27, 452A64 (1990)
Factors Influencing Survival of Mammalian Cells Exposed to Hypothermia IV. Effects of Iron Chelation M. A. J. ZIEGER, D. J. GLOFCHESKI, J. R. LEPOCK, AND J. KRUUV Guelph-Waterloo Program for Graduate Work in Physics and Department of Biology, University of Waterloo, Waterloo, Ontario, Canada N2L 3Gl Survival of V-79 Chinese hamster cells was assessed by colony growth assay after hypothermic exposure in the presence of iron chelators. At YC, maximum protection from hypothermic damage was achieved with a 50 p&f concentration of the intracellular ferric iron chelator Desferal. A 3-hr prehypothermic incubation with 50 pM Desferal followed by replacement with chelator-free medium at 5°C also provided some protection. This was not observed when the extracellular chelator DETAPAC (50 m was used prior to cold storage. Treating W-stored cells with Desferal just prior to rewarming was ineffective, but treating cells with Dcsferal during hypothermia exposure after a significant period of unprotected cold exposure ultimately increased the surviving fraction. Submaximal protection during hypothermia was achieved to various degrees with extraceMar chelators at YC, including 50 pM DETAPAC and 110pM EDTA. EGTA (110 pM) had little effect. The sensitization of cells at 5°C with 200 @! FeCl, could be reduced or eliminated with Desferal in accordance with a f : 1 binding ratio. At 1O”C, 50 JLM Desferal, 50 FM DETAPAC, and 110 ~.LMEDTA were as or less effective in protecting cells than at 5°C. An Arrhenius plot of cell inactivation rates shows a break at 7-W, corresponding to maximum survival for control cells and cells in 50 t.& Desferal; however, the amount of protection offered by the chelator increases with decreasing temperature below about WC, and sensitization increases above that point. it has not previously been shown that iron chelators protect against cellular hypothermia damage which is uncomplicated by previous or simultaneous ischemia. This may be relevant to the low-temperature storage of transplant organs, in which iron of intracellular origin and in the perfusate may be active and damaging. Q 1990 Academic Press. Inc.
The effects of cooling on complex integrated cell metabolism may include the uncoupling of reaction pathways with unpredictable consequences (32). The aftermath of some of these consequences is a decrease in mammalian cell survival as a function of time of hypothermia (0 to +25”C) exposure, even when the cells are not subjected to hypoxia before or during the hypothermic exposure (24). The mechanisms of cell killing by hypothermia are not fully understood but could involve damage by free radicals, Free radicak are naturally produced as a consequence of oxygen metabolism, and are considered “universal” mediators of cell and tissue damage, since their reactions are believed to play a part in a great variety Received April 3, 1989; accepted August 22, 1989.
of physiological and pathological processes, including ischemiakeperfusion damage to transplant organs (10, 20). Iron has been shown in vitro and in vivo to catalyze powerful and biologicaIly destructive oxidants (2, 8, 19, 21, 40), and since iron is present intracellularly in a variety of forms and may also be inadvertently present in the extracellular environment (i.e., capillary bed) via the perfusate, its role as a potential mediator of damage to transplant organs or cells may be significant (14, 16, 17, 28). We have used mammalian cells in culture and colony survival assays to study hypothermic damage free from the effects of previous or simultaneous hypoxia. We have discovered a previously unreported mechanism of hypothermic damage, one that is iron mediated, occurs extracellularly and
4.52 001 l-2240/90 $3.00 Copyright 0 1990 by Academic Press, Inc. Au rights of reproduction in my form reserved.
intracellularly, and can be prevented to varying degrees by iron chelation. MATERIALS
Cell Culture and Media Preparation
The cells used were originally derived from the V-79 S-171 line of Chinese hamster fibroblasts; the subline was S-171-Wl (13). The plating efficiency of this line is about 8&95%. The cells were maintained at 37°C in an atmosphere of 95% air + 5% carbon dioxide in exponential growth phase using Eagle’s basal medium (BME powder formula with Hanks’ salts) supplemented with antibiotics (penicillin G, 78 IUlml, and streptomycin sulfate, 78 pg/ml), sodium bicarbonate (2.75 g/liter), and 10% fetal calf serum (all components from GIBCO Laboratories, Grand Island, NY). The resulting mixture will, henceforth, be called regular growth medium. Cells were subject to hypothermic stress in regular growth medium [without sodium bicarbonate (SBC)] buffered by 25 mM Hepes [4-(Z-hydroxyethyl)-l-piperazineethanesulfonic acid, sodium salt], henceforth to be called Hepes medium; therefore, the 5% CO, atmosphere was not required to maintain physiological pH. A concentrated stock solution containing the desired hypothermic protectant was prepared in Hepes medium (with one exception), then filter sterilized with a 0.22~p,m Millex-GS filter unit (from Millipore Products Division, Bedford, MA), and subsequently diluted to the desired experimental concentration in Hepes medium. The exception, FeCl,, was dissolved in distilled H,O to prevent precipitation of iron. In this case, the respective control medium was diluted to the same minimal extent with distilled H,O. Iron-free Hepes medium was prepared by substituting regular distilled deionized water with that from the Milli-Q Water System (Millipore Canada Limited). Desferal (deferoxamine mesylate) was purchased from Ciba-Geigy Canada Ltd.,
DETAPAC (diethylenetriaminepentaacidic acid) and EDTA (ethylenediaminetetraacetic acid) from Sigma Chemical Company (St. Louis, MO), EGTA (ethylenebisoxyethylenenitrilotetraacetic acid) from J. T. Baker Chemical Company (Phillipsburg, NJ), and FeCI, (ferric chloride) from Fisher Scientific. Cell Survival Assessment by Colony Growth Assay
Two days before an experiment, cells were detached from the surface of glass culture flasks by a 2- to 3-min room temperature treatment with a solution containing 0.01% trypsin, 0.1% EDTA in Ca2’- and MgZt-free Hepes-buffered saline, Cells were resuspended in regular medium, gently aspirated, and then counted in a Celloscope (Particle Data Corp.). Following this, 2 x lo6 cells were resuspended in a total volume of 10 ml, and then reincubated for 24 hr, after which this procedure was repeated, ensuring that cells would be in exponential growth phase for the experiment. On the day of the experiment, cells were trypsinized, resuspended in Hepes medium, aspirated, and then counted. At this time, cell concentrations were adjusted to suit the treatment (period of cold exposure with/without protectant) by inoculating enough single cells into the experimental flasks so that l&200 colonies would result per plastic flask at the end of the colony assay period (34). [Each experiment included four data points or periods of cold exposure (“cold dose”) per survival curve, and each point was done in triplicate.] The cells were then incubated to allow them to attach and repair trypsin damage for 3 hr in regular growth medium, preequilibrated to 37°C and pH 7.4. At this point the medium was removed and replaced with either room temperature equilibrated Hepes medium or experimental medium. Flasks containing cells destined for 25, 20, 15, 10, or 0°C exposure were sealed and placed in the cold in a circulating water bath (&OS”C); for 5°C
exposure, flasks were placed unsealed in a cabinet drawer in a walk-in incubator (+-OYC). In some 5°C experiments, Desferal in Hepes medium was injected into flasks with a syringe during the period of cold exposure. After exposure to the cold, the flasks were incubated for 3 hr at 37°C so that any unattached, viable cells could reattach. “Plating efficiency” flasks (in triplicate), containing cells and Hepes or experimental medium, were returned to the 37°C incubator for 3 hr, instead of being subject to cold, to determine the plating efficiency (PE) of the cells in respective media free from hypothermic stress. The medium was then removed from both the experimental and control flasks (as well as the plating efficiency flasks) and replaced by regular growth medium after which the flasks were incubated at 37°C for 6-8 days to allow growth of single cells into colonies. The resulting colonies were stained with methyIene blue and counted. Throughout the course of the experiment, the exposure of cells in Hepes medium to fluorescent light was minimized by keeping test tubes and flasks wrapped or covered. At no time were cells in Hepes medium exposed to direct fluorescent light; only indirect lighting was used during periods of cell manipulation. [We have previously shown that fluorescent light reacts with Hepes-buffered medium to significantly reduce cell survival (44, 45).] The fraction survival, as a function of “cold dose” (or time at a particular temperature), is obtained by multiplying the “colony surviving fraction” at that dose by 100 and dividing through by the plating efflciency (expressed as a percentage). All points were plotted with standard errors, unless they were smaller than the points. Cell Survival Curves and Arrhenius Plots The cell survival curve consists of a semilogarithmic plot of fraction survival versus cold dose. In accord with standard
terminology (9), the survival curve can be represented by two numbers, n and &. Extrapolation of the linear portion of the curve to the log axis will yield the extrapolation number, n. The Do is defmed as the time or cold dose required to reduce the survival by a factor of l/c on the linear portion of the semilog graph. II,, is inversely related to the cold sensitivity of the cehs. If the curve has a “shoulder,” the size of the shoulder is a quantitative estimate of the amount of sublethal damage a cell can accumulate before this damage becomes lethal. This size is given by the parameter D,, the quasithreshold “dose,” which can be found by extrapolation of the linear portion of this curve to the point where it meets the horizontal line drawn through 100% survival (Le., fraction survival = 1). Alternatively, D, = D, X Inn. Arrhenius plots can be used to model cell inactivation according to the inactivation of single targets within the cell, such as proteins, lipids, or deoxyribonucleic acid (DNA). The logarithm of the cell inactivation rate (k = l/Do) is plotted versus reciprocal temperature? where k can be considered “the rate of cell killing,” or for a critical target in the cell, say, a protein, the rate of inactivation, For simple chemical reactions, the reaction rate is an exponential function of temperature, k = t?-Ea’RT, where E, is the activation energy or “energy barrier” of the reaction, and is calculated from the slope of the plot (E, is also a measure of the temperature coefftcient of the reaction); A is a collision frequency constant; and R is the gas constant. Molecular inactivations also conform to the Arrhenius equation. Hence, the logarithm of cell inactivation constants plotted against reciprocal temperature should yield a straight line if there is a direct relationship between cell killing and inactivation of a molecular target. From this model, the activation energy for cell killing is equal to the activation energy for inactivation of the rate-limiting critical target (26).
PROTECTION BY IRON CHELATION
Figures 1 and 2 show survival curves for control cells and Desferal-treated cells during 0, 5, 10, 15, 20, and 25°C exposures. Survival in Hepes (control) medium reaches a maximum at lO”C, and decreases above and below this temperature. This is indicative of two different hypothermic damage mechanisms above and below 10°C (24). In a 50 PM Desferal medium, the same maximum level of survival is achieved at 5 and lO”C, and it is significantly higher than in the control medium. The greatest protection by Desferal occurs at O”C, and decreases with increasing temperature until 2O”C, at which point sensitization occurs (i.e., at 20 and 25°C). Arrhenius plots (Fig. 3) for control cells in Hepes medium and cells in 50 PM Desferal medium during hypothermia are qualitatively similar to previously published plots (24). These curves have discontinuities around 7 to 8”C, which correspond to maximum D,‘s in the respective media, or minima in the rates of cell inactivation. Above the break inactivation rates conform to the Arrhenius equation, increasing exponentially with increasing temperature. The inactivation energies for control cells and Desferal-treated cells in the region above
days FIG. 1. Fraction survival of cells stored in Hepes
medium at 0°C (squares), 5°C (octagons), 10°C (triangles), 15°C (diamonds), 20°C (inverted triangles), and 25°C (circles)
days FIG. 2. Fraction survival of cells stored in Hepes medium supplemented with 50 +W Desferal at 0°C (squares), 5°C (octagons), 10°C (triangles), 15°C (diamonds), 20°C (inverted triangles), and 25°C (circles).
the break are 10.67 and 26.16 kcal/mol, respectively. Below the break, the inactivation energies have “negative” values of 57.89 and 44.17 kcal/mol for control and Desferal-treated cells, respectively. The two curves cross at 19.17”C, above which Desferal sensitizes cells to hypothermia, and below which Desferal protects. The logarithmic plot of the ratios of cell inactivation rates, kcontroJkDesferal, is linear (Fig. 3B) when plotted versus the inverse of the temperature. A 3-hr prehypothermic incubation in 50 PM Desferal followed by replacement with control medium for the 5°C exposure is sufficient to protect cells against damage. This is shown in Fig. 4. The mechanism may be intracellular, since Desferal is reported to accumulate within the cell at 37°C (5, 25), and the corresponding cold exposure medium was chelator free. In fact, Desferal was not present during the hypothermic exposure in any of the experiments shown in Fig. 4. Cells exposed to the external iron chelator DETAPAC during the 5°C exposure show slightly better survival than those receiving only the prehypothermic treatment with Desferal, with the best survival resulting from prehypothermic intracellular chelation plus hypothermic extracel-
ZIEGER ET AL.
10-l 3 70
l/T ( X 1000 l/K )
I 3 62
l/T ( X 1000 1,‘K )
FIG. 3. (A) Arrhenius plot of ceil inactivation rates in Hepes medium (closed circles) and in 50 M Desferal-supplemented Hepes medium (open circles). (B) Logarithmic plot of the ratios of cell inacas a function of reciprocal temperature. tivation rates (keon~Ol/kDcafernl)
Mar chelation. We have previously shown (44) that cell survival at 5°C in control medium or 50 &Sf DETAPAC medium could not be improved by a 3-hr 37°C prehypothermic treatment with 50 +I4 DETAPAC (results not shown here). This is contrary to the protection offered by a Desferal pre-
treatment, and provides evidence for the different mechanismsof action of these two chelators. Figure 5 shows survival curves for cells exposed to 5°C in 25 p& DETAPAC vs 12.5 a Desferal + 12.5 Qf DETAPAC. Both media contain 25 &V of extracellular iron chelatability, but only the latter contains any direct intracellular ability to chelate iron, and survival in it is significantly higher, providing testament to internal damage/protection mechanisms. The manner in which cells can be sensi-
days FIG. 4. Fraction survival of cells stored at 5°C with a prehypothermic Desferal treatment, but with no Desferal present during the hypothermic exposure, A 3-hr prehypothermic treatment of cells with 50 t.tM Desferal in the plating medium followed by no added protection at 5°C (Hepes medium only) protects cells ([email protected]
) when compared to the control (circles). Similarly, a 3-hr prehypothermic treatment of cells with 50 t&f Desferal in the plating medium followed by 50 @I DETAPAC at 5°C (i.e., no Desferal, inverted trianpies) improves survival compared to 50 PM DETAPAC at 5°C with no prehypothermic Desferal treatment (diamonds).
days FIG. 5. Fraction survival of cells stored at 5°C in Hepes medium supplemented with either 12.5 fl Desferal + 12.5 pN DETAPAC (squares) or 25 t&j DETAPAC atone (triangles). The control (Hepes medium at 5°C) is represented by circles.
tized may help elucidate the actual damage/ protection mechanisms of hypothermia. If the chelation of iron in the culture medium protects cells from hypothermic damage, then the addition of ferric ions to the medium should sensitize cells. Figure 6 shows that 200 t.&f FeCl, sensitizes cells to hypothermia compared to a control medium of equal osmolarity (363 p&I NaCl). Also, 50 tGI4Desferal partially desensitizes the cells to 200 pI14Fe&, and 250 @f Desferal + 20.0 00 ‘40 80 12.0 16.0 200 w Fe& promotes survival to the 50 days @I Desferal supplementation level. These FIG. 7. Fraction survival of cells stored at 5°C in results confirm a potential role for ferric Hepes medium made with substantially iron-free water ions in external sensitization which is re- and supplemented with Desferal. Protection results for versible by 1:1 chelation with Desferal. cells stored in Hepes medium made with iron-free waThis does not preclude any internal damage ter (inverted triangles) compared to cells stored in regular Hepes medium made with distilled deionized wamechanisms. To investigate the source of sensitizing ter (control, circles). There are negligible differences in survival between an iron-free water/Desferal meiron, the cell culture medium was prepared dium (triangles) and a Desferal medium made with regwith iron-free water (Milli-Q Water Sys- ular distilled deionized water (squares). tem). Figure 7 shows the protective effect of iron contaminant-free medium at 5°C; however, survival is not as high as with 50 t&f Desferal in regular medium. This suggests that the extra protection by Desferal may be due to chelation of serum components and/or chelation in the intracellular medium. Trace iron contaminants introduced via glassware and pipets may preclude the higher survival of a truly iron-free medium. Nevertheless, there are negligible differences in survival between an iron-free water/Desferal medium and a Desferal medium made with regular distilled deionized water. To investigate the mechanism of action days by Desferal, 50 t.GI4of the chelator was added to cells at different points in the 16 FIG. 6. Fraction survival of cells stored at 5°C in FeCI,, and Desferal. Cells in Hepes medium supple- to 19-day-long exposure to 5°C. Desferal mented with 200 pM FeCI, are sensitized to hypo(50 kIt4) added just prior to rewarming has thermia (inverted triangles) compared to the control no protective effect (see Fig. 8A); there(circles). This sensitization is reduced by 1:1 iron chefore, the damaging event occurs during hylation with Desferal-ells in Hepes medium supplemented with 200 pkf FeCl, + 50 uM Desferal (dia- pothermia, and is not associated with the monds) have a survival that is intermediate between warming-up process, The drop in survival those of control and 200 &W FeCl,-treated cells. Cells relative to the control (cells in Hepes mein Hepes medium supplemented with 200 t&f FeCl, + dium) occurs consistently with the addition 250 @I Desferal (squares) are protected to a similar of extra medium to cold cells. When extra extent as those supplemented with 50 fl L&feral control medium is added to control cells at only (triangles).
days FIG. 8. Fraction survival of cells stored at 5°C in Hepes medium (open circles) and in 50 k&f Desferal added at Day 0 of the cold exposure (open squares). (A) Desferal added to Hepes medium at 5°Cjust prior to rewarming (closed squares, dashed line) has no protective effect. (B) Desferal (closed squares, dotted line) and Hepes medium (closed circles, dotted line) added to control cells on Day 2 of the cold exposure-already by Day 4, Desferal protects from the sensitization of medium addition, and ultimately increases survival. (C) Desferal (closed squares, dashed line) and Hepes medium (closed circles, dashed line) added to control cells on Day 6 of the cold exposure.
Day 2 or 6, fewer survivors are seen within chelators. IntraceIIular and extracellular 1 to 2 days (see Figs. 8B and C). If the ferric iron chelation with 50 pJ4 Desferal added medium contains Desferal, this de- promotes the highest survival, producing a crease in survival is still apparent but less Do more than three times greater than that severe, and within 3 to 6 days extra protec- of the contro1 medium. Desferal (100 CIM) tive action by Desferal can be seen as the failed to promote higher survival (44) (resurvival curves begin to diverge. Also, Fig. sults not shown here); since Desferal che8 shows greater resolution in the shoulder lates ferric iron in a 1: 1 ratio, the sensitizaregion of survival curves for Desferal- tion of cells to 5°C is by 50 &%f iron or less. treated and control cells at 5°C than the pre- Extracellular iron (ferric and ferrous) chevious figures. Both curves exhibit positive lation with 50 FM DETAPAC protects to a Dq’s, and the protective action by Desferal lesser, though still significant, degree, is seen within 2 days. while 110 PM EDTA provides a minor imThe 5°C survival curves in Fig. 9 illus- provement in cell survivaI. Because EDTA trate various degrees of protection by iron is also a general divalent chelator, cell sur-
BY IRON CHELATION
protection (Do increases 1.4~ at 10°C and 1.2x at 533. DISCUSSION
10-3kTixFzTdo Clays FIG. 9. Fraction survival of cells stored at 5°C in Hepes medium (control, circles), and supplemented with 50 PM Desferal (squares), 50 $4 DETAPAC (triangles), 110 +LM EDTA (diamonds), and 110 &M EGTA (inverted triangles).
vival was assessedin a medium with a different divalent chelator of lower ferrous but higher Ca” affinity; Fig. 9 shows that 110 FM EGTA produces no significant protection. The 10°C survival curves in Fig. 10 illustrate protection by iron chelators. Desferal (50 CLM)increasesD, by just over two times relative to the control medium, which is less significant than the protection at 5°C. DETAPAC (50 pM) provides less protection at 10 than at 5°C (D, increases 1.6X versus 2.0~), while EDTA provides similar
days FIG. 10. Fraction survival of cells stored at 10°C in Hepes medium (control, circles), and supplemented with 50 w Desferal (squares), 50 @I DETAPAC (triangles), or 110 fl EDTA (diamonds).
It has previously been reported that 10°C was the optimum storage temperature for single, attached V-79 cells in regular (sodium bicarbonate-buffered) medium (24). The present study has shown this to also hold true in Hepes-buffered regular medium. Supplementation of Hepes medium with 50 fl Desferal protected cells at temperatures below 2O”C,with the same maximal survival attained at 5 and 10°C. The Arrhenius plot of inactivation rates of cells in regular medium contains a break at approximately 7”C, which corresponds to the minimum inactivation rate. This feature has been shown by others, and marks distinct damage mechanisms at higher and lower temperatures (24). Above 7°C (range I), the damage mechanism has an inactivation rate which conforms to the Arrhenius equation (k increases exponentially with temperature); the activation energy (EJ for control cells is 10.7 kcal/mol. Presumably this represents the E, for inactivation of the critical target for hypothermic cell killing. This value is in the range of temperature coefficients of metabolic processes and much less than that for protein denaturation. Desferal(50 PM) increasesE, to 26.2 kcaVmo1, resulting in progressively less protection with increasing temperature and sensitization above about 19°C. Below 7°C (range II), we see the peculiarity of an “apparent negative” activation energy, and progressively greater inactivation rates at lower temperatures. The only way an apparent negative activation energy can be obtained is from a two-step process or mechanism. We have previously given evidence that one of these steps, probably the initial step, involves a membrane lipid phase transition (24). E, for control cells is similar to the previously published value (“ - “57.9 versus “ - “61 kcal/mol) (24). The effect of Desferal on the reaction rates
associated with the mechanism of cell killing below 7°C is to “decrease” E, to -44.2 kcal/mol . Numerous cellular events at hypothermic temperatures have been reported. For example, sharp increasesin the rates of DNA and protein synthesis occur between 18 and 21.6”C, and are unaccompanied by cell division, possibly accounting for increased sensitivity at higher temperatures, i.e., 25°C (30) Electron spin resonance studies of V-79 cell membranes show a lipid phase transition centered at approximately 20°C (36), and refluidizing the membrane with BHT at or below this temperature facilitates the repair of sublethal radiation damage (31). Ward et al. have shown that single-strand breaks in DNA and consequential killing of V-79 cells by superphysiological amounts of HZO, is significantly more damaging at 37 than at 0°C; they hypothesized that site-specific *OH damage induced at 37°C was accompanied by misrepair, while the repair mechanism was not active during the 0°C exposure (39). We now have evidence that there is an ironrelated damage mechanism at low temperature. Thus, hypothermic damage to V-79 cells is probably a result of multiple simultaneous mechanisms (perhaps metabolic and free radical), as well as endogenousrepair mechanisms. The process of hypothermic protection by Desferal is likely complicated as well; the iron chelator probably interacts with regular metabolism (3), as well as repair processes and pathological (free radical) reactions. The result is a varied protective response to hypothermic cell damage as a function of temperature (Fig. 3A). However, since we observe linearity in the log plot of kcontra,/knesferti versus reciprocal temperature (Fig. 3B), the mechanism of action of Desferal must have a single temperature coefftcient throughout the temperature range of 0 to +25”C. From Fig. 3A, it is also apparent that Desferal alters the temperature coefftcient for hypothermic cell killing (otherwise the line in
Fig. 3B, while linear, would have a zero slope). Hence, the decreasein protection as the temperature is raised, eventually resulting in sensitization above 19”C, is due to the difference in the temperature coefficients for hypothermic cell killing in the presence and absence of Desferal. The physiological success of iron chelation is dependent on many factors, including the optimal coordination of iron by the chelator, an effective binding affinity for iron, especially in the presence of competing or interfering ions, the bulk of the chelator and its possible effect on the kinetics of iron mobilization, and the environment of iron chelation (33). Both Desferal and DETAPAC have very high affinities for ferric iron [standard and effective stability constants are given by (33)], and in view of the excess chelator concentrations used, the difference in hypothermic protection offered probably relates to the environment of iron chelation. Regular medium supplemented with 50 @I Desferal protects maximally against hypothermic damage. DETAPAC (50 $M) also protects at 5°C but less effectively than Desferal. This extra level of protection by Desferal may be due to the inhibition of two distinct damage mechanisms:one external and one internal. A 3-hr prehypothermic treatment with Desferal, followed by its absence during hypothermic (5°C) exposure, provides substantial though submaximalprotection from cell death. A 3-hr prehypothermic treatment with DETAPAC followed by its absence during hypothermia failed to protect cells. However, 50 PM DETAPAC does offer protection when administered during hypothermia, and the addition of a DETAPAC prehypothermic treatment does not increase this level of survival, suggesting that the positive effect of DETAPAC is not related to an intracellular damage protection mechanism. Conversely, a Desferal pretreatment followed by only DETAPAC at 5°C protects better than DETAPAC at 5°C alone. Also, survivaI at 5°C in 12.5 t&
Desferal + 12.5 a DETAPAC greatly exceedsthat in 25 r.liwDETAPAC alone, even though both media contain roughly equal extracellular chelating power. These experiments imply that iron mediates both intracellular and extracellular hypothermic damage mechanisms. If there were indeed only an extracellular damage mechanism, it seemsunlikely that Desferal could become sticiently internally concentrated during the 3-hr prehypothermic treatment to subsequently release enough unbound surplus chelator to the regular medium during hypothermia to protect cells from extracelluiar mediated damage. The extracellular chelators EDTA and EGTA provide less signticant protection at 5°C than Desferal or DETAPAC. This may be due to lower specificity for iron, resulting in the binding of competing ions like calcium. Also, these chelates contain an open coordination site and hence are not as successful at restricting the reactivity of bound iron (15). With this feature, EDTA can promote the iron-catalyzed production of free radicals under certain conditions (II, 15, 18, 38) by increasing the concentration of iron in solution (15), which otherwise would be precipitated. The sensitivity of cells to hypothermia is intensified by the addition of 200 fl FeCl, to regular Hepes medium, which must have sufficient chelators to keep the extra iron in solution for the duration of the experiment (no precipitate was observed after 16 days cold storage). These chelators may be lowmolecular-weight medium components or organic contaminants in the water, and must bind iron weakly since the sensitizing effect of Fe3+ is completely inhibited by 1:1 chelation with Desferal. The concentration of trace iron in our medium (10 ).LMor less) is only slightly higher than physiological levels but is sufficient to sensitize cells at 5°C. Reducing the amount of iron and organic contaminants (potential chelators) with the Milli-Q Water System modestly
BY IRON CHELATION
improves survival. Residues of iron contaminants in culture flasks and pipets may account for some of the shortfall in survival with respect to the maximum attainable with Desferal, but a decrease in the external iron level may have little effect on intracellular iron and potential intracellular damagemechanisms,Iron, when added in a safe association with the transport protein transfenin, was found to be nontoxic at 5°C (44); it is not known if transfenin-bound iron is internalized at this temperature. Trace amounts of contaminating iron in transplant organ perfusates have been shown to promote biochemical and functional deterioration in baboon and pig hearts following hypothermic perfusion (42, 43). It has been speculated that metals in the perfusate may be a source of free radicals that are damaging to the vascular endothelium even before reperfusion-generated free radicals appear (4). Such primary lesions may be amplified during reperfusion by the infiltration of neutrophils attracted to the site of injury where they generate superoxide and injure tissue (7). We have now shown evidence of an iron-promoted extracellular damage mechanism in cultured cells exposed to hypothermia under nonischemic conditions. This low-temperature damage mechanism may be free radical mediated, and may contribute to the much-cited ischemia/repetfusion injury to transplant organs. Our results reaffirm the strict requirement for ultraclean perfusates in a transplant organ storage regime. Cells obtain iron from extracellular transfen-in, and much of this iron is used in the mitochondria to synthesize heme for mitochondrial enzymes, for widely distributed enzymes like cytochrome P450, or for specialized proteins such as hemoglobin and myoglobin (22). Iron which is in excess of metabolic requirements accumulates in the transient storage protein ferritin (22). Jacobs has described an intracellular transit iron pool consisting of low-molecular-
weight iron chelates which maintain an equilibrium between iron uptake, iron storage, and iron utilization within the cell (22). This pool may be the target of intracellular chelation by DesferaI (5, 6, 35, 41). Alternatively, iron binding by Desferal may occur in lysosomes where it complexes iron released from ferritin under the conjugate actions of acidic pH and lysosomal enzymes (25). Cytoplasmic ferritin is deemed an unlikely source of chelatable iron as the channels in this protein limit free access of the relatively large chelator (Desferal) to the iron core (23). Some intracellular forms of low-molecular-weight iron chelates can support free radical production in vitro (If, 12, 15, 38) and lipid peroxidation in vivo (1, 29). Also, it has been proposed that the lysosomal pool is the source of ferric iron necessary for the killing of hepatocytes by H202 via the Haber-Weiss reaction (27, 37). Free radicals are produced in the cell by many metabolic reactions, and natural scavenging mechanisms exist to keep these metaboiites in check (10). It is possible that hypothermic stress alone can create a metabolic imbalance by impairing endogenous scavenging mechanisms or by overloading the cell with dangerous free radicals. Such an overload may be the result of iron being released from a safe association or storage site to a more active chelate, capable of catalyzing the production of ‘OH or similarly reactive free radicals, and promoting lipid peroxidation. Desferd was shown to reduce lipid peroxidation in cold (OOC)ischemic rabbit kidneys when added to the initial flush and cold storage solutions (17). Signs of lipid peroxidation predominated when cold storage was followed by in viva reperfusion (16); treatment of the tissue homogenates of cold-stored porcine kidneys with Desferal just prior to in vitro incubation reduced this damage (14). The authors hypothesized that the observed lipid peroxidation products
resulted from the iron-catalyzed breakdown of lipid hydroperoxides produced in the cold ischemic preservation period by the reaction of low-density lipoproteins with free radicals (14). However, their experimental methods did not allow them to determine the exact site of damage (intraceliular or extracellular), nor could they identify the specific stress responsible. In our experiments, treating cold-stored cells (5°C) with DesferaI just prior to rewarming was an ineffective means of protection. This failure of the chelator to act as a “rescue agent” and the prevalence of fully oxygenated conditions (44) preclude an “ischemia/reperfusion” damage mechanism. Desferal(50 l.tM) added on the second or sixth day of the cold exposure ultimately salvages survival for some of the cells not yet killed. From Figs. 8B and C we see that by Day 6, and possibIy even by Day 2, the surviving cells in Hepes control medium (open circles) have accumulated significant damage (survivals lie on the linear part of the curve). Without the intervention of the iron chelator, subsequent survivals would lie further along this part of the curve. However, within 2 days of the addition of extra medium, we see that Desferal minimizes the sensitization experienced by control cells, and within 3 to 6 days, Desferal facilitates a decrease in the rate of killing as witnessed by the decrease in the survival curve slope, The events corresponding to these curve features may be immediate protection from extracellular mediated damage, with intracellular protection taking effect only after enough chelator has diffused through a solidified plasma membrane. Thus, we have identified a distinct hypothermic damage mechanism(s) using a nonischemic tissue culture system. This unique form of injury is iron mediated, and our evidence shows that it occurs in both intracellular and extracellular environments. The precise physiological and biochemical mechanisms are yet to be identi-
fied, but may be elucidated with further investigations of the protective actions of iron chelators and the kinetics of iron mobilization from the cell. ACKNOWLEDGMENT
This research was supported by grants from the Natural Sciences aud Engineering Research Council of Canada. REFERENCES
1, Anderson, D. K., and Means, E. D. Iron-induced lipid peroxidation in spinal cord: Protection with mannitol and methylprednisolone. Free Radicals Biof. Med. 1, 59-64 (1985). 2. Aust, S. D., Morehouse, L. A., and Thomas, C. E. Role of metals in oxygen radical reactions. Free Radicals Eiol. Med. 1, 3-25 (1985). 3. Barankiewicz, J., and Cohen, A. Impairment of nucleotide metabolism by iron-chelating deferoxamine. Biochem. Pharmacol. 36, 2343-2347 (1987). 4. Bennet, J. F., Bry, W. I., Collins, G. M., and Halasz, N. A. The effects of oxygen free radicals on the preserved kidney. CvyoBiology 24, 264-269 (1987). 5. Bottomley, S. S., Wolfe, L. C., and Bridges, K. R. Iron metabolism in K562 erythroleukemic cells. J. Biol. Chem. 260, 6811-6815 (1985). 6. Bridges, K. R., and Cudkowicz, A. Effect of iron chelators on the transferrin receptor in K562 cells. J. Biol. Chem. 259, 12970-12977(1984). 7. Bulkley, G. 3. Free radical-mediated reperfusion injury: A selective review. hit. J. Cancer 55 (Suppl. VIII), 66-73 (1987). 8. Dunford, H. B. Free radicals in iron-containing systems. Free Radicals Biol. Med 3, 405421 (1987). 9. Elkind, M. M., and Whitmore, G. F. “The Radiobiology of Cultured Mammalian Cells,” pp. 7115. Gordon & Breach, New York, 1967. 10. Feher, J., Csomos, G., and Vereckei, A. “Free Radicals: Reactions in Medicine.” SpringerVerlag, Berlin/Heidelberg, 1987. Il. Flitter, W., Rowley, D. A., and Halliwell, B. Superoxide-dependent formation of hydroxyl radicals in the presence of iron salts: What is the physiological iron chelator? FEBS Left. 158, 310-312 (1983). 12. Floyd, R. A., and Lewis, C. A. Hydroxyl free radical formation from hydrogen peroxide by ferrous iron-nucleotide complexes. Biochemisrry 22, 2645-2649 (1983). 13. Frim, J., Kruuv, I., Frey, H. E., and Raaphorst,
BY IRON CHELATION
G. P. Survival of unprotected mammalian plateau-phase cells following freezing in liquid nitrogen. Cryobiology 13, 475-483 (1976). 14. Fuller, B. J., Myles, M., Healing, G., Simpkin, S., Palmer, A., Lunec, J., and Green, C. J. Investigation of protective agents against oxidative stress during hypothermic storage of pig kidneys. C&o-Letters 7, 103-112 (1986). 15. Graf, E., Mahoney, J. R., Bryant, R. G., and Eaton, J. W. Iron-catalyzed hydroxyl radical formation: Stringent requirement for free iron coordination site. J. Biol. Chem. 259, 3621% 3624 (1984). 16. Green, C. J., Healing, G., Lunec, J., Fuller, B. J., and Simpkin, S. Evidence of free-radicalinduced damage in rabbit kidneys after simple hypothermic preservation and autotransplantation. Transplantation 41, 161-165 (1986). 17. Green, C. J., Healing, G., Simpkin, S., Fuller, B. J., and Lunec, J. Reduced susceptibility to lipid peroxidation in cold ischemic rabbit kidneys after addition of desferrioxamine, marmitol, or uric acid to the fiush solution. CryobioIogy 23, 35&365 (1986). 18. Gutteridge, J. M. C., Richmond, R., and Halliwell, B. Inhibition of the iron-catalyzed formation of hydroxyl radicals from superoxide and of Lipid peroxidation by desfenioxamine. Biothem. J. 184, 469-472 (1979). 19. Halliweil, B., and Gutteridge, J. M. C. Oxygen toxicity, oxygen radicals, transition metals and disease. Biochem. J. 219, l-14 (1984). 20. Halliwell, B., and Gutteridge, J. M. C. “Free Radicals in Biology and Medicine.” Oxford Univ. Press, (Clarendon), London/New York, 198.5. 21. Halliwell, B., and Gutteridge, J. M. C. Oxygen free radicals and iron in relation to biology and medicine: Some problems and concepts. Arch. Biochem. Biophys. 246, 501-514 (1986). 22. Jacobs, A. An intracellular transit iron pool. Iron metabolism, Cibn Found. Symp. 51, 91-100 (1977). 23. Kontoghiorghes, G. J., Chambers, S., and Hoffbrand, A. V. Comparative study of iron mobilization from haemosiderin, ferritin and iron (III) precipitates by chelators. Biochem. J. 241, 87-92 (1987). 24. Kruuv, J., Glofcheski, D., Cheng, K. H., Campbell, S. D., Al-Quysi, H. M. A., Nolan, W. T., and Lepock, J. R. Factors influencing survival and growth of mammalian cells exposed to hypothermia. I. Effects of temperature and membrane lipid perturbers. J. Cell. Physiol. 115, 179-185 (1983).
25. Laub, R., Schneider, Y.-J., Octave, J.-N., Trouet, A., and Crichton, R. R. Cellular pharmacology
ZIEGER ET AL. of deferrioxamine B and derivaties in cultured rat hepatocytes in relation to iron mobilization. Biochem. Pharmacol. 34, 1175-1183 (1985). Lepock, J. R., Frey, H. E., Rodahl, A. M., and Kruuv, J. Thermal analysis of CHL V79 cells using differential scanning calorimetry: Implications for hyperthermic cell killing and the heat shock response. J. Cell Physiol. 137, 14-24 (1988). Mello Filho, A. C., Hoffman, M. E., and Meneghini, R. Cell killing and DNA damage by hydrogen peroxide are mediated by intracellular iron. Biochem. J. 218, 273-275 (1984). Myers, C. L., Weiss, S. J., Kirsh, M. M., Shepard, B. M., and Shlafer, M. Effects of supplementing hypothermic crystalloid cardioplegic solution with catalase, superoxide dismutase, allopurinol, or deferoxamine on functional recovery of giobally ischemic and reperfused isolated hearts. J. Thorac. Cardiovasc. Surg. 91, 281-289 (1986). Nayini, N. R., White, B. C., Aust, S. D., Huang, R. R., Indrieri, R. J., Evans, A. T., Bialek, H., Jacobs, W. A., and Komara, J. Postresuscitation iron delocalization and malondialdehyde production in the brain following prolonged cardiac arrest. Free Radicals Biul. Med. 1, 11l-l 16 (1985). Nelson, R. J., Kruuv, .I., and Frey, H. E. Effect of sub-optimal temperatures on survival of mammalian cells. Exp. Cell Res. 68, 247-252 (1971). Nolan, W. T., Thompson, J. E., Lepock, J. R., and Kruuv, J. Effect of membrane lipid perturbers on the temperature dependence of repair of sublethal and potentially lethal radiation damage. Inz. J. Radial. Biol. 39, 195-205 (1981). Pegg, D. E. The biology of cell survival in vitro. “Organ Preservation for Transplantation” (Karow and Pegg Eds.), 2nd ed., pp. 31-52. Marcel Dekker, New York, 1981. Pitt, C. G., and Martell, A. E. The design of chelating agents for the treatment of iron overload. In “Inorganic Chemistry in Biology and Medicine” (Martell, Ed.), pp. 277-312. American Chemical Society, Washington, DC, 1980. Puck, T. T., and Marcus, P. I. A rapid method for viable cell titration and clone production with HeLa cells in tissue culture: The use of X-
irradiated cells to supply conditioning factors. (1955). Rogers, J., and Munro, H. Translation of ferritin light and heavy subunit mRNAs is regulated by intracellular chelatable iron levels in rat hepatoma cells. Proc. Natl. Acad. Sci. USA 84, 2277-2281 (1987). Rule, G. S., Law, P., Kruuv, J., and Lepock, J. R. Water permeability of mammalian cells as a function of temperature in the presence of dimethylsulfoxide: Correlation with the state of membrane lipids. J. Cell Physiol. 103, 407416 (1980). Starke, P. E., Gilbertson, J. D., and Farber, J. L. Lysosomal origin of the fenic iron required for cell killing by hydrogen peroxide. Biockem. Biophys. Res. Commun. 133, 371-379 (1985). Sutton, H. C. Efficiency of chelated iron compounds as catalysts for the Haber-Weiss reaction. Free Radicals Biol. Med. 1, 195-202 (1985). Ward, J. F., Evans, J. W., Limoli, C. L., and Calabro-Jones, P. M. Radiation and hydrogen peroxide induced free radical damage to DNA. Brit. 1. Cancer 55 (Suppl. VIII). 105-112 (1987). Weiss, G. P., Oxygen, ischemia, and inflammation. Acta Physiol. Stand. Suppl. 548, 9-37 (1986). White, G. P., Bailey-Wood, R., and Jacobs, A. The effect of chelating agents on cellular iron metabolism. Clin. Sci. Mol. Med. 50, 145-152 (1976). Wicomb, W. N., Cooper, D. K. C., and Novitzky, D. Loss of myo&dial viability following hypothermic perfusion storage from CORtaminating trace elements in the perfusate. Proc. Natl. Acad. Sci. USA 41,432-437
43, 23-29 (1987).
43. Wicomb, W. N., Novitzky, D., Cooper, D. K. C., and Rose, A. G. Forty-eight hours hypothermic perfusion storage of pig and baboon hearts. J. Sarg. Res. 40, 276-284 (1986). 44. Zieger, M. A. J. “The Effect of Hypothermia on Cultured Mammalian Cells.” M.Sc. thesis, University of Waterloo, Waterloo, Ontario, 1988. 45. Zieger, M. A. J., Glofcheski, D. J., Lepock, J. R., and Kruuv, J. Factors influencing survival of mammalian cells exposed to hypothermia. V. Effects of Hepes, free radicals and H,O, under light and dark conditions. CryoBiology, (in press).