2018 Conference on Sergio Ita, PhD Retroviruses and UC San Diego Opportunistic Infections SMRT SEQUENCING OF FULL-LENGTH POL AMPLICONS TO 9500 Gilman Drive 0679 Abstract # 2188 INVESTIGATE HIV-1 DRUG RESISTANCE La Jolla, CA 92093, USA Poster # 556 Sergio Ita1, Ben Murrell1, Gemma Caballero2, Caroline Ignacio1, Sanjay Mohan1, Venkatesh Kumar1, [email protected] March 7, 2018 Doug Richman1,2, Davey Smith1,2, Gabriel A. Wagner1, San Diego, CA USA 1University of California San Diego, La Jolla, CA, USA, 2Veterans Affairs San Diego Healthcare System, San Diego, CA, USA
Background Table 1. Participant Characteristics Results
Time from EDI (months)* ART start • Antiretroviral therapy (ART) reduces HIV viral loads to undetectable levels by Participant ID Log HIV RNA (copies/ml) • We produced full-length pol sequences from multiple participants. targeting the three enzymes encoded by HIV-1 pol : (months after EDI)* Pre-ART sampling Post-ART sampling • The consensus sequences of HQCSs from two participants • Reverse Transcriptase (RT), Protease (PR), and Integrase (IN) 222-47n 14 13 5.61 • Broad implementation of ART will be key to reducing worldwide HIV-1 transmission, matched bulk Sanger sequencing (Figure 1). 24 5.60 but will also likely lead to increased prevalence of drug resistance mutations • Sequences submitted to the Stanford HIV drug resistance (DRMs). 222-3yn 2 2 6.92 database revealed mutations at DRM sites in 6 participants 61 4.55 • Current DRM testing relies on Sanger sequencing which is unable to detect variants (Figure 2). that comprise less than 20% of the viral population. 222-5d9 1 1 NS 6.99 • Deep sequencing platforms such as MiSeq (Illumina) can lower the limit of detection 222-46m 31 6 4.77 • We were able to link mutations at sites of drug resistance from all to 1% of the viral population, but can only sequence short reads and may not 42 4.60 four drug classes targeting RT, PR, and IN. capture DRM across the whole gene. 222-4d6 4 3 NS 6.00 Figure 3. Full-length pol HQCSs produced by SMRT PacBio 222-47c 3 NS 34 4.86 Objective 222-433 22 3 NS 4.88 sequencing from multiple participants 222-3mf N/A 26 NS 4.68
To use the long-read Single Molecule Real Time (SMRT, Pacific *N/A=Not applicable, NS=Not sequenced
222_5d9_LP2_02102017.fastqseq_11_33
222_5d9_LP2_02102017.fastqseq_31_24
222_5d9_LP2_02102017.fastqseq_64_36
222_5d9_LP2_02102017.fastqseq_7_24
222_5d9_LP2_02102017.fastqseq_38_12
222_5d9_LP2_02102017.fastqseq_1_289
222_5d9_LP2_02102017.fastqseq_23_14
222_5d9_LP2_02102017.fastqseq_17_11 222_5d9_LP2_02102017.fastqseq_65_25
222_5d9_LP2_02102017.fastqseq_54_11 222_5d9_LP2_02102017.fastqseq_18_42
Biosciences®) deep sequencing platform to investigate linkage of 222_5d9_LP2_02102017.fastqseq_20_31
222_5d9_LP2_02102017.fastqseq_14_33
222_5d9_LP2_02102017.fastqseq_52_31
222_5d9_LP2_02102017.fastqseq_30_18
222_5d9_LP2_02102017.fastqseq_6_37 222_5d9_LP2_02102017.fastqseq_67_11 222_3yn_LP84.fastqseq_7_22
222_3yn_LP84.fastqseq_5_13 222_5d9_LP2_02102017.fastqseq_22_25
222_3yn_LP84.fastqseq_10_24
222_5d9_LP2_02102017.fastqseq_66_23 222_5d9_LP2_02102017.fastqseq_48_14 222_3yn_LP84.fastqseq_2_18 222_3yn_LP84.fastqseq_1_17
222_3yn_LP84.fastqseq_4_33 222_3yn_LP84.fastqseq_8_15
222_5d9_LP2_02102017.fastqseq_39_37 222_3yn_LP1_02102017.fastqseq_3_17
222_3yn_LP84.fastqseq_9_27 222_5d9_LP2_02102017.fastqseq_49_27 222_3yn_LP1_02102017.fastqseq_1_7418
222_3yn_LP1_02102017.fastqseq_2_16
222_3yn_LP84.fastqseq_3_32222_3yn_LP84.fastqseq_6_10 222_5d9_LP2_02102017.fastqseq_34_41
Figure 1. Full-length pol sequences match bulk sequences 222_5d9_LP2_02102017.fastqseq_63_12
HIV-1 DRMs across full-length pol sequences. 222_5d9_LP2_02102017.fastqseq_15_39 222_46m_LP82.fastqseq_3_22
222_46m_LP87.fastqseq_3_10 222_5d9_LP2_02102017.fastqseq_32_50 222_46m_LP82.fastqseq_1_11
222_46m_LP82.fastqseq_2_16 222_5d9_LP2_02102017.fastqseq_13_27 222_46m_LP87.fastqseq_2_20
222_46m_LP87.fastqseq_1_29
222_5d9_LP2_02102017.fastqseq_50_23
222_5d9_LP2_02102017.fastqseq_24_23
222_5d9_LP2_02102017.fastqseq_19_23
222_5d9_LP2_02102017.fastqseq_62_19 222_5d9_LP2_02102017.fastqseq_60_14
P767_LP95.fastqseq_5_81
222_5d9_LP2_02102017.fastqseq_5_11
222_5d9_LP2_02102017.fastqseq_46_33 P767_LP95.fastqseq_9_127 222_5d9_LP2_02102017.fastqseq_29_43 P767_LP95.fastqseq_13_356
P767_LP95.fastqseq_29_40 222_5d9_LP2_02102017.fastqseq_3_17 P767_LP95.fastqseq_44_47
P767_LP95.fastqseq_1_242P767_LP95.fastqseq_19_54 222_5d9_LP2_02102017.fastqseq_58_25 P767_LP95.fastqseq_34_48
222_5d9_LP2_02102017.fastqseq_16_27 P767_LP95.fastqseq_25_33
222_5d9_LP2_02102017.fastqseq_40_40 P767_LP95.fastqseq_38_51 222_5d9_LP2_02102017.fastqseq_61_17 P767_LP95.fastqseq_64_15
P767_LP95.fastqseq_10_83
222_5d9_LP2_02102017.fastqseq_25_19 P767_LP95.fastqseq_67_44 222_5d9_LP2_02102017.fastqseq_55_19 P767_LP95.fastqseq_8_33
PP6_LP93.fastqseq_5_12 P767_LP95.fastqseq_60_10 PP6_LP93.fastqseq_3_59 PP6_LP93.fastqseq_4_38 222_5d9_LP2_02102017.fastqseq_2_1145 PP6_LP93.fastqseq_2_128
PP6_LP93.fastqseq_1_2155 P767_LP95.fastqseq_11_21
222_5d9_LP2_02102017.fastqseq_10_26 Methods 222_5d9_LP2_02102017.fastqseq_8_12 HXB2
P767_LP95.fastqseq_40_12 HIV-1 HXB2 222_5d9_LP2_02102017.fastqseq_33_43
P767_LP95.fastqseq_65_74
222_5d9_LP2_02102017.fastqseq_59_22 P767_LP95.fastqseq_54_54 222_5d9_LP2_02102017.fastqseq_56_15 P767_LP95.fastqseq_59_11
Reference P767_LP95.fastqseq_23_121
222_5d9_LP2_02102017.fastqseq_44_20 P767_LP95.fastqseq_37_149
222_5d9_LP2_02102017.fastqseq_12_29 P767_LP95.fastqseq_46_34
222_5d9_LP2_02102017.fastqseq_57_39 P767_LP95.fastqseq_63_22 222_5d9_LP2_02102017.fastqseq_21_27 P767_LP95.fastqseq_2_743
• Participants were selected from the San Diego Primary Infection Resource 222_5d9_LP2_02102017.fastqseq_4_40 P767_LP95.fastqseq_66_171 P767_LP95.fastqseq_58_17
222_5d9_LP2_02102017.fastqseq_27_32 P767_LP95.fastqseq_62_76
222_5d9_LP2_02102017.fastqseq_45_27 P767_LP95.fastqseq_28_24 222-3yn- PacBio 222_5d9_LP2_02102017.fastqseq_36_27
P767_LP95.fastqseq_26_93 222_5d9_LP2_02102017.fastqseq_42_33 P767_LP95.fastqseq_27_68
P767_LP95.fastqseq_39_10 Consortium (SD PIRC) who had: 222_5d9_LP2_02102017.fastqseq_53_40 P767_LP95.fastqseq_24_40
222_5d9_LP2_02102017.fastqseq_35_21 P767_LP95.fastqseq_48_11
222-3yn - Sanger 222_5d9_LP2_02102017.fastqseq_37_21 P767_LP95.fastqseq_52_62
222_5d9_LP2_02102017.fastqseq_47_35 P767_LP95.fastqseq_36_153
222_5d9_LP2_02102017.fastqseq_51_37 P767_LP95.fastqseq_33_10
• At least two longitudinal study visits (one before and one after ART initiation) 222_5d9_LP2_02102017.fastqseq_43_13 P767_LP95.fastqseq_51_29 222_5d9_LP2_02102017.fastqseq_28_42 P767_LP95.fastqseq_57_148 P767_LP95.fastqseq_15_73
222-5d9– PacBio 222_5d9_LP2_02102017.fastqseq_9_16 P767_LP95.fastqseq_18_38
222_5d9_LP2_02102017.fastqseq_26_12 P767_LP95.fastqseq_30_12 • Blood viral load of at least 1000 copies/ml at the post-ART visit P767_LP95.fastqseq_22_20 222_5d9_LP2_02102017.fastqseq_41_47
PP8_LP94.fastqseq_25_44 P767_LP95.fastqseq_43_76
PP8_LP94.fastqseq_1_112 P767_LP95.fastqseq_21_74
222-5d9 - Sanger PP8_LP94.fastqseq_10_74 P767_LP95.fastqseq_49_22
PP8_LP94.fastqseq_23_55 P767_LP95.fastqseq_4_269 • HIV-1 RNA was extracted from blood plasma samples, cDNA generated, and P767_LP95.fastqseq_7_211
PP8_LP94.fastqseq_18_84
PP8_LP94.fastqseq_27_92 P767_LP95.fastqseq_53_116
PP8_LP94.fastqseq_39_16 P767_LP95.fastqseq_14_53 PP8_LP94.fastqseq_31_91 P767_LP95.fastqseq_16_40
coding regions within pol (HXB2 1736-5074) were PCR-amplified. PP8_LP94.fastqseq_28_10 P767_LP95.fastqseq_70_26
PP8_LP94.fastqseq_38_71 P767_LP95.fastqseq_35_53
PP8_LP94.fastqseq_26_45 P767_LP95.fastqseq_45_11
PP8_LP94.fastqseq_4_19 P767_LP95.fastqseq_47_36
PP8_LP94.fastqseq_19_22 P767_LP95.fastqseq_61_24 • Amplicons were sequenced on a PacBio RSII machine to produce circular PP8_LP94.fastqseq_2_190 P767_LP95.fastqseq_6_156
PP8_LP94.fastqseq_22_13 P767_LP95.fastqseq_17_132 PP8_LP94.fastqseq_12_84 P767_LP95.fastqseq_55_10
PP8_LP94.fastqseq_9_152 P767_LP95.fastqseq_12_14
PP8_LP94.fastqseq_37_10 P767_LP95.fastqseq_41_86
consensus sequences (CCS). P767_LP95.fastqseq_56_11 PP8_LP94.fastqseq_21_29
PP8_LP94.fastqseq_11_34 P767_LP95.fastqseq_3_115
PP8_LP94.fastqseq_32_43 P767_LP95.fastqseq_20_78 PP8_LP94.fastqseq_30_46 P767_LP95.fastqseq_50_20
PP8_LP94.fastqseq_5_82 P767_LP95.fastqseq_42_54
• Fastq files were filtered and processed using in-house variant reconstruction PP8_LP94.fastqseq_7_68 P767_LP95.fastqseq_69_10 PP8_LP94.fastqseq_29_39
PP8_LP94.fastqseq_17_37 P767_LP95.fastqseq_32_30
PP8_LP94.fastqseq_35_37 P767_LP95.fastqseq_31_18
222_47n_LP57.fastqseq_9_13
PP8_LP94.fastqseq_8_39 P767_LP95.fastqseq_68_16 222_47n_LP57.fastqseq_2_24
PP8_LP94.fastqseq_20_78
PP8_LP94.fastqseq_24_71
algorithms, inferring high quality consensus sequences (HQCSs), which were 222_47n_LP57.fastqseq_7_181
PP8_LP94.fastqseq_6_89
222_47n_LP57.fastqseq_5_11
PP8_LP94.fastqseq_36_17 222_47n_LP57.fastqseq_8_16
PP8_LP94.fastqseq_13_26 222_47n_LP57.fastqseq_12_21 222_47n_LP57.fastqseq_10_14 PP8_LP94.fastqseq_15_53PP8_LP94.fastqseq_34_19
PP8_LP94.fastqseq_14_41 222_47n_LP85.fastqseq_29_23
PP8_LP94.fastqseq_33_22 222_47n_LP85.fastqseq_15_23
PP8_LP94.fastqseq_3_15 222_47n_LP57.fastqseq_1_757 used for phylogeny and frequency analysis 222_47n_LP85.fastqseq_26_24
PP8_LP94.fastqseq_16_25 222_47n_LP57.fastqseq_6_17
222_47n_LP57.fastqseq_11_13
222_47n_LP85.fastqseq_25_26
222_47n_LP85.fastqseq_12_65 222_47n_LP85.fastqseq_7_40
• Sequences were analyzed for DRMs using the Stanford HIV Drug Resistance 222_47n_LP85.fastqseq_24_15
222_47n_LP85.fastqseq_27_23
H497_LP96.fastqseq_4_333 222_47n_LP57.fastqseq_3_47
222_47n_LP57.fastqseq_4_19
H497_LP96.fastqseq_18_32 222_47n_LP85.fastqseq_18_30 222_47n_LP85.fastqseq_32_10
H497_LP96.fastqseq_14_34 222_47n_LP85.fastqseq_23_16
H497_LP96.fastqseq_9_134 222_47n_LP85.fastqseq_31_27
H497_LP96.fastqseq_16_36 222_47n_LP85.fastqseq_14_30 222_47n_LP85.fastqseq_16_10
Figure 2. DRM linkage across full-length HIV-1 pol sequences H497_LP96.fastqseq_5_328 222_47n_LP85.fastqseq_9_40 222_47n_LP85.fastqseq_30_25 222_47n_LP85.fastqseq_1_184
H497_LP96.fastqseq_19_22 222_47n_LP85.fastqseq_17_57
222_47n_LP85.fastqseq_5_46
222_47n_LP85.fastqseq_6_24 222_47n_LP85.fastqseq_13_11 H497_LP96.fastqseq_6_143H497_LP96.fastqseq_3_168
Database:(https://hivdb.stanford.edu/). 222_47n_LP85.fastqseq_3_78 222_47n_LP85.fastqseq_21_34 H497_LP96.fastqseq_17_15
222_47c_LP45.fastqseq_2_28 222_47n_LP85.fastqseq_8_38 222_46d_LP83.fastqseq_1_173 222_47c_LP45.fastqseq_1_30
222_47n_LP85.fastqseq_2_26 H497_LP96.fastqseq_15_13 222_47n_LP85.fastqseq_10_41 H497_LP96.fastqseq_7_13
222_47n_LP85.fastqseq_22_14 222_47n_LP85.fastqseq_20_22 222_47n_LP85.fastqseq_19_29
222_47n_LP85.fastqseq_11_30 222_47n_LP85.fastqseq_28_35 H497_LP96.fastqseq_8_40H497_LP96.fastqseq_13_60
222_47n_LP85.fastqseq_4_43 H497_LP96.fastqseq_11_76H497_LP96.fastqseq_2_242 H497_LP96.fastqseq_1_359 H497_LP96.fastqseq_12_32 • Samples also underwent standard bulk sequencing of partial pol (ViroSeq) for H497_LP96.fastqseq_10_109 comparison to HQCSs. 0.0080 Single Molecule Real-Time (SMRT) Sequencing (PacBio ®)
Sites: PR 33, 46, 82, 88 Figure 3. A phylogeny of pol sequences, colored by participant, estimated by maximum likelihood from full-length pol high quality consensus sequence (HQCS). Bubbles represent the proportion of circular consensus sequences that mapped to each HQCS. Inset:A phylogenetic tree of variation at position 184 in reverse transcriptase of 222-47n (pre-ART). We detected the M184I drug resistance mutation on a single outlying haplotype variant, which also had evidence of 18 APOBEC-mediated hypermutations. Sites: NNRTI: 103, 106, 138, 188, 190 Conclusions
o SMRT sequencing is a high-throughput approach to providing full-length sequences of the HIV-1 pol gene. o Sites: NRTI: 67, 70, 74, 184, 210, 215, 219 PacBio HQCSs matched bulk Sanger sequencing of partial pol regions. o Full-length pol sequences provided a method to detect and link DRMs across the entire length of pol.
References Acknowledgments 1. Clutter D.S., Jordan M.R., Bertagnolio S., R.W. Shafer. HIV-1 drug resistance and resistance testing. This work was supported by UCSD CFAR grant A1036214,K23 award MH105231 (G.A.W.), Infect Genet Evol. 2016 Aug 29. NIH/NIAID award R00AI120851 (B.M.), and S.I. is supported by NIH IRACDA K12 GM068524. 2. D.D. Richman. Antiviral drug resistance. Antiviral Res. 2006 Sep;71(2-3):117-21. Sites: INI: 138, 140 3. Gianella S., D.D. Richman. Minority variants of drug-resistant HIV. J Infect Dis. 2010 Sep 01;202(5):657- 66. Figure 2. Linkage analysis of DRM sites across pol. Sequences from 11 participants and 2 panel samples (PP6 and 4. Fisher R., van Zyl G.U., Travers S.A., Kosakovsky Pond S.L., Engelbrech S., Murrell B., Scheffler K., D. Smith. Deep sequencing reveals minor protease resistance mutations in patients failing a protease PP8) with mutations at protease (PI), non-nucleoside RT inhibitor (NNRTI), nucleoside RT inhibitor (NRTI), and inhibitor regimen. J Virol. 2012 Jun;86(11):6231-7. integrase inhibitor (INI) DRM sites were analyzed for evidence of resistance. Within each drug class, common DRM 5. Smith M.L., Murrell B., Eren K., Ignacio C., Landais E., Weaver S., et al. Rapid Sequencing of complete sites were grouped into motifs and represented as different colors. env genes from primary HIV-1 samples. Virus Evolution. 2016 Jul 1.