Point Mutation in the Stalk of Angiotensin-Converting Enzyme Causes a Dramatic Increase in Serum Angiotensin-Converting Enzyme But No Cardiovascular Disease

The study was approved by the hospital ethics committee. The procedures followed were in accordance with institutional guidelines. All plasma ACE activities determined in the last 5 years in the UMC Nijmegen (n=≈2900) were reviewed, and 4 unrelated patients were identified with serum ACE activities >80 U/L. In addition, 4 other unrelated patients were identified elsewhere in the Netherlands. First-degree (and in some instances second-degree) relatives of these patients were asked to participate in the clinical evaluation.

After giving written informed consent, these family members (not the probands) were given an interview and physical examination. An ECG was recorded, and blood pressure was recorded automatically every 3 minutes during 20 minutes (Dinamap, Critikon Inc). Serum samples were taken for determination of ACE activity, erythrocyte sedimentation rate, hematology and blood chemistry (including blood minerals, kidney function, liver enzymes, albumin, and glucose), and thyroid-stimulating hormone. Blood samples for renin, angiotensinogen, aldosterone, and angiotensin I and II determination were taken and processed as described.^18,19 Aldosterone was measured in plasma by a radioimmunoassay kit (DPC). In one family (No. 6, Figure 1) an echocardiography (Diasonics Vingmed Ultrasound, System V) was performed in all 12 family members (8 with elevated ACE). In the left parasternal short- and long-axis views, the cardiac dimensions were assessed; in the apical 2-, 4-, and 5-chamber views, the wall motions and valvular function were assessed. In 7 selected individuals 300 mL of blood was drawn in tubes containing heparin for additional in vitro analysis of isolated blood cells.

Serum ACE and ACE genotype were determined in 166 randomly selected, unrelated individuals from an epidemiological survey in Rotterdam.

Activated peripheral blood lymphocytes (PBL) and immature dendritic cells (DC) were generated from 4 individuals with elevated ACE and 3 nonaffected family members (subjects indicated in Figure 1) as described.²⁰ Cells were cultured for 6 days, with a complete change of medium at day 4. Nonadherent PBL were activated for 3 days with PHA (1 μg/mL) and IL-2 (20 U/mL) in the same culture medium. Flow cytometric analysis of the leukocyte populations was performed with the primary antibody (10 μg/mL) and FITC-conjugated GAM IgG F(ab′)2 (Zymed) as the secondary antibody on a FACScan flow cytometer (Becton and Dickinson & Co). An isotype-matched mAb was used as a control.

Colorimetric determination of ACE activity in the plasma was performed with a kit by Fujirebio Inc, which uses a p-hydroxy-Hip-His-Leu substrate.²¹ The reference range in our laboratory is 8.3 to 31.4 U/L (n=215). Fluorimetric assay of ACE activity in serum, plasma, culture fluids, or lysate of cultured cells was performed by measuring the release of His-Leu from the substrates Hip-His-Leu and Z-Phe-His-Leu.^22,23 In isolated cells, ACE levels were corrected for the amount of cells and volume of culture medium.

These assays were performed as described in References 24 and 25, respectively.^24,25

The ACE gene was sequenced in 3 unrelated individuals by means of an automatic sequencer (ABI). After the discovery of the mutation, genotyping of other individuals was performed by polymerase chain reaction (PCR)-based restriction fragment length polymorphism assessment. Genomic DNA was amplified with Goldstar Taq (Eurogentec), with the use of a forward primer ATGTTGAGCTACTTCAAGCCGGC and a reverse primer GTTAAGACCCAAGGCTGGAGGT, resulting in a 141-bp fragment. This fragment was digested with AcyI (Boehringer Mannheim). The C-to-T mutation at position 3705 eliminates this AcyI restriction site. The same primers were used for reverse transcription-PCR to confirm the mutation at mRNA level and to estimate the amount of ACE mRNA isolated from the cultured cells described above. Haplotyping of mutation-containing alleles of the ACE gene was performed by typing of two single nucleotide polymorphisms, 577 bp upstream (C/T, frequencies 0.54/0.46) and 154 bp downstream (A/G, frequencies 0.50/0.50) of the mutation, respectively.²⁶ A PCR (forward primer CAGCCTTGACTGGCAGATTT, reverse CAGTGTTCCCATCCCAGTCT) spanning 1508 bp around the mutation and the two flanking polymorphisms was performed. The wild-type amplimer was digested by AcyI, and the resulting fragments were removed by agarose electrophoresis. The 5′ polymorphic marker on the undigested fragment was typed by single base extension with a commercial kit (SnaPshot, Applied Biosystems). The 3′ marker was characterized by digestion with BlpI (New England Biolabs).

Eight Dutch individuals were identified with ACE values exceeding 4 times the upper limit of normal. In all of them, ACE activity was measured because of a suspicion of sarcoidosis. The patients had a variety of complaints, including fatigue, arthralgia, nephrolithiasis and hypercalciuria, restrictive pulmonary disease, and premature stroke. Despite an extensive workup, this diagnosis could be made in none. Subsequent analysis revealed additional individuals with hyper-ACE in each of the 8 families participating in the study (ACE=105±23 U/L, ranging from 53 to 134 U/L, n=28). The Table summarizes the relevant clinical and laboratory data in these family members. The hyper-ACE segregation pattern was compatible with autosomal dominant inheritance (Figure 1).

Elevated ACE was not accompanied by any apparent clinical abnormality. In each group, 2 individuals had hypertension (diastolic blood pressure >90 mm Hg), 1 individual (female, 68 years of age, with normal ACE level) had heart failure and angina pectoris, and 1 individual (female, 72 years of age, with elevated ACE) had angina pectoris and proven coronary artery disease. Renin, aldosterone (Table), and angiotensinogen (not shown) levels did not differ significantly. Angiotensin II levels were slightly increased in subjects with high serum ACE (Table). The results of all other blood analyses performed were without abnormalities (not shown). ECG recordings were normal, except for the 2 patients with known heart disease. There were no signs of left ventricular hypertrophy on ECG (Table). In one family, echocardiography of all 12 members (8 with elevated ACE) was without abnormalities.

We determined the activity, amount, and possible structural abnormalities of serum ACE in affected and nonaffected family members. The increase of ACE activity (4- to 6-fold) was similar with application of different substrates (Table comparing 20 affected with 21 nonaffected individuals, P<0.0001; Figure 2A comparing 4 affected with 3 nonaffected individuals, P<0.01) and correlated with the increase found in the ACE ELISA (not shown). Moreover, the ratio between the hydrolysis of the substrates Hip-His-Leu and Z-Phe-His-Leu in affected subjects was normal (not shown), suggesting that neither one of two catalytic domains of ACE was altered.^25,28 The binding characteristics of a panel of anti-ACE mAbs recognizing 9 different epitopes of ACE were not different between affected and nonaffected family members (not shown). This binding pattern, as assessed by immunoprecipitation, is very sensitive even to subtle changes in the conformation of ACE molecule.²⁴

In Figure 2, cells obtained from 4 affected and 3 nonaffected individuals were tested. Fluorescently activated cell sorter (FACS) analysis of leukocytes and dendritic cells from hyper-ACE and normal individuals demonstrated that they express a similar amount of ACE on the cell surface (Figure 2C). Also, the total amount of cell-associated ACE (Figure 2B) and ACE mRNA (not shown) was identical. In contrast, in both the activated PBL and immature DC, a 5- to 6-fold increase in the amount of secreted ACE was observed (Figure 2D, P<0.01). These findings implicate that the rate of ACE shedding by activated PBL and DC is affected in the hyper-ACE individuals.

Sequence analysis in 3 affected, unrelated, individuals yielded a C-T mutation at nucleotide position 3705, which leads to the replacement of a proline residue by a leucine at amino acid position 1199 in the stalk region of the mature ACE protein (Figure 3) [nt and aa numbering according to Reference 5].⁵ All affected individuals were heterozygous for this mutation, and the mutation was not present in the family members and 166 unrelated Dutch individuals with normal ACE levels. Moreover, sequence analysis of the complete coding region of the ACE gene in 3 affected individuals revealed that besides known polymorphisms,²⁶ the C-T mutation at position 3705 is the only mutation present in the coding sequence of the ACE gene. Linkage analysis of this mutation with the hyper-ACE phenotype showed a highly significant maximum LOD score of 6.63 at a θ of 0.00. All affected individuals had an identical haplotype carrying the mutation with T at position 577 bp upstream and G at position 154 bp downstream. Finally, analysis of RNA isolated from the DC from affected individuals confirmed the mutation (not shown).