Anti-aging Gene Klotho Deficiency Promoted High Fat Diet-induced Arterial Stiffening via Inactivation of AMP-activated Protein Kinase

Abstract

Klotho was originally discovered as an aging-suppressor gene. The objective of this study is to investigate if klotho gene deficiency affects high fat diet-induced arterial stiffening. Heterozygous Klotho deficient (KL^+/−) mice and WT littermates were fed on high fat diet (HFD) or normal diet (ND). HFD increased pulse wave velocity (PWV) within 5 weeks in KL^+/− mice but not in WT mice, indicating that klotho deficiency accelerates and exacerbates HFD-induced arterial stiffening. A greater increase in blood pressure was found in KL^+/− mice fed on HFD. Protein expressions of phosphorylated AMPKα (pAMPKα), phosphorylated eNOS (peNOS), and Mn-SOD were decreased while protein expressions of collagen I, TGFβ1, and Runx2 were increased in aortas of KL^+/− mice fed on HFD. Interestingly, daily injections of an AMPKα activator, AICAR, abolished the increases in PWV, blood pressure, and blood glucose in KL^+/− mice fed on HFD. Treatment with AICAR for 2 weeks not only abolished the downregulation of pAMPKα, peNOS, and Mn-SOD levels but also attenuated the increased levels of collagen I, TGFβ1, Runx2, superoxide, elastic lamellae breaks, and calcification in aortas of KL^+/− mice fed on HFD. In cultured mouse aortic smooth muscle cells (SMCs), cholesterol plus KL deficient serum decreased phosphorylation levels of AMPKα and LKB1 (an important upstream regulator of AMPKα activity) but increased collagen I synthesis, which can be eliminated by activation of AMPKα by AICAR. In conclusions, Klotho deficiency promoted HFD-induced arterial stiffening and hypertension via downregulation of AMPKα activity.

Introduction

Arterial stiffening is one of the aging-related disorders which can be accelerated by metabolic syndromes, diabetes, and arteriosclerosis.^{1, 2} Arterial stiffening leads to isolated systolic hypertension.¹ Pulse wave velocity (PWV) is a non-invasive measure of arterial stiffening.³ Abundant epidemiological studies have demonstrated that arterial stiffening is an independent predictor of cardiovascular outcomes such as myocardial infarction, cognitive decline in aging, stroke, and chronic kidney diseases.^4–7 Two longitudinal studies indicated that arterial stiffness predicts an increase in systolic blood pressure and incident hypertension.^{8, 9}

Klotho (KL) gene was originally identified as a putative anti-aging gene in mice.¹⁰ The Klotho gene was named after the purported Greek goddess Klotho who spins the thread of life.¹¹ KL homozygous deficient mice carry a disruption in the promoter region of the KL gene, leading to extensive premature aging phenotypes including severe hyperphosphatemia, ectopic soft tissue calcification, and early death (<10 weeks).^{10, 11} Overexpression of KL, however, extended life span in mice.^{12, 13} The mouse full length KL gene encodes a single-pass transmembrane domain of 1014 amino acids (130 kDa).¹¹ The short-form KL (≈65 kDa) can be generated by alternative RNA splicing or proteolytic cleavage.¹¹ KL protein is predominantly expressed in the kidney and slightly expressed in parathyroid glands and brain choroid plexus.¹¹ Both human and mouse KL gene are alternatively spliced after exon 3,¹⁴ which encodes the KL1 repeat of Klotho, called secreted Klotho (sKL). Soluble KL might include the truncated extracellular domain (KL1 and KL2), the KL1 fragment, and the KL2 fragment.^11,15 Secreted and soluble KL circulates in the blood and regulates function in organs and cells (e.g., vascular cells) that do not express KL.¹¹ KL plays important roles in a variety of physiological and pathological processes including modulation of Wnt signal transduction, anti-oxidation, and renal ion channels and transporters.^{11, 15, 16}

Aging and aging-related medical complications (metabolic syndrome, hypertension, diabetes, chronic kidney disease) are associated with a decreased ratio of elastin/collagen (arterial remodeling) and arterial calcification (elastocalcinosis) which contribute significantly to arterial stiffening.^17–20 Transforming growth factor β1 (TGF-β1) via binding to its receptors induces a variety of gene expressions in vascular smooth muscle cells that may result in collagen deposition and calcification.^{21, 22} The prevalence of arterial stiffening is increased with age while anti-aging gene klotho levels are decreased with age.¹ However, whether KL deficiency plays a role in the development of arterial stiffening is not fully understood.

5′ adenosine monophosphate-activated protein kinase (AMPK) has been shown to be essential in regulating vascular homeostasis. AMPKα is a serine/threonine kinase that regulates cellular energy homeostasis through its enzymatic activity stimulated by phosphorylation of threonine-172 in the catalytic alpha subunit.^{23, 24} The phosphorylation status at threonine-172 is often used as an indicator of the activation state of AMPKα.²⁵ In addition, activation of AMPKα through 5-aminoimidazole-4-carboxamide-3-ribonucleoside (AICAR), an adenosine mimetic, was shown to decrease mean arterial pressure and vascular tone in hypertensive rats.²⁶ AMPKα inactivation has been found in a model of calcification in rat aortic smooth muscle cells and Metformin has been shown to inhibit calcification via the activation of the AMPKα pathway.²⁷ However, whether AMPKα is involved in the pathogenesis of arterial stiffening is not clear.

The purpose of the study is to investigate if Klotho deficiency plays a role in the pathogenesis of arterial stiffening in animals fed on high fat diet (HFD). We found that KL deficiency plus HFD accelerated and exacerbated arterial stiffening, which was associated with AMPKα inactivation in the arteries. In addition, we further assessed the effect of activation of AMPKα by AICAR on HFD-induced arterial stiffening in KL-deficient mice.

Methods

Animal studies

Heterozygous KL^+/− mutant mice with more than 9 generations in 129/Sv background were kindly provided Dr. Kuro-o.¹⁰ Briefly, all mice were housed in cages at room temperatures (25±1°C) and were provided with laboratory chow (Cat# 5053, PicoLab) and tap water ad libitum. This project was approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Oklahoma Health Sciences Center. For details, see the Online Supplemental Methods and Data.

Statistical Analysis

Data were analyzed using a one-way ANOVA. The Newman-Keuls procedure was used to assess differences between means. Data were expressed as mean ± SEM. A probability value with p<0.05 were considered significant.

Results

KL-deficiency accelerated and exacerbated HFD-induced arterial stiffening and hypertension

To explore whether KL deficiency affects the development of arterial stiffening in animals with metabolic syndrome, KL heterozygous mutant mice and their wild type littermates were fed on HFD (Fig. 1A). HFD increased PWV as early 5 weeks in KL^+/− deficient mice but not in WT mice (Fig. 1B). These results suggested, for the first time, that klotho deficiency promotes HFD-induced arterial stiffening.

Figure 1. KL deficiency accelerated and exacerbated HFD-induced arterial stiffening and hypertension.

(A), scheme of the experimental designs. (B), pulse wave velocity (PWV) of arterial flow. (C), arterial systolic blood pressure. (D), blood glucose levels. (E), body weights. N=8 * p<0.05, **p<0.01, *** p<0.001 vs. WT-normal diet (ND); ⁺p<0.05, ⁺⁺p<0.01, ⁺⁺⁺p<0.001 vs. KL^+/−-ND; ^p<0.05, ^^p<0.01, ^^^ p<0.001 vs. WT-HFD.

Interestingly, HFD induced a greater increase in systolic blood pressure (BP) by 5 weeks in KL deficient mice compared with WT mice fed on HFD (Fig. 1C), indicating that KL deficiency exacerbates HFD-induced systolic hypertension. Systolic BP was elevated to a lesser extent in KL deficient mice on normal diet (ND) and WT mice on HFD (Fig. 1C). Similar changes were found in mean arterial blood pressure (Fig. S1A). It is noticed, however, that diastolic BP was not elevated in KL^+/− mice until 6 weeks after HFD (Fig. S1B), suggesting that HFD-induced elevation of systolic BP and diastolic BP in KL^+/− mice is isolated.

Hyperglycemia is an important feature of metabolic syndrome. HFD slightly increased blood glucose levels in both KL deficient and WT mice (Fig. 1D). Interestingly, KL deficient mice with HFD developed an earlier increase in blood glucose levels, compared to WT mice on HFD (Fig. 1D). No significant changes in insulin sensitivity and glucose tolerance were found in these animals (Fig. S1C&D). These data indicated that animals on HFD might be still in the early stage in the development of metabolic syndrome.

Obesity is another hallmark of metabolic syndrome. HFD tended to increase body weight in both genotypes but no significance was found until the 11^th week. KL deficient mice fed on HFD started to demonstrate a significantly greater increase in body weight, compared to the WT-ND and the KL^+/−-ND (Fig. 1E).

KL deficiency exacerbated HFD-induced expression of aortic collagen I and TGFβ1 but decreased levels of aortic pAMPKα, peNOS, and Mn-SOD levels in mice fed on HFD

During the 6th week of HFD, one third of the animals were sacrificed for assessing the molecular changes associated with arterial stiffening. HFD increased collagen I and TGFβ1 expression in WT mice (Fig. 2A). KL deficient mice on normal diet had greater levels of TGFβ1 and collage I compared with WT mice on normal diet (Fig. 2A). Interestingly, KL deficiency further increased HFD-induced TGFβ1 and collagen I expression (Fig. 2A). An increase in TGFβ1 and collagen I expression could contribute to arterial stiffening. Unexpectedly, there was no significant difference in tropoelastin levels among different animal groups (Fig. 2A). Taken together, these results suggested that KL deficiency further promotes HFD-induced collagen synthesis and arterial stiffening.

Figure 2. KL deficiency exacerbated HFD-induced expression of aortic collagen I and TGFβ1 and led to downregulation of aortic pAMPKα, peNOS, and Mn-SOD levels in mice fed on HFD for 6 weeks.

(A), Collagen I, Tropoelastin, Runx2 protein levels in aorta. (B), pAMPKα(T172), AMPKα, peNOS(S1177), eNOS, and Mn-SOD protein levels in aorta. (C), plasma total cholesterol levels. (D), plasma LDL-Cholesterol. (E), plasma calcium levels. (E), plasma phosphorus levels. N=5. *p<0.05, **p<0.01, *** p<0.001 vs. WT-ND; ⁺p<0.05, ⁺⁺p<0.01, ⁺⁺⁺p<0.001 vs. KL^+/−-ND; ^p<0.05, ^^p<0.01, ^^^ p<0.001 vs. WT-HFD

Interestingly, KL deficient mice on HFD displayed decreased phosphorylation levels of AMPKα and eNOS (Fig. 2B), suggesting that KL deficiency plus HFD downregulates AMPKα and eNOS activities. KL deficient mice on HFD also demonstrated lower levels of Mn-SOD (Fig. 2B). We confirmed that KL^+/− deficient mice had about half of short form KL protein in plasma (Fig. S1E&F).

No significant elastic fiber breaks or obvious calcification were found in cross-sections of aortas at this stage of HFD feeding by 6 weeks of HFD feeding (Fig. S2A–D). Therefore, arterial stiffening observed in KL deficient animals on HFD preceded the noticeable elastin degradation and arterial calcification at this early stage.

HFD increased plasma total cholesterol and LDL-cholesterol in mice

HFD increased total plasma cholesterol and LDL-cholesterol in both WT and KL deficient mice to a similar extent (Fig. 2C&D), indicating that KL deficiency did not affect HFD-induced increases in plasma cholesterol. There was no significant change in plasma free fatty acids and triglycerides after 6-week HFD feeding in either WT or KL deficient mice (Fig. S2E&F).

Since hypercalcemia and hyperphosphotemia have been found in KL homozygous (−/−) mutant mice,¹⁰ we measured plasma levels of calcium and phosphorus in these animals. No significant changes in plasma calcium and phosphorus were found in these animals (Fig. 2E&F), suggesting that arterial stiffening found in KL^+/− deficient mice fed on HFD might not be attributed to changes in plasma calcium and phosphorus.

Given that arterial stiffening might be accelerated by chronic kidney disease, we measured blood albumin and blood urea nitrogen levels. No significant differences in blood albumin and blood urea nitrogen levels were observed (Fig. S2G&H), suggesting that arterial stiffening in KL deficient mice on HFD might not be caused by kidney dysfunction.

AICAR alleviated arterial stiffening, hypertension, and blood glucose levels in KL deficient mice fed on HFD

Since HFD decreased activation of AMPKα, we treated animals with AICAR, an analog of AMP that activates AMPK.²⁴ Interestingly, AICAR treatment decreased PWV in KL deficient mice on HFD to the control levels within one week (Fig. 3A). These results indicated that activation of AMPK is sufficient to eliminate arterial stiffening in KL deficient animals on HFD. In addition, AICAR abolished the increases in systolic arterial BP in all treated groups (Fig. 3B), suggesting that AMPK might be an effective therapeutic target for hypertension. Similar results were observed for mean and diastolic arterial BP (Fig. S3A&B). Furthermore, AICAR abolished the mild increase in blood glucose levels induce by HFD in both WT and KL deficient mice (Fig. 3C), suggesting that activation of AMPK by AICAR might regulate glucose homeostasis in animals fed on HFD. Lastly, AICAR also decreased body weight in KL deficient mice fed on HFD (Fig. 3D). Therefore, activation of AMPKα improved not only arterial stiffening but also metabolic syndrome in KL deficient mice.

Figure 3. AICAR alleviated arterial stiffening, hypertension, and high blood glucose levels in KL deficient mice fed with HFD.

(A), pulse wave velocity (PWV) of arterial flow. (B), arterial systolic blood pressure. (C), blood glucose levels. (D), body weights. N=5–8. *p<0.05, **p<0.01 vs. WT-ND-Saline; ⁺p<0.05, ⁺⁺p<0.01 vs. KL^+/−-ND-Saline; ^$$p<0.01 vs. WT-HFD-Saline; ^p<0.05, ^^ p<0.01 vs. KL^+/−-HFD-Saline.

AICAR slightly decreased blood total cholesterol and LDL-cholesterol in KL deficient mice fed on HFD

AICAR treatments slightly but significantly decreased blood total cholesterol and LDL-cholesterol in KL deficient mice but not in WT mice fed on HFD (Fig. S3C&D). These data suggested that KL deficient mice might be more sensitive to AICAR in terms of downregulation of HFD-induced hypercholesterolemia which may partially contributes to a decrease in arterial stiffening by AICAR in KL deficient mice. Although KL homozygous mice (−/−) suffer from hypercalciumia and hyperphosphotemia,¹¹ serum levels of calcium and phosphorus were in normal range in KL heterozygous (+/−) mice (Fig. S3E&F). No significant changes in serum calcium and phosphorus were observed after AICAR treatments (Fig. S3E&F).

AICAR increased pAMPKα, peNOS, and Mn-SOD levels and decreased superoxide production, collagen I levels, and TGFβ1 expression in aortas

AICAR increased phosphorylation levels of AMPKα and eNOS in all treated groups although it did not alter protein expression of AMPKα and eNOS (Fig. 4A&B). These results suggested that AICAR treatments might decrease arterial stiffening and hypertension by activating AMPKα and eNOS. In addition, AICAR treatments also increased Mn-SOD in all three groups (Fig. 4C). Interestingly, the urinary level of nitrite/nitrate (index of nitric oxide) was increased during weeks 1 and 2 of treatment with AICAR (Fig. S4), suggesting that activation of AMPKα enhances bioavailability of nitric oxide (NO).

Figure 4. AICAR treatments increased pAMPKα(T172), peNOS(S1177), and Mn-SOD levels in mouse aorta.

(A), pAMPKα and AMPKα levels. (B), peNOS and eNOS levels. (C), Mn-SOD levels. N=5–8. *p<0.05, **p<0.01, ***p<0.001 vs. WT-ND-Saline; ⁺p<0.05, ⁺⁺p<0.01, ⁺⁺⁺p<0.001 vs. KL^+/−-ND-Saline; $$p<0.01, $$$p<0.001 vs. WT-HFD-Saline; ^p<0.05, ^^ p<0.01, ^^^p<0.001 vs. KL^+/−-HFD-Saline.

Interestingly, KL deficiency plus HFD for 15 weeks significantly increased superoxide levels in aortas which can be abolished by AICAR (Fig. 5A&B). AICAR completely abolished klotho deficiency- and HFD-induced upregulation of collagen I and TGFβ1 in aortas (Fig. 6A–C). These data suggested that the beneficial effects of AICAR on arterial stiffening and hypertension may be mediated via the AMPKα-eNOS-MnSOD-Superoxide-TGFβ1-collagen I pathway.

Figure 5. AICAR alleviated arterial superoxide production in KL deficient mice with HFD.

(A), representative images of aortic DHE staining. (B), quantification of DHE staining for mouse aorta. N=5–8. ***p<0.001 vs. WT-ND-Saline; ⁺⁺⁺p<0.001 vs. KL^+/−-ND-Saline; ^^^p<0.001 vs. KL^+/−-HFD-Saline.

Figure 6. AICAR down-regulated collagen I, Runx2, and TGFβ1 levels in mouse aortas.

(A), representative western blots fro collagen I, tropoelastin, Runx2, and TGFβ1. Quantification of Collagen I (B), TGFβ1 (C), Runx2 (D), and tropoelastin (E). N=5–8. *p<0.05, ** p<0.01, ***p<0.001 vs WT-ND-Saline; ⁺p<0.05, ⁺⁺p<0.01, ⁺⁺⁺p<0.001 vs KL^+/−-ND-Saline; $$p<0.01, $$$p<0.001 vs WT-HFD-Saline; ^p<0.05, ^^ p<0.01, ^^^p<0.001 vs KL^+/−-HFD-Saline.

AICAR decreased aortic Runx2 levels and abolished arterial calcification and elastic fiber breaks in KL deficient mice on HFD

Arterial calcification and elastic fiber breaks have been considered as important factors contributing to the pathogenesis of arterial stiffening.^{28, 29} AICAR significantly decreased protein expression of Runx2 (a marker of osteoblasts) but not tropoelastin in aortas in KL deficient mice fed on HFD (Fig. 6A,D,E). Interestingly, KL deficient mice fed on HFD for 15 weeks developed arterial calcification and more arterial elastic fiber breaks which can be abolished by AICAR (Fig. S5A–D). Therefore, these results demonstrated that HFD-induced calcification and elastic fiber fragmentation in arteries of KL deficient mice may be mediated by downrgulation of AMPKα. Further, activation of AMPKα by AICAR might alleviate arterial calcification via decreasing arterial Runx2, a key transcription factor in the regulation of bone formation.

KL protein deficiency plus cholesterol inactivate AMPKα via downregulation of LKB1 activity

To further study the molecular mechanisms of direct regulation of collagen I protein expression by KL and to avoid in vivo complications, MOVAS, a cell line of mouse aortic smooth muscle cells, were used. Interestingly, we found that water-soluble cholesterol dose-dependently increased precursor of collagen I in MOVAS (Fig. S6A&B). Neither full length KL nor short form KL was detected in MOVAS (Fig. S7A). KL gene was not detected in MOVAS (Fig. S7B). Since FBS used in cell culture contained both full length and short form KL, we removed about 50% of KL in both forms from FBS using an IP purification kit (Fig. S7C–E). Interestingly, cholesterol plus KL deficiency indeed decreased phosphorylation of AMPKα (Fig. S6C–E), whereas AICAR treatments increased pAMPKα levels (Fig. S6C–E). As shown in Fig. S6F&G, KL-deficient FBS potentiated the cholesterol-induced increase in collagen I. AICAR treatments abolished the increase in collagen I induced by cholesterol alone or KL-deficient FBS plus cholesterol in MOVAS (Fig. S6F&G). Interestingly, cholesterol plus KL deficiency also attenuated phosphorylation level of LKB1 and protein level of LKB1 in MOVAS (Fig. S6H–K). Cholesterol plus KL deficiency did not affect CaMKKα and CaMKKβ levels in MOVAS (Fig. S7F–I). LKB1 and CaMKKβ are considered key activators of AMPKα in various mammalian cells ³⁰. Taken together, these results indicated that cholesterol plus KL deficiency might inactivate AMPKα likely via decreasing LKB1 activity, resulting in collagen I accumulation in aortic smooth muscle cells.

Discussion

Aging is associated with a decline in klotho levels and an increase in prevalence of arterial stiffening and hypertension.¹ At age 70 years, the serum level of klotho is about one half of what it was at 40 years.^{11, 31} This study demonstrated, for the first time, that one half deficiency of klotho, an aging-suppressor gene, accelerated and exacerbated HFD-induced arterial stiffening and hypertension (Fig. 1B&C). This finding advances the current understanding of aging-related arterial stiffness and hypertension.¹ An increase in PWV and elevation of systolic BP occur at approximately the same time in KL(+/−) mice fed on HFD (e.g., 5 weeks). Therefore, arterial stiffening may not be due to elevation of BP because hypertension-associated arterial remodeling is a slow and chronic process. By contrast, arterial stiffening would lead to an immediate elevation of systolic BP. Indeed, systolic BP was elevated as early as 5 weeks of HFD while diastolic BP was not increased until 6 weeks of HFD in KL deficient mice (Fig. 1C, Fig. S1B). The recent Framingham study showed that large artery stiffness precedes the development of hypertension.³² This report indicated that arterial stiffening may be the cause of hypertension.³² Two longitudinal studies have demonstrated that arterial stiffness predicts an increase in systolic blood pressure and incident hypertension.^{8, 9} This is especially true for aging-related arterial stiffening.¹ Kidney function was normal (Fig. S2G&H), excluding the involvement of renal dysfunction in HFD-induced arterial stiffness and hypertension in KL(+/−) mice.

Although HFD caused arterial stiffening only in KL(+/−) mice, it increased total plasma cholesterol levels in both KL(+/−) and WT mice to a similar extent (Fig. 2). Therefore, KL deficiency makes animals more susceptible to HFD-induced arterial damage. KL is predominately expressed in kidneys.^10–12 We showed that KL^+/− deficient mice have one half of KL in kidneys compared to WT mice.³³ Kidneys are the major source of circulating klotho.³⁴ Indeed, plasma levels of KL protein were decreased by 50% in KL^+/− deficient mice (Fig. S1). These results suggest that a reduction in the circulating KL may synergize with increased plasma LDL-cholesterol levels to promote HFD-induced arterial stiffening and hypertension. Therefore, KL deficient mice are a unique animal model for studying the mechanism of arterial stiffness associated with metabolic syndrome.

It is noticed that one half klotho deficiency plus HFD diminished activities of AMPKα (pAMPKα) in aortas (Fig. 2). Interestingly, activation of AMPKα by AICAR abolished the increases in PWV and blood pressure in KL^+/− mice fed on HFD (Fig. 2, Fig. S3). To the best of our knowledge, this is the first report demonstrating that inactivation of AMPKα may be involved in the pathogenesis of arterial stiffening and hypertension. These interesting findings suggest that AMPKα might be a potential therapeutic target for arterial stiffening and hypertension. The impaired AMPKα activity may mediate the downregulation of eNOS activity (peNOS) and Mn-SOD expression which can be eliminated by activation of AMPKα by AICAR (Fig. 4). Normalization of eNOS activity and Mn-SOD expression was associated with abolishment of the increased superoxide levels in aortas of KL^+/− mice fed on HFD (Fig. 5). Activation of AMPKα also attenuated the increased levels of TGFβ1 and collagen I in aortas from KL^+/− deficient mice fed on HFD. These results strongly suggest that inactivation of AMPKα is an important upstream factor in the regulation of arterial stiffening in KL deficient mice and that activation of AMPKα is sufficient to restore arterial compliance which is associated with normalization of TGFβ1 and collagen I levels. Arterial collagen deposition has been believed to be an important factor contributing to the pathogenesis of arterial stiffening.² Thus, one of ultimate goals in the treatment of arterial stiffening is to block collagen synthesis.

It is expected that eNOS activity was increased by one-week treatment with AICAR because activation of AMPKα by AICAR would functionally interact with eNOS and upregulate its activity (phosphorylation) via LKB1³⁰ which occurs quickly. Indeed, the nitric oxide (NO) level was increased within one week of treatment with AICAR (Fig. S4). Therefore, the early and quick drop in PWV and BP within one week of treatment may be partially attributed to relaxation of blood vessels due to increased bioavailability of NO. Thus, there may be a functional component of arterial stiffness due to increased vascular tension in KL^+/− mice treated with HFD. By contrast, the structural recovery of blood vessels is a relatively slow process although it contributes to the attenuation of arterial stiffening and hypertension by AICAR.

Klotho deficiency did not significantly affect metabolic parameters (cholesterol, body weight, phosphorus) under normal or HFD diets (Fig. 2). AICAR only slightly decreased the HFD-induced increase in plasma cholesterol levels (Fig. S3). It has been well documented that AICAR increases fatty acid oxidation via activation of the ACC1 and/or ACC2 pathways in skeletal muscles and liver.^35–38 AICAR may also inhibit cholesterol synthesis in the liver.³⁹ Metformin, another AMPK activator also decreases the cholesterol levels.⁴⁰ It is noticed, however, that the slight drop of plasma cholesterol levels cannot explain the significant attenuating effect of AICAR on HFD-induced arterial stiffening and hypertension (Fig. 3). On the other hand, AICAR can directly activate AMPKα in vascular cells which, in turn, increases eNOS activity (Fig. 4) leading to improvement in arterial stiffening and hypertension in KL^+/− mice fed on HFD. Indeed, AICAR almost abolished the increase in PWV and BP, suggesting that the rescuing effect of AICAR may be primarily mediated by its vascular effect.

Klotho gene deficiency or mutation is associated with arterial stiffening and hypertension.^{1, 11} A decrease in plasma Klotho is used as a biomarker of arterial stiffening in patients with chronic kidney disease.⁴¹ Aging-related hypertension is largely due to arterial stiffening or remodeling.¹ Unfortunately, the current antihypertensive drugs are mainly designed to reduce peripheral resistance and are not adequate to alter the pathological process of arterial stiffening or even selectively reduce systolic BP in isolated systolic hypertension.¹⁸ Thus, this study suggests that supplementation with Klotho protein or pharmacological activation of AMPKα may be a novel therapeutic strategy to alleviate arterial stiffening and hypertension. The anti-hypertensive effect of AICAR may be mediated by activation of eNOS and decreased arterial stiffening.

To assess the direct effect of cholesterol in SMCs, water-soluble cholesterol was added to the cultured SMCs. Cholesterol loading by water-soluble cholesterol has been shown to induce mouse aortic SMCs into a macrophage-like state.⁴² Water-soluble cholesterol delivered cholesterol rapidly and directly to the plasma membrane.⁴³ Thus, cholesterol loading to cultured aortic SMCs might recapture the effects of high LDL-cholesterol on aortic SMCs in animals fed on HFD. KL protein-deficient serum exacerbated cholesterol-induced collagen I protein expression, and AICAR abolished the promoting effects of KL deficiency on cholesterol-induced collagen I expression in SMCs (Fig. S6). AMPK has been reported to inhibit TGFβ-induced fibrogenic responses (collagen I) in hepatic stellate cells by targeting transcriptional coactivator p300.⁴⁴ Adenoviral transduction of constitutively active AMPKα was sufficient to prevent TGFβ-induced collagen I and fibronectin in cultured fibroblasts.⁴⁵ Therefore, these data strongly suggest that AICAR decreased arterial stiffening largely via suppressing cholesterol-induced collagen I synthesis.

Interestingly, KL deficiency plus cholesterol loading also attenuated the phosphorylation level of LKB1 and protein expression of LKB1 in SMCs (Fig. S6). LKB1 is one of the key activators of AMPKα in various mammalian cells.³⁰ Therefore, these results suggest that KL deficiency plus cholesterol might inactivate AMPKα likely through decreasing LKB1 activity.

In this study, BP was measured using a computerized volume-pressure recording (VPR) tail-cuff method, a non-invasive and high-throughput measurement technique. It facilitates long-term monitoring of BP in unanesthetized animals. This method has been confirmed to be in good agreement with the radiotelemetry measurement and recommended by the American Heart Association (AHA).⁴⁶ The repeatable measurements of BP over a 15-week period are a strong guarantee for the reliability of the BP data. This method can reliably monitor BP and is a common method for monitoring BP in our laboratory.^47–50

KL deficiency and HFD did not affect aortic elastin protein expression levels (Figs. 2&6) but increased elastic lamellae breaks (Fig. S5). The increased elastin streak fragmentation, which seems to be mediated by downregulation of AMPKα, would also contribute to arterial stiffening. Another interesting finding is that KL deficient mice fed on HFD developed arterial calcification (Fig. S5). Arterial calcification could contribute to arterial stiffening. RUNX2 is a reliable marker of osteoblasts and has been used as an indicator of bone formation. Therefore, KL deficiency plus HFD leads to arterial calcification likely via inactivation of AMPKα because AICAR abolished the upregulation of RUNX2 and arterial calcification in KL deficient mice fed on HFD (Figs. 6, S5).

The serum levels of klotho decrease with age after age 40.³¹ By contrast, the prevalence of arterial stiffening and hypertension increases with age.¹ Klotho^+/− mice were used which mimics a half klotho reduction in the aged population.³¹ Klotho homozygous (−/−) mice develop extensive aging phenotypes and die before the age of 8 weeks (body weight = 8 grams).¹⁰ Klotho homozygous mice also demonstrate severe hyperphosphatemia, emphysema, and soft tissue calcification.^{11, 12} As a result, klotho homozygous mice were not used for this study.

Perspective

This study revealed a previously unidentified role of KL deficiency in promoting the development of HFD-induced arterial stiffening and hypertension. Protein expression of pAMPKα, peNOS, and Mn-SOD was downregulated while protein expression of collagen I, Runx2, and TGFβ1 was upregulated in aortas of KL^+/− mice fed on HFD which can be abolished by activation of AMPKα by AICAR. Therefore, KL deficiency might promote HFD-induced arterial stiffening via down-regulation of AMPKα activity which leads to up-regulation of collagen I levels in aortas. Therapeutic activation of AMPKα might be a novel strategy for alleviating arterial stiffening and hypertension. It should be mentioned that hypertension is a complicated and multifactorial disorder. The antihypertensive effect of AMPK activation may not be applied to other forms of hypertension that do not involve reduction of AMPK activity.

Novelty and Significance.

1. What is new?

1. It is new and interesting that haplodeficiency of klotho gene accelerates the development of high fat diet (HFD)-induced arterial stiffening and hypertension.

2. This study demonstrates, for the first time, that klotho gene deficiency plus HFD downregulates vascular AMPKα expression and activity.

2. What is relevant?

1. It is significant that klotho deficiency promotes HFD-induced arterial stiffening and hypertension which are prevalent cardiovascular disorders associated with aging.

2. This study reveals that activation of AMPKα may be a new therapeutic strategy for arterial stiffening, an independent risk factor for cardiovascular mortality and morbidity.

3. Summary

Klotho deficiency promoted HFD-induced arterial stiffening and hypertension. The promoting effects of Klotho deficiency on arterial stiffening might be mediated by downregulation of AMPKα activity.