Uptake and Release of Glycine in the Guinea Pig Cochlear Nucleus After Axotomy of Afferent or Centrifugal Fibers

Glycine may be an inhibitory transmitter in the mammalian cochlear nucleus (CN). This study attempts to determine if cochlear and/or centrifugal projections to the CN use glycine as a transmitter. The high-affinity uptake and electrically evoked release of exogenous [14C]glycine were measured in vitro in the three major subdivisions of the guinea pig CN: the anteroventral, posteroventral, and dorsal cochlear nuclei (AVCN, PVCN, and DCN, respectively). [14C]Glycine (3.4 microM) was taken up by each subdivision, reaching tissue concentrations six to seven times that in the medium. Subsequent electrical stimulation evoked a Ca2+-dependent release of [14C]glycine from each subdivision. These activities were compared in subdivisions fr0m unlesioned animals, and from animals with lesions of centrifugal or cochlear projections to the CN. Two knife-cut lesions were made to interrupt centrifugal projections to the CN lying in the right acoustic striae and trapezoid body. In one group of animals, centrifugal fibers projecting mainly to the right AVCN and PVCN were severed, which reduced [14C]glycine uptake and release by 44-53% in these subdivisions, but not in the right DCN. In another group of animals, fibers projecting mainly to the right PVCN and DCN were severed, which reduced [14C]glycine uptake and release by 33-47% in these subdivisions, but not in the right AVCN. In CN subdivisions contralateral to either lesion there was no significant change in [14C]glycine uptake or release. Neither of these lesions altered the uptake or release of D-[3H]aspartate in the right or the left CN. Ablation of the left cochlea, which presumably destroyed cochlear nerve fibers unilaterally, had no effect on [14C]glycine uptake and release. These observations suggest that centrifugal projections contribute a proportion of the glycinergic synaptic endings in the CN. In addition, some glycinergic endings probably arise from neurons intrinsic to the CN. The cochlear nerve contains very few, if any, glycinergic fibers.     

Evidence for a Glutamatergic Pathway from the Guinea Pig Auditory Cortex to the Inferior Colliculus

Abstract: We attempt to provide evidence that the projection from the guinea pig auditory cortex (AC) to the inferior colliculus (IC) may contain glutamatergic or GABAergic fibers. Seven days after unilateral AC aspiration, histological studies indicated almost complete AC destruction and preterminal degeneration of fibers and terminal fields in the dorsal cortex (DCIC), external cortex (ECIC), and central nucleus (CNIC) of the IC ipsilateral to the ablated AC. Contralaterally, degeneration appeared in the DCIC. AC ablation depressed the electrically evoked Ca²⁺-dependent release of d -[³H]aspartate ( d -[³H]Asp) in the ipsilateral DCIC, ECIC, and CNIC, and d -[³H]Asp uptake in the CNIC. Together with other evidence that the corticocollicular pathway is excitatory, these findings suggest that this projection may contain glufamatergic and/or aspartatergic (Glu/Asp-ergic) fibers. Glutamic acid decarboxylase immunoreactivity was not apparent in presumed pyramidal cells of layer V of the AC retrogradely labeled with biotinylated dextran injected into the ipsilateral IC. Thus, corticocollicular neurons probably do not synthesize GABA and may not be GABAergic. However, AC ablation depressed [¹⁴C]GABA release from the ipsilateral DCIC and ECIC, and [¹⁴C]GABA uptake in the DCIC. These findings are consistent with the atrophy or down-regulation of some subcortical neurons that mediate GABAergic transmission in the IC.     

Uptake and Release of d-Aspartate, GABA, and Glycine in Guinea Pig Brainstem Auditory Nuclei

Abstract: This study attempts to determine if the medial (MSO) and lateral superior olive (LSO), medial nucleus of the trapezoid body (MNTB), ventral nucleus of the lateral lemniscus (VNLL), and central nucleus of the inferior colliculus (ICc) contain glutamatergic synaptic endings. Micropunch and microdissection procedures provided fresh samples of these auditory nuclei for the measurement of the high-affinity uptake and electrically evoked release of exogenous d -[³H]ASP. The study also determined if the LSO and MSO contain glycinergic synaptic endings by measuring uptake and release of [¹⁴C]-Gly in these nuclei, and whether the MNTB, VNLL, and ICc contain GABAergic endings by assessing the uptake and release of [¹⁴C]GABA in these structures. Several strategies optimized the evoked Ca²⁺-dependent release of the labeled amino acids. These included the enhancement of high-affinity uptake during loading of the markers into the tissues, inhibition of uptake during the subsequent measurement of release, and use of an electrical stimulus current that evoked maximal Ca²⁺-dependent release. Each of these nuclei manifested the high-affinity uptake and the evoked Ca²⁺-dependent release of d -[³H]Asp, suggesting the presence of synaptic endings that may use Glu or Asp as a transmitter. Similar findings suggest the presence of glycinergic synaptic endings in the LSO and MSO, and of GABAergic synaptic endings in the MNTB, VNLL, and ICc.     

Uptake and Release of D-Aspartate in the Guinea Pig Cochlear Nucleus

Abstract: This study attempted to determine if l -glutamate (L-Glu) and/or l -aspartate (L-Asp) might be the transmitters of neurons that provide synaptic endings to the cochlear nucleus of the medulla. The uptake and release of D-[³H]aspartate (D-Asp), a putative marker for l -Glu and l -Asp, were measured in the guinea pig cochlear nucleus before and after destruction of the cochlear afferents by cochlear ablation. The cochlear nucleus was dissected into the anteroventral (AVCN), posteroventral (PVCN), and dorsal (DCN) cochlear nuclei. Subdivisions from unlesioned animals took up D-Asp, achieving concentrations in the tissues that were 13–20 times that in the medium. Subsequently, electrical stimulation evoked a Ca²⁺-dependent release of part of the D-Asp from each subdivision. Disarticulation of the middle ear ossicles, which attenuates acoustic stimulation, produced a modest inhibition of D-Asp release in each subdivision, but did not alter the uptake of D-Asp. Cochlear ablation strongly depressed both the uptake and the release of D-Asp in each subdivision, presumably as a result of destruction of the cochlear nerve endings in the cochlear nucleus. Nevertheless, after lesions, there was a preservation of the uptake and release of D-Asp in the DCN relative to the AVCN and PVCN. These residual activities in the DCN may be mediated by the axonal endings of the granule cells of the cochlear nucleus. The present findings support the hypothesis that the granule cells of the cochlear nucleus, as well as the cochlear nerve fibers, use l -Glu and/or l -Asp as transmitters.     

Ethylenediamine as a Specific Releasing Agent of γ-Aminobutyric Acid in Rat Striatal Slices

Abstract: Following incubation with [¹⁴C]y-aminobutyric acid (GABA) or [³H]dopamine, slices of rat striatum were superfused with media containing 36 mM K⁺ or ethylenediamine (EDA), 1 or 5 mM. Both K⁺ and EDA induced a release of [¹⁴C]GABA, the K⁺-induced release being largely Ca²⁺-dependent, while the EDA-induced release was not. Whereas K⁺ also evoked a Ca²⁺-dependent release of [³H]dopamine, EDA evoked no release of dopamine. EDA may therefore have potential as a specific GABA releasing agent.     

THE RELEASE OF AMINO ACIDS FROM THE HEMISECTED SPINAL CORD DURING STIMULATION

Abstract— Isolated frog or toad hemicords were incubated for 40 min with either [¹⁴C]glycine, [³H]GABA, l -[¹⁴C]glutamate. l -[¹⁴C]aspartate, l -[¹⁴C]serine, l [¹⁴C]threonine or l -[³H]leucine, and the release of these compounds from the cord was measured under resting conditions and during electrical stimulation. Stimulation of spinal roots produced no significant change in the efflux of any of the compounds tested. Direct stimulation of the rostral cord however, produced a large increase in the efflux of [¹⁴C]glycine, [³H]GABA, l -[¹⁴C]glutamate and l -[¹⁴C]aspartate. These increased effluxes were calcium dependent, the effects of stimulation being reduced in a calcium-free, or magnesium-supplemented (10 mM) medium. Stimulation failed to produce an increase in the efflux of l -[¹⁴C]serine, l -[¹⁴C]threonine, l -[¹⁴H]leucine, [¹⁴C]mannitol or [¹⁴C]urea. These results are consistent with the suggestions that glycine, GABA, glutamate and aspartate may be synaptic transmitters in the spinal cord.     

NEUROTRANSMITTER UPTAKE INTO SLICES OF RAT CEREBRAL CORTEX IN VITRO: EFFECT OF SLICE SIZE

Abstract— The uptake of [¹⁴C]GABA, [¹⁴C]taurine, [³H] β -alanine and [¹⁴C]dopamine was compared in slices of rat cerebral cortex of three different sizes (0.1 × 0.1 × 2 mm, 0.2 × 0.2 × 2 mm and 0.4 × 0.4 × 2 mm prepared with a mechanical tissue chopper). [¹⁴C]Taurine and [³H] β -alanine uptake increased whereas [¹⁴C]GABA uptake decreased with increasing slice size. [¹⁴C]Dopamine uptake was optimal in 0.2 × 0.2 × 2 mm slices. Increasing slice size was shown to decrease inhibition of [³H] β -alanine and [¹⁴C]GABA uptake by l -2,4-diaminobutyric acid. Lactate dehydrogenase activity increased with increasing slice size indicating decreased tissue damage or increased cellular integrity. The possibility that varying slice size can be used to distinguish between neuronal and glial uptake is discussed. It is suggested that taurine uptake in the cerebral cortex is predominantly glial.     

Quantification of the GABA Shunt and the Importance of the GABA Shunt Versus the 2-Oxoglutarate Dehydrogenase Pathway in GABAergic Neurons

Abstract: We investigated the activity of the cerebral GABA shunt relative to the overall cerebral tricarboxylic acid (TCA) cycle and the importance of the GABA shunt versus 2-oxoglutarate dehydrogenase for the conversion of 2-oxoglutarate into succinate in GABAergic neurons. Awake mice were dosed with [1-¹³C]glucose, and brain extracts were analyzed by ¹³C NMR spectroscopy. The percent enrichments of GABA C-2 and glutamate C-4 were the same: 5.0 ± 1.6 and 5.1 ± 0.2%, respectively (mean ± SD). This, together with previous data, indicates that the flux through the GABA shunt relative to the overall cerebral TCA cycle flux equals the GABA/glutamate pool size ratio, which in the mouse is 17%. It has previously been shown that under the experimental conditions used in this study, the ¹³C labeling of aspartate from [1-¹³C]glucose specifically reflects the metabolic activity of GABAergic neurons. In the present study, the reduction in the formation of [¹³C]aspartate during inhibition of the GABA shunt by γ-vinyl-GABA indicated that not more than half the flux from 2-oxoglutarate to succinate in GABAergic neurons goes via the GABA shunt. Therefore, because fluxes through the GABA shunt and 2-oxoglutarate dehydrogenase in GABAergic neurons are approximately the same, the TCA cycle activity of GABAergic neurons could account for one-third of the overall cerebral TCA cycle activity in the mouse. Treatment with γ-vinyl-GABA, which increased GABA levels dramatically, caused changes in the ¹³C labeling of glutamate and glutamine, which indicated a reduction in the transfer of glutamate from neurons to glia, implying reduced glutamatergic neurotransmission. In the most severely affected animals these alterations were associated with convulsions.     

In vivo neurochemical profiling of rat brain by 1H-[13C] NMR spectroscopy: cerebral energetics and glutamatergic/GABAergic neurotransmission

The quantification of excitatory and inhibitory neurotransmission and the associated energy metabolism is crucial for a proper understanding of brain function. Although the detection of glutamatergic neurotransmission in vivo by ¹³C NMR spectroscopy is now relatively routine, the detection of GABAergic neurotransmission in vivo has remained elusive because of the low GABA concentration and spectral overlap. Using ¹H-[¹³C] NMR spectroscopy at high magnetic field in combination with robust spectral modeling and the use of different substrates, [U-¹³C₆]-glucose and [2-¹³C]-acetate, it is shown that GABAergic, as well as glutamatergic neurotransmitter fluxes can be detected non-invasively in rat brain in vivo .     

METABOLISM OF [14C]LEUCINE AND [14C]ACETATE IN SENSORIMOTOR CORTEX, THALAMUS, CAUDATE NUCLEUS AND CEREBELLUM OF THE CAT

Abstract— —In the head of the caudate nucleus, the relative specific activity of glutamine (glutamic acid specific activity = 1) was less than 1 with intravenous [¹⁴C]leucine as the tracer metabolite. This is in contrast to observations made in other brain areas (cortex, hippocampus, thalamus, pons, and medulla) where the relative specific activity of glutamine was greater than 1. This is also in contrast to findings when [l-¹⁴C]acetate was utilized as the tracer; under these conditions, in all brain areas, including the head of the caudate nucleus, the relative specific activity of glutamine was greater than 1. It is inferred that the differences in metabolism of [¹⁴C]leucine and [¹⁴C]acetate in the head of the caudate from that in other brain areas reflect differences in compartmentation of the glutamate-glutamine system.     

Effect of Tetanus Toxin on Transmitter Release from the Substantia Nigra and Striatum In Vitro

Abstract: The effect of tetanus toxin on the uptake and release of radiolabelled transmitters from slices prepared from substantia nigra (SN) and striatum of rats has been investigated. Tetanus toxin-500–750 mouse lethal doses (MLD)-injected into the SN 6 h before preparing the slices significantly reduced the calcium-dependent, potassium-evoked release of [³H]GABA. Endogenous GABA levels in the SN and [³H]GABA uptake by nigral slices were unaffected by pretreatment with the toxin. Injections of tetanus toxin (1000–2000 MLD) into the striatum significantly reduced the calcium-dependent, potassium-evoked release of [¹⁴C]GABA and also [³H]dopamine, but had no effect on the K⁺-evoked release of [³H]5-hydroxytryptamine or [¹⁴C]acetylcholine. It is concluded that tetanus toxin inhibits GABA release directly and not by interference with synthesis or inactivation processes.     

SUBCELLULAR DISTRIBUTION OF ENDOGENOUS AND [3H] γ-AMINOBUTYRIC ACID IN RAT CEREBRAL CORTEX

Abstract— Slices of rat cerebral cortex were labelled by incubation with [³H]γ-aminobutyric acid (GABA) and homogenized in isotonic sucrose. The subcellular distributions of endogenous GAB A, [³H]GABA and glutamate decarboxylase (GAD) were studied by density gradient centrifugation. The subcellular distributions of the labelled and endogenous amino acid were remarkably similar, indicating that [³H]GABA is taken up into the endogenous GABA pool. About 40 per cent of both endogenous and [³H]GABA were recovered in particles which were tentatively identified as synaptosomes from their equilibrium density and sensitivity to osmotic shock. In slices labelled with [³H]GABA and [¹⁴C]α-aminoisobutyric (AIB) acid, significantly more [³H]GABA was recovered in paniculate fractions than [¹⁴C]AIB. About 80 per cent of the enzyme GAD was also recovered in the same particle fractions which contained [³H]GABA and endogenous GABA. Evidence is presented which suggests that a loss of particle-bound GABA occurs during subcellular fractionation procedures.     

GAMMA-HYDROXYBUTYRATE DEGRADATION IN THE BRAIN IN VIVO: NEGLIGIBLE DIRECT CONVERSION TO GABA

Abstract— The metabolism of γ-hydroxybutyrate (GHB) was studied by following the fate of [1-¹⁴C]GHB in mouse brain after an intravenous injection. Cerebral uptake of GHB was rapid and this substance disappeared from brain tissue with a half-life of approx 5 min. Degradation of [1-¹⁴C]GHB took place in the brain since ¹⁴C was incorporated in amino acids associated with the tricarboxylic acid cycle: the labelling pattern was consistent with the oxidation of GHB via succinate through the cycle, rather than with β-oxidation of GHB. Conversion of [¹⁴C]GHB into [¹⁴C]GABA prior to oxidation was negligible, thus it is unlikely that the pharmacological action of GHB would be mediated through GABA formation. [¹⁴C]GHB oxidation also elicited the signs of metabolic compartmentation of the tricarboxylic acid cycle in the brain (glutamine/glutamate specific radioactivity ratio was about 4).     

THE EFFECT OF ELECTRICAL STIMULATION AND HIGH POTASSIUM CONCENTRATIONS ON THE EFFLUX OF [3H] γ-AMINOBUTYRIC ACID FROM BRAIN SLICES

Abstract— Brain slices were incubated with [³H]GABA in a medium containing aminooxyacetic acid to prevent metabolism of [³H]GABA by GABA-glutamate transaminase. The slices, which rapidly accumulated radioactivity, were then continuously perfused and the efflux of [³H]GABA from the tissue was measured. The spontaneous efflux of [³H]GABA consisted of an initial rapid phase followed by a much slower release of [³[H]GABA. After 40 min perfusion 90 per cent of the radioactivity remained in the tissue.
The slices were depolarized by electrical stimulation or by perfusion with a medium containing a high potassium concentration (40 mM). These procedures caused a striking increase in the efflux of [³H]GABA. The increased efflux produced by potassium, but not that produced by electrical stimulation, was dependent on calcium ions in the medium. The effect of electrical stimulation on [³H]GABA release was considerably reduced by a raised concentration (10 mM) of magnesium in the medium.
High potassium concentrations and electrical stimulation did not cause an increase in the efflux of [¹⁴C]urea, L-[³H]leucine or [¹⁴C]α-amino-isobutyric acid from brain slices. These results are consistent with the suggestion that GABA may be an inhibitory transmitter in the cerebral cortex.     

Metabolic Precursors and Compartmentation of Cerebral GABA in Vigabatrin-Treated Rats

Abstract: The metabolic precursors and cerebral compartmentation of the augmented GABA pool induced by vigabatrin, an irreversible inhibitor of GABA transaminase, have been investigated by ¹³C NMR. Adult rats receiving rat chow ad libitum were given either drinking water only or drinking water containing 2.5 g/L vigabatrin for 7 days. Both groups of animals were infused either with [1,2-¹³C₂]acetate (15 µmol/min/100 g body weight), an exclusive precursor of GABA formation through the glial glutamine pathway, or with [1,2-¹³C₂]glucose (15 µmol/min/100 g body weight), a substrate that can produce GABA through the glial glutamine pathway or by direct metabolism in the neurons. The brains were frozen in situ, extracted with perchloric acid, and analyzed by ¹³C NMR. In vigabatrin-treated animals [¹³C]glutamine, a common intermediate for [¹³C]GABA synthesis from glucose or acetate, was accumulated to similar amounts during infusions with [1,2-¹³C₂]glucose or [1,2-¹³C₂]acetate. However, [¹³C]GABA accumulation was sevenfold higher during [1,2-¹³C₂]glucose infusions or twofold higher during [1,2-¹³C₂]acetate infusions. These results show that the direct pathway of GABA formation by neuronal metabolism of glucose predominates over the alternative pathway through glial glutamine. Near-equilibrium relationships of the aminotransferases of GABA and aspartate imply that the observed [¹³C]GABA accumulation occurs initially in the neuronal compartment.     

Glutamine and Glucose as Precursors of Transmitter Amino Acids: Ex Vivo Studies

Abstract: Radiolabelled glutamine and glucose were infused into lateral ventricles of rats in order to label transmitter amino acid pools in vivo . Brain regions close to the lateral ventricle (hippocampus, corpus striatum, hypothalamus) were labelled more effectively than more distant structures such as cerebral cortex or cerebellum. All regions were labelled to much the same extent over 30-150 min by [U-¹⁴C]glucose, [U-¹⁴C]glutamine, or [³H]glutamine administered alone or together in doublelabel experiments when allowance was made for any differences in precursor specific radioactivities. Slices of cerebral cortex or hippocampus from brains labelled in vivo were incubated and stimulated in vitro with veratrine (75 μ M ); tetrodotoxin (1 μ M ) was present in the control medium. Single-label experiments showed that [U-¹⁴C]- glutamine was more effective than [U-¹⁴C]glucose for labelling releasable glutamate and GABA. Double-label experiments showed that [³H]glutamine and [U-¹⁴C]- glucose given together in vivo labelled glutamate and GABA releasable in vitro to a similar extent. Both types of experiment empbasise the large contribution made by glutamine in vivo to pools of transmitter glutamate and GABA.     

Now I know my ABC. A systems neurochemistry and functional metabolomic approach to understanding the GABAergic system

Here, we describe use of a reductionist brain model, the brain tissue slice, to generate snapshots of functional metabolism in response to a pharmacological (GABAergic) perturbation. Tissue slices prepared from Guinea pig cerebral cortex were incubated for 1 h in the presence of [3-¹³C]-pyruvate and ligands with affinity for GABA receptors. The resultant patterns of ¹³C flux and metabolite levels were measured by ¹³C/¹H NMR spectroscopy, generating 'metabolic fingerprints' for each ligand. Effects of agonists and effectors at GABA receptors (A, B, and C types) were examined, compared to those of exogenous GABA and evaluated using multivariate statistical models. Data clusterings did not directly correlate with GABA receptor types but produced at least five distinct groups ranked according to their affinity for GABA. As our experimental model retains, to a large extent, the structure and function of normal brain tissue, the generated database can be used to assess GABAergic ligands and make unique inferences relevant to their modes of action in brain.     

Lumiflavin and Lumichrome Transport in the Central Nervous System

Abstract: The transport of the lipid-soluble sugarless flavins, [¹⁴C]lumiflavin and [¹⁴C]lumichrome, into and from the isolated choroid plexus and brain slices was studied in vitro. The isolated choroid plexus accumulated both [¹⁴C] flavins by a saturable, energy-requiring process that did not depend on binding or intracellular metabolism of the [¹⁴C] flavins. Both sugar-containing and sugarless flavins, as well as cyclic organic acids, significantly inhibited [¹⁴C]lumiflavin and [¹⁴C]Iumichrome uptake by the isolated choroid plexus. Within 2.5 min, 75% of the [¹⁴C]lumiflavin accumulated by the isolated choroid plexus was released into the medium. Brain slices accumulated [¹⁴C]lumiflavin by a saturable process that did not meet all the criteria for active transport. Ninety-five percent of the [¹⁴C]lumiflavin accumulated by brain slices was released into the medium within 7.5 min. In vivo , 2 h after the intraventricular injection of 6.5 nmol [¹⁴C]lumiflavin, almost all of the [¹⁴C]flavin was cleared from the CNS. Addition of 3.5 μmol FMN to the intraventricular injectate significantly decreased the clearance of [¹⁴C]lumiflavin from the CNS. These studies document that the sugarless flavins are transported by the flavin transport systems in the CNS.     

N-Methyl-d-Aspartate Releases γ-Aminobutyric Acid from Rat Striatum In Vivo: A Microdialysis Study Using a Novel Preloading Method

Abstract: In vivo microdialysis was used in conjunction with a novel dual-label preloading method to monitor changes in extracellular levels of γ-aminobutyric acid (GABA) and glutamate due to N -methyl- d -aspartate (NMDA) infusion in the striatum of conscious, unrestrained rats. [¹⁴C]GABA and [³H]glutamate were applied in the dialysis stream for a preloading period of 30 min, after which dialysis perfusion was continued for up to 6 h and dialysate samples were collected for analysis by liquid scintillation spectrometry. NMDA (300 μ M in the dialysate) caused significant rises in both ¹⁴C and ³H content measured in the dialysates, the majority of which remained associated with the preloaded GABA and glutamate, respectively. The NMDA-evoked release of both GABA and glutamate was blocked by the specific NMDA receptor antagonist 3-[(±)-2-carboxypiperazin-4-yl]propyl-1-phosphonic acid (CPP), indicating that the response was receptor mediated. The NMDA-stimulated release of glutamate was also totally abolished by concomitant application of the adenosine agonist 2-chloroadenosine or by prior frontal decortication. However, these two treatments caused little change in NMDA-evoked GABA release. These results show that NMDA causes release of GABA from the striatum in vivo by an NMDA receptor-mediated mechanism and that the majority of this release is not secondary to glutamate release from terminals of the corticostriate pathway. In addition, they confirm the results of previous studies investigating the effect of NMDA on endogenous glutamate release.     

Isoprenoid biosynthesis in bacteria: Two different pathways?

Abstract The biosynthesis of isopentenylpyrophosphate, a central intermediate of isoprenoid formation, was investigated in six different bacterial organisms. Cell-free extracts of Myxococcus fulvus, Staphylococcus carnosus, Lactobacillus plantarum and Halobacterium cutirubrum converted [¹⁴C]acetyl-CoA or [¹⁴C]hydroxymethylglutaryl-CoA to [¹⁴C]mevalonic acid. Furthermore, [¹⁴C]mevalonic acid, [¹⁴C]mevalonate-5-phosphate and [¹⁴C]mevalonate-5-pyrophosphate were metabolized to [¹⁴C]isopentenylpyrophosphate in bacteria. In contrast, no intermediates of this reaction sequence could be detected using cell-free extracts of Zymomonas mobilis and Escherichia coli . These results indicate that at least two different pathways for the biosynthesis of isopentenylpyrophosphate are present in bacteria.