- Email: [email protected]
Cell response to hyperbaric oxygen treatment
Int. d. Oral Maxillofac. Surg. 1997; 26:82 86 grflzted #~ Denmark. All rights reserved
Copyright 9 Munksgaard 1997 bTtemationaUoumalof
Oral 8C MaxillofacialSurgery ISSN 0901-5027
Leading article
Cell response to hyperbaric oxygen treatment
Paul C. Tornpach, 1 Daniel Lew, 1 J. Lynn Stoll 2 Departments of 1Oral and Maxillofacial Surgery and 2Anesthesia, University of Iowa Hospitals and Clinics, Iowa City, Iowa, USA
P. C. Tompach, D. Lew, J. L. Stoll: Cell response to hyperbaric oxygen treatment. Int. J. Oral Maxillofac. Surg. 1997; 26. 82-86. 9 Munksgaard, 1997 Abstract. Wound healing involves matrix deposition, angiogenesis, and new tissue
growth. Cellular activity during healing is related to tissue oxygen levels. Since wound healing requires oxygen, the purpose of this study was to investigate the effects of hyperbaric oxygen (HBO) on cells involved in wound healing. Cultured endothelial cells and fibroblasts were exposed to HBO. The effect of varied partial pressure, oxygen saturation, and duration and frequency of exposure to HBO on cell proliferation was determined by 3H-labeled thymidine incorporation. HBO causes an increase in the partial pressure of oxygen in the medium of cultured cells, leading to increased endothelial cell and fibroblast proliferation. Increased endothelial cell proliferation occurred after 15 min of HBO. Fibroblasts required 120 rain of HBO to produce a response. A second exposure to HBO on the same day produced no additional increase in cell proliferation. A 120-min HBO exposure stimulated fibroblast proliferation for 72 h after the exposure. An increase in pressure from 2.4 to 4.0 atmospheres absolute did not enhance the proliferative response. These studies begin to elucidate the effects of HBO on cells 'involved in wound healing and establish a scientific foundation upon which to develop more efficacious and cost-effective HBO therapeutic protocols.
Wound healing is a complex process involving inflammatory proliferative, and remodeling phases 4. The coordinated participation of constituent cells during each phase of this reparative process is critical for efficient wound closure. Successful healing represents the sum of this integrated series of physiologic events mounted in response to tissue damage. The inflammatory phase begins immediately upon injury and persists for several days. Blood vessel disruption leads to platelet activation and the release of substances which promote cell migration into the wound 14. Inflammatory cells phagocytose bacteria and release biologically active substances to assist in tissue debridement and initiate granulation tissue. The proliferative phase of wound healing is characterized by rapid fibroblast growth and increased synthesis of colla-
gen and proteoglycans in response to chemotactic factors released during the inflammatory phase. Ingrowth of capillaries accompanies fibroblast proliferation. New vascular channels fuse across the wound or join adjacent capillary buds to form granulation tissue ~7. Fibroblasts continue to produce collagen which accumulates during the remodeling phase, and wound contraction occurs due to the presence of myofibroblasts. The migration of fibroblasts occurs in concert with capillary buds, reflecting the interdependence of fibroblast function and oxygenation. The capillaries provide oxygen required to synthesize the collagen matrix secreted by fibroblasts. A delicate balance exists between neovascularization providing oxygen and collagen matrix production, which in turn supports new capillary growth 2~
Key words: hyperbaric oxygen; wound healing; cell proliferation; in vitro. Accepted for publication 1 September 1996
Molecular oxygen is one of the critical nutrients of the wound, and it plays a central role in the reparative process 13. Collagen synthesis, matrix deposition, angiogenesis, epithelialization, and bacterial killing all require molecular oxygen during the reparative process. During collagen synthesis, oxygen is a substrate for the hydroxylation of lysine and proline, a step required for the release of collagen from cells 7. Under anoxic conditions, fibroblasts produce an intracellular polypeptide collagen precursor, but fail to release it 9. Maturation and cross-linking of collagen increases linearly with an elevated ambient oxygen concentration 3. Angiogenesis and epithelialization rates are also oxygen-dependent. Oxygen supply controls the rate of epithelialization in normal and ischemic wounds when oxygen is administered at
H B O in wound healing 1-2 a t m o s p h e r e s absolute (ATA) 6. Angiogenesis is driven by a g r a d i e n t o f oxygen whereby high arterial Po~ drives angiogenesis into hypoxic spacesq2. Oxidative bacterial killing m e c h a n isms within leukocytes require molecular oxygen 11. The generation o f oxygenderived free radicals results in the destruction of bacterial m e m b r a n e s . The p r o d u c t i o n rate of toxic radicals a n d the ability to kill via oxidative m e c h a n isms is directly p r o p o r t i o n a l to local oxygen tension 16. This r e q u i r e m e n t for oxygen in w o u n d healing is the rationale for h y p e r b a r i c oxygen (HBO) therapy. It results in a n increase in tissue oxygen tension a n d improves collagen synthesis, angiogenesis, epithelialization, a n d resistance to bacteria in p r o b l e m wounds. Since these processes are closely related to tissue oxygen tension, relieving w o u n d hypoxia with H B O therapy accelerates w o u n d healing by increasing the oxygen tension s . C u r r e n t protocols for H B O therapy are empirical; the m e c h a n i s m of action a n d optimal t h e r a p e u t i c use o f H B O r e m a i n poorly u n d e r s t o o d . O u r knowledge of the m o s t a p p r o p r i a t e design of this therapy is h i n d e r e d by the lack of detailed mechanistic i n f o r m a t i o n on the effects of H B O on c o n s t i t u e n t cells in the woundhealing process. F u r t h e r m o r e , complex cell-to-cell interactions a n d the activity of soluble factors controlling w o u n d healing in vivo m a k e it difficult to determine the direct effects of oxygen on cells themselves a n d to predict optimal oxygenation conditions for the function of these cells. Therefore, o u r l a b o r a t o r y has c o n d u c t e d in vitro experiments designed to investigate the effects o f H B O o n cellular physiology a n d establish a scientific basis for the optimal use of this therapy. The p u r p o s e of this study was to clarify the role of oxygen in w o u n d healing by evaluating the effect of varied partial pressure, oxygen saturation, a n d dura t i o n a n d frequency of exposure to H B O on the proliferation of c u l t u r e d endothelial cells a n d fibroblasts as models for angiogenesis a n d connective-tissue f o r m a t i o n .
7~ble l. Effect of HBO on partial pressure of oxygen in medium of cultured cells
used between passages 6 and 12, and fibroblasts between passages l0 and 20.
Time (min)
Cell growth determination
0 2 15 30 60 90 a
Oxygen (%)
Pressure ~
21 100 100 I00 100 100
0 14 14 14 14 14
fo~ b 167 270 576 1010 >101I >i011
Equivalent to m of sea water, b mm Hg.
USA). Cells were plated and grown to the desired degree of confluence under normoxic conditions. Cultures were then exposed to hyperbaric gas mixtures for various time periods at various frequencies (1-2 exposure/ day). All experiments were performed at 2.4 ATA, unless otherwise noted. Other parameters under study include the long-term effects of a single exposure, the effects of additional exposures on subsequent days, and the effects of varying pressures (1 4 ATA). In these studies, cell proliferation was measured by incorporation of 3H-labeled thymidine into cellular DNA. Cell cultures
Bovine aortic endothelial cells were isolated and characterized as previously described 2~. The endothelial cells were grown in Dulbecco's Minimal Essential Medium (DMEM) (GIBCO, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (Hyclone, Logan, UT, USA), MEM Non-essential Amino Acids (GIBCO), MEM Vitamin Solution (GIBCO), 15 mM 4-(2-hydroxyethyl)-l-piperazineethane sulfonic acid (HEPES) (Sigma, St. Louis, MO, USA), 2 mM Lglutamine (Sigma), and 50 ,uM gentamicin (Schering Corp., Kenilworth, N J, USA). Detroit 551 human skin fibroblasts (ATCC Ccl 1 I0) were obtained from ATCC, Rockville, MD, USA. Cells were maintained similarly in Eagle's MEM with Earle's salts (GIBCO) with MEM Non-essential Amino Acids, 1.0 mM sodium pyruvate, and 10% fetal bovine serum. Both cell types were grown at 37~ in a humidified atmosphere containing 5% CO> Stocks were subcultured weekly by trypsinization. For experiments, endothelial cells were
The studies described in this paper were performed with cultured endothelial cells and fibroblasts in an animal research hyperbaric chamber (Mechidyne Systems, Houston, TX,
Cell growth was determined by incorporation of 3H-labeled thylnidine (New England Nuclear) into trichloroacetic acid-precipitable material, as previously described 21. Cultures of bovine aortic endothelial cells or human skin fibroblasts were grown in I2-well plates to the desired degree of confluency. Experimental cultures were exposed to HBO for one or more treatments of 15-120 rain, according to the experimental protocol; control nontreated cultures were maintained in the incubator. The cultures were maintained at 37~ in the HBO chamber during experiments by a water-jacketed aluminum block which holds the tissue cultures plates; the temperature of the water is regulated by a cardiac bypass machine pump outside the chamber, which circulates water into the chamber and through the water jacket of the aluminum block. After the HBO exposures, the experimental cultures were returned to the incubator. Eighteen hours after the treatment, 1 /zC of 3H-labeled thymidine was added to each well for 6 h. The medium was then removed, the cells were washed, and cold 5% trichloroacetic acid was added. The trichloroacetic acid precipitable material was solubilized in 0.25 N NaOH for 1 h at 37~ and an aliquot was transferred to a scintillation vial and neutralized with 6 N HC1; then 5 ml scintillation cocktail (Research Products International, Mt Prospect, IL, USA) was added, and the radioactivity was measured in a Beckman LS 7000 scintillation spectrophotometer. Quenching was monitored by channels ratio estimates.
Results Effect of HBO on the partial pressure of oxygen (Po2) in the medium of cultured cells
To measure the effect of H B O o n the P o 2 o f the culture medium, a small-bore catheter was t h r e a d e d t h r o u g h a p o r t in the H B O chamber, with the tip placed in the culture medium. Samples were w i t h d r a w n before the t r e a t m e n t began,
Table 2. Effect of increasing time of exposure to HBO on 3H-labeled thymidine incorporation by bovine aortic endothelial cells and human skin fibroblasts Endothelial cells Expt. 1
Material and methods General experimental design
83
Control 1• rain HBO 2• min HBO 1• min HBO 1 • 120 rain HBO
98 500+7800 152000+13000 157000• 136000_+6200 n.d.
Expt. 2 dpm/weli 65 700~_2300 101 000_+4000 105000_+2400 90600_+1900 n.d.
Fibroblasts Expt. 1 3220_+400 3220_+680 n.d. 3860_+560 5360_+840
Values represent mean_+SEM of six individual wells, n.d. =not determined.
Expt. 2 2330_+550 2470_+450 n.d. 3220+700 4100+440
T o m p a c h et al.
84 8OOO
j
7~176176
5000 i
~dosTROL
i
-i~- HBO
increased chamber pressure may have been harmful to the cells since 3Hlabeled thymidine incorporation at 4.0 ATA was only 79% of the control value. In contrast, these data show that the standard HBO treatment at 2.4 ATA effectively stimulates cell proliferation (Table 3). Discussion
~ooo] 3000 •
Iooo
~
.11/
i
Days
Fig. 1. Graph representing incorporation of 3H-labeled thymidine by HBO-treated cultures
and control cultures not exposed to HBO at 24, 47, and 72 h after treatment. Human skin fibroblasts were subjected to single 120-min exposure to HBO (day 0). Significant increase in 3H-labeled thymidine incorporation was noted 24 h after treatment, with further increases persisting at 48 and 72 h after initial HBO treatment.
and at various times after the chamber was pressurized equivalent to a depth of 14 m of sea water with 100% oxygen. Samples were analyzed for oxygen tension immediately after aspiration fi'om the chamber with an IL-1306 pH/blood gas analyzer (Instrumentation Laboratories, Lexington, MA, USA). The results shown in Table 1 demonstrate the rapid increase in Po2 of the culture medium with HBO treatment. Effect of HBO on cell growth
When 3H-labeled thymidine was added to the cultures immediately before HBO treatment at 2.4 ATA, there was no change in thymidine incorporation by endothelial cells or fibroblasts. However, a 50% increase in 3H-labeled thymidine incorporation was consistently noted with both cell types 18 24 h after 15-120-min HBO treatment (Table 2). After 15-min exposure to HBO, a significant increase in 3H-labeled thymidine incorporation by endothelial cells was noted. Maximal levels of thymidine
Table 3. Effect of chamber pressure on 3Hlabeled thymidine incorporation by endothelial cells
3H-labeled thymidine incorporation dpm/well Control 2.4 ATA 4 ATA
74500_+4900 110000_+2200 59 000• 1800
% of control 100_+6.5 148_+3.0 79_+2.5
incorporation were reached by 60 min. Longer HBO treatment of up to 120 rain resulted in little or no additional increases in 3H-labeled thymidine incorporation by endothelial cells. Fibroblasts showed little or no response after HBO treatments of 15 60 min, and required a 120-min exposure to HBO to produce a response (Table 2). When endothelial cells and fibroblasts were exposed to a second HBO treatment on the same day, no additional increase in 3H-labeled thymidine incorporation was seen. In the experiment shown in Fig. 1, human skin fibroblasts were subjected to a single 120-rain exposure to HBO (day 0). 3H-labeled thymidine incorporation was measured in experimental cultures and in control cultures not exposed to HBO at 24, 48, and 72 h after the treatment. Incorporation of 3Hlabeled thymidine by HBO-treated cultures was signifcantly increased when compared with control cultures 24 h after the treatment. A further increase in proliferation was seen in the HBOtreated cells at 48 h, and an even greater increase was noted 72 h after the HBO exposure. In the experiment shown in Table 3, endothelial cells were exposed to HBO for 60 rain at the standard chamber pressure of 2.4 ATA or at a pressure increased to 4.0 ATA. Increasing the chamber pressure from 2.4 to 4.0 ATA resulted in a significantly decreased thymidine incorporation. The effects of the
Wound healing normally proceeds through a complex but orderly series of events4. The initial inflammatory response involves platelet aggregation, coagulation, and attraction of neutrophils and macrophages into the injured area. Tissue repair occurs under the stimulation of multiple cytokines and growth factors released from platelets, macrophages, keratinocytes, fibroblasts, and endothelial cells TM. Wound repair includes the creation of an extracellular matrix of fibronectin and hyaluronic acid, and proceeds under the continued presence of stimulatory cytokines and growth factors to f o r m ' granulation tissue. Macrophages, fibroblasts, and blood vessels simultaneously move into the wound as a unit4. The mild hypoxia and/or increased lactate levels found in normally healing wounds may facilitate wound healing by inducing the synthesis of collagen precursors and activating macrophages to stimulate angiogenesis6. A common problem in nonhealing wounds is a nonstimulatory level of hypoxia due to inadequate perfusion or excessive consumption from active metabolic processes or infection. The Po 2 of normal healing tissue is 30 50 mmHg, while the Po2 of nonhealing wound tissue is usually less than 20 mmHg 2. The effects of hypoxia have been well documented in wound healing and include decreased fibroblast migration and decreased collagen synthesis with impaired hydroxylation of lysine and proline6. Under hypoxic conditions, fibroblasts produce an intracellular peptide collagen precursor, but fail to release it. Collagen synthesized under hypoxic conditions is produced at a slow rate and has poor mechanical strength compared to that produced in a normoxic environment19. Maturation and cross-linking of collagen increases linearly with an elevated ambient oxygen concentration3. As shown by studies in rabbit ear chambers using controlled oxygen tensions, angiogenesis is driven by a gradient of
H B O in wound healing
oxygen 12. High arterial 0 2 appears to drive angiogenesis into hypoxic spaces in coordination with fibroblast collagen production and the release of an angiogenic substance by macrophages. Hypoxic wounds are also susceptible to infections. Neutrophils and macrophages can phagocytose bacteria, but their bactericidal ability is impaired by hypoxia and the inability of neutrophils to produce oxygen radicals 1~ Bacteria may flourish and further increase the demand for oxygen in an already hypoxic wound. The adjunctive use of HBO therapy can positively alter the compromised wound healing process 5,13. Exposure of patients to HBO increases the amount of oxygen that is present within tissue. Delivery of oxygen from the ambient air to the cardiovascular system and eventually to the mitochondria is described as the oxygen pathway s. The Po2 is decreased at each step of this pathway. The oxygen cascade describes this decrease in Po2 from 160 mmHg in dry ambient air to 1-3 mmHg in the mitochondria ~5. The capability of HBO therapy to increase the delivery of oxygen may be appreciated by comparing the Po2 at different sites in the oxygen pathway. The alveolar partial pressure of oxygen (PAo2) is approximately 102 mmHg while breathing air at atmospheric pressure 1. The PAo2 can be increased to 673 mmHg whiie breathing 100% oxygen, and to 1813 mmHg while breathing 100% oxygen at 2.5 ATA, an 18-fold increase over breathing air at atmospheric pressure. This additional oxygen is distributed by the circulating blood primarily dissolved in the plasma. This correlates with an increase in the wound tissue Po2 from 5 20 mmHg while breathing air at 1.0 ATA to 200-400 mmHg while breathing 100% oxygen at 1.0 ATA to a Po2 of 800-1100 mmHg while exposed to 100% oxygen at 2.5 ATA Is. By using cell cultures exposed to hyperbaric gas mixtures, we have identified conditions which produce increased stimulation of cell types present in healing wounds. Cell proliferation was measured in cultured endothelial cells as a measure of angiogenesis and in fibroblasts as a measure of connective-tissue formation. Our data indicate that HBO at 2.4 ATA causes no change in the incorporation of 3H-labeled thymidine by endothelial cells or fibroblasts when thymidine is added immediately before the treatment. However, a
50% increase in 3H-labeled thymidine incorporation was consistently observed with both cell types 18-24 h after a 15-120-min treatment. This finding is significant since it suggests an underlying mechanism requiring induction of gene expression. Experiments are currently underway to examine how HBO affects other cellular functions which are implicated in wound healing. These functions include free radical formation, prostaglandin production, leukocyte adhesion to endothelium, nitric oxide production, and the induction of oncogenes. A significant increase in 3H-labeled thymidine incorporation by endothelial cells was seen after an HBO treatment as short as 15 min. Longer treatments up to 120 min showed no additional increases in 3H-labeled thymidine incorporation by endothelial cells. These data are interesting since they suggest that, at least for cultured cells and perhaps also for superficial wounds, optimal oxygenation and proliferative responses may be obtained under less rigorous HBO regimens than current standard practice. A second HBO exposure on the same day produced no additional increase in 3H-labeled thymidine incorporation by endothelial cells after 24 h when compared to control cultures. This result further suggests an underlying mechanism involving induction of gene expression and the synthesis of RNA and/ or protein products. The intracellular generation of oxygen radicals may also initiate a series of cellular processes which regulate cell proliferation and growth. If this finding holds true for fibroblasts and for functional studies of cell synthetic activity, it is significant since it suggests that the second of two daily treatments in current therapeutic regimens may prove to be superfluous. A single 120-min treatment with HBO stimulated proliferation of human skin fibroblasts for at least 72 h after the initial exposure. This result suggests that a single HBO exposure may initiate a series of cellular events whose effects may persist for several days. The persistent effects of a single HBO exposure and the elicited cellular responses are areas for further study. The results of experiments reported here indicate that both endothelial cell and fibroblast growth are augmented by HBO treatment. Thus, within the design of our experimental model, the growth of endothelial cells and fibro-
85
blasts in our in vitro model may parallel the behavior of constituent cells in superficial healing wounds. Initial investigations concerning the in vitro behavior of cells that participate in wound healing are important since they afford the ability to study the direct effects of oxygen on cells themselves. Low oxygen tensions may be required to activate certain cells in the wound-healing process, while the function of other cells may be impaired by low oxygen tensions. In vivo, it is difficult to determine the exact oxygen level to which cells involved in the wound-healing process are exposed. This makes it difficult to predict optimal oxygen tensions for the function of these cells. Determining the growth and biosynthetic activities of key celt types relative to in vitro oxygen tensions sets the stage for further studies concerning the complex cell-tocell interactions and subtle physiologic effects which occur in vivo. Although our data on the effect of varying chamber pressure are incomplete, there is evidence that enhanced oxygenation occurs at lower pressures than the standard 2.4 ATA, and that higher pressures may be damaging to cells. If these initial findings are confirmed by further studies, it may be possible to obtain optimal healing of superficial vascularized wounds by fewer and shorter HBO treatments, perhaps at lower chamber pressures. Such a change in the therapeutic regimen could reduce the incidence of undesirable effects while also reducing costs.
References 1. BASSETTBE, BENNETTPB. Introduction to the physical and physiologicalbasis of hyperbaric therapy. In: DAvis JC, HUNT TK, eds.: Hyperbaric oxygen therapy. Bethesda, MD: Undersea Medical Society, 1977:11 24. 2. CHANG N, GOODSONWH, GOTTRUPE HURTTK. Direct measurement of wound and tissue oxygen tension in postoperative patients. Ann Surg 1983: 197:470 8. 3. CHVAPIL M, HURYCH J, EHRLICHOVA E.
The influence of varying oxygen tensions upon proline hydroxylation and the metabolism of collagenous and noncollagenous proteins in skin slides. Z Physiol Chem 1968: 349:211. 4. CLARKRAE Basics of cutaneous wound repair. J Dermatol Surg Oncol i993: 19: 693 706. 5. DAVISJC. Hyperbaric oxygen therapy. J Intensive Care Med 1989: 4:55 7. 6. HUNTTK, PAI MR The effect of varying ambient oxygen tensions on wound me-
86
Tompach et al.
tabolism and collagen synthesis. Surg Gynecol Obstet 1972: 135:561 7. 7. HUTTON J J, TAPPEL AL, UDENFRIEND S. Cofactor and substrate requirements of collagen proline hydroxylase. Arch Biochem 1967: 118: 23140. 8. JAIN KK, FISCICER B. Oxygen in physiology and medicine. Springfield, IL: Thomas, 1989. 9. JuvA K, PROCKOPD J, COOPERGW, LASH JW. Hydroxylation of proline and the intracellular accumulation of a polypeptide precursor of collagen. Science 1966: 152: 924. 10. KLEBANOFF S. Oxygen metabolism and the toxic properties of phagocytes. Ann Intern Med 1980: 93: 480-9. 1i. KN1GHTONDR, HALLIDAYB, HUNT TK. Oxygen as an antibiotic: a comparison of the effects of inspired oxygen concentration and antibiotic administration on in vivo bacterial clearance. Arch Surg 1986: 121: 191-5. 12. KNIGHTON DR, SILVER IA, HUNT TK. Regulation of wound healing angiogenesis effect of oxygen gradients and in-
spired oxygen concentration. Surgery 1981: 90: 262-70. 13. LAVAN FB, HURT TK. Oxygen and wound healing. Clin Plastic Surg 1990: 17:463 72. 14. MCGRATH MH. Peptide growth factors and wound healing. Clin Plastic Surg 1990: 17:421 32. 15. NUNN JE Oxygen. In: Nunn's applied respiratory physiology. 4th ed. Oxford: Butterworth-Heinemann, 1993: 247-305. 16. RABKINJ, HUNT TK. Infection and oxygen. In: DAVIS J, HUNT TK, eds.: Problem wounds: the role of oxygen. New York: Elsevier, 1988:1 16. 17. SCHOEFL GI. Studies in inflammation. III. Growing capillaries: their structure and permeability. Virchows Arch A Pathol Anat Histopathol 1963: 337:97 102. 18. SHEFFIELD PJ, HEIMBACH RD. Respiratory physiology. In: DEHART RL, ed.: Fundamentals of aerospace medicine. Philadelphia: Lea and Febiger, 1985:72 109. 19. SHEFFIELD PJ. Tissue oxygen measure-
lnents. In: DAVIS JC, HUNT TK, eds.: Problem wounds: the role of oxygen. New York: Elsevier, 1988:17 51. 20. SILVERIA. Cellular microenvironment in healing and non-healing wounds. In: HUNT TK, HEPENSTALL RB, PINES E, ROVEE D, eds.: Soft and hard tissue repair: biological and clinical aspects. New York: Praeger, 1984: 50-66. 21. aTOLL LL, SPECTOR AA. Interaction of platelet-activating factor with endothelial and vascular smooth muscle cells in coculture. J Cell Physiol 1989: 139: 253-61.
Address: Paul C. Tompach, DDS, PhD Department of Hospital Dentistry Division of Oral and Maxillofacial Surgery E202 GH University of Iowa Hospitals and Clinics 200 Hawkins Drive Iowa City, 1A 52242 USA
Copyright 9 Munksgaard 1997 bTtemationaUoumalof
Oral 8C MaxillofacialSurgery ISSN 0901-5027
Leading article
Cell response to hyperbaric oxygen treatment
Paul C. Tornpach, 1 Daniel Lew, 1 J. Lynn Stoll 2 Departments of 1Oral and Maxillofacial Surgery and 2Anesthesia, University of Iowa Hospitals and Clinics, Iowa City, Iowa, USA
P. C. Tompach, D. Lew, J. L. Stoll: Cell response to hyperbaric oxygen treatment. Int. J. Oral Maxillofac. Surg. 1997; 26. 82-86. 9 Munksgaard, 1997 Abstract. Wound healing involves matrix deposition, angiogenesis, and new tissue
growth. Cellular activity during healing is related to tissue oxygen levels. Since wound healing requires oxygen, the purpose of this study was to investigate the effects of hyperbaric oxygen (HBO) on cells involved in wound healing. Cultured endothelial cells and fibroblasts were exposed to HBO. The effect of varied partial pressure, oxygen saturation, and duration and frequency of exposure to HBO on cell proliferation was determined by 3H-labeled thymidine incorporation. HBO causes an increase in the partial pressure of oxygen in the medium of cultured cells, leading to increased endothelial cell and fibroblast proliferation. Increased endothelial cell proliferation occurred after 15 min of HBO. Fibroblasts required 120 rain of HBO to produce a response. A second exposure to HBO on the same day produced no additional increase in cell proliferation. A 120-min HBO exposure stimulated fibroblast proliferation for 72 h after the exposure. An increase in pressure from 2.4 to 4.0 atmospheres absolute did not enhance the proliferative response. These studies begin to elucidate the effects of HBO on cells 'involved in wound healing and establish a scientific foundation upon which to develop more efficacious and cost-effective HBO therapeutic protocols.
Wound healing is a complex process involving inflammatory proliferative, and remodeling phases 4. The coordinated participation of constituent cells during each phase of this reparative process is critical for efficient wound closure. Successful healing represents the sum of this integrated series of physiologic events mounted in response to tissue damage. The inflammatory phase begins immediately upon injury and persists for several days. Blood vessel disruption leads to platelet activation and the release of substances which promote cell migration into the wound 14. Inflammatory cells phagocytose bacteria and release biologically active substances to assist in tissue debridement and initiate granulation tissue. The proliferative phase of wound healing is characterized by rapid fibroblast growth and increased synthesis of colla-
gen and proteoglycans in response to chemotactic factors released during the inflammatory phase. Ingrowth of capillaries accompanies fibroblast proliferation. New vascular channels fuse across the wound or join adjacent capillary buds to form granulation tissue ~7. Fibroblasts continue to produce collagen which accumulates during the remodeling phase, and wound contraction occurs due to the presence of myofibroblasts. The migration of fibroblasts occurs in concert with capillary buds, reflecting the interdependence of fibroblast function and oxygenation. The capillaries provide oxygen required to synthesize the collagen matrix secreted by fibroblasts. A delicate balance exists between neovascularization providing oxygen and collagen matrix production, which in turn supports new capillary growth 2~
Key words: hyperbaric oxygen; wound healing; cell proliferation; in vitro. Accepted for publication 1 September 1996
Molecular oxygen is one of the critical nutrients of the wound, and it plays a central role in the reparative process 13. Collagen synthesis, matrix deposition, angiogenesis, epithelialization, and bacterial killing all require molecular oxygen during the reparative process. During collagen synthesis, oxygen is a substrate for the hydroxylation of lysine and proline, a step required for the release of collagen from cells 7. Under anoxic conditions, fibroblasts produce an intracellular polypeptide collagen precursor, but fail to release it 9. Maturation and cross-linking of collagen increases linearly with an elevated ambient oxygen concentration 3. Angiogenesis and epithelialization rates are also oxygen-dependent. Oxygen supply controls the rate of epithelialization in normal and ischemic wounds when oxygen is administered at
H B O in wound healing 1-2 a t m o s p h e r e s absolute (ATA) 6. Angiogenesis is driven by a g r a d i e n t o f oxygen whereby high arterial Po~ drives angiogenesis into hypoxic spacesq2. Oxidative bacterial killing m e c h a n isms within leukocytes require molecular oxygen 11. The generation o f oxygenderived free radicals results in the destruction of bacterial m e m b r a n e s . The p r o d u c t i o n rate of toxic radicals a n d the ability to kill via oxidative m e c h a n isms is directly p r o p o r t i o n a l to local oxygen tension 16. This r e q u i r e m e n t for oxygen in w o u n d healing is the rationale for h y p e r b a r i c oxygen (HBO) therapy. It results in a n increase in tissue oxygen tension a n d improves collagen synthesis, angiogenesis, epithelialization, a n d resistance to bacteria in p r o b l e m wounds. Since these processes are closely related to tissue oxygen tension, relieving w o u n d hypoxia with H B O therapy accelerates w o u n d healing by increasing the oxygen tension s . C u r r e n t protocols for H B O therapy are empirical; the m e c h a n i s m of action a n d optimal t h e r a p e u t i c use o f H B O r e m a i n poorly u n d e r s t o o d . O u r knowledge of the m o s t a p p r o p r i a t e design of this therapy is h i n d e r e d by the lack of detailed mechanistic i n f o r m a t i o n on the effects of H B O on c o n s t i t u e n t cells in the woundhealing process. F u r t h e r m o r e , complex cell-to-cell interactions a n d the activity of soluble factors controlling w o u n d healing in vivo m a k e it difficult to determine the direct effects of oxygen on cells themselves a n d to predict optimal oxygenation conditions for the function of these cells. Therefore, o u r l a b o r a t o r y has c o n d u c t e d in vitro experiments designed to investigate the effects o f H B O o n cellular physiology a n d establish a scientific basis for the optimal use of this therapy. The p u r p o s e of this study was to clarify the role of oxygen in w o u n d healing by evaluating the effect of varied partial pressure, oxygen saturation, a n d dura t i o n a n d frequency of exposure to H B O on the proliferation of c u l t u r e d endothelial cells a n d fibroblasts as models for angiogenesis a n d connective-tissue f o r m a t i o n .
7~ble l. Effect of HBO on partial pressure of oxygen in medium of cultured cells
used between passages 6 and 12, and fibroblasts between passages l0 and 20.
Time (min)
Cell growth determination
0 2 15 30 60 90 a
Oxygen (%)
Pressure ~
21 100 100 I00 100 100
0 14 14 14 14 14
fo~ b 167 270 576 1010 >101I >i011
Equivalent to m of sea water, b mm Hg.
USA). Cells were plated and grown to the desired degree of confluence under normoxic conditions. Cultures were then exposed to hyperbaric gas mixtures for various time periods at various frequencies (1-2 exposure/ day). All experiments were performed at 2.4 ATA, unless otherwise noted. Other parameters under study include the long-term effects of a single exposure, the effects of additional exposures on subsequent days, and the effects of varying pressures (1 4 ATA). In these studies, cell proliferation was measured by incorporation of 3H-labeled thymidine into cellular DNA. Cell cultures
Bovine aortic endothelial cells were isolated and characterized as previously described 2~. The endothelial cells were grown in Dulbecco's Minimal Essential Medium (DMEM) (GIBCO, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (Hyclone, Logan, UT, USA), MEM Non-essential Amino Acids (GIBCO), MEM Vitamin Solution (GIBCO), 15 mM 4-(2-hydroxyethyl)-l-piperazineethane sulfonic acid (HEPES) (Sigma, St. Louis, MO, USA), 2 mM Lglutamine (Sigma), and 50 ,uM gentamicin (Schering Corp., Kenilworth, N J, USA). Detroit 551 human skin fibroblasts (ATCC Ccl 1 I0) were obtained from ATCC, Rockville, MD, USA. Cells were maintained similarly in Eagle's MEM with Earle's salts (GIBCO) with MEM Non-essential Amino Acids, 1.0 mM sodium pyruvate, and 10% fetal bovine serum. Both cell types were grown at 37~ in a humidified atmosphere containing 5% CO> Stocks were subcultured weekly by trypsinization. For experiments, endothelial cells were
The studies described in this paper were performed with cultured endothelial cells and fibroblasts in an animal research hyperbaric chamber (Mechidyne Systems, Houston, TX,
Cell growth was determined by incorporation of 3H-labeled thylnidine (New England Nuclear) into trichloroacetic acid-precipitable material, as previously described 21. Cultures of bovine aortic endothelial cells or human skin fibroblasts were grown in I2-well plates to the desired degree of confluency. Experimental cultures were exposed to HBO for one or more treatments of 15-120 rain, according to the experimental protocol; control nontreated cultures were maintained in the incubator. The cultures were maintained at 37~ in the HBO chamber during experiments by a water-jacketed aluminum block which holds the tissue cultures plates; the temperature of the water is regulated by a cardiac bypass machine pump outside the chamber, which circulates water into the chamber and through the water jacket of the aluminum block. After the HBO exposures, the experimental cultures were returned to the incubator. Eighteen hours after the treatment, 1 /zC of 3H-labeled thymidine was added to each well for 6 h. The medium was then removed, the cells were washed, and cold 5% trichloroacetic acid was added. The trichloroacetic acid precipitable material was solubilized in 0.25 N NaOH for 1 h at 37~ and an aliquot was transferred to a scintillation vial and neutralized with 6 N HC1; then 5 ml scintillation cocktail (Research Products International, Mt Prospect, IL, USA) was added, and the radioactivity was measured in a Beckman LS 7000 scintillation spectrophotometer. Quenching was monitored by channels ratio estimates.
Results Effect of HBO on the partial pressure of oxygen (Po2) in the medium of cultured cells
To measure the effect of H B O o n the P o 2 o f the culture medium, a small-bore catheter was t h r e a d e d t h r o u g h a p o r t in the H B O chamber, with the tip placed in the culture medium. Samples were w i t h d r a w n before the t r e a t m e n t began,
Table 2. Effect of increasing time of exposure to HBO on 3H-labeled thymidine incorporation by bovine aortic endothelial cells and human skin fibroblasts Endothelial cells Expt. 1
Material and methods General experimental design
83
Control 1• rain HBO 2• min HBO 1• min HBO 1 • 120 rain HBO
98 500+7800 152000+13000 157000• 136000_+6200 n.d.
Expt. 2 dpm/weli 65 700~_2300 101 000_+4000 105000_+2400 90600_+1900 n.d.
Fibroblasts Expt. 1 3220_+400 3220_+680 n.d. 3860_+560 5360_+840
Values represent mean_+SEM of six individual wells, n.d. =not determined.
Expt. 2 2330_+550 2470_+450 n.d. 3220+700 4100+440
T o m p a c h et al.
84 8OOO
j
7~176176
5000 i
~dosTROL
i
-i~- HBO
increased chamber pressure may have been harmful to the cells since 3Hlabeled thymidine incorporation at 4.0 ATA was only 79% of the control value. In contrast, these data show that the standard HBO treatment at 2.4 ATA effectively stimulates cell proliferation (Table 3). Discussion
~ooo] 3000 •
Iooo
~
.11/
i
Days
Fig. 1. Graph representing incorporation of 3H-labeled thymidine by HBO-treated cultures
and control cultures not exposed to HBO at 24, 47, and 72 h after treatment. Human skin fibroblasts were subjected to single 120-min exposure to HBO (day 0). Significant increase in 3H-labeled thymidine incorporation was noted 24 h after treatment, with further increases persisting at 48 and 72 h after initial HBO treatment.
and at various times after the chamber was pressurized equivalent to a depth of 14 m of sea water with 100% oxygen. Samples were analyzed for oxygen tension immediately after aspiration fi'om the chamber with an IL-1306 pH/blood gas analyzer (Instrumentation Laboratories, Lexington, MA, USA). The results shown in Table 1 demonstrate the rapid increase in Po2 of the culture medium with HBO treatment. Effect of HBO on cell growth
When 3H-labeled thymidine was added to the cultures immediately before HBO treatment at 2.4 ATA, there was no change in thymidine incorporation by endothelial cells or fibroblasts. However, a 50% increase in 3H-labeled thymidine incorporation was consistently noted with both cell types 18 24 h after 15-120-min HBO treatment (Table 2). After 15-min exposure to HBO, a significant increase in 3H-labeled thymidine incorporation by endothelial cells was noted. Maximal levels of thymidine
Table 3. Effect of chamber pressure on 3Hlabeled thymidine incorporation by endothelial cells
3H-labeled thymidine incorporation dpm/well Control 2.4 ATA 4 ATA
74500_+4900 110000_+2200 59 000• 1800
% of control 100_+6.5 148_+3.0 79_+2.5
incorporation were reached by 60 min. Longer HBO treatment of up to 120 rain resulted in little or no additional increases in 3H-labeled thymidine incorporation by endothelial cells. Fibroblasts showed little or no response after HBO treatments of 15 60 min, and required a 120-min exposure to HBO to produce a response (Table 2). When endothelial cells and fibroblasts were exposed to a second HBO treatment on the same day, no additional increase in 3H-labeled thymidine incorporation was seen. In the experiment shown in Fig. 1, human skin fibroblasts were subjected to a single 120-rain exposure to HBO (day 0). 3H-labeled thymidine incorporation was measured in experimental cultures and in control cultures not exposed to HBO at 24, 48, and 72 h after the treatment. Incorporation of 3Hlabeled thymidine by HBO-treated cultures was signifcantly increased when compared with control cultures 24 h after the treatment. A further increase in proliferation was seen in the HBOtreated cells at 48 h, and an even greater increase was noted 72 h after the HBO exposure. In the experiment shown in Table 3, endothelial cells were exposed to HBO for 60 rain at the standard chamber pressure of 2.4 ATA or at a pressure increased to 4.0 ATA. Increasing the chamber pressure from 2.4 to 4.0 ATA resulted in a significantly decreased thymidine incorporation. The effects of the
Wound healing normally proceeds through a complex but orderly series of events4. The initial inflammatory response involves platelet aggregation, coagulation, and attraction of neutrophils and macrophages into the injured area. Tissue repair occurs under the stimulation of multiple cytokines and growth factors released from platelets, macrophages, keratinocytes, fibroblasts, and endothelial cells TM. Wound repair includes the creation of an extracellular matrix of fibronectin and hyaluronic acid, and proceeds under the continued presence of stimulatory cytokines and growth factors to f o r m ' granulation tissue. Macrophages, fibroblasts, and blood vessels simultaneously move into the wound as a unit4. The mild hypoxia and/or increased lactate levels found in normally healing wounds may facilitate wound healing by inducing the synthesis of collagen precursors and activating macrophages to stimulate angiogenesis6. A common problem in nonhealing wounds is a nonstimulatory level of hypoxia due to inadequate perfusion or excessive consumption from active metabolic processes or infection. The Po 2 of normal healing tissue is 30 50 mmHg, while the Po2 of nonhealing wound tissue is usually less than 20 mmHg 2. The effects of hypoxia have been well documented in wound healing and include decreased fibroblast migration and decreased collagen synthesis with impaired hydroxylation of lysine and proline6. Under hypoxic conditions, fibroblasts produce an intracellular peptide collagen precursor, but fail to release it. Collagen synthesized under hypoxic conditions is produced at a slow rate and has poor mechanical strength compared to that produced in a normoxic environment19. Maturation and cross-linking of collagen increases linearly with an elevated ambient oxygen concentration3. As shown by studies in rabbit ear chambers using controlled oxygen tensions, angiogenesis is driven by a gradient of
H B O in wound healing
oxygen 12. High arterial 0 2 appears to drive angiogenesis into hypoxic spaces in coordination with fibroblast collagen production and the release of an angiogenic substance by macrophages. Hypoxic wounds are also susceptible to infections. Neutrophils and macrophages can phagocytose bacteria, but their bactericidal ability is impaired by hypoxia and the inability of neutrophils to produce oxygen radicals 1~ Bacteria may flourish and further increase the demand for oxygen in an already hypoxic wound. The adjunctive use of HBO therapy can positively alter the compromised wound healing process 5,13. Exposure of patients to HBO increases the amount of oxygen that is present within tissue. Delivery of oxygen from the ambient air to the cardiovascular system and eventually to the mitochondria is described as the oxygen pathway s. The Po2 is decreased at each step of this pathway. The oxygen cascade describes this decrease in Po2 from 160 mmHg in dry ambient air to 1-3 mmHg in the mitochondria ~5. The capability of HBO therapy to increase the delivery of oxygen may be appreciated by comparing the Po2 at different sites in the oxygen pathway. The alveolar partial pressure of oxygen (PAo2) is approximately 102 mmHg while breathing air at atmospheric pressure 1. The PAo2 can be increased to 673 mmHg whiie breathing 100% oxygen, and to 1813 mmHg while breathing 100% oxygen at 2.5 ATA, an 18-fold increase over breathing air at atmospheric pressure. This additional oxygen is distributed by the circulating blood primarily dissolved in the plasma. This correlates with an increase in the wound tissue Po2 from 5 20 mmHg while breathing air at 1.0 ATA to 200-400 mmHg while breathing 100% oxygen at 1.0 ATA to a Po2 of 800-1100 mmHg while exposed to 100% oxygen at 2.5 ATA Is. By using cell cultures exposed to hyperbaric gas mixtures, we have identified conditions which produce increased stimulation of cell types present in healing wounds. Cell proliferation was measured in cultured endothelial cells as a measure of angiogenesis and in fibroblasts as a measure of connective-tissue formation. Our data indicate that HBO at 2.4 ATA causes no change in the incorporation of 3H-labeled thymidine by endothelial cells or fibroblasts when thymidine is added immediately before the treatment. However, a
50% increase in 3H-labeled thymidine incorporation was consistently observed with both cell types 18-24 h after a 15-120-min treatment. This finding is significant since it suggests an underlying mechanism requiring induction of gene expression. Experiments are currently underway to examine how HBO affects other cellular functions which are implicated in wound healing. These functions include free radical formation, prostaglandin production, leukocyte adhesion to endothelium, nitric oxide production, and the induction of oncogenes. A significant increase in 3H-labeled thymidine incorporation by endothelial cells was seen after an HBO treatment as short as 15 min. Longer treatments up to 120 min showed no additional increases in 3H-labeled thymidine incorporation by endothelial cells. These data are interesting since they suggest that, at least for cultured cells and perhaps also for superficial wounds, optimal oxygenation and proliferative responses may be obtained under less rigorous HBO regimens than current standard practice. A second HBO exposure on the same day produced no additional increase in 3H-labeled thymidine incorporation by endothelial cells after 24 h when compared to control cultures. This result further suggests an underlying mechanism involving induction of gene expression and the synthesis of RNA and/ or protein products. The intracellular generation of oxygen radicals may also initiate a series of cellular processes which regulate cell proliferation and growth. If this finding holds true for fibroblasts and for functional studies of cell synthetic activity, it is significant since it suggests that the second of two daily treatments in current therapeutic regimens may prove to be superfluous. A single 120-min treatment with HBO stimulated proliferation of human skin fibroblasts for at least 72 h after the initial exposure. This result suggests that a single HBO exposure may initiate a series of cellular events whose effects may persist for several days. The persistent effects of a single HBO exposure and the elicited cellular responses are areas for further study. The results of experiments reported here indicate that both endothelial cell and fibroblast growth are augmented by HBO treatment. Thus, within the design of our experimental model, the growth of endothelial cells and fibro-
85
blasts in our in vitro model may parallel the behavior of constituent cells in superficial healing wounds. Initial investigations concerning the in vitro behavior of cells that participate in wound healing are important since they afford the ability to study the direct effects of oxygen on cells themselves. Low oxygen tensions may be required to activate certain cells in the wound-healing process, while the function of other cells may be impaired by low oxygen tensions. In vivo, it is difficult to determine the exact oxygen level to which cells involved in the wound-healing process are exposed. This makes it difficult to predict optimal oxygen tensions for the function of these cells. Determining the growth and biosynthetic activities of key celt types relative to in vitro oxygen tensions sets the stage for further studies concerning the complex cell-tocell interactions and subtle physiologic effects which occur in vivo. Although our data on the effect of varying chamber pressure are incomplete, there is evidence that enhanced oxygenation occurs at lower pressures than the standard 2.4 ATA, and that higher pressures may be damaging to cells. If these initial findings are confirmed by further studies, it may be possible to obtain optimal healing of superficial vascularized wounds by fewer and shorter HBO treatments, perhaps at lower chamber pressures. Such a change in the therapeutic regimen could reduce the incidence of undesirable effects while also reducing costs.
References 1. BASSETTBE, BENNETTPB. Introduction to the physical and physiologicalbasis of hyperbaric therapy. In: DAvis JC, HUNT TK, eds.: Hyperbaric oxygen therapy. Bethesda, MD: Undersea Medical Society, 1977:11 24. 2. CHANG N, GOODSONWH, GOTTRUPE HURTTK. Direct measurement of wound and tissue oxygen tension in postoperative patients. Ann Surg 1983: 197:470 8. 3. CHVAPIL M, HURYCH J, EHRLICHOVA E.
The influence of varying oxygen tensions upon proline hydroxylation and the metabolism of collagenous and noncollagenous proteins in skin slides. Z Physiol Chem 1968: 349:211. 4. CLARKRAE Basics of cutaneous wound repair. J Dermatol Surg Oncol i993: 19: 693 706. 5. DAVISJC. Hyperbaric oxygen therapy. J Intensive Care Med 1989: 4:55 7. 6. HUNTTK, PAI MR The effect of varying ambient oxygen tensions on wound me-
86
Tompach et al.
tabolism and collagen synthesis. Surg Gynecol Obstet 1972: 135:561 7. 7. HUTTON J J, TAPPEL AL, UDENFRIEND S. Cofactor and substrate requirements of collagen proline hydroxylase. Arch Biochem 1967: 118: 23140. 8. JAIN KK, FISCICER B. Oxygen in physiology and medicine. Springfield, IL: Thomas, 1989. 9. JuvA K, PROCKOPD J, COOPERGW, LASH JW. Hydroxylation of proline and the intracellular accumulation of a polypeptide precursor of collagen. Science 1966: 152: 924. 10. KLEBANOFF S. Oxygen metabolism and the toxic properties of phagocytes. Ann Intern Med 1980: 93: 480-9. 1i. KN1GHTONDR, HALLIDAYB, HUNT TK. Oxygen as an antibiotic: a comparison of the effects of inspired oxygen concentration and antibiotic administration on in vivo bacterial clearance. Arch Surg 1986: 121: 191-5. 12. KNIGHTON DR, SILVER IA, HUNT TK. Regulation of wound healing angiogenesis effect of oxygen gradients and in-
spired oxygen concentration. Surgery 1981: 90: 262-70. 13. LAVAN FB, HURT TK. Oxygen and wound healing. Clin Plastic Surg 1990: 17:463 72. 14. MCGRATH MH. Peptide growth factors and wound healing. Clin Plastic Surg 1990: 17:421 32. 15. NUNN JE Oxygen. In: Nunn's applied respiratory physiology. 4th ed. Oxford: Butterworth-Heinemann, 1993: 247-305. 16. RABKINJ, HUNT TK. Infection and oxygen. In: DAVIS J, HUNT TK, eds.: Problem wounds: the role of oxygen. New York: Elsevier, 1988:1 16. 17. SCHOEFL GI. Studies in inflammation. III. Growing capillaries: their structure and permeability. Virchows Arch A Pathol Anat Histopathol 1963: 337:97 102. 18. SHEFFIELD PJ, HEIMBACH RD. Respiratory physiology. In: DEHART RL, ed.: Fundamentals of aerospace medicine. Philadelphia: Lea and Febiger, 1985:72 109. 19. SHEFFIELD PJ. Tissue oxygen measure-
lnents. In: DAVIS JC, HUNT TK, eds.: Problem wounds: the role of oxygen. New York: Elsevier, 1988:17 51. 20. SILVERIA. Cellular microenvironment in healing and non-healing wounds. In: HUNT TK, HEPENSTALL RB, PINES E, ROVEE D, eds.: Soft and hard tissue repair: biological and clinical aspects. New York: Praeger, 1984: 50-66. 21. aTOLL LL, SPECTOR AA. Interaction of platelet-activating factor with endothelial and vascular smooth muscle cells in coculture. J Cell Physiol 1989: 139: 253-61.
Address: Paul C. Tompach, DDS, PhD Department of Hospital Dentistry Division of Oral and Maxillofacial Surgery E202 GH University of Iowa Hospitals and Clinics 200 Hawkins Drive Iowa City, 1A 52242 USA