A Comparison of Solutions for Lung Preservation Using Pulmonary Alveolar Type II Cell Viability

A Comparison of Solutions for Lung Preservation Using Pulmonary Alveolar Type I1 Cell Viability Mitsuhiro Hachida, M.D., Dave S. B. Hoon, Ph.D., and Donald L. Morton, M.D.

ABSTRACT Many special solutions have been developed to protect the ischemic lung in preparation for transplantation. To determine an effective solution, we isolated pulmonary alveolar type I1 cells from rat lungs. These cells play an important role in sodium transport and the production of surfactant; thus, they are crucial to the respiratory physiology of the lung. In this study, we examined in vitro the effect of various solutions such as Collins’ solution, Collins-Sacks solution, and glucose-insulin-potassium solution on alveolar type I1 cell viability. The cell viability was examined with a trypan blue dye exclusion test and I3H1thymidine uptake proliferation assay after 24 and 72 hours of incubation. The alveolar type I1 cells in the glucose-insulin-potassium solution had greater viability compared with cells cultured in either Collins’ or CollinsSacks solution. This study demonstrates that glucoseinsulin-potassium solution has the least toxic effect on isolated alveolar type I1 cells compared with other preserving solutions. Numerous attempts have been made at lung preservation, and special solutions for lung preservation (pulmoplegic solutions) have been developed to protect the lung during ischemia [ 1, 21. However, there is still little information available about the effects of pulmoplegic solutions on the ischemic lung. The primary reason may be that the evaluations have been performed only with global lung ischemia models followed by lung transplantation, which include many varying factors. Pulmonary alveolar cells, which are predominantly pulmonary type I1 cells, play a major role in pulmonary physiological functions such as sodium transport. They are further involved in the production of surfactant [3]. To assess the detailed effects of the composition of pulmoplegic solutions, we designed an in vitro experiment focusing on the viability of the alveolar cells as an indicator. In this study, we extracted pulmonary alveolar type I1 cells from rat lungs and evaluated the effects of Collins’ solution, Collins-Sacks (CS) solution, and glucose-insulin-potassium (GIK) solution on pulmonary alveolar cells in vitro.

Material and Methods

Extraction of Pulmonary Epithelial Cells Two balanced salt solutions were prepared using a previously described method [4, 51. Solution 1 contained 136.0 mM NaCl, 2.2 mM Na,HPO,, 5.3 mM KC1, 10 mM N-(2-hydroxyethyl)-piperazine-N’-2-ethanesulfonic acid buffer, and 5.6 mM glucose (pH 7.4 at 25°C). Solution 2 was Solution 1 plus 1.9 mM CaC1, and 1.3 mM MgSO,. An emulsion of bovine serum albumin and fluorocarbon in Solution 2 was also prepared. An elastase solution (7 U/ml in Solution 2) was used to isolate the type I1 cells. Elastase was obtained from Washington Biochemical (Freehold, NJ). Harvesting Cells Lungs were removed from male-specific pathogen-free Sprague-Dawley rats (weight, 180-220 gm). Solution I was used to perfuse the lungs to remove blood and to lavage the airways. The excised lungs were then filled with fluorocarbon/albumin emulsion and incubated at 37°C for 20 minutes. After repeated lavage with Solution 1, the elastase solution was instilled into the lungs to total lung capacity. The lungs were again incubated for 20 minutes at 37°C and were subsequently minced and sequentially filtered to obtain a crude cell mixture. Viability was determined by the trypan blue dye exclusion test. Cell counts were obtained using a hemocytometer. A discontinuous metrizamide buoyant density gradient was next prepared by layering metrizamide solutions. The crude cell mixture was layered on top of the gradient and centrifuged at 160 g for 20 minutes at 4°C (Sorvall RT6000; DuPont, Wilmington, DE). Type I1 cells were found in a broad band throughout the gradient, while alveolar macrophages, polymorphonuclear leukocytes, and clumped red blood cells pelleted. Cells from the gradient band were removed and centrifuged twice in cold Solution 2 for 10 minutes at 200 g. Aliquots were then resuspended in Eagle modified minimum essential medium with 10%fetal bovine serum. Approximately 28 x lo6 cells per pair of rat lungs were obtained from the metrizamide density gradient in which 67% were type I1 cells with a viability exceeding 95%. The other types of cells were macrophage/monocyte, 8.8 & 1.1%;lymphocytes, 19.2 ? 1.8%; eosinophils, 1.5 0.2%; and monocytes, 0.97 k 0.21%. After harvesting the cells from the rats, the number of extracted cells was counted sequentially and 5 x lo4 were plated into 200 ~1 of each pulmoplegic solution. Subsequently, they were incubated at 37°C and harvested after 24 or 72 hours.

*

From the Division of Surgical Oncology, Department of Surgery, UCLA School of Medicine, Los Angeles, CA. Accepted for publication Dec 29, 1987. Address reprint requests to Dr. Hachida, Division of Surgical Oncology, 9th Floor, Louis Factor Bldg, UCLA School of Medicine, Los Angeles, CA 90024.

643 Ann Thorac Surg 45:643-646, June 1988. Copyright 0 1988 by The Society of Thoracic Surgeons

644 The Annals of Thoracic Surgery Vol45 No 6 June 1988

Assessment of Viubility Eight experiments for each solution were performed. The isolated cells were seeded in 96-well microplates (Linbro; Flow Laboratories, CA) with 200 pl of each solution. The cells were incubated for 24 or 72 hours in a humidified 7% carbon dioxide atmosphere. Cells were then harvested with 10% trypsin and assayed for viability (61. The viable cells were counted under the microscope within 10 minutes. The mean number of viable cells in four separate fields was calculated in each experiment. To detect the metabolic function of cells in the various solutions, [3H]thymidine uptake proliferation assay was performed. These isolated cells were also seeded in 96-well microplates (Linbro) in the same fashion. The cells were incubated for 24 and 72 hours in a humidified 7% carbon dioxide atmosphere and then pulsed for 6 hours with [3H]thymidine 20 pCi/well. Supernatants from the microwells were harvested with a PHD harvester (Cambridge Technology, Cambridge, MA), and the incorporation of [3H]thymidine was measured by liquid scintillation counting. Data were analyzed on the basis of counts per minute (cpm). The mean count per minute value of five duplicates was used for all calculations. Preparation of Solutions The pulmoplegic solutions had the following compositions: Collins’ solution (gm/L)

KH,PO, K,P0,-3HZ0 KCl

NaHCO,

2.05 9.7

1.12 0.84

CS Solution (gm/L)

K,HPO, KHJ’O,

Ethylenediaminetetraacetic acid Mannitol

NaHCO, KHCO, MgC1,.6H2O

7.4 4.76 0.075 37.5

1.26 1.0 1.62

Insulin (unitsiL) NaH,PO,

Na,HPO, KC1

Mannitol

c1-

Solution

pH

Na+ (mEq)

K+ (mEq)

collins1 Collins-Sacks

7.4 7,4

10 17

115 115

15

180

40

410

GIK

7.4

10

40

35

350

(mEq)

Osmolarity

GIK = glucose-insulin-potassium.

statistical comparison between the solutions, the unpaired Student‘s t test was used.

Results

Trypun Blue Exclusion The pulmonary alveolar cells were harvested after 24 and 72 hours of incubation at 37°C. The mean viability after 24 hours of incubation was 79.9 k 7.4% in Collins’ solution, 86.1 & 8.7% in CS solution, and 96.1 6.8% in GIK solution (N = 8). After 72 hours of incubation, the cell viability was 66.6 ? 3.6% in Collins’ solution, 62.6 2 2.2% in CS solution, and 84.7 & 4.4% in GIK solution (N = 8) (Fig I). Although the number of viable cells decreased in each solution, the viability of cells in GIK solution was significantly higher than that in the other two solutions for both incubation periods ( p < 0.005).

*

l3HIThymidine lncorporation After 24 hours of incubation in each solution, the [3H]thymidine uptake was 3,502 2 646 cpm in Collins’ solution, 5,695 ? 851 cpm in CS solution, and 8,993 & 132 cpm in the GIK solution ( p < 0.01). After 72 hours of incubation, [3H]thymidine uptake was 2,716 & 267 cpm in Collins’ solution, 5,359 ? 360 cpm in CS solution, and 7,015 ? 1,355 cpm in GIK solution ( p < 0.05) (Fig 2). The metabolite activity, as measured with the [3H]thymidine incorporation assay, also showed a better survival in GIK solution. Thus, these results suggest that GIK solution might be less toxic to the pulmonary alveolar cells than either Collins’ or CS solution.

Comment

GIK Solution (gm/L)

Glucase

Solutions for Lung Preservation

5.0 60 0.6 0.4

1.5

2.5

Electrolyte solution for kidney preservation (Travenol, Deerfield, IL) was used as Collins’ solution. CS solution was made by the University of California at Los Angeles Pharmacology Department, and GIK solution was made by our laboratory. The pH, Na+, K+, and C1- concentration, and osmolarity of each solution were measured and are shown in the Table. The results are presented as means ? SEM. For

The major role of a pulmoplegic solution is to provide prompt hypothermia to the whole lung without toxicity to the various types of specialized cells in the lung tissue, such as vascular endothelial cells and pulmonary alveolar type I1 cells. Since the lung is an anatomically complex organ composed of a heterogeneous population of cells [5], it is difficult to investigate the effects of the pulmoplegic solution on these various cells. Currently, there is no reliable experimental method to compare the effects of the pulmoplegic solution, except by surgical procedure. However, the practical preservation protocol involves too many factors to accurately analyze the effects of pulmoplegic solutions during lung preservation on such a heterogeneous cellular population. Hino and colleagues [6] successfully evaluated lung cell via-

645 Hachida et al: Solutions for Lung Preservation

f f

UHDUS

Fig 1. In vitro viability of the pulmonary alveolar type 11 cells, cultured in three different pulmoplegic solutions, was assessed by the trypan blue exclusion test and expressed as percentage of viable cells. Glucose-insulin-potassium (GIK) solution showed the best viability after 24 and 72 hours of incubation. Asterisk indicates significant difference (p < 0.005) in viability.

bility after transplantation using the trypan blue dye exclusion method. Their study showed a good relationship between cell viability and graft viability after lung preservation followed by transplantation [6]. However, the cell viability and the cell metabolic function of particular cell types were not evaluated. During preservation, the pulmoplegic solution remains in the pulmonary capillary bed. Therefore, vascular endothelial cells are primarily exposed to the pulmoplegic solution. The pulmoplegic solutions tested have been used successfully for kidney and heart preservation, demonstrating beneficial effects in suppressing ischemic injury to such organs. No critical damage to either the capillaries or the endothelial cells has been reported. However, as many micropores exist between the pulmonary capillary bed and the alveolar epithelial cells [3], pulmonary epithelial cells could be exposed to the solution through the pores in an alveolar-capillary unit. In comparison to systemic endothelia, the pulmonary endothelium is known to be “leaky.“ Fluid and some colloid molecules normally pass from plasma to the interstitium of the lung tissue [9, 101. Thus, these studies suggest that the effects of solutions on the alveolar type I1 cells may be more critical than the effects on endothelial cells during lung preservation. Although pulmonary alveolar type I1 cells constitute only 10 to 15% of the total lung cells, they have recently been isolated in relatively pure yield [5, 101. These cells are important for synthesis, storage, and secretion of the alveolar surfactant [3]. In addition, they are the apparent stem cells for the damaged alveolar epithelium [5].

I

COLLINS 6 SACKS

72HDUS

These pulmonary alveolar cells are also related to sodium transport in the alveolar-capillary unit [9, 131, and alveolar type I1 cells are exceptionally vulnerable to physiological stress, like ischemia [12]. Injury of these cells due to ischemia or pulmoplegic solution causes respiratory distress; severe tissue deterioration exhibits similar symptoms to those induced by the respiratory distress syndrome [12]. This process suggests that protection of these alveolar epithelial cells is critical in the development of a better pulmoplegic solution. Various types of solutions have been considered, such as those of Collins and colleagues [14] and Sacks and associates [15]. Collins’ and CS solutions are the most frequently used solutions for lung and heart-lung preservation [2]. GIK solution was originally developed for cardioplegia and showed beneficial effects in the enhancement of anaerobic glycolysis of the ischemic heart [7, 81. According to previous animal experiments, solutions used successfully in lung preservation were of the intracellular type with low sodium and high potassium concentrations and large amounts of phosphate bases. The CS solution has approximately the same intracellular ionic composition as Collins’ solution, but it contains mannitol instead of dextrose and has a higher osmolarity. GIK solution contains different compositions from conventional solutions and consists of glucose, insulin, potassium, phosphate buffer, and mannitol. The glucose in GIK solution enhances the rate of anaerobic glycolysis, reverses the ion loss, alters impaired membrane electrophysiology, decreases plasma free fatty acid concentration, and alters plasma osmolarity [16]. Furthermore, insulin reduces sodium permeability and stimulates active Na+ efflux [17, 181. Wichert [19] indicated that glucose levels in lung tissue declined much faster than any other biochemical indicators during ischemia. Glucose was no longer found after 120 and 180 minutes of lung ischemia at 37°C. These studies indicate that lung tissue almost exclusively metabolizes glucose to

646 The Annals of Thoracic Surgery Vol45 No 6 June 1988

1 p3 COLLINS'

3c

'*T

1

COLLINS' P SACKS'

3H-TdR UPTAKE

(cpm)

24 HMRS

Fig 2 . The viability of the cells cultured in the solutions was determined using the thymidine uptake assay ([3H]-TdR).The cells were cultured for 24 and 72 hours in the presence of diferent pulmoplegic solutions. Single (p < 0.01) and double asterisks (p < 0.05) indicate significant differences among the solutions for each incubation time. (GIK = glucose-insulin-potassium.~

meet its energy demand. Moreover, Hess and colleagues [20] suggested that GIK solution may act as a scavenger of oxygen free radicals. Based on our results, cells incubated in GIK solution had 96.1 and 84.7% viability after 24 and 72 hours of incubation, respectively, and showed a significantly better survival rate during t3H]thymidine incorporation test. We presume the effects of GIK solution for membrane electrophysiology create the least toxic environment for the alveolar epithelial cells. This assumption suggests that these beneficial effects of GIK solution in vitro may relate to improved protection of ischemic lung cells in vivo. We are extremely grateful to Professor E. Crandall and Mrs. S. Brown for consultations. We also thank Mr. Lee Wilcox for assistance with extracting cells and Mr. Anatoly Bulkin for preparing the manuscript.

References 1. Pinskar KL, Montefusco C, Yipintsoi T, et al: Total in vivo functional adequacy of canine lung autografts after 24-hr preservation. Transplant Proc 11:599, 1979 2. Haverich A, Scott WC, Jamieson SW: Twenty hours of lung preservation: a review. Heart Transplant 4:234, 1985 3. Said SI: The Pulmonary Circulation and Acute Injury. New York, Futura, 1985, pp I S 2 7 4. Dobbs LG, Mason RJ: Pulmonary type I1 cells isolated from rats: release of phosphatidylcholine in response to P-adrenergic stimulation. J Clin Invest 63:378, 1979

72 rrrmRs

5. Brown S, Goodman E, Crandall E: Type I1 alveolar epithelial cells in suspension: separation by density and velocity. Lung 162:271, 1984 6. Hino K, Grogan JB, Hardy JD: Viability of stored lungs. Transplantation 6:25, 1968 7. Majid PA, Sharma B, Meeran MK, et al: Insulin and glucose in the treatment of heart failure. Lancet 2937, 1972 8. Opie LH: Effect of glucose and insulin and potassium infusions on arterio-venous differences of glucose and of free fatty acids and on tissue metabolic changes in dogs with developing myocardial infarction. Am Heart J 38:310, 1979 9. Staul NC: Lung Biology in Health and Disease. Vol7. New York, Marcel Dekker, 1978, pp 236-239 10. Schneeberger EE, Karnovsky MJ: The ultrastructural basis of alveolar capillary membrane permeability to peroxidase used as a tracer. J Cell Biol37:781, 1968 11. Mason RJ, Williams MC, Greenleaf RO, et al: Isolation and properties of type I1 alveolar cells from rat lung. Am Rev Respir Dis 115:1015, 1977 12. Shapiro DL, Harrison RA, Trout CA: Clinical application of respiratory care. Second edition. Chicago, Year Book, 1979, 1416 13. Goodman BE, Fleischer RS, Crandall ED: Evidence for active Na+ transport by cultured monolayers of pulmonary alveolar epithelial cells. Am J Physiol 245:C78, 1983 14. Collins GM, Bravo-Shugarman M, Terasaki PI: Kidney preservation for transplantation. Lancet 2:1219, 1969 15. Sacks SA, Petritsch PH, Kaufman JJ: Canine kidney preservation using a new perfusate. Lancet 1:1024, 1973 16. Opie LH: The glucose hypothesis: relation to acute myocardial ischemia. J Mol Cell Cardiol 1:107, 1970 17. Creese R: Sodium exchange in rat muscle. Nature 201:505, 1964 18. Keman RP: The role of lactate in the active excretion of sodium by frog muscle. J Physiol (Lond) 162129, 1962 19 Wichert P: Studies on the metabolism of ischemic rabbit lungs. J Thorac Cardiovasc Surg 63984, 1972 20. Hess ML, Okabe E, Poland J, et al: Glucose, insulin, potassium protection during the course of hypothermic global ischemia and perfusion: a new proposal mechanism by the scavenging of free radicals. J Cardiovasc Pharmacol 5:35, 1983