Red Algal Sulfated Galactan Binds and Protects Neural Cells from HIV-1 gp120 and Tat

The potential neuroprotective capacity of four different sulfated glycans: Botryocladia occidentalis-derived sulfated galactan (BoSG) (MW > 100 kDa), Lytechinus variegatus-derived sulfated fucan (LvSF) (MW~90 kDa), high-molecular weight dextran sulfate (DxS) (MW 100 kDa), and unfractionated heparin (UFH) (MW~15 kDa), was assessed in response to the HIV-1 proteins, R5-tropic glycoprotein 120 (gp120) and/or trans-activator of transcription (Tat), using primary murine neurons co-cultured with mixed glia. Compared to control-treated cells in which HIV-1 proteins alone or combined were neurotoxic, BoSG was, among the four tested sulfated glycans, the only one capable of showing significant concentration-dependent neuroprotection against Tat and/or gp120, alone or combined. Surface plasmon resonance-based data indicate that BoSG can bind both HIV-1 proteins at nM concentrations with preference for Tat (7.5 × 10−8 M) over gp120 (3.2 × 10−7 M) as compared to UFH, which bound gp120 (8.7 × 10−7 M) over Tat (5.7 × 10−6 M). Overall, these data support the notion that sulfated glycan extracted from the red alga B. occidentalis, BoSG, can exert neuroprotection against HIV-1 Tat and gp120, potentially via direct molecular interactions.


Introduction
Human immunodeficiency virus type 1 (HIV-1) infection is associated with neurological symptomatology, including cognitive deficits, affective disorders, and motor impairment [1]. Neurological symptoms, collectively termed 'neuroHIV', persist despite treatment with combined antiretroviral therapeutics (cART), likely due to poor accumulation of cART within the central nervous system where viral neurotoxins promote inflammation and neuronal damage [2,3]. The most well-characterized neurotoxic HIV-1 proteins include the HIV-1 envelope protein, glycoprotein 120 (gp120), and the HIV-1 regulatory protein, trans-activator of transcription (Tat). Through shared and separate neuroinflammatory and excitotoxic mechanisms, gp120 and Tat promote synaptodendritic damage and neuronal death [4][5][6]. Expression of either protein is enough to recapitulate many aspects of neuro-HIV in rodent models [7,8]. Thus, the identification of centrally active and safe therapeutics that can ameliorate gp120/Tat-mediated neurotoxicity is important for the improvement of current cART strategies.
Herein, we evaluate the potential neuroprotective properties of four structurally distinct sulfated glycans against the HIV-1 proteins gp120 and Tat. We examined unfractionated heparin (UFH) [25], 100 kDa-dextran sulfate (DxS(100)) [26], the tetrasaccharide repeating sulfated fucan isolated from the sea urchin Lytechinus variegatus (LvSF) [27], and the disaccharide repeating sulfated galactan isolated from the red alga Botryocladia occidentalis (BoSG) [28]. We conducted an in vitro live/dead cell-based assay using cocultured primary striatal medium spiny neurons and mixed glia obtained from mice, as well as a binding assessment using surface plasmon resonance (SPR) to begin to assess the neuroprotective capacity of these compounds and their potential interactions with HIV-1 proteins. We anticipated the sulfated glycans to ameliorate gp120 and Tat-mediated neurotoxicity and to bind these proteins directly. Furthermore, we suggest that the present study yields new insights on the sulfated glycan structural components that are important for HIV interactions.

Binding Capacity of BoSG and UFH of HIV-1 Proteins
In follow-up to the functional effects for BoSG's neuroprotection against HIV-1 Tat and/or gp120, SPR measurements were performed to assess the individual binding affinities of BoSG for either protein. Data were compared to UFH as a well-known standard ( Figure 5). After biotinylation (reaction described in Materials and methods), both BoSG and UFH were immobilized on a high-affinity streptavidin (SA) sensor chip. Then, the HIV-1 proteins were injected to the flow cell of a SA chip at a flow rate of 10 μL/min at different final nanomolar concentrations (see color-coded curves). Binding measurements were conducted at 25 °C. Successful immobilization of the sugars could be confirmed by a ~200 resonance unit (RU) increase in the sensor chip. The resultant binding kinetics were demonstrated by the curves of response difference as a function of time and protein concentrations in the sensorgrams. From these sensorgrams, kinetic values such as association rate constant (ka), dissociation rate constant (kd), and dissociation constants (KD) for the four intermolecular complexes were generated (Table 1).

Binding Capacity of BoSG and UFH of HIV-1 Proteins
In follow-up to the functional effects for BoSG's neuroprotection against HIV-1 Tat and/or gp120, SPR measurements were performed to assess the individual binding affinities of BoSG for either protein. Data were compared to UFH as a well-known standard ( Figure 5). After biotinylation (reaction described in Materials and methods), both BoSG and UFH were immobilized on a high-affinity streptavidin (SA) sensor chip. Then, the HIV-1 proteins were injected to the flow cell of a SA chip at a flow rate of 10 µL/min at different final nanomolar concentrations (see color-coded curves). Binding measurements were conducted at 25 • C. Successful immobilization of the sugars could be confirmed by ã 200 resonance unit (RU) increase in the sensor chip. The resultant binding kinetics were demonstrated by the curves of response difference as a function of time and protein concentrations in the sensorgrams. From these sensorgrams, kinetic values such as association rate constant (ka), dissociation rate constant (kd), and dissociation constants (K D ) for the four intermolecular complexes were generated ( Table 1). The SPR-derived results have indicated that BoSG can bind to both HIV proteins at similar molar ranges of heparin, between tens of nanomolar to a few micromolar concentration. BoSG bound tighter to Tat (7.5 × 10 −8 M) compared to UFH-Tat (5.7 × 10 −6 M), UFH-gp120 (8.7 × 10 −7 M), and BoSG-gp120 (3.2 × 10 −7 M) ( Table 1). These data indicate that BoSG presents higher specificity of interaction with Tat than gp120, and this supports the observation from the cell-based assay (Figures 2 and 3). The result from the SPR study also supports the finding that BoSG presents stronger affinity for both HIV-1 proteins than UFH (two orders of magnitude in the case of Tat and almost three-fold more in the case of gp120), justifying the necessity of lower concentrations of BoSG to inhibit the neurotoxic effects of the HIV proteins in the cell-based assay ( Table 1). As confirmed by SPR, results have significantly indicated stronger, although not very pronounced, affinity of BoSG for the HIV-1 proteins as compared to UFH.  Standard deviation values, after ± symbol, were obtained from global fitting of five different injections using a 1:1 Langmuir model.
The SPR-derived results have indicated that BoSG can bind to both HIV proteins at similar molar ranges of heparin, between tens of nanomolar to a few micromolar concentration. BoSG bound tighter to Tat (7.5 × 10 −8 M) compared to UFH-Tat (5.7 × 10 −6 M), UFH-gp120 (8.7 × 10 −7 M), and BoSG-gp120 (3.2 × 10 −7 M) ( Table 1). These data indicate that BoSG presents higher specificity of interaction with Tat than gp120, and this supports the observation from the cell-based assay (Figures 2 and 3). The result from the SPR study also supports the finding that BoSG presents stronger affinity for both HIV-1 proteins than UFH (two orders of magnitude in the case of Tat and almost three-fold more in the case of gp120), justifying the necessity of lower concentrations of BoSG to inhibit the neurotoxic effects of the HIV proteins in the cell-based assay ( Table 1). As confirmed by SPR, results have significantly indicated stronger, although not very pronounced, affinity of BoSG for the HIV-1 proteins as compared to UFH.  Table 1. Kinetic values from the molecular interactions between UFH and BoSG with Tat and gp120.

Discussion
The present data upheld the hypothesis that sulfated glycans are a class of natural products that can functionally ameliorate HIV-1 Tat and/or gp120-mediated neurotoxicity. In the current study, a set of four different sulfated glycans were used: UFH as a standard control as widely exploited in many pharmacological works investigating sulfated glycans and/or GAGs [25]; DxS(100) as the second control since this particular sulfated glycan has been widely investigated in the past regarding its anti-HIV activity [12], and the two marine unique compounds: the sea-urchin-derived sulfated fucan from L. variegatus, LvSF [27] and the sulfated galactan derived from the red alga B. occidentalis, BoSG [28]. These two marine sulfated glycans show different chemical and biological features. The LvSF is regular and defined in terms of structure and lacks biological effect as an anticoagulant and antithrombotic agent, while BoSG is more sulfated and heterogeneous; and exhibits great effects when investigated as an anticoagulant and antithrombotic agent [29]. Hence, structure-function relationship studies in neuroHIV can be achieved using this set of four structurally and functionally different sulfated glycans. Compared to the other tested sugars, the major structural difference in BoSG is that it is composed solely of galactose units.
Overall, data reveal BoSG to exert greater neuroprotection compared to the controls, DxS(100) and UFH. Importantly for therapeutic consideration, this red algal sulfated galactan demonstrates no increased bleeding effects as opposed to the well-documented hemorrhagic side-effects of DxS and UFH after intravenous administration [30]. DxS and UFH are also known to present thrombocytopenic effects in treatments [31,32], including those in HIV therapy [32]. As opposed to LvSF, which showed significant neuroprotective action only against the proteins combined at the highest concentration, BoSG was capable of significantly inhibiting in a concentration-dependent manner both HIV proteins in all tested conditions, proteins alone and/or combined. The decision to conduct further SPR analyses compared to the control UFH was made, therefore, in favor of only BoSG. Data from SPR confirm preferential binding of BoSG with Tat over gp120. Together, the data from this investigation provide new insights in the consideration of the marine sulfated glycans, particularly the red algal sulfated galactan, BoSG, for their capacities to interact with and ameliorate HIV-1 protein neurotoxicity.
Previously, the role of HIV-1 gp120 in the binding of host cell HSPGs was identified as a key process for infection [9,11]. Several sites of molecular interaction may mediate these effects. HSPG binding is partly modulated by a conformational change in gp120 that is induced through its interaction with CD4 [11]. The V3 loop (CD4-induced epitope) of gp120 is essential for binding to HS and for the initial process of T-cell infection [9]. The V3 loop contains a binding domain for CD4 that, when activated, induces a conformational change that is necessary for gp120-HSPG interactions to occur [9]. As such, the HSPG interaction is thought to follow the initial step of gp120 binding to CD4 during viral-cell attachment [9].
Through residue mapping experiments using mutagenesis and SPR spectroscopy, the gp120 amino acids Arg-419, Lys-421, and Lys-432 have been identified to be highly important for the binding of HS in the presence of CD4 [11]. These positively charged amino acids comprise the typical binding site of GAG-binding proteins [33]. In addition to this GAG binding site, other binding sites were also mapped and found to comprise the V2 loop, a portion of the C-terminal, and the CD4-induced bridging sheet, together with the V3 loop [11]. Interestingly, three of these gp120-GAG-binding sites undergo conformational change upon interaction with CD4 and are directly involved in co-receptor recognition [11]. These data suggest that gp120-HSPG-binding sites interact with HIV-1 co-receptors, making them an intriguing target for novel HIV-1 entry inhibitors.
Molecular interactions between HIV-1 Tat and HSPGs are also likely to occur. Following infection, transcribed HIV-1 Tat acts on the HIV-1 LTR, greatly increasing the efficiency of viral transcription. However, as a soluble protein that can be secreted from infected cells, Tat also exerts effects on the extracellular environment and nearby uninfected cells that further catalyze HIV infectivity through multiple mechanisms [34]. Tat may bind to a component of the extracellular milieu, the HSPG, to increase its effective concentration at the cell-surface and to gain access to nearby uninfected cells [14]. Tat-HSPG interactions may occur via binding of the basic region of Tat, serving as the HS (or GAG) binding domain [14]. This prior work suggested that HS was responsible for the attachment, sequestration, and retention of Tat at the cell surface, but the molecular player(s) responsible for cellular uptake were not known until the discoveries made 4 years later. These findings revealed HS acts as a cell membrane surface receptor for Tat and its consequent cellular internalization [16]. Furthermore, the binding and uptake of Tat can be prevented by using competitive soluble heparin or GAG lyases that can cleave the HS chain [16]. These data support a role for HS in HIV-1 viremia.
Given demonstrations of HS binding to and internalizing Tat during the HIV infection, many other works have appeared in the literature discussing the aspects of the HS-Tat interactions and their biochemical consequences in HIV/AIDS [13,15,17,18]. Among these, two of great relevance include demonstrations that (1) the binding and clustering of Tat molecules by HSPGs occurs at low-micromolar concentrations [18] and (2) a quaternary complex containing Tat homodimers with two HSPGs (HS/Tat-Tat/HS) are physiologically assembled in order to promote adhesion, migration, and extravasation of lymphocytes across the endothelium during HIV infection [17]. The current results provide evidence that the marine sugars, particularly BoSG, can efficiently compete with HSPGs required for HIV neuroinfection.
Future in vivo studies to assess efficacy and safety of BoSG must be also conducted to advance the status of this marine sulfated glycan as a potential drug candidate in phase I clinical trials. Other open questions may be related to bioavailability and to how this marine molecule can efficiently reach the central nervous system (CNS) for the desired effects. Data from the literature have demonstrated that BoSG to exert venous antithrombotic actions when intravenously administered [35]. This observation supports the notion that BoSG can be bioactive in the body. Although the blood-brain barrier (BBB) permeability for native BoSG and low MW derivatives is virtually unknown, data exist in the literature demonstrating low MW heparin derivatives to efficiently pass through the BBB [36]. The production and exploitation of low MW BoSG derivatives may prove to be a promising strategy to increase the efficacy and bioavailability of BoSG in the CNS. Melo et al. have already shown the paths for production and biomedical use of such BoSG oligosaccharides [37]. Future investigations aimed to assess the influence of native BoSG and its derived oligosaccharides on HIV-1 viremia are also required, as are investigations on the effects at other cell types.

Sulfated Glycans
Purification of BoSG and LvSF were conducted as previously described [27,28]. Briefly, 20 mg of crude polysaccharides obtained after papain digestion were extracted from the respective tissues (red algal body wall and sea urchin egg jelly) and applied to a DEAEcellulose column (2.5 cm diameter × 20 cm width) that was equilibrated with 50 mM sodium acetate buffer (pH 5.0) and washed with 300 mL of the same buffer containing 0.2 M NaCl and 10 mM EDTA. Thereafter, the column was eluted by a linear gradient prepared by mixing 2000 mL of 50 mM sodium acetate buffer (pH 5.0) containing 0.2 M NaCl and 10 mM EDTA with 200 mL of 1.2 M NaCl in the same buffer. The flow rate of the column was 12 mL/h. Fractions of 1.5 mL were collected and checked by metachromatic properties [38]. The chromatograms of the two materials resulted in three major fractions. The third fraction of each chromatography corresponded to BoSG and LvSF. Prior to submission of samples, either for cell-based in vitro assay or SPR studies, the purity level and the structural integrity of BoSG and LvSF were checked by 1D 1 H (400 MHz) NMR spectra acquired with 32 scans at the 400 MHz Bruker Avance III HD equipped with 5 mm BBFO RT probe. The resultant NMR spectra indicated purity of >98%, no residual contamination from solvents, and an NMR signal pattern as expected with the previous publications [27,28]. For NMR analysis,~5 mg of sugar samples were individually dissolved in 550 µL of deuterium oxide (D 2 O) "100%" (D 99.96%), purchased from Cambridge Isotope Laboratories, Inc. (Andover, MA, USA), and transferred to a 3 mm NMR tube for data acquisition. The NMR spectra, as well as the polyacrylamide gel electrophoresis for assessment of MW distribution of the two marine sample preparations, BoSG and LvSF, can be found in [39], in which the same sample preparations were studied.

Live/Dead Assay
Neuron viability was assessed using a LIVE/DEAD ® Viability/Cytotoxicity Kit (#L-7013, Molecular Probes, Eugene, OR, USA) per manufacturer instructions and as described [42,43]. Briefly, experimental treatments (sulfated glycans and/or HIV-1 proteins) were applied to primary co-cultures and live/dead assay was performed 20 h post treatment. Prior work utilizing time-lapse microscopy (0-60 h) identified the 20 h time-point as the earliest time when Tat-treated cells significantly diverged on the measure of viability [41]. A working solution of propidium iodide (ex/em: 535/617 nm) and Hoechst 33342 (ex/em: 360⁄460 nm) was prepared by diluting the stock solutions in Hank's Balanced Salt Solution containing Ca 2+ and Mg 2+ (HBSS; 1:50 and 1:10,000 dilution, respectively). Cell media was replaced with the propidium iodide/Hoechst 33342 working solution and incubated (37 • C; 5% CO 2 ) for 15 min in the dark. Cells were washed three times with HBSS and imaged on a Nikon Ti2-E inverted microscope (Nikon, Melville, NY, USA) with a motorized stage. Cells were quantified from images captured at 40 × magnification (S Plan Fluor, 0.6 NA). Viability was assessed by calculating the proportion of necrotic cells (# propidium iodide + cells/# Total Cells) × 100).

Statistical Analyses
Cell death was separately analyzed for BoSG and additional sulfate glycans by twoway ANOVA with compound and concentration as the between subjects' factors. Fisher's protected least significant difference post-hoc tests determined group differences following main effects. Interactions were delineated via simple main effects and main effect contrasts with alpha controlled for multiple comparisons. Analyses were considered significant when p ≤ 0.05. The reaction mixture was heated at 70 • C for 24 h and followed by further addition of 10 mg NaCNBH 3 .The reaction mixture was heated at 70 • C for an additional 24 h. After cooling to room temperature, the mixture was desalted with the spin column (3000 MWCO). Biotinylated BoSG and UFH were collected, freeze-dried, and used for SA chip preparation. The biotinylated BoSG and UFH were immobilized to streptavidin (SA) chip based on the manufacturer's protocol. In brief, 20 µL solution of biotinylated BoSG (0.1 mg/mL) or 40 µL solution of biotinylated UFH (2 mg/mL) in HBS-EP running buffer was injected over flow cell 2 (FC2) of the SA chip at a flow rate of 10 µL/min. The successful immobilization of these sulfated glycans was confirmed by the observation of a~200 to 400 resonance unit increase in the sensor chip. The control flow cell (FC1) was prepared by 1 min injection with saturated biotin.

Measurement of Interaction between Sulfated Glycans and Proteins Using BIAcore
The HIV-1 proteins (Tat or gp120) were diluted in HBS-EP buffer (0.01 M HEPES, 0.15 M NaCl, 3 mM EDTA, 0.005% surfactant P20, pH 7.4). Different dilutions of protein samples were injected at a flow rate of 40 µL/min. At the end of the sample injection, the same buffer was flowed over the sensor surface to facilitate dissociation. After a 3 min dissociation time, the sensor surface was regenerated by injection of 40 µL of 2 M NaCl. The response difference (RU) was monitored as a function of time (sensorgram) at 25 • C. The resulting sensorgrams were used for interaction kinetics and binding affinity determination: association rate constant: ka; dissociation rate constant: kd; and binding equilibrium dissociation constant: K D , K D = kd/ka, by global fitting using 1:1 Langmuir binding model from Biaevaluation software 4.0.1 (GE healthcare, Uppsala, Sweden).

Conclusions
The HIV-1 proteins gp120 and Tat are required for viral infection and replication of the virus. In addition, they may promote neurotoxicity despite cART treatment. Heparan sulfate or heparin are known to be neuroprotective against these two HIV proteins. Herein, we identify the neuroprotective effects of four different sulfated glycans and reveal a sulfated galactan derived from the red alga B. occidentalis as a compound with potential for the treatment of HIV-1 Tat/gp120-mediated neurotoxicity. Compared to the other tested sugars, the major structural difference in BoSG is that it is composed solely of galactose units, while UFH, DxS(100), and LvSF are composed of hexosamine and hexuronic acid, glucose, or fucose respectively. Sulfated galactans of different structures, such as the homogenous sulfated galactan from the sea urchin Glyptocidaris crenularis [44] or the 3linked 4-sulfated galactan derived from Codium isthmocladum [45], remain to be investigated in comparison to sugars of different monosaccharide compositions. Further study is needed to confirm the neuroprotective effect of the galactose-composed polysaccharides against HIV-1 proteins. Moreover, the anti-viremia capacity of BoSG and its potential interactions with cART remain questions for future investigation.