J.
861
gen. Virol. (1985), 66, 861-870. Printed in Great Britain
Key words: scrapie/hydrophobiciO'/uw, inactivation/neutral lipid
Characterization of Lipids in Membrane Vesicles from Scrapie-infected
Hamster Brain
By C. D E E S , T. L. G E R M A N ,
W. F. W A D E
AND R. F. M A R S H *
Department o f Veterinary Science, 1655 Linden Drive, University o f Wisconsin-Madison,
Madison, Wisconsin 53706, U.S.A.
(Accepted 31 December 1984)
SUMMARY
The lipid compositions of m e m b r a n e vesicles from scrapie-infected and uninfected
hamster brains were examined before and after detergent extraction. No differences
were observed in polar lipids, glycolipids, gangliosides or neutral lipids examined by
thin-layer chromatography. Analysis of detergent-extracted CsC1 gradient fractions
with high scrapie infectivity failed to reveal any glycerolphosphatides, although neutral
lipids were demonstrated. The major neutral lipid associated with detergent-extracted
m e m b r a n e vesicles from both infected and uninfected brain was an unidentified lipid
which was found to absorb u.v. radiation strongly from 250 to 300 nm wavelengths.
Membrane neutral lipids that strongly absorb u.v. radiation at wavelengths normally
used to inactivate viruses may protect a small nucleic acid essential for scrapie
infectivity.
INTRODUCTION
Many unusual properties have been attributed to the scrapie agent, including unusual
resistance to heat, u.v. and ionizing radiation, and chemical inactivation (Millson et al., 1976).
The resistance of the agent to treatments that inactivate nucleic acids has led to speculation that
the agent is a unique pathogen with no nucieic acid component ( A l p e r e t a L , 1967; Lewin, I982;
Prusiner, 1982). Previous studies have shown that the majority of scrapie infectivity is
associated with cell m e m b r a n e (Millson et al., 1971 ; Semancik et al., 1976), and that treatments
which disrupt m e m b r a n e integrity, e.g. organic solvents and high concentrations of ionic
detergents, reduce infectivity (Millson et al., 1976). The close association with cellular
components is likely to contribute to the unusual properties of the agent and it has been proposed
that cell lipids may protect a small nucleic acid from u.v. inactivation (Latarjet, 1979;
Kimberlin, 1982). This study reports on attempts to partially characterize the lipid composition
of membrane vesicles from scrapie-infected hamster brain.
METHODS
Agent and bioassay. The scrapie agent used in these studies was serially passaged in outbred hamsters after
adaptation from the Chandler strain of mouse scrapie (Kimberlin & Marsh, 1975), Infectivity was quantified by
the method of incubation interval assay (Prusiner et al., 1981a) after intracerebral inoculation of weanling male
outbred hamsters purchased from Harlan Sprague Dawley (Indianapolis, Ind., U.S.A.).
Preparation ~l membrane t'esicles. Membrane vesicles were prepared on iodinated density gradient medium as
previously described (Marsh et al., 1984). Briefly, plasma membrane-enriched homogenates from scrapie-infected
or 12-week-old age-matched healthy control hamsters were sonicated, then separated on Nycodenz ® using rate
zonal centrifugation. Fractions enriched for large membrane vesicles, containing 108 LDso/ml of scrapie
infectivity and 180 ~,tg/mlof protein, were used for lipid studies. In some instances, vesicle fractions were extracted
with (I-57o Triton X-100 and re-fractionated by equilibrium density centrifugation in CsCI. A CsC1 fraction
containing 108.4 LDsQ/ml of scrapie infectivity was used for further studies (Marsh et al., 1984).
Lipid extraction procedures. Lipids were extracted from membrane vesicles using a biphasic chloroform/methanol/water system (Bligh & Dyer, 1959), or with single washings with butanol or pentanol. Detergent-extracted
0000-6386 © 1985 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 16:44:14
862
c . DEES AND OTHERS
membrane vesicle fractions from CsCI density gradients were extracted by the Bligh & Dyer method, and by single
solvent extraction with petroleum ether (3 × ). CsCI gradient fractions were also extracted with perchloric acid.
One hundred microlitres of 70~o perchloric acid was added to 1 ml of a CsC1 fraction from scrapie-infected or
uninfected hamster brain. Immediately, 1 ml of 0-6 M-perchloric acid was added and perchloric acid-insoluble
lipids and proteins were removed by centrifugation at 100000g for 1 h at 4 'C. Supernatants were neutralized by
careful addition of 0.1 M-NaOH to pH 7-2. All lipids extracted from membrane vesicles by solvent systems or
perchloric acid were examined by thin-layer chromatography (TLC).
TLC solid supports and driving soh'ent O'sterns. All TLC of polar lipids, including ganglioside and glycolipid
chromatography, was performed on pre-coated silica gel G (250 btm) 20 × 20 cm plates. Neutral lipids were
examined using pre-coated plates of similar size and depth, except that the solid support was silica gel H without
added binders. The driving solvent system for polar lipid analysis was chloroform/methanol/water (65:25:4 by
vat.). Propanol/water (7:3 by vol.) was the driving solvent system for ganglioside glycolipid analysis, and the
solvent system for neutral lipid analysis was petroleum ether/diethyl ether/glacial acetic acid (80:20:1 by vol.).
Precoated silica gel G or H plates were washed twice with chloroform/methanol (2 : 1 by vol.) and activated by
drying in an oven at II0"C for I h prior to use.
Visualization, quantification and identification oflipids. After TLC in the appropriate solvent system, lipids were
visualized by exposure to saturated iodine vapour or by charring (Shands & Noble, 1980). The amount of
individual lipid was estimated using the charring scintillation method (Shands & Noble, 1980). Neutral lipids were
also visualized by fluorescence under short- or long-wave u.v. irradiation with a Model UVS1-58 Mineralight ®
lamp. Glycolipids were identified by c¢-naphthol staining using the method of Siakotos & Rouser (Kates, 1975).
Gangliosides were stained with the Svennerholm, Miettinen & Takki-Luukainen resorcinol stain (Kates, 1975).
Phospholipids were stained by a modified Dittmer-Lester stain, and Marinetti's ninhydrin stain was used to
identify amino hpids (Kates, 1975). Beis-Dragendorf stain was used to identify choline-containing lipids (Kates,
1975). Sterols and sterol esters were visualized by a Lowry ferric chloride spray (Kates, 1975). Trace
glycerolphosphatides and polar lipids associated with CsCI density gradient fractions were radiolabelled
simultaneously with perchloric acid-soluble proteins (Greenwood et al., 1963). Lipids that were radiolabelled with
~2~1 were examined using TLC and the polar lipid driving solvent system. TLC plates were visualized by
autoradiography.
U.v. absorption spectra. Neutral lipids were extracted from CsCI fractions by washing with petroleum ether
(3 x ). The extracted neutral lipids were dried with sterile nitrogen gas, brought to a known volume in spectroscopy
grade hexane and examined for u.v. absorbance. The major neutral lipid associated with CsCI fractions was
parUally purified by normal phase HPLC on a DuPont Zorbax Sil column with 0.1 °o ethanol/hexane solvent
(Nelson. 1975). Partially purified neutral lipid was examined for its u.v. absorbance spectra in hexane.
Sensitivity to alkaline hydroO'sis. Sensitivity to base hydrolysis was performed as described previously (Kates,
1975). Extracted lipids from various fractions were subjected to mild alkaline hydrolysis and untreated lipids were
compared to base-treated lipids by TLC using the appropriate support and solvent driving system as described
above.
Effects ~?l soh'ents, dioxane, attd lipid-binding antibiotics on scrapie inJectiviO'. The effects of Bligh & Dyer and
butanol extraction on scrapie infectivity associated with membrane vesicles was examined. Biomolecules (nucleic
acids, proteins, carbohydrates, and complex glycolipids) that partitioned into methanol-water after extraction
were dialysed for 3 days against water at 4 ~C to remove traces of methanol. Samples were concentrated in a
Savant Speed Vac Concentrator @ and bioassayed for infectivity. Membrane vesicles were also washed once with
butanol or pentanol. Traces of solvents in the water layers were removed by dialysis before concentration of the
samples for bioassay.
The effect of petroleum ether washing on scrapie infectivity associated with CsC1 fractions was determined
using procedures similar to those for butanol. The effect of dioxane treatment on scrapie infectivity was
determined by adding 10°-o, 15°-~,or 25 °.o (v/v) dioxane to samples from CsCI gradients. Samples were bioassayed
directly without dialysis. Polymyxin B and gramicidin S were added at a concentration of 4 × 103 units per mg of
protein to membrane vesicle suspensions to determine their effects on scrapie infectivity. As in the studies with
dioxane, the lipid-binding antibiotics were not removed prior to infectivity assays.
RESULTS
Polar lipids in membrane vesicles
Membrane vesicles before detergent extraction contained a wide variety of polar lipids,
including phosphate-bearing glycerolphosphatides (phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, and phosphatidylinositol). No qualitative differences
between scrapie-infected and uninfected samples were observed.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 16:44:14
Lipids of scrapie-inJbcted membrane vesicles
(a)
(a)
(b)
(b)
(c)
(d)
(e)
(e)
(d)
(e)
(b)
~O
6
•
(f)
(f)
(g)
(h)
(g)
(h)
30 3.0 ?0
1.I
~,.I
1.
I
I
I
].0
tQ
g
(i)
Q 2.a 2.4 12.ql,
P
1. B
(i)
q
|
2.•
863
S
0
O
O
I •
m
Fig. 1. (a) TLC of neutral lipids extracted from membrane vesicles from uninfected and scrapieinfected hamster brain by three solvent systems (lanes a and b, Bligh & Dyer; lanes e andJ~ butanol;
lanesg and h, pentanol). Identifications and R¢ values for these lipids are given in Table I. Lanes (c), (at)
and (i) contain reference neutral lipids. Lanes t,a), (e) and (g) contain neutral lipids from uninfected
hamster brain, lanes (b), (J) and (h), from scrapie-infected brain. (b) Diagrammatic representation of
extracted lipids seen in (a). Lipids are identified by their lane and number designation in Table 1.
Neutral lipids in membrane vesicles
The neutral lipid c o m p o s i t i o n of m e m b r a n e vesicles from scrapie-infected b r a i n also was not
qualitatively different when c o m p a r e d to vesicles from uninfected brain (Fig. l a, b; T a b l e 1).
Fig. 1 (a, b) shows that a highly fluorescent, very n o n - p o l a r lipid is a m a j o r neutral lipid. The
lipid is unidentified but had a relative m i g r a t i o n similar to c a r o t e n o i d fat-soluble v i t a m i n s such
as r e t i n y l p a l m i t a t e (Table 1). The lipid was highly fluorescent under short-wave irradiation, and
only slightly less so with long-wave excitation, as e x p e c t e d for a carotenoid-like c o m p o u n d . T h e
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 16:44:14
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 16:44:14
CN
TG
ST
5.
4.
3.
2.
1.
67
28
25
21
16
Rr
FAME
FFA
ST
1,3 DG
1,2 DG
Designation
2.
1.
No.
77
48
Rr
~
STE
TG
Designation
Neutral lipid
standards
(lane d)
3.
2.
1.
No.
91
47
34
Rr
~k
CN
TG
FFA
Designation
Butanol
extraction
(lanes e, f )
3.
2.
1.
No.
92
47
35
Rv
~k
CN
TG
FFA
Designation
Pentanol
extraction
(lanes g, h, i)
* Fig. I : lanes (at, (el and (g) contain neutral lipids from uninfected hamster brain : lanes (b), (f) and (h) contain neutral lipids from scrapie-infected hamster brain;
lane (i) contains cholesterol (R~ 22) and squalene (R~ 86).
? No. refers to the corresponding number and lane in Fig. 1 (b).
Rr is reported × 100 in petroleum ether/diethyl ether/acetic acid (80:20:1 by vol.) on silica gel H.
§ Lipids were presumptively identified by relative migration. Key for lipids: CN, carotenoid; TG, triglyceride; ST, sterol; F A M E , fatty acid methyl ester; DG,
diglyceride; STE, sterol ester; FFA, free fatty acid.
93
41
22
3.
2.
1.
No.
Jk
Des gnation§
A
Rt ~
No.?
Neutral lipid
standards
(lane c)
Bligh & Dyer
extraction
(lanes a. h)*
T a b l e I. Presumptive identification o]' neutral lipids f r o m membrane vesicles f r o m scrapie-inJected and uninJected hamster brain
865
Lipids of scrapie-inJected membrane vesicles
(a)
(b)
(c)
(d)
(e)
(f)
(g)
rr!
Fig. 2. TLC of gangliosides from scrapie-infected and uninfected hamster brain membrane vesicles. No
qualitative differences in gangliosides were observed (small arrows indicate spots stained with
resorcinol for gangliosides and positive for carbohydrates by c~-naphthol). Large arrows indicate spots
that stain yellow for phosphatides with c~-naphthol. Solvent key: membrane vesicles extracted
according to Bligh & Dyer (lanes a and b); extracted with butanol (lanes c and d); extracted with
pentanol (lanes e and J). Lane (g) contains a mixture of reference gangliosides. Lanes (a), (c) and (e) are
from uninfected brain; (b), (d) and (f) from scrapie-infected brain.
lipid took up iodine readily which is consistent with a carotenoid compound. However, this lipid
was not base-labile like retinylpalmitate. The compound did not have a relative migration
similar to any known tocopherol or D vitamin by TLC or H P L C (unpublished data). The lipid
was also found associated with CsC1 fractions of detergent-extracted membrane vesicles (Fig. 4).
Gangtiosides and glycolipids in membrane vesicles
Gangliosides were found in the total lipid extracts removed from membrane vesicles by Bligh
& Dyer extraction, and by butanol and pentanol even though these techniques are not
appropriate for complete quantitative extraction of gangliosides (Fig. 2). No qualitative
differences in the ganglioside composition of healthy and scrapie-infected tissues were evident.
One lipid did not stain with resorcinol but was positive for carbohydrate with c~-naphthol stain.
The c~-naphthol-positive glycolipid was found in both scrapie-infected and uninfected tissue.
Polar lipids in CsCl fractions
No polar lipids, including glycerolphosphatides, could be found in lipid extracts from CsC1
gradient fractions by TLC and conventional visualization methods such as charring or
phosphate stain. Radioiodination of lipids did show that perchloric acid-soluble lipids that were
mobile in a TLC polar lipid solvent system were associated with the CsC1 fractions (Fig. 3). The
lipids had very low mobility, suggesting that they are extremely polar and highly water-soluble.
They may be highly complex brain glycolipids, but further study is required for their
identification. Two very small spots with relative mobilities equal to that of phosphatidylcholine
and phosphatidylethanolamine could be found in the radioiodinated samples suggesting that
there may be trace amounts of glycerolphospholipids associated with the CsCI fractions.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 16:44:14
866
C. DEES AND OTHERS
(a)
(b)
(e)
(d)
PAl
PE B
I
Pc ti
sM I
!
I
Fig. 3. Perchloric acid-soluble polar lipids from CsC1 fractions. Perchloric acid-soluble radioiodinated
polar lipids that are associated with uninfected (lane c) and scrapie-infected (lane d) hamster brain, gl,
Glycolipid; nl, neutral lipid. Lane (a) contains reference gangliosides and lane (b) reference polar
phospholipids (PA, phosphatidic acid ; PE, phosphatidylethanolamine; PC, phosphatidylcholine; SM,
sphingomyelin).
Neutral lipids in CsCl fractions
In contrast to the small amounts of polar lipids associated with the detergent-extracted CsC1
fractions, a relatively large amount of neutral lipid was not removed by detergent extraction
(Fig. 4). Minor amounts of free fatty acids, diglycerides and monoglycerides were found in CsC1
gradient fractions from both heaithy and scrapie-infected tissue (Fig, 4, Table 2). The majority
of neutral lipid remaining after detergent extraction and fractionation on CsC1 gradients was a
single unidentified, very non-polar lipid. This lipid exhibited fluorescence under short-wave u.v.
and readily absorbed iodine when exposed to iodine vapour. While we cannot positively identify
the lipid, its relative mobility in solvent systems (Fig. 4) and by H P L C (unpublished data)
suggests that it may be a carotenoid or some other vitamer.
U.v. absorbance oJ neutral lipids from CsCl.fractions
Total neutral lipids associated with CsC1 fractions from both infected and uninfected brain
tissue were found to have a broad absorbance spectrum over the u.v. range usually used to
inactivate viruses (250 to 280 nm). As noted above, the strong fluorescence exhibited by the
major neutral lipid suggested that it was responsible for the u.v. absorption spectrum seen in the
total lipid extract. This was confirmed after partial purification by normal phase HLPC.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 16:44:14
Lipids of scrapie-injected membrane vesicles
(a)
(b)
(c)
867
(d)
Fig. 4. Neutral lipids in CsCI fractions of detergent-extracted membrane vesicles from uninfected (lane
c) and scrapie-infected (lane at) brain. Lanes (a) and (b) contain cholesterol (choi) and standard steroI
ester (ste) markers.
Table 2. Quantification of neutral lipids associated with CsCI fractions from scrapie-infected and
uninfected brain
% Total neutral lipid
;k
Lipid class
Unidentified
(carotenoid ?)
Free fatty acid
1,3 Diglyceride
1,2 Diglyceride
Monoglyceride
Uninfected Scrapie-infected
79.95*
6.40
5.56
3-36
4.60
77.89
7.67
6.49
3-60
4.03
* Quantity of the individual classes of lipids was estimated by a quench char method. Charring of each
individual lipid was compared to charring of known reference lipids of that class with the exception of the
unidentified lipid which was compared to a mixture of carotenoids.
Eflects of soh,ent extraction on infectivity
Table 3 shows the results of solvent extraction, lipid-binding antibiotics, and dioxane on the
infectivity of the scrapie agent in m e m b r a n e vesicles or CsC1 fractions. Simple solvent
extraction of membrane vesicles with butanol reduced infectivity from 108 LDs0 per ml to 106"5.
Pentanol, which is less efficient in extracting lipids, was also slightly less efficient in reducing
scrapie infectivity. Bligh & Dyer lipid extraction was found to be most effective in reducing
infectivity in membrane vesicles. Polymyxin B and gramicidin S, which can bind and disperse
glycerolphosphatides in membranes, failed to have any effect on the infectivity of the agent.
Dioxane, which can alter the structure of m e m b r a n e lipoproteins thus interfering with
hydrophobic bonds, failed to reduce the infectivity of the scrapie agent in CsC1 fractions.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 16:44:14
868
C. DEES AND OTHERS
Table 3. Effects of solvents, lipid-binding antibiotics, and dioxane on scrapie infectivity in
membrane vesicles bejbre or ajter detergent extraction and fractionation on CsCl gradients
Preparation
Before detergent
extraction
After detergent
extraction
Treatment
Bligh& Dyer extraction
Butanol
Pentanol
Polymyxin Bt
Gramicidin S~"
Petroleumether extraction
Dioxane (10~£)
Dioxane (15°,0)
Dioxane (25°/o)
Starting
titre*
8
8
8
7.4
7.4
7.8
8
8
8
Titre after
treatment
5-8
6.5
7
7
7
6.5
7
7
8
* All titres are log~o LDso per ml as determined by incubation interval assay (Prusiner et al., 1981a).
The concentration of polymyxin B and gramicidin S was 4 x 103 units per mg protein,
DISCUSSION
Extensive qualitative analysis of the phospholipids, neutral lipids, and gangliosides extracted
from membrane vesicles prepared from scrapie-infected or uninfected brain failed to reveal any
discernible differences. This finding is consistent with recent studies using electron spin
resonance which showed that spin probes in the hydrophobic core of phospholipid bilayers of
scrapie-infected membrane were unperturbed except for slight effects at the terminal stages of
the disease (Viret et al., 1981). However, spin probes associated with hydrophobic portions of
membrane proteins were perturbed early in the disease, suggesting rearrangement of certain
membrane proteins (Viret et al., 1981). In this study, we cannot account for the perturbation of
the lipid-associated spin probe in the terminal stages of the disease. However, subtle alteration
of the fatty acid side chains of glycerolphosphatides might account for the previously observed
phenomena. We have not compared the fatty acid composition of infected cell membranes with
fatty acids found in uninfected cell membranes. Other studies suggest alterations in fluidity of
scrapie-infected membranes consistent with alterations in fatty acid composition (Rutter et al.,
1981).
Previous studies havc shown that glycosides are incrcased and ganglioside composition is
altered in Creutzfeldt-Jakob disease (Annunziata & Federico, 1981; Yu & Manuelidis, 1978).
We did not find any qualitative changes in the ganglioside composition of membrane vesicles
from infected hamsters. The observation that the high passaged hamster-adapted scrapie agent
produces very little spongiform degeneration (R. F. Marsh, unpublished results) may account
for these differences.
The lipids that remain with the detergent-insoluble membrane residue after fractionation on
CsC1 density gradients was also examined. CsCI gradient fractions with high scrapie infectivity
were remarkably free of glycerolphosphatides that, for the most part, are responsible for
maintaining bilayer integrity. Traces of several glycerolphosphatides could be found after
radioiodination of lipids. Other highly polar and water-soluble lipids were also detected after
radioiodination. These lipids may be complex brain glycolipids that are not quantitatively
removed by detergent or solvent extraction. However, radioiodination of trace lipids does not
provide uniform labelling. Therefore, further studies are required to quantify and identify the
trace lipids and highly polar lipids associated with CsCI fractions. Furthermore, if the scrapie
agent requires an essential lipid protein relationship or a membrane bilayer for infectivity, it is
unclear at this time how the bilayer is maintained with the majority of glycerolphosphatides
removed. Infectivity may be maintained in CsCI fractions without glycerolphosphatides by
micelles formed from the trace glycerolphosphatides in combination with membrane proteins
and detergents. If a significant amount of highly water-soluble brain glycolipid remains, it may
participate in the formation of an essential bilayer or provide an essential boundary lipid
required for infectivity. Previous studies on the Na+,K + ATPase of cytoplasmic membranes
have shown that as little as 1 ~ of the original essential polar lipid is required for function of the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 16:44:14
Lipids of scrapie-infected membrane vesicles
869
enzyme after solubilization (Jorgenson, 1982). Trace amounts of glycerolphosphatides found
associated with CsCI fractions, similar to studies with the Na+,K + ATPases, may be enough
lipid to maintain infectivity. However, other studies suggest that glycerolphosphatides may not
be required for infectivity. Gramicidin S and polymyxin B, which bind to and disrupt biological
membranes by interaction with essential glycerolphosphatides (Cooperstock & Riegle, 1981;
Vostroknutova et al., 1981), were shown in this study to be ineffective in reducing scrapie
infectivity. Infectivity of the scrapie agent in CsC1 density gradient fractions was also
unaffected by dioxane which can disrupt hydrophobic lipid protein interactions.
Infectivity of the scrapie agent in CsC1 fractions was reduced by simple solvent washing of the
material with petroleum ether which removed neutral lipids. Reduction of infectivity by
petroleum ether may result from damage to proteins that are essential for scrapie infectivity
(Cho, 1983L as opposed to removal of an essential lipid or disruption of lipid-protein
interactions. Highly effective lipid extraction procedures (Bligh & Dyer, butanol) were also
effective in reducing scrapie infectivity, but these procedures may also result in extensive
denaturation of proteins or extraction of highly hydrophobic membrane proteins. Previous
studies have claimed that the scrapie agent contains a hydrophobic protein (Prusiner et al.,
1981 b). Therefore, the effects of solvents and detergents on reducing scrapie infectivity might be
due to removal or solubilization of a hydrophobic protein, or proteolipid, rather than removal of
essential lipid. We have been unable to find any proteolipid that is removed by solvent
extraction of CsCI fractions (Dees et al., 1985). Loss of infectivity of the scrapie agent by
solvents or detergent treatment is most likely due to denaturation of essential proteins, or to
disruption of essential lipid-protein interactions in membranes containing the scrapie agent.
Further studies are necessary to define precisely the effects on scrapie infectivity of treatments
that disrupt hydrophobic interactions.
We also found that although CsC1 fractions with high scrapie infectivity are relatively free of
glycerolphosphatides, a number of neutral lipids are not removed by detergent extraction. The
major neutral lipid has characteristics of carotenoids, like retinylpalmitate, but the lipid is not
affected by alkaline hydrolysis that would hydrolyse the ester bond in a retinylpalmitate-like
vitamer. This major neutral lipid deserves further study since it absorbs u.v. radiation at
wavelengths normally used to inactivate viruses by inducing photodimerization of bases in
nucleic acids. Thus, previous suggestions that the scrapie agent may contain an essential nucleic
acid that is protected from inactivation by u.v. radiation by cellular lipids may be correct
(Latarjet, 1979; Kimberlin, 1982).
These studies were supported by the College of Agricultural and Life Sciences, University of Wisconsin
Madison, and by a Romnes Faculty Fellowship awarded to R. F. M. by the Wisconsin Alumni Research
Foundation. A Research Assistantship to C. Dees was provided by the Department of Pathobiological Sciences,
University of Wisconsin, School of Veterinary Medicine.
REFERENCES
ALPER, T., CRAMP,W.
A., HAIG, D. A. & CLARKE, M. C.
(1967). Does the scrapie agent replicate without a nucleic acid?
Nature, London 214, 764 765.
ANNUNZIATA,P. & EEDERICO, A. (1981). Brain glycosidases in Creutzfeldt-Jakob disease. Journal of the Neurological
Scienees 49, 325-328.
BLIGH, E. G. & DYER, W. J. (1959). A rapid method of total lipid extraction and purification. Canadian Journal oJ
Biochemistry and Physiology 55, 297 302.
CHO, n. J. (1983). Inactivation of the scrapie agent by Pronase, Canadian Journal o! Comparatice Medicine 47,494
496.
COOPERSTOCK, M. & RIEGLE, L. (1981). Polymyxin B inactivation of lipopolysaccharide in vaccines of G r a m negative bacteria, h~/bcthm and Immunity 33, 315 318.
DEES, C., GERMAN,T. L., WADE, W. F. & MARSH,R. F. (1985). Characterization of proteins in m e m b r a n e vesicles from
scrapie-infected hamster brain. Journal of General Virology 66, 851 859.
GREENWOOD, F. C., HUNTER, W. M. & GLOVER, J. S, (1963). The preparation of ~3 ~I-labelled h u m a n growth hormone
of high specific radioactivity. Biochemical Journal 894 114123.
JORGENSON, P. L. (1982). Mechanism of the Na+,K + pump. Protein structure and conformation of the pure
(Na+,K+)-ATPase. Biochimica et biophysiea acta 694, 27-68.
KATES, M. (1975). Techniques of lipidology. In Laboratory Techniques in Biochemisto' and Molecular Biology, pp.
435-444. Edited by T. S. Work & E. Work. A m s t e r d a m : North-Holland.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 16:44:14
870
C. D E E S A N D
OTHERS
KIMBERLIN, R. H. (1982). Reflections on the nature of the scrapie agent. Trends in Biochemical Science 7, 393-394.
KIMBERLIN, R. H. & MARSH, R. F. (1975). Comparison of scrapie and transmissible mink encephalopathy in
hamsters. I. Biochemical studies of the brain during development of disease. Journal of InJbctious Diseases
131, 97-103.
LATARJET, R. (1979). Inactivation of the agents of scrapie, Creutzfeldt-Jakob disease and kuru by radiations. In
Slow Transmissible Diseases of the Nervous System, vol. 2, pp. 397-407. Edited by S. B. Prusiner & W. J.
Hadlow. New York: Academic Press.
LEWIN, P. K. (1982). Infectious peptide: postulated mechanisms of protovirion replication in scrapie. Canadian
Medical Association Journal 127, 471-472.
MARSH, R, F., DEES.C., CASTLE,B. E., WADE, W. F. & GERMAN,T. L. (1984). Purification of the scrapie agent by density
gradient centrifugation. Journal oJ General V~)'ology65, 415 421.
MILLSON,G. C., HUNTER, G. D. & KIMBERLIN,R. 14. (1971). An experimental examination of the scrapie agent in cell
mixtures. II. Association of scrapie activity with membrane fractions. Journal of Comparative Pathology 81,
255 265.
MILLSON,G. C., 14UNTER,G. D. & KIMBERLIN,R. 14. (1976). The physico-chemical nature of the scrapie agent. In Slow
Virus Diseases oJ Animals and Man, pp. 243-266. Edited by R. H. Kimberlin. Amsterdam & Oxford : NorthHolland.
NELSON, G. J. (1975). Isolation and purification of lipids from animal tissues. In Analysis of Lipids and Lipoproteins,
pp. 1 22. Edited by E. G. Perkins. Champaign: American Oil Chemists Society.
PRUSINER, S. B. (1982). Novel proteinaceous infectious particles cause scrapie. Science 216, 136-144.
PRUSINER, S. B., COCHRAN,S. P., GROTH, D. F., BOWMAN,K. A. & MARTINEZ,H. M. (1981 a). Measurement of the scrapie
agent using an incubation time interval assay. Annals oJ Neurology 11, 353 358,
PRUSINER, S. B., McKINLEY,M. P., GROTH, D. F., BOWMAN,K. A., MOCK,N. I., COCHRAN,S. P. & MASIARZ,F. R. (1981 b).
Scrapie agent contains a hydrophobic protein. Proceedings of the National Academy of Sciences, U.S.A. "78,
6675-6679.
RUTTER, G., ASHER,O. M., ROHWER,R. G., GIBBS,C. L & GAJDUSEK,D. C. (1981 ). Increased concanavalin A capping in
cells from brains of scrapie-infected hamsters. Archives of Virology 68, 129 133.
SEMANCIK,J. S., MARSH, R. F., GEELEN, J. L. & HANSON,R. P. (1976). Properties of the scrapie agent endomembrane
complex from hamster brain. Journal of" Virology 18, 693-700.
S14ANDS,J. H. & NOBLE, R. C. (1980). Quantitation of lipid mass by liquid scintillation counting procedure following
charring on thin-layer plates. Analytical Biochemistry 101, 427-434.
VIRET, J,, DORMONT, D., MOLLE, D., COURT, L., LETERRIER, F., CATHALA,F., GIBBS, C. J. & GAJDUSEK,D. C. (1981).
Structural modifications of nerve membranes during experimental scrapie evolution in mouse. Biochemical
and Biophysical Research Communications 101, 830 836.
VOSTROKNUTOVA,G. N., BULGAKOVA,V. G., UDALVA,T. P., SEPETOV, N. F. & SIBEL'DINA,L. A. (1981). Localization of
gramicidins on the cytoplasmic membrane of Bacillus brevis and its effect on the activity of membrane
enzymes. Biokhimiya 46, 657-666.
YU, R. U. & MANUELIDIS,E. E. (1978). Ganglioside alterations in guinea pig brains at end stages of experimental
Creutzfeldt-Jakob disease. Journal of the Neurological Sciences 35, 15-23.
(Received 18 September 1984)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 16:44:14