INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Angiotensin-converting enzyme (ACE, peptidyl-dipeptidase A, kinase II; EC 3.4.15.1) is probably the most widely used laboratory test in sarcoidosis, and it has been demonstrated that untreated patients with more active clinical disease tend to have a higher serum ACE level (1). It has recently been shown that the ACE gene contains a polymorphism based on the presence (insertion [I]) or absence (deletion [D]) of a nonsense DNA fragment (4). The polymorphism is located in intron 16, so the ACE itself does not differ due to genotype, but the polymorphism accounts for 47% of the total phenotypic variance in serum ACE level. The genotype is classified into three types: deletion homozygotes, DD; insertion homozygotes, II; and heterozygotes, DI. The serum ACE level of DD type is reported to be about double that of II type. That of DI type is intermediate (4, 5). The DD genotype has been reported as a genetic risk factor for myocardial infarction, dilated cardiomyopathy, left ventricular hypertrophy, and IgA nephropathy (6). However, polymorphism of the ACE gene has seldom been studied in sarcoidosis (10). Therefore, in the present study, we investigated the possible association between polymorphism and sarcoidosis. First, we examined the distribution of genotypes both in sarcoidosis and in normal controls to find a genetic risk factor. Second, the serum ACE level of each genotype was assessed both in sarcoidosis and in normal controls. Finally, we investigated whether the polymorphism was involved with the clinical manifestation of sarcoidosis, especially prognosis.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Population

Two hundred and seven sarcoidosis patients from central Japan were examined. Sarcoidosis was diagnosed based on the clinical picture and the presence of epithelioid cell granulomas in biopsy specimen from the lung, skin, or lymph node. Sixty-one of the cases were males and 146 were females. They had a mean age of 56.2 ± 15.6 yr (mean ± SD) at the first visit to our hospital. There were 141 patients in roentgenographic stage I, 31 in stage II, and 17 in stage III, according to the classification system defined by DeRemee (11). Disease extent was assessed by chest radiography, high resolution CT scan, bronchoalveolar lavage, ⁶⁷Ga lung uptake, ²⁰¹Tl myocardial scintigraphy, abdominal ultrasonography, and Holter ECG.

As normal control subjects, 314 unrelated healthy subjects who lived in the same area of Japan were selected. They consisted of 158 males and 156 females with a mean age of 51.3 ± 12.9 yr. They did not have any abnormalities based on physical examination, chest radiography, blood pressure, ECG, urinalysis, and routine laboratory blood testing, and none were receiving medication at the time of evaluation. Informed consent was obtained from all patients and normal control subjects (Table 1).

Determination of ACE Genotypes

The D and I alleles were identified on the basis of polymerase chain reaction (PCR) amplification of the respective fragments from intron 16 of the ACE gene and size fractionation and visualization by electrophoresis. DNA was extracted from peripheral leukocytes with standard techniques. PCR was performed with 20 pmoles of each primer: sense oligo 5'CTGGAGACCACTCCCATCCTTTCT3' and anti-sense oligo: 5'GATGTGGCCATCACATTCGTCAGAT3' in a final volume of 25 µl, containing 1.5 mM MgCl₂, 50 mM KCl, 10 mM Tris-HCl pH 8.3, 0.2 mM of each dNTP, and 1.25 unit of Taq polymerase (Perkin Elmer-Cetus, Norwalk, CT). The DNA was amplified for 30 cycles with denaturation at 94° C for 30 s, annealing at 58° C for 30 s and extension at 72 ° C for 1 min, followed by final extension at 72 ° C for 5 min (DNA Thermal Cycler 480, Perkin Elmer-Cetus) (12, 13). PCR products were electrophoresed in 2% agarose-gel with 5 µg ethidium bromide per milliliter. The amplification products of the D and I alleles were identified by 300-nm ultraviolet trans-illumination as distinct bands (D allele: 191 bp; I allele: 478 bp) (Figure 1A).

View larger version (62K): [in this window] [in a new window]   Figure 1.   Determination of ACE genotypes. The left lane of each panel contains markers. Panel A shows part of a representative 2% agarose gel stained with ethidium bromide and photographed under ultraviolet transillumination after PCR amplification with insertion-spanning primers. The upper band of 478 bp is the I allele and the lower band of 191 bp is the D allele (arrows). The II type is shown as a single upper band, the DD type as a single lower band, and the DI type as a double band. Panel B shows the results of screening for erroneous assignment of the DD genotype to DI samples with use of an insertion-specific primer. The sample in lane 1 is a standard for II and that in lane 2 is a standard for DI, followed by four samples (lanes 3 through 6) identified as DD with the insertion-spanning primer. A DI sample previously misclassified as DD is in lane 5.

Because the D allele in heterozygous samples is preferentially amplified, each sample found to have the DD genotype was subjected to a second independent PCR amplification with a primer pair that recognizes an insertion-specific sequence (hace 5a, 5'TGGGACCACAGCGCCCGCCACTAC3'; hace 5c, 5'TCGCCAGCCCTCCCATGCCCATAA3'), with identical PCR conditions except for an annealing temperature of 67 ° C. The reaction yields a 335-bp amplicon only in the presence of an I allele, and no product in samples homozygous for DD (14, 15). This procedure demonstrated that approximately 2% of samples (5/240) with the DI genotype were misclassified as DD with the insertion-spanning primer (Figure 1B).

Serum ACE Level Measurement

Serum ACE level was measured by a colorimetric method (colorimetric assay kit, Fujizoki Assay, Tokyo, Japan) using p-hydroxyhippuryl- L-histidyl-L-leucine as the substrate (16).

Statistical Analysis

The allele ratio and genotype distribution of sarcoidosis patients and normal control subjects, and clinical manifestation of the disease among the three genotypes were analyzed with the chi-square test. Analysis of serum ACE level was performed using the Mann-Whitney U test for comparison of the two groups, and the Kruskal-Wallis test for the three genotypes and roentgenographic stages (17). A p value < 0.05 was considered significant. Values are expressed as means ± SD.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Genotype Distribution

Of the 314 normal control subjects, 136 had the II genotype (43.3%), 139 the DI type (44.3%), and 39 the DD type (12.4%). The I allele/D allele (I/D) ratio was 0.654/0.346. Of the 207 sarcoidosis patients, 77 were type II (37.2%), 101 were DI (48.8%), and 29 were DD (14.0%). The I/D ratio was 0.616/0.384. The observed genotype distribution was in agreement with the Hardy-Weinberg proportion. There were no significant differences in the genotype distribution and I/D ratio between normal controls and sarcoidosis cases (Table 2).

Serum ACE Level

In normal control subjects, the average serum ACE level of II, DI, and DD individuals were 10.8 ± 3.1, 13.8 ± 4.3, and 17.2 ± 4.0 (IU/l), respectively. For 174 of the 207 sarcoidosis patients, serum ACE levels were measured at the first visit to our hospital, the average values for II, DI, and DD type being 21.4 ± 7.9, 23.9 ± 7.2, and 27.3 ± 7.5 (IU/l), respectively. Significant differences among the three genotypes were found for both sarcoidosis patients (p < 0.01) and normal control subjects (p < 0.0001). The serum ACE levels in the sarcoidosis cases were significantly increased compared with those for the respective normal control subjects of each genotype (p < 0.0001) (Table 3). Regarding the roentgenographic stage, we observed an increase in serum ACE level, according to the order I < II < III (p = 0.092). The average values for stages I, II, and III were 22.9 ± 7.3, 25.3 ± 8.8, and 25.7 ± 8.4 (IU/l), respectively (Figure 2).

View larger version (20K): [in this window] [in a new window]   Figure 2.   Roentgenographic and serum ACE level. The open columns show the average serum ACE level at the first visit to our hospital for each roentgenographic stage. A trend for increase was observed according to the order I < II < III (p = 0.092).

We defined the 95% confidence interval of our normal control data for each genotype as the new normal range of serum ACE level (i.e., II type: 9.0 ~ 12.5; DI type; 10.5 ~ 17.1; and DD type: 11.9 ~ 22.5 IU/l), and compared the sensitivity of this new range with that of the conventional normal range (i.e., 8.3 ~ 21.4 IU/l) for diagnosis of sarcoidosis. The overall sensitivity of the conventional normal range and the new normal range is 60.8% and 83.0%, respectively. This new normal range was overall 22% more sensitive than the conventional normal range, and 39% more so for the II type (Figure 3).

View larger version (23K): [in this window] [in a new window]   Figure 3.   Comparison of sensitivity of the conventional and new normal ranges. The open and shaded columns show sensitivity of the new normal range and conventional normal range for diagnosis. Serum ACE was measured by a colorimetric method (colorimetric assay kit, Fujizoki Assay) using p-hydroxyhippuryl-L-histidyl-L-leucine as the substrate. Conventional normal range: 8.3 ~ 21.4 (IU/l). New normal range: for II, 9.3 ~ 11.8; for DI, 10.7 ~ 16.6; for DD, 11.5 ~ 22.6 (IU/l). The sensitivity of the new normal range was 22% higher overall than that of the conventional normal range, and 39% higher for the II type.

Association of Polymorphism with Clinical Manifestation

Next, we examined for any relationship between organ involvement and genotype. The percentage of eye involvement with the II, DI, and DD type cases was 72.7%, 75.2%, and 65.5%, respectively; that of skin involvement was 20.8%, 25.7%, and 24.1%, respectively. Heart involvement was observed in 11.7% of II, 8.9% of DI, and 6.9% of DD type patients. These results indicate that there is no specific organ involvement associated with a certain genotype. Regarding multiorgan involvement and genotype, the percentage of patients demonstrating lesion in three or more organs for the II, DI, and DD types were 26.0%, 25.7%, and 20.7%, respectively. No significant differences were found (Table 4). There was no significant correlation between the roentgenographic stage of patients at the first visit to our hospital and the ACE genotype (data not shown).

In order to determine whether the prognosis of sarcoidosis was linked to the genotype, two or more specialists evaluated the disappearance ratio of abnormal shadow on chest radiography 3 and 5 yr after disease onset. The number of evaluable patients was 151 and 139, respectively. The disappearance ratios after 3 yr for the II, DI, and DD type were 33.3%, 25.3%, and 31.8%, respectively, and after 5 yr were 42.3%, 36.4%, and 50.0%. No significant differences were evident among the three genotypes (Table 5). As chronic severe cases, we selected 16 patients who had abnormal shadow on chest radiography for 8 or more yr and two or more extrapulmonary organ involvements, with 8, 6, and 2 cases of II, DI, and DD type, respectively. The genotype distribution did not significantly differ from those of normal control subjects (² = 0.315, df = 2, p  = 0.854) or total sarcoidosis patients (² = 1.06, df = 2, p = 0.589).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ACE is a widely distributed zinc-metalloproteinase occurring, for example, as a membrane-bound ectoenzyme on the surface of vascular endothelial cells and renal epithelial cells and as a circulating enzyme in plasma. It plays an important role in blood pressure homeostasis by activating angiotensin I into angiotensin II and inactivating bradykinin. The ACE gene spans 21 kilobases, is located on the 17th chromosome q23, and consists of 26 exons and 25 introns. The polymorphism exists in intron 16. The length of the insertion is 287 bp, and it is a repetition of a meaningless Alu family configuration (5, 18). It has not been determined why the polymorphism influences the serum ACE level, but some authors have suggested that the insertion/deletion may be in linkage disequilibrium with regulatory elements of the ACE gene, or that the insertion itself might modify the splicing process of the ACE precursor mRNA by interfering with the lariat formation step (4, 19). Some investigators have indicated that the I/D ratio depends upon race. In Caucasians, the I/D ratio was reported to be 0.41/0.59 by Rigat and coworkers (4), 0.4449/0.5551 by Lindpaintner and colleagues (15), and 0.400/0.600 by Arbustini and coworkers (10), which means the D allele is dominant. As confirmed in the present study, however, in Japanese the I allele has been shown to be dominant by Furuya and associates (0.65/0.35) (20).

With regard to sarcoidosis, Arbustini and colleagues (10) examined 61 patients and 80 healthy control subjects for this polymorphism in 1996. They reported no significant difference in I/D ratio or genotype distribution between healthy control subjects and patients, identical to our findings. Regarding the serum ACE level, however, the same investigators observed only a trend for increase in serum ACE level, according to the order II < DI < DD, whereas our data at the first visit to our hospital showed a significant difference. This might stem from the fact that they examined fewer cases, as well as from their use of the highest serum ACE level during the disease course to assess the relationship with the ACE genotype. Secondly, there may be a racial difference (1). Finally, they did not verify the mistyping that is said to occur for 4 to 5% of the DI type cases, which might also have affected their results (14, 15, 21). Regarding clinical manifestations, we investigated individual organ involvement, with special attention to the heart, because ACE plays a key role in the renin-angiotensin and kallikrein-kinin system. Our results, however, showed no significant variation with the genotype.

In following patients, a predictive marker for prognosis of sarcoidosis is desired. We did not observe any clear effects on prognosis, although we conjectured that with the DD type the disease might be more prolonged (6).

Clinically, the serum ACE level is useful for the evaluation of disease activity and for diagnosis in sarcoidosis (2, 22). However, the reported sensitivity is not high, with average incidence of elevated serum ACE level in sarcoidosis being in the order of 50% to 60% (3, 24, 26). In our study, the sensitivity of the conventional normal range (8.3-21.4 IU/l) for diagnosis was also 60.8% in total and only about 50% for the II type. The fact that the conventional normal range is far greater than the respective ranges of the three genotypes may be one reason for this lower sensitivity. In fact, our new genotype-specific normal range proved to be more sensitive than the conventional normal range by 22% in general, and by 39% for the II type, in particular. For precise assessment, determination of ACE genotype and the normal range of each genotype are necessary. Especially in those cases that have borderline ACE level with suspicion of sarcoidosis, it would appear necessary to determine the genotype.

Subtraction of the mean serum ACE level of normal control subjects from values for sarcoidosis patients of the corresponding genotype at the first visit to our hospital gave a balance of 10.6 ± 7.9, 10.1 ± 7.2, and 10.1 ± 7.5 for the II, DI, and DD type, respectively. The fact that these are almost identical indicates that the degree of elevation of serum ACE level due to the effect of the disease does not depend on the genotype. In sarcoidosis, elevated serum ACE is believed to be produced by epithelioid cells of the granulomas and alveolar macrophages (2, 24). The mechanism of control, which is apparently not influenced by the polymorphism, remains to be elucidated.