Application of liquid chromatography-nuclear magnetic resonance spectroscopy to the identification of natural products.

LC-NMR combines the separation power of high-performance liquid chromatography (HPLC) with the superior structural information content of nuclear magnetic resonance (NMR). These two techniques traditionally have been the primary tools used by natural products chemists to isolate and determine the structures of molecules of interest. Recent advances in NMR technology have allowed the practical application of LC-NMR, thus providing natural products chemists with a hyphenated technique which combines the two most important tools in their field. A brief review of the literature describing how LC-NMR has been applied to natural products research is followed by a specific example illustrating how this technique was used to identify the marine alkaloid aaptamine (1). Aaptamine was identified as the active component in the crude dichloromethane extract of the sponge Aaptos sp. which had been determined to possess inhibitory activity against the enzyme glutamine:fructose-6-phosphate amidotransferase (GFAT) by a high throughput screening (HTS) effort. Isolated aaptamine (1) exhibited an IC50 = 120 microM against this enzyme. The experience gained from the identification of aaptamine was used to define a strategy for the use of LC-NMR in a natural products HTS program.

LC-NMR: a new tool to expedite the dereplication and identification of natural products.

The rapid identification of known or undesirable compounds from natural products extracts - "dereplication" - is an important step in an efficiently run natural products discovery program. Dereplication strategies use analytical techniques and database searching to determine the identity of an active compound at the earliest possible stage in the discovery process. In the past few years, advances in technology have allowed the development of tandem analytical techniques such as liquid chromatography mass spectrometry (LC-MS), LC-MS-MS, liquid chromatography nuclear magnetic resonance (LC-NMR), and LC-NMR-MS. LC-NMR, despite its lower sensitivity as compared to LC-MS, provides a powerful tool for rapid identification of known compounds and identification of structure classes of novel compounds. LC-NMR is especially useful in instances where the data from LC-MS are incomplete or do not allow confident identification of the active component of a sample. LC-NMR has been used to identify the marine alkaloid aaptamine as the active component in an extract of the sponge Aaptos sp. This extract had been identified as an enzyme inhibitor by a high throughput screening (HTS) effort. Isolated aaptamine exhibited an IC(50)=120 µM against this enzyme. Strategies for the identification of aaptamine and for the use of LC-NMR in a natural products HTS program are discussed.

Inhibitory activity of unsaturated fatty acids and anacardic acids toward soluble tissue factor-factor VIIa complex.

Five compounds, which inhibited the amidolytic activity of soluble tissue factor/activated factor VII complex (sTF/VIIa), were isolated from two traditional Chinese medicinal plants commonly used in the treatment of cardiovascular and cerebrovascular diseases. The active compounds were found to be linolenic, linoleic, and oleic acids from roots of Salvia miltiorrhiza; and two anacardic acids, 6-(8'Z-pentadecenyl)- and 6-(10'Z-heptadecenyl)-salicylic acids, from leaves of Ginkgo biloba. The IC50 values were in the range 30-80 micromol/L. Palmitic acid, isolated from roots of Salvia miltiorrhiza, and 2-[(3',7',11',15'-tetramethyl)-2'E,6'E,10'E, 14'E-hexadecatetraenyl]-1,4-hydroquinone, isolated from the marine sponge Adocia viola, did not inhibit sTF/VIIa. Further expansion of the structure-activity relationship to include anacardic acids, 6-(8'Z,11'Z-heptadecadienyl)- and 6-(8'Z, 11'Z, 14'Z-heptadecatrienyl)-salicylic acids from leaves of Anacardium spondias, and other fatty acids demonstrated that at least one cis double bond was essential for inhibitory activity, and that fatty acids containing two or three cis double bonds were optimal. Evidence from preincubation studies implied that these fatty acids may exert their effect by binding to VIIa and consequently preventing binding of sTF to VIIa.

Potentiation of nitric oxide-mediated vascular relaxation by SC52608, a superoxide dismutase mimic.

Nitric oxide (NO) produced by the vascular endothelium is an endogenous contributor to the regulation of vascular relaxation and the maintenance of blood pressure. The effective half-life of NO and the relaxation of aortic rings by NO is enhanced by a reduction in the concentration of superoxide radicals with superoxide dismutase (SOD). In the current study, SC52608, a newly synthesized SOD mimic with a manganese core, was tested for its ability to potentiate the activity of NO both in vitro and in vivo. SC52608 relaxation of rat aortic segments was endothelium dependent as well as concentration dependent. The maximum relaxation following KCl contraction was 44% with 300 microM SC52608. Cyclic GMP concentrations in the segments were increased 1.6- and 3.2-fold with 5 and 300 microM SC52608, respectively. N-monomethyl-I-arginine pretreatment of aortic rings abolished the relaxation and cyclic GMP accumulation mediated by SC52608. In a smooth muscle cell reporter system of nitric oxide synthase activity, SC52608 potentiated the increase in cyclic GMP elicited by NO in a concentration-dependent manner with a maximum increase of 5.2-fold at 100 microM. Injection of SC52608 into conscious, restrained rats resulted in a dose-dependent decrease of blood pressure. Therefore, the data suggest that SC52608 potentiates the actions of nitric oxide on vascular tone, cyclic GMP, and blood pressure by enhancing the half-life of NO through a mechanism that mimics the action of SOD.

Potentiation of osteoclast bone-resorption activity by inhibition of nitric oxide synthase.

We have examined the effects of modulating nitric oxide (NO) levels on osteoclast-mediated bone resorption in vitro and the effects of nitric oxide synthase (NOS) inhibitors on bone mineral density in vivo. Diaphorase-based histochemical staining for NOS activity of bone sections or highly enriched osteoclast cultures suggested that osteoclasts exhibit substantial NOS activity that may account for basal NO production. Chicken osteoclasts were cultured for 36 hr on bovine bone slices in the presence or absence of the NO-generating agent sodium nitroprusside or the NOS inhibitors N-nitro-L-arginine methyl ester and aminoguanidine. Nitroprusside markedly decreased the number of bone pits and the average pit area in comparison with control cultures. On the other hand, NOS inhibition by N-nitro-L-arginine methyl ester or aminoguanidine dramatically increased the number of bone pits and the average resorption area per pit. In a model of osteoporosis, aminoguanidine potentiated the loss of bone mineral density in ovariectomized rats. Aminoguanidine also caused a loss of bone mineral density in the sham-operated rats. Inhibition of NOS activity in vitro and in vivo resulted in an apparent potentiation of osteoclast activity. These findings suggest that endogenous NO production in osteoclast cultures may regulate resorption activity. The modulation of NOS and NO levels by cells within the bone microenvironment may be a sensitive mechanism for local control of osteoclast bone resorption.

Manganese potentiation of nitric oxide-mediated vascular relaxation.

The half-life of nitric oxide (NO) and the relaxation of aortic-rings are enhanced by superoxide dismutase. Manganese and manganese-containing preparations have been reported to mimic superoxide dismutase activity. In the current study, manganese was tested for its ability to potentiate the activity of NO both in vitro and in vivo. Manganese relaxation of aortic segments was endothelium dependent as well as concentration dependent. Cyclic GMP concentrations in the segments were increased 2- and 4-fold with 5 and 300 microM manganese, respectively. N-Monomethyl-L-arginine pretreatment of aortic rings abolished the relaxation and cyclic GMP accumulation mediated by manganese. Infusion of manganese into conscious, restrained rats resulted in a decrease of blood pressure which was abolished by N-nitro-L-arginine pretreatment. Therefore, manganese may prolong the half-life of NO by a mechanism that mimics the action of superoxide dismutase resulting in potentiation of NO actions in vascular tissue.

Fructose 2,6-bisphosphate: changes during neonatal maturation and aging of rat and potential role in regulation of glucose utilization.

During the 6 days following birth, tissue levels of fructose-2,6-P2 in rat brain, liver, muscle, heart and kidney did not significantly change. However, by the tenth day postpartum fructose-2,6-P2 levels in brain, heart, and skeletal muscle increased approximately 50% and attained adult values. During maturation of liver, adult levels of fructose-2,6-P2 were not achieved until 3-4 weeks after birth or approximately at the time of maximum rates of gluconeogenesis. Renal fructose-2,6-P2 levels in the neonate were initially elevated and 2-3 weeks after birth decreased approximately 2.5-fold to adult values. With the exception of the pons-medulla, which showed no significant changes in fructose-2,6-P2 amounts, levels of this regulatory sugar from aging brain regions were generally decreased. The fructose-2,6-P2 levels from heart atria of old rats (24-30 month) were also significantly decreased. In diaphragm, the fructose-2,6-P2 levels were increased at 12 months of age and at 27 months of age were twice the level at 3 months. The fructose-2,6-P2 levels during the aging of liver, skeletal muscle (EDL and soleus), spleen, thymus, kidney, testis and lung were not significantly altered.

Regulation of brain 6-phosphofructo-1-kinase: effects of aging, fructose-2,6-bisphosphate, and regional subunit distribution.

Total 6-phosphofructo-1-kinase (PFK) activity, amounts of each type of PFK subunit, and levels of fructose-2,6-P2 in the cerebral cortex, midbrain, pons-medulla, and cerebellum of 3, 12, and 25 month rats were measured. Further, the role of fructose-2,6-P2 in the regulation of brain PFK activity was examined. A positive correlation was found to exist between the reported losses of glucose utilization as measured by 2-deoxy-D-glucose uptake and PFK activity in each region. That is, both parameters decreased to their lowest level by 12 months of age and remained decreased and fairly constant thereafter. Fructose-2,6-P2 levels did not appear to directly correlate with regional changes in glucose utilization. Also, region-specific and age-related alterations of the PFK subunits were found although these changes apparently did not correlate with decreased glucose utilization. Brain PFK is apparently saturated with fructose-2,6-P2 due to the high endogenous levels, and it contains a large proportion of the C-type subunit which dampens catalytic efficiency. Consequently, brain PFK could exist in a conformational state such that it can readily consume fructose-6-P rather than in an inhibited state requiring activation. This may explain, in part, the ability of brain to efficiently but conservatively utilize available glucose in energy production.

Selective inhibition of the inducible nitric oxide synthase by aminoguanidine.

Overproduction of the free radical nitric oxide (NO) has been implicated in the pathogenesis of a variety of inflammatory and immunologically mediated diseases as well as complications of diabetes. In the present study we have demonstrated that aminoguanidine selectively inhibits the cytokine-inducible isoform of NO synthase which appears to be responsible for the excess production of NO linked to these disease states. By using organ, cell, and enzyme-based measurements we have shown that aminoguanidine is equipotent to NG-monomethyl-L-arginine (L-NMA) as an inhibitor of the cytokine-induced isoform of NO synthase but is 10 to 100-fold less potent as an inhibitor of the constitutive isoform. Thus, aminoguanidine may be useful as a selective inhibitor of the inducible NO synthase in the treatment of disease states characterized by the pathological overproduction of NO.

Age-related changes in subunit composition and regulation of hepatic 6-phosphofructo-1-kinase.

6-Phosphofructo-1-kinase (PFK) isoenzyme pools from livers of fetal, neonatal, young adult (3 months) and aged (24 months) rats were studied. Near-term liver PFK isoenzyme pools were composed of nearly equal quantities of all three subunits. During the 30 days after birth, the total activity increased by 25%; the amount of the L-type, M-type or C-type subunit was increased 3-fold, was unchanged, or was decreased by 80% respectively. In aged rats, compared with young adults, total PFK activity was unchanged, but the L-type, M-type or C-type subunit decreased by 24%, increased by 39%, or increased by 338% respectively. During neonatal maturation, the changing subunit composition of the hepatic isoenzyme pools led to a decreased susceptibility to ATP inhibition, to a greater apparent affinity for fructose 6-phosphate, and to increased sensitivity to fructose 2,6-bisphosphate. Also, these alterations correlated with the measured increases in fructose 2,6-bisphosphate and the reported optimal rate of hepatic glycolysis/gluconeogenesis.

Physiological relevance of the changing subunit composition and regulatory properties of the 6-phosphofructo-1-kinase isozyme pools during heart and muscle development.

During postnatal development, the subunit compositions of the 6-phosphofructo-1-kinase isozyme pools of heart and skeletal muscle are known to change. The isozyme pools from fetal muscle were composed of the L-type (60%), and M-type (36%) and C-type (4%) subunits and the isozymes from fetal and early neonatal heart contain nearly equal amounts of all three subunits. During postnatal development of both tissues, the proportion of the M-type subunit increases until it is the only type present in adult muscle and the major subunit in adult heart (75%). The isozyme pool from fetal muscle exhibit a decreased affinity for fructose-6-P and a greater susceptibility to ATP inhibition compared to the M-rich isozymes which are subsequently present. The isozyme pools from fetal and early neonatal heart, if compared to the M-rich isozymes which are present later during heart development and to the fetal muscle isozymes, exhibited the least affinity for fructose-6-P and the greatest susceptibility to ATP inhibition. Comparison of the isozyme pools containing little or no C-type subunit with those from fetal and early neonatal heart clearly indicates that the presence of substantial levels of the C-type subunit imposed a decreased ability for fructose-2,6-P2 to both lower affinity for fructose-6-P and antagonize sensitivity to ATP inhibition. Although still not thoroughly appreciated, it appears that the changing nature of the isozyme pools in these tissues permits regulation of glucose metabolism in a manner which allows efficient utilization of nutritional opportunities and which adequately meets the energy requirements of each tissue at different stages of development.

Physiological implications of the alteration of 6-phosphofructo-1-kinase isozyme pools during brain development and aging.

The 6-phosphofructo-1-kinase (PFK) isozyme pools from brains of fetal, neonatal, young adult (3 months) and aged (30 months) rats were studied using chromatographic and immunological techniques. Also, the changing subunit composition of each isozyme pool was determined by sodium dodecyl sulfate polyacrylamide gel electrophoresis on 6% slab gels and by immunoblotting with subunit-specific antibodies. The total PFK activity increased over seven-fold during the 30 days following birth, and the L-type, M-type, and C-type subunits increased approximately 2-fold, 7-fold, and 24-fold, respectively. In the near-term fetal brain and early neonatal brain, the L-type and M-type subunits were the predominant forms and were present in approximately equal amounts. During the second second week of postnatal brain maturation, the levels of the M-type and C-type subunit began to significantly increase. Consequently, during postnatal development, the isozyme pools switched from L-M-rich forms to M-C-rich forms. In aged brain relative to the young adult (3 months) brain, the 20% loss of total activity was associated with 27% and 18% losses of the M-type and C-type subunits, respectively. Examination of the regulatory properties of the various PFK isozyme pools revealed that at the low concentration of fructose-6-P and high level of ATP which are thought to occur in vivo, fructose-2,6-P2 was required for measurable PFK activity.(ABSTRACT TRUNCATED AT 250 WORDS)

Analysis of the phosphofructokinase subunits and isoenzymes in human tissues.

The 6-phosphofructo-1-kinase (PFK) subunits and isoenzymes were studied in human muscle, heart, brain, liver, platelets, fibroblasts, erythrocytes, placenta and umbilical cord. In each tissue, the subunit types in the native isoenzymes were characterized by immunological titration with subunit-specific antibodies and by column chromatography on QAE (quaternary aminoethyl)-Sephadex. Further, the subunits of the partially purified native isoenzymes were resolved by SDS/polyacrylamide-gel electrophoresis, identified by immunoblotting, and quantified by scanning gel densitometry of silver-stained gels and immunoblots. Depending on the type of tissue, one to three subunits were detected. The Mr values of the L, M and C subunits regardless of tissue were 76,700 +/- 1400, 82,500 +/- 1640 and 86,500 +/- 1620. Of the tissues studied, only the muscle PFK isoenzymes exhibited one subunit, which was the M-type subunit. Of the other tissues studied, the PFK isoenzymes contained various amounts of all three subunits. Considering the properties of the native PFK isoenzymes, it is clear that, in human tissues, they are not simply various combinations of two or three homotetrameric isoenzymes, but complex mixtures of homotetramers and heterotetramers. The kinetic/regulatory properties of the various isoenzyme pools were found to be dependent on subunit composition.

Nature of the subunits of the 6-phosphofructo-1-kinase isoenzymes from rat tissues.

The nature of the PFK (6-phosphofructo-1-kinase) isoenzymes in many rat tissues was examined by immunological and chromatographic techniques and by measurement of their subunit compositions. It was revealed that, except for diaphragm and skeletal muscle, these complex isoenzymic populations contained different amounts of the three subunit types and were nearly tissue-specific. Apparently this tissue specificity is due to different concentrations of the tetramers, which in turn are controlled by the types and amounts of each subunit that are available to associate randomly.

Alteration of 6-phosphofructo-1-kinase isozyme pools during heart development and aging.

The nature of 6-phosphofructo-1-kinase isozyme pools in fetal, neonatal, young adult (3 months), and aged (30 months) rat hearts was studied using chromatographic and immunological techniques. Furthermore, the changing subunit composition of each isozyme pool was determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis on 6% slab gels and by immunoblotting with subunit-specific antibodies. Although all three subunit types were expressed in heart throughout life, total activity and the nature of the isozyme pools varied during neonatal development and in aged heart. In fetal heart, the complex tetramers containing all three subunits appeared to be the major isozyme types. As the heart matured to the young adult stage, the M-type subunit increased over 6-fold; whereas the changes in the other two subunits were considerably less. These data indicate that during neonatal heart maturation the isozymic pools progressively exhibited increased amounts of the tetrameric forms containing two or more M-type subunits. In aged heart relative to the young adult (3 months) heart, the total activity and proportion of M-type subunit in the isozymes were decreased; and consequently, the amounts of the M-rich isozymes were decreased. The shifts in the types of isozymes during heart maturation and subsequent aging were primarily due to changes in availability of the M-type subunit to participate in random assembly of the tetrameric isozymes.

Regulation of skeletal muscle 6-phosphofructo-1-kinase during aging and development.

The purpose of this paper is to provide insight into the alterations of 6-phosphofructo-1-kinase (PFK) activity and isozyme types of rat skeletal muscle during development and aging. PFK isozymes are tetramers which may be comprised of one or any combination of the three subunit types, L, M, and C. The effects of fusion or terminal differentiation of cultured rat L6 myoblasts leading to formation of myotubles does not have a noticeable effect on total PFK activity. However, the amount of M-type subunit was increased; and the level of the C-type subunit decreased. These subunit changes caused shifts in the isozymic types. The ultimate effects of prenatal development of PFK were characterized in the near-term fetal muscle. This stage of development was accompanied by a significant loss of the C-type subunit and by two-fold increases in the L-type and M-type subunits which accounted for the 40% increase in total PFK activity. After birth, the M-type subunit increased dramatically as did the total PFK activity. Since the L-type and C-type subunits were gradually lost during the subsequent 3 weeks, the homotetramer of the M-type subunit (M4) was the only type which is present in mature muscle. M4 persisted as the only detectable form in skeletal muscle during the remainder of life, but the total PFK activity and amount of M4 decreased after 18 months of age. The decreased total PFK activity in aged skeletal muscle suggested that the expression of PFK genes may have reverted to an immature state when total PFK activity was lower. As shown by both the immunological analysis and direct quantification of subunit types, this clearly did not occur. That is, the loss of PFK activity in aged muscle is a consequence of decreased levels of the M-type subunit and not reappearance of other subunit types such as found in maturing muscle. Further, our examination of aged skeletal muscle indicates that no significant structural changes in M-type subunits had occurred and that inactive or partially active proteins which could crossreact with the M-type subunit were not detectable. It is suggested that the loss of M4 could cause a depression of the glycolytic rate leading to diminished ability of senile muscle to accommodate extreme energy demands.

Insulin-mediated regulation of heart atrial and ventricular 6-phosphofructo-1-kinase.

Atrial 6-phosphofructo-1-kinase activity from the hearts of diabetic rats was decreased by 50%, but ventricular 6-phosphofructo-1-kinase activity was found not to be insulin-sensitive. This decrease in atrial 6-phosphofructo-1-kinase activity during diabetes was characterized by diminished levels of all three types of 6-phosphofructo-1-kinase subunits. As shown by immunological titration and column chromatography, the population of native 6-phosphofructo-1-kinase isozymes in the ventricles was not measurably affected during insulin deprivation. However, the atrial isozyme population in diabetic rat heart appeared to contain, on a relative basis, higher levels of the isozymic forms containing the L-type subunit. Measurement of the levels of this subunit indicated that in diabetic atria it was less affected than the other subunits. In the ventricles, insulin deficiency did not promote significant losses of fructose-2,6-P2; but, in diabetic rats, the atrial levels of this activator were decreased by 80% and subsequently restored by insulin treatment. These data suggest that any insulin-mediated effects on ventricular 6-phosphofructo-1-kinase activity and resultant effects on ventricular glycolysis do not appear to be exerted through changes in enzyme concentration, but probably through changes in modulators other than fructose-2,6-P2. In contrast to the ventricles, it appears that insulin exerts its effects on atrial 6-phosphofructo-1-kinase activity and, in part, influences atrial glycolysis through alteration of fructose-2,6-P2 levels, enzyme concentration, and isozymic content.

Characterization of the rat heart 6-phosphofructo-1-kinase isozymes.

By elution with a discontinuous gradient from QAE-Sephadex, the rat muscle 6-phosphofructo-1-kinase (PFK) isozyme (M4) and the major rat liver PFK isozyme (L4) could be completely separated. Subjecting heart supernatant fluid to this treatment indicated that all of the heart PFK activity was found in the first wash, where M4 eluted, indicating little, if any, L4 was present. However, about 50% of the heart PFK activity was immunoprecipitated by L4 anti-IgG demonstrating the presence of the L-type subunit. A purification procedure was developed which yielded an enzyme preparation of high specific activity and resulted in a recovery of 50% to 60% of the original PFK activity. Three proteins were detected in this PFK preparation that exhibited apparent mol wt of 87 500, 85 000 and 80 000. The 85 000 and 80 000 Dalton components corresponded to the subunits of M4 and L4, respectively. The third protein was thought to be a distinct subunit (C-type) since it exhibited the highest molecular weight and was present in an exhaustively washed immunoprecipitate of purified heart PFK. From seven different heart PFK preparations, the relative distributions of the L-, M-, and C-type subunits were 0.80 +/- 0.1, 4.5 +/- 0.7, and 0.7 +/- 0.1, respectively. A comparison of the kinetic properties of L4, M4, and purified heart PFK isozymes clearly demonstrated that all three preparations exhibited different regulatory properties. In ventricular and atrial preparations the total PFK activity and the relative amounts of each subunit were drastically different suggesting regional differences between the distributions of PFK isozymes and consequently in regulation of PFK activities and glycolysis.

Nature of the rat brain 6-phosphofructo-1-kinase isozymes.

The complex nature of the brain 6-phosphofructo-1-kinase isozymes was examined by elution with a discontinuous gradient from QAE (quaternary aminoethyl)-Sephadex. In the first wash (150 mM NaCl), where the rat muscle 6-phosphofructo-1-kinase isozyme (M4) eluted, about 40% of the total brain 6-phosphofructo-1-kinase activity washed through without exhibiting a sharp peak. In the second elution (300 mM NaCl), the remaining activity eluted in a sharp peak that preceded where the major rat liver 6-phosphofructo-1-kinase isozyme (L4) eluted. Enzyme activity in brain extracts or purified brain isozymes was titrated above 90% with M4 anti-IgG and 20% with L4 anti-IgG. A purification procedure was developed which resulted in a recovery of 70 to 80% of the original enzyme activity in brain 100,000 X g supernatant fluids. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis on slab gels and detection by silver staining indicated that three components were present with apparent molecular weights of 87,500, 85,000, and 80,000. The 85,000- and 80,000-dalton components corresponded to the subunits of M4 and L4, respectively. The third component (C type) was thought to be an actual subunit since it exhibited the highest molecular weight and was present in an exhaustively washed immunoprecipitate of the purified brain isozymes. From 10 different purifications of the brain enzyme, the subunit distributions of the liver, muscle, and C-type subunit were 1.4 +/- 0.2, 4.9 +/- 0.5, and 3.9 +/- 0.3, respectively. A comparison of the kinetic properties of purified liver, muscle, and brain isozymes clearly demonstrated that all three preparations had quantitatively different regulatory properties. All three subunits were present in different regions of the brain, and region-specific changes in total activity and the relative amounts of each subunit were observed. This study suggests that brain 6-phosphofructo-1-kinase is a complex mixture of homotetramers and hybrids which are composed of different amounts of the three subunits.

Purification of homogeneous rat phosphofructokinase isozymes with high specific activities.

The purification of rat muscle and liver phosphofructokinase (PFK) isozymes has been greatly facilitated by column chromatographic separation on immobilized Cibacron Blue F3GA. The homogeneous liver PFK isozyme exhibited a specific activity of greater than 200 units per mg of protein which is nearly two-fold greater than has been previously reported for this isozyme. The yields for this isozyme exceeded 40% of the original activity and the molecular weight of its subunit was about 85,000 as determined by SDS-polyacrylamide gel electrophoresis. The muscle PFK isozyme's specific activity was approximately 265 units/mg of protein which also is about twice the greatest specific activity previously reported. The overall yield for muscle PFK exceeded 50% of the original activity, and the molecular weight of its subunit was approximately 82,000. Using each homogeneous isozyme, antibodies were produced in rabbits; and the immunoglobin-G (IgG) fraction from the sera of these rabbits was highly specific for the PFK isozyme used as an antigen.