- Email: [email protected]
Volume overload hypertrophy elicited by cold and its effects on myocardial capillarity
1
Respiration Physiology (1985) 59, 1-14 Elsevier
VOLUME OVERLOAD HYPERTROPHY ELICITED BY COLD AND ITS EFFECTS ON MYOCARDIAL CAPILLARITY
SUSAN R. KAYAR and NATALIO BANCHERO Department of Physiology, University of Colorado School of Medicine, Denver, CO 80262, U.S.A.
Abstract. Capillarity and fiber cross-sectional areas were measured in the hearts of guinea pigs exposed to
cold early during growth. Twelve male guinea pigs were kept at 5 + 1 °C for 4-18 weeks. Hearts were perfusion fixed via the aorta with a 2.5 % glutaraldehyde, 1% formaldehyde-buffered solution, blocks were cut from left (LV) and right (RV) ventricles, post-fixed in OsO4, dehydrated and embedded in Spurr medium. Blocks were cut transversely to fiber orientation, 0.5/zm thick, stained with Toluidine Blue and photographed at 400 ×. Heart weights, number and location of capillaries and fiber cross-sectional areas (FCSA) were measured in cold-acclimated animals and in normothermic controls. Growth rates for all guinea pigs were similar. Acclimation to cold caused modest LV and RV hypertrophy. The greater LV weight seemed due to longer fibers of normal FCSA, whereas the greater RV weight was due to larger FCSA. Capillary density, capillary-to-fiber ratio and number of capillaries around the fibers were similar in the two groups of animals. Mean and maximal diffusion distances in cold-acclimated animals were not different from controls. Thus the myocardial hypertrophy induced by chronic volume overload was fully compensated for by increases in capillarity commensurate with increases in fiber girth. Acclimation Capillary Cold
Diffusion Guinea pig Heart
Hype~'trophy Myocardium Temperature
Exposure to a cold environment causes a sustained elevation in oxygen consumption and cardiac output and the latter in turn induces modest bilateral cardiac hypertrophy (Bui and Banchero, 1980). Rodents reared in a cold environment show substantial increases in the capillarity of skeletal muscle (Heroux and St. Pierre, 1957; Sillau et al., 1981). This study examines the progression of myocardial hypertrophy in growing guinea pigs exposed continuously to a 5 °C environment and its effects on myocardial capillarity. Myocardial hypertrophy induced either by sustained volume overload as in cold exposure or intermittent volume overload as in exercise training, typically is not Accepted for publication 8 October 1984 0034-5687/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
2
S.R. KAYAR AND N. BANCHERO
associated with large changes in ventricular mass. In studies of left ventricular hypertrophy induced by aorto-caval fistula (Rakusan et al., 1980) and treadmill exercise (Tharp and Wagner, 1982), myofibers increased in girth but there were no increases in the ratio of capillary density to fiber density in the hypertrophied rat hearts compared to age-matched controls. This indicated that no compensatory capillary proliferation had occurred in these rats. Heroux and St. Pierre (1957) found no changes in myocardial capillary density in rats exposed to a 6 °C environment for 4 weeks, although changes in myofiber size or heart weight were not reported. There is evidence, however, that myocardial hypertrophy of any origin induced in young, growing animals can be accompanied by significant compensatory increases in capillarity (Rakusan, 1971; Tomanek etal., 1982). Since it is known that in muscles of normal growing animals fiber size is the most important factor determining capillarity and oxygen diffusion distances (Sillau and Banchero, 1978; Hudlicka, 1982), then it becomes particularly important to analyze the capillarity in myocardial hypertrophy as a function of the size of the myofibers. In this study, weanling guinea pigs were maintained in a 5 °C environment for 4-18 weeks. Blood samples were analyzed weekly for hematocrit and hemoglobin concentration. Animals were sacrificed after 4, 6, 8, 13, 16 or 18 weeks of cold exposure to determine heart weights, and heart tissue samples were morphometrically analyzed for fibers and capillaries. From this analysis we offer some speculations on the factors involved in capillary proliferation.
Materials and methods Guinea pigs (Cavia porcellus, Hartley strain) were obtained from Camm Research laboratories (Wayne, New Jersey). Control animals (n = 18)which were of both genders were obtained at a variety of body weights from weanling (250-300 g) to adult dimensions (1000 g). Since no significant differences (P > 0.20) were found in heart weight or body weight that were attributable to gender, data from males and females were pooled. Experimental animals (n = 12) which were all males were obtained at a body weight of 250-300 g. One week after their arrival in the laboratory, the experimental animals were placed in a chamber in which the temperature was lowered to 5 + 1 ° C. A small volume of air was blown continuously into one end of the chamber to maintain 0 2 and CO2 at normal levels. Animals were housed within the chamber in cages 20 x 40 x 20 cm in dimension, two young animals to a cage, or one animal over 700 g to a cage. Food and water were available to the animals ad libitum, and cage litter was changed at least every other day, necessitating regular, brief exposure of animals to room temperature. On the average, animals were exposed to room temperature for less than 8 h per week. Timed lamps provided a 12 h light/12 h dark cycle. Growth rate of the animals in the cold was similar to controls (P > 0.50). At weekly intervals, experimental animals were removed from the cold chamber and blood samples were drawn for hematocrit (Hct) and hemoglobin concentration ([Hb])
MYOCARDIAL CAPILLARITY IN COLD EXPOSURE
3
determinations. Blood was obtained by clipping a toenail near its base, and if necessary blood flow was later stanched by applying a drop of collodion to the toenail. Hematocrit was determined from duplicate blood samples of approximately 40/~1 each, centrifuged in capillary tubes for 6 rain. Hemoglobin concentration was determined from duplicate samples of 20 #1 each by the cyanmethemoglobin spectrophotometric method. At sacrifice the guinea pigs were anesthetized with sodium pentobarbital, 35 mg/kg IP. A polyethylene catheter was inserted into the left carotid artery and advanced to the arch of the aorta, and a second catheter was placed in the fight jugular vein and also advanced toward the heart. One milliliter of sodium heparin (1000 units) was introduced slowly via the jugular catheter. A solution of isotonic saline with 0.5~o heparin and 0.5~o lidocaine was then perfused via the carotid line, at a pressure of approximately 100 torr, with the jugular catheter opened as a drain. After 20 min of saline perfusion, most blood had drained and the animal's heart had stopped. A solution of 2.5 ~/o glutaraldehyde and 1~o formaldehyde in 0.1 M phosphate buffer, pH 7.35 was then perfused via the carotid line, also at a pressure of 100 torr. This fixative was chosen for its reputedly superior fixing properties over either a glutaraldehyde or a formaldehyde solution alone (Karnovsky, 1965). Tissues fixed with this solution appeared well-preserved under the electron microscope, with no obvious swelling or shrinkage of myofibrillar mitochondria. No correction factor for shrinkage was calculated since true anatomical dimensions were unknown. Myofibers appeared uncontracted when viewed longitudinally under the electron microscope. After 20 rain of perfusion with the fixative, the heart was removed, blotted dry and weighed to the nearest milligram while moist. The atria were discarded, right and left ventricular walls were cut free of the septum and weighed. Each ventricular wall was cut into four approximately equal bands from base to apex. Several blocks 1-2 mm to a side were cut from the central portion of each band. The blocks were then further fLxed by immersion in the glutaraldehyde-formaldehyde solution for 1 h, rinsed in 2 ~ buffered sucrose for 45 min, and postfLxed in 1~o buffered OsO4 pH 7.35 for 1 h. Following a rinse in distilled water, the blocks were stained with 2~/o uranyl acetate, rinsed again, dehydrated with acetone and embedded in Spurr low-viscosity medium. At the time of embedding, tissue blocks were examined carefully under a dissecting microscope and two blocks from each band of a ventricle were selected and embedded such that primary fiber orientation was vertical in the epicardial region of one block and in the endocardial region of the other block. Thus there were four epicardial and four endocardial samples from each ventricle, for a total of 16 tissue samples per animal. Tissue blocks were cut with a Sorvall II Ultramierotome to a thickness of approximately 0.5 ~tm. Tissue sections were stained with Toluidine Blue and examined and photographed on color transparency fire at 400 x under the light microscope. One or two photographs per section were taken only from regions in which fibers and capillaries appeared to be cut in cross-section, with capillary profiles circular and little or no signs of striations in the fibers. Photographic fields were also selected for minimal interstitial gaps between fibers (large gaps presumably being signs of mechanical damage to tissues during handling or fixation), and for the absence if possible of blood vessels larger in
4
S.R. KAYAR AND N. BANCHERO
diameter than capillary dimensions (approximately 18/~m). From endocardial samples, photographs were not taken from regions that were clearly papillary muscles. The total epicardial or endocardial sample from a given ventricle could thus contain up to 8 fields and was discarded if fewer than 3 suitable fields could be obtained. In most cases there were 20-25 fields sampled per animal. Analysis of the tissue photographs was performed with the aid of an x-y digitizer (Datatizer, GTCO Corp., Rockville, MD) on line with a Nova 4 Computer. Fiber cross-sectional areas (FCSA) were calculated by computer integration after outlining fiber profiles; fiber perimeters and maximal fiber diameters were calculated simultaneously. Fiber density was estimated from FCSA and the average area of a frame occupied by capillaries and interstitium, the latter being estimated by point-counting the first 200 frames. The number of capillaries immediately around the border of each fiber (CAF) was also recorded. The total number of capillaries and the x-y coordinates of their locations were scored for calculation of capillary density (CD) and capillary array. Data on the fibers and capillaries were also combined to calculated capillary density to fiber density ratios (C'F). The distribution of capillaries relative to each other was determined by the closest individual method (Kayar et al., 1982). A computer-stored uniform grid of 25 points was placed over a photograph of the tissues, and the distance from each grid point to the nearest capillary was recorded. A mathematical analysis of the distribution of these distance measurements yields what Kayar et al. (1982) have defined as the mean (R) and maximal (R95) oxygen diffusion distances in these tissues. Comparison of these R and R95 values to values calculated from the CD were used to indicate which of several possible geometric models best applied to the arrangement of capillaries in these tissues. Statistical analysis was performed by Student's t-test and linear regression. Regression lines were compared using Fieller's theorem (Zerbe et aL, 1982) when slopes were unequal; otherwise, y-intercepts were compared by a t-test.
Results
Cold acclimation caused a significant (P < 0.001) hypertrophy of both left (LV) and right (RV) ventricles (fig. 1). However, the degree of this hypertrophy was modest. After 4 weeks of exposure, cold-acclimated animals had ventricular weights approximately 50~ greater than those of control guinea pigs of the same body weight and after 18 weeks, ventricles of cold-acclimated animals were approximately 15 ~o heavier than in controls. In the RV of control animals, FCSA increased linearly with ventricular weight (fig. 2). The FCSA in the RV of cold-acclimated animals was similar (P > 0.50) to that of controls of the same ventricular weight, although the relationship between FCSA and ventricular weight was not significant. In the LV, FCSA increased linearly with ventricular weight for control and cold-acclimated animals. The slopes of the regression were the same (P < 0.001), but the y-intercept for the regression for the cold animals
MYOCARDIAL CAPILLARITY IN COLD EXPOSURE
3.4
y : 1.08+ (231X 10-3)r,
•
f c°'° \ :3/" r=0.88
3.0
o
5
~
o
.o X "
.,.//
2.6
0 // /"~ •
0
, ,,,o-'~ o
2.2
1.8 1.4
•
~
//~,;
o
,,y.
,-"
-.
, .-
~
LControl
.,,
r= 0.97
• ~'0,-
I.C %/
Whole HeQrt
I
I
I
I -
Cold o 3 0 0 1.2- y=O50+(O.85xlO" )x ~•'" .w'" •~ r=0.79 ~ ~.'" 5 . / ~ ' r " •
)
t.o~-
0.8 0.6 0.4
o
.o~..-
"i
•~Control y=O.OB.(H4x10-')x
.~... J•~
r=0.95
Left Ventricle I
I
I
Cold o.6
y= o.zz+ (oszxlo-~)x .~-.~_°'~r.~" r=0.93
0.4. 02
"~
oo'--" . ~
•~.=~
~ •
• -
I
400
•
[Control
y : 0 0 3 + ( 0 . 6 2 x lO-')x
Right Ventricle 0
1 ,.
o
r=0.85 i
I
600 800 Body Weight (g)
I
I000
Fig. 1. Whole heart weight and left and right ventricular weights v s body weight for control guinea pigs and for guinea pigs reared in a 5 °C environment. Control group (O) in this case includes data from 36 animals.
was significantly lower (P < 0.01) than for controls (fig. 2). Because there was LV hypertrophy this smaller FCSA at the same ventricular weight may indicate a disproportionate increase in the length of the fibers in the LV of the cold-acclimated animals. Capillary density in both ventricles of control guinea pigs decreased with increasing FCSA (fig. 3). In the RV of cold-acclimated animals, mean CD was similar (P > 0.20) to that of controls but there was no significant correlation between CD and FCSA. In the LV of cold-acclimated animals, there was a significant regression between CD and FCSA, and this relationship was similar (P > 0.10) to controls. Both C : F and CAF increased with FCSA in both ventricles of control guinea pigs (figs. 4 and 5). Data from cold-acclimated guinea pigs were similar to those measured in controls ( C : F left ventricle, P > 0 . 1 0 ; C : F right ventricle, P > 0 . 1 0 ; CAF left ventricle, P > 0.50; CAF right ventricle, P > 0.10), indicating that the total number of
6
S.R. KAYAR AND N. BANCHERO 400
Guineo Pig M y o c a r d l a m Left Venlrlcle
E
:L
~3oo
c o
•
!
.
200
A I •
0
o
•
• E p i c a r d l u m , Conlrols • Endocaromm,Controls ~ Eplcordium,Cold [] Endocclrd,u m, Cold
I00
~
y=59.00+170.37X
\
r= 0.81
Controls ~ y = 8 8 . 6 3 + 170.67x r = 0.80 0 -¢/
I
l
I
I
I
0.4
0.6
0.8
1.0
1.2
..... l
1.4
400 Right V e n t r i c l e
E
3OO c O
T-
i
IOOL? •
01"//120 •
Controls y=90.Ol+207.12x r : 0.78 I
0.4
Ventricular
l
0.6
1
0.8
1.0
Weight (g)
Fig. 2. Fiber cross-sectional area v s ventricular weight for the left (upper panel) and right (lower panel) ventricles of control guinea pigs and of guinea pigs reared in a 5 °C environment.
capillaries had increased during growth only as a function of increasing myofiber girth, but that cold had no direct effect on capillarity. Likewise R and R95 measured directly in cold-acclimated animals were not different from controls, and could be approximated + 0.5 #m for R as 0.40(CD- 1/2) and + 2 #m for R95 as 0.71(CD- 1/2). Hematocrit and [Hb] were elevated modestly in the cold-acclimated animals. Mean Hct and [Hb] of the young guinea pigs before cold exposure were 41.7 +_ 0 . 5 ~ and
M Y O C A R D I A L CAPILLARITY IN COLD E X P O S U R E
4000!
• •
/---Cold
z~ / • /y=2297.76 p,/
7
+ 154022 --'-E'--
r =0.48
E E 3000•
•
-
,,,_~l
e@
Pig Myocordium Guineo Left Ventricle & Epicordium, Contr01s • Endocerdium,ControLs Epicardiu m, Cold oEndocardium,Cold
2000 Q.
0
E
D
4000
-
3000
-
I I00
•
/^
. I
I#~
x
r = O. 63
I
I 300
J 400
]L=l~,i~ •
¢..
J87334
y= 1929.53+
200
•
•
L- L; 0 I1 t r o l $
•
E~
A•
>,
Controls ~)000
--
Right Ventr,cle
"7~
y = 2451.72'
98781 - -
r = 0.50
X
O 0
0
I i00 Fiber Cross
I 200 Sectional
I 300 Area(/~m
I 400 2)
Fig. 3. Capillarydensity v s fiber cross-sectionalarea for left (upper panel) and right (lower panel) ventricles of control guinea pigs and of guinea pigs reared in a 5 ° C environment. 13.1 + 0.2 g/100 ml blood, respectively. These values increased after the In'st week and were always between 47 and 50~o (mean = 47.6~o) and between 14 and 16 g/100 ml (mean = 14.9), respectively, in all subsequent weeks of cold exposure.
~]
S.R. KAYAR AND N. BANCHERO 1 . 4 --
0
1.2 --
.e-
0 n-
1.0-
Guinea Pig M y o c o r d i u m Left V e n t r i c l e A E p i c a r d i u m Controls
[3 & / 5 %'.," .~" /
• Endocard,um,Controls ~ Epicardium,Cold u Endocardium,Cold
./ /
•
Z~~m
.o
0.80 I--
0.6-
.,,~,/~'t ~ r / J ~ °AIlt
Lcontro ' y= 0 . 3 0 + ( 2 . 5 9 x IO-')x
\Cold
r = O. 83
t~
0
m
-& 0 . 4 -
y=O
°
"
19+(3.43x10
-3)x
r=0.86 0.2C
I I00
0
I 200
I 300
J 400
Right Ventricle
o
1.2-
O
n-
1.0 -
.9_ "
0.8-
O I--
o = ";, °
(..)
A e~m/~r'"
0.6-
/
"co.,,ol
/ 0.4
0.2-
-
/'
/ LC01d
y--0.18+ r =0.85
*
(3.33x10-')x
y=0.14+(3.88+10-3)x
r=077 I
O0
I
I
J
I00 200 300 400 Fiber Cross Sectionol Are(](/~m 2)
Fig. 4. Capillary-to-fiber ratio vs fiber cross-sectional area for left (upper panel) and right (lower panel) ventricles of control guinea pigs and of guinea pigs reared in a 5 °C environment.
9
MYOCARDIAL CAPILLARITY IN COLD EXPOSURE
M Q)
Guinea Pi 9 M y o c a r d i u m Left V e n t r i c l e • Epicardium, Controls • Endocardium,Controls tx E p i e a r d i u m , C o l d
4
t..
t,..
a
e
!
U A
o Endocordium,Cold
.Q
•
,,~ / ~ i y
o.b. 0
o
&~l / / 3 ~
o
•o :0~
o
~i-~_~;~ -m=~ T | =
E
-1
Cold ~
z
m
y= 1.49+
°-.
\
C ont.r01s ~ .. 1 . 6 5 + ((6 . 0. 2 x 1 0 "s ) x yv= 1.6 r= 0 . 7 9
(6.73xlO'n)x
r=O.81
0
I , I00
Right Ventricle
=
Q
4
I 200
I 300
J 400
Cold y- 0.8 :)+(1.08x lO-2)x r=0.77
"7. O b.
oo
3
q-'10 0 @ 0
E< z=
2 -
~ ~ o n t r o l .=ill y= 1.32+ ( 7 . 6 3 X l O - S ) X r =0.84 I I00 200 Fiber Cross Sectional L
0
I 300 Areo(~m
I
400 2}
Fig. 5. Number of capillaries around a fiber vs fiber cross-sectional area for left (upper panel) and right (lower panel) ventricles of control guinea pigs and of guinea pigs reared in a 5 °C environment.
Discussion
In studies of cardiac hypertrophy induced by pressure-overload such as in chronic hypoxic exposure or pulmonary artery banding, a number of investigators have concluded that the increased myofibrillar mass is not compensated for by an increase in the number of capillaries, and the diffusion distances for oxygen lengthen (Roberts
10
S.R. KAYAR AND N. BANCHERO
and Wearn, 1941; Rakusan, 1971; Turek etaL, 1972; Murray etal., 1979; Rakusan et aL, 1981). In our own study of right ventricular hypertrophy induced by hypoxia we found significant increases in capillarity only in young animals, but these differences disappeared as the myofibers grew past the maximal normal cross-sectional area (Kayar and Banchero, 1984). Right ventricular hypertrophy due to pressure-overload can be of massive proportions (greater than 200 ~o) and in such cases cardiac performance may be severely impaired (Bui and Banchero, 1980). In the present experiment chronic cold exposure, which causes a two-fold increase in Vo~ and a sustained elevation of cardiac output, increases the work of the heart. This leads to a significant but modest degree of hypertrophy of both ventricles. Hypertrophy of the LV of cold-acclimated guinea pigs does not appear to be due to proportionate increases in FCSA and fiber length since the ventricular weight-FCSA relationship for cold-acclimated animals was significantly different from that of controls. The explanation we favor is that hypertrophy of the LV in the cold-acclimated guinea pigs was due to disproportionately greater lengthening of the LV fibers, while the increase in FCSA relative to body weight was similar to that in controls (fig. 6). Astorri et al. (1971) have found that fiber length changes and width changes in the hypertrophy of human hearts can in fact occur at different times, with length increasing first. If one assumes that tissue weight (W) is proportional to tissue volume (assuming constant tissue density) and that changes in the weight of a ventricle are accounted for by proportional increases in FCSA and fiber length (L) then: AL
: ( / ~ W ) 1/3
AFCSA
and
= ( A L ) 2 = ( A W ) 2/3
If the weight gain in normal growth of the LV is due to proportional changes in FCSA and length, then as the LV increases from 0.4 to 1.2 g, or by a factor of 3, we would fi
400
P,g Myocardium Left Ventricle
Guinea
• Epicardium, Controls m F n d o c a rd t u r n , C o n t r o l s z~ E p i c a r d l u m , C o l d a Endocardmm, Cold
g 300 o c o lJ q)
A
0')
200 i/1 o
,3 .1o
i2_
I00
/1 200
I 400
I 600 Body
I 800
I . I000
W e i g h t {g}
Fig. 6. Relationship between fiber cross-sectional area and body weight for the left ventricle of control guinea pigs and of guinea pigs reared in a 5 °C environment.
MYOCARDIAL CAPILLARITY IN COLD EXPOSURE
11
expect an increase in FCSA of 3 2/3 = 2.1. The actual change in FCSA measured in controls was from 157 to 294 #m2 (a factor of 1.9), which is in good agreement with this prediction. Fiber length change in controls should thus be by a factor of 31/3 = 1.44. However, in the youngest cold-acclimated animals mean LV weight was 0.84 g while LV weight was 0.54 g in controls of the same body weight and age. This is an increase in ventricular weight by a factor of 1.56. If FCSA and length were increasing proportionately in the cold-induced hypertrophy of the LV, then FCSA should have been 1.562/3 = 1.35 times greater in the cold-acclimated animals. Instead, FCSA for controls of LV weight = 0.54 g was 181 pan2 and FCSA for cold-acclimated guinea pigs of LV weight = 0.84 g was 202/an 2, an increase by a factor of only 1.12. This suggests that actual fiber length change due to hypertrophy was by a factor of 1.34 (1.56/1.12) rather than by a factor of 1.16 (1.561/3) if FCSA and length changes were proportionate to each other. By the same reasoning, in the oldest cold-acclimated animals, LV weight was 1.28 g while it was 1.11 g in controls of the same body weight, which is a factor of 1.16. The FCSA in the cold-acclimated animals should have been greater by a factor of 1.10 (1.162/3) if fiber growth in length and area were proportionate, but since FCSAs were nearly identical at these respective LV weights, then actual fiber length must have been greater by a factor of 1.16/1 = 1.16, rather than by 1.05 (1.161/3). Thus we conclude that the volume overload to the LV resulted in fibers that were of normal FCSA but were probably longer than in controls of the same body weight. However, an alternative explanation for the heavier LVs in the cold-acclimated animals which we cannot rule out is that a small degree of fiber hyperplasia was present in this ventricle. Hypertrophy in the RV is entirely attributable to increases in the size of individual fibers, rather than by hyperplasia. This was demonstrated by counting the number of fiber profiles in a complete cross-section of the RV free wall. In the youngest control animal, the oldest control animal, and the oldest cold-acclimated animal, there were 35-50 fibers from the epicardial surface to the endocardial surface. The same conclusion may be drawn indirectly from the unique relationship between FCSA and ventricular weight. However normal fiber growth in the RV may not be by proportionate changes in length and FCSA as it is in the control LV. As RV weight increases from 0.2 to 0.7 g, (factor of 3.5) there should be an increase in FCSA of 2.31 (3.52/3). Actual change in FCSA measured for this change in ventricular weight was from 131 to 235 #m 2, (a factor of 1.79). This is in poor agreement with our prediction and suggests that some disproportionate length increases may be occurring. Fiber length change should account for the rest of the increase in ventricular weight which should be by a factor of 1.96 (3.5/1.79). Thus during both the normal growth of these guinea pigs from weanlings to adults and also during the modest hypertrophy from cold exposure for up to 18 weeks, fibers in the RV become nearly double in FCSA and should also double in length. The difference between the changes in fiber length and FCSA in the RV and the LV is undoubtedly a reflection of the different modes of contraction, work loads and wall tensions in the two ventricles. In the normal cardiac cycle, contraction of the LV reduces its lateral diameter with relatively tittle shortening of its vertical (base-apex) axis,
12
S.R. KAYAR AND N. BANCHERO
whereas contraction of the RV involves ventricular shortening with relatively little pulling of the free wail toward the septum. The RV is a volume pump which can increase its volume output easily, whereas the LV performs more as a pressure pump which can respond more easily to increases in outflow resistance (Brecher and Gailetti, 1963). Thus in chronic cold exposure and volume overload, the RV evidently can respond to the greater work load by fiber hypertrophy in the same geometric proportions of fiber length and FCSA that it uses in normal growth of the animal. The LV evidently responds to the greater blood volume in its chamber by increasing fiber length more than that in normal growth, which considering the primary direction of orientation of the major fiber bands, would generally increase the length of the LV and not increase wall thickness. This should distribute the greater blood volume in a manner that would not change wall tension. Capillarity in both ventricles relative to FCSA was entirely normal in cold-acclimated guinea pigs. Neither CD, C : F , CAF nor directly-measured diffusion distances were different from controls. Thus the modest hypertrophy of both ventricles was fully compensated for by increases in the number of capillaries relative to the larger fiber areas. However, we have calculated that there is probably a disproportionate lengthening of fibers that accounts for the increased LV weight in the cold-acclimated animals. Ventricular weight and also body weight, however, are poor independent variables in regressions with CD, CAF and C : F, which makes it difficult to assess these relationships between capillarity and body mass. This is not surprising if one takes into consideration that the radial component of oxygen diffusion outward from capillaries to tissues is more important to oxygen delivery rate than capillary length or total tissue volume. We believe from the magnitude of the FCSA and length differences calculated above that capillarity in the LVs of the cold-acclimated guinea pigs is not impaired. It is not at all certain how an increase in fiber length of 10-30~ more than that contributed by normal growth would affect capillary length. If one assumes that this causes an equal lengthening of the capillaries, then from the Krogh equation, Akmai etal. (1978) calculated that a modest increase in the total length of a capillary should affect blood Po2 at the venous end of the capillary by not more than a few torr. In contrast to the heart, the gastrocnemius and soleus muscles of growing guinea pigs acclimated to a 5 °C environment had considerably higher capillarity than normothermic controls (Sillau et aL, 1980). We hypothesize that the difference in the response of the capillarity of the heart vs the skeletal muscles of the hindlimb is a function of tissue Po2. Low tissue Po2 is the promoter of angiogenesis in wounds (Remensnyder and Majno, 1968) and there is reasonable evidence to suggest that hypoxia is also important in the elicitation of endothelial growth in cultures. In cold exposure, the hindlimbs themselves are cold, as would be the blood perfusing the muscles. This would result in a leftward shift in the Hb-O 2 dissociation curve. While such a shift does not affect arterial oxygen saturation, it would result in a lower venous Po2, even for a normal arterio-venous oxygen difference. If, for example, blood temperature fell from 38 to 30 ° C, at an oxygen extraction of 5 vol. ~o, venous Po2 would be approximately 13 torr lower at 30 °C (Banchero et al., 1984). Thus hindlimb tissue Po: may be significantly
MYOCARDIAL CAPILLARITY IN COLD EXPOSURE
13
lower in the cold-acclimated guinea pigs than in normal, thereby stimulating the capillary angiogenesis reported by Sillau et al. (1980) and by Heroux and St. Pierre (1957). On the other hand, environmental temperature does not affect the core temperature of guinea pigs in a cold environment (Banchero et al., 1981). The arterio-venous oxygen difference across the heart is about 11 vol. Y/o,which is more than twice the oxygen difference for the entire body. Thus additional oxygen requirements of the myocardium above resting values must be satisfied mainly by increases in coronary blood flow (CBF) since Vo2 = CBF (a - v 02 difference). The doubling of cardiac output observed in chronic cold exposure requires additional cardiac work, and therefore also requires sustained increases in CBF, but should not significantly change coronary sinus Po~ or myocardial tissue Po~. We believe that it is this constancy of the myocardial tissue Po2 that led to our finding no changes in myocardial capillarity as a function of cold exposure. A number of other studies of capillarity (for review see Hudlicka, 1982) have also found indications that tissue Po2 is involved in the elicitation of capillary proliferation. However, because blood flow rate is usually also increased in these situations some investigators believe that the mechanical effect of increased blood flow is important in capillary proliferation. It has not been clear whether it is either decreased tissue Po2 or increased blood flow alone, or these two factors combined that are sufficient for capillary proliferation. In our experiment, capillarity and oxygen diffusion distances were unchanged from values expected for myocardial fibers of a given FCSA, in a situation in which blood flow was chronically elevated, but tissue Po~ should not have been affected.
Acknowledgements This work was supported by NIH HL-18145 and HL 28849. S.R.K. was supported by NIH Postdoctoral Fellowship HL 06527.
References Akmal, K., D.F. Bruley and N. Banchero (1978). Multi-capillary model for oxygen transport to skeletal muscle. In: Oxygen Transport to Tissue. III, edited by I.A. Silver, M. Erecinska and H. I. Bicher. New York, Plenum Publishing Co., pp. 139-147. Astorri, E., A. Chizzola, O. Visioli, P. Anversa, G. Olivetti and L. Vitali-Mazza (1971). Right ventricular hypertrophy - a cytometric study on 55 human hearts. J. Mol. Cell. Cardiol. 2:99-110. B anchero, N., A.J. Lechner, L. Aquin, A.H. Sillau and M.V. Bui (1981 ). Resistance to hypothermia during chronic cold exposure. Acta Univ. Carolinae BioL 1979: 233-235. Banchero, N., S.R. Kayar and C. I. Blake (1984). Factors involved in the ehcitation of increased capillarity in muscle. Fed. Proc. 43: 1006. Brecher, G.A. and P.M. GaUetti (1963). Functional anatomy of cardiac pumping. Section 2. Circulation Vol. II. In: Handbook of Physiology. Washington, DC, American Physiological Society, pp. 790-791. Bui, M.V. and N. Banchero (1980). Effects of chronic exposure to cold or hypoxia on ventricular weights and ventricular myoglobin concentrations in guinea pigs during growth. Pfliigers Arch. 385: 155-160.
14
S.R. KAYAR AND N. BANCHERO
Heroux, O. and J. St. Pierre (1957). Effect of cold acclimation on vascularization of ears, heart, liver and muscles of white rats. Am. J. Physiol. 188: 163-168. Hudlicka, O. (1982). Growth of capillaries in skeletal and cardiac muscle. Circ. Res. 50: 451-461. Karnovsky, M.J. (1965). A formaldehyde-glutaraldehyde fixative of high osmolarity for use in electron microscopy. J. Cell Biol. 27: 137A-138A. Kayar, S.R., P.G. Archer, A.J. Lechner and N. Banchero (1982). The closest individual method in the analysis of the distribution of capillaries. Microvasc. Res. 24: 326-341. Kayar, S.R. and N. Banchero (1984). Myocardial capillarity in acclimation to hypoxia. Pfliigers Arch. (in press). Murray, P.A., H. Baiz, M.C. Fishbein and S.F. Vatner (1979). Effects of experimental right ventricular hypertrophy on myocardial blood flow in conscious dogs. J. Clin. Invest. 64: 421-427. Rakusan, K. (1971). Quantitative morphology of capillaries of the heart. Methods Achier. Exp. Pathol. 5: 272-286. Rakusan, K., J. Morafiec and P.Y. Hatt (1980). Regional capillary supply in the normal and hypertrophied rat heart. Microvasc. Res. 20:319-326. Rakusan, K., Z. Turek and F. Kreuzer (1981). Myocardial capillaries in guinea pigs native to high altitude (Junin, Peru, 4,105 m). Pfliigers Arch. 391: 22-24. Remensnyder, J. P. and G. Majno (1968). Oxygen gradients in healing wounds. Am. J. Pathol. 52: 301-323. Roberts, J.T. and J.T. Wearn (1941). Quantitative changes in the capillary-muscle relationship in human hearts during normal growth and hypertrophy. Am. Heart J. 21: 617-633. Sillau, A.H. and N. Banchero (1978). Skeletal muscle fiber size and capillarity. Proc. Soc. Exp. Biol. Med. 158: 288-291. Sillau, A.H., L. Aquin, A.J. Lechner, M.V. Bui and N. Banchero (1980). Increased capillary supply in skeletal muscle of guinea pigs acclimated to cold. Respir. Physiol. 42: 233-245. Tharp, G.D. and C.T. Wagner (1982). Chronic exercise and cardiac vascularization. Eur. J. Appl. Physiol. 48: 97-104. Tomanek, R. J., J. C. Searls and P.A. Lachenbruch (1982). Quantitative changes in the capillary bed during developing, peak, and stabilized cardiac hypertrophy in the spontaneously hypertensive rat. Circ. Res. 51 : 295-304. Turek, Z., M. Grandtner and F. Kreuzer (1972). Cardiac hypertrophy, capillary and muscle fiber density, muscle fiber diameter, capillary radius and diffusion distance in the myocardium of growing rats adapted to a simulated altitude of 3500 m. Pfliigers Arch. 335: 19-28. Zerbe, G.O., P.G. Archer, N. Banchero and A.J. Lechner (1982). On comparing regression lines with unequal slopes. Am. J. Physiol. 242: R178-RI80.
Respiration Physiology (1985) 59, 1-14 Elsevier
VOLUME OVERLOAD HYPERTROPHY ELICITED BY COLD AND ITS EFFECTS ON MYOCARDIAL CAPILLARITY
SUSAN R. KAYAR and NATALIO BANCHERO Department of Physiology, University of Colorado School of Medicine, Denver, CO 80262, U.S.A.
Abstract. Capillarity and fiber cross-sectional areas were measured in the hearts of guinea pigs exposed to
cold early during growth. Twelve male guinea pigs were kept at 5 + 1 °C for 4-18 weeks. Hearts were perfusion fixed via the aorta with a 2.5 % glutaraldehyde, 1% formaldehyde-buffered solution, blocks were cut from left (LV) and right (RV) ventricles, post-fixed in OsO4, dehydrated and embedded in Spurr medium. Blocks were cut transversely to fiber orientation, 0.5/zm thick, stained with Toluidine Blue and photographed at 400 ×. Heart weights, number and location of capillaries and fiber cross-sectional areas (FCSA) were measured in cold-acclimated animals and in normothermic controls. Growth rates for all guinea pigs were similar. Acclimation to cold caused modest LV and RV hypertrophy. The greater LV weight seemed due to longer fibers of normal FCSA, whereas the greater RV weight was due to larger FCSA. Capillary density, capillary-to-fiber ratio and number of capillaries around the fibers were similar in the two groups of animals. Mean and maximal diffusion distances in cold-acclimated animals were not different from controls. Thus the myocardial hypertrophy induced by chronic volume overload was fully compensated for by increases in capillarity commensurate with increases in fiber girth. Acclimation Capillary Cold
Diffusion Guinea pig Heart
Hype~'trophy Myocardium Temperature
Exposure to a cold environment causes a sustained elevation in oxygen consumption and cardiac output and the latter in turn induces modest bilateral cardiac hypertrophy (Bui and Banchero, 1980). Rodents reared in a cold environment show substantial increases in the capillarity of skeletal muscle (Heroux and St. Pierre, 1957; Sillau et al., 1981). This study examines the progression of myocardial hypertrophy in growing guinea pigs exposed continuously to a 5 °C environment and its effects on myocardial capillarity. Myocardial hypertrophy induced either by sustained volume overload as in cold exposure or intermittent volume overload as in exercise training, typically is not Accepted for publication 8 October 1984 0034-5687/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
2
S.R. KAYAR AND N. BANCHERO
associated with large changes in ventricular mass. In studies of left ventricular hypertrophy induced by aorto-caval fistula (Rakusan et al., 1980) and treadmill exercise (Tharp and Wagner, 1982), myofibers increased in girth but there were no increases in the ratio of capillary density to fiber density in the hypertrophied rat hearts compared to age-matched controls. This indicated that no compensatory capillary proliferation had occurred in these rats. Heroux and St. Pierre (1957) found no changes in myocardial capillary density in rats exposed to a 6 °C environment for 4 weeks, although changes in myofiber size or heart weight were not reported. There is evidence, however, that myocardial hypertrophy of any origin induced in young, growing animals can be accompanied by significant compensatory increases in capillarity (Rakusan, 1971; Tomanek etal., 1982). Since it is known that in muscles of normal growing animals fiber size is the most important factor determining capillarity and oxygen diffusion distances (Sillau and Banchero, 1978; Hudlicka, 1982), then it becomes particularly important to analyze the capillarity in myocardial hypertrophy as a function of the size of the myofibers. In this study, weanling guinea pigs were maintained in a 5 °C environment for 4-18 weeks. Blood samples were analyzed weekly for hematocrit and hemoglobin concentration. Animals were sacrificed after 4, 6, 8, 13, 16 or 18 weeks of cold exposure to determine heart weights, and heart tissue samples were morphometrically analyzed for fibers and capillaries. From this analysis we offer some speculations on the factors involved in capillary proliferation.
Materials and methods Guinea pigs (Cavia porcellus, Hartley strain) were obtained from Camm Research laboratories (Wayne, New Jersey). Control animals (n = 18)which were of both genders were obtained at a variety of body weights from weanling (250-300 g) to adult dimensions (1000 g). Since no significant differences (P > 0.20) were found in heart weight or body weight that were attributable to gender, data from males and females were pooled. Experimental animals (n = 12) which were all males were obtained at a body weight of 250-300 g. One week after their arrival in the laboratory, the experimental animals were placed in a chamber in which the temperature was lowered to 5 + 1 ° C. A small volume of air was blown continuously into one end of the chamber to maintain 0 2 and CO2 at normal levels. Animals were housed within the chamber in cages 20 x 40 x 20 cm in dimension, two young animals to a cage, or one animal over 700 g to a cage. Food and water were available to the animals ad libitum, and cage litter was changed at least every other day, necessitating regular, brief exposure of animals to room temperature. On the average, animals were exposed to room temperature for less than 8 h per week. Timed lamps provided a 12 h light/12 h dark cycle. Growth rate of the animals in the cold was similar to controls (P > 0.50). At weekly intervals, experimental animals were removed from the cold chamber and blood samples were drawn for hematocrit (Hct) and hemoglobin concentration ([Hb])
MYOCARDIAL CAPILLARITY IN COLD EXPOSURE
3
determinations. Blood was obtained by clipping a toenail near its base, and if necessary blood flow was later stanched by applying a drop of collodion to the toenail. Hematocrit was determined from duplicate blood samples of approximately 40/~1 each, centrifuged in capillary tubes for 6 rain. Hemoglobin concentration was determined from duplicate samples of 20 #1 each by the cyanmethemoglobin spectrophotometric method. At sacrifice the guinea pigs were anesthetized with sodium pentobarbital, 35 mg/kg IP. A polyethylene catheter was inserted into the left carotid artery and advanced to the arch of the aorta, and a second catheter was placed in the fight jugular vein and also advanced toward the heart. One milliliter of sodium heparin (1000 units) was introduced slowly via the jugular catheter. A solution of isotonic saline with 0.5~o heparin and 0.5~o lidocaine was then perfused via the carotid line, at a pressure of approximately 100 torr, with the jugular catheter opened as a drain. After 20 min of saline perfusion, most blood had drained and the animal's heart had stopped. A solution of 2.5 ~/o glutaraldehyde and 1~o formaldehyde in 0.1 M phosphate buffer, pH 7.35 was then perfused via the carotid line, also at a pressure of 100 torr. This fixative was chosen for its reputedly superior fixing properties over either a glutaraldehyde or a formaldehyde solution alone (Karnovsky, 1965). Tissues fixed with this solution appeared well-preserved under the electron microscope, with no obvious swelling or shrinkage of myofibrillar mitochondria. No correction factor for shrinkage was calculated since true anatomical dimensions were unknown. Myofibers appeared uncontracted when viewed longitudinally under the electron microscope. After 20 rain of perfusion with the fixative, the heart was removed, blotted dry and weighed to the nearest milligram while moist. The atria were discarded, right and left ventricular walls were cut free of the septum and weighed. Each ventricular wall was cut into four approximately equal bands from base to apex. Several blocks 1-2 mm to a side were cut from the central portion of each band. The blocks were then further fLxed by immersion in the glutaraldehyde-formaldehyde solution for 1 h, rinsed in 2 ~ buffered sucrose for 45 min, and postfLxed in 1~o buffered OsO4 pH 7.35 for 1 h. Following a rinse in distilled water, the blocks were stained with 2~/o uranyl acetate, rinsed again, dehydrated with acetone and embedded in Spurr low-viscosity medium. At the time of embedding, tissue blocks were examined carefully under a dissecting microscope and two blocks from each band of a ventricle were selected and embedded such that primary fiber orientation was vertical in the epicardial region of one block and in the endocardial region of the other block. Thus there were four epicardial and four endocardial samples from each ventricle, for a total of 16 tissue samples per animal. Tissue blocks were cut with a Sorvall II Ultramierotome to a thickness of approximately 0.5 ~tm. Tissue sections were stained with Toluidine Blue and examined and photographed on color transparency fire at 400 x under the light microscope. One or two photographs per section were taken only from regions in which fibers and capillaries appeared to be cut in cross-section, with capillary profiles circular and little or no signs of striations in the fibers. Photographic fields were also selected for minimal interstitial gaps between fibers (large gaps presumably being signs of mechanical damage to tissues during handling or fixation), and for the absence if possible of blood vessels larger in
4
S.R. KAYAR AND N. BANCHERO
diameter than capillary dimensions (approximately 18/~m). From endocardial samples, photographs were not taken from regions that were clearly papillary muscles. The total epicardial or endocardial sample from a given ventricle could thus contain up to 8 fields and was discarded if fewer than 3 suitable fields could be obtained. In most cases there were 20-25 fields sampled per animal. Analysis of the tissue photographs was performed with the aid of an x-y digitizer (Datatizer, GTCO Corp., Rockville, MD) on line with a Nova 4 Computer. Fiber cross-sectional areas (FCSA) were calculated by computer integration after outlining fiber profiles; fiber perimeters and maximal fiber diameters were calculated simultaneously. Fiber density was estimated from FCSA and the average area of a frame occupied by capillaries and interstitium, the latter being estimated by point-counting the first 200 frames. The number of capillaries immediately around the border of each fiber (CAF) was also recorded. The total number of capillaries and the x-y coordinates of their locations were scored for calculation of capillary density (CD) and capillary array. Data on the fibers and capillaries were also combined to calculated capillary density to fiber density ratios (C'F). The distribution of capillaries relative to each other was determined by the closest individual method (Kayar et al., 1982). A computer-stored uniform grid of 25 points was placed over a photograph of the tissues, and the distance from each grid point to the nearest capillary was recorded. A mathematical analysis of the distribution of these distance measurements yields what Kayar et al. (1982) have defined as the mean (R) and maximal (R95) oxygen diffusion distances in these tissues. Comparison of these R and R95 values to values calculated from the CD were used to indicate which of several possible geometric models best applied to the arrangement of capillaries in these tissues. Statistical analysis was performed by Student's t-test and linear regression. Regression lines were compared using Fieller's theorem (Zerbe et aL, 1982) when slopes were unequal; otherwise, y-intercepts were compared by a t-test.
Results
Cold acclimation caused a significant (P < 0.001) hypertrophy of both left (LV) and right (RV) ventricles (fig. 1). However, the degree of this hypertrophy was modest. After 4 weeks of exposure, cold-acclimated animals had ventricular weights approximately 50~ greater than those of control guinea pigs of the same body weight and after 18 weeks, ventricles of cold-acclimated animals were approximately 15 ~o heavier than in controls. In the RV of control animals, FCSA increased linearly with ventricular weight (fig. 2). The FCSA in the RV of cold-acclimated animals was similar (P > 0.50) to that of controls of the same ventricular weight, although the relationship between FCSA and ventricular weight was not significant. In the LV, FCSA increased linearly with ventricular weight for control and cold-acclimated animals. The slopes of the regression were the same (P < 0.001), but the y-intercept for the regression for the cold animals
MYOCARDIAL CAPILLARITY IN COLD EXPOSURE
3.4
y : 1.08+ (231X 10-3)r,
•
f c°'° \ :3/" r=0.88
3.0
o
5
~
o
.o X "
.,.//
2.6
0 // /"~ •
0
, ,,,o-'~ o
2.2
1.8 1.4
•
~
//~,;
o
,,y.
,-"
-.
, .-
~
LControl
.,,
r= 0.97
• ~'0,-
I.C %/
Whole HeQrt
I
I
I
I -
Cold o 3 0 0 1.2- y=O50+(O.85xlO" )x ~•'" .w'" •~ r=0.79 ~ ~.'" 5 . / ~ ' r " •
)
t.o~-
0.8 0.6 0.4
o
.o~..-
"i
•~Control y=O.OB.(H4x10-')x
.~... J•~
r=0.95
Left Ventricle I
I
I
Cold o.6
y= o.zz+ (oszxlo-~)x .~-.~_°'~r.~" r=0.93
0.4. 02
"~
oo'--" . ~
•~.=~
~ •
• -
I
400
•
[Control
y : 0 0 3 + ( 0 . 6 2 x lO-')x
Right Ventricle 0
1 ,.
o
r=0.85 i
I
600 800 Body Weight (g)
I
I000
Fig. 1. Whole heart weight and left and right ventricular weights v s body weight for control guinea pigs and for guinea pigs reared in a 5 °C environment. Control group (O) in this case includes data from 36 animals.
was significantly lower (P < 0.01) than for controls (fig. 2). Because there was LV hypertrophy this smaller FCSA at the same ventricular weight may indicate a disproportionate increase in the length of the fibers in the LV of the cold-acclimated animals. Capillary density in both ventricles of control guinea pigs decreased with increasing FCSA (fig. 3). In the RV of cold-acclimated animals, mean CD was similar (P > 0.20) to that of controls but there was no significant correlation between CD and FCSA. In the LV of cold-acclimated animals, there was a significant regression between CD and FCSA, and this relationship was similar (P > 0.10) to controls. Both C : F and CAF increased with FCSA in both ventricles of control guinea pigs (figs. 4 and 5). Data from cold-acclimated guinea pigs were similar to those measured in controls ( C : F left ventricle, P > 0 . 1 0 ; C : F right ventricle, P > 0 . 1 0 ; CAF left ventricle, P > 0.50; CAF right ventricle, P > 0.10), indicating that the total number of
6
S.R. KAYAR AND N. BANCHERO 400
Guineo Pig M y o c a r d l a m Left Venlrlcle
E
:L
~3oo
c o
•
!
.
200
A I •
0
o
•
• E p i c a r d l u m , Conlrols • Endocaromm,Controls ~ Eplcordium,Cold [] Endocclrd,u m, Cold
I00
~
y=59.00+170.37X
\
r= 0.81
Controls ~ y = 8 8 . 6 3 + 170.67x r = 0.80 0 -¢/
I
l
I
I
I
0.4
0.6
0.8
1.0
1.2
..... l
1.4
400 Right V e n t r i c l e
E
3OO c O
T-
i
IOOL? •
01"//120 •
Controls y=90.Ol+207.12x r : 0.78 I
0.4
Ventricular
l
0.6
1
0.8
1.0
Weight (g)
Fig. 2. Fiber cross-sectional area v s ventricular weight for the left (upper panel) and right (lower panel) ventricles of control guinea pigs and of guinea pigs reared in a 5 °C environment.
capillaries had increased during growth only as a function of increasing myofiber girth, but that cold had no direct effect on capillarity. Likewise R and R95 measured directly in cold-acclimated animals were not different from controls, and could be approximated + 0.5 #m for R as 0.40(CD- 1/2) and + 2 #m for R95 as 0.71(CD- 1/2). Hematocrit and [Hb] were elevated modestly in the cold-acclimated animals. Mean Hct and [Hb] of the young guinea pigs before cold exposure were 41.7 +_ 0 . 5 ~ and
M Y O C A R D I A L CAPILLARITY IN COLD E X P O S U R E
4000!
• •
/---Cold
z~ / • /y=2297.76 p,/
7
+ 154022 --'-E'--
r =0.48
E E 3000•
•
-
,,,_~l
e@
Pig Myocordium Guineo Left Ventricle & Epicordium, Contr01s • Endocerdium,ControLs Epicardiu m, Cold oEndocardium,Cold
2000 Q.
0
E
D
4000
-
3000
-
I I00
•
/^
. I
I#~
x
r = O. 63
I
I 300
J 400
]L=l~,i~ •
¢..
J87334
y= 1929.53+
200
•
•
L- L; 0 I1 t r o l $
•
E~
A•
>,
Controls ~)000
--
Right Ventr,cle
"7~
y = 2451.72'
98781 - -
r = 0.50
X
O 0
0
I i00 Fiber Cross
I 200 Sectional
I 300 Area(/~m
I 400 2)
Fig. 3. Capillarydensity v s fiber cross-sectionalarea for left (upper panel) and right (lower panel) ventricles of control guinea pigs and of guinea pigs reared in a 5 ° C environment. 13.1 + 0.2 g/100 ml blood, respectively. These values increased after the In'st week and were always between 47 and 50~o (mean = 47.6~o) and between 14 and 16 g/100 ml (mean = 14.9), respectively, in all subsequent weeks of cold exposure.
~]
S.R. KAYAR AND N. BANCHERO 1 . 4 --
0
1.2 --
.e-
0 n-
1.0-
Guinea Pig M y o c o r d i u m Left V e n t r i c l e A E p i c a r d i u m Controls
[3 & / 5 %'.," .~" /
• Endocard,um,Controls ~ Epicardium,Cold u Endocardium,Cold
./ /
•
Z~~m
.o
0.80 I--
0.6-
.,,~,/~'t ~ r / J ~ °AIlt
Lcontro ' y= 0 . 3 0 + ( 2 . 5 9 x IO-')x
\Cold
r = O. 83
t~
0
m
-& 0 . 4 -
y=O
°
"
19+(3.43x10
-3)x
r=0.86 0.2C
I I00
0
I 200
I 300
J 400
Right Ventricle
o
1.2-
O
n-
1.0 -
.9_ "
0.8-
O I--
o = ";, °
(..)
A e~m/~r'"
0.6-
/
"co.,,ol
/ 0.4
0.2-
-
/'
/ LC01d
y--0.18+ r =0.85
*
(3.33x10-')x
y=0.14+(3.88+10-3)x
r=077 I
O0
I
I
J
I00 200 300 400 Fiber Cross Sectionol Are(](/~m 2)
Fig. 4. Capillary-to-fiber ratio vs fiber cross-sectional area for left (upper panel) and right (lower panel) ventricles of control guinea pigs and of guinea pigs reared in a 5 °C environment.
9
MYOCARDIAL CAPILLARITY IN COLD EXPOSURE
M Q)
Guinea Pi 9 M y o c a r d i u m Left V e n t r i c l e • Epicardium, Controls • Endocardium,Controls tx E p i e a r d i u m , C o l d
4
t..
t,..
a
e
!
U A
o Endocordium,Cold
.Q
•
,,~ / ~ i y
o.b. 0
o
&~l / / 3 ~
o
•o :0~
o
~i-~_~;~ -m=~ T | =
E
-1
Cold ~
z
m
y= 1.49+
°-.
\
C ont.r01s ~ .. 1 . 6 5 + ((6 . 0. 2 x 1 0 "s ) x yv= 1.6 r= 0 . 7 9
(6.73xlO'n)x
r=O.81
0
I , I00
Right Ventricle
=
Q
4
I 200
I 300
J 400
Cold y- 0.8 :)+(1.08x lO-2)x r=0.77
"7. O b.
oo
3
q-'10 0 @ 0
E< z=
2 -
~ ~ o n t r o l .=ill y= 1.32+ ( 7 . 6 3 X l O - S ) X r =0.84 I I00 200 Fiber Cross Sectional L
0
I 300 Areo(~m
I
400 2}
Fig. 5. Number of capillaries around a fiber vs fiber cross-sectional area for left (upper panel) and right (lower panel) ventricles of control guinea pigs and of guinea pigs reared in a 5 °C environment.
Discussion
In studies of cardiac hypertrophy induced by pressure-overload such as in chronic hypoxic exposure or pulmonary artery banding, a number of investigators have concluded that the increased myofibrillar mass is not compensated for by an increase in the number of capillaries, and the diffusion distances for oxygen lengthen (Roberts
10
S.R. KAYAR AND N. BANCHERO
and Wearn, 1941; Rakusan, 1971; Turek etaL, 1972; Murray etal., 1979; Rakusan et aL, 1981). In our own study of right ventricular hypertrophy induced by hypoxia we found significant increases in capillarity only in young animals, but these differences disappeared as the myofibers grew past the maximal normal cross-sectional area (Kayar and Banchero, 1984). Right ventricular hypertrophy due to pressure-overload can be of massive proportions (greater than 200 ~o) and in such cases cardiac performance may be severely impaired (Bui and Banchero, 1980). In the present experiment chronic cold exposure, which causes a two-fold increase in Vo~ and a sustained elevation of cardiac output, increases the work of the heart. This leads to a significant but modest degree of hypertrophy of both ventricles. Hypertrophy of the LV of cold-acclimated guinea pigs does not appear to be due to proportionate increases in FCSA and fiber length since the ventricular weight-FCSA relationship for cold-acclimated animals was significantly different from that of controls. The explanation we favor is that hypertrophy of the LV in the cold-acclimated guinea pigs was due to disproportionately greater lengthening of the LV fibers, while the increase in FCSA relative to body weight was similar to that in controls (fig. 6). Astorri et al. (1971) have found that fiber length changes and width changes in the hypertrophy of human hearts can in fact occur at different times, with length increasing first. If one assumes that tissue weight (W) is proportional to tissue volume (assuming constant tissue density) and that changes in the weight of a ventricle are accounted for by proportional increases in FCSA and fiber length (L) then: AL
: ( / ~ W ) 1/3
AFCSA
and
= ( A L ) 2 = ( A W ) 2/3
If the weight gain in normal growth of the LV is due to proportional changes in FCSA and length, then as the LV increases from 0.4 to 1.2 g, or by a factor of 3, we would fi
400
P,g Myocardium Left Ventricle
Guinea
• Epicardium, Controls m F n d o c a rd t u r n , C o n t r o l s z~ E p i c a r d l u m , C o l d a Endocardmm, Cold
g 300 o c o lJ q)
A
0')
200 i/1 o
,3 .1o
i2_
I00
/1 200
I 400
I 600 Body
I 800
I . I000
W e i g h t {g}
Fig. 6. Relationship between fiber cross-sectional area and body weight for the left ventricle of control guinea pigs and of guinea pigs reared in a 5 °C environment.
MYOCARDIAL CAPILLARITY IN COLD EXPOSURE
11
expect an increase in FCSA of 3 2/3 = 2.1. The actual change in FCSA measured in controls was from 157 to 294 #m2 (a factor of 1.9), which is in good agreement with this prediction. Fiber length change in controls should thus be by a factor of 31/3 = 1.44. However, in the youngest cold-acclimated animals mean LV weight was 0.84 g while LV weight was 0.54 g in controls of the same body weight and age. This is an increase in ventricular weight by a factor of 1.56. If FCSA and length were increasing proportionately in the cold-induced hypertrophy of the LV, then FCSA should have been 1.562/3 = 1.35 times greater in the cold-acclimated animals. Instead, FCSA for controls of LV weight = 0.54 g was 181 pan2 and FCSA for cold-acclimated guinea pigs of LV weight = 0.84 g was 202/an 2, an increase by a factor of only 1.12. This suggests that actual fiber length change due to hypertrophy was by a factor of 1.34 (1.56/1.12) rather than by a factor of 1.16 (1.561/3) if FCSA and length changes were proportionate to each other. By the same reasoning, in the oldest cold-acclimated animals, LV weight was 1.28 g while it was 1.11 g in controls of the same body weight, which is a factor of 1.16. The FCSA in the cold-acclimated animals should have been greater by a factor of 1.10 (1.162/3) if fiber growth in length and area were proportionate, but since FCSAs were nearly identical at these respective LV weights, then actual fiber length must have been greater by a factor of 1.16/1 = 1.16, rather than by 1.05 (1.161/3). Thus we conclude that the volume overload to the LV resulted in fibers that were of normal FCSA but were probably longer than in controls of the same body weight. However, an alternative explanation for the heavier LVs in the cold-acclimated animals which we cannot rule out is that a small degree of fiber hyperplasia was present in this ventricle. Hypertrophy in the RV is entirely attributable to increases in the size of individual fibers, rather than by hyperplasia. This was demonstrated by counting the number of fiber profiles in a complete cross-section of the RV free wall. In the youngest control animal, the oldest control animal, and the oldest cold-acclimated animal, there were 35-50 fibers from the epicardial surface to the endocardial surface. The same conclusion may be drawn indirectly from the unique relationship between FCSA and ventricular weight. However normal fiber growth in the RV may not be by proportionate changes in length and FCSA as it is in the control LV. As RV weight increases from 0.2 to 0.7 g, (factor of 3.5) there should be an increase in FCSA of 2.31 (3.52/3). Actual change in FCSA measured for this change in ventricular weight was from 131 to 235 #m 2, (a factor of 1.79). This is in poor agreement with our prediction and suggests that some disproportionate length increases may be occurring. Fiber length change should account for the rest of the increase in ventricular weight which should be by a factor of 1.96 (3.5/1.79). Thus during both the normal growth of these guinea pigs from weanlings to adults and also during the modest hypertrophy from cold exposure for up to 18 weeks, fibers in the RV become nearly double in FCSA and should also double in length. The difference between the changes in fiber length and FCSA in the RV and the LV is undoubtedly a reflection of the different modes of contraction, work loads and wall tensions in the two ventricles. In the normal cardiac cycle, contraction of the LV reduces its lateral diameter with relatively tittle shortening of its vertical (base-apex) axis,
12
S.R. KAYAR AND N. BANCHERO
whereas contraction of the RV involves ventricular shortening with relatively little pulling of the free wail toward the septum. The RV is a volume pump which can increase its volume output easily, whereas the LV performs more as a pressure pump which can respond more easily to increases in outflow resistance (Brecher and Gailetti, 1963). Thus in chronic cold exposure and volume overload, the RV evidently can respond to the greater work load by fiber hypertrophy in the same geometric proportions of fiber length and FCSA that it uses in normal growth of the animal. The LV evidently responds to the greater blood volume in its chamber by increasing fiber length more than that in normal growth, which considering the primary direction of orientation of the major fiber bands, would generally increase the length of the LV and not increase wall thickness. This should distribute the greater blood volume in a manner that would not change wall tension. Capillarity in both ventricles relative to FCSA was entirely normal in cold-acclimated guinea pigs. Neither CD, C : F , CAF nor directly-measured diffusion distances were different from controls. Thus the modest hypertrophy of both ventricles was fully compensated for by increases in the number of capillaries relative to the larger fiber areas. However, we have calculated that there is probably a disproportionate lengthening of fibers that accounts for the increased LV weight in the cold-acclimated animals. Ventricular weight and also body weight, however, are poor independent variables in regressions with CD, CAF and C : F, which makes it difficult to assess these relationships between capillarity and body mass. This is not surprising if one takes into consideration that the radial component of oxygen diffusion outward from capillaries to tissues is more important to oxygen delivery rate than capillary length or total tissue volume. We believe from the magnitude of the FCSA and length differences calculated above that capillarity in the LVs of the cold-acclimated guinea pigs is not impaired. It is not at all certain how an increase in fiber length of 10-30~ more than that contributed by normal growth would affect capillary length. If one assumes that this causes an equal lengthening of the capillaries, then from the Krogh equation, Akmai etal. (1978) calculated that a modest increase in the total length of a capillary should affect blood Po2 at the venous end of the capillary by not more than a few torr. In contrast to the heart, the gastrocnemius and soleus muscles of growing guinea pigs acclimated to a 5 °C environment had considerably higher capillarity than normothermic controls (Sillau et aL, 1980). We hypothesize that the difference in the response of the capillarity of the heart vs the skeletal muscles of the hindlimb is a function of tissue Po2. Low tissue Po2 is the promoter of angiogenesis in wounds (Remensnyder and Majno, 1968) and there is reasonable evidence to suggest that hypoxia is also important in the elicitation of endothelial growth in cultures. In cold exposure, the hindlimbs themselves are cold, as would be the blood perfusing the muscles. This would result in a leftward shift in the Hb-O 2 dissociation curve. While such a shift does not affect arterial oxygen saturation, it would result in a lower venous Po2, even for a normal arterio-venous oxygen difference. If, for example, blood temperature fell from 38 to 30 ° C, at an oxygen extraction of 5 vol. ~o, venous Po2 would be approximately 13 torr lower at 30 °C (Banchero et al., 1984). Thus hindlimb tissue Po: may be significantly
MYOCARDIAL CAPILLARITY IN COLD EXPOSURE
13
lower in the cold-acclimated guinea pigs than in normal, thereby stimulating the capillary angiogenesis reported by Sillau et al. (1980) and by Heroux and St. Pierre (1957). On the other hand, environmental temperature does not affect the core temperature of guinea pigs in a cold environment (Banchero et al., 1981). The arterio-venous oxygen difference across the heart is about 11 vol. Y/o,which is more than twice the oxygen difference for the entire body. Thus additional oxygen requirements of the myocardium above resting values must be satisfied mainly by increases in coronary blood flow (CBF) since Vo2 = CBF (a - v 02 difference). The doubling of cardiac output observed in chronic cold exposure requires additional cardiac work, and therefore also requires sustained increases in CBF, but should not significantly change coronary sinus Po~ or myocardial tissue Po~. We believe that it is this constancy of the myocardial tissue Po2 that led to our finding no changes in myocardial capillarity as a function of cold exposure. A number of other studies of capillarity (for review see Hudlicka, 1982) have also found indications that tissue Po2 is involved in the elicitation of capillary proliferation. However, because blood flow rate is usually also increased in these situations some investigators believe that the mechanical effect of increased blood flow is important in capillary proliferation. It has not been clear whether it is either decreased tissue Po2 or increased blood flow alone, or these two factors combined that are sufficient for capillary proliferation. In our experiment, capillarity and oxygen diffusion distances were unchanged from values expected for myocardial fibers of a given FCSA, in a situation in which blood flow was chronically elevated, but tissue Po~ should not have been affected.
Acknowledgements This work was supported by NIH HL-18145 and HL 28849. S.R.K. was supported by NIH Postdoctoral Fellowship HL 06527.
References Akmal, K., D.F. Bruley and N. Banchero (1978). Multi-capillary model for oxygen transport to skeletal muscle. In: Oxygen Transport to Tissue. III, edited by I.A. Silver, M. Erecinska and H. I. Bicher. New York, Plenum Publishing Co., pp. 139-147. Astorri, E., A. Chizzola, O. Visioli, P. Anversa, G. Olivetti and L. Vitali-Mazza (1971). Right ventricular hypertrophy - a cytometric study on 55 human hearts. J. Mol. Cell. Cardiol. 2:99-110. B anchero, N., A.J. Lechner, L. Aquin, A.H. Sillau and M.V. Bui (1981 ). Resistance to hypothermia during chronic cold exposure. Acta Univ. Carolinae BioL 1979: 233-235. Banchero, N., S.R. Kayar and C. I. Blake (1984). Factors involved in the ehcitation of increased capillarity in muscle. Fed. Proc. 43: 1006. Brecher, G.A. and P.M. GaUetti (1963). Functional anatomy of cardiac pumping. Section 2. Circulation Vol. II. In: Handbook of Physiology. Washington, DC, American Physiological Society, pp. 790-791. Bui, M.V. and N. Banchero (1980). Effects of chronic exposure to cold or hypoxia on ventricular weights and ventricular myoglobin concentrations in guinea pigs during growth. Pfliigers Arch. 385: 155-160.
14
S.R. KAYAR AND N. BANCHERO
Heroux, O. and J. St. Pierre (1957). Effect of cold acclimation on vascularization of ears, heart, liver and muscles of white rats. Am. J. Physiol. 188: 163-168. Hudlicka, O. (1982). Growth of capillaries in skeletal and cardiac muscle. Circ. Res. 50: 451-461. Karnovsky, M.J. (1965). A formaldehyde-glutaraldehyde fixative of high osmolarity for use in electron microscopy. J. Cell Biol. 27: 137A-138A. Kayar, S.R., P.G. Archer, A.J. Lechner and N. Banchero (1982). The closest individual method in the analysis of the distribution of capillaries. Microvasc. Res. 24: 326-341. Kayar, S.R. and N. Banchero (1984). Myocardial capillarity in acclimation to hypoxia. Pfliigers Arch. (in press). Murray, P.A., H. Baiz, M.C. Fishbein and S.F. Vatner (1979). Effects of experimental right ventricular hypertrophy on myocardial blood flow in conscious dogs. J. Clin. Invest. 64: 421-427. Rakusan, K. (1971). Quantitative morphology of capillaries of the heart. Methods Achier. Exp. Pathol. 5: 272-286. Rakusan, K., J. Morafiec and P.Y. Hatt (1980). Regional capillary supply in the normal and hypertrophied rat heart. Microvasc. Res. 20:319-326. Rakusan, K., Z. Turek and F. Kreuzer (1981). Myocardial capillaries in guinea pigs native to high altitude (Junin, Peru, 4,105 m). Pfliigers Arch. 391: 22-24. Remensnyder, J. P. and G. Majno (1968). Oxygen gradients in healing wounds. Am. J. Pathol. 52: 301-323. Roberts, J.T. and J.T. Wearn (1941). Quantitative changes in the capillary-muscle relationship in human hearts during normal growth and hypertrophy. Am. Heart J. 21: 617-633. Sillau, A.H. and N. Banchero (1978). Skeletal muscle fiber size and capillarity. Proc. Soc. Exp. Biol. Med. 158: 288-291. Sillau, A.H., L. Aquin, A.J. Lechner, M.V. Bui and N. Banchero (1980). Increased capillary supply in skeletal muscle of guinea pigs acclimated to cold. Respir. Physiol. 42: 233-245. Tharp, G.D. and C.T. Wagner (1982). Chronic exercise and cardiac vascularization. Eur. J. Appl. Physiol. 48: 97-104. Tomanek, R. J., J. C. Searls and P.A. Lachenbruch (1982). Quantitative changes in the capillary bed during developing, peak, and stabilized cardiac hypertrophy in the spontaneously hypertensive rat. Circ. Res. 51 : 295-304. Turek, Z., M. Grandtner and F. Kreuzer (1972). Cardiac hypertrophy, capillary and muscle fiber density, muscle fiber diameter, capillary radius and diffusion distance in the myocardium of growing rats adapted to a simulated altitude of 3500 m. Pfliigers Arch. 335: 19-28. Zerbe, G.O., P.G. Archer, N. Banchero and A.J. Lechner (1982). On comparing regression lines with unequal slopes. Am. J. Physiol. 242: R178-RI80.