SUPPLEMENTARY INFORMATION

Transcription

1 In the format provided by the authors and unedited. SUPPLEMENTARY INFORMATION ARTICLE NUMBER: DOI: 1.138/NMICROBIOL An HIV-1 antibody from an elite neutralizer implicates the fusion peptide as a site of vulnerability Marit J. van Gils*, Tom L.G.M. van den Kerkhof, Gabriel Ozorowski, Christopher A. Cottrell, Devin Sok, Matthias Pauthner, Jesper Pallesen, Natalia de Val, Anila Yasmeen, Steven W. de Taeye, Anna Schorcht, Stephanie Gumbs, Inez Johanna, Karen Saye-Francisco, Chi-Hui Liang, Elise Landais, Xiaoyan Nie, Laura K. Pritchard, Max Crispin, Garnett Kelsoe, Ian A. Wilson, Hanneke Schuitemaker, Per Johan Klasse, John P. Moore, Dennis R. Burton, Andrew B. Ward, Rogier W. Sanders* * Corresponding authors. m.j.vangils@amc.uva.nl; r.w.sanders@amc.uva.nl This PDF file includes: Supplementary Figures 1-9 Supplementary Tables 1-6 NATURE MICROBIOLOGY Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

2 a b Geometric mean IC 5 titer virus isolation mab isolation elite broad months post-sc Geomean MGRM-C-26 92BR2 92TH21 93IN95 94UG13 JRCSF NL43 Supplementary Figure 1. CD4+ T-cell count, viral load, and heterologous neutralization response in the elite neutralizer during the course of infection. (A) Viral load is recorded on the left y-axis and plotted in black, the CD4+ T-cell count on the right y-axis and shown in red. On the x-axis time after seroconversion (SC) is shown. Antiretroviral therapy was taken during the period indicated in grey. (B) The longitudinal geometric mean IC 5 titers were measured using a six-virus panel. The time points at which virus and mab were isolated are indicated by arrows. Figure modified from 24, donor D1295 of the Amsterdam Cohort Studies (IDU2 in 24 ).

3 a H2C11 ACS241 H2D5 H3C7 H1E1 H4C6 ACS242 H4B9 H1F3 ACS243 H1D4 ACS244 H2F7 H4D3 H4G2 H2D6 H4C7 H2F3 H1G11 H1C3 ACS221 H1D9 H1C1 H1C12 H1E12 ACS231 H4B3 ACS24 H1E3 ACS25 H4E7 ACS21 H1G7 IgGHV 3-3 family H2G2 ACS22 H1F11 ACS23 H2E1 H2F11 H4B5 H3F5 H4G6 H3B9 H1E9 H3D8 H3C11 ACS212 H2D3 H3F7 H3G3 ACS211 H1F7 H4G9 ACS271 H4G1 H2D11 H1E11 H4B8 H1C11 H1B11 H3G8 H2D1 H2E4 H2C6 H3F1 H3B7 H3E9 H1E7 H3F6 H3D9 H3C6 H3D6 H1C7 ACS261 H4D11 ACS251 H1B2 H1D7 ACS252 H1B3 IgG IgHV b Heavy chain FR CDR FR CDR FR CDR3 ----FR4---- VH3-3*3 QVQLVESGGGVVQPGRSLRLSCAASGFTFSSYGMHWVRQAPGKGLEWVAVISY.DGSNKYYADSVKGRFTISRDNSKNTLYLQMNSLRAEDTAVYYCARD... ACS21 E----Q G----T------A--DF GSGH-QHQS-SE------S-T---E--I---H-G---T K-RLGRAWNIGGRLEYHYYGMDVWGQGTTATVSS ACS22 E G A-KDF GGGH-QHQS-SE------A-T---E--K---H-DR--T K-RLGRPWNIGGRLVYYYYGMDVWGQGTTATVSS ACS23 E G----T------A-KDF GSGH-QHQA-SE------S-T---E--R---H-GR--T K-RLGRAWNIGGRLEYHYYGMDVWGQGPAATVSS! ACS24 E----Q G P-KDF Q GGGH-LHQS-SE------Y-----E----F-N-DNV-T---G-----K-RLGRPWNMGRRIEYHYYGMDVWGQGTTATVSS ACS25 E G A-KDF Q GGGH-LHQS-SE------Y-----E--K-F-N-DNV-T---G-----K-RLGRPWNMGRRIEYHYYGMDVWGQGTTATVSS Light chain FR _CDR1_ FR CDR FR CDR3 ---FR4---- VK1-33 DIQMTQSPSSLSASVGDRVTITCQASQDISNYLNWYQQKPGKAPKLLIYDASNLETGVPSRFSGSGSGTDFTFTISSLQPEDIATYYCQQYDNLPP... ACS21 --R---A-V S KKS----R---S---E---H---I-Q----AA-TA-----R-S-V-NK-----V---F--H-A-F-FTFGPGTKVNIK ACS22 --R---A-V KKS----R---S---E---H---I-Q-----A-TA-----H-S-V-NK-----VG--F--E-E--QFTFGPGTKVNIK ACS23 --R---A-V KKS----R---S---E---H---I-Q-----A-TA-----R-S-V-NK-----V---F--E-E--QFTFGPGAKVNIK ACS24 --RI--A-V KKS----R-----G-E---H---I-Q-----G-TA-----H-S-V-NKV----V---F----A--QFTFGPGTKVNIK ACS25 --R--PA-V KKS----R-----G-E---H---I-Q-----G-TA-----H-S-V-NK-----V---F----A--QFTFGPGTKVNIK.1

4 Supplementary Figure 2. Phylogenetic tree of the IgH V genes. (A) maximum-likelihood tree of all the productive IgH V gene sequences was constructed using PhyML. Germline IgH V1-69 was added as an outlier. All the mabs that were cloned and tested are indicated by name. The ACS21-26 clonal family, which uses the IgH V3-3 gene, is emphasized. (B) Alignment of the IgH and IgLκ gene amino acid sequences of ACS21-ACS25 with the most closely related germline and V gene regions indicated above.

5 a Neutralization titer (µg/ml) IC5 IC8 b Max neutralization (%) ACS22 Supplementary Figure 3. Neutralization characteristics of ACS22. (A) Neutralization IC 5 and IC 8 values and (B) maximum percentage of virus neutralization (only shown for viruses with full neutralization curve) of ACS22 on a reference panel of 75 viruses (Supplementary Table 2). IC 5, IC 8 andmaximum percentage of virus neutralization are depicted as medianwith interquartile range.

6 1F11 ACS22 N u m b e r o f P r o t e in s ACS21 ACS G 7 ACS21 5 N u m b e r o f P r o t e in s a D is p la c e m e n t ( lo g ) [ACS21MFI] PI=1.183 [ACS22MFI] PI=.5462 b High polyreactivity Low polyreactivity c units/ml 1 4E1 ACS22 Humira 1-1 Jo B pen C Sc l-7 SS -B SS -A Sm P N R 1U sn R N P/ Sm 1 d e anti-cardiolipin.8 OD (45nm).6 Humira ACS22 4E1.4 2 µm Positive 2 µ m ACS Ab [µg/ml] 2 µ m Negative 2 D is p la c e m e n t ( lo g ) 2 µ m PG9

7 Supplementary Figure 4. Polyreactivity and autoreactivity of ACS22. (A) Protein microarray binding by bnabs was tested to determine polyreactivity and autoreactivity. Representative ProtoArray summary for protein arrays blotted with ACS22 (left), ACS21 (right), or the 151K control mab. The relative mean fluorescence signal intensity (MFI) in the 151K array is plotted on the y-axis, and the MFI of the test Ab array on the x-axis. Each dot represents the average of duplicate array proteins. A diagonal line indicates equal binding by a test mab and 151K. Internal controls for loading of mab and secondary detection reagent were equally bound by Ab pairs (boxes). Dashed lines indicate the signal/background ratio of 5 that is defined as the cut-off for autoreactivity. The Gaussian mean of all array protein displacements is termed the polyreactivity index (PI). A PI value of.21 implies there is a 2-fold-stronger overall binding by the test Ab than the control 151K mab, and is defined as the threshold of polyreactivity. (B) ELISA reactivity against soluble membrane proteins (SMP) and soluble cytoplasmic proteins (SCP) to indicate polyreactivity. mabs 4E1, PG9 and 2F5 were included for comparison. (C) ELISA reactivity against nuclear antigens and (D) against cardiolipins with mab 4E1 for comparison. (E) ACS22 bind to nuclear or cytoplasmic self-antigens in the antinuclear antibody HEp-2 cell (ANA-HEp-2) indirect immunofluorescence assay. mab PG9 was included for comparison. Negative and positive control sera were provided by the manufacturer.

8 b OD 45nm 1.5 1F11 IgG SOSIP.664 SOSIP 1. gp12 gp Ab µg/ml d OD 45nm Response Difference (RU) VRC1 36% ANC ACS22 + ACS % PGT % -2 2 PGT % 2 35O % -2 25% % 3BC % ACS22 + 3BC315 ACS O22 ACS22 + PGT151-3 ACS22+ PGT145 ACS22 + 8ANC ACS22 + VRC VRC1 + ACS mock 1 ACS22 6 JRFL wt f 3 JRF uncleaved 5 11 Time (s) MFI MFI 3 1 2G12 - cell surface binding BG55 SOSIP S241N 1 WT uncleaved Foldon uncleaved NFL GnT-/N88Q N234S S241N N611/N625Q/N637Q ACS22 - cell surface binding BG55 SOSIP ACS22 - BG55 SOSIP e 1-3 PGT151 - BG55 SOSIP c 9 PGT145 - BG55 SOSIP VRC1 - BG55 SOSIP Time (s) SOSIP.664 trimer gp12-gp41ecto protomer gp12 monomer Time (s) PGT121 - BG55 SOSIP.664 2G12 - BG55 SOSIP.664 OD 45nm 6 Response Difference Response difference (RU) (RU) a 3 Response difference (RU) ACS22 - BG55 99% 1 98% 39% 2 68% 1 71% -1 8ANC195 + ACS22 PGT151 + ACS O22 + ACS22 PGT145 + ACS22 3BC315 + ACS Time (s)

9 Supplementary Figure 5. Characteristics of the ACS22 epitope. (A) ACS22 binding to D7324-tagged BG55 gp12 monomers and SOSIP.664 trimers in ELISA (data displayed is representative of 3 individual experiments). (B) Binding of ACS22 to BG55 SOSIP.664 trimers (green), gp12-gp41 ECTO protomers (blue) and gp12 monomers (pink) was measured using SPR (data displayed is representative of 2 individual experiments). (C) Binding of ACS22 and control bnabs to BG55 SOSIP.664 trimers (blue) and BG55 SOSIP.664 S241N trimers (purple) (data displayed is representative of 2 individual experiments). (D) Binding of ACS22 and control bnabs to BG55 SOSIP.664 mutants N88Q, N234S, S241N, N611Q+N625Q+N637Q, GnT-/- produced BG55 SOSIP.664, uncleaved BG55 gp14 (RRRRRR cleavage site replaced by SEKS) with and without a foldon trimerization domain, and uncleaved BG55 native-flexible-linker (NFL) trimer in ELISA (data displayed is representative of 3 individual experiments). (E) Binding of 2G12 and ACS22 to cell surface associated JR-FL wild-type gp16 or a gp16 that cannot be cleaved (REKR replaced with SEKS), as expressed as the mean fluorescence intensity on the y-axis (data displayed is representative of 3 individual experiments in duplo). (F) SPR analysis of the competitive binding of the indicated bnabs to chip-immobilized, His-tagged BG55 SOSIP.664 trimers. ACS22 was injected for 2s, followed by a second bnab (as indicated on each plot) for an additional 2s (data displayed is representative of 2 individual experiments). Percentage of residual binding of the bnab is indicate in each graph.

10 a 87 D1295.8m.2D6 WKWGIMLLGILMICSAAEQLWVTVYYGVPVWKEATTTLFCASDARAYDTEVHNVWATHACVPTDPNPQEVVLKNVTENFNMWKNNMVEQMHEDIISLWDQSLKPCVKLTPLCVTLNCTDLGNATNT...NATNTTNSSRGTMEGGEIKNC D1295.8m.2D N G R-T---TSG----NNSN-G-I-----M--- D1295.8m.2G E S K----R S AMC E R S D1295.8m.2D6 SFNITTSMRDKVQKEYALFYKLDVVPIKNDN..TSYRLISCNTSVITQACPKVSFEPIPIHYCAPAGFAILKCNDKKFNGTGPCTNVSTVQCTHGIRPVVSTQLLLNGSLAEEEVVIRSANFTDNAKIIIVQLNKSVEINCTRPNNNTRK D1295.8m.2D ED K D1295.8m.2G RI GK AMC D1295.8m.2D6 D1295.8m.2D7 D1295.8m.2G9 AMC11 D1295.8m.2D6 D1295.8m.2D7 D1295.8m.2G9 AMC11 SIHI..GPGRAFYTTGEIIGDIRQAHCNISGTKWNDTLKQIFVKLKEQFGNKTIVFNHSSGGDPEIVMHSFNCGGEFFYCNSTQLFNSTW..NDTTGSNYTGT...IVLPCRIKQIVNMWQEVGKAMYAPPIKGQIRCSSNITGLILI G E V TW NYTGT V V D--P R V FP FPPR RDGGTNRSE...NTEIFRPGGGDMRDNWRSELYKYKVVKIEPLGIAPTKAKRRVVQREKRAVGIGAVFLGFLGAAGSTMGAASMTLTVQARLLLSGIVQQQNNLLRAIEAQQHLLQLTVWGIKQLQARVLAVERYLKDQQLLGIWGCSGK ----K-E K K----.D K K D1295.8m.2D6 LICTTAVPWNTSWSNKSYNQIWNNMTWMEWEREIDNYTSLIYTLIEDSQNQQEKNEQELLELDKWASLWNWFDITKWLWYIKIFIMIVGGLIGLRIVFTVLSIVNRIRQGYSPLSFQTPLPTPRGPDRPEGIEEEGGERDRDRSDRLVTG D1295.8m.2D L D1295.8m.2G T E AMC D1295.8m.2D6 D1295.8m.2D7 D1295.8m.2G9 AMC11 LLALIWVDLRSLCLFSYHRLRDLLLIVTRIVELLGRRGWGVLKYWWNLLQYWSQELRNSAVSLLNATAIAVAEGTDRAIEVLQRAFRAILHIPVRIRQGLERALV F-----A K V---S V F V---S V F V---S V ! b H66R 31 C1 A316W V1 V1 V2 V2 C2 V3 C3 V4 C4 V5 C5 R6 Q567K L543Q I559P HR1 HR2 AMC11 SOSIP.v4.2 A51C S-S T65C c Marker" AMC11"SOSIP.v4.2" Marker" AMC11"SOSIP.v4.2" Marker" AMC11"SOSIP.v4.2" d 25" 15" 1" 75" 669" 44" 232" AMC11 SOSIP.v4.2 +"DTT" 8"DTT" e f AMC11 SOSIP.v4.2 3Å g EU AMC11 SOSIP.v4.2 M5 M6 M7 M8 M9 6 5 AMC11 SOSIP.v Minutes

11 Supplementary Figure 6. Biochemical and biophysical properties of PGT145-purified AMC11 SOSIP.v4.2. (A) Sequence alignment of the 3 isolated viral variants of patient D1295 (also called ACS2) at time point 8 months post seroconversion and the consensus of these 3 sequences, called AMC11. (B) Linear representation of the SOSIP v4.2 Env construct, with modifications to the original AMC11 consensus sequence indicated in red. (C) The AMC SOSIP v4.2 proteins were analyzed by SDS-PAGE under reducing (+DTT) and non-reducing (-DTT) conditions, and by BN-PAGE. Bands corresponding to gp12 (cleaved) and gp14 (uncleaved) proteins on the SDS-PAGE gel, and to trimers on the BN-PAGE gel, are indicated. (D) SEC profile of PGT145-purified AMC11 SOSIP.v4.2 proteins on a Superose 6 column. Retention volume is plotted on the x-axis and 28nm absorption on the y-axis. (E) Negativestain EM analyses showing 2D class averages of AMC11 SOSIP.v4.2 trimers. (F) DSC analyses of AMC11 SOSIP.v4.2 with the thermal denaturation midpoint temperature (T m values) of the peaks calculated using a two-state model. The temperature is plotted on the x-axis and the heat energy in μj/s on the y-axis. n=1. (G) Glycan profile of AMC11 SOSIP.v4.2 trimers. The peaks corresponding to the oligomannose glycans (Man 5-9 GlcNAc 2 ) are marked M5 to M9, and their peak areas (as a percentage of total glycans) are illustrated by the various blue-shaded sectors of the pie charts. The remaining peaks correspond to complex and hybrid-type glycans; these peak areas are indicated by grey-shaded sectors on the pie charts. The retention time is plotted on the x-axis and the HILIP-UPLC signal on the y-axis. n=1. The percentages of Man 5-9 GlcNAc 2 and Man 9 GlcNAc 2 glycans are listed in Supplementary Table 5.

12 AMC11 SOSIP.v4.2/PGV4 A a 348 micrographs micrograph motion correction projection image polishining 3D refinement 5 nm reference-free picking reference-free 2D classification classes 3, 4, 5, 6 (C3 symmetry) B b 1 Å reference: 67, projection images 3D refinement c C Relative angular distribution of projection images 3D classification classes 3, 4, 5, 6 (53,5 projection images) 3D refinement Supplementary Figure 7. Cryo-EM raw data quality, flow diagram of image processing, and resolution assessment of final density. (A) Data processing flow diagram. (B) The resolution of the AMC11 SOSIP.v4.2/PGV4 reconstruction was determined to be 6.2 Å according to the gold standard FSC.143 cutoff. (C) Angular distribution of projection images indicates the particles were isotropically dispersed. Slight over-representation of particular orientations are indicated by the tall red bars.

13 a b 3Å 3Å c ~21 Å d 9 Top view Side view Supplementary Figure 8. NS-EM refinement. NS-EM reconstruction of AMC11 SOSIP.v4.2 trimers in complex with ACS22 Fab. (A) Reference-free 2D class averages and (B) 2D back projections of the final model. (C) The estimated resolution using an FSC cutoff of.5 is 21 A. (D) Top and side views of the final 3D model.

14 a Response Difference (RU) 3 1 ACS VRC1 1 5 PGT ACS22 + ACS22 ACS22 + VRC1 ACS22+ PGT PGT151-2 ACS22 + PGT AMC11 SOSIP.v O22-2 ACS O BC315 8% 8% 41% 7% 3% 54% ACS22 + 3BC % 99% 26% 1% 55% -1-2 VRC1 + ACS22 PGT145 + ACS22 PGT151 + ACS22 35O22 + ACS22 3BC315 + ACS Time (s) b 3 2G12 - AMC11 SOSIP.v4.2 3 VRC1 - AMC11 SOSIP.v4.2 OD 45nm WT E83A E87A N88Q N229K N241K A512W I515W G516W S528W PGT145 - AMC11 SOSIP.v4.2 PGT151 - AMC11 SOSIP.v4.2 ACS22 - AMC11 SOSIP.v OD 45nm Supplementary Figure 9. ACS22 interactions with autologous Env. (A) SPR analysis ofthe competitive binding of the selected bnabs to chip-immobilized, His-tagged AMC11 SOSIP.v4.2 trimers. ACS22 was injected for 2 s, followed by a second bnab (as indicated on each plot) for an additional 2 s (data displayed is representative of 2 individual experiments). Percentage of residual binding of the bnab is indicate in each graph. (B) Binding of ACS22 and control bnabs to various AMC11 SOSIP.4.2 trimers measured by ELISA (data displayed is representative of 3 individual experiments).

15 Supplementary Table 1: Binding and neutralization properties of mabs isolated from patient D1295. BG55 SOSIP.664 SOSIP.664 D368R Binding (ELISA) a Neutralization Tier 1 (IC5) b Neutralization Tier 2 (IC5) FACS sort c BG55 gp12 gp12 ΔV3 SF162 Bal Q23env17 92RW JRFL JRCSF 92BR2 MGRM- C26 d 92TH21 94UG13 93IN95 BG55 AMC11 HIV Bait mab ID IgHV gene B B A A B B B B C AE A C A B ACS261 IgHV1-18 ± ± ± ± ND ND ND ND ND ND ND ND ND ND ND ND ND ND 94UG13 gp12 ACS271 IgHV1-58 ± ± ± ± ND ND ND ND ND ND ND ND ND ND ND ND ND ND BG55 SOSIP.664 ACS251 IgHV ± ± < >25 > >25 >25 >25 >25 >25 >25 ND >25 >25 MGRM-C26 gp12 ACS252 IgHV1-69 ± ± ± ± ND ND ND ND ND ND ND ND ND ND ND ND ND ND MGRM-C26 gp12 ACS21 IgHV >25 >25 >25 >25 > >25 >25 >25 >25.54 >5 >25 BG55 SOSIP.664 ACS22 IgHV >25 >25 >25 >25 < >25 >25 >25 < BG55 SOSIP.664 ACS23 IgHV >25 >25 >25 >25 >25 < >25 >25 >25 <.1 >25 ND BG55 SOSIP.664 ACS24 IgHV >25 >25 >25 >25 < >25 >25 >25 >25 ND >25 >25 BG55 SOSIP.664 ACS25 IgHV >25 >25 >25 >25 > >25 >25 >25 >25 <.1 >25 ND BG55 SOSIP.664 ACS231 IgHV ± ± 1.11 >25 >25 >25 >25 < >25 >25 >25 <.1 >25 ND BG55 SOSIP.664 ACS221 IgHV ND ND ND ND ND ND ND ND ND ND ND ND ND ND BG55+MGRM-C26 ACS241 IgHV ± ± ± 3.46 >25 >25 >25 >25 >25 >25 >25 >25 >25 >25 ND >25 >25 Not determined ACS242 IgHV ± >25 > >25 >25 >25 >25 >25 >25 ND >25 >25 MGRM-C26 gp12 ACS243 IgHV4-34 ± - ± ± ND ND ND ND ND ND ND ND ND ND ND ND ND ND BG55+94UG13 ACS244 IgHV4-34 ± - ± ± ND ND ND ND ND ND ND ND ND ND ND ND ND ND All ACS211 IgHV < >25 > >25 >25 >25 >25 >25 >25 ND >25 >25 All ACS212 IgHV <.2.26 >25 >25.21 >25 >25 >25 >25 >25 >25 ND >25 >25 All The binding and neutralization properties of the 17 isolated mabs from elite neutralizer D1295, which have both high-level IgG production and strong binding to BG55 SOSIP.664. Binding was assessed using D7324-capture ELISA with BG55 SOSIP.664, BG55 SOSIP.664 D368R, BG55 gp12 and BG55 gp12 ΔV3. Neutralization capacity of tier 1 and tier 2 viruses of the mabs was assesses in the TZM-bl assay. a Binding to BG55 SOSIP.664, BG55 gp12 or mutant; strong binding in green (+), modest binding in orange (±) and no binding (-) in red. b Neutralization titers as IC 5 are determined for tier 1 and tier 2 viruses. ND: not determined. Strong neutralization is colored in green, intermediate neutralization in orange and low to no neutralization in red. c Proteins that were bound by the memory B cell from which the mab was cloned. d Virus subtype of the virus above.

16 Supplementary Table 2: Neutralization of 75 virus isolates by ACS22 and other bnabs IC5 IC8 Virus Clade ACS22 PG9 PGV4 PGT121 PGT151 PGT143 VRC26-8 ACS B > >1 >1 >1 QH B >1 > >1 >1 >1 SC B > >1 >1 PVO.4 B.43 > TRO.11 B >1 > >1.996 >1 >1 AC1..29 B >1.97 > >1 >1 RHPA B.12 > >1.237 REJO B > >1 >1 TRJO B >1 > CAAN5342.A2 B.439 > >1 >1 >1 WEAU_d15_41_517 B (T/F).49 ND ND ND ND ND ND _11_C3_161 B (T/F) > >1 >1 >1 154_7_TC4_1499 B (T/F) >1 ND ND ND ND ND ND >1 156_1_TA11_1826 B (T/F) >1 ND ND ND ND ND ND >1 112_11_TC21_3257 B (T/F) > >1.8 >1 >1 624_8_TA5_4622 B (T/F) > >1 >1 >1 6244_13_B5_4576 B (T/F) >1 > >1 >1 >1 > _14_D3_4589 B (T/F) >1 >1 > >1 >1 >1 SC5_8C11_2344 B (T/F) > >1 >1 Du422.1 C > > ZM214M.PL15 C.51 > >1.7 >1 ZM249M.PL1 C 1.65 ND ND ND ND ND ND >1 ZM53M.PB12 C > >1 ZM19F.PB4 C ND >1 >1 ND CAP21.2..E8 C >1 > >1 HIV C >1 ND ND ND ND ND ND >1 HIV C > >1 Ce393_C3 C (T/F) >1.13 >1 >1 >1 >1.131 >1 Ce21_F5 C (T/F) >1 >1.21 >1 >1 >1 >1 >1 Ce73154_2A2 C (T/F) >1.13 > F_C1G C (T/F) >1 >1 >1.41 >1 >1.26 >1 249M_B1 C (T/F) >1 > > E5(Rev-) C (T/F).563 ND ND ND ND ND ND Ce _1B3 C (T/F) ND >1.124 >1 ND CNE19 BC > >1.27 >1 CNE2 BC >1.61 >1.1 >1 >1 >1 >1 CNE21 BC ND ND CNE17 BC > >1 CNE3 BC ND ND ND ND ND ND >1 CNE52 BC > >1 >1 >1 CNE53 BC.1 ND ND ND ND ND ND.38 CNE58 BC ND.29.2 > ND Q23.17 A > >1 Q461.e2 A > >1 >1 >1.126 >1 Q769.d22 A > >1 >1 >1 >1 >1 26.v5.c36 A >1 >1 >1 T CRF2_AG > > > CRF2_AG > >1 >1 >1.16 > CRF2_AG >1 T25-4 CRF2_AG >1.2 > >1 T CRF2_AG >1 >1.738 >1.13 >1 >1 >1 T278-5 CRF2_AG >1 ND ND ND ND ND ND > c1 CRF1_AE.21 >1 >1 >1 > C18.c3 CRF1_AE > >1 > >1 R2184.c4 CRF1_AE >1 ND ND ND ND ND ND >1 R1166.c1 CRF1_AE > >1 > >1 R3265.c6 CRF1_AE >1 >1 >1 >1 >1 >1 C3347.c11 CRF1_AE >1.46 >1 >1 >1 >1 >1 >1 C4118.c9 CRF1_AE >1 >1.1.2 >1 CNE5 CRF1_AE > >1 > >1 BJOX9.2.4 CRF1_AE >1.7 > BJOX CRF1_AE (T/F) >1 >1 >1 >1.136 BJOX1.6.2 CRF1_AE (T/F) >1 >1 >1 >1 >1 BJOX CRF1_AE (T/F) >1 >1 >1 >1 >1.149 BJOX CRF1_AE (T/F).5 ND ND ND ND ND ND.111 X1193_c1 G > >1 X1254_c3 G ND >1 >1 ND X288_c9 G >1 >1 >1.3 >1 >1 >1 >1 P1981_C5_3 G >1.4 >1 X1632_S2_B1 G > >1.61 >1.5 >1 A7412M1.vrc12 D ND ND c1 D > >1.1 >1 >1 > v2.c59 CD >1 >1 >1 > v1.c2 CD ND > >1 >1 >1 ND 6811.v7.c18 CD >1 ND ND ND ND ND ND >1 89-F1_2_25 CD >1.42 >1 >1 >1.1.1 >1 331.v1.c24 AC.217 ND ND ND ND ND ND >1 641.v3.c23 AC >1 ND ND ND ND ND ND >1 654.v4.c1 AC >1 > v4.c1 AC >1 > v3.c3 ACD >1 > >1 >1 >1 >1 313.v3.c1 ACD.685 ND ND ND ND ND ND >1 Strong neutralization is colored in red, intermediate neutralization in yellow and weak neutralization in green. ND; not determined

17 Supplementary Table 3: JR-CSF virus mutants affecting ACS22 neutralization. JRCSF mutant fold reduction ACS22 neutralization fold reduction PGT121 neutralization JRCSF mutant fold reduction ACS22 neutralization fold reduction PGT121 neutralization JRCSF mutant fold reduction ACS22 neutralization fold reduction PGT121 neutralization D78A.5 1. N262A* ND.9 N386A*.3 ND N8A N276A* N461A*.4 ND V182A 1.5 ND T283A.3 ND E462A.8 ND D185A K293A.3 ND S463A* N186A E322A 3.5 ND E466A K187A 1. ND S334A.5 ND L467A 4.7 ND N188A*.8 ND Q337A.6 ND M626A T19A*.1 ND N339A*.9 ND E647A.6.8 K236A T341A*.9 ND T244S.6.9 R35A.2 ND kifunensine.3 ND ACS22 neutralization of JR-CSF virus mutants containing the listed single amino-acid substitutions, with those that result in deletion of a glycan site indicated with an *. The values shown are the fold-reduction in neutralization titer for each mutant compared to the wildtype virus (n=3). The color-scheme is based on the intensity of the red shading increasing as the neutralization capacity decreases. The highest ACS22 concentration tested was 1 µg/ml and the IC 5 value for ACS22 was.83 µg/ml. Supplementary Table 4: Neutralization IC 5 values of early autologous clones and the consensus AMC11 virus. D1295.8m.2D6 D1295.8m.2D7 D1295.8m.2G9 AMC11 VRC PGT PGT PGT O G ACS22 >5 > Strong neutralization is colored in red, intermediate neutralization in yellow and weak neutralization in green.

18 Supplementary Table 5: Biophysical properties of AMC11 SOSIP.v4.2 trimers. Trimer morphology Thermostability Glycan profile NS-EM DSC HILIC-UPLC Native-like trimers (%) Closed NL trimers (%) Open NL trimers (%) Tm ( C) Man 9 GlcNAc 2 glycans (%) Oligomannose glycans (%) AMC11 SOSIP.v The biophysical properties of 293F cell-expressed, PGT145-puriefied AMC11 SOSIP.v4.2 were assessed using NS-EM to dertermine native-like trimer formation, DSC to quantify thermostability (T m ), and glycan profiling to measure glycan content. Supplementary Table 6: The binding parameters for ACS22 in SPR analysis. ACS22 Fab BG55 SOSIP.664 (n=2) AMC11 SOSIP.v4.2 (n=2) kon koff Kd(app) (1/Ms) (1/s) (nm) ± ± ± 1.1 ± ± ± ±.7 ± Sm K d(app) ; apparent dissociation constant K on ; the on-rate constant K off ; the off-rate constant S m ; the stoichiometric number