Targeted Genomic Disruption of H-Ras Induces Hypotension Through a NO-cGMP-PKG Pathway–Dependent Mechanism
The aim of the present experiments was to evaluate the differences in arterial pressure between H-Ras lacking mice and control mice and to analyze the mechanisms involved in the genesis of the differences. H-Ras lacking mice and mouse embryonic fibroblasts from these animals were used. Blood pressure was measured using 3 different methods: direct intraarterial measurement in anesthetized animals, tail-cuff sphygmomanometer, and radiotelemetry. H-Ras lacking mice showed lower blood pressure than control animals. Moreover, the aorta protein content of endothelial nitric oxide synthase, soluble guanylyl cyclase, and cyclic guanosine monophosphate–dependent protein kinase was higher in H-Ras knockout mice than in control animals. The activity of these enzymes was increased, because urinary nitrite excretion, sodium nitroprusside–stimulated vascular cyclic guanosine monophosphate synthesis, and phosphorylated vasoactive-stimulated phosphoprotein in aortic tissue increased in these animals. Furthermore, mouse embryonic fibroblasts from H-Ras lacking mice showed higher cyclic guanosine monophosphate–dependent protein kinase promoter activity than control cells. These results strongly support the upregulation of the nitric oxide-cyclic guanosine monophosphate pathway in H-Ras–deficient mice. Moreover, they suggest that H-Ras pathway could be considered as a therapeutic target for hypertension treatment.
- arterial pressure
- H-Ras protein
- nitric oxide
- cyclic GMP
- soluble guanylyl cyclase
- cGMP-dependent protein kinase
Small GTP-binding proteins (G proteins) are monomeric G proteins with a molecular mass of 20 to 40 kDa. A small G protein acts as a molecular switch between inactive GDP-bound and active GTP-bound cycles. Ras proteins are small G proteins highly conserved from yeast to humans. Three functional Ras genes are ubiquitously expressed in mammals. These genes are located on different chromosomes and encode four 21-kDa proteins: H-Ras, N-Ras, K-Ras4A, and K-Ras4B.1–3 Ras mediates its effect on cell proliferation mainly through the activation of its effector Raf, initiating the mitogen-activated protein kinase cascade.4 In addition, a variety of Ras effectors have been identified, such as a phosphatidylinositol 3-kinase (PI3K) and Raf-Erk1/2.5
Ras isoforms have a very high degree of homology (≈80%) at the protein level. H- and K-Ras have a >90% homology excluding the last 25 amino acids. Most of the differences between these proteins are found within the hypervariable region at the carboxyl end of the molecule. However, the different Ras isoforms seem to have different functions.6 Mice knocked out for H-Ras, N-Ras, or both isoforms are viable, whereas mutations in K-Ras are lethal, indicating that K-Ras is not only essential but also sufficient for normal mouse development.7–9
Until recently, the relationships between Ras and the regulation of blood pressure have scarcely been studied. Previous reports suggested a role for Ras in the cellular response to angiotensin II,10,11 but no definite relationship between this protein and blood pressure was established. However, genetic manipulation of the different isoforms of the Ras gene has underscored the relevance of this protein as a hemodynamic regulator. Thus, it has been reported that the knock-in of the H-Ras coding sequence at the K-Ras locus (HRas-KI) is viable, although a dilated cardiomyopathy associated with arterial hypertension was shown.12 More recently, it has been reported that transgenic mice for a constitutively activated form of H-Ras have hypertension and heart hypertrophy,13 thus suggesting a role for H-Ras in the cardiovascular system. However, an analysis of the intrinsic cardiovascular mechanisms involved in the genesis of this hypertension has not been extensively performed.
The present experiments were devoted to answer part of these questions, using a different experimental approach. H-Ras lacking mice are viable and apparently normal, but no study has evaluated these animals from a hemodynamic point of view. We hypothesized that H-Ras−/− mice may show changes in arterial pressure that were opposite to those observed in the mice overexpressing Ras. Because the nitric oxide (NO)/cyclic cGMP/cGMP-dependent protein kinase (PKG) signaling pathway seems to be one of the more relevant systems involved in the regulation of blood pressure,14–18 we planned to explore this system in our H-Ras knock-out mice.
A complete description of the methods used in the present study is provided in the online Data Supplement at http://hyper.ahajournals.org.
H-Ras–deficient mice (H-Ras−/−) were obtained as previously reported.7
Mouse embryonic fibroblasts (MEFs) were isolated using previously described methods. Studies were performed on confluent monolayers at passages 2 to 4.19
Blood Pressure Measurements
Blood pressure was measured using 3 different methods: direct intra-arterial measurement in anesthetized animals,20 tail-cuff sphygmomanometer,21,22 and radiotelemetry.23 Blood pressure was measured under basal conditions and after the administration of the following treatments: acetylcholine (ACh) (1 μg/kg body weight [BW] intraperitoneal [IP]), sodium nitroprusside (SNP) (2 μg/kg BW IP), db-cGMP (5 mg/kg BW IP), NG-nitro-l-arginine methyl ester (l-NAME) (20 mg/kg BW per day, drinking water for 2 weeks), 1H-[1,2,4]oxadiazole[4,3-a]quinoxalin-1-one (ODQ) (5 mmol/kg BW IP), DT-3 (500 μg/kg BW IP), and chaetomellic acid A (1.5 mg/kg BW IP at 24 and 16 hours before blood pressure measurement and euthanasia).24
Preparation of the Aortic Rings and Vascular Reactivity Studies
The preparation of the aortic rings and vascular reactivity studies were performed as described.22 Norepinephrine-contracted aortic rings were treated with increased concentrations of SNP and db-cGMP, and arterial wall tension was recorded.
Protein Extraction and Western Blot Analysis
Tissues were studied under basal conditions. Cells were studied under basal conditions and after treatments with SNP (10−4 mol/L, 15 minutes) or 8-Br-cGMP (10−4 mol/L, 15 minutes). In the SNP experiments, to assess the specificity of the observed effects, cells were preincubated with ODQ (10−6 mol/L, 30 minutes). Tissues or cells were washed in PBS and solubilized for 30 minutes at 4°C. Immunoblotting was performed as described.25
RNA Extraction and Analysis of mRNA Expression by Northern Blot
Thoracic aortas were harvested from H-Ras+/+ and H-Ras−/− mice, fixed with formalin, and subjected to increasing concentrations of ethanol and xylene. Tissues were embedded in paraffin, cut into serial sections 3 to 5 μm thick, incubated with the anti–endothelial nitric oxide synthase (eNOS) (dilution 1:100), soluble guanylyl cyclase (sGC)-α1 (dilution 1:100), sGC-β1 (dilution 1:100), and PKG-I antibodies (dilution 1:200), then with biotin-labeled goat anti-mouse and anti-rabbit antibodies, and subsequently exposed to the avidin–peroxidase complex. Finally, diaminobenzidine was added to serve as substrate. Samples were counterstained with Mayer’s hematoxylin. Negative controls were obtained by omitting primary antibodies.
Nitrites were analyzed using the Griess method.27 Animals were housed in individual metabolic cages and their urine was collected for 24 hours.
Measurement of cGMP
Tissues or cells were studied under basal conditions and after treatment with SNP (10−4 mol/L, 15 minutes), with or without ODQ (10−6 mol/L, 30 minutes). cGMP was determined as previously described.22
Transient Transfection and Luciferase Assays
A total of 3.5×105 cells/well were plated in six-well plates 24 hours before transfection. MEFs were incubated 4 hours at 37°C with 2 mL Opti-MEM medium containing complexes of 2 μL of Lipofectamine, 0.1 μg of human PKG-I reporter,28 and 0.01 μg of Renilla luciferase reporter as an internal control. The transfected cells were then incubated with complete growth medium for 24 hours, washed with PBS, and lysed. Finally, firefly luciferase activity of the PKG-I reporter was measured with a luminometer (FB12 Berthold luminometer) and normalized against the Renilla luciferase activity using the dual-luciferase reporter assay system. Protein concentration was determined by BCA.
The number of experiments performed is reflected in the legends of the figures because this number was never more than 10; nonparametric statistics were used for comparisons (Friedman and Wilcoxon tests for paired data and Kruskal–Wallis or Mann–Whitney tests for nonpaired data). The data are expressed as the means±SEM. A value of P<0.05 was considered statistically significant.
Data obtained by either direct cannulation of anesthetized animals or tail-cuff measurements and telemetry in conscious animals revealed that H-Ras−/− mice showed lower systolic, diastolic, and mean arterial pressure than H-Ras+/+ mice. Heart rate measurements taken in anesthetized animals were significantly lower in H-Ras−/− than in WT mice, whereas it was significantly higher when measurements were obtained in conscious animals (Table). Telemetry measurements provided additional information about the basal circadian rhythm of mouse blood pressure. Wild-type animals exhibited higher systolic blood pressure in the dark than in the light period (118±2 and 102±1 mm Hg, respectively, P<0.01), a finding that was also observed in H-Ras−/− mice (106±1 and 96±1 mm Hg, P<0.01). In each period of the day, systolic blood pressure was significantly lower in the knock-out mice.
The intraperitoneal administration of ACh, SNP, or db-cGMP induced a significant reduction of systolic arterial pressure (SAP) in both the H-Ras+/+ and H-Ras−/− mice (Figure 1A through 1F). After these treatments, the SAP values showed a tendency toward being significantly lower in the H-Ras−/− mice than in the wild-type animals. These differences were statistically significant when mice received ACh and SNP, and the tail-cuff method was used (Figure 1A and 1C), and with both tail-cuff measurements and telemetry after db-cGMP treatment (Figure 1E and 1F). The blockade of NO with l-NAME, sGC with ODQ, and PKG with DT-3 increased SAP in both groups of animals, and the differences in blood pressure disappeared (Figure 2A through 2F). To confirm these findings, vascular reactivity studies were performed in isolated aortic rings. Precontracted rings from both strains of animals were incubated with different concentrations of SNP or db-cGMP, and the tension of the rings was recorded. Both compounds induced a significantly increased relaxation of the rings that was significantly more marked in the H-Ras−/− mice (Figure S1A and S1B in the online Data Supplement).
Because these pharmacological experiments suggested an upregulation of the NO-cGMP-PKG signaling pathway in the vascular walls of the H-Ras−/− mice, we measured the aortic protein content of eNOS, the α1 and β1 subunits of sGC, and PKG-I (Figure 3A through 3F). The protein content of all of these proteins was significantly higher in H-Ras−/− mice than in control animals. Moreover, these quantitative changes in the protein content determined functional modifications in H-Ras−/− mice. Thus, they excreted a higher amount of nitrites in urine (Figure S2A), their aortic ring synthesized higher amounts of cGMP when treated with SNP (Figure S2B), and phosphorylation of the PKG substrate VASP on Ser239 increased in their aortic walls (Figure S2C).
To gain insight into the mechanisms responsible for the changes observed in H-Ras−/− mice, some experiments were performed in MEFs from these animals. As occurred in their vascular walls, the protein content of sGC-α1, sGC β1, and PKG-I was significantly higher in the MEFs from H-Ras−/− mice (Figure S3A) than in the control MEFs. The H-Ras lacking cells, stimulated by SNP, showed greater cGMP synthesis (Figure S3B) and VASP phosphorylation (Figure S3C) than their respective controls. Moreover, when 8-Br-GMP was used as an agonist of PKG, MEFs without H-Ras also showed an increased VASP phosphorylation (Figure S3D) when compared with control cells. After the confirmation that the changes observed in the cGMP-PKG pathway were comparable in aortic walls and in MEFs, preliminary experiments were performed to evaluate the relationship between H-Ras deficiency and increased PKG-I content. The lack of H-Ras induced a significant increase of the PKG-I mRNA content (Figure S3E), probably related to an increase of the transcriptional activity of the human PKG-I 2 Kb promoter region (Figure S3F).
Finally, the ability of chaetomellic acid A, an inhibitor of Ras farnesylation with rather high specificity for H-Ras,24,29–30 to modify blood pressure was tested in wild-type animals. Mice treated with the inhibitor exhibited lower blood pressure than control animals (106±1 mm Hg versus 116±2 mm Hg, P<0.01), their aortic rings showed an increased relaxing response after treatment with different agonists of the NO-sGC-PKG pathway (Figure S4A and S4B). Moreover, they also showed an increased sGC and PKG content in aortic walls (Figure S4C and S4D).
The analysis of blood pressure in mice is a relatively complex procedure, which is done mainly by using the classic indirect method based on tail sphygmomanometers, with the subsequent possible interferences attributable to movement restriction. A direct intraarterial pressure measurement requires the animals to be anesthetized, with the subsequent possible interferences of anesthesia. These methodological problems increase when the expected changes in blood pressure are quantitatively small.
Thus, to study the changes in AP in H-Ras–deficient mice, we used an approach that combined these 2 techniques and added telemetry,20–23 a procedure that allows to measure arterial blood pressure and heart rate in conditions that may be considered almost completely physiological. Regardless of the procedure performed, H-Ras−/− mice showed a lower AP than control animals, with a reduction in SAP ranging from 7 to 13 mm Hg. This difference was observed either in the active period (dark) or in the inactive period (light).
Lower blood pressure in H-Ras−/− mice may be explained by decreased circulating volume, deficient cardiac function, or slightly higher vasodilation. In turn, this higher vasodilatation may be the consequence of deficient vasoconstriction or increased vascular relaxation. Because H-Ras is expressed in vascular smooth muscle31 and endothelial cells,32 it could be expected a direct effect of H-Ras deficiency in vascular tone. Therefore, we set out to test the possibility that increased vasodilatation, caused by upregulation of the NO-cGMP-PKG pathway, could be responsible for the changes detected in AP.
When compared with their wild-type littermates, H-Ras–deficient mice showed lower AP, and exhibited increased hypotensive responses to the administration of drugs that stimulate endogenous NO release, or activate sGC or PKG. The vascular protein content of eNOS, α1 and β1 subunits of sGC, and PKG I was higher in H-Ras−/− mice than in wild-type animals, and the activity of these enzymes was also increased, as suggested by increased urinary nitrite excretion, SNP-stimulated vascular cGMP synthesis, elevated phosphorylated VASP in thoracic aortic tissue, and the response of isolated aortic rings to variable concentrations of SNP and db-cGMP. Moreover, the blood pressure of H-Ras–deficient mice reached values comparable to those of the control animals when eNOS, sGC, or PKG were blocked, a finding that would not be observed if the differences in AP between both kind of animals were not attributable to changes in the NO-cGMP-PKG pathway. Taken together, these results strongly support the upregulation of the NO-sGC-PKG pathway in H-Ras–deficient mice and the functional relevance of this upregulation in the genesis of the hypotension of these animals.
Parts of the present results do not appear to be thoroughly explained. Blood pressure changes after ACh and SNP treatments were similar in both experimental groups when telemetry was used, whereas a significant difference in this parameter was observed when arterial pressure was measured with the sphygmomanometer. The possibility that the circadian rhythm of AP could be involved in the genesis of these apparent discrepancies was considered, but measurements of blood pressure were performed at the same time with both procedures, and both strains of animals showed significant changes in systolic AP between the dark and light cycles, which could interfere with the effects of ACh and SNP, although the differences were quantitatively higher in the control animals. Unfortunately, the treatments administered to the animals involved an intraperitoneal injection, with the subsequent interferences with the basal circadian rhythm and significantly increased stress.
The complexity of the analysis of the in vivo results led us to perform a number of experiments in a cellular system, MEFs from H-Ras–deficient and control mice. Obviously, eNOS cannot be studied in these cells, but the differences in sGC and PKG were assessable. The results were completely comparable to those obtained in mice: The lack of H-Ras induced the overexpression of both sGC isoforms and PKG-I, and these overexpressed enzymes were functionally active. An additional advantage of these experiments on cells is that they allowed us to better understand the relationships between H-Ras deficiency and enzyme overexpression. A preliminary analysis performed for PKG-I suggest that the changes in the protein content could be related to an increased activation of the transcription of the gene caused by increased promoter activity.
Our results cannot be compared with previous studies, because no similar experimental approaches have previously been performed. However, Jeon et al indirectly explored the relationship between H-Ras and eNOS when studying the mechanisms involved in the endothelial dysfunction observed in APE1/ref-1 mice.33 These animals showed a significantly lower H-Ras, and phospho-S1177 eNOS protein content in aortic tissue, but an increased eNOS content, suggesting a deficiency in the activation of the enzyme. Studies in COS-7 cells led the authors to propose a relationship between H-Ras deficiency and impaired eNOS phosphorylation. It could be argued that our H-Ras–deficient animals could also exhibit an increased but functionally deficient eNOS, but the in vivo response to ACh and the analysis of the nitrite excretion do not support this possibility. Moreover, it has been reported that the administration of FPT III, a Ras-GTPase inhibitor, significantly decreased mean arterial blood pressure in spontaneously hypertensive rats treated with l-NAME,34 an observation that supports the relevance of our findings.
The present results provide new information about the mechanisms involved in the regulation of the NO-sGC-PKG pathway. In fact, the presented data support an inverse relationship between H-Ras and the expression level of the proteins of this pathway. A careful analysis of these mechanisms may complete our knowledge about the regulation of this important cellular system, mainly concerning the targets of the NO, such as sGC and PKG, which have been less extensively studied.
The data obtained point to the possibility that the inhibition of H-Ras may be used as a way to control blood pressure or to improve tissue perfusion in some cardiovascular diseases. In this sense, the effects of chaetomellic acid administration decreasing blood pressure, increasing aortic relaxing response, and increasing vascular sGC and PKG in control animals suggest that pharmacological inhibition of Ras, and particularly of H-Ras, could constitute a useful therapeutic tool in hypertension.
Sources of Funding
This work was supported by the “Ministerio de Educacion y Ciencia” (SAF 2004-07845-C02, SAF 2004-07845-C02-01, SAF 2007-6389, and SAF 2007-623471), the “Ministerio de Sanidad” (FISS PI070695 and RETICS RD6/0016 [RedinRen]), and Junta de Castilla y León (SA001/C05 and SA029/A05 and Excellence Group GR100).
- Received February 26, 2010.
- Revision received March 8, 2010.
- Accepted July 8, 2010.
Macara IG, Lounsbury KM, Richards SA, McKiernan C, Bar-Sagi D. The Ras superfamily of GTPases. FASEB J. 1996; 10: 625–630.
Takai Y, Sasaki T, Matozaki T. Small GTP-binding proteins. Physiol Rev. 2001; 81: 153–208.
Santos E, Nebreda AR. Structural and functional properties of ras proteins. FASEB J. 1989; 3: 2151–2163.
Esteban LM, Vicario-Abejón C, Fernández-Salguero P, Fernández-Medarde A, Swaminathan N, Yienger K, Lopez E, Malumbres M, McKay R, Ward JM, Pellicer A, Santos E. Targeted genomic disruption of H-ras and N-ras, individually or in combination, reveals the dispensability of both loci for mouse growth and development. Mol Cell Biol. 2001; 21: 1444–1452.
Johnson L, Greenbaum D, Cichowski K, Mercer K, Murphy E, Schmitt E, Bronson RT, Umanoff H, Edelmann W, Kucherlapati R, Jacks T. K-ras is an essential gene in the mouse with partial functional overlap with N-ras. Gene Dev. 1997; 11: 2468–2481.
Ohtsu H, Suzuki H, Nakashima H, Dhobale S, Frank GD, Motley ED, Eguchi S. Angiotensin II Signal Transduction Through Small GTP-Binding Proteins Mechanism and Significance in Vascular Smooth Muscle Cells. Hypertension. 2006; 48: 534–540.
Potenza N, Vecchione C, Notte A, De Rienzo A, Rosica A, Bauer L, Affuso A, De Felice M, Russo T, Poulet R, Cifelli G, De Vita G, Lembo G, Di Lauro R. Replacement of K-Ras with H-Ras supports normal embryonic development despite inducing cardiovascular pathology in adult mice. EMBO Rep. 2005; 6: 432–437.
Shesely EG, Maeda N, Kim HS, Desai KM, Krege JH, Laubach VE, Sherman PA, Sessa WC, Smithies O. Elevated blood pressures in mice lacking endothelial nitric oxide synthase. Proc Natl Acad Sci U S A. 1996; 93: 13176–13181.
Sausbier M, Schubert R, Voigt V, Hirneiss C, Pfeifer A, Korth M, Kleppisch T, Ruth P, Hofmann F. Mechanisms of NO/cGMP-dependent vasorelaxation. Circ Res. 2000; 87: 825–830.
Friebe A, Koesling D. Regulation of nitric oxide-sensitive guanylyl cyclase. Circ Res. 2003; 93: 96–105.
Koeppen M, Feil R, Siegl D, Feil S, Hofmann F, Pohl U, de Wit C. CGMP-dependent protein kinase mediates NO- but not acetylcholine-induced dilations in resistance vessels in vivo. Hypertension. 2004; 44: 952–955.
Jerkic M, Rivas-Elena JV, Prieto M, Carron R, Sanz-Rodriguez F, Perez-Barriocanal F, Rodriguez-Barbero A, Bernabeu C, Lopez-Novoa JM. Endoglin regulates nitric oxide-dependent vasodilatation. FASEB J. 2004; 18: 609–611.
Perez-Rivero G, Ruiz-Torres MP, Rivas-Elena JV, Jerkic M, Diez-Marques ML, Lopez-Novoa JM, Blasco MA, Rodriguez-Puyol D. Mice deficient in telomerase activity develop hypertension because of an excess of endothelin production. Circulation. 2006; 114: 309–317.
Ruiz-Torres MP, Griera M, Chamorro A, Diez-Marques ML, Rodriguez-Puyol D, Rodriguez-Puyol M. Tirofiban increases soluble guanylate cyclase in rat vascular walls: pharmacological and pathophysiological consequences. Cardiovasc Res. 2009; 82: 125–132.
Tornavaca O, Pascual G, Barreiro ML, Grande MT, Carretero A, Riera M, Garcia-Arumi E, Bardaji B, Gonzalez-Nuñez M, Montero MA, Lopez-Novoa JM, Meseguer A. Kidney androgen-regulated protein transgenic mice show hypertension and renal alterations mediated by oxidative stress. Circulation. 2009; 119: 1908–1917.
Sabbatini M, Santillo M, Pisani A, Paterno R, Uccello F, Seru R, Matrone G, Spagnuolo G, Andreucci M, Serio V, Esposito P, Cianciaruso B, Fuiano G, Avvedimento EV. Inhibition of Ras/ERK1/2 signaling protects against postischemic renal injury. Am J Physiol Renal Physiol. 2006; 290: 1408–1415.
Diez-Marques ML, Ruiz-Torres MP, Griera M, Lopez-Ongil S, Saura M, Rodriguez-Puyol D, Rodriguez-Puyol M. Arg-Gly-Asp (RGD)-containing peptides increase soluble guanylate cyclase in contractile cells. Cardiovasc Res. 2006; 69: 359–369.
Chomczynski P, Sacchi N. Single-step methods of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biol Chem. 1987; 162: 156–159.
Blanco FJ, Grande MT, Langa C, Oujo B, Velasco S, Rodriguez-Barbero A, Perez-Gomez E, Quintanilla M, López-Novoa JM, Bernabeu C. S-endoglin expression is induced in senescent endothelial cells and contributes to vascular pathology. Circ Res. 2008; 103: 1383–1392.
Sellak H, Choi C, Browner N, Lincoln TM. Upstream stimulatory factors (USF-1/USF-2) regulate human cGMP-dependent protein kinase I gene expression in vascular smooth muscle cells. J Biol Chem. 2005; 280: 18425–18433.
Gibbs JB, Pompliano DL, Mosser SD, Rands E, Lingham RB, Singh SB, Scolnick EM, Kohl NE, Oliff A. Selective inhibition of farnesyl-protein transferase blocks ras processing in vivo. J Biol Chem. 1993; 268: 7617–7620.
Ramos KS. H-RAS controls phenotypic profiles of vascular smooth muscle cells and the pathogenesis of vascular proliferative disorders. Circ Res. 2009; 104: 1139–1141.
Serban D, Leng J, Cheresh D. H-ras regulates angiogenesis and vascular permeability by activation of distinct downstream effectors. Circ Res. 2008; 102: 1350–1358.
Jeon BH, Gupta G, Park YC, Qi B, Haile A, Khanday FA, Liu YX, Kim JM, Ozaki M, White AR, Berkowitz DE, Irani K. Apurinic/apyrimidinic endonuclease 1 regulates endothelial NO production and vascular tone. Circ Res. 2004; 95: 902–910.
Benter IF, Francis I, Khan I, Cojocel C, Juggi JS, Yousif MH, Canatan H, Alshawaf EH, Akhtar S. Signal transduction involving Ras-GTPase contributes to development of hypertension and end-organ damage in spontaneously hypertensive rats-treated with l-NAME. Pharmacol Res. 2005; 52: 401–412.