L-Serine lowers while glycine increases blood pressure in chronic L-NAME-treated and spontaneously hypertensive rats

C

Original article 2339

L-Serine lowers while glycine
increases blood pressure inchronic L-NAME-treated and spontaneously hypertensive ratsRamesh C. Mishraa, Saswati Tripathya, Dale Questc, Kaushik M. Desaia,Jawed Akhtarb, Indravadan D. Dattanib and Venkat Gopalakrishnana
Objective To determine the acute hemodynamic effects of

the nonessential amino acid, glycine, and its precursor,

L-serine, in normotensive and hypertensive rats.

Methods Changes in mean arterial pressure and heart rate

evoked by comparable intravenously administered doses

(0.3–3.0 mmol/kg) of L-serine, D-serine and glycine were

examined in anaesthetized normotensive 14-week-old male

Sprague–Dawley, Wistar–Kyoto (WKY) rats, spontaneously

hypertensive rats and WKY rats subjected to chronic nitric

oxide synthase inhibition by treatment with NG nitro

L-arginine methyl ester (0.7 mg/ml in drinking water for

5 days).

Results L-Serine evoked a greater maximal fall in mean

arterial pressure [L-serine vs. D-serine in Sprague–Dawley

rats, mean W standard error of the mean values (mmHg):

30 W 3 vs. 20 W 5, P < 0.05; in control WKY rats: 46 W 3 vs.

30 W 4, P < 0.05; in NG nitro L-arginine methyl ester-treated

WKY rats: 93 W 6 vs. 41 W 5, P < 0.01; in spontaneously

hypertensive rats: 81 W 7 vs. 39 W 5 P < 0.01]. The effects of

L-serine were significantly reduced in rats pretreated with a

combination of apamin and charybdotoxin, inhibitors of the

small conductance and intermediate conductance calcium-

activated potassium (KCa) channels. Glycine elicited a dose-

dependent fall in mean arterial pressure in normotensive

WKY rats (25 W 4; P < 0.01) and evoked pressor responses in

both spontaneously hypertensive rats (29 W 3; P < 0.01) and

NG nitro L-arginine methyl ester-pretreated hypertensive

WKY (39 W 5; P < 0.01) rats. Both the depressor and pressor

opyright © Lippincott Williams & Wilkins. Unauth

0263-6352 � 2008 Wolters Kluwer Health | Lippincott Williams & Wilkins

responses to glycine were abolished by pretreatment with

the N-methyl D-aspartate receptor antagonist, MK-801.

Conclusion The profound stereo-selective

antihypertensive effect of L-serine is neither mediated

nor mimicked by glycine. It does not require N-methyl

D-aspartate receptor activation by glycine but likely involves

activation of endothelial KCa channels. L-Serine is a potential

antihypertensive agent. J Hypertens 26:2339–2348 Q 2008

Wolters Kluwer Health | Lippincott Williams & Wilkins.

Journal of Hypertension 2008, 26:2339–2348

Keywords: amino acids, blood pressure, glycine, hypertension, L-serine,mean arterial pressure, MK-801, nitric oxide synthase inhibitor, N-methylD-aspartate receptor, spontaneously hypertensive rats

Abbreviations: ARA-S, N-archidonoyl-L-serine; ChTx, charybdotoxin COXcyclooxygenase; IKCa, intermediate conductance calcium activatedpotassium channels; L-NAME, NG nitro L-arginine methyl ester; MAP, meanarterial pressure; NMDA, N-methyl D-aspartate; NOS, nitric oxide synthase;SHR, Spontaneously Hypertensive Rat; SKCa, small conductance calciumactivated potassium channels; WKY, Wistar-Kyoto

aDepartment of Pharmacology and bDivision of Cardiology, Department ofMedicine, Royal University Hospital, College of Medicine, University ofSaskatchewan, Saskatoon, Saskatchewan, Canada and cTexas Tech University,Lubbock, Texas, USA

Correspondence to Venkat Gopalakrishnan, MSc, PhD, Professor, Head,Department of Pharmacology, College of Medicine, University of Saskatchewan,Saskatoon, Saskatchewan, S7N 5E5, CanadaTel: +1 306 966 6293; fax: +1 306 966 6220; e-mail: [email protected]

Received 28 March 2008 Revised 5 July 2008Accepted 30 July 2008

IntroductionRecently, we demonstrated that the nonessential amino

acid, L-serine, promotes concentration-dependent endo-

thelium-mediated vasodilatation in phenylephrine con-

stricted third-order branches of rat mesenteric arterioles

in vitro and evokes a reversible and dose-dependent fall

in mean arterial pressure (MAP) following acute intrave-

nous infusion in vivo. Both the in-vitro and in-vivo

responses to L-serine were higher in Sprague–Dawley

rats rendered hypertensive by pretreatment with the

nitric oxide synthase (NOS) inhibitor, NG nitro L-arginine

methyl ester (L-NAME) [1,2]. In vivo, L-serine is con-

verted to D-serine and glycine [3,4]. Glycine is known to

promote N-methyl D-aspartate (NMDA) receptor acti-

vation resulting in increased Ca2þ influx [3]. Numerous

studies support the notion that glycine exerts anti-

hypertensive, vasodilator, antiapoptotic, anti-inflamma-

tory and antioxidant effects and protects against

hypoxia-induced and cyclosporine-induced renal injury

[5–18]. In contrast, the hemodynamic effects of L-serine

have not been addressed. Some of the cardiovascular

effects of glycine have been attributed to stimulation of

renal/endothelial NMDA receptors or elevation of nitric

oxide levels [12,15,18]. Long-term oral administration of

L-threonine, a precursor of L-serine and glycine, has been

shown to cause hypertension [19]. Thus, it is possible that

long-term hemodynamic effects of L-serine may differ

from its effects of acute administration because during

chronic treatment, L-serine may be bioconverted to glycine

and/or D-serine, which in turn could modify the net hemo-

dynamic effects. To address these issues, first of all, it is

important to determine the acute in-vivo effects of

orized reproduction of this article is prohibited.

DOI:10.1097/HJH.0b013e328312c8a3

mailto:[email protected]

http://dx.doi.org/10.1097/HJH.0b013e328312c8a3

C

2340 Journal of Hypertension 2008, Vol 26 No 12

comparable doses of L-serine, D-serine, glycine and

L-threonine in normotensive and hypertensive rats. We

observed that the acute administration of L-serine or

glycine reduces MAP by different mechanisms in normo-

tensive rats and that the MAP of hypertensive rats is

reduced by L-serine and increased by glycine.

MethodsAnimals and methodsThe present study approved by the University of Saskatch-

ewan review committee conformed to the guidelines

stipulated by the Canadian Council on Animal Care.

Fourteen-week-old male Sprague–Dawley rats (350–

380 g), Wistar–Kyoto (WKY) rats (320–350 g) and spon-

taneously hypertensive rats (SHRs, 250–270 g) were

obtained from Charles River (St. Constant, Quebec,

Canada). One group of WKY rats received the NOS

inhibitor, L-NAME (0.7 mg/ml in drinking water ad libi-

tum) for 5 days [1,20,21]. All other rats received plain

water. Although it is generally accepted that WKY is

the most appropriate and the closest control group for

comparison of data with the SHR strain, several studies

encourage inclusion of another normotensive strain,

namely Sprague–Dawley rats besides WKY rats. Thus,

in the present study, SHRs and chronic L-NAME-treated

WKY rats served as hypertensive rat models, whereas

untreated WKY and Sprague–Dawley rats served as

normotensive control groups for comparison. MAP and

heart rate (HR) were determined in rats anaesthetized with

thiopental sodium [100 mg/kg, intraperitoneally (i.p.)] as

described earlier [1,20,21]. The femoral artery was cannu-

lated and connected to a pressure transducer to record

changes in MAP and HR using a PowerLab data acqui-

sition system (AD Instruments Pvt. Ltd., Sydney, Austra-

lia). The femoral vein was cannulated to administer the

amino acids (prepared in physiological saline, pH adjusted

to 7.3) given as bolus injections (0.4 ml/kg) for each infu-

sion. After ensuring stable baseline MAP and HR, the

responses to comparable infusions (between 0.3 and

3.0 mmol/kg) of L-serine, D-serine, glycine and L-threonine

were determined. Enough time was allowed between

responses for the MAP to return to baseline. The maxi-

mum dose of L-serine was limited to 2.0 mmol/kg in the

chronic L-NAME-treated rats after finding that recovery of

baseline was difficult following the profound fall in MAP

attained with the 3.0 mmol/kg dose. The responses to

either glycine or L-serine were also determined before

and 30 min after slow infusions given over a period of

15 min of either a combination of apamin, a small con-

ductance Ca2þ-dependent Kþ channel (SKCa) inhibitor,

with charybdotoxin (ChTx), an intermediate conductance

Ca2þ-dependent Kþ channel (IKCa) inhibitor [75 mg/kg

intravenously (i.v.) of each], or the NMDA antagonist,

MK-801 (75 mg/kg. i.p.) using an infusion pump.

Studies were also performed in a select group of normo-

tensive WKY rats in which the effect of increasing doses

opyright © Lippincott Williams & Wilkins. Unautho

(0.3–3.0 mmol/kg, i.v.) of glycine infusion was examined

both prior to and 1 h after acute infusion of L-NAME

(100 mg/kg, i.v. given over a period of 10 min) to compare

the responses to glycine infusion in chronic L-NAME-

treated hypertensive WKY group vs. acute L-NAME-

treated hypertensive rats. The dose of L-NAME was

premised on previous work [22].

MaterialsL-Serine, D-serine, glycine, L-threonine, L-NAME and

MK-801 were obtained from Sigma-Aldrich Canada Ltd.

(Oakville, Ontario, Canada), apamin and ChTx were

purchased from EMD Biosciences, Inc. (La Jolla,

California, USA) and thiopental sodium from Abbott

Laboratories Ltd (St. Laurent, Quebec, Canada).

Statistical analysisThe change in MAP following infusion of each dose was

plotted to generate dose–response curves to respective

amino acids. The data are expressed as mean�SEM

(n¼ 5–7 rats per group). Differences between the groups

were tested for significance by one-way analysis of

variance, followed by Tukey’s post-hoc test. The differ-

ences were considered significant when the P value was

less than 0.05.

ResultsBaseline MAP and HR were similar in normotensive

Sprague–Dawley and WKY strains. As expected, MAP

was significantly higher (P< 0.01) in L-NAME-

pretreated Sprague–Dawley and WKY rats and untreated

SHRs (Table 1). The data from a single experiment

performed in parallel in a normotensive WKY rat

(Fig. 1a, upper panel), a chronic L-NAME-pretreated

hypertensive WKY rat (Fig. 1b, lower panel) and a

SHR strain (Fig. 2a, upper panel) revealed that acute

infusion of comparable doses of either D-serine or L-

serine evoked reversible, dose-dependent falls in MAP

in all these strains. However, compared with D-serine, the

responses to L-serine were consistently more pronounced

at each incremental dose in these rats (Fig. 1a, b and

Fig. 2a). The return of MAP to baseline was protracted in

the chronic L-NAME-treated WKY group following a

profound fall in MAP with the highest (3.0 mmol/kg)

dose in initial experiments, so the maximum dose of

L-serine infusion was limited to 2.0 mmol/kg in this group

(Fig. 1b and Fig. 3c). The baseline MAP was much higher

in the SHR strain compared with L-NAME-pretreated

hypertensive WKY rats (Figs 1 and 2 and Table 1).

Despite the higher baseline MAP in the SHR group,

the fall in MAP evoked by D-serine and L-serine was

relatively lower in SHRs than in L-NAME-treated WKY

rats (compare data in Fig. 1b vs. Fig. 2a). The pooled

values from several D-serine and L-serine dose–response

curves revealed that L-serine was more potent and effi-

cacious than D-serine in lowering MAP. L-Serine showed

the following rank order of efficacy in lowering MAP:

rized reproduction of this article is prohibited.

C

Opposite effects of serine and glycine on BP Mishra et al. 2341

Table 1 A comparison of mean arterial pressure (mmHg) and heart rate (beats per minute) recorded before and 30 s after infusion of either L-serine or glycine (2–3 mmol/kg) in 14-week-old male Sprague–Dawley, L-NAME-pretreated Sprague–Dawley, Wistar–Kyoto, L-NAME-pretreated Wistar–Kyoto and spontaneously hypertensive rat strains

Baseline L-Serine Glycine

Strain MAP HR MAP HR MAP HR

SD 111�4 364�11 83�5y 386�10 91�3 381�12SD (L-NAME treated) 142�3M 343�8 47�4yy 391�9 180�5y 321�7WKY (control) 114�3 381�10 69�4y 410�8 89�2y 399�9WKY(L-NAME treated) 149�4MM 377�12 52�3yy 427�11 183�4y 354�13SHR 169�4MM 401�8 88�4yy 443�10 198�5y 383�8

Each data point is mean�SEM (n¼5–7 rats/group). HR, heart rate; L-NAME, NG nitro L-arginine methyl ester; MAP, mean arterial pressure; SHR, spontaneouslyhypertensive rat; SD, Sprague–Dawley; WKY, Wistar–Kyoto. MP<0.05 vs. SD (control) group; MMP<0.01 vs. WKY (control) group; yP<0.05; yyP<0.01 vs. respectivebaseline value in the same group prior to infusion of either L-serine or glycine.

chronic L-NAME-pretreated WKY>SHR>untreated

WKY>Sprague–Dawley rats (Fig. 3). The L-serine-

evoked fall in MAP was not accompanied by significant

increases in HR in any of these groups (Table 1). The


Fig. 1

This is a typical experiment that compares the fall in mean arterial pressureL-serine. (a) Infusion in a 14-week-old male normotensive WKY (control) ra(0.3–2.0 mmol/kg) in a hypertensive WKY (L-NAME) rat subjected to chronwater) for 5 days. Due to profound and protracted hypotension at higher d2.0 mmol/kg when it was infused in L-NAME-treated rats. Similar responseand L-NAME-treated WKY rats. The response patterns to both isomers weSprague–Dawley rats (n¼5–7). L-NAME, NG nitro L-arginine methyl ester;Wistar–Kyoto.

exact EC50 value for L-serine and D-serine in reducing

MAP could not be determined because a supramaximal

effect could not be established due to the insolubility of

these amino acids at concentrations of more than


to acute comparable doses (0.3–3.0 mmol/kg) of D-serine followed byt. (b) The changes in MAP to D-serine followed by L-serine infusionic NOS inhibition by pretreatment with L-NAME (0.7 mg/ml in drinkingoses, the maximal dose of D-serine or L-serine used was limited topatterns were seen for D-serine and L-serine in six WKY (control)re also comparable to that of WKY control group in 14-week-old maleMAP, mean arterial pressure; NOS, nitric oxide synthase; WKY,

C


Fig. 2

80

120

160

200D-serine (mmol/kg)

--0.3

--0.5 1.0

-- ----

2.0 3.0

80

120

160

200

D-serine (mmol/kg)

--0.3

-- -- -- --0.5 1.0 2.0 3.0M

AP

(m

mH

g)

(a)

(b)

SHR

SHR

5 min

5 min

L-serine (mmol/kg)

-- -- -- -- --0.3 2.01.00.5 3.0

---- -- ----0.50.3 2.01.0 3.0

L-serine (mmol/kg)

This is a typical experiment that compares the fall in mean arterial pressure to acute intravenous infusion of increasing doses (0.3–3.0 mmol/kg)of D-serine followed by L-serine in a 14-week-old male spontaneously hypertensive rat in the upper panel (a). (b) Responses to similar doses ofD-serine followed by L-serine infused in the same animal after pretreatment with apamin and ChTx combination (75 mg/kg of each given i.v.by slow infusion over a 15-min period). Similar results were obtained in six age-matched male SHRs. The MAP remained at a steady level between160 and 170 mmHg throughout the study period. ChTx, charybdotoxin; i.v., intravenously; MAP, mean arterial pressure; SHR, spontaneouslyhypertensive rat.

3.0 mmol/kg. However, the estimated EC50 values,

assuming that the maximum effect is reached at

3 mmol/kg in all of the groups, were in the ranges of

0.8–1.1 mmol/kg and 1.4–2.3 mmol/kg for L-serine and

D-serine, respectively. Pretreatment with apamin and

ChTx inhibited the fall in MAP evoked by D-isomers

and L-isomers of serine in both WKY and SHR strains,

and the degree of inhibition was higher for L-serine in the

SHR strain as shown in a single experiment (Fig. 2b,

lower panel and Fig. 4a, b).

Like L-serine, comparable doses of glycine evoked a dose-

dependent fall in MAP (maximal fall 25� 4 mmHg at

3.0 mmol/kg dose) without significant changes in HR in

normotensive WKY rats (Fig. 5, left panel). Pretreatment

with the selective NMDA antagonist, MK-801, did not

affect the basal MAP; it prevented the hypotensive

response to glycine but not to L-serine (Fig. 5, right panel).

However, infusion of increasing doses of glycine evoked a

consistent dose-dependent elevation in MAP in both

SHRs and chronic L-NAME-treated hypertensive rats


(Fig. 6a, b, left panels). The glycine-induced maximal

pressor response at 3.0 mmol/kg in chronic L-NAME-

treated WKY rats was 39� 5 mmHg (P< 0.01) and

29� 3 mmHg (P< 0.01) in the SHR strain. Thus, it was

higher even though the SHR group had higher basal MAP

prior to glycine infusion. The hypertensive effect of

glycine was also blocked subsequent to MK-801 treatment

in the same rats (Fig. 6a, b, right panels). Pooled mean

�SEM data from several dose–response curves for the

depressor response in the normotensive WKY rats and the

pressor response in the hypertensive groups and the block-

ade of the glycine responses after MK-801 pretreatment

are shown (Fig. 7). The dose-dependent hypotensive

effect of glycine observed in the normotensive WKY strain

is shown (Fig. 8, left panel). Following acute infusion of

L-NAME (100 mg/kg i.v.) in those rats, the MAP increased

and stabilized in 1 h. Increasing doses of glycine failed to

elevate or decrease MAP at this stage, while infusion of

increasing doses of L-serine elicited a dose-dependent

hypotension (Fig. 8, right panel). Similar results were

reproduced in a minimum of five WKY rats.


C


Fig. 3

Dose–response curves comparing the fall in mean arterial pressure evoked by D-serine (*) and L-serine (*). (a) Fourteen-week-old malenormotensive Sprague–Dawley rats or (b) WKYvs. (c) hypertensive WKY rats that received L-NAME (0.7 mg/ml in drinking water for 5 days)treatment or (d) SHR strains. Each data point represents mean�SEM (n¼5–7 rats/group). Due to profound and protracted hypotension at higherdoses, the maximal dose of D-serine or L-serine used was limited to 2 mmol/kg when it was infused in L-NAME-treated rats. L-NAME, NG nitroL-arginine methyl ester; MAP, mean arterial pressure; SHR, spontaneously hypertensive rat; WKY, Wistar–Kyoto. �P<0.05 and ��P<0.01 vs. datapoint for D-serine dose in the same group. yP<0.05 and yyP<0.01 vs. data for basal MAP prior to infusion of D-serine or L-serine in the same group.

Infusion of L-threonine in the same dose range had no

effect on MAP or HR in any group (Fig. 9). Responses

to amino acids were very similar between WKY and

Sprague–Dawley rats, so only data for WKY strain are


Fig. 4

0

20

40

60

80

L-SD-S

*

D-SL-S

Apa+ChTxControl

†

(a)

Fall

in M

AP

WKY

††

Graphical representation of pretreatment with apamin and charybdotoxin inspontaneously hypertensive rat strains. Each bar represents mean�SEM m�P<0.05, ��P<0.01 vs. data for D-serine in the same strain. ChTx, charybdrat; WKY, Wistar–Kyoto. yP<0.05 and yyP<0.01 vs. respective data for D-sand charybdotoxin (75 mg/kg, i.v.) in the same group.

shown. The responses to all amino acids could be repro-

duced a second time, and changing the order in which the

amino acids were infused did not influence the responses

they evoked.


L-SD-S D-SL-S

Apa+ChTxControl

0

20

40

60

80 **

(b)

SHR

† ††

hibiting the fall in mean arterial pressure in both Wistar–Kyoto andaximum fall in MAP determined in five to seven rats for each group.otoxin; MAP, mean arterial pressure; SHR, spontaneously hypertensiveerine or L-serine values prior to slow infusion of a combination of apamin

C


Fig. 5

Glycine

0.5− −

1.0

− −−− − −−2.0 3.0

1.00.50.3 3.02.0

L-serine (mmol/kg)

80

100

120

140

Glycine (mmol/kg)

−−−− −0.3 3.02.01.00.5

MA

P (

mm

Hg

)

MK-801 (75 mg/kg) i.p.

WKY (control)

5 min 5 min

A typical experiment shows the fall in mean arterial pressure evoked by increasing doses (0.3–3.0 mmol/kg, intravenously) of glycine before and30 min after the infusion of N-methyl D-aspartate-selective antagonist, MK-801 (75 mg/kg, i.p.), given over a period of 15 min. MK-801 blocked theresponses to glycine but not to L-serine. The time scale on the X axis is reduced for the right hand side panel MK-801 treatment as the data werecompressed to accommodate data for glycine followed by responses to L-serine (a total of nine responses). Similar results were reproduced in sevenage-matched male normotensive Sprague–Dawley and WKY rats. i.p., intraperitoneally; MAP, mean arterial pressure; NMDA, N-methyl D-aspartate;WKY, Wistar–Kyoto.

DiscussionComparison of the antihypertensive effects of L-serineand D-serineThe key findings of this study are summarized as follows:

(1) T

opy

he D-isoform of serine is less potent and less

efficacious than the naturally occurring L-isoform in

evoking a reversible and dose-dependent fall in MAP

but does not change HR.

(2) T
he fall in MAP evoked by L-serine was more
profound in two well established models of hyper-

tension (L-NAME-induced and SHR models) than in

normotensive rats.

(3) T
he L-serine-induced fall in MAP was blocked by a
combination of apamin and ChTx in both normo-

tensive and hypertensive rats. These data are

consistent with our recent report using Sprague–

Dawley rats rendered hypertensive by chronic

treatment with L-NAME [1]. The fall in MAP

evoked by L-serine involves activation of apamin and

ChTx-sensitive SKCa and IKCa channels that are

predominantly expressed in the vascular endo-

thelium [1].

(4) I
nfusion of similar doses of glycine, a metabolite of
L-serine, such as L-serine, evoked a dose-dependent

fall in MAP in normotensive WKY rats, but unlike

D-serine and L-serine, it elevated MAP in both models

of hypertension. Both depressor and pressor responses

to glycine were prevented in rats pretreated with the

NMDA-selective antagonist, MK-801.

(5) W
hen MAP was elevated following acute i.v. infusion
of L-NAME in a normotensive WKY rat, glycine led

to elevation of MAP within 1 h. Unlike chronic

L-NAME-treated rats, infusion of glycine failed to

right © Lippincott Williams & Wilkins. Unauthoriz

elicit a pressor response in normotensive WKY rats. It

appears that the pressor effect of glycine manifests

only in hypertensive rats, such as SHRs and chronic

L-NAME-treated rats.

(6) A
cute infusion of L-threonine, a precursor of L-serine,
did not affect MAP or HR responses in any group of

rats in the present study.

(7) T
hese data confirm the stereo-selective nature of
L-serine in reducing blood pressure (BP), showing

that the L-isomer is more potent and efficacious than

the D-isomer. The depressor effect was not mimicked

or shared by closely related amino acids such as its

precursor, L-threonine, nor its metabolite, glycine,

nor is it mediated by vascular/renal NMDA receptors,

but it does involve activation of apamin and ChTx-

sensitive KCa channels.

(8) F
inally and most importantly, L-serine and glycine
have opposite effects in hypertensive rats. Thus,

these amino acids alter BP by different mechanisms,

such that the responses to glycine, but not L-serine,

are prevented by MK-801.

N-Archidonoyl-L-serine (ARA-S), a brain-derived lipid

mediator, was recently shown to promote endothelium-

dependent but cyclooxygenase (COX)-independent

and nitric oxide-independent vasodilatation of rat aorta

and mesenteric arteries through stimulation of a novel

cannabinoid receptor present on endothelial cells [23].

That study did not address whether ARA-S evoked a fall

in MAP or whether L-serine alone causes the vasodilata-

tion. In the present study, the depressor response to

L-serine was larger in L-NAME-treated rats than in

SHRs. We have previously reported that L-serine evokes

ed reproduction of this article is prohibited.

C


Fig. 6

(a)

80

100

120

140

160

180

−−−−−

3.02.01.00.50.3

Glycine (mmol/kg)

−− − − −−1.00.50.3 1.03.02.0

Glycine (mmol/kg) L-serineM

AP

(m

mH

g)

MK-801 (75 mg/kg) i.p.

WKY (L-NAME)

5 min

5 min

100

140

180

220Glycine (mmol/kg)

−−−

− −

1.00.50.3

2.03.0

(b)

SHR

− − − − −1.00.50.3 3.02.0

Glycine (mmol/kg)

Similar responses were reproduced in five to seven age-matched animals from each group. A typical experiment compares the increase in meanarterial pressure with increasing doses of glycine (0.3–3.0 mmol/kg, intravenously) infusion. (a) NG nitro L-arginine methyl ester-treated WKY ratbefore (left upper panel) and after (right upper panel) the administration of N-methyl D-aspartate antagonist, MK-801. The pressor responses toglycine were absent after MK-801 treatment, but the response to L-serine persisted. (b) Pressor responses to increasing doses of glycine infusionand the lack of pressor responses to the same doses of glycine infusion after the administration of MK-801. i.p., intraperitoneally; L-NAME, NG nitroL-arginine methyl ester; NMDA, N-methyl D-aspartate.

COX-independent and nitric oxide-independent, but

endothelium-dependent, vasodilatation in rat mesenteric

arterioles through activation of SKCa and IKCa present on

the endothelium [1]. The significance of this finding was

highlighted [2].

Opposite effects of glycine in the regulation of meanarterial pressureBesides a dietary source, both glycine and D-serine can be

synthesized in vivo from L-serine [3,4,24]. Glycine is a

classical agonist that promotes NMDA receptor-mediated

Ca2þ influx in the brain, blood vessels, heart and kidney

[3,12,25]. Glycine-evoked renal vasodilatation and in-

creased glomerular filtration rate were abolished by the

inclusion of either a NMDA antagonist or a NOS inhibitor

[12,15,18,26]. The addition of 1% glycine to the diet for a

period of 2–4 weeks normalized BP in sucrose-fed hyper-

tensive rats [16]. Central administration of glycine caused a

fall in MAP within 3 min [27]. Glycine is reported to

hyperpolarize endothelial cells through activation of


chloride channel and to reduce oxidative stress and inflam-

mation [4,8,9]. These data led to speculation that glycine is

an antihypertensive agent [17]. However, there are no

reports on the effects of acute infusion of glycine in

normotensive or genetic SHRs or in nitric oxide-comprom-

ised (L-NAME-induced) hypertensive states. In the pre-

sent study, acute infusion of glycine led to a rapid, revers-

ible and dose-dependent fall in MAP in normotensive

WKY rats that is blocked by the NMDA antagonist,

MK-801. The same doses of glycine had the opposite

effect such that they elevated MAP in both hypertensive

groups. This pressor response to glycine was also sensitive

to blockade by MK-801. The hypertensive effect of

glycine was higher in long-term or chronic L-NAME-

pretreated WKY rats than in SHRs. These data suggest

that in the normotensive state, glycine predominantly

activates the endothelial/renal NMDA receptors that

promote Ca2þ-dependent NOS activation that in turn

leads to nitric oxide-mediated vasodilatation/hypotension.

However, when nitric oxide-mediated vasodilatation and


C


Fig. 7

Line graphs provide the dose–response relationships for glycine-evoked changes in mean arterial pressure either before (*) or after MK-801 (*) infusion. (a) WKY, (b) L-NAME-pretreated hypertensive WKYor (c) SHR strains. Each data point is mean�SEM (n¼5-7 rats/group).L-NAME, NG nitro L-arginine methyl ester; MAP, mean arterial pressure;SHR, spontaneously hypertensive rat; WKY, Wistar–Kyoto. �P<0.05and ��P<0.01 vs. data of basal MAP prior to infusion of glycine in eachgroup.

hypotension are attenuated in chronic L-NAME-treated

rats, glycine stimulates NMDA receptors on the vascular

smooth muscle cells to evoke vasoconstriction that mani-

fests as a pressor response in the whole animal. Thus,

elevation in MAP evoked by glycine was more pronounced


in nitric oxide-compromised hypertensive rats, the same

model in which L-serine evoked a profound depressor

effect. Despite endothelial dysfunction, nitric oxide-

mediated responses are not fully inhibited in the SHR

strain, so the pressor effect of glycine was accordingly less.

Our in-vivo observation is consistent with a report that

in vitro, addition of glycine evoked dose-dependent vaso-

constriction of rat pial arterioles, an effect that was blocked

by an NMDA antagonist [28]. D-Serine, a weaker NMDA

agonist, did not evoke an increase in MAP but instead

produced a dose-dependent fall in MAP, albeit with much

lower efficacy relative to L-serine, thus, its effect on BP is

not mediated by NMDA receptor activation. On the basis

of the data gathered in the present study, it seems reason-

able to conclude that the acute hypotensive response to

L-serine is not mimicked or mediated by glycine and is not

linked to activation of vascular NMDA receptors.

L-Threonine infusion does not affect mean arterialpressureL-Serine, D-serine and glycine can be derived in vivo from

either glucose metabolism or dietary intake of L-threo-

nine. Administration of 8% L-threonine, but not glycine,

in the diet for a period of 6 weeks increased acetaldehyde

concentrations and elevated systolic BP by 50 mmHg

[19]. The present data confirmed that acute adminis-

tration of L-threonine failed to alter MAP or HR in all

rat models studied.

Comparison of antihypertensive effect of L-serinein NG nitro L-arginine methyl ester treated WKY andspontaneously hypertensive rat modelsThe normal plasma concentration of L-serine is 130 mmol/l

[3,4]. Our data show hypotensive responses at doses

between 0.3 and 3.0 mmol/kg. Thus, the pharmacological

concentrations of L-serine and its related amino acids

employed in this study are much higher than the physio-

logical plasma concentration of L-serine. Although the

hypotensive response to the highest dose of L-serine tested

was accompanied by a marginal increase in HR, it was not

statistically significant (Table 1). This led us to speculate

that at higher concentrations, L-serine might depress the

reflex increase in HR expected in response to the profound

fall in MAP encountered in the hypertensive models.

However, serial infusions of L-serine ex vivo in a perfused

Langendorff heart preparation isolated from a chronic

L-NAME-pretreated rat failed to show substantive

changes in coronary perfusion pressure, HR or left

ventricular pressure (unpublished observation). The

maximum MAP-lowering effect of L-serine was more

pronounced in the chronic L-NAME model (93 mmHg

at 2 mmol/kg dose) than in the SHR group (81 mmHg at

3 mmol/kg dose) despite SHRs having higher baseline

MAP (prior to L-serine infusion) than L-NAME-induced

hypertensive rats. It can thus be concluded that the

maximal MAP-lowering effect of L-serine is not a function

of the initial baseline MAP.


Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.


Fig. 9

80

100

120

140L-threonine (mmol/kg)

− − − − −0.3 0.5 2.01.0 3.0

100

140

160

120

−− −−−

L-threonine (mmol/kg)

0.3 1.00.5 3.02.0

(b)(a)

100

120

140

160

180L-threonine (mmol/kg)

2.01.00.50.3 3.0−−−−−

(c)

MA

P (

mm

Hg

)

WKY (control) WKY (L-NAME)

SHR

5 min

5 min 5 min

A representative experiment depicts that infusion of L-threonine (0.3–3.0 mmol/kg) failed to alter mean arterial pressure recording. (a) NormotensiveWistar–Kyoto rat; (b) chronic L-NAME-treated hypertensive WKY rat; and (c) a SHR strain. These results were reproduced in five to seven differentanimals from each group. L-NAME, NG nitro L-arginine methyl ester; MAP, mean arterial pressure; SHR, spontaneously hypertensive rat; WKY,Wistar–Kyoto.

Fig. 8

A typical experiment compares the fall in mean arterial pressure to increasing doses of glycine (0.3–3.0 mmol/kg, intravenously) before (left handpanel) and the lack of responses to the same doses of glycine 1 h after NG nitro L-arginine methyl ester (100 mg/kg, intravenously) infusion in anormotensive Wistar–Kyoto rat. While glycine responses were blunted, L-serine responses persisted following acute L-NAME infusion. The X axistime scale was compressed for the right hand side panel to accommodate eight responses. Similar data were reproduced in five WKY rats. L-NAME,NG nitro L-arginine methyl ester; MAP, mean arterial pressure; WKY, Wistar–Kyoto.

C


In conclusion, our findings suggest that pharmacological

doses of L-serine have a substantial BP-lowering effect,

particularly in hypertensive states. The therapeutic

benefit of oral administration of L-serine in doses more

than 400 mg/kg/day in patients with depression, schizo-

phrenia and chronic fatigue syndrome has been pur-

ported [3,4,29]. Oral formulations of more polar amino

acids such as L-serine are absorbed more rapidly and

extensively in the form of nonpolar small peptides than

an equimolar mixture of free amino acids [30,31]. The

data from our present acute infusion studies performed in

two rat models of hypertension pertain to the direct

effects of L-serine on vascular KCa channels, whereas

the net long-term effects of oral chronic L-serine treat-

ment in the future should be interpreted in the light of

potential effects mediated by bioconversion to glycine or

D-serine or both.

AcknowledgementsThis work was supported by grants-in-aid (MOP-67060)

from the Canadian Institutes of Health Research (CIHR)

and the Heart & Stroke Foundation of Saskatchewan.

There was no conflict of interest.

References1 Mishra RC, Tripathy S, Desai KM, Quest D, Lu Y, Akhtar J, Gopalakrishnan

V. Nitiric oxide synthase inhibition promotes endothelium dependentvasodilatation and antihypertensive effect of L-serine. Hypertension 2008;51:791–796.

2 Jalil JE. Blood pressure reduction, potassium channels, and theendothelium: insights from L-serine. Hypertension 2008; 51:626–628.

3 de Koning TJ, Snell K, Duran M, Berger R, Poll-The B-T, Surtees R. L-Serinein disease and development. Biochem J 2003; 371:653–661.

4 de Koning TJ, Klomp LW. Serine-deficiency syndromes. Curr Opin Neurol2004; 17:197–204.

5 Heyman S, Spokes K, Rosen S, Epstein FH. Mechanism of glycineprotection in hypoxic injury: analogies with glycine receptor. Kidney Int1992; 42:41–45.

6 Zhong Z, Arteel GE, Connor HD, Yin M, Frankenberg MV, Stachlewitz RF,et al. Cyclosporin A increases hypoxia and free radical production in ratkidneys: prevention by dietary glycine. Am J Physiol Renal Physiol 1998;275:F595–F604.

7 Zhong Z, Connor HD, Yin M, Moss N, Mason RP, Bunzendahl H, et al.Dietary glycine and renal denervation prevents cyclosporin A-inducedhydroxyl radical production in rat kidney. Mol Pharmacol 1999; 56:455–463.

8 Matilla B, Mauriz JL, Culebras JM, Gonzalez-Gallego J, Gonzalez P. Glycine:a cell-protecting antioxidant nutrient. Nutr Hosp 2002; 17:2–9.

9 Wheeler M, Stachlewitz RF, Yamashina S, Ikejima K, Morrow AL, ThurmanRG. Glycine-gated chloride channels in neutrophils attenuate calcium influxand superoxide production. FASEB J 2000; 14:476–484.

10 Zhang Y, Ikejima K, Honda H, Kitamura T, Takei Y, Sato N. Glycineprevents apoptosis of rat sinusoidal endothelial cells caused bydeprivation of vascular endothelial growth factor. Hepatology 2000;32:542–546.

11 Jackson AA, Dunn RL, Marchand MC, Lanley-Evans SC. Increased systolicblood pressure in rats induced by a maternal low-protein diet is reversed bydietary supplementation with glycine. Clin Sci (Lond) 2002; 103:641–642.

12 Deng A, Valdivielso JM, Munger KA, Blantz RC, Thomson SC. VasodilatoryN-methyl-D-aspartate receptors are constitutively expressed in rat kidney.J Am Soc Nephrol 2002; 13:1381–1384.

13 Brawley L, Torrens C, Anthony FW, Itoh S, Wheeler T, Jackson AA, et al.Glycine rectifies vascular dysfunction induced by dietary protein imbalanceduring pregnancy. J Physiol 2003; 554:497–504.

14 Slamowitz LA, Deng A, Hammes JS, Gabbai F, Thomson SC.Glomerulotubular balance, dietary protein, and the renal response toglycine in diabetic rats. Am J Physiol Regul Integr Comp Physiol 2002;282:R1096–R1103.


15 Slamowitz LA, Gabbai FB, Khang SJ, Satriano J, Thareau S, Deng A, et al.Protein intake regulates the vasodilatory function of the kidney and NMDAreceptor expression. Am J Physiol Regul Integr Comp Physiol 2004;287:R1184–R1189.

16 Hafidi ME, Perez I, Zamora J, Soto V, Carvajal-Sandoval G, Banos G.Glycine intake decreases plasma free fatty acids, adipose cell size, andblood pressure in sucrose-fed rats. Am J Physiol Regul Integr CompPhysiol 2004; 287:R1387–R1393.

17 Hafidi ME, Perez I, Banos G. Is glycine effective against elevated bloodpressure? Curr Opin Clin Nutr Metab Care 2006; 9:26–31.

18 Qui C, Engels K, Samsell L, Baylis C. Renal effects of acute amino acidinfusion in hypertension induced by chronic nitric oxide blockade.Hypertension 1995; 25:61–66.

19 Vasdev S, Whalen M, Ford CA, Longerich L, Prabhakaran V, Parai S.Ethanol- and threonine-induced hypertension in rats: a commonmechanism. Can J Cardiol 1995; 11:807–815.

20 Shinde UA, Desai KM, Yu C, Gopalakrishnan V. Nitric oxide synthaseinhibition exaggerates the hypotensive response to ghrelin: role of calcium-activated potassium channels. J Hypertens 2005; 23:779–784.

21 Desai KM, Gopalakrishnan V, Hiebert LM, McNeill JR, Wilson TW. EDHF-mediated rapid restoration of hypotensive response to acetylcholine afterchronic, but not acute, nitric oxide synthase inhibition in rats. Eur JPharmacol 2006; 546:120–126.

22 Rees DD, Palmer RM, Schulz R, Hodson HF, Moncada S. Characterizationof three inhibitors of endothelial nitric oxide synthase in vitro and in vivo. Br JPharmacol 1990; 101:746–752.

23 Milman G, Maor Y, Abu-Lafi S, Horowitz M, Gallily R, Batkai S, et al.N-Arachidonoyl L-serine, an endocannabinoid-like brain constituent withvasodilatory properties. Proc Natl Acad Sci U S A 2006; 103:2428–2433.

24 Takahashi K, Hayashi F, Nishikawa T. In vivo evidence for the link betweenL- and D-serine metabolism in rat cerebral cortex. J Neurchem 1997;69:1286–1290.

25 Leung JC, Travis BR, Verlander JW, Sandhu SK, Yang S-G, Zea AH, et al.Expression and developmental regulation of the NMDA receptor subunits inthe kidney and cariovascular system. Am J Physiol Regul Integr CompPhysiol 2002; 283:R964–R971.

26 De Nicola L, Blantz RC, Gabbai FB. Nitric oxide and angiotensin II.Glomerular and tubular interaction in the rat. J Clin Invest 1992; 89:1248–1256.

27 Sgaragli G, Pavan F. Effects of amino acid compounds injected intocerebral fluid spaces on colonic temperature, arterial blood pressure andbehaviour of the rat. Neuropharmacology 1972; 11:45–56.

28 Huang Q-F, Gebrewold A, Zhang A, Altura BT, Altura BM. Role of excitatoryamino acids in regulation of rat pial microvasculature. Am J Physiol RegulIntegr Comp Physiol 1994; 266:R158–R163.

29 Addington JW. L-Serine: treatment for chronic fatigue syndrome (CFIDS).http://www.immunesupport.com/library/showarticle.cfm/ID/162/[Accessed 18 March 2008].

30 Zaloga GP, Siddiqui RA. Biologically active dietary peptides. Mini Rev MedChem 2004; 4:815–821.

31 Rerat A, Simoes-Nunes C, Mendy F, Vaissade P, Vaugelade P. Splanchnicfluxes of amino acids after duodenal infusion of carbohydrate solutionscontaining free amino acids or oligopeptides in the nonanaesthetized pig.Br J Nutr 1992; 68:111–1138.


http://www.immunesupport.com/library/showarticle.cfm/ID/162/

Documents

L-Serine lowers while glycine increases blood pressure in chronic L-NAME-treated and spontaneously hypertensive rats