Mice Treated Subcutaneously with Mouse LPS-Converted PrPres or LPS Alone Showed Brain Gene Expression Profiles Characteristic of Prion Disease

Previously, we showed that bacterial lipopolysaccharide (LPS) converts mouse PrPC protein to a beta-rich isoform (moPrPres) resistant to proteinase K. In this study, we aimed to test if the LPS-converted PrPres is infectious and alters the expression of genes related to prion pathology in brains of terminally sick mice. Ninety female FVB/N mice at 5 weeks of age were randomly assigned to 6 groups treated subcutaneously (sc) for 6 weeks either with: (1) Saline (CTR); (2) LPS from Escherichia coli 0111:B4 (LPS), (3) one-time sc administration of de novo generated mouse recombinant prion protein (moPrP; 29-232) rich in beta-sheet by incubation with LPS (moPrPres), (4) LPS plus one-time sc injection of moPrPres, (5) one-time sc injection of brain homogenate from Rocky Mountain Lab (RLM) scrapie strain, and (6) LPS plus one-time sc injection of RML. Results showed that all treatments altered the expression of various genes related to prion disease and neuroinflammation starting at 11 weeks post-infection and more profoundly at the terminal stage. In conclusion, sc administration of de novo generated moPrPres, LPS, and a combination of moPrPres with LPS were able to alter the expression of multiple genes typical of prion pathology and inflammation.


Introduction
Transmissible spongiform encephalopathies (TSEs), or prion diseases are a group of fatal neurodegenerative disorders that are characterized by misfolding of cellular prion protein (PrP C ) into a pathogenic form, the scrapie prion protein (PrP Sc ). The scrapie or the misfolded form of the protein is commonly found in the brain tissue of affected animals and humans. This led to the development of the prion-only hypothesis [1]. Then, the main challenge for the scientific community was to prove that indeed the misfolded protein (PrP Sc ) was the real causative agent of prion disease. Despite, controversies regarding the potential role of a misfolded protein in the etiology of prion diseases, several reports during the last decade proved that brain homogenates coming from sick animals, or an in vitro generated misfolded protein might indeed cause typical prion disease in experimental animal models [2][3][4][5][6]. There is one drawback of all the reported investigations: they have used un-natural conditions to generate the infectious prions like serial rounds of protein misfolding cyclic amplification (PMCA) and long incubations periods, presence of RNA or other synthetic forms like polyriboadenylic acid, poly-anions like glucosaminoglycans (GAGs), denaturing acidic conditions, and synthetic lipids like 1-palmitoyl-2-oleoylphosphatidylglycerol (POPG) or phosphatidylethanolamine [3,[6][7][8].
The main issue with all these successful in vitro investigations is that the odds of these After one week of adaptation time, FVB/N mice at 5 weeks of age were randomly assigned to 6 treatment groups: (1) six weeks sc administration of saline at 11 µL/h, (2) six weeks sc injection of lipopolysaccharide from Escherichia coli 0111:B4 LPS at 0.1 µg/g BW at 11 µL/h, (3) one-time sc injection of de novo generated mouse recombinant prion protein (moPrP; 29-232) rich in β-sheet by incubation with LPS (moPrP res ) at 45 µg/mouse, (4) six weeks sc injection of LPS at 0.1 µg/g BW at 11 µL/h plus one-time sc injection of moPrP res at 45 µg/mouse, (5) one-time sc injection of brain homogenate from Rocky Mountain Laboratory scrapie strain at 10 7 ID 50 units of scrapie prions, and (6) six weeks sc administration of LPS plus one-time sc injection of RML at 10 7 ID 50 units of scrapie prions. For the groups that involve saline or LPS sc administration, ALZET ® osmotic minipumps (ALZET, Cupertino, CA, USA) were used. The moPrP res and RML were subcutaneously injected at the start of the experiment, right after ALZET ® osmotic mini pumps implantation.
The ALZET ® osmotic pumps were implanted sc to mimic the continuous but minimal access of LPS to the body through the oral route. The sc infusion with ALZET ® osmotic pumps was done for 6 weeks continuously to eliminate the need for frequent animal handling and repetitive injection schedules. The sc infusion was performed according to the manufacturer's procedure. Briefly, mice were anesthetized by isoflurane (Baxter Corporation, Mississauga, ON, Canada) and once the mouse was not responsive to tail pinch, low and continuous isoflurane, and oxygen were supplied using the anesthetic machine (Matrx by Midmark corporation, Versailles, OH, USA). After shaving the hair and disinfection, a cut was made at the back of the mice using sterile surgical scissors. The ALZET ® osmotic pumps were inserted and the opening was closed with sutures (ALZET Osmotic Pumps, Cupertino, CA, USA). After 6 weeks, the same procedure was repeated to take the empty pumps out and close the skin opening with sutures.

Euthanasia at 11 Weeks Post-Inoculation and Terminal Stage
A total of 30 FVB/N female mice, 5 from each treatment group were euthanized at 11 weeks of age. Mice were euthanized by overdose inhalation of isoflurane. After taking out the brain tissue from the skull it was sagittally cut into two symmetrical halves using a sterile scalpel. One half was snap-frozen and stored at −86 • C until RNA purification. The euthanasia of terminal stage mice was conducted at different time points. During the entire experimental period, mice were checked on a daily basis for the onset and progression of prion disease. Mice were euthanized when the clinical signs were consistent and progressive. Terminal stage sickness was characterized by kyphosis, ataxia, dysmetria, tremor, head tilt, tail rigidity, bradykinesia, proprioceptive deficits, stupor, loss of deep pain sensation, and loss of weight. The euthanasia and brain samples processing procedures were the same as the described for euthanasia at 11 weeks.

Total RNA Isolation
Total RNA was isolated from brain tissue (n = 3) using SV Total RNA Isolation System (Promega Corporation, Madison, WI, USA) following the manufacturer's protocol. Briefly, brain samples were thawed at room temperature and 30 mg was taken for RNA isolation. Then samples were lysed in 175 µL RNA lysis buffer (RLA + BME) in sterile tubes. After adding 350 µL RNA dilution buffer, samples were heated at 70 • C for 3 min. Samples were then centrifuged for 10 min at 14,000× g (model 5430R, Eppendorf, Hamburg, Germany). The clear lysate was collected to a new collection 5 mL tube (Promega, Madison, WI, USA) and 200 µL (95%) ethanol (Commercial Alcohol, Winnipeg, MB, Canada) was added. After transferring the lysate to spin basket assembly (Promega, Madison, WI, USA) and centrifuged for 1 min, the lysate was washed with RNA wash solution. On-column DNA digestion was subsequently done using DNase I to remove any DNA contamination. After subsequent washing with washing buffers, total RNA was eluted with 100 µL nucleasefree water.

Custom-Made PCR Array
Custom profiler RT 2 PCR array was designed to investigate the expression of 84 genes in 96-genes in 1 sample per plate PCR array format. The 96 wells of the PCR array consisted of specific primers for 84 genes, genomic DNA contamination control (GDC), reverse transcription control (RTC), and positive PCR control. Genes related to prion pathology and inflammation were selected from the literature. The gene symbol and ResSeq number for mouse were listed in an Excel sheet and submitted to SABiosciences (SABiosciences, Frederick, MD, USA). The custom profiler RT 2 PCR array was manufactured by SAbioscience (SABiosciences, Frederick, MD, USA) and run in StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA).

First Strand cDNA Synthesis and Quantitative PCR (qPCR) Assay
All RNA samples were checked for their concentration and purity (based on 260/280 nm measurement) using Nanodrop 8000 instrument (peqLab Biotechnologies GmbH, Erlangen, Germany). An equal amount of RNA (0.65 µg) from all the treatment groups were used for first strand cDNA synthesis using RT 2 miRNA First Strand kit (SABiosciences, Frederick, MD, USA) following the manufacturer's protocol. The first genomic DNA elimination mix was prepared by combining 0.65 µg total RNA and 2 µL of 5× gDNA elimination buffer. Then, ddH 2 O was added to a final volume of 10 µL. After gently mixing the contents with a pipette (Eppendorf, Hamburg, Germany) followed by brief centrifugation (Eppendorf, Hamburg, Germany), the mix was incubated at 42 • C for 5 min. Next, the RT cocktail was prepared by combining 4 µL of 5X RT buffer 3 (BC3), 1 µL of primer and external control mix (P2), 2 µL of RT enzyme mix 3 (RE3) and 3 µL of ddH 2 O to a total volume of 10 µL reaction for each sample. Then, 10 µL of RT cocktail was added to each 10 µL of genomic DNA elimination mix. After mixing gently with a pipette, the mix was incubated at 42 • C for 15 min and immediately heated at 95 • C for 5 min to stop the reaction. Finally, 91 µL of ddH 2 O was added to each 20 µL cDNA synthesis reaction and stored at −20 • C until analysis.
Expression profiling of selected genes in mice brain tissue was conducted using a custom RT 2 -PCR array (SABioscience, Frederic, MD, USA) following manufacturers' protocol. The 96-well custom RT 2 -PCR array was designed to quantify a total of 84 genes related to prion pathology and inflammation. Five housekeeping genes (B2M, Actb, Gapdh, Gusb, and HSP90ab1), 3 reverse transcription controls (RTC), 3 positive PCR controls, and mouse genomic DNA contamination controls were included in the 96-well plate. Prior to real-time PCR profiling 91 µL DNase/RNase-free water was added to each of the 20 µL first strand cDNA products from each biological replicates in the treatment and the negative control group. A PCR master mix was prepared using the 102 µL diluted cDNA template, 1350 µL RT 2 SYBR Green PCR master mix, and 1248 µL DNase/RNase-free water. Twentyfive microliter of this mix was distributed to each well of the 96-well plate containing sequence-specific primer sets for each gene and the respective controls. Following brief centrifugation (Eppendorf, Hamburg, Germany), the plate was loaded onto StepOnePlus Real-Time PCR System (Applied Biosystems, Darmstadt, Germany) and run with a thermal program of initial heating at 95 • C for 10 min followed by 40 cycles of 95 • C for 15 s and 60 • C for 1 min. The specificity of amplification was controlled using a melting curve generated at the end of the qRT-PCR protocol.

Protein Expression and Purification
The expression and purification of mouse recombinant moPrP (29-232) were prepared as per the procedure described previously [19]. Shortly, a synthetic gene including a 22-residue N-terminal fusion tag containing 6x His and a thrombin cleavage site (MGSSH-HHHHHSSGLVPRGSHML) was synthesized by DNA 2.0 (Menlo Park, CA, USA). The gene was cloned into a pET15b expression vector between XhoI and EcoRI restriction sites and heat shock transformed into Escherichia coli strain BL21 (DE3). For expression, the transformed cells were grown in 100 mL Luria-Bertani broth plus 100 µg/mL ampicillin overnight to generate a starter culture. Between 1% and 2% of this starter culture was then used to inoculate 1 L of Luria-Bertani media (giving a starting D 600 of 0.1). The cells were allowed to reach a D 600 between 0.6 and 1.0 before induction with 1 mM isopropyl thiogalactoside. Twelve to eighteen hours later, the cells were harvested by centrifugation at 1600× g for 25 min at 4 • C. In addition, 15 N -labeled moPrP (90-232) was also expressed and purified from M9 media (1.0 g/L) 15 NH4 Cl) for the collection of heteronuclear NMR data. The inclusion of the 6x His tag afforded a standardized nickel affinity purification strategy for the protein construct.

Statistical Analysis
The mRNA expression of selected genes included in the custom PCR array was analyzed using the ∆∆CT method from the PCR array data analysis web portal (http: //www.sabiosciences.com/pcrarraydataanalysis.php accessed on 15 June 2015). The software calculates average threshold cycle values for the biological replicates from both the treatment and control groups for all the genes in the custom PCR array. Those genes having >35 ct values and undetected ones were removed from the analysis. Data were normalized by correcting all comparative threshold (CT) values for the average ct value of endogenous controls present in the array. After sequential computations, the software calculates fold regulation and associated t-test p-value. Thus, fold change values >1.5 and p ≤ 0.1 were taken as cutoff values to identify genes that are differentially expressed in the treatment groups as compared to the saline-treated negative control group.

Results and Discussion
Previously, we showed that LPS from E. coli 0111:B4 instantly converts recombinant Syrian hamster and mouse prion protein (ShPrP C and mPrP C ) into a beta-rich isoform PrP res resistant to proteinase K [17]. In the present study, we tested the hypothesis that the moPrP res , administered once sc, would cause prion disease in FVB/N female mice and alter the expression of genes related to prion disease and inflammation in the brain of mice as early as 11 weeks pi and terminal stage. Indeed, moPrP res caused sickness and death in 60% of the experimental mice with clinical signs similar to prion disease and affected the expression of genes in the brain at 11 weeks pi and more profoundly at the terminal stage. Moreover, combinations of moPrP res with LPS or treatment with RML, RML + LPS, and LPS alone also affected the expression of multiple genes at 11 weeks pi and at a terminal stage as will be discussed in detail below ( Figure 1).

Subcutaneous Injection of LPS, moPrP res , RML and Combinations of LPS Altered Gene Expression in the Brains of Mice at 11 Weeks Post-Infection
The mRNA expression of genes related to prion disease and inflammation at 11 weeks pi were analyzed in all treatment groups including LPS, moPrP res , RML, and combinations of LPS treated groups in the brain tissue (Table 1). Results indicated that moPrP res treated mice showed up-regulation of Ifi2912a and Nos2 and down-regulation of Apoe, Sprn, and C1qb at 11 weeks pi. Subcutaneous administration of LPS for a period of 6 weeks downregulated Apoe, GbP4, Grn, Sod1, Bax, and Ccl17. Combination of moPrP res and LPS up-regulated 5 genes (Tlr3, Tlr6, Nos2, Il1a, and Il1f10) and down-regulated C4b. On the other hand, RML treatment down-regulated C1qb, C4b, Grn, Anp32a, and Sod1, whereas a combination of RML with LPS down-regulated C1qb, Rtp4, and Sod1 and up-regulated H2-T23 and Ccl25.
Interestingly, treatment with moPrP res down-regulated Sprn (shadow of prion protein) at 11 weeks pi indicating its early alteration during the disease development. The product of this gene, Sho, is typically decreased during scrapie disease [14]. This suggests that Sprn may be used as a marker for the progression and development of prion and prionlike diseases.

Subcutaneous Injection of LPS, moPrP res , RML and Combinations of LPS Altered Gene Expression in the Brains of Mice at 11 Weeks Post-Infection
The mRNA expression of genes related to prion disease and inflammation at 11 weeks pi were analyzed in all treatment groups including LPS, moPrP res , RML, and combinations of LPS treated groups in the brain tissue (Table 1). Results indicated that moPr-P res treated mice showed up-regulation of Ifi2912a and Nos2 and down-regulation of Apoe, Sprn, and C1qb at 11 weeks pi. Subcutaneous administration of LPS for a period of 6 weeks down-regulated Apoe, GbP4, Grn, Sod1, Bax, and Ccl17. Combination of moPrP res and LPS up-regulated 5 genes (Tlr3, Tlr6, Nos2, Il1a, and Il1f10) and down-regulated C4b. On the other hand, RML treatment down-regulated C1qb, C4b, Grn, Anp32a, and Sod1, whereas a combination of RML with LPS down-regulated C1qb, Rtp4, and Sod1 and up-regulated H2-T23 and Ccl25.   Moreover, C1qb was commonly down-regulated by moPrP res , RML, and RML/LPS treatments at 11 weeks pi. The complement protein C1qb binds to a very wide range of non-self and altered-self-molecules. For instance, amyloid proteins are a group of alteredself-proteins to which C1qb binds. Mice deficient in C1qb, or temporarily depleted of complement component C3 by treatment with cobra venom factor showed longer survival following scrapie infection, confirming that interactions between prion and complement at the very early stages of infection affect the course of neurodegeneration [20].
The up-regulation of Nos2 by moPrP res and moPrP res /LPS treatments at 11 weeks pi signals the production of nitric oxide where its overproduction may compromise energy production in neurons and initiate the process of neurodegeneration [21]. The downregulation of Sod1 in LPS, RML, and RML/LPS groups may be a protective response of the host to the development of neurodegeneration. Sod1 knockout animals show no signs of neurodegeneration while generation of transgenic animals expressing mutant Sod1 leads to motor neuron degeneration without reduction in Sod1 activity [22].

Reduction of Sprn mRNA Expression at Terminal Stage
Interestingly, the results showed that only one gene (i.e., Sprn-Shadoo (Sho) of prion protein) was differentially expressed by all treatments (moPrP res , LPS, moPrP res /LPS, RML, and RML/LPS) applied in this study (Tables 2-4). Down-regulation of Sprn was more pronounced in the group of mice treated with moPrP res /LPS (−3.46-fold) versus −3.20-fold in the LPS-treated mice, −3.13-fold in RML-treated mice as well as −2.39and −1.94-fold in mice treated with moPrP res or RML/LPS, respectively. Previously, Sho was described as a neuronal glycoprotein and a member of the mammalian prion protein family, including PrP C and Doppel (Dpl) [29]. Recently, it was shown that there is a strong decrease of Sho protein in the brain of mice clinically ill with prion disease and that Sho has protective effects against prion disease [15,16]. Moreover, lowered levels of endogenous Sho have been shown to trace an early response of PrP Sc buildup in the CNS and that Sho downregulation is a common and typical event in neurodegeneration conditions caused by prion strain isolates [14]. Moreover, another study showed a relationship whereby the degree of down-regulation of the mature Sho protein was inversely related to concentrations of PrP Sc [15]. Therefore, down-regulation of Sprn in mice treated with moPrP res , RML, LPS, and combinations of moPrP res and RML with LPS supports the hypothesis that moPrP res and other treatments caused prion-like disease in infected mice. The most intriguing observation was the down-regulation of Sprn by chronic subcutaneous LPS treatment. To the best of our knowledge, this is the first report that shows that chronic sc administration of LPS can lower Sprn expression in the brain tissue.  In a series of articles published recently [14][15][16]30,31], it is indicated that a typical finding during prion disease is decreased Sho and PrP C in the brain of animals or humans affected by prion disease. A decrease of Sho and PrP C are typical findings of prion disease and not present in other misfolding disorders like cytoplasmic accumulation of Tau in Tg(P301L)23027 mice [32], or parenchymal and vascular accumulation of Aβ and Bri peptide in TgCRND8 and TgADanPP7 mice, respectively [33][34][35]. Indeed, expression of Prnp also was decreased by moPrP res , moPrP res /LPS, RML, and RML/LPS treatments. These findings strongly support the clinical observations of prion neurodegenerative disease in mice under those treatments. Of note is the lack of significant effect of LPS on the Prnp gene, although LPS treatment numerically decreased expression of Sprn.

Decreased Expression of ApoE and Bax in moPrP res , moPrP res /LPS, and LPS Treatments
Two other important genes that are known to be associated with prion disease (i.e., ApoE and Bax) also were differentially expressed in 3 treatments involving moPrP res , moPrP res /LPS, and LPS compared with CTR mice (Tables 2 and 3). The gene encoding apoE was down-regulated −2.53-fold by LPS, more than −3-fold by moPrP res , and more than −10-fold by moPrP res /LPS treatment. It has been widely documented that apoE is present in high concentrations in neurons following brain injury [36]. Moreover, some studies have shown that increased ApoE expression leads to synaptic regeneration [37]. In contrast, in ApoE-deficient mice, synaptic plasticity and regeneration are impaired [38].
Down-regulation of ApoE in our experiment suggests the contribution of ApoE downregulation in the development of brain neurodegenerative disease in terminally ill mice treated with moPrP res , LPS, and moPrP res /LPS. The mechanism by which the ApoE gene is affected during brain neurodegeneration warrants further investigation.
Another important protein that has been shown consistently to participate in prion disease is Bax. Bax was down-regulated by moPrP res , moPrP res /LPS, LPS, and RML scrapie isolate treatments in our study. In a recent study, it was shown that Bax deletion in Tg(PrP∆32-134) mice delays the development of clinical illness and slows down the apoptosis of cerebellar granule cells [39]. The Bax molecule resides in the cytoplasm and responds to various stimuli by migration to the mitochondria [40] where it can cause cytochrome C release [41], thereby activating apoptotic protease activating factor (Apaf)-1 dimerization and the apoptotic cascade. Bax is a powerful executioner of neurons and PrP has been shown to protect neurons from Bax-mediated cell death [42]. Our data suggest that down-regulation of Bax by these treatments might be a host response to slow down the process of neuronal cell death.
Down-regulation of the Atp1b1 gene by moPrP res /LPS, RML, and RML/LPS treatments agrees with previous research indicating that failure of Na + /K + -ATPase (i.e., sodium/potassium pump) is implicated in the pathogenesis of several neurodegenerative disorders [43]. Atp1b1 encodes the β1 unit of Na + /K + -ATPase, which is important in repolarization of neuronal plasma membrane after excitatory depolarization, the transmission of action potentials in neurons, and it is associated with caveolae and intracellular signal transduction events [44].
The neural cell adhesion molecule (Ncam), encoded by Ncam1, is a cell membrane constituent with three isoforms. Ncam1 plays significant roles in interneuronal and glianeuronal adhesion phenomena, cell-cell recognition, development of the nervous system, synaptic plasticity, memory and learning as well as re-myelination and post-injury regeneration [45][46][47]. Therefore, down-regulation of this gene by moPrP res /LPS, RML, and RML/LPS suggests that multiple neuronal functions have been affected by these treatments. These 3 treatments also differentially expressed Prkaca (Protein kinase, cAMP-dependent, catalytic, alpha), which is suggestive of apoptotic processes occurring in the brain. The protein kinase encoded by this gene binds cAMP and initiates phosphorylation of various downstream substrates and has been shown to function as a neuroprotective molecule with anti-apoptotic activity [48,49].

Down-Regulation of SOD1 in moPrP res , RML, and RML/LPS Treatments
Treatment of mice with moPrP res , RML, and RML/LPS also decreased the expression of Sod1. Decreased superoxide dismutase (Sod) activity has been shown previously in prion protein knockout mice and cell cultures [50][51][52][53], and a loss of Sod function is one of the proposed mechanisms of prion disease pathogenesis. Alterations in the activity and redistribution to mitochondria have been reported for Sod1 mutants associated with genetic amyotropic lateral sclerosis in humans and mice [54] indicating that mislocalization and altered activity of the Sod enzymes can have deleterious consequences for neurons. Sod activity is lowered in 4 neurodegenerative diseases including AD, PD, ALS, and DS [55].

Up-Regulation of Immune Response Genes by moPrP res Treatment
Other genes up-regulated by moPrP res treatment were Ifitm3, Myd88, Tlr4, Ccl5, Tnf, Fcgr3, Lyz2, Mdk, Tlr1, and Il1ra. Up-regulation of genes related to innate immunity like Tlr4, Myd88, Tnf, Tlr1, and Il1rn were characteristic for the moPrP res treatment only. Interestingly, mutations in the TLR4 have been shown to expedite prion disease [56]. Moreover, Myd88 -/mice were fully affected by scrapie as wild-type mice, suggesting that protective effects of activation of TLR4 signaling do not result from direct interaction with prions [57]. On the other hand, MyD88 is a cytoplasmic adapter protein that associates as an obligate functional partner with all members of the TLR including TLR4 and interleukin-1 receptor (IL-1R) family [58,59]. Moreover, overexpression of Tnf is important because TNF stimulates the NF-κB transcription factor, which induces the expression of proinflammatory cytokines, and promotes the synthesis of neuronal survival factors such as calbindin, manganese superoxide dismutase, and anti-apoptotic Bcl-2 protein [60]. This cytokine also can stimulate microglia glutaminase to release glutamate, thus, generating excitotoxicity [61] and promoting the development of neurodegenerative diseases.
Treatment with moPrP res up-regulated Ccl5, a gene that has been implicated in neurodegenerative diseases including AD. Elevated expression of Ccl5 in cerebral microcirculation is associated with the recruitment of immune cells [62]. It has also been demonstrated in cell culture models that Ccl5 can stimulate chemotaxis, increase nitric oxide secretion, and attenuate IL-10 and insulin growth factor (IGF)-1 production in activated microglia [63]. In addition to its chemoattractant function, Ccl5 also has a modulatory effect on microglia activation [61]. Treatment of neurons with Ccl5 has resulted in an increase in cell survival and a neuroprotective effect against the toxicity of thrombin and sodium nitroprusside [64]. It is obvious that mice treated with moPrP res are responding to infection with enhanced Ccl5 gene expression to attract and activate microglia at the site of inflammation.
Of note was the overexpression of Mdk (Midkine) by moPrP res , which encodes a protein that is important in the attraction of immune cells to the site of inflammation. In line with this Mdk -/mice show significantly suppressed recruitment of inflammatory cells [65]. Midkine also plays a significant role as an anti-apoptotic factor for neuronal cells during apoptosis induced by serum starvation of neurons [66].
Although the role of Lyz2 in brain neurodegenerative diseases is not clear, it is interesting to note that a recent report indicates that human lysozyme was able to prevent amyloid aggregation of the Aβ peptides [67]. Therefore, it is tempting to speculate that up-regulation of Lyz2 in moPrP res -treated mice in our experiment might provide more lysozyme to the host to counteract the establishment of prion fibrils. However, this warrants further investigation.
Subcutaneous treatment with moPrP res also up-regulated Fcgr3. Our results agree with previous research demonstrating up-regulation of this gene in scrapie-infected animals [10,68]. Fcγ receptors (FcγRs) for IgG are expressed on a wide variety of immune cells, linking cellular and humoral immunity. Engagement of activating FcγRs, associated with the common γ-chain, triggers effector cell responses, such as antibody-dependent cell-mediated cytotoxicity, phagocytosis, reactive oxygen production, and release of inflammatory mediators [69]. During inflammation, FcγRs play important roles in leukocyte recruitment and activation. More specifically FcγR3 mediates neutrophil tethering and adhesion in response to immune complexes during inflammation [70,71]. Up-regulation of the Fcgr3 gene in the brain of mice treated with moPrP res in our experiment suggests host responses to increase recruitment of leukocytes to the site of inflammation.
The other two genes up-regulated by moPrP res treatment were Tlr1 and Ilr1a. Tlr1 and Tlr2 have been shown to be primary receptors for Aβ and activation of inflammatory processes [72]. In addition, genetic deletion of IL1ra (also known as IL-1rn) in mice increases microglia activation and reduces neuronal survival in response to intracerebroventricular infusion of Aβ in human subjects [73].

Immune Response Genes Uniquely Regulated by LPS Treatment
Subcutaneous chronic infusion of LPS differentially expressed 19 genes. Out of these genes, only 6 were uniquely differentially expressed by LPS treatment including Fcgr2b, Ccl17, Ccl19, Ccl25, Grn, and H2-T23. Down-regulation of the Fcgr2b gene is in alignment with previous reports that Fcgr2b knockout mice are protected from neurodegeneration in AD [72]. Additionally, it has been reported that the FcγRIIb is a receptor of Aβ 1-42 . Furthermore, FcγRIIb is significantly up-regulated in the hippocampus area of the AD brains and in neuronal cells exposed to synthetic Aβ and that genetic depletion of Fcgr2b rescues memory impairments in an AD mouse model. Mice treated with LPS in our experiment are responding with down-regulation of Fcgr2b potentially to prevent neurodegenerative processes in the brain [74].
It is interesting to observe that two chemokine genes (Ccl17 and Ccl25) were upregulated, whereas Ccl19 was down-regulated by LPS. The function of chemokines is to regulate leukocyte trafficking at the site of inflammation. In an earlier study, it was shown that mice deficient in Ccl17 lowered deposition of Aβ in the brain and were protected against neuronal loss and cognitive deficits. In addition, in absence of Ccl17, there is accelerated uptake and degradation of Aβ [75]. Interestingly, receptors for both Ccl17 and Ccl25 are in the hippocampal neurons, the part of the brain that is related to learning and memory [76]. On the other hand, Ccl19 was reported to be constitutively transcribed in the normal human CNS and that its transcription was elevated in MS patients. The locally produced Ccl19 might be involved in the maintenance of different types of immune cells in the brain. Therefore, subcutaneous treatment with LPS apparently affected the expression of brain chemokines to attract immune cells to the sites of inflammation.
Another gene down-regulated by LPS was Grn. This gene encodes the protein granulin. Granulin is expressed in neurons and microglia [77]. Mutations in the Grn have been shown to give rise to frontotemporal lobar degeneration (FTLD) [78]. Granulin inhibits TNF and promotes the up-regulation of Th2 cytokines such as IL-4, IL-10, and IL-5 [79]. Progranulin was shown to play a role in wound healing, where it triggers inflammation and increases the accumulation of neutrophils, macrophages, fibroblasts, and the formation of new blood vessels [80]. This suggests that down-regulation of Grn by LPS has influenced the etiopathology of the neurodegeneration process in the treated animals.
It is interesting to note that LPS and RML commonly affected 8 genes (Sprn, Ncam1, Tlr3, Ache, Fcgr3, Ly86, Anp32a, and Bax). Intriguingly, LPS differentially expressed 8 genes that have been previously reported in experimental scrapie disease in rodents [81]. This implicates LPS as a potential co-factor in prion disease and suggests that the effects of LPS on the etipathogenesis of neurodegeneration involve similar pathogenic mechanisms with scrapie agents.

Conclusions
In conclusion, data from this study indicate that moPrP res generated in vitro by incubation with LPS was able to cause a gene expression signature of neurodegenerative disease with a resemblance with prion disease. For instance, moPrP down-regulated two genes typically related to the development of prion diseases like Sprn and Prnp. The combination of moPrP res with LPS also differentially expressed genes typically related to prion disease especially Sprn, Prnp, and Prnd. Of interest is the signature of genes differentially expressed by chronic subcutaneous administration of LPS that commonly expressed 8 genes with RML, typical of the prion disease like down-regulation of Sprn and up-regulation of Prnd. RML and a combination of RML with LPS also affected a variety of genes that are involved in scrapie and that have been reported previously by other investigators. More research is warranted to study the potential role of chronic LPS and LPS-converted PrP res in the development of neurodegenerative diseases in humans and animals.  Institutional Review Board Statement: All the protocols used during the study were in accordance with the Canadian Council on Animal Care and were approved by the Animal Care and Use Committee (Animal Use Protocol #660), Health Sciences, University of Alberta.