Transgenic Mice Overexpressing Human STIM2 and ORAI1 in Neurons Exhibit Changes in Behavior and Calcium Homeostasis but Show No Signs of Neurodegeneration

The maintenance of proper cytosolic Ca2+ level is crucial for neuronal survival, and dysregulation of Ca2+ homeostasis is found in a variety of neurological disorders, including Alzheimer’s disease. According to the “Ca2+ hypothesis of aging”, Ca2+ disturbances precede the onset of AD symptoms and lead to neurodegeneration. STIM and ORAI proteins are involved in neuronal physiological and pathological processes as essential components of the store-operated Ca2+ entry. Our previous data suggested that overexpression of STIM2 and ORAI1 might increase basal neuronal cytosolic Ca2+ level. We generated double transgenic mice overexpressing these two genes in neurons, expecting that the increased basal Ca2+ concentration will lead to premature neurodegeneration. We observed changes in Ca2+ homeostasis and electrophysiological properties in acute brain slices of STIM2/ORAI1 neurons. However, we did not observe any augmentation of neurodegenerative processes, as tested by Fluoro-Jade® C staining and assessment of amyloidogenesis. The battery of behavioral tests did not show any signs of accelerated aging. We conclude that changes of calcium homeostasis induced by overexpression of STIM2 and ORAI1 had no substantial adverse effects on neurons and did not lead to early neurodegeneration.


Introduction
Store-operated Ca 2+ entry (SOCE) is a canonical Ca 2+ entry pathway that is activated in non-excitable cells when the ER Ca 2+ is depleted. SOCE plays an essential role in the maintenance of cellular Ca 2+ homeostasis and regulates a variety of processes, such as Ca 2+ oscillations, gene expression, mitochondrial function, and apoptosis. We have shown that proteins of the SOCE pathway are expressed and functionally active also in neurons [1,2]. The Ca 2+ sensors that convey information about the Ca 2+ load of the ER lumen are Stromal Interaction Molecules (STIM1, STIM2). STIMs sense luminal Ca 2+ concentration via an EF hand Ca 2+ -binding domain. In response to store depletion, they rearrange into punctate structures close to the plasma membrane, where they activate store-operated channels (SOCs), including members of the ORAI family (ORAI1-3) of Ca 2+ -influx channels [3,4]. This results in Ca 2+ entry into the cell. maintenance of SOCE by controlling the redox-dependent inactivation of ORAI1 [32]. On the other hand, it was shown that STIM2 was downregulated in brains of AD patients, whereas its presence was critical for the maintenance of mushroom spines in mouse models of AD [33,34].
Based on our earlier results [2] we hypothesized that mice that overexpress STIM2 and ORAI1 in neurons might exhibit early signs of neurodegeneration because of the increased basal Ca 2+ levels. The mice with such features would confirm Khachaturian's hypothesis of sporadic AD [19,35] and be a suitable model to study mechanisms of this disorder. In this paper we describe a new transgenic mouse line overproducing STIM2/ORAI1 in neurons. Despite the lack of obvious signs of accelerated neurodegeneration, the mice exhibit some changes in behavior and their neurons show minor impairments in basal synaptic transmission. These results highlight the importance of SOCE machinery in neuronal cells.

Overexpression of ORAI1 and STIM2 in Neurons Leads to Altered Ca 2+ Response in a Modified Ca 2+ Addback Assay in CA1 Hippocampal Region
We generated double transgenic mice overproducing ORAI1 and STIM2 in neurons as described in Materials and Methods. Real-time quantitative PCR (qPCR) analysis was performed to check the expression levels of human transgenes in the brain of these mice and compared the level of transcripts with that of endogenous Stim2 and Orai1. The results show a robust overexpression of human STIM2 and ORAI1 in the cortex and hippocampus of mice of both sexes ( Figure 1A,B). Acute brain slices of hippocampi that were isolated from 5-week-old wildtype and transgenic STIM2/ORAI1 mice were analyzed by fluorescent imaging. Calcium measurements and Fura-2 acetomethyl ester (AM) indicator loading were performed as described before [36]. To assess Ca 2+ homeostasis in the STIM2/ORAI1 overproducing neurons, we used a Ca 2+ addback assay that was modified to enable the measurement of Ca 2+ release from the ER [37]. As shown in Figure 1C,F, stimulation by glutamate led to similar peaks in both variants tested. However, the signal decay following the pulse of glutamate tended to be slower in transgenic neurons ( Figure 1D). The resulting difference in Ca 2+ cytosolic concentration persisted during the subsequent application of a Ca 2+ chelating agent, ethylene glycol-bis(β-aminoethyl ether)-N,N,N ,N -tetraacetic acid (EGTA) ( Figure 1C,E). Also, a moderately increased influx of calcium ions upon Ca 2+ addback was observed in neurons overproducing ORAI1 and STIM2. Recently, we have found similar results in neurons of single transgenic line with overproduction of ORAI1 [36]. This suggests that incorporation of STIM2 did not significantly change the Ca 2+ dynamics of neurons overproducing ORAI1 channels.

Impairment of Basal Synaptic Transmission in Adult Female Mice Overexpressing STIM2 and ORAI1
To check basic electrophysiological properties of neurons that overproduced STIM2 and ORAI1, we recorded local field potentials in the CA3-CA1 hippocampal projection using acute brain slices. The comparison of input-output curves with those in wildtype animals revealed a modest decrease in basal synaptic transmission in slices that were isolated from transgenic females (Figure 2A, upper row). However, population spike amplitudes were virtually identical in both tested groups ( Figure 2B, upper row). Similarly, we did not detect significant changes in short-term plasticity (measured by paired-pulse ratio, Figure 2C, upper row). An identical set of experiments that was performed on age-matched male mice revealed no changes in the tested parameters ( Figure 2, lower row). Our results suggest that neuronal excitability is similar in STIM2/ORAI1 and wildtype hippocampi and that the minor impairments in synaptic function are specific to female mice. Following our observation of a modest impairment in synaptic transmission, we checked the phosphorylation state of GluR1, an AMPA receptor subunit, and found no major differences between the two genetic variants ( Figure 2D). The measurements were performed using Fura-2 AM dye that was loaded into the CA1 neurons of the acute hippocampal brain slices. About 20 pyramidal neurons per slice (n) were analyzed; 1-2 slices from one animal were analyzed. The total number of analyzed slices was equal to 9 and 12 for wildtype and transgenic variants, respectively. (D) Time-course of signal decay following stimulation by glutamate that was fit by a logarithmic function. Student's t-test was used to check statistical significance of the observed differences; p-values are provided above the respective charts. (E) Quantification of baseline F340/F380 values and following the application of Ca 2+ chelator, EGTA; at 13th and 25th min of the measurement, a.u. (F) Quantification of signal amplitudes that were observed following treatment by glutamate and cyclopiazonic acid (CPA), and Ca 2+ addition; arbitrary units (a.u.).

Figure 2.
Basic electrophysiological properties of hippocampal neurons measured by local field potential recordings from acute brain slices that were isolated from 11-month-old female and male wildtype and STIM2/ORAI1 mice, upper panel: females; lower panel: males. (A) Input-output curves reflecting basal synaptic transmission in CA3-CA1 projection recorded from stratum radiatum. (B) Input-output curves of population spike responses recorded from stratum pyramidale in CA1 region. (C) Paired-pulse ratios of fEPSP slopes measured at different interstimulus intervals. Student's t-test was used to check statistical significance of the observed differences; p-values are displayed above the respective charts. n = 16 wildtype and n = 19 STIM2/ORAI1 slices (females); n = 9 wildtype and STIM2/ORAI1 slices (males). (D) The level of GluR1 phosphorylation was analyzed by Western blot in tissue homogenates from wild type and transgenic mice of both sexes. Three mice per sex and genetic variant were analyzed.

Modest Changes in Behavior of STIM2/ORAI1 Animals
To check whether the observed alterations in Ca 2+ homeostasis and the moderate impairment in synaptic transmission translates into changes in behavior of adult mice (6-month-old), we performed a battery of tests. Female transgenic mice exhibited enhanced willingness to explore the arena in Open Field test, which could indicate increased exploratory drive or decreased anxiety ( Figure 3B). Transgenic males spent less time on the rotating cylinder in the RotaRod test ( Figure 3A,B). Also, male transgenic animals tended to spend more time in the open arms during the elevated-plus maze (EPM) test ( Figure 3C), which suggests lower anxiety. Interestingly, whereas cued and contextual learning abilities were comparable across sexes and genetic variants ( Figure 3D,E), male transgenic mice exhibited less freezing during the first 2 min of the cued conditioning protocol, before the stimulus was applied. This observation may correspond to the observed lower anxiety levels in the EPM test. In the Novel Object Recognition test, we observed differences in the discrimination of the old and novel objects in the case of male mice, but not in females ( Figure 3F). Altogether, our results suggest that overproduction of STIM2 and ORAI1 had a modest effect on the behavior of animals. Apparently, the observed changes in Ca 2+ homeostasis and electrophysiological properties did not have a significant impact on learning and memory of transgenic mice and did not trigger premature pathology.

No Features of Neurodegeneration in Tg(STIM2/ORAI1)Ibd Mice
Following the initial idea of the project, we wondered whether overexpression of STIM2 and ORAI1 genes would trigger accelerated neurodegenerative processes because of the expected changes in Ca 2+ homeostasis. To evaluate this hypothesis, we performed staining with the Fluoro-Jade ® C dye that specifically labels degenerating neurons ( Figure 4A). Additionally, potential deposition of amyloid plaques was checked by immunostaining ( Figure 4B). No differences between wildtypes and STIM2/ORAI1 mice were observed, which suggests that overproduction of these two proteins had no effect on neurodegenerative processes.

Discussion
Accumulating evidence suggests that STIMs and ORAIs, the key proteins responsible for SOCE, play an important role in neuronal physiology. To explore further the role of neuronal SOCE (nSOCE), we have recently created several transgenic mouse lines that overproduce SOCE proteins specifically in brain neurons. Using these mice, we have shown that the activities of ORAI1 and STIM1 may regulate basal synaptic transmission and synaptic plasticity [38,39]. Together with studies by other groups [40][41][42][43][44], our results show that SOCE proteins play a role in fundamental neuronal processes. The proper functioning of synaptic transmission and plasticity is in part conveyed through intricate control of Ca 2+ dynamics in neurons [45]. We have previously found that in cultured cortical neurons, the overproduction of STIM2 together with ORAI1 resulted in an increased basal cytoplasmic Ca 2+ concentration [2]. Along the lines of so-called 'Ca 2+ hypothesis of aging', this observation prompted us to create a transgenic mouse line with neuronal-specific overproduction of these two proteins. This could serve as a useful model to validate whether the expected elevated Ca 2+ level in neurons facilitated premature neurodegeneration, a hallmark of AD and other neurological disorders [19].
In contrast to our findings in cultured neurons [2], Ca 2+ imaging that was performed on hippocampal slices isolated from double transgenic STIM2/ORAI1 mice did not reveal significant changes in basal cytosolic Ca 2+ level ( Figure 1C,E). This may result from the differences between biological models that were used in the original and the current study (neuronal cultures [2] vs ex vivo tissue-this study) or from the existence of compensatory mechanisms that are often observed in transgenic animals [46]. Nevertheless, our Ca 2+ imaging experiments showed an altered Ca 2+ response in neurons following a strong physiological stimulus (the application of glutamate). The changes were akin to the ones that we have recently observed in single transgenic ORAI1 mice [36]-a slower relaxation of the signal following glutamate peak and the persistence of increased cytosolic Ca 2+ level in Ca 2+ chelating conditions. Based on the mechanism proposed by Chen Engerer at al [47], this supports our hypothesis that overproduction of ORAI1 may have altered the refilling of ER Ca 2+ stores. Interestingly, the co-expression of STIM2 together with ORAI1 seemed to have little effect on the observed changes in Ca 2+ homeostasis. Indeed, the previously described Ca 2+ dynamics in hippocampal neurons overexpressing STIM2 alone was similar to that of wild-type neurons, with the exception of elevated Ca 2+ influx in the 'Ca 2+ re-addition protocol' that is used to assess the magnitude of SOCE [36]. Field potential recordings have revealed minor impairments in basal synaptic transmission of double transgenic female STIM2/ORAI1 mice (Figure 2A). This change was absent in male STIM2/ORAI1 mice. In our transgenic model, STIM2/ORAI1 mice, the alterations in Ca 2+ homeostasis that we have observed did not correlate with detectable signs of neurodegeneration. Fluoro-Jade ® C staining (which specifically detects degenerating neurons) and immunostaining for Aβ revealed no detrimental changes in hippocampal neurons from these mice (Figure 4). Despite modest impairments in synaptic transmission, we did not observe any changes in the phosphorylation state of the Ser845 GluR1 subunit of the AMPA receptor ( Figure 2D). Likewise, transgenic animals performed similarly in most behavioral tests that assess cognitive skills. However, male animals of the transgenic lines were more willing to enter open arms during elevated plus maze test, which could be a sign of decreased anxiety. Also, transgenic males spent less time on the rotating cylinder in the RotaRod test. This may indicate problems with motor coordination. Moreover, male mice of STIM2/ORAI1 manifested altered recognition of the novel object. Together with our previous report [38], these findings provide further indications that some actions of SOCE proteins may be sex-dependent.
Despite the lack of expected premature neurodegeneration in the STIM2/ORAI1 mice, our novel model can be a useful tool for further studies of the role of SOCE proteins in neurons. This is especially relevant in the context of conflicting evidence of STIM2 actions in neuronal pathophysiology [10,31,33,[48][49][50][51][52], the involvement of ORAI1 in neurotransmission [53] and the growing body of evidence that links Ca 2+ homeostasis to AD-related neurodegeneration.

Animal Care
The mice were housed under standard conditions on a 12 h/12 h light/dark cycle with food and water available ad libitum.

Generation of FVB/NJ-Tg(STIM2/ORAI1)Ibd Transgenic Mice
A coding sequence of human STIM2 from pEX-CMV-SP-STIM2(15-746) plasmid (a gift from Tobias Meyer-Addgene plasmid # 18868) [54] was cloned to the Thy-1.2 expression cassette [55,56]. The STIM2 coding region and flanking Thy1.2 sequence were sequenced to confirm that no point mutations were introduced during the cloning steps. Next, the Thy1.2-STIM2 construct was digested with PvuI and NotI to excise plasmid sequences, and the linearized fragment was gel-purified and used for standard pronuclear microinjection [57] and injected into pronuclei of fertilized eggs that were derived from FVB/NJ. The procedure was performed in the Laboratory of Animal Models at the Nencki Institute of Experimental Biology (Warsaw, Poland). Quantitative Real-Time PCR confirmed specific expression of the transgene in the brain within the established FVB/NJ-Tg(STIM2)Ibd lines. Generation of the FVB/NJ-Tg(ORAI1)Ibd line was described elsewhere [38]. Double transgenic mice were generated by crossing homozygotes of male FVB/NJ-Tg(STIM2)Ibd with the female FVB/NJ-Tg(ORAI1)Ibd. To identify transgenes, genotyping was performed with the following primers: Thy_Forw_genotype 5 -TCTGAGTGGCAAAGGACCTTAGG-3 , ORAI1_Rev_genotype 5 -TGGTCCTGTAAGCGGGCAAAC-3 , STIM2_Rev_genotype 5 -GCTGCTT ATTCTGGCAACACTTGG-3 .

Immunohistochemistry
Fluoro-Jade ® C (staining of degenerating neurons) and Amylo-Glo ® RTD™ (staining of amyloid plaques in tissue sections) procedures were performed according to the manufacturer's protocols. Briefly, the brain sections were mounted on Super Frost Ultra Plus slides (Thermo Scientific, Hungary), dried at 50-60 • C on a slide warmer for at least 30 min to 1hr for good adhesion. Next, in case of Fluoro-Jade ® C staining, slides were immersed in a basic alcohol solution comprising 1% sodium hydroxide in 80% ethanol for 5 min. They were then transferred to a new Coplin jar containing freshly prepared 70% ethanol for 2 min and next rinsed for 2 min with 0.1M PB (phosphate buffer). Following incubation in 0.06% potassium permanganate solution for 10 min for fluorescent background blocking and contrast optimization, the slides were transferred for 10 min to a 0.0001% solution of the Fluoro-Jade ® C dissolved in 0.1% acetic acid. The slides were then rinsed three times in 0.1 M PB for 1 min, air dried on slide warmer at 50 • C for at least 5 min. Next, dry slides were cleared by brief immersion in xylen (POCH, Poland) and coverslipped with non-aqueous mounting medium Entellan ® new (Millipore, Germany). In case of Amylo-Glo ® RTD™ procedure, the dried slides were transferred into a 70% solution of ethanol for 5 min, then rinsed in 0.1 M PB for 2 min and incubated for 10 min in the prepared 1× Amylo-Glo ® RTD™ staining solution. Next the slides were transferred into 0.9% saline solution for 5 min, briefly rinsed in 0.1 M PB, air dried, dehydrated through a graded series of ethanol concentrations, cleared in xylene (POCH, Poland) and coverslipped with Entellan ® new (Millipore, Germany). Images were acquired with a 10× objective on Zeiss Airyscan detector. Three individuals per variant per sex were analyzed.

Behavioral Analysis
Behavioral experiments were performed with a group of 25 males (12 wild-types, 13 transgenes) and 27 females (12 wild-types, 15 transgenes, age PD180) according to the following protocol: Rotarod test (RR), open field test (OF), elevated plus maze (EPM), fear conditioning, and novel object recognition test (NOR).
The tests were described extensively elsewhere [38,39]. In brief, in the rotarod test the mice were forced to maintain their balance on a rotating rod (30 mm diameter) that was placed 20 cm above the apparatus floor. The time (latency) it took for the mouse to fall from the rod (0-40 rotations per minute (rpm)) in a 300 s session was recorded. In the open-field test, the animals were placed in the middle of a circular arena (64 cm diameter, 30 cm high walls) and their activity was recorded for 10 consecutive minutes. In EPM, testing was performed using an apparatus that consisted of two opposing open arms and two opposing closed arms, (both 35 cm × 5 cm). The open and closed arms were arranged perpendicular to each other, with a central platform at the intersection of the arms. Animal movements were recorded for 5 consecutive minutes.
In fear conditioning test, a speaker was used to deliver the acoustic conditioned stimulus (CS). Sensory stimuli were adjusted to generate two contexts (context A and context B). For context A, a house light that was mounted on the ceiling of the cage was illuminated, and the room lights remained on. For context B, the room and chamber house lights were turned off. The ventilation fans were turned off and a 60 W red light provided illumination. The animals underwent three phases of training: fear conditioning, cue testing, and context testing. For fear conditioning, the animals were placed in the conditioning chambers in context A. They were exposed to five tones followed by foot shocks, with a 60 s interval. Twenty-four hours after the conditioning session, the mice underwent cue testing in a novel context (context B). Two minutes after being placed in context B, a tone was delivered. Twenty-four hours after cue testing, the animals were subjected to context testing. They were placed in the same context as in the fear conditioning phase (context A) for 300 s. Fear-related behavior (i.e., freezing) in response to the tone and context stimuli during training and testing was assessed by an automated video-based analysis system [58].
The novel-object recognition test was conducted in boxes (40 cm × 40 cm × 30 cm) with even lighting conditions (70 lux). Two laboratory glass bottles (6 cm diameter, 8 cm height) were used as Objects A, and a door stopper (6 cm diameter, 7 cm height, round in shape, stainless steel) was used as Object B. On day 1 of the tests, a habituation trial was performed. The mice were allowed to explore the empty test box for 5 min. The learning trial was conducted twenty-four hours after the habituation trial. The animals were placed in the test box with two identical Objects A that were positioned in opposite corners. The mouse was allowed to explore both for 5 min before it was placed back into its home cage. After 24 h, the mouse was placed in the test box with familiar Object A and novel Object B. The mouse was allowed to explore the objects for 5 min. The videos were recorded and analyzed using a video-tracking system (EthoVision, Noldus) to extract the behavioral data.
The differences between the groups were evaluated using the nonparametric Mann-Whitney U test. The data are expressed mean ± SEM. Data preparation and analysis were performed using R software (https://www.r-project.org).
4.9. Brain Slice Preparation, Ca 2+ Imaging, and Electrophysiology 12 ± 1 month-old and P25 ± 5 mice were used for electrophysiological experiments and Ca 2+ imaging, respectively. Following cervical dislocation, hippocampi from the right hemisphere were retrieved and 350 µm thick sections were prepared according to the protective recovery method and as described previously [36]. For Ca 2+ imaging experiments, CA1 pyramidal layer was labeled with the use of Fura-2 AM indicator and a modified Ca 2+ -addback assay was performed as described before [36]. The concentration of glutamate, EGTA and CPA were 100 µM, 2.5 mM and 20 µM, respectively. The authors wish to note that the imaging experiments that are described in the present paper and in the paper [36] were performed simultaneously, but the results concerning the double transgenic STIM2/ORAI1 line were not included in the latter work. The wildtype controls are the same for both sets of data, therefore, they overlap with our previously published work.
Local field potential recordings were performed in a submerged chamber that was filled with artificial cerebrospinal fluid (aCSF) that contained, in mM: 126 NaCl, 2.6 KCl, 20 glucose, 1.25 NaH 2 PO 4 , 25 NaHCO 3 , 1.5 MgSO 4 , and 2.5 CaCl 2 . The pH was adjusted to 7.3-4 at room temperature with O 2 /CO 2 95%/5% (v/v) mixture (carbogen) and HCl, and the osmolality was 315 ± 5 mOsm/kg H 2 0. aCSF was carbogenated throughout the recordings and heated with an in-line heater (catalog no. 64-0102, controlled by a TC-324B temperature control unit, Warner Instruments) to maintain the temperature in the recording chamber at 25 ± 1 • C. The rate of solution flow was 7 mL/min. To obtain synaptic responses, Schaffer collaterals were stimulated with a bipolar electrode (catalog no. CBARC75, FHC, USA). Stimuli of increasing intensity (0-100 µA, every 12.5 µA) were delivered with an ISO-200 stimulator (Circlelabs) at a frequency of 0.1 Hz. Field excitatory postsynaptic potentials (fEPSPs) were recorded in the stratum radiatum and stratum pyramidale of the CA1 area of the hippocampus using standard Ag/AgCl electrodes that were connected to CV-7B headstages (Molecular Devices). Glass micropipettes that were mounted on the headstages were pulled from borosilicate glass (Warner Instruments, 0.86 mm inner diameter, 1.50 mm outer diameter) with the use of a horizontal puller (P-1000, Sutter Instruments). Resistance of the pipette tips was between 2 and 4 MΩ when filled with aCSF.
The signals were amplified with Multiclamp 700B amplifier and acquired at 20 kHz with a Digidata 1550B acquisition card (Molecular Devices) using a 3 kHz low-pass filter. Prior to analysis, the signals were filtered at 1 kHz. To calculate popspike activity, a built-in function of the AxoGraph program (developed by John Clements) was used. To analyze synaptic activity, initial linear part of the slope of fEPSP was calculated in Clampfit (Molecular Devices). Further data analyses were performed in Microsoft Excel and GraphPad Prism 5 (USA).

Conclusions
Together with our previously published study, our data show that overproduction of SOCE proteins influence neuronal activity and may exert sex-dependent effects. Despite the observed impairments in synaptic transmission in transgenic female STIM2/ORAI1 mice and minor behavioral alterations in transgenic male animals, we did not detect any signs of accelerated neurodegeneration.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.