Diffusional transport of the aminoterminal propeptide of type III procollagen in the interstitium of the globally ischaemic cat myocardium

Clinica Chimica Acta 255 (1996) 183-194 ELSEVIER

Diffusional transport of the aminoterminal propeptide of type III procollagen in the interstitium of the globally ischaemic cat myocardium Nis Baun Host a'b'*, Per Sejrsen c, Lars Thorbjorn Jensen b, Stig H a u n s o a aDepartment of Medicine B, Division of Cardiology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark bDepartment of Medicine, Division of Rheumatology 232, University of Copenhagen, Hvidovre Hospital, Copenhagen, Denmark CInstitute of Medical Physiology, University of Copenhagen, School of Medicine, Copenhagen, Denmark Received 11 March 1996; revised 12 July 1996; accepted 21 July 1996

Abstract

Local repair after acute myocardial infarction appears to be reflected by levels in serum of the aminoterminal propeptide of type III procollagen (serum-PIIINP). Furthermore, serumPIIINP has recently been reported to provide information on prognosis after acute myocardial infarction. However, no attention has yet been paid to the resistance to diffusion offered by the myocardial interstitium. We determined the diffusion coefficient of PIIINP in the interstitium of the globally ischaemic interstitium of the cat (D~7) by means of a 'true transient diffusion' method, and compared with the free diffusion in water (D37). D~7 (in cm 2 s - i. 10-5) was 0.0157 + 0.0005 (mean _ SEM) (n = 13), and D37 was 0.0624 _+0.0024 (n = 12). The mean diffusive progression during 20 min of the concentration profile of [~25I]PIIINP into the tissue was calculated to be 0.19 mm. The D~7 of albumin is practically identical to the D~7 of PIIINP, and the myocardium offers a similar resistance to diffusion of PIIINP and albumin, as expressed from the ratio D37/D~7 of approximately 4 for both

* Corresponding author, Department of Medicine B, Division of Cardiology, Rigshospitalet, University of Copenhagen, DK-2100 Copenhagen, Denmark. Tel.: + 45 35452340; fax: + 45 31383186; home tel.: +45 39692210. 000%8981/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved PH S0009-8981(96)06406-6

184

N.B. Host et al. / Clinica Chimica Acta 255 (1996) 183-194

molecules. PIIINP has a molecular weight of 42 000 Da, is rod shaped and has an overall negative charge. These characteristics explain the similarity in diffusion coefficients of PIIINP and albumin, which has a molecular weight of 69 000 Da. Albumin is known to pass the membrane of the continuous capillaries of the heart, making it very likely that direct exchange of PIIINP between interstitium and capillary plasma can also occur. During one hour of interstitial diffusion PI|INP will have traversed a distance calculated to correspond to 15 20 capillaries. Therefore, the results support the concept of serum-PIIINP as a direct marker of events taking place locally in the myocardium following acute myocardial infarction. Keywords:

Myocardium; Interstitium; Collagen; Diffusion; Transport; Cat

1. Introduction During recent years circulating markers of the extracellular matrix have attracted attention. This especially pertains to the aminoterminal propeptide of type III procollagen (PIIINP), which is cleaved off in a stochiometric fashion, as type II! collagen is synthesized from type III procollagen during repair following tissue injury [1]. P I I I N P in serum (serum-PIIINP) has been found to mirror fibrillogenesis in a variety o f inflammatory conditions [2-5]. Serum-PIIINP appears directly to reflect the process of repair after acute myocardial infarction (AMI) [6]. Furthermore, we have recently reported a possible association between serum-PIIINP and the prognosis after AMI, as we found serum-PIIINP on admission and for the following few days after A M I to be higher in patients with poor outcome than in patients surviving a study period of one year [7]. During experimental formation of granulation tissue P I I I N P increases locally in the wound fluid, but is reflected serologically only if the organism is in steady state [8]. In addition, the relation between P I I I N P as a m a r k e r o f scar formation after A M ! and the extent of myocardial damage expressed by enzyme release is altered by thrombolytic therapy, which induces degradation of collagen [9,10]. Type III collagen is ubiquitous in the organism and is constantly being degraded and renewed [1] and, at present, knowledge of whether the P I I I N P found in serum following A M I originates from the m y o c a r d i u m is lacking. The transport o f P I I I N P from a confined location to the circulation has recently been investigated [1 1]. However, as yet no attention has been paid to the transport of the propeptide in the myocardial interstitium, even though knowledge of this process seems a prerequisite for the interpretation o f sequential changes in serum-PIIINP as indicative of events taking place locally in the myocardium. The aim of the present study was to examine the resistance to diffusion of P I I I N P offered by the feline myocardial interstitium during conditions of

N.B. Host et al. / Cliniea Chimica Acta 255 (1996) 183-194

185

global ischaemia by means of a 'true transient method' and to compare with the free diffusion coefficient in water.

2. Materials and methods

The experiments were performed on cats specially bred for experimental use and treated in accordance with the national guiding principles in the care and use of animals. The study was approved by the local authorities on animal studies. Three cats with an average weight of 3.4 kg were used. The animals were anesthesized with chloralose (80 mg), the hearts were excised, and the right ventricle carefully dissected from the left ventricle and placed under a gauze cloth wet with isotonic saline at room temperature. The left ventricular cavity was opened, and the anterior wall was cut free from the septum and the papillary muscles. Cylindrically shaped blocks were then prepared in random order from the right and left ventricles (8 m m in diameter and 2 - 4 m m in thickness).

2.1. Preparation of diffusion solution Porcine P I I I N P purified from lymph was used for iodination. The porcine PIIINP is known to be identical to the h u m a n propeptide concerning molecular size and immunoreactivity [12]. Na125I was purchased from Amersham (Buckinghamshire, UK). Iodogen was purchased from Pierce and Warriner (Chester, UK). Cryotubes for iodination were purchased from Nunc (Osted, Denmark). Sephadex G-25 Coarse and Superfine were purchased from Pharmacia Fine Chemicals (Uppsala, Sweden). The P I I I N P was labelled with ~25I by means of the Iodogen method [13], separating labelled antigen from free iodine by gel filtration on a 0.6 x 28 cm Sephadex G-25 Coarse column. The calculated specific radioactivity averaged 1.7 × 1012 Bq" g - ~ (45 Ci" g - 1) PIIINP. On the day of experimentation the tracer was passed through an additional gel filtration (column 0.6 × 28 cm, Sephadex G-25 Superfine) to remove radioactivity other than that bound to PIIINP. After gel filtration, it was observed that the tracer eluted in a single smooth peak. In order to ensure high activity of the tracer, all experiments were performed within 60 days of iodination.

2.2. Diffusion procedure 2.2.1. Diffusion in myocardial tissue The isotope solution was heated to 37°C, and the tissue block placed in a diffusion chamber (13.5 m m in inner diameter and 35 m m in height (Fig.

186

N.B. Host et al. / Clinica Chimica Acta 255 (1996) 183-194

1), placed in a bath of streaming water at 37°C. The tissue block was heated for 3 min and then the prewarmed isotope solution was placed on the epicardial surface for 360 s. At the end of the diffusion period the mixing chamber was emptied of isotope solution and removed and a 0.5 m m thick brass plate was placed in tight connection with the free end of the steel tube. The tissue block was then frozen immediately by immersion in isopentane precooled in liquid nitrogen. The central core of the tissue block was sawn out of the frozen cylinder (Fig. 1), mounted on the guide plate in a cryostat microtome (Tissue-Tek II, Miles, Naperville, IL, USA) and cut in slices with a thickness of 2 0 / z m in parallel with the epicardial surface. 2.2.2. Diffusion in agar Agar was heated to boiling and subsequently cooled to 60°C, whereafter a brass tube (5 m m in inner diameter) was filled with the agar solution (giving an outer diameter of 4.9 mm). A thin glass plate was pressed against the free end of the brass tube to give the agar cylinder a plane surface at

13.5 mm ,I

i=

.............

E

Indicator solution

4--~" 4mm

[

J

Fig. 1. Diffusion chamber used for diffusion in myocardial tissue. The chamber consists of a stainless steel tube surrounded by a cylindrical mixing chamber of plastic mounted around the free end of a the steel tube and tightened by an interposed short tube of rubber. The dimensions of the central core of myocardial tissue used for slicing and counting are given on the right side of the figure.

N.B. Host et al./ Clinica Chimica Acta 255 (1996) 183-194

187

that level and then removed. The diffusion procedure was performed similarly to that of the myocardial tissue except that the cylindrical mixing chamber was 12 m m in inner diameter and 30 m m in height. Slices of 20/~m (2 in each vial) of tissue or agar were counted in a liquid scintillation spectrometer (Mark III Liquid Scintillation System, Model 6880, Searle Analytic Inc., USA). Measurement was continued until 100 min had elapsed. The activity ranged from 2000 to 12 000 cpm in the first vials. Corrections were made for isotope spillover, decay and counting efficiency.

2.3. Calculations and statistics The data were depicted as the inverse error function to the complementary value of the normalised concentration, which by graphical presentation versus diffusion distance forms a straight line. The slope of the line was calculated by the method of least squares. As the concentration at the surface level (Co) was unknown, the computer was programmed to find a solution by an iterative procedure, where Co = 1.0 at the distance x = 0. The diffusion coefficient, D, (D~7 in tissue and D37 in agar) was calculated at the arbitrary concentration level of c = 0.1 using the formula:

C = Co (1 - erf(x(2w/-~t ) -1))

(1)

where t is the diffusion time in seconds. The diffusion coefficient in agar was multiplied by (1 + (2/3)q~), where ~p is the fraction 0.05 occupied by agar in 1.5% agar gel, to yield the free diffusion coefficient in water at 37°C, D37 [14]. The mean distance, x, in cm, of the concentration profile from the surface into the myocardium or agar block of the test substance employed was calculated by means of the Einstein-Schmolochowsky solution of the Fick diffusion equation: x = 2x/~

(2)

where D is the diffusion coefficient of the test substance in either water (D37) or tissue (D~7) in cm 2 s-1, and t is the diffusion time interval in seconds. When this equation is applied to tissue diffusion data, the value of D~7 includes contributions from all obstacles to diffusion in the tissue. However, the contribution from one of these factors, the tortuosity factor 2, can be corrected for. 2 denotes the relative increase in the mean diffusion path length caused by the architecture of the fibre matrix and cardiac myocytes. The corrected diffusion coefficient in the water phase of the interstitium (D~7) is calculated as: D~7 = D~7(2 2)- 1, where 2 is assumed to have a value of 1.44 [15]. The degree of elipsoidity of the P I I I N P molecule was evaluated by plotting the diffusion coefficient in water versus the

188

N.B. Host et al. / Clinica Chimica Acta 255 (1996) 183 194

reciprocal value of the cube root of the molecular weight (a Stokes-Einstein plot). F r o m the deviation of the value of the diffusion coefficient from that of a spheric molecule it is possible to estimate a ratio between the two radii in an elipsoidal molecular configuration. Values are presented as mean _+ SEM.

3. Results

The mean time from heart excision to the diffusion procedure ranged from 18 to 170 min, average 80.3 min ( n = 13). Fig. 2(a and b) show representative curves of the concentration profiles of [125I]PIIINP in cat myocardium and agar. Both concentration profiles show a straight line down to the 0.01% level of the normalized concentration. The apparent diffusion coefficient in cat myocardium, 037, the free diffusion coefficient in w a t e r , 037 , of [~25I]PIIINP and the ratio between the two are given in Table 1, which also gives the diffusion coefficients of the reference substance [14C]sucrose, and of [~31I]albumin. [14C]sucrose and [131I]albumin are known to be distributed solely in the extracellular space and transported only by diffusion [ 16,17]. Table 2 shows the average distances in m m of the concentration profile of I25I-PIIINP in myocardium obtained by diffusion without convective transport, and by free diffusion in water during 20 and 60 min. Calculations were made using Eq. (2). Values for [~4C]sucrose and [131I]albumin are given for comparison. Fig. 3 shows a Stokes-Einstein plot, in which the diffusion coefficient in water for PIIINP is depicted versus the reciprocal value of the cube root of the molecular weight and compared with a number of molecules of known steric configuration. The straight line through the value of PIIINP is seen to approximate closely a ratio of 1/13 between the two radii of an elipsoid, indicating that the shape of PIIINP is like that of a thin rod.

4. Discussion

The present study describes the interstitial transport by diffusion of PIIINP in the feline myocardium during global ischaemia, ensuring the exclusion of interference with convective transport. In addition, the results allow the shape of the PIIINP molecule to be estimated as being close to that of an elongated elipsoid with a ratio between the two radii of approximately 1/13. This is in accordance with previous findings [12], and provides evidence that iodination by the iodogen method does not alter the

N.B. Host et al. / Clinica Chimica Acta 255 (1996) 183-194

0.2 l

i"

O.d ,

0.6

m

,

0.8

m

,

1.0

l

,

1.2

I

,

l.d

,

,

,

189

1.6 ,

,'n,'n

i

4

0. i

o

0.01

o o

oo c,

o

Ooo

o°o

o

o

ooo o

oo°

o

o o

o

o

o

0.001

o

O.OO01

0

0.2

0.~

m

I

1-

0.6 ,

m

0.8 .

m

1.0 ,

m

1.2 ,

r

l.d ,

m

1.6 ,

~nm

I

+

O O O O

O

O O OOOOOO

0.0

O OO

O OOOoo OOOOo

0.001-

0.0001-

Fig. 2. Representative curves of the concentration profiles of [12sI]PIIINP in cat myocardium (a) and agar (b) using a diffusion time interval of 360 s. The abscissa is the distance from the epicardial surface/agar surface in cm, and the ordinate is the inverse error function to the complementary value of the normalised concentration. m o l e c u l a r shape o f P I I I N P significantly in contrast to i o d i n a t i o n by the C h l o r a m i n e - T m e t h o d [18]. The m a i n results, given in Table 1, s h o w that the apparent diffusion coefficient o f P I I I N P (D~7) in the m y o c a r d i u m decreased to 25% o f the free diffusion coefficient in water at 37°C. F o r

190

N.B. Host et al. / Clinica Chimica Acta 255 (1996) 183 194

Table 1 Apparent diffusion coefficient in cat myocardium, D'37 , and free diffusion coefficient in water, 037 , for ~zsI-PIIINP. Values for the reference substances 14C-sucrose, and xzSI-albumin in cat myocardium are given for comparison. Values are given as mean _+ SEM. Indicator

D'37 (cm2s-1- 105)

D37 (cm2s "1 " 10-5)

Ratio D37/D~7

MW (D)

~25I-PIIINP

0.0157 + 0.0005 (n = 13) 0.120 _+0.0080 0.0140 _+ 0.0003

0.0624 _+ 0.0024 (n = 12) 0.6748 _+0.0053 0.0635 + 0.0008

3.97

43 000

5.63 4.54

366.3 a 69 000 b

laC-sucrose 13q-albumin

aHaunso et al. [16]. bSejrsen et al. [17].

comparison,

the corresponding values of the reference substance and [131I]albumin are shown. It is notable, that the 0 ; 7 of albumin is practically identical to the D ; 7 of PIIINP, and that the myocardium offers a similar resistance to diffusion of P I I I N P and albumin, as expressed from the ratio D37/D~7. P I I I N P has a molecular weight of 42 000 Da, but owing to its rod-shape behaves like a larger molecule in gel filtrations with an apparent molecular weight between 65 000 and 150 000 Da [1,11]. Furthermore, PIIINP has an i overall negative charge with a pKa-value of approximately 3.5 [11], which may facilitate its transport in the interstitial space. These characteristics may explain the similarity in diffusion coefficients of PIIINP and albumin, which has a molecular weight of 69 000 Da. Large molecules such as albumin with an equivalent globular molecular radius of 35.5 A. may pass the membrane of continuous capillaries [19], and [14C]sucrose

Table 2 The average distance of the concentration profile, x, from the border of the ischaemic zone into the ischaemic myocardium, and the average diffusion distance in water, during 20 and 60 min, calculated by means of the Einstein-Schmolochowsky solution of the Fick diffusion equation: ~ ' = x / 2 D t Substance

125I-PIIINP 14C-sucrose

131I_albumin aHaunso et al. [16]. bSejrsen et al. [17].

Ischaemic myocardium (mm)

Water (mm)

X20 min

X60 min

X20 rain

X60 min

0.19 0.54 0.18

0.33 0.93 0.32

0.39 1.27 0.39

0.67 2.20a 0.68 b

191

N.B. Host et al. / Clinica Chimica Acta 255 (1996) 183-194

~-" 0.8 O

1251-alb. • 1311-alb. -o 1251-PIIINP

E 0.6

o v

-~- 5 1 C r - E D T A

c

-Y- 1 4 C - s u c r o s e

o

0 o c

0.4

Globular Elips.1/8

.o

- - Elips.1/13

•~ 0.2

0.0

R e c i p r o c a l c u b e root of MW

0.2

Fig. 3. A Stoke-Einstein plot, in which the diffusion coefficient in water for PIIINP is depicted versus the reciprocal value of the cube root of the molecular weight and compared with a number of molecules of known globular or elipsoidal configuration. Three straight lines are given (top to bottom): the line corresponding to molecules of globular configuration and lines corresponding to elipsoids of 1/8 and 1/13 ratio between radii respectively.

consequently direct exchange of PIIINP between interstitium and capillary plasma may also occur. However, the transport of PIIINP from peripheral tissue to the circulating blood is believed to take place mainly through the local extracellular matrix and subsequently via the lymphatics [20]. Considering the evaluation of PIIINP as a marker of local events in the myocardial interstitium, we found that PIIINP during one hour will be transported 0.33 mm by diffusion in the myocardium. The capillary density in the human myocardium averages 2200 capillaries (mm-2), and equals conditions in other mammals [21,22]. Assuming that the myocardial capillary network has the structure of a homogenous grid with an equally large distance between all capillaries, that is equilateral triangles in all planes with 47 (the square root of 2200) capillaries aligned in a cube of 1 mm 3, this can be translated into an intercapillary distance of approximately 21 /~m [22]. This means that during one hour a PIIINP-molecule will have traversed a distance corresponding to approximately 15 capillaries. Lymphatic capillaries originating in the interstitial spaces have been described in the myocardium [23]. These vessels, however, are sparsely and unevenly distributed compared with the extensive capillary bed, and are found only in

192

N.B. Host et al. / Clinica Chimica Acta 255 (1996) 183-194

connection with connective tissue elements [24,25]. Therefore, cardiac lymphatics would be expected to be of very little importance to the transport of a molecule the size of PIIINP. It is apparent from our results that exchange of P I I I N P may take place via blood capillaries. The process of interstitial transport will probably even be accelerated to some extent by the mechanical action of the heart during in vivo conditions [26,27]. In conclusion, the present results provide evidence that PIIINP is transported through the ischaemic myocardial interstitium at a rate, which supports the concept of serum-PIIINP as a direct marker of events taking place locally in the myocardium following non-thrombolysed A M I [61.

Acknowledgements The excellent technical assistance of Grete Nielsen and Annette Orth is gratefully acknowledged. The study was supported by grants from the Danish Heart Foundation, the Danish Medical Research Council, and The NOVO Nordisk Foundation.

References [1] Risteli L, Risteli J. Noninvasive methods for the detection of organ fibrosis. In: Rojkind M, ed. Connective tissue in health and disease. Boca Raton, FL: CRC Press, 1990:61 98. [2] Bentsen KD, Lanng C, Horslev-Petersen K, Risteli J. The aminoterminal propeptide of type III procollagen and basement membrane components in serum during wound healing in man. Acta Chir Scand 1988;154:97 101. [3] Horslev-Petersen K, Bentsen KD, Junker P, Lorenzen I. Serum aminoterminal type III procollagen peptide in rheumatoid arthritis. Relationship to disease activity, treatment, and development of joint erosions. Arthr Rheum 1986;29:592-599. [4] Horslev-Petersen K, Bentsen KD, Junker P, Mathiesen FK, Hansen TM, Lorenzen I. Serum aminoterminal type III procollagen peptide in inflammatory and degenerative rheumatic disorders. Relationship to physical activity. Clin Rheumatol 1988;7:61-68. [5] Horslev-Petersen K, Bentsen KD, Engstr6m-Laurent A, Junker P, Halberg P, Lorenzen I. Serum amino terminal type III procollagen peptide in rheumatoid arthritis. Relationship to clinical and serological parameters of inflammation during 8 and 24 months treatment with levamisole, penicillamine or azathioprine. Ann Rheum Dis 1988;47:116127.

[6] Jensen LT, Horslev-Petersen K, Tort P e t al. Serum aminoterminal type III procollagen peptide reflects repair after acute myocardial infarction. Circulation 1990;81:52-57.

N.B. Host et al. / Clinica Chimica Acta 255 (1996) 183 194

193

[7] Host NB, Jensen LT, Bendixen PM, Jensen SE, Koldkj~er O, Simonsen EE. The aminoterminal propeptide of type III procollagen provides new information on prognosis following acute myocardial infarction. Am J Cardiol 1995;76:869-873. [8] Jensen LT, Garbarsch C, Horslev-Petersen K, Schuppan D, Kim KY, Lorenzen I. Collagen metabolism during wound healing in rats. The aminoterminal propeptide of type III procollagen in serum and wound fluid in relation to formation of granulation tissue. APMIS 1993;101:557-564. [9] Peuhkurinen K J, Risteli L, Melkko JT, Linnaluoto M, Jounela A, Risteli J. Thrombolytic therapy with streptokinase stimulates collagen breakdown. Circulation 1991;83:1969-1975. [10] Host NB, Hansen SS, Jensen LT, Husum D, Nielsen JD. Thrombolytic therapy of acute myocardial infarction alters collagen metabolism. Cardiology 1994;85:323-333. [11] Jensen LT, Henriksen JH, Olesen HP, Risteli J, Lorenzen I. Lymphatic clearance of synovial fluid in conscious pigs: the aminoterminal propeptide of type III procollagen. Eur J Clin Invest 1993;23:778-784. [12] Jensen LT, Risteli J, Nielsen MD et al. Purification of porcine aminoterminal propeptide of type III procollagen from lymph and use for lymphatic clearance studies in pigs. Matrix 1992;11:73-79. [13] Salacinski PRP, McLean C, Sykes JEC, Clement-Jones VV, Lowry PJ. Iodination of proteins, glycoproteins, and peptides using a solid-phase oxidizing agent, 1,3,4,6-tetrachloro-37,6~-diphenyl glycoluril (iodogen). Anal Biochem 1981;117:136 146. [14] Schantz EJ, Lauffer MA. Diffusion measurements in agar gel. Biochemistry 1962;1:658-663. [15] Suenson M, Richmond DR, Bassingthwaighte JB. Diffusion of sucrose, sodium, and water in ventricular myocardium. Am J Physiol 1974;227:1116 1123. [16] Haunso S, Sejrsen P, Svendsen JH. Transport of beta-blockers and calcium antagonists by diffusion in cat myocardium. J Cardiovasc Pharmacol 1991;17:357-364. [17] Sejrsen P, Paaske W, Henriksen O. Capillary permeability of [131I]-albumin in skeletal muscle. Microvasc Res 1985;29:265-281. [18] Brocks DG, Steinert C, Gerl M, Knolle J, Neubauer HP, G/inzler V. A radioimmunoassay for the N-terminal propeptide of rat procollagen type III. Application to the study of the uptake of the N-terminal propeptide of procollagen type III in isolated perfused rat liver. Matrix 1993;13:381-387. [19] Paaske WP, Sejrsen P. Permeability of continuous capillaries. Dan Med Bull 1989;36:570-590. [20] Jensen LT, Henriksen JH, Risteli J, Olesen HP, Nielsen MD, Lorenzen I. Fate of circulating amino-terminal propeptide of type III procollagen in conscious pigs. Am J Physiol 1993;265:R139-R145. [21] Stoker ME, Gerdes AM, May JF~ Regional differences in capillary density and myocyte size in the normal human heart. Anat Rec 1982;202:187-191. [22] Bassingthwaighte JB, Yipintsoi T, Harvey RB. Microvasculature of the dog left ventricular myocardium. Microvasc Res 1974;7:229-249. [23] Ljungqvist A, Mandache E, Unge G. Ultrastructural aspects of cardiac lymphatic capillaries in experimental cardiac hypertrophy. Microvasc Res 1975;10:1-7. [24] Schmid-Sch6nbein GW. Microlymphatics and lymph flow. Phys Rev 1990;70:987-1021. [25] Esperanca-Pina JA. Microvascular aspects of myocardial lymphatic vessels in the dog by scanning electron microscopy. In: Cells and Tissues: A three-dimensional approach by modern techniques in microscopy. Alan R. Liss, 1989:435-441.

194

N.B. Host et al. / Clinica Chimica Acta 255 (1996) 183-194

[26] Han Y, Vergroesen I, Goto M, Dankelman J, Van der Ploeg CP, Spaan JA. Left ventricular pressure transmission to myocardial lymph vessels is different during systole and diastole. Pflugers Arch 1993;423:448-454. [27] Lfillmann H, Timmermans PBMWM, Ziegler A. Accumulation of drugs by resting or beating cardiac tissue. Eur J Pharmacol 1979;60:277-285.