NR AETD
AU Gordon,I.; Abdulla,E.M.; Campbell,I.C.; Whatley,S.A.
TI Phosmet induces up-regulation of surface levels of the cellular prion protein
QU Neuroreport 1998 May 11; 9(7): 1391-5
PT journal article
AB Chronic (2 day) exposure of human neuroblastoma cells to the organophosphate pesticide phosmet induced a marked concentration-dependent increase in the levels of PrP present on the cell surface as assessed by biotin labelling and immunoprecipitation. Levels of both phospholipase C (PIPLC)-releasable and non-releasable forms of PrP were increased on the plasma membrane. These increases appear to be due to post-transcriptional mechanisms, since PrP mRNA levels as assessed by Northern blotting were unaffected by phosmet treatment. These data raise the possibility that phosmet exposure could increase the susceptibility to the prion agent by altering the levels of accessible PrP.
VT
Key words: BSE; Organophosphates; Prion diseases; Prion protein
Introduction
Prion diseases are a group of neurodegenerative disorders which include Creutzfeldt-Jakob disease (CJD) in man, scrapie in sheep and bovine spongiform encephalopathy (BSE) in cows. These are characterized by similar neuropathological features and transmissibility both within and between species.[1] Both the transmissibility and the pathology of prion diseases have been shown by a large body of research to be dependent on the cellular prion protein (PrPc) an extracellular phosphatidylinositol-anchored glycoprotein.[1] The conversion of PRPC protein to PRPSC, its disease-specific isoform, is a key process in disease development and involves conformational change.[2]
The appearance of BSE in the mid-1980s, apparently through cross-contamination from sheep products, raised fears regarding the safety of beef and bovine products for humans; this has been emphasized by the more recent emergence of a new variant form of human CJD (vCJD) which is related to BSE in terms of 'strain' type.[3,4] The exact causes of the BSE epidemic, and the mechanisms of species cross-contamination are therefore of great importance to human health. The hypothesis that BSE arose from ingestion of scrapie-contaminated meat and bone meal products (MBM) in the early 1980s has gained widespread acceptance and was based on the significant correlation between MBM ingestion and the incidence of BSE.[5] Purdey,[6] however, has proposed an alternative hypothesis that widespread use of organophosphate pesticides may have acted as an environmental trigger to the BSE epidemic. He argues that BSE incidence correlates both temporally and geographically with systemic, high-dose usage of one in particular of this class of compounds, phosmet. Although Purdey's original hypothesis proposes that phosmet may act directly on PrPc to result in the emergence of the pathogenic PrPsc form, an alternative hypothesis is that, if involved in the BSE outbreak, phosmet could have acted indirectly by altering susceptibility to the causal agent. This could occur by several mechanisms; for example by altering either the intracellular distribution of PrP or its levels in the cell. The purpose of this research has been to investigate the effects of phosmet on cellular prion protein in a cell culture system. We report the effects of subtoxic doses of phosmet on levels of PrP on the plasma membrane and on PrP mRNA levels in human neuroblastoma SKNSH cells.
Materials and Methods
All cell culture reagents were obtained from the Sigma Chemical Company, Dorset, UK. Monoclonal antibody-producing cell line 611 (raised against hamster scrapie fibrils) was a gift from Professor S.B. Prusiner.
Human SKNSH neuroblastoma cells were grown in a humidified 5% CO2/95% air atmosphere in Dulbecco's modified Eagle's medium supplemented with 5% (v/v) fetal bovine serum, 5% (v/v) horse serum and penicillin/streptomycin at 100 units/ml.
All cultures were treated and harvested at confluence. Cells were exposed to phosmet (dimethyl-S-phthalimido-methyl phosphorothiodothionate, Imidan, Phthalophose, obtained from Philip Harris, UK) at either 2 or 12 ppm for 48 h.
Biotinylation of plasma membrane proteins was performed essentially as described by Shyng et al.[7] All procedures were performed at 4°. After washing in Dulbecco's phosphate buffered saline (DPBS, Sigma), cells were biotinylated with sulfo-N-hydroxysuccinimide-biotin (Sigma) at 1.5 mg/mi for 30 min followed by washing four times with 30 mM glycine in PBS. Cells were then incubated with phosphioinositide-specific phospholipase C (PIPLC, Sigma) at a concentration of 1 unit/ml for 2 h in PBS to cleave phosphioinositide anchors. PIPLC-released protein was removed and remaining cells were lysed by addition of 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, 0.5% (w/v) Triton-X-100 and 0.5% (w/v) sodium deoxycholate. After clearing (1000 x g, 10 min), proteins were immunoprecipitated from the incubation medium and from the cell lysates by addition of protein A-sepharose (Sigma) together with 5 µl ascites fluid containing monoclonal antibody 611[8] at 4° overnight. After washing three times in 50 mM Tris-HCI, pH 7.4, the immunopurified proteins were extracted from the beads by boiling in SDS gel sample buffer.[9]
Protein estimations were performed on initial cell lysates using a commercially supplied Coomassie blue dye-based protein reagent (Pierce, UK) according to manufacturers instructions. Biotinylated protein preparations were separated on 10% (w/v) polyacrylamide gels[9] based on equal equivalent amounts of original total cellular protein. After blotting onto Immobilon membranes (Whatman, UK), Western analysis[9] was performed using an enhanced chemiluminescence detection system (Amersham, UK) according to manufacturers instructions. Detection was obtained using streptavidin-conjugated horseradish peroxidase (HRP).
Plasma membranes were purified essentially as described for cultured cells by Butters and Hughes.[10] All procedures were performed at 4°. Neuroblastoma cells were washed twice with calcium and magnesium-free Dulbecco's phosphate buffered saline (CMFDPBS Sigma), detached from the dish surface in CMFDPBS containing 1 mM EDTA and harvested by low speed centrifugation (750 x g, 10 min). Pellets were resuspended and homogenized by five strokes of a teflon/glass homogenizer (500 r.p.m., 0.2 mm clearance) in 250 mM sucrose supplemented with 1 mM EDTA, 0.1 mM phenylmethyl sulfonyl fluoride, leupeptin (1 µg/ml) and pepstatin (1 µg/ml). After low speed centrifugation (10 min, 750 x g) membranes were pelleted by centrifugation at 17,000 x g for 60 min. Membranes were resuspended by fresh homogenisation and layered on top of stepped gradients containing 1.3 M and 0.9 M sucrose. After centrifugation at 100,000 x g for 1 h, the interface band was harvested, diluted two-fold with distilled water and recentrifuged for 30 min at 100,000 x g. Western blotting was performed following protein estimations (100 µg plasma membrane protein per lane) using antibody 611(0.5 µg IgG/ml) and HRP-conjugated secondary antibody according to manufacturers' protocols.
Total cellular RNA was extracted by the acid phenol guanidinium thiocyanate method of Chomczynmski and Sacchi.[11] Total cellular RNA (5 µg) was electrophoresed on denaturing formaldehyde agarose gels by standard protocols[9] and blotted onto nylon membranes (Hybond, Amershan). Probing of blots was performed again using standard protocols, using a radiolabelled purified fragment of the human PrP coding sequence (nucleotides 281-797 of EMBL sequence M13899, corresponding to the C-terminal 160 amino acids of the prion protein). Blots revealed a single band equivalent to the known size of PrP mRNA (results not shown).
Image analysis of Northern and Western blots was performed using an interactive computerized image facility (IBAS 2000, Kontron Instruments, Watford, UK). The integrated grey value, a function of area and optical density was calculated for each band or area in question after subtraction of the background grey value.
Results
Preliminary experiments on the effect of phosmet on human SKNSH neuroblastoma cells established that at the concentrations used there was no inhibition of cell growth in logarithmic phase (results not shown). All the effects of phosmet observed here are therefore assumed to be sub-cytotoxic. Confluent SKNSH cells were treated with phosmet for 48 h. Figure 1 shows the results of Western blotting of biotinylated cell surface proteins immunoprecipitated with antiPrP antibody which are released into the medium by PIPLC digestion under control conditions and after treatment with phosmet. In agreement with other reports,[12] PrPc migrates as a heterogeneous set of bands between 30 and 39 kDa in polyacrylamide gels. Visual inspection indicates that phosmet elicits a marked concentration-dependent increase in the amount of labelled PrP released from the cells.
Densitometric quantitation of the PrP area indicates an increase of ~5-10-fold (Fig. 1) after treatment at 12 p.p.m. In addition, after treatment at the higher phosmet concentration there was a reproducible shift in the distribution of PrPc isoforms from predominantly high and low mass isoforms to forms which were more evenly spread through the size range.
Other authors report that the proportion of PrP which releasable by PIPLC is variable and dependent on the cell line.13 Since PIPLC-resistant PrP may be related to either chemical modification or changes in the physical state on the membrane, we also assessed the PIPLC-resistant fraction of labelled PrP (Fig. 2). This figure also illustrates a concentration-dependent increase of PrP in the non-releasable fraction which reaches ~5-10-fold. This represents a similar increase to that seen in the releasable fraction, and we therefore cannot find evidence for a change in the relative proportion of releasable and non-releasable PrP in the cells. Changes in the distribution of PrP isoforms due to phosmet treatment are also visible in this non-released PrP fraction which are comparable to changes seen in the PIPLC-releasable fraction.
In addition to surface labelling techniques, plasma membrane fractions were also prepared using sucrose gradients. Figure 3 shows the effects of phosmet on PrPc on the plasma membrane fractions isolated from control and treated cells. Results again indicate a concentration-dependent increase in PrP detectable in the plasma membrane, though this was less marked than for surface-labelled protein, being just under 2-fold at 12 p.p.m. phosmet.
The increase in PrPc observed in the accessible fraction may be due either to selective increase in PrP on the membrane, or to a generalized increase in the population of phosphionositide (PI)-anchored proteins. Comparison of samples relative to yield of PIPLC-released protein, however, gave essentially identical results (not shown), indicating that the effect is selective for PrP within the PI-anchored protein population. This observation also argues against generalized increases in amounts of plasma membrane per cell after treatment. Measurement of either total protein per cell or plasma membrane yield also indicated no significant changes after treatment (results not shown).
Several mechanisms may account for the effects of phosmet on surface PrP of cultured cells: these include changes in either synthesis and degradation of the protein, or alterations in its intracellular distribution. Since expression is usually linked to relative mRNA levels, these were measured by Northern blotting procedures after extraction of total cellular RNA from cultures. No significant differences were found between PrP mRNA levels in untreated cultures and those treated with either 2 p.p.m. or 12 p.p.m. phosmet (14.9 +/- 3.0, 15.8 +/- 2.3 and 15.5 +/- 4.4, respectively, expressed as arbitrary units +/- standard deviations from three separate experiments). The increase seen here must therefore be due to post-transcriptional mechanisms.
FIG. 1. Effects of phosmet treatment on PIPLC-releasable PrP of SKNSH cells. Treatment and surface labelling were performed as described in Materials and Methods. After PIPLC digestion, released protein was immunoprecipitated and Western blotting was performed as described. (a) Untreated control cultures: (b) 2 p.p.m. phosmet; (c) 12 p.p.m. phosmet. Figures denote migration of molecular mass markers. Graph represents amount of PrP in arbitrary units measured by image analysis of two independent experiments.
FIG. 2. Effects of phosmet treatment on non-releasable PrP of SKNSH cells. Treatment and surface labelling was performed as described in Materials and Methods. After PIPLC digestion, cells were lysed, PrP immunoprecipitated and Western blotting was performed as described in Materials and Methods. (a) Untreated control cultures; (b) 2 p.p.m. phosmet; (c) 12 p.p.m. phosmet. Figures denote migration of molecular mass markers. The graph represents amount of PrP in arbitrary units per unit protein, and hence cell, as measured by image analysis of two independent experiments.
FIG. 3. Western blot of plasma membrane preparations. Membrane preparations and Western blotting was performed as described in Materials and Methods. (a) Untreated control cultures; (b) 2 p.p.m. phosmet; (c) 12 p.p.m. phosmet. Graph represents amount of PrP in arbitrary units measured by image analysis of two separate experiments.
Discussion
The hypothesis that BSE arose from ingestion of scrapie-contaminated meat and bone meal products (MBM) in the early 1980s has gained widespread acceptance and was based on the significant correlation between MBM ingestion and the incidence of BSE.[5] Several uncertainties remain, however, about the exact relationship between MBM ingestion and BSE incidence, highlighted by the surprisingly low reported incidence of BSE in European countries despite the large amounts of suspect MBM exported to them, and the large number of BSE-affected cattle born after the UK ban on animal-derived offal protein entering the ruminant food chain.[6,14-16] These uncertainties have allowed several alternative hypotheses of BSE aetiology to emerge, such as the notion that BSE is a modification of a pre-existent cattle illness.[14] In contrast, Purdey[6] has hypothesied that organophosphate pesticides may act as an environmental trigger and argues that the incidence of BSE correlates both temporally and geographically with systemic, high-dose phosmet usage. He further proposes that phosmet is able to cause modifications of PrPc to produce its pathogenic form, PrPsc.
It has been established by a large body of evidence that the development of prion disorders in both animal models and human illness is dependent on both the sequence and the expression of PrP,[1,17,18] which has been reported to localize almost exclusively on the cell surface.[13] We report that phosmet treatment causes a dose-dependent increase in PrPc of the plasma membrane of SKNSH human neuroblastoma cells. The effect is more pronounced in the surface-biotinylated PrP fraction than in the plasma membrane fraction which was harvested by sucrose gradient purification. The reasons for this discrepancy might relate to a number of factors including the differences in quantitation due to the use of either streptavidin~biotin binding or antibody-antigen binding to detect PrP on the blot, as well as the relative impurity of plasma membrane preparations. Indeed, others have shown that chick PrP expressed in mammalian cells cycles between the cell surface and an early endocytic compartment,[19] which contains approximately half of the total PrP. This endocytic compartment would be present in the membrane preparations described here, and may therefore mask changes in external, cell surface PrP. The upregulation of PrP observed here may arise by several mechanisms, including differences in the kinetics of synthesis and degradation or to slowing of the rate of endocytosis. Redistribution of cellular compartments is not without precedent, as organophosphates have been reported by others to cause a redistribution of cellular compartments from the Golgi apparatus to the external membrane.[20] We have not been able to detect changes in PrP levels in a crude microsomal fraction of the cells (results not shown). Changes in the distribution between the external membrane and the endocytic compartment may, however, be suggested by our data.
The data presented in this study may be predicted to be relevant to the transmission of prion disorders. First, other data indicate that the incubation period of experimentally transmitted prion diseases is inversely proportional to the gene dosage of PrP.[18] This appears to be the case at expression levels of PrPc close to normal, as evidenced by the doubling of incubation times when gene dosage is halved.[21] Second, the effects of phosmet reported here may be contrasted with the effects of polyanions which have been shown to delay the progression of prion disorders. Thus, pentosan sulphate, congo red and related compounds have been shown to reduce the amount of PIPLC-releasable PrP in cultured cells by stimulation of endocytosis[7] and to inhibit prion replication both in vitro and in vivo.[22-24] The increase in the levels of PrP reported here, if replicated in vivo, would therefore be expected to increase the risk to illness on challenge by the transmissible agent and also to shorten incubation time. While the results presented here, therefore, provide no evidence for the hypothesis that phosmet may directly cause emergence of BSE through de novo production of PrPsc, it represents the first experimental evidence that phosmet could modify susceptibility to the prion disease agent. Finally, 'strain' type in prion diseases has been shown to correlate with the isoform distribution of protease-resistant PrP, which may be due to protein glycosylation patterns.[3,25] It is of interest that treatment with phosmet appears to change the PrPc isoform distribution and tempting to hypothesize that phosmet use may have contributed to the emergence of the new 'strain' of prion disease represented by BSE and vCJD.[3,4] The BSE outbreak may have ultimately arisen from a number of factors unique to the UK but the experiments presented here suggest that further investigation of the relationship between these observed changes and the effects of phosmet in vivo would be valuable.
Conclusion
We report an increase in the externally accessible fraction of PrP of cultured human SKNSH neuro-blastoma cells after phosmet treatment. These data are the first experimentally defined association between phosmet treatment and PrP and may be relevant to the risk for prion illnesses on exposure to the prion agent. Further research is needed to establish whether this represents a common effect of the class of organophosphate pesticides as well as the relevance of these phenomena to the UK BSE epidemic.
References
1. Prusiner SB. Annu Rev Microbiol 48, 655-686 (1994).
2. Horwich AL and Weissman HJ. Cell 89, 499-510 (1997).
3. Collinge J, Sidle KC, Meads J et al. Nature 383, 685-690 (1996).
4. Bruce ME, Will RG, Ironside JW et al. Nature 389, 498-501 (1997).
5. Bradley R and Wilesmith JW. Br Med Bull 49, 932-959 (1993).
6. Purdey M. Med Hypoth 46, 429-454 (1996).
7. Shyng SL, Lehmann S, Moulder KL and Harris DA. J Biol Chem 270, 30221-30229 (1995).
8. Whatley SA, Powell JF, Politopoulou G et al. NeuroReport 6, 2332-2337 (1995).
9. Sambroke J, Fritsch EF and Maniatis T. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, Cold Spring Harbor Press (1989).
10. Butters TD and Hughes RC. Biochem J 140, 469-478 (1994).
11. Chomczynmski P and Sacchi N. Anal Biochem 162, 155-159 (1987).
12. Borchelt DR, Scott M, Taraboulous A et al. J Cell Biol 10, 743-752 (1990).
13. Stahl N, Borchelt DR and Prusiner SB. Glycolipid anchors ol the cellular prion protein. In: Turner AD. ed. Molecular and Cell Biology of Membrane Proteins. Horwood. Herts, UK, 1990: 189-216 (1990).
14. Collee JG and Bradley R. Lancet 349, 635-641 (1997).
15. Collee JG and Bradley R. Lancet 349, 715-721 (1997).
16. Butler D. Nature 382, 4 (1996).
17. Prusiner SB. Ann Neurol 35, 385-395 (1994).
18. Prusiner SB, Scott M, Foster D et aL Cell 63, 673-686 (1990).
19. Shyng SL, Huber MT and Harris DA. J Biol Chem 268, 15922-15928 (1993).
20. Medda S, Stevens AM and Swank RT. Cell 50, 301-310 (1987).
21. Büeler H, Aguzzi A, Sailer A et al. Cell 73, 1339-1347 (1993).
22. Diringer H and Ehlers BJ. Gen Virol 72, 457-60 (1991).
23. Caughey B and Race RE. J Neurochem 59, 768-771 (1992).
24. Caughey B and Raymond GJ. J Virol 67, 643-650 (1993).
25. Bessen RA and March RF. J Virol 66, 2095-2101 (1992).
ACKNOWLEDGEMENTS: We would like to thank Mark Purdey and David Male for helpful discussion. This work was funded by the Independent Research Fund, the Mark Purdey Research Fund, the Network for Social Change, and a donation from A.M. Walker.
Received 3 February 1998;
accepted 15 February 1998
IN In der menschlichen Neuroblastomzelllinie SKNSH wurde nur eine Bande der mRNA für das Prionprotein gefunden. Zumindest in der logarithmischen Phase wurde die Vermehrung der Neuroblastomzellen durch das Insektizid Phosmet in den verwendeten Konzentrationen von bis zu 12 ppm nicht beeinträchtigt. Es erwies sich also in dieser Konzentration als nicht cytotoxisch. Im Western blot waren die Banden des mit Phospholipase freisetzbaren und des übrigen auf der Zelloberfläche markierbaren Prionproteins nach zweitägiger 12 ppm Phosmet-Behandlung densitometrisch jeweils 5-10-fach stärker. Außerdem waren nach der 12 ppm Phosmet-Behandlung die Banden mit den höchsten und den niedrigsten scheinbaren Molekularmassen zugunsten mittlerer Molekularmassen weniger dominant. Bei ohne Markierung aus Plasmamembranen isoliertem Prionprotein war nach 12 ppm-Phosmet-Behandlung nur knapp eine Verdopplung erkennbar. Der beobachtete Effekt von Phosmet betraf selektiv das Prionprotein. Andere Phospholipid-verankerte Proteine, die Gesamtproteinmenge und die Plasmamembran waren nicht vermehrt. Die Menge der mRNA für das Prionprotein veränderte sich nicht. Die Arbeitsgruppe von Prof. Prusiner hat eine Zelllinie 661, welche Antikörper gegen Hamster-Scrapie-Fibrillen produziert.
MH Cell Membrane/metabolism; Dose-Response Relationship, Drug; Gene Expression Regulation, Neoplastic/*drug effects; Human; Insecticides, Organophosphate/*pharmacology; Kinetics; Membrane Proteins/biosynthesis; Neuroblastoma; Phosmet/*pharmacology; Prions/*biosynthesis; Support, Non-U.S. Gov't; Tumor Cells, Cultured
AD Department of Neuroscience, Institute of Psychiatry, London, UK
SP englisch
PO England
OR Prion-Krankheiten G
ZF kritische Zusammenfassung von Roland Heynkes