Effects of Food Restrietion on Mechanical Properties ...

Journal ofGerontology: BIOLOGICAL Sc/ENCES 1999, Val. 54A, No. 10, B441-B447

Copyright 1999 by The Gerontological Society ofAmerica

Effects of Food Restrietion on Mechanical Properties of Arterial System inAdult and Middle-Aged Rats Kuo-Chu Chang, Chai-Yee Chow, Ying-I Peng, Tong-J Chen, andYuan-Feen Tsai DepartmentofPhysiology,CollegeofMedicine, National Taiwan University, Taipei, Taiwan.

I

T has been shown that food restriction without malnutrition is a reproducible and effective means of retarding the aging process in rodents (1). Food restriction increases the life span of normotensive Wistar rats and spontaneously hypertensive rats (2). Histological examination of heart, kidneys, and brain shows that freely fed hypertensive rats die of end-organ damage associated with high blood pressure. On the contrary, deaths of food-restricted hypertensive rats appear to be due to changes associated with old age, rather than specific lesions due to hypertension. Food restriction can reduce sympathetic support of blood pressure and enhance baroreflex sensitivity in normotensive rats (3,4) and spontaneously hypertensive rats (5,6). However, the effects of food restriction on the mechanical properties of the vasculature have not been fully explored. Because the mechanical properties of the vasculature can be reflected in the aortic pressure-flow relation, measurements of the aortic input impedance are regarded as essential for the analysis of arterial mechanics (7-10). The impedance is determined by the physical properties of the vasculature, including its elastic properties and size, as weil as by the viscosity and density of the fluid within (7,8). The model-based approach, such as wave transmission model based on 'l-tube topology, has been used to explain the arterial wave propagation and reflection phenomena (11,12). In mammalian animals, changes in the vascular diameter and elastic tapering play an important role in determining the wave reflection and then the vascular impedance. Consequently, the wave transmission model based on tapered T-tube topology is considered a proper method to relate pulsatile pressure and flow signals measured in the ascending aorta (13,14). This kind of analysis will provide insight into the wave transmission and reflection phenomena in the arterial system. Earlier findings on the changes of arterial blood pressure to rats showed great diversity in response to food restriction. Food

restriction tends to decrease blood pressure in hypertensive rats, whereas its effects on normotensive rats seem to be inconsistent. Thomas and colleagues (4) demonstrated that caloric restriction reduced mean arterial blood pressure of old restricted rats. By contrast, Yu (1) pointed out that food restriction did not retard the age-related increase in rats' systolic blood pressure. Recently, we reported that arterial blood pressure decreased in Long-Evans male rats with age (14). Although a fall in total peripheral resistance and aortic characteristic impedance was observed in l S-month-old rats, there was no significant change in aortic distensibility in those rats. Consequently, we determined the effects of food restriction on the mechanical properties of the vasculature in Long-Evans male rats of different ages. The exponentially tapered T-tube model was employed to relate pulsatile pressure and flow signals measured in the ascending aorta. Hemodynamic parameters such as aortic characteristic impedance, arterialload compliance, and wave transit time were derived to delineate changes in arterial mechanical properties produced by food restriction. MATERIALS AND METHODS

Subjects The specific pathogen-free, Long-Evans male rats used in this study were obtained from the colony maintained in the barrier facilities at the Animal Center of Medical College, National Taiwan University. Rats aged 12 and 18 months were referred to as adult and middle-aged rats, respectively. The adult (n = 10) and middle-aged (n = 10) ad libitum-fed groups were allowed free access to the Purina chow and water, and housed two to three per cage in a 12-h Iight-dark cycle animal room. Periodic checks of the cages and body weights ensured that the food was administered properly. Rats that started food restric-

B441

Downloaded from http://biomedgerontology.oxfordjournals.org/ at National Taiwan Univ. Hospital on November 23, 2011

The effects oJ'food restrictionon the mechanicalproperties oJthe vasculatureweredetermined in Long-Evans male rats with different ages. Rats that beganfood restrictionat the ages oJ6 months and 12 months werefed on alternatedaysfor 6 months. Rats at the ages oJ12 and 18 months werereferredto as adult and middle-agedrats and wereanesthetizedand thoracotomized. The exponentially tapered T-tube model was employed to relate pulsatile pressure and fiow signals measured in the ascending aorta. In each age group, food restriction eliciteda decrease in body weight as weil as basal heart rate but showed no significant change in cardiac output. Arterial bloodpressure, total peripheral resistance, and aorticcharacteristic impedancewerenot affectedby food restriction in middle-aged rats.However; adultfood-restricted rats exhibited lowermean arterial bloodpressure (99.1 ± 3.1 mmHg) than did adult ad libitum-fed rats (110.7 ± 3.0 mmHg). Totalperipheralresistancewasreducedfrom 0.645 ± 0.045mmHg-min-kg/ml in adult ad libitum-Jed rats to 0.492 ± 0.030 mmllg-min-kg/ml in adult food-restricted rats. Moreover; aortic characteristic impedance ofadult food-restricted rats (0.014 ± 0.001 mmHg-min-kg/ml) was lower than that ofadult ad libitum-fed rats (0.024 ± 0.002 mmHg-min-kg/ml). Neither age nor diet exerted effects on wavetransit time and produced no changes in aorticdistensibility. In conclusion, food restriction may elicitsignificantchangesin the mechanicalproperties ofboth Windkesselvessels and resistance arteriolesin adultrats,but not in middle-aged rats.

B442

CHANGETAL.

tion at the ages of 6 months and 12 months were fed on alternate days until adult (n::: 10) and middle age tn > 10), respectively. Animals at the ages of 12 and 18 months were anesthetized and thoractomized for the study of changes in the arterial mechanics caused by food restriction.

Model Parameters lnferred by the Exponentially Tapered T-tube System All model parameters were estimated and analyzed by the exponentially tapered T-tube model according to the procedure previously described (13,14). In brief, an asymmetric T-tube model with vascular nonuniformity was used to relate the pulsatile pressure and flow waves in the ascending aorta. The exponentially tapered T-tube model and its terminal complex load

HeadEnd

cOlDplexJ'"T" Q,,(t) P(t)

)~~ ~

(Zu>--c-ZcI'ap(qlJoi)m--.

~

Q(t)

QbC.t)

Complex Load

1 Complex Load i=h,b

BodyEnd Figure 1. Ttube arterial system model with asymmetrie head and body cireulation paths. Eaeh path eonsists of a nonuniform lossless transmission tube and a eomplex terminal load. Properties of eaeh tube include an input eharaeteristie impedanee Zci and a eharaeteristiedelay time 'Ti for wave transmission from one end ofthe tube to the other. The quantity exp((qOo) w---+ 00)is used to relate the eharaeteristie impedanee at the distal end of the tube to that at the inlet of the tube. Complex loads possess Ra;. CL;' and Rp ; as deseribed in text. Reprodueed with permission (13).

Downloaded from http://biomedgerontology.oxfordjournals.org/ at National Taiwan Univ. Hospital on November 23, 2011

Catheterization Each rat was anesthetized with sodium pentobarbital (35 mg/kg, i.p.). The femoral vein was cannulated for the administration of supplemental pentobarbital (30 mg/kg every 2 h). Tracheotomy was performed to provide artificial ventilation with a tidal volume of 6-8 ml/kg and respiratory rate of 50-70 breaths/min (15). The ehest was opened through the right second intercostal space. An electromagnetic flow probe (model 100 series, internal circumference 8 mm, Carolina Medical Electronics, King, NC) was placed around the ascending aorta to measure the pulsatile aortic flow. A Millar catheter with a high-fidelity pressure sensor (model SPC 320, size 2F, Millar Instruments, Houston, TX) was used to measure the pulsatile aortic pressure. Before inserting the catheter, the pressure sensor was prewarmed in 37°C saline for at least 1 h. The catheter was inserted via the isolated right carotid artery into the ascending aorta. The catheter tip position of the pressure transducer was adjusted 1-2 mm distal to the downstream edge of the electromagnetic flow probe to avoid interference with the flow measurements. After withdrawing the catheter from each rat, the catheter was reimmersed in the bath to check for baseline drift. At the end of the experiment, the pressure reading from the sensor barely submerged in the saline at atmospheric pressure and room temperature was used as the zero pressure reference (16). The electrocardiogram (ECG) oflead II was recorded with a Gould (Cleveland, OH) ECG/Biotach amplifier. The analogue waveforms were sampled at 500 Hz using a 12-bit simultaneously sampling analog-to-digital (ND) converter interfaced to a personal computer. Selection of signals (5-10 beats at steady state) was made on the basis of the following criteria: (a) recorded beats with optimal velocity profile that was characterized by a steady diastolic level, maximal systolic amplitude, and minimallate systolic negative flow; (b) beats with an RR interval, cardiac cycle length, less than 5% different from the average value for all recorded beats; and (c) exclusion of ectopic and postectopic beats. The selective beats were averaged in the time domain, using the peak R wave, the peak wave of QRS (complex for ventricular depolarization), of ECG as a fiducial point. Timing between the pulsatile pressure and flow signals, due to spatial distance between the flow probe and proximal aortic pressure transducer, was corrected by a time-domain approach, in which the foot of the pressure waveform was realigned with that of the flow (17). The resulting pressure and flow signals were subjected to the aortic input impedance analysis.

are shown in Figure 1. This model consists of two sections of different lengths. The shorter section represents the circulation of head, neck, and upper limbs (head or upper body circulation), and the longer section represents the circulation of trunk and lower limbs (body or lower body circulation). The subscripts h and b represent the head and body circulation, respectively. Properties of the t" tube, i equals h or b, include a characteristic impedance (ZcJ at the entrance of the tube and a transmission time (T) Ti is the time for a wave to propagate from one end of the tube to the other. Properties of the load are given by the load elements, which are high-frequency tube-matching impedance element (Roi ) , load compliance (Cu), and terminal resistance (RpJ. The relation between the characteristic impedance at the distal end of the tube and that at the entrance ofthe tube is quantified by the tapering index (qiOOJw---+ 00 (13). In this system, the parallel combination of the head circulation and body circulation is defined as the global circulation. Peripheral resistance of the global circulation (Rp ) was calculated as mean pressure divided by mean flow, The terminal resistance of each tube was computed usingthe upper-to-Iower body resistance ratio of 2.33 for normotensive state and 3.17 for hypotensive state (14). With setting (qhOOh) w ---+ 00 equal to (qbOOb) w ---+ 00, the model parameters, such as aortic characteristic impedance (Zc)' vascular tapering index (qOO) w---+ 00, wave transit time (Th,Tb), and arterialload compliance (Clh,Clh ) , were estimated using the equations developed for the exponentially tapered T-tube model. The characteristic impedance at the inlet of each tube was calculated using the upper-to-Iower body characteristic impedance ratio of 1.14 for normotensive condition and 1.20 for hypotensive condition (14). The high-frequency tube-matching impe-

B443

VASCUIAR DYNAMIC CHANGES BY FOOD RESTRICTION

systemic arterial system with the exponentially tapered T-tube model.

Statistics Results are expressed as means ± SE. Because cardiac output is significantly related to body shape, this variable was normalized to body weight when comparison was made between ad libitum-fed and food-restricted rats. Other hemodynamic variables derived from cardiac output were also normalized to body weight to detect the effects of food restriction on these parameters. A two-way analysis of variance (ANOVA) was employed to determine the effects of food restriction on the mechanical properties of the vasculature in adult and middle-aged rats. Simple effects analysis was used when significant interaction between diet and age occurred. Differences among means within levels of a factor were determined by Tukey's honestly significant difference (HSD). Significant differences were assumed at the level of p < .05. RESULTS

The effects of food restriction and age on body weight, basal heart rate, and arterial blood pressure, as weIl as cardiac output, are shown in Table 2. Food restriction was associated with a decrease in body weight in each age group. There was a significant (p < .01) interaction between age and diet in their effects on body weight, reflected in the fact that adult rats showed a significantly lower body weight than did middle-aged rats only in the foodrestrictedcondition, not in the ad lib condition. Basal heart rate was decreased by both age and dietary restriction. No interaction between the effects of age and diet on basal heart rate was detected.

Table 1. Indexes of Fitnessfor the Exponentially TaperedT-tubeModel-Dcrived Data to AdultAd Libiturn and RestrictedRats, and Middle-agedAd Libiturnand RestrictedRats

Rats

Adult Ad Libitum (A) (n = 10)

Adult Restricted (B) (n = 10)

Middle-aged Ad Libitum (C) (n = 10)

Middle-aged Restricted (D) (n = 10)

5.86± .57 .95 ± .22 .9899± ,(XH6

6.33 ± .39 .94 ±.18 .9904 ± .0014

6.19 ± .28 .99± .17 .9896 ± .0014

6.03 ± .36 .92± .20 .9878 ± .0014

Nonnalized root-mean-square error (e*), X 10-4 Standard error of the estimate (%) Coefficient of determination

Notes: All values are expressed as means ± SE. These indexes are the linear regression parameters ofthe model output pressure over the measured pressure.

Table2. Effectsof Diet and Age on Basic Hernodynamic Data Measuredand Calculatedin AdultAd Libiturn and RestrictedRats, and Middle-agedAd Libiturnand RestrictedRats Adult (12 months)

Rats Body weight Heartrate Systolic pressure Diasto1ic pressure Mean pressure Cardiac output Stroke vo1ume Periphera1 resistance

Midd1e-aged (18 months)

Ad Libiturn (A) (n = 10)

Rrestricted (B) (n = 10)

Ad Libitum (C) (n = 10)

Restricted (D) (n = 10)

AvsB

503.0± 15.3 346.7 ± 8.6 127.2 ±2.6 93.7 ± 3.4 110.7 ± 3.0 179.3 ± 13.9 .517 ± .037 .645 ± .045

301.5 ± 10.1 308.7 ± 10.5 108.7 ±2.9 87.5 ± 3.7 99.1 ± 3.1 205.7 ± 12.0 .673 ± .048 .496 ± .030

508.2± 17.1 312.6± 12.1 115.8 ± 1.8 83.6 ± 2.1 100.8 ± 2.1 209.5 ± 10.1 .672 ± .024 .492± .026

392.7 ± 5.4 276.8 ±9.9 114.1 ± 2.8 85.1 ± 3.0 98.2 ± 2.5 224.8 ± 16.9 .808 ± .046 .460±.037