Am J Physiol Regul Integr Comp Physiol 290: R1429 –R1435, 2006.
First published December 22, 2005; doi:10.1152/ajpregu.00676.2005.
Are large amounts of sodium stored in an osmotically inactive form during
sodium retention? Balance studies in freely moving dogs
Erdmann Seeliger, Mechthild Ladwig, and H. Wolfgang Reinhardt
Institut für Physiologie, Charité Campus Mitte, Berlin, Germany
Submitted 19 September 2005; accepted in final form 20 December 2005
during Na⫹ retention, large portions of Na⫹ are usually stored
in an osmotically inactive form. This notion questions the
prevailing theory and was already noticed as an imminent
change of paradigm (18, 23).
To examine the relationship between changes in TBSodium
and those in TBWater, we performed balance studies of four
days duration in dogs. In a first set of experimental protocols,
we studied the effects of changes in Na⫹ intake, in analogy to
Heer’s and Titze’s studies (16, 41– 43). However, because the
effects of changes in Na⫹ intake on TBSodium are usually
very small in normal dogs and rats (17, 36, 41), other experimental approaches are necessary to induce more pronounced
changes in TBSodium. We wanted to study alterations in
TBSodium that cover the whole range from moderate deficit
(about ⫺10%) to large surplus [about ⫹50%; TBSodium
without bone Na⫹ is about 34 mmol/kg body mass (26)]. To
this end, we performed another set of experimental protocols,
and, in addition, we reanalyzed and included results of specific
protocols from previously published studies (10, 35, 37, 38).
As we had observed in previous studies (28, 35) that TBSodium changes are often accompanied by changes in TBPotassium, we routinely determined K⫹ balances.
MATERIALS AND METHODS
the osmolality of body fluids is
controlled within tight boundaries, as mechanisms of osmocontrol ensure that alterations in osmolality are effectively
adjusted by the excretion or intake of water. Thus the organism
maintains its content of isotonic fluid, that is, of total body
water (TBWater), by maintaining its content of osmolytes, in
particular, total body sodium (TBSodium) and total body
potassium (TBPotassium). According to this prevailing theory,
osmocontrol adjusts TBWater to the body’s present content of
the major cations, Na⫹ and K⫹.
The debate has attracted increasing attention since the publication of a series of papers by Titze et al. (41– 43), in which
a new concept of Na⫹ homeostasis was put forward. From a
balance study in humans performed by Heer et al. (16) and
their own studies in rats the authors (41– 43) concluded that,
A total of 81 chronically instrumented female Beagles, about 2
years of age, and weighing 12–16 kg, were studied by standardized
methods described in detail in previous papers (10, 24, 29, 30, 34, 36,
37). On completion of the experiments, implants were removed and
the dogs were given to suitable private individuals. The studies were
approved by the Berlin Government according to the German Animal
Protection Law.
The dogs were equipped with a urinary bladder catheter, an
inflatable aortic cuff above the renal arteries (used for servocontrolled
reduction of renal perfusion pressure in specific protocols), one
femoral vein catheter (used for infusions) and two femoral artery
catheters. The left femoral artery catheter was advanced into the
abdominal aorta well above the renal arteries (used for plasma
sampling), the tip of the right catheter was placed directly below the
renal arteries (used to obtain the pressure signal for servocontrol of
renal perfusion pressure). The lines were exteriorized in the nape
region. The dogs were allowed at least 3 weeks for recovery. Catheterrelated infections were prevented with a catheter-restricted antibioticlock technique (24). Daily assessments of general status, and daily
measurements of body temperature, weight, and erythrocyte sedimentation rate ensured that only healthy dogs were studied.
The dogs were housed individually in large kennels (9 m2) in a
sound protected, air-conditioned animal room. For reasons of social
well-being, dogs in adjacent kennels accompanied the dog under
investigation. During the 4-day balance study, the chronic lines of the
Address for reprint requests and other correspondence: E. Seeliger, Institut
für Physiologie, Charité Universitätsmedizin Berlin CCM, Tucholskystr. 2,
10117 Berlin, Germany (E-Mail: [email protected]).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
water-electrolyte balance; osmocontrol; total body sodium; total body
potassium; sodium-potassium-exchange
IT IS WIDELY ACCEPTED THAT
http://www.ajpregu.org
0363-6119/06 $8.00 Copyright © 2006 the American Physiological Society
R1429
Downloaded from http://ajpregu.physiology.org/ by 10.220.32.246 on July 31, 2017
Seeliger, Erdmann, Mechthild Ladwig, and H. Wolfgang Reinhardt. Are large amounts of sodium stored in an osmotically inactive
form during sodium retention? Balance studies in freely moving dogs.
Am J Physiol Regul Integr Comp Physiol 290: R1429 –R1435, 2006. First
published December 22, 2005; doi:10.1152/ajpregu.00676.2005.—Alterations in total body sodium (TBSodium) that covered the range from
moderate deficit to large surplus were induced by 10 experimental
protocols in 66 dogs to study whether large amounts of Na⫹ are stored
in an osmotically inactive form during Na⫹ retention. Changes in
TBSodium, total body potassium (TBPotassium), and total body water
(TBWater) were determined by 4-day balance studies. A rather close
correlation was found between individual changes in TBSodium and
those in TBWater (r2 ⫽ 0.83). Changes in TBSodium were often
accompanied by changes in TBPotassium. Taking changes of both
TBSodium and TBPotassium into account, the correlation with
TBWater changes became very close (r2 ⫽ 0.93). The sum of changes
in TBSodium and TBPotassium was accompanied by osmotically
adequate TBWater changes, and plasma osmolality remained unchanged. Calculations reveal that even moderate TBSodium changes
often included substantial Na⫹/K⫹ exchanges between extracellular
and cellular space. The results support the theory that osmocontrol
effectively adjusts TBWater to the body’s present content of the major
cations, Na⫹ and K⫹, and do not support the notion that, during Na⫹
retention, large portions of Na⫹ are stored in an osmotically inactive
form. Furthermore, the finding that TBSodium changes are often
accompanied by TBPotassium changes and also include Na⫹/K⫹
redistributions between fluid compartments suggests that cells may
serve as readily available Na⫹ store. This Na⫹ storage, however, is
osmotically active, since osmotical equilibration is achieved by opposite redistribution of K⫹.
R1430
OSMOTICALLY ACTIVE SODIUM-POTASSIUM BALANCING IN DOGS
Protocols
Dogs were randomly assigned to protocols. Two control protocols
without interventions served to obtain reference data for balance
calculation and plasma parameters: protocol CoLSI in dogs on LSI
(n ⫽ 7 dogs) and protocol CoHSI in dogs on HSI (n ⫽ 8).
In two protocols, the effects of steplike changes in Na⫹ intake were
studied. For protocol LtoHSI, dogs were on LSI during the first 2
study days, on day 3, Na⫹ intake was switched to HSI, which was also
fed on day 4 (n ⫽ 8), and for protocol HtoLSI, Na⫹ intake was
switched in the opposite direction, from HSI on the first 2 study days
to LSI on days 3 and 4 (n ⫽ 7).
In five protocols, different degrees of surplus in TBSodium
(⫹TBS) were induced, using endogenous stimulation of the reninangiotensin-aldosterone system via servocontrolled 20% reduction in
renal perfusion pressure (rRPP, for details, see Refs. 29 and 34) in
combination with different Na⫹ intake or low-dose infusion of aldosterone (Aldocorten, Ciba, 10 pg 䡠 min⫺1 䡠 kg⫺1) and ANG II (Hypertensin, Ciba, 4 ng 䡠 min⫺1 䡠 kg⫺1). The protocols are named according
to the average change in TBSodium achieved over the 4 days (mean
value in mmol Na⫹/kg bm, at the end of day 4). For protocol
⫹1.3TBS, rRPP was applied in dogs on LSI (n ⫽ 6); for protocol
⫹2.6TBS, rRPP was applied in dogs during a switch decrease in Na⫹
intake, as in protocol HtoLSI (n ⫽ 7); for protocol ⫹3.9TBS, rRPP
was applied in dogs on HSI (n ⫽ 7) (37); for protocol ⫹8.8TBS,
rRPP ⫹ aldosterone was applied in dogs on HSI (n ⫽ 7) (35); for
protocol ⫹13.1TBS, rRPP⫹aldosterone⫹angiotensin was applied in
dogs on HSI (n ⫽ 6) (35).
In two protocols, different degrees of deficit in TBSodium (-TBS)
were induced. For protocol ⫺0.9TBS, continuous ACE-inhibition
(ACE-I: Captopril, Bristol-Myers Squibb, New York, NY; 3.5
␮g 䡠 min⫺1 䡠 kg⫺1) was used to block endogenous renin-angiotensinaldosterone system stimulation by rRPP in dogs on HSI (n ⫽ 5), as
described by (10). For protocol ⫺3.4TBS, TBSodium was reduced by
peritoneal dialysis in the morning of the first study day in dogs on LSI
(n ⫽ 6), according to a previously described method (36). By
peritoneal dialysis, TBSodium was reduced by exactly 3.50 mmol
Na⫹ per kg bm, whereas TBWater and TBPotassium were not
changed by the procedure itself. After completion of the dialysis
(around 11:00 AM on the first study day), however, all three variables
were free to change according to endogenous control processes and,
on days 2– 4, the effects of ACE-I.
Finally, in protocol ⫾TBS, a large surplus of TBSodium was
induced by rRPP⫹aldosterone⫹angiotensin in dogs on HSI on day 1,
at the end of which, rRPP and the hormone infusions were stopped to
study whether normal TBSodium is regained within the remaining 3
days (n ⫽ 7) (38).
Balance Calculations and Statistics
Cumulative balances were used to determine the changes in
TBSodium, TBPotassium, and TBWater. Because of the 5 prestudy
days on the respective Na⫹, K⫹, and water intake, dogs are in balance
at the beginning of the 4-day study period. Dogs in the control
protocols (CoHSI, CoLSI) remain in balance; thus the data from their
4-day study provide the reference for balance calculation in the other
protocols. The amounts of extrarenal Na⫹, K⫹, and water loss are
known to depend on Na⫹ intake. Therefore, the extrarenal losses
assessed by protocol CoHSI provided the reference for dogs on HSI,
those assessed by protocol CoLSI provided the reference for dogs on
LSI. Extrarenal loss is calculated by the difference between 24-h
intake and 24-h urinary excretion in the control protocols. As room
temperature and moisture were kept constant, and the dogs’ body
temperature and physical activity did not change, extrarenal losses in
the other protocols are assumed to equal those observed in the
respective control protocol. Therefore, individual 24-h balances for
dogs studied in the other protocols were calculated as differences
between the excretion in the respective control protocol and the
individual dog’s excretion on the respective day. All fluid administered via arterial and venous lines was accounted for in balances. The
individual dog’s cumulative balance was calculated by summing up
the 24-h balance values over the 4 consecutive study days.
Daily measurements of the dogs’ body mass were done to ensure
that, even in protocols with massive changes in TBWater (e.g., surplus
in protocol ⫹13.1TBS, deficit in protocol ⫺3.4TBS), the accuracy of
water balance data was not compromised, as would be the case, if
such changes in TBWater would alter the amount of extrarenal water
loss, especially, evaporative loss with breathing (the major portion of
extrarenal water loss in dogs). Throughout the protocols, water balance data completely corresponded with body mass data.
Statistics were calculated using Number Cruncher Statistical Software (Hintze, Kaysville, UT). Differences between the control protocol, CoHSI or CoLSI, and the protocols with the respective Na⫹
intake were assessed by unpaired t-test with Bonferoni’s multiple
comparison adjustment (P ⬍ 0.05/m, where m is the number of
comparisons made). Data are presented as means ⫾ SE.
RESULTS
To determine whether changes in TBSodium are accompanied by osmotically adequate changes in TBWater, a correlation analysis (Fig. 1A) was done from data of all dogs (except
control protocols). The data were best fitted to a simple linear
regression. This overall analysis reveals a relatively close
correlation between individual changes of TBSodium and
those of TBWater: the coefficient of determination r2 indicates
that 83% of TBWater changes are attributable to changes of
TBSodium. However, scanning individual data, we found deviations accumulated in dogs studied in four protocols, namely
in protocols ⫹1.3TBS, ⫹8.8TBS, ⫺0.9TBS, and ⫺3.4TBS.
Accordingly, the coefficient of determination was 91%, when
the data of these protocols were not included.
Comparison between Fig. 1A and 1B reveals that the correlation of all dogs’ data becomes markedly closer, when
changes of both TBSodium and TBPotassium are taken into
account. Changes in TBPotassium were observed in the majority of protocols. In three protocols, including ⫹1.3TBS and
⫺3.4TBS, significant amounts of K⫹ were retained; in four
AJP-Regul Integr Comp Physiol • VOL
290 • MAY 2006 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ by 10.220.32.246 on July 31, 2017
accustomed dog were connected to a swivel system that allows free
movement within the kennel (10, 30). From the swivel, the lines were
led to the adjoining laboratory. All actions necessary to run the
experiment (except feeding) were performed from the laboratory
without drawing the attention of the dogs in the animal room.
Beginning 5 days before the experiment, food intake was controlled
with regard to amount [per kg body mass (bm)], composition (including the whole intake of Na⫹, K⫹, and water), feeding time (0830 –
0900), and completeness of intake as detailed by Seeliger et al. (37).
In all experiments, K⫹ intake was 3.5 mmol per kg bm per day, and
water intake was 100 ml per kg bm per day. Na⫹ intake varied with
protocols (see Protocols), it was either 0.5 mmol per kg bm per day
(low sodium intake, LSI), or 5.5 mmol per kg bm per day (high
sodium intake, HSI).
Urine was continuously collected throughout the 4 study days by
means of a computerized collection system (30). Urine volume was
measured gravimetrically. Blood samples were taken at the end of day
4 as described in detail by Reinhardt et al. (29). The blood withdrawn
was always replaced by an equal amount of stored blood collected
from the respective dog about 2 wk before the experiments. Concentrations of Na⫹ and K⫹ in urine and plasma ([Na⫹]pl, [K⫹]pl) were
determined by flame photometry, plasma osmolality ([Osmol]pl) by
freezing point depression.
OSMOTICALLY ACTIVE SODIUM-POTASSIUM BALANCING IN DOGS
R1431
Fig. 1. Changes of total body sodium (TBSodium) vs. changes of total body
water (TBWater) (A), the sum of changes of TBSodium and total body
potassium (TBPotassium) vs. changes of TBWater (B), the sum of changes of
TBSodium and TBPotassium vs. plasma osmolality (C; correlation not significant, r2 ⫽ 0.03). Data represent the individual dogs’ values at the end of 4
study days (n ⫽ 66 dogs studied in 10 protocols).
protocols, including ⫹8.8TBS and ⫺0.9TBS, significant
amounts of K⫹ were lost. Collectively, 93% of TBWater
changes are attributable to simultaneous changes in both
TBSodium and TBPotassium. Thus the sum of changes in
TBSodium and TBPotassium was accompanied by osmotically adequate TBWater changes, as also revealed by the
numerical values of the regression equation (Fig. 1B, inset; for
15 mmol of cations, the computed value is 109 ml of water). In
accordance, no significant correlation is found between [Osmol]pl and the sum of changes in TBSodium and TBPotassium
(Fig. 1C), and in none of the protocols did [Osmol]pl change
significantly (Table 2). Furthermore, in the collective correlation analysis between individual changes in TBSodium⫹
TBPotassium and those in TBWater (Fig. 1B), the absolute
term (the intercept) of the regression equation took on a
slightly positive value (not significantly different from zero); if
a considerable portion of these cations had been regularly
stored in an osmotically inactive form, the intercept would
have taken on a negative value.
Descriptive statistics (means ⫾ SE) for the protocols are
given in Fig. 2 and Tables 1 and 2. Please note that, in the
following, all data concerning volumes and amounts are given
per 1 kg of body mass. Step changes in Na⫹ intake (protocols
LtoHSI and HtoLSI) did not leave sustained changes in
TBSodium or TBWater after two days; however, a significant
K⫹ retention was observed in the HtoLSI protocol (Table 1).
Fig. 2. Changes of TBSodium (right-hatched bars), changes of TBPotassium
(left-hatched bars), and the sum of changes of TBSodium and TBPotassium
(cross-hatched bars), all related to the left ordinates; and changes of TBWater
(open bars), related to the right ordinates, in protocols ⫹1.3TBS (A), ⫹8.8TBS
(B), ⫺3.4TBS (C), and ⫺0.9TBS (D). Average changes (means ⫾ SE) at the
end of 4 study days. Please note the scaling of the ordinates: a 1.5-mmol
change in the content of cations (left ordinates) corresponds with a 10 ml
change in water content (right ordinates); thus the gray dashed lines indicate
approximate isotonicity of the retained or lost fluid (see RESULTS section). For
all changes: P ⬍ 0.05/m, where m is the number of comparisons vs. the
respective control protocol (CoHSI or CoLSI)
AJP-Regul Integr Comp Physiol • VOL
290 • MAY 2006 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ by 10.220.32.246 on July 31, 2017
In protocol ⫹1.3TBS, Na⫹ and water were retained. As
shown by Fig. 2A, however, the amount of retained Na⫹
(⌬TBSodium) is far too small compared with the retained fluid
volume (⌬TBWater) to give an isotonic fluid. As ⌬TBWater
represents a volume (the retained one), and ⌬TBSodium represents an amount of Na⫹, a virtual Na⫹ concentration of the
retained fluid can be calculated: [Na⫹]retained fluid ⫽ ⌬TBSodium/⌬TBWater ⫽ 1.33 mmol Na⫹/19.3 ml H2O ⫽ 69
mmol/l. This is only half normal [Na⫹]extracell. Adding the
same concentration of anions for electro-neutrality, the osmotic
concentration of the retained fluid would be about 50% of
isotonic fluid (⬃300 mosmol/kgH2O; isotonicity represented
by dotted lines in Fig. 2). Thus, judging from Na⫹ and water
retention only, about half of the retained water appears to be
osmotically free. However, dogs retained K⫹ alongside Na⫹.
Taking the retained amounts of both cations into account
(⌬TBSodium⫹⌬TBPotassium), isotonicity is almost exactly
achieved (see Fig. 2A). Calculated [K⫹]retained fluid ⫽ ⌬TBPotassium/⌬TBWater ⫽ 65 mmol/l is 16-fold higher than normal
[K⫹]extracell. Addition of this fluid with low [Na⫹] and very
R1432
OSMOTICALLY ACTIVE SODIUM-POTASSIUM BALANCING IN DOGS
Table 1. Changes of TBSodium, changes of TBPotassium,
the sum of changes of TBSodium and TBPotassium, and
changes of TBWater in protocols CoLSI, CoHSI, LtoHSI,
HtoLSI, ⫹2.6TBS, ⫹3.9TBS, ⫹13.1TBS, and ⫾TBS
Protocol
⌬TBSodium,
mmol/kg bm
⌬TBPotassium,
mmol/kg bm
⌬TBSodium ⫹
⌬TBPotassium,
mmol/kg bm
⌬TBWater,
ml/kg bm
CoLSI
CoHSI
LtoHSI
HtoLSI
⫹2.6TBS
⫹3.9TBS
⫹13.1TBS
⫾TBS
0.0⫾0.2
0.0⫾0.1
0.1⫾0.3
⫺0.3⫾0.2
2.6⫾0.6*
3.9⫾0.6*
13.1⫾1.8*
⫺0.5⫾0.5
0.0⫾0.1
0.0⫾0.2
0.0⫾0.1
0.7⫾0.2*
0.4⫾0.5
0.2⫾0.3
⫺1.3⫾0.6*
⫺0.6⫾0.4*
0.0⫾0.2
0.0⫾0.2
0.1⫾0.4
0.4⫾0.2*
3.0⫾0.9*
4.1⫾0.5*
11.8⫾2.3*
⫺1.1⫾0.8*
0⫾2
0⫾2
5⫾5
4⫾5
24⫾8*
33⫾4*
89⫾17*
⫺6⫾4*
high [K⫹] to normal extracellular fluid is expected to decrease
[Na⫹]extracell and to dramatically increase [K⫹]extracell. However, neither [Na⫹]pl nor [K⫹]pl changed significantly (Table
2). The most probable explanation for this finding is an almost
quantitative exchange of Na⫹ and K⫹ between extracellular
and cellular space, as revealed by calculations (see Appendix);
about 1.16 mmol of K⫹ left the extracellular space, entering
the cells, whereas about 1.27 mmol of Na⫹, coming from
cellular fluid, entered the extracellular space.
In protocol ⫹8.8TBS, the amount of retained Na⫹ is far too
large compared with the retained fluid volume to give an
isotonic fluid (Fig. 2B). Calculated [Na⫹]retained fluid ⫽
⌬TBSodium/⌬TBWater ⫽ 227 mmol/l is equal to about 150%
of normal [Na⫹]extracell. Thus a large portion of retained Na⫹
(the ⌬TBSodium portion above the line of isotonicity in Fig.
2B) appears to be osmotically inactive. However, dogs lost
large amounts of K⫹. Taking both cations into account, the
retained amount of Na⫹ and the lost amount of K⫹
(⌬TBSodium⫹⌬TBPotassium), isotonicity is almost exactly
achieved (Fig. 2B). In accordance, [Osmol]pl remained unchanged (Table 2). [K⫹]pl decreased dramatically (Table 2).
Most intriguingly, however, the lost amount of K⫹ (3.39
mmol) is more than fourfold larger than the amount of K⫹ that
is contained in the whole extracellular space (ECK⫹ ⫽ 0.8
mmol, assuming that normal [K⫹]extracell is 4 mmol/l and
normal extracellular fluid volume (ECFV) is 20% of body
mass, that is, ECFV ⫽ 0.200 l). Thus, even if all extracellular
K⫹ had been lost (resulting in [K⫹]extracell of 0 mmol/l), the
major portion of the lost K⫹ must have come from a source
other than extracellular space, most probably, from inside the
cells. Calculations revealed that an almost quantitative exchange of Na⫹ and K⫹ must also have occurred in this
protocol, this time, however, in the opposite direction: about
2.32 mmol of K⫹ left the cells, while about 2.55 mmol of Na⫹
entered the cells (see Appendix).
In protocol ⫺3.4TBS, the lost amount of Na⫹ is too large
compared with the lost fluid volume to give isotonic fluid (Fig.
2C). Calculated [Na⫹]lost fluid ⫽ ⌬TBSodium/⌬TBWater ⫽
203 mmol/l is equal to about 140% of normal [Na⫹]extracell, but
[Na⫹]pl did not decrease significantly (Table 2). Because of K⫹
retention, however, isotonicity is almost exactly achieved (Fig.
2C). K⫹ retention led to a marked increase in [K⫹]pl. However,
DISCUSSION
The present results corroborate the prevailing theory that
osmocontrol effectively adjusts TBWater to the body’s present
content of the major cations, Na⫹ and K⫹. Alterations in
TBSodium that cover the range from moderate deficit to large
surplus were induced by various methods, yet the sum of
simultaneous changes in TBSodium and TBPotassium was
accompanied by osmotically adequate changes in TBWater.
This result does not support the notion that, during Na⫹
retention, large portions of Na⫹ (up to 75%) are usually stored
in an osmotically inactive form (41– 43).
It has long been known that Na⫹ retention is often accompanied by K⫹ loss, and, vice versa, Na⫹ loss with K⫹ retention. This was observed with changes in Na⫹ intake (1, 9, 19,
27, 32) and with TBSodium changes induced by a variety of
interventions (3, 20, 28, 31, 37). Accordingly, opposite
changes in TBSodium and TBPotassium occurred in several of
the present protocols. However, parallel changes in TBSodium
and TBPotassium were also observed. In one protocol both
Na⫹ and K⫹ were retained, and in another, both Na⫹ and K⫹
were lost.
Furthermore, changes in TBSodium apparently often include
redistribution of substantial amounts of Na⫹ and K⫹ between
extracellular and cellular space, as indicated by the calculations
from results of four protocols, two that showed an increase in
TBSodium and two that showed a decrease in TBSodium. In
each case, the calculation revealed that an (almost) quantitaTable 2. Plasma Na⫹ concentration, plasma K⫹
concentration, and plasma osmolality
Protocol
[Na⫹]pl, mmol/l
[K⫹]pl, mmol/l
[Osmol]pl,
mosmol/kgH2O
CoLSI
CoHSI
LtoHSI
HtoLSI
⫹1.3TBS
⫹2.6TBS
⫹3.9TBS
⫹8.8TBS
⫹13.1TBS
⫺3.4TBS
⫺0.9TBS
⫾TBS
143.6⫾1.0
145.1⫾1.3
147.0⫾0.7
142.2⫾1.3
142.9⫾0.9
146.2⫾0.4†
144.6⫾0.9
147.6⫾0.9
148.6⫾1.5*
141.3⫾1.4
145.5⫾1.0
143.1⫾0.8
4.12⫾0.06
4.03⫾0.09
4.14⫾0.08
4.17⫾0.08
4.17⫾0.03
4.23⫾0.05
4.11⫾0.06
2.92⫾0.13*
3.03⫾0.15*
4.58⫾0.13†
4.11⫾0.04
4.09⫾0.08
301.8⫾0.7
302.3⫾1.3
303.8⫾0.5
302.5⫾0.9
301.3⫾0.6
302.3⫾1.1
301.0⫾1.3
303.1⫾0.5
304.4⫾1.8
300.1⫾1.4
301.1⫾0.9
303.2⫾0.8
Data are expressed as means ⫾ SE at the end of 4 study days; *Significant
vs. protocol CoHSI. †Significant vs. protocol CoLSI. P ⬍ 0.05/m, where m is
the number of comparisons.
AJP-Regul Integr Comp Physiol • VOL
290 • MAY 2006 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ by 10.220.32.246 on July 31, 2017
Average changes are expressed as means ⫾ SE per kilogram of body mass
(kg bm) at the end of 4 study days; *P ⬍ 0.05/m, where m is the number of
comparisons, vs. the respective control protocol (CoHSI or CoLSI).
the extent of [K⫹]pl elevation should have been much larger, if
all changes were limited to the extracellular space. Again, the
cellular space appears to be involved. The same applies for
protocol ⫺0.9TBS (see Fig. 2D), in which the deficit in
TBSodium was accompanied by loss of K⫹ (see Appendix for
both protocols).
In protocols ⫹2.6TBS and ⫹3.9TBS, moderate amounts of
Na⫹ and water were retained, while changes in TBPotassium
did not reach statistical significance (Table 1). In protocol
⫹13.1TBS, the largest amounts of Na⫹ and water were retained, and a significant amount of K⫹ was lost. In protocol
⫾TBS, normal TBSodium was restored, yet small deficits in
TBPotassium and TBWater were observed.
OSMOTICALLY ACTIVE SODIUM-POTASSIUM BALANCING IN DOGS
increase in Na⫹ intake immediately resulted in positive 24-h
Na⫹ balances for a couple of days; this Na⫹ retention was not
accompanied by osmotically adequate water retention. Titze et
al. (41– 43) compared groups of rats that had been offered food
of different Na⫹ content for several weeks and reported that
differences in TBSodium were not accompanied by adequate
differences in TBWater, as assessed by postmortem drying and
ashing procedures. Unfortunately, TBPotassium or its changes
were not assessed in these past studies (16, 41– 43). In a very
recent study in rats, however, Titze et al. (40) assessed TBPotassium in addition to TBSodium and TBWater, by drying and
ashing. In one protocol of this new study, they administered
DOCA and drinking fluid of 1% NaCl concentration for 5 wk
to induce a marked rise in TBSodium. This protocol has certain
experimental conditions in common with our present protocol
⫹8.8TBS: elevated mineralocorticoid levels and high Na⫹
intake. In the DOCA⫹NaCl rats, Titze et al. (40) found body
weight (wet and dry wt) reduced, TBWater/wet wt ratio increased, TBSodium increased, and TBPotassium decreased.
Quantitative estimations indicated that one portion of the
retained Na⫹ was osmotically active (accompanied by water),
and another portion was osmotically balanced by K⫹ loss. This
is in line with our protocol ⫹8.8TBS (see Fig. 2B). In contrast
to our results, however, Titze’s estimations indicated that a
third portion of the retained Na⫹ in the DOCA⫹NaCl rats was
osmotically inactive (neither accompanied by water, nor exchanged for K⫹). Possible reasons for this discrepancy include,
but are not limited to, study duration (4 days vs. 5 wk) and
species difference. It is possible that osmotically inactive Na⫹
storage does not occur within 4 days even during massive Na⫹
accumulation (as in protocol ⫹8.8TBS) but may require weeks
of TBSodium surplus to be induced. The data of Heer’s study
in humans (16) that hitherto appeared to demonstrate that
osmotically inactive Na⫹ storage is a rapid process, can no
longer be regarded as positive proof for this storage, because
K⫹ balances were not assessed. Furthermore, kinetics of Na⫹
homeostasis and the response to changes in Na⫹ intake vary
considerably among species (8, 16, 17, 25, 33, 41). Thus it is
also conceivable that osmotically inactive Na⫹ storage during
Na⫹ retention takes place in rats but not in dogs. In conclusion,
further studies are needed that address the time course of
TBSodium changes, involve different species, and must include measurements of TBPotassium or its changes.
Perspective
Pathways for transmembranal movements of Na⫹ and K⫹
are well known. The mechanism(s) that connect external Na⫹
balance with transmembranal Na⫹/K⫹ exchange, however, are
largely unknown, as was also noted for K⫹ redistribution
during primary TBPotassium changes (22). Are small changes
in Na⫹ concentration the trigger for this exchange? Is it
controlled by hormones? Considering the known effects of
aldosterone and ANG II on transmembranal cation traffic in
various nonrenal cells, it is also conceivable that Na⫹/K⫹
redistributions may rely on nonrenal effects of these hormones.
However, comparing protocols with Na⫹/K⫹ redistributions,
we found no clear-cut relationship between hormonal changes
and the direction of Na⫹/K⫹ redistributions. For instance,
aldosterone plasma concentrations are markedly elevated both
in protocol ⫹8.8TBS (35) and protocol ⫹1.3TBS (E. Seeliger
AJP-Regul Integr Comp Physiol • VOL
290 • MAY 2006 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ by 10.220.32.246 on July 31, 2017
tive, osmotically neutral Na⫹/K⫹ exchange between the fluid
compartments must have occurred. Because this redistribution
was observed even with moderate TBSodium changes and
occurred rather rapidly and because Na⫹ moved into cells in
two protocols, and out of cells in two others, we conclude that
cells may serve as a readily available Na⫹ store. This Na⫹
storage would be osmotically active, as osmotical equilibration
is achieved by opposite changes in cellular K⫹ content.
It is generally known that primary changes in TBPotassium
almost regularly include compartmental redistribution of K⫹
(7, 22); even the transient TBPotassium increase resulting from
a K⫹-rich meal is accompanied by rapid cellular K⫹ uptake.
Several early reports also support the concept that changes in
TBSodium often include Na⫹/K⫹ exchange between cells and
extracellular space; however, this concept, published decades
ago, appears largely forgotten. In 1968, Reinhardt and coworkers (3) found that large amounts of cellular Na⫹ had been
exchanged against extracellular K⫹ in a protocol comparable
to our ⫺3.4TBS protocol. Laragh (21) observed a similar
Na⫹/K⫹ exchange in hyponatremic patients following K⫹
administration. In their seminal 1952 study on “mineralocorticoid escape,” Relman and Schwartz (31) noted a Na⫹/K⫹
redistribution in the opposite direction. Several authors reported that, during administration of “K⫹-wasting” diuretics,
extracellular Na⫹ enters cells in an osmotically neutral exchange for cellular K⫹, that is, water does not follow Na⫹,
which substantially contributes to the associated hyponatremia
(4 – 6, 14).
A considerable amount of Na⫹ resides in the bones. A small
portion of bone Na⫹ belongs to the fluid phase that is in
equilibrium with extracellular fluid, (i.e., this Na⫹ is osmotically active), whereas the major portion belongs to the crystal
phase (i.e., this Na⫹ is osmotically inactive) (Refs. 11 and 15).
We cannot exclude the possibility that the latter portion contributed to the Na⫹/K⫹ exchanges observed. Under physiological conditions, the rate of exchange of crystal phase Na⫹, as
estimated from whole body isotope measurements, is low
[⬍1% of exchangeable TBSodium per day (12)]. Under conditions of prolonged severe hyponatremia and acidosis crystal
phase Na⫹ is lost, whereas short-term hyponatremia reduces
fluid phase Na⫹ only (15). Considering the time frame of our
experiments in conjunction with the finding that plasma Na⫹
concentration was largely unaltered, it thus appears not likely
that crystal phase Na⫹ accounted for a significant portion of
Na⫹/K⫹ exchanges. On the other hand, we cannot rule out
from the present results that crystal phase bone and other
osmotically inactive or neutral storage [e.g., cartilage glycosaminoglycans (13)] may play a role in long-term homeostasis.
Furthermore, our calculations of transmembranal Na⫹/K⫹
exchange allow global, but not tissue-specific conclusions. It is
conceivable that this exchange does not occur (to the same
extent) on all cells, but (preferentially) on cells of specific
tissues, since intracellular Na⫹ and K⫹ concentrations differ
considerably among tissues (26), and compartmental redistributions of K⫹ during primary TBPotassium changes occur
preferentially in skeletal muscles (7, 22). Clearly, studies that
include electrolyte measurements in various cells are needed to
test this notion.
Titze’s notion of osmotically inactive Na⫹ storage originated from a balance study in humans by Heer et al. (16) and
own studies in rats (41– 43). Heer et al. observed that an abrupt
R1433
R1434
OSMOTICALLY ACTIVE SODIUM-POTASSIUM BALANCING IN DOGS
and H. W. Reinhardt, unpublished observation); however, Na⫹
ions were shifted from extracellular space into cells in protocol
⫹8.8TBS but redistributed in the opposite direction in protocol
⫹1.3TBS. Considering that even moderate TBSodium changes
were accompanied by substantial transmembranal Na⫹/K⫹
exchange, which must have changed intracellular Na⫹ and K⫹
concentrations markedly, thereby possibly altering membrane
potentials of a variety of cells, these questions certainly warrant further investigations.
APPENDIX: EXAMPLE CALCULATIONS OF COMPARTMENTAL
Naⴙ/Kⴙ REDISTRIBUTIONS
ACKNOWLEDGMENTS
We thank K. Dannenberg, S. Molling, D. Bayerl, T. Rebeschke, A.
Bierwagen, and H. Kändler for expert technical assistance.
GRANTS
This work was supported by the Deutsche Forschungsgemeinschaft.
REFERENCES
1. Adler GK, Moore TJ, Hollenberg NK, and Williams GH. Changes in
adrenal responsiveness and potassium balance with shifts in sodium
intake. Endocr Res 13: 419 – 445, 1987.
2. Altman PL and Dittmer DS. Total body water: mammals other than man.
In: Biological Data Book, edited by Altman PL and Dittmer DS. Bethesda,
MD: Federation of Amercian Societies for Experimental Biology, 1974, p.
1989 –1992.
3. Behrenbeck D, Dörge A, and Reinhardt HW. Untersuchungen an
wachen Hunden über die Einstellung der Natriumbilanz III. Elektrolytbilanzen und Natriumrejection nach akutem Na-Entzug durch Peritonealdialyse oder wiederholter Mannitolinfusion. Pflügers Arch 300: 226 –243,
1968.
4. Bergström J. The effect of hydrochlorothiazide and amiloride administered together on muscle electrolytes in normal subjects. Acta Med Scand
197: 415– 419, 1975.
5. Bergström J, Alvestrand A, Fürst P, Hultman E, Sahlin K, Vinnars E,
and Widström A. Influence of severe potassium depletion and subsequent
repletion with potassium on muscle electrolytes, metabolites and amino
acids in man. Clin Sci Mol Med Suppl 51: 589 –599, 1976.
6. Bevegard S, Castenfors J, Danielson M, and Bergström J. Effect of
saluretic therapy on muscle content of water and electrolytes in relation to
hemodynamic variables. Acta Med Scand 202: 379 –384, 1977.
7. Bia MJ and DeFronzo RA. Extrarenal potassium homeostasis. Am J
Physiol Renal Fluid Electrolyte Physiol 240: F257–F268, 1981.
8. Boemke W, Palm Ü, Corea M, Seeliger E, and Reinhardt HW.
Endogenous variations and sodium intake-dependent components of diurnal sodium excretion patterns in dogs. J Physiol 476: 547–552, 1994.
9. Boemke W, Palm Ü, Kaczmarczyk K, and Reinhardt HW. Effect of
high sodium and high water intake on 24 h-potassium balance in dogs. Z
Versuchstierkd 33: 179 –185, 1990.
10. Boemke W, Seeliger E, Rothermund L, Corea M, Pettker R, Mollenhauer G, and Reinhardt HW. ACE inhibition prevents Na and water
retention and MABP increase during reduction of renal perfusion pressure.
Am J Physiol Regul Integr Comp Physiol 269: R481–R489, 1995.
11. Edelman IS, James A, Baden HA, and Moore F. Electrolyte composition of bone and the penetration of radiosodium and deuterium oxide into
dog and human bone. J Clin Invest 33: 122–131, 1954.
AJP-Regul Integr Comp Physiol • VOL
290 • MAY 2006 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ by 10.220.32.246 on July 31, 2017
For calculation, the mean values obtained at the end of day 4 in
protocol ⫹1.3TBS are used with the index “4”, and those of the
control protocol (CoLSI) with the index “Co”. For simplicity, plasma
concentrations of Na⫹ and K⫹ are equated to extracellular concentrations (ignoring Gibbs-Donnan distribution). All data concerning
volumes and amounts are given per 1 kg bm. Because normal ECFV
is ⬃20% of body mass in dogs (27, 39), ECFVCo is assumed to be
0.200 l/kg bm. Thus control extracellular amount of Na⫹
(ECNa⫹Co) ⫽ ECFVCo 䡠 [Na⫹]plCo ⫽ 0.200 l 䡠 143.6 mmol/l ⫽ 28.7
mmol, and the control extracellular amount of K⫹ is
ECK⫹Co ⫽ ECFVCo 䡠 [K⫹]plCo ⫽ 0.200 l 䡠 4.12 mmol/l ⫽ 0.82
mmol.
If we assume that all changes in the ⫹1.3TBS protocol occur
within and are limited to extracellular space, then, after 4 study days,
ECFV4 ⫽ ECFVCo ⫹ ⌬TBWater ⫽ 0.200 liter ⫹ 0.0193 liter ⫽
0.2193 liter, ECNa⫹4 ⫽ ECNa⫹Co ⫹ ⌬TBSodium ⫽ 28.7 mmol ⫹
1.33 mmol ⫽ 30.0 mmol, and ECK⫹4 ⫽ ECK⫹Co ⫹ ⌬TBPotassium ⫽ 0.82 mmol ⫹ 1.25 mmol ⫽ 2.07 mmol.
Extracellular ⫽ plasma [Na⫹] and [K⫹] after 4 study days are
calculated: calculated [Na⫹]pl4 ⫽ ECNa⫹4 / ECFV4 ⫽ 30.0 mmol/
0.2193 liter ⫽ 137.1 mmol/liter. Thus, compared with control,
[Na⫹]pl should have decreased by ⌬[Na⫹]pl ⫽ ⫺6.5 mmol/l.
Calculated [K⫹]pl4 ⫽ ECK⫹4 / ECFV4 ⫽ 2.07 mmol/0.2193 l ⫽
9.46 mmol/l.
Thus, compared with control, [K⫹]pl should have dramatically
increased by ⌬[K⫹]pl ⫽ 5.34 mmol/l.
However, the measurements in the dogs of protocol ⫹1.3TBS
reveal that neither [Na⫹]pl nor [K⫹]pl was significantly changed; there
were only small numerical deviations in the mean values from those
of CoLSI (see Table 2): measured ⌬[Na⫹]pl ⫽ ⫺0.7 mmol/l; measured ⌬[K⫹]pl ⫽ 0.05 mmol/l. Taking these small changes into
account, there remain large differences between the calculated and the
measured concentration changes. Compared with calculation, actual
[Na⫹]pl had seemingly increased by 5.8 mmol/l, whereas actual
[K⫹]pl had seemingly decreased by 5.29 mmol/l.
The assumption, that all changes were limited to the extracellular
space, is obviously invalid, as the most probable explanations rely on
the involvement of the intracellular space.
Normal [K⫹]extracell could have been maintained either by net K⫹
flux from extracellular space into cells, or by expansion of ECFV (net
water flux from cells). The extracellular K⫹ efflux necessary to
maintain [K⫹]extracell is calculated by Necessary ⌬ECK⫹ ⫽
ECFV4 䡠 seeming decrease in [K⫹]pl ⫽ ⫺1.16 mmol. The extracellular
water influx necessary to maintain [K⫹]extracell is calculated by Necessary change in the amount of extracellular water (⌬ECH2O) ⫽
⫺1 䡠 ECFV4 䡠 seeming decrease in [K⫹]pl/[K⫹]pl4 ⫽ 0.2870 liters.
Likewise, normal [Na⫹]extracell could have been maintained either
by net Na⫹ flux from cells into extracellular space: Necessary
⌬ECNa⫹ ⫽ ECFV4 䡠 seeming increase in [Na⫹]pl ⫽ 1.27 mmol, or by
a small contraction of ECFV (net water flux into cells): Necessary
⌬ECH2O ⫽ ⫺1 䡠 ECFV4 䡠 seeming increase in [Na⫹]pl/[Na⫹]pl4 ⫽
⫺0.0089 liters. As the extent of ECFV expansion (⫹140%) that
would be necessary to maintain [K⫹]extracell is not plausible and,
moreover, is contrary to the small ECFV contraction (⫺5%) necessary
to maintain [Na⫹]extracell, there remains only one explanation: an
(almost) quantitative exchange of Na⫹ and K⫹ took place. About 1.16
mmol of K⫹ left the extracellular space, entering the cells, while about
1.27 mmol of Na⫹, coming from cellular fluid, entered the extracellular space.
For protocol ⫹8.8TBS, the calculation leads to similar conclusions.
Here, normal [Na⫹]extracell could have been maintained by expansion
of ECFV (⌬ECH2O ⫽ 0.017 liter), or by Na⫹ flux into cells
(⌬ECNa⫹ ⫽ ⫺2.55 mmol); normal [K⫹]extracell by ECFV contraction
(⌬ECH2O ⫽ ⫺0.794 liter), or by K⫹ flux into extracellular space
(⌬ECK⫹ ⫽ 2.32 mmol). The extent of ECFV contraction necessary to
maintain [K⫹]extracell exceeds the sum of cellular and extracellular
water (⬃0.600 liter; Ref. 2), and the amount of K⫹ lost exceeds
ECK⫹ fourfold. As explanation remains an (almost) quantitative
exchange of Na⫹ and K⫹, this time, however, in the opposite
direction. About 2.32 mmol of K⫹ left the cells, while about 2.55
mmol of Na⫹ entered the cells.
In protocol ⫺3.4TBS, the changes in ECFV necessary to simultaneously maintain both [Na⫹] and [K⫹] contradict each other. Thus,
⬃0.63 mmol of K⫹ must have entered the cells, whereas about 0.89
mmol of Na⫹ left the cells. In protocol ⫺0.9TBS, the exchange went
in the opposite direction: about 0.62 mmol of K⫹ must have left the
cells, while about 0.76 mmol of Na⫹ entered the cells.
OSMOTICALLY ACTIVE SODIUM-POTASSIUM BALANCING IN DOGS
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
AJP-Regul Integr Comp Physiol • VOL
water balance during 4 days of servo-controlled reduction of renal perfusion pressure. Am J Physiol Heart Circ Physiol 266: H650 –H657, 1994.
Reinhardt HW, Palm Ü, Mohnhaupt R, Dannenberg K, and Boemke
W. Computer-assisted long-term measurements of urinary output and
other biological data. Am J Physiol Regul Integr Comp Physiol 258:
R274 –R280, 1990.
Relman AS and Schwartz WB. The effect of DOCA on electrolyte
balance in normal man and its relation to sodium chloride intake. Yale
J Biol Med 24: 540 –558, 1952.
Rogacz S, Hollenberg NK, and Williams GH. Role of angiotensin II in
the hormonal, renal, and electrolyte response to sodium restriction. Hypertension 9: 289 –294, 1987.
Sagnella GA, Markandu ND, Singer DR, and MacGregor GA. Kinetics of renal sodium excretion during changes in dietary sodium intake in
man–an exponential process? Clin Exp Hypertens A 12: 171–178, 1990.
Seeliger E, Andersen JL, Bie P, and Reinhardt HW. Elevated renal
perfusion pressure does not contribute to natriuresis induced by isotonic
saline infusion in freely moving dogs. J Physiol 559: 939 –951, 2004.
Seeliger E, Boemke W, Corea M, Encke T, and Reinhardt HW.
Mechanisms compensating Na and water retention induced by long-term
reduction of renal perfusion pressure. Am J Physiol Regul Integr Comp
Physiol 273: R646 –R654, 1997.
Seeliger E, Lohmann K, Nafz B, Persson PB, and Reinhardt HW.
Pressure-dependent renin release: effects of sodium intake and changes of
total body sodium. Am J Physiol Regul Integr Comp Physiol 277: R548 –
R555, 1999.
Seeliger E, Persson PB, Boemke W, Mollenhauer G, Nafz B, and
Reinhardt HW. Low-dose nitric oxide inhibition produces a negative
sodium balance in conscious dogs. J Am Soc Nephrol 12: 1128 –1136,
2001.
Seeliger E, Safak E, Persson PB, and Reinhardt HW. Contribution of
pressure natriuresis to control of total body sodium: balance studies in
freely moving dogs. J Physiol 537: 941–947, 2001.
Swan R, Madisso A, and Pitts RF. Measurement of extracellular fluid
volume in nephrectomized dogs. J Clin Invest 33: 1447–1456, 1954.
Titze J, Bauer K, Schafflhuber M, Dietsch P, Lang R, Schwind KH,
Luft FC, Eckardt KU, and Hilgers KF. Internal sodium balance in
DOCA-salt rats: a body composition study. Am J Physiol Renal Physiol
289: F793–F802, 2005.
Titze J, Krause H, Hecht H, Dietsch P, Rittweger J, Lang R, Kirsch
KA, and Hilgers KF. Reduced osmotically inactive Na storage capacity
and hypertension in the Dahl model. Am J Physiol Renal Physiol 283:
F134 –F141, 2002.
Titze J, Lang R, Ilies C, Schwind KH, Kirsch KA, Dietsch P, Luft FC,
and Hilgers KF. Osmotically inactive skin Na⫹ storage in rats. Am J
Physiol Renal Physiol 285: F1108 –F1117, 2003.
Titze J, Shakibaei M, Schafflhuber M, Schulze-Tanzil G, Porst M,
Schwind KH, Dietsch P, and Hilgers KF. Glycosaminoglycan polymerization may enable osmotically inactive Na⫹ storage in the skin. Am J
Physiol Heart Circ Physiol 287: H203–H208, 2004.
290 • MAY 2006 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ by 10.220.32.246 on July 31, 2017
12. Edelman IS and Leibman J. Anatomy of body water and electrolytes.
Am J Med 27: 256 –277, 1959.
13. Farber SJ. Mucopolysaccharides and sodium metabolism. Circulation
21: 941–947, 1960.
14. Fichman MP, Vorherr H, Kleeman CR, and Telfer N. Diuretic-induced
hyponatremia. Ann Intern Med 75: 853– 863, 1971.
15. Forbes GB, Tobin RB, Harrison A, and McCoord A. Effect of acute
hypernatremia, hyponatremia, and acidosis on bone sodium. Am J Physiol
209: 825– 829, 1965.
16. Heer M, Baisch F, Kropp J, Gerzer R, and Drummer C. High dietary
sodium chloride consumption may not induce body fluid retention in
humans. Am J Physiol Renal Physiol 278: F585–F595, 2000.
17. Holtzman EJ, Braley LM, Williams GH, and Hollenberg NK. Kinetics
of sodium homeostasis in rats: rapid excretion and equilibration rates.
Am J Physiol Regul Integr Comp Physiol 254: R1001–R1006, 1988.
18. Humphreys MH. Salt intake and body fluid volumes: have we learned all
there is to know? Am J Kidney Dis 37: 648 – 652, 2001.
19. Krieger JE, Liard JF, and Cowley AW Jr. Hemodynamics, fluid
volume, and hormonal responses to chronic high-salt intake in dogs. Am J
Physiol Heart Circ Physiol 259: H1629 –H1636, 1990.
20. Krieger JE, Roman RJ, and Cowley AW Jr. Hemodynamics and blood
volume in angiotensin II salt-dependent hypertension in dogs. Am J
Physiol Heart Circ Physiol 257: H1402–H1412, 1989.
21. Laragh J. The effect of potassium chloride on hyponatremia. J Clin Invest
33: 807– 818, 1954.
22. McDonough AA, Thompson CB, and Youn JH. Skeletal muscle regulates extracellular potassium. Am J Physiol Renal Physiol 282: F967–
F974, 2002.
23. Nguyen MK and Kurtz I. Are the total exchangeable sodium, total
exchangeable potassium and total body water the only determinants of the
plasma water sodium concentration? Nephrol Dial Transplant 18: 1266 –
1271, 2003.
24. Palm Ü, Boemke W, Bayerl D, Schnoy N, Juhr NC, and Reinhardt
HW. Prevention of catheter-related infections by a new, catheter-restricted
antibiotic filling technique. Lab Anim 25: 142–152, 1991.
25. Palm Ü, Boemke W, and Reinhardt HW. Rhythmicity of urinary
sodium excretion, mean arterial blood pressure, and heart rate in conscious
dogs. Am J Physiol Heart Circ Physiol 262: H149 –H156, 1992.
26. Pitts RF. Physiology of the kidney and body fluids. Chicago, IL: Year
Book Medical Publisher, 1974, p. 11–34.
27. Reinhardt HW and Behrenbeck D. Untersuchungen an wachen Hunden
über die Einstellung der Natriumbilanz I. Die Bedeutung des Extracellulärraumes für die Einstellung der Natrium-Tagesbilanz. Pflügers Arch 295:
266 –279, 1967.
28. Reinhardt HW, Boemke W, Palm Ü, and Kaczmarczyk G. What
causes escape from sodium retaining hormones? Acta Physiol Scand Suppl
591: 12–17, 1990.
29. Reinhardt HW, Corea M, Boemke W, Pettker R, Rothermund L,
Scholz A, Schwietzer G, and Persson PB. Resetting of 24-h sodium and
R1435