Phenotypic Mutation 'sublytic' (pdf version)
Allelesublytic
Mutation Type missense
Chromosome7
Coordinate30,415,225 bp (GRCm39)
Base Change T ⇒ C (forward strand)
Gene Atp4a
Gene Name ATPase, H+/K+ exchanging, gastric, alpha polypeptide
Synonym(s) H+K+-transporting alpha 1, H+/K+-ATPase alpha
Chromosomal Location 30,411,634-30,424,959 bp (+) (GRCm39)
MGI Phenotype FUNCTION: [Summary is not available for the mouse gene. This summary is for the human ortholog.] The protein encoded by this gene belongs to a family of P-type cation-transporting ATPases. The gastric H+, K+-ATPase is a heterodimer consisting of a high molecular weight catalytic alpha subunit and a smaller but heavily glycosylated beta subunit. This enzyme is a proton pump that catalyzes the hydrolysis of ATP coupled with the exchange of H(+) and K(+) ions across the plasma membrane. It is also responsible for gastric acid secretion. This gene encodes a catalytic alpha subunit of the gastric H+, K+-ATPase. [provided by RefSeq, Jul 2008]
PHENOTYPE: Homozygous mutation of this gene results in achlorhydria, hypergastrinemia, and abnormalities of the parietal cells. Mice homozygous for an ENU-induced allele exhibit iron-deficiency anemia. [provided by MGI curators]
Accession Number

NCBI RefSeq: NM_018731; MGI: 88113

MappedYes 
Amino Acid Change Serine changed to Proline
Institutional SourceBeutler Lab
Gene Model not available
AlphaFold no structure available at present
SMART Domains Protein: ENSMUSP00000005692
Gene: ENSMUSG00000005553
AA Change: S282P

DomainStartEndE-ValueType
Pfam:H-K_ATPase_N 2 42 5.4e-23 PFAM
Cation_ATPase_N 52 126 2.26e-18 SMART
Pfam:E1-E2_ATPase 144 375 1.1e-57 PFAM
Pfam:Hydrolase 380 739 5.3e-16 PFAM
Pfam:HAD 383 736 1.9e-18 PFAM
Pfam:Cation_ATPase 436 531 1.6e-24 PFAM
Pfam:Cation_ATPase_C 809 1019 4.8e-43 PFAM
Predicted Effect probably benign

PolyPhen 2 Score 0.317 (Sensitivity: 0.90; Specificity: 0.89)
(Using ENSMUST00000005692)
SMART Domains Protein: ENSMUSP00000131964
Gene: ENSMUSG00000005553
AA Change: S282P

DomainStartEndE-ValueType
Pfam:H-K_ATPase_N 2 42 4.9e-28 PFAM
Cation_ATPase_N 52 126 2.26e-18 SMART
Pfam:E1-E2_ATPase 145 376 1e-62 PFAM
Pfam:Hydrolase 380 730 9.3e-25 PFAM
Pfam:HAD 383 727 2.1e-15 PFAM
Pfam:Hydrolase_like2 436 531 4e-25 PFAM
Pfam:Cation_ATPase_C 800 1010 1.5e-42 PFAM
Predicted Effect probably benign

PolyPhen 2 Score 0.317 (Sensitivity: 0.90; Specificity: 0.89)
(Using ENSMUST00000170371)
Predicted Effect probably benign
Meta Mutation Damage Score Not available question?
Is this an essential gene? Probably nonessential (E-score: 0.085) question?
Phenotypic Category Autosomal Recessive
Candidate Explorer Status loading ...
Single pedigree
Linkage Analysis Data
Penetrance 100% 
Alleles Listed at MGI

All alleles(3) : Targeted, knock-out(1) Targeted, other(2)

Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL00339:Atp4a APN 7 30412629 missense possibly damaging 0.95
IGL01327:Atp4a APN 7 30412675 missense possibly damaging 0.96
IGL01510:Atp4a APN 7 30420216 missense probably benign 0.02
IGL01763:Atp4a APN 7 30414943 missense probably benign 0.20
IGL02061:Atp4a APN 7 30414454 missense probably damaging 1.00
IGL02435:Atp4a APN 7 30416482 missense probably benign
IGL02903:Atp4a APN 7 30415344 missense probably benign 0.00
IGL03181:Atp4a APN 7 30424129 missense probably benign 0.02
IGL03350:Atp4a APN 7 30420292 missense probably damaging 1.00
atypical UTSW 7 30414781 missense possibly damaging 0.84
IGL03097:Atp4a UTSW 7 30422462 missense probably benign 0.14
R0095:Atp4a UTSW 7 30420160 missense probably damaging 0.99
R0121:Atp4a UTSW 7 30419526 missense probably benign 0.00
R0140:Atp4a UTSW 7 30419526 missense probably benign 0.00
R0241:Atp4a UTSW 7 30416560 missense probably benign 0.00
R0437:Atp4a UTSW 7 30419526 missense probably benign 0.00
R0624:Atp4a UTSW 7 30418424 missense probably benign
R1164:Atp4a UTSW 7 30417117 missense probably benign 0.00
R2105:Atp4a UTSW 7 30419793 critical splice donor site probably null
R2272:Atp4a UTSW 7 30414925 nonsense probably null
R2327:Atp4a UTSW 7 30419666 missense probably benign 0.16
R2881:Atp4a UTSW 7 30419650 missense probably benign 0.16
R2990:Atp4a UTSW 7 30419650 missense probably benign 0.16
R2992:Atp4a UTSW 7 30419650 missense probably benign 0.16
R2993:Atp4a UTSW 7 30419650 missense probably benign 0.16
R3123:Atp4a UTSW 7 30419650 missense probably benign 0.16
R3125:Atp4a UTSW 7 30419650 missense probably benign 0.16
R3441:Atp4a UTSW 7 30419650 missense probably benign 0.16
R3442:Atp4a UTSW 7 30419650 missense probably benign 0.16
R3686:Atp4a UTSW 7 30419650 missense probably benign 0.16
R3687:Atp4a UTSW 7 30419650 missense probably benign 0.16
R3845:Atp4a UTSW 7 30416540 missense probably null 0.99
R4027:Atp4a UTSW 7 30424377 splice site probably null
R4072:Atp4a UTSW 7 30414757 missense probably benign 0.09
R4433:Atp4a UTSW 7 30419650 missense probably benign 0.16
R4454:Atp4a UTSW 7 30419650 missense probably benign 0.16
R4457:Atp4a UTSW 7 30419650 missense probably benign 0.16
R4458:Atp4a UTSW 7 30419650 missense probably benign 0.16
R4510:Atp4a UTSW 7 30423678 nonsense probably null
R4511:Atp4a UTSW 7 30423678 nonsense probably null
R4576:Atp4a UTSW 7 30417147 missense probably benign 0.25
R4656:Atp4a UTSW 7 30419373 intron probably benign
R4661:Atp4a UTSW 7 30419650 missense probably benign 0.16
R4662:Atp4a UTSW 7 30419650 missense probably benign 0.16
R4852:Atp4a UTSW 7 30423693 missense probably benign 0.10
R4892:Atp4a UTSW 7 30411899 missense probably benign 0.07
R4907:Atp4a UTSW 7 30418517 missense possibly damaging 0.66
R5024:Atp4a UTSW 7 30415289 missense possibly damaging 0.82
R5254:Atp4a UTSW 7 30414955 missense probably damaging 1.00
R5318:Atp4a UTSW 7 30414754 missense probably damaging 1.00
R5340:Atp4a UTSW 7 30420231 missense probably benign
R5484:Atp4a UTSW 7 30420097 unclassified probably benign
R5729:Atp4a UTSW 7 30411851 missense possibly damaging 0.48
R5762:Atp4a UTSW 7 30418521 missense probably damaging 0.99
R5797:Atp4a UTSW 7 30412074 missense probably damaging 1.00
R6030:Atp4a UTSW 7 30421941 missense probably damaging 0.99
R6030:Atp4a UTSW 7 30421941 missense probably damaging 0.99
R6077:Atp4a UTSW 7 30415344 missense probably benign 0.00
R6243:Atp4a UTSW 7 30415382 missense possibly damaging 0.68
R6346:Atp4a UTSW 7 30414781 missense possibly damaging 0.84
R6459:Atp4a UTSW 7 30411887 missense probably benign 0.00
R6515:Atp4a UTSW 7 30411903 missense possibly damaging 0.78
R6773:Atp4a UTSW 7 30414802 missense probably damaging 0.98
R6854:Atp4a UTSW 7 30414433 missense probably benign 0.29
R7215:Atp4a UTSW 7 30416785 missense possibly damaging 0.61
R7271:Atp4a UTSW 7 30421944 missense probably benign 0.16
R7340:Atp4a UTSW 7 30416155 missense possibly damaging 0.94
R7457:Atp4a UTSW 7 30420192 missense probably benign 0.08
R7593:Atp4a UTSW 7 30424105 missense probably benign 0.08
R7712:Atp4a UTSW 7 30414978 missense probably damaging 0.96
R7762:Atp4a UTSW 7 30419461 missense probably damaging 0.96
R8714:Atp4a UTSW 7 30420013 missense probably damaging 0.99
R9324:Atp4a UTSW 7 30415207 missense probably benign 0.02
Z1177:Atp4a UTSW 7 30417265 missense possibly damaging 0.47
Z1186:Atp4a UTSW 7 30416782 missense probably benign
Mode of Inheritance Autosomal Recessive
Local Stock Sperm, gDNA
MMRRC Submission 034317-UCD
Last Updated 2016-05-13 3:09 PM by Stephen Lyon
Record Created 2009-05-26 12:00 AM
Record Posted 2012-04-11
Other Mutations in This Stock Stock #: 7510 Run Code:
Validation Efficiency: 77/92

GeneSubstitutionChr/LocMutationPredicted EffectZygosity
Ankmy2 T to C 12: 36,207,411 (GRCm39) V19A probably benign Het
Camk4 G to T 18: 33,289,892 (GRCm39) A180S probably null Homo
Slc28a1 T to C 7: 80,819,017 (GRCm39) V622A probably benign Het
Phenotypic Description
Figure 1.  Sublytic mice develop severe anemia.  (A) Morphology of RBCs from a wild-type and sublytic mouse.  RBCs were examined after Wright-Giemsa staining, and photographed at a 63x magnification (oil) under light microscopy.  (B) BM cells (top panels) and splenocytes (bottom panels) prepared from wild-type and sublytic mice were labeled with Ter119 and CD71 Abs and analyzed by flow cytometry.  Values indicate the percentage of cells in each quadrant.  (C) Reticulocytes were stained with thiazole orange and counted by flow cytometry.  The percentage of reticulocytes in the blood is shown (n = 6 for both wild-type and sublytic mice).  (D) Serum erythropoietin was measured by ELISA (n = 3 for both wild-type and sublytic mice).  Data represent means ± SEM.  (E)  Whole spleens from a sublytic homozygous and a wild-type mouse at approximately 9 months of age.  (F) Sections of spleens from wild-type and sublytic mice stained with H & E are shown at a 20x magnification.  (G) In vivo RBC survival.  RBCs were labeled with biotin in vivo by injecting mice with N-hydroxysuccinimide-biotin at 6 weeks of age.  The survival of RBCs was followed by measuring the percentage of labeled erythrocytes by flow cytometry.  Figure obtained from (1).
Figure 2.  Sublytic erythrocytes are resistant ot osmotic lysis.  One microliter of fresh blood was added to 0.2 mL of a series of hypotonic solutions ranging from 0.1%-0.85% NaCl.  The percentage of lysed erythrocytes was caluclated based on absorbance.  The degree of lysis in 0.1% and 0.85% NaCl was considered to be 100% and 0%, respectively.  Data represent the means ± SEM. Figure obtained from (1).
Figure 3.  Sublytic mice are iron deficient.  (A) Serum iron levels (n = 5 for wild-type mice and n = 3 for sublytic mice, P = .0357).  (B) Liver iron stores (n = 4 for wild-type mice and n = 5 for sublytic mice, P = .0159). (C) Sections of spleens from sublytic and wild-type mice stained for iron with Prussian blue and shown at a 20x magnification. (D) Hamp expression measured by quantitatve RT-PCR using total RNA isolated from liver.  Expression was normalized to that of endogenous S18 transcripts (n = 5 for wild-type mice and n = 3 for sublytic mice, P = .0357). (E) Gastrointestinal iron absorption.  Mice were maintained on a regular diet or on an iron-deficient diet for 1 month.  Animals were fasted for 12 hours before gastric instillation of 59FeCl3.  Iron absorption was determined 24 hours after gastric feeding (n = 6 for wild-type mice and n = 5 for sublytic mice).  The percentage of absorption was calculated as the amount of radioactivity in the whole animal minus the gastrointestinal tract, divided by the amount instilled by gavage. Note that wild-type mice were rendered anemic, as determined by complete blood count analysis, by feeding with the iron-deficient diet for 1 month.  Graphical data were expressed as means ± SEM. *P < .05. Figure obtained from (1).       

Figure 4.  Atp4asublytic causes parietal cysts and achlorhydria.  (A) Stomachs of age- and sex-matched sublytic and wild type mice.  (B) Sections of stomachs from sublytic and wild type mice stained with H&E.  Numerous parietal cell cysts developed in the gastric mucosa of sublytic mice.  Arrow indicates parietal cells.  (C) pH of the gastric contents of sublytic and wild-type mice. Samples were collected 45 minutes after subcutaneous injection of histamine HCl.  Each point represents an individual mouse.  Figure obtained from (1).

Figure 5.  Maintenance on acidified drinking water rescues anemia in sublytic mice.  Sublytic homozygous mice 2-4 months of age (n = 4) were maintained for a period of 3 months on drinking water acidfied to pH 2.5-2.8 with hydrochloric acid.  C57BL/6J mice (n = 6) were maintained for the same period on normal drinking waster.  Data shown are from analysis after 3 months of treatment. (A) Complete blood counts. *P < 0.05 by Student's t test.  (B) Section of spleen from sublytic homozygote stained for iron with Prussian blue and shown at a 20x magnification. (C) Serum iron levels and (D) Hamp expression measured by quantitiatve RT-PCR using total RNA isolated from liver.  Expression was normalized to that of endogenous 18S transcripts. P = 0.0571 by Student t test.  Figure obtained from (1).

The sublytic phenoytpe was identified in two ENU-mutagenized G3 sisters (C7510, C7511) among a litter of three mice during a screen to identify hematological phenovariants (1).  Blood cell analysis revealed that sublytic mice were severely anemic, exhibiting reduced hematocrit, hemoglobin (total and mean corpuscular hemoglobin), and mean corpuscular volume (MCV) (1).  Blood smears from sublytic mice showed hypochromic red blood cells (RBCs), with anisocytosis and poikilocytosis (Figure 1A).  Leukocytes were normal in number and appearance. Reciprocal bone marrow transplantation between sublytic and wild type mice demonstrated that the development of anemia was independent of the hematopoietic compartment.

The severity of anemia in sublytic mice changed with age.  Four to six week old sublytic mice had increased numbers of Ter119highCD71high/med erythroid progenitor cells in the bone marrow and spleen (Figure 1B-D), associated with increased serum erythropoietin and reiculocytosis, and sometimes with thrombocytosis. Recovery of RBC numbers, hemoglobin level, and hematocrit were observed in response to the increased erythropoiesis, and approached wild type levels after about eight weeks of age and were maintained for several months. Some sublytic mice later developed severe anemia (greatly reduced hemoglobin, RBC numbers, and serum iron) after about ten months of age. These animals also developed spenomegaly (Figure 1E), and splenic achitechture was disrupted by extramedullary erythropoiesis (Figure 1F).  The half-life of erythrocytes in eight week old sublytic mice was found to be slightly reduced from 26 to 24 days (Figure 1G).  Additional analysis of blood from sublytic mice demonstrated that their erythrocytes were more resistant to lysis with hypotonic saline than those of wild type mice (Figure 2(1)

Sublytic mice showed a reduction in serum ion levels (Figure 3A) as well as a reduction in iron stores in both the spleen (Figure 3B) and liver (Figure 3C) (1).  Furthermore, the livers of sublytic mice had reduced levels of Hamp mRNA encoding the peptide hormone hepcidin, which controls systemic iron levels by downregulating ferric iron export into the plasma and absorption through the duodenal epithelium (Figure 3D) (1).   After one month on a low iron diet, sublytic mice failed to increase iron absorption, whereas wild type mice rendered anemic under the same conditions doubled their iron absorption (Figure 3E).

Sublytic mice also had hard, enlarged stomachs (Figure 4A), with cysts in gastric glands as seen in histological sections (Figure 4B(1).  In contrast to the acidity (pH 3) of the gastric contents of wild type mice, a pH close to neutral was measured for the gastric contents of sublytic mice (Figure 4C). 

The anemia phenotype of sublytic mice was rescued by maintaining the mice on a regimen of HCl-acidified drinking water (pH 2.5-2.8) for three months.  Partial reversal of anemia was observed as early as one month after treatment, and a complete normalization in RBC indices was observed after three months (Figure 5A) (1).  Furthermore, iron deposits in the spleen (Figure 5B) as well as an increase in serum iron (Figure 5C) and Hamp transcript levels (Figure 5D) were observed in sublytic mice given acidified water.

Nature of Mutation

The sublytic mutation was mapped on the basis of anemia by bulk segregation analysis (BSA) of F2 intercross offspring using C57BL/10J as the mapping strain (n=10 with mutant phenotype, 16 with normal phenotype).  The mutation showed strongest linkage with two Chromosome 7 markers that had equivalent synthetic LOD scores of 3.07:  B10SNPS0105 at 7014667 bp and B10SNP2G0049 at 28467081 bp (1).  B10SNP2G0049 showed a higher BSA linkage score than B10SNPS0105 (3.49 vs. 3.32).  In addition, a trend towards linkage was observed on Chromosome 5, with a peak synthetic LOD of 1.58 at 67382789 bp (B10SNPS0079).

Whole genome SOLiD sequencing of a sublytic homozygote identified a total of three mutations that could be validated by capillary sequencing out of 92 discrepant calls covered three or more times (validation efficiency 77/92).  One of them, a heterozygous mutation in Slc28a1, was found to be on Chromosome 7, 59.79 Mb from the peak marker (B10SNP2G0049).  The Atp4a gene was selected as a candidate on Chromosome 7 that when mutated causes enlargement of the stomach and formation of large cysts in parietal gastric glands (2;3as observed in sublytic mice, and was directly sequenced.  A homozygous T to C transition was detected at position 863 of the Atp4a transcript (exon 7 of 22 total exons), which results in a serine to proline change at position 282 of the encoded gastric H,K-ATPase α subunit.  The mutation had been detected by SOLiD sequencing among “N pattern” nucleotides (those covered 1 or 2 times).  The mutation occurs at position Chr 7: 31500819 bp, located 3.03 Mb from the peak marker.  A portion of the Atp4a transcript is shown with the mutated nucleotide in red:

848 ATTGGGCGCATCGCCTCGCTGGCCTCGGGTGTG

277 -I--G--R--I--A--S--L--A--S--G--V-

F2 progeny from mapping crosses were genotyped for the Atp4a mutation.   100% of anemic mice harbored the Atp4a mutation, while non-anemic mice were either wild type or heterozygous for the mutation.  The Slc28a1 mutation was excluded by fine mapping, which defined a critical region located between 7014667 and 38216957 bp.  We conclude that the T to C transition in Atp4a is the mutation responsible for the sublytic phenotype.

Illustration of Mutations in
Gene & Protein
Protein Prediction

 

Figure 6. Domain and 3D structure of ATP4A.  The phosphorylation (P), nucleotide-binding (N), and actuator (A) domains are positioned cytoplasmically.  Both the A and P domains are discontinuous, divided by helices of the M domain and by the N domain, respectively.  The invariant TGE sequence motif in the A domain contacts the phosphorylation site (the aspartate residue of the conserved sequence DKTGTLT) during the ion pumping cycle.  The sublytic serine to proline mutation at position 282 lies in the A domain.  UCSF Chimera model based on PDB: 3IXZ, Abe et al., Embo J., 28, 1637-43 (2009).  Click on the 3D structure to view it rotate.  

Figure 7.  The E1/E2 model of the P-type ATPase ion translocation cycle.  See text for details.

Figure 8.  In the E1 state, interactions with residues in the M domain promote the binding of Ion1 from the cytoplasmic side of the membrane.  This promotes a conformational change in the P domain that results in the autophosphorylation of the critical aspartate residue by Mg-ATP, which is brought to the phosphorylation site by the N domain.  A conformational change involving the P domain and a rotation of the A domain converts the autophosphorylated E1-P state to the E2-P state.  From the E2-P state the reverse reaction, i.e., hydrolysis of the phosphoryl group and phosphorylation of ADP, cannot occur.  In E2-P, access to the ion binding site from the cytoplasm is close and the high affinity Ion1-binding site is disrupted, causing Ion1 to be released to the outside of the cell through an exit channel.  The ion binding site now has high affinity for Ion2, which enters from the outside of the cell. Hydrolysis of the phosphorylated aspartate brings the enzyme to the E2 state, in which the ion binding site opens to the inside of the cell and Ion2 is released, resulting in the E1 state and the beginning of another cycle.  Figure adapted from reference (4).

Acid secretion in the stomach is controlled by the H,K-ATPase, which catalyzes the electroneutral exchange of luminal potassium ions for cytoplasmic protons.  The H,K-ATPase is one of the P-type ATPases, a large group of evolutionarily related ion pumps that use the free energy of ATP hydrolysis to drive transport and establish ion gradients across membranes (4).  In all P-type ATPases both the N- and C-termini are located on the cytoplasmic side of the membrane such that these proteins contain an even number of transmembrane segments.  Four well-defined, conserved protein domains exist in P-type ATPases: the phosphorylation (P) domain, nucleotide-binding (N) domain, actuator (A) domain, and membrane (M) domain (4-8).  The P, N, and A domains are positioned cytoplasmically, whereas the M domain spans the plasma membrane.  The H,K-ATPase is closely related to other cation transport ATPases including the Na,K-ATPase and the sarcoplasmic reticulum Ca2+-ATPase (SERCA).  It consists of two subunits in [αβ]2 heterodimer oligomer stoichiometry: the 110 kD catalytic α subunit encoded by Atp4a (1033 aa in mouse) that contains all the elements of P-type ATPases (Figure 6), and the 35 kD accessory β subunit encoded by Atp4b (294 aa in mouse) (9;10).  The β subunit is a highly glocosylated type II single-span membrane protein with a short N-terminal cytoplasmic tail and a large C-terminal extracellular domain (11-13); it has been shown to regulate assembly, trafficking, and insertion of the complex into the membrane (14;15).  The α subunit is 98% identical in sequence between mice and humans; the β subunit is 83.8% identical between mice and humans. 

The 3D structure of the H,K-ATPase α and β oligomer deduced by electron crystallography (16demonstrates an overall conformation similar to those of other P-type ATPases (6-8).  The α subunit has a characteristic cytoplasmic domain consisting of P, N, and A domains, and a transmembrane M domain with ten membrane-spanning helices (Figure 6).  The smaller β subunit consists of an ectodomain, a single transmembrane helix that contacts residues of the α subunit M domain, and an rod-like N-terminal tail.  Biochemical studies demonstrated that a cluster of carboxylic amino acids located in the middle of M4, M5, M6, and M8 contains the ion-binding domain in the α subunit (17).  One amino acid, lysine 791, is critical for the specificity of the H,K-ATPase in outward transport of the hydronium ion. 

P-type ATPases cycle through a series of conformational changes to translocate ions.  The Post-Albers or E1/E2 model of the reaction cycle (18;19has been widely used to explain the functional transitions of P-type ATPases that facilitate ion transport.  The model proposed two conformational states, E1 and E2, where the E1 state has high-affinity binding sites for the ion to be transported from the cytoplasmic side of the membrane, and the E2 state has high-affinity binding sites for the ion to be transported from the extracellular side of the membrane (Figure 7).  In the first step, ion1 binds to E1 from inside the cell, triggering autophosphorylation of the enzyme by Mg2+-ATP on a conserved aspartate residue (see below), which leads to the phosphorylated E1-P state.  E2-P is formed by rate-limiting conformational changes to E1-P.  E2-P has reduced affinity for ion1, which is thus released outside the cell.  E2-P is unable to phosphorylate ADP, meaning that the reaction cannot proceed in the reverse direction from E2-P.  Ion2 binds to E2-P from the outside of the cell, leading to hydrolysis of the phosphorylated aspartate, which triggers release of ion2 inside the cell and a return to the E1 state.  The utility of the E1/E2 model has recently been questioned because of some inaccuracies identified experimentally (20), but it still seems to be generally accepted and appears in many current publications.

The P domain is the catalytic core of P-type ATPases, containing a conserved sequence (DKTGTLT) in which the aspartate residue is reversibly autophosphorylated to form the high energy E1-P intermediate of the reaction cycle (21).  The N domain is linked to the P domain by a conserved hinge of two antiparallel strands, and serves to bind Mg2+-ATP and deliver it to the phosphorylation site in the P domain.  The A domain does not interact with ions or nucleotides, but undergoes a rotation during the E1-P to E2-P transition that places it in contact with the phosphorylation site, inducing conformational changes of the P and M domains that may serve to cause release of ion1 and binding of ion2.  The M domain consists of ten membrane-spanning helices that surround the ion-binding sites, and is directly linked to the P domain through helices M4 and M5.  Reciprocal movements of M domain helices have been proposed to open, in turn, the ion binding cavities facing the cytoplasm and extracellular space, and create high affinity binding sites for different ions through the reorientation of coordinating side chains.  Amino acid sequences of M domains are the least conserved of all P-type ATPase domains, reflecting the distinct ionic specificities of each transporter.  A schematic depiction of the catalytic cycle of P-type ATPases conforming to the E1/E2 model is shown in Figure 8.

The two heterodimers of the H,K-ATPase oligomer function out of phase with each other (9).  The ratio of H+ ions transferred to ATP molecules hydrolyzed has been measured at 1 or 2 H+ ions per ATP, and may be pH dependent (22-24).  Following the E1/E2 model, the enzyme has cytoplasmic-open E1 and luminal-open E2 states with high affinity for H+ and K+, respectively.  Upon autophosphorylation on Asp386 (mouse numbering), the H,K-ATPase exists predominantly in the E2-P, ADP-non-reactive form; the reverse reaction from E2-P to E1-P is strongly disfavored (25;26).  This property may account for the ability of the H,K-ATPase to maintain the extremely steep H+ gradient across the gastric mucosa.  In the 3D structure (16), the N-terminal tail of the β subunit was shown to contact the α subunit P domain in the E2-P state; deletion of the β subunit N-terminal tail was shown to result in a significantly lower proportion of forward reaction products (E1-P àE2-P).  These findings suggest that the β subunit N terminus prevents the reverse reaction from E2-P to E1-P, perhaps by holding E2-P in an energetically lower, stabilized state (16).

The sublytic serine to proline mutation at position 282 lies in the A domain.  The primary sequence of the A domain is divided into two unequal parts by the sequences for helices M1 and M2.  The sublytic mutation is in the second part of the domain, close to its C terminus.

Expression/Localization

The α-subunit of the gastric H,K-ATPase is present in tubulovesicular and canalicular membranes of parietal cells of the stomach (27-29).  It has been estimated that the H,K-ATPase represents up to 10% of parietal cell proteins (30).  H,K-ATPase is also expressed in the kidney (31-33).

Background

The gastric mucosa secretes into the stomach a highly acidic fluid (pH 2-3) that denatures proteins, activates the proenzyme pepsinogen (secreted by chief cells), and kills or inhibits the growth of many foodborne organisms.  This gastric acid consists mainly of isotonic HCl (0.15 N) and is produced and discharged by parietal cells through the function of the H,K-ATPase, which pumps acid into the stomach lumen against a gradient of greater than a million fold.  Because H+ is transported in exchange for K+, no net charge accumulates across the plasma membrane (34).  The H,K-ATPase is the target of proton pump inhibitors (PPIs), drugs such as omeprazole (Losec; AstraZeneca) and lansoprazole (Prevacid; TAP Pharmaceuticals) that inhibit gastric acid secretion for the treatment of peptic ulcers, reflux oesophagitis, gastroesophageal reflux disease (GERD), Barrett’s esophagus, and Zollinger-Ellison syndrome, as well as the eradication of Helicobacter pylori as part of combination regimens (35;36).

Figure 9.  Parietal cell activation.  (A) Parietal cells secrete hydrochloric acid and are located within gastric glands in the stomach. (B) In an unstimulated parietal cell, the apical canaliculi extend into the cell and contain a few, short microvilli.  Mitochondria and tubulovesicles containing the resting state H, K-ATPase (green) are located in the cytoplasm.  Upon stimulation, the tubulovesicles are recruited to the membrane and the canaliculi and microvilli deepen, increasing surface area for acid secretion.

Parietal cells are morphologically and functionally specialized for acid production.  At their apical membrane, parietal cells possess a series of small canals (canaliculi) that invaginate from the cell surface and project throughout the cell interior.  In the non-secreting or resting parietal cell, the canaliculi are lined with short microvilli, and throughout the cytoplamic space are an abundance of membranous structures, called tubulovesicles, constituting some 50% of the total membrane mass and taking the form of vesicles, tubules, and cisternal sacs (Figure 9) (37).  These tubulovesicles contain virtually all of the resting state H,K-ATPase, which is inactive for H+ transport because tubulovesicles lack an endogenous pathway for bringing K+ inside the vesicle where it can be exchanged for H+ (38;39).  Upon stimulation of acid secretion (see below), the H,K-ATPase-rich tubulovesicles migrate and fuse to the canalicular membrane, resulting in a dramatically expanded apical membrane surface with elongated microvilli and dilated canalicular spaces, and a correspondingly reduced cytoplasmic tubulovesicular membrane area (40-42).  This membrane fusion event also recruits large numbers of the H,K-ATPase to the apical membrane, where it transports H+ out of the cell in exchange for transport of K+ into the cell (38).  The H,K-ATPase operates in parallel with the K+ channel KCNQ1, which is required to recycle K+ to the luminal surface of the proton pump (43-45).  Inhibition or deletion of KCNQ1 blocks acid secretion by parietal cells in vitro and in vivo.

Acid secretion by parietal cells is stimulated by extracellular signals delivered through receptors on the basolateral membrane.  The histamine H2 receptor, cholinergic muscarinic M3 receptor, and gastrin CCK-B receptor are the three receptors known to stimulate acid secretion, with histaminergic stimulation being a central player.  These receptors activate a number of intracellular signaling pathways, including those of protein kinase A (PKA), protein kinase C (PKC), Ca2+-calmodulin kinase II (CaMKII), and PI3 kinase, that lead to upregulation of H,K-ATPase mRNA and fusion of H,K-ATPase-containing tubulovesicles with canalicular membranes [reviewed in (46;47)].  Membrane fusion is mediated by SNAREs and their associated proteins including Rab proteins.  Re-sequestration of H,K-ATPases back into tubulovesicles negatively regulates acid secretion and is carried out through coat protein-mediated endocytosis (e.g. clathrin-mediated endocytosis), together with dynamin and other associated proteins (46).

The gastric H,K-ATPase was first cloned from rat in 1986 (48), and later from pig, rabbit, dog, human, and mouse (49-53).  In 2000, the phenotype of the knockout mouse was first reported (3).  Atp4a-/- mice failed to secrete gastric acid, resulting in a near neutral pH (6.9) of the gastric contents, compared to a pH of 3.17 in wild type littermates.  Atp4a-/- mice produced elevated levels of gastrin both at the transcript and protein levels.  Parietal cells and chief cells were present in normal numbers in Atp4a-/- mice, and expressed the H,K-ATPase β subunit and pepsinogen, respectively.  However, both cell types displayed changes in morphology compared to wild type cells.  Parietal cells of Atp4a-/- mice contained dilated canaliculi with sparse short and stiff microvilli, and some contained massive stores of cytoplasmic glycogen.  At the ultrastructural level, Atp4a-/- parietal cells displayed decreased canalicular folds, and normally abundant tubulovesicles were replaced with a few rigid round vesicles (54).  The microvilli of Atp4a-/- parietal cells displayed alterations in the actin cytoskeleton.  Atp4a-/- chief cells had a reduced number of granules and endoplasmic reticulum.  These phenotypes are similar to, but do not precisely replicated, those of Atp4b-/- mice lacking the H,K-ATPase β subunit (55;56).

Deficiency of the H,K-ATPase α subunit has been implicated in the development of stomach hyperplasia (2;3).  Stomach weight and thickness in Atp4a-/- mice were significantly increased compared to those of wild type mice by 8 months of age, and continued to increase until at least 20 months of age.  Atp4a-/- mice also had more cells per gastric gland than wild type mice.  Cyst formation was observed by 3 months of age, but with age, the formation of countless, and occasionally very large, cysts was widespread.  In addition, incomplete intestinal metaplasia, ciliated metaplasia, a shift in mucins from neutral to acidic, and inflammation were observed in Atp4a-/-mice.  Transcripts encoding gastric growth and oncogenic factors, including those for Reg IIIγ, Reg IIIδ, and osteopontin,were upregulated in stomach tissue from Atp4a-/- mice.  However, nuclear atypia and the invasion of epithelial cells into the muscularis mucosa were absent, as was metastasis into adjacent organs.

The H,K-ATPase is the major autoantigen targeted in autoimmune gastritis (57), an inflammatory disease of the stomach epithelia leading to pernicious anemia caused by the loss of parietal cells and consequently a lack of intrinsic factor, which is required to absorb vitamin B12 from the gut (58).  Autoimmune gastritis is characterized by a monocytic infiltrate of the gastric mucosa, loss of parietal and zymogenic cells, and T and B cell responses to the H,K-ATPase.  A CD4+ T cell response to the α and β subunits of the H,K-ATPase is necessary and sufficient to induce autoimmune gastritis (59-61).  In mice, autoimmune gastritis develops in susceptible strains after neonatal thymectomy (62), thymectomy combined with administration of immunosuppressive drugs or irradiation in adult mice (63), or after CD4+ T cell transfer to lymphopenic recipients (64).  It is now known that failure to eliminate H,K-ATPase specific CD4+ T cells, and loss of CD4+ T cell tolerance to the H,K-ATPase leads to the disease.  Exposure of T cells to the gastritogenic H,K-ATPase antigen outside of the thymus is important for purging of autoreactive CD4+ T cells (64-66), and for the maintenance of H,K-ATPase specific CD4+CD25+ regulatory T cells (65;67).

Putative Mechanism

Evidence from numerous studies indicates that gastric acid secretion is important for optimal absorption of non-heme iron.  Iron is obtained exclusively from the diet, and is absorbed in the proximal portion of the duodenum by intestinal enterocytes (68).  Approximately two-thirds of dietary non-heme iron is in the form of ferric (Fe3+) salts, which are precipitated in solutions with a pH greater than 3 (69).  In the stomach, gastric acid maintains the acidic environment necessary to enforce solubility, and permit subsequent chelation and/or reduction, of ferric iron.  Clinical studies as early as 1932 suggested that gastric acidity might have a role in the absorption of iron, because the response to oral iron supplements in patients with iron deficiency was greater in patients with normal gastric acidity than those with achlorhydria [see Discussion in (70)].  Further studies showed that prolonged achlorhydria resulting from gastric resection or atrophic gastritis can result in iron deficiency and anemia (71-73).  However, patients with Zollinger–Ellison syndrome treated long term (5.7 years average) with omeprazole did not develop iron deficiency or decrease body iron stores (70).

For absorption in the less acidic proximal small intestine, soluble ferric iron must be reduced to ferrous (Fe2+) iron, which enters enterocytes through divalent metal transporter 1 (DMT1), the primary transmembrane non-heme iron transporter in intestinal epithelial cells (74;75).  DMT1 is a cotransporter requiring protons to move Fe2+ across the cell membrane (76;77).  These protons are also provided by gastric acid flowing into the proximal portion of the duodenum where DMT1 is most highly expressed.

The sublytic mice are able to sense and respond to iron-deficiency by down-regulating Hamp transcription; however, they are unable to upregulate intestinal iron absorption (1). The data strongly suggest that the inability of sublytic mice to absorb iron is due to poor solubilty of ferric iron and inefficient DMT1 function, which are caused by reduced availability of protons in the stomach (1)Sublytic mice displayed a 3-fold increase in DMT1 transcript levels, consistent with this hypothesis (1).  An increase in stomach inflammation, as observed in Atp4a knockout mice (2), may also contribute to the anemia phenotype observed in the sublytic model.  The increased surface-to-volume ratio of microcytic sublytic erythrocytes may contribute to the reduced osmotic fragility observed (1).  

Primers Primers cannot be located by automatic search.
Genotyping

Sublytic genotyping is performed by amplifying the region containing the mutation using PCR, followed by sequencing of the amplified region to detect the single nucleotide change.

Primers

sublytic (F): 5’-ACACATGAGAGTCCCCTTGAGACC -3’

sublytic (R): 5’- GTTTGGGCTGACTTCCCACAGTAG -3’

PCR program

1) 95°C             2:00

2) 95°C             0:30

3) 56°C             0:30

4) 72°C             1:00

5) repeat steps (2-4) 29X

6) 72°C             7:00

7)  4°C              ∞

Primers for sequencing

sublytic_seq(F): 5’- TCCACCATGTGTCTGGAGG -3’

sublytic_seq(R): 5’- GGTCCAGGACCCTGTTTATTCAC -3’

The following sequence of 756 nucleotides (from Genbank genomic region NC_000073.5 for linear genomic sequence of Atp4a, sense strand) is amplified:

3280                                           a cacatgagag tccccttgag

3301 acccgcaaca tcgccttctt ctccaccatg tgtctggagg gtctgtgaag catcgttagc

3361 ctgtcctgaa gccacacaga ccccatactt tccataatag cgccttttct gttctgtgtg

3421 atgtgtgcca caagcctccc tatcacagag cacctttctg gtgctgccca cctcaccctg

3481 agctctttct ccccttgttg ccactgcagg aacagctcag ggtttggtgg tgagcaccgg

3541 cgatcgcacc atcattgggc gcatcgcctc gctggcctcg ggtgtggaaa acgagaagac

3601 tccgattgca atcgagattg aacattttgt ggacatcatt gccggcctgg ccatcctctt

3661 cggtgccaca ttctttgtgg tggccatgtg tattggctat accttccttc gggccatggt

3721 cttcttcatg gccattgtgg tagcctatgt gcctgagggg ctgctggcaa ctgtcacagt

3781 gagtaaggga gaaggggtgg gggggggtgc agggagcaga gagctgcttg ttcagtcatc

3841 ccaccagacg tccctactac ctgtccattc attcccctcc cccaccctac atcatccatc

3901 cattcagtga acatttatca ggatgctctg ggctggcccc gtgtcataca ccagagccca

3961 cgggtgaata aacagggtcc tggaccaagg aagatgtggg aagatgttgg gctactgtgg

4021 gaagtcagcc caaac

Primer binding sites are underlined; sequencing primer binding sites are highlighted in gray; the mutated T is indicated in red.

References
Science Writers Eva Marie Y. Moresco, Anne Murray
Illustrators Diantha La Vine
AuthorsXin Du, Oren Milstein, Bruce Beutler