Neutral Amino Acid Transport in Hepatocytes Isolated from ...

THEJOURNALO F BIOLOGXCAL CHEMISTRY Vol. 257, No. 24, Issue of December 25, pp. 14960-14967, 1982 Printed in U.S.A.

Neutral Amino Acid Transport inHepatocytes Isolated from Streptozotocin-induced Diabetic Rats* (Received for publication, July 30, 1982)

Ellen F. Barber, Mary E. Handlogten, Thomas A. VidaS, and Michael S. Kilbergg From the Departmentof Biochemistry and Molecular Biology, J.Hillis Miller Health Center, University of Florida, Gainesuille, Florida 32610

In the present study, transport by Systems A, ASC, and N was shown to be elevated in hepatocytes isolated from diabetic rats. After the cells were placed in primary culture, the System ASC activity declined rapidly, while the decay of Systems A and N was slower and dependent on protein synthesis. The elevated 2-aminoisobutyric acid uptake was the resultof an increase in the V,, of a single, high affinity system, presumably System A. The stimulation of System A could not be accounted for by an increase in the affinity for Na+;in fact, the apparent K,,, for the ion was actually greater in the experimental cells. Release from trans-inhibition was also eliminated as a possible explanation. The data suggest that during the development of the diabetic state the liver is triggered to induce the activity of System A by synthesizing the necessary protein components. Treatment of the cultured hepatocytes with insulin could partially reverse the stimulation due to diabetes, indicating that the induction of System A may be the resultof the hyperglucagonemia associated with the disease. In supportof this hypothesis, the cells from the diabetic rats were resistant to further stimulation of System A by glucagon, yet they did respond to high levels of insulin or to amino acid starvation. Glucagon does not appear to be involved in the induction of System N in the diabetic animal because this system is not responsive to either glucagon or insulin when tested in vitro.

Regulation of hepatic gluconeogenesis by the pancreatic hormones insulin and glucagon has been extensively studied both in vivo and in uitro by many investigators and much of the latter work has involved isolated rat hepatocytes. These studies have shown that glucagon can stimulate the incorporation of amino acid carbons into glucose (1,2) andphysiological concentrations of insulin can antagonize this effect (3).A great deal of effort has been put forth to determine the exact site or sites of control by these hormones and various enzymatic steps have been proposed as possible points of hormonal regulation (4). One possible mechanism by which these hormones may regulate gluconeogenesis has received less attention, namely, transport of amino acid precursors across the * Thesestudies were supported by Grant AM-28374 from the National Institutes of Health, the Institute of Arthritis, Diabetes, Digestive and Kidney Diseases. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. *Present address: Department of Biochemistry, University of Texas Health Science Center, Harry Hines Blvd., Dallas, TX. 8 To whom correspondence should be addressed.

hepatocyte plasma membrane. As early as 1970 Exton et al. (5) postulated that substrate supply might play a significant role in the control of the gluconeogenic process. Recently, it has been demonstrated that in hepatocytes isolated from fed or fasted rats,thetransport of alanine into the cell can represent the rate-limiting step in its overall metabolism, particularly at physiological concentrations of the amino acid (6, 7). Glucagon has been shown to stimulate the transport of neutral amino acids because of its action on hepatic System A activity (8,9). Given that diabetics generally show a hyperglucagonemia in addition to the reduction in insulin levels (lo), it was postulated that System A activity wouldbe elevated in the livers of diabetic rats.This possibility was fiist demonstrated in an isolated perfused liver preparation (11) and more recently c o n f i i e d for isolated hepatocytes (12). Insulin treatment in uivo,prior to testing for transport activity inuitro, resultedin reversal of the stimulatedSystem A activity (11, 12). Furthermore, when rat livers were isolated from diabetic donors and perfused with insulin in uitro, the enhanced transport rates were also suppressed (11).These results suggested that insulin and glucagon were acting in an antagonistic fashion to regulate the activity of hepatic amino acid transport. On the other hand,in freshly isolated rat hepatocytes, exogenously added glucagon and insulin do not show such antagonism for control of transport (13). The present report, the first in a series characterizing the transport of neutral amino acids by rat hepatocytes isolated from diabetic rats, shows the effect of experimental diabetes on 1) the activity of several neutral amino acid transport systems during the initial 24 h of primary culture, 2) the amino acid and Na’ kinetics for System A, 3) the effect of stimulation of transport by insulin and glucagon, and 4) the possible control of System N by factors other than glucagon which may be present in the diabetic rat. MATERIALSANDMETHODS

Experimental Diabetes-Diabetes was induced by an intraperitoneal injection of streptozotocin at 10 mg/100g body weight.The drug was diluted in 50 mM sodium citrate, pH 4.5, just prior to use. After 2 to 4 days, the animals were tested for hyperglycemia by measuring their serum glucose levels (14). Rats were considered chronically diabetic when the serum glucose value was greater than 300 mg/100 ml of serum. The average serum glucose values obtained for the animals from which the cells were taken for the present studies were 129 f 38 and 383 & 65 mg/100 ml for the control and diabetic rats, respectively. In most instances the serum from the diabetic animals also showed a marked lipemia. Hepatocyte Isolation and Culture-All of the rats used in the experiments described were fasted overnight and the hepatocytes were isolated between 7 and 9 a.m. on the day of the experiment. Hepatocytes were isolated as described previously (15, 16). Hepatocyte primary cultures were established by placing the cell suspension in collagen-coated 12-well (675,000 cells/well) or24-well(300,000

14960

Amino Acid Transport

in Hepatocytes from Diabetic Rats

3 - 1

14961

cells/well) Costar clustertrays (15).After 1 to 2 h, the initial plating 240 medium was changed either to fresh Waymouths medium lacking serum or to the appropriate buffer as indicated for each experiment. Transport Assays-Aminoacid uptake wasmeasuredby the method of Gazzola et al. (17). The Na’-free buffers (Krebs-Ringer bicarbonate or Krebs-Ringer-phosphate,pH 7.4) weremadeby re0 dloborir placing the corresponding Na+ salts with choline chloride, choline bicarbonate, or choline phosphate. Following the transport assays, normol the cells were solubilized by the addition of either 0.3 ml (12-well 0 trays) or 0.2 ml (24-welltrays) of a 0.2 N NaOH, 0.2% sodium dodecyl c 15 min at room sulfate solution to each well. After incubation for temperature, one-half of the extract was placed in scintillation vials for determination of radioactivity, while the remaining extract was left in each well to measure protein contentby a nlodification of the Lowry procedure (15). The Na+-dependent transport of both AIB’ and histidine was shown to be linear for at least 3 min in the cells isolated from normal or diabetic rats (data not shown) and similar results have been reported for glutamine in a previous study. The amount of protein per IO6 cells was not significantly different forthe two cell types (normal = 0.91 0.15, diabetic = 0.93 & 0.17mgof protein/1o6 cells). Analysis ofData-The data were calculatedand analyzed withthe aid of a microcomputer utilizing software which incorporates standard 1 1 1 I I 0 6 12 18 24 statistical analyses. The transport kinetic parameters were calculated with FORTRAN software designed to describe uptake by one satuHours rable and one nonsaturable component(18).The software calculated FIG. 1. System A activity in hepatocytes isolated from norand subtracted the nonsaturable uptake (i.e. the slope of the linear region of the velocity uersus [SI plot) prior to the calculation of the mal or diabetic rats. The Na’-dependent transport of 50 PM AIB was measured for 1 min at specific intervals after the cells had been kinetic constants. Materials-The highlypurifiedglucagonandinsulinwere the placed in primary culture. The Na+-dependent velocity was detergenerous gifts of Lilly. Male Sprague-Dawleyrats weighing between mined by subtracting the rate of uptake in the absence of Na’ from 100 and 200 g were obtained from a colonymaintainedby the that found inthe presence of Na’. The Na’-free medium was prepared University of Florida, Division of Animal Resources. [3H]2-Aminoiso- by replacing the corresponding Na’ salts with choline chloride, choacid, ~-[3,4-~H]gluta-line bicarbonate, or choline phosphate. The results are given as the butyric acid, [l-’4C]2-(methylamino)-isobutyric averages & S.D. of at least three determinations. mine, and ~-[G-~H]histidine were purchased from New England Nuclear and unlabeled AIB was from Eastman Kodak. L-Amino acids, Type I collagenase, and antibiotics were obtained from Sigma. Waymouth’s medium was purchasedas a powder from Flow Lahratories. protein). The decrease in cell water was nearly the same for K. C. Biologicalswas the source of the fetal calf serumand the both cell types and should not affect the comparison of the scintillation cocktail (No. 3a70B) was obtained from Research Prod- transport rates for any one particular time after culture. ucts International. Kinetics ofAIB Transport-The kinetics of AIB transport in hepatocytes from normal, diabetic, and adrenalectomizedRESULTS diabetic donor rats was measured over the amino acid concentration range of 0.05 to 50 mM. The Lineweaver-Burke plot SystemAActivity in Culture-When hepatocytesfrom diabetic rats were placed in primary culture the activity of for the data from each of the three types of cells was linear (Fig. 2 ) . This result System A, as measured by the uptake of AIB, was consider- over the entire concentration range tested ably enhanced in comparison to the activity in the cells from does not necessarily demonstrate homogeneity of AIB transport, butinhibition analysis using MeAIB suggests that, under non-diabeticdonors (Fig. 1). The elevatedtransportrate decayed with time during the first 48 h of culture such that our conditions, Na’-dependent AIB uptake is mediated priafter this period the uptake of AIB by the two cell types was marily by System A. Computer analysis of the data showed indistinguishable (data not shown). In support of our previous that the stimulationof AIB uptake by the cells isolated from of a the diabetic rats resulted from an increase in the V,, results (15), the transport rate by System A in the control transport component having a K,,, for AIB of 1to 2 mM. This cells remained relatively constant during the time studied. The decay of the activity in the cells from the diabetic animalsresult is in disagreement with the dataof Samson et al. (12) who have shown increased AIB uptake infreshlyisolated was rapid during the first 6 to 8 h and then continued to decay, but at a much slower rate. Despite this initial rapid hepatocytes from diabeticdonors. They ascribed the effect to decline in activity, the transport in the hepatocytes isolated an increase in theV,, for a process having a K, of 35 to 40 from the diabetic rats was still significantly elevated after 24 mM, although this value must represent an estimate in that h of culture (Fig. 1). The decline in velocity during the fist 8 the highest concentration of AIB tested in their studies (30 h appears tofollow fist order kineticswhich yields ahalf-life m)was below this range. Morin et al. (20) have shown that if rat hepatocytes are maintainedin culture for 20 h, the low of approximately 2 h. Monitoring System A activity with MeAIB gives essentially the same results (data not shown). affinity system for AIB transport (K, = 30 to 50 mM) is not The intracellular water content, estimatedby the method of detectable and the high affinity system ( K , = 1 to 2 mM) Kletzien et al. (19), was the same for the hepatocytes isolated predominates. In contrast, cells from the same preparation, from the normal and streptozotocin-treated animals and re- when tested as freshly isolated cell suspensions, exhibit both mained relatively constant for most of the culture period the low affinity and the highaffinity systems (20). A full (average values for 1 to 18 h were 1.6 f 0.3 and 1.6 f 0.4 pl/ characterization of these two systems has not been reported, mg of protein for the cells from the normal and diabetic although Freychet and his collaborators have demonstrated donors, respectively), butlower values wereseen after24 h of that MeAIB inhibits both agencies. It is not clear at the culture (normal = 1.1 & 0.1,diabetic = 0.93 f 0.1 ,ul/mg of present how the singular term “System A ’ should be applied in this situation. Recently, Kelley and hisco-workers (21) have published an The abbreviationsused are: AIB, 2-aminoisobutyricacid MeAIB, 2-(methylamino)-isobutyricacid. extensive study of AIB transport kineticsfor cultured rat

.-

*

’

I \

Amino Acid Transport in Hepatocytes from Diabetic Rats

14962 'ool"---l

ao

-

0

t

5

IO

15

20

compared AIB uptake in cultured cells isolated from normal and adrenalectomized rats (16). In that study, it was shown that the effectsof adrenalectomy on amino acid transport in isolated hepatocytes canpersist for at least 24 h after thecells have been placed in primary culture. Studies involving alanine transport in normal and diabetic rats yielded results similar to those for AIB shown in Fig. 2, but the K,,, for alanine was higher than that found for AIB and the increase in the V,,, due to the inductionof diabetes was not as large (Table I). These differences may reflect the heterogeneity in alanine transport; it is well recognized that System ASC also contributes to the hepatic uptake of this amino acid and kinetic measurements may not be sufficient to distinguish the ASC-mediated uptake from the transport of alanine by System A (22, 23). Nevertheless, the results demonstrate that the uptake of alanine, a major gluconeogenic isolated precursor, is significantly elevated in the hepatocytes from diabetic rats. Relation between AIB Transport and Na+ Levels-The role of Na+ was tested by measuring AIB uptake at four different amino acid concentrations in the presence of varying amounts of extracellular Na+ between 15 and 140 mM. The 11, show that data, depicted in Fig. 3 and summarized in Table the K,,, for Na+ was 30 to 40 mM for the hepatocytes isolated from the non-diabetic rats, whereas the value for the cells fromthestreptozotocin-treatedanimals wassignificantly higher, between 50 and 70 mM. It is not clear at the present for K,,, a t 5.0 mM AIB in the control time if the higher estimate cells represents a trend that would continue at higher levels of the amino acid substrate. The calculatedvalue for Vmaxat each concentration of AIB was greater in the cells from the diabetic animals, although thedegree of stimulation declined with increasing levels of AIB (Table 11).A replot of the data taking AIBas the substrate generateda family of lines which showed that the estimates forK,,, (approximately 1 to 2 mM) for AIB remained relatively constant at all concentrations of Na+ tested (data not shown). Derepression a n d Release from Trans-inhibition-The hepatic System A activity has been shown to be inhibited by intracellular amino acids which are transported by that system (24). Increasedamino acid uptake by System A through release from trans-inhibition during aminoacid depletion can be mistakenly attributed toa direct physiological stimulation. The data shown in Fig. 4 illustrate the effect of amino acid

FIG. 2. The kinetics of Na+-dependent AIB transport by hepatocytes isolated from normal (A), diabetic (B), or adrenalectomized-diabetic (C) rats. The Na'-dependent uptake of AIB over the concentration range of 0.05 to 50 mM was measured for 1 TABLEI min at 37 "C. The resultswere calculated with the aid of FORTRAN Kinetic constants for Nu'-dependent AIB transport by hepatocytes software as described under "Materials and Methods."The nonsatuisolated from normal or diabetic donor rats rableuptakerates, subtracted priortocalculation of the kinetic The procedures used to measure the transport kinetics of 0.05 to constants, were 736, 258, and 73 pmol. mg" of protein. min" .mM" for the cells fromthe normal, diabetic, and adrenalectomized-diabetic50 mM AIB are discussed in the legend to Fig. 2 and in the section rats, respectively. The kinetic constants obtained are summarizedin under "Materials and Methods." The alanine assays were made in the same manner as for AIB except that the cells were treated with Table I. 0.5 mM aminooxyacetate for 30 min prior to the transport assays and hepatocytes showing that AIB uptake occurs by what appears the uptake measurements were for 30 s. Both of these changes were made to minimize the possible effects of alanine metabolism. to bea single component having a K,,, of approximately 1 mM. Condition of Vmax K"2 Vmax/Km Furthermore, their results indicate that the pancreatic horcell donor mones, insulin and glucagon, act on this system to increase pmol.mg" mM protein, min" in Our data, summarized the Vmaxwith little or no change K,,,. proposal. Whenthe in Table I, areconsistentwiththat AIB 2.3 t 0.3 690 t 80 300 Normal kinetics of AIB transport was measured using hepatocytes 1.5 t 0.1 2800 f 90 1867 Diabetic isolated from animals treated with streptozotocin several days 1440 t 170 282 5.1 t 0.6 Adrenalectomizedafter undergoing bilateral adrenalectomy, the activity of Sysdiabetic tem A was elevated, but the increase was significantly less pmol.mg" mM protein. 30 s" than that found in cells from diabetic rats which had not been adrenalectomized (Table I).Kinetically these cells showed an Alanine 2230 k 900 558 4.0 f 1.3 Normal increase in V,.,, but the K,,, was increased as well (2.3 versus 3830 t 430 1064 3.6 f 0.3 Diabetic 5.1mM). Asimilareffect on K,,, wasobserved when we

Amino Acid

& []

I

Transport in Hepatocytes from Diabetic Rats

[+I.

mm"

m"'

FIG. 3. The relation between the extracellular Na+ concentration and the uptake rate of AIB by hepatocytes isolated from either normal or streptozotocin-treated animals. The Na'dependent transport of 0.1, 0.5, 2.0, or 5.0 mM AIB was measured in the presence of various extracellular Na' concentrations between 15 and 140mM. The osmolarity of the extracellular solution was kept constant by the addition of the appropriate amount of choline chloride. The uptake rate in the absence of Na' was subtracted from each value prior to analysis for kinetic constants. These constants, summarized in Table 11, were obtained from linear regression computer analysis. Note that the substrate under test is AIB; the rate of Na' transport per se was not measured (see text).

TABLE I1 The kinetics of Nu' transport as estimated by the uptake of radioactively l a b e l d .A TB The uptake of AIB was measured at the concentrations shown, in the presence of varying amounts of extracellular Na'. The apparent K , and V , , values for Na' were then calculated by utilizing the velocity of AIB uptake and the extracellular Na' concentration as the "substrate." The validity of this calculation is based on the observation that thetranslocation ratio of System A is 1 Na' for each AIB. The double reciprocal plots of the datasummarized in this table are shown in Fig. 3. Condition of cell donor

Normal

Diabetic

Apparent K,,, for Na+

CAW

VmaX for AIB

uptake

4). The latter observation indicates that these cells continue to respond to external factors such as amino acid starvation which can alter nutrient transport activity (23). Hormone Treatment in Vitro-Hepatocytes from normal or diabetic rats were placed inprimary culture in the presence of varying amounts of insulin between the concentrations of 10"' and lop7M. The cells fromthe normal rats responded to increasing levels of insulin ina concentration-dependent fashion as reported by several investigators (Fig. 5). The transport of AIB was stimulated by physiological concentrations of the hormone, and the effectwasmaximal at M (data not shown). In contrast, the amino acid uptake by the hepatocytes isolated from the diabetic rats was not enhanced by physiological concentrations of insulin, rather it was decreased by insulin levels of10"' or lo-' M (Fig. 5). If the amount of insulin was increased to lo-* M or more, the AIB transport was stimulated so that theuptake rates were faster than those found in the absence of hormone; note that the rate in the absence of insulin is already several times faster than that of the cells fromthe control animals. Thus, System A activity in hepatocytes isolated from diabetic rats responds to insulin in a biphasic fashion. The hepatocytes from both normal and diabetic donors were also tested for their sensitivity to submaximal levels of glucagon. The hormone was added to the cells for 3 h at concentrations of lo-" to lo-* M, the concentration range where the stimulatory action of the hormone is nearly linear (9, 16). After incubation in the presence of the hormone, the normal cells showedenhanced AIB transport after treatment 50

T,

1

rnM

0.1 0.5

2.0 5.0

32 37 31 59

73 420 1500 4200

0.1

71

0.5

2.0

64 52

230 870 2600

5.0

73

660

starvation in the presence or absence of 0.1 mM cycloheximide for hepatocytes from normal or diabetic rat donors. For the fiist 90 (normal) or 120 (diabetic) min of incubation in the amino acid-free medium there was no significant increase in the uptake of AIB. Including the cycloheximide for these time periods did not alterthe response (Fig. 4). These results suggest that under the conditions of these experiments transinhibition was not a critical factor. Between 1.5 and 4 h the uptake by the control cell population increased by 4- to &fold when they were incubated in amino acid-free medium and this stimulation was almost completely abolished by the presence of cycloheximide. In contrast, the increase in AIB uptake due to amino acid starvation, although detectable, was suppressed in the cells from the streptozotocin-treated animals. It should be noted that the small stimulation that did result after amino acid deprivation of the hepatocytes from the diabetic rats was effectively blocked by cycloheximide (Fig.

:

I

/

201

/O

E

7n

1

1

pmol.rng;'qrotein. rnm

rnM

:

14963

""_ 0

E" Diabotie

'

1

2

3

I4ou.r

FIG. 4. Release from trans-inhibition and derepression in hepatocytes from normal or diabetic rats. Cells were cultured for 2 h in plating medium as described under "Materials and Methods." The medium was then changed to fresh Waymouth's medium lacking serum (amino acid-rich, AAR),Krebs-Ringer bicarbonate buffer (amino acid-free, AAF), or Krebs-Ringer bicarbonate buffer (AAF) containing 0.1 mM cycloheximide. After an additional 2 h, fresh medium of the same composition was placed on the remaining cells. The Na+-dependent uptakeof 50 PM AIB was tested for 1 min at the time points indicated. The results are the averages of a t least three determinations and the standard deviations, omitted for clarity, were usually 10%or less.


14964

observation that System ASC is not responsive to either hormonal or adaptational regulation(26). System N Activity in Culture-In the rat hepatocyte, the Nac-dependent uptakeof both glutamine and histidine occurs almost exclusively by System N (27), although some media- 240 tion of glutamine uptake by System A has been shown after amino acid starvation (28). The activity of System N was monitored during the initial 24 h of culture for hepatocytes isolated fromboth normal and diabetic rats. Na'-dependThe - 160 ent uptake of histidine was increased by nearly 4-fold in the cells from the animals injected with streptozotocin (Fig. 7). This stimulated uptake rate declined considerably over the initial 24-h period of culture, yet always remained greater - 80 than the rate seen in the controlcells. The observation that 3 System N activity was enhanced in thecells isolated from the diabetic rats was unexpected, considering that preliminary experiments, testing this agency for stimulation by the pancreatic hormones, had shown the system tobe unaffected by either insulin or glucagon (27). The persistence of the en[Insulin], -109 M hanced transport activity in culture arguesagainst ascribing FIG. 5. The effect of insulin treatmentin vitro on the uptake the effect to trans-stimulationas discussed above for System of AIB. Isolated hepatocytes were placed in culture in the presence of the indicated concentrations of insulin for 2 to 3 h. The Na+dependent uptake of 50 PM AIB was assayed for 1 min at 37 "C. The rates of uptake in the absence of the hormone were 8.9 +- 1.9 and 60 7.3 pmol.mg" of protein.min" for the control and experimental cells, respectively. Although they are different in absolute magnitude, both scales shown on the ordinates reflectthe percentage of transport when compared to the rate observed for the cells from the diabetic donor in the absence of insulin. The decrease in transport by the diabetic cells after treatment with lo-' M was significant at the p < 0.01 level. The results are the averages k S.D. of four individual assays.

$ U

\

1

loa

with glucagon a t levels as low as lo-" M (7.5 versus 10.3pmol0 mg"of proteinemin"), whereas the hepatocytes from the diabetic rats did not show an increase in System A uptake activity at hormone concentrations below lo-' M. At a glucaa gon concentration of lo-' M the uptake of the experimental Hours cells was enhanced by only 15% (26.9 uersus 30.8 pmol.mg-' FIG. 6. Transport activity of System ASC at specific interof protein rnin"), while that of the normal hepatocytes was vals after establishment of hepatocyte primary cultures.Cells taken from normal or diabetic rats were used to monitor the Na'stimulated by more than2-fold (7.5 uersus 18.6 pmol, mg" of proteinamin"). The results suggest that if the cells isolated dependent transport of 50 PM cysteine (CysH), a system-specific from the diabetic rats can respond to glucagon at all, the substrate for the hepatic System ASC. In these assays 10 mM dithiothreitol was included to keep the cysteine in the reduced state. The effectiveness of the hormone is decreased significantly. data are shown as the averages +- S.D. of at least three determinations. System ASC Activity in Culture-The transport activityof System ASC was monitored for the 24 h immediately after establishment of primary cultures of hepatocytes isolated from normal and diabetic rats. The Na'-dependent uptake of T cysteine, a System ASC-specific probe in rat hepatocytes (23), was measured ina manner similar to that used to follow System A as described in the legend to Fig. 1. The transport of cysteine was significantly higher in the cells isolated from the streptozotocin-treated animals after 1 to 2 h of culture (Fig. 6 ) . The increased transport by these cells declined during the initial 2-h period such that after3 h in culture there was no difference in the uptake rate between the two cell types. After 24 h in culture, the transport of cysteine by the hepatocytes from the diabetic rats was actually less than that of the normalcells (Fig. 6), a marked contrast to the observation made for System A activity (Fig. 1).System ASC is known to be sensitive to trans-stimulation (25) and the enhanced uptake by the experimental cells may be the result of differences in Hours the amino acid pools of the two cell populations. Samson et FIG. 7 . Changes in the System N activity during the initial al. (12) have also reported a slight increase in the MeAIB24 h of culture forcells from normalor diabetic animals. After insensitive portion (ie. System ASC) of alanine uptake by establishment of primary cultures of rat hepatocytes, the Na'-defreshly isolated suspensionsof hepatocytes from diabetic rats. pendent uptake of 50 PM histidine (His) was measured for 1 min at The lack of an effect in the cells of the diabetic rats, other 37 "C at the times indicated. The data are reported as the averages than that producedby intracellular pools, would support the +- S.D. of four assays.

i

-

Amino Acid Transport in

N Nm . KKl lPP

+llmM

Y.Als

Hepatocytes from Diabetic Rats

0

CholKlP

FIG. 8. Uptake of System N substrates in the presence of MeAIB. The Na+-dependent uptakeof either glutamine or histidine, at a substrate concentration of 50 ELM,was measured in the presence or absence of 25mM MeAIB. The 30-s assays were performed in the appropriate buffers as shown. Choline chloride (12.5mM) was added to the assays lacking MeAIB to compensate for the osmotic effect of the inhibitor. The results are the averages f S.D. of triplicate determinations. KRP, Krebs-Ringer phosphate buffer.

TABLEI11 Stimulation of amino acid transport in normal hepatocytes by insulin or glucagon

Hepatocytes from normal rats were maintained in primary culture to the hormones.After this forseveralhourspriortoexposure equilibrationperiod, the platingmedium was changed toKrebsRinger bicarbonate with or without the indicated hormones each at a concentration of 10" M. For the first 3 h, the buffer also contained 5 mM AIBto keep the System A activity in the repressed state. During the last 1h of incubation inthe presence of the hormones, the AIBwas omitted to eliminate possible trans effects. The Na'-dependent uptake of each of the amino acids indicated, at a substrate concentration of 50 PM, was measured for 30 s (histidine and glutamine) or 1 min (AIB and MeAIB) at 37 "C. The results given are from two separate trials, the first involving the uptake of AIB and histidine and the second testing the uptake of MeAIB and glutamine.

14965

thediabeticrats (Fig. 8). Although a slightinhibition of glutamine uptake by these cells was seen in the presence of MeAIB, the majorityof the transport activitywas insensitive to the inhibitor and presumed, therefore, to represent System N-mediated transport. System N Activity inHormone-treated Cells-As mentioned above, our earlier work had suggested that System N was not responsive to the pancreatic hormones, insulin and glucagon. The data depicted in Table I11 c o n f i i t h erelative inertness of this agency to these two hormones. In the fist experiment, the uptake of glutamine was tested after normal hepatocytes hadbeen incubated in thepresence of maximally stimulating levels of insulin or glucagon (lop7M) for 4 h. Neither hormone increased the transport of this amino acid, although insulin was c o n f i i e d t o be active on System A (Table 111).In a second experiment, the uptakeof either AIB or histidine was assayed after treatment of the cells with the combination of dexamethasone and glucagon. These two hormones have been shown to act on cultured hepatocytesin a synergistic manner to produce greater stimulation of System A activity than that produced by glucagon alone (26). The combination of dexamethasone plus glucagon was effective in stimulating AIB transport, but not that of histidine (Table

111). DISCUSSION

Our interest ir. studying the regulation of hepatic amino acid transport in diabetic animals stems from the hypothesis that enhanced transport of the amino acids into the hepatocyte would result in an increased supply of gluconeogenic precursors. This elevation of gluconeogenic substrates, particularly in combination with glucagon stimulation of one or more of the key enzymes inthe gluconeogenic pathway, would increase further the already uncontrollablehyperglycemia associated with insulin deficiency. The data presented demonstrate the effect of experimental diabetes on the hepatic Substrate under test, 50 p~ transport of both nonmetabolizable amino acids and naturally Hormone added occurring ones such as alanine, cysteine, glutamine, and hisAIB Histidine MeAIB Glutamine tidine. All three of the Na+-dependent systems tested, A, ASC, pmol.mg"protein.unit time" and N, were found to be altered in some fashion by the None 28.4 c 3.2 266 f 21 3.6 f152 1.8 f7 streptozotocin treatment, although we believe that the in14.7 Insulin f 2.4 98 f 6 crease in uptake by System ASC is the result of trans effects Glucagon 128 f 27 Dexamethasone 87.1 f 3.7 298 f 30 rather than induction of the synthesis of proteins necessary plus glucagon for that activity. Note that the difference in the decay rates for Systems ASC and N can be taken as additional evidence of the uniquenessof these two agencies. ASC. Furthermore, trans effects are independent of protein Although hyperglucagonemia has been proposed as one of synthesis and thedecay of System N in the hepatocytes from the major manifestations of the diabetic state (10) and it is the diabetic rats can effectively be blocked by includingcyclo- well known that glucagon has the ability to stimulate System heximide in the culture medium.' The latter observation is A transport in liver tissue (26), it remains to be established reminiscent of the cycloheximide protection of System A that the changes in System A activity seenin the cells isolated activity after stimulation byglucagon in vitro (29). from the diabetic ratsis the direct resultof elevated levels of Inhibition of System N Substrates by MeAIB-To elimi- plasma glucagon. Based on the kinetics of AIB uptake, the nate the possibility that the enhanced uptakeof histidine by work of Freychetandhiscollaborators suggests that the the hepatocytes isolated from the diabetic rats was the result stimulation of AIB uptake by exogenously added glucagon of transport of the amino acid by System A, the uptake of occurs by a different process than does the transport stimueither glutamine or histidine was measured in the presence lated after streptozotocin treatment. Using freshly isolated and absence of a large excess of MeAIB. Fig. 8 shows the hepatocytes as cell suspensions they reported that exogesignificant elevation in the transport of both of these amino nously added glucagon stimulated the synthesis of a previacids by the cells from the diabetic animals. In the presence ously undetectable component having a K,,, for AIB of about of 25 mM MeAIB, the uptake of neither amino acid, a t a 1 mM (30), whereas hepatocytes isolated from diabetic rats substrate concentrationof 50 PM, was inhibited for the normalexhibited increased transport due to stimulation of a low cells, c o n f i i i n g t h especificity of these substrates to System affinity system having a K,,, for AIB of more than30 mM (12). N. The uptake of histidine was similarly unaffected by the These authors have recently shown that during culture, the System A-specific substrate in the hepatocytes isolated from low affinity system is lost and consequently only the high affinity component can be detected (20). M. E. Handlogten and M. S. Kilberg, unpublished observations. Our results confirm the observationby Freychet and his co-

14966


workers that only the high affinity agency for AIB uptake can be detected in primary cultures of rat hepatocytes, but are in conflict with the data of those investigators with respect to which of these two systems for AIBtransport is stimulated by streptozotocin treatment. Our studies indicate that the high affinity agency is elevated after induction of the diabetic state, whereas those of Samson et UI!. (12) suggest that in freshly isolated hepatocytes from diabetic rats thelow affinity agency is enhanced and the high affinity system is not even detectable. Furthermore, we find that alanine transport by the cells isolated from the diabetic donors also occurs by a process with a K , for the amino acid of less than 5 mM (Table I). Hepatocytes isolated from adrenalectomized rats show a decreased sensitivity to glucagon (16, 31)and glucocorticoids exhibit a permissive effect with regard to glucagon induction of AIB transport in cultured hepatocytes (21, 32).Consistent with these findings are the present results which suggest a possible role for the glucocorticoids in the development of stimulated transport in the liver tissue of diabetics. Although streptozotocin treatment alone caused a large increase in the transport capacity ( Vmax/K,,,) of System A, if the animals had been subjected to bilateral adrenalectomy prior to treatment with the drug, the K,,, and the V,,, changed in an inverse relation such that there was little or no net effect onthe V,,,,,/ K,,, ratio (Table I). One of the postulated mechanisms by which glucagon might cause stimulation of transport is by increasing the magnitude of the membrane potential (33). Changes in the flux of both Na+ and K+ are known to occur immediately after exposure of liver to glucagon (34). An argument against this proposal is that a hormone-mediated increase in the membrane potential should, in theory, enhance all of the electrogenic Na+-dependent transport systems, not just System A as is seen experimentally. An alternate possibility is that hormone treatment causes a decrease in the K,,, for Na+ uptake by System A only, the net result being a stimulation of amino acid uptake by that particular agency. Our results indicate that rather than a measurable decrease in the K , for Na+ after induction of the diabetic state, the estimated K , for transport of this ion by System A was actually increased by nearly 2-fold. Hence the stimulation of transport by System A in the hepatocytes isolated from the diabetic animals does not appear to be the result of changes in the interaction of Na+ with that system. The System A activity in the hepatocytes isolated from the diabetic rats did increase in a protein synthesis-dependent fashion after the cells had been starved for amino acids. That response indicates that the transport system was not stimulated to amaximal level.Stimulation of the system with high concentrations of insulin also supports this proposal. In contrast, when the cells from the diabetic rats were exposed to glucagon, no stimulation of transport occurred even though the hormone was shown to be effective on control cells. The inability of the cells to respond to glucagon mayresult because the hepatocytes are refractory to this particular stimulus. This lack of stimulation by glucagon may arise from a shortage of receptors on the surface of the plasma membranes of these cells, chronic exposure to elevated plasma glucagon levels may cause the loss of surface receptors in the liver by downregulation (12, 35). In contrast to glucagon, insulin administration invitro elicited a biphasic response in hepatocytes isolated from diabetic rats. The first phase, occurring at low levels of hormone, results in a partialreversal of the diabetes-induced stimulation of transport, while higher concentrations of insulin cause a further increase in the uptake by this system. Administration of insulin in vivo had been shown previously to reverse the stimulation of hepatic amino acid transport in diabetic rats

(11,12),but the present results represent the f i s t demonstration that insulin can act directly on the liver to counteract the effect of diabetes on transport. Other investigators have shown that insulin treatment in vitro can reverse the diabetes-induced rates of hepatic gluconeogenesis, ureogenesis,ketogenesis, and release of amino acids (3, 36).The observation that insulin can cause the reversal of the diabetes-induced transport is important because several laboratories, including our own: have found that insulin does not antagonize the stimulation of System A by glucagon when both hormones are added in vitro (13, 37). These results suggest that the treatment of isolated hepatocytes with glucagon may not represent a good model for studies on diabetes-induced transport. Further evidence that streptozotocin-treatment produces alterations in hepatic amino acid transport not mimicked by the in vitro addition of glucagon is the finding that System N activity is stimulated in the hepatocytes isolated from the diabetic rats, but is relatively inert to stimulation by glucagon added in vitro. In summary, our results demonstrate the usefulness of primary cultures of rat hepatocytes as a model system to study the effects of experimental diabetes on amino acid transport. The retention of enhanced transport activity by Systems A and N for several hours after cultures are initiated allows sufficienttime to study the regulation of these systems and their role in governing hepatic metabolism in both the normal and the disease state. Further work will be necessary to establish the magnitude of the impact of the stimulated amino acid transport on hepatic metabolism in the diabetic animal. REFERENCES 1. Mallette, L. E., Exton, J. H., and Park, C.R. (1969)J. Biol. Chem.

244,5713-5723 2. Claus. T. H.. Pilkis, S. J., and Park, C.R. (1975)Biochim. Bionhvs. -r - -" Acia 404,110-123 3. Claus, T. H., and Pilkis, S. J. (1976) Biochim. Biophys. Acta 421, 246-262 4. Claus, T. H., and Pilkis, S. J. (1981) in Biochemical Actions of Hormones (Litwack, G.,ed) Vol. 8, pp. 209-271,Academic Press, New York 5. Exton, J. H., Mallette, L. E., Jefferson, L. S., Wong, E. H. A., Friedmann, N., Miller, T. B.,Jr., and Park, C.R. (1970) Recent Prog. Horm. Res. 26,411-457 6. Sips, H. J., Groen, A. K., and Tager, J. M. (1980) FEBS Lett. 119,271-274 7. McGivan, J. D., Ramsell, J . C., and Lacey, J . H. (1981) Biochim. Biophys. Acta 644, 295-304 8. Pariza, M. W., Butcher, F. R., Kletzien, R. F., Becker, J. E., and Potter, V. R. (1976) Proc. Natl. Acad. Sci. U.S. A . 73, 45114515 9. LeCam, A,, and Freychet, P. (1976) Biochem. Biophys.Res. Commun. 72,893-901 10. Unger, R. H. (1978) Metab. Clin. Exp. 27, 1691-1706 11. Kilberg, M. S., and Neuhaus, 0. W. (1977) J. Supramol. Struct. 6,191-204 12. Samson, M., Fehlmann, M., Dolais-Kitabgi, J., and Freychet, P. (1980) Diabetes 29, 996-1000 13. LeCam, A,, and Freychet, P. (1978) Diabetologia 15, 117-123 14. Feteris, W. A. (1965) Am. J. Med. Technol. 31.17-21 15. Weissbach, L., Handlogten, M. E., Christensen, H. N., and Kilberg, M. S. (1982) J. Biol. Chem. 257, 12006-12011 16.Kilberg,M. S., Vida, T. A., andBarber, E. F. (1982) J. Cell. Physiol., in press 17. Gazzola, G. C., Dall'Asta, V., Franchi-Gazzola, R., and White, M. F. (1981) Anal. Biochem. 115,368-373 18. Handlogten, M. E., Weissbach, L., andKilberg,M. S. (1982) Biochem. Biophys. Res. Commun. 104,307-313 19. Kletzien, R. F., Pariza, M. W., Becker, J. E., and Potter, V. R. (1975) Anal. Biochem. 68,537-544 20.Morin,O., Fehlmann, M., and Freychet, P. (1982) Mol. Cell. Endocrinol. 25,339-352

Amino Acid Transport

in Hepatocytes from Diabetic Rats

21. Kelley, D. S., Campbell, H. A., and Potter, V. R. (1982) J. Cell, Physiol. 112,67-75 22. Edmondson, J. W., Lumeng, L., and Li, T.-K. (1979) J. Biol. Chem. 2S4, 1653-1658 23. Kilberg, M. S., Handlogten, M. E., and Christensen, H. N. (1981) J. Biol. Chem. 256,3304-3312 24. Kelley, D. S., and Potter, V. R. (1978) J. Biol. Chem. 253, 90099017 25. Thomas, E. L., and Christensen, H. N. (1971)J.Biol. Chem. 246, 34. 1682-1688 26. Kilberg, M. S. (1982) J. Membr. Biol. 69, 1-12 27. Kilberg, M. S., Handlogten, M. E., and Christensen, H. N. (1980) J. Biol. Chem. 255,4011-4019 28. Handlogten, M. E., Kilberg, M. S., and Christensen, H. N. (1982) J. Biol. Chem. 257,345-348 29. Pariza, M. W., Butcher, F. R., Kletzien, R. F., Becker, J. E., and Potter, V. R. (1976) Proc. Natl. Acad. Sci. U. S. A . 73, 4511-

14967

4515 30. Fehlmann, M., Le Cam, A., and Freychet, P. (1979)J. Biol. Chem. 254, 10431-10437 31. Dolais-Kitabgi, J., Faure, T., and LeCam, A. (1981) Cell. Mol. Endocrinol. 24,315-323 32. Kletzien, R. F., Pariza, M. W., Becker, J. E., and Potter, V. R. (1975) Nature 256,46-47 33. Edmondson, J. W., and Lumeng, L. (1980) Biochem.Biophys. Res. Commun. 96.61-68 Friedmann, N., and Dambach, G. (1980)Biophys. Biochim. Acta 596,180-185 35. Meier, J. M., McGarry, J. D., Faloona, G. R., Unger, R.H., and Foster, D. W. (1972) J. Lipid Res. 13, 228-233 36. Lombardo, Y. B., Hron, W. T., andMenahan, L. A. (1978) Diabetologia 14.47-51 37. Kelley, D. S., Shull, J. D., and Potter, V. R. (1980)J. Cell. Physiol. 103, 159-168