Point mutations in the upstream region of the α-galactosidase A gene exon 6 in an atypical variant of Fabry disease

Summary Single point mutations in the upstream region of exon 6 of the -galactosidase A gene were found in two Japanese cases of the cardiac form of Fabry disease; ³⁰¹ArgGln (⁹⁰²GA) in a case that has already been published and ²⁷⁹GlnGlu (⁸³⁵CG) in a new case. They both expressed markedly low, but significant, amounts of residual activity in COS-1 cells. In contrast, two unrelated cases with classic Fabry disease were found to have different point mutations, which showed a complete loss of enzyme activity in a transient expression assay; ³²⁸GlyArg (⁹⁸²GA) in the downstream region of exon 6 in one case and two combined mutations, ⁶⁶GluGln (¹⁹⁶GC)/¹¹²ArgCys (³³⁴CT), in exon 2 in the other. We conclude, on the basis of the results recorded in this study and those in previous reports, that the pathogenesis of atypical Fabry disease is closely associated with point mutations in the upstream region of exon 6 of the -galactosidase A gene.     

Polyvalent GalNAcα1→Ser/Thr (Tn) and Galβ1→3GalNAcα1→Ser/Thr (T <Subscript>α</Subscript>) as the most potent recognition factors involved in <Emphasis Type="Italic">Maclura pomifera</Emphasis> agglutinin–glycan interactions

Summary The agglutinin isolated from the seeds of Maclura pomifera (MPA) recognizes a mucin-type disaccharide sequence, Gal13GalNAc (T) on a human erythrocyte membrane. We have utilized the enzyme-linked lectinosorbent assay (ELLSA) and inhibition assay to more systematically analyze the carbohydrate specificity of MPA with glyco-recognition factors and mammalian Gal/GalNAc structural units in lectin–glycoform interactions. From the results, it is concluded that the high densities of polyvalent GalNAc1Ser/Thr (Tn) and Gal13GalNAc1Ser/Thr (T) glycotopes in macromolecules are the most critical factors for MPA binding, being on a nanogram basis 2.0 × 10⁵, 4.6 × 10⁴ and 3.9 × 10⁴ more active than monovalent Gal, monomeric T and Tn glycotope, respectively. Other carbohydrate structural units in mammalian glycoconjugates, such as human blood group Sd (a⁺) related disaccharide (GalNAc14Gal) and P^k/P₁ active disaccharide (Gal14Gal) were inactive. These results demonstrate that the configurations of carbon-4 and carbon-2 are essential for MPA binding and establish the importance of affinity enhancement by high-density polyvalencies of Tn/T glycotopes in MPA–glycan interactions. The overall binding profile of MPA can be defined in decreasing order as high density of polyvalent Tn/T (M.W. > 4.0 × 10⁴) >> Tn-containing glycopeptides (M.W. < 3.0 × 10³) > monomeric T/Tn and P (GalNAc13Gal) > GalNAc > Gal >> Man, LAra, DFuc and Glc (inactive). Our findings should aid in the selection of this lectin for elucidating functions of carbohydrate chains in life processes and for applications in the biomedical sciences.     

Mutational specificity of ethyl methanesulfonate in excision-repair-proficient and -deficient strains of Drosophila melanogaster

Summary The vermilion gene was used as a target to determine the mutational specificity of ethyl methanesulfonate (EMS) in germ cells of Drosophila melanogaster. To study the impact of DNA repair on the type of mutations induced, both excision-repair-proficient (exr ⁺) and excision-repair-deficient (exr ^–) strains were used for the isolation of mutant flies. In all, 28 mutants from the exr ⁺ strain and 24 from the exr ^– strain, were characterized by sequence analysis. In two mutants obtained from the exr ⁺ strain, small deletions were observed. All other mutations were caused by single base-pair changes. In two mutants double base-pair substitutions had occurred. Of the mutations induced in the exr ⁺ strain, 22 (76%) were GCAT transitions, 3 (10%) ATTA transversions, 2 (6%) GCTA transversions and 2 (6%) were deletions. As in other systems, the mutation spectrum of EMS in Drosophila is dominated by GCAT transitions. Of the mutations in an exr ^– background, 12 (48%) were GCAT transitions, 7 (28%) ATTA transversions, 5 (20%) GCTA transversions and 1 (4%) was a ATGC transition. The significant increase in the contribution of transversion mutations obtained in the absence of an active maternal excision-repair mechanism, clearly indicates efficient repair of N-alkyl adducts (7-ethyl guanine and 3-ethyl adenine) by the excision-repair system in Drosophila germ cells.     

Possible intrachromosomal duplication in a case of trisomy 9p

Summary A 5-year-old boy with multiple minor anomalies and mental retardation was found to have chromosomal condition of 46,XY,inv dup(9p)(pterp13::p21p24::p13qter). The clinical features of the propositus fit well with those of trisomy 9p which have been established to be a clinical entity.     

One-bond carbon-proton coupling constants: Angular dependence in β-linked oligosaccharides

Summary The angular dependence of¹J_C,H in model compounds related to -linked oligosaccharides has been established by FPT INDO quantum chemical calculations. Values calculated for models of (1 1)-, (1 2)-, (1 3)- and (1 4)-linked disaccharides were compared, and the effect of the orientation of HO-2 elucidated. The angular dependence of¹J_C,H on the torsional angles ^H and ^H and the solvent dielectric constant (s) was characterized in the form:¹J_C,H = A cos2+B cos + C sin2 + D since + E + Fe. The¹J_C,H values, measured by DEPT methods for C-1-H-1 and C-X-H-X in cellobiose, cyclic trisaccharide and hexopyranoses were used to adjust the calculated angular dependences. Based on the occurrence of the conformers for agarobiose, neoagarobiose, mannobiose and methyl -xylobioside, the thermodynamically averaged <¹J_C,H > values were calculated. The results obtained (<¹J_C-1,H-1 > 162.4, <¹J_{C-4, H-4} > 147.6 Hz for methyl -xylobioside; <¹J_C-1,H-1 > 162.4 and <¹J_C-4,H-4] > 147.6 Hz for mannobiose; <¹J_C-1,H-1 > 162.8 Hz for neo agarobiose and <¹J_C-1,H-1 > 163.2 Hz for agarobiose) agree well with the experimental values of 162.7, 147.5, 160.4, 147.2, 160.9 and 165.7 Hz, respectively.     

Mutation to rifampicin resistance at the beginning of the RNA polymerase beta subunit gene in Escherichia coli

Summary The unusual recombinant plasmid pRC19 carrying the N-terminal fragment of the Escherichia coli RNA polymerase rpoB gene was found to specify high level rifampicin resistance of E. coli cells. Sequence analysis of this plasmid revealed one substitution only: transversion GT, leading to amino acid substitution Val¹⁴⁶Phe. This mutational change marks the second domain of the subunit involved in rifampicin binding.     

Regional mapping of catalase and Wilms tumor—aniridia,genitourinary abnormalities,and mental retardation triad loci to the chromosome segment 11p1305→p1306

Summary Gene dosage effects for catalase (CAT) were studied in two unrelated patients with an interstitial deletion involving 11p13 to determine precisely the sites of the genes for CAT and the Wilms tumor—aniridia, genitourinary abnormalities, and mental retardation triad (WAGR) in the 11p13 band. Case 1 had the aniridia-Wilms tumor association, and case 2 showed the AGR triad. The karyotypes identified by high resolution banding techniques were 46,XY,del(11)(pterp13::p11.11qter) for case 1 and 46,XY,t(2;17) (q23;q25), del(11) (pterp13::p11.2 qter) for case 2. In both cases, the distal breakpoints of the deleted chromosomes 11 appeared to have occurred on the middle portion of 11p13 (11p1305p1306). The level of erythrocyte CAT activities in case 1 was reduced (47% of normal), while that in case 2 was normal. The results suggested not only that both the CAT and WAGR should be mapped to chromosome region 11p1305p1306, but also that in this region the CAT locus is more distally placed than the WAGR locus. Because of the proximity of the two gene loci, assays of erythrocyte CAT may be useful to identify a submicroscopic deletion in some patients with sporadic aniridia and to predict a risk of developing Wilms tumor.     

Molecular characterization of β-thalassemia in Czechoslovakia

Summary We have identified different -thalassemia mutations in 93 members of 34 families of Czech or Slovakian descent using gene amplification, hybridization with specific ³²P-labeled oligonucleotide probes, sequencing of amplified DNA, and gene mapping. The GA mutation at IVS-I-1 was found in 18 families; other Mediterranean mutations were IVS-II-1 (GA), IVS-II-745 (CG), IVS-I-110 (GA), and codon 39 (CT); these were present in 9 additional families. The GT mutation at codon 121, known to cause Heinzbody -thalassemia, was present in 3 families, and the frameshift at codons 82/83 (-G), first described in the Azerbaijanian population, in 2 families. A newly discovered allele was a frameshift at codons 38/39 (-C). One -thalassemia allele was incompletely characterized. We observed in 2 families a TC mutation at position +96 UTR (untranslated region) relative to the termination codon; this mutation likely is a rare polymorphism, -Thalassemia was rare; only one person carried the -^3.7 heterozygosity, and one other had a yet to be identified -thalassemia-1, while seven had the ^{anti 3.7} triplication.     

Structures of the N-linked sugar chains in the PAS-6 glycoprotein from the bovine milk fat globule membrane

The structures of the N-linked sugar chains in the PAS-6 glycoprotein (PAS-6) from the bovine milk fat globule membrane were determined. The sugar chains were liberated from PAS-6 by hydrazinolysis, and the pyridylaminated sugar chains were separated into a neutral (6N) and two acidic chains (6M and 6D), the acidic sugar chains then being converted to neutral sugar chains (6MN and 6DN). 6N was separated into two neutral fractions (6N13 and 6N5.5), while 6MN and 6DN each gave a single fraction (6MN13 and 6DN13). The structure of 6N5.5, which was the major sugar chain in PAS-6, is proposed to be Man16 (Man13) Man14GlcNAc14GlcNAc-PA; 6N13, 6MN13 and 6DN13 are proposed to be Gal13Gal14GlcNAc12Man16 (Gal13Gal14GlcNAc12Man13) Man14GlcNAc14 (Fuc16)GlcNAc-PA;6M and 6D had 1 or 2 additional NeuAc residues at the non-reducing ends of 6MN13 and 6DN13, respectively. © 1998 Rapid Science Ltd     

Plastid Sequence Evolution: A New Pattern of Nucleotide Substitutions in the Cucurbitaceae

Nucleotide substitutions (i.e., point mutations) are the primary driving force in generating DNA variation upon which selection can act. Substitutions called transitions, which entail exchanges between purines (A=adenine, G=guanine) or pyrimidines (C=cytosine, T=thymine), typically outnumber transversions (e.g., exchanges between a purine and a pyrimidine) in a DNA strand. With an increasing number of plant studies revealing a transversion rather than transition bias, we chose to perform a detailed substitution analysis for the plant family Cucurbitaceae using data from several short plastid DNA sequences. We generated a phylogenetic tree for 19 taxa of the tribe Benincaseae and related genera and then scored conservative substitution changes (e.g., those not exhibiting homoplasy or reversals) from the unambiguous branches of the tree. Neither the transition nor (A+T)/(G+C) biases found in previous studies were supported by our overall data. More importantly, we found a novel and symmetrical substitution bias in which Gs had been preferentially replaced by A, As by C, Cs by T, and Ts by G, resulting in the GACTG substitution series. Understanding this pattern will lead to new hypotheses concerning plastid evolution, which in turn will affect the choices of substitution models and other tree-building algorithms for phylogenetic analyses based on nucleotide data.     

Characterization of the sugar-binding specificity of the toxic lectins isolated from Abrus pulchellus seeds

The sugar-binding specificity of the toxic lectins from Abrus pulchellus seeds was investigated by combination of affinity chromatography of glycopeptides and oligosaccharides of well-defined structures on a lectin-Sepharose column and measurement of the kinetic interactions in real time towards immobilized glycoproteins. The lectins showed strong affinity for a series of bi- and triantennary N-acetyllactosamine type glycans. The related asialo-oligosaccharides interact more strongly with the lectins. The best recognized structures were asialo-glycopeptides from fetuin. Accordingly, the kinetic interaction with immobilized asialofetuin was by far the most pronounced. Human and bovine lactotransferrins and human serotransferrin interacted to a lesser extent. The interaction with asialofetuin was inhibited by galactose in a dose dependent manner. Lactose, N-acetyllactosamine and lacto-N-biose exhibited similar degree of inhibition while N-acetylgalactosamine was a poor inhibitor. These results suggested that the carbohydrate-binding site of the Abrus pulchellus lectins was specific for galactose and possess a remarkable affinity for the sequences lactose [-D-Gal-(14)-D-Glc], N-acetyllactosamine [-D-Gal-(14)-D-GlcNAc] and lacto-N-biose [-D-Gal-(13)-D-GlcNAc].     

Energy conservation in whole cells of the methylotrophic bacterium Methylophilus methylotrophus

Respiratory chain phosphorylation has been investigated in the methylotrophic bacterium Methylophilus methylotrophus following the addition of oxidisable substrates to aerobic, whole cell suspensions. Initial-rate experiments showed that ATP synthesis occurred at the overall expense of AMP and inorganic phosphate via the sequential action of the ATP phosphohydrolase and adenylate kinase; some of the nascent ATP was rapidly used to synthesis nonadenine nucleoside triphosphates. After being corrected for ATP turnover, Pi/O quotients of 0.46 to 0.54, 0.77 and 1.37 nmol/ng-atom O were obtained for the oxidation of methanol dehydrogenase-linked substrates (methanol, ethanol and acetaldehyde), duroquinol and formate (NAD⁺-linked) respectively. These values were proportional to the H⁺/O and/or K⁺/O quotients exhibited by these substrates, and yielded an average H⁺/ATP (H⁺/Pi) quotient of 4.2 ng-ion H⁺/nmol. Steady-state experiments showed that the extent of cellular energisation varied with the respiration rate but was always in the order methanol > duroquinol > acetaldehyde, thus indicating that under these longer-term conditions methanol was completely oxidised to yield PQQH₂ and 2NAD(P)H. These results are discussed in terms of the various reactions which lead to the generation or utilisation of the protonmotive force in this organism.Abbreviations FCCP carbonylcyanide p-trifluoromethyxyphenyl-hydrazone - bulk phase, transmembrane electrochemical potential difference of protons ( ) - pH bulk phase, transmembrane pH difference (pH_in–pH_out) - bulk phase, transmembrane electrical potential difference (_in - _out) - [P] concentration of anhydride phosphate bonds in adenine nucleotides (2[ATP]+[ADP]) - FPLC fast protein liquid chromatography - PQQ pyrroloquinoline quinone - Gp phosphorylation potential     

Absorbance difference spectra of the S-state transitions in Photosystem II core particles

Redox changes of the oxygen evolving complex in PS II core particles were investigated by absorbance difference spectroscopy in the UV-region. The oscillation of the absorbance changes induced by a series of saturating flashes could not be explained by the minimal Kok model (Kok et al. 1970) consisting of a 4-step redox cycle, S₀ S₁ S₂ S₃ S₀, although the values of most of the relevant parameters had been determined experimentally. Additional assumptions which allow a consistent fit of all data are a slow equilibration of the S₃ state with an inactive state, perhaps related to Ca²⁺-release, and a low quantum efficiency for the first turnover after dark-adaptation. Difference spectra of the successive S-state transitions were determined. At wavelengths above 370 nm, they were very different due to the different contribution of a Chl bandshift in each spectrum. At shorter wavelengths, the S₁ S₂ transition showed a difference spectrum similar to that reported by Dekker et al. 1984b and attributed to an Mn(III) to Mn(IV) oxidation. The spectrum of absorbance changes associated with the S₂ S₃ transition was similar to that reported by Lavergne 1991 for PS II membranes. The S₀ S₁ transition was associated with a smaller but still substantial absorbance increase in the UV. Differences with the spectra reported by Lavergne 1991 are attributed to electrostatic effects on electron transfer at the acceptor side associated with the S-state dependence of proton release in PS II membranes.Abbreviations Bis-Tris (bis[2-hydroxyethyl]imino-tris[hydroxymethyl]methane) - DCBQ 2,5-dichloro-p-benzoquinone - DCMU 3-(3,4-dichlorophenyl)-1,1-dimethylurea - PS II Photosystem II - Q_A secondary electron acceptor of PS II - S₀ to S₄ redox state of the oxygen evolving complex - Z secondary electron donor of PS II     

Cytochrome P-450-dependent catabolism of triethanolamine in Rhodotorula mucilaginosa

The yeast Rhodotorula mucilaginosa was able to grow in media containing triethanolamine or diethanolamine as the sole nitrogen source. During growth in the presence of triethanolamine, extracts of yeast cells contained increased levels of cytochrome P-450 dependent monooxygenase which catalyzed the oxidative N-dealkylation of aminoalcohols. Formation of diethanolamine, ethanolamine and glyoxylate from triethanolamine was demonstrated, and the identity of the products was verified by thin layer chromatography. These observations suggested the following scheme of triethanolamine catabolism: triethanolamine diethanolamine + glycolaldehyde, diethanolamine ethanolamine + glycolaldehyde, ethanolamine NH₃ + glycolaldehyde glycolate glyoxylate glycerate pathway.     

β-D-Glycan synthases and the CesA gene family: lessons to be learned from the mixed-linkage (1→3),(1→4)β-D-glucan synthase

Cellulose synthase genes (CesAs) encode a broad range of processive glycosyltransferases that synthesize (14)-D-glycosyl units. The proteins predicted to be encoded by these genes contain up to eight membrane-spanning domains and four `U-motifs' with conserved aspartate residues and a QxxRW motif that are essential for substrate binding and catalysis. In higher plants, the domain structure includes two plant-specific regions, one that is relatively conserved and a second, so-called `hypervariable region' (HVR). Analysis of the phylogenetic relationships among members of the CesA multi-gene families from two grass species,Oryza sativa and Zea mays, with Arabidopsis thaliana and other dicotyledonous species reveals that the CesA genes cluster into several distinct sub-classes. Whereas some sub-classes are populated by CesAs from all species, two sub-classes are populated solely by CesAs from grass species. The sub-class identity is primarily defined by the HVR, and the sequence in this region does not vary substantially among members of the same sub-class. Hence, we suggest that the region is more aptly termed a `class-specific region' (CSR). Several motifs containing cysteine, basic, acidic and aromatic residues indicate that the CSR may function in substrate binding specificity and catalysis. Similar motifs are conserved in bacterial cellulose synthases, the Dictyostelium discoideum cellulose synthase, and other processive glycosyltransferases involved in the synthesis of non-cellulosic polymers with (14)-linked backbones, including chitin, heparan, and hyaluronan. These analyses re-open the question whether all the CesA genes encode cellulose synthases or whether some of the sub-class members may encode other non-cellulosic (14)-glycan synthases in plants. For example, the mixed-linkage (13)(14)-D-glucan synthase is found specifically in grasses and possesses many features more similar to those of cellulose synthase than to those of other -linked cross-linking glycans. In this respect, the enzymatic properties of the mixed-linkage -glucan synthases not only provide special insight into the mechanisms of (14)-glycan synthesis but may also uncover the genes that encode the synthases themselves.     

Examining the relationship between secondary structure and antibody recognition in immunopeptides from foot-and-mouth disease virus

Summary The conformation of a peptide that represents antigenic site A of foot-and-mouth disease virus strain C-S8c1 (residues 136–156 of VP1; YTASARGDLAHLTTTHARHLP) has been studied by circular dichroism and compared with three analogs that reproduce amino acid substitutions at position 146 (HisArg, Gln or Asp) which affect antibody recognition. Four other peptides, incorporating replacements at position 147 predicted to maintain (LeuIle, Nle and Ala) or disrupt (LeuGly) helical structure at this site, have also been studied. In aqueous solution or in 4 M urea, the spectra of all eight peptides were typical of aperiodic conformation and independent of concentration or pH. However, upon addition of solvents such as methanol or hexafluoroisopropanol, spectral patterns evidenced significant levels (ca. 50%) of helical structure. The single residue substitutions at positions 146 and 147 caused minor to significant variations in the calculated amount of -helix of the peptides. An attempt to relate these changes in helical content to the antigenic behaviour of the peptides towards five monoclonal antibodies elicited with virus and mapping at site A could not find any straightforward correspondence between the two sets of results. The parent peptide and its His¹⁴⁶Arg analog were also analyzed by circular dichroism in the presence of the Fab fragment of SD6, a monoclonal antibody mapping at site A and much less reactive with viruses carrying the referred mutation. Although a peptide-antibody interaction was evident from spectral changes, careful inspection of the difference spectra (peptide-Fab minus Fab) of both peptides failed to detect any significant distinction between them that could be attributed to their different immunoreactivity. While these findings do not necessarily conflict with previous reports that the interaction of antigenic site A with antibodies is mediated to some extent by the adoption of a helix structure, they suggest that, at least for C-serotype viruses, other structural features in addition to a helical conformation are critically involved in antigenic recognition.     

Three chromosomes' (7;9;22) rearrangement and the origin of the Philadelphia chromosome

Summary A woman with chronic myelocytic leukemia had the Philadelphia chromosome and a complex four-break—three-chromosome rearrangement. The q32q34 portion of chromosome 9 is translocated to band q22 of chromosome 7, and at the end of this segment is attached the deleted q11 qter portion of chromosome 22. A review of 12 cases of the Philadelphia chromosome originating by the rearrangement of three or more chromosomes reveals that chromosomes 9 and 22 are always involved, while the third chromosome is a different one in each case. We discuss the hypothesis that the 22q segment is always specifically attached to band 9q34 wherever this portion of 9q is transposed.Address for offprint requests: Prof. M. Fraccaro, Gruppo Euratom, Via Forlanini, 14, I-27100 Pavia, Italy     

Possible complex translocation t(9; 14; 13)(q12;pl?;q31) in mother of a child with 9p-trisomy syndrome

Summary A mentally retarded boy with trisomy 9p is described. This trisomy arose through aberrant segregation of translocation chromosome during meiosis in his mother, who has a complex translocation involving chromosomes 9, 13, and 14. Based on both G-, Q-banding, and DNA replication patterns, the patient's karyotype was identified as 47,XY,-13, +(9;13) (9pter9q12::13q3113qter), +t(13;14) (13pter13q31::14pl?14pter). We suppose his mother's karyotype to be 46,XX,-9,-13,-14,+t(9;13) (9pterq12::13q3113qter), +t(13;14) (13pter13q31::14pl?14pter), +t(9;14) (9qter9q12::14pl?14qter). His phenotypically normal brother and sister are also carriers, having the same translocation chromosome as their mother. Clinical findings of the patient included peculiar face with hypertelorism, prominent nasal bridge and deformed helix, marked delay of osseous development, hypoplastic phalangia in fingers and toes, dysplastic nails and absence of digital triradii.     

Clinical and cytogenetic studies of the Prader-Willi syndrome: Evidence of phenotype-karyotype correlation

Summary Twenty-seven patients with the presumed diagnosis of Prader-Willi syndrome (PWS) were studied clinically and cytogenetically. The patients were classified into three study groups on the basis of their clinical pictures: group 1 with 12 children meeting the strict diagnostic criteria for PWS; group 2 with nine floppy infants and young children, aged 3 years or less, without obesity and hyperphagia; and group 3 with six older children in whom some characteristic features of the syndrome were absent. High-resolution GTG banding of prometaphase chromosomes revealed del(15)(q11.1;q12) in eleven and t(15;15)(qterp11.2::q12qter) in one of the twelve group 1 patients. In group 2, four patients had del(15)(q11.1;q12), one had t(15;15)(qterp11.1::q13qter), and the remaining four had normal karyotypes. The deleted segment common to the 17 patients with the chromosome aberrations was confined to subband 15q11.2. On the other hand, all six group 3 patients had normal karyotypes. These findings indicated that when strictly defined PWS is absolutely correlated with chromosome 15 aberrations, i.e, there is a positive phenotype-karyotype correlation, and that the aberrations are etiologically related to the syndrome. Parental origin of the deleted chromosome was determined in seven patients, with OFQ-heteromorphisms being used as hereditary markers. The deleted chromosome originated from the paternal chromosome 15 in six patients and from the maternal 15 in one.     

A Mathematical Model of the Cerebellar-Olivary System I: Self-Regulating Equilibrium of Climbing Fiber Activity

We use a mathematical model to investigate how climbing fiber-dependent plasticity at granule cell to Purkinje cell (grPkj) synapses in the cerebellar cortex is influenced by the synaptic organization of the cerebellar-olivary system. Based on empirical studies, grPkj synapses are assumed to decrease in strength when active during a climbing fiber input (LTD) and increase in strength when active without a climbing fiber input (LTP). Results suggest that the inhibition of climbing fibers by cerebellar output combines with LTD/P to self-regulate spontaneous climbing fiber activity to an equilibrium level at which LTP and LTD balance and the expected net change in grPkj synaptic weights is zero. The synaptic weight vector is asymptotically confined to an equilibrium hyperplane defining the set of all possible combinations of synaptic weights consistent with climbing fiber equilibrium. Results also suggest restrictions on LTP/D at grPkj synapses required to produce synaptic weights that do not drift spontaneously.