Dr. Roger Paredes

1. MICROBIOME & HIV
Roger Paredes
Unitat VIH i irsiCaixa
Hospital Germans Trias i Pujol
Badalona

2. FLSida: Isabel Bravo, Pep Coll, Carla Estany, Cristina Herrero
BCN Checkpoint: Jorge Saz
IrsiCaixa: Bea Mothe, Jorge Carrillo, Christian Brander, Julià
Blanco, Bonaventura Clotet
Vall d’Hebron: Manel Crespo, Jordi Navarro, Ariadna Torrela
UVIC-UCC: Malu Calle
Marc Noguera
Yolanda
Guillén
Cristina
Rodríguez
Chiara
Mancuso
Muntsa
Rocafor
t
Maria
Casadell
à
Marion
a
Parera
Rocío
Bellid
o
Javier
River
a
Gràcies!

3. Diverse microbiomes, but also diverse
factors fueling the HIV pandemic
Europeans
Americans
Asians
N=33
Sanger
Danes
N=85
Illumina
US
N=145
454
Arumugam et al, Nature 2011 Piot et al, Lancet 2015

4. Brenchley J M et al. J Exp Med 2004;200:749-759
HIV infection destroys the GALT
% CD4+ T-cells

5. Microbial translocation in HIV
b
Secretory IgA
HEV
HEV
Increased TNF
production
Lossof tight
junctions
Decreased IgA
production
Increased TNF
and IFNg production;
failed ionic balance HIV infection
Villus
CD8+
Tcell in ltration
Crypt
Decreased ratio of
villusheight to
crypt depth
Abnormal enterocyte
di erentiation
CD4+
Tcell loss;
preferential
TH17 cell lossEnterocyte
apoptosis
small
at is
cal
din, a
uten. It
during
d foods.
trophy,

6. Immune senescence
Inflammageing
CV disease
Metabolic disorders
Osteoporosis
Chronic kidney
disease
Liver disease (NASH)
Cognitive problems
HIV persistence
Is there an HIV-associated disbyosis
perpetuating systemic inflammation?
GALT +
GI wall
damage
Bacterial
translocation
HIV
Disbyosis?
GUT BLOOD
Inflammation
Residual HIV
replication
Other
diseases, Tx
Other viral
infections

7. Barcelona: Test dataset
156 subjects
• 27 HIV negative, mostly MSM
• 129 HIV positive with ≠ phenotypes
• 100 (64%) MSM, 41 (26%) HTS, 15 (10%)
PWID
Stockholm: External Validation
83 subjects
• 7 HIV negative, all HTS
• 76 HIV positive (including 3 EC)
• 18 (22%) MSM, 55 (66%) HTS, 10 (12%)
PWID
Barcelona: Internal
Validation: same
analysis 1 month later
* Study reviewed by our Community Advisory

8. BCN patients
N=156
Microbiota
(16S, Shotgun
Illumina Sequencing)
Exome genotyping
Diet
HIV
negative
n=27
Elite control
n=8
Viremic
control
n=11
Naive
CD4>500
n=15
Early ART
n=13
Inmune
concordant
n=55
Inmune
discordant
n=18
Late
Presenter
n=11
• 18-60 years old
• BMI 18.5-30
• no ATB in last 3 months
(except LP)
• no surgery or active GI
diseases
No ART
VL< 50 c/mL >1year
No ART
VL 50-2000 c/mL >1year
No ART
VL >2000 c/mL
CD4+> 500 cells/mm3
ART within 6 mo from HIV-1 infection
ART > 2 years
VL <50 c/mL
CD4+ >500 cells/mm3
ART > 2 years
VL <50 c/mL
CD4+ <300 cells/mm3
No ART
CD4+ <200

9. Late
presenter
Discordan
t
Concorda
nt
Early
treated
ART naive
Viremic
controller
Elite
controller
CD4+ counts (cells/mm3) 100 263 761 785 701 783 940
0
100
200
300
400
500
600
700
800
900
1000
Subject’s characteristics (IV)
CD4+ count by HIV phenotype
AIDS
Increased
Clinical
Complications
Same 3-year
mortality than
HIV-negatives
On ART
N=11 N=18 N=53 N=13 N=15 N=11 N=8

10. IGC
GGs
Fecal samples
Shotgun Sequencing16S rDNA sequencing
Trimmomatic QC
Mothur, QIIME MetaPhlAn2
IGC
mapping
OTU clustering Taxonomy classification Gene content
bwa
de novo
assembly
Virulence factors
and AB resistance
contentOTU table Phylotypes Bacterial
content
Ecological
parameters
(Richness &
Diversity)
Bacterial content
PICRUST
Gene content
MOCAT
Gene annotation
MetaGeneMark
bwa
Virulence
factors and Abx
resistance
Metadata
Gene content
VFDB
ORDINATION: PCoA, NMDS
KEGG
Functional profile
MetaHIV
catalogue
ARDB
VEGAN,
PHYLOSEQ,
ADE4
BIOMARKERS: LEFSe, BinNeg
PREDICTORS: LASSO, MetaDistance, Dirichlecht Multinomial
Methods

11. a-diversity

12. a-diversity: HIV phenotypes

13. Spearman correlation by genus abundance
Only significantvalues are shown,Holm’s-corrected p<0.05

14. Distribution of the study population
2 4 6 8 10
0.100.200.300.40
Silhouette coefficient
Number of clusters
Silhouettecoefficient
−1.0 −0.5 0.0 0.5
−0.50.00.5
NMDS + PAM (Bray−Curtis)
NMDS 1
NMDS2
ellipse: 95%

15. ADONIS test of factors explaining
distance variation (Bray-Curtis)
Variable Univariate Multivariate*
R2 p-value R2 p-value
HIV Risk Group 0.367 <0.001 0.373 <0.001
Gender 0.152 <0.001 0.008 0.407
Feces consistency 0.049 <0.001 0.009 0.335
Residency 0.044 0.014 0.008 0.421
Ethnicity 0.037 0.022 0.018 0.156
HIV serostatus 0.035 0.003 0.011 0.054
Altered Abdominal transit 0.031 0.023 0.005 0.238
* terms added sequentially (first to last)

16. Risk Practice and HIV Status
16S rDNA sequencing (Bray-Curtis)
0.5
0.0
0.5
HIV_Status
negative
positive
Risk_Group
hts
msm
pwid
Genus level
−1.0
−0.5
0.0
0.5
1.0
NMDS2
HIV_Status
negative
positive
Risk_Group
hts
msm
pwid
Species level
−0.5
0.0
0.5
−1.5 −1.0 −0.5 0.0 0.5 1.0
NMDS1
NMDS2
Genus level − Month 0
−0.5
0.0
0.5
−2 −1 0 1
NMDS1
NMDS2
Genus level − Month 1
N=155
Ellipse 95%
N=108
Ellipse 95%

17. Concordance BCN0 – BCN1
−1.0 −0.5 0.0 0.5
−0.50.00.51.0
Procrustes errors
Dimension 1
Dimension2
PROTEST Statistics:
Procrustes Sum of Squares (m2): 0.3475
Correlation in a symmetric Procrustes
rotation: 0.8078
Significance: 0.001
Permutation: free
Number of permutations: 999
Jackson, D. A. Protest - A Procrustean
Randomization Test of Community
Environment Concordance. Ecoscience 2,
297–303 (1995).

18. Distribution of the study population
frequency
0.00.20.40.60.81.0
Distance from Bacteroides representative
Prevotella
Faecalibacterium
Bacteroides
Lachnospiraceae_unclassified
Succinivibrio
Alloprevotella
Ruminococcaceae_unclassified
Blautia
Parabacteroides
Alistipes
Others
No MSM
MSM
Positive
Negative
Cluster 1 (Bacteroides)
Cluster2 (Prevotella)
Distance fom Bacteroides representative

19. Distribution of the study population

20. Dysbiosis by Risk group

21. Magnitude of associations

22. Diet evaluation
2 complementary questionnaires given to the study
participant. Dietitian/nutritionist explained the
questionnaires before and reviewed them after
completion
• Nutrient amounts & energy intake
• Short-term questionnaire
• Prospective consecutive 3 to 5 day registry (including one
weekend day) of all food & drinks specifying as much as
possible ways of preparation & amounts
• Energy and nutritional calculations: Software ADN
(“Anàlisi de Dades Nutricionals”), v1.0 beta 2. Compiles
existing & validated nutritional composition tables.
• Food consumption frequency assessment
• Mid-term questionnaire
• Analysed as portions of food/week
• Capitulo 6: Métodos de valoración del consumo alimentario. En: Nutrición y Dietética Clínica. J. Salas-Salvadó, A. Bonada, R. Trallero, ME Saló, R. Burgos. 2ª edición.
Elsevier Masson. 2008
• Anex 12. Cuestionarios para realizar una historia dietetica. En: Nutrición aplicada y dietoterapia. M. Muñoz, J. Aranceta, I.Garcia-Jalón. 2ª edición. Ediciones Universidad de
Navarra, EUNSA. Pamplona 2004.
Nutritional data were transformed to adjust for
total energy consumption by fitting a linear
model and taking residuals as new nutritional
values (mean= 0 and SD=1)
E.g.: The relative contribution of 100 gr of proteins is
different if the total energy intake is 5000 kJ or 9000 kJ
Energy Folate
Increased energy intake MSM
vs. Non-MSM and Prevotella
vs. Bacteroides. P<0.01 in both
comparisons

23. Diet

24. By HIV phenotype

25. By Risk Group

26. By Microbiome Cluster

27. GENE RICHNESS

28. Gene Richness & HIV
150 155 160
KEGG pathways
richness (PICRUST−Humman)
0e+00
1e−06
2e−06
250000 500000 750000
Gene richness
density
Gene richness shotgun IGC
250 275
ys from enzimes in KEGG (IGC)
0e+00
1e−04
2e−04
3e−04
9000 11000 13000 15000
density
eggNOG richness (IGC)
0e+00
3e−04
6e−04
9e−04
3000 3500 4000 4500 5000
Enzimes
density
Enzimes KEGG richness (PICRUST)
0.00
0.04
0.08
0.12
145 150 155 160
KEGG pathways
density
KEGG Pathways richness (PICRUST−Humman)
0e
1e
2e
density
0.00000
0.00025
0.00050
0.00075
0.00100
3500 4000 4500 5000 5500
Enzimes
density
Enzimes KEGG richness (IGC)
0.000
0.005
0.010
0.015
0.020
225 250 275
Metabolic pathways
density
Metabolic pathways from enzimes in KEGG (IGC)
0e
1e
2e
3e
density

29. All samples LGC HGC
HIV-1 Negative 11 16
HIV-1 Positive 92 37
MSM samples LGC HGC
HIV-1 Negative 8 15
HIV-1 Positive 46 31
Only MSM
Chi square, p=0.05
Rrisk = 1.72 (1.0-3.5)
Other factors linked to LGC:
• Older ager, female gender, caucasian ethnicity, Profile (viremic controller and
HIV-negative higher in HGC), Nadir CD4+, creatinin, FA, fibrinogen and
platelets.
HIV infection linked to LGC
All subjects
Chi square, p<0.001
Rrisk = 1.75 (1.1-3.1)

30. Gene Richness & HIV phenotype

31. Kruskal, p = 0.003
P = 0.05
P = 0.005
P = 0.02
Gene Richness & nadir CD4

34. High vs. low gene counts by function

35. -0.1 1.0
n/a n/a
1.0 1.0
n/a n/a
10.5 0.006
n/a n/a
0.87** 0.5
n/a n/a
-3.6 0.3
n/a n/a
0.8 0.3
n/a n/a
6.92 0.02
n/a n/a
1.62 0.7
n/a n/a
2.7 0.4
n/a n/a
0.77** 0.2
n/a n/a
5.25 0.2
n/a n/a
1.8 0.7
n/a n/a
0.6 0.8
n/a n/a
Figure 4. Several genes important in neutrophil
function were significantly upregulated among
MSM who engaged in CRAI ≤24 hours prior
indica ng a possible role for neutrophils in response to
CRAI.
Figure 5. Gene Set Enrichment Analysis using the MSigDB
database (www.broadinstitute.org/gsea/msigdb/collection.jsp )
was conducted to identify significant immunologic pathways that
were differentially expressed in the rectal mucosa comparing
MSM who engaged in CRAI ≤24 hours prior and MSM who
abstained from CRAI for ≥ 72 hours. Pathways of interest
included DNA Replication (A), TH17 vs. naïve CD4 T cells (B),
and Reactome Antigen Processing and Cross Presentation (C).
p=0.011
A
B C
e MSM engaging in CRAI showed a distinct mRNA gene expression and CD8+ T lymphocyte profile as
engaged in anal intercourse.
te inflammatory response to CRAI, possibly driven by neutrophils and monocytes, that may be the result of
g intercourse with exposure to pro-inflammatory semen and the gut microbiota.
e frequency of IFNg producing CD8+ T cells increased, suggesting a pro-inflammatory microenvironment.
enses and inflammation, also likely play an important role in the acute and chronic response.
determine how these factors may impact HIV susceptibility or mucosal vaccine response in the rectal
The Effect of Condomless Receptive Anal Intercourse on the Rectal Mucosa in MSM
Colleen F. Kelley MD, MPH; Chandni Duphare, Hyun-Woo Lee; Kirk Easley; Jing Yang; Gregory Tharp;
Mark J. Mulligan MD; Patrick S. Sullivan DVM, PhD; Steven Bosinger PhD; Rama R. Amara PhD
Center for AIDS Research at Emory University, Atlanta, GA, USA
q The risk of HIV transmission per exposure event for receptive
anal intercourse is 1.38%, more than 12-fold higher than other
routes of sexual transmission. Nearly 70% of HIV transmissions
among MSM are attributed to rectal exposure.
q The vast majority of mucosal HIV transmission research has
been conducted in the female genital tract or in non-human
primates and extrapolated to rectal transmission among MSM.
q Sexual intercourse with semen exposure is known to cause an
inflammatory reaction with influx of HIV target cells in the
female genital tract.
q Therefore, we conducted a study to examine the rectal mucosal
effects of condomless receptive anal intercourse (CRAI) in
MSM
BACKGROUND:
METHODS:
Colleen F. Kelley MD, MPH
The Hope Clinic
500 Irvin Ct. Ste. 200
Decatur, GA 30030
colleen.kelley@emory.edu
404-712-1823
Acknowledgements: K23AI108335 (Kelley) and Emory CFAR P30 AI050409
Poster #286
q We enrolled 41 HIV negative MSM who were engaging in CRAI
with an HIV negative partner and 21 men who had never engaged
in anal intercourse (controls) into this study.
q Peripheral blood and rectal biopsy samples were collected from 2
time points separated by a median of 9 weeks. MSM abstained
from CRAI for ≥72 hours prior to visit 1. MSM engaged in CRAI
≤24 hours prior to visit 2.
q PBMCs were isolated by Ficoll density and rectal MMCs by
collagenase digestion. Cells were stained with surface antibodies
for CD4+ and CD8+ cell phenotyping and stimulated with PMA/
Ionamycin (Figure 1) to evaluate cytokine expression and
analyzed with Flowjo software. Linear mixed effects models were
used to examine differences in CD4+ and CD8+ cellular
phenotype and cytokine expression between MSM engaging in
CRAI and controls.
q From a subset of subjects (18 MSM and 10 controls), RNA was
extracted from one rectal pinch biopsy and sequenced with
Illumina HiSeq. Data were analyzed with DESeq to examine
differential mRNA gene expression and Gene Set Enrichment
Analysis with MSigDB database.
RESULTS:
CONCLUSIONS:
Figure 1. Representative gating strategy for PBMCs and rectal MMCs for CD4+ and CD8+ cell phenotyping (A) and mitogen
stimulation (B) to evaluate cytokine expression.
A.
Table 2. Results of rectal MMC phenotyping for MSM engaging in CRAI and men who never engaged in AI
(controls) demonstrate overall lower CD38 expression on CD4+ T cells in MSM engaging in CRAI and higher
Ki67 expression on CD8+ T cells. Upon mitogen stimulation, rectal CD8+ T cells from MSM engaging in CRAI
overall produced higher levels of IFNγ and TNFα compared to controls. With the exception of a decline in
CD38 expression on CD4+ T cells, there were no significant differences with timing of CRAI among MSM. Data
were analyzed by repeated measure modeling controlling for time, time*group interaction, and technician.
Immunologic
Indices
n
Baseline
Mean
(95% CI)
n CRAI mean
(95%CI)
Overall model-
based mean
p-
value
CRAI:
V1 vs. V2
mean
difference
or ratio**
p-
value
Rectal Memory
CD4+ cells
%CCR5+
CRAI 41 61.0(56.7,65.3) 38 62.8(58.4,67.1) 61.9(58.4,65.4) -0.1 1.0
Control 21 60.2(54.3,66.1) 17 64.2(56.5,71.9) 62.0(56.9,67.1) 1.0 n/a n/a
*%Ki67+
CRAI 41 2.5(2.0,3.0) 38 2.5(1.9,3.3) 2.5(2.0,3.0) 1.0 1.0
Control 21 2.1(1.4,3.2) 17 1.9(1.4,2.6) 2.0(1.5,2.6) 0.2 n/a n/a
%CD38
CRAI 40 45.0(38.2,51.9) 36 34.5(28.8,40.2) 39.7(34.6,44.8) 10.5 0.006
Control 21 49.1(41.5,56.7) 17 50.0(38.6,61.4) 49.5(42.4,56.7) 0.03 n/a n/a
*%CCR5+Ki67+
CRAI 41 1.5(1.2,1.8) 38 1.7(1.3,2.3) 1.58(1.31,1.91) 0.87** 0.5
Control 21 1.4(0.9,2.1) 17 1.2(0.8,1.7) 1.28(0.92,1.78) 0.3 n/a n/a
%α4β7+
CRAI 38 57.0(51.7,62.4) 38 53.8(48.1,59.4) 55.4(51.5,59.3) -3.6 0.3
Control 21 58.0(51.7,64.3) 17 60.0(49.7,70.2) 59.0(53.1,64.9) 0.3 n/a n/a
Rectal Memory
CD8+ cells
*%Ki67+
CRAI 41 3.7(3.0,4.7) 38 4.5(3.3,6.2) 4.1(3.3,5.0) 0.8 0.3
Control 21 3.5(2.3,5.1) 17 2.0(1.4,3.0) 2.6(1.9,3.7) 0.04 n/a n/a
%CD38+
CRAI 40 71.1(66.2,76.0) 36 64.2(57.7,70.7) 67.6(62.6,72.6) 6.92 0.02
Control 21 69.4(60.9,77.9) 17 70.4(61.0,79.9) 69.9(63.2,76.6) 0.5 n/a n/a
Stimulated rectal
CD4+ cells
%IFNγ+
CRAI 35 21.6(17.9,25.3) 32 22.6(18.8,26.5) 22.1(19.2,25.1) 1.62 0.7
Control 18 16.8(12.1,21.5) 16 24.2(18.1,30.3) 20.5(16.5,24.5) 0.5 n/a n/a
%TNFα+
CRAI 35 30.8(25.7,35.9) 32 33.6(28.2,39.0) 32.2(28.5,35.9) 2.7 0.4
Control 18 27.1(20.1,34.1) 16 31.9(25.1,38.8) 29.5(25.2,33.9) 0.4 n/a n/a
*%IL-17+
CRAI 35 2.0(1.7,2.7) 32 2.5(1.8,3.6) 2.2(1.7,2.9) 0.77** 0.2
Control 18 1.4(0.9,2.3) 16 1.7(1.1,2.6) 1.6(1.1,2.2) 0.1 n/a n/a
Stimulated rectal
CD8+ cells
%IFNγ+
CRAI 34 58.3(51.3,65.2) 32 53.0(45.7,60.3) 55.6(50.1,61.2) 5.25 0.2
Control 18 37.6(28.5,46.7) 16 45.0(33.1,56.7) 41.3(33.3,49.3) 0.005 n/a n/a
%TNFα+
CRAI 33 40.8(34.4,47.2) 32 39.0(32.5,45.5) 39.9(35.1,44.7) 1.8 0.7
Control 18 31.3(21.9,40.6 16 32.5(24.7,40.3) 31.2(26.1,37.6) 0.04 n/a n/a
%IFNγ+TNFα+
CRAI 33 25.8(20.9,30.8) 32 25.2(19.2,31.2) 25.5(21.0,30.0) 0.6 0.8
Control 17 14.2(7.9,20.4) 14 17.9(11.6,24.3) 16.1(11.4,20.7) 0.005 n/a n/a
* report the geometric mean (95%CI) **ratio of visit 1 and 2 geometric means reported
Figure 2. Global mRNA gene expression data from 1 rectal pinch
biopsy was generated by RNASeq and analyzed with DESeq.
Principal components analysis showed distinct separation between
MSM who abstained from CRAI for ≥ 72 hours (blue ellipse) and
MSM who engaged in CRAI ≤ 24 hours prior (red ellipse). Data from
both control visits (green and purple ellipses) clustered together
demonstrating similarity in mRNA gene expression over time in the absence
of CRAI.
Figure 3. Fifty-four genes were differen ally expressed (q<0.05,
0.5<FC<1.5) between MSM who abstained from CRAI for ≥ 72 hours
and controls, including those implicated in ssue remodeling (CAPN8,
CAM1, COL6A3, COL8A1, LOX, PAPPA, MMP3), cell prolifera on (C8orf4,
EDNRB, FGF7, HGF, ZNRD1, FGFBP1) and immune ac va on (CD109,
CD80, CRLF1, LILRA3, LILRA6, MASP1, OASL). An addi onal 19 unique
genes were differen ally expressed when comparing MSM who engaged
CRAI ≤24 hours prior vs. MSM who abstained from CRAI for ≥72 hours and
also implicated genes important in ssue remodeling (HYAL1, MMP1), cell
prolifera on (AGR2, CDC6, CGREF1, WDR4) and immune ac va on
(S100A9). *genes were downregulated
Figure 4. Several genes important in neutrophil
function were significantly upregulated among
MSM who engaged in CRAI ≤24 hours prior
indica ng a possible role for neutrophils in response to
CRAI.
Figure 5. Gene Set Enrichment Analysis using the MSigDB
database (www.broadinstitute.org/gsea/msigdb/collection.jsp )
was conducted to identify significant immunologic pathways that
were differentially expressed in the rectal mucosa comparing
MSM who engaged in CRAI ≤24 hours prior and MSM who
abstained from CRAI for ≥ 72 hours. Pathways of interest
included DNA Replication (A), TH17 vs. naïve CD4 T cells (B),
and Reactome Antigen Processing and Cross Presentation (C).
p=0.011
A
B C
q The rectal mucosa of HIV negative MSM engaging in CRAI showed a distinct mRNA gene expression and CD8+ T lymphocyte profile as
compared to men who have never engaged in anal intercourse.
q Our data show evidence of an acute inflammatory response to CRAI, possibly driven by neutrophils and monocytes, that may be the result of
microtrauma/mucosal injury during intercourse with exposure to pro-inflammatory semen and the gut microbiota.
q With chronic exposure to CRAI, the frequency of IFNg producing CD8+ T cells increased, suggesting a pro-inflammatory microenvironment.
Th17 cells, critical for mucosal defenses and inflammation, also likely play an important role in the acute and chronic response.
q Further research will be needed to determine how these factors may impact HIV susceptibility or mucosal vaccine response in the rectal
mucosa of HIV negative MSM and how the gut microbiota may contribute.
Table 1. Description of the study population. MSM engaging in CRAI were slightly older than
MSM who never engaged in AI (controls). Almost all MSM engaging in CRAI reported use of
lubricants for sex and many reported enema use.
Characteristic MSM engaging in
CRAI
(n=41)
Men never engaged
in AI
(n=21)
p-value
Median age in years
(25th
, 75th
)
28 (25.5, 33.9) 24 (23.5, 30.0) 0.02
Race n(%)
White 33 (80.0) 14 (66.7)
Black 6 (14.6) 2 (9.5)
Other 2 (4.9) 5 (23.8) 0.11
Lubricant use n(%) 39 (95.1) n/a
Enema use n(%) 18 (43.9) n/a
Median CRAI episodes
in previous month
(25th
, 75th
)
5 (4,8) n/a
!
B. Rectal MMC
s mula ons with
PMA/Ionamycin
IFNγ IL-2 IL-4 IL-17 IL-21 TNFα
CD3
Cytokine
IFNγ IL-2 IL-4 IL-17 IL-21 TNFα
Cytokine
CD4+
CD8+
Non-s mulated
S mulated
Non-s mulated
S mulated
Forward
sca er
Sidescaer
Live/Dead
Sidescaer
Sidescaer
CD3 CD8
CD4
12
92.9
50.1
66.8
20
0.242 0.0191
0.0573
0.0319 0.0382 0.102
0.341 0.149
0.0213
0.107 0.0426 0.256
27 27.1
1.21
4.3 0.411 36.4
50.6 9.8
0.98
2.43 0.948 32.4
CD3
69.3
Lymphocytes
99.8
99.6
46.2
27.3
44
56 65.234.7
17.3
0.0715 25.4
74.10.422
0.136 18.2
81.40.357
0.0143 1.75
97.80.479
0.1 17.3
82.20.393
1.22 0.543
18.379.9
0.563 0.449
36.762.3
40.5 16.1
2122.4
Blood
23.4 88
73.6
33.2
16.1
82.4
17.6 8217.8
Rectal
MMCs
24.6
0.508 64.2
34.11.16
0.145 40
58.31.52
0.0483 4.54
93.71.62
0.628 46.8
51.51.04
2.22 2.37
38.756.6
0.501 2.71
80.915.9
8.72 59.6
247.72
Live/Dead
Sidescaer
Sidescaer
CD45Forward
sca er
Sidescaer
CD8
CD4
CD45RO
CCR7
CD4+ Cells CD8+ Cells
CD45RO
CCR7
Gated on total memory CD4+ cells (excluding CCR7+CD45RO-)
Gated on total memory CD8+ cells
(excluding CCR7+CD45RO-)
Gated on total memory CD4+ cells (excluding CCR7+CD45RO-)
Gated on total memory CD8+ cells
(excluding CCR7+CD45RO-)
CD45RO
α4β7
CD45RO
CCR5
CD45RO
Ki67
CD45RO
CD38
CCR5
Ki67
CD45RO
Ki67
CD45RO
CD38
Forward
sca er
Sidescaer
Live/Dead
Sidescaer
Sidescaer
CD45 CD8
CD4
CD45RO
CCR7
CD45RO
CCR7
CD45RO
α4β7
CD45RO
CCR5
CD45RO
Ki67
CD45RO
CD38
CCR5
Ki67
CD45RO
Ki67
CD45RO
CD38
A B
“Evidence of an acute inflammatory response to CRAI, possibly driven by
neutrophils and monocytes, that may be the resuly of microtrauma/mucosal
injury during intercourse with exposure to pro-inflammatory semen and the gut
microbiota”
DNA replication
Reactome & Antigen processingTh17 vs naive CD4+ cells

36. Conclusions
• Human activities are potential confounders of microbiome studies: sexual
practice seems to be one of them.
• The fecal microbiota of gay men in Europe is richer and has a distinct
composition.
• HIV-1 infection is independently associated with reduced gut bacterial
richness, more so in immune discordant subjects.
• Low gene counts correlate with immune, inflammatory and metablic
parameters
• Interventions to increase gut bacterial richness might offer novel avenues to
improve HIV-1-associated immune dysfunction.

37. "There are trillions of bacterial cells living on my body, but I still
get lonely sometimes."
Gràcies!