63

Biochimica et Biophysica Acta, 542 (1978) 63--76

© Elsevier/North-Holland Biomedical Press

BBA 28597 THE ROLE OF CALCIUM AND CYCLIC ADENOSINE 3',5'-MONOPHOSPHATE IN THE REGULATION OF GLYCOGEN METABOLISM IN PHAGOCYTOZING HUMAN POLYMORPHONUCLEAR LEUKOCYTES *

TROELS HERLIN, CARSTEN SAND PETERSEN and VIGGO ESMANN

Marselisborg Hospital, Department of Medicine, DK-8000 Aarhus C. (Denmark) (Received November 21st, 1977)

Summary Incubation of human polymorphonuclear leukocytes in a glucose-free KrebsRinger bicarbonate buffer for 2 h resulted in glycogen depletion, decreased phosphorylase activity and increased synthase-R activity. Addition of dialyzed latex particles to starved leukocytes revealed a very rapid ingestion rate (halfmaximal ingestion within 30 s). This uptake is followed by glycogenolysis associated with an immediate two-fold increase in phosphorylase a activity and a synthase-R to -D conversion within 30 s. Furthermore, in rapid time-course experiments with phagocytozing cells we found that the concentration of cyclic AMP increased by 93% within 15 s and returned to baseline values at 1 min. In a medium without added calcium and with 1 mM ethyleneglycol-bis(~-aminoethylether)-N,N'-tetraacetic acid, phagocytosis was blocked, cyclic AMP formation decreased by 50% and phosphorylase activation was abolished, but the conversion of synthase-R to -D was preserved. Addition of calcium ions to cells suspended in a calcium-free buffer without added latex results in phosphorylase activation and glycogenolysis, but not in cyclic AMP increase or synthase-R to oD conversion. Measurements of 4SCa efflux during phagocytosis suggest an initial increase in cytosolic calcium obtained by a release of membrane-bound 4SCa. Activation of phosphorylase during phagocytosis is thus presumably due to an increase in cytosol Ca 2+ and subsequent activation of phosphorylase kinase, and is independent of the simultaneous increase in concentration of cyclic AMP. Phosphorylation of synthase R to the D form does not depend on the presence of Ca 2÷ in the extracellular phase.

* Part of this w o r k has b e e n p r e s e n t e d in a b s t r a c t f o r m at the 1 1 t h FEBS Meeting, C o p e n h a g e n , 1 9 7 7 [ 3 5 ] , A b b r e v i a t i o n s cAMP, cyclic AMP; E G T A , e t h y l e n e g l y c o l - b i s ( / 3 - a m i n o e t h y l e t h e r ) - N , N ' - t e t r a a c e t i c acid.

64 Introduction Glycogen synthase exists in human polymorphonuclear leukocytes in a phosphorylated, physiologically inactive, D form and a non-phosphorylated, active, I form [1,2]. It has been realized, however, that the activity of synthase-! present in intact cells in insufficient to account for glycogen synthesis under several metabolic conditions and that, instead, a kinetically defined, partially phosphorylated enzyme form, synthase-R, with an activation constant for the Glc-6-P intermediate between that of synthase-I and -D was responsible for glycogen synthesis [3]. Similarly, with the purified muscle enzyme, the M0s value for Glc-6-P has been found to be a function of the alkali-labile phosphate c o n t e n t [4] and it has been discovered that specific sites of the subunits of the enzyme are phosphorylated by cAMP-dependent and cAMP-independent protein kinases [5,6]. These kinases are also present in leukocytes, [7], but their exact contributions to the phosphorylation of the I-enzyme and the number and positions of phosphate groups of the R-enzyme (and thus whether there are more than just three discrete species) remains to be established. It has been observed with the glycogen particle, however, that the initial rate of the I to R conversion is increased by cAMP (Saugmann, P., unpublished observations). Recent studies [8,9] have revealed t h a t glycogen breakdown during phagocytosis is associated with an immediate activation of phosphorylase and inactivation of synthase-R (i.e., R to D conversion). These coordinated and inverse changes upon ingestion of latex particles might suggest a preceding activation of the cAMP-dependent protein kinase. Indeed, a cAMP elevation has been found 5 min after the addition of latex particles [10], but this effect was refuted by working with a more purified preparation of polymorphonuclear leukocytes [11], and it was shown that the cAMP stimulation revealed in the primer work was caused by contaminating monocytes. However, an investigation of the initial ingestion rate of lipopolysaccharide particles and paraffin oil by guinea pig granulocytes [12,13] has shown that the omission of Ca 2÷ and Mg 2÷ from the medium partially inhibited the ingestion rate and that the addition of EDTA completely blocked the engulfment of the particles. Furthermore, studying chemotaxis in polymorphonuclear leukocytes, Boucek and Snyderman [14] measured an influx of Ca 2÷. These observations have led to our assumption of a Ca-dependent mechanism for engulfment during phagocytosis and further to the obligate requirement of Ca > to activate phosphorylase through activation of phosphorylase kinase. This is similar to the known Ca 2+ dependency of contraction and phosphorylase activation in skeletal [15--17] and heart muscle [18,19]. The present work was undertaken to elucidate the role of Ca > and cAMP during phagocytosis and subsequent glycogenolysis. Methods and Materials

Experimental and analytical procedures Human polymorphonuclear leukocytes were isolated [20] and incubated at 37°C in a glucose-free Krebs-Ringer bicarbonate buffer (although with 1.3 mM

65 CaC12 unless otherwise mentioned). Under aerobic conditions (94.4% 02/5.6% CO2) 10--15 ml suspensions were incubated with shaking in siliconized conical flasks in a final cell concentration of 1.2 • 107 cells/ml. After a period of 2--2.5 h, 500 pl latex particles per 10 ml Krebs-Ringer bicarbonate buffer were added and the reactions were followed over a 30-min period. At intervals, 900-pl samples were withdrawn using siliconized pipettes and were mixed into 150 pl prechilled buffer ( 50 mM Tris. HC1, pH 7.6 at 20°C, 50 mM EDTA and 300 mM NaF) and crude homogenates were made by sonic disruption in the cold. Routinely, the concentration of glycogen and protein and the activities of glycogen synthase and phosphorylase were assayed from the homogenates. Glycogen synthase was determined according to Thomas et al. [21] but with 0.4 mM UDP-[U-14C]glucose ( 3 - - 4 . 1 0 6 cpm/pmol). Synthase-I was assayed in the absence of Glc-6-P and Na2SO4, synthase-R was assayed with 0.67 mM Glc-6-P, which constitutes 85% of synthase-R and 50% of synthase-I) activity [9], and finally total synthase activity was assayed with 67 mM Glc-6-P. The activity was expressed in munits/mg protein equal to nmol glucose incorporated per min per mg protein at 30 ° C. Glycogen phosphorylase a was determined in the direction of glycogen synthesis [22] with 10 mM [U-14C]glucose-l-phosphate (3--4. l 0 S cpm/pmol). 1 munit corresponds to I nmol glucose incorporated per min at 30°C. Recently, in kinetic studies of leukocyte phosphorylase [23], it has been found that Km for Pi is 85--100 mM for the phosphorylase b form and consequently this enzyme is totally inactive in vivo. Therefore, it is sufficient to consider changes in the activity of phosphorylase a. Determination of the glycogen concentration was performed with a sensitive filter paper technique [24]. Protein was measured by a Lowry method with human albumin as a standard. cAMP determinations. The leukocytes were incubated in a Krebs-Ringer bicarbonate buffer adjusting the cell concentration to 2 . 4 . 107/ml and assays for cAMP determinations were made according to Capito and Hedeskov [25], although with some modifications. At appropriate times, 300 pl of the suspension were placed in glass tubes immersed in a boiling water bath and 300 pl of boiling phosphate buffer, containing 0.25 M potassium phosphate buffer, pH 5.5, 2 mM theophylline and 6 mM 2-mercaptoethanol, were added. The tubes were then boiled for 10 min, whereafter t h e y were kept on ice for the rest of the procedure. After sonication for 15 s, the denaturated protein was removed by centrifugation (2000 rev./min for 5 min at 4°C) and the supernatants were stored at --20°C or assayed immediately. 100-pl samples were assayed by the protein-binding competition m e t h o d described by Gilman [26]. Recovery of a standard solution (1 or 10 pmol/100 pl) after this procedure was 97.9 + 3.35% (S.E., n = 6) and 93.8 + 8.87% (n = 6), respectively. 4SCa efflux during phagocytosis. The cells were isolated as described above, although t h e y were resuspended in a Ca- and Mg-free buffer. The cells were then labeled with 4SCa by incubation with 83 pCi 4SCa per 1 • 108 cells (7.3 Ci/g Ca equal to a final concentration of 33 pM CaC12) for 1 h at 37°C. The cells were washed once with 50 ml of a Ca- and Mg-free buffer and twice with 50 ml of a normal Krebs-Ringer bicarbonate buffer (1.2 mM Mg2÷ and 1.3 mM Ca2*).

66 The concentration was then adjusted to 1.2 • 107 cells/ml using normal KrebsRinger bicarbonate. After 15 min, latex particles were added to the cells and at various intervals 450 pl were withdrawn and instantly placed on a 0.45-pro Millipore filter under constant suction. The retained cells were washed with 1 ml of Krebs-Ringer bicarbonate buffer and calcium efflux was determined by the a m o u n t of 4SCa present per ml of filtrate. The efflux is expressed as a percentage of the filtrate activity at zero time. The rate o f uptake o f latex particles. A modification of the method of Roberts and Quastel [27] was adopted to evaluate the rates of ingestion of latex spherules. Phagocytosis was initiated by the addition of 500 pl latex to 10 ml suspension, giving a cell : latex ratio of approx. 1 : 100. Then 500 pl of the suspension were pipetted into 10 ml ice-cold saline to stop phagocytosis. The cells were centrifuged and washed three times at 4°C with saline to remove extracellular latex. After removal of the supernatant, the cell pellet was resuspended in 0.5 ml isotonic NaC1, 50 pl were placed in 2 ml Methyl Violet and intracellular latex particles were counted with a 1000-times magnification. The cell pellet, without Methyl Violet, was then air-dried, the ingested latex particles were extracted from the cells with 3 ml p-dioxan, and the quantity of latex was determined spectrophotometrically at 253 nm against p-dioxan as blank. The tendency of polystyrene beads to stick to the surface should be prevented by the washing procedures. Although the inaccuracy in evaluating phagocytosis by counting the particles has been at issue [12], we found it relatively easy to distinguish between intra- and extracellular particles.

Materials Latex beads (Difco, 0.81 t~m) were dialyzed for 24 h, changing the dialysate three times. They were dialyzed against either distilled water or a Krebs-Ringer bicarbonate buffer without or with 1.2 mM MgC12, 1.3 mM CaCl~, I mM EDTA or 1 mM EGTA (see later for details), cAMP test kits were purchased from Boehringer, Mannheim. [U-14C]Glucose 1-phosphate and UDP-[U-lac]glucose were obtained from the Radiochemical Centre, Amersham. 4SCa was from Rise Isotoplaboratorium, Denmark. Other chemicals mentioned were of analytical grade and either from Sigma or Merck. Results

Starvation o f leukocytes for 2 h Freshly isolated leukocytes have almost all synthase in the D form and most phosphorylase in the a form. Incubation of the cells for 2 h without glucose induces a state sensitive to both activating and inactivating stimuli. When the cells were incubated in a normal, Ca- and Mg-containing Krebs-Ringer bicarbonate buffer, a glycogen breakdown of 29% (Table I), a decrease in phosphorylase a activity of 23% and an increase in synthase-R of 120% (Table II) were obtained. Synthase-I activity remained low (data not shown), but a small increment from 0.04 + 0.007 munits/mg (n = 5) to 0.075 + 0.010 munits/mg protein (n = 16, P < 0.01) was noted as a consequence of starvation and corresponded to 1--2% of total glycogen synthase activity. When the leukocytes, after separation from plasma, were washed twice and

67 TABLE

I

CHANGES IN GLYCOGEN VATION PERIOD OF 2 h

CONTENT

IN DIFFERENT

Mean of duplicate determinations in n experiments. MgSO 4; +EDTA, 1 mM EDTA; +EGTA, 1 mM EDTA. Krebs-Ringer buffer

bicarbonate

Glycogen

+Ca +Mg --Ca +Mg --Ca--Mg --Ca --Mg +EDTA --Ca +Mg +EGTA

INCUBATION

Abbreviations:

concentration

BUFFERS

DURING

A STAR-

+Ca, 1 . 3 m M C a C I 2 ; + M g , 1 . 2 m M

( ~ g p e r 1 0 8 cells ± S . E . )

Decrease in glycogen conten

0 min

n

120 min

n

(%)

801 + 52 810 722 707 699 660 706 6 1 1 _+ 4 6

5

5 9 0 _+ 1 8 5 4 1 -+ 2 5

16 8

29 26

523 ± 15

6

26

498 + 47

7

27

4 4 7 -+ 21

4

27

2 2

2 3

prepared in a Ca-free or a Ca- and Mg-free buffer (Table II), it was observed that the initial activity of phosphorylase a was low, particularly when Mg was also removed, and did not decrease further during the subsequent 2 h of incubation. Fig. 1 illustrates that the effect on phosphorylase a activity of completely removing Ca from Ca-free buffer by adding EGTA is rapid (15 min). Synthase-R activity increased normally during incubation in the absence of Ca from the buffer, b u t the response was conspicuously absent when the Mg was also removed (Table II). In spite of the depression of phosphorylase a activity seen in Ca- and Mg-free buffers, there was no change in the rate of glycogen b r e a k d o w n during the incubation (Table I).

TABLE

II

CHANGES IN ENZYME DURING A STARVATION

ACTIVITIES OF PERIOD OF 2 h

GLYCOGEN

n samples each representing duplicate determinations. legend to Table I for concentration of ions. Krebs-Ringer bicarbonate buffer

Phosphorylase a (munits/mg protein

+ S.E.)

0 min

n

120 min

+Ca +Mg

7.89 ± 0.67

6

--Ca +Mg

5.43 4.10 3.43 3.32

6.10 ± 0.42 (P < 0.05) 3.97 + 0.27

--Ca +Mg +EGTA

---Ca - - M g ---Ca - - M g + E D T A

1.45 2.43 1.53 1.47

2 2

2 2

2.97 2.61 (0.05

PHOSPHORYLASE

a AND

SYNTHASE

R

Significant changes are indicated by a P value. See

Synthase R (munits/mg protein

+ S.E.)

n

0 rain

n

120 rain

n

20

0 . 3 7 -+ 0 . 0 3

6

0.81 + 0.07 (P < 0.001) 0 . 4 1 _+ 0 . 0 4 ( < 0.01) 0.65 0.60

18

7

2

i o