Biomechanical response of arterial wall to DOCA–salt hypertension in growing and middle-aged rats

ARTICLE IN PRESS

Journal of Biomechanics 40 (2007) 1583–1593 www.elsevier.com/locate/jbiomech www.JBiomech.com

Biomechanical response of arterial wall to DOCA–salt hypertension in growing and middle-aged rats Kozaburo Hayashia,b,, Takanori Sugimotoc a

Research Institute of Technology, Okayama University of Science, 1-1 Ridai-cho, Okayama 700-0005, Japan b Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka, Japan c Division of Bioengineering, Department of Mechanical Science and Bioengineering, Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka, Japan Accepted 14 July 2006

Abstract To study arterial remodeling in response to hypertension, Deoxycortico-sterone acetate (DOCA)–salt hypertension was induced in immature (aged 16 weeks) and middle-aged (48 weeks) rats, and biomechanical properties and wall dimensions of common carotid arteries were determined. Arterial segments were excised at 10 or 16 weeks postoperatively from the immature rats and at 16 weeks from the middle-aged ones. In vitro pressure–diameter tests were performed under normal (in Krebs–Ringer solution), active (norepinephrine), and passive (papaverine) conditions. Non-treated, age-matched rats (26, 32, and 64 weeks) were used to obtain control data. Wall thickness at in vivo blood pressure level was increased by hypertension at all ages; however, there were no significant changes in inner diameter. In hypertensive rats, arterial outer diameter was smaller under normal condition than under passive condition, indicating the increase of smooth muscle tone by hypertension. Diameter reduction developed by norepinephrine was increased by hypertension, which was significant above 100 mmHg; however, there were no significant differences between hypertensive and normotensive arteries, if compared at respective in vivo blood pressures. No significant differences were observed in wall stiffness at in vivo pressure. Wall hoop stress at in vivo blood pressure had a significant positive correlation with the pressure in 26-week old arteries. However, there were no differences in the stress between hypertension and normotension in 32- and 64-week old arteries. These results were essentially similar to previous ones observed in Goldblatt hypertension and in younger animals. Age-related differences in arterial wall remodeling were not clearly observed. r 2006 Elsevier Ltd. All rights reserved. Keywords: Arterial remodeling; Wall hypertrophy; Wall stress; Elasticity and stiffness; Smooth muscle contraction

1. Introduction Geometry, structure, and mechanical properties of living organs and tissues change in response to such environmental changes as zero gravity, hyperbaric conditions, or rehabilitation (Fung, 1993). This phenomenon is called tissue remodeling. Many experimental studies and clinical observations have shown that the chronic elevation of blood pressure changes dimensions and properties of Corresponding author. Research Institute of Technology, Okayama University of Science, 1-1 Ridai-cho, Okayama 700-0005, Japan. Tel./fax: +81 86 256 9403. E-mail address: [email protected] (K. Hayashi).

0021-9290/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbiomech.2006.07.021

arterial wall (Fung, 1993; Hayashi et al., 1996, 2001; Holzapfel and Ogden, 2002; Humphrey, 2002). One of the specific biomechanical phenomena occurring in response to hypertension is wall thickening, which restores wall hoop stress to normal level (e.g. Matsumoto and Hayashi, 1994). Arterial elasticity at in vivo working pressure gradually changes to optimal level. Moreover, vascular tone is increased by hypertension (e.g. Fridez et al., 2001). Many of these studies have been done using animal models, including Goldblatt hypertension (GH in short; e.g. Matsumoto and Hayashi, 1994), Deoxycorticosterone acetate (DOCA)–salt hypertension (e.g. Berthon et al., 2002), aortic banding/ligation (e.g. Fridez et al., 2001), and spontaneously hypertensive rats (SHR; e.g. Marque et al.,

ARTICLE IN PRESS 1584

K. Hayashi, T. Sugimoto / Journal of Biomechanics 40 (2007) 1583–1593

1999). In GH, the serum concentration of angiotensin is increased by the constriction of one of the renal arteries; angiotensin contracts arteries, resulting in systemic hypertension. In this model, angiotensin II directly acts on arterial wall, and develops wall hypertrophy and vascular smooth muscle (VSM) contraction (Levy et al., 1988). On the other hand, SHR has different circulatory function, wall compositions, and vascular anatomy and morphology from the normal rat (Greenwald and Berry, 1978). Therefore, these two models and the aortic banding/ligation model cannot be regarded as very good models of human hypertension. In DOCA–salt treated animals, Na+ and water are absorbed in the kidney, which increases circulating blood volume and results in hypertension. This process is similar to that in human essential hypertension (Zuckerman and Yin, 1989). Moreover, the mechanism for hypertension in this model is not directly associated with arterial wall. Therefore, DOCA–salt model is more suitable for the study of the response of arterial wall to hypertension in relation to clinical hypertension. In previous studies, however, DOCA–salt hypertension model has been less commonly used for the study of arterial wall response to hypertension than GH and SHR models. Moreover, these models have been applied to relatively young, immature animals. Blood pressure and/or flow change during maturation (e.g. Kawasaki et al., 1987; Lee et al., 1972; Sugimoto et al., 2003), which may affect the remodeling of arterial wall. Furthermore, the ability of remodeling may depend upon age. In the present study, DOCA–salt hypertension was induced in immature and middle-aged rats, and biomechanical properties and wall dimensions of common carotid arteries were determined to see: (1) whether arterial wall remodeling occurs in growing and middle-aged, DOCA– salt hypertensive animals, and if so, (2) whether the phenomenon is age-related. 2. Methods 2.1. Animals and induction of hypertension We used two age groups of male Wistar rats: (1) growing group consisting of 23 animals aged 16 weeks, and (2) middle-aged group of 6 animals aged 48 weeks. Because only 8-week old rats are commercially available, we have to keep rats in our laboratory for a long period of time to obtain middle-aged animals, for example for 40 weeks to start experiments on 48-week old animals. Compared with growing ones, therefore, much less middle-aged rats were used to save time and cost. After a silicone rubber stick (size, 50  10  3 mm3) impregnated with DOCA (100 mg/kg body weight) was subcutaneously implanted at the dorsal side, each animal was given 1% NaCl water ad libitum (Ormsbee and Ryan, 1973). The dose of DOCA was determined from our preliminary experiments. Systolic blood pressure was measured with tail-plethysmography once a week with

the accuracy of 74.5 mmHg. For biomechanical tests, the animals initially aged 16 weeks were killed at 10 (group 16H10) or 16 (16H16) weeks after, and those aged 48 weeks were killed at 16 weeks after (48H16). Non-treated 5, 7, and 6 rats aged 26, 32, and 64 weeks, respectively, were used to obtain age-matched control data (groups 26C, 32C, and 64C); the data from 32C and 64C have been already reported elsewhere (Sugimoto et al., 2003). All animal procedures were done under the Guideline for Animal Experiments, Graduate School of Engineering Science, Osaka University. 2.2. Arterial specimens At the age of 26, 32 or 64 weeks, a catheter-tip, telemetric pressure transducer (0.7 mm in diameter) was inserted from the right femoral artery under anesthesia, and positioned in the abdominal aorta proximal from the renal artery. Then, blood pressure was measured at a slightly better accuracy compared to the above-mentioned tail-plethysmography, although the differences between the data measured with these two methods were statistically insignificant. The left common carotid artery (LCC) was then exposed, and marked with gentian violet on the surface at the interval of 2.5 mm. These marks were used to determine the in vivo axial extension ratio. Then, the LCC of 20 mm in length was excised; a proximal 10–15 mm long segment was used for biomechanical testing; the remaining portion was used to measure arterial dimensions under no load condition. Finally, each rat was killed by gradual and continuous administration of pentobarbital sodium, although the dose depended upon animals. 2.3. Biomechanical tests Experimental procedures and apparatus for biomechanical tests were similar to those reported previously (Sugimoto et al., 2003). Briefly, each proximal LCC segment was mounted onto a pressure–diameter tester in Krebs–Ringer solution of 37 1C, and stretched to the in vivo length in reference to the marks put on the surface. After the preconditioning of inflation–deflation loops between 0 and 250 mmHg at the rate of 3.3 mmHg/s, a reproducible internal pressure (Pi)-external diameter (D0) curve was determined at 1.7 mmHg/s. A small difference in the curve between inflation and deflation, i.e. hysteresis, was observed because of VSM tone; the data obtained during inflation were used as the data under a normal condition with physiological VSM tone (normal VSM tone). Internal pressure was then increased to 100 mmHg, maintained at the level, and norepinephrine was added to 7  107 M. After the maximal contraction was confirmed, pressure was lowered to 0 mmHg. Then, inflation–deflation was done again. A large hysteresis was observed in the Pi–D0 curve; the data during inflation were used as the data under an active condition (maximal VSM contraction).

ARTICLE IN PRESS K. Hayashi, T. Sugimoto / Journal of Biomechanics 40 (2007) 1583–1593

1585

Internal pressure was again maintained at 100 mmHg, and papaverine was added to 104 M. After relaxation became maximal, inflation–deflation loops between 0 and 250 mmHg were performed to obtain a reproducible pressure–diameter curve. Although there was no significant difference in the curve between inflation and deflation, the inflation curve was used as the data under a passive condition (maximal VSM relaxation). All the biomechanical tests for each vessel were completed within 12 h post-mortem.

tion (Laplace). In such a thin-wall vessel, the maximum hoop stress (at the innermost wall) calculated from a thickwalled equation (Lame) is only by approximately 5% larger than the hoop stress obtained from the thin-walled equation. Therefore, we can use the thin-walled theory as the first approximation. Arterial stiffness was expressed by stiffness parameter, b0 (Hayashi et al., 1980)

2.4. Dimensional measurements under no load

where DD0 is the increment of external diameter induced by the increment of pressure, DPi, at Pi. Elastic property of wall material was represented by the incremental elastic modulus, Hyy, proposed by Hudetz (1979) for orthotropic incompressible materials   D0 DPi D2i H yy ¼ 2 2 þ Pi D0 . (4) DD0 D0  D2i

Arterial rings of 0.5 mm in thickness were cut out from each distal LCC segment. They were placed in Krebs–Ringer solution of 37 1C for 20 min. Their cross-sectional images at no load state under normal VSM tone were recorded on an image analyzer via a microscope and a CCD camera, from which external and internal diameters were determined.

b0 ¼

DPi =Pi , DD0 =D0

(3)

2.6. Statistical analysis

2.5. Data analysis Arterial contractility was evaluated with diameter response (Cox, 1976) dD0 D0  D00 ¼ , D0 D0

(1)

where D0 is external diameter at Pi under the passive condition, and D0 0 is that under the normal or active condition. This parameter represents the degree of basal VSM tone (normal condition) or the contractility of wall (active condition). Wall normal stress in the circumferential direction (wall hoop stress), sy , was calculated from (Humphrey, 2002) Pi Di , (2) 2T where T and Di are wall thickness and internal diameter at Pi , respectively. Di and, therefore, T ( ¼ (D0Di)/2) were calculated from the other wall dimensions measured under no load condition, D0 at Pi, and the assumption of wall incompressibility (Fung, 1993; Sugimoto et al., 2003). Because the ratio of wall thickness to radius was small, i.e. approximately 0.1, we used this thin-wall stress formulasy ¼

All data were expressed as means7standard deviations (SDs). Statistical comparisons between two groups were performed with Student’s t-test. Correlations between continuous two variables (in vivo blood pressure and wall stress) were evaluated by simple linear regression. Differences and correlation were considered to be significant if po0:05 (5%). 3. Results Systolic blood pressure in DOCA animals very gradually increased until 2 weeks after treatment, progressively increased between 2 and 6 weeks, and became constant thereafter. Final systolic blood pressures in DOCA–salt treated animals were significantly higher than those in agematched control animals (Table 1). Therefore, we can say that all the DOCA–salt treated animals became hypertensive. Approximately 60% of DOCA–salt treated rats (64% in group 16H10, 50% in 16H16, and 67% in 48H16) had systolic blood pressure above 160 mmHg. To cancel the effects of different body weight changes among animals, wall dimensions were normalized by the

Table 1 Systolic blood pressure measured immediately before arterial excision in each experimental group Growing groups

Middle-aged groups

Age (wk)

26

32

Group

26C

16H10

32C

16H16

64C

48H16

SBP n

120.277.9 5

174.1724.5 11

124.977.2 7

164.0718.0 12

129.277.2 6

164.8711.2 6

SBP ¼ Systolic blood pressure in mmHg (Mean7SD); n ¼ number of animals.

64

ARTICLE IN PRESS K. Hayashi, T. Sugimoto / Journal of Biomechanics 40 (2007) 1583–1593

1586

Control

Hypertension

250

200

150

100

50

0 26C

16H10 32C 26

(A)

BW=Body weight

(Mean ± SD) BW-normalized wall thickness T/BW1/3 (mm/g1/3)

BW-normalized internal diameter Di/BW1/3 (mm/g1/3)

* p < 0.05 vs. Control

16H16 64C

32 Age (wks)

15 ∗

10

5

0 26C

48H16 64

16H10 32C 26

(B)

16H16 64C

32 Age (wks)

48H16 64

Fig. 1. Effects of hypertension on the internal diameter (A) and wall thickness (B) of the common carotid artery. These dimensions are normalized by body weight (BW) of each animal. Hypertension increased wall thickness, but did not change internal diameter. See Table 1 for experimental groups.

cube root of each body weight (Matsumoto and Hayashi, 1994). At each age in both experimental and control rats, thus normalized, non-loaded external and internal diameters and wall thickness showed a tendency of gradually increasing with blood pressure measured immediately before death (data not shown). There were no significant differences in normalized internal diameter at systolic pressure between hypertensive and control arteries at all ages (Fig. 1). Normalized wall thickness tended to be larger in hypertensive arteries than in control ones; the difference was significant at the age of 64 weeks (48H16 vs. 64C). In control rats, averaged pressure (Pi)–diameter (D0) curves under the normal condition were similar to those under the passive condition (Fig. 2). In hypertensive animals, however, the curves under the normal condition were shifted towards the left from those under the passive condition except for high and low pressures, which indicates increased VSM tone in hypertensive arteries. The diameter reduction developed by norepinephrine (active condition) tended to be larger in hypertensive than in control rats. The pressures at flexion points in Pi–D0 curves under the active condition (Nagasawa et al., 1980) were much higher in hypertensive than in control animals. These phenomena are more clearly seen from Fig. 3. Higher VSM tone (normal condition) and larger vascular contraction (active condition) were observed in hypertensive arteries than in control ones. In addition, the curves under the active condition in hypertensive arteries were shifted towards the right from control curves between 100

and 200 mmHg, indicating that hypertension significantly increased wall contractility at high pressures. Diameter response under the normal condition at each working blood pressure tended to be larger in hypertensive than in control arteries (Fig. 4A). However, there were no significant differences under the active condition between hypertensive and control arteries (Fig. 4B). Hypertensive arteries had higher stiffness at 100 mmHg than control ones, although they had lower stiffness at 200 mmHg (Fig. 5); the differences between the two arteries were smaller if compared at respective systolic blood pressures. These results imply that hypertension reduces arterial wall stiffness at high blood pressures and restores it to normal level. The incremental elastic modulus of wall material was smaller in hypertensive arteries than in control ones at 200 mmHg (Fig. 6). Those differences were statistically significant in 16H10 and 16H16 animals. However, if compared at respective in vivo working pressures, there were no significant differences in the modulus. If we see the relationship between the incremental elastic modulus and the wall circumferential strain, the modulus was larger in hypertensive arteries than in normotensive ones at the same strains (Fig. 7). Wall stress in the circumferential direction (hoop stress) at in vivo blood pressure had a significant correlation with the pressure in 26-week old arteries (Fig. 8); the elevation of blood pressure increased the stress. However, such significant correlations were not observed in 32- and 64-week old arteries; wall stress was independent of blood pressure.

ARTICLE IN PRESS K. Hayashi, T. Sugimoto / Journal of Biomechanics 40 (2007) 1583–1593

Active

Normal

1587

(Mean ± SD)

Passive

250 26C 200

32C

64C

16H16

48H16

150

Internal pressure Pi (mmHg)

100 50 0 250 16H10

200 150 100 50 0 0.8

1.2

1.6

2.0

0.8

1.2 1.6 2.0 External diameter Do (mm)

0.8

1.2

1.6

2.0

Fig. 2. Pressure–diameter curves of control (26C, 32C, and 64C) and hypertensive (16H10, 16H16, and 48H16) arteries under VSM active, normal, and passive conditions. See Table 1 for experimental groups.

* p < 0.05 vs. Control 26 wks 16H10

32 wks 16H16

26C

(Mean ± SD) 64 wks 48H16

32C

64C

0.15 *

* Normal

*

Diameter response δDo/Do

0.10

*

*

*

*

*

Normal

*

*

* *

*

*

0.05

Normal

*

*

*

*

*

*

**

0 0.30 Active

Active

*

* *

* *

0.20

*

*

Active * *

* *

*

* *

0.10 *

**

0 0

50

100

150

200

250

0

50 100 150 200 250 Internal pressure Pi (mmHg)

0

50

100

150

200

250

Fig. 3. Diameter response at each pressure in hypertensive (16H10, 16H16, and 48H16) and control (26C, 32C, and 64C) arteries under VSM normal (upper three figures) and active (lower three figures) conditions. See Table 1 for experimental groups.

ARTICLE IN PRESS K. Hayashi, T. Sugimoto / Journal of Biomechanics 40 (2007) 1583–1593

1588

Control

Hypertension

* p < 0.05 vs. Control 0.08

(Mean ± SD) 0.25

Normal

(A)

(B)

Active

26C

16H10 32C

Diameter response δDo/Do

* 0.20 0.06 0.15 0.04 0.10 0.02 0.05

0

0 26C

16H10 32C 26

16H16 64C

32 Age (wks)

48H16

26

64

16H16 64C 32

48H16 64

Age (wks)

Fig. 4. Effects of hypertension on diameter response at each in vivo blood pressure level. (A) VSM normal condition (normal VSM tone); (B) VSM active condition (maximal VSM contraction). See Table 1 for experimental groups.

Control

Hypertension

* p < 0.05 vs. Control (Mean ± SD) Pi = 100mmHg 6

(A)

Pi = 200mmHg 40

(B)

(C)

*

5 Stiffness parameter β'

Pi = Ps 15

* 4

30 10

3

20

2

5

*

*

10 1 0 26C 16H10 32C 16H16 64C 26

32 Age (wks)

48H16 64

0

0 26C 16H10 32C 16H16 64C 48H16 26

32 Age (wks)

64

26C

16H10 32C 16H16 64C 26

32 Age (wks)

48H16 64

Fig. 5. Effects of hypertension on arterial wall stiffness at pressures of 100 mmHg (A), systolic blood pressure (Ps (B)), and 200 mmHg (C). Data were obtained under VSM normal condition. See Table 1 for experimental groups.

4. Discussion 4.1. Experimental models Blood pressure in DOCA animals increased until 6 weeks post surgery, and reached a plateau thereafter until

16 weeks, which indicates steady release of DOCA. In our previous study using 8- to 9-week old rats treated for GH, blood pressure gradually increased until 6 weeks post surgery, and became constant thereafter until 16 weeks (Matsumoto and Hayashi, 1994). Therefore, the rates of increase in blood pressure were similar in both models.

ARTICLE IN PRESS K. Hayashi, T. Sugimoto / Journal of Biomechanics 40 (2007) 1583–1593

Control

1589

Hypertension

* p < 0.05 vs. Control (Mean ± SD) 0.8

3

Incremental elastic modulus Hθθ (MPa)

(A)

8

(B)

(C)

Pi = Ps

Pi = 100mmHg

Pi = 200mmHg

0.6

6 2

4

0.4

1

*

* 2

0.2

0.0

0

0 26C 16H10 32C 16H16 64C 26

32 Age (wks)

48H16

26C 16H10 32C 16H16 64C 48H16

26C 16H10 32C 16H16 64C 48H16

64

26

32 Age (wks)

26

64

32 Age (wks)

64

Fig. 6. Effects of hypertension on arterial elastic modulus at pressures of 100 mmHg (A), systolic blood pressure (Ps (B)), and 200 mmHg (C). Data were obtained under VSM normal condition. See Table 1 for experimental groups.

Incremental elastic modulus Hθθ (MPa)

15

(A)

26 wks 26C 16H10

(B)

(C)

32 wks 32C 16H16

64 wks 64C 48H16

10

5

0 0

1

2

0

1 Circumferential strain εθθ

2

0

1

2

Fig. 7. Relationships between incremental elastic modulus and wall circumferential strain in control (26C, 32C, and 64C) and hypertensive (16H10, 16H16, and 48H16) arteries at ages of 26 weeks (A), 32 weeks (B), and 64 weeks (C). Data were obtained under VSM normal condition. See Table 1 for experimental groups.

Approximately 60% of DOCA–salt treated animals had systolic blood pressure above 160 mmHg; this percentage was slightly smaller compared to GH (70%; Matsumoto and Hayashi, 1994). In GH model, angiotensin directly affects arterial wall, and develops wall hypertrophy and VSM contraction (Levy et al., 1988). Therefore, it is not clear whether changes in arterial wall are induced by hypertension. In contrast, in DOCA–salt model, the absorption of Na+ and water in the kidney increases circulating blood volume, which results in hypertension. DOCA has no direct influence on arterial wall, and the mechanism for hypertension is similar to that in human essential hypertension

(Zuckerman and Yin, 1989). Therefore, we can say that this model is better than the other ones for the study of arterial response to hypertension, in particular to human hypertension. 4.2. Arterial wall remodeling There were no significant differences in internal diameter between hypertensive and control arteries regardless of ages (Fig. 1). However, wall thickness was larger in hypertensive arteries, which appeared more clearly in older animals. Wall stress in the circumferential direction (hoop stress) had a significant positive correlation with pressure

ARTICLE IN PRESS K. Hayashi, T. Sugimoto / Journal of Biomechanics 40 (2007) 1583–1593

Circumferential stress σθ (kPa)

1590

300

(A)

(B)

26 wks 26C 16H10

32 wks 32C 16H16

(C) 64 wks 64C 48H16

200

100

0 100

150

200

250

r = 0.390 (p = 0.210) m = 0.392

r = 0.402 (p = 0.088) m = 0.684

r = 0.775 (p < 0.05) m = 1.102

100 150 200 250 Systolic pressure Ps (mmHg)

100

150

200

250

Fig. 8. Simple linear plot of normal wall stress in the circumferential direction (hoop stress) vs. in vivo systolic blood pressure measured immediately before arterial excision at ages of 26 weeks (A), 32 weeks (B), and 64 weeks (C). Significant positive correlations were observed in (A), but not in (B) and (C). See Table 1 for experimental groups. Table 2 Comparison of the present results with those from previous studies using rat models

Model Initial age Restoration of wall stress Restoration of wall elasticity

Present study

Berthon et al. (2002)

Sumitani et al. (1997)

Matsumoto and Hayashi (1994)

DH 16, 48 wks 10o, o16 wks o10 wks

DH 4 wks 58 wks o8 wks

GH 12 wks o12 wks —

GH 8 wks o2 wks 8–16 wks

DH ¼ DOCA–salt hypertension; GH ¼ Goldblatt hypertension.

in 26-week old arteries (Fig. 8). However, this was not the case in 32- and 64-week arteries, which is attributable to wall thickening in hypertensive arteries as can be seen from Eq. (2) and Fig. 1. That is, arterial wall hypertrophy occurs in hypertension as if it recovers wall stress to normal level. This phenomenon is regarded as an optimal operation of arterial wall for functional adaptation, which is unique to living tissues and organs (Fung, 1993; Hayashi et al., 1996). A number of experimental studies (e.g. Berthon et al., 2002; Matsumoto and Hayashi, 1994; Sumitani et al., 1997; Wolinsky, 1971, 1972) have shown that arteries change dimensions and properties in response to hypertension. This phenomenon has been observed in clinical cases as well (e.g. Girerd et al., 1994; Roman et al., 1992). One of the specific biomechanical manifestations is wall hypertrophy. Due to this, wall hoop stress is restored to or reduced towards normal level (Table 2). There are several studies on the remodeling of arterial wall using DOCA model. Berry and Greenwald (1976) induced DOCA–salt hypertension in Wistar rats aged 4 weeks. At 16 weeks after treatment, they observed significant increase in aortic wall thickness. However, they did not mention about wall stress. Therefore, we estimated hoop stress from their data, and found that it was at control level at 8 and 16 weeks after treatment, but not at 4 weeks. The restoration of wall stress to control level occurred earlier than in the present study, possibly because of different initial ages of animals. More recently, Berthon et al. (2002) induced DOCA–salt hypertension in Sprague

Dawley rats aged 4 weeks. At 8 weeks after, circumferential wall stress in the common carotid artery was lower than a normal value. This result implies that the stress returned to control level within 8 weeks, which is essentially similar to the result obtained by Berry and Greenwald (1976). From these present and previous results, we can say that wall hoop stress is restored to a control level by 10–20 weeks after the induction of hypertension irrespective of the species of rats and the model of hypertension. In addition to such a stress hypothesis, there is a counterpart, strain hypothesis (Takamizawa and Hayashi, 1987). Recently, Guo and Kassab (2004) have claimed that the latter hypothesis may be more suitable to the remodeling principle than the former one. However, it is difficult to experimentally determine whether blood vessels respond to stress or strain. In relation to tissue remodeling and optimal operation of arterial wall, we measured opening angles (Matsumoto and Hayashi, 1996) in all the specimens to estimate residual stresses. However, we observed no significant differences in the angle between hypertensive and normotensive arteries, and also no significant correlations between the angle and blood pressure. 4.3. VSM tone and arterial contractility VSM basal tone under the normal condition was higher in hypertensive arteries than in normotensive ones (Figs. 2–4). Fridez et al. (2001, 2002) induced hypertension

ARTICLE IN PRESS K. Hayashi, T. Sugimoto / Journal of Biomechanics 40 (2007) 1583–1593

in 8-week old Wistar rats by aortic ligation, and observed that VSM basal tone in the common carotid artery rapidly increased within several days after treatment. This is similar to the present results, although the age of animals, the model of hypertension, and the duration of hypertension are different between the two studies. These results imply that VSM cells are stimulated by hypertension. Many cells including VSM cells change their function and morphology in response to stress (e.g. Frangos, 1993) and, therefore, VSM cells are considered to be a sensing and effecting element for the adaptation process of arterial wall. Cipolla et al. (2002) have shown that increased pressure induces rapid actin polymerization in VSM cells, which results in strong cell contraction. Recently, Humphrey and Wilson (2003) have theoretically demonstrated that the constrictive response of arterial wall to blood pressure change decreases the magnitude and transmural gradients of wall stress and returns the stress towards its homeostatic value. They called the response ‘‘compensatory vasoconstriction’’. Diameter reduction developed by norepinephrine (active condition) was much larger in hypertensive rats than control ones, if compared at the same pressures (Figs. 2 and 3). The pressures at flexion points in Pi–D0 curves were higher in hypertensive rats (Fig. 2). Moreover, dD0/D0Pi curves from hypertensive arteries were shifted towards the right from control curves above 100 mmHg (Fig. 3). These results indicate that arterial contractility increased in response to hypertension. However, there were almost no changes in diameter response, if compared at working blood pressures (Fig. 4, right). This is also a phenomenon showing the optimal operation of arterial wall. There are several articles reporting high contractility in hypertensive arteries. For example, Cox (1979) induced GH (146 mmHg in systolic blood pressure) or DOCA–salt hypertension (162 mmHg) in 15-week old Wistar rats. After 8 weeks, the maximal diameter response in the carotid artery under an active condition was increased by hypertension in both models, although the increase was somewhat greater in GH. The dD0/D0Pi curves obtained from hypertensive arteries under this condition were shifted towards the right from control curves. Fridez et al. (2001) observed in 8-week old Wistar rats that the contractility of the common carotid artery significantly increased at 8 days after the induction of hypertension by aortic ligation. The present results were essentially similar to these previous results. On the other hand, Kamm et al. (1989) observed no significant changes in the mechanical and contractile behavior of the carotid artery even at 7 weeks after the induction of DOCA–salt hypertension in the 2- to 3-month old swine. Such a large difference between their result and the other results may be ascribed to differences in the species of animals and their relative ages. Histologically, Olivetti et al. (1980) observed hypertrophy but not hyperplasia of VSM cells in the thoracic aorta of young adult Wistar rats at 8 days after aortic constriction. Fridez et al. (2003) obtained essentially

1591

similar results from the common carotid artery in 8-week old Wistar rats. Cox (1982) observed increase of relative cell volume in the rat carotid artery by DOCA–salt hypertension, and thought it as one of the reasons for increased active diameter response. In the study mentioned above, Kamm et al. (1989) observed no change in the fraction of VSM cells in the carotid arterial media, which indicates increase in the total volume of cells because wall hypertrophy occurs. Furthermore, many in vitro culture studies have shown that stress affects the proliferation and function of VSM cells (Standley et al., 1999; Sumpio et al., 1988). Like these, VSM cells seem to take an important role in the arterial wall remodeling developed by hypertension. We have performed histological studies on the arterial specimens used in the present experiments, the results from which will be reported elsewhere. 4.4. Wall stiffness and elasticity Hypertensive arteries were less stiff than control at 200 mmHg, whereas they were stiffer at 100 mmHg (Fig. 5); however, there were no significant differences at respective blood pressures. These results imply that wall stiffness at in vivo pressures is restored towards normal. This phenomenon is regarded as a functional adaptation of arterial wall to hypertension. Essentially similar results were observed in DOCA–salt hypertensive rats by Berthon et al. (2002) (Table 2). In 8- to 9-week old rats, Matsumoto and Hayashi (1994) observed no significant difference in aortic stiffness at in vivo blood pressures between GH and normotension at 4 weeks after treatment; however, this was not the case at 2 weeks. In hypertensive rats, the elastic modulus of wall was smaller than in control ones at high pressures if compared at the same pressures (Fig. 6C). At respective in vivo working pressures, however, there were no significant differences in the modulus (Fig. 6B). As stated above, Cox (1979) induced GH and DOCA–salt hypertension in 15-week old rats, and observed no significant differences in passive mechanical properties of the common carotid artery between normal and hypertensive animals at 8 weeks after. Later, he observed in 12-week old Wistar rats that wall elastic modulus increased between 2 and 12 weeks after treatment for DOCA–salt hypertension (Cox, 1982). However, wall elasticity at in vivo working pressures was not dealt with. Berry and Greenwald (1976) induced DOCA–salt hypertension in 4-week old rats, and observed decrease in the elastic modulus of wall at high pressures. More recently, Matsumoto and Hayashi (1994, 1996) have demonstrated in GH rats that the elastic modulus of wall at in vivo pressure came to a normal value after relatively long period of time (Table 2). The similar tendency was also seen in the present study (Fig. 6). These results indicate that arterial wall stiffness and elasticity at in vivo working pressure are restored to normal level at relatively early period after the induction of hypertension for growing animals, in comparison to the

ARTICLE IN PRESS 1592

K. Hayashi, T. Sugimoto / Journal of Biomechanics 40 (2007) 1583–1593

results obtained by others from much younger animals. Age-related differences in the restoration of wall stiffness and elasticity to normal level were not clearly observed in the present study. Acknowledgements The authors appreciate Dr. Hiroshi Miyazaki for his suggestions on experiments and Ms. Emiko Shimizu for her assistance in data analyses. This research work was financially supported in part by the Grant-in-Aid for Scientific Research (A) (2) (nos. 12308147 and 15200036) from the Ministry of Education, Culture, Sports, Science and Technology (Monbu-Kagakusho), Japan. References Berry, C.L., Greenwald, S.E., 1976. Effects of hypertension on the static mechanical properties and chemical composition of the rat aorta. Cardiovascular Research 10, 437–451. Berthon, N., Laurant, P., Hayoz, D., Fellmann, D., Brunner, H.R., Berthelot, A., 2002. Magnesium supplementation and deoxycorticosterone acetate–salt hypertension: effect on arterial mechanical properties and on activity of endothelin-1. Canadian Journal of Physiology and Pharmacology 80, 553–561. Cipolla, M.J., Gokina, N.I., Osol, G., 2002. Pressure-induced actin polymerization in vascular smooth muscle as a mechanism underlying myogenic behavior. FASEB Journal 16, 72–76. Cox, R.H., 1976. Mechanics of canine iliac artery smooth muscle in vitro. American Journal of Physiology 230, H462–H470. Cox, R.H., 1979. Alterations in active and passive mechanics of rat carotid artery with experimental hypertension. American Journal of Physiology 237, H597–H605. Cox, R.H., 1982. Time course of arterial wall changes with DOCA plus salt hypertension in the rat. Hypertension 4, 27–38. Frangos, J.A. (Ed.), 1993. Physical Forces and the Mammalian Cell. Academic, New York. Fridez, P., Makino, A., Miyazaki, H., Meister, J-J., Hayashi, K., Stergiopulos, N., 2001. Short-term biomechanical adaptation of the rat carotid to acute hypertension: contribution of smooth muscle. Annals of Biomedical Engineering 29, 26–34. Fridez, P., Makino, A., Kakoi, D., Miyazaki, H., Meister, J.-J., Hayashi, K., Stergiopulos, N., 2002. Adaptation of conduit artery vascular smooth muscle tone to induced hypertension. Annals of Biomedical Engineering 30, 905–916. Fridez, P., Zulliger, M., Bobard, F., Montorzi, G., Miyazaki, H., Hayashi, K., Stergiopulos, N., 2003. Geometrical, functional, and histomorphometric adaptation of rat carotid artery in induced hypertension. Journal of Biomechanics 36, 671–680. Fung, Y.C., 1993. Biomechanics: Mechanical Properties of Living Tissues. Springer, New York. Girerd, X., Mourad, J.-J., Copie, X., Moulin, C., Acar, C., Safar, M., Layrent, S., 1994. Noninvasive detection of an increased vascular mass in untreated hypertensive patients. American Journal of Hypertension 7, 1076–1094. Greenwald, S.E., Berry, C.L., 1978. Static mechanical properties and chemical composition of the aorta of spontaneously hypertensive rats: a comparison with the effects of induced hypertension. Cardiovascular Research 12, 364–372. Guo, X., Kassab, G.S., 2004. Distribution of stress and strain along the porcine aorta and coronary arterial tree. American Journal of Physiology 286, H2361–H2368. Hayashi, K., Handa, H., Nagasawa, S., Okumura, A., Moritake, K., 1980. Stiffness and elastic behavior of human intracranial and extracranial arteries. Journal of Biomechanics 13, 175–184.

Hayashi, K., Kamiya, A., Ono, K. (Eds.), 1996. Biomechanics : Functional Adaptation and Remodeling. Springer, Tokyo. Hayashi, K., Stergiopulos, N., Meister, J.-J., Greenwald, S.E., Rachev, A., 2001. Techniques in the determination of the mechanical properties and constitutive laws of arterial walls. In: Leondes, C. (Ed.), Cardiovascular Techniques (Biomechanical Systems Techniques and Applications, Vol. II). CRC Press, Boca Raton, FL pp. 6-1–6-61. Holzapfel, G.A., Ogden, R.W. (Eds.), 2002. Biomechanics of Soft Tissue in Cardiovascular Systems. Springer, Heidelberg. Hudetz, A.G., 1979. Incremental elastic modulus for orthotropic incompressible arteries. Journal of Biomechanics 12, 651–655. Humphrey, J.D., 2002. Cardiovascular Solid Mechanics: Cells, Tissues, and Organs. Springer, New York. Humphrey, J.D., Wilson, E., 2003. A potential role of smooth muscle tone in early hypertension: a theoretical study. Journal of Biomechanics 36, 1595–1601. Kamm, K.E., Gerthofffer, W.T., Murphy, R.A., Bohr, D.F., 1989. Mechanical properties of carotid arteries from DOCA hypertensive swine. Hypertension 13, 102–109. Kawasaki, T., Sasayama, S., Yagi, S., Asakawa, T., Hirai, T., 1987. Noninvasive assessment of the age related changes in stiffness of major branches of the human arteries. Cardiovascular Research 21, 678–687. Lee, J.C., Karpeles, L.M., Downing, S.E., 1972. Age-related changes of cardiac performance in male rats. American Journal of Physiology 222, 432–438. Levy, B.I., Michel, J.-B., Salzmann, J.-L., Azizi, M., Poitevin, P., Safar, M., Camilleri, J.-P., 1988. Effects of chronic inhibition of converting enzyme on mechanical and structural properties of arteries in rat renovascular hypertension. Circulation Research 63, 227–239. Marque, V., Kieffer, P., Atkinson, J., Lartaud-Idjouadiene, I., 1999. Elastic properties and composition of the aortic wall in old spontaneously hypertensive rats. Hypertension 34, 415–422. Matsumoto, T., Hayashi, K., 1994. Mechanical and dimensional adaptation of rat aorta to hypertension. Transactions of the ASME, Journal of Biomechanical Engineering 116, 278–283. Matsumoto, T., Hayashi, K., 1996. Stress and strain distribution in hypertensive and normotensive rat aorta considering residual strain. Transactions of the ASME, Journal of Biomechanical Engineering 118, 62–73. Nagasawa, S., Handa, H., Okumura, A., Naruo, Y., Okamoto, S., Moritake, K., Hayashi, K., 1980. Mechanical properties of human cerebral arteries. Part 2: vasospasm. Surgical Neurology 14, 285–290. Olivetti, G., Anversa, P., Melissari, M., Loud, A.V., 1980. Morphometry of medial hypertrophy in the rat thoracic aorta. Laboratory Investigation 42, 559–565. Ormsbee III, H.S., Ryan, C.F., 1973. Production of hypertension with desoxycorticosterone acetate-impregnated silicone rubber implants. Journal of Pharmacological Science 62, 255–257. Roman, M.J., Saba, P.S., Pini, R., Spitzer, M., Pickering, T.G., Rosen, S., Alderman, M.H., Devereux, R.B., 1992. Parallel cardiac and ventricular adaptation in hypertension. Circulation 86, 1909–1918. Standley, P.R., Obards, T.J., Martina, C.L., 1999. Cyclic stretch regulates autocrine IGF-I in vascular smooth muscle cells: implications in vascular hyperplasia. American Journal of Physiology 276, E697–E705. Sugimoto, T., Miyazaki, H., Hayashi, K., 2003. Age-related changes in the morphology and mechanics of arterial wall in the rat. JSME International Journal, Series C 46, 1312–1320. Sumitani, M., Cabral, A.M.S., Michelini, L.C., Krieger, E.M., 1997. In vivo adaptive response of the aorta to hypertension and aging. American Journal of Physiology 273, H96–H103. Sumpio, B.E., Banes, A.J., Linkm, G.W., Johnson Jr., G., 1988. Enhanced collagen production by smooth muscle cells during repetitive mechanical stretching. Archives of Surgery 123, 1233–1236.

ARTICLE IN PRESS K. Hayashi, T. Sugimoto / Journal of Biomechanics 40 (2007) 1583–1593 Takamizawa, K., Hayashi, K., 1987. Strain energy density function and uniform strain hypothesis for arterial mechanics. Journal of Biomechanics 20, 7–17. Wolinsky, H., 1971. Effect of hypertension and its reversal on the thoracic aorta of male and female rats: morphological and chemical studies. Circulation Research 28, 622–637.

1593

Wolinsky, H., 1972. Long-term effects of hypertension on the rat aortic wall and their relation to concurrent aging changes: morphological and chemical studies. Circulation Research 30, 301–309. Zuckerman, B.D., Yin, F.C.P., 1989. Aortic impedance and compliance in hypertensive rats. American Journal of Physiology 257, H553–H562.