The effects of d - and l -glyceraldehyde on glucose oxidation, insulin secretion and insulin biosynthesis by pancreatic islets of the rat

384

Biochimica et Biophysica Acta, 399 (1975) 384--394

© Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands

BBA 27704 THE EFFECTS OF D- AND L-GLYCERALDEHYDE ON GLUCOSE OXIDATION, INSULIN SECRETION AND INSULIN BIOSYNTHESIS BY PANCREATIC ISLETS OF THE RAT

KANTI JAIN, JOHN LOGOTHETOPOULOS and PETER ZUCKER Banting and Best Department of Medical Research, University of Toronto, Toronto, M5G, 1L6, Toronto, Ontario (Canada)

(Received December 18th, 1974) (Revised manuscript received April 17th, 1975)

Summary D-glyceraldehyde stimulated insulin secretion from isolated rat pancreatic islets in static incubation and perifusion systems. At low concentrations (2--4 mM) D-glyceraldehyde was a more potent secretagogue than glucose. The insulinotropic action of 15 mM D-glyceraldehyde was not affected by D-mannoheptulose, was potentiated by cytochalasinB (5 pg/ml) and theophylline (4 mM), and was inhibited by both adrenalin (2 #M) and somatostatin (10 pg/ml). D-glyceraldehyde at a concentration of 1.5 mM produced a 10-fold increase of L-[ 4,5 -3 H] leucine incorporation into proinsulin and insulin without a significant increase into other islet proteins. Glucose at 1.5 mM did not stimulate proinsulin biosynthesis. D-Glyceraldehyde at concentrations higher than 1.5 mM, in marked contrast to glucose, progressively inhibited incorporation of labelled leucine into proinsulin + insulin and other islet proteins. DGlyceraldehyde also inhibited the oxidation of glucose. L-Glyceraldehyde did not stimulate proinsulin biosynthesis and had less effect than the D-isomer on insulin release and glucose oxidation. The results strongly suggest that metabolites below D-glyceraldehyde-3-P are signals for insulin biosynthesis and release. Interaction of D-glyceraldehyde with a "membrane receptor" cannot, however, be excluded with certainty.

Introduction Glucose is both an energy source and a specific stimulus for insulin secretion, insulin biosynthesis and ~-cell replication. This dual function of glucose obscures the initial molecular events in insulin secretion and biosynthesis, so that whether cell receptors exist to recognize glucose, some metabolite(s) of

385 glucose, or an enzyme substrate complex remains uncertain [1--6]. Furthermore, there is also uncertainty as to whether possibly existing receptors are located only in the cell membrane or are present in the cytosol. The rate of oxidation of monosaccharides b y the pancreatic islets correlates well with their effect on insulin secretion and biosynthesis [5,7]. Moreover, insulin secretion and biosynthesis are inhibited by inhibitors of glucose metabolism [8--10]. These experimental facts strongly support the view that insulin secretion and biosynthesis are initiated by the metabolism of the sugar. Experimental data have also been published suggesting a direct interaction between monosaccharides and receptors as the primary molecular event [11]. Ashcroft et al. [12] have shown that D-glyceraldehyde maintains levels of ATP as effectively as glucose in incubated mouse pancreatic islets. It also exhibited an insulinotropic effect. Hellman et al. [13] also reported studies of the effects of D-glyceraldehyde on glucose oxidation and insulin secretion by pancreatic islets of the obese hyperglycemic mouse (ob/ob). In experiments designed to study the effect of glyceraldehyde on proinsulin biosynthesis we found a marked inhibitory effect of racemic glyceraldehyde on glucose-induced proinsulin biosynthesis. This led us to investigate and compare the effects of the isomers of glyceraldehyde on glucose oxidation, insulin secretion, and insulin biosynthesis b y rat pancreatic islets.

Materials and Methods

Materials Collagenase was obtained from Worthington Biochemical Co. Ltd; crystallized bovine albumin from Miles Lab. Inc.; a-[4,5-3H]leucine, L-[4,5 -3 H]lysine and Aquasol from New England Nuclear Corporation; D-[U -~ 4C]glucose, hyamine, PPO and POPOP from Amersham-Searle Corporation; monopeak bovine insulin from the Connaught Laboratories; rat insulin from the Novo Research Institute; D-glyceraldehyde (as 90% syrupy solution in water) from Koch-Light Lab. Ltd; L-glyceraldehyde from Senn Chemical Co.; and somatostatin from Ayerst Co., Ltd. All other chemicals were obtained from Fisher Scientific Co., and were of the highest purity available.

Preparation of pancreatic islets Fed adult male and female Wistar rats were used and were killed by decapitation. The islets were isolated by a modification of the collagenase method of Lacy and Kostianovsky [14]. Two to four pancreases were processed simultaneously. 12--24 tubes containing 10--12 well-preserved islets free from exocrine tissue were assigned randomly to different treatments, 5--7 tubes per treatment. This permitted statistically valid results from a single experimental set.

Oxidation of glucose Output of 24 CO2 from uniformly labelled glucose was measured by a specially adapted radio-spirometric device. The procedure described in a previous publication was followed [15].

386

Insulin secretion. Static tube incubations Batches of 10--15 islets were incubated at 37°C in 0.3 ml of Krebs-Ringer buffer containing 4 mM glucose and 500 mg% albumin under continuous gassing with 95% 02 + 5% CO2. After 45 min these solutions were replaced with 0.5 ml of buffer containing the appropriate additions of glucose, glyceraldehyde and modifiers of insulin secretion. After 60 min the medium was separated from the islets by gentle centrifugation and kept frozen at --70°C until radioimmunoassay. The islets were extracted with 0.5 ml of cold acid/alcohol (64% ethanol, 31% Hz O, 2% HC1) for 48 h at 4°C. Insulin in the medium and in the extract was measured b y the back titration radioimmunoassay procedure of Wright et al. [16] using bovine insulin and guinea pig anti-bovine antibody. A solution of rat crystalline insulin containing 20 //units per 0.1 ml was repeatedly immunoassayed with this heterologous system and gave a mean value of 17.2 + 0.6 (n = 10). Perifusion of isolated islets A perifusion system slightly modified from that described by Lacy et al. [ 17 ] was used with plastic millipore holders serving as perifusion chambers. 30 islets were delivered on the millipore filter (pore diameter 8/am) through the opening of the screw top of the chamber with a siliconized transfer pipet. Buffer was continuously removed with a syringe attached b y a silicone tubing to the lower portion of the chamber. The fluid was maintained at the neck opening so that no air bubbles were introduced into the chamber. Constantly gassed perifusion medium at 37°C was fed into an air tight plastic syringe by gravity through a short polyethylene tube, maintaining at the b o t t o m of the syringe a constant volume of 0.5 ml. The syringe was attached to the perifusion chamber and both were immersed in a water bath at 37°C. The system was connected to a double-headed Harvard peristaltic pump with a narrow silicone rubber tubing. A steady flow rate of 0.8--0.9 ml per min was maintained. Four chambers were perifused simultaneously and allocated randomly to control perifusion solutions or solutions containing the c o m p o u n d s to be tested. Proinsulin and insulin biosynthesis The islets were preincubated in 300 /aml of Krebs-Ringer bicarbonate buffer (pH 7.35) containing 4 mM glucose and 100 mg% bovine albumin under a constant flow of 95% 02 + 5% CO2. After 45 min this medium was completely removed and replaced by 100 /al of buffer containing 20/aCi L-[4,5 -3 H ] leucine (40 Ci/mmol) or 20 pCi L-[4,5 -3 H] lysine (16.7 Ci/mmol) and 19 other naturally occurring aminoacids except the labelled one [18]. After a further incubation for 45 min the radioactive medium was removed and the islets washed thrice with cold buffer containing 3 mM leucine or lysine. The islets were extracted with 1 ml acid alcohol, for 48 h. Radioactive proinsulin and insulin was quantitatively precipitated by a double antibody procedure [18]. Incorporation of labelled amino acid into non proinsulin-insulin proteins of the extracted islets were estimated by trichloroacetic acid precipitation and subsequent trapping of the precipitate on 0.03-/am pore-size miUipore filters [18]. Expression of results In the insulin secretion experiment the total content in the islets of each

387 tube (secreted and extracted) was estimated. Secreted insulin was expressed per 103 punits of insulin content. This represented the approximate insulin content of a medium-sized islet used in our experiments. In the experiments with labelled glucose or amino acids an approximate volume of each islet was estimated b y measuring the two transverse diameters and the height of the islet using a calibrated grid in the ocular of the stereomicroscope. The total volume of islets in each tube was then calculated. Incorporation of L-[4,5 -3 H] leucine into proteins or 14 CO2 o u t p u t from [U -14 C]glucose was expressed per 107 #m 3 of islet volume. This value was chosen as it approximately corresponded to a medium-sized islet obtained from the rat pancreases used in the experiments. Results

Glucose oxidation D-Glyceraldehyde strongly inhibited the 14CO2 output from [U -14C]glucose by rat pancreatic islets. The effect was dependent on the concentration of D-glyceraldehyde. L-Glyceraldehyde exerted a smaller inhibitory effect (Table I). Insulin secretion In the perifusion system D-glyceraldehyde stimulated insulin secretion. The secretion profiles were closely similar to those obtained with glucose. An abrupt increase in the rate of insulin secretion was caused by switching from 4 mM glucose to 15 mM D-glyceraldehyde (Fig. 1). The secretion rate of insulin increased with time, but was not maintained as well as with high glucose in the final collection periods. When the perifusate of the first period consisted of 4 mM D-glyceraldehyde, the rate of insulin secretion by the end of 30 min was much higher than when the perifusate contained 4 mM glucose (Fig. 2). As a result, on switching from 4 mM to 15 mM D-glyceralde-

TABLE I E F F E C T O F D- A N D L - G L Y C E R A L D E H Y D E ON 14CO2 P R O D U C T I O N F R O M [UoI4C] G L U C O S E P a n c r e a t i c islets w e r e i n c u b a t e d for 9 0 m i n in 1 5 0 #1 of m e d i u m c o n t a i n i n g 1 ~tCi o f [ U - 1 4 C ] g l u c o s e . M e a n s + S.E. N u m b e r o f t u b e s in b r a c k e t s .

D-Glucose (raM)

Additions (raM)

14CO 2 f o r m a t i o n ( p m o l g l u c o s e / h p e r 10 7 p m 3 o f islet v o l u m e * )

4.5 4.5

-D-Glyceraidehyde

(15)

16 + 0.9 6 + 0.4

(6)** (6)**

15.O 15.0 15.0 15.0 15.0

-D-Glyceraldehyde D-Glyceraidehyde L-Glyceraldehyde Pyruvate

(15) (4) (15) (15)

85-+ 9.6 27 -+ 9.0 78 + 7.5 43 + 5.9 80 + 7.2

(17)** (18)** (6) (6) (6)

* This v o l u m e c o r r e s p o n d s a p p r o x i m a t e l y to an average-sized islet used in t h e e x p e r i m e n t s . **P~0.001.

388 B

A

26C

IL

24C

T

22C

2

o) 2©C

0

03

18C

TF--T-J

16C c

T , T

14C

E

12C o. c 10C D

8C

~

6c

~ D oL U

4C

TTT

T

T

n-

T

2C

//

i

30

/ 65

i

35 40 45

55

7a5

85

MID F i g . 1. E f f e c t o f D- a n d L - g l y c e r a l d e h y d e o n i n s u l i n r e l e a s e b y p e n f u s e d i s l e t s o f t h e r a t . F o u r p e r i f u s i o n c h a m b e r s c o n t a i n i n g 30 islets were p e r i f u s e d s i m u l t a n e o u s l y at a rate of 0 , 8 - - 0 . 9 m l ] m i n . On t h e 3 5 t h r a i n ( v e r t i c a l a r r o w ) S o l u t i o n A w a s c h a n g e d t o S o l u t i o n B. C o l l e c t i o n p e r i o d s are s h o w n i n t h e a b c i s s a . R a t e s o f i n s u l i n s e c r e t i o n o n t h e o r d i n a t e are a v e r a g e r a t e s f o r e a c h c o l l e c t i o n p e r i o d . M e a n s -+ S.E. f o r t h r e e p e r i f u s i o n e x p e r i m e n t s ( s i x c h a m b e r s ) f o r e a c h t r e a t m e n t are g i v e n . 4 m M g l u c o s e "-* 1 5 m M D~glyceraldehyde, ; 4 m M g l u c o s e "-* 1 5 m M L - g l y e e r a l d e h y d e , . . . . . .

hyde the change in the secretion rate of insulin was less striking than when 4 mM glucose was used in the first period of perifusion. These observations were corroborated with the static incubations of islets. D-Glyceraldehyde

A

200r 9 180P 0 160

....

i _

T, ~-

,

T

T, T

o. 140 E E 120 ~100 £ c E ~

2

60

T

T

T

T

T

20

//

i

30

35

/

d5

4 0 45

65

715

85

Min Fig. 2. E f f e c t o f D - a n d L - g l y c e r a l d e h y d e o n i n s u l i n r e l e a s e b y p e r l f u s e d i s l e t s o f t h e rat. F o r e x p e r i m e n t a l d e t a i l s s e e l e g e n d o f Fig. 1. 4 m M D - g l y c e r a l d e h y d e -* 1 5 m M D - g l y c e r a l d e h y d e , ; 4 mM Dg l y c e r a ] d e h y d e --* 1 5 m M b - g l y c e r a l d e h y d e . . . . . . .

389 T A B L E II EFFECT

OF GLUCOSE

RELEASE

AND OF THE ISOMERS

OF PANCREATIC

OF GLYCERALDEHYDE

ON RATES OF INSULIN

ISLETS OF RAT

D a t a o b t a i n e d w i t h i s l e t s f r o m t h e s a m e e o l i a g e n a s e - d i g e s t i o n o f r a t p a n c r e a s e s are g r o u p e d M e a n s +- S . E . N u m b e r o f t u b e s i n b r a c k e t s .

Concentration

a

D-Glyceraldehyde

--

d

--

15

4

c

L-Glyceraldehyde

4

--

b

Insulin secreted (~units]h per 10 3 punits of islet insulin**)

in medium (mM)

D-Glucose

6 9 -+ 2 . 6 *

--

--

111

--

± 9.7*

1 6 -+ 1 . 6 "

(6)

4

--

--

15

--

--

--

15

-3

---

---

--

2

--

14

± 2.8*

(6)

--

3

--

35

-+ 3 . 1 "

(6)

-

-

4

I01

± 6.2*

59

-+ 6 . 0 *

3 -+ 2 . 2 5 +- 1 . 4 "

(6) (6) (6)

(6) (6)

19 ± 2.7*

(6)

--

58

-+ 2 . 6 *

(5)

-

--

2 6 -+ 5 . 0

(6) (5)

--

4

-

--

15

--

77

± 3.6

(6)

--

--

15

2 3 -+ 5 . 1

(6)

20 20 20

--10

-10

8 9 -+ 2 . 2 88 ± 0.5 9 7 -+ 2.3

(6) (6) (6)

* P <

together.

-

-

0.001.

** A p p r o x i m a t e

insulin content of an average-sized islet.

proved to have a stronger secretagogic effect than glucose at low molarities (Table II). In the perifusion system L-glyceraldehyde had a much smaller effect on insulin secretion than D-glyceraldehyde (Figs 1 and 2). D-Mannoheptulose B 200F

g

T,T T'

T

[

T

n2olo. 100

8o

_c ,,- 6 0 o E

40

o 20 ~:

o

C., T : -

i

30

T T

:, i

i

35 4 0

T

i

45

55

65

5

85

MIN F i g . 3. E f f e c t o f D - m a n n o h e p t u l d s e o n g l u c o s e a n d D - g l y c e r a l d e h y d e - i n d u c e d i n s u l i n r e l e a s e b y p a n c r e a t i c i s l e t s o f t h e r a t . F o r e x p e r i m e n t a l d e t a i l s see l e g e n d o f F i g . 1 . 4 m M g l u c o s e -~ 1 5 m M D - g l y c e r a l d e h y d e + 10 mM D-mannoheptulose, ; 4 m M g l u c o s e .~ 2 0 m M g l u c o s e + 1 0 m M D - m a n n o h e p t u l o s e . . . . . . .

390 TABLE III EFFECT OF SOMATOSTATIN, ADRENALIN, CYTOCHALASIN RELEASE INDUCED BY GLUCOSE OR D-GLYCERALDEHYDE

AND THEOPHYLLINE IN RAT PANCREATIC

ON INSULINISLETS

M e a n s + S.E. N u m b e r o f t u b e s i n b r a c k e t s . I n h i b i t o r s a n d p o t e n t i a t o r s o f i n s u l i n r e l e a s e w e r e p r e s e n t o n l y during the incubation period.

Concentration in medium (mM) D-Glucose

D-Glyceraldehyde

15 15

---

--

15

--

15

Additions

-Somatostatin

Insulin secreted (/aunits/h per 1 0 3 p u n i t s o f i s l e t - i n s u l i n * *)

(10 #g/ml)

1 0 5 +58 +-

5.6 8.0*

(6) (6)

9 6 -+ 6 3 -+

7.0 2.1"

(6)

(10 #g/ml)

1 2 4 -+ 1 2 . 0 5 8 -+ 6 . 2 *

(6)

(6)

--

Somatostatin

--

15

--

15

--

15

--

15

C y t o c h a l a s i n (5 p g / m l )

+- 5.6 1 4 0 -+ 4 . 5 *

15 15

---

-C y t o c h a l a s i n (5 p g / m l )

110+1 7 3 +-

5.6 7.2*

(6) (6)

15 15

---

-T h e o p h y l l i n e (4 r a M )

8 4 +1 2 0 -+

5.7 4.9*

(6) (6)

--

15

--

15 * P <

--

(6)

A d r e n a l i n (2 p M ) --

103

--

Theophylline

(4 mM)

7 5 +9 8 +-

(8)

(6)

3.0

(6)

2.7*

(6)

0.001.

** A p p r o x i m a t e

insulin content of an average-sized islet.

(10 mM) added to the perifusate completely inhibited the effect of 2 0 mM glucose on insulin release (Fig. 3). D-Mannoheptulose did not have any obvious effect on the secretagogic effect of D-glyceraldehyde (Fig. 3). Results with static incubations of pancreatic islets are shown in Table II. A strong secretagogic effect of D-glyceraldehyde was again demonstrated; L-glyceraldehyde exhibited a much weaker b u t nevertheless definite effect. Adrenalin and somatostatin inhibit the glucose effect on insulin secretion [19,20], they also inhibited insulin release by D-glyceraldehyde. Cytochalasin B [21] and theophylline [22], both potentiators of glucose-induced insulin release, also potentiated the effect of D-glyceraldehyde (Table III). Incorporation o f L-[4,5 -3 H] leucine into proinsulin + insulin and non immunoprecipitable islet proteins Results are presented in Table IV. D-Glyceraldehyde at concentrations of 0.25--2 mM was much more effective than glucose in increasing the incorporation of L-[4,5 -3 H]leucine into proinsulin + insulin. The rate of proinsulin and insulin biosynthesis with 0.2--2 mM D-glyceraldehyde was about 7--10 times higher than that of islets incubated at zero or 1.5 mM glucose. The stimulatory effect of glucose on proinsulin + insulin biosynthesis appeared above concentra-

391 TABLE

IV

EFFECT INTO

OF

D-, L - G L Y C E R A L D E H Y D E

PROINSULIN

+ INSULIN

Means ± S.E. Number

Concentration

INTO

GLUCOSE

ON

NON-PROINSULIN

L-[4,5-3H] L E U C I N E + INSULIN

INCORPORATION

ISLET

PROTEINS

of tubes in brackets.

in medium

D-Glucose

AND

AND

dpm

(mM)

D-Glyceraldehyde

L-Glyceraldehyde

X 1 0 - 1 p e r 1 0 7 btm 3 o f i s l e t v o l u m e

Proinsulin and insulin

Other proteins

4

--

--

3 2 8 -+ 2 7

764 ± 46

(6)

--

0.25

--

1 3 6 -+ 1 5

6 3 6 -+ 5 7

(6)

6 9 6 -+ 6 8

(6)

331

(19)

--

--

2 5 9 -+ 2 3

--

1.5

--

8 +

1

1.5"

--

--

9-+

3

345-+ 53

(12)

4

--

--

167-+ 11

5 0 6 +- 7 6

(18)

+ 14

451 + 87

(6)

433 + 386-+ 226 + 224-* 235 ± 118-+

(6) (12) (6) (6) (6)

--

--

86

+45

--

0.5

------

1.0 1.5 3.0 6.0 10.0

------

--

15.0

--

--

--

1.5

7+

1

341

+ 32

(6)

--

--

4.0

I0-+

4

261

-+ 2 8

(6)

--

--

5-+

2

179

+ 41

(6)

9 3 +- 2 5 105-+ 4 7 9 -+ 2 3 43-+ 6 42-+ 4 18-+ 2

10.0

78 60 48 25 31 52

(6)

* Pooled data from separate experiments.

TABLE

V

EFFECT LEUCINE INSULIN

OF D-, L-GLYCERALDEHYDE OR

L-[4,5-3H] LYSINE

INTO

AND GLUCOSE PROINSULIN

ON THE INCORPORATION

+ INSULIN

AND

INTO

OF L-[4,5-3H]-

NON-PROII~SULIN

+

ISLET PROTEINS

M e a n -+ S . E . N u m b e r

Concentration

of tubes in brackets.

in medium

dpm X 10 -1 per 10 7 #m 3 of islet

(mM)

volume D-Glucose

D-Glycer-

L-Glycer-

D-Manno-

Proinsulin and

Other

aldehyde

aldehyde

heptulose

insulin

proteins 775 + 852-+ 600-+

a

20 20 20

-5 --

--5

----

285-+ 33 240-+ 18 212-+ I I

b

20 20 20

-10 --

--

----

410-+

-10 -----

-10 -10

196-* 13 19-+ 4 151-+ 13 152-+ I i 144-+ 15

c

5 5 ---

--1.5 1.5

d**

10

--

--

--

--

--

10

--

10

--

--

--

--

I0

--

--

* P ~ 0.001. ** E x p e r i m e n t s

with L-[4,5-3H] lysine.

6 145-+ 12 139-+ 13

31-+ 32±

2

(6) (6) (6)

991+ 153 447 + 11 3 6 3 -+ 1 3

(12) (12) (6)

*

498 + 347 + 368-+ 358 +

47 14 13 35

(6) (6) (6) (6)

•

618-+

50

(6)

279-+

42

(6)

635-+

65

(6)

230±

14

(6)

*

5

194-+ 13

47 52 68

.

392 tions of 1.5 mM. At 4 mM glucose it was 10--15 times higher than at 1.5 mM or zero glucose. D-Glyceraldehyde had its maximal stimulatory effect at concentrations between 1 and 2 mM. At concentrations above 2 mM, a progressive decrease in the rate of proinsulin + insulin biosynthesis was observed. Incorporation of radioactive leucine into proinsulin and insulin was approximately 5 times less and into islet proteins 3 times less at 15 mM than at 1.5 mM D-glyceraldehyde. In contrast to its D-isomer, L-glyceraldehyde at 1.5 and 4 mM did n o t affect the incorporation of L-[4,5-3H]leucine into proinsulin and insulin or other islet proteins. Both D- and L~glyceraldehyde at higher concentration (10 mM) markedly inhibited the stimulation of proinsulin + insulin biosynthesis by glucose. The effect was concentration dependent and appeared to be equally strong for both isomers (Table V). D-Mannoheptulose (10 mM) almost completely abolished the incorporation of L-[4,5 -3 H] teucine into proinsulin and insulin induced by 10 mM glucose, b u t it did not inhibit the effect of D-glyceraldehyde (Table V). Discussion In agreement with Ashcroft et al. [12] and Hellman et al. [13] D-glyceraldehyde proved a potent insulin releasing agent in rat pancreatic islets. At low concentrations, D-glyceraldehydewas a stronger secretagogue than glucose. A non-physiological mechanism of insulin release by D-glyceraldehydecan be excluded for the following reasons: (a) L-glyceraldehyde had a much smaller releasing effect; (b) insulin secretion reverted to base levels after cessation of perifusion with D-glyceraldehyde (15 mM) at the same rate as after glucose perifusion (Logothetopoulos, J. and Jain, K., unpublished); (c) a series of positive and negative modifiers of glucose-induced insulin secretion exerted similar effects on insulin secretion stimulated by D-glyceraldehyde [ 1 3 ] . In our experiments somatostatin and adrenalin inhibited, and theophyllin.e and cytochalasinB potentiated the effect of D-glyceraldehyde,as they did for glucose. D-Glyceraldehyde, in contrast to pyruvate, inhibited the o u t p u t of t 4 CO2 from uniformly labelled glucose by rat islets. A profile of steady-state concentrations of intermediates of glycolysis would be required to define the level and mechanism of this inhibition. In rat liver and adipose tissue the initial enzymatic reactions of D-glyceraldehyde metabolism have been discussed b y Landau et al. [23] and Antony et al. [24]. The phosphorylated derivatives, D-glyceraldehyde 3-phosphate, dihydroxyacetone phosphate and phosphoglyceric acid are formed. All three can feed into the Embden-Meyerhoff pathway. It is unlikely that glucose is synthesized from these three-carbon precursors leading to intracellular concentrations comparable to those established when islets are incubated in glucosecontaining media. L-Glyceraldehyde proved a weaker insulin secretagogue and inhibitor of glucose oxidation than the D-isomer. Its effect is unlikely to be due to a small content of D-glyceraldehyde in the commercial preparation or their solutions. Racemization does not seem to occur between D- and L-isomers [25,26]. L-Glyceraldehyde may feed into the glycolytic pathway by several metabolic modifications [23,24].

393 The effect of the t w o isomers on the incorporation of L-[4,5 -3 H] leucine into proinsulin + insulin and other non-proinsulin proteins are of special interest. L-Glyceraldehyde had no enhancing effect. D-Glyceraldehyde, at a concentration of 0.5 raM, produced a 10-fold increase in the incorporation of L-[4,5 -3 H] leucine into proinsulin + insulin. Incorporation into other islet proteins did not show a significant change. We did not measure the intracellular specific activities of L-[4,5-3H]leucine. However, the great increase in the incorporation of radioactive leucine into proinsulin and insulin compared with incorporation into other proteins indicates that specific changes in the net rate of proinsulin biosynthesis occur. In agreement with Pipeleers et al. [27] 1.5 mM glucose did not significantly stimulate proinsulin + insulin biosynthesis. A steep increase was observed at 4 mM glucose (12--15 fold). D-Glyceraldehyde at concentrations above 2 mM inhibited the incorporation of L-[4,5-3H] leucine into proinsulin + insulin and into other islet proteins. D- and L-glyceraldehyde at 10 mM concentration also inhibited the enhancing effect of 20 mM glucose on the incorporation of L-[4,5-3H] leucine into proinsulin + insulin and " o t h e r " islet proteins. The effects of high concentrations of the isomers on protein synthesis b y the islets cannot be attributed to an inhibition of leucine transport into the cells. Similar results were obtained with L-[4,5-3H]lysine (Table V) an amino acid which does not share the membrane-transport mechanism of leucine [28,29]. An inhibitory effect of high concentrations of the isomers on the incorporation of exogenous leucine into tissue proteins was also found with rat diaphragm-muscle and with pituitaries (Logothetopoulos, J. and Jain, K., unpublished). Incorporation of L-[4,5-3H] leucine into proteins of exocrine pancreatic tissue was not affected, however, by D- or L-glyceraldehyde at 10 mM concentrations. The stimulation of insulin secretion and biosynthesis b y D-glyceraldehyde strongly suggest that an interaction of a glucose metabolite below or at the level of D-glyceraldehyde-3-P with a putative membrane or cytosol receptor(s) may produce the signal for increased secretion and biosynthesis of insulin in the ~-cell. Control sites for insulin secretion at levels below glyceraldehyde-3-P have also been implicated in experiments with inhibitors of insulin secretion which were also inhibitors of glycolysis at the level of glyceraldehyde-3-P dehydrogenase or glycerokinase [30--33]. This would explain why D-mannoheptulose does not inhibit the action of D-glyceraldehyde on insulin secretion and synthesis but effectively blocks the effect of glucose by preventing its phosphorylation and possibly membrane transport. It could be proposed, however, that intact D-glyceraldehyde interacts with "receptors" producing the signal which triggers insulin secretion and biosynthesis in/3-cells which are also supplied with energy through the metabolism of the three-carbon unit [12]. Acceptance of this second "glucose-receptor" model would have to assume a higher affinity of the " r e c e p t o r " for D-glyceraldehyde than for D-glucose since D-glyceraldehyde stimulated secretion and synthesis of insulin at molarities at which glucose was without any effect. The threshold difference between glucose and D-glyceraldehyde would be easier explained, however, by a "metabolite-receptor" model assuming a higher rate of D-glyceraldehyde-3-P production from D-glyceraldehyde than from glucose. We have no explanation

394

to offer at present for the inhibitory effect of high concentrations of L- and I)-glyceraldehyde on protein synthesis. As discussed by Ashcroft et al. [12] a high rate of phosphorylation of D-glyceraldehyde may deplenish cells from ATP or GTP. It is possible that such a mechanism may prevent a stronger effect on proinsulin + insulin biosynthesis at high concentration of D-glyceraldehyde. Acknowledgement This work was supported by a grant from the Medical Research Council of Canada. Mr Hua Sik Lee contributed excellent technical assistance. References 1 Matschinsky, F.M., Landgraf, t~., Ellerman, J. and Kotler-Bra)tburg, J. (1972) Diabetes 21, 555--569 2 Ashcroft, S.J.H., Hedescov, C.J. and Randle, P~I, (1970) Biochem. J. l l S , 143--154 3 Matschinsky, F.M., Ellerman, J., K.rzanowski, J., Kotler-Brajtburg, J., Landgraf, R. and Fertel, R.. (1971) J. Biol. Chem. 246, 1007--1011 4 Randle, P~I. and Hales, C.N. (1972) in Handbook of Physiology (Steiner, D.F. and Freinkel, N., eds), Section 7, Vol. 1, pp. 219--235, American Physiological Society, Washington, D.C. 5 Ashcroft, S.J.H., Bassett, J.M. and Randle, P.J. (1972) Diabetes 21, 538--545 6 Dean, P.M. and Matthews, E.K. (1970) J. Physiol. 210, 255--264 7 Lin, B.J. and Haist, R.E. (1971) Can. J. Physiol. Pharmacol. 49, 559--567 8 Milner, R.D.G. and Hales, C.N. (1969) Biochem. J. 1 1 3 , 4 7 3 - - 4 7 9 9 Ashcroft, S.J.H. and Randle, P~I. (1970) Biochem. J. 119, 5--15 10 Lin, B.J. and Haist, R.E. (1969) Can. J. Physiol. Pharmacol. 4 7 , 7 9 1 - - 8 0 1 11 Matschinsky, F.M. and Ellerman, J. (1973) Biochem. Bioph. Res. Commun. 50, 193--199 12 Ashcroft, S.J.H., Weerasinghe, L.C.C. and Randle, P.J. (1973) Biochem. J. 132, 223--231 13 Hellman, B., Idahl, L.A., Lernmaxk, A., Sehlin, J. and Taljedal, I.oB. (1974) Arch. Biochem. Biophys. 162, 448--457 14 Lacy, P.E. and Kostianovsky, M. (1967) Diabetes 16, 35--39 15 Cole, E.H. and Logothetopoulos, J. (1974) Diabetes 2 3 , 4 6 9 - - 4 7 3 16 Wright, P.H., Makula, D.R., Vichick, D. and Sussman, K.E. (1971) Diabetes 20, 33--45 17 Lacy, P.E., Walker, M.M. and Fink, C.J. (1972) Diabetes 2 1 , 9 8 7 - - 9 9 8 18 Zucker, P. and Logothetopoulos, J. (1975) Diabetes 24, 194--200 19 Coore, H.G. and Randle, P.J. (1964) Biochem. J. 93, 66--78 20 Curry, D., Bennett, L.L. and Li, C.H. (1974) Biochem. Biophys. Res. Commun. 58, 885--889 21 Orci, B.L., Gabbay, K.H. and Malaisse, W.J. (1972) Science 175, 1128 22 Malalsse, W~I., Malaisse*Lagae, F. and Mayhew, D. (1967) J. Clin. Invest. 46, 1724--1734 23 Landau, B.R., Merlevede, W. and Williams, H.R. (1963) J. Biol. Chem. 2 3 8 , 8 6 1 - - 8 6 7 24 Antony, G., White, L.W. and Landau, B.R. (1969) J. Lipid Res. 10, 521--527 25 Meyerhof, O., Lohmann, K. and Shuster, P. (1936) Biochem. Z. 286, 319--332 26 Baer, E. and Fisher, H.O.L. (1939) J. Am. Chem. Soc. 6 1 , 7 6 1 - - 7 6 5 27 Pipeleers, D.G., Marichal, M. and Malaisse, W.J. (1973) Endocrinology 93, 1001--1011 28 Christensen, H.N. (1969) Some Special Kinetic Problems of Transport Adv. Enzymol. 32, 1--20 29 Lin, B.J. and Haist, R.E. (1973) Diabetes 22, Suppl. 1, Abstr. 323 30 Hellman, B. (1970) Diabetologia 6 , 1 1 0 - - 1 2 0 31 Georg, K.H., Sussman, K.E., Leitner, J.W. and Kirsh, W.M. (1971) Endocrinology 8 9 , 1 6 9 - - 1 7 6 32 Georg, R.H. and Sussman, K.E. (1971) Fed. Proc. 30, Abstr. 249 33 Pipeleers, D.G., Marichal, M. and Malaisse, W.J. (1973) Abstract Eighth Annual Meeting, European Association for the Society of Diabetes, Dlabetologia 9, 86