- Email: [email protected]
Preimplantation alteration of adenine nucleotides in cryopreserved heart valves
Preimplantation Alteration of Adenine Nucleotides in Cryopreserved Heart Valves Patrick W. Domkowski, MS, Robert H. Messier, Jr, MD, Donald G. Crescenzo, MD, Hamdy S. Aly, MD, PhD, Anwar S. Abd-Elfattah, PhD, Stephen L. Hilbert, PhD, Robert B. Wallace, MD, and Richard A. Hopkins, MD Georgetown Universitv School of Medicine, Washington, DC;Medical College of Virginia, Richmond, Virginia; and Food and Drug Admynistration, R o c k h e , Maryland
+
+
To assess the initial metabolic phase of cellular injury from cardiac valve processing, high-energy phosphate concentrations were analyzed in valve leaflets subsequent to critical processing steps. Using a porcine model, valves were processed in a manner identical to human homografts, with 58 randomly assigned to five groups representing distinct preparation phases. Group I (controls) sustained 40 minutes of warm ischemia concluded by liquid nitrogen immersion. Remaining groups similarly endured 40 minutes of ischemia, but were subsequently prepared according to stepwise design: 11, warm ischemia 24 hours of 4OC ischemia; 111, warm ischemia 24 hours of 4OC antibiotic disinfection; IV, warm ischemia 24 hours at 4OC (without antibiotics) cryopreservation (-1OC/min cryoprotected freezing); and
V, warm ischemia disinfection cryopreservation. At each regimen's conclusion leaflet extracts were assayed by high-performance liquid chromatography for highenergy adenine nucleotides (adenosine triphosphate, adenosine diphosphate, adenosine monophosphate) and catabolites. A 47% and 86% decrease in cellular adenosine triphosphate level was observed in group 111 and group V leaflets, respectively. The level of total adenine nucleotides was maintained up to cryopreservation; thereafter a 74% decrease was noted. Catabolite analysis confirmed incomplete degradation of adenine nucleotides indicating cellular metabolic resilience throughout standard homograft preparation in valves previously exposed to 40 minutes of warm ischemia. (Ann Thorac Surg 1993;55:413-9)
C
ide, irradiation, and glutaraldehyde treatment. Most have been demonstrated to damage the tissue and thereby negatively influence long-term homograft durability [7-91. Conversely, less rigorous techniques such as maintenance in culture medium (4°C) for up to 6 weeks have provided satisfactory results [lo-121. At present, however, the majority of harvested human valves are disinfected using antibiotic formulations having low toxicity and preserved by controlled-rate freezing (-l"C/min) with 10% dimethylsulfoxide as a cryoprotectant (13, 141. Cryopreservation is a well-defined and reproducible technique [151, but other "preimplantation processing" steps occurring between valve harvest and implantation are less uniform. Details of one effective method include a rinsing of the procured valve in saline solution and storage in cold (4°C)nutrient medium for 24 hours, during which transport to processing facilities and antibiotic disinfection are usually accomplished. Valves are then cryopreserved as described and maintained in vapor phase liquid nitrogen until the time of transplantation, when they are thawed by means of a dilutional warming procedure in the operating room. Each step between donor death and implantation retains a potential for homograft alteration through injury to the cellular and matrix components. Morphologic criteria and in vitro cell viability assays have been used as a means of monitoring the effects of processing on the valvular tissue [3]; we were interested in quantitating high-energy phosphate
+
+ +
+
ongenital and acquired cardiac valvular pathology continues as a prevalent clinical problem, often requiring prosthetic replacement. Because of the morbidity associated with the conferred "prosthetic valve disease," transplanted human valves are under increasing scrutiny as they have demonstrated certain specific advantages over xenograft and mechanical replacement heart valves [l].With increased utilization, those factors potentially influencing prolonged valve durability, the major controversial issue, remain incompletely characterized. It has been proposed by some that such determinants may be related to the postimplantation cellular integrity of homograft leaflet fibroblasts [24]. Since the initial homograft aortic valve replacements by Ross and Barratt-Boyes in 1962 [5,6], many methods have been employed for preparation before storage of such tissue. Historically, these have included various methods of chemical disinfection and sterilization including highconcentration antibiotics, ppropriolactone, ethylene oxAccepted for publication May 18, 1992. Presented in part to the Society for Cryobiology, Leuven, Belgium, July 1991. The opinions or assertions contained herein are the private views of the authors and are not to be construed as conveying either an official endorsement or criticism by the US Department of Health and Human Services, the Food and Drug Administration. Address reprint requests to Dr Hopkins, Department of Surgery, UHC, Georgetown University, 3800 Reservoir Rd, NW, Washington, DC 20007.
0 1993 by The
Society of Thoracic Surgeons
0003-4975/93/$6.00
414
DOMKOWSKl ET AL NUCLEOTIDES AND CRYOPRESERVEDVALVES
concentrations as a function of the various processing steps. Intracellular concentration changes of adenine nucleotides essential for anabolic cellular function may serve as an estimate of the metabolic status of the cell. This type of metabolic assay of high-energy phosphate measures ischemic injury presumably during a prelethal injury phase. This may function as an initial measurement of ischemic stress, rather than using a less sensitive indicator of frank cell death. The purpose of this experiment was to accurately measure these phosphorylated adenine nucleotides and their products of catabolism, thereby evaluating the extent of such injury occurring consequent to cold ischemia, antibiotic disinfection, and cryopreservation.
Material and Methods Animal Preparation Fifty-eight semilunar cardiac valves were obtained from 9-month-old female Yorkshire and Poland China swine after a 40-minute period of corporeal warm ischemia between death of the animal and cardiotomy at the slaughterhouse. Aortic and pulmonary valve conduits were dissected, roots opened between adjacent cusps, and the leaflets subjected to vigorous scraping with a rubber policeman to remove endothelium. Random scanning electron microscopy was performed to verify endothelium removal. The excised leaflets were either stored in liquid nitrogen (-190°C) or refrigerated (4°C) as defined by the experimental protocol.
Experimental Groups Group I (n = 12) leaflets were immediately immersed in liquid nitrogen after harvest (ie, control; 40-minute warm ischemic time) to establish baseline adenine nucleotide concentrations. Group I1 (n = 12) leaflets were placed in serum-free RPMI 1640 nutrient medium (Gibco Co, Grand Island, NY) at 4°C for 24 hours (cold ischemic time; analogous to cold transport of donor tissue to processing centers, but in the absence of antibiotics by design) followed by rapid freezing by immersion in liquid nitrogen. Group I11 (n = 11) leaflets were subjected to 4°C antibiotic disinfection (cold disinfection time, virtually analogous to clinical protocol) followed by immersion in liquid nitrogen. The antibiotic formulation consisted of a serum-free RPMI 1640 nutrient media solution containing cefoxitin, 240 pg/mL; lincomycin, 120 pg/mL; polymyxin B sulfate, 100 pg/mL; and vancomycin, 50 pg/mL [15]. Group IV (n = 12) was subjected to cold ischemia (identical to group I1 valves, 24 hours at 4"C, in antibiotic-free media) but then cryopreserved as follows: 10% each of dimethylsulfoxide and fetal calf serum in RPMI 1640, frozen at a rate of -1"C/min to -60°C (CryoMed 1010 Micro computer), and stored in vapor phase liquid nitrogen (-190OC) (CryoMed freezer model CMS-450A, New Baltimore, MI) [15]. Group V (n = 11) leaflets were subjected to all of the processing steps, modeling a completed homograft preparation with brief warm ischemia (ie, 40 minutes of warm ischemia, 24 hours of cold antibiotic disinfection, and cryopreservation). Leaflets were thawed in single-valve trileaflet sets, and one cusp
Ann Thorac Surg 1993;55:41%9
was randomly selected for analysis of all adenine nucleotide metabolites by high-performance liquid chromatography.
Ex traction Procedure Using cold (4°C) instruments to manipulate the tissue, the selected leaflet from each valve was homogenized by hand for 30 minutes at 10°C in 12% trichloracetic acid solution. The mixture was separated by centrifugation (10 minutes, 3,000 rpm) and the soluble extract removed and neutralized using a tri-n-octy1amine:freon (1:2) mixture. After vortexing, the organic phase was separated by centrifugation (10 minutes, 3,000 rpm) and stored in 1.5-mL polypropylene microcentrifuge tubes (Baxter, McGaw Park, IL) at -70°C. The postextraction pellets of denatured leaflet protein were dissolved in 0.6 N sodium hydroxide for protein determination according to the method described by Lowry and associates (161 (Gilford Response UV-VIS spectrophotometer [Oberlin, OH] at 725 nm).
High-Performance Liquid Chromatography Analysis Aliquots (80 pL) were loaded onto a Waters Intelligent Sampler Processor model 710B high-performance liquid chromatography column (Waters Associates, Milford, MA). A solvent delivery system model 6000A (Waters Associates) along with a solvent selector model 101 (Alltech Associates, Deerfield, IL) was used. Step gradient separation of the adenine pool metabolites (adenosine triphosphate [ATE'], adenosine diphosphate [ADP], adenosine monophosphate [AMP], oxidized form of nicotinamide adenine dinucleotide "AD+], adenosine, inosine, hypoxanthine, and xanthine) was achieved using a NOVA-Pak-A (C18; 5-pm particle size; 10 x 8 mm) chromatographic column inside a radial compressor module model RCM 100 (Waters Associates) (Fig 1). Ammonium phosphate (100mmoYL; pH 5.5) was used to elute ATP, ADP, uric acid, hypoxanthine, and xanthine; NAD+ was eluted with 7% methanol; and inosine and adenosine were eluted with 40% methanol. The adenine pool metabolites were integrated and quantitated t:, external standards through a visible/ultraviolet detector model 490 and a data module model 730 (Waters Associates). Retention time, peak area, and amounts of all eluted adenine metabolites were recorded on each chromatogram and normalized to the average concentration of protein in the units of adenine metabolite per milligram of protein % standard error of the mean. Chemical reagents and buffers were purchased from Sigma Chemical Company (St. Louis, MO). High-performance liquid chromatographygrade ammonium phosphate and methanol were purchased from Fisher Scientific (Fair Lawn, NJ). Distilled and deionized water was processed through the Milli-Q System Millipore (Bedford, MA).
Sta tistical Analysis The statistical analysis program SAS (Statistical Analysis System Institute, Cary, NC) processed on an IBM 386 computer was used to determine statistical significance, where p was less than 0.05 using a one-way analysis of
DOMKOWSKI ET AL NUCLEOTIDES AND CRYOPRESERVED VALVES
Ann Thorac Surg 1993;55:41%9
415
f 2.5 a
Ic ~
F
Q 1.5 & a
ATP A
<
-
2
D
P
21 u)
HX
5 05
E
E
X
2 L
0
I
II
111
Processed Groups
IV
V
Fig 2 . Progressive degradation of adenosine triphosphate (ATP) existing at the completion of each processing phase: (1) warm ischemia, (11) cold transport, (111) antibiotic disinfection, (IV) cryopresmation, and ( V ) cryopreservation with antibiotics. (* p < 0.05 versus group I; p < 0.05, group ZV versus group I1 and group V versus group 111.)
NAD IN0 ADEN
adenosine + inosine 1, Figs 24).
Fig 7 . Representative peaks of adenine nucleotide pool metabolites as generated by step-gradient high-performance liquid chromatographic elution. Retention time, peak area, and amounts of each metabolite are recorded from each chromatogram. (ADEN = adenosine; ADP = adenosine diphosphate; AMP = adenosine monophosphate; ATP = adenosine triphosphate; HX = hypoxanthine; IN0 = inosine; NAD = nicotinaniide adenine dinucleotide, oxidized form; X = xanthine.)
+ hypoxanthine + xanthine) (Table
Group I: 40 Minutes of Warm Ischemia (Control) After 40 minutes of warm ischemia 12 randomly assayed leaflets contained the following baseline concentrations: ATP = 1.78 f 0.25 nmoYmg protein, TAN = 2.59 f 0.33 nmoYmg protein, and TDP = 3.51 2 0.41 nmoYmg protein (see Figs 2-4). Values for all metabolites are listed in Table 1.
variance between all possible combinations of each step of the experimental model. A Scheffk’s multiple comparison procedure was used to identify significant differences between treatment groups.
Group 11: 40 Minutes of Warm Ischemia 4°C Ischemia (No Antibiotics)
+ 24 Hours of
A 17% reduction in ATP (see Fig 2) was observed in group I1 leaflets (1.48f 0.12 nmoYmg protein) as compared with group I levels, indicative of some ATP degradation induced by this step. In addition, there was an 8% increase in TAN (2.815 0.34 nmoYmg protein) (seeFig 3) and a 6% increase in TDP (3.75f 0.87nmol/mg protein) (see Fig 4) as compared with group I. This commenced a trend of ATP degradation to lower energy adenine nucleotides
Results Valve leaflets (excluding endothelium) were analyzed after each processing step of a representative homograft valve cryopreservation protocol to quantify the following metabolites: ATP, total adenine nucleotides (TAN = AMP + ADP + ATP), and total diffusible purines (TDP = Table 1 . Metabolites ( n m o h x proteiny
ATP ADP AMP Adenosine Inosine Hypoxanthine Xanthine NAD+
Group 111
Group I (WIT only)
Group I1 (WIT + 24 h 4°C ischemia)
(WIT + 24 h 4°C antibiotic disinfection)
Group IV (WIT + 4°C ischemia + cryopreservation)
Group V (WIT + 4°C disinfection + cryopreservation)
1.78 2 0.25 0.59 f 0.07 0.40 f 0.10 0.18 +- 0.06 1.50 f 0.30 1.70 f 0.25 0.00 f 0.00 0.50 f 0.04
1.48 f 0.12 0.55 f 0.05 0.78 f 0.21 1.20 f 0.29 0.72 f 0.20 0.78 f 0.16 1.10 f 0.63 0.33 f 0.04
0.91 f 0.13b 0.33 f 0.07 1.00 f 0.40 6.10 ? 3.50 0.87 f 0.20 0.73 f 0.24 0.10 f 0.08 0.29 f 0.12
0.46 f O . l l b
0.25 f 0.04b 0.25 f 0.03 0.18 f 0.07 3.10 f 0.70 0.71 f 0.13 0.07 k 0.03 0.05 2 0.04 0.23 2 0.04
0.27 k 0.05 0.14 5 0.07 0.52 f 0.10 0.50 0.50 0.12 f 0.06 0.09 f 0.08 0.07 f 0.04
*
Adenine nucleotide metabolite concentrations (* standard error of the mean) existing at the completion of each step of preimplantation processing. p < 0.05 versus group I by analysis of variance.
ADP = adenosine diphosphate; AMP = adenosine monophosphate; dinucleotide, oxidized form; WIT = warm ischemia.
ATP
=
adenosine triphosphate;
NAD+ = nicotinamide adenine
416
.-c Q 2 a L
DOMKOWSKI ET AL NUCLEOTIDES AND CRYOPRESERVED VALVES
Ann Thorac Surg 1993;55:41>9
3.5
T
3
I
II
111
Processed Groups
IV
V
Fig 3 . Total adenine nucleotides (TAN)existing at the completion of each processing phase: (I) warm ischemia, (II)cold transport, (III) antibiotic disinfection, (IV)cryopreserwtion, and N)cryopreservalion with antibiotics. (* p < 0.05 versus groups I , 11, and 111.)
and ultimately to purine byproducts (both TAN and TDP were increased). However, statistical significance was not reached by this group of valves.
Group Ill: 40 Minutes of Warm Ischemia of 4°C Ischemia With Antibiotics
+ 24 Hours
Antibiotic disinfection resulted in a 49% reduction (see Fig 2) in ATP with respect to control leaflets (0.91 f 0.13 nmoYmg protein; p 5 0.0001 versus group I). However, TAN concentrations in group 111 (2.28 2 0.48 nmoYmg protein) were not significantly reduced. It should be emphasized that through the completion of the antibiotic step, total high-energy phosphate metabolites (TAN) were statistically maintained at baseline 40-minute levels (see Fig 3), indicating that during these stages ATP degradation proceeds mainly to its lower energy phosphates (ie, ADP and AMP). However, levels of diffusible purines were 55% higher (7.83 + 1.84 nmoYmg protein) than in group I (p = 0.0006) (see Fig 4). This accumulation of byproducts is consistent with degradation of ATP to TAN and partial TAN degradation to adenosine, inosine, hypoxanthine, and xanthine.
Group IV: 40 Minutes of Warm Ischemia + 24 Hours of 4°C Antibiotic-Free Ischemia + Cyopreservation The consequences of cryopreservation were evaluated in groups IV and V, which differed by the presence or
I
II
111
Processed Groups
IV
V
Fig 4 . Total diffusible purines (TDP) at the completion of each phase of cardiac valve transplant preparation: (I)warm ischemia, (IIcold ) antibiotic disinfection, (IV)cryopresmation, and N) transport, fIII) cyopreservatwn with antibiotics. C‘ p < 0.05 versus groups I , 11, and V by analysis of variance.)
absence of antibiotics. In group IV, the valves exposed to cryopreservation without antibiotics, there was a significant depletion of ATP by 74% (0.46 f 0.11 nmol/mg protein) compared with control (p 5 0.0001) (see Fig 2). The 69% reduction (p = 0.0001) in ATP between groups I1 and IV (see Fig 2) quantitates the effect of cryopreservation and warm and cold ischemia without exposure to antibiotics. Similarly, cryopreservation without antibiotics effected a 66% TAN depletion from baseline (0.87 f 0.15 nmoYmg protein) (see Fig 3). With the addition of cryopreservation to 24 hours of cold ischemia (group IV), there was a 69% diminution (p = 0.OOOl) in TAN from antibiotic-free cold ischemia levels (group II). There was a significant change in total diffusible purines (1.24 f 0.49 nmoYmg protein) in group IV (p < 0.05 versus baseline).
Group V : 40 Minutes of Warm Ischemia, 24 Hours of 4°C Antibiotic Disinfection, and Cryopreservation (Complete Cryopreservation Protocol) Analysis of ATP in group V (0.25 f 0.04 nmoYmg protein) measures the cumulative effects of all steps involved in the preparation of clinical homograft cardiac valves. There was an 86% (p = 0.OOOl) and 83% (p = 0.OOOl) reduction in ATP as compared with baseline (group I) and 24 hours of cold ischemia without antibiotics (group 11), respectively (see Fig 2). Cells in leaflets exposed to both antibiotics and cryopreservation demonstrated a 73% reduction in ATP (p = 0.0001) versus leaflets experiencing 24 hours of cold ischemia with antibiotics (group III) (see Fig 2). The effect of all steps (group V) on TAN levels (0.68 2 0.09 nmoYmg protein) was simiiar to that seen with group IV. Concentrations of AMP, ADP, and ATP in the fully processed leaflets demonstrated a 74% (p = 0.0001) reduction as compared with group I (see Fig 3). The total diffusible purine concentration in fully processed leaflets (3.56 f 0.82 nmoYmg protein) was 55% less than the purine level in leaflets assayed after antibiotic disinfection alone (see Fig 4). The NAD+ levels remained essentially unchanged (0.49 f 0.06 nmoYmg protein) until the cryopreservation steps, which resulted in a slight reduction.
Comment This study of porcine cardiac valves was designed to
mimic favorable harvest conditions in which the warm ischemic time was limited to 40 minutes, correlating with a brief period between cessation of donor heartbeat and procurement, such as might occur in a multiple organ harvest setting. Experimental tissue exposed to this degree of ischemia conveniently served as control groups with which to compare metabolite levels after later preimplantation processing steps. Twenty-four hours of cold ischemia did not cause reductions in total high-energy phosphate concentrations from baseline (see Fig 3). However, 24 hours of cold ischemia resulted in a decrease in the ATP level that, although not statistically significant, was subsequently observed throughout processing. This finding indicates that although the ATP level is diminished at this stage, the majority of degradation occurs
Ann Thorac Surg 1993;55:41>9
only through to ADP and AMP (reflected in total adenine nucleotide concentrations; confirmed by the measured increases in ADP and AMP) due to incomplete breakdown of ATP or partial restoration by alternate pathways. The purpose of this particular experimental grouping, however, was to differentiate the cold ischemic effects alone (group 11) from cold ischemia combined with disinfection (group 111).
Antibiotics and Processing-Associated lnjury Many aggressive antibiotic regimens that include antifungal and high-dose antibiotic mixtures have been detrimental to fibroblast viability [17], but antibiotic combinations similar to the one employed in this study appear relatively benign as demonstrated by tissue culture and proline uptake assessment [ 18-20]. Independent of antibiotic toxicity [21, 221, this study has substantiated that metabolic depletion is, in fact, occurring during 24 hours of cold ischemia and antibiotic disinfection. However, in this study the majority of ATP degradation occurring during antibiotic disinfection generates only to ADP and AMP, which is noteworthy considering that it is the conservation of total high-energy phosphates (not ATP alone) that is critical to a cell's ability to maintain basal function and integrity [23]. Such changes are potentially reversible (eg, conversion of ADP and AMP to ATP) and appear to be likely due to the ischemia associated with the time necessary for disinfection rather than chemical injury. The effect of disinfection disappears when comparing cryopreservation with and without antibiotics, because cryopreservation after disinfection results in an additional 46% depletion of ATP. Regardless of whether prior antibiotic disinfection treatment is used, apparently the cryopreservation phase is responsible for the largest decrease in ATP. This information can be used to develop disinfection strategies that may be even less metabolically costly. Total avoidance of antibiotic disinfection with the subsequent enhanced risk for transmission of infectious disease to the recipient does not appear to be worth the minimal "gain" of reduced metabolic cellular leaflet injury and its theoretical effect on long-term performance.
DMSO Cryopreservation and Cellular Metabolic Injury With subsequent cryopreservation, significant reductions in high-energy phosphates were observed. Cryopreservation alone, subsequent to only warm ischemia and 24 hours of 4°C ischemia without antibiotics, was responsible for a substantial depletion of ATP (74%)(see Fig 2), which was greater than that imparted by cold ischemia (17%)or cold disinfection (47%).The effects of all preprocessing steps were synergistic in the diminution of adenine nucleotides. Nevertheless, leaflet cells still retained measurable amounts of high-energy phosphates after all of the processing steps (see Fig 3).
Metabolic Depletion, Cell Injury, Viability, and Durability Determination of the relationship between valvular leaflet fibroblast viability and the long-term function and durability of homografts has been the focus of a few investi-
DOMKOWSKI ET AL NUCLEOTIDES AND CRYOPRESERVED VALVES
417
gations and numerous postulations [17, 24, 251. Before this interesting question can be addressed, the initial response of the leaflet fibroblast cell population to ischemia and subsequent processing steps must be defined. Adenine nucleotide quantitation is not by itself a viability assay; however, it is a specific indicator of the overall metabolic state of the leaflet fibroblast cell population. Our data define a progressive diminution of high-energy phosphate with each step of preimplantation processing. It is reasonable to assume that as these nucleotides are increasingly depleted, a percentage (without attempting to assign a value) of fibroblasts will lose their ability to structurally maintain cell membranes, thus leading to irreversible injury. It is possible that only after relatively complete depletion of high-energy phosphates in the cell population (which this study did not find, even in the fully processed groups) will irreversible metabolic injury be manifest in the majority of fibroblasts. Unlike specific morphological assays [26], which examine irrefutable structural criteria for injury or death of fibroblasts, the aim of this investigation was to delineate the energy reserve of the leaflet fibroblast population. Few studies have carefully analyzed the early injury response (before irreversible cell injury), a period when such insults may be reversible and responsive to technical modifications, thereby potentially mitigating later effects on ultimate viability. Such changes in high-energy phosphate status might also precede reductions in energy-dependent cell functions such as proliferation, protein transport, and synthetic activity [19]. The hypothesis that cellular viability is a critical determinant of long-term clinical performance rests on the notion that it is the extent of cellular viability at the time of implantation that is the key issue [4]. If harvesting and processing protocols are to be designed to maximize cellular viability, then studies like this one are fundamental to understanding the effects of preimplantation processing on heart valve tissue [26-281. An alternative concept might be that it is the extent of functional viability (ie, fibroblast capable of proliferation and structural protein synthesis) some weeks or months after transplantation that is critical to enhancing valve durability. This being the case, processing methods that result in cells that not only "appear" viable but that have significant metabolic reserve with which to weather the peritransplantation stress could lead to better leaflet fibroblast functional recovery.
High-Performance Liquid Chromatography High-performance liquid chromatography assays have been employed to quantitate adenine nucleotide concentrations in myocardial cells after ischemia and reperfusion injury [29]. This study demonstrates the feasibility of this method for detecting nanomolar concentrations of the adenine nucleotide metabolites in cardiac valvular leaflets as well. Other methods for adenine nucleotide determination include enzymatic analysis and nuclear magnetic resonance; however, the former depends on generation or consumption of a substrate in subcellular reactions, whereas the latter requires fairly large samples [30,311.
418
DOMKOWSK~ET AL NUCLEOTIDES AND CRYOPRESERVED VALVES
ATP
/-t
&
1
ADP
( 31
& ADENOSINE & INOSINE 14, & HYPOXANTHINE (51 & XANTHINE (6: ' (7) & URIC ACID
Fig 5 . Degradation of adenosine triphosphate (ATP) to its lower energy metabolites and constitutive purine bases. Circled numbers indicated principal enzymes responsible for ATP catabolism: 1 = ATP kinases, 2 = adenylate kinase, 3 = cytosol 5'-nucleotidase, 4 = adenosine deaminase, 5 = purine nucleoside phosphoylase, 6 = xanthine oxidase, 7 = xanthine oxidase. (ADP = adenosine diphosphate; AMP = adenosine monophosphate.)
The progression of biochemical changes in the leaflet fibroblasts has been examined with proton ('H) and phosphorous (3'P) magnetic resonance spectroscopy, and demonstrated a time-dependent depletion of ATP stores reflected by an increase in inorganic phosphate; after 2 hours of harvest-related ischemia, increases in lactate concentration were measured by the nuclear magnetic resonance technique indicating the onset of anaerobic These data are quite consistent with our metabolism [N). high-performance liquid chromatography results. An advantage of the high-performance liquid chromatography technique resides in its utility for analysis of the entire high-energy phosphate pool in a "snapshot" fashion throughout the degradatory cascade [32] (Fig 5). All catabolism reactions up to and including the production of adenosine are completely reversible and allow for the possibility of adenosine phosphorylation into a highenergy metabolite and reutilization. Conversely, the freely diffusible purine inosine cannot be converted to adenosine because mammalian cells do not contain appreciable amounts of the necessary inosine kinase [31].Phosphorylated moieties are restricted to the intracellular milieu. The remaining components of the adenine nucleotide pool, the diffusible purine bases adenosine, inosine, hypoxanthine, and xanthine, were quantitatively assessed in this study to confirm the high-energy phosphate degradation seen. As total high-energy phosphates are depleted, there should be a concomitant increase in total purine concentrations. This phenomenon was observed through the antibiotic phase in our study. Compared with baseline purine levels, leaflets exposed only to cold ischemia versus those subjected to cold ischemia plus antibiotic disinfection experienced a 6% and 55% increase in purine levels, respectively (see Fig 4). Thus, the greatest increase in TDP was experienced by those groups with the longest decrease in high-energy phosphates. This confirms a certain portion of the TAN pool is in fact proceed-
Ann Thorac Surg 1993;5541>9
ing to various metabolites on the lower end of the degradatory pathway (see Fig 5). It must be noted, however, that this linear increase in TDP was blunted at the cryopreservation step. The dilutional liquid media technique for removal of dimethylsulfoxide during thaw may be responsible for a "wash out" of these bases as seen in groups IV and V. Furthermore, the mandatory immersion during disinfection may also account for the lower TDP levels found in both nonantibiotic and antibiotic-treated cryopreserved leaflets.
Time- and Process-Dependent Depletion of High-Energy Adenine Nucleotides The ATP and TAN concentrations measured at the completion of each step of the cryopreservation process indicate progressive metabolic insult to the valvular leaflet cells. This injury becomes quantitatively important at the antibiotic disinfection step when considering only ATP, whereas TAN depletion does not become significant until later, after cryopreservation. These methods analyze and characterize metabolic effects, which are potentially reversible and thus may be useful in monitoring technological modifications in homograft processing. These detailed biochemical studies suggest that cardiac valvular leaflet fibroblasts are remarkably metabolically resilient and withstand significant perturbation of energy metabolism during processing. The cell population demonstrates measurable energy reserve when the obligatory processing stages are preceded by relatively short harvestassociated warm ischemia (40 minutes) [33]. The observed metabolic response to the summation of ischemia, antibiotic disinfection, and cryopreservation suggests that these leaflet matrix cells retain reduced but not exhausted intracellular high-energy adenine nucleotide stores. Therefore, at the time of implantation, the population of matrix leaflet cells within the homograft cusps, although significantly altered, are not completely metabolically exhausted by the processing steps inherent in current cryopreservation methods. This study was supported by research grants from the American Heart Association (Capital Affiliate) and LifeNet, Inc. The assistance of Horst's Meat Co, Hagerstown, MD, is gratefully acknowledged. The assistance of Lila C. Evans is gratefully acknowledged.
References Hopkins RA. Historical development of the use of homograft valves. In: Hopkins RA, ed. Cardiac reconstructions with allograft valves. New York: Springer-Verlag, 1989:5. OBrien MF, Stafford EG, Gardner MA, Pohlner PG, McGiffin DC. A comparison of aortic replacement with viable cryopreserved and fresh allograft valves, with a note on chromosomal studies. J Thorac Cardiovasc Surg 1987;94: 812-23. VanDerKamp AW, Visser WJ, VanDongen JM, Nauta J, Galjaard H. Preservation of aortic heart valves with maintenance of cell viability. J Surg Res 1981;30:47-56. McGiffin DC, OBrien MF, Stafford EG, Gardner MA, Pohlner PG. Long-term results of the viable cryopreserved al-
DOMKOWSKI ET AL NUCLEOTIDES AND CRYOPRESERVED VALVES
Ann Thorac Surg 19!33;554159
lograft aortic valve: continuing evidence for superior valve durability. J Cardiac Surg 1988;3(Suppl):289-96. 5. Ross DN. Homograft replacement of the aortic valve. Lancet 1962;2:487. 6. Barratt-Boyes BG. Homograft aortic valve replacement and aortic incompetence and stenosis. Thorax 1964;19:131-50. 7. Campalani G, Chalmers JC, Weaver EM. Aortic valve replacement with frozen irradiated homografts: an 18-year experience. Eur J Cardiothorac Surg 1989;3:558-61. 8. Wallace RB, Giuliana ER, Trus JL. Use of aortic valve homograft for aortic valve replacement. Circulation 1971;43: 36572. 9. Heimbecker RO, Aldridge HE, Lemire G. The durability and fate of aortic valve grafts. J Cardiovasc Surg 1968;9:511-7. 10. Barratt-Boyes BG, Roche AH, Subramanyan R, Pemberton JR. Long-term follow-up of patients with the antibiotic sterilized aortic homograft valve inserted freehand in the aortic position. Circulation 1987;75:768-77. 11. Khanna SK, Ross JK, Monro JL. Homograft aortic valve replacement: seven years' experience with antibiotic-treated valves. Thorax 1981;36:33&7. 12. Thompson R, Yacoub M, Ahmed M, Somerville W, Towers M. The use of "fresh' unstented homograft valves for replacement of the aortic valve. J Thorac Cardiovasc Surg 1980;79:896-903. 13. OBrien MR, Johnston N, Stafford G, et al. A study of the cells in the explanted viable cryopreserved allograft valve. J Cardiac Surg 1988;3(Suppl):279-87. 14. Angell WW, Angell JD, Oury JH. Long-term follow-up of viable frozen aortic homografts. A viable homograft bank. J Thorac Cardiovasc Surg 1987;93:815-22. 15. Lange PL, Hopkins RA. Allograft valve banking: techniques and technology. In: Hopkins RA, ed. Cardiac reconstruction with allograft valves. New York: Springer-Verlag, 1989: 44-58. 16. Lowry OH, Rosenbrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:26575. 17. Yankah AC, Hetzer R. Procurement and viability of cardiac valve allografts. In: Yankah AC, Hetzer R, Miller DC, Ross DN, Somerville J, Yacoub MH, eds. Cardiac valve allografts 1962-1987. New York: Springer-Verlag, 1988:23-6. 18. Hu JF, Gilmer L, Hopkins RA, Wolfinbarger L Jr. Effects of antibiotics on cellular viability in porcine heart valve tissue. Cardiovasc Res 1989;23:960-4. 19. Hu JF, Gilmer L, Hopkins RA, Wolfinbarger L Jr. Assessment of cellular viability in cardiovascular tissue as studied with 3H proline and "H inulin. Cardiovasc Res 1990;24:528-31. 20. Lockey E, Nawal A, Gonzalez-Lavin L, Ross DN. A method
21.
22. 23.
24.
25.
26.
27.
28. 29.
30. 31. 32.
33.
419
of sterilizing and preserving fresh allograft heart valves. Thorax 1972;27398400. Barratt-Boyes BG. Long term follow-up of aortic valvular grafts. Br Heart J 1971;33(SuppI):60. Moore CH, Martelli V, Al-Janabi N, Ross DN. Analysis of homograft valve failure in 311 patients followed up to 10 years. Ann Thorac Surg 1975;20:274-81. Abd-Elfattah AS, Jessen ME, Hanan SA, Tuchy G, Wechsler AS. Is adenosine 5'-triphosphate derangement or freeradical-mediated injury the major cause of ventricular dysfunction during reperfusion? Circulation 1990;82(Suppl): 341-50. Bank HL, Schmehl MK, Brockbank KM. Endothelial and fibroblast viability assays for tissue allografts. In: Yankah AC, Hetzer R, Miller DC, Ross DN, Somerville J, Yacoub MH, eds. Cardiac valve allografts 1962-1987. New York: SpringerVerlag, 1988:43-8. Heacox AE, McNally RT, Brockbank KM. Factors affecting the viability of cryopreserved allograft heart valves. In: Yankah AC, Hetzer R, Miller DC, Ross DN, Somerville J, Yacoub MH, eds. Cardiac valve allografts 1962-1987. New York: Springer-Verlag, 1988:37-42. Crescenzo DG, Hilbert SL, Bamck MK, et al. Donor heart valves: electron microscopic and morphometric assessment of cellular injury induced by warm ischemia. J Thorac Cardiovasc Surg 1992;103:253-7. Barratt-Boyes BG, Roche AHG. A review of aortic valve homografts over a six and one half year period. Ann Surg 1969;170:483. OBrien M. The viable cryopreserved allograft aortic valves. J Cardiac Surg 1987;2(Suppl):153-67. Hull-Ryde EA, Lewis WR, Veronee CD, Lowe JE. Simple step gradient elution of the major high-energy compounds and their catabolites in cardiac muscle using high performance liquid chromatography. J Chromatogr 1986;377:165-74. St. Louis J, Corcoran P, Rajan S, et al. Effects of warm ischemia following harvesting of allograft cardiac valves. Eur J Cardiothorac Surg 1991;5:458-65. Swain JL, Holmes EW. Nucleotide metabolism in cardiac muscle. In: Fozzard AH, ed. The heart and cardiovascular system. New York: Raven, 1986:913. Abd-Elfattah AS, Wechsler AS. Superiority of HPLC to assay for enzymes regulating adenine nucleotide pool intermediated metabolism: 5'-nucleotides, adenylate deaminase, adenosine deaminase and adenylosuccinate lyase: a simple and rapid determination of adenosine. J Liquid Chromatogr 1987; 10:2653-94. Messier RH, Domkowski PW, Aly H, et al. High energy phosphate depletion in leaflet matrix cells during processing of cryopreserved cardiac valves. J Surg Res 1992;52:483-8.
+
+
To assess the initial metabolic phase of cellular injury from cardiac valve processing, high-energy phosphate concentrations were analyzed in valve leaflets subsequent to critical processing steps. Using a porcine model, valves were processed in a manner identical to human homografts, with 58 randomly assigned to five groups representing distinct preparation phases. Group I (controls) sustained 40 minutes of warm ischemia concluded by liquid nitrogen immersion. Remaining groups similarly endured 40 minutes of ischemia, but were subsequently prepared according to stepwise design: 11, warm ischemia 24 hours of 4OC ischemia; 111, warm ischemia 24 hours of 4OC antibiotic disinfection; IV, warm ischemia 24 hours at 4OC (without antibiotics) cryopreservation (-1OC/min cryoprotected freezing); and
V, warm ischemia disinfection cryopreservation. At each regimen's conclusion leaflet extracts were assayed by high-performance liquid chromatography for highenergy adenine nucleotides (adenosine triphosphate, adenosine diphosphate, adenosine monophosphate) and catabolites. A 47% and 86% decrease in cellular adenosine triphosphate level was observed in group 111 and group V leaflets, respectively. The level of total adenine nucleotides was maintained up to cryopreservation; thereafter a 74% decrease was noted. Catabolite analysis confirmed incomplete degradation of adenine nucleotides indicating cellular metabolic resilience throughout standard homograft preparation in valves previously exposed to 40 minutes of warm ischemia. (Ann Thorac Surg 1993;55:413-9)
C
ide, irradiation, and glutaraldehyde treatment. Most have been demonstrated to damage the tissue and thereby negatively influence long-term homograft durability [7-91. Conversely, less rigorous techniques such as maintenance in culture medium (4°C) for up to 6 weeks have provided satisfactory results [lo-121. At present, however, the majority of harvested human valves are disinfected using antibiotic formulations having low toxicity and preserved by controlled-rate freezing (-l"C/min) with 10% dimethylsulfoxide as a cryoprotectant (13, 141. Cryopreservation is a well-defined and reproducible technique [151, but other "preimplantation processing" steps occurring between valve harvest and implantation are less uniform. Details of one effective method include a rinsing of the procured valve in saline solution and storage in cold (4°C)nutrient medium for 24 hours, during which transport to processing facilities and antibiotic disinfection are usually accomplished. Valves are then cryopreserved as described and maintained in vapor phase liquid nitrogen until the time of transplantation, when they are thawed by means of a dilutional warming procedure in the operating room. Each step between donor death and implantation retains a potential for homograft alteration through injury to the cellular and matrix components. Morphologic criteria and in vitro cell viability assays have been used as a means of monitoring the effects of processing on the valvular tissue [3]; we were interested in quantitating high-energy phosphate
+
+ +
+
ongenital and acquired cardiac valvular pathology continues as a prevalent clinical problem, often requiring prosthetic replacement. Because of the morbidity associated with the conferred "prosthetic valve disease," transplanted human valves are under increasing scrutiny as they have demonstrated certain specific advantages over xenograft and mechanical replacement heart valves [l].With increased utilization, those factors potentially influencing prolonged valve durability, the major controversial issue, remain incompletely characterized. It has been proposed by some that such determinants may be related to the postimplantation cellular integrity of homograft leaflet fibroblasts [24]. Since the initial homograft aortic valve replacements by Ross and Barratt-Boyes in 1962 [5,6], many methods have been employed for preparation before storage of such tissue. Historically, these have included various methods of chemical disinfection and sterilization including highconcentration antibiotics, ppropriolactone, ethylene oxAccepted for publication May 18, 1992. Presented in part to the Society for Cryobiology, Leuven, Belgium, July 1991. The opinions or assertions contained herein are the private views of the authors and are not to be construed as conveying either an official endorsement or criticism by the US Department of Health and Human Services, the Food and Drug Administration. Address reprint requests to Dr Hopkins, Department of Surgery, UHC, Georgetown University, 3800 Reservoir Rd, NW, Washington, DC 20007.
0 1993 by The
Society of Thoracic Surgeons
0003-4975/93/$6.00
414
DOMKOWSKl ET AL NUCLEOTIDES AND CRYOPRESERVEDVALVES
concentrations as a function of the various processing steps. Intracellular concentration changes of adenine nucleotides essential for anabolic cellular function may serve as an estimate of the metabolic status of the cell. This type of metabolic assay of high-energy phosphate measures ischemic injury presumably during a prelethal injury phase. This may function as an initial measurement of ischemic stress, rather than using a less sensitive indicator of frank cell death. The purpose of this experiment was to accurately measure these phosphorylated adenine nucleotides and their products of catabolism, thereby evaluating the extent of such injury occurring consequent to cold ischemia, antibiotic disinfection, and cryopreservation.
Material and Methods Animal Preparation Fifty-eight semilunar cardiac valves were obtained from 9-month-old female Yorkshire and Poland China swine after a 40-minute period of corporeal warm ischemia between death of the animal and cardiotomy at the slaughterhouse. Aortic and pulmonary valve conduits were dissected, roots opened between adjacent cusps, and the leaflets subjected to vigorous scraping with a rubber policeman to remove endothelium. Random scanning electron microscopy was performed to verify endothelium removal. The excised leaflets were either stored in liquid nitrogen (-190°C) or refrigerated (4°C) as defined by the experimental protocol.
Experimental Groups Group I (n = 12) leaflets were immediately immersed in liquid nitrogen after harvest (ie, control; 40-minute warm ischemic time) to establish baseline adenine nucleotide concentrations. Group I1 (n = 12) leaflets were placed in serum-free RPMI 1640 nutrient medium (Gibco Co, Grand Island, NY) at 4°C for 24 hours (cold ischemic time; analogous to cold transport of donor tissue to processing centers, but in the absence of antibiotics by design) followed by rapid freezing by immersion in liquid nitrogen. Group I11 (n = 11) leaflets were subjected to 4°C antibiotic disinfection (cold disinfection time, virtually analogous to clinical protocol) followed by immersion in liquid nitrogen. The antibiotic formulation consisted of a serum-free RPMI 1640 nutrient media solution containing cefoxitin, 240 pg/mL; lincomycin, 120 pg/mL; polymyxin B sulfate, 100 pg/mL; and vancomycin, 50 pg/mL [15]. Group IV (n = 12) was subjected to cold ischemia (identical to group I1 valves, 24 hours at 4"C, in antibiotic-free media) but then cryopreserved as follows: 10% each of dimethylsulfoxide and fetal calf serum in RPMI 1640, frozen at a rate of -1"C/min to -60°C (CryoMed 1010 Micro computer), and stored in vapor phase liquid nitrogen (-190OC) (CryoMed freezer model CMS-450A, New Baltimore, MI) [15]. Group V (n = 11) leaflets were subjected to all of the processing steps, modeling a completed homograft preparation with brief warm ischemia (ie, 40 minutes of warm ischemia, 24 hours of cold antibiotic disinfection, and cryopreservation). Leaflets were thawed in single-valve trileaflet sets, and one cusp
Ann Thorac Surg 1993;55:41%9
was randomly selected for analysis of all adenine nucleotide metabolites by high-performance liquid chromatography.
Ex traction Procedure Using cold (4°C) instruments to manipulate the tissue, the selected leaflet from each valve was homogenized by hand for 30 minutes at 10°C in 12% trichloracetic acid solution. The mixture was separated by centrifugation (10 minutes, 3,000 rpm) and the soluble extract removed and neutralized using a tri-n-octy1amine:freon (1:2) mixture. After vortexing, the organic phase was separated by centrifugation (10 minutes, 3,000 rpm) and stored in 1.5-mL polypropylene microcentrifuge tubes (Baxter, McGaw Park, IL) at -70°C. The postextraction pellets of denatured leaflet protein were dissolved in 0.6 N sodium hydroxide for protein determination according to the method described by Lowry and associates (161 (Gilford Response UV-VIS spectrophotometer [Oberlin, OH] at 725 nm).
High-Performance Liquid Chromatography Analysis Aliquots (80 pL) were loaded onto a Waters Intelligent Sampler Processor model 710B high-performance liquid chromatography column (Waters Associates, Milford, MA). A solvent delivery system model 6000A (Waters Associates) along with a solvent selector model 101 (Alltech Associates, Deerfield, IL) was used. Step gradient separation of the adenine pool metabolites (adenosine triphosphate [ATE'], adenosine diphosphate [ADP], adenosine monophosphate [AMP], oxidized form of nicotinamide adenine dinucleotide "AD+], adenosine, inosine, hypoxanthine, and xanthine) was achieved using a NOVA-Pak-A (C18; 5-pm particle size; 10 x 8 mm) chromatographic column inside a radial compressor module model RCM 100 (Waters Associates) (Fig 1). Ammonium phosphate (100mmoYL; pH 5.5) was used to elute ATP, ADP, uric acid, hypoxanthine, and xanthine; NAD+ was eluted with 7% methanol; and inosine and adenosine were eluted with 40% methanol. The adenine pool metabolites were integrated and quantitated t:, external standards through a visible/ultraviolet detector model 490 and a data module model 730 (Waters Associates). Retention time, peak area, and amounts of all eluted adenine metabolites were recorded on each chromatogram and normalized to the average concentration of protein in the units of adenine metabolite per milligram of protein % standard error of the mean. Chemical reagents and buffers were purchased from Sigma Chemical Company (St. Louis, MO). High-performance liquid chromatographygrade ammonium phosphate and methanol were purchased from Fisher Scientific (Fair Lawn, NJ). Distilled and deionized water was processed through the Milli-Q System Millipore (Bedford, MA).
Sta tistical Analysis The statistical analysis program SAS (Statistical Analysis System Institute, Cary, NC) processed on an IBM 386 computer was used to determine statistical significance, where p was less than 0.05 using a one-way analysis of
DOMKOWSKI ET AL NUCLEOTIDES AND CRYOPRESERVED VALVES
Ann Thorac Surg 1993;55:41%9
415
f 2.5 a
Ic ~
F
Q 1.5 & a
ATP A
<
-
2
D
P
21 u)
HX
5 05
E
E
X
2 L
0
I
II
111
Processed Groups
IV
V
Fig 2 . Progressive degradation of adenosine triphosphate (ATP) existing at the completion of each processing phase: (1) warm ischemia, (11) cold transport, (111) antibiotic disinfection, (IV) cryopresmation, and ( V ) cryopreservation with antibiotics. (* p < 0.05 versus group I; p < 0.05, group ZV versus group I1 and group V versus group 111.)
NAD IN0 ADEN
adenosine + inosine 1, Figs 24).
Fig 7 . Representative peaks of adenine nucleotide pool metabolites as generated by step-gradient high-performance liquid chromatographic elution. Retention time, peak area, and amounts of each metabolite are recorded from each chromatogram. (ADEN = adenosine; ADP = adenosine diphosphate; AMP = adenosine monophosphate; ATP = adenosine triphosphate; HX = hypoxanthine; IN0 = inosine; NAD = nicotinaniide adenine dinucleotide, oxidized form; X = xanthine.)
+ hypoxanthine + xanthine) (Table
Group I: 40 Minutes of Warm Ischemia (Control) After 40 minutes of warm ischemia 12 randomly assayed leaflets contained the following baseline concentrations: ATP = 1.78 f 0.25 nmoYmg protein, TAN = 2.59 f 0.33 nmoYmg protein, and TDP = 3.51 2 0.41 nmoYmg protein (see Figs 2-4). Values for all metabolites are listed in Table 1.
variance between all possible combinations of each step of the experimental model. A Scheffk’s multiple comparison procedure was used to identify significant differences between treatment groups.
Group 11: 40 Minutes of Warm Ischemia 4°C Ischemia (No Antibiotics)
+ 24 Hours of
A 17% reduction in ATP (see Fig 2) was observed in group I1 leaflets (1.48f 0.12 nmoYmg protein) as compared with group I levels, indicative of some ATP degradation induced by this step. In addition, there was an 8% increase in TAN (2.815 0.34 nmoYmg protein) (seeFig 3) and a 6% increase in TDP (3.75f 0.87nmol/mg protein) (see Fig 4) as compared with group I. This commenced a trend of ATP degradation to lower energy adenine nucleotides
Results Valve leaflets (excluding endothelium) were analyzed after each processing step of a representative homograft valve cryopreservation protocol to quantify the following metabolites: ATP, total adenine nucleotides (TAN = AMP + ADP + ATP), and total diffusible purines (TDP = Table 1 . Metabolites ( n m o h x proteiny
ATP ADP AMP Adenosine Inosine Hypoxanthine Xanthine NAD+
Group 111
Group I (WIT only)
Group I1 (WIT + 24 h 4°C ischemia)
(WIT + 24 h 4°C antibiotic disinfection)
Group IV (WIT + 4°C ischemia + cryopreservation)
Group V (WIT + 4°C disinfection + cryopreservation)
1.78 2 0.25 0.59 f 0.07 0.40 f 0.10 0.18 +- 0.06 1.50 f 0.30 1.70 f 0.25 0.00 f 0.00 0.50 f 0.04
1.48 f 0.12 0.55 f 0.05 0.78 f 0.21 1.20 f 0.29 0.72 f 0.20 0.78 f 0.16 1.10 f 0.63 0.33 f 0.04
0.91 f 0.13b 0.33 f 0.07 1.00 f 0.40 6.10 ? 3.50 0.87 f 0.20 0.73 f 0.24 0.10 f 0.08 0.29 f 0.12
0.46 f O . l l b
0.25 f 0.04b 0.25 f 0.03 0.18 f 0.07 3.10 f 0.70 0.71 f 0.13 0.07 k 0.03 0.05 2 0.04 0.23 2 0.04
0.27 k 0.05 0.14 5 0.07 0.52 f 0.10 0.50 0.50 0.12 f 0.06 0.09 f 0.08 0.07 f 0.04
*
Adenine nucleotide metabolite concentrations (* standard error of the mean) existing at the completion of each step of preimplantation processing. p < 0.05 versus group I by analysis of variance.
ADP = adenosine diphosphate; AMP = adenosine monophosphate; dinucleotide, oxidized form; WIT = warm ischemia.
ATP
=
adenosine triphosphate;
NAD+ = nicotinamide adenine
416
.-c Q 2 a L
DOMKOWSKI ET AL NUCLEOTIDES AND CRYOPRESERVED VALVES
Ann Thorac Surg 1993;55:41>9
3.5
T
3
I
II
111
Processed Groups
IV
V
Fig 3 . Total adenine nucleotides (TAN)existing at the completion of each processing phase: (I) warm ischemia, (II)cold transport, (III) antibiotic disinfection, (IV)cryopreserwtion, and N)cryopreservalion with antibiotics. (* p < 0.05 versus groups I , 11, and 111.)
and ultimately to purine byproducts (both TAN and TDP were increased). However, statistical significance was not reached by this group of valves.
Group Ill: 40 Minutes of Warm Ischemia of 4°C Ischemia With Antibiotics
+ 24 Hours
Antibiotic disinfection resulted in a 49% reduction (see Fig 2) in ATP with respect to control leaflets (0.91 f 0.13 nmoYmg protein; p 5 0.0001 versus group I). However, TAN concentrations in group 111 (2.28 2 0.48 nmoYmg protein) were not significantly reduced. It should be emphasized that through the completion of the antibiotic step, total high-energy phosphate metabolites (TAN) were statistically maintained at baseline 40-minute levels (see Fig 3), indicating that during these stages ATP degradation proceeds mainly to its lower energy phosphates (ie, ADP and AMP). However, levels of diffusible purines were 55% higher (7.83 + 1.84 nmoYmg protein) than in group I (p = 0.0006) (see Fig 4). This accumulation of byproducts is consistent with degradation of ATP to TAN and partial TAN degradation to adenosine, inosine, hypoxanthine, and xanthine.
Group IV: 40 Minutes of Warm Ischemia + 24 Hours of 4°C Antibiotic-Free Ischemia + Cyopreservation The consequences of cryopreservation were evaluated in groups IV and V, which differed by the presence or
I
II
111
Processed Groups
IV
V
Fig 4 . Total diffusible purines (TDP) at the completion of each phase of cardiac valve transplant preparation: (I)warm ischemia, (IIcold ) antibiotic disinfection, (IV)cryopresmation, and N) transport, fIII) cyopreservatwn with antibiotics. C‘ p < 0.05 versus groups I , 11, and V by analysis of variance.)
absence of antibiotics. In group IV, the valves exposed to cryopreservation without antibiotics, there was a significant depletion of ATP by 74% (0.46 f 0.11 nmol/mg protein) compared with control (p 5 0.0001) (see Fig 2). The 69% reduction (p = 0.0001) in ATP between groups I1 and IV (see Fig 2) quantitates the effect of cryopreservation and warm and cold ischemia without exposure to antibiotics. Similarly, cryopreservation without antibiotics effected a 66% TAN depletion from baseline (0.87 f 0.15 nmoYmg protein) (see Fig 3). With the addition of cryopreservation to 24 hours of cold ischemia (group IV), there was a 69% diminution (p = 0.OOOl) in TAN from antibiotic-free cold ischemia levels (group II). There was a significant change in total diffusible purines (1.24 f 0.49 nmoYmg protein) in group IV (p < 0.05 versus baseline).
Group V : 40 Minutes of Warm Ischemia, 24 Hours of 4°C Antibiotic Disinfection, and Cryopreservation (Complete Cryopreservation Protocol) Analysis of ATP in group V (0.25 f 0.04 nmoYmg protein) measures the cumulative effects of all steps involved in the preparation of clinical homograft cardiac valves. There was an 86% (p = 0.OOOl) and 83% (p = 0.OOOl) reduction in ATP as compared with baseline (group I) and 24 hours of cold ischemia without antibiotics (group 11), respectively (see Fig 2). Cells in leaflets exposed to both antibiotics and cryopreservation demonstrated a 73% reduction in ATP (p = 0.0001) versus leaflets experiencing 24 hours of cold ischemia with antibiotics (group III) (see Fig 2). The effect of all steps (group V) on TAN levels (0.68 2 0.09 nmoYmg protein) was simiiar to that seen with group IV. Concentrations of AMP, ADP, and ATP in the fully processed leaflets demonstrated a 74% (p = 0.0001) reduction as compared with group I (see Fig 3). The total diffusible purine concentration in fully processed leaflets (3.56 f 0.82 nmoYmg protein) was 55% less than the purine level in leaflets assayed after antibiotic disinfection alone (see Fig 4). The NAD+ levels remained essentially unchanged (0.49 f 0.06 nmoYmg protein) until the cryopreservation steps, which resulted in a slight reduction.
Comment This study of porcine cardiac valves was designed to
mimic favorable harvest conditions in which the warm ischemic time was limited to 40 minutes, correlating with a brief period between cessation of donor heartbeat and procurement, such as might occur in a multiple organ harvest setting. Experimental tissue exposed to this degree of ischemia conveniently served as control groups with which to compare metabolite levels after later preimplantation processing steps. Twenty-four hours of cold ischemia did not cause reductions in total high-energy phosphate concentrations from baseline (see Fig 3). However, 24 hours of cold ischemia resulted in a decrease in the ATP level that, although not statistically significant, was subsequently observed throughout processing. This finding indicates that although the ATP level is diminished at this stage, the majority of degradation occurs
Ann Thorac Surg 1993;55:41>9
only through to ADP and AMP (reflected in total adenine nucleotide concentrations; confirmed by the measured increases in ADP and AMP) due to incomplete breakdown of ATP or partial restoration by alternate pathways. The purpose of this particular experimental grouping, however, was to differentiate the cold ischemic effects alone (group 11) from cold ischemia combined with disinfection (group 111).
Antibiotics and Processing-Associated lnjury Many aggressive antibiotic regimens that include antifungal and high-dose antibiotic mixtures have been detrimental to fibroblast viability [17], but antibiotic combinations similar to the one employed in this study appear relatively benign as demonstrated by tissue culture and proline uptake assessment [ 18-20]. Independent of antibiotic toxicity [21, 221, this study has substantiated that metabolic depletion is, in fact, occurring during 24 hours of cold ischemia and antibiotic disinfection. However, in this study the majority of ATP degradation occurring during antibiotic disinfection generates only to ADP and AMP, which is noteworthy considering that it is the conservation of total high-energy phosphates (not ATP alone) that is critical to a cell's ability to maintain basal function and integrity [23]. Such changes are potentially reversible (eg, conversion of ADP and AMP to ATP) and appear to be likely due to the ischemia associated with the time necessary for disinfection rather than chemical injury. The effect of disinfection disappears when comparing cryopreservation with and without antibiotics, because cryopreservation after disinfection results in an additional 46% depletion of ATP. Regardless of whether prior antibiotic disinfection treatment is used, apparently the cryopreservation phase is responsible for the largest decrease in ATP. This information can be used to develop disinfection strategies that may be even less metabolically costly. Total avoidance of antibiotic disinfection with the subsequent enhanced risk for transmission of infectious disease to the recipient does not appear to be worth the minimal "gain" of reduced metabolic cellular leaflet injury and its theoretical effect on long-term performance.
DMSO Cryopreservation and Cellular Metabolic Injury With subsequent cryopreservation, significant reductions in high-energy phosphates were observed. Cryopreservation alone, subsequent to only warm ischemia and 24 hours of 4°C ischemia without antibiotics, was responsible for a substantial depletion of ATP (74%)(see Fig 2), which was greater than that imparted by cold ischemia (17%)or cold disinfection (47%).The effects of all preprocessing steps were synergistic in the diminution of adenine nucleotides. Nevertheless, leaflet cells still retained measurable amounts of high-energy phosphates after all of the processing steps (see Fig 3).
Metabolic Depletion, Cell Injury, Viability, and Durability Determination of the relationship between valvular leaflet fibroblast viability and the long-term function and durability of homografts has been the focus of a few investi-
DOMKOWSKI ET AL NUCLEOTIDES AND CRYOPRESERVED VALVES
417
gations and numerous postulations [17, 24, 251. Before this interesting question can be addressed, the initial response of the leaflet fibroblast cell population to ischemia and subsequent processing steps must be defined. Adenine nucleotide quantitation is not by itself a viability assay; however, it is a specific indicator of the overall metabolic state of the leaflet fibroblast cell population. Our data define a progressive diminution of high-energy phosphate with each step of preimplantation processing. It is reasonable to assume that as these nucleotides are increasingly depleted, a percentage (without attempting to assign a value) of fibroblasts will lose their ability to structurally maintain cell membranes, thus leading to irreversible injury. It is possible that only after relatively complete depletion of high-energy phosphates in the cell population (which this study did not find, even in the fully processed groups) will irreversible metabolic injury be manifest in the majority of fibroblasts. Unlike specific morphological assays [26], which examine irrefutable structural criteria for injury or death of fibroblasts, the aim of this investigation was to delineate the energy reserve of the leaflet fibroblast population. Few studies have carefully analyzed the early injury response (before irreversible cell injury), a period when such insults may be reversible and responsive to technical modifications, thereby potentially mitigating later effects on ultimate viability. Such changes in high-energy phosphate status might also precede reductions in energy-dependent cell functions such as proliferation, protein transport, and synthetic activity [19]. The hypothesis that cellular viability is a critical determinant of long-term clinical performance rests on the notion that it is the extent of cellular viability at the time of implantation that is the key issue [4]. If harvesting and processing protocols are to be designed to maximize cellular viability, then studies like this one are fundamental to understanding the effects of preimplantation processing on heart valve tissue [26-281. An alternative concept might be that it is the extent of functional viability (ie, fibroblast capable of proliferation and structural protein synthesis) some weeks or months after transplantation that is critical to enhancing valve durability. This being the case, processing methods that result in cells that not only "appear" viable but that have significant metabolic reserve with which to weather the peritransplantation stress could lead to better leaflet fibroblast functional recovery.
High-Performance Liquid Chromatography High-performance liquid chromatography assays have been employed to quantitate adenine nucleotide concentrations in myocardial cells after ischemia and reperfusion injury [29]. This study demonstrates the feasibility of this method for detecting nanomolar concentrations of the adenine nucleotide metabolites in cardiac valvular leaflets as well. Other methods for adenine nucleotide determination include enzymatic analysis and nuclear magnetic resonance; however, the former depends on generation or consumption of a substrate in subcellular reactions, whereas the latter requires fairly large samples [30,311.
418
DOMKOWSK~ET AL NUCLEOTIDES AND CRYOPRESERVED VALVES
ATP
/-t
&
1
ADP
( 31
& ADENOSINE & INOSINE 14, & HYPOXANTHINE (51 & XANTHINE (6: ' (7) & URIC ACID
Fig 5 . Degradation of adenosine triphosphate (ATP) to its lower energy metabolites and constitutive purine bases. Circled numbers indicated principal enzymes responsible for ATP catabolism: 1 = ATP kinases, 2 = adenylate kinase, 3 = cytosol 5'-nucleotidase, 4 = adenosine deaminase, 5 = purine nucleoside phosphoylase, 6 = xanthine oxidase, 7 = xanthine oxidase. (ADP = adenosine diphosphate; AMP = adenosine monophosphate.)
The progression of biochemical changes in the leaflet fibroblasts has been examined with proton ('H) and phosphorous (3'P) magnetic resonance spectroscopy, and demonstrated a time-dependent depletion of ATP stores reflected by an increase in inorganic phosphate; after 2 hours of harvest-related ischemia, increases in lactate concentration were measured by the nuclear magnetic resonance technique indicating the onset of anaerobic These data are quite consistent with our metabolism [N). high-performance liquid chromatography results. An advantage of the high-performance liquid chromatography technique resides in its utility for analysis of the entire high-energy phosphate pool in a "snapshot" fashion throughout the degradatory cascade [32] (Fig 5). All catabolism reactions up to and including the production of adenosine are completely reversible and allow for the possibility of adenosine phosphorylation into a highenergy metabolite and reutilization. Conversely, the freely diffusible purine inosine cannot be converted to adenosine because mammalian cells do not contain appreciable amounts of the necessary inosine kinase [31].Phosphorylated moieties are restricted to the intracellular milieu. The remaining components of the adenine nucleotide pool, the diffusible purine bases adenosine, inosine, hypoxanthine, and xanthine, were quantitatively assessed in this study to confirm the high-energy phosphate degradation seen. As total high-energy phosphates are depleted, there should be a concomitant increase in total purine concentrations. This phenomenon was observed through the antibiotic phase in our study. Compared with baseline purine levels, leaflets exposed only to cold ischemia versus those subjected to cold ischemia plus antibiotic disinfection experienced a 6% and 55% increase in purine levels, respectively (see Fig 4). Thus, the greatest increase in TDP was experienced by those groups with the longest decrease in high-energy phosphates. This confirms a certain portion of the TAN pool is in fact proceed-
Ann Thorac Surg 1993;5541>9
ing to various metabolites on the lower end of the degradatory pathway (see Fig 5). It must be noted, however, that this linear increase in TDP was blunted at the cryopreservation step. The dilutional liquid media technique for removal of dimethylsulfoxide during thaw may be responsible for a "wash out" of these bases as seen in groups IV and V. Furthermore, the mandatory immersion during disinfection may also account for the lower TDP levels found in both nonantibiotic and antibiotic-treated cryopreserved leaflets.
Time- and Process-Dependent Depletion of High-Energy Adenine Nucleotides The ATP and TAN concentrations measured at the completion of each step of the cryopreservation process indicate progressive metabolic insult to the valvular leaflet cells. This injury becomes quantitatively important at the antibiotic disinfection step when considering only ATP, whereas TAN depletion does not become significant until later, after cryopreservation. These methods analyze and characterize metabolic effects, which are potentially reversible and thus may be useful in monitoring technological modifications in homograft processing. These detailed biochemical studies suggest that cardiac valvular leaflet fibroblasts are remarkably metabolically resilient and withstand significant perturbation of energy metabolism during processing. The cell population demonstrates measurable energy reserve when the obligatory processing stages are preceded by relatively short harvestassociated warm ischemia (40 minutes) [33]. The observed metabolic response to the summation of ischemia, antibiotic disinfection, and cryopreservation suggests that these leaflet matrix cells retain reduced but not exhausted intracellular high-energy adenine nucleotide stores. Therefore, at the time of implantation, the population of matrix leaflet cells within the homograft cusps, although significantly altered, are not completely metabolically exhausted by the processing steps inherent in current cryopreservation methods. This study was supported by research grants from the American Heart Association (Capital Affiliate) and LifeNet, Inc. The assistance of Horst's Meat Co, Hagerstown, MD, is gratefully acknowledged. The assistance of Lila C. Evans is gratefully acknowledged.
References Hopkins RA. Historical development of the use of homograft valves. In: Hopkins RA, ed. Cardiac reconstructions with allograft valves. New York: Springer-Verlag, 1989:5. OBrien MF, Stafford EG, Gardner MA, Pohlner PG, McGiffin DC. A comparison of aortic replacement with viable cryopreserved and fresh allograft valves, with a note on chromosomal studies. J Thorac Cardiovasc Surg 1987;94: 812-23. VanDerKamp AW, Visser WJ, VanDongen JM, Nauta J, Galjaard H. Preservation of aortic heart valves with maintenance of cell viability. J Surg Res 1981;30:47-56. McGiffin DC, OBrien MF, Stafford EG, Gardner MA, Pohlner PG. Long-term results of the viable cryopreserved al-
DOMKOWSKI ET AL NUCLEOTIDES AND CRYOPRESERVED VALVES
Ann Thorac Surg 19!33;554159
lograft aortic valve: continuing evidence for superior valve durability. J Cardiac Surg 1988;3(Suppl):289-96. 5. Ross DN. Homograft replacement of the aortic valve. Lancet 1962;2:487. 6. Barratt-Boyes BG. Homograft aortic valve replacement and aortic incompetence and stenosis. Thorax 1964;19:131-50. 7. Campalani G, Chalmers JC, Weaver EM. Aortic valve replacement with frozen irradiated homografts: an 18-year experience. Eur J Cardiothorac Surg 1989;3:558-61. 8. Wallace RB, Giuliana ER, Trus JL. Use of aortic valve homograft for aortic valve replacement. Circulation 1971;43: 36572. 9. Heimbecker RO, Aldridge HE, Lemire G. The durability and fate of aortic valve grafts. J Cardiovasc Surg 1968;9:511-7. 10. Barratt-Boyes BG, Roche AH, Subramanyan R, Pemberton JR. Long-term follow-up of patients with the antibiotic sterilized aortic homograft valve inserted freehand in the aortic position. Circulation 1987;75:768-77. 11. Khanna SK, Ross JK, Monro JL. Homograft aortic valve replacement: seven years' experience with antibiotic-treated valves. Thorax 1981;36:33&7. 12. Thompson R, Yacoub M, Ahmed M, Somerville W, Towers M. The use of "fresh' unstented homograft valves for replacement of the aortic valve. J Thorac Cardiovasc Surg 1980;79:896-903. 13. OBrien MR, Johnston N, Stafford G, et al. A study of the cells in the explanted viable cryopreserved allograft valve. J Cardiac Surg 1988;3(Suppl):279-87. 14. Angell WW, Angell JD, Oury JH. Long-term follow-up of viable frozen aortic homografts. A viable homograft bank. J Thorac Cardiovasc Surg 1987;93:815-22. 15. Lange PL, Hopkins RA. Allograft valve banking: techniques and technology. In: Hopkins RA, ed. Cardiac reconstruction with allograft valves. New York: Springer-Verlag, 1989: 44-58. 16. Lowry OH, Rosenbrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:26575. 17. Yankah AC, Hetzer R. Procurement and viability of cardiac valve allografts. In: Yankah AC, Hetzer R, Miller DC, Ross DN, Somerville J, Yacoub MH, eds. Cardiac valve allografts 1962-1987. New York: Springer-Verlag, 1988:23-6. 18. Hu JF, Gilmer L, Hopkins RA, Wolfinbarger L Jr. Effects of antibiotics on cellular viability in porcine heart valve tissue. Cardiovasc Res 1989;23:960-4. 19. Hu JF, Gilmer L, Hopkins RA, Wolfinbarger L Jr. Assessment of cellular viability in cardiovascular tissue as studied with 3H proline and "H inulin. Cardiovasc Res 1990;24:528-31. 20. Lockey E, Nawal A, Gonzalez-Lavin L, Ross DN. A method
21.
22. 23.
24.
25.
26.
27.
28. 29.
30. 31. 32.
33.
419
of sterilizing and preserving fresh allograft heart valves. Thorax 1972;27398400. Barratt-Boyes BG. Long term follow-up of aortic valvular grafts. Br Heart J 1971;33(SuppI):60. Moore CH, Martelli V, Al-Janabi N, Ross DN. Analysis of homograft valve failure in 311 patients followed up to 10 years. Ann Thorac Surg 1975;20:274-81. Abd-Elfattah AS, Jessen ME, Hanan SA, Tuchy G, Wechsler AS. Is adenosine 5'-triphosphate derangement or freeradical-mediated injury the major cause of ventricular dysfunction during reperfusion? Circulation 1990;82(Suppl): 341-50. Bank HL, Schmehl MK, Brockbank KM. Endothelial and fibroblast viability assays for tissue allografts. In: Yankah AC, Hetzer R, Miller DC, Ross DN, Somerville J, Yacoub MH, eds. Cardiac valve allografts 1962-1987. New York: SpringerVerlag, 1988:43-8. Heacox AE, McNally RT, Brockbank KM. Factors affecting the viability of cryopreserved allograft heart valves. In: Yankah AC, Hetzer R, Miller DC, Ross DN, Somerville J, Yacoub MH, eds. Cardiac valve allografts 1962-1987. New York: Springer-Verlag, 1988:37-42. Crescenzo DG, Hilbert SL, Bamck MK, et al. Donor heart valves: electron microscopic and morphometric assessment of cellular injury induced by warm ischemia. J Thorac Cardiovasc Surg 1992;103:253-7. Barratt-Boyes BG, Roche AHG. A review of aortic valve homografts over a six and one half year period. Ann Surg 1969;170:483. OBrien M. The viable cryopreserved allograft aortic valves. J Cardiac Surg 1987;2(Suppl):153-67. Hull-Ryde EA, Lewis WR, Veronee CD, Lowe JE. Simple step gradient elution of the major high-energy compounds and their catabolites in cardiac muscle using high performance liquid chromatography. J Chromatogr 1986;377:165-74. St. Louis J, Corcoran P, Rajan S, et al. Effects of warm ischemia following harvesting of allograft cardiac valves. Eur J Cardiothorac Surg 1991;5:458-65. Swain JL, Holmes EW. Nucleotide metabolism in cardiac muscle. In: Fozzard AH, ed. The heart and cardiovascular system. New York: Raven, 1986:913. Abd-Elfattah AS, Wechsler AS. Superiority of HPLC to assay for enzymes regulating adenine nucleotide pool intermediated metabolism: 5'-nucleotides, adenylate deaminase, adenosine deaminase and adenylosuccinate lyase: a simple and rapid determination of adenosine. J Liquid Chromatogr 1987; 10:2653-94. Messier RH, Domkowski PW, Aly H, et al. High energy phosphate depletion in leaflet matrix cells during processing of cryopreserved cardiac valves. J Surg Res 1992;52:483-8.