.02

0.00 CD3δTG CD3γTG CD3εTG CD3ζTG CD4TmCh CD4TmCh CD4TmCh CD4TmCh

d g e z Supplement 1. TCR-CD3 complex containing mEGFP tagged CD3 subunits assemble with appropriate stoichiometry. Whole cell lysates from M12 cell lines were analyzed using a flow-based flourophore-linked immunosorbant assay (FFLISA). Cell lysates were immunoprecipitated (IP’d) with anti-TCR (anti-TCRβ mAb H57)- coated latex beads as described in the methods. Detection of TCR (anti-TCRα mAb APC), CD3ε (2C11 PE) and CD3xT-mEGFP subunits (by GFP intensity) on the beads was used to assess complex assembly. (A) Proportional relationship between CD3ε and CD3xTG. Dot plots of flow cytometry analysis of beads are shown as labeled. The same populations are shown in the left and right plots but with different populations in the foreground. This was done to emphasize the equivalent slope of the diagonal, which reflects the proportional distribution of the IP’d subunits. (B) Appropriate stoichiometry of TCR-CD3. Bars represent the ratio of mEGFP (gMFI) to CD3ε (gMFI) for beads with matched TCRα intensity (i.e. equivalent TCR IP), and reflect the proportion of CD3 subunits per complex. Results are representative of three experiments.

A B C

99.1Å 99.1Å

99.1Å

D E

Supplement 2. CD3γ and CD3ζ are likely to be >100Å from CD4 in a V-like arch. (A–C) Surface rendered representations of the TCR-pMHC-CD4 ternary complex showing CD4 (salmon), class II MHC (green), TCRα (gray) and TCRβ (magenta) from the (A-B) side and (C) a view as if looking out from the T-cell surface. The ternary TCR-pMHC-CD4 structure was used as a staring point (PDB ID 3T0E). Our experiments were performed using the 5c.c7 TCR engaging MCC:I-Ek, so we aligned the structure of the highly related 226 TCR bound to MCC:I-Ek (PDB ID 3QIU) via the MHC to consider the docking of the cytochrome c related TCRs on their cognate ligand. Finally, we aligned the 2C TCR structure (PDB ID 1TCR) to the 226 TCR to take advantage of the well refined constant regions, with an intact inter-chain disulfide bond, to better appreciate the mouse TCR constant regions. The net result shown is the 2C TCR shown with the MHC and CD4 from the original ternary structure. The distance shown (99.1Å) is the distance measured from the most C-terminal residue of the TCRα subunit to the most C-terminal residue of CD4 using PyMol (PyMOL Molecular Graphics System, Version 1.7.6.2. Schrödinger, LLC). (D-E) Models of the TCR-CD3 complex that are consistent with the FRET data presented here would likely place CD3γ and CD3ζ >100Å from CD4 in a V-like arch. Shaded spheres represent general positioning of the ectodomains of the CD3 heterodimers (D) clustered on the face of the TCR inside the arch with CD4 in the order δ:ε:ε:γ or (E) placed on opposite sides of the TCR. The CD3γε (PDB ID 1SY6) and CD3δε (PDB ID 1XIW) structures were used to approximate their placement on opposite side of the TCR as depicted in (32), and used to adjust our model (1) in consideration of the recent EM structure (20) such that the CD3 heterodimers are nested further below the TCR. Solid spheres represent rough approximation of where the transmembrane domains would emerge from the membrane into the cytoplasm (CD3δ: cyan; CD3γ: orange; CD3ζ: red; CD3ε: marine). For these placements, we used the C-terminal end of the CD3 subunit G-strands and the most C-terminal residue of CD4. To place the CD3 heterodimers in relationship to the TCR subunits we used the transmembrane domain heterotrimer of the NKG2C receptor as a guide (PDB ID 2L35). This was also used as a guide to place the CD3ζζ subunits ~140º offset from CD3δε based on previous work (21). Yellow dashed line represents the boundary of a circle centered at CD4 (salmon dot) with a radius of ~100Å.