Chiozzini Surface-bound Tat inhibits antigen-specific CD8 T cell activation in an integrin-dependent...

CE: Swati; AIDS-D-14-00246; Total nos of Pages: 12;

AIDS-D-14-00246

Surface-bound Tat inhib
its antigen-specific CD8R
T cell activation in an integrin-dependent manner

Chiara Chiozzinia, Barbara Collacchia, Filomena Nappia,b,

Tanja Bauerc, Claudia Arenaccioa,d, Antonella Tripicianoa,

Olimpia Longoa, Fabrizio Ensolie, Aurelio Cafaroa, Barbara Ensolia and

Maurizio Federicoa

Copyright © L

aNational AIDS CItaly, cInstitute ofdDepartment of ScVirology and Imm

Correspondence tRome, Italy.

Tel: +9 06 49903Received: 27 Febr

DOI:10.1097/QAD

IS

Objective: The identification of still unrevealed mechanisms affecting the anti-HIVCD8þ T cell response in HIV-1 infection.

Design: Starting from the observation that anti-Tat immunization is associated withimproved CD8þ T cell immunity, we developed both in-vitro and ex-vivo assays tocharacterize the effects of extra-cellular Tat on the adaptive CD8þ T cell response.

Methods: The effects of Tat on CD8þ T cell activation were assayed using CD8þ T cellclones specific for either cellular (MART-1) or viral (HIV-1 Nef) antigens, and HIV-1Gag-specific CD8þ T cells from HIV-1 patients.

Results: The interaction between CD8þ T lymphocytes and immobilized Tat, but not itssoluble form, inhibits peptide-specific CD8þ T lymphocyte activation. The inhibitiondoes not depend on Tat trans-activation activity, but on the interaction of the Tat RGDdomain with a5b1 and avb3 integrins. Impaired CD8þ T cell activation was alsoobserved in cocultures of CD8þ T cells with HIV-1 infected cells. Anti-Tat Abs abrogatethe inhibitory effect, consistently with the evidence that extracellular Tat accumulateson the cell membrane of virus-producing cells. The Tat-induced inhibition of cellactivation associates with increased apoptosis of CD8þ T cells. Finally, the inhibition ofcell activation also takes place in Gag-specific CD8þ T lymphocytes from HIV-1infected patients.

Conclusion: Our results support the idea that CD8þ T cell apoptosis induced bysurface-bound extracellular Tat can contribute to the dysregulation of the CD8þ T celladaptive response against HIV as well as other pathogens present in AIDS patients.

� 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins

AIDS 2014, 28:000–000

Keywords: apoptosis, CD8þ T cells, interferon-gamma, integrins, Tat

Introduction

The peak of viremia after primary HIV infection iscounteracted by the antiviral CD8þ T lymphocyteadaptive immune response, which in most cases is unable

ippincott Williams & Wilkins. Unaut

enter, bNational Center for Immunobiologicals,Virology, Helmholtz Zentrum Munchen, Germaience, University Roma Tre, and eIstituti Fisioterunology, Rome, Italy.

o Dr Maurizio Federico, National AIDS Center,

248; fax: +39 06 49903002; e-mail: maurizio.feuary 2014; revised: 20 June 2014; accepted: 23

.0000000000000389

SN 0269-9370 Q 2014 Wolters Kluwer H

to clear all infected cells [1,2]. Occasionally, the CD8þ

T lymphocyte response is associated with a more effectivecontrol of viremia such as when the immune response isdirected against HIV-1 Tat [3–5] or against HIV-1 Gagregions critical for high viral fitness, as observed in HIV

horized reproduction of this article is prohibited.

Research and Evaluation, Istituto Superiore di Sanita, Rome,n Center for Environmental Health, Munchen, Germany,apici Ospetalieri, San Gallicano Hospital, Core Laboratory of

Istituto Superiore di Sanita, Viale Regina Elena 299, 00161

[email protected] 2014.

ealth | Lippincott Williams & Wilkins 1

mailto:[email protected]

http://dx.doi.org/10.1097/QAD.0000000000000389

Co


AIDS-D-14-00246

2 AIDS 2014, Vol 00 No 00

‘elite controllers’, many of whom express HLA-B27,HLA-57 or HLA-58 [6]. Several studies have indicated thatin ‘elite controllers’, HIV-1 specific CD8þ T lymphocyteshave a superior capacity to produce cytokines andchemokines than CD8þ T lymphocytes from viremicpatients [7]. A key role of the CD8þ T lymphocyteresponse is also indicated by studies in nonhuman primates[8,9] and in HIV-2 infection, wherein thenatural control ofinfection is associated with a polyfunctional Gag-specificCD8þT cell response [10]. Thus, evidence suggests that aneffective CD8þ T cell response may contribute incontaining the spread of HIV and progression to disease.

HIV-1 Tat is a key regulatory HIV-1 protein essential fortranscription of viral RNA [11]. In the absence of Tat, onlyshort noncoding viral transcripts are produced. Similar toother HIV regulatory proteins, Tat exhibits additionalrelevant properties. For example, Tat is released by infectedcells by a leaderless secretory pathway [12–14] and ispresent in tissues and sera of HIV-infected persons [14,15].Tat can enter bystander cells, in which it transactivates viraland/or cellular genes expression [16]. Extracellular Tatbinds heparan sulphate proteoglycans of the extracellularmatrix through its basic domain [14,17], as well as a5b1,avb3 and avb5integrins through its C-terminal RGDdomain [18–20]. The Tat-integrin binding leads toefficient cell internalization of Tat in dendritic cells,macrophages and endothelial cells [15,21,22]. Binding ofTat to integrins is involved in HIV-1 transmission todendritic cells through the formation of a virus entrycomplex between Tat and oligomeric Env [23]. Integrinsare heterodimeric cell membrane proteins that respond tointracellular signals and transduce signals to the cytoplasmupon binding to extracellular ligands [24–26]. Integrins areinvolved in cellmigration, the formation of immunologicalsynapses and cytoskeletal rearrangement. All of thesefunctions are critical for the T cell adaptive immuneresponse.

In light of the overall improvement of immune functionsin Tat-immunized HIV-1 patients under HAART werecently described [27], we investigated the effects ofextracellular Tat on the CD8þ T cell adaptive immuneresponse. Here, we show that the interaction betweenantigen-specific CD8þ T lymphocytes and surface-bound Tat, but not its soluble form, leads to decreasedpeptide-specific CD8þ T cell activation in an a5b1 andavb3 integrin-dependent manner. These findingsindicate that Tat acts as a negative regulator of theantigen-specific CD8þ T cell immune response, support-ing the notion that Tat represents a privileged therapeutictarget in the fight against AIDS.

Materials and methods

Cell cultures and HIV-1 infections293T cells were grown in DMEM as well as 10% heat-inactivated FCS. MCF-7 [28], Jurkat cells and both

pyright © Lippincott Williams & Wilkins. Unautho

HLA-A02 and HLA–B7 human B-lymphoblastoid celllines (B-LCLs) were grown in RPMI medium supple-mented with 10% FCS. B-LCLs were generated by in-vitro EBV infection of ex-vivo B lymphocyte cultures.Both isolation and expansion of CD8þ T cell clonesspecific for MART-1 (melanoma-associated recognizedby T cells) and HIV-1 Nef have been previously described[29,30]. MART-1-specific CD8þ T cells recognize theHLA-A.02-restricted AAGIGILTV27–35 peptidesequence, whereas Nef-specific CD8þ T cells recognizethe HLA-B7-restricted TPGPGVRYPL128–137 pep-tide. Both antigen-specific CD8þ T-cell clones werecultivated in RPMI as well as 10% AB human serum(Gibco) and regularly monitored for their specificity.

The generation of HIV-1 rev-pNL4-3 molecular clonewas already described [31]. HIV-1 preparations pseudo-typed with the G protein from vesicular stomatitis virus(VSV-G) were obtained as previously reported [32]. Viruspreparations were titrated in terms of HIV-1 CAp24content using quantitative ELISA (Innogenetic). Infectionswith pseudotyped HIV-1 derivatives were carried out byspinoculation at 150g for 30 min at room temperature.

CD8R T cell activation assaysThe CD8þ T cell activation was evaluated in terms ofrelease of interferon-gamma (IFN-g) by ELISA (Immuno-logical Sciences) and/or in ELISPOT assays. For thesoluble Tat assay, endotoxin-free clade B recombinantHIV-1 Tat (Advanced Biological Laboratories, ABL) testedfor its biologic activity as previously described [21], wasused. For the tests with immobilized Tat, wells were coatedas already reported [33] with either wt or Tatcys22,(ABL),that is, a Tat protein with a mutation in cysteine 22 andlacking the trans-activation function [34]. As control,human serum albumin (HSA), HIV-1 Env gp120, HIV-1Nef [35] and human hepatitis C virus NS3 (ProSpec)recombinant proteins were used. In some instances, beforecell addition, Tat-coated wells were treated with affinitypurified anti-Tat polyclonal Abs (ANT0001; Diatheva). Inother instances, CD8þ T cells were preincubated with amix of anti-a5b1 (JBS5, Millipore) and avb3 (LM609;Millipore) integrin mAbs (1 mg/ml each) blocking thebinding of the Tat RGD domain to integrins. Tat wasloaded onto the membranes of Jurkat cells by incubating106 cells with 1 mg/ml recombinant Tatcys22 for 30 min at48C on a rotating plate. Thereafter, the cells wereextensively washed and incubated with 2% v/v parafor-maldheyde (PFA) in 1�PBS for 30 min at 48C.

FACS analysisIntegrins on the surface of CD8þ T cells were detected byFACS analysis with the above referenced mAbs for a5b1and avv3 integrins, and P1F6 mAb (Millipore) for avb5integrin. HLA-A.02 on the surface of MCF-7 cells wasdetectedby FITC-conjugated anti–HLA-A.02 mAb (OneLambda, Inc.). Tat on the cell membrane was detectedusing the anti-Tat clone 4138 mAb. An RGD-specific

rized reproduction of this article is prohibited.


AIDS-D-14-00246

Tat effects on CD8R T cells Chiozzini et al. 3

anti-Tat 4B4C4 mAb (Diatheva) was used to detect Tat onthe surface of PFA-fixed cells. Activated CD8þ T cells inPBMCs from HIV-1 patients were identified by anti-CD25 and mAb (eBioscience). Both intracellular HIV-1Gag and Nef product were detected after cell permeabi-lization as already reported [32].

Western blot assayWestern blot analysis of MCF-7 cells infected with (VSV-G) HIV-1 was performed as previously described [32].The Abs used in immunoblots were polyclonal rabbitanti-HIV-1 Gag CAp24 #4250, rabbit polyclonal anti-Tat ANT0001, sheep polyclonal anti-Nef clone ARP444(a generous gift from Dr Mark Harris) and antib-actinmAb AC-74 (Sigma).

Immunofluorescence assayFor fluorescence microscopy analysis, 5� 104 cells wereseeded on chamber slides and infected with wt (VSV-G)HIV-1. Forty-eight hours later, the cells were labelledwith both 2G12 anti-Env gp120 human mAb and anti-Tat clone 4138 mAb.

Apoptosis assayApoptosis was detected by measuring the binding ofFITC-conjugated Annexin-V (BD Pharmingen) follow-ing the manufacturers’ recommendations.

Isolation and cultivation of PBMCs from AIDSpatientsWe isolated PBMCs from three HIV-1 infected asympto-matic patients enrolled in a multicentric observationalstudy (ISS OBS T-002; ClinicalTrial.gov NCT01024556).The patients were of either sex, at least 18 years old, undersuccessful HAART, that is chronic suppression of HIVinfection with a plasma viremia less than 50 copies/ml inthe last 6 months and without a history of virologicrebound. Whole blood was collected in vacutainermononuclear Cells Preparation Tubes (CPT) (BDDiagnostics Preanalytical Systems Belliver IndustrialEstate) containing sodium citrate, and processed accordingto the manufacturer’s instructions. PBMCs were thenisolated and cultured in RPMI plus 10% AB human serumfor 7 days in the presence of a mix of HIV-1 Gag peptidesrepresenting most frequent potential T epitopes (PTEs)[36]. In parallel, monocytes were separated from PBMCsusing anti-CD14 microbeads (Miltenyi) and were culturedin RPMI plus 10% AB human serum for 7 days to obtainmonocyte-derived macrophages (MDMs). CD8þ T cellswere separated from peptide-stimulated PBMCs usinganti-CD8 microbeads (Miltenyi).

Statistical analysisWhen appropriate, data are presented as meanþ standarddeviation (SD). In some instances, the paired Student’s t-test was used and confirmed using the nonparametricWilcoxon rank sum test. A P value of less than 0.05 wasconsidered significant.

Copyright © Lippincott Williams & Wilkins. Unaut

Results

IFN-g release from peptide-stimulated CD8þ T lympho-cytes is inhibited by immobilized Tat but not itssoluble form.

The HLA-A.02 CD8þ T clone specific for the MART-1antigen was incubated with different concentrations(i.e. from 1 to 100 ng/ml) of clade B recombinant HIV-1Tat or HSA. After 5 h of incubation, MART-1 specificCD8þ T cells were cocultured with HLA-matched B-LCLs previously treated with 1–10 ng/ml of the HLA-A.02-restricted MART-1 peptide. After overnight incu-bation, supernatants were assayed for IFN-g contentby ELISA. Figure 1a shows the results obtained bytreatment with the highest concentration of Tat. Nodifferences in IFN-g levels were detected in supernatantsfrom cocultures comprising Tat-treated, MART-1specific CD8þ T cells compared with HSA or mock-treated CD8þ T cells. Similar results have been obtainedwith lower doses of Tat and by readding Tat in cocultures(data not shown). In contrast, when MART-1 specificCD8þ T cells were incubated for 5 h in wells coated withrecombinant Tat before coculturing with B-LCLs, therelease of IFN-g was strongly reduced (Fig. 1b). Theseresults were reproduced using an HLA-B7 CD8þ T cellclone specific for HIV-1 Nef (not shown). Importantly,no differences in IFN-g production were observedwhen CD8þ T cells were preincubated in wells coatedwith other viral and nonviral proteins, such as HIV-1 Envgp120, HCV-NS3 and bovine serum albumin (data notshown).

This first set of experiments suggested that the contactbetween CD8þ T lymphocytes and immobilized Tat, butnot its soluble form, results in a defective antigen-specific activation.

The Tat-induced inhibition of IFN-g release does notdepend on its trans-activation activity but is mediated bythe interaction of Tat-RGD domain with a5b1 andavb3 integrins.

Apart from trans-activating HIV RNA transcription, Tatalso increases the transcription of several host cell genes[37,38], possibly contributing to the inhibitory effect weobserved. Although we observed that soluble Tat alonedoes not influence the peptide-dependent activation ofCD8þ T lymphocytes, synergic effects of immobilizedTat with molecules possibly detaching from coated wellscannot be formally excluded. To address this issue, B-LCLs were cocultured with MART-1 specific CD8þ Tcells, which had been preincubated in wells coated withTatcys22, a mutant lacking trans-activation activity [34].The contact with Tatcys22mutant inhibited the release ofIFN-g from the cocultures as efficiently as wt Tat(Fig. 2a), ruling out a role for Tat trans-activation in theimpaired IFN-g release by the CD8þT cells. On the basis


Co


AIDS-D-14-00246

4 AIDS 2014, Vol 00 No 00

(a)

(b)

0

100

200

HSA

10 1 0peptide ng/ml

Tat

nd

10 1 0

Ctrl

10 1 0

ndnd

Soluble Tat

pg/m

l γ-I

FN

105

CD

8+ T

cel

ls

Immobilized Tat

0

100

200

HSA TatCtrl

*

peptide ng/ml

*

pg/m

l γ-I

FN

105

CD

8+ T

cel

ls

10 1 0 10 1 010 1 0

ndndnd

Fig. 1. Contact with immobilized Tat, but not its solubleform, decreases the release of interferon-gamma from pep-tide-stimulated CD8R T cells. (a) IFN-g levels in supernatantsfrom overnight cocultures of HLA-A.02 B-LCL loaded or notwith the indicated amounts of HLA-A.02-restricted MART-1peptide, and MART-1 specific CD8þ T cells previously incu-bated for 5 h with 100 ng/ml of recombinant Tat, equalamounts of HSA or left untreated (Ctrl). Results were calcu-lated from data obtained in three independent experiments.(b) IFN-g levels in supernatants from overnight cocultures ofHLA-A.02 B-LCL loaded with the indicated amounts ofMART-1 peptide, and MART-1 specific CD8þ T cells pre-viously incubated for 5 h in wells coated with Tat, HSA or leftuncoated (Ctrl). The results were calculated from dataobtained in five independent experiments. �P<0.05; nd,IFN-g levels below the sensitivity threshold of the ELISA assay(i.e. 15 pg/ml).

of this result, the subsequent experiments wereperformed using the Tatcys22mutant.

As further steps of the study, we were interested indetermining the specificity of Tat-induced inhibition ofIFN-g release and the possible engagement of Tat withintegrins known to bind the C-terminal RGD domain ofTat. The specificity of the effect induced by Tat wasevaluated by adding anti-Tat or unrelated control Abs onTat-coated wells before CD8þ T cell seeding. Throughthe measurement of IFN-g in the supernatants ofB-LCL/CD8þ T cell cocultures, we concluded that


the anti-Tat Abs block the inhibitory effect of Tat onCD8þ T cells (Fig. 2b).

Next, we investigated the possible involvement of integrinsin the generation of Tat-induced CD8þ T cell inhibition.We noticed that a5b1 and avb3 integrins were expressedon the surface of MART-1-specific CD8þT cells (Fig. 2c),although the avb5integrin was not detectable (data notshown). Treatment of MART-1 specific CD8þT cells witha mix of Abs masking the RGD binding domains of botha5b1 and avb3 integrins [23] prior to incubation withimmobilized Tatcys22 and subsequent coculture with B-LCLs abolished the inhibition of CD8þ T cell activationinduced by Tat (Fig. 2c). Thus, the binding of Tat tointegrins is critical for the inhibition of CD8þ Tcell activation.

Interferon-gamma release from peptide-stimulated CD8R T lymphocytes is inhibited byTat present on the cell membranes of HIV-1infected cellsIn order to reproduce the previous results in a morephysiological setting, we performed cocultures of CD8þ

T lymphocytes with HIV-1 infected cells displaying Taton the cell membrane. To this end, HLA-A.02 (Fig. 3a)human breast cancer MCF-7 cells were used as bothAPCs and target of HIV-1 infection. MCF-7 cells werechosen owing to the unacceptably low cell viability weobserved when B-LCLs were infected by HIV-1pseudotyped with VSV-G (data not shown). MCF-7cells were infected with (VSV-G) wt HIV-1 or with a rev-defective (VSV-G) HIV-1 strain, as the lack of functionalRev increases multispliced transcripts thereby enhancingthe production of regulatory HIV-1 proteins, includingTat [39]. Consistently, the infection of MCF-7 cells with50 ng CAp24 equivalent of (VSV-G) Drev HIV-1 per 105cells led to higher production of both Tat and Nef thancells infected with the same amount of wt (VSV-G) HIV-1 (Fig. 3b-C). Challenge with pseudotyped HIV-1allowed the recovery of more than 90% of infected cells asshown by FACS analysis for Nef expression (Fig. 3c).Importantly, Tat efficiently bound the cell membranes ofinfected cells as shown by FACS analysis of intact cells(Fig. 3d) and by the at least partial colocalization of Tatand Env gp120 in wt HIV-1 infected cells in immuno-fluorescence analysis (Fig. 3e). Twenty-four hours afterviral challenge, cells were treated for 3 h with 10 ng/ml ofHLA-A.02 restricted MART-1 peptide. Thereafter,MART-1 specific CD8þ T cells were added in thepresence or absence of anti-Tat polyclonal Abs or, ascontrol, irrelevant species-specific Abs. Infected cocul-tures were incubated overnight, and the supernatantsassayed for IFN-g release. Figure 3f shows the resultsobtained in two independent experiments performed intriplicate. Decreased IFN-g production was clearlyobserved in cocultures of MCF-7 cells infected with(VSV-G) wt or Drev HIV-1 compared with controlcocultures. Notably, the inhibitory effect was not



AIDS-D-14-00246


(a)

(b)

0

100

200HSA wt TatCtrl

peptide ng/ml

Tatcys22

0

HSA+ctrl

Abs

HSA+αTat A

bs

Tat cys2

2+c

trl A

bs

Tat cys2

2+αTat

Abs

HSA+ctrl

Abs

HSA+αInte

grim

Abs

Tat cys2

2+c

trl A

bs

Tat cys2

2+αIn

tegr

im A

bs

100

200

**

**

Anti–α5β1 integrin Anti–αVβ3 integrin

100

101

102

103

104

FL1-H10

010

110

210

310

4

FL1-H

Cel

l cou

nts

Cou

nts

3025

2015

105

0

Cou

nts

3025

2015

105

0

*

0

100

200

*

(c)

10 1 0 10 1 010 1 0 10 1 0

ndndndnd

pg/m

l γ-I

FN

105

CD

8+ T

cel

lspg

/ml γ

-IF

N10

5 C

D8+

T c

ells

pg/m

l γ-I

FN

105

CD

8+ T

cel

ls

Fig. 2. The inhibitory effect of Tat is specific, does not rely on its transactivation activity and depends on the interaction withintegrins. (a) IFN-g levels in supernatants from overnight cocultures of HLA-A.02 B-LCL loaded with the indicated amounts ofMART-1 peptide, and MART-1 specific CD8þ T cells previously incubated for 5 h in wells coated with wt Tat, Tatcys22, HSA or leftuncoated (Ctrl). Nd, IFN-g levels below the sensitivity threshold of the ELISA assay (i.e. 15 pg/ml). (b) IFN-g levels in supernatantsfrom overnight cocultures of HLA-A.02 B-LCL loaded with 1 ng/ml of MART-1 peptide, and MART-1 specific CD8þ T cellspreviously incubated for 5 h in wells that had been coated with Tatcys22 and then treated with either anti-Tat polyclonal or controlAbs. (c) Left: FACS analysis for the presence of a5b1 and avb3 integrins on the cell membranes of MART-1 specific CD8þ T cells.Bars indicate the range of positivity, as determined by labelling cells with isotype-matched, unspecific Abs. Results arerepresentative of four independent determinations. Right: IFN-g levels in supernatants from overnight cocultures of HLA-A.02B-LCL loaded with 1 ng/ml of MART-1 peptide, and MART-1 specific CD8þ T cells preincubated with a mix of anti-a5b1 and anti-avb3 integrin mAbs or isotype-matched control Abs before the 5-h incubation in wells coated with HSA or Tatcys22. The Abs werethen readded to the cocultures. The IFN- g concentrations were calculated from data obtained in three (a) and two (b and c)independent experiments performed in triplicates. �P< 0.05.

detectable when anti-Tat Abs were added to MCF-7infected cells and re-added during coculture incubation.IFN-g was not detectable in supernatants from MCF-7cells cocultivated with MART-1 specific CD8þ T cells inthe absence of peptide (not shown).

We concluded that the contact between CD8þ T cellsand cell membrane bound Tat expressed by HIV-1infected cells can inhibit antigen-specific lymphocyteactivation.


Interaction with cell membrane associated Tatleads to an increased apoptosis in peptide-stimulated CD8R T lymphocytesEngagement of the T-cell receptor (TCR), especially inprestimulated CD8þ T cells, can trigger either activation-induced cell death (AICD) involving Fas-mediatedapoptosis [40,41], or cell survival when followed by afurther activation [42]. Thus, we investigated whether theTat-dependent inhibition of CD8þ T lymphocyte activa-tion is associated with increased apoptosis in CD8þT cells.


Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.


AIDS-D-14-00246

6 AIDS 2014, Vol 00 No 00

(a)

(b)

(c)

kDa

12

Moc

k

wtH

IV-1

Moc

k

∆rev

HIV

-1

∆rev

HIV

-1

wtH

IV-1

αGag+α Nef+α β actin Abs

αTat

kDa

Tat

NefGagCAp24

Gagp55β-actin

66

46

30

21

FITC

FL1-H10

010

110

210

310

4

Cel

l cou

nts

Cel

l cou

nts

100

8060

4020

0

αHLA-A.02

IgG

MCF -7 cells

αNef-FITC

Mock ∆rev HIV-1wtHIV-1

mfi: 49 mfi: 91

SS

C

mfi: 3.8

FL1-H10

010

110

210

310

4

SS

C-H

1000

800

600

400

200

0

FL1-H10

010

110

210

310

4

SS

C-H

1000

800

600

400

200

0

FL1-H10

010

110

210

310

4

SS

C-H

1000

800

600

400

200

0mfi: 3.8 mfi: 49 mfi: 91

Fig. 3. Contact with Tat bound to the cell membrane of HIV-1-infected cells inhibits the release of interferon-gamma frompeptide-stimulated CD8R T cells. (a) FACS analysis of the expression of HLA-A.02 on the cell surface of MCF-7 cells. (b) Westernblot analysis of lysates from MCF-7 cells infected with (VSV-G) wt or DrevHIV-1. Thirty micrograms of cell proteins were run on12% PAGE, blotted and incubated with the indicated Abs. Specific products are indicated on the left of each panel, and molecularweight markers on the right. Results are representative of four independent experiments. (c) FACS analysis for the presence ofintracytoplasmic Nef in HIV-1 infected MCF-7 cells. Cells were infected with (VSV-G) wt or DrevHIV-1. Forty-eight hours later, thecells were permeabilized and labelled with anti-Nef MATG mAb, and then with FITC-conjugated goat antimouse IgGs. Barsindicate the fluorescence intensity of uninfected or HIV-1 infected cells incubated with isotype IgGs. Mean fluorescence intensities(mfi) are indicated in each plot. The results are representative of three independent experiments. (d) FACS analysis for the presenceof Tat on the cell membranes of HIV-1 infected MCF-7 cells. The cells were infected as described for (c). Forty-eight hours later, thecells were labelled with an anti-Tat mAb, and then with FITC-conjugated goat antimouse IgGs. Bars indicate the fluorescenceintensity of uninfected or HIV-1 infected cells incubated with isotype IgGs. Mean fluorescence intensities (mfi) are indicated ineach plot. The results are representative of five independent experiments. (e) Detection of both Env gp120 and Tat on the cellmembranes of HIV-1 infected MCF-7 cells. Cells infected with (VSV-G) wt HIV-1 were first labelled with an anti-Tat mouse mAband an anti-Env gp120 human mAb, and subsequently with the indicated Alexa Fluor secondary Abs. Finally, cells were fixed andlabelled with DAPI. A representative field for each fluorescence image is reproduced and the merged image is shown on the right.(f) IFN-g levels in supernatants from MCF-7 infected cells cocultured with MART-1 specific CD8þ T cells. HLA-A.02 MCF-7 cellswere infected with (VSV-G) wt or DrevHIV-1 and, 48 h later, incubated for 1 h at 48C with polyclonal anti-Tat Abs or isotype IgGs.The cells were then cocultured with MART-1 specific CD8þ T cells in the presence of newly added anti-Tat Abs and 1 ng/ml of theHLA-A.02-restricted MART-1 peptide. After overnight incubation, supernatants were harvested, clarified and IFN-g levelsmeasured by ELISA assay. Results from two independent experiments with triplicate conditions are shown. �P<0.05.


AIDS-D-14-00246


(d)

(e)

SS

C

αTat-FITC

Mock rev HIV-1wtHIV-1

mfi: 2.6 mfi: 7.2 mfi: 20.9

αTat+Alexa fluor 488

αEnv+Alexa fluor 568 Merge

wtHIV-1 infected MCF-7 cells

DAPI

FL1-H10

010

110

210

310

4

SS

C-H

1000

800

600

400

200

0

FL1-H10

010

110

210

310

4

SS

C-H

1000

800

600

400

200

0

FL1-H10

010

110

210

310

4

SS

C-H

1000

800

600

400

200

0

Moc

k

wt H

IV-1

∆rev

HIV

-1

Moc

k

wt H

IV-1

∆ rev

HIV

-1

Moc

k

wt H

IV-1

∆rev

HIV

-1

0

75

150

IgG Anti-Tat Abs

Exp. II

*

*

0

75

150

Moc

kw

t HIV

-1

Moc

kw

t HIV

-1

Untreated IgG Anti -Tat Abs

Exp. I

∆rev

HIV

-1

∆rev

HIV

-1

**

**

pg/m

l γ-I

FN

105

CD

8+ T

cel

lspg

/ml γ

-IF

N 1

05 C

D8+

T c

ells

(f)

Fig. 3. (Continued ).

To this aim, we performed a FACS-based assay in whichCD8þ T cells were cocultured with HLA-matched B-LCLs, and cell membrane associated Tat supplied by PFA-fixed human lymphoblastoid Jurkat cells previouslyincubated with recombinant Tat. B-LCLs were treatedfor 3 h with 1–100 ng/ml of the appropriate peptide andthen cocultured with CD8þ T lymphocytes and Jurkat


cells previously incubated with either Tatcys22 or HSA,and PFA-fixed. Consistent with the previous results,cocultivation with Jurkat cells binding Tatcys22 (Jurkat-Tatcys22) led to decreased IFN-g production from CD8þ

T cells compared with control cocultures comprisingHSA-binding Jurkat cells (JurkatHSA) (Fig. 4a). Thegating strategy to identify and analyse the three cocultured


Co


AIDS-D-14-00246

8 AIDS 2014, Vol 00 No 00

0

500

250

B-LCLs X X X X

CD8+ T cells X X X X X

Jurkat cells X X

Peptide X X

*

JurkatHSA

JurkatTatcys22

SF

U/ 1

05 C

D8+

T c

ells

(a)

Fig. 4. Increased levels of apoptosis correlate with the Tat-dependent inhibition of CD8R T cell activation. (a) IFN-g specificCD8þ T cell response as detected by the ELISPOT assay after overnight cultures/cocultures of MART-1 specific CD8þ T cells withB-LCLs and Jurkat cells preincubated with Tatcys22 (JurkatTatcys22) or HSA (JurkatHSA) and PFA-fixed. The cocultures were run inthe presence or absence of the MART-1 peptide. Results are representative of two experiments performed in triplicates. (b) Dot plotFACS analysis of CD8þ T cells, B-LCLs and PFA-fixed Jurkat cells as evaluated just before coculturing. Cells were labelled with PE-conjugated anti-CD8 mAb, and then treated with Annexin V-FITC and 7-AAD. The analysis was either restricted to single celltypes, or carried out on the whole cocultures. Bottom: FACS analysis of the presence of cell membrane-associated Tatcys22 in PFA-fixed Jurkat cells using a RGD-specific anti-Tat mAb. (c) Annexin-V labelling of CD8þ T cells from triple cocultures carried out inthe absence or in the presence of the indicated amounts of peptides. PFA-fixed JurkatHSA or JurkatTatcys22 were cocultivatedovernight with HLA-A.02 MART-1 specific or HLA-B7 Nef-specific CD8þ T cells and HLA-matched B-LCLs. Results arerepresentative of two experiments performed in duplicates. �P<0.05.

cell types as well as the assessment of cell membrane boundTat on JurkatTatcys22 in flow cytometry, as measured byFACS using a RGD-specific anti-Tat mAb [43], are shownin Fig. 4b. The amount of Tat detected on JurkatTatcys22was comparable to that detected on cells treated with thesame amount of recombinant Tat and maintained at 48C(data not shown), demonstrating that the PFA treatmentdid not alter the integrin binding domain of Tat.For apoptosis evaluation, cocultures after overnightincubation were stained with PE-conjugated anti-CD8mAb, and then with AnnexinV-FITC/7-AAD.Both MART-1 specific and HIV-1 Nef-specific CD8þ

T cells were assayed. In both cases, we noticed significantlyhigher levels of apoptosis in CD8þ T cells coculturedwith JurkatTatcys22 than in those cocultured withJurkatHSA (Fig. 4c). The effect was not detectable inthe presence of the highest peptide concentration (i.e.100 ng/ml), likely because of the overwhelming peptidestimulus. Tat-dependent CD8þ T cell apoptosis wasnot observed in the absenceof the appropriate peptide (datanot shown).

These results suggest that apoptosis of CD8þ

T lymphocytes could be part of the mechanismby which cell membrane associated extracellularTat acts against the CD8þ T cell adaptive immuneresponse.


Diminished interferon-gamma release from Gag-specific CD8R T lymphocytes isolated from HIV-1 patientsTo assess the biological relevance of our findings, theeffects of immobilized Tat were investigated using CD8þ

T lymphocytes from HAART-treated, HIV-1 infectedpatients. HIV-1 specific CD8þ T lymphocytes wereexpected to persist in PBMCs from treated patients, in theabsence of high levels of spontaneous cell apoptosisfrequently observed in cultures of PBMCs from viremicpatients. PBMCs were isolated and cultured in thepresence of a mix of HIV-1 Gag peptides representingmost frequent potential T-cell epitopes [36]. In parallel,monocytes were isolated and cultured alone to obtainMDMs to be used as APCs. Seven days later, CD8þ Tlymphocytes were isolated from PBMCs. A small aliquotof PBMCs was assayed for CD8/CD25 expression. FACSanalysis (Fig. 5a) showed the presence of a subpopulationof activated CD8þ T cells. In contrast, no CD25þ/CD8þ cells were detected in unstimulated PBMCs, aswell as in PBMCs from healthy donors treated with theGag peptides (data not shown). Consistent with the verylow viral load of HAART-treated HIV-1 infected donors(i.e. <50 RNA copies/ml), the percentage of HIV-1expressing cells during the whole culture periodremained below the sensitivity threshold of the FACS-based assay for the detection of intracytoplasmic HIV-1



AIDS-D-14-00246


(b)

annexinV-FITC

CD8+Tlymphocytes

PFA-fixedJurkat cells

B-LCLs

7-A

AD

αCD

8-P

E

FL1-H10

010

110

210

310

4100

101

102

103

104

FL3

-H

FL1-H10

010

110

210

310

4100

101

102

103

104

FL2

-H

FL1-H10

010

110

210

310

4100

101

102

103

104

FL2

-HFL1-H

100

101

102

103

104

100

101

102

103

104

FL2

-H

FL1-H10

010

110

210

310

4

100

101

102

103

104

FL3

-H

FL1-H10

010

110

210

310

4100

101

102

103

104

FL3

-H

AnnexinV-FITC

7-A

AD

αCD

8-P

EJurkat

B-LCLs

B-LCLs+CD8+

Jurkat

CD8+

100

101

102

103

104

FL3

-H

FL1-H10

010

110

210

310

4

Jurkat-HSA

Jurkat-TATcys22

Cel

l cou

nts

Cou

nts

αRGD Tat-FITC

020

4060

8010

0

FL1-H10

010

110

210

310

4

100

101

102

103

104

FL2

-H

FL1-H10

010

110

210

310

4


CAp24 (data not shown). CD8þ T lymphocytes werecultivated for 5 h in either HSA or Tat-coated wells, andthen cocultured overnight with MDMs previously pulsedwith Gagmix peptides in an IFN-g ELISPOT microwellplate. Interestingly, we reproducibly observed a significantreduction in the number of spot-forming units (SFUs) incocultures of CD8þ T cells exposed to Tat compared withthose exposed to HSA (Fig. 5b). Thus, the antigen-specificresponse of CD8þ T lymphocytes from HIV-1 patients isinhibited by the presence of surface-bound Tat. No cellactivation was observed when cells from healthy donorswere cultured under the same conditions (data not shown).

This finding adds further relevance to the previous dataobtained with cell lines and CD8þ T cell clones,


meanwhile enforcing the idea that surface-bound Tat actsas an inhibitory factor of the adaptive immune response.

Discussion

We established that the interaction between extracellular,immobilized Tat and CD8þ T lymphocytes leads todecreased peptide-specific CD8þ T cell activation. Thisdecrease was stronger when the contact between CD8þ Tcells and Tat occurred in Tat-coated wells than the Tatpresent on the surface of either HIV-infected orrecombinant Tat-treated and PFA-fixed cells. Thisdifference might depend on a larger accessibility to Tatfor CD8þ T cells in coated wells, as each CD8þ T cell can


Co


AIDS-D-14-00246

10 AIDS 2014, Vol 00 No 00

ng/ml of peptide

0

20

50

Mart -1-specific CD8+ T cells

Nef -1-specific CD8+ T cells

*

*

% o

f ann

exin

V-p

ositi

veC

D8+

T c

ells

% o

f ann

exin

V-p

ositi

veC

D8+

T c

ells

ng/ml of peptide

0 1 10 100

+B-LCLs

w/o B-LCLs +B-LCLs

30

40

10

0

20

50

30

40

10

0

JurkatHSAJurkatTatcys22

*

0 1 10 1000

w/o B-LCLs

*

(c)


I

0

300

150

*

*

(a)

(b)

HSATatcys22II

*

III

αCD

8-P

E

αCD25-FITC

Ctrl HIV-1 Gag peptide mix

0.1 2.9

FL1-H10

010

110

210

310

4100

101

102

103

104

FL2

-H

FL1-H10

010

110

210

310

4100

101

102

103

104

FL2

-H

SF

U/1

06 C

D8+

T c

ells

Fig. 5. Contact with Tat inhibits interferon-gamma releasefrom CD8R T lymphocytes from HIV-1 infected patients. (a)FACS analysis of both CD25 and CD8 expression on PBMCsfrom a representative AIDS patient after 7 days of culture inthe presence or absence (Ctrl) of HIV-1 Gag peptides. Per-centages of double-positive cells are indicated. (b) IFN-gELISPOT analysis of cocultures of Gag peptide loaded macro-phages with CD8þ T lymphocytes isolated from peptide-stimulated PBMCs. Monocyte-derived macrophages weretreated for 5 h with the HIV-1 Gag peptides, and then co-cultured in ELISPOT microwells with CD8þ T lymphocytespreviously incubated for 5 h in Tat or HSA-coated wells. Theplate was developed after overnight incubation. The resultsrefer to cocultures of cells from three different HIV-1 patients(I–III). �P<0.05.

potentially interact with Tat molecules adhering to thewell, although the contact with cell membrane associatedTat relies on the relative disposition of Tat-expressing cellsin the microwells. These observations could mirror eventsoccurring in vivo. In fact, Tat bound to extracellular matrixof tissues, wherein the virus actively replicates, could be amore efficient vehicle of the inhibitory effect on CD8þ Tlymphocytes than Tat displayed on the cell surface.

The percentage of apoptotic CD8þ T lymphocytes incocultures with peptide-loaded B-LCLs was significantlyincreased in the presence of Tat. Apoptosis assays wereperformed using PFA-fixed cells. The use of an anti-TatmAb specific for the Tat RGD domain [43] showed thatPFA fixation did not alter the integrins binding site, thusconfirming the validity of this system as already describedin cell signalling experiments [44,45]. Increased apoptosiswas not observed in the absence of the peptide antigen,suggesting that TCR-signalling is required and may act insynergywith the integrin signals triggered by Tat,which, inturn, might amplify intracellular pathways or activatealternative signals whose identification deserves furtherinvestigations.

The CD8þ T cells we used resemble, at least functionally,effector memory lymphocytes. After their expansion,


they returned to a resting state upon withdrawal ofstimulatory factors. As the contact between effectorCD8þ T lymphocytes, APCs and HIV-infected cells inthe presence of extracellular Tat is conceivably a frequentevent in tissues wherein HIVactively replicates (i.e. lymphnodes, gut), our results appear of relevance in under-standing some aspects of the pathogenesis of thedysfunctions of the adaptive immune response in HIV-1 infected patients. In fact, the sustained production of Tatby HIV-infected cells may lead to an accumulation ofextracellular Tat on the cell membranes of both infectedand bystander cells, as well as extracellular matrix. Theinteraction between effector CD8þ T cells and extra-cellular Tat can trigger intracellular signals, which, insynergy with TCR stimulation, can induce apoptosis and,by consequence, selective depletion of effector CD8þ

T lymphocytes.



AIDS-D-14-00246


Anti-Tat immune responses inversely correlate with viralload and AIDS development in HIV-1 as well in HIV-2infected people [46–48]. Furthermore, anti-Tat immu-nization has been associated with improved CD8þ T cellresponses against HIV-1 Env, Candida and CEF (CMV,EBV and Flu) antigens [27], which correlated withrestoration in the number and functions of CD8þ Tcentral memory cells. On the basis of the data presentedhere, we speculate that neutralization of extracellular Tatcan reduce apoptosis within effector memory CD8þ Tcells, favouring the homeostatic restoration of the centralmemory CD8þ T cells compartment and function.

Our data point to a novel function of extracellular Tat thatfurther stresses the key role of this viral product inimmune dysregulation and AIDS pathogenesis, enforcingthe concept that the induction of an effective immuneresponse against Tat, which is rare in natural HIVinfection, may contribute to the restoration of theimmunological equilibrium in HIV-infected individuals.

Acknowledgements

C.C. performed all in-vitro CD8þT cell activation assays;B.C. carried out the ex-vivo CD8þ T cell activationassays; F.N. prepared and tested wells coated with differentrecombinant proteins; T.B. provided B-LCLs and Nef-specific CD8þ T cells; C.A. produced and titrated HIV-1preparations; A. T. prepared PBMCs from HIV-1 infectedpatients; O.L. collected and analysed clinical data of HIV-1 patients; F.E. supervised the clinic management of HIV-1 patients and participated in writing the manuscript;A.C. assayed the biologic activity of recombinant Tatpreparations and contributed to writing the manuscript;B.E. supervised the experiments and participated inwriting the manuscript; M.F. designed the experiments,analysed the data and wrote the manuscript.

We thank Lucia Gabriele and Carmen Rozera, Depart-ment of Hematology, Oncology, and Molecular Medi-cine, Istituto Superiore di Sanita, Rome, Italy, for kindlyproviding MART-1 specific CD8þ T cells. We are alsoindebted to Mario Falchi and Pietro Arciero for theirexcellent technical support.

This work was supported by grants from AIDS Nationaland ‘Ricerca Finalizzata’ projects, both from the Ministryof Health, Italy, Anti-Tat clone 4138 mAb, recombinantHIV-1 Env gp120 protein, PTE HIV-1 Gag peptides,polyclonal rabbit anti-HIV-1 Gag CAp24 #4250, and2G12 anti-HIV-1 Env gp120 human mAb were obtainedfrom NIH AIDS Research and Reference ReagentProgram.

Conflicts of interestThe authors declare no conflicts of interest.


References

1. Walker B, McMichael A. The T-cell response to HIV. ColdSpring Harb Perspect Med 2012; 2:a007054.

2. McDermott AB, Koup RA. CD8R T cells in preventing HIVinfection and disease. AIDS 2012; 26:1281–1292.

3. van Baalen CA, Pontesilli O, Huisman RC, Geretti AM, KleinMR, de Wolf F, et al. Human immunodeficiency virus type 1Rev- and Tat-specific cytotoxic T lymphocyte frequencies in-versely correlate with rapid progression to AIDS. J Gen Virol1997; 78:1913–1918.

4. Allen TM, O’Connor DH, Jing P, Dzuris JL, Mothe BR, Vogel TU,et al. Tat-specific cytotoxic T lymphocytes select for SIV escapevariants during resolution of primary viraemia. Nature 2000;407:386–390.

5. Cao J, McNevin J, Malhotra U, McElrath MJ. Evolution of CD8RT cell immunity and viral escape following acute HIV-1 infec-tion. J Immunol 2003; 171:3837–3846.

6. Zaunders J, van Bockel D. Innate and adaptive immunity inlong-term nonprogression in HIV disease. Front Immunol 2013;4:95.

7. Betts MR, Harari A. Phenotype and function of protective T cellimmune responses in HIV. Curr Opin HIV AIDS 2008; 3:349–355.

8. Hansen SG, Vieville C, Whizin N, Coyne-Johnson L, Siess DC,Drummond DD, et al. Effector memory T cell responses areassociated with protection of rhesus monkeys from mucosalsimian immunodeficiency virus challenge. Nat Med 2009;15:293–299.

9. Hansen SG, Ford JC, Lewis MS, Ventura AB, Hughes CM,Coyne-Johnson L, et al. Profound early control of highly patho-genic SIV by an effector memory T-cell vaccine. Nature 2011;473:523–527.

10. de Silva TI, Peng Y, Leligdowicz A, Zaidi I, Li L, Griffin H, et al.Correlates of T-cell-mediated viral control and phenotype ofCD8R T cells in HIV-2, a naturally contained human retroviralinfection. Blood 2013; 121:4330–4339.

11. Romani B, Engelbrecht S, Glashoff RH. Functions of Tat: theversatile protein of human immunodeficiency virus type 1.J Gen Virol 2010; 91:1–12.

12. Ensoli B, Buonaguro L, Fiorelli V, Gendelman R, Wingfield P,Gallo RC. Release, uptake, and effects of extracellular humanimmunodeficiency virus type 1 Tat protein on cell growth andviral transactivation. J Virol 1993; 67:277–287.

13. Rayne F, Debaisieux S, Yezid H, Lin YL, Mettling C, Konate K,et al. Phosphatidylinositol-(4,5)-bisphosphate enables efficientsecretion of HIV-1 Tat by infected T-cells. EMBO J 2010;29:1348–1362.

14. Chang HC, Samaniego F, Nair BC, Buonaguro L, Ensoli B. HIV-1Tat protein exits from cells via a leaderless secretory pathwayand binds to extracellular matrix-associated heparan sulfateproteoglycans through its basic region. AIDS 1997; 11:1421–1431.

15. Ensoli B, Gendelman R, Markham P, Fiorelli V, Colombini S,Raffeld M, et al. Synergy between basic fibroblast growth factorand HIV-1 Tat protein in induction of Kaposi’s sarcoma. Nature1994; 371:674–680.

16. Futaki S, Suzuki T, Ohashi W, Yagami T, Tanaka S, Ueda K,et al. Arginine-rich peptides. An abundant source of mem-brane-permeable peptides having potential as carriers forintracellular protein delivery. J Biol Chem 2001; 276:5836–5840.

17. Tyagi M, Rusnati M, Presta M, Giacca M. Internalization ofHIV-1 Tat requires cell surface heparan sulfate proteoglycans.J Biol Chem 2001; 276:3254–3261.

18. Barillari G, Gendelman R, Gallo RC, Ensoli B. The Tat protein ofhuman immunodeficiency virus type 1, a growth factor forAIDS Kaposi sarcoma and cytokine-activated vascular cells,induces adhesion of the same cell types by using integrinreceptors recognizing the RGD amino acid sequence. ProcNatl Acad Sci U S A 1993; 90:7941–7945.

19. Barillari G, Sgadari C, Fiorelli V, Samaniego F, Colombini S,Manzari V, et al. The Tat protein of human immunodeficiencyvirus type-1 promotes vascular cell growth and locomotion byengaging the alpha 5 beta 1 and alpha v beta 3 integrins and bymobilizing sequestered basic fibroblast growth factor. Blood1999; 94:663–672.


Co


AIDS-D-14-00246

12 AIDS 2014, Vol 00 No 00

20. Barillari G, Sgadari C, Palladino C, Gendelman R, Caputo A,Morris CB, et al. Inflammatory cytokines synergize with theHIV-1 Tat protein to promote angiogenesis and Kaposi’s sar-coma via induction of basic fibroblast growth factor and thealpha(v)beta(3). J Immunol 1999; 163:1929–1935.

21. Fanales-Belasio E, Moretti S, Nappi F, Barillari G, Micheletti F,Cafaro A, et al. Native HIV-1 Tat protein targets monocyte-derived dendritic cells and enhances their maturation, func-tion, and antigen-specific T cell responses. J Immunol 2002;168:197–206.

22. Fanales-Belasio E, Moretti S, Fiorelli V, Tripiciano A, PavoneCossut MR, Scoglio A, et al. HIV-1 Tat addresses dendritic cellsto induce a predominant Th1-type adaptive immune responsethat appears prevalent in the asymptomatic stage of infection.J Immunol 2009; 182:2888–2897.

23. Monini P, Cafaro A, Srivastava IK, Moretti S, Sharma VA,Andreini C, et al. HIV-1 Tat promotes integrin-mediated HIVtransmission to dendritic cells by binding Env spikes andcompetes neutralization by anti-HIV antibodies. PLoS One2012; 7:e48781.

24. Springer TA, Dustin ML. Integrin inside-out signaling and theimmunological synapse. Curr Opin Cell Biol 2012; 24:107–115.

25. Evans R, Patzak I, Svensson L, De Filippo K, Jones K, McDowallK, et al. Integrins in immunity. J Cell Sci 2009; 122:215–225.

26. Hogg N, Patzak I, Willenbrock F. The insider’s guide to leu-kocyte integrin signalling and function. Nat Rev Immunol 2011;11:416–426.

27. Ensoli B, Bellino S, Tripiciano A, Longo O, Francavilla V,Marcotullio S, et al. Therapeutic immunization with HIV-1Tat reduces immune activation and loss of regulatory T-cellsand improves immune function in subjects on HAART. PLoSOne 2010; 5:e13540.

28. Soule HD, Vazguez J, Long A, Albert S, Brennan M. A humancell line from a pleural effusion derived from a breast carci-noma. J Natl Cancer Inst 1973; 51:1409–1416.

29. Rivoltini L, Kawakami Y, Sakaguchi K, Southwood S, Sette A,Robbins PF, et al. Induction of tumor-reactive CTL from per-ipheral blood and tumor-infiltrating lymphocytes of melanomapatients by in vitro stimulation with an immunodominantpeptide of the human melanoma antigen MART-1. J Immunol1995; 154:2257–2265.

30. Di Bonito P, Grasso F, Mochi S, Petrone L, Fanales-Belasio E,Mei A, et al. Antitumor CD8R T cell immunity elicited by HIV-1-based virus-like particles incorporating HPV-16 E7 protein.Virology 2009; 395:45–55.

31. Nappi F, Schneider R, Zolotukhin A, Smulevitch S, MichalowskiD, Bear J, et al. Identification of a novel posttranscriptionalregulatory element by using a rev- and RRE-mutated humanimmunodeficiency virus type 1 DNA proviral clone as a mo-lecular trap. J Virol 2001; 75:4558–4569.

32. Federico M, Percario Z, Olivetta E, Fiorucci G, Muratori C,Micheli A, et al. HIV-1 Nef activates STAT1 in human mono-cytes/macrophages through the release of soluble factors.Blood 2001; 98:2752–2761.

33. Nappi F, Chiozzini C, Bordignon V, Borsetti A, Bellino S,Cippitelli M, et al. Immobilized HIV-1 Tat protein promotesgene transfer via a transactivation-independent mechanismwhich requires binding of Tat to viral particles. J Gene Med2009; 11:955–965.


34. Balboni PG, Bozzini R, Zucchini S, Marconi PC, Grossi MP,Caputo A, et al. Inhibition of human immunodeficiency virusreactivation from latency by a tat transdominant negativemutant. J Med Virol 1993; 41:289–295.

35. Alessandrini L, Santarcangelo AC, Olivetta E, Ferrantelli F,d’Aloja P, Pugliese K, et al. T-tropic human immunodeficiencyvirus (HIV) type 1 Nef protein enters human monocyte-macro-phages and induces resistance to HIV replication: a possiblemechanism of HIV T-tropic emergence in AIDS. J Gen Virol2000; 81:2905–2917.

36. Li F, Malhotra U, Gilbert PB, Hawkins NR, Duerr AC, McElrathJM, et al. Peptide selection for human immunodeficiency virustype 1 CTL-based vaccine evaluation. Vaccine 2006; 24:6893–6904.

37. Li JCB, Lee DCW, Cheung BKW, Lau ASY. Mechanisms for HIVTat upregulation of IL-10 and other cytokine expression: kinasesignaling and PKR-mediated immune response. FEBS Lett 2005;579:3055–3062.

38. Stettner MR, Nance JA, Wright CA, Kinoshita Y, Kim WK,Morgello S, et al. SMAD proteins of oligodendroglial cellsregulate transcription of JC virus early and late genes coordi-nately with the Tat protein of human immunodeficiency virustype 1. J Gen Virol 2009; 90:2005–2014.

39. Riggs NL, Guatelli JC. Production and characterization of high-titer stocks of rev-defective HIV-1. Virology 1996; 217:602–606.

40. Strasser A, Jost PJ, Nagata S. The many roles of FAS receptorsignaling in the immune system. Immunity 2009; 30:180–192.

41. Krammer PH, Arnold R, Lavrik IN. Life and death in peripheral Tcells. Nat Rev Immunol 2007; 7:532–542.

42. Guerrero AD, Welschhans RL, Chen M, Wang J. Cleavage ofantiapoptotic Bcl-2 family members after TCR stimulationcontributes to the decision between T cell activation andapoptosis. J Immunol 2013; 190:168–173.

43. Valvatne H, Szilvay AM, Helland DE. A monoclonal antibodydefines a novel HIV type 1 Tat domain involved in trans-cellular trans-activation. AIDS Res Hum Retroviruses 1996;12:611–619.

44. Trotta R, Puorro KA, Paroli M, Azzoni L, Abebe L, Eisenlohr LC,et al. Dependence of both spontaneous and antibody-depen-dent, granule exocytosis-mediated NK cell cytotoxicity onextracellular signal-regulated kinases. J Immunol 1998;161:6648–6656.

45. Gismondi A, Jacobelli J, Mainiero F, Paolini R, Piccoli M, Frati L,et al. Functional role for proline-rich tyrosine kinase 2 in NK cell-mediated natural cytotoxicity. J Immunol 2000; 164:2272–2276.

46. Re MC, Vignoli M, Furlini G, Gibellini D, Colangeli V, Vitone F,et al. Antibodies against full-length Tat protein and some low-molecular-weight Tat-peptides correlate with low or undetect-able viral load in HIV-1 seropositive patients. J Clin Virol 2001;21:81–89.

47. Rezza G, Fiorelli V, Dorrucci M, Ciccozzi M, Tripiciano A,Scoglio A, et al. The presence of anti-Tat antibodies is pre-dictive of long-term nonprogression to AIDS or severe immu-nodeficiency: findings in a cohort of HIV-1 seroconverters.J Infect Dis 2005; 191:1321–1324.

48. Rodriguez SK, Sarr AD, Olorunnipa O, Popper SJ, Gueye-Ndiaye A, Traore I, et al. The absence of anti-Tat antibodiesis associated with risk of disease progression in HIV-2 infec-tion. J Infect Dis 2006; 194:760–763.


Chiozzini Surface-bound Tat inhibits antigen-specific CD8 T cell activation in an integrin-dependent...

Documents

Transcript of Chiozzini Surface-bound Tat inhibits antigen-specific CD8 T cell activation in an integrin-dependent...