- Email: [email protected]
An inositol phosphoglycan from Trypanosoma cruzi inhibits ACTH action in calf adrenocortical cells
Cellular Signalling Vol. 7, No. 4, pp. 331-339, 1995. Copyright © 1995 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0898~5568/95 $9.50 + 0.00
Pergamon 0898-6568(95)00008-9
AN INOSITOL
PHOSPHOGLYCAN ACTION
FROM
TRYPANOSOMA
IN CALF ADRENOCORTICAL
CRUZI INHIBITS
ACTH
CELLS
M A R I A DEL C. VILA,* E D U A R D O N. COZZA,* C A R L O S L I M A , t M A R I A I. R A M I R E Z t and R O S A M. D E . L E D E R K R E M E R t *Departamento de Qufmica Bioldgica and tDepartamento de Qu/mica Org~inica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Argentina (Received 9 December 1994; and accepted 23 December 1994) A b s t r a c t - - W e describe the effect of an inositol phosphoglycan (IPG) purified from Trypanosoma cruzi on the stimulation of aldosterone and cAMP production by ACTH in calf adrenocortical cells. T. cruzi IPG has two galactofuranose residues (Galf) which are not frequent in other IPGs. The effect of IPG with galactofuranose residues (IPG Galf) and IPG without these residues (IPG) was investigated. It was found that IPG Galf slightly decreased the stimulation of aldosterone and cAMP production by ACTH, whereas IPG significantly inhibited ACTH-mediated accumulation of both aldosterone and cAMP. The inhibition of aldosterone content in ACTH-treated cells by IPG was dose dependent. It was also found that the pretreatment of calf adrenocortical cells with IPG inhibited the accumulation of aldosterone provoked by ACTH and dibutyryladenosine-3',5'-cyclic monophosphate (db-cAMP). On the other hand, the activation of a GPI (glycosyl phosphatidylinositol)-phospholipaseC by ACTH was evaluated. First it was found that the release of ceramide from a GPI-like molecule: a glycoinositol-phosphoceramide (LPPG) purified from T. cruzi is increased in ACTH-treated cells. Second, the release of alkaline phosphatase, a GPI-anchored enzyme, to the extracellular medium was increased in these cells by ACTH. These data suggest that ACTH activates a phospholipase C in calf adrenocortical cells, releasing IPG, which in turn may inhibit, or modulate ACTH action. Key words: Glycosyl-phosphatidylinositol,ACTH, aldosterone biosynthesis.
INTRODUCTION
described which seems to be very similar to this IPG [3-5]. In keeping with this, it was reported that antibodies raised against the IPG from the variant surface glycoprotein (VSG) of Trypanosoma brucei blocked the insulin-induced stimulation of pyruvate dehydrogenase in BC3H-1 myocytes [6], and that an IPG fragment from V S G mimics the antilipolytic action of insulin in adipocytes as well as the inhibition of glucose-6phosphatase and fructose-l,6-biphosphatase in hepatocytes [7]. It was also found that the IPG obtained from the GPI anchor of human erythrocyte acetylcholinesterase mimics some effect of insulin on intact hepatocytes [8]. Accordingly, insulin stimulates the release of alkaline phosphatase, a protein anchored to the cell membrane by GP1 [9]. Insulin also generates IPG from free GPI which is not serving as protein anchor [4, 10, 1 I]. In addition, IPG, like insulin, seems to acti-
Recently, several glycosyl phosphatidylinositols (GPIs) have received a great deal of attention due to their function as anchors of membrane proteins in a variety of systems [1, 2]. This family of molecules consists of a phosphatidylinositol linked to n o n - N - a c e t y l a t e d g l u c o s a m i n e w h i c h is then attached to additional monosaccharides of different composition. The hydrolysis of GPI by phosp h o l i p a s e C releases an inositol p h o s p h o g l y c a n (IPG). A mediator of insulin action has been
Correspondence to: M. del C. Vila, Depto. Qufmica Biol6gica, Pabell6n 2, Piso: 4, Ciudad Universitaria, Cap. Fed. 1428, Argentina. Abbreviations: GPI--glycosyl phosphatidylinositol, IPG-inositol phosphoglycan, LPPG--lipopeptidophosphoglycan, Galf---galactofuranose residue. 331
332
M, DEL C. VILA et al.
vate low Km phosphodiesterase and inhibit adenylate cyclase and protein kinase A [12-14[. This suggests that IPG could inhibit cAMP-mediated signalling [8, 15, 16]. IPG was also proposed to mediate the action of nerve growth factor [17], interleukin-2 [11, 18] and thyroid-stimulating hormone [ 19]. GPI exists in bovine adrenocortical cells and it was proposed that it may be involved in transmembrane-signalling processes in this system [20]. Our previous data suggest [21] that ACTH increases the hydrolysis of GPI in calf adrenal glomerulosa cells. On the other hand, ACTH stimulates the synthesis of aldosterone in glomerulosa cells by a cAMP-mediated mechanism and IPG was reported to inhibit cAMP signalling. If ACTH-mediated hydrolysis of GPI resulted, as it was found with other hormones and growth lectors, in an increase in IPG, the released IPG might block ACTH action. To examine this hypothesis we used an IPG of chemically defined structure released from lipopeptidophosphoglycan (LPPG) from Trypanosoma cruzi. The lipopeptidophosphoglycan LPPG from Tr)7?anosoma cruU is a glycoinositol phosphoceramide whose complete structure was recently reported by Lederkremer et al. [22, 23]. This molecule can be included as a member of the GPI family since it contains the structural motif Man-o~I-4GlcNH2o~1-6 myo-inositol- 1-PO4-1ipid. Indeed, LPPG is very closely related to GPI protein anchor structures although it represents the first example of an IPG core structure linked to a ceramide, containing mainly palmitoylsphinganine, lignoceroylsphinganine and palmitoylsphingosine. Hydrolysis of LPPG by bacterial PI-specific phospholipase C releases ceramide and IPG with cross-reacting determinant activity due to the presence of the consensus sequence for IPG. However, this IPG has some differences from those seen associated with other anchoring molecules, namely the presence of terminal galactofuranose units and an aminoethylphosphonic acid substituent (Fig. 1). In this paper, we investigated in calf adrenocortical cells: (a) the effect of the IPG released from the LPPG of T. cruzi, on the stimulation of
aldosterone production by ACTH, and (b) if ACTH increases the hydrolysis of GPI by activation of a phospholipase C.
MATERIALS AND METHODS ACTH 1-24 (Cortrosyn) was purchased from Organon (West Orange, NJ, U.S.A.). Collagenase (type I) was purchased from Worthington Biochemicals (Freehold, NJ, U.S.A.). p-Nitro-phenylphosphate was a gift from Wiener, Argentina. Cell preparation and treatments Calf adrenal glands were obtained at a local slaughterhouse. Adrenals were freed of fat and the outer 500 ~m was sliced off with a Stadie-Riggs microtome in order to obtain adrenocortical cells as described previously ]211. Cells were resuspended in Ham F-12 medium supplemented with 1.58 mM CaCI, (Ham F12/Ca2+) and counted; aliquots of 1 x 10~ cells were incubated in 0.5 ml of Ham F-12/Ca2+ at 37°C for I h (unless otherwise stated) with the treatments indicated in each case. After incubation under an atmosphere of 95% 02/5% CO2, cells were placed on ice and immediately centrifuged at 4°C (2500 rpm x 10 rain) and supernatants were used to measure aldosterone and cAMP. Aldosterone was measured by radioimmunoassay [24]. Aliquots of supernatants were immediately placed in a boiling water bath for 2 rain and frozen to measure cAMP by radio-immunoassay [25]. Isolation o/' ~H-labelled LPPG Epimastigote forms of Tr3'panosoma cruu were incubated with [3H]palmitic acid as previously reported [26] and the labelled LPPG was purified as described in [22]. Galactofuranose residues of LPPG were selectively removed by hydrolysis with 0.02 N TFA acid for 2.5 h at 100°C [27]. After this time the sample was evaporated in a Savant concentrator, with several additions of water, until neutral. All the other components were stable under the conditions used. Determination qf ceramide releasedJhom [SHILPPG in ca!f adrenocortical cells Calf adrenocortical cells were incubated at 37°C for 8 min in Harn F-12/Ca 2+ with 20,000 cpm of [~HILPPG without galactofuranose residues in the presence or absence of 1 nM ACTH. Incubation was stopped with the addition of 0.5 ml of ether: after vortexing and centrifuging, the organic phase was separated. Extraction with ether was repeated twice, organic phases were pooled and analysed by
Effect of inositol phosphoglycan on ACTH action
333
AEP
I
6 Galf~l-3Manal-2Man(zl-2Man~1-6Mano~l-4-GIcN(~l-6myo-lnol-PO4-Ceramide
GalJ~1
LPPG
0.02N TFA 2.5 h, 100~
AEP I 6 Manal-2Manal-2Manal-6Manal-4-GIcNal-6myo-lnol-PO4-Ceramide
PI-PLC
AEP I 6 Manal-2Manal-2Manal-6Manal-4-GIcNal-6myo-lnol-PO 4
Jl"
Ceramide
IPG Fig. l. Structure of LPPG and IPG purified from 7". cruzi [22, 23].
TLC with hexane:isopropanol (93:7, v/v). Bands, 1 cm wide, were scraped off the plate and counted for radioactivity.
Determination of alkaline phosphatase activity Adrenocortical cells were incubated with or without 1 nM ACTH for the times indicated. After incubation, the cell suspensions were centrifuged (900 rpm x 8 rain) and supernatants were used to measure alkaline
phosphatase activity. The incubation mixture contained in a final volume of l ml 0.1 M carbonate buffer (pH 10), p-nitrophenylphosphate 2 mg/ml and an aliquot of the supernatant. After 15 rain of incubation at 37°C, the reaction was stopped by the addition of 2 ml of 0.1 M NaOH. The product was determined spectrophotometrically at 410 rim. An enzyme unit was defined as the amount of enzyme that produced an increase in the absorbance of 0.1 under these assay conditions.
334
M. D E E C. V I L A et al.
RESULTS The IPG obtained from the LPPG of T. cruzi has the same core as other IPGs but it is substituted by one more mannose and two Galf which are not frequent in other molecules of this family. These Galf can be selectively hydrolysed as indicated in Methods, in order to obtain a structure which consists of phosphoinositol glucosamine glycosidically linked to four mannosyl residues. The terminal mannose residue has the same linkage (~1, 2), as the third mannose in the conserved core. The effect of both compounds IPG Galf and IPG without Galf (IPG) on the ACTH-stimulated aldosterone accumulation was studied in calf adrenocortical cells. As in shown in Table 1, ACTH increased aldosterone production in these cells. IPG Galf produced a non-significant decrease in the accumulation of aldosterone promoted by ACTH while IPG significantly decreased this accumulation. In contrast, neither IPG Galf nor IPG affected the aldosterone content of control cells. As cAMP is the second messenger in ACTH action, the effect of both IPGs on the accumulation of cAMP elicited by A C T H was also studied. As can be seen in Table 1, ACTH increased cAMP content. ACTH-mediated cAMP accumulation was slightly modified by IPG Galf while IPG significantly decreased it. IPG Galf or IPG did not affect the basal level of cAMP. Table I. Effect of IPG and IPG G a l f on the stimulation of aldosterone and c A M P content by A C T H in c a l f adrenocortical cells
Treatment
Aldosterone (ng/assay)
None ACTH IPG G a l f A C T H + 1PG G a l f IPG A C T H + IPG
2.0 5.0 2.0 4.4 1.9 3.1
+ 0.2 _+0.4 _+ 0.1 _+0.3 _+ 0.2 _+0.3*
cAM P (pmoles/assay) 0.43 0.89 0.43 0.74 0.41 0.55
_+ 0,06 +_0,04 _+ 0.07 - 0.05 _+ 0.02 -+ 0.05*
C a l f adrenocortical cells were treated with vehicle, 1 n M A C T H , 20 p M IPG or 20 ~ M IPG Gall, alone or in c o m b i n a tion as indicated in the table, After 1 h o f incubation, aldosterone and c A M P were m e a s u r e d by r a d i o - i m m u n o a s s a y (see Methods). Results are m e a n s _+ S.E: for three determinations. *Significantly different f r o m A C T H P _<_0.05 (Student's ttest).
Thus, there is a good correlation between the effect of both IPGs on ACTH-mediated aldosterone and cAMP accumulation. Besides, IPG is a better inhibitor of ACTH action than IPG Galf. For this reason, IPG was used in the following experiments. The inhibition of ACTH-mediated aldosterone accumulation by IPG was dose dependent (Fig. 2) with maximal inhibition occurring at 20 laM IPG. The effect of IPG on the accumulation of aldosterone promoted by dibutyryl-cAMP (dbcAMP) was also investigated. Adrenocortical cells were pretreated with vehicle or IPG for 15 rain before adding ACTH, db-cAMP or Ham F12/Ca 2+ and then aldosterone production was measured, IPG pretreatment inhibited aldosterone accumulation promoted by both ACTH and cAMP (Table 2). In a previous paper we reported the hydrolysis of GPI by ACTH in calf adrenocortical cells [21]. To evaluate the possibility that ACTH-mediated hydrolysis of GPI is due to the activation of a phospholipase C, two different approaches were followed. First, we investigated the release of alkaline phosphatase to the incubation medium, since the release of this GPI-anchored enzyme from the cell surface has been used to evaluate the activation of GPI phospholipase C [9, 28]. As it can be seen in Fig. 3, there is a rapid and signifi-
Table 2. Effect of the pretreatment with IPG on the a c c u m u l a tion of aldosterone p r o d u c e d by A C T H or d b - c A M P
Pretreatment
Treatment
Vehicl6 Vehicle Vehicle IPG IPG IPG
Ham F-12 ACTH db-cAMP H a m F- 12 ACTH db-cAMP
Aldosterone (ng/assay) 4.7 8.9 8.0 4.7 5.2 6.0
_+0.6 _+0.8 +--0.4 _+ 0.1 - 0.4* -+ 0.3**
C a l f adrenocortical cells were pre-incubated for 15 min with vehicle or 20 ~aM IPG, a n d then treated for 45 rain with H a m F-12, 1 nM A C T H or 1 m M dibutyryl c A M P (dbc A M P ) . After the incubation, aldosterone was m e a s u r e d as indicated in Methods. Results are means + S.E. o f duplicate determinations of a representative experiment. *P _< 0.05 with respect to A C T H pretreated with vehicle. **P <- 0.05 with respect to d b - c A M P pretreated with vehicle (Student's t-test).
Effect of inositol phosphoglycanon ACTH action
1.1
Control ACTH
335
• •
m
0.9 v ®
o 0 0
~r
o.7 0.5
,
0
m
!
20
I 40
I
I 60
I
80
Fig. 2. Dose-dependent inhibition of ACTH-mediated aldosterone accumulation by IPG. Calf adrenocortical cells were treated with different concentrations of IPG in the absence (control) or in the presence of 1 nM ACTH. After 1 h of incubation, aldosterone was determined by radioimmunoassay as indicated in Methods. *P < 0.05 (Student's ttest). cant increase of alkaline phosphatase activity released to the medium with A C T H treatment. Second, adrenocortical cells were incubated with [3H]LPPG from T. cruzi without Galf, in the presence and absence of A C T H to analyse if [3H] ceramide is released. After incubation, the cell suspension was extracted with ether and the organic phase was analysed by TLC. As can be seen in Fig. 4 the content of [3H]cerarnide was increased in the ACTH-treated cells. An increase of the radioactivity that remained in the origin of the plate was also found in the treated ceils. This radioactivity is not due to ceramide phosphate since we found that when LPPG is treated with rat plasma, as a source of GPI-phospholipase D, the ceramide phosphate released is not extracted by ether (Lederkremer R. et al., submitted for publication). The compounds more polar than ceramide, that remain in the origin of the plate, were not further investigated.
DISCUSSION In a previous paper we have shown that A C T H increases GPI hydrolysis in calf adrenocortical cells. Our present findings show that ACTHmediated hydrolysis of GPI is due to the activation of a phospholipase C since we found an increase in the alkaline phosphatase released to the incubation media by ACTH treatment. Furthermore, the activation of a phospholipase C in intact cells has been directly assayed using a substrate with a chemically defined structure: a glycoinositolphosphoceramide, LPPG. It was previously demonstrated [22] that ceramide is released from LPPG by a PI-specific phospholipase C. The increase in the release of ceramide from LPPG by A C T H confirmed the activation of a phospholipase C by the hormone. A GPI-specific phospholipase C was purified from Trypanosoma brucei [29] and rat liver
336
M. DEL C. V I L A et al.
4.5 Control ACTH
rJ t l
•
®E
• •
I
o gu D,.
4.0 D.w
~d ®'= •-- e,<
3.5
t 0
!
i
2
i
I
4
i
6
i
i
8
i
10
time (rain.)
Fig. 3. Effect of ACTH on the release of alkaline phosphatase from calf adrenocortical cells. Calf adrenocortical cells were treated with 1 nM ACTH or vehicle (control) for the times indicated. After incubation, the cell suspensions were centrifuged and alkaline phosphatase activity was measured in the supernatants as indicated in Methods. Values shown are means +_ S.E. of duplicate determinations of a representative experiment repeated three times with similar results. *Significantly different from control, P < 0.05 (Student's t-test).
plasma membrane [30]. ACTH is probably activating a GPI-phospholipase C present in calf adrenocortical cells. As a result of the activation of GPI-phospholipase C by ACTH, IPG would be released to the incubation media and then, transported into the cell by a specific uptake system [6, 31]. Our results show that the IPG obtained from T. cruzi LPPG inhibits ACTH-stimulated aldosterone production and c A M P accumulation. The presence of G a l l in IPG, which is a rare feature of the IPG isolated from T. cruzi, decreased its effect. In contrast, the IPG without the Galf, is a more potent inhibitor of A C T H action. This IPG (see Fig. 1) contains the core structure of other IPGs described [1, 2] consisting of phosphoinosi-
tol glycosidically linked to a tetrasaccharide composed of one glucosaminyl and three mannosyl residues. The dose of IPG necessary to inhibit ACTH action (20 ~tM) is similar to that reported for other IPGs that provoked different biological effects [7, 8]. W e found that IPG does not affect the basal content of c A M P in calf adrenocortical cells but inhibits A C T H - m e d i a t e d increase. This is consistent with the fact that IPG was reported to inhibit adenylyl cyclase and activate phosphodiesterase [12, 13]. When calf adrenocortical cells were pretreated with IPG before the addition of A C T H or dbcAMP, both responses were blocked. These find-
Effect of inosilol phosphoglycan on ACTH action
337
600
Control ACTH A
E e~ &
400
"u
C
Q
-& _1
II •
P
200
|
0
!
10
|
20
distance (cm)
Fig. 4. Effect of ACTH on the hydrolysis of [3H]LPPG by calf adrenocortical cells. Calf adrenocortical cells were incubated with [3H]LPPG as indicated in Methods, in the absence (control) or in the presence of ACTH 1 nM. After 8 min of incubation lipids were extracted with ether and analysed by TLC in hexane: isopropanol (93:7, v/v). Bands l cm wide were scraped off and counted for radioactivity. The positions of authentic ceramide, C and palmitic acid, P are shown. This experiment was repeated twice with similar results.
ings suggest that a site distal to c A M P production is inhibited by IPG. This site might be protein kinase A. Protein kinase A was found to be directly inhibited by IPG [14], although very recently Deeg et al. [8] reported that, using IPG derived from human erythrocyte acetylcholinesterase, they could not reproduce this inhibition suggesting that the difference in the results might be due to structural differences in the IPGs tested. At present our data suggest that a postc A M P event is inhibited by IPG but further studies are necessary to elucidate the site of action of IPG. The fact that IPG may block the d b - c A M P mediated increase in aldosterone production strongly suggests that this is a post-ACTH-bind-
ing effect of IPG rather than a decrease in the binding of A C T H to its receptor. Accordingly, we found that the inhibition of A C T H response by IPG was maintained even if the dose of A C T H was increased from 1 nM to 1 g M , suggesting that the effect of IPG was not due to a competitive inhibition of the binding of A C T H to its receptor (data not shown). Interestingly, it was also found that when IPG was added simultaneously with A C T H or dbcAMP, A C T H response was inhibited but not that of d b - c A M P (data not shown). This probably reflects the tact that d b - c A M P effect, but not A C T H effect, is faster than that of IPG and once the d b - c A M P effect was triggered it could not be inhibited by a simultaneous addition of 1PG. Even
338
M. DEL C. VILA et al.
when our data suggest that the release of IPG is rapid (Fig. 3), its transport and its intracellular effects may not be as fast. It has recently been reported that insulin is able to inhibit aldosterone accumulation produced by ACTH, but not by angiotensin, which is known to act through a cAMP-independent mechanism, in bovine glomerulosa cells. Insulin is also able to inhibit ACTH response in fasciculata-reticularis cells. This seems to be due to the ability of insulin to decrease cAMP content [32]. Taking into account our findings and the fact that insulin is known to hydrolyse GPI [4, 10], this effect of insulin in bovine adrenocortical cells might be mediated by IPG. Moreover, Klein et al. also found that insulin acts on the synthesis, degradation and intracellular action of cAMP and this is in agreement with the sites of action proposed for IPG. In conclusion, our results suggest that ACTH activates a GPI-phospholipase C, releasing IPG which, in turn, may be a signal to modulate or inhibit ACTH action. Consistently, the inhibition of other cAMP-dependent effects by other IPGs has been recently reported [7, 8, 15, 16]. Further studies are in progress to evaluate if IPG inhibits ACTH-mediated accumulation of other steroid hormones in adrenocortical cells. Acknowledgements--This work received financial support from: Fundaci6n Antorchas, Consejo Nacional de Investigaciones Cientfficas y Tdcnicas and Universidad de Buenos Aires. We are grateful to the laboratory of Dr Walter Colli (Sao Paulo, Brazil) for providing the T. cruzi cells and to Dr Celso GomezSanchez for the monoclonal antibody for aldosterone radioimmunoassay.
REFERENCES 1. Doering T. L., Masterson W. J., Hart G. W. and Englund P. T. (1990) J. biol. Chem. 265, 611-614. 2. Ferguson M. A. J. (1992) Biochem. Soc. Trans. 20, 243-256. 3. Mato J. M. (1989) Cell. Signal. 1, 143-146. 4. Saltiel A. R., Fox J. A., Sherline P. and Cuatrecasas P. (1986) Science 233, 967-972. 5. Vila M. C., Milligan G., Standaert M. L. and Farese R. V. (1990) Biochemistry 29, 8735-8740.
6. Romero G., Gamez G., Huang L. C., Lilley K. and Luttrell L. (1990) Proc. natn. Acad. Sci. U.S.A. 87, 1476-1480. 7. Misek D. E. and Saltiel A. R. (1992) J. biol. Chem. 267, 16266-16273. 8. Deeg M. A., Brass E. P. and Rosenberry T. L. (1993) Diabetes 42, 1318-1323. 9. Romero G., Luttrell L., Rogol A., Zeller K., Hewlett E. and Larner J. (1988) Science 240, 509-,511. 10. Mato J. M., Kelly K. L., Abler A. and Jarret L. (1987) J. biol. Chem. 262, 2131-2137. 11. Merida I., Pratt J. C. and Gaulton G. N. (1990) Proc. ham. Acad. Sci. U.S.A. 87, 9421-9425. 12. Saltiel A. R. (1987) Endocrinology 120, 967-972. 13. Martiny L., Dib K., Haye B., Correze C., Jacquemin C. and Lambert B. (199l) FEBS Lett. 286, 105-109. 14. Villalba M., Kelly K. L. and Mato J. M. (1988) Biochim. biophys. Acta 968, 69-76. 15. Alvarez J. F., Cabello M. A., Felfu J. E. and Mato J. M. (1987) Biochem. biophys. Res. Commun. 147, 765-771. 16. Alemany S., Mato J. and Stralfors P. (1987) Nature 330, 77-79. 17. Represa J., Avila M. A., Miner C., Giraldez F., Romero G., Clemente R., Mato J. and Varela-Nieto I. (1991) Proc. natn. Acad. Sci. U.S,A. 88, 8016-8019, 18. Eardly D. E. and Koshland M. E. (1991) Science 251, 78-81. 19. Martiny L., Antonicelli F., Thuilliez B., Lambert B., Jacquemin C. and Haye B. (1990) Cell. Signal. 2, 21-27. 20. lgarashi I. and Chambaz E. M. (1987) Biochem. biophys. Res. Commun. 145, 249-256. 21. Cozza E. N., Vila M. C., Gomez-Sanchez C. E. and Farese R. V. (1988) Bioehem. biophys. Res. Commun. 157, 585-589. 22. Lederkremer R. M., Lima C., Ramirez M. I. and Casal O. L. (1990) Eur. J. Biochem. 192, 337-345. 23. Lederkremer R. M., Lima C., Ramirez M. 1., Ferguson M. A. J., Homans S, W. and ThomasOates J. ( 1991 ) J. biol. Chem. 266, 23670-23675. 24. Gomez-Sanchez C. E., Foecking M. F., Ferris M. W., Chavarri M. R., Uribe L. and Gomez-Sanchez E. P. (1987) Steroids 49, 581-587. 25. Steiner A. L., Parker C. W. and Kipnis D. M. (1972) J. biol. Chem. 247, 1106-1113. 26. Lederkremer R. M., Lima C. E., Ramirez M. I., Goncalvez M. F. and Colli W. (1993) Eur. J. Biochem. 218, 929-936. 27. Lederkremer R. M., Casal O. L., Alves J. M. and Colli W. (1980) FEBS Lett. 116, 25-29. 28. Roberts J. M., Kenton P. and Johnson P. M. (1990) FEBS Lett. 267, 213-216.
Effect of inositol phosphoglycan on ACTH action 29. Fox J. A., Duszenko M., Ferguson M. A., Low M. G. and Cross G. A. M. (1986) J. biol. Chem. 261, 15767-15771. 30. Fox J. A., Soliz N. M. and Saltiel A. R. (1987) Proc. natn. Acad. Sci. U.S.A. 84, 2663-2667. 31. Alvarez J. F., Sanchez-Arias J. A., Guadafio A.,
339
Est6vez F., Varela I., Felfu J. E. and Mato J. M. (1991) Biochem. J. 274, 369-374. 32. Klein N. A., Andersen R. N., Casson P. R., Buster J. N. and Kramer R. E. (1992) J. Steroid. Biochem. Molec. Biol. 41, 11-20.
Pergamon 0898-6568(95)00008-9
AN INOSITOL
PHOSPHOGLYCAN ACTION
FROM
TRYPANOSOMA
IN CALF ADRENOCORTICAL
CRUZI INHIBITS
ACTH
CELLS
M A R I A DEL C. VILA,* E D U A R D O N. COZZA,* C A R L O S L I M A , t M A R I A I. R A M I R E Z t and R O S A M. D E . L E D E R K R E M E R t *Departamento de Qufmica Bioldgica and tDepartamento de Qu/mica Org~inica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Argentina (Received 9 December 1994; and accepted 23 December 1994) A b s t r a c t - - W e describe the effect of an inositol phosphoglycan (IPG) purified from Trypanosoma cruzi on the stimulation of aldosterone and cAMP production by ACTH in calf adrenocortical cells. T. cruzi IPG has two galactofuranose residues (Galf) which are not frequent in other IPGs. The effect of IPG with galactofuranose residues (IPG Galf) and IPG without these residues (IPG) was investigated. It was found that IPG Galf slightly decreased the stimulation of aldosterone and cAMP production by ACTH, whereas IPG significantly inhibited ACTH-mediated accumulation of both aldosterone and cAMP. The inhibition of aldosterone content in ACTH-treated cells by IPG was dose dependent. It was also found that the pretreatment of calf adrenocortical cells with IPG inhibited the accumulation of aldosterone provoked by ACTH and dibutyryladenosine-3',5'-cyclic monophosphate (db-cAMP). On the other hand, the activation of a GPI (glycosyl phosphatidylinositol)-phospholipaseC by ACTH was evaluated. First it was found that the release of ceramide from a GPI-like molecule: a glycoinositol-phosphoceramide (LPPG) purified from T. cruzi is increased in ACTH-treated cells. Second, the release of alkaline phosphatase, a GPI-anchored enzyme, to the extracellular medium was increased in these cells by ACTH. These data suggest that ACTH activates a phospholipase C in calf adrenocortical cells, releasing IPG, which in turn may inhibit, or modulate ACTH action. Key words: Glycosyl-phosphatidylinositol,ACTH, aldosterone biosynthesis.
INTRODUCTION
described which seems to be very similar to this IPG [3-5]. In keeping with this, it was reported that antibodies raised against the IPG from the variant surface glycoprotein (VSG) of Trypanosoma brucei blocked the insulin-induced stimulation of pyruvate dehydrogenase in BC3H-1 myocytes [6], and that an IPG fragment from V S G mimics the antilipolytic action of insulin in adipocytes as well as the inhibition of glucose-6phosphatase and fructose-l,6-biphosphatase in hepatocytes [7]. It was also found that the IPG obtained from the GPI anchor of human erythrocyte acetylcholinesterase mimics some effect of insulin on intact hepatocytes [8]. Accordingly, insulin stimulates the release of alkaline phosphatase, a protein anchored to the cell membrane by GP1 [9]. Insulin also generates IPG from free GPI which is not serving as protein anchor [4, 10, 1 I]. In addition, IPG, like insulin, seems to acti-
Recently, several glycosyl phosphatidylinositols (GPIs) have received a great deal of attention due to their function as anchors of membrane proteins in a variety of systems [1, 2]. This family of molecules consists of a phosphatidylinositol linked to n o n - N - a c e t y l a t e d g l u c o s a m i n e w h i c h is then attached to additional monosaccharides of different composition. The hydrolysis of GPI by phosp h o l i p a s e C releases an inositol p h o s p h o g l y c a n (IPG). A mediator of insulin action has been
Correspondence to: M. del C. Vila, Depto. Qufmica Biol6gica, Pabell6n 2, Piso: 4, Ciudad Universitaria, Cap. Fed. 1428, Argentina. Abbreviations: GPI--glycosyl phosphatidylinositol, IPG-inositol phosphoglycan, LPPG--lipopeptidophosphoglycan, Galf---galactofuranose residue. 331
332
M, DEL C. VILA et al.
vate low Km phosphodiesterase and inhibit adenylate cyclase and protein kinase A [12-14[. This suggests that IPG could inhibit cAMP-mediated signalling [8, 15, 16]. IPG was also proposed to mediate the action of nerve growth factor [17], interleukin-2 [11, 18] and thyroid-stimulating hormone [ 19]. GPI exists in bovine adrenocortical cells and it was proposed that it may be involved in transmembrane-signalling processes in this system [20]. Our previous data suggest [21] that ACTH increases the hydrolysis of GPI in calf adrenal glomerulosa cells. On the other hand, ACTH stimulates the synthesis of aldosterone in glomerulosa cells by a cAMP-mediated mechanism and IPG was reported to inhibit cAMP signalling. If ACTH-mediated hydrolysis of GPI resulted, as it was found with other hormones and growth lectors, in an increase in IPG, the released IPG might block ACTH action. To examine this hypothesis we used an IPG of chemically defined structure released from lipopeptidophosphoglycan (LPPG) from Trypanosoma cruzi. The lipopeptidophosphoglycan LPPG from Tr)7?anosoma cruU is a glycoinositol phosphoceramide whose complete structure was recently reported by Lederkremer et al. [22, 23]. This molecule can be included as a member of the GPI family since it contains the structural motif Man-o~I-4GlcNH2o~1-6 myo-inositol- 1-PO4-1ipid. Indeed, LPPG is very closely related to GPI protein anchor structures although it represents the first example of an IPG core structure linked to a ceramide, containing mainly palmitoylsphinganine, lignoceroylsphinganine and palmitoylsphingosine. Hydrolysis of LPPG by bacterial PI-specific phospholipase C releases ceramide and IPG with cross-reacting determinant activity due to the presence of the consensus sequence for IPG. However, this IPG has some differences from those seen associated with other anchoring molecules, namely the presence of terminal galactofuranose units and an aminoethylphosphonic acid substituent (Fig. 1). In this paper, we investigated in calf adrenocortical cells: (a) the effect of the IPG released from the LPPG of T. cruzi, on the stimulation of
aldosterone production by ACTH, and (b) if ACTH increases the hydrolysis of GPI by activation of a phospholipase C.
MATERIALS AND METHODS ACTH 1-24 (Cortrosyn) was purchased from Organon (West Orange, NJ, U.S.A.). Collagenase (type I) was purchased from Worthington Biochemicals (Freehold, NJ, U.S.A.). p-Nitro-phenylphosphate was a gift from Wiener, Argentina. Cell preparation and treatments Calf adrenal glands were obtained at a local slaughterhouse. Adrenals were freed of fat and the outer 500 ~m was sliced off with a Stadie-Riggs microtome in order to obtain adrenocortical cells as described previously ]211. Cells were resuspended in Ham F-12 medium supplemented with 1.58 mM CaCI, (Ham F12/Ca2+) and counted; aliquots of 1 x 10~ cells were incubated in 0.5 ml of Ham F-12/Ca2+ at 37°C for I h (unless otherwise stated) with the treatments indicated in each case. After incubation under an atmosphere of 95% 02/5% CO2, cells were placed on ice and immediately centrifuged at 4°C (2500 rpm x 10 rain) and supernatants were used to measure aldosterone and cAMP. Aldosterone was measured by radioimmunoassay [24]. Aliquots of supernatants were immediately placed in a boiling water bath for 2 rain and frozen to measure cAMP by radio-immunoassay [25]. Isolation o/' ~H-labelled LPPG Epimastigote forms of Tr3'panosoma cruu were incubated with [3H]palmitic acid as previously reported [26] and the labelled LPPG was purified as described in [22]. Galactofuranose residues of LPPG were selectively removed by hydrolysis with 0.02 N TFA acid for 2.5 h at 100°C [27]. After this time the sample was evaporated in a Savant concentrator, with several additions of water, until neutral. All the other components were stable under the conditions used. Determination qf ceramide releasedJhom [SHILPPG in ca!f adrenocortical cells Calf adrenocortical cells were incubated at 37°C for 8 min in Harn F-12/Ca 2+ with 20,000 cpm of [~HILPPG without galactofuranose residues in the presence or absence of 1 nM ACTH. Incubation was stopped with the addition of 0.5 ml of ether: after vortexing and centrifuging, the organic phase was separated. Extraction with ether was repeated twice, organic phases were pooled and analysed by
Effect of inositol phosphoglycan on ACTH action
333
AEP
I
6 Galf~l-3Manal-2Man(zl-2Man~1-6Mano~l-4-GIcN(~l-6myo-lnol-PO4-Ceramide
GalJ~1
LPPG
0.02N TFA 2.5 h, 100~
AEP I 6 Manal-2Manal-2Manal-6Manal-4-GIcNal-6myo-lnol-PO4-Ceramide
PI-PLC
AEP I 6 Manal-2Manal-2Manal-6Manal-4-GIcNal-6myo-lnol-PO 4
Jl"
Ceramide
IPG Fig. l. Structure of LPPG and IPG purified from 7". cruzi [22, 23].
TLC with hexane:isopropanol (93:7, v/v). Bands, 1 cm wide, were scraped off the plate and counted for radioactivity.
Determination of alkaline phosphatase activity Adrenocortical cells were incubated with or without 1 nM ACTH for the times indicated. After incubation, the cell suspensions were centrifuged (900 rpm x 8 rain) and supernatants were used to measure alkaline
phosphatase activity. The incubation mixture contained in a final volume of l ml 0.1 M carbonate buffer (pH 10), p-nitrophenylphosphate 2 mg/ml and an aliquot of the supernatant. After 15 rain of incubation at 37°C, the reaction was stopped by the addition of 2 ml of 0.1 M NaOH. The product was determined spectrophotometrically at 410 rim. An enzyme unit was defined as the amount of enzyme that produced an increase in the absorbance of 0.1 under these assay conditions.
334
M. D E E C. V I L A et al.
RESULTS The IPG obtained from the LPPG of T. cruzi has the same core as other IPGs but it is substituted by one more mannose and two Galf which are not frequent in other molecules of this family. These Galf can be selectively hydrolysed as indicated in Methods, in order to obtain a structure which consists of phosphoinositol glucosamine glycosidically linked to four mannosyl residues. The terminal mannose residue has the same linkage (~1, 2), as the third mannose in the conserved core. The effect of both compounds IPG Galf and IPG without Galf (IPG) on the ACTH-stimulated aldosterone accumulation was studied in calf adrenocortical cells. As in shown in Table 1, ACTH increased aldosterone production in these cells. IPG Galf produced a non-significant decrease in the accumulation of aldosterone promoted by ACTH while IPG significantly decreased this accumulation. In contrast, neither IPG Galf nor IPG affected the aldosterone content of control cells. As cAMP is the second messenger in ACTH action, the effect of both IPGs on the accumulation of cAMP elicited by A C T H was also studied. As can be seen in Table 1, ACTH increased cAMP content. ACTH-mediated cAMP accumulation was slightly modified by IPG Galf while IPG significantly decreased it. IPG Galf or IPG did not affect the basal level of cAMP. Table I. Effect of IPG and IPG G a l f on the stimulation of aldosterone and c A M P content by A C T H in c a l f adrenocortical cells
Treatment
Aldosterone (ng/assay)
None ACTH IPG G a l f A C T H + 1PG G a l f IPG A C T H + IPG
2.0 5.0 2.0 4.4 1.9 3.1
+ 0.2 _+0.4 _+ 0.1 _+0.3 _+ 0.2 _+0.3*
cAM P (pmoles/assay) 0.43 0.89 0.43 0.74 0.41 0.55
_+ 0,06 +_0,04 _+ 0.07 - 0.05 _+ 0.02 -+ 0.05*
C a l f adrenocortical cells were treated with vehicle, 1 n M A C T H , 20 p M IPG or 20 ~ M IPG Gall, alone or in c o m b i n a tion as indicated in the table, After 1 h o f incubation, aldosterone and c A M P were m e a s u r e d by r a d i o - i m m u n o a s s a y (see Methods). Results are m e a n s _+ S.E: for three determinations. *Significantly different f r o m A C T H P _<_0.05 (Student's ttest).
Thus, there is a good correlation between the effect of both IPGs on ACTH-mediated aldosterone and cAMP accumulation. Besides, IPG is a better inhibitor of ACTH action than IPG Galf. For this reason, IPG was used in the following experiments. The inhibition of ACTH-mediated aldosterone accumulation by IPG was dose dependent (Fig. 2) with maximal inhibition occurring at 20 laM IPG. The effect of IPG on the accumulation of aldosterone promoted by dibutyryl-cAMP (dbcAMP) was also investigated. Adrenocortical cells were pretreated with vehicle or IPG for 15 rain before adding ACTH, db-cAMP or Ham F12/Ca 2+ and then aldosterone production was measured, IPG pretreatment inhibited aldosterone accumulation promoted by both ACTH and cAMP (Table 2). In a previous paper we reported the hydrolysis of GPI by ACTH in calf adrenocortical cells [21]. To evaluate the possibility that ACTH-mediated hydrolysis of GPI is due to the activation of a phospholipase C, two different approaches were followed. First, we investigated the release of alkaline phosphatase to the incubation medium, since the release of this GPI-anchored enzyme from the cell surface has been used to evaluate the activation of GPI phospholipase C [9, 28]. As it can be seen in Fig. 3, there is a rapid and signifi-
Table 2. Effect of the pretreatment with IPG on the a c c u m u l a tion of aldosterone p r o d u c e d by A C T H or d b - c A M P
Pretreatment
Treatment
Vehicl6 Vehicle Vehicle IPG IPG IPG
Ham F-12 ACTH db-cAMP H a m F- 12 ACTH db-cAMP
Aldosterone (ng/assay) 4.7 8.9 8.0 4.7 5.2 6.0
_+0.6 _+0.8 +--0.4 _+ 0.1 - 0.4* -+ 0.3**
C a l f adrenocortical cells were pre-incubated for 15 min with vehicle or 20 ~aM IPG, a n d then treated for 45 rain with H a m F-12, 1 nM A C T H or 1 m M dibutyryl c A M P (dbc A M P ) . After the incubation, aldosterone was m e a s u r e d as indicated in Methods. Results are means + S.E. o f duplicate determinations of a representative experiment. *P _< 0.05 with respect to A C T H pretreated with vehicle. **P <- 0.05 with respect to d b - c A M P pretreated with vehicle (Student's t-test).
Effect of inositol phosphoglycanon ACTH action
1.1
Control ACTH
335
• •
m
0.9 v ®
o 0 0
~r
o.7 0.5
,
0
m
!
20
I 40
I
I 60
I
80
Fig. 2. Dose-dependent inhibition of ACTH-mediated aldosterone accumulation by IPG. Calf adrenocortical cells were treated with different concentrations of IPG in the absence (control) or in the presence of 1 nM ACTH. After 1 h of incubation, aldosterone was determined by radioimmunoassay as indicated in Methods. *P < 0.05 (Student's ttest). cant increase of alkaline phosphatase activity released to the medium with A C T H treatment. Second, adrenocortical cells were incubated with [3H]LPPG from T. cruzi without Galf, in the presence and absence of A C T H to analyse if [3H] ceramide is released. After incubation, the cell suspension was extracted with ether and the organic phase was analysed by TLC. As can be seen in Fig. 4 the content of [3H]cerarnide was increased in the ACTH-treated cells. An increase of the radioactivity that remained in the origin of the plate was also found in the treated ceils. This radioactivity is not due to ceramide phosphate since we found that when LPPG is treated with rat plasma, as a source of GPI-phospholipase D, the ceramide phosphate released is not extracted by ether (Lederkremer R. et al., submitted for publication). The compounds more polar than ceramide, that remain in the origin of the plate, were not further investigated.
DISCUSSION In a previous paper we have shown that A C T H increases GPI hydrolysis in calf adrenocortical cells. Our present findings show that ACTHmediated hydrolysis of GPI is due to the activation of a phospholipase C since we found an increase in the alkaline phosphatase released to the incubation media by ACTH treatment. Furthermore, the activation of a phospholipase C in intact cells has been directly assayed using a substrate with a chemically defined structure: a glycoinositolphosphoceramide, LPPG. It was previously demonstrated [22] that ceramide is released from LPPG by a PI-specific phospholipase C. The increase in the release of ceramide from LPPG by A C T H confirmed the activation of a phospholipase C by the hormone. A GPI-specific phospholipase C was purified from Trypanosoma brucei [29] and rat liver
336
M. DEL C. V I L A et al.
4.5 Control ACTH
rJ t l
•
®E
• •
I
o gu D,.
4.0 D.w
~d ®'= •-- e,<
3.5
t 0
!
i
2
i
I
4
i
6
i
i
8
i
10
time (rain.)
Fig. 3. Effect of ACTH on the release of alkaline phosphatase from calf adrenocortical cells. Calf adrenocortical cells were treated with 1 nM ACTH or vehicle (control) for the times indicated. After incubation, the cell suspensions were centrifuged and alkaline phosphatase activity was measured in the supernatants as indicated in Methods. Values shown are means +_ S.E. of duplicate determinations of a representative experiment repeated three times with similar results. *Significantly different from control, P < 0.05 (Student's t-test).
plasma membrane [30]. ACTH is probably activating a GPI-phospholipase C present in calf adrenocortical cells. As a result of the activation of GPI-phospholipase C by ACTH, IPG would be released to the incubation media and then, transported into the cell by a specific uptake system [6, 31]. Our results show that the IPG obtained from T. cruzi LPPG inhibits ACTH-stimulated aldosterone production and c A M P accumulation. The presence of G a l l in IPG, which is a rare feature of the IPG isolated from T. cruzi, decreased its effect. In contrast, the IPG without the Galf, is a more potent inhibitor of A C T H action. This IPG (see Fig. 1) contains the core structure of other IPGs described [1, 2] consisting of phosphoinosi-
tol glycosidically linked to a tetrasaccharide composed of one glucosaminyl and three mannosyl residues. The dose of IPG necessary to inhibit ACTH action (20 ~tM) is similar to that reported for other IPGs that provoked different biological effects [7, 8]. W e found that IPG does not affect the basal content of c A M P in calf adrenocortical cells but inhibits A C T H - m e d i a t e d increase. This is consistent with the fact that IPG was reported to inhibit adenylyl cyclase and activate phosphodiesterase [12, 13]. When calf adrenocortical cells were pretreated with IPG before the addition of A C T H or dbcAMP, both responses were blocked. These find-
Effect of inosilol phosphoglycan on ACTH action
337
600
Control ACTH A
E e~ &
400
"u
C
Q
-& _1
II •
P
200
|
0
!
10
|
20
distance (cm)
Fig. 4. Effect of ACTH on the hydrolysis of [3H]LPPG by calf adrenocortical cells. Calf adrenocortical cells were incubated with [3H]LPPG as indicated in Methods, in the absence (control) or in the presence of ACTH 1 nM. After 8 min of incubation lipids were extracted with ether and analysed by TLC in hexane: isopropanol (93:7, v/v). Bands l cm wide were scraped off and counted for radioactivity. The positions of authentic ceramide, C and palmitic acid, P are shown. This experiment was repeated twice with similar results.
ings suggest that a site distal to c A M P production is inhibited by IPG. This site might be protein kinase A. Protein kinase A was found to be directly inhibited by IPG [14], although very recently Deeg et al. [8] reported that, using IPG derived from human erythrocyte acetylcholinesterase, they could not reproduce this inhibition suggesting that the difference in the results might be due to structural differences in the IPGs tested. At present our data suggest that a postc A M P event is inhibited by IPG but further studies are necessary to elucidate the site of action of IPG. The fact that IPG may block the d b - c A M P mediated increase in aldosterone production strongly suggests that this is a post-ACTH-bind-
ing effect of IPG rather than a decrease in the binding of A C T H to its receptor. Accordingly, we found that the inhibition of A C T H response by IPG was maintained even if the dose of A C T H was increased from 1 nM to 1 g M , suggesting that the effect of IPG was not due to a competitive inhibition of the binding of A C T H to its receptor (data not shown). Interestingly, it was also found that when IPG was added simultaneously with A C T H or dbcAMP, A C T H response was inhibited but not that of d b - c A M P (data not shown). This probably reflects the tact that d b - c A M P effect, but not A C T H effect, is faster than that of IPG and once the d b - c A M P effect was triggered it could not be inhibited by a simultaneous addition of 1PG. Even
338
M. DEL C. VILA et al.
when our data suggest that the release of IPG is rapid (Fig. 3), its transport and its intracellular effects may not be as fast. It has recently been reported that insulin is able to inhibit aldosterone accumulation produced by ACTH, but not by angiotensin, which is known to act through a cAMP-independent mechanism, in bovine glomerulosa cells. Insulin is also able to inhibit ACTH response in fasciculata-reticularis cells. This seems to be due to the ability of insulin to decrease cAMP content [32]. Taking into account our findings and the fact that insulin is known to hydrolyse GPI [4, 10], this effect of insulin in bovine adrenocortical cells might be mediated by IPG. Moreover, Klein et al. also found that insulin acts on the synthesis, degradation and intracellular action of cAMP and this is in agreement with the sites of action proposed for IPG. In conclusion, our results suggest that ACTH activates a GPI-phospholipase C, releasing IPG which, in turn, may be a signal to modulate or inhibit ACTH action. Consistently, the inhibition of other cAMP-dependent effects by other IPGs has been recently reported [7, 8, 15, 16]. Further studies are in progress to evaluate if IPG inhibits ACTH-mediated accumulation of other steroid hormones in adrenocortical cells. Acknowledgements--This work received financial support from: Fundaci6n Antorchas, Consejo Nacional de Investigaciones Cientfficas y Tdcnicas and Universidad de Buenos Aires. We are grateful to the laboratory of Dr Walter Colli (Sao Paulo, Brazil) for providing the T. cruzi cells and to Dr Celso GomezSanchez for the monoclonal antibody for aldosterone radioimmunoassay.
REFERENCES 1. Doering T. L., Masterson W. J., Hart G. W. and Englund P. T. (1990) J. biol. Chem. 265, 611-614. 2. Ferguson M. A. J. (1992) Biochem. Soc. Trans. 20, 243-256. 3. Mato J. M. (1989) Cell. Signal. 1, 143-146. 4. Saltiel A. R., Fox J. A., Sherline P. and Cuatrecasas P. (1986) Science 233, 967-972. 5. Vila M. C., Milligan G., Standaert M. L. and Farese R. V. (1990) Biochemistry 29, 8735-8740.
6. Romero G., Gamez G., Huang L. C., Lilley K. and Luttrell L. (1990) Proc. natn. Acad. Sci. U.S.A. 87, 1476-1480. 7. Misek D. E. and Saltiel A. R. (1992) J. biol. Chem. 267, 16266-16273. 8. Deeg M. A., Brass E. P. and Rosenberry T. L. (1993) Diabetes 42, 1318-1323. 9. Romero G., Luttrell L., Rogol A., Zeller K., Hewlett E. and Larner J. (1988) Science 240, 509-,511. 10. Mato J. M., Kelly K. L., Abler A. and Jarret L. (1987) J. biol. Chem. 262, 2131-2137. 11. Merida I., Pratt J. C. and Gaulton G. N. (1990) Proc. ham. Acad. Sci. U.S.A. 87, 9421-9425. 12. Saltiel A. R. (1987) Endocrinology 120, 967-972. 13. Martiny L., Dib K., Haye B., Correze C., Jacquemin C. and Lambert B. (199l) FEBS Lett. 286, 105-109. 14. Villalba M., Kelly K. L. and Mato J. M. (1988) Biochim. biophys. Acta 968, 69-76. 15. Alvarez J. F., Cabello M. A., Felfu J. E. and Mato J. M. (1987) Biochem. biophys. Res. Commun. 147, 765-771. 16. Alemany S., Mato J. and Stralfors P. (1987) Nature 330, 77-79. 17. Represa J., Avila M. A., Miner C., Giraldez F., Romero G., Clemente R., Mato J. and Varela-Nieto I. (1991) Proc. natn. Acad. Sci. U.S,A. 88, 8016-8019, 18. Eardly D. E. and Koshland M. E. (1991) Science 251, 78-81. 19. Martiny L., Antonicelli F., Thuilliez B., Lambert B., Jacquemin C. and Haye B. (1990) Cell. Signal. 2, 21-27. 20. lgarashi I. and Chambaz E. M. (1987) Biochem. biophys. Res. Commun. 145, 249-256. 21. Cozza E. N., Vila M. C., Gomez-Sanchez C. E. and Farese R. V. (1988) Bioehem. biophys. Res. Commun. 157, 585-589. 22. Lederkremer R. M., Lima C., Ramirez M. I. and Casal O. L. (1990) Eur. J. Biochem. 192, 337-345. 23. Lederkremer R. M., Lima C., Ramirez M. 1., Ferguson M. A. J., Homans S, W. and ThomasOates J. ( 1991 ) J. biol. Chem. 266, 23670-23675. 24. Gomez-Sanchez C. E., Foecking M. F., Ferris M. W., Chavarri M. R., Uribe L. and Gomez-Sanchez E. P. (1987) Steroids 49, 581-587. 25. Steiner A. L., Parker C. W. and Kipnis D. M. (1972) J. biol. Chem. 247, 1106-1113. 26. Lederkremer R. M., Lima C. E., Ramirez M. I., Goncalvez M. F. and Colli W. (1993) Eur. J. Biochem. 218, 929-936. 27. Lederkremer R. M., Casal O. L., Alves J. M. and Colli W. (1980) FEBS Lett. 116, 25-29. 28. Roberts J. M., Kenton P. and Johnson P. M. (1990) FEBS Lett. 267, 213-216.
Effect of inositol phosphoglycan on ACTH action 29. Fox J. A., Duszenko M., Ferguson M. A., Low M. G. and Cross G. A. M. (1986) J. biol. Chem. 261, 15767-15771. 30. Fox J. A., Soliz N. M. and Saltiel A. R. (1987) Proc. natn. Acad. Sci. U.S.A. 84, 2663-2667. 31. Alvarez J. F., Sanchez-Arias J. A., Guadafio A.,
339
Est6vez F., Varela I., Felfu J. E. and Mato J. M. (1991) Biochem. J. 274, 369-374. 32. Klein N. A., Andersen R. N., Casson P. R., Buster J. N. and Kramer R. E. (1992) J. Steroid. Biochem. Molec. Biol. 41, 11-20.