Download Altered Scrapie Infectivity Estimates by Titration and Incubation

J. gen. Virol. (1983), 64, 2045-2050. Printed in
Great Britain
2045
Key words: scrapie infectivity/incubation period/effect of detergents
Altered Scrapie Infectivity Estimates by Titration and Incubation Period in
the Presence of Detergents
By R O B E R T A. S O M E R V I L L E * t AND R I C H A R D I. C A R P
New York State Institute for Basic Research in Developmental Disabilities,
1050 Forest Hill Road, Staten Island, New York 10314, U.S.A.
(Accepted 25 May 1983)
SUMMARY
During experiments on the purification of scrapie infectivity, changes were found in
the dynamics of scrapie titration. After exposure to detergent, infectivity estimates by
both endpoint titration and incubation period were altered. The addition of detergent
to the diluent used in titration resulted in at least a 100-fold increase in the infectivity
estimate. This suggests that the amount of scrapie in a sample, as measured by serial
dilution and titration, may be underestimated to different extents, depending on the
biochemical milieu of the inoculum. Membrane fractions treated with detergents
before dilution exhibited longer incubation periods than untreated fractions for the
same number of infectious units of scrapie. This demonstrates that detergent
treatments and possibly other biochemical manipulations can cause changes in the
response of the host to the inoculum that are not detectable if incubation periods alone
are used to estimate scrapie 'titre'.
Scrapie is an infectious disease of the central nervous system of sheep and goats. Experimental
transmission of scrapie to mice has allowed the large scale measurement of scrapie by the
standard microbiological technique of titration to a limiting dilution to give an LDs0. It was
apparent from early studies in mice that incubation periods were extended as the amount of
infectivity was reduced (Chandler, 1963; Hunter et al., 1963; Mould et al., 1967). When all
variables such as strain, sex and age of mouse, strain of scrapie agent, and the amount and
chemical treatment of inoculum are held constant, then titre estimates and incubation periods
are remarkably reproducible (Outram, 1976). Comparative measurements of incubation periods
of strictly equivalent samples have been performed on many occasions (e.g. Dickinson et al.,
1969; Kimberlin & Walker, 1980).
Attempts have also been made to use estimates of titre derived solely from measurements of
incubation period to compare samples that are not strictly equivalent (Hunter & Millson, 1964;
Prusiner et al., 1980a, b, 1982). This involves the assumption that a dose-response curve
established for the starting homogenate could be used as the standard from which to determine
the titre from the incubation period, even for samples subjected to different biochemical
manipulations.
If certain of the parameters for measuring scrapie infectivity are changed then the estimated
titre and the relationship between titre and incubation period are altered. For example, titration
of infectivity using an intraperitoneal route of infection rather than an intracerebral route can
lead to at least a 1 loglo unit decrease in the titre estimate and a prolongation of incubation
period for the same dose (i.e. number of infectious units measured) (Kimberlin & Walker, 1978).
Several treatments of brain homogenate have previously been found to affect the expected
relationship between incubation period and dose of infectivity. Longer incubations for
equivalent doses of scrapie infectivity were found after boiling the inoculum (Dickinson &
Fraser, 1969). Although Prusiner et al. (1982) found no effect on the titre estimate when
inoculum had been heated to temperatures of 100 °C, there was a 1 log10 unit difference between
t Present address: ARC and MRC Neuropathogenesis Unit, West Mains Road, Edinburgh EH9 3JF, U.K.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:53:31
0022-1317/83/0000-5651 $02.00 © 1983 SGM
2046
S h o r t communication
I
I
I
I
I
I
I
I
I
f
I
104
A0
r~
5
102
.........................................
Z
.-..._
-.}
I
I
~,l
ix- -'r-I
l
190
210
230
Time (days)
Fig. 1. Effect of detergents on dose-response curves. Brains from C57BL/6J mice infected
intracerebrally with the 139A strain of scrapie were fractionated according to the method of Cohen et
aL (1977)as modified by R. A. Somerville,P. A. Merz & R. I. Carp (manuscript in preparation). SPM
fractions were treated with detergent and either subjected to differential centrifugation or overlaid on
25 to 55% metrizamide gradients and centrifuged at 80000g for 16 h. Aliquots of the followingfractions
were titrated: brain homogenate, synaptosomal fraction and SPM fraction (A
A); synaptosomal
fraction plus S3-14, SPM fraction plus octyl glucoside and overlay from the top of metrizamide
gradients after both detergent treatments (O---C)); fractions from the gradients after centrifugation
and the pellet obtained after octyl glucoside treatment and centrifugation (O .... 0). Samples were
titrated by intracerebral injection of 0-03 ml of 10-fold serial dilutions in PBS. Mice were monitored
twice weekly from day 130 to day 200 and once weekly until day 280 post-infection for clinical signs of
scrapie, the incubation period being designated the time at which definite clinical signs of scrapie were
first observed. Histological examination was performed on two occasions to confirm when a clinical
diagnosis of scrapie was not clear. Titres were calculated by the Reed-Muench method (Dougherty,
1964) and the dose for each dilution determined. The regression lines indicated for each group of data
were calculated for all values with less than 102 infectious doses.
I
I
I
150
170
I
the LD50 and the incubation period estimate of titre after treatment at 12l °C. SDS or LiC1
treatments of b r a i n homogenate have also produced skewing of the dose-response curve
(Kimberlin, 1977). It has also been noted that measurement of infectivity after its partial
purification with lysolecithin resulted in intracerebral estimates of titre similar to those obtained
before treatment, but intraperitoneal measurements were reduced by 3 log10 units after
treatment (G. C. Millson, personal communication). Here, we present data obtained in
biochemical experiments (involving the use of detergents) in which the treatments changed the
measured LDs0 or altered the relationship between incubation period and dose.
I n two experiments on tile purification of scrapie infectivity, fractionations were performed to
yield synaptosomal and synaptic plasma m e m b r a n e (SPM) fractions (Fig. 1). Detergent [n-octylfl-D-glucopyranoside (octyl glucoside) or sulphobetaine 3-14 ($3-14)] was added to these
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:53:31
Short communication
2047
Table 1. Effects of detergent in the diluent on infectivity titres and incubation periods
(a) Titration of brain homogenate*
Addition of NLS (0.1~)
No detergent added
Dilution
10-z
10-5
10-6
10-7
10-s
Titre§
'Mice/group
with scrapie
6/6
6/6
6/6
6/6
4/6
t>8'3
10-2
10-5
10-6
6/6
4/4
5/5
10-7
Titre§
O/1
i>6-5
Incubation
period ( + S.E.M.)I"
133 + 0-6
142 + 0-8
159 __+2"7
169 + 10.0
175 + 3-14
Mice/group
with scrapie
5/5
6/6
6/6
1/5
0/6
6.4
Incubation
period ( + S.I~.M.)
133 + 1.2
154 + 0.6
181 _+ 8'7
210
--
(b) Titration of a detergent-treated membrane fraction*
139 + 2-0
6/6
t44 + 2-0
197 + 9-0
0/6
-238 + 5-2
0/6
---
0/6
--
p~:
NS
<0.001
<0"02
---
NS
----
1>2.5, ~<4-5
* Scrapie-infected brain was fractionated as described for Fig. 1. The synaptosomal fraction was treated with
$3-14 and overlaid on a metrizamide gradient. Brain homogenate and a 2 ml fraction from the centre of the
gradient were titrated as in Fig. 1. NLS was added to the diluent for the titration of one aliquot of each sample and
omitted from the other.
t" S.E.M., Standard error of the mean.
:~Probability that differences in incubation periods at the same dilution occurred by chance (Student's t-test).
NS, Not significant.
§ Titre, logl0LDs0 per 0.03 ml.
fractions which were then subjected to differential or density gradient centrifugation. Aliquots
of these fractions were titrated for scrapie infectivity and the infectious dose at each dilution
calculated (Fig. 1). The relationship between dose-response curves and incubation period was
compared by analysing three groups o f data pooled from these experiments: group 1, before
detergent treatment (homogenate, synaptosomal and SPM fractions); group 2, samples
containing detergent (membrane fractions containing detergent and samples from the overlay of
density gradients where detergent remains after centrifugation); group 3, samples previously
exposed to detergent (the pellet obtained after centrifugation and samples from the density
gradient). For each group the relation between infectious dose and incubation period was
compared. Data in each group were analysed by means of a multiple regression procedure
(Cohen & Cohen, 1975), with groups represented by dummy coding. In c o m m o n with most other
scrapie dose-response curves a point in inflection occurred between 102 and 103 units of
infectivity. Since the majority of the values were below 102 units only these data were included in
the regression analysis.
There was a significant linear relationship between dose of scrapie and incubation time (Fig.
1) which did not vary with group (r = 0-92, 0-94, 0.98 respectively). Regardless of dosage,
significant differences were obtained in incubation period between groups 1 and 2 (P < 0-01)
and between groups 1 and 3 (P < 0.001). The mean incubation period for samples containing
detergent (group 2) was 11 days longer than for untreated material (group 1) for the same
operational doses of infectivity. A 19 day difference was found when untreated fractions (group
1) were compared with fractions previously exposed to detergent (group 3).
In a separate experiment, the effect of the detergent N-lauroyl sarcosinate (NLS) added to the
diluent was tested (Table 1). Two samples were titrated each with and without detergent
included in the diluent. Titration of a scrapie-infected brain homogenate with 0.1 ~o NLS
included in the diluent (phosphate-buffered saline, PBS) resulted in a titre estimate at least 2
log10 units higher than that obtained from the sample which was diluted in PBS alone (Table
1 a). Dilutions to the endpoint were not performed in the samples with unexpectedly high titres.
It is probable that a complete LDs0 estimate of the dilution series with N L S would give a value
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:53:31
2048
Short communication
higher than 10 t° infectious units/g. In the second part of this experiment (Table 1 b), detergent
(S3-14)-treated synaptosomal material obtained from the scrapie brain homogenate was titrated
in the presence and absence of NLS. Again, the addition of detergent to the diluent resulted in
an LDs0 estimate at least 100-fold higher than in the sample in which detergent was omitted.
Comparison of the incubation periods of each dilution shows that, whereas at low dilution
(Table l a and b; 10-2 dilution) the incubation periods are similar despite a 2 log10 unit
difference in effective dose, the incubation periods at higher dilutions in NLS (Table 1a; 10-5,
10-6) were significantly shorter than those obtained in the absence of NLS. Because an endpoint
was not reached in two of these titrations it is not possible to compare fully the dose-response
curves from this experiment. Assuming the value obtained for the homogenate (Table 1a) with
detergent is near to or a little higher than the true LDs0, it is clear that for similar doses at high
dilution there is little difference in incubation period whether detergent is present or absent.
However, at low dilution (10-2) the incubation period is similar despite a 2 logx0 unit difference
in dose. The detergent-treated membrane fraction showed, similarly to Fig. 1, a prolongation of
incubation period for equivalent doses when compared to the homogenate. A few deaths
occurred in animals injected with NLS soon after injection (Table 1b) presumably because of the
toxic effect of the detergent.
It is clear from these results that the measurement of scrapie infectivity by titration can be
profoundly affected by the chemical environment of the infectivity in the inoculum. A priori,
chemical treatments may act on the infectious agent itself, for example by destroying it; they
may alter the environment of the infectious unit such that its ability to infect the host is changed,
and chemicals in the inoculum may act on the host to alter its response to infectious material. In
Table 1, the elevated titre observed in the detergent-containing samples could be due to a
combination of these factors. Specifically in these experiments, tissue disaggregation could be
more effective in the presence of detergents, such that aggregates containing more than one
potential infectious unit are dispersed, thereby increasing the effective number of infective
units; the ability to infect the injected animal might be enhanced by the presence of detergent,
possibly by opening up the blood-brain barrier. In addition, the adherence of infectivity to
dilution tubes, syringes, etc. may be reduced in the presence of detergent. It has been found that
significant losses of infectivity can occur when inocula are left to stand for a period of 4 h in glass
bottles or syringes before injection (A. G. Dickinson, personal communication). Titres of brain
homogenates have been elevated by extensive homogenization (Malone et al., 1978) and,
similarly, sonication of a brain homogenate at each step in the dilution series elevated the titre
17-fold (Rohwer & Gajdusek, 1980).
The data obtained in these experiments also show changes in the relationship between dose
and incubation period. Changes in dose-response curves have been noted in unfractionated
tissue after physical or chemical treatments, and have recently been reported for partially
purified fractions exposed to sodium deoxycholate (Lax et al., 1983). In one case, treatment of a
partially purified fraction with Triton X-100 did not affect the titre estimate or the dose-response
curve (Somerville et al., 1980). Two effects of detergent on incubation period were observed in
the present experiments. For samples diluted in the presence of NLS the data show that the
presence of detergent increases the calculated number of infective units present at low dilutions
(2.5 x 104 infectious units in the 10-2 dilution in the absence of detergent and at least 2 x 106
infectious units in the presence of NLS) without changing the incubation period. In contrast,
there are significant differences in incubation period between the samples with and without
detergent at higher dilutions of the same starting material. Presumably, these are due primarily
to the higher effective dose present in the sample with detergent added to the diluent. Overall,
the addition of detergent to the diluent in this experiment increased the number of infective
units present in each dilution, altering the relationship between dose, incubation period and the
degree of dilution. It would be impossible to predict the differences in titre illustrated here solely
from a measurement of incubation period. The second effect produced on dose-response curves
is demonstrated in Fig. 1 where a shift in the dose-response curve took place following chemical
treatment of the sample, in this case exposure to detergent before dilution. The effects found
here are probably comparable to those described by Lax et al. (1983). Since the samples in this
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:53:31
Short communication
2049
experiment were diluted under identical conditions it is unlikely that the mechanics of the
dilution process played a major role in producing the effect shown. Nevertheless, it is possible
that significant amounts of detergent in group 2 would be carried through the dilution process so
that at low dilutions effects similar to those shown in Table 1 may have applied to a certain
extent. Clearly, there was a significant shift in the dose-response curves after treatment, which
can only be ascribed to changes in the presentation of the inoculum to the host and a
consequential alteration in the pathogenesis of the disease.
In conclusion, addition of detergent to the diluent for scrapie titration enhanced the efficiency
of the titration process such that titre estimates at least 100-fold higher were obtained.
Biochemical manipulation of the samples by treating with detergents resulted in altered doseresponse curves. These two results have fundamental implications for the measurement of
scrapie infectivity. Of particular importance will be the re-evaluation of experiments in which
various biochemical treatments were intended to inactivate infectivity, since those treatments
may have altered the interaction of infectivity with its environment or changed the scrapie
agent's ability to infect the host rather than affected the structure or properties of the infectious
molecule p e r se. In addition, estimates of the relative specific activity of partially purified
fractions from scrapie-affected brains may not be correct, if the titre in those fractions and the
starting homogenate were underestimated to different extents. These results also compromise
attempts to estimate titre solely from incubation period. Except where strictly equivalent
samples are being compared, assumptions that dose incubation period relationships are
constant cannot be made. Accordingly, the effect of any changes in dose-response curve must be
accounted for, particularly in biochemical work, before attempting to use incubation periods as
a 'quick' assay for scrapie infectivity.
The authors are grateful to Dr Eugene Sersen for statistical analysis. W e also t h a n k Ms Sharon Callahan for
technical assistance and Mrs Leatrice Kolodney for typing the manuscript.
REFERENCES
CHANDLER, R. L. (1963). Experimental scrapie in the mouse. Research in Veterinary Science 4, 276-285.
COHEN, J. & COHEN, P. (1975). Applied multiple regression/correlation analysis for the behavioral science.
Hillsdale, N.J. : Lawrence Erlbaum Associates.
COHEN, R. S., BLOMBERG,F., BERZINS,K. & SIEKOVITZ,P. (1977). The structure of postsynaptic densities isolated from
dog cerebral cortex. Journal of Cell Biology 75, 181 203.
DICKINSON,A. G. & FRASER,H. (1969). Modification of the pathogenesis of scrapie in mice by treatment of the agent.
Nature, London 222, 892-893.
DICKINSON, A. G., MEIKLE, V. M. H. & FRASER, H. (1969). Genetical control o f the concentration of the M E 7 scrapie
agent in the brain of mice. Journal of Comparative Pathology 79, 15-22.
DOUGHERTY, R. M. (1964). Animal virus titration techniques. In Techniques in Experimental Virology, pp. 169-223.
Edited by R. J. C. Harris. London & N e w York: Academic Press.
HUNTER, G. D. & MILLSON,G. C. (1964). Further experiments on the comparative potency of tissue extracts from
mice infected with scrapie. Research in Veterinary Science 5, 149-153.
HUNTER, G. D., MILLSON, G. C. & CHANDLER, R. L. (1963). Observations on the comparative infectivity of cellular
fractions derived from homogenates of mouse-scrapie brain. Research in Veterinary Science 4, 543-549.
KIMBERLIN, R. H. (1977). Biochemical approaches to scrapie research. Trends in Biochemical Science 2, 220-223.
KIMBERLIN, R. H. & WALKER,C. A. (1978). Pathogenesis of mouse scrapie: effect of route of inoculation o f infectivity
titres and dose response curves. Journal of Comparative Pathology 88, 39-47.
KIMBERLIN, R. H. & WALKER,C. A. (1980). Pathogenesis of mouse scrapie: evidence for neural spread of infection to
the CNS. Journal of General Virology 51, 183-187.
LAX, A. J., MILLSON, G. C. & MANNING, E. I. (1983). C a n scrapie titres be calculated accurately from incubation
periods? Journal of General Virology 64, 971-973.
MALONE, T. G., MARSH, R. F., HANSON, R. P. & SEMANClK, J. S. (1978). M e m b r a n e free scrapie activity. Journal of
Virology 25, 933-935.
MOULD, D. L., DAWSON,A. McL. & SMITH, W. (1967). Determination of the dosage-response curve of mice inoculated
with scrapie. Journal of Comparative Pathology 77, 387 391.
OUTRAM,G. W. (1976). The pathogenesis of scrapie in mice. In Slow Virus Diseases of Animals and Man, pp. 324357. Frontiers of Biology series, vol. 44. Edited by R. H. Kimberlin. A m s t e r d a m & New York: NorthHolland.
PRUSINER, S. B., C(~HRAN, S. P., GROTH, D., HADLAY, D., MARTINEZ, H. M. & HADLOW, W. J. (1980a). A g i n g of the
nervous system and prolonged incubation periods of the spongiform encephalopathies. In Aging of the Brain
and Dementia, pp. 205-216. Edited by L. Amaducci, A. N. Davison & P. Antuono. N e w York: R a v e n Press.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:53:31
Short communication
2050
PRUSINER, S. B., GROTH, D. F., COCHRAN, S. P., MASIARZ, F. R., McKINLEY, M. P. & MARTINEZ, H. M. (1980b). M o l e c u l a r
properties, partial purification, and assay by incubation period measurements of the hamster scrapie agent.
Biochemistry 19, 4883~4891.
PRUSINER, S. B., COCHRAN, S. P., GROTH, D. F., DOWNEY, D. E., BOWMAN, K. A. & MARTINEZ, H. M. (1982). M e a s u r e m e n t
of the scrapie agent using an incubation time interval assay. Annals of Neurology 11, 353-358.
ROHWER, R. G. & GAJOOSEK,D. C. (1980). Scrapie - virus or viroid. The case for a virus. In Search for the Cause of
Multiple Sclerosis and Other Chronic Diseases of the Central Nervous System. Edited by A. Boese. Weinheim:
Verlag Chemie.
SOMERVILLE, R. A., MILLSON, G. C. & KIMBERLIN, R. H. (1980). Sensitivity o f s c r a p i e i n f e c t i v i t y to d e t e r g e n t s a n d 2-
mercaptoethanol, lntervirology 12, 126-129.
(Received 2 March 1983)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 08:53:31