Creatine kinase and renal sodium excretion in African and European men on a high sodium diet

Abstract

Creatine kinase (CK) rapidly regenerates ATP for Na⁺/K⁺‐ATPase driven sodium retention throughout the kidney. Therefore, we assessed whether resting plasma CK is associated with sodium retention after a high sodium diet. Sixty healthy men (29 European and 31 African ancestry) with a mean age of 37.2 years (SE 1.2) were assigned to low sodium intake (< 50 mmol/d) during 7 days, followed by 3 days of high sodium intake (> 200 mmol/d). Sodium excretion (mmol/24‐h) after high sodium was 260.4 (28.3) in the high CK tertile versus 415.2 (26.3) mmol/24‐h in the low CK tertile (P < .001), with a decrease in urinary sodium excretion of 98.4 mmol/24‐h for each increase in log CK, adjusted for age and African ancestry. These preliminary results are in line with the energy buffering function of the CK system, but more direct assessments of kidney CK will be needed to further establish whether this enzyme enhances sodium sensitivity.

1. INTRODUCTION

Sodium plays a major role in the regulation of blood pressure and there is ample evidence linking higher sodium intake with increased blood pressure levels and associated cardiovascular risk.1, 2, 3 There is a wide interindividual variation in the response to sodium intake, and some people display a slower urinary excretion of sodium after a standard sodium load.4, 5 However, the pathophysiological mechanisms leading to interindividual differences in sodium excretion have not been well defined.1, 2, 3, 4, 5

We propose that these interindividual differences in renal sodium handling may be related to the variation in activity of the enzyme creatine kinase (CK) observed in humans.6, 7, 8, 9 CK catalyses the rapid transfer of a phosphoryl group from creatine phosphate to ADP, thereby forming creatine and ATP:6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18

$$\text{Creatine phosphate}+\text{MgADP}\leftrightarrow \text{Creatine}+\text{MgATP}.$$

CK connects sites of ATP production (glycolysis and mitochondrial oxidative phosphorylation) with subcellular sites of ATP utilization, including myosin ATPase and myosin light chain kinase at the contractile proteins, and Ca²⁺‐ATPase and Na⁺/K⁺‐ATPase at cellular membranes, where it rapidly regenerates ATP in situ from phosphocreatine.6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 These ATPases preferably use ATP regenerated by CK.6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 Thus, CK is thought to enhance ATP buffer capacity for cardiovascular contractility and renal sodium retention and increase hypertension risk.6, 7, 17, 18

Healthy tissue continuously “leaks” CK to the systemic circulation, and resting plasma CK activity is thought to proportionally represent tissue activity of the enzyme.7, 19, 20 In line with this, plasma CK was found to be a main predictor of blood pressure in the general population, independent of age, sex, BMI, or ethnicity, with an increase in systolic and diastolic blood pressure of respectively 14 and 8 mm Hg per log CK increase.7 Subsequently, contractility of human resistance arteries was shown to be CK‐dependent.17 In addition, CK expression in human resistance arteries was highly correlated with blood pressure.18 However, to our knowledge the association between creatine kinase and sodium excretion in humans has not previously been studied.

Sodium reabsorption throughout the kidney is energetically driven by basolateral Na⁺/K⁺‐ATPase, which couples hydrolysis of ATP to the active exchange of 3 intracellular Na⁺ ions for 2 K⁺ ions.21, 22 CK is tightly bound near this enzyme, where it rapidly regenerates ATP for tubular sodium reabsorption.11, 12, 13, 14 High CK activity may thus enhance ATP buffer capacity for this highly energy‐demanding process (Figure 1).6, 11, 12, 13, 14, 15, 16, 21, 22 Therefore, we hypothesized that resting plasma CK activity is inversely related to sodium excretion, and assessed predictors of urinary sodium excretion after a high sodium diet.

Figure 1.

Creatine kinase drives molecular motors of renal sodium reabsorption. Main mechanisms of Na⁺ transport in the proximal tubule, thick ascending limb (TAL), distal tubule, and collecting duct, modified from Greger et al and others.11, 12, 13, 14, 21, 22 In all parts of the nephron, basolateral Na⁺/K⁺‐ATPase is the primary force for the vectorial transport of Na⁺ from the tubular lumen to the blood compartment, by coupling hydrolysis of ATP to the active exchange of 3 intracellular Na⁺ ions for 2 K⁺ ions.21, 22 Ample evidence indicates that creatine kinase (CK) is functionally coupled to renal Na⁺/K⁺‐ATPase and that ATP produced by co‐localized CK is preferentially used for the high and fluctuating ATP demand of Na⁺ transport across the tubular epithelial cells.11, 12, 13, 14 Thus, we proposed that differential CK activity might affect sodium handling.6, 7 Apical and Basolateral, apical and basolateral cellular membrane; CA, carbonic anhydrase; CKmi, cyt, mitochondrial and cytoplasmic creatine kinase; Cr(P), creatine (phosphate); ENaC, epithelial Na⁺ channel; Matrix, mitochondrial matrix; IMS, mitochondrial intermembrane space; NHE3, Na⁺/H⁺ exchanger; OXPHOS, oxidative phosphorylation

2. METHODS

2.1. Participants and protocol

Our local institutional review board approved the protocol (NL 32565.018.10). All of the procedures were in accordance with institutional and international guidelines. All participants gave written informed consent to participate in the study. As men have higher CK activity and greater contrasts in plasma and tissue CK based on ethnicity,6, 7, 8, 9, 20 we included healthy men, 18 to 50 years old, normotensive or with uncomplicated untreated primary hypertension, of self‐reported European or African ancestry. We included untreated hypertension as we previously showed that hypertensives are at the high end of the CK spectrum.23 However, treated and controlled blood pressure is associated with lower CK.6, 7, 23 Participants with treated or secondary hypertension, glucose, lipid spectrum, thyroid, kidney, or liver abnormalities, cardiovascular, neuromuscular, or endocrine disorders, vasculitis, HIV infection, infectious hepatitis, or using CK‐increasing drugs including statins were excluded. Participants were instructed to abstain from heavy exercise 3 days before the baseline visit to our hospital, to obtain a resting plasma CK activity.7 We included a 7‐day low‐sodium period to “wash out” previous sodium intake and standardize all participants to the same level of sodium intake24, 25 before the high sodium intervention, which was the main study intervention. After the baseline visit, participants were instructed to adhere to a low sodium diet (LS; < 50 mmol Na⁺ per day) during 7 days followed by a high sodium diet (HS; > 200 mmol Na⁺ per day) during 3 days (day 8 to 10). During HS, a minimal daily amount of sodium (200 mmol) with sodium cubes containing 1 gram of salt (NaCl) each was provided to the participants by the research physician. A dietician was consulted before start of the protocol. During the study, participants were supported daily by the research physician. At baseline on day 1 and day 4 of LS, overnight urine was sampled to monitor dietary compliance. On day 7 (LS) and day 10 (HS), all participants collected 24‐h urine; and after an overnight fast, body weight and resting/sitting blood pressure were measured and blood was sampled.

2.2. Study procedures

Physical examination included height, weight, and blood pressure levels. Office blood pressure was measured with an Omron M4 oscillometric device (Omron Healthcare Europe BV, Hoofddorp, Netherlands), in a quiet room with the subject seated (following a 5‐min rest period). An appropriately adjusted cuff size was used on the non‐dominant arm, supported at heart level. Blood pressure was measured to a maximum of 6 times. The first measurement was discarded. Mean blood pressure was calculated as the mean of 2 consecutive readings with a maximum of 5 mm Hg, or with the smallest difference after a maximum of 6 measurements.7 Body mass index (BMI) was calculated as weight (in kg) divided by the height (in m rounded to the nearest cm) squared. We assessed plasma CK after 3 days of rest as a proxy for tissue CK activity as previously reported.7 Furthermore, we assessed plasma and 24‐h urine creatinine, sodium, potassium, and urea; plasma thyroid‐stimulating hormone (TSH), fasting glucose, and fasting lipids (total cholesterol, low‐density lipoprotein, high‐density lipoprotein, and triglycerides). Except for TSH, all laboratory analyses were performed on a Modular Cobas 8000 (Roche Diagnostics, Darmstadt, Germany). Plasma CK, glucose, total cholesterol, and triglycerides were estimated by enzymatic spectrophotometric; high‐density lipoprotein‐cholesterol by colorimetric/spectrophotometric; plasma creatinine and urea by kinetic spectrophotometric; and sodium and potassium by indirect ion‐selective electrode methods. TSH was assessed with an Access Immunoanalyzer (Beckman Coulter Inc., Fullerton, CA, USA).

2.3. Data analysis and statistics

2.3.1. Outcomes

The primary outcome was the association between CK and urinary sodium excretion after a high sodium diet. We studied baseline, LS, and HS variables for body weight, sitting blood pressure, heart rate, and 24‐h sodium excretion. We used 24‐h creatinine excretion as a measure of the accuracy of 24‐h urine sampling. We assessed correlates of 24‐h sodium excretion after a HS diet, such as age, ethnicity, BMI, and CK before entering relevant variables into regression analysis.

2.3.2. Sample size calculation

We expected that persons with high baseline CK excrete less sodium on HS than those with low baseline CK. Based on previously reported differences in sodium retention of 10‐50% between sodium sensitive persons and controls,24, 26 we calculated to need 40‐60 people to detect an independent association between CK and urinary sodium on HS, with an anticipated effect size between 0.2 and 0.3, 3 to 5 predictors, a 1‐sided alpha of 0.05, and a 1‐beta of 0.80.

2.3.3. Statistical analyses

Because plasma CK distribution was reported to be extremely skewed to the right, we planned to remove outliers using Dixon statistics and perform a logarithmic transformation to the base of 10 to achieve a more symmetrical distribution.7, 9 Unpaired and paired t‐tests were used respectively for between‐group (low versus high CK tertile, European versus African ancestry) and within‐group (LS versus HS) comparisons, respectively. To assess associations, 1‐tailed Pearson product‐moment correlation coefficients (r) were calculated for 24‐h urinary sodium excretion versus potential predictors of this variable, including log plasma CK, ethnicity, age, and BMI. Significant predictors (at P < .05) were entered into the regression analysis using forced entry. As CK was expected to be highly correlated with ancestry,6, 7, 8, 9, 10, 20 we planned to use projection to latent structures (PLS) regression analysis suited for collinear variables to assess the predictors of sodium excretion. We considered a 1‐sided probability value of < 0.05 to be significant when the direction of the outcome was hypothesized or known, otherwise we used a 2‐sided test. We did not impute missing data. Data were analyzed with SPSS statistical software package for Windows, Version 22.0 (SPSS Inc., Chicaco, IL, USA) and XLSTAT 2017 (Addinsoft, New York, NY, USA). Data are presented as mean with the standard error unless stated otherwise.

3. RESULTS

We recruited 70 men to participate in this study. One man of European ancestry was excluded at baseline because of hyperlipidaemia and 9 men (8 of African ancestry) dropped out after baseline assessments. Of these, 6 refused to participate with the low sodium intervention after the first visit and 3 dropped out during the low sodium intervention as they found the LS diet difficult to keep with their specific lifestyle. None of the participants had any symptoms or signs during the low sodium intervention of the study. There was no significant difference in age, baseline blood pressure, or other baseline characteristic between dropouts and patients who finished the study.

The baseline characteristics of the included participants are shown in Table 1. Sixty men (31 of African ancestry) were included, with a mean age of 37.2 (1.2) years and a mean BMI of 24.9 (0.4) kg/m². Mean sitting systolic/diastolic blood pressure (SBP/DBP) at baseline was 125.2 (1.6)/75.8 (1.3) mm Hg, with a heart rate of 61.0 (1.1) beats per minute. Eight participants had untreated high blood pressure at inclusion (SBP ≥ 140 or DBP ≥ 90 mm Hg). None of the participants had symptoms or signs of hypothyroidism at careful clinical examination, or used prescription medication for thyroid or other disease. Crude CK activities ranged from 63 to 1648 IU/L (median, 205 IU/L). We identified 1 possible outlier (1648 IU/L, in 1 participant of African ancestry) and after applying Dixon's one‐third rule, this value was excluded from further analysis. As the data still showed significant skewness and kurtosis as expected,7, 9 a logarithmic transformation was performed to acquire a normal distribution.

Table 1.

Baseline characteristics of the participants

Participants	Total (n = 60)	European (n = 29)	African (n = 31)
Age, yeara	37.2 (1.2)	33.1 (1.5)	41.3 (1.7)
BMI, kg/m2 a	24.9 (0.4)	23.8 (0.5)	25.8 (0.7)
CK, IU/Lb , c	205 (63 to 1648)	129 (63 to 468)	324 (116 to 1648)
Sodium, mmol/La , c	142.2 (0.3)	142.8 (0.4)	141.7 (0.3)
Potassium, mmol/La , c	4.2 (0.0)	4.2 (0.0)	4.2 (0.1)
Creatinine, μmol/La , c	95.0 (1.6)	91.8 (2.0)	98.0 (2.3)
Glucose, mmol/La , c	5.6 (0.1)	5.5 (0.1)	5.6 (0.1)
Cholesterol, mmol/La , c	4.7 (0.1)	4.9 (0.1)	4.6 (0.1)
TSH, mU/La , c , d	2.14 (0.20)	2.13 (0.29)	2.15 (0.27)
SBP, mm Hga	125.2 (1.6)	123.5 (1.8)	126.8 (2.7)
DBP, mm Hga	75.8 (1.3)	74.7 (1.6)	77.0 (2.0)
Hypertension, n	8	2	6

European, African, refer to participant's self‐reported ancestry.

BMI, body mass index; CK, plasma creatine kinase after 3 days of rest; TSH, thyroid‐stimulating hormone; SBP and DBP, systolic and diastolic blood pressure; Hypertension, untreated hypertension with systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥ 90 mm Hg.

^aMean (SE).

^bMedian (Range).

^cFasting plasma concentration.

^dN = 40 participants.

Monitoring of overnight urine on day 4 of the LS diet showed adequate adherence to the diet with lowering sodium excretion (data not shown). General and haemodynamic parameters at LS (day 7) and HS (day 10) are shown in Table 2A. Mean 24‐h urinary sodium excretion during LS and HS was 31.5 (3.5) and 320.0 (21.1) mmol/24‐h, respectively (P < .001), indicating adequate dietary compliance. Mean body weight and SBP increased significantly during HS as expected, whereas DBP and other parameters did not significantly change.

Table 2.

Outcome parameters by sodium intake

	Low sodium	High sodium	P‐value
(A) All participants (n = 60)
Body weight, kg	79.2 (1.5)	81.0 (1.6)	<.01
Systolic blood pressure, mm Hg	120.9 (1.3)	125.1 (1.4)	<.01
Diastolic blood pressure, mm Hg	73.6 (0.9)	74.6 (1.1)	ns
Heart rate, bpm	66.0 (1.4)	61.1 (1.3)	<.01
Plasma sodium, mmol/La	139.8 (0.2)	142.1 (0.2)	<.01
Plasma potassium, mmol/La	4.1 (0.0)	4.1 (0.0)	ns
Plasma creatinine, μmol/La	96.8 (1.6)	92.4 (1.5)	<.01
Urinary sodium, mmol/24‐h	31.5 (3.5)	320.0 (21.1)	<.01
Urinary potassium, mmol/24‐h	72.4 (4.2)	75.2 (4.6)	ns
Urinary creatinine, mmol/24‐h	14.6 (0.6)	14.9 (0.6)	ns
(B) Low CK (n = 19) vs high CK (n = 20)
Systolic blood pressure, mm Hg
Low CK	117.8 (2.6)	124.2 (2.6)	<.05
High CK	123.3 (2.6)	125.9 (2.7)	.05
Diastolic blood pressure, mm Hg
Low CK	73.7 (1.6)	74.9 (1.9)	ns
High CK	74.3 (1.5)	74.7 (1.7)	ns
Urinary sodium, mmol/24‐h
Low CK	28.0 (4.3)	415.2 (26.3)	<.01
High CK	34.5 (7.6)	260.4 (28.3)b	<.01
Urinary potassium, mmol/24‐h
Low CK	92.1 (8.2)	93.8 (8.0)	ns
High CK	62.4 (5.7)b	69.1 (7.5)b	ns
Urinary creatinine, mmol/24‐h
Low CK	15.4 (1.1)	16.1 (0.8)	ns
High CK	14.1 (0.9)	15.1 (1.4)	ns
(C) European (n = 29) vs African (n = 31)
Systolic blood pressure, mm Hg
European	119.1 (1.8)	124.1 (1.7)	<.01
African	122.6 (1.9)	126.0 (2.2)	<.05
Diastolic blood pressure, mm Hg
European	72.9 (1.2)	74.2 (1.6)	ns
African	74.2 (1.3)	75.0 (1.5)	ns
Urinary sodium, mmol/24‐h
European	24.7 (3.3)	439.1 (21.9)	<.01
African	38.1 (5.9)	208.6 (20.4)c	<.01
Urinary potassium, mmol/24‐h
European	87.9 (5.7)	95.9 (5.9)	ns
African	57.5 (4.5)c	55.9 (4.9)c	ns
Urinary creatinine, mmol/24‐h
European	15.3 (0.8)	16.2 (0.7)	ns
African	13.7 (0.9)	13.7 (1.0)	ns

All data are mean (SE), at low sodium diet (day 7) or high sodium diet (day 10).

CK, creatine kinase; low CK, lowest third CK values; high CK, highest third CK values. European, African, participants of European or African ancestry; ns, not significant (P > .05).

^aFasting plasma concentration. Missing values (in 0‐7 participants per outcome, median 0) were not imputed.

^b P < .05 for difference between high CK and low CK.

^c P < .05 for difference between participants of European vs African ancestry.

Parameters during LS and HS for the low (I) versus the high (III) CK tertile are shown in Table 2B and Figure 2. Urinary sodium excretion after HS was significantly lower in the high CK tertile as compared to the low tertile, respectively 260.4 (28.3) versus 415.2 (26.3) mmol/24‐h (P < .001). In line with this, the correlation coefficient r between resting plasma CK and 24‐h urinary sodium after HS was −0.48 (P < .001). The other parameters that significantly correlated with 24‐h urinary sodium excretion (mmol/24‐h) after HS were age and African ethnicity (r respectively −0.29; P = .013 and −0.71; P < .001). African ethnicity and CK were also highly correlated (r = 0.67; P < .001);6, 7, 9 and in line with previous reports on ethnic differences in sodium handling,24, 25, 26 the 24‐h urinary sodium excretion during HS was lower in participants of African ancestry (208.6 [20.4] mmol/24‐h versus 439.1 [21.9] for European ancestry [P < .001; Table 2C]). BMI did not correlate with urinary sodium excretion. Potassium excretion was also lower with African ancestry (as previously described)25 and lower in participants with high CK. Adjustment for 24‐h creatinine excretion, as a measure for urine sampling accuracy, did not affect the direction or magnitude of the outcomes (data not shown).

Figure 2.

Dot plot of 24‐h urinary sodium excretion after a high sodium diet in the low (n = 19) versus the high (n = 20) log creatine kinase (CK) tertile. Tertile I, lowest third; Tertile III, highest third CK values. Horizontal lines are means with standard error bars. Mean sodium excretion is lower in the high CK tertile (P < .001)

In PLS multivariable regression analysis for collinear variables, using log CK, age, and ancestry as predictors, we found an intercept of 617.1 mmol/24‐h. Sodium excretion with HS decreased with 98.4 mmol for each increase in log CK and with 1.8 mmol per year increase in age. European versus African ancestry respectively increased or reduced sodium excretion with a further 83.9 mmol; a model equation of: Na⁺ excretion HS (mmol/24‐h) = (617.1) ‐ (98.4*Log CK) ‐ (1.8*Age) +/− (83.9 for European versus African ancestry). The correlation and regression coefficients of the variables with their confidence intervals are depicted in Table 3.

Table 3.

PLS regression analysis of 24‐h sodium excretion after high sodium intake

Variable	Correlation coefficient	Regression coefficient	Confidence interval (95%)
			Lower bound	Upper bound
Intercept		617.1	531.3	702.9
LogCK	−0.5	−98.4	−134.4	−62.4
Age	−0.3	−1.8	−3.2	−0.3
EA	0.7	83.9	59.7	108.0
AA	−0.7	−83.9	−108.0	−59.7

Correlation coefficients and coefficients of the projection to latent structures (PLS) regression analysis with 95% confidence intervals.

AA, African ancestry (n = 31); EA, European ancestry (n = 29).

4. DISCUSSION

In this study we report that resting plasma CK is inversely associated with urinary sodium excretion after a high sodium diet. Other predictors of sodium excretion were age and ethnicity, as expected from previous reports.24, 25, 26 CK is known to rapidly provide ATP in situ for sodium reabsorption in the renal tubules (Figure 1).6, 11, 12, 13, 14 In the kidney, Na⁺/K⁺‐ATPase plays a key role in both ion homeostasis and blood pressure regulation.6, 11, 12, 13, 14, 21, 22, 27, 28, 29 Proximal tubule sodium handling accounts for 60%‐70% of reabsorption of all filtered sodium, 20%‐30% of the filtered load is absorbed in the thick ascending loop of Henle, and 5%‐10% in the distal tubule.21, 27, 28 Importantly, in all parts of the nephron, Na⁺/K⁺‐ATPase resides at the basolateral surface where it provides the force for the vectorial transport of sodium from the tubular lumen to the blood compartment by coupling hydrolysis of ATP to the active exchange of 3 intracellular Na⁺ ions for 2 K⁺ ions.21, 27

Evidence indicates that CK is functionally coupled to renal basolateral Na⁺/K⁺‐ATPase and that ATP produced by co‐localized CK is preferentially used for the high and fluctuating ATP demand of sodium transport across the tubular epithelial cells.11, 12, 13, 14 ATPase activity is higher when ATP is regenerated by creatine phosphate rather than by glycolysis. Furthermore, hydrolysis of creatine phosphate is inhibited by ouabain, an inhibitor of Na⁺/K⁺‐ATPase. Also, calcium transport across the plasma membrane is faster when creatine phosphate rather than glycolysis models provides ATP. These data further suggest functional coupling between CK and ATP requiring enzymes.10, 11, 12, 13, 14, 15, 16

Thus, high CK activity in the kidney tubule cells may lead to increased availability of ATP for the active process of sodium reabsorption.6, 7, 29 This thought is in line with the work of Lifton et al on molecular causes of hypertension with a focus on sodium handling in the kidney, stating that “increased sodium reabsorption leads to higher blood pressure.”27 The lower excretion of potassium in the setting of augmented sodium reabsorption is not completely understood, but is mainly attributed to lower sodium delivery to the distal nephron.25

In the absence of tissue damage, plasma CK at rest is thought to reflect tissue CK activity.6, 7, 8, 19, 20 Healthy tissue loses a small fraction of intracellular CK into the interstitial space, which is transported to the bloodstream via the lymphatic system. This release is proportional to tissue CK activity.7, 8, 19, 20 The coordinated expression of CK in different tissues yields resting plasma CK a valid measure of tissue CK, also representing tissues relatively low in CK.6, 7 Our findings are in accord with previous studies showing that in the absence of muscle damage, resting plasma CK is a useful measure of tissue CK and predicts systolic and diastolic blood pressure in the population, independent of ethnicity.7, 23, 30

Relatively high tissue and plasma CK activity is found in persons of African ancestry.6, 7, 8, 9, 20, 29, 31 Also, persons of African ancestry are known to have an enhanced ability to retain sodium.24, 25, 26 This is accompanied by lower potassium excretion,25 whereas renin activity is, on average, lower in persons with African versus European ancestry.25, 26, 29 The enhanced capacity of the kidney to retain sodium is thought to play a major role in the greater prevalence of (sodium‐sensitive) hypertension in this population subgroup.24, 25, 26, 29

Although the association we found between CK and sodium retention does not imply causation, our preliminary findings are in line with the existing evidence of CK rapidly providing ATP for tubular sodium retention.10, 11, 12, 13, 14 High CK might lead to greater ATP buffer capacity for renal sodium retention regardless of ancestry or ethnicity.6, 7, 23, 29 The findings of this study add to the body of evidence indicating that CK enhances ATP buffer capacity near ATPases, resulting in higher blood pressure.6, 7, 23, 29, 30, 31, 32 We previously showed that human vascular contractility is CK‐dependent and that CK expression in resistance arteries has a near‐perfect correlation with blood pressure.17, 18, 29 Furthermore, properties of skeletal muscle may affect blood pressure. High skeletal muscle CK activity is associated with a predominance of “fast” type II fibers, lower oxidative capacity, and capillary rarefaction, which may contribute to increased peripheral resistance and blood pressure.6, 31 Thus, the aggregated evidence indicates that high CK activity may enhance ATP buffer capacity in different tissues and facilitate the development of hypertension.6, 7, 8, 17, 18, 20, 23, 29, 31, 32 In line with this, high resting plasma CK correlates with failure of antihypertensive therapy23, 29 and treatment with an oral systemic CK inhibitor reduces blood pressure in spontaneously hypertensive rats, with evidence of attenuated vascular contractility and sodium retention.32

This study has several strengths and limitations. The main strength is that we included healthy men from different ethnic groups representing a wide spectrum of cellular CK activity6, 8, 20 and we were able to show for the first time that those with high CK excrete less sodium on a high sodium diet.

During the diet, the participants were closely monitored and supported by the research physician. With HS, the minimal daily amount of sodium (200 mmoles) was provided by the study staff to the participants. However, as the participants were not hospitalized during the 10 days of the study, we cannot exclude that the dietary intake of sodium during HS was below the target of minimally 200 mmoles sodium per day, but the participants seem to have followed the instructions well, as reflected by an adequate 24‐h creatinine excretion and mean sodium excretion (during LS or HS intake) being well below 50 mmol or above 200 mmol/24‐h, respectively.

The current experiment was limited to acute sodium handling. A long‐term balance study may render more information than presented in this paper. Renin and aldosterone were not studied, but many previous studies addressed these currently well‐known hormonal aspects of salt secretion, including Luft et al25, 26 who reported consistently lower renin in persons of African ancestry, before and after salt loading, without major differences in aldosterone compared to persons of European ancestry.

As in previous studies, we used resting plasma CK as an indirect measure of tissue CK.7, 8, 20, 23, 29 This is a validated, but somewhat “diluted” surrogate measure of tissue CK.7, 8, 18 Therefore, we standardised for exercise. Plasma CK is elevated up to 3 days with regular exercise, but up to a week after strenuous eccentric exercise, where the muscle lengthens and contracts at the same time against an external load. This leads to disruption of muscle fibers and highly elevated plasma CK—up to 10.000 IU/L during a week or longer.7, 33 However, none of the participants stated that they were involved in such eccentric exercise. Still, we cannot exclude an exercise‐induced component in the CK values and this could have weakened the association between CK and urinary sodium excretion.7 Furthermore, our previous studies showed that the correlation of plasma CK with blood pressure is a magnitude lower than of vascular CK with blood pressure.7, 18 Thus, the true association between renal tissue CK and sodium retention could be higher than estimated in this study.

Nonetheless, we chose to use plasma CK as a first step to analyze the association with sodium excretion, as with currently available techniques, assessing kidney CK would need invasive measures, which should not be considered without evidence of at least an association. Future studies could assess renal CK activity in relation to sodium handling as a more direct measure. In addition, the recently reported specific oral CK inhibitor for human use, beta‐guanidinopropionic acid34 might provide a new tool to further assess the potential role of CK in sodium handling of the human kidney in relation to blood pressure.

Finally, CK is strongly entangled with ancestry.6, 7, 8, 9, 20, 29, 30, 31 In particular, with rather small sample sizes, it can be statistically challenging to separate these 2 variables in analyses, as persons of African ancestry with low CK, as well as persons of European ancestry with high CK will more likely be underrepresented, yielding CK and ancestry to be collinear variables. Therefore, we used PLS regression analysis to incorporate these collinear variables into the same model. Importantly, ancestry or race/ethnicity is in itself not an explanation for sodium retaining capacity, a highly energy‐dependent process. With CK, we provide a biologically plausible mechanism for the enhanced sodium retention with a dose effect relationship.

In summary, our data indicate that participants with high CK activity display reduced urinary sodium excretion after a high sodium diet. CK is known to promote high blood pressure through greater vascular contractility, and to regenerate ATP near tubular Na⁺/K⁺‐ATPase for sodium transport.6, 7, 8, 11, 12, 13, 14, 17, 18, 29, 32 Wide interindividual variation in cellular CK activity and inherent variation in capacity to promote molecular motor activity is rather unique to humans, and this has clearly been shown to affect human cellular function.6, 7, 8, 17, 18, 20, 23, 29, 30, 31 The presented data provide first indications that differential CK activity might help explain the observed variation in sodium excretion in humans beyond ancestry or skin color. However, the pathophysiology of sodium retention is not resolved with our clinical study and mathematical exercise in collinear variables. Although the molecular function of CK in sodium transport is clear,6, 7, 8, 10, 11, 12, 13, 14, 15, 16 our preliminary finding of an association between CK and sodium excretion need further confirmation in more direct assessments of renal CK and in studies of longer duration. Also beta‐guanidinopropionic acid, the recently developed CK inhibitor for human use,34 might further our understanding of the significance of CK for the human physiology and pathophysiology of renal sodium handling.

CONFLICTS OF INTEREST

LMB is an inventor on patent WO/2012/138226 (filed).

Brewster LM, Oudman I, Nannan Panday RV, et al. Creatine kinase and renal sodium excretion in African and European men on a high sodium diet. J Clin Hypertens. 2018;20:334–341. 10.1111/jch.13182