Vol. 139, No. 2, 1986

September 16, 1986

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Pages 612-618

CALCIUM UPTAKE BY INTRACELLULAR COMPARTMENTS IN PERMEABILISED ENTEROCYTES

EFFECT OF INOSITOL 1,4,5 TRISPHOSPHATE

Gloria Velasco I, Stephen B. Shears 2 , Robert H. Michell 2

and Pedro S. Lazo I

i Departamento de Bioquimica, Universidad de Oviedo, 33071 Oviedo, Spain

2 Department of Biochemistry, University of Birmingham, Birmingham BI5 2TT, UK

Received July 29, 1986

Treatment of rat small intestine with EDTA produced isolated enterocytes with plasma membranes which were permeable to small ions. When resuspended in a medium designed to resemble the intracellular medium, Ca 2+ was accumulated into the cells. Both mitoehondrial and a non-mitochondrial (presumably endoplasmic reticulum) compartments were responsible for sequestering the cation, as indicated by the effects of the mitochondrial inhibitors oligomycin and antimycin and of the Ca-ATPase inhibitor sodium orthovanadate assayed at low (0.9 pM) and high ( 12 ~M) free Ca 2+ concentrations. Addition of inositol (1,4,5) trisphosphate induced a rapid release of Ca 2+ from the non mitochondrial compartment. The effect of inositol trisphosphate was concentration dependent and showed 50% of maximal release at 2 M. Neither cyclic AMP nor dibutryl cyclic AMP caused release of Ca 2+. These findings lend novel support to the possibility that Ca-mediated control of ionic transport in the small intestine is exerted through the phosphatidylinositol-protein kinase C transduction mechanism. © 1986 Academic Press, Inc.

Intestinal secretion of Na~ CI- and water is considered to be

controlled by changes in intracellular concentrations of both cyclic AMP and

Ca since a number of secretagoEues (VIP, cholera toxin, prostaElandin E I) act

by increasing the intracellular concentration of cyclic AMP while others

(carbachol,serotonin or substance P) are dependent upon extracellular C~ +

(for a review see I). Agonists such as acetylcholine and serotonin,reEulate

secretion in other tissues by interacting with membrane receptors. Upon

interaction of the effector with the plasma membrane, polyphosphoinositides

are degraded by a phospholipase producing Ins (1,4,5)P 3 and diacylglycerol

(2,3). Diacylglycerol appears to be implicated in the activation of protein

kinase C (4) while Ins(l,4,5)P 3 seems to mobilise Ca 2+ from intracellular

pools (5) thereby triggering Ca-dependent intracellular responses. Thus, it

Abbreviations. Ins(l,4,5)P :inositol(l,4,5)trisphosphate; EDTA:ethilene- diaminetetraaeetie acid; EGTA: ethilene glycol bis ( 6 -aminoethylether)- N,N,N',N', tetraacetic acid.

0006-291 X/86 $1.50 Copyright © 1986 by Academic Press, Inc. All rights of reproduction in any form reserved. 612

Vol, 139, No. 2, 1986 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

was important to know if this transduction mechanism is involved in the Ca

mediated ionic secretion in the small intestine.

We have used isolated enterocytes which have been shown to have a

permeabilised plasma membrane to study the uptake of Ca 2+ by intracellular

compartments. We also show that Ins(l,4,5)P$ induces Ca 2+ release from the

non- mitochondrial compartment. This findin~ strongly suggest that the

Ca-mediated control of ionic secretion in the small intestine is exerted

through the phosphatidylinositol-protein kinase C transduction mechanism.

Part of this work has been presented in preliminary form (6).

MATERIALS AND METHODS

Materials ATP, cyclic AMP, dibutiryl cyclic AMP, sodium orthovanadate,

oligomycin, antimycin, phosphocreatine, A~phosphocreatine kinase and A23187 were purchased from Sigma Chemical Co. ~CaCI 2 was obtained from Amersham. Ins(l,4,5)P 3 was prepared from human erythrocytes according to Downes et al. (7) and assayed for total phosphate (inorganic plus organic) according to Ames (8). This value was corrected for inorganic phosphate (9).

Isolation of enterocytes

Enterocytes from rat small intestine were isolated by the followin~ modification of the method of Watford et al (i0).

Two buffers were used: (A) the medium of Krebs and Henseleit (ii) from which CaCI 2 was omitted, (B) the same to which 0,25% dialysed serum albumin and 5 mM EDTA were added. Rats were decapitated and the small intestine was removed. The lumen was rinsed with 0.15 M NaCI plus i mM EDTA. Then, the intestine was filled with buffer B and the ends ligated, incubated with shakinE at 37~ C for 15 min in a flask containing buffer A. The intestine was then opened drained and washed with ice-cold buffer B. It was refilled with buffer B and patted with finEer tips on ice. Cells were released into the lumen and collected into plastic tubes. They were then washed and resuspended in the appropriate incubation buffer.

Assay of Ca uptake

Enterocytes isolated from rat small intestine as described were resuspended at concentrations of 2-5 x 106 cells/ml in a medium having the following composition (in mM): Hepes, 20; KCI, 150; MES04, i; EGTA, 0.2 and the concentration of CaCI 2 required to obtain the desired free Ca 2+ concentration. The pH was 7.0. 45CACI (600-900 mCi/ml) was added at a concentration of 5 ~Ci/ml. When required, I mM ATP, 6 mM phosphocreatine and 8 U/ml of phosphocreatinekinase were also added. A231287 was added in dimethyl sulfoxide and mitochondrial inhibitors as ethanolic solutions. In both cases the final concentration of the solvent never exceeded 0.1% (v/v). The cells were incubated at 37~ C and when required 0.i ml aliquots were withdrawn, diluted 20 fold with 20 mM Hepes-NaOH pH 7.0 containing 0.3 M mannitol and 1.5 mM EGTA and filtered through GF/C filters. The filters were washed with 2 ml of the same buffer and after dryinE, the radioactivity determined by liquid scintillation counting. The cells were always used within 30 minutes of their preparation.

Determination of free Ca 2+ concentrations

The concentration of ionic species in C 2+- EGTA buffers were calculated with a computer program accordin~ to Fabiato and Fabiato (12) Logarithms of the apparent association constants at pH 7.0 and 370C for CaEGTA,CaATP, MEEGTA and MyATP were 6.28, 3.70, 1.60 and 4.00 respectively.

613

Vol. 139, No. 2, 1986 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

RESULTS

Ca 2+ uptake by intracellular compartments in permeabilized enterocytes

Enterocytes isolated by treatment of the intestine with EDTA

appear permeabilized to Ca2+ and other ions, while enterocytes isolated by

treatment of the intestine with hyaluronidase appear to retain their original

permeability properties. Thus, the former do not accumulate a-methyl

Flucoside, the transport of which is coupled to the Na + gradient, while the

latter accumulate the sugar 30-40 fold over the medium concentration (Velasco

et al., in preparation). Also while cells isolated by hyaluronidase treatment

of the intestine release Ca 2+ into the medium, cells isolated by EDTA

treatment accumulate Ca 2+. 45

Fig 1 shows the time course of Ca uptake by permeabilised 2+

enterocytes isolated by EDTA treatment of the intestine at two different Ca

concentrations. It can be observed that in both cases the uptake was strongly

stimulated by ATP. This indicates that intracellular compartments are

responsible for requesterin~ the cation, since Ca 2+ homeostasis is controlled

by Ca-ATPases and other transport systems associated to or~anelle membranes

and plasma membranes. The former accumulate cytosolic C~ + , whilst the latter

pump Ca 2+ out of the cell. When cells were suspended in a medium containinF

0.9 M free Ca 2+, sodium orthovanadate (an inhibitor of Ca-ATPase) reduced

Ca 2+ uptake more than 90% while the mitochondrial inhibitors oli~omycin and

antimicyn caused less than 5% inhibition of the uptake (Fi~. IA). On the

other hand at 12 M free Ca 2+ both orthovanadate and the mitochondrial

inhibitors caused approximately 50% inhibition. Thus, the intracellular

1.5

w

,o o 1.0

+

3 0.5

¢=

A 8

8 o

/,

c

10 20 Time ( mi~',

B

10 20

Time (min}

Figure 1.45Ca uptake by permeabilised enterocytes. Enterocytes were isolated as described in Methods anda5Ca uptake

was assayed at 0.9 UM free Ca 2+ (0.13 mM CaCI2 added to the incubation medium) (A) and at 12 ~M free Ca 2+ (0.22 mM CaCI 2 added) (B). The control incubations (0] contained no ATP. To the three other series, ATP and the ATP regenerating system were added. Oligomycin (5 ~g/ml) and antimycin (6 ~/ml) was added to (A) and 1 mM Na orthovanadate was added to (A). A representative experiment is shown.

614

Vol. 139, No. 2, 1986 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

1,5

i/I

1"0 o

0 .5 o ¢;

A

I I

tn ~u w

÷

o E

10 20

T ime ( rn in ) Time (rain)

B

------o o o o L~

10 20

Figure 2.1ns(l,4,5)P~-mediated release of C~+from permeabilised enterocytes. (A) 4 ~a uptake was assayed at 0.9 ~ M free Ca 2+ . ATP and the ATP

regenerating system was added to two series (O,A) and,.~hen indicated by the arrow, lns(l,4,5)P~ was added to one of them (&). (B)4~a uptake was assayed at 12 ~M free Ca ~+ in the absence of inhibitors(e), in the presence of mitoehondrial inhibitors (A) and in the presence of Na orthovanadate (~). When indicated by the arrow, Ims(l,4,5)P 3 (6.25 ~M) was added to the three series (e,A,~).No additions were made to the control series (o). A repre- sentative experiment is shown.

compartments sequesterinE Ca 2+ in an ATP dependent manner very likely

corresponds to endoplasmic reticulum (hiEh affinity) and mitochondria (low

affinity). In all the conditions tested and in the presence of the calcium

ionophore A23187, no Ca 2+ uptake was observed. AnaloEously, when the

ionophore was added to cells which were actively sequesterinE Ca 2+ , 100% of

the accumulated cation was immediately released (not sohwn).

Release of accumulated Ca 2+by Ins(l,4,5)P 3

Ins(l,4,5)P 3 has been shown to mediate Ca 2+ release from

endoplasmic reticulum in a variety of tissues (12). The findin~ that

enterocytes isolated by EDTA treatment of the intestine have a permeabilised

plasma membrane has permitted us to test wether Ins(l,4,5)P 3 releases Ca 2+

from intracellular stores. FiE 2 shows that when Ins(l,4,5)P 3 was added to

permeabilized enterooytes in which orEanelles were actively accumulatinE

Ca 2+, a rapid release of the cation, was induced. When cells were resuspended

in a medium with 0.9 ~M free Ca 2+ ,either in the presence or in the absence

of mitochondrial inhibitors, the effect of Ins(l,4,5)P 3 was clearly observed

(FiE 2A). When the experiment was carried out at 12 ~M Ca~+,Ims(l,4,5)~ also

induced a release of the accumulated Ca. The release was also observed in the

presence of mitochondrial inhibitors, but not when cells were incubated in

the presence of sodium orthovanadate (FiE 2B).

The effect of Ins(l,4,5)P 3 was concentration dependent for

concentrations lower than 10 ~M, with a 50% of maximum effect beinE obtained

at a concentration of about 2 ~M (Fi~ 3).

615

Vol. 139, No. 2, 1986 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

60

30

..... I I

10 20

[ns(1,~,5)P3(pH)

Figure 3.Concentration dependence of Ins(l,4,5)P3 -induced Ca2+release from permeabilised enterocytes.

4~Ca uptake was assayed in the presence of ATP and ATP regenerating system at 0.9 ~M free Ca 2+. When the steady state was reached, Ins(l,4,5)P3 at the indicated concentrations was added and the release of Ca 2+ determined. The release is depicted as a percentage of total ATP-dependent Ca 2+ uptake. Data are derived from 3 experiments.

Lack of effect of cyclic AMP on Ca 2+ stores

Cyclic AMP has been shown to induce the release of Ca 2+ from

mitochondria (14) and it has been suEEested that this nucleotide mediates

intestinal secretion by releasinE Ca 2+ from intracellular stores which, in

turn, would act by chanEin~ the plasma membrane conductance to ions (15). We

have tested this hypothesis usin~ permeabilised enterocytes. Neither cyclic

AMP nor its dibutyril derivative induced Ca 2+ release from intracellular

compartments in any of the conditions used for the experiments reported here,

i.e. either with mitochondrial or non-mitochondrial pools beinE mainly

responsible for the uptake of Ca2+(not shown).

DISCUSSION

A variety of hormones and neurotransmitters utilize an inositol

lipid dependent signal transduction system. The production of the second

messengers Ins(l,4,5)~ and 1,2-diacyl~lycerol, and the resultant increase in

cytosolic free Ca 2+ and protein kinase C activity, leads to cell activation

(2-4). Much evidence indicates that intracellular Ca 2+ plays an important

role in the reEulation of intestinal ion transport (i). However, no data are

yet available on intracellular levels of Ca 2+ in enterocytes and it is not

known if the neurohumoral substances which induce a Ca 2+ mediated secretion

stimulate the turnover of inositol lipids. The intracellular Ca 2+ selective

indicator Quin 2 (16) has been used to determine intracellular Ca 2+

concentrations in many cells. Quin 2 measurements of intracellular Ca 2+ were

in our hands not possible because of rapid extracellular hydrolysis of the

ester Quin 2AH and because Quin 2 was not retained within the cells even when

616

Vol. 139, No. 2, 1986 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

they were prepared with hyaluronidase. It is therefore important that

treatment of the intestine with EDTA produces enterocytes with permeabilised

plasma membranes. This has allowed us to study the effect of exogenous

Ins(l,4,5)P 3 on intracellular stores responsible for sequestering Ca 2+ .

The data demonstrate that enterocytes contain a mitochondrial and

non-mitochondrial system which accumulate Ca 2+ in an ATP dependent manner.

Ins(l,4,5)P3-induced Ca 2+ release was observed in the presence of oligomycin

plus antimycin, but was abolished by vanadate (Fig.2). Therefore, Ins(l,4,5)P 3

released Ca 2+ from the non-mitochondrial rather than the mitochondrial C~ +

pool. It is interesting that in the absence of inhibitors Ca 2+ release was

even observed at 12 ~M Ca 2+ when non-mitochondrial Ca 2+ uptake was around 50%

of the total (Fig. 2). This contrast with data obtained from permeabilised

hepatecytes when Ca 2+ concentrations were increased so that non-mitochondrial

Ca 2+ uptake was only 10% of the total (17). In the latter case, Ins(l,4,5)~

-induced Ca 2+ release was two small to be detected (19). The effect of

Ins(l,4,5)P 3 was concentration dependent, with half maximal release being

obtained at 2 ~M. This value is somewhat higher than the value found in other

cell types, where it ranges from 0.i to 1.0 ~M, with the only known exception

beinK insulinoma microsomes in which a value of 3 ~M was found (13). These

results provide novel evidence that an inositol lipid-dependent signal

transduction system is operating in epithelial cells of the small intestine

indicating that not only exocytotic secretion but also ionic transport across

the plasma membrane can be under control of this transduction system. The

finding that enterocytes contain protein kinase C (Velasco et al.,

unpublished data) and the fact that phorbol esters induce anion secretion

(19) lend support to this possibility. Ca 2+ release from endoplasmic

reticulum might regulate ionic transport either through protein kinase C or

by complexing with ealmodulin which is thought to be involved in the control

of NaCI absorption (20). The use of permeabilised cells will permit studies

on the contribution of intracellular organelles to Ca 2+ homeostasis in the

small intestine and its control by a variety of effectors. In this regard,

further research has to be carried out to know whether secretagoHues which

induce a Ca 2+ mediated secretion stimulate the formation of Ins(l,4,5)P 3.

There are some indications that cyclic AMP and Ca 2+ may interact

in the control on ionic transport in the mammalian intestine (1,15) althouyh

it is not certain which aspect of Ca 2+ homeostasis may be affected by cyclic

AMP. The observation that neither cyclic AMP nor its dibutyryl derivative

induce Ca 2+ release from intracellular stores suggest that any possible

interaction between the two second messengers should occur at a different

level than at the control of intracellular Ca 2+ pools.

617

Vol. 139, No. 2, 1986 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Acknowledgements. We are ~rateful to Dr. F. Barros for his criticism. Thanks to the British Council and the M.E.C. (Spain) for granting an Acci6n Inte~rada and the M.R.C.(UK) and the F.I.S. and C.A.I.C.Y.T.(Spain) for financial support. G.V. was a recipient of a fellowship from the F.I.S.

REFERENCES

i. Don owitz, M. and Welsh, M.J. (1986) Ann. Rev. Physiol. 48, 135-150. 2. Berrid~e, M.J. (1984) Biochem. J. 220, 345-360. 3. Downes, C.P. and Miehell, R.H. (1985) In "Molecular Mechanisms of

Transmembrane Signalling" (Cohen, P. and Houslay, M.D., eds.) pp. 3-56, Elsevier, Amsterdam.

4. Nishisuka, Y. (1984). Science 225, 1365-1369. 5. Streb, H., Irvine, R.E., Berridge, M.J. and Schulz, I. (1983) Nature

306, 67-69 6. Velasco, G., Shears, S.B., Miehell, R.H. and Lazo, P.S. (1986) Biochem.

Soo. Transactions. In press. 7. Downes, C.P., Mussat, M.C. and Michell, R.H. (1982) Biochem. J. 203,

169-177. 8. Ames, B.N. (1966) Meth. Enzymol. 8, 115-118. 9. Black, M.J. and Jones, M.E. (19857 Anal. Biochem. 135, 233-238.

i0. Warlord, M., Lund, P. and Krebs, H.A. (1979) Biochem. J. 178, 589-596. ii. Krebs, H.A. and Henseleit, K. (1932) Hoppe-Seyler's, Z. Physiol. Chem.

210, 33-36. 12. Fabiato, M. and Fabiato, F. (1979) J. Physiol. 75, 463-505. 13. Berridge, M.J. and Irvine, R.F. (1984) Nature 312, 315-321. 14. Arshad, J.H. and Holdsworth, E.S. (1980) J. Memb. Biol. 5~7, 207-212. 15. Fondaearo, J.D. (1986) Am. J. Physiol. 250, 61-68 16. Tsien, R.V., Pozzan, T. and Rink, T.J. (i982) J. Cell. Biol. 94,

325-334. 17. Burgess, G.M., Mckinney, J.S., Fabiato, A., Leslie, B.A. and Putney,

J.W. (1983) J. Biol. Chem. 258, 15336-15395 18. Burgess, G.M., Godfrey, P.P., Mckinney, J.S., Berrid~e, M.J., Irvine,

R.F., Putney, J.W. (1984) Nature 309, 63-66. 19. Fondacaro, J.D. and Henderson, L.S. (1985) Am. J. Physiol. 249,

G422-G426. 20. Donowitz, M. (1983) Am. J. Phys io l . 245, G165-G177.

618