Metabolism and toxicity of hydrochlorofluorocarbons: current knowledge and needs for the future.

Hydrochlorofluorocarbons (HCFCs) are being developed as replacements for chlorofluorocarbons (CFCs) that deplete stratospheric ozone. The depletion of stratospheric ozone may increase the intensity of ultraviolet radiation at the earth's surface, which may be associated with global, adverse human health effects. The greater tropospheric lability of HCFCs, which is due to the presence of C-H bonds, reduces HCFC migration to the stratosphere; HCFCs should, therefore, cause less depletion of stratospheric ozone than CFCs. HCFCs under development include HCFC-22 (chlorodifluoromethane), HCFC-123 (2,2-dichloro-1,1,1-trifluoroethane), HCFC-132b (1,2-dichloro-1,1-difluoroethane), HCFC-134a (1,1,1,2-tetrafluoroethane), HCFC-141b (1,1-dichloro-1-fluoroethane, and HCFC-142b (1-chloro-1,1-difluoroethane). With the exception of HCFC-22, which is already in use, the metabolism and toxicity of HCFCs have not been studied in detail. By analogy to chlorinated ethanes, predictions can be made about the possible metabolism of HCFCs, but there are insufficient data available to predict rates of metabolism. Although most HCFCs appear to show low acute toxicity, some HCFCs are mutagenic in the Ames test. Hence, future research on HCFCs should include studies on the in vivo and in vitro metabolism of HCFCs as well as on their toxicity in in vivo and in vitro systems.


Introduction
Chlorofluorocarbons (CFCs) currently in use are perhalogenated alkanes that find extensive use in four major applications: as refrigerants, as blowing agents in the manufacture of foam plastics, as cleaning fluids, and as propellants. These applications exploit the useful physical properties of CFCs, including low boiling points, specific heats and heats of vaporization, high insulating value, low surface tension and viscosity, and high vapor densities. Moreover, CFCs are nonflammable and relatively free ofadverse health effects; reviews about the toxicity of CFCs have been published (1)(2)(3).
There is currently much concern about the potential effects of CFCs on stratospheric ozone (4). Because of their chemical stability, CFCs do not decompose in the troposphere but rise to the stratosphere where ultraviolet radiation catalyzes the release ofchlorine atoms that destroy stratospheric ozone. Stratospheric ozone is an important barrier against ultraviolet radiation and reduces the amount of ultraviolet radiation that may otherwise reach the earth. Depletion of the ozone layer may lead to increased ultraviolet radiation at the earth's surface, and global adverse human health effects associated with increased exposure to ultraviolet radiation, such as an increase in skin cancer and cataract formation, may be seen (5,6). These concerns have prompted a search for CFC replacements that are less damaging to the ozone layer. Commercially important CFCs include the perhalomethanes and ethanes fluorotrichloromethane (CFC-ll),* dichlorodifluoromethane (CFC-12), 1,1,2-trichloro-l,2,2-trifluoroethane (CFC-113), 1,2-dichloro-1,1,2,2-tetrafluoroethane (CFC-114), and 2-chloro-1,1,1,2,2-pentafluoroethane (CFC-115). Some CFCs have estimated atmospheric lifetimes in the tens to hundreds ofyears. Bromine-containing fluorocarbons, the so-called halons, have greater ozone-depleting potential than the CFCs, but are used to a lesser extent. A cap on the production ofCFC-ll, 12,113,114, and 115 and a production freeze on halon-1211, 1301, and 240'2 was mandated by the Montreal Protocol on Substances that Deplete the Ozone Layer and by Environmental Protection Agency regulations that became effective July 1, 1989 (8,9); hence there is a sense of urgency to develop CFC replacements.
Hydrochlorofluorocarbons (HCFCs) have been targeted as replacements for CFCs currently in use because of the greater lability imparted by the presence of C-H bonds, which makes them susceptible to oxidation in the troposphere, and because SThe numeical codes for fluorocarbons have been established by the American Society of Refrigerating Engineers (7). Briefly, the first digit on the right is the number offluorine atoms in the compound, the second digit from the right is one more than the number of hydrogen atoms in the compound, and the third digit from the right, which is omitted when the digit is zero, is one less than the number ofcarbon atoms in the compound; the number ofchlorine atoms is found by subtracting the sum ofthe fluorine and hydrogen atoms from the total number ofatoms that can be connected to the carbon atoms. In the case ofgeometric isomers, the most symmetrical isomer is indicated by the number alone; as isomers become more unsymmetrical, the letters a, b, c, etc., are appended. Symmetry is determined by summing the atomic weights ofthe substituents attached to each carbon and by subtracting the smaller sum from the larger sum; the smaller the difference the more symmetrical the compound. their physical properties are similar to CFCs. CFC replacements cannot, however, be named with certainty. The toxicological, environmental, and basic thermodynamic properties of most CFC replacements have not been fully investigated, and further investigations may make some potentially useful compounds unsuitable for commercial development. Moreover, cost-effective synthetic methods that can be applied to the commercial-scale production ofHCFCs are not available for all compounds. Information from several sources indicates that the HCFCs listed in Table 1, and whose chemical structures are shown in Figure 1, are candidates for development (8)(9)(10)(11)(12)(13)(14)(15).
The objective of this review is to present current knowledge about the metabolism andtoxicity ofHCFCs, to speculate about the possible metabolic fate ofHCFCs and abouthow metabolism may contribute to toxicity, and to indicate some future research needs.

Principles of Metabolism of Halogenated Hydrocarbons
Work conducted over the past two decades has defined the general metabolic pathways of halogenated hydrocarbons and their association with the toxic effects ofhalogenated hydrocarbons. These principles, which should be equally applicable to HCFCs, have been summarized (16,17 Briefly, HCFCs may undergo C-H a-bond oxidation or oxygenation reactions catalyzed by cytochromes P-450; the expected initial metabolites would be geminal halohydrins, which may lose HX to form acyl halides or aldehydes. Such reactions may prove to be the dominant pathways of metabolism of HCFCs and may serve as detoxication or bioactivation reactions; the acyl halides may acylate nucleophilic sites in protein, as has been shown for chlorinated analogs (18,19). In the absence ofoxygen, some HCFCs may undergo cytochrome P-450-catalyzed reduction reactions, as demonstrated for other polyhaloalkanes (20)(21)(22)(23); the expected initial metabolites would be haloalkanes in which one halogen has been replaced by a hydrogen or haloethenes formed by the didehalogenation of haloalkanes.
Haloalkenes formed by the didehalogenation of HCFCs may undergo bioactivation through the cysteine conjugate (3-lyase pathway, which is initiated by glutathione S-conjugate formation and leads to the release of reactive metabolites (24,25). Certain HCFCs or their metabolites may be substrates for the glutathione S-transferases, and, depending on the compound, this may serve as a detoxication or bioactivation reaction. Indeed, recent studies indicate that fluorocarbons may be more subject to nucleophilic attack by sulfur nucleophiles than thought previously (26,27), indicating that the glutathione S-transferase-catalyzed addition of glutathione to fluorocarbons may warrant investigation. Finally, although HCFCs cannot serve as substrates for glucuronylor sulfotransterases, phase I functionalization reactions may yield metabolites that are conjugated by, for example, glucuronide or sulfate ester formation.

Current Knowledge and Speculations about Metabolism
Published information on the metabolism and toxicity of HCFCs that have been designated as replacements for CFCs is summarized below, and possible metabolic routes for HCFCs, based on current knowledge about halocarbon metabolism, are presented. HCFC-22. The metabolism and toxicology of chlorodifluoromethane has been summarized (28). HCFC-22 shows low acute toxicity in various animal species; a concentration of 20% HCFC-22 is not lethal to any species tested (see Litchfield and Longstaff (28) for references to the original literature). HCFC-22 undergoes little metabolism in vivo; less than 0.03 % of an inhaled dose of 500 ppm [14C]or [36C1]HCFC-22 was metabolized (28). Pharmacokinetic studies in rats demonstrated no detectable in vivo metabolism of HCFC-22 (29). In vitro studies with [36CI]HCFC-22 did not show the release of detectable chloride ion, indicating that little metabolism takes place.
HCFC-22 (50,000 ppm for 5 hr/day for 8 weeks) does not affect male fertility in the rat nor is there evidence of dominant lethality (30). Exposure of rats to 50,000 ppm HCFC-22 (6 hr/day on days 6 to 15 ofpregnancy) produced a low incidence of microphthalmia and anophthalmia in rats; this effect was not observed in rabbits (28).
HCFC-22 is mutagenic in the Ames test in Salmonella typhimurium strains TA1535 and TA100 (31); a positive response was obtained in both the absence and presence of S-9 fractions from Aroclor 1254-treated rats. Similar results were obtained in other experiments (28). HCFC-22 is not mutagenic in Schizosaccharomyces pombe or in Saccharomyces cerevsiae or in a host-mediated assay with S. pombe orS. cerevisiae (32). Similarly, HCFC-22 was negative in an unscheduled DNA synthesis assay inthe human heteroploid EUE cell line and in V-79 Chinese hamster cells (32) or in a CHO cell line (28,32). The ability of HCFC-22 to induce chromosome damage in rat bone marrow cells was studied (28); an apparent increase in chromosomal damage was seen at the lowest exposure level but not at two higher exposure levels. Dominant lethal assays ofHCFC-22 in mice have been conducted (28); although significant ects were seen, the effects were not systematically related to duration ofexposure or to dose, and the compound produced its effect in the premeiotic phase.
Recent studies report that no treatment-related effects were seen in Sprague-Dawley rats or Swiss mice exposed to HCFC-22 by inhalation (1,000 or 5,000 ppm for 4 hr/day, 5 days/week for 104 weeks) (33). HCFC-22 given by gavage (300 mg/kg, 5 days/week for 52 weeks) did not induce tumors during the 125-week observation period (34,35).
HCFC-23, I2, and DI The lowest observed lethal concentration of HCFC-123 is 14 pph/4 min in the mouse (36); apparently data on the mammalian toxicity ofHCFC-124 and 125 have not been reported. HCFC-123, 124, and 125 are not mutagenic in the Ames test with S. typhimurium TA1535 or TA100 as the test strains (35).
2-Chloro-2,2-difluoroethanol undergoes conjugation to form a glucuronide and a sulfate ester. No detectable alkylation of hepatic proteins was seen (46). These results are consistent with the generalization that protein acylation may be associated with the metabolism of geminal dihaloethanes to acyl halides; HCFC-132b should not be metabolized to an acyl halide and would not, therefore, be expected to acylate proteins. The toxicity of2-chloro-2,2-difluoroethanol has not been investigated, but the fluorinated analog 2,2,2-trifluoroethanol is toxic (47).
HCFC-142b. The lowest observed lethal concentration of HCFC-142b in the rat is 50 pph/30 min (60). HCFC-142b undergoes dechlorination (0.6%) when incubated with rat hepatic microsomes (44). HCFC-142b is mutagenic in the Ames test with S. typhimurium TA1535 or TA100 as the test organisms (35) and in a BHK21 cell-transformation assay (34,35). Rats exposed to HCFC-142b by inhalation (100, 10,000, or 20,000 ppm for 6 hr/day, 5 days/week for 13 or 15 weeks) did not show compound-related effects in a bone marrow cytogenetic assay or in a dominant lethal assay (61). No treatment-related effects were observed in a 90-day study of rats and dogs exposed by inhalation to HCFC-142b (1,000 or 10,000 ppm for 6 hr/day, 5 days/week for 90 days) (62) or in a 104-week study ofrats exposed by inhalation to HCFC-142b (1,000, 10,000, or 20,000 ppm for 6 hr/day, 5 days/week for 104 weeks) (61).
Apart from the observation that HCFC-142b undergoes dechlorination (44), the metabolism of HCFC-142b has not been investigated. The cytochrome P-450-dependent oxidative metabolism of HCFC-142b may yield 2-chloro-2,2-difluoroethanol as the initial metabolite, and chlorodifluoroacetaldehyde cI Fc H x >t°H X and chlorodifluoroacetic acid may be formed by the further metabolism ofthe trihaloethanol (Fig. 4). It should be noted that the toxicity of alcoholic metabolites may warrant investigation because of the observation that the analog 2,2,2-trifluoroethanol is toxic (47).

Future Research Needs
Toxicology With the exception of HCFC-22, whose toxicity has already been extensively investigated, there are limited published toxicological data available about the HCFCs that have been proposed as replacements for CFCs. Hence in vivo toxicity studies on HCFCs are needed; these studies should include shortand long-term toxicity studies, including, at least, developmental and reproductive toxicity, neurotoxicity, carcinogenic potential, mutagenicity, and immunotoxicity. The objective ofthese studies should be to identify target organs that are affected and that may signal potential human health hazards. Toxicity studies in in vitro systems (freshly isolated cells and cultured cell lines) should be conducted in parallel with in vivo toxicity studies, and the choice of cells or cell lines to be used should reflect the target organs observed in in vivo studies. These in vitro studies should also include mutagenicity studies in submammalian test systems. Finally, in vitro systems may find utility in exploring possible mechanisms of toxicity.
As noted above, recent studies indicate that metabolites of HCFC-123 acylate hepatic proteins; because modification of cellular macromolecules may be associated with cytotoxicity, future studies should examine fully the extent and nature of the interactions of HCFC metabolites with cellular macromolecules. In addition, preliminary studies also indicate that 2-chloro-2,2-difluoroethanol is a metabolite of HCFC-132b; because the analog 2,2,2-trifluoroethanol is toxic (47), future studies should address the potential toxicity of2,2,2-trihaloethanol metabolites of HCFCs.
Strategies for toxicity testing CFC replacements have been proposed (63). Also, toxicity studies are currently underway; the multi-industry Program for Alternative Fluorocarbon Toxicity Testing (PAFTT) is developing toxicity profiles for several HCFCs (8).

Metabolism
Limited metabolic data are available for most HCFCs proposed as replacements for CFCs. Hence metabolism studies should include investigations on the pharmacokinetics of uptake and elimination ofHCFCs as well as investigations on the in vivo metabolic fate ofHCFCs, including identification ofmetabolites, their possible interaction with cellular constituents, and the mechanisms and routes of clearance of metabolites. In vivo studies on the metabolism of HCFCs may exploit the utility of 19F-NMR spectrometry in identifying free and tissue-bound metabolites, as noted above (37,46,59). The objective ofin vitro metabolism studies, particularly in identified target organs or derived cell systems, should be to identify the enzymes catalyzing the metabolism of HCFCs, possible associations between metabolism and toxicity , and interactions ofHCFC metabolites with cellular macromolecules. Ultimately, studies on HCFC metabolism should be conducted with purified enzymes or enzyme systems so that reaction mechanisms can be explored in detail. Because the HCFCs constitute a group ofclosely related analogs, computational analysis of data from structure-metabolism studies may be used to improve predictability of the metabolic fate ofHCFCs, interactions ofHCFC metabolites with cellular macromolecules, and relationships between HCFC metabolism and toxicity.
Although there is a sense of urgency to develop HCFCs as replacements for CFCs because ofmandated reductions in CFC use, it is important to note that CFC replacements will likely enjoy widespread commercial use, which may be associated with human exposure to HCFCs. Therefore, careful toxicological and metabolic evaluation of HCFCs should not be compromised to allow early introduction of HCFCs into commerce.
The author thanks Michael J. Olson, General Motors Research Laboratories, for providing preprints ofpublications and Sandra E. Morgan for preparing the manuscript. Research on HCFC metabolism conducted in the author's laboratory was supported in part by National Institute of Environmental Health Sciences grant ES05407.