Rare mutations in SLC12A1 and SLC12A3 protect against hypertension by reducing the activity of renal...

C

CE: Namrta; HJH/202096; Total nos of Pages: 9;

HJH 202096

Original article 1

Rare mutations in SLC12A1 an
d SLC12A3 protect againsthypertension by reducing the activity of renalsalt cotransportersRocıo Acunaa, Lilia Martınez-de-la-Mazaa, Jose Ponce-Coriaa, Norma Vazqueza,Penelope Ortal-Vitea, Diana Pacheco-Alvarezb, Norma A. Bobadillaa andGerardo Gambaa
Objectives Screening for variants in SLC12A1 and

SLC12A3 genes, encoding the renal NaR:ClS (NCC) and

Naþ:Kþ:2ClS (NKCC2) cotransporters, respectively, in 3125

members of the Framingham Heart Study (FHS) revealed

that carrying a rare mutation in one of these genes was

associated with a significant reduction in blood pressure, in

the risk of arterial hypertension, and of death due to

cardiovascular disease. Because near 60% of the rare

mutations identified have not been related to Bartter’s or

Gitelman’s disease, the consequence of such mutations on

cotransporter activity is unknown.

Methods We used the heterologous expression system of

Xenopus laevis oocytes, microinjected with wild-type or

mutant NCC or NKCC2 cRNAs, to examine the effect of

these inferred NCC and NKCC2 mutations on the

cotransporters’ functional properties. Cotransporter activity

was defined as the diuretic-sensitive radioactive tracer

uptake and response to known modulators was assessed.

Results Basal NCC activity was significantly reduced in all

NCC mutants and, excluding NCC-S186F, response to

WNK3, WNK4, or intracellular chloride depletion was

conserved. Similarly, basal activity was reduced in six out of

nine NKCC2 mutants and response to WNK3 was

maintained. No effect on protein expression was seen,

except for NCC-S186F, which was significantly reduced.

Conclusions The rare NCC or NKCC2 mutations

found in the FHS significantly reduced the basal

opyright © Lippincott Williams & Wilkins. Unauth

0263-6352 � 2011 Wolters Kluwer Health | Lippincott Williams & Wilkins

activity of the cotransporters. This observation supports

that even a small, but chronic reduction of NCC or

NKCC2 function results in a lower blood pressure and

decreased risk of hypertension in otherwise healthy

individuals in the general population. J Hypertens 29:000–

000 Q 2011 Wolters Kluwer Health | Lippincott Williams &

Wilkins.

Journal of Hypertension 2011, 29:000–000

Keywords: blood pressure, distal convoluted tubule, genetic variation, NCC,NKCC2, thiazide diuretics, thick ascending limb

Abbreviations: DCT, distal convoluted tubule; FHS, Framingham HeartStudy; GWAS, genome-wide association studies; NCC, the renal Naþ:ClS

cotransporter; NKCC2, the renal specific Naþ:Kþ:2ClS cotransporter; ROMK,rat outer medulla potassium channel; SLC12A1, solute carrier cotransporterfamily 12 member A1; SLC12A3, solute carrier cotransporter family 12member A3; TAL, thick ascending limb of Henle’s loop; WNK3, with no lysinekinase 3; WNK4, with no lysine kinase 4

aMolecular Physiology Unit, Instituto Nacional de Cardiologıa Ignacio Chavez,Instituto Nacional de Ciencias Medicas y Nutricion Salvador Zubiran, and Institutode Investigaciones Biomedicas, Universidad Nacional Autonoma de MexicoTlalpan and bEscuela de Medicina, Universidad Panamericana, Mexico City,Mexico

Correspondence to Gerardo Gamba, MD, PhD, Molecular Physiology Unit, Vascode Quiroga No. 15, Tlalpan 14000, Mexico City, MexicoTel: +5255 5513 3868; fax: +5255 5655 0382;e-mail: [email protected] or [email protected]

Received 22 July 2010 Revised 20 September 2010Accepted 21 October 2010

IntroductionSodium reabsorption by renal transport mechanisms has a

critical role in long-term blood pressure control, which is

illustrated by rare monogenic syndromes altering salt

handling in the kidney and considerably affecting blood

pressure regulation [1,2]. The bumetanide-sensitive

Naþ:Kþ:2Cl� cotransporter (NKCC2) in the apical mem-

brane of the thick ascending limb of Henle’s loop (TAL)

and the thiazide-sensitive Naþ:Cl� cotransporter (NCC)

in the distal convoluted tubule (DCT) are two major

salt reabsorption pathways in the nephron, accounting

for the reabsorption of 35% of the salt filtered at the

glomerulus. The activity of both cotransporters has an

effect on final urinary salt excretion, consequently influen-

cing long-term blood pressure levels. Loss-of-function

mutations in the NKCC2 gene, SLC12A1, or in the

NCC gene, SLC12A3, cause type I Bartter’s syndrome

and Gitelman’s syndrome, respectively, with both diseases

featuring arterial hypotension along with electrolyte

abnormalities.

The renal salt wasting and hypotension accompanying

these two rare autosomic recessive diseases represent a

lower extreme in the spectrum of salt reabsorption and

blood pressure, and the study of these blatant phenotypes

has provided critical insight into the role of salt transport

mechanisms in blood pressure maintenance. The more

subtle phenotype associated with the heterozygous state

of either of these Mendelian diseases has been less

examined, although carriers of Gitelman’s syndrome

orized reproduction of this article is prohibited.

DOI:10.1097/HJH.0b013e328341d0fd

mailto:[email protected]

mailto:[email protected]

http://dx.doi.org/10.1097/HJH.0b013e328341d0fd

C


HJH 202096

2 Journal of Hypertension 2011, Vol 00 No 00

mutations have been reported to display lower blood

pressure than controls [3,4]. Recently, Ji et al. [5] studied

the effect of heterozygosity for rare independent

mutations in salt-handling proteins on blood pressure.

This group developed criteria to identify NCC, NKCC2

and ROMK (a potassium channel involved in renal

sodium reabsorption) rare variants under purifying selec-

tion, by searching for low-frequency mutations in amino

acids under complete phylogenetic conservation, which

were then corroborated by protein prediction programs.

By applying these criteria to coding DNA sequence

variants in 3125 participants of the Framingham Heart

Study (FHS), 49 heterozygous carriers (representing

1.5% of the studied population) of 30 different NCC,

NKCC2, or ROMK mutations were detected. As a group,

carriers were found to have lower blood pressure than the

general population and their noncarrier siblings, as well as

a decreased risk of developing hypertension and

mortality from cardiovascular disease.

Out of 30 variants described, 18 (five in NCC and nine in

NKCC2) had not been previously reported as disease-

causing mutations and, hence, are inferred to reduce

protein activity by affecting the cotransporters’ primary

structure. Carriers for these variants had a similar effect on

blood pressure and hypertension prevalence as carriers

harboring biochemically proven loss-of-function muta-

tions in the same study [5], suggesting, indeed, decreased

activity in cotransporters with inferred mutations. None-

theless, because the effect of these variants on the cotran-

sporters’ functional properties was not analyzed, experi-

mental testing is necessary to demonstrate that these

inferred mutations impact cotransporter activity [6]. If

so, this may be evidence that, in the general population,

slightly decreased renal salt reabsorption influences blood

pressure regulation. Additionally, the functional and bio-

chemical corroboration of the criteria used by Ji et al. [5] to

detect rare independent mutations in the general popu-

lation would support the use of this methodology in

genomic studies of polygenic diseases [7].

MethodsSite-directed mutagenesisMutations describedby Ji et al. were reproduced in ratNCC

and rat NKCC2 by site-directed mutagenesis (Quik-

Change; Stratagene) on corresponding amino acids, using

previously described rNCC/pSPORT1 and rNKCC2/

pSPORT1 as a template [8]. All primers were custom made

(Sigma) and all mutations were corroborated by automatic

DNA sequencing. To avoid undesired mutations, a fully

sequenced restriction fragment harboring the wanted

mutation was subcloned into wild-type NCC or NKCC2.

In-vitro rNCC and rNKCC2 cRNA translationTo prepare cRNA, all rNCC and rNKCC2 clones were

digested at the 30-end using Not I and Xba I (Invitrogen),

respectively. cRNA was transcribed in vitro using a T7

opyright © Lippincott Williams & Wilkins. Unautho

RNA polymerase mMESSAGE kit (Ambion). Transcrip-

tion product integrity was confirmed on agarose gels, and

concentration was determined by absorbance reading at

260 nm (DU 640; Beckman, Fullerton, California, USA)

and by densitometric analysis of the corresponding band

in ethidum bromide-stained agarose gels. cRNA was

stored frozen in aliquots at �808C until use.

Xenopus oocytes expression systemIn the present study, we used the heterologous expression

system in Xenopus laevis oocytes to assess the effect of rare

mutations in NCC or NKCC2 on the activity and some

functional properties of the cotransporters. This expres-

sion system has been shown to be an excellent tool for a

robust and reproducible expression of NCC and NKCC2

in our hands [8–18] and in other laboratories [19–24]. In

contrast, both NCC and NKCC2 functional expression in

transfected mammalian cells has been unsuccessful in

many laboratories and the few reports obtained [25,26],

exhibit a low level of activity over basal uptake and thus,

are not appropriate for detection of subtle differences

between wild-type and mutant cotransporter activity.

Xenopus laevis frog oocytes were surgically harvested from

anesthetized adult females under 0.17% tricaine and

incubated for 1 h under vigorous shaking in calcium-free

frog Ringer-ND96 (mmol/l: 96 NaCl, 2 KCl, 1 MgCl, and

5 HEPES/Tris, pH 7.4), supplemented with collagenase

B (2 mg/ml). The oocytes were manually defolliculated

and incubated overnight in ND96 at 188C containing

5 mg/100 ml of gentamicin. The following day, mature

oocytes were injected with 50 nl of water with or without

cRNA from wild-type or mutant rNCC at a concentration

of 0.5 mg/ml (i.e. 25 ng cRNA/oocyte) or rNKCC2 at

0.25 mg/ml (i.e. 12.5 ng cRNA/ooctye). After injection,

oocytes were incubated for 3 days at 188C or, when

specified, for 2 days at 298C in ND96 with gentamicin.

FunctionalanalysisoftheNaþ:Cl�cotransporterwasdeter-

mined by assessing tracer 22Naþ uptake (New England

Nuclear) in groups of at least 10 oocytes. 22Naþ uptake was

measured using our usual protocol: a 30-min incubation

period in an isotonic Kþ and Cl�-free medium (mmol/l: 96

Naþ gluconate, 6.0 Ca2þ gluconate, 1.0 Mg2þ gluconate,

5 HEPES/Tris, pH 7.4) with 1 mmol/l ouabain, 100 mmol/l

bumetanide, and 100 mmol/l amiloride, followed by a 60-

min uptake in a Kþ-free isotonic medium. The isotonic

medium contained (in mmol/l) 40 NaCl, 56 N-methyl-D-

glucamine(NMDG)-Cl,1.8CaCl2,1MgCl2,5HEPES,pH

7.4,and1 mmol/louabainaswellas100 mmol/lbumetanide,

100 mmol/l amiloride,and2.5 mCi 22Naþ. Todetermine the

thiazide-sensitive portionof theobserved uptake, a parallel

group of oocytes for each experimental group was exposed

to the same uptake protocol in the presence of 100 mmol/l

metolazone in both incubation periods.

Functional analysis of the Naþ:Kþ:2Cl� cotransporter

was measured by assessing 86Rbþ uptake in groups of

rized reproduction of this article is prohibited.

C


HJH 202096

Characterization of rare NCC and NKCC2 mutants Acuna et al. 3

at least 10 oocytes according to our usual protocol for

NKCC2: a 30-min incubation period in the same isotonic

Kþ and Cl� -free medium described previously, followed

by a 60-min uptake in an isotonic medium containing

Naþ, Kþ and Cl� (mmol/l: 90 NaCl, 10 KCl, 1.8 CaCl2, 1

MgCl, and 5 HEPES/Tris, pH 7.4 and 1 mmol/l ouabain).

As for NCC, to determine the bumetanide-sensitive

portion of the observed uptake, for every experimental

group a corresponding group of oocytes was exposed to

the uptake protocol in the presence of 100 mmol/l bume-

tanide in both periods of incubation. Because X. laevisoocytes express an endogenous NKCC1-type Naþ-Kþ-

2Cl– cotransporter, every experiment included control

water-injected oocytes with and without bumetanide.

All uptakes were performed at 328C. At the end of the

uptake period, oocytes were washed five times in ice-cold

uptake solution without the isotope to remove extracellu-

lar fluid tracer. After the oocytes were dissolved in 10%

SDS, tracer activity was determined for each oocyte by

b-scintillation counting.

Western blottingWestern blotting was used to compare wild-type and

mutant rNCC and rNKCC2 total protein in cRNA-

injected oocytes. In brief, groups of 15 oocytes injected

with water or cRNA were homogenized in 5 ml/oocyte of

homogenization buffer, centrifuged twice at 100g for

10 min at 48C, and the supernatant was recollected. Oocyte

protein (three oocytes/lane) was heated in sample buffer

containing 6% SDS, 15% glycerol, 0.3% bromophenol

blue, 150 mmol/l Tris, pH 7.6, and 2% b-mercaptoethanol

and resolved by SDS-PAGE (7.5%). The proteins were

transferred to a polyvinylene difluoride membrane (Amer-

sham Pharmacia Biotech) at 10 V for 60 min in a buffer

containing 25 mmol/l Tris, 190 mmol/l glycine, 0.1% SDS,

and 20% (vol/vol) methanol. Prestained Rainbow markers

(Amersham) were used as molecular mass standards.

SNAP i.d. protein detection system (Millipore) was used

for blocking, antibody addition and membrane washing.

Bands were detected by using the Immun-Star Chemilu-

minescent Protein Detection System (Bio-Rad).

Statistical analysisStatistical significance is defined as two-tailed P< 0.05,

and the results are presented as means �SE of normal-

ized data, taking wild-type NCC or NKCC2 basal activity

as 100%. The significance of the differences between

groups was tested by one-way ANOVA with Dunn’s

correction for multiple comparisons.

The study was approved by an institutional review com-

mittee.

ResultsThe activity of NCC mutants is reducedJi et al. [5] described 15 independent NCC mutations

associated with lower hypertension prevalence in the


FHS cohort, only five of which had not been previously

studied as biochemical loss-of-function or as Gitelman’s

disease-causing mutations. The putative localization of

these mutations is depicted in the topological model of

NCC in Fig. 1a. Because of the high degree of identity at

the protein level between human and rat NCC (near to

90%), with all mutations involving conserved residues,

along with our previous experience studying mutations

causing Gitelman’s syndrome in rat NCC [27], we gener-

ated and characterized these independent mutations in

rat NCC.

A representative experiment of functional expression in

X. laevis oocytes injected with wild-type or mutant NCC

cRNA is presented as the uptake of 22Naþ in the absence

or presence of metolazone in Supplementary Figure S1,

http://links.lww.com/HJH/A58. Several experiments

were performed and NCC function is reported as meto-

lazone-sensitive 22Naþ uptake, taking wild-type NCC

uptake as 100% (Fig. 1b). Only NCC mutation S186F

exhibited no significant activity, whereas the four remain-

ing mutants (L153F, A230T, F493L and G777E) had a

lower, albeit significant function, although it varied

among the different studied mutants. As shown in

Fig. 1b, the basal activity of NCC mutations is: G777E

(78%) > L153F (59%) > F493L (46%) > A230T (26%)

> S186F (0%). To further support our results, we repro-

duced mutation G779E, located in a less conserved

region of the NCC C-terminal domain, both in rat and

in human NCC with a similar outcome (Supplemental

Figure S2, http://links.lww.com/HJH/A59).

Wild-type and mutant NCC protein expression is similarThe monomeric form of wild-type and mutant NCC was

detected at the expected weight of 110 kDa in the

immunoblot. A representative western blot with total

protein expression measured by densitometry and nor-

malized to wild-type NCC protein is depicted in Fig. 1c.

Total protein expression does not vary substantially

between wild-type NCC and L153F, A230T, F493L,

and G777E mutants, suggesting that decreased protein

expression is not the source of these mutants’ lower

function. In contrast, mutation S186F resulted in a sig-

nificantly lower protein expression when compared to

wild-type NCC. Additionally, wild-type NCC and

mutants L153F, A230T, F493L and G777E show the

characteristic diffuse band visible above the better-

defined protein monomer band at 110 kDa in the immu-

noblot, corresponding to the glycosylated form of the

cotransporter. In this regard, we have previously ascer-

tained that N-glycosylation of NCC at residues N404 and

N424 is necessary for proper folding and protein traffick-

ing to the plasma membrane [16,27]. This glycosylation

band is absent in S186F NCC mutant, implying that the

lack of glycosylation in this variant would prevent its

transport to plasma membrane. In this regard, we have

previously studied NCC disease-causing mutations in


http://links.lww.com/HJH/A58


Copyright © Lippincott Williams & Wilkins. Unautho


HJH 202096


Fig. 1

(a) Topological model proposed for NCC. A central hydrophobicdomain containing 12 putative transmembrane segments is flanked by ashort amino-terminal and a long carboxyl-terminal domain. Mutationsdetected by Ji et al. [5] proven to cause biochemical loss of function orGitelman’s disease (squares) and inferred loss-of-function mutations(circles) are depicted. (b) Functional expression of wild-type and mutantNCC in Xenopus laevis oocytes. The thiazide-sensitive portion of theuptake was computed for each group. Then, the mean value observedin wild-type NCC was taken as 100% and data in the mutant groupswas normalized accordingly. (�) P<0.001 vs. wild-type NCC; n¼40oocytes per bar. (c) Wild-type and mutant NCC protein expression.A representative Western blot of proteins extracted from wild-type andmutant NCC-injected oocytes. Proteins corresponding to four oocytesper lane were resolved in SDS-PAGE gels and transferred to anitrocellulose membrane. Bands were exposed using monoclonalantibodies against FLAG epitope. The graph shows the densitometricanalyses of the Western blot.

which cotransporter activity was not completely abol-

ished [27]. In that study, trafficking toward the plasma

membrane as well as kinetic properties of mutant cotran-

sporters were carefully defined. We observed that no

mutation affected kinetic properties. Since the Km for

Naþ and Cl� in the wild-type cotransporter is below

5 mmol/l and uptake solutions contain up to 90 mmol/l

NaCl, we believe that inferred mutations, comparable to

Gitelman mutations, are unlikely to change NCC affinity

for cotransported ions. In contrast, it was reported by this

same study [27] as well as by De Jong et al. [22], that all

Gitelman mutations with residual activity exhibited a

significant reduction in the amount of NCC in the plasma

membrane, suggesting a defect in the cotransporter’s

trafficking. It is possible that this may also be happening

with NCC harboring the inferred mutations observed in

FHS population.

Wild-type and mutant NCC response to stimulation byWNK3 and intracellular chloride depletion and to WNK4inhibitionSodium reabsorption in the distal nephron is regulated by

the kinases of the WNK [With No lysine (K)] family. We

have previously demonstrated that co-expression of wild-

type NCC with WNK3 in X. laevis oocytes enhances

NCC activity, at least in part by increasing NCC cell

surface expression [18]. Similarly, NCC activity is also

stimulated by intracellular chloride depletion, which can

be accomplished by co-injection of NCC with KCC2, an

isotonically active Kþ:Cl� cotransporter that generates a

constitutively active chloride efflux pathway [28]. On the

contrary, we have also substantiated that WNK4 inhibits

NCC activity by diminishing its cell surface expression

[17]. Thus, in the present study we explored if rare NCC

mutations observed in the FHS population affected NCC

regulation by known modulators. When wild-type NCC

cRNA was co-injected with either WNK3 or KCC2

cRNA, cotransporter basal activity was increased to a

level above 200%, as reported in Fig. 2a. Contrariwise,

addition of WNK4 cRNA to the injection cocktail

resulted in a significant decrease of activity to about

25% (Fig. 2b). These effects were completely lost in

mutant S186F (data not shown), which indicates that, in

addition to its absence of basal activity, S186F does not

respond to NCC stimulators or inhibitors.

In contrast to what occurred with S186F, NCC mutations

with residual function had an increase in activity when

co-expressed with WNK3 or KCC2 to induce intracellular

chloride depletion (Fig. 2a). Thus, although NCC

mutants had a lower basal activity than wild-type

NCC, they responded properly to stimulation by

WNK3 or intracellular chloride depletion (Fig. 2a).

Finally, wild-type NCC and all mutants with residual

function were similarly inhibited by WNK4 (Fig. 2b).

These results denote that mutations with residual func-

tion remain responsive to stimulatory and inhibitory


C


HJH 202096


Fig. 2

300

(a) (b)

Per

cent

age

of N

CC

acti

vity

Per

cent

age

of N

CC

acti

vity

120

100

80

60

40

20

0

250

200

150

100

50

0

NCC

NCC NCC + WNK3 NCC + KCC2

L153F A230T F493L G777ENCC L153F A230T F493L G777E

WNK4 − + − + − + − + − +

* * *

*

**

*

* *

*

Wild-type and mutant NCC response to stimulation by (a) WNK3 or intracellular Cl- depletion and (b) WNK4 inhibition. (�) P<0.001 vs. basalactivity; n¼15 oocytes per bar. (b) When co-expressed with WNK4, wild-type and mutant NCC have a decrease of activity to 25% of basal function(P<0.01 vs. absence of WNK4).

modulators, whereas mutations, in which function is fully

abolished, do not.

Functional expression of wild-type and mutant NKCC2Ten different mutations in NKCC2 associated to lower

blood pressure and hypertension prevalence were found

in the FHS cohort [5]. Only one variant, a frameshift

mutation, had been previously described as causing type

I Bartter’s syndrome. The remaining nine mutations,

represented in Fig. 3a, are located all along the NKCC2

protein. NKCC2 mutations were reproduced in the cor-

responding amino acids of rat NKCC2 and cotransporter

activity was measured in functional expression studies as

the bumetanide-sensitive uptake of 86Rbþ in X. laevisoocytes injected with wild-type or mutant NKCC2

cRNA. NKCC2 harboring mutation T231M, R298W,

L501V, P565H, Y1066C, or P1079A exhibited lower

activity when compared to wild-type NKCC2. However,

P250A and P344L mutants behaved like wild-type

NKCC2 and mutation N395S had an increased activity.

As presented in Fig. 3b, the activity profile was N395S

(172%) > P250A (101%) � wild-type NKCC2 (100%) �P344L (94%)>T231M (73%)> P1079A (58%)>L501V

(52%)> Y1066C (34%)> P565H (15%)>R298W (11%).

Total protein expression was assessed by Western blot

and relative protein expression, as measured by densito-

metry and normalized to wild-type NKCC2 protein, is

depicted in Fig. 3c. As with NCC, there is no substantial

difference between wild-type and mutant NKCC2 total

protein expression implying that the variation observed

in mutant cotransporters’ activity is not due to altered

protein expression, but probably related either to

decreased trafficking to the plasma membrane [27] or

to diminished intrinsic cotransporter activity [23,29].

Mutants P250A, P344L and N395S did not, unlike the

remaining NKCC2 mutations, have decreased activity.

Some protein processing defects have been revealed to


be temperature-sensitive, such as the DF508 mutation in

the cystic fibrosis transmembrane conductance regulator

(CFTR) chloride channel that behaves like wild-type

CFTR at 188C, the standard temperature for incubating

X. laevis oocytes after cRNA injection, but lacks activity

when cells are incubated at higher temperatures [30].

Therefore, we kept oocytes injected with wild-type

NKCC2 or P250A, P344L and N395S mutants’ cRNA

at a higher temperature (298C) for 2 days after injection to

trigger a temperature-sensitive defect. At this tempera-

ture, bumetanide-sensitive 86Rbþ uptake increased in

wild-type NKCC2 as well as in mutant clones, although

no substantial difference was observed between exper-

iments in which oocytes were incubated at 188C and

those at 298C (Supplemental Figure S3A, http://links.

lww.com/HJH/A60). As incubating X. laevis oocytes

above 298C seriously damages these cells, we are unable

to test mutant function at higher temperatures. Con-

sequently, we cannot completely eliminate the possib-

ility that a temperature-sensitive processing defect in

these clones may become apparent at 378C. However,

as no substantial change was observed between cotran-

sporter function at 18 and at 298C, this seems unlikely for

P250A, P344L and N395S mutations.

Interestingly, mutant N395S consistently had an increased

basal activity when compared to wild-type NKCC2. To

determine if the enhanced cotransporter function

depended on mutation of N395 to any amino acid residue

or if it was caused specifically by mutation to a serine, we

studied a new mutant in which asparagine was substituted

by alanine, N395A. When expressed in oocytes incubated

at 188C, mutation N395A did not reproduce the increased

activity seen with N395S (Supplemental Figure S3B,

http://links.lww.com/HJH/A60). In addition, N395A func-

tion was not statistically different from wild-type NKCC2

activity. This suggests that increased activity in N395S is

caused specifically by the substitution of N395 for serine,





C


HJH 202096


Fig. 3

(a) NKCC2 topological model. The structure proposed for NKCC2 isvery similar to that of NCC. Rare NKCC2 mutations linked to lower riskof hypertension are found all along the protein and inferred loss-of-function mutations are indicated. (b) Basal NKCC2 mutant activityexpressed in percentage according to basal wild-type NKCC2 activity.Most mutations (T231M, R298W, L501V, P565H, Y1066C andP1079A) exhibited the expected loss of function when measuredagainst wild-type NKCC2, although mutations P250A, P344L andN395S had an unanticipated effect on cotransporter activity, (�)P<0.001 vs. wild-type NKCC2; (§) P<0.01 vs. wild-type NKCC2;n¼20 oocytes per bar. (c) Western blot of total NKCC2 expressionwith the corresponding densitometry, when wild-type and mutantNKCC2 show a similar quantity of total NKCC2 protein expression.Proteins were extracted from Xenopus oocytes injected with the sameamount of cRNA from the different studied groups.

Fig. 4

400

300

200

100

0NKCC2 T231M P250A R298W P344L N395S L501V P565H Y1066C P107PA

Per

cent

age

of b

asal

act

ivit

y

Wild-type and mutant NKCC2 activity is stimulated by WNK3. Xenopuslaevis oocytes express NKCC1, an endogenous bumetanide-sensitiveNaþ:Kþ:2Cl- cotransporter responsive to WNK3 stimulation [32].Oocytes injected with H2O or H2O with WNK3 were included in theexperiments (data not shown) and endogenous NKCC1 activity (with orwithout WNK3) was deducted from the uptake figures of injectedNKCC2 with or without WNK3. (�) P<0.001 vs. basal activity; n¼15oocytes per bar.

implying that ‘gain-of-function’ mutations are possible

within the SLC12A1 gene.

Response of wild-type and mutant NKCC2 to WNK3We have previously substantiated that WNK3 is a modu-

lator of all members of the SLC12 family [18,31,32].

NKCC2 co-expression with WNK3 in X. laevis oocytes

results in a significant increase of NKCC2 cotransporter

activity. We analyzed if regulation of NKCC2 by WNK3


was conserved in all the studied NKCC2 mutations, as we

did for NCC mutants. As presented in Fig. 4, when co-

expressed with WNK3, wild-type and mutant NKCC2

exhibited an increase in the cotransporter activity,

suggesting that NKCC2 mutants remain responsive to

WNK3 stimulation.

DiscussionIn this study, we analyzed the functional properties of

rare independent mutations in SLC12A1 and SLC12A3genes that have been linked to lower blood pressure and

decreased risk for arterial hypertension [5]. All five NCC

mutations and six out of nine NKCC2 mutations resulted

in a significant reduction of cotransporter activity. We

acknowledge that mutations were studied using the rat

cDNA for NCC and NKCC2. However, we believe that

this introduces no bias as there is a very high identity

between rat and human sequences at the protein level on

these transporters (>90%) and mutations occurred in

residues that are highly conserved among species [33].

In addition, in the only case for NCC in which nearby

residues varies substantially, we introduced the same

mutation in the human cDNA and the observed results

were similar to those obtained with rat NCC cDNA. By

the means of western blot analysis we observed that in

most cases, but one, mutations did not affect protein

expression levels or the glycosylation status of the cotran-

sporters. Thus, reduced activity is not due to a serious

impairment in protein synthesis or stability, leaving the

explanation to a possible problem with protein trafficking

or a reduced intrinsic activity of the cotransporter. In this

regard, previous studies have shown that Gitelman’s

mutation in NCC exhibiting some degree of activity

affect the protein trafficking to the cell surface. Because


C


HJH 202096


the reduction of activity in mutants analyzed in this study

is not as dramatic as in Gitelman’s or Bartter’s mutations,

the analysis of cell surface expression of the cotransporter

would require fine techniques that are beyond the scope

of this study.

Our results support the proposal that even a small, but

chronic reduction of NCC or NKCC2 function entails a

higher urinary salt excretion, resulting in a lower blood

pressure and decreased risk of hypertension in otherwise

healthy individuals in the general population. Thus, our

study complements the observations in the FHS popu-

lation by providing a mechanism to explain the reduced

blood pressure and/or protection against hypertension in

otherwise healthy individuals harboring the inferred

mutations. In addition, there has been a debate for years

about the mechanisms by which therapy with loop or

thiazide diuretics decreases blood pressure [34]. Our

observations suggest that the low blood pressure

observed in FHS participants is due to a reduced activity

of the transporters that are the target for these diuretics,

supporting the proposal that it is the chronic natriuretic

effect of these drugs what causes the decrease in blood

pressure in treated hypertensive patients.

Three NKCC2 mutations did not exhibit loss of function

when compared to wild-type NKCC2. The finding that

NKCC2-N395S is a mutation with increased, instead of

lower, activity may be due to the strategy used by Ji et al.[5] since their methodology sought for mutations in which

amino acid substitutions would affect protein function,

without distinguishing its direction (i.e. between loss and

gain-of-function mutations). Although most mutations

are likely to be disruptive, some may enhance protein

function, which could be the case for NKCC2-N395S.

Supporting this conclusion is our observation that sub-

stituting N395 for another residue did not increase

NKCC2 activity. Blood pressure in carriers of mutation

NKCC2-N395S is slightly lower than expected, which we

believe could be due to chance or to other factors not

taken into account. When patients were analyzed as a

group, the impact of all mutations on blood pressure

levels and protection against hypertension was striking.

However, as only two out of 37 patients in which

mutations in SLC12A1 or SLC12A3 were detected, har-

bored this mutation, the effect of these individuals on the

data of the entire carrier group may be minimal.

The effect of residue substitution in NKCC2-P250A and

NKCC2-P344L may not be visible in X. laevis oocytes

due either to temperature-dependent processing defects

[30], to the lack of the physiological context of the TAL

(for instance, absence of apical-basal polarity in oocytes)

or it may be a false-positive. On one hand, unfortunately,

all attempts until now to express functional NCC or

NKCC2 in polarized epithelia have been unsuccessful.

Consequently, it is still possible that these two mutants

exhibit decreased function in the TAL cells. On the other


hand, the criteria used in the FHS analysis [5] anticipate a

false-positive rate of 10%, which may vindicate NKCC2-

P250A and NKCC2-P344L as tolerated mutations

with no effect on cotransporter function. Remarkably,

five out of five inferred mutations in NCC and six out

nine inferred mutations in NKCC2 indeed exhibited

decreased functional activity. Therefore, even if

NKCC2-P250A and NKCC2-P344L mutants were fully

functional in the TAL, we believe that our results corro-

borate the methodology and criteria used by Ji et al. [5],

which was effective in detecting NCC and NKCC2

mutations with a biochemical effect that correlated with

a physical and epidemiological observation, with a poten-

tial false-positive rate of 10%.

Interestingly, there is a noteworthy difference in levels of

activity between cotransporters harboring mutations pre-

viously described as disease-causing (for NCC as well as

for NKCC2) and cotransporters carrying most of the

inferred mutations analyzed in the present study. Bart-

ter’s or Gitelman’s disease-causing mutations expressed

in X. laevis oocytes exhibit activity below 50% of wild-

type cotransporter function, most frequently between 0

and 25% [19,22,23,27]. In contrast, some of the mutations

characterized in this study display levels of activity well

over 50%. Conceivably, these NCC and NKCC2

mutations with high residual activity may not have been

described in association with Gitelman’s or Bartter’s

syndrome, as they might induce a subclinical phenotype

(such as slightly lower blood pressure or a decreased risk

of hypertension), unlikely to trigger genetic work-up, or

mild forms of disease [35]. Thus, in contrast to the study

of a genetic disease, in which the point of departure is a

symptomatic patient who prompts the discovery of a

specific mutation, the methodology proposed by Ji

et al. [5] progresses from genotype to phenotype, allowing

detection of mutations that might otherwise go unnoticed.

Despite the difference in biochemical activity between

cotransporters harboring proven and inferred mutations,

interestingly there is no disparity in blood pressure

between both groups in the FHS [5]. This is analogous

to what occurs with Gitelman’s syndrome, a highly

heterogeneous disease in which the same NCC mutation

can lead to a wide range of severity in age of onset and

clinical manifestations, including blood pressure levels.

Thus, although rare NCC or NKCC2 mutations have a

clear impact on blood pressure in the general population,

these results hint to the complexity of blood pressure

maintenance. Additional elements, such as epigenetics,

other modulatory genes or environmental factors, may

modify the severity of the phenotype produced by NCC

or NKCC2 dysfunction.

In combination with previously published studies, our

findings support the ‘rare variant, common disease’

model for arterial hypertension; NCC and NKCC2 rare

variants, not associated to disease but exhibiting loss of


C


HJH 202096


function, are linked to a lower blood pressure and

decreased prevalence of arterial hypertension in the

general population. Although certain genetic variants

may be rare, it is expected that each individual may be

carrying an average number of ten or more rare variants in

his or her whole genome [36,37]. For instance, if 1.5% of

3125 studied individuals carried a rare variant in a hyper-

tension-associated gene, when only three genes were

taken into account, it is possible that a significant pro-

portion of the general population may carry a rare variant

in one or more of the multiple genes involved in blood

pressure maintenance.

The ‘rare variant, common disease’ model does not

exclude the possibility that common allelic variants also

play a significant role in the genetics of arterial hyperten-

sion. Indeed, genome-wide association studies (GWAS)

have identified common variants in the atrial and brain

natriuretic peptides with modest effects on blood pres-

sure levels and hypertension prevalence [38]. In another

example, a recent GWAS identified the gene STK39encoding SPAK kinase as a hypertension-associated gene

[39]. Interestingly, SPAK is expressed in the TAL and

DCT [40] and has a significant role in modulating the

activity of NKCC2 and NCC. Therefore, both theories

not only coexist, but may also overlap, as the effect of rare

variants could be modulated by common variants.

In conclusion, corroboration of the method and criteria

used by Ji et al.[5] to identify rare genetic variants altering

protein function and correlating with a clinical obser-

vation in the present study, supports its application for

detecting both alleles causing disease [41], and alleles

with a milder effect on disease predisposition. With

whole-exome sequencing [42], this approach may be used

to identify rare variants in multiple genes, including

genes causing Mendelian diseases with altered blood

pressure levels (such as other types of Bartter’s syndrome

or Gordon’s disease), as well as in genes or locus detected

by GWAS to be linked to changes in blood pressure (such

as ANP and BNP genes NPPA and NPPB or SPAK gene

STK39). Furthermore, this methodology may prove use-

ful in fields beyond blood pressure maintenance and

arterial hypertension, by detecting rare variants associ-

ated with disease predisposition or subclinical traits.

AcknowledgementsThe work was supported in part by the Leducq Founda-

tion Transatlantic Network on Hypertension, the

National Institutes of Health Grant DK-64635, and El

Consejo Nacional de Ciencia y Tecnologıa (CONACYT-

Mexico) Grant 59992 (to G.G.).

There are no conflicts of interest.

References1 Guyton AC. Blood pressure control: special role of the kidneys and body

fluids. Science 1991; 252:1813–1816.

2 Lifton RP, Gharavi AG, Geller DS. Molecular mechanisms of humanhypertension. Cell 2001; 104:45–556.


3 Cruz DN, Simon DB, Nelson-Williams C, Farhi A, Finberg K, Burleson L,et al. Mutations in the Na-Cl cotransporter reduce blood pressure inhumans. Hypertension 2001; 37:1458–1464.

4 Fava C, Montagnana M, Rosberg L, Burri P, Almgren P, Jonsson A, et al.Subjects heterozygous for genetic loss of function of the thiazide-sensitivecotransporter have reduced blood pressure. Hum Mol Genet 2008;17:413–418.

5 Ji W, Foo JN, O’Roak BJ, Zhao H, Larson MG, Simon DB, et al. Rareindependent mutations in renal salt handling genes contribute to bloodpressure variation. Nat Genet 2008; 40:592–599.

6 Devuyst O. Salt wasting and blood pressure. Nat Genet 2008; 40:495–496.

7 Wagner CA. How much is blood pressure in the general populationdetermined by rare mutations in renal salt-transporting proteins? J Nephrol2008; 21:632–634.

8 Gamba G, Miyanoshita A, Lombardi M, Lytton J, Lee WS, Hediger MA,Hebert SC. Molecular cloning, primary structure and characterization oftwo members of the mammalian electroneutral sodium-(potassium)-chloride cotransporter family expressed in kidney. J Biol Chem 1994;269:17713–17722.

9 Meade P, Hoover RS, Plata C, Vazquez N, Bobadilla NA, Gamba G, HebertSC. cAMP-dependent activation of the renal-specific Naþ-Kþ-2Cl-cotransporter is mediated by regulation of cotransporter trafficking. Am JPhysiol Renal Physiol 2003; 284:F1145–F1154.

10 Plata C, Meade P, Hall AE, Welch RC, Vazquez N, Hebert SC, Gamba G.Alternatively spliced isoform of the apical Na-K-Cl cotransporter geneencodes a furosemide sensitive Na-Cl cotransporter. Am J Physiol RenalPhysiol 2001; 280:F574–F582.

11 Plata C, Meade P, Vazquez N, Hebert SC, Gamba G. Functional propertiesof the apical Naþ-Kþ-2Cl- cotransporter isoforms. J Biol Chem 2002;277:11004–11012.

12 Ponce-Coria J, San Cristobal P, Kahle KT, Vazquez N, Pacheco-Alvarez D,De Los Heros P, et al. Regulation of NKCC2 by a chloride-sensingmechanism involving the WNK3 and SPAK kinases. Proc Natl Acad SciU S A 2008; 105:8458–8463.

13 Gamba G, Saltzberg SN, Lombardi M, Miyanoshita A, Lytton J, Hediger MA,et al. Primary structure and functional expression of a cDNA encoding thethiazide-sensitive, electroneutral sodium-chloride cotransporter. Proc NatlAcad Sci USA 1993; 90:2749–2753.

14 Monroy A, Plata C, Hebert SC, Gamba G. Characterization of the thiazide-sensitive Na(þ)-Cl(-) cotransporter: a new model for ions and diureticsinteraction. Am J Physiol Renal Physiol 2000; 279:F161–F169.

15 Vazquez N, Monroy A, Dorantes E, Munoz-Clares RA, Gamba G. Functionaldifferences between flounder and rat thiazide-sensitive Na- Clcotransporter. Am J Physiol Renal Physiol 2002; 282:F599–F607.

16 Hoover RS, Poch E, Monroy A, Vazquez N, Nishio T, Gamba G, Hebert SC.N-glycosylation at two sites critically alters thiazide binding and activity ofthe rat thiazide-sensitive Na(þ):Cl(-) cotransporter. J Am Soc Nephrol2003; 14:271–282.

17 Wilson FH, Kahle KT, Sabath E, Lalioti MD, Rapson AK, Hoover RS, et al.Molecular pathogenesis of inherited hypertension with hyperkalemia: theNa-Cl cotransporter is inhibited by wild-type but not mutant WNK4. ProcNatl Acad Sci U S A 2003; 100:680–684.

18 Rinehart J, Kahle KT, De Los HP, Vazquez N, Meade P, Wilson FH, et al.WNK3 kinase is a positive regulator of NKCC2 and NCC, renal cation-Cl-cotransporters required for normal blood pressure homeostasis. Proc NatlAcad Sci U S A 2005; 102:16777–16782.

19 Kunchaparty S, Palcso M, Berkman J, Velazquez H, Desir GV, Bernstein P,et al. Defective processing and expression of thiazide-sensitive Na-Clcotransporter as a cause of Gitelman’s syndrome. Am J Physiol 1999;277:F643–F649.

20 Yang CL, Angell J, Mitchell R, Ellison DH. WNK kinases regulatethiazide-sensitive Na-Cl cotransport. J Clin Invest 2003; 111:1039–1045.

21 De Jong JC, Willems PH, Mooren FJ, van den Heuvel LP, Knoers NV,Bindels RJ. The structural unit of the thiazide-sensitive NaCl cotransporteris a homodimer. J Biol Chem 2003; 278:24302–24307.

22 De Jong JC, Van Der Vliet WA, van den Heuvel LP, Willems PH, Knoers NV,Bindels RJ. Functional expression of mutations in the human NaClcotransporter: evidence for impaired routing mechanisms in Gitelman’ssyndrome. J Am Soc Nephrol 2002; 13:1442–1448.

23 Starremans PG, Kersten FF, Knoers NV, van den Heuvel LP, Bindels RJ.Mutations in the human Na-K-2Cl cotransporter (NKCC2) identified inBartter syndrome type I consistently result in nonfunctional transporters.J Am Soc Nephrol 2003; 14:1419–1426.

24 Gimenez I, Forbush B. Regulatory phosphorylation sites in the N-terminusof the renal Na-K-Cl cotransporter (NKCC2). Am J Physiol Renal Physiol2005; 289:F1341–F1345.


C


HJH 202096


25 De Jong JC, Willems PH, van den Heuvel LP, Knoers NV, Bindels RJ.Functional expression of the human thiazide-sensitive NaCl cotransporter inMadin-Darby canine kidney cells. J Am Soc Nephrol 2003; 14:2428–2435.

26 Richardson C, Rafiqi FH, Karlsson HK, Moleleki N, Vandewalle A, CampbellDG, et al. Activation of the thiazide-sensitive Naþ-Cl- cotransporter by theWNK-regulated kinases SPAK and OSR1. J Cell Sci 2008; 121:675–684.

27 Sabath E, Meade P, Berkman J, De Los Heros P, Moreno E, Bobadilla NA,et al. Pathophysiology of functional mutations of the thiazide-sensitiveNa-Cl cotransporter in Gitelman disease. Am J Physiol Renal Physiol2004; 287:F195–F203.

28 Pacheco-Alvarez D, San Cristobal P, Meade P, Moreno E, Vazquez N,Munoz E, et al. The Na-Cl cotransporter is activated and phosphorylated atthe amino terminal domain upon intracellular chloride depletion. J BiolChem 2006; 281:28755–28763.

29 Moreno E, Tovar-Palacio C, De Los Heros P, Guzman B, Bobadilla NA,Vazquez N, et al. A single nucleotide polymorphism alters the activity of therenal Naþ:Cl- cotransporter and reveals a role for transmembrane segment4 in chloride and thiazide affinity. J Biol Chem 2004; 279:16553–16560.

30 Denning GM, Anderson MP, Amara JF, Marshall J, Smith AE, Welsh MJ.Processing of mutant cystic fibrosis transmembrane conductanceregulator is temperature-sensitive. Nature 1992; 358:761–764.

31 De Los Heros P, Kahle KT, Rinehart J, Bobadilla NA, Vazquez N, SanCristobal P, et al. WNK3 bypasses the tonicity requirement for K-Clcotransporter activation via a phosphatase-dependent pathway. Proc NatlAcad Sci U S A 2006; 103:1976–1981.

32 Kahle KT, Rinehart J, De Los Heros P, Louvi A, Meade P, Vazquez N, et al.WNK3 modulates transport of Cl- in and out of cells: implications forcontrol of cell volume and neuronal excitability. Proc Natl Acad Sci U S A2005; 102:16783–16788.


33 Gamba G. Molecular physiology and pathophysiology of the electroneutralcation-chloride cotransporters. Physiol Rev 2005; 85:423–493.

34 Hughes AD. How do thiazide and thiazide-like diuretics lowerblood pressure? J Renin Angiotensin Aldosterone Syst 2004;5:155–160.

35 Pressler CA, Heinzinger J, Jeck N, Waldegger P, Pechmann U, Reinalter S,et al. Late-onset manifestation of antenatal Bartter syndrome as a result ofresidual function of the mutated renal Naþ-Kþ-2Cl- co-transporter. J AmSoc Nephrol 2006; 17:2136–2142.

36 Wheeler DA, Srinivasan M, Egholm M, Shen Y, Chen L, McGuire A, et al.The complete genome of an individual by massively parallel DNAsequencing. Nature 2008; 452:872–876.

37 Bodmer W, Bonilla C. Common and rare variants in multifactorialsusceptibility to common diseases. Nat Genet 2008; 40:695–701.

38 Arora P, Newton-Cheh C. Blood pressure and human geneticvariation in the general population. Curr Opin Cardiol 2010; 25:229–237.

39 Wang Y, O’Connell JR, McArdle PF, Wade JB, Dorff SE, Shah SJ, et al.Whole-genome association study identifies STK39 as a hypertensionsusceptibility gene. Proc Natl Acad Sci U S A 2009; 106:226–231.

40 Rafiqi FH, Zuber AM, Glover M, Richardson C, Fleming S, Jovanovic S, et al.Role of the WNK-activated SPAK kinase in regulating blood pressure.EMBO Mol Med 2010; 2:63–75.

41 Lifton RP. Individual genomes on the horizon. N Engl J Med 2010;362:1235–1236.

42 Choi M, Scholl UI, Ji W, Liu T, Tikhonova IR, Zumbo P, et al. Geneticdiagnosis by whole exome capture and massively parallel DNA sequencing.Proc Natl Acad Sci U S A 2009; 106:19096–19101.


Rare mutations in SLC12A1 and SLC12A3 protect against hypertension by reducing the activity of renal...

Documents

Transcript of Rare mutations in SLC12A1 and SLC12A3 protect against hypertension by reducing the activity of renal...