Natural Antibodies to CCR5 from Breast Milk Block Infection of Macrophages and Dendritic Cells with Primary R5-Tropic HIV-11
http://www.100md.com
免疫学杂志 2005年第11期
Abstract
In the present study, we demonstrate that breast milk of 66% and 83% of HIV-seronegative and seropositive women, respectively, contains natural Abs of the secretory IgA and IgG isotypes directed against the CCR5 coreceptor for R5-tropic strains of HIV-1. Abs to CCR5 were affinity purified on a matrix to which a synthetic peptide corresponding to the second extracellular loop of CCR5 had been coupled. The purified Abs bound to the CCR5 peptide in a dose-dependent fashion and to both native CCR5 expressed by Chinese hamster ovary cells transfected with CCR5 gene, macrophages, and immature dendritic cells. Although the avidity differed, the amount of anti-CCR5 Abs did not significantly differ between breast milk of HIV-seropositive and -seronegative women. Purified anti-CCR5 Abs inhibited up to 75% infection of macrophages and dendritic cells with HIVBaL and HIVJR-CSF. Our observations provide evidence for a role of natural Abs to CCR5 in breast milk in controlling transmissibility of HIV through breastfeeding.
Introduction
The CCR5 chemokine receptor is the major coreceptor that is associated with mucosal transmission of R5-tropic HIV-1 during sexual intercourse and postnatal transmission through breastfeeding (1, 2, 3). The importance of CCR5 for HIV-1 transmission is evidenced by the observation that individuals homozygous for a defective CCR5 allele remain uninfected despite repeated exposure to HIV (4). The CCR5 molecule is thus a target for novel therapeutic strategies aimed at blocking HIV-1 entry into cells (5, 6, 7). We have previously demonstrated that natural Abs from therapeutic preparations of IgG (i.v. Ig) contain natural Abs directed against CCR5 that inhibit infection of human macrophages and CD4+ T lymphocytes with laboratory and primary R5-tropic isolates of HIV-1 in vitro (8). Breast milk contains high amounts of immunoglobulins, predominantly of the secretory IgA (sIgA) 4 and to a lesser degree of the IgG and sIgM isotypes that are produced in the absence of deliberate immunization and independently of exposure to Ags. Most natural Abs in the breast milk of healthy women are self-reactive Abs (9, 10).
In the present study, we demonstrate that breast milk contains Abs directed against CCR5 molecule. Following affinity purification, these Abs were shown to bind to the native CCR5 molecule and to inhibit infection of macrophages and dendritic cells by R5-tropic HIV-1.
Materials and Methods
Inhibition of HIV-1 infection by Abs to CCR5 peptide
Macrophages and dendritic cells (5 x 105 cells) in 48-well plates were infected with 5 ng/ml p24 HIVBal, HIVJR-CSF, and HIVNDK for 3 h at 37°C. The cells were washed and then cultured for 6 days. Supernatants were collected every 2 days, and HIV p24 levels were measured by ELISA. In some experiments, cells were preincubated with purified anti-CCR5 Abs (100, 250, 500, and 1000 μg/ml) or with corresponding amounts of anti-CCR5-depleted immunoglobulins used as negative control. As positive control of inhibition of HIV infection, purified anti-CCR5 from i.v. Ig, or the chemokines RANTES and SDF-1 (500 ng/ml) were added to cells before infection with HIV.
Results
Human breast milk contains natural Abs directed against CCR5
We first investigated the reactivity with the CCR5 peptide of immunoglobulins in each of the 18 breast milk samples from HIV-seronegative women and 12 samples from HIV-seropositive women by ELISA. Fifteen of 18 (83%) and 8 of 12 (66%) samples contained over 1 μg/ml anti-CCR5 Ig (Fig. 1). Based on the total Ig content of breast milk samples from HIV-seronegative and -seropositive women, the mean specific anti-CCR5 activity was calculated as being 0.24 and 0.30 in breast milk samples of HIV-seronegative and HIV-seropositive women, respectively.
As shown in Fig. 3A, Abs purified from breast milk of seropositive women exhibited an 10-fold stronger anti-CCR5 reactivity than those purified from pooled milk of seronegative women. Anti-CCR5 IgG that had been affinity-purified from i.v. Ig, used as a positive control, also bound with lesser intensity to the CCR5 peptide than affinity-purified Abs from breast milk of HIV-seropositive women. No binding to CCR5 peptide was observed when anti-CCR5 completely depleted the fraction of breast milk was tested. Affinity-purified anti-CCR5 Abs from breast milk of healthy and HIV-seropositive women and i.v. Ig reacted specifically with the CCR5 peptide but not with irrelevant peptide corresponding to a sequence of DC-SIGN molecule. Furthermore, addition of CCR5 peptide to mAb 2D7, that recognize the sequence of CCR5-peptide, before incubation with breast milk purified anti-CCR5 Abs induced a total inhibition of binding to CCR5-peptide (Fig. 3B).
To investigate the anti-CCR5 peptide Abs response in depth, IgG and IgA were purified by means of anti-- and anti--chain antibodies. Our results showed that depletion of sIgA from IgA Abs fraction, by means of anti-secretory component Abs insolubilized on Sepharose beads, induced a complete abolition of binding to CCR5 peptide indicating that 100% of purified IgA directed against CCR5 peptide are sIgA. No significant difference in reactivity against CCR5 peptide was observed between IgG and sIgA purified from the same pool of breast milk for both seropositive and seronegative women. (Fig. 3, C and D).
We further compared the relative avidity of affinity-purified anti-CCR5 Abs using a KSCN dissociation assay. The molarities of KSCN required to dissociate 50% of total anti-CCR5 Abs from the CCR5 peptide were 1.2 M and 0.5 M for Abs purified from HIV-seropositive and HIV-seronegative breast milk pools, respectively (Fig. 4A). No difference in the relative avidity was observed between IgG and sIgA directed to CCR5 peptide purified from the same pool of breast milk of seropositive and seronegative women (Fig. 4B). Thus, the molarities of KSCN required to dissociate 50% of sIgA and IgG anti-CCR5 Abs from the CCR5 peptide were 0.58 and 0.67 M for Abs purified from HIV-seronegative and 0.95 and 1.45 M for Abs purified from HIV-seropositive women, respectively.
Affinity-purified anti-CCR5 Abs from breast milk recognize the native CCR5 molecule
As shown in Fig. 5, affinity-purified anti-CCR5 Abs from HIV-seronegative and -seropositive breast milk pools bound in a dose-dependent fashion to the native CCR5 molecule expressed by CHO cells that had been transfected with the CCR5 gene. A plateau of binding was reached at a concentration of 100 μg/ml Abs that stained 40 and 70% of CHO-CCR5+ cells for Abs from seronegative and seropositive pools, respectively (Fig. 5A). The anti-CCR5-depleted fraction of breast milk stained <5% of CCR5+ CHO cells. Anti-CCR5 peptide Abs tested at 100 μg/ml recognized 7–9% of CXCR4+CCR5– CHO cells (Fig. 5B). Mouse mAbs to CCR5 (2D7) and CXCR4 (12G5) used at 10 μg/ml stained 98 and 78% of CCR5+ and CXCR4+ cells, respectively (Fig. 5C). In addition, the incubation of affinity-purified anti-CCR5 Abs with saturating amounts of CCR5 peptide before using the Abs in a binding assay, resulted in staining of <6% of CCR5+ CHO cells (Fig. 5D). In our culture conditions, 45% of macrophages and 25% of dendritic cells expressed CCR5 coreceptor. Used at 50 and 100 μg/ml, the affinity-purified anti-CCR5 Abs from seronegative women stained 9 and 25% of MDM and 7 and 18% of iMDDC, respectively. Similarly, affinity-purified anti-CCR5 Abs from seropositive patients used at 50 and 100 μg/ml, stained 14 and 22% of MDM and 13 and 15% of iMDDC, respectively (Fig. 5E). No binding was observed when anti-CCR5-depleted Abs were used.
Natural anti-CCR5 Abs from breast milk inhibit CCR5-dependent infection of macrophages and dendritic cells
Affinity-purified anti-CCR5 Abs from the breast milk pool of HIV-seronegative women dose-dependently inhibited infection of both MDM and iMDDC with HIVBaL and HIVJR-CSF. Thus, at a concentration of 1000 μg/ml, the anti-CCR5 Abs inhibited up to 75% infection with HIVBaL and HIVJR-CSF of macrophages (Fig. 6A) and dendritic cells (Fig. 6B). At a concentration of 250 μg/ml anti-CCR5 peptide Abs from breast milk of HIV-seropositive women were more efficient in inhibition of HIV infection than Abs from breast milk of HIV-seronegative women. RANTES, used as a positive control, inhibited at 85 and 90% infection of macrophages and dendritic cells, respectively. No significant inhibition was observed when using the anti-CCR5-depleted fraction of breast milk.
Discussion
Here we report that natural Abs directed against CCR5, the coreceptor for R5-tropic HIV-1, are present in human breast milk and are capable of inhibiting infection of human macrophages and dendritic cells with primary R5-tropic HIV in vitro.
The relative content in anti-CCR5 Abs was found to differ between samples of breast milk from individual women. Anti-CCR5 activity was found by means of ELISA in 66% and 83% of breast milk samples of HIV-seropositive and -seronegative women, respectively. We purified the Abs to CCR5 from pooled breast milk of both HIV-seropositive and -seronegative women by means of affinity chromatography on a synthetic peptide corresponding to a 20-amino acid region (II.E/C-CCR5) derived from the second extracellular loop of CCR5. This loop together with the N-terminal sequence (aa 1–20) and the first extracellular loop (aa 89–102) is involved in the binding of gp120 and of natural CCR5 ligands to the CCR5 molecule (6, 11, 12, 13). We focused on the second extracellular loop of CCR5 that was shown to be immunogenic and induce a B and T cell response. Affinity-purified Abs from breast milk bound specifically to CCR5 peptide as demonstrated by the lack of binding to an irrelevant peptide and by the abrogation of the binding in the presence of 2D7 mAb directed against the CCR5 peptide.
Affinity-purified Abs from breast milk bound to the immobilized CCR5 peptide in a dose-dependent manner, with total Abs from seropositive women exhibiting a 10-fold higher reactivity than those from HIV-seronegative women. However, no difference was observed when comparing the binding of IgG and IgA purified from the same pool of breast milk. Anti-CCR5 activity was found in IgA, IgG, and to a small degree in IgM. Depletion of sIgA by means of anti-secretory component Abs induced a complete abolition of binding of purified IgA to CCR5 peptide indicating that 100% of IgA directed against CCR5 peptide are sIgA.
By determining a dissociation index of immune complexes with KSCN, anti-CCR5 Abs purified from breast milk of HIV-seropositive women seemed to have a 2.4-fold higher avidity for the CCR5 peptide than Abs from HIV-seronegative women. However, no difference was observed when we compared the relative avidity index for the CCR5 peptide of IgG and sIgA purified from the same pool of breast milk. The increased avidity against CCR5 of Abs from breast milk of infected women may reflect polyclonal B cell stimulation in the mammary mucosae and the expansion of more avid natural Ig-producing B cell clones.
Antibodies purified from breast milk of HIV-seropositive women exhibited a 2-fold higher ability to bind to the native CCR5 molecule expressed by CHO cells than Abs purified from HIV-seronegative breast milk. The binding of affinity-purified anti-CCR5 Abs to CCR5+ CHO cells was abrogated by presaturation of the Abs with CCR5 peptide. Compared with CHO CCR5+ cells, the binding of affinity-purified anti-CCR5 peptide Abs to primary cultured macrophages and immature dendritic cells was less efficient. Using macrophages and immature dendritic cells, a slight difference in binding to CCR5 was observed between anti-CCR5 peptide Abs purified from breast milk of HIV-seropositive and -seronegative women. Although the avidity differed, the amount of anti-CCR5 Abs did not significantly differ between breast milk of HIV-seropositive and -seronegative women.
HIV infection is predominantly transmitted through mucosae (14). The CCR5 chemokine receptor functions as the major coreceptor for the R5-tropic strains that are primarily transmitted through the mucosal route (15). The role of CCR5 in HIV-1 transmission has been evidenced by observations that individuals homozygous for a defective CCR5 allele (CCR5 32) remained uninfected despite repeated exposure to HIV (4), and that protection against HIV infection was associated with the presence of natural anti-CCR5 Abs in serum in some exposed uninfected individuals in sero-discordant couples (16). Anti-CCR5 Abs were shown to down-modulate surface expression of CCR5 and to neutralize the infectivity of HIV R5-tropic strains, providing a basis for the acquisition of resistance to infection (16). After crossing the epithelial barrier, HIV spreads rapidly, through contact between dendritic cells and CD4 T lymphocytes (17). Hence, the blockade of CCR5 with mucosal anti-CCR5 Abs could result in inhibition of spreading of R5-tropic viruses.
Transmission of HIV-1 to the infant through breastfeeding is a major cause of new pediatric HIV-1 infections worldwide. Although extended breastfeeding accounts for approximately one- to two-thirds of infant HIV infections, most breastfed infants remain uninfected, despite prolonged and repeated exposure to HIV-1. The risk of transmission of HIV-1 through breastfeeding has been reported to be between 5 and 23% (18), depending on a number of factors, including viral load in breast milk and the intensity and quality of the specific mucosal anti-HIV immune response (19). Both free and cell-associated HIV-1 are detected in early and mature breast milk of HIV-infected women (3). Breast milk of HIV-seropositive women also contains high levels of sIgA and IgG to env-encoded HIV surface glycoproteins (20). The Ab response to HIV in breast milk appears to be compartmentalized, when analyzed compared with that in serum (21). Anti-HIV Abs in breast milk were not found to be a relevant factor in protection against transmission of HIV through breastfeeding (22). We thus examined the effect of anti-CCR5 Abs purified from breast milk on infection of macrophages and dendritic cells that express CCR5 (23) and are considered as primary targets for HIV following trans-epithelial passage of the virus (24). Here we demonstrate that natural anti-CCR5 Abs from breast milk of HIV-seropositive and -seronegative women, inhibit infection of macrophages and dendritic cells with the R5-tropic strains HIVBaL and HIVJR-CSF in a dose-dependent manner. At similar concentrations, the anti-CCR5-depleted fraction of breast milk inhibited <10% of infection of the cells. Such "residual" inhibitory effect could be due to the presence of anti-CCR5 Abs in the effluents of affinity columns that recognize extramembrane CCR5 domains outside the second extracellular loop. Concentration of anti-CCR5 peptide Abs needed to induce a significant inhibition of HIV infection was always higher than that inducing a maximum binding on CHO-CCR5 cells, macrophages, and immature dendritic cells. This discrepancy may be the result of either the amount of the virus used or of the higher affinity of HIV (gp160) to CCR5 receptor than that of purified natural anti-CCR5 peptide antibodies. We found a difference in the inhibitory capacity toward infection between Abs from breast milk of HIV-seronegative and -seropositive women. Thus, at the same amount, Abs purified from seropositive women exhibited a higher capability of inhibiting HIV infection.
Our observations suggest that anti-CCR5 Abs in breast milk may control postnatal transmissibility of HIV through breastfeeding. The relationship between the presence/amounts and avidity of anti-CCR5 Abs and postnatal transmission of HIV-1 remains to be investigated in clinical cohorts. Abs to both viral-neutralizing glycoprotein epitopes (25, 26) and to chemokine coreceptors have been suggested to play a role in protection against HIV infection (6, 27). The presence of anti-CCR5 Abs has previously been documented in sera of HIV-seropositive (28), HIV-exposed uninfected (16), healthy (8), and homozygous CCR5 32 individuals (29), who had been repeatedly exposed to CCR5-expressing cells through sexual activity. Furthermore, Abs to CCR5 molecule induced by vaccination in macaques were shown to play an important role in inhibition of infection with SIV (30). Anti-CCR5 Abs and CCR5-binding chemokine such as RANTES that are secreted in breast milk with a considerable variability among women (31, 32) may compete for the binding to CCR5 and inhibit infection of peripheral blood monocytic cells with R5 but not with X4-tropic primary isolates of HIV-1. Interestingly, Abs to the second extracellular loop of CCR5 block CCR5-mediated HIV transmission without affecting CCR5 signaling by chemokines (33), which could represent an advantage of passive therapy with such Abs warranting further research.
Acknowledgments
We thank Pr. F Barre-Sinoussi and Dr. G Gresenguet for HIV-1 strains and for the gift of seropositive milk, respectively. We thank Dr Ali Amara for CHO-CCR5 cell lines, Dr. A Beretta for CCR5 peptide gift. We thank J Bayry and S Delignat for participation in the anti-CCR5 antibodies purifications.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by Agence Nationale de Recheche sur le Sida (France).
2 H.B. and V.L. are considered co-first authors.
3 Address correspondence and reprint requests to Dr. Hicham Bouhlal, INSERM U743, Equipe d’Immunité et Biothérapies Muqueuses et Université René Descartes Paris V; Institut Biomédical des Cordeliers, 15, rue de l’Ecole de Médecine, 75006 Paris. E-mail address: hicham.bouhlal{at}u430.bhdc.jussieu.fr
4 Abbreviations used in this paper: sIgA, secretory IgA; DC-SIGN, dendritic cell-specific ICAM-3-grabbing nonintegrin; CHO, Chinese hamster ovary; KSCN, potassium thiocyanate.
Received for publication May 12, 2004. Accepted for publication March 29, 2005.
References
Zhu, T., H. Mo, N. Wang, D. S. Nam, Y. Cao, R. A. Koup, D. D. Ho. 1993. Genotypic and phenotypic characterization of HIV-1 patients with primary infection. Science 261: 1179-1181.
Williams, S. B., T. P. Flanigan, A. W. Artenstein, T. C. VanCott, D. Smith, K. Mayer, R. A. Koup. 1999. CCR5 genotype and human immunodeficiency virus (HIV)-specific mucosal antibody in seronegative women at high risk for HIV infection. J. Infect. Dis. 179: 1310-1312.
Becquart, P., N. Chomont, P. Roques, A. Ayouba, M. D. Kazatchkine, L. Belec, H. Hocini. 2002. Compartmentalization of HIV-1 between breast milk and blood of HIV-infected mothers. Virology 300: 109-117.
Liu, R., W. A. Paxton, S. Choe, D. Ceradini, S. R. Martin, R. Horuk, M. E. MacDonald, H. Stuhlmann, R. A. Koup, N. R. Landau. 1996. Homozygous defect in HIV-1 coreceptor accounts for resistance of some multiply exposed individuals to HIV-1 infection. Cell 86: 367-377.
Wu, L., W. A. Paxton, N. Kassam, N. Ruffing, J. B. Rottman, N. Sullivan, H. Choe, J. Sodroski, W. Newman, R. A. Koup, C. R. Mackay. 1997. CCR5 levels and expression pattern correlate with infectability by macrophage-tropic HIV-1, in vitro. J. Exp. Med. 185: 1681-1691.(Hicham Bouhlal2,3,*, Vane)
In the present study, we demonstrate that breast milk of 66% and 83% of HIV-seronegative and seropositive women, respectively, contains natural Abs of the secretory IgA and IgG isotypes directed against the CCR5 coreceptor for R5-tropic strains of HIV-1. Abs to CCR5 were affinity purified on a matrix to which a synthetic peptide corresponding to the second extracellular loop of CCR5 had been coupled. The purified Abs bound to the CCR5 peptide in a dose-dependent fashion and to both native CCR5 expressed by Chinese hamster ovary cells transfected with CCR5 gene, macrophages, and immature dendritic cells. Although the avidity differed, the amount of anti-CCR5 Abs did not significantly differ between breast milk of HIV-seropositive and -seronegative women. Purified anti-CCR5 Abs inhibited up to 75% infection of macrophages and dendritic cells with HIVBaL and HIVJR-CSF. Our observations provide evidence for a role of natural Abs to CCR5 in breast milk in controlling transmissibility of HIV through breastfeeding.
Introduction
The CCR5 chemokine receptor is the major coreceptor that is associated with mucosal transmission of R5-tropic HIV-1 during sexual intercourse and postnatal transmission through breastfeeding (1, 2, 3). The importance of CCR5 for HIV-1 transmission is evidenced by the observation that individuals homozygous for a defective CCR5 allele remain uninfected despite repeated exposure to HIV (4). The CCR5 molecule is thus a target for novel therapeutic strategies aimed at blocking HIV-1 entry into cells (5, 6, 7). We have previously demonstrated that natural Abs from therapeutic preparations of IgG (i.v. Ig) contain natural Abs directed against CCR5 that inhibit infection of human macrophages and CD4+ T lymphocytes with laboratory and primary R5-tropic isolates of HIV-1 in vitro (8). Breast milk contains high amounts of immunoglobulins, predominantly of the secretory IgA (sIgA) 4 and to a lesser degree of the IgG and sIgM isotypes that are produced in the absence of deliberate immunization and independently of exposure to Ags. Most natural Abs in the breast milk of healthy women are self-reactive Abs (9, 10).
In the present study, we demonstrate that breast milk contains Abs directed against CCR5 molecule. Following affinity purification, these Abs were shown to bind to the native CCR5 molecule and to inhibit infection of macrophages and dendritic cells by R5-tropic HIV-1.
Materials and Methods
Inhibition of HIV-1 infection by Abs to CCR5 peptide
Macrophages and dendritic cells (5 x 105 cells) in 48-well plates were infected with 5 ng/ml p24 HIVBal, HIVJR-CSF, and HIVNDK for 3 h at 37°C. The cells were washed and then cultured for 6 days. Supernatants were collected every 2 days, and HIV p24 levels were measured by ELISA. In some experiments, cells were preincubated with purified anti-CCR5 Abs (100, 250, 500, and 1000 μg/ml) or with corresponding amounts of anti-CCR5-depleted immunoglobulins used as negative control. As positive control of inhibition of HIV infection, purified anti-CCR5 from i.v. Ig, or the chemokines RANTES and SDF-1 (500 ng/ml) were added to cells before infection with HIV.
Results
Human breast milk contains natural Abs directed against CCR5
We first investigated the reactivity with the CCR5 peptide of immunoglobulins in each of the 18 breast milk samples from HIV-seronegative women and 12 samples from HIV-seropositive women by ELISA. Fifteen of 18 (83%) and 8 of 12 (66%) samples contained over 1 μg/ml anti-CCR5 Ig (Fig. 1). Based on the total Ig content of breast milk samples from HIV-seronegative and -seropositive women, the mean specific anti-CCR5 activity was calculated as being 0.24 and 0.30 in breast milk samples of HIV-seronegative and HIV-seropositive women, respectively.
As shown in Fig. 3A, Abs purified from breast milk of seropositive women exhibited an 10-fold stronger anti-CCR5 reactivity than those purified from pooled milk of seronegative women. Anti-CCR5 IgG that had been affinity-purified from i.v. Ig, used as a positive control, also bound with lesser intensity to the CCR5 peptide than affinity-purified Abs from breast milk of HIV-seropositive women. No binding to CCR5 peptide was observed when anti-CCR5 completely depleted the fraction of breast milk was tested. Affinity-purified anti-CCR5 Abs from breast milk of healthy and HIV-seropositive women and i.v. Ig reacted specifically with the CCR5 peptide but not with irrelevant peptide corresponding to a sequence of DC-SIGN molecule. Furthermore, addition of CCR5 peptide to mAb 2D7, that recognize the sequence of CCR5-peptide, before incubation with breast milk purified anti-CCR5 Abs induced a total inhibition of binding to CCR5-peptide (Fig. 3B).
To investigate the anti-CCR5 peptide Abs response in depth, IgG and IgA were purified by means of anti-- and anti--chain antibodies. Our results showed that depletion of sIgA from IgA Abs fraction, by means of anti-secretory component Abs insolubilized on Sepharose beads, induced a complete abolition of binding to CCR5 peptide indicating that 100% of purified IgA directed against CCR5 peptide are sIgA. No significant difference in reactivity against CCR5 peptide was observed between IgG and sIgA purified from the same pool of breast milk for both seropositive and seronegative women. (Fig. 3, C and D).
We further compared the relative avidity of affinity-purified anti-CCR5 Abs using a KSCN dissociation assay. The molarities of KSCN required to dissociate 50% of total anti-CCR5 Abs from the CCR5 peptide were 1.2 M and 0.5 M for Abs purified from HIV-seropositive and HIV-seronegative breast milk pools, respectively (Fig. 4A). No difference in the relative avidity was observed between IgG and sIgA directed to CCR5 peptide purified from the same pool of breast milk of seropositive and seronegative women (Fig. 4B). Thus, the molarities of KSCN required to dissociate 50% of sIgA and IgG anti-CCR5 Abs from the CCR5 peptide were 0.58 and 0.67 M for Abs purified from HIV-seronegative and 0.95 and 1.45 M for Abs purified from HIV-seropositive women, respectively.
Affinity-purified anti-CCR5 Abs from breast milk recognize the native CCR5 molecule
As shown in Fig. 5, affinity-purified anti-CCR5 Abs from HIV-seronegative and -seropositive breast milk pools bound in a dose-dependent fashion to the native CCR5 molecule expressed by CHO cells that had been transfected with the CCR5 gene. A plateau of binding was reached at a concentration of 100 μg/ml Abs that stained 40 and 70% of CHO-CCR5+ cells for Abs from seronegative and seropositive pools, respectively (Fig. 5A). The anti-CCR5-depleted fraction of breast milk stained <5% of CCR5+ CHO cells. Anti-CCR5 peptide Abs tested at 100 μg/ml recognized 7–9% of CXCR4+CCR5– CHO cells (Fig. 5B). Mouse mAbs to CCR5 (2D7) and CXCR4 (12G5) used at 10 μg/ml stained 98 and 78% of CCR5+ and CXCR4+ cells, respectively (Fig. 5C). In addition, the incubation of affinity-purified anti-CCR5 Abs with saturating amounts of CCR5 peptide before using the Abs in a binding assay, resulted in staining of <6% of CCR5+ CHO cells (Fig. 5D). In our culture conditions, 45% of macrophages and 25% of dendritic cells expressed CCR5 coreceptor. Used at 50 and 100 μg/ml, the affinity-purified anti-CCR5 Abs from seronegative women stained 9 and 25% of MDM and 7 and 18% of iMDDC, respectively. Similarly, affinity-purified anti-CCR5 Abs from seropositive patients used at 50 and 100 μg/ml, stained 14 and 22% of MDM and 13 and 15% of iMDDC, respectively (Fig. 5E). No binding was observed when anti-CCR5-depleted Abs were used.
Natural anti-CCR5 Abs from breast milk inhibit CCR5-dependent infection of macrophages and dendritic cells
Affinity-purified anti-CCR5 Abs from the breast milk pool of HIV-seronegative women dose-dependently inhibited infection of both MDM and iMDDC with HIVBaL and HIVJR-CSF. Thus, at a concentration of 1000 μg/ml, the anti-CCR5 Abs inhibited up to 75% infection with HIVBaL and HIVJR-CSF of macrophages (Fig. 6A) and dendritic cells (Fig. 6B). At a concentration of 250 μg/ml anti-CCR5 peptide Abs from breast milk of HIV-seropositive women were more efficient in inhibition of HIV infection than Abs from breast milk of HIV-seronegative women. RANTES, used as a positive control, inhibited at 85 and 90% infection of macrophages and dendritic cells, respectively. No significant inhibition was observed when using the anti-CCR5-depleted fraction of breast milk.
Discussion
Here we report that natural Abs directed against CCR5, the coreceptor for R5-tropic HIV-1, are present in human breast milk and are capable of inhibiting infection of human macrophages and dendritic cells with primary R5-tropic HIV in vitro.
The relative content in anti-CCR5 Abs was found to differ between samples of breast milk from individual women. Anti-CCR5 activity was found by means of ELISA in 66% and 83% of breast milk samples of HIV-seropositive and -seronegative women, respectively. We purified the Abs to CCR5 from pooled breast milk of both HIV-seropositive and -seronegative women by means of affinity chromatography on a synthetic peptide corresponding to a 20-amino acid region (II.E/C-CCR5) derived from the second extracellular loop of CCR5. This loop together with the N-terminal sequence (aa 1–20) and the first extracellular loop (aa 89–102) is involved in the binding of gp120 and of natural CCR5 ligands to the CCR5 molecule (6, 11, 12, 13). We focused on the second extracellular loop of CCR5 that was shown to be immunogenic and induce a B and T cell response. Affinity-purified Abs from breast milk bound specifically to CCR5 peptide as demonstrated by the lack of binding to an irrelevant peptide and by the abrogation of the binding in the presence of 2D7 mAb directed against the CCR5 peptide.
Affinity-purified Abs from breast milk bound to the immobilized CCR5 peptide in a dose-dependent manner, with total Abs from seropositive women exhibiting a 10-fold higher reactivity than those from HIV-seronegative women. However, no difference was observed when comparing the binding of IgG and IgA purified from the same pool of breast milk. Anti-CCR5 activity was found in IgA, IgG, and to a small degree in IgM. Depletion of sIgA by means of anti-secretory component Abs induced a complete abolition of binding of purified IgA to CCR5 peptide indicating that 100% of IgA directed against CCR5 peptide are sIgA.
By determining a dissociation index of immune complexes with KSCN, anti-CCR5 Abs purified from breast milk of HIV-seropositive women seemed to have a 2.4-fold higher avidity for the CCR5 peptide than Abs from HIV-seronegative women. However, no difference was observed when we compared the relative avidity index for the CCR5 peptide of IgG and sIgA purified from the same pool of breast milk. The increased avidity against CCR5 of Abs from breast milk of infected women may reflect polyclonal B cell stimulation in the mammary mucosae and the expansion of more avid natural Ig-producing B cell clones.
Antibodies purified from breast milk of HIV-seropositive women exhibited a 2-fold higher ability to bind to the native CCR5 molecule expressed by CHO cells than Abs purified from HIV-seronegative breast milk. The binding of affinity-purified anti-CCR5 Abs to CCR5+ CHO cells was abrogated by presaturation of the Abs with CCR5 peptide. Compared with CHO CCR5+ cells, the binding of affinity-purified anti-CCR5 peptide Abs to primary cultured macrophages and immature dendritic cells was less efficient. Using macrophages and immature dendritic cells, a slight difference in binding to CCR5 was observed between anti-CCR5 peptide Abs purified from breast milk of HIV-seropositive and -seronegative women. Although the avidity differed, the amount of anti-CCR5 Abs did not significantly differ between breast milk of HIV-seropositive and -seronegative women.
HIV infection is predominantly transmitted through mucosae (14). The CCR5 chemokine receptor functions as the major coreceptor for the R5-tropic strains that are primarily transmitted through the mucosal route (15). The role of CCR5 in HIV-1 transmission has been evidenced by observations that individuals homozygous for a defective CCR5 allele (CCR5 32) remained uninfected despite repeated exposure to HIV (4), and that protection against HIV infection was associated with the presence of natural anti-CCR5 Abs in serum in some exposed uninfected individuals in sero-discordant couples (16). Anti-CCR5 Abs were shown to down-modulate surface expression of CCR5 and to neutralize the infectivity of HIV R5-tropic strains, providing a basis for the acquisition of resistance to infection (16). After crossing the epithelial barrier, HIV spreads rapidly, through contact between dendritic cells and CD4 T lymphocytes (17). Hence, the blockade of CCR5 with mucosal anti-CCR5 Abs could result in inhibition of spreading of R5-tropic viruses.
Transmission of HIV-1 to the infant through breastfeeding is a major cause of new pediatric HIV-1 infections worldwide. Although extended breastfeeding accounts for approximately one- to two-thirds of infant HIV infections, most breastfed infants remain uninfected, despite prolonged and repeated exposure to HIV-1. The risk of transmission of HIV-1 through breastfeeding has been reported to be between 5 and 23% (18), depending on a number of factors, including viral load in breast milk and the intensity and quality of the specific mucosal anti-HIV immune response (19). Both free and cell-associated HIV-1 are detected in early and mature breast milk of HIV-infected women (3). Breast milk of HIV-seropositive women also contains high levels of sIgA and IgG to env-encoded HIV surface glycoproteins (20). The Ab response to HIV in breast milk appears to be compartmentalized, when analyzed compared with that in serum (21). Anti-HIV Abs in breast milk were not found to be a relevant factor in protection against transmission of HIV through breastfeeding (22). We thus examined the effect of anti-CCR5 Abs purified from breast milk on infection of macrophages and dendritic cells that express CCR5 (23) and are considered as primary targets for HIV following trans-epithelial passage of the virus (24). Here we demonstrate that natural anti-CCR5 Abs from breast milk of HIV-seropositive and -seronegative women, inhibit infection of macrophages and dendritic cells with the R5-tropic strains HIVBaL and HIVJR-CSF in a dose-dependent manner. At similar concentrations, the anti-CCR5-depleted fraction of breast milk inhibited <10% of infection of the cells. Such "residual" inhibitory effect could be due to the presence of anti-CCR5 Abs in the effluents of affinity columns that recognize extramembrane CCR5 domains outside the second extracellular loop. Concentration of anti-CCR5 peptide Abs needed to induce a significant inhibition of HIV infection was always higher than that inducing a maximum binding on CHO-CCR5 cells, macrophages, and immature dendritic cells. This discrepancy may be the result of either the amount of the virus used or of the higher affinity of HIV (gp160) to CCR5 receptor than that of purified natural anti-CCR5 peptide antibodies. We found a difference in the inhibitory capacity toward infection between Abs from breast milk of HIV-seronegative and -seropositive women. Thus, at the same amount, Abs purified from seropositive women exhibited a higher capability of inhibiting HIV infection.
Our observations suggest that anti-CCR5 Abs in breast milk may control postnatal transmissibility of HIV through breastfeeding. The relationship between the presence/amounts and avidity of anti-CCR5 Abs and postnatal transmission of HIV-1 remains to be investigated in clinical cohorts. Abs to both viral-neutralizing glycoprotein epitopes (25, 26) and to chemokine coreceptors have been suggested to play a role in protection against HIV infection (6, 27). The presence of anti-CCR5 Abs has previously been documented in sera of HIV-seropositive (28), HIV-exposed uninfected (16), healthy (8), and homozygous CCR5 32 individuals (29), who had been repeatedly exposed to CCR5-expressing cells through sexual activity. Furthermore, Abs to CCR5 molecule induced by vaccination in macaques were shown to play an important role in inhibition of infection with SIV (30). Anti-CCR5 Abs and CCR5-binding chemokine such as RANTES that are secreted in breast milk with a considerable variability among women (31, 32) may compete for the binding to CCR5 and inhibit infection of peripheral blood monocytic cells with R5 but not with X4-tropic primary isolates of HIV-1. Interestingly, Abs to the second extracellular loop of CCR5 block CCR5-mediated HIV transmission without affecting CCR5 signaling by chemokines (33), which could represent an advantage of passive therapy with such Abs warranting further research.
Acknowledgments
We thank Pr. F Barre-Sinoussi and Dr. G Gresenguet for HIV-1 strains and for the gift of seropositive milk, respectively. We thank Dr Ali Amara for CHO-CCR5 cell lines, Dr. A Beretta for CCR5 peptide gift. We thank J Bayry and S Delignat for participation in the anti-CCR5 antibodies purifications.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by Agence Nationale de Recheche sur le Sida (France).
2 H.B. and V.L. are considered co-first authors.
3 Address correspondence and reprint requests to Dr. Hicham Bouhlal, INSERM U743, Equipe d’Immunité et Biothérapies Muqueuses et Université René Descartes Paris V; Institut Biomédical des Cordeliers, 15, rue de l’Ecole de Médecine, 75006 Paris. E-mail address: hicham.bouhlal{at}u430.bhdc.jussieu.fr
4 Abbreviations used in this paper: sIgA, secretory IgA; DC-SIGN, dendritic cell-specific ICAM-3-grabbing nonintegrin; CHO, Chinese hamster ovary; KSCN, potassium thiocyanate.
Received for publication May 12, 2004. Accepted for publication March 29, 2005.
References
Zhu, T., H. Mo, N. Wang, D. S. Nam, Y. Cao, R. A. Koup, D. D. Ho. 1993. Genotypic and phenotypic characterization of HIV-1 patients with primary infection. Science 261: 1179-1181.
Williams, S. B., T. P. Flanigan, A. W. Artenstein, T. C. VanCott, D. Smith, K. Mayer, R. A. Koup. 1999. CCR5 genotype and human immunodeficiency virus (HIV)-specific mucosal antibody in seronegative women at high risk for HIV infection. J. Infect. Dis. 179: 1310-1312.
Becquart, P., N. Chomont, P. Roques, A. Ayouba, M. D. Kazatchkine, L. Belec, H. Hocini. 2002. Compartmentalization of HIV-1 between breast milk and blood of HIV-infected mothers. Virology 300: 109-117.
Liu, R., W. A. Paxton, S. Choe, D. Ceradini, S. R. Martin, R. Horuk, M. E. MacDonald, H. Stuhlmann, R. A. Koup, N. R. Landau. 1996. Homozygous defect in HIV-1 coreceptor accounts for resistance of some multiply exposed individuals to HIV-1 infection. Cell 86: 367-377.
Wu, L., W. A. Paxton, N. Kassam, N. Ruffing, J. B. Rottman, N. Sullivan, H. Choe, J. Sodroski, W. Newman, R. A. Koup, C. R. Mackay. 1997. CCR5 levels and expression pattern correlate with infectability by macrophage-tropic HIV-1, in vitro. J. Exp. Med. 185: 1681-1691.(Hicham Bouhlal2,3,*, Vane)