INTRODUCTION
Malaria, particularly disease caused by
Plasmodium falciparum, remains a huge problem, and its control is challenged by resistance to most available therapies (
1). New antimalarial drugs, ideally directed against new targets, are needed. One potential target is the proteasome, a protein degradation complex of eukaryotes and actinomycetes that degrades proteins targeted for removal after tagging with ubiquitin or other protein tags (
2). A proteasome complex is composed of multi-subunit structures, generally a 20S core particle and one or two 19S regulatory particles on one or both ends of the 20S barrel (
2). The 19S regulatory particle recognizes ubiquitinated proteins and transfers them to the 20S core particle for degradation (
2). The proteasome has been validated as a target for drugs to treat multiple myeloma and mantle cell lymphoma, and three proteasome inhibitors, bortezomib, carfilzomib, and ixazomib, are approved for these indications (
3). Additionally, the proteasomes of pathogens have been identified as potential targets for drugs against a number of infectious diseases, including those caused by bacteria and protozoan parasites (
4).
In eukaryotes, the 20S core particle contains 14 α and 14 β subunits arranged in a α7β7β7α7 barrel-shaped complex, with 7 different α subunits, including the α-type-6 subunit, forming the outer ring, and 7 different β subunits forming the inner rings. Outer rings regulate the entrance of protein substrates, and proteolytic activity exists in the β rings (
5). Only β1, β2, and β5 are catalytically active, with caspase-like, trypsin-like, and chymotrypsin-like proteolytic activities, respectively (
6), and the four noncatalytic subunits (β3, β4, β6, and β7) function in assembly and provide structural support for the proteasome complex (
7). The RPN10 subunit in the 19S regulatory particle is an essential canonical ubiquitin receptor that recognizes multiubiquitin chains (
8). The C-terminal ubiquitin interacting motif of this subunit identifies ubiquinated substrates to shuttle to the proteasome (
2,
5). The 19S regulatory particle contains a 19S PfRPT4 subunit, which is an ATPase located at the base of the 19S particle that interacts with the α-ring of the core particle (
5). PF3D7_0808300 encodes a putative
P. falciparum ubiquitin regulatory protein (
9); increased copy number was associated
in vitro with low grade resistance to WLL-vinyl sulfone (WLL-VS) and WLW-VS (
10).
The
P. falciparum proteasome is similar to that of other eukaryotes, with moderate homology between human and
P. falciparum catalytic subunits (27% identity for β1, 53% for β2, and 51% for β5; [
11]), but differences in structure and substrate specificity have facilitated the design of
P. falciparum-targeted inhibitors. Among proteasome inhibitors that have been studied for the treatment of malaria are asparagine ethylenediamines (
12), peptide vinyl sulfones (
13), and peptide boronates (
14,
15). Proteasome inhibitors are typically selective for a particular catalytic subunit, but some can act against multiple subunits (
12,
16). Asparagine ethylenediamines are noncovalent, reversible inhibitors that specifically target the β5 subunit (
12). Peptide boronates are covalent and slowly reversible inhibitors, and a validated class, with two FDA approved drugs, bortezomib and ixazomib (
3,
15). Peptide vinyl sulfones are covalent and irreversible inhibitors, including WLL-VS, which targets the β2 and β5 subunits, and WLW-VS, which targets the β2 subunit (
17). Peptide epoxyketones are another validated class of proteasome inhibitors, with one FDA-approved drug (
18). Compounds from all of these classes have been developed to selectively target the
P. falciparum proteasome over its human counterpart (
17,
19,
20).
In considering the P. falciparum proteasome as a target for antimalarials, it is important to characterize naturally occurring variation in inhibitor susceptibility among circulating parasites. We therefore measured ex vivo susceptibilities of fresh Ugandan isolates to a range of proteasome inhibitors. We also sequenced target P. falciparum proteasome subunits and explored associations between identified genetic polymorphisms and inhibitor susceptibilities of Ugandan parasites.
DISCUSSION
We studied susceptibilities of fresh Ugandan P. falciparum isolates to a panel of proteasome inhibitors and searched for associations between susceptibilities and genetic polymorphisms in proteasome targets. A number of the inhibitors had potent activity against P. falciparum isolates circulating recently in Uganda. A small number of genetic polymorphisms were identified in the proteasome subunits of the Ugandan isolates. The S214F mutation in the β2 subunit was associated with decreased susceptibility to peptide boronates, but not other tested classes of proteasome inhibitors, although the infrequency of this mutation limited the analysis. Overall, a number of proteasome inhibitors offered potent activity against P. falciparum circulating in Uganda, and genetic variation in the β2 and β5 subunit targets was uncommon.
The 7 tested proteasome inhibitors mostly showed potent
ex vivo activity, with values similar to those previously seen against cultured
P. falciparum laboratory strains for asparagine ethylenediamine (
12,
21), macrocyclic peptide, and peptide boronate inhibitors (
21,
22). Asparagine ethylenediamines and macrocyclic peptides had median IC
50s <50 nM against the Ugandan isolates, suggesting potential as antimalarial lead compounds. Peptide boronates had varied activity, with median IC
50 values of 56 to 668 nM against the Ugandan isolates. Among the isolates studied, strong correlations were seen between results for chemically related inhibitors, suggesting shared determinants of activity.
Previous studies showed selection for a number of polymorphisms in different proteasome subunits in
P. falciparum laboratory strains incubated with proteasome inhibitors (
10,
12,
15,
22). To assess genotype variability in circulating Ugandan isolates, we sequenced the β2 and β5 catalytic proteasome subunits, which are predicted inhibitor targets, and 5 regulatory subunits. We identified a small number of polymorphisms in the β2 and β5 subunits, and we searched for associations between these and other polymorphisms and
ex vivo susceptibilities. The two identified mutations in the β2 subunit, S214F and I204T, were both found in isolates with mixed genotypes. Isolates with one of these mutations, β2 S214F, had decreased susceptibility to two peptide boronates, MMV1579506 and MMV1794229, but the small number of available mutant parasites limited the analysis. The S214F mutation is located in the C-terminal tail of the β2 subunit, which wraps around the β3 subunit (
Fig. 3; [
23]), but the basis of altered inhibitor sensitivity is unclear. Two identified mutations in the β5 subunit, A142S and D150E, were not associated with alterations in susceptibility to proteasome inhibitors. Two mutations in the RPN10 subunit and one mutation in PF3D7_0808300 were common, but these were not associated with altered susceptibility to tested compounds. No polymorphisms were seen in the 19S PfRPT4 subunit, the α-type-6, subunit, or the fourth exon of the β6 subunit, in which mutations have previously been selected by proteasome inhibitors
in vitro (
10,
12,
22).
To further study the relevance of the β2 214F mutation, which was identified in Ugandan isolates, we assessed susceptibilities of culture adapted Ugandan strains to multiple proteasome inhibitors. β2 214F mutant parasites had decreased susceptibility to MMV1579506 and MMV1794229 compared to that of W2 strain and laboratory-selected β2 49E mutant parasites, but there were no significant differences in susceptibilities between β2 S214F wild type and mutant Ugandan strains. There were no significant differences in susceptibilities to other tested inhibitors between β2 wild type and mutant parasites.
This study had important limitations. First, different proteasome inhibitors were available for ex vivo analysis for only limited time frames, limiting our sample size for key compounds. One class, the peptide vinyl sulfones, was not available until after evaluation of field isolates, and therefore was only studied against selected culture adapted clones. Second, only small numbers of isolates with mutations in the β2 or β5 subunit were identified, limiting our ability to characterize the impacts of these polymorphisms. Third, most mutant isolates had mixed wild type/mutant genotypes, as typical in a region with high multiplicity of infection, further limiting our ability to characterize impacts of mutations.
In summary, multiple ethylenediamine and peptide boronate proteasome inhibitors demonstrated potent activity against Ugandan P. falciparum isolates. Modest variations in activities were generally not explained by polymorphisms seen in the β2, β5, RPN10, and PF3D7_0808300 subunits, although the β2 S214F mutation was associated with decreased ex vivo activity of two inhibitors in the few isolates with this mutation. Importantly, resistance-mediating mutations previously selected in vitro were not seen in Ugandan isolates. Overall, our results add confidence that, although studied proteasome inhibitors can select for resistance in vitro, they are likely to show consistently strong activity against P. falciparum isolates now circulating in East Africa.
ACKNOWLEDGMENTS
This work was supported by grants from The National Institutes of Health (AI139179 to P.J.R., AI075045 to P.J.R., AI143714 to G.L., T37MD003407 to U. California, Berkeley) and the Medicines for Malaria Venture (RD-15-0001 to P.J.R.). We thank study participants and staff members of the clinics where samples were collected. We also thank Matthew Bogyo, Stanford University for generously providing compounds for study and Barbara Stokes and David Fidock, Columbia University, for providing the V1/S K13C580Y strain with the β2 49E mutation.
S.G., O.K., S.C., P.K.T., O.B., M.O., S.O., T.K., and R.A.C. assisted in study design, performed ex vivo IC50 assays, and archived data. M.D. and J.L. provided project administrative and logistical support. S.G., O.K., M.D.C., O.A., and J.A.B. performed and analyzed genotyping studies. G.L. assisted in generating the protein model. S.G., O.K., R.A.C., and B.R.B. verified and analyzed data and performed statistical analysis. R.A.C., S.L.N., L.A.K., G.L., and P.J.R. provided guidance in study design and intellectual support. S.G., O.K., and P.J.R. wrote the draft of the manuscript and all authors corrected and approved the final manuscript.
M.D. is employed by Medicines for Malaria Venture (MMV), who cofunded this study.