Inhibition of the Enhancement of Infection of Human Immunodeficiency Virus by Semen-Derived Enhancer of Virus Infection Using Amyloid-Targeting Polymeric Nanoparticles

Abstract

The semen-derived enhancer of virus infection (SEVI) is natural amyloid material that has been shown to substantially increase viral attachment and infectivity of HIV in cells. We previously reported that synthetic monomeric and oligomeric amyloid-targeting molecules could form protein-resistive coatings on SEVI and inhibit SEVI- and semen-mediated enhancement of HIV infectivity. While oligomeric amyloid-binding compounds showed substantial improvement in apparent binding to SEVI compared to monomeric compounds, we observed only a modest correlation between apparent binding to SEVI and activity for reducing SEVI-mediated HIV infection. Here, we synthesized amyloid-binding polyacrylate-based polymers and polymeric nanoparticles of comparable size to HIV virus particles (~150 nm) to assess the effect of sterics on the inhibition of SEVI-mediated enhancement of HIV infectivity. We show that these polymeric materials exhibit excellent capability to reduce SEVI-mediated enhancement of HIV infection, with the nanoparticles exhibiting the greatest activity (IC₅₀ value of ~4 μg/mL, or 59 nM based on polymer) of any SEVI-neutralizing agent reported to date. The results support that the improved activity of these nanomaterials is likely due to their increased size (diameters = 80-200 nm) compared to amyloid-targeting small molecules, and that steric interactions may play as important a role as binding affinity in inhibiting viral infection mediated by SEVI amyloids. In contrast to the previously reported SEVI neutralizing, amyloid-targeting molecules (which required concentrations at least 100-fold above the K_d to observe activity), the approximate 1:1 ratio of apparent K_d to IC₅₀ for activity of these polymeric materials, suggests the majority of polymer molecules that are bound to SEVI contribute to the inhibition of HIV infectivity enhanced by SEVI. Such size-related effects on physical inhibition of protein-protein interactions may open further opportunities for the use of targeted nanomaterials in disease intervention.

The most common mode for transmission of HIV is through sexual contact.¹ Recent studies have shown that several peptides found in high abundance in semen can form aggregated amyloid species capable of facilitating viral attachment and internalization in cells.^2–8 For instance, the amyloid form of a peptide derived from prostatic acid phosphatase, PAP 248-286 (known as the semen-derived enhancer of virus infection or SEVI), has been reported to increase the rate of HIV infection by up to 400,000-fold, making this naturally abundant amyloid species a particularly attractive target for development of molecules that can reduce the spread of HIV.^2,5,9–14

Previously, we reported that monomeric and oligomeric amyloid-targeting 6-methylbenzothiazole aniline (BTA) compounds could form protein-resistive coatings^15–19 on amyloids and could inhibit SEVI-mediated HIV infection.^20,21 We showed that while the monomeric BTA molecules were effective at inhibiting HIV infection, oligomeric versions of BTA exhibited improved activity for inhibiting SEVI-mediated HIV infection. These BTA oligomers also exhibited enhanced binding to the SEVI amyloid putatively through multivalent binding.²¹ Interestingly, while the binding of the BTA oligomers to SEVI dramatically increased compared to a monomeric BTA analog (e.g., by 590-fold in the case of a BTA pentamer), the improvement in the functional activity of these oligomers to reduce HIV infection was considerably less pronounced (e.g., by only 65-fold in the case of a BTA pentamer).

From these previous results, we hypothesized that the improved effect of the BTA oligomers on reducing SEVI-mediated HIV infection could be a result of the increased size of the BTA oligomers compared to BTA monomers, rather than their improved binding to SEVI. In order to test this hypothesis, here we investigated the effect of polymeric nanoparticles (NPs) containing covalently-attached amyloid-binding compounds on SEVI-enhanced HIV infection (Figure 1). We designed these polymeric NPs to be approximately the same size as HIV virions²² (~150 nm) to assess the effect of steric interferences on the reduction of SEVI enhanced HIV infection by an amyloid-binding agent. We found that these polymeric NPs had greatly (over 200-fold) increased potency in inhibiting SEVI-enhanced HIV infection in cells despite retaining similar (or slightly lower) binding properties to SEVI compared to an amyloid-binding monomer. The effects of these NPs on reduction of SEVI-mediated HIV infection were greater than the activity found for amyloid-binding oligomers with much stronger avidity. The results suggest that consideration of steric bulk in the design of SEVI-binding molecules could be an important factor along with thermodynamic binding for neutralizing the effects of natural, aggregated amyloid proteins on sexual transmission of HIV.

Figure 1.

Illustration of the inhibition of binding of HIV virions and SEVI amyloid fibrils using amyloid-binding small molecules or amyloid-binding polymeric nanoparticles. Inhibiting HIV-SEVI interactions results in reduction of SEVI-enhanced HIV infection in cells.

RESULTS AND DISCUSSION

Design, synthesis and characterization of polyacrylate-based polymers

Prior work has shown that polyacrylamide-based polymers carrying sialic acid groups could bind to Influenza virus particles and effectively inhibit the direct binding of the virus to cells through steric inhibition.²³ Inspired from these previous studies, we hypothesized that amyloid-binding polymers or polymeric nanoparticles could create a significant steric barrier to block interactions between SEVI and HIV virions, and, thus, effectively inhibit SEVI-mediated HIV infection. Here, we synthesized polyacrylate-based polymers carrying amyloid-binding BTA moieties. We chose to use polyacrylate-based polymers because they are non-toxic and synthetically accessible in useful quantities. Polyacrylate polymers are also known to form spherical nanoparticles of 100-200 nm in diameter,^24,25 which is approximately the same size as HIV virions.²² Moreover, their capability to form nanoparticles of relatively narrow dispersity makes it possible to more accurately assess efficacy for reducing HIV infection compared to a mixture of polymers with large variability in size, shape, and aggregation state.

To develop a polyacrylate-based polymer carrying amyloid-binding BTA moieties, we synthesized an analog of BTA-EG₆ (1, Figure 2)^16,19 comprising an acrylate group (2). This BTA-acrylate monomer 2 was then subjected to free-radical conditions^26,27 at elevated temperature in the presence of acrylic acid and a catalytic amount of AIBN (radical initiator) to generate the polyacrylate polymer 3 (Figure 2). Analysis of polymer 3 by static light scattering-multi angle light scattering (SLSMALS) revealed that this polymer had a number average molecular weight (M_n) of 68K, with a polydispersity index (PDI) of 1.63 (Table 1). Examination of polymer 3 by ¹H-NMR suggested that 56% of the residues in the polymer contained BTA groups.

Figure 2.

Scheme for the synthesis of BTA-containing polyacrylate-based polymer 3 and control polymer 5 (lacking amyloid-binding groups).

Table 1.

Properties of polyacrylate polymers 3 and 5, and nanoparticles (NP) derived from polymer 3.

Polymer	BTA Incorporation (%)	Mn	Mw	PDI	Apparent Kd to SEVI (μg/mL)	NP Diameter DLS (nm)	NP PD DLS	NP Diameter EM (nm)
Amyloid Binding Polymer, 3	56	68000	110000	1.63	14 ± 0.06	218	0.19	83
Control Polymer 5	0	13000	18000	1.42	-	-	-	-

As a control, we also synthesized polyacrylate-based polymer 5 (Figure 2), containing hexaethylene glycol groups but lacking the BTA moieties. This control polymer was designed to test whether amyloid-binding groups are necessary for reducing SEVI-mediated HIV infection. While this control polymer had a lower M_n and M_w than 3, it had a similar PDI as the BTA-containing polyacrylate-based polymer 3 (Table 1).

Generation and characterization of polymeric nanoparticles

The BTA-containing polyacrylate polymer 3 is substantially larger than the previously reported oligomers of BTA,²¹ and, thus, was expected to be useful to probe the effects of sterics on reducing SEVI-mediated HIV infection. However, the hydrophobic nature of 3 makes this polymer highly prone to uncontrolled aggregation in aqueous solutions. Formulation of polymer 3 into nanoparticles was, therefore, pursued with the goal of significantly narrowing the size and shape dispersity of aggregates, thereby making it easier to quantitatively assess the activity of these polymers in SEVI-mediated HIV infectivity assays.

In order to convert polymer 3 into nanoparticles, we used a previously reported solvent evaporation method^28,29 to generate spherical nanoparticles with average diameters of ~200 nm (PDI = 0.19, see Figure S1 in the Supporting Information), as determined by dynamic light scattering (DLS). Analysis of these nanoparticles by Cryo-electron microscopy (Cryo-EM, Figure 3A), suggested a smaller average diameter of the spherical nanoparticles of 83 nm. Since larger particles tend to scatter more light than smaller particles, estimations of size of nanoparticles by DLS will often reflect a greater weight given to larger particles that may be present in the sample even if they represent only a minor fraction of particles.³⁰ Thus, we consider the estimation of nanoparticle diameter obtained from Cryo-EM images as a more accurate method to determine the average size of nanoparticles in the sample. Regardless of the discrepancy in size of the nanoparticles as estimated by DLS or Cryo-EM, spherical polymeric nanoparticles with diameters between 83 and 200 nm are of similar shape and size as HIV virions.²²

Figure 3.

Representative Cryo-EM or non-Cryo TEM images of polymeric nanoparticles, SEVI fibrils, or a mixture of nanoparticles and SEVI fibrils. (A) Cryo-EM images of polymeric nanoparticles carrying amyloid-binding BTA groups. (B) Cryo-EM images of SEVI fibrils. (C) Unstained TEM images of a sample containing SEVI fibrils and polymeric nanoparticles. The insets represent higher magnification images of the respective samples.

We expected that nanoparticles in this size range should be large enough to impart significant steric interactions between virus particles and SEVI amyloid fibrils. When we incubated these particles with a solution of SEVI fibrils (Figure 3B), unstained transmission electron microscopy (TEM) images revealed a population of nano-sized objects that were consistent with SEVI fibrils coated with nanoparticles (Figure 3C). The nano-sized objects shown in Figure 3C were not present in samples containing nanoparticles (Figure 3A) or SEVI fibrils (Figure 3B) alone. These results from imaging, therefore, support that nanoparticles derived from polymer 3 can bind and coat SEVI fibrils, and, thus, may be capable of sterically inhibiting interactions between SEVI and HIV virions.

Lastly, we evaluated the stability of the polymeric nanoparticles derived from polymer 3 by DLS. When incubated at 25°C or 37°C in deionized water, we observed a slight decrease (< 10%) in the average size of the particles over the first 6 hours (at either temperature). Over a 48 hour period, we observed a decrease in average particle size by ~18% at 25°C and ~23% at 37°C (see Figure S2 in the Supporting Information). The polydispersity (PD) of the nanoparticles, however, remained unchanged at either temperature over the 2-day period. For comparison, polymeric nanoparticles with poor stability have been characterized with large observed changes in size (by > 30%) and PD (by > 10-fold) over a period of less than a few hours.³¹ Given that SEVI-mediated HIV infection in cells takes place within a 2 hour period,^20,21 the nanoparticles derived from polymer 3 appear to have suitable stability for evaluation in infectivity assays.

Binding studies of BTA-containing polyacrylate polymer 3 and control polymer 5 with SEVI fibrils

We quantified the interaction of polymer 3 and SEVI fibrils using a previously reported centrifugation method.^20,21 This assay revealed that polymer 3 bound to SEVI amyloid fibrils with an apparent K_d = 14 μg/mL (Table 1, see Figure S3 in the Supporting Information). If we use the number average molecular weight (M_n) as an estimate of the molecular weight (MW) of the polymer, the apparent K_d value is ~0.21 μM. For comparison, the reported K_d value for the binding of BTA-EG₆ to SEVI fibrils is 0.127 μM²⁰ (see also Table 2). We attribute this slightly weaker apparent K_d for polymer binding to SEVI compared to monomer to the likely possibility that the majority of the amyloid-binding BTA moieties are buried within the hydrophobic bulk of the polymer rather than remaining accessible to the aqueous environment and available for binding to an amyloid surface. While we expect nanoparticles to exhibit an apparent binding to SEVI that would be similar to polymer 3, the centrifugation assay used in these studies could not be applied to measure apparent binding constants for the nanoparticles due to the colloidal nature of the nanoparticle suspensions (which results in pelleting of the nanoparticles upon centrifugation when either bound or unbound to SEVI).

Table 2.

Comparison of BTA containing compounds

Compound	Apparent Kd (μM)	IC50 (μM)	IC50 normalized to [BTA]tot (μM)	Ratio (Kd:IC50)
BTA-EG6 (Monomer) (1)	0.127 ± 0.022a	13c	13	1:102
BTA-Pentamer (6)	0.0004 ± 0.0002b	0.20d	1	1:500
Polymer	0.206 ± 0.0009	0.21d	23	1:1
Nanoparticle	-	0.059d	6	-

^aFor details, see reference 20

^bFor details, see reference 21

^cIC₅₀ values reflect reduction of SEVI-enhanced infection of HIV_IIIB in CEM-M7 cells

^dIC₅₀ values reflect reduction of SEVI-enhanced infection of HIV_IIIB in TZM-bl cells

In order to evaluate whether control polymer 5 (which lacks fluorescent amyloid-binding groups) exhibits any significant non-specific binding to SEVI, we performed a competition assay where we incubated a constant concentration of 100 μg/mL of polymer 3 and 10 μg of SEVI amyloid fibrils in water in the presence of increasing concentrations of control polymer 5. This competition assay made it possible to evaluate the binding between polymer 5 and SEVI by measuring the decrease in the fluorescence of polymer 3 if competitively displaced from the SEVI surface.^20,21 As expected, the results show that polymer 5 does not exhibit any significant binding to SEVI at concentrations ≤ 1 mg/mL (i.e., at concentrations that were 10-fold higher than polymer 3, see Figure S4 in the Supporting Information).

BTA-containing polymeric nanoparticles effectively inhibit SEVI-mediated HIV infection

We measured the half maximal inhibitory concentration (IC₅₀) of HIV infectivity with and without the presence of SEVI amyloid fibrils for our polyacrylate polymer 3 and our polymeric nanoparticles using a luciferase reporter cell line (TZM-bl cells²¹) that was exposed to infectious HIV_IIIB. For BTA-containing polyacrylate polymer 3, we found an IC₅₀ value of ~14 μg/mL (Figure 4 A,B) for inhibition of SEVI-enhanced HIV infection. This IC₅₀ value corresponds to a concentration of ~200 nM polymer, which is similar in activity to the previously reported most potent amyloid-targeting BTA-pentamer 6²¹ (Figure 5) for inhibiting SEVI-mediated infection of HIV. Additionally, polymer 3 was significantly more active on a per mole basis in this infection assay compared to BTA-EG₆ 1.²⁰ When we consider the activity of polymer 3 on a per BTA basis (IC₅₀ = 23 μM), however, the polymers appeared to have activity that was similar to BTA-EG₆ 1 (IC₅₀ = 13 μM). This lower than expected apparent activity for polymer 3 on a per BTA basis may be due to the heterogeneity of polymer aggregates that are likely formed in solution, resulting in a mixture of species with varying activity for binding to SEVI (and subsequent inhibition of SEVI-mediated HIV infection).

Figure 4.

Representative graphs displaying the efficacy and toxicity of the polymers 3 and 5 as well as nanoparticles (NP) formulated from 3. (A) Graph showing the reduction of SEVI-mediated enhancement of HIV_IIIB infection in TZM-bl cells in the presence of polymer 3 at 1.5, 4.4, 13.3, 40 and 120 μg/mL as estimated by cellular luciferase expression levels normalized to HIV infectivity in the absence of both SEVI and polymer 3. (B) Dose-response curve of data shown in A normalized to HIV infectivity in the presence of SEVI only. Fitting the curve gives an IC₅₀ value of 14 μg/mL, which is equivalent to 206 nM. (C) Graph showing reduction of SEVI-mediated enhancement of HIV_IIIB infection in TZM-bl cells in the presence of 1.5, 4.4, 13.3, 40 and 120 μg/mL BTA-containing nanoparticles (NPs) as estimated by cellular luciferase expression levels normalized to HIV infectivity in the absence of both SEVI amyloid fibrils and NPs. (D) Dose-response curve of data show in C normalized to HIV infectivity in the presence of SEVI only. Fitting the curve gives an IC₅₀ value of 4 μg/mL, which is equivalent to 59 nM polymer. (E) Graph showing the reduction of SEVI-mediated enhancement of HIV_IIIB infection in TZM-bl cells in the presence of 40 and 120 μg/mL polymer 5 (negative control) as estimated by cellular luciferase expression levels normalized to HIV infectivity in the absence of both SEVI amyloid fibrils and polymer 5. (F) Graph showing the toxicity of polymers 3 and 5 and NPs in 3EC1 ectocervical cells as estimated by fluorescence of Alamar Blue normalized to cells alone. Polymer 3 and NPs dosed at 7.5, 30, and 120 μg/mL and control polymer 5 dosed at 40 and 120 μg/mL. These results show no statistically significant change in cell viability at concentrations of polymers 3 and 5 and NPs under 40 μg/mL. Statistical analysis by ANOVA with Turkey’s post test. ** p< 0.001.

Figure 5.

Structure of a previously reported BTA-Pentamer 6²¹

Interestingly, the BTA-containing polymeric nanoparticles exhibited an IC₅₀ value of ~4 μg/mL (Figure 4 C,D), which corresponded to a concentration of 59 nM polymer. While the molar activity for these nanoparticles is the most active of any SEVI-binding species reported to date for reducing SEVI-enhanced HIV infection, the activity of these nanoparticles on a per BTA basis (IC₅₀ = 6 μM) is also twice as active as BTA-EG ²⁰₆ and over three times more active than polymer 3. We attribute the increased activity of these nanoparticles relative to polymer 3 to the increased uniformity of shape and size of the nanoparticles compared to a complex mixture of aggregates that likely form with polymer 3 in aqueous solutions. The more narrow dispersity of shape and size of the nanoparticles presumably results in a higher concentration of polymeric aggregates with consistent activity for binding to SEVI. The control polymer 5 was, as expected, completely ineffective at inhibiting SEVI-mediated HIV infection at concentrations below 40 μg/mL (~ 2 μM), emphasizing the importance of an amyloid-binding moiety on the polymer or nanoparticle for activity (Figure 4E).

In order to rule out that the observed activity of the polymers and nanoparticles reported here could be affected by unexpected toxicity of these materials to cells, we exposed ectocervical cells to high concentrations of 3, 5 and the BTA-containing polymeric nanoparticles for 24 h and measured cell viability using an Alamar Blue assay.²⁰ We found that none of the polymers or nanoparticles displayed significant toxicity at concentrations below 40 μg/mL (Figure 4F).

In order to evaluate the effect of sterics of SEVI-binding molecules on reduction of SEVI-mediated HIV infection, we compared the K_d values for binding to SEVI and IC₅₀ values for inhibiting SEVI-enhanced HIV infection of BTA-EG₆ (monomer, 1),²⁰ previously reported pentameric BTA molecule 6 (Figure 5),²¹ and polymer 3. Table 2 shows that the apparent K_d value of BTA-pentamer 6 to SEVI fibrils is >300-fold lower (i.e., stronger) than the K_d value of BTA-EG₆ 1. The IC₅₀ value for reducing SEVI-enhanced HIV infection, however, was only 65-fold lower for pentamer 6 compared to monomer 1. Comparing the ratio of K_d:IC₅₀ for BTA-EG₆ monomer (~1:100) and BTA-pentamer (~1:500), therefore, illustrates that significantly reducing the apparent K_d of the SEVI-binding group is not accompanied by a corresponding strong reduction in SEVI-mediated HIV infection. For polymer 3, the apparent K_d for binding to SEVI is slightly higher (i.e., less tightly bound) than the monomer, but polymer 3 exhibits an IC₅₀ that is 65-fold lower than monomer 1 for reducing SEVI-enhanced HIV infection. The ratio of K_d:IC₅₀ for polymer 3 is thus approximately 1:1 (Table 2). This drastic increase in ratio of K_d:IC₅₀ for polymer 3 compared to BTA-EG₆ monomer 1 or BTA-pentamer 6 suggests that the IC₅₀ values are not well correlated with the apparent K_d values for binding to SEVI. The approximate 1:1 ratio of K_d:IC₅₀ for polymer 3 suggests that the majority of polymer molecules that are bound to SEVI contribute to the inhibition of HIV infectivity enhanced by SEVI. In contrast, the monomeric- and pentameric-BTA compounds required a concentration for inhibiting SEVI-mediated HIV infection that is at least 100-fold higher than the concentration required for binding to the amyloid surface. These observations suggest that steric bulk on the amyloid-binding agent greatly improves efficiency for inhibiting SEVI-mediated HIV infection.

CONCLUSION

In conclusion, we developed a polymer comprising amyloid-binding BTA moieties that maintained good affinity for SEVI amyloid fibrils. Nanoparticles generated from this polymer were similar in size and shape to HIV virions, and effectively inhibited SEVI-mediated infectivity of HIV in cells. These polymers and nanoparticles were much more efficient at inhibiting SEVI-enhanced HIV infection compared to previously reported monomeric and pentameric versions of BTA, suggesting that steric hindrance can play a significant role in the design of effective SEVI neutralizing agents. These results further support the use of amyloid-targeting agents as prophylactic microbicide supplements for reducing HIV transmission through sexual contact. Since polymeric nanoparticles can be used as vehicles for controlled delivery of a variety of drugs,^29,32,33 the results reported here pave the way for development of multifunctional polymeric materials that can be targeted to amyloids associated with HIV transmission and possibly other diseases.^2,19,34,35

METHODS

Synthesis of BTA-EG₆-Acrylate (2)

In a flame-dried round bottom flask purged with N₂, BTA-EG₆ 1¹⁶ (500 mg, 0.99 mmol) was dissolved in 25 mL of dichloromethane (DCM). The reaction mixture was cooled to −78°C in a dry ice bath (acetone/CO₂) and triethylamine was added (138 μL, 0.99 mmol) over 5 min. After incubation for an additional 5 min, a solution of acryloyl chloride (67 μL, 0.83 mmol) in DCM (1 mL) was added dropwise. After warming to room temperature (r.t.) over 24 h, the reaction mixture was concentrated to dryness and purified via silica gel chromatography using ethyl acetate (EtOAc) as the eluent. The product was dried under vacuum to afford an orange oil (334 mg, 72% yield). ¹H-NMR (400 MHz, CDCl₃): δ 7.89 (3H, dd, J = 8.6, 10.1 Hz), 7.63 (1H, s), 7.23 (1H, d, J = 9 Hz), 6.66 (2H, d, J = 8.7 Hz), 6.40 (1H, dd, J = 1.9, 17 Hz), 6.11 (1H, dd, 10.1, 17 Hz), 5.82 (1H, dd, 1.9, 10 Hz), 4.30 (2H, t, J = 4.9 Hz), 3.67 (20H, m), 3.37 (2H, t, J = 4.7 Hz), 2.47 (3H, s). ¹³C NMR (125 MHz, CDCl₃) 167.85, 166.31, 152.47, 150.63, 134.71, 134.40, 131.10, 129.00, 128.23, 127.62, 122.69, 121.90, 121.32, 112.67, 70.62, 70.60, 70.58, 70.57, 70.56, 70.33, 69.29, 69.10, 63.81, 43.17, 21.62; ESI-MS (m/z) calculated for C₂₉H₃₈N₂O₇S [M] 558.24; found [M + H] 559.23

Synthesis of Poly[acrylic acid-co-benzothiazole aniline hexa(ethylene glycol) acrylate] (3)

Acrylic acid (32 μL, 0.47 mmol), 2 (263 mg, 0.47 mmol), and AIBN (4 mg, 0.024 mmol) were dissolved in butyl acetate (1 mL). The reaction mixture was degassed by gently bubbling dry N₂ gas with stirring for 30 min. After degassing, the reaction vial was sealed under N₂ atmosphere and placed in an oil bath at 65°C for 16 h. After cooling to r.t., the cloudy solution was placed in a 50 mL conical vial. The reaction vial was rinsed with DCM (2 mL) and added to the conical vial. The polymer was precipitated by addition of hexanes (30 mL) and the solution was decanted. The polymer was re-dissolved in DCM and again precipitated with hexanes. The product was dried under vacuum to afford a yellow solid (100 mg, 34% yield). ¹H-NMR (400 MHz, CDCl₃): δ 7.82 (3H, bs), 7.56 (1H, bs), 7.21 (1H, bs), 6.62 (2H, bs), 4.16 (2H, bs), 3.29 (20H, bs), 2.42 (3H, bs), 1.97 (1H, bs), 1.69 (2H, bs). Integration of aromatic protons in the region δ 6.62-7.82 ppm was used to determine incorporation of monomer 2 (see Supporting Information for details).

Synthesis of Hexaethylene Glycol Monomethyl Ether Acrylate (4)

In a flame-dried round bottom flask purged with N₂, hexaethylene glycol monomethyl ether (304 mg, 1.03 mmol) was dissolved in DCM (5 mL). The reaction mixture was cooled to −78°C in a dry ice bath (acetone/CO₂) and triethylamine was added (172 μL, 1.23 mmol) over 5 min. After another 5 min, a solution of acryloyl chloride (100 μL, 1.23 mmol) in DCM (1 mL) was added dropwise. After warming to r.t. over 24 h, the reaction mixture was concentrated to dry and purified via silica gel chromatography using EtOAc as the eluent. The product was dried under vacuum to afford an orange oil (200 mg, 56% yield). ¹H-NMR (500 MHz, CDCl₃): δ 6.43 (1H, dd, J = 1.4, 17.4 Hz), 6.15 (1H, dd, J = 10.4, 17.3 Hz), 5.84 (1H, dd, J = 1.4, 10.4 Hz), 4.31 (2H, t, J = 5.3 Hz), 3.74 (2H, t, J = 5 Hz), 3.65 (18H, m), 3.54 (2H, t, 4.5 Hz), 3.37 (3H, s). ¹³C NMR (125 MHz, CDCl₃) δ 166.33, 131.23, 128.37, 72.03, 70.73, 70.68, 70.63, 69.24, 63.84, 59.19; ESI-MS (m/z) calculated for C₁₆H₃₀O₈ [M] 350.19; found [M + H]⁺ 351.33, [M + NH₄]⁺ 368.32 and [M + Na]⁺ 373.36.

Synthesis of Poly(hexaethylene glycol monomethyl ether acrylate) (5)

Compound 6 (180 mg, 0.51 mmol) and AIBN (2 mg, 0.013 mmol) were dissolved in butyl acetate (1 mL). The reaction mixture was degassed by gently bubbling dry N₂ gas with stirring for 30 min. After degassing, the reaction vial was sealed under N₂ atmosphere and placed in an oil bath at 65°C for 16 h. After cooling to room temperature, the cloudy solution was placed in a 50 mL conical vial. The reaction vial was rinsed with DCM (2 mL) and added to the conical vial. The polymer was precipitated by addition of hexanes (30 mL) and the solution was decanted. The polymer was re-dissolved in DCM and again precipitated with hexanes. The product was dried under vacuum to afford a colorless oil. ¹H NMR (400 MHz, CDCl₃): δ 4.17 (2H, bs), 3.68 (22H, bm), 3.38 (3H, bs), 2.31 (1H, bs), 1.78 (2H, bs), 1.62 (1H, bs).

Polymer Analysis

Polymer polydispersity index (PDI) and molecular weight were determined by size-exclusion chromatography (Phenomenex Phenogel 5u 10, 1K - 75K, 300 × 7.80 mm in series with a Phenomenex Phenogel 5u 10, 10K – 1000K, 300 × 7.80 mm (0.05 M LiBr in DMF, 0.75 mL/min, 60°C)) using a Shimadzu LA-10AT pump equipped with a UV detector (Hitachi-Elite LaChrom L-2420), a multi-angle light scattering detector (DAWN-HELEOS, Wyatt Technology) and a refractive index detector (Hitachi L-2490). Data analysis was performed using the ASTRA software package.

Nanoparticle Formulation

We dissolved 2 mg of polymer 3 in 2 mL of tetrahydrofuran (THF). This solution was added dropwise to an equal volume of deionized (DI) water while stirring vigorously. After stirring for 30 min, the THF was removed in vacuo and the remaining 1 mg/mL nanoparticle solution was analyzed via dynamic light scattering.

Nanoparticle Analysis by Dynamic Light Scattering

Particle size determination by dynamic light scattering was performed on a Wyatt DynaPro NanoStar (Wyatt Technology, Santa Barbara, CA) instrument using a disposable cuvette (Eppendorf UVette 220 nm–1,600 nm). Samples were analyzed using a concentration of nanoparticles of 1 mg/mL (with respect to polymer concentration) and data processed using Wyatt DYNAMICS V7 software. Data exported for final plotting using GraphPad Prism 5 (GraphPad Software, Inc., La Jolla, CA) and a representative plot of signal intensity versus radius is shown in Figure S1 in the supporting information.

Nanoparticle and SEVI Analysis by Cryo-EM

Nanoparticles were re-suspended in a volume (50 μL) of water with a concentration of 20 mg/mL nanoparticle (with respect to polymer concentration) prior to imaging. For SEVI, the amyloid fibrils were suspended at a concentration of 10 mg/mL (with respect to peptide concentration) in PBS. Small aliquots (< 5 μL) of each sample were separately applied to holey grids, which had been glow discharged and plasma cleaned. These Cryo-EM samples were vitrified using liquid ethane as the cryogen. Frozen samples were transferred into a pre-cooled cryo-transfer holder to maintain low temperature. Images acquired on a FEI Tecnai G2 Sphera operated at 200 keV using a Gatan Ultrascan 1000 UHS 4 MP CCD camera.

Preparation of SEVI Fibrils

PAP248–286 was dissolved in PBS at a concentration of 10 mg/mL. Fibrils were formed by agitation in an Eppendorf Thermomixer at 1400 rpm and 37 °C for 72 h. The presence of fibrils was confirmed by a previously described Congo Red spectroscopic assay.^21,36

TEM Imaging of SEVI Fibrils Incubated with Nanoparticles (NP)

In order to obtain TEM images of SEVI fibrils incubated with NP, a 5 μL solution of NPs (5 mg/mL) of was incubated with a 2 μL solution of SEVI (2 mg/mL) in water over an air plasma oxidized EM grid. The mixture was agitated via pipette on the grid for approximately 5 min. After 5 min, excess water was wicked away using filter paper and the dried sample was subjected to TEM imaging on a FEI Tecnai G2 Sphera using a Gatan Ultrascan 1000 UHS 4 MP CCD camera.

Measurement of the Apparent Dissociation Constant, K_d, of (3) to SEVI Fibrils

The binding of polymer 3 to SEVI amyloid fibrils was estimated using a centrifugation assay described previously.^20,21 Briefly, 200 μL of various concentrations of 3 in 5 % DMSO/water were incubated in the presence or absence of 10 μg of SEVI to give a final volume of 201 μL of solution. These incubations were performed in duplicate runs and allowed to equilibrate for 12 h at room temperature. After equilibration, each solution was centrifuged at 16,000 g for 30 min at 4°C. The supernatants were separated from the pelleted fibrils, and 200 μL of fresh water was added to resuspend the pellets. 100 μL aliquots of each re-suspended pellet was pipetted into a cuvette (ultramicrocuvette,10-mm light path, Hellma®, Müllheim, Germany), and the fluorescence of the bound molecule was determined at 355 nm excitation and 420 nm emission using a spectrofluorometer (Photon Technology International, Inc., Birmingham, NJ). Each measurement experiment was repeated at least 3 times. Error bars represent standard deviations from the mean. The graph shown in Figure S3 in the supporting information shows the fluorescence intensity versus concentration of polymer 3; the data were fitted using the following one-site specific binding algorithm to determine K_d: Y = Bmax X/(K_d + X), where X is the concentration of 3, Y is the specific binding fluorescence intensity, and Bmax corresponds to the apparent maximal observable fluorescence upon binding of BTA polymer to SEVI fibrils. The data were processed using GraphPad Prism 5 (GraphPad Software, Inc., La Jolla, CA).

Evaluation of SEVI-Mediated Enhancement of HIV-1 Infectivity of TZM-bl Cells in the Presence of BTA Polymer and Nanoparticles

TZM-bl cells (in DMEM culture medium supplemented with 10% FBS, 50 units/mL penicillin, and 50 μg/mL streptomycin) were seeded on 96-well flatbottomed tissue culture plates at a density of 4×10³ cells/well. Plates were incubated for 12 h (in a humidified atmosphere of 95% air, 5% CO2 at 37°C) to promote attachment of cells to the wells. HIV_IIIB virions were then pretreated for 10 min at room temperature with 15 μg/mL SEVI fibrils in the presence or absence of polymer 3, 5, or polymeric nanoparticle. Treated virions were then added to the plated TZM-bl cells and were incubated for 2 h at 37°C. After incubation, the cells were washed with DPBS and the media was replaced. Infection was assayed after 72 hours by quantifying luciferase expression with PerkinElmer Britelite Plus and measuring luminescence with a microplate reader (DTX880, Beckman Coulter). All data are represented as the mean ± standard deviation of triplicate measurements. ANOVA with Tukey’s post test was employed in all analyses of data. A p-value < 0.05 was considered statistically significant.

Toxicity Assay

Cervical epithelial (ectocervical) cells (3EC1 cell line) were treated for 24 h with polymer 3, polymeric nanoparticles, and polymer 5 at concentrations up to 40 μg/mL. After 24 h cell viability was analyzed by measuring cellular metabolic activity using the resazurin cytotoxicity assay (Alamar Blue; Invitrogen), in accordance with the manufacturer’s protocol. Cells were also treated with 0.1% Nonoxynol-9 as a positive control for cytotoxicity.