CRYOBIOLOGY
13, 448-454
( 1976)
Functional Preservation of the Mammalian IV. Functional Effects of Perfusion with Sulfoxide (DMSO) at Normothermia A. M. KAROW, JR., of Pharmacology
Departments
Augusta,
and
AND
Kidney. Dimethyl l
A. H. JESKE 2
Surgery, Medical Georgiu 30902
College
of Georgia,
The potential value of dimethyl sulfoxide vation (S). Workers in our laboratory have (DMSO) in protecting whole kidneys previously reported the functional (3) and against freezing injury has been frequently ultrastructural (6) effects of the DMSOdemonstrated in both renal and nonrenal free perfusate on the rabbit kidney at nortissues. Previous work has suggested that mothermia for periods as long as 2 hr. Also, such cryoprotection may require rather high the ultrastructural effects of DMSO perDMSO concentrations, e.g., 2.1 M or 15 fusion have been reported (7). volumes percent, which are known to alter renal uhrastructure (7, 9) and, in view of MATERIALS AND METHODS the results presented here, can alter funcThe left kidney was removed from New tion of isolated mammalian kidneys. The Zealand albino rabbits and connected to a present study represents the most complete nonrecirculating perfusion system. Details evaluation to date of renal functional paof the surgery and perfusion apparatus have rameters in the isolated kidney during perbeen previously decribed (3, 7). fusion with cryoprotective concentrations of Each liter of perfusate contained NaCl, DMSO. The characterization of the poten96.2 mM; KCl, 40.3 mM; NaCHOs, 11.9 mM; tial functional and structural alterations CaCla, 1.7 mM; MgS04, 12.5 mM; dextrose, produced in isolated kidneys is important 11.1 mM; heparin, 2006 units/liter; creatiin evaluating cryoprotectant toxicity and in nine sulfate, 25 mg/liter; p-aminohippuric selecting suitable cryoprotectants for the acid (PAH), 1 mg%; and isoxsuprine-HCl, intact kidney. 10 mg/liter. Isoxsuprine hydrochloride We investigated in the isolated rabbit (Vasodilan, Mead Johnson) was added to kidney the effects of various concentrations prevent the initial vasospasm which otherof DMSO (1.4, 2.1, and 2.8 M) in a K+wise often occurs in this preparation (4). Mg2+-rich perfusate at normothermia. It is All solutions were filtered (Millipore, 0.22 our belief that a period of normothermic pm) immediately prior to use. Perfusate perfusion will facilitate DMSO distribution was oxygenated with 97% 023% COa, rethroughout the kidney prior to cryopresersulting in a final pH of 7.35-7.38 and an Received January 20, 1975. oxygen tension of approximately 550 mm 1 Supported by USPHS-NIH grants, AM 17816 Hg. All experiments were done at 37°C. 01 and GM 08472-13. Perfusates containing DMSO were made 2Research performed as partial fulfillment of by replacing, on a volume basis, 10, 15, and requirements for the Doctor of Philosophy in 20% of solvent water with DMSO, resultpharmacology. 448 Copyright rights
W
1976 by Academic
Press, Inc. reserved.
o3 reproduction in any form
KIDNEY
PERFUSION
ing in final DMSO concentrations of 1.4, 2.1, and 2.8 M, respectively. After the kidney was connected to the perfusion circuit, the vasculature was allowed to flush for an additional 5 min in order to stabilize flow rate and to establish urine flow. A lo-min control period was then initiated, during which perfusion of the kidney with the control perfusate (0.0 M DMSO was continued. At the end of the control period, the reservoir was exchanged for one containing DMSO. Perfusion with DMSO was then continued for precisely 50 min. No attempt was made to raise DMSO concentration gradually in the perfusate. Control kidneys were perfused the entire time without DMSO. The following data were obtained for each IO-min interval of all experiments: perfusate flow rate (F), perfusion pressure (P), vascular resistance (P/ (F X viscosity) ), glomerular filtration rate (GFR by creatinine clearance), urine flow (V), sodium reabsorption, para-aminohippurate (PAH ) clearance, sodium clearance, fractional water excretion (VIGFR) and filtration fraction (GFRIF) , Also, kidney weight gain was determined as percentage of change in wet weight during 60 min of perfusion. Differences in experimental values were considered to be statistically significant if t values (Student’s) correspond to probability (P) values of 0.05 or less. RESULTS
The functional performance of the perfused kidneys is summized in Table 1 and Figs. 14. Control
Kidneys
In the control kidneys (no DMSO), perfusate flow rate increased over the 60min period, and this gradual change in ilow rate was accompanied by reciprocal decreases in vascular resistance, GFR, and urine flow rate (Fig. 1). Despite the fact that absolute values for these parameters
WITH
449
DMSO
L
1
‘01
10 CONTROL
(O.OM
20 OMSO)
30
40
50
60
TIME MIN
Fro 1. Plots of vascular resistance, GFR, and urine flow rate in control perfusion studies of the isolated rabbit kidney. The increase in urine flow shown at the 20-min time is signii?cant (P < 0.05) by the paired t test. Differences between lo- and 60-min values for the three parameters shown are statistically significant (P < 0.05). Vertical bars represent 1.0 SEM above and below the data points.
did not appear to vary widely, the changes which occurred in these parameters from 10 to 60 min as compared with the paired t test were statistically significant (P < 0.05). Since these control studies indicated that significant changes in various renal functional parameters could occur during 60-min perfusion without DMSO as a variable, further statistical comparisons between this group and kidneys perfused with various concentrations of DMSO were made using the unpaired t test, with control values serving as predictors of renal functional parameters at the various times. This was done in an attempt to eliminate effects of perfusion time on renal function so that the effects of DMSO as a single variable could be more precisely elucidated. Throughout the 60-min perfusion period the sodium reabsorption ranged from 2.14 * 0.28 to 1.56 + 0.42 eol/g/min (mean * SEM ) and the fractional water reabsorp-
450
KAROW
Effect DMSO conce~d&&3n~
of DMSO
on Rabbit
Kidneys
AND
JESKE
TABLE
1
Perfused
with
Time (min)
Perfu&e pressure (mm Hd
a K+-Mgz+-Rich
Solution
b&3+ Cl~~GWKX bl/dmin)
at 37”Ca PAH Cl~~PJFX! (ml/g/min)
0.0
10 20 30 40 50 60
0.721 0.748 0.768 0.777 0.781 0.794
f f f f f f
0.045 0.046 0.046 0.045 0.046 0.048
61.93 60.77 56.70 55.80 53.77 52.77
r!z f f f * f
3.96 4.09 3.93 3.77 3.49 3.49
0.101 0.092 0.091 0.095 0.079 0.074
f f f f f f
0.016 0.016 0.018 0.018 0.020 0.018
0.431 f 0.433 f 0.442 f 0.416 f 0.414 * 0.368 f
0.035 0.037 0.034 0.033 0.041 0.031
1.4
10 20 30 40 50 60
0.872 0.936 0.959 1.02 1.03 1.05
f 31 f f f f
0.071 0.086 0.097 0.108 0.102* 0.102*
59.43 58.87 58.27 49.93 50.67 52.77
f f f zk f xk
2.06 1.65 3.09 2.62 2.97 2.96
0.086 0.117 0.109 0.122 0.111 0.095
f f f rt f f
0.016 0.014 0.019 0.019 0.010 0.010
0.514 0.516 0.495 0.444 0.369 0.310
f f f f f &
0.064 0.058 0.063 0.042 0.033 0.028
2.1
10 20 30 40 50 60
0.865 0.830 0.777 0.759 0.732 0.752
f 0.056 f 0.062 z!zi 0.083 f 0.095 f 0.080 f 0.058
58.00 53.00 51.83 42.67 44.50 41.50
f rt f h f f
3.11 3.34 3.24 4.69* 5.65 5.31
0.072 f 0.075 f 0.069 f 0.079 + 0.077 f 0.077 zt
0.016 0.016 0.012 0.015 0.009 0.008
0.398 0.372 0.270 0.286 0.201 0.134
f f f f f zk
0.091 0.077 0.057* 0.059 0.034* 0.020*
2.8
10 20 30 40 50 60
0.759 0.851 0.889 0.964 0.930 0.921
31 f zk f f f
67.29 70.71 68.24 55.38 62.52 75.80
f xk f f f f
2.72 3.12 3.21 3.98 4.28 3.63
0.085 0.114 0.064 0.108 0.115 0.096
0.019 0.030 0.016 0.025 0.007 0.011
0.421 0.435 0.477 0.435 0.198 0.123
zk zt f f f f
0.071 0.072 0.061 0.054 0.023* 0.019*
0.025 0.041 0.052 0.067* 0.060 0.057
f f f f * +
a Values expressed as mean f SEM. Values significantly different (P < 0.05) from control group are noted by *. b DMSO concentration indicates each of four experimental groups. Kidneys in each group were perfused with the K+-Mgz+-rich perfusate containing the designated concentration of DMSO beginning at 10 min and continuing throughout the remaining 50-min period. N = 8 in 2.1 M group and 10 in all other groups.
tion ranged from 84.6 I+ 2.3 to 95.2 * 6.9%. At the end of the 60-min perfusion period the wet weight of the control kidneys increased 35.2 * 4.9%. Effect of 1.4 M DMSO DMSO (1.4 M) resulted in significant increases in per&sate flow rates (P < 0.05) at the 50-min and 60-min time periods (Table 1) . Vascular resistance declined in a reciprocal manner, although resistance values were not significantly different from control values at any of the times (Fig. 2).
Perfusion with 1.4 M DMSO produced biphasic changes in GFR, which increased significantly above control levels at 40 min (P < 0.05) and thereafter declined. Urine flow followed GFR, although urine flow rates were not significantly changed at any time in comparison with controls (Fig. 2). Na+ reasbsorption in kidneys perfused with 1.4 M DMSO differed significantly from control levels only at the 30-min period. The clearance of PAH remained relatively stable in this group and did not change significantly in relation to control values. Kidneys perfused with 1.4 M
KIDNEY
PERFUSION
WITH
451
DMSO
DMSO gained an avera,ge of 56.8% wet weight which is signiffcantly greater (P < 0.001) than control weight gain. Effects
of 2.1
M
DMSO
Surprisingly, perfusion with 2.1 M DMSO did not alter renal hemodynamics, GFR, urine flow rate, or Na+ reabsorption when compared with controls (Fig. 3). Nor was kidney weight gain significantly increased over control values. On the other hand, 2.1 M DMSO resulted in a significant depression of the clearance of PAH (Table 1) at 30 min (P < O.OW), 50 min (P < 0.005), and 60 min (P < 0.001). Effects of 2.8
M
DMSO
Perfusion with 2.8 M DMSO induced biphasic changes in both perfusate flow rate ( Table 1) and vascular resistance ( Fig. 4). ,147
T
l.4M
IO 20 t DMSO
do
TIME MIN
20
36
TIME MIN
4’0
56
FIG. 3. Plots of vascular resistance, GFR, urine flow rate against time in rabbit kidneys fused with ‘2.1 M DMSO. No values shown significantly different from those of control neys (Fig. 1). DMSO was introduced into perfusion circuit at the lo-min time. Vertical represent 1.0 SEM above and below the points.
T T
51
IO t 2.1 M DMSO
I
1
1
4b
do
6b
FIG. 2. Plots of vascular resistance, GFR, and urine flow rate against time in kidneys perfused with 1.4 M DMSO. DMSO was introduced into the perfusion circuit at the lo-min time. The 40-min value for GFR is significantly higher (I’ < 0.05) by the unpaired t test than the corresponding value in the control group ( Fig. 1). Vertical bars represent 1.0 SEM above and below the data points.
60
and perwere kidthe bars data
Perfusate flow rate steadily increased up to 40 min, at which time flows were significantly greater than control values (P < 0.05) and then declined toward, but did not reach, the initial values. Changes in perfusate flow rate were accompanied by reciprocal changes in vascular resistance, which declined abruptly at 40 min and then rose to nearly initial values. Changes in vascular resistance, however, were not statistically significant. Figure 4 illustrates our finding that 2.8 M DMSO caused biphasic variations in GFR and urine flow rate. Of the latter changes, only the 40-min value for GFR was significantly different from controls. Na+ reabsorption increased significantly above control values at 40 min (P < 0.05) but was otherwise unchanged in relation to
KAROW
452
AND
JESKE
resistance changes occurring during prolonged DMSO perfusion (e.g., 2 hr) observed by others (10) may be due to capillary endothelial degradation. Generally, changes in vascular resistance u .09 J were accompanied by reciprocal changes in .I7 .I6 GFR and urine flow rate (Figs. l-3). In A kidneys perfused with 1.4 and 2.8 M DMSO, GFR values at the 40-min time were significantly greater than corresponding control values, at which time values for vascular resistance had reached their lowest 101 1 T level. Statistical analysis showed no significant differences between values of filtration fraction for any of the groups at any given time period. Therefore, the reciprocal relationship between GFR and resistance in kidneys perfused with 2.8 M to $0 io 4,o 5b do TIME t DMSO appears to have occurred as a result MN 2.sl-i OMSO of changes in afferent arteriolar resistance, FIG. 4. Plots of vascular resistance, GFR, and so that the significant increase in GFR at urine flow rate against time in kidneys perfused 40 min resulted from a significant increase with 2.8 M DMSO. Urine flow and GFR varied in perfusate flow rate or a decrease in afferreciprocally with changes in vascular resistance. ent arteriolar resistance, at a constant filValues shown here were significantly different from control values (Fig. 1) only at the 40-min time. tration fraction. This relationship also apVertical bars represent 1.0 SEM above and below pears to hold in kidneys perfused with 1.4 the data points. M DMSO up to 40 min, at which time a significant increase in GFR occurred along control levels. As observed in the 2.1 M with a decrease in vascular resistance and DMSO group, the clearance of PAH was an increase in perfusate flow rate, at a severely depressed by 2.8 M DMSO at 50 relatively unchanged filtration fraction. min (P < 0.005) and 60 min (I’ < 0.001). Following the 40-min measurement in the Kidney weight gain in this group averaged 1.4 M DMSO group, however, perfusate 63.20/o, which significantly exceeded control flow rates were significantly higher than values for this parameter. control values, while values of GFR declined. Since filtration fraction was not DISCUSSION significantly different from control values at DMSO is capable of altering renal hemothese 50- and 60-min time periods, one dynamics. Since our electron microscopy would have expected GFR to increase with (7) previously demonstrated the capillary the increased perfusate flow rates. A reendothelium to be intact ultrastructurally distribution of vascular flow within the at all concentrations of DMSO, we believe kidney may have occurred, so that perfusate that the resistance changes which occurred flow shifted from cortical regions to medulduring the first 50 min of DMSO perfusion lary regions. However, there is no direct are due to nonstructural mechanisms, such evidence in this study for such a redistribuas osmotic fluid shifts, edema formation, a tion of flow. Results from kidneys perfused direct vasoaction of DMSO on vascular with 2.1 M DMSO are compatible with the smooth muscle, or an indirect action via the stated relationships between vascular dylocal anesthetic effect of DMSO. Vascular
!k$i] --‘_“w’, 5
KIDNEY
PERFUSION
namics and GFR in this preparation, since vascular resistance and perfusate flow rate and GFR did not differ significantly from controls, nor did values of filtration fraction, at any of the time periods. Variations in urine flow rate during perfusion with DMSO closely parallel changes in GFR. Regulation of urine flow in rabbits is dual in nature, consisting of tubular reabsorption and glomerular filtration ( 11)) the latter becoming important in this species when body fluids are markedly expanded (2). If perfusion in vitro is capable of shifting the regulation of rabbit urine flow from tubular reabsorption to glomerular filtration, then the relationship between GFR and urine flow observed in our experiments is to be expected. The absence of any significant difference in the fractional water excretion between control kidneys and those perfused with DMSO suggests that the change in urine flow during DMSO perfusion was mediated by changes in GFR. Na+ was reabsorbed by kidneys in all groups as indicated by the fact that the ratio of Na+ clearance to GFR (creatinine clearance) was always less than 1.0. The clearances of substances reabsorbed by the kidney (e.g., Nat) should be less than that of substances filtered and not reabsorbed (e.g., creatinine). The absolute rate of Na+ reabsorption in kidneys perfused with DMSO was never significantly below that in control kidneys and actually increased during one period in 1.4 and 2.8 M DMSOperfused kidneys, The increases in Na+ reabsorption were probably related to the simultaneous increases in GFR. Changes in GFR rather than changes in Na+ handling therefore seem to be responsible for changes in urine flow rate. Some DMSO-induced changes in renal function seem to be predicated upon changes in GFR. We previously reported (7, 8) results of electron microscopy showing the glomerular ultrastructure to be resistant to DMSO perfusion. The GFR
WITH
DMSO
453
changes may well be due to DMSO-induced alterations in renal hemodynamics rather than fine structure degeneration of the glomerular barrier. PAH clearance was depressed during the final 20 min of perfusion with 2.1 and 2.8 M DMSO. DMSO may have increased tubular membrane permeability and thereby enhanced back-diffusion of PAH from the tubular fluid to the peritubular fluid, even if secretory processes remained intact. We previously reported (7) results of electron microscopy .showing proximal tubular alterations in kidneys perfused with these concentrations of DMSO. Another explanation for the depression of PAH clearance is the potential ability of DMSO to alter enzymes essential to PAH transport. Increases in renal weight during perfusion correlate with accumulation of edema fluid observed histologically (6, 7). Increases in weight may be due to restoration of vascular, interstitial, and tubular fluid volumes following exsanguination. After these “normal” volumes are restored, continuing weight gains may be due to abnormal accumulations of fluid in the same compartments secondary to altered hemodynamics or capillary permeability or to intracellular edema secondary to a perfusion-induced defect in cell ion and volume regulatory mechanisms. The exact time during perfusion at which weight gain changed from restoration of normal fluid volumes to excessive accumulation cannot be determined from our results. DMSO exacerbated these weight changes, perhaps as a result of DMSO-enhanced transcapillary fluid transudation. While colloid-free perfusates result in edema in perfused kidneys, the edema does not necessarily impair subsequent renal function after reimplantation. For example, Humphries (5) was able to preserve dog kidneys with colloid-free media as well as with diluted blood or colloid-containing plasma perfusates. In hearts, perfusioninduced edema of up to 40% weight gains
454
KAROW
AND
were rapidly reversed after reimplantation in the experiments of Copeland et al. ( 1).
JESKE
2.
SUMMARY
Rabbit kidneys were perfused at 37°C with various concentrations of DMSO in a K+-Mgz+-rich perfusate. The effects of DMSO on various functional parameters of the rabbit kidney perfused for 60 min were compared with the functional effects of perfusion without DMSO under the same conditions. DMSO produced deviations in vascular resistance and per&sate flow rate from control values. In kidneys perfused with 1.4 and 2.8 M DMSO these vascular changes resulted in changes in GFR at relatively unchanged filtration fractions. The closely parallel relationship between changes in GFR and urine flow rate in all groups indicates that perfusion per se or perfusion with DMSO may shift the regulation of urine flow rate from tubular reabsorption, which obtains in the in who situation, to glomerular filtration. This view was supported by the relatively unchanged parameters of Na+ reabsorption and fractional water excretion during perfusion with all concentrations of DMSO. Additionally, DMSO perfusion resulted in significantly greater weight gains than those observed in kidneys perfused without DMSO, and significantly depressed clearances of PAH, with 2.1 and 2.8 M DMSO. REFERENCES 1. Copeland, J, G., Jones, J., Spragg, R., and Stinson, E. B. Successful method for preservation of canine hearts for 24-28 hours In “Fourth International Transplantation Con-
3.
4.
5.
6.
7.
8.
9.
10.
11.
ference,” p. 54. Grune and Stratton, New York, 1972. Dicker, S. E., and Heller, H. Relationship between glomerular filtration rate and urine flow in the rabbit. Science 112, 340-341 (1950). Fonteles, M. C., Jeske, A. H., and Karow, A. M., Jr. Functional preservation of the mammalian kidney. I. Normothermia, low-flow perfusion. 1. Surg. Res. 14, 7-15 ( 1973). Fonteles, M. C., Jeske, A. H., and Karow, A. M., Jr. Blockade of vasoconstriction by isoxsuprine (Vasodilan) in the isolated perfused rabbit kidney. Res. Commun. Chem. Pathol. Pharmacol. 5, 333-343 ( 1973). Humphries, A. L., Jr. Problems with various perfusates for kidney preservation. Cyobiology 5, 447-451 ( 1969). Jeske, A. H., Fonteles, M. C., and Karow, A. M., Jr. Functional preservation of the mammalian kidney. II. Ultrastructure with low-flow perfusion at normothermia. J. surg. Res. 15, 4-13 ( 1973). Jeske, A. H., Fonteles, M. C., and Karow, A. M., Jr. Functional preservation of the mammalian kidney. III. Ultrastructural effects of perfusion with dimethylsulfoxide (DMSO). Cyobiology 11, 170-181 (1974). Karow, A. M., Jr., Holst, H. I., and Ecker, H. A. Organ cryopreservation: Renal and cardiac experience. In “Organ Preservation for Transplantation.” (A. M. Karow, Jr., G. Abouna, and A. L. Humphries, Jr., Eds.), p. 274. Little, Brown, Boston, 1974. Malinin, G. I. Cytotoxic effect of dimethylsulfoxide on the ultrastructure of cultured rhesus kidney cells. Cyobiology 10, 22-32 (1973). Pegg, D. E. Perfusion of rabbit kidneys with cryoprotective agents. Cryobiology 9, 411419 (1972). Renkin, E. M., and Gilmore, J. P. Glomerular filtration. In “Handbook of Physiology.” (J. Orloff and R. W. Berliner, Eds. ), Sect. 8. Renal Physiology, pp. 18%248. American Physiological Society, Washington, D. C., 1973.
13, 448-454
( 1976)
Functional Preservation of the Mammalian IV. Functional Effects of Perfusion with Sulfoxide (DMSO) at Normothermia A. M. KAROW, JR., of Pharmacology
Departments
Augusta,
and
AND
Kidney. Dimethyl l
A. H. JESKE 2
Surgery, Medical Georgiu 30902
College
of Georgia,
The potential value of dimethyl sulfoxide vation (S). Workers in our laboratory have (DMSO) in protecting whole kidneys previously reported the functional (3) and against freezing injury has been frequently ultrastructural (6) effects of the DMSOdemonstrated in both renal and nonrenal free perfusate on the rabbit kidney at nortissues. Previous work has suggested that mothermia for periods as long as 2 hr. Also, such cryoprotection may require rather high the ultrastructural effects of DMSO perDMSO concentrations, e.g., 2.1 M or 15 fusion have been reported (7). volumes percent, which are known to alter renal uhrastructure (7, 9) and, in view of MATERIALS AND METHODS the results presented here, can alter funcThe left kidney was removed from New tion of isolated mammalian kidneys. The Zealand albino rabbits and connected to a present study represents the most complete nonrecirculating perfusion system. Details evaluation to date of renal functional paof the surgery and perfusion apparatus have rameters in the isolated kidney during perbeen previously decribed (3, 7). fusion with cryoprotective concentrations of Each liter of perfusate contained NaCl, DMSO. The characterization of the poten96.2 mM; KCl, 40.3 mM; NaCHOs, 11.9 mM; tial functional and structural alterations CaCla, 1.7 mM; MgS04, 12.5 mM; dextrose, produced in isolated kidneys is important 11.1 mM; heparin, 2006 units/liter; creatiin evaluating cryoprotectant toxicity and in nine sulfate, 25 mg/liter; p-aminohippuric selecting suitable cryoprotectants for the acid (PAH), 1 mg%; and isoxsuprine-HCl, intact kidney. 10 mg/liter. Isoxsuprine hydrochloride We investigated in the isolated rabbit (Vasodilan, Mead Johnson) was added to kidney the effects of various concentrations prevent the initial vasospasm which otherof DMSO (1.4, 2.1, and 2.8 M) in a K+wise often occurs in this preparation (4). Mg2+-rich perfusate at normothermia. It is All solutions were filtered (Millipore, 0.22 our belief that a period of normothermic pm) immediately prior to use. Perfusate perfusion will facilitate DMSO distribution was oxygenated with 97% 023% COa, rethroughout the kidney prior to cryopresersulting in a final pH of 7.35-7.38 and an Received January 20, 1975. oxygen tension of approximately 550 mm 1 Supported by USPHS-NIH grants, AM 17816 Hg. All experiments were done at 37°C. 01 and GM 08472-13. Perfusates containing DMSO were made 2Research performed as partial fulfillment of by replacing, on a volume basis, 10, 15, and requirements for the Doctor of Philosophy in 20% of solvent water with DMSO, resultpharmacology. 448 Copyright rights
W
1976 by Academic
Press, Inc. reserved.
o3 reproduction in any form
KIDNEY
PERFUSION
ing in final DMSO concentrations of 1.4, 2.1, and 2.8 M, respectively. After the kidney was connected to the perfusion circuit, the vasculature was allowed to flush for an additional 5 min in order to stabilize flow rate and to establish urine flow. A lo-min control period was then initiated, during which perfusion of the kidney with the control perfusate (0.0 M DMSO was continued. At the end of the control period, the reservoir was exchanged for one containing DMSO. Perfusion with DMSO was then continued for precisely 50 min. No attempt was made to raise DMSO concentration gradually in the perfusate. Control kidneys were perfused the entire time without DMSO. The following data were obtained for each IO-min interval of all experiments: perfusate flow rate (F), perfusion pressure (P), vascular resistance (P/ (F X viscosity) ), glomerular filtration rate (GFR by creatinine clearance), urine flow (V), sodium reabsorption, para-aminohippurate (PAH ) clearance, sodium clearance, fractional water excretion (VIGFR) and filtration fraction (GFRIF) , Also, kidney weight gain was determined as percentage of change in wet weight during 60 min of perfusion. Differences in experimental values were considered to be statistically significant if t values (Student’s) correspond to probability (P) values of 0.05 or less. RESULTS
The functional performance of the perfused kidneys is summized in Table 1 and Figs. 14. Control
Kidneys
In the control kidneys (no DMSO), perfusate flow rate increased over the 60min period, and this gradual change in ilow rate was accompanied by reciprocal decreases in vascular resistance, GFR, and urine flow rate (Fig. 1). Despite the fact that absolute values for these parameters
WITH
449
DMSO
L
1
‘01
10 CONTROL
(O.OM
20 OMSO)
30
40
50
60
TIME MIN
Fro 1. Plots of vascular resistance, GFR, and urine flow rate in control perfusion studies of the isolated rabbit kidney. The increase in urine flow shown at the 20-min time is signii?cant (P < 0.05) by the paired t test. Differences between lo- and 60-min values for the three parameters shown are statistically significant (P < 0.05). Vertical bars represent 1.0 SEM above and below the data points.
did not appear to vary widely, the changes which occurred in these parameters from 10 to 60 min as compared with the paired t test were statistically significant (P < 0.05). Since these control studies indicated that significant changes in various renal functional parameters could occur during 60-min perfusion without DMSO as a variable, further statistical comparisons between this group and kidneys perfused with various concentrations of DMSO were made using the unpaired t test, with control values serving as predictors of renal functional parameters at the various times. This was done in an attempt to eliminate effects of perfusion time on renal function so that the effects of DMSO as a single variable could be more precisely elucidated. Throughout the 60-min perfusion period the sodium reabsorption ranged from 2.14 * 0.28 to 1.56 + 0.42 eol/g/min (mean * SEM ) and the fractional water reabsorp-
450
KAROW
Effect DMSO conce~d&&3n~
of DMSO
on Rabbit
Kidneys
AND
JESKE
TABLE
1
Perfused
with
Time (min)
Perfu&e pressure (mm Hd
a K+-Mgz+-Rich
Solution
b&3+ Cl~~GWKX bl/dmin)
at 37”Ca PAH Cl~~PJFX! (ml/g/min)
0.0
10 20 30 40 50 60
0.721 0.748 0.768 0.777 0.781 0.794
f f f f f f
0.045 0.046 0.046 0.045 0.046 0.048
61.93 60.77 56.70 55.80 53.77 52.77
r!z f f f * f
3.96 4.09 3.93 3.77 3.49 3.49
0.101 0.092 0.091 0.095 0.079 0.074
f f f f f f
0.016 0.016 0.018 0.018 0.020 0.018
0.431 f 0.433 f 0.442 f 0.416 f 0.414 * 0.368 f
0.035 0.037 0.034 0.033 0.041 0.031
1.4
10 20 30 40 50 60
0.872 0.936 0.959 1.02 1.03 1.05
f 31 f f f f
0.071 0.086 0.097 0.108 0.102* 0.102*
59.43 58.87 58.27 49.93 50.67 52.77
f f f zk f xk
2.06 1.65 3.09 2.62 2.97 2.96
0.086 0.117 0.109 0.122 0.111 0.095
f f f rt f f
0.016 0.014 0.019 0.019 0.010 0.010
0.514 0.516 0.495 0.444 0.369 0.310
f f f f f &
0.064 0.058 0.063 0.042 0.033 0.028
2.1
10 20 30 40 50 60
0.865 0.830 0.777 0.759 0.732 0.752
f 0.056 f 0.062 z!zi 0.083 f 0.095 f 0.080 f 0.058
58.00 53.00 51.83 42.67 44.50 41.50
f rt f h f f
3.11 3.34 3.24 4.69* 5.65 5.31
0.072 f 0.075 f 0.069 f 0.079 + 0.077 f 0.077 zt
0.016 0.016 0.012 0.015 0.009 0.008
0.398 0.372 0.270 0.286 0.201 0.134
f f f f f zk
0.091 0.077 0.057* 0.059 0.034* 0.020*
2.8
10 20 30 40 50 60
0.759 0.851 0.889 0.964 0.930 0.921
31 f zk f f f
67.29 70.71 68.24 55.38 62.52 75.80
f xk f f f f
2.72 3.12 3.21 3.98 4.28 3.63
0.085 0.114 0.064 0.108 0.115 0.096
0.019 0.030 0.016 0.025 0.007 0.011
0.421 0.435 0.477 0.435 0.198 0.123
zk zt f f f f
0.071 0.072 0.061 0.054 0.023* 0.019*
0.025 0.041 0.052 0.067* 0.060 0.057
f f f f * +
a Values expressed as mean f SEM. Values significantly different (P < 0.05) from control group are noted by *. b DMSO concentration indicates each of four experimental groups. Kidneys in each group were perfused with the K+-Mgz+-rich perfusate containing the designated concentration of DMSO beginning at 10 min and continuing throughout the remaining 50-min period. N = 8 in 2.1 M group and 10 in all other groups.
tion ranged from 84.6 I+ 2.3 to 95.2 * 6.9%. At the end of the 60-min perfusion period the wet weight of the control kidneys increased 35.2 * 4.9%. Effect of 1.4 M DMSO DMSO (1.4 M) resulted in significant increases in per&sate flow rates (P < 0.05) at the 50-min and 60-min time periods (Table 1) . Vascular resistance declined in a reciprocal manner, although resistance values were not significantly different from control values at any of the times (Fig. 2).
Perfusion with 1.4 M DMSO produced biphasic changes in GFR, which increased significantly above control levels at 40 min (P < 0.05) and thereafter declined. Urine flow followed GFR, although urine flow rates were not significantly changed at any time in comparison with controls (Fig. 2). Na+ reasbsorption in kidneys perfused with 1.4 M DMSO differed significantly from control levels only at the 30-min period. The clearance of PAH remained relatively stable in this group and did not change significantly in relation to control values. Kidneys perfused with 1.4 M
KIDNEY
PERFUSION
WITH
451
DMSO
DMSO gained an avera,ge of 56.8% wet weight which is signiffcantly greater (P < 0.001) than control weight gain. Effects
of 2.1
M
DMSO
Surprisingly, perfusion with 2.1 M DMSO did not alter renal hemodynamics, GFR, urine flow rate, or Na+ reabsorption when compared with controls (Fig. 3). Nor was kidney weight gain significantly increased over control values. On the other hand, 2.1 M DMSO resulted in a significant depression of the clearance of PAH (Table 1) at 30 min (P < O.OW), 50 min (P < 0.005), and 60 min (P < 0.001). Effects of 2.8
M
DMSO
Perfusion with 2.8 M DMSO induced biphasic changes in both perfusate flow rate ( Table 1) and vascular resistance ( Fig. 4). ,147
T
l.4M
IO 20 t DMSO
do
TIME MIN
20
36
TIME MIN
4’0
56
FIG. 3. Plots of vascular resistance, GFR, urine flow rate against time in rabbit kidneys fused with ‘2.1 M DMSO. No values shown significantly different from those of control neys (Fig. 1). DMSO was introduced into perfusion circuit at the lo-min time. Vertical represent 1.0 SEM above and below the points.
T T
51
IO t 2.1 M DMSO
I
1
1
4b
do
6b
FIG. 2. Plots of vascular resistance, GFR, and urine flow rate against time in kidneys perfused with 1.4 M DMSO. DMSO was introduced into the perfusion circuit at the lo-min time. The 40-min value for GFR is significantly higher (I’ < 0.05) by the unpaired t test than the corresponding value in the control group ( Fig. 1). Vertical bars represent 1.0 SEM above and below the data points.
60
and perwere kidthe bars data
Perfusate flow rate steadily increased up to 40 min, at which time flows were significantly greater than control values (P < 0.05) and then declined toward, but did not reach, the initial values. Changes in perfusate flow rate were accompanied by reciprocal changes in vascular resistance, which declined abruptly at 40 min and then rose to nearly initial values. Changes in vascular resistance, however, were not statistically significant. Figure 4 illustrates our finding that 2.8 M DMSO caused biphasic variations in GFR and urine flow rate. Of the latter changes, only the 40-min value for GFR was significantly different from controls. Na+ reabsorption increased significantly above control values at 40 min (P < 0.05) but was otherwise unchanged in relation to
KAROW
452
AND
JESKE
resistance changes occurring during prolonged DMSO perfusion (e.g., 2 hr) observed by others (10) may be due to capillary endothelial degradation. Generally, changes in vascular resistance u .09 J were accompanied by reciprocal changes in .I7 .I6 GFR and urine flow rate (Figs. l-3). In A kidneys perfused with 1.4 and 2.8 M DMSO, GFR values at the 40-min time were significantly greater than corresponding control values, at which time values for vascular resistance had reached their lowest 101 1 T level. Statistical analysis showed no significant differences between values of filtration fraction for any of the groups at any given time period. Therefore, the reciprocal relationship between GFR and resistance in kidneys perfused with 2.8 M to $0 io 4,o 5b do TIME t DMSO appears to have occurred as a result MN 2.sl-i OMSO of changes in afferent arteriolar resistance, FIG. 4. Plots of vascular resistance, GFR, and so that the significant increase in GFR at urine flow rate against time in kidneys perfused 40 min resulted from a significant increase with 2.8 M DMSO. Urine flow and GFR varied in perfusate flow rate or a decrease in afferreciprocally with changes in vascular resistance. ent arteriolar resistance, at a constant filValues shown here were significantly different from control values (Fig. 1) only at the 40-min time. tration fraction. This relationship also apVertical bars represent 1.0 SEM above and below pears to hold in kidneys perfused with 1.4 the data points. M DMSO up to 40 min, at which time a significant increase in GFR occurred along control levels. As observed in the 2.1 M with a decrease in vascular resistance and DMSO group, the clearance of PAH was an increase in perfusate flow rate, at a severely depressed by 2.8 M DMSO at 50 relatively unchanged filtration fraction. min (P < 0.005) and 60 min (I’ < 0.001). Following the 40-min measurement in the Kidney weight gain in this group averaged 1.4 M DMSO group, however, perfusate 63.20/o, which significantly exceeded control flow rates were significantly higher than values for this parameter. control values, while values of GFR declined. Since filtration fraction was not DISCUSSION significantly different from control values at DMSO is capable of altering renal hemothese 50- and 60-min time periods, one dynamics. Since our electron microscopy would have expected GFR to increase with (7) previously demonstrated the capillary the increased perfusate flow rates. A reendothelium to be intact ultrastructurally distribution of vascular flow within the at all concentrations of DMSO, we believe kidney may have occurred, so that perfusate that the resistance changes which occurred flow shifted from cortical regions to medulduring the first 50 min of DMSO perfusion lary regions. However, there is no direct are due to nonstructural mechanisms, such evidence in this study for such a redistribuas osmotic fluid shifts, edema formation, a tion of flow. Results from kidneys perfused direct vasoaction of DMSO on vascular with 2.1 M DMSO are compatible with the smooth muscle, or an indirect action via the stated relationships between vascular dylocal anesthetic effect of DMSO. Vascular
!k$i] --‘_“w’, 5
KIDNEY
PERFUSION
namics and GFR in this preparation, since vascular resistance and perfusate flow rate and GFR did not differ significantly from controls, nor did values of filtration fraction, at any of the time periods. Variations in urine flow rate during perfusion with DMSO closely parallel changes in GFR. Regulation of urine flow in rabbits is dual in nature, consisting of tubular reabsorption and glomerular filtration ( 11)) the latter becoming important in this species when body fluids are markedly expanded (2). If perfusion in vitro is capable of shifting the regulation of rabbit urine flow from tubular reabsorption to glomerular filtration, then the relationship between GFR and urine flow observed in our experiments is to be expected. The absence of any significant difference in the fractional water excretion between control kidneys and those perfused with DMSO suggests that the change in urine flow during DMSO perfusion was mediated by changes in GFR. Na+ was reabsorbed by kidneys in all groups as indicated by the fact that the ratio of Na+ clearance to GFR (creatinine clearance) was always less than 1.0. The clearances of substances reabsorbed by the kidney (e.g., Nat) should be less than that of substances filtered and not reabsorbed (e.g., creatinine). The absolute rate of Na+ reabsorption in kidneys perfused with DMSO was never significantly below that in control kidneys and actually increased during one period in 1.4 and 2.8 M DMSOperfused kidneys, The increases in Na+ reabsorption were probably related to the simultaneous increases in GFR. Changes in GFR rather than changes in Na+ handling therefore seem to be responsible for changes in urine flow rate. Some DMSO-induced changes in renal function seem to be predicated upon changes in GFR. We previously reported (7, 8) results of electron microscopy showing the glomerular ultrastructure to be resistant to DMSO perfusion. The GFR
WITH
DMSO
453
changes may well be due to DMSO-induced alterations in renal hemodynamics rather than fine structure degeneration of the glomerular barrier. PAH clearance was depressed during the final 20 min of perfusion with 2.1 and 2.8 M DMSO. DMSO may have increased tubular membrane permeability and thereby enhanced back-diffusion of PAH from the tubular fluid to the peritubular fluid, even if secretory processes remained intact. We previously reported (7) results of electron microscopy .showing proximal tubular alterations in kidneys perfused with these concentrations of DMSO. Another explanation for the depression of PAH clearance is the potential ability of DMSO to alter enzymes essential to PAH transport. Increases in renal weight during perfusion correlate with accumulation of edema fluid observed histologically (6, 7). Increases in weight may be due to restoration of vascular, interstitial, and tubular fluid volumes following exsanguination. After these “normal” volumes are restored, continuing weight gains may be due to abnormal accumulations of fluid in the same compartments secondary to altered hemodynamics or capillary permeability or to intracellular edema secondary to a perfusion-induced defect in cell ion and volume regulatory mechanisms. The exact time during perfusion at which weight gain changed from restoration of normal fluid volumes to excessive accumulation cannot be determined from our results. DMSO exacerbated these weight changes, perhaps as a result of DMSO-enhanced transcapillary fluid transudation. While colloid-free perfusates result in edema in perfused kidneys, the edema does not necessarily impair subsequent renal function after reimplantation. For example, Humphries (5) was able to preserve dog kidneys with colloid-free media as well as with diluted blood or colloid-containing plasma perfusates. In hearts, perfusioninduced edema of up to 40% weight gains
454
KAROW
AND
were rapidly reversed after reimplantation in the experiments of Copeland et al. ( 1).
JESKE
2.
SUMMARY
Rabbit kidneys were perfused at 37°C with various concentrations of DMSO in a K+-Mgz+-rich perfusate. The effects of DMSO on various functional parameters of the rabbit kidney perfused for 60 min were compared with the functional effects of perfusion without DMSO under the same conditions. DMSO produced deviations in vascular resistance and per&sate flow rate from control values. In kidneys perfused with 1.4 and 2.8 M DMSO these vascular changes resulted in changes in GFR at relatively unchanged filtration fractions. The closely parallel relationship between changes in GFR and urine flow rate in all groups indicates that perfusion per se or perfusion with DMSO may shift the regulation of urine flow rate from tubular reabsorption, which obtains in the in who situation, to glomerular filtration. This view was supported by the relatively unchanged parameters of Na+ reabsorption and fractional water excretion during perfusion with all concentrations of DMSO. Additionally, DMSO perfusion resulted in significantly greater weight gains than those observed in kidneys perfused without DMSO, and significantly depressed clearances of PAH, with 2.1 and 2.8 M DMSO. REFERENCES 1. Copeland, J, G., Jones, J., Spragg, R., and Stinson, E. B. Successful method for preservation of canine hearts for 24-28 hours In “Fourth International Transplantation Con-
3.
4.
5.
6.
7.
8.
9.
10.
11.
ference,” p. 54. Grune and Stratton, New York, 1972. Dicker, S. E., and Heller, H. Relationship between glomerular filtration rate and urine flow in the rabbit. Science 112, 340-341 (1950). Fonteles, M. C., Jeske, A. H., and Karow, A. M., Jr. Functional preservation of the mammalian kidney. I. Normothermia, low-flow perfusion. 1. Surg. Res. 14, 7-15 ( 1973). Fonteles, M. C., Jeske, A. H., and Karow, A. M., Jr. Blockade of vasoconstriction by isoxsuprine (Vasodilan) in the isolated perfused rabbit kidney. Res. Commun. Chem. Pathol. Pharmacol. 5, 333-343 ( 1973). Humphries, A. L., Jr. Problems with various perfusates for kidney preservation. Cyobiology 5, 447-451 ( 1969). Jeske, A. H., Fonteles, M. C., and Karow, A. M., Jr. Functional preservation of the mammalian kidney. II. Ultrastructure with low-flow perfusion at normothermia. J. surg. Res. 15, 4-13 ( 1973). Jeske, A. H., Fonteles, M. C., and Karow, A. M., Jr. Functional preservation of the mammalian kidney. III. Ultrastructural effects of perfusion with dimethylsulfoxide (DMSO). Cyobiology 11, 170-181 (1974). Karow, A. M., Jr., Holst, H. I., and Ecker, H. A. Organ cryopreservation: Renal and cardiac experience. In “Organ Preservation for Transplantation.” (A. M. Karow, Jr., G. Abouna, and A. L. Humphries, Jr., Eds.), p. 274. Little, Brown, Boston, 1974. Malinin, G. I. Cytotoxic effect of dimethylsulfoxide on the ultrastructure of cultured rhesus kidney cells. Cyobiology 10, 22-32 (1973). Pegg, D. E. Perfusion of rabbit kidneys with cryoprotective agents. Cryobiology 9, 411419 (1972). Renkin, E. M., and Gilmore, J. P. Glomerular filtration. In “Handbook of Physiology.” (J. Orloff and R. W. Berliner, Eds. ), Sect. 8. Renal Physiology, pp. 18%248. American Physiological Society, Washington, D. C., 1973.
