Combinatory effects of citrinin and ochratoxin A in immortalized human proximal tubule cells

Ochratoxin A (OTA) and citrinin (CIT) are two mycotoxins often occurring together in grains and cereals. Although both are nephrotoxic and can induce apoptosis, combination effects have not been examined up to now. Therefore, the aim of this study was to take a close look at the interactions of citrinin and OTA in cultured human proximal tubule-derived cells (IHKE cells). The cytotoxicity of both mycotoxins was studied, measuring the metabolic activity and the cell number. Furthermore, caspase 3-activation as a marker for apoptosis was examined for both mycotoxin alone and in combination. The results show that citrinin had an antagonistic effect on ochratoxin A induced caspase 3-activation in concentrations of 2.5 and 5 μmol/l. Higher concentrations (7.5 and 15 μmol/l) lead to additive effects, lower citrinin concentrations (0.25 and 1 μmol/l) did not show any effect at all. The observed decrease in caspase 3-activity was specific for the combination with OTA, since the combination of citrinin with cisplatin did not show any effect. Citrinin did not influence of the OTA-induced apoptosis when added two hours after applying ochratoxin A. Also the combination of both toxins decreased the uptake of OTA into the cells which might be an explanation for the antagonistic effect of citrinin in certain concentrations. However, the transport into cells can not be the only explanation. so further examinations are necessary.

Effects of the mycotoxin fumonisin B(1) on cell death in human kidney cells and human lung fibroblasts in primary culture

Fumonisins are mycotoxins produced by Fusarium verticillioides. The toxic effects of fumonisin B(1) (FB(1)) at the cellular level consist of a mixture of both necrosis and apoptosis. We studied the effect of FB(1) in human lung fibroblasts (NHLF) and human kidney epithelial cells (RPTEC) in primary culture. Apoptotic and necrotic cell death, collagen and fibronectin secretion were determined mainly after 14 days' exposure. The protein content of NHLF and RPTEC cells was slightly increased after 14 days' exposure to low FB(1) concentrations (0.1 or 1 microm). Caspase-3 activity tended to increase in NHLF and to decrease in RPTEC cells with higher FB(1) concentrations after 14 days' exposure. LDH release was slightly decreased in both cell types after 14 days. Collagen I and III secretion was enhanced in NHLF cells. Collagen III was decreased in RPTEC. Collagen IV was not changed in both cell types. Fibronectin secretion was uninfluenced in RPTEC and interim increased in NHLF. Furthermore LC-MS/MS studies did not give any hints for a metabolism of FB(1). Therefore, the main risk of prolonged FB(1) exposure seems to be altered collagen secretion pattern.

Induction of apoptosis in cultured human proximal tubule cells by fumonisins and fumonisin metabolites

Fumonisin B1 (FB1) causes apoptosis in a variety of cell types and tissues but the apoptotic potential of other fumonisins and fumonisin metabolites has not been determined and the underlying mechanisms are not completely understood. In our studies we exposed human proximal tubule-derived cells (IHKE cells) to FB1, fumonisin B2 (FB2), fumonisin B3 (FB3), hydrolyzed fumonisin B1 (HFB1) and N-palmitoyl-hydrolyzed fumonisin B1 (N-Pal-HFB1) and investigated caspase-3 activation, chromatin condensation and DNA fragmentation. Exposure to 10 micromol/L FB1 for 24 h led to a significant increase in caspase-3 activity, chromatin condensation and to DNA fragmentation. All other tested compounds did not show any significant activation of caspase-3 activity nor chromatin condensation and DNA-fragmentation. Furthermore, we examined if a sphinganine accumulation is correlated with an induction of apoptosis in IHKE cells. Therefore we used a liquid chromatography/electrospray ionization-mass spectrometry(LC/ESI-MS)-method using phytosphingosine as an internal standard to determine sphinganine and sphingosine concentrations in IHKE cells. Whereas a significant increase of sphinganine (up to 7000% compared to control cells) was observed with all fumonisin-derivates, sphingosine levels nearly remained unchanged indicating that all substrates inhibited ceramide synthase effectively. These results demonstrate that all compounds let to increased sphinganine levels in IHKE cells but only FB1 was able to induce apoptosis. We conclude that the inhibition of the ceramide synthase is not per se a predictor whether or not fumonisins induce apoptosis.

Programmed cell death by interaction of ochratoxin A with other nephrotoxins

We determined the effects of co-exposure of OTA and other (potentially) nephrotoxic substances on renal tubule cells (IHKE, from the proximal tubule; MDCK-C7 cells, representing principal cells of the collecting duct) by measuring the activation of caspase-3, an enzyme with key roles in the process of programmed cell death or apoptosis.Substances used in combination with OTA included: cadmium, a known nephrotoxin with apoptotic effects, cisplatin (anticancer drug, also a known nephrotoxin), cyclosporin A (an immunosuppresive agent), H2O2 (generated during oxidative cell stress), amphotericin B (antifungal agent), and fumonisin B1. Because of the great number of possible combinations, a method based on the 96-well format was developed which allowed a fast and cost-efficient succesive measurement of caspase-3 activity and protein content in one well as well as necrotic effects in parallel.We found that cells responded differently to the various combinations: for example, 1 µM OTA combined with 100 µM cisplatin had an antagonizing effect on caspase-3 activation in IHKE cells but potentiating effects in MDCK-C7. DNA ladder formation confirmed the results. Necrotic effects were very small and additive. We conclude that the co-exposure of renal cells to OTA with other substances can enhance or reduce the apoptotic potential of one substance alone depending on the substance and on the cell line investigated. A "harmless" substance can thus convert to a potent cell toxic substance when combined with OTA or other mycotoxins.These first findings show the necessity to further investigate the combined effects of OTA or other mycotoxins with other substances or of mycotoxins with each other.

Liquid chromatography/electrospray ionisation-mass spectrometry method for the quantification of sphingosine and sphinganine in cell cultures exposed to fumonisins

Fumonisins, mycotoxins produced by Fusarium verticillioides, are potent inhibitors of the de novo sphingolipid biosynthesis via inhibition of the key enzyme ceramide synthase. The cellular response to a fumonisin exposure is obvious as an alteration of the ratio of the sphingoid bases sphingosine (SO) and sphinganine (SA). We developed a new column liquid chromatography/electrospray ionisation-mass spectrometry (LC-ESI-MS) method for the rapid, simultaneous and quantitative determination of these bases in cell cultures of immortalised human kidney epithelial cells (IHKE cells). For sample preparation, cell lysates were only diluted, centrifuged and directly used for LC-MS measurements. Quantification was carried out using phytosphingosine (PSO) as an internal standard. Detecting the protonated molecule [M+H](+) signals of SO (m/z 300) and SA (m/z 302) in the selected ion monitoring (SIM) mode, detection limits of 10 pg for SO (signal-to-noise ratio S/N=3:1) and 25 pg for SA (S/N=3:1) were established. The average recovery for SO and SA was higher than 90% for control IHKE-cells, respectively. The developed LC-ESI-MS method allows the sensitive, selective and rapid monitoring of sphingosine and sphinganine in cell matrices with a drastically reduced time for sample preparation.

Preliminary results on Ochratoxin A concentrations in blood of patients with various kidney diseases in Germany

Ochratoxin A (OTA) is supposed to induce renal diseases in man and animals and a correlation between renal diseases and OTA concentration in blood is suspected. Therefore, we measured OTA concentrations in blood of subjects suffering from various renal diseases as e.g. interstitial nephritis or mesangial proliferating glomerular nephritis (GN) and compared them with the blood concentration of healthy individuals. We found OTA in 87% of all samples. There was no significant difference between OTA concentrations of healthy individuals and patients but some renal diseases (e.g. chronic glomerular nephritis) showed increased numbers of samples containing more than 1.5 nmol/l OTA in sera. In contrast, in samples from patients suffering from membranous or focal-sclerotic glomerular nephritis no concentrations above 1.5 nmol/l were found. Our preliminary results show that OTA is abundant in nearly all serum samples but some renal diseases show increased numbers of samples with high (>1.5 nmol/l) OTA concentrations.

Nephritogenic ochratoxin A interferes with mitochondrial function and pH homeostasis in immortalized human kidney epithelial cells

The ubiquitous nephritogenic and carcinogenic fungal metabolite ochratoxin A (OTA) has been shown to interact with renal cell function at low nanomolar concentrations. This is possibly brought about through changes in cellular pH (pHc) homeostasis and mitochondrial function. We assessed the effect of nanomolar concentrations of OTA on pHc homeostasis and the possible involvement of mitochondria using immortalized human kidney epithelial (IHKE1) cells. Within seconds OTA evoked a decrease of pHc with a threshold concentration of 0.1 nmol/l, followed by a sustained alkalinization. Acidification was the same in bicarbonate and non-bicarbonate Ringer solution. When Ca2+ entry across the plasma membrane was prevented, virtually no OTA-induced pH changes could be observed. Inhibition of Na+/H+-exchange (NHE, Na+-free solution) and H+-ATPase (bafilomycin A1) did not reduce the OTA-induced acidification. By contrast, determination of NHE activity as a function of pHc revealed that OTA stimulates NHE (maximal flux increases) in a Ca2+-dependent manner. OTA exposure did not increase lactic acid production, indicating that anaerobic glycolysis was not enhanced. Inhibiting complexes I, III and IV of the mitochondrial electron transport chain (ETC) with rotenone, antimycin A and CN- prevented the OTA-induced acidification almost completely. Completely inhibiting F1FO-ATPsynthase with oligomycin reduced the effect of OTA by approximately equal 50%. In addition, OTA induced a hyperpolarization of the mitochondrial membrane potential (psim) in a Ca2+-dependent manner. Furthermore, OTA exposure resulted in a mitochondria-dependent increase of the cellular ATP content. We conclude that OTA activates mitochondria and NHE by interfering with cellular Ca2+ homeostasis. Stimulation of mitochondrial metabolism leads to enhanced "proton production". Anaerobic glycolysis is not enhanced.

The nephrotoxin ochratoxin A induces apoptosis in cultured human proximal tubule cells

To test the apoptotic potential of the nephrotoxic mycotoxin ochratoxin A (OTA), we exposed human proximal tubule-derived cells (IHKE cells) for various times to OTA concentrations close to those occurring during dietary exposure (from 2 to 100 nmol/L) and investigated caspase 3 activation, chromatin condensation, and DNA fragmentation. OTA induced a time- and concentration-dependent activation of caspase 3: concentrations as low as 5 nmol/L OTA caused a slight but significant increase in caspase 3 activity after 7 days of OTA exposure. Exposure to 10 nmol/L OTA for 72 or 24 h led to a significantly increased activity of caspase 3 in human proximal tubule-derived cells. Radical scavengers such as N-acetylcysteine had no effect on OTA-induced caspase 3 activation. Chelation of intracellular calcium with 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid tetrakis (acetoxymethylester) (BAPTA-AM) also showed no effect. Exposure to 30 nmol/L or more OTA led to DNA fragmentation and chromatin condensation in IHKE cells. Cultured renal epithelial MDCK-C7 and MDCK-C11 or OK cells also showed increased caspase 3 activity after OTA exposure. We conclude that exposure to low OTA concentrations can lead to direct or indirect caspase 3 activation and subsequently to apoptosis in cultured human proximal tubule cells and in other renal epithelial cell lines of different origins.

Ochratoxin A induces JNK activation and apoptosis in MDCK-C7 cells at nanomolar concentrations

Ochratoxin A (OTA) is a ubiquitous fungal metabolite with nephritogenic, carcinogenic, and teratogenic action. Epidemiological studies indicate that OTA may be involved in the pathogenesis of different forms of human nephropathies. Previously we have shown that OTA activates extracellular signal-regulated kinases 1 and 2, members of the mitogen-activated protein kinases (MAPK) family, in the C7-clone but not in the C11-clone of renal epithelial Madin-Darby canine kidney (MDCK) cells. Here we show that nanomolar concentrations of OTA lead to activation of a second member of the MAPK family, namely, c-jun amino-terminal-kinase (JNK) in MDCK-C7 cells but virtually not in MDCK-C11 cells, as determined by kinase assay and Western blot. Furthermore, OTA potentiated the effect of tumor necrosis factor-alpha on JNK activation. In parallel to its effects on JNK, nanomolar OTA induced apoptosis in MDCK-C7 cells but not in MDCK-C11 cells, as determined by DNA fragmentation, DNA ladder formation, and caspase activation. In addition, OTA potentiated the proapoptotic action of tumor necrosis factor-alpha. Our data provide additional evidence that OTA interacts in a cell type-specific way with distinct members of the MAPK family at concentrations where no acute toxic effect can be observed. Induction of apoptosis via the JNK pathway can explain some of the OTA-induced changes in renal function as well as part of its teratogenic action.

Ochratoxin A-binding proteins in rat organs and plasma and in different cell lines of the kidney

In order to detect cellular proteins which bind the mycotoxin ochratoxin A (OTA) we coupled OTA covalently to horseradish peroxidase (HRP). The peroxidase activity of the conjugate was used to detect these proteins in Western (ligand) blot analysis. Only signals caused by OTA binding to proteins were viewable. HRP alone detected no proteins and OTA-HRP binding could be inhibited by free OTA. Several proteins from the rat intestine, liver, spleen, and kidney were detected by OTA. Also rat plasma proteins bind OTA which confirms previous findings. In all renal cell lines investigated (MDCK-C11, OK, LLC-PK1, IHKE, and SKPT) there are several proteins which bind OTA. Comparison of the PonceauS stain on the nitrocellulose sheet with the signal obtained from OTA-HRP unveiled proteins with high specific OTA binding. Especially, proteins with molecular masses between 55 and 60 kDa, 40 and 45 kDa and 25 and 30 kDa showed OTA binding in all samples. OTA was partially displaced by aspartame and phenylalanine from some but not all proteins. Binding to cytosolic and organellar proteins was comparable in all investigated cell lines. In the OK cell organellar compartment a 62 kDa protein is preferentially detected by OTA-HRP although virtually no protein band is detectable. In conclusion we have found a method to clearly detect proteins which bind OTA. With this new method we proved that OTA has the potential to bind to several proteins yet specific binding differs dramatically. Thus, highly specific binding of OTA possibly makes certain proteins a preferential target of OTA toxicity. Furthermore, binding contributes to intracellular accumulation of OTA, thus leading to a prolonged half life in the mammalian body and emphasises the toxic potential of this fungal metabolite.

Apical uptake of radiolabelled ochratoxin A into Madin-Darby canine kidney cells

Uptake of ochratoxin A (OTA) across the apical cell membrane of collecting duct cells is the first step in reabsorption and partly mediated by proton-dipeptide cotransport. As the remaining part of apical OTA uptake remained unclear, we studied the characteristics of apical uptake of tritium-labelled OTA (3H-OTA) in MDCK-C11 cells in detail. Uptake of 3H-OTA was pH- and temperature-dependent and led to intracellular accumulation of OTA. Lowering pH led to an increase and lowering temperature (4 degrees C) to a decrease of OTA uptake. Besides dipeptides, the beta-lactam antibiotics cephalexin and ceftibuten inhibited the 3H-OTA uptake also confirming the role of the proton dipeptide cotransporter. In addition, substrates of organic anion transporter, taurocholate and methotrexate, inhibited 3H-OTA uptake in part. Aspartylphenylalanine methyl ester (aspartame) had no inhibitory effect on 3H-OTA uptake. Uptake of OTA was not dependent on sodium. Sixty minutes of preincubation with phorbol 12-myristate 13-acetate (PMA) led to increased apical uptake of OTA. The PMA effects were inhibited by ethylisopropylamilorid (EIPA). We conclude that apical uptake of OTA occurs by Na+-independent transport. One part of the uptake is mediated by proton-dipeptide cotransport (30%, dipeptide-inhibitable), by organic anion transporter (20%, taurocholate-inhibitable) and by diffusion (20%, responsible for uptake at 4 degrees C). The remaining part occurs by as yet unidentified but pH-dependent transport mechanisms. An acidic urine in distal parts of the nephron provides thus the main risk for OTA uptake leading to its reabsorption and consequently alkalinisation of the urine should help to prevent this reabsorption.

Renal transepithelial secretion of ochratoxin A in the non-filtering toad kidney

Renal excretion is an important way of elimination for the nephrotoxin ochratoxin A (OTA). Because binding to proteins hinders filtration, excretion is mainly due to proximal tubular secretion. The goal of this study was to investigate the kinetics of secretion as well as the extent of urine and tissue accumulation in situ using the non-filtering amphibian kidney model. Transepithelial secretion is a saturable process (K(m) = 0.63.10(-6) mol/l) and leads to a concentration-dependent accumulation of OTA in the tubular lumen and in renal tissue. Maximum accumulation achieved is approximately 14-fold as compared to the perfusate concentration. There was no accumulation in the tubular lumen as compared to renal tissue (lumen-to-tissue concentration ratio approximately 1). Tissue and tubular lumen accumulation were reduced to approximately 40% of control in the presence of 10(-3) mol/l p-aminohippurate (PAH). Addition of 10(-3) mol/l alpha-ketoglutarate (KG) to PAH-containing perfusate did not lead to a further reduction of secretion. By contrast, addition of 10(-2) mol/l L-phenylalanine (L-Phe) reduced secretion further to approximately 25% of control. In the presence of 10(-3) mol/l probenecid tissue accumulation was reduced to 7% and tubular lumen accumulation to 1% of control. Lumen-to-tissue concentration ratio decreased to 0.15 in the presence of probenecid, indicating an inhibitory action at the luminal membrane. Addition of albumin to the perfusate, reduced secretion to only 50% of control, whereas the concentration of free OTA was reduced below 1% as compared to control. The results of this study show that transepithelial secretion is an effective way for accumulation of OTA in the tubular lumen and thus its urinary excretion. Transport via the basolateral organic anion and a basolateral amino acid carrier are the active steps in transepithelial secretion. Luminal exit of OTA is a passive process. Furthermore, tissue accumulation by the active transport across the basolateral membrane supports the toxic action of OTA on proximal tubular cells. Due to the qualitative similarity of organic anion transport our findings should also apply for the mammalian kidney.

Albumin endocytosis in OK cells: dependence on actin and microtubules and regulation by protein kinases

We used proximal tubule-derived opossum kidney (OK) cells to determine the dependence of albumin endocytosis on regulation by protein kinases and on the cytoskeleton. Uptake was observed only across the apical but not the basolateral membrane and exceeded uptake in collecting duct-derived Madin-Darby canine kidney cells 14-fold. Inhibition of endocytosis via clathrin-coated vesicles but not via caveolae abolished uptake. Cytochalasin D reduced uptake to < 5% of control, and inhibition of microtubule polymerization by nocodazole reduced uptake to approximately 55% of control. Activation of protein kinase A (PKA) by adenosine 3',5'-cyclic monophosphate, forskolin, or parathyroid hormone (PTH) reduced uptake to approximately 65% of control. Protein kinase C (PKC) activation did affect uptake to a similar extent as PKA activation but with a certain delay. Stimulation of PKG by guanosine 3',5'-cyclic monophosphate did not affect albumin endocytosis. The inhibitor of tyrosine kinases (TRK), genistein, induced an increase of uptake to approximately 160% of control. Reexocytosis of albumin was enhanced by PKC activation but not by PKA activation. TRK inhibition reduced the rate of reexocytosis. We conclude that albumin endocytosis in OK cells requires the integrity of the actin cytoskeleton. Microtubules facilitate endocytosis. Uptake is regulated by PKA, PKC, and TRK, yet with different time course and by different mechanisms, e.g., reexocytosis. Possibly TRK activity serves in a negative feedback loop to limit albumin endocytosis via a stimulation of reexocytosis.

Apical-to-basolateral transepithelial transport of Ochratoxin A by two subtypes of Madin-Darby canine kidney cells

In this study we investigated the transepithelial transport of Ochratoxin A (OTA), a potent nephrotoxin, across monolayers of two collecting duct-derived cells clones (Madin-Darby canine kidney cells (MDCK)-C7 and MDCK-C11 cells, resembling principal and intercalated cells, respectively) either from the apical to the basolateral side or vice versa. We cultured cells on permeable supports and compared the transport rates of OTA, p-aminohippuric acid (PAH) and fluorescein-labelled inulin. Monolayers of both cell clones translocated OTA from the apical to the basolateral side but not in the opposite direction. Transport rate across MDCK-C11 cell monolayers was 2.9-fold the transport rate across MDCK-C7 cell monolayers. OTA transport was temperature-dependent being reduced from 77.5 pmol/cm2 per h to 10.1 pmol/cm2 per h in MDCK-C11 and from 27.0 pmol/cm2 per h to 7.6 pmol/cm2 per h in MDCK-C7 cells when temperature was decreased from 37 degrees C to 4 degrees C. In both cell clones, the dipeptides carnosine and glycylsarcosine but not the amino acids glycine or phenylalanine had an inhibitory effect on OTA transport. In both cell clones, transepithelial transport of OTA was dependent on the apical pH (pK(a) of OTA = 7.1). In an environment mimicking the transepithelial in vivo pH gradient to some extent with more acidic pH on the apical side than on the basolateral side, transport was 4-fold higher in both cell clones as compared to conditions when pH was 7.4 in both bath solutions. In the absence of a pH gradient, transport rates were similar to that at 4 degrees C. Apical uptake of [3H]OTA was inhibited by carnosine and by glycylsarcosine and the uptake of [3H]carnosine was inhibited by OTA. Our results indicate that OTA is transported across the apical membrane of MDCK cells by both non-ionic diffusion and by a H+-dipeptide cotransporter. Thus, reabsorption of OTA in the collecting duct contributes to the observed long half life of OTA in the mammalian body.

Accumulation of ochratoxin A in rat kidney in vivo and in cultivated renal epithelial cells in vitro

In this study we determined the distribution of ochratoxin A (OTA) in renal tissue as well as its content and binding pattern in different cell types stemming from the proximal tubule (OK cells) and collecting duct (MDCK cells) of the kidney. To obtain the net amount of OTA in renal tissue, the relative amount in the vascular compartment was calculated and subtracted. After acute administration of OTA to male Wistar rats, the highest concentrations were detected in the papilla and inner medulla. Concentrations in the outer medulla and in the cortical tissue were 50% lower. Sub-chronic (6 day) exposure to OTA resulted in a similar distribution in young rats (6 weeks old) but in a different distribution in adult rats (2 year old). Cultured MDCK-C11 cells (representing intercalated cells) showed a higher OTA content after 48 h of exposure to 1 mumol/1 OTA than another MDCK cell subtype (MDCK-C7, representing principal cells) and OK cells. Additionally, the capability of cell extracts to bind OTA was the highest in MDCK-C11 cells followed by cell extracts of OK cells and cell extracts of MDCK-C7 cells. In two cell types studied, OK and MDCK-C11 cells, OTA was bound preferentially by organellar components and less by cytosolic ones.

Turnover and transformation of mitochondrial acetyl-CoA acetyltransferase into CoA-modified forms

Rat liver mitochondrial acetyl-CoA acetyltransferase (acetoacetyl-CoA thiolase, EC 2.3.1.9) exists additionally in the CoA-modified forms A1 and A2. After a pulse of radioactivity using [35S]methionine in hepatocytes, the highest radioactivity was obtained in the unmodified enzyme. Over the chase time, the radioactivity in the unmodified enzyme decreased, but simultaneously increased in both CoA-modified forms, thus proving that the fully active unmodified enzyme exists before the partially active modified forms A1 and A2. Also, the specific radioactivity (ratio % radioactivity/% immunoreactive area) of A1 > A2 demonstrates a sequential CoA modification of form A1 to form A2. Acetyl-CoA acetyltransferase was degraded with an apparent half-life of 38.0 h: the modified forms A1 and A2 have half-lives of 24.5 and 7.2 h. The physiological meaning of the CoA modification of acetyl-CoA acetyltransferase is not yet understood.

Identification of the CoA-modified forms of mitochondrial acetyl-CoA acetyltransferase and of glutamate dehydrogenase as nearest-neighbour proteins

A 52 kDa protein could only be co-purified with the CoA-modified forms of acetyl-CoA acetyltransferase (acetoacetyl-CoA thiolase) (EC 2.3.1.9) from rat liver mitochondria. Immunoprecipitations of these modified forms with anti-(acetyl-CoA acetyltransferase) IgG or anti-(52 kDa protein) IgG yielded, in addition to the appropriate proteins, the 52 kDa protein or the CoA-modified form of acetyl-CoA acetyltransferase (41 kDa) respectively. This was demonstrated by SDS/PAGE and immunoblots. The modified forms containing the 52 kDa protein could be cross-linked by 1,5-difluoro-2,4-dinitrobenzene to a high-molecular-mass complex containing both the 41 kDa and 52 kDa proteins. The 52 kDa protein was identified as mitochondrial glutamate dehydrogenase (EC 1.4.1.3) by amino acid sequence analysis. The results of co-immunoprecipitation and cross-linking characterize the CoA-modified forms of acetyl-CoA acetyltransferase and the glutamate dehydrogenase as nearest-neighbour proteins.