System A amino acid transport activity in membrane vesicles and reconstituted proteoliposomes


Material Information

System A amino acid transport activity in membrane vesicles and reconstituted proteoliposomes
Physical Description:
156 leaves : ill. ; 29 cm.
Schenerman, Mark Allen, 1959-
Publication Date:


Subjects / Keywords:
Amino Acid Activation   ( mesh )
Biological Transport   ( mesh )
Cell Membrane   ( mesh )
Biochemistry and Molecular Biology thesis Ph.D   ( mesh )
Dissertations, Academic -- Biochemistry and Molecular Biology -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 1986.
Bibliography: leaves 149-155.
Statement of Responsibility:
by Mark Allen Schenerman.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000560488
oclc - 17831258
notis - ACY6040
System ID:

This item is only available as the following downloads:

Full Text







This dissertation is dedicated to my parents, whose constant

love and support made this work possible.


I would like to thank my mentor, Dr. Michael S.

Kilberg, for his patience and understanding even in the

midst of my impatience. I would also like to thank Dr.

Efraim Racker for his kind advice and hospitality during my

visits to Ithaca. Finally, I would like to thank Mary

Handlogten, Donna Bracy, and Tom Chiles for their friendship

and assistance during the course of my studies.



LIST OF TABLES .....................................

LIST OF FIGURES ....................................

ABBREVIATIONS USED..................................





Materials and Methods......................
Discussion .. ............. ....... ....... .


Introduction........................... ....
Materials and Methods.......................
Results ....................................




B ENZYME ASSAYS...............................


















BIBLIOGRAPHY........................................ 149

BIOGRAPHICAL SKETCH.................................. 156


1-1 Characteristics of Na+-Dependent Neutral
Amino Acid Transport Systems ................... 3

2-1 Enzyme Activities in the Plasma Membrane-
Enriched Fraction ............................... 27

2-2 Alkali-Ion Specificity for Systems A and ASC
in Rat Liver Plasma Membrane Vesicles........... 38

2-3 Effect of Gramicidin or Monensin on AIB Uptake.. 51

2-4 Trans-Inhibition in Isolated Plasma Membrane
Vesicles....................................... 53

2-5 Glucagon Stimulation of System A in Rat
Hepatocytes is Retained in Isolated Plasma
Membrane Vesicles............................... 59

2-6 Enzyme Activities in Membranes from Control
and Glucagon-Treated Hepatocytes............... 61

3-1 Enzyme Marker Activities in EAT Cell Membranes.. 91

3-2 Reconstitution of System A Activity into
Proteoliposomes Following Detergent
Extraction of EAT Cell Membranes................ 105


2-1 Time-Dependent Uptake of AIB by Rat Liver
Plasma Membrane Vesicles...................... 30

2-2 MeAIB Inhibition of Na+-Dependent AIB
Uptake by Isolated Vesicles.................. 33

2-3 Alanine, Cysteine, Histidine and Glycine
Transport by Rat Liver Plasma Membrane
Vesicles..................................... 35

2-4 Effect of pH on Alanine Uptake in Membrane
Vesicles..................................... 42

2-5 Effect of the Extravesicular Osmolarity on
AIB Uptake.................................... 44

2-6 Effect of Incubation Temperature on the
Na -Dependent Uptake of AIB................... 47

2-7 Relation Between Membrane Protein
Concentration and AIB Uptake.................. 49

2-8 System A Activity in Plasma Membrane Vesicles
from Control or Glucagon-Injected Rats....... 57

2-9 Decay of System A Activity in Membrane
Vesicles Incubated at 40C or -70C........... 64

2-10 Decay of Systems A, N and ASC in Vesicles from
Glucagon-Treated and Normal Hepatocytes...... 67

2-11 Flow Chart of HepG2 Membrane Preparation...... 69

3-1 Flow Chart of the Detergent-Extraction and
Reconstitution of System A Transport Activity
Using the Freeze-Thaw Procedure............... 86

3-2 Titration of the Cholate to Protein Ratio for
Reconstitution of System A Amino Acid
Transport in EAT Cell Membranes.............. 94


3-3 Titration of the Lipid to Protein Ratio for
Reconstitution of System A Activity Using
EAT Cell Membranes ............................. 97

3-4 Determination of the Optimal Period of
Sonication for Reconstituted Proteoliposomes
from EAT Cell Membranes...................... 99

3-5 Determination of the Optimal Concentration of
C E for Extraction of EAT Cell
Mebbane Proteins Prior to Reconstitution.... 102

3-6 Temperature Stability of the Membrane Protein
Extract...................................... 108

3-7 Time Course of AIB Uptake into Proteoliposomes
in the Presence and Absence of Valinomycin... 111

3-8 Osmotic Sensitivity of the Reconstituted
Proteoliposomes.............................. 113

3-9 Measurement of the Intravesicular Volume of
the Reconstituted Proteoliposomes Using
3-0-Methyl-Glucose........................... 116

3-10 SDS-Polyacrylamide Gel Electrophoresis of
EAT Cell Membranes and Reconstituted
Proteoliposomes.............................. 122


AIB 2-aminoisobutyric acid

C12E9 polyoxyethylene-9-lauryl ether

cAMP adenosine 3':5'-cyclic monophosphate

CHAPS 3-[(3-cholamidopropyl)dimethyl-ammonio]-
CHX cycloheximide

CMC critical micellar concentration

EAT Ehrlich ascites tumor

EDTA ethylenediamine tetraacetic acid

EGTA ethyleneglycol-bis-(B-amino-ethyl ether)
N, N'-tetraacetic acid

FBS fetal bovine serum

GABA Y-aminobutyric acid

HEPES 4-(2-hydroxyethyl)-l-piperazineethane-
sulphonic acid

HepG2 human hepatoma cell line

MeAIB 2-(methylamino)-isobutyric acid

MEM Eagle's minimal essential medium

NADPH reduced nicotinic dinucleotide phosphate

NaKRB sodium-containing Kreb's-Ringer bicarbonate

NP-40 non-ionic industrial detergent

NRK-49F normal rat kidney cell line

octyl -













inorganic phosphate

phenylmethylsulfonyl fluoride

relative specific activity


standard deviation

sodium dodecyl sulfate


trichloroacetic acid

polyacrylamide gel electrophoresis

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy




May 1986

Chairman: Dr. Michael S. Kilberg
Major Department: Biochemistry and Molecular Biology

System A amino acid transport activity and its regula-

tion has been studied in plasma membrane vesicles isolated

from rat liver. Enzyme markers were used to show that the

membrane preparation was enriched in plasma membrane yet

showed minimal contamination from endoplasmic reticulum mem-

branes and no contamination by mitochondrial membranes. The

plasma membrane vesicles actively accumulated amino acids

through Systems A, ASC, N and Gly, all of which are depen-

dent on an imposed trans-membrane Na+ gradient. Amino

acid uptake was into an osmotically-sensitive space and was

inhibited by ionophores that collapse the transmembrane

Na+ gradient.

Regulation of System A-mediated transport activity was

studied by stimulating transport activity using various

effectors, then isolating membrane vesicles from those

cells. When plasma membrane vesicles were isolated from

either glucagon-treated or normal hepatocytes, it was dis-

covered that the membrane vesicles from cells exposed to the

hormone partially retained the stimulated transport activ-

ity. The lack of retention of all of the enhanced activity

in the membrane vesicles is explained, in part, by a rapid

loss of activity during storage at 40C (half-life = 13 h).

Neither protease inhibitors nor dithiothreitol blocked this


A procedure for reconstitution of transport activity

was developed using Ehrlich ascites tumor cell membranes.

The membrane proteins were solubilized in cholate and urea

which were exchanged for the non-ionic detergent polyoxy-

ethylene-9-lauryl ether (C12Eg) by dialysis. After

mixing with sonicated asolectin and cholate, the detergent-

extract was reconstituted by a freeze-thaw procedure. The

proteoliposomes were collected by centrifugation and tested

for System A activity by a rapid filtration assay. Reconsti-

tuted transport activity was determined to be optimal when

the cholate to protein ratio (w/w) was 3:1 and the lipid to

protein ratio (w/w) was 20:1. The development of this pro-

cedure for reconstitution will facilitate further studies

toward purifying the protein(s) associated with System A

because reconstitution can be used to monitor the increase

in specific activity of the carrier during purification.

The non-ionic nature of the detergent C12E9 will

also facilitate the separation of proteins in the detergent-

extract according to their charge.



Amino acids are pivotal contributors to many metabolic

pathways in cells including gluconeogenesis, glycolysis, and

the Kreb's cycle. For example, amino acids released from

muscle provide the carbon skeletons for de novo glucose

synthesis in the liver through the glucose-alanine cycle

(Felig, 1973). The supply of amino acid precursors for glu-

coneogenesis has long been considered to be a control point

for the entire pathway (Exton et al., 1970). It has been

shown more recently that the rate-limiting step in alanine

metabolism in rat hepatocytes is the transport of alanine

into the cells (Sips et al., 1980a). In this way, the intra-

cellular concentration of amino acids is tightly regulated

according to the nutritional needs of the cell.

Amino acids can be transported into cells through two

types of carrier-mediated pathways. Eukaryotic cell trans-

port can either be by active accumulation, that is, coupled

to a trans-membrane Na gradient (Na -dependent), or

by facilitated transport not driven by ion fluxes (Na -

independent). Neutral amino acid transport was first

observed as two distinct pathways in Ehrlich ascites tumor

(EAT) cells and the pathways were designated by Oxender and

Christensen (1963) as System A (Na+-dependent) and

System L (Na -independent). Since that time, numerous

other amino acid transport systems have been defined and

characterized in animal cells (Christensen, 1984). The

present work will focus on Na -dependent neutral amino

acid transport systems in isolated membranes derived from

rat hepatocytes and EAT cells.

At least five Na+-dependent neutral amino acid

transport systems have been described in animal cells. Some

of the characteristics of these systems are summarized in

Table 1-1. System A prefers amino acids having short,

polar, or linear sidechains such as alanine, serine, methio-

nine, and glycine. Two non-metabolizable amino acids have

proven to be particularly useful for characterizing System A

activity. The Na+-dependent uptake of 2-aminoisobutyric

acid (AIB) and its N-methylated derivative, 2-(methylamino)-

isobutyric acid (MeAIB), have proven to be highly specific

for System A activity, particularly in rat hepatocytes

(Kilberg et al., 1985a). System A is stereospecific for L-

amino acids and is strongly inhibited as the pH of the

medium is lowered (Kilberg et al., 1980). The alkali-ion

specificity of System A depends somewhat on the cell line

tested because Li is not acceptable as a substitute for

Na in rat hepatocytes (Edmondson et al., 1979), but

Li -for-Na substitution is well tolerated by the

activity in EAT cells (Christensen and Handlogten, 1977).

System A activity has been observed in all nucleated cells



E r
V)I ~

4 C

) a,
a E
>4 0

'.. 4 0 0
ca a


E a

44 4
434 44

.0 -
.0 .0

0 0
a a





a E


aa a


r J4->
0Q Q c

a3 <
0- Q

u >

a a) I 4 .

S* 0 m a *- >:- D C,-
S,- =- o + 0- *-0 00
EL 04- a_ w l Oa u4 4

fo.0 au-a a 4- aj a, ar 4a
X 0 3 4- 0 440 c 0 00
wv4 44404 I- -LJi- V l 4 U 0 IU

0 0
a c

a~ ar
>4 >4


h O

> L5

4 -

">4 (/I



i0 0 <
4- .0
*a I-

System ASC was also originally characterized in EAT

cells (Christensen et al., 1967). It was detected because

the System A-specific probe, MeAIB, did not completely inhi-

bit Na -dependent alanine, serine, or cysteine uptake.

System ASC prefers neutral amino acids with small, polar

sidechains with particular affinity for those containing

oxygen or sulfur atoms, such as serine, threonine, and cys-

teine. In rat hepatocytes, Na -dependent cysteine

uptake was observed to be totally insensitive to MeAIB inhi-

bition and cysteine is considered, therefore, to be a selec-

tive probe for hepatic System ASC activity (Kilberg et al.,

1979). System ASC is not as sensitive as System A to lower-

ing the pH of the medium and, in liver tissue, it readily

accepts Li -for-Na substitution. System ASC also

shows a high degree of stereospecificity for L-amino acids

and appears to be ubiquitous, as it has been found in all

eukaryotic cells so far tested (Shotwell et al., 1983).

System Gly is a Na -dependent glycine-specific

transport system first described in rabbit reticulocytes

(Winter and Christensen, 1965), but is also present in

pigeon erythrocytes, rat hepatocytes, and hepatoma cells.

Sodium-dependent glycine uptake is not restricted entirely

to System Gly in rat hepatocytes because MeAIB inhibits a

portion of the transport (Christensen and Handlogten, 1981).

System 3 is a Na -dependent transport system which

supports the uptake of B-amino acids. Taurine has served

as a model substrate for System B, but B-alanine can also

be used as a selective substrate. System B has been well

characterized in EAT cells (Christensen, 1964) and rat hepa-


System N was first described in rat hepatocytes as a

system which was specific for Na -dependent asparagine,

glutamine, and histidine uptake (Kilberg et al., 1980).

System A mediates a portion of the transport of these sub-

strates in rat hepatocytes, but this component can be elimin-

ated by measuring uptake in the presence of an excess of the

System A-specific probe, MeAIB. System N is highly sensi-

tive to decreasing the pH of the medium and also exhibits a

high degree of specificity for L-amino acids. System N has

also been shown to occur in fetal rat hepatocytes and the

rat hepatoma cell line H4-II-EC3 (Vadgama and Christensen,


Neutral amino acid transport activity is regulated by

three different processes: trans-effects, hormones, and sub-

strate availability. The transport systems affected by

these factors are indicated in Table 1-1. Combined, the

characteristics of specificity and regulation of each trans-

port system make it possible to uniquely identify a given

transport system operating in a particular cell line.

Regulation of transport activity by trans-effects is an

acceleration or deceleration of amino acid uptake caused by

an elevated concentration of substrates inside the cell. An

acceleration of the transport rate, or trans-stimulation, is

a common characteristic of System ASC. In other words, when

cells are pre-loaded with a System ASC substrate then washed

free of external substrate and tested for System ASC uptake,

the activity is increased (Gazzola et al., 1980). A deceler-

ation of the transport rate, or trans-inhibition, is a

common characteristic of System A. In this case, when cells

are loaded with a System A substrate, washed free of exter-

nal substrate, and tested for System A uptake, the activity

is lower (Kelley and Potter, 1978). Trans-inhibition of

System A is cycloheximide-insensitive indicating de novo

protein synthesis is not required (Kilberg et al., 1985b).

Regulation of transport activity by amino acid depriva-

tion, sometimes called adaptive regulation, has been

observed for Systems A and N. The increase in activity is

detected after a time lag of 1-2 h, is blocked by cyclohexi-

mide, and is kinetically-defined as an increase in Vmax

(Kilberg et al., 1985b). The implication is that the

increased transport activity caused by the response to amino

acid deprivation is due to a de novo synthesis and inser-

tion of new carriers into the plasma membrane. A lowering

of the Km of the carrier for Na+ or amino acid is not

consistent with the data collected from several different

laboratories (Guidotti et al., 1978; Shotwell et al., 1983).

A wide variety of hormones affect the activity of

System A. Other systems, such as System N, are also

affected by hormones but no transport systems other than

System A show such a large response after hormone treatment

(Kilberg, 1982; Shotwell et al., 1983; Kilberg et al.,

1985a). Some of the hormones which induce hepatic System A

activity include growth hormone, growth factors, glucocorti-

coids, catecholamines, glucagon, and insulin. The

stimulatory effects of most of the hormones tested were

kinetically-defined as an increase in Vmax (Shotwell et al.,

1983). Insulin- and glucagon-dependent stimulation of

System A is also blocked by cycloheximide, actinomycin, and

tunicamycin indicating that the continuous synthesis of RNA

and glycoprotein is required to express glucagon-stimulated

activity (Kilberg et al., 1985a). The increase in Vmax and

the dependence on de novo glycoprotein synthesis has led

to the proposal that hormonal induction of System A trans-

port results in the insertion of a greater number of active

carrier molecules in the plasma membrane.

The first 15-30 min after exposure of freshly isolated

hepatocytes to glucagon is characterized by a stimulation of

System A activity which is independent of de novo pro-

tein synthesis. After 30 min, cycloheximide eliminates any

further increase in transport activity (Edmondson and

Lumeng, 1980). There are several possible explanations for

the protein synthesis-independent portion of the hormone-

stimulated transport activity. One possibility is that glu-

cagon is causing a redistribution of the trans-membrane ion

gradients resulting in increased electrogenic transport.

Friedmann and Dambach (1980) showed that glucagon-treatment

of rat liver resulted in a hyperpolarization of the cells

and an increase in the membrane potential from 39.0 mV to

47.2 mV. Bradford et al. (1985) showed that the membrane

potential increased in isolated hepatocytes after treatment

with dibutyryl cAMP using a 35C1 uptake procedure to

measure the potential. The increase in potential was

maintained for at least 40 min and could be partly

responsible for the corresponding increase in alanine uptake

induced by glucagon during the first 15-30 min of exposure.

Presumably, a change in membrane potential would affect all

Na+-dependent electrogenic transport systems and this

has not been reported, implying that the change in membrane

potential is not the only factor contributing to the

increased System A activity.

Another possible explanation for the protein synthesis-

independent increase in System A activity after glucagon-

treatment would be migration of cryptic carriers to the

plasma membrane as has been described for the glucose

carrier (Simpson and Cushman, 1985). Convincing evidence

supporting this model has not as yet been provided and

awaits adequate methods of subcellular fractionation as well

as methods for testing transport activity in various mem-

brane fractions. Reconstitution of glucose transport was

used by Suzuki and Kono (1980) to show the insulin-

stimulated movement of glucose carriers from an intra-

cellular compartment to the plasma membrane. Similar

studies could be undertaken for System A once a reliable

reconstitution process has been developed.

Post-translational modifications could also be respon-

sible for regulating the protein synthesis-independent por-

tion of the hormone-stimulated activity. Nilsen-Hamilton

and Hamilton (1979) observed that treatment of 3T3-fibro-

blast membrane vesicles with cAMP resulted in increased AIB

uptake. However, when the vesicles were treated under

hypotonic conditions with cAMP-deoendent protein kinase, an

inhibition of AIB uptake was observed. As expected, the
32P incorporation was increased in many proteins after

the addition of protein kinase and no conclusion could be

reached correlating specific protein phosphorylation and the

System A carrier protein.

Amino acid and hexose transport have frequently been

observed to be increased in tumorigenic cells (Parnes and

Isselbacher, 1978). In particular, it has been reported

that the Na -dependent uptake of AIB, a specific probe

of System A activity in most cells, is increased in trans-

formed cells (Isselbacher, 1972; Foster and Pardee, 1969).

Increased rates of glycolysis and System A transport activ-

ity were detected when "normal cells" were transformed

either by transfection with specific oncogenes or by expo-

sure of the cells to the low molecular weight polypeptides

called "transforming growth factors" (TGF). Exposure of rat

fibroblasts (rat-l) or myc-transformed rat-1 cells to the

23 kDa polypeptide called TGF-B greatly increased the rate

of System A transport (Racker et al., 1985).

Stimulation of amino acid transport and glycolysis was also

observed when a normal rat kidney cell line (NRK-49F) was

exposed to TGF-B (Boerner et al., 1985). The increased

activity reported in the experiments where TGF-B was used

to transform the cells was blocked by cycloheximide. The

cycloheximide sensitivity of the transformation-stimulated

System A activity appeared to be analogous to the cyclohexi-

mide sensitivity detected during the hormone- and

starvation-induced increase in transport activity. These

results raise the possibility that the same stimulatory mech-

anism (i.e., increased number of carriers being inserted

into the plasma membrane) may be operating to execute the

hormone-, starvation-, and transformation-induced System A


The relationship between growth factors, which stimu-

late System A activity, and tumorigenesis is beginning to be

understood through several different observations reviewed

recently by Weinberg (1985). When the sequence of the

platelet-derived growth factor, which is known to stimulate

System A activity (Owen et al., 1982), was compared to the

sequence of the v-sis oncogene product, considerable homol-

ogy was observed (Doolittle et al., 1983). Epidermal growth

factor (EGF) is another peptide which has been reported to

stimulate amino acid transport activity in human fibroblasts

(Hollenberg and Cuatrecasas, 1975). The cDNA clone for the

epidermal growth factor receptor was recently sequenced and

has been found to be homologous to a large portion of the

v-erbB oncogene product (Ullrich et al., 1984). It would

seem that certain growth factors, which are known to stimu-

late System A activity or the receptors for those growth

factors, have homology to oncogene products. This may imply

some unknown link between transforming-ability and the abil-

ity to stimulate System A activity which still remains to be


Glucagon- or starvation-induced transport activity has

been observed to decay if the stimulated cells are exposed


to amino acids. In fact, certain amino acids, such as aspar-

agine, cause a protein synthesis-dependent repression of

activity. Other amino acids, such as AIB, cause a rapid

decline in transport activity which is not protein

synthesis-dependent and is probably attributable to trans-

inhibition (Handlogten et al., 1985). The decay process is

blocked by cycloheximide and actinomycin (Handlogten and

Kilberg, 1984) implying that de novo protein and RNA syn-

thesis are required for the decay to occur. Interestingly,

the decay process is also blocked by a-amanitin, an inhibi-

tor of RNA polymerase II at the concentrations used in the

study, but not by cordycepin or ara-A, inhibitors of poly(A)

polymerase (Handlogten et al., 1985). These results suggest

that a mRNA must be synthesized de novo for the decay

process to operate and that the mRNA has the unusual feature

of lacking a poly(A) extension. The major group of mRNA

molecules known to lack poly(A) code for the histone pro-

teins. The role of these proteins in controlling the decay

process is, at present, unclear. What process is actually

responsible for causing the loss in System A activity is

unknown also. Whether it is a protease, post-translational

modification, internalization, or some other process remains

to be elucidated.

The purpose of this study is to learn more about the

hormonal and adaptational regulation of System A, and to

learn about some of the molecular aspects of System A-asso-

ciated proteins. At present, we can only examine the regula-

tion of transport activity in whole cells by observing the

effects of inhibitors of protein and RNA synthesis on trans-

port activity; therefore, many questions remain as to the

exact mechanisms that exist for the regulation of System A

transport. For example, how does a hormone like glucagon

activate transport activity in a protein synthesis-indepen-

dent manner? What other evidence besides kinetics and

cycloheximide-sensitivity can be collected to support the

notion that hormone-induction, amino acid-deprivation, or

cellular transformation causes an increased number of System

A carrier molecules to be inserted into the plasma membrane?

How does the amino acid-dependent decay process relate to

the regulation of transport activity? How is transport regu-

lated by trans-effects?

The strategy we have used to begin our study of these

questions is to prepare isolated membrane vesicles which

mediate System A transport. These membrane vesicles can be

used in studies of trans-effects, hormonal, and adaptational

control to establish if these modes of regulation are associ-

ated with the plasma membrane. For example, the argument

that hormones, amino acid-deprivation, or transformation

causes an increased number of System A carriers to be inser-

ted into the plasma membrane would be strengthened if it

could be determined that transport activity was increased in

membrane vesicles prepared from those induced cells. Like-

wise, if trans-inhibition could be demonstrated in isolated

membrane vesicles, it would support the notion that trans-

inhibition is a membrane-associated phenomenon.

Having perfected this artificial system for assaying

the transport of amino acids, the next part of the strategy

is to reconstitute System A activity after extracting it

from its native membrane environment and replacing it into

artificial lipid bilayers. If this assay system could be

developed, it would then be possible to perform several puri-

fication steps on the membrane-extract and assay purifica-

tion by reconstitution. Purification and identification of

System A-associated proteins would be an important prelude

to the elucidation of the mechanism of amino acid transport

because studies could be done on the carrier in an isolated

state in an artificial membrane. The consequences of reveal-

ing the regulation and mechanism of System A transport activ-

ity would be medically relevant because its activity is

increased not only in tumor cells but also in hepatic cells

from diabetic animals.



Use of Membrane Vesicles for Study of Nutrient Transport

Membrane vesicles from various cell types and tissues

have proven to be extremely useful for studying nutrient

transport processes. The earliest studies of active trans-

port systems in membrane vesicles were reported using bacter-

ial membranes (Kaback, 1960). Transport systems for various

sugars and amino acids were characterized by using bacterial

vesicles prepared by osmotic lysis and testing uptake using

a rapid filtration system (Kaback, 1974). The development

of methods for preparation of vesicles from bacterial cells

led other groups to develop similar techniques for eukaryo-

tic cells.

Nutrient transport in membrane vesicles from eukaryotic

cells and tissues has been reviewed recently in several ex-

cellent articles (Lever, 1980; Murer and Kinne, 1980; Sachs

et al., 1980). Sodium-dependent transport systems for D-

hexoses, L-amino acids, bile acids, and various ions have

been demonstrated in vesicles from brush border as well as

basolateral membranes. Most transport systems tested show

similar specificities in membrane vesicles when compared to

intact cells as has been demonstrated for the Na -

dependent glucose transporter in intestinal brush border

vesicles (Kessler and Semenza, 1983).

Several criteria have been applied to membrane vesicles

to determine decisively if solute is accumulated in the

intravesicular space via a carrier-mediated process: 1)

exchange diffusion of the solute with an internal solute, 2)

temperature dependence of influx and efflux, 3) osmotic sen-

sitivity of the vesicles for accumulation of solute, 4) sat-

uration kinetics of the substrate and, 5) selective effects

on ion-dependent transport caused by ionophores.

Advantages of Studying Transport in Membrane Vesicles

There are several advantages to studying transport phe-

nomena in isolated membrane vesicles versus intact cells.

Membrane vesicles offer the possibility of examining trans-

port processes in the absence of intracellular metabolism.

Many nutrients, such as amino acids or glucose, are trans-

ported into a cell and then rapidly metabolized, complica-

ting transport studies considerably. Membrane vesicles also

allow one to introduce defined media to either side of the

membrane, thus facilitating carrier function studies from

the cis or trans side. Various electrochemical driving

forces can be introduced and the effects on the carrier

readily observed. Vesicles are, additionally, ideal systems

for monitoring electrochemical potential-sensitive and pH-

sensitive dyes. Membrane vesicles which show transport

activity are also often useful as starting material for

reconstitution studies.

Drawbacks to the use of membrane vesicles for transport

studies include the following: 1) the orientation of the

carrier is not always known and, 2) the proteins necessary

for activity may be partially or completely degraded during

the course of vesicle isolation, making transport studies

difficult. In many instances, the disadvantages of using

membrane vesicles are greatly outweighed by the amount of

information that can be obtained through their use.

Homogeneity of Plasma Membrane Vesicles

The hepatocyte is a polarized cell, that is, its plasma

membrane is separated into functional domains with special-

ized functions. The bile canalicular domain is the barrier

between the cell and the bile-collecting ducts. It is

through this membrane surface that bile acids produced in

the hepatocyte are secreted into the bile duct. The sinusoi-

dal or blood-domain is the border between the cell and the

blood-carrying vessels. It is through this membrane surface

that peptide hormones such as insulin or glucagon interact

with the hepatocyte. The final domain, referred to as the

contiguous surface, is the region of hepatocyte plasma mem-

brane which is found between two hepatocyte cells. It is in

this domain that cell-cell communication occurs through gap


The most commonly used technique for determining homoge-

neity in membrane vesicle preparations is through the use of

enzyme markers. The level of subcellular contamination can

also be quantitated using enzyme marker assays (Evans,

1980). Indicators of plasma membrane purity include 5'-

nucleotidase, Na+,K+-ATPase, alkaline phosphatase,

leucine aminopeptidase, and adenylate cyclase. Endoplasmic

reticulum contamination can be quantitated through the use

of glucose-6-phosphatase or NADPH:cytochrome c reductase

assays. Mitochondrial contamination can be determined using

either succinate:cytochrome c reductase as an indicator of

the inner mitochondrial membrane or monoamine oxidase as an

indicator of the outer mitochondrial membrane. Contamina-

tion by Golgi remnants can be quantitated using galactosyl

transferase and lysosomal contamination can be determined

using acid phosphatase. Nuclear contamination can be deter-

mined using DNA as a marker and cytoplasmic contamination

can be quantitated using latent lactate dehydrogenase activ-

ity. A qualitative determination of contamination by sub-

cellular organelles can also be performed using transmission

electron microscopy.

Enzyme markers can also be used to distinguish the func-

tional domains of the hepatocyte plasma membrane. The bile-

canalicular surface contains most of the activity for leu-

cine aminopeptidase and 5'-nucleotidase (Roman and Hubbard,

1983), whereas glucagon-activated adenylate cyclase is found

primarily on the blood-sinusoidal surface (Wisher and Evans,

1975) and Na+,K -ATPase is located primarily on the

contiguous and sinusoidal surfaces of the rat liver plasma

membrane (Poupon and Evans, 1979).

Isolation of Membrane Vesicles from Rat Liver

Several groups have succeeded in isolating plasma

membrane-enriched vesicles from rat liver (Neville, 1968;

Ray, 1970; Touster et al., 1970). One group in particular

has addressed the functional polarity of the hepatocyte and

has developed a technique for separating the canalicular por-

tion of the plasma membrane from the blood-sinusoidal and

contiguous membrane surfaces (Wisher and Evans, 1975). More

recently, membranes from the three separate domains have

been prepared to even greater homogeneity through the use of

sucrose step-gradients (Hubbard et al., 1983). The best

separation of plasma membrane domains reported so far used

rate-zonal centrifugation and resulted in a 64-fold enrich-

ment of canalicular membrane markers and a 34-fold enrich-

ment of basolateral (a mixture of sinusoidal and contiguous

membrane surfaces) membrane markers (Meier et al., 1984b)

over the activity detected in the homogenate. Centrifuga-

tion through a Percoll gradient has also proven useful in

purifying plasma membrane vesicles from rat liver (Prpic et

al., 1984).

Amino acid transport has been studied to a limited

extent in membrane vesicles prepared from rat liver. The

first report of Na -dependent amino acid transport

assayed alanine uptake after isolation of membranes by dis-

continuous sucrose gradient centrifugation (Van Amelsvoort

et al., 1978). Another group (Meier et al., 1984a), has

shown that about equal amounts of sodium-dependent amino

acid transport are found in the basolateral and the canalicu-

lar surfaces. Increased Na -dependent alanine transport

has been reported in rat liver vesicles after starvation of

the animals (Quinlan et al., 1982) or after treatment of

hepatocytes with dibutyryl cAMP (Samson and Fehlmann, 1982).

In our work, we have demonstrated that membrane vesi-

cles from rat liver can be prepared and these vesicles activ-

ely accumulate amino acids. The vesicles were also prepared

from glucagon-treated hepatocytes as well as amino acid-

starved hepatocytes in order to show that increased System A

transport observed in the intact cells was retained in the

membrane vesicles. Vesicles were also prepared from a human

hepatoma cell line (HepG2) and those vesicles retained

increased System A activity as is observed in intact cells.

Materials and Methods


The radio-labelled compounds used were [carboxyl-

14C] inulin, [methyl-3H1 2-aminoisobutyric acid (AIB),

ICN Pharmaceuticals; L-[ H] cystine, Schwarz/Mann; L-[2,5-

3H] histidine, [2-3 H glycine, L-[2,3-3H] alanine, and

3-O-methyl-D-[U14C] glucose, Amersham. Filters used for

transport assays were either Millipore type HAWP (0.45 pm)

or Gelman type GN-6 (0.45 pm). Highly purified glucagon

was a generous gift from Dr. Mary Root of Lilly Laborator-

ies. Fetal bovine serum (FBS) and Eagle's Minimal Essential

Medium (MEM) were obtained from Flow Laboratories. All


other chemicals were reagent grade or better and were

obtained from Sigma Chemical Company. Rats were from a col-

ony maintained by the University of Florida Animal Resources


Hepatocyte Isolation and Transport Assay

Hepatocytes were isolated from male Sprague-Dawley rats

(100-200 g) as described previously (Kilberg et al., 1983).

Usually, more than 90% of the cells were viable as deter-

mined by the trypan blue exclusion assay. Both the control

and the hormone-treated rats were fasted overnight prior to

cell isolation. The experimental animals were injected with

1 mg of glucagon per 100 g body weight 4 h before surgery.

A small portion of both control and hormone-treated hepato-

cytes was suspended in Na -containing Kreb's-Ringer bi-

carbonate buffer (NaKRB) containing 0.1 mM cycloheximide

(CHX) and placed in monolayer culture. Following a 2 h cul-

ture period, the activity of System A was determined by

assaying the Na -dependent transport of 50 pM AIB for 1

min at 370C as described by Kilberg et al. (1983). The

remaining cells were resuspended in 40 ml of ice-cold Buffer

A (0.25 M sucrose, 0.2 mM MgC12, 10 mM HEPES-KOH, pH 7.5)

for preparation of a plasma membrane-enriched subcellular


Rat Liver Plasma Membrane Isolation

Plasma membrane vesicles were prepared as described by

Van Amelsvoort et al. (1978). The liver of a 24 h-fasted

male Sprague-Dawley rat (150 to 200 g) was perfused with an

iso-osmotic homogenization buffer consisting of 0.25 M

sucrose, 0.2 mM MgCl2, 10 mM HEPES-KOH, pH 7.5 (Buffer

A). All subsequent procedures were carried out at 40C.

The blanched liver was removed and homogenized with 22

strokes using a Potter-Elvehjem homogenizer with a motor-

driven, loose-fitting teflon pestle. After the addition of

EDTA to a final concentration of 1 mM, the homogenate was

forced through a nylon screen (75 pm) and then centrifuged

at 1000xg for 10 min. The supernatant and the loose upper-

layer of the pellet were saved and the remaining solid

pellet was resuspended in Buffer A containing 1 mM EDTA.

This suspension was centrifuged again at l000xg for 10 min.

The resulting supernatant was collected as before, pooled

with the previous one, and then centrifuged at 20,000xg for

30 min. The loose upper-layer of the pellet was collected

and resuspended in Buffer A containing 1 mM EDTA by passing

the material through a 19 ga. needle six times. This mem-

brane fraction was purified further by placing it on a dis-

continuous sucrose gradient composed of 20%(w/v) and

39.5%(w/v) sucrose each containing 10 mM HEPES-KOH, pH 7.5.

The gradients were centrifuged at 50,000xg for 2.5 h. Vesi-

cles enriched in plasma membrane were collected from the

20%/39.5% sucrose interface and diluted 1:1 with 0.2 mM

MgC12, 10 mM HEPES-KOH, pH 7.5. These membranes were

then pelleted by centrifugation at 100,000xg for 40 min.

The resulting pellet was resuspended by vortexing in Buffer

A to a final concentration of approximately 10 mg pro-

tein/ml. The overall yield of the procedure was about 2-3

mg protein per g liver (wet weight). Vesicles could be

stored for up to one month at -700C with minimal loss of

transport activity.

Preparation of Membrane Vesicles from Cultured Hepatocytes

When membrane vesicles were prepared from cultured

cells following substrate starvation, the freshly isolated

hepatocytes were placed in 150 mm collagen-coated dishes

(Kilberg et al., 1983) in NaKRB (amino acid-free medium) or

NaKRB containing 20 mM asparagine (amino acid-supplemented

medium) at a density of 27 million viable cells per dish.

The cells were incubated at 370C in a humidified atmosphere

of 5% C02/95% air for 6 h and then each dish was rinsed

with 10 ml of phosphate buffered saline (154 mM NaCl, 10 mM

Na2HPO4, brought to pH 7.5 with HC1). The cells were

scraped into 5 ml of Buffer A and homogenized by 25 strokes

with the Potter-Elvehjem homogenizer with a tight-fitting

teflon pestle. Membrane vesicles were isolated as described

above. The total yield from 10 dishes of cultured cells was

approximately 3 mg of membrane protein.

Isolation of Membrane Vesicles from Human Hepatoma Cells

A human hepatoma cell line (HepG2) was grown to confluence

in fourteen 150 mm Petri dishes in MEM supplemented with 5%

FBS The dishes were rinsed twice with PBS and then

the cells were scraped into a total of 30 ml of Buffer A.

The suspension was homogenized using 10 strokes of a

tight-fitting Potter-Elvehjem homogenizer and the homogenate

was brought to 1 mM in EDTA. The homogenate was centrifuged

at 1000xg for 10 min. The supernatant (Sl) and the pellet

(PI) were both saved. The S1 fraction was centrifuged at

45,000xg for 30 min, and the resulting supernatant (S2) was

discarded. The corresponding pellet (P2) was resuspended in

1 ml of Buffer A and saved. Membranes contained in the

nuclear pellet were prepared using the initial pellet (P1)

which was placed on top of two sucrose step gradients

containing 39.5% sucrose as the lower layer and 20% sucrose

as the middle layer. The gradients were centrifuged at

50,000xg for 2.5 h. The white, fluffy material at 20%/39.5%

interface was removed and was diluted 1:1 with 0.2 mM

MgC12, 10 mM HEPES, pH 7.5. The suspension was

centrifuged at 100,000xg for 1 h to pellet the vesicles.

The final pellet (P3) was resuspended in 1 ml of Buffer A.

The protein content of the fractions was measured by a Lowry

assay. The total protein yield of P2 was approximately 7 mg

and that of P3 was approximately 3 mg.

Cells could also be grown in roller culture (2 L) in
HEPES-MEM, pH 7.45 containing 5% FBS. The cells were
removed by treating with 2 mM EDTA for 2 h at 370C then
scraping with a rubber policeman into 40 ml of PBS.

Vesicle Transport Assay

Just prior to use, the membrane vesicles were diluted

with Buffer A to a final concentration of 2.5 mg protein per

ml and then incubated at 220C for 15 min. To initiate

amino acid uptake, 20 pl (50 pg protein) of the vesicle

suspension was added to 20 pl of Buffer A supplemented with

10 mM MgC12, 120 mM of either NaSCN or KSCN, and 200 pM

radioactively-labelled amino acid. These two solutions will

be referred to as Na+- and K -uptake buffers, respective-

ly. Where indicated in the figure legends, 60 mM Na2SO4

or K2SO4 was used to replace the corresponding thiocyan-

ate salts. Uptake was terminated by the addition of 1 ml of

ice-cold Buffer A containing 100 mM NaC1 (stop-buffer). The

mixture was vortexed immediately and passed over a 0.45 pm

nitrocellulose filter. The filter was washed with another 3

ml of ice-cold stop-buffer and then analyzed for trapped

radioactivity in 5 ml of Bray's scintillation cocktail

(Bray, 1960). Unless otherwise indicated in the figure

legends, the results were from a single membrane preparation

and the S.D. of triplicate assays was less than 10% of the


Enzyme Marker Assays

The activities of 5'-nucleotidase (Morre, 1971), glu-

cose-6-phosphatase (Swanson, 1955), succinate:cytochrome c

reductase (Kilberg and Christensen, 1979), and cytochrome ox-

idase (Kilberg and Christensen, 1979) were assayed by

previously described methods. Inorganic phosphate was

determined by the method of Fiske and Subbarow (1925).

Fluoride-stimulated (10 mM NaF) adenylate cyclase activity

was measured by a modification of the procedure described by

Wisher and Evans (1975). The cAMP produced was detected by

a protein-binding assay supplied as a kit by Amersham Corp.

Tests for contamination of the final membrane fraction by

intracellular membranes showed a similar profile to that

obtained by Van Amelsvoort et al. (1978).

Determination of the Extravesicular Volume

The extravesicular volume of pelleted membrane vesicles
was determined using 14C-inulin. Before use, the

14C-inulin stock (1 mg inulin/ml; 7 pCi/ml) was fil-

tered through a Gelman filter (0.22 pm) to remove particu-

lates and the filtrate was counted for radioactivity.

Vesicles (2.48 mg/500 pl) were mixed with 400 pl of Buffer

A and 100 pl of 14C-inulin and the mixture was vor-

texed. After 10 pl was removed for determination of the

total radioactivity, the suspension was centrifuged at

100,000xg for 1 h and 100 pl of the supernatant was removed

for determination of the total radioactivity. The sides of

the centrifuge tube and the surface of the pellet were

washed 3 times with 3 ml aliquots of ice-cold Buffer A,

then the pellet was resuspended in 1 ml of Buffer A and 200

pl of the suspension was removed for determination of the

total radioactivity. The total dpm remaining in the pellet

after the washes was divided by the total dpm present in the

initial suspension before centrifugation. This value was

divided by the total mg of protein present in the suspension

to give 23.2 pl of extravesicular volume per mg protein for

the 100,000xg pellet.

Protein Determination

Vesicle protein was determined by a modification of the

method of Bensadoun and Weinstein (1976). Approximately

10-50 pg of membrane protein was suspended in 1 ml of 0.1%

sodium dodecyl sulfate (SDS). Following a 10 min incubation

at 220C, the protein was precipitated by adding 750 pl of

ice-cold 24%(w/v) trichloroacetic acid (TCA) and then

pelleted by centrifugation at 12,000xg for 20 min. The pro-

tein content of the pellet was measured by a modification of

the Lowry technique as described previously (Kilberg et al.,

1983). Bovine serum albumin (5 to 100 pg) was used as the



Enzyme Marker Activities in Rat Liver Membrane Vesicles

Enzyme marker analysis of the membrane vesicles pre-

pared from rat liver revealed that the vesicles are enriched

approximately 10-fold in the plasma membrane marker enzymes

5'-nucleotidase and adenylate cyclase (Table 2-1). A marker

enzyme for microsomal contamination, glucose-6-phosphatase,

showed 3-fold enrichment. There was a 10-fold reduction in

the level of the mitochondrial enzyme marker activities,

Enzyme Activities in the Plasma Membrane-Enriched Fraction

Membrane vesicles were tested for the presence of particular
enzyme markers for plasma membrane, endoplasmic reticulum, and
mitochondrial inner membrane. The activities of 5'- nucleotidase
and glucose-6-phosphatase are expressed in terms of pmol Pi
formed per mg protein per h. Adenylate cyclase activity is
expressed as pmol cAMP formed per mg protein per h. Succin-
ate:cytochrome c reductase and cytochrome c oxidase activities
are expressed as nmol cytochrome c reduced per mg protein per
min. Relative specific activity (R.S.A.) is determined by divi-
ding the specific activity of the enzyme in the plasma membrane-
enriched fraction by the specific activity in the homogenate.
The data are the averages + S.D. of triplicate determina-

Enzyme Activity Homogenate Vesicles R.S.A.

5'-nucleotidase 4.1 + 0.2 38.9 + 1.8 9.4

Adenylate cyclase 9.1 + 3.6 94.2 + 8.9 10.4

Glucose-6-phosphatase 10.2 + 1.5 31.9 + 1.6 3.1

Succinate:cytochrome c 14.0 + 0.31 1.32 + 0.09 0.1

Cytochrome c 13.2 + 0.35 1.01 + 0.18 0.1
oxi dase

succinate:cytochrome c reductase and cytochrome c oxidase

(Table 2-1). These enzyme marker activities are consistent

with those observed by Sips et al. (1980b).

Time-Dependent Uptake of AIB

Fig. 2-1 depicts the time-course of AIB transport by

isolated membrane vesicles. AIB uptake showed a Na -

dependent overshoot in the presence of a Na+ gradient

with maximal transport at 3 min. By 40 min, the accumula-

tion of AIB reached a steady-state level, presumably because

the Na+ gradient had been dissipated. The uptake in the

absence of Na+ was essentially hyperbolic in nature, and

reached a plateau after 15 min (Fig. 2-1).

The presence of a Na -dependent overshoot suggests
transport of the amino acid against a concentration gradi-

ent. To confirm this hypothesis, the intravesicular water

space was estimated. A value of 1.2 pl/mg protein was

determined by the 3-0-methyl-glucose method of Kletzien et

al. (1975). Using this value, it was calculated that the

steady state distribution ratio for AIB (AIBin/AIBout

in the absence of Na was slightly less than one.

Hence, an accumulated level of the amino acid greater than

240 pmol per mg protein represents transport against a con-

centration gradient.

Fig. 2-1. Time-Dependent Uptake of AIB by Rat Liver Plasma
Membrane Vesicles. Membrane vesicles (?.5 mg protein/mi)
were diluted into either Na () or K (*) uptake
buffer containing 200 pM radioactively-labelled AIB. After
incubation at 220C for the time indicated, an aliquot (50
pg) was removed and assayed for trapped radioactivity as
described in Methods section. The difference between the
transport in the presence of NaSCN or KSCN is shown as the
Na -dependent uptake (A). The data are the averages +
S.D. of three determinations.


& 320

E 240


1 80

0 10 20 30 40 50 60



MeAIB Inhibition of AIB Uptake

Fig. 2-2 illustrates the inhibition of Na+-dependent AIB

uptake by MeAIB, a System A-specific analog (Kilberg et al.,

1981). The inhibition was concentration dependent yielding

an apparent Ki of 0.6 + 0.2 mM. This value is in good

agreement with unpublished data from our laboratory indi-

cating that the Km for the high-affinity Na -dependent

component of MeAIB uptake by isolated hepatocytes is about

0.3 mM. Although an excess of MeAIB does not completely

block AIB uptake (Fig. 2-2), it is clear that most (> 85%)

of the Na -dependent AIB uptake occurs by the MeAIB-

inhibitable route (i.e., System A). These data support the

use of AIB as a selective substrate for System A in isolated

rat liver plasma membrane vesicles.

Sodium-Dependent Transport of Naturally-Occurring Amino

The uptake of four individual amino acids was measured

to obtain evidence for the presence of Systems A, ASC, N,

and Gly in the isolated membrane vesicles. The rate of

alanine transport in the presence of a Na+ gradient was

considerably greater than that in the absence of Na+ (Fig.

2-3A). The time-course of alanine uptake showed a rapid

Na -dependent overshoot, about 50% of which was not inhibi-

table by an excess of MeAIB. These data, in agreement with

results from intact hepatocytes (Kilberg et al., 1979) and

isolated liver membranes prepared in other laboratories

(Samson and Fehlmann, 1982; Sips et al., 1980b), indicate

Fig. 2-2. MeAIB Inhibition of Na -Dependent AIB Uptake
by Isolated Vesicles. Rat liver plasma membranes were
assayed for Na -dependent AIB transport in the presence of
the System A-specific probe, MeAIB, at the indicated
concentrations. The Na -dependent uptake of AIB was
measured as described in the Methods section. Computer
analysis of the data by a FORTRAN program as described by
Kilberg et al. (1983) yielded an estimated Ki of 0.6 +
0.2 mM for MeAIB. The results are reported as the averages
+ S.D. of assays in triplicate.

:200 ,


.120 app. Ki =0.6*0.2mM.

E so



0 5 10 15 20 25


Fig. 2-3. Alanine, Cysteine, Histidine and Glycine Trans-
port by Rat Liver Plasma Membranes. Membrane vesicles were
used to measure the transport of the indicated substrate.
The transport of alanine (A), cysteine (plus 1 mM dithio-
threitol) (B), and histidine (C) was tested in the presence
or absence of 5 mM MeAIB. Glycine (0) uptake was assayed in
the presence or absence of 5 mM threonine. The substrate
concentration in each case was 200 pM and transport was
measured at 220C. .The difference between uptake iq Na
( ) and that in K ( ) was taken as the total Na -
dependent transport (A). The Na -dependent component in
the presence of the inhibitor (MeAIB or threonine) is also
shown (V). See the text for interpretation of these data as
evidence for the specific transport systems indicated. The
results are presented as the averages + S.D. of three







S 1100




2 3 4 5 0 1 2 3 4


that the hepatic Na -dependent uptake of alanine, at a sub-

strate concentration of 200 pM, is about equally divided

between Systems A and ASC.

Additional evidence for the presence of System ASC was

obtained by examining cysteine transport. The Na+depen-

dent uptake rate for cysteine was considerably less than

that seen for alanine (e.g., 150 versus 450 pmol per mg pro-

tein per 15 s). As reported previously for intact hepato-

cytes (Kilberg et al., 1979), the Na+-dependent transport

of cysteine by the membrane vesicles was not inhibited by an

excess of MeAIB (Fig. 2-3B). These results demonstrate the

ability to monitor specifically hepatic System ASC in rat

liver vesicles by measuring Na -dependent cysteine


The Na+-dependent transport of glutamine and histi-

dine by intact hepatocytes is mediated to a large extent by

System N (Kilberg et al., 1980). In the absence of MeAIB,

histidine uptake by the vesicles (Fig. 2-3C) showed a rapid

Na+-dependent overshoot that decayed at a slow rate. When

Na -dependent histidine uptake was assayed in the presence

of an excess of MeAIB, conditions that provide a specific

test for System N activity (Kilberg et al., 1980), a rapid

overshoot was observed with an accumulation of more than 400

pmol per mg protein at 15 s. These results suggest the pre-

sence of System N activity in the isolated vesicles, but

also indicate that histidine transport in this membrane pre-

paration is not restricted entirely to System N.

Hepatic Na -dependent glycine transport is mediated

by Systems A, ASC, and Gly (Christensen and Handlogten,

1981). System Gly activity can be assayed selectively by

measuring Na -dependent glycine uptake in the presence of

an amino acid that can inhibit efficiently the other two sys-

tems. We have chosen threonine for this purpose because, in

cultured hepatocytes, this amino acid is transported effec-

tively by both Systems A and ASC (Kilberg et al., 1985a).

When Na -dependent glycine uptake was measured in the ves-

icles, approximately 40% escaped inhibition by threonine

(Fig. 2-30). Although these data show that a significant

portion of glycine uptake occurs by Systems A and ASC, they

also demonstrate that System Gly activity can be measured

readily in isolated liver plasma membrane vesicles.

Further Evidence for Heterogeneity of Alanine Transport

For isolated rat hepatocytes, transport by System ASC

is partially retained when Li+ replaces Na+ as the

extracellular alkali ion, whereas little or no System A-

mediated uptake occurs in the presence of lithium (Edmondson

et al., 1979). Such system selectivity for Li is just

the reverse in the Ehrlich ascites tumor cell (Christensen

and Handlogten, 1977). To determine whether this cell-spe-

cific property of System A was observed in isolated liver

membranes, Na and Li -dependent amino acid trans-

port was measured. Li+ substitution for Na+ caused

an 80% decrease in AIB uptake (Table 2-2). Lithium-

dependent transport depends to some degree on amino acid


Alkali-Ion Specificity for Systems A and ASC
in Rat Liver Plasma Membrane Vesicles

Membrane vesicles (2.5 mg protein/ml) were diluted into
Buffer A containing 10 mM MgC1l and one of the following:
120 mM NaC1, 120 mM LiC1, or 120 mM KC1. Incubation of the
vesicles at 22C in the presence of 200 pM substrate was
for 1 min (alanine) or 5 min (AIB). All other assay condi-
tions were the same as described in the text. The veloci-
ties are expressed in terms of pmol amino acid accumulated
per mg protein per unit time. In parentheses, the Li -
dependent data are shown as a percent of control values
determined in Na The results are reported as the aver-
ages + S.D. of three determinations.

Substrate Alkali ion Velocity Na or Li -
Tested dependent uptake

AIB Na+ 371 + 12 196 (100)

Li+ 217 + 7 42 (21)

K+ 175 + 16 -

Alanine Na+ 654 + 37 542 (100)

Li+ 526 + 57 414 (76)

K+ 112 + 10

Alanine Na+ 405 + 55 293 (100)

+ 5 mM MeAIB Li+ 438 + 42 326 (111)

K+ 112 + 10 -

structure (Christensen and Handlogten, 1977); therefore,

alanine was chosen as an additional test substrate.

Lithium-dependent alanine transport via System ASC has been

demonstrated for isolated rat hepatocytes (Kilberg et al.,

1981; Kilberg et al., 1979; Edmondson et al., 1979),

although Quinlan et al. (1982) reported that the rate of

Li -dependent alanine uptake by liver plasma membrane

vesicles was only 15% of the corresponding rate in Na.

Alanine uptake mediated by System A, estimated by subtrac-

ting the alkali-ion dependent velocity in the presence of

MeAIB from that seen in the absence of the inhibitor, was

249 and 88 pmol per mg protein per min in Na and Li ,

respectively (Table 2-2). In contrast, the rates of alanine

transport by System ASC, as monitored by MeAIB-insensitive

uptake, were the same in the presence of Li+ and Na+

(Table 2-2). These data are in agreement with those

obtained with intact hepatocytes and demonstrate that in iso-

lated vesicles System ASC accepts Li+ to a greater degree

than does System A. They also provide additional evidence

for the heterogeneity of alanine transport in the isolated

plasma membrane vesicles.

One of the established tests for distinguishing between

Systems A and ASC is the greater degree of inhibition of the

former system by increased H+ concentration in the uptake

buffer (LeCam and Freychet, 1977). To further demonstrate

the heterogeneity of Na+-dependent alanine transport in

isolated liver membrane vesicles, we assayed alanine uptake

in the presence or absence of MeAIB at pH values between 5.5

and 8.0 (Fig. 2-4). Decreasing the pH caused an inhibition

of System A-mediated Na+-dependent alanine transport by

about 67% over the pH range tested, whereas Na -

dependent alanine uptake via System ASC remained relatively

constant (Fig. 2-4).

Characterization of Amino Acid Uptake into Membrane Vesi-

An important criterion of carrier-mediated transport is

the stereospecificity of the process. When the Na -

dependent uptake of radioactively-labelled 200 pM L-alanine

was measured in the presence of 5 mM of unlabelled alanine,

the L-isomer caused a 54% inhibition of transport (control =

553 + 39, plus inhibitor = 256 + 12 pmol per mg pro-

tein per min), whereas the D-isomer caused only a 17% inhibi-

tion (control = 553 + 39, plus inhibitor = 457 + 36

pmol per mg protein per min). Kinetic analysis of L-alanine

transport is consistent with the observed level of inhibi-

tion by the L-isomer. Likewise, experiments testing inhibi-

tion of Na -dependent L-histidine uptake by the corre-

sponding L- or D-isomers showed that they inhibited trans-

port by 94% and 17%, respectively. These results demon-

strate the stereospecificity of amino acid transport by the

isolated membranes.

To determine if the plasma membrane vesicles were tight-

ly sealed, the intravesicular space was altered by changing

the osmolarity of the incubation medium. Fig. 2-5 illus-

trates that the uptake of AIB in sodium or in potassium was

Fig. 2-4. Effect of pH on Alanine Uptake in Membrane
Vesicles. Na -dependent alanine uptake was assayed in the
presence (V) or absence (A) of 5 mM MeAIB. The pH of the
uptake buffers was varied from 5.9 to 8.1. All other assay
conditions were the same as those described in the Methods
section. Data are presented as the averages + S.D. for
triplicate determinations of the Na -dependent transport.


e 330
9L A + ASC
E 220

110 -

5.5 6.1 6.7 7.3 7.9 8.5

Fig. 2-5. Effect of the Extravesicular Osmolarity on AIB
Uptake. Membrane vesicles were diluted into Na ( )
or K -( *) uptake buffer containing sucrose from 0.2 to
0.9 M and radioactively-labelled AIB (200 pM). Each of
these suspensions was divided into 3 tubes (50 pg each) and
incubated at 40C for 1 h. The accumulation of the amino
acid was assayed as described in the Methods section. The
data are the averages + S.D. of three determinations.



. 320
a 240

J 160 ,-

o 80 --

0 1 2 3 4 5

[sucrose] M"1

decreased when the concentration of sucrose in the incuba-

tion medium was increased. The inversely linear relation

between the sucrose concentration and the steady state amino

acid accumulation indicates that AIB is being transported

into an osmotically-sensitive compartment, presumably the

intravesicular space. Extrapolation of the data to infinite

sucrose concentration suggests that some leakage of sucrose

or some non-specific binding to the vesicles does occur

(Fig. 2-5).

The Na+-dependent uptake of AIB into vesicles was

dependent on temperature changes and protein concentration.

Sodium-dependent uptake of AIB after a 6 min incubation at

15C was only 78% of that at 22C, whereas the uptake at

4C was only 11% of the uptake observed at 220C (Fig.

2-6). To provide evidence that the uptake of amino acids

was mediated by a membrane-bound protein or protein-complex,

the effect of increasing the concentration of protein con-

tent in each assay was measured. Fig. 2-7 shows that,

between 10 and 200 pg of protein, the rate of AIB uptake

varied linearly with the amount of vesicle protein included

in the assay. This relation was observed for both total and

sodium-dependent transport.

Two lines of evidence support the conclusion that the

Na+-dependent uptake of amino acids by the membranes is

the result of an imposed trans-membrane Na+ gradient.

First, when the extravesicular Na concentration is

increased the rate of transport also increases. Secondly,

if the sodium gradient is dissipated by incubation of the

Fig. 2-6. Effect of Incubation Temperature on the
Na -Dependent Uptake of AIB. Membrane vesicles (2.5 mg
protein/ml) were incubated at either 220C (* ), 150C, ( )
or 4C (A) for 15 min prior to assay of System A activity
as described in the Methods section. The uptake was per-
formed as usual except that the uptake buffer was pre-
incubated at the indicated temperature and the uptake was
performed for the indicated period of time at the appropri-
ate temperature.

* 160

' 80

E 40

0 2 4 6

Fig. 2-7. Relation Between Membrane Protein Concentration
and AIB Uptake. Membrane vesicles were diluted into Na -
or K -uptake buffer containing 200 pM AIB. The amount
of membrane protein in the assay was varied from 10 to 200
pg. Uptake of the amino acid was measured for 6 min at
22C in Na containing buffer( 0). The Na -dependent
AIB uptake is also shown (A). The data are the averages
+ 5.0. of triplicate determinations. Where not shown,
The standard deviation bars are within the symbol.

50 -





0 40 80 120 160 200

ug vesicle protein

vesicles with a Na -selective ionophore such as gramici-

din or monensin (Pressman, 1976), the rate of Na -

dependent transport is decreased significantly (Table 2-3).

To demonstrate that the membrane vesicles retained

their native permeability with respect to anions, and to

show that the transport occurred by an electrogenic process,

Na+-dependent alanine uptake was measured in the presence

of different counter-anions. If Na -dependent uptake of

200 pM alanine in NaSCN-containing buffer was set equal to

100% (698 pmol per mg protein per min), the Na -dependent

transport rates in NaCl or Na2S04 were 86% and 73%,

respectively. The ability of an anion to cross a lipid

bilayer depends on the lipophilic nature of the anion; order-

ing the anions used from most lipophilic to least lipophilic

is as follows: thiocyanate > chloride > sulfate. Hence, the

order of effectiveness for these anions in increasing trans-

port activity parallels their relative lipophilic properties

and illustrates the electrogenic nature of the process (more

lipophilic anions permeate the membrane more rapidly causing

an increase in membrane potential).

Trans-Inhibition of System A in Membrane Vesicles

The activity of System A is decreased considerably when

the cytoplasmic concentration of its substrates is elevated

(Kelley and Potter, 1978). This phenomenon, referred to as

"trans-inhibition", is cycloheximide-insensitive and gener-

ally is thought to occur because the amino acids bind to the

carrier and "lock" it in the cytoplasmic orientation (White



Effect of Gramicidin or Monensin on AIB Uptake

Membrane vesicles (2.5 mg protein/mi) were incubated at
220C for 15 min with either 2% ethanol, 20 pg/ml gramici-
din in 2% ethanol, or 20 pg/ml monensin in 2% ethanol. The
suspensions were then diluted (1:1) into either Na or
K -uptake buffer containing 200 pM of radioactively-
labelled AIB. The transport was measured in the usual
manner as described in the text. The velocities are
expressed as pmol AIB per mg protein per 6 min and are the
averages + S.D. of three individual assays.

Additions Alkali ion Velocity Na+-dependence

Ethanol Na+ 232 + 8 118

Ethanol K+ 114 + 16

Gramicidin Na+ 146 + 3 29

Gramicidin K+ 117 + 13 ---

Monensin Na+ 130 + 5 8

Monensin K+ 122 + 6 ---

and Christensen, 1983; Kilberg et al., 1985a). An alternate

explanation for trans-inhibition is possible. In response

to increased cellular levels of System A substrates, the

carrier may be rapidly internalized by a process analogous

to the response observed for the adipocyte glucose carrier

after removal of insulin from the medium (Lienhard, 1983;

Simpson and Cushman, 1985). Likewise, if the elevated amino

acid levels were decreased, the carrier would be shuttled

back to the plasma membrane location resulting in a protein

synthesis-independent increase in transport. The latter

response, referred to as "release from trans-inhibition",

has been demonstrated with intact hepatocytes by several

laboratories. This "shuttle hypothesis" is made even more

plausible by recent evidence for a serum-dependent shuttling

of System A carriers in human fibroblasts (Gazzola et al.,


Isolated plasma membrane vesicles provide an excellent

model system to test whether trans-inhibition involves inter-

nalization of endocytic vesicles, and therefore requires in-

tact cells. If the phenomenon can be demonstrated in plasma

membrane vesicles, the results would argue that trans-

inhibition is, indeed, a membrane-associated response and

not dependent on extensive cellular machinery or architec-

ture. The results of Table 2-4 show that vesicles previ-

ously loaded with MeAIB exhibited a 26% reduction in measur-

able System A activity. When 0-glutamine was used, an amino

acid with little or no saturable transport in hepatocytes,

no decrease in activity was observed. These data indicate


Trans-Inhibition in Isolated Plasma Membrane Vesicles

Membrane vesicles (5 mg of protein) were incubated at 4C
for 2 h in Buffer A containing either no additions (con-
trol), 25 mM D-glutamine, or 25 mM MeAIB. After the incuba-
tion period, 0.5 ml of ice-cold Buffer A was added to each
condition. The samples were immediately vortexed and centri-
fuged in a microcentrifuge (15,000xg) for 5 min at 40C.
The supernatants were removed and 1 ml of ice-cold Buffer A
was added back without disturbing the pellet. The tubes
were centrifuged again in the microcentrifuge at 40C for 2
min. The supernatants were removed and the pellets were
resuspended in 200 pl of ice-cold Buffer A. The uptake of
200 pM MeAIB for 1 min at 220C was immediately tested as
described in the Methods section. The velocities are
expressed as pmol MeAIB per mg protein per min and are repor-
ted as the averages + S.D. of triplicate determinations.
The values in parentheses are the percent of control for
the Na -dependent velocity. Theoretical calculations
indicate that leakage of MeAIB into the extravesicular space
could not account for the degree of inhibition observed.
Actual determinations of the extravesicular concentration of
MeAIB were not performed and the MeAIB trapped in the extra-
vesicular space may contribute to the inhibition observed.

Loading Alkali Total Na+-dependent
Condition Ion velocity velocity

Control Na+ 173 + 4 136 (100)

K+ 37 + 3

25 mM D-glutamine Na+ 166 + 14 136 (100)

K+ 30 + 8 --

25 mM MeAIB Na+ 130 + 9 101 (74)*

K+ 29 + 7 ---

* p < 0.025

that trans-inhibition can occur in isolated membrane

vesicles and probably does not depend on internalization of

the System A carrier as is observed for the down-regulation

of the insulin receptor. It is important to note that these

data could also be explained by competitive inhibition of

AIB uptake caused by MeAIB trapped in the extravesicular

volume. The data should be regarded with caution until the

concentration of MeAIB in the extravesicular volume is

actually determined.

Effect of Glucagon-Treatment in vivo and Amino Acid-
Starvation in vitro on System A Activity in Plasma Membrane

It is well documented that glucagon-treatment, either

in vivo (Handlogten and Kilberg, 1984) or in vitro

(LeCam and Freychet, 1976; Pariza et al., 1976), causes a

protein synthesis-dependent increase in hepatic System A

transport activity. Our laboratory has provided evidence

that the hormone-induced molecule responsible for the stim-

ulation of System A activity is a glycoprotein (Barber et

al., 1983). The stimulation by glucagon is generally consid-

ered to be the result of de novo synthesis of carriers or

carrier-associated molecules and their subsequent insertion

into the plasma membrane. To determine whether the System

A-associated glycoprotein induced by glucagon is also

located in the plasma membrane, the System A activity in

membrane vesicles from hepatocytes taken from either normal

or glucagon-treated rats was first tested using thiocyanate

as the counter-anion in the uptake buffer, but the

Na+-dependent overshoot was too rapid to measure. In

order to demonstrate the peak of Na -dependent AIB uptake,

sulfate was used to replace thiocyanate as the counter-anion

in the uptake buffers. The lower lipophilicity of the

sulfate anion slowed the Na -dependent uptake so that the

peak of the Na+-dependent overshoot could be measured


Fig. 2-8 illustrates the time course for the Na -

dependent uptake of AIB in normal or glucagon-treated vesi-

cles*. It is evident that the Na+-dependent AIB

uptake in the glucagon-stimulated vesicles is increased at

least 2-fold over the normal vesicles. The System A activ-

ity peaks at approximately 3 min in the control membranes,

whereas the maximal uptake occurs at only 1 min in the

glucagon-treated vesicles. The higher rate of decline of

the Na -dependence in the glucagon-treated vesicles

compared to the normal vesicles is probably due to either a

more rapid dissipation of the trans-membrane Na+ gradi-

ent or a Faster equilibration of AIB across the membrane.

All subsequent assays of AIB transport that were designed to

make comparisons between control and hormone-treated vesi-

cles were performed for 15 s in sulfate-containing buffers.

These data indicate that the glucagon-induced activity is

retained in the membrane vesicles isolated from those same

cel s.

For brevity, the vesicles prepared from the hepato-
cytes that were isolated from the glucagon-injected rats
will be referred to as "glucagon-treated vesicles".

Fig. 2-8. System A Activity in Plasma Membrane Vesicles
from Control or Glucagon-Injected Rats. Membrane vesicles
were isolated from control rats (A) or rats injected with 1
mg glucagon/100 g body weight 4 h prior to membrane isola-
tion ( V). Following the isolation, the vesicles were
immediately tested for Na -dependent AIB transport as
described in the section on Materials and Methods using
either 60 mM Na SO4 or K SO4 uptake buffer.
The time of uptake at 22 C was varied from 10 s to 30 min.
The results are expressed as the averages of triplicate
determinations and the standard deviations, omitted for
clarity, were generally less than 10%.

.5 300

2 250

a 200
A 150

o 100



The System A activity in primary cultures of rat hepato-

cytes is also induced if the cells are incubated in an amino

acid-free medium (Kelley and Potter, 1978). Like glucagon-

dependent stimulation, this process, referred to as adaptive

regulation, is thought to result from increased synthesis of

a System A-associated glycoprotein located in the plasma mem-

brane (Barber et al., 1983). To test this hypothesis, cul-

tured hepatocytes were incubated for 6 h in amino acid-free

medium (NaKRB) or NaKRB supplemented with 20 mM asparagine

and then membrane vesicles were prepared from those cells.

The System A activity was enhanced nearly 6-fold in the vesi-

cles from the starved cells. The rate of Na -dependent

AIB uptake in the vesicles from starved (no amino acid) and

fed (20 mM asparagine) cells was 189 + 14 and 33 + 6

pmol per mg protein per min, respectively. The degree of

induction for System A transport activity measured in intact

cells is generally 5- to 10-fold.

After several different membrane preparations had been

tested for the level of glucagon-dependent stimulation of

Na -dependent AIS transport, it appeared as though the

amount of hormone-induced activity for intact cells was not

reflected in the membrane vesicles. To monitor the degree

of hormone stimulation, hepatocytes from glucagon-injected

rats were assayed for System A activity and then plasma mem-

brane vesicles were isolated from the same preparation of

cells. The data shown in Table 2-5 are representative of

many experiments. In this instance, the intact cells showed

a 30-fold increase in System A transport following glucagon

c E 0 +u =
SIU C- 4-

S- L 0 > 0 *04-

.> /) ( >L > 0

+ I- O ) >C 0

A- .0 in f Q .0
CC0 a L L O

i ITto< ONC:,
o C- m CS m C

i I,- *-

o-'1 E 0 r- L -

w Iom ouo
0S >CL L.C ( CL C

Cl V C O Ea -C **
SCv S0 ) C 3 -

MC W v

0 .E* *- E (0 0- 4 0
E CO C 4- +' 4)
4- E U CO er ,
CL 0 0U C L O4 VI C-
0I L3 CM 0 ) LO U

3 00 C 0 J O 0

C 4L" L 0 C C C 0
w- O U U C C U
O UO CO0 4

= QC'o = 0CC

C L C ( +O o0
*- 3 Oy C-r-- (D ^ "I O

S- oa o E c o 0) C
C E L. S E- S C

(0 **- Q.r- L. **~ C ET0 .) 40
4- 4- l.. OT 0 0 C t0
C j c 1) 4-' C' c

o U I C- Ca 3 o

f 4-) 4.4 +/1 l fl' 4
C J t 43 C c '- )4 o

S r -*r- Lcon o 4,

+U > (J T3 u
C/ < CC *~ *C- C .
+ CE C1)0 0 1.. 3 -
f0 4 L. U *C .C to

oCE*-r +> */ cu>
.C '4- CI 4-)Ct)3 30)






CO >







a)1 V
C: 01

)- .0 ,-
4- EV
=oM oE :
- U

treatment, whereas the membrane vesicles showed a 13-fold


The apparent incomplete retention of stimulated System

A activity in the membrane vesicles compared to the intact

cells may be due to a difference in the composition of the

control and glucagon-treated vesicles with respect to the

three domains known to exist on the hepatocyte surface

(Wisher and Evans, 1975; Hubbard et al., 1983). It has been

reported that Na+-dependent alanine transport activity is

about equally distributed on the cell surface (Van

Amelsvoort et al., 1980; Meier et al., 1984b), but these

studies were not performed on hormone-induced membranes.

We assayed plasma membrane marker enzymes to determine

if the amount of canalicular and basolateral membrane in the

control and glucagon-treated vesicles were similar. The

control vesicles showed an enrichment of 6-fold for 5'-

nucleotidase canalicularr) and 2-fold for fluoride-stimula-

ted adenylate cyclase (basolateral) when compared to the

homogenate, whereas the vesicles from glucagon-treated rats

showed a 9-fold enrichment of 5'-nucleotidase activity and a

5-fold enrichment of fluoride-stimulated adenylate cyclase

(Table 2-6). These data indicate that there are differences

in the absolute content of canalicular and basolateral mem-

brane between the two vesicle populations, but the glucagon-

treated vesicles actually show a higher degree of enrichment

for both markers. The ratio of the 5'-nucleotidase activity

to the fluoride-stimulated adenylate cyclase activity

appears to be similar for both preparations (control = 2.5,


Enzyme Activities in Membranes from Control and
Glucagon-Treated Hepatocytes
Membrane protein (100 pg) was analyzed for 5'-nucleotidase
and adenylate cyclase activity as described in the Methods
section. Values for 5'-nucleotidase activity are expressed
as pmol P formed per mg protein per h, whereas those
for adenylate cyclase activity are expressed as nmol cAMP
formed per mg protein per h. Adenylate cyclase was stimula-
ted using 10 mM NaF. The non-stimulated level of adenylate
cyclase ranged from 0.02-0.05 and 0.13-0.29 in the homogen-
ate and membrane vesicles, respectively, and has been sub-
tracted from the results to yield the data shown. The
results are given as the averages of three determinations
and the standard deviations were typically less than 15%.

Donor Rat Homogenate Membrane Enrichment
Enzyme Tested Vesicles


5'-nucleotidase 2.10 12.0 5.7

adenylate cyclase 0.26 0.60 2.3


5'-nucleotidase 1.90 17.3 9.1

adenylate cyclase 0.22 1.11 5.1


glucagon-treated = 1.8). Hence, a difference in the compo-

sition with respect to plasma membrane domains is probably

not responsible for the decreased glucagon-dependent stimula-

tion reported in Table 2-6.

Inactivation of Glucagon-Stimulated System A Activity

The decreased retention of stimulated System A activity

in membrane vesicles might also be explained if the System A

carrier complex is irreversibly inactivated during the iso-

lation procedure. In order to test for inactivation,

freshly isolated membrane vesicles from normal or glucagon-

treated rats were incubated at 4*C for 48 h. At specific

intervals, the System A activity was assayed in both vesicle

preparations. The results indicate that at 40C there is a

time-dependent loss of System A activity in the glucagon-

treated vesicles (Fig. 2-9). The decay of the hormone-

stimulated transport consisted of a single exponential compo-

nent. The preparation of the membranes requires about 6 h

and we assume that the decay was ongoing during that time.

The value for the half-life of the decay was approximately

13 h. In several studies, maintenance of the glucagon-

stimulated vesicles at -70C for 48 h protected most, if

not all, of the initial activity (Fig. 2-9). In contrast,

the basal rate of System A-mediated uptake seen in the con-

trol vesicles did not change during a 24 h incubation at

4C (Fig. 2-9).

A mixture of 2 mM PMSF, 2 mM EDTA, trasylol (30 trypsin

inhibitor units/ml), 0.1 mM leupeptin and 0.5 mM bacitracin

Fig. 2-9. Decay of System A Activity in Membrane Vesicles
Incubated at 40C or -70C. Isolated membrane vesicles
either from glucagon-treated (*,V) or control rats(A) were
incubated at 40C (AV) or -70 ( ) for the indicated
times. Vesicles stored at -70C were in aliquots and were
thawed only once prior to assay. At the indicated times,
Na -dependent AIB uptake was measured as described in
Materials and Methods using either 60 mM Na SO or
K SO uptake buffer. The results are expressed as a
percent of the initial Na -dependent uptake in the
glucagon-treated vesicles (423 + 42 pmol AIB accumulated
per mg protein per 15 s) and are averages of triplicate
determinations. Standard deviations are typically less than
20% of the mean.


u 75

e 50


0 10 20 30 40

added to the vesicle suspension did not slow the decay of

System A activity in the membrane vesicles maintained at

4C. Dithiothreitol (1 mM) also did not have any effect on

the decay of transport activity. Similar results were

obtained if the vesicles were subjected to a freeze-thaw

cycle in the presence of the inhibitors to trap them inside

the intravesicular space.

To determine if the degradative process observed was

specific for System A, we tested the stability of Systems N

and ASC in control and glucagon-treated vesicles during incu-

bation at 40C. The glucagon-treated vesicles showed simi-

lar patterns of decay for Systems A and N (Fig. 2-10A).

Transport by System A and System N was enhanced in the

glucagon-treated vesicles when compared to the control mem-

branes. Interestingly, these data are consistent with ear-

lier observations in cultured hepatocytes suggesting that

System N is stimulated by glucagon in vivo (Barber et al.,

1983) but not in vitro (Kilberg et al., 1980). The esti-

mated half-life of the decay of the hormone-stimulated activ-

ity of System A and N was approximately 13 h. It is diffi-

cult to be certain if System ASC is decaying either in the

glucagon-treated vesicles or in the control vesicles due to

the low basal rate of activity. Membrane vesicles isolated

from control hepatocytes also appeared to show decay of

Systems A and N (Fig. 2-10B). The half-life value for

System N was calculated to be about 13 h; however, in the

case of Systems A and ASC these measurements were too impre-

cise to report due to the low level of activity present.

Fig. 2-10. Decay of Systems A, N, and ASC in Vesicles from
Glucagon-Treated and Normal Hepatocytes. Membrane vesicles
from glucagon-treated (A) or normal control rats (B) were
incubated at 4*C for the indicated times. After each period
of incubation, the vesicles were tested for amino acid
transport activity as described in Fig. 2-3. System A (*),
System N (a), and System ASC (A) were tested as described
in the Methods section. The results are expressed as the
mean + S.D. of triplicate determinations of the
Na -dependent pmol of amino acid accumulated per mg pro-
tein per 15 s. Where not indicated, the error bars are con-
tained within the symbols.


o 600


I 200

0 5 10 15 20

0 5 10 15 20

System A Transport Activity in Membrane Vesicles from Human
Hepatoma Cells
Several investigators have reported that glycolysis and

nutrient transport activities are increased in transformed

cell lines (Parnes and Isselbacher, 1978). Amino acid trans-

port, in particular, is increased in membrane vesicles pre-

pared from SV40-transformed 3T3 cells (Lever, 1976). Our

laboratory has studied System A activity in a number of hepa-

toma cell lines from both human and rat. For example, a

human hepatoma cell line (HepG2) shows an increased basal

transport rate of nearly 12-fold when compared to the normal

hepatocyte (38.6 + 6.1 versus 444 + 22 pmol AIB per

mg protein per min for the normal and hepatoma cells, respec-

tively). To test whether or not the increased System A-

mediated transport activity observed in the human hepatoma

cells is retained in isolated membranes prepared from HepG2

cells, several membrane vesicle fractions were prepared as

described in the Methods section. Each of the subcellular

membrane fractions tested displayed Na -dependent AIB

transport activity (Fig. 2-11). The P1 fraction (l000xg

pellet) showed a sodium-dependence of 137 + 18 pmol AIB

per mg protein per min, while the P2 (45,000xg pellet) and

the P3 fractions (discontinuous sucrose gradient pellet)

showed 614 + 24 and 631 + 18 pmol AIB per mg protein

per min, respectively (Fig. 2-11). It appears, from these

results, that either sucrose-gradient centrifugation or a

differential centrifugation procedure at 45,000xg will

result in a similar level of specific activity for System A-

mediated transport. The data from either of these two



S1 (ND) P1 (137 + 18)


S2 (0) P2 (614 + 24) SUCROSE-STEP

P3 (631 + 18)

Fig. 2-11. Flowchart of HepG2 Membrane Preparation. HepG2
membranes were prepared as shown above and described in the
Methods section. The supernatant fractions (S) and the
pellet fractions (P) are indicated for each step. Fractions
which were tested for System A transport activity are indi-
cated in parentheses as the Na -dependent pmol of AIB
accumulated per mg protein per min + S.D. for triplicate
determinations. Fractions which were not tested for activity
are indicated by NO.

fractions indicate that System A activity is greater in

membrane vesicles from human hepatoma cells when compared to

the transport observed in vesicles from normal rat

hepatocytes (Fig. 2-8). When vesicles from the HepG2 cells

were prepared by similar centrifugation methods, but using

either bath sonication or nitrogen cavitation for cellular

disruption, the results were the same. Furthermore,

membrane vesicles from a rat hepatoma, H4-II-EC3, also

retained elevated rates of System A activity (data not

shown). These data argue that the increased transport

activity observed in transformed liver cell lines is

retained in membrane vesicles prepared from those cells.


Membrane vesicles have been prepared from either rat

liver tissue or isolated rat hepatocytes. These vesicles

actively accumulated AIB and more than 85% of the AIB accumu-

lation was inhibited by the System A-specific analog, MeAIB.

These data indicate that AIB is useful as a selective probe

for System A activity in isolated membrane vesicles. The

accumulation of AIB was energized by the imposed trans-

membrane Na gradient. The uptake of AIB was against a

concentration gradient as indicated by the distribution

ratio which was greater than one at the maximal overshoot


The vesicles also actively accumulated typical sub-

strates of the other sodium-dependent amino acid transport

systems found in rat liver. Alanine, cysteine, histidine,

and glycine were used as model substrates for Systems A,

ASC, N, and Gly, respectively. The specificities of these

amino acids for their respective transport systems were simi-

lar to those observed in intact hepatocytes (Kilberg, 1982).

The only notable difference was that Na -dependent hist-

idine uptake showed a greater degree of inhibition by MeAIB

in membrane vesicles than in whole cells (Kilberg et al.,


Several of the criteria that Lever (1980) has applied

to establish active transport of nutrient molecules by mem-

brane vesicles were tested in this system. Uptake of alan-

ine and histidine was stereospecific and AIB was actively

accumulated into an enclosed, osmotically-sensitive space.

AIB uptake was also dependent on temperature and protein con-

centration. AIB uptake was inhibited by the ionophores gram-

icidin and monensin which collapse the trans-membrane

Na+ gradient by either forming channels for Na+ ions

gramicidinn) or carrying Na+ ions across the membrane

(monensin). The electrogenicity of transport was demonstra-

ted by observing that the use of less permeant counter-

anions in the buffer resulted in lower transport velocities.

The membrane vesicle preparation was shown to be

enriched in plasma membrane through measurements of 5'-

nucleotidase and adenylate cyclase activities. Enzyme

marker studies indicated that the vesicles had some contamin-

ation by microsomes, but little or no contamination by mito-

chondria. Collectively, the results demonstrate that these


membrane vesicles are useful for amino acid transport stu-

dies by providing a system free of many of the restrictions

and complications imposed by intact cells. The charac-

teristics that we have described for this vesicle system are

consistent with those of other amino acid transport studies

using isolated rat liver membranes (Sips et al., 1980b;

Samson and Fehlmann, 1982; Quinlan et al., 1982; Meier et

al., 1984b). The demonstration of active amino acid trans-

port and its properties represents the first step toward

using isolated rat liver membranes to study the regulation

of System A by trans-effects, hormones, and adaptive con-


Membrane vesicles facilitate the study of trans-

inhibition of System A because, presumably, endocytosis and

recycling cannot occur due to the removal of intracellular

ultrastructures such as the microtubular network. Using

these membrane vesicles, trans-inhibition was detected when

the vesicles were loaded with AIB but not when the vesicles

were loaded with D-glutamine, an amino acid with no satur-

able Na -dependent uptake. Our results support the hypo-

thesis that the System A carrier is locked into an internal

orientation by binding of AIB (or a Na /AIB complex) at

the intracellular surface of the plasma membrane.

It is well established that glucagon treatment and star-

vation of hepatocytes for amino acids results in a protein

synthesis-dependent increase in System A activity (Kilberg,

1982; Shotwell et al., 1983). The increase observed in both

cases is kinetically defined as an increase in the Vmax of

the carrier rather than a change in the Km for the test sub-

strate (Shotwell et al., 1983). Taken together, the kinetic

effects and the protein synthesis-dependency are interpreted

to indicate that the stimulated transport activity is due to

a greater number of active carrier molecules in the plasma

membrane. We have utilized isolated membrane vesicles to

provide additional support to that proposal. Vesicles iso-

lated from amino acid-starved hepatocytes displayed elevated

System A transport at a similar magnitude to that observed

in intact cells, a result that is consistent with the hypo-


A large number of experiments were performed in which

rats were treated with glucagon in vivo and then membrane

vesicles were prepared from isolated hepatocytes. Increased

transport activity was always detected in the freshly iso-

lated vesicles and the level of stimulation paralleled, but

typically did not equal, the degree of induction observed in

the intact hepatocytes. Several explanations are possible

for the apparent incomplete retention of glucagon-stimulated

activity. We have eliminated some of these through experi-

mentation including differences in membrane composition of

the vesicles preparations from normal and glucagon-treated

cells and differences in the time at which the maximal

Na+-dependent transport is observed. It is also clear

from our studies that irreversible inactivation of System A

activity in the membrane vesicles from hormone-treated hepa-

tocytes contributes in a significant way to the loss of ele-

vated transport activity in the isolated membranes.

When Systems N and ASC were tested for inactivation in

membrane vesicles, System N decayed with a half-life similar

to that for System A. It could not be determined accurately

if System ASC was decaying at a significant rate. These

data imply that the inactivation process affects other trans-

porters as well as System A and may, in fact, result from a

non-specific effect on vesicle integrity. It cannot be

determined accurately if the inactivation process is also

occurring in vesicles from normal cells because the data at

each time point are not statistically different from each

other. Another unknown is the status of the membrane perme-

ability with respect to Na An increase in permeabil-

ity to Na+ ions could result in an apparent inactivation

of the carrier.

Membrane vesicles isolated from human hepatoma cells

(HepG2) showed increased System A activity when compared to

normal rat hepatocytes. Such a comparison is difficult

because we do not know the basal activity of transport in

normal human liver tissue. However, it is interesting to

note that the level of Na -dependent AIB uptake in HepG2

vesicles is comparable to the transport activity in

glucagon-induced rat liver vesicles. Our laboratory has

made direct comparisons between normal rat hepatocytes in

culture and several rat hepatoma cell lines. In every case,

the hepatoma cells contain enhanced transport activity. The

data obtained with isolated membrane vesicles are consistent

with the hypothesis that transformation induces new System A

carriers to be inserted into the plasma membrane, similar to


the proposed mechanism for glucagon-induction and adaptive

control of System A. Ideally, one would like to test System

A activity in a particular cell line before and after trans-

formation. Such an experiment was done using Rat-1 cells

transformed by the myc oncogene (Racker et al., 1985) and

NRK-49F cells transformed by transforming growth factor-B

(Boerner et al., 1985). Each cell line exhibited increased

System A activity following transformation; the induction

required de novo protein synthesis.



Reconstitution of ion transport systems has proven to

be a powerful method of resolving the transporters into

their functional components. The elegance of taking a

multi-protein complex apart, separating it into its indivi-

dual components, and then recombining the components to

regain activity is apparent for the mitochondrial ATPase

(Racker, 1976). A number of comprehensive reviews have

appeared over the last few years that discuss various

methods of performing reconstitution of ion transporters or

membrane receptors (Levitzki, 1985; Eytan, 1982; Hokin,

1981; Racker, 1979).

Reconstitution of transport processes is a technique

that must be developed in a series of stages. The first

stage is the establishment of a membrane source that con-

tains large quantities of activity. The second stage is the

demonstration of transport activity in an artificial proteo-

liposome system by combining membrane fragments and sonica-

ted liposomes (membrane-fragment reconstitution). The final

stage requires removal of the transporter from its native


membrane by detergent-extraction. Reconstitution is

achieved by combining the extracted proteins with lipid and

then removing the detergent. Recovery of activity from an

inactive detergent-extract is a convincing argument for


Several methods of reconstitution after detergent-

extraction of membrane proteins have proven to be highly

successful. The first reconstitution of oxidative phosphor-

ylation was performed using a cholate-dialysis technique

(Kagawa and Racker, 1971). The technique slowly removes

detergent by dialysis against detergent-free buffers which

facilitates the insertion of proteins into liposomes. Since

1971, a wide variety of transport systems have been reconsti-

tuted using detergent-dialysis techniques (Racker, 1979).

Although these methods have been useful for many systems,

they require the proteins to be in contact with detergent

for long periods of time (18-98 h) often resulting in inacti-


To circumvent the inactivation problem caused by long

periods of dialysis, rapid reconstitution techniques were

devised. One example is the detergent-dilution procedure

which has been useful for the reconstitution of cytochrome

oxidase activity (Racker, 1972) and the reconstitution of

coupled oxidative phosphorylation (Racker, 1975). The tech-

nique is simple and rapid in that it only requires dilution

of the detergent-extract so as to lower the detergent concen-

tration below its critical micellar concentration (CMC)

followed by centrifugation to collect the resulting


Another rapid technique to remove detergents with

micelle molecular weights less than 10 kDa, such as cholate

and deoxycholate, was described by Brunner et al. (1976).

The procedure requires passing the detergent-extract over a

Sephadex G-50 column. The gel filtration matrix retards the

small detergent molecules and allows the large lipid and pro-

tein molecules to fuse into proteoliposomes and pass through

in the void volume. McCormick et al. (1984) used this proce-

dure prior to a freeze-thaw step to achieve reconstituted

amino acid transport activity from EAT cell membranes.

A procedure which has shown great success for the glu-

cose transporter from red blood cells (Kasahara and Hinkle,

1976) requires a freeze-thaw step coupled with sonication.

This technique involves combining sonicated lipid with solu-

bilized membrane proteins followed by rapid freezing in a

dry-ice-ethanol bath. After thawing at room temperature,

the mixture is sonicated for short periods of time resulting

in a 3-fold greater activity than without the sonication.

Alfonso et al. (1981) have observed that the activity of the

mitochondrial proton-pump is increased even further when

small amounts of cholate are added before the freeze-thaw

step. Pick (1981) suggests that the fusion process which

occurs during the rapid freezing and thawing is a result of

the formation of crystalized water molecules in two frozen

planes separated by the membrane domain. The bilayer is

easily fractured, exposing hydrophobic regions which can

fuse to form large liposomes.

Another technique used successfully to achieve recon-

stitution of transport processes is the Ca+ -fusion pro-

cedure. This alternate procedure capitalizes on the observa-

tion that liposomes which contain phosphatidylethanolamine

and phosphatidylserine or cardiolipin rapidly fuse in the

presence of Ca+2 (Miller and Racker, 1976). This tech-

nique has proven useful for reconstitution of cytochrome

oxidase activity from mitochondria and has also found wide

use in drug delivery systems (Szoka and Papahadjopoulos,


Reconstitution of Na -dependent amino acid trans-

port activity has been reported in kidney brush border mem-

branes (Kinne and Faust, 1977; Takahashi et al., 1985),

SV-40-transformed 3T3 fibroblasts (Nishino et al., 1978),

and EAT cell membranes (McCormick et al., 1984; Cecchini et

al., 1977). Reconstitution of amino acid transport activity

from kidney brush border membranes has been achieved,

although the level of Na -dependence was low. Kinne and

Faust (1977) extracted membranes with Triton X-100, then

removed excess Triton by passing the detergent-extract over

Bio-Bead SM-2 columns. The detergent-extract was reconsti-

tuted by combining it with lipid and sonicating the mixture.

Sodium-dependent alanine uptake was quantitated, but less

than a 2-fold difference was reported for uptake in Na+

versus K -containing buffers. The authors did not dis-

cuss the margin of error present in the assays so the data


may not be statistically different. The activity reported

may not be representative of System A activity because

Na+-dependent alanine uptake may be through several dif-

ferent systems including System ASC.

Reconstitution of System A from SV-40-transformed mouse

fibroblasts membranes, extracted with 2% cholate was accom-

plished by Nishino et al. (1978), by passing the extracted

proteins over a Sephadex G-50 column, and then subjecting

them to a freeze-thaw step. Sodium-dependent AIB uptake was

determined by a rapid filtration technique and was reported

to be at least 3-fold higher in sodium thiocyanate than in

choline chloride.

Reconstitution of System A in EAT cell membranes has

been reported after cholate-extraction followed by detergent

dilution (Cecchini et al., 1977) or cholate-extraction fol-

lowed by dialysis (Bardin and Johnstone, 1978). Later

reports from Johnstone's group demonstrated the existence of

Na -dependent AIB uptake after extraction of EAT mem-

brane proteins in 2.5% cholate/4 M urea followed by dialysis

to 0.25% cholate, Sephadex G-50 chromatography, and a

freeze-thaw step (McCormick et al., 1984).

Little progress has been made toward the purification

of the System A carrier using reconstitution as an assay.

Cecchini et al. (1978) reported that an EAT cell membrane-

extract retained Na -dependent alanine uptake after

ammonium sulfate fractionation, Biogel P-60 chromatography

and DEAE-cellulose chromatography. Polyacrylamide gel elec-

trophoresis of the final extract showed only 15 or so

Coomassie-staining bands. McCormick et al. (1985) reconsti-

tuted Na+-dependent AIB transport from EAT cell mem-

branes that had been extracted with cholate/urea as

described above and showed an apparent enrichment in a pro-

tein of 125 kDa in the proteoliposomes when compared to the

native plasma membranes. Using reconstitution as an assay,

Takahashi et al. (1985) have determined by radiation-

inactivation studies that the target size for the Na -

dependent glucose transporter from kidney brush border mem-

branes is one-million daltons and 1.2 million daltons for

the Na+-dependent alanine transporter.

The object of the present studies was to devise a

rapid, efficient system to reconstitute System A activity

from EAT cell membranes. Techniques for reconstitution of

System A described by McCormick et al. (1984) and Cecchini

et al. (1978) were attempted in our laboratory and in the

laboratory of Dr. Efraim Racker but no reproducible activity

could be obtained. In collaboration with Dr. Racker, we

developed a novel reconstitution technique for System A

activity which was rapid and reproducible. Having estab-

lished the technique, future research in the laboratories of

both Dr. Kilberg and Dr. Racker will focus on the purifica-

tion of the transporter using reconstitution as an assay.


Materials and Methods


Cholic acid was obtained from Sigma Chemical Company

and was recrystallized three times in ethanol as described

by Kagawa and Racker (1971). Ultrapure urea was obtained

from Pierce. Protein concentration by ultrafiltration in a

stirred cell was performed using an Amicon model 8010 or

8050 cell fitted with a YM-30 membrane. Asolectin was

obtained from Associated Concentrates and was stored at

-200C. (U-14C)-sucrose in 20% ethanol was obtained

from ICN. The detergent NP-40 was obtained from Particle

Data Laboratories, Inc. and octyl-glucoside was supplied by

Calbiochem. Hydroxylapatite (Bio-Gel HTP) and all reagents

for electrophoresis were obtained from Bio-Rad. All experi-

ments requiring sonication were performed at 220C in a bath

sonicator supplied by Laboratory Supplies Co., Inc. All

other reagents were obtained from Sigma Chemical Company.

Preparation of EAT Cell Membrane

EAT cell membranes were prepared as described by Racker

et al. (1984). 75 Swiss white mice (Charles River--19-21 g)

were injected intraperitoneally with 0.25 ml of EAT cell sus-

pension which had been removed from mice and filtered

through two layers of cheesecloth. After 8-10 days, the

mice were sacrificed by cervical dislocation, the intraperi-

toneal cavity was carefully opened, and the ascites fluid

was collected and filtered through two layers of cheese-

cloth. The ascites fluid was typically whitish-yellow in

color and was not tinged at all with blood. Any bloody

fluid was discarded. The cells were diluted in 10 volumes

of wash buffer (140 mM NaCl, 5 mM KC1, 1 mM MgCl2, 10 mM

HEPES, pH 7.4) and placed in GSA rotor bottles. The cell

suspension was centrifuged at 650xg in a GSA rotor for 5

min. The supernatant was decanted and the pellets were

resuspended in a small volume of wash buffer. The pellet

suspension was centrifuged again at 650xg for 5 min (table-

top centrifuge) in order to pack the cells. The cells were

frozen in 45 ml centrifuge tubes and stored at -70C. Fro-

zen cells (150 g) were used for each preparation of mem-

branes. The frozen cells were thawed in warm water and

poured into 1.5 L of 10 mM CaC12 (dihydrate; Fisher ACS,

pH 8.0) at room temperature. The cells were stirred slowly

at 40C for 5 h and were homogenized by rapidly forcing

through a Logeman homogenizer with the nozzle half closed

(this and all subsequent steps were done at 40C). The homo-

genate was poured into 6 GSA rotor bottles and centrifuged

at 650xg for 5 min. The pellets were discarded and the

supernatant was collected and centrifuged at 16,000xg for 40

min in the GSA rotor. The pellets were gently homogenized

in 30 ml of sucrose-EDTA-PMSF buffer (0.3 M sucrose, 1 mM

EDTA, 1 mM PMSF, pH 7.4) using a typewriter brush and a

loose-fitting Potter-Elvehjem homogenizer. The suspension

was incubated on ice for 1 h then it was centrifuged at

27,000xg for 30 min. The supernatant was decanted and the

pellet was resuspended in 30 ml of sucrose-EDTA-PMSF as

before. The pellet was gently homogenized and centrifuged

again at 27,000xg for 30 min. The pellet was resuspended as

before and centrifuged at 27,000xg for 30 min. The final

pellet was resuspended in a small volume of sucrose-EDTA-

PMSF and frozen at -700C in 0.5 ml aliquots (total yield =

200-300 mg per 150 g of packed cells).

Detergent-Extraction of EAT Cell Membrane

EAT cell membrane was extracted with detergent as

described in the figure legends or in the text. The details

of the protocol for optimized extraction of the EAT mem-

branes using cholate and urea by the technique of McCormick

et al. (1984) is described here. Twenty mg of EAT cell mem-

brane protein was mixed with 10 ml of solubilization buffer

(2.5% cholic acid, 4 M urea, 0.1 mM EDTA, 100 mM NaCl, 5 mM

Tris-HC1, pH 7.4) at a protein concentration of 2 mg/ml at

4C for 30 min. The mixture was centrifuged at 125,000xg

for 45 min. The supernatant was removed and placed in a dia-

lysis bag, then the suspension was dialyzed overnight

against 100 volumes of dialysis buffer (0.2% C12E9,

5 mM Tris-HCl, 0.1 mM MgCl2, 0.1 mM CaCl2, 100 mM

KC1, 1 pM PMSF, pH 7.45) at 40C. The dialysate could be

stored at -70C for up to one month with minimal loss of

activity. After the determination of protein content by a

modified Lowry procedure (Bensadoun and Weinstein, 1976),

the suspension was prepared for reconstitution.

Freeze-Thaw Reconstitution and Transport Assay

Reconstitution of transport activity was performed by

mixing 0.5 mg of protein from either the membrane-fragment

preparation or the detergent-extract, 10 mg of sonicated aso-

lectin (stock = 40 mg/ml) and 1.5 mg of K+ cholate

(stock = 10% w/v). The final volume of the mixture ranged

from 0.7-1.0 ml. The mixture was frozen in liquid nitrogen

and thawed at room temperature, then the suspension was

diluted with 2 ml of K+-uptake buffer (120 mM KC1, 10 mM

MgCl2, 10 mM HEPES, pH 7.45) and centrifuged at

125,000xg for 45 min. The pellet was resuspended in 200 pl

of K+-uptake buffer with a stirring rod and gentle vor-

texing. A summary of the reconstitution procedure, inclu-

ding further details, is shown in Fig. 3-1.

Amino acid transport buffers were prepared as 2X stocks

using NaCl (120 mM NaC1) or KCI (120 mM KC1) uptake buffer

and 200 pM AIB. To initiate uptake, 20 p1 of vesicles

were added to 20 pl of uptake buffer. The mixture was vor-

texed and incubated at 22C for 1 min. To stop the uptake,

1 ml of ice-cold stop buffer (154 mM NaCl, 10 mM

Na2HPO4, adjusted to pH 7.45 with HC1) was added and

the suspension was vortexed. The mixture was immediately

filtered over a Gelman nitrocellulose filter (25 mm diameter

and 0.45 pm pore size) and the filter was washed once with

3 ml of ice-cold stop buffer. The filter was placed into a

15 ml plastic scintillation vial and the radioactivity

trapped on the filter was determined after adding 5 ml of

Bray's scintillation cocktail. Transport activity is



0.2% C E 2.5% CHOLATE & 4 M UREA
FOR 30 MI~2A0 40C FOR 30 MIN AT 40C
\ I





Fig. 3-1. Flowchart of the Procedure for Detergent-
Extraction and Reconstitution of System A Transport Activity
Using the Freeze-Thaw Technique.

expressed as pmol of AIB trapped per mg protein per min.

Unless otherwise indicated in the figure legends, the

results were from a single membrane preparation and the S.D.

of triplicate determinations was less than 10% of the mean.

Concentration of Detergent-Extracts Using Ultrafiltration

Detergent-extracts (1-2 mg protein/ml) were placed in

an Amicon ultrafiltration cell (10 ml or 50 ml size) fitted

with a YM-30 membrane. The apparatus was immersed in a tray

of ice water and the tray was placed on a magnetic stirring

plate. Nitrogen gas was introduced into the cell at 60 psi

while stirring was continued on ice. Concentration was

allowed to continue until the volume was reduced to 1-2 ml.

The concentrated extract was removed with a Pasteur pipette.

Sucrose Trapping in Reconstituted Proteoliposomes

Detergent-extract (0.5 mg protein/413 pl), 10 mg of

asolectin (250 pl) and 1.5 mg (15 pl) of K+ cholate

were combined with 5 pl (8.28 x 106 dpm) of 14C-

sucrose (0.746 pCi/pl; 1.21 mM) and then frozen in liquid

nitrogen. After thawing, the suspension was diluted with 2

ml of K+-uptake buffer and 100 u1 was removed for deter-

mination of radioactivity. A control sample was prepared in

which the 1C-sucrose was added after the freeze-thaw

step. Each suspension (100 pl) was passed over a Sephadex

G-50 column (1x16 cm; flow rate = 1.33 ml/min) equilibrated

with K+-uptake buffer. Fractions (0.22 ml) were col-

lected in plastic scintillation mini-vials and then analyzed

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd