Nanomedicine for prion disease treatment

Abstract

Despite their devastating impact, no effective therapeutic yet exists for prion diseases at the symptomatic stage in humans or animals. Progress is hampered by the difficulty in identifying compounds that affect PrP^Sc and the necessity of any potential therapeutic to gain access to the CNS. Synthetic polymers known as dendrimers are a particularly promising candidate in this area. Studies with cell culture models of prion disease and prion infected brain homogenate have demonstrated that numerous species of dendrimers eliminate PrP^Sc in a dose and time dependent fashion and specific glycodendrimers are capable of crossing the CNS. However, despite their potential a number of important questions remained unanswered such as what makes an effective dendrimer and how dendrimers eliminate prions intracellularly. In a number of recent studies we have tackled these questions and revealed for the first time that a specific dendrimer can inhibit the intracellular conversion of PrP^C to PrP^Sc and that a high density of surface reactive groups is a necessity for dendrimers in vitro anti-prion activity. Understanding how a therapeutic works is a vital component in maximising its activity and these studies therefore represent a significant development in the race to find effective treatments for prion diseases.

Keywords: :

• prion
• dendrimer
• therapeutic
• scrapie cell assay
• neurodegeneration
• PrP N-terminal
• amyloid
• nanomedicine

This article refers to:

Dendrimers

Dendrimers are repetitively branched synthetic molecules with a wide variety of biological applications. They generally consist of a core from which monomers branch out in a symmetrical, structured fashion, resulting in a spherical three dimensional morphology (Fig. 1). The number of branch points defines the generation of dendrimer while the surface functional groups are the primary factor in determining the ionic charge. Like many great discoveries, dendrimers anti-prion activity was serendipitously discovered when it was found that the poly(amidoamine) (PAMAM) dendrimer in a transfection reagent eliminated prions from ScN2a cells.¹ Since then numerous other dendrimer species have demonstrated anti-prion activity including phosphorous dendrimers, maltose based glycodendrimers (mPPI), Poly(propyleneimine) (PPI) and the dendrimer like hyperbranched polymer poly(ethyleneimine) (PEI).^1-5 The ability of these dendrimers to eliminate PrP^Sc from infected cells is dose and time dependent and increases with the generation number of the dendrimer. In addition, mPPI, PPI, PAMAM and high MW PEI are all capable of eliminating protease resistant PrP^Sc from RML infected brain homogenate, making these synthetic polymers the only known therapeutics to be effective against prions in both an intracellular and in vitro setting.^1,3,4

Figure 1. Poly(amidoamine) dendrimer (PAMAM) structure and nomenclature

Despite their immense potential, many crucial questions surrounding dendrimer prion interactions remained unanswered at the time we began our study. For example the chemical properties of a dendrimer which define its anti-prion activity and the exact mechanism by which dendrimers eliminate PrP^Sc intracellularly were both unknown. In a number of recent manuscripts we have addressed these questions and advanced our understanding of the complex factors governing dendrimer/prion interactions.

Properties Which Govern Anti-Prion Activity of Dendrimers

The anti-prion efficacy of a dendrimer was traditionally believed to be based upon its cationic charge density.⁶ However in a recent study we demonstrated that anionic, cationic and polar dendrimers are all capable of exhibiting anti-prion activity and that a high density of reactive surface groups is a necessity.⁷

To achieve this, we produced a number of novel dendrimers through the surface group modification of a neutral maltose poly(propyleneimine) dendrimer (mPPI). The modified dendrimers varied in size, structure, charge and surface group composition and were examined against a range of prion strains to determine the effect of the modifications on dendrimer efficacy. Surprisingly, we found that dendrimers with an anti-prion effect could be cationic, anionic or neutral in charge as long as a high density of reactive surface groups was present, indicating dendrimers may interact with patches of minor surface charge on the PrP^Sc molecule opposite to the major surface charge. Support for dendrimers performing such a feat can be found in previous studies with proteins other than PrP.⁸ Even more surprising was the observation that the same prion strains were always the most susceptible to elimination, regardless of dendrimer charge. This suggests that upon interaction with protease resistant PrP^Sc, dendrimers destabilize the protein through a somewhat generic mechanism. One possible explanation is that dendrimers act as water structure perturbing solutes—a highly destabilizing event for most protein tertiary structures. Again support for dendrimers achieving this feat can be found in previous studies with proteins other than PrP.⁹

Having established the properties of a dendrimer which influence its in vitro anti-prion activity, we next investigated the efficacy of a particularly promising maltose dendrimer in an intracellular setting.

Anti-Prion Efficacy of mPPIg5

Maltose poly(propyleneimine) generation five (mPPIg5) is a particularly promising dendrimer with remarkably low cytotoxicity.³ We investigated its efficacy in eliminating PrP^Sc from scrapie infected neuroblastoma cells. This is traditionally assessed by monitoring the elimination of protease resistant PrP^Sc over time through the use of PK digestion (to remove PrP^C) and immunoblotting (to detect the remaining protease resistant PrP^Sc). As immunoblotting is a laborious technique and at best semi-quantitative, we developed a high throughput and highly quantitative assay known as the modified scrapie cell assay (mSCA), to measure the efficacy of anti-prion drugs in a cell culture setting.¹⁰ The mSCA is based upon the standard scrapie cell assay (sSCA), an Elispot based assay for determining the level of infectious prions in a sample.¹¹ In the sSCA, susceptible cells are exposed to prion-containing samples, grown to confluence and passaged three times. The cells are subsequently attached to an Elispot plate and undergo numerous treatments including a protease digestion step to remove PrP^C and an immuno-detection step to detect protease resistant PrP^Sc. We altered the protease digestion step and some of the incubation stages of the sSCA to produce an assay more suitable for detecting a decrease in prion infected cells over time following treatment with a therapeutic agent. The result was the creation of the mSCA, a high-throughput and fully quantitative method for measuring the efficacy of anti-prion drugs. This assay was used to demonstrate that despite its low cytotoxicity, mPPIg5 is a highly efficient anti-prion therapeutic with an activity comparable to the more established anti-prion compounds STI571 and suramin.¹⁰

Intracellular Method of Action of mPPIg5

Having established the efficacy of mPPIg5, we next set about determining its intracellular mode of action. Previous studies suggested that dendrimers with an anti-prion effect act directly on PrP^Sc in lysosomes to render them sensitive to proteases. This assumption was based on three principle pieces of evidence:

1. Dendrimers (PPI) localize in lysosomes.⁴

2. The alkalising lysomotrophic agent chloroquine, inhibits the elimination of prions by branched polyamines (PEI).¹

3. The in vitro elimination of protease resistant PrP^Sc by dendritic polymers (PEI and PPI) was greatest at an acidic pH.^1,4

Collectively, these findings certainly suggest that dendrimers with anti-prion activity achieve their effect by increasing the degradation of PrP^Sc within lysosomes. However when examined individually, each of the arguments is less robust. For example, PPI dendrimers localizing predominantly in lysosomes is not surprising as most endocytosed materials will end up there. Chloroquine’s inhibitory effect on branched polyamines (PEI) cannot be automatically attributed to a rise in lysosomal pH as it has numerous other effects on cells such as altering protein trafficking and endocytosis.¹² Furthermore, NH₄Cl, another alkalising lysomotrophic agent, does not inhibit the effect of PEI¹ suggesting alkalisation of acidic compartments alone is not sufficient. Finally, the finding that the in vitro elimination of protease resistant PrP^Sc by dendrimers was greatest at an acidic pH is also not conclusive as the in vitro anti-prion action of dendrimers appears to involve the denaturation of PrP^Sc and so, an acidic pH enhancing this is not surprising. In addition, an acidic pH is not necessary for the in vitro anti-prion effect dendrimers, it just optimizes it, and so the necessity of an acidic environment intracellularly cannot be automatically assumed.^7,13 Most pertinently, a small number of dendrimers, including mPPIg5, have been shown to be effective in an intracellular environment but not always in an in vitro one, even when tested against the same prion strain.^10,14 If dendrimers do indeed act directly against protease resistant PrP^Sc to render it sensitive to protease degradation, how can one explain a dendrimer working intracellularly on PrP^Sc but not in vitro? Was it possible that dendrimers act through a different mechanism intracellularly? To examine this and to determine how mPPIg5 was eliminating prions from cells, we took advantage of the unique intracellular processing of PrP^Sc.

Exploiting the Intracellular Cleavage of PrP^Sc

PrP^Sc has residues 23–88 cleaved from its N-terminus approximately one hour after its production.¹⁵ The truncated PrP^Sc formed by this event is known as PrP^27-30 and accumulates in lysosomes.¹⁵ For the sake of clarity, we use the term “Full Length PrP^Sc (FL PrP^Sc)” to refer to non-truncated protease resistant PrP^Sc and PrP^27-30 to refer to truncated protease resistant PrP^Sc. PrP^Sc is used to refer to all disease isoforms of the prion protein.

In summary: PrP^C → FL PrP^Sc → PrP^27-30

As FL PrP^Sc is short lived, its intracellular level is closely related to its production rate. If the synthesis of FL PrP^Sc is interfered with, the level of detectable FL PrP^Sc drops rapidly. Thus the level of FL PrP^Sc can be seen as a marker of the conversion of PrP^C to misfolded forms. NH₄CL treatment of cells inhibits the truncation of FL PrP^Sc.¹⁵ Thus in prion infected cells NH₄Cl treatment leads to a rapid increase in FL PrP^Sc as long as the conversion of PrP^C to FL PrP^Sc is not inhibited.

We exploited this fact to determine how a drug eliminates PrP^Sc from cells. We reasoned, if a compound inhibits the conversion of PrP^C to FL PrP^Sc, no difference between cells treated in the presence and absence of NH₄Cl should be observed; FL PrP^Sc is not being produced so NH₄Cl induced inhibition of truncation has no effect.

On the other hand, if an anti-prion compound solely achieves its effect by increasing the degradation of PrP^Sc isoforms (including FL PrP^Sc), prion infected cells treated in the presence of NH₄Cl should produce a higher level of FL PrP^Sc than cells treated in the absence of NH₄Cl. There are two possible reasons for this. (1) NH₄Cl may raise the pH of the lysosome and consequently inhibit the anti-prion action of a drug. In this instance, PrP^Sc isoforms (including FL PrP^Sc) would not be degraded and so an increase in the level of FL PrP^Sc would be observed in cells treated in the presence of NH₄Cl over cells treated in the absence of NH₄Cl. (2) In the presence of NH₄Cl, an anti-prion drug may still degrade PrP^Sc isoforms. However NH₄Cl treated cells will possess an increased level of FL PrP^Sc. Cells treated with an anti-prion drug in the presence of NH₄Cl will therefore display a higher level of FL PrP^Sc than cells treated in the absence of NH₄Cl as there is more FL PrP^Sc to degrade.

Thus it can be seen that by comparing a compounds effect on cellular levels of FL PrP^Sc in the presence and absence of NH₄Cl over a short time frame, we were able to determine whether a drug increases the degradation of PrP^Sc or inhibits the conversion of PrP^C to the misfolded PrP^Sc.

nSCA for Determining Compounds Mode of Action

To produce a method capable of quantifying the level of FL PrP^Sc in treated cells, the N-terminal specific scrapie cell assay was created (nSCA). This is another modified version of the sSCA and specifically detects FL PrP^Sc—it does not detect PrP^C or PrP^27-30. Two drugs with a known method of action, suramin and STI571, were used to validate the assay and demonstrate that our system is capable of determining an anti-prion therapeutics’ mode of action. Finally, the system was used to examine how mPPIg5 eliminates PrP^Sc from cells. The assay demonstrated that mPPIg5 is a highly potent drug which inhibits the conversion of PrP^C to PrP^Sc—the first time this has been demonstrated for a dendrimer. Through the use of pulse chase, confocal microscopy and phosphatidylinositol-specific phospholipase C we demonstrated that mPPIg5 does not interfere with the biogenesis or processing of the regular PrP isoform. Thus it can be seen that maltose dendrimers are a highly promising class of therapeutics, which specifically target misfolded PrP and prevent the conversion of PrP^C to PrP^Sc.¹⁰

There are a number of possible explanations for how mPPIg5 achieves this. mPPIg5 may competitively inhibit interactions between PrP^C and PrP^Sc. It may interfere with PrP^Sc trafficking to such an extent that it is no longer available for the conversion of PrP^C. Alternatively it may interfere with components of PrP^Sc formation which have not yet been identified such as short-lived PrP^C → PrP^Sc intermediates or potential co-factors like RNA which are necessary for PrP^Sc formation.¹⁶ However, in our opinion the most likely explanation is that mPPIg5 alters the structure of PrP^Sc to such an extent that it is no longer capable of initiating the mis-folding of PrP^C.

Evidence for Dendrimers Altering Structure of PrP^Sc

Evidence for mPPIg5 and other dendrimers effecting the structure of PrP^Sc comes from in vitro studies with prion infected brain homogenate, performed in our laboratory and elsewhere, demonstrating that various dendrimers appear to denature protease resistant PrP^Sc and render it susceptible to proteolysis.^1,4,7,13 Additionally, electron microscopy and Fourier transformed infrared spectroscopy studies have demonstrated that PPI dendrimer treatment of purified PrP^Sc leads to disaggregation and a loss in β-sheet structure.⁴ It is possible that this change in conformation inhibits PrP^Sc’s ability to convert PrP^C to misfolded forms. This is supported by in vitro fibrilization studies that demonstrate dendrimer treatment of PrP peptides inhibits their ability to aggregate.²

Dendrimers altering the conformation of PrP^Sc may explain all anti-prion activity that has been reported for dendrimers in both an in vitro and intracellular setting. One can envisage dendrimers interacting with PrP^Sc (both intracellularly and in vitro) and altering its structure in a process that is governed by the original confirmation of the PrP^Sc molecule (influenced by the type of prion strain)¹³ and on the chemical nature of the dendrimer (dependant on a combination of the dendrimers size, structure, charge and density of reactive surface groups).⁷ In some scenarios this interaction may alter the conformation of PrP^Sc to such an extent that it loses its ability to initiate the misfolding of PrP^C and becomes susceptible to proteolysis. However, in other situations the change in conformation may be more subtle and PrP^Sc, while losing its ability to induce misfolding of PrP^C, is not altered to the extent whereby it becomes sensitive to proteolysis. This would explain how some dendrimers fail to reduce protease resistant PrP^Sc in vitro but can be effective against the same prion strains intracellularly. Additionally, this model allows dendrimers to eliminate prions intracellularly by increasing their degradation as the original experiments in this area suggest^1,4,5 and/or eliminate prions by inhibiting the conversion of PrP^C to PrP^Sc as some of the more recent studies in this area, including our own, suggest.^10,14

Dendrimers altering the conformation of misfolded PrP may also explain in vitro studies which have shown that numerous dendrimers (including mPPI) inhibit the fibrilization of PrP peptides in vitro.^2,17-19 Dendrimers increasing the fibril breakage rate and dendrimers blocking the ends of growing fibrils are two of the explanations offered for this phenomena.¹⁸ While both of these scenarios are possible in an intracellular setting, dendrimers blocking the end of growing fibrils cannot explain the in vitro ability of various dendrimers to reduce protease resistant PrP^Sc in prion infected brain homogenate.^1,3-5 Conversely, specific dendrimers increasing the fibril breakage rate can explain this ability and can again be linked back to a dendrimers ability to alter the conformation of misfolded PrP. While it is entirely possible that particular dendrimers achieve their anti-prion effect through a variety of mechanisms, we believe dendrimers altering the conformation of PrP^Sc is the most likely explanation as it can explain the anti-prion activity observed for dendrimers in a variety of settings.

Lack of mPPIg5 Effect on PrP^C—Positive or Negative?

The lack of an effect of mPPIg5 on the biogenesis and processing of PrP^C is particularly interesting as it can be viewed as a negative or positive feature of mPPIg5. On the positive side, the role of the highly conserved PrP^C is not yet known and so alterations to its life cycle are undesirable. That said, studies with PrP^−/− knockout mice demonstrate that such mice develop normally (though they are somewhat inconclusive in regards to synaptic function) and some of the most promising results in prolonging the incubation time of prion disease in animal models have involved the ablation of PrP^C.²⁰ Additionally, a drugs ability to interfere with PrP^C may mean it also has potential in other neurodegenerative diseases where PrP^C is thought to play a role. For example PrP^C is implicated in Alzheimer diseases as a potential receptor for Aβ and antibodies directed against PrP^C have successfully inhibited the disruption of long-term potentiation (LTP) in rat hippocampus induced by Aβ-containing aqueous extracts of Alzheimer disease.²² PrP^C has even been implicated in mediating neurotoxic signaling by sensitizing cells to the toxic effects of β-sheet-rich conformers.²¹ As far as prion diseases are concerned, mPPIg5′s specific effect on PrP^Sc and lack of an effect on PrP^C appears to be the most promising outcome. While not necessarily the direct cause of neurodegeneration in prion diseases, PrP^Sc is certainly the main component of the infectious agent²³ and so a therapeutic that specifically eliminates PrP^Sc with minimum toxicity to the cell and with no detectable effect on endogenous PrP^C is, from a theoretical viewpoint at least, ideal.

Dendrimer Potential in Treatment of Other Neurodegenerative Diseases

In the realm of neurodegenerative disorders, the therapeutic potential of dendrimers is not just limited to prion diseases. In vitro studies have demonstrated that dendrimers can interact with various other polypeptides and that they have a particular affinity for peptides of amyloid structure.²⁴ This suggests dendrimers have potential against a wide range of disease associated proteins such as SOD1 mutants in Amyotrophic Lateral Sclerosis, α-synuclein aggregates in Parkinson disease and Htt protein in Huntington disease. Dendrimers have already been shown to have an in vitro effect on Aβ peptides. Aβ is the main constituent of amyloid plaques in Alzheimer disease and Aβ oligomers are believed to be neurotoxic. Studies have shown that various dendrimers (including mPPI) can inhibit the in vitro fibrilization of Aβ_1–28 (a fragment of Aβ) and reduce the toxicity of the fibrils formed.^18,25-27 In addition, phosphorous dendrimers have recently been shown to affect the aggregation of MAP-Tau protein.²⁷ Hyper-phosphorylated tau is a major component of the neurofibrillary tangles present in Alzheimer disease, frontotemporal dementia and the tauopathies.²⁸ The ability of dendrimers to effect the in vitro aggregation of Aβ and MAP-tau is very promising as an almost identical effect is observed for various dendrimers acting against PrP peptides.^2,17-19 This raises hopes that dendrimers may be able to accelerate the clearance and/or prevent the aggregation of a host of misfolded proteins intracellularly.

Conclusion

Dendrimer/prion interactions represent a fascinating area in neurodegenerative disease research which has allowed us to address many questions about the fundamental nature of prions. The therapeutic potential of dendrimers is immense due to the fact that numerous varieties target amyloid proteins and maltotriose modified PPI dendrimers have recently been shown to be capable of crossing the blood to brain barrier.²⁹

In our recent work we have demonstrated that anti-prion activity is not limited to cationic dendrimers and that a high density of reactive surface groups is a necessity. Furthermore we showed for the first time that a dendrimer is capable of inhibiting the intracellular conversion of PrP^C to PrP^Sc. Understanding how a drug works is a vital component in maximising its performance. By establishing that mPPIg5 interferes with the conversion of PrP^C to PrP^Sc, our study should help determine what other drugs may enhance this effect. Future work will require the use of animal models of prion disease to fully determine the potential of dendrimers in the treatment of prion and other neurodegenerative disorders.

Abbreviations:
Aβ	=	Amyloid Beta peptide
mPPI	=	maltose modified poly(propyleneimine) dendrimer
mPPIg5	=	maltose modified poly(propyleneimine) generation five dendrimer
PAMAM	=	poly(amidoamine) dendrimer
PEI	=	hyperbranched poly(ethyleneimine)
PPI	=	poly(propyleneimine) dendrimer
SOD1	=	superoxide dismutase 1
FL PrPSc	=	non-truncated protease resistant PrPSc. PrP27-30 (in this manuscript) = N-terminally truncated protease resistant PrPSc

Acknowledgments

This work was supported by University College Dublin, the Irish Research Council for Science, Engineering and Technology (IRCSET) the European Union (Dendrimers in Biomedical Applications—COST TD0802, Neuroprion—FOOD-CT-2004–506579), the Irish Department of Agriculture, Food and Rural Development (FIRM 01-R&D-D-160), the Saxon Ministry for Science and Art and The German Ministry for Education and Science, as well as the DFG SFB 596 (to Jorg Tatzelt).

Disclosure of Potential Conflicts of Interest

No potential conflict of interest was disclosed.