Tri Phosphate-Dependent Doliehol Kinase and Dolichol Phosphatase

[CANCER RESEARCH 48. 3418-3424, June 15, 1988] Cytidine 5'-Tri phosphate-dependent Doliehol Kinase and Dolichol Phosphatase Activities and Levels of Dolichyl Phosphate in Microsomal Fractions from Highly Differentiated Human Hepatomas1

Ivan Eggens, Johan Ericsson, and Ã–rjanTollbom Department of Pathology at Huddinge Hospital, Karolinska Institute!, S-I4I 86 Huddinge, Â¡I.E.], and Department of Biochemistry, Arrhenius Laboratory, University of Stockholm, S-106 91 Stockholm, Sweden [J. E., Ã–.T.]

ABSTRACT synthesis is the free alcohol and that this compound is subse quently phosphorylated to give dolichyl phasphate (15). The Homogenates and microsomal fractions prepared from biopsies of possibility was also raised that the level of dolichyl-P is regu highly differentiated human hepatocellular carcinomas were found to lated by dolichol kinase and dolichol phosphatase activities contain low levels of dolichol in comparison with control tissue. In contrast, the amount of dolichyl phosphate in tumor homogenates was (15). The existence of dolichol mono- and pyrophosphatases have unchanged and actually increased in the microsomal fraction. The pattern of individual polyisoprenoids, both in the free and the phosphorylated been reported by several authors and their activities are found dolichol fractions of hepatomas, did not exhibit any major alterations in several intracellular organelles, although at different levels compared to the control. (16-18). It has also been reported earlier that dolichol can be The rates of incorporation of ['H|mevalonic acid into dolichol and phosphorylated in vitro via a CTP-mediated kinase, which is dolichyl phosphate in hepatomas were low. The dolichol monophospha- situated on the outer surface of microsomal membranes (19, tase activities in microsomal fractions from hepatomas and controls did 20). However, no investigation of the substrate specificities of not show any major differences, whereas the activity of the CTP-depend- the kinase or phosphatase for polyisoprenoid compounds has ent dolichol kinase was increased in tumor microsomes. Glycosylation of endogenous dolichyl phosphate and of total protein using certain nucleo- been performed. tide-activated sugars was found to be slightly elevated in microsomal Recent reports using rat liver tissue found, however, no fractions from the tumor itself when compared to the control. The reasons evidence that dolichol serves as a precursor for dolichyl-P. It for the differences in the levels of polyisoprenoids in hepatomas and was also suggested that the immediate precursors of dolichol control tissue are discussed. and dolichyl-P are the a-unsaturated polyprenol and the cor responding polyprenyl phosphate, respectively (21, 22). An other study using primary cultures of rat hepatocytes indicated INTRODUCTION that dolichyl-P is synthesized primarily through a de novo Several cell lines derived from human hepatocellular carci process and that only about 10% of newly synthesized dolichyl- nomas demonstrate biosynthetic capabilities similar to those of P is produced by phosphorylation of the free alcohol (23). normal liver parenchyma! cells and many of these cell lines are In the present study we have measured and compared the capable of synthesizing and secreting most major plasma pro biosynthesis of dolichyl-P and dolichol as well as certain related teins (1). Nearly all plasma proteins are glycoproteins, the enzyme activities in microsomal fractions from highly differ majority of which are formed by ,/V-linked glycosylation, and entiated hepatomas and from control hepatic tissue. The rea investigations of certain hepatoma lines have revealed defects sons for the differences in dolichyl-P and dolichol levels ob in their glycosylation of different glycoproteins (2, 3). Phospho served are discussed. rylated dolichols are involved in glycoprotein synthesis and it is therefore of interest to study these compounds in hepatomas. MATERIALS AND METHODS These polyisoprenoid compounds are present in all tissues and intracellular membranes, either as a free alcohol, esterified with Isolation and Characterization of the Material. Our material was a fatty acid, and/or in phosphorylated form (4-7). derived from 5 highly differentiated human hepatocellular carcinomas (hepatomas), from the morphologically normal region outside the hep The function of the free alcohol in biological membranes has atoma tumor and the tumor itself, and from 5 normal livers. The not yet been established, but in model membranes dolichol is patients involved were 45-55 years of age. The tissue samples were proposed to influence membrane properties such as fluidity and utilized immediately after surgery and part of the material was then permeability (8-10). Phosphorylated dolichol participates in rapidly frozen to â€”70Â°C.Inselecting samples from the tumor, care was the biosynthesis of W-glycosidically linked oligosaccharide taken to exclude fibrotic or hemorrhagic areas. chains and it was suggested earlier that the level of this com All tumor tissues examined were also macroscopically free of necro pound may be rate limiting for protein glycosylation under sis. All samples were subjected to careful histolÃ³gica!examination and certain conditions (11-14). those samples which after later examination were found to contain The exact pathways and mechanisms for the biosynthesis and areas of necrosis or fibrosis were not included in this study. Samples were homogenized with an Ultra-Turrax blender prior to regulation of the levels of polyisoprenoids are still not com pletely understood. When dolichol and dolichyl-P2 were studied lipid extraction. Homogenates and microsomal fractions were prepared as described earlier (24, 25). The values presented in the tables are the in sea urchin embryos using radioactively labeled precursors, means of 2 different experiments each involving 5 different liver sam the results indicated that the principal end product of de novo ples handled independently. Each sample consisted of 3 or 4 pieces of tissue picked randomly from the area of interest. Received 9/2/87; revised 2/18/88; accepted 2/26/88. The costs of publication of this article were defrayed in part by the payment Isolation and Quantitation of Dolichyl-P and Dolichol. The tissue was of page charges. This article must therefore be hereby marked advertisement in homogenized, dolichol-23 and dolichyl-23 phosphate were added as accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1This work was supported by Centrala FÃ¶rsÃ¶ksdjursnÃ¤mnden,theSwedish internal standards, and lipids were extracted with CMW (1:1:0.3). HC1 was then added to the mixture to give a final concentration of 0.1 M Cancer Society, and the Swedish Medical Research Council. 2The abbreviations used are: dolichyl-P, dolichyl phosphate; CMW, chloro- and acid hydrolysis was performed, tirsi for 60 min at room tempera form:melhanol:water; HPLC, high-performance liquid chromalography. ture, then for 45 min at 55Â°Cand finally, for 10 min at 100'C. The 3418

suspension was then neutralized with NaOH and evaporated to dryness. added and the mixture shaken in a vortex and incubated at 37*C for 15 Alkaline hydrolysis was performed with methanol:water:KOH (60%) min. The reaction was stopped by adding 7 ml chloroform:methanol (1:1:0.5) for 45 min at 90Â°C.The pH was subsequently adjusted to 7.0 (2:1), 1 ml H2O, and 0.5 ml 0.1 M MgCl2. After shaking, this mixture and chloroform was added to obtain a CMW ratio of 3:2:1. After was incubated at 40"C for 20 min. After centrifugation the upper phase washing the lower phase with theoretical upper phase chloro- was discarded and the lower phase washed twice with theoretical upper form:methanol:water (3:48:47) the sample was evaporated to dryness. phase. This washed extract, which contained the polyprenols, was The residue was then dissolved in 200 ^1 CM (2:1) and 10 ml MW evaporated and placed on a silica gel thin-layer chromatography plate. (98:2) containing 20 IHM11.!'( )4 were added. This mixture was placed The plate was developed using hexane:ether:acetic acid (75:30:1.5) and onto a CIS Sep-Pak column which had been equilibrated with 20 DIM the spot containing dolichol (identified using standards) was stained H3PO4 and thereafter washed first with 10 ml MW (98:2) containing with iodine. This spot was scraped off the plate and transferred to 20 mM H3PO4 followed by 10 ml MW (98:2). scintillation vials. One ml methanol was added to the sample, which Dolichol and dolichyl-P were eluted from the CIS Sep-Pak column was then heated to 60Â°Cfor 30 min and finally, supplemented with 10 with CM (2:1). This fraction was then supplemented with ammonia to ml Aqualuma Plus, whereafter the radioactivity was determined. give a final concentration of 1% and placed onto a silica Sep-Pak Glycosylation of Endogenous Dolichyl-P and Protein. Microsomes (7 column. Dolichol was eluted with CM (2:1) containing 1% ammonia mg protein) were incubated in a total volume of 1 ml. To measure and dolichyl-P with CM (1:1:0.3). The lipid solutions were evaporated glycosylation with GDP-mannose, the medium contained 30 mM Tris- to dryness and the residues dissolved in CM (2:1) (dolichol) or CM HCl, pH 7.8, 1 mM MnCl2, 12.5 mM j3-mercaptoethanol, 0.4% Triton (2:1) containing 40 mM HC1 (dolichyl-P). The samples were then X-100, 2 mM AMP, and 0.4 ^Ci GDP-[MC]mannose (80 mCi/mmol; analyzed by HPLC using a Hewlett-Packard Hypersil ODS 3 pm Radiochemical Centre, Amersham, England). In the case of glycosyla reversed phase column. A convex gradient was used (No. 5, Waters 660 tion with jV-acetylglucosamine, the medium contained 30 mM Tris- solvent program) from the initial 2-propanol-methanol-H2O (40:60:5) HCl, pH 7.8, 2.5 mM EDTA, 12.5 mM jS-mercaptoethanol, 10 mM in pump system A to 40% hexane-2 propanol (70:30) in pump system MnCl2, 1.5 mM ATP, and 0.8 MCiUDP-fCyV-acetylglucosamine (300 B at a flow rate of 1.0 ml/min with a program time of 45 min. For mCi/mmol; Radiochemical Centre). When nucleotide-activated glucose measurement of dolichyl-P the solvent mixtures in pump systems A was used as substrate, the incubation mixture contained 30 mM Tris- and B also contained 20 mM H3PO4. The eluate was monitored at 210 HCl, pH 7.8, 12.5 mM /3-mercaptoethanol, 10 mM MnCl2, 1.5 mM ATP, and 0.8 Â¿iCiUDP-['"C]glucose (200 mCi/mmol; Radiochemical nm. For measurement of the dolichol amount only, in order to ensure as Centre). The microsomes alone were preincubated for 5 min after which the complete system was incubated at 30"C for 15 min. complete recovery as possible, the homogenate was initially subjected to alkaline hydrolysis as described above, whereafter the lipid mixture At the end of this period, dolichyl monophosphate sugars were was purified on a CIS Sep-Pak column prior to HPLC analysis. isolated by extraction with CM (2:1) and dolichyl pyrophosphate- Assays of Dolichol and Dolichyl-P Synthesis. A total of 1 ml homog oligosaccharides by extraction with CMW (1:1:0.3) as described under enate (35 mg protein suspended in 0.25 M sucrose) was incubated in a the kinase assay above. The protein pellet was washed with 1 ml Ilo final medium volume of 5 ml with constant shaking. The medium and solubilized with 2% sodium dodecyl sulfate. Radioactivity was contained 100 mM KH2PO4, 5 mM MgCl2, 100 units pyruvatekinase measured as described below. The CM and CMW extracts were also (Sigma), 10 mM phosphoenolpyruvate, 5 mM NADH, and 5 mM ATP analyzed by HPLC and thin layer procedures (26). The major portion and had a pH of 7.5. This medium was incubated with 125 Â¿Â¿Ci|'II|- of the radioactivity was associated with dolichyl-P-sugar and dolichyl mevalonolactone for 40 min at 30Â°C.Dolichol-23 and dolichyl-23-P pyrophosphate-oligosaccharides, respectively (not shown). were added prior to alkaline hydrolysis. In this experimental system Determinations of Radioactivity and Enzymatic and Chemical Meas urements. Mixtures of dolichol and dolichyl-P were dissolved in 10 ml acid hydrolysis was not performed after the incubation, since such treatment destroys a-unsaturated polyprenyl phosphates. Lipids were Aqualuma Plus. Proteins were dissolved first in 1 ml 2% sodium dodecyl thereafter extracted and isolated by HPLC and radioactivity was deter sulfate and this solution then supplemented with 10 ml Aqualuma Plus. mined. Radioactivity was determined by scintillation counting. Marker enzyme Assay of CTP-dependent Dolichol hÃÃ±aseActivity.In order to meas activities were determined using procedures described previously (27, ure kinase activity ['H]dolichol-15 (2 x IO6 dpm; specific activity, 8.7 28). The lipid values in the tables are expressed on a wet weight and/ Ci/mmol) and 40 Â»igunlabeleddolichol-15 were mixed with 50 //I 1% or protein basis. Protein was determined using the biuret procedure Triton X-100, after which this mixture was evaporated to dryness. The (29). residue was shaken with a mixture containing 190 IHMTris-HCl, pH 7.5, 5 mM /3-mercaptoethanol, 5 mM CaCl2, 25 mM NaF, and 40 MM ("TP in a total volume of 1 ml. The incubation was started by the RESULTS addition of microsomes (7 mg protein) which had been preincubated for 5 min at 37'C and was carried out at 37Â°Cfor 15 min. At the end Tissue Characterization. The material in this report consisted solely of surgical biopsies from highly differentiated human of this period, 7 ml chloroform: meth:mol (2:1) and 0.5 ml 0.1 M MgCl2 were added, after which the mixture was incubated further at 40"C for hepatocellular carcinomas and from normal hepatic tissue. The 20 min. The clear lower phase was removed and the remaining protein controls consisted of morphologically normal tissue surround pellet was extracted twice again in the same manner with 7 ml CM ing the hepatoma tumor and a separate group of normal liver (2:1) and 0.5 ml 0.1 M MgCl2. After pooling the lower phases and biopsies from healthy individuals. Effort was made to exclude washing with theoretical upper phase, the extract was evaporated and areas of necrosis and fibrosis. In all cases histolÃ³gica! exami placed on a silica gel thin-layer chromatography plate. The plate was nation was performed. developed using CMW (6:25:4) and the spot containing dolichyl-P In the case of isolated microsomal fractions, electron micros (identified using standards) was stained with iodine. This spot was copy revealed vesicles corresponding mainly to rough and scraped off the plate and transferred to scintillation vials. One ml methanol was added to this sample, which was then heated to 60Â°Cfor smooth microsomes and no clear differences between the con trols and hepatomas could be seen (not shown). Since mito 30 min and subsequently supplemented with 10 ml Aqualuma Plus and chondria are reported to be decreased in number and to show the radioactivity was determined. Assay of Dolichol Monophosphatase Activity. When dolichol mon changes in structure in certain hepatomas (30), the possibility ophosphatase activity was measured, [3H]-dolichyl-19-phosphate that whole or fragmented mitochondria might contaminate the (100,000 dpm) in chloroform:methanol (2:1) and 50 ^1 0.1% Triton X- microsomal fractions from such tumors was considered. Enzy 100 were mixed and dried. One mM EDTA, 10 mM dithiothreitol, and matic studies of microsomal fractions (Table I) revealed no 63 HIMTris-HCl with a pH of 6.5 in a final volume of 50 Â¿ilwereadded major differences in the hepatoma and control groups. In to the residue. Microsomes (0.35 mg protein) in a volume of 50 fil were microsomes from hepatomas only minor decreases in NADPH- 3419

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 1988 American Association for Cancer Research. DOLICHOL KINASE AND PHOSPHATASE IN HUMAN HEPATOMAS cytochrome c reducÃase and cytochrome c oxidase activities lowered. In Ihe case of dolichyl-P ihe levels in hepaloma ho compared to controls were observed. In the mitochondrial mogenales were basically the same as Ihose in non-hepaloma fraction the cytochrome c oxidase activity was more than 100- conlrols. fold higher and the NADPH-cytochrome c reducÃaseactivily The isoprenoid pallern of individual free dolichols in conlrol was 20 limes lower in comparison wilh ihe microsomal fraclion liver homogenales was Ihe same as reported previously, i.e., Ihe in all groups (noi shown in Table 1). In some experiment Ihe major componenls were dolichol-19 and -20 (24, 31). In our lysosomal and plasma membrane conlenls of Ihe microsomal group of highly differenlialed hepalomas no changes in ihis fractions were analyzed and no major differences belween ihe distribution of individual free polyisoprenoids could be de- hepaloma and conlrol groups could be seen. lecled. The dislribution of individual dolichyl phosphates in From ihe dala above il was concluded lhal Ihe microsomal controls was the same as lhal of Ihe free alcohol, which is also fractions from Ihe 3 differenl groups (hepatoma and 2 controls) in agreemenl wilh earlier preliminary reports (31). The dolichyl- contained approximalely Ihe same amounl of microsomes and P pallern was unallered in hepatoma lissue. showed no differences in milochondrial or olher organelle The level of dolichol in ihe microsomal fraclion from control conlamination. These findings agree well wilh earlier observa- liver was aboul 1.4 fig/mg prolein (Table 3) and approximalely lions lhal highly differenlialed hepalomas conlain a well-devel the same in morphologically normal regions oulside Ihe hepa oped endoplasmic reliculum resembling lhal found in normal loma tumors, whereas Ihe dolichol level in microsomal fractions hepatocytes (30). from ihe lumor ilself was more lhan 30% lower. In case of Ihe Dolichol and Dolichyl-P in Control Tissue and Hepalomas. microsomal dolichyl-P level, ihe non-lumor regions in hepalo The lolal level of dolichol (Table 2) in homogenales of conlrol mas and ihe normal livers exhibiled similar values, whereas Ihe corresponding value for Ihe lumors was Iwice as greal. The liver is aboul 500 //g/g lissue, which is in agreemenl wilh earlier patterns of individual dolichol and dolichyl-P in microsomes reports (24, 31). The concenlralion of dolichol in homogenales from normal livers, morphologically normal lissue oulside Ihe from morphologically normal liver lissue oulside Ihe tumors hepaloma lumors and from ihe lumors ihemselves were basi was slighlly lower lhan in lissue from normal livers, whereas cally Ihe same, allhough minor decreases in ihe relalive levels Ihe level of dolichol in the hepatoma tumors was draslically of dolichol-21 and dolichyl-21-P were noliced in Ihe hepaloma lumors. The pallern of individual dolichols in conlrol micro Table I Enzymatic activities of microsomal fractions prepared from human control liver, from morphologically normal liver tissue surrounding the hepatoma somes was differenl from lhal of conlrol homogenales, wilh tumor, and from the tumor itself slighlly higher relalive percenlages in dolichol-21 and -22 in All tissues were diagnosed histologically prior to subfractionation and analysis. Ihe microsomes (Tables 2 and 3). On ihe olher hand, Ihe Values are the means Â±SE of 10 experiments. No significant differences were obtained with the Student's t test. Experimental details are presented under distribution of individual dolichyl-P was similar in conlrol "Materials and Methods." microsomes and homogenales. Activity in microsomal fractions Incorporation of |'H|Mevalonic Acid into Dolichol and Doli Morphologically chyl Phosphate in Liver Slices. In ihis particular series of normal liver experimenls only alkaline hydrolysis was ulilized (i.e., no acid tissue outside hydrolysis) in order lo ensure lhal bolh polyprenyl and dolichyl- Control the hepatoma Hepatoma Enzyme liver tumor tumor P could be measured reliably. Il was possible lo label polypren- NADPH-cytochrome ere- 66 Â±3.0 69 Â±3.8 63 Â±2.8 ols in conlrol livers and ihe values for dolichol synthesis in ductase (nmol cyto Table 4 consequently represenl ihe sums of labeled a-unsalu- chrome c reduced/min/ raled and saturaled polyprenols in each group. The incorpora- mg protein) Cytochrome c oxidase 14Â±1.6 15Â±1.9 12Â±1.4 lion of radioaclivily inlo dolichyl-P in conlrols was found lo be (nmol cytochrome c oxi- slighlly higher lhan lhal inlo Ihe dolichol (calculated on a gram dized/min/mg protein) lissue basis), in agreement with earlier investigations on rat

Table 2 Levels of dolichol and dolichyl-P and the distributions of individual polyisoprenoids in homogenates from control liver, from morphologically normal liver lissue outside Ihe hepatoma lumor, and from the tumor ilself All tissues were diagnosed histologically before analysis. Values are the means Â±SE of 10 experiments. Levels of significance were determined by Student's / test compared with control values. Experimental details are presented under "Materials and Methods."

Total amounts of Polyprenol composition (% of total) dolichol and dolichyl-P in homogenate Sample wet wt) D17" D18 D19 D20 D21 D22 Control liver Dolichol 490 Â±15.8 5 Â±1.2 13 Â±2.3 39 Â±3.1 32 Â±3.4 9 Â±1.5 2 Â±0.3 Dolichyl-P 14.8 Â±1.2 2 Â±0.3 9 Â±1.5 40 Â±3.4 33 Â±3.5 13 Â±2.0 3 Â±0.5

Morphologically normal liver tissue outside the hepatoma tumor Dolichol 455 Â±33.2 3 Â±0.4 14 Â±2.3 39 Â±3.2 33 Â±2.8 9+1.6 2Â±0.3 Dolichyl-P 14.1 Â±1.1 2 Â±0.3 8 Â±1.0 41 Â±4.2 38 Â±2.9 8 Â±1.5 3 Â±0.4

HepatomatumorDolicholDolichyl-P61 Â±5.9*13.9 Â±0.2C2 +1.56Â± Â±3.541 Â±2.637 1.411 Â±0.1''3 Â±1.02 Â±0.311 1.043 Â±3.436 + 2.57Â± Â±1.61 Â±0.4 " Number of isoprenes. " P< 0.001.

3420

Table 3 Levels ofdoUchol and dolichyl phosphate and the distribution of individual polyisoprenoids in microsomal fractions prepared from control liver, from morphologically normal liver tissue outside the hepatoma tumor, and from the tumor itself The samples were diagnosed histologically before analysis by high-performance liquid chromatography. Values are the means Â±SE of 10 experiments. Levels of significance were determined by Student's t test compared with control values. Experimental details are presented under "Materials and Methods."

amounts of total)DIT1 composition (% of dolichol and dolichyl-P in the microsomal fraction SampleControl (ng/mgprotein)1.42 liver Dolichol Â±0.1 Â±0.2 1.0 Â±3.2 Â±3.1 Â±2.0 Â±0.94 Dolichyl-PTotal 0.11 Â±0.02Polyprenol 1 Â±0.1D187Â± 8Â± 1.1D1933 34 Â±3.5D2038 38 Â±2.9D2116 15 Â±1.9D225 Â±0.7

Morphologically normal liver tissue outside the hepatoma tumor Dolichol Â±0.2 Â±0.2 1.2 Â±3.1 Â±3.037 Â±1.8 Â±0.8 Dolichyl-PHepatoma 0.13Â±0.010.95 1Â±0.12 7Â±0.97Â±0.8 39 Â±3.039 Â±2.937 13 Â±1.811 3Â±0.44

tumor Dolichol Â±0.1* Â±0.3e Â±3.2 Â±3.0 Â±1.410 Â±0.4 Dolichyl-P1.38 0.22 Â±0.03''1 1 Â±0.28Â± 7 Â±0.934 38 Â±2.836 41 Â±3.215 Â±1.2e5 3 Â±0.3 " Number of isoprenes. "P<0.01. CP<0.05. " P< 0.001.

Table 4 incorporation off'HJmevalonic acid into dolichol and dolichyl Table 5 Dolichol kinase activity in microsomal fractions prepared from control phosphate in slices from control liver and from the hepatoma tumor liver, from morphologically normal liver tissue outside the hepatoma tumor, and All tissues were diagnosed histologically before use. Values are the means Â± from the tumor itself SE of 10 experiments. Levels of significance were determined by Student's t test All tissues were diagnosed histologically before analysis. Values are the means compared with control values. Experimental details are presented under "Mate Â±SE of 10 experiments. The level of significance was determined by Student's / rials and Methods." test compared with the control value. Experimental details are presented under "Materials and Methods." Dolichyl Dolichol synthesis phosphate synthesis Dolichol kinase activity in cpm/g cpm/Mg cpm/g cpm//ig microsomal fractions Sample homogenate dolichol homogenate dolichyl-P Sample (cpm/mg protein/min) Control liver Â±165 Â±0.2 Â±185 Â±26 95 Â±18"2.6 1.5 Â±0.1"178065 Â±15.4Â°3074.7 Â±0.4Â° Control liver 4986 Â±385 Hepatoma tumor1573 Morphologically normal liver tissue 4972 Â±358 Â°P<0.001. outside the hepatoma tumor Hepatoma tumor 7218 Â±425Â° hepatocytes (21). The specific radioactivity of dolichyl-P frac tions from controls was more than 100-fold higher than that of Table 6 Dolichol monophosphatase activities in microsomal fractions prepared the dolichol fraction. from control liver, from morphologically normal liver tissue outside the hepatoma In the case of hepatoma tumors, however, clear decreases in tumor, and from the tumor itself labeling of the dolichol and dolichyl-P fractions (on a gram All tissues were diagnosed histologically before analysis. Values are the means Â±SE of 10 experiments. No significant differences were obtained with Student's tissue basis) to about 6 and 3.5%, respectively, of the control t test. Experimental details are presented under "Materials and Methods." value were observed. The specific radioactivity in the dolichyl- Dolichol monophosphatase P fraction from hepatomas was decreased in comparison to activities in microsomes controls but still remained higher than that of the dolichol (dolichol formed, fraction. When incorporation of mevalonic acid into polyiso- Sample cpm/mg protein/10 min) prenoid compounds in microsomal fractions and homogenates Control liver 3943 Â±285 Morphologically normal liver tissue 4086 Â±315 were compared, the results were basically similar (not shown). outside the hepatoma tumor Dolichol Kinase Activity. The dolichol kinase activity of the Hepatoma tumor 4282 Â±386 different microsomal fractions was tested using dolichol-IS as substrate. This activity in microsomal fractions from non-tumor hepatomas, although a minor increase in the hepatomas was regions in hepatoma livers did not differ from that of the other noticed in the dolichyl-P-glucose fraction. In the case of incor control non-hepatoma livers, while tumor microsomes demon poration of sugar into total protein the only difference noticed strated about a 30% higher kinase activity (Table 5). Dolichol was an increase of about 15-20% in preparations from the kinase activity using dolichol-11 as a substrate was the same as hepatoma tumors when labeling was performed with activated that with dolichol-15 in all cases (not shown). mannose or glucose. Dolichol Monophosphate Activity. Microsomal dolichol mon ophosphatase activity was tested with monophosphorylated dolichol-19 as substrate (Table 6). No major differences could DISCUSSION be observed between controls and hepatomas. Glycosylation of Endogenous Dolichyl Monophosphate and Considerable effort has been focused on defining changes in Protein. Three different activated sugars, i.e.. glucosamine, the surface membranes of hepatoma cells and the relationship marinÃ³se,and glucose, were chosen to investigate glycosylation of such changes to the process of glycoprotein synthesis (32). of endogenous dolichyl-P and total protein in controls and An important compound involved in glycoprotein synthesis is hepatoma microsomes (Table 7). Incorporation of sugar into dolichyl-P, which is an obligatory intermediate in the synthesis the dolichyl-P fraction was basically similar in the controls and of A'-linked oligosaccharide chains (4-7). Recent investigations 3421

Table 7 Glycosylalion of endogenous dolichyi phosphate and of protein in microsomal fractions from control liver, from morphologically normal liver tissue outside the hepatoma tumor, and from the tumor itself All tissues were diagnosed hislologically before analysis. Values are the means Â±SE of 10 experiments. Levels of significance were determined by Student's t test compared with control values. Experimental details are presented under "Materials and Methods." Dolichyl-P monosaccharide Total protein

SampleControl protein)3055 liverMorphologically Â±265 Â±18 UDP-A/-acetylglucosamine 1009 Â±85 186 Â±20 UDP-glucoseGDP-mannose 925 Â±873125 208 Â±21255

normal liver tissue outside Â±255 Â±24 tumorHepatomathe hepatoma UDP-A/-acetylglucosamine 1011 Â±72 195 Â±18 UDP-glucoseGDP-mannose 1025Â±653145 215Â±20289

tumorSubstrateGDP-mannose Â±269 Â±21Â° UDP-iV-acetylglucosamine 1060Â± 82 190 Â±17238 UDP-glucose(cpm/mg 1257 Â±91Â°225 Â±22 have revealed changes in the levels of dolichol and dolichyl-P previous studies), but also the fact that we used only highly in hyper piustit- nodules in rats (14) and in human hepatomas differentiated hepatomas, whereas the former material con (31, 33). In the studies on human hepatomas, changes in the sisted of hepatomas of varying degrees of differentiation. pattern of individual dolichols were also found (31). It has been established that different animal species have The investigations on human hepatomas have used mainly different polyisoprenoid patterns, but the functional signifi autopsy samples consisting of a mixture of highly and poorly cance of these patterns is still unknown (7). In a recent inves differentiated tumors. In the present investigation our aim was tigation it was found that the chain length of the polyisoprenes to obtain information about the composition and biosynthesis does not influence sugar transfer during completion of the of dolichol and dolichyl-P in fresh surgical biopsies from highly oligosaccharide chain (35). To date, only a few examples of differentiated human hepatocellular carcinomas. Earlier mor changes in the pattern of individual polyisoprenoids have been phological studies on certain hepatomas, usually less differen reported. Livers of rats exposed to phthalate esters exhibited tiated, have revealed differences in the amount and structure of an increase in the relative levels of longer dolichols (36), their mitochondria and endoplasma- reticulum, but the cells of whereas a group of human hepatomas of varying degrees of highly differentiated hepatomas are basically similar in appear differentiation showed a relative enrichment in shorter doli ance to normal hepatocytes (30). chols (31). The cells from the highly differentiated hepatomas studied In the present study of highly differentiated hepatomas the here were morphologically similar to control hepatocytes, in patterns of individual polyisoprenoids in both the free and line with the earlier electron microscopic study (30). The simi phosphorylated dolichol fractions in homogenates were similar larities between our control livers and hepatomas were also to those of the control livers. confirmed by analyzing the microsomal fractions by electron In the case of analyzing polyisoprenoid levels, the ratio of microscopy and on the basis of marker enzyme activities. Con dolichol to dolichyl-P in control microsomes (about 13) differs sequently, it seemed adequate to isolate and compare the mi as expected from that in control homogenates (about 33), since crosomal fractions from control tissue and highly differentiated dolichyl-P plays a major role in microsomal glycoprotein syn hepatomas. The control groups in this report consisted not only of normal thesis. The dolichol concentration in microsomes from normal liver tissue from healthy individuals but also of morphologically liver biopsies collected from healthy individuals was similar to normal tissue outside the hepatoma tumor. The selection of that in microsomes isolated from the regions outside the hep such non-tumor regions from hepatoma-bearing individuals as atoma tumor, while tumor microsomes showed a decrease in an additional control allowed us to compare the tumor itself dolichol levels by more than 30%. The low dolichol level in and a morphologically normal area of the same liver, i.e., under tumor microsomes could be explained by a lowered de novo similar influence of hormonal, dietary, and other factors. synthesis, since we observed a low level of incorporation of In our measurements of the polyisoprenoid synthesis we mevalonate into dolichols in the hepatomas. The low dolichol utilized liver slices. The possibility that this system is not an level in tumors could theoretically also be explained by an adequate measure of in vivo synthesis was considered. However, increased degradation, but a degradative process has not yet when comparing the [-'Hjmevalonate incorporation into doli been established for dolichol. chol by different tissues in vivo and by slices the results in a In contrast to the free alcohol, phosphorylated dolichol was recent investigation showed a good agreement between these present at an unaltered level in tumor homogenates and at an techniques (34). increased level in microsomes from hepatomas. The low rate Previous investigations on polyisoprenoid levels in human of dolichyl-P synthesis was thus surprising but indicated that hepatomas revealed a low level of dolichol (31) and also a some mechanism(s) other than alterations in de novo synthesis decreased amount of dolichyl-P (33). The dolichol levels in the may determine the level of dolichyi phosphate. Experimental tumors studied here were also drastically decreased in compar findings here support such an idea since the increase in the ison to the controls, while the levels of dolichyi phosphate did dolichyi phosphate level of tumor microsomes was accom not show any major differences. The explanation for these panied by an increase in the dolichol kinase activity. The different results with respect to dolichyl-P may not be only relatively unaltered dolichol phosphatase activity in hepatoma biological variability and/or the origin of the samples (tissue microsomes was also consistent with this idea. The regions collected during surgery in this report, autopsy cases in the outside the hepatoma tumors and the different normal liver 3422

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 1988 American Association for Cancer Research. DOLICHOL KJNASE AND PHOSPHATASE IN HUMAN HEPATOMAS samples showed both similar dolichyl-P levels and enzyme REFERENCES activities. It thus appears probable that in the hepatomas de novo Knowles, B. B., Howe, C. C., and Aden, D. P. Human hepatocellular synthesis is less important in regulating the dolichyl-P level carcinoma cell lines secrete the major plasma proteins and hepatitis B surface antigen. Science (Wash. DC), 209: 497-499, 1980. than are the dolichol kinase and dolichol phosphatase activities. Carlson, J., Eriksson, S., Aim, R., and KjellstrÃ¶m,T. Biosynthesis of abnor However, other explanations for the observed dolichyl-P levels mally glycosylated n ,â€¢Â¡iniiirypsinbya human hepatoma cell line. Hepatology, 4:235-241, 1984. in tumors are theoretically possible, including differences in the Aim, R., and Eriksson, S. Biosynthesis of abnormally glycosylated hepatoma degradation of dolichyl-P, in the dietary uptake of polyiso- secretory proteins in cell cultures. FEBS Lett., 190: 157-160, 1985. l'anuli. A. J., and Leloir, L. F. The role of lipid intermediates in the prenes, and in distribution of dolichyl-P in the body. The low glycosylation of proteins in the eucaryotic cell. Biochem. Biophys. Acta, 559: rate of dolichyl-P synthesis may thus be coupled with a slow 1-37, 1979. Struck, D. K., and Lennarz, W. J. In: W. J. Lennarz (ed.), The Biochemistry degradation, but the largely unchanged phosphatase activity in of Glycoproteins And Proteoglycans, pp. 35-83. New York: Plenum Press, microsomes does not support this idea. 1980. Under pathological conditions an increased uptake of polyi- 6. Hemming, F. W. In: J. W. Porter and S. L. Spurgeon (eds.), Biosynthesis of soprenoids from the diet could influence the dolichyl-P level. Isoprenoid Compounds, Vol. 2, pp. 305-354. New York: John Wiley and Sons, 1983. However, this explanation seems unlikely since the uptake of 7. Dallner, G., and Hemming, F. W. Lipid carriers in microsomal membrane. In: C. P. Lee, G. Schatz, and G. Dallner (eds.), Mitochondria and Micro polyisoprenes from the diet is normally very limited and occurs somes, pp. 655-681. Reading, MA: Addison-Wesley Publishing Co., 1981. exclusively with short dolichols, which under normal condi 8. Lai, C.-S., and Schutzbach, J. S. Localization of dolichols in phospholipid tions, may be phosphorylated to some extent. It was recently membranes. FEBS Lett., 203: 153-156, 1986. 9. Vigo, C., Grossman, S. H., and Drost-Hansen, W. Interaction of dolichol also reported that these dietary polyisoprenes do not seem to and dolichyl phosphate with phospholipid bilayers. Biochim. Biophys. Acta, be condensed to longer dolichols (37). 774:221-226, 1984. Since polyisoprenoid compounds are transported out of hep- 10. Valtersson, C., van Duijn, G., Verkleij, A. J., Chojnacki, T., de Kruijff, B., and Dallner, G. The influence of dolichol, dolichol esters and dolichyl atocytes to the blood and bile (34, 38), it would be theoretically phosphate on phospholipid polymorphism and fluidity in model membranes. possible to regulate the dolichyl-P level by altering cellular J. Biol. Chem., 260: 2724-2751, 1985. 11. Mills, J. T., and Adamany, A. M. Impairment of dolichyl saccharide synthesis secretion or uptake of these lipids. However, this regulatory and dolichol-mediated glycoprotein assembly in the aortic smooth muscle cell in culture by inhibitors of cholesterol biosynthesis. J. Biol. Chem., 253: mechanism seems unlikely, since it was recently suggested that 5270-5273, 1978. this type of transport occurs only to a very limited extent (34). 12. Carson, D. D., Earles, B. J., and Lennarz, W. J. Enhancement of protein Upon analysis of the distribution of individual polyisoprenes glycosylation in tissue slices by dolichyl phosphate. J. Biol. Chem., 256: 11552-11557, 1981. in both the free and phosphorylated forms, in microsomes no 13. Potter, J. E. R., James, M. J., and Kandutsch, A. A. Sequential cycles of differences between control tissue and hepatomas could be cholesterol and dolichol synthesis in mouse spleens during phenylhydrazine- induced erythropoesies. J. Biol. Chem., 256: 2371-2376, 1981. found, with the exception of a minor decrease in the relative 14. Eggens, I., Eriksson, L. C., Chojnacki, T., and Dallner, G. Role of dolichyl amounts of dolichol-21 and dolichyl-21-P in hepatoma tumors. phosphate in regulation of protein glycosylation in 2-acetylaminofluorene- The mechanism behind these minor alterations in chain length induced carcinogenesis in rat liver. Cancer Res., 44: 799-805, 1984. 15. Rossignol, D. P., Scher, M., Waechter, C. J., and Lennarz, W. J. Metabolie remains to be investigated. interconversion of dolichol and dolichyl phosphate during development of Finally, the question of changes in dolichyl-P levels and the sea urchin embryo. J. Biol. Chem., 25Â«:9122-9127, 1983. 16. Appelkvist, E. L., Chojnacki, T., and Dallner, G. Dolichol (C55) mono- and accompanying effects on glycosylation was investigated earlier pyrophosphatases of the rat liver. Biosci. Rep., /. 619-626, 1981. in the sea urchin, where the level of dolichyl-P was suggested 17. Rip, J. W., Rupar, A., Chandhary, N., and Carroll, K. K. Localization of a dolichyl phosphate phosphatase in plasma membranes of rat liver. J. Biol. to be rate limiting (IS). In addition, a previous study on chem Chem., 256:1929-1934, 1981. ically induced liver nodules in rats showed that a lowered 18. Keller, R. K., Adair, W. L., Cafmeyer, Jr., N., Simion, F. A., Fleischer, B., dolichyl-P-level was paralleled by a decreased rate of glycosyl and Fleischer, S. Characterization of polyisoprenyl phosphate phosphatase activity in rat liver. Arch. Biochem. Biophys., 249: 207-214, 1986. ation (14). Our data with a relatively high level of dolichyl 19. Allen, C. M., Kalin, J. R., Sack, J., and Verizzo, D. CTP-dependent dolichol phosphate in hepatoma microsomes or stable levels in tumor phosphorylation by mammalian cell homogenates. Biochemistry, 17: 5020- 5026, 1978. homogenates, together with the slightly increased glycosylation 20. Burton, W. A., Scher, M. G., and Waechter, C. J. Enzymatic phosphorylation in these fractions, were in line with this idea. of dolichol in central nervous tissue. J. Biol. Chem., 254: 7129-7136, 1979. 21. EkstrÃ¶m,T. J., Chojnacki, T., and Dallner, G. Metabolic labeling of dolichol The changes in microsomal dolichyl-P levels and glycosyla and dolichyl phosphate in isolated hepatocytes. J. Biol. Chem., 259: 10460- tion do not, however, prove that the polyisoprenoid levels are 10468, 1984. 22. Keller, R. K. The mechanism and regulation of dolichyl phosphate biosyn rate limiting, since changes in the amounts and activities of the thesis in rat liver. J. Biol. Chem., 267: 12053-12059, 1986. glycosylating enzymes cannot be excluded. 23. Ã„strand, I.-M., Fries, E., Chojnacki, T., and Dallner, G. Inhibition of dolichyl phosphate biosynthesis by compactin in cultured rat hepatocytes. Eur. J. It is thus possible that the level of dolichyl phosphate under Biochem., /55: 447-452, 1986. specific pathological conditions is partly influenced by changes 24. I oliiÂ«)!]!,<)..and Dallner, G. Dolichol and dolichyl phosphate in human in the CTP-dependent dolichol kinase and phosphatase activi tissues. Br. J. Exp. Pathol., 67: 757-764, 1986. 25. Autuori, F., Svensson, H., and Dallner, G. Biogenesis of microsomal mem ties and that the amount of this phosphorylated polyisoprenoid brane glycoproteins in rat liver. J. Cell Biol., 67:687-699, 1975. may influence the rate of protein glycosylation in the liver. 26. Behrens, N. H., and Tabora, E. Dolichol intermediates in the glycosylation Further studies on dolichyl-P synthesis and degradation, on the of proteins. Methods Enzymol., 50:402-439, 1978. 27. Eriksson, L. C. Studies on the biogenesis of endoplasmic reticulum in the specificities of dolichol kinases and phosphatases, and on the liver cell. Acta Pathol. Microbio!. Scand. Suppl. 239, pp. 1-72, 1973. specific microsomal glycosyltransferases under pathological 28. Beaufay, H., Amar-Costesec, A., Feytmans, E., Thines-Sempoux, D., Wibo, M., Robbi. M., and Berthet, J. Analytical study of microsomes and isolated and normal conditions are, however, necessary to clarify these subcellular membranes from rat liver. J. Cell Biol., 61: 188-200, 1974. 29. Cornali, A. G., Bardawill, C. J., and David, M. M. Determination of serum questions. proteins by means of the Biuret reaction. J. Biol. Chem., /77: 751-766, 1949. 30. Ma, M. H., and Blackburn, C. R. B. Fine structure of primary liver tumors ACKNOWLEDGMENTS and tumor-bearing livers in man. Cancer Res., 33: 1766-1774, 1973. 31. Eggens, I., Chojnacki, T., Kenne, L., and Dallner, G. Separation, quantifi The authors wish to thank Dr. Sven-Olof Bohman for preparing the cation and distribution of dolichol and dolichyl phosphate in rat and human electron micrographs which were discussed in the text. tissues. Biochim. Biophys. Acta, 757: 335-368, 1983. 3423

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 1988 American Association for Cancer Research. DOLICHOL KINASE AND PHOSPHATASE IN HUMAN HEPATOMAS 32. Yogeeswaran, G. Cell surface glycolipids and glycoproteins in malignant G. Reaction of optically active 5-and A-forms of dolichyl phosphates with transformation. Adv. Cancer Res., 38: 289-338, 1983. activated sugars. Biochem. Biophys. Res. Commun., 130:460-466, 1985. 33. Eggens, I., and Elmberger, G. Polyprenol content in primary human liver 36- Canning, A. E., Brunk, U., and Dallner, G. Phtalate esters and their effect carcinoma. Acta Chem. Scand., B 39,4:326-328, 1985. ,7 r"h' '^V PTn^ r^Th '', t4' rn- , ,.,,.â€¢ ,. _. , â€ž ...... _ , . U â€ž r* , .. 37. Chojnacki, T., and Dallner, G. The uptake of dietary polyprenols and their 34. Elmberger, G., Kalen, A., Appelqvist, E. L., and Dallner, G. In varo and in modification to active dolichols by the rat liver. 3. Biol. Chem., 25Â«:916- vivo synthesis of dolichol and other main mevalonate products in various 923, I9g3. organs of the rat. Eur. J. Biochem., 168: 1-11, 1987. 38. Elmberger, G., and Engfeldt, P. Distribution of dolichol in human and rabbit 35. Low, P., Peterson, E., Mizuno, M., Takigawa, T., Chojnacki, T., and Dallner, blood. Acta Chem. Scand., B 39, 4: 323-325, 1985.

3424

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 1988 American Association for Cancer Research. Cytidine 5′-Triphosphate-dependent Dolichol Kinase and Dolichol Phosphatase Activities and Levels of Dolichyl Phosphate in Microsomal Fractions from Highly Differentiated Human Hepatomas

Ivan Eggens, Johan Ericsson and Örjan Tollbom

Cancer Res 1988;48:3418-3424.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/48/12/3418

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/48/12/3418. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.