Traffic 2005; 6: 488–501 Blackwell Munksgaard
Copyright
#
Blackwell Munksgaard 2005
doi: 10.1111/j.1600-0854.2005.00293.x
HIV-1 Trafficking to the Dendritic Cell–T-Cell Infectious Synapse Uses a Pathway of Tetraspanin Sorting to the Immunological Synapse Eduardo Garcia1, Marjorie Pion1, Annegret Pelchen-Matthews2, Lucy Collinson2, JeanFrancois Arrighi1, Guillaume Blot1, Florence Leuba1, Jean-Michel Escola1, Nicolas Demaurex3, Mark Marsh2 and Vincent Piguet1,* 1
Department of Dermatology and Venereology, University Hospital of Geneva, Geneva, Switzerland 2 MRC Laboratory for Molecular Cell Biology and Cell Biology Unit, and Department of Biochemistry and Molecular Biology, University College London, Gower Street, London WC1E 6BT, UK 3 Department of Cell Physiology and Metabolism, University of Geneva Medical Center, Geneva, Switzerland *Corresponding author: Vincent Piguet,
[email protected] Dendritic cells (DCs) are essential components of the early events of HIV infection. Here, we characterized the trafficking pathways that HIV-1 follows during its capture by DCs and its subsequent presentation to CD4+ T cells via an infectious synapse. Immunofluorescence microscopy indicates that the virus-containing compartment in mature DCs (mDCs) co-labels for the tetraspanins CD81, CD82, and CD9 but contains little CD63 or LAMP-1. Using ratio imaging of pH-reporting fluorescent virions in live DCs, we show that HIV-1 is internalized in an intracellular endocytic compartment with a pH of 6.2. Significantly, we demonstrate that the infectivity of cell-free virus is more stable at mildly acidic pH than at neutral pH. Using electron microscopy, we confirm that HIV-1 accumulates in intracellular vacuoles that contain CD81 positive internal membranes but overlaps only partially with CD63. When allowed to contact T cells, HIV-1-loaded DCs redistribute CD81, and CD9, as well as internalized HIV-1, but not the immunological synapse markers MHC-II and T-cell receptor to the infectious synapse. Together, our results indicate that HIV-1 is internalized into a non-conventional, non-lysosomal, endocytic compartment in mDCs and further suggest that HIV-1 is able to selectively subvert components of the intracellular trafficking machinery required for formation of the DC–T-cell immunological synapse to facilitate its own cell-to-cell transfer and propagation. Key words: dendritic cells, endosomes, HIV, infectious synapse, trans infection Received 14 March 2005, revised and accepted for publication 23 March 2005, published on-line 29 April 2005
488
Dendritic cells (DCs) are believed to be crucial mediators of the early events in HIV-1 infection following sexual transmission [reviewed in (1,2)]. Dendritic cells reside in the skin and mucosal tissues in a resting ‘immature’ state until they encounter pathogens. Upon exposure to a variety of stimuli, DCs are activated to a mature antigen presenting state (3). Changes during DC maturation considerably modify the DC intracellular trafficking machinery allowing, for example, the rapid translocation of MHC-II from lysosomes to the cell surface [reviewed in (3)]. Maturation is closely linked with the migration of DCs from peripheral tissues to secondary lymphoid organs. Within these tissues, activated mature DCs (mDCs) interact with antigen-specific T cells to initiate immune responses (4,5). HIV-1 infects Langerhans cells (LCs) and other types of myeloid DCs both in vivo and in vitro [reviewed in (1,6)], although this infection is inefficient compared with CD4þ T cells. In addition, DCs can capture HIV-1 in an infectious form and transfer this virus to CD4þ T cells in trans, leading to massive levels of HIV-1 replication in DC–T-cell clusters (7). Indeed, this DC-mediated trans infection is believed to be the most efficient route for HIV-1 infection of T cells [reviewed in (1,8)]. In immature DC (iDCs) subtypes, the C-type lectin DC-SIGN (CD209) is the principal molecule mediating HIV-1 trans infection to CD4þ T cells (9,10). In DC-SIGN negative DC subsets, such as LCs, HIV capture can occur through other C-type lectins, such as the mannose receptor and Langerin (11). Importantly, HIV-1 captured by DCs remains infectious for several days in vitro, whereas free virus rapidly loses infectivity (9,12). Although significant viral degradation occurs after HIV-1 capture by DCs (13), the mechanisms that mediate the prolonged retention of DC-associated viral infectivity are currently unclear. Nevertheless, virus internalization by DCs appears to be a prerequisite for efficient transfer of HIV-1 infection to T cells in trans and may explain why trypsin treatment of HIV-exposed DCs does not decrease the efficiency of DC-mediated virus transmission to T cells (12). Upon contact with uninfected CD4þ T cells, internalized HIV-1 recycles rapidly to sites of contact between DCs and T cells (10,13,14). By analogy to the immunological synapse involved in antigen presentation (15), these sites of virus transfer have been termed ‘infectious’ or ‘virological’ synapses [reviewed in (16)]. The focusing of
HIV-1 Localization in Human Dendritic Cells
virions at the synapse may contribute to the observed efficient infection of T cells by HIV-loaded DCs (7,17). In order to understand the mechanisms involved in DCmediated trans infection in more detail, we here describe a morphological study of a compartment into which HIV-1 is sequestered following capture and internalization in monocyte-derived DCs. Using a combination of ratio imaging of pH-reporting fluorescent virions, confocal and electron microscopy, we show that HIV-1 is efficiently captured and internalized by both immature and mature DCs, at least in part via clathrin-mediated endocytosis. Surprisingly, after internalization, HIV-1 does not accumulate in lysosomes but localizes in a mildly acidic compartment (pH 6.2). Furthermore, confocal and electron microscopic studies demonstrate that this compartment is distinct from the classical late endosome/multivesicular body (MVB) compartment but contains tetraspanins such as CD81 and CD9. Finally, we show that upon contact with T cells, internalized HIV-1 redistributes rapidly to form infectious synapses in which the tetraspanins CD81 and CD9 are also observed. Given the apparent role of CD81 as a component of the immunological synapse (18,19), we suggest that HIV-1 is able to exploit a pathway responsible for the delivery of key components involved in immunological synapse formation and T-cell activation to facilitate its transfer to CD4þ T cells.
Results Mature DCs capture and transfer HIV-1 through an infectious synapse Peripheral blood monocytes were induced to differentiate into iDCs in the presence of GM-CSF and IL-4 and subsequently activated with lipopolysaccharide (LPS) to obtain mDCs (LPS-mDCs). As expected, mDCs expressed high cell-surface levels of classical markers associated with DC maturation such as CD83 (Figure 1A). To examine whether mDCs could capture HIV-1, we used a well-characterized FACS-based assay to measure viral capture by detecting intracellular accumulation of the viral p24gag protein. After incubating LPS-mDCs with HIV-1 for 2 h at 37 C, p24gag could be detected in more than 50% of the cells (Figure 1B). Similar results were obtained for iDCs (data not shown). The observed FACS signals were likely to indicate internalized HIV-1 because similar results were seen when virus-pulsed DCs were treated with trypsin to remove surface-bound HIV (data not shown). We also tested whether LPS-mDCs pulsed with virus could enhance transfer of HIV-1 infection to target cells in trans as reported (7,9). For this purpose, we incubated LPS-mDCs with HIV-1 for 2 h at 37 C and then measured viral transfer to Jurkat CD4þ T cells in a single round infection assay. As expected, HIV-1 infection could be transferred from mDCs to target cells in trans Traffic 2005; 6: 488–501
(Figure 1C). To determine whether infection of T cells occurred via formation of an infectious synapse between DCs and CD4þ T cells, we analysed cell conjugates by immunofluorescence. Monocyte-derived LPS-mDCs were loaded with HIV-1, washed, and co-cultured with Jurkat cells for 30 min before fixation, permeabilization, and staining with appropriate antibodies. In mDCs, HIV was taken up into a compartment that appeared clustered on one side of the cell (Figure 1D, left). When the mDCs encountered T cells, the virus re-distributed to the zone of contact between the DC and the T cell. Infectious synapses were considered to have formed when >75% of the virus was focused in this contact zone. After 30 min of incubation of HIV-1-pulsed LPS-mDCs and uninfected T cells, approximately 40% of mDCs were observed transferring HIV-1 through infectious synapses to CD4þ T cells as previously reported (10). Significantly, bona fide DC–Tcell immunological synapse markers [MHC-II (HLA-DR) and T-cell receptor (CD3)] were not enriched in the DC–T-cell infectious synapse (Figure 1D, center and right). HIV-1 accumulates in a mildly acidic endocytic compartment Measurement of the pH in endocytic organelles has revealed that the acidity of the lumen increases from pH 6.2–6.8 in early and recycling endosomes to 5.5–6.1 in late endosomes/MVBs and 5.0–5.5 in lysosomes (20,21). To study the fate of HIV-1 internalized by DCs, we used ratio imaging of pH-reporting fluorescent virions in live DCs. For this purpose, we treated HIV-1 virions with aldrithiol (AT)-2 to inactivate infectivity (for live experiments). Then, we labeled AT-2-treated HIV-1 virions with a pH-sensitive fluorescent probe (FITC) using a similar method to that described for adenovirus (22). A significant fraction of the AT-2-treated HIV-1 particles labeled with FITC retained their capacity to interact with HIV-1 receptors and undergo fusion (data not shown). Furthermore, AT-2-treated HIV or SIV captured by DCs can recycle to DC–T-cell infectious synapses (13), indicating that AT-2 treatment does not alter the trafficking of HIV captured by DCs. Subsequently, we incubated HIV-1-AT-2-FITC with iDCs or LPS-mDCs for 2 h at 37 C. The cells were then washed with PBS and allowed to adhere to coverslips for 1 h. The pH of the intracellular compartment containing HIV-1-AT-2FITC was assessed by ratio fluorescence imaging of the internalized pH-sensitive FITC in living cells. HIV-1-AT-2FITC accumulated in structures that were more scattered in iDCs (data not shown) and more clustered in mDCs (Figure 2A). Strikingly, when we quantified the pH of internalized FITC-AT-2-HIV in endocytic organelles in mDCs, we observed that the virus accumulated in an intracellular compartment with a mean pH of 6.12. A Gaussian fit analysis of the results confirmed that a majority of HIV-1 containing endocytic vesicles had a pH of 6.24 (Figure 2B). Similarly, in iDCs, HIV-1-AT-2-FITC accumulated in a compartment with a pH of 6.22 (data not shown). This result was significant because our previous studies, using a 489
Garcia et al.
102 fl2
0
101
101
102
103
104
100
101
102
10
103
p24gag
DC-SIGN
0 ck
100
20
Mo
100
CD83
HIV-1
30
HI V1
Mock
103
104
C Percentage of p24gag + T cells
B 128
A
D HLA-DR
HLA-DR
HIV-1
HIV-1
CD3
HIV-1 T
mDC
mDC
similar method, demonstrated that the DC-specific receptor DC-SIGN accumulated in vesicles with a pH of 5.47 in iDCs (compatible with late endosomes/lysosomes) and 6.45 in mDCs (early endosomes) (23). Thus, HIV-1-AT-2FITC captured by DCs was targeted to an intracellular endocytic compartment with an internal pH similar to that of early endosomes but distinct from late endosomes and lysosomes. Because HIV-1 accumulates in organelles with a pH of approximately 6.2, we tested the effect of different pH media on HIV-1 infectivity. Infectious virus was incubated for up to 5 days at 37 C in media adjusted to pH values ranging from 5.0 to 7.5. Surprisingly, though the infectivity of virus incubated at all pHs declined, virus treated at pH 5.5, 6.0, or 6.5 was preserved significantly longer than virus incubated at neutral or above (pH 7–7.5) or more acidic (pH 5.0) (Figure 2C). 490
mDC
T
Figure 1: Mature dendritic cells (mDCs) capture and transfer HIV1 infection to CD4+ T cells via an infectious synapse. A) FACS analysis of lipopolysaccharide-matured mDCs. LPS-mDCs were positive for DC-SIGN and CD83. B) LPSmDCs were pulsed with HIV-1 for 2 h at 37 C, then fixed, stained intracellular HIV p24gag, and analysed by FACS. About 50% of the cells were positive for p24gag. C) LPS-mDCs transfer HIV-1 infectivity to Jurkat CD4þ T cells in trans. LPS-mDCs were incubated with HIV-1 and co-cultured with noninfected Jurkat cells treated with Indinavir (1 mM). Forty-eight hours post – co-culture, viral transfer was determined by flow cytometric analysis of p24gag on CD3þ cells. D) LPS-mDCs were incubated with HIV-1 for 2 h at 37 C. HIV-1 accumulates in an intracellular ‘viral endosome’ (D, left). Upon encountering Jurkat CD4þ T cells, HIV-1 is redistributed from this intracellular compartment to the zone of contact (infectious synapse) between the DC and the CD4þ T cell (D, center and right). Immunological synapse markers [MHC-II (HLA-DR, center) and T-cell receptor (CD3, right)] do not appear enriched in the infectious synapse. This result is representative of approximately 20 infectious synapses in each condition. (green, immunostaining of HIV-1 p24gag; red (left and center), HLA-DR; and red (right), CD3] Bar ¼ 5 mm.
Next, we compared the degradation of HIV-1 in DCs with that of a well-characterized endocytic tracer, horseradish peroxidase (HRP), using a FACS-based assay. In iDCs, HRP was poorly degraded, but upon LPS-induced DC maturation, HRP was degraded at a faster rate, in agreement with the findings of (21) (Figure 2D). We also observed that the internalized HRP co-localized at least in part with LAMP-1 in iDCs and mDCs, consistent with lysosomal degradation for this protein (data not shown). Interestingly, no loss of signal was observed for HIV-1 over the first 4 h after internalization either in iDCs or mDCs. However, after 24 h the cell-associated HIV-1 signal was decreased by approximately 90% in iDCs and approximately 50% in mDCs (Figure 2D). Although we cannot rule out in this assay that loss of signal is due to HIV-1 recycling to the DC surface, it may be also due to DC-mediated viral degradation. This result indicates that HIV-1 virions are retained in iDCs and mDCs over the first Traffic 2005; 6: 488–501
HIV-1 Localization in Human Dendritic Cells B
7.0
6.0
mDC
5.0
106 105 104 103
50 40 30 20 10 0
mDC 5.0
5.5
6.0
6.5
7.0
D 5.0 5.5 6.0 6.5 7.0 medium
pH
Total signal (%)
Infectious units/ml
C
Number of FITC + vesicles
A pH
102 101
Detection threshold 0 1 2 3 4 Days post-incubation
5
180 160 140 120 100 80 60 40 20 0
0h HRP
HIV 4h 24 h
iDC
mDC
iDC
mDC
˚
Figure 2: HIV-1 accumulates in a mildly acidic dendritic cell (DC) intracellular compartment. LPS-mDCs were incubated for 2 h at 37 C with HIV-1-AT-2-FITC to allow internalization. The pH of HIV-1-AT-2-FITC positive vesicles was measured. A) A composite image of one representative LPS-mDC integrating fluorescence and pH scaled in pseudocolors (side bar) is shown. The HIV-1-AT-2-FITC-positive vesicular structures exhibit the blue coloration indicative of a pH of 6.1–6.2. Bar ¼ 5 mm. B) Distribution of HIV-1-AT-2-FITC-positive vesicles along the endocytic pH gradient in mDCs, with Gaussian fit (red line). C) Mildly acidic pH stabilizes HIV-1 infectivity. Equal aliquots of HIV-1 were mixed with DC culture medium adjusted to specific pH values ranging from 5.0 to 7.0 or with non-adjusted culture medium (pH 7.5–7.8) (medium) at 37 C for various periods of time. Infectious units per millilitre contained in supernatants were then determined. Each value represents the mean of two independent experiments. D) Quantitative decay of intracellular HRP and HIV-1 p24gag in DC-SIGN þ iDCs and LPS-mDCs. Horseradish peroxidase and p24gag amounts measured at each time point are expressed as a percentage relative to the 100% starting points. Histograms represent three independent experiments SEM.
hours and are also detectable after 24 h in the absence of viral replication. Furthermore, the loss of the HIV p24gag signal differed from that of a classical endocytic tracer targeted to lysosomes. Intracellular localization of HIV-1 captured by mDCs To further analyse the compartment in which HIV-1 accumulates after internalization by DCs, we labeled cells with several established markers of endocytic compartments, including EEA1 (early endosomes), TGN46 (trans-Golgi network), lysobisphosphatidic acid (LBPA), CD63 (late endosome/MVB), CD81, CD9, HLA-DM, MHC-II (MHC-II compartment) and LAMP-1 (lysosomes). HIV-1-GFP was incubated with LPS-mDC for 2 h at 37 C, to allow viral capture and internalization. Cells were then washed with PBS, allowed to adhere to coverslips for 1 h at 37 C, fixed, and stained with appropriate antibodies and analysed by immunofluorescence microscopy. Most of the cellular markers analysed did not show significant colocalization with HIV-1 (Figure 3A). However, some tetraspanins (CD81, CD82, CD9, and CD53) did show Traffic 2005; 6: 488–501
significant overlap with internalized HIV-1 in the LPSmDC (Figure 3A and S1 available online at http:// www.traffic.dk/suppmat/6_6c.asp). By contrast, the internalized virus showed only limited overlap with the late endosome/MVB marker CD63 and no co-staining with LBPA or LAMP-1. In iDCs, HIV-1 was transiently distributed to scattered peripheral vesicles (Figure S2A available online at http:// www.traffic.dk/suppmat/6_6c.asp) that did not co-localize with any of the marker antibodies tested (data not shown). However, after 4–5 h incubation with iDCs, the virus started to cluster in a perinuclear compartment (data not shown). This clustering became very obvious after 24 h (Figure 3B and S2 available online at http://www.traffic.dk/ suppmat/6_6c.asp), suggesting that the virus induced DC maturation. At this point, the clustered intracellular HIV in these HIV-1 treated DCs (HIV-mDCs) also co-localized with CD81, but not with late endosome or lysosome markers (CD63 and LAMP-1), with HLA-DM or with MHC-II HLA-DR (Figure 3B). Some other morphological 491
Garcia et al. A
EEA1
MHC-II
TGN46
CD63
LBPA
LAMP-1
CD81
CD9
CD53
changes similar to those observed during LPS-induced DC maturation (in the absence of HIV-1) were observed in HIVmDCs when compared to iDC, e.g. MHC-II was redistributed from intracellular compartments to the DC surface, while CD81 and CD9 were removed from the cell surface to an intracellular location (Figure 3A,B and S3 available online at http://www.traffic.dk/suppmat/6_6c.asp). Thus, after 24 h, HIV-1 induced at least partial maturation in the DC (compared to full maturation with LPS) (data not shown), consistent with other studies (24,25). To analyse the co-localization of HIV-1 with late endosomal/MVB markers more quantitatively, we used immuno fluorescence and confocal microscopy. HIV-mDCs or LPSmDCs were processed for immunofluorescent labeling as described above. In HIV-mDCs, pixel analysis indicated that approximately 90% of HIV-1 co-localized with CD81, approximately 20% of HIV-1 co-localized with CD63, and less than 10% with LAMP-1, HLA-DM, or HLA-DR (Figure 3C, center). As expected, approximately 70–80% of HLA-DM and CD63 co-localized with LAMP-1 in these cells, showing a typical late endosome/lysosome distribution (Figure 3C, right). Observations with LPS-mDCs were similar to HIV-mDCs. Pixel analysis indicated that approximately 60–80% of HIV-1-GFP co-localized with CD81, CD82, and CD9 (Figure 3C, left) and that up to 30–50% 492
Figure 3: Continued on next page.
of the total staining for CD81, CD82, CD53, and CD9 overlapped with HIV-1-GFP in LPS-mDCs (data not shown). Only approximately 5% of HIV-1 co-localized with LBPA (Figure 3C, left). HIV-1 co-localized partly with CD63 (approximately 35% of virus overlapped with CD63). However, the bulk of CD63 staining did not overlap with HIV-1 (only 5% of the total CD63 staining co-localized with HIV-1, data not shown) consistent with observations in HIV-mDCs. These data indicate (i) that internalized HIV-1 reorganizes endocytic compartments in iDCs in a manner that is similar to LPS and (ii) that HIV-1 is located in a ‘viral endosome’ that is distinct from early endosomes (defined by EEA1) or ‘classical’ late endosomes/lysosomes (defined by CD63, LBPA, and LAMP-1). This subcompartment is characterized by the presence of the tetraspanins CD81, CD82, and CD9 and the absence of LAMP-1. Of note, this CD81þ/LAMP-1– compartment is also present in LPS-mDCs (in the absence of HIV-1, see below and Figure 6). Therefore, we use in the present article the term ‘viral endosome’ operationally, pending further functional studies of this novel endocytic compartment. Analysis of the HIV-1-containing endosome compartment by electron microscopy To examine the structure of the virus-containing endosomal compartments in more detail, we analysed iDCs or Traffic 2005; 6: 488–501
HIV-1 Localization in Human Dendritic Cells B HIV-1
CD81
LAMP-1
HIV-1
CD63
LAMP-1
HIV-1
HLA-DM
LAMP-1
HIV-1
HLA-DR
LAMP-1
100
90
90
90
80
80
80
70
70
70
40 30
Percentage of HIV-1 co-localized with cellular markers (HIV-mDC)
CD 63
CD 81
-1 MP LA
HL
CD 6
CD 8
LB
Percentage of HIV-1 co-localized with cellular markers (LPS-mDC)
HL A-D R
0
3
0
A-D M
0
2
10
CD 9 CD 53
20
10
CD 81
20
10
PA
20
1
30
CD 8
40
50
R
30
60
A-D
40
50
HL
50
60
HL A-D M
60
Co-localization (%)
100
Co-localization (%)
100
CD 63
Co-localization (%)
C
Percentage of cellular markers co-localized with LAMP-1 (HIV-mDC)
Figure 3: Analysis of HIV-1 intracellular compartment. Immature DCs were incubated for 24 h at 37 C with HIV-1 (HIV-mDC), or LPS-mDCs were incubated for 2 h at 37 C with HIV-1, to allow internalization. A) LPS-mDCs loaded with HIV-1 were analysed by immunofluorescence microscopy. One representative LPS-mDC is depicted here with the corresponding cellular markers [green, HIV-1GFP; red, cellular markers; and blue, DAPI (nucleus)] Bar ¼ 5 mm. B) HIV-mDCs loaded with HIV-1 were analysed by confocal microscopy. One representative HIV-mDC is depicted here with the corresponding cellular markers (green, immunostaining for HIV-1 p24gag; red, cellular markers; and blue, LAMP-1) Bar ¼ 5 mm. C) Quantification of the percentage of HIV-1 co-localized with the cellular markers in LPS-mDCs (left) and in HIV-mDCs (center) (confocal images for LPS-mDCs and HIV-mDCs). Quantification of the percentage of LAMP-1 co-localized with the cellular markers in HIV-mDCs (right).
Traffic 2005; 6: 488–501
493
Garcia et al.
LPS-mDCs by immunolabeling and electron microscopy. Cells were pulsed with HIV-1 for 2 h at 37 C, fixed, and processed for cryosectioning and immunolabeling. Labeling with antibodies against the viral matrix protein (p17/MA) identified numerous virions as electron-dense, slightly irregular particles of diameter 100–130 nm, some of which contained a darker center representing the viral core. Virus particles were found at the cell surface, often tangled deeply among the numerous membrane protrusions and microvilli, or in pockets, folds or deeper invaginations of the plasma membrane. In addition, viruses were seen in coated vesicles, indicating that HIV-1 capture and internalization by DCs occurred at least in part via clathrin-mediated endocytosis (Figure 4). In iDCs, some labeled virus particles were seen in small vesicles (approximately 200-nm diameter) throughout the cytoplasm but frequently found close to the plasma membrane. By contrast, in the mDCs, large numbers of viruses were observed in more complex vacuoles ranging in size from 0.4 to 1.8 mm that are likely to represent the viral endosome observed by immunofluorescence (Figure 5A). These virus-containing structures often had a rounded or elongated appearance and some seemed to consist of clusters of several vacuoles, although these could be interconnected in adjacent planes of section. Although some of these virus-containing vacuoles were close to the cell surface, they did not have obvious connections to the plasma membrane; analysis of cells pulsed on ice with HRP indicated that at least 20% of the virus vacuoles were not accessible from the cell surface. The virus vacuoles on mDCs often contained other intraluminal membrane structures including small vesicles of 50–80-nm diameter resembling the intraluminal vesicles of MVBs (black arrow in Figure 5A). On some vacuoles, we observed coated structures resembling clathrin-coated pits apparently fusing into or budding away from the compartment (see Figure 5B).
When LPS-mDCs cryosections were double stained for various cellular markers and HIV p17/MA, we found that the virus-containing vacuoles consistently labeled for the tetraspanin CD81, which was usually seen on the small internal vesicles (Figure 5B, black arrows). Similarly, we could detect some CD63 on the small vesicles (Figure 5C, black arrows). Although the CD63 gold particle densities on the vesicles were comparable with the labeling seen for CD81, the bulk of the cellular CD63 was seen over more juxtanuclear MVBs and lysosome structures, which were intensely labeled but did not contain virus. Thus, as suggested by the immunofluorescence labeling (Figure S3 available online at http://www.traffic.dk/suppmat/6_6c.asp), virus-containing vacuoles represent a subpopulation of the CD63 containing structures present in these cells. The virus-containing endosomes were also weakly labeled by an antibody against MHC class II (Figure 5D). In contrast, prominent labeling for this antigen was seen at the cell surface (Figure 5D, white arrows), as expected for mDCs. The MHC class-II staining in the virus-containing vacuoles appeared to be associated mainly with the internal membranes (black arrow) and not the limiting membrane, suggesting that the virus-containing vacuoles are not continuous with the plasma membrane but are discrete cytoplasmic structures. The results described here are compared with results from confocal immunofluorescence and immunofluorescence on semithin crysection experiments summarized in Table 1. Thus, the compartment to which HIV-1 is sequestered after internalization into mDCs has the appearance of a MVB with internal membranes and small intraluminal vesicles that contain various tetraspanin molecules and some MHC class-II antigens (see Table 1). Although these vesicles have characteristics similar to the vesicles in MVBs, the virus-containing endosome appears to be distinct from the main MVB and lysosome compartment in these cells. We refer to this compartment as the ‘viral endosome’.
Figure 4: HIV is internalized by mDCs via clathrin-mediated endocytosis. Ultrathin cryosections of HIV-1-pulsed LPS-mDCs were labeled with antibodies against HIV p17 (PAG 10 nm, left-hand panel) or p17 (PAG 15 nm) plus CD81 (PAG 5 nm, centre and right panels). Virus particles could frequently be seen in coated vesicles, suggesting that HIV-1 is internalized, at least in part, through clathrindependent endocytosis. Bars ¼ 100 nm.
494
Traffic 2005; 6: 488–501
HIV-1 Localization in Human Dendritic Cells p17-10
CD81 - 5 p17 - 15
A
B
CD63 - 5 p17 - 15
MHC-II - 5 p17 - 15
C
D
Figure 5: Ultrastructure of the viral endosome compartment. A) Ultrathin cryosections of HIV-1-pulsed LPS-mDCs were labeled with antibodies against the HIV-1 matrix protein MA/p17 and PAG 10 nm. The large vacuole contains numerous labeled virus particles, while the black arrow identifies one of the small internal vesicles. (B, C, D) Sections were double labeled for HIV-1 p17 (PAG 15 nm) and the cellular markers B) CD81; C) D63; or D) MHC class II with PAG 5 nm. Black arrows show internal vesicles or membranes labeled with the markers. Note the coated buds on the limiting membrane of the vacuoles shown in B (white arrowheads). In D), strong labeling for MHC-II is observed at invaginations of the plasma membrane nearby (white arrows). Note the coated bud on this membrane (black arrowhead). Bars ¼ 200 nm.
Table 1: Summary of co-localizations between cellular markers and HIV-1 in LPS-mDCs Cellular markers
IF/confocal
IF on cryosection
EM
CD81 CD9 CD63 CD53 CD82 MHC-II LAMP-1 LBPA EEA1 TGN46
þþþ þþþ þ þ þþ – – – – –
þþþ þþþ þ ND ND – – ND ND
Yesa Yesa Yesb ND ND Yesc ND ND ND ND
þþþ, strong; þþ, medium; þ, weak, þ/–; or –, very weak or none; ND: not defined, IF: immunofluorescence, EM: electron microscopy. a A majority of the intracellular maker co-localizes with HIV-1. b The majority of CD63 is in MVB/lysosomes. Only a minority of intracellular CD63 co-localizes with HIV-1. c The majority of MHC-II is at the plasma membrane. Weak staining of intracellular MHC-II is observed only on the internal membranes of the viral endosome and never on the limiting membrane of the viral endosome.
Traffic 2005; 6: 488–501
HIV-1 and CD81 recycle to the infectious synapse HIV-1 is rapidly routed to the DC surface when cells pulsed with the virus encounter CD4þ T cells. To analyse the pathway of HIV-1 trafficking from the viral endosome to the DC–T-cell infectious synapse, we used our infectious synapse assay (see above and Figure 1D). Because HIV-1 did not co-localize in the infectious synapse either with the T-cell receptor or MHC-II but shared trafficking pathways with some tetraspanins such as CD81, we analysed the distribution of the tetraspanins CD81, CD9, and CD63 as well as LAMP-1 at the infectious synapse. In LPS-mDCs that had not been exposed to HIV, approximately 90% of the CD81 staining (quantified by confocal microscopy) was in an intracellular compartment that did not co-localize with LAMP-1, while 10% was at or close to the cell surface. CD63 co-localized extensively with LAMP-1 in the same conditions (Figure 6A, lines 1 and 2). Adding CD4þ T cells induced some re-location of CD81 to the cell surface in a small proportion of the cells but did not significantly alter the distribution of CD63 or LAMP-1 (Figure 6A, lines 3 and 4). As staining for CD81 495
Garcia et al.
–HIV-1
A
CD81
LAMP-1
CD63
LAMP-1
CD81
LAMP-1
CD63
LAMP-1
+T
+HIV-1
B
HIV-1
CD81
LAMP-1
HIV-1
CD63
LAMP-1
HIV-1
CD81
LAMP-1
+T HIV-1
CD63
LAMP-1
Figure 6: HIV-1 subverts the trafficking pathways of components of the DC–T-cell immunological synapse. A) Distribution of CD81 and LAMP-1 (Lines 1 and 3) or CD63 and LAMP-1 (Lines 2 and 4) in LPS-mDCs alone (in absence of HIV-1, Lines 1 and 2) or after incubation with Jurkat CD4þ T cells (Lines 3 and 4). B) LPS-mDCs were incubated with HIV-1 to allow internalization and incubated with Jurkat CD4þ T cells for 30 min to allow infectious synapse formation. The pattern of HIV-1, CD81, and LAMP-1 (Lines 1 and 3) or HIV-1, CD63, and LAMP-1 (Lines 2 and 4) is shown in LPS-mDCs alone (Lines 1 and 2) or after incubation with the Jurkat CD4þ T cells (Lines 3 and 4). CD81 redistributes from its intracellular pool to the infectious synapse. This result is representative of three independent experiments. (green, immunostaining of HIV-1 p24gag; red, cellular markers; and blue, LAMP-1). Bar ¼ 5 mm.
496
Traffic 2005; 6: 488–501
HIV-1 Localization in Human Dendritic Cells
and CD9 was very weak in CD4þ T cells [Figure 6 (data not shown)], our assay mainly follows CD81 and CD9 on the DC side of the synapse. In LPS-mDCs pulsed with HIV-1 (in the absence of T cells), HIV co-localized with CD81, but not with CD63 or LAMP-1 (Figure 6B, lines 1 and 2). Strikingly, when the virus-pulsed LPS-mDCs were incubated with CD4þ T cells, the intracellular CD81 and CD9 disappeared and were completely redistributed to the infectious synapses [Figures 6B (line 3) and 7]. In contrast, there was no apparent redistribution of CD63 or LAMP-1 (Figure 6B, line 4). We quantified the percentage of infectious synapses that showed redistribution and focusing of CD81 and CD9. Strikingly, 90–100% of DC–T-cell conjugates presenting virus at their zone of contact also relocated CD81 and CD9 in the synapse zone (Figure 7). Interestingly, even in the absence of virus, some DC–T-cell conjugates (approximately 30–40%) showed a partial redistribution of CD9 or CD81 from intracellular compartments to the DC–T-cell contact zone. This indicates that HIV-1 stimulates the redistribution of CD9 and CD81 to the DC–T-cell contact zone in a similar way to that which occurs during formation of antigen-dependent immunological synapses (19).
Discussion
Mock
9 CD
81 CD
63 CD
9 CD
CD
CD
81
100 90 80 70 60 50 40 30 20 10 0 63
Redistribution (%)
Our results demonstrate that, in immature and mature DCs, intact HIV-1 particles are captured and internalized into an intracellular endocytic compartment with novel properties that may facilitate cell-to-cell transmission of infectious virus. A major role for DCs in facilitating HIV-1 spread within infected individuals has been proposed
+HIV-1
Figure 7: HIV-1 stimulates the redistribution of tetraspanins to the infectious synapse. Quantification of the redistribution of tetraspanins from intracellular pools to DC–T-cell zone of contact in the presence (right) or absence of virus (left). Results are representative of three or four independent experiments including SD.
Traffic 2005; 6: 488–501
(7,17). Moreover, studies from several laboratories have indicated that endocytosis of infectious virus is important for this activity (12). However, except for the fact that HIV1 can be recycled to the specialized areas of DC–T-cell zone of contact, termed infectious or virological synapses, the fate of the internalized virus within DCs has remained unclear (10,13,14). Here, we show that the internalized virus accumulated in a clustered intracellular compartment characterized by the presence of the tetraspanins CD81, CD82, and CD9. Although this compartment contained some (though not the majority) of the cellular CD63, it was distinct from HLA-DM and LAMP-1-containing lysosomes. Immunoelectron microscopy confirmed that HIV-1 particles accumulated in intracellular vacuoles that contained some intraluminal vesicles reminiscent of exosomes. pH measurements indicated that this compartment has a mildly acidic pH, and studies with cell-free virus suggested that this is the optimum pH to maintain HIV-1 infectivity. When HIV-1-loaded DCs were allowed to contact T cells, the virus, together with the markers CD81 and CD9, was relocated to the infectious synapse. In addition, we noticed that HIV-1 treatment could induce reorganization of the endocytic compartments in iDCs similar to that observed for LPS-induced activation. HIV-1 treatment induced the translocation of MHC-II to the cell surface and the intracellular accumulation of the tetraspanins CD81 and CD9. This result is consistent with the fact that HIV might induce at least some degree of DC maturation, possibly via Toll-like receptor 8 (26), in a similar manner to LPS-induced maturation via TLR-4. Although HIV-1-induced DC maturation is not as extensive as that following LPS treatment (data not shown), several changes associated with DC maturation have been observed after HIV-1 binding, including cytokine secretion and cell migration (24,25,27). Maturation alters the endocytic trafficking in DCs, e.g. shutting down some pathways such as macropinocytosis (28). We, therefore, compared the degradation of HIV-1 in DCs to that of HRP, a well-characterized endocytic tracer. In iDCs, HRP was poorly degraded, but upon LPS-induced DC maturation, HRP was degraded at a faster rate, consistent with the finding that DC maturation activates lysosomal function (21). Interestingly, no loss of the HIV-1 p24 signal was observed over the first 4 h after internalization either in iDCs or mDCs. However, after 24 h, DC-associated HIV-1 degradation occurred faster in iDCs when compared with LPS-mDCs, in agreement with (13). Although we cannot rule out in this assay that loss of signal is due to some HIV-1 recycling to the cell surface, DC-mediated viral degradation is the most likely explanation for our results. Together, our data indicate that the properties of the viral endosome during DC developmental stages are distinct from the ‘classical’ lysosomes to which HRP is targeted. Interestingly, we could observe HIV-1 in coated 497
Garcia et al.
vesicles (Figure 4), suggesting that, at least in part, HIV-1 could reach the viral endosome by clathrin-mediated endocytosis, a pathway that is reported not to be affected by DC maturation (28). However, the precise events that allow HIV-1 to reach the viral endosome and avoid lysosomal degradation remain to be identified. One possibility is that HIV-1 uses clathrin-mediated endocytosis to reach the viral endosome directly. Alternatively, the virus may be delivered to early endosomes, or even late endosomes, and be actively sorted from these compartments to the viral endosome. After demonstrating that HIV-1 accumulates in a compartment of pH of 6.1–6.2, we tested the direct effect of media with a pH ranging from 5.0 to 7.5 on the infectivity of cell-free HIV-1. Although most infectivity was lost over time, we showed that virus incubated in mildly acidic pH medium (approximately 6.0) retained infectivity significantly longer than virus incubated in neutral/more alkaline (7.5) or more acidic conditions (5.0) (Figure 2C). These results may provide an alternative explanation for the results of Kwon et al. (12), who showed that agents that neutralize endosomal pH and affect the proper endosomal trafficking in DCs also prevent HIV-1 transmission to T cells. Correct trafficking of HIV-1 through the endocytic pathway after internalization is obviously essential for virus transmission, and perturbation of endosomal pH might influence this trafficking. However, the finding that HIV-1 infectivity is retained better at a mildly acidic pH, similar to that found in viral endosomes, raises the possibility that increasing endosomal pH could reduce the infectivity of virus sequestered in the DC viral endosome. This result is important, because even if a minimal fraction of viral infectivity is retained at pH 6.0 after 3–4 days, very small amounts of virus can be transferred from DCs to T cells in trans, a process known as ‘trans enhancement of HIV-1 infection to T cells’ (9,12,29). Nevertheless, given that the contents of endosomal compartments and extracellular culture supernatants are very different, the infectivity of HIV-1 retained within the viral endosome will require further analysis. Characterization of the compartment where HIV-1 accumulates by immunofluorescence showed that it shares some features with late endosomes/MVBs in both LPSmDCs and HIV-mDCs, but that it is clearly distinct from ‘classical’ late endosomes or lysosomes. Internalized HIV-1 did not significantly co-localize with well-characterized cellular marker proteins including EEA-1 (early endosomes), TGN46 (trans-Golgi network), and LBPA or LAMP-1(lysosomes), extending the results from others showing that HIV-1 did not co-localize with early endosomes (transferrin) or lysosomes (LAMP-1) (12,13,30). However, HIV-1 did co-localize with a number of tetraspanins (CD81, CD82, and CD9) and with a subpopulation of the cellular CD63. Interestingly, recent observations demonstrated that in HIV-infected human primary macrophages, a cell type related to DCs, assembling HIV-1 can bud directly 498
into a late endosome/MVB compartment that also contains CD63 and CD81 (31). Furthermore, the cellular machinery involved in MVB formation (the ESCRT machinery) has been found to be required to complete HIV-1 assembly (32). These observations have lead to the proposal that HIV-1 might subvert similar trafficking pathways for viral budding in macrophages and for transfer of viral infection from DCs to T cells (31,33). However, in macrophages HIV-1 buds into the endosome compartment, while in the DCs, the virus reaches its intracellular compartment by endocytosis in the absence of viral replication. We have also demonstrated that on encountering T cells, DCs can translocate HIV-1 from this intracellular compartment to the DC–T-cell infectious synapse. The presence of a synapse between a virus-carrying cell and an uninfected target cell is not restricted to HIV-1 in the DC–T-cell situation (7,10,13,14) and may well be a general mechanism of viral propagation (34,35). Cell-to-cell transmission is likely to favor HIV-1 replication because it avoids the ratelimiting step of virus diffusion prior to attachment. Furthermore, cell-to-cell transmission may reduce viral neutralization by antibodies and complement (36) and potentially allows for T-cell activation concurrently with viral infection. As such, the presence of cellular antigens implicated in T-cell activation in the infectious synapse is potentially important. Remarkably, upon contact with CD4þ T cells, HIV-1-pulsed LPS-mDCss transported their intracellular pools of virus, as well as the tetraspanins CD81 and CD9, to the infectious synapse. The tetraspanin CD81 has been linked to several functions including intracellular signaling (37) and modulation of T-cell activation (18). Interestingly, in antigen-presenting cells, CD81 facilitates MHC class-IImediated antigen presentation (38), and CD81 redistributes to the central zone of the antigen-dependent immunological synapse both on the APC side and on the T-cell side (19). Interestingly, in the HIV-1-loaded DC, bona fide immunological synapse markers such as HLA-DR and CD3 did not cluster in the infectious synapse. Further analysis is required to determine the impact of the selective recruitment of CD81 and CD9 to the DC–T-cell zone of contact (in the absence of MHC-II and of the T-cell receptor) on CD4þ T-cell activation and HIV-1 replication. The fact that HIV-1 in DCs appears to follow, at least in part, the trafficking pathway that CD81 uses to redistribute from its intracellular pools to the immunological synapse identifies a clear relationship between the DC–T-cell infectious synapse and a DC–T-cell immunological synapse and suggests that HIV-1 ‘highjacks’ a pathway involved in trafficking components of the immunological synapse, in order to mediate infection of T cells in trans. In LPS-mDCs (in the absence of HIV-1), CD81 and CD9 were also observed clustered intracellularly in a similar pattern and did not co-localize with CD63 or LAMP-1, Traffic 2005; 6: 488–501
HIV-1 Localization in Human Dendritic Cells
suggesting that HIV-1 might target a pre-existing tetraspanin-rich endosomal compartment. The function of this CD81þ/CD9þ but CD63low/LAMP-1– vesicle-containing endosome is unclear, but it might be implicated in antigen processing, e.g. antigen degradation rates in this compartment might be lower than in classical lysosomes, and this may be a way to store antigens for prolonged periods of time. Interestingly, much of the CD81 in the viral endosome was seen associated with the intraluminal exosome-like vesicles and not the limiting membrane. Similarly, the MHC class II and CD63 labeling was also mainly associated with the internal membranes. Whether these antigens are released into the synapse as exosomes remains unclear but warrants further investigation. The tetraspanin-rich compartment may allow antigen sharing between DCs by transferring some antigens from DCs to other DCs or other antigen-presenting cells and might then be exploited as an escape route for viruses such as HIV-1 to avoid lysosomal degradation. In conclusion, our studies identify a trafficking pathway that is shared by molecules that function in the DC–T-cell immunological synapse (CD81 and CD9) and by HIV-1 captured by DCs, allowing it to be transported in a retrograde manner from its viral endosome to the DC–T-cell infectious synapse. The elucidation of HIV-1 trafficking in DCs and of DC–T-cell infectious synapse formation begins to provide us with insights into the interactions between retroviruses and the highly organized endocytic machinery of DCs, cells that are central for immune responses and HIV-1 transmission.
Materials and Methods Preparation of human primary DCs Monocytes from buffy coats were obtained according to institutional guidelines of the ethical committee of the University of Geneva. Monocytes were induced to differentiate into iDC for 6 days with 50 ng/mL GM-CSF and IL-4 or into mDC by further addition of LPS (20 ng/mL) for the last 2 days (LPS-mDC). Alternatively, iDCs were ‘matured’ by pulsing them with HIV-1 for 24 h (MOI ¼ 5) and called HIV-mDC. Dendritic cells were harvested at day 6, analysed by flow cytometry, and used in subsequent assays. Additional technical details are available in (10,29).
Viral stocks Viral stocks production and viral titers were described previously (29). To track HIV-1 particles, we prepared GFP-labeled HIV-1 X4 (HIV-1-GFP) by incorporation of WxxF-GFP into virions through interaction with HIV-1-VPR in a similar manner as in (39). HIV-1-AT-2-FITC was generated using a modified version of the protocol used by Greber et al. to study adenoviral entry (22) and described in supplemental online Material and Methods available at http://www.traffic.dk/suppmat/6_6c.asp.
Antibodies and reagents Most antibodies used in this study have been previously described (29). The rabbit polyclonal anti-LAMP-1 was a gift from M. Fukuda (Cancer Research Center, La Jolla, CA, USA) (40). Additional antibodies are described in supplemental online Material and Methods available at http://www.traffic.dk/suppmat/6_6c.asp.
Traffic 2005; 6: 488–501
Flow cytometric analysis Flow cytometric analysis was performed as described (10,29).
Viral capture and transfer assays Viral capture and transfer assays were performed as described previously (10) with minor modifications available in supplemental online Material and Methods available at http://www.traffic.dk/suppmat/6_6c.asp.
PH measurement studies The pH of the organelles to which internalized HIV-1-AT-2-FITC virions were targeted was measured by ratio fluorescence imaging of a pHsensitive probe as previously described (23,41).
Variation of cell-free medium pH and effect on HIV Infectivity Aliquots of HIV-1 were added to DC culture medium adjusted to pH 5.0, 5.5, 6.0, 6.5, and 7.0 at 37 C. Viral infectivity was monitored at 1-day intervals using a single round infectivity assay on CD4þ HeLa P4-2 cells.
Proteolysis assays Degradation assays in living cells were performed as in (21) with minor modifications available in supplemental Material and Methods available at http://www.traffic.dk/suppmat/6_6c.asp.
Immunofluorescence microscopy and confocal microscopy To localize HIV-1, LPS-mDCs (105 cells/condition) were loaded with HIV-1 GFP (MOI ¼ 10) for 2 h at 37 C. HIV-mDCs were pulsed with HIV-1 (MOI ¼ 5) for 24 h, washed twice in PBS, and left to adhere on poly L-lysine-treated (Sigma-Aldrich, St. Louis, MO, USA) glass coverslips for 1 h at 37 C. Cells were then fixed 20 min at room temperature in 3% paraformaldehyde, permeabilized with 0.05% saponin (Sigma-Aldrich), and washed with PBS containing 0.2% bovine serum albumin (BSA; SigmaAldrich) and human IgG (20 mg/condition). Cells were stained with primary antibodies and secondary donkey anti-mouse coupled to rhodamine (Jackson ImmunoResearch Laboratories, West Grove, PA, USA). Nuclei were stained with DAPI (Molecular Probes, Eugene, OR, USA). Alternatively, triple labeling of HIV-mDCs was done as follows: iDCs pulsed with HIV-1 (MOI ¼ 5) were stained with primary antibodies against CD81, HLA-DM [both monoclonal and from BD PharMingen (San Diego, CA, USA)], CD63 [monoclonal (1B5)] and LAMP-1 [polyclonal; a gift from M. Fukuda (Cancer Research Center)]. After extensive washes in BSA/ saponin-containing PBS, cells were then stained with secondary donkey anti-mouse antibodies coupled to rhodamine or secondary donkey antirabbit antibodies coupled to Cy-5 (Jackson ImmunoResearch Laboratories). In order to avoid unspecific labeling, cells were incubated 20 min at room temperature in PBS containing BSA, saponin, and mouse serum (0.5 mg/mL). Finally, HIV-1-p24gag was detected using a monoclonal anti-HIV-1-p24gag (KC57) coupled to FITC (Coulter, Miami, FL, USA). Infectious synapse assays were performed as previously described (10) with minor modifications available in supplemental Material and Methods available at http://www.traffic.dk/suppmat/6_6c.asp.
Immunolabeling of cryosections for electron microscopy Immunolabeling of cryosections for electron microscopy was performed with minor modifications from (31). Details are available in supplemental online Material and Methods available at http://www.traffic.dk/suppmat/ 6_6c.asp.
499
Garcia et al.
Acknowledgments We thank Q. Sattentau, D. Trono, and U. Greber for helpful discussions. We thank J. Gruenberg for providing us with the anti-LBPA antibody (6C4) and M. Fukuda for the polyclonal anti-LAMP-1 antibody. We thank S. Arnaudeau for assistance during analysis of confocal images. This work was supported by the Swiss National Science Foundation grant no. 3345–67200.01, Leenaards Foundation, NCCR oncology and the Geneva Cancer League to VP. VP is the recipient of a ‘Professor SNF’ position (PP00A68785). MM, AP-M, and LC are supported by the UK Medical Research Council.
References 1. Steinman RM, Granelli-Piperno A, Pope M, Trumpfheller C, Ignatius R, Arrode G, Racz P, Tenner-Racz K. The interaction of immunodeficiency viruses with dendritic cells. Curr Top Microbiol Immunol 2003;276: 1–30. 2. Pope M, Haase AT. Transmission, acute HIV-1 infection and the quest for strategies to prevent infection. Nat Med 2003;9:847–852. 3. Mellman I, Steinman RM. Dendritic cells: specialized and regulated antigen processing machines. Cell 2001;106:255–258. 4. Banchereau J, Steinman RM. Dendritic cells and the control of immunity. Nature 1998;392:245–252. 5. Lanzavecchia A, Sallusto F. Regulation of T cell immunity by dendritic cells. Cell 2001;106:263–266. 6. Piguet V, Blauvelt A. Essential roles for dendritic cells in the pathogenesis and potential treatment of HIV disease. J Invest Dermatol 2002;119:365–369. 7. Pope M, Betjes MG, Romani N, Hirmand H, Cameron PU, Hoffman L, Gezelter S, Schuler G, Steinman RM. Conjugates of dendritic cells and memory T lymphocytes from skin facilitate productive infection with HIV-1. Cell 1994;78:389–398. 8. Geijtenbeek TB, van Kooyk Y. DC-SIGN: a novel HIV receptor on DCs that mediates HIV-1 transmission. Curr Top Microbiol Immunol 2003;276:31–54. 9. Geijtenbeek TB, Kwon DS, Torensma R, van Vliet SJ, van Duijnhoven GC, Middel J, Cornelissen IL, Nottet HS, KewalRamani VN, Littman DR, Figdor CG, van Kooyk Y. DC-SIGN, a dendritic cell-specific HIV-1-binding protein that enhances transinfection of T cells. Cell 2000;100:587–597. 10. Arrighi JF, Pion M, Garcia E, Escola JM, van Kooyk Y, Geijtenbeek TB, Piguet V. DC-SIGN-mediated infectious synapse formation enhances X4 HIV-1 transmission from dendritic cells to T cells. J Exp Med 2004;200:1279–1288. 11. Turville SG, Cameron PU, Handley A, Lin G, Pohlmann S, Doms RW, Cunningham AL. Diversity of receptors binding HIV on dendritic cell subsets. Nat Immunol 2002;3:975–983. 12. Kwon DS, Gregorio G, Bitton N, Hendrickson WA, Littman DR. DC-SIGN-mediated internalization of HIV is required for transenhancement of T cell infection. Immunity 2002;16:135–144. 13. Turville SG, Santos JJ, Frank I, Cameron PU, Wilkinson J, MirandaSaksena M, Dable J, Stossel H, Romani N, Piatak M Jr, Lifson JD, Pope M, Cunningham AL. Immunodeficiency virus uptake, turnover, and 2-phase transfer in human dendritic cells. Blood 2004;103: 2170–2179. 14. McDonald D, Wu L, Bohks SM, KewalRamani VN, Unutmaz D, Hope TJ. Recruitment of HIV and its receptors to dendritic cell-T cell junctions. Science 2003;300:1295–1297. 15. Dustin ML, Colman DR. Neural and immunological synaptic relations. Science 2002;298:785–789. 16. Piguet V, Sattentau Q. Dangerous liaisons at the virological synapse. J Clin Invest 2004;114:605–610.
500
17. Cameron PU, Freudenthal PS, Barker JM, Gezelter S, Inaba K, Steinman RM. Dendritic cells exposed to human immunodeficiency virus type-1 transmit a vigorous cytopathic infection to CD4þ T cells. Science 1992;257:383–387. 18. Miyazaki T, Muller U, Campbell KS. Normal development but differentially altered proliferative responses of lymphocytes in mice lacking CD81. Embo J 1997;16:4217–4225. 19. Mittelbrunn M, Yanez-Mo M, Sancho D, Ursa A, Sanchez-Madrid F. Cutting edge: dynamic redistribution of tetraspanin CD81 at the central zone of the immune synapse in both T lymphocytes and APC. J Immunol 2002;169:6691–6695. 20. Gruenberg J. The endocytic pathway: a mosaic of domains. Nat Rev Mol Cell Biol 2001;2:721–730. 21. Trombetta ES, Ebersold M, Garrett W, Pypaert M, Mellman I. Activation of lysosomal function during dendritic cell maturation. Science 2003;299:1400–1403. 22. Suomalainen M, Nakano MY, Keller S, Boucke K, Stidwill RP, Greber UF. Microtubule-dependent plus- and minus end-directed motilities are competing processes for nuclear targeting of adenovirus. J Cell Biol 1999;144:657–672. 23. Engering A, Geijtenbeek TB, van Vliet SJ, Wijers M, van Liempt E, Demaurex N, Lanzavecchia A, Fransen J, Figdor CG, Piguet V, van Kooyk Y. The dendritic cell-specific adhesion receptor DC-SIGN internalizes antigen for presentation to T cells. J Immunol 2002;168:2118–2126. 24. Granelli-Piperno A, Golebiowska A, Trumpfheller C, Siegal FP, Steinman RM. HIV-1-infected monocyte-derived dendritic cells do not undergo maturation but can elicit IL-10 production and T cell regulation. Proc Natl Acad Sci USA 2004;101:7669–7674. 25. Fonteneau JF, Larsson M, Beignon AS, McKenna K, Dasilva I, Amara A, Liu YJ, Lifson JD, Littman DR, Bhardwaj N. Human immunodeficiency virus type 1 activates plasmacytoid dendritic cells and concomitantly induces the bystander maturation of myeloid dendritic cells. J Virol 2004;78:5223–5232. 26. Heil F, Hemmi H, Hochrein H, Ampenberger F, Kirschning C, Akira S, Lipford G, Wagner H, Bauer S. Species-specific recognition of single-stranded RNA via toll-like receptor 7 and 8. Science 2004;303: 1526–1529. 27. Wilflingseder D, Mullauer B, Schramek H, Banki Z, Pruenster M, Dierich MP, Stoiber H. HIV-1-induced migration of monocyte-derived dendritic cells is associated with differential activation of MAPK pathways. J Immunol 2004;173:7497–7505. 28. Garrett WS, Chen LM, Kroschewski R, Ebersold M, Turley S, Trombetta S, Galan JE, Mellman I. Developmental control of endocytosis in dendritic cells by Cdc42. Cell 2000;102:325–334. 29. Arrighi JF, Pion M, Wiznerowicz M, Geijtenbeek TB, Garcia E, Abraham S, Leuba F, Dutoit V, Ducrey-Rundquist O, van Kooyk Y, Trono D, Piguet V. Lentivirus-mediated RNA interference of DC-SIGN expression inhibits human immunodeficiency virus transmission from dendritic cells to T cells. J Virol 2004;78:10848–10855. 30. Trumpfheller C, Park CG, Finke J, Steinman RM, Granelli-Piperno A. Cell type-dependent retention and transmission of HIV-1 by DC-SIGN. Int Immunol 2003;15:289–298. 31. Pelchen-Matthews A, Kramer B, Marsh M. Infectious HIV-1 assembles in late endosomes in primary macrophages. J Cell Biol 2003;162: 443–455. 32. Morita E, Sundquist WI. Retrovirus budding. Annu Rev Cell Dev Biol 2004;20:395–425. 33. Amara A, Littman DR. After Hrs with HIV. J Cell Biol 2003;162: 371–375. 34. Igakura T, Stinchcombe JC, Goon PK, Taylor GP, Weber JN, Griffiths GM, Tanaka Y, Osame M, Bangham CR. Spread of HTLV-I between lymphocytes by virus-induced polarization of the cytoskeleton. Science 2003;299:1713–1716. 35. Jolly C, Kashefi K, Hollinshead M, Sattentau QJ. HIV-1 cell to cell transfer across an Env-induced, Actin-dependent synapse. J Exp Med 2004;199:283–293.
Traffic 2005; 6: 488–501
HIV-1 Localization in Human Dendritic Cells 36. Ganesh L, Leung K, Lore´ K, Levin R, Panet A, Schwartz O, Koup R, Nabel G. Infection of specific dendritic cells by CCR5-tropic HIV-1 promotes cell-mediated transmission of virus resistant to broadly neutralizing antibodies. J Virol (in press). 37. Tarrant JM, Robb L, van Spriel AB, Wright MD. Tetraspanins: molecular organisers of the leukocyte surface. Trends Immunol 2003;24: 610–617. 38. Engering A, Pieters J. Association of distinct tetraspanins with MHC class II molecules at different subcellular locations in human immature dendritic cells. Int Immunol 2001;13:127–134. 39. Turelli P, Doucas V, Craig E, Mangeat B, Klages N, Evans R, Kalpana G, Trono D. Cytoplasmic recruitment of INI1 and
Traffic 2005; 6: 488–501
PML on incoming HIV preintegration complexes: interference with early steps of viral replication. Mol Cell 2001;7: 1245–1254. 40. Carlsson SR, Roth J, Piller F, Fukuda M. Isolation and characterization of human lysosomal membrane glycoproteins, h-lamp-1 and h-lamp-2. Major sialoglycoproteins carrying polylactosaminoglycan. J Biol Chem 1988;263:18911–18919. 41. Piguet V, Gu F, Foti M, Demaurex N, Gruenberg J, Carpentier JL, Trono D. Nef-induced CD4 degradation: a diacidic-based motif in Nef functions as a lysosomal targeting signal through the binding of betaCOP in endosomes. Cell 1999;97:63–73.
501
