SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

HIV budding requires the interaction with cell factors involved in the biogenesis of exosomes. This implies the possibility that viral products undergo exosome incorporation. While this has been already described for both Gag and Nef HIV-1 proteins, no conclusive results on HIV genome have been produced so far. Here, we report that unspliced, but not single or double spliced, HIV-1 RNA species are incorporated in exosomes. Deletion mutant analysis indicated that the presence of a stretch of sequences within the 5′ end of the Gag p17 open reading frame is sufficient for HIV-1 RNA exosome incorporation. These sequences were found associating with exosomes also out of the HIV-1 context, thus indicating that the diversion towards the vesicular compartment can occur without need of additional HIV-1 sequences. Finally, the incorporation of genomic HIV-1 RNA in exosomes significantly increased when producer cells express HIV-1 defective for viral genome packaging. Manipulating infected cells to favour the selective incorporation in exosomes of genomic HIV-1 RNA might have therapeutic implications.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Exosomes are lipid bilayer vesicles of 50–100 nm which form intracellularly upon inward invagination of endosome membranes (for a recent review, see Gyorgy et al., 2011). This leads to the formation of intraluminal vesicles (ILVs) which then become part of multivesicular bodies (MVBs). These can undergo either to lysosomal degradation or to the release of their vesicular contents in the extracellular milieu upon fusion with plasma membrane. Vesicles released through this mechanism are defined exosomes. It has been proposed that exosomes can be generated also through direct extrusion of plasma membrane (Booth et al., 2006). While originally exosomes were thought to be simply addressed to secretion of waste cell material, with time it became clear that exosomes are part of the intercellular communication network (Mathivanan et al., 2010). Accordingly, they incorporate messenger (m)RNAs, microRNAs and proteins which have been found functional in target cells. There is no consensus about the existence of a selective mechanism of macromolecule incorporation in exosomes. This was because both transcriptomic and proteomic differential analyses of exosome contents generated inconsistent results. In fact, both significant (Skog et al., 2008) and not relevant (Gibbings et al., 2009) differences in the composition between exosomes and cytoplasm of exosome-producing cells have been reported.

HIV-1 buds by interacting with a number of cell factors also involved in exosome biogenesis, i.e. Tsg101, Alix and other components of the endosomal sorting complex required for transport (Usami et al., 2009). This fact, together with stringent similarities between virions and exosomes in both protein and lipid composition (Brugger et al., 2006), led to the concept that HIV buds by hijacking MVB-forming cell machinery. The convergence of exosome and HIV biogenesis would imply that viral products can be incorporated in exosomes. This was already proven for both Gag (Fang et al., 2007) and Nef HIV-1 proteins (Muratori et al., 2009; Lenassi et al., 2010). HIV-1 Gag molecules associate with exosomes seemingly by virtue of their higher-order oligomerization. On the other hand, Nef is incorporated in exosomes by anchoring exosome lipid raft microdomains, i.e. membrane regions rich in cholesterol, through both its N-terminal myristoylation and a stretch of basic amino acid within the alpha helix 1. Conversely, no conclusive evidences regarding possible association of HIV-1 RNA with exosomes have been reported until now. Considering the already documented activity of RNA molecules following exosome-mediated delivery in target cells (Valadi et al., 2007; Skog et al., 2008; Kosaka et al., 2010; Pegtel et al., 2010), the transmission by exosomes of active HIV-1 RNA molecules to bystander cells would have relevant biologic significances.

Here, we report evidences that unspliced but not spliced HIV-1 RNA can be incorporated in exosomes. This allowed the identification of RNA sequences important for exosome incorporation.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Purification, characterization and quantification of exosomes from supernatants of HIV-1-expressing cells

To investigate whether HIV-1 RNA associates with exosomes, supernatants from U937HIV-1 cells, i.e. a cell population chronically infected with the X4-tropic HTLVIIIB HIV-1 strain (Muratori et al., 2007), were harvested, and nanovesicles contained herein were concentrated by differential centrifugations. Finally, vesicles were purified by ultracentrifugation on 6–18% iodixanol density gradients to separate exosomes from viral particles.

First, we analysed the exosome distribution within the gradient fractions by assaying the acetylcholinesterase (AchE) activity. This enzyme specifically marks exosome vesicles (Rieu et al., 2000) while it is basically absent in HIV-1 virions (Cantin et al., 2008). According to previously reported data (Cantin et al., 2008), we found the strongest AchE activity in low-density fractions (i.e. fractions 4–7; Fig. 1A). On the other hand, HIV-1 particles were detected by Env gp120 ELISA in high-density fractions, i.e. starting from fraction 9 (Fig. 1A). The electron microscope analysis of vesicle input and of vesicles pooled from AchE-positive fractions showed that iodixanol gradients were apparently effective in separating nanovesicles from HIV-1 particles (Fig. 1B).

figure

Figure 1. Analysis of vesicles purified from supernatants of HIV-1-expressing cells.

A. Analysis of both AchE activity and Env gp120 contents in fractions from 6% to 18% iodixanol gradients loaded with vesicles obtained by differential centrifugations of supernatants of U937HIV-1 cells. The results are the mean values of duplicate conditions representative of four experiments for the AchE assay, and of at least 11 experiments for the HIV-1 Env gp120 detection.

B. TEM analysis of vesicle input of gradients (on the left), and of vesicles from the pools of gradient fractions with highest AchE activity (on the right). Bars indicate 100 nm.

C. Lack of HIV-1 infectivity in exosome fractions. Pools of both exosome and HIV-1 fractions were ultracentrifuged and resuspended in complete medium. The whole exosome samples or 500 ng of CAp24 equivalent of HIV-1 samples was used to challenge 5 × 104 CEM-Rev cells. As control, cells were also infected with 500 ng of CAp24 equivalent 5 × 104 cells of semi-purified HIV-1. HIV-1-infected cell cultures were carried out in the presence or absence of 1 μg ml−1 T-20. The results were calculated from seven independent experiments. Data are expressed as mean values + SD of the percentages of GFP-positive cells as compared with HIV-1-infected cells.

D. CD63 detection in exosome fractions. AchE highly positive fractions from iodixanol gradients were pooled, incubated with surfactant-free white aldehyde/sulfate latex beads, labelled with anti-CD63 mAbs and finally FACS analysed. Side and forward scatters (SSC and FSC respectively) of beads after the incubation with exosomes are depicted on the left panel together with the selected gate. The fluorescence levels of beads incubated with exosomes and then labelled with FITC-conjugated either isotype control IgG or anti-CD63 mAb are reported in the middle and right panels respectively. The results are representative of four independent measures carried out on two pools of exosomes. M1 marks the range of positivity. Percentages of positivity are indicated.

E. Detection of GM1 on pools of AchE and HIV-1 Env gp120-positive fractions from iodixanol gradients. The same amounts of AchE-positive vesicles analysed in D and scaled amounts of vesicles from HIV-1 gp120-positive fractions were bound to aldehyde/sulfate latex beads and labelled with FITC-CTX-B. Then, the former were incubated with PE-conjugated anti-CD63 mAbs, while the latter were permeabilized and incubated with PE-conjugated anti-HIV-1 CAp24 mAb. Among these, shown is the analysis representative of samples having PE fluorescence levels similar to those detected in AchE-positive vesicles. Both mean fluorescence intensities (mfi) of CTX-B and double positivity percentages are indicated. The results are representative of three independent experiments carried out on vesicles from two iodixanol gradient preparations.

F. GM1 FACS analysis of purified HIV-1 VLPs. HIV-1 VLPs purified by iodixanol gradients were quantified by anti-CAp24 ELISA, and the indicated amounts were labelled with FITC-CTX-B and scored by FACS after incubation with aldehyde/sulfate latex beads. The results are presented as percentages of fluorescent beads + SD as calculated from 15 independent experiments carried out with three different VLP preparations.

G. GM1 FACS analysis of 50 μl of iodixanol gradient fractions. The results are presented as percentages of fluorescent beads, calculated as means of duplicate measures, and are representative of three independent experiments.

Download figure to PowerPoint

The results depicted in Fig. 1A tentatively excluded obvious contaminations of viral vesicles within exosome fractions. Anyway, to exclude the possible presence of also minute amounts of infectious HIV-1, exosome fractions were tested by an infectivity assay. In detail, Rev-CEM indicator cells were challenged with pools of AchE-positive fractions recovered from iodixanol gradients loaded with vesicles from 30 ml of supernatants of U937HIV-1 cells. As control, cells were infected with 500 ng of CAp24 equivalent 5 × 104 cells of HIV-1 either recovered from high-density gradient fractions or semi-purified by 20% sucrose cushion. The treatment with exosomes did not produce increase in the expression of the GFP reporter gene starting to 96 h (Fig. 1C) until 2 weeks (not shown) post challenge. On the contrary, GFP was readily expressed in cells challenged with HIV-1 (Fig. 1C). This result further supports the idea that exosome fractions did not contain infectious viral particles.

Vesicles comprised within AchE-positive fractions were further analysed for the presence of CD63, i.e. a tetraspannin typically associating with exosomes (Escola et al., 1998). AchE-positive fractions were pooled, incubated with surfactant-free aldehyde/sulfate latex beads, labelled with anti-CD63 Abs and finally analysed by flow cytometry. The FACS analysis (Fig. 1D) indicated that CD63-positive vesicles were part of AchE-positive fractions, thus enforcing the conclusion that low-density gradient fractions indeed comprised authentic exosomes.

We next sought to analyse the gradient fractions in terms of relative amounts of vesicle contents. Mostly, exosomes are identified through the detection of associated cell proteins, while HIV-1 viral particles are routinely quantified in terms of viral protein contents, e.g. Gag CAp24, Env gp120. This precludes a comparative quantification of vesicles recovered from experimental manipulations of samples including both exosomes and HIV-1. To address this point, we attempted to quantify the vesicles by measuring the amounts of a lipid component shared by exosomes and HIV-1, i.e. the membrane-associated monosialotetrahexosylganglioside (GM1). This is a ganglioside typically comprised within lipid raft microdomains which are part of both membrane exosomes (Rabesandratana et al., 1998) and HIV-1 envelope (Nguyen and Hildreth, 2000). However, the relative amounts of GM1 associated with exosomes and HIV-1 particles are unknown. To establish this, exosomes and HIV-1 virus-like particles (VLPs) were labelled with FITC-labelled subunit B of cholera toxin (CTX-B), and either anti-CD63 mAbs for exosomes or anti-CAp24 mAbs for HIV-1. The FACS analysis of labelled vesicles bound to aldehyde/sulfate latex beads (Fig. 1E) showed that basically all CD63- and CAp24-positive vesicles bound CTX-B. The differences in the mean fluorescence intensities (mfi) revealed about 2.5-fold higher contents of GM1 in HIV-1 than in exosomes. This difference did not appear depending on the vesicle-producing cell type (not shown). We concluded that CTX-B labelling efficiently sensed GM1 in both exosomes and HIV-1 particles.

To establish the sensitivity threshold of the GM1 assay, the FACS analysis was carried out on different amounts of HIV-1 VLPs purified by iodixanol gradients (Figs 1F and S1). The lowest amount of HIV-1 VLPs reproducibly detectable was 37.5 ng of CAp24 equivalent. Within the range from 37.5 to 300 ng of CAp24, we found a linear correlation between amounts of VLPs and percentage of CTX-B-positive vesicle–bead complexes. When equal volumes (i.e. 50 μl) of fractions from iodixanol gradients loaded with vesicles obtained by differential centrifugations of U937HIV-1 supernatants were assayed for GM1 contents, two families of GM1-positive vesicles were distinguishable on the basis of the respective gradient flotation (Fig. 1G). Less dense vesicles accumulating in AchE-positive fractions can be identified as exosomes. Denser vesicles floating in Env gp120-positive fractions comprised HIV-1 particles.

In sum, iodixanol density gradients performed on vesicles harvested from supernatants of HIV-1-infected cells identified two families of GM1-positive vesicles which were identified by their flotation properties as well as by respective lipid and protein markers.

Full-length but not spliced HIV-1 RNA associates with exosomes

To look at the association of viral RNA to the vesicles recovered from iodixanol gradients, each fraction was threefold diluted with an isotonic solution and ultracentrifuged. Pelletted vesicles were then lysed and processed by oligo-dT directed retrotranscription. Samples were finally PCR-amplified with SDfw–Gagrevl primers (Fig. 2). Notably, we observed a clear signal in fractions 4–7 (Fig. 3A), i.e. the fractions where exosome vesicles accumulated at highest levels as indicated by the GM1 detection assay (Fig. 1G). Similar results were obtained using pairs of Nef- or Vif-specific primers (Fig. S2). These signals were no longer detectable when fractions were incubated for 10 min at room temperature (r.t.) with Triton X-100 at the final concentration of 0.1% v/v before ultracentrifugation (not shown), consistent with the idea that the amplified HIV-1-related sequences were associated with vesicle-based particulates. The HIV-1 specificity of products amplified from exosome fractions was demonstrated by nested PCR carried out with Gagfw1/Gagrev2 primers (Fig. 3B). Importantly, no signals were detectable in nested PCR carried out on pools of exosome fractions amplified without prior reverse transcription (Fig. 3B, lane RT−). Consistently, no amplification products were detectable when purified exosomes were assayed by PCR using U3/LTRfw and Gagrev1 primers (Fig. 3C). Hence, the PCR products detected in exosome fractions did not originate from the amplification of viral or proviral DNA. No signals were detected also when the retrotranscription was performed before exosome lysis (not shown), thus excluding the presence of HIV-1 RNA somehow adhering to the external side of exosomes.

figure

Figure 2. Mapping and sequences of primers used for PCR-based analyses. On the top, the map of primers recognizing different LTR–gag HIV-1 sequences is depicted. In the HIV-1 map, shown are the positions of the transcription start site (+1), of the four stem loops SL1 to SL4 composing the packaging site, of the major splice donor (SD), of the Gag start codon, of Δ8–87 and Δ8–126 Gag deletions and of the truncation of gag region in the pHR’ LV. On the bottom, shown are the sequences of primers, their orientations and their nucleotide mapping considering the transcription start as +1 nt. As for Gagfw2 and Gagrev4 primers, the added restriction sites, i.e. BamHI and XhoI respectively, are underlined.

Download figure to PowerPoint

figure

Figure 3. Full-length HIV-1 RNA associates with exosomes.

A. RT-PCR analysis of fractions from iodixanol gradients loaded with vesicles concentrated from supernatants of HIV-1-expressing cells. The PCR amplifications were performed using SDfw and Gagrev1 primers. MW, weight markers; Ctrl+, amplification of 100 ng of total RNA from HIV-1-expressing cells. The results are representative of at least five independent experiments.

B. Nested PCR analysis of samples amplified as described in A. One microlitre of RT-PCR-amplified samples underwent 10 cycles of additional PCR amplification using Gagfw and Gagrev2 primers. Pool exo RT−, nested PCR performed on pools of exosome fractions previously amplified by PCR as described for A, but without previous retrotranscription. Molecular weight markers (MW) are shown on the left. The results are representative of four independent experiments.

C. HIV-1 DNA detection in exosome fractions. Pools of exosome fractions were PCR-amplified using U3/LTRfw and Gagrev1 primers. As controls, 100 ng of either DNA or total RNA from HIV-1-expressing cells was used. The results are representative of two independent experiments.

D. GM1 FACS analysis of 50 μl of fractions from the iodixanol gradients loaded with vesicles from equal volumes of supernatants of HIV-1-expressing cells concentrated by either 20% sucrose cushion or differential centrifugations. Purified HIV-1 VLPs were used as reference samples (inset). The results are presented as percentages of fluorescent beads calculated as means of duplicate measures.

E. RT-PCR analysis of fractions from the iodixanol gradients described in D. The PCR amplifications were performed using SDfw and Gagrev1 primers. The results are representative of two independent experiments.

F. Sensitivity curve of RT-PCR assay carried out on 1:3 serial dilutions of 10 ng of total cell RNA from HIV-1-infected cells using SDfw as forward primer, and Gagrev1, Vifrev or Nefrev as reverse primers to amplify Gag-, Vif- and Nef-specific sequences respectively. Ctrl−, RT-PCR performed on 100 ng of total RNA from uninfected cells. MW are shown on the right.

G. Analysis of HIV-1 vif transcripts in exosomes. Samples from pools of exosome fractions were retrotranscribed (RT+), and then PCR-amplified using SDfw as forward primer and either Gagrev1 (on the left side) or Vifrev (on the right side) as reverse primers. As control, 100 ng of total RNA from both HIV-1-expressing and control uninfected cells was assayed. The amplification was carried out also without previous retrotranscription (RT−). On the right side, the migration of specific PCR products is indicated. MW are indicated on the left. The results are representative of two independent experiments.

H. Analysis of HIV-1 nef RNA in exosomes. Samples from pools of exosome fractions were retrotranscribed (RT+) and PCR-amplified using SDfw as forward primer and either Gagrev1 (on the upper side) or Nefrev (on the bottom side) as reverse primers. As control, 100 ng of total RNA from HIV-1-expressing cells was retrotranscribed and amplified. Both samples were amplified also in the absence of previous retrotranscription (RT−). On the right side, the migration of specific PCR products is indicated. The results are representative of three independent experiments. MW are indicated on the left.

I. RT-qPCR on gradient fractions carried out with SDfw and Gagrev3 primers. The HIV-1 RNA copy numbers of samples were calculated in comparison with a standard RNA from HIV-1-infected cells whose HIV-1 RNA copy number was previously accurately determined using PCR standards from the AIDS Research and Reference Reagent Program. The results are expressed as mean values + SD of the percentages of HIV-1 RNA copy numbers with respect to the fractions with highest contents of HIV-1 RNA, and were calculated from the data obtained testing samples from five independent experiments. HIV-1 RNA copy numbers at the peak fractions varied from 104 to 9 × 104 μl−1.

Download figure to PowerPoint

HIV-1 RNA was amplified from vesicles free of particles associated with HIV-1 Env gp120. This in principle did not exclude that the HIV-1 RNA could originate from contaminating, non-infectious HIV-1 lacking Env gp120. This possibility was evaluated by analysing vesicle preparations obtained through ultracentrifugation on 20% sucrose cushion of supernatants from HIV-1-infected cells. This method was already proven to efficiently exclude exosomes from the viral pellet (Park and He, 2010). Pelletted vesicles were loaded on iodixanol gradients, and the fractions were assayed for both vesicle amounts and presence of HIV-1 RNA. Fractions of gradients loaded with vesicles obtained by the differential centrifugation were used as control. As expected, the amounts of vesicles from low-density fractions appeared reduced in gradients loaded with vesicles from 20% sucrose cushion as compared with those loaded with vesicles from differential centrifugations (Fig. 3D). More importantly, the clear signals in exosome fractions of gradients loaded with vesicles from differential centrifugations were no longer detectable in low-density fractions of gradients loaded with vesicles recovered from the pellet of 20% sucrose cushions (Fig. 3E). Hence, the reduction of the exosome input led to a dramatic decrease of the HIV-1 RNA contents in low-density gradient fractions. This result runs against the hypothesis that the HIV-1 RNA detected in exosome fractions originated from whatever contaminating HIV-1 particle.

Next, we were interested in establishing whether also spliced HIV-1 RNAs associate with exosomes. To optimize the sensitivity of the RT-PCR assay, exosomes and, as control, HIV-1 fractions from iodixanol gradients were pooled before testing. The PCR assay was carried out using equal volumes of a unique oligo-dT retrotranscribed template, and using SDfw–Vifrev, SDfw–Nefrev and, as control, SDfw–Gagrev1 primers. The overall sensitivity of the assay was evaluated by amplifying dilutions of cDNA generated by oligo-dT directed retrotranscription of 10 ng of total cell RNA from U937HIV-1 cells. The RNA sequences most efficiently amplified seemed the Nef-specific ones, while the lowest signals were generated by the Gag RNA amplification (Fig. 3F). These results could be a consequence of a more efficient retrotranscription of shorter RNAs and/or of the relative amounts of the three HIV-1 RNA species in chronically infected cells (Canki et al., 2001). When the same amplification protocol was applied to exosomes, no Vif- and Nef-specific amplifications were detectable, in the presence, however, of strong Gag-specific products (Fig. 3G and H). These results strongly supported the idea that spliced HIV-1 RNAs do not associate with exosomes efficiently.

Finally, we quantified the HIV-1 RNA copy number in exosomes as compared with what detectable in purified HIV-1. To this aim, equal volumes of each fraction from iodixanol gradients were retrotranscribed and then amplified by real-time PCR (RT-qPCR) using SDfw–Gagrev3 primers. The HIV-1 RNA copy numbers were calculated by running in parallel standard samples containing previously enumerated copy numbers of unspliced HIV-1 RNA. As reported in Fig. 3I, the fraction most enriched in exosomes (see the results from the GM1 detection assay; Fig. 1G) contained around 9% of the HIV-1 RNA copy number measured in the fraction with the highest content in viral vesicles (i.e. fraction 13).

Taken together, these data indicate that full-length but not spliced HIV-1 RNA efficiently associates with exosomes produced by HIV-1-expressing cells.

Identification of sequences within the Gag p17 open reading frame important for HIV-1 RNA incorporation in exosomes

We next tried to identify the region of HIV-1 genome important for exosome incorporation. By analysing different HIV-1 mutants, the most significant results have been achieved through the comparative analysis of HIV-1 RNA contents in exosomes produced by cells expressing either Δ8–87 or Δ8–126 Gag HIV-1 mutants. These are HIV-1 molecular clones where the Gag p17 matrix open reading frame was deleted at the amino acids 8–87 and 8–126 respectively (Reil et al., 1998). 293T cells were transfected with wt, Δ8–87 or Δ8–126 Gag HIV-1 molecular clones in a way that cell cultures expressed them at similar levels, i.e. from 75% to 95% positive cells, as measured by intracytoplasmic HIV-1 Gag CAp24 FACS assay (Fig. S3). Supernatants were treated as here above described to concentrate and purify exosomes. Notably, RT-PCR analysis failed to detect unspliced HIV-1 transcripts in exosome fractions from gradients loaded with vesicles from Δ8–126 Gag HIV-1-expressing cells. On the contrary, these transcripts were readily amplified in exosomes from both wt and Δ8–87 Gag HIV-1-expressing cells (Figs 4A, B and S4). Importantly, these results were reproduced when cDNA from gradient fractions were assayed by RT-qPCR (Fig. 4C).

figure

Figure 4. Lack of HIV-1 RNA incorporation in exosomes from cells expressing the Δ8–126 Gag HIV-1.

A. GM1 FACS analysis of 50 μl of each fraction from iodixanol gradients loaded with vesicles from supernatants of cells expressing wt, Δ8–87 Gag or Δ8–126 Gag HIV-1 genomes. Purified HIV-1 VLPs were used as reference samples (inset). The results are presented as percentages of fluorescent beads calculated as means of duplicates.

B. RT-PCR analysis of samples from fractions of the iodixanol gradients described for A. PCR amplifications were performed using SDfw and Gagrev3 primers. The results are expressed through densitometric analysis of the PCR signals where the intensity of signals is expressed as arbitrary units. The results are representative of three independent experiments.

C. RT-qPCR on gradient fractions using SDfw and Gagrev3 primers. The HIV-1 RNA copy numbers were calculated as compared with the standard HIV-1 RNA described for Fig. 3. The results are expressed as mean values + SD of the percentages of HIV-1 RNA copy numbers with respect to the fractions with highest contents of HIV-1 RNA. HIV-1 RNA copy numbers at the peak fractions varied from 4 × 104 to 9.5 × 104 μl−1. The results are representative of three independent experiments.

Download figure to PowerPoint

Overall, these results suggested that sequences codifying the HIV-1 Gag 87–126 region, i.e. from nt 557 to nt 674, are important for exosome incorporation of unspliced HIV-1 RNA.

Unspliced RNA expressed by lentiviral vectors associates with exosomes

The fact that the presence of a region of Gag p17 RNA is important for HIV-1 RNA exosome incorporation does not exclude the involvement of additional viral genome sequences within the remainder gagpol sequences which are part of unspliced HIV-1 RNA. To test this possibility, we extended our investigations on RNA expressed by lentiviral vectors (LVs). In the second generation pHR’ LV, the Gag–Pol sequences are truncated at nt 663, thus leaving almost intact the whole region within the Gag p17 matrix open reading frame we identified to be involved in the HIV-1 RNA exosome incorporation. We looked at the incorporation of HIV-1-related sequences in exosomes released by TNFα-treated CEMss cells expressing the human low-affinity nerve growth factor receptor truncated in its intracytoplasmic domain (L-ΔNGFr) under the control of the 5′ HIV-1 LTR of a pHR’ LV (CEMHN cells) (Muratori et al., 2002). Total cell RNA was analysed by Northern blot for the production of HIV-1 LTR-promoted transcripts. As shown in Fig. 5A, both unspliced and spliced LV RNAs were detectable, the latter appearing, as expected, more represented. Of note, the fact that unspliced LV mRNA was produced in the absence of Rev was consistent with previous findings showing a fairly detectable production of full-length HIV-1 RNA from rev-defective HIV-1 (Hadzopoulou-Cladaras et al., 1989). The RT-PCR analysis of fractions from iodixanol gradients carried out using SDfw–Gagrev1 primers revealed the presence of unspliced transcripts in exosome fractions (Fig. 5B and C). Conversely, no spliced LV RNA was detected when the exosome samples were amplified by Gagfw2 and a reverse primer recognizing the 5′ end of L-ΔNGFr (not shown).

figure

Figure 5. Unspliced RNA expressed by LVs associates with exosomes.

A. Analysis of LV transcripts in CEMHN and parental CEMss cells after TNFα treatment. Thirty micrograms of total RNA was analysed by Northern blot using L-ΔNGFr- or GAPDH-specific probes. The migration of both unspliced and spliced LV-promoted mRNAs is indicated. MW are also indicated.

B. GM1 FACS analysis of 50 μl of fractions from iodixanol gradients loaded with vesicles released by CEMHN cells. Purified HIV-1 VLPs were used as reference samples (inset). The results are presented as percentages of fluorescent beads calculated as means of duplicate measures.

C. RT-PCR analysis of fractions from iodixanol gradients loaded with vesicles recovered from supernatants of CEMHN cells. PCR amplification was performed using SDfw and Gagrev1 primers. Ctrl+, RT-PCR carried out on 100 ng of total RNA from CEMHN cells. The migration of specific PCR products was indicated on the left. MW are indicated on the right. The results are representative of three independent experiments.

Download figure to PowerPoint

These results, together with what observed in exosomes from HIV-1-expressing cells, strongly suggested that the presence of sequences from nt 557 to 663 of HIV-1 RNA is sufficient to deliver the viral genome into exosomes.

HIV-1557–663 sequences expressed in a non-viral context are incorporated in exosomes

We next attempted to achieve a formal demonstration that HIV-1557–663 sequences are efficiently delivered to exosomes also out of the lentiviral context. To this end, these sequences were cloned in a eukaryotic vector in both sense and anti-sense orientations (Fig. 6A). Two hundred and ninety-three cells were transfected with either construct, and the expression of HIV-1557–663 sequences was checked by Northern blot assay (not shown). Vesicles from the supernatants of transiently transfected cells were harvested, concentrated by differential centrifugations and exosomes were purified by iodixanol density gradient. Exosome fractions were assayed for the presence of HIV-1557–663 sequences by RT-PCR. While no RNA sequences were amplified in exosomes from cells expressing the anti-sense construct (not shown), a signal of the expected molecular weight was observed in exosomes from cells expressing the HIV-1557–663 sequences in sense orientation (Fig. 6B). No PCR products were detectable when exosome samples were amplified without retrotranscription (not shown).

figure

Figure 6. HIV-1557–663 RNA sequences associate with exosomes.

A. Scheme of the vector expressing the HIV-1557–663 sequences in either sense (s) or anti-sense (as) orientations.

B. RT-PCR analysis of exosome fractions from iodixanol gradients loaded with vesicles concentrated from supernatants of 293 cells transiently expressing the HIV-1557–663 sequences in s orientation. Samples were amplified with Gagfw2 and Gagrev4 primers. MW are indicated on the right. The results are representative of two independent experiments.

C. RT-PCR analysis on exosomes from cells stably expressing the HIV-1557–663 sequences. Exosomes from G418 selected 293 cells expressing the HIV-1557–663 sequences in either s or as orientation were analysed by RT-PCR using either Gagfw2 (for s samples) or Gagrev4 (for as samples) as forward primers, and oligo-dT as reverse primer, without (RT−) or with (RT+) previous retrotranscription. The RT-PCR analysis was also carried out on 100 ng of total RNA from cells expressing the HIV-1557–663 sequences in either orientation and, as control, from 293 cells expressing the void vector (Ctrl−). MW are indicated on the right. The results are representative of three independent experiments made on two exosome preparations.

D. Ten cycles-nested PCR with Gagfw2 and Gagrev4 primers carried out on 1 μl of the indicated exosome samples amplified as described in C. MW are indicated on the left. The results are representative of three independent experiments made on two sets of exosome preparations.

Download figure to PowerPoint

However, it would be the case that the incorporation of HIV-1557–663 sequences in exosomes was mostly a consequence of the high levels of ectopic transcription typically occurring in transient transfection setting. To address this point, cell lines stably expressing the vectors coding HIV-1557–663 sequences were isolated by G418 selection. As expected, both 293-based cell lines constitutively produced mRNA containing the HIV-1557–663 sequences in either orientation (Fig. 6C). More importantly, RT-PCR analysis performed on exosomes recovered by differential centrifugations demonstrated that sense but not anti-sense HIV-1557–663 sequences associated with exosomes (Fig. 6C). The specificity of the detected signals was demonstrated by nested PCR (Fig. 6D).

Taken together, these data supported the hypothesis that HIV-1557–663 sequences are efficiently incorporated in exosomes.

Decreased HIV-1 RNA packaging in viral particles correlates with increased levels of viral genome incorporation in exosomes

Our results are consistent with the idea of a specific mechanism of delivery of full-length HIV-1 RNA to vesicular compartment. Looking for the biological significance of our findings, we investigated possible correlations between HIV-1 genome packaging and its association with exosomes. To this end, we evaluated the efficiency of genomic HIV-1 RNA incorporation in exosomes in the presence of reduced levels of packaging in viral particles. The HIV packaging signal is composed of four stem loops (SL1 to SL4). The deletion of SL3, which is important for the interaction with Gag nucleoprotein, drastically reduces HIV-1 RNA packaging (Ramalingam et al., 2011). Exosomes from cells expressing either wt or ΔSL3 HIV-1 molecular clones at comparable levels (Fig. 7A) were purified by iodixanol gradients and quantified in terms of GM1 contents. Then, equal volumes of exosome fractions were analysed by RT-qPCR. The HIV-1 RNA copy numbers were then normalized for the respective exosome contents. As shown in Fig. 7B, we found that the viral RNA amounts in exosomes from cells expressing ΔSL3 HIV-1 were about twofold increased with respect with exosomes from wt HIV-1-expressing cells. As expected, the analysis of high-density gradient fractions showed that, on the contrary, ΔSL3 HIV-1 RNA associated with viral particles less efficiently than wt HIV-1 RNA (Fig. 7B). According to what was already observed with wt HIV-1, no spliced HIV-1 RNA was detected in exosomes from ΔSL3 HIV-1-expressing cells (not shown).

figure

Figure 7. ΔSL3-deleted HIV-1 genome associates with exosomes more efficiently than the wt counterpart.

A. Gag CAp24 FACS analysis of 293T cells transfected with either wt or ΔSL3 HIV-1 molecular clones as compared with mock-transfected cells. The percentages of positive cells are indicated.

B. Quantification of HIV-1 RNA contents in exosomes from 293T cells transfected with either wt or ΔSL3-deleted HIV-1 molecular clones. Gradient fractions were assayed for the HIV-1 RNA contents by RT-qPCR using Gagfw2Gagrev1 primers. The HIV-1 RNA copy numbers per microlitre were adjusted for the vesicle contents as measured by the GM1 FACS assay. The results are presented as mean values + SD, and are representative of two independent experiments.

Download figure to PowerPoint

This result indicates that reduced packaging of genomic HIV-1 RNA in nascent viral particles leads to its increased diversion to exosomes.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We identified HIV-1 RNA sequences which are efficiently incorporated in exosomes. Exosomes and HIV-1 are structurally quite similar vesicles, being the density of exosomes 1.13 to 1.21 g ml−1 (Thery et al., 2001), while that of HIV-1 is 1.16 to 1.18 g ml−1 (Wang et al., 1999). Virions are slightly larger than exosomes, i.e. 100–120 nm versus 50–100 nm.

Hence, in the analysis of exosomes produced by HIV-1-expressing cells, the effective discrimination between exosomes and viral particles represented a key experimental point considering the high sensitivity of PCR techniques we used as readouts.

We took advantage of literature data which point out the effectiveness of iodixanol density gradients in separating exosomes from viral vesicles (Cantin et al., 2008). The amounts of vesicles in the fractions of iodixanol gradients were evaluated in terms of GM1 contents. To the best of our knowledge, this method was not previously described. More commonly, in fact, exosomes are identified and quantified in terms of their protein contents, which can vary according to exosome-producing cells, state of cell culture and the quality of exosome purification. Tetraspannins, integrins, heat shock proteins, β-actin, acetylcholinesterase, classes I and II MHC are the most frequently used markers of exosomes (for reviews, see Thery et al., 2009; Chaput and Thery, 2011). GM1 is part of lipid raft microdomains which are included in membranes of both exosomes (de Gassart et al., 2003) and HIV-1 (Nguyen and Hildreth, 2000). Hence, its measurement was fairly instrumental for comparing the vesicle contents in the fractions from iodixanol gradients. In addition, since GM1 is a ubiquitous molecule, its detection is expected to identify and quantify exosomes whatever the cell source.

RT-PCR assays reproducibly revealed the presence of unspliced HIV-1 RNA in exosomes. Since these sequences were amplified upon oligo-dT directed retrotranscription, we concluded that authentically full-length HIV-1 RNA can incorporate in exosomes. The detection of full-length but not spliced HIV-1 RNA in exosomes suggested the existence of a specific mechanism of viral RNA sorting into cell vesicles. In fact, one would expect that in the case of unspecific incorporation of HIV-1 RNA in exosomes as waste material, both single and double spliced HIV-1 RNA would be represented in exosomes at levels at least similar to those of full-length RNA, considering their lower size and higher levels of production in infected cells.

Data obtained using both HIV-1 deletion mutants and LVs allowed us to establish that HIV-1 RNA incorporation in exosomes relies on the presence of a region of about 100 nt within the 5′ half of Gag p17 matrix open reading frame. HIV-1557–663 sequences appear quite well conserved among both laboratory and clinic clade B HIV-1 isolates (not shown). Interestingly, HIV-1557–663 sequences comprise nearly 50% of the previously described HIV-1 instability (INS)-1 sequences (Schwartz et al., 1992a,b). These are AU-rich sequences, a feature typical of short half-life cell mRNAs. Their instability is counteracted, as for other HIV-I INS sequences, by Rev-RRE binding. Interestingly, it was found that in the absence of Rev, INS HIV-1 sequences affect the viral protein expression rather than the levels of cytoplasmic RNA (Schwartz et al., 1992a). For this reason, it was proposed that Rev acts on the intracellular localization of mRNA rather than on its stability. On the basis of our findings, we propose that INS-1 sequences divert HIV-1 full-length RNA to exosomes, thus subtracting it to polysome loading and/or viral packaging. Likely, the activity of RNA binding protein(s) specifically recognizing the HIV-1557–663 sequences is involved in targeting HIV-1 genome to exosomes.

Concerning the biological significance of our findings, exosomes incorporating the genomic HIV-1 RNA appeared to be not infectious. On the other hand, we found that decreased levels of HIV-1 RNA packaging in viral particles correlated with increased association with exosomes. Likely, this can be a consequence of the increased availability of genomic HIV-1 RNA. Considering also the already characterized intersection between HIV-1 budding and ILV/exosome biogenesis, this might at least in part mirror the competition for one or more cell component(s). More generally, it may be conceivable that the delivery of genomic HIV-1 RNA to exosomes is part of a cell defence mechanism devoted to the elimination of foreign viral genomes. A deep characterization of this mechanism would open the possibility to engineer HIV target cells in a way that the HIV genome would be preferentially diverted to cell vesicular compartments with obvious therapeutic advantages.

Finally, it would be of interest investigating whether exosomes can incorporate RNA of other virus species. Possible sequence or secondary structure homologies with the HIV-1557–663 region might imply the existence of a common mechanism of diversion of viral RNAs to exosomes. This would have relevant applicative implications in the field of early diagnosis of infectious diseases.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Molecular constructs

For the analysis of HIV-1 RNA incorporation in exosomes, the HXBH10 HIV-1 molecular clone (Terwilliger et al., 1989) and its derivatives Δ8–87 Gag and Δ8–126 Gag defective for the respective amino acid regions within the p17 matrix N-terminus (Reil et al., 1998) have been employed. The pNL4-3.Luc and its ΔSL3 derivative (Ramalingam et al., 2011) were used for analysing the influence of the HIV-1 packaging signal in viral genome exosome incorporation. For the ectopic expression of HIV-1557–663 sequences, these have been PCR-amplified from the HXBH10 HIV-1 molecular clone using Gagfw2 and Gagrev4 primers (Fig. 2), cloned in pTarget vector (Invitrogen) in either sense or antisense orientations and accurately sequenced.

Cell cultures and challenges

HEK293, HEK293T and GPR37 cells (Sparacio et al., 2001) were grown in Dulbecco's modified Eagle's medium plus 10% heat-inactivated FCS. U937, U937HIV-1 (Muratori et al., 2007) CEMss, CEMHN (Muratori et al., 2002) and Rev-CEM cells (i.e. indicator cells expressing GFP in the presence of both HIV-1 Tat and Rev) (Wu et al., 2007) were grown in Roswell Park Memorial Institute (RPMI) medium supplemented with 10% FCS. Transfections were made by the Lipofectamine 2000 (Invitrogen)-based method. Challenges of cells with either iodixanol gradient-purified exosomes or HIV-1 were carried out by spinoculation at 400 g for 30 min at r.t. using 5 × 104 cells seeded in 50 μl in 96-well plates. Then, the adsorption was prolonged for additional 2 h at 37°C and, finally, 150 μl of complete medium was added. T-20 was obtained from NIH AIDS Research and Reference Reagent Program.

Vesicle purification

Cell culture supernatants were processed on the basis of already described methods for exosome recovery. In detail, 20 to 50 ml of supernatants were centrifuged at 500 g for 10 min and filtered with 0.22 μM pore size. Then, the supernatants underwent differential centrifugations consisting in a first ultracentrifugation at 10 000 g for 30 min. Supernatants were then harvested and ultracentrifuged at 70 000 g for 1 h. The pelleted vesicles were resuspended in 1 × PBS, and ultracentrifuged again at 70 000 g for 1 h. Afterwards, the pellet was resuspended in 200 to 400 μl of 1 × PBS and subjected to discontinuous iodixanol (Axis-Shield) gradient. This was performed essentially as described (Dettenhofer and Yu, 1999). Briefly, concentrated vesicles were ultracentrifuged at 200 000 g for 1.5 h at 4°C in an SW41 Ti rotor (Beckman) through a 6% to 18% iodixanol density gradient formed by layering iodixanol in 1.2% increments. Then, 0.7 ml of fractions was collected starting from the top. Half of each fraction was then diluted with two volumes of 0.9% sodium chloride and ultracentrifuged for 30 min at 95 000 r.p.m. in a TL-100 tabletop ultracentrifuge. Finally, the pellet was resuspended in 50 μl of Tris-HCL pH 7.4 10 mM, NaCl 100 mM, EDTA 1 mM (TNE) 0.1% Triton X-100. Pools of fractions were obtained by centrifuging 0.3 ml of each fraction in TL-100 ultracentrifuge for 60 min at 95 000 r.p.m. Pellets were then resuspended in 70 μl of either TNE 0.1% Triton X-100 for PCR analysis or complete medium for infection experiments. HIV-1 VLPs were recovered from supernatants of GPR37 cells induced by ponasterone A as described (Sparacio et al., 2001). VLPs and, occasionally, supernatants from HIV-1-expressing cells were concentrated by ultracentrifugation at 70 000 g for 2.5 h on 20% sucrose cushion. VLPs were then purified by iodixanol density gradient.

Acetylcholinesterase activity assay

The vesicle-associated AchE activity was evaluated through the Amplex Red kit (Molecular Probes) following the manufacturer's recommendations. The AchE activity was measured as mU ml−1, where 1 mU is defined as the amount of enzyme which hydrolyses 1 pmol of acetylcholine to choline and acetate per minute at pH 8.0 at 37°C.

The FACS analysis of cells and nanovesicles

Intracytoplasmic HIV-1 CAp24 levels were measured by labelling cells with the KC57-RD anti-CAp24 mAb (Coulter) upon permeabilization with Cytofix/Cytoperm solutions (BD Pharmingen) as previously described (Muratori et al., 2007). For GM1 detection on vesicles, samples were incubated at 37°C for 2 h with FITC-CTX-B (Sigma-Aldrich). Therefore, 5 μl of surfactant-free white aldehyde/sulfate latex beads (Invitrogen Molecular Probes) was added and incubated overnight at r.t. on a rotating plate. Finally, the beads were washed, resuspended in 1 × PBS–2% v/v formaldehyde and FACS analysed.

Double staining of nanovesicles was performed by incubating them with 5 μl of surfactant-free white aldehyde/sulfate latex beads overnight at r.t. on a rotating plate. Afterwards, nanovesicle–bead complexes were washed and incubated at 37°C for 2 h with 1:50 diluted FITC-conjugated CTX-B. Then, the samples were washed and either incubated with PE-conjugated anti-CD63 mAb (BD Pharmingen) 1 h at 37°C for exosomes or permeabilized with Cytofix/Cytoperm solutions, and labelled for 30 min at 4°C with PE-conjugated anti HIV-1 CAp24 KC57 mAb for HIV-1. Finally, the beads were washed, resuspended in 1 × PBS–2% v/v formaldehyde and FACS analysed.

Quantification of HIV-1 proteins

The HIV-1 CAp24 contents in both VLP and HIV-1 preparations were measured by quantitative ELISA (Innogenetics). HIV-1 gp120 was detected by the Antigen Capture Assay (Advanced BioScience Laboratories).

RNA detection: RT-PCR, RT-qPCR and Northern blot analysis

Primers used for all PCR-based analysis are listed in Fig. 2. Cell RNA was extracted using the SV Total RNA Isolation System (Promega) according to the manufacturer's instructions. One hundred nanograms of purified RNA was reverse transcribed using oligo-dT primers with 15 U of avian myeloblastosis virus reverse transcriptase (AMV-RT) (Promega), and then PCR-amplified. Exosomes were lysed in TNE 0.1% v/v Triton X-100, incubated 10 min at 95°C and then reverse transcribed in 10 μl of total volume using oligo-dT primers. Then, 1 μl of cDNA was subjected to denaturation for 5 min at 95°C followed by PCR amplification for 30 cycles (30 s at 95°C, 30 s at 60°C and 30 s at 72°C) in 25 μl of total volume. Primers were used at 0.2 pmol per reaction. When oligo-dT was used as reverse primer in cDNA amplifications, the annealing temperature was adjusted to 50°C. Relative quantification of the PCR product intensity was performed using the Quantity One Analysis software of Fluor-S Image (Bio-Rad). As for RT-qPCR, 1 μl of cDNA was amplified in triplicate using Power SYBR Green PCR Master Mix (Applied Biosystems) in a total volume of 20 μl of reaction. qPCR was carried out using ABI 7500 FAST System (Applied Biosystems). Northern blot assays were carried out on total cell RNA as already described (Chelucci et al., 1995).

Electron microscopy

The vesicle analysis at the transmission electron microscope (TEM) was carried out by adsorbing droplets of exosome samples on carbon-coated Formvar nickel grids at 4°C for 1 h. Before analysis with a Philips 208 electron microscope, the grids were stained with 1% uranyl acetate solution in water.

Statistical analysis

When appropriate, data are presented as mean + standard deviation values (SD).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

This work was supported by grants from the AIDS national project of the Ministry of Health, Rome, Italy. T-20 and PCR standards were obtained from NIH AIDS Research and Reference Reagent Program. We thank A. Baur, Department of Dermatology, University of Erlangen/Numberg, Erlangen, Germany, for helpfully discussing our results. We also thank Vinayaka Prasad, Department of Microbiology and Immunology, Albert Einstein College of Medicine, Bronx, NY, for kindly providing the ΔSL3 HIV-1 molecular clone.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Booth, A.M., Fang, Y., Fallon, J.K., Yang, J.M., Hildreth, J.E., and Gould, S.J. (2006) Gag bud from endosome-like domains of the T cell plasma membrane. J Cell Biol 172: 923935.
  • Brugger, B., Glass, B., Haberkant, P., Leibrecht, I., Wieland, F.T., and Krausslich, H.G. (2006) The HIV lipidome: a raft with an unusual composition. Proc Natl Acad Sci USA 103: 26412646.
  • Canki, M., Thai, J.N., Chao, O., Ghorpade, A., Potash, M.J., and Volsky, D.J. (2001) Highly productive infection with pseudotyped human immunodeficiency virus type 1 (HIV-1) indicates no intracellular restrictions to HIV-1 replication in primary human astrocytes. J Virol 75: 79257933.
  • Cantin, R., Diou, J., Belanger, D., Tremblay, A.M., and Gilbert, C. (2008) Discrimination between exosomes and HIV-1: purification of both vesicles from cell-free supernatants. J Immunol Methods 338: 2130.
  • Chaput, N., and Thery, C. (2011) Exosomes: immune properties and potential clinical implementations. Semin Immunopathol 33: 419440.
  • Chelucci, C., Hassan, H.J., Locardi, C., Bulgarini, D., Pelosi, E., Mariani, G., et al. (1995) In vitro human immunodeficiency virus-1 infection of purified hematopoietic progenitors in single-cell culture. Blood 85: 11811187.
  • Dettenhofer, M., and Yu, X.F. (1999) Highly purified human immunodeficiency virus type 1 reveals a virtual absence of Vif in virions. J Virol 73: 14601467.
  • Escola, J.M., Kleijmeer, M.J., Stoorvogel, W., Griffith, J.M., Yoshie, O., and Geuze, H.J. (1998) Selective enrichment of tetraspan proteins on the internal vesicles of multivesicular endosomes and on exosomes secreted by human B-lymphocytes. J Biol Chem 273: 2012120127.
  • Fang, Y., Wu, N., Gan, X., Yan, W., Morrell, J.C., and Gould, S.J. (2007) Higher-order oligomerization targets plasma membrane proteins and HIV gag to exosomes. PLoS Biol 5: e158.
  • de Gassart, A., Geminard, C., Fevrier, B., Raposo, G., and Vidal, M. (2003) Lipid raft-associated protein sorting in exosomes. Blood 102: 43364344.
  • Gibbings, D.J., Ciaudo, C., Erhardt, M., and Voinnet, O. (2009) Multivesicular bodies associate with components of miRNA effector complexes and modulate miRNA activity. Nat Cell Biol 11: 11431149.
  • Gyorgy, B., Szabo, T.G., Pasztoi, M., Pal, Z., Misjak, P., Aradi, B., et al. (2011) Membrane vesicles, current state-of-the-art: emerging role of extracellular vesicles. Cell Mol Life Sci 68: 26672688.
  • Hadzopoulou-Cladaras, M., Felber, B.K., Cladaras, C., Athanassopoulos, A., Tse, A., and Pavlakis, G.N. (1989) The rev (trs/art) protein of human immunodeficiency virus type 1 affects viral mRNA and protein expression via a cis-acting sequence in the env region. J Virol 63: 12651274.
  • Kosaka, N., Iguchi, H., Yoshioka, Y., Takeshita, F., Matsuki, Y., and Ochiya, T. (2010) Secretory mechanisms and intercellular transfer of microRNAs in living cells. J Biol Chem 285: 1744217452.
  • Lenassi, M., Cagney, G., Liao, M., Vaupotic, T., Bartholomeeusen, K., Cheng, Y., et al. (2010) HIV Nef is secreted in exosomes and triggers apoptosis in bystander CD4+ T cells. Traffic 11: 110122.
  • Mathivanan, S., Ji, H., and Simpson, R.J. (2010) Exosomes: extracellular organelles important in intercellular communication. J Proteomics 73: 19071920.
  • Muratori, C., Schiavoni, I., Melucci-Vigo, G., Olivetta, E., Santarcangelo, A.C., Pugliese, K., et al. (2002) Inducible expression of the deltaNGFr/F12Nef fusion protein as a new tool for anti-human immunodeficiency virus type 1 gene therapy. Hum Gene Ther 13: 17511766.
  • Muratori, C., Sistigu, A., Ruggiero, E., Falchi, M., Bacigalupo, I., Palladino, C., et al. (2007) Macrophages transmit human immunodeficiency virus type 1 products to CD4-negative cells: involvement of matrix metalloproteinase 9. J Virol 81: 90789087.
  • Muratori, C., Cavallin, L.E., Kratzel, K., Tinari, A., De Milito, A., Fais, S., et al. (2009) Massive secretion by T cells is caused by HIV Nef in infected cells and by Nef transfer to bystander cells. Cell Host Microbe 6: 218230.
  • Nguyen, D.H., and Hildreth, J.E. (2000) Evidence for budding of human immunodeficiency virus type 1 selectively from glycolipid-enriched membrane lipid rafts. J Virol 74: 32643272.
  • Park, I.W., and He, J.J. (2010) HIV-1 is budded from CD4+ T lymphocytes independently of exosomes. Virol J 7: 234.
  • Pegtel, D.M., Cosmopoulos, K., Thorley-Lawson, D.A., van Eijndhoven, M.A., Hopmans, E.S., Lindenberg, J.L., et al. (2010) Functional delivery of viral miRNAs via exosomes. Proc Natl Acad Sci USA 107: 63286333.
  • Rabesandratana, H., Toutant, J.P., Reggio, H., and Vidal, M. (1998) Decay-accelerating factor (CD55) and membrane inhibitor of reactive lysis (CD59) are released within exosomes during In vitro maturation of reticulocytes. Blood 91: 25732580.
  • Ramalingam, D., Duclair, S., Datta, S.A., Ellington, A., Rein, A., and Prasad, V.R. (2011) RNA aptamers directed to human immunodeficiency virus type 1 Gag polyprotein bind to the matrix and nucleocapsid domains and inhibit virus production. J Virol 85: 305314.
  • Reil, H., Bukovsky, A.A., Gelderblom, H.R., and Gottlinger, H.G. (1998) Efficient HIV-1 replication can occur in the absence of the viral matrix protein. EMBO J 17: 26992708.
  • Rieu, S., Géminard, C., Rabesandratana, H., Sainte-Marie, J., and Vidal, M. (2000) Exosomes released during reticulocyte maturation bind to fibronectin via integrin alpha4beta1. Eur J Biochem 267: 583590.
  • Schwartz, S., Felber, B.K., and Pavlakis, G.N. (1992a) Distinct RNA sequences in the gag region of human immunodeficiency virus type 1 decrease RNA stability and inhibit expression in the absence of Rev protein. J Virol 66: 150159.
  • Schwartz, S., Campbell, M., Nasioulas, G., Harrison, J., Felber, B.K., and Pavlakis, G.N. (1992b) Mutational inactivation of an inhibitory sequence in human immunodeficiency virus type 1 results in Rev-independent gag expression. J Virol 66: 71767182.
  • Skog, J., Wurdinger, T., van Rijn, S., Meijer, D.H., Gainche, L., Sena-Esteves, M., et al. (2008) Glioblastoma microvesicles transport RNA and proteins that promote tumour growth and provide diagnostic biomarkers. Nat Cell Biol 10: 14701476.
  • Sparacio, S., Pfeiffer, T., Schaal, H., and Bosch, V. (2001) Generation of a flexible cell line with regulatable, high-level expression of HIV Gag/Pol particles capable of packaging HIV-derived vectors. Mol Ther 3: 602612.
  • Terwilliger, E.F., Cohen, E.A., Lu, Y.C., Sodroski, J.G., and Haseltine, W.A. (1989) Functional role of human immunodeficiency virus type 1 vpu. Proc Natl Acad Sci USA 86: 51635167.
  • Thery, C., Boussac, M., Veron, P., Ricciardi-Castagnoli, P., Raposo, G., Garin, J., and Amigorena, S. (2001) Proteomic analysis of dendritic cell-derived exosomes: a secreted subcellular compartment distinct from apoptotic vesicles. J Immunol 166: 73097318.
  • Thery, C., Ostrowski, M., and Segura, E. (2009) Membrane vesicles as conveyors of immune responses. Nat Rev Immunol 9: 581593.
  • Usami, Y., Popov, S., Popova, E., Inoue, M., Weissenhorn, W., and Göttlinger, H.G. (2009) The ESCRT pathway and HIV-1 budding. Biochem Soc Trans 37: 181184.
  • Valadi, H., Ekstrom, K., Bossios, A., Sjostrand, M., Lee, J.J., and Lotvall, J.O. (2007) Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol 9: 654659.
  • Wang, J.J., Horton, R., Varthakavi, V., Spearman, P., and Ratner, L. (1999) Formation and release of virus-like particles by HIV-1 matrix protein. AIDS 13: 281283.
  • Wu, Y., Beddall, M.H., and Marsh, J.W. (2007) Rev-dependent indicator T cell line. Curr HIV Res 5: 394402.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
cmi12046-sup-0001-figS1.TIF2351K

Fig. S1. Histograms from a representative GM1 FACS analysis of purified HIV-1 VLPs. HIV-1 VLPs purified by iodixanol gradients were quantified by anti-CAp24 ELISA, and the indicated amounts were labelled with FITC-CTX-B and scored by FACS after incubation with aldehyde/sulfate latex beads. M1 marks the range of positivity. Both percentages of fluorescent beads and mfi are indicated.

cmi12046-sup-0002-figS2.TIF2166K

Fig. S2. Vif- and Nef-specific RT-PCR analysis of fractions from iodixanol gradients loaded with vesicles concentrated from supernatants of HIV-1-expressing cells. The PCR amplifications were performed using either Viffw and Vifrev primers (on the top) or Neffw and Nefrev primers (on the bottom). Weight markers (MW) are indicated on the left. The results are representative of two independent experiments.

cmi12046-sup-0003-figS3.TIF1826K

Fig. S3. HIV-1 Gag CAp24 FACS analysis in 293T cells transfected with either wt, Δ8–87 or Δ8–126 HIV-1 molecular clones. The histograms of mock-transfected cells are also reported. M1 marks the range of positivity. The percentages of positive cells are indicated. The results are representative of three independent experiments.

cmi12046-sup-0004-figS4.TIF2166K

Fig. S4. Gel electrophoresis of samples from RT-PCR analysis of fractions from iodixanol gradients loaded with vesicles from supernatants of cells expressing wt, Δ8–87 Gag or Δ8–126 Gag HIV-1. PCR amplifications were performed using SDfw and Gagrev3 primers. Ctrl+, RT-PCR carried out on 100 ng of total RNA from wt HIV-1-expressing cells. Weight markers (MW) are indicated on the left. The results are representative of three independent experiments.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.