Altered Expression of Urea Cycle Enzymes in Amyloid-β Protein Precursor Overexpressing PC12 Cells and in Sporadic Alzheimer’s Disease Brain

Abstract

Urea cycle enzymes may play important yet poorly characterized roles in Alzheimer’s disease (AD). Our previous results showed that amyloid-β (Aβ) affects urea cycle enzymes in rat pheochromocytoma (PC12) cells. The aim of the present study was to investigate the changes in arginases, other urea cycle enzymes, and nitric oxide synthases (NOSs) in PC12 cells transfected with AβPP bearing the double ‘Swedish’ mutation (APPsw, K670M/N671L) and in postmortem sporadic AD brain hippocampus; the mutation intensifies Aβ production and strongly associates with AD neuropathology. mRNA expression was analyzed using real-time PCR in cell cultures and DNA microarrays in hippocampal CA1 area of human AD brains. Arginase activity was measured spectrophotometrically, and arginine, ornithine, and citrulline levels by high-performance liquid chromatography. Our data demonstrated that the expression and activity of arginases (Arg1 and Arg2), as well as the expression of argininosuccinate synthase (Ass) were significantly reduced in APPsw cells compared to control. However, argininosuccinate lyase (Asl) was upregulated in APPsw cells. Real-time PCR analysis revealed significant elevation of neuronal nitric oxide synthase (Nnos) mRNA in APPsw cells, without changes in the endothelial Enos, whereas inducible Inos was undetectable. The changes were found to follow closely those observed in the human hippocampal CA1 region of sporadic AD brains. The changes in enzyme expression were accompanied in APPsw cells by significantly elevated citrulline, ornithine, and arginine. Our findings demonstrate that AβPP/Aβ alters arginine metabolism and induces a shift of cellular homeostasis that may support the oxidative/nitrosative stress observed in AD.

Keywords

Alzheimer’s disease amyloid β amyloid-β protein precursor arginase arginine argininosuccinate lyase argininosuccinate synthase citrulline nitric oxide synthase ornithine

INTRODUCTION

Alzheimer’s disease (AD) is the most widespread age-related neurodegenerative disorder characterized by long period of silent development and devastating outcome. Despite decades of research effort, the diagnosis of AD is still difficult and no efficient treatments are currently available. Age is the main risk factor, and a relatively minor fraction of AD patients suffer from the early-onset AD. This vastly complicates the research effort, although both familial and sporadic AD are characterized by a similar set of symptoms and share the neuropathological characteristics that include synaptic dysfunction, neuron loss, inflammatory microglial reaction against the dying cells, and damage to brain microvasculature. Extracellular accumulation of amyloid-β (Aβ) is the biochemical hallmark of AD. Aβ is generated via sequential cleavage of the membrane protein AβPP (amyloid-β protein precursor) by the proteases β- and γ-secretase. Despite clear association of mutations in the gene coding for AβPP with the rare, early-onset AD [1], the mechanism underlying Aβ overproduction, aggregation, and toxicity in sporadic AD is not well understood. The ‘Swedish’ double AβPP mutation (APPsw) affects two adjacent amino acids at the β-cleavage site (K670M/N671L) leading to dominant early AD with very high penetrance [2, 3]. Patients bearing this mutation display intense accumulation of Aβ peptides in brain parenchyma and cerebral amyloid angiopathy in all cortical lobes and in the cerebellum. The mutation disturbs proteolytic AβPP processing (increasing the β-cleavage, [4]); cells expressing APPsw produce ∼2×to 7×more Aβ, depending on the cell line [5, 6]. APPsw-expressing animal models [7, 8] recapitulate a number of crucial AD features including gradually accumulating Aβ deposits in the cerebral cortex and hippocampus, alterations in multiple neurotransmitter systems and signaling pathways ([7 , 10]) along with progressive behavioral abnormalities. Disturbances in the free radical second and retrograde messenger nitric oxide (NO) are frequently noted in neurodegenerative disorders where abnormal neurotransmission and/or deregulated immune response plays a role, as is observed in AD. NO synthases (NOSs) have been increasingly implicated in the pathophysiology of AD [11]. Both constitutive isoforms, neuronal (nNOS) and endothelial (eNOS) can contribute to synaptic plasticity [12] and the pathological role of iNOS in the deregulated immune response in AD has long been recognized. However, the changes observed in the metabolism of the substrate for NO synthesis, the amino acid L-arginine, and its by-product L-citrulline vary heavily between research works; both compounds are crucial cellular metabolites and their disturbances might heavily impact the pathophysiology of AD.

Arginine and citrulline are metabolites of the arginine cycle, also termed urea cycle. Argininosuccinate synthase and -lyase (ASS and ASL, respectively) recycle citrulline back into arginine. Arginases (ARG-1, -2) metabolize arginine into the polyamine precursor ornithine; apart from substrate competition they can modulate NOS activity and product spectrum [13 –15]. The cycle is typically incomplete in healthy brain due to lack of ornithine carbamoilotransferase (OTC). Normally OTC produces citrulline from ornithine [13] in a reaction that also allows introduction of CPS I (carbamoylphosphate synthase 1) – synthesized carbamylphosphate (and thus NH₄) into the cycle (Fig. 1). Other enzymes of the cycle undergo expression in the brain and locally deliver the necessary metabolites. Arginases are present in several neuron types, some glia and vascular endothelium [16, 17]) while ASS and ASL have been immunodetected predominantly in neurons in multiple brain areas [13]. A number of works has shown varying changes in their expression in AD as well as induction of OTC in some cell types [18 –20]. However, the available information is incoherent [6 , 22] and thus cannot provide clear conclusion [23]. This necessitates careful examination of the direct effects of Aβ/AβPP and their separation from the more distal influences which could come from AD-linked tissue/organism-level phenomena such as, for example, immune activation, or cerebrovascular pathology. Recent work on PC12 cells has demonstrated that elevated Aβ/AβPP has led to creation of pro-nitrosative conditions by simultaneous increase in the expression and activity of NO synthases and downregulation of the expression of arginases [23, 24].

Fig.1

Arginine metabolism. Arginine is a substrate of NOS which generates NO and citrulline; the latter is recycled back into arginine by sequential action of ASS and ASL. Alternatively, arginases metabolize arginine into ornithine, which may become a substrate for polyamine biosynthesis. In most brain cells and many neuronal and related cell lines the cycle is incomplete due to lack of OTC. ARG, arginases; ASL, argininosuccinate lyase; ASS, argininosuccinate synthase; CPS, carbamoylphosphate synthetase; NO, nitric oxide; NOS, NO synthases; ODC, Ornithine decarboxylase; OTC, Ornithine transcarbamoilase.

The aim of this study was to analyze the effect of overexpression in PC12 cells of AβPP carrying double ‘Swedish’ mutation (APPsw, K670M/N671L) associated with early-onset AD on the expression of NOS isoforms, arginases, and other urea cycle enzymes as well as the level of the cycle amino acid intermediates. The mRNA expression of the enzymes was then measured in the postmortem samples of human CA1 area of the hippocampus in AD and control brain tissue using DNA microarrays.

MATERIALS AND METHODS

Cell culture

Rat PC12 pheochromocytoma cell lines stably transfected with expression construct containing human AβPP gene with the double “Swedish” mutation (APPsw, K670M/N671L) [24] were used along with empty vector-transfected controls. APPsw cells display nearly 2×higher Aβ level than wt AβPP -transfected cells [6]. The cells were cultured at 37°C, 5% CO₂ in a humidified incubator in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum, 5% heat-inactivated horse serum, 2 mM L-glutamine, 50 U/ml penicillin, 50 μg/ml streptomycin, and 400 μg/ml G418.

Human brain tissue

Human brain samples were selected from archived tissues or extracts at the LSU Neuroscience Center, New Orleans LA, from the University of California at Irvine, CA, and from the Oregon Health Sciences Center, Portland, OR. Additional details of these human brain tissues have been previously reported [25]. Human tissue use was in accordance with the stipulations of institutional review board at the LSU Health Sciences Center and donor institutions [25 –28]. All RNAs were isolated from sporadic AD/control samples having a mean postmortem interval (PMI; death to brain-freezing interval) of 2.9 h or less as PMI can affect RNA quality [25 –27]. AD tissues used in these studies had a clinical dementia rating (CDR; an index of cognitive decline) ranging from a CDR of 0.5 to 3.0, indicating mild- to a severe-stage; CERAD/NIH criteria were used [27, 28].

Preparation of samples for western blot analysis

To obtain cellular extracts for Aβ measurement, control PC12 and APPsw-expressing cells were seeded in equal numbers onto 0.1% PEI -coated 6-well dishes. After 24 h incubation at 37°C in low-serum medium (DMEM with 2% FBS, 50 U/ml penicillin/streptomycin and 2 mM L-glutamine), cells were washed 3×with phosphate buffered saline (PBS), lyzed in 1×Lysis Buffer (Cell Signaling, Danvers, MA, USA) and denatured in Laemmli sample buffer.

To obtain the cell culture medium for AβPP measurement, PC12 and APPsw-expressing cells were seeded in equal numbers onto 6-well 0.1% PEI-coated plates and the growth medium was changed to a free-serum medium (DMEM supplemented with 50 U/ml penicillin/streptomycin and 2 mM L-glutamine). After 24 h the medium was harvested, centrifuged for 5 min at 400×g at 4°C and the supernatant frozen in liquid nitrogen for storage. Protein was precipitated from thawed medium with trichloroacetic acid (TCA, final concentration 20%) and bovine serum albumin (50 μg/vial, as carrier protein). After 30 min incubation at room temperature (RT) the samples were centrifuged at 13 000×g for 4 min. The pellets were washed twice with cold (freshly taken from –20°C) acetone, dried at 95°C for 10 min and denatured in 0.55 M Tris-based Laemmli sample buffer.

Western blot analysis of AβPP and Aβ peptide levels

The quantities of AβPP and Aβ were estimated by western blot. Samples were electrophoresed in equal amounts (protein measurements according to Bradford) through 10% (AβPP) or 16.5% (Aβ) SDS-PAGE/Tricine polyacrylamide gels as described previously [23]. Transfer onto nitrocellulose membranes was done for 30 or 120 min at 50 to 120 V to achieve optimum efficiency for various molecular weights. The membranes were washed for 5 min in Tris-buffered saline (TBS)-Tween buffer (TBST; 100 mM Tris-HCl, pH 7.6, 140 mM NaCl, 0.1% Tween 20), blocked for 30 min at RT with 5% non-fat dry milk in TBST and washed further 3×for 5 min in TBST. Detection of AβPP and Aβ was performed with mouse monoclonal Ig (1 : 250, clone#20.1, Santa Cruz Biotechnology, Santa Cruz, CA, cat.# sc-53822). The antibody was dissolved in TBST and membranes were incubated overnight at 4°C. After 3 washes (5 min) in TBST the membranes were incubated with secondary antibody for 2 h at RT (1 : 2000 anti-mouse IgG, GE Health Care UK, Little Chalfont, Buckinghamshire, UK) in a 5% non-fat dry milk/TBST. Signals were visualized with ECL (Amersham Biosciences, Bath, UK) under standard conditions and quantitated using Phoretix software, TotalLab, Newcastle upon Tyne, UK. After stripping, the membranes were re-probed for GAPDH (used as a loading control).

Gene expression measurement: Real-time polymerase chain reaction

The cells were washed twice and scraped down in PBS and centrifuged (3 min, 1000×g). RNA was isolated using TRI-reagent and DNA digested with DNase I according to manufacturer’s protocols (Sigma-Aldrich). RNA content and purity was assessed spectrophotometrically (A260/A280 nm ratio). Reverse transcription was performed with High Capacity Reverse Transcription Kit according to the instructions (Applied Biosystems, Foster City, CA, USA). Real-time PCR was done with TaqMan Gene Expression Assays (Applied Biosystems, Foster City, CA, USA) on ABI PRISM 7500 apparatus according to the manufacturer’s instructions (Applied Biosystems, Foster City, CA, USA). Gene expression was quantified using ddCt method and normalized against actin B (Actb).

Microarray measurement of gene expression using DNA arrays

Total DNA, RNA and proteins were isolated from control and AD-affected human brains in primary culture using TRIzol (Invitrogen) as previously described by our laboratory [29 –34]; RNA quality was assessed using an Agilent Bioanalyzer 2100 (Lucent Technologies/Caliper Technologies; Palo Alto, CA, USA) and RNA integrity numbers (RIN) values were typically 8.0–9.0 indicating high quality total RNA [27 , 33–36]. Control and AD brain RNA samples were labelled and hybridized and analyzed using GeneChips (Affymetrix, Palo Alto CA, USA) as previously described by our group [27 , 32–36].

Spectrophotometric measurement of arginase activity

Arginase activity was measured as described previously [23]. Cells were rinsed twice in PBS and scraped in 1 ml of PBS. After centrifugation for 10 min at 400×g the cell pellet was homogenized in 0.5 ml of 10 mM Tris-HCl, pH 7.5 containing 5 mM MnCl₂ and 100 mM KCl. The reaction was started by addition of 100 μl of cell homogenate to the reaction mixture containing: 100 μl of barbiturate buffer, pH 9.2 (50 μM sodium barbiturate, 3 mM HCl), 20 μl of 50 mM MnCl₂, and 20 μl of 200 mM arginine (final volume 240 μl). Blank controls did not contain arginine. The reaction was carried out for 30 min at 37°C and stopped by adding 260 μl of 20% trichloroacetic acid. Then arginine was added to blank samples. The samples were centrifuged for 10 min at 15,500×g and 400 μl of supernatant was added to 400 μl of 99.5% acetic acid and 200 μl of ninhydrin reagent (25 mg/ml ninhydrin, 2.4 M H₃PO₄) and boiled at 100°C for 60 min. Absorbance was measured at 515 nm in a Shimadzu spectrophotometer against 5 μM to 200 μM ornithine curve.

Measurements of arginine and its metabolites: High-performance liquid chromatography (HPLC)

Cells were scraped into 1 ml of PBS, centrifuged, re-suspended in 45 mM phosphate buffer pH 6.2 with 10% acidic methanol and sonicated (10 cycles 3 s each, 10% power, BioLogics 150 V/T). After centrifugation (10 min at 13,000×g) the supernatant was stored at –20°C and used for HPLC amino acid analysis; protein pellet was re-suspended in 0.5 ml of 1 M NaOH for protein measurement (Bradford).

Sample derivatization was done in a timed reaction with o-phtalaldehyde and mercaptoethanol as described previously [23]. Derivatized samples (25 μl) were injected onto a 150×4.6 mm 5X Hypersil Gold BDS C18 column and solved in mobile phase of 50 mM phosphate buffer (KH₂PO₄/K₂HPO₄) containing 10% v/v methanol, pH 6.2 (solvent A), and methanol (solvent B) with fluorescence detection.

Determination of the effect of APPsw expression on nitrite level

Nitrite levels were determined in cell medium using Griess reagent as described previously [37]. Briefly, cell culture medium supernatant or serial dilution of NaNO₂ standard (linear range 0–100 μM) were plated onto a microtiter plate and total nitrite in each sample was then determined by adding 50 μl of the Griess reagent 1 (1% sulfanilamide in 1M HCl), followed by reagent 2 (0.1% N-(1-napthyl)-ethylenediamine). After 10 min incubation in the dark at RT, optical density was measured at 540 nm.

MTT reduction test for the viability of cells treated with a nitric oxide donor

Control PC12 and APPsw-expressing cells were seeded at 2×10⁴ cells/well onto collagen-coated 96-well dishes, cultured for 24 h at 37°C, then switched to low-serum medium (DMEM with 2% FBS, 50 U/ml penicillin/streptomycin and 2 mM L-glutamine). DEANO (sodium 2-(N, N-diethylamino)-diazenolate-2-oxide) was added to final concentration of 0.1 mM at indicated time points (24, 16, 6, 3, and 0 h from the end of experiment). Then MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) was added; after 2 h the medium was removed, formazan was solubilized with DMSO, and measured at 595 nm.

Statistical analysis

The results were expressed as mean values±S.E.M. Differences between the means were analyzed using Student’s t-test between two groups, or two-way analysis of variance (ANOVA) with Tukey post-hoc test among multiple groups. Statistical significance was accepted at p < 0.05. The statistical analyses were performed using Graph Pad Prism version 5.0 (Graph Pad Software, San Diego, CA).

RESULTS

APPsw cells produce more AβPP and Aβ than APPwt line

Our previous results indicated 2-fold elevated cellular Aβ in PC12 cells overexpressing human wild type AβPP (APPwt) as compared to control cells [23]. In the present study, we showed that PC12 line expressing the APPsw mutation produced 463±186% of the endogenous Aβ comparing to control PC12 cells (Fig. 2A, p < 0.05), which was approximately twice the value found in cells transfected with wild-type AβPP [23]. In addition, here we observed that AβPP is released from APPsw cells into the extracellular space along its cleavage product Aβ, as previously observed in, for example, neurons or SH-SY5Y neuroblastoma [38, 39]. The amounts of secreted AβPP were 80×higher in APPsw cell line than in controls (Fig. 2B, p < 0.001).

Fig.2

Aβ peptide and AβPP levels in control PC12 and APPsw cells. Total immunoreactivity of Aβ in cellular extracts (A) and secreted AβPP in the culture media (B) were analyzed with Western blotting in control cells (PC12 cells transfected with an empty vector; lanes marked as ‘a’) and cells transfected with double Swedish-mutated AβPP (APPsw; lanes marked as ‘b’) as described in Materials and Methods. Densitometry results of Aβ were normalized against GAPDH as loading control; values of secreted AβPP were compared to the intensity of Ponceau S staining. Example blots are shown. Data represent mean values±S.E.M. for 4 independent experiments. ^*p < 0.05; ^***p < 0.001 versus control PC12 cells, Student’s t-test.

NO metabolism is altered in APPsw-expressing cells

To examine the alterations of NO synthases that metabolize arginine into NO and citrulline, we compared the expression of the three genes coding for NOS isoforms (Nnos, Enos, Inos). Real-time PCR measurements showed a moderate but significant rise of Nnos RNA expression in APPsw cells as compared to mock, empty vector- transfected PC12 cells (137.1±10.6% of control, p < 0.05) without changes in Enos (Fig. 3A). Inos mRNA was not detectable in either cell line (not shown). In parallel with the transcriptional analysis, we looked at whether the changes in gene expression might have an impact on the levels of nitrites released to the cell medium. We observed that nitrites (measured using Griess reaction) were elevated by 28.6% in APPsw cells as compared to mock, empty vector- transfected PC12 cells (Fig. 3B). This was in agreement with previous results, where APPsw-expressing cells displayed elevated NO synthesis [24].

Fig.3

The effect of AβPP on the mRNA expression of NOS isoforms and on nitrite levels. A) Total RNA was isolated from PC12 and APPsw cells, treated with DNase, and reverse-transcribed in equal amounts as stated in Materials and Methods. Real-time PCR results for Nnos and Enos mRNAs ere expressed as relative quantity, Rq normalized against actin b expression. Inos mRNA was not detectable (not shown). Data represent mean values±S.E.M. for ≥3 independent experiments. ^*p < 0.05 versus PC12, Student’s t-test. B) Nitrites in cell culture medium from control and APPsw lines were measured using Griess reaction against NaNO₂ standard curve. Data represent mean values±S.E.M. for 3 independent experiments. ^**p < 0.01 versus PC12, Student’s t-test.

Arginases and Ass mRNAs are reduced while Asl is increased in APPsw cells

Besides NOS, arginine is also the substrate of arginases that hydrolyze it to urea and ornithine. This allows OTC to close the cycle by re-creating citrulline from ornithine. Our results show that the expression of arginases, Arg1 and Arg2 was lower in APPsw cells than in controls (Arg1 mRNA was reduced by 32%, p < 0.05; Arg2 by 40%, p < 0.01; Fig. 4A). Correspondingly, we observed that total activity of arginases was reduced by nearly 57% in APPsw cells (from 1.85 to 0.80 nmol/mg protein/min; Fig. 4B).

Fig.4

The effect of AβPP on the mRNA expression and activity of arginases. A) Total RNA was isolated from PC12 and APPsw cells, treated with DNase, and reverse-transcribed in equal amounts as stated in Materials and Methods. Real-time PCR results for Arg1 and Arg2 mRNAs ere expressed as relative quantity, Rq normalized against actin b expression. Data represent mean values±S.E.M. for ≥3 independent experiments. ^*p < 0.05; ^**p < 0.01 versus PC12, Student’s t-test. B) Arginase activity was determined spectrophotometrically against ornithine curve in extracts of control PC12 and APPsw cells in the presence of exogenously added arginine using ninhydrin reagent. The results are means±S.E.M. from 4-5 independent experiments. ^*p < 0.05 versus PC12, Student’s t-test.

NOS product citrulline is recycled to arginine by ASS and ASL. Ass mRNA was significantly decreased in APPsw cells (by 16%, p < 0.05; Fig. 5). However, Asl gene appeared to undergo higher expression in APPsw cells than in control cells (by about 44%, p = 0.01; Fig. 5). Ornithine decarboxylase (Odc), the rate-limiting enzyme in polyamine biosynthesis that decarboxylates ornithine to putrescine, was not changed in APPsw cells (data not shown).

Fig.5

mRNA levels of ASS and ASL in PC12 and APPsw cells. Total RNA was isolated from PC12 and APPsw cells, treated with DNase, and reverse-transcribed in equal amounts as stated in Materials and Methods. Real-time PCR results for argininosuccinate synthase-1 (Ass) and argininosuccinate lyase (Asl), were expressed as relative quantity, Rq normalized against actin b expression. Data represent mean values±S.E.M. for ≥3 independent experiments. ^*p < 0.05; ^**p < 0.01 versus PC12, Student’s t-test.

Citrulline, ornithine, and arginine are elevated in APPsw cells

Subsequently we investigated whether the changes in gene expression might have an impact on the levels of corresponding amino acid intermediates of the urea cycle. Citrulline, ornithine, and arginine were found to be significantly (p < 0.05) elevated in APPsw cells comparing to controls by 33%, 21%, and 42%, respectively, as measured using HPLC (Fig. 6). Tryptophan (used as the internal standard) was not significantly different between the cell lines (data not shown).

Fig.6

The levels of citrulline, ornithine, and arginine in PC12 and APPsw cells. The cellular levels of citrulline, ornithine, and arginine were measured in control PC12 and APPsw cells by HPLC with fluorescence detection after extraction and derivatization as described in Materials and Methods. The results are means±S.E.M. from ≥6 independent experiments. ^*p < 0.05 versus PC12, Student’s t-test.

APPsw cells are more sensitive to NO donor-induced death

APPsw cells display significantly compromised viability, as tested by the MTT reduction assay after 24 h of culture (p < 0.001 versus PC12, Fig. 7A). The alteration of cellular metabolism that probably favors NO production might also have significant impact on cellular viability in conditions of additional stress, which occurs in the AD brain. Therefore, we have assessed cellular viability in the presence of the NO donor DEANO at various time points. APPsw cells appeared to react stronger to the treatment with DEANO (0.1 mM). At 24 h in the presence of the donor, there was a significant difference in the DEANO effect between control and AβPP-overexpressing lines (p < 0.001, 67% surviving APPsw cells versus 89% PC12, Fig. 7B).

Fig.7

The influence of AβPP expression and treatment with DEANO on cell viability. A) The survival of PC12 and APPsw cells at 24 h time point. Cell viability was assessed using MTT reduction test. The results are means±S.E.M.; ^***p < 0.001 versus PC12, Student’s t-test. B) The effect of DEANO treatment for various amounts of time (as compared to untreated, i.e., ‘t = 0’ cells of each line). The results are means±S.E.M.; ^***p < 0.001 versus PC12 at the same time point, ANOVA with Tukey post-hoc test.

Global gene expression patterns in the hippocampal CA1 region of control versus age-matched AD brain gave comparable results to the APPsw studies. For example, ARG1, ARG2, and ASS exhibited downregulation, ENOS and the control markers β-actin and α-tubulin showed no significant change while ASL and NNOS exhibited significant upregulation (Fig. 8).

Fig.8

mRNA levels of arginases (ARG1, ARG2), ASS, ASL and nitric oxide synthases (ENOS, NNOS and controls (β-actin and α-tubulin) in control and sporadic AD brains. A) DNA array study – results displayed in standard cluster diagram format [26 –31]. B) Quantified in bar graph format; RNAs were isolated from samples of hippocampal CA1 of sporadic AD cases (PMI of 2.9 h or less), CDR between 0.5 and 3.0 and from control CA1 using DNA array (GeneChip) analysis as described in Materials and Methods. A dashed black horizontal line has been placed at +1 (upregulated gene expression) and -1 (downregulated gene expression) for ease of comparison; data represent mean values±S.E.M. for 3–5 independent experiments. ^*p < 0.05; ^**p < 0.01, Student’s t-test.

DISCUSSION

The potential involvement of NO and nitrosative stress in AD [6] including AβPP/Aβ processing [40] and the extensive links between NOS and the enzymes of the urea cycle have prompted us to analyze in vitro the influence of AβPP/Aβ on NO and arginine metabolism (Fig. 9). PC12 cell lines transfected with the wild-type AβPP (APPwt) [6] were used in our previous work. These cells also displayed higher NO synthesis, stemming from elevated nNOS and eNOS [23], resulting in higher nitrite levels observed in the cell culture medium (Fig. 3). The co-occurring downregulation of mRNAs coding for arginases (Arg1, Arg2) and the reduced arginase activity [23] have appeared to create a favorable environment for the activation of constitutive NO synthases (arginases compete for NOS substrate and can change NOS’ product range). Other urea cycle enzymes were also deregulated in APPwt cells [23].

Fig.9

Depiction of the potential contribution of AβPP and/or Aβ peptides to altered ARG1, ARG2, ASS, ASL or NNOS signaling in the APPsw (K670M/N671L) amyloid over-expressing murine model. Multiple known stressors including reactive oxygen and nitrogen species (ROS, RNS), tumor necrosis factor alpha (TNFα), interleukin 1-beta (IL-1β), hypoxia, age and/or AD stimulate tandem β- and γ-secretase of AβPP to generate Aβ peptides (small dark circles); Aβ peptides appear to decrease ARG1, ARG2, and ASS and increase ASL and NNOS abundance with no change in ENOS. These changes in turn are associated with increases in citrulline, ornithine and arginine and ROS- and RNS-mediated increases in inflammatory neurodegeneration. These or related mechanisms may operate (i) directly, via Aβ peptide signaling; and/or (ii) indirectly through pathogenic factor-stimulated changes in the expression of arginases or other urea cycle enzymes. These neuropathogenic signals further promote amyloid aggregation and inflammatory degeneration characteristic of age-related neurological diseases including AD that exhibit defective Aβ₄₂ clearance mechanisms and amyloidogenesis; excessive ROS and/or RNS signaling may drive a self-sustaining, self-reinforcing feedback loop; see text for further details.

The expression of APPsw bearing double “Swedish” K670M/N671L mutation in its β-cleavage site (which additionally enhances Aβ production about two-fold) leads to a further significant rise in NOS activity [6]. However, in our current work we find that only Nnos mRNA was elevated in APPsw cells (Fig. 3A). Literature data suggests that Aβ (rather than AβPP itself) might be involved in the deregulation of NO production, signaling and/or toxicity. NOS changes observed in cells overexpressing AβPP were found to be sensitive to secretase inhibition [24], which points to Aβ amount (rather than AβPP alone) as the decisive factor. Our results suggest that NO synthesis might be dependent on Aβ/AβPP on several levels. Possible pathways of the observed influence include ‘saturation’ of the Aβ-to-NOS effect in the presence of the much higher Aβ loads in APPsw cells. The elevated stress in APPsw cells reaching a certain threshold could also trigger new, efficient responses. An effect of the K670M/N671L mutation on physiological signaling mechanisms triggered by AβPP is also possible. The interplay of such mechanisms with the toxicity of Aβ could be responsible in part for the enormous variability of NO disturbances that is observed in AD and its numerous experimental models [6 , 23]. However, the changes of NOS expression we measured in human postmortem hippocampal CA1 tissue followed the trends observed in APPsw cell culture, as NNOS was upregulated while ENOS remained unchanged (Fig. 6).

Like previously in the APPwt cells [23], we found lower Arg1 and Arg2 mRNAs in the APPsw line (Fig. 4A), which resulted in reduced total arginase activity (Fig. 4B). Reduced expression was also observed in CA1 hippocampal tissue (Fig. 8). However, ornithine accumulated in the APPsw cell model, just like in the APPwt cell line (Fig. 6). It might probably arise due to the action of other ornithine-utilizing enzymes (e.g., engaged in glutamate/glutamine/proline metabolism [41]). Moreover, results have been published suggesting that their regulatory mechanisms might be linked to those of the urea cycle genes [42]; however, elaboration of this question falls outside the scope of the current work.

Ass mRNA reduction in APPsw cells was similar to that present in the APPwt line, and could be responsible for the observed moderate accumulation of its substrate citrulline, assuming that the enzyme’s specific activity is unchanged [23]. Again, the trend has been confirmed to occur also in the CA1 tissue samples.

A notable difference between the two lines was the significant ∼50% upregulation of Asl expression in APPsw cells. This change suggests a somewhat unique regulation mechanism, possibly driven by the Aβ-related stress reaching a certain threshold and triggering new, additional cellular responses. ASL takes part in the replenishment of the NOS substrate arginine through the re-cycling of citrulline (which we found to accumulate moderately in the APPsw-overexpressing cells). As ASL mRNA was also upregulated in the human hippocampus, the APPsw model appears to follow the changes of ornithine metabolism in CA1 more closely than APPwt cells used in our earlier work [23]. It also suggests that the unique behavior of Asl mRNA should be attributed to Aβ/AβPP levels rather than some specific influence of the ‘Swedish’ mutation on Aβ/AβPP function, as the examined tissues come from sporadic AD cases. The reduced Ass mRNA does not seem to impact the pathway significantly, although it is replicated in the hippocampal material.

Thus, the observed changes fit into the general trend towards pro-nitrosative, and most probably maladaptive changes that keep the levels of NOS substrate arginine from falling in our cell model. These metabolic disturbances have an impact of cellular viability, as shown by the reduced survival of APPsw cells treated with the nitric oxide donor DEANO. It also suggests how the stress-related mechanisms might operate in human AD brains. Previous works on the alterations of arginine metabolism in AD and its models have drawn a contradictory and inconclusive image. Results vary from insignificant reduction [43] to marked increase [44] in the concentrations of arginine, citrulline, and ornithine in the CSF of AD subjects [45, 46]. Arginase activity measured in the plasma dropped in AD [47] while intracerebroventricularly administered pre-aggregated Aβ_25–35 selectively increased hippocampal ARG-2 protein and total arginase activity. Like in our work, the results of Liu et al. [22] suggest an inverse relationship between NOS and arginases. However, in that case elevated ARG-2 protein level (and arginase activity) predominated in the AD brain regions analyzed (hippocampus, superior frontal gyrus and cerebellum). Rather surprisingly, no changes were shown for hippocampal or cerebellar arginine while it increased in the superior frontal gyrus; citrulline was unchanged while ornithine was dramatically reduced in all three anatomical regions. Thus, Liu et al. [22] hypothesized that ornithine transcarbamylase induction, possibly in response to the changes in the tissue environment in AD, might mask the influence of NOS and ARG-2 on amino acid levels in the brain regions analyzed; OTC is absent from our cells and from most healthy brain tissue. Liu et al. also noted high variation between patients and, importantly, in the postmortem delay. Thus, tissue handling appears to be especially critical factor for any quantitative assessment of arginine metabolism in the AD brains.

In our hands, the effect of double Swedish-mutated AβPP on enzyme expression appear to be similar to the effect of wild-type AβPP, with the exception of Asl mRNA which increases only in APPsw cells [6]. The levels of amino acid intermediates of the urea cycle are significantly affected as a result. The APPsw model seems to mostly replicate the changes in the human AD hippocampal CA1 studies, and the level of Aβ production seems to be the most important factor (as APPwt line follows the tissue changes less closely). There is a general trend in enzymatic changes that alters the cellular milieu in favor of excessive NO production, though particular processes may vary depending probably on stress intensity.

Footnotes

ACKNOWLEDGMENTS

This study was supported by the statutory budget (theme no. 5) of the Department of Cellular Signalling, Mossakowski Medical Research Centre, Polish Academy of Sciences, Warsaw, Poland and in part by NIH NIA AG038834, Bethesda MD, USA.

Authors’ disclosures available online ().

References

Goate

, Chartier-Harlin

, Mullan

, Brown

, Crawford

, Fidani

, Giuffra

, Haynes

, Irving

, James

(1991) Segregation of a missense mutation in the amyloid precursor protein gene with familial Alzheimer’s disease. Nature 349, 704–706.

Axelman

, Basun

, Winblad

, Lannfelt

(1994) A large Swedish family with Alzheimer’s disease with a codon 670/671 amyloid precursor protein mutation. A clinical and genealogical investigation. Arch Neurol 51, 1193–1197.

Lannfelt

, Bogdanovic

, Appelgren

, Axelman

, Lilius

, Hansson

, Schenk

, Hardy

, Winblad

(1994) Amyloid precursor protein mutation causes Alzheimer’s disease in a Swedish family. Neurosci Lett 168, 254–256.

Haass

, Lemere

, Capell

, Citron

, Seubert

, Schenk

, Lannfelt

, Selkoe

(1995) The Swedish mutation causes early-onset Alzheimer’s disease by beta-secretase cleavage within the secretory pathway. Nat Med 1, 1291–1296.

Johnston

, O’Neill

, Lannfelt

, Winblad

, Cowburn

(1994) The significance of the Swedish APP670/671 mutation for the development of Alzheimer’s disease amyloidosis. Neurochem Int 25, 73–80.

Chalimoniuk

, Stolecka

, Cakała

, Hauptmann

, Schulz

, Lipka

, Leuner

, Eckert

, Muller

, Strosznajder

(2007) Amyloid beta enhances cytosolic phospholipase A2 level and arachidonic acid release via nitric oxide in APP-transfected PC12 cells. Acta Biochim Pol 54, 611–623.

Ikarashi

, Harigaya

, Tomidokoro

, Kanai

, Ikeda

, Matsubara

, Kawarabayashi

, Kuribara

, Younkin

, Maruyama

, Shoji

(2004) Decreased level of brain acetylcholine and memory disturbance in APPsw mice. Neurobiol Aging 25, 483–490.

Kragh

, Nielsen

, Li

, Du

, Lin

, Schmidt

, Bøgh

, Holm

, Jakobsen

, Johansen

, Purup

, Bolund

, Vajta

, Jørgensen

(2009) Hemizygous minipigs produced by random gene insertion and handmade cloning express the Alzheimer’s disease-causing dominant mutation APPsw. Transgenic Res 18, 545–558.

Choi

JHK

, Kaur

, Mazzella

, Morales-Corraliza

, Levy

, Mathews

(2013) Early endosomal abnormalities and cholinergic neuron degeneration in amyloid-β protein precursor transgenic mice. J Alzheimers Dis 34, 691–700.

10.

Tajeddinn

, Persson

, Maioli

, Calvo-Garrido

, Parrado-Fernandez

, Yoshitake

, Kehr

, Francis

, Winblad

, Höglund

, Cedazo-Minguez

, Aarsland

(2015) 5-HT1B and other related serotonergic proteins are altered in APPswe mutation. Neurosci Lett 594, 137–143.

11.

Strosznajder

, Jeśko

, Strosznajder

(2000) Effect of amyloid beta peptide on poly(ADP-ribose) polymerase activity in adult and aged rat hippocampus. Acta Biochim Pol 47, 847–854.

12.

Staschewski

, Kulisch

, Albrecht

(2011) Different isoforms of nitric oxide synthase are involved in angiotensin-(1-7)-mediated plasticity changes in the amygdala in a gender-dependent manner. Neuroendocrinology 94, 191–199.

13.

Wiesinger

(2001) Arginine metabolism and the synthesis of nitric oxide in the nervous system. Prog Neurobiol 64, 365–391.

14.

Kim

, Bugaj

, Oh

, Bivalacqua

, Ryoo

, Soucy

, Santhanam

, Webb

, Camara

, Sikka

, Nyhan

, Shoukas

, Ilies

, Christianson

, Champion

, Berkowitz

(2009) Arginase inhibition restores NOS coupling and reverses endothelial dysfunction and vascular stiffness in old rats. J Appl Physiol 107, 1249–1257.

15.

Yang

, Ming

X-F

(2013) Arginase: The emerging therapeutic target for vascular oxidative stress and inflammation. Front Immunol 4, 149.

16.

, Iyer

, Kern

, Rodriguez

, Grody

, Cederbaum

(2001) Expression of arginase isozymes in mouse brain. J Neurosci Res 66, 406–422.

17.

Choi

, Park

, Ahn

, Lee

, Shin

(2012) Immunohistochemical study of arginase 1 and 2 in various tissues of rats. Acta Histochem 114, 487–494.

18.

Morrison

, Cao

, Kish

(1998) Ornithine decarboxylase in human brain: Influence of aging, regional distribution, and Alzheimer’s disease. J Neurochem 71, 288–294.

19.

Haas

, Storch-Hagenlocher

, Biessmann

, Wildemann

(2002) Inducible nitric oxide synthase and argininosuccinate synthetase: Co-induction in brain tissue of patients with Alzheimer’s dementia and following stimulation with beta-amyloid 1-42 in vitro. Neurosci Lett 322, 121–125.

20.

Colton

, Mott

, Sharpe

, Xu

, Van Nostrand

, Vitek

(2006) Expression profiles for macrophage alternative activation genes in AD and in mouse models of AD. J Neuroinflammation 3, 27.

21.

Liu

, Jing

, Collie

, Campbell

, Zhang

(2011) Pre-aggregated Aβ(25-35) alters arginine metabolism in the rat hippocampus and prefrontal cortex. Neuroscience 193, 269–282.

22.

Liu

, Fleete

, Jing

, Collie

, Curtis

, Waldvogel

, Faull

RLM

, Abraham

, Zhang

(2014) Altered arginine metabolism in Alzheimer’s disease brains. Neurobiol Aging 35, 1992–2003.

23.

Jęśko

, Wilkaniec

, Cieślik

, Hilgier

, Gąssowska

, Lukiw

, Adamczyk

(2016) Altered arginine metabolism in cells transfected with human wild-type beta amyloid precursor protein (βAPP). Curr Alzheimer Res 13, 1030–1039.

24.

Keil

, Bonert

, Marques

, Scherping

, Weyermann

, Strosznajder

, Müller-Spahn

, Haass

, Czech

, Pradier

, Müller

, Eckert

(2004) Amyloid beta-induced changes in nitric oxide production and mitochondrial activity lead to apoptosis. J Biol Chem 279, 50310–50320.

25.

Lukiw

, Cui

, Yuan

, Bhattacharjee

, Corkern

, Clement

, Kammerman

, Ball

, Zhao

, Sullivan

, Hill

(2010) Acyclovir or Aβ42 peptides attenuate HSV-1-induced miRNA-146a levels in human primary brain cells. Neuroreport 21, 922–927.

26.

, Cui

, Hill

, Bhattacharjee

, Zhao

, Lukiw

(2011) Increased expression of miRNA-146a in Alzheimer’s disease transgenic mouse models. Neurosci Lett 487, 94–98.

27.

Cui

, Li

, Zhao

, Bhattacharjee

, Lukiw

(2010) Differential regulation of interleukin-1 receptor-associated kinase-1 (IRAK-1) and IRAK-2 by microRNA-146a and NF-kappaB in stressed human astroglial cells and in Alzheimer disease. J Biol Chem 285, 38951–38960.

28.

Lukiw

, Zhao

, Cui

(2008) An NF-kappaB-sensitive micro RNA-146a-mediated inflammatory circuit in Alzheimer disease and in stressed human brain cells. J Biol Chem 283, 31315–31322.

29.

Cui

J-G

, Zhao

, Lukiw

(2005) Isolation of high spectral quality RNA using run-on gene transcription; application to gene expression profiling of human brain. Cell Mol Neurobiol 25, 789–794.

30.

Lukiw

, LeBlanc

, Carver

, McLachlan

, Bazan

(1998) Run-on gene transcription in human neocortical nuclei. Inhibition by nanomolar aluminum and implications for neurodegenerative disease. J Mol Neurosci 11, 67–78.

31.

Cui

J-G

, Hill

, Zhao

, Lukiw

(2007) Expression of inflammatory genes in the primary visual cortex of late-stage Alzheimer’s disease. Neuroreport 18, 115–119.

32.

Lukiw

, Bazan

(2010) Inflammatory, apoptotic, and survival gene signaling in Alzheimer’s disease. A review on the bioactivity of neuroprotectin D1 and apoptosis. Mol Neurobiol 42, 10–16.

33.

Lukiw

(2004) Gene expression profiling in fetal, aged, and Alzheimer hippocampus: A continuum of stress-related signaling. Neurochem Res 29, 1287–1297.

34.

Colangelo

, Schurr

, Ball

, Pelaez

, Bazan

, Lukiw

(2002) Gene expression profiling of 12633 genes in Alzheimer hippocampal CA1: Transcription and neurotrophic factor down-regulation and up-regulation of apoptotic and pro-inflammatory signaling. J Neurosci Res 70, 462–473.

35.

Pogue

, Dua

, Hill

, Lukiw

(2015) Progressive inflammatory pathology in the retina of aluminum-fed 5xFAD transgenic mice. J Inorg Biochem 152, 206–209.

36.

Lukiw

(2012) NF-κB-regulated micro RNAs (miRNAs) in primary human brain cells. Exp Neurol 235, 484–490.

37.

Adamczyk

, Czapski

, Kaźmierczak

, Strosznajder

(2009) Effect of N-methyl-D-aspartate (NMDA) receptor antagonists on alpha-synuclein-evoked neuronal nitric oxide synthase activation in the rat brain. Pharmacol Rep 61, 1078–1085.

38.

Liao

M-C

, Muratore

, Gierahn

, Sullivan

, Srikanth

, De Jager

, Love

, Young-Pearse

(2016) Single-cell detection of secreted Aβ and sAPPα from human IPSC-derived neurons and astrocytes. J Neurosci 36, 1730–1746.

39.

Jesko

, Okada

, Strosznajder

, Nakamura

(2014) Sphingosine kinases modulate the secretion of amyloid β precursor protein from SH-SY5Y neuroblastoma cells: The role of α-synuclein. Folia Neuropathol 52, 70–78.

40.

Austin

, Santhanam

, Katusic

(2010) Endothelial nitric oxide modulates expression and processing of amyloid precursor protein. Circ Res 107, 1498–1502.

41.

Yoneda

, Roberts

, Dietz

(1982) A new synaptosomal biosynthetic pathway of glutamate and GABA from ornithine and its negative feedback inhibition by GABA. J Neurochem 38, 1686–1694.

42.

Levillain

, Diaz

J-J

, Blanchard

, Déchaud

(2005) Testosterone down-regulates ornithine aminotransferase gene and up-regulates arginase II and ornithine decarboxylase genes for polyamines synthesis in the murine kidney. Endocrinology 146, 950–959.

43.

Fonteh

, Harrington

, Tsai

, Liao

, Harrington

(2007) Free amino acid and dipeptide changes in the body fluids from Alzheimer’s disease subjects. Amino Acids 32, 213–224.

44.

Kaiser

, Schoenknecht

, Kassner

, Hildebrandt

, Kinscherf

, Schroeder

(2010) Cerebrospinal fluid concentrations of functionally important amino acids and metabolic compounds in patients with mild cognitive impairment and Alzheimer’s disease. Neurodegener Dis 7, 251–259.

45.

Virarkar

, Alappat

, Bradford

, Awad

(2013) L-arginine and nitric oxide in CNS function and neurodegenerative diseases. Crit Rev Food Sci Nutr 53, 1157–1167.

46.

Whiteley

(2014) Arginine metabolising enzymes as targets against Alzheimer’s’ disease. Neurochem Int 67, 23–31.

47.

Vural

, Sirin

, Yilmaz

, Eren

, Delibas

(2009) The role of arginine-nitric oxide pathway in patients with Alzheimer disease. Biol Trace Elem Res 129, 58–64.