THE JOURNAL OF EXPERIMENTAL ZOOLOGY 274:291-299 (1996) Subcellular Organization of Intermediary Metabolism in the Hepatopancreas of the Terrestrial Snail, Cepaea nemoralis: A Cytosolic P-HydroxybutyrateDehydrogenase J.A. STUART AND J.S. BALLANTYNE Department of Zoology, University of Guelph, Guelph, Ontario, N l G ZWl, Canada ABSTRACT The subcellular distributions of key enzymes of the tricarboxylic acid cycle, electron transport chain, ketone body, amino acid, and carbohydrate metabolism were studied in the hepatopancreas of the terrestrial snail Cepaea nemoralis. The presence of mitochondria1 carnitine octanoyl transferase, carnitine palmitoyl transferase, and 3-hydroxyacyl-CoAdehydrogenase indicate an active lipid catabolic pathway in this tissue. Activities of enzymes of ketone body metabolism are similar in magnitude to those of carbohydrate metabolism, suggesting an important metabolic role for ketone bodies in C. nemoralis. Two enzymes that, in mammals, catalyze the synthesis of ketone bodies, hydroxymethylglutaryl-CoA synthase and hydroxymethylglutaryl-CoA lyase, are not detectable in Cepaea hepatopancreas. The activity of 3-oxoacid-CoAtransferase and a portion of acetoacetyl-CoA thiolase activity are found in the mitochondria. f3-hydroxybutyrate dehydrogenase is a cytosolic enzyme in this tissue, and preferentially oxidizes the 1,-isomer of phydroxybutyrate. A portion of glutamate dehydrogenase activity is also cytosolic. The subcellular organization distribution of enzymes of intermediary metabolism in Cepaea hepatopancreas suggests adaptation to periodic anoxia. @ 1996 Wiley-Liss, Inc. The study of intermediary metabolism in gastropod molluscs has been generally biased toward carbohydrate-based metabolism in marine species (see Livingstone and de Zwann, '83 for review). The metabolism of terrestrial snails is believed t o be carbohydrate based, and carbohydrate metabolism has been shown t o play a key role in sustaining estivation (Rees and Hand, '90, '93). However, relatively little is known about the participation of other energy substrates in the intermediary metabolism of gastropods, especially in non-marine species. The organization and roles of lipid and ketone body metabolism in these organisms have not been examined in detail. A single study by Meyer et al.('86) examined the metabolism of ketone bodies, acetoacetate (Am),and p-hydmqbutyrak (BHB), in the freshwater gastropod Biomphalaria glabruta. High activities of enzymes of ketone body metabolism in various tissues and hemolymph levels of Acac and BHB suggest a significant role for ketone bodies in routine energy metabolism in B. glabruta. In the hepatopancreas of this freshwater snail, the enzymes of ketone body metabolism including 0-hydroxybutyrate dehydrogenase (BHBDH),are localized within the mitochondria (Meyer et al., '86). 0 1996 WILEY-LISS, INC. In preliminary studies of ketone body metabolism in the terrestrial snail Cepaea nernoralis, we found that BHBDH, which in virtually all animals exists bound to the inner membrane of the mitochondrion, is cytosolic in hepatopancreas. We thus further examined the subcellular distribution of enzymes of ketone body metabolism in this tissue. To determine the relative importance of ketone body metabolism in the intermediary metabolism of this terrestrial gastropod, we also measured the activities and subcellular distributions of enzymes of the tricarboxylic acid cycle, the electron transport chain, and carbohydrate, amino acid, and lipid metabolism in hepatopancreas. MATERIALS AND METHODS Experimental animals C. nernoralis were collected on the University of Guelph campus in the spring of 1994. Snails were kept in a laboratory terrarium at 20" f 2°C for up t o 1week prior to experimentation and fed Received August 22, 1995; revision accepted January 15, 1996. Address reprint requests to J.S. Ballantyne, Department of Zoology, University of Guelph, Guelph, Ontario, N1G 2W1, Canada. 292 J.A. STUART AND J.S. B A L L A ” E ments Inc., Saddlebrook, NJ). Rates of reactions involving NADH were followed at 340 nm (millimolar extinction coefficient &a40 = 6.22). Reactions involving 5,5’-dithio-bis-(2-nitrobenzoicacid) Tissue preparation (DTNB) were followed at 412 nm (millimolar exSnails were deshelled, the hepatopancreas was tinction coefficient ~ 4 1 213.6). Acetoacetyl-CoA excised, a n d t h e intestine was removed and thiolase, 3-oxoacid transferase, and hydroxyplaced immediately in 3 mL of ice-cold mito- methylglutaryl-CoA synthase were measured by chondrial isolation buffer containing 20 mM N- following the accumulation or disappearance of [2-hydroxyethyl]piperazine-N-[2-ethanesulfonic acetoacetyl-CoA at 303 nm (millimolar extinction acid] (HEPES) and 100 mM sucrose, pH 7.5. All coefficient &303 13.8). Enzyme activities are extissue preparation procedures were carried out on pressed as units per gram of wet tissue wet where ice, Hepatopancreas was homogenized with three 1 unit equals 1 pmol of substrate converted to passes of a Potter-Elvejhem homogenizer with a product per minute. For convenience of compariTeflon pestle attached to a drill press, operating son, enzyme activities measured in isolated mitoat c 100 rpm. The homogenate was centrifuged chondria were also expressed as activities per at 200g for 10 minutes to remove cell debris and tissue wet weight. All enzyme assays were buffnuclei. A 1 mL sample of this supernatant, the ered in 50 mM imidazole, adjusted to appropriate “whole homogenate,” was taken and treated as pH with KOH and HC1. The following optimal asdescribed below. The remaining homogenate was say conditions were determined with respect to centrifuged at 10,OOOg for 10 minutes. The result- substrate and cofactor concentrations: a n t pellet, the “mitochondrial fraction,” was resuspended in 1 mL of mitochondrial isolation TCA cycle: buffer and treated as described below. This fracMalate dehydrogenase (MDH) (E.C. 126.96.36.199): tion contained virtually all mitochondria, based 0.15 mM NADH, 5 mM oxaloacetate (omiton activities of cytochome C oxidase, and citrate ted for control), pH 7.5 synthase. The supernatant from this step was cenCitrate synthase (CS) (E.C. 188.8.131.52): 0.1 mM trifuged at 40,OOOg for 10 minutes to remove nonDTNB, 0.3 mM acetyl-CoA, 0.5mM oxaloacsoluble materials; it yielded a pellet too small to etate (omitted for control), pH 8.0 use for enzyme determinations. The resultant su- Electron transport chain Cytochrome c oxidase (CCO) (E.C. 184.108.40.206): 50 pernatant was considered the “cytoplasmic fraction.” A 1mL sample of this fraction was collected. pM reduced cytochrome C (omitted for control), pH 8.0 All fraction samples were sonicated with five bursts of 5 seconds at 60% power output, 25 W, Lipid metabolism on a Vibra-Cell sonicator (Sonics & Materials Inc., 3-Hydroxyacyl-CoA dehydrogenase (HOAD) (E.C. 220.127.116.11): 0.2 mM NADH, 0.1 mM aceDanbury, CT) and subsequently used in enzyme assays. toacetyl-CoA (omitted for control), pH 8.0 Carnitine palmitoyl transferase (CPT) (E.C. Enzyme assays 18.104.22.168): 0.1 mM DTNB, 50 pM palmitoylCoA, 100 mM KCl, 5 mM L-carnitine (omitOptimal substrate and cofactor concentrations were determined for all enzymes assayed with the ted for control), pH 8.0 Carnitine octanoyl transferase (COT) (E.C. supernatant of a crude homogenate of C. nemoralis 22.214.171.124): 0.1 mM DTNB, 50 pM octanoylhepatopancreas. For these determinations, hepatoCoA, 100 mM KC1, 5 mM L-carnitine (omitpancreases of deshelled snails were excised, placed in 2 mL of ice-cold 50 mM imidazole, homogenized ted for control), pH 8.0 Malic enzyme (ME) (E.C. 126.96.36.199): 1.0 MgC12, with three passes of a Polytron PTlO tissue ho0.4 mM NADP, 1mM of either D- or L-malate mogenizer (Knematica Gmbh., Lucerne, Switzerland), and centrifuged at 10,OOOg. (individually) (omitted for control), pH 7.5 ATP-citrate lyase (ATPCL) (E.C. 188.8.131.52): 20 mM Maximal enzyme activities were determined usMgC12, 0.1 mM NADH, 1 mM DTT, 0.4 mM ing a Hewlett Packard HP8452 diode array specCoA, 10 mM ATP, 25 units MDH, 20 mM citrophotometer (Hewlett Packard, Mississauga, trate (omitted for control), pH 8.0 Ontario, Canada), equipped with a thermostated Isocitrate dehydrogenase (ICDH) (E.C. 184.108.40.206): cell changer maintained at 20°C with a Haake D8 5 mM MgC12, 0.4 mM nicotinamide adenine circulating water bath (Haaek Buchler Instrulettuce ad libitum. Humidity within the terrarium was kept between 50% and 100%with moist sphagnum moss. KETONE BODY METABOLISM IN C. N E M O U I S diphosphate (NADP), 0.5 mM threo-DsLsisocitrate (omitted for control), pH 8.0 Ketone body metabolism Acetoacetyl-CoA thiolase (AAT) (E.C. 220.127.116.11): 10 mM MgC12, 0.16 acetoacetyl-CoA, 0.2 mM CoA (omitted for control), pH 8.0 3-Oxoacid-CoA-transferase: 0.4 mM succinylCoA, 5 mM MgC12, 5 mM iodoacetamide, 10 mM acetoacetate (omitted for control), pH 8.0 BHBDH (E.C. 18.104.22.168): 2 mM DTT, 2 mM NAD, 40 mM DL-P-hydroxybutyrate (omitted for control), pH 8.0 Hydroxymethylglutaryl-CoA synthase (HMS) (E.C. 22.214.171.124) (modified from Quant et al., '89): 10 mM MgC12,2 mM DTT, 5 mM acetyl phosphate, 10 pM acetoacetyl-CoA, 100 pM acetylCoA and 10 units phosphotransacetylase (PTA) (both omitted for control), pH 8.0 Hydroxymethylglutaryl-CoA lyase (HML) (E.C. 126.96.36.199) (modified from Ballantyne and Berges, '91): 10 mM MgC12, 0.2 mM NADH, 1 unit BHBDH, 0.2 mM HMG-CoA (omitted for control), pH 8.0 Amino acid metabolism Glutamate dehydrogenase (GDH) (E.C. 188.8.131.52): 250 mM ammonium acetate, 0.1 ,mM EDTA, 0.05 mM NADH, 1 mM ADP, 7 mM a-ketoglutarate (aKG) (omitted for control), pH 8.0 Glutamate-pyruvate transaminase (GPT) (E.C. 184.108.40.206): 200 mM L-alanine, 0.2 mM NADH, 0.025 mM pyridoxal-5'-phosphate, 1unit lactate dehydrogenase (LDH), 10.5 mM aKG (omitted for control), pH 7.5 Glutamate-oxaloacetate transaminase (GOT) (E.C. 220.127.116.11): 0.2 mM NADH, 0.25 mM pyridoxal-5'-phosphate, 30 mM aspartate, 1 unit MDH, 3.5 mM aKG (omitted for control), pH 7.5 Carbohydrate metabolism LDH (E.C. 18.104.22.168): 0.4 mM NADH, 1 mM pyruvate (omitted for control), pH 7.5 Pyruvate kinase (PK) (E.C. 22.214.171.124): 10 mM MgC12,0.4 mM NADH, 1.3 mM ADP, 2 units LDH, 10 mM phosphoenolpyruvate (omitted for control), pH 7.5 Hexokinase (HK) (E.C. 126.96.36.199): 10 mM MgC12, 0.16 mM NADP, 1mM ATP, 2 units glucose6-phosphate dehydrogenase, 10 mM D-glucose (omitted for control), pH 7.5 Glucose-6-phosphate dehydrogenase (G6PDH) (E.C. 188.8.131.52): 7 mM MgC12,0.4 mM NADP, 1 mM glucose-6-phosphate (omitted for control), pH 7.5 a-Glycerophosphate dehydrogenase (aGPDH) 293 (E.C. 184.108.40.206): 0.2 mM NADH, 0.4 mM dihydroxyacetone phosphate (omitted for control), pH 8.0. The maximal activities of BHBDH with the Dor L-stereoisomers were determined. BHBDH activity with each isomer is the mean of six measurements, determined on homogenates prepared as described above. RESULTS The levels of enzymes of various metabolic pathways and their proportional distribution between mitochondria and cytosol provide a qualitative view of the metabolic organization of C. nemoralis hepatopancreas. The activities of all enzymes measured in the hepatopancreas are presented in Table 1.The percentage of this activity recovered in the summed mitochondrial and cytosolic fractions, and the percentage of this amount appearing in either the mitochondrial or cytosolic fractions individually, are presented in Table 2. TABLE 1 . Maximal activities of enzymes in C. nemoralis heDatoDancreas' Enzvme Whole homogenate activitv TCA cycle cs ICDH MDH Oxidative metabolism CCO Ketone body metabolism HMS HML AAT 30AT BHBDH Lipid metabolism COT CPT HOAD ATPL ME Amino acid metabolism GDH GPT GOT Carbohydrate metabolism PK HK LDH G6PDH a-GPDH 0.629 & 0.089 8.848 f 3.787 40.718 f 7.535 2.676 2 0.432 ND ND 3.909 2 0.810 0.795 2 0.202 0.825 2 0.096 0.032 2 0.020 0.149 2 0.043 0.296 0.051 0.067 2 0.025 ND 0.198 2 0.051 3.263 2 0.933 11.653 +: 0.830 0.420 2 0.051 0.298 f 0.024 4.264 f 1.325 2.072 f 0.761 1.394 2 0.216 'Activities (pmollminig wet tissue weight) are means of six separate determinations, f SE. ND, not detected. J.A. SWARTAND J.S. B A L , L A " E 294 TABLE 2. Recovery of whole homogenate enzyme activities in summed mitochondrial and cytosolic fractions and distribution of enz.yme activities between mitochondrial and cytosolic fractions' % of recovered recovered in summed fractions activity in mitochondrial fraction 9k of recovered activity in cytosolic fraction 104.9 9.1 87.8 9.2 76.2 f 11.9 90.6 f 2.1 22.9 0.9 17.7 1.0 9.4 f 2.1 77.1 -1- 0.9 82.3 2 1.0 98.9 r 6.6 98.2 2 6.6 1.8 2 0.3 ND ND 133.7 f 17.6 108.7 2 11.9 150.6 f 37.0 ND ND 33.3 f 6.5 92.8 f 1.4 0.0 ND ND 66.7 -1- 6.5 7.2 2 1.4 100.0 100.0 57.0 2 18.0 107.8 2 29.0 200.4 2 65.1 ND 100.0 100.0 83.4 2 8.0 74.3 2 11.5 ND 0.0 0.0 16.6 2 8.0 25.7 11.5 ND 129.2 2 30.7 100.1 f 3.8 96.3 7.8 59.4 2 8.4 28.6 3.1 23.1 1.3 40.6 f 8.4 71.4 f 3.1 76.9 2 1.3 117.0 f 10.9 84.9 f 14.6 103.5 f 15.4 92.4 27.5 88.4 f 7.2 0.0 8.8 4.2 3.4 f 0.9 10.5 0.3 2.1 f 0.6 100.0 90.2 f 4.8 96.6 z 0.9 89.5 0.3 97.9 f 0.6 % Activity Enzyme TCA cycle cs ICDH MDH Oxidative metabolism cco Ketone body metabolism HMS HML AAT 30AT BHBDH Lipid metabolism COT CPT HOAD ATPL ME Amino acid metabolism GDH GPT GOT Carbohydrate metabolism PK HK LDH G6PDH uGPDH 'Values represent average percent recovery of six separate determinations. ND, not determined. TCA cycle Enzymes that can provide substrate and reducing equivalents (NADPH) to fuel cytoplasmic lipid CS and MDH activities are detectable in the synthesis were examined. Malic enzyme, which mitochondrial fraction of the hepatopancreas. Alcatalyzes the formation of pyruvate from malate though CS appears t o be exclusively mitochondrial with the concomitant generation of NADPH, is un(91% in mitochondrial fraction), most of the MDH detectable in all cellular fractions. NADP+-depenactivity is cytosolic (82%). dent ICDH also provides NADPH for cytosolic lipid synthesis. The relatively high ICDH activity is priElectron transport chain marily cytosolic, with a small proportion of total CCO activity is exclusively mitochondrial, with activity detected in the mitochondrial fraction. 98% of its activity in this fraction. ATPCL catalyzes what is believed to be an important step in lipid synthesis in some biological sysLipid metabolism tems (Newsholme and Leech, '83), the production The enzymes involved in the transport and of acetyl-CoA from citrate produced in and exported P-oxidation of fatty acids are substantially mi- from the mitochondria. ATPCL is present in approxitochondrial. CPT and COT, which catalyze the mately equal proportions in the mitochondrial and the carnitine-dependent transport of long and me- cyt~solicfractions. The summed activities in these two dium chain fatty acids, respectively, into the mi- fractions is about twice that measured in the whole tochondrion, are not detectable in the cytosolic homogenate. The basis for this is not known. fraction. HOAD catalyzes the P-oxidation of fatty Ketone body metabolism acids. HOAD is primarily a mitochondrial activActivities of BHBDH, AAT, and 30AT are of the ity, with 16.6% of recovered activity appearing in the cytosolic fraction. same order of magnitude as those of key enzymes KETONE BODY METABOLISM IN C.N E M O W Z S (PK and HK) of carbohydrate metabolism. BHBDH, normally a mitochondrial membrane-bound enzyme, is exclusively cytosolic in C. nemoralis hepatopancreas, with no detectable mitochondrial activity. 30AT, which catalyzes the reversible oxidation of acetoacetate, is mitochondrial, with less than 10% appearing in the cytosolic fraction. AAT produced acetoacetyl-CoA from acetyl-CoA. AAT activity is largely cytosolic, with about 30% of total activity appearing in the mitochondrial fraction. HMS and HML catalyze the synthesis of acetoacetate from acetoacetyl-CoA in mammals, through the transient production of hydroxymethylglutaryl-CoA. Neither of these enzymes is detectable in any cellular fraction. The validity of the assays was verified using rat and mouse liver (data not shown). Amino acid metabolism The amino acid transaminases, GPT and GOT, have high proportions of cytoplasmic activity, with lesser activities appearing in the mitochondrial fraction. Activity of GOT is greater t h a n GPT, GDH, which catalyzes the oxidative deamination of glutamate to a-ketoglutarate, is approximately equally distributed between mitochondrial and cytosolic fractions. Carbohydrate metabolism HK, which catalyzes the first reaction in the intracellular metabolism of glucose, is present in the cytosol, with minor contamination of the mitochondrial fraction. PK, which catalyzes what is considered to be one of the rate-limiting steps in glycolysis, the production of pyruvate from phosphoenolpyruvate, is detectable exclusively in the cytosol. Virtually no LDH activity is seen in the mitochondrial fraction. All LDH activity measured in the whole homogenate is recovered in the cytosol. aGPDH is involved in the cytoplasmic arm of the a-glycerophosphate shuttle, which participates in balancing mitochondrial and cytoplasmic redox. The activity of this enzyme is also virtually exclusively cytosolic. G6PDH catalyzes the first committed reaction of the pentose phosphate pathway and generates NADPH, which can be used to drive lipid synthesis. GGPDH is entirely cytoplasmic in C. nemoralis hepatopancreas. Stereospecificity of BHBDH for D- or L-BHB Maximal BHBDH activity with L-BHB was optimized at a n L-BHB concentration of 20 mM. The mean activity with the L-isomer was 1.758 2 0.235 pmoVmidg wet tissue weight. Optimal BHBDH activity with D-BHB was achieved with 40 mM 295 D-BHB. Mean activity with the D-isomer was 0.515 f 0.039 ymol/min/g wet tissue weight. DISCUSSION As the major storage site for lipid (Voogt, '831, the hepatopancreas plays a n important role in the energy metabolism of terrestrial snails (Heeg, '77; Krupanidhi et al., '78). Although the tissue itself is capable of lipid catabolism, as we have demonstrated, it is unclear if lipid stored in the hepatopancreas can be transported to other tissues. The most metabolically active transport form of lipid in many organisms are non-esterified fatty acids (NEFAs). The very low solubility of NEFAs i n aqueous solutions requires that they be transported bound t o a carrier. There is no evidence of a fatty acid binding protein in gastropod molluscs, and the transport of fatty acids in hemolymph may not contribute substantially to mollusc energy metabolism (Allen, '77). Lipid carbon stored in hepatopancreas may therefore be transported in another, more soluble, form. In some animals this is achieved through the partial oxidation of lipid in the hepatopancreas to more freely soluble metabolic intermediates, like ketone bodies, which can be readily exported t o peripheral tissues (Zammit and Newsholme, '79). This may be a strategy for distribution of stored energy reserves in Cepaea. Based on enzyme activities, the metabolism of ketone bodies in the hepatopancreas is substantial. In Cepaea hepatopancreas, activities of enzymes of ketone body metabolism are of the same order of magnitude as those of carbohydrate metabolism. In most other animals these activities are only about 10% of activities of key enzymes of carbohydrate metabolism (Newsholme and Leech, '83j. Similarly, the concentration of BHB in fed B. glabrata hemolymph (0.6 mM) (Meyer et al., '86) is comparable t o that of glucose (0.62 mM) (see Livingstone and de Zwaan, 83). This suggests that ketone bodies are possibly as important as carbohydrates in the intermediary metabolism of freshwater and terrestrial gastropods. The data of Meyer et al. ('86) support a role for ketone bodies in routine energy metabolism in B. glabrata, as opposed to their more specialized role i n starvation in most other animals (Newsholme and Leech, '83). The organization of the pathways of ketone body metabolism in C. nemoralis differs from that of mammals in several respects. In most mammals, ketogenesis occurs primarily through reactions catalyzed by HMS and HML, whereas oxidation occurs through 30AT (Newsholme and Leech, '83). HMS and HML are undetectable in both the mitochon- J.A. STUART AND J.S.BALLANTYNE 296 BHB t-- A CYTOSOL Acac c-- C E I E Q s oxaloacetatew glucose 4- Fig. 1. Proposed organization of aerobic metabolism, such as could occur following a bout of anoxia, in C. nemoralis hepatopancreas. (For clarity, some enzymes and intermediates have been omitted.) Intermediates in bold letters indicate anaerobic endproducts and importediexported substrates. Note particularly that anaerobic endproducts lactate and alanine are reoxidized by LDH and a GPTIGDH couple, respec- tively. NADH generated in these processes is reoxidized by cytosolic BHBDH and gluconeogenesis, and redox balance is maintained. Pyruvate incorporation into glucose andor glycogen is maximized by: 1) exchange with Acac; 2) maintenance of high mitochondrial redox, promoting malate export; and 3) low mitochondrial concentrations of oxaloacetate, promoting synthesis of Acac for export from fatty acids. Abbreviations AAT aKG Acac BHB BHBDH CPT GOT acetoacetyl-CoA thiolase a-ketoglutarate acetoacetate P-hydroxybutyrate P-hydroxybutyrate dehydrogenase carnitine palmitoyl transferase glutamate-oxaloacetate transaminase GDH GPT HOAD LDH MDH 30AT glutamate dehydrogenase glutamate-pyruvate transaminase L-3-hydroxyacyl-CoAdehydrogenase lactate dehydrogenase malate dehydrogenase 3-oxoacid-CoAtransferase drial and cytoplasmic fractions of Cepaea hepato- tively prevents the loss of acetoacetyl-CoA via the pancreas, indicating that this pathway of ketone combined actions of acetyl-CoAhydrolase and AAT, body synthesis is not present. The lack of HMS and by virtually eliminating the accumulation of CoA HML in Cepaea hepatopancreas contrasts with the through an acetyl-CoA regenerating system. The results of Meyer et al. ('86), who measured HMS assay used by Meyer et al. ('86) did not account for and HML activities in the hepatopancreas of the acetoacetyl-CoA lost through the latter pathway and freshwater snail B. glabratus. We assayed HMS us- therefore may have substantially overestimated the ing the method of Quant et al. ('891, which effec- activity of HMS in B. glabrata. KETONE BODY METABOLISM IN C. NEMORALIS Ketone body synthesis and degradation do not depend on the presence of HMS and HML. Acac can also be synthesized through the reversal of the reaction catalyzed by 30AT (Fig. 1). Correspondingly, the levels of 30AT are threefold higher in Cepaea hepatopancreas than in rat liver (0.21 pmol/min/g) (Zammit and Newsholme, ,791, which may indicate a major role for this enzyme in ketogenesis. A ketogenic role for 30AT has been suggested in fish (Philips and Hird, '771, where activity of this enzyme is also quite high, for example, 12 pmol/min/g in bass (Zammit and Newsholme, '79). In starved elasmobranch fish, an increase in blood ketone body levels is paralleled by a threefold increase in the activity of liver 30AT (Zammit and Newsholme, '79). Thus, ketogenesis in Cepaea hepatopancreas must be catalyzed by AAT, 30AT, and BHBDH (Fig. l),without the involvement of HMS or HML. Another important finding of the present study is the subcellular distribution of BHBDH, which is exclusively cytosolic in C. nemoralis hepatopancreas (Fig. 1).In most animals it is a mitochondrial enzyme, found associated with the inner mitochondria1 membrane (Wang et al., '88). A cytosolic BHBDH is, however, found also in ruminant liver and kidney (see Zammit, '90, for review). Zammit ('90) suggests that cytosolic BHBDH may promote gluconeogenesis from pyruvate under ketogenic conditions. Gluconeogenesis is the main source of glucose and glycogen in ruminants, as virtually no glucose is produced by ruminant gut (Heitmann et al., '87). Gluconeogenesis in Cepaea hepatopancreas, in the fed state, is probably less important than in ruminants. Gastropods possess a variety of carbohydrases, and transport of sugars from the gut has been demonstrated (Livingstone and de Zwaan, '83), suggesting that, unlike ruminants, gut is an important source of glucose in gastropods in the fed state. Gluconeogenesis is, however, very important during recovery from anoxia, when glycogen reserves are depleted (Livingstone and DeZwaan, '83). Cytosolic BHBDH may play a role in promoting gluconeogenesis in these situations. Most terrestrial snails have a well-developed anaerobic metabolism (Livingstone and DeZwaan, ,831,adapted to a life history that includes prolonged anoxia during hibernation (Oudejans and van der Horst, '74) and intermittent bouts of anaerobic metabolism during estivation (Barnhart and McMahon, '88). During periods of anoxia, terrestrial gastropods deplete their endogenous glycogen reserves and accumulate alanine and 297 large amounts of lactate in the hepatopancreas (Churchill and Storey, '89). Upon resumption of aerobic metabolism, both anaerobic endproducts are reoxidized to pyruvate, which may then be used t o regenerate glucose and glycogen (Fig. 1). Ketogenesis in Cepaea hepatopancreas from stored lipid (Fig. 11,occurring in parallel with the reincorporation of pyruvate into carbohydrates, would stimulate this process in two ways. First, it would provide intramitochondrial Acac for export, which would stimulate pyruvate uptake through the mitochondria monocarboxylate carrier (Zammit, '90). Second, cytosolic reduction of Acac t o BHB would aid in maintenance of cytosolic redox through reoxidation of NADH generated by reformation of pyruvate from lactate, via LDH, and from alanine, through the combined actions of GPT and cytosolic GDH (Fig. 1).Cytoplasmic redox might otherwise become too highly reduced following anaerobic bouts, as hepatopancreas must reoxidize endogenous pools of alanine and lactate, as well as pools of these substrates transported from other tissues via the hemolymph (Livingstone and DeZwann, '83; Churchill and Storey, '89) (Fig. 1). A role for lipid in anaerobic metabolism has been variously suggested (Oudejans and van der Horst, '74; Zs-Nagy, '79; DeZwaan, '83), and there is some evidence that fatty acids may function as terminal electron acceptors. In hepatopancreas of the estivating giant African snails Achatina, free fatty acid content cycles over 24 days of estivation, showing periods of increase and decrease (Umezerike and Iheanacho, '83). Interestingly, 0, consumption and COBemission show similar cyclic fluxes during estivation, as snails alternate between aerobic and anaerobic metabolism (Barnhart and McMahon, '88). These data support a possible biphasic role for lipids in estivation, with anaerobic synthesis providing an electron sink and aerobic partial oxidation, leading to the synthesis of ketone bodies. The cytoplasmic BHBDH of C. nemoralis hepatopancreas is also similar to that of ruminant liver in that it preferentially oxidizes the L-stereoisomer of BHB (Williamson and Kuenzel, '71). Oxidation of L-BHB may not be the result of non-specificity of the cytosolic BHBDH. A necessary step t o clarify the relative physiological importance of the L- and D-isomers is the measurement of hemolymph levels of both stereoisomers of BHB. Other aspects of the subcellular organization of C. nemoralis hepatopancreas indicating adaptation t o episodic anoxic conditions include the significant proportions of GDH, GPT, and GOT activities that occur in the cytosol (Hochachka, 298 J.A. STUART AND J.S.BALLANTYNE aKG CYTOSOL - NADH NAD+ +rucdnate Fig. 2. Proposed pathways for the generation of anaerobic endproducts in C. nenoralis hepatopancreas. (For clarity, some enzymes and intermediates have been omitted.) Intermediates in bold indicate accumulated anaerobic endproducts. Note particularly the cyclic fluxes of glutamate and aKG, and redox coupling. NADH produced through anaerobic glycolysis is reoxidized via LDH and via the cytosolic GPT/GDH couple. 'SO). The predominantly anaerobic rat tapeworm, Hymenolepis diminuta (Cestoda) has an exclusively cytosolic GDH (Mustafa et al., '78), with lactate and alanine among the accumulated anaerobic endproducts (Mustafa et al., '78). These authors have suggested that the shift of GDH from mitochondria to cytosol may facilitate the maintenance of cytosolic redox in H . diminuta. Similarly, the subcellular distributions of GDH and GPT in Cepaea hepatopancreas suggest that the balance of cytosolic redox in this tissue could be achieved through the reduction of pyruvate to alanine, via a coupled cytosolic GPT and GDH (Fig. 2). This arrangement would not deplete glutamate or aKG pools, since it provides cyclic fluxes of these metabolites. This pathway results in the net consumption of NH3 (Walsh and Henry, '911, which may be of further benefit in providing a proton sink, reducing the rate of acidification of the cytosol. This proposed strategy differs from that of anaerobic marine molluscs, which use a much larger intracellular aspartate pool (12 pmoVg wet weight in Busycon contrarium ventricle; Ellington, '81) to regenerate glutamate from aKG (Fig. 2). With aspartate pools an order of magnitude lower in tissues of terrestrial snails (0.85 pmoVg wet weight in 0. lactea hepatopancreas; Churchill and Storey, '89), glutamate pools could be thus sustained for only the early stages of anoxia. In summary, the organization of ketone body metabolism in the hepatopancreas of the terrestrial snail C. nemoralis differs from that of freshwater snails and most mammals and is similar in several respects t o that of ruminants. Based on our observations, we suggest that there are links between ketone body metabolism and gluconeogenesis and amino acid metabolism in this tissue. The subcellular distribution of enzymes of intermediary metabolism in this tissue are consistent with adaptations to periodic anoxia, which typify terrestrial snail life history. Further study of the tissue distribution of these enzymes and hemolymph concentrations of substrates are required t o define the roles of ketone bodies in the intermediary metabolism of these organisms. ACKNOWLEDGMENTS This research was funded through a Natural Sciences and Engineering Research Council of Canada (NSERC) operating grant t o J.S.B. LITERATURE CITED Allen, S.V.V. (1977) Interorgan transport of lipids in the blood of the gumboot chiton Cryptochiton stelleri (Middendorff). Comp. Biochem. Physiol., 57A341-46. Ballantyne, J.S., and J.A. Berges (1991) Enzyme activities of gill, hepatopancreas, mantle and adductor muscle of the oyster (Crassostrea virginica) after changes in diet and salinity. Can. J. Fish. Aq. Sci., 48:1117-1123. KETONE BODY METABOLISM IN C. N E M O M Z S Barnhart, M.C., and B.R. McMahon (1988)Depression of aerobic metabolism and intracellular pH by hypercapnia in land snails, Otala lactea. J. Exp. Biol., 138:289-299. Churchill, T.A., and K.B. Storey (1989)Intermediary energy metabolism during dormancy and anoxia in the land snail Otala lactea. Phys. Zool., 62:1015-1030. DeZwaan, A. (1983) Carbohydrate catabolism in bivalves. In: The Mollusca, Vol. 10, Metabolic Biochemistry and Molecular Biomechanics. P.W. Hochachka, ed. Academic Press, Toronto, pp. 137-175. Ellington, W.R. (1981) Energy metabolism during hypoxia in the isolated, perfused ventricle of the whelk, Busycon contrarium Conrad. J. Comp. Physiol. 142:457-464. Heeg, J. (1977)Oxygen consumption and the use of metabolic reserves during starvation and aestivation in Bulinus (Physopsis)africanus (Pulmonata: Planorbidae). Malacolagia, 16:549-560. Heitmann, R.N., D.J. Dawes, and S.C. Senseing (1987) Hepatic ketogenesis and peripheral ketone body utilization in the ruminant. J . Nutr., 117:1174-1180. Hochachka, P.W. (1980) Living Without Oxygen: Closed and Open Systems in Hypoxia Tolerance. Harvard University Press, Cambridge. Krupanidhi, S., V.V. Reddy, and B.P. Naidu (1978) Organic composition of tissues of the snail Cryptozona ligulata (Ferussac) with special reference t o aestivation. Indian J. Exp. Biol., 16:611-612. Livingstone, D.R., and A. de Zwaan (1983)Carbohydrate metabolism of gastropods. In: The Mollusca, Vol. 1, Metabolic Biochemistry and Molecular Biomechanics. P.W. Hochachka, ed. Academic Press, Toronto, pp. 177-242. Meyer, R., W. Becker, and M. Klimkewitz (1986) Investigations on the ketone body metabolism in Biomphalaria glabrata: Influence of starvation and of infection with Schistosoma mansoni. J. Comp. Physiol., 156:563-571. Mustafa, T., R. Komlunieck, and D.F. Mettrick (1978) Cytosolic glutamate dehydrogenase in adult Hymenolepis diminuta (cestoda). Comp. Biochem. Physiol., 61B:219-222. Newsholme, E.A., and A.R. Leech (1983) Biochemistry for the Medical Sciences. John Wiley & Sons, New York. Oudejans, R.C.H.M., and D.J. van der Horst (1974)Aerobic-anaerobic biosynthesis of fatty acids and other lipids from glycolytic intermediates in the pulmonate land 299 snail Cepaea nemoralis (L.). Comp. Biochem. Physiol., 47B :139-147. Philips, J.W., and F.J.R. Hird (1977) Ketogenesis in vertebrate livers. Comp. Biochem. Physiol., 57B:133-138. Quant, P.A., P.K. Tubbs, and M.D. Brand (1989) Treatment of rats with glucagon or mannoheptulose increases mitosynthase activchondrial 3-hydroxy-3-methylglutaryl-CoA ity and decreases succinyl-CoA content in liver. Biochem. J., 262:159-164. Rees, B.B., and S.C. Hand (1990) Heat dissipation, gas exchange and acid-base status in the land snail Oreohelix during short-term estivation. J. Exp. Biol., 15277-92. Rees, B.B., and S.C. Hand (1993) Biochemical correlates of estivation tolerance in t h e mountainsnail Oreohelix (Pulmonata: Oreohelicidae). Biol. Bull., 184:230-242. Umezurike, G.M., and E.N. Iheanacho (1983) Metabolic adaptations in aestivating giant african snail (Achatina achatina). Comp. Biochem. Physiol., 74B:493-498. Voogt, P.A. (1983)Lipids: Their distribution and metabolism. In: The Mollusca, Vol. 10, Metabolic Biochemistry and Molecular Biomechanics. P.W. Hochachka, ed. Academic Press, Toronto, pp. 329-370. Walsh, P.J., and R.P. Henry (1991) Carbon dioxide and ammonia metabolism and exchange. In: Biochemistry and Molecular Biology of Fishes, Vol. 1,Phylogenetic and Biochemical Perspectives. P.W. Hochachka and T.P. Mommsen, eds. Elsevier, New York, pp. 181-208. Wang, S., E. Martin, J. Cimino, G. Omann, and M. Glaser (1988)Distribution of phospholipids around gramicidin and D-P-hydroxybutyrate dehydrogenase as measured by resonance energy transfer. Biochemistry, 27:2033-2039. Williamson, D.H., and P. Kuenzel (1971) The nature of the ‘cytoplasmic 3-hydroxybutyrate dehydrogenase’ from sheep kidney. Biochem. J., 121:169-570. Zammit, V.A. (1990)Ketogenesis in the liver of ruminantsadaptations t o a challenge. J. Agr. Sci., 115:155-162. Zammit, V.A., and E.A. Newsholme (1979) Activities of enzymes of fat and ketone-body metabolism and effects of starvation on blood concentrations of glucose and fat fuels in teleost and elasmobranch fish. Biochem. J., 184:313-322. Zs-Nagy, A. (1979) Cytosomes (yellow pigment granules) of molluscs as cell organelles of anoxic energy production. Int. Rev. Cytol., 49:331-377.