HIV-1 subtype C in Ethiopia and India HIV-1 subtype C Isolated - - PowerPoint PPT Presentation

HIV-1 subtype C in Ethiopia and India

HIV-1 subtype C

• Isolated 1988, Ethiopia (Sönnerborg et al, 1990)
• Dominating subtype globally: 50%
• Biological characteristics

– 3 NF-κB sites (Sönnerborg, 1995) – 4 NF-κB strains in India (Bachu, 2012) – Higher set-point after acute infection (Novitisky 2011) (?) – HIV drug resistance pattern, K65R – Insertions p6 during ART failure – Co-receptor usage

HIV in Ethiopia

• First HIV seropositive identified in 1984
• 2001: 2 million adults + 200,000 children; 2011: 800,000 persons
• HIV prevalence

– 1% 1989, 5.2% 1996; 5.6% 2005, 3.5% 2007, 2.6% 2009/11 – Urban centers: 11.5% 2003, 5.5% 2009 – Rural areas: 4% 2003, 1.4% 2009

• In 2005, a national ART program was rolled out
• In 2011, 336 000 treated
• Ethiopia is one of the six sub-Saharan African countries showing

a decline of >25% in new HIV infections

• Delatorre. PLoS One, July 2012

1,981 HIV-1C pol sequences: ”South African clade” ”Heterogeneous clades” ”East African clade” ” Subtype C` or Ethiopian 2 clade

Two HIV-1C clades in East Africa

• Delatorre. PLoS One, July 2012

Based on 1,981 HIV-1C pol sequences:

• >70% belongs to a

single regional-specific monophyletic group,

• riginating from

Burundi in 1960

• A second major

Ethiopian subtype C lineage

• 170 plasma from treatment naive Ethiopian AIDS-patients, 2007-2008
• Advanced immunosuppression (average CD4 cells 104, range 4-199)
• The RNA was reverse transcribed and then PCR-amplified using specific

primers for the pol-gene (ref HXB2 inner primers: forward 2135-2158, reverse 3315-3338).

• Sequencing of the env-gene was attempted on the samples for which pol-

sequencing had been succesful (ref HXB2 inner primers: forward 7003-7021, reverse 7648-7668).

Ethiopian samples

pol-seq env-seq

HIV-1 subtypes in Ethiopia

Rega RIP pol (n= 127) C 127 126 C/A1/F2

• 1

env (= 102*) C 73 82 A1 1 1 ”close to C” 9

• *19 of these had good sequence only for V3 and could not be subtyped or used for

phylogenetic analysis

• 1. http://dbpartners.stanford.edu/RegaSubtyping/
• 2. http://www.hiv.lanl.gov/content/sequence/RIP/RIP.html

• 83 Ethiopian samples with good sequence for both pol (average seq length

939 bp) and env (average seq length ≈480 bp)

• References from LANL pre-made alignments [6], 45 ref-seq for different

subtypes + 28 subtyp C seq from different geographic regions

• Phylogenetic analysis was performed in BEAST [7], evolutionary model

GTR+I+Γ, run for 10 million generations, sampling every 10 000 gen.

• The posterior probability value (corresponding to bootstrap in non-Bayesian

phylogenetics) was 1 (100%) for the differentiation between the C and C’ clusters for both pol and env in this analysis.

Phylogenetic analysis (prel), methods

• 6. http://www.hiv.lanl.gov/content/sequence/NEWALIGN/align.html
• 7. http://beast.bio.ed.ac.uk/

Tree, pol-seq C’ C

H A, 01_AE G, 02_AG B D F K J

Blue= ref subtype C-seq Cyan=C-seq from different regions Green=Ethiopian seq from this study which are either C or C’ for both pol and env Other colours – potentially recombinant viruses

Tree, env-seq C’ C

01_AE A, 02_AG G B D F K J

Blue= ref subtype C-seq Cyan=C-seq from different regions Green=Ethiopian seq from this study which are either C or C’ for both pol and env Other colours – potentially recombinant viruses

H J

Out of 83 Ethiopian samples sequenced for pol and env, the following distribution was found in the C and C’ clades:

• 34 viruses (41%): C-clade for both pol and env (C - C)
• 28 viruses (34%): C’-clade for both pol and env (C’- C’)
• ne C-C and one C’-C’ virus moved substantially within the clades and

associated with strains from Europe/India and South America rather than with other Ethiopian strains for the env-region

• 10 viruses (12%): C’ for pol and C for env (C’- C)
• 8 viruses (9.6%): C for pol and C’ for env (C - C’)
• One strain was found to be a C-A1 recombinant

Prel phylogenetic analysis, results

• The two Ethiopian lineages have resulted from independent founder

strains, originated in eastern and southern African regions

• In total, 18/83 (23%) Ethiopian strains analyzed fall into distinctly

different clades for pol and env -> recombinants between C and C’ appear to be common in Ethiopia

• In contrast, only 1/28 (3.6%) of the randomly selected subtype C strains

from other geographic regions outside Ethiopia appeared to be a recombinant, this was a C’- C strain from Israel

• These results indicate that circulating recombinant forms C-C’ and C’-C

are spreading in Ethiopia alongside the non-recombinant clades C-C and C’-C’

Conclusion

• 102 V3-sequences were tested by:

₋ geno2pheno [4] ₋ WebPSSM (three algorithms) [5] ₋ CoRSeqV3-C [3] ₋ (the 11/25+net charge rule, 2011)

• No test repeat
• In the cases of ambiguous base pairs (double peaks) all possible translations

were analyzed. ₋ most sequences (73/102) had only one translation while 13 had 2, and 16 had between 3 and 32 possible translations (3: N=1; 4: N=7; 8: N=2; 16: N=3; 28: N=1; 32: N=2).

Co-receptor tropism, methods

• 3. http://www.burnet.edu.au/facilities/8; Gorry et al.
• 4. http://coreceptor.bioinf.mpi-inf.mpg.de/
• 5. http://indra.mullins.microbiol.washington.edu/webpssm/

Co-receptor tropism, results

Total: 102 V3 All three algorithms G2P FPR 20 WebPSSM (sinsi C) CoRSeqV3-C R5 70 (68.6%) 82 (80.4%) 83 (81.4%) 89 (87.3%) X4 6 (5.9%) 20 (19.6%) 19 (18.6%) 13 (12.7%) Discordant 26 (25.5%)

NA NA NA

X4 prevalence: WebPSSM sinsi B: 4/102 X4R5: 5/102 11/25: 2/102

Co-receptor tropism, results

(70 samples consequently designated as R5 not shown)

Sample # V3-LOOP Number of alternative translations Geno2Pheno, coreceptor CoRSeqV3-C WebPSSM sinsi-C >Et111 CTRPNNNTRRGIGIGPGHTFYTTGQIIGDIRKAYC 1 5 CXCR4 CXCR4 X4 N= 6 >Et86 CTRPNNNTRKSIGIGPGRAFYATGDIIGNIRQAHC 1 5,4 CXCR4 CXCR4 >Et136 CTRPSNNTRRNIGIGPGQTFFAPGTIIGDIRRAYC 1 7,8 CXCR4 CXCR4 >Et191 CTRPNNNTRKGIRIGPGQMFYAAGEIIGNVRQAFC 1 13,2 CXCR4 CXCR4 >Et124 CTRPNNNTRKSIRIGPGQAF[HY]ATG[KNRSEDG]IIGDIRQA[HY]C 28 14,4 CXCR4 (9) CXCR4 (3) >Et73 CTRPSNNTRKSVEIGPGRAIYATGDIIGDIRQAHC 1 15,4 CXCR4 CXCR4 >Et107 CTRPNNNTSKSIRIGPGQAFYATERIIGNIRQAYC 1 4,6 CCR5 CXCR4 >Et62 CTRPNNNTRKGIGIGPGQMFYATDAIIGDMRDNC 1 4,8 CCR5 CXCR4 >Et21 CTRPNNNTIESIRIGPGQAFYAT[TR]RIIGDIRQAYC 2 5 CCR5 CXCR4 (2) >Et150 CTRPNNNTRKSMRIGPGQTFYATGKIIGDIRKAYC 1 5 CXCR4 CCR5 >Et49 CTRPGNNTRRSVRIGPGQAFYTTGEIIGDIRRAHC 1 5,3 CCR5 CXCR4 >Et78 CTRPNNNTRKGIGIGPGQTFYAAEEIIGNIRNAYC 1 6,9 CCR5 CXCR4 >Et68 CVRPGNNTRKSIRIGPGQAFYAEGGIIGDVRQAHC 1 10,5 CCR5 CXCR4 >Et35 CARPGNNTRKSMRIGPGQTFYATGDIIGNIRKAHC 1 11,7 CXCR4 CCR5 >Et20 CMRPGNNRRKSIRIGPGQTFYATGEIIGDIRQAYC 1 3,7 CCR5 CCR5 ? N= 8 >Et82 CARPGNNTRKSTRIGPGQTFYATGDIIGDIRQAHC 1 6,7 CCR5 CCR5 >Et105 CTRP[NS]NNTRKSMRIGPGQTFYAMGVIGDIRQAYC 2 8,7 CCR5 CCR5 >Et31 CTRPNNNTRKSIRIGPGQAFYTTNIIGDIRQAHC 1 20,9 CCR5 CXCR4 >Et88 CERPNNNTRKSIRIGPGQAFYTTGQIIGDIRQAHC 1 23,6 CCR5 CXCR4 >Et96 CTRPNNNTRKSIRIGPGQAFYATGDIIGDIRHAFC 1 51,9 CXCR4 CCR5 >Et142 C[TM]RPNNNTRESIGIGPGQTLFAIGAIIGDIRQAHC 2 84,5 CCR5 CXCR4 (2) >Et128 CTRLNNNTRKSIRIGPGQTFYATGGIIGNIRLAHC 1 95,5 CXCR4 CCR5 >Et92 C[IM]RPNNNTRKSMRIGPGQTFYATGEIIGDIRQAHC 2 15,4 CCR5 CCR5 R5? N= 10 >Et166 C[TM]RPNNNTRKS[IV]RIGPGQAFYATG[ED]I[IV]G[ND]IRQAHC 32 16,4 CCR5 CCR5 >Et149 CTRPNNNTRKSVRIGPGQTYYATGDIIGDIRRAHC 1 19,4 CCR5 CCR5 >Et81 CTRPNNNTRKSI[RG]IGPGQ[TA]FYATGEIIG[ND]IRKA[HY]C 16 21 CXCR4 (2) CCR5 >Et63 CTRPTNNTRKSVRIGPGQ[TA]FYATGEVIGDIR[KN]AHC 4 21,8 CCR5 CXCR4 (2) >Et135 CTRPSNNTRTSMRIGPG[QP*S]TFYATGDIIGDIRKAYC 3 25,3 CXCR4 (2) CCR5 >Et152 C[TI]RPGNNTR[KR]SVRIGPGQ[AV]FY[TA]TGEIIGDIREAHC 16 25,6 CCR5 CXCR4 (2) >Et47 CTRPN[NS]NTRKSMRIGPGQTFYATGDI[IV]G[KNED]IRQA[HY]C 32 63,1 CXCR4 (4) CCR5 >Et84 CIRPSNNTR[KE]S[IM]RIGPGQTFYATG[ND]I[TI]GDLRQAHC 16 79,5 CCR5 CXCR4 (4) >Et97 CTRP[NS]NNTRRGIRIGPGQTFFAIGEIIGDIRQAHC 2 90,3 CCR5 CXCR4 (1)

Numbers within parantheses show the number of possible translations that were categorized as X4-tropic

X4? N= 8

Conclusion, Co-receptor tropism

• CCR5-inhibitors are likely to be effective in around 75%-85% of ART-naïve

Ethiopian subtype C infected AIDS-patients

• The classification of the co-tropism was disconcordant in 25.5% of the

sequences between G2P, WebPSSM sinsi C and CoRSeqV3-C

• The patients with X4- strains had lower CD4-counts (average±stdev:

86.2±65.1), compared with the undetermined (91.8±48.0) and R5-using strains (108.0±49.7), however, these differences were not statistically significant.

HIV in India

• First HIV seropositive identified in 1986, Tamil Nadu
• Estimate of 2.4 million infections as of 2009
• A prevalence of 0.31%
• Despite the low prevalence, India is the third largest country
• Currently 334 000 on ART

Figure 2. State wise adult HIV-1 prevalence in India (2009 estimates). Data adapted from National AIDS Control Programme Phase III, State Fact Sheets, March 2012

Manipur Andra Pradesh Tamil Nadu Karnataka Madhya Pradesh Punjab Haryana 168 HIV-1 seropositive subjects from 7 different provinces

Indian HIV-1 infected subjects, 2007-2011

Methods

• HIV-1 subtypes were determined

– using two or three genes, gag, pol, and env – maximum likelihood phylogenetic tree, RIP 3.0, Rega 2.0;

• Estimating tMRCA

– Bayesian coalescent-based approach was used to reconstruct the time of introduction of the Indian HIV-1C. – three viral genes were used (gag, pol, env)

HIV-1 subtypes in India

• 6.0% BC
• 4.2% A1C
• Closest B: China/Thai
• Closest A1: East Afr
• A large magnitude
• f genetic diversity
• at least five

different B-C and three different A1- C recombinants

tMRCA of HIV-1C reported from different regions

Multiple enhancer motifs in HIV-1C

• Three NFkB bindings sites were described in

Ethiopian HIV-1 subtype C strains in 1995 (Johansson, Sherefa, Sönnerborg; AIDS Res Hum Retro)

Schematic model portraying a novel strategy the NF-κB mediated replication competence.

Subtype B

Distribution of HIV-1C with 4 NkB binding sites in India

Conclusions

• HIV-1 subtype C was confirmed as the dominating strain in

India.

• A significant increase in recombinant strains was observed,

indicating an evolving heterogeneous viral epidemic in India.

• The Indian HIV-1 subtype C epidemic originated nearly four

decades ago from a single or few genetically related African lineages, and since then largely evolved independently and formed a separate monophyletic clade segregating from the larger subtype C clade

• A significant increase in strains with 4 NFkB binding sites is

seen in India of so far unknown clinical relevance