Phytochemistry, zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Vol. 20, No. 4, pp. 5% 567, 1981. 0031-9422/81/040553-15 $05.00/O Printed in Great Britam. 0 1981 Pergamon Press Ltd. REVIEW zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQ CA RBON ISOTOPE FRACTIONATION MARION Departments of Chemistry and Biochemistry, IN PLANTS H. O’LEARY University of Wisconsin, (&wised received 1 September Key Word Index-Isotope fractionation; carboxylation; respiration; C, plants: C, plants; CAM plants; isotope discrimination. Madison, WI 53706, U.S.A. 1980) metabolites; carbon fixation; carbon dioxide; Abstract-Plants with the C?, C, and crassulacean acid metabolism (CAM) photosynthetic pathways show characteristically different discriminations against 13C during photosynthesis. For each photosynthetic type, no more than slight variations are observed within or among species. CAM plants show large variations in isotope fractionation with temperature, but other plants do not. Different plant organs, subcellular fractions and metabolites can show widely varying isotopic compositions. The isotopic composition of respired carbon is often different from that of plant carbon, but it is not currently possible to describe this effect in detail. The principal components which will affect the overall isotope discrimination during photosynthesis are diffusion of CO,, interconversion of CO, and HCO;, incorporation of CO, by phosphoenolpyruvate carboxylase or ribulose bisphosphate carboxylase, and respiration. The isotope fractionations associated with these processes are summarized. Mathematical models are presented which permit prediction of the overall isotope discrimination in terms of these components. These models also permit a correlation of isotope fractionations with internal CO, concentrations. Analysis of existing data in terms of these models reveals that CO, incorporation in C, plants is limited principally by ribulose bisphosphate carboxylase, but CO, diffusion also contributes. In C, plants, carbon fixation is principally limited by the rate of CO, diffusion into the leaf. There is probably a small fractionation in C, plants due to ribulose bisphosphate carboxylase. In plant physiology and geochemistry applications this is more commonly expressed as a 613C value, in units per mil (“/,J: zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPO INTRODUCTION Atmospheric carbon dioxide contains about 1.1% of the heavier carbon isotope 13C and 98.9% of the lighter isotope “C. Plants discriminate against 13C during photosynthesis in ways which reflect plant metabolism and environment [l-8]. The purpose of this review is to summarize the data concerning this discrimination and to provide a chemical and physiological basis for understanding the fractionation in terms of the various components of plant metabolism. This review is organized as follows: I. Definition of isotopic composition and isotope fractionation. II. Review of reported isotopic compositions and fractionations in plants, including effects of metabolic and environmental factors. III. Theoretical treatment of isotope fractionations, including measured fractionations for component processes and mathematical models for the fractionation. IV. Interpretation of isotope studies in plants. ~13qo/,,) R (sample) R (standard) -1 xlooo. (2) I Absolute isotope ratios are troublesome to obtain, and for most purposes it is adequate to give ii’ 3C values relative to some standard. The standard in general use is PDB (belemnite from the Pee Dee Formation in South Carolina; [1,9]). The isotope ratio for PDB is 13C/‘2C = 0.01124 [9]. Organic matter is invariably depleted in 13C compared to PDB, so 613C values of organic materials are negative. A less negative figure means richer in 13C, or ‘heavier’. Standards for calibration of isotope-ratio mass spectrometers can be obtained from the U.S. National Bureau of Standards. In the absence of industrial activity, atmospheric CO, has a 613C value between -6.4 and - 7.0”/,, [9,1 l-131. However, in industrial areas this value may be significantly more negative because of combustion of coal and petroleum, with 613C values near -3O” /,, [14]. Although the 613C value for atmospheric CO, is believed to have remained approximately constant over geologic time [14], increasing combustion of fossil fuels is causing a shift toward slightly more negative values. The S13C value for atmospheric CO, decreased by approximately 0.6”/,, in the period 1956-1978 [13]. CO, in greenhouses and in dense forests may be significantly more negative than -- 7”/,, because of the contribution of respired CO,. Leaves obtained near the ground in dense forests are several per mil more negative than ordinary C, plants or leaves from higher elevations in the same forests [14a, 14b]. Expression of isotopic composition and isotope fractionation Isotopic compositions are measured by means of a specially designed mass spectrometer equipped with two collectors, two amplifiers and a bridging circuit. The output from the mass spectrometer is an isotope ratio compared to some standard. After minor corrections for instrumental effects and for the presence of I70 in the sample, the ratio is converted to the carbon-13 abundance ratio R [9, lo]: R = ‘3C0,/12C0,. = (1) 553 554 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA M. H. O’LEARY In an attempt to verify their model, Park and Epstein measured the isotopic composition of ‘internal CO,’ obtained by treatment of leaves with acid. However, the origin of the CO, evolved in these experiments and the Discrimination = _.__ _~~~ ~~~ ~~~ _ ~ ~, relevance of its isotopic composition to the internal CO, 1 + 613C(source) + 1000 pool are both unknown. These experiments do not appear (3) to have been repeated. Errors in the literature have often resulted from the failure Park and Epstein also studied erects of light intensity of investigators to assign the proper algebraic sign to their and CO, concentration. They observed that CO, respired discrimination values. in the dark is up to 8 “/,, heavier than whole plant carbon. Chemists investigating isotope fractionations most They subsequently extended these observations and also often express their results in terms of ‘isotope effects’, zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJI reported that lipids are enriched in “C by as much as for reactions of the k**/k13, the ratio of rate constants 8”/,, compared to the whole plant [22]. respective isotopic substances. If the source is a C, Und c, sufficiently large reservoir that is not appreciably depleted by product formation, then The plants studied by Craig [ 17,181 and by Park and Epstein [21,22] were C, plants, which incorporate COZ from the atmosphere by carboxyiation of ribulose Isotope &ect = k1*/k13 -_ R (source)jR (product). (4) bisphosphate. In the mid-1960s, the C, pathway was Thus the discrimination factor is given by discovered [23,24]. C, plants incorporate COz by the carboxylation of phosphoenolpyruvate. The carboxylk’3/k12). Discrimination = 1000 x (1 - zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA (5) ation product is transported from the outer layer of In most chemical reactions the lighter isotopic species photosynthetic cells (mesophyll cells) to the inner layer reacts more rapidly than the heavier isotopic species. (the bundle sheath), where decarboxylation and refixation When this occurs, the discrimination has a positive value by ribulose bisphosphate carboxylase occur. Isotope studies demonstrated that these plants show less negative and the isotope effect is greater than unity. Carbon fS”C values than C, plants [25-271. isotope discriminations in enzymatic reactions are commonly in the range O-20” /,,, though values as large as This dill’erence in isotopic composition has become one 60”/,, are occasionally observed [ 15,161. of the standard methods by which C, plants can be wo uld like to point out two In concluding this section I zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA distinguished from C, plants, and a great variety of plant difficulties with the expression of isotopic compositions types have been studied [8,28-371. Troughton et cd. [37] and isotope fractionations. As noted, errors in algebraic reported a mean 6’ 3C value for C, plants of - 13.5k 1.5; sign have frequently been made as a result of the for C, plants the value is -28.1 f 2.5. unfortunate fact that ii13C values are almost invariably The distinction between C, and C, pathways of negative. Second, a number with units of ‘per mil’ (“ /,,) photosynthesis has been particularly useful in studies of may be of either of two types: it may be an isotopic c‘,3,C1 hybrid .-1rri/‘/c~\-specleh. These hybrids show a composition value (cf. equation 2) relative to some variety of leaf anatomies characteristic of intermediate standard (commonly PDB); alternatively, it may be an stages between C, and C,, but the (j13C values obtained isotope discrimination (cf. equation 3) reflecting a are invariably indicative of the C, pathway [3%40]. difference in isotopic composition between source and CA M plunts product. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Isotope discrimination is generally expressed as a difference in the 613C value between source and product (1 ): 613C(source) - cil”C (product) ISOTOPIC CO,Mf’OSITIONS OF PLANTS Early observations concerning carbon isotopic compositions of plant materials were made by Craig [17. 181,Wickman [19] and Baertschi [20]. Craig, in the course of an extensive survey of carbon isotopic compositions of natural materials, observed that most plant materials had a relatively constant rS’3C value near - 27”/,, [17]. He observed that there seemed to be no important species or geographical effects (interestingly, he reported one grass with a 613C value of - 12”!,,, in retrospect clearly a C, plant). In a subsequent paper, he discussed his results in more detail and speculated on the probable source of the discrimination. He considered possible contributions from environment, CO, diffusion, chemical absorption of CO, and respiration [18]. Craig’s description of the carbon isotope discrimination was confirmed and extended by Park and Epstein [21], who showed that the key carboxylating enzyme in plants, ribulose bisphosphate carboxylase, discriminates against 13C0,. They suggested that the primary cause of the difference in isotopic composition between plants and atmospheric COZ is the isotope discrimination by this enzyme. Succulent plants which exhibit crassulacean acid metabolism (CAM) may either fix atmospheric carbon in the manner of C, plants (by use of ribulose bisphosphate carboxylase) or else in a time-separated C,-like sequence in which phosphoenolpyruvate is carboxylated, then reduced, in the dark, forming malate, which accumulates in the vacuole. In the following light period. this malate is decarboxylated and the CO2 thus formed is fixed by ribulose bisphosphate carboxylase [41,42]. Following earlier reports that CAM plants show widely varying carbon isotope ratios [26], several groups nearly simultaneously concluded that the isotopic composition of CAM plants reflects operation of the different carboxylation options [43- 461. Osmond et al. [43] correlated isotopic compositions of KN/u/v/~o(J tltri,~,‘r,no,lriLIrlcileaves with results of pulsechase experiments and concluded that 8’ “C values are a useful indicator of photosynthetic pathway, with ii’% values becoming si_pnificantly more positive as the plant shifts from predommnntlq light to predominantly dark CO, lixation. Bender rt ctl. [44] surveyed a wide variety of CAM plants and used correlation with environmental variables to reach the same conclusion. Lerman and Queiroz [4.5] used variable photoperiods with young leaves of K. h/os.sfXliunn to manipulate carbon fixation Carbon isotope fractionation in plants 555 pathways and observed corresponding changes in 613C hemprichii, Halophila spinulosa and Thalassia testudinum values. Medina and Troughton [46] observed that for a observed are [66- 681. The small isotope fractionations variety of Bromeliaceae species, fi13C values near - 13”/,, presumably due to slow diffusion of CO, in water. Under were associated with dark CO2 fixation and values near these conditions, as with algae under certain conditions - 25”/,, were associated with the absence of dark fixation. [3], the CO, supply becomes limiting and the expected C, M esembryanSimilar studies have been performed for zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGF isotope fractionation is not expressed. themum species [47-491. Effect of fertilization Osmond et al. [50,51] studied K. duigremontiana grown under various temperature and light regimes. The nutritional status of a plant might affect the observed isotope discrimination. For timothy grass Isotopic compositions showed the expected correlations (Phleum pratense L.), well-nourished plants showed more with gas exchange results; under conditions where carbon positive 613C values than did plants deficient in nitrogen fixation occurs principally in the dark, S13C values and/or potassium. The difference was about 2”/,, for become less negative. The same plant species was studied plants grown under a 13’/7O temperature regime and by Lerman et al. [52], who observed that younger leaves diminished to less than lo/,, for a 32”/26” regime [69]. are more C,-like. In the same study were reported isotopic Although sodium is a required micronutrient in C, plants, compositions for a soluble fraction (said to be mostly no sodium dependency was found in Kochia childsii or malate), starch and insoluble material. Further fractionaChloris barbata [69a]. tion of materials from the same species was reported by Deleens and Garnier-Dardart [53]. The most systematic Plant organs approach to environmental variables and their effect on Most measurements of S13C values are made on leaves. 613C values in K. daigremontiana is that of Osmond et al. It is important that investigators specify whether this is [54], who measured the time course of 613C values for the case because there may be some variation in isotopic plants grown under various water, light and temperature composition for various organs. In tomato, stems and roots regimes. are a few tenths per mil less negative than leaves [21]. Thus it appears that when CAM plants function strictly Seeds are generally more positive (up to lO” /,,) than in the C3 mode they have fi13C values near those of C, leaves [35,65]. Potato tubers are about 2”/,, more plants, or -2l” /,,; a CAM plant engaging in only dark positive than leaves [35]. Epidermal and mesophyll fixation should have a ci’ 3C value near - 13”/,,. This view tissues in C, plants differ by a few tenths per mil [70]. has been further strengthened by Nalborczyk et al. [55], who showed that when K. daigremontiana plants are Eflkcts of temperature furnished with CO, only in the light (thus eliminating Plants could be useful indicators of past climate if dark carboxylation) they show S13C values of - 26”/,,. If carbon isotope ratios varied in a systematic way with with the same plants are supplied with CO, only during the growth temperature. A number of attempts have been dark period, the 613C value is - ll” /,,. made to correlate S13C variations of tree rings with Isotope fractionations have been used to study the climatic variations, but with only limited success [4]. occurrence of the CAM photosynthetic mode in relation Troughton [35] reported that 613C values become to various environmental effects [56-621. Osmond et al. slightly more negative (by up to 2’jo0) with increasing [63] have used 613C values to identify CAM species temperature in several C, and C, plants. Similar results among alpine plants. Winter [64] has used 613C values to have been obtained by Smith et al. [71,72] and by Bender identify CAM species among Madagascar succulents. and Berge [69]. Other studies have failed to find such an effect [36,73]. If the temperature effect is real, it must be Interspecies variations very small. The largest factor affecting carbon isotopic composiCAM plants can show large variations in 813C values tions is the existence of the C,, C, and CAM with temperature as a result of changes in the balance photosynthetic options. Other environmental variables between dark fixation and light fixation. In K. can sometimes affect 613C values (see below). It is not daigremontiana, an increase in day temperature from 17 to currently possible to say whether, in the absence of 31” results in a decrease of about 8”/,, in 613C values [54]. environmental differences, various species of, for example, The temperature variation of isotopic composition in C, plants will show different 613C values. Variations in micro-organisms presumably results from the change in the range of 2-5” /,, are still within the realm of possibility. CO, availability with temperature [74-761. Isotopic variations in micro-organisms can also reflect different Intraspecies variations CO, fixation options [77]. Other studies of microLowden [65] failed to find significant differences in organisms have been summarized by Benedict [3]. 613C values for several strains of Zea mays. Data of Troughton [35] on several C, and C, plants suggest that variations of up to 3”/,, might be observed for various strains of the same species. No extensive study of this type appears to have been undertaken. Aquatic plants The ()_13Cvalues of various aquatic plants are often significantly more positive than those of terrestrial C, plants and have sometimes been interpreted as indicating the operation of the C, photosynthetic pathway. Recent pulse-chase studies indicate that in most cases the C, pathway is operating. Species studied include Thalassia EfSect of salinity Smith and Epstein [78] measured 6°C values for a variety ofC, and C, plants from salt marshes but made no correlation with salinity. Card et al. [79] showed no effect of salinity on 613C values for Zea muys, Gomphrenn globosa, Raphanus sativus, Triticum aestivum, Salicornia virginica and Spartina foliosa. In a more extensive study, Guy et al. [80] showed that for the halophytes Salicornia rubra and Puccinellia nuttalliana, europaea subsp. increasing salt concentrations resulted in decreased isotope discrimination both in field samples and in 556 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA M. H. O’LEARY growth chamber samples. minations was IO” /,,. EffectofCO, The total concentrationoncarbon range of discri- isotopefiachonation Park and Epstein [21] reported that for tomato the carbon isotope fractionation increases slightly with increasing CO, concentrations, but the total range of fractionations reported was only about 2”j,,. Carbon isotope fractionations in algae subjected to various CO, regimes have been reviewed by Benedict [3]. At low CO, levels, blue-green algae show only a very small isotope fractionation [74,81-831. At low cell densities and CO, concentrations above 0.5%, a maximum fractionation of about 24”/,, was obtained [74]. Fractionation was reduced at higher temperatures [74]. The concentration effect is principally a reflection of the availability of CO,. When the CO, level is high the cells have an opportunity to discriminate between “C and 13C, and a large fractionation is observed. When the CO, level is low, growth is largely limited by CO, availability and the cells use all available CO,, independent of its isotopic nature. Park and Epstein [21] showed variation in 613C values of tomato intensity. Smith et al. [71] observed variations in whole plant S’ 3C values for C, plants. Acaciafamesiana values 5”’/00 with increasing light intensity, showed a 2”/,, variation in the other less than a 2”/,, plants with light seemingly random with light intensity decreased by about but Festuca rubra direction. Solsent deuterium @ct Uphaus and Katz [84] showed that when Nicotiana in water containing tabacum plants were cultured increased amounts of deuterium, the carbon isotope fractionation was decreased. The fractionation was approximately linear with deuterium content, being 21”/,, at 0 y< deuterium and lY/,, at 60 y;, deuterium. The plants would not grow above this concentration of deuterium. The authors pointed out that this change in fractionation was most likely a consequence of widespread changes in the cell, and they made no attempt to interpret the results quantitatively. Replacement of hydrogen by deuterium in cultures of Chlorella rulgaris [84] and C. pyrenoidosa [81] results in similar decreases in carbon isotope fractionation. Respired carbon The isotopic composition of a plant is controlled by the isotopic composition of the CO, source, the isotope, fractionation accompanying CO, incorporation, and the isotopic composition and quantity of the CO, lost through respiratory processes. A number of measurements of d13C values for respired carbon have been made by trapping the CO, released by a plant in a CO,-free atmosphere and measuring its isotopic composition. It should be noted that not all CO, which is formed as a result ofrespiration and other CO,-forming processes within the plant is actually released to the environment; some of this CO, is refixed. Refixation can fractionate carbon isotopes. Consequently, the isotopic composition measured for ‘respired carbon’ may differ from that of total carbon formed by respiratory processes. The isotopic composition of CO, released from plants in the dark differs only slightly from that of the whole plant. Thus, CO, released by dark respiration in tomato was 2-Y/,, more positive than the whole plant [22]. For Triticum aesticum, released carbon was 5”/,, more positive than the leaf [37]. Carbon dioxide released by seedlings of wheat, radish or pea was about lo/,, more negative than the seedling, although in the case of squash the released CO, was lo/,, more positive than the whole plant [85]. For tobacco, the released carbon was lo/,, more positive than the starch pool (which should have nearly the same isotopic composition as the leaves) [S5a]. Carbon dioxide released by Pinus radiata was 40ion more negative than the whole plant 1371. In four species of C, plants, released carbon was 0 7” o. more ncpatilc than the ~liolc plant [37,SS]. There appears to be a greater difference in d13C values between whole leaves and carbon released in the light. For Gossypium hirsutum and Triticum clestilum (both C, plants) the released carbon was IO and 12”/,,, respectively, more positive than the leaf [37]. For Paspalum dilatatum and Zetr nzaq’s (both C, plants) the released carbon was 3 and 6”i,,. respectively, more ilegutice than the leaf [37]. The large differences observed in C, plants may be due to the occurrence of partial refixation of respired CO,, leading to a discrimination against ’ 3C and the release of relatively positive carbon to the atmosphere. The measurement of the isotopic composition of released CO, in plants suffers from a kind of botanical ‘uncertainty principle’. In order to measure the isotopic composition of released CO,, it is necessary to subject the plant to a CO,-free atmosphere for a period of time. However, stomata1 opening is subject to control by CO, levels, so the extent of refixation of respired CO, will be different in the CO,-free environment than it is in a normal environment. The extent of refixation will also be affected by the fact that there is no atmospheric COz to compete with respired CO, for ribulosc bisphosphate carboxylase. Thus, we cannot be sure that the isotopic composition of ‘respired CO,’ in a CO,-free environment is the same as would be obtained in a normal atmosphere. Photorespiration in C, plants [85b] probably also affects isotopic compositions. In the case of ‘4triplex patula (a C, species) a change from 20”. oxygen to 4”” oxygen caused the whole plant ci’ 3C value to become more positive by 2-4” /,, [86]. A shift in the opposite direction was observed in soybean leaves by Smith et al. [71]. No oxygen effect was observed in the C, plant Atriplex rosea [86]. At least some of the variation in the isotopic composition of respired carbon may be due to whether the source of CO, for respiration is carbohydrate or lipid [85,87]. Plant lipids are significantly more negative than other components (see below). Although decarboxylations often show significant discriminations against 13C [ 151, the extent to which this discrimination is expressed during respiration is not known. Values of (Y3C in lichens showed seasonal variations which were suggested to be due to enrichment of *‘C during the winter (when carbon is being added through photosynthesis) and depletion of “C during the summer (when carbon is being lost through respiration) [88 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONM I. Few isotopic compositions have been measured for respired carbon in CAM plants. Nalborclyk et (11. [55] Carbon isotope fractionation in plants 557 K. daigremontiana the CO, released in reported that for zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA source C02. The difference between these results and the light was “even less negative than that normally those of Deleens and Garnier-Dardart [53] may reflect obtained for C, plants” , but no details were given. This either contributions of other acidic compounds in the observation is not entirely surprising. The amount of CO, latter case or variable contributions of respired carbon in released under these conditions is probably small the two studies. In the same study, O’Leary and Osmond compared to the total amount being formed by [99] observed that malate isolated from Bryophyllum decarboxylation of malate [55,89]. The recycling as the source tub$orum had the same isotopiccomposition mechanism in CAM plants can allow the internal CO, CO, and was about 5”/,, heavier than the whole leaf. concentration in the light to reach quite high levels Several constituents of a C, plant (cotton) and a C, [90,91], presumably with relatively modest CO, leakage. plant (sorghum) were studied by Whelan et al. [92]. In Under these conditions the internal CO, is principally both cases the lipids were about 5” ,1,, lighter than the used to form phosphoglyceric acid under the influence of whole leaf. In the case of cotton, the amino acid fraction ribulose bisphosphate carboxylase. The latter enzyme was 6”/,, lighter than the whole leaf. Other metabolites shows a large (20-4O” /,,) discrimination against ‘%ZO, were generally within 2”/,, of the whole leaf value. The (see below) and as a result the internal CO, pool at same study included values for a number of individual steady-state may be substantially enriched in ‘%O,. It metabolites, including aspartate, glutamate, alanine, was the leakage of this material which was presumably malate and glucose. Significantly, malate in the C_, plant observed by Nalborczyk et al. [%I. was more positive than other metabolites or the whole leaf and only about 2”/,, more negative than the source co,. Isotopic compositions of metabolites Interpretation of plant isotopic compositions would be Following the discovery that plants show systematic most simple if metabolites of a particular type were of differences in isotopic composition from the source CO?, similar isotopic compositions. However, this does not a number of investigators have studied the isotoptc appear to be the case. Instead, a variety of studies have compositions of various subfractions of the plant. These suggested that different metabolites of a particular class studies span anatomical variations (leaf vs stem vs root), have quite different isotopic compositions. Degens et al. metabolic fractions (starch vs lipid, etc.), individual [93] reported that different carbohydrates show different compounds (glutamate vs aspartate vs malate, etc.), and, isotopic compositions in marine plankton. Whelan et al. in a few cases, individual carbon atoms (carboxyl groups [92] showed that glutamate is several per mil different of various acids, etc.). from aspartate in both C, and C, plants. Malic acid and Park and Epstein [21,22] reported that lipids in citric acid differ by about 4”/,, in cotton [92]. Crystalline tomato are enriched in “C compared to total plant oxalic acid secreted by several cactus species was found to organic matter. Similar observations have been made for have a 613C value within lo/,, of the atmosphere, about cotton and sorghum [92], marine plankton [93], K. 5”/,, more positive than plant fibers. Interestingly, oxalic daigremontiana [53], potato tuber [94,95], other plants acid from spinach was also significantly more positive than [22,96], and micro-organisms [81]. The difference the whole plant [lOO]. between whole leaf carbon and lipid is often near 5O/,,, but The most thorough and convincing case for large may be as large as lo’/,,. This depletion may be caused by variations among metabolites of a given class comes from the isotope fractionation associated with decarboxylation the work of Abelson and Hoering [81] on photosynthetic of pyruvic acid [97]. micro-organisms. Individual amino acids were purified Other gross differences among metabolites are less and their isotopic compositions were measured. In all striking. Lerman et al. [52] conducted a partial cases the amino acids were more negative than the source separation of components from K. daigremontiana. carbon, but the 613C values for individual amino acids Aqueous extracts (presumed to be mostly malate) fell near varied over a wide range. In the case of Chlorella - 12” /,,, whereas starch from the same plants was near pyrenoidosa, for example, some typical values are glutamic - 15°/,,0, and the insoluble residue was - 16 to - 18” /,,. acid - 18.7, aspartic acid -6.6, serine - 5.7, alanine Systematic differences were noted with leaf age. In - 10.3, leucine - 22.7, tyrosine - 19.8. Thus, any subsequent work from the same laboratory, Deleens and approach to isotopic fractionations based on pools of Garnier-Dardart [53] conducted extensive chromatoamino acids would be doomed to failure. graphic fractionations of material from the same source. However, the situation is even more complicated than As always, the lipid fraction was significantly lighter than this; isotopic compositions are far from uniform even for other fractions. For a mature leaf the isotopic all of the carbons of a particular amino acid. Abelson and composition of starch was virtually the same as that of the Hoering [81] used enzymatic and chemical methods to whole leaf. Cellulose was about 3”/,, more negative than decarboxylate their amino acids, thus allowing them to the whole leaf. Significantly, the acidic fraction which had determine the isotopic composition of the carboxyl accumulated at the end of the dark period, presumed to be carbons and then (by difference) all remaining carbons. mostly malate, was about l”/0, lighter than the COZ source. Large differences were often observed. For example, in Values for a younger leaf were, in some cases, a few per mil Chlorella pyrenoidosa the carboxyl carbon of glutamic lighter, reflecting a greater contribution of C3 photoacid is heavier than the remaining carbons by 12”/,,; for synthesis in the younger leaf. Similar results have been threonine the difference is 20”/,,; for lysine it is reported for Kalnnchoe blossfeldiana cv Tom Thumb [98]. 2301,~. Similar studies do not appear to have been conducted O’Leary and Osmond [99] have recently purified malate from K. daigremontiana at the end of the light with higher plants. In the recent work of O’Leary and period and at the end of the dark period. The evening Osmond [99] carbon-4 of malic acid was found to be l-4” /,, heavier than the other three carbons in material sample was about 2”/,, lighter than the source CO,. The morning sample ofmalate was about lo/,, lighter than the purified from K. daigremontiana and B. tubjjlorum. M. H. O’LEARY 558 Summary of jixtors afecting ? iL3C calues It is very clear that a distinction can be made between C, plants and C, plants on the basis of their 6i3C values. The operation of the various photosynthetic options in CAM plants can also be clearly distinguished. But how many of the other variables discussed in the preceding sections can give rise to significant variations in 613C values? A caution must first be raised with regard to the experimental data. Most investigators appear to assume that the 613C value of their source CO, is -7” /,,, but measurements of the source are seldom reported. This assumption may not be correct for plants grown in growth chambers, in greenhouses or at sites close to industrial or urban CO, sources. Variations in di3C for different plants at a single site, rather than absolute values, provide a more adequate experimental approach. Further, most literature reports of dr3C values fail to provide any estimate of error or even any statement of the number of repetitions of a particular measurement. Modern isotope-ratio mass spectrometers are capable of measuring d13C values for CO, with a reproducibility substantially better than &O.l” /,,. Although under ideal circumstances repeated measurements on plant materials may yield errors in the range of &0.3” /,, or better, it is likely that errors in most reported values should be considered to be closer to & l”/,,. On the other hand, it is clear from the preceding discussion that there are small environmental and intrinsic variations in 6i” C values. Carefully controlled experimentation is required for unambiguous delineation of these effects. Because of the possibility for internal comparison, the variations among various plant fractions and individual metabolites are more clearly established. The difference between lipids and other materials is particularly clear. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONM often near 20”/,, [15], but the thermodynamic fractionation in the same process is only about 3”/,, [loll. Our models for isotope fractionation in plants are based on the cumulative fractionations associated with various individual steps (such as stomata1 diffusion, CO,-HCO; equilibration, carboxylation, etc.). Conversion of material from one state to the next may result in either a kinetic or a thermodynamic fractionation, depending on whether the rate of interconversion of the two states is slow or fast compared to the rates of other steps in the sequence. This distinction is extremely important for the construction of models because thermodynamic fractionations are additive, whereas kinetic ones are not. Treatment of isotope fractionations in processes which are near, but not at, isotopic equilibrium is a key factor in the discussion to follow. Fortunately, there is a simple relationship between kinetic and thermodynamic isotope fractionations. For a given transformation of the type AgB, the equilibrium (6) is defined as zyxwvutsrqponmlkjihg constant Kl2 = =I I’zsl, P2Al in which the superscripts refer to the appropriately substituted isotopic species and brackets indicate concentrations. The equilibrium isotope effect for such a process is given by (8) In terms of discrimination to factors, this is very nearly equal D,, = D, - Dz. THEORETICAL TREATMENT OF CARBON FRACTIONATION To date, attempts to explain carbon isotope fractionations in plants have been focused on the carbon-fixing enzymes ribulose bisphosphate carboxylase and phosphoenolpyruvate carboxylase. However, many chemical and some physical processes are capable of fractionating isotopes and it is likely that several of those processes may contribute to the overall fractionation. In this section I give estimates of the magnitudes of the fractionations expected in these various processes. I then describe the models which can be used in conjunction with these fractionations in order to predict actual 6i ‘C values. The mathematical treatment requires use of isotope effects (cf. equation 4), whereas most plant physiologists are more familiar with isotope discriminations (equation 3) so I will give the fractions in both forms. Thermodynamics (9) ISOTOPE L’Skinetic fractions It is important at the outset to note that there are two kinds of isotope fractionations: thermodynamic and kinetic. Thermodynamic fractionations are isotope fractionations occurring for processes which are at equilibrium. Kinetic fractionations are fractionations occurring because different isotopic species are transformed at different rates. These two types of fractionations are often quite different. For example, the kinetic fractionation associated with the enzymatic conversion of a carboxylic acid into CO1 and some other material is Thus, if we know the equilibrium isotope effect and the kinetic isotope effect in one direction, we can calculate the kinetic isotope effect in the other direction. When should we use kinetic fractionations and when should we use equilibrium fractionations? It is permissible to use the equilibrium fractionation orz/y when a particular transformation is at equilibrium; that is, the rate of the transformation must be rapid compared to the rates of steps preceding and following it. The situation is exacerbated by the fact that isotopic equilibrium is attained many times more slowly than chemical equilibrium. Thus, when in doubt, it is safer to use a kinetic approach. “C e$i%cts r\ 14C &ects Isotope fractionations involving assumed to be twice as large as those diffusion processes this assumption chemical processes the correct factor COMPONENT i4C are generally involving 13C. For is correct, but for is 1.9 [102,103]. PROCESSES Isotope fractionations for various components of carbon metabolism in plants are summarized in Table 1 and described in detail in the following sections. Gas difSusiorl Isotope fractionation in gaseous diffusion has been misunderstood frequently in the plant physiology Carbon isotope fractionation in plants 559 Table 1. Expected values ofisotope fractionations in various components CO, fixation process of the Isotope discrimination Step Gas-phase diffusion of CO, Dissolution of CO,* Liquid-phase diffusion of CO, or HCO; CO, hydration* Carboxylation of phosphoenolpyruvate Relative to HCO; Relative to CO, Carboxylation of ribulose bisphosphatet Respiratory decarboxylations Isotope effect (k12/k” ) (“/,.A zyxwvutsrqponmlkjihgfedcbaZYXWVU 4.4 -0.9 0.0 - 7.0 2.0 -5.0 30 O-20 1.0044 0.9991 l.ooo 0.993 1.002 0.995 1.03 1.0&1.02 * Equilibrium isotope effect. t Exact magnitude uncertain. literature. diffusion Many authors have concluded that the ratio of coefficients of “CO, and 13C0, should be the anhydrase [log], it would not be expected that such a large fractionation in absorption would occur in a plant. A similar phenomenon was observed by Gaastra [llO], who attempted to measure the boundary-layer resistance to CO, transfer at the interface between air and a leaf replica saturated with KOH solution. His data indicated that CO, absorption at the interface was not complete, again presumably because of the relatively slow hydration of co,. ratio of the square roots of the masses (thus approximately 1 lo/,,), presumably because this relationship holds for the relative diffusion coefficients of water and CO* [ 104, 1051. This relationship, however, is fortuitous, and the correct relationship for gases diffusing in air is the ratio of the square roots of the reduced masses of CO, (mass 44 or 45) and air (mass 28.8) [105,106]. This provides an isotope discrimination of 4.4”/,, for CO, Liquid dt@ iiion diffusing in air. This factor is independent of temperature Theoretical treatments are currently inadequate for and, within broad limits, independent of pressure and predicting the ratio of diffusion coefficients of “CO, and CO, concentration. ‘%O, in aqueous solution, but a variety of experimental This isotope fractionation will occur both during the and theoretical data indicate that the effect should be no diffusion of CO, through the boundary layer, through the larger than a few tenths per mil ( [l lo]; R. Mills, personal stomata, and into the internal air spaces of the leaf and communication). during the reverse diffusion of internal CO, back to the atmosphere. There is, of course, no equilibrium isotope Hydration of CO, fractionation in diffusion. To the extent that diffusion is The isotope effect on the hydration of CO, has been rapid compared to other components of the overall measured a number of times, most recently by Mook et al. carbon absorption process, no fractionation will be [107], who presented data covering a range of observed. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA temperatures. At 25” the isotope fractionation between Disso lutio n o f CO, gaseous CO, and dissolved HCO; is -7.9” /,,. The The equilibrium carbon isotope fractionation in ,t3h” fractionation decreases slightly with increasing temperadissolution of CO, in water is -0.9°/00, with C ture. Note that at equilibrium, 13C concentrates in HCOS. Sign errors involving this fractionation have concentrating in the dissolved phase [107]. A proper caused numerous errors in interpretation of isotope value for the kinetic fractionation has not been measured, fractionations. but a logical upper limit for this value is the gas-phase No satisfactory estimates are available for the diffusional isotope fractionation of 4.4” /,,. Isotope magnitude of the kinetic isotope fractionation associated fractionations in physical processes are ordinarily quite with the hydration of CO,. Isotope fractionations in small. We expect that dissolution of CO, will ordinarily reactions in which CO, is formed can be as large as be at equilibrium, so the kinetic fractionation should not 60” /,,, and it is possible that such an effect might occur in be important. CO, hydration. However, most plants appear to contain Craig [17] and Baertschi [20] measured isotope carbonic anhydrase in sufficient levels to maintain fractionations near 16”/,, for the dissolution of CO, in equilibrium [109]. Under these conditions no kinetic aqueous Ba(OH), solution, but Craig [18] correctly isotope fractionation should be observed. pointed out that these fractionations have little to do with the situation in the plant. The experiment was conducted Phosphoenolpyruvate carboxylase in such a way that the rate-determining step was not The initial carboxylation in C, plants and in CAM dissolution, but rather reaction of CO, with OH-. The plants during dark fixation is catalysed by phosphoenollatter reaction is relatively slow [ 1081 and should show a pyruvate carboxylase. This enzyme requires HCO; as large isotope effect. Because plants contain carbonic 560 M. H. O’LFARY studies. Further, the isotopic difference between starting substrate [112,113]. The carbon isotope fractionation materials and products is multiplied by a factor of six (for associated with carboxylation has been measured several the six carbon atoms) in order to calculate the isotope times by combustion methods [114-l 161. Schmidt et al. fractionation. Obviously, small errors in the results [117] measured the isotope fractionation by observing generate large errors in the calculated isotope fractionaisotopic changes in a limited pool of source CO,. O’Leary zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIH tion This analysis also assumes that all materials are et al. (unpublished results) compared the isotopic absolutely pure: otherwise the combustion analyses are in composition ofcarbonofthe carboxylation product with error. Ribulose bisphosphate obtained commercially is that of the source HCO;. All three methods indicate that quite impure and even when purified, it decomposes the isotope fractionation is in the range 2.0~2.5” /,,. rapidly [121a]. Earlier studies failed to provide rigorous Another way to consider the isotope fractionation in criteria of purity for starting materials and products. PEP carboxylase is to assume that sufficient carbonic Another problem in studies of this isotope fractionation anhydrase is present that the CO,-HCO; interchange is arises from the fact that the substrate for the at isotopic equilibrium. This interchange contributes carboxylation is CO,, rather than HCO; [122]. In order -7”/,, to the isotope fractionation; thus if we use for the isotopic fractionation measurements to be valid, it dissolved CO, as our reference state in PEP carboxylase, is necessary that the interconversion of CO, and HCO; the overall fractionation becomes - 5” /00. This alternative be at isotopic equilibrium. Thus, carbonic anhydrase is often useful, for it saves the necessity of introducing an additional step into the mathematical modelling. must be present. Oxygenation also provides a potential source of Ribulose bisphosphate carboxylase problems [85b]. If oxygen is not rigorously excluded Park and Epstein [21] recognized the key role of during the measurements, isotope fractionations can ribulose bisphosphate carboxylase in carbon isotope occur because of the carboxylation/oxygenation parfractionation and measured an isotope fractionation of titioning. Since oxygenation also produces 3-phospho17”/,, compared to the total CO,-HCO; pool. This glyceric acid, this is a serious problem. An alternative approach to the isotope fractionation has corresponds to a fractionation of only 9” i,,0 compared to CO,. Subsequent measurements have invariably given been taken by Schmidt et ul. [117], who conducted the isotope larger values. Deleens et al. [ 1151 reported carboxylation in the presence of limiting COZ and fractionations in the range 4%99” /,, for carboxylase from measured the change in isotope ratio of the source CO, as spinach and maize. The details of these experiments have the carboxylation proceeded. By this method, enzymes never been reported. Christeller et rrl. [l IS] reported a from spinach and maize gave isotope fractionations of discrimination of 27”/,, for the carboxylase from soybean lg”/,, with a very small temperature dependence. Thus, ribulose bisphosphate carboxylase shows a large at 25”. The discrimination was nearly independent of temperature. discrimination against r3C02, but the exact magnitude of the discrimination is in doubt. In this author’s opinion, In one of the most careful studies, Estep et al. [119] studied isotope fractionations with purified enzyme and the best value is near 30”/,,, but the uncertainty is near with enzyme from freshly lysed spinach chloroplasts. In *lo”/,,. both cases the fractionation was 36”/,, at 30”. The Finally, one must ask whether the isotope fractionation carboxylase from A gmenellum quadruplicatum gave an by ribulose bisphosphate carboxylase must be the same isotope fractionation of 32”/,,. Fractionations in the under all conditions. The answer to this question is clearly range 2%38” j,, were observed for carboxylases from no. Extensive work of O’Leary and collaborators [15,16] other organisms. Experiments with the spinach enzyme has shown that the isotope fractionations resulting from revealed that the fractionation increased to 42”/,, when enzymatic reactions may vary with pH, temperature, Mn2+ was substituted for MgZf. Substitution of Ni2+ metal ion and other variables. Thus, some of the variation reduced the fractionation to 30”/,,. Estep et al. [120] in isotope fractionations found above may reflect real variations with reaction conditions. Fortunately, vasubsequently reported that the carboxylase from Cylindrotheca sp. gave a fractionation of 33”j,,. riation of fractionation with substrate concentration is not generally expected [16]. a carbon isotope Whelan ~1 a/. [114] reported fractionation of 34” oo_ relative to dissolved CO1 for the enzyme from Sorghum. Subsequent studies from the same Respiratory processes laboratory [121] presented a careful analysis of the Enzymatic decarboxylations in oitro often show factors likely to affect the fractionation. An isotope significant carbon isotope fractionations [I 51. However, fractionation of 27”!,, was obtained for the enzyme from based on the isotopic composition of respired carbon. the cotton at 35”. extent to which these fractionations are expressed in ciao All the studies reported above have been performed by appears to be small. Nonetheless, our picture of the combustion methods, in which the key issue is the isotopic compositions of plants will not be complete until comparison of the isotopic compositions of the substrates we are able to define accurately the quantity of carbon lost CO, and ribulose bisphosphate with that of the product by respiration and the isotopic composition of that 3-phosphoglyceric acid. A number of factors complicate carbon. Particularly in C3 plants, where because of photothe analysis ofsuch a system. In the first place, the isotopic respiration the total amount of respired carbon may be difference between starting materials and products is large, this is an important consideration. zyxwvutsrqponmlkjihgfe assumed to reflect only the isotopic fractionation associated with the source CO,. In fact, this is true only if INTEGRATED MODELS ribulose bisphosphate is quantitatively consumed in the reaction, so that there can be no isotope fractionation As noted earlier, fractionations may be of two types, connected with the carbons of this substrate. This factor kinetic and thermodynamic. The task for the present does not appear to have been accounted for in the earlier section is to provide a general approach for integrating Carbon isotope fractionation in plants 561 these fractionation factors to explain the isotope fractionations in the various steps in the carbon discriminations which might be observed under various absorption process, and the relative rates of these various conditions. The thermodynamic/kinetic distinction prosteps. If no fractionation of isotopes occurs as a result of vides a key to this approach. either respiratory processes or translocation, then the Most modelling of isotope fractionation to date has equations derived here will apply not only to the specific been based on the simple assumption of additivity of site(s) incorporated during carboxylation, but to the fractionation factors [2,3,35-j. This is equivalent to whole leaf as well. assuming that all steps except the last one in the particular sequence under consideration are at equilibrium; thus, we Two-step scheme are dealing with a set of thermodynamic factors plus one Consider a carboxylation system which can be final kinetic factor. This model is occasionally correct, but expressed as only two steps. In the first step, atmospheric more often it represents an unacceptable simplification. CO, [called CO,(ext)] diffuses into the plant and Another extreme model has sometimes been used. In becomes internal CO, [called CO,(i)]. This diffusion will to some extent be reversible. In the second step, CO, is certain cases an intermediate may be totally sequestered used to form the first carboxylation product R-CO;. If and have no metabolic fate except to be carried through the next step in the pathway. Under these circumstances carboxylation occurs by way of phosphoenolpyruvate the isotope fractionation associated with that next step carboxylase, then the second step must include CO, will not be expressed. This occurs. for example, in the hydration. The scheme is summarized as decarboxylation-carboxylation sequence in the bundle CO,(ext) &CO,(i) 2 R--CO;. sheath cells of C, plants. Even though the carboxylation (IO) k, of ribulose bisphosphate in the bundle sheath cells should show a large isotope fractionation, this fractionation is Transport of material from one state to the next is not expressed because virtually all of the CO, is carried described by rate constants k,, k, and k, .For example, the through carboxylation. However, as noted later, the rate of conversion of CO,(ext) to CO,(i) is given by the intracellular CO, pool which accumulates under such product of k, and some power of the concentration of conditions may have a very abnormal isotopic composiCO,(i). Under most circumstances it will be adequate to tion. assume that this dependence is on the first power of the The dynamics of carbon assimilation processes in plant concentration. Provided that the power is first or lower, it physiology has most often been considered drops out of the subsequent considerations of isotope [104,123,124] in terms of a resistivity model in which fractionation. various metabolic intermediates are connected by Rate constant k, describes the diffusion ofexternal CO, into the intercellular air space. It thus may include any resistances which represent the hindrance to metabolic flow through each particular step. The total throughput of external boundary layer diffusion, stomata1 diffusion, and perhaps an internal diffusion component. Of these, the the system can then be calculated by analogy with analysis stomata1 resistance to diffusion is generally the most of electrical networks. However, this is not the best approach to computing isotope fluxes because it is important. This diffusion will show a carbon isotope fractionation of 4.4” /,,. Rate constant k, reflects outward necessary when considering isotope fluxes to include diffusion of CO,(i) and thus should also show a carbon specifically the rate of return of material from one isotope fractionation of 4.4” /,,. metabolic pool to the preceding pool: the resistivity model The second step (k3) represents absorption of CO, at does not allow this. The overall modelling of isotope fractionation must be the air-liquid interface, transport of CO, into the cell, equilibration of CO, with HCO;, and carboxylation. We divided into three parts, which will be considered in turn: assume for the present that absorption and transport are 1. The initial steps in carbon incorporation. All steps rapid, although direct measurement of the rates of these jirst irreversible step, most often the through the zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA processes by gas exchange methods has not been possible carboxylation step, must be included. This approach as yet; most analyses have combined these steps with the is sufficient to explain isotopic compositions of whole plants, provided that respiratory processes do not carboxylation to give one overall liquid-phase contribufractionate carbon. Isotopic compositions of in- tion [104,123]. As noted, the isotope fractionations associated with absorption and liquid-phase diffusion dividual metabolites or individual carbon atoms should be extremely small. cannot properly be understood at this level. In the case of C, photosynthesis, hydration of CO, to 2. Branch points. Isotope fractionations subsequent to form HCO; must precede carboxylation. As noted the initial steps occur at branch points in metabolic previously, we assume that carbonic anhydrase is present pathways-points at which the intermediate either in. sufficient concentration to maintain this step at may suffer two different fates (giving the possibility equilibrium. Thus, we can logically lump the carboxylfor different isotope fractionations in the two ation and hydration steps together. directions) or else where a very large pool of an It is implicit in this treatment that the total magnitude intermediate accumulates (allowing for an isotope of the pool represented by CO,(ext) is sufficiently large fractionation from a relatively infinite pool). that no change in its concentration or isotopic 3. Non-branching intermediates. Such situations do not actually provide overall isotope fractionation, but it is composition occurs as a result of photosynthetic activity. If this condition is not met, then the apparent isotope necessary to discuss such cases in order to achieve a fractionation will be smaller than predicted. This was proper view of the isotopic composition of a observed, for example, in the work of Berry and metabolic intermediate. Troughton [125] who showed that under conditions I will first derive equations which relate the isotopic where plants absorb all of the CO, supplied, no isotope composition of photosynthetically-introduced carbon to fractionation is observed. the isotopic composition of the source, the isotope zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPON O)LEARY M. H. 567 The rate of change of the concentration time is given by d[CO,(i)j dr of CO,(i) and Ir,. The ratio !x:‘,Xj Ais then an ‘isotope e&t’ particular step, and we will abbreviate this as with E, _ X”,k’.‘. = X.1[CO,(eXt)] - (k, I -t X,) [CO,(i)]. (11) Stomata1 r&stance in higher plants appears to be regulated in such a way as to maintain the concentration of CO,(i) approximately constant [126]. Certainly at steady-state the change in concentration with time should be very small and it is possible to set equation 11equal to zero. Thus we find that 1 for that (15) and so on for the other isotope effects. Note that these are all kinetic, rather than thermodynamic, isotope effects. The isotope effect for the whole process. I\” k” (overall), IS obtained by writing equations for formation of R-CO; for each isotope spccics. combining. and integrating Thus, the overall isotope fractionation in this two-step scheme is a composite of four factors: isotope discriminations E,, E,. E,. and the ratio+ i,, /\?. The item of interest is actually the rate of formation of The \alucs of the isotope etfccts \\hich would be R -~C’O;. which is given by appropriate for the application ot‘cquat~on Ih to studies of C, and C, carboxylation are summarircd in Table 2. ] d [R-CO; 1 k, iCO,ii)] (13) with C, The value of E, to be cased in connection dt carboxylation is uncertain because of the uncertainty in (In the general case, we should expect that the rate of the correct isotope effect for ribulose hisphosphate formation of the carboxylation product should also carboxylase. No correction for CO, hydration is depend on the concentration of the other carboxylation necessary in this case. For C, carboxylat&n. E aincludes substrate, phosphoenolpyruvate or ribulose bisphosdissolution, CO, hydration and carboxylation comphate; however, as we will shortly take the ratio of two ponents. expessions of the sort given above. any concentration Values obtained by use ofequation 16can be compared depcndencc cancels out.) The internal CO1 concentration with experimental isotope effects calculated from can be substituted from equation 12 into this last equation 4 or 5. The remaining unknown is the ratio I\,;r(,. expression The ratio h,,k, is a very important quantity. In chemical kinetics it is often called a ‘partitioning factor‘ dLR--?,:I _ ~__V-i ~[Co,(ext)], because it reflects the partitioning of an intermediate [in (14) dr k, +k, this case, CO,(i)] between further reaction and return to the preceding state. When this ratio is large. COz (i) Although this last expression purports to give the rate of mostly reacts further and the carbon fixation rate is formation of carboxylation products in terms of the limited by diffusion; when the ratio is small. COL(i) concentration of external CO,. it is unlikely that in this form the equation has any real use. However, the returns to the starting state much more often than it important point is that this equation can provide a proper undergoes carboxylation and the carbon fixation rate is starting point for explaining the isotope fractionation limited by carboxylation. Previous models [3.21 ] have occurring in these steps. It is possible that the rate failed to take specific account of this partitioning. It is useful to consider two limiting casts in which constants given above may be different for carbon-12 and equation 16 might be used. If the first step and its reverse for carbon-l 3. We will call the isotopic rate constants for the firs1 step /,fL and it;“. respectively and similarly for zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGF are rapid compared to the subsque~~t Ii, carbou\.I;ltion step. Table Step 2. Isotope Processes Gas-phase effects for use in connection diffusion into the leaf CO, dissolution hydration -i: This number I-ibulose t This rate-constant for wrboml 3 would ratio be only tlishtly is tentative dllferent. and transport, (in c‘, plants), bizphosphate is for carbon-12. I because of the uncertainty carboxylase. although CO, carboxylation the ratio I6 equation c, plants included Reverse of step with <‘j plants I .0044 I .0044 I .0044 I .0044 I .01” (1‘N-1 in the isotope clrect fol Carbon isotope fractionation in plants 563 then zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA k,/k, approaches zero and the observed isotope processes often show significant isotope fractionation fractionation becomes zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA [151. such fractionation might not be expressed in riro because no other fate is accessible to the decarboxylation k’2/k’3 (overall) = E,E,/E,. (17) substrate. Consider, for example, a scheme in which an The fractionation is then the product of the equilibrium intermediate (I) is converted into some carboxylic acid fractionation in the first step (= E,/E2) and the kinetic (R-CO;) which then undergoes decarboxylation: fractionation in the second. This might be the case, for example, when CO, diffusion in and out of the leaf is rapid 1: R-CO, 4 R-H + CO,. (20) compared to the subsequent carboxylation step. In that case, E, = E, (the isotope fractionation due to diffusion is If the conversion of I to R-CO; is irreversible, or nearly the same in and out of the leaf) and the observed so, then any isotope fractionation which should fractionation becomes equal to the fractionation conaccompany the decarboxylation will not be expressed, nected with carboxylation. and the isotopic composition of the respired carbon will The opposite limiting case is the one in which internal be the same as that of the corresponding carbon atom in 1. CO, always undergoes carboxylation and never returns It is interesting to note that the isotopic composition of to the atmosphere. In this case k,/k, becomes very large the carboxyl group of R -CO; at steady-state may not be and the observed fractionation becomes equal to E,, the the same as that of the intermediate from which it is kinetic fractionation in the first step. formed or the CO, which is produced. The isotopic composition of the carboxylic acid at steady-state is given Isotope fractionation and stomata1 opening by For the carbon fixation mechanism described by R(R-CO;) = (E,/E,)(R(I)) (21) equation 10, we expect that the rate of the first step will be controlled primarily by stomata1 resistance, and this will where (R(R-CO;) and R(I) are the appropriate isotope be reflected both in k, and in k,. Later I show that in many ratios. Qualitatively, this means that the isotopic composition of an intermediate will adjust itself to plants the partition factor k,/k, is not too different from unity. When that is true. changes in stomata1 resistance compensate for the difference between the isotope effect will result in significant changes in the partition ratio and on its formation and that on its decomposition; the larger thus also in the 613C value for the plant (cf. equation 16). the difference in isotope effects, the larger the difference in An increase in stomata1 resistance will result in a isotopic compositions. A similar phenomenon holds when corresponding any or all of these conversions are reversible. decrease in rate constants k, and k, and The important consequence of this fact is that the consequently an increase in k,/k,. In C, plants an increase isotopic compositions of individual metabolites (or, more in stomata1 resistance will cause 613C values to become properly, of particular sites) in plants reflect not only the more negative. In C, plants an increase in stomata1 isotope fractionations in their formation, but also the resistance will cause d13C values to become more positive. fractionations in their metabolism. This fact may Three-step model influence, for example, the isotopic composition of malate If the above two-step model becomes inadequate for isolated from C, plants [92]. Malic enzyme shows an describing carbon isotope fractionation (for example, if isotope fractionation of 30”/,, for carbon-4 of malate (M. CO, hydration or liquid diffusion must be explicitly H. O’Leary and C. Roeske, unpublished results), so the considered), it is possible to use the same kind of steadyisotopic composition of the malate isolated may or may state assumption (setting expressions of the form of not reflect the isotope fractionation associated with its equation 11 equal to zero) for more complex situations. formation. This same problem does not occur with malate For example, consider the scheme isolated at the end of the dark period from CAM plants because the malate is not being metabolized under those conditions. The small malate pool remaining at the end of C02(ext)~C02(a)~C0,(b)~ R-CO;, (18) k, ki the light period is slightly different from the pool at the end of the dark period [99] because of isotope in which CO,(a) and CO,(b) represent two successive fractionations connected with the metabolism of malate. states of CO2 during photosynthesis. In this case the overall isotope fractionation is given by Branch points $ (overall) _ E 1 E,E,IJ%& + (EJE2) WU + (WJ (k&J l+ k, :k, + &/k2) (WkJ’ (19) As described earlier, a number of cases are known in which individual carbon atoms or individual metabolites have an isotopic composition significantly different from the plant as a whole. This phenomenon often arises as a result of branch points in a metabolic pathway. For example, consider metabolites A, IS, C and D in the following scheme. Thus the overall fractionation depends on the individual isotope effects E, through E, and on partitioning factors k,Jk, and k,Jk,. Fractionations subsequent to carboxylation The isotopic composition of whole plant carbon is governed by the isotope fractionation which accompanies CO, absorption and by any isotope fractionation which accompanies respiration. Even though decarboxylative We assume for the moment that the conversions are all irreversible but the argument is not appreciably different if the conversions are reversible. The isotopic composition of product D compared to that of A (R(A)/R(D)) is given 564 M. H. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONM O’LEA RY under striking that the range ofvariation ofvalues ofk,/k, by equation 16. Compounds C and D will have different a variety of conditions must be quite small. In terms of k, and k, isotopic compositions if the isotope effects on zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA equation 21, the internal CO, concentration in C, plants are different. The operation of this phenomenon in lipid must be maintained nearly constant, independent of formation has been suggested by DeNiro and Epstein environment. This, of course, is consistent with the 1971. homeostatic hypothesis [ 1261. One important and unknown factor complicates the above analysis. We have assumed that respiratory There is currently a considerable interest in the processes make no contribution to the isotopic composivariation of stomata1 resistance in CAM plants in tion of C, plants. Because of the extent of photorespiraresponse to CO, concentrations [127,128]. Gas exchange tion in C, plants, it would take only a small respiratory and direct measurements by gas measurements fractionation in order to make a significant ditference in chromatography of CO,(i) have provided useful inforthe overall isotopic composition of the plant. There are mation in this regard. Carbon isotope fractionations differing reports as to whether or not there is isotope provide another useful approach. fractionation connected with respiration, and this issue Consider the model given in equation 10, in which k, must be considered open. The effect of oxygen on ti”C and k2 represent inward and outward diffusion of CO,, values in C, plants [X7] suggests that photorespiration We expect that k, = k,; and k, represents carboxylation. has an important influence on isotopic compositions. that is, the resistance to CO, diffusion is the same in both Thus our best guess at this point is that C, plants show directions. Equation 12 can be rearranged to give zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGF a large isotope discrimination because of the discrimination in the carboxylation step. However. diffusion is iCO,(ext)l partially rate-limiting and this serves to make the irl rim -~= 1 + k,,L,. (23) [CO,(i)1 fractionation smaller than the carboxylasc fractionation. The ratio k3,‘k2 is, of course, obtained from the isotope c, planrs fractionation measurements, so a direct correlation is The isotope effects given in Table 2 can be used to expected between gas exchange measurements and calculate expected d”C values for various diffusion/ isotope fractionation measurements. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA carboxylation partitionings. If the rate of carbon fixation is entirely limited by carboxylation, then h,!k2 (equation CURRENT INTERPRETATION OF ISOTOPE FRACTIONATION 16) is small. k' ‘jk” (overall) = 0.994, and the predicted IN PLANTS <ir3C value for C, plants is near O”,‘,,. This is. of course, very different from what is observed, and the common The purpose of this section is to interpret existing data assumption that isotope fractionation in C, plants simply on isotopic compositions of plants and metabolites in reflects carboxylation is incorrect. terms of the models given in the preceding sections of this On the other hand, if ditfusion is limiting and review. carboxylation is relatively fast (I\,,/<, large), then whole plant carbon should have a (i’“C value near - I lo;,, in C, plants incorporate CO, by carboxylation of the absence of any respiratory fractionation. Models in ribulose bisphosphate. C, plants show a large discrimiwhich neither ditfusion nor carboxylation is entirely nation against ‘%Y. as does ribulose bisphosphate limiting give (i13C values between 0 and - 1I. carho\ylasc. and most authors have assumed that the The diffusion limited case is, of course. much closer to isotopic composition of C, plants simply reflects isotope what is observed, but it is important to note that the fractionation due to ribulose bisphosphate carboxylase. predicted value for the diffusion limited case 1- 1I” :,,) is The key question is, what value should we use for the slightly different from the accepted C, value (- 14” /,,). isotope fractionation due to ribulose bisphosphate One mechanism by which this discrepancy could be carboxylase? The range of possible values is large, diminished is isotope fractionation during respiration, covering at least 20-40” /,,. The fractionation of carbon but the limited studies available for C, plants fail to during C, photosynthesis is 20”/,,, and ifthe fractionation support this explanation [37]. for ribulose bisphosphate carboxylase is also 20”/,,, then A more likely possibility is that some leakage of COz we would say that k,jk, is small, CO, diffusion in and out occurs during decarboxylation and refixation of COz in of the leaves is rapid compared to carboxylation, the the bundle sheath in C, plants. Qualitatively, this results stomata1 resistance is very small, and CO,(i) should be in the expression ofa small portion of the expected isotope near 330ppm. discrimination for ribulose bisphosphate carboxylase: the However, CO,(i) is smaller than this [126], and the more CO, is lost by this process, the more expression of ribulose bisphosphate carboxylase fractionation is prothe ribulose bisphosphate carboxylase fractionation bably larger than 20”/,,. An alternative model is one in would be observed. Quantitatively. we can use equation which diffusion plays a limited role in reducing the 22, in which B is the bundle sheath CO, pool. D is CO? magnitude of the isotope fractionation. If we assume a which has been fixed by ribulose bisphosphatc carbovalue of 30” lo0 for the carboxylase fractionation, then the xylase, and C is the CO, which escapes. Then E, = 1.004, partitioning factor is 0.6, and stomata1 resistance E, = 1.000, E, = 1.03, and a leakage of j”II of the CO, contributes significantly to the overall carboxylation rate. pool would shift the predicted 8’ ‘C value from - 1I to This combination of factors gives a CO,(i) value of -13” /,,. It is unlikely that such a small leakage would 200 ppm, consistent with what is generally observed for C, have been detected in any experiments performed to date. plants [ 1261. If this model is correct, then some variation Another problem with this model is that it predicts in (jlJC values in C, plants might have been expected as a CO,(i) values approaching zero. whereas experimental result of variations in the ratio k,<:k,. However. it is values are closer to IO0 ppm ‘I 261. This can be 565 Carbon isotope fractionation in plants accommodated if diffusion is mostly, but not entirely, limiting in CO, absorption and the bundle sheath leakage is somewhat greater than the 5 y0 estimated above. 9. 10. An important consequence of the conclusion that diffusion primarily limits carbon fixation in C, plants is that the photosynthetic rate for C, plants appears not to be limited by the amount of phosphoenolpyruvate carboxylase or by the rate of synthesis of phosphoenolpyruvate. Increased efficiency of carbon assimilation in C, plants will not be brought about by finding plants with a higher level of carboxylase. Instead, it will be necessary either to increase CO, levels or to find a plant with a lower diffusive barrier to CO, transport; such a decreased barrier would likely result in a lower water efficiency for the plant. 11. 12. 13. 14. 14a. 14b. Craig, H. (1957) Geochim. Cosmochim. Acta 12, 133. Mook, W. G. and Grootes, P. M. (1973) Int. J. M ass Spectrom. Ion Phys. 12, 213. Keeling, C. D. (1961) Geochim. Cosmochim. Acta 24,277. Kroopnick, P. (1974) Earth Planet. Sci. Letters 22, 397. Keeling, C. D., Mook, W. G. and Tans, P. P. (1979) Nature 277, 121. Hoefs, J. (1973) Stable Isotope Geochemistry. p. 87. Springer, New York. Vogel, J. C. (1978) Oecof. Plant. 13, 89. Medina, E., and Minchin, P. (1980) O ecologia (Berlin) 45, 371. 15. 16. O’Leary, Tamelen, York. O’Leary, M. H. (1977) Bio-Organic Chemistry (van E. E., ed.) Vol. I, p. 259. Academic Press, New M. H. (1978) Transition States of Biochemical R. L., eds.) p. 285. Plenum, New York. Craig, H. (1953) Geochim. Cosmochim. Acta 3, 53. Craig, H. (1954) J. Geol. 62, 115. Wickman, F. E. (1952) Geochim. Cosmochim. Acta 2,243. Baertschi, P. (1953) Helu. Chim. Acta 36, 773. Park, R. and Epstein, S. (1960) Geochim. Cosmochim. Acta 21, 110. Park, R. and Epstein, S. (1961) PIant Physiol. 36, 133. Kortschak, H. P., Hart, C. E. and Burr, G. 0. (1965) Plant Physiol. 40, 209. Hatch, M. D. and Slack, C. R. (1970) Anna. Rev. Plant CAM zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA plants R. D. and Schowen, Processes (Gandour, When CAM plants engage in direct daytime CO, fixation, they show isotope fractionations which are 17. identical with those shown by C, plants. I have earlier 18. concluded that C3 plants probably show some diffusional 19. resistance to carbon fixation, and the identical isotope 20. fractionations indicate that CAM plants, too, must show 21. some diffusional resistance to carbon fixation. CAM 22. plants have a much lower density of stomata than C, 23. plants [128], but this density does not seem to translate into a greater diffusive resistance to CO, transport. 24. When CAM plants engage in dark CO, fixation, their overall isotope fractionation is like that of C, plants. Physiol. 21, 141. However, carbon-4 of malate formed in the dark [99] is Bender, M. M. (1968) Radiocarbon 10, 468. 25. significantly more positive than the most positive values Bender, M. M. (1971) Phytochemistry 10, 1239. 26. which have been observed for CAM plants, indicating that 27. Smith, B. N. and Epstein, S. (1971) Plant Physiof. 47,380. additional fractionation follows the initial carbon fixation. Brown, W. V. and Smith, B. N. (1972) Nature 239, 345. 28. This may either occur as a result of leakage during 29. Smith, B. N. and Turner, B. L. (1975) Am. J. Botany. 62, refixation of CO, in the light or else as a result of isotope 541. fractionation during respiration. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Raynal, J. (1972) C. R. Acad. Sci. Paris Ser. D 275, 2231. 30. 31. particularly grateful to C. Barry Osmond for advice concerning many aspects of plant physiology. Graham Farquhar suggested the relationship between isotope Acknowledgements- l am Smith, B. N. and Brown, W. V. (1963) Am. J. Botany 60, 505. 32. Tregunna, E. B., Smith, B. N., Berry, J. A. and Downton, W. J. S. (1969) Can. J. Botany 48, 1209. fractionations and CO,(i) values. I thank Joe Berry, Margaret Tieszen, L. L., Senyimba, M. M., Imbamba, S. K. and 33. Bender, Bruce Smith, Joe Holtum, John Hayes, Klaus Winter Troughton, J. H. (1979) O ecologia (Berlin) 37, 337. and Park Nobel for useful comments on the manuscript. Work in Waller, S. S. and Lewis, J. K. (1979) J. Range Manage. 32, 34. my laboratory was supported by NSF grants PCM77-00812, 12. PCM79-22697 and 2AS-21. This article was written in part Troughton, J. H. (1972) Proc. 8th Int. Congr. on 35. during the tenure of a visiting fellowship at the Australian Radiocarbon p. 420. The Royal Society of New Zealand, National University. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Wellington. Troughton, J. H., Card, K. A. and Bjorkman, 0. (1974) 36. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. Carnegie Inst. W ashington Yearb. 73, 780. Troughton, J. H., Card, K. A. and Hendy, C. H. (1974) 37. J. C. (1975) Environmental and Biological Control zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIH Carnegie Inst. W ashington Yearb. 73, 768. of Photosynthesis (Marcelle, R., ed.) p. 323. W. Junk, The Hatch, M. D., Osmond, C. B., ‘Troughton, J. H. and 38. Hague. Bjorkman, 0. (1972) Carnegie Insr. W ashington Yearb. 71, Troughton, J. H. (1979) Encyclopedia of Plant Physiology, 135. New Series, Vol. 6 (Gibbs, M. and Latzko, E., eds.) p. 140. Bjorkman, O., Troughton, J. H. and Nobs, M. A. (1974) 39. Springer, New York. Benedict, C. R. (1978) Annu. Rev. Plant Physiol. 29, 67. Brookhauen Symp. Biol. 25, 206. Bjorkman, O., Nobs, M. A., Pearcy, R., Boynton, J. and Lerman, J. C. and Troughton, J. (1975) Proc. 2nd Int. 40. Berry, J. (1971) Photosynthesis and Photorespiration Co& on Stable Isotopes (Klein, E. R. and Klein, P. D., (Hatch, M. D., Osmond, C. B. and Slatyer, R. O., eds.) p. eds.) p. 630. USERDA CONF-751027. 18. Wiley-Interscience, New York. Smith, B. (1972) Bioscience 22, 226. Osmond, C. B. (1978) Annu. Reu. Plant Physiol. 29, 379. Degens, E. T. (1969) Organic Geochemistry (Eglinton, G. 41. Kluge, M. and Ting, I. P. (1978) Crassufacean Acid and Murphy, M. T. J., eds.) p. 304. Springer, New York. 42. M erabolism, Springer, New York. Schmidt, H.-L. and Winkler, F. J. (1979) Ber. Dtsch. Bot. Osmond, C. B., Allaway, W. G., Sutton, B. G., Troughton, 43. Ges. 92, 185. J. H., Queiroz, O., Liittge, U. and Winter, K. (1973) Lerman, J. C. and Raynal, J. (1972) C. R. Acad. Sci. Paris Nature 246, 41. Ser. D 275, 1391. Lerman, 566 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 51. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 69a 70. 71. 72. 73. 74. 75. M. H. O’LEARY 76. Deuser. W. G.. Degens. E. T. and Guillard. R. R. L. (19681 Bender, M. M., Rouhani, I., Vines, H. M. and Black, C. C., Adcr 32, 657. Groch6tr. C’o.\/r~oc~hb,~. Jr. (1973) Plant Physiol. 52, 427. 77. Sirevag, R., Buchanan, B. B., Berry, J. A. and Troughton. J. Lerman, J. C. and Queiroz, 0. (1974) Science 183, 1207. H. (1977) Arch. Miuohiol. 112. 35. Medina, E. and Troughton, J. H. (1974) Plant Sci. Lerters 78. Smith, B. N. and Epstein, S. (1970) Plant Phpinl. 46, 738. 2. 357. 79. Card, K. A., Mahall, B. and Troughton, J. H. (1974) Winter, K., Liittge, LJ., Winter, E. and Troughton, J. H. Curnegie lnsr. Washingrorl )?trrh. 73. 784. (I 978) Oecoloaicr (Berlin) 34, 225. 80. Guy. R. D., Reid, D. M. and Krouse. H. R. (1980) Winter, K. and Troughton, J. H. (1978) Z. PfianzenOecologia (Berlin) 44, 24 1. phpk/. 88, 153. 81. Abelson. P. H.andH0ering.T. C. (196l)Proc. Nurl. ilcutl. Bloom, A. J. and Troughton, J. H. (1979) Oerologia SC/. L’.S.A. 47. 623. (Berlin) 38, 35. 82. Calder. J. A. and Parker. P. L. (1973) GeochLn. zyxwvu Osmond, C. B. (1975) Encironrnental and Biological C o sm o c him . Ac rrr 37. 133. (Marcel]. R., ed.) p. 311. W. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLK Control of‘ Photosynthesis 83. Degens, E. T.. Guillard. R. R. L., Sackett, W. M. and Junk. The Hague. Hellebust, J. A. (1968) I)ec~p Serr Res. 15, 1. Allaway, W. G., Osmond, C. B. and Troughton. J. H. 84. Uphaus, R. A. and Katz. J. J. (1967) Science 155, 324. (1974) Mechanisms of Regulation of Plant Growth 85. Smith, B. N. (1975) Proc. 2rrd fnt. Conf on Stable Isotopes (Bieleski, R. L., Ferguson, A. R. and Cresswell, M. M., E. R. and Klein. P. D.. eds.) p. 2X7. (Klein, eds.) p. 195. The Royal Society of New Zealand, USERDA CONE--751027 Wellington. 85a. Hsu, J. C. and Smith. B. N. (1972) P/am Cell Pk~ziol. 13. Lerman, J. C., Deleens, E.. Nato, A. and Moyse, A. (1974) 689. P/ant Phjsiol. 53, 581. 85b. Jensen, R. G. and Bahr. J. T. (1977) ilnnu. Ref.. Plum Deleens. E. and Gamier-Dardart, J. (1977) Planra 135, Phjsiol. 28, 379. 241. J. H. and Bjiirkman, 0. (1972) Osmond. C. B.. Bender, M. M. and Burris, R. H. (1976) 86. Berry, J., Troughton, Carnegie Inst. Wshington Yearh. 71, 158. Aust. J. Plant Pk~siol. 3, 787. 87. Jacobson, 13.S.. Laties, G. G.. Smith, B. N., Epstein, S. and Nalborczyk, E., LaCroix, L. J. and Hill, R. D. (1975) Can. Laties, B. (1970) Biochim. Biophj,s. Acfa 216, 295. J. BoranJ 53, 1132. A., Nissenbaum. A., Galun, M. and Waisel. 88. Shomer-Ilan, Eickmeier, W. G. and Bender, M. M. (1976) Oecologia Y. (1979) Plant Ph_wio/. 63, 201. (Berlin) 2.5, 341. 89. Marcelle, R. (1975) Enc.ironmc~nt[II and Biologkal Control Troughton, J. H., Mooney, H. A., Berry, J. A. and Verity, r~Photo,s~nthesi.s (Marcelle. R., ed.) p. 349. W. Junk. The D. (1977) Oecologia (Berlin) 30, 307. Hague. Rundel, P. W., Rundel, J. A., Ziegler, H. and Stichler. W. W., Ting, 1. P. and Sternberg. L. 0. (1979) 90. Cockburn, (1979) Oecologia (Berlin) 38, 45. Plant Phq’siol. 63, 1029. Mooney, H., Troughton, J. H. and Berry, J. A. (1974) Carnegie Inst. Washington Yearh. 73, 793. 91. Spalding, M. H., Stumpf, D. K., Ku, M. S. B., Burris, R. H. and Edwards. C. E. (1979) .Aust.J. P/am. Phjsiol. 6, 557. Mooney, H. A., Troughton. J. H. and Berry, J. A. (1977) Whelan, T., Sackett, W. M. and Benedict, C. R. (1970) 92. Oeco/ogia (Berlin) 30, 295. Biochem. Biophw. RKS. Comm~m. 41. 1205. Schulze, E.-D., Ziegler, H. and Stichler, W. (1976) Degens, E. T., Behrendt, M., Gotthardt. G. and 93. Oecologia (Berlin) 24, 323. Reppmann, E. ( 1968) Deep Sea Rra. 15. 11. Szarek, S. R. and Troughton, J. H. (1976) Phnt Pk~sio/. Jacobson, B. S., Smith, B. N., Epstein, S. and Laties, G. G. 94. 58, 367. (1970)J. Gen Ph~xiol. 55, 1. Osmond, C. B., Ziegler, H., Stichler, W. and Trimborn, P. Smith, B. N. a nd Jacobson, B. S. (1976) Plant Gel/ Phxsiol. 95. (1975) Orcologia (Berlin) 18, 209. 17, 1089. Winter, K. (1979) Oecologiu (Berlin) 40, 103. 96. Ziegler, H. (1979) Ber. Dtsch. Bat. Ges. 92, 169. Lowdon, J. A. (1969) Radiocarbon 11, 391. DeNiro, M. J. and Epstein, S. (1977) %ier~ce 197. 261. 97. Andrews. T. J. and Abel, K. M. (1979) P/am Physiol. 63, Deleens, E., Garnier-Dardart, J. and Queiroz. 0. (1979) 9x. 650. Plautu 146, 441. Wong, W. W. L., Benedict, C. R., Newton, R. J. and 99. O’Leary, M. H. and Osmond. C. B. (1980) Plant Physiol. Bhaskaran, S. (1979) P/ant Ph.vsio/. 63 (suppl.), I. 66, 931. Benedict, C. R., Wong, W. W. L. and Wong. J. H. H. (1979) 100. Rivera. E. R. and Smith, B. N. (1979) P/mt Ph~siol. 64. Plum Physiol. 65, 512. 966. Bender, M. M. and Berge, A. J. (1979) Oe&ogia (Berlin) 44. I 17. 101. O’Leary, M. H. and Yapp. C. J. (1978) Biochem. Bioph!x Res. Cornmun. 80, 155. Boag, T. S. and Brownell, P. F. (1979) ,4ust. J. Pht Yankwich, P. E., Promislow, A. L. and Nystrom, R. F. Ph~~,sio/. 6. 43 I 102. (1954) J. Am. Chrm. SW. 76, 5893. Willmer, C. M. and Firth, P. (1980) J. Exp. Botany 31, 1. Smith, B. N., Oliver, J. and McMillan, C. (1976) Bat. Gaz. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLK 103. Stern. M. J. and Vogel. P. C. (1971 ) J. Chrru. Pilyc. 55. 137, 99. 2007. Jarvis, P. G. (1971 ) Pltrnt Photos)mhetic Produc.rion 104. Smith, B. N., Herath, H. M. W. and Chase, J. B. (1973) Manual of xMethodh (Sestak, Z., Catsky, J. and Jarvis. P. Plant Cell Phj,sio/. 14, 177. Troughton, J. H. and Card, K. A. (1975) Planta (Berlin) 123, 185. Pardue. J. W., Scalan, R. S., Van Baalen, C. and Parker, P. L. (1976) Geochirn. Cosmochim. Acta 40, 309. Sackett, W. M., Eckelmann, W. R., Bender, M. L. and Be, A. W. H. (1965) Science 148, 235 105. 106. G., (eds.) p. 566. W. Junk, The Hague. Mason, E. A. and Marrero. T. R. (1970) .4(/1,. Al. Mol. /‘hp. 6, 155. Hirschfelder. J. 0.. Curtiss, C. I;. and Bird, R. B. (1954) ~Moleculor Theory of Ga.ws amI Liquids. pp. 527. 539. Wiley. New York Carbon isotope fractionation in plants 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 567 Mook, W. G., Bommerson, J. C. and Staverman, W. H. 119. Estep, M. F., Tabita, F. R., Parker, P. L., and Van Baalen, Planet. Sci. Letters 22, 169. (1974) Earth zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA C. (1978) Plant Physiol. 61, 680. 120. Estep, M. F., Tabita, F. R. and Van Baalen, C. (1978) .I. Edsall, J. T. (1969) CO,: Chemical, Biochemical and Phycol. 14, 183. Phy siofogical Aspects (Forster, R. E., Edsall, J. T., Otis, A. B. and Roughton, F. J. W., eds.). NASA SP-188, p. 15. 121. Wong, W. W., Benedict, C. R. and Kohel, R. J. (1979) Poincelot, R. P. (1979) Ency clopedia of Plant Phy siology Plant Physiol. 63, 852. 121a. Paech, E., Pierce, J., McCurry, S. D. and Tolbert, N. E. New Series, Vol. 6 (Gibbs, M. and Latzko, E., eds.) p. 230. Springer, New York. (1978) Biochem. Biophys. Res. Commun. 83, 1084. Gaastra, P. (1959) M eded. Lanbouwhogesch. W ageningen 122. Cooper, T. G., Filmer, D. L., Wishnick, M. and Lane, M. 59, 1. D. (1969) J. Biol. Chem. 244, 1081. Mills, R. and Harris, K. R. (1976) Chem. Sot. Rev. 5,215. 123. Nobel, P. S. (1974) Biophy sical Plant Phy siology , p. 325. Freeman, San Francisco. Cooper, T. G. and Wood, H. G. (1971) J. Biol. Chem. 246, 124. Nobel, P. S. and Hartsock, T. L. (1978) Plant Physiol. 61, 5488. Coombs, J., Maw, S. L. and Baldry, C. W. (1975) Plant Sci. 510. Letters 4, 97. 125. Berry, J. A. and Troughton, J. H. (1974) Carnegie Inst. W ashington Yearb. 73, 785. Whelan, T., Sackett, W. M. and Benedict, C. R. (1973) 126. Wong, S. C., Cowan, I. R. and Farquhar, G. D. (1979) Plant Physiol. 51, 1051. Deleens, E., Lerman, J. C., Nato, A. and Moyse, A. (1974) Nature 282, 424. 127. Wong, S. C., Cowan, I. R. and Farquhar, G. D. (1978) Proc. 3rd Int. Congr. on Photosynthesis, Rehovot, Israel, p. 1267. Elsevier. Amsterdam. Plant Physiol. 62, 670. 128. Black, C. C., Goldstein, L. D., Ray, T. B., Kestler, D. P., Reibach, P. H. and Benedict, C. R. (1977) Plant Physiol. 59, and Mayne, B. C. (1976) CO, M etabolism and Plant 564. Productivity (Burris, R. H. and Black, C. C., eds.) p. 113. Schmidt, H.-L., Winkler, F. J., Latzko, E. and Wirth, E. University Park Press, Baltimore. (1978) Isr. J. Chem. 17, 223. Christeller, J. T., Laing, W. A. and Troughton, J. H. (1976) Plant Physiol. 57, 580.