Abstract

   Arterial calcification is a powerful predictor of both the events and
   mortality associated with cardiovascular diseases in chronic kidney
   disease (CKD) patients. GLS1 (glutaminase 1), a rate‐limiting enzyme
   catalyzing the conversion of glutamine to glutamate, is disordered in
   various cardiovascular diseases. However, the potential interplay
   between GLS1‐mediated glutamate production and arterial calcification
   remains poorly understood. Here, LC‐MS/MS analysis of CKD patients’
   samples shows an abnormally elevated activity of GLS1, reflected by the
   increased glutamate/glutamine ratio. Moreover, GLS1 activity is
   positively correlated with arterial calcification progression, and its
   expression is upregulated in calcified arteries. Treatment with GLS1
   inhibitors or knockdown of GLS1 alleviates osteogenic reprogramming. In
   contrast, glutamate administration boosts the development of arterial
   calcification. Mechanistically, GLS1 redundancy‐regulated glutamate
   superfluity stimulates the activation of N‐methyl‐d‐aspartate receptors
   (NMDAR), leading to Ca^2+ influx and extracellular regulated protein
   kinases (ERK) phosphorylation, followed by the nuclear translocation of
   β‐Catenin and acceleration of osteogenic reprogramming of vascular
   smooth muscle cells (VSMCs) in further. This research defines GLS1 as a
   key contributor to arterial calcification. Glutamate, a major product
   of GLS1‐mediated glutamine metabolism, exerts a deleterious effect on
   arterial calcification by activating NMDAR and subsequently triggering
   Ca^2+ influx, which in turn exacerbates β‐Catenin‐regulated osteogenic
   reprogramming in VSMCs.

   Keywords: arterial calcification, GLS1, glutamate, NMDAR, osteogenic
   reprogramming, β‐Catenin
     __________________________________________________________________

   GLS1 is a novel contributor to arterial calcification. GLS1‐catalyzed
   glutamate exerts the promoting effects on osteogenic reprogramming in
   arteries. NMDAR, a glutamate receptor, is also activated and
   overexpressed during arterial calcification. GLS1‐mediated redundancy
   in glutamate promotes calcium influx and subsequent ERK phosphorylation
   by activating NMDAR, leading to nuclear translocation of β‐catenin and
   thus progression of arterial calcification.

   graphic file with name ADVS-12-2414252-g001.jpg

1. Introduction

   Arterial calcification, defined as the ectopic deposition of calcium
   and phosphate crystals mainly composed of hydroxyapatite in arteries,
   is highly prevalent in patients with chronic kidney disease (CKD) and
   contributes to adverse clinical events.^[ [60]^1a,b ^] Cardiovascular
   disease is the main cause of death in CKD patients.^[ [61]^2 ^]
   Arterial calcification consequently remains an increased risk of
   all‐cause death and adverse cardiovascular outcomes, beyond traditional
   risk factors.^[ [62]^3a,b ^] Owing to a lack of systematic
   understanding of the pathological mechanisms, there is no clear and
   efficient anti‐calcifying strategy.

   Previous studies have linked arterial calcification to metabolic
   disorders,^[ [63]^4a,b,c ^] especially abnormal amino acid
   metabolism.^[ [64]^4 , [65]^5 ^] There has been a reported connection
   between CKD and altered glutamine metabolism. Studies demonstrated that
   CKD was related to higher levels of glutamine and glutamate in rat
   blood, possibly owing to disturbance nitrogen utilization and
   accelerated protein breakdown.^[ [66]^6a,b,c ^] The deamidation of
   glutamine to glutamate is catalyzed by glutaminase1 (GLS1). The
   metabolic process is called glutaminolysis, in which GLS1 serves as a
   rate‐limiting enzyme.^[ [67]^7a,b,c ^] The glutamate/glutamine ratio is
   widely used to reflect the activity of GLS1.^[ [68]^8 ^] Multiple
   studies have shown that GLS1 dysregulation participates in the
   pathogenesis of age‐associated disorders,^[ [69]^9 ^] pulmonary
   arterial hypertension (PAH),^[ [70]^10 ^] vascular stiffness,^[ [71]^11
   ^] heart failure,^[ [72]^12 ^] and calcific aortic valve disease.^[
   [73]^13 ^] In addition, GLS1 abnormality was also associated with the
   process of osteogenic differentiation.^[ [74]^14a,b ^] However, little
   is known about whether GLS1 has participated in the process of arterial
   calcification.

   As a pleiotropic molecular, glutamate binds to its receptors, including
   ionotropic receptors AMPAR
   (α‐amino‐3‐hydroxy‐5‐methyl‐4‐isoxazole‐propionate receptors), NMDAR
   (N‐methyl‐d‐aspartate receptors) and metabotropic receptors
   (metabotropic glutamate receptor, mGlu1‐8) to exert functions.^[
   [75]^15 ^] NMDAR is a glutamate‐gated, voltage‐dependent,
   heterotetrameric transmembrane channel, with subunits classically
   consisting of two GluN1 and two GluN2A or GluN2B.^[ [76]^16 ^]
   Glutamate‐NMDAR axis participates in bone resorption,^[ [77]^17 ^] and
   NMDAR activation is necessary for bone formation,^[ [78]^18 ^] which
   indicates that the glutamate‐NMDAR axis may be a vital regulator of
   osteogenic reprogramming. Because of the similar mechanisms of bone and
   vascular mineralization, including the increased expression of critical
   osteogenic markers such as Runt‐related transcription factor 2 (RUNX2),
   bone morphogenetic protein 2 (BMP2) and osteocalcin (OCN),^[ [79]^19a,b
   ^] we assumed that GLS1 and glutamate may be involved in the
   pathological process of arterial calcification.

   Wnt/β‐Catenin signaling has been widely studied for its involvement in
   arterial calcification. Stimulated by various factors, unphosphorylated
   β‐Catenin (an active form of β‐Catenin) accumulate in the cytoplasm and
   translocate to the nucleus.^[ [80]^20a,b ^] Nuclear β‐Catenin binds to
   T‐cell factor/lymphoid enhancer‐binding factor (TCF/LEF), functioning
   as a coactivator to drive the transcription of master osteogenic
   regulators, including RUNX2^[ [81]^21 ^] and msh homeobox 2 (MSX2),^[
   [82]^22 ^] and thus promoting arterial calcification. In this study, we
   investigated the detrimental role of GLS1 in arterial calcification.
   Using clinical samples from donors, we verified that GLS1 was increased
   in calcified arteries, and GLS1‐mediated metabolism was abnormally
   elevated in CKD patients. We identified that GLS1 reduction impeded
   VSMCs osteogenic reprogramming and subsequent arterial calcification,
   whereas glutamate supplementation exhibited the converse effect,
   accelerating these processes. Mechanistically, stimulation of NMDAR by
   glutamate‐mediated Ca^2+ influx and subsequent extracellular regulated
   protein kinases (ERK) activation, which promoted osteogenic
   reprogramming of VSMCs in a β‐Catenin dependent manner, ultimately
   leading to arterial calcification. These findings emphasized the
   harmful roles of GLS1 and glutamate in arterial calcification.

2. Results

2.1. GLS1 and Its Catalyzed Production Glutamate Are Positively Associated
with Arterial Calcification

   KEGG pathway enrichment analysis using the MetaboAnalyst database
   revealed that blood metabolites causally associated with arterial
   calcification are primarily involved in glutamate metabolism
   (Figure [83]S1A, Supporting Information). Clinical research found that
   CKD patients were prone to developing stroke,^[ [84]^23 ^] cognitive
   impairment,^[ [85]^24 ^] Alzheimer's disease,^[ [86]^25 ^]
   schizophrenia,^[ [87]^26 ^] and so on. Significantly, the comorbidities
   of the above‐mentioned diseases are all due to abnormal glutamate
   signaling. Arterial calcification, as the most common pathological
   process in CKD patients, was positively correlated with the occurrence
   of stroke.^[ [88]^27 ^] In our clinical study, the baseline
   characteristics of enrolled populations are shown in Table [89]S1
   (Supporting Information). Consistently, Self‐Rating Depression Scale
   (SDS) Score elevated with the increasing plasma glutamate (Glu) levels
   from CKD patients (Figure [90]S2A, Supporting Information), and there
   was also a positive correlation between SDS score and arterial
   calcification (Figure [91]S2B, Supporting Information). Consistently,
   Glu levels were upregulated in both CKD patients and AP mice (a CKD
   mouse model with high adenine and high phosphorus diet) (Figure [92]
   1A,B). GLS1‐mediated conversion of glutamine to glutamate is a primary
   pathway for glutamate production. Accordingly, glutamine (Gln) levels
   were also measured (Figure [93]1C,D). We found that the ratio of
   Glu/Gln, regarded as an index of GLS1 activity, was dramatically
   increased in both CKD patients and AP mice (Figure [94]1E,F). Further
   analysis exhibited that plasma Glu levels (Figure [95]1G) were
   positively associated with calcium Agaston score (evaluated by Siemens
   Syngo CT Workplace software), and plasma Gln levels were inversely
   correlated with calcium Agaston score (Figure [96]1H). More
   meaningfully, there was a positive correlation between plasma Glu/Gln
   ratios and calcium Agaston score (Figure [97]1I). Immunohistochemical
   staining data exhibited that GLS1 expression from patients was enhanced
   in calcified arteries than in noncalcified arteries (Figure [98]1J).
   Correlation analysis indicated that the positive area of GLS1
   expression was directly proportional to the positive area stained with
   Von Kossa (Figure [99]1K; Figure [100]S3A, Supporting Information).
   Additionally, we evaluated the expression of GLS1 in calcified arteries
   from mice administered with AP diet or subjected to Vitamin‐D[3]
   (VitD[3]) injection.^[ [101]^28a,b,c ^] Immunofluorescence staining and
   immunohistochemical staining from mice also showed the increased
   expression of GLS1 in calcified aortic (Figure [102]1L,M). GLS1 mRNA
   was obviously elevated in calcified arteries of mice (Figure [103]1N).
   The GLS1 protein levels were strikingly enhanced in calcified aortas,
   as evidenced by increased osteogenic markers including RUNX2 and OCN,
   along with decreased contractile property marker smooth muscle‐22α
   (SM22α) (Figure [104]1O,P; Figure [105]S3B,C, Supporting Information).
   Taken together, these findings implied that an upregulation of GLS1 was
   regarded as an essential clue to the development of arterial
   calcification.

Figure 1.

   Figure 1
   [106]Open in a new tab

   GLS1 (glutaminase 1) and its catalyzed production of glutamate are
   positively associated with arterial calcification. A) Plasma levels of
   Glu (glutamate) in patients with CKD (CKD, n = 65) and control donors
   (healthy, n = 58). B) Serum levels of Glu were determined in mice with
   either control (n = 12) or AP (high adenine and high phosphorus diet)
   (n = 15). C) Plasma levels of Gln (glutamine) in patients with CKD
   (CKD, n = 65) and control donors (healthy, n = 58). D) Serum levels of
   Gln were detected in control (n = 12) or AP (n = 15) mice. E) Plasma
   levels of Glu/Gln ratios in patients with CKD (CKD, n = 65) and control
   donors (healthy, n = 58). F) Analyzing serum levels of Glu/Gln ratio in
   control (n = 12) or AP (n = 15) mice. G–I) The correlation analysis
   between plasma levels of Glu, Gln, and Glu/Gln ratios and whole aortic
   Agatston scores in CKD patients (n = 53). J) Representative Von Kossa
   staining (upper) and immunohistochemical staining of GLS1 (lower) in
   arteries with or without calcification from patients (n = 5). Scale
   bar: 100 µm. K) Spearman's correlation analysis for relative expression
   of GLS1 and calcification lesion size (% of SMC layer) in patients with
   or without calcification (n = 5). L) Representative immunohistochemical
   images of GLS1 staining in control or VitD[3] (500 IU kg^−1, once a day
   for three consecutive days)‐induced arterial calcification (n = 3).
   Scale bar: 200µm. M) Representative immunofluorescence images in mice
   aortas from control or VitD[3]‐induced arterial calcification groups (n
   = 4). Scale bar: 200 µm. N) mRNA levels of GLS1 in aortas from control
   or VitD[3]‐induced arterial calcification (n = 4). O) The expression of
   GLS1 protein was confirmed from mice aortas in either the control or
   VitD[3]‐induced arterial calcification groups via representative
   immunoblots (n = 4). P) Representative immunoblots of GLS1 levels from
   mice aortas in control or AP‐induced arterial calcification groups (n =
   5). Data are shown as means ± SEM. Two‐tailed, unpaired Student's
   t‐test was used in (A) – (F) and (N); Spearman's rank correlation
   coefficient was used in (G), (H), (I), (K).

2.2. Inhibition of GLS1 Retards the Development of Arterial Calcification In
Vivo

   The expression of GLS1 was upregulated during arterial calcification,
   so it is interesting to determine whether GLS1 downregulation might
   alleviate the development of arterial calcification. A specific GLS1
   inhibitor (BPTES, 0.25 mg/20 g/200 µL, three times a week) was used to
   confirm the role of GLS1 in the progression of arterial calcification
   in vivo. Compared to negative controls, interference with GLS1
   obviously retarded arterial calcification, as determined by Alizarin
   Red staining (Figure [107] 2A). Calcium quantitative assay
   (Figure [108]2B) exhibited that BPTES reduced the calcium content of
   calcified arteries compared with VitD[3]‐induced arterial
   calcification. Consistently, Von Kossa staining (Figure [109]2C,D) also
   revealed that BPTES administration suppressed arterial calcification
   when compared to the VitD[3] or AP group. At the same time, serum
   alkaline phosphatase (ALP), blood urea nitrogen (BUN) and serum
   creatinine (CREA), factors known to be associated with increased
   occurrence of arterial calcification, were also improved in mice
   (Figure [110]2E–G). In western blot analysis, osteogenic markers
   (RUNX2, OCN) were substantially prompted during arterial calcification,
   while were alleviated after GLS1 inhibition (Figure [111]2H–K). Taken
   overall, these data indicated that GLS1 deficiency retarded arterial
   calcification.

Figure 2.

   Figure 2
   [112]Open in a new tab

   Inhibition of GLS1 retards the development of arterial calcification in
   vivo. A) Representative images of Alizarin Red staining from whole
   aortas in mice with VitD[3] (500 IU kg^−1)‐induced arterial
   calcification (n = 4). B) The calcium content of mice's aortas after
   injection with VitD[3] is analyzed quantitatively (n = 5). C,D)
   Representative images of Von Kossa staining from the aortic ring in
   mice (n = 3). Scale bar: 200 µm. E–G) Measurement of serum ALP, BUN,
   and CREA levels were performed on the indicated groups of mice (n =
   10). H,I) Representative immunoblotting and statistical diagram
   demonstrated that BPTES (GLS1 inhibitor, 0.25 mg/20 g/200 µL, three
   times a week) significantly retarded VitD[3]‐induced arterial
   calcification (n = 5). J,K) Immunoblotting analysis and statistical
   diagram showed that BPTES evidently attenuated CKD‐related arterial
   calcification (n = 5). Data are presented as means ± SEM. One‐way
   analysis of variance (ANOVA) with Bonferroni's test was used in (B),
   (E–G), and (K); or Dunnett's test was used in (I).

2.3. GLS1 Accelerates Osteogenic Reprogramming of VSMCs

   Osteogenic reprogramming of VSMCs is the key process during arterial
   calcification,^[ [113]^29 ^] so we further tested the role of GLS1 in
   osteogenic reprogramming of VSMCs in vitro. BPTES is an allosteric
   inhibitor of GLS1 that alters protein structure and may expose sites
   susceptible to degradation, resulting in changes in GLS1 protein
   levels.^[ [114]^13 , [115]^30 ^] Osteogenic reprogramming of VSMCs
   induced by sodium phosphate dibasic (Pi, 3.0 mm) was largely
   antagonized with BPTES treatment when compared with the negative
   control (Pi + DMSO) group by Alizarin Red staining (Figure [116] 3A;
   Figure [117]S4A, Supporting Information) and quantitative analysis of
   calcium content (Figure [118]3B). Western blot results showed that the
   osteogenic markers were induced by Pi but were reversed after GLS1
   inhibitor treatment (Figure [119]3C,D). CB‐839 behaved as a primarily
   noncompetitive inhibitor by inhibiting enzyme activity without
   affecting protein levels. It also exhibited the protective effect on
   VSMCs osteogenic reprogramming compared to the group of Pi+DMSO using
   Alizarin Red staining (Figure [120]3E; Figure [121]S4B, Supporting
   Information) and western blot (Figure [122]3F,G). Consistently, upon Pi
   stimulation, GLS1‐siRNA intervention (Figure [123]S4C,D, Supporting
   Information) also achieved the same anti‐calcification effect on VSMCs
   compared to siNC, as evidenced by calcium content assay
   (Figure [124]3H), Alizarin Red staining (Figure [125]3I;
   Figure [126]S4E, Supporting Information) and western blot analysis
   (Figure [127]3J,K). In contrast, as a result of GLS1 overexpression
   (Figure [128]S4F,G, Supporting Information), vascular mineralization
   was dramatically exacerbated (Figure [129]3L–O; Figure [130]S4H,
   Supporting Information). Collectively, these data represented that GLS1
   elevation acted as a promoting role in the osteogenic reprogramming of
   VSMCs during arterial calcification.

Figure 3.

   Figure 3
   [131]Open in a new tab

   GLS1 accelerates osteogenic reprogramming of VSMCs. A) Representative
   images of Alizarin Red staining from VSMCs incubated with sodium
   phosphate dibasic (Pi, 3.0 mm) and GLS1 inhibitor (BPTES) for 7 days (n
   = 3). B) Quantitative analysis of calcium content in Pi‐mediated
   osteogenic transdifferentiation of VSMCs cocultured with GLS1 inhibitor
   (BPTES) (n = 3). C, D) Representative immunoblots showed the effects of
   GLS1 inhibitor (BPTES) on VSMCs osteogenic transdifferentiation induced
   by Pi (n = 4). E) VSMCs treated with CB‐839 (1 µm) were stained for
   mineralization by Alizarin Red staining (n = 4). F, G) VSMCs suffered
   osteogenic transdifferentiation caused by Pi in the absence or presence
   of CB‐839, as shown in a representative western blot analysis and
   statistical diagram (n = 4). H) VSMCs were pretransfected with
   si‐negative control (siNC) or GLS1‐siRNA (si1, si2, si3), and then
   exposed to Pi for 7 days before the quantitative analysis of calcium
   content was conducted (n = 3). I) Pi‐mediated osteogenic
   transdifferentiation of VSMCs pretransfected with siNC or GLS1‐siRNA
   were observed through Alizarin Red staining (n = 3). J, K)
   Pretransfected with siNC or GLS1‐siRNA, VSMCs were treated with Pi for
   7 days, as depicted in a representative western analysis (n = 4). L)
   Representative images of Alizarin Red staining in Pi triggered
   osteogenic transdifferentiation of VSMCs preinfected with
   Lentivirus‐vector (LV‐vector) or Lentivirus‐GLS1 (LV‐GLS1) (n = 3). M)
   Quantitative analysis of calcium content in Pi led to osteogenic
   transdifferentiation of VSMCs preinfected with LV‐vector or LV‐GLS1 (n
   = 3). N, O) Representative immunoblots reflected the effects of GLS1
   overexpression on VSMCs osteogenic transdifferentiation with a
   treatment of Pi for 7 days (n = 5). Data are expressed as means ± SEM.
   One‐way ANOVA with Bonferroni's test was used in (B), (D), (G), (H),
   and (M); or Dunnett's test was used in (K) and (O).

2.4. GLS1‐Related Glutamate Accelerates Osteogenic Reprogramming

   Considering that GLS1 is defined as a rate‐limiting enzyme for
   glutamine metabolism, the above results reflected a strong link between
   GLS1‐regulated glutamine metabolism and arterial calcification.
   GLS1‐mediated glutamine metabolism catalyzes the production of various
   metabolites, mainly including glutamate, ammonia, α‐ketoglutarate
   (αKG), etc. Therefore, we incubated VSMCs in an osteogenic medium with
   a GLS1 inhibitor and supplemented the medium with several downstream
   GLS1‐catablized metabolites, including glutamate, NH[4]Cl (a supplement
   of ammonia), and DM‐αKG (a supplement of αKG). As shown in Figure [132]
   4A, only glutamate reversed the improved effects of VSMCs osteogenic
   transdifferentiation caused by BPTES. Furthermore, calcium deposition
   (Figure [133]4B) and western blot analysis exhibited a consistent trend
   (Figure [134]4C,D). To better confirm that GLS1‐regulated glutamate
   accelerated osteogenic reprogramming, mice injected with VitD[3] to
   induce arterial calcification were treated with 2% glutamate by gavage
   daily. As expected, glutamate supplements dramatically deteriorated the
   anti‐calcification effect of BPTES in vivo (Figure [135]4E). Aortas
   were also examined using western blot analysis to validate the
   promoting role of glutamate in GLS1‐mediated arterial calcification
   (Figure [136]4F,G), while glutamate did not affect the level of GLS1
   protein. Altogether, these findings suggested that GLS1‐related
   glutamate played a crucial role in aggravating the development of
   arterial calcification.

Figure 4.

   Figure 4
   [137]Open in a new tab

   GLS1‐related glutamate accelerates osteogenic reprogramming. A)
   Representative images of Alizarin Red staining with replenishment of
   GLS1‐mediated primary metabolites on the basis of treating with GLS1
   inhibitor (n = 3), (NH[4]Cl: a supplement of ammonia, DM‐ αKG: a
   supplement of αKG). Scale bar: 200 µm. B) Quantitative analysis of
   calcium content with replenishment of GLS1‐catalyzed primary
   metabolites based on treating with GLS1 inhibitor (n = 3). C,D)
   Representative immunoblots showed the effects of glutamate supplements
   on VSMCs osteogenic transdifferentiation in the presence of BPTES (n =
   5). E) Representative images of Alizarin Red staining from whole aortas
   (upper) and Von Kossa staining from aortic ring (lower) in mice that
   received VitD[3] injection (n = 4). Scale bar: 200 µm. F‐G)
   Representative immunoblotting showed the impact of glutamate
   replenishment on arterial calcification on the basis of exposing to
   BPTES in the aortas (n = 3). Data are shown as means ± SEM. One‐way
   ANOVA with Dunnett's test was used in (B), (D) and (G).

2.5. Glutamate Aggravates Osteogenic Reprogramming of VSMCs by Activating
NMDAR

   In regard to GLS1‐related glutamate‐promoted osteogenic reprogramming
   of VSMCs during arterial calcification, we explored the underlying
   mechanism by which glutamate worked. The binding of glutamate released
   into the extracellular environment to its corresponding receptors on
   the cell membrane is a classic mode of action for glutamate.^[ [138]^31
   ^] Therefore, we first detected the level of extracellular glutamate
   during osteogenic reprogramming of VSMCs. As shown in Figure [139] 5A,
   compared to the control group, an increase in extracellular glutamate
   levels is induced in VSMCs treated by Pi. More so, extracellular
   glutamate was further upregulated owing to overexpression of GLS1
   compared with Pi treatment group. Subsequently, a glutamate transporter
   inhibitor (DL‐TBOA) was employed, and the operation brought about a
   noteworthy accumulation in glutamate levels in VSMCs supernatant
   (Figure [140]5B). Consistently, overexpression of GLS1 exacerbated the
   calcification of VSMCs, and simultaneous treatment with DL‐TBOA further
   enhanced this pro‐calcification effect by GLS1, as evidenced by
   Alizarin Red staining (Figure [141]S5A, Supporting Information). All
   these results suggested that there are functional glutamate receptors
   for sensing extracellular glutamate and facilitating downstream
   signaling transduction to promote calcification. Traditionally,
   glutamate binds and activates to its receptor to exert functions.^[
   [142]^31 ^] Several antagonists targeting glutamate receptors were
   added to the osteogenic medium. Interestingly, only the NMDAR
   antagonist (MK801) reversed the accelerated function of
   GLS1‐overexpression‐induced VSMCs osteogenic reprogramming
   (Figure [143]5C–F), and the further experiments exhibited the
   anti‐calcification effect of MK801 in a dose‐dependent manner
   (Figure [144]5G–J). Notably, MK801 treatment did not affect the level
   of GLS1 protein. As a multimeric, the subunit of GluN1 determined the
   activity and availability of total NMDAR. As expected, mRNA
   (Figure [145]S5B, Supporting Information) and protein of GluN1
   (Figure [146]5K; Figure [147]S5C, Supporting Information) in trend were
   consistent with those of the osteogenic marker (RUNX2). Besides, GluN1
   expression was significantly elevated in calcified arteries from
   patients (Figure [148]5L; Figure [149]S5D,E, Supporting Information).
   This represented the first indication that the expression and activity
   of NMDAR, especially GluN1, were upregulated in arterial calcification,
   playing a worsening role in its progression. In a word, activation of
   NMDAR by GLS1‐related glutamate was indispensable for GLS1‐caused VSMCs
   osteogenic reprogramming.

Figure 5.

   Figure 5
   [150]Open in a new tab

   Glutamate aggravates osteogenic reprogramming of VSMCs by activating
   NMDAR. A) Detection of extracellular glutamate released into the
   supernatant of cultured VSMCs from control and Pi with or without GLS1
   overexpression (n = 3). B) Measurement of extracellular glutamate
   released into the supernatant of cultured VSMCs from Pi with
   overexpression of GLS1 in the absence or presence of DL‐TBOA (a
   glutamate transport inhibitor, 100 µm) for 7 days (n = 5). C,D) After
   overexpression of GLS1, VSMCs were incubated with Pi, and then exposed
   to glutamate receptor inhibitors ([151]LY341495: mGlu inhibitor, 10 µm;
   CNQX: AMPAR inhibitor, 10 mm; MK801: NMDAR inhibitor, 10 µm),
   respectively. The osteogenic reprogramming of VSMCs was investigated
   using Alizarin Red staining (n = 3) and Calcium quantitative analysis
   (n = 3). Scale bar: 200 µm. E,F) In the presence of GLS1
   overexpression, western analysis of RUNX2 was performed in VSMCs
   treated with Pi and 3 different glutamate receptor inhibitors (n = 3).
   G,H) Dose‐response relationship for the effect of MK801 on VSMCs with
   overexpression of GLS1 and Pi treatment was determined by calcium
   content assay (n = 3) and Alizarin Red staining (n = 3). Scale bar: 200
   µm. I,J) Western Blots analysis of VSMCs overexpressing GLS1 and
   treated with Pi, both in the absence and presence of MK801 (n = 4). K)
   The expression of GluN1 protein levels from VSMCs in the indicated
   group was measured by western blot assay (n = 5). L) Representative Von
   Kossa staining (upper) and immunohistochemical staining of GluN1
   (lower) in arteries with or without calcification from patients (n =
   5). Scale bar: 100 µm. Data are represented as means ± SEM. One‐way
   ANOVA with Bonferroni's test was used in (A), (B), (D), and (J), or
   Dunnett's test was used in (F) and (G).

2.6. Glutamate‐Activated NMDAR Induces Nuclear Translocation of β‐Catenin by
Mediating Ca^2+ influx

   Activation of NMDAR facilitates calcium influx.^[ [152]^32 ^] As
   predicted, BPTES did lead to a reduction in intracellular calcium
   (Figure [153] 6A). Further to that, glutamate supplements prevented
   BPTES‐antagonized calcium influx during osteogenic reprogramming of
   VSMCs, and MK801 completely abolished the aforementioned effect.
   Moreover, BAPTA‐AM (an intracellular calcium chelator) intervention
   reversed the intracellular Ca^2+ induced by the activation of the
   glutamate‐NMDAR axis. Ca^2+ acts as a secondary messenger to activate a
   series of kinases.^[ [154]^33 ^] Thus, the fluctuations of kinase were
   detected. Western blot analysis showed that, compared to Pi treatment
   alone, phosphate‐ERK (p‐ERK) protein levels in VSMCs were
   down‐regulated by BPTES treatment. Furthermore, on the basis of BPTES
   treatment, these levels were re‐stimulated by the addition of glutamate
   and further inhibited by MK801 (Figure [155]6B,C). Surprisingly,
   changes of p‐ERK levels across different treatments were the most
   significant among the series of kinase detected (Figure [156]S6A,B,
   Supporting Information). Notably, ERK inhibitors (U0126 and SCH772984)
   also attenuated the effect of osteogenic reprogramming of VSMCs with
   GLS1 overexpression (Figure [157]6D,E; Figure [158]S6C,D, Supporting
   Information). It concluded that GLS1‐related glutamate promoted
   osteogenic reprogramming of VSMCs by activating NMDAR mediated‐Ca^2+
   influx, and further facilitated activation of ERK. It is well
   established that activated ERK is associated with the stability of
   β‐Catenin, thereby promoting its nuclear translocation.^[ [159]^34a,b
   ^] Stimulated by various factors, β‐catenin on the cell membrane is
   released to the cytoplasm, where its active form (nonphosphorylated
   β‐Catenin) is stabilized and translocated to the nucleus, functioning
   as a transcriptional activator of downstream genes such as RUNX2 and
   MSX2.^[ [160]^20 , [161]^21 , [162]^22 , [163]^35 ^] Accordingly, we
   investigated the change of active β‐Catenin protein. It showed that
   BPTES administration markedly reduced the level of active β‐Catenin
   protein, whereas additional glutamate supplement promoted the increased
   expression of its active form (Figure [164]6F,G). Furthermore, we
   conducted the nuclear/cytoplasmic fractionation experiment to examine
   the subcellular distribution of β‐Catenin and observed the increased
   nuclear active β‐Catenin in VSMCs treated with Pi and incubated with
   BPTES as well as glutamate, compared to Pi and BPTES treatment only
   (Figure [165]S7A,B, Supporting Information). Similarly, after the
   glutamate supplement, increased active β‐Catenin protein was also
   observed in VSMCs cocultured with Pi and CB‐839 (Figure [166]S7C,D,
   Supporting Information). Then, MK801 was employed to test whether
   β‐Catenin was regulated by the related glutamate‐NMDAR axis. As
   predicted, MK801 was resistant to the upregulation of active β‐Catenin
   protein (Figure [167]S8A,B, Supporting Information). Concurrently, ERK
   inhibitors displayed the same trend as MK801 (Figure [168]S9A,B,
   Supporting Information). As shown in immunofluorescence images
   (Figure [169]6H), a low level of active β‐Catenin was observed in
   untreated cells. Under Pi treatment, the level of active β‐Catenin was
   upregulated in both cytoplasm and nucleus, especially in the nucleus.
   With the overexpression of GLS1, active β‐Catenin levels in nucleus
   were further increased compared to Pi group, indicating the increased
   nuclear translocation promoted by GLS1. In contrast, ERK inhibitor
   counteracted the promoting effect of GLS1 on active β‐Catenin levels in
   the presence of Pi treatment. In addition, the effect of overexpression
   GLS1 in VSMCs was alleviated by silencing β‐Catenin (Figure [170]6I;
   Figure [171]S10A–C, Supporting Information). Accordingly, GLS1 triggers
   osteogenic reprogramming of VSMCs in a β‐Catenin manner.

Figure 6.

   Figure 6
   [172]Open in a new tab

   Glutamate‐activated NMDAR induces nuclear translocation of β‐Catenin by
   mediating Ca^2+ influx. A) Measurement of intracellular Ca^2+
   concentration in VASMCs after the indicated treatment (n = 5). B,C)
   Representative immunoblotting and statistical diagram depicted the
   levels of p‐ERK proteins following the indicated treatment (n = 5). D)
   Alizarin Red staining was conducted to observe the osteogenic
   reprogramming of VSMCs treated with Pi and overexpressing GLS1 in the
   absence or presence of ERK inhibitors (U0126, 10 µm; SCH772984, 10 µm)
   in vitro (n = 5). Scale bar: 200 µm. E) Representative western blotting
   showed the effects of ERK inhibitors (U0126 and SCH772984) on
   osteogenic reprogramming of VSMCs treated with Pi and overexpressing
   GLS1 (n = 5). F,G) Western Blot analysis of active β‐Catenin protein
   expression from aortas in mice following the indicated treatment (n =
   4). H) Changes in nuclear translocation of active β‐Catenin were shown
   by representative immunofluorescence images (n = 4). Scale bar: 20 µm.
   I) Representative immunoblotting of active β‐Catenin protein levels
   with indicated treatment (n = 4). All values are expressed as means ±
   SEM. One‐way ANOVA with Bonferroni's test was used in (A), and
   Dunnett's test was used in (C) and (G).

3. Discussion

   In this present study, we demonstrated a previously unidentified role
   of GLS1 in the development of arterial calcification. Above all, the
   increased plasma ratio of Glu/Gln, a reflection of GLS1 activity, was
   first recognized in CKD patients compared with healthy controls. More
   so, plasma levels of Glu and Glu/Gln ratio were positively correlated
   with the progression of arterial calcification. Further experiments
   represented that inhibition or knockdown of GLS1 alleviated VSMCs
   osteogenic reprogramming and the subsequent arterial calcification,
   which were reversed by glutamate supplementation in vivo and in vitro.
   Mechanistically, GLS1‐mediated glutamate activated NMDAR, an
   unidentified signaling pathway in the field of arterial calcification,
   to promote Ca^2+ influx and consequent ERK activation, leading to
   arterial calcification by promoting nuclear translocation of β‐Catenin.
   A schematic in Figure [173] 7 illustrates GLS1‐catalyzed glutamate
   production in promoting arterial calcification.

Figure 7.

   Figure 7
   [174]Open in a new tab

   Schematic illustration of GLS1‐catalyzed glutamate production in
   promoting arterial calcification via activating NMDAR/Ca^2+/β‐Catenin
   pathway. GLS1 as a novel contributor to arterial calcification.
   GLS1‐catalyzed glutamate exerts the promoting effects on osteogenic
   reprogramming in arteries. NMDAR, a glutamate receptor, is also
   activated and overexpressed in arterial mineralization. GLS1
   redundancy‐mediated glutamate superfluity promotes calcium influx and
   subsequent ERK phosphorylation by activating NMDAR, leading to nuclear
   translocation of β‐Catenin and thus progression of arterial
   calcification.

   GLS1, the first rate‐controlling enzyme in the glutaminolysis pathway,
   converts glutamine to glutamate to yield multiple metabolites for
   anabolism in various cells.^[ [175]^36 ^] GLS1‐regulated glutamine
   metabolism has been shown to nourish various cardiovascular diseases.
   GLS1 inhibitor (CB‐839) was effective in reducing vascular collagen
   content, vascular stiffness, and proliferation, as well as histologic
   and hemodynamic parameters of pulmonary hypertension.^[ [176]^37 ^]
   Furthermore, BPTES partially prevented aortic valve thickness,
   fibrosis, and calcification.^[ [177]^13 ^] Inhibition of glutaminolysis
   (CB‐839) in activated myofibroblasts, protects against cardiac
   functional decline and fibrosis.^[ [178]^12 ^] GLS1‐related
   glutaminolysis aggravates age‐associated disorder. BPTES administration
   ameliorated age‐associated glomerulosclerosis, kidney dysfunction, lung
   fibrosis, SA–β‐gal–positive population, and improved age‐associated
   weakness of grip strength and short‐hanging endurance in aged mice.^[
   [179]^9 ^] In addition, glutaminolysis‐related productions were
   reported to promote osteogenic differentiation of mesenchymal stem
   cells.^[ [180]^14 , [181]^38 ^] Given the above situation, whether
   glutaminolysis is involved in the pathogenesis of arterial
   calcification attracts our attention. Therefore, assessed by the level
   of plasma Glu/Gln ratio, the GLS1 activity was higher in CKD patients
   than in healthy controls for the first time. Furthermore, the plasma
   Glu/Gln ratio was positively associated with aortic Agaston calcium
   score in CKD patients. It was consistent with a universal upregulation
   in GLS1 expression in the arteries from calcified patients or calcified
   mice. To further clarify the role of GLS1 in arterial calcification,
   GLS1 inhibitor administration has been proven to significantly impede
   arterial calcification. These results indicate the effect of VSMCs GLS1
   abundance on calcific vasculopathy. It is suggested that targeting GLS1
   serves as a new insight into antagonistic arterial calcification.

   GLS1‐regulated glutaminolysis derives an abundance of metabolic
   products, the primary metabolite such as glutamate, ammonium, and
   αKG.^[ [182]^39 ^] Recent studies have reported that ammonium derived
   from GLS1‐driven glutaminolysis was responsible for age‐associated
   disorder. Enhanced GLS1‐mediated glutaminolysis induced ammonia
   production, which neutralized the lower pH and improved the survival of
   the senescent cells.^[ [183]^9 ^] Additional research illustrated that
   glutaminolysis‐related glutamate derivatives including αKG, proline,
   and asparagine promoted the proliferation of MSCs and osteogenic
   differentiation.^[ [184]^14a,b ^] A study has revealed that
   GLS1‐mediated glutamate was associated with the inflammation in the
   development of psoriasis.^[ [185]^40 ^] In our research, we conducted
   incubated experiments using the original and primary metabolites to
   conclude that only GLS1‐induced glutamate was the culprit of VSMC
   osteogenic reprogramming. Conversely, overexpression of GLS1‐induced
   glutamate sufficiency pushed VSMCs osteogenic reprogramming and
   consequence arterial calcification. Moreover, glutamate administration
   in vivo also showed that glutamate was the initiator of
   glutaminolysis‐facilitated arterial calcification, which was consistent
   with previous research.^[ [186]^18 ^] The reason why different
   metabolites catalyzed by GLS1 play different roles may be due to
   different cell types and stimulation conditions. It is suggested that
   under the state of GLS1‐mediated glutamine metabolism, the same amount
   of glutamate is more harmful than ammonia, which has drawn our
   attention. Further experiments are needed to verify the deeper work.
   Given the above evidences, glutamate as a major component of a
   ginomoto, may shed light on intake of ginomoto in daily diet.

   Glutamate is a ligand, which is decoded by an array of glutamate
   receptors to realize diverse functions. To investigate the receptor of
   responsibility without bias, multiple inhibitors for glutamate
   receptors were employed. Interestingly, only treatment with MK801, an
   NMDAR inhibitor, showed the anti‐calcification effect despite increased
   glutamate mediated by GLS1 overexpression, indicating that the
   glutamate‐NMDAR axis was responsible for GLS1‐related glutamate evoking
   arterial calcification. Hinoi E et al. found that sustained exposure to
   the NMDAR antagonist (MK801) significantly impeded osteogenesis in
   osteoblast.^[ [187]^18 ^] NMDAR is assembled as tetramers, and
   modulation of GluN1 plays a key role in controlling the total NMDAR
   pool available for activation.^[ [188]^41 ^] Dumas et al. revealed that
   NMDAR exists in human pulmonary arterial smooth muscle cells, of which
   GluN1 overexpressed and overactivated to stimulate the development of
   PAH.^[ [189]^42 ^] Consistent with this, we also detected an increase
   in GluN1 protein during arterial calcification. In conclusion, these
   results displayed a decisive role of the glutamate‐NMDAR axis in
   GLS1‐mediated arterial calcification.

   NMDAR activation evokes Ca^2+ influx,^[ [190]^31 , [191]^32 , [192]^41
   ^] and we also found the increased Ca^2+ influx with overexpression of
   GLS1. Thus, several Ca^2+‐dependent protein kinases were further
   detected. Interestingly, only activated ERK was significantly detected,
   while inhibiting ERK attenuated the effect of GLS1 on arterial
   calcification, indicating that ERK served as the bridge of GLS1‐related
   glutamate/NMDAR axis‐induced arterial calcification. Activated ERK
   promotes the stability of active β‐Catenin and its translocation to
   nucleus.^[ [193]^34a ^] As presented in our immunofluorescence results,
   accumulation of active β‐Catenin was transported to nucleus. Therefore,
   β‐Catenin was also confirmed as a major role of GLS1 mediated arterial
   calcification. This effect may be related to the binding of TCF/LEF
   with β‐Catenin in nucleus, and the subsequent stimulation of the
   transcription of RUNX2.

   The present study shows that GLS1 is a deteriorating regulator in
   arterial calcification through glutamate products. It is worth noting
   that the activation and upregulation of GluN1 in arterial calcification
   are also reported for the first time. However, the regulatory mechanism
   between GluN1 and arterial calcification is yet unclear. In addition,
   potential confounding factors of the enrolled population, such as the
   patient's diet and basic metabolic condition, were not controlled in
   this study. Further clinical trials are needed to address these
   confounding factors and reduce the impact on the study outcomes.

   In conclusion, clinical investigation emphasized the correlation of
   increased plasma glutamate and GLS1 activity with the development of
   arterial calcification. We identified the initiating factor of GLS1 in
   promoting the process of arterial calcification. Further speaking,
   GLS1‐related glutamate, NMDAR activation that was verified as an
   unidentified regulatory molecule in vascular calcification, accelerated
   the osteogenic reprogramming of VSMCs and induced arterial
   calcification progression in a β‐Catenin manner. These findings provide
   novel perspectives on the prevention and inhibition of clinical
   arterial calcification.

4. Experimental Section

Clinical Study Population and Human Artery Samples

   This cross‐section clinical study was conducted from July 2023 to
   January 2024 at The Eighth Affiliated Hospital of Sun Yat‐sen
   University, Shenzhen, China. A total of 123 participants (58 healthy
   subjects and 65 CKD patients) were enrolled, and their plasma samples
   were detected by LC‐MS/MS. Among CKD patients, 12 individuals lacked
   chest CT data. For the remaining 53 patients with available CT data,
   the statistical analysis of calcification scores was performed using
   the calcium Agaston score. The diagnostic criterion of CKD^[ [194]^43
   ^] was based on eGFR decline persisting for 3 months or longer, and
   specific manifestation of eGFR less than 60 mL min^−1/1.73 m^2.
   Patients were excluded if they had acute kidney disease (AKD), elevated
   liver enzymes, acute cardiovascular diseases (atrial fibrillation,
   myocardial infarction, unstable angina, heart failure, and stroke),
   malignancies, acute infectious diseases, history of taking
   α‐ketoglutarate or phosphate binders, vitamin D analogs, or the
   corticosteroids within 3 months before recruitment, were currently
   pregnant or wished to get pregnant throughout the study. Healthy
   participants were matched age and gender, and excluded if they had CKD,
   AKD, elevated liver enzymes, acute cardiovascular diseases (atrial
   fibrillation, myocardial infarction, unstable angina, heart failure,
   and stroke), malignancies, acute infectious diseases, history of taking
   α‐ketoglutarate or corticosteroids within 3 months before recruitment.
   Artery samples were originated from patients who underwent an operation
   of renal cell carcinoma. The clinical trial protocol and human samples
   collection were performed with the Declaration of Helsinki and approved
   by the Ethics Committee of The Eighth Affiliated Hospital of Sun
   Yat‐sen University (2023‐100‐01). Informed Consent Form have been
   obtained from all participants or legal representatives.

Assessment of Calcium Score

   Among the cohort, CKD patients received a chest CT scan and calculated
   Agatston scores to quantify the severity of aortic calcification. The
   CT images were reconstructed using slices of 1 mm thickness to evaluate
   aortic calcification with Siemens Syngo CT Workplace software.
   Calcification score is calculated using the formula based on the weight
   coefficients corresponding to CT values.^[ [195]^44 ^] More details
   have been described in previous research.^[ [196]^45 ^] Arterial
   calcification was identified as an Agatston score or Volume score
   greater than 0.

LC‐MS/MS Measurement

   Peripheral venous blood sample was collected from participants for at
   least 8 h of fasting. The operation procedure of plasma was adhered to
   a centrifugal speed of 3000×g for 15 min at 4 °C, followed by 5000×g
   for 10 min, and plasmas were accumulated for subsequent research or
   stored at −80 °C until analysis. LC‐MS/MS detection for samples was
   operated by Wuhan Metware Biotechnology Co., Ltd.
   ([197]www.metware.cn). The final results were provided. The operating
   parameters of LC‐MS/MS summarized are listed in Table [198] 1 .

Table 1.

   Operating parameters of LC‐MS/MS.
   Parameter Amide
   Liquid Chromatography LC‐ESI‐MS/MS system (UPLC, ExionLC AD,
   [199]https://sciex.com.cn/)
   Column Waters ACQUITY UPLC BEH Amide (2.1 × 100 mm, i.d.1.7 µm)
   Temperature 40 °C
   Injection Volume 2 µL
   Flow Rate 0.4 mL min^−1
   Mobile Phase A Water with 10 mm Ammonium acetate and 0.3% Ammonium
   hydroxide
   Mobile Phase B 90% acetonitrile/water (V/V)
   Gradient The gradient was started at 95% B (0–1.2 min), decreased to
   70% B (8 min), 50% B (9–11 min), finally ramped back to 95% B
   (11.1–15 min)
   Mass Spectrometry AB Sciex QTRAP 6500+ System,
   ([200]https://sciex.com/)
   Ion Source ESI±
   Source Temperature 550 °C
   Ionspray Voltage Positive, 5500 V; Negative, −4500 V
   Ion Source Gas1 45 psi
   Ion Source Gas2 55 psi
   Curtain Gas 35 psi
   Scan Type Scheduled multiple reaction monitoring
   [201]Open in a new tab

Animal Experiments

   All animal experiments abided by animal ethical standards and were
   approved by the Ethics Committee of Zhongshan School of Medicine, Sun
   Yat‐sen University (SYSU‐ACUC‐2022‐B0053). 8‐week‐old C57BL/6J male
   mice weighing 20–25 g were purchased from the Laboratory Animal Center
   of Sun Yat‐Sen University. Mice were housed in a temperature (23–25 °C)
   and humidity‐controlled colony room, maintained on a 12‐h light/dark
   cycle (08:00–20:00 light on). Mice were randomly divided into each
   group (n = 15). For CKD‐related‐arterial calcification, mice were
   exposed to a high adenine (0.15%) and high phosphorus (1.9%) diet
   (GDMLAC, AP diet) for 12 w. Alternatively, Vitamin D[3] (Sangon
   Biotech, 500 IU kg^−1) was intraperitoneally injected into mice
   continuously for 3d to induce arterial calcification. Mice were
   maintained on a standard chow diet until sacrifice on day 10.
   Meanwhile, On the first day of inducing arterial calcification, mice
   were intraperitoneally injected with BPTES (Selleck, 0.25 mg/20
   g/200 µL) or vehicle control (200 µL of 10% DMSO in corn oil) three
   times a week until they were euthanized. In addition, glutamate
   treatment was given a daily gavage of 2% glutamate (Aladdin) dissolved
   in water. Animals were anesthetized with isoflurane and then euthanized
   by cervical dislocation under deep anesthesia.

Assessment of Mice Serum Glutamate and Glutamine

   Mice serum was centrifuged and collected for measurement of glutamate
   and glutamine using a Yellow Spring Instrument 2950D‐1 (YSI Inc./Xylem
   Inc., Yellow Springs).

Cell Culture

   Primary human VSMCs were purchased from the American Type Culture
   Collection (ATCC) and cultured in high‐glucose Dulbecco's modified
   Eagle's medium (DMEM, Gibco) containing 10% fetal bovine serum (FBS,
   MIKX), 1% penicillin/streptomycin (P/S, Gibco). VSMCs at passages 3–10
   were handled in all cell experiments. To induce VSMCs calcification,
   3.0 mm sodium phosphate dibasic (Pi, Sigma–Aldrich) was added to the
   cellular culture medium for 7d. The medium was changed every 2 or 3
   days.

   VSMCs were transfected with small interfering RNA (siRNA), GLS1 siRNA,
   or β‐ Catenin siRNA according to a direction of lipofectamine 3000
   regent (Invitrogen). The sequences of siRNA are listed in Table [202]S2
   (Supporting Information). For overexpression, VSMCs were infected with
   lentiviral vector and lentiviral GLS1 following the manufacturer's
   instructions for lentivirus usage. The efficiencies of transfection or
   infection were identified by real‐time polymerase chain reaction (PCR)
   or Western blot. Treatments were carried out simultaneously with Pi
   stimulation: BPTES (Selleck); CB‐839 (Selleck, 1 µm); MK801 (APExBIO,
   10 µm); CNQX (APExBIO, 10 µm); [203]LY341495 (APExBIO, 10 µm);
   SCH772984 (APExBIO, 10 µm); U0126‐EtOH (APExBIO, 10 µm); BAPTA‐AM
   (Selleck, 5 mm); DL‐TBOA (MCE, 100 µm) was selected as the working
   concentration.

Western Blot

   Tissue or VSMCs samples were added with RIPA lysis (Beyotime)
   containing phosphatase (Roche PhoSTOP) and protease inhibitors
   (Beyotime). The final concentration of protein was analyzed using BCA
   Protein Assay Kits (Thermo Fisher Scientific). Proteins with different
   molecular weights were separated on sodium dodecyl sulfate
   (SDS)–polyacrylamide gel (Epizyme), and then transferred to PVDF
   membrane (Millipore). The PVDF membranes were blocked with 5% BSA
   solution at room temperature for 1 h, followed by incubation with
   specified antibodies overnight at 4 °C. After incubating secondary
   antibodies, the membranes were developed with Chemiluminescence imaging
   system (ChemiDoc Touch or Amersham Imager 600 or minichemi910). The
   bands’ intensity was analyzed by Image J software. The antibodies are
   listed in Table [204]S3 (Supporting Information).

RNA Extraction, Reverse Transcription, and Quantitative Real‐Time Polymerase
Chain Reaction (qPCR)

   Total RNA was extracted by RNA Purification Kit (EZBioscience)
   according to the manufacturer's protocols. RNA concentration was tested
   using Nanodrop 2000. For qPCR analysis, cDNAs were obtained and
   real‐time PCR amplifications were performed according to the protocol
   (EZBioscience). The relative expressions were represented as Ct values
   standardized by ACTB, and the fold changes were assessed by the 2^−ΔΔCt
   method. The sequences of primers are listed in Table [205]S4
   (Supporting Information).

Calcification Quantification

   Calcification of the whole‐mount aorta, from arcus aortae to the
   bifurcation of iliac artery, was determined by Alizarin Red staining
   (leagene). Artery tissues were fixed in 95% ethanol for 24 h, then
   stained with 0.003% Alizarin Red in 1% sodium hydroxide (Guangzhou) for
   30 h. Two percent sodium hydroxide was used to wash tissue twice before
   taking photos. Calcium depositions of aorta were evaluated with the
   Calcium Assay Kit (Nanjing Jiancheng), which was based on the released
   calcium after decalcified with 0.6 mmol/l hydrochloric acid (Guangzhou)
   at 37 °C for 48 h. The contents of calcium were calibrated to the dry
   weight of artery tissues. Aortic ring slice with calcification was
   determined by the Von Kossa staining. Briefly, aortas were treated with
   5% silver nitrate (Macklin) and exposed to ultraviolet light for
   30 min, rinsed, and incubated with 5% sodium thiosulfate (GENMED) to
   show the staining brown. As for VSMCs calcification, was also
   identified by Alizarin Red staining (leagene) as previously
   described.^[ [206]^28 , [207]^45 ^] Calcium deposition of VSMCs is
   detected using a Calcium Assay Kit (Nanjing Jiancheng) after
   decalcifying. The contents of calcium were corrected for protein
   contents.^[ [208]^45 ^]

Detection of Glutamate in Supernatant

   The concentration of glutamate in the supernatant was detected by
   Glutamate Assay Kit‐WST (DOJINDO). In general, 100 µL supernatant was
   taken from the cell culture medium, and diluted with ultrapure water.
   The experiment was conducted according to the manufacturer's protocol.

Intracellular Ca^2+ Measurement

   VSMCs counts were adjusted to 3000/well in a 96‐well black opaque plate
   and gave the corresponding interventions. Fluo‐4‐AM (Beyotime) was
   added to a final concentration of 2.5 µm. The solution was gently mixed
   and incubated for 30 min at 37 °C in the dark. Cleaning it up and
   detecting it with the fluorescence microplate reader.

Nuclear‐Cytoplasmic Separation

   Nuclear and cytoplasmic separation was carried out according to the
   protocol (Beyotime). Briefly, cells were collected in a 1.5 mL
   centrifuge tube and centrifuged to obtain cell precipitation. The cells
   were re‐suspended in cytoplasmic lysis buffer and rested on the ice for
   20 min. The supernatant was taken as cytoplasmic part after high‐speed
   centrifugation (4 °C, 15 000 g for 5 min). The remaining cell
   precipitates were suspended in the nuclear protein extraction reagent
   and were completely mixed at low temperature. The supernatant was taken
   as a nuclear component after high‐speed centrifugation (4 °C, 15 000 g
   for 10 min).

Immunohistochemical Staining

   Aortic tissues were fixed in 4% paraformaldehyde and embedded in
   paraffin to be cut into 5‐µm sections. After being baked,
   deparaffinization and hydration were performed with xylene and alcohol.
   Then, heat‐induced antigen retrieval unmasked aortic tissues. Three
   percent H[2]O[2] was used to eliminate endogenous peroxidase activity.
   The sections were blocked in 5% BSA (Solarbio) at room temperature for
   30min. After that, the regions of aorta were incubated with specified
   primary antibodies overnight at 4 °C and enhanced enzyme‐linked goat
   anti‐mouse/rabbit IgG polymer (Zhongshan Company) at room temperature
   for 30min. Diaminobenzidine was selected for visualization. After being
   hydrated again and sealed, images were captured.

Immunofluorescence Staining

   Immunofluorescence observation of GLS1 and SM22α in mouse aortic tissue
   was carried out. Parts of the details are described in the section of
   Immunohistochemical staining. After being incubated with primary
   antibodies, Alexa Fluor 594–conjugated secondary antibodies (Beyotime)
   and Alexa Fluor 488–conjugated secondary antibodies (Beyotime) were
   chosen to incubate. Finally, containing DAPI (Servicebio) sealing agent
   was stained nuclei. Microscopy was performed with Leica. As for VSMCs,
   1× PBS was used to wash 3 times after treatments. Then, 4%
   paraformaldehyde solution was applied to fix for 15 min. Next, VSMCs
   were removed from the paraformaldehyde and washed in PBS 3 times.
   Permeabilized with 0.1% Triton‐X, cells were washed another 3 times
   with PBS and then incubated with 5% BSA for 30 minutes. After that,
   cells were incubated with rabbit anti‐active β‐Catenin (CST) overnight
   at 4 °C, Alexa Fluor 488–labeled secondary antibodies (Beyotime) for 1
   h at room temperature, and DAPI (Servicebio) for 5 min at room
   temperature. Following this, cells were washed three times with PBS.
   Imaging was performed using LSM 880 with Airyscan.

Statistical Analyses

   All statistical analyses were conducted using GraphPad Prism 8.0.
   Values were expressed as means ± SEM. Comparisons between the two
   groups were done using two‐tailed Student's t‐tests or nonparametric
   Mann–Whitney U test. Multiple comparisons were tested by ANOVA followed
   by Bonferroni's or Dunnett's post‐hoc analysis. Additionally, the
   correlation was confirmed using Spearman's rank correlation
   coefficient. p < 0.05 was considered statistically significant.

Ethical Statement

   All animal experiments abided by animal ethical standards and were
   approved by the Ethics Committee of Zhongshan School of Medicine, Sun
   Yat‐sen University (SYSU‐ACUC‐2022‐B0053). The clinical trial protocol
   and human samples collection were performed with the Declaration of
   Helsinki and approved by the Ethics Committee of The Eighth Affiliated
   Hospital of Sun Yat‐sen University (2023‐100‐01). All participants or
   legal representatives gave written informed consent before entering the
   study.

Conflict of Interest

   The authors declare no conflict of interest.

Author Contributions

   Z.Z., B.D., D.H., and J.M. contributed equally to this work. Z.Z.,
   J.T., and H.H. conceived the research. Z.Z., B.D., D.H., and J.M.
   performed the major experiments. Y.K., H.Z., C.X., T.Y., X.Z., and Z.Z.
   contributed to the collection of clinical populations and blood
   samples. Z.H. and J.C. collected the arterial samples. Z.Z., A.H., and
   R.W. analyzed and interpreted the clinical data. Z.Z., Y.C., and H.H.
   drafted and revised the manuscript, with comments from all authors.
   H.Y. supervised English spelling. All authors approved the final
   version of the article.

Supporting information

   Supporting Information
   [209]ADVS-12-2414252-s001.pdf^ (1.8MB, pdf)

Acknowledgements