J. Steroid Biochem. Molec. Biol. Vol. 42, No. 6, pp. 617-623, 1992
0960-0760/92 $5.00+ 0.00 Copyright ~ 1992Pergamon Press Ltd
Printed in Great Britain.All rights reserved
LOW SODIUM INTAKE ENHANCES SENSITIVITY OF I1-DEOXYCORTISOL AND DEOXYCORTICOSTERONE TO ACTH IN ACTH-SUPPRESSED NORMAL SUBJECTS CLAUDIO E. KATER, EDWARDG. BIGLIERI* and ILAN IRONY San Francisco General Hospital Medical Center and the General Clinical Research Center, University of California, San Francisco, CA 94110, U.S.A. and the Division of Endocrinology Escola Paulista de Medicina Saio Paulo, SP 04034, Brazil (Receh~ed 10 July 1991; received for publication 27 March 1992)
Summary--Continued administration of ACTH to patients with hypopituitarism produced normal increases in steroids dependent on microsomai cytochrome P4502t and P450jT~ but reduced responses of steroids dependent on mitochondrial cytochrome P4501t~.zs. To explore possible mechanisms and to determine whether this dissociation occurs with short-term ACTH suppression, we have examined the steroid responses to ACTH after 1 h in 12 normal subjects after equilibration on sodium intakes of 124 mmoi/d [normal sodium diet (NSD)], 22 mmol/d [low sodium diet (LSD)], and 240 mmol/d [high sodium diet (HSD)] before and during continued ACTH suppression with dexamethasone (DEX). Two distinct patterns of steroid responses were observed. Deoxycorticosterone (DOC) responses were initially reduced during LSD-DEX but eventually returned to the NSD-control (NSD-CONT) values; in contrast 18-hydroxydeoxycorticosterone and corticosterone remained suppressed, ll-Deoxycortisol and 21-deoxycortisoi showed patterns similar to DOC, with a return to normal ACTH responses on LSD-DEX. Basal cortisoi levels were reduced and the ACTH response was unchanged by LSD. HSD-DEX reduced basal levels of all steroids as well as their ACTH responses. LSD and/or increased activity of the renin-angiotensin system have a significant impact on 17a- and 21-hydroxylation functions in the zona fasciculata to maintain a normal ACTH response of microsomally dependent steroids under these conditions. In contrast, on HSD-DEX with the renin-angiotensin system suppressed, there is generalized impairment of steroid responses to ACTH.
INTRODUCTION
There are three histologic and physiologically distinct zones in the adrenal cortex that secrete steroid hormones. Aldosterone production by the zona glomerulosa (ZG) is primarily regulated by the renin-angiotensin system (RAS). In the ACTH-dependent zona fasciculata (ZF) the 17-hydroxy steroid biosynthetic pathway produces cortisol (F), l l-deoxycortisol (S), and 21-deoxycortisol (21-dF). The ZF 17-deoxy pathway produces deoxycorticosterone (DOC), corticosterone (B), and 18-hydroxydeoxycorticosterone (18-OHDOC) under the control of both ACTH and a second possible regulator [1]. DOC and B are obligatory intermediates in both ZG and ZF 17-deoxy pathways, but their plasma levels and circadian rhythms reflect *To whom correspondence should be addressed at: Clinical Study Center, Building I00, Room 321, San Francisco General Hospital, I001 Potrero Avenue, San Francisco, CA 94110, U.S.A.
ZF origin and ACTH dependence [1-3]. Under normal conditions, the Z G contribution to plasma levels of DOC and B is insignificant. Multiple intracellular mechanisms are involved in ACTH stimulation of steroidogenesis. Most important is increased synthesis of mRNA for cytochrome P450 hydroxylating enzymes [4-9]. Other factors, such as the stimulation of synthesis of sterol carrier proteins, cholesterol esterases, and proteins responsible for synthesis of phospholipids, or inhibition of cholesterol ester synthetase are involved in transporting cholesterol into the mitochondrion [10-12]. All of these actions make more cholesterol available for steroid production. In the absence of ACTH, as in hypopituitarism or chronic ACTH suppression by glucocorticoid analogs, ZF function in reduced with diminished protein synthesis, steroidogenic enzymes, and 17-hydroxy steroid output [12]. While ACTH affects the ZG, RAS preserves aldosterone production in conditions where
617
618
CLAUDIO E. KXI~R et al.
ACTH is absent or suppressed [13]. The mechanisms involved in the RAS regulation of ZG steroidogenesis differ from those of ACTH [14-16]. The steroids in the ZF 17-deoxy pathway are virtually identical to those in the ZG pathway and it might be argued that angiotensin II could participate in the control of ZF "mineralocorticoids" [17]. However, the ZF lacks corticosterone methyloxidase type II which is responsible for 18-dehydrogenation and aldosterone production. Previous observations from our group [18, 19] have suggested that non-ACTH factors participate in the regulation of the 17-deoxy pathway, along with ACTH, to achieve the normal responses of the 17-deoxy steroids to ACTH stimulation. Additional studies implied that the hypothalamus-pituitary region could be the source of a regulating factor [20]. In patients with panhypopituitarism [21], ZF microsomal cytochromes P4502j and P450t7 = are preserved. Similarly, steroids dependent on mitochondrial cytochrome P450t,~..s demonstrated reduced responses after continued ACTH stimulation. The present studies were designed to evaluate the influence of sodium restriction and high sodium intakes in maintaining ACTH responsiveness of specific ZF steroid hydroxylating functions in normal subjects during suppression of endogenous ACTH by dexamethasone (DEX). ACTH administered to subjects on a low sodium diet (LSD) increases ZF steroidogenesis of steroids dependent on microsomal cytochromes P45017 and P45021, whereas increases in those steroids requiring mitochondrial cytochrome P450~,B.~a are markedly impaired. High sodium intake reduces the ACTH response of all steroids. EXPERIMENTAL
Twelve healthy volunteers [11 male/l female, 21-55 years of age (median age of 34.5)] were admitted to the General Clinical Research Center at the San Francisco General Hospital Medical Center. Informed written consent was obtained according to the requirements of the Committee on Human Research of the University of California, San Francisco. Six participated in the normal and LSD studies and six different subjects in the high sodium diet (HSD) studies. The first study protocol consisted of testing adrenocortieal steroid sensitivity to exogenous ACTH on three different sequential metabolic
conditions: (I) normal sodium diet control period (NSD-CONT); (2) NSD-DEX suppression, and (3) LSD-DEX suppression. As a separate study, after control values were established, a HSD was initiated together with DEX suppression (HSD-DEX). Each study period lasted 5 to 8 days. In study 1, six subjects underwent the NSDCONT protocol. They ingested a fixed metabolic diet containing approx. 22mmol of sodium and 70-90 mmol of potassium per day, to which sodium chloride (NaCI) tablets were added to establish a sodium intake of approx. 124mmol/d. After 5-7 days of NSD-CONT, they were given DEX (0.5 mg p.o. every 6 h), which was continued throughout the remainder of the study (NSD-DEX period). Sodium tablets were discontinued after 7 days to establish a LSD (22mmol sodium/d) (LSD-DEX period). As a separate study, 6 additional normal subjects ingested similar diets, NSD-CONT 2, followed by HSD either be receiving additional NaCI tablets (270mmol sodium/d, n = 3), or NaCi tablets (165mmol sodium/d) plus fludrocortisone (Florinef, 0.2 mg p.o. every 12h, n =3) plus DEX (0.5 p.o. every 6h) HSD-DEX. To determine adrenal responsiveness to a maximum ACTH stimulus during endogenous ACTH suppression (NSD-DEX period) and during RAS activation (LSD-DEX period) or suppression (HSD-DEX period), we performed acute stimulation tests every other morning throughout the study. Cosyntropin (Cortrosyn, 250 #g i.v. bolus) was administered to fasting and overnight recumbent subjects and blood samples were collected before and 60 min after the injection. Sodium balance and ACTH suppression were monitored daily in all subjects by: (1) changes in body weight (BW), (2) 24-h urinary excretion of sodium and potassium, (3) 24-h urinary excretion rates of free cortisol (UFF) and free aldosterone (UFA), and (4) plasma renin activity (PRA), and sodium and potassium concentrations were determined every other day. Plasma and urinary electrolytes were measured by routine internal standard flame photometry. PRA was measured by the RIA of generated angiotensin I [22]. After extraction and HPLC isolation, UFF was measured by a competitive protein-binding assay and UFA was measured by specific RIA[2]. Plasma F, S, 21dF, B, DOC, 18-OHDOC, 18-OHB, and aldosterone were measured by RIA in all samples following
619
Low sodium intake enhances sensitivityof normal subjects extraction, solvent partitioning, and HPLC separation. Methodology and reference values for this laboratory have been published [2, 22]. All samples from the same subject were assayed as a single batch after completion of the study. For statistical purposes all non-detectable values were arbitrarily considered equal to the limit of sensitivity for each particular assay, divided by the square root of 2 (1.412)[2]. Non-parametric analysis of variance (the Kruskall-Wallis rank sum test) was used to compare data. When significant, ANOVA was complemented by the Scheffe's and Dunn's tests of contrasts to ascertain differences between study periods [23]. RESULTS
Sodium and potassium balance
Although a different group of normal subjects was used for the high sodium protocol, all steroid values were virtually identical. In Fig. 2, the steroid responses to ACTH study 2 were compared to their own NSD-CONT study 2. Cortisol, S and 21 dF (the Z F 17-hydroxy pathway) (Table 1). All three steroids showed virtually complete suppression of basal levels when dexamethasone was given: NSD-, LSD-, and HSD-DEX. Plasma F demonstrated relatively fixed responses to ACTH stimulation. The control response of 632 + 22 nmol/l decreased significantly to 367 + 36 (NSD + DEX) nmol/1. Surprisingly, the response at the end of the HSD-DEX period was even further and significantly reduced to 254 _+ 28 nmol/l (see Table 1). Both basal S and 21 dF were significantly suppressed while on DEX and their responses to ACTH did not significantly decrease during
Daily urinary sodium, UFA and U F F during the study periods are depicted in Fig. 1. The NSD-DEX period (average sodium intake of 124mmol/d) was associated with moderate sodium retention and an average BW gain of 1% compared with the NSD-CONT period. [= Sodium balance was achieved on the 4th day of LSD-DEX resulting in a BW reduction of approx. 1.5-2%, and on the 6th day of the HSD-DEX study resulting in a BW gain of 2-3% (Fig. 1). Plasma sodium and potassium concentrations (Table 1) showed no significant 275o change during study periods except for a significant decrease in plasma potassium during HSD-CONT 2. i lSS0
UFA, UFF excretion and PRA UFA showed minimal change during NSDCONT and -DEX and incremental increases as the LSD-DEX period progressed. On HSDDEX, UFA significantly decreased by half from its NSD-CONT period (Fig. 1). PRA levels (Table 1) showed little change during the NSDCONT and -DEX, but increased significantly on LSD-DEX and were reduced by HSD-DEX. U F F excretion was reduced during all DEX treatment period. Increases in U F F and UFA were seen on the days the 1 h ACTH test was administered (Fig. 1).
Plasma steroid responses to ACTH on different sodium intakes Table 1 and Fig. 2 show the basal and 60 rain responses of plasma steroids to the ACTH test performed at the end of each study period.
UFA
I 1110
/
0
275[" UFF
UFF urinary~ e corUsol
UFAurinaryfree|idosmroneI I
T
16S 110
~
"fdl! 0
i
.
i .
IZX
.
.
,:
SUPPRESSION
i I
Fig. 1. Twenty-four hour urinary sodium (retool/d) UFA
and UFF excretion during each treatment period in 6 normal subjectson NSD and LSD, and 6 on HSD. *Day I h ACTH t~t performed.
620
CLAUDIO E . KATER
+1 + l +1 + t +1 +t ~
the NSD-DEX period (3.05+ 1.0 and 1.18 + 0.3nmol/1). By the end of the LSD-DEX period, S and 21 dF were higher (4.7 + 1.5 and 1.9 +0.4nmol/l) than those of NSD-CONT (4.17 + i.5 and 1.4 + 0.3 nmol/l), respectively. While no significant change in S or 21 dF occurred comparing the NSD-CONT to LSDDEX, F peaks were markedly reduced. Although DEX was continued for a second week during the LSD period, basal levels of ZF steroids showed no further decrease. Like F, the HSD-DEX period showed significant impairment of S and 21dF responses to ACTH (1.3 + 0.2 and 0.94 + 0.12 nmol/l, respectively).
+1 +1 +1 +1
~~-~~ ,-e.
+1+1+1+1+1+1+1+~+~+=+1
d z
~ ' ~ 2
. . . .
-=
÷1 +1 ÷1 +1 +1 +1 +1 +1 +l +1 +l
0960-0760/92 $5.00+ 0.00 Copyright ~ 1992Pergamon Press Ltd
Printed in Great Britain.All rights reserved
LOW SODIUM INTAKE ENHANCES SENSITIVITY OF I1-DEOXYCORTISOL AND DEOXYCORTICOSTERONE TO ACTH IN ACTH-SUPPRESSED NORMAL SUBJECTS CLAUDIO E. KATER, EDWARDG. BIGLIERI* and ILAN IRONY San Francisco General Hospital Medical Center and the General Clinical Research Center, University of California, San Francisco, CA 94110, U.S.A. and the Division of Endocrinology Escola Paulista de Medicina Saio Paulo, SP 04034, Brazil (Receh~ed 10 July 1991; received for publication 27 March 1992)
Summary--Continued administration of ACTH to patients with hypopituitarism produced normal increases in steroids dependent on microsomai cytochrome P4502t and P450jT~ but reduced responses of steroids dependent on mitochondrial cytochrome P4501t~.zs. To explore possible mechanisms and to determine whether this dissociation occurs with short-term ACTH suppression, we have examined the steroid responses to ACTH after 1 h in 12 normal subjects after equilibration on sodium intakes of 124 mmoi/d [normal sodium diet (NSD)], 22 mmol/d [low sodium diet (LSD)], and 240 mmol/d [high sodium diet (HSD)] before and during continued ACTH suppression with dexamethasone (DEX). Two distinct patterns of steroid responses were observed. Deoxycorticosterone (DOC) responses were initially reduced during LSD-DEX but eventually returned to the NSD-control (NSD-CONT) values; in contrast 18-hydroxydeoxycorticosterone and corticosterone remained suppressed, ll-Deoxycortisol and 21-deoxycortisoi showed patterns similar to DOC, with a return to normal ACTH responses on LSD-DEX. Basal cortisoi levels were reduced and the ACTH response was unchanged by LSD. HSD-DEX reduced basal levels of all steroids as well as their ACTH responses. LSD and/or increased activity of the renin-angiotensin system have a significant impact on 17a- and 21-hydroxylation functions in the zona fasciculata to maintain a normal ACTH response of microsomally dependent steroids under these conditions. In contrast, on HSD-DEX with the renin-angiotensin system suppressed, there is generalized impairment of steroid responses to ACTH.
INTRODUCTION
There are three histologic and physiologically distinct zones in the adrenal cortex that secrete steroid hormones. Aldosterone production by the zona glomerulosa (ZG) is primarily regulated by the renin-angiotensin system (RAS). In the ACTH-dependent zona fasciculata (ZF) the 17-hydroxy steroid biosynthetic pathway produces cortisol (F), l l-deoxycortisol (S), and 21-deoxycortisol (21-dF). The ZF 17-deoxy pathway produces deoxycorticosterone (DOC), corticosterone (B), and 18-hydroxydeoxycorticosterone (18-OHDOC) under the control of both ACTH and a second possible regulator [1]. DOC and B are obligatory intermediates in both ZG and ZF 17-deoxy pathways, but their plasma levels and circadian rhythms reflect *To whom correspondence should be addressed at: Clinical Study Center, Building I00, Room 321, San Francisco General Hospital, I001 Potrero Avenue, San Francisco, CA 94110, U.S.A.
ZF origin and ACTH dependence [1-3]. Under normal conditions, the Z G contribution to plasma levels of DOC and B is insignificant. Multiple intracellular mechanisms are involved in ACTH stimulation of steroidogenesis. Most important is increased synthesis of mRNA for cytochrome P450 hydroxylating enzymes [4-9]. Other factors, such as the stimulation of synthesis of sterol carrier proteins, cholesterol esterases, and proteins responsible for synthesis of phospholipids, or inhibition of cholesterol ester synthetase are involved in transporting cholesterol into the mitochondrion [10-12]. All of these actions make more cholesterol available for steroid production. In the absence of ACTH, as in hypopituitarism or chronic ACTH suppression by glucocorticoid analogs, ZF function in reduced with diminished protein synthesis, steroidogenic enzymes, and 17-hydroxy steroid output [12]. While ACTH affects the ZG, RAS preserves aldosterone production in conditions where
617
618
CLAUDIO E. KXI~R et al.
ACTH is absent or suppressed [13]. The mechanisms involved in the RAS regulation of ZG steroidogenesis differ from those of ACTH [14-16]. The steroids in the ZF 17-deoxy pathway are virtually identical to those in the ZG pathway and it might be argued that angiotensin II could participate in the control of ZF "mineralocorticoids" [17]. However, the ZF lacks corticosterone methyloxidase type II which is responsible for 18-dehydrogenation and aldosterone production. Previous observations from our group [18, 19] have suggested that non-ACTH factors participate in the regulation of the 17-deoxy pathway, along with ACTH, to achieve the normal responses of the 17-deoxy steroids to ACTH stimulation. Additional studies implied that the hypothalamus-pituitary region could be the source of a regulating factor [20]. In patients with panhypopituitarism [21], ZF microsomal cytochromes P4502j and P450t7 = are preserved. Similarly, steroids dependent on mitochondrial cytochrome P450t,~..s demonstrated reduced responses after continued ACTH stimulation. The present studies were designed to evaluate the influence of sodium restriction and high sodium intakes in maintaining ACTH responsiveness of specific ZF steroid hydroxylating functions in normal subjects during suppression of endogenous ACTH by dexamethasone (DEX). ACTH administered to subjects on a low sodium diet (LSD) increases ZF steroidogenesis of steroids dependent on microsomal cytochromes P45017 and P45021, whereas increases in those steroids requiring mitochondrial cytochrome P450~,B.~a are markedly impaired. High sodium intake reduces the ACTH response of all steroids. EXPERIMENTAL
Twelve healthy volunteers [11 male/l female, 21-55 years of age (median age of 34.5)] were admitted to the General Clinical Research Center at the San Francisco General Hospital Medical Center. Informed written consent was obtained according to the requirements of the Committee on Human Research of the University of California, San Francisco. Six participated in the normal and LSD studies and six different subjects in the high sodium diet (HSD) studies. The first study protocol consisted of testing adrenocortieal steroid sensitivity to exogenous ACTH on three different sequential metabolic
conditions: (I) normal sodium diet control period (NSD-CONT); (2) NSD-DEX suppression, and (3) LSD-DEX suppression. As a separate study, after control values were established, a HSD was initiated together with DEX suppression (HSD-DEX). Each study period lasted 5 to 8 days. In study 1, six subjects underwent the NSDCONT protocol. They ingested a fixed metabolic diet containing approx. 22mmol of sodium and 70-90 mmol of potassium per day, to which sodium chloride (NaCI) tablets were added to establish a sodium intake of approx. 124mmol/d. After 5-7 days of NSD-CONT, they were given DEX (0.5 mg p.o. every 6 h), which was continued throughout the remainder of the study (NSD-DEX period). Sodium tablets were discontinued after 7 days to establish a LSD (22mmol sodium/d) (LSD-DEX period). As a separate study, 6 additional normal subjects ingested similar diets, NSD-CONT 2, followed by HSD either be receiving additional NaCI tablets (270mmol sodium/d, n = 3), or NaCi tablets (165mmol sodium/d) plus fludrocortisone (Florinef, 0.2 mg p.o. every 12h, n =3) plus DEX (0.5 p.o. every 6h) HSD-DEX. To determine adrenal responsiveness to a maximum ACTH stimulus during endogenous ACTH suppression (NSD-DEX period) and during RAS activation (LSD-DEX period) or suppression (HSD-DEX period), we performed acute stimulation tests every other morning throughout the study. Cosyntropin (Cortrosyn, 250 #g i.v. bolus) was administered to fasting and overnight recumbent subjects and blood samples were collected before and 60 min after the injection. Sodium balance and ACTH suppression were monitored daily in all subjects by: (1) changes in body weight (BW), (2) 24-h urinary excretion of sodium and potassium, (3) 24-h urinary excretion rates of free cortisol (UFF) and free aldosterone (UFA), and (4) plasma renin activity (PRA), and sodium and potassium concentrations were determined every other day. Plasma and urinary electrolytes were measured by routine internal standard flame photometry. PRA was measured by the RIA of generated angiotensin I [22]. After extraction and HPLC isolation, UFF was measured by a competitive protein-binding assay and UFA was measured by specific RIA[2]. Plasma F, S, 21dF, B, DOC, 18-OHDOC, 18-OHB, and aldosterone were measured by RIA in all samples following
619
Low sodium intake enhances sensitivityof normal subjects extraction, solvent partitioning, and HPLC separation. Methodology and reference values for this laboratory have been published [2, 22]. All samples from the same subject were assayed as a single batch after completion of the study. For statistical purposes all non-detectable values were arbitrarily considered equal to the limit of sensitivity for each particular assay, divided by the square root of 2 (1.412)[2]. Non-parametric analysis of variance (the Kruskall-Wallis rank sum test) was used to compare data. When significant, ANOVA was complemented by the Scheffe's and Dunn's tests of contrasts to ascertain differences between study periods [23]. RESULTS
Sodium and potassium balance
Although a different group of normal subjects was used for the high sodium protocol, all steroid values were virtually identical. In Fig. 2, the steroid responses to ACTH study 2 were compared to their own NSD-CONT study 2. Cortisol, S and 21 dF (the Z F 17-hydroxy pathway) (Table 1). All three steroids showed virtually complete suppression of basal levels when dexamethasone was given: NSD-, LSD-, and HSD-DEX. Plasma F demonstrated relatively fixed responses to ACTH stimulation. The control response of 632 + 22 nmol/l decreased significantly to 367 + 36 (NSD + DEX) nmol/1. Surprisingly, the response at the end of the HSD-DEX period was even further and significantly reduced to 254 _+ 28 nmol/l (see Table 1). Both basal S and 21 dF were significantly suppressed while on DEX and their responses to ACTH did not significantly decrease during
Daily urinary sodium, UFA and U F F during the study periods are depicted in Fig. 1. The NSD-DEX period (average sodium intake of 124mmol/d) was associated with moderate sodium retention and an average BW gain of 1% compared with the NSD-CONT period. [= Sodium balance was achieved on the 4th day of LSD-DEX resulting in a BW reduction of approx. 1.5-2%, and on the 6th day of the HSD-DEX study resulting in a BW gain of 2-3% (Fig. 1). Plasma sodium and potassium concentrations (Table 1) showed no significant 275o change during study periods except for a significant decrease in plasma potassium during HSD-CONT 2. i lSS0
UFA, UFF excretion and PRA UFA showed minimal change during NSDCONT and -DEX and incremental increases as the LSD-DEX period progressed. On HSDDEX, UFA significantly decreased by half from its NSD-CONT period (Fig. 1). PRA levels (Table 1) showed little change during the NSDCONT and -DEX, but increased significantly on LSD-DEX and were reduced by HSD-DEX. U F F excretion was reduced during all DEX treatment period. Increases in U F F and UFA were seen on the days the 1 h ACTH test was administered (Fig. 1).
Plasma steroid responses to ACTH on different sodium intakes Table 1 and Fig. 2 show the basal and 60 rain responses of plasma steroids to the ACTH test performed at the end of each study period.
UFA
I 1110
/
0
275[" UFF
UFF urinary~ e corUsol
UFAurinaryfree|idosmroneI I
T
16S 110
~
"fdl! 0
i
.
i .
IZX
.
.
,:
SUPPRESSION
i I
Fig. 1. Twenty-four hour urinary sodium (retool/d) UFA
and UFF excretion during each treatment period in 6 normal subjectson NSD and LSD, and 6 on HSD. *Day I h ACTH t~t performed.
620
CLAUDIO E . KATER
+1 + l +1 + t +1 +t ~
the NSD-DEX period (3.05+ 1.0 and 1.18 + 0.3nmol/1). By the end of the LSD-DEX period, S and 21 dF were higher (4.7 + 1.5 and 1.9 +0.4nmol/l) than those of NSD-CONT (4.17 + i.5 and 1.4 + 0.3 nmol/l), respectively. While no significant change in S or 21 dF occurred comparing the NSD-CONT to LSDDEX, F peaks were markedly reduced. Although DEX was continued for a second week during the LSD period, basal levels of ZF steroids showed no further decrease. Like F, the HSD-DEX period showed significant impairment of S and 21dF responses to ACTH (1.3 + 0.2 and 0.94 + 0.12 nmol/l, respectively).
+1 +1 +1 +1
~~-~~ ,-e.
+1+1+1+1+1+1+1+~+~+=+1
d z
~ ' ~ 2
. . . .
-=
÷1 +1 ÷1 +1 +1 +1 +1 +1 +l +1 +l
