. 34
( 45 .)


passaged strains isolated from five natural scrapie sheep have been found to
differ between sources (16).
5.2. Strain Typing of BSE Isolates
BSE has been transmitted to mice from a series of unrelated cattle sources,
collected at different times during the epidemic, from widely separated loca-
232 Bruce

BSE 1-7 pooled l
0 A n Hi

FSE l-3 pooled l IukY ml

Kudu HA
0 n

n to&t w

200 300
100 400 500 700 800

Mean incubation period (days + S.E.M.)

Fig. 4. Incubation periods m a panel of mouse strains on primary transmrssron from
seven cattle with BSE (pooled data), three cats with fehne sponglform encephalopathy
(FSE) (pooled data), one greater kudu, and one nyala The mouse strams are RI11
(SznP™) (0), C57BL (SW&˜) (O), VM (Szncr™) (A), IM (SzncP™) (A), and C57BL x VM
(SznCS™q (rn).

tions within the United Kingdom (I I, Z7). Seven such transmtssions have been
completed. The results of these experiments were remarkably similar to each
other and differed from those m all previous and contemporary transmtsstons
of scrapie. All seven BSE sources have produced a charactertsttc pattern of
incubation periods and pathology m a standard panel of inbred mouse strains
and crosses (see Figs. 4 and 5). There are large and consistent differences m
mcubatron period between mouse strains of different Sine genotypes and also,
surprisingly, between mouse strains of the same Sine genotype (e.g., C57BL
and RIB). Further passages m Szrzcs7 Sin&” mace have produced two strains,
30 1C and 3OlV, that differ from all strains derived from sheep or goat scrapte
(ZZ) (see Ftg. 1).
The umformtty of the BSE transmissions shows that each cow was infected
with the same major strain of agent The conststency of the pathology reported
m cattle with BSE also suggests the mvolvement of a single or a limited num-
ber of strains (IS). The major strain of BSE in cattle appears to be different
from the strains causing scrapte m sheep. This does not necessarily undermine
the widely held assumption that BSE originated from rendered sheep scrapie
offal in feed; a possible explanation is that the high temperatures involved n-r
rendering and subsequent passage through cattle have selected variant strains
from sheep scrapre.
Stram Typrng Stud/es of Scrapie and BSE


1 2 3 4 5 6 7 8 9
Scoring Area
Fig. 5. Lesion profiles m RI11 mace m transmrssrons from cattle, cats, greater kudu,
and nyala, using data from the experiments shown In Frg 4 No whrte matter vacuola-
non was seen m any of these experrments.

5.3. Strain Typing
of Novel Spongiform Encephalopathies from Other Species
Recently, transmtsslons to mice have been achieved from other species with
novel spongiform encephalopathies, suspected to be related to the BSE ept-
demrc; the sources were three domestrc cats, a greater kudu, and a nyala. The
results of all five transmissions were strikmgly stmrlar to results from cattle
sources, indicating a common source of infection among these species (2 Z,I9)
(Ftgs. 4 and 5). BSE from cattle also has been transmitted experrmentally to
sheep, goats, and prgs and then from each of these species to mice, again, the
results were similar to direct transmissions of BSE (11). These studies show,
first, that the BSE agent IS unchanged when passaged through a range of spe-
cres and, second, that the donor species has little specific mfluence on the dis-
ease characteristics of BSE on transmission to mice.
It IS clear from the preceding that transmission and strain typing studies can
be used to answer epidemrologlcal questions, for example, to estabhsh links
between the natural drseases in different specres or m drfferent countrtes. One
obvious application would be to test any future suspicion that BSE has spread
to another species, for example, to humans. However, it should be stressed that
the methods are cumbersome and slow and that such studies can be undertaken
234 Bruce
only on a limited scale. Also, this approach can confirm but not refute a sus-
pected lurk, because there is always the possibility that passage through a new
species has selected a variant strain.

6. The Molecular Basis of Strain Variation
The molecular nature of the agent 1s still a matter for speculation There are
three mam hypothetical models*

1 A “pnon” (20), conslstmg solely of modrtied forms of the host protein, PrP,
2 A “vmno” (21), conslstmg of an mfectlon-specific mformatlonal molecule (possibly
a nucleic acid) that IS closely associated with and protected by abnormal PrP, and
3 An unusual but conventlonal V˜I,IS (22)

The existence of multiple strains of agent dictates that any proposed struc-
ture should have a repbcable mformational component Furthermore, the sta-
bility of strains on passage through different host genotypes or species sets a
number of conditrons for the validity of any particular model. If the agent con-
tams its own nucleic acid, strain variatton would be analogous to that seen m
conventional microorgamsms. It IS more difficult to envisage how a protem
alone could specify stram diversity.
According to the “priori” hypothesis, the pathogen IS PrP, which has been
modified m some specific way (3). This abnormal protein is suggested to
Induce the same modification m new host PrP molecules, by direct mteraction.
The modtfication currently is thought to be conformattonal (24), with specific
conformations determmmg strain properties. There is recent evidence from m
vitro studies of two hamster-passaged strains of transmissible murk encepha-
lopathy that such a model might be feasible (2.5).
If strain specificity resides solely m PrP structure, there must be as many
spectfic modtficattons as there are distinct strains and each must be able to
“repltcate” itself accurately over many serial passages, apart from predictably
generating other specific modifications. Multiple forms of modified PrP must
be capable of retammg their separate identities when passaged as mixtures and
such mixtures must be resolvable by btologtcal cloning. Furthermore, it must
be possible to reproduce strain-spectfic modifications faithfully m PrP mol-
ecules with different ammo acid sequences. It remains to be seen whether PrP
alone can fulfill all these criteria, or whether a separate mformational molecule
1s required, such as a nucletc acid.

The author would like to thank Irene McConnell, Patricia McBride, and their
staff for all their hard work over the years, and Aileen Chree and Laurence
Doughty for help m preparing the manuscript.
Strain Typing Stud/es of Scrapie and BSE 235

1 Bruce, M. E , McConnell, 1, Fraser, H , and Dickmson, A G. (1991) The disease
characterlstlcs of different strams of scrapie m Sznc congemc mouse hnes Impll-
catlons for the nature of the agent and host control of pathogenesls. J Gen Vu-01
Bruce, M E., Fraser, H , McBride, P. A., Scott, J. R., and Dlckmson, A G (1992)
The basisof stram variation rn scrapie, m Prcon Diseases of&mans and Animals
(Prusmer, S. B , Collmge, J , Powell, J., and Anderton, B., eds.), Elhs Horwood,
ChIchester, pp. 497-508
Dickinson, A. C., Melkle, V. M. H., and Fraser, H (1968) Identlficatlon of a gene
which controls the mcubatlon period of somestramsof scrapie m mice J Comp
Path01 78, 293-299
4 Dlckmson, A, G. and Melkle, V M H. (197 1) Host-genotype and agent effects m
scraple mcubatlon change m allelic mteractlon with different strams of agent
Mol Gen Genet 112,73-79
Westaway, D , Goodman P A., Mirenda, C A , McKinley, M. P , Carlson, G A ,
and Prusmer, S. B (1987) Dlstmct prlon proteins m short and long scraple incu-
bation period mice Cell 51,65 l-662,
Hunter, N , Dann, J. C., Bennett, A. D , Somerville, R. A , McConnell, I , and
Hope, J (1992) Are Sine and the PrP gene congruent? Evidence from PrP gene
analysis m Slnc congemc mice J Gen Vwol 73,275 l-2755.
Fraser, H and Dickinson, A. G (1968) The sequential development of the bram
lesions of scraple m three strainsof mice. .I Comp Path01 78, 301-3 11
8 Bruce, M E , McBride, P A , and Farquhar, C F (1989) Precisetargetmg of the
pathology of the slaloglycoprotcm, PrP, and vacuolar degeneration m mouse
scraple. Neurosck Lett 102, l-6.
Bruce, M. E , McBride, P. A., Jeffrey, M., and Scott, J R (1994) PrP m pathology
and pathogenesism scrapte-infected mice Mol Neurobiol 8, 105-l 12
10. Dickinson, A. G. (1976) Scrapie m sheepand goats, in Slow Vwus Dzseasesof
AnzmalsandMan (KImberlin, R. H., ed.), North Holland, Amsterdam, pp 209-24 1
11 Bruce, M E., Chree, A , McConnell, I., Foster, J., Pearson, G , and Fraser, H.
(1994) Transmlsslonof bovine sponglform encephalopathy and scraple to mice:
strain variation and the speciesbarrier Phzl. Trans R. Sot Lond B 343,405-4 11,
12 Bruce, M. E. and Dlckmson, A G (1987) Biological evidence that scraple agent
has an independentgenome J Gen. Viral 68,79-89.
13 Klmberlm, R H., Walker, C. A , and Fraser, H. (1989) The genomlc IdentIty of
different strams of mouse scraple IS expressed m hamsters and preserved on
relsolatlon in mice J Gen. Vwol 70,2017-2025.
14 Klmberlm, R H., Cole, S , and Walker, C. A. (1987) Temporary and permanent
modifications to a smgle strain of mousescrapleon transmissionto rats and ham-
sters J Gen Vzrol 68, 1875-l 881
15 Fraser, H. (1983) A survey of primary transmissionof Icelandic scraple (nda) to
mice, m Vwus Non Conventronnels et Affections du Systtme Nerveux Central
(Court, L A., ed), Masson, Pans, pp. 34-46
236 Bruce
16 Carp, R I and Callahan, S M (1991) Vartatton m the charactertstxs of 10 mouse-
passaged scrapre lines derived from five scrapte-posmve sheep J Gen Vzrol 72,
17 Fraser, H , Bruce, M E , Chree, A., McConnell, I , and Wells, G A H (1992)
Transmtsston of bovine spongtform encephalopathy and scrapte to mice J Gen
Vwol 73, 1891-l 897
18 Wells, G A H , Hawkms, S A C., Hadlow, W J , and Spencer, Y I (1992)The
dtscovery of bovme spongtfotm encephalopathy and observations on the vacuolar
changes, m Prlon Diseases of Humansand Animals (Prusmer, S B., Collmge, J ,
Powell, J., and Anderton, B., eds ), Elhs Horwood, Chrchester, pp. 256-274
19 Fraser, H., Pearson, G. R., McConnell, I , Bruce, M. E , Wyatt, J M., and
Gruffydd-Jones, T J (1994) Transmtsstonof felme spongiform encephalopathy
to mice. Vet Ret 134,449.
20 Prusmer, S. B (1982) Novel protemaceousmfecttous particles causescrapte SCI-
ence 216, 13&144
21 Dlckmson, A G and Outram, G W. (1988) Genettc aspectsof unconventtonal
vnus infecttons™ the basts of the vumo hypothesis, m CubaFoundatzon Sympo-
sium 13.5 Novel Infectrous Agents and the Central Nervous System (Bock, G and
Marsh, J , eds.), Wiley, Chtchester, pp 63-83
22 Rohwer, R G ( 1991) The scrapte agent: “a vu-usby any other name ” Curr Top
Mcroblol Immunol 172, 195-232
23 Prusiner, S B. (1992) Prton biology, m Przon Diseasesof Humans and Animals
(Prusmer, S. B., Collmge, J , Powell, J., and Anderton, B , eds.), Ellis Horwood,
Chtchester, pp. 533-567
24 Baldwin, M. A , Pan, K -M , Nguyen, J , Huang, Z , Groth, D., Serban, A , et al
(1994) Spectroscoptccharactertzatton of conformattonal differences between PrPC
and PrPSCPhrl Trans R Sot Lond B 343,435-441
25 Bessen,R. A , Koctsko, D. A., Raymond, G. J , Nandan, S., Lansbury, P. T., and
Caughey, B. (1995) Non-genetic propagation of stram-specific properties of
scrapte prton protem Nature 375698-700
PrP-Deficient Mice in the Study
of Transmissible Spongiform Encephalopathies
Jean C. Manson

1. Introduction
The Transmissible Spongiform Encephalopathies (TSEs), such as scraple,
BSE, and Creutzfeldt-Jakob disease, are associated with alteratlons in the neu-
ral membrane protein or prion protein (PrP). This chapter will outline the gene
targeting approaches that have been used to mutate the murine PrP gene,
resulting in mice with reduction or absence of the PrP protein. It will then
describe how these transgenlc animals can contribute to our understanding of
the role of PrP m agent rephcatlon, the pathology of the TSEs, and the normal
function of PrP
1.1. PrP Gene Expression in Infected and Uninfected Animals
The PrP protein is a protease-sensitive cell surface glycoprotein anchored in
the membrane by a glycomositol phospholipid (PrPc) (I), but during the course
of scraple infection the PrP protein aggregates and accumulates in and around
the cells of the brain as protease-resistant deposits (PrPsc). The distribution of
the PrP protein m infected brain is dependent on both the host genotype and the
strain of scraple (2). The amount of PrPSC detected in the brains of mice or
hamsters mfected with scrapie is tenfold greater than the amount of PrPC
detected m the uninfected brain (3).
PrP mRNA 1sdetected m neuronal cells throughout the brain with different
amounts of mRNA being detected m different populations of neuronal cells
(4). In contrast to the alterations m PrP protem detected during disease, there 1s
no difference in the amount or localization of PrP mRNA in brains either
uninfected or infected with different strains of scrapie, as detected by Northern
analysis m hamsters (5) or by in situ hybridization m mice (4).
From Methods m Molecular Medmne Pnon CWseases
Edited by H Baker and R M Ridley Humana Press Inc , Totowa, NJ

238 Manson
1.2. Allelic Variants of PrP
The Sznc gene has been shown to be the maJor gene controllmg survival
time of mice exposed to scrapie (6,7) and ammals homozygous for the alleles
of Sznc s7 and Sine p7, have alleltc forms of the PrP gene The PrP allele with
ammo acid 108 Leu and 189 Thr is associatedwith short mcubation times when
infected with Chandler isolate of scrapie (Pm-p”), whereas the allele with 108 Phe
and 189 Val (Pm-pb) IS associated with long mcubation periods with the same


. 34
( 45 .)