Lennarz W.J., Lane M.D. (eds.) Encyclopedia of Biological Chemistry. Four-Volume Set . V. 4

Подождите немного. Документ загружается.

spatially restricted microdomains. During Ca

2þ

inﬂux-

dependent reﬁll of internal Ca

2þ

stores there is minimal

diffusion of Ca

2þ

in the subplasma membrane region

due to rapid uptake into the ER by the SERCA pump,

indicating close apposition between the ER and plasma

membrane at the site of Ca

2þ

inﬂux. SERCA, mitochon-

dria, and PMCA together determine the efﬁciency of

SOCC by removing Ca

2þ

from the vicinity of the

channel, thereby decreasing Ca

2þ

-dependent feedback

inhibition of the channel. PMCA, mitochondria, and

SERCA have been functionally localized to the micro-

domain where SOCE is occurring. Furthermore, PMCA

and SERCA have been shown to be immunoprecipitated

with TRPC channels. Thus, the architecture of such

2þ

signaling microdomains facilitates direct physical,

or functional, coupling between the molecular com-

ponents that are involved in regulating plasma mem-

brane SOCC. Current models support the idea that

SOCC and functionally associated proteins are strategi-

cally localized by the action of scaffolding proteins

which allow dynamic regulation of cellular function

(Figure 3).

A candidate for such a microdomain is caveolar lipid

raft domains which have been shown to be involved in

the regulation of SOCE in several cell types. These

domains are functionally and biochemically distinct

microdomains formed by the lateral packing of glyco-

sphingolipids and cholesterol within the membrane

bilayer. It has been proposed that arrangement of the

lipids and scaffolding proteins within LRD forms a

platform for the assembly of a number of proteins into

signaling complexes. This compartmentalization of the

signaling molecules can increase the rate of interactions

and enhance crosstalk networks. Importantly, key

protein and nonprotein molecules involved in the Ca

2þ

signaling cascade, such as PIP

q/11

, PMCA, several

TRPC proteins, and IP

R-like protein, and Ca

2þ

signaling events such as receptor-mediated turnover of

PIP

have been localized to caveolar microdomains in

the plasma membrane. Studies have also revealed that

(1) agonist-stimulated Ca

2þ

signal in endothelial cells

originates in speciﬁc areas of the PM that are enriched in

caveolin-1, and (2) intact lipid rafts are required for

activation of SOCE. Thus, it has been proposed that

caveolae might regulate the spatial organization of Ca

2þ

signaling by contributing to the localization of Ca

2þ

signaling complex as well as the site of Ca

2þ

entry.

It is now well recognized that Ca

2þ

signaling proteins

are assembled in multiprotein complexes. This ﬁnding is

again supported by studies carried out with TRP

proteins. It was earlier shown that the well-studied

TRP prototype, the Drosophila TRP, is assembled in a

signaling complex by the scaffolding action of INAD, a

multi-PDZ domain containing protein. Mammalian

TRPC channels are also assembled in multimeric

protein complexes that are associated with key Ca

2þ

signaling proteins such as G

q/11

,IP

R, PLC

, PMCA,

and SERCA as well as scaffolding proteins such as

FIGURE 3 Proposed calcium-signaling microdomain: Functionally associated proteins are localized in distinct structures and assembled together

by scaffolding proteins as well as cytoskeletal interactions. The architecture of the region enables the coupling between ER and PM that is involved

in activation of SOCE. Protein components are labeled in the ﬁgure.

STORE-OPERATED MEMBRANE CHANNELS: CALCIUM 121

caveolin-1, homer, and NHERF. Accessory proteins have

been identiﬁed not only for TRPC channels, but also for

PMCA, SERCA, and IP

Rs. These proteins encompass a

variety of functions such as trafﬁcking, phosphorylation,

dephosphorylation, scaffolding, etc. Thus, further infor-

mation regarding the protein components and assembly

of the mammalian Ca

2þ

-signaling complex will provide

a better understanding of SOCE. Such knowledge will

impact several important areas of physiology and

pathophysiology.

FURTHER READING

Berridge, M. J. (1995). Capacitative Ca

2þ

entry. Biochem. J. 312,

1–11.

Isshiki, M., and Anderson, R. G. W. (2003). Function of caveolae in

2þ

entry and Ca

2þ

-dependent signal transduction. Trafﬁc 4,

717–723.

Montell, C., Birnbaumer, L., and Flockerzi, V. (2002). The TRP

channels, a remarkable functional family. Cell 108, 595– 598.

Muallem, S., and Wilkie, T. M. (1999). G-protein dependent Ca

2þ

signaling complexes in polarized cells. Cell Ca

2þ

26, 173–180.

Parekh, A. B. (2003). Store-operated Ca

2þ

entry: Dynamic interplay

between endoplasmic reticulum, mitochondria, and plasma mem-

brane. J. Physiol. 547, 333–348.

Putney, J. W., Jr., Broad, L. M., Braun, F.-J., Lievremont, J.-P., and Bird,

G. J., St. (2001). Mechanisms of capacitative Ca

2þ

entry. J. Cell Sci.

114, 2223– 2229.

Venkatachalam, K., von Rossum, D. B., Patterson, R. L., Ma, H.-T.,

and Gill, D. L. (2002). The cellular and molecular basis of store-

operated Ca

2þ

inﬂux. Nat. Cell Biol. 4, 263– 272.

BIOGRAPHY

Indu Ambudkar is Chief of Secretary Physiology Section at the

National Institute of Dental and Craniofacial Research, NIH,

in Bethesda, Maryland. She holds both Masters and Ph.D. degrees in

Biochemistry from Lucknow and Madurai Kamaraj Universities in

India and received postdoctoral training at the University of Maryland,

School of Medicine. Her primary research interest is regulation of

cellular Ca

2þ

homeostasis and more recently store-operated Ca

2þ

inﬂux in salivary epithelial cells. Her work has contributed signiﬁ-

cantly to the understanding of SOCE, and the role of TRPC proteins

in this process.

122 STORE-OPERATED MEMBRANE CHANNELS: CALCIUM

Substrate Binding, Catalysis, and

Product Release

W. Wallace Cleland

University of Wisconsin, Madison, Wisconsin, USA

Enzymes catalyze reactions by forming complexes with their

substrate(s), accelerating the chemical reaction by lowering the

activation energy for the reaction, and then releasing

the product(s). The enzyme has to have sufﬁcient afﬁnity for

a substrate to form a complex with it at the physiological

concentration of this molecule. Conversely, however, too great

an afﬁnity for the substrate usually results in too great an

afﬁnity for the product, so that product release becomes rate

limiting. The maximum rate of an enzyme-catalyzed reaction

is limited, in fact, by the relationship between the equilibrium

constant and the kinetic constants called the Haldane

relationship, and one element of this restriction is the afﬁnity

of the substrate.

The Haldane Relationship

As noted above, the Haldane relationship allows

calculation of the maximum rate of an enzyme-catalyzed

reaction. For the simple conversion of substrate A into

product P, the initial rate of the forward reaction in the

absence of P is

¼ 2d½A



dt ¼ V

½A



ðK

þ½AÞ ½1

and that in the back reaction is

¼ 2d½P



dt ¼ V

½P



ðK

þ½PÞ ½2

where V

and V

are maximum velocities in forward and

reverse reaction, and K

and K

are Michaelis constants

for A and P (i.e., apparent dissociation constants in the

steady state).

The Haldane for this mechanism is the ratio of the

V=K values in forward and reverse directions:

¼½P



½A

¼ðV



ðV

¼ V



ðV

Þ½3

V=K; with units of reciprocal time, is the apparent ﬁrst-

order rate constant for reaction at low substrate

concentration, so eqn. [3] is analogous to the one for a

nonenzymatic reaction where K

would equal k

i.e., the ratio of rate constants in forward and reverse

directions.

To optimize the turnover number of an enzyme

) in this case, also known as k

cat

, with units of

reciprocal time, we can raise the two V/K values in

constant ratio until one V/KE

value (with units of

) reaches the limit set by diffusion. V/KE

(or k

cat

/K) is a second-order rate constant for productive

combination of enzyme and substrate and subsequent

reaction to give product, and the combination of enzyme

and substrate can only go as fast as these two can diffuse

together. This limit is , 10

for small sub-

strates, and somewhat less for larger ones like ATP.

Once the V

value in eqn. [3] is as high as it can

go, the only way to increase V

is to increase K

well. But K

cannot exceed the physiological level of A,

or the enzyme will not be at least half-saturated. Thus

has an upper limit, and enzymes that are part of

metabolic pathways are optimized to operate at this

limit. The actual turnover numbers vary, depending on

and the physiological level of the substrate, but

evolution has brought these enzymes to the limit set by

the Haldane relationship. On the other hand, enzymes

that are involved in control, as opposed to ones in

metabolic pathways, often sacriﬁce speed for control

characteristics, and do not have turnovers numbers at

the Haldane limit.

Catalysis

The catalytic process on an enzyme starts once the

substrate is bound. As noted above, the enzyme has to

have sufﬁcient afﬁnity to bind the substrate at its

physiological concentration. The enzyme then has to

lower the activation energy of the reaction sufﬁciently

for it to go at a rate approaching the maximum given by

the Haldane relationship. It is commonly said that the

enzyme accomplishes this by binding the transition state

structure more tightly than the ground state of the

substrate, but this is really a deﬁnition of catalysis.

Molecules resembling the transition state structure are

often bound much more tightly than the substrate (see

entry on transition state analogues). This is either

because the geometry of the transition state is a better

ﬁt, or includes inherently tighter interactions (stronger

hydrogen bonds, for example; see entry on low barrier

hydrogen bonds). Alternatively, the differential binding

results because the substrate is destabilized when it is

bound (sterically deformed, placed in an unfavorable

electrostatic environment, etc.), but this destabilization

is relieved in the transition state. Of course, destabiliza-

tion in one part of the substrate must be matched by

tight binding in another part, or the substrate will not

have sufﬁcient afﬁnity for the enzyme at its physiological

level. OMP decarboxylase is a classic example, where a

substrate analogue with –COO

replaced by – O

binds 10

-fold more tightly to the enzyme. The

difference is ascribed to charge repulsion between

the carboxyl group of the substrate and an aspartate

on the enzyme.

Free-Energy Proﬁles

The energetics of what happens during an enzymatic

reaction is illustrated by a free-energy proﬁle. Figure 1

shows the free-energy proﬁle at equilibrium for the

nonenzymatic reaction of A to P. The activation energy,

‡

, corresponds to the height of the barrier, and DG

for the reaction ð¼ 2RT ln K

Þ is the difference

between the levels marked A and P. For a corresponding

enzymatic reaction, one must decide what levels of A

and P to use. Figure 2 shows an equilibrium free-energy

proﬁle where the concentration of A is above its

dissociation constant, so that formation of EA has an

equilibrium constant greater than unity, but where P is

less than its dissociation constant, so that EP dis-

sociation is favorable. DG

‡

is given by the height of

the barrier above the level of EA, not E þ A. The

turnover number (V/E

or k

cat

) is determined by DG

‡

If one picks different levels of A or P, the difference

between E þ A and EA, or E þ P and EP, will differ as

shown in Figure 3. The dissociation level of A (K

)is

convenient, but then the level of P is determined by K

an equilibrium free-energy proﬁle is plotted, and will not

usually match K

, its dissociation constant. Free-energy

proﬁles may, of course, be plotted for nonequilibrium

concentrations of reactants, but in any case it is critical

to state the concentrations of reactions that are assumed

in plotting the proﬁle. When there are two or more

substrates or products involved in the reaction, one must

pick suitable levels of all reactants (and state them!) in

order to plot a free-energy proﬁle.

So far we have considered only simple free-

energy proﬁles with a single transition state connecting

FIGURE 1 Free-energy proﬁle at equilibrium for an uncatalyzed

reaction. DG

‡

is the activation energy and DG ¼ 2RT ln K

; the

energy difference between product and substrate.

FIGURE 2 Free-energy proﬁle at equilibrium for an enzymatic

reaction. The symbols have the same meaning as in Figure 1.

FIGURE 3 Free-energy proﬁle showing the difference in the levels of

(E þ A) or (E þ P) as the concentrations of A or P are changed. K

and

are the dissociation constants of A from EA and of P from EP.

124 SUBSTRATE BINDING, CATALYSIS, AND PRODUCT RELEASE

EA and EP. In practice, enzymes have different confor-

mations at the various stages of the catalytic cycle. Thus,

they have open forms where the reactants are free to

bind and dissociate, and closed forms in which catalysis

takes place. The chemistry may also take place in more

than one step, so that intermediates are present. Figure 4

shows a proﬁle where a conformation change converts

the open EA form to a closed, precatalytic EA

form.

This undergoes a catalytic reaction to give an EI

intermediate complex, which then is further converted

to the closed EP

form. A ﬁnal conformation change

gives the open EP form, from which P can dissociate.

Rate-Limiting Step for V/K

The two independent kinetic constants for an enzymatic

reaction are the maximum velocity, V, and the ratio of V

and the Michaelis constant, V/K. There is a separate V/K

for each substrate. Each of these parameters varies

separately with pH, ionic strength, temperature, and the

concentrations of other substrates, products, or inhibi-

tors. The Michaelis constant, K, while it measures the

apparent dissociation constant of the substrate in

the steady state, is not an independent constant, but

just the ratio of V and V/K. Thus when one speaks of

“rate-limiting steps” one must specify whether one is

referring to V or V/K.

The rate-limiting step for V/K is the one with the

highest barrier in the free-energy proﬁle. V/K involves

the combination of enzyme and substrate and reaction

through the ﬁrst irreversible step, which normally is

release of the ﬁrst product. Later steps that involve

further conformation changes in the enzyme and release

of other products do not affect V/K. Thus the slow

release of NADH from dehydrogenases, which often

limits the maximum velocities of these enzymes, has no

effect on V/K. As a result, the V/K value often is limited

more by the chemistry of the reaction than V,and

isotope effects on the chemistry are more fully expressed

(see entry on kinetic isotope effects). Except for some

slow mutants, full rate-limitation by the chemical

reaction is unusual, however, and the conformation

changes that precede and follow the chemical

reaction are usually partly rate-limiting. The equation

for an isotope effect on V/K is:

ðV=KÞ¼ð

k þ c



ð1 þ c

þ c

Þ½4

where x deﬁnes the isotope effect (D, T, 13, 15, 18 for

deuterium, tritium,

O) and the leading

superscript indicates the ratio of the parameters for light

and heavy isotopes.

k is the intrinsic isotope effect on

the chemical step, and

the equilibrium isotope

effect on the reaction. The constants c

and c

are

forward and reverse commitments to catalysis and are

the ratio of the rate constant for the isotope sensitive

step to the net rate constant for release from the enzyme

of either the substrate (for c

) or the ﬁrst product (for c

For the simple mechanism

E þ A O

EA O

EPQ

EQ þ P !

E þ P þ Q

½5

¼ðk

Þð1 þ k

Þ c

¼ðk

Þð1 þ k

Þ½6

The commitments thus consist of partition ratios of

intermediates and these ratios correspond to the

differences in barrier heights in a free-energy proﬁle. In

the proﬁle shown in Figure 4, c

and c

will be small if the

isotope effect measured involves reaction of EI to EP

because the barrier between EI and EP

is the highest

one. But if the isotope effect measured is on reaction of

to EI, c

will be small, but c

will be large. Thus one

will see largely the equilibrium isotope effect, as the EA

to EI step approaches equilibrium.

In the deﬁnition of c

in eqn. [5], k

is the internal

part of the commitment (c

f-in

) and k

/(k

) is the

external part (c

f-ex

). A sticky substrate is one where the

net rate constant for reaction of the initial collision

complex to give products is faster than the rate constant

for dissociation (k

). This ratio is called the stickiness

ratio (S

) and is given by S

¼ c

f-ex

=ð1 þ c

f-in

þ c

Þ in

terms of the parameters of eqn. [5].

Rate-Limiting Steps for V

The rate-limiting step for the maximum velocity depends

on the deﬁnition of rate limiting. One deﬁnition is the

“slowest” step, or the one with the highest individual

barrier in the forward direction. Another deﬁnition is the

“least-conductive” step, which is the one that sees the

highest total barrier to reach an irreversible step. A third

deﬁnition is the “most-sensitive” step. This is the one

FIGURE 4 Free-energy proﬁle showing conformation changes in the

EA and EP complexes, as well as an intermediate between the two

activated complexes. The levels of (E þ A) and (E þ P) are not shown.

SUBSTRATE BINDING, CATALYSIS, AND PRODUCT RELEASE 125

which causes the greatest percentage change in the

turnover number for a given percent change in

the forward rate constant. An isotope effect on this

step is more fully expressed than one on any other step.

The isotope effect on V for mechanism 3 is

V ¼ð

k þ c



ð1 þ c

þ c

Þ½7

where

; and c

have the same meaning as in eqn.

[5], but

¼½k

=ðk

þ k

Þ½1=k

þð1=k

Þð1 þ k

þ 1=k



½8

Note that c

; unlike c

, does not consist of partition

ratios. Rather k

, reduced by the factor k

=ðk

þ k

Þ; is

compared to k

, and to each of the net rate constants

after the chemical step. A low value of k

, for example,

can result in a large value of c

; and a greatly reduced

expression of the isotope effect on V.

The “slowest,” “least-conducting,” and “most-sensi-

tive” steps may be the same in an enzymatic reaction,

but need not be, so one must deﬁne what one means

when using the term “rate-limiting” for the maximum

velocity.

FURTHER READING

Cleland, W. W. (1982). An analysis of haldane relationships. Meth.

Enzymol. 87, 366–369.

Cleland, W. W. (1986). Enzyme kinetics as a tool for determination of

enzyme mechanisms. In Investigation of Rates and Mechanisms of

Reactions (C. F. Bernasconi, ed.) 4th edition, Part 1, pp. 791–870.

John Wiley & Sons, New York.

Cleland, W. W., and Northrop, D. B. (1999). Energetics of substrate

binding, catalysis, and product release. Meth. Enzymol. 308, 3–27.

BIOGRAPHY

W. Wallace Cleland is M. J. Johnson Professor of Biochemistry at the

University of Wisconsin and a co-director of the Institute for Enzyme

Research. His research interest is in the use of kinetic methods for

the determination of enzyme mechanisms, and in particular the use of

isotope effects. He has a Ph.D. from the University of Wisconsin and

was a postdoctoral Fellow at the University of Chicago. He is a

member of the National Academy of Sciences and the American

Academy of Arts and Sciences.

126 SUBSTRATE BINDING, CATALYSIS, AND PRODUCT RELEASE

Sugar Nucleotide Transporters

Carlos B. Hirschberg

Boston University School of Dental Medicine, Boston, Massachusetts, USA

Sugar nucleotide transporters are proteins in the membrane

of the Golgi apparatus. Their role is to translocate, from the

cytosol into the Golgi lumen, nucleotide substrates for the

glycosylation of proteins and lipids. Mutants in the above

transporters have biochemical and developmental pheno-

types, resulting in diseases such as leukocyte adhesion

deﬁciency II.

A Requirement for Sugar

Nucleotide Transport into

the Golgi Apparatus Lumen

Approximately half of all proteins in eukaryotic

cells are secreted or membrane-bound. These proteins

are processed through the secretory pathway where

80% undergo post-translational modiﬁcations such

as glycosylation, sulfation, and phosphorylation in

the lumen of the Golgi apparatus. Intact nucleotide

sugars, nucleotide–sulfate, and ATP, the substrates

for these reactions, enter the lumen of the Golgi

apparatus via speciﬁc transporter proteins (Figure 1).

These are multi-transmembrane spanning proteins

which function as antiporters with the corresponding

nucleoside monophosphates and thereby concentrate

the nucleotide derivatives in the Golgi lumen relative

to their concentration in the cytosol (Figure 1).

Evidence for this mechanism has been obtained

through biochemical studies with rat liver-derived

Golgi vesicles as well as with different species of

yeast. In the latter, gene disruption of a Golgi lumenal

GDPase, which generates GMP (the antiporter

molecule coupled to entry of GDP-mannose into the

yeast Golgi lumen), results in severe undermannosyla-

tion of proteins and lipids in vivo, reduced transport

of the GDP-mannose into Golgi vesicles in vitro,

as well as in a failure to make hyphae in Candida

albicans.

Mutants in Nucleotide Sugar

Transport—Biochemical

and Developmental Phenotypes

Including Diseases

Mutant mammalian cells grown in tissue culture have

been described as having 95% reduced transport of

speciﬁc nucleotide sugars into their Golgi apparatus,

while their proteins and lipids have a drastic reduction in

the speciﬁc sugar transported by the corresponding

nucleotide derivative transporters. Multicellular orga-

nisms such as C. elegans, Drosophila melanogaster,

plants, and humans that have mutations in the above

proteins also have distinct morphological phenotypes

affecting development of limbs, wings, brain, etc.

Studies of these mutants have shown that some of the

transporters can translocate more than one nucleotide

sugar, whereas other transporters retain high substrate

speciﬁcity such as differentiating between sugar epimers.

Mutants in nucleotide sugar transport have also been

described in Leishmania donovani and in yeasts such as

S. cerevisiae, K. lactis,andC. albicans.

Structure of Golgi Nucleotide

Sugar Transporters

So far the primary amino acid sequence, together with

the substrate speciﬁcity of approximately two dozen

Golgi membrane nucleotide sugar transporters, has

been determined. In most cases this was done by

correcting the phenotypes of mutant cells or mutant

multicellular organisms with libraries containing wild-

type DNA or cDNA from homologous or heterologous

organisms. In most cases after phenotypic correction,

the speciﬁc DNA (or cDNA) was expressed in yeast

or mammalian cells, Golgi vesicles were isolated, and

transport of different nucleotide sugars assayed

in vitro. Yeast is a particularly attractive system for

heterologous expression of nucleotide sugar transporter

genes, because it has few endogenous nucleotide sugar

transport activities. This approach led to the identiﬁ-

cation of multisubstrate nucleotide sugar transporters

from C. elegans. In some instances a particular cDNA

was also tested for its ability to correct the phenotype

of mutant cells where the substrate speciﬁcity of the

defective nucleotide sugar is known.

The fact that the above approaches for determining

the substrate speciﬁcity of particular transporters have

worked is based on the following points:

1. the transporters DNA (or cDNA) can express the

proteins in a heterologous system;

2. the targeting signals allowing the protein to

localize to the Golgi apparatus are functional in different

cell types; and

3. the targeted transporter is active once it reaches

the Golgi apparatus of the recipient cell and can

therefore affect phenotypic correction.

To date, the relationship of amino acid sequences and

substrate speciﬁcities of nucleotide sugar transporters

can be summarized as follows:

1. substrate speciﬁcities of nucleotides sugar trans-

porters cannot be inferred even when amino acid

sequences are , 50% identical;

2. only sequence identities of over 80% allow

accurate predictions of substrate speciﬁcity; and

3. transporters from different organisms that have

the same substrate speciﬁcity may have only 20%

overall amino acid sequence identity.

Due to the above reasons the reader is cautioned that

annotations about the identity (substrate speciﬁcity) of

nucleotide sugar transporters in databases should be

looked at very carefully!

Topography of Nucleotide

Sugar Transporters in

the Golgi Membrane

So far the only detailed topographic study of a

nucleotide sugar transporter has been done with that

for CMP-sialic acid in mammalian cells. Using a

combination of tagged epitopes and immunocyto-

chemistry, the transporter was found to have ten trans-

membrane domains and both its amino and carboxy

termini facing the cytosol. A study on the topography

of the Kluyveromyces lactis UDP-acetylglucosamine

transporter using membrane accessibility of a peptide

antibody and amino and carboxy termini epitope

tags also found that both termini face the cytosol

and that this transporter has either six or eight trans-

membrane domains. It is important to stress that none

of the existing algorithms in the literature are designed

speciﬁcally to address the issue of protein topography

in the Golgi apparatus membrane which is somewhat

thinner than the plasma membrane; most algorithms

are designed for predicting topography in the

plasma membrane.

FIGURE 1 Golgi membrane nucleotide sugar, PAPS, and ATP transporters have been detected in mammals, yeast, protozoa, nematodes, insects,

and plants. They are antiporters with the corresponding nucleoside monophosphate, except for PAPS, where the antiporter is not known, and for

ATP, where it is either AMP or ADP or both.

128 SUGAR NUCLEOTIDE TRANSPORTERS

Regulation of Macromolecular

Glycosylation by Nucleotide

Sugar Transporters

Studies with mutant mammalian cells in nucleotide

sugar transporters grown in tissue culture as well as with

ﬁbroblasts and lymphoblasts from patients with leuko-

cyte adhesion deﬁciency (LAD) syndrome II showed that

decreased concentrations of nucleotide sugar in the

Golgi lumen result in selective hypoglycosylation of

proteins rather than across the board reduction. Thus,

mutant MDCK cells which have a 95% reduction in

transport of UDP-galactose into their Golgi lumen have

a corresponding decrease in galactosylation of bulk

glycoproteins, glycolipids, and keratan sulfate but not in

heparan and chondroitin sulfates. LAD II patients

appear to have normal O-fucosylation of Notch protein,

whereas N-fucosylation of bulk proteins is drastically

reduced. Although the detailed mechanism for these

biochemical phenotypes is not understood, an attractive

hypothesis is that when supply of nucleotide sugars is

limited due to impairment in their transport, glycosyla-

tion reactions with lower K

s would proceed, whereas

those with higher K

s would not.