Acquisition of drug resistance and dependence by prions

Abstract

We have reported that properties of prion strains may change when propagated in different environments. For example, when swainsonine-sensitive 22L prions were propagated in PK1 cells in the presence of swainsonine, drug-resistant variants emerged. We proposed that prions constitute quasi- populations comprising a range of variants with different properties, from which the fittest are selected in a particular environment. Prion populations developed heterogeneity even after biological cloning, indicating that during propagation mutation-like processes occur at the conformational level. Because brain-derived 22L prions are naturally swainsonine resistant, it was not too surprising that prions which had become swa sensitive after propagation in cells could revert to drug resistance. Because RML prions, both after propagation in brain or in PK1 cells, are swainsonine sensitive, we investigated whether it was nonetheless possible to select swainsonine-resistant variants by propagation in the presence of the drug. Interestingly, this was not possible with the standard line of PK1 cells, but in certain PK1 sublines not only swainsonine-resistant, but even swainsonine-dependent populations (i.e. that propagated more rapidly in the presence of the drug) could be isolated. Once established, they could be passaged indefinitely in PK1 cells, even in the absence of the drug, without losing swainsonine dependence. The misfolded prion protein (PrP(Sc)) associated with a swainsonine-dependent variant was less rapidly cleared in PK1 cells than that associated with its drug-sensitive counterpart, indicating that likely structural differences of the misfolded PrP underlie the properties of the prions. In summary, propagation of prions in the presence of an inhibitory drug may not only cause the selection of drug-resistant prions but even of stable variants that propagate more efficiently in the presence of the drug. These adaptations are most likely due to conformational changes of the abnormal prion protein.

Figure 1. Different subclones of PK1 cells mediate susceptibility to swainsonine to different extents.

A. Genealogy of PK1-derived subclones. B. Standard Scrapie Cell Assay of brain-derived RML prions on AMO18, AMO10, PK1 and 2E4 cells in the absence (blue line) or presence (red, dashed line) of swainsonine (swa; 1 µg/ml). RIs are the reciprocals of the dilutions required to yield 750 PrP^res-positive cells per 20000 cells. Q_swa, the ratio of RI_cell vs. RI_cell+swa, indicates the inhibitory effect of swa on the propagation of RML prions in the individual cell lines.

Figure 2. Scheme displaying the emergence of swa-resistant and swa-dependent RML prions, and their transmission through various cell lines.

Initially, PK1 or PK1-derived cells were infected with RML prions, cultured in the presence of swa, and prions secreted into the conditioned medium (CM) were concentrated (CCM) and used to infect fresh batches of cells, also in the constant presence of swa. This cycle was repeated at least once more. The resulting prions were further propagated under various conditions. Large circles indicate cell lines; small circles indicate prions; horizontal arrows represent propagation of infected cells and vertical arrows transfer of prions; <#> indicates a prion sample whose swa resistance is reported in Table 1 and/or in one of the further figures. A. CCM from PK1 cells (large green disks), collected after the first infection (<1>) in the presence of swa (red arrows), contained low titers of swa-sensitive prions (small blue disks). After transfer to a fresh batch of PK1 cells in the presence of swa, infectivity dropped below detectability (<?>). B. AMO10 cells (large yellow disks) infected with RML in the presence of swa yielded swa-resistant prions (small red disks) (<2>); upon further transmission to AMO10 cells (large yellow disks) in the presence of swa the prions remained swa resistant; upon transmission to PK1 cells (large green disks) in the presence of swa the prions developed swa dependence. When transferred in the absence of swa the prions became swa sensitive (small, blue disks), semi-resistant (small orange disks) or swa dependent (small black disks) depending on whether they had previously been cultured with (red arrows) or without swa (blue arrows), and on the cell line (CAD cells violet disks). C. In the case of 2E4 cells, swa-dependent prions (small black disks) were recovered after the first transfer of CCM in the presence of swa (<3>); these prions remained swa dependent regardless of culture conditions or cell line.

Figure 3. Swa-resistant and swa-dependent RML prions can only be selected in certain cell sublines.

PK1 cells and cells of the PK1-derived sublines AMO10 and 2E4 were infected with RML prions, cultured in the presence of swa for 2 splits, and prions secreted into the conditioned medium (CM) were concentrated (CCM) and used to infect fresh batches of cells in the constant presence of swa (see scheme in Figure 2). CCM recovered from these cultures was analyzed by the SSCA on PK1 cells in the absence (blue line) or presence (red, dashed line) of swa. The proportion of PrP^res positive cells was plotted against the logarithm of the inocula dilutions and RIs were determined as reciprocal of the dilutions required to yield 750 PrP^res positive cells per 20000 cells. Q_swa, the ratio of RI_cell vs. RI_cell+swa, indicates the inhibitory effect of swa on the analyzed prion sample, to be compared with swa-sensitive, brain-derived RML prions (rightmost graph). Prions could not be detected in CM recovered from PK1 cells (<?>). Prions recovered from AMO10 cells (<2>) were found to be swa resistant, that is, they were propagated to the same extent in the presence or absence of swa. Prions recovered from 2E4 cells (<3>) were “swa dependent”, i.e. were propagated more efficiently in the presence of swa. n.a., not applicable.

Figure 4. Standard Scrapie Cell Assay of swa-resistant and swa-dependent RML prions propagated in the presence or absence of swa.

AMO10 and 2E4 cells were infected with RML prions, cultured in the presence of swa, and prions secreted into the conditioned medium (CM) were concentrated (CCM) and used to infect fresh batches of cells, also in the constant presence of swa. This cycle was repeated once more; the infected cells were cultured in the presence of swa for an extended period of time and then divided into two batches, of which one continued to be propagated in the presence of swa for nine splits while the other was propagated in parallel in the absence of swa (see Figure 2B, C). CCM from these cultures was then analyzed by the SSCA on PK1 cells in the absence (blue line) or presence (red, dashed line) of swa. RIs are the reciprocals of the dilutions required to yield 1000 PrP^res positive cells per 20000 cells. Q_swa = RI_cell/RI_cell+swa indicates the inhibitory effect of swa. A. Swa-resistant AMO10-derived prions that were constantly propagated in the presence of swa (<13>) remained swa resistant, but reverted to semi-resistance when propagated for nine splits in the absence of swa (<12>). B. Swa-dependent 2E4-derived prions remained swa dependent whether propagated for 9 splits in the presence of swa (<16>) or in its absence (<17>). C. Swa-sensitive, brain-derived RML prions. Q_swa for RML^brain was unusually high in this assay and that of Figure 5B, leading to the low Q_rel values shown in Table 1.

Figure 5. Standard Scrapie Cell Assay of swa-resistant and swa-dependent RML prions transferred to PK1 cells and propagated in the presence or absence of swa.

AMO10 and 2E4 cells were infected with RML prions, cultured in the presence of swa, and prions secreted into the conditioned medium (CM) were concentrated (CCM) and used to infect fresh batches of cells in the constant presence of swa. This cycle was repeated once more, whereupon the prions were transferred to PK1 cells in the presence or absence of swa (Figure 2B, C). CCM was analyzed by the SSCA on PK1 cells in the absence (blue line) or presence (red, dashed line) of swa. A. Swa-resistant AMO10-derived prions developed swa dependence when propagated in PK1 cells in the presence of swa (<4>); when propagated in PK1 cells in the absence of swa, they reverted to swa sensitivity (<5>). In contrast, swa-dependent 2E4-derived prions remained swa dependent in PK1 cells, whether they were propagated in the presence (<7>) or absence (<8>) of swa. RIs are the reciprocals of the dilutions yielding 500 PrP^res positive cells per 20000 cells. Q_swa = RI_cell/RI_cell+swa reflects inhibition by swa and may be compared to the value for swa-sensitive, brain-derived RML prions. B. AMO10- or 2E4-derived prions that had acquired or retained swa dependence after propagation in PK1 cells in the presence of swa were either continuously propagated in the presence of swa (<10>, <14>) or cultured for eleven splits in the absence of swa (<11>, <15>). Once swa dependent, they remained dependent, whether propagated in the presence or absence of swa. RIs are the reciprocals of the dilutions required to yield 1000 PrP^res positive cells per 20000 cells. Q_swa for RML^brain is unusually high in this assay and that of Figure 4C, leading to the low Q_rel values shown in Table 1.

Figure 6. Standard Scrapie Cell Assay of swa-resistant and swa-dependent RML prions transferred to CAD cells in the absence of swa.

Swa-resistant AMO10-derived prions and swa-dependent 2E4-derived prions were transferred to CAD cells and propagated in the absence of swa for 8 splits. Concentrated conditioned medium from these cultures was then analyzed by the SSCA on PK1 cells in the absence (blue line) or presence (red, dashed line) of swa. RIs are the reciprocals of the dilutions yielding 1000 PrP^res-positive cells per 20000 cells. Q_swa = RI_cell/RI_cell+swa reflects inhibition by swa and may be compared with the value for swa-sensitive, brain-derived RML prions (rightmost panel). Swa-resistant AMO10-derived prions reverted to swa sensitivity (<18>), while swa-dependent 2E4-derived prions remained dependent (<19>).

Figure 7. Standard Scrapie Cell Assay of swa-resistant and swa-dependent RML prions in the presence of various prion inhibitors.

Swa-resistant AMO10-derived prions (<6>), swa-dependent 2E4-derived prions (<9>) and swa-sensitive, brain-derived prions were analyzed by the SSCA on PK1 cells in the absence (blue line) or presence (red, dashed line) of inhibitor. RIs are the reciprocals of the dilutions yielding 1000 PrP^res-positive cells per 20000 cells. Q_drug = RI/RI_+drug indicates the inhibitory effect of a drug on the RI of a particular prion isolate (in PK1 cells) and may be compared with the Q_drug value of brain-derived RML prions. A. While swa-resistant AMO10-derived prions are also resistant to kifunensine (kifu), they are semi-resistant to compound B (cpdB) and sensitive to curcumine (curc) or Congo Red (CR). B. Swa-dependent, 2E4-derived prions are resistant to kifu, curc and cpdB, but inhibited by CR. C. The propagation of swa-sensitive brain-derived RML prions, used as control, is strongly inhibited by kifu, curc and CR.

Figure 8. Standard Scrapie Cell Assay in the presence of various prion inhibitors (ECPA) of AMO10- and 2E4-derived RML prions after propagation in mice.

Swa-resistant AMO10-derived prions, swa-dependent 2E4-derived prions and authentic brain-derived RML prions were propagated in C57BL6 mice and analyzed by ECPA on PK1 cells in the absence (blue line) or presence of either swa (red, dashed line) or kifu (green, dashed line). RIs are the reciprocals of the dilutions required to yield 500 PrP^res positive cells per 20000 cells. Q_drug = RI/RI_+drug indicates the inhibitory effect of the drug on the RI of a prion sample and may be compared with the Q_drug value of brain-derived RML prions. A. Brain-passaged AMO10-derived prions (<AMO10-brain-a>, <AMO10-brain-b>) were sensitive to swa, but semi-resistant to kifu. B. Brain-passaged 2E4-derived prions were inhibited by swa, however, one sample (<2E4-brain-c>) contained prions that were semi-resistant to kifu, one sample (<2E4-brain-b>) contained kifu-sensitive prions, and one sample (<2E4-brain-a>) contained prions that were strongly inhibited by kifu. C. Two experimental control brains (brain1 and brain2) of mice that were inoculated with authentic RML prions and a homogenate of pooled brains (lab stock) were analyzed as controls; the prions were swa-sensitive and strongly inhibited by kifu.

Figure 9. Depletion of PrP^res following inhibition of PrP^res synthesis in PK1 cells infected with either RML^brain or RML^{(AMO10+swa)2/(PK1+swa)14x}.

PK1 cells were infected with brain-derived, swa-sensitive RML prions or swa-dependent prions derived from RML propagated sequentially in AMO10 cells and PK1 cells in the presence of swa (RML^{(AMO10+swa)/(PK1+swa)14x}), grown for 8 days and seeded into two 15-cm dishes to one of which 2 µg swa/ml were added. Three days after swa addition, the cells from each plate were transferred to ten 15-cm dishes, resulting in ten dishes with and ten without swa. One day after the third split, two dishes were harvested (time zero), 1 unit PIPLC/ml media was added to four dishes of each condition, and one dish per condition was harvested after 7, 12,18 and 24 hours. Cells were counted at each time point and equal amounts of protein were subjected to western blot analysis to determine PrP^res levels. From PrP^res level changes and growth rate PrP^res synthesis and PrP^res clearance were calculated as shown in Table 3. A. Western-blots showing PrP^res signals at indicated time points. Samples were loaded in triplicate gels; one representative blot is shown. B. After correction for differences in luminescence intensity between gels, the intensity of PrP^res at each time point was normalized to the PrP^res intensity at the 0 hour time point and plotted, beginning at the 7 hour time point, as a function of time on a log2 scale. C. In order to determine growth rate and partitioning rate (see Table 3), total cell numbers from each time point were plotted as a function of time on a log2 scale.

Figure 10. Hypothetical mechanisms for prion mutation and cell tropism.

A. The conformation of a PrP^Sc unit in an aggregate may be “locked” even if it is not the thermodynamically most favored one. The shorter the aggregate, the more the conformation of its subunits may be subject to thermal fluctuation; accretion of PrP^C stabilizes the variant conformation, resulting in “mutation” of the prion. B. PrP^C may assume a variety of conformations, of which only one or a few can readily accrete to a particular seed. Different conformers may be present at different steady state concentrations and could be favored by the nature of the N-glycans and/or by interaction with cellular factors (such as small RNAs). Cell tropism may reflect the preponderance of different sets of conformations in distinct cell types. Panel A is from ref. and Panel B from ref. .