Eo

VALF

ARRA

KHSRTS

1

n

E8f

I

0h

Journal of Steroid Biochemistry & Molecular Biology 138 (2013) 63– 71

Contents lists available at SciVerse ScienceDirect

Journal of Steroid Biochemistry and Molecular Biology

jo ur nal home page: www.elsev ier .com/ locate / j sbmb

ffect of sex steroid hormones on replication and transmissionf major HIV subtypes

iswanath Ragupathy, Krishnakumar Devadas, Shixing Tang, Owen Wood, Sherwin Lee,rmeta Dastyer, Xue Wang, Andrew Dayton, Indira Hewlett ∗

aboratory of Molecular Virology, Division of Emerging Transfusion Transmitted Diseases, Center for Biologics Evaluation and Research,ood and Drug Administration, Bethesda, MD 20892, USA

a r t i c l e i n f o

rticle history:eceived 25 October 2012eceived in revised form 26 February 2013ccepted 1 March 2013

eywords:IVubtypeseplication kineticsransmission kineticsex hormones

a b s t r a c t

Background: The HIV epidemic is expanding worldwide with an increasing number of distinct viral sub-types and circulating recombinant forms (CRFs). Out of 34 million adults living with HIV and AIDS, womenaccount for one half of all HIV-1 infections worldwide. These gender differences in HIV pathogene-sis may be attributed to sex hormones. Little is known about the role of sex hormone effects on HIVSubtypes pathogenesis. The aim of our study was to determine sex hormone effects on replication andtransmissibility of HIV subtypes.Methods: Peripheral blood mononuclear cells (PBMC) and monocyte derived dendritic cells (MDDC) frommale and female donors were infected with HIV subtypes A–D and CRF02 AG, CRF01 AE, MN (lab adapted),Group-O, Group-N and HIV-2 at a concentration of 5 ng/ml of p24 or p27. Virus production was evaluatedby measuring p24 and p27 levels in culture supernatants. Similar experiments were carried out in thepresence of physiological concentrations of sex steroid hormones. R5/X4 expressions measured by flowcytometry and transmissibility was evaluated by transfer of HIV from primary dendritic cells (DC) toautologous donor PBMC.Results: Our results from primary PBMC and MDDC from male and female donors indicate in the absenceof physiological concentrations of hormone treatment virus production was observed in three clusters;high replicating virus (subtype B and C), moderate replicative virus (subtype A, D, CRF01 AE, Group N)and least replicative virus (strain MN). However, dose of sex steroid hormone treatment influenced HIVreplication and transmission kinetics in PBMC, DCs and cell lines. Such effects were inconsistent betweendonors and HIV subtypes. Sex hormone effects on HIV entry receptors (CCR5/CXCR4) did not correlatewith virus production.

Conclusions: Subtypes B and C showed higher replication in PBMC from males and females and weretransmitted more efficiently through DC to male and female PBMC compared with other HIV-1 subtypes,HIV-1 Group O and HIV-2. These findings are consistent with increased worldwide prevalence of subtype Band C compared to other subtypes. Sex steroid hormones had variable effect on replication or transmissionof different subtypes. These findings suggest that subtype, gender and sex hormones may play a crucialrole in the replication and transmission of HIV.

Published by Elsevier Ltd.

. Introduction

The HIV epidemic is expanding worldwide with an increasingumber of distinct viral subtypes and circulating recombinant

∗ Corresponding author at: Laboratory of Molecular Virology, Center for Biologicsvaluation and Research, Food and Drug Administration, Building 29B, Room 4NN16,800 Rockville Pike, Bethesda, MD 20892, USA. Tel.: +1 301 827 0795;ax: +1 301 480 7928.

E-mail addresses: Ragupathy.Viswanath@fda.hhs.gov (V. Ragupathy),ndira.hewlett@fda.hhs.gov (I. Hewlett).

960-0760/$ – see front matter. Published by Elsevier Ltd.ttp://dx.doi.org/10.1016/j.jsbmb.2013.03.002

forms (CRFs) [1]. Until now, research has been focused primarilyon subtype B, the most prevalent subtype in North America andEurope. However, the global distribution of subtypes and CRFs isvaried and dynamic in nature and could contribute to the emergingdiversity of HIV variants. These variants potentially affect ratesof disease progression [2]. According to UNAIDS [3] almost halfof all HIV-1 infected individuals worldwide are women. Studiescomparing the course of HIV-1 infection between women and

men have demonstrated considerable sex differences in the man-ifestations of HIV-1 disease [4]. Although the rate of progressionto AIDS is similar among men and women, plasma HIV-1 RNAlevels in women are significantly lower than in men [5,6]. These

6 hemistry & Molecular Biology 138 (2013) 63– 71

gusrdsmte

aIgit[iiebswp1Abccb

tvfsgiHcacd

2

2m

aIotorfccfihct(atr

-500 0

0

5000

10000

15000

20000

25000

30000

35000

Subtype B Subtype C CRF01_AE Group -N Subtype D Subtype A MN

Le

ast

Sq

ua

re M

ea

ns

Cluster 1

Cluster 2 Cluster 3

*

***

Fig. 1. Dynamics of HIV-1 subtypes in peripheral blood mononuclear cells (PBMCs).

4 V. Ragupathy et al. / Journal of Steroid Bioc

ender based disparities were confirmed in a long-term followp study [7] from seroconversion to clinical outcomes throughubsequent follow up. This study revealed significant statisticalacial differences present in the earliest stages of infection andisappeared at 6 months of follow up for patients on HAART. In thisame study untreated women had lower viral loads compared withen after 6 months and reported fewer symptoms associated with

he acute retroviral syndrome, which may delay or complicatearly diagnosis in this population.

Physiological hormonal levels fluctuate in the peripheral bloodnd reproductive organs of women during the menstrual cycle [8].t has been hypothesized that during the follicular phase high estro-en levels and higher immunity may have protective effects againstnvading pathogens, while during the luteal phase, high proges-erone levels and reduced immunity may favor microbial invasion9]. These fluctuations in hormone levels are thought to play anmportant role in susceptibility and immune responses to HIV-1nfection in women. In vitro studies demonstrate sex hormones,strogen and progesterone, prevent HIV-1 infection in monocytesut not in lymphocytes suggesting cell specific effects of femaleex hormones [10]. In another study estradiol and progesteroneere shown to regulate HIV-1 replication in PBMC during the mid-roliferative phase by increasing replication, while decreased HIV-

replication was observed during the mid-secretory phase [11].nother in vitro study demonstrated that beta-estradiol inhibitedoth HIV-1 replication in primary human peripheral blood lympho-ytes (PBL) and the antiretroviral efficacy of stavudine (D4T) in PBLultures [12]. These studies suggested that HIV-1 infection coulde modulated either by hormones or host genetic factors.

It is apparent that women are more susceptible to HIV-1 infec-ion, and prognosis rates vary with diverse population of HIV-1ariants and that these gender differences may likely be a selectiveactor for the evolution of HIV-1 variants. Hence a better under-tanding of the biologic factors leading to these differences mayive insight into the potential impact of sex-specific differencesncluding hormonal effects on the replication and transmission ofIV. In this study, we attempted to determine whether the repli-ation of specific major HIV subtypes would differ in pathogenesisnd modulated by sex hormone levels using susceptible in vitroell culture systems including PBMC and MDDC (monocyte derivedendritic cell).

. Results

.1. Dynamics of HIV-1 subtypes in peripheral bloodononuclear cells (PBMCs)

It is known that subtypes could differ in infectivity and mayccount for geographic variations in distribution of HIV subtypes.n our experiments cells were obtained from normal blood donorsf Caucasian origin. To evaluate the replication of different HIV sub-ypes, cells were infected with similar concentrations of virus basedn p24 antigen or p27 antigen levels. Least square measure (Fig. 1)evealed that subtype B replicated well in cells from both males andemales followed by subtype C and other subtypes. We identified 3lusters based on statistically significant differences in replicationapacity. Subtype B and C yielded high p24 values and were classi-ed as a high replicating cluster 1, subtype A, D, CRF01 AE, Group-Nad moderate p24 values compared to cluster 1 (p < 0.05) and werelassified as cluster 2 and MN (strain), a lab adapted strain was clus-

er 3 and showed the least p24 production compared to cluster 1p < 0.001) in PBMC. Other subtypes such as F2, CRF02 AG, Group-Ond HIV-2 did not replicate efficiently (data not shown) suggestinghat either higher input of virus or longer duration of infection wasequired.

HIV subtypes replication in four male and two female donor PBMCs forms threeclusters. Replication of HIV subtypes expressed as least square means varies signif-icantly. *p < 0.05 cluster 1 vs 2; ***p < 0.001 cluster 1 vs 3.

2.2. Estrogen and progesterone effect are consistent in femaledonor cells with laboratory adapted strain MN

Even though the lab adapted MN strain replicated less efficientlyin primary cells than T cell lines, this strain has been extensivelyused for studies of HIV pathogenesis. We therefore chose to furtherstudy hormonal effects using this strain because of measurable p24levels at different time points. Infection of cells with similar levels ofvirus resulted in a significant 1 log reduction in replication in femalecells treated with progesterone (p = 0.0021, Fig. 2A), while estrogentreatment had no effect. On the contrary, estrogen, progesteroneand testosterone treated male cells showed 1–2 logs increase inHIV replication (p < 0.05, Fig. 2A).

2.3. Hormone effect showed inconsistency between donors andsubtypes

The three primary strains subtype A, B, and C replicated effi-ciently with 1–4 logs of virus production in all donors as indicated,however they were differentially regulated by female and male hor-mones. Hormone effect varied between donors for subtype A wherelow concentrations of progesterone suppressed subtype A replica-tion by 2 logs in one donor and favored replication for donor W1(p < 0.001, Fig. 2B). This may be due to differences in host factors. Inone male and female donor, subtypes B and C had no effect with hor-mone treatment except for subtype C where virus production wasreduced by one log in cells treated with high estrogen (p = 0.041,Fig. 2B). CRF01 AE, Group-O, Group-N also showed inconsistentvariation in replication kinetics with or without progesterone andestrogen hormone treatment (data found in the Additional file1).

2.4. Significant donor to donor variation in HIV subtypestransmission from dendritic cells (DCs) to autologous peripheralblood mononuclear cells

Because varied responses were observed in PBMC with or with-out hormone treatment we wanted to study whether hormoneshad any effect on dendritic cells which are located in the mucosae(including the oral and vaginal mucosal surfaces) and the lym-phoid tissues as they have been proposed to be first target cellsfor sexual transmission of HIV or SIV [13]. Dendritic cells were firstinfected with virus concentrations similar to that used for PBMC

and the monolayer was washed and co-cultured with autologusPBMC (uninfected) cells. Day 4 and 7 culture sups were analyzedfor virus production by measuring HIV-1 p24 and HIV-2 p27 lev-els. We observed statistically significant (p < 0.001) donor to donor

V. Ragupathy et al. / Journal of Steroid Biochemistry & Molecular Biology 138 (2013) 63– 71 65

B

Strain MN(M1)

Day 4 Day 7 Day 121

2

3

4

No H ormon e

Low Estr ogen

High Est rog e nProg e ste rone

Low Te stost erone

High Test ost eron eLo

g H

IV-1

p2

4 p

g/m

l

Strain MN(W2)

Day 4 Day 7 Day 12-1

0

1

2

3

4

No H ormon e

Low Estroge nHigh Est rog e nLow Prog est e rone

High P roge steron e

Test ost e ron e

Lo

g H

IV-1

p2

4 p

g/m

lStrain MN

(W1)

Day 4 Day 7 Day 120

1

2

3

4

No H ormone

Low E strogen

High E strogen

Low P rogesterone

High P rogesterone

Testosterone

Lo

g H

IV-1

p2

4 p

g/m

l

A

Sub type C(M1)

Day 4 Day 7 Day 120

1

2

3

4

5

No Hormone

Low Estrogen

High Estrogen

Progesterone

Low Testosterone

High Testosterone

Log H

IV-1

p24 p

g/m

l

Sub type A(W2)

Day 4 Day 7 Day 120

1

2

3

4

5

6

No Hormone

Low Estrogen

Hig h Estrogen

Low Progesterone

Hig h Progesterone

Testosterone

Log H

IV-1

p24 p

g/m

l

Subtype B(W1)

Day 4 Day 7 Day 12

0

1

2

3

4

5

6

No Hormone

Low Estrog en

High Estrog en

Low Prog esteron e

High Progesterone

Testosterone

Log H

IV-1

p24 p

g/m

l

Sub type A(W1)

Day 4 Day 7 Day 12-1

0

1

2

3

4

5

No Hormon e

Low Estrog en

High Estr ogen

Low Prog esteron e

High Progesterone

Testosterone

Log H

IV-1

p24 p

g/m

l

Fig. 2. Estrogen and progesterone effect is consistent in female donor peripheral blood mononuclear cells with laboratory adapted strain MN. (A) Female donors (W1 and W2)had consistent down regulation of HIV-1 MN by high progesterone and up-regulation by low estrogen. In male donor (M1) hormones augment HIV replication as comparedto no hormone treated cells. (B) Hormone effects showed inconsistency between donors and subtypes. PBMCs estrogen treated cells differentially modulate HIV replicationkinetics of subtype A for donors W1 and W2. Estrogen or progesterone treated cells from donor W2 favor subtype A replication, but in donor W1 estrogen has no effectcompared with untreated (no hormone). Subtype C infected cells from male donor M1 and high estrogen treated cells down regulate 1 log HIV replication compared withuntreated. Sex hormones did not have any effect on subtype B infected cells of women donor W1.

v2avtHs

ta2oow7(dcmCbA

ariations (1–3 logs) for subtypes CRF02 AG, subtype B, C, HIV- and Group O. Each of 5 donors was grouped for the subtypend data for one time point (day 7) where we could measureirus production for all the subtypes are presented in Fig. 3. Cer-ain male donors (M4 and M5) did not show measurable virus forIV-2 and in a donor (M1) Group O infectivity could not be mea-

ured.In addition, we tested quantity of virus in DC after 2 h of infec-

ion, replication kinetics and role of ICAM-1 in virus transmission toutologous PBMC. DCs from each donor infected with subtypes for

h and washed, lysed to quantify virus titer. We observed ∼300 pgf HIV-2 in cells from both males and females but <100 pg forther subtypes. HIV subtypes C, HIV-2 and Group O replicationas observed based on increase of virus titer from Day 4 to Day

while other subtypes B and CRF02 AG did not replicate in DCdata available in Additional file 1). The involvement of ICAM-1 wasetermined by blocking infected DCs with anti-ICAM antibody ando-culture. These studies showed that 0.5–1 log reduction in trans-

ission to autologous PBMCs was observed for CRF02 AG, B and

subtypes while HIV-2 and Group O infections were completelylocked as indicated by virus levels at 7 days post infection (p < 0.05,dditional file 1).

2.5. Effect of hormones on HIV transmission kinetics

Similar to PBMC, treatment of DC co-culture with hormonesresulted in a consistent 1 log down regulation (p < 0.05) of HIVsubtypes transmission in estrogen-treated cells when virus levelswere measured at two different time points while progesteronetreatment favored replication (shown in Fig. 4A). Group O virustransmission was delayed and could be quantified only at day 7. Inmale cells treated with testosterone (high), a similar 1 log reduc-tion (p < 0.05) of HIV transmission (as estrogen in cells from femaledonor) was observed (Fig. 4B). In both cases hormone effect wasconsistent between the donors.

2.6. Hormones effect on HIV replication in Jurkat cells

As noted above, inconsistencies were observed in virus kineticswith primary cells. We therefore performed similar studies usingsusceptible T cell lines to control for donor to donor variations and

to ascertain whether specific subtypes were modulated by specifichormones. When virus kinetics of strains used in previous primaryPBMC experiments were tested on H9 and Jurkat cell lines, noneof primary viruses produced measurable infectivity (at 1 ng of p24

66 V. Ragupathy et al. / Journal of Steroid Biochemis

B

A

1.0E+00

1.0E+01

1.0E+02

1.0E+03

1.0E+04

1.0E+05

1.0E+06

CRF02_AG Sub type C Sub type B HIV-2 Group O

W1

W2

W3

W4

W5

HIV

- I/

II p

24

/p2

7 p

g/m

l

1.0E+00

1.0E+01

1.0E+02

1.0E+03

1.0E+04

1.0E+05

CRF02 _AG Subtype C Subtype B HIV-2 Group O

M1

M2

M3

M4

M5

HIV

- I/

II p

24

/p2

7 p

g/m

l

Fig. 3. Significant donor to donor variation in HIV subtypes transmission fromdendritic cells (DCs) to autologous peripheral blood mononuclear cells. Virus trans-mission from DC to PBMC was inferred from quantification of HIV-I/II p24/p27 titeraWi

ddlahsm

2

hdclmttsNe

3

os1no

t 7 days post infection from culture sups. (A) Dendritic cells from female donors1-W5; (B) dendritic cells from male donors M1-M5. HIV clades with missing bars

ndicate these donors resistant for that subtype of HIV.

ose) except the laboratory adapted strain MN. We therefore con-ucted hormone studies in the H9 (data not shown) and Jurkat cell

ines using the MN strain. Lower concentrations of progesteronend testosterone favored HIV replication (p < 0.05; p < 0.01), whileigher concentrations of estrogen, progesterone and testosteroneignificantly down regulated HIV replication confirming that hor-ones modulate HIV kinetics of replication (p < 0.05, Fig. 5).

.7. Hormonal effect on HIV entry receptors in Jurkat cells

Since we observed from our study using cell lines that lowormone concentrations do modulate virus kinetics we wished toetermine whether these effects were due to variations in levels ofellular virus receptors CCR5/CXCR4. Jurkat cells were treated withower hormone concentrations, stained for CD4, CCR5 and CXCR4

arkers and analyzed after 2 h of hormone treatment. Hormonereatment did not significantly alter receptor levels except that cellsreated with estrogen alone or in combination with progesteronehowed a variation of 1–2% change in CCR5 receptor levels (Fig. 6).o significant change was observed with testosterone and CXCR4xpression (data not shown).

. Discussion

This is the first study that has comprehensively analyzed effectsf hormones on the replication and transmission kinetics of HIV

ubtypes using in vitro cell culture systems. Even though we used2 donor cells from either gender, HIV subtypes infectivity wasot successful with all primary viruses or donors. We present herenly the cells (both donors) that had measurable p24 levels and its

try & Molecular Biology 138 (2013) 63– 71

effects in the presence or absence of sex hormones. In our studywe used two hormone levels that represent menstrual stages inwomen and two levels of testosterone for males in the context ofHIV infection. Our study confirmed that primary T cells from USblood donors were highly supportive of replication of predominantNorth American strain subtype B followed by the globally dominantsubtype C. Our study also confirmed that when multiple subtypeswere used for infection at similar concentrations only subtype Bestablished high infectivity in both male and female cells. Thisobservation was further supported by earlier US study where HIV-1clade B and Clade C infectivity was compared in human monocytesderived from PBMC. A panel of Clade B viruses had higher replica-tive advantage compared to clade C viruses among donor cells in US[14]. Based on our findings we propose that host HLA haplotypescould play a role in restricting infectivity of divergent subtypes andthis needs to be investigated in the future.

Several studies have shown that female sex steroid hormonescould affect HIV replication [15] either at the cellular [16,17] ormolecular level [18] these studies were performed using a singlesubtype or a cell line. We wanted to explore whether hormoneeffects were associated with a particular HIV subtype or the sex ofthe donor. Our results indicate that PBMC from female donor cellsinfected with lab adapted strains show a consistently lower levelof replication with high progesterone (luteal phase of menstrualcycle). However at lower concentrations progesterone treated cells(follicular phase) showed no change in HIV replication. Our resultsare consistent with earlier report [19] where low titers of HIV-1MNwas consistently inhibited in progesterone-treated cultures due toinhibition of IL-2 activated CCR5 expression. Furthermore lower orhigher concentration of estrogen treated cells show no differencesin replication but male cells treated with low estrogen shows signif-icant 2 log increase in virus production (Fig. 2A). Even though we didnot studied in-depth molecular mechanisms for the effect of hor-mone on HIV replication, previous studies indicated low estrogenactivates HIV LTR. Hence we assume dose of a hormone responsiblefor transcriptional regulation of HIV [11,20]. However, infection ofdifferent donor PBMCs with primary viruses did not yield similarresults particularly for subtypes A, B or C (Fig. 2B). Our results indi-cate that hormonal regulation is dependent on individual subtypeor target cell and we conclude that it may not be possible to gen-eralize these effects in broad terms for HIV infection and hormonaleffects in PBMC.

World-wide sexual transmission of HIV accounts for more than80% of new infections [3] and dendritic cells are the first target cellsthat HIV encounters after transmission to a new host. Even thoughhormone effects were not consistent among different donors of pri-mary PBMC we wanted to explore whether such effects were seenin other cell types. Five different subtypes varied in infectivity in 10donors DC of each gender (Fig. 3). In almost all the donors subtypeB and C replicated efficiently compared with other subtypes. Thisobservation correlates with population based molecular epidemi-ology data reinforcing the observation that the worldwide epidemicis dominated by these 2 strains.

In regard to virus transmission, DC to PBMC transmission mech-anisms have been extensively explored and discussed [21,13].Most of the studies confirm that after binding to DC-SIGN the virusinternalized is either degraded or transmitted through ICAM-1,LFA-1 interaction [22] by formation of infectious synapse. Whenwe measured the quantity of virus associated with DCs except forHIV-2, all other subtypes had <100 pg of p24 suggesting that HIV-2may present mostly at the surface or internalized into DC. Thisobservation was further supported by culturing DC alone without

co-culture and measurement of virus production at 2 intervals thatresulted in only HIV-1 replication in multiple donors and subtypeC and Group O replication in one donor. DCs transfer HIV to CD4+

T-cells with an interaction of ICAM-1, 2 or ICAM-3 and LFA forming

V. Ragupathy et al. / Journal of Steroid Biochemistry & Molecular Biology 138 (2013) 63– 71 67

B

A

Sub type CM1

Day 4 Day 72

3

4

5

No Hormone

Low Estrogen

High Estrogen

Low Testosterone

High Testosterone

Prog esterone

Lo

g H

IV-1

p2

4 p

g/m

l

HIV-2M2

Day 4 Day 7-1

0

1

2

3

4

No Hormone

Low Estrog en

High Estrogen

Low Testosteron e

High Testosteron e

Progesteron e

Lo

g H

IV-2

p2

7 p

g/m

l

Sub type CM3

Day 4 Day 71

2

3

4

5

No Hormon e

Low Estrogen

High Estrogen

Low Testosterone

High Testosterone

Progesterone

Lo

g H

IV-1

p2

4 p

g/m

l

Sub typ e BM2

Day 4 Day 72

3

4

No Hormone

Low Estrogen

High Estrogen

Low Testosterone

High Testosterone

Progesterone

Lo

g H

IV-1

p2

4 p

g/m

l

Sub type BW1

Day 4 Day 72

3

4

5

No Hormone

Low Estrog en

High Estrogen

Low Progesteron e

High Prog esterone

Testosteron e

Lo

g H

IV-I

p2

4 p

g/m

lSubtype C

W1

Day 4 Day 7-1

0

1

2

3

4

5

No Hormon e

Low Estrog en

High Estrog en

Low Prog esteron e

High Prog esteron e

Testosterone

Lo

g H

IV-I

p2

4 p

g/m

l

Group OW1

Day 4 Day 7-1

0

1

2

3

No Hormone

Low Estrogen

High Estrogen

Low Progesterone

High Progesterone

Testosteron e

Lo

g H

IV-I

p2

4 p

g/m

l

HIV-2W5

Day 4 Day 7-1

0

1

2

3

4

No Hormone

Low Estrogen

High Est rog en

Low Progesterone

High Progesterone

Testosterone

Lo

g H

IV-1

p2

7 p

g/m

l

Subtype BW2

Day 4 Day 72

3

4

5

No Hormon e

Low Estrog en

High Estrog en

Low Progesteron e

High Progesterone

Testosteron e

Lo

g H

IV-1

p2

4 p

g/m

l

Fig. 4. Estrogen and testosterone down regulates cell to cell HIV-1 transmission from dendritic cells to autologous peripheral blood mononuclear cells. (A) High estrogent d B ini own

B

vIcIbpEis

reatment down regulates cell to cell transmission of HIV-1 subtypes C, HIV-2, annfected with subtype B or Group O. (B) High testosterone treated cells significantly doth figures (A) and (B) virus titer was indicated for two time points.

irologic synapse. In order to study this mechanism we blockedCAM-1 on DCs previously infected with different subtypes ando-cultured with PBMC. Our results indicate HIV subtypes differ inCAM mediated transmission. HIV-2 depended solely on ICAM-1ut other subtypes had reduced transmission efficiency which was

reviously demonstrated in cell lines with lab adapted strain [23].ven though anti-ICAM antibody was consistently maintainedn co-culture experiments we detected p24 levels in cultureupernatants suggesting involvement of other molecules [21,24].

fected donors compared to control. No effect of estrogen treatment for donor W1regulate cell to cell transmission HIV subtypes C, B, HIV-2 in cells from male donors.

Hormone treatment and dendritic cells play a crucial role inmediating transmission in both males and females and higher lev-els of estrogen reduce transmission efficiency of subtypes B, C andHIV-2. However in certain female donors Group O and subtype Bhormone effects were significant at lower levels and promoted

replication at higher levels. These results suggest that HIV sub-type and hormones may have to be taken into consideration withfemales of reproductive age. In males when high doses of testos-terone showed significant reduction in transmission of all subtypes

68 V. Ragupathy et al. / Journal of Steroid Biochemis

0

5

10

15

20

25

30

HIV

-1 p

24

pg

/ml x

10

00

Day 4

Day 7

Day 12**

*

*

**

Fig. 5. Hormones effect on HIV replication in Jurkat cells. Bar graph shows lymphoidcell line infectivity by HIV was significantly down regulated at higher concentrationsof estrogen, progesterone and testosterone treated cells. Lower concentration ofpa

adpbcaHb

In summary, our study demonstrates that HIV diversity and sex

Fsh

rogesterone and testosterone treated cells up-regulated HIV replication. *p < 0.05nd **p < 0.01, compared with no hormone treated cells.

nd donors. Previously it was reported that estrogen/progesteroneoses are crucial for transcriptional regulation of HIV and weonder in our study testosterone effects on HIV replication maye at molecular level [11,18]. Hormonal levels (including contra-eptive use) and HIV acquisition studies in humans [25,26] and

nimal models [27] showed that higher progesterone levels favorIV production. Our results suggest that these observations maye individualized i.e., HIV subtype and host genetics may play a

ig. 6. Hormone effects on HIV entry receptor CCR5 levels in Jurkat cells. Jurkat cells wurface marker CCR5. The percentage of CCR5 cells increase with estrogen alone or in comormone). Progesterone and testosterone treated cells do not have any effect on CCR5 lev

try & Molecular Biology 138 (2013) 63– 71

role in acquisition. Lastly, we observe donor to donor variationsin both cell types discussed above. To determine whether hor-mones modulate HIV kinetics or not we used Jurkat and H9 celllines where such variations are not a contributing factor to theobservations. When we used similar virus concentrations (5 ng/mlof p24), incubation and infectivity conditions in PBMC or DC noneof the primary viruses established infection except the lab adaptedMN strain. Exploring hormone effects on HIV replication kineticswe observed dose dependent effects where, higher concentrationsreduced replication kinetics and lower concentration favored HIVreplication (Fig. 5). However studies with primary cells do notsupport this observation. This might explain that host geneticsmay play a major role in infectivity of HIV strains. Our futureexperiments will focus on this area of research. Hormone effectswere consistent between two cell line hence we present data onlyfor Jurkat cell line. Many studies report that HIV receptors CD4,CCR5 and CXCR4 were modulated by hormones [28,29]. In ourexperiments using cell lines where we found that low hormonelevels promoted replication, the effects were not due to changesin CD4+ CCR5 or CXCR4 levels although cells treated with estro-gen had slightly higher CCR5 expression which did not correlatewith infectivity. Progesterone treatment did not show any changesto expression of CCR5 or CXCR4 in both of our primary cells or T-cell line. But earlier report [19] indicated that IL-2 activated PBMCwhen treated with progesterone at a concentration of 10−5–10−6

resulted in reduced expression of CCR5 and CXCR4. This may be dueto donor to donor variations in cytokine expression levels.

hormones may have an impact on viral replication and transmissionkinetics. The replication of subtypes significantly differed in bothPBMC and DC cell types with high replicative advantage for subtype

ith or without hormone treatment are represented in this histogram for the cellbination with progesterone was consistent when compared to untreated cells (noels.

hemis

BrtmioToatopic

4

4(

toiwdttwlPtflsim12q

4(

bFwfho2wmcgb

4

CH[fOm

V. Ragupathy et al. / Journal of Steroid Bioc

and C in male and female derived cells. Hormones modulated HIVeplication and transmission kinetics but showed significant donoro donor variation which is the major limitation of this study. Hor-

one levels modulate HIV kinetics and may play a major role innfection of women of reproductive age. DC to PBMC transmissionf HIV-2 was dependent on ICAM-1 but not for the other subtypes.he variations observed in our study are not due to hormone effectsn cellular HIV receptor levels but may involve other cellular mech-nisms. Larger cohort studies in different geographic locations withheir prevalent strains are needed to better understand the effectsf subtype and hormone status on virus transmission and diseaserogression. The findings of our study may facilitate future stud-

es with regard to hormonal contraception and HIV acquisition inontext of Women’s health.

. Materials and methods

.1. Isolation and culture of peripheral blood mononuclear cellsPBMC

Prior to blood draw informed consent was obtained from all ofhe healthy, normal blood donors according to the ethical principlesf international ethical guidelines for biomedical research involv-ng human subjects. PBMC from six male and six female donors

ere isolated from buffy coat received from HIV seronegative bloodonors from the NIH Blood bank. Most of these donors were athe reproductive age between 25 and 45 years. Other identities ofhese donors were not notified to us. Hence significant findingsere could not be repeated on these donors. We used our standard

aboratory protocol for PBMC isolation from buffy coat [12]. Briefly,BMC were isolated by Ficoll/Hypaque density gradient centrifuga-ion. Monocytes (>95%) were removed by adherence to the cultureasks and the remaining peripheral blood lymphocytes (PBL) weretimulated with 2 �g/ml PHA for 3 days to activate T cells beforenfection. The PBL or PBMC were cultured in RPMI 1640 supple-

ented with 15% FBS, 2 mM glutamine, 100 units/ml of penicillin,00 units/ml of streptomycin, and 5 units/ml of human Interleukin-

(Roche, NJ). Cells were either stored frozen or infected with equaluantities of each of viral isolates.

.2. Isolation and culture of monocyte-derived dendritic cellsMDDC)

MDDC were isolated from each of five female and maleuffy coat follows similar method of our PBMC isolation untilicoll–Hypaque gradient (density = 1.070 g/ml) and afterwards cellsere split into two, one part of cells (1 × 107/ml) stored at L2N

or co-culture experiments and another part layered over a slightyper-osmolar Percoll gradient (density = 1.064 g/ml) to separateut lymphocytes. Subsequently, serum free plastic adherence for

h at 37c yielded >90% of pure monocytes [30]. Dendritic cells (DC)ere differentiated by culturing with M-CSF/IL4 containing RPMIedia with 10% FBS serum for 6 days. After 6 days, media were

hanged and R848 added (2 �g/ml chemical ligand from Invitro-en, CA, USA) for DC maturation for 48 h and subsequently switchedack to M-CSF/IL4 media [21] for infectivity.

.3. PBMC infection with HIV subtypes

PBMC’s are infected with (5 ng p24/ml) HIV subtypes A–D andRF02 AG, CRF01 AE, MN (lab adapted), Group O, Group N andIV-2. (These isolates are obtained from the FDA subtype panel

31]. Subtype A and D isolates are from Uganda, B from USA, Crom Senegal, CRF01 AE and CRF02 AG are from Cameroon, Group

from Germany and N from Sinnousi). Prior to infectivity experi-ents, PBMC cells were prepared at a concentration of 1 million/ml

try & Molecular Biology 138 (2013) 63– 71 69

in serum free media and aliquots of 3 million cells per tube weremade for infection with each of the 11 HIV strains noted above induplicates. The remaining 7 million cells served as the uninfectedcontrol. After 2 h of infection at 37 ◦C, 5% CO2, cells were washedtwice with 10 ml of 1× Phosphate Buffer Saline (PBS) and cell pelletwas re-suspend with 3 ml of culture media (RPMI with 15% FBS).About 0.5 ml (0.5 million cells) from each aliquot of infected cellswas distributed into 24 well plates. Similar methods were followedfor infection of frozen cells after spontaneous thawing.

4.4. MDDC infection with HIV subtypes

Because we could not obtain sufficient cells to conduct studieswith all HIV subtypes we limited our study of DC to autologus PBMCtransmission to five strains including subtypes B, C, CRF02 AG,Group O and HIV-2. DC’s were first infected with HIV strains for2 h of infection at 37 ◦C, 5% CO2, cells were washed twice with1×PBS by gentle swirling followed by complete removal of PBS bypipet and co-cultured with uninfected autologous PBMC. Cell den-sity and hormone concentrations were similar to that used in thePBMC kinetics study. Replication and transmission kinetics weredetermined from culture supernatants harvested at day 4 and 7.Since inter cellular adhesion molecule – 1 (ICAM) is thought to beinvolved in DC to T-cell transmission of HIV, we used 2000 ng/mlof anti-ICAM antibodies [32] to study whether transmission kinet-ics depended on ICAM-1. In our experiments infected DCs weretreated with anti-ICAM-1 antibody for 2 h prior to co-culture andmaintained throughout the culture.

4.5. T-cell lines

Since donor to donor variability is a well-known factor inHIV infectivity we also studied hormone effects on HIV replica-tion kinetics in established T cell lines, Jurkat and H9 receivedfrom the NIH AIDS Reagent program and cultured in RPMI 1640supplemented with 10% FBS, 2 mM glutamine, 100 units/ml of peni-cillin, and 100 units/ml streptomycin. Hormone concentrationsused were similar to those used in PBMC and DC studies.

4.6. Physiological hormone concentrations used

Two different physiological concentrations of sex steroid hor-mones were used to treat infected cells; progesterone (Sigma,P7556) 200 pM (low) & 500 pM (high), ß-estradiol (Sigma, E2257)2 nM (low) & 20 nM (High), testosterone (Sigma, T5411) 4 nm (low)& 40 nM (High). Rational for not pretreating the cells was to ensurean equal quantity of virus infects cells and to analyze the effectof hormone during the log phase of virus replication. Further, hor-mone concentrations were maintained throughout the culture afterinfection. To study the effect of testosterone and progesterone oncells from female and male donors, we used a single low concentra-tion of these hormones for each donor. The concentrations selectedwere all within the normal range found in plasma. Further we chose2 different concentrations because during a menstrual cycle one ofthe hormone is elevated at different stages and we wish to comparethe virus replication at these two levels.

4.7. Determination of infectivity titers

1 ml of culture supernatant was harvested at 4, 7 and 12 dayspost infection for p24 or p27 measurements in duplicates. For HIV-1 p24 quantification we used the Perkin–Elmer HIV-1 p24 Elisa

Kit, Cat No: NEK050B001 and for HIV-2 quantification the Zep-tometrix SIV p27 Antigen Elisa kit, Cat No 0801169 was used. Allprocedures were followed as per the manufacturer instructions.Because these kits were optimized for quantification in a narrow

7 hemis

rptfp

4

ctcdmeaf

4

ldulmA(a

C

am

A

twavawsc

A

JTOcAA

A

f2

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

0 V. Ragupathy et al. / Journal of Steroid Bioc

ange 200–12.5 pg/ml for HIV-1 p24 and 2000–62.5 pg/ml for HIV-227 most of the culture supernatants were diluted in culture mediao 1:10, 1:100 and 1:1000 for quantification. Data were analyzedrom excel spread sheet with graph pad software v5.0 and p24 or27 values were plotted in log scale.

.8. Flow cytometry analysis for HIV entry receptors

Flow cytometric analysis was performed for CD4+ CCR5/CXCR4ells to study whether hormones modulate levels of these recep-ors. These studies were performed with a single donor for primaryells and Jurkat cells were treated with hormone concentrationsiscussed above for receptor level changes. Cells were treated withale hormone testosterone and female hormones progesterone,

strogen either alone or combination for measurement of CCR5nd CXCR4 expression levels. Cells express CD4+ were gated firstollowed by CCR5+ or CXCR4+ cells.

.9. Statistical analysis

Overall viruses’ replication kinetics in both genders we usedeast square means (LSM) in order to best fit XY-curve. LSM data iserived from means of p24 values in different hormone treated andntreated culture sups and means from different donors were ana-

yzed in a statistical analysis system. Analysis of datasets containore than two groups was achieved by one-way repeated measureNOVA. Analysis of two groups of data sets statistical significance

unpaired student t-test) was measured at p < 0.05 (*), p < 0.01 (**)nd p < 0.0001 (***) very significant.

ompeting interests

The authors certify that there is no conflict of interest withny financial organization regarding the material discussed in theanuscript.

uthor contributions

V.R. drafted the manuscript and made substantial contributionso conception, design, analysis and interpretation of data, ST helpedith study design and hormone concentrations for infectivity, O.W.

nd S.L. provided cells and viruses for this study, K.D. and W.X. pro-ided suggestions for the work and reviewed the manuscript, Ar.D.ssisted with p24 analysis, A.D. assisted in design of the study, I.H.as involved in concept, design, data interpretation, and supervi-

ion of the project and provided input in revising the manuscriptritically for publication.

cknowledgments

The authors wish to acknowledge Dr. Ming Zhang and Dr.iangqin Zhao and Dr. Robin Biswas for review of the manuscript.his work was supported by an intramural grant from the FDA’sffice of Women’s Health. The findings and conclusions in this arti-le have not been formally disseminated by the Food and Drugdministration and should not be construed to represent anygency determination or policy.

ppendix A. Supplementary data

Supplementary data associated with this article can beound, in the online version, at http://dx.doi.org/10.1016/j.jsbmb.013.03.002.

[

[

[

try & Molecular Biology 138 (2013) 63– 71

References

[1] J. Hemelaar, E. Gouws, P.D. Ghys, S. Osmanov, Global trends in molecular epi-demiology of HIV-1 during 2000–2007, Aids 25 (5) (2011) 679–689.

[2] S.C. Ray, T.C. Quinn, Sex and the genetic diversity of HIV-1, Nature Medicine 6(1) (2000) 23–25.

[3] Unaids report on the global aids epidemic, 2010.[4] A. Meier, J.J. Chang, E.S. Chan, R.B. Pollard, H.K. Sidhu, S. Kulkarni, T.F. Wen, R.J.

Lindsay, L. Orellana, D. Mildvan, S. Bazner, H. Streeck, G. Alter, J.D. Lifson, M.Carrington, R.J. Bosch, G.K. Robbins, M. Altfeld, Sex differences in the toll-likereceptor-mediated response of plasmacytoid dendritic cells to HIV-1, NatureMedicine 15 (8) (2009) 955–959.

[5] M. Gandhi, P. Bacchetti, P. Miotti, T.C. Quinn, F. Veronese, R.M. Greenblatt, Doespatient sex affect human immunodeficiency virus levels? Clinical InfectiousDiseases 35 (3) (2002) 313–322.

[6] T.R. Sterling, D. Vlahov, J. Astemborski, D.R. Hoover, J.B. Margolick, T.C. Quinn,Initial plasma HIV-1 RNA levels and progression to AIDS in women and men,New England Journal of Medicine 344 (10) (2001) 720–725.

[7] A.L. Meditz, S. MaWhinney, A. Allshouse, W. Feser, M. Markowitz, S. Little, R.Hecht, E.S. Daar, A.C. Collier, J. Margolick, J.M. Kilby, J.P. Routy, B. Conway, J.Kaldor, J. Levy, R. Schooley, D.A. Cooper, M. Altfeld, D. Richman, E. Connick,Sex, race, and geographic region influence clinical outcomes following primaryHIV-1 infection, Journal of Infectious Disease 203 (4) (2011) 442–451.

[8] F. Brambilla, A. Speca, I. Pacchiarotti, M. Biondi, Hormonal background ofphysiological aggressiveness in psychologically healthy women, InternationalJournal of Psychophysiology 75 (3) (2010) 291–294.

[9] C.R. Wira, J.V. Fahey, A new strategy to understand how HIV infects women:identification of a window of vulnerability during the menstrual cycle, Aids 22(15) (2008) 1909–1917.

10] A.S. Bourinbaiar, R. Nagorny, X. Tan, Pregnancy hormones, estrogen and pro-gesterone, prevent HIV-1 synthesis in monocytes but not in lymphocytes, FEBSLetters 302 (3) (1992) 206–208.

11] S.N. Asin, A.M. Heimberg, S.K. Eszterhas, C. Rollenhagen, A.L. Howell, Estradioland progesterone regulate HIV type 1 replication in peripheral blood cells, AIDSResearch and Human Retroviruses 24 (5) (2008) 701–716.

12] M. Zhang, Q. Huang, Y. Huang, O. Wood, W. Yuan, C. Chancey, S. Daniel, M. Rios,I. Hewlett, K.A. Clouse, A.I. Dayton, Beta-estradiol attenuates the anti-HIV-1efficacy of Stavudine (D4T) in primary PBL, Retrovirology 5 (2008) 82.

13] L. Wu, V.N. KewalRamani, Dendritic-cell interactions with HIV: infection andviral dissemination, Nature Reviews Immunology 6 (11) (2006) 859–868.

14] A.A. Constantino, Y. Huang, H. Zhang, C. Wood, J.C. Zheng, HIV-1 clade B andC isolates exhibit differential replication: relevance to macrophage-mediatedneurotoxicity, Neurotoxicity Research 20 (3) (2011) 277–288.

15] A.W. Lee, D. Mitra, J. Laurence, Interaction of pregnancy steroid hormones andzidovudine in inhibition of HIV type 1 replication in monocytoid and placentalHofbauer cells: implications for the prevention of maternal-fetal transmissionof HIV, AIDS Research and Human Retroviruses 13 (14) (1997) 1235–1242.

16] L.D. Munoz, M.J. Serramia, M. Fresno, M.A. Munoz-Fernandez, Progesteroneinhibits HIV-1 replication in human trophoblast cells through inhibition ofautocrine tumor necrosis factor secretion, Journal of Infectious Diseases 195(9) (2007) 1294–1302.

17] D. Mavoungou, V. Poaty-Mavoungou, M.Y. Akoume, B. Ongali, E. Mavoungou,Inhibition of human immunodeficiency virus type-1 (HIV-1) glycoprotein-mediated cell-cell fusion by immunor (IM28), Virology Journal 2 (2005) 9.

18] D. Katagiri, H. Hayashi, A.F. Victoriano, T. Okamoto, K. Onozaki, Estrogenstimulates transcription of human immunodeficiency virus type 1 (HIV-1),International Immunopharmacology 6 (2) (2006) 170–181.

19] N. Vassiliadou, L. Tucker, D.J. Anderson, Progesterone-induced inhibition ofchemokine receptor expression on peripheral blood mononuclear cells cor-relates with reduced HIV-1 infectability in vitro, Journal of Immunology 162(12) (1999) 7510–7518.

20] A. Gilles, A. Miquelis, J. Luis, E. Faure, Activation of transcription from the humanimmunodeficiency virus type 1 (HIV-1) long terminal repeat by the vasoactiveintestinal peptide (VIP), Italian Journal of Biochemistry 47 (2) (1998) 101–110.

21] F. Groot, T.M. van Capel, M.L. Kapsenberg, B. Berkhout, E.C. de Jong, Opposingroles of blood myeloid and plasmacytoid dendritic cells in HIV-1 infection ofT cells: transmission facilitation versus replication inhibition, Blood 108 (6)(2006) 1957–1964.

22] L. de Witte, A. Nabatov, M. Pion, D. Fluitsma, M.A. de Jong, T. de Gruijl, V.Piguet, Y. van Kooyk, T.B. Geijtenbeek, Langerin is a natural barrier to HIV-1transmission by Langerhans cells, Nature Medicine 13 (3) (2007) 367–371.

23] J.H. Wang, C. Kwas, L. Wu, Intercellular adhesion molecule 1 (ICAM-1), but notICAM-2 and -3, is important for dendritic cell-mediated human immunodefi-ciency virus type 1 transmission, Journal of Virology 83 (9) (2009) 4195–4204.

24] N.J. Venkatachari, S. Alber, S.C. Watkins, V. Ayyavoo, HIV-1 infection of DC:evidence for the acquisition of virus particles from infected T cells by antigenuptake mechanism, PLoS One 4 (10) (2009) e7470.

25] C.S. Morrison, P.L. Chen, C. Kwok, B.A. Richardson, T. Chipato, R. Mugerwa, J.Byamugisha, N. Padian, D.D. Celentano, R.A. Salata, Hormonal contraceptionand HIV acquisition: reanalysis using marginal structural modeling, Aids 24(11) (2010) 1778–1781.

26] C.S. Morrison, B.A. Richardson, F. Mmiro, T. Chipato, D.D. Celentano, J. Luoto, R.Mugerwa, N. Padian, S. Rugpao, J.M. Brown, P. Cornelisse, R.A. Salata, Hormonalcontraception and the risk of HIV acquisition, Aids 21 (1) (2007) 85–95.

27] S.A. Vishwanathan, P.C. Guenthner, C.Y. Lin, C. Dobard, S. Sharma, D.R.Adams, R.A. Otten, W. Heneine, R.M. Hendry, J.M. McNicholl, E.N. Kersh, High

hemis

[

[

[

[

V. Ragupathy et al. / Journal of Steroid Bioc

susceptibility to repeated, low-dose, vaginal SHIV exposure late in the lutealphase of the menstrual cycle of pigtail macaques, Journal of Acquired ImmuneDeficiency Syndromes 57 (4) (2011) 261–264.

28] F. Dominguez, A. Galan, J.J. Martin, J. Remohi, A. Pellicer, C. Simon, Hormonal andembryonic regulation of chemokine receptors CXCR1, CXCR4, CCR5 and CCR2B

in the human endometrium and the human blastocyst, Molecular HumanReproduction 9 (4) (2003) 189–198.

29] J.S. Sheffield, G.D. Wendel Jr., D.D. McIntire, M.V. Norgard, The effect of pro-gesterone levels and pregnancy on HIV-1 coreceptor expression, ReproductionScience 16 (1) (2009) 20–31.

[

try & Molecular Biology 138 (2013) 63– 71 71

30] M.C. de Almeida, A.C. Silva, A. Barral, M. Barral Netto, A simple method forhuman peripheral blood monocyte isolation, Memorias do Instituto OswaldoCruz 95 (2) (2000) 221–223.

31] S. Lee, O. Wood, R.E. Taffs, J. Hu, A. Machuca, A. Vallejo, I. Hewlett,Development and evaluation of HIV-1 subtype RNA panels for the standard-

ization of HIV-1 NAT assays, Journal of Virological Methods 137 (2) (2006)287–291.

32] C.D. Rizzuto, J.G. Sodroski, Contribution of virion ICAM-1 to human immuno-deficiency virus infectivity and sensitivity to neutralization, Journal of Virology71 (6) (1997) 4847–4851.