Steroid Recognition Site Research Articles

Multi-generational pharmacophore modeling for ligands to the cholane steroid-recognition site in the β1 modulatory subunit of the BKCa channel

Large conductance, voltage- and Ca2+-gated K+ (BKCa) channels play a critical role in smooth muscle contractility and thus represent an emerging therapeutic target for drug development to treat vascular disease, gastrointestinal, bladder and uterine disorders. Several compounds are known to target the ubiquitously expressed BKCa channel-forming α subunit. In contrast, just a few are known to target the BKCa modulatory β1 subunit, which is highly expressed in smooth muscle and scarce in most other tissues. Lack of available high-resolution structural data makes structure-based pharmacophore modeling of β1 subunit-dependent BKCa channel activators a major challenge. Following recent discoveries of novel BKCa channel activators that act via β1 subunit recognition, we performed ligand-based pharmacophore modeling that led to the successful creation and fine-tuning of a pharmacophore over several generations. Initial models were developed using physiologically active cholane steroids (bile acids) as template. However, as more compounds that act on BKCa β1 have been discovered, our model has been refined to improve accuracy. Database searching with our best-performing model has uncovered several novel compounds as candidate BKCa β1 subunit ligands. Eight of the identified compounds were experimentally screened and two proved to be activators of recombinant BKCa β1 complexes. One of these activators, sobetirome, differs substantially in structure from any previously reported activator.

Cerebrovascular Dilation via Selective Targeting of the Cholane Steroid-Recognition Site in the BK Channel β1-Subunit by a Novel Nonsteroidal Agent

The Ca(2+)/voltage-gated K(+) large conductance (BK) channel β1 subunit is particularly abundant in vascular smooth muscle. By determining their phenotype, BK β1 allows the BK channels to reduce myogenic tone, facilitating vasodilation. The endogenous steroid lithocholic acid (LCA) dilates cerebral arteries via BK channel activation, which requires recognition by a BK β1 site that includes Thr169. Whether exogenous nonsteroidal agents can access this site to selectively activate β1-containing BK channels and evoke vasodilation remain unknown. We performed a chemical structure database similarity search using LCA as a template, along with a two-step reaction to generate sodium 3-hydroxyolean-12-en-30-oate (HENA). HENA activated the BK (cbv1 + β1) channels cloned from rat cerebral artery myocytes with a potency (EC₅₀ = 53 μM) similar to and an efficacy (×2.5 potentiation) significantly greater than that of LCA. This HENA action was replicated on native channels in rat cerebral artery myocytes. HENA failed to activate the channels made of cbv1 + β2, β3, β4, or β1T169A, indicating that this drug selectively targets β1-containing BK channels via the BK β1 steroid-sensing site. HENA (3-45 μM) dilated the rat and C57BL/6 mouse pressurized cerebral arteries. Consistent with the electrophysiologic results, this effect was larger than that of LCA. HENA failed to dilate the arteries from the KCNMB1 knockout mouse, underscoring BK β1's role in HENA action. Finally, carotid artery-infusion of HENA (45 μM) dilated the pial cerebral arterioles via selective BK-channel targeting. In conclusion, we have identified for the first time a nonsteroidal agent that selectively activates β1-containing BK channels by targeting the steroid-sensing site in BK β1, rendering vasodilation.

Effects of some neurosteroids injected into some brain areas of WAG/Rij rats, an animal model of generalized absence epilepsy

Neurosteroids are synthesized in the brain and have been demonstrated to modulate various cerebral functions. Allopregnanolone (3α-hydroxy-5α-pregnan-20-one), a naturally occurring neurosteroid, and ganaxolone (3α-hydroxy-3β-methyl-5α-pregnan-20-one), a synthetic derivative, are two neurosteroids acting as positive allosteric modulators of the GABA A receptor complex acting on a specific steroid recognition site. Both agents antagonize generalized tonic-clonic seizures in various animal models of epilepsy. Pregnenolone sulphate (3β-hydroxy-5α-pregnen-20-one 3-sulphate; PS) is a negative allosteric modulator of GABA A receptors and a positive modulator of the NMDA receptors. We have evaluated the effects of such compounds in a genetic animal model of absence epilepsy, the WAG/Rij rat. Animals were chronically implanted with five frontoparietal cortical electrodes for electrocorticogram (EEG) recordings and bilateral guide cannulae into specific brain areas of the cortico-thalamic circuit in order to evaluate the effects of these compounds on the number and duration of epileptic spike-wave discharges (SWDs). The focal and bilateral microinjection of the two GABA A positive modulators into some thalamic nuclei (nucleus ventralis posteromedialis, nucleus reticularis thalami, nucleus ventralis posterolateralis was usually able to significantly worsen the occurrence of SWDs in WAG/Rij rats. Whereas both compounds were able to reduce the number and duration of SWDs when microinjected into the peri-oral region of the primary somatosensory cortex. The effects of PS were more complex depending on both the dose and the site of administration, generally, at low doses in thalamic nuclei and cortex, PS induced an increase of absence activity and a reduction at higher doses. These findings suggest that neurosteroids might play a role in absence epilepsies and that it might depend on the involvement of specific neuronal areas.

Neuroactive steroids have multiple actions to potentiate GABAA receptors.

The effects of neuroactive steroids on the function of GABAA receptors were studied using cell-attached records of single channel activity recorded from HEK293 cells transfected with alpha1 beta2 gamma2L subunits. Activity was elicited with a half-maximal (50 microM) concentration of GABA. Two steroids were studied in detail: ACN ((3alpha,5alpha,17beta)-3-hydroxyandrostane-17-carbonitrile) and B285 ((3alpha,5beta,17beta)-3-hydroxy-18-norandrostane-17-carbonitrile). Four effects on channel activity were seen, two on open time distributions and two on closed times. When clusters of openings were elicited in the absence of steroid, the open time distribution contained three components. ACN produced concentration-dependent alterations in the open time distribution. The prevalence of the longest duration class of open times was increased from about 15% to about 40% (EC50 about 180 nM ACN), while the duration of the longest class increased from 7.4 ms to 27 ms (EC50 about 35 nM ACN). B285 also increased the prevalence of the longest duration open times (EC50 about 18 nM B285) but increased the duration only at concentrations close to 10 microM. The differences in the actions of these two steroids suggest that the effects on proportion and duration of the long duration open time component are produced by independent mechanisms and that there are separate recognition sites for the steroids which are associated with the two functional actions. The closed time distributions also showed three components in the absence of steroid. The rate of occurrence of the two brief duration closed time components decreased with increasing ACN, with an EC50 of about 50 nM ACN. In contrast, B285 did not reduce the rate of occurrence of the brief closings until high concentrations were applied. However, both B285 and ACN reduced the rate of occurrence of the activation-related closed state selectively, with comparable IC50 concentrations (about 40 nM ACN, 20 nM B285). As in the case for action on open times these data suggest that there are two recognition sites and two independent mechanisms, perhaps the sites and mechanisms associated with actions on open times. The presence of 1 microM ACN had no effect on the estimated channel opening rate or on the apparent affinity of the receptor for GABA. Mutation of the carboxy terminus of the gamma2 subunit, but not the alpha1 or beta2 subunits, abolished the ability of ACN to increase the duration of OT3 but had no effect on the reduction of the rate of occurrence of the activation-related closed state. These observations are also consistent with the idea that there is more than one distinguishable steroid recognition site on the GABAA receptor.

Membrane-initiated steroid signaling (MISS): genomic steroid action starts at the plasma membrane

Plasma membrane (PM) steroid recognition sites are thought to be responsible only for rapid, non-genomic responses without any link to the nuclear receptor-mediated genomic effects of steroids. We focused on a PM “glucocorticoid-importer” (GC-importer) that imports GC into rat liver cells. This site interacts also with particular gestagens (progesterone, P; medroxyprogesterone, MP; ethynodiol, Ethy) and estrogens (ethinylestradiol, EE 2; mestranol), which do not bind to the nuclear GC receptor (GR). To elucidate the role of the GC-importer, we transfected a rat wild-type hepatocyte (CC-1) and a hepatoma cell line, unable to import GC (MH 3924), with a GC↔GR-responsive luciferase (luc)-reporter gene. Selected steroids were tested for their ability to induce or inhibit luc expression. Corticosterone (B) and dexamethasone (Dex), but also the GC-antagonists cortexolone (Cortex), P and MP, induced luc. Even the PM-impermeable BSA-derivatives of B, Dex and Cortex did so to almost the same extent as the free steroids. MH 3924 cells respond stronger than CC-1 to luc inducing steroids. Luc expression was inhibited by RU 38 486, but also by EE 2 and Ethy. The thiol reactive mesylate-derivatives of B, Dex and Cortex induced to a considerably lesser extent than the free or BSA-steroids. The thiol reagent mersalyl blocks cellular entry of GC and inhibits luc induction in CC-1 cells. Incubation with EE 2 and B of PM-vesicles, isolated from liver cells, resulted in a decrease of the density of two 75 and 52 kDa G-proteins reflecting a diminished exchange of GDP by GTP. Conclusion: the PM-residing GC-importer, now renamed “ Steroid Hormone Recognition and Effector Complex” (SHREC) is an interdependent part of the complete GC signal propagation in which G-proteins are involved. Free SH-groups of SHREC are a prerequisite for genomic GC activity. Specific interactions between SHREC and GC-agonist/-antagonist trigger steroid-dependent signaling. However, import of the ligand into the cell terminates it. Thus, the PM-related non-genomic steroid responses are clearly linked to the GR-related genomic effects.

Subunit dependent modulation of GABA A receptor function by neuroactive steroids

Neurosteroids are potent, endogenous modulators of GABA A receptor function in the central nervous system. The endogenous progesterone metabolite allopregnanolone (ALP) and the synthetic steroid compound alphaxalone (AFX) have been shown to both directly activate and potentiate GABA A receptor-activated membrane current ( I GABA). The role of different α and γ subunit subtypes in modulation of I GABA by ALP and AFX was investigated using recombinant GABA A receptor isoforms expressed in Xenopus oocytes. Changing or removal of the α subunit subtype altered the efficacy of both ALP and AFX (α2β1γ2L>α1β1γ2L≫β1γ2L) to potentiate I GABA, but did not alter the potency of the neuroactive steroids at these receptor isoforms. The efficacy of ALP to enhance I GABA was also dependent on the γ subunit subtype (α1β1γ3>α1β1γ2L=α1β1γ1). AFX also had higher efficacy in the α1β1γ3 receptor isoform compared to α1β1γ1. In contrast to ALP, the potency of AFX was greater in the α1β1γ3 and α1β1γ1 receptor isoforms compared to α1β1γ2L. This study provides evidence that the α subunit subtype determines the efficacy, but not the potency, of these neuroactive steroids to potentiate I GABA. The γ3 subunit subtype increases the maximal efficacy of neuroactive steroids compared to other γ subunit subtypes. These results suggest that the heteromeric assembly of different GABA A receptor isoforms containing different subunit subtypes results in multiple steroid recognition sites on GABA A receptors that in turn produce distinctly different modulatory interactions between neuroactive steroids acting at the GABA A receptor.

Modulation of GABAA receptor function by neuroactive steroids: evidence for heterogeneity of steroid sensitivity of recombinant GABAA receptor isoforms

Neuroactive steroids are potent, selective allosteric modulators of gamma-aminobutyric acid type A (GABAA) receptor function in the central nervous system, and may serve as endogenous anxiolytic and analgesic agents. In order to study the influence of subunit subtypes of the GABAA receptor on modulation of receptor function by neuroactive steroids, we expressed human recombinant GABAA receptors in Xenopus oocytes. GABA-activated membrane current, and the modulatory effects of the endogenous neurosteroid 5alpha-pregnan-3alpha-ol-20-one (allopregnanolone) and the synthetic steroid anesthetic 5alpha-pregnan-3alpha-ol-11,20-dione (alphaxalone) were measured using two-electrode voltage-clamp recording techniques. Allopregnanolone had similar effects to potentiate GABA-activated membrane current in the alpha1beta1gamma2L and alpha1beta2gamma2L receptor isoforms. In contrast, alphaxalone was much more effective as a positive allosteric modulator on the alpha1beta1gamma2L receptor isoform. In the absence of the gamma2L subunit subtype, allopregnanolone had much greater efficacy, but its potency was decreased. Allopregnanolone was much more effective on the alpha1beta1 receptor isoform compared with the alpha1beta2 receptor isoform. The potency for alphaxalone to potentiate the GABA response was not altered in the absence of the gamma2L subunit subtype, although its efficacy was greatly enhanced. Both allopregnanolone and alphaxalone produced nonparallel leftward shifts in the GABA concentration-response relationship in the absence of the gamma2L subunit, decreasing the EC50 concentration of GABA and increasing the maximal response. Only alphaxalone increased the maximal GABA response when the gamma2L subunit subtype was present. The 3beta-pregnane isomers epipregnanolone and isopregnanolone both inhibited the ability of allopregnanolone and alphaxalone to potentiate GABAA receptor function. However, the degree of block produced by the 3beta-pregnane steroid isomers was dependent on the type of receptor isoform studied and the neuroactive steroid tested. Isopregnanolone, the 3beta-isomer of allopregnanolone, was significantly more effective as a blocker of potentiation caused by allopregnanolone compared with alphaxalone in all receptor isoforms tested. Epipregnanolone had a greater efficacy as a blocker at the alpha1beta2gamma2L receptor isoform compared with the alpha1beta1gamma2L receptor isoform, and also produced a greater degree of block of potentiation caused by allopregnanolone compared with alphaxalone. Our results support the hypothesis that the heteromeric assembly of different GABAA receptor isoforms containing different subunit subtypes results in multiple steroid recognition sites on GABAA receptors, which in turn produces distinctly different modulatory interactions between neuroactive steroids acting at the GABAA receptor. The alpha and gamma subunit subtypes may have the greatest influence on allopregnanolone modulation of GABAA receptor function, whereas the beta and gamma subunit subtypes appear to be most important for the modulatory effects of alphaxalone.Key words: GABAA receptor, neurosteroid, allopregnanolone, alphaxalone, Xenopus oocyte.

Modulation of GABA(A) receptor function by neuroactive steroids: evidence for heterogeneity of steroid sensitivity of recombinant GABA(A) receptor isoforms.

Neuroactive steroids are potent, selective allosteric modulators of gamma-aminobutyric acid type A (GABA(A)) receptor function in the central nervous system, and may serve as endogenous anxiolytic and analgesic agents. In order to study the influence of subunit subtypes of the GABA(A) receptor on modulation of receptor function by neuroactive steroids, we expressed human recombinant GABA(A) receptors in Xenopus oocytes. GABA-activated membrane current, and the modulatory effects of the endogenous neurosteroid 5alpha-pregnan-3alpha-ol-20-one (allopregnanolone) and the synthetic steroid anesthetic 5alpha-pregnan-3alpha-ol-11,20-dione (alphaxalone) were measured using two-electrode voltage-clamp recording techniques. Allopregnanolone had similar effects to potentiate GABA-activated membrane current in the alpha1beta1gamma2L and alpha1beta2gamma2L receptor isoforms. In contrast, alphaxalone was much more effective as a positive allosteric modulator on the alpha1beta1gamma2L receptor isoform. In the absence of the gamma2L subunit subtype, allopregnanolone had much greater efficacy, but its potency was decreased. Allopregnanolone was much more effective on the alpha1beta1 receptor isoform compared with the alpha1beta2 receptor isoform. The potency for alphaxalone to potentiate the GABA response was not altered in the absence of the gamma2L subunit subtype, although its efficacy was greatly enhanced. Both allopregnanolone and alphaxalone produced nonparallel leftward shifts in the GABA concentration-response relationship in the absence of the gamma2L subunit, decreasing the EC50 concentration of GABA and increasing the maximal response. Only alphaxalone increased the maximal GABA response when the gamma2L subunit subtype was present. The 3beta-pregnane isomers epipregnanolone and isopregnanolone both inhibited the ability of allopregnanolone and alphaxalone to potentiate GABA(A) receptor function. However, the degree of block produced by the 3beta-pregnane steroid isomers was dependent on the type of receptor isoform studied and the neuroactive steroid tested. Isopregnanolone, the 3beta-isomer of allopregnanolone, was significantly more effective as a blocker of potentiation caused by allopregnanolone compared with alphaxalone in all receptor isoforms tested. Epipregnanolone had a greater efficacy as a blocker at the alpha1beta2gamma2L receptor isoform compared with the alpha1beta1gamma2L receptor isoform, and also produced a greater degree of block of potentiation caused by allopregnanolone compared with alphaxalone. Our results support the hypothesis that the heteromeric assembly of different GABA(A) receptor isoforms containing different subunit subtypes results in multiple steroid recognition sites on GABA(A) receptors, which in turn produces distinctly different modulatory interactions between neuroactive steroids acting at the GABA(A) receptor. The alpha and gamma subunit subtypes may have the greatest influence on allopregnanolone modulation of GABA(A) receptor function, whereas the beta and gamma subunit subtypes appear to be most important for the modulatory effects of alphaxalone.

3β-OH-5β-Pregnan-20-One Enhances [3H]Gaba Binding in Developing Chick Optic Lobe

The neurosteroids are synthesized in the CNS and act mainly through allosteric modulation of the GABAA receptor. Structure-activity relationship studies in mammalian CNS have shown that a 3α-hydroxyl group and a 5α-reduced A-ring are striking features for their biological activity, while the 3β,5β structures as in 3β,5β-P are completely inactive. In this work we report the enhancing activity of epipregnanolone on [3H]GABA binding to its receptor sites in the chick optic lobe. Concentration-effect curves for this neurosteroid showed a concentration-dependent activity with different potencies at the three developmental stages studied, the hatching stage being the most sensitive to the steroid stimulatory effect. The displacement of a potent 3α,5α steroid by epipregnanolone indicated that this steroid behaved as a partial agonist of the steroid recognition site. Considering the developmental profile for steroidogenesis in avian tissues and the biological relevance of 5β-reduced steroids in early development, we propose that 3 or its 3α-epimer, pregnanolone, instead of the potent 3α,5α neurosteroids, modulates GABAA receptors in the chick optic lobe during development.

Steroid modulation of GABA A receptors in an amphibian brain

Steroids can modulate γ-aminobutyric acid (GABA A) receptor function in rat brains, but the physiological relevance of this mechanism is still unclear. To determine whether this phenomenon is widespread among vertebrates, we investigated steroid modulation of GABA A receptors in amphibian brain tissue. Equilibrium binding parameters for t-butylbicyclophosphorothionate ([ 35S]TBPS) and [ 3H]flunitrazepam were similar in Tricha glanulosa and mammalian brains, as was the allosteric regulation of [ 35S]TBPS and [ 3H]flunitrazepam binding in GABA. The rank order and absolute potencies of steroids to inhibit [ 35S]TBPS binding and enhance [ 3H]flunitrazepam binding were also similar in Taricha and rat brains. As in mammalian studies, physiological concentrations of corticosterone had no effect on ligand binding of GABA-stimulated Cl − uptake. In autoradiogrphic studies, 3α-hydroxy-5α-pregnan-20-one inhibited [ 35S]TBPS binding sites in all brain regions examined, whereas corticosterone had no effect on [ 35S]TBPS binding. These studies suggest that the steroid recognition sites on GABA A receptors bave been highly conserved through vertebrate evolution and thus portend physiologically important functions. However, the pharmacological profiles for the GABA A receptor and high-affinity coritcosteroid receptor are apparently different, suggesting there are multiple types of steroid recognition sites on neuronal membranes.

GABAA RECEPTORS AND ACTIONS OF NEUROSTEROIDS

GABA acting via GABAA receptors exerts inhibitory actions on neurons. Certain endogenous steroids (neurosteroids) are modulators of GABAA receptors. Tetrahydroprogesterone (THP) and tetrahydrodeoxycorticosterone (THDOC) behave as allosteric agonists of the GABAA receptors, while pregnenolone sulfate (PS) and dehydroepiandrosterone sulfate (DHEAS) act as antagonists. The neurosteroids alter ligand binding, chloride transport and electrophysiological responses to GABA, acting at unique steroid recognition sites at the GABAA receptors. Certain behavioral effects of the neurosteroids are consistent with their GABA-agonistic or -antagonistic properties. For example, THP and THDOC manifest hypnotic and anxiolytic actions, while PS reduces barbiturate-induced sleep time. Physiologically changing levels of GABAergic steroids, during development, stress, phases of ovarian cycle or pregnancy may alter GABAA receptor function, thus influencing neuronal excitability and CNS arousal. For example, pregnancy and puerperium are associated with alterations of GABAA receptor function that seem attributable to steroid actions. GABA-ergic steroids also regulate uterine contractibility. This phenomenon may play a role in quieting the uterus during pregnancy and in its activation during parturition. (Review: Majewska, Prog.Neurobiol.38:379, 1992.)

Identification and characterization of a pregnane steroid recognition site that is functionally coupled to an expressed GABAA receptor.

Based on the pharmacological and biochemical evidence to date, especially that derived from the recombinantly expressed receptor studies, the suggestion that a novel GBRC-linked steroid recognition site exists becomes a cogent argument. The high affinity of the steroid site for certain naturally occurring metabolites of progesterone and glucocorticoids favors a physiologic role for these steroids in the regulation of brain excitability. Clearly, investigations of such a regulatory role is warranted. If present, it provides an important example of endocrine control of a major inhibitory neurotransmitter in the CNS. Moreover, as we gain a greater understanding of the molecular organization of the GBRC, the putative steroid site provides a novel target for the rational design of therapeutic agents for the treatment of anxiety, epilepsy, and insomnia.

Anticonvulsant steroids and the GABA/benzodiazepine receptor-chloride ionophore complex

The ability of steroids to influence brain excitability is well documented. Certain 3 alpha-hydroxylated pregnanes are known to possess anticonvulsant and sedative-hypnotic/anesthetic properties. It has been observed that the seizure susceptibility in menstruating women with catamenial epilepsy appears to be correlated with changes in ovarian steroid levels. However, the underlying mechanism of these steroid influences on brain activity has only been recently revealed by pharmacological studies. These studies have provided compelling evidence for the presence of a novel steroid recognition site on the GABAA-benzodiazepine receptor complex (GBRC). Steroids may interact with this site with high affinity and stereospecificity to enhance chloride channel conductance in a manner similar to that produced by benzodiazepines (BZs) or barbiturates. The existence of such a steroid site on the GBRC is further supported by recent experiments involving the transfection of GABAA receptor cDNAs into a human embryonic kidney cell line. Based on the knowledge of the structure-activity requirements for the interaction of steroids with this novel recognition site, it is conceivable that the development of new anticonvulsant steroids with high therapeutic indices can be achieved.

A steroid recognition site is functionally coupled to an expressed GABA A-benzodiazepine receptor

Pregnane steroids, particularly 3α-hydroxylated metabolites of progesterone, are known to have rapid and profound effects on brain excitability. Recent evidence suggests that the γ-aminobutyric acid (GABA A)-benzodiazepine receptor-Cl − ionophore complex may mediate these actions. The data further suggest that these steroids modulate the complex through a novel site independent of other known sites on the complex. The hypothesis that this site is on the GABA A-benzodiazepine receptor-Cl − ionophore complex is tested in the present study by determining its presence on transiently expressed GABA A-benzodiazepine receptors.

Anticonvulsant profile of the progesterone metabolite 5α-pregnan-3α-ol-20-one

5α-Pregnan-3α-ol-20-one (3α-OH-DHP) is a naturally occuring metabolite of progesterone that can modulate brain excitability through a specific steroid recognition site on the GABA/benzodiazepine receptor-chloride ionophore complex. The anticonvulsant properties of 3α-OH-DHP were determined using standardized anticonvulsant screening tests in mice. This steroid was found to be effective against metrazol-, (+)-bicuculline- and picrotoxin-induced seizures. The steroid has maximum potency against (+)-bicuculline-induced convulsions and no activity against maximal electroshock and strychnine-induced seizures. These findings support the hypothesis that therapeutically useful anticonvulsant steroids active at the putative steroid recognition site associated with the GABA/benzodiazepine receptor-chloride ionophore complex can be identified.

SEQUENCES OF INTEREST: Human Placental 17β-Hydroxysteroid Dehydrogenase is Homologous to NodG Protein ofRhizobium meliloti

The amino acid sequence of human placental 17 beta-hydroxysteroid dehydrogenase (17 beta-OH-steroid dehydrogenase) was found to be similar to that of the NodG protein of Rhizobium meliloti. The computer-based comparison score is 11.5 SD higher than that obtained with 2500 comparisons of randomized sequences of these proteins. The probability of getting such a score by chance is 6 x 10(-31). 17 beta-OH-steroid dehydrogenase is also similar to Klebsiella aerogenes ribitol dehydrogenase and Escherichia coli glucitol-6-phosphate dehydrogenase. We propose that the steroid recognition site on 17 beta-OH-steroid dehydrogenase evolved from an ancestral recognition site for polyols such as ribitol and glucitol-6-phosphate.