The FASEB Journal

The FASEB Journal Copy of e-mail notification

z388810

Your article (# 11-199711 ) from The FASEB Journal is available for downloading. _______________________ The FASEB Journal is published by the Federation of American Societies for Experimental Biology (FASEB) Dear author, Please click on this URL address http://rapidproof.cadmus.com/RapidProof/retrieval/index.jsp then Login: your e-mail address Password: 5eoqCY3V865V (You may copy and paste this case-sensitive password.) The site contains one file. You will need to have Adobe Acrobat® Reader software (Version 4.0 or higher) to read it. This free software is available for user downloading at http://www.adobe.com/products/acrobat/readstep.html. If you have any problems with downloading your article from the Rapid Proof site, please contact [email protected]. Please include your article number (8810) with all correspondence. This file contains instructions for correcting the PDF proof electronically and a copy of the page proof for your article. The proof contains 9 pages. Please read over your article carefully. Please be sure to respond to any queries that appear on the last page of the proof. Proofread the following elements of your article especially carefully: - Non-English characters and symbols - Tables - Equations and mathematical symbols - Figures (including figure and caption placement): * If you need to provide the editorial office with a revised figure, please indicate this with a comment on the incorrect figure and write 'NEW FIGURE FILE REQUIRED' in the comment and include the new figure as an attachment to your email. * Color figures have the word 'COLOR' in the margins. Grayscale (black and white) figures have no tag. If you wish to change a color figure to grayscale, include a comment instructing the figure to print in grayscale. * Publication charges will be calculated based on your changes to this proof. Please note that figures are published the same way in the online and print versions of the journal. Authors may not publish figures in color online while publishing the same figures in grayscale in print or vice-versa. Please use the comments and notes features in Adobe Acrobat (instructions provided in the PDF file to be downloaded) and return the annotated PDF to [email protected] within 24 hours. If you are unable to use Adobe Acrobat and your changes are minimal, you may summarize them in an email (sent to the same fjproduction email address), clearly indicating the location of each change. If you are unable to use Adobe Acrobat and have changes too extensive to summarize in an e-mail, please scan your marked-up proof and send it to [email protected]. If you are unable to provide your proof corrections in any of the above manners, please fax your corrections to: 1-717-738-9479.

Fax PUBLICATION FORMS to: 1-301-634-7809 Publication forms sent by any other means or faxed to any other number may result in substantial publication delays. NOTE: Proofs or publication forms retained by the author for an excessive length of time may not be published online in a timely way and may need to be scheduled for a later print issue. If you have any problems or questions, please contact me. Always include your article number in all correspondence. Sincerely, Mary Kiorpes Hayden FASEB Office of Publications 9650 Rockville Pike Bethesda, MD 20814-3998 phone: 301-634-7151 fax: 301-634-7809 [email protected]

Adding Comments and Notes to Your PDF To facilitate electronic transmittal of corrections, we encourage authors to utilize the comments and notes features in Adobe Acrobat. The PDF provided has been “commentenabled,” which allows you to utilize the comments and notes features, even if using only the free Adobe Acrobat reader (see note below regarding acceptable versions). Adobe Acrobat’s Help menu provides additional details on the tool. When you open your PDF, the comments/notes/edit tools are clearly shown on the tool bar (though icons may differ slightly among versions from what is shown below). For purposes of correcting the PDF proof of your journal article, the important features to know are the following: x

Use the Text Edits tool (

) to insert, replace, or delete text.

o To insert text, place your cursor at a point in text and select “Insert Text at Cursor” from the text edits menu. Type your additional text in the pop-up box.

o To replace text (do this instead of deleting and then re-inserting), highlight the text to be changed, select “Replace Selected Text” from the text edits menu, and type the new text in the pop-up box.

o To delete text, highlight the text to be deleted and select “Cross Out Text for Deletion” from the text edits menu (see graphic above).

x

Use the Sticky Note tool ( ) to describe changes that need to be made (e.g., changes in bold, italics, or capitalization use; altering or replacing a figure) or to answer a question or approve a change that was posed by the editor. To use this feature, click on the sticky note tool and then click on a point in the PDF where you would like to make a comment, then type your comment in the pop-up box.

x

Use the Callout tool ( ) to point directly to changes that need to be made. Try to put the callout box in an area of white space so that you do not obscure the text, as in the example below.

x

Use the Highlight tool ( ) to indicate font problems, bad breaks, and other textual inconsistencies. Describe the inconsistencies with the callout tool (shown) or a sticky note. One callout (or sticky note) can describe many changes.

An alternate method is to select the appropriate text with your cursor, select “Add Note to Selected Text” from the text edits menu, and then type your note in the pop-up box (the selected text is highlighted automatically).

As with hand-annotated proof corrections, the important points are to communicate changes clearly and thoroughly; to answer all queries and questions; and to provide complete information for us to make the necessary changes to your article so it is ready for publication. To utilize the comments/notes features on this PDF you will need Adobe Reader version 7 or higher. This program is freely available and can be downloaded from http://get.adobe.com/reader/

AUTHOR INSTRUCTIONS

Copyright Transfer and Publication Costs Approval Form All authors are required to sign the following copyright transfer and cost agreement form prior to publication. If you have not yet submitted this form to the editorial office, please fax it to 301-6347809 as soon as possible. Multiple forms may be submitted for the same article. Instructions for calculating publication charges are also attached for your convenience. Please read that sheet carefully. Do not fax your estimate sheet to the editorial office.

The FASEB Journal

Mandatory Copyright Transfer and Publication Costs Approval Form Manuscript No.: Title: Author Names (Please Print All Names):

Signatures Below Certify Compliance With the Following Statements: Copyright Transfer. In consideration of the acceptance of the above work for publication, I do herby assign and transfer to the Federation of American Societies for Experimental Biology (FASEB) all rights, titles, and interest in and to the copyright in The FASEB Journal. This includes preliminary display/posting of the abstract of the accepted article in electronic form before publication. The journal grants the author permission to provide a copy of the accepted manuscript to NIH upon acceptance for Journal publication, with public release in PubMed Central twelve months after final print publication by The FASEB Journal. This form must be signed by all authors. If any changes in authorship (order, deletions, or additions) occur after the manuscript is submitted, agreement by all authors for such changes must be on file with FASEB. An author’s name may only be removed at his/her own request and with written consent from all of the other authors, as well as final approval by the Editor-in-Chief. Material prepared by employees of the US Government in the course of their official duties cannot be copyrighted; work prepared by employees of the British or British Commonwealth government in the course of their official duties is subject to Crown Copyright and cannot be transferred to FASEB. Nevertheless, authors must sign the form to indicate acceptance of all terms other than copyright transfer. Please check if this article was written as part of the official duties of an employee of the US government. Please check if this article was written as part of the official duties of an employee of the British or British Commonwealth government. Authorship Responsibilities. I attest that: 1. the manuscript is not currently under consideration, in press, or published elsewhere, and the research reported will not be submitted for publication elsewhere until a final decision has been made as to its acceptability by The FASEB Journal (posting of submitted material on a web site or by any other electronic means may be considered prior publication—note this in your cover letter); 2. the manuscript is truthful, original work without fabrication, fraud, or plagiarism; 3. I have made an important scientific contribution to the study and am thoroughly familiar with the primary data and; 4. I have read the complete manuscript and take responsibility for the content and completeness of the manuscript and understand that I share responsibility if the paper, or part of the paper, is found to be faulty or fraudulent. Conflict of Interest Disclosure. All funding sources supporting the work and all institutional or corporate affiliations of mine are acknowledged. Except as disclosed on a separate attachment, I certify that I have no commercial associations (e.g., consultancies, stock ownership, equity interests, patent-licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article, and that I accept full responsibility for the conduct of the trial, had full access to all the data, and controlled the decision to publish. Author Fees. I agree to pay the following publication charges. Page charges are $80 per printed page for the first 8 pages; $160 per page for the 9th page and beyond. Color figures are $350 each. Articles containing eight or more figures and/or tables will be charged an additional $150 per figure and table. Supplemental units uploaded to the journal website are $160 each (maximum of 4 units). All figures are in U.S. dollars. Author Signatures. For more than 10 authors, use an extra sheet. Multiple forms are acceptable. *NOTE: Electronic/digital signatures of any kind are not acceptable; any form containing these signatures will be returned. This form must be signed directly by the author.* 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

Print Name: Print Name: Print Name: Print Name: Print Name: Print Name: Print Name: Print Name: Print Name: Print Name:

Signature: Signature: Signature: Signature: Signature: Signature: Signature: Signature: Signature: Signature:

Date: Date: Date: Date: Date: Date: Date: Date: Date: Date:

Signed forms should be faxed to 301-634-7809 or scanned and emailed to Mary Hayden at [email protected]. E-mail questions to Mary Hayden at [email protected] or call 301-634-7151.

AUTHOR INSTRUCTIONS

Calculating Publication Costs The FASEB Journal now uses an online bill pay system for publication and reprint order fees. The corresponding author will receive a separate e-mail containing a link to a web page where charges can be paid by credit card. The e-mail link also will provide information about invoices, check payment, and wire transfers. Publication charges are calculated based on the final version of the article and not this proof. Color figures have the word “COLOR” in the margins. Grayscale (black and white) figures have no tag. If you wish to change a color figure to grayscale, cross out the word “color” and write in “grayscale.” Please note that figures are published the same way in the online and print versions of the journal. Authors may not publish figures in color online while publishing the same figures in print or vice-versa. For your convenience and records, the following table should help you estimate publication charges. Please account for any changes you have made to your proofs. Please do not fax this sheet to the editorial office. Your charges will be calculated automatically based on the final article.

Quantity Per page cost, first 8 pages

Unit Price ×

$80 per page

Total = +

Per page cost, 9th page and beyond

×

$160 per page

= +

Color figures

×

$350 per figure

= +

Any combination of figures and/or tables over 7

×

$150 per figure or table

= +

Individual supplemental files

×

$160 per unit, maximum of four units

=

Total Publication Charges: This sheet is for estimation purposes only. Final costs are calculated using the final printed version of the article. All prices are in U.S. dollars. Open access fees are calculated separately. Reprints can be purchased when paying publication charges online.

tapraid4/z38-faseb/z38-faseb/z3800712/z388810d12z xppws S⫽1 3/26/12 4:46 Ms. No.: 11-199711 Input-lak

The FASEB Journal • Research Communication

Circadian expression of H,K-ATPase type 2 contributes to the stability of plasma Kⴙ levels Amel Salhi,* Gabriel Centeno,† Dmitri Firsov,† and Gilles Crambert*,1 *Université Pierre et Marie Curie Paris 6, Institut National de la Santé et de la Recherche Médicale (INSERM), Centre National Recherche Scientifique (CNRS) Unité Mixte de Recherche en Santé (UMRS) 872, Equipe 3, Laboratoire de Génomique, Physiologie et Physiopathologie Rénales ERL Paris, France; and †Department of Pharmacology and Toxicology, University of Lausanne, Lausanne, Switzerland Maintenance by the kidney of stable plasma K values is crucial, as plasma Kⴙ controls muscle and nerve activity. Since renal Kⴙ excretion is regulated by the circadian clock, we aimed to identify the ion transporters involved in this process. In control mice, the renal mRNA expression of H,K-ATPase type 2 (HKA2) is 25% higher during rest compared to the activity period. Conversely, under dietary Kⴙ restriction, HKA2 expression is ⬃40% higher during the activity period. This reversal suggests that HKA2 contributes to the circadian regulation of Kⴙ homeostasis. Compared to their wild-type (WT) littermates, HKA2null mice fed a normal diet have 2-fold higher Kⴙ renal excretion during rest. Under Kⴙ restriction, their urinary Kⴙ loss is 40% higher during the activity period. This inability to excrete Kⴙ “on time” is reflected in plasma Kⴙ values, which vary by 12% between activity and rest periods in HKA2-null mice but remain stable in WT mice. Analysis of the circadian expression of HKA2 regulators suggests that Nrf2, but not progesterone, contributes to its rhythmicity. Therefore, HKA2 acts to maintain the circadian rhythm of urinary Kⴙ excretion and preserve stable plasma Kⴙ values throughout the day.—Salhi, A., Centeno, G., Firsov, D., Crambert, G. Circadian expression of H,K-ATPase type 2 contributes to the stability of plasma Kⴙ levels. FASEB J. 26, 000 – 000 (2012). www.fasebj.org ABSTRACT

ⴙ

Key Words: diurnal variation 䡠 potassium balance 䡠 electrolyte 䡠 epithelial ion transport Stability of plasma K⫹ levels is crucial for a variety of physiological functions and is particularly important for the electrical properties of cell membranes in both excitable and nonexcitable tissues. This parameter is strongly and mainly controlled by the kidney, which is able to adapt its K⫹ excretion to the dietary K⫹ intake.

To achieve this, the filtered K⫹ is almost completely reabsorbed in the proximal tubules and the loops of Henle. The distal part of the nephron then plays an important role to finely tune the amount of excreted K⫹; it is equipped with ion transporters that mediate either K⫹ secretion [renal outer medulla potassium (ROMK) channels] or K⫹ reabsorption [H,K-ATPase type 1 and 2 (HKA1 and HKA2)] (for review, see ref. 1). Under normal conditions, the distal part of the nephron is mainly a site of K⫹ secretion, which allows it to eliminate the amount of K⫹ ingested daily. In circumstances like K⫹ restriction, HKA2 is stimulated in the kidney, presumably to limit the loss of K⫹, although analysis of the renal phenotype of HKA2-deficient mice failed to validate this hypothesis (2). Regarding urinary K⫹ excretion, its hormonal regulation has been intensively studied, and results have highlighted the role of aldosterone and angiotensin II (3), mainly as regulators of ROMK function. More recently, HKA2 has been shown to be acutely regulated by pH (through a G-protein-coupled receptor, GPR4; ref. 4), kallikrein (5), interaction with tetraspan CD63 (6), and chronically by progesterone (7) and oxidative stress responsive transcription factor Nrf2 (8). Renal function and ion homeostasis are not only regulated by “reactive” processes involving hormones but also by “predictive” mechanisms depending on the clock system (9). This system is composed of molecular oscillators driven by a central pacemaker located in the suprachiasmatic nucleus of hypothalamus. The molecular basis of these oscillators is a set of self-autonomous transcriptional/translational feedback loops involving PAS domain transcription factors named Clock, Bmal1, and Npas1 and their feedback repressors Per1, Per2, Cry1, and Cry2. Each of these genes may also control AQ: 1 output gene expression in a tissue-dependent manner 1

Abbreviations: CCD, cortical collecting duct; ENaC, epithelial sodium channel; HKA1, H,K-ATPase type 1; HKA2, H,KATPase type 2; KO, knockout; PPIA, peptidylprolyl isomerase A; PR, progesterone receptor; ROMK, renal outer medulla potassium; WT, wild type; ZT, zeitgeber time 0892-6638/12/0026-0001 © FASEB

Correspondence: Centre de Recherche des Cordeliers, Génomique, Physiologie et Physiopathologie Rénales, Equipe 3 UMRS 872, ERL 7226, 15 rue de l’Ecole de Médecine, 75270 Paris cedex, France. E-mail: [email protected] doi: 10.1096/fj.11-199711 This article includes supplemental data. Please visit http:// www.fasebj.org to obtain this information. 1


conferring rhythmic and specific physiological functions. As a matter of fact, recent investigations showed that canonical clock genes have a robust circadian variation of their expression in the kidney that can be regulated by food and light cues (10, 11), indicating that renal function is under the control of the circadian clock. Recently, Zuber et al. (12) described the global transcriptional variations of distal nephron transcripts during the circadian cycle both in wild-type (WT) and in clock-deficient mice, which allow for identification of a significant number of clock-controlled genes. Many of these genes are involved in the homeostatic function of the kidney and encode for ion and solute transporters and their regulators. The same researchers also demonstrated that disruption of the circadian clock affects water, Na⫹, and K⫹ homeostasis (13). As described above, renal functions are, therefore, dependent on the circadian cycle system. It remains now to clearly identify the protagonists, their exact involvement, and their physiological relevance in the context of the renal adaptation to the activity/rest cycle. Interestingly, both reactive and predictive regulatory systems are interlinked. Indeed, Gumz et al. (14, 15) recently showed that the stimulation of Na⫹ transport through aldosterone-induced epithelial sodium channel (ENaC) expression depends on the Per1 factor. Regarding urinary K⫹ excretion, it is known to follow a strong circadian variation, with an increased excretion during the period of activity independent of nutritional behavior (16, 17). This renal adaptation to K⫹ intake variations between activity and rest periods enables the organism to maintain a stable plasma K⫹ level (18). In rodents, this circadian cycle is maintained even during nonfeeding periods, whereas the circadian excretion of Na⫹ is abolished (19). A high-K⫹ diet changes the amplitude but not the period of the rhythmic urinary excretion (19). Adrenal steroids are involved in the circadian excretion of K⫹, more particularly corticosterone (17). Interestingly, the circadian K⫹ excretion is sensitive to amiloride, which suggests the predominance of the ENaC/ROMK partnership all along the 24-h cycle (20). The purpose of our study is therefore to investigate whether the actors that promote K⫹ reabsorption in the distal part of the nephron (namely, HKA1 and HKA2) are regulated by the circadian system and contribute to cyclic urinary K⫹ excretion, and therefore to the stability of the plasma K⫹ value.

MATERIALS AND METHODS In vivo studies Experiments were performed on C57BL6 WT and knockout (KO) mice for the clock gene (21) or for the HKA2␣ subunit gene (2). A colony of clock-deficient mice was established from breeding pairs of clock⫹/⫺ heterozygous mice originally generated by Debruyne et al. (21). These mice were placed under a strict 12-h light-dark alternation with ad libitum access to 2

Vol. 26

July 2012

food and water. The day was divided into zeitgeiber time (ZT) periods. ZT0 corresponds to the time when the light is turned on and ZT12 to the time when it is turned off. For a rodent, the activity period therefore corresponds to the part of the day from ZT12 to ZT0, whereas the rest period is from ZT0 to ZT12. WT and HKA2-deficient mice were fed the standard laboratory diet (0.28% Na⫹ and 0.6% K⫹, referred to as normal conditions; Safe, Augy, France) or a low-K⫹ diet AQ: 2 (0.28% Na⫹ and 0% K⫹) for 5 d. Physiological analysis The metabolic studies were performed after 2 d of animal adaptation to encaging. Urine samples were collected for 4-h periods, and urinary creatinine concentrations were determined on an automatic analyzer (Konelab 20i; Thermo, Cergy Pontoise, France). Urinary Na⫹ and K⫹ concentrations were determined by flame photometry (IL943; Instrumentation Laboratory, Le Pré Saint Gervais, France) and urinary NH4⫹ concentration was measured by the colorimetric method of Berthelot. Stools produced during the whole activity or rest periods were collected, dried, brushed (to eliminate food contaminant), and homogenized in distilled water (5 ml/g). Proteins were then precipitated using trichloroacetic acid (20%) and removed by centrifugation (10,000 g; 10 min at 4°C). Potassium contents were then measured on the supernatant by using a flame photometer (IL 943; Instrumentation Laboratory). Plasma K⫹ concentrations were measured by tail incision on anesthetized animal with an ABL77 pH/blood-gas analyzer (Radiometer, Lyon, France). All animal procedures were carried out in accordance with the French legislation for animal care and experimentation. Quantitative PCR RNAs were extracted from whole kidneys using the TRI reagent (Invitrogen, Villebon sur Yvette, France) following the manufacturer’s instructions. Total RNA (1 ␮g) was then reverse-transcribed using the first-strand cDNA synthesis kit for RT-PCR (Roche Diagnostics, Meylan, France) according to the manufacturer’s instructions. Real-time PCRs were performed on a LightCycler (Roche Diagnostics). No DNA was detected in samples that did not undergo reverse transcription or in blank run without cDNA. In each run, a standard curve was obtained using serial dilution of stock cDNA prepared from mouse kidney total RNA. Specific primers for targeted transcripts were chosen using the LC Probe Design 2.0 software (Roche Diagnostics). Immunolabeling on microdissected tubules Mice fed normal or low-K⫹ diet were sacrificed either at the end of the activity period (between ZT20 and ZT0) or at the end of the rest period (between ZT8 and ZT12). Microdissected cortical collecting ducts (CCDs) were then treated as recently described (22, 23) and incubated with an anti-HKA2 antibody (24) previously characterized (kind gift from Nikolai N. Modyanov, University of Toledo, Toledo, OH, USA). After mounting, the slides were observed on a confocal microscope (Zeiss Observer.Z1; Carl Zeiss, Oberkochen, Germany), and images were analyzed by ImageJ software (U.S. National Institutes of Health, Bethesda, MD, USA). Progesterone measurement Blood (100 –150 ␮l) was taken by retroorbital puncture on anesthetized mice (pentobarbital, 50 mg/kg) and placed in

The FASEB Journal 䡠 www.fasebj.org

SALHI ET AL.

tapraid4/z38-faseb/z38-faseb/z3800712/z388810d12z xppws S⫽1 3/26/12 4:46 Ms. No.: 11-199711 Input-lak TABLE 1. Cosine wave analysis of physiological parameters under normal and low-K⫹ diets NK diet Parameter

Plasma progesterone K⫹/creatinine WT HKA2-null Urine volume WT HKA2-null

LK diet

Acrophase (ZT, h)

Amplitude (%)

r2

Acrophase (ZT, h)

Amplitude (%)

r2

12

76

0.965

11

46

0.904

20 21

52 41

0.867 0.806

22 21

36 56

0.828 0.980

20 20

43 70

0.881 0.753

20 20

48 90

0.820 0.720

Data were fitted with a cosine wave equation, using a nonlinear least-squares regression, as described in Material and Methods. From this cosine, the time of the peak (acrophase) and the amplitude of the variation around the mean (%) were calculated. NK, normal K⫹; LK, low K⫹.

tubes containing heparin (2% final; Sanofi Winthrop, Gentilly, France). The samples were then centrifuged for 15 min at 1700 g, and the plasma was placed in new tubes and frozen before hormone measurements. Plasma (50 ␮l) was taken to measure progesterone by radioimmunoassay (Immunotech, Beckman Coulter, Marseille, France), according to the manufacturer’s instructions.

ZT0 and ZT12 (light switched on) corresponds to the 12-h rest period, whereas that between ZT12 and ZT0 (light switched off) corresponds to the 12-h activity period]. Moreover, the expression of the core clock gene Per1 clearly exhibited a circadian pattern with a peak of expression at ZT11 and an amplitude of ⬃60% (Table 2), similar to what has been measured by others T2 in rodent kidney (10, 11).

Statistical analysis Results were first analyzed using a 1-way ANOVA (Prism 5; GraphPad, San Diego, CA, USA) to determine whether one or more groups exhibited significant variations compared to the whole set of data. To test for a circadian variation of the parameters investigated (gene expression, ion excretion, or hormone level), we then fitted the data with a cosine wave equation, using a nonlinear least-squares regression (KaleidaGraph; Synergy Software, Reading, PA, USA; ref. 10) which allowed us to calculate the acrophase and to evaluate the circadian aspect of the phenomenon (defined as data set fitting with a cosine wave having a period of 24 h). For statistical comparison between 2 groups, we used Student’s t test analysis.

Renal expression of HKA2 but not HKA1 is under the control of the circadian clock As shown in Supplemental Fig. S1, HKA1 mRNA expression does not display significant variations all along the 24-h period either in WT (Supplemental Fig. S1A) or in clock-deficient mice (Supplemental Fig. S1B), and the absence of clock does not modify the overall HKA1 expression (Supplemental Fig. SC). In WT mice, the expression of HKA2 does exhibit a circadian pattern, with a higher level during the rest period than the activity period (Fig. 1A, C) and an acrophase around F1 ZT11 (Table 1). The amplitude of this variation is 15% around the daily mean. This pattern disappears in clock-deficient mice (Fig. 1B, C). The absence of rhythmicity of HKA2 mRNA expression in clock-deficient mice is unlikely to result in global modifications of circadian behavior, since those mice still exhibit robust behavioral rhythmicity (25). At the protein level, we observed that the number of cells exhibiting a labeling of HKA2 at the luminal side of the cortical collecting tubule (absent in HKA2-null mice; see Supplemental

RESULTS

T1

We first evaluated whether renal function is subjected to circadian rhythms of renal function in our experimental conditions. As shown in Table 1, both WT and HKA2-deficient mice exhibited a circadian profile of urine volume and creatinine excretion, which both exhibited a peak of excretion around ZT20 [as mentioned in Materials and Methods, the interval between

TABLE 2. Cosine wave analysis of gene expression under normal and low-K⫹ diets NK diet Gene

Per 1 HKA2 PR Nrf2

LK diet 2

Acrophase (ZT)

Amplitude (%)

r2

11 22 21 18

56 45 26 18

0.869 0.912 0.599 0.720

Acrophase (ZT)

Amplitude (%)

r

11 11 12 8

64 15 26 15

0.957 0.954 0.802 0.820

Data were fitted with a cosine wave equation, using a nonlinear least-squares regression, as described in Material and Methods. From this cosine, the time of the peak (acrophase) and the amplitude of the variation around the mean (%) were calculated. NK, normal K⫹; LK, low K⫹. RENAL H,K-ATPase REGULATES CIRCADIAN K⫹ BALANCE

3


plasma K⫹ variation (P⫽0.004; paired Student’s t test; Fig. 3E) of ⬃0.5 mM between ZT20 (4.2⫾0.1 mM) and ZT8 (3.7⫾0.2 mM). Thus, HKA2 contributes to stabilizing plasma K⫹ values along the 24-h period. Circadian expression of HKA2 is affected by chronic dietary Kⴙ restriction Because a known situation leading to HKA2 stimulation is dietary K⫹ restriction, we investigated how it affects

AQ: 6

F2

Figure 1. Expression of HKA2 in WT and clock-deficient mice fed normal K⫹ diet. A, B) Expression of HKA2 relative to peptidylprolyl isomerase A (PPIA), relative to its daily mean, was measured at 6 different times of the day in WT (A) or clock-deficient mice (B). The interval between ZT0 and ZT12 (light on) is the rest period for rodents, whereas the interval between ZT12 and ZT0 (light off, indicated by gray zone) is the activity period. Dotted line represents the daily mean value (equal to 1). Dot size is proportional to se; n ⫽ 5. Intergroup comparisons (1-way ANOVA) showed significant variability (P⫽0.022) for WT mice but not for clock-deficient mice. C) Mean expression of HKA2/PPIA over 24 h (daily mean) and during the periods of activity (ZT16, ZT20, and ZT0) and rest (ZT4, ZT8, and ZT12) in WT (open bars) and clock-deficient mice (solid bars); n ⫽ 15. Results are shown as means ⫾ se. **P ⬍ 0.01; nonpaired Student’s t test.

Fig. S2) was much higher at the end of the rest period when compared to the end of activity period (Fig. 2). HKA2 is involved in the circadian urinary excretion of Kⴙ under normal conditions

F3

Regarding urinary K⫹ excretion, WT mice fed the normal diet (Fig. 3A and Table 1) excreted almost 75% of the total amount of K⫹ excreted daily during the activity period, with a peak in excretion around ZT20. The HKA2-deficient mice also exhibited a circadian urinary excretion of K⫹ (Fig. 3B and Table 1). However, during the rest period, the HKA2-deficient mice excreted more K⫹ compared to WT mice, as shown in Fig. 3C. This observation is specific to K⫹, since neither the circadian urinary excretion of Na⫹ (Supplemental Fig. S3) nor that of NH4⫹ (Supplemental Fig. S3D, F) was significantly disrupted by the lack of HKA2. Since HKA2-deficient mice exhibited a higher loss of K⫹ during the rest period than the WT mice, we speculate that it may affect the levels of plasma K⫹. As shown in Fig. 3D, WT mice had the same plasma K⫹ values at ZT20 (activity period) and at ZT8 (rest period), 3.9 ⫾ 0.1 and 3.8 ⫾ 0.2 mM, respectively. In contrast, HKA2-deficient mice displayed a significant 4

Vol. 26

July 2012

C O L O R Figure 2. Expression of HKA2 in CCDs of WT mice fed normal-K⫹ diet. A–D) HKA2 localization in mouse CCDs isolated at the end of the activity period (ZT20 –ZT0; A, C) or at the end of the rest period (ZT8 –ZT12; B, D). After labeling with DAPI (nucleus in blue; A, B) and an anti-HKA2 antibody (green; C, D), slides were examined by confocal microscopy. Short arrows indicate cell luminal-side labeling; arrows indicate lumen. Asterisks indicate nonspecific labeling of nucleus. E) Number of cells exhibiting apical-oriented labeling normalized to the total number of nuclei in the same image. Results are shown as means ⫾ se (activity period: n⫽10 images analyzed from 3 mice; rest period: n⫽13 pictures analyzed from 3 mice). **P ⬍ 0.01; nonpaired Student’s t test.


SALHI ET AL.


Figure 3. Urinary K⫹ excretion and plasma K⫹ values during the day in WT and HKA2-deficient mice fed normal K⫹ diet. A, B) Urinary K⫹/creatinine ratio for a 4-h period relative to daily mean for WT mice (A) and HKA2-deficient mice (B). Dot size is proportional to se. Intergroup comparisons (1-way ANOVA) showed significant variability in WT (P⬍0.01) and HKA2-deficient mice (P⬍0.01); n ⫽ 6. C) Urinary K⫹/ creatinine ratio for WT mice (open bars) and HKA2-deficient mice (solid bars) at different periods of time; n ⫽ 6. *P ⬍ 0.05; nonpaired Student’s t test. D, E) Plasma K⫹ values for WT (n⫽5; D) and HKA2-deficient mice (n⫽6; E) fed a normal diet during the activity period (ZT20) and the rest period (ZT8). **P ⬍ 0.01; paired Student’s t test. Results are shown as means ⫾ se.

F4

F5

the circadian expression of this transporter by feeding mice a low-K⫹ diet for 5 d. As shown in Fig. 4A and Table 2, HKA2 expression still exhibited a circadian profile. However, it was reversed (acrophase around ZT20) and had a larger amplitude (45%) compared to what was observed in normal conditions. In other words, after 5 d of K⫹ restriction, HKA2 expression was higher during the activity period than during the rest period (Fig. 4B). The number of cells displaying a luminal HKA2 labeling was 40% higher in CCDs of mice sacrificed at the end of the activity period (ZT20 – ZT0) than at the end of the rest period (ZT8 –ZT12; Fig. 5). However, the difference between both periods

C O L O R Figure 5. Expression of HAK2 in CCDs of WT mice fed low-K⫹ diet. A–D) HKA2 localization in mouse CCDs isolated at the end of the activity period (ZT20 –ZT0; A, C) or at the end of the rest period (ZT8 –ZT12; B, D). After labeling with DAPI (nucleus in blue; A, B) and an anti-HKA2 antibody (green; C, D), slides were examined by confocal microscopy. Arrowheads indicate cell luminal-side labeling; arrows indicate lumen. Asterisks indicate nonspecific labeling of nucleus. E) Number of cells exhibiting an apical-oriented labeling normalized to the total number of nuclei in the same image. Results are shown as means ⫾ se (activity period: n⫽6 images analyzed from 2 mice; rest period: n⫽16 images analyzed from 3 mice). *P ⬍ 0.05; nonpaired Student’s t test.

was smaller than in normal diet conditions, and its significance was borderline (P⫽0.042), probably because the membrane expression of the HKA2 was already stimulated during K⫹ depletion independently of the circadian regulation. HKA2 is involved in the circadian urinary excretion of Kⴙ under dietary Kⴙ restriction ⫹

Figure 4. Expression of HAK2 in WT mice fed low-K diet (5 d). A) Expression of HKA2/PPIA, relative to its daily mean, was measured at 6 different times of the day in WT mice; n ⫽ 5. Dot size is proportional to se. Intergroup comparisons (1-way ANOVA) showed significant variability (P⫽0.039). B) Mean expression of HKA2/PPIA during periods of activity (ZT16, ZT20, and ZT0) and rest (ZT4, ZT8, and ZT12); n ⫽ 15. Results are shown as means ⫾ se. **P ⬍ 0.01; nonpaired Student’s t test. RENAL H,K-ATPase REGULATES CIRCADIAN K⫹ BALANCE

To our knowledge, the circadian excretion of K⫹ has not been investigated under dietary K⫹ restriction. The preservation of a circadian rhythm for renal function in this condition was proved by the urine volume and creatinine excretion profiles observed for both mouse strains (Table 1). Per1 gene also conserved its pattern of expression, with an acrophase around ZT11 and similar amplitude of variation (around 60%), as compared to normal conditions (Table 2). As shown in Fig. 6A, F6 5


Figure 6. Urinary K⫹ excretion and plasma K⫹ values during the day in WT and HKA2-deficient mice fed low-K⫹ diet. A, B) Urinary K⫹/creatinine ratio for a 4-h period relative to daily mean for WT mice (A) and HKA2-deficient mice (B). Dot size is proportional to se. Intergroup comparisons (1-way ANOVA) showed significant variability in WT (P⬍0.01) and HKA2-deficient mice (P⬍0.01); n ⫽ 6. C) Urinary K⫹/ creatinine ratio for WT mice (open bars) and HKA2-deficient mice (solid bars) at different periods of time; n ⫽ 6. *P ⬍ 0.05; nonpaired Student’s t test. D, E) Plasma K⫹ values for WT (n⫽5; D) and HKA2-deficient mice (n⫽6; E) fed a low-K⫹ diet (for 5 d) during the activity period (ZT20) and the rest period (ZT8). *P ⬍ 0.05; paired Student’s t test. Results are shown as means ⫾ se.

urinary K⫹ excretion varied with a circadian pattern similar to what was observed under normal-K⫹ diet conditions in WT mice (acrophase around ZT21, Table 2). However, the amplitude of these variations is lower under low-K⫹ diet conditions than under normal conditions (36 and 52%, respectively). Interestingly, the HKA2-deficient mice fed a low-K⫹ diet also exhibited a circadian K⫹ excretion (Fig. 6B and Table 2) with a higher loss of K⫹ during the activity period than the WT mice (Fig. 6C). Potassium restriction induced a decrease in plasma K⫹ values in both WT and HKA2deficient mice compared to normal conditions. These plasma K⫹ values remained identical (Fig. 6D) between activity and rest periods (3.33⫾0.06 and 3.38⫾0.14 mM, respectively) in WT mice. However, HKA2-deficient mice displayed a slightly but significantly (P⫽0.043; paired Student’s t test; Fig. 4E) lower plasma K⫹ value during the activity period (2.70⫾0.06 mM) than during the rest period (3.00⫾0.16 mM). Transcriptional regulators of HKA2 expression are affected by the circadian cycle

F7

Among the different regulators of HKA2, we focused on those directly affecting its transcriptional expression: progesterone and its nuclear receptor [progesterone receptor (PR)[; ref. 7] and Nrf2 (8). The plasma level progesterone in mice fed a low-K⫹ diet (Fig. 7B) was moderately increased from 50 to 100% at each time point when compared to the normal 6

Vol. 26

July 2012

diet condition (Fig. 7A). However, the circadian profile of progesterone was similar in both conditions, with a constant increase during the rest period to reach a maximum at ZT12 (Table 1) and a constant decrease during the activity period, with a minimum at ZT0. In the kidney, PR is expressed in the distal part of the nephron (26). The mean expression of PR, over the 24-h period, was stimulated by a 5-d K⫹ restriction (Fig. 7C). The expression profile of PR under normal-K⫹ conditions was similar to that of progesterone levels, with minimum and maximum values around ZT0 and ZT12, respectively (Fig. 7D) and an amplitude of the variation close to 30%. This distribution along the 24-h period was modified by a low-K⫹ diet (Fig. 7E) and was no longer in phase with the plasma progesterone level; its maximal expression occurred at ZT0 (Table 1), whereas this time corresponded to the lowest level of circulating progesterone. Another transcription factor, Nrf2, which binds antioxidant-responsive elements on gene promoters, has been recently shown to participate in the expression of HKA2 (8). Compared to normal conditions, dietary K⫹ restriction did not affect the daily mean expression of Nrf2 (Fig. 8A), but did affect its circadian pattern. F8 Under normal conditions, Nrf2 displayed a circadian expression profile with a small but significant amplitude (15%; Table 1 and Fig. 8B) and an acrophase around ZT8, which preceded the peak of HKA2 expression by 4 h. Dietary K⫹ restriction reversed the circadian expression of Nrf2, which then displayed a peak at

Figure 7. Circadian variation of progesterone and renal PR under normal diet conditions and low-K⫹ diet conditions. A, B) Plasma progesterone concentrations were measured at 6 different times in mice fed normal chow (A) or a low-K⫹ diet (B); n ⫽ 5. Intergroup comparisons (1-way ANOVA) showed significant variability in both conditions (P⬍0.01). C) Daily mean expression of PR in kidney of mice receiving a normal-K⫹ (NK) or a low-K⫹ (LK) diet relative to cyclophilin A (PPIA) expression; n ⫽ 5. **P ⬍ 0.01; nonpaired Student’s t test. D, E) Expression of PR/PPIA, relative to its daily mean, was measured at 6 different times of the day in WT mice under normal (D) or low-K⫹ conditions (E); n ⫽ 5. Dotted line represents daily mean value (equal to 1). Dot size is proportional to se. Intergroup comparisons (1-way ANOVA) showed significant variability (P⬍0.01) for both conditions. Results are shown as means ⫾ se.


SALHI ET AL.


Figure 8. Circadian variation of Nrf2 under normal diet conditions and low-K⫹ diet conditons. A) Mean expression of Nrf2 relative to PPIA expression over 24 h (daily mean) and during the periods of activity (ZT16, ZT20, and ZT0) and rest (ZT4, ZT8, and ZT12) in WT mice fed normal K⫹ (NK; open bars) and low-K⫹ (LK; shaded bars) diets; n ⫽ 15. *P ⬍ 0.05, **P ⬍ 0.01; nonpaired Student’s t test. B, C) expression of Nrf2/PPIA, relative to its daily mean, was measured at 6 different times of the day in WT mice fed normal (B) or low-K⫹ diets (C); n ⫽5. Dotted line represents daily mean value (equal to 1).Dot size is proportional to se. Intergroup comparisons (1-way ANOVA) showed significant variability (P⬍0.01) for both conditions. Results are shown as means ⫾ se.

ZT18 (Fig. 8C). Interestingly, as under normal conditions, this peak occurred again 4 h before the HKA2 and PR acrophases.

DISCUSSION Potassium homeostasis is mainly dependent on the ability of the kidney to adapt its K⫹ excretion to the daily K⫹ intake (3). Many decades ago, it was shown that urinary K⫹ excretion follows a circadian profile, with the largest loss during the period of activity, whereas it is slowed during the rest period, when K⫹ intake is low or null (16, 19). In this study, we established the circadian expression of a renal ion transporter, HAK2, its contribution to the circadian urinary excretion of K⫹ , and its effect on the stability of plasma K⫹ values during the day. ⴙ

Circadian regulation of K balance Despite the repeated description of a circadian profile for urinary K⫹ excretion, the molecular actors involved in this process have remained unknown until now. The robust circadian variation of urinary K⫹ excretion was shown to be sensitive to amiloride (20), indicating the involvement of the coupled electrogenic pathway that mediates Na⫹ reabsorption by ENaC and K⫹ secretion RENAL H,K-ATPase REGULATES CIRCADIAN K⫹ BALANCE

by ROMK. Recent advances have contributed to the understanding of this system by demonstrating the circadian expression of the ␣ subunit of ENaC channel, as well as some of its regulators (12, 14). ROMK expression is also directly under the regulation of the circadian clock with an increased level during the activity period (ZT20; refs. 12, 27). In searching for other transporters that may contribute to this process, we analyzed the data provided by Zuber et al. (12), focusing mainly on K⫹ transporters, and found that the HKA2 transcript may display a circadian expression. However, this transcript did not meet the cutoff criteria chosen by the researchers and was therefore not retained as a circadian gene candidate in their study. When we directly investigated the hypothesis of the circadian expression of the HKA2, we found that it indeed follows a circadian rhythm that is abolished in clock-deficient mice and that this circadian profile is phase-shifted under low-K⫹ diet conditions. The amplitude of variation around the daily mean of expression is small, as expected by the transcriptomic data of Zuber et al. (12), but significant. As mentioned above, under the normal diet conditions, the excretion of K⫹ during the activity period is mainly dependent on ROMK channel activity, with HKA2 exerting no measurable effect, since total urine excretion of K⫹ is similar in WT and HKA2-null mice (Table 3). T3 However, during the rest period, the increased expression of HKA2 is concordant with the necessity to reduce the urinary loss of K⫹ when K⫹ intake is also reduced. Inability to express this transporter during this period almost doubles the urinary K⫹ loss. As shown in Table 3, the urinary K⫹ excretion is always the major component of the total K⫹ loss (whatever the period, the strains, or the diet that we take into account). For instance, the difference in urinary K⫹ loss during the rest period between WT and HKA2-null mice under normal-K⫹ diet conditions is 47 ␮mol/d and only 16 ␮mol through fecal loss. It is therefore likely that the renal defects of HKA2-null mice contribute to a large part of the decrease plasma K⫹ value observed during the rest period in normal-K⫹ diet conditions. The renal adaptation to the activity/rest periods enables the TABLE 3. Urinary and fecal K⫹ excretion in mice fed a normal or a low-K⫹ diet K⫹ excretion (␮mol/12 h)

NK diet Urinary Fecal LK diet Urinary Fecal

WT

KO

Activity

Rest

Activity

233 ⫾ 25 54 ⫾ 6

77 ⫾ 17 8⫾1

282 ⫾ 9 54 ⫾ 6

23 ⫾ 1.5 1 ⫾ 0.3

6.5 ⫾ 1.3 1.7 ⫾ 0.7

26 ⫾ 1* 1.1 ⫾ 0.6

Rest

124 ⫾ 18** 24 ⫾ 4** 6 ⫾ 0.7 2 ⫾ 0.4

Results are shown as means ⫾ se (n ⫽ 6 mice). Total K⫹ contents were measured on urine and fecal samples collected during the activity and the rest period from WT or HKA2-null (KO) mice. NK, normal K⫹; LK, low K⫹. *P ⬍ 0.05, **P ⬍ 0.01 vs. WT; nonpaired Student’s t-test.

7


kidney to maintain a constant plasma K⫹ value throughout the day. In this study, we demonstrated that the absence of HKA2 is associated with a urinary K⫹ leak during the rest period, indicating a transitory inefficient adaptation of the kidney, leading to a decrease in the plasma K⫹ value. Under low-K⫹ diet conditions, ROMK channels are inhibited (for review, see ref. 28), which may reduce the amplitude of variation in urinary K⫹ excretion during the 24-h period, as we observed in the present study. Under these conditions, where K⫹ is absent from food, the cyclic expression of HKA2 is reversed in the kidney compared to normal conditions. We propose that this inversion of the HKA2 expression acts to limit K⫹ loss during the activity period, and the inability to do so leads to a higher loss of urinary K⫹ during the activity period, which is correlated with a lower plasma K⫹ value. Under low-K⫹ diet conditions, the fecal excretion is extremely low (Table 3), and we probably reach the limit of sensitivity of the methods we used (from collection to K⫹ measurement). The results show, however, that, if not negligible, fecal excretion is always lower than urinary K⫹ loss. Here again, the main determinant of the plasma K⫹ value is therefore the renal ability to excrete more or less K⫹. Meneton et al. (2), who described the HKA2-null mice for the first time, observed that the fecal excretion of K⫹ was important when those mice were fed a low-K⫹ diet for 18 d, a period of time much longer than the one we have chosen (only 5 d). It is therefore possible that when the mice are fully depleted, the urinary excretion falls to a value similar to that of the fecal excretion. To equilibrate the ins and the outs is not enough, to be on time matters

F9

The basic requirement for a homeostatic system is to excrete the amount of a compound that corresponds exactly to its intake level in order to avoid rapid and lasting variation of the internal milieu. In the present study, we showed that another feature of an efficient homeostatic system is the way excretion is distributed throughout the day. Regarding K⫹ homeostasis, the question is therefore not only how much K⫹ is excreted but also when it is excreted. As illustrated in Fig. 9, under normal conditions, WT mice perfectly equilibrate their K⫹ outputs to their K⫹ input in both periods of the day. However, HKA2-null mice are not able to excrete K⫹ on time, leading to discrepancies between K⫹ intake and K⫹ excretion during activity and rest periods. These discrepancies then lead to variations of the internal milieu K⫹ concentration. When the HKA2deficient mice were first investigated, the observation that the 24-h K⫹ excretion was not affected either under normal conditions or under K⫹ restriction, compared to WT littermates, was puzzling and challenged the physiological relevance of such transporters in the kidney (2). In 2008, Wingo and colleagues (29) investigated the effect of HKA2 and HKA1 on the ability of intercalated cells to extrude protons by mea8

Vol. 26

July 2012

Figure 9. Circadian regulation of K⫹ balance under normal conditions. Schematic illustration of the regulation of the K⫹ homeostasis in WT (A) and HKA2-null mice (B) during activity and rest periods. Exact values for K⫹ excretion are listed in Table 3. In WT mice, the total excretion of K⫹ in both periods correlates with the K⫹ intakes (in⫽out). However, excretion of K⫹ in HKA2-null mice in the different periods of the day does not cope with the K⫹ intake in the same periods (in⬎out during activity or in⬍out during rest), leading to instability of the plasma K⫹ value.

suring fluxes across the CCD in KO mice for both genes using in vitro microperfusion techniques. They concluded that HKA1 and HKA2 contribute to acid excretion by the CDDs under normal conditions and are, therefore, both functional in microperfused mouse CCDs. So, on the one hand, the functionality of HKA2 in the kidney was proven, but on the other, evidence for its physiological effect was lacking. In the present study, we resolved this paradox by demonstrating that the presence of renal HKA2 is necessary for the efficient adaptation of the kidney to the day-night or activity-rest cycles. Possible mechanisms for HKA2 circadian variation The molecular mechanism that underlies the circadian expression of HKA2 and its reversion during dietary K⫹ restriction is not yet defined and will need further investigation. Here, we showed that progesterone, which participates in the renal adaptation to dietary K⫹ restriction (7), is unlikely to be at the origin but is rather a target of this regulatory system. Indeed, the circulating progesterone level undergoes a cyclic variation with a profile that is not modified under low-K⫹ diet conditions. As for PR, its variation of expression parallels and does not precede that of HKA2 in all dietary conditions. We may therefore speculate that a common circadian signal regulates both HKA2 and PR expression in the kidney. Interestingly, the renal expression of Nrf2, another transcriptional regulator of HKA2, displays a circadian profile that is modified by dietary K⫹ restriction. This factor is stimulated by oxidative stress and reactive oxygen species, which are known to be increased by a low-K⫹ diet and known to induce the inhibition of ROMK (30). Our results


SALHI ET AL.


therefore suggest that Nrf2 could be one of the molecular triggers that determine the time at which HKA2 expression peaks. In addition, overall HKA2 expression could be amplified by progesterone. Further investigations will be needed to identify all the molecular participants that lead to circadian expression of HKA2 and its modification during dietary K⫹ restriction. All together, this work provides the first formal identification of a renal ion transporter involved in the circadian excretion of K⫹ and in the stability of plasma K⫹ values. The authors thank Dr. Nikolai N. Modyanov (University of Toledo, Toledo, OH, USA) for his generous gift of anti-HKA2 antibody. The authors thank Alain Doucet and Aurélie Edwards for helpful reading of the manuscript and fruitful discussion and Lydie Cheval for technical help. This study was supported by grants from the French Society of Nephrology (G.C. and A.S.) and by the program of the Transatlantic Network on Hypertension of the Leducq Fondation (G.C.). The authors declare no conflicts of interest.

13.

14.

15.

16. 17.

18. 19. 20.

REFERENCES 1. 2.

3. 4.

5.

6.

7.

8.

AQ: 3 9. 10.

11. 12.

Wang, W. H., and Giebisch, G. (2008) Regulation of potassium (K) handling in the renal collecting duct. Pflügers Arch. 458, 157–168 Meneton, P., Schultheis, P. J., Greeb, J., Nieman, M. L., Liu, L. H., Clarke, L. L., Duffy, J. J., Doetschman, T., Lorenz, J. N., and Shull, G. E. (1998) Increased sensitivity to K⫹ deprivation in colonic H,K-ATPase-deficient mice. J. Clin. Invest. 101, 536 –542 Giebisch, G., Krapf, R., and Wagner, C. (2007) Renal and extrarenal regulation of potassium. Kidney Int. 72, 397–410 Codina, J., Opyd, T. S., Powell, Z. B., Furdui, C. M., Petrovic, S., Penn, R. B., and Dubose, T. D., Jr. (2011) pH-dependent regulation of the ␣-subunit of H⫹-K⫹-ATPase (HK␣2). Am. J. Physiol. Renal Physiol. 301, F536 –F543 El Moghrabi, S., Houillier, P., Picard, N., Sohet, F., Wootla, B., Bloch-Faure, M., Leviel, F., Cheval, L., Frische, S., Meneton, P., Eladari, D., and Chambrey, R. (2010) Tissue kallikrein permits early renal adaptation to potassium load. Proc. Natl. Acad. Sci. U. S. A. 107, 13526 –13531 Codina, J., Li, J., and Dubose, T. D., Jr. (2005) CD63 interacts with the carboxy terminus of the colonic H⫹-K⫹-ATPase to decrease [corrected] plasma membrane localization and 86Rb⫹ uptake. Am. J. Physiol. Cell Physiol. 288, C1279 –C1286 Elabida, B., Edwards, A., Salhi, A., Azroyan, A., Fodstad, H., Meneton, P., Doucet, A., Bloch-Faure, M., and Crambert, G. (2011) Chronic potassium depletion increases adrenal progesterone production that is necessary for efficient renal retention of potassium. Kidney Int. 80, 256 –262 Lee, C. B., Lee, Y. S., Lee, J. Y., Lee, S. E., and Ahn, K. Y. (2012) Nrf2 and Sp family synergistically enhance the expression of ion transporters in potassium-depleted conditions. J. Nephrol. In press Moore-Ede, M. C. (1986) Physiology of the circadian timing system: predictive versus reactive homeostasis. Am. J. Physiol. Regul. Integr. Comp. Physiol. 250, R737–R752 Wu, T., Ni, Y., Dong, Y., Xu, J., Song, X., Kato, H., and Fu, Z. (2010) Regulation of circadian gene expression in the kidney by light and food cues in rats. Am. J. Physiol. Regul. Integr. Comp. Physiol. 298, R635–R641 Yamamoto, T., Nakahata, Y., Soma, H., Akashi, M., Mamine, T., and Takumi, T. (2004) Transcriptional oscillation of canonical clock genes in mouse peripheral tissues. BMC Mol. Biol. 5, 18 Zuber, A. M., Centeno, G., Pradervand, S., Nikolaeva, S., Maquelin, L., Cardinaux, L., Bonny, O., and Firsov, D. (2009)

RENAL H,K-ATPase REGULATES CIRCADIAN K⫹ BALANCE

21.

22.

23.

24.

25. 26.

27. 28. 29.

30.

Molecular clock is involved in predictive circadian adjustment of renal function. Proc. Natl. Acad. Sci. U. S. A. 106, 16523–16528 Nikolaeva, S., Pradervand, S., Centeno, G., Zavadova, V., Tokonami, N., Maillard, M., Bonny, O., and Firsov, D. (2012) Pleiotropic regulation of renal sodium handling by the circadian clock. J. Am. Soc. Nephrol. In press Gumz, M. L., Stow, L. R., Lynch, I. J., Greenlee, M. M., Rudin, A., Cain, B. D., Weaver, D. R., and Wingo, C. S. (2009) The circadian clock protein Period 1 regulates expression of the renal epithelial sodium channel in mice. J. Clin. Invest. 119, 2423–2434 Gumz, M. L., Cheng, K. Y., Lynch, I. J., Stow, L. R., Greenlee, M. M., Cain, B. D., and Wingo, C. S. (2010) Regulation of ␣ENaC expression by the circadian clock protein Period 1 in mpkCCD(c14) cells. Biochim. Biophys. Acta 1799, 622–629 Moore-Ede, M. C., and Herd, J. A. (1977) Renal electrolyte circadian rhythms: independence from feeding and activity patterns. Am. J. Physiol. 232, F128 –F135 Moore-ede, M. C., Schmelzer, W. S., Kass, D. A., and Herd, J. A. (1977) Cortisol-mediated synchrinization of circadian rhythm in urinary potassium excretion. Am. J. Physiol. Regul. Integr. Comp. Physiol. 233, R230 –R238 Rabinowitz, L., Berlin, R., and Yamauchi, H. (1987) Plasma potassium and diurnal cyclic potassium excretion in the rat. Am. J. Physiol. 253, F1178 –F1181 Rabinowitz, L., Wydner, C. J., Smith, K. M., and Yamauchi, H. (1986) Diurnal potassium excretory cycles in the rat. Am. J. Physiol. 250, F930 –F941 Dalton, P. L., and Rabinowitz, L. (1989) Amiloride effect on diurnal cyclic Na and K excretion in rats. Am. J. Physiol. Regul. Integr. Comp. Physiol. 256, R510 –R517 Debruyne, J. P., Noton, E., Lambert, C. M., Maywood, E. S., Weaver, D. R., and Reppert, S. M. (2006) A clock shock: mouse CLOCK is not required for circadian oscillator function. Neuron 50, 465–477 Fila, M., Brideau, G., Morla, L., Cheval, L., Deschenes, G., and Doucet, A. (2011) Inhibition of K⫹ secretion in the distal nephron in nephrotic syndrome: possible role of albuminuria. J. Physiol. 589, 3611–3621 Azroyan, A., Morla, L., Crambert, G., Laghmani, K., Ramakrishnan, S., Edwards, A., and Doucet, A. (2012) Regulation of pendrin by cAMP: possible involvement in beta adrenergicdependent NaCl retention. [E-pub ahead of print] Am. J. Physiol. Renal Physiol. doi: 10.1152/ajprenal.00403 Pestov, N. B., Korneenko, T. V., Shakhparonov, M. I., Shull, G. E., and Modyanov, N. N. (2006) Loss of acidification of anterior prostate fluids in Atp12a-null mutant mice indicates that nongastric H-K-ATPase functions as proton pump in vivo. Am. J. Physiol. Cell Physiol. 291, C366 –C374 DeBruyne, J. P., Weaver, D. R., and Reppert, S. M. (2007) CLOCK and NPAS2 have overlapping roles in the suprachiasmatic circadian clock. Nat. Neurosci. 10, 543–545 Grimont, A., Bloch-Faure, M., El Abida, B., and Crambert, G. (2009) Mapping of sex hormone receptors and their modulators along the nephron of male and female mice. FEBS Lett. 583, 1644 –1648 Firsov, D., Tokonami, N., and Bonny, O. (2012) Role of the renal circadian timing system in maintaining water and electrolytes homeostasis. Mol. Cell. Endocrinol. 349, 51–55 Wang, W. H. (2006) Regulation of ROMK (Kir1.1) channels: new mechanisms and aspects. Am. J. Physiol. Renal Physiol. 290, F14 –F19 Lynch, I. J., Rudin, A., Xia, S. L., Stow, L. R., Shull, G. E., Weiner, I. D., Cain, B. D., and Wingo, C. S. (2008) Impaired acid secretion in cortical collecting duct intercalated cells from H-K-ATPase-deficient mice: role of HK␣ isoforms. Am. J. Physiol. Renal Physiol. 294, F621–F627 Babilonia, E., Wei, Y., Sterling, H., Kaminski, P., Wolin, M., and Wang, W. H. (2005) Superoxide anions are involved in mediating the effect of low K intake on c-Src expression and renal K secretion in the cortical collecting duct. J. Biol. Chem. 280, 10790 –10796 Received for publication November 2, 2011. Accepted for publication March 19, 2012.

9

AQ: 4

AQ: 5

JOBNAME: AUTHOR QUERIES PAGE: 1 SESS: 1 OUTPUT: Mon Mar 26 04:46:27 2012 /tapraid4/z38-faseb/z38-faseb/z3800712/z388810d12z

AUTHOR QUERIES AUTHOR PLEASE ANSWER ALL QUERIES AQ1— Please ensure that italics are used correctly and consistently to distinguish genes from proteins, etc., per CBE style, throughout the abstract and text. AQ2— Please confirm or revise all instances where supplier information has been added AQ3— Ref. 8: Please update if possible. If article is an e-pub ahead of print, please include doi. AQ4 — Ref. 13: Please update if possible. If article is an e-pub ahead of print, please include doi. AQ5— Ref. 23: Please update if possible. AQ6 — Figs. 1, 4: PPIA explained correctly in captions as edited? Please confirm or revise.

1