Bioscience Reports, Vol. 6, No. 7, 1986
Effectivity of Dolichyl Phosphates with
Different Chain Lengths as Acceptors of
Nucleotide Activated Sugars
P. L~w, ~ E. Peterson, 1 M. Mizuno 2, T. Takigawa 2,
T. Chojnacki 3 and G. Dallner 1'4'5
Received July 25, 1986
KEY WORDS: dolichyl-P; glycosyl transferases; microsomes.
Chemical synthesis of different S-forms of dolichyl-P was performed in order to
investigate the use of these polyprenes in mannosyl, glucosyl and glucosaminyl
transferase reactions. Determination of the Vmaxvalues for a series of dolichyl-P
demonstrated that the velocities of transferase reactions with all those dolichyl-P
derivatives present in animal tissues are largely the same. The apparent Km values for
the various dolichyl-P in the transferase system studied differed, but this property does
not appear to have physiological importance.
INTRODUCTION
Dolichol is present in all tissues, cells (with the exception of erythrocytes) and
organelles (1). In contrast, dolichyl phosphate has a more restricted distribution and is
enriched mainly in the endoplasmic reticulum. The reason for this localization is
obviously the well-defined function of dolichyl-P in the biosynthesis of Nglycosidically-linked oligosaccharide chains (2).
Since the main function of dolichyl-P is to act as an intermediate in glycosyl
transferase reactions, it is of great importance to establish to what extent the structure
of this polyprene influences its carrier function.
1 Department of Biochemistry, University of Stockholm, Stockholm, Sweden.
2 Central Research Laboratories, Kuraray Co., Ltd., Okayama, Japan.
3 Institute of Biochemistry and Biophysics, Polish Academy of Sciences.
4 Department of Pathology, Karolinska Institutet, Stockholm, Sweden.
5 To whom correspondence should be addressed.
677
0144-8463/86/0700-0677505.00/0 9 1986PlenumPublishingCorporation
678
Low er al.
The concentration ofdolichyl-P may be rate-limiting in a number ofglycosylation
processes, e.g. gastrulation in the sea urchin (5). A regulatory role is also supported by
other experimental findings: in inflammation-induced terpentine both the amount of
dolichyl-P and the level of protein glycosylation are increased (6); while in 2acetylaminofluorene-induced tumors (7), and in hepatic tissue during di(2ethylhexyl)phthalate treatment (8), both the amount of dolichyl-P and protein
glycosylation are decreased. In addition the presence of these polyprenes in
membranes may modify these structures and thereby affect the transfer of activated
sugar from one location to another. In model membranes dolichyl-P gives
destablization and increased phospholipid fluidity, properties which may elicit and/or
facilitate the transmembrane movement of glycosylated intermediates (3, 4).
In addition of its amount, the structure of dolichyl-P also modulates the glycosyl
transfer process. Fully unsaturated polyprenyl phosphates are much less effective as
glycosyl carriers than are the a-saturated forms (9, 10). In animal tissues, the three
isoprene residues at the co-end of the dolichyl chain are in the t r a n s configuration, while
the remaining residues are in the cis form. We do not at present know the significance
of this pattern, but available experimental findings seem to indicate that sugar transfer
is not influenced by such structural features (11). Recent studies demonstrated that the
optically active isomers of dolichol, the S- and R-forms, have different capacities as
sugar acceptors (12, 13). The S-form appears to be required in the lipid intermediateregulated biosynthesis of oligosaccharides.
In contrast to findings with ubiquinone, all pools of dolichyl-P analyzed to date
are mixtures of polyisoprenes containing different numbers of isoprene residues. There
are considerable species differences in this respect, but similar patterns are found in
different tissues of the same species (12). Under certain experimental conditions and in
primary human liver cancer the pattern of isoprenoid distribution is changed (8, 13),
but we do not yet know the functional importance of these findings.
Obviously, it is of great importance to determine the function of specific
isoprenoid patterns in dolichyl-P. It is possible that individual sugars are translocated
by defined forms of dolichyl-P with specific chains length. Recent developments have
made it possible to chemically synthesize various a-saturated dolichols exclusively in
S-form which allow investigation of the catalytic properties of these isoprenes as
acceptors in glucosyl transferase reactions (16). In the present study we have analyzed
these properties.
MATERIAL AND METHODS
Microsomes were prepared from livers of starved rats as described previously (17).
For measurement of glycosyl transferase activity the appropriate dolichyl phosphate
was dissolved in chloroform-methanol (2:1) and this solvent was then evaporated in
the presence of MgC12 and EDTA. The dried mixtures were dissolved in 0.5 ~o Triton
X-100. These samples were added to the incubation tube, which contained 30 mM
Tris-HC1, pH 7.8, 5 mM EDTA, 10 mM MgC12 and 2 mM ATP (final concentrations)
in a total volume of 100/A.
Microsomes were preincubated for 15 min at 0~ in Triton X-100 before addition
Dolichyl Phosphates as Sugar Acceptors
679
to the incubation mixture. Each tube received 0.03 mg (mannosyl transferase) or
0.12 mg (glucosaminyl and glucosyl transferases) microsomal protein. The final Triton
concentration was adjusted to 0.04 % in the case of GDP-mannosyl and UDP-glycosyl
transferases and, 0.18 % for the UDP-GlcNAc transferase. The incubation was started
by the addition of the nucleotide sugars: 0.25#Ci GDP-[14C]mannose
(290 mCi/mmol), 0.25 ~Ci UDP-[14C]GlcNAc (260 mCi/mmol) or 0.25/~Ci UDP[14C]glucose (296 mCi/mmol), all from Amersham.
After incubation at 30~ (3 rain for mannosyl and 10 rain for glucosaminyl and
glucosyl transferases) the reaction was stopped by addition of 1.0 ml chloroformmethanol (2:1) and 0.1 ml water. The mixture was extracted at 40~ for 30 rain. After
centrifugation the upper layer was discarded and the lower phase was washed with
theoretical upper phase by rinsing twice and mixing and centrifuging once. This
washing procedure was repeated three times.
For the synthesis of dolichols with 8, 11, 13, 18, 19 and 23 isoprene residues,
polyprenols with one fewer isoprene unit were isolated. Polyprenol 7 was prepared
from wood of Betula verrucosa, polyprenol 10 from leaves of Laurus nobilis, polyprenol
12 from leaves ofRhus typhina, polyprenols 17 and 18 from leaves of Gingko biloba and
polyprenola 22 from leaves of Sorbus suecica as described earlier (17). The acetate
forms were then used for addition of the S-form of a saturated isoprene unit using the
Gringard coupling reaction (16). Dolichols with 16 and 22 isoprene residues were
isolated from human liver. Phosphorylation of dolichol was performed according to
Danilov and Chojnacki (18).
Dolichyl phosphates were identified by HPLC on a C18 Resolve column
(Waters). The solvent systems used were: A. 2-propanol:methanol:water, 40: 60:5 and
B. hexane: 2-propanol, 70: 30, both containing 20 mM phosphoric acid. The dolichyl
phosphates were eluted with a 30-min gradient program using a flowrate of 1.3 ml/min
and progressing from 0 to 40% of system B.
RESULTS AND DISCUSSION
Two of the dolichyl-P used in these experiments (D 16-P and D 22-P) were
isolated from human liver, i.e. "natural", while the rest were synthesized chemically.
Comparison of some of the chemically-synthesized derivatives with those isolated from
liver using aH NMR, 13C NMR, IR and thin-layer chromatography did not reveal any
differences. The purity of the phosphorylated dolichols was also established by HPLC
(Fig. 1).
The V~,x and apparent K~ values of the glycosyl transferases of rat liver
microsomes specific for the substrates GDP-mannose, UDP-glucose and UDPGlcNAc for the different dolichyl phosphates were determined. Appropriate protein
amounts, incubation times and detergent concentrations were established in control
experiments.
For determination of Vma~values for the individual dolichyl-P in the various
glycosyl transferase systems, the Lineweaver-Burk plot was employed (Fig. 2). For all
forms of dolichyl-P the velocity of mannosyl transferase was about ten times higher
than that of glucosyl transferase, while the latter velocity was twice as great as that of
680
L~w et al.
r~
0o
r~
L
Fig. 1. High performance liquid chromatography of the dolichyl-P derivatives
used as sugar acceptors. Left to right: D8-P, D1 l-P, D13-P, D16-P, D18-P, D19-P,
D22-P and D23-P.
GlcNAc transferase. There were considerable differences in the Vm~ values for
individual dolichyl-P forms in the mannosyl transferase reaction. The lowest value was
obtained with the shortest acceptor dolichyl 8-P (D8-P), followed by a continuously
increasing Vm,xusing acceptors of increasing length. Thus, D 19-P gave a value six times
higher than D8-P, while a slight decrease was observed with D22-P and D23-P. In the
case of glycosyl transferase the picture is similar, but the difference between the Vm~x
values for the shortest and longest dolichol derivatives is much less, i.e. only two-fold.
With UDP-GlcNAc as substrate, the Vm,xvalue for D8-P and D13-P differed 4-fold,
but then remained at the same level for all the longer dolichyl-P.
Determination of the apparent K~ values revealed that GlcNAc transferase has a
higher affinity for the various dolichyl-P than do glucosyl or mannosyl transferase
(Fig. 3). Interestingly, the KmS for the shortest and longest dolichyl-P derivatives with
mannosyl transferase are similar, but the values for derivatives with intermediate
lengths are somewhat higher. The situation is very different for both glucosyl and
glucosaminyl transferases. In these cases there is a continuous increase in the apparent
Km values upon moving from the short to the long dolichyl-P, the maximal increase
being 4-6-fold.
Hypothetically, it is possible that these differences in V~x and apparent Km values
do not reflect different catalytic efficiencies or affinities of the transferases for the
different dolichyl-P derivatives, but can be explained as the result of the solubility
properties of these derivatives. Dolichyl-P with different chain lengths may form mixed
micelles in the presence of Triton X-100 which differ in size and form and,
consequently, exhibit varying interaction with the transferases. In order to test this
Do[ichyl Phosphates as Sugar Acceptors
681
B. alucose
A Hc~nnose
c
2
o~
o
0.02"
t3n
E
c
xe
1
-6
0.01-
E
c
J
<.#.
-/.,,',
8
1~
2'4
8
1~
2~
C.G[cNAc
002E
r
o
c_
0.01"
E
Number of isoprene units
Fig. 2. Vmax values for various dolichyl-P derivatives in glycosyl
transferase reactions. A. mannosyl, B. glucosy], and C, glucosaminyl
transferase reactions. All values are the means of 7-10 separate
experiments,
possibility the influence of Triton X-100 concentration on the three glycosyl
transferases using D11-, DI8- and D23-P as acceptors was determined (Fig. 4). The
pattern was very different for each transferase, involving both activation and
inhibition. On the other hand, no differences between different dolicyly-P in the case of
an individual transferase were detected. Thus, the differences in Vmax and K mvalues for
the dolichyl-P tested are most probably not caused by variations in micelle properties,
but rather reflect interaction of these derivatives with the enzyme proteins.
The results described here demonstrate that those forms of dolichyl-P which are
present in microsomal membranes give similar V~axvalues in the transferase systems
studied. This finding would seem to exclude the possibility that polyprenes with
specific chain lengths interact with specific nucleotide-activated sugars. The fact that
short dolichyl-P derivatives display lower velocities in transferase reactions may be
explained by their restricted mobility or other hindrance to their interaction with the
enzymes.
In contrast, the Km values are very much influenced by the acceptor used and at
present we have no explanation for this phenomenon. In model systems dolichols with
increasing chain lengths increase phospholipid fluidity (4), and it is thus also possible
682
L~Sw et aL
30-
A. Monnose
20
:~ 15-
8. Glucose
/
10
.//,
y,
~
2~
52
=Z,I"
7~4
8
16
24
Number of isoprene units
Fig. 3.
Apparent K m values for vario/as dolichyl-P derivatives in glycosyl
transferase reactions. See legend to Figure 2.
that the chain length of dolichyl-P in a micellar system also affects proteinphospholipid-dolichyl-P interaction. From a physiological point of view the relatively
large differences in the affinities of the transferase enzymes for these polyprenyl
phosphates has no deciding importance, since the microsomal content of dolichyl-P
exceeds the apparent K m values. Therefore, the Vmaxvalue is a more important factor in
the regulation of glycosylation of the lipid intermediate.
The question remains as to why a family of dolichyl-P of different chain lengths is
present in microsomes, a situation not existing for other isoprenoid compounds, for
example, mitochondrial ubiquinone (19). The increase in longer dolichols in rat liver
after administration of plasticizers and the increase of shorter polyprenes in human
primary liver cancer suggest the involvement of individual forms of dolichol and
dolichyl-P in pathological membrane and/or metabolic processes, which will be an
area for future study.
ACKNOWLEDGEMENT
This work was supported by the Swedish Medical Research Council. T.
Chojnacki was a visiting professor supported by the Swedish Medical Research
Council during part of the study.
Do]ichyl Phosphates as Sugar Acceptors
A PIQnnose
1284
2.~
I BG~ucose
683
18
?
o
•
E
c~_
~A
I
006 012
IL
]
[IGtCNAc
005
01
18
e,n
x
0'i
012
Triton,%
Fig. 4. Effect of the concentration of Triton X-100 on glycosyl transferase
activities. The substrates used were GDP-mannose (A), UDP glucose (B) and
UDP-GlcNAc (C).
REFERENCES
1. Dallner, G., and Hemming, F W. (1981 ). In: Mitochondria and Microsomes (C. P. Lee, G. L. Schatz, G.
Dallner, Eds.), Addison-Wesley, Reading, pp. 655 681.
2. Struck, D. K., and Lennarz, W. J. (1980). In: The Biochemistry of Glycoproteins and Proteoglycans
(W. J. Lennarz, Ed.), Plenum Press, New York, pp. 35 73.
3. McClosky, M. A., and Troy, F. A. (1980). Biochemistry 19:2061-2066.
4. Valterson, C., van Duyn, G., Verkleij, A. J., Chojnacki, T., de Kruijff, B., and Dallner, G. (1985). J. Biol.
Chem. 260:2742-2751.
5. Carson, D. D., Earles, B. J., and Lennarz, W. J. (1981). J. Biol. Chem. 256:11552-11557.
6. Coo/bear, T., and Mookerjea, S. (1981). J. Biol. Chem. 256:4529-4535.
7. Eggens, I., Eriksson, L. C., Chojnacki, T., and Dallner, G. (1984). Cancer Res. 44:799-805.
8. Edlund, C., Ganning, A. E., and Dallner, G. (1986). Chem.-Biol. Interactions 57:255-270.
9. Mankowski, T., Sasak, W., Janczura. E., and Chojnacki, T. (1977). Arch. Biochem. Biophys. 181:393401.
I0. Palamarczyk, G., Lehle, L, Mankowski, T., Chojnacki, T., a~d Tanner, W. (1980). Euro 3. Biochem.
t0l :531-540.
I 1. Mankowski, T., Sasak, W., and Chojnacki, T. (1975). Biochem. Biophys. Res. Commun. 65:1292-1297.
!2. Ch•jnacki, T., Pa•amarczyk• G.• Jank •wski, W.• Krajewska-R ych•ik• • Szk •pinska, A., and V•gtman,
T. (1984). Biochem. Biophys. Acta 793:187-192.
13. L6w, P., Peterson, E., Mizuno, M., Takigawa, T., Chojnacki, T., and Dallner, G. (t985). Biochem.
Biophys. Res. Cormnun. 130:460-466.
14. Hemming, F. W. (1983). Biosynthesis of Isoprenoid Compounds (J. W. Porter, and S. L. Spurgeon, Eds.),
John Willey, New York, pp. 305-354.
t5. Eggens, I., Chojnacki, T., Kenne, L., and Dallner, G. (1983). Biochem. Biophys. Acta 751:355-398.
16. Suzuki, S., Mori, F., Takigawa, T., Ibata, K., Nivorgawa, Y., Niskida, T., Mizuno, M., and Fanatra, Y.
(1983). Tetrahedron Lett. 24:5103-5106.
i7. Chojnacki, T., Jankowski, W., Mankowski, T., and Sasak, W. (1975). Anal. Biochem. 68:114-119.
!8. Danilov, L., and Chojnacki, T. (1981). FEBS Lett. 131:310-312.
19. Hemming, F. W. (1974). Biochemistry of Lipids fT. W. Goodwin, Ed.), Butterworlhs, London, Vol. 4:
pp. 39-98.