Human Immunodeficiency Virus (HIV) is a sexually transmitted virus that can cause a serious disease that weakens the immune system. The virus is most commonly transmitted between individuals in semen, the male reproductive fluid. Semen contains deposits of protein fragments called amyloid fibrils, which can increase the transmission of HIV by trapping viral particles. This helps the virus to attach to the membranes surrounding human cells, which increases the risk of infection. Therefore, therapies that reduce the levels of amyloid fibrils in semen might be able to reduce the transmission of HIV.

Drugs that prevent amyloid formation are already being developed because structurally similar fibrils can also form in the brains of individuals with neurodegenerative diseases. One such molecule—called CLR01—works by binding to particular sites on the proteins that form fibrils in the brain. This inhibits fibril formation and slowly disassembles the fibrils that have already formed. CLR01 physically interacts with these residues in a way that resembles a tweezer.

The peptides in the amyloid fibrils in semen also have these sites, which suggests that CLR01 might also disrupt amyloid fibrils from forming in semen. Here Lump and Castellano et al. show that CLR01 can both disrupt fibril formation and remodel fibrils that have already formed. In addition, CLR01 prevents HIV particles from interacting with these fibrils and can displace the virus particles that have already bound to the fibrils. In the presence of CLR01, human cells exposed to semen that contained HIV were less likely to become infected with the virus.

Unexpectedly, CLR01 also directly destroys HIV and other enveloped viruses such as HCV or HSV particles by disrupting the membranes that surround the virus. Therefore, Lump and Castellano et al.'s findings reveal that CLR01 has considerable potential to be used as an agent for reducing the transmission of HIV and other sexually transmitted viral diseases.

The majority of new HIV-1 infections are transmitted via sexual intercourse, and semen is the main vector for viral spread. Far from being a passive vehicle, semen potently enhances HIV infectivity (Münch et al., 2007; Kim et al., 2010). This HIV-enhancing activity is attributed to seminal amyloid fibrils (Münch et al., 2007; Kim et al., 2010; Roan et al., 2011; Usmani et al., 2014) that form by self-assembly of proteolytic fragments of prostatic acid phosphatase (PAP248-286 and PAP85-120) and the homologous proteins semenogelin 1 (SEM1) and semenogelin 2 (SEM2) (Münch et al., 2007; Roan et al., 2011; Arnold et al., 2012; Castellano and Shorter, 2012; Münch et al., 2014). Seminal amyloid fibrils are highly cationic, and the positively charged fibrils capture HIV virions, increase viral attachment rates to target cells, and augment fusion (Roan et al., 2009; Arnold et al., 2012; Usmani et al., 2014). By doing so, fibrils promote HIV infection in vitro by several orders of magnitude, whereas the corresponding monomeric peptides have no effect (Münch et al., 2007; Roan et al., 2011; Arnold et al., 2012). Importantly, the stimulatory effect of seminal amyloid is the greatest at low virus concentrations (Münch et al., 2007), and semen and PAP248-286 fibrils (termed SEVI for Semen-derived Enhancer of Virus Infection) may facilitate vaginal virus transmission after exposure to low viral doses (Miller et al., 1994; Neildez et al., 1998; Münch et al., 2013). HIV transmission rates are relatively low, occurring as infrequently as 1 in 200 to as low as 1 in 10,000 coital acts (Gray et al., 2001). Thus, counteraction of infectivity promoting amyloids in semen should reduce or even prevent HIV transmission via the sexual route.

The lysine- and arginine-specific molecular tweezer, CLR01 (Figure 1A,B) (Fokkens et al., 2005; Klärner et al., 2006, 2010), inhibits amyloid fibrillization by engaging specific lysine, arginine, or both residues within a variety of disease-associated amyloidogenic proteins including amyloid-β protein (Aβ), tau, islet amyloid polypeptide, transthyretin, and α-synuclein (Sinha et al., 2011; Attar et al., 2012; Prabhudesai et al., 2012; Sinha et al., 2012; Acharya et al., 2014; Ferreira et al., 2014; Lopes et al., 2015; Zheng et al., 2015). Furthermore, CLR01 has even been found to slowly remodel preformed Aβ and α-synuclein fibrils over the course of several weeks (Sinha et al., 2011; Prabhudesai et al., 2012). CLR01 binds lysine residues with a K_d of ∼10 µM and also binds arginine residues, albeit with ∼10-fold lower affinity (Fokkens et al., 2005; Dutt et al., 2013). The unprecedented high specificity of CLR01 for basic amino acids relies on a unique binding mode in which the tweezer draws the cationic side chains into its torus-shaped cavity and engages the ammonium cation of lysine or the guanidinium cation of arginine with its anionic phosphate group in a tight ion pair (Figure 1B) (Klärner and Schrader, 2013). No other amino acids fulfill the requirements for this threading mechanism. The structure of the CLR01-lysine complex and the precise mechanism of lysine threading into the CLR01 guest cavity and subsequent ion pairing have been extensively characterized by NMR spectroscopy, crystal structure, molecular dynamics, and quantum mechanics/molecular mechanics (QM/MM) calculations (Bier et al., 2013; Dutt et al., 2013; Klärner and Schrader, 2013). Importantly, CLR01 appears only to complex with readily accessible lysine or arginine residues on protein surfaces, as evidenced by crystal structures and NMR experiments (Bier et al., 2013). This restriction makes CLR01 more selective for lysine or arginine residues found in intrinsically unfolded proteins or protein sequences.

Figure 1

Download asset Open asset

Since amyloidogenic seminal peptides are particularly rich in lysine and arginine residues (Roan et al., 2009; Arnold et al., 2012; Castellano and Shorter, 2012) (Figure 1C–E, Lys and Arg residues are highlighted in red), we hypothesized that CLR01 might interfere with their HIV-enhancing activity. Here, we establish that CLR01 inhibits amyloidogenesis of PAP and SEM peptides, neutralizes the cationic surface charge of seminal amyloid, and rapidly remodels preformed SEVI and PAP85-120 fibrils. Strikingly, CLR01 also exhibits a direct antiviral effect by selectively disrupting the membrane of enveloped viruses. Thus, CLR01 represents an unprecedented candidate for further development as a microbicide as it not only inactivates HIV and other enveloped viruses but also antagonizes host-encoded seminal amyloids that enhance viral infection.

Despite the development of several different classes of topical microbicides, none have proven safe and effective at HIV prevention. The failure of topical microbicide candidates in previous clinical trials has been attributed to lack of adherence (Van Damme et al., 2008; Marrazzo et al., 2015), adverse effects (McGowan et al., 2011), and a greatly diminished antiviral efficacy in the presence of semen (Neurath et al., 2006; Zirafi et al., 2014). This bodily fluid is not only the main vector for HIV transmission but also contains cationic amyloid fibrils that markedly increase viral infectivity (Castellano and Shorter, 2012; Münch et al., 2014). Various biological polyanions such as heparin or other glycosaminoglycans can prevent these cationic fibrils from enhancing HIV infection (Roan et al., 2009). Unfortunately, however, such anionic polymers have been unsuccessful in past clinical microbicide trials due to their poor bioavailability and induction of inflammatory responses in the genital tract, which actually augment HIV transmission by recruiting HIV-susceptible target cells to the genital mucosa (Lüscher-Mattli, 2000; van de Wijgert and Shattock, 2007). We have suggested that future microbicide endeavors should focus on agents that simultaneously and safely target HIV and the host factors that are exploited by the virus to facilitate its transmission (Castellano and Shorter, 2012; Zirafi et al., 2014; Roan and Münch, 2015). Here, we report that CLR01, a lysine- and arginine-specific molecular tweezer (Fokkens et al., 2005), not only counteracts the infection-enhancing activity of seminal amyloids and semen, but also directly destroys HIV virions (Figure 10). CLR01 is a highly promising topical microbicide candidate because it possesses potent antiviral and anti-amyloid activity, displays minimal toxicity in vivo (Prabhudesai et al., 2012; Attar et al., 2014; Ferreira et al., 2014), and is efficacious in human seminal fluid.

Figure 10

Download asset Open asset

It was surprising that CLR01 not only affected the formation and function of seminal amyloids but also displayed a broad and direct antiviral activity against HIV and other enveloped viruses (Figure 8). These diverse activities made us concerned that CLR01 might disrupt amyloidogenesis and viral infection via a non-specific mechanism involving the formation of colloidal CLR01 aggregates (McGovern et al., 2002, 2003; Shoichet, 2006; Feng et al., 2008). Nevertheless, multiple lines of evidence argue against a non-specific, colloidal mechanism of CLR01 activity. First, using a variety of biophysical techniques, we were unable to detect significant quantities of colloidal CLR01 aggregates under the conditions employed in our experiments (Figure 7). Second, the anti-amyloid and antiviral activity of CLR01 was unaffected by high concentrations of BSA (Figure 2—figure supplement 4; Figure 3—figure supplement 2; Figure 5—figure supplement 2 and 3; Figure 9A), which would adsorb and quench the activity of any potential colloidal micelles (McGovern et al., 2002; Feng et al., 2008). Third, attempts to preclear CLR01 solutions by centrifugation to remove any colloidal aggregates (McGovern et al., 2003; Feng et al., 2008) had no effect on CLR01 anti-amyloid or antiviral activity (Figure 2—figure supplement 4; Figure 3—figure supplement 2; Figure 5—figure supplement 4). Fourth, the shallow CLR01 dose–response curves are not typical of colloidal aggregate inhibitors, which typically exhibit steep dose–response curves (Shoichet, 2006). Finally, a feature of small molecules that form colloidal aggregates is that they are non-specific inhibitors (McGovern et al., 2002, 2003; Feng et al., 2008). By contrast, CLR01 displays specificity for amyloidogenesis by peptides that harbor lysine and arginine residues. Thus, if the lysine and arginine residues of PAP248-286 or PAP85-120 are replaced with alanine, then CLR01 can neither inhibit fibril assembly nor remodel preformed fibrils (Figures 2G–I, 3I–L). Likewise, the antiviral activity of CLR01 is restricted to enveloped viruses and CLR01 is ineffective against non-enveloped viruses (Figure 8). This specificity is inconsistent with an indiscriminate, non-specific mechanism of action mediated by colloidal CLR01 aggregates. Importantly, the anti-amyloid and antiviral activity of CLR01 could be eliminated by excess lysine or poly-L-lysine. Thus, the lysine-binding cavity of CLR01 plays an important role in both anti-amyloid and antiviral activity.

The ability of CLR01 to antagonize seminal amyloid in the presence of BSA and to prevent HIV infection in the presence of seminal fluid (Figure 9) raises another question. Why is CLR01 activity not quenched by lysine or arginine residues on the surface of BSA or on other proteins abundant in seminal fluid? Here, it is important to consider the profound difference between freely accessible lysines on intrinsically unstructured domains, which can all be complexed by the tweezers, and lysines in folded proteins whose surface often prohibits the sterically demanding tweezer molecule to approach a lysine residue close enough for the threading mechanism (Figure 1B). The interaction between free lysine and CLR01 has a K_d of ∼10 µM, with fast on and off rates (Fokkens et al., 2005; Bier et al., 2013; Dutt et al., 2013). On folded proteins, however, only a few lysines are accessible for the relatively large molecular skeleton of the tweezers. For example, only 5 of 17 surface lysines in a 14-3-3 protein are most likely to be complexed by CLR01 (Bier et al., 2013). Even in an unstructured small protein, such as Aβ, CLR01 has clear preference for the free Lys16 over Lys28 (Sinha et al., 2011) because the latter is engaged in stabilizing hydrophobic interactions and salt bridges (Petkova et al., 2002; Lazo et al., 2005; Xiao et al., 2015). Moreover, because of the high on and off rates, CLR01 is expected to only disrupt relatively weak interactions as well as events that depend on critical and accessible lysine or arginine residues. Therefore, the tweezer has shown selectivity in multiple cases that on first impression may appear counter-intuitive. For example, inhibition of abnormal protein aggregation is typically achieved at ∼1:1–1:5 protein:CLR01 concentration (Sinha et al., 2011, 2012; Acharya et al., 2014), whereas disruption of controlled protein polymerization, for example, of tubulin, requires ∼55-fold excess CLR01 (Attar et al., 2014) and enzyme inhibition requires ∼1000-fold excess CLR01 (Talbiersky et al., 2008). Moreover, in vivo, CLR01 shows no toxicity at doses 2–3 orders of magnitude above those needed to enable clearance of amyloid (Attar et al., 2012; Prabhudesai et al., 2012; Attar et al., 2014; Ferreira et al., 2014). Thus, selectivity for lysine-bearing amyloid conformers in complex biological fluids is achieved by CLR01.

We suggest that the lysine-rich nature of seminal amyloid peptides (Figure 1C–E) makes them unusually sensitive to CLR01 activity. Indeed, substoichiometric CLR01 concentrations (relative to peptide monomers) suffice to inhibit PAP248-286, PAP85-120, and SEM1(45-107) fibrillization (Figure 2D–F). Remarkably, substoichiometric CLR01 concentrations (relative to peptide monomers) also sufficed to effectively remodel SEVI and PAP85-120 fibrils (Figure 3G,H). Curiously, SEM1(45-107) fibrils could not be remodeled by CLR01. Why SEM1(45-107) fibrils were refractory to CLR01 remains unclear and requires further study. CLR01 may require longer time periods to remodel SEM1(45-107) fibrils as with remodeling of Aβ fibrils, which required several weeks (Sinha et al., 2011). It will also be important to determine whether some lysine and arginine residues in PAP248-286, PAP85-120, and SEM1(45-107) are more critical than others for CLR01 activity.

How does CLR01 directly antagonize viral infection? Since the CLR03 control, which has a similar negative charge as CLR01 (Sinha et al., 2011), was entirely ineffective at blocking viral infection, a purely polyanion-like mechanism that prevents the interaction of virions and cells by increasing charge repulsions (Neurath et al., 2002; Weber et al., 2005) could be excluded. Instead, we demonstrate that CLR01 selectively disrupts membranes containing elevated levels of sphingomyelin and cholesterol (Figure 6). Envelopes of HIV-1, herpes viruses, and HCV differ significantly from the cellular plasma membrane as they are more highly enriched in these and other lipids (van Genderen et al., 1994; Brügger et al., 2006; Chan et al., 2008; Lorizate et al., 2009, 2013; Merz et al., 2011). Indeed, the HIV membrane possesses a highly unusual lipid composition, which is distinct even from the detergent-resistant microdomains typically found in the plasma membrane, indicating that HIV budding involves a specific lipid raft clustering process (Brügger et al., 2006). For example, although HIV membrane resembles lipid raft microdomains in that it is enriched for saturated lipids, phosphatidylserine, plasmalogen-phosphatidylethanolamine, cholesterol, and sphingolipids, it is also enriched for the unusual sphingolipid dihydrosphingomyelin (Brügger et al., 2006; Lorizate et al., 2009, 2013). The unique lipid composition of viral membranes may render them uniquely sensitive to CLR01.

Importantly, unlike various non-ionic and anionic surfactants, CLR01 has minimal effects on cell viability (Figure 4—figure supplement 3). Thus, CLR01 might leverage a therapeutic opportunity provided by the physiological difference between static viral membranes and biogenic cellular membranes capable of self-repair. Static viral membranes are unable to withstand disruption by CLR01, whereas cells can likely repair lipid-raft-rich regions of the plasma membrane that might be disrupted by CLR01. In this regard, CLR01 may possess similarities to other broad-spectrum antivirals such as LJ001, a lipophilic aryl methyldiene rhodanine derivative, which selectively disrupt diverse viral membranes but are ineffective against non-enveloped viruses, and also do not disrupt cell membranes (Wojcechowskyj and Doms, 2010; Wolf et al., 2010; Vigant et al., 2013). We investigated whether the tweezer might recognize the charged lipid head groups but found via NMR chemical-shift experiments that this was not the case (data not shown). However, experiments on various phase boundaries strongly indicated that the tweezer could migrate into lipid bilayers due to its amphiphilic character (data not shown). A pronounced destabilizing effect may be triggered if this happens at the edge of lipid rafts that are enriched in virion membranes. Importantly, HSV-2, HCMV, and HCV can also be transmitted via sexual intercourse (Handsfield et al., 1985; Rapp, 1989; Tedder et al., 1991). Thus, a CLR01-based microbicide may not only protect from HIV-1 acquisition but also from other major human pathogens.

CLR01 binding to lysine residues could potentially disrupt normal protein function leading to toxicity or side effects. However, CLR01 showed no signs of cytotoxicity in this study and has previously been safely employed in multiple cell and animal models (Sinha et al., 2011; Attar et al., 2012; Prabhudesai et al., 2012; Attar et al., 2014). For example, transgenic mice that were given 10 mg CLR01 a day per 1 kg body weight intraperitoneally for 30 days showed no apparent toxicity or adverse effects (Attar et al., 2014). Moreover, it is conceivable that the application of CLR01 on mucosal surfaces should be even safer compared to its systemic administration.

Unlike previous microbicide candidates that block HIV infection by a single mechanism, a CLR01-based formulation would interfere with various essential steps in sexual HIV transmission (Figure 10). On a minute time frame, the tweezer destroys infectious HIV particles that are present in semen, prevents the formation of amyloid–virus complexes by neutralization of the fibril surface charge, and also displaces pre-bound virions from fibrils (Figure 10). On a longer time frame (hours), CLR01 not only prevents the formation of infection-enhancing seminal amyloids upon ejaculation but also remodels fibrils that are already abundant in semen after ejaculation or formed during semen liquefaction (Figure 10). Given the time frame of these effects, we suggest that the primary mechanism of CLR01 action is via its direct effect on the virus, and that the anti-amyloid effects, although potent, play a secondary role. Regardless, these combined antiviral and anti-amyloid activities and the encouraging safety data make CLR01 a promising broad-spectrum, topical microbicide against HIV-1 and other sexually transmitted viruses. We anticipate that large-scale synthesis of CLR01 could be relatively inexpensive (less than $1 per mg), which would enable facile development as a broad-spectrum antiviral microbicide.

1. 1. Acharya S
   2. Safaie BM
   3. Wongkongkathep P
   4. Ivanova MI
   5. Attar A
   6. Klarner FG
   7. Schrader T
   8. Loo JA
   9. Bitan G
   10. Lapidus LJ

   (2014) Molecular basis for preventing alpha-synuclein aggregation by a molecular tweezer

   The Journal of Biological Chemistry 289:10727–10737.

   https://doi.org/10.1074/jbc.M113.524520

   • Google Scholar
2. 1. Aloia RC
   2. Jensen FC
   3. Curtain CC
   4. Mobley PW
   5. Gordon LM

   (1988) Lipid composition and fluidity of the human immunodeficiency virus

   Proceedings of the National Academy of Sciences of USA 85:900–904.

   https://doi.org/10.1073/pnas.85.3.900

   • Google Scholar
3. 1. Arnold F
   2. Schnell J
   3. Zirafi O
   4. Stürzel C
   5. Meier C
   6. Weil T
   7. Ständker L
   8. Forssmann WG
   9. Roan NR
   10. Greene WC
   11. Kirchhoff F
   12. Münch J

   (2012) Naturally occurring fragments from two distinct regions of the prostatic acid phosphatase form amyloidogenic enhancers of HIV infection

   Journal of Virology 86:1244–1249.

   https://doi.org/10.1128/JVI.06121-11

   • Google Scholar
4. 1. Attar A
   2. Chan WT
   3. Klarner FG
   4. Schrader T
   5. Bitan G

   (2014) Safety and pharmacological characterization of the molecular tweezer CLR01-a broad-spectrum inhibitor of amyloid proteins' toxicity

   BMC Pharmacology & Toxicology 15:23.

   https://doi.org/10.1186/2050-6511-15-23

   • Google Scholar
5. 1. Attar A
   2. Ripoli C
   3. Riccardi E
   4. Maiti P
   5. Li Puma DD
   6. Liu T
   7. Hayes J
   8. Jones MR
   9. Lichti-Kaiser K
   10. Yang F
   11. Gale GD
   12. Tseng CH
   13. Tan M
   14. Xie CW
   15. Straudinger JL
   16. Klärner FG
   17. Schrader T
   18. Frautschy SA
   19. Grassi C
   20. Bitan G

   (2012) Protection of primary neurons and mouse brain from Alzheimer's pathology by molecular tweezers

   Brain 135:3735–3748.

   https://doi.org/10.1093/brain/aws289

   • Google Scholar
6. 1. Bier D
   2. Rose R
   3. Bravo-Rodriguez K
   4. Bartel M
   5. Ramirez-Anguita JM
   6. Dutt S
   7. Wilch C
   8. Klärner FG
   9. Sanchez-Garcia E
   10. Schrader T
   11. Ottmann C

   (2013) Molecular tweezers modulate 14-3-3 protein-protein interactions

   Nature Chemistry 5:234–239.

   https://doi.org/10.1038/nchem.1570

   • Google Scholar
7. 1. Bjelkmar P
   2. Larsson P
   3. Cuendet MA
   4. Hess B
   5. Lindahl E

   (2010) Implementation of the CHARMM force field in GROMACS: analysis of protein stability effects from correction maps, virtual interaction sites, and water models

   Journal of Chemical Theory and Computation 6:459–466.

   https://doi.org/10.1021/ct900549r

   • Google Scholar
8. 1. Brügger B
   2. Glass B
   3. Haberkant P
   4. Leibrecht I
   5. Wieland FT
   6. Krausslich HG

   (2006) The HIV lipidome: a raft with an unusual composition

   Proceedings of the National Academy of Sciences of USA 103:2641–2646.

   https://doi.org/10.1073/pnas.0511136103

   • Google Scholar
9. 1. Castellano LM
   2. Shorter J

   (2012) The surprising role of amyloid fibrils in HIV infection

   Biology 1:58–80.

   https://doi.org/10.3390/biology1010058

   • Google Scholar
10. 1. Chan R
    2. Uchil PD
    3. Jin J
    4. Shui G
    5. Ott DE
    6. Mothes W
    7. Wenk MR

    (2008) Retroviruses human immunodeficiency virus and murine leukemia virus are enriched in phosphoinositides

    Journal of Virology 82:11228–11238.

    https://doi.org/10.1128/JVI.00981-08

    • Google Scholar
11. 1. Coan KE
    2. Shoichet BK

    (2007) Stability and equilibria of promiscuous aggregates in high protein milieus

    Molecular Biosystems 3:208–213.

    https://doi.org/10.1039/b616314a

    • Google Scholar
12. 1. Daura X
    2. Gademann K
    3. Jaun B
    4. Seebach D
    5. van Gunsteren WF
    6. Mark AE

    (1999)

    Peptide folding: when simulation meets experiment

    Angewandte Chemie International Edition 38:236–240.

    • Google Scholar
13. 1. Dutt S
    2. Wilch C
    3. Gersthagen T
    4. Talbiersky P
    5. Bravo-Rodriguez K
    6. Hanni M
    7. Sanchez-Garcia E
    8. Ochsenfeld C
    9. Klarner FG
    10. Schrader T

    (2013) Molecular tweezers with varying anions: a comparative study

    The Journal of Organic Chemistry 78:6721–6734.

    https://doi.org/10.1021/jo4009673

    • Google Scholar
14. 1. Feng BY
    2. Toyama BH
    3. Wille H
    4. Colby DW
    5. Collins SR
    6. May BC
    7. Prusiner SB
    8. Weissman J
    9. Shoichet BK

    (2008) Small-molecule aggregates inhibit amyloid polymerization

    Nature Chemical Biology 4:197–199.

    https://doi.org/10.1038/nchembio.65

    • Google Scholar
15. 1. Ferreira N
    2. Pereira-Henriques A
    3. Attar A
    4. Klarner FG
    5. Schrader T
    6. Bitan G
    7. Gales L
    8. Saraiva MJ
    9. Almeida MR

    (2014) Molecular tweezers targeting transthyretin amyloidosis

    Neurotherapeutics 11:450–461.

    https://doi.org/10.1007/s13311-013-0256-8

    • Google Scholar
16. 1. Fokkens M
    2. Schrader T
    3. Klarner FG

    (2005) A molecular tweezer for lysine and arginine

    Journal of the American Chemical Society 127:14415–14421.

    https://doi.org/10.1021/ja052806a

    • Google Scholar
17. 1. French KC
    2. Makhatadze GI

    (2012) Core sequence of PAPf39 amyloid fibrils and mechanism of pH-dependent fibril formation: the role of monomer conformation

    Biochemistry 51:10127–10136.

    https://doi.org/10.1021/bi301406d

    • Google Scholar
18. 1. Frohm B
    2. DeNizio JE
    3. Lee DS
    4. Gentile L
    5. Olsson U
    6. Malm J
    7. Akerfeldt KS
    8. Linse S

    (2015) A peptide from human semenogelin I self-assembles into a pH-responsive hydrogel

    Soft Matter 11:414–421.

    https://doi.org/10.1039/c4sm01793e

    • Google Scholar
19. 1. Gerl MJ
    2. Sampaio JL
    3. Urban S
    4. Kalvodova L
    5. Verbavatz JM
    6. Binnington B
    7. Lindemann D
    8. Lingwood CA
    9. Shevchenko A
    10. Schroeder C
    11. Simons K

    (2012) Quantitative analysis of the lipidomes of the influenza virus envelope and MDCK cell apical membrane

    The Journal of Cell Biology 196:213–221.

    https://doi.org/10.1083/jcb.201108175

    • Google Scholar
20. 1. Goldschmidt L
    2. Teng PK
    3. Riek R
    4. Eisenberg D

    (2010) Identifying the amylome, proteins capable of forming amyloid-like fibrils

    Proceedings of the National Academy of Sciences of USA 107:3487–3492.

    https://doi.org/10.1073/pnas.0915166107

    • Google Scholar
21. 1. Gray RH
    2. Wawer MJ
    3. Brookmeyer R
    4. Sewankambo NK
    5. Serwadda D
    6. Wabwire-Mangen F
    7. Lutalo T
    8. Li X
    9. vanCott T
    10. Quinn TC
    11. Rakai Project Team

    (2001) Probability of HIV-1 transmission per coital act in monogamous, heterosexual, HIV-1-discordant couples in Rakai, Uganda

    Lancet 357:1149–1153.

    https://doi.org/10.1016/S0140-6736(00)04331-2

    • Google Scholar
22. 1. Haid S
    2. Novodomská A
    3. Gentzsch J
    4. Grethe C
    5. Geuenich S
    6. Bankwitz D
    7. Chhatwal P
    8. Jannack B
    9. Hennebelle T
    10. Bailleul F
    11. Keppler OT
    12. Poenisch M
    13. Bartenschlager R
    14. Hernandez C
    15. Lemasson M
    16. Rosenberg AR
    17. Wong-Staal F
    18. Davioud-Charvet E
    19. Pietschmann T

    (2012) A plant-derived flavonoid inhibits entry of all HCV genotypes into human hepatocytes

    Gastroenterology 143:213–222. e5.

    https://doi.org/10.1053/j.gastro.2012.03.036

    • Google Scholar
23. 1. Handsfield HH
    2. Chandler SH
    3. Caine VA
    4. Meyers JD
    5. Corey L
    6. Medeiros E
    7. McDougall JK

    (1985) Cytomegalovirus infection in sex partners: evidence for sexual transmission

    The Journal of Infectious Diseases 151:344–348.

    https://doi.org/10.1093/infdis/151.2.344

    • Google Scholar
24. 1. Hauber I
    2. Hohenberg H
    3. Holstermann B
    4. Hunstein W
    5. Hauber J

    (2009) The main green tea polyphenol epigallocatechin-3-gallate counteracts semen-mediated enhancement of HIV infection

    Proceedings of the National Academy of Sciences of USA 106:9033–9038.

    https://doi.org/10.1073/pnas.0811827106

    • Google Scholar
25. 1. Herzog G
    2. Shmueli MD
    3. Levy L
    4. Engel L
    5. Gazit E
    6. Klarner FG
    7. Schrader T
    8. Bitan G
    9. Segal D

    (2015) The Lys-specific molecular tweezer, CLR01, modulates aggregation of the mutant p53 DNA binding domain and inhibits its toxicity

    Biochemistry.

    https://doi.org/10.1021/bi501092p

    • Google Scholar
26. 1. Hess B
    2. Kutzner C
    3. van der Spoel D
    4. Lindahl E

    (2008) GROMACS 4: Algorithms for highly efficient, load-balanced, and scalable molecular simulation

    Journal of Chemical Theory and Computation 4:435–447.

    https://doi.org/10.1021/ct700301q

    • Google Scholar
27. 1. Jorgensen WL
    2. Chandrasekhar J
    3. Madura JD
    4. Impey RW
    5. Klein ML

    (1983) Comparison of simple potential functions for simulating liquid water

    The Journal of Chemical Physics 79:926–935.

    https://doi.org/10.1063/1.445869

    • Google Scholar
28. 1. Kim KA
    2. Yolamanova M
    3. Zirafi O
    4. Roan NR
    5. Staendker L
    6. Forssmann WG
    7. Burgener A
    8. Dejucq-Rainsford N
    9. Hahn BH
    10. Shaw GM
    11. Greene WC
    12. Kirchhoff F
    13. Münch J

    (2010) Semen-mediated enhancement of HIV infection is donor-dependent and correlates with the levels of SEVI

    Retrovirology 7:55.

    https://doi.org/10.1186/1742-4690-7-55

    • Google Scholar
29. 1. Klärner FG
    2. Kahlert B
    3. Nellesen A
    4. Zienau J
    5. Ochsenfeld C
    6. Schrader T

    (2006) Molecular tweezer and clip in aqueous solution: unexpected self-assembly, powerful host-guest complex formation, quantum chemical 1H NMR shift calculation

    Journal of the American Chemical Society 128:4831–4841.

    https://doi.org/10.1021/ja058410g

    • Google Scholar
30. 1. Klärner FG
    2. Schrader T

    (2013) Aromatic interactions by molecular tweezers and clips in chemical and biological systems

    Accounts of Chemical Research 46:967–978.

    https://doi.org/10.1021/ar300061c

    • Google Scholar
31. 1. Klärner FG
    2. Schrader T
    3. Polkowska J
    4. Bastkowski F
    5. Talbiersky P
    6. Campana Kuchenbrandt M
    7. Schaller T
    8. de Groot H
    9. Kirsch M

    (2010)

    Effect of molecular clips and tweezers on enzymatic reactions by binding coenzymes and basic amino acids

    Pure and Applied Chemistry 82:991–999.

    • Google Scholar
32. 1. Koutsoudakis G
    2. Kaul A
    3. Steinmann E
    4. Kallis S
    5. Lohmann V
    6. Pietschmann T
    7. Bartenschlager R

    (2006) Characterization of the early steps of hepatitis C virus infection by using luciferase reporter viruses

    Journal of Virology 80:5308–5320.

    https://doi.org/10.1128/JVI.02460-05

    • Google Scholar
33. 1. Kreppel F
    2. Biermann V
    3. Kochanek S
    4. Schiedner G

    (2002) A DNA-based method to assay total and infectious particle contents and helper virus contamination in high-capacity adenoviral vector preparations

    Human Gene Therapy 13:1151–1156.

    https://doi.org/10.1089/104303402320138934

    • Google Scholar
34. 1. Lazo ND
    2. Grant MA
    3. Condron MC
    4. Rigby AC
    5. Teplow DB

    (2005) On the nucleation of amyloid beta-protein monomer folding

    Protein Science 14:1581–1596.

    https://doi.org/10.1110/ps.041292205

    • Google Scholar
35. 1. Lindenbach BD
    2. Evans MJ
    3. Syder AJ
    4. Wölk B
    5. Tellinghuisen TL
    6. Liu CC
    7. Maruyama T
    8. Hynes RO
    9. Burton DR
    10. McKeating JA
    11. Rice CM

    (2005) Complete replication of hepatitis C virus in cell culture

    Science 309:623–626.

    https://doi.org/10.1126/science.1114016

    • Google Scholar
36. 1. Lockhart A
    2. Ye L
    3. Judd DB
    4. Merritt AT
    5. Lowe PN
    6. Morgenstern JL
    7. Hong G
    8. Gee AD
    9. Brown J

    (2005) Evidence for the presence of three distinct binding sites for the thioflavin T class of Alzheimer's disease PET imaging agents on beta-amyloid peptide fibrils

    The Journal of Biological Chemistry 280:7677–7684.

    https://doi.org/10.1074/jbc.M412056200

    • Google Scholar
37. 1. Lopes DH
    2. Attar A
    3. Nair G
    4. Hayden EY
    5. Du Z
    6. McDaniel K
    7. Dutt S
    8. Bravo-Rodriguez K
    9. Mittal S
    10. Klärner FG
    11. Wang C
    12. Sanchez-Garcia E
    13. Schrader T
    14. Bitan G

    (2015) Molecular tweezers inhibit islet amyloid polypeptide assembly and toxicity by a new mechanism

    ACS Chemical Biology.

    https://doi.org/10.1021/acschembio.5b00146

    • Google Scholar
38. 1. Lorizate M
    2. Brügger B
    3. Akiyama H
    4. Glass B
    5. Muller B
    6. Anderluh G
    7. Wieland FT
    8. Krausslich HG

    (2009) Probing HIV-1 membrane liquid order by Laurdan staining reveals producer cell-dependent differences

    The Journal of Biological Chemistry 284:22238–22247.

    https://doi.org/10.1074/jbc.M109.029256

    • Google Scholar
39. 1. Lorizate M
    2. Sachsenheimer T
    3. Glass B
    4. Habermann A
    5. Gerl MJ
    6. Krausslich HG
    7. Brügger B

    (2013) Comparative lipidomics analysis of HIV-1 particles and their producer cell membrane in different cell lines

    Cellular Microbiology 15:292–304.

    https://doi.org/10.1111/cmi.12101

    • Google Scholar
40. 1. Lüscher-Mattli M

    (2000) Polyanions—a lost chance in the fight against HIV and other virus diseases?

    Antiviral Chemistry & Chemotherapy 11:249–259.

    https://doi.org/10.1177/095632020001100401

    • Google Scholar
41. 1. Mackerell AD Jr
    2. Feig M
    3. Brooks CL III

    (2004) Extending the treatment of backbone energetics in protein force fields: limitations of gas-phase quantum mechanics in reproducing protein conformational distributions in molecular dynamics simulations

    Journal of Computational Chemistry 25:1400–1415.

    https://doi.org/10.1002/jcc.20065

    • Google Scholar
42. 1. Marrazzo JM
    2. Ramjee G
    3. Richardson BA
    4. Gomez K
    5. Mgodi N
    6. Nair G
    7. Palanee T
    8. Nakabiito C
    9. van der Straten A
    10. Noguchi L
    11. Hendrix CW
    12. Dai JY
    13. Ganesh S
    14. Mkhize B
    15. Taljaard M
    16. Parikh UM
    17. Piper J
    18. Mâsse B
    19. Grossman C
    20. Rooney J
    21. Schwartz JL
    22. Watts H
    23. Marzinke MA
    24. Hillier SL
    25. McGowan IM
    26. Chirenje ZM
    27. VOICE Study Team

    (2015) Tenofovir-based preexposure prophylaxis for HIV infection among African women

    The New England Journal of Medicine 372:509–518.

    https://doi.org/10.1056/NEJMoa1402269

    • Google Scholar
43. 1. McGovern SL
    2. Caselli E
    3. Grigorieff N
    4. Shoichet BK

    (2002) A common mechanism underlying promiscuous inhibitors from virtual and high-throughput screening

    Journal of Medicinal Chemistry 45:1712–1722.

    https://doi.org/10.1021/jm010533y

    • Google Scholar
44. 1. McGovern SL
    2. Helfand BT
    3. Feng B
    4. Shoichet BK

    (2003) A specific mechanism of nonspecific inhibition

    Journal of Medicinal Chemistry 46:4265–4272.

    https://doi.org/10.1021/jm030266r

    • Google Scholar
45. 1. McGowan I
    2. Gomez K
    3. Bruder K
    4. Febo I
    5. Chen BA
    6. Richardson BA
    7. Husnik M
    8. Livant E
    9. Price C
    10. Jacobson C
    11. MTN-004 Protocol Team

    (2011) Phase 1 randomized trial of the vaginal safety and acceptability of SPL7013 gel (VivaGel) in sexually active young women (MTN-004)

    AIDS 25:1057–1064.

    https://doi.org/10.1097/QAD.0b013e328346bd3e

    • Google Scholar
46. 1. Merz A
    2. Long G
    3. Hiet MS
    4. Brügger B
    5. Chlanda P
    6. Andre P
    7. Wieland F
    8. Krijnse-Locker J
    9. Bartenschlager R

    (2011) Biochemical and morphological properties of hepatitis C virus particles and determination of their lipidome

    The Journal of Biological Chemistry 286:3018–3032.

    https://doi.org/10.1074/jbc.M110.175018

    • Google Scholar
47. 1. Miller CJ
    2. Marthas M
    3. Torten J
    4. Alexander NJ
    5. Moore JP
    6. Doncel GF
    7. Hendrickx AG

    (1994)

    Intravaginal inoculation of rhesus macaques with cell-free simian immunodeficiency virus results in persistent or transient viremia

    Journal of Virology 68:6391–6400.

    • Google Scholar
48. 1. Müller JA
    2. Münch J

    (2016) Reporter assay for semen-mediated enhancement of HIV-1 infection

    Bio-protocol 6:e1871.

    https://doi.org/10.21769/BioProtoc.1871

    • Google Scholar
49. 1. Münch J
    2. Rücker E
    3. Ständker L
    4. Adermann K
    5. Goffinet C
    6. Schindler M
    7. Wildum S
    8. Chinnadurai R
    9. Rajan D
    10. Specht A
    11. Giménez-Gallego G
    12. Sánchez PC
    13. Fowler DM
    14. Koulov A
    15. Kelly JW
    16. Mothes W
    17. Grivel JC
    18. Margolis L
    19. Keppler OT
    20. Forssmann WG
    21. Kirchhoff F

    (2007) Semen-derived amyloid fibrils drastically enhance HIV infection

    Cell 131:1059–1071.

    https://doi.org/10.1016/j.cell.2007.10.014

    • Google Scholar
50. 1. Münch J
    2. Sauermann U
    3. Yolamanova M
    4. Raue K
    5. Stahl-Hennig C
    6. Kirchhoff F

    (2013) Effect of semen and seminal amyloid on vaginal transmission of simian immunodeficiency virus

    Retrovirology 10:148.

    https://doi.org/10.1186/1742-4690-10-148

    • Google Scholar
51. 1. Münch J
    2. Standker L
    3. Forssmann WG
    4. Kirchhoff F

    (2014) Discovery of modulators of HIV-1 infection from the human peptidome

    Nature Reviews. Microbiology 12:715–722.

    https://doi.org/10.1038/nrmicro3312

    • Google Scholar
52. 1. Nanga RP
    2. Brender JR
    3. Vivekanandan S
    4. Popovych N
    5. Ramamoorthy A

    (2009) NMR structure in a membrane environment reveals putative amyloidogenic regions of the SEVI precursor peptide PAP(248–286)

    Journal of the American Chemical Society 131:17972–17979.

    https://doi.org/10.1021/ja908170s

    • Google Scholar
53. 1. Neildez O
    2. Le Grand R
    3. Cheret A
    4. Caufour P
    5. Vaslin B
    6. Matheux F
    7. Theodoro F
    8. Roques P
    9. Dormont D

    (1998) Variation in virological parameters and antibody responses in macaques after atraumatic vaginal exposure to a pathogenic primary isolate of SIVmac251

    Research in Virology 149:53–68.

    https://doi.org/10.1016/S0923-2516(97)86900-2

    • Google Scholar
54. 1. Nelson R
    2. Sawaya MR
    3. Balbirnie M
    4. Madsen AO
    5. Riekel C
    6. Grothe R
    7. Eisenberg D

    (2005) Structure of the cross-beta spine of amyloid-like fibrils

    Nature 435:773–778.

    https://doi.org/10.1038/nature03680

    • Google Scholar
55. 1. Neurath AR
    2. Strick N
    3. Li YY

    (2002) Anti-HIV-1 activity of anionic polymers: a comparative study of candidate microbicides

    BMC Infectious Diseases 2:27.

    https://doi.org/10.1186/1471-2334-2-27

    • Google Scholar
56. 1. Neurath AR
    2. Strick N
    3. Li YY

    (2006) Role of seminal plasma in the anti-HIV-1 activity of candidate microbicides

    BMC Infectious Diseases 6:150.

    https://doi.org/10.1186/1471-2334-6-150

    • Google Scholar
57. 1. Ochsenbauer C
    2. Edmonds TG
    3. Ding H
    4. Keele BF
    5. Decker J
    6. Salazar MG
    7. Salazar-Gonzalez JF
    8. Shattock R
    9. Haynes BF
    10. Shaw GM
    11. Hahn BH
    12. Kappes JC

    (2012) Generation of transmitted/founder HIV-1 infectious molecular clones and characterization of their replication capacity in CD4 T lymphocytes and monocyte-derived macrophages

    Journal of Virology 86:2715–2728.

    https://doi.org/10.1128/JVI.06157-11

    • Google Scholar
58. 1. Okabe T
    2. Kawata M
    3. Okamoto Y
    4. Mikami M

    (2001) Replica-exchange Monte Carlo method for the isobaric–isothermal ensemble

    The Journal of Chemical Physics 335:435–439.

    https://doi.org/10.1016/S0009-2614(01)00055-0

    • Google Scholar
59. 1. Olsen JS
    2. DiMaio JT
    3. Doran TM
    4. Brown C
    5. Nilsson BL
    6. Dewhurst S

    (2012) Seminal plasma accelerates semen-derived enhancer of viral infection (SEVI) fibril formation by the prostatic acid phosphatase (PAP248-286) peptide

    The Journal of Biological Chemistry 287:11842–11849.

    https://doi.org/10.1074/jbc.M111.314336

    • Google Scholar
60. 1. Owen DH
    2. Katz DF

    (2005) A review of the physical and chemical properties of human semen and the formulation of a semen simulant

    Journal of Andrology 26:459–469.

    https://doi.org/10.2164/jandrol.04104

    • Google Scholar
61. 1. Papkalla A
    2. Münch J
    3. Otto C
    4. Kirchhoff F

    (2002) Nef enhances human immunodeficiency virus type 1 infectivity and replication independently of viral coreceptor tropism

    Journal of Virology 76:8455–8459.

    https://doi.org/10.1128/JVI.76.16.8455-8459.2002

    • Google Scholar
62. 1. Patriksson A
    2. van der Spoel D

    (2008) A temperature predictor for parallel tempering simulations

    Physical Chemistry Chemical Physics 10:2073–2077.

    https://doi.org/10.1039/b716554d

    • Google Scholar
63. 1. Petkova AT
    2. Ishii Y
    3. Balbach JJ
    4. Antzutkin ON
    5. Leapman RD
    6. Delaglio F
    7. Tycko R

    (2002) A structural model for Alzheimer's beta -amyloid fibrils based on experimental constraints from solid state NMR

    Proceedings of the National Academy of Sciences of USA 99:16742–16747.

    https://doi.org/10.1073/pnas.262663499

    • Google Scholar
64. 1. Prabhudesai S
    2. Sinha S
    3. Attar A
    4. Kotagiri A
    5. Fitzmaurice AG
    6. Lakshmanan R
    7. Ivanova MI
    8. Loo JA
    9. Klärner FG
    10. Schrader T
    11. Stahl M
    12. Bitan G
    13. Bronstein JM

    (2012) A novel ‘molecular tweezer’ inhibitor of alpha-synuclein neurotoxicity in vitro and in vivo

    Neurotherapeutics 9:464–476.

    https://doi.org/10.1007/s13311-012-0105-1

    • Google Scholar
65. 1. Rapp F

    (1989)

    Sexually transmitted viruses

    The Yale Journal of Biology and Medicine 62:173–185.

    • Google Scholar
66. 1. Reiss S
    2. Rebhan I
    3. Backes P
    4. Romero-Brey I
    5. Erfle H
    6. Matula P
    7. Kaderali L
    8. Poenisch M
    9. Blankenburg H
    10. Hiet MS
    11. Longerich T
    12. Diehl S
    13. Ramirez F
    14. Balla T
    15. Rohr K
    16. Kaul A
    17. Bühler S
    18. Pepperkok R
    19. Lengauer T
    20. Albrecht M
    21. Eils R
    22. Schirmacher P
    23. Lohmann V
    24. Bartenschlager R

    (2011) Recruitment and activation of a lipid kinase by hepatitis C virus NS5A is essential for integrity of the membranous replication compartment

    Cell Host & Microbe 9:32–45.

    https://doi.org/10.1016/j.chom.2010.12.002

    • Google Scholar
67. 1. Roan NR
    2. Müller JA
    3. Liu H
    4. Chu S
    5. Arnold F
    6. Sturzel CM
    7. Walther P
    8. Dong M
    9. Witkowska HE
    10. Kirchhoff F
    11. Münch J
    12. Greene WC

    (2011) Peptides released by physiological cleavage of semen coagulum proteins form amyloids that enhance HIV infection

    Cell Host & Microbe 10:541–550.

    https://doi.org/10.1016/j.chom.2011.10.010

    • Google Scholar
68. 1. Roan NR
    2. Münch J

    (2015)

    Improving preclinical models of HIV microbicide efficacy

    Trends in Microbiology.

    • Google Scholar
69. 1. Roan NR
    2. Münch J
    3. Arhel N
    4. Mothes W
    5. Neidleman J
    6. Kobayashi A
    7. Smith-McCune K
    8. Kirchhoff F
    9. Greene WC

    (2009) The cationic properties of SEVI underlie its ability to enhance human immunodeficiency virus infection

    Journal of Virology 83:73–80.

    https://doi.org/10.1128/JVI.01366-08

    • Google Scholar
70. 1. Schiedner G
    2. Hertel S
    3. Kochanek S

    (2000) Efficient transformation of primary human amniocytes by E1 functions of Ad5: generation of new cell lines for adenoviral vector production

    Human Gene Therapy 11:2105–2116.

    https://doi.org/10.1089/104303400750001417

    • Google Scholar
71. 1. Shattock RJ
    2. Moore JP

    (2003) Inhibiting sexual transmission of HIV-1 infection

    Nature Reviews. Microbiology 1:25–34.

    https://doi.org/10.1038/nrmicro729

    • Google Scholar
72. 1. Shoichet BK

    (2006) Interpreting steep dose-response curves in early inhibitor discovery

    Journal of Medicinal Chemistry 49:7274–7277.

    https://doi.org/10.1021/jm061103g

    • Google Scholar
73. 1. Sievers Sa
    2. Karanicolas J
    3. Chang HW
    4. Zhao A
    5. Jiang L
    6. Zirafi O
    7. Stevens JT
    8. Münch J
    9. Baker D
    10. Eisenberg D

    (2011) Structure-based design of non-natural amino-acid inhibitors of amyloid fibril formation

    Nature 475:96–100.

    https://doi.org/10.1038/nature10154

    • Google Scholar
74. 1. Sinha S
    2. Du Z
    3. Maiti P
    4. Klarner FG
    5. Schrader T
    6. Wang C
    7. Bitan G

    (2012) Comparison of three amyloid assembly inhibitors: the sugar scyllo-inositol, the polyphenol epigallocatechin gallate, and the molecular tweezer CLR01

    ACS Chemical Neuroscience 3:451–458.

    https://doi.org/10.1021/cn200133x

    • Google Scholar
75. 1. Sinha S
    2. Lopes DH
    3. Du Z
    4. Pang ES
    5. Shanmugam A
    6. Lomakin A
    7. Talbiersky P
    8. Tennstaedt A
    9. McDaniel K
    10. Bakshi R
    11. Kuo PY
    12. Ehrmann M
    13. Benedek GB
    14. Loo JA
    15. Klärner FG
    16. Schrader T
    17. Wang C
    18. Bitan G

    (2011) Lysine-specific molecular tweezers are broad-spectrum inhibitors of assembly and toxicity of amyloid proteins

    Journal of the American Chemical Society 133:16958–16969.

    https://doi.org/10.1021/ja206279b

    • Google Scholar
76. 1. Sinzger C
    2. Hahn G
    3. Digel M
    4. Katona R
    5. Sampaio KL
    6. Messerle M
    7. Hengel H
    8. Koszinowski U
    9. Brune W
    10. Adler B

    (2008) Cloning and sequencing of a highly productive, endotheliotropic virus strain derived from human cytomegalovirus TB40/E

    The Journal of General Virology 89:359–368.

    https://doi.org/10.1099/vir.0.83286-0

    • Google Scholar
77. 1. Sugita Y
    2. Okamoto Y

    (1999) Replica-exchange molecular dynamics method for protein folding

    Chemical Physics Letters 314:141–151.

    https://doi.org/10.1016/S0009-2614(99)01123-9

    • Google Scholar
78. 1. Talbiersky P
    2. Bastkowski F
    3. Klarner FG
    4. Schrader T

    (2008) Molecular clip and tweezer introduce new mechanisms of enzyme inhibition

    Journal of the American Chemical Society 130:9824–9828.

    https://doi.org/10.1021/ja801441j

    • Google Scholar
79. 1. Tedder RS
    2. Gilson RJ
    3. Briggs M
    4. Loveday C
    5. Cameron CH
    6. Garson JA
    7. Kelly GE
    8. Weller IV

    (1991) Hepatitis C virus: evidence for sexual transmission

    BMJ 302:1299–1302.

    https://doi.org/10.1136/bmj.302.6788.1299

    • Google Scholar
80. Book

    1. Tikhonov AN
    2. Arsenin VY

    (1977)

    Solution of III-posed problems

    Washington, DC: Halsted Press.

    • Google Scholar
81. 1. Usmani SM
    2. Zirafi O
    3. Müller JA
    4. Sandi-Monroy NL
    5. Yadav JK
    6. Meier C
    7. Weil T
    8. Roan NR
    9. Greene WC
    10. Walther P
    11. Nilsson KP
    12. Hammarström P
    13. Wetzel R
    14. Pilcher CD
    15. Gagsteiger F
    16. Fändrich M
    17. Kirchhoff F
    18. Münch J

    (2014) Direct visualization of HIV-enhancing endogenous amyloid fibrils in human semen

    Nature Communications 5:3508.

    https://doi.org/10.1038/ncomms4508

    • Google Scholar
82. 1. Van Damme L
    2. Govinden R
    3. Mirembe FM
    4. Guédou F
    5. Solomon S
    6. Becker ML
    7. Pradeep BS
    8. Krishnan AK
    9. Alary M
    10. Pande B
    11. Ramjee G
    12. Deese J
    13. Crucitti T
    14. Taylor D
    15. CS Study Group

    (2008) Lack of effectiveness of cellulose sulfate gel for the prevention of vaginal HIV transmission

    The New England Journal of Medicine 359:463–472.

    https://doi.org/10.1056/NEJMoa0707957

    • Google Scholar
83. 1. van de Wijgert JH
    2. Shattock RJ

    (2007) Vaginal microbicides: moving ahead after an unexpected setback

    AIDS 21:2369–2376.

    https://doi.org/10.1097/QAD.0b013e3282ef83fd

    • Google Scholar
84. 1. van Genderen IL
    2. Brandimarti R
    3. Torrisi MR
    4. Campadelli G
    5. van Meer G

    (1994) The phospholipid composition of extracellular herpes simplex virions differs from that of host cell nuclei

    Virology 200:831–836.

    https://doi.org/10.1006/viro.1994.1252

    • Google Scholar
85. 1. Vigant F
    2. Lee J
    3. Hollmann A
    4. Tanner LB
    5. Akyol Ataman Z
    6. Yun T
    7. Shui G
    8. Aguilar HC
    9. Zhang D
    10. Meriwether D
    11. Roman-Sosa G
    12. Robinson LR
    13. Juelich TL
    14. Buczkowski H
    15. Chou S
    16. Castanho MA
    17. Wolf MC
    18. Smith JK
    19. Banyard A
    20. Kielian M
    21. Reddy S
    22. Wenk MR
    23. Selke M
    24. Santos NC
    25. Freiberg AN
    26. Jung ME
    27. Lee B

    (2013) A mechanistic paradigm for broad-spectrum antivirals that target virus-cell fusion

    PLOS Pathogens 9:e1003297.

    https://doi.org/10.1371/journal.ppat.1003297

    • Google Scholar
86. 1. Weber J
    2. Desai K
    3. Darbyshire J
    4. Microbicides Development P

    (2005) The development of vaginal microbicides for the prevention of HIV transmission

    PLOS Medicine 2:e142.

    https://doi.org/10.1371/journal.pmed.0020142

    • Google Scholar
87. 1. Weise K
    2. Kapoor S
    3. Denter C
    4. Nikolaus J
    5. Opitz N
    6. Koch S
    7. Triola G
    8. Herrmann A
    9. Waldmann H
    10. Winter R

    (2011) Membrane-mediated induction and sorting of K-Ras microdomain signaling platforms

    Journal of the American Chemical Society 133:880–887.

    https://doi.org/10.1021/ja107532q

    • Google Scholar
88. 1. Weise K
    2. Triola G
    3. Brunsveld L
    4. Waldmann H
    5. Winter R

    (2009) Influence of the lipidation motif on the partitioning and association of N-Ras in model membrane subdomains

    Journal of the American Chemical Society 131:1557–1564.

    https://doi.org/10.1021/ja808691r

    • Google Scholar
89. 1. Wojcechowskyj JA
    2. Doms RW

    (2010) A potent, broad-spectrum antiviral agent that targets viral membranes

    Viruses 2:1106–1109.

    https://doi.org/10.3390/v2051106

    • Google Scholar
90. 1. Wolf MC
    2. Freiberg AN
    3. Zhang T
    4. Akyol-Ataman Z
    5. Grock A
    6. Hong PW
    7. Li J
    8. Watson NF
    9. Fang AQ
    10. Aguilar HC
    11. Porotto M
    12. Honko AN
    13. Damoiseaux R
    14. Miller JP
    15. Woodson SE
    16. Chantasirivisal S
    17. Fontanes V
    18. Negrete OA
    19. Krogstad P
    20. Dasgupta A
    21. Moscona A
    22. Hensley LE
    23. Whelan SP
    24. Faull KF
    25. Holbrook MR
    26. Jung ME
    27. Lee B

    (2010) A broad-spectrum antiviral targeting entry of enveloped viruses

    Proceedings of the National Academy of Sciences of USA 107:3157–3162.

    https://doi.org/10.1073/pnas.0909587107

    • Google Scholar
91. 1. Xiao Y
    2. Ma B
    3. McElheny D
    4. Parthasarathy S
    5. Long F
    6. Hoshi M
    7. Nussinov R
    8. Ishii Y

    (2015) Abeta(1-42) fibril structure illuminates self-recognition and replication of amyloid in Alzheimer's disease

    Nature Structural & Molecular Biology 22:499–505.

    https://doi.org/10.1038/nsmb.2991

    • Google Scholar
92. 1. Ye Z
    2. French KC
    3. Popova LA
    4. Lednev IK
    5. Lopez MM
    6. Makhatadze GI

    (2009) Mechanism of fibril formation by a 39-residue peptide (PAPf39) from human prostatic acidic phosphatase

    Biochemistry 48:11582–11591.

    https://doi.org/10.1021/bi901709j

    • Google Scholar
93. 1. Yolamanova M
    2. Meier C
    3. Shaytan AK
    4. Vas V
    5. Bertoncini CW
    6. Arnold F
    7. Zirafi O
    8. Usmani SM
    9. Müller JA
    10. Sauter D
    11. Goffinet C
    12. Palesch D
    13. Walther P
    14. Roan NR
    15. Geiger H
    16. Lunov O
    17. Simmet T
    18. Bohne J
    19. Schrezenmeier H
    20. Schwarz K
    21. Ständker L
    22. Forssmann WG
    23. Salvatella X
    24. Khalatur PG
    25. Khokhlov AR
    26. Knowles TP
    27. Weil T
    28. Kirchhoff F
    29. Münch J

    (2013) Peptide nanofibrils boost retroviral gene transfer and provide a rapid means for concentrating viruses

    Nature Nanotechnology 8:130–136.

    https://doi.org/10.1038/nnano.2012.248

    • Google Scholar
94. 1. Young LM
    2. Saunders JC
    3. Mahood RA
    4. Revill CH
    5. Foster RJ
    6. Tu LH
    7. Raleigh DP
    8. Radford SE
    9. Ashcroft AE

    (2015) Screening and classifying small-molecule inhibitors of amyloid formation using ion mobility spectrometry-mass spectrometry

    Nature Chemistry 7:73–81.

    https://doi.org/10.1038/nchem.2129

    • Google Scholar
95. 1. Zheng X
    2. Liu D
    3. Klarner FG
    4. Schrader T
    5. Bitan G
    6. Bowers MT

    (2015) Amyloid beta-protein assembly: the effect of molecular tweezers CLR01 and CLR03

    The Journal of Physical Chemistry. B 119:4831–4841.

    https://doi.org/10.1021/acs.jpcb.5b00692

    • Google Scholar
96. 1. Zirafi O
    2. Kim KA
    3. Roan NR
    4. Kluge SF
    5. Muller JA
    6. Jiang S
    7. Mayer B
    8. Greene WC
    9. Kirchhoff F
    10. Münch J

    (2014) Semen enhances HIV infectivity and impairs the antiviral efficacy of microbicides

    Science Translational Medicine 6:262ra157.

    https://doi.org/10.1126/scitranslmed.3009634

    • Google Scholar
97. 1. Zoete V
    2. Cuendet MA
    3. Grosdidier A
    4. Michielin O

    (2011) SwissParam: a fast force field generation tool for small organic molecules

    Journal of Computational Chemistry 32:2359–2368.

    https://doi.org/10.1002/jcc.21816

    • Google Scholar

Thank you for sending your work entitled “A molecular tweezer antagonizes seminal amyloids and HIV infection” for consideration at eLife. Your article has been favorably evaluated by Randy Schekman (Senior editor) and three reviewers, one of whom, Jeffery W Kelly, is a member of our Board of Reviewing Editors.

The Reviewing editor and the other reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.

All reviewers concur that the significance of the work is potentially there. The concerns center around some of the data and critical control experiments relevant to the mechanism that seem to be missing.

All three reviewers expressed concerns centered around the mechanism of action of CRL01.

The authors postulate that classic host-guest complexes are forming whereas the reviewers suspect colloidal aggregation on the part of the CLR01 itself. Colloidal aggregation of organic small molecules can disrupt protein fibrils, by protein sequestration by the colloid (Feng et al., Nature Chem. Biol., 2008). Such a mechanism must be controlled for, especially given the peculiar efficacy of the compounds against the virion membranes themselves. Second, the authors should look to see if free lysine competes for CLR01 binding, disrupting its ability to act on the fibrils and the virions. There are few statistical analyses performed, so the relevance of comparing EC50 values is often unclear. On top of that, many of the raw data sets are of too low quality to produce accurate EC50 values, e.g. the lack of a saturation plateau. All three reviews are pasted below and the questions and suggestions in the reviews should be addressed if the authors choose to submit a revised manuscript.

Reviewer #1:

These results appear to be incredible, and they may be. There are several critical controls that appear to be lacking, as indicated below. If the data hold up after these controls, then I am enthusiastic. Amyloid fibrils in seminal fluid composed of at least 4 cationic proteins are established to enhance HIV transmission through intercourse by binding virons augmenting fusion efficiency. The Lys-binding molecular tweezer CLR01 delays the aggregation of PAP248-286 and SEM1(45-107) fibrillization, half-maximal inhibitory concentrations (IC50) of CLR01 inhibition exhibited were 2.6 μM and 18.8 μM, respectively. By contrast, the IC50 to impede PAP85-120 assembly was significantly higher at 970 μM. Have the authors controlled for the possibility that CLR01 is functioning through colloid formation? It is important that the authors show that CLR01 is not impeding the adsorption of the aggregates to the EM grid, which can distort what is in solution. If these reactions are run for more than 30h, does aggregation into fibrils eventually occur? Does seeded aggregation eventually occur? The critical control that CLR01 is not displacing ThT from the fibrils needs to be performed. The analog in which all eight lysine and arginine residues are replaced by alanine appears to be far from a perfect control in that this variant is anticipated to be much more amyloidogenic? In Figure 3, the critical control that CLR01 is not displacing ThT from the fibrils and inhibiting fibril adsorption to the grids needs to be run-the rapidity of disassembly is surprising, is it really disassembly? The result that CLR01 destroys retroviral particles with IC50 values between 10-20 μM by compromising virion integrity further suggests that CLR01 might be functioning as a colloid. CLR01 decreased viral infectivity of semen in a concentration-dependent manner, while CLR03 had no impact on semen-mediated infection enhancement. At a CLR01 concentration of 31 μM, the enhancing effect of semen was completely abolished. Thus, CLR01 prevents HIV infection in the presence of semen. If this is occuring through monomeric CLR01 this is incredible, but I suspect the authors' inhibitor is a micelle or colloid. The data are so impressive that I am concerned that these results are unlikely due to the monomeric molecular tweezer, but my hypothesis could be incorrect.

Reviewer #2:

Seminal amyloids are known to concentrate HIV virions and enhance infectivity. In this report, the authors use a recently identified, general inhibitor of amyloid formation (the molecular tweezer CLR01) to understand whether it also inhibits model seminal amyloids: SEVI, PAP85-120 and SEM1. CLR01 is known to work by docking to exposed lysine and arginine residues and this molecule has found use as a probe for biophysical chemistry and protein folding studies. Lump et al. first test CLR01 as a possible inhibitor of seminal amyloids in vitro, using EM and a ThT assay. They then test whether the molecules can reverse pre-formed amyloids using similar methods. Next, they test whether CLR01 might block binding of mock virions to fibrils and find that the compound seems to disrupt virions themselves. They use AFM and microscopy in vesicles and virus particles to test this idea – concluding that CLR01 seems to disrupt virion membranes, similar to a detergent. Somehow, this membrane disruption effect seems to be exclusive to viral membranes, but the mechanism remains cloudy (and somewhat confusing). They find that CLR01 blocks the infectivity-enhancing activity of seminal fibrils, both with and without addition of BSA. Based on these results, the possibility of using CLR01 for the prevention of viral spread are discussed. However, there are concerns (and even contradictions) about the lack of mechanistic knowledge and the nature of the CLR01-amyloid interaction is not well understood. The null hypotheses – that CLR01 is a non-specific lysine interactor or even a detergent – are not adequately tested. Without deeper mechanistic insight and more rigorous exploration of the binding events, publication of the work seems premature.

1) There are few statistical analyses performed, so the relevance of comparing EC50 values is often unclear. On top of that, many of the raw data sets are of too low quality to produce accurate values. For example, the value of 970 uM (Figure 2E) has no saturation plateau, so one assumes that the value has a very large error. Similar problems can be seen in the raw data for Figure 2F, Figure 2J, Figure 3G, Figure 3H and all of Figure 5. These results cannot be adequately fit because of sparse datapoints, so comparing (or even reporting) the half-maximal values is not advised.

2) One of the major problems with the manuscript is the uncertain mechanism of the compounds and the selectivity of the recognition. If CLR01 is simply blocking cationic charge of lysine and arginine residues, as the zeta potential and all-alanine mutant results imply, than one could presumably achieve this outcome faster, cheaper and more effectively with abundant biological polyanions, such as heparin or nucleic acids (DNA/RNA). What is the effect of these biological polyanions on the assembly of seminal amyloids and infectivity?

3) The hypothesis of CLR01 binding to individual lysine and arginine residues is not experimentally tested outside of a rather crude comparison of an all-alanine replacement peptide (which likely doesn't conform to the same structure as the native peptide). Does free lysine, arginine or other free organic cations (or even divalent cations) compete for CLR01 binding? Does systematic replacements of lysine residues would provide insight into what the key interactions might be? What is the affinity of these interactions? Presumably there are countless free lysine residues in true seminal fluid, so how can CLR01 have the selectivity required to identify the seminal amyloids (especially given the rather weak EC50 values in the mid-micromolar)?

4) The ability of the compounds to block fibril-induced virion activity in the presence of BSA doesn't appear to make logical sense. CLR01 was created as a general lysine/arginine probe – so why wouldn't the compound be adsorbed onto the surface of BSA, which carries surface-exposed nucleophiles? The affinity of CLR01 for the seminal amyloid peptides and BSA (as a stand-in for biological fluids) needs to be carefully performed – by NMR, ITC or other method.

5) The lack of CLR01-induced toxicity in mammalian cells is confusing, given the effects on vesicles and virion membranes (Figure 6). The mammalian cells have numerous cholesterol-rich, raft-like domains, so that argument doesn't work. Presumably, the mammalian membranes would be disrupted just like in the vesicle and virion studies, but not leading to toxicity. What is going on here? Is calcium flux activated? Do the cells change shape? What is the affinity of CLR01 for viral membranes, vesicles and mammalian membranes? What lipid is being bound? By what mechanism do the membranes dissolve? Normally, such questions could be addressed in subsequent work, but the readers of this document would be expected to have major concerns about the logical gas and contradictions in the datasets, making careful and unbiased discussion an important requirement.

6) What is the CMC equivalent of CLR01? In other words, at what concentration does it form micro-aggregates or micelles? Some (perhaps many?) of the observed activities could be a result of non-specific activities, as described by the Shoicet group. DLS studies and detergent competition experiments need to be performed.

Reviewer #3:

Lump and colleagues describe the activity of a host-guest cavitand-like molecule, CLR01, in disassembling seminal protein fibrils and directly disrupting the membrane of HIV (and related) virions, much reducing HIV virus infectivity. The cavitand-like CLR01 acts by binding to lysine or arginine amino acids (the guests), with previously measured affinities in the 20 μM range, and presumably this activity is what leads to de-fibrilization (although exactly why binding should lead to defibrilization is unclear; presumably binding is improved in the soluble form, though this is not stated in the manuscript). In principle this is an interesting story, and certainly a novel use of a cavitand-like molecule, one that might inspire other efforts to repurpose such molecules, which are well-known to chemists, for similar effects. The manuscript could have substantial impact. Before it is published, I suggest the authors address the following concerns:

1) My major concern regards the mechanism of action. The authors postulate that classic host-guest complexes are forming with the cationic amino acids, as shown in Figure 1B, and this is important to the specificity and ultimate usefulness of the compounds. Currently, however, the manuscript lacks two controls that would help establish this mechanism:

A) First, the authors should control for colloidal aggregation on the part of the CLR01 itself. Colloidal aggregation of organic small molecules can disrupt protein fibrils, by protein sequestration in the small molecule colloid (Feng et al., Nature Chem. Biol., 2008). Such a mechanism must be controlled for, especially given the peculiar efficacy of the compounds against the virion membranes themselves (this is hard to square with a host-guest type interaction, but would make sense for colloidal aggregates). To do so, the authors should do one or more of the following i. Look to see if CLR01 is forming colloids in the 10 - 30 μM range, for instance by dynamic light scattering. ii. Add a mild, non-ionic detergent such as Tween-100 to disrupt the colloidal aggregates. This would right-shift the efficacy curves (assuming the detergent doesn't disrupt the fibrils directly). iii. Before addition to the fibrils, spin down the aqueous “solution” of CRL01 (not the DMSO one) in a microfuge at its highest speed for 20 minutes. Use the supernatant as the CLR01 solution for efficacy testing (colloids will typically pellet over this time). iv. Add 10 mg/ml BSA to the fibril mixture, before addition of CLR01. BSA will prophylacticly coat colloids, disrupting their actions.

B) Second, the authors should look to see if free lysine competes for CLR01 binding, disrupting its ability to act on the fibrils and the virions. If it does not, or the effect is small, this would cast doubt on the mechanism proposed by the authors. This to me is an obvious control, and I worry that it was in the manuscript and I just overlooked it – if I did I apologize. If not, it should be done.

Both controls should be done for both the effects on the protein fibrils and on the direct effects of the virus membrane. Whereas I hesitate to put authors to more work, I believe that these controls will substantially strengthen the manuscript, and should take more than a few days to perform.

2) Minor points:

A) I thought many of the dose response curves could be higher in quality, and should be re-done at better resolution.

B) In the third paragraph of the subsection “CLR01 disaggregates preformed SEVI and PAP85-120 fibrils” the authors write that CLR01 acts sub-stoichiometrically. How do they know what the concentration of the relevant fibril species is? What is the stoichiometry?

C) In the Discussion, the authors speculate on the use of CLR01 in vivo, as a drug or drug lead. Probably the only way such a molecule could be used is topically. It will not be orally absorbed (in all likelihood), and an intravenous route would lead to low compliance. If they are to speculate on this, as they do now, they should alert the readership to these points.

D) I was unconvinced by the Concentration-Response curves, which are crucial to the efficacy arguments for CLR01. For Figure 2D, the curve should go through 0 (not 25%...how is this possible at no compound?). More observations are needed in the transition region, and the authors should certainly sample every half-log, not every log10 in concentration. For Figure 2J, they should calculate a curve that actually goes through their points. I assume they fixed the Hill coefficient to 1, which doesn't seem to be supported by the curve. Here too, they need more points. In Figure 5, the Y-axis of the concentration-response curves should be on a linear scale, not a log scale. If it were, the reader would quickly notice that the curves are quite steep, which can be informative, and sometimes concerning.

Notwithstanding these concerns, the manuscript is interesting and could have wide impact. If the authors can address the concerns of the missing controls, I would support its publication.

All reviewers concur that the significance of the work is potentially there. The concerns center around some of the data and critical control experiments relevant to the mechanism that seem to be missing.

All three reviewers expressed concerns centered around the mechanism of action of CRL01.

The authors postulate that classic host-guest complexes are forming whereas the reviewers suspect colloidal aggregation on the part of the CLR01 itself. Colloidal aggregation of organic small molecules can disrupt protein fibrils, by protein sequestration by the colloid (Feng et al., Nature Chem. Biol., 2008). Such a mechanism must be controlled for, especially given the peculiar efficacy of the compounds against the virion membranes themselves.

We thank the editor and reviewers for raising this issue. We agree that it is critical to exclude the possibility that CLR01 is acting via a non-specific mechanism that involves colloidal CLR01 aggregates. Colloidal aggregates formed by certain organic small molecules can inhibit fibril assembly by various amyloidogenic proteins, but to the best of our knowledge have not been reported to disrupt preformed amyloid fibrils (also mentioned by reviewer 3) (Feng et al., 2008). Furthermore, in all our experiments performed with CLR01 during the past 8 years, no hint of colloidal aggregate formation has been found.

Here, to rule out any participation of colloidal aggregates, we performed five independent experiments to assess potential aggregate formation by CLR01 (documented and discussed in detail in the revised manuscript (subsection “CLR01 does not form colloidal aggregates”), new Figure 7). All experiments were performed in exactly the same buffer compositions as used in the manuscript:

1) NMR dilution titrations provided experimental evidence for a very weak CLR01 dimer formation (K_a = ∼60M^-1) in PBS buffer (10 mM sodium phosphate, 137 mM NaCl, 2.7 mM KCl, pH 7.4), with ∼5% CLR01 dimers present at 500 µM (Figure 7A; Figure 7–figure supplement 1). Dimerization was totally absent in HEPES buffer (25 mM HEPES, 150 mM KOAc, 10 mM Mg(OAc)₂, pH 7.4) (Figure 7B; Figure 7–figure supplement 2). Thus, CLR01 is predominantly monomeric.

2) Diffusion NMR (DOSY) experiments in both buffers revealed hydrodynamic radii of ∼0.9 to 1.0 nm, slightly above the monomeric species (Figure 7C, 7D; Figure 7–figure supplement 3).

3) Microcalorimetric dilution titrations revealed only minute endothermic heat changes, strongly contradicting an extensive aggregation process (data not shown).

4) Pyrene fluorescence revealed that no critical micelle concentration (cmc) could be determined in a wide concentration range (0-500 µM CLR01) encompassing fibril assembly and remodeling conditions (Figure 7E).

5) Dynamic Light Scattering (DLS), which overemphasizes the presence of large aggregates because the intensity of the scattered light is proportional to the square of the mass, revealed a very small fraction of CLR01 formed particles with hydrodynamic radius R_H ∼5–8 nm in PBS at concentration between 200–1,000 μM, plus a very minor fraction of particles with a R_H ∼ 20–40 nm (Figure 7F). We did not observe CLR01 particles with an R_H of ∼95-400 nm, which is the size range typically associated with colloidal small-molecule aggregates (McGovern et al., 2002). By contrast, no CLR01 particles could be detected at 10 or 50 μM in PBS (data not shown). Moreover, no CLR01 particles were detected in HEPES buffer up to 1 mM (Figure 7F).

Thus, under all assay concentrations the vast majority of the tweezer molecules are monomeric. Minute amounts of dimers or higher order species may be present at high concentrations of CLR01 in some cases in PBS buffer (fibril assembly and cell culture buffer), but are absent from the HEPES buffer (fibril remodeling buffer). Thus, it is highly unlikely that colloidal CLR01 aggregates contribute meaningfully to the observed activity of the compound.

We have also performed experiments to exclude the possibility that the minute quantities of CLR01 that might form colloidal aggregates were the active species. Thus, the ability of CLR01 to: (a) inhibit fibril assembly by PAP248-286, PAP85-120, and SEM1(45-107); (b) remodel preformed SEVI and PAP85-120 fibrils; and (c) directly inhibit HIV infection (in the absence of seminal amyloid) was not affected by:

1) BSA (10-50mg/ml), which would adsorb to the surface of any colloidal CLR01 aggregates and quench their activity (Figure 2–figure supplement 4; Figure 3–figure supplement 2; Figure 5–figure supplement 2 and 3; Figure 9A) (Feng et al., 2008; McGovern et al., 2002).

2) Preclearing buffers containing CLR01 by centrifugation at 16,100-20,000g for 20min, which would remove any colloidal CLR01 aggregates (Figure 2–figure supplement 4; Figure 3–figure supplement 2; Figure 5–figure supplement 4) (McGovern et al., 2003).

These data strongly suggest that the small quantities of colloidal CLR01 aggregates that might be present are not the active species.

Finally, a feature of small molecules that form colloidal aggregates is that they are non-specific inhibitors (Feng et al., 2008; McGovern et al., 2002; McGovern et al., 2003). By contrast, however, CLR01 displays specificity for amyloidogenesis by peptides that harbor lysine and arginine residues. Thus, if the lysine and arginine residues of PAP248-286 or PAP85-120 are replaced with alanine, then CLR01 can neither inhibit fibril assembly nor remodel preformed fibrils (Figure 2-I, 3I-L). This specificity is not consistent with an indiscriminate, non-specific mechanism of action mediated by colloidal CLR01 aggregates.

Second, the authors should look to see if free lysine competes for CLR01 binding, disrupting its ability to act on the fibrils and the virions.

We have now established that free lysine or poly-L-lysine disrupt the anti-amyloid and antiviral activities of CLR01 (Figure 2–figure supplement 3; Figure 3–figure supplement 1; Figure 4–figure supplement 1; Figure 5–figure supplement 1). These data provide further support that the lysine-binding cavity of CLR01 plays an important role in both the anti-amyloid and antiviral activities of CLR01.

There are few statistical analyses performed, so the relevance of comparing EC50 values is often unclear. On top of that, many of the raw data sets are of too low quality to produce accurate EC50 values, e.g. the lack of a saturation plateau.

We have now included additional replicates and tested additional concentrations of CLR01 to improve the accuracy of our IC₅₀ and EC₅₀ measurements (Figure 2D, E, F, L, 3G, H). We now also qualify differences with statistics where appropriate throughout the manuscript (Figure 2–figure supplement 1, 3-5; Figure 3–figure supplement 1, 2; Figure 5–figure supplement 1, 3 and 4; Figure 6A and 6B; Figure 8A-D; Figure 9A, B).

Reviewer #1:

These results appear to be incredible, and they may be. There are several critical controls that appear to be lacking, as indicated below. If the data hold up after these controls, then I am enthusiastic. Amyloid fibrils in seminal fluid composed of at least 4 cationic proteins are established to enhance HIV transmission through intercourse by binding virons augmenting fusion efficiency. The Lys-binding molecular tweezer CLR01 delays the aggregation of PAP248-286 and SEM1(45-107) fibrillization, half-maximal inhibitory concentrations (IC50) of CLR01 inhibition exhibited were 2.6 μM and 18.8 μM, respectively. By contrast, the IC50 to impede PAP85-120 assembly was significantly higher at 970 μM. Have the authors controlled for the possibility that CLR01 is functioning through colloid formation?

As outlined above in response to editor, we now provide evidence that CLR01 is not functioning through colloidal aggregate formation.

It is important that the authors show that CLR01 is not impeding the adsorption of the aggregates to the EM grid, which can distort what is in solution.

The reviewer raises a good point. When we mix CLR01 (or buffer) with preformed SEVI, PAP85-120, or SEM1(45-107) fibrils for 10 min (a time at which no fibril remodeling occurs; Figure 3A-C) and then adsorb them to the EM grid, we observe abundant fibrils in both the CLR01 and buffer control conditions (Figure 2–figure supplement 2). Furthermore, it is evident in Figure 3C, that CLR01 does not affect adsorption of SEM1(45-107) fibrils to the EM grid. Collectively, these observations strongly suggest that CLR01 does not impede adsorption of peptide conformers to the grid.

If these reactions are run for more than 30h, does aggregation into fibrils eventually occur? Does seeded aggregation eventually occur?

We have run the spontaneously fibril assembly reactions for 68 h for PAP248-286 (Figure 2A) and 72 h for SEM1(45-107) (Figure 2C) and observe very little fibril assembly in the presence of CLR01. We have now also run the PAP85-120 fibril assembly reaction for 68 h and observe very little fibril assembly in the presence of CLR01 (Figure 2B). Thus, under these conditions CLR01 is a potent inhibitor, which prevents spontaneous PAP248-286, PAP85-120, and SEM1(45-107) fibril assembly.

The critical control that CLR01 is not displacing ThT from the fibrils needs to be performed.

We thank the reviewer for suggesting this key control. We have addressed this question in two ways. First, we have monitored early time points (0 min, 10 min and 20 min) in SEVI and PAP85-120 fibril remodeling by CLR01 (Figure 3A, B). At these times, robust ThT fluorescence is still observed indicating that CLR01 is not simply displacing ThT from SEVI or PAP85-120 fibrils. Indeed, fibrils are still prevalent at these early times (Figure 2–figure supplement 2). Thus, CLR01 is not simply competing for ThT binding as we would expect an instantaneous effect on ThT fluorescence as is observed, for example, with another small molecule EGCG and ABeta40 fibrils (Palhano et al., 2013). In this regard, it is also important to note that ThT fluorescence of SEM1(45-107), SEVI(Ala), and PAP85-120(Ala) fibrils is unaffected by CLR01 at various times (Figure 3C, 3I, 3L). Thus, CLR01 is not a general inhibitor or competitor of ThT binding to amyloid fibrils.

In a second approach, we employed a ThT displacement assay (Lockhart et al., 2005). Here, preformed SEVI, PAP85-120, or SEM1(45-107) fibrils (5 µM monomer) were preincubated with ThT (25 µM) for 30min at room temperature. Buffer, CLR01 (250 µM), or a known competitor of ThT binding, BTA-1 (250 µM) (Lockhart et al., 2005) were then added and incubated for 10min at room temperature. ThT displacement was then assessed by fluorescence. BTA-1 effectively displaced ThT, whereas CLR01 did not (Figure 2–figure supplement 1). These findings establish that CLR01 does not simply displace ThT from SEVI, PAP85-120, or SEM1(45-107) fibrils. Thus, any reduction in ThT fluorescence caused by CLR01 can be attributed to an inhibition of fibril assembly or remodeling of fibrils into conformers unable to bind ThT. Inhibition of fibril assembly by CLR01 has also been assessed by EM (Figure 2A-C). Likewise, fibril remodeling by CLR01 has also been assessed by: EM (Figure 3A-C), Proteostat staining, a red fluorescent aggregate sensing dye (Figure 3A-C), seeding activity of remodeled conformers (Figure 3F), and circular dichroism (Figure 3D, E). By employing multiple approaches, we thus establish that CLR01 inhibits amyloid assembly of PAP248-286, PAP85-120, and SEM1(45-107), and remodels preformed SEVI and PAP85-120 fibrils.

The analog in which all eight lysine and arginine residues are replaced by alanine appears to be far from a perfect control in that this variant is anticipated to be much more amyloidogenic?

We have demonstrated that PAP248-286(Ala) and now the PAP85-120(Ala) peptide readily assemble into ThT-reactive fibrils in the presence of CLR01 (Figure 2G-I). Moreover, SEVI(Ala) and PAP85-120(Ala) fibrils are resistant to remodeling by CLR01 (Figure 3I-L). Thus, the activity of CLR01 cannot be attributed to a general ability to displace ThT from fibrils (and see above response). Indeed, CLR01 also has no effect on the ThT fluorescence of preformed SEM1(45-107) fibrils (Figure 3C). Rather, the PAP248-286(Ala) and PAP85-120(Ala) results suggest that CLR01 is unable to inhibit fibrillization or remodel fibrils when the target polypeptide is devoid of lysine and arginine residues. The PAP248-286(Ala) and PAP85-120(Ala) peptides spontaneously assemble into amyloid more rapidly than the wild-type peptides (Figure 2A, B, G, I). However, we have been unable to establish conditions (e.g. higher CLR01 concentrations) where CLR01 prevents assembly of PAP248-286(Ala) and PAP85-120(Ala) into amyloid fibrils. Likewise, we have been unable to establish conditions (e.g. higher CLR01 concentrations) where CLR01 remodels SEVI(Ala) or PAP85-120(Ala) fibrils. These results are consistent with our suggestion that CLR01 is inactive against polypeptides devoid of lysine or arginine residues.

In Figure 3, the critical control that CLR01 is not displacing ThT from the fibrils and inhibiting fibril adsorption to the grids needs to be run-the rapidity of disassembly is surprising, is it really disassembly?

As discussed above, we have demonstrated that CLR01 does not affect amyloid adsorption to EM grids (Figure 2–figure supplement 2) or displace ThT from fibrils (Figure 2–figure supplement 1). Although rapid, we now show that fibril remodeling by CLR01 is not instantaneous. Indeed, SEVI and PAP85-120 fibrils resist remodeling for ∼20 min (Figure 3A, B). However, to be more cautious we now avoid the term fibril disassembly, and instead use the term fibril remodeling, which more accurately describes the observed effects.

The result that CLR01 destroys retroviral particles with IC50 values between 10-20 μM by compromising virion integrity further suggests that CLR01 might be functioning as a colloid. CLR01 decreased viral infectivity of semen in a concentration-dependent manner, while CLR03 had no impact on semen-mediated infection enhancement. At a CLR01 concentration of 31 μM, the enhancing effect of semen was completely abolished. Thus, CLR01 prevents HIV infection in the presence of semen. If this is occuring through monomeric CLR01 this is incredible, but I suspect the authors' inhibitor is a micelle or colloid. The data are so impressive that I am concerned that these results are unlikely due to the monomeric molecular tweezer, but my hypothesis could be incorrect.

As outlined in the response to the editor, CLR01 does not form colloids, micelles, or other large aggregates (Figure 7). We agree that the µM activity of CLR01 against HIV-1 in the presence of semen is remarkable. However, this is not due to formation of colloidal CLR01 aggregates as human seminal fluid contains abundant quantities of protein, which would be anticipated to quench the activity of a non-specific colloidal aggregate by adsorption (Feng et al., 2008). Moreover, CLR01 activity against the virus is preserved in the presence of BSA (10 mg/ml; Figure 5–figure supplement 2 and 3) and is retained in the soluble fraction after centrifugation of CLR01 (Figure 5–figure supplement 4), two established methods to eliminate the activity of non-specific small molecule colloids (Feng et al., 2008). Furthermore, CLR01 exerts powerful inhibitory effects on amyloid accumulation of Abeta and TTR in vivo in mouse models (Attar et al., 2012; Ferreira et al., 2014) and alpha-synuclein aggregation in zebrafish models (Prabhudesai et al., 2012). This potent activity in vivo makes a colloidal-based mechanism extremely unlikely. Taken together, these data suggest that CLR01 is not acting via a mechanism involving non-specific colloidal CLR01 aggregates.

Reviewer #2:

[… The authors] find that CLR01 blocks the infectivity-enhancing activity of seminal fibrils, both with and without addition of BSA. Based on these results, the possibility of using CLR01 for the prevention of viral spread are discussed. However, there are concerns (and even contradictions) about the lack of mechanistic knowledge and the nature of the CLR01-amyloid interaction is not well understood. The null hypotheses – that CLR01 is a non-specific lysine interactor or even a detergent – are not adequately tested. Without deeper mechanistic insight and more rigorous exploration of the binding events, publication of the work seems premature.

We agree that further mechanistic insight would enhance our study. Hence, we have now added several experiments that provide deeper insight into the mechanism of CLR01 action. We can now exclude the possibility that CLR01 is acting via a non-specific mechanism mediated by colloidal CLR01 aggregates or detergent-like CLR01 micelles (see response to editor). We also demonstrate that the lysine-binding cavity of CLR01 plays an important role in both the anti-amyloid and antiviral activities of CLR01 (Figure 2–figure supplement 3; Figure 3–figure supplement 1; Figure 4–figure supplement 1; Figure 5–figure supplement 1). Indeed, it is important to note that the specificity of CLR01 binding to lysine or arginine is well established. CLR01 was tested against almost all naturally occurring amino acid residues in model compounds, and does not bind to any other amino acid side chain except for lysine (K_d∼10 µM) and arginine (K_d∼50 µM) (Fokkens et al., 2005; Klärner et al., 2006; Klärner et al., 2010). The structure of the CLR01-lysine complex and the precise mechanism of lysine threading into the CLR01 guest cavity and subsequent ion pairing have been extensively characterized by NMR spectroscopy, crystal structure, molecular dynamics, and quantum mechanics/molecular mechanics (QM/MM) calculations (Bier et al., 2013; Dutt et al., 2013; Klärner and Schrader, 2013). Importantly, CLR01 has no effect on formation or remodeling of SEVI(Ala) or PAP85-120(Ala) fibrils in which all lysine and arginines residues are exchanged to alanines (Figure 2G-I, 3I-L). Thus, CLR01 does not inhibit amyloid formation or remodel amyloid fibrils in a non-specific way but rather requires that the polypeptides bear arginine or lysine residues. Moreover, the presence of arginine or lysine residues in the target peptide also does not guarantee that CLR01 will be able to remodel amyloid fibrils. Indeed, CLR01 could not remodel preformed SEM1(45-107) fibrils (Figure 3C), indicating a specific effect of CLR01 on SEVI and PAP85-120 fibrils (Figure 3A, B). Importantly, we now also show that free lysine or poly-L-lysine abrogate the anti-amyloid activity of CLR01, further demonstrating specificity (Figure 2–figure supplement 3; Figure 3–figure supplement 1; Figure 4–figure supplement 1). Moreover, CLR01 does not exert cytotoxic effects on cells, unlike non-ionic and anionic detergents such as TX-100, nonoxyl 9, or SDS (Figure 4–figure supplement 3). Collectively, these data establish that CLR01 is not acting in a non-specific manner that involves detergent-like CLR01 micelles.

1) There are few statistical analyses performed, so the relevance of comparing EC50 values is often unclear. On top of that, many of the raw data sets are of too low quality to produce accurate values. For example, the value of 970 uM (Figure 2E) has no saturation plateau, so one assumes that the value has a very large error. Similar problems can be seen in the raw data for Figure 2F, Figure 2J, Figure 3G, Figure 3H and all of Figure 5. These results cannot be adequately fit because of sparse datapoints, so comparing (or even reporting) the half-maximal values is not advised.

We have now included additional replicates and tested additional concentrations of CLR01 to improve the accuracy of our IC₅₀ and EC₅₀ measurements (Figure 2D, E, F, L, 3G, H). We now also qualify differences with statistics where appropriate throughout the manuscript (Figure 2–figure supplement 1, 3-5; Figure 3–figure supplement 1,2; Figure 5–figure supplement 1, 3 and 4; Figure 6A and 6B; Figure 8A-D; Figure 9A, B).

The data shown in Figure 5 were derived from triplicate infections with seven different concentrations of the tweezer including controls (infected and uninfected cells). The inhibition curve follows a typical sigmoidal shape and thus allows a meaningful determination of IC₅₀ values using Graphpad Prism software_. We have extensive experience in the determination of antiviral activities and IC₅₀ values, including the preclinical and clinical development of drugs (Forssmann et al., 2010; Münch et al., 2014; Venken et al., 2011; Zirafi et al., 2014; Zirafi et al., 2015).

2) One of the major problems with the manuscript is the uncertain mechanism of the compounds and the selectivity of the recognition. If CLR01 is simply blocking cationic charge of lysine and arginine residues, as the zeta potential and all-alanine mutant results imply, than one could presumably achieve this outcome faster, cheaper and more effectively with abundant biological polyanions, such as heparin or nucleic acids (DNA/RNA). What is the effect of these biological polyanions on the assembly of seminal amyloids and infectivity?

The reviewer raises an interesting point. However, it is important to note that CLR01 and CLR03 have identical negative charge, but only CLR01 exerts an anti-amyloid or antiviral effect (Figure 2A-C, 3A-C, 5A, 6A, 8A-C, 9A, B). We can therefore exclude the possibility that the effects of CLR01 are a trivial consequence of being anionic. Simply being anionic is not sufficient for anti-amyloid or antiviral activity, as CLR03 lacks activity (Figure 2A-C, 3A-C, 5A, 6A, 8A-C, 9A, B). This can be rationalized by the fact that, at the molecular level, the interaction of CLR01 with Lys residues is based on the formation of inclusion complexes in which Lys is trapped inside the tweezer’s cavity (Bier et al., 2013; Dutt et al., 2013). The chemical structure of CLR03 does not allow for the formation of such inclusion complexes. Thus, as shown by the simulations reported in this paper, CLR03 interactions with the Lys residues of PAP are not preserved, unlike CLR01, which forms conserved inclusion complexes (Figure 1F, G).

With regard to the effect of biological polyanions on HIV infectivity, it is well established that heparin (Ito et al., 1987; Rider, 1997; Vives et al., 2005) or oligodeoxynucleotides (Matsukura et al., 1987; Stein et al., 1993; Vaillant et al., 2006) can inhibit HIV-1 infection by binding to viral envelope glycoproteins, cellular receptors, or both. By contrast, CLR01 directly disrupts the viral membrane thereby eliminating the virus (Figure 6A-D). Hence, CLR01 and polyanions directly inhibit HIV infection by very different mechanisms.

It has also been demonstrated that heparin binds SEVI or PAP85-120 fibrils by electrostatic interactions, rendering the fibril surface negatively charged, thereby preventing enhancement of HIV infection (Arnold et al., 2012; Roan et al., 2009). Even though CLR01 is also negatively charged, it binds to SEVI, PAP85-120, or SEM1(49-107) fibrils via specific interaction with lysine or arginine residues making the fibrils neutrally charged (Figure 4A). In addition, CLR01 prevents SEVI, PAP85-120, or SEM1(45-107) fibril formation (Figure 2A-C) and remodels preformed SEVI and PAP85-120 fibrils (Figure 3A, B), whereas polyanions such as heparin accelerate the formation of SEVI fibrils (Tan et al., 2013). Thus, CLR01 and polyanions inhibit the viral infection enhancing activity of amyloids fibrils by different mechanisms.

Polyanions, including heparin and various other glycosaminoglycans, have been investigated as HIV microbicides for decades, as they are also thought to exert direct anti-HIV activity by inhibiting the binding of the viral envelope glycoprotein, gp120, to its major cell surface receptor CD4 (Callahan et al., 1991; Rider, 1997; Witvrouw and De Clercq, 1997). Unfortunately, such anionic polymers have proven to be unsuccessful in past clinical trials due to their poor bioavailability and induction of inflammatory responses in the genital tract, which instead augment HIV transmission by recruiting HIV-susceptible target cells to the genital mucosa (Lüscher-Mattli, 2000; van de Wijgert and Shattock, 2007). Moreover, it has recently become clear that the HIV-enhancing activity of semen diminishes the efficacy of diverse anti-HIV microbicides, including several polyanions (Zirafi et al., 2014). By contrast, CLR01 retains potent antiviral activity in the presence of semen (Figure 9B). Thus, CLR01 antagonizes seminal amyloid and HIV via distinct mechanisms to polyanions and retains high activity in semen, which is likely an advantageous property for a potential microbicide candidate. We now discuss some of these issues in the Discussion.

3) The hypothesis of CLR01 binding to individual lysine and arginine residues is not experimentally tested outside of a rather crude comparison of an all-alanine replacement peptide (which likely doesn't conform to the same structure as the native peptide).

We have established that CLR01 is ineffective in preventing amyloid formation by PAP248-286(Ala) (Figure 2G, H), and is also unable to remodel preformed SEVI(Ala) fibrils (Figure 3I-K). We have now extended these experiments to a PAP85-120 peptide, PAP85-120(Ala), where all the lysines and arginines are replaced by alanine. CLR01 is unable to inhibit the amyloidogenesis of PAP85-120(Ala) (Figure 2I) and is unable to remodel preformed PAP85-120(Ala) fibrils (Figure 3L). These findings strongly suggest that lysine and arginine residues are critical for CLR01 activity against PAP248-286 and PAP85-120.

It is important to note that it is well established that CLR01 selectively binds lysine in amyloid fibrils (Acharya et al., 2014; Attar et al., 2012; Ferreira et al., 2014; Prabhudesai et al., 2012; Sinha et al., 2012; Sinha et al., 2011; Zheng et al., 2015). We have examined several complexes between CLR01 and peptides/proteins with multiple lysines (Fokkens et al., 2005; Klärner et al., 2006; Klärner et al., 2010). 2D NMR and MS experiments have revealed, for example, that the tweezer molecule sits exclusively on both lysines and to a smaller extent on the single arginine of Αβ1-40 (Sinha et al., 2011). In addition, we have shown that some enzymes with lysine residues around the active site can be inhibited by CLR01, and start to work again, after addition of free lysine (Talbiersky et al., 2008). As mentioned above, the interaction between CLR01 and lysine has been studied and it is structurally well characterized: threading of the lysine side chain through the tweezer cavity, plus ion pair formation between the lysine ammonium cation and the tweezer phosphate anion produces a K_d of ∼10 µM for the 1:1 complex between one lysine side chain and one tweezer molecule, as determined by NMR and fluorescence titration (Bier et al., 2013; Klärner and Schrader, 2013; Lopes et al., 2015). Collectively, these precedents provide further support for a critical role for CLR01 interactions with lysine or arginine residues in the effects we observe on seminal amyloid peptides.

Does free lysine, arginine or other free organic cations (or even divalent cations) compete for CLR01 binding?

To provide evidence that the anti-amyloid activity of CLR01 depends on the specific interaction with lysine, we performed competition experiments with free lysine or poly-L-lysine. As now shown in the revised manuscript, pretreatment of CLR01 with free lysine or poly-L-lysine abrogates the tweezers’ ability to: block amyloid formation (Figure 2–figure supplement 3), remodel amyloid (Figure 3–figure supplement 1), neutralize amyloid surface charge (Figure 4–figure supplement 1), and directly antagonize viral infection (Figure 5–figure supplement 1).

Does systematic replacements of lysine residues would provide insight into what the key interactions might be? What is the affinity of these interactions?

The reviewer makes great suggestions here. It will be extremely interesting to systematically replace the various lysine and arginine residues in PAP248-286, PAP85-120, and SEM1(45-107) to determine which lysines or arginines are critical for CLR01 to prevent amyloid assembly by these peptides. Likewise, it will be very interesting to systematically replace the various lysine and arginine residues in PAP248-286 and PAP85-120 to determine which are critical for fibril remodeling by CLR01. How these alterations affect the affinity of CLR01 for the various peptides will also be very interesting to determine. However, this is a very large set of studies, which we suggest is beyond the scope of this initial manuscript. We plan to perform these for a separate study where we dissect CLR01 mechanism in further detail.

Presumably there are countless free lysine residues in true seminal fluid, so how can CLR01 have the selectivity required to identify the seminal amyloids (especially given the rather weak EC50 values in the mid-micromolar)?

The reviewer raises a very interesting point. The interaction between a single lysine residue and CLR01 has a K_d of ∼10µM, with fast on and off rates. On proteins, however, only a few lysines are well accessible for the relatively large molecular skeleton of the tweezers. For example, ITC and MD+QM/MM calculations have revealed that only 5 of 17 surface lysines in a 14-3-3 protein are complexed by CLR01 (Bier et al., 2013). In addition, a key point is that, because of the high on–off rate, CLR01 disrupts only relatively weak interactions as well as events that depend on critical and accessible lysine or arginine residues. Therefore, the tweezer has shown selectivity in multiple cases that on first impression may appear counter-intuitive. For example, inhibition of abnormal protein aggregation is achieved at 1:1–1:5 protein:CLR01 concentration (Acharya et al., 2014; Sinha et al., 2012; Sinha et al., 2011), whereas disruption of controlled protein polymerization, e.g., of tubulin, requires 55-fold excess CLR01 (Attar et al., 2014) and enzyme inhibition requires ∼1,000-fold excess CLR01 (Talbiersky et al., 2008). Moreover, in vivo, CLR01 shows no toxicity at doses 2–3 orders of magnitude above those needed to enable clearance of amyloid (Attar et al., 2014; Attar et al., 2012; Ferreira et al., 2014; Prabhudesai et al., 2012). Thus, selectivity for lysine-bearing amyloid conformers in complex biological fluids is achieved by CLR01. We now discuss the issue (Discussion).

To experimentally address this question, we have tested the effect of BSA on the ability of the tweezer to antagonize virus infection. Neither coated nor free BSA affected the antiviral activity of CLR01 (Figure 5–figure supplements 2 and 3). Of note, all cell culture experiments (Figure 4G, 5A, 8A-C, 9B) were performed in the presence of 10% FCS which contains abundant proteins, further demonstrating that free lysines at the surface of bulk, structured proteins do not affect CLR01 function. Likewise, CLR01 retains antiviral activity in the presence of seminal fluid (Figure 9B), which also contains abundant proteins bearing lysine residues. Moreover, inhibition of amyloid assembly or amyloid remodeling by CLR01 was unaffected by BSA, indicating that surface lysines and arginines on BSA are not effective competitors of CLR01 activity (Figure 2–figure supplement 4; Figure 3–figure supplement 2).

4) The ability of the compounds to block fibril-induced virion activity in the presence of BSA doesn't appear to make logical sense. CLR01 was created as a general lysine/arginine probe – so why wouldn't the compound be adsorbed onto the surface of BSA, which carries surface-exposed nucleophiles? The affinity of CLR01 for the seminal amyloid peptides and BSA (as a stand-in for biological fluids) needs to be carefully performed – by NMR, ITC or other method.

There is a profound difference between freely accessible lysines on non-structured protein domains, which can all be complexed by the tweezers, and folded proteins, whose surface often does not allow the sterically-demanding tweezer molecule to approach a lysine residue close enough for the threading mechanism. The interaction of CLR01 with seminal fibrils has been discussed above. To experimentally address whether BSA affects the antiviral activity of the tweezer and its ability to antagonize fibril-mediated infectivity enhancement (the two main mechanisms to prevent sexual viral transmission) we performed competition experiments. As mentioned above, we found that physiologically relevant concentrations of coated or free BSA do not affect the direct antiviral activity of CLR01 (Figure 5–figure supplements 2 and 3). These results are in agreement with our previous data showing that the tweezer is active in the presence of semen (Figure 9B). BSA also failed to antagonize inhibition of amyloid assembly or amyloid remodeling by CLR01, suggesting that surface lysines and arginines on BSA are ineffective competitors of CLR01 (Figure 2–figure supplement 4; Figure 3–figure supplement 2).

5) The lack of CLR01-induced toxicity in mammalian cells is confusing, given the effects on vesicles and virion membranes (Figure 6). The mammalian cells have numerous cholesterol-rich, raft-like domains, so that argument doesn't work. Presumably, the mammalian membranes would be disrupted just like in the vesicle and virion studies, but not leading to toxicity. What is going on here? Is calcium flux activated? Do the cells change shape? What is the affinity of CLR01 for viral membranes, vesicles and mammalian membranes? What lipid is being bound? By what mechanism do the membranes dissolve? Normally, such questions could be addressed in subsequent work, but the readers of this document would be expected to have major concerns about the logical gas and contradictions in the datasets, making careful and unbiased discussion an important requirement.

As mentioned by the reviewer, we suggest that a thorough analysis of the interaction of the tweezer with cellular and viral membranes is beyond the scope of this study. We like to point out, however, that we have extensively studied possible cytotoxic effects of CLR01 in vitro and in vivo, but have never observed significant effects at concentrations up to 500 µM (Figure 4–figure supplement 3). We now also show in new Figure 4–figure supplement 3, that CLR01 is far less cytotoxic than various detergents, including Triton X-100, SDS, or N9. The tweezer also has been tested in vivo in multiple experiments, as mentioned above, and was found to have a high safety margin in mice (Attar et al., 2014). As recommended, we now discuss possible mechanisms of membrane disruption in the revised Discussion. Finally, it is important to note that the HIV membrane possesses a highly unusual lipid composition, which is distinct even from the detergent-resistant microdomains typically found in the plasma membrane, indicating that HIV budding involves a specific lipid raft clustering process (Brügger et al., 2006). For example, although the HIV membrane resembles lipid raft microdomains in that it is enriched for saturated lipids, phosphatidylserine, plasmalogen-phosphatidylethanolamine, cholesterol, and sphingolipids, it is unusually enriched for the unusual sphingolipid dihydrosphingomyelin (Brügger et al., 2006; Lorizate et al., 2009). This unusual lipid raft composition could make the viral membrane uniquely susceptible to the action of CLR01. In our future work, we intend to address this interesting point in greater detail.

6) What is the CMC equivalent of CLR01? In other words, at what concentration does it form micro-aggregates or micelles? Some (perhaps many?) of the observed activities could be a result of non-specific activities, as described by the Shoicet group. DLS studies and detergent competition experiments need to be performed.

As outlined in detail in the response to the editor, we performed a series of experiments to exclude that CLR01 forms detergent-like micelles or colloidal aggregates. Thus, we show that even at millimolar concentrations the tweezers are present as predominantly monomeric species; only in PBS buffer does a very weak dimerization occur and minute amounts of higher-order species can be observed by DLS, due to the strong bias of this technique for large particles. No cmc could be determined using pyrene fluorescence. These experiments are summarized in the new Figure 7. We have also performed a series of experiments to exclude the possibility that the minute quantities of higher order CLR01 structures are the active species, i.e.: CLR01 activity is not affected by BSA (which would adsorb to the colloid and quench activity) or by preclearing any colloidal aggregates by centrifugation (Figure 2–figure supplement 4; Figure 3–figure supplement 2; Figure 5–figure supplement 2, 3, and 4). Likewise, CLR01 is highly active in the presence of FCS (Figure 4E, 8A-C) and semen (Figure 9B), complex multiprotein mixtures that would also be predicted to quench the non-specific activity of colloidal CLR01 aggregates. Unfortunately, we were unable to perform the suggested experiments with non-ionic detergents as these exerted inhibitory effects on amyloidogenesis by seminal peptides, which conflated analysis. A similar issue of detergent incompatibility with amyloid assembly was encountered by Feng et al. (Feng et al., 2008).

Reviewer #3:

[…] The manuscript could have substantial impact. Before it is published, I suggest the authors address the following concerns:

1) My major concern regards the mechanism of action. The authors postulate that classic host-guest complexes are forming with the cationic amino acids, as shown in Figure 1B, and this is important to the specificity and ultimate usefulness of the compounds. Currently, however, the manuscript lacks two controls that would help establish this mechanism:

A) First, the authors should control for colloidal aggregation on the part of the CLR01 itself. Colloidal aggregation of organic small molecules can disrupt protein fibrils, by protein sequestration in the small molecule colloid (Feng et al., Nature Chem. Biol., 2008). Such a mechanism must be controlled for, especially given the peculiar efficacy of the compounds against the virion membranes themselves (this is hard to square with a host-guest type interaction, but would make sense for colloidal aggregates). To do so, the authors should do one or more of the following i. Look to see if CLR01 is forming colloids in the 10 - 30 μM range, for instance by dynamic light scattering. ii. Add a mild, non-ionic detergent such as Tween-100 to disrupt the colloidal aggregates. This would right-shift the efficacy curves (assuming the detergent doesn't disrupt the fibrils directly). iii. Before addition to the fibrils, spin down the aqueous “solution” of CRL01 (not the DMSO one) in a microfuge at its highest speed for 20 minutes. Use the supernatant as the CLR01 solution for efficacy testing (colloids will typically pellet over this time). iv. Add 10 mg/ml BSA to the fibril mixture, before addition of CLR01. BSA will prophylacticly coat colloids, disrupting their actions.

As outlined above in the response to editor, we now provide evidence that CLR01 is not functioning through colloidal aggregate formation. Thus, CLR01 is predominantly monomeric and only forms minute quantities of colloidal aggregates in PBS buffer, and does not form them at all in HEPES buffer (Figure 7). Even then, CLR01 activity is unaffected by BSA or by preclearing aqueous solutions of CLR01 (incidentally, stock CLR01 solutions were made in water or PBS and not DMSO, p. 23) via centrifugation (Figure 2–figure supplement 4; Figure 3–figure supplement 2; Figure 5–figure supplement 2, 3, and 4), indicating that even if present, colloidal CLR01 particles cannot be the active species. Unfortunately, we were unable to perform the suggested experiments with non-ionic detergents as these exerted inhibitory effects on amyloidogenesis by seminal peptides, which conflated analysis. A similar issue of detergent incompatibility with amyloid assembly was encountered by Feng et al. (Feng et al., 2008).

B) Second, the authors should look to see if free lysine competes for CLR01 binding, disrupting its ability to act on the fibrils and the virions. If it does not, or the effect is small, this would cast doubt on the mechanism proposed by the authors. This to me is an obvious control, and I worry that it was in the manuscript and I just overlooked it – if I did I apologize. If not, it should be done.

Both controls should be done for both the effects on the protein fibrils and on the direct effects of the virus membrane. Whereas I hesitate to put authors to more work, I believe that these controls will substantially strengthen the manuscript, and should take more than a few days to perform.

As outlined above in the response to the editor, we performed these important controls. As expected from all previously published data on CLR01, we found that free lysine and poly-L-lysine abrogate the anti-amyloid and antiviral effects of CLR01 (Figure 2–figure supplement 3; Figure 3–figure supplement 1; Figure 4–figure supplement 1; Figure 5–figure supplement 1).

2) Minor points:

A) I thought many of the dose response curves could be higher in quality, and should be re-done at better resolution.

We have now included additional replicates and tested additional concentrations of CLR01 to improve the accuracy of our IC₅₀ and EC₅₀ measurements (Figure 2D, E, F, L, 3G, H).

B) In the third paragraph of the subsection “CLR01 disaggregates preformed SEVI and PAP85-120 fibrils” the authors write that CLR01 acts sub-stoichiometrically. How do they know what the concentration of the relevant fibril species is? What is the stoichiometry?

By substoichiometric we meant relative to peptide monomers not fibrils. We now clarify this point in the manuscript (subsection “CLR01 remodels preformed SEVI and PAP85-120 fibrils”).

C) In the Discussion, the authors speculate on the use of CLR01 in vivo, as a drug or drug lead. Probably the only way such a molecule could be used is topically. It will not be orally absorbed (in all likelihood), and an intravenous route would lead to low compliance. If they are to speculate on this, as they do now, they should alert the readership to these points.

Of course a topical application of CLR01 as microbicide is the only possible application, we hope that we have clarified this point in the revised manuscript (see first paragraph in the Discussion).

D) I was unconvinced by the Concentration-Response curves, which are crucial to the efficacy arguments for CLR01. For Figure 2D, the curve should go through 0 (not 25%...how is this possible at no compound?). More observations are needed in the transition region, and the authors should certainly sample every half-log, not every log10 in concentration.

We have now adjusted our fits so the curves go through zero. We have now included additional replicates and tested additional concentrations of CLR01 to improve the accuracy of our IC₅₀ and EC₅₀ measurements (Figure 2D, E, F, L, 3G, H).

For Figure 2J, they should calculate a curve that actually goes through their points. I assume they fixed the Hill coefficient to 1, which doesn't seem to be supported by the curve. Here too, they need more points.

The reviewer is correct, we had fixed the Hill coefficient to 1. We now fit with a variable slope and have included additional replicates and tested additional concentrations of CLR01 to improve the accuracy of our IC₅₀ measurement in Figure 2L.

In Figure 5, the Y-axis of the concentration-response curves should be on a linear scale, not a log scale. If it were, the reader would quickly notice that the curves are quite steep, which can be informative, and sometimes concerning.

As exemplified by Figure 5A, a semi-log representation does not provide more information, also with respect to the steepness of the curves. As we published all our research on antivirals as log-log diagrams e.g. Zirafi et al. (Zirafi et al., 2014; Zirafi et al., 2015), we would prefer (if possible) to keep the figures in their original log-log representation.

Author response image 1

Download asset Open asset

Author details

1. Edina Lump

   Institute of Molecular Virology, Ulm University Medical Center, Ulm, Germany

   Contribution

   EL, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Contributed equally with

   Laura M Castellano

   Competing interests

   No competing interests declared.

2. Laura M Castellano

   1. Department of Biochemistry and Biophysics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, United States
   2. Pharmacology Graduate Group, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, United States

   Contribution

   LMC, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Contributed equally with

   Edina Lump

   Competing interests

   No competing interests declared.

3. Christoph Meier

   Institute of Organic Chemistry III/Macromolecular Chemistry, Ulm University, Ulm, Germany

   Contribution

   CM, Conception and design, Acquisition of data, Analysis and interpretation of data

   Competing interests

   No competing interests declared.

4. Janine Seeliger

   Physical Chemistry I—Biophysical Chemistry, Department of Chemistry and Chemical Biology, Technical University of Dortmund, Dortmund, Germany

   Contribution

   JS, Conception and design, Acquisition of data, Analysis and interpretation of data

   Competing interests

   No competing interests declared.

5. Nelli Erwin

   Physical Chemistry I—Biophysical Chemistry, Department of Chemistry and Chemical Biology, Technical University of Dortmund, Dortmund, Germany

   Contribution

   NE, Conception and design, Acquisition of data, Analysis and interpretation of data

   Competing interests

   No competing interests declared.

6. Benjamin Sperlich

   Physical Chemistry I—Biophysical Chemistry, Department of Chemistry and Chemical Biology, Technical University of Dortmund, Dortmund, Germany

   Contribution

   BS, Conception and design, Acquisition of data, Analysis and interpretation of data

   Competing interests

   No competing interests declared.

7. Christina M Stürzel

   Institute of Molecular Virology, Ulm University Medical Center, Ulm, Germany

   Contribution

   CMS, Conception and design, Acquisition of data, Analysis and interpretation of data

   Competing interests

   No competing interests declared.

8. Shariq Usmani

   Institute of Molecular Virology, Ulm University Medical Center, Ulm, Germany

   Contribution

   SU, Conception and design, Acquisition of data, Analysis and interpretation of data

   Competing interests

   No competing interests declared.

9. Rebecca M Hammond

   1. Department of Biochemistry and Biophysics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, United States
   2. Biology Department, Swarthmore College, Swarthmore, United States

   Contribution

   RMH, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   No competing interests declared.

10. Jens von Einem

    Institute of Virology, Ulm University Medical Center, Ulm, Germany

    Contribution

    JE, Conception and design, Acquisition of data, Analysis and interpretation of data

    Competing interests

    No competing interests declared.

11. Gisa Gerold

    Institute of Experimental Virology, Twincore, Centre for Experimental and Clinical Infection Research, Hannover, Germany

    Contribution

    GG, Conception and design, Acquisition of data, Analysis and interpretation of data

    Competing interests

    No competing interests declared.

12. Florian Kreppel

    Institute of Gene Therapy, Ulm University Medical Center, Ulm, Germany

    Contribution

    FK, Conception and design, Acquisition of data, Analysis and interpretation of data

    Competing interests

    No competing interests declared.

13. Kenny Bravo-Rodriguez

    Max-Planck-Institut für Kohlenforschung, Mülheim an der Ruhr, Germany

    Contribution

    KB-R, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

    Competing interests

    No competing interests declared.

14. Thomas Pietschmann

    Institute of Experimental Virology, Twincore, Centre for Experimental and Clinical Infection Research, Hannover, Germany

    Contribution

    TP, Conception and design, Acquisition of data, Analysis and interpretation of data

    Competing interests

    No competing interests declared.

15. Veronica M Holmes

    Division of Infectious Diseases, Department of Medicine, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, United States

    Contribution

    VMH, Conception and design, Drafting or revising the article, Contributed unpublished essential data or reagents

    Competing interests

    No competing interests declared.

16. David Palesch

    Institute of Molecular Virology, Ulm University Medical Center, Ulm, Germany

    Contribution

    DP, Conception and design, Acquisition of data, Analysis and interpretation of data

    Competing interests

    No competing interests declared.

17. Onofrio Zirafi

    Institute of Molecular Virology, Ulm University Medical Center, Ulm, Germany

    Contribution

    OZ, Conception and design, Acquisition of data, Analysis and interpretation of data

    Competing interests

    No competing interests declared.

18. Drew Weissman

    Division of Infectious Diseases, Department of Medicine, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, United States

    Contribution

    DW, Conception and design, Drafting or revising the article, Contributed unpublished essential data or reagents

    Competing interests

    No competing interests declared.

19. Andrea Sowislok

    Department of Chemistry, University of Duisburg-Essen, Essen, Germany

    Contribution

    AS, Acquisition of data, Analysis and interpretation of data, Contributed unpublished essential data or reagents

    Competing interests

    No competing interests declared.

20. Burkhard Wettig

    Department of Chemistry, University of Duisburg-Essen, Essen, Germany

    Contribution

    BW, Acquisition of data, Analysis and interpretation of data, Contributed unpublished essential data or reagents

    Competing interests

    No competing interests declared.

21. Christian Heid

    Department of Chemistry, University of Duisburg-Essen, Essen, Germany

    Contribution

    CH, Acquisition of data, Analysis and interpretation of data, Contributed unpublished essential data or reagents

    Competing interests

    No competing interests declared.

22. Frank Kirchhoff

    1. Institute of Molecular Virology, Ulm University Medical Center, Ulm, Germany
    2. Ulm-Peptide Pharmaceuticals, Ulm University, Ulm, Germany

    Contribution

    FK, Conception and design, Analysis and interpretation of data, Drafting or revising the article

    Competing interests

    No competing interests declared.

23. Tanja Weil

    1. Institute of Organic Chemistry III/Macromolecular Chemistry, Ulm University, Ulm, Germany
    2. Ulm-Peptide Pharmaceuticals, Ulm University, Ulm, Germany

    Contribution

    TW, Conception and design, Analysis and interpretation of data, Drafting or revising the article

    Competing interests

    No competing interests declared.

24. Frank-Gerrit Klärner

    Institute of Organic Chemistry, University of Duisburg-Essen, Essen, Germany

    Contribution

    F-GK, Conception and design, Analysis and interpretation of data, Drafting or revising the article

    Competing interests

    F-GK: Co-inventor of International Patent Application No. PCT/US2010/026419, USA Patent 8,791,092, European Patent Application 10 708 075.6

25. Thomas Schrader

    Department of Chemistry, University of Duisburg-Essen, Essen, Germany

    Contribution

    TS, Conception and design, Analysis and interpretation of data, Drafting or revising the article

    Competing interests

    TS: Co-inventor of International Patent Application No. PCT/US2010/026419, USA Patent 8,791,092, European Patent Application 10 708 075.6

26. Gal Bitan

    1. Department of Neurology, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, United States
    2. Brain Research Institute, University of California at Los Angeles, Los Angeles, Los Angeles, United States
    3. Molecular Biology Institute, University of California, Los Angeles, United States

    Contribution

    GB, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents

    Competing interests

    GB: Co-inventor of International Patent Application No. PCT/US2010/026419, USA Patent 8,791,092, European Patent Application 10 708 075.6

27. Elsa Sanchez-Garcia

    Max-Planck-Institut für Kohlenforschung, Mülheim an der Ruhr, Germany

    Contribution

    ES-G, Conception and design, Analysis and interpretation of data, Drafting or revising the article

    Competing interests

    No competing interests declared.

28. Roland Winter

    Physical Chemistry I—Biophysical Chemistry, Department of Chemistry and Chemical Biology, Technical University of Dortmund, Dortmund, Germany

    Contribution

    RW, Conception and design, Analysis and interpretation of data, Drafting or revising the article

    Competing interests

    No competing interests declared.

29. James Shorter

    1. Department of Biochemistry and Biophysics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, United States
    2. Pharmacology Graduate Group, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, United States

    Contribution

    JS, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

    For correspondence

    jshorter@mail.med.upenn.edu

    Competing interests

    No competing interests declared.

30. Jan Münch

    1. Institute of Molecular Virology, Ulm University Medical Center, Ulm, Germany
    2. Ulm-Peptide Pharmaceuticals, Ulm University, Ulm, Germany

    Contribution

    JM, Conception and design, Analysis and interpretation of data, Drafting or revising the article

    For correspondence

    Jan.Muench@uni-ulm.de

    Competing interests

    No competing interests declared.

• 5,537

  Page views

• 713

  Downloads

• 62

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, Scopus, PubMed Central.