Sites of gibberellin biosynthesis in pea seedlings

Potential sites of gibberellin biosynthesis in 10-day-old `Alaska' pea (Pisum sativum L.) seedlings were investigated using a cell-free ezyme system capable of incorporating [¹⁴C]-mevalonic acid into ent-kaurene. In peas, ent-kaurene is assumed to be a committed intermediate in the gibberellin biosynthetic pathway. Comparative results from enzyme assays using extracts from shoot tips, leaf blades, internodes, and root tips indicate that the highest capacity for ent-kaurene (and presumably gibberellin) synthesis is in those tissues with the greatest potential for growth. The highest rates were obtained with extracts prepared from the fifth (youngest) internode, the fourth (youngest) expanded leaf, and the shoot tip itself. This report represents the first direct evidence that the enzymes responsible for early stages in gibberellin biosynthesis occur in internode tissues with potential for rapid elongation.     

Conversion of geranylgeranyl pyrophosphate to ent-kaurene in enzyme extracts of sonicated chloroplasts

Farnesyl pyrophosphate-[¹⁴C] and geranylgeranyl pyrophosphate-[¹⁴C] were biosynthesized from mevalonic acid-[2-¹⁴C] by cell-free enzyme extracts of pea (Pisum sativum) cotyledons containing MgCl₂, MnCl₂, ATP and AMO-1618. Maximum yields of farnesyl pyrophosphate were obtained after 30 min incubation while geranylgeranyl pyrophosphate was the primary product after 180 min. Biosynthesized geranylgeranyl pyrophosphate-[¹⁴C] served as an efficient substrate for ent-kaurene biosynthesis in reaction mixtures containing cotyledon enzymes when AMO-1618 was omitted. Enzyme extracts from green pea shoot tips and chloroplasts also converted geranylgeranyl pyrophosphate to ent-kaurene in very low yields. Ent-kaurene production from mevalonic acid-[2-¹⁴C] in extracts of pea shoot tips was also enhanced by addition of chloroplast enzymes. This evidence indicates that kaurene synthetase is present in pea chloroplasts and adds to the possibility that some gibberellin biosynthesis may be compartmentalized in those organelles.     

ent-Kaurene Biosynthesis in Cell-Free Extracts of Excised Parts of Tall and Dwarf Pea Seedlings

Investigations on the sites of ent-kaur-16-ene (ent-kaurene) biosynthesis were conducted with cell-free extracts from several excised parts of 10-, 13-, and 16-d-old tall and dwarf pea (Pisum sativum L.) seedlings. [¹⁴C]Mevalonic acid was incorporated into ent-kaurene in cell-free extracts from young developing leaves and elongating internodes of tall (`Alaska') and dwarf (`Progress No.9') pea seedlings at all three stages of development. ent-Kaurene biosynthesis also occurred readily in cell-free extracts from shoot tips, petioles, and stipules near the young elongating internodes. The ent-kaurene-synthesizing activity found in young developing tissues declined as tissues matured. Little or no activity was detectable in enzyme extracts from cotyledons and root tips at different stages. In light grown tall pea internodes ent-kaurene-synthesizing activity was low as they began to elongate, reached a maximum when the internodes reached about 2 cm in length and declined as they matured. Activity in extracts of dwarf shoot tips and internodes was generally lower than in equivalent tall plants, but the activity in dwarf leaves and stipules was somewhat higher than in tall plants. With the exception of root tips, there is a strong correlation between growth potential of a tissue and the rate of ent-kaurene biosynthesis in extracts from that tissue.     

Biosynthesis of (-)-Kaurene in Cell-free Extracts of Immature Pea Seeds

Mevalonate-¹⁴C was incorporated into (—)-kaurene-¹⁴C in cell-free extracts of immature pea (Pisum sativum L.) seeds. The identification of ¹⁴C-product as (—)-kaurene was based on: A) comparison with authentic (—)-kaurene on thin-layer and gas-liquid chromatography; and B) oxidation of ¹⁴C-product and (—)-kaurene with osmium tetroxide to form the common derivative kaurane-16,17-diol. The enzyme system is heat labile and is dependent upon ATP and Mg²⁺ or Mn^2-, with Mn²⁺ being a more effective activator than Mg²⁺. The reaction rate was proportional to enzyme concentration in reaction mixtures containing 0.45 to 1.8 mg protein n/ml, and was linear with time through 120 minutes in standard reaction mixtures. Enzyme preparations from immature seeds of tall and dwarf peas appeared to synthesize (—)-kaurene at the same rate. Synthesis of (—)-kaurene was readily inhibited by Amo-1618. (2-Chloroethyl)-trimethylammonium chloride (CCC) also inhibited (—)-kaurene synthesis; however, approximately 1000-fold higher concentrations of CCC were required to evoke the same percentages of inhibition as Amo-1618.     

Induction ofent-kaurene biosynthesis by low temperature in dwarf peas

Germinating pea (Pisum sativum L.) seeds of two dwarf cultivars, “Progress No. 9” and “Green Arrow”, and two tall cultivars, “Alaska” and “Alderman”, were treated with low temperature (3–5°C) for 14 days and then transferred to normal growing conditions (19–21°C for 16 h/14.5–16.5°C for 8 h) for an additional 10 days. Biosynthesis of [¹⁴C]ent-kaurene from [¹⁴C]2-mevalonic acid (2-MVA) was assayed in cell-free enzyme extracts prepared from shoot tips 10 days after cold treatment and was compared with activity in enzyme extracts prepared from noncold-treated, 10-day-old control plants. Shoot lengths of cold-treated plants were measured throughout a 35-day period and compared with shoot lengths of plants grown without cold treatment for 25–35 days. Low temperature induced a five-to 10-fold enhancement ofent-kaurene, hence potentially gibberellin (GA), biosynthesis in seedlings of the two dwarf cultivars but not in the tall cultivars. However, the lack of an increase in growth rate in the cold-treated dwarfs indicated that endogenous GA biosynthesis remained blocked at some point beyondent-kaurene in the biosynthetic pathway. Since the late-flowering “Alderman” cultivar did not exhibit enhanced biosynthesis ofent-kaurene, it appears that if vernalization in late-flowering cultivars of peas is correlated with enhanced GA biosynthesis, it is not the early part of the biosynthetic pathway which is affected.     

Apparent endogenous circadian rhythm inent-kaurene biosynthesis

Net synthesis of [¹⁴C]ent-kaurene from [¹⁴C]2-mevalonic acid was assayed in cell-free enzyme extracts prepared from Alaska pea (Pisum sativum L.) seedlings throughout 44 h of a regimen consisting of a 16-h day and an 8-h night. Activities generally followed an upward trend during the dark period and a downward trend during the photoperiod. Activity was also assayed in enzyme extracts prepared at intervals during a 12-h photoperiod and a following, continuous 36-h dark period after entrainment of plants to a regimen of 12-h days and 12-h nights.Ent-kaurene synthesis activity again followed an upward trend in enzyme extracts prepared during what would have been the entrainment dark period, and a downward trend during the entrainment photoperiod. The apparent endogenous rhythm ofent-kaurene biosynthesis may have implications for the regulation of gibberellin biosynthesis.     

Correlations of Growth Rate and De-etiolation with Rate of Ent-Kaurene Biosynthesis in Pea (Pisum sativum L.)

Biosynthesis of the gibberellin precursor ent-kaurene-¹⁴C from mevalonic acid-2-¹⁴C was assayed in cell-free extracts of shoot tips of etiolated and light-grown Alaska (normal) and Progress No. 9 (dwarf) peas (Pisum sativum L.). During ontogeny of light-grown Alaska peas, kaurene-synthesizing activity increased from an undectectable level in 3-day-old epicotyls to a maximum in shoot tips of 9-day-old plants and remained relatively constant thereafter until postanthesis. The capacity for kaurene synthesis in extracts from shoot tips of 10-day-old etiolated Alaska seedlings increased approximately exponentially during the first 12 hr of de-etiolation in continuous high intensity white light and remained relatively constant during the succeeding 24 hr of irradiation. Extracts from light-grown Alaska (normal) shoot tips possessed greater capacity for kaurene synthesis than did extracts from light-grown Progress No. 9 (dwarf) shoot tips. Extracts from shoot tips of either light-grown cultivar displayed greater kaurene-synthesizing capacity than was observed in extracts from their dark-grown counterparts. It is concluded that gibberellin biosynthesis in pea shoot tips is subject to partial regulation by factors controlling the rate of biosynthesis of kaurene.     

Apparent endogenous circadian rhythm inent-kaurene biosynthesis

Net synthesis of [¹⁴C]ent-kaurene from [¹⁴C]2-mevalonic acid was assayed in cell-free enzyme extracts prepared from Alaska pea (Pisum sativum L.) seedlings throughout 44 h of a regimen consisting of a 16-h day and an 8-h night. Activities generally followed an upward trend during the dark period and a downward trend during the photoperiod. Activity was also assayed in enzyme extracts prepared at intervals during a 12-h photoperiod and a following, continuous 36-h dark period after entrainment of plants to a regimen of 12-h days and 12-h nights.Ent-kaurene synthesis activity again followed an upward trend in enzyme extracts prepared during what would have been the entrainment dark period, and a downward trend during the entrainment photoperiod. The apparent endogenous rhythm ofent-kaurene biosynthesis may have implications for the regulation of gibberellin biosynthesis.     

Comparison between Biosynthesis of ent-Kaurene in Germinating Tomato Seeds and Cell Suspension Cultures of Tomato and Tobacco

Biosynthesis of ent-kaurene was investigated in extracts of cell suspension cultures derived from tobacco callus (Nicotiana tabacum L.), tomato callus (Solanum lycopersicum L.), and in germinating tomato seeds. Incubation of extracts derived from the two cell cultures with either isopentenyl pyrophosphate-¹⁴C or with ¹⁴C-labeled mevalonate, followed by alkaline phosphatase hydrolysis, resulted in the formation of trans-geranylgeraniol-¹⁴C and trans-farnesol-¹⁴C. The corresponding pyrophosphates of trans-geranyl-geraniol-¹⁴C and trans-farnesol-¹⁴C were also detected. No detectable amount of ent-kaurene-¹⁴C was produced by these enzymatic preparations when trans-geranylgeranyl-¹⁴C pyrophosphate served as substrate. However, copalyl-¹⁴C pyrophosphate served as a substrate for the production of ent-kaurene. Cell-free extracts derived from germinating tomato seeds catalyzed the formation of ent-kaurene-¹⁴C from mevalonate-¹⁴C, isopentenyl-¹⁴C pyrophosphate, trans-geranylgeranyl-¹⁴C pyrophosphate, and copalyl-¹⁴C pyrophosphate.     

Ent-kaurene,occurrence and metabolism in Hordeum distichon

Barley grains contain hydrocarbons, including a material indistinguishable from ent-kaurene by GLC, and which after appropriate chemical conversions contain material behaving like ent-kauran-16,17-diol, ent-kaurene norketone and ent-17-nor-kaurane on TLC and GLC. The presence of ent-kaurene was confirmed by conversion to ent-kauran-16-ol and, following formation of acetate-[³H], recrystallization to constant specific activity with unlabelled carrier. In the initial ca. 15 hr of germination, preceding the rise in endogenous gibberellins, the level of ent-kaurene falls. Exogenous ent-kaurene-[¹⁴C] was not metabolized by intact barley grains. ent-Kauran-16,17-epoxide was formed non-enzymically by boiled extracts. Unboiled homogenates also formed ent-kauran-17-ol and ent-kauran-16,17-diol. The diol appeared to be formed from the epoxide, but the ent-kauran-17-ol was not. No recognized gibberellin precursors were detected. Nevertheless, endogenous ent-kaurene may be the stored biosynthetic precursor of gibberellins in germinating barley grains.     

Gibberellin estimation and biosynthesis in germinating Hordeum distichon

Gibberellic acid (GA₃) and 13-deoxy-gibberellic acid (GA₇) were identified in extracts of germinating barley as their ¹⁴C-methyl esters. The maximal level of GA₃ was estimated by an isotopic dilution procedure to be 1·5 ng per grain. Germinating barley incorporated 2-¹⁴C-mevalonic acid into several terpenes, whose specific radioactivities were measured, but incorporation into GA₃ could not be detected. Cell-free embryo extracts from germinating barley converted 2-¹⁴C-mevalonic acid into isopentenol, dimethylallyl alcohol, farnesol and squalene, while ¹⁴C-isopentenyl pyrophosphate was incorporated into geraniol, farnesol, geranylgeraniol and squalene. There was no detectable incorporation into the gibberellin intermediate ent-kaurene.     

ent-Kaurene biosynthesis in a cell-free system from wheat (Triticum aestivum L.) seedlings and the localisation of ent-kaurene synthetase in plastids of three species

A cell-free system capable of converting [¹⁴C]geranylgeranyl diphosphate to ent-[¹⁴C]kaurene and to an unidentified acid-hydrolysable compound was obtained from the basal portions of 5-d-old shoots of wheat seedlings (Triticum aestivum L.). By means of marker enzyme activities, the synthesis of ent-kaurene and the unknown compound could be quantitatively assigned to a plastid fraction obtained by Percoll-gradient centrifugation of the homogenate. The enzyme activities were located within the plastids, probably in the stroma, because they withstood trypsin treatment of the intact plastids, and the plastids had to be broken to release the activity, which was then obtained in soluble form. Plastid membranes had no activity. Plastid stroma preparations obtained from pea (Pisum sativum L.) shoot tips and pumpkin (Cucurbita maxima L.) endosperm also yielded ent-kaurene synthetase activity, but did not form the unknown compound. The exact nature of the active plastids was not ascertained, but the use of methods for proplastid isolation was essential for full activity, and the active tissues are all known to contain high proportions of proplastids, developing chloroplasts or leucoplasts. We therefore believe that ent-kaurene synthesis may be limited to these categories. Mature chloroplasts from the wheat leaves did not contain ent-kaurene synthetase activity and did not yield the unknown component. Incorporation of [¹⁴C]geranylgeranyl diphosphate into ent-[¹⁴C]kaurene and the unknown component was assayed by high-performance liquid chromatography with on-line radiocounting. ent-[¹⁴C]Kaurene was identified by Kovats retention index and full mass spectra obtained by combined gas chromatography-mass spectrometry. The unknown component was first believed to be copalyl diphosphate, because it yielded a compound on acid hydrolysis, which migrated like copalol on high-performance liquid chromatography and gave a mass spectrum very similar to that of authentic copalol. However, differences in the mass spectrum and in retention time on capillary gas chromatography excluded identity with copalol. Furthermore, the unhydrolysed compound was not converted to ent-kaurene by a cell-free system from C. maxima endosperm as copalyl diphosphate would have been.Abbreviations ADH alcohol dehydrogenase - AMO 1618 2isopropyl-4-(trimethylammoniumchloride)-5-methylphenyl piperi-dine-1-carboxylate - BSA bovine serum albumin - DTT dithioth-reitol - GA_n gibberellin A_n - GAPDH NADP⁺-glyceraldehyde 3-phosphate dehydrogenase - GC-MS combined gas chromatography-mass spectrometry - GGPP all trans-isomer of geranyl-geranyl diphosphate - KS ent-kaurene synthetase - MDH malate dehydrogenase - MAA mevalonate activating activity - SOR shikimate oxidoreductase We thank Mrs. Gudrun Bodtke and Mrs. Dorothee Dasbach for able technical assistance, Prof. L.N. Mander (Australian National University, Canberra, Australia) for ent-[²H₂]kaurene and Dr. Yuji Kamiya (RIKEN, Saitama, Japan) for geranylgeraniol and copalol. The work was supported by the Deutsche Forschungsgemeinschaft.     

Presence of endogenous ent-kaurene in a microsomal preparation from Cucurbita maxima L. Endosperm and implications for kinetic studies of ent-kaurene oxidase

Microsomal and soluble cell-free extracts prepared from liquid endosperm of Cucurbita maxima L. were found to contain high concentrations of endogenous ent-kaurene and ent-kaurenol by gas chromatography-mass spectrometry-chemical ionization with deuterated internal standards. Increases in the levels of ent-kaurenol, ent-kaurenoic acid, and ent-7-hydroxykaurenoic acid are correlated with a decline in the amount of endogenous ent-kaurene following a 10 min incubation of microsomes with NADPH and FAD. The rate of oxidation of radiolabeled ent-kaurene by the microsomal fraction was determined, and the need to account for endogenous substrate is shown. Endogenous ent-kaurene present in soluble extracts had the effect of diluting the [¹⁴C]ent-kaurene synthesized from [¹⁴C]mevalonic acid, resulting in reduced specific radioactivity of the product. The dilution of [¹⁴C]ent-kaurene was more pronounced in extracts with higher endogenous ent-kaurene levels or when the reactions were run in the presence of O₂ and NADPH. Evidence is presented that suggests differential metabolism of endogenous ent-kaurene and radiolabeled ent-kaurene in both microsomal and soluble extracts.Abbreviations Kaurene ent-kaur-16-ene - MVA mevalonic acid - kaurenol ent-kaur-16-en-19-ol - kaurenoic acid ent-kaur-16-en-19-oic acid - EtOAc ethyl acetate - MeOH methanol - GC-MS-CI gas chromatography-mass spectrometry-chemical ionization - 13-OH KA ent-13-hydroxykaur-16-en-19-oic acid - 7-OH kaurenoic acid ent-7-hydroxykaur-16-en-19-oic acid - kaurenal ent-kaur-16-en-19-al - Me(x) methyl ester of x - TMS(x) trimethylsilyl ether or ester of x - GA(x) gibberellin A(x)     

Effects of Protein Synthesis Inhibitors on ent-Kaurene Biosynthesis during Photomorphogenesis of Etiolated Pea Seedlings

Excised shoot tips from 10-day-old etiolated pea (Pisum sativum L. cv. Alaska) seedlings were incubated in solutions of chloramphenicol, cycloheximide, and lincomycin at different concentrations during periods of 0, 4, 8, and 12 hours of irradiation with high intensity white light. Enzyme extracts were prepared from the whole shoot tips and compared with extracts from nontreated shoot tips for their capacity to synthesize ent-kaurene from mevalonate. In control samples, kaurene synthesis increased during the first 8 hours of irradiation and decreased after 12 hours. Chlorophyll content increased steadily up to 12 hours of irradiation. Chloramphenicol and cycloheximide reduced both kaurene synthesis and chlorophyll formation to a similar extent during all periods of irradiation, the reduction being greatest after 8 hours of irradiation. Lincomycin, a specific inhibitor of the formation of chloroplast ribosomes in detached pea shoot tips, did not significantly affect kaurene synthesis activity but strongly inhibited chlorophyll formation. It is tentatively concluded that the increase in kaurene synthesis activity during normal photomorphogenesis in pea seedlings is due to photoinduction of de novo synthesis of one or more proteins involved in the biosynthetic pathway from mevalonate to kaurene.     

Carotene biosynthesis by a cell extract of Aspergillus giganteus mut alba

A cell extract prepared from lyophilized mycelia of light-grown cultures of Aspergillus giganteus mut alba converted [2-¹⁴C]mevalonic acid into phytoene, lycopene, β-carotene and squalene, but from similar preparations from dark grown cultures formed only squalene. The carotenogenic activities of the cell extracts varied with the age of the cultures. Phytoene synthetase was located in the cytosolic fraction, whereas the dehydrogenation and cyclisation steps were catalysed by membrane-bound enzymes. Dithiothreitol, ATP, Mn²⁺, Mg²⁺, NAD and NADP were essential for the formation of carotenes from mevalonic acid, whilst FAD was required for phytoene metabolism. Oxygen enhanced the conversion of phytoene into other carotenes.     

Concentrations ofent-kaurene and squalene in vegetative rice shoots

Ent-kaurene (ent-kaur-16-ene) and squalene were analyzed in extracts of the shoots of three cultivars of rice (Oryza sativa L.) of 14 and 28 days of age by gas chromatography-mass spectrometry (GS-MS) and GC-selected ion monitoring (GS-SIM).Ent-kaurene occurred at approximate concentrations of <1 to 13 ng/g f.w. in 14-day-old plants and 26 to 147 ng/g f.w. in 28-day-old plants. Shoots of the dwarf cultivar Waito-C contained much lessent-kaurene than the other two cultivars at both developmental stages. The level ofent-kaurene in the dwarf cultivar Tan-ginbozu was similar to that in the normal cultivar Nihonbare at 14 days but was lower than in Nihonbare at 28 days. Trace amounts ofent-isokaurene (ent-kaur-15-ene) were also found in the extracts of all three cultivars of shoots at 28 days. Squalene occurred in approximate concentrations from as low as 19 ng/g f.w. in 28-day-old Waito-C shoots to as much as 626 ng/g f.w. in 14-day-old Nihonbare shoots. Both Tan-ginbozu and Waito-C shoots contained less squalene than Nihonbare shoots at both developmental stages.     

Separation of Light-Induced [C]ent-Kaurene Metabolism and Light-Induced Germination in Grand Rapids Lettuce Seeds

The effect of light on the metabolism of [¹⁴C]kaurene in light-requiring lettuce seeds (Lactuca sativa L. cv Grand Rapids) was investigated. Seeds were soaked in a solution of [¹⁴C]ent-kaurene in methylene chloride with 0.01% Tween-20, dried, and incubated in 20% polyethylene glycol (PEG) to prevent seedling development. Labeled metabolites were extracted and analyzed by high performance liquid chromatography and gas chromatography-radio counting. [¹⁴C]ent-Kaurenol and [¹⁴C]ent-kaurenal were identified in seeds incubated in constant white light, while no ethyl acetate-soluble metabolites were found in seeds incubated in the dark. In time course experiments using acid scarified seeds, metabolism began after 18 hours of incubation and greatly increased after 24 hours of incubation in 20% PEG. By 48 hours, several unidentified, more polar metabolites were found. Germination was induced in seeds imbibed in 20% PEG by 4 hours of red or 4 hours of white light following 20 hours in the dark, and was fully reversed by 2 hours of far red light. However, in metabolism experiments, [¹⁴C]ent-kaurene oxidation was observed only with constant white light. These results indicate that although ent-kaurene oxidation is a light sensitive step in the biosynthesis of gibberellins in Grand Rapids lettuce seeds, ent-kaurene metabolism is not required for light-induced germination.     

Synthesis of dehydrosqualene in microsomal fraction of Saccharomyces cerevisiae.

When the microsomal fraction of Saccharomyces cerevisiae was incubated with farnesyl pyrophosphate or presqualene pyrophosphate in the presence of Mn²⁺, dehydrosqualene was formed. Incubation of the reaction mixture in the presence of NADPH gave squalene, not dehydrosqualene, as the product. Little dehydrosqualene was formed when Mn²⁺ was replaced with Mg²⁺. These observations suggest that dehydrosqualene formation is closely associated with squalene synthesis in yeast, which synthesizes neither carotenes nor related pigments.     

Comparison ofEnt-kaurene synthetase A and B activities in cell-free extracts from young tomato fruits of wild-type andgib-1, gib-2, andgib-3 tomato plants

The nonallelicgib-1 andgib-3 tomato (Lycopersion esculentum Mill.) mutants are gibberellin deficient and exhibit a dwarfed growth habit. Previous work has shown that this dwarfed growth pattern can be reversed by the application of a number of gibberellins and their precursors, includingent-kaurene (ent-kaur-16-ene). This indicates that they are blocked in gibberellin biosynthesis at a step prior toent-kaurene metabolism. The normal accumulation of carotenoids observed in these mutants suggests a functionally normal isoprenoid pathway.Ent-kaurene is synthesized from geranylgeranyl pyrophosphate in a two-step process with copalyl pyrophosphate as an intermediate.In vitro assays using young fruit extracts from wild-type andgib-2 plants resulted in the conversion of geranylgeranyl pyrophosphate to copalyl pyrophosphate, and the conversion of copalyl pyrophosphate toentkaurene. Similar assays usinggib-1 plants indicated a reduced ability for synthesis of copalyl pyrophosphate from geranylgeranyl pyrophosphate, and thus a reducedent-kaurene synthetase A activity. Furthermore,gib-3 extracts demonstrated a reduced ability to synthesizeent-kaurene from copalyl pyrophosphate, and thus a reducedent-kaurene synthetase B activity. These results establish the enzymatic conversion of geranylgeranyl pyrophosphate to copalyl pyrophosphate, and copalyl pyrophosphate toent-kaurene, as the sites of the mutations ingib-1 andgib-3 tomatoes, respectively. We also note that tomato fruit extracts contain components which are inhibitory toent-kaurene synthesis.     

Membrane-associated Glycosyl Transferases in Cotyledons of Pisum sativum: Differential Effects of Magnesium and Manganese Ions

In crude particulate fractions isolated from pea (Pisum sativum) cotyledons, the transfer of radioactivity from GDP-[¹⁴C]mannose to glycolipid appears to be preferentially stimulated by Mn²⁺ while the transfer to lipid-free residue is enhanced by Mg²⁺. In contrast, the transfer of radioactivity from UDP-N-acetyl-[¹⁴C]glucosamine to glycolipid shows preferential stimulation by Mg²⁺ while the transfer to lipid-free residue prefers Mn²⁺. These results are accounted for by the differential stimulation by Mg²⁺ and Mn²⁺ of glycosyl transferases associated with subcellular membranes which were separated by isopycnic sucrose density centrifugation.