Comparat ive Studies of GIucose-fed and GIucose-starved Hamster Cell Cultures: Responses in Galactose Metabolism1 C. WILLIAM CHRISTOPHER, WENDY W. COLBY, DONNA ULLREY HERMAN M. KALCKAR* Department of Biological Chemistry, Haruard Medical School and the Section of Cell Metabolism in the john Collins Warren Laboratories of the Huntington Memorial Hospital of Harvard University at the Massachusetts General Hospital, Boston, Massachusetts 02114 AND

ABSTRACT The metabolic flow of trace amounts of D-[14C]-galactosewas followed in cultures of transformed and untransformed hamster cells over a period ranging from five minutes to two hours. The results of chromatographic and enzymatic analyses of the soluble pools are described. Non-glycolytic cells (previously deprived of sugar for periods of up to 24 hours) convert D-galactose to galactose-1-phosphate and uridine diphosphoglucuronic acid in 10 to 20 minutes. In the same short assay time, glycolytic cells which have been maintained for 24 hours in media containing glucose or galactose convert D-galactose to uridine diphosphogalactose and uridine diphosphoglucose (ratio 1.4: 1).Longterm deprivation of sugar also results in 3- to 4-fold increases in the uptake of galactose. In addition, the incorporation of galactose label into chloroformmethanol soluble material appears to be influenced by the culture conditions of' the untransformed cells while incorporation in the transformed cells appears unaffected. When cycloheximide is included in the maintenance medium for extended periods, the non-glycolytic cells also show increases in galactose uptake rates but the glucose-fed, glycolytic cells lose uptake ability. UDPhexose is the main galactose metabolic peak in the soluble pools of the cycloheximidetreated, glycolytic and the cycloheximide-treated, non-glycolytic cells. The results of these experiments suggest that uptake of galactose and its subsequent metabolism are under separate control.

Cultured cells, transformed by oncogenic viruses, show enhanced transport of amino acid analogs (Foster and Pardee, '69; Isselbacher, '72) and hexose uptake (Hatanaka et al., '69; Martin et al., '71; Hatanaka, '74). Another phenomenon of enhanced hexose uptake has also been observed when untransfonned or transformed cells are maintained in culture media free of glucose. This was first observed by Amos and his co-workers (Martineau et al., '71) when they cultured chick embryo fibroblasts (CEF)3in media devoid of glucose. Thus, when cells which have limited reserves of glycogen are deprived of glucose, they cannot benefit from glycolysis. When tested for glucose, 2-deoxyglucose (2-dG) or 3-0-methylgluJ. CELL. PHYSIOL., 90: 387-406.

Received May 17, '76. Accepted June 29, '76. 'This work was supported by grants from the American Cancer Society BC-120 and the National Institutes of Health AM-05507 and the National Science Foundation BMS71-01291 A03. This is publication No. 1517 of the Cancer Commission of Harvard University. *Dr. Herman M. Kalckar was Fogarty Scholar in residence at the National Institutes of Health and is Visiting Professor of Biological Chemistry at the John Collins Warren Laboratories of the Huntington Memorial Hospital of Harvard University. 3Abbreviations used: CEF, chick embryo fibroblast; NIL, stable line of hamster fibroblasts; BHK, baby hamster kidney; py, polyoma; Gal, D-galactose; Gal-1-P, u-Dgalactose-1-phosphate; G or Glc, D-glucose; G-6-P, a-Dglucose-6-phosphate; G-1-P, a-D-glucose-1-phosphate; 2-dG, 2-deoxy-D-glucose; 3-O-meG, 3-0-methyl-D-glucose; NAD, nicotinamide adenine dinucleotide; NADH, reduced NAD; NADP, nicotinamide adenine dinucleotide phosphate; UDPGlcUA, uridine 5'-diphosphoglucuronic acid; UDPGal, uridine 5' diphosphogalactose; UDPG, iiridine 5'-diphosphoglucose; MEM, minimum essential medium (Eagle); PBS, Dulbecco's phosphate buffered saline.

387

388

C . CHRISTOPHER, W. COLBY, D . ULLREY AND H . KALCKAR

cose (3-0-meG)uptake rates, the glucosestarved, non-glycolytic cells showed dramatically increased rates over sugar fed, glycolytic cells (Martineau et al., '72; Kalckar and Ullrey, '73; Hatanaka, '73; Shaw and Amos, '73; Ullrey et al., '75; Kletzien and Perdue, '75; Rossow et al., '75; Christopher et al., '75). This type of perturbation not only affects facilitated diffusion (energy independent transport) of sugars, but as was demonstrated by Ullrey et al. ('75), also enhances active transport of amino acids. Perturbations of cultures-whether they are elicited by virus transformation or by glucose deprivation-appear to affect not only transport but also many of the early steps of hexose metabolism. Indeed, although 2-deoxyglucose and 3-0-methylglucose are the commonly used hexose analogs for the study of uptake and transport regulations, for some studies of the kinetics of regulation of physiological sugars (like glucose and galactose) these analogs have limited application. Thus, transformations enhance the glucose transport system (Weber, '73; Eckhardt and Weber, '74) but also the metabolic steps at least as far as the formation of fructose-1,6-bisphosphate (Fodge and Rubin, '73). As far as the metabolism of galactose is concerned, the steps including UDPhexose formation (and also its fission) are enhanced in NILpy cultures as compared with the untransformed NIL cultures (Ullrey et al., '75). Moreover, in cultures of hamster cells (BHK and NIL lines and their corresponding polyoma (py) transformed clones) the apparent catabolic repression of hexose uptake is particularly well expressed when the substrate for the uptake measurement is galactose (Kalckar and Ullrey, '73; Ullrey et al., '75). In comparison, the rate of galactose uptake in the glucose-starved NIL cultures is increased by an average of 4-to 5-fold over that of glucose-fed cultures (Ullrey et al., '75). Certain advantages are provided b y the use of physiological sugars in the study of

regulations of hexose metabolism in cultured cells. The pathway of galactose metabolism may be called "amphibolic" (Davis, '61) because of the formation of nucleotide galactose (UDPGal) makes possible the performance of a physiological double feature; anabolic formation of oligosaccharides and catabolic formation of glucose-6-phosphate (G-6-P) (Robinson et al., '66). In our previous study of differences between cultures maintained in media containing glucose and those deprived of glucose (Ullrey et al., '75), we could not help noticing the sharp differences in the early amphibolic pathway of galactose. Since the study of the flow of galactose in animal cells could be used as a good model system for studying the regulation of hexose uptake, and since extensive catabolic feedback inhibitions could govern this flow, we felt it was important to identify the galactose accumulation products. In addition, the well known changes in hexose uptake caused by oncogenic virus transformation of animal cells may be the result of significant changes in the pattern of galactose metabolism. For example, the early galactose catabolism might be reflected in the biosynthesis of simple and complex galactosyl compounds. Thus, a focus on a relatively simple hexose metabolic system may well turn out to be a useful device for studying the differences between transformed and untransformed cells. In this paper we report analyses of galactose metabolism in cultures of untransformed and polyoma transformed hamster cells. The comparisons were made using cells that were maintained in medium containing glucose and in glucose-free medium and also using cells maintained for 18 to 24 hours in the presence of cycloheximide. MATERIALS AND METHODS

Chemicals, enzymes, and radioisotopes Inorganic and organic salts were dissolved in deionized water and were the highest purity available. Nucleotides

GALACTOSE METABOLISM I N HAMSTER CELL CULTURES

used for UV standards (NAD, UDPGlc, UDPGal, UDPGlcUA, and UDPxylose) and cycloheximide were obtained from Sigma Chemical Co. Also obtained from Sigma were the following purified enzymes: galactose dehydrogenase [Dga1actose:NAD oxidoreductase (E.C. 1.1.1.48)], UDPG dehydrogenase [UDPglucose:NAD oxidoreductase (E.C. 1.1.1.22)] and G-6-P dehydrogenase [Dglucose-6-phosphate:NADP oxidoreductase (E.C. 1.1.1.49)]. UDPGlcUA lyase [UDPglucuronate carboxy-1,yase (E.C. 4.1.1.35)] from Crytococcus laurentii was a gift from Dr. D. S. Feingold, University of Pittsburgh. Glucostat reagent was from Worthington Biochemical Corp. Aquasol and the following uniformly labeled [‘*C(U)] biochemicals were purchased from New England Nuclear: D-galactose, galactose-1-phosphate, uridine diphosphate galactose, uridine diphosphate glucose and uridine diphosphate glucuronic acid. Cells and culture conditions Hamster cells of the line NIL and the polyoma transformed NIL, NILpy, (generously provided by Maureen T. Gammon of the Gastrointestinal Unit at The Massachusetts General Hospital) were grown in 60 mm Falcon Petri dishes or 25 mm glass shell vials (Arthur H. Thomas, Co.) in Dulbecco’s Modified Eagle’s Medium supplemented with 10% fetal calf serum (Microbiological Associates). Cells were allowed to grow in this medium for three or four days without a change of medium. Twenty to 24 hours prior to the uptake tests (see below), the medium was changed to Eagle’s Minimal Essential Medium (MEM) with double the concentration of amino acids and vitamins plus 10% dialyzed fetal calf serum (Gibco). The normal glucose concentration in this medium was 10 mM (2 mg per ml) and, depending upon the experiment, was omitted or replaced by other sugars as indicated in the text. Before changing the medium, the cells were twice washed with 10 ml of the medium

389

in which they were to be incubated for the next 20 to 24 hours. Cultures were routinely checked for mycoplasma infections. After an 18 to 24 hour exposure of representative cultures to 3H-thymidine, inspection of developed photographic emulsions revealed that the label appeared exclusively in the cell nuclei. The absence of extranuclear thymidine incorporation indicated that the cells were free of mycoplasma. Uptake tests After the 20 to 24 hour incubation in fresh medium, the medium was removed and the cells were rinsed three times at ambient room temperature with DulbecCO’S phosphate buffered saline (PBS) pH 7.4. The cells which remained anchored to the surface were incubated in PBS containing the radioactive sugars; the time and temperature of the incubation are indicated for individual experiments in the text. Routine conditions were to incubate the cells in PBS containing between 1 p M and 50 p M D-[14C(U)] galactose for 20 minutes at 23°C. After the uptake incubation, the medium was removed by aspiration and the cells were rapidly washed three times with ice-cold PBS. Two milliliters of 70% ethanol were then added to each dish (0.25 ml to each vial) and allowed to remain at room temperature for at least one hour. An aliquot of the ethanol extract (0.1 ml) was placed in a Beckman LS 233 scintillation counter in 10 ml Aquasol. Triplicate 10 pl aliquots of the uptake mix in PBS was diluted in ethanol, was counted under the same conditions as just described and these counts were used to compute CPM per nmole available in the test solution. Denatured (ethanol insoluble) cells were dissolved in 0.1 N NaOH and the NaOH-soluble material was used for protein determination according to the method of Lowry et al. (’51).Counts per minute per mg were converted to nmoles taken up per mg protein. Incorporation of the sugars into ethanol insol-

390

C . CHRISTOPHER, W . COLBY, D. ULLREY A N D H. KALCKAR

uble material was determined by counting a neutralized aliquot of the NaOH soluble material in the same scintillation system described above.

Chromatography Aliquots of the ethanol soluble extract were concentrated under a stream of nitrogen gas at room temperature and then taken up in 50 pl of 70% ethanol. All of the resuspended radioactivity along with non-radioactive UDPGal and UDPGlcUA was dried on Whatman chromatography paper ( 1 mm). T h e UDPsugars were included as internal controls for comparison of migration distances with markers spotted separately and chromatographed in parallel. The separate marker channels also included free 14C-galactose and 14C-Gal-l-P. Chromatography (descending) was for 16 to 18 hours at room temperature using the Paladini-Leloir ('52) ethanol acetate system as previously described (Kalckar and Ullrey, '73; Ullrey et al., '75). Following chromatography, the sheets were marked for short wave UV absorption regions using an Ultra Violet Products, Inc. viewer and cut into 4 cm x 56 cm strips. The strips were then scanned on a Packard Model 7201 Radiochromatogram Scanner. Enzymatic analyses Protein-free, ethanol soluble filtrates from cells deprived of glucose for 24 hours were divided into several portions. One portion was dried and chromatographed directly (see above). Another portion (0.7 ml) was dried and taken up in 0.5 ml of 0.1 N HC1 then hydrolyzed for 30 minutes at 100". The hydrolysate was neutralized with NaOH and then 0.1 ml of 0.1 M Tris buffer (pH 8.6) was added. To 0.2 ml of the neutralized, buffered hydrolysate was added NAD and D-galactose to make the final concentrations 5 mM and 0.5 mM respectively and the final volume 0.22 ml. This reaction mix was chromatographed directly. To another 0.22 ml volume of reaction mix-

ture was added 5 pl (50 punits) galactose dehydrogenase. The mixture was stirred and allowed to incubate at 23" for one hour. Following the incubation, 50 p1 of concentrated NH,OH were added and the mixture was allowed to stand for 30 minutes before application to the chromatography paper. A third portion of the original filtrate (0.5ml) was dried and taken up in 0.15 ml of 67 mM Tris buffer (pH 7.4) containing 1.3 mM NAD and 3.3 mM UDPGlcUA. Twenty microliters of purified UDPGlcUA lyase was added to the mixture and incubated for two hours at 37". The reaction was stopped by the addition of 0.5 ml 100%ethanol. To a control portion of the original filtrate, ethanol was added before the addition of lyase. Aliquots of the final mixture (0.15 ml) were dried and chromatographed. A fourth portion of the original filtrate (0.5 ml) was dried and then taken up in 65 p1 phosphate buffered saline (pH 7.4) containing 0.3 mM G-6-P, 0.3 mM G-1-P, 0.3 mM Gal-1-P, 3 mM NADP, 0.03 mM G-1,6-diphosphate, 2 pl phosphoglucomutase and 2 pg G-6-P dehydrogenase. The assay mixture was incubated for one hour at 22" and then 10 pl concentrated NH,OH were added to stop the reaction. After 30 minutes, the assay mixture was dried on chromatography paper and chromatographed as described above.

Lipid extraction Ethanol soluble extracts of cells were separated in chloroform-methanol according to the method of Folch et al. ('57). At least 20 volumes of chloroformmethanol (CM, 2:l) were added to the dried ethanol extract (redissolved in water) and thoroughly mixed. No precipitate was observed. After 15 minutes, 0.2 volumes of 20 mM KC1 was added and mixed by shaking. The upper, aqueous phase was removed and the lower, chloroform layer was then twice washed with 0.4 volumes chloroform-methanolKC1 solution (3:49:48) and the washes combined with the aqueous phase. Each

GALACTOSE METABOLISM I N HAMSTER C E L L C U L T U R E S

phase was prepared for counting and paper chromatography by drying under N, and redissolving in 70% ethanol. RESULTS

Uptake and chromatographic patterns Hamster cells, maintained for extended periods in medium free of glucose, do not excrete significant amounts of lactate (see table VI in Ullrey et al., '75) and for this reason are called "nonglycolytic". These non-glycolytic cells take up between three and four times more galactose in 20 minutes than cells maintained in media containing glucose (glycolytic conditions) (Kalckar and U11rey, '73; Ullrey et al., '75). Figure 1 shows the typical radiochromatogram profiles of the ethanol soluble metabolites from glucose-fed, glycolytic cells (fig. 1A) and non-glycolytic cells (fig. 1B). The glycolytic condition results in an accumulation of radioactivity as a single peak which migrates from the origin to a position coincident with UDPhexose markers, UDPG or UDPGal. By comparison, two peaks of radioactive metabolites obtained from the soluble pool of cells deprived of sugar for 24 hours can be chromatographically separated. Neither of the peaks is UDPhexose. The predominant peak which corresponds to approximately 90 to 95%of the soluble pool of these non-glycolytic cells is hexose monophosphate and is indistinguishable from the Gal-1-P marker. The second peak, typical of starved cells and comprising the remaining 5-10% of the soluble material, migrates to the position of the UDPGlcUA marker. There is little if any detectable UDPhexose present in sugar starved cells. Occasionally a variable amount of free galactose is observed in the pools of either glycolytic or nonglycolytic cells. Glucose-starved polyoma transformed NIL cells invariably have patterns of galactose metabolism identical with the untransfonned NIL cells. Likewise, well fed NILpy cells most often have a pattern identical with glucose-fed NIL cells even

39 1

though their uptake rate is two to four times higher. However, on occasion, glucose-fed NIL and NILpy cultures contain three peaks of radioactive galactose catabolites in the soluble pools; some hexose monophosphate, a predominant amount of UDPhexose and some UDPGlcUA (fig. 1C). Often, but not necessarily always, this mixed profile appears in cultures whose cell density is particularly high. This mixed profile suggests that perhaps some cultures may deplete their supply of glucose to a concentration below a critical low level (U11rey et al., '75). However, direct measurement of the glucose remaining in the supporting medium by the Glucostat test revealed that the concentration of glucose was above the level (approximately 100 p g per ml) which results in nonglycolytic profiles. Thus, the reasons for the appearance of these mixed profiles is not clear.

ldentijication of catabolites The nucleotide sugar peak of glycolytic NIL cells was analyzed using the coupled UDPG dehydrogenase and UDPG-UDPGal epimerase system described by Robinson (Robinson et al., '66) and was found to contain UDPGal and UDPG in a ratio of 1.4: 1. This is in good agreement with a previous report described by us (Kalckar et al., '73). In non-glycolytic cultures essentially all of the slow-moving peak of radioactivity which co-migrated with UDPGlcUA (fig. 1B) was identified as UDPGlcUA. The concentrated filtrate extracted from glucose-starved NIL cells was incubated with excess UDPGlcUA lyase in order to convert any of the UDPGlcUA in the slow migrating metabolite to a new compound (UDPxylose). UDPxylose migrates faster than UDPGlcUA but slower than hexose monophosphate (the only other constituent of the soluble pool in non-glycolytic cells). Chromatographic analysis of a representative experiment is shown in figure 2. It can be seen that the position of the slow-moving

392

C. CHRISTOPHER, W. COLBY, D. ULLREY A N D H . KALCKAR

b

I

L

I

\P V\

MIGRATION DISTANCE

f ern )

Fig. 1 Chromatographic patterns of galactose metabolites from glycolytic and non-glycolytic hamster cells. NIL cultures were prepared for 14C-galactose uptake tests as described under MATERIALS AND METHODS. The incubation was for 20 minutes at 23”. After washing with PBS the reaction was stopped with 70% ethanol. The extract was concentrated under a stream ofN2. Paper chromatography and scanning of the chromatograms were performed as described under MATERIALS AND METHODS. Fifty nmoles each of unlabeled UDPGlcUA and UDPGal were used as standards and mixed with the labeled ethanol extract prior to chromatography. UDPGlcUA and UDPGal were also mixed with additional standards, L4C-galactose-l-Pand 14C-galactose,and this mixture was chromatographed in a separate lane. The UDP-sugar standards were detected by their absorption in short wave (U.V.) light. Panel A shows the separation of the ethanol extract from near confluent cells maintained in M E M plus 2 mg glucose per rnl; panel B, near confluent cells maintained in glucose-free MEM; panel C, confluent cells maintained in MEM plus 2 mg glucose per ml.

(UDPGlcUA) peak was shifted to a faster peak from the non-glycolytic cells as moving peak (UDPxylose) after incuba- UDPGlcUA. tion with UDPGlcUA lyase. This specific The hexose monophosphate peak was enzymatic test as well as thin-layer identified as predominantly Gal-1-P by chromatography on polyethyleneimine the following combination of enzymatic cellulose in LiC1-acetic acid as described and chromatographic analysis prior to the previously by Randerath and Randerath usual paper chromatography, the concen(’67), clearly identifies the slow-moving trated ethanol extract from non-glycolytic

GALACTOSE METABOLISM IN HAMSTER C E L L C U L T U R E S

0

393

0

0

" !

Y

5

UDPxy lose

*

2 SO 5

[

I

t 5

8 I

I

I

I

I

I

25 MI GIOA N O 1 D/STANC€ { cm ) 20

Fig. 2 Enzymatic analysis of the slow moving radioactive galactose metabolite from .non-glycolytic cultures. Near confluent NIL cultures were incubated 20 hours at 37" in culture medium free of glucose and the uptake of '4C-galactose was for 20 minutes at 23"as described in figure 1. The concentrated ethanol extract was incubated with UDPGlcUA lyase from Crytococcus laurentii for two hours at 37" and the enzyme was denatured by the addition of ethanol (B). In the control (A), ethanol was added prior to the lyase and incubated as above. After concentrating the protein free mixture under N2, paper chromatography and scanning were performed as described in MATERIALS AND METHODS. The following non-radioactive standards were mixed with the samples, chromatographed separately and detected by their absorption in U.V. light: NAD, UDPGlcUA, UDPGal, and UDPxylose. The only change in the chromatographic profile of the non-glycolytic cultures caused by the lyase occurred between the UDPGlcUA and the UDPGal markers and only this region is shown.

NIL cells was hydrolyzed in 0.1 N HC1 for generated by the acid hydrolysis step was 30 minutes at 100" and prepared for en- quantitatively converted to a distinctly zymatic modification. Briefly, one aliquot of the neutralized hydrolysate (control) was taken up in assay buffer without enzymes while another aliquot was incubated with excess galactose dehydrogenase and NAD. After exposure of the mixture to ammonia and evaporation, the samples were chromatographed. As appears in figure 3, incubation with NAD and galactose dehydrogenase (followed by ammoniacal splitting of any lactone) altered the chromatographic profile of the hexose monophosphate peak as follows: (1) the majority of the hexose monophosphate peak was converted to a fast moving peak which migrated to a position indistinguishable from the free sugar marker (D-galactose), (2) the free sugar

slower migrating peak by the action of galactose dehydrogenase and its new position corresponded with the position of galactonate. Incubation of the unhydrolyzed filtrate with the coupled enzyme system, phosphoglucomutase (plus Mgz+ and glucose-1,6-biphosphate) and G-6-P dehydrogenase (plus NADP) failed to convert any of the hexose monophosphate to a slower migrating phosphogluconate (fig. 4).

Metabolic profile ratio For convenience it is possible to express the physiologic conditions of a culture (even one containing a mixed metabolic profile) as being glycolytic or non-glycolytic by applying values ob-

394

C . C H R I S T O P H ER , W . COLBY, D. ULLREY AND H . KALCKAR

tained from measurements of the areas under each metabolite peak to the following equation: UDPhexose Metabolic profile ratio = Gal-1-P + UDPGlcUA

A metabolic profile ratio of greater than 1 indicates cells that have been maintained in medium which can support glycolysis. A ratio of less than 1 is indicative of nonglycolytic cultures. Table 1 summarizes a series of experiments designed to compare uptake rates with metabolic patterns. In those experiments glucose in the supporting medium was either exUDPGal ~~l-l-p 0 -

i

A

Galactose H

h

'

STANDARD ETHANOL EXTRACT

n, I

Y + I

I TA DA

I

D'A

+

J 15

Fig. 3 Chemical and enzymatic analysis of the 14C-hexose monophosphate peak from non-glycolytic cultures: hot acid hydrolysis and galactose dehydrogenase modification. Near confluent NIL cultures were incubated in glucose free culture medium and allowed to take up 14C-galactose as described in figure 1. Prior to chromatography the protein free extract was divided into several aliquots and modifications were performed as described in MATERIALS AND METHODS. One aliquot was chromatographed directly (A). Another aliquot was made 0.1 N HC1 and hydrolyzed at 100" for 30 minutes. The hydrolysate was neutralized with NaOH and taken up in 67 mM Tris buffer (pH 8.6) containing NAD and a portion chromatographed (B). To another portion of the hydrolysate containing buffer, NAD and galactose was added 50 munits galactose dehydrogenase and this was incubated at 23" for one hour (C). Authentic 14C-galactose-l-P was hydrolyzed in hot acid, modified by galactose dehydrogenase under the same conditions, and chromatographed separately. Chromatography and scanning procedures were the same as described in figure 1. For clarity only the region between the hexose monophosphate and the free sugar markers is shown. Procedures for chromatography of the standard markers are given in figure 1.

sTANDARrm I

HYDROLYZED EXTRACT

4

perimentally omitted or replaced with galactose. It can be seen that NIL cells deprived of sugar take up about four times more galactose than glucose fed cells and the profile ratio of the starved cells is 0.06. Tranformed cells have the usual 2- to %fold higher uptake rates even when their metabolism is clearly glycolytic and the uptake of galactose is 4- to 5-fold higher than NIL sugar fed cells when they are in non-glycolytic conditions (table 1).Galactose behaves as a metabolic reducing sugar when substituting for glucose in t h e culture medium of NIL cells. As is shown in table 1, galactose is taken up at about the same rate as it is by glucose fed cells and the metabolic profile ratio is 2.8. Since uptake is usually much higher in transformed cells than it is in untransformed cells, it is often difficult to determine if increases in these cells are due to the transformation effect (Hatanaka, '74) or to sugar starvation effects (Ullrey et al., '75). It is here that the galactose metabolic profile ratio becomes a useful indicator.

25

MIGRATION DISTANCE ( c m Figure 3

35

GALACTOSE METABOLISM I N HAMSTER C E L L C U L T U R E S

As can be seen in table 1,the NILpy cells fed D-galactose for 24 hours have a glycolytic pattern (ratio 4) in spite of a very high uptake rate. UDPGal Gal-1-P

Galactose

"s I

4

're 0

STANDARD G-6-P t

ENzYMFS

, STANDARD

25 35 MIGRATION DISTRANCE f e m Fig. 4 Enzymatic analysis of the 14C-hexose monophosphate peak from non-glycolytic cultures: treatment with phosphoglucomutase and G-6-P dehydrogenase. A portion of the unhydrolyzed ethanol extract from non-glycolytic cultures was concentrated and taken up in phosphate buffered saline containing G-6-P, G-1-P, Gal-1-P, G-1,6-diphosphate and NADP and then an aliquot chromatographed directly (A). To the remaining aliquot was added phosphoglucomutase and G-6-P dehydrogenase and incubated for one hour at 23". The reaction was stopped with NH,OH and the mixture was chromatographed (B). Details are given in MATERIALS AND METHODS. Standard 14C-G-6-P, l4C-G-1-P and 14C-Gal-1-P were incubated under the same conditions (C). The region of the chromatograms between UDPGal and free galactose is shown. 15

395

Uncoordinated control of uptake and metabolism The correlation of galactose uptake rate enhancement and non-glycolytic metabolic profiles is not absolute. It was seen earlier (table 1) that high uptake rates can exist with glycolytic cells such as NILpy cells. The same effect can be brought about in untransformed cells if they are first rendered non-glycolytic and then refed high concentrations of glucose or galactose. These NIL cells, when suddenly supplied with glucose, respond by completely converting their metabolic profile from the typical non-glycolytic pattern (containing Gal-1-P and UDPGlcUA) to the typical glycolytic pattern (only UDPhexose) in less than 30 minutes. When the cells are suddenly fed with medium containing galactose, they also completely convert from a nonglycolytic to a glycolytic metabolism in about one hour (table 2). Yet, within this span of time in excess reducing sugars, the enhanced uptake level persists. Thus, the changes in metabolism demonstrated by the shifts in accumulated metabolic products occurs in a more rapid way than changes in uptake rates. A converse example was discovered when cycloheximide was used to stop protein synthesis by greater than 95%. When cycloheximide (10pg per ml) was present in the medium of either NIL or NILpy cells from the start of a 24 hour sugar deprivation period, a variable increase in the rate of uptake was always observed (table 3). In addition, although these cells were incubated in the complete absence of glucose, UDPhexose appeared as the sole accumulation product following a 20 minute uptake of 50 p M D-galactose (table 3,fig. 5). Glucosefed cells, treated for long periods with cycloheximide, consistently lose uptake activity. In some experiments the loss was greater than 90% in the 24-hour treatment period. The reasons why the glycolytic cells decrease their uptake rates, whereas the non-glycolytic cells increase their uptake rates are not

396

C. C H R I S T O P H E R , W. COLBY, D . ULLREY AND H . KALCKAR TABLE 1

Galactose uptake and metabolism Culture condition

Number of samples

Galactose uptake (relative)

NIL

+ Glucose + Galactose

NILPY

+ Glucose + Galactose

24 33 32 12 10 12

100 137 407 298 343 474

Cell line

No sugar

No sugar

UDPhexose Gal-1-P

+ UDPGlcUA

4.41 ? 2.68 2.76 ? 1.39 0.058 ? 0.045 2.96 5 1.60 3.95 ? 2.25 0.15 r 0.12

Hamster cells (NIL and NILpy) were grown to near confluency in 60 mm plastic dishes and then maintained in glucose-free medium (no sugar), in medium containing 2 mg glucose per ml or 2 mg galactose per milliliter for 20-24 hours as described in MATERIALS AND METHODS. Uptake of 15 pM D-[14C]-galactose was for 20 minutes at 23". Extraction of the soluble pool with 70% ethanol and chromatographic procedures are described in MATERIAL AND METHODS. TABLE 2

Reversal of metabolic projile Culture condition Cell line

First incubation

Second incubation

Galactose uptake (nmoledmg)

Main metabolic peak ~

NIL

+ Glucose (16 hours) No sugar No sugar

No sugar NILpy

+ Glucose (24 hours) No sugar No sugar No sugar

+ Glucose (60 min) No sugar (60 min) + Glucose (30 min) + Glucose (60 min) + Glucose (60 min) No sugar (60 min) + Glucose (30 min) + Glucose (60 min)

UDPhexose Gal-1-P

+ UDPGlcUA

~

0.516

UDPhexose

1.56

1.891

Gal-I-P

0.089

1.404

UDPhexose

2.87

1.740

UDPhexose

1.36

0.339

UDPhexose

2.87

0.908

Gal-1-P

0.076

0.806

UDPhexose

1.68

1.202

UDPhexose

4.30

The cells were incubated overnight (16 hours) in glucose-free medium (no sugar) or medium containing 2 mg glucose per milliliter. The cells were washed and incubated for 30 minutes or 60 minutes in fresh medium (second incubation). After the second incubation the cells were again washed and allowed to take up 15 pM D-[W]-galactose for 20 minutes at 23". Procedures for uptake, ethanol extraction, and chromatography are described in MATERIALS A N D METHODS.

known, Likewise, the reasons why NILpy cells appear less responsive to cycloheximide treatment than the untransformed NIL cells are not known. A one-hour exposure of glycolytic or nonglycolytic cultures to cycloheximide did not affect their previously established uptake or metabolism of galactose. In cultures which had a mixed metabolic profile to start (with UDPhexose the pre-

dominant peak), the same pattern was observed after 24 hours in glucose-free medium containing cycloheximide. Phosphohydrolase activity Because of the absence of the UDPhexose peak and the abundance of the Gal-1-P peak in chromatograms of pools of non-glycolytic cells, the possibility that endogenous UDPsugars were being

397

GALACTOSE METABOLISM I N HAMSTER C E L L C U L T U R E S TABLE 3

Effect of cycloheximide on galactose uptake and metabolism Culture condition (24-hour incubation) Cell line (* Glucose)

( 2 Cyclohex)

Galactose uptake (nmoledmg)

Main me tabo1ic peak

0.274 0.032 1.458 0.725 0.654 0.184 1.236 .82

UDPhexose UDPhexose Gal-I-P UDPhexose UDPhexose UDPhexose Gal-1-P UDPhexose

UDPhexose Gal-1-P

+ UDPGlcUA 3.27 4.20 0.047 4.75 0.805 4.63 .099

2.93

Cells were maintained for 24 hours in glucose-free medium or medium containing 2 mg glucose per milliliter and either no cycloheximide or 10 pg cycloheximide per ml. The cells were washed, allowed to take up galactose, extracted with ethanol and soluble material counted and chromatographed as described in MATERIALS AND METHODS.

-CYCLOHEXIMIDE

#ON-GL YCOL Y JIC

+ CYCLOHEX IM I DE

NON-GL YCOL Y TIC

F

X

GL YCOL Y J f C

I

I NUN-GL YCOL Y TIC

NON-GLYCOL rrfc

15

20

25

30

35

MIGRA T l O N D/STANC€ f e r n ) Fig. 5 Effect of cycloheximide on galactose metabolism in glycolytic and non-glycolytic cultures. Near confluent NIL cultures (upper paneIs) and confluent NILpy cultures (lower panels) were prepared for incubation for 24 hours at 37" in MEM or glucose free M E M as described in MATERIALS AND METHODS. Half of the cultures were incubated in the absence of cycloheximide (A and C) and the other half incubated in the culture media containing 10 pg cycloheximide per ml (B and D). Uptake and chromatography were as described in figure 1.

398

C . C H R I S T O P H E R , W . COLBY, D . ULLREY AND H . KALCKAR

hydrolyzed by these cells to produce hexose monophosphates and free nucleotides was considered and tested. It was reasoned that if phosphohydrolase activity was greater in non-glycolytic cells, it might help to explain the shifts in metabolic profiles that are brought on by sugar deprivation. Glycolytic and nonglycolytic cultures were washed and then frozen and thawed three times in order to lyse the cells. To the lysate thus produced was added PBS containing UDP(14C)galactoseand the mixture was incubated at 23" for one hour. Control cultures were washed (but not lysed) and incubated with UDP(14C)galactoseunder the same conditions. Contrary to the expected results, table 4 shows that the glycolytic NIL cells (having a low galactose uptake level and a high metabolic profile ratio) are the most active in splitting UDPGal to Gal-1-P. In addition, the transformed NILpy cells showed an overall lower phosphohydrolase activity than the untransformed NIL cells. Moreover, the non-glycolytic NILpy cells were about equal in hydrolytic activity to the sugar fed NILpy cells and these levels of activity were similar to the low levels of sugar starved untransformed cells. N o hydrolysis was evident from similarly treated whole cells (table

4) indicating that the incubation conditions in PBS did not cause the release of Gal-1-P from UDPGal and hydrolysis did not occur to any appreciable degree on the outside of the cells. Based on the amount of Gal-1-P proportional to the sum of UDPGal plus Gal-1-P on the chromatograms, in one hour only 20-25% of the available UDPGal was hydrolyzed by even the most active lysate. Incorporation of galactose a. Insoluble fraction A survey of incorporation of l4Cgalactose residues into the ethanol insoluble fraction was made over a 2-hour period. The overall levels of incorporation of label into glycolytic and nonglycolytic NIL cells was, in general, not a good reflection of their respective uptake properties. The untransformed glucosestarved or glucose-fed cells incorporated about the same amount of label while uptake was at least five times higher in the sugar starved cells than it was in the fed cells (figs. 6A, B). In contrast, incorporation by transformed (NILpy) cells more closely reflected uptake in that the incorporation by starved cells was higher than by fed cells at the end of two hours (figs. 6C, D). However, the difference was due to an initial delay of about 30 minutes in

TABLE 4

In ljitro hydrolysis of UDP galactose UDPhexose Cell line

Culture condition (+ Glucose)

Galactose uptake (nmolesimg)

0.226 1.162 .456 1.26

Gal-1-P + UDPGlcUA

UDPGalactose Hydrolyzed (nmoledmg)

area'

elution*

intact cells

cell lysate

8.15 0.019 1.11

10.3 0.039 1.34