design of tcr structural variants that retain or invert the … · tcr signaling (18, 19). in this...

Normal Activation SignalDesign of TCR Structural Variants That Retain or Invert the

Jee-Young Mock, Julyun Oh, Jason Yi, Mark E. Daris, Agnes Hamburger and Alexander Kamb

http://www.immunohorizons.org/content/5/5/349https://doi.org/10.4049/immunohorizons.2100033doi:

2021, 5 (5) 349-359ImmunoHorizons

This information is current as of September 9, 2021.

MaterialSupplementary

s.2100033.DCSupplementalhttp://www.immunohorizons.org/content/suppl/2021/05/26/immunohorizon

Referenceshttp://www.immunohorizons.org/content/5/5/349.full#ref-list-1

, 18 of which you can access for free at: cites 50 articlesThis article

Email Alertshttp://www.immunohorizons.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

ISSN 2573-7732.All rights reserved.1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is an open access journal published byImmunoHorizons

by guest on September 9, 2021

http://ww

w.im

munohorizons.org/

Dow

nloaded from


http://ww

w.im

munohorizons.org/

Dow

nloaded from

https://doi.org/10.4049/immunohorizons.2100033

http://www.immunohorizons.org/content/5/5/349

http://www.immunohorizons.org/content/suppl/2021/05/26/immunohorizons.2100033.DCSupplemental

http://www.immunohorizons.org/content/suppl/2021/05/26/immunohorizons.2100033.DCSupplemental

http://www.immunohorizons.org/content/5/5/349.full#ref-list-1

http://www.immunohorizons.org/alerts

http://www.immunohorizons.org/


Design of TCR Structural Variants That Retain or Invert theNormal Activation Signal

Jee-Young Mock, Julyun Oh, Jason Yi,1 Mark E. Daris, Agnes Hamburger, and Alexander KambA2 Biotherapeutics, Agoura Hills, CA

ABSTRACT

We designed variant human TCRs composed of the full-length TCRa/b or extracellular and transmembrane domains of the associated

CD3 subunits fused to polypeptides derived from proteins thought to either enhance or inhibit normal T cell function. First, we

showed that the C termini of both the TCR a- and b-chains can accommodate specific additional sequences, without abrogating

complex formation or acute sensitivity of the receptor. Replacement of ITAMs with ITIM-containing intracellular domains inverted

the TCR signal (i.e., created a ligand-dependent inhibitory receptor). The normal signaling function of the CD3 complex was

transferable to the TCR by eliminating all CD3 ITAMs and grafting three to six ITAMs onto the C termini of the a/b-chains, with no

effect on acute sensitivity. The observation that TCR variants of such diverse C-terminal composition can fold and function as

signaling receptors demonstrates substantial structural and functional malleability of TCRs. These results add to knowledge about

TCR structure–function with regard to acute signaling and may provide a route to use TCRs in different ways for T cell therapy.

ImmunoHorizons, 2021, 5: 349–359.

INTRODUCTION

The TCR is an extraordinary signaling device with the ca-pacity to trigger T cell response to cellular or artificial sur-faces that display as few as one to two Ag molecules (1, 2).The mechanism by which the TCR achieves its high sensi-tivity is still controversial. Several models of receptor func-tion have been proposed, none of which account for allaspects of TCR signaling (3). The high conservation of theprimary and quaternary structures suggests a preciselytuned machine, in which exact geometries may be impor-tant to achieve sensitivity (4). At a minimum, it is thoughtthat all TCR/CD3 subunits (a, b, g, d, e, z) are required forfunction in human T cells, and this multisubunit structureis conserved over perhaps 300 million years of evolution,back to the divergence of mammals and avians (5).

Although lacking important structure-function details, TCRsignaling is understood in outline. Ligand binding activates acascade of phosphorylation events that initially center aroundthe CD3 ITAMs, intracellular sequence motifs that are phos-phorylated during signal transduction. Activated TCRs ulti-mately recruit a variety of kinases and adapter proteins withina microcluster of ligand-bound TCRs (6). The role of ITAMs indevelopment of the TCR repertoire in vivo has been exploredby removing individual or groups of ITAMs (7–12). Meanwhilea largely independent branch of research has demonstrated themodularity of T cell signaling components, notably through de-sign of CARs, monomeric signaling molecules that use CD3zITAMs, and other intracellular domains (ICDs) derived fromcostimulatory receptors [e.g., CD28, 4-1BB (13, 14)]. Elegantwork in vitro has confirmed a quantitative relationship betweenthe number of ITAMs and signal transduction using CAR

Received for publication April 2, 2021. Accepted for publication April 2, 2021.

Address correspondence and reprint requests to: Jee-Young Mock and Alexander Kamb, A2 Biotherapeutics, 30301 Agoura Road, Suite 210, Agoura Hills, CA 91301.E-mail addresses: [email protected] (J.-Y. M.) and [email protected] (A.K.)1Current address: ImmPACT-Bio, Camarillo, CA.

ORCID: 0000-0002-4656-3357 (J.-Y. M.).

Abbreviations used in this article: APL, altered-peptide ligand; gRNA, guide RNA; HIA, heat-inactivated; ICD, intracellular domain; invTCR, inverter TCR; KIR, KIR3DL2;LIR-1, LILRB1; PD-1, programmed cell death protein 1; pen/strep, penicillin and streptomycin; pMHC, peptide MHC; sgRNA, single gRNA.

The online version of this article contains supplemental material.

This article is distributed under the terms of the CC BY 4.0 Unported license.

Copyright © 2021 The Authors

https://doi.org/10.4049/immunohorizons.2100033 349

ImmunoHorizons is published by The American Association of Immunologists, Inc.

RESEARCH ARTICLE

Clinical and Translational Immunology


http://ww

w.im

munohorizons.org/

Dow

nloaded from

http://orcid.org/0000-0002-4656-3357

mailto:[email protected]

mailto:[email protected]

http://orcid.org/0000-0002-4656-3357

https://www.immunohorizons.org/lookup/suppl/doi:10.4049/immunohorizons.2100033/-/DCSupplemental

https://creativecommons.org/licenses/by/4.0/


constructs (15). TCRs and CARs are also subject to negativemodulation of their signal, notably by ITIM-containing inhibi-tory receptors such as PD-1. ITIMs are a family of phosphory-lation-attracting domains that oppose ITAM signaling (see forreview Refs. 16, 17). These motifs have been co-opted in the de-sign of artificial monomeric inhibitory receptors (iCAR andTmod) that produce ligand-dependent inhibition of CAR andTCR signaling (18, 19).

In this study, we set out to test the limits of TCR structure–function with respect to acute signaling. We concentrated onthe ITAM and ITIM domains that are known to initiate andmodulate acute response of T cells. Through a series of domaingrafting and swapping experiments, we show that despite theTCR�s multisubunit, evolutionarily conserved structure: 1) theTCR/CD3 ICDs can be substantially altered without affectingcomplex formation or ligand binding; 2) these variants confernormal or divergent signaling outputs, depending on the ICDs;and 3) replacement of the 10 CD3 ITAMs with three to sixITAMs fused to the a/b TCR subunits forms a receptor withindistinguishable acute sensitivity compared with its wild-typeTCR counterpart. These results add to knowledge about TCRstructure–function vis-�a-vis signaling and may provide a routeto use TCRs in different ways for T cell therapy, for example,as peptide MHC (pMHC)-regulated inhibitory receptors.

MATERIALS AND METHODS

Plasmid constructionAll variants of A*02:01-NY-ESO-1 directed TCR were derivedfrom 1G4 a95:LY/wtb (20). TCRa/b fusions containing theNY-ESO-1–responsive activating and inhibitory constructs werecreated as previously described (19). Briefly, ICDs from CD28,CD4, CD8a, 4-1BB, various truncations of Lck, Fyn, ZAP70,and LAT were fused to TCRa/b with either S, GGS, (G4S)2,(G4S)3, or (G4S)4 as linkers. ITIM-fused TCRa/b constructswere created by fusing the ICDs of LILRB1 (LIR-1; R483-H649), programmed cell death protein 1 (PD-1; S192-L287), orKIR3DL2 (KIR; N365-F454) with a (G4S)4 linker for TCRa and(G4S)3 linker for TCRb. For CD3 subunits directly fused to in-hibitory domains, the extracellular domain and transmembranedomain of the CD3 subunits (CD3) were directly fused to LIR-1(R483-H649). For CD3 constructs whose ITAMs were replacedby ITIMs, the conserved ITAM motif (YxxL/I) of each CD3subunit was replaced with the region spanning LIR-1 ITIM3and 4 (RESI 611-647). The rest of the CD3 backbone was leftintact. All constructs used in this study are described in detailin Supplemental Table I. All constructs were assembled usingGolden Gate Assembly.

Cell cultureJurkat cells encoding an NFAT luciferase reporter gene (BPSBioscience) were maintained in RPMI media supplementedwith 10% heat-inactivated (HIA) FBS (inactivated at 56�C for 1h), 1% penicillin and streptomycin (pen/strep), and 0.4 mg/ml

Geneticin. T2 cells were obtained from the American Type Cul-ture Collection and maintained in IMDM supplemented with20% HIA FBS and 1% pen/strep.

Generation of CD3-knockout Jurkat NFAT luciferasereporter linesGuide RNAs (gRNA) for CRISPR knockout were obtainedfrom Synthego. EnGen Spy Cas9 containing SV40 nuclearlocalization sequence at the N and C termini was pur-chased from New England Biolabs. Prior to transfection,single gRNA (sgRNA) was resuspended to 100 mM. Thedissolved sgRNA was further diluted to 30 mM in 1� Tris–EDTA buffer purchased from Synthego. gRNA sequences tar-geting CD3d are as follows: G*G*U*CCAGGAUGCGUUUUCCC, C*A*U*CACAUGGGUAGAGGGAA, and C*C*U*CUAUAG-GUAUCUUGAAG. gRNA sequences targeting CD3e are as fol-lows: U*U*U*CAGAUCCAGGAUACUGA, G*A*U*GAGGAUGAUAAAAACAU, and U*A*U*UAUGUCUGCUACCCCAG. Theasterisk (*) represents 2'-O-methyl analogs and 3'-phosphor-thioate internucleotide linkages reported to improve stabilityand knockout efficiency (21). First, each sgRNA targeting CD3dor CD3e was coincubated with Cas9 protein at room tempera-ture for 15 min. The final concentration of sgRNA was 3.6 mMand Cas9 2.6 mM in total 10 ml of volume of R buffer providedby the Neon Transfection Kit (Thermo Fisher Scientific). Afterthe coincubation, the RNA protein complexes targeting CD3dand CD3e were combined. A total of 15 ml of the RNA proteincomplex was then combined with 15 ml of Jurkat NFAT lucif-erase reporter cells resuspended at 2e6/ml in the Neon R buffer(Thermo Fisher Scientific). Using the 10-ml format Neon Elec-troporation System, the Jurkat cells were transfected at 1400 V,10 ms, three pulses. Transfected cells were immediately trans-ferred to prewarmed RPMI 1640 supplemented with 20% HIAFBS and 0.1% pen/strep and incubated for 48 h at 37�C and 5%CO2 (19).

CD3d/e knockout cells were stained with PE-conjugatedSK7 (BioLegend) to confirm lack of CD3 expression, expanded,and functionally validated to be CD3d/e-null using genetic com-plementation as described (Fig. 3A). Briefly, CD3d/e knockoutcells were transfected using the 100-ml format Neon Electropo-ration System as described above with TCRa and TCRb subu-nits targeting A*02:01-NY-ESO-1 (clone 1G4 a95:LY) (20) witheither CD3d alone, CD3e alone, or with both CD3d and CD3e.Rescue of functional activity was measured by coculturing withpeptide-loaded T2s as was done previously (19) and describedbelow. The pool of CD3d/e knockout cells generated was thenfurther engineered to remove CD3g and CD3z as above andfunctionally validated (Fig. 3A). U*C*U*CAUUUCAGGAAAC-CACU, C*A*G*AAGCCAAAAAUAUCACA, and A*U*U*UUUUUUUAUCUUCAGUU are gRNA sequences used to target CD3g. G*A*A*ACUCUUUGUGUAUGUGU, A*C*A*GUUGCCGAUUACAGGUA, and A*A*A*GGACAAGAUGAAGUGGA are gRNAsequences used to target CD3z.

350 GENERATION OF AN INVERTER TCR ImmunoHorizons



http://ww

w.im

munohorizons.org/

Dow

nloaded from

http://www.immunohorizons.org/lookup/suppl/doi:10.4049/immunohorizons.2100033/-/DCSupplemental


Jurkat cell assaysJurkat NFAT luciferase assays were carried out as previouslydescribed (19). Briefly, modified NY-ESO-1(v) peptide(SLLMWITQV) (Genscript) was serially diluted 3-fold startingat 100 mM and loaded onto 1e4 T2 cells resuspended in 15 ml ofRPMI 1640, 1% BSA, and 0.1% pen/strep and incubated over-night. Jurkat NFAT luciferase cells were electroporated usingthe 100 ml format Neon Transfection Kit (1500 V, 10 ms, threepulses) with 1 mg of inhibitory TCRa/b variant DNAs andITAM-null CD3 subunits with 0.5–1 mg of DNA of CD19 CARactivator per 1e6 cells. Cells were cultured in RPMI 1640 sup-plemented with 20% HIA FBS and 0.1% pen/strep for �18–24h posttransfection. Jurkat cells were counted and resuspendedin RPMI 1640 supplemented with 10% HIA FBS and 0.1% pen/strep. A total of 1e4 resuspended Jurkat cells were coculturedwith T2 cells loaded with varying amounts of NY-ESO-1(v)peptide for 4–6 h. Luciferase activity was measured usingONE-Step Luciferase Assay System (BPS Bioscience).

RESULTS

C termini of the TCRa and b subunits can be fused toactivating and inhibitory domains without affecting acutesensitivityWe first tested the feasibility of fusing exogenous sequencesto the C termini of a benchmark NY-ESO-1 TCR (22); GS link-ers of varying lengths were also tested. A variety of constructsderived from signaling molecules thought to enhance or inhib-it TCR function were tested in this context (SupplementalTable I). A subset of these ICDs fused to the C terminus ofTCRa or TCRb did not harm acute TCR-mediated NFAT lu-ciferase response in Jurkat cells when cocultured with pepti-de-loaded T2 cells (Fig. 1); in particular, domains from CD28,CD4, 4-1BB, LCK, FYN, and LAT produced functional TCRswith minimal shift of EC50 when fused to the C terminus ofeither or both TCR chains. These domains encompassed bothglobular and extended structures. Surprisingly, no definitiverules were observed regarding which domains were accom-modated well by the TCR complex, but some trends were ap-parent; for example, longer (GGGGS)x3 or x4 linkers werepreferred to shorter S or (GGS)x1 (Fig. 1B), and b-chain wasslightly more preferred (Fig. 1C). Furthermore, the NY-ESO-1TCRa and b fused via their C termini to KIR, PD-1, or LIR-1ITIM domains did not harm the TCR EC50 in this context(Fig. 2, Supplemental Fig. 1A). To summarize these results, 1)C termini of TCRs tolerated significant engineering; 2)b-chain fusions were potentially more stable and sensitive, aswere longer G4S linkers; and 3) ITIM-containing domains didnot produce detectable change in dose-response.

Creation of CD3-deficient Jurkat cells as an assay systemfor TCR/CD3 variantsThe lack of EC50 shift in TCRs fused to inhibitory domainsin wild-type Jurkat cells suggested that the 10 endogenous

CD3 ITAMs may overpower any inhibitory effect exertedby ITIMs. To explore other TCR designs, including invert-er TCRs (invTCRs) that might invert the normal activationsignal, it was necessary to generate a Jurkat cell line defi-cient in CD3 expression as the basis for an assay. Jurkatcells express CD3, which poses a problem for assessmentof the functional effect of added subunits because of thepossibility of structurally mixed TCR complexes. To pro-vide a clean assay background devoid of wild-type CD3subunit function, we used CRISPR/Cas9 with a pool ofgRNAs to target the genomic CD3 loci in Jurkat cells (seeMaterials and Methods). After transfection of the gRNAs/Cas9 RNA protein complex, Jurkat cells were grown andcloned by single-cell dilution. The sequence relatednessand complexity of the CD3-encoding loci hindered assess-ment of genetic knockout by standard Sanger sequencingmethods. We therefore did a complementation test to de-tect residual subunit function in two of the clones (clones24 and 32). CD3 subunits without activating intracellularITAM domains (DITAMCD3) were expressed in selectedknockout clones. If the clone lacked functional CD3 subu-nits, then expression of DITAMCD3 (null) subunits wouldlead to TCR/CD3 surface expression but no NFAT lucifer-ase response when stimulated by A*02:01-NY-ESO-1. Morespecifically, we searched for functional complementationof individual CD3 subunits by transfecting all combina-tions of CD3 subunits, with one subunit missing. Clones 24and 32 were confirmed as CD3-null on this basis and wereselected for detailed study of the invTCRs (Fig. 3A,Supplemental Fig. 1B).

Reconstitution of invTCR function in CD3-deficient JurkatcellsHaving generated a DCD3 Jurkat cell line, we sought toconfirm our hypothesis that endogenous CD3 ITAMs sup-press the inhibitory effects of TCRa/b-ITIM fusions. Wetherefore coexpressed either wild-type NY-ESO-1 TCRaand b (wild-type TCRa/b) or NY-ESO-1 TCRa and b fusedvia their C termini to KIR, PD-1, or LIR-1 ITIM domainsas above (Fig. 2, Supplemental Fig. 1A), along withDITAMCD3 subunits, in CD3-null Jurkat cells (Fig. 3B). Itwas only possible to measure an inhibitory signal fromITIMs in our system indirectly by its effect on activation(19). To test for ligand-dependent inhibition of an activat-ing signal, we coexpressed a CD19 CAR to serve as the ac-tivating receptor. We thus determined the IC50 of theITIM-fused TCRa/b in the presence of CD191 T2 cells.The different TCRa/b subunits expressed if and only ifcoexpressed with DITAMCD3 subunits (Fig. 3C, top pan-el). As expected, wild-type TCRa/b, when combined withDITAMCD3, showed no activation/inhibition upon ligand binding(Fig. 3B, Supplemental Fig. 1C, black closed circle). TCRa/b-PD-1,TCRa/b-KIR, and TCRa/b-LIR-1, in contrast, showed reproduc-ible, ligand-dependent inhibition of CD19 CAR activation (Fig. 3B,

ImmunoHorizons GENERATION OF AN INVERTER TCR 351



http://ww

w.im

munohorizons.org/

Dow

nloaded from








Supplemental Fig. 1C, green, light blue, and purple). These resultsconfirm that the CD3-null Jurkat cells are functionally compro-mised for CD3 function, ITAMs dominate ITIMs, and TCRa/bcan signal as an inhibitory receptor via C-terminally fused ITIMdomains if ITAMs are eliminated from the complex.

We next sought to improve the potency of the invTCRin CD3-deficient Jurkat cells. To do this, we focused onthe ICD of LIR-1, possibly the most-potent member of theinhibitory class of T cell modulators (19, 23). LIR-1 ITIM-containing CD3 subunits were expressed in clone 32 cells,along with NY-ESO-1 TCRa and b, resulting in surface ex-pression of TCR/CD3 (Fig. 4A). We also expressed TCR/CD3 chains with all four different CD3 subunits fused toITIM-containing domains. We determined the IC50 valuesof the inverter TCR-CD3 complex against CD19 CAR, asabove. Expressing LIR-1–fused CD3d/e or CD3g/z showedsimilar inhibition to fully reconstituted invTCR-CD3 com-plex (Fig. 4A, Supplemental Fig. 1D). The reconstitutedTCR IC50 was compared with previously characterized in-hibitory constructs (�blockers�) that use either an Ftcr(TCR ligand-binding domain) or scFv against HLA-A*02:01-NY-ESO-1 pMHC (19). The blocking activity in the pres-ence of CD19 CAR was similar among the scFv, Ftcr, andinvTCR constructs (Fig. 4B, Supplemental Fig. 1E). Thus,the switch of ITAMs for ITIMs converted the NY-ESO-1TCR into a ligand-dependent inhibitory receptor similar to,but no more potent than, other blocker constructs composedof CAR backbones tested previously.

We next compared the LIR-1–based invTCR to a more na-tive-like inhibitory domain (Fig. 4C). Inhibitory curves wereagain generated using the CD19 CAR to mediate activation inclone 32 cells. To generate a more native-like structure with re-gard to ICD sequence length, the CD3 ITAMs were replacedby 36 (LIR-1 611-646) residues spanning LIR-1�s ITIM 3 and 4(orange). We included these ITIMs because of their importancein SHP-1 binding (24). CD3 ICDs range from 45 to 112 residues,whereas the full-length LIR-1 ICD is 168 residues. The ITAM3/4 domain constructs were in this range, 61–168 residues (see Mate-rials and Methods). Both LIR-1 sequence–containing invTCRsblocked well (Fig. 4C, Supplemental Fig. 1F). This result suggeststhat LIR-1 611-646 is sufficient to recruit SHP-1 and this inhibitoryfunction is transferable.

A

B

C

0.0 0.5 1.0 1.5 2.00

1

2

3

Relative EC50

Rel

ativ

epr

obe

bind

ing

TCRα-fusions

TCRα/β-fusions

CD28((G4S)2)

CD28((G4S)2)

4-1BB(L)

α-4-1BB(L), β-CD28

CD28FYN(L)

CD4(L)

CD28

Lck(2-509; Y509A)Lck(73-509)CD28(GGS)

TCRβ-fusions

short

long

short,

short

long, long

short,

long0

1

2

3

Rel

ativ

epr

obe

bind

ing

short

long

short,

short

long, long

short,

long0.0

0.5

1.01.01.52.0

Rel

ativ

eEC

50

α fusio

ns

β fusio

ns

α + β fu

sions

0

1

2

3

Rel

ativ

epr

obe

bind

ing

α fusio

ns

β fusio

ns

α + β fu

sions

0.0

0.5

1.0

1.5

2.0

Rel

ativ

eEC

50

FIGURE 1. A*02:01 NY-ESO-1 directed TCR C-terminal modifications

have negligible effect on acute TCR activity.

(A) Fluorescently labeled HLA-A*02:01-NY-ESO-1(v) tetramer complexes

(probe) binding, normalized to the benchmark TCR, is plotted on the y-

axis versus relative EC50 on the x-axis (see Materials and Methods). Dot-

ted lines align to benchmark TCR values. (B) Replotted from

data shown in (A) to compare different linker lengths. Singular short or

long data points derived from either TCRa or b C-terminal fusions

paired with corresponding wild-type subunit. (Short, short), (long, long),

and (short, long) data points derived from both subunit replacement

with C-terminal fusion constructs. (C) Replotted from data shown in (B)

to compare TCRa versus TCRb fusions. Fusions to TCR b are better tol-

erated, and longer linkers are preferred. Data are representative of one

to three independent experiments. Outliers were repeated, whereas

constructs similar to or worse than benchmark TCR were abandoned

after the first trial.




http://ww

w.im

munohorizons.org/

Dow

nloaded from






Charge-swapped and ITAM-shuffled TCR/CD3 complexesfunction in Jurkat cell assaysIn the process of further exploration of invTCRs, we uncoveredadditional structure-function relationships within the TCR/CD3 complex. These findings followed from efforts to demon-strate activity of the invTCR against a TCR as opposed to aCAR. As mentioned above, one challenge is the interminglingof wild-type and ITIM-fused CD3 subunits, potentially creatinga complex mixture of structures with differing ratios of ITIMsand ITAMs. Conclusions about structure-function relationshipsof the invTCR are clouded by these hybrid molecules. We thusattempted to create a structure that could assemble and func-tion independently of a wild-type receptor. The CD3 complexand its assembly with the TCR depends on charge-pair interac-tions within the membrane-spanning domains of the subunits(25, 26). We used this structural information to design charge-swapped subunits intended to attract correct subunit partnersbut repel corresponding wild-type subunits (Fig. 5A). We fo-cused particularly on the charge-pair interactions betweenTCRbK288 and CD3eD137/gE122 because they appeared to bethe most amenable to engineering within the TCR and CD3 por-tions of the complex. The TCRaR253–CD3zD15 (SupplementalFig. 2A) and TCRaK258–CD3dD111/eD137 interactions provedmore difficult to disrupt and rescue. Furthermore, when thesetwo mutations were combined, expression and activity were nev-er fully recovered (Supplemental Fig. 2C). In contrast, the posi-tively charged K288 residue of TCRb, which interacts withcorresponding negatively charged residues of CD3e (D137) andCD3g (E122), was more tractable. These charge pairs were inter-changed between subunits to generate distinct TCRb–CD3egcomplexes.

We tested the design premise of a charge-swapped twinTCR by expressing the twin subunits, otherwise wild type insequence, in the presence of NY-ESO-1 TCRa/b. EC50 valueswere measured for these constructs in Jurkat cells (Fig. 5A). Inwild-type Jurkat cells, either TCRa/b or TCRa and K288DTCRb were coexpressed. As expected, TCRa/b showed robustfunction upon ligand binding (Fig. 5B, black closed circle).However, replacing the b subunit with the K288D mutant dra-matically reduced the functional activity of the TCR, with�700-fold right shift compared with wild type (Fig. 5B, blackclosed circle versus open circle). This functional activity couldbe rescued by coexpression of the matching CD3g mutantE122R or E122K (Fig. 5B, gray and blue compared with opencircle). Surprisingly, expression of CD3g E122K without itscharge-swapped partner CD3e D137K/R completely rescued ac-tivity of the K288D TCRb in the presence of the wild-typeCD3d, e, and z subunits, suggesting some flexibility in the10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102

0

20000

40000

60000

80000

100000

NY-ESO-1 Peptide Concentration (uM)

Lum

ines

cenc

e(R

LU)

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 1020

20000

40000

60000

80000

100000


Lum

ines

cenc

e(R

LU)

A

B

C

TCRαTCRα-PD-1

TCRβ+TCRβ+

TCRα-KIR TCRβ+

TCRα-LIR-1 TCRβ+

α β α β α β α β



2

TCRαTCRα

TCRβ+TCRβ-PD-1+

TCRα TCRβ-KIR+

TCRα TCRβ-LIR-1+

TCRαTCRα-PD-1

++

TCRα-KIR +

TCRα-LIR-1 +

TCRβTCRβ-PD-1

TCRβ-KIR

TCRβ-LIR-1

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 1020

20000

40000

60000

80000

100000


Lum

ines

cenc

e(R

LU)

FIGURE 2. C-terminal ITIM fusions to TCRa or TCRb have negligible

effect on acute TCR activity.

Jurkat NFAT luciferase cells were transfected with either TCRa (A) or

TCRb (B) or both (C) fused to the ICD of PD-1 (green), KIR (light blue),

or LIR-1 (purple). Single subunit fusions were paired with corresponding

wild-type subunit. At �18–24 h posttransfection, transfected Jurkat cells

were cocultured with T2 cells loaded with titrating amounts of NY-ESO-

1(v) peptide for 6 h, and NFAT luciferase response was measured.




http://ww

w.im

munohorizons.org/

Dow

nloaded from





organization of subunits in the TCR to balance the charges inthe transmembrane segments (Fig. 5B, black open circle versusmagenta).

We next constructed an invTCR equivalent to the charge-swapped twin TCR (Fig. 5C). We first tested whether the mu-tant invTCR containing LIR-1 ITIM3/4 in a CD3g/z backbonewas functional against a CD19 CAR. As expected, the trans-membrane charge-swapped mutant invTCR complex blockedCD19 CAR-mediated activation in the presence of CD191 targetcells (Fig. 5D and Supplemental Fig. 2E). We then sought totest this mutant invTCR against a wild-type TCR in wild-typeJurkat cells. If successful, this construct would allow us to mea-sure the blocking potency of an invTCR on a second TCR. Un-fortunately, the level of subunit mixing of a charge-swappedTCRb with wild-type CD3g/z subunits was too high to sepa-rate activation from blockade when dosed with the invTCR tar-get ligand (Supplemental Fig. 2D). We speculate thatintroduction of additional variants to enhance the specificity ofsubunit interactions would ultimately resolve this difficulty.

In light of these results, we took an alternative approach,which ultimately succeeded in testing the hypothesis thatinvTCRs may inhibit TCR/CD3 activating receptors. This ap-proach also yielded another remarkable result: namely, thatacute signaling can be transferred from the CD3 ITAMs to theTCR a/b-chains. To do this, we attempted to create a modified,active TCR and its reciprocal invTCR using a pool of CD3-nullsubunits shared between them. The concept was to generate anactive TCR variant in which the ITAMs were supplied not byCD3 but by fusion to the TCRa- and b-chain C termini (Fig.6A). Surprisingly, this design not only functioned but also pro-duced TCR variants with identical EC50 compared with wildtype coexpressed with wild-type CD3 subunits (Fig. 6B,Supplemental Fig. 3A, 3B). Thus, three to six ITAMs shuffledfrom CD3 to TCR were sufficient to produce a receptor withindistinguishable acute-signaling behavior in Jurkat cells.

The ITAM-shuffled TCR/CD3 enabled us to test if theinvTCR could inhibit activity of a TCR variant, as we hadshown for a CAR. To achieve this, we used a/b-chains fused toITIM, not ITAM, domains (Fig. 6A). These constructs were co-transfected into Jurkat cells with DITAMCD3 subunits to forminvTCR complexes. In addition, the TCRa/b ITAM fusionswere cotransfected to generate activating receptors in whichthe CD3-null subunits were shared with both the ITAM-shuf-fled activator and the invTCR. Surface expression of both theITAM-shuffled TCR and invTCR was comparable to previousexperiments (Fig. 6C). The IC50 of the NY-ESO-1 invTCR

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 1020

20000

40000

60000

80000


Lum

ines

cenc

e(R

LU)

CD3 KO JurkatA

B

C

Electroporate TCRα/β + various combinations of

ΔITAM CD3 subunits

Electroporate TCRα/β + wtCD3 subunits

α β

ζ

ε γδ ε

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 1020

10000

20000

30000

40000


Lum

ines

cenc

e(R

LU)

ΔITAMCD3(δ,ε,γ)

ΔITAMCD3(δ,ε,γ,ζ)

ΔITAMCD3(ε,γ,ζ)

ΔITAMCD3(δ,γ,ζ)

ΔITAMCD3(δ,ε,ζ)

wtCD3(δ,ε,γ,ζ)

Q10.49

Q213.8

Q39.83

Q475.8

101

102

103

104

105

Comp-PE-A

101

102

103

104

105

Com

p-A

PC

-A

Q10.22

Q29.16

Q315.2

Q475.5

101

102

103

104

105

Comp-PE-A

101

102

103

104

105

Com

p-A

PC

-A

Q11.99

Q220.0

Q33.90

Q474.1

101

102

103

104

105

Comp-PE-A

101

102

103

104

105

Com

p-A

PC

-A

Q10.43

Q20.22

Q30.057

Q499.3

101

102

103

104

105

Comp-PE-A

101

102

103

104

105

Com

p-A

PC

-A

Q11.22

Q26.95

Q30.36

Q491.5

101

102

103

104

105

Comp-PE-A

101

102

103

104

105

Com

p-A

PC

-A

Q11.04

Q214.4

Q34.87

Q479.7

101

102

103

104

105

Comp-PE-A

101

102

103

104

105

Com

p-A

PC

-A

αCD3E

A2-

NY

-ES

O-1

αCD3E

A2-

NY

-ES

O-1

αFMC63

A2-

NY

-ES

O-1

wtCD3(δ,ε,γ,ζ)

CD19 CAR CD19 CAR +wtTCRα/β

CD19 CAR +TCRα/β-PD-1

CD19 CAR +TCRα/β-KIR

CD19 CAR +TCRα/β-LIR-1

ΔITAMCD3(δ,ε,γ)

ΔITAMCD3(δ,ε,γ,ζ)

ΔITAMCD3(ε,γ,ζ)

ΔITAMCD3(δ,γ,ζ)

ΔITAMCD3(δ,ε,ζ)

CD19 CAR + wtTCRα/β

CD19 CAR + TCRα/β-PD-1

CD19 CAR + TCRα/β-KIR

CD19 CAR + TCRα/β-LIR-1

CD19 CAR only

Q13.39

Q212.5

Q37.83

Q476.2

101

102

103

104

105

APC-A

101

102

103

104

105

PE

-A

Q19.75

Q213.3

Q35.71

Q471.2

101

102

103

104

105

APC-A

101

102

103

104

105

PE

-A

Q15.66

Q219.3

Q38.43

Q466.7

101

102

103

104

105

APC-A

101

102

103

104

105

PE

-A

Q18.76

Q216.6

Q36.03

Q468.7

101

102

103

104

105

APC-A

101

102

103

104

105

PE

-A

Q10.12

Q20.092

Q331.9

Q467.9

101

102

103

104

105

APC-A

101

102

103

104

105

PE

-A

Q11.61

Q213.6

Q35.65

Q479.2

101

102

103

104

105

APC-A

101

102

103

104

105

PE

-A

Q12.04

Q220.0

Q35.24

Q472.7

101

102

103

104

105

APC-A

101

102

103

104

105

PE

-A

Q11.67

Q222.1

Q34.98

Q471.2

101

102

103

104

105

APC-A

101

102

103

104

105

PE

-A

Q12.65

Q220.4

Q34.60

Q472.3

101

102

103

104

105

APC-A

101

102

103

104

105

PE

-A

Q10.66

Q20.016

Q30.14

Q499.2

101

102

103

104

105

APC-A

101

102

103

104

105

PE

-A

γα β

ζ

ε δ ε

TCRα/β + ΔITAMCD3δ/ε/γ/ζ

TCRα/β-PD-1 + ΔITAMCD3δ/ε/γ/ζ

TCRα/β-KIR + ΔITAMCD3δ/ε/γ/ζ

TCRα/β-LIR-1 + ΔITAMCD3δ/ε/γ/ζ

α β

ζ

ε γδ ε

ζ

ε γδ εα β

ζ

ε γδ εα β

ζ

ε γδ εα β

FIGURE 3. Generation of DITAMCD3 Jurkat NFAT luciferase reporter

line for inverter TCR evaluation.

(A) CD3 subunits that lack activating intracellular ITAM domains

(DITAM CD3) were expressed in complete functional knockout clone

32. Each DITAMCD3 subunit was omitted to see if the remaining en-

dogenous subunits would rescue functional activity. The complete

set of wild-type CD3 subunits were coexpressed as a positive con-

trol (black). (B) Inhibitory effects of constructs with PD-1 (green), KIR

(light blue), or LIR-1 ICD (purple) fused to the TCRa/b in clone 32.

Note the maximal inhibition is never complete and somewhat vari-

able. This is at least partly caused by a fraction of activator-only Jur-

kat cells. (C) Surface expression of CD19 CAR and NY-ESO-1 invTCR.




http://ww

w.im

munohorizons.org/

Dow

nloaded from





(73 nM) in the context of the ITAM-shuffled TCR (Fig. 6C,Supplemental Fig. 3C, square) was similar to its IC50 measuredwith the CD19 CAR activator (168 nM) (Fig. 6C, triangle), dem-onstrating that invTCRs inhibit TCRs as they do CARs.

DISCUSSION

Since the invention of CARs, it has been known that ICDs fromCD3 subunits and other activating immune receptors can bestitched together to serve in T cells as signaling components ofmonomeric membrane proteins with Ab-based ligand-bindingdomains (13, 14). These first-, second-, and third-generationCARs convert Ag binding into signals that stimulate T cell prolif-eration and cytotoxicity much like native TCRs (see Ref. 27 forreview). Second- and third-generation CARs have been used inthe clinic as components of cell therapeutics and achieved note-worthy success (28–31). Most engineering for therapeutic pur-poses has been restricted to CARs, as opposed to TCRs, giventhe CAR�s tolerance for alterations (32). TCRs have been typical-ly used in their native form or with minor sequence changesthat involve introduction of stabilizing disulfide bonds or swap-ping of mouse constant domains to facilitate desirable H-L-chainpairing in human cells (33–35). Notable exceptions are: 1) chime-ric TCRs with an scFv fused to the N terminus of the TCR/CD3subunits, creating a TCR/CD3 hybrid receptor that transducesnon-pMHC ligand binding into TCR signaling (36) and 2) single-variable-domain TCRs that function in the context of a TCR-likeCD3 complex to signal (37). These studies demonstrate the resil-ience of the N terminus of the TCR with regard to structuralperturbations. However, they also reveal that the extraordinarysensitivity of the TCR cannot be simply co-opted by grafting adifferent ligand-binding domain onto the structure. The ligand-binding domain itself, rather than the receptor signaling sequen-ces and structures, appears to play a dominant role in dictatingreceptor sensitivity. The observation that TCRs can accommo-date charge-swapped subunits that preserve net charge neutrali-ty, and even a situation in which one transmembrane residue isswitched from negative to positive (CD3g E122K), further sug-gests more plastic structure-function behavior than might be ex-pected from the high conservation of TCR/CD3 primarysequences in the transmembrane segments and ICDs.

We sought to build TCRs with distinct signaling properties:for example, which invert ligand-binding signals that normallyactivate T cell response. We were surprised by how tolerant

Q12.27

Q237.0

Q36.48

Q454.2

101

102

103

104

105

PE-A

101

102

103

104

105

AP

C-A

Q12.17

Q241.6

Q37.31

Q448.9

101

102

103

104

105

PE-A

101

102

103

104

105

AP

C-A

Q10.046

Q20.012

Q32.31

Q497.6

101

102

103

104

105

PE-A

101

102

103

104

105

AP

C-A

CD3E Positive5.78E-3

101

102

103

104

105

PE-A

0

200

400

600

800

Cou

nt

CD3E Positive58.4

101

102

103

104

105

PE-A

0

50

100

150

200

Cou

nt

CD3E Positive56.6

101

102

103

104

105

PE-A

0

50

100

150

Cou

nt

α β

ζ

ε γδ ε α β

ζ

ε γδ ε

AP

C-A

αVβ13.1

A2-

NY

-ES

O-1

αCD3E

TCRα/β + CD3δ/ε-LIR-1 +ΔITAMCD3γ/ζ

A

B

TCRα/β + CD3δ/ε/γ/ζ-LIR-1

α β

ζ

ε γδ εα β


TCRα/β + CD3δ/ε/γ/ζ-LIR-1 scFv-LIR-1

IC50: 6 nM IC50: 54 nM IC50: 9 nM

Ftcr-LIR-1

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 1020

20000

40000

60000

80000

100000


Lum

ines

cenc

e(R

LU) CD19 CAR only



scFv-LIR-1

Ftcr-LIR-1

CD19 CAR only

CD19 CAR + CD3δ/ε-LIR-1

CD19 CAR + CD3δ/ε/γ/ζ-LIR-1

α β

ζ

ε γδ ε



TCRα/β + CD3δ/ε/γ/ζ-LIR-1_3/4

γα β

ζ

ε γδ ε α β

ζ

ε γδ ε α β

ζ

ε δ ε

2

CD19 CAR only

ΔITAM CD3δ/ε/γ/ζ

CD3δ/ε/γ/ζ-LIR-1

CD3δ/ε/γ/ζ-LIR-1_3/4

C

Q15.07

Q229.8

Q30.81

Q464.3

101 102 103 104 105

PE-A

101

102

103

104

105

AP

C-A

Q16.35

Q227.6

Q30.58

Q465.5

101 102 103 104 105

PE-A

101

102

103

104

105

AP

C-A

Q15.35

Q226.3

Q30.55

Q467.8

101

102

103

104

105

PE-A

101

102

103

104

105

AP

C-A

αVβ13.1

A2-

NY

-ES

O-1

210-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 100

50000

100000

150000


Lum

ines

cenc

e(R

LU)

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 1020

20000

40000

60000

80000

100000

Lum

ines

cenc

e(R

LU)

FIGURE 4. Reconstitution of inverter TCR in DITAMCD3 Jurkat NFAT

luciferase cell line.

(A) Inverter CD3 subunits were generated by replacing the CD3 ICD

with the ICD of LIR-1. These inverter CD3 subunits were expressed in

clone 32 along with TCRa/b against A*02:01-NY-ESO-1. To determine

the IC50 of the inverter TCR, they were coexpressed along with activat-

ing CD19 CAR, which reacts with CD19 expressed on the surface of

T2s. (B) The reconstituted inverter TCR IC50 was compared with previ-

ously characterized inhibitory Ftcr- and scFv-LIR-1 fusions against

A*02:01-NY-ESO-1 pMHC (19). Inhibitory activity against CD19 CAR

was very similar between scFv-LIR-1, Ftcr-LIR-1, and invTCR. (C) Two

types of ICDs were appended to inverter CD3 domains, and their IC50s

were compared: LIR-1 ICD (purple) or the region that contains ITIM 3

and 4 in the native CD3 backbone (orange).




http://ww

w.im

munohorizons.org/

Dow

nloaded from



the TCR structure is with regard to C-terminal modification,accommodating a variety of C-terminal fusions, including glob-ular domains and extended sequences. Potential activating do-mains were selected from the set of proteins known toparticipate directly in T cell activation via the TCR, spanningthe main arms of downstream signaling (38). Interestingly,none of the TCR fusions improved acute sensitivity. This sug-gests that recruitment of the proteins tested in our TCR-fusionexperiments, including CD28, is not rate-limiting for TCR acti-vation in vitro. In contrast, it is known that CARs with ITAM-containing ICDs derived from CD3z or FceRIg are not suffi-cient to fully activate proliferation in acute primary human andmouse T cell assays (39–41). The addition of a costimulatorydomain in tandem provides either a quantitatively or qualita-tively different signal that boosts acute response in both prima-ry T and Jurkat cells (41). Notably, our assays were performedin Jurkat cells, which express many, but not all, of the proteinsused to create the fusions. Although Jurkat cells do not containall the components identified through study of normal T cell bi-ology [e.g., PTEN, SYK, SHP-1, CTLA-4 (42)], we have shownthat acute receptor sensitivity measured in Jurkat cells corre-lates with primary T cells (32). Thus, the lack of acute effect inJurkat cells may be explained by the parent molecules� avail-ability for recruitment into the signaling complex. Alternatively,the fusion proteins may not be positioned in an optimal way tofunction in acute signaling. Indeed, it is possible that the fu-sions compromise the function of TCRs in some fashion, whichonly a subset of fusions are able to partially offset.

The result that the acute signaling function of CD3 can betotally replaced by ITAMs on the TCR a/b-chains is remark-able but consistent with data from careful comparisons ofCARs and TCRs that reveal very little difference in responseproperties when corrected for functional sensitivity to Ag. Weshould point out that this statement may be controversial, butwe believe that many studies focus on binding affinity, ratherthan functional sensitivity. We and others have shown thatthese parameters are only weakly correlated in CARs, as inTCRs [(12, 32); see Ref. 43 for review]. Our results are consis-tent with quantitative experiments of James (15) who investi-gated sensitivity in Jurkat cells using a CAR-like scaffold. Heshowed that a single ITAM could mediate some activation,with a big inflection in the response curve at three ITAMs/TCR. His results are extended in this study to include TCRswith different ITAM numbers with respect to sensitivity.

ε

+

- ++

TCRβK288D

NY-ESO-1 Peptide concentration (uM)

Lum

ines

cenc

e(R

LU)

εD137K + γE122R

εD137K + γE122KεD137R + γE122R

wtTCRα + mutTCRβwtTCRαβ

γE122K onlyεD137R + γE122R

γα β

ζ

ε δ ε

βε

γ+ --

K288 K288D/E

E122

D137

- +βε

γ

+

E122K/R

D137K/R

α β

ζ

C

D

B

A

β

γK288D

R253D

E122KD15K

α

-

ζζ

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 1020

50000

100000

150000


Lum

ines

cenc

e(R

LU)

mutTCRα/β + mutCD3γ/ζ-LIR-1_3/4

CD19 CAR only

γ

10 -8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 100

5000

10000

15000

20000

ε δ

2

FIGURE 5. Construction of charge-swapped twin TCRs.

(A) Cartoon representation of engineered TCR transmembrane interac-

tions. In the native TCR–CD3 complex, the positively charged K288 of

TCRb interacts with corresponding negatively charged residues of

CD3e (D137) and CD3g (E122). These charge-pairs were mutated to

K288D and D137K/R and E122K/R to generate distinct TCR complexes.

(B) Jurkat NFAT luciferase activity of charge-swapped TCRb–CD3e/g

complex. In wild-type Jurkat cells, either wild-type TCRa/b or wild-

type TCRa and K288D TCRb were coexpressed with or without corre-

sponding CD3e/g mutants. Rescue of NFAT response was measured.

(C) Cartoon representation of ITIM3/4-substituted transmembrane

domain mutant TCR–CD3 complex. TCRa-CD3z transmembrane mu-

tation (cyan star) and TCRb-CD3g transmembrane mutation (black

star) are described in dotted boxes. (D) Jurkat NFAT luciferase activity

of charge-swapped invTCRa/b–CD3g/z complex. In clone 32 Jurkat

cells, CD19 CAR was expressed with or without charge-swapped TCRb

with ITIM-fused charge-swapped CD3g/z. Inhibitory activity was mea-

sured by coculturing with NY-ESO-1(v) peptide-loaded T2 cells.




http://ww

w.im

munohorizons.org/

Dow

nloaded from


To induce inhibitory rather than activating signals, wesubstituted ITIM sequences for the CD3 ITAM domains. Weshowed that we could create an invTCR with sensitivity towarda CD19 CAR within the range of other potent blocker modulesseen before but not as sensitive as the best activating TCRs(19). Allen and colleagues (44, 45) first described a phenome-non they called altered-peptide ligands (APLs), wherebypMHC agonists could be converted into nonagonists�even an-tagonists�by single amino acid substitutions in the peptide.They subsequently explained this behavior by differences inoff-rates of the TCR/pMHC complexes; faster off-rate variantsproduced lower activation (46). However, the details of theAPL mechanism with respect to signaling are still imperfectlyunderstood (47). Some have postulated that APLs trigger a neg-ative signal from the engaged TCR, but there is evidenceagainst this view (48). We believe that the invTCRs describedin this study are examples of TCRs that directly transformligand binding into a negative signal, mediated via the ITIMmechanism of T cell checkpoint control.

Apart from the focus on in vitro sensitivity and not T cell de-velopment, our study has limitations. We measured acute sensi-tivity of variant TCRs in Jurkat cells, and it is possible that someof the constructs tested in this study (e.g., the fusions to CD28)may provide benefit in long-term contexts for primary T cells, in-cluding cell therapy where exhaustion is thought to limit efficacyin vivo (49). The studies focused on a single TCR, the ultrasensi-tive optimized clinical NY-ESO-1 TCR. It is possible that some ofthe results may not generalize to other TCRs or to more nativesituations in which expression levels are regulated endogenously.Finally, we studied the response of Jurkat cells using the NFATpromotor fusion and did not assess other signaling pathways in-volved in TCR function. In addition, we were not able to achievethe sensitivity level on the inhibitory side with invTCRs, whichwild-type TCRs display routinely for activation.

Ease of engineering has prompted use of CARs in dual-sig-nal integrators for synthetic cellular logic gates (50). The poten-tial of creating two TCRs whose subunits do not comingleusing charge-swapped transmembrane domains may provide anavenue for TCRs to be used in these logic systems. TCRs havethe virtue that they arise in the body with exquisite sensitivityand selectivity against pMHC Ags. The capacity of TCRs to ac-commodate C-terminal fusions offers opportunities to modifyTCR behavior in cis, for example, through inverting specificpMHC signals. Such behavior, if optimized, may prove usefulfor AND NOT signal integration in which the inhibitory signalis derived from pMHCs, such as minor histocompatibility Ags(6, 19). If the results generalize to other TCRs (e.g., to class-II–restricted TCRs), modified TCRs might provide an

α β α β+/-

ζ

ε γδ ε+

A

B

C

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 1020

20000

40000

60000

80000


Lum

ines

cenc

e(R

LU)

wtTCRα/β +ΔITAMCD3δεγζ

wtTCRα/β +wtCD3δεγζ

TCRα/β-ζITAM +ΔITAMCD3δεγζ

KRAS TCRα/β-ζITAM

KRAS TCRα/β-ζITAM +NY-ESO-1 TCRα/β-LIR-1

NY-ESO-1 TCRα/β-LIR-1CD19 CAR +

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 1020

20000

40000

60000

80000

100000


Lum

ines

cenc

e(R

LU)

KRAS wtTCRα/β

Q111.1

Q20.094

Q35.20E-3

Q488.8

101

102

103

104

105

Comp-PE-A

101

102

103

104

105

Com

p-A

PC

-A

Q111.9

Q20.093

Q35.18E-3

Q488.0

101

102

103

104

105

Comp-PE-A

101

102

103

104

105

Com

p-A

PC

-A

Q10.39

Q23.41

Q313.1

Q483.1

101

102

103

104

105

Comp-PE-A

101

102

103

104

105

Com

p-A

PC

-A

Q10.33

Q20.48

Q318.6

Q480.6

101

102

103

104

105

Comp-PE-A

101

102

103

104

105

Com

p-A

PC

-A

A2-NY-ESO-1

A11

-KR

AS

KRASwtTCRα/β

KRASTCRα/β-ζITAM

KRASTCRα/β-ζITAM

+ +NY-ESO-1

TCRα/β-LIR-1

CD19 CAR

NY-ESO-1TCRα/β-LIR-1

KRASTCRα/β-ζITAM

NY-ESO-1TCRα/β-LIR-1 ΔITAMCD3δεγζ

FIGURE 6. Characterization of C-terminally ITAM- or ITIM-fused

TCRa/b.

(A) Cartoon representation of ITAM-shuffled KRAS TCRa/b (KRAS TCRa/

b-zITAM) expressed with or without ITIM-shuffled TCRa/b (NY-ESO-1

TCRa/b-zITIM) in cells in the presence of DITAMCD3 subunits. (B) NFAT

luciferase response of wild-type TCRa/b coexpressed with either

DITAMCD3 (open circle) or wild-type CD3 (closed circle) and ITAM-

shuffled TCRa/b coexpressed with DITAMCD3 subunits (triangle) in Jur-

kat cells. (C) IC50 curves of ITIM-shuffled NY-ESO-1 TCRa/b against

either CD19 CAR or ITAM-shuffled KRAS TCRa/b. To test ITIM-shuf-

fled NY-ESO-1 TCRa/b activity against ITAM-shuffled KRAS

TCRa/b, transfected Jurkat cells were cocultured with T2s expressing

both A*02:01 and A*11:01 loaded with 1 mM KRAS G12D peptide and ti-

trating amounts of NY-ESO-1 peptide.




http://ww

w.im

munohorizons.org/

Dow

nloaded from


alternative means to modulate response to specific Ags. For ex-ample, invTCRs might be used to inhibit reactivity toward spe-cific self-antigens in a subset of a patient�s T cells in an Ag-dependent manner without eliminating portions of the TCRrepertoire. From a practical perspective, the invTCR will likelyneed further development for use in primary T cells becausethe presence of some ITAMs in the complex (i.e., from wild-type CD3z and CD3g) prevents signal inversion (i.e., creates anactivator; Supplemental Fig. 2D). With CAR activators, thisproblem can be eliminated by removal of wild-type CD3 ex-pression in the host cell. If invTCRs are paired with TCRs asopposed to CARs, proper subunit segregation will need to beenforced, for example, by improved specificity of the charge--pair interactions described in this article to explore invTCRfunction.

In conclusion, by exploring C-terminal and transmembranevariants of the TCR, we have uncovered a surprising degree ofstructure-function flexibility. Specifically, we have shown that1) the C termini of the TCR can be fused to additional sequen-ces, preserving acute function; 2) the ITAM domains can beeliminated from all CD3 subunits and fused to the TCRa/bsubunits to create a receptor that is very similar to the wild-type TCR in acute assays; 3) ITAMs dominate ITIMs if theyare present in the same TCR/CD3 complex, but if the ITAMsare replaced by ITIMs, signaling is inverted and the TCR be-comes a ligand-gated inhibitory receptor; and 4) conserved,charged residues in the transmembrane domains of the TCR/CD3complex can be swapped to create functional TCRs, a further in-dication of the robust mechanism that operates in acute TCR sig-naling. These results suggest opportunities for design of TCR-based receptors that extend well beyond those that have beenused in T cell therapeutics to date.

DISCLOSURES

The authors are members of a company (A2 Biotherapeutics)focused on the cell therapy modality for cancer.

ACKNOWLEDGMENTS

We are grateful to Dr. Mark Davis for discussion and Bella Lee, Brean-na Luna, and Shrimika Madhavan for technical assistance.

REFERENCES

1. Manz, B. N., B. L. Jackson, R. S. Petit, M. L. Dustin, and J. Groves.2011. T-cell triggering thresholds are modulated by the number ofantigen within individual T-cell receptor clusters. Proc. Natl. Acad.Sci. USA 108: 9089–9094.

2. Irvine, D. J., M. A. Purbhoo, M. Krogsgaard, and M. M. Davis. 2002. Di-rect observation of ligand recognition by T cells. Nature 419: 845–849.

3. Courtney, A. H., W. L. Lo, and A. Weiss. 2018. TCR signaling: mecha-nisms of initiation and propagation. Trends Biochem. Sci. 43: 108–123.

4. Borroto, A., D. Abia, and B. Alarc�on. 2014. Crammed signaling mo-tifs in the T-cell receptor. Immunol. Lett. 161: 113–117.

5. Gouaillard, C., A. Huchenq-Champagne, J. Arnaud, C. L. Chen Cl,and B. Rubin. 2001. Evolution of T cell receptor (TCR) alpha betaheterodimer assembly with the CD3 complex. Eur. J. Immunol. 31:3798–3805.

6. Hwang, J. R., Y. Byeon, D. Kim, and S. G. Park. 2020. Recent in-sights of T cell receptor-mediated signaling pathways for T cell acti-vation and development. Exp. Mol. Med. 52: 750–761.

7. Rozdzial, M. M., B. Malissen, and T. H. Finkel. 1995. Tyrosine-phos-phorylated T cell receptor zeta chain associates with the actin cytoskele-ton upon activation of mature T lymphocytes. Immunity 3: 623–633.

8. Haks, M. C., T. A. Cordaro, J. H. van den Brakel, J. B. Haanen, E. F.de Vries, J. Borst, P. Krimpenfort, and A. M. Kruisbeek. 2001. A re-dundant role of the CD3 gamma-immunoreceptor tyrosine-based ac-tivation motif in mature T cell function. J. Immunol. 166: 2576–2588.

9. Haks, M. C., E. P�epin, J. H. van den Brakel, S. A. Smeele, S. M. Bel-kowski, H. W. Kessels, P. Krimpenfort, and A. M. Kruisbeek. 2002.Contributions of the T cell receptor-associated CD3gamma-ITAMto thymocyte selection. J. Exp. Med. 196: 1–13.

10. Shores, E., V. Flamand, T. Tran, A. Grinberg, J. P. Kinet, and P. E.Love. 1997. Fc epsilonRI gamma can support T cell developmentand function in mice lacking endogenous TCR zeta-chain. J. Immu-nol. 159: 222–230.

11. Holst, J., H. Wang, K. D. Eder, C. J. Workman, K. L. Boyd, Z. Baquet,H. Singh, K. Forbes, A. Chruscinski, R. Smeyne, et al. 2008. Scalablesignaling mediated by T cell antigen receptor-CD3 ITAMs ensureseffective negative selection and prevents autoimmunity. Nat. Immu-nol. 9: 658–666.

12. Gudipati, V., J. Rydzek, I. Doel-Perez, V. D. R. Goncalves, L. Scharf,S. K€onigsberger, E. Lobner, R. Kunert, H. Einsele, H. Stockinger,et al. 2020. Inefficient CAR-proximal signaling blunts antigen sensi-tivity. Nat. Immunol. 21: 848–856.

13. Gross, G., T. Waks, and Z. Eshhar. 1989. Expression of immunoglobu-lin-T-cell receptor chimeric molecules as functional receptors withantibody-type specificity. Proc. Natl. Acad. Sci. USA 86: 10024–10028.

14. Roberts, M. R., L. Qin, D. Zhang, D. H. Smith, A. C. Tran, T. J. Dull,J. E. Groopman, D. J. Capon, R. A. Byrn, and M. H. Finer. 1994. Tar-geting of human immunodeficiency virus-infected cells by CD81 Tlymphocytes armed with universal T-cell receptors. Blood 84:2878–2889.

15. James, J. R. 2018. Tuning ITAMmultiplicity on T cell receptors can con-trol potency and selectivity to ligand density. Sci. Signal. 11: eaan1088

16. Jubel, J. M., Z. R. Barbati, C. Burger, D. C. Wirtz, and F. A. Schild-berg. 2020. The role of PD-1 in acute and chronic infection. Front.Immunol. 11: 487.

17. Getahun, A., and J. C. Cambier. 2015. Of ITIMs, ITAMs, and ITA-Mis: revisiting immunoglobulin Fc receptor signaling. Immunol. Rev.268: 66–73.

18. Fedorov, V. D., M. Themeli, and M. Sadelain. 2013. PD-1- andCTLA-4-based inhibitory chimeric antigen receptors (iCARs) divertoff-target immunotherapy responses. Sci. Transl. Med. 5: 215ra172.

19. Hamburger, A. E., B. DiAndreth, J. Cui, M. E. Daris, M. L. Munguia,K. Deshmukh, J. Y. Mock, G. E. Asuelime, E. D. Lim, M. R. Kreke,et al. 2020. Engineered T cells directed at tumors with defined alle-lic loss. Mol. Immunol. 128: 298–310.

20. Robbins, P. F., Y. F. Li, M. El-Gamil, Y. Zhao, J. A. Wargo, Z. Zheng,H. Xu, R. A. Morgan, S. A. Feldman, L. A. Johnson, et al. 2008. Sin-gle and dual amino acid substitutions in TCR CDRs can enhance an-tigen-specific T cell functions. J. Immunol. 180: 6116–6131.

21. Hendel, A., R. O. Bak, J. T. Clark, A. B. Kennedy, D. E. Ryan, S. Roy,I. Steinfeld, B. D. Lunstad, R. J. Kaiser, A. B. Wilkens, et al. 2015.Chemically modified guide RNAs enhance CRISPR-Cas genome ed-iting in human primary cells. Nat. Biotechnol. 33: 985–989.

22. Rapoport, A. P., E. A. Stadtmauer, G. K. Binder-Scholl, O. Goloube-va, D. T. Vogl, S. F. Lacey, A. Z. Badros, A. Garfall, B. Weiss, J. Fin-klestein, et al. 2015. NY-ESO-1-specific TCR-engineered T cells




http://ww

w.im

munohorizons.org/

Dow

nloaded from



mediate sustained antigen-specific antitumor effects in myeloma.Nat. Med. 21: 914–921.

23. Borges, L., M. L. Hsu, N. Fanger, M. Kubin, and D. Cosman. 1997. Afamily of human lymphoid and myeloid Ig-like receptors, some ofwhich bind to MHC class I molecules. J. Immunol. 159: 5192–5196.

24. Bell�on, T., F. Kitzig, J. Say�os, and M. L�opez-Botet. 2002. Mutationalanalysis of immunoreceptor tyrosine-based inhibition motifs of the Ig-like transcript 2 (CD85j) leukocyte receptor. J. Immunol. 168: 3351–3359.

25. Dong, D., L. Zheng, J. Lin, B. Zhang, Y. Zhu, N. Li, S. Xie, Y. Wang,N. Gao, and Z. Huang. 2019. Structural basis of assembly of the hu-man T cell receptor-CD3 complex. [Published erratum appears in2021 Nature 590: E52.] Nature 573: 546–552.

26. Call, M. E., J. Pyrdol, M. Wiedmann, and K. W. Wucherpfennig.2002. The organizing principle in the formation of the T cell recep-tor-CD3 complex. Cell 111: 967–979.

27. Meng, X., R. Jing, L. Qian, C. Zhou, and J. Sun. 2020. Engineeringcytoplasmic signaling of CD28z CARs for improved therapeuticfunctions. Front. Immunol. 11: 1046.

28. Neelapu, S. S., C. A. Jacobson, O. O. Oluwole, J. Munoz, A. Deol,D. B. Miklos, N. L. Bartlett, I. Braunschweig, Y. Jiang, J. J. Kim,et al. 2020. Outcomes of older patients in ZUMA-1, a pivotal studyof axicabtagene ciloleucel in refractory large B-cell lymphoma.Blood 135: 2106–2109.

29. Neelapu, S. S., F. L. Locke, N. L. Bartlett, L. J. Lekakis, D. B. Miklos,C. A. Jacobson, I. Braunschweig, O. O. Oluwole, T. Siddiqi, Y. Lin,et al. 2017. Axicabtagene ciloleucel CAR T-cell therapy in refractorylarge B-cell lymphoma. N. Engl. J. Med. 377: 2531–2544.

30. Schuster, S. J., M. R. Bishop, C. S. Tam, E. K. Waller, P. Borchmann,J. P. McGuirk, U. J€ager, S. Jaglowski, C. Andreadis, J. R. Westin,et al. 2019. Tisagenlecleucel in adult relapsed or refractory diffuselarge B-cell lymphoma. N. Engl. J. Med. 380: 45–56.

31. Abramson, J. S. 2020. Anti-CD19 CAR T-cell therapy for B-cell non-Hodgkin lymphoma. Transfus. Med. Rev. 34: 29–33.

32. Xu, H., A. E. Hamburger, J. Y. Mock, X. Wang, A. D. Martin, T. To-katlian, J. Oh, M. E. Daris, K. R. Negri, G. B. Gabrelow, et al. 2020.Structure-function relationships of chimeric antigen receptors inacute T cell responses to antigen. Mol. Immunol. 126: 56–64.

33. Parkhurst, M. R., J. C. Yang, R. C. Langan, M. E. Dudley, D. A. Na-than, S. A. Feldman, J. L. Davis, R. A. Morgan, M. J. Merino, R. M.Sherry, et al. 2011. T cells targeting carcinoembryonic antigen canmediate regression of metastatic colorectal cancer but induce severetransient colitis. Mol. Ther. 19: 620–626.

34. Morgan, R. A., N. Chinnasamy, D. Abate-Daga, A. Gros, P. F. Rob-bins, Z. Zheng, M. E. Dudley, S. A. Feldman, J. C. Yang, R. M. Sher-ry, et al. 2013. Cancer regression and neurological toxicity followinganti-MAGE-A3 TCR gene therapy. J. Immunother. 36: 133–151.

35. Jin, B. Y., T. E. Campbell, L. M. Draper, S. Stevanovi�c, B. Weissbrich,Z. Yu, N. P. Restifo, S. A. Rosenberg, C. L. Trimble, and C. S. Hinrichs.

2018. Engineered T cells targeting E7 mediate regression of humanpapillomavirus cancers in a murine model. JCI Insight 3: e99488

36. Baeuerle, P. A., J. Ding, E. Patel, N. Thorausch, H. Horton, J. Gierut,I. Scarfo, R. Choudhary, O. Kiner, J. Krishnamurthy, et al. 2019. Syn-thetic TRuC receptors engaging the complete T cell receptor for po-tent anti-tumor response. Nat. Commun. 10: 2087.

37. Oh, J., D. T. Warshaviak, M. Mkrtichyan, M. L. Munguia, A. Lin, F.Chai, C. Pigott, J. Kang, M. Gallo, and A. Kamb. 2019. Single variabledomains from the T cell receptor b chain function as mono- and bi-functional CARs and TCRs. Sci. Rep. 9: 17291.

38. Cantrell, D. A. 2002. T-cell antigen receptor signal transduction. Im-munology 105: 369–374.

39. Brocker, T., and K. Karjalainen. 1995. Signals through T cell recep-tor-zeta chain alone are insufficient to prime resting T lymphocytes.J. Exp. Med. 181: 1653–1659.

40. Brocker, T. 2000. Chimeric Fv-zeta or Fv-epsilon receptors are notsufficient to induce activation or cytokine production in peripheralT cells. Blood 96: 1999–2001.

41. Finney, H. M., A. N. Akbar, and A. D. Lawson. 2004. Activation ofresting human primary T cells with chimeric receptors: costimula-tion from CD28, inducible costimulator, CD134, and CD137 in serieswith signals from the TCR zeta chain. J. Immunol. 172: 104–113.

42. Gioia, L., A. Siddique, S. R. Head, D. R. Salomon, and A. I. Su. 2018.A genome-wide survey of mutations in the Jurkat cell line. BMC Ge-nomics 19: 334.

43. Wu, L., Q. Wei, J. Brzostek, and N. R. J. Gascoigne. 2020. Signalingfrom T cell receptors (TCRs) and chimeric antigen receptors(CARs) on T cells. Cell. Mol. Immunol. 17: 600–612.

44. Evavold, B. D., and P. M. Allen. 1991. Separation of IL-4 productionfrom Th cell proliferation by an altered T cell receptor ligand. Sci-ence 252: 1308–1310.

45. Evavold, B. D., J. Sloan-Lancaster, and P. M. Allen. 1994. Antagonismof superantigen-stimulated helper T-cell clones and hybridomas byaltered peptide ligand. Proc. Natl. Acad. Sci. USA 91: 2300–2304.

46. Kersh, G. J., and P. M. Allen. 1996. Structural basis for T cell recognitionof altered peptide ligands: a single T cell receptor can productively rec-ognize a large continuum of related ligands. J. Exp. Med. 184: 1259–1268.

47. Candia, M., B. Kratzer, and W. F. Pickl. 2016. On peptides and alteredpeptide ligands: from origin, mode of action and design to clinical ap-plication (immunotherapy). Int. Arch. Allergy Immunol. 170: 211–233.

48. Stotz, S. H., L. Bolliger, F. R. Carbone, and E. Palmer. 1999. T cell recep-tor (TCR) antagonism without a negative signal: evidence from T cell hy-bridomas expressing two independent TCRs. J. Exp. Med. 189: 253–264.

49. Gaissmaier, L., M. Elshiaty, and P. Christopoulos. 2020. Breakingbottlenecks for the TCR therapy of cancer. Cells 9: 2095

50. Dannenfelser, R., G. M. Allen, B. VanderSluis, A. K. Koegel, S. Levin-son, S. R. Stark, V. Yao, A. Tadych, O. G. Troyanskaya, and W. A.Lim. 2020. Discriminatory power of combinatorial antigen recogni-tion in cancer T cell therapies. Cell Syst. 11: 215–228.e5.




http://ww

w.im

munohorizons.org/

Dow

nloaded from


design of tcr structural variants that retain or invert the … · tcr signaling (18, 19). in this...

Documents