|Home | About | Journals | Submit | Contact Us | Français|
Alcian Blue (AB), a phthalocyanine derivative, is able to prevent infection by a wide spectrum of human immunodeficiency virus type 1 (HIV-1), HIV-2, and simian immunodeficiency virus strains in various cell types [T cells, (co)receptor-transfected cells, and peripheral blood mononuclear cells]. With the exception of herpes simplex virus, AB is inactive against a broad variety of other (DNA and RNA) viruses. Time-of-addition studies show that AB prevents HIV-1 infection at the virus entry stage, exactly at the same time as carbohydrate-binding agents do. AB also efficiently prevents fusion between persistently HIV-1-infected HUT-78 cells and uninfected (CD4+) lymphocytes, DC-SIGN-directed HIV-1 capture, and subsequent transmission to uninfected (CD4+) T lymphocytes. Prolonged passaging of HIV-1 at dose-escalating concentrations of AB resulted in the selection of mutant virus strains in which several N-glycans of the HIV-1 gp120 envelope were deleted and in which positively charged amino acid mutations in both gp120 and gp41 appeared. A mutant virus strain in which four N-glycans were deleted showed a 10-fold decrease in sensitivity to the inhibitory effect of AB. These data suggest that AB is likely endowed with carbohydrate-binding properties and can be considered an important lead compound in the development of novel synthetic nonpeptidic antiviral drugs targeting the glycans of the envelope of HIV.
Targeting the entry process of human immunodeficiency virus (HIV), including drugs that bind to the receptor CD4, to a coreceptor, CCR5 or CXCR4, or to gp160 envelope is a valuable approach to prevent or suppress HIV infections. The first clinically used entry inhibitor, enfuvirtide (T20; Fuzeon) binds to the transmembrane gp41 of the envelope of HIV (21), thus preventing the required conformational changes of the envelope to successfully complete viral entry. Most recently, the CCR5 antagonist maraviroc was approved by the FDA for the treatment of HIV-infected individuals (16). Both enfuvirtide and maraviroc prove that HIV entry can be efficiently targeted by drugs that act at different stages in the entry process.
The envelope of HIV consists of two subunits: the surface gp120 and the transmembrane gp41. Both units are highly glycosylated (23, 24), which is essential for the virus to escape immune surveillance (28). A broad variety of carbohydrate-binding agents (CBAs), such as the plant lectins Hippeastrum hybrid agglutinin (HHA), Galanthus nivalis agglutinin (GNA), and Urtica dioica agglutinin (UDA) or the prokaryotic cyanovirin-N (CV-N) and actinohivin, bind to the glycans that are present on the envelope of HIV and inhibit the viral entry process (4, 5). The majority of the natural CBAs are proteins (i.e., lectins), which may have some major drawbacks: (i) it is technically not easy and it is relatively costly to produce and purify these proteins on a large scale, (ii) they have poor, if any, oral bioavailability, and (iii) they can trigger an immune response that compromises their eventual antiviral efficacy (1). Therefore, (semi)synthetic low-molecular-weight compounds that are also able to bind glycans and prevent virus infection would be a valuable alternative.
Recently, pradimicin A (PRM-A), an antifungal nonpeptidic antibiotic (26), was described to possess lectin-like properties and bind to the glycans of HIV gp120 (32) and proved able to efficiently prevent HIV infection (31). HIV selected under escalating PRM-A concentrations can escape this drug pressure by deleting multiple N-glycosylation sites in gp120 (10). This was earlier also observed to occur in HIV-1 strains selected under pressure of peptidic CBAs, such as HHA, GNA, UDA, and cyanovirin-N (5, 6). Thus, the CBAs not only prevent virus infection by binding to glycans on the envelope of HIV, but also they can force the virus to delete its envelope N-glycans in order to escape drug pressure, resulting in the exposure of previously hidden immunogenic epitopes. This phenomenon is interesting given the fact that the glycans on the HIV gp120 envelope play a very important role in shielding the virus from the immune system and in limiting the neutralizing antibody response to HIV (39).
Here, we report on a synthetic compound, the phthalocyanine Alcian Blue (AB), that is endowed with anti-HIV activity due to its lectin-like properties. It prevents entry of HIV into its target cells and selects for mutant virus strains that have several deletions in N-glycosylation sites in gp120.
Alcian Blue 8GX (molecular weight, 1,298.88; lot no. 074K1779) (Fig. (Fig.1)1) and dextran sulfate 5000 (DS5000) were obtained from Sigma (St. Louis, MO). PRM-A (molecular weight, 838) was isolated and purified from the fermentation broth of the Actinomyces species Actinomadura hibisca TP-A0016 (12, 26). The mannose-specific plant lectins GNA, HHA, and UDA were derived and purified from these plants as previously described (33, 34). Polyvinyl alcohol sulfate (PVAS) was kindly provided by S. Görög (Budapest, Hungary).
Alcian Blue was subjected to high-performance liquid chromatography (HPLC) analysis using the following gradient: 2 min of 2% acetonitrile (ACN) in H2O plus 0.1% tetrafluoroacetate (buffer B); 10-min linear gradient to 20% ACN in buffer B; 5-min linear gradient to 35% ACN in buffer B; 10-min linear gradient to 50% ACN in buffer B; 5-min isocratic elution; 10-min linear gradient 2% ACN in buffer B; 5-min equilibration.
Selected fractions were analyzed by HPLC-mass spectrometry (MS) using an HPLC Waters 12695 apparatus connected to a Waters Micromass ZQ spectrometer. The column employed was a Sunfire C18 (4.6 by 50 mm; 3.5 μm). The elution was performed with a linear gradient from 10% CH3CN (0.08% formic acid):90% H2O (0.1% formic acid) to 100% CH3CN (0.08% formic acid) in 5 min.
Matrix-assisted laser desorption/ionization-MS measurements on selected fractions were performed with a Voyager DE-PRO mass spectrometer (Applied Biosystems, Foster City, CA) equipped with a pulsed nitrogen laser (wavelength, 337 mm; 3-ns pulse width; 3-Hz frequency) and a delayed extraction ion source. Ions generated by laser desorption were introduced into a time-of-flight analyzer (1.3-m flight path) with an acceleration voltage of 20 kV, 72% grid voltage, 0.01% ion guide wire voltage, and a delay time of 100 ns in the reflector positive ion mode. Mass spectra were obtained over the m/z range of 700 to 1,700 u. External mass calibration was applied using the monoisotopic [M+H]+ values of des-Arg1-Bradykinin, angiotensin I, and Glu1-fibrinopeptide B of calibration mixture 1, Sequazyme peptide mass standards kit (Applied Biosystems). 2,5-Dihydroxybenzoic acid (>98%; Fluka) at 10 mg ml−1 in acetonitrile was used as the matrix. The sample was mixed with the matrix at a ratio of ~1:5 (vol/vol), and 1 μM of this solution was spotted onto a flat stainless steel sample plate and dried in air.
Human T-lymphocytic CEM, MT4, C8166, and HUT-78 cells were cultivated in RPMI 1640 medium (Invitrogen, Merelbeke, Belgium) supplemented with 10% heat-inactivated fetal calf serum (Lonza Verviers SPRL, Verviers, Belgium), 2 mM l-glutamine, and 20 μg/ml gentamicin (Invitrogen).
HIV-1(IIIB) was kindly provided by R. C. Gallo (at that time at the National Cancer Institute, NIH, Bethesda, MD) and HIV-2 (ROD) was provided by L. Montagnier (at that time at the Pasteur Institute, Paris, France). Simian immunodeficiency virus SIVmac251 was provided by H. Egberink (Utrecht, The Netherlands).
The methodologies of the anti-HIV and anti-SIV assays have been described previously (9). Briefly, CEM cells (4.5 × 105 cells/ml) were suspended in fresh culture medium and infected with HIV-1 at 100 times the 50% cell culture infective dose per ml of cell suspension in the presence of appropriate dilutions of the test compounds. The compound concentrations were present during the entire incubation period of the virus-infected cell cultures. After 4 to 5 days of incubation at 37°C, giant cell formation was recorded microscopically in the CEM cell cultures. The 50% effective concentration (EC50) corresponds to the compound concentration required to prevent syncytium formation by 50% in the virus-infected CEM cell cultures. MT4 cells (5 × 105 cells/ml) were suspended in fresh culture medium and infected with SIV at 100 times the 50% cell culture infective dose per ml of cell suspension in the presence of appropriate dilutions of the test compounds. After 4 days of incubation at 37°C, virus-induced destruction of the MT4 cells was recorded by the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] dye viability staining method (27). Identical assays were performed in the presence of 2.5 mg/ml mannan (Sigma). The cytostatic/cytotoxic activity of AB was determined as follows: CEM cell cultures (~3 × 105 cells/ml) were exposed to serial concentrations of AB (i.e., 100, 20, 4, and 0.8 μg/ml) for 3 to 4 days and incubated at 37°C in a humidified CO2-controlled incubator. After this incubation time, the number of cells was counted using an automated Coulter Counter (Analis, Harpenden Hertz, United Kingdom). The 50% inhibitory concentration represents the inhibitory concentration required to inhibit CEM cell proliferation by 50%.
Primary clinical isolates representing different HIV-1 clades (Table (Table1)1) were all kindly provided by L. Lathey from BBI Biotech Research Laboratories, Inc., Gaithersburg, MD, and their coreceptor use (R5 or X4) was determined (19). Antiviral testing of these isolates in peripheral blood mononuclear cells (PBMC) was performed as described previously (10). Briefly, PBMC from healthy donors were stimulated with phytohemagglutinin (PHA) at 2 μg/ml (Sigma, Bornem, Belgium) for 3 days at 37°C. The PHA-stimulated blasts were then seeded at 0.5 × 106 cells per well into a 48-well plate containing various concentrations of compound in cell culture medium (RPMI 1640) containing 10% fetal calf serum and interleukin-2 (25 U/ml; R&D Systems Europe, Abingdon, United Kingdom). The virus stocks were added at a final dose of 250 pg of p24/ml. Cell supernatant was collected at day 12, and HIV-1 core antigen in the culture supernatant was analyzed by using a p24 antigen (Ag) enzyme-linked immunosorbent assay kit (Perkin-Elmer, Boston, MA).
Persistently HIV-1-infected HUT-78 cells (designated HUT-78/HIV) were washed to remove free virus from the cell culture medium, and 5 × 104 cells (50 μl) were transferred to 96-well microtiter plates. Next, 5 × 104 Sup-T1 cells (50 μl) and an appropriate concentration of test compound (100 μl) were added to each well. The mixed cell cultures were cultured at 37°C in a CO2-controlled atmosphere. The first syncytia arose after about 6 h of cultivation. After 16 to 20 h, marked syncytium formation was observed, and the number of syncytia was examined and quantified under a microscope.
The time-of-drug-addition experiments were performed as follows (26): MT-4 cells were infected with HIV-1(IIIB) at a multiplicity of infection of 0.5. The test compounds were added at different time points after virus infection (from 0 to 25 h). Viral p24 Ag production was determined at 31 h postinfection by enzyme-linked immunosorbent assay (Perkin-Elmer, Brussels, Belgium). The reference compounds were added at 100 times the 50% inhibitory concentration obtained in the MT-4 cells/MTT assay.
Wild-type Raji/0 and Raji/DC-SIGN cells (17) were suspended in cell culture medium at 1 × 106 cells/500 μl. A 100-μl aliquot of HIV-1(IIIB) was added in the presence of 400 μl of test compound or medium (mock treatment). After 60 min of incubation at 37°C, the cells were carefully washed three times to remove unbound virus particles and resuspended in 1 ml of cell culture medium. The captured virus on the Raji/DC-SIGN cells was quantified by p24 determination. From this suspension, 200 μl was added to uninfected 2 × 105 C8166 cells (800 μl) in a 48-well plate. The cocultures were incubated at 37°C, giant cell formation was microscopically estimated the next day, p24 determinations in the supernatants were performed.
The procedure for the selection of AB-resistant HIV-1(IIIB) strains was comparable with that described for PRM-A against HIV-1 (10). Briefly, CEM cell cultures were infected with HIV-1(IIIB) and were seeded in 48-well plates in the presence of AB at a concentration equal to one- to twofold the respective EC50. Subcultivations occurred every 3 to 4 days by transferring 50 μl of the drug-exposed HIV-infected cells to 950 μl of uninfected, freshly prepared CEM cell cultures. For the generation of drug-resistant virus mutants, escalating (twofold higher) drug concentrations were administered when in the previous cell culture the virus afforded a full cytopathogenic effect.
Proviral DNA was extracted from cell pellets using the DNeasy tissue kits (Qiagen, Hilden, Germany). The genotypes of both the gp120 and gp41 genes were determined by this assay, as described previously (35).
Alcian Blue 8GX has been shown to be a tetrakis(tetramethylisothiouronium)copper phthalocyanine tetrachloride (molecular weight, 1,298.86; exact mass, 1295.28) consisting of mainly four geometrical isomers predominantly substituted at position 4 of each isoindole residue (Fig. (Fig.1)1) (30). It has been described that the percentage of the dye in commercial samples can vary among manufacturers (29). Moreover, other materials can also be present, including dextrin, inorganic salts, and boric acid, among others (25). Therefore, an HPLC analysis was performed to obtain different fractions from the commercial samples. The fractions were then tested in antiviral (HIV) assays and were subject to structural analysis. The analysis of UV peak fractions 42 and 44 of Alcian Blue (Sigma) by HPLC revealed major peaks at 3.06 and 3.15 min, respectively. In the electrospray-MS analysis of these samples, the base peak was observed at m/z 576. This peak can be considered characteristic of the presence of the phthalocyanine core structure and has also been observed in previous analysis of Alcian Blue (13). No molecular ion peak was detected, and only very-low-intensity peaks whose putative origins will be discussed later were observed at higher m/z values (1,026 and 1,068). Both fractions were further analyzed by MALDI, because this technique has been successfully employed for the characterization of the molecular mass of phthalocyanines (11). As show in Fig. Fig.22 for fraction 42, the MALDI mass spectrum corroborated the presence of the active ingredient of Alcian Blue, that is, a peak with 100% intensity at m/z 1153.39 (M-4HCL+2H)+ (calculated for C56H64CuN16S4 1151.37) with the expected isotopic pattern, including the copper contribution. The other two major peaks at 1069.29 and 1026.24 may correspond to two or three losses of the dimethylamino groups of the tetramethylisothiouronium side chains, since a certain degree of in-source fragmentation has been described for other phthalocyanines (11). Both fractions 42 and 44 were shown to be inhibitory against HIV-1(IIIB) in CEM cell cultures.
The antiviral activity of AB (taken from Sigma) was determined against HIV-1(IIIB), HIV-2(ROD), and SIVmac251 in cell culture. AB was able to inhibit HIV-1(IIIB) infection in CEM cell cultures at an EC50 of 3.7 ± 2.0 μM (Table (Table1).1). It was not toxic at 77 μM. AB was active against HIV-2 (ROD), but at a fivefold-higher concentration than HIV-1 (17 ± 4.4 μM). In U87/CD4.CCR5.CXCR4 cell cultures, AB inhibited SIVmac251 infection at an EC50 of 12 ± 3.2 μM. AB had an antiviral activity that was only 2- to 10-fold below that of PRM-A, but it was clearly less active than the plant lectins GNA, HHA, or UDA. In PMBC, AB blocked HIV-1 (laboratory strain BaL, NL4.3, and HE) at an EC50 of around 6.4 μM and was also inhibitory against several members of different X4 and R5 HIV-1 clades (Table (Table1).1). AB was also evaluated against a broad variety of other DNA and RNA viruses, including herpes simplex virus type 1 (HSV-1) and HSV-2, vaccinia virus, vesicular stomatitis virus, reovirus 1, respiratory syncytium virus, Semliki Forest virus, Sindbis virus, influenza viruses A and B, feline infectious peritonitis virus, and Punta Toro virus. AB showed an activity against HSV-1 and HSV-2 comparable to its activity against HIV-1 (EC50 of around 7.7 μM). It showed also minor activity against vaccinia virus and vesicular stomatitis virus (EC50s of 35 μM for both viruses), but AB did not show any activity against any of the other viruses tested.
The inhibitory activity against syncytium (giant cell) formation between persistently HIV-1-infected HUT-78 lymphocytic cells and uninfected Sup-T1 cells was determined for AB by using HHA, UDA, and PRM-A as control drugs. AB efficiently prevented syncytium formation, with an EC50 of 6.2 ± 2.2 μM, while the EC50s of HHA, UDA, and PRM-A were 0.048 ± 0.0028 μM, 0.51 ± 0.26 μM, and 4.1 ± 1.6 μM, respectively.
Since DC-SIGN-mediated capture of HIV has been thought to represent an important venue for HIV transmission (8, 9, 22), AB was also evaluated for its ability to inhibit capture of HIV-1 by DC-SIGN-expressing Raji/DC-SIGN cells and subsequent transmission of the virus to C8166 CD4+ T cells. The compound dose dependently inhibited capture of HIV-1 by the Raji/DC-SIGN cells at an EC50 of 73 ± 6.0 μM. HHA and UDA had EC50s ranging between 0.98 and 2.1 μM.
When the HIV-1-exposed Raji/DC-SIGN cells that had been preexposed to the different concentrations of the CBAs were cocultured with uninfected T-lymphocytic C8166 cells after the CBAs had been removed by carefully washing of the Raji/DC-SIGN cells, syncytium formation was reduced or even prevented in those cocultures to which the higher concentrations of the test compounds had been administered. The EC50 of virus transmission for AB was 55 ± 11 μM, whereas the EC50s for HHA, GNA, and UDA were within the higher nanomolar or lower micromolar range.
In order to obtain more insights into the timing of antiviral action of AB in the replicative cycle, a time-of-drug-addition experiment was performed. MT4 cell cultures were exposed to HIV-1(IIIB) and the compounds were added at different time points (i.e., at 0, 1, 2, 3,… 9, 24, and 25 h) postinfection, as indicated in Fig. Fig.3.3. After 31 h, p24 levels were measured and used as a parameter for the degree of virus replication. In these experiments, virus production can be efficiently inhibited when the test compound is added at a time point well before its antiviral target is operative in the viral life cycle. When the compound is added after its antiviral target becomes operative, virus replication is not inhibited. DS8000, a polyanionic adsorption inhibitor, and AMD3100, a CXCR4 coreceptor antagonist, were likewise unable to block virus infection when added at 1 h (or later) post-virus infection (Fig. (Fig.3).3). Likewise the CBAs HHA, UDA, and PRM-A act on a very early step in the viral life cycle, since addition later than 1 h post-virus infection resulted in the loss of their capacity to block virus replication. In contrast, the addition of nevirapine, a reverse transcriptase inhibitor, could be delayed up to 4 h without losing its antiviral potential. Alcian Blue, like the CBAs, acts on a very early step in the life cycle of HIV-1, since its antiviral potential was heavily compromised as soon as it was added to the virus-infected cell cultures at a time point later than 1 h postinfection (Fig. (Fig.33).
HIV-1(IIIB)-infected CEM T-cell cultures in 48-well plates were exposed to an AB concentration that corresponded to one to two times its EC50. Every 3 to 4 days, 50 μl of the drug-exposed virus-infected cell suspensions was transferred to 950 μl of freshly prepared CEM cell culture. Giant cell formation was recorded microscopically, and once full cytopathic effect was observed, the drug concentration was increased (Fig. (Fig.4).4). Virus isolates were collected at different time points during the selection process. Up to 80 passages were needed to obtain a virus isolate (number 7) that was able to replicate in the presence of AB at concentrations that were at least ≥10-fold the EC50 for wild-type virus (Fig. (Fig.44).
At different time points during the selection process of AB-resistant HIV-1(IIIB) strains, viral isolates were collected from the infected CEM cell cultures (Fig. (Fig.4)4) and the envelope gene, which encodes gp120 and gp41, was amplified and sequenced. The data were compared to the env genetic background of wild-type HIV-1(IIIB), which was subcultured in parallel but in the absence of AB. The mutational patterns are displayed in Table Table2.2. It took up to 70 passages for the first N-glycan deletions to appear in isolate 5. From then on, two additional mutations at N-glycosylation sites appeared, and by passage 80, four N-glycans were deleted in gp120 (isolate 7) (Fig. (Fig.5).5). Two of these glycans were originally defined as high-mannose type, and two glycans were determined to be complex type (24). In addition, an extra glycosylation motif was created by a proline-to-serine mutation at position 460 in isolate 4, but this glycosylation motif was not preserved during the further course of the selection experiments.
Besides the glycan deletions in HIV-1 gp120, a number of additional (nonglycosylation site) mutations in gp120 and also gp41 were observed (Table (Table2).2). Whereas most of the nucleotide mutations did not change the nature of the amino acid or resulted in homologous amino acid changes, a number of them selected for different amino acids in gp160, replacing the original amino acids for basic (positively charged) amino acids (i.e., I182→K; N348→K; N631→K; E643→K; Q545→H).
The AB-resistant HIV-1(IIIB) isolate (7) was evaluated for its sensitivity against the CBAs HHA, GNA, UDA, and PRM-A, and the polyanions DS5000 and PVAS were included as controls (Fig. (Fig.6).6). The EC50s of the compounds for the AB-resistant HIV-1(IIIB) isolate were determined as performed for wild-type virus. Although four N-glycan deletions were observed in AB-resistant isolate 7, the phenotypic activities with HHA, GNA, UDA, and PRM-A remained unchanged. Also, as expected, the antiviral activities of the polyanions DS5000 and PVAS and the CXCR4 antagonist AMD3100 were nearly similar. Compared with wild-type virus, the mutant HIV-1(IIIB) isolate 7 did show a marked decrease in sensitivity toward AB: it became approximately 13-fold less sensitive to the inhibitory activity of this phthalocyanine derivative.
The CBAs are an interesting new category of antiviral compounds that target the glycans on the envelope of viruses like HIV (for a review, see references 4 to 6). However, since most CBAs are of a peptidic nature, which may cause problems such as high production costs, unfavorable pharmacokinetics, and possible immunological reactions, it is important to look for novel nonpeptidic low-molecular-weight compounds that behave as CBAs and prevent HIV infection. Recently, the antiviral properties of pradimicin A, an α-1,2-mannose-binding nonpeptidic semisynthetic antibiotic that binds to the glycans of HIV-1 gp120 and selects for glycan deletions in HIV-1 gp120, were described (12).
In this study, we showed for the first time that a synthetic prototype compound within the structural class of the phthalocyanines, designated Alcian Blue, selectively inhibits retrovirus infection, presumably by binding to the glycans of the HIV-1 gp120 envelope. In contrast with the vast majority of CBAs that are endowed with anti-HIV activity, this compound is a synthetic nonpeptidic agent. Time-of-addition experiments clearly showed that AB acts on an antiviral target that is operative in an early stage of the viral life cycle (i.e., entry/fusion) but at a slightly later time point than DS8000, an attachment (adsorption) inhibitor of the HIV infection process. In fact, the time-of-addition study showed that AB acts at exactly the same time point as the CBAs HHA, UDA, and PRM-A. Our observation that AB dose dependently prevents syncytium formation (fusion) between persistently HIV-1-infected T cells and uninfected T lymphocytes is in agreement with the time-of-drug addition data, as well as with the mechanism of action of CBAs. This property may have the advantage that, if virus escaped the entry block afforded by AB, the transmission of virus from the virus-infected cell through cell-cell contact can still be blocked in the presence of the drug.
HHA shows specificity for internal (α-1,3 and α-1,6) mannoses, while PRM-A preferentially binds to α-1,2-mannose residues (7, 32). The glycoprotein gp120 of HIV is not only highly glycosylated, but also, ~ 33% of the N-glycans are high-mannose type (24). This is remarkable, since human cells usually do not express high-mannose-type sugars on their glycoproteins and if they do, the high-mannose-type glycans never occur as dense on the glycoprotein as on gp120. In fact, the C-type mannose-binding lectin, which is part of the innate immune system, recognizes high-mannose-type sugars, initiating the lectin pathway (20). However, whereas mannan reverses the anti-HIV activity of the CBAs HHA, GNA, and PRM-A (12), high concentrations of mannan were not able to reverse the antiviral activity of AB (data not shown). Also, only 50% of the N-glycans deleted upon increasing concentrations of AB were high-mannose type. These data may indicate that AB may not preferentially recognize high-mannose-type glycans on HIV-1 gp120. The carbohydrate spectrum of AB still needs to be determined. In this respect, it is still unclear why AB tends to be more active against HIV-1 strains (EC50, 3.7 to 9.3 μM) than HIV-2(ROD) (EC50, 17 μM) or SIVMac251 (EC50, 12 μM). The fact that the HIV-1(IIIB) and HIV-2(ROD) strains contain an equal amount of N-linked glycosylation sites may rule out that the differences in activity are simply due to the number (density) of N-glycans on the viral envelope. Glycan specificity, which has not been characterized for AB (38), may play a more important role in its eventual antiviral efficacy, as well as the importance of the interaction of specific (glycosylated) envelope epitopes with the viral receptors. In fact, CBAs turned out to be potent inhibitors of Dengue virus entry and infection, whereas the viral E-envelope protein only contains two potential N-glycans (1).
Besides mannose-binding lectin, other human mannose-binding lectins interact with gp120 of HIV. DC-SIGN, a lectin present on dendritic cells, functions in dendritic cell recognition and mediates uptake of pathogens, leading to antigen presentation to T cells (17, 18). Alike the CBAs HHA, GNA, and UDA (11), AB was able to inhibit the binding of HIV-1 to DC-SIGN and the subsequent virus transmission to T cells. These observations indicate that CBAs such as Alcian Blue may not only have the potential to prevent cell-free HIV infection and syncytium formation between HIV-infected cells and noninfected cells but also the capture and subsequent transmission of the virus by DC-SIGN-expressing cells. These properties are important in view of a potential microbicidal application for this drug class.
Another important similarity with the CBAs such as HHA, UDA, or PRM-A is the selection of mutant virus strains that predominantly show deletions in the glycans on HIV gp120. Interestingly, an accumulation of four N-glycan deletions occurred under exposure of HIV-1 to escalating AB concentrations (Fig. (Fig.5).5). These four N-glycosylation site positions in gp120 have also been shown to be deleted under HHA, GNA, and/or UDA pressure (5). While the plant lectins HHA and GNA showed a clear preference for deleting high-mannose-type N-glycans, this preference was not so (statistically) clear for AB, since only two out of four deleted glycans were high-mannose types. The selection of the AB-resistant virus strain containing four deletions in N-linked glycans of gp120 took up to 80 subcultivations in the presence of escalating drug concentrations. Similar long selection times have been observed for other CBAs as well, including GNA and HHA (65 to 70 passages) (2) and UDA (90 passages) (3), whereas somewhat shorter selection times were required for PRM-A (15 to 20 passages) (10). However, in contrast with the AB-resistant virus strains, no additional positively charged amino acid changes have been observed in their gp120/gp41 envelope for the virus strains resistant to HHA, GNA, UDA, and PRM-A.
Also noteworthy is the observation that none of the CBAs tested showed cross-resistance toward the AB-resistant HIV-1(IIIB) isolate 7 (Fig. (Fig.6).6). A possible explanation for this observation could be that HHA, GNA, UDA, and PRM-A need a higher number of high-mannose-type glycan deletions in gp120 to become phenotypically resistant. The lack of cross-resistance against the other CBAs is also in line with our previous findings that many CBAs have a high genetic barrier, requiring multiple (≥4) glycan deletions in HIV-1 gp120 before phenotypic resistance becomes measurable. Alike the CBAs, AB is also endowed with a relatively high genetic barrier. It took up to 70 passages (~245 days) for the first N-glycan mutations to appear and up to 80 passages (~ 280 days) for HIV-1 to become moderately (~ 10-fold) resistant to AB (Fig. (Fig.6).6). In this respect, AB closely mimics the antiviral mechanism of action and resistance profile of the plant and prokaryotic lectins.
The cationic phthalocyanine derivative Alcian Blue has been used to quantify glycosaminoglycans by its ability to form complexes with such carbohydrates (39). It was also previously shown by Vzorov et al. (36, 37) to inhibit HIV in cell culture. Many other structurally related porphyrins, like the metalloporphyrin-ellipticine complexes (15) and the negatively charged carboxyphenyl porphyrine derivatives (14), have been described as potent inhibitors of HIV infection. It is clear that the class of porphyrins, either cationic, neutral, or anionic, may show promise as potential anti-HIV drug leads for systemic as well as microbicidal applications. However, it seems likely that cationic and anionic derivatives, although they both block virus entry by interacting with gp120, may inhibit HIV infectivity by different mechanisms. The mutational pattern observed under escalating Alcian Blue concentrations provided evidence that this compound blocks viral infection by interacting with the envelope glycans. Such a mechanism of action is very interesting because it provokes a unique resistance profile by forcing the virus to delete its glycans on its envelope. By doing this, it is speculated that AB triggers the immune system to produce a neutralizing antibody or cellular immune response against the uncovered immunogenic epitopes and, thus, AB may represent a unique prototype agent for further exploration as a potential antiviral agent. However, not all mutations in the envelope of HIV-1 occurred in N-glycosylation motifs (Table (Table2).2). Although most of the mutations in the envelope resulted in changes to an identical or homologous amino acid, five mutations led to the creation of positive charges by creating a lysine or histidine residue in the envelope. It may be of particular importance to notice that three out of the five created positive charges in the HIV-1 envelope occurred in gp41. The significance of these observations is currently still unclear. It cannot be excluded that the appearance of the positively charged amino acids may afford a repulsing effect against the (positively charged) AB molecules. Surface plasmon resonance interaction studies of AB with immobilized gp120 and gp41 to clarify this issue failed due to a significant level of aspecific binding of AB to the envelope-coated chip (B. Hoorelbeke, personal communication). Thus, although Alcian Blue undoubtedly targets the glycans on the envelope of HIV-1, this may not be its sole mode of antiviral activity. It cannot be excluded that CBAs such as AB also interact with glycoproteins that are present on the cell surface and which play a direct or indirect role in the entry process of HIV. In this respect, we have observed an interaction of AB with one of the coreceptors of HIV (i.e., CCR5) by fluorescence-activated cell sorter analysis, but the significance of this observation in the eventual antiviral efficacy of AB is currently unclear and will be the subject of further studies.
More than 25 years after the discovery of HIV and AIDS, the search for an efficient vaccine is still ongoing and is more technically challenging than ever expected. Microbicides are chemical agents that can be used intravaginally by women. The ideal microbicide should be easy to use, active against most HIV strains, and safe, but also economically affordable for people in the developing world (8, 22). In this respect, CBAs are especially promising, since they have a high genetic barrier and enable “neutralization” of a broad variety of HIV clades (5, 10). Alcian Blue, belonging to the chemical class of phthalocyanines, opens the way to the exploration of other (cationic) phthalocyanines in the development of novel antiviral drugs targeting the glycans on the envelope glycoproteins of HIV. It has the same interesting properties as the CBAs HHA, UDA, and PRM-A and, thus, AB can be considered a prototype compound in the development of new CBA microbicides of a nonpeptidic nature.
This work was supported by grants from the K.U. Leuven Centers of Excellence (EF-05/15), the Geconcerteerde Onderzoeksacties (no. 05/19), and the Fonds voor Wetenschappelijk Onderzoek (G.0485.08) and the CSIC-Intramural PIF 08-022.
We thank Rebecca Provinciael, Sandra Claes, Leen Ingels, Kristien Erven, Cindy Heens, and Leentje Persoons for excellent technical assistance and Christiane Callebaut for fine editorial help. V.L. thanks the CSIC and the FSE for an I3P predoctoral contract.
Published ahead of print on 31 August 2009.