Chronobiology of the renin-angiotensin-aldosterone system ... · blood flow and glomerular...

Chronobiology International, Early Online: 1–16, (2013)! Informa Healthcare USA, Inc.ISSN: 0742-0528 print / 1525-6073 onlineDOI: 10.3109/07420528.2013.807275

Chronobiology of the renin-angiotensin-aldosterone system in dogs:relation to blood pressure and renal physiologyy

Jonathan P. Mochel1,2, Martin Fink2, Mathieu Peyrou3, Cyril Desevaux4, Mark Deurinck5,Jerome M. Giraudel6, and Meindert Danhof1

1Department of Pharmacology, Leiden-Academic Centre for Drug Research, Leiden, The Netherlands, 2Department ofModeling and Simulation, Novartis Campus St. Johann, Basel, Switzerland, 3Department of Therapeutics Research,Novartis Centre de Recherche Sante Animale SA, Saint-Aubin, Switzerland, 4Department of Pre-Clinical Safety, NovartisCentre de Recherche Sante Animale SA, Saint-Aubin, Switzerland, 5Department of Pre-clinical Safety, Novartis Institutesof Biomedical Research, Basel, Switzerland, and 6Yarrandoo Research and Development Centre, Novartis Animal HealthAustralasia Pty. Limited, Kemps Creek, Australia

The renin-angiotensin-aldosterone system (RAAS) plays a pivotal role in the regulation of blood pressure and volumehomeostasis. Its contribution to the development of cardiovascular diseases has long been recognized. Extensiveliterature has shown that peptides of the RAAS oscillate with a circadian periodicity in humans, under strong influenceof posture, sleep, and age. Although observations of time-variant changes in the renin cascade are available in dogs,no detailed chronobiological investigation has been conducted so far. The present studies were designed to explorethe circadian variations of plasma renin activity (RA) and urinary aldosterone-to-creatinine ratio (UA:C) in relationto blood pressure (BP), sodium (UNa, UNa,fe), and potassium (UK, UK,fe) renal handling. Data derived from intensiveblood and urine sampling, as well as continuous BP monitoring, were collected throughout a 24-h time period,and analyzed by means of nonlinear mixed-effects models. Differences between the geometric means of day andnight observations were compared by parametric statistics. Our results show that variables of the renin cascade, BP,and urinary electrolytes oscillate with significant day-night differences in dogs. An approximately 2-fold (1.6–3.2-fold)change between the average day and night measurements was found for RA (p50.001), UA:C (p¼ 0.01), UK,fe

(p¼ 0.01), and UNa (p¼ 0.007). Circadian variations in BP, albeit small (less than 10 mm Hg), were statisticallysignificant (p50.01) and supported by the model-based analysis. For all variables but UNa and UNa,fe, the levels werehigher at night than during the day. The data also indicate that blood pressure oscillates in parallel to the RAAS, suchthat, as opposed to healthy humans, BP does not drop at night in dogs. The postprandial decrease in RA is assumed tobe related to body fluid volume expansion secondary to water and sodium intake, whereas the reduction of UA:C

reflects aldosterone-stimulated secretion by the renin-angiotensin II pathway. UNa and UNa,fe peaked in the afternoon,about 7–8 h after food intake, which is consistent with the ‘‘impulse-response pattern’’ of sodium excretion describedin previous publications. Finally, UK and UK,fe mirrored aldosterone-mediated potassium secretion in the kidneytubules. To describe the circadian variations of the various variables, two different mathematical representations wereapplied. A cosine model with a fixed 24-h period was found to fit the periodic variations of RA, UA:C, UK, UK,fe, and BPwell, whereas changes in UNa and UNa,fe were best characterized by a surge model. The use of nonlinear mixed effectsallowed estimation of population characteristics that can influence the periodicity of the RAAS. Specifically, sodiumintake was found to interact with the tonic and the phasic secretion of renin, suggesting that varying feeding timecould also impact the chronobiology of the renin cascade. (Author correspondence: [email protected])

Keywords: Chronotherapy, Circadian periodicity, Covariate, Nonlinear mixed-effects modeling

INTRODUCTION

Within the mammalian kidney, the juxtaglomerular

apparatus (JGA) consists of tubular and vascular struc-

tures whose role is critical in the regulation of renal

blood flow and glomerular filtration rate (Thurau, 1966).

As part of the JGA, the macula densa (MD) is an area

of specialized cells that respond to changes in sodium

chloride concentration in the tubular fluid by mediating

yPart of the results of these studies were presented in abstract form at the European Association for Veterinary Pharmacology andToxicology (EAVPT), July 8–12, 2012, Noordwijkerhout, The Netherlands.Correspondence: Jonathan Mochel, MSc, DVM, ECVPT Resident, Novartis Campus St. Johann, WSJ-27.6.022, 4002 Basel,Switzerland. Tel.: +41 61 696 9077; Fax: +41 61 324 5874; E-mail: [email protected]

Submitted January 28, 2013, Returned for revision May 16, 2013, Accepted May 17, 2013

1

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

mailto:[email protected]

renin release from the granular cells of the JGA. Renin

release triggers a cascade of biochemical events, with

the subsequent formation of angiotensin I, angiotensin

II, and aldosterone. The so-called renin-angiotensin-

aldosterone system (RAAS) plays a pivotal role in the

control of blood pressure (BP) and volume homeostasis

by regulating sodium and potassium exchanges in the

kidney tubules. Its contribution to the development

of hypertension and chronic heart failure has long

been recognized (Kunita et al., 1976; Nicholls et al., 1993;

Pedersen et al., 1995; Roche et al., 2002). Further, it has

been shown that inhibition of the renin-angiotensin

pathway is an effective approach to treat these disorders

(Allikmets, 2007; Turnbull, 2003).

It has been established that RAAS peptides oscillate

with a circadian periodicity in rodents and humans

(Cugini et al., 1981, 1984, 1985, 1986, 1987; Kawasaki

et al., 1990). Several physiological influences have been

shown to influence day-night variations of the renin-

angiotensin cascade, including alterations in posture

(Muller et al., 1958), sleep cycles (Brandenberger et al.,

1985, 1994), and age (Cugini et al., 1985).

Parallel to day-night changes in RAAS peptides, data

from telemetry monitoring in humans have consistently

demonstrated that BP has a reproducible circadian

rhythm, with highest levels measured in the morning,

and lowest values around midnight (Smolenski & Haus,

1994; Staessen et al., 1992).

Although observations of time-variant changes in

RAAS peptides in dogs have been published (Corea

et al., 1996; Gordon & Lavie, 1985; Reinhardt et al., 1996),

no detailed characterization of the chronobiology of

these variables is currently available in the literature.

In addition, the question of whether BP oscillates over

the 24-h span in dogs is still a matter of debate

(Miyazaki et al., 2002; Piccione et al., 2005, Soloviev

et al., 2006).

Benefits of acquiring knowledge on the chronobiol-

ogy of the RAAS and BP in dogs are twofold. First,

it provides a scientific rationale to support further

research on administration time-dependent efficacy

of treatment in the management of RAAS-related

diseases, such as chronic canine valvular heart disease.

This condition is characterized by the thickening

and shortening of the atrioventricular valves, and affects

about 75% of dogs over the age of 16 (Guglielmini, 2003).

Second, human and canine heart failure share common

pathophysiological features. Specifically, activation of

the renin cascade is one of the key neurohumoral

responses to the reduced cardiac output observed in

both species (Sayer et al., 2009; Watkins et al., 1976).

Therefore, knowledge on the circadian periodicity of

renin, in relation to BP and renal physiology in dogs,

also constitutes valuable information to the understand-

ing of RAAS biology in humans.

This investigation offers a comprehensive description

of the chronobiology of the renin cascade in dogs, in

relation to BP and renal sodium-potassium handling,

using a nonlinear mixed-effects modeling approach

(also referred to as population modeling).

MATERIALS AND METHODS

Sample collection for measurement of plasma and

urinary variables and telemetry recordings were per-

formed in two separate studies and distinct groups

of animals, in order to preclude manipulation-related

disturbances (e.g., venipuncture) on blood pressure

(Baumgart, 1991).

AnimalsThe studies were performed in compliance with a

registered Swiss permit covering animal experiments

for cardiovascular research in dogs as approved by

the Cantonal Animal Welfare Committee and the

Veterinary Services. The study protocols were designed

to use the fewest number of animals possible while

being consistent with the scientific needs of the

study, and conformed to international ethical standards

(Portaluppi et al., 2010).

In the first experiment (Study a; Figure 1), blood

and urinary samples for measurement of RA, urinary

aldosterone, and renal sodium-potassium exchanges

were taken from 18 adult (9 males and 9 females)

healthy, non-neutered beagle dogs weighing between

12.0 and 18.0 kg (Marshall Europe, Green Hill,

Montichiari, Italy).

In the second experiment (Study b; Figure 1), systolic

(SBP) and diastolic (DBP) arterial blood pressures

were continuously recorded from 6 adult healthy,

non-neutered telemetered male beagle dogs weighing

from 10.4 to 15.2 kg (Marshall Europe, Green Hill).

Suitability for inclusion was evaluated by a physical

examination and confirmed by measuring selected

hematological (red and white blood cells counts, hemo-

globin [Hb], hematocrit [Hct]) and clinical chemistry

(albumin, total protein, alanine aminotransferase,

aspartate aminotransferase, blood urea nitrogen, cre-

atinine) parameters in blood. Telemetry signal quality

and BP waveform analysis were used as additional

criterion for telemetry dogs.

Housing ConditionsPrior to the experiments, the dogs were acclimatized

to the experimental facility for a week. Animals were

housed in pens (about 2 m2/animal) containing granu-

late bedding material and an additional elevated plat-

form for resting. On the sampling days, dogs were

housed in metabolism cages. Over many sessions, the

dogs were trained to rest in metabolism cages for up to

12 h. Dogs from the telemetry study were group housed

and pair housed on days of BP recording. The study

rooms had natural daylight and additional artificial light

of similar intensity (400 lux) from 07:00 to 19:00 h. Room

temperature and relative humidity were within the

target ranges of 17–23 �C and 35–75%, respectively.

2 J. P. Mochel et al.

Chronobiology International

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

Drinking water quality was compliant with the Swiss

Federal Regulation on Foodstuff, and was offered ad

libitum. One week before and throughout the studies,

water and food intake were recorded on a daily basis.

The dogs were offered a normal-sodium diet (Biomill

Adult Medium, 0.5% sodium; Biomill, Herzogenbuchsee,

Switzerland) once daily at 07:00 h. Depending on the

size of the ration, the individual daily sodium intake

ranged from 34 to 80 mEq. The amount of food given per

dog was kept constant throughout the experiments.

Experimental ProcedureStudy a: Periodicity of the RAAS and UrinaryElectrolytes in DogsBlood specimens were collected every 2 h (starting from

07:00 h) into 1.2 - or 2.7-mL S-Monovette tubes

(Sarstedt, Newton, NC, USA). Due to the known sensi-

tivity of the renin-angiotensin cascade to posture and

external stimuli (Muller et al., 1958), specific precau-

tions were taken: dogs were kept and maintained in

the same position (up and standing) during blood

collection, sampling was performed in a sound-

protected room, and low-intensity lighting was used

for night sampling.

Blood samples were cooled on ice immediately

after withdrawal and centrifuged under refrigeration

(2� 1 �C, 15 min) within 30 min of sampling. Plasma was

then transferred into cooled propylene aliquots, snap-

frozen, and stored at �80 �C before determination

of RA. Urine samples were collected from the metabol-

ism cage every 4 h (starting from 07:00 h) into cooled

Erlenmeyer flasks and transferred into two distinct

plain tubes, for determination of urinary (i) aldosterone

(UAL) (stored at �80 �C), (ii) sodium (UNa) (stored at

4 �C), and (iii) potassium (UK) (stored at 4 �C) concen-

trations. Urinary aldosterone-to-creatinine ratios (UA:C)

were derived by measuring plasma and urinary creatin-

ine, as proposed by Gardner et al. (2007). The glomeru-

lar filtration rate (GFR), as well as sodium and

potassium fractional excretions (UNa,fe and UK,fe,

respectively), was determined using the renal clearance

of creatinine1 as described by Lefebvre et al. (2008).

Study b: Periodicity of Blood Pressure in DogsThe telemetry system consisted of an implantable

transmitter (chronic use TL11M2-D70-PCT implant;

Data Sciences International [DSI], St. Paul, MN, USA),

cage receivers (RMC-1 general purpose receiver for

metal cages; DSI), ambient pressure monitor (APR-1

ambient pressure reference; DSI), data exchange matrix

(Dataquest ART data exchange matrix; DSI), and an

electronic data acquisition system (DQ ART 2.1

Dataquest acquisition and analysis system; DSI).

The transmitter contained a pressure sensor for BP

measurement, an electric potential sensor for electro-

cardiogram (ECG) recording, and a thermometer for

measuring core body temperature. The pressure sensor

was connected to a fluid-filled urethane catheter

inserted in the arteria femoralis with the tip placed in

the arteria inguinalis. ECG leads were implanted intra-

muscularly in an Einthoven lead II configuration. The

transmitted signal was captured by the cage receivers

and transformed by the data exchange matrix. After

surgery, the dogs were allowed to recover for 1 month.

The procedures were in compliance with the recom-

mendations of the BVAAWF/FRAME/RSPCA/UFAW

Joint Working Group on Refinement regarding telemetry

(Morton et al., 2003). Beginning around 07:00 h and

ending about 24 h later, the electrocardiogram and

arterial BP and core body temperature were recorded

continuously. Data from every third quarter of an hour

were analyzed and averaged to derive an hourly value.

Intensive blood and urine samplingover 24-hour periods

Period of acclimation Regular diet(07.00am)

Regular diet(07.00am)

Period of acclimation

Continuous telemetry recordingover a 24-hour time window

Study a. Periodicity of the RAAS and urinaryelectrolytes in dogs (n: 18 dogs)

Study b. Periodicity of blood pressure in dogs(n: 6 dogs)

FIGURE 1. Study outline. All animals were healthy adult, non-neutered beagle dogs fed a normal-sodium diet at 07:00 h. (a) Periodicity

of the RAAS and urinary electrolytes in dogs (Study a). Blood and urinary samples for measurement of RA, urinary aldosterone, and

renal sodium-potassium exchanges were taken from 18 beagle dogs (9 males and 9 females) weighing between 12.0 and 18.0 kg. Samples

were collected every 2 and 4 h for plasma and urinary specimens, respectively. (b) Periodicity of blood pressure in dogs (Study b).

Systolic and diastolic arterial BPs were continuously recorded from 6 telemetered male beagle dogs weighing from 10.4 to 15.2 kg.

1Because of limited tubular secretion and reabsorption (Guignard & Drukker, 1999; O’Connell et al., 1962), creatinine clearance canbe used to provide estimates of the glomerular filtration rate (GFR) in dogs.

RAAS and Blood Pressure Chronobiology in Dogs 3

! Informa Healthcare USA, Inc.

Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

Analytical MethodsRenin activity was determined by measuring the rate

of angiotensin I (AI) formation after incubation of

endogenous renin and angiotensinogen in plasma

(2 h, 37 �C, pH 7.2). AI concentrations were measured

after liquid-solid extraction using a validated enzyme

immunoassay (EIA) (S-1188 Angiotensin I-EIA kit; host:

rabbit high-sensitivity European Conformity (CE)-

marked; Bachem, Bubendorf, Switzerland). Analyses

were performed in duplicates; values with a coefficient

of variance (CV%) below 25% were retained for statis-

tical evaluation. UAL concentrations were determined

with a liquid chromatography–tandem mass spectrom-

etry method using an isotope dilution technique. Urine

samples were not subjected to acid hydrolysis before

extraction, that is, only nonconjugated aldosterone

concentrations were measured. Calibration standards

ranging from 0.05 to 10.0 ng/mL were used for quanti-

fication. Sodium and potassium concentrations were

quantified using an ion-selective electrode measure-

ment method (Olympus AU 400; Beckmann Coulter

International, Nyon, Switzerland). A colorimetric Jaffe

reaction was used for quantification of plasma and

urinary creatinine concentrations.

Data AnalysisQuantification of Day-Night DifferencesComparison of area under the curves (AUCs). For rich

data (i.e., RA, SBP, and DBP), the individual area

under the curve of day (AUCs[07:00–19:00]) vs. night

(AUCs[19:00–07:00]) observations were derived using the

linear trapezoidal rule, and compared by paired t tests

using R version 2.15 (The R Foundation for Statistical

Computing, Vienna, Austria). The homogeneity of vari-

ances between day and night AUCs was assessed

using the Bartlett’s test. p values were adjusted for

multiple comparisons using the Sidak procedure, and

were reported for a¼ 0.05. Distributions were inspected

for normality and skewness.

Comparison of day vs. night urinary spot samples. For

sparse data (i.e., UA:C, UNa, UNa,fe, UK, and UK,fe), day

(11:00 and 15:00 h) vs. night (23:00 and 03:00 h) meas-

urements derived from urinary spot samples were

compared by paired t tests. The Bartlett’s test was

used for evaluating homoscedasticity between day and

night levels. p values were adjusted for multiple com-

parisons using the Sidak correction, and were reported

for a¼ 0.05.

Chronobiological AnalysisChi-square statistics for testing the zero-amplitude

hypothesis.. For rhythm detection, a p value was

derived from the difference in objective function value

(OFV) between the fit of a straight line approximation

of the mean (1 model parameter) and that of a cosine

function (3 model parameters). In nonlinear mixed-

effects modeling, the OFV is approximated to minus

2 times the logarithm of the likelihood of the data given

the model, with a lower value indicating a better model

(Sheiner & Ludden, 1992). It is used to draw statistical

inference about the goodness-of-fit of the mathematical

models. A negative OFV would indicate that the likeli-

hood is greater than 1.

The difference in OFV between two models follows

an asymptotic chi-square distribution, with degrees

of freedom equal to the difference in the number of

parameters between contending models. A periodic

rhythm was considered as statistically significant for a

drop in OFV45.9 (for a risk level a¼ 0.05).

Nonlinear mixed-effects (population) modeling. Nyc-

themeral variations in RA, BP, and urinary variables

were characterized by means of a nonlinear mixed-

effects modeling approach, using NONMEM version 7.2

(Icon Development Solutions, Ellicott City, MD, USA).

Individual model parameters were obtained post hoc as

empirical Bayes estimates.

Nonlinear mixed-effects models were written using

the following format (Equation 1):

yij ¼ f �i, tij� �

þ g �i, tij,��

� "ij, j ¼ 1, . . . ni

�i ¼ � � exp�i, i ¼ 1, . . . , Nð1Þ

where yij is the observed variable (e.g., RA, UA:C)

measured on the ith individual at time tij, �i is the

vector of individual parameters, f(�i, tij) is the value

of that observed variable at time tij for an individual

with parameters �i, and "ij is an independent random

variable. The function g(�i, tij,�) is the standard devi-

ation of the error of a given measurement at time tij. In

population modeling f(�i, tij) is referred to as structural

model (error-free), whereas g(�i, tij,�) is the residual

error model (combining unexplained variability and

measurement noise). � represents the typical value

(population average) of a model parameter. The sources

of variation between the individual parameters �i can be

explained by population characteristics (i.e., covariates)

that can be included additionally or proportionally to m.

The independent random variables �i represent the

unexplained difference between the value of the indi-

vidual parameters �i and their average �. Random

variables yij, "ij, �i, and �i were assumed to be log-

normally distributed.

The first-order conditional estimation method with

interaction was used for all analyses. Covariate search

was performed using the stepwise covariate model

building tool of Perl-speaks-NONMEM (Mats Karlsson,

Niclas Jonsson, and Andrew Hooker, 2008) with forward

inclusion based on p¼ 0.05 and afterwards backward

exclusion based on p¼ 0.01. Analyzed covariates were

sex, dietary sodium, and body weight.

Standard goodness-of-fit diagnostics, including

population and individual predictions vs. observations,

and the distributions of weighted residuals over time,

were performed to assess the adequacy of selected



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

models. Graphical assessment was performed using the

R-based software Xpose version 4.1 (Jonsson & Karlsson,

1999) in R version 2.15.

Model selection was based on statistical significance

between competing models using the objective function

value obtained from NONMEM, graphical evaluation,

and validity of parameter estimates. Residual error

estimates from the population models were used as

supportive information for evaluation of lack of fit.

The normality and independence of residuals were

evaluated using the Shapiro-Wilk and Durbin-Watson

tests, respectively.

Correlations Between signalsPhase relationships between observed variables were

investigated using a circular version of the Pearson’s

product moment correlation in R version 2.15

(Agostinelli & Lund, 2011; R package ‘‘circular’’:

Circular Statistics).

Differences between the average phases of blood

pressure, plasma, and urinary variables were character-

ized by introducing a phase shift in the population

models. Chi-square statistics were used to decide on

the statistical significance of the drop in OFV when

including this additional model parameter.

p values were adjusted for multiple comparisons

using the Sidak procedure, and were reported for

a¼ 0.05.

RESULTS

Study a: Periodicity of the RAAS and UrinaryElectrolytes in DogsQuantification of Day-Night DifferencesFor RA, UA:C, UK, UK,fe, UNa, and UNa,fe, the p values

derived from the Bartlett’s test were greater than the

chosen alpha level of 0.05. Therefore, the null hypothesis

that the variances of day and night measurements were

identical could not be rejected.

As presented in Figures 2 and 3, in dogs fed early

in the morning RA, UA:C, UK, and UK,fe oscillated with

circadian changes over the 24-h span. Values were low

in the morning, rose through the afternoon, and peaked

in the evening (around 21:00, 19:00 and 19:00 h for RA,

Time (hours)

Ren

in a

ctiv

ity (

pg/m

L/h)

7 am 1 pm 7 pm 1 am 7 am

4080

120

160

200

Darkness

Time (hours)

Urin

ary

aldo

ster

one

to c

reat

inin

e ra

tio (

pg/m

g)


020

4060

80

Darkness

AUC (Day) AUC (Night)

Ren

in a

ctiv

ity (

pg/m

L/h)

050

015

0025

0035

00

**

Measurement (Day) Measurement (Night)

Urin

ary

aldo

ster

one

to c

reat

inin

e ra

tio (

pg/m

g)

020

4060

80

**

FIGURE 2. Circadian changes in renin activity and urinary aldosterone-to-creatinine ratio in dogs. (Top) Geometric mean plasma renin

activity (pg/mL/h) (left panel) and urinary aldosterone-to-creatinine ratio (pg/mg) (right panel) vs. time profile in dogs fed a normal-

sodium diet (0.5% sodium) at 07:00 h (n¼ 18). Vertical bars indicate 1 standard error of the geometric mean. The data show that RA and

UA:C oscillate with circadian changes over the 24-h span. Measurements were low in the morning, rose through the afternoon, and peaked

in the evening (around 21:00 and 19:00 h for RA and UA:C, respectively), with an apparently shorter time to return to baseline for UA:C levels

compared with RA. (Bottom) Area under the curve of day (AUCs[07:00–19:00]) vs. night (AUCs[19:00–07:00]) renin activity (left panel) and urinary

aldosterone-to-creatinine ratio (right panel) in dogs fed a normal-sodium diet (0.5 % sodium) at 07:00 h (n¼ 18). Vertical bars indicate 1

standard error of the geometric mean. Significant differences are indicated by double asterisks (**). Night measurements were on average

2.0 and 3.2 times significantly greater than day measurements for RA (p50.001) and UA:C (p¼ 0.01), respectively.



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

UA:C, and UK,fe, respectively), with an apparently shorter

time to return to baseline for UA:C, UK, and UK,fe levels

compared with RA. Night measurements were on aver-

age 2.0 and 3.2 times significantly greater than day

measurements for RA (p50.001) and UA:C (p¼ 0.01),

respectively (Figure 2).

UK and UK,fe oscillated in parallel, showing simi-

lar dynamics to UA:C and a peak around 19:00 h

(Figure 3). Day UK and UK,fe values were on average

1.2 (p40.05) and 1.6 (p¼ 0.01) times smaller than night

measurements. In contrast, UNa and UNa,fe levels were

high from morning to the middle of the afternoon

(peak around 15:00 h), and decreased from the latter half

of the afternoon to the early morning (Figure 3). UNa and

UNa,fe values were respectively 1.9 (p¼ 0.007) and 1.3

(p¼ 0.05) times higher during daytime.

Chi-Square Statistics for Testing the Zero-AmplitudeHypothesisThe cosinor fit of the data was statistically significant

(p� 0.001) for RA, UA:C, UK, UK,fe, UNa, and UNa,fe,

supporting the hypothesis of time-varying dynamics

with a 24-h period (Table 1). In contrast, no circadian

●

●

●

●

●●

●

Time (hours)


0.0

0.2

0.4

0.6

0.8

040

8012

016

0

●

●

●

●

●●

●

●

●

Sodium fractional excretion (%)Urinary sodium concentration (mmol/L)

Darkness

●

●

●

●

●

●

●

Time (hours)


010

2030

050

100

150

●

●

●

●

●●

●

●

●

Potassium fractional excretion (%)Urinary potassium concentration (mmol/L)

Darkness

Day Night

Sod

ium

frac

tiona

l exc

retio

n (%

)

0.0

0.1

0.2

0.3

0.4

0.5

Day Night

Urin

ary

sodi

um c

once

ntra

tion

(mm

ol/L

)

020

4060

8010

0

**

Day Night

Pot

assi

um fr

actio

nal e

xcre

tion

(%)

05

1015

20

**

Day NightUrin

ary

pota

ssiu

m c

once

ntra

tion

(mm

ol/L

)

020

4060

8010

0

FIGURE 3. Circadian changes in sodium and potassium renal elimination (urinary concentration and fractional excretion) in dogs. (Top)

Geometric mean sodium (left panel) and potassium (right panel) renal elimination vs. time profile in dogs fed a normal-sodium diet (0.5%

sodium) at 07:00 h (n¼ 18). Solid line: fractional excretion (%, left axis); dashed line: urinary concentration (mmol/L, right axis). Vertical

bars indicate 1 standard error of the geometric mean. UNa,fe and UNa levels were high from morning to the middle of the afternoon (peak

observed around 15:00 h), and decreased from the latter half of the afternoon to the early morning. In contrast, UK,fe and UK showed a

profile very similar to UA:C, with a peak around 19:00 h and a rather short time to return to baseline values compared with RA. (Bottom) Day

vs. night urinary electrolyte measurements in dogs fed a normal-sodium diet (0.5 % sodium) at 07:00 h (n¼ 18). From left to right: sodium

fractional excretion (%), sodium concentration (mmol/L), potassium fractional excretion (%), and potassium concentration (mmol/L).

Vertical bars indicate 1 standard error of the geometric mean. Significant differences are indicated by double asterisks (**).UNa,fe and UNa

values were respectively 1.3 (p¼ 0.05) and 1.9 (p¼ 0.007) times higher during daytime. Unlike sodium, day UK,fe and UK values were on

average 1.6 (p¼ 0.001) and 1.2 (p40.05) times smaller than night measurements.

TABLE 1. Comparison of objective function value (OFV) for statistical testing of the zero-amplitude hypothesis.

OFV RA UA:C UK,fe UNa,fe UNa UK GFR SBP DBP

OFV (straight line) �51 240 �196 27 40 �16 28 �353 �333

OFV (cosine model) �203 203 �263 �36 1 �36 27 �370 �343

Difference �152 �37 �67 �63 �39 �20 �1 �17 �10

p value 50.001 50.001 50.001 50.001 50.001 50.001 NS 50.001 0.001

For rhythm detection, a p value was derived from the difference in OFV between the fit of a straight line approximation of the mean and

that of a cosine function. A periodic rhythm was considered as statistically significant for a drop in OFV45.9 (for a risk level of a¼ 0.05).

Model estimates of the amplitudes can be found in Table 2.



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

changes were found for the endogenous creatinine

clearance (GFR).

Nonlinear Mixed-Effects ModelingA cosine model (Figure 4) was found to fit the periodic

nature of plasma renin activity, aldosterone, and potas-

sium urinary elimination well, as shown by the standard

goodness-of-fit diagnostics, and the individual predic-

tions in Figures 5 and 6.

The selected structural model was written as follows

(Equation 2):

f ðtijÞ ¼ Mi � 1þ Ai � cos tij � i

� �� 2�

�i

� �� ð2Þ

where f(tij) is the predicted RA, UA:C, UK, or UK,fe value

at time tij, Mi is the mesor (daily average of rhythm) for

the ith individual, Ai is the amplitude of the cosine, i

is the acrophase (or time of peak), and � i is the fixed

24-h period of the cosine for that individual.

The residual error model combined an additive and

a proportional error term. Estimates of residual errors

(CV%) from the population models were 30%, 80%, 18%,

and 30% for RA, UA:C, UK, and UK,fe, respectively. The p

values derived from the Shapiro-Wilk and Durbin-

Watson tests were greater than the chosen alpha level

of 0.05. Hence, the null hypothesis that the residuals

were independent and came from a normally distrib-

uted population could not be rejected.

Changes in UNa and UNa,fe were best characterized by

means of a surge function (Nagaraja et al., 2003). The

structural model was written as follows (Equation 3):

gðtijÞ ¼ Bi � 1þ Ai � cos tij � i

� �� 2�

�i

� �� þ 1

� �� wi

ð3Þ

where g(tij) is the predicted UNa or UNa,fe level, Bi

is the baseline for the ith individual, Ai is the ampli-

tude of the cosine, i is the acrophase, � i is the

fixed 24-h period of the cosine, and wi represents

the width of the surge function for that individual

(Figure 4).

A proportional error model was used to account for

the residual noise in the measurement of UNa and UNa,fe.

Estimates of residual errors (CV%) from the population

models were 20% and 30% for UNa and UNa,fe, respect-

ively. The p values derived from the Shapiro-Wilk and

Durbin-Watson tests were greater than the chosen alpha

level of 0.05. Therefore, the null hypothesis that the

residuals were independent and came from a normally

distributed population could not be rejected.

Time (hours)

Obs

erve

d va

riabl

e (u

nit)

7 am 1 pm 7 pm 1 am 7 am 1 pm 7 pm 1 am 7 am

8010

012

014

0

M: mesor

τ: period

τ: period

Ψ: acrophase

Ψ: acrophase

A: amplitude

A: amplitude

Time (hours)

Obs

erve

d va

riabl

e (u

nit)

7 am 1 pm 7 pm 1 am 7 am 1 pm 7 pm 1 am 7 am

8010

012

014

0

B: baseline w: width

FIGURE 4. Model parameters of a cosine and a surge function. (Top) The shape of a cosine model is determined by a set of parameters:

(M, A, , and �), where M is the mesor (daily average of rhythm), A is the amplitude of the cosine, is the acrophase (or time of peak), and �

is the period (herein fixed to a value of 24 h). (Bottom) The structure of a surge function is similar to that of a cosine, with the substitution of

the mesor by the baseline (initial value of rhythm, B), and the addition of another parameter: the width of the surge (w). A cosine model was

found to fit the periodic variations of RA, UA:C, UK, UK,fe, and BP well, whereeas circadian changes in UNa and UNa,fe were best characterized

by means of a surge function, reflecting an afternoon peak sodium excretion followed by a monotonous decay, rather than periodic

oscillations around the clock.



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

Standard goodness-of-fit diagnostics and individual

predictions can be found in Figures 5 and 6.

Population parameter estimates, relative standard

errors (RSEs), and 90% confidence intervals (CIs) are

listed in Table 2. The precision of the final model

parameters was considered satisfactory (RSE530%).

Data from the various variables were fitted simultan-

eously, leveraging the richness of the plasma data (RA)

to derive the acrophase of the (more sparse) urinary

variables. The model estimated a 3-h delay between the

acrophase of RA and that of UA:C, UK, or UK,fe. The peak

UNa and UNa,fe was estimated to lie around 14:30 h.

Estimates from the model further indicate that UA:C

oscillated with a larger relative amplitude (ca. 58% of the

mesor) compared with RA, UK, and UK,fe (respectively

26%, 24%, and 21% of the mesor/baseline value).

CorrelationsStatistics confirmed the positive correlation between

the phase of UA:C and that of UK,fe (r¼ 0.3, p¼ 0.009).

The correlation between plasma renin activity and

the renal elimination of aldosterone was also supported

by the Pearson’s product moment correlation test

(r¼ 0.2, p¼ 0.03). Finally, a positive correlation was

found between the phase of UK,fe and that of UNa,fe

(r¼ 0.4, p50.001). Results are summarized in Table 3.

Influence of Population Characteristics on the Periodicityof RAAS-Related VariablesDietary sodium intake and sex were found to be

significant covariates to explain part of the between-

subject variability in RA, UA:C, and urinary electrolytes

(Table 4).

Results from the covariate analysis indicate that

sodium intake (which was proportional to the size of

the ration) had a significant effect on the mesor of UA:C

(p50.001), RA (p50.001), UK,fe (p50.001), the baseline

of UNa,fe (p50.001), and the amplitude of RA (p50.001)

and UNa,fe (p50.001). The effect of sodium intake on the

periodicity of RA is further illustrated in Figure 7, using

predictions from the population model. Specifically, the

mesor RA of a typical 13-kg healthy female dog fed

around 190 g of dry food per day (i.e., 34 mEq Na/24 h)

would be 140 pg/mL/h, with an amplitude of 27 pg/mL/

h. Increasing the portion size from 190 to 440 g (i.e., 80

mEq Na/24 h) per day would result in a decrease of both

3.5 4.0 4.5 5.0

12

34

56

7

Renin activity

Population prediction

Obs

erva

tion

(log)

1 2 3 4 5 6

12

34

56

7

Individual prediction

Obs

erva

tion

(log)

0 5 10 15 20

−2−1

01

2

Time (hours)

CW

RE

S

2.0 2.5 3.0 3.5 4.0 4.5

01

23

45

Urinary aldosterone:creatinine

Population predictionO

bser

vatio

n (lo

g)

0 1 2 3 4 5

01

23

45


Obs

erva

tion

(log)

0 5 10 15 20

−2−1

01

Time (hours)

CW

RE

S

2.1 2.3 2.5 2.7

2.0

2.5

3.0

Potassium fractional excretion


Obs

erva

tion

(log)

2.0 2.2 2.4 2.6 2.8

2.0

2.5

3.0

Individual predictionO

bser

vatio

n (lo

g)

0 5 10 15 20

−3−2

−10

12

Time (hours)

CW

RE

S

−2.0 −1.5 −1.0 −0.5

−3.0

−2.0

−1.0

0.0

Sodium fractional excretion


Obs

erva

tion

(log)

−2.0 −1.5 −1.0 −0.5

−3.0

−2.0

−1.0

0.0


Obs

erva

tion

(log)

0 5 10 15 20

−3−2

−10

12

Time (hours)C

WR

ES

4.4 4.6 4.8 5.0

4.0

4.5

5.0

Systolic & Diastolic BP


Obs

erva

tion

(log)

4.0 4.4 4.8 5.2

4.0

4.5

5.0


Obs

erva

tion

(log)

0 5 10 15 20

−4−2

02

Time (hours)

CW

RE

S

FIGURE 5. Standard goodness-of-fit diagnostics. Scatter plot of population (top panel) and individual predictions (middle panel) vs.

observations (log scale), and conditional weighted residuals (bottom panel, CWRES) of population predictions. From left to right: RA, UA:C,

UK,fe, UNa,fe, and BP. Solid black line: identity line. Dashed green line: regression line. For CWRES, the x-axis represents time after food

ingestion (e.g., time 0 is 07:00 h). Because of the very good agreement between the fractional excretion and the urinary concentration of

electrolytes, only fractional excretions are represented herein. Note: Population predictions are estimates of the average (plasma, urinary)

concentration, enzyme activity, etc. An adequate model presents the following characteristics: (i) the line of identity is aligned with the

regression line (for both individual and population predictions), whereas (ii) the residues (differences between observations and

predictions) are centered on a mean value of 0, with (iii) an homogeneous dispersion around the mean.



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

the mesor (from 140 to 100 pg/mL/h) and the amplitude

(from 27 to 12.8 pg/mL/h) of RA oscillations.

Sex had a significant impact on the amplitude of RA

(p50.001), UK,fe (p50.01), and UNa,fe (p50.001). An

almost 5-fold difference in the amplitude of RA was

observed between male and female dogs (Figure 7).

Finally, body weight had limited influence on the

periodicity of the RAAS, with a statistically significant

effect on the amplitude of RA only (p¼ 0.01).

Study b: Periodicity of Blood Pressure in DogsDBP and SBP oscillated parallel to RA throughout

the observation span (Figure 8). Accordingly, DBP and

SBP were found to be highly correlated, as indicated by

the significance of the Pearson’s product moment

correlation test (r¼ 0.8, p50.001) (Table 3).

AUCs[07:00–19:00] were on average 6.5% and 7% smaller

than AUCs[19:00–07:00] for DBP (p¼ 0.007) and SBP

(p50.001), respectively. This corresponds to an average

nocturnal increase of 5 mm Hg for DBP and 9 mm Hg

for SBP. The p values derived from the Bartlett’s test

were greater than the chosen alpha level of 0.05,

indicating similar variances between day and night

AUCs for SBP and DBP.

Concurrently, the model fit of blood pressure data

led to the rejection of the hypothesis that SBP and DBP

measurements (p¼ 0.01) were constant over time

(Table 1). Similar to RA, a cosine model with a fixed

24-h period was found to describe the periodic nature of

DBP and SBP well, as shown by the quality of the

population and individual fits (Figures 5 and 6).

The selected structural model was the same as

indicated in Equation 2, but with f(tij) being the

predicted DBP or SBP value at time tij

Based on the drop in OFV between competing

models, the average phase of blood pressure measure-

ments was found to be statistically different from that

of plasma and urinary variables (p¼ 0.04, p¼ 0.002, and

p50.001 for RA, UA:C and UK,fe, and UNa,fe, respectively).

Noise in DBP and SBP measurements was modeled

using a proportional error term. Estimates of residual

errors (CV%) from the population models were 7%

and 10% for DBP and SBP, respectively. Based on the

results of the Shapiro-Wilk and Durbin-Watson tests,

the null hypothesis that the residuals were independent

and came from a normally distributed population could

not be rejected.

Body weight and dietary sodium were not found to

have an effect on BP model parameters. Population

parameter estimates, RSEs, and 90% CIs can be found

in Table 2. The precision of the final model parameters

appeared highly satisfactory (RSE520%).

0 5 10 15 20

12

34

56

Renin activity

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

0 5 10 15 20

23

45

67

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

0 5 10 15 20

23

45

67

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

0 5 10 15

01

23

45

6

Urinary aldosterone:creatinine

Time (hours)O

bser

vatio

n, P

redi

ctio

n (lo

g)

0 5 10 15

01

23

45

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

0 2 4 6 8 10 12

01

23

45

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

0 5 10 15

01

23

4

Potassium fractional excretion

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

0 5 10 15

01

23

4

Time (hours)O

bser

vatio

n, P

redi

ctio

n (lo

g)

0 5 10 15 20

01

23

4

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

4 6 8 10 12 14 16

−0.8

−0.4

0.0

0.4

Sodium fractional excretion

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

0 5 10 15 20

−0.

8−

0.4

0.0

0.4

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

0 2 4 6 8 10 12

−0.8

−0.4

0.0

0.4

Time (hours)O

bser

vatio

n, P

redi

ctio

n (lo

g)

0 5 10 15 20

4.0

4.2

4.4

4.6

4.8

5.0

Systolic blood pressure

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

0 5 10 15 20

4.0

4.2

4.4

4.6

4.8

5.0

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

0 5 10 15 20

4.6

4.8

5.0

5.2

5.4

Time (hours)

Obs

erva

tion,

Pre

dict

ion

(log)

FIGURE 6. Individual prediction time-course profiles based on individual parameter estimates (obtained as empirical Bayes estimates).

Scatter plot of observed (open circles, log scale) and predicted (continuous blue line) individual data vs. time after food (hour). Dashed line:

population predictions. Out of clarity only a subset of 3 individuals per variable are represented herein (ordered in columns). From left to

right: RA, UA:C, UK,fe, UNa,fe, and SBP.



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

DISCUSSION

Various authors have amply reviewed the time-varying

changes of the renin-angiotensin cascade in humans. In

contrast, little is known about the chronobiology of

RAAS peptides in dogs. Gordon & Lavie (1985) described

nocturnal increases in renin activity with concomitant

peaks in urine osmolality and potassium excretion in

four adult, female mongrel dogs fed a normal-sodium

diet. More work on the circadian variations of BP in dogs

has been published. Most of these investigations failed

to demonstrate daily rhythmicity in BP (Miyazaki et al.,

2002; Soloviev et al., 2006), although Piccione et al.

(2005) disclosed strong evidence that BP and heart rate

oscillated around the clock.

TABLE 2. Parameter estimates of the population model, relative standard errors of the means (RSEs), and

90% confidence intervals (CIs).

Parameter Point estimate (unit) RSE 90% CI

Renin activity (RA)

Mesor 96 (pg/mL/h) 12% (76–116)

Amplitude 25 (pg/mL/h) 20% (17–34)

Acrophase 22:40 (h) 9% (20:40–23:50)

Urinary aldosterone-to-creatinine ratio (UA:C)

Mesor 20 (pg/mg) 19% (14–26)

Amplitude 11.5 (pg/mg) 28% (6.0–17.0)

Acrophase 19:40 (h) 9% (18:10–20:20)

Potassium fractional excretion (UK,fe)

Mesor 12.8 (%) 4% (11.9–13.7)

Amplitude 2.7 (%) 17% (1.9–3.5)

Acrophase 19:40 (h) 9% (18:10–20:20)

Urinary potassium concentration (UK)

Mesor 88 (mmol/L) 11% (72–104)

Amplitude 21 (mmol/L) 26% (12–30)

Acrophase 19:40 (h) 9% (18:10–20:20)

Sodium fractional excretion (UNa,fe)

Baseline 0.20 (%) 9% (0.17–0.23)

Amplitude 0.27 (%) 27% (0.15–0.39)

Acrophase 14:30 (h) 4% (13:30–15:30)

Width 4:00 (h) 16% (2:54–5:04)

Urinary sodium concentration (UNa)

Baseline 49 (mmol/L) 14% (37–61)

Amplitude 88 (mmol/L) 15% (66–110)

Acrophase 14:30 (h) 4% (13:30–15:30)

Width 4:00 (h) 16% (2:54–5:04)

Systolic BP (SBP)

Mesor 141 (mm Hg) 6% (127–155)

Amplitude 7.0 (mm Hg) 8% (6.0–8.0)

Acrophase 23:30 (h) 4% (22:43–00:17)

Diastolic BP (DBP)

Mesor 82 (mm Hg) 3% (78–86)

Amplitude 5.0 (mm Hg) 16% (3.6–6.4)

Acrophase 23:30 (h) 4% (22:43–00:17)

TABLE 3. Correlation table indicating Pearson’s product-moment

correlation (circular version) coefficient� and p value (in

parenthesis).

RA UA:C UK,fe UNa,fe

RA 1

UA:C r¼ 0.2 (p¼ 0.03) 1

UK,fe NS r¼ 0.3 (p¼ 0.009) 1

UNa,fe NS NS r¼ 0.4 (p50.001) 1

SBP DBP

SBP 1

DBP r¼ 0.8 (p50.001) 1

NS¼nonsignificant p value; 1¼diagonal elements of the correl-

ation table.

TABLE 4. Summary information on the effect of dietary sodium

and sex on the mesor/baseline and the amplitude of RA, UA:C, as

well as potassium and sodium renal excretion (UK,fe and UNa,fe).

RA UA:C UK,fe UNa,fe

(A) Effect of dietary Na on model parameters

Low to high Na intake ratio

Mesor or baseline 1.4** 3.8** 0.8** 1.9**

Amplitude 2.1** 1 1 0.25**

(B) Effect of sex on model parameters

Male-to-female ratio

Mesor or baseline 1 1 1 2**

Amplitude 4.5** 1 0.7** 0.3**

(A) Model parameter (M, B, A) ratio of low (34 mEq Na) to high

(80 mEq Na) daily sodium intake. (B) Model parameter (M, B, A)

ratio of male to female dogs. Note that body weight had limited

influence on the periodicity of the RAAS. The covariate search

was performed using the stepwise covariate model building tool

of Perl-speaks-NONMEM. **p50.01.



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

The present studies offer a comprehensive charac-

terization of the chronobiology of the renin cascade in

dogs in relation to BP and renal sodium-potassium

exchanges, using a nonlinear mixed-effects modeling

approach. The core value of population modeling lies in

its ability to separate the (between- and within-subject)

variability from the measurement error (noise), in order

to determine population characteristics (i.e., covariates)

that are able to explain the sources of variation between

individuals.

Telemetry recordings were performed in a distinct

cohort of animals to preclude manipulation-related

disturbances, such as venipuncture, on BP. Extensive

literature has demonstrated the influence of environ-

mental factors (e.g., noise, stress) on BP in humans

(Attarchi et al., 2012; Baumgart, 1991; Lindquist et al.,

1997; Rocha et al., 2002). The relation of stress to

elevated heart rate and BP in dogs is well known, as

most recently confirmed by Hoglund et al. (2012). In this

observational study, both SBP and DBP increased

when owners left dogs in the examination room while

recordings were made by the veterinarian alone.

The RAAS and Urinary Electrolytes Exhibit a ClearCircadian Periodicity in DogsOur data document a clear circadian rhythmicity of RA,

UA:C, UNa, UNa,fe, UK, and UK,fe in trained and relaxed

healthy dogs, under standardized conditions. Our

results are consistent with previous investigations

in dogs (Corea et al., 1996), horses (Clarke et al., 1978,

1988), and humans (Cugini et al., 1981, 1985), which

underlines the similarity of blood fluid homeostasis

between mammalian species. The endogenous clear-

ance of creatinine, used as estimate of the GFR, showed

no periodic rhythmicity, which is consistent with other

publications in dogs (e.g., Uechi et al., 1994).

In accordance with the usual feeding pattern in

domesticated dogs, the animals in these studies were

fed once daily at 07:00 h. The effect of feeding time

on the periodicity of the renin cascade remains unclear,

as illustrated by the conflicting results between Kunita

et al. (1976) and Ikonomov et al. (1981). Kunita et al.

have reported significant differences in the periodicity

of renin and aldosterone when meals were taken at

night instead of the usual times of the day in five

healthy volunteers. These results dispute data from

Ikonomov et al. showing that diurnal changes in food

intake do not affect the rhythmicity of renin and sodium

excretion. Investigations in horses (Clarke et al., 1978,

1988) and sheep (Blair-West & Brook, 1969) have led

to the conclusion that episodic feeding caused substan-

tial variations of the RAAS, as opposed to continuous

feeding. To the authors’ knowledge, no information is

currently available in dogs.

Time (hours)

Pre

dict

ed r

enin

act

ivity

(pg

/mL/

h)


4060

8010

012

014

0

Male dog

Time (hours)

Pre

dict

ed r

enin

act

ivity

(pg

/mL/

h)


4060

8010

012

014

0

Female dog

Time (hours)

Pre

dict

ed r

enin

act

ivity

(pg

/mL/

h)


6010

014

018

0 Dietary Na: 34mEq Na

Time (hours)

Pre

dict

ed r

enin

act

ivity

(pg

/mL/

h)


6010

014

018


Time (hours)

Pre

dict

ed r

enin

act

ivity

(pg

/mL/

h)


6010

014

018


FIGURE 7. Effect of sex and dietary sodium on the periodicity of renin activity in dogs. Predictions based on the covariate analysis.

(Top) Renin activity (pg/mL/h) predictions from the population model for a male (left panel) and a female (right panel) dog with similar

daily sodium intake (80 mEq Na), and similar body weight (13 kg). (Bottom) Renin activity (pg/mL/h) predictions from the population

model in a 13-kg female dog fed a normal-sodium diet of increasing size, with daily sodium intake ranging from 34 to 80 mEq Na. As

also shown in Table 4, results of the covariate analysis indicate that sex has a significant effect on the amplitude of RA, whereas sodium

intake also has an influence on the mesor of RA oscillations (i.e., the greater the amount of dietary sodium, the smaller the mesor and

the amplitude of RA).



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

A cosine model with a fixed 24-h period was found to

fit the periodic variations of RA, UA:C, UK, and UK,fe well,

as suggested by the quality of the standard goodness-

of-fit diagnostics and the individual predictions.

In contrast, circadian changes in UNa and UNa,fe were

best characterized by means of a surge model, as

described by Nagaraja et al. (2003), reflecting an after-

noon peak sodium excretion followed by a monotonous

decay, rather than periodic oscillations around the

clock.

RA and UA:C measurements were low in the morning,

rose through the afternoon, and peaked in the evening.

The morning decrease in RA is assumed to be related

to body fluid volume expansion secondary to water and

sodium intake, referred to as ‘‘postprandial decrease

of RA’’ in previous publications (Kaczmarczyk et al.,

1980; Seeliger et al., 1999). The similitude of RA and UA:C

signals, supported by the Pearson’s product moment

correlation test, reflects aldosterone-stimulated secre-

tion by the renin-angiotensin II pathway.

In contrast, UNa and UNa,fe were high from morning

to the middle of the afternoon, and peaked around

15:00 h. This phenomenon has been termed ‘‘impulse-

response pattern’’ of sodium excretion (Boemke et al.,

1995), and is characterized by a peak urinary elimination

4–8 h after meal ingestion. During the evening and night,

UNa was maintained on very low levels. The decrease

in UNa,fe, together with constant GFR, suggests that

tubular, rather than glomerular, events were primarily

involved in the reduced elimination of sodium during

night hours. This assumption is supported by the

concomitant activation of known sodium-conserving

mediators (i.e., UA:C) acting on the kidney tubules.

Variations in UK and UK,fe mirror aldosterone-

mediated excretion of potassium in the distal tubules.

Tubular transport is indeed the main modality of

potassium exchange in the kidneys, whereas sodium

reabsorption also occurs in the proximal kidney tubules,

independently of aldosterone. This divergence is further

supported by the positive correlation between UA:C

and UK,fe, and the nonsignificant correlation between

UA:C and UNa,fe.

Dietary Sodium Influences the Tonic and thePhasic Secretion of ReninThe relation of elevated RA levels to increased renin

secretion has been established by Schricker et al. (1994).

Specialized cells of the MD act as a sensing device

monitoring changes in sodium chloride concentrations,

so that granular cells of the JGA can adjust their net

Time (hours)

Sys

tolic

blo

od p

ress

ure

(mm

Hg)


120

130

140

150

160

Darkness

Time (hours)

Dia

stol

ic b

lood

pre

ssur

e (m

mH

g)


7080

9010

0

Darkness


Sys

tolic

blo

od p

ress

ure

(mm

Hg)

1200

1400

1600

1800

2000

**


Dia

stol

ic b

lood

pre

ssur

e (m

mH

g)

600

700

800

900

1100

1300

**

FIGURE 8. 24-Hour time-course profiles of systolic and diastolic blood pressures in dogs. (Top) Geometric mean systolic (left panel) and

diastolic (mm Hg) (right panel) blood pressures in dogs fed a normal-sodium diet (0.5% sodium) at 07:00 h (n¼ 6). Vertical bars indicate 1

standard error of the geometric mean. SBP and DBP oscillated parallel to RA over the observation span. Blood pressure increased in the first

half of the night, before returning to baseline in the early morning. (Bottom) Area under the curve of day (AUCs[07:00–19:00]) vs. night

(AUCs[19:00–07:00]) systolic (left panel) and diastolic (right panel) blood pressures in dogs fed a normal-sodium diet (0.5% sodium) at 07:00 h

(n¼ 6). Vertical bars indicate 1 standard error of the geometric mean. Significant differences are indicated by double asterisks (**).

AUCs[07:00–19:00] were on average 7% and 6.5% smaller than AUCs[19:00–07:00] for SBP (p50.001) and DBP (p¼ 0.007), respectively.



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

secretion of renin to maintain an appropriate RA

level (Laragh & Sealey, 2011). Herein, results from the

covariate analysis show that dietary sodium interacts

with the renin cascade, not only by influencing the

tonic (i.e., mesor), but also the phasic (i.e., amplitude)

secretion of renin (i.e., the greater the amount of

sodium intake, the smaller the mesor and the ampli-

tude of RA). Note that sex was also found to

be a significant source of variability, with a 4.5-fold

-difference between the amplitude of male and female

dogs.

The urinary aldosterone-to-creatinine ratio has been

validated as a reproducible measure of 24-h urinary

aldosterone excretion in dogs (Gardner et al., 2007). UA:C

reflects aldosterone production over several to many

hours, eliminating minute-to-minute variation, as

seen with plasma aldosterone concentrations. From

the covariate analysis, sodium intake had a significant

effect on the mesor, but not on the amplitude of UA:C

oscillations, indicating that dietary sodium influences

the tonic, but possibly not the phasic, secretion of

aldosterone. This should, however, be interpreted with

caution given the relatively small changes in dietary

sodium investigated in our experiment.

Blood Pressure and Renin Activity Display SimilarFluctuations Around the ClockThe levels of DBP and SBP observed in this experiment

were in agreement with previous investigations in dogs

(Mishina & Watanabe, 2008; Miyazaki et al., 2002), and

reference values reported in humans (Baumgart, 1991).

Further, the high level of correlation between DBP and

SBP was in the same order of magnitude as reported by

Gavish et al. (2008) in ambulatory patients.

Circadian variations in DBP and SBP, albeit small

(5 and 9 mm Hg, respectively), were statistically signifi-

cant and supported by the model-based analysis.

Blood pressure increased during the first half of the

night, before returning to baseline in the early morning,

thereby showing similar fluctuations to RA around the

clock. Renin has been shown to play a pivotal role in BP

regulation. In a study by Passo et al. (1971), increases in

renin secretion were associated with a substantial rise

in BP in 16 dogs. The main contribution of the RAAS

to BP regulation is mediated by the sodium-retaining

effects of aldosterone, and the powerful vasoconstrictor

effect of angiotensin II. The role of RAAS activation

in the development of hypertension in dogs has been

elucidated in a renal failure model by Mishina &

Watanabe (2008). In their study, RA, angiotensin II,

and aldosterone were significantly elevated in associ-

ation with a noticeable increase in BP, indicating that

the RAAS was involved in the development of nephron

loss-associated hypertension.

The nocturnal increase in BP observed in our experi-

ment supports earlier findings from Piccione et al.

(2005) in 10 healthy beagle dogs. These results are in

contradiction with investigations in human healthy

volunteers, where lower BP levels have been reported

during periods of darkness allocated to sleeping

(Fernandez et al., 2009). In humans, Smyth et al.

(1969) have shown that sleep is characterized by signs

of parasympathetic predominance, with a noticeable

reduction of heart rate and BP. According to Trinder

et al. (2001), sleep displays a strong ultradian rhythm

characterized by the regular occurrence of two funda-

mental states, alternating with a ca. 100-min period: the

non-rapid (NREM) and the rapid (REM) eye movement

sleep. NREM is associated with low BP and bradycardia,

whereas REM is characterized by a substantial increase

in heart rate and BP. Later investigations from Murali

et al. (2003) have shown that sympathetic activity was

increased to levels above awaking values during REM

sleep.

Although dogs have a natural tendency to be crepus-

cular, they can easily become diurnal through human

interactions. In essence, domesticated dogs are mainly

active during the day, while sleeping at night. Similar

to humans, REM sleep has been described in dogs

(Hendricks & Morrison, 1981; Hendricks et al., 1989).

In theory, nocturnal increases of BP could be a sign

of REM predominance in this species. Differences in

posture (dogs were monitored whilst freely moving

allowing for different body positions, and not main-

tained recumbent as in human studies), may also

explain part of the observed difference in BP dynamics

(Muller et al., 1958). In addition, the activation of

sodium-conserving mechanisms (i.e., elevated UA:C)

may have been a prerequisite to increase the levels of

SBP and DBP at night.

It has been argued that increased sodium excretion is

preceded by an elevation of BP (Coleman & Guyton,

1969; Hall et al., 1980). This mechanism, referred to as

‘‘pressure natriuresis,’’ would represent a powerful

means of stabilizing BP via reduction of the extracellular

fluid volume (Hall, 1986). Herein, an elevation of

UNa and UNa,fe during daytime has been reported in

the absence of increasing BP. Likewise, a reduction of

sodium elimination was observed at night in spite of an

increase in BP. These observations are entirely consist-

ent with those of other studies in dogs (Andersen et al.,

2000; Bie & Sandgaard, 2000; Sandgaard et al., 2000),

indicating that the RAAS controls sodium homeostasis

even when opposing changes in BP occur. According to

Bie & Damkjaer (2010), in order for the pressure

natriuresis mechanism to be involved, BP must change

in response to varying sodium intake. This has not been

reported in intact animals (Kjolby et al., 2005, 2008),

as shown by the nonsignificant effect of dietary sodium

on BP in our experiment.

Body weight and sodium intake were not found to

have an effect on BP model parameters. However,

the effect of sex on the chronobiology of BP could not

be assessed in Study b, since only male dogs were

included in the experiment.



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

Tuning in to the Body’s Rhythms to Adapt DrugDosing SchedulesDeeper understanding of circadian rhythms can have a

substantial impact on the therapeutic management of

RAAS-related diseases by determining the time of drug

administration that would optimize efficacy while

minimizing the occurrence of adverse effects. This

concept, referred to as chronotherapy, is currently

being used for the treatment of human rheumatoid

arthritis (Staessen et al., 1992), lung cancer (Mazzoccoli

et al., 2012), and cardiovascular diseases (Nicholls et al.,

1993). An increasing number of investigations on the use

of angiotensin-converting enzyme inhibitors (ACEIs) in

hypertension have shown a greater reduction of BP with

bedtime administration as compared with morning

dosing (Hermida & Ayala, 2009; Palatini et al., 1993).

Sole & Martino (2009) have demonstrated that heart and

vessel growth and remodeling were dynamic and

occurred more actively during the period normally

allocated to sleep. In mice, administrations of the

ACEI captopril at sleeping hours significantly improved

cardiovascular function and reduced adverse remodel-

ing, whereas no effects were reported when the drug was

given during active hours of the day (Martino et al.,

2011). In a study by Nozawa et al. (2006), temocapril

(another ACEI) prolonged the survival of spontaneously

hypertensive rats, with a maximum effect after dosing

during the resting period, and a minimum effect after

dosing at the active period. The authors concluded that

treatment with an ACEI at night may be a more effective

dosing regimen in patients with hypertension.

Another therapeutic approach in the management of

heart failure and hypertension is to continuously assess

not only the medical response, but also the development

of adverse effects. The optimal treatment time can vary

considerably between patients, as shown by the recent

work of Watanabe et al. (2013) in hypertensive patients

under losartan/hydrochlorothiazide (L/H) (angiotensin

II receptor blocker/thiazide diuretics) combination

therapy. In their study, L/H taken few hours before

bedtime in a 61-yr-old man induced circadian hyper-

amplitude tension (CHAT), a condition associated with

an increased cardiovascular disease risk. For yet another

patient, CHAT was exacerbated when L/H was given

during the day, but was alleviated when the same dose

of treatment was taken in the evening. In all instances,

optimization of therapy based on the most appropriate

time of drug administration should be investigated on

an individual basis (Cornelissen & Halberg, 2009).

SUMMARY AND PERSPECTIVES

This research offers the first chronobiological charac-

terization of the RAAS in relation to BP and renal

sodium-potassium handling in dogs. Our data demon-

strate that BP oscillates in parallel to the RAAS, which is

coherent with existing publications in dogs, but reflects

a clear (and not elucidated) difference from humans.

Cosine and surge models were able to describe and

predict the time-variant changes of the experimental

data with high accuracy, as shown by the quality of the

standard goodness-of-fit diagnostics.

The use of nonlinear mixed-effects allowed borrow-

ing information from the densely sampled plasma

values (RA) to improve knowledge on the time variations

of the urinary variables. The model-based approach

provided insights into the relation of dietary sodium,

sex, and body weight to renin and aldosterone secretion,

which would have been impossible using standard

statistical approaches. Specifically, sodium intake

appears to be an important determinant of RA chrono-

biology, and we are currently investigating the influence

of varying feeding time on the chronobiology of the

RAAS. This would support earlier findings from Itoh

et al. (1996) who showed that the effect of dietary

sodium on BP depends on the circadian timing of its

intake in humans.

Recent experiments in human patients have pointed

out the importance of administration time-dependent

effects of treatment in the management of RAAS-related

diseases. These results suggest that additional investi-

gations on the chronobiology of the RAAS and BP are

required in diseased dogs under ACEI to determine

whether it is possible to improve drug therapy of heart

failure or hypertension by selecting the appropriate time

of treatment.

ACKNOWLEDGMENTS

The investigations on RAAS periodicity were con-ducted at the Centre de Recherche Sante AnimaleSA (CRA) of Novartis Animal Health, located inSaint-Aubin, Switzerland. The telemetry study wasperformed at Novartis Institutes of BiomedicalResearch, Novartis Pharma AG, Basel, Switzerland.

DECLARATION OF INTEREST

With the exception of Pr. Meindert Danhof, theauthors of this paper are Novartis employees. Theexperiments were supported by Novartis AnimalHealth, Basel, Switzerland.

The authors report no conflicts of interest. Theauthors alone are responsible for the content andwriting of the paper.

REFERENCES

Allikmets K. (2007). Aliskiren—an orally active renin inhibitor.

review of pharmacology, pharmacodynamics, kinetics, and

clinical potential in the treatment of hypertension. Vasc

Health Risk Manag, 3, 809–15.

Andersen JL, Andersen LJ, Sandgaard NC, Bie P. (2000). Volume

expansion natriuresis during servo control of systemic blood

pressure in conscious dogs. Am J Physiol Regul Intergr Comp

Physiol, 278, R19–27.



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

Attarchi M, Dehghan F, Safakhah F, et al. (2012). Effect of exposure

to occupational noise and shift working on blood pressure

in rubber manufacturing company workers. Ind Health, 50,

205–13.

Baumgart P. (1991). Circadian rhythm of blood pressure: Internal

and external time triggers. Chronobiol Int, 8, 444–50.

Bie P, Damkjaer M. (2010). Renin secretion and total body sodium:

Pathways of integrative control. Clin Exp Pharmacol Physiol, 37,

e34–42.

Bie P, Sandgaard NC. (2000). Determinants of the natriuresis after

acute, slowsodium loading in conscious dogs. Am J Physiol

Regul Intergr Comp Physiol, 278, R1–10.

Blair-West JR, Brook AH. (1969). Circulatory changes and renin

secretion in sheep in response to feeding. J Physiol, 204, 15–30.

Brandenberger G, Follenius M, Goichot B, et al. (1994).

Twenty-four-hour profiles of plasma renin activity in relation

to the sleep-wake cycle. J Hypertens, 12, 277–83.

Brandenberger G, Follenius M, Muzet A, et al. (1985). Ultradian

oscillations in plasma renin activity: Their relationships to

meals and sleep stages. J Clin Endocrinol Metab, 61, 280–4.

Clarke LL, Ganjam VK, Fichtenbaum B, et al. (1988). Effect of

feeding on renin-angiotensin-aldosterone system of the horse.

Am J Physiol, 254, R524–30.

Clarke LL, Moore J, Garner H. (1978). Diurnal variations of plasma

aldosterone in horses: Dependence on feeding schedule

[abstract]. Physiologist 21, 21.

Coleman TG, Guyton AC. (1969). Hypertension caused by salt

loading in the dog. Onset transients of cardiac output and other

circulatory variables. Circ Res, 25, 153–60.

Corea M, Seeliger E, Boemke W, Reinhardt HW. (1996).

Diurnal pattern of sodium excretion in dogs with and without

chronically reduced renal perfusion pressure. Kidney Blood

Press Res, 19, 16–23.

Cornelissen G, Halberg F. (2009). Treatment with open eyes:

Markers-guided chronotheranostics. In Youan BC (ed.).

Chronopharmaceutics: science and technology for biological

rhythm-guided therapy and prevention of diseases. Hoboken,

NJ: Wiley, pp. 257–323.

Cugini P, Scavo D, Cornelissen G, et al. (1981). Circadian rhythms

of plasma renin, aldosterone and cortisol on habitual and low

dietary sodium intake. Horm Res, 15, 7–27.

Cugini P, Centanni M, Lucia P, et al. (1984). Parametric methods

applied to the standardization of gestational reference intervals

for blood renin activity, aldosterone and cortisol using the

transversal and longitudinal approach. Quad Sclavo Diagn, 20,

245–57.

Cugini P, Murano G, Lucia P, et al. (1985). Circadian rhythms

of plasma renin activity and aldosterone: Changes related to

age, sex, recumbency and sodium restriction. Chronobiologic

specification for reference values. Chronobiol Int, 2, 267–76.

Cugini P, Halberg F, Lucia P, et al. (1986). Role of the macula densa

and beta-adrenergic system in the control of the circadian

rhythm of renin and aldosterone. Recenti Prog Med, 77, 570–2.

Cugini P, Murano G, Lucia P, et al. (1987). Effects of a mild and

prolonged restriction in sodium or food intake on the circadian

rhythm of aldosterone and related variables. Chronobiol Int, 4,

245–50.

Fernandez JR, Hermida RC, Mojon A. (2009). Chronobiological

analysis techniques. Application to blood pressure. Philos Trans

A Math Phys Eng Sci, 28;367, 431–45.

Gardner SY, Atkins CE, Rausch WP, et al. (2007). Estimation of 24-h

aldosterone secretion in the dog using the urine aldosterone:

Creatinine ratio. J Vet Cardiol, 9, 1–7.

Gavish B, Ben-Dov IZ, Bursztyn M. (2008). Linear relationship

between systolic and diastolic blood pressure monitored over

24 h: Assessment and correlates. J Hypertens, 26, 199–209.

Gordon CR, Lavie P. (1985). Day-night variations in urine excre-

tions and hormones in dogs: Role of autonomic innervation.

Physiol Behav, 35, 175–81.

Guglielmini C. (2003). Cardiovascular diseases in the ageing dog:

Diagnostic and therapeutic problems. Vet Res Commun, 27,

555–60.

Guignard JP, Drukker A. (1999). Why do newborn infants have a

high plasma creatinine? Pediatrics, 103, e49.

Hall JE. (1986). Control of sodium excretion by angiotensin II:

Intrarenal mechanism and blood pressure regulation. Am J

Physiol, 250, R960–72.

Hall JE, Guyton AC, Smith Jr MJ, Coleman TG. (1980). Blood

pressure and renal function during chronic changes in sodium

intake: Role of angiotensin. Am J Physiol, 239, F271–80.

Hendricks JC, Morrison AR. (1981). Normal and abnormal sleep in

mammals. J Am Vet Med Assoc, 178, 121–6.

Hendricks JC, Lager A, O’Brien D, Morrison AR. (1989). Movement

disorders during sleep in cats and dogs. J Am Vet Med Assoc,

194, 686–9.

Hermida RC, Ayala DE. (2009). Chronotherapy with the angioten-

sin-converting enzyme inhibitor ramipril in essential hyperten-

sion: Improved blood pressure control with bedtime dosing.

Hypertension 54, 40–6.

Hoglund K, Hanas S, Carnabuci C, et al. (2012). Blood pressure,

heart rate, and urinary catecholamines in healthy dogs

subjected to different clinical settings. J Vet Intern Med, 26,

1300–8.

Ikonomov O, Stoynev AG, Shisheva A, Tarkolev N. (1981).

Circadian rhythms of plasma renin activity and the plasma

concentration of aldosterone and insulin in the rat: Effect

of renal denervation and food restriction. Agressologie 22,

247–53.

Itoh K, Kawasaki T, Cugini P. (1996). Effects of timing of salt

intake to 24-hour blood pressure and its circadian rhythm.

Ann N Y Acad Sci, 783, 324–5.

Kaczmarczyk G, Echt M, Mohnhaupt R, et al. (1980). Postprandial

volume regulation and renin-angiotensin system in conscious

dogs. Adv Physiol Sci, 11, 611–20.

Kawasaki T, Cugini P, Uezono K, et al. (1990). Circadian variations

of total renin, active renin, plasma renin activity and plasma

aldosterone in clinically healthy young subjects. Horm Metab

Res, 22, 636–639.

Kjolby MJ, Kompanowska-Jezierska E, Wamberg S, Bie P. (2005).

Effects of sodium intake on plasma potassium and renin

angiotensin aldosterone system in conscious dogs. Acta

Physiol Scand, 184, 225–34.

Kjolby M, Bie P. (2008). Chronic activation of plasma renin is log-

linearly related to dietary sodium and eliminates natriuresis in

response to a pulse change in total body sodium. Am J Physiol

Regul Intergr Comp Physiol, 294, R17–25.

Kunita H, Obara T, Komatsu T, et al. (1976). The effects of dietary

sodium on the diurnal activity of the renin-angiotensin-

aldosterone system and the excretion of urinary electrolytes.

J Clin Endocrinol Metab, 43, 756–9.

Laragh JH, Sealey JE. (2011). The plasma renin test reveals the

contribution of body sodium-volume content (V) and renin-

angiotensin (R) vasoconstriction to long-term blood pressure.

Am J Hypertens, 24, 1164–80.

Lefebvre HP, Dossin O, Trumel C, Braun JP. (2008). Fractional

excretion tests: A critical review of methods and applications

in domestic animals. Vet Clin Pathol, 37, 4–20.

Lindquist TL, Beilin LJ, Knuiman MW. (1997). Influence of lifestyle,

coping, and job stress on blood pressure in men and women.

Hypertension 29, 1–7.

Martino TA, Tata N, Simpson JA, et al. (2011). The primary

benefits of angiotensin-converting enzyme inhibition on car-

diac remodeling occur during sleep time in murine pressure

overload hypertrophy. J Am Coll Cardiol, 57, 2020–8.

Mazzoccoli G, Sothern RB, Parrella P, et al. (2012). Comparison

of circadian characteristics for cytotoxic lymphocyte subsets

in non-small cell lung cancer patients versus controls. Clin Exp

Med, 12, 181–94.



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

Mishina M, Watanabe T. (2008). Development of hypertension and

effects of benazepril hydrochloride in a canine remnant kidney

model of chronic renal failure. J Vet Med Sci, 70, 455–60.

Miyazaki H, Yoshida M, Samura K, et al. (2002). Ranges of diurnal

variation and the pattern of body temperature, blood pressure

and heart rate in laboratory beagle dogs. Exp Anim, 51, 95–8.

Morton DB, Hawkins P, Bevan R, et al. (2003). Refinements in

telemetry procedures. Seventh report of the BVAAWF/FRAME/

RSPCA/UFAW Joint Working Group on Refinement, Part A.

Lab Anim, 37, 261–99.

Muller AF, Manning EL, Riondel AM. (1958). Influence of position

and activity on the secretion of aldosterone. Lancet 1, 711–13.

Murali NS, Svatikova A, Somers VK. (2003). Cardiovascular physi-

ology and sleep. Front Biosci, 8, s636–52.

Nagaraja NV, Pechstein B, Erb K, et al. (2003). Pharmacokinetic/

pharmacodynamic modeling of luteinizing hormone (LH)

suppression and LH surge delay by cetrorelix after single and

multiple doses in healthy premenopausal women. J Clin

Pharmacol, 43, 243–51.

Nicholls MG, Richards AM, Crozier IG, et al. (1993). Cardiac

natriuretic peptides in heart failure. Ann Med, 25, 503–5.

Nozawa M, Sugimoto K, Ohmori M, et al. (2006). Dosing time-

dependent effect of temocapril on the mortality of stroke-prone

spontaneously hypertensive rats. J Pharmacol Exp Ther, 316,

176–81.

O’Connell JMB, Romeo JA, Mudge GH. (1962). Renal tubular

secretion of creatinine in the dog. Am J Physiol, 203, 985–90.

Ohmori M, Fujimura A. (2005). ACE inhibitors and chronotherapy.

Clin Exp Hypertens, 27, 179–85.

Palatini P, Mos L, Motolese M, et al. (1993). Effect of evening versus

morning benazepril on 24-hour blood pressure: A comparative

study with continuous intraarterial monitoring. Int J Clin

Pharmacol Ther Toxicol, 31, 295–300.

Passo SS, Assaykeen TA, Otsuka K, et al. (1971). Effect of

stimulation of the medulla oblongata on renin secretion in

dogs. Neuroendocrinology, 7, 1–10.

Pedersen HD, Koch J, Poulsen K, et al. (1995). Activation of the

renin-angiotensin system in dogs with asymptomatic and

mildly symptomatic mitral valvular insufficiency. J Vet Intern

Med, 9, 328–31.

Piccione G, Caola G, Refinetti R. (2005). Daily rhythms of blood

pressure, heart rate, and body temperature in fed and fasted

male dogs. J Vet Med A Physiol Pathol Clin Med, 52, 377–81.

Portaluppi F, Smolensky MH, Touitou Y. (2010). Ethics and

methods for biological rhythm research on animals and

human beings. Chronobiol Int, 27, 1911–29.

Reinhardt HW, Seeliger E, Lohmann K, et al. (1996). Changes of

blood pressure, sodium excretion and sodium balance due to

variations of the renin-angiotensin-aldosterone system. J Auton

Nerv Syst, 57, 184–7.

Rocha R, Porto M, Morelli MY, et al. (2002). Effect of environmental

stress on blood pressure during the work shift. Rev Saude

Publica, 36, 568–75.

Roche BM, Schwartz D, Lehnhard RA, et al. (2002). Changes in

concentrations of neuroendocrine hormones and catechol-

amines in dogs with myocardial failure induced by rapid

ventricular pacing. Am J Vet Res, 63, 1413–17.

Sandgaard NC, Andersen JL, Bie P. (2000). Hormonal regulation of

renal sodium and water excretion during normotensive sodium

loading in conscious dogs. Am J Physiol Regul Intergr Comp

Physiol, 278, R11–18.

Sayer MB, Atkins CE, Fujii Y, et al. (2009). Acute effect

of pimobendan and furosemide on the circulating renin-

angiotensin-aldosterone system in healthy dogs. J Vet Intern

Med, 23, 1003–6.

Schricker K, Holmer S, Hamann M, et al. (1994). Interrelation

between renin mRNA levels, renin secretion, and blood pres-

sure in two-kidney, one clip rats. Hypertension, 24, 157–62.

Seeliger E, Lohmann K, Nafz B, et al. (1999). Pressure-dependent

renin release: Effects of sodium intake and changes of total

body sodium. Am J Physiol, 277, R548–55.

Sheiner LB, Ludden TM. (1992). Population pharmacokinetics/

dynamics. Annu Rev Pharmacol Toxicol, 32, 185–209.

Smolensky MH, Haus E. (2001). Circadian rhythms and clinical

medicine with applications to hypertension. Am J Hypertens,

14, 280S–90S.

Smyth HS, Sleight P, Pickering GW. (1969). Reflex regulation of

arterial pressure during sleep in man. A quantitative method

of assessing baroreflex sensitivity. Circ Res, 24, 109–21.

Sole MJ, Martino TA. (2009). Diurnal physiology: Core principles

with application to the pathogenesis, diagnosis, prevention,

and treatment of myocardial hypertrophy and failure. J Appl

Physiol, 107, 1318–27.

Soloviev MV, Hamlin RL, Shellhammer LJ, et al. (2006). Variations

in hemodynamic parameters and ECG in healthy, conscious,

freely moving telemetrized beagle dogs. Cardiovasc Toxicol, 6,

51–62.

Spies CM, Cutolo M, Straub RH, et al. (2011). Prednisone chrono-

therapy. Clin Exp Rheumatol, 29, S42–5.

Staessen J, Guo C, De Cort P, et al. (1992). Mean and range of the

ambulatory pressure in normotensive subjects. Chin Med J

(Engl), 105, 328–33.

Thurau K. (1966). Influence of sodium concentration at macula

densa cells on tubular sodium load. Ann N Y Acad Sci, 139,

388–99.

Trinder J, Kleiman J, Carrington M, et al. (2001). Autonomic activity

during human sleep as a function of time and sleep stage.

J Sleep Res, 10, 253–64.

Turnbull F. (2003). Blood Pressure Lowering Treatment Trialists’

Collaboration. Effects of different blood-pressure-lowering

regimens on major cardiovascular events: Results of prospec-

tively-designed overviews of randomised trials. Lancet 362,

1527–35.

Uechi M, Terui H, Nakayama T, et al. (1994). Circadian variation

of urinary enzymes in the dog. J Vet Med Sci, 56, 849–54.

Van Orshoven NP, Oey PL, van Schelven LJ, et al. (2004).

Effect of gastric distension on cardiovascular parameters:

Gastrovascular reflex is attenuated in the elderly. J Physiol,

555, 573–83.

Watanabe Y, Halberg F, Otsuka K, Cornelissen G. (2013). Toward

a personalized chronotherapy of high blood pressure and a

circadian overswing. Clin. Exp. Hypertens. Advance online

publication. doi: 10.3109/10641963.2013.780073. Epub 2013

Mar 29.

Watkins Jr L, Burton JA, Haber E, et al. (1976). The renin-

angiotensin-aldosterone system in congestive failure in

conscious dogs. J Clin Invest, 57, 1606–17.



Chr

onob

iol I

nt D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y N

ovar

tis P

harm

a on

08/

09/1

3Fo

r pe

rson

al u

se o

nly.

Chronobiology of the renin-angiotensin-aldosterone system ... · blood flow and glomerular...

Documents

Transcript of Chronobiology of the renin-angiotensin-aldosterone system ... · blood flow and glomerular...