. 35
( 45 .)


isolate of scrapie (8). The Sine gene has been shown to be linked closely to
the PrP gene by RFLP analysis (9) and Sine congemc mice encode different
PrP proteins (IO), suggesting PrP may indeed be the product of the Sznc locus,
Allehc forms of PrP have also been linked to incubation periods of scrapte
m hamsters (12) and sheep (12), and are associated with the incidence of
Gerstmann-Straussler syndrome (GSS) and Creutzfeldt-Jakob disease m
humans (I 3-15)
7.3. PrP Gene Dosage and Scrapie Disease
Transgemc models have been produced to investigate the mechanisms of
these diseases. Introduction of hamster PrP genes into mice has shown that
mcreasmg the copy number of the PrP gene reduces the incubation period of
the disease and that the species type of PrP expressed alters susceptibihty of
the mice to specific isolates of scrapie (Z6). Transgenic mice with high copy
numbers of the Pm-pb allele of the murme PrP gene also were shown to have
shorter mcubation periods when inJected with the Chandler scrapte isolate than
their nontransgemc littermates (2 7). Transgemc mice with high copy numbers
of the murme PrP gene contammg a codon 101 prolme to leucme mutation
spontaneously develop neurodegeneration, spongiform changes m the brain,
and astrogliosts (IS). However, overexpression of the wild-type PrP gene has
also been show to lead to a lethal neurological disease mvolvmg spongiform
changes m the bram and muscle degeneration (19).
Although these experiments have shown that increasing the copy number of
the PrP gene leads to shortening m incubation periods of the disease, random
integration of multrple coptes presents with climcal artifacts that may not
accurately reflect the disease process. More appropriate models to examme the
effect of PrP gene dosage and the role of mutations m the PrP gene may be
produced by the mtroduction of specific alterations mto the endogenous PrP
gene using gene targeting techmques.
2. Methods to Produce PrP-Deficient Mice
Introduction of foreign DNA mto cells can lead to recombmatton of the
DNA with homologous sequencesof the endogenous DNA. This homologous
recombmatton event, known as gene targeting, has been used to introduce
PrP- Deficien t Mice 239

mutations into genes m murme embryonic stem cells. These cells then can be
used to produce mice with one or two copies of the mutated gene Thts allows
the assessmentof the specific mutations within a gene m the mouse
Thts method of production of transgemc mice has enabled the function of
many genes to be assessedby deletion of the gene products. As a tool m the
field of the TSEs, gene targeting has already addressed a number of important
questions through the mtroduction of mutations m the PrP gene that have lead
to the production of PrP-deficient mice.
2.1. Targeting Vectors
Two dtfferent strategies have been used to introduce mutations mto the PrP
gene, leading to the production of PrP-deficient mice (20,21). Both approaches
have used a replacement targeting vector with a positive selection marker for
DNA integration, the neomycin phosphotransferase gene (neo) (Fig. 1). In the
first approach PrP sequencesused m the vector were designed to replace 552 bp
of the coding sequences of PrP (posmons 10-562) with GeOunder control of
the thymidine kmase (tk) promoter (Fig. IA) (20). An alternative strategy was
to insert neo under control of the metallothtoneme promoter mto the PrP cod-
mg sequence. This led to homologous recombrnatton at the PrP locus without
deletton of any PrP sequences(Fig. 1B). In this approach the tk gene was used
as a negative selection against random integration of the DNA (21).
2.2. Targeting Mouse Embryonic Stem Cells
The targeting vectors were linearized outstde the PrP sequences and
electroporated into embryonic stem cells derived from (A) agouti 129/SV(ev)
or (B) 129/0la mice. Selectton was carried out and targeted clones were
selected m (A) by PCR and in (B) by Southern analysts of genomrc DNA. The
frequency of homologous recombmatton m (A) was 1 m 5000 G4 18-resistant
colonies compared wtth 1 m 800 in (B). These differences may reflect the dtf-
ference m the PrP sequencesused in the targeting vectors.
2.3. Production of Mice Heterozygous
and Homozygous for the Mutant Allele
The mutant cells lines produced by Btieler and coworkers (20) were inJected
mto 4-d-old blastocysts from C57BL/6J. The chimeric male mice produced
were crossed with C57BL females to produce an outbred lme of mice both
heterozygous (PrPO”) and homozygous (PrPO™O) the mutant allele.
In the experiments carried out by Manson et al, (21) the mutant cell lines
were injected mto 4-d-old blastocysts from C57BLKBA mice and the male
chimeric mice bred wtth 129/Ola females to produce an inbred lme of mice
heterozygous (PrP-I™) and homozygous (PrP”) for the mutant allele. I29/0la
240 Manson

Targeting vector B Targeting vector A

I,..,- JfK

PrP gene I
exon I exon II exon III

Mutated PrP gene locus using targeting vector El

Mutated PrP gene locus using targeting vector A

Fig. 1. Targeting strategies for the production of PrP null mice. Vector A. Neomycin
phosphotransferase gene (neo) under the control of the thymidine kinase promoter,
was inserted into exon 3 of the PrP coding region replacing 552 bp of PrP sequence
that had been removed (positions 10-562). The vector was linearized outside the PrP
sequence and electroporated into 129/SV//ev mouse embryonic stem cells. Homolo-
gous recombination of the targeting vector with the endogenous PrP locus resulted in
a mutated PrP gene locus lacking 552 bp of PrP and with neo inserted into exon 3.
Vector B. Neo under control of the metallothioneine promoter was inserted into exon
3 of the PrP gene. The thymidine kinase gene (tk) was inserted 5™ of the PrP sequences
to select against random integration. The vector was linearized outside the PrP coding
region and electroporated into 129/01a mouse embryonic stem cells. Homologous
recombination of this vector with the endogenous locus results in the lossof the tk
and a mutated PrP gene with neo gene inserted into exon 3. No PrP sequences have
been removed in this strategy.

mice are poor breeders and this approach is considerably slower than outbreed-
ing. Nevertheless, it was considered important for the scrapie experiments in
these animals to produce an inbred line of mice. In this way differences between
wild-type and mutant mice can be attributed solely to the PrP gene.
2.4. PrP mRNA Analysis of Mutant Mice
PrP mRNA analysis of PrP mutant mice by Northern blot has shown both
the inbred (PrP+˜-) and outbred (PrP+˜O) heterozygous lines have approx 50%
PrP-Deficient Mice
of the wild-type PrP mRNA. In sztu hybridization studtes of the inbred mace
have shown the reductton in PrP mRNA is throughout the brain (22). No PrP
mRNA was detected in the brains of the inbred mice using probes either 3™ or 5™
of the neo gene. The outbred mice were shown to contam substantial quanttttes
of a chimerlc mRNA containing the neo and residual PrP sequences. Reduced
levels of the PrP-neo RNA molecules were also detected m the outbred het-
erozygous mace (20).
2.5. PrP Protein Analysis of Mutant Mice
Protein extracted from membrane fractions of mouse brains was analyzed
by Western blotting. No PrP protein was detected in either of the homozygous
mutant fines. The PrP protein detected in the heterozygous lines was qualita-
trvely identical to that detected m the wild-type mice but was significantly
reduced in amount. Quantitation of the PrP protein m the outbred heterozygous
mice estimated it to be 50% of that detected in the wild-type mice (20) whereas
m the inbred heterozygous line the estimated amount of PrP protein was approx
70% of the wild-type mice (22). These differences may srmply reflect the prob-
lems of accurate quantitation by thts technique or alternatively may reflect actual
differences in protein amount owing to the different targeting strategies used.

3. Use of PrP-Deficient Mice in Scrapie Research
3.1. The Role of PrP in Disease Pathology
3. I. 1. Onset of Disease
Mice with no functional copies of the PrP gene have been shown to survive
scrapre infection when inJected with Chandler Isolate of scrapie (23) or the
ME7 strain (22). No disease symptoms were observed m these mice over 500 d
after moculatton, whereas termmal stages of the disease are detected in wild-
type mice around 150 d. These results have clearly demonstrated that PrP gene
expression IS required for the development of scrapie disease.
In macewith one functional copy of the PrP gene, the time of onset of disease
is delayed (22-241, but the amount by which the onset is delayed in the inbred
heterozygotes appears to differ, depending on the strain of scrapie (Table 1). It
has not yet been established tf differences between the outbred and inbred PrP-
deficient mouse lines also will lead to differences in timing of the onset of
disease, when injected with the same strain of scrapie.
Whereas these experiments have shown that decreasing the copy number of
the PrP genes lengthens the incubation period, they have also shown that there
is no simple formula relating the PrP gene copy number and incubation period
of disease. The strams of mice used m the gene targeting experiments contain
the Pm-pa allele of the PrP gene. The mcubation period for the 1290la PrP+˜-

Table 1
Incubation Period of PrP+˜+, PrP+˜-, and PW- Mice Infected with Scrapiea
Mouse Sine d/p7
PrP+˜- PrP™- (Pm-pa/pm-pb)
genotype PrP+˜+

ME7 154+O(n=4) 280 14 (n = 8) >530 251 f2
284 3˜0 (n = 4)
>440c (n = 2 1)
301c 154+ l™(n= 14) 230+2™(n=21) 505 f 5b
Sine s7lp7
genotype PrP+˜+ PrPO™+ (Pm-pa/Pm-pb)

RML 171 k 11 415f30(n= 15) >600 268f4
OThe time-pomt at the terminal stages of scraple for a group of ammals of each genotype was used to
give an estimate of mcubatlon period of ME7,3OlC strams m Inbred lines of mice (129/0la PrP+˜+, PrP+˜;
and Prp-) and chandler isolate m the outbred lmes (PrP+˜+, PrPO™+, and PrPO™O) These mcubatlon periods
are compared with the mcubatlon penod for each scrapte stram/lsolate m Bnc s7/p7 (Pm-pa/Pm-pb) mice
to determine the role of the different alleles of PrP m determmmg the mcubatlon period of the disease
*Isolate passaged three times m C57Bl mice
˜Isolate passaged four times m C57Bl mice (Incubation periods for Bnc s7/p7 and Sznc p7/p7 Iden-
txal to passage 3)

infected with the ME7 strain of scrapie is approximately twtce that of the
129/0la PrP+˜+and IS simtlar to that obtained with a heterozygous mouse con-
taming a Pm-pa and a Pm-pb allele of the PrP gene. These results suggest that
for the ME7 strain the Pm-pa allele of the PrP gene 1sthe predominant allele
mvolved m either generating the ME7 agent or interacting with it. This 1snot,
however, the case with other strains of scrapte. PrP”˜+ mice injected with the
RML scrapie isolate had incubation periods (400-465) more prolonged than
the heterozygous Pm-pa/Pm-pb mice (268 d) (23,24) and PrP” infected with
30 1C isolate have mcubatron periods considerably shorter than the Pm-pa/Pm-pb
mice (Table 1).
Gene dosage 1sclearly not the only factor involved in the mcubatron period
of disease. The mteractton of the different alleles of the PrP gene wtth speck
strains of scrapre and also genes other than PrP can be investigated using the
PrP-deficient mrce, into which multiple copies of the Pm-pa or Pm-pb allele
have been mserted. This type of approach has been used to show that the Pm-p”
allele of PrP is the major determinant of mcubatron period m mice infected
with the RML isolate of scrapre smce mcreasmg copy numbers of the Pm-pa
allele leads to a reduction in mcubation periods. However, expresston of the
Pm-pb allele appears to confer resistance to disease in mice infected with 87V
strain of scrapie and increasesthe incubation time with 22A strain infection (25).
PrP-Defhent Mice
3.1.2. PrP Accumulation in Deficient Mice
In PrP+˜- mice infected with the ME7 strain of scrapie, PrP deposition startsm
the samebrain area asthe wild-type mice and can be detectedas early as 50 d. The
pattern of PrP deposition in the brain of heterozygotesfollows an identical course,
but builds up more slowly than m the wild-type mice. By the terminal stagesof
disease the amount detected in the brain by immunohistochemical techniques 1s
identical in the PrP+˜- and PrP+˜+ mice (22). These results contrast with the PrP”˜+
and PrP+˜+mice injected with the RML isolate in which the heterozygote animals
were shown to have levels of PrPscdetected by immunoblotting, comparable to
their terminally ill counterparts for many months prior to clinical diseaseand death
(26). These expenments indicate that PrPSC accumulation 1snot limited by the lev-
els of PrPc and raise questions asto whether there is a causal relationship between
PrPSC accumulation in the brain and the development of the clinical disease.
3 1.3. PrP Gene Dosage and Severity of Disease
The timing of disease onset and progression of disease symptoms is delayed
in the PrP-deficient mice (22,26). However, there appears to be no difference
m the severity of the clinical signs and the pathology in the brain at the termi-
nal stages of diseases (22). In PrP+˜- mice injected with ME7 the pattern and
degree of vacuolatlon was similar to that detected in the wild-type mice and
disease symptoms, although delayed, were no less severe than those seen in the
wild-type mice (22) (Fig. 2).
Suggestions have been put forward for the use of therapies that regulate the
production of PrP as a means of controllmg these disease (23) These findings
suggest that the use of such measures may delay the onset and progression of
the disease but have no effect on the final severity of disease.
3.2. The Role of PrP in Agent Replication
PrPO™O mice maculated with the RML isolate of scrapte were reslstant to
disease (23,24). Brain homogenates from these animals at 2, 8, 12, and 25 wk
after infectivity showed no infectivity. A sample from 20 wk showed low titer
infectivity m a group of mice and in a repeat experiment in only one of a group
of mice (23,27). Although these results indicate that PrP is necessary for agent
replication, whether contamination, residual infectivity, or low level propaga-
tion account for the occasional case of scrapie remains to be established.
Initial experiments with the PrP+˜+and PrP”˜+ mice infected with the RML
isolate of scrapie indicate that mfectlvity in the wild-type and PrP-deficient
mice reaches similar maximal levels by 20 wk postinfectlon (26). These results
suggest that the reduced levels of PrPC m the heterozygous mice do not limit
agent replication, but whether there 1sany alteration in the rate of rephcation
requires more detailed analysis.
244 Manson

1 2 3 4 5 6 7 6 9

Brain Areas

Fig. 2. Leston profile for 129/01a PrP+˜+ and PrP+˜- mice Infected wtth ME7 stram
of scrapte The amount of vacuolatlon m tune areas of the bram at the terminal stages
of dtsease were used to calculate the leston profile. A leston profile was produced for


. 35
( 45 .)