Available via license: CC BY 4.0
Content may be subject to copyright.
Page 1/25
Select Gene Expression Across Variable PrPSc
Permissive Ovine Microglia
Valerie McElliott
University System of Georgia https://orcid.org/0000-0002-2405-8333
Kelcey Dinkel
Washington State University
Zachary Nesbit
University System of Georgia
James B. Stanton ( jbs@uga.edu )
https://orcid.org/0000-0002-7661-2631
Research article
Keywords: Classical scrapie, Prion permissibility, Ovine microglia, Fibronectin, Survivin, PRNP
Posted Date: August 26th, 2019
DOI: https://doi.org/10.21203/rs.2.13583/v1
License: This work is licensed under a Creative Commons Attribution 4.0 International License.
Read Full License
Page 2/25
Abstract
Abstract Abstract Background: Transmissible spongiform encephalopathies (TSEs) are a group of fatal,
neurodegenerative diseases that affect multiple species, including sheep, cattle, and humans. A
misfolded, pathogenic isoform (PrPD) of the normal, host-encoded, cellular prion protein (PrPC) is the
causative agent for TSEs. While there have been advances in understanding TSEs, antemortem
diagnostic tests are limited in many species, and there are no effective treatment protocols. Filling these
reagent gaps will require knowledge of the molecular pathophysiology of PrPD accumulation. Previous
work has suggested that the extracellular matrix (i.e., bronectin 1) and physiological functions (i.e., cell
division) maybe key factors for cellular prion permissibility, at least in specic cell culture models. Using a
natural scrapie isolate, six immortalized, ovine microglial clones, of varying permissiveness to classical
scrapie were evaluated for differential gene expression in seven genes based on previous RNASeq studies
(bronectin 1 [FN1], follistatin-like 1 [FSTL1], osteonectin [SPARC], survivin [BIRC5], syndecan 4 [SDC4],
AXL receptor tyrosine kinase [AXL], and prion protein [PRNP]), and to determine correlations with prion
permissibility. Results: Signicant differential gene expression was frequently observed for survivin,
follistatin-like 1 and osteonectin between clones, and when evaluated relative to PRNP expression.
However, only bronectin 1 and survivin were signicantly correlated with prion permissibility, and only
when evaluated relative to PRNP expression. Inoculation had a signicant effect on follistatin-like 1,
syndecan 4, and osteonectin. Conclusions: Similar to previous studies in other systems, bronectin and
mitotic rate show promise as potential determinants of prion permissibility in ovine microglia. As
determinants of prion permissibility, the expression of bronectin 1 and survivin coupled with PRNP could
be utilized as biomarkers for detection of prion permissibility phenotype in ovine microglia, and perhaps
other cell culture models of prion disease.
Background
Transmissible spongiform encephalopathies (TSEs) are a group of progressive, fatal neurodegenerative
diseases that affect cattle (bovine spongiform encephalopathy), small ruminants (scrapie), cervids
(chronic wasting disease), felids (feline spongiform encephalopathy), humans (Creutzfeldt-Jakob disease
and Kuru) [1, 2], and now camelids [3]. Classical scrapie is the oldest described prion disease, dating back
to over 250 years ago [4]. Due to the public concern for all prion diseases, and the economic and nancial
devastation associated with this disease, eradication efforts have been implemented in various countries
for the control and elimination of classical scrapie.
TSEs are caused by a misfolded, pathogenic isoform (PrPD; D for disease-associated) of the normal, host-
encoded, cellular prion protein (PrPC; C for cellular) [5-10]. In this disease, PrPD can promote autocatalytic
conversion of normal PrPC to PrPD in a perpetuating, positive feedback cycle [5-9]. While most TSEs are
initiated after exposure to PrPD from an infected animal, PrPC has also been reported to spontaneously
convert to PrPD [10-12]. Although there are variations in several of the key components contributing to
prion infection in hosts, which include heterogeneity of PrPD strains [13], neuroanatomical distribution of
Page 3/25
lesions and PrPD deposits [2, 13], and post-translational PrPC modications [9], the common denominator
is the conversion of PrPC to its pathological isoform, PrPD.
PrPC is a protein encoded by the host gene PRNP which is evolutionarily conserved amongst mammals
[1, 14, 15] and is constitutively expressed in several cell types, but notably in adult neurons [16], glial cells
[17], lymphocytes, and monocytes [18]. In sheep, PrPC is 256 amino acids long [15, 19]. Polymorphisms
in PrPC have been reported in codons 136 (A/alanine or V/valine), 154 (R/arginine or H/ histidine) and/or
171 (Q/glutamine, R/arginine, or H/histidine) that increase susceptibility (i.e. VRQ/VRQ, ARQ/VRQ,
ARQ/ARQ) or confer resistance (ARR/ARR) to classical scrapie in sheep [15, 20, 21].
Although there are variations in several of the key components contributing to prion infection in hosts,
which include heterogeneity of PrPD strains [13], neuroanatomical distribution of lesions and PrPD
deposition [2, 13], and post-translational PrPC modications [9], the common denominator is the
conversion of PrPC to its pathological isoform, PrPD. While certain cofactors such as glycosaminoglycans
[22], and lipids [23] have been suggested to facilitate this conversion, the intricate mechanisms
contributing to this conformational change are not completely elucidated. Hence, this generalized paucity
of knowledge in prion disease pathogenesis hinders the development of diagnostic aids, and of most
importance, treatments for prion disease.
Various experimental model systems have been used to study and further characterize classical scrapie.
In vivo models provide excellent systems to study the clinical aspects of the disease, and serve as the
nal test for potential therapies [24-26]. However, with in vivo models, assessment of cellular and
extracellular factors that may inuence disease onset or progression could be challenging to identify. Cell
culture models serve as practical and informative ex-vivo systems, providing an avenue to study and
dissect the role of cellular and extracellular constituents in prion disease pathogenesis, that cannot be
dissected at the tissue or animal level. Furthermore, these models allow for the development and testing
of treatments prior to animal testing. While transgenic cell culture models, such as ovinized rabbit kidney
epithelial cells (Rov) [27], and ovinized mouse cerebellar astrocytes [1] allow for the expression of PrPC
from a normal prion host (e.g., sheep), and have provided useful information in prion research, these
models cannot account for species-to-species variation in all non-PrPC proteins. To increase the chances
that PrP-non-PrP protein interactions are species matched, non-transgenic cell culture lines can be used.
Immortalized, ovine microglia cells [28, 29] are an example of a non-transgenic cell culture line, and were
hence utilized in this study. Microglia are innate, myeloid immune cells of the brain that have similar
functions to macrophages [30, 31]. These resident phagocytic cells have also been localized to the
neuroanatomical vicinity of PrPD deposits and contain intracellular PrPD, which contributes to prion
protein degradation but also potential dissemination [31]. As a possible mode of prion propagation in the
nervous system, the use of microglia could provide further insight on prion disease pathogenesis.
Using N2a cells inoculated with a mouse-adapted scrapie isolate (ScN2a), Marbiah
et al.
demonstrated
that the downregulation of 9 extracellular matrix genes through silencing increased susceptibility to prion
Page 4/25
infection, and when bronectin 1 was inhibited from binding to integrin α8, prion propagation was
enhanced [32]. Similar investigations into gene expression were also performed using a subline of ovine
microglia established within our laboratory. Munoz-Gutierrez
et al.
evaluated the transcriptomic prole in
immortalized clones of this subline, between one permissive and one relatively resistant to scrapie prions,
and twenty-two genes were differentially expressed [33]. Additionally, several other genes were potentially
differentially expressed or regulated (i.e., AXL, BIRC5, SPARC) [33]. Due to the limited nature of a single
pairwise analysis, the relevance of these ndings is unknown. For this study, we expanded the analysis of
AXL, BIRC5, FSTL1, FN1, SPARC, SDC4 and PRNP into three additional immortalized, ovine microglia
clones of varying prion permissibility, and one clone that reverted from highly to intermediately
permissive from a previous study [28] to determine if their expression correlated with prion permissibility.
Results
3.1 Assessment of gene expression normality for Mock and Utah-inoculated clones
Seven genes with a potential role in prion permissibility were assessed with RT-qPCR,
and normalized to 18s rRNA and hPRT1b. Gene expression data from Mock-inoculated clones showed a
normal distribution (Goodness-of-Fit;
p
> 0.05). Utah-inoculated clones were not normally distributed for
SDC4 (Goodness-of-Fit;
p
< 0.001) and SPARC (Goodness-of-Fit;
p
< 0.01) but were normally distributed
for the remaining genes. For SDC4 and SPARC, raw expression values from Utah-inoculated clones were
transformed with Box-Cox Y transformation test, and these values were deemed as normally distributed
after assessment with histograms and Goodness of Fit test (
p
> 0.05) and subsequently utilized in
parametric statistical analyses as such.
3.2 Transcript levels in Mock-inoculated clones
To determine if the RNA levels of the targeted genes signicantly differed between clones, fold changes in
normalized gene expression were statistically compared amongst all of the clones from at least 3
independent replicates. Statistically signicant differences in expression were observed for BIRC5 (
p
<
0.001), SPARC (
p
< 0.001) and FSTL1 (
p
< 0.01) (Figure 1). For BIRC5, differential gene expression was
signicantly decreased for both 439Late
and 439Early compared to all other clones (
p
< 0.05), and this decreased expression was a four-to-eight-
fold change compared to clone 438. Expression of FSTL1 in clones 439Late (
p
< 0.05) and 439Early (
p
<
0.05) was signicantly decreased compared to clone 434, and clone 439Late was also signicantly
decreased compared to clone 438 (
p
< 0.05), but these changes were less than two-fold compared to
clone 438. Additionally, clone 440 had signicantly decreased expression of FSTL1 compared to clones
438 (
p
< 0.05) and 434 (
p
< 0.05), but these changes were also less than two-fold compared to clone 438.
With SPARC, clones 439Late and 439Early had signicantly diminished expression compared to clones
438 (
p
< 0.0001 and
p
< 0.01) and 434 (
p
< 0.001 and
p
< 0.05), and these changes were two-to-three-fold
compared to clone 438. Additionally, 439Late SPARC expression was signicantly decreased compared
Page 5/25
to clone 441 (
p
= 0.0378). Furthermore, clone 438 had signicantly increased SPARC expression
compared to clones 441 (
p
< 0.05) and 440 (
p
< 0.01), and clone 434 was also signicantly increased
compared to 440 (
p
< 0.05), but these changes were less than two-fold compared to clone 438.
Signicant differential gene expression was lacking for AXL, FN1, PRNP and SDC4 in all pairwise
comparisons.
Since clone 439 exhibited variation in permissibility over time, differences in gene expression were also
assessed for 439Late relative to the most permissive clone, 439Early in all evaluated genes. Only in
BIRC5 was there a two-fold change expression in 439Late relative to 439Early, but this change was not
statistically signicant (
p
> 0.05).
There were no statistically signicant correlations between mean normalized gene expression of Mock-
inoculated clones and prion permissibility (p > 0.05).
3.3 Transcript levels in Utah-inoculated clones
As described in section 3.2, analysis of fold changes in gene expression and correlations with prion
permissibility were also performed for Utah-inoculated clones for at least 3 independent replicates, with
the exception of clone 439Early for BIRC5 in which data from only 2 out of 3 independent replicates was
available to assess (one replicate had undetectable transcript levels, consistent with low expression levels
detected in the other replicates). Individual, signicant variations in gene expression were present for
PRNP (
p
< 0.05) and SPARC (
p
< 0.01) (Figure 2). A single, signicant increase in PRNP expression was
observed for clone 439Late as compared to clone 438 (
p
< 0.05); however, this change was less than two-
fold compared to clone 438. With SPARC, expression levels were signicantly increased for clone 439Late
as opposed to clone 438 (
p
< 0.01), and clone 434 (
p
< 0.01), but this variation was also less than two-
fold compared to clone 438. Although a two-fold change in BIRC5 expression was present for 439Late
and 439Early relative to clone 438, these differences were not statistically signicant (
p
> 0.05) due to the
low expression levels in these clones. BIRC5 tended to amplify late in RT-qPCR with Utah-inoculated
clones. Signicant differential gene expression was also lacking for AXL, FN1, FSTL1, and SDC4 in all
pairwise comparisons. No signicant differential expression existed for clones 439Late versus 439Early
in any of the examined genes.
As also described for the Mock-inoculated clones, signicant correlations were not observed between
mean normalized gene expression and prion permissibility (
p
> 0.05).
3.4 Transcript levels of Mock and Utah-inoculated clones combined
The data for the Mock and Utah-inoculated clones were combined to increase the sample sizes. For
FSTL1, SPARC, and SDC4, the effect of inoculation was signicant (
p
< 0.05), and the interaction of
inoculation status and clone were also signicant for SDC4 and SPARC (
p
< 0.05), so these genes were
not included in this aspect of the study. However, data could be combined for AXL, FN1, BIRC5, and PRNP.
For BIRC5 (
p
< 0.001) and PRNP (
p
< 0.01), the effect of clone remained statistically signicant with two-
Page 6/25
way ANOVA, frequently involving the same original clones as described in sections 3.2 and 3.3, but with
some modications to be addressed. When data was combined for BIRC5, the signicant variation in
gene expression of clones 439Early and 439Late, compared to clone 440 were no longer signicant (
p
>
0.05). In PRNP, although clone 439Early lacked a two-fold change in expression to other clones, collective
data indicated that 439Early is weakly but signicantly (
p
< 0.05) increased relative to 438. No signicant
correlations were identied between genes (
p
> 0.05) or with prion permissibility (
p
> 0.05).
Since inoculation status was inuential to gene expression for FSTL1, SPARC, and SDC4 with two-way
ANOVA, the ratio of normalized mean expression derived from both treatment groups (i.e., Mock-
inoculated/Utah-inoculated) was constructed to evaluate if inoculation with scrapie prions inuenced
transcript levels of these three genes (Figure 3).
For SPARC, transcript levels were approximately three-fold higher for clones 438 and 434 (
p
< 0.05) as
compared to clone 439Late subsequent to inoculation with scrapie prions.
3.5 Gene expression values relative to PRNP in Mock-inoculated clones
Although normalized PRNP expression was determined to lack two-fold changes in expression and
displayed absent to minimal, signicant variation amongst the clones in this, as well as another previous
study [28], the inuence of target genes relative to PRNP expression of other genes was evaluated to
determine if target gene/PRNP ratios were signicant relative to prion permissibility. Raw target
gene/PRNP ratios were calculated using the relative quantity of expression (i.e., non-normalized) for
target genes and PRNP, per clone. Fold changes in expression and correlations with prion permissibility
were statistically evaluated as previously described in section 3.2. Statistically signicant differences in
expression were observed for BIRC5/PRNP (
p
< 0.001), SPARC/PRNP (
p
< 0.001), FSTL1/PRNP (
p
< 0.01)
and SDC4/PRNP (
p
< 0.01) (Figure 4). In BIRC5/PRNP, differential gene expression of 439Late and
439Early were signicantly decreased compared to clones 438 (
p
< 0.001 and
p
< 0.01), and 441 (
p
< 0.01
and
p
< 0.05), and this decreased expression was a ve-to-eight-fold change for 439Late and 439Early
compared to clone 438. Earlier in the study, signicant fold changes with BIRC5 in 439Early and 439Late
were also described for both treatment groups, independent of PRNP expression. Moreover, for
BIRC5/PRNP, clone 439Late had signicantly diminished expression compared to clones 434 (
p
< 0.01)
and 440 (
p
< 0.05). SPARC/PRNP had signicant decreased expression for clones 439Late and 439Early
compared to clones 438 (
p
< 0.001 and
p
< 0.01) and 434 (
p
< 0.001 and
p
< 0.05), and a two-to-three-fold
change in expression existed in clones 439Late and 439Early compared to clone 438. Also, for
SPARC/PRNP, clones 440 and 441 had signicantly diminished expression compared to 438 (
p
< 0.01
and
p
< 0.05), and clone 440 was signicantly decreased from 434 (
p
< 0.05), but these changes were
slightly less than two-fold relative to clone 438. In FSTL1, clone 439Late had signicantly diminished
expression from clones 438 (
p
< 0.05) and 434 (
p
< 0.05), but these differences were less than two-fold.
For SDC4/PRNP expression, 439Late was signicantly decreased compared to clones 438 (
p
< 0.01), 440
(
p
< 0.05), and 434 (
p
< 0.05), but these differences were also less than two-fold relative to clone 438.
Relative to PRNP expression, signicant differential gene expression remained absent for AXL and FN1.
Page 7/25
Although an approximately three-fold change in BIRC5/PRNP expression was present in clone 439Late
compared to 439Early, this difference was not signicant (
p
> 0.05). No signicant differences in
expression of clones 439Early versus 439Late were present for any of the remaining gene ratios.
No signicant correlations to prion permissibility were identied amongst all gene ratios.
3.6 Gene expression values relative to PRNP in Utah-inoculated clones
Fold change data and correlations to prion permissibility were constructed and statistically assessed as
previously described in section 3.5. Statistically signicant differences in expression were observed for
AXL/PRNP (
p
< 0.05) (Figure 5). For AXL/PRNP, clone 441 had signicantly diminished expression
compared to clone 438 (
p
< 0.05), but this change was less than two-fold. Clones of BIRC5/PRNP lacked
signicant differential expression, but an
approximately three-to-six-fold change in expression was noted in clones 439Late and 439Early relative
to clone 438. In FN1/PRNP, there was also an absence of signicant differential expression amongst the
clones, but two-fold changes in expression were present in clones 439Late, 439Early, 434, and 441,
relative to clone 438, and mean expression was highest in the least permissive clone (i.e., 438). In
FSTL/PRNP, no signicant differential expression was observed amongst the clones, but a two-fold
change was present in clone 439Late relative to clone 438. Signicant differential gene expression was
lacking for SDC4/PRNP and SPARC/PRNP.
For BIRC5/PRNP, clone 439Late had a two-fold higher transcript level relative to 439Early, but this change
lacked statistical signicance. No additional signicant differences in expression were present in clone
439Early versus 439Late for any of the remaining gene ratios.
No signicant correlations with prion permissibility were detected amongst the gene ratios.
3.7 Gene expression values relative to PRNP in Mock and Utah-inoculated clones
Target gene/PRNP expression ratios were pooled from Mock and Utah-inoculated clones for AXL/PRNP,
BIRC5/PRNP and FN1/PRNP, and the impact of clone and inoculation status with expression, as well as
correlations to prion permissibility were evaluated using similar procedures as described in section 3.4
(Figure 6). Neither inoculation status, nor the
interaction of clone with inoculation status were inuential to expression of genes selected for this
portion of the study, so treatment groups were pooled to increase sample sizes. For AXL/PRNP, the effect
of clone was signicant (
p
< 0.01) for clones 438 and 441. In BIRC5/PRNP, although the effect of clone
remained signicant to gene expression (
p
< 0.001) with many of the same clone comparisons depicted
in section 3.5, further signicant differential comparisons included clones 439Early and 434 (
p
< 0.01),
clones 441 and 439Late (
p
< 0.01), and clones 440 and 438 (
p
< 0.05). Furthermore, BIRC5/PRNP
differential expression became insignicant between clones 440 and 439Late (
p
> 0.05). Of interest, when
FN1/PRNP was assessed jointly for both treatment groups, the effect of clone became signicant (
p
<
Page 8/25
0.01), and expression of clone 438 was signicantly increased compared to 439Late (
p
< 0.05), 441 (
p
<
0.01) and 439Early (
p
< 0.05), with clones 441 and 439Early previously characterized as highly
permissive to scrapie prion infection.
Additionally, BIRC5/PRNP (r = -0.7187, adjusted
p
-value < 0.05), and FN1/PRNP (r = -0.7841, adjusted
p
-
value < 0.01) were strongly and negatively correlated with prion permissibility (Table 1).
Table 1. Significant correlations between target genes/PRNP and prion permissibility.
a Data was derived from at least 3 independent replicates. A
p
-value < 0.05 was considered significant. Ratio
expression was scaled to 438.
b Adjusted
p
-value were derived from the formula (Pearson’s
p
-value x (total number of
p
-values/
p
-value rank)).
The value obtained from this calculation was compared to the previous adjusted
p
-value, and the smaller of the
two was recorded. This value is also referred to as the Benjamini-Hochburg
p
-value
(www.biostathandbook.com/multiplecomparisons.html).
Discussion
Identifying denitive determinants of prion permissibility is key in unraveling prion disease pathogenesis,
and it also acts as a catalyst in the development of preventative measures and effective treatment
protocols. Multiple studies of classical scrapie have implemented the use of cell culture models, and
from these studies, potential determinants of prion permissibility have been identied [28, 32, 33]. In this
study, 6 genes (AXL, BIRC5, FN1, FSTL1, SPARC, SDC4) that previously demonstrated either differential
gene expression or the potential to affect prion permissibility between two clones (i.e., 438 and 439) of
contrasting prion permissibility [33], were further characterized. The previous results were expanded into a
set of six sheep microglia clones of varying prion permissibility, all derived from the same subline as
formerly stated. Raw gene expression levels were evaluated, as were gene expression values relative to
PRNP (i.e., target gene/PRNP).
Overall, there were no signicant differences in gene expression observed between clones 439Late and
439Early. Signicant correlations with prion permissibility were identied in expression ratios only (i.e.,
target gene/PRNP), and no signicant correlations were detected between genes. Clone was the most
common, inuential factor to gene expression.
As previously stated, clone 439 was originally determined to have high permissibility to infection with
naturally-derived classical scrapie isolates [29], but over time and subsequent subpassage, a shift to a
more intermediate permissibility phenotype manifested [28]. Thus, this provided an opportunity to
Page 9/25
temporally compare a clone to itself. Clone 439Late was consistently decreased by a two-to-three-fold
change from 439Early with BIRC5 raw expression and BIRC5/PRNP ratios, but these changes were not
statistically signicant. Unlike clone 439Early, clone 439Late was signicantly decreased in FSTL1 raw
expression, FSTL1/PRNP and SDC4/PRNP expression ratios, but these variations were also less than
two-fold. Similar to a previous study including 439Late and 439Early [28], these limited transcriptional
analysis studies failed to nd signicant difference. A transcriptome-wide approach could be used in
future studies to determine the differences between the 439Late and 439Early that result in the change of
permissibility phenotype, as well as the rest of the clones, which have not been evaluated on a
transcriptomic level (i.e., 434, 440, 441).
FN1/PRNP expression was strongly and negatively correlated with prion permissibility in ovine microglia.
Furthermore, with pooled, Mock and Utah-inoculated FN1/PRNP expression, the effect of clone became
signicant with decreased expression in clones 439Late, 439Early, and 441 compared to clone 438.
These ndings suggest that the expression prole of FN1, relative to PRNP may contribute to prion
permissibility. Marbiah
et al.
observed that FN1 silencing enhanced the levels of PrPD accumulation in
prion-susceptible murine neuroblastoma cells (i.e., R7 cells) [32]. The inuence of clone parallels ndings
reported by Munoz-Gutierrez
et al.
in which FN1 expression was downregulated in highly-permissive clone
439 compared to poorly-permissive clone 438, both infected with scrapie prion [33]. Collectively, these
ndings suggest that transcript levels of FN1 inuence prion permissibility and susceptibility, and that
this applies across different prion isolates and types of cell culture model systems. While the mechanism
of bronectin’s inuence on permissibility is unknown, possible causes include prion protein binding or
extracellular matrix remodeling. Growth factors have been reported to bind to FN1, becoming concealed
and able to evade proteolytic degradation, with subsequent accumulation in the FN1 brillar matrix [34].
Potentially, prion protein could have a similar interaction with FN1. Additional studies of FN1 are
warranted to determine if similar correlations with prion permissibility exist in other cell culture systems
(i.e., Rov or cpRK13[35]) and prion strains (i.e., natural scrapie isolates X124 or SSBP-1, Nor98, human
prion isolates). Furthermore, future studies to silence FN1 or deletion of FN1 from culture media in order
to create cell culture models for human prions may be warranted. With these subsequent studies, further
knowledge may be provided to strengthen the classication of FN1 as a biomarker of prion permissibility.
BIRC5 had consistently low expression in clones 439Late and 439Early. This is consistent with previous
results [33]. However, BIRC5 expression was very low (nearly undetectable). While BIRC5/PRNP
expression was strongly and negatively correlated with prion permissibility, this correlation was likely
skewed by the results of the related 439Early and 439Late clones. Thus, the signicance of these
comparisons, involving such a poorly expressed gene is questionable. Since BIRC5 has a role in
apoptosis (i.e., it directly inhibits activation of caspase-9, and formation of the apoptosome, suppressing
the intrinsic signaling pathway of apoptosis) [36], future studies may be best targeted toward apoptosis,
rather than BIRC5 specically.
SPARC and SPARC/PRNP expression were signicantly different in multiple pairwise comparisons, but
lacked a correlation with prion permissibility. While likely not important to prion permissibility amongst
Page 10/25
the clones, it is possible that the permissibility of different clones have different genetic determinants.
With specialized functions in cellular adhesion, proliferation, migration [37, 38], and matrix
metalloproteinase activity [38], it is possible that decreased expression of SPARC may result in
pathological abnormalities in these functions that enhance prion infectivity. Booth
et al.
identied
differential expression of SPARC between non-infected and scrapie-infected brain tissue from C57BL/6
mice experimentally inoculated (i.e., intracerebral) with one of two mouse-adapted prion strains (i.e., ME7,
79a) [39]. Interestingly, SPARC was also signicantly affected by the inoculation status of the microglia
clones. These ndings indicate that for some genes, inoculation inuences transcript levels, which may
potentially contribute to prion permissibility, in contrast to other genes studied in ovine microglia [28, 33].
It is also possible that the pre-inoculation state of the cell is critical for initial prion replication. To the
authors’ knowledge, this report depicts the rst characterization of SPARC in prion permissive ovine
microglia clones.
PRNP expression was signicantly increased in inoculated clone 439Late compared to 438, and
collectively in Mock and Utah-inoculated clones 439Late and 439Early compared to clone 438, but these
variations in expression were less than two-fold. Previous studies using these clones also identied some
variation in expression that was of a similar magnitude in one study [28], but in the remaining study, the
results were not statistically signicant [33]. Additionally, in this study, many signicant differential
comparisons were present in target gene/PRNP expression ratios (i.e., FN1/PRNP, BIRC5/PRNP,
FSTL1/PRNP). These ndings suggest that transcript level ratios of certain target genes and PRNP may
also inuence prion permissibility. Future gene expression studies of prion diseases are suggested to
evaluate gene transcript levels relative to PRNP, to account for the potential of PRNP levels interacting
with the target gene levels.
FSTL1, AXL and SDC4 results demonstrated sporadic pairwise comparisons that were signicant, but
these genes lacked patterns or correlations. Based on these results, it is unlikely that FSTL1, AXL, or SDC4
affect prion permissibility.
Some limitations were present in this study. First, only 2 out of 3 replicates of Utah-inoculated 439Early
could be assessed for BIRC5, which hindered full assessment of BIRC5 for this clone; however, this was
due to the very low transcript levels for BIRC5, with one replicate being so low as to be undetectable.
Second, only a single scrapie isolate was used (i.e., Utah), and assessed for differential gene expression
amongst the clones. Currently, only clone 439Early from the subline has been assessed for
permissiveness to other natural scrapie isolates (i.e., Pullman 1, Pullman 3, X124, 13-7) [29]. If the
remaining clones are also discovered to be permissive to these isolates, evaluation and comparison of
the transcriptomic proles between clones and scrapie isolates would be benecial. This would allow the
identication of similar or disparate genes that could be differentially expressed, providing more insight
on prion disease pathogenesis in ovine microglia. Thirdly, assessment of gene expression and
correlations with prion permissibility were only performed for ovine microglia of VRQ/VRQ genotype,
which is characterized by a rapid incubation period and fast accumulation of PrPSc. Extrapolation of the
results from this study towards the development of an ARQ/ARQ sheep prion culture system could aid in
Page 11/25
the development of the rst cell culture system for slowly accumulating PrPSc, which is more
characteristic of most human prion diseases.
Conclusions
Currently, denitive determinants of prion permissibility are limited, but one extracellular matrix gene
identied in this study (FN1), shows considerable promise in becoming an acknowledged determinant of
prion permissibility. The studies previously performed by Marbiah
et al.
, Munoz-Gutierrez
et al.
, and this
current study suggest that alterations in FN1 expression levels consistently correlate with prion
permissibility and susceptibility across two distinct cell culture model systems. For this study, FN1/PRNP
expression was strongly and negatively correlated with prion permissibility. These ndings suggest prion
permissibility is not dependent on individualized gene expression in ovine microglia cells, but other
factors may be involved to include cell of origin, type of scrapie isolate or strain, post-transcriptional or
post-translational factors, and possibly additive effects of gene expression (e.g., target gene levels
relative to PRNP levels).
Future studies are needed to fully characterize the functionality of these genes and their role in prion
permissibility in different cell culture model systems, with various prion isolates or strains, and
transcriptomic analysis.
Methods
2.1 Cell culture line
In previous studies, ovine microglia were isolated from the fetal brain of a pregnant, near-term, Suffolk-
cross ewe [40], and immortalized (hTERT) with construction of sublines, and cloning procedures
implemented for a single subline (subline H) in our lab [29], which was approved by the Institutional
Animal Care and Use Committee of Washington State University, Pullman (ASAF04575). Six ovine
microglia clones from subline H were further evaluated in the experimental study (438, 440, 434, 439Late,
441, and 439Early), arranged in order of diminishing permissibility to scrapie prion infection [28]. Clone
438 was least permissive; clones 440, 434 and 439Late of intermediate permissiveness, and clones 441
and 439Early as highly permissive [28, 29, 33]. 439Late and 439Early refer to a single clone whose
permissibility changed after continual passage [41], and are treated as two separate clones in the study.
These clones were previously inoculated with either a natural, sheep-derived scrapie isolate (Utah)
(derived from Animal Disease Research Unit; USDA Agricultural Research Service, Pullman, WA from a
natural infected sheep in Utah, USA), or uninfected ovine brain homogenate (Mock) [28, 29]. Cells were
cultured under standard conditions of Opti-MEM media (Gibco), supplemented with 10% complement-
inactivated fetal bovine serum (Atlanta Biologicals), 10,000 units/mL penicillin (Hyclone), 10,000 µg/mL
streptomycin (Hyclone), and 200 mM L-Glutamine (Hyclone). Cells were passaged every 3 to 4 days at a
1/5 dilution.
Page 12/25
2.2 Extraction, collection and analysis of RNA
Cell suspensions for use with ve experiments were previously collected at passage 5 (P-5) for Mock-
inoculated clones, and for Utah-inoculated clones at passages: P-2 (experiment 2.3Redo), P-4 (experiment
2.2), P-5 (experiment 1), P-6 (experiment 2.3), and P-7 (experiment 2.7) [28]. Over the course of all
experiments, a total of at least 3 independent replicates of each clone, per inoculation status, were tallied.
Cell suspensions were homogenized with QIAshredders (Qiagen) according to manufacturer’s
instructions. RNA was extracted from lysates using RNeasy Mini Kit (Qiagen), per manufacturer’s
directions. Eluted RNA was collected into DNA LoBind microcentrifuge tubes (Eppendorf) and stored at
-80 °C until further use. When ready to use, RNA purity and concentration was measured using NanoDrop
2000 spectrophotometer (Thermo Scientic).
2.3 DNase-treatment of RNA
RNA was standardized at 10 µg/50 µL, and residual DNA was removed using Invitrogen DNA-free DNase
Treatment and Removal kit (Fisher Scientic). A total volume of 50 µL was prepared per sample using 5
µL of 10x DNase I buffer, 1 µL of rDNase I, eluted RNA, and nuclease-free water. Each sample was
incubated at 37 °C for 30 minutes before adding 5 µL of DNase Inactivation reagent. After addition of the
inactivation reagent, the sample was mixed frequently and incubated at room temperature (~ 25 °C) for 2
minutes. Subsequent to centrifugation of the sample at 10,000 x g for 1.5 minutes, RNA was collected
and transferred to a new, 0.5 mL DNA LoBind microcentrifuge tube (Eppendorf). RNA was either
immediately stored at -80 °C, or purity and concentration were measured using NanoDrop 2000
spectrophotometer (Thermo Scientic), before storage for later use.
2.4 First strand cDNA synthesis
DNase-treated RNA was standardized to 1 µg/20 µL, and rst strand cDNA synthesis was performed
using SuperScript III First-Strand Synthesis Supermix for qRT-PCR (ThermoFisher Scientic). Each sample
consisted of 10 µL of 2x RT reaction mix, 2 µL of RT enzyme mix, 1 µg of DNase- treated RNA, and DEPC-
treated water for a total volume of 20 µL. Samples were gently mixed, then incubated at 25 °C for 10
minutes, 50 °C for 30 minutes, and 85 °C for 5 minutes. After incubation, samples were chilled on ice for 5
minutes, and then 1 µL of
E. coli
RNase H was added before an additional incubation at 37° C for 20
minutes. After the last incubation, samples were chilled on ice for 5 minutes, and immediately stored at
-80 °C.
2.5 Primers
Genes identied from previous RNAseq data in two clones of varying prion permissibility [33] were further
assessed (Table 2).
Table 2. Primers for qRT-PCR and RT-PCR.
Page 13/25
a Primers were designed using either presumed or curated mRNA from selected genes in the database program
of NIH, US National Library of Medicine, National Center for Biotechnology Information (NCBI) [42].
b PRNP primer sequences were derived from a previous publication [43].
c FN1 primer sequences were designed using PriFi [44].
These annotated genes include bronectin 1 (
FN1
), follistatin-like 1 (
FSTL-1
), osteonectin (
SPARC
),
survivin (
BIRC5
), syndecan 4 (
SDC4
), AXL receptor tyrosine kinase (
AXL
), and prion protein (
PRNP
). 18s
ribosomal RNA (
18s rRNA
) and hypoxanthine phosphoribosyltransferase 1 (
hPRT1
) are both reference
genes that have shown stable expression as internal controls in quantitative PCR analyses [45-47], and
were hence utilized as reference genes in this study. With the exception of PRNP [43] and FN1 [44], which
were previously published, primers were designed for each gene according to a reference sequence listed
for the gene using the database program from the National Center of Biotechnology Information [42].
Standard curves for each primer set were performed to determine eciency, conrm predictability across
dilutions (r2 > 0.85), and establish a single, repeatable product melting temperature. The resulting
products were also evaluated by standard agarose gel electrophoresis to conrm the product size.
2.6 Sequencing of PCR products
All PCR products were cloned and sequenced to conrm amplication of the correct gene target.
Superscript IV rst strand cDNA synthesis system kit for RT-PCR (ThermoFisher Scientic) was used,
according to manufacturer’s instructions. Each reaction contained 50 ng of RNA, derived from naïve ovine
microglia clone 439. First strand cDNA samples were then immediately used or stored at -20 °C.
Conventional PCR was implemented using Advantage II PCR Enzyme System (Clontech) according to
manufacturer’s directions. Each reaction was composed of 40 μL of nuclease-free water, 5 μL of 10X PCR
buffer, 1 μL of 50x dNTPs, 1 μL forward primer, 1 μL reverse primer, either 1 μL of Taq polymerase or
nuclease free water, and 1 μL of cDNA. The PCR protocol was as follows: 95 °C for 5 minutes; 30 cycles
of 95 °C for 30 seconds, 55 °C for 30 seconds, and 72°C for 30 seconds; and 72 °C for 7 minutes. Gel
Page 14/25
electrophoresis was used to assess the size and solitary nature of the products. Bands not exposed to
ultraviolet imaging were cut from the gel, and the DNA was puried using QIAEX II Agarose Gel Extraction
Kit (QIAGEN) according to manufacturer’s instructions. Eluted, puried DNA was ligated into a pGEM-T
easy vector system (Promega) per manufacturer’s directions. Two microliters of each ligation reaction
were transformed into 50 µL of JM109 high eciency competent cells and plated, according to
manufacturer’s instructions with few modications. Cells were heat shocked at 43°- 45 °C, transformation
cultures were incubated for 2 hours at 37 °C with a shaking speed of 420 rpm, and approximately 200 µL
of transformation cultures were plated onto LB/Ampicillin plates. After an overnight incubation at 37 °C,
up to 9 clones were selected per plate, and evaluated for the presence of the correct gene product using
conventional PCR and gel electrophoresis. Selected clones were isolated on an additional LB/Ampicillin
agar plate, and placed in a 37 °C incubator for 8 – 12 hours. For each gene, a single clone of the correct
size was further isolated in 3 µL of ampicillin, 3 mL of sterile LB medium, and a pipette tip containing the
colony, in an overnight 37 °C incubation with a shaking speed of 420rpm. After incubation, DNA was
eluted (QIAprep Spin Miniprep Kit, QIAGEN), per manufacturer’s directions, and measured for
concentration and purity using NanoDrop 2000 spectrophotometer (Thermo Scientic). Eluted DNA was
submitted for commercial Sanger sequencing using GENEWIZ (www.genewiz.com) or MCLAB
(www.mclab.com). Forward and reverse sequences were aligned using software programs Mega 6
(Molecular Evolutionary Genetics Analysis; www.megasoftware.net) and Geneious Prime
(www.geneious.com), and the resulting consensus sequences were aligned to the National Center of
Biotechnology Information (www.ncbi.nlm.nih.gov) nt database, using megablast.
2.7 qPCR
Each reaction contained 10 μL SsoAdvanced Universal SYBR Green Supermix, 0.4 μL forward primer, 0.4
μL reverse primer, and 7.2 μL nuclease-free water. Either 2.5 μL of appropriately diluted cDNA or nuclease-
free water for negative control was added to each reaction, and a single non-reverse transcriptase (NRT)
control was run for each clone and gene. Reactions were then run with CFX96 Touch Real-Time PCR
Detection System (Bio-Rad) with the following protocol for all primers: 95 °C for 30 minutes; 35 times of
95 °C for 15 seconds then 60 °C for 30 seconds; followed by 95 °C for 10 minutes, 60 °C for 5 minutes,
and an increase in temperature in 0.5 °C increments from 60 °C to 95 °C.
2.8 Data Analysis and Statistical computation
Gene expression data normalized to reference genes 18s rRNA and hPRT1 was retrieved and evaluated
from at least 3 independent replicates using the software program CFX Manager 3.1 (www.bio-rad.com:
Bio-Rad, Hercules, CA). Statistical analyses were performed using JMP version 14 (www.jmp.com; SAS
Institute, Cary, NC.). However, for Utah-inoculated clone 439Early, only 2 of the 3 independent replicates
had the acquisition of data for BIRC5 (i.e., undetectable levels of transcript in one replicate). Relative
prion permissibility was previously determined [28]. Normality of the data distribution was assessed per
inoculation status, using histograms and Goodness-of-t test. For data that was not normally distributed,
Box-Cox Y transformation was used for approximation to normality, and this newly congured data was
Page 15/25
assessed for normality using histograms and Goodness of Fit test. Fold changes were computed
manually using raw normalized expression (Mode in CFX Manager: Normalized expression (ΔΔ Cq))
collectively, and then relative to the previously identied least permissive clone, 438 [28], for each gene by
treatment group. Statistical signicance of fold change (i.e., clone) was assessed using One-way ANOVA
and
post hoc
Tukey-Kramer HSD for multiple pairwise comparisons (
p
< 0.05). Correlations of mean
normalized gene expression between genes, and with prion permissibility per treatment group, were
calculated using multivariate analysis with Pearson’s correlation coecient (Pearson’s r), and Benjamini-
Hochberg multiple comparison correction. The effect of inoculation status, and the interaction of clone
and inoculation status on gene expression were analyzed using pooled, raw expression data by gene in
Two-way ANOVA and
post-hoc
Tukey HSD for multiple pairwise comparisons (
p
< 0.05). For genes in
which inoculation status was not inuential to expression, mean normalized gene expression data were
pooled by gene and clone, and correlations with prion permissibility were calculated using Pearson’s r and
Benjamini-Hochberg multiple comparison correction. qRT-PCR analyses and statistical calculations as
previously described were also performed for relative density expression (Mode in CFX Manager: Relative
quantity (ΔCq)) of target genes relative to PRNP (target gene/PRNP ratios) using at least 3 independent
replicates.
Abbreviations
Scrapie prion protein (PrPSc), transmissible spongiform encephalopathies (TSEs), cellular prion protein
(PrPC), disease-associated prion protein (PrPD), prion protein (PrP), bronectin 1 (FN1), follistatin-like 1
(FSTL1), osteonectin (SPARC), survivin (BIRC5), syndecan 4 (SDC4), AXL receptor tyrosine kinase (AXL),
prion protein gene (PRNP), valine-arginine-glutamine (VRQ), alanine-arginine-glutamine (ARQ), alanine-
arginine-arginine (ARR), ovinized rabbit kidney epithelial cells (Rov), murine neuroblastoma cells
inoculated with mouse-adapted scrapie isolate (ScN2a), 18s rRNA (18s ribosomal RNA), hypoxanthine
phosphoribosyltransferase 1 (hPRT1), E (Early), L (Late), human telomerase (hTERT), caprinized rabbit
kidney epithelial cells (cpRK13), reverse transcriptase quantitative polymerase chain reaction (RT-qPCR)
Declarations
a) Ethics approval and consent to participate
A previously established immortalized (hTERT) ovine microglia cell subline (Subline H) was utilized [29,
40], which was approved by the Institutional Animal Care and Use Committee of Washington State
University (ASAF04575).
b) Consent for publication
Not applicable
c) Availability of data and material
Page 16/25
All data generated or analysed during this study are included in this published article.
d) Competing interests
The authors declare that they have no competing interests.
e) Funding
We would like to thank the following funding sources: University of Georgia Oce of the Vice President
for Research [Veterinary Medical Experimental Station Research Competitive Grant], and USDA Animal
Health Formula Funds (1007561). These sources contributed to acquisition of research materials,
performance of the study, and writing of this manuscript.
f) Author’s contributions
JBS provided guidance on experimental design, collection and analysis of data. KD inoculated and
maintained cell culture clones used in this study. ZN performed validation and eciency testing of some
primers. VRM performed the experiments, collected and analyzed the data. All authors have read and
approved this manuscript.
g) Acknowledgements
The authors would like to thank Jian Zhang, Alma Pena-Briseno, Salman Butt, and Kelsey Young for their
time, efforts and technical assistance with this study.
References
1. Grassmann, A., Wolf, H., Hofmann, J., Graham, J., Vorberg,I.,
Cellular Aspects of Prion Replication In
Vitro.
Viruses, 2013. 5: p. 374-405.
2. Imran, M., Mahmood, S.,
An overview of animal prion diseases.
Virology Journal 2011. 8: p. 493-500.
3. Babelhadj, B., DiBari, M.A., Pirisinu, L., Chiappini, B., Gaouar, S.B.S., Riccardi, G., Marcon, S., Agrimi, U.,
Nonno, R., Vaccari, G.,
Prion disease in Dromedary Camels, Algeria.
Emerging Infectious Diseases,
2018. 24(6): p. 1029 - 1035.
4. Schneider, K., Fangerau, H., Michaelson, B., Raab, W.H.M.,
The early history of the transmissible
spongiform encephalopathies exemplied by scrapie.
Brain Research Bullentin, 2008. 77(6): p. 343-
355.
5. Prusiner, S.B.,
Novel Proteinaceous Infectious Particles Cause Scrapie.
Science, 1982. 216(4542): p.
136-144.
. Grith, J.S.,
Self replication and scrapie.
Nature, 1967. 215(5105): p. 1043-1044.
7. Caughey, B.W., Dong A., Bhat, K.S., Ernst, D., Hayes, S.F. and Caughey, W.S.,
Secondary structure
analysis of the scrapie-associated protein PrP 27–30in water by infrared spectroscopy.
Biochemistry,
1991. 30: p. 7672 - 7680.
Page 17/25
. Pan, K.M., Baldwin, M., Nguyen, J., Gasset, M., Serban, A., Groth. D., Mehlhorm, I., Huang, Z., Fletterick,
R.J., Cohen, F.E.,
Conversion of alpha-helices into beta-sheets features in the formation of the scrapie
prion proteins.
Proc Natl Acad Sci USA, 1993. 90(23): p. 10962 - 10966.
9. Zhou, Z., Xiao, G.,
Conformational conversion of prion protein in prion diseases.
Acta Biochim
Biophys Sin, 2013. 45(6): p. 465 - 476.
10. Legname, G., Baskakov, I.V., Nguyen, H.O., Riesner, D., Cohen, F.E., DeArmond, S.J., Prusiner, S.B.,
Synthetic mammalia prions.
Science, 2004. 305(5684): p. 673 - 676.
11. Castilla, J., Saa, P., Hetz, C., Soto, C. ,
In vitro generation of infectious scrapie prions.
Cell, 2005.
121(2): p. 195 - 206.
12. Telling, G.C.,
Transgenic mouse models of prion diseases.
Methods in Molecular Biology, 2008. 459:
p. 249 - 263.
13. Thackray, A.M., Hopkins, L., Spiropoulos, J., Bujdoso, R.,
Molecular and Transmission Characteristics
of Primary-Passaged Ovine Scrapie Isolates in Conventional and Ovine PrP Transgenic Mice.
Journal
of Virology, 2008. 82(22): p. 11197-11207.
14. Rheede, T.V., Smolenaars, M.M.W., Madsen, O., de Jong, W.W.,
Molecular Evolution of the Mammalian
Prion Protein.
Molecular Biology and Evolution, 2002. 20(1): p. 111-121.
15. Goldmann, W., Hunter, N., Foster, J.D., Salbaum, J. M., Beyreuther, K., Hope, J.,
Two alleles of a neural
protein gene linked to scrapie in sheep.
Proceedings of the National Academy of Sciences of the
United States of America, 1990. 87: p. 2476 - 2480.
1. Brown, D.R., Schmidt, B., Kretzschmar, H.A.,
Role of microglia and host prion protein in neurotoxicity
of a prion protein fragment.
Nature, 1996. 380: p. 345 - 347.
17. Moser, M., Colello, R.J., Pott, U., Oesch, B. ,
Developmental expression of the prion protein in glial
cells.
Neuron, 1995. 14: p. 509 - 517.
1. Linden, R., Martins, V.R., Prado, M. A., Cammarota, M., Izquierdo, I., Brentani, R.R.,
Physiology of the
prion protein.
Physiological Reviews, 2008. 88: p. 673 - 728.
19. Tranulis, M.A.,
Inuence of the prion protein gene, Prnp, on scrapie susceptibility in sheep.
APMIS,
2002. 110: p. 33 - 43.
20. Hunter, N., Foster, J.D., Goldmann, W., Stear, M.J., Hope, J., Bostock, C,
Natural scrapie in a closed
ock of Cheviot sheep occurs only in specic PrP genotypes.
Archives of Virology, 1996. 141(5): p.
809 - 824.
21. Clouscard, C., Beaudry, P., Elsen, J.M., Milan, D., Dussaucy, M., Bounneau, C., Schelcher, F., Chatelain,
J., Launay, J.M., Laplanche, J.L.,
Different allelic effects of the codons 136 and 171 of the prion
protein gene in sheep with natural scrapie.
Journal of General Virology, 1995. 76: p. 2097 - 2101.
22. Snow, A.D., Kisilevsky, R., Willmer, J., Prusiner, S.B, DeArmond, S.J.,
Sulfated glycosaminoglycans in
amyloid plaques of prion diseases.
Acta Neuropathologica, 1989. 77(4): p. 337 - 342.
23. Baron, G.S., Caughey, B.,
Effect of glycosylphosphatidylinositol anchor-dependent and -independent
prion protein association with model raft membranes on conversion to the protease-resistant
Page 18/25
isoform.
Journal of Biological Chemistry, 2003. 278(17): p. 14883 - 14892.
24. Arsac, J.N., Baron, T.,
Distinct Transmissibility Features of TSE Sources Derived from Ruminant Prion
Disease by the Oral Route in a Transgenic Mouse Model (TgOvPrP4) Overexpression the Ovine Prion
Protein.
PLOS One, 2014. 9(5): p. 1-8.
25. Beekes, M., McBride, P.A.,
Early accumulation of pathological PrP in the enteric nervous system and
gut-associated lymphoid tissue of hamsters orally infected with scrapie.
Neuroscience Letter, 2000.
278(3): p. 181- 184.
2. Hu, P.P., Morales, R., Duran-Aniotz, C., Moreno-Gonzalez, I., Khan, U., Soto, C.,
Role of Prion Replication
in the Strain-dependent BrainRegional Distribution of Prions.
THE JOURNAL OF BIOLOGICAL
CHEMISTRY, 2016. 291(24): p. 12880-12887.
27. Solassol, J., Crozet, C., Lehmann, S.,
Prion propagation in cultured cells.
British Medical Bulletin,
2003. 66: p. 87 - 97
2. Dinkel, K.D., Schneider, D.A., Muñoz-Gutiérrez, J.F., McElliott, V.R., Stanton, J.B.,
Correlation of cellular
factors and differential scrapie prion permissiveness in ovine microglia.
Virus Research, 2017. 240:
p. 69-80.
29. Muñoz-Gutiérrez, J.F., Schneider, D.A., Baszler, T.V., Greenlee, J.J., Nicholson, E.M., Stanton, J.B. ,
hTERT-immortalized ovine microglia propagate natural scrapie isolates.
Virus Research, 2015. 198: p.
35 - 43.
30. Kettenmann, H., Hanisch, U.K., Noda, M., Verkhratsky, A.,
Physiology of Microglia.
Physiological
Reviews, 2011. 91: p. 461-553.
31. Aguzzi, A., Zhu, C.,
Microglia in prion disease.
The Journal of Clinical Investigation, 2017. 127(9): p.
3230 - 3239.
32. Marbiah, M.M., Harvey, A., West, B.T., Louzolo, A., Banerjee, P., Alden, J., Grigoriadis, A., Hummerich, H.,
Kan, H.M., Cai, Y., Bloom, G.S., Jat, P., Collinge, J., Klohn, P.C. ,
Identication of a gene regulatory
network associated with prion replication.
The EMBO Journal, 2014. 33(14): p. 1527-1547.
33. Muñoz-Gutiérrez, J.F., Pierlé, S.A., Schneider, D.A., Baszler, T.V., Stanton, J.B.,
Transcriptomic
Determinants of Scrapie Prion Propagation in Cultured Ovine Microglia.
PLoS ONE, 2016. 11(1): p. 1 -
20.
34. Zollinger, A.J., Smith, M.L.,
Fibronectin, the extracellular gene.
Matrix Biology, 2016. 60 - 61: p. 27 - 37.
35. Dassanayake, R.P., Zhuang, D., Truscott, T.C., Madsen-Bouterse, S.A., O'Rourke, K.I, Schneider, D.A.,
A
transfectant RK13 cell line permissive to classical caprine scrapie prion propagation.
Prion, 2016.
10(2): p. 153 - 164.
3. Khan, Z., Khan, A.A., Yadav, H., Prasad, G.B.K.S., Bisen, P.S.,
Survivin, a molecular target for
therapeutic interventions in squamous cell carcinoma.
Cellular and Molecular Biology Letters, 2017.
22(8): p. 1 - 32.
37. Lloyd-Burton, S.M., York, E.M., Anwar, M.A., Vincent, A.J., Roskams, A.J.,
SPARC Regulates
Microgliosis and Functional Recovery following Cortical Ischemia.
The Journal of Neuroscience,
2013. 33(10): p. 4468 - 4481.
Page 19/25
3. Bradshaw, A.D.,
The role of SPARC in extracellular matrix assembly.
Journal of Cell Communication
and Signaling, 2009. 3: p. 239 - 246.
39. Booth, S., Bowman, C., Bamgartner, R., Sorensen, G., Robertson, C., Coulthart, M., Phillipson, C.,
Somorjai, R.L.,
Identication of central nervous system genes involved in the host response to the
scrapie agent during preclinical and clinical infection.
Journal of General Virology, 2004. 85: p. 3459 -
3471.
40. Stanton, J.B., Knowles, D.P., O'Rourke, K.I., Herrmann-Hoesing, L.M., Mathison, B.A., Baszler, T.V.,
Small
Ruminant Lentivirus Enhances PrPSc Accumulation in Cultured Sheep Microglial Cells.
Journal of
Virology, 2008. 82(20): p. 9839 - 9847.
41. Dinkel, K.D., Stanton, J.B., Boykin, D.W., Stephens, C.E., Madsen-Bouterse, S.A., Schneider, D.A., ,
Antiprion Activity of DB772 and Related Monothiophene- and FuranBased Analogs in a Persistently
Infected Ovine Microglia Culture System.
Antimicrobial Agents and Chemotherapy, 2016. 60(9): p.
5467 - 5482.
42. Ye, J., Coulouris, G., Zaretskaya, I., Cutcutache, I., Rozen, S., Madden, T.,
Primer-BLAST: A tool to
design target-specic primers for polymerase chain reaction.
BMC Bioinformatics, 2012. 13(134).
43. Stanton, J.B., Schneider, D.A., Dinkel, K.D., Balmer, B.F., Baszler, T.V., Mathison, B.A., Boykin, D.W.,
Kumar, A. ,
Discovery of a Novel, Monocationic, Small-Molecule Inhibitor of Scrapie Prion
Accumulation in Cultured Sheep Microglia and Rov Cells.
PLOS One, 2012. 7(11): p. 1 - 11.
44. Fredslund, J., Schauser, L., Madsen, L.H., Sandal, N., Stougaard, J.,
PriFi: using a multiple alignment
of related sequences to nd primers for amplication of homologs.
Nucleic Acids Research, 2005.
33: p. W516-W520.
45. Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy, N., De Paepe, A., Speleman, F.,
Accurate
normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal
control genes.
Genome Biology, 2002. 3(7): p. 1 - 12.
4. Kozera, B., Rapacz, M.,
Reference genes in real-time PCR.
Journal of Applied Genetics, 2013. 54: p.
391-406.
47. Kuchipudi, S.V., Tellabati, M., Nelli, R.K., White, G.A., Perez, B.B., Sebastian, S., Slomka, M.J., Brookes,
S.M., Brown, I.H., Dunham, S.P., Chang, K-C,
18S rRNA is a reliable normalisation gene for real time
PCR based on inuenza virus infected cells.
Virology Journal, 2012. 9(230): p. 1 - 7.
Figures
Page 20/25
Figure 1
Fold change in gene expression for Mock-inoculated clones. Fold changes were assessed using
normalized gene expression relative to the least permissive clone (438) for Mock-inoculated clones.
Expression data was normalized to 18s rRNA and hPRT1, scaled to 438, and then the geometric mean
and geometric standard deviation factor were calculated for graphical representation. Solid, black
horizontal lines (i.e., 21 or 2 -1) demarcate the two-fold change threshold. The x-axis is ordered by
increasing prion permissibility phenotype (left to right). The horizontal axes are partitioned into panels by
gene (AXL, BIRC5, FN1, FSTL1, PRNP, SDC4, SPARC). Error bars indicate 1 geometric standard deviation
factor from the geometric mean. In differentially expressed genes, clones that do not share the same
letter had signicant differential mean expression (p < 0.05). E = Early, L = Late
Page 21/25
Figure 2
Fold change in gene expression for Utah-inoculated clones. Fold changes were assessed using
normalized gene expression, relative to the least permissive clone (438) for Utah-inoculated clones.
Expression data was normalized to 18s rRNA and hPRT1, scaled to 438, and then the geometric mean
and geometric standard deviation factor were calculated for graphical representation. Solid, black,
horizontal lines (i.e., 21 or 2 -1) demarcate the two-fold change threshold. The x-axis is ordered by
increasing prion permissibility phenotype (left to right). The horizontal axes are partitioned into panels by
gene (AXL, BIRC5, FN1, FSTL1, PRNP, SDC4, SPARC). Error bars indicate 1 geometric standard deviation
factor from the geometric mean. In differentially expressed genes, clones that do not share the same
letter had signicant differential mean expression (p < 0.05). E = Early, L = Late
Page 22/25
Figure 3
Effect of inoculation on transcript levels of FSTL1, SDC4, SPARC. The effect of inoculation on transcript
levels was calculated by dividing the Mock-inoculated value by the Utah-inoculated value. Geometric
mean and geometric standard deviation were calculated for graphical representation. Solid, black
horizontal lines (i.e., 21 or 2 -1) demarcate the two-fold change threshold. The x-axis is ordered by
increasing prion permissibility phenotype (left to right). The horizontal axis is partitioned into panels by
gene (FSTL1, SDC4, SPARC). Error bars indicate 1 geometric standard deviation factor from the geometric
mean. In differentially-expressed genes, clones that do not share the same letter had signicant
differential mean expression (p < 0.05). E = Early, L = Late
Page 23/25
Figure 4
Fold change in target gene/PRNP expression in Mock-inoculated clones. Fold changes were assessed
using relative expression data, scaled to the least permissive clone (438) for Mock-inoculated clones.
Relative expression ratios were constructed (i.e., target gene/PRNP expression), and then the geometric
mean and geometric standard deviation were calculated for graphical representation. Solid, black
horizontal lines (i.e., 21 or 2 -1) demarcate the two-fold change threshold. The x-axis is ordered by
increasing prion permissibility phenotype (left to right). The horizontal axes are partitioned into panels by
gene (AXL, BIRC5, FN1, FSTL1, SDC4, SPARC). Error bars indicate 1 geometric standard deviation factor
from the geometric mean. In differentially-expressed genes, clones that do not share the same letter had
signicant differential mean expression (p < 0.05). E = Early, L = Late
Page 24/25
Figure 5
Fold change in expression of target gene/PRNP expression in Utah-inoculated clones. Fold changes were
assessed using relative expression data, scaled to the least permissive clone (438) for Utah-inoculated
clones. Relative expression ratios were constructed (i.e., target gene/PRNP expression), and then the
geometric mean and geometric standard deviation factor were calculated for graphical representation.
Solid, black, horizontal lines (i.e., 21 or 2 -1) demarcate the two-fold change threshold. The x-axis is
ordered by increasing prion permissibility phenotype (left to right). The horizontal axes are partitioned
into panels by gene (AXL, BIRC5, FN1, FSTL1, SDC4, SPARC). Error bars depict 1 geometric standard
deviation factor from the geometric mean. In differentially-expressed genes, clones that do not share the
same letter had signicant differential mean expression (p < 0.05). E = Early, L = Late.
Page 25/25
Figure 6
Comparison of Target gene/PRNP expression for Mock and Utah-inoculated clones. Expression ratios
were pooled from Mock and Utah-inoculated clones, relative to PRNP expression, for genes in which
inoculation was insignicant. Geometric mean and geometric standard deviation factor were calculated
per ratio for graphical representation. Solid, black, horizontal lines (i.e., 21 or 2 -1) demarcate the two-fold
change threshold. The x-axis is ordered by increasing prion permissibility phenotype (left to right). The
horizontal axis is partitioned into panels by gene (AXL, BIRC5, FN1). Error bars depict 1 geometric
standard deviation factor from the geometric mean. In differentially-expressed genes, clones that do not
share the same letter had signicant differential mean expression (p < 0.05). E = Early, L = Late.
