Phosphoantigen-induced conformational change ofbutyrophilin 3A1 (BTN3A1) and its implication onVγ9Vδ2 T cell activationSiyi Gua, Joseph R. Sachlebena,b, Christopher T. Boughterc, Wioletta I. Nawrockaa, Marta T. Borowskaa,Jeffrey T. Tarraschd,e, Georgios Skiniotisd,e,1, Benoît Rouxa,c, and Erin J. Adamsa,f,g,2

aDepartment of Biochemistry and Molecular Biophysics, University of Chicago, Chicago, IL 60637; bBiomolecular NMR Facility, University of Chicago,Chicago, IL 60637; cGraduate Program in Biophysical Sciences, University of Chicago, Chicago, IL 60637; dLife Sciences Institute, University of MichiganMedical School, Ann Arbor, MI 48109; eDepartment of Biological Chemistry, University of Michigan Medical School, Ann Arbor, MI 48109; fCommittee onImmunology, University of Chicago, Chicago, IL 60637; and gCommittee on Cancer Biology, University of Chicago, Chicago, IL 60637

Edited by Philippa Marrack, Howard Hughes Medical Institute, National Jewish Health, Denver, CO, and approved July 21, 2017 (received for review May6, 2017)

Human Vγ9Vδ2 T cells respond to microbial infections as well ascertain types of tumors. The key initiators of Vγ9Vδ2 activation aresmall, pyrophosphate-containing molecules called phosphoanti-gens (pAgs) that are present in infected cells or accumulate intra-cellularly in certain tumor cells. Recent studies demonstrate thatinitiation of the Vγ9Vδ2 T cell response begins with sensing of pAgvia the intracellular domain of the butyrophilin 3A1 (BTN3A1) mol-ecule. However, it is unknown how downstream events can ulti-mately lead to T cell activation. Here, using NMR spectrometry andmolecular dynamics (MD) simulations, we characterize a globalconformational change in the B30.2 intracellular domain ofBTN3A1 induced by pAg binding. We also reveal by crystallogra-phy two distinct dimer interfaces in the BTN3A1 full-length intra-cellular domain, which are stable in MD simulations. Theseinterfaces lie in close proximity to the pAg-binding pocket andcontain clusters of residues that experience major changes ofchemical environment upon pAg binding. This suggests that pAg bind-ing disrupts a preexisting conformation of the BTN3A1 intracellulardomain. Using a combination of biochemical, structural, and cellu-lar approaches we demonstrate that the extracellular domains ofBTN3A1 adopt a V-shaped conformation at rest, and that lockingthem in this resting conformation without perturbing their mem-brane reorganization properties diminishes pAg-induced T cell ac-tivation. Based on these results, we propose a model in which aconformational change in BTN3A1 is a key event of pAg sensingthat ultimately leads to T cell activation.

human Vγ9Vδ2 T cells | butyrophilin 3A1 | phosphoantigen |conformational change

Gamma delta (γδ) T cells exhibit restricted T cell receptor(TCR) use and tissue-specific residence, and they play critical

roles in immune responses to infection and tumorigenesis. Inhuman peripheral blood, a major γδ T cell subset is calledVγ9Vδ2, based on the variable gene segments (TRGV9 andTRDV2) used in the recombined TCR. This population can beactivated and expanded to about 50% of total blood T cells uponinfection with certain microbes, including the bacterial pathogensMycobacterium tuberculosis and Mycobacterium leprae and pro-tozoal parasites such as Plasmodium falciparum (1). Vγ9Vδ2T cells can quickly launch specific and effective immune re-sponses toward these pathogens (1). Moreover, they show sim-ilar reactivity toward certain types of cancer cells, resulting inwidening and intensifying interest in the use of gamma deltaT cells for cancer immunotherapy (2). However, despite inten-sive research over the past 30 y, the molecular mechanismsgoverning Vγ9Vδ2 T cells’ recognition of infected and malignantcells are still poorly understood, thus impeding the overall un-derstanding of Vγ9Vδ2 T cell immunity and development of itspotential medical applications.

Vγ9Vδ2 T cells are specifically activated by a set of pyro-phosphate metabolites collectively named phosphoantigens(pAgs), which are present in both infected and malignant targetcells (3). These pAgs are sensed by the butyrophilin 3A1(BTN3A1) protein, a member of the BTN3A family with threedifferent isoforms (A1, A2, and A3) that confer pAg-mediatedreactivity toward target cells by Vγ9Vδ2 T cells (4). Unrelated toMHC molecules, BTN3A proteins are type-I membrane proteinswith two Ig-like extracellular domains with structural homologyto the B7 superfamily of proteins (5). The antibody 20.1, specificto the BTN3A extracellular domains, is capable of activatingVγ9Vδ2 T cells even in the absence of pAgs (4, 5). Previousstructural studies on the BTN3A Ig-like extracellular domainsand their complex with 20.1 showed two possible conformationsof extracellular domains: a “V-shaped” form, which is compati-ble with 20.1 binding and has the potential to oligomerize, and a“head-to-tail” form, of which the dimer interface overlaps withthe 20.1 binding site (6). However, it is unknown whether thesetwo dimer forms exist in the full-length BTN3A molecule in thecellular environment, and whether they play a role in pAg-induced T cell activation.While it is still unclear how the extracellular domains of BTN3A

contribute to T cell activation, the intracellular B30.2 domain ofBTN3A1 has been proven to play a critical role in pAg detection

Significance

Gamma delta T cells, a group of immune cells that exhibitfeatures from both innate and adaptive immunity, possesssignificant potential in clinical applications such as treatment ofmicrobial infections and cancer immunotherapy. To fully un-derstand their biology and harness them in the clinic it is im-perative to dissect the molecular mechanisms involved in theirrecognition of infected and tumor cells. In this paper we focuson Vγ9Vδ2 T cells, a major subset of human gamma delta T cellsin blood and investigate the phosphoantigen-induced, MHC-independent molecular mechanisms governing their activation.

Author contributions: S.G. and E.J.A. designed research; S.G., J.R.S., C.T.B., W.I.N., M.T.B.,and J.T.T. performed research; S.G., J.R.S., C.T.B., W.I.N., M.T.B., J.T.T., G.S., and B.R. an-alyzed data; S.G., C.T.B., and E.J.A. wrote the paper; and G.S. and B.R. providedresearch resources.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

Data deposition: The atomic coordinates and structure factors have been deposited in theProtein Data Bank, www.wwpdb.org (PDB ID code 5HM7).1Present address: Department of Molecular and Cellular Physiology and Department ofStructural Biology, Stanford University School of Medicine, Stanford, CA 94305.

2To whom correspondence should be addressed. Email: ejadams@uchicago.edu.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1707547114/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1707547114 PNAS | Published online August 14, 2017 | E7311–E7320

IMMUNOLO

GYAND

INFLAMMATION

PNASPL

US

Dow

nloa

ded

by g

uest

on

Apr

il 4,

202

0

(4, 7). pAgs bind directly to a positively charged pocket in the in-tracellular B30.2 domain of BTN3A1 (8, 9). Other proteins im-portant for pAg-induced T cell activation, such as RhoB GTPaseand periplakin, are also reported to interact with the intracellulardomain (10, 11). Moreover, the BTN3A1 full-length intracellulardomain (BFI), including the membrane proximal region locatedN-terminal to the B30.2 domain, undergoes a conformationalchange upon pAg binding (9). However, it is unknown how ex-actly pAg binding triggers a conformational change of BFI andhow this ultimately leads to Vγ9Vδ2 TCR engagement and T cellstimulation.Here we present structural, biophysical, computational, and

functional data dissecting the pAg-induced conformational changeof the intracellular domain of BTN3A1. Using NMR spectrometryand molecular dynamics (MD) simulations, we show that theBTN3A1 B30.2 domain undergoes a global conformational changeupon pAg binding. We also reveal two distinct dimer interfacesof the BFI domain through crystallography. Mapping residueswith significant chemical shift perturbation (CSP), obtained byNMR, onto the crystal structure of BFI reveals changes acrossthe B30.2 domain, many of which are located in the dimer in-terfaces. Together with additional supporting data from MDsimulations, we propose that the binding of pAg induces changesin the dimer interface of the intracellular domain that can po-tentially propagate to the extracellular domain of BTN3A1.Combining approaches such as EM, cross-linking, and functionalassays, we then demonstrate that the extracellular domains ofBTN3A1 adopt a V-shaped conformation at rest. We furtherfound that locking the extracellular domains in this restingconformation without perturbing their membrane reorganizationproperties diminishes pAg-induced T cell activation, suggestingthat rearrangement of BTN3A1 proteins is critical to Vγ9Vδ2T cell activation. Altogether, our data strongly support a modelin which pAg-triggered conformational change of BTN3A1 is anessential molecular event leading to Vγ9Vδ2 T cell activation.

ResultspAg Induces a Global Conformational Change of the BTN3A1 IntracellularB30.2 Domain.Previous biophysical and structural studies have shownthat pAgs bind directly to the BTN3A1 intracellular B30.2 domain(8, 9). In our attempts to obtain B30.2–pAg complex crystalsthrough ligand soaking we observed that B30.2 apo crystals dissolveupon pAg addition, hinting that pAg binding may cause confor-mational changes of the B30.2 domain which disrupt crystal packing(8). To explore this possibility, we applied NMR techniques to theB30.2–pAg complex to reveal detailed structural information aboutthe B30.2 domain upon pAg binding. Major CSP from multipleresidues within the B30.2 domain were observed comparing the 15N,1H- heteronuclear single quantum coherence (HSQC) spectra of15N-single labeled B30.2 with increasing concentrations of 1-hydroxy-2-methylpent-2-enyl-pyrophosphonate (cHDMAPP), kindly pro-vided by C. Belmant, Innate Pharma, Marseille, France, a syntheticparalog to the naturally occurring microbial pAg 1-hydroxy-2-methylpent-2-enyl-pyrophosphate (HMBPP) (Fig. 1A). The major-ity of these peaks exhibit fast-exchange behavior, whereas some peaksshow intermediate to slow exchange behavior, suggesting a potentialconformational change in addition to the ligand binding event. Tounderstand where these CSPs reside on the 3D structure of theB30.2 domain, we performed 3D NMR spectroscopy for backboneresonance assignments of both B30.2 with and without pAg,using 2H, 15N, 13C triple-labeled B30.2 protein sample. About80% of the peaks appearing in the HSQC spectra were assigned,and the assigned residues covered almost the entire structure ofB30.2 (Figs. S1 and S2). For residues with assignment in bothapo and complex forms, we calculated the weighted average CSPof 15N and 1H between the free and pAg-bound B30.2 domain(Fig. 1B). By mapping the residues undergoing large CSP (higherthan SD) onto the crystal structure of B30.2 domain [Protein Data

Bank (PDB) ID code 4N7I] we found that many of these residuesare located near the pAg binding pocket, likely due to changes ofthe chemical environment in the presence of pAg (Fig. 1C). In-terestingly, there were several residues residing in remote sitesaway from the binding pocket, indicating that there is a non-local conformational change of B30.2 domain upon pAg bind-ing (Fig. 1C). Notably, some residues are either only assigned inapo or in pAg-bound form, likely due to the line-broadeningeffect from intermediate or slow chemical exchange (Fig. S2).

Y352

1H (ppm)

15N

(ppm

)

105

110

115

120

125

135

130

7.08.09.010.0 6.0

00.250.50.7512cHDMAPP : B30.2:

H351

L434

T402/413

Y352

90̊

180̊

M394

H351 M394

T402

T413 L434

Residue

CSP

CSP2XStd.1XStd.

1.21.00.80.60.4

0.00.2

301

311

321

331

341

351

361

371

381

391

401

411

421

431

441

451

461

471

481

A

B

C

Fig. 1. (Upper) Binding of pAg cHDMAPP induces a conformational changein the BTN3A1 B30.2 domain. (A) 15N, 1H- HSQC spectra of 15N single-labeledB30.2 domain with increasing concentrations of cHDMAPP. The ligand-to-protein ratio is color-coded as indicated above the spectra. Representativeenlarged views for six residues experiencing major CSP are also shown(Lower). (B) Histogram of weighted average CSP of 15N and 1H between freeand ligand-bound (1:1) B30.2 domain. The SD and twofold SD are shown asgray lines on the histogram. (C) CSP mapping onto the B30.2 domainstructure (PDB ID code 4N7I). The B30.2 domain is shown in yellow surfacerepresentation and cHDMAPP is shown in cyan stick representation in thepAg-binding pocket. The residues with major CSP located near thecHDMAPP-binding pocket (Left) or away from the ligand-binding pocket(Right, rotated 90° or 180° respectively from the left view) are shown in red.

E7312 | www.pnas.org/cgi/doi/10.1073/pnas.1707547114 Gu et al.

Dow

nloa

ded

by g

uest

on

Apr

il 4,

202

0

Interestingly, these residues reside mostly away from the pAg-binding pocket, again suggesting a conformational change distantfrom the pAg-binding pocket.To further investigate the putative pAg-induced conforma-

tional change, we performed all-atom MD simulations of theBTN3A1 intracellular domain in the presence or absence of pAgcHDMAPP, starting from the crystal structure model. Rmsdbackbone analysis of these simulations, which reveal structuralshifts in the protein backbone over time, show that overall thecrystal structure is stable in solution, with a mean deviation be-tween crystal structure and simulation of 1.81 Å2 for the aposystems and 1.57 Å2 for the pAg-bound systems. Notably, single-residue rmsd analysis of the backbone showed that the sameresidues that were identified in the CSP data accounted for themajority of this variation (Fig. 2A). The specific residues showingthe most prominent shifts in silico were along two distinct flex-ible loops encompassing residues 393–397 and residues 410–419.Visualization of these trajectories show that pAg keeps thebinding pocket of the intracellular domain in a more contractedstate, while the apo systems are free to adopt a more open,flexible conformation which gives atomic-level insight in to thepAg proximal chemical shifts outlined above (Fig. 2B). Together,the NMR and computational data both strongly support the ideaof a global conformational change of the B30.2 domain triggeredby pAg binding.

Tyrosine 352 Is Critical for pAg Binding to B30.2 Domain. Among allof the residues with major CSP in the B30.2 domain, tyrosine 352(Y352) showed the largest perturbation (Fig. 1B). Because of itslocation near the pAg-binding pocket, we postulated that itmight be important for pAg binding. To test this hypothesis, weused isothermal titration calorimetry (ITC) to measure thebinding between the B30.2 alanine mutant of Y352 (Y352A) andboth cHDMAPP and the weaker-potency, native pAg iso-pentenyl pyrophosphate (IPP). Previous ITC measurements havemeasured the affinities (dissociation constant, Kd) of these pAgsfor the B30.2 domain to be ∼1 μM and ∼1 mM, respectively (8).Compared with WT B30.2 protein, the mutant Y352A showed asignificant defect in binding to both pAg species. In the case ofcHDMAPP, the Kd for the protein–ligand interaction exhibited asixfold increase (weaker affinity) whereas the mutant interactionwith IPP was undetectable (Fig. 3A and Table S1). To determineif the Y352A mutant had similar effects on pAg-mediatedVγ9Vδ2 T cell activation, we transfected either full-length WTBTN3A1 or the Y352A mutant into HEK293 cells with theBTN3A genes (A1, A2, and A3) knocked out via CRISPR-Cas9 targeting. We then stimulated the Jurkat cells transducedwith Vγ9Vδ2 G115 TCR by coculturing them with BTN3A1-transfected cells that were pretreated with either the microbialpAg HMBPP (the naturally occurring form of cHDMAPP) orthe aminobisphosphate (NBP) pamidronate that causes in-tracellular IPP accumulation through inhibition of the enzymefarnesyl diphosphate synthase. Consistent with our biophysicalmeasurements, the HMBPP-induced T cell activation was di-minished for the Y352A mutant, but the defect was not as sig-nificant as the NBP-induced IPP activation, which was severelyaffected (Fig. 3B). The 20.1 antibody (20.1 mAb)-elicited Vγ9Vδ2response for the Y352A mutant was reduced compared withWT, but the difference was not statistically significant (Fig. 3B).Together, these results indicate that residue Y352 in the B30.2domain is critical for pAg binding.

The 1.9-Å Structure of BTN3A1 BFI Reveals Two Potential DimerInterfaces. To obtain a more complete view of the pAg-inducedconformational change we performed crystallization of the BFIdomain with cHDMAPP soaking, in light of recent works fromHsiao et al. (9) and Rhodes et al. (11) suggesting the importanceof the membrane proximal region of BTN3A1 intracellular do-

0

1

2

3

4

5

Apo

324

351

352

353

393

394

395

397

402

405

406

407

410

411

412

413

414

415

416

419

434

Residue

0

1

2

3

4

5

RM

SD (Å

2)

pAg Bound

Apo

pAg Bound

pAgcHDMAPP

324

351

352

353

393

394

395

397

402

405

406

407

410

411

412

413

414

415

416

419

434

Residue

RM

SD (Å

2)

A

B

Fig. 2. All-atom MD simulations reveal dynamics and structural differencesbetween B30.2 domain apo and pAg-bound state. (A) Single amino acidbackbone rmsd of the residues identified to experience high CSP upon pAgbinding. The thick horizontal line in each plot is the mean plus or minus SDof the backbone rmsd of the rest of the protein. (B) Coordinates of theBTN3A1 intracellular domain with (blue) or without (red) pAg cHDMAPP(shown in cyan stick) after 100 ns of MD simulations are shown in cartoonrepresentation. The segments 393–397 and 410–419 that exhibit majorbackbone RMSD shift are shown as ribbon representation in red (apo) orblue (pAg-bound).

Gu et al. PNAS | Published online August 14, 2017 | E7313

IMMUNOLO

GYAND

INFLAMMATION

PNASPL

US

Dow

nloa

ded

by g

uest

on

Apr

il 4,

202

0

main. Crystals diffracted to 1.9 Å and the structure was refinedto an Rwork/Rfree of 20.5%/23.4% (Fig. 4A and Table S2). Un-fortunately, the electron density for cHDMAPP was too weak tobe interpreted unambiguously. However, the BFI crystalized in adifferent space group with two molecules in the asymmetric unitrather than one compared with the B30.2 domain structure pub-lished previously (8). The dimer interface in the asymmetric unit,referred to as Dimer I hereafter, involves more than 20 residuesfrom both monomers (Fig. 4B, Upper). Among these, the side

chain hydroxyl group of Y352, the residue that undergoes thelargest CSP and is critical for pAg binding and T cell activation(Figs. 1B and 3), is one of the major contributors to this dimerinterface, making hydrogen bonds with the backbone of E327 andD328 from the other monomer (Fig. 4B, Upper). Other importantinterface contacts are hydrophobic and van der Waals interactionsmade by W350, W391, and L471 from one monomer with a clusterof residues from the other monomer (Fig. 4B, Upper).In addition to the Dimer I interface described above an ad-

ditional dimer was identified, similar in orientation to that notedin the original B30.2 structure (PDB ID code 4N7I). In the BFIstructure, the symmetrical N-terminal α-helical interface visual-ized in the original B30.2 domain structure is broken apart by theinsertion of the membrane proximal region from residue K277 toS297. Unlike other parts of the membrane proximal region, thissegment is well ordered in the BFI structure, clearly adopting analpha helical structure. It interacts with the N terminus of theother monomer, mostly making hydrogen bonds and salt bridges,of which most are side-chain-specific (Fig. 4B, Lower). This in-terface involving 28 residues is hereafter referred to as Dimer II.Using MD simulations we probed the stability of these dimer

forms. With the two BFI dimer interfaces as starting points for oursimulations each system was equilibrated and then run at 293.15 Kand 303.15 K until an accumulated total trajectory time of 0.5 μswas reached (simulation details are given inMaterials and Methods).In the case of both interfaces, multiple contacts persist throughoutthe entire 0.5-μs trajectory, suggesting that both Dimer I and DimerII are stable over this time period (Fig. 4B). Visualizations of thesimulated trajectories and analysis highlighting persistent contactsindicate that the hydrophobic interactions of the Dimer I interfaceand the interhelix hydrogen bonds and salt bridges of the Dimer IIinterface are both stable (Fig. 4B). Therefore, combined evidencefrom crystallographic and MD simulation experiments suggest thatBFI can adopt two dimer configurations.The dimer configurations revealed in the BFI structure reveal

interesting features upon further analysis. Intriguingly, the Di-mer I interface is very close to the pAg-binding pocket (Fig. 4).Moreover, in or near this dimer interface we also find a signifi-cant fraction of residues with major CSP or with differentialassignments for the apo and the complex HSQC spectra. Forexample, Y352, a key residue in the Dimer I interface, showedthe largest CSP as mentioned before (Fig. 1B). In addition,among the residues buried in the Dimer I interface, residuesL324, W350, H351, C353, K393, M394, and T395 experiencesignificant changes in chemical environment (Fig. 4C, Left).Notably, the same is true in the Dimer II interface for residuesN300, L434, T438, S442, F443 and Y444 (Fig. 4C, Right). Thisindicates that the pAg-induced conformation change is concen-trated at these putative BFI dimer interfaces.

The Extracellular Domains of Full-Length BTN3A Adopt a V-ShapedConformation in Lipid Nanodiscs. Since Vγ9Vδ2 T cells must detectsignals on the extracellular side of target cells, the intracellulardetection of pAg by BTN3A must be somehow transmitted to theextracellular space for detection. The conformation of the BTN3Aextracellular domains has been speculated to be important forVγ9Vδ2 T cell activation (6). As mentioned earlier, the extracel-lular domains of BTN3A1 can adopt two alternative dimer con-formations, namely a “V-shaped” form and a “head-to-tail” form,according to crystal structure models (PDB ID code 4F80) (6).However, previous in vitro FRET assays on recombinant BTN3A1extracellular domains provided evidence only for the existence ofthe V-shaped dimer (6).To fully examine the possibility of these two putative dimer

conformations in a more physiologically relevant setting, we recon-stituted full-length BTN3A dimers expressed in insect cells intonanodiscs, which are a synthetic model membrane system composedof a phospholipid bilayer surrounded by two amphipathic scaffold

cHDMAPP IPP

40

20

0Nor

mal

ized

CD

69 M

FI (%

)

Isotype HMBPP NBPPBS

WT Y352A

*

0 1 2 0 1 2 3 4 5Molar Ratio

kcal

mol

-1 o

f inj

ecta

nt 0-2-4-6-8

-10-12

0-2-4-6-8

-10-12

WT Y352A

WT Y352A

Buffer Buffer

A

B

ns

ns

Fig. 3. Y352 in the B30.2 domain is important for pAg binding and pAg-induced T cell activation. (A) ITC binding isotherm traces of the BTN3A1B30.2 domain WT protein and Y352 alanine mutant (Y352A) with eithercHDMAPP or IPP. The binding measurements were all performed using 100 μMprotein in the cell and 2 mM ligand in the syringe. The interactions betweenthe WT/Y352A with cHDMAPP (Left) and their interactions with IPP (Right) areshown. Within the same group of ligand, the isotherm traces were plotted onthe same y axis for comparison. The buffer controls are shown in open square,the WT traces in gray triangle (cHDMAPP) or square (IPP), and the Y352 tracesin red triangle (cHDAMPP) or square (IPP). The binding curves shown in allplots were fitted assuming one-site binding. (B) Histogram showing CD69 up-regulation level of Jurkat G115 Vγ9Vδ2 TCR transductants in response totreatment of WT or Y352A BTN3A1 transfected HEK293 cells in BTN3A CRISPRKO background with either PBS, isotype control antibody, anti-CD277 agonistantibody (20.1 mAb), NBP pamidronate, or HMBPP. G115 transfectants treatedwith 1 μg/mL PHA or PBS served as positive or negative controls, and theCD69 level from all of the other experimental groups were normalizedaccording to the controls. Means of three independent experiments with theSD are shown. P values indicated in the plot were calculated using the un-paired Student’s t test method [*P < 0.05, not significant (ns) P > 0.05].

E7314 | www.pnas.org/cgi/doi/10.1073/pnas.1707547114 Gu et al.

Dow

nloa

ded

by g

uest

on

Apr

il 4,

202

0

proteins (Fig. 5A). Interestingly, although the expression level wasequivalent, the purification process of BTN3A1/A2 heterodimerwas much more successful than the BTN3A1 homodimers, leadingto significantly higher yield of the protein product with greaterstability (Fig. 5B and Fig. S3). For this reason, nanodiscs of theBTN3A1/A2 heterodimer, instead of the BTN3A1 homodimer,were visualized by negative-stain EM followed by single-particleanalysis to examine the conformation of the extracellular domainsin the context of full-length BTN3A dimers. Consistent with thepreviously published FRET data, the V-shaped conformation wasreadily observable in several 2D class averages (Fig. 5C). In con-trast, the head-to-tail conformation could not be unequivocallydetected, even though the mean diameter of the reconstitutedBTN3A dimer nanodisc is about 13 nm, which is large enough toaccommodate the head-to-tail conformation (∼8 nm). We conclude

that in the absence of other cellular components and/or exogenouslyadded pAgs the extracellular domains of full-length BTN3A in alipid bilayer form V-shaped dimers.

The Extracellular Domains of BTN3A1 in the Native Cellular EnvironmentAdopt the V-Shaped Conformation at Rest. We next investigatedwhether the V-shaped conformation of BTN3A1 exists in the nativecellular environment. To this end, we performed disulfide trappingby designing a pair of cysteine residues in BTN3A1 that should forman intermolecular disulfide bond if the protein adopts the V-shapedconformation (Fig. 6A). Residues D124 and S207 in the V-shapeddimer interface were mutated to cysteines in the context of amCherry-tagged full-length BTN3A1, and this construct was trans-fected into the BTN3A-KO HEK293 cells described previously.Cells expressing the engineered construct were either treated with

Dimer I Dimer II

Dimer I Dimer II

BFI domain Stem B30.2 domain277 297 298 483

T438L434

F443

N300Y444

S442

L324

W350Y352

K393T395

M394

C353

H351

YW W LR TN

V AVEL QSD KA QRS

Dimer I

E W AA LY

IE W Y SQ ERDimer II

Hydrophobic/VDW H-bond/Salt bridge Contact probability 1

F R F T D S Y G S H T

334 349 350 352 391 471 472

324 325 326 327 328 331 332 333 334 335 336 339 477

298 299 301 302 305 306 307 364 433 438 440 442 444 449 450 451 454

* *

*

D435

E437

281 284 285 287 288 289 291 292 294R

* * *

1-0.8 0.8-0.5

L478

A

B

C

Fig. 4. The 1.9-Å structure of BTN3A1 BFI reveals two potential dimer interfaces. (A) Overview of two dimer conformations observed in the crystal lattice of theBTN3A1 BFI domain. The domain architecture is presented above. One monomer is colored yellow with the extended membrane proximal region colored inorange, and the other is colored light blue. The dimer observed in the asymmetric unit (Left) is referred to as Dimer I. The other dimer (Right) similar to thepreviously published B30.2 structure is referred to as Dimer II. The pAg-binding pockets are indicated by cHDMAPP model colored in cyan. (B) Contact maps withinthe dimer interfaces of Dimer I (Upper) and Dimer II (Lower). Residues in purple and yellow panels are from different monomers within the dimer. The starsindicate residues that undergo major CSP. Dashed lines are hydrophobic or van der Waal interactions and solid lines are salt bridges or hydrogen bonds. Contactlines are color-coded by contact percentage calculated from MD simulations. Contact percentage is defined as the number of simulation frames in which tworesidues are separated by 5 Å or less divided by the total number of frames in the simulation. (C) Residues with major CSP and differential assignment in the apoand complex HSQC spectra are concentrated in two dimer interfaces of BFI. The residues located in or near either Dimer I or Dimer II interface are shown as boldred lines and labeled in the context of the whole dimer. The pAg-binding pockets are indicated by pAg cHDMAPP model colored in cyan.

Gu et al. PNAS | Published online August 14, 2017 | E7315

IMMUNOLO

GYAND

INFLAMMATION

PNASPL

US

Dow

nloa

ded

by g

uest

on

Apr

il 4,

202

0

the pAg HMBPP or left untreated and then subjected to flowcytometry to verify proper expression on the cell surface (Fig. S4)and Western blot analysis to detect dimerization (Fig. 6B). Asexpected, all constructs appear as monomeric species when reducingagent is included in the protein-loading buffer to disrupt disulfidebonds (Fig. 6B, Upper). However, under nonreducing conditions,which leave disulfide bonds intact, WT BTN3A1 appears as amonomer, whereas the D124/S207C mutant migrates as a homoge-nous high-molecular-weight species, with a molecular weight consis-tent with a BTN3A1 dimer (Fig. 6B, Lower). Notably, dimerization ofthe D124/S207 was observed both with and without HMBPP treat-ment. These results demonstrate that BTN3A1 forms V-shaped di-mers on the cell surface and are consistent with our in vitro data.

Locking BTN3A1 at the V-Shaped Dimer Interface Diminishes pAg-Induced T Cell Activation. To test whether the conformation of theextracellular domains of BTN3A1 plays a role in pAg-inducedT cell activation we assayed the disulfide-linked BTN3A1 mutantfor its ability to stimulate Vγ9Vδ2 T cells in their basal state andupon addition of pAgs (HMBPP or NBP) or the 20.1 antibody.Intriguingly, we observed that locking BTN3A1 together at theV-shaped dimer interface (D124/S207C) significantly impairedboth pAg-induced and 20.1-induced T cell activation (Fig. 6C).This observation is consistent with the notion that the V-shapedconformation is the resting state of BTN3A1 and is not in itselfsufficient for T cell activation.It has been reported previously that BTN3A1 becomes immo-

bilized on the cell surface in the presence of 20.1 antibody, sug-gesting that BTN3A1 membrane reorganization is also importantfor Vγ9Vδ2 T cell activation (4). Thus, we speculated that theD124/S207C mutant might prevent T cell activation by disruptingproper immobilization of BTN3A1 on the cell surface. To formallytest this prediction, we performed fluorescence recovery afterphotobleaching (FRAP) experiments, wherein mCherry-taggedBTN3A1 is photobleached within a small area of the cell sur-face and the recovery of fluorescence is observed as an indicator ofBTN3A1 mobility within the membrane. Contrary to our pre-diction, we observed that the D124/S207C mutant exhibited nodefect in immobilization in the presence of the 20.1 antibody (Fig.6D and Fig. S5). This is evidence that even though important toVγ9Vδ2 T cell activation, simply immobilizing BTN3A1 at the cellmembrane is insufficient to stimulate Vγ9Vδ2 T cells. Based onthe cumulative work presented here, we propose that the activa-tion of Vγ9Vδ2 T cells requires additional features beyondmembrane reorganization on the target cell, including a confor-mational rearrangement of the BTN3A1 extracellular domain.

DiscussionHuman Vγ9Vδ2 T cell recognition of infection and tumorigen-esis is of great interest in immunology and directly relevant forimmunotherapeutic applications. Despite the significant poten-tial of these cells, a clear understanding of the molecularmechanisms underlying pAg-induced T cell activation still re-mains elusive. The structural, biophysical, and functional datapresented here provide insights into the events after intracellularpAg sensing critical for Vγ9Vδ2 T cell activation.First, we showed by NMR spectrometry that upon pAg binding

the BTN3A1 intracellular B30.2 domain undergoes a confor-mational change. A significant fraction of residues experiencingmajor CSP lie in or near the ligand-binding pocket but also in-clude residues located in other regions of the B30.2 domain.Complementary to the experimental data, all-atom MD simulations

-2.00

3.00

8.00

13.00

18.00

23.00

28.00

0.00 5.00 10.00 15.00 20.00 Elution volume (ml)

~13 nm

BTN3A1 monomer

BTN3A2 monomer

E3D1 scaffold protein

lipids

25 nm

A28

0 (m

Au)

BTN3A

Empty

A

B

C

Fig. 5. The extracellular domains of BTN3A full-length protein recon-stituted in lipid nanodiscs form a V-shaped conformation. (A) A schematicrepresentation of the BTN3A1/A2 heterodimer nanodisc. All componentsare labeled in the corresponding colors. (B) Gel filtration chromatographs ofthe reconstituted BTN3A1/A2 heterodimer nanodiscs (blue) overlaid with theempty nanodiscs control (gray). (C) Negative-staining EM images of the

BTN3A1/A2 heterodimer in nanodiscs. On the left are the class averages of3,968 out of the 4,359 picked particles. A close-up view of the selected classaverages is shown at the bottom.

E7316 | www.pnas.org/cgi/doi/10.1073/pnas.1707547114 Gu et al.

Dow

nloa

ded

by g

uest

on

Apr

il 4,

202

0

of the intracellular domain monomer with and without pAgcHDMAPP provide insight into the structural changes that arecausing these chemical shifts. Backbone rmsd analysis derived fromthe MD simulations are consistent with the NMR data and alsosuggest that the pAg-proximal CSPs, namely those of residues 393–397 and 410–419, are due to a decrease in flexibility of the bindingpocket upon pAg binding.These backbone motions, which show overall flexibility in the

apo simulations, can be seen in a relatively short (100 ns)timescale but do not rule out the possibility of slower, globalchanges in the protein backbone. Consistent with this idea, ourNMR data revealed many residues exhibiting broadened signalseither in the apo or the complex spectra possibly due to in-termediate or slow chemical exchange, thus hindering their as-signments of both spectra. Although these residues cannot besubjected to CSP analysis due to missing assignments and likelycannot be probed with all-atom MD simulations, they still revealvaluable information on where the chemical environment changeoccurs during ligand binding, and intriguingly, most are locatedremotely from the pAg-binding pocket (Fig. S2).What could be the main consequence of this conformational

change? By mapping residues with major CSP and residues withdifferential assignments in the newly solved crystal structure ofthe BFI domain we found that many of these residues concen-trated in two dimeric interfaces revealed in the structure.According to the MD simulations, the clusters of residues withmajor CSP and rmsd shifts reside in or near the Dimer I in-terface of the BFI. This overlap between interfacial residues andthose that shift upon the addition of pAg, as well as the in-volvement of a residue required for pAg binding and T cell ac-tivation, Y352, further implicates a connection between pAgbinding and intracellular conformational change (12).Y352 makes a major contribution to the Dimer I interface

interaction through two hydrogen bonds via the hydroxyl group,and the contacts are shown to be long-lived (they persist fornearly the entire simulated trajectory). Based on the location ofthis residue in relation to the orientation of pAg in the bindingsite (8), it might engage the isopentenyl chain of pAgs with itshydroxyl group, which could potentially compete with its dimerinterface interactions. Indeed, Y352 experiences the largest CSPupon pAg binding, implying that its surrounding chemical envi-ronment changes substantially in the presence of pAg. Theseobservations suggest that Y352 might act as a switch that directlylinks the pAg binding to the change of BTN3A1 intracellulardomain dimer conformation. Since other residues within thesame dimer interface also experience major change in theirchemical environment in the presence of pAg, this pAg-sensingswitch is likely to be a synergic effect among all these residuesincluding Y352. This is consistent with our observation thatsingle mutation of Y352 is not sufficient to completely abrogatethe pAg-induced T cell activation.The membrane proximal region of BFI is an important addition

to the previously published B30.2 domain structure (8). It is in-volved in a new, N-to N-terminal dimer interface (Dimer II) andadds stability to this particular interface based on the result fromMD simulations. While the BFI interface is stable in simulationfor the full 0.5-μs trajectory, the truncated B30.2 interface quicklydissociates (∼100 ns). Although we were unable to resolve theentire membrane proximal region in the crystal structure, pre-viously published work from our laboratory and others suggeststhat this region interacts with other protein(s) that are important

A

C

B

D124/S207

WT D124/S207C HMBPP - +

Reducing condition

Non-reducing condition

D

S-S

190

75 M

190

75

Nor

mal

ized

CD

69

MFI

(%)

HMBPP NBP PBS Isotype 20.1mAb

50 40 30 20 10 0

D

PBS PBS 20.1mAb 20.1mAb

- +

* *

**

WT D124/S207C

WT D124/S207C

S b S b

ns ns

Imm

obile

Fra

ctio

n (%

)

100 80 60 40 20 0

**** ****

Fig. 6. A proper conformation of the BTN3A1 extracellular domains is im-portant for pAg-induced Vγ9Vδ2 T cell activation. (A) Positions of the residues(shown in stick representation) that were mutated to cysteine for disulfidebond formation in the putative dimer interface of the BTN3A1 extracellulardomains. The proposed full-length cross-linked BTN3A1 in the cell membraneis also shown in cartoon representation. (B) Western blots revealing theBTN3A1 protein species from WT BTN3A1 and S207A/D124A BTN3A1 mutanttransfected HEK293 cells in BTN3A CRISPR KO background, either treatedwith HMBPP or left untreated. (Upper) A PM fraction sample from each groupwas treated with reducing protein loading buffer with 5 min heating at95 °C. (Lower) The PM fraction samples were treated with nonreducingprotein loading buffer. “M” marks the monomer species and “D” marksthe dimer species. (C) Histogram shows CD69 up-regulation level of JurkatG115 transductants in response to treatment of WT or S207/D124C mutantBTN3A1 transfected HEK293 cells in BTN3A CRISPR KO background with eitherPBS, isotype control antibody, anti-CD277 agonist antibody (20.1 mAb),pamidronate (NBP), or HMBPP. G115 transfectants treated with 1 μg/mL PHAor PBS served as positive or negative control, and the CD69 level from all of theother experimental groups was normalized according to the controls. Meansof three independent experiments with the SD are shown. P values indicatedin the plot were calculated using the unpaired Student’s t test method (*P <0.05, **P < 0.01). (D) FRAP analysis of cells expressing BTN3A1 (WT) or S207/D124C mutant after incubation for 30 min with anti-CD277 agonist antibodyor PBS. Data are presented as the value for immobile fraction, measured asdescribed under “FRAP Analysis” in SI Materials and Methods. WT-PBS: n = 19;

WT-20.1mAb: n = 24; S207/D124C-PBS: n = 22; S207/D124C-20.1mAb: n = 20.Bars represent mean values and SD. P values indicated in the plot werecalculated using the unpaired Student’s t test method [****P < 0.0001, notsignificant (ns) P > 0.05].

Gu et al. PNAS | Published online August 14, 2017 | E7317

IMMUNOLO

GYAND

INFLAMMATION

PNASPL

US

Dow

nloa

ded

by g

uest

on

Apr

il 4,

202

0

for pAg-induced T cell activation and may adopt a more struc-tured conformation when bound to its partner(s) (10, 11). Con-sistent with the idea that a conformational change likely occurs inthis dimer interface, residues in this region were shown to undergomajor CSP in previous studies (9). Because of its location in be-tween the intracellular and the extracellular domains, this regionmight be critical in translating the pAg-induced conformationalchange of BTN3A1 from inside to outside of the target cells,which could then be detected by Vγ9Vδ2 T cells. Structural de-termination of the full stem region with both apo and pAg-boundB30.2 domain could potentially reveal more detailed structuralinformation on how this process would occur.How is a conformational change of the intracellular domain of

BTN3A1 transduced to the extracellular region to be sensed byVγ9Vδ2 T cells? A critical role of BTN3A extracellular domains inthe activation process has been proposed, since the antibody20.1 against the BTN3A extracellular domains can stimulateVγ9Vδ2 T cells independent of pAgs. Two distinct dimer confor-mations, namely a V-shaped and a head-to-tail form, were observedin the crystal structure model of the BTN3A1 extracellular domains(6). We originally speculated that the binding of 20.1 or binding ofpAg in the BTN3A1 intracellular domain drives a conformationconversion between the two. Previous investigations of soluble ver-sions of the BTN3A1 extracellular domains via FRET revealed ev-idence for only the V-shaped form (6). Here we examined the dimerconformations in the context of the full-length BTN3A1 protein in areconstituted lipid system as well as in the native cellular environ-ment. Consistent with our previous data, we found direct evidencefor the existence of the V-shaped dimer form. The presence of theV-shaped extracellular domains in solution, in lipid-like nanodiscs,and in cellular membranes without any exogenously addedpAg stimulants indicates that this form is a resting state of theBTN3A1 molecule. Furthermore, locking this dimer interface withengineered disulfide bonds produced BTN3A1 that is defective inits ability to mediate pAg-induced and 20.1-induced T cell activa-tion. Interestingly, the Y352A mutant also showed a defect in bothpAg-mediated and 20.1-mediated T cell stimulation. These resultssuggest that compromised dimer interfaces, either in the extracel-lular or the intracellular domains, can affect the global configura-tion of BTN3A1, and thus affect Vγ9Vδ2 T cell stimulation. Wealso assessed the membrane mobility of this mutant under differentconditions using FRAP and found that it behaves much like the WTBTN3A1. Of note, in our FRAP analysis we were unable to observea significant difference between PBS-treated and pamidronate-treated BTN3A1, both for the WT and the mutant BTN3A1 (Fig.S5). Since pAg-induced membrane reorganization was not impairedin the locked V-shaped mutant, these results suggest that eventhough membrane reorganization is important for pAg-inducedT cell activation, itself alone is not sufficient to stimulate Vγ9Vδ2T cells, and a conformational rearrangement of the extracellulardomain is indeed critical for the activation process.Outside of the crystallographic data where we observed the

head-to-tail dimer we have no data supporting the existence orthe functional relevance of this dimer in full-length, cell-surfaceexpressed BTN3A. We originally speculated this could be theresting-state dimer. This is because the dimer interface of thehead-to-tail dimer overlaps with the 20.1-binding site, and thusbinding of the agonist 20.1 is incompatible with this dimer con-formation and would presumably destabilize it on the cell surface(6). However, the data presented here do not support theaforementioned hypothesis. Alternatively, based on the 20.1–BTN3A1 extracellular domain complex structure where we ob-served a subtle rotational shift in the V-shaped dimer when incomplex with 20.1 (6), we hypothesize that a similar shift toBTN3A1 extracellular domains may be induced by pAg bindingto the intracellular domain. This “rotational shift” model isconsistent with the data we present here. Additional studies willbe needed to determine whether such a state exists and, if so, to

uncover the unique structural features that enable it to activateVγ9Vδ2 T cells.Overall, clarifying the nature of the BTN3A1 conformational

change induced by pAgs is essential for understanding how anintracellular pAg-sensing event is transduced to outside of thetarget cells for possible detection by Vγ9Vδ2 T cells. Elucidationof the molecular mechanisms of pAg-induced T cell activationnot only broadens and deepens our understanding of the myriadways in which γδ T cells can be activated but also lays a foun-dation for practical applications such as the development oftherapeutics targeting certain infectious diseases and cancers.

Materials and MethodsNMR Spectrometry. NMR spectra were acquired with Bruker AVANCE IIIHD600-MHz NMR spectrometer equippedwith a room-temperature TXI probe. Allspectra were acquired on partially deuterated samples at a temperature of25 °C. It was found that at 600 MHz, HN transverse relaxation-optimizedspectroscopy (TROSY) showed higher resolution than the normal HN HSQC.Assignments were thus obtained by analysis of TROSY versions of the standardtriple resonance experiments including HNCO, HNcaCO, HNCA, HNCACB,HNcoCA, HNcoCACB, and NcocaNH. Spectra were processed in NMRpipe (12)and then assignments were done with CARA (13). The average CSPs were

calculated according to the formula ΔδNH=ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiΔδH2 + ðΔδN=5Þ2

qfor the resi-

dues assigned both in the free and ligand-bound B30.2 domain.

ITC. Proteins samples for ITC were purified over a Superdex 200 into 20 mMHepes, pH 7.4, and 150 mM NaCl. Concentration of peak fractions was de-termined from A280 using the theoretical extinction coefficient. The fractionswere pooled and diluted to 50 μM. IPP was obtained from Echelon Biosciences,and cHDMAPP was obtained from Innate Pharma. ITC data were collectedusing 50 μM B30.2 WT and Y352A mutant in the cell and 2 mM pAg in thesyringe on a MicroCal iTC200 (GE Healthcare Life Sciences) at 25 °C using aninitial injection of 0.4 μL followed by 19 2-μL injections. ITC binding fits weredetermined after reference subtraction of injections of the ligand into bufferonly from the experimental results. Fits were determined using the MicrocalOrigin software. The stoichiometry (N) of the IPP-B30.2 interaction was set toone during fitting due to weak affinity.

Crystallization. Large rod-shaped crystals of apo BTN3A1 BFI were obtainedusing hanging drops consisting of 1 μL of 12.5 mg/mL protein purified asabove with 4 mM β-mercaptoethanol in the protein solution and 1 μL ofmother liquor containing 0.1 M Hepes, pH 7.0, 0.2 M MgCl2, and 22%PEG3350. Crystals were transferred into drops containing mother liquor witheither an additional 1 mM, 5 mM, or 10 mM cHDMAPP and incubated atroom temperature overnight. Of note, these crystals did not show any visibledamage. Crystals were then cryoprotected by sequentially transferring todrops containing the mother liquor with an additional 5%, then 10%, andfinally 20% glycerol before freezing. The cryoprotectant for each crystalcontained 1 mM, 5 mM, or 10 mM cHDMAPP according to pAg-soakingcondition of the crystal.

Data Collection and Processing. X-ray datasets for the BFI-pAg crystals werecollected at the Advanced Photon Source (APS) beamline 23-ID-B on aMAR300

CCD at 100 °K. A complete dataset was collected over 360 images using a0.5° oscillation at a wavelength of 1.033 × 10−10 m. HKL2000 and Mosflmwere used to index, integrate, and scale the data (14). An initial molecularreplacement solution was obtained using PHASER with PDB ID code 4N7I as asearch model (15, 16). The initial model was built using ARP/wARP and im-proved using Coot with manual building followed by automated individualsite, NCS, and B-factor refinement using Refmac5. All structural figures weregenerated using Pymol (DeLano Scientific).

All-Atom MD Simulations. All simulations performed were prepared using theCHARMM-GUI Input Generator (17–19). All systems, starting from the BFIcrystal structure, were fully hydrated with TIP3P water molecules and neu-tralized with 0.15 M KCl. All simulations were carried out in simulation boxeswith periodic boundary conditions (20) using the additive PARAM36 forcefield from the CHARMM (Chemistry at HARvard Macromolecular Mechanics)(19). pAg, which has not been parameterized and is not found in the BFIstructure, was modeled in based on the binding location determined bySandstrom et al. (8) with custom parameter files based upon other well-characterized diphosphate parameters. Simulation specifics varied for eachsystem and are described in the SI Materials and Methods but used a

E7318 | www.pnas.org/cgi/doi/10.1073/pnas.1707547114 Gu et al.

Dow

nloa

ded

by g

uest

on

Apr

il 4,

202

0

combination of NAMD, OpenMM, and AMBER (18, 21–24). For all simulatedsystems run on the Midway Computing Cluster at the University of Chicagoat least two replicas were run to confirm independence of results on initialvelocity assignments.

BTN3A Nanodisc Production. Reconstitution of detergent-solubilized His-tagged BTN3A1/A2 dimers was performed according to the protocols de-scribed by the Sligar laboratory with minor modifications (25). BTN3A1/A2 dimer was reconstituted into nanodiscs using either MSP 1D1 or E3D1(Addgene) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC):TX-100 mixed micelles described in the SI Materials and Methods. Excess ofdetergent was removed by overnight incubation at 4 °C with Bio-Beads SM-2Adsorbents (Bio-Rad). BTN3A nanodiscs were separated from proteolipo-somes and empty nanodiscs first by Ni-NTA resin pull-down (Qiagen) fol-lowed by Superdex 200 10/300 GL column (GE Healthcare) in 50 mM Hepes,pH 7.5, with 50 mM NaCl. As a control, empty nanodiscs were prepared usingthe same procedure. Gel filtration fractions corresponding to BTN3A1/A2 nanodiscs were concentrated and an aliquot was shipped on ice to G.S.for negative-stain EM. The remainder sample was supplemented with 5%(wt/vol) sucrose, flash-frozen in liquid nitrogen, and stored at −80 °C.

Negative-Stain EM. BTN3A1/A2 nanodiscs, produced as described above, wereprepared for EM using the conventional negative staining protocol (26) andimaged at room temperature with a Tecnai T12 electron microscope operated at120 kV using low-dose procedures. Images were recorded at a magnification of71,138× and a defocus value of ∼1.5 μm on a Gatan US4000 CCD camera. Allimages were binned (2 × 2 pixels) to obtain a pixel size of 4.16 Å on the specimenlevel. Particles were manually excised using e2boxer (part of the EMAN 2 soft-ware suite) 200 (27). Two-dimensional reference-free alignment and classifica-tion of particle projections was performed using ISAC (28); 4,359 projections ofBTN3A1/A2 nanodisc were subjected to ISAC, producing 70 classes consistentover two-way matching and accounting for 3,968 particle projections.

In Vivo Cross-Linking. The BTN3A1 S207/D124C-RFP mutant cell lines were culti-vated in a 150-mm TC-treated cell culture dish (Falcon) with 25 mL Dulbecco’smodification of Eagle’s medium media with 4.5 g/L glucose, L-glutamine, sodiumpyruvate, and 10% FBS (Corning). The cells were either treated with 5 μMHMBPP (Sigma) for 2 h at 37 °C or left untreated before harvesting using a largecell scraper (Corning). The cells were then washed three times with 1× PBS andincubated in PBS, pH 8.0, contained with 0.1 M iodoacetamide (Thermo Scien-tific) in the dark at room temperature for 30min. After incubation, the cells werelysed and the plasmamembrane (PM) fraction was isolated using the Minute PMprotein isolation kit (Invent Biotechnologies, Inc.). The PM fraction was then splitinto two samples: one was treated with NuPAGE LDS sample buffer (Novex)supplemented with 600 mM β-mercaptoethanol and heated at 95 °C for 5 min,and the other was treated with NuPAGE LDS sample buffer (Novex) alonewithout heating. All samples were then subjected to SDS/PAGE andWestern blotanalysis. The primary antibody used was a rabbit polyclonal antibody againstresidues 344–374 of BTN3A1 (LSBio) and the secondary antibody was a horse-radish peroxidase-conjugated goat polyclonal antibody against rabbit IgG (EMD;Millipore). The blot was detected using SuperSignal West Pico chemiluminescentsubstrate (Thermo Scientific).

Functional Assays. For CD69 expression assays target cells were plated at 70%confluence in a 96-well plate (∼10,000 cells in total) and pretreated for 2 h at37 °C with either CD277-specific mAb (BioLegend) at 5 μg/mL, isotype controlmAb (BioLegend) at 5 μg/mL, NBP pamidronate (Sigma) at 125 μM, orHBMPP (Sigma) at 5 μM per well. Treated cells were next extensively washedand cocultured together with Jurkat J.RT3-T3.5 T cells transfected withVγ9Vδ2 G115 TCR (100,000 cells in total) at 37 °C in complete RPMI-1640Medium. G115 transductants alone treated with 1 μg/mL phytohemagglu-tinin (PHA) served as the positive control whereas transductants left un-treated served as the negative control. After 4 h, cells were harvested,stained with fluorochrome-labeled Vδ2 TCR-specific mAb (BioLegend) andCD69-specific mAb (BioLegend) analyzed by flow cytometry. Data werecollected on a Fortessa 4-15 or a LRII cytometer (BD Biosciences) and ana-lyzed with FlowJo software (TreeStar). All of the experimental data werenormalized to the positive control (100%) and the negative control (0%)using the following equation:

Normalized MFI=MFIðiÞ−MFIðnegative  controlÞ

MFIðpostive  controlÞ−MFI  ðnegative  controlÞ×100%,

where MFI is the mean fluorescence intensity of each experimental group orthe controls.

More details are provided in SI Materials and Methods.

FRAP. mCherry-fused WT and D124/S207C mutant BTN3A1 transfected cellswere cultivated and imaged in 35-mm, no. 1.5 coverslip, 10-mm glass-bottomdish (MatTek). The cells were subjected to pretreatment either with PBS,250 μM pamidronate (Sigma), or 5 μg/mL purified anti BT3.1 20.1mAb(Biolegend) for 2 h at 37 °C. Then the cells were imaged under Leica SP5IISTED-CW supperresolution laser scanning confocal microscope (with 63×/1.4UV oil objective). A 561-nm DPSS laser was used at 15% intensity for imagingand a 592-nm depletion laser was used at 80% intensity for photobleaching.The selected rectangular areas within the cell membrane were photo-bleached for five frames (200 ms per frame, 2-s interval). Images were col-lected every 5 s for 15 s (three frames, 200 ms per frame) before bleachingand every 2 s for 180 s (90 frames, 200 ms per frame) after bleaching. Theimages were processed using ImageJ software and the data were analyzedusing PRISM6. More details are provided in SI Materials and Methods.

ACKNOWLEDGMENTS. We thank the staff of the APS at GM/CA-CAT (23-ID)for their use and assistance with X-ray beamlines and Dr. Ruslan Sanishvili inparticular for help and advice during data collection, Dr. John Leonard andDr. Sobhan Roy for advice and insightful discussion, Dr. C. Belmant fromInnate Pharma for providing cHDMAPP, Dr. Caitlin Castro for help and adviceduring CRISPRi KO cell line generation and cellular functional assays, andDr. Christine Labno at the Integrated Light Microscopy Core Facility, Univer-sity of Chicago, for help and advice in conducting FRAP experiments. Thiswork was supported by NIH Grant R01 AI115471. The crystallographic datawere collected, processed and interpreted during the 2015 CCP4/APS Schoolin macromolecular crystallography held in APS. All NMR spectra were col-lected in the Biomolecular NMR Core Facility at the University of Chicago.

1. Morita CT, Jin C, Sarikonda G, Wang H (2007) Nonpeptide antigens, presentation

mechanisms, and immunological memory of human Vgamma2Vdelta2 T cells: Dis-

criminating friend from foe through the recognition of prenyl pyrophosphate anti-

gens. Immunol Rev 215:59–76.2. Gober HJ, et al. (2003) Human T cell receptor gammadelta cells recognize endogenous

mevalonate metabolites in tumor cells. J Exp Med 197:163–168.3. Lang F, et al. (1995) Early activation of human V gamma 9V delta 2 T cell broad cy-

totoxicity and TNF production by nonpeptidic mycobacterial ligands. J Immunol 154:

5986–5994.4. Harly C, et al. (2012) Key implication of CD277/butyrophilin-3 (BTN3A) in cellular stress

sensing by a major human γδ T-cell subset. Blood 120:2269–2279.5. Abeler-Dörner L, Swamy M, Williams G, Hayday AC, Bas A (2012) Butyrophilins: An

emerging family of immune regulators. Trends Immunol 33:34–41.6. Palakodeti A, et al. (2012) The molecular basis for modulation of human Vγ9Vδ2 T cell

responses by CD277/butyrophilin-3 (BTN3A)-specific antibodies. J Biol Chem 287:

32780–32790.7. Vavassori S, et al. (2013) Butyrophilin 3A1 binds phosphorylated antigens and stim-

ulates human γδ T cells. Nat Immunol 14:908–916.8. Sandstrom A, et al. (2014) The intracellular B30.2 domain of butyrophilin 3A1 binds

phosphoantigens to mediate activation of human Vγ9Vδ2 T cells. Immunity 40:

490–500.9. Hsiao CH, et al. (2014) Synthesis of a phosphoantigen prodrug that potently activates

Vγ9Vδ2 T-lymphocytes. Chem Biol 21:945–954.

10. Sebestyen Z, et al. (2016) RhoB mediates phosphoantigen recognition by Vγ9Vδ2T cell receptor. Cell Rep 15:1973–1985.

11. Rhodes DA, et al. (2015) Activation of human γδ T cells by cytosolic interactions ofBTN3A1 with soluble phosphoantigens and the cytoskeletal adaptor periplakin.J Immunol 194:2390–2398.

12. Delaglio F, et al. (1995) NMRPipe: A multidimensional spectral processing systembased on UNIX pipes. J Biomol NMR 6:277–293.

13. Keller RLJ, Meier BH, Pervushin K (2005) Optimizing the process of nuclear magneticresonance spectrum analysis and computer aided resonance assignment. PhD disser-tation (Naturwissenschaften, Eidgenössische Technische Hochschule, ETH Zürich,Zurich), no. 15947.

14. Otwinowski Z, Minor W (1997) Processing of X-ray diffraction data collected in os-cillation mode. Methods Enzymol 276:307–326.

15. McCoy AJ, et al. (2007) Phaser crystallographic software. J Appl Cryst 40:658–674.16. Weinert C, Grütter C, Roschitzki-Voser H, Mittl PR, Grütter MG (2009) The crystal

structure of human pyrin b30.2 domain: Implications for mutations associated withfamilial Mediterranean fever. J Mol Biol 394:226–236.

17. Jo S, Kim T, Im W (2007) Automated builder and database of protein/membranecomplexes for molecular dynamics simulations. PLoS One 2:e880.

18. Lee J, et al. (2016) CHARMM-GUI input generator for NAMD, GROMACS, AMBER,OpenMM, and CHARMM/OpenMM simulations using the CHARMM36 additive forcefield. J Chem Theory Comput 12:405–413.

19. Jo S, Kim T, Iyer VG, Im W (2008) CHARMM-GUI: A web-based graphical user interfacefor CHARMM. J Comput Chem 29:1859–1865.

Gu et al. PNAS | Published online August 14, 2017 | E7319

IMMUNOLO

GYAND

INFLAMMATION

PNASPL

US

Dow

nloa

ded

by g

uest

on

Apr

il 4,

202

0

20. Jorgensen WL, Chandrasekhar J, Madura JD, Impey RW, Klein ML (1983) Comparisonof simple potential functions for simulating liquid water. J Chem Phys 79:926–935.

21. Phillips JC, et al. (2005) Scalable molecular dynamics with NAMD. J Comput Chem 26:1781–1802.

22. Eastman P, et al. (2013) OpenMM 4: A reusable, extensible, hardware independentlibrary for high performance molecular simulation. J Chem Theory Comput 9:461–469.

23. Friedrichs MS, et al. (2009) Accelerating molecular dynamic simulation on graphicsprocessing units. J Comput Chem 30:864–872.

24. Salomon-Ferrer R, Götz AW, Poole D, Le Grand S, Walker RC (2013) Routine micro-second molecular dynamics simulations with AMBER on GPUs. 2. Explicit solventparticle mesh Ewald. J Chem Theory Comput 9:3878–3888.

25. Ritchie TK, et al. (2009) Chapter 11 - Reconstitution of membrane proteins in phos-pholipid bilayer nanodiscs. Methods Enzymol 464:211–231.

26. Filizola M (2015) G Protein-Coupled Receptors in Drug Discovery: Methods and Pro-tocols (Springer, New York), 2nd Ed.

27. Tang G, et al. (2007) EMAN2: An extensible image processing suite for electron mi-croscopy. J Struct Biol 157:38–46.

28. Yang Z, Fang J, Chittuluru J, Asturias FJ, Penczek PA (2012) Iterative stable alignmentand clustering of 2D transmission electron microscope images. Structure 20:237–247.

29. Hopkins CW, Le Grand S, Walker RC, Roitberg AE (2015) Long-time-step molecular dy-namics through hydrogen mass repartitioning. J Chem Theory Comput 11:1864–1874.

30. Hoover WG (1985) Canonical dynamics: Equilibrium phase-space distributions. PhysRev A Gen Phys 31:1695–1697.

31. Nose S, Klein ML (1983) A study of solid and liquid carbon tetrafluoride using theconstant pressure molecular-dynamics technique. J Chem Phys 78:6928–6939.

32. Darden T, York D, Pedersen L (1993) Particle mesh Ewald - An N.Log(N) method forEwald sums in large systems. J Chem Phys 98:10089–10092.

33. Ryckaert JP, Ciccotti G, Berendsen HJC (1977) Numerical-integration of cartesianequations of motion of a system with constraints: Molecular dynamics of n-alkanes.J Comput Phys 23:327–341.

E7320 | www.pnas.org/cgi/doi/10.1073/pnas.1707547114 Gu et al.

Dow

nloa

ded

by g

uest

on

Apr

il 4,

202

0